From c2f0c7c356dc9ae15419f00c725a2fcc58eeff58 Mon Sep 17 00:00:00 2001 From: Steve Grubb Date: Fri, 6 May 2005 12:38:39 +0100 Subject: The attached patch addresses the problem with getting the audit daemon shutdown credential information. It creates a new message type AUDIT_TERM_INFO, which is used by the audit daemon to query who issued the shutdown. It requires the placement of a hook function that gathers the information. The hook is after the DAC & MAC checks and before the function returns. Racing threads could overwrite the uid & pid - but they would have to be root and have policy that allows signalling the audit daemon. That should be a manageable risk. The userspace component will be released later in audit 0.7.2. When it receives the TERM signal, it queries the kernel for shutdown information. When it receives it, it writes the message and exits. The message looks like this: type=DAEMON msg=auditd(1114551182.000) auditd normal halt, sending pid=2650 uid=525, auditd pid=1685 Signed-off-by: Steve Grubb Signed-off-by: David Woodhouse --- include/linux/audit.h | 25 +++++++++++++++++-------- 1 file changed, 17 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/linux/audit.h b/include/linux/audit.h index 19f04b049798..baa80760824c 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -28,14 +28,16 @@ #include /* Request and reply types */ -#define AUDIT_GET 1000 /* Get status */ -#define AUDIT_SET 1001 /* Set status (enable/disable/auditd) */ -#define AUDIT_LIST 1002 /* List filtering rules */ -#define AUDIT_ADD 1003 /* Add filtering rule */ -#define AUDIT_DEL 1004 /* Delete filtering rule */ -#define AUDIT_USER 1005 /* Send a message from user-space */ -#define AUDIT_LOGIN 1006 /* Define the login id and informaiton */ -#define AUDIT_KERNEL 2000 /* Asynchronous audit record. NOT A REQUEST. */ +#define AUDIT_GET 1000 /* Get status */ +#define AUDIT_SET 1001 /* Set status (enable/disable/auditd) */ +#define AUDIT_LIST 1002 /* List filtering rules */ +#define AUDIT_ADD 1003 /* Add filtering rule */ +#define AUDIT_DEL 1004 /* Delete filtering rule */ +#define AUDIT_USER 1005 /* Send a message from user-space */ +#define AUDIT_LOGIN 1006 /* Define the login id and information */ +#define AUDIT_SIGNAL_INFO 1010 /* Get information about sender of signal*/ + +#define AUDIT_KERNEL 2000 /* Asynchronous audit record. NOT A REQUEST. */ /* Rule flags */ #define AUDIT_PER_TASK 0x01 /* Apply rule at task creation (not syscall) */ @@ -161,6 +163,11 @@ struct audit_rule { /* for AUDIT_LIST, AUDIT_ADD, and AUDIT_DEL */ #ifdef __KERNEL__ +struct audit_sig_info { + uid_t uid; + pid_t pid; +}; + struct audit_buffer; struct audit_context; struct inode; @@ -190,6 +197,7 @@ extern void audit_get_stamp(struct audit_context *ctx, extern int audit_set_loginuid(struct task_struct *task, uid_t loginuid); extern uid_t audit_get_loginuid(struct audit_context *ctx); extern int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode); +extern void audit_signal_info(int sig, struct task_struct *t); #else #define audit_alloc(t) ({ 0; }) #define audit_free(t) do { ; } while (0) @@ -200,6 +208,7 @@ extern int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mo #define audit_inode(n,i) do { ; } while (0) #define audit_get_loginuid(c) ({ -1; }) #define audit_ipc_perms(q,u,g,m) ({ 0; }) +#define audit_signal_info(s,t) do { ; } while (0) #endif #ifdef CONFIG_AUDIT -- cgit v1.2.3 From ea9c102cb0a7969df5733d34f26e0b12c8a3c889 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Sun, 8 May 2005 15:56:09 +0100 Subject: Add CONFIG_AUDITSC and CONFIG_SECCOMP support for ppc32 Signed-off-by: David Woodhouse --- arch/ppc/Kconfig | 17 +++++++++++++++++ arch/ppc/kernel/entry.S | 16 +++++++++------- arch/ppc/kernel/ppc_ksyms.c | 2 -- arch/ppc/kernel/ptrace.c | 40 ++++++++++++++++++++++++++++++++++++---- include/asm-ppc/thread_info.h | 7 +++++++ init/Kconfig | 2 +- 6 files changed, 70 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/arch/ppc/Kconfig b/arch/ppc/Kconfig index 600f23d7fd33..cd752a3cf3bd 100644 --- a/arch/ppc/Kconfig +++ b/arch/ppc/Kconfig @@ -1083,6 +1083,23 @@ source "drivers/zorro/Kconfig" source kernel/power/Kconfig +config SECCOMP + bool "Enable seccomp to safely compute untrusted bytecode" + depends on PROC_FS + default y + help + This kernel feature is useful for number crunching applications + that may need to compute untrusted bytecode during their + execution. By using pipes or other transports made available to + the process as file descriptors supporting the read/write + syscalls, it's possible to isolate those applications in + their own address space using seccomp. Once seccomp is + enabled via /proc//seccomp, it cannot be disabled + and the task is only allowed to execute a few safe syscalls + defined by each seccomp mode. + + If unsure, say Y. Only embedded should say N here. + endmenu config ISA_DMA_API diff --git a/arch/ppc/kernel/entry.S b/arch/ppc/kernel/entry.S index 5f075dbc4ee7..661523707e8c 100644 --- a/arch/ppc/kernel/entry.S +++ b/arch/ppc/kernel/entry.S @@ -202,7 +202,7 @@ _GLOBAL(DoSyscall) rlwinm r11,r11,0,~_TIFL_FORCE_NOERROR stw r11,TI_LOCAL_FLAGS(r10) lwz r11,TI_FLAGS(r10) - andi. r11,r11,_TIF_SYSCALL_TRACE + andi. r11,r11,_TIF_SYSCALL_T_OR_A bne- syscall_dotrace syscall_dotrace_cont: cmplwi 0,r0,NR_syscalls @@ -237,7 +237,7 @@ ret_from_syscall: SYNC MTMSRD(r10) lwz r9,TI_FLAGS(r12) - andi. r0,r9,(_TIF_SYSCALL_TRACE|_TIF_SIGPENDING|_TIF_NEED_RESCHED) + andi. r0,r9,(_TIF_SYSCALL_T_OR_A|_TIF_SIGPENDING|_TIF_NEED_RESCHED) bne- syscall_exit_work syscall_exit_cont: #if defined(CONFIG_4xx) || defined(CONFIG_BOOKE) @@ -277,7 +277,8 @@ syscall_dotrace: SAVE_NVGPRS(r1) li r0,0xc00 stw r0,TRAP(r1) - bl do_syscall_trace + addi r3,r1,STACK_FRAME_OVERHEAD + bl do_syscall_trace_enter lwz r0,GPR0(r1) /* Restore original registers */ lwz r3,GPR3(r1) lwz r4,GPR4(r1) @@ -291,7 +292,7 @@ syscall_dotrace: syscall_exit_work: stw r6,RESULT(r1) /* Save result */ stw r3,GPR3(r1) /* Update return value */ - andi. r0,r9,_TIF_SYSCALL_TRACE + andi. r0,r9,_TIF_SYSCALL_T_OR_A beq 5f ori r10,r10,MSR_EE SYNC @@ -303,7 +304,8 @@ syscall_exit_work: li r4,0xc00 stw r4,TRAP(r1) 4: - bl do_syscall_trace + addi r3,r1,STACK_FRAME_OVERHEAD + bl do_syscall_trace_leave REST_NVGPRS(r1) 2: lwz r3,GPR3(r1) @@ -627,8 +629,8 @@ sigreturn_exit: subi r1,r3,STACK_FRAME_OVERHEAD rlwinm r12,r1,0,0,18 /* current_thread_info() */ lwz r9,TI_FLAGS(r12) - andi. r0,r9,_TIF_SYSCALL_TRACE - bnel- do_syscall_trace + andi. r0,r9,_TIF_SYSCALL_T_OR_A + bnel- do_syscall_trace_leave /* fall through */ .globl ret_from_except_full diff --git a/arch/ppc/kernel/ppc_ksyms.c b/arch/ppc/kernel/ppc_ksyms.c index 2ccb58fe4fc3..d59ad07de8e7 100644 --- a/arch/ppc/kernel/ppc_ksyms.c +++ b/arch/ppc/kernel/ppc_ksyms.c @@ -55,7 +55,6 @@ #define EXPORT_SYMTAB_STROPS extern void transfer_to_handler(void); -extern void do_syscall_trace(void); extern void do_IRQ(struct pt_regs *regs); extern void MachineCheckException(struct pt_regs *regs); extern void AlignmentException(struct pt_regs *regs); @@ -74,7 +73,6 @@ extern unsigned long mm_ptov (unsigned long paddr); EXPORT_SYMBOL(clear_pages); EXPORT_SYMBOL(clear_user_page); EXPORT_SYMBOL(do_signal); -EXPORT_SYMBOL(do_syscall_trace); EXPORT_SYMBOL(transfer_to_handler); EXPORT_SYMBOL(do_IRQ); EXPORT_SYMBOL(MachineCheckException); diff --git a/arch/ppc/kernel/ptrace.c b/arch/ppc/kernel/ptrace.c index 59d59a8dc249..e7aee4108dea 100644 --- a/arch/ppc/kernel/ptrace.c +++ b/arch/ppc/kernel/ptrace.c @@ -27,6 +27,9 @@ #include #include #include +#include +#include +#include #include #include @@ -455,11 +458,10 @@ out: return ret; } -void do_syscall_trace(void) +static void do_syscall_trace(void) { - if (!test_thread_flag(TIF_SYSCALL_TRACE) - || !(current->ptrace & PT_PTRACED)) - return; + /* the 0x80 provides a way for the tracing parent to distinguish + between a syscall stop and SIGTRAP delivery */ ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD) ? 0x80 : 0)); @@ -473,3 +475,33 @@ void do_syscall_trace(void) current->exit_code = 0; } } + +void do_syscall_trace_enter(struct pt_regs *regs) +{ + if (test_thread_flag(TIF_SYSCALL_TRACE) + && (current->ptrace & PT_PTRACED)) + do_syscall_trace(); + + if (unlikely(current->audit_context)) + audit_syscall_entry(current, AUDIT_ARCH_PPC, + regs->gpr[0], + regs->gpr[3], regs->gpr[4], + regs->gpr[5], regs->gpr[6]); +} + +void do_syscall_trace_leave(struct pt_regs *regs) +{ + secure_computing(regs->gpr[0]); + + if (unlikely(current->audit_context)) + audit_syscall_exit(current, + (regs->ccr&0x1000)?AUDITSC_FAILURE:AUDITSC_SUCCESS, + regs->result); + + if ((test_thread_flag(TIF_SYSCALL_TRACE)) + && (current->ptrace & PT_PTRACED)) + do_syscall_trace(); +} + +EXPORT_SYMBOL(do_syscall_trace_enter); +EXPORT_SYMBOL(do_syscall_trace_leave); diff --git a/include/asm-ppc/thread_info.h b/include/asm-ppc/thread_info.h index f7f01524e8a8..e3b5284a6f91 100644 --- a/include/asm-ppc/thread_info.h +++ b/include/asm-ppc/thread_info.h @@ -77,12 +77,19 @@ static inline struct thread_info *current_thread_info(void) #define TIF_POLLING_NRFLAG 4 /* true if poll_idle() is polling TIF_NEED_RESCHED */ #define TIF_MEMDIE 5 +#define TIF_SYSCALL_AUDIT 6 /* syscall auditing active */ +#define TIF_SECCOMP 7 /* secure computing */ + /* as above, but as bit values */ #define _TIF_SYSCALL_TRACE (1< Date: Wed, 11 May 2005 10:54:05 +0100 Subject: Move ifdef CONFIG_AUDITSYSCALL to header Remove code conditionally dependent on CONFIG_AUDITSYSCALL from audit.c. Move these dependencies to audit.h with the rest. Signed-off-by: Chris Wright Signed-off-by: David Woodhouse --- include/linux/audit.h | 4 +++- kernel/audit.c | 12 ++---------- kernel/auditsc.c | 7 +++---- 3 files changed, 8 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/linux/audit.h b/include/linux/audit.h index baa80760824c..58c5589b531f 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -192,7 +192,7 @@ extern void audit_inode(const char *name, const struct inode *inode); /* Private API (for audit.c only) */ extern int audit_receive_filter(int type, int pid, int uid, int seq, void *data, uid_t loginuid); -extern void audit_get_stamp(struct audit_context *ctx, +extern int audit_get_stamp(struct audit_context *ctx, struct timespec *t, unsigned int *serial); extern int audit_set_loginuid(struct task_struct *task, uid_t loginuid); extern uid_t audit_get_loginuid(struct audit_context *ctx); @@ -206,6 +206,8 @@ extern void audit_signal_info(int sig, struct task_struct *t); #define audit_getname(n) do { ; } while (0) #define audit_putname(n) do { ; } while (0) #define audit_inode(n,i) do { ; } while (0) +#define audit_receive_filter(t,p,u,s,d,l) ({ -EOPNOTSUPP; }) +#define audit_get_stamp(c,t,s) ({ 0; }) #define audit_get_loginuid(c) ({ -1; }) #define audit_ipc_perms(q,u,g,m) ({ 0; }) #define audit_signal_info(s,t) do { ; } while (0) diff --git a/kernel/audit.c b/kernel/audit.c index dc4aba21f30a..c18b769e23a2 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -416,12 +416,8 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) return -EINVAL; /* fallthrough */ case AUDIT_LIST: -#ifdef CONFIG_AUDITSYSCALL err = audit_receive_filter(nlh->nlmsg_type, NETLINK_CB(skb).pid, uid, seq, data, loginuid); -#else - err = -EOPNOTSUPP; -#endif break; case AUDIT_SIGNAL_INFO: sig_data.uid = audit_sig_uid; @@ -636,15 +632,11 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx) return NULL; } -#ifdef CONFIG_AUDITSYSCALL - if (ab->ctx) - audit_get_stamp(ab->ctx, &t, &serial); - else -#endif - { + if (!audit_get_stamp(ab->ctx, &t, &serial)) { t = CURRENT_TIME; serial = 0; } + audit_log_format(ab, "audit(%lu.%03lu:%u): ", t.tv_sec, t.tv_nsec/1000000, serial); return ab; diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 680bb928343b..94338abf76f5 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -992,7 +992,7 @@ void audit_inode(const char *name, const struct inode *inode) context->names[idx].rdev = inode->i_rdev; } -void audit_get_stamp(struct audit_context *ctx, +int audit_get_stamp(struct audit_context *ctx, struct timespec *t, unsigned int *serial) { if (ctx) { @@ -1000,10 +1000,9 @@ void audit_get_stamp(struct audit_context *ctx, t->tv_nsec = ctx->ctime.tv_nsec; *serial = ctx->serial; ctx->auditable = 1; - } else { - *t = CURRENT_TIME; - *serial = 0; + return 1; } + return 0; } extern int audit_set_type(struct audit_buffer *ab, int type); -- cgit v1.2.3 From c1b773d87eadc3972d697444127e89a7291769a2 Mon Sep 17 00:00:00 2001 From: Chris Wright Date: Wed, 11 May 2005 10:55:10 +0100 Subject: Add audit_log_type Add audit_log_type to allow callers to specify type and pid when logging. Convert audit_log to wrapper around audit_log_type. Could have converted all audit_log callers directly, but common case is default of type AUDIT_KERNEL and pid 0. Update audit_log_start to take type and pid values when creating a new audit_buffer. Move sequences that did audit_log_start, audit_log_format, audit_set_type, audit_log_end, to simply call audit_log_type directly. This obsoletes audit_set_type and audit_set_pid, so remove them. Signed-off-by: Chris Wright Signed-off-by: David Woodhouse --- include/linux/audit.h | 16 ++++++++++------ kernel/audit.c | 48 +++++++++++++++--------------------------------- kernel/auditsc.c | 23 +++++++---------------- security/selinux/avc.c | 2 +- 4 files changed, 33 insertions(+), 56 deletions(-) (limited to 'include') diff --git a/include/linux/audit.h b/include/linux/audit.h index 58c5589b531f..405332ebf3c6 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -216,11 +216,14 @@ extern void audit_signal_info(int sig, struct task_struct *t); #ifdef CONFIG_AUDIT /* These are defined in audit.c */ /* Public API */ -extern void audit_log(struct audit_context *ctx, - const char *fmt, ...) - __attribute__((format(printf,2,3))); +#define audit_log(ctx, fmt, args...) \ + audit_log_type(ctx, AUDIT_KERNEL, 0, fmt, ##args) +extern void audit_log_type(struct audit_context *ctx, int type, + int pid, const char *fmt, ...) + __attribute__((format(printf,4,5))); -extern struct audit_buffer *audit_log_start(struct audit_context *ctx); +extern struct audit_buffer *audit_log_start(struct audit_context *ctx, int type, + int pid); extern void audit_log_format(struct audit_buffer *ab, const char *fmt, ...) __attribute__((format(printf,2,3))); @@ -240,8 +243,9 @@ extern void audit_send_reply(int pid, int seq, int type, void *payload, int size); extern void audit_log_lost(const char *message); #else -#define audit_log(t,f,...) do { ; } while (0) -#define audit_log_start(t) ({ NULL; }) +#define audit_log(c,f,...) do { ; } while (0) +#define audit_log_type(c,t,p,f,...) do { ; } while (0) +#define audit_log_start(c,t,p) ({ NULL; }) #define audit_log_vformat(b,f,a) do { ; } while (0) #define audit_log_format(b,f,...) do { ; } while (0) #define audit_log_end(b) do { ; } while (0) diff --git a/kernel/audit.c b/kernel/audit.c index c18b769e23a2..060b554f481e 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -140,18 +140,6 @@ struct audit_buffer { struct audit_context *ctx; /* NULL or associated context */ }; -void audit_set_type(struct audit_buffer *ab, int type) -{ - struct nlmsghdr *nlh = (struct nlmsghdr *)ab->skb->data; - nlh->nlmsg_type = type; -} - -static void audit_set_pid(struct audit_buffer *ab, pid_t pid) -{ - struct nlmsghdr *nlh = (struct nlmsghdr *)ab->skb->data; - nlh->nlmsg_pid = pid; -} - struct audit_entry { struct list_head list; struct audit_rule rule; @@ -344,7 +332,6 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) void *data; struct audit_status *status_get, status_set; int err; - struct audit_buffer *ab; u16 msg_type = nlh->nlmsg_type; uid_t loginuid; /* loginuid of sender */ struct audit_sig_info sig_data; @@ -396,19 +383,13 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) loginuid); break; case AUDIT_USER: - ab = audit_log_start(NULL); - if (!ab) - break; /* audit_panic has been called */ - audit_log_format(ab, + audit_log_type(NULL, AUDIT_USER, pid, "user pid=%d uid=%d length=%d loginuid=%u" " msg='%.1024s'", pid, uid, (int)(nlh->nlmsg_len - ((char *)data - (char *)nlh)), loginuid, (char *)data); - audit_set_type(ab, AUDIT_USER); - audit_set_pid(ab, pid); - audit_log_end(ab); break; case AUDIT_ADD: case AUDIT_DEL: @@ -560,12 +541,10 @@ static void audit_buffer_free(struct audit_buffer *ab) spin_unlock_irqrestore(&audit_freelist_lock, flags); } -static struct audit_buffer * audit_buffer_alloc(struct audit_context *ctx, - int gfp_mask) +static struct audit_buffer * audit_buffer_alloc(int gfp_mask) { unsigned long flags; struct audit_buffer *ab = NULL; - struct nlmsghdr *nlh; spin_lock_irqsave(&audit_freelist_lock, flags); if (!list_empty(&audit_freelist)) { @@ -587,12 +566,6 @@ static struct audit_buffer * audit_buffer_alloc(struct audit_context *ctx, if (!ab->skb) goto err; - ab->ctx = ctx; - nlh = (struct nlmsghdr *)skb_put(ab->skb, NLMSG_SPACE(0)); - nlh->nlmsg_type = AUDIT_KERNEL; - nlh->nlmsg_flags = 0; - nlh->nlmsg_pid = 0; - nlh->nlmsg_seq = 0; return ab; err: audit_buffer_free(ab); @@ -605,11 +578,12 @@ err: * syscall, then the syscall is marked as auditable and an audit record * will be written at syscall exit. If there is no associated task, tsk * should be NULL. */ -struct audit_buffer *audit_log_start(struct audit_context *ctx) +struct audit_buffer *audit_log_start(struct audit_context *ctx, int type, int pid) { struct audit_buffer *ab = NULL; struct timespec t; unsigned int serial; + struct nlmsghdr *nlh; if (!audit_initialized) return NULL; @@ -626,12 +600,19 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx) return NULL; } - ab = audit_buffer_alloc(ctx, GFP_ATOMIC); + ab = audit_buffer_alloc(GFP_ATOMIC); if (!ab) { audit_log_lost("out of memory in audit_log_start"); return NULL; } + ab->ctx = ctx; + nlh = (struct nlmsghdr *)skb_put(ab->skb, NLMSG_SPACE(0)); + nlh->nlmsg_type = type; + nlh->nlmsg_flags = 0; + nlh->nlmsg_pid = pid; + nlh->nlmsg_seq = 0; + if (!audit_get_stamp(ab->ctx, &t, &serial)) { t = CURRENT_TIME; serial = 0; @@ -828,12 +809,13 @@ void audit_log_end(struct audit_buffer *ab) /* Log an audit record. This is a convenience function that calls * audit_log_start, audit_log_vformat, and audit_log_end. It may be * called in any context. */ -void audit_log(struct audit_context *ctx, const char *fmt, ...) +void audit_log_type(struct audit_context *ctx, int type, int pid, + const char *fmt, ...) { struct audit_buffer *ab; va_list args; - ab = audit_log_start(ctx); + ab = audit_log_start(ctx, type, pid); if (ab) { va_start(args, fmt); audit_log_vformat(ab, fmt, args); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 94338abf76f5..d089263253a7 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -648,7 +648,7 @@ static void audit_log_exit(struct audit_context *context) int i; struct audit_buffer *ab; - ab = audit_log_start(context); + ab = audit_log_start(context, AUDIT_KERNEL, 0); if (!ab) return; /* audit_panic has been called */ audit_log_format(ab, "syscall=%d", context->major); @@ -680,7 +680,7 @@ static void audit_log_exit(struct audit_context *context) while (context->aux) { struct audit_aux_data *aux; - ab = audit_log_start(context); + ab = audit_log_start(context, AUDIT_KERNEL, 0); if (!ab) continue; /* audit_panic has been called */ @@ -701,7 +701,7 @@ static void audit_log_exit(struct audit_context *context) } for (i = 0; i < context->name_count; i++) { - ab = audit_log_start(context); + ab = audit_log_start(context, AUDIT_KERNEL, 0); if (!ab) continue; /* audit_panic has been called */ audit_log_format(ab, "item=%d", i); @@ -1005,22 +1005,13 @@ int audit_get_stamp(struct audit_context *ctx, return 0; } -extern int audit_set_type(struct audit_buffer *ab, int type); - int audit_set_loginuid(struct task_struct *task, uid_t loginuid) { if (task->audit_context) { - struct audit_buffer *ab; - - ab = audit_log_start(NULL); - if (ab) { - audit_log_format(ab, "login pid=%d uid=%u " - "old loginuid=%u new loginuid=%u", - task->pid, task->uid, - task->audit_context->loginuid, loginuid); - audit_set_type(ab, AUDIT_LOGIN); - audit_log_end(ab); - } + audit_log_type(NULL, AUDIT_LOGIN, 0, + "login pid=%d uid=%u old loginuid=%u new loginuid=%u", + task->pid, task->uid, task->audit_context->loginuid, + loginuid); task->audit_context->loginuid = loginuid; } return 0; diff --git a/security/selinux/avc.c b/security/selinux/avc.c index 85a6f66a873f..9e71a1bbe011 100644 --- a/security/selinux/avc.c +++ b/security/selinux/avc.c @@ -549,7 +549,7 @@ void avc_audit(u32 ssid, u32 tsid, return; } - ab = audit_log_start(current->audit_context); + ab = audit_log_start(current->audit_context, AUDIT_KERNEL, 0); if (!ab) return; /* audit_panic has been called */ audit_log_format(ab, "avc: %s ", denied ? "denied" : "granted"); -- cgit v1.2.3 From add671412021b68c3b4f2882b0d10a56e2dcdabe Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 11 May 2005 11:36:21 +0100 Subject: Add missing asm-ppc/seccomp.h. Must learn to use git properly. Signed-off-by: David Woodhouse --- include/asm-ppc/seccomp.h | 10 ++++++++++ 1 file changed, 10 insertions(+) create mode 100644 include/asm-ppc/seccomp.h (limited to 'include') diff --git a/include/asm-ppc/seccomp.h b/include/asm-ppc/seccomp.h new file mode 100644 index 000000000000..666c4da96d87 --- /dev/null +++ b/include/asm-ppc/seccomp.h @@ -0,0 +1,10 @@ +#ifndef _ASM_SECCOMP_H + +#include + +#define __NR_seccomp_read __NR_read +#define __NR_seccomp_write __NR_write +#define __NR_seccomp_exit __NR_exit +#define __NR_seccomp_sigreturn __NR_rt_sigreturn + +#endif /* _ASM_SECCOMP_H */ -- cgit v1.2.3 From 6f2f38128170814e151cfedf79532e19cd179567 Mon Sep 17 00:00:00 2001 From: Brad Campbell Date: Thu, 12 May 2005 15:07:47 -0400 Subject: [PATCH] libata basic detection and errata for PATA->SATA bridges This patch works around an issue with WD drives (and possibly others) over SiL PATA->SATA Bridges on SATA controllers locking up with transfers > 200 sectors. Signed-off-by: Brad Campbell --- drivers/scsi/libata-core.c | 35 +++++++++++++++++++++++++++++++++-- include/linux/ata.h | 1 + include/linux/libata.h | 1 + 3 files changed, 35 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c index 0b5d3a5b7eda..8b5a3f00083d 100644 --- a/drivers/scsi/libata-core.c +++ b/drivers/scsi/libata-core.c @@ -1186,6 +1186,37 @@ err_out: DPRINTK("EXIT, err\n"); } + +static inline u8 ata_dev_knobble(struct ata_port *ap) +{ + return ((ap->cbl == ATA_CBL_SATA) && (!ata_id_is_sata(ap->device->id))); +} + +/** + * ata_dev_config - Run device specific handlers and check for + * SATA->PATA bridges + * @ap: Bus + * @i: Device + * + * LOCKING: + */ + +void ata_dev_config(struct ata_port *ap, unsigned int i) +{ + /* limit bridge transfers to udma5, 200 sectors */ + if (ata_dev_knobble(ap)) { + printk(KERN_INFO "ata%u(%u): applying bridge limits\n", + ap->id, ap->device->devno); + ap->udma_mask &= ATA_UDMA5; + ap->host->max_sectors = ATA_MAX_SECTORS; + ap->host->hostt->max_sectors = ATA_MAX_SECTORS; + ap->device->flags |= ATA_DFLAG_LOCK_SECTORS; + } + + if (ap->ops->dev_config) + ap->ops->dev_config(ap, &ap->device[i]); +} + /** * ata_bus_probe - Reset and probe ATA bus * @ap: Bus to probe @@ -1208,8 +1239,7 @@ static int ata_bus_probe(struct ata_port *ap) ata_dev_identify(ap, i); if (ata_dev_present(&ap->device[i])) { found = 1; - if (ap->ops->dev_config) - ap->ops->dev_config(ap, &ap->device[i]); + ata_dev_config(ap,i); } } @@ -4014,6 +4044,7 @@ EXPORT_SYMBOL_GPL(ata_scsi_release); EXPORT_SYMBOL_GPL(ata_host_intr); EXPORT_SYMBOL_GPL(ata_dev_classify); EXPORT_SYMBOL_GPL(ata_dev_id_string); +EXPORT_SYMBOL_GPL(ata_dev_config); EXPORT_SYMBOL_GPL(ata_scsi_simulate); #ifdef CONFIG_PCI diff --git a/include/linux/ata.h b/include/linux/ata.h index f178894edd04..ca5fcadf9981 100644 --- a/include/linux/ata.h +++ b/include/linux/ata.h @@ -224,6 +224,7 @@ struct ata_taskfile { }; #define ata_id_is_ata(id) (((id)[0] & (1 << 15)) == 0) +#define ata_id_is_sata(id) ((id)[93] == 0) #define ata_id_rahead_enabled(id) ((id)[85] & (1 << 6)) #define ata_id_wcache_enabled(id) ((id)[85] & (1 << 5)) #define ata_id_has_flush(id) ((id)[83] & (1 << 12)) diff --git a/include/linux/libata.h b/include/linux/libata.h index 505160ab472b..d33e70361a7d 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -420,6 +420,7 @@ extern void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg, extern unsigned int ata_dev_classify(struct ata_taskfile *tf); extern void ata_dev_id_string(u16 *id, unsigned char *s, unsigned int ofs, unsigned int len); +extern void ata_dev_config(struct ata_port *ap, unsigned int i); extern void ata_bmdma_setup (struct ata_queued_cmd *qc); extern void ata_bmdma_start (struct ata_queued_cmd *qc); extern void ata_bmdma_stop(struct ata_port *ap); -- cgit v1.2.3 From 7d17c1d606f6e89778f05554ddea43791d5c92a0 Mon Sep 17 00:00:00 2001 From: Date: Thu, 12 May 2005 19:45:25 -0400 Subject: [netdrvrs] Use netif_carrier_* instead of IFF_RUNNING --- drivers/net/au1000_eth.c | 10 ++-------- drivers/net/bmac.c | 2 -- drivers/net/sk98lin/skge.c | 8 +++----- drivers/net/tlan.c | 4 ++-- drivers/net/tokenring/ibmtr.c | 11 +++-------- drivers/net/wan/lmc/lmc_main.c | 8 ++------ include/linux/if.h | 2 +- 7 files changed, 13 insertions(+), 32 deletions(-) (limited to 'include') diff --git a/drivers/net/au1000_eth.c b/drivers/net/au1000_eth.c index 5a2efd343db4..c82b9cd1c924 100644 --- a/drivers/net/au1000_eth.c +++ b/drivers/net/au1000_eth.c @@ -1681,10 +1681,6 @@ static int au1000_init(struct net_device *dev) control |= MAC_FULL_DUPLEX; } - /* fix for startup without cable */ - if (!link) - dev->flags &= ~IFF_RUNNING; - aup->mac->control = control; aup->mac->vlan1_tag = 0x8100; /* activate vlan support */ au_sync(); @@ -1709,16 +1705,14 @@ static void au1000_timer(unsigned long data) if_port = dev->if_port; if (aup->phy_ops->phy_status(dev, aup->phy_addr, &link, &speed) == 0) { if (link) { - if (!(dev->flags & IFF_RUNNING)) { + if (!netif_carrier_ok(dev)) { netif_carrier_on(dev); - dev->flags |= IFF_RUNNING; printk(KERN_INFO "%s: link up\n", dev->name); } } else { - if (dev->flags & IFF_RUNNING) { + if (netif_carrier_ok(dev)) { netif_carrier_off(dev); - dev->flags &= ~IFF_RUNNING; dev->if_port = 0; printk(KERN_INFO "%s: link down\n", dev->name); } diff --git a/drivers/net/bmac.c b/drivers/net/bmac.c index 734bd4ee3f9b..00e5257b176f 100644 --- a/drivers/net/bmac.c +++ b/drivers/net/bmac.c @@ -1412,7 +1412,6 @@ static int bmac_open(struct net_device *dev) bp->opened = 1; bmac_reset_and_enable(dev); enable_irq(dev->irq); - dev->flags |= IFF_RUNNING; return 0; } @@ -1425,7 +1424,6 @@ static int bmac_close(struct net_device *dev) int i; bp->sleeping = 1; - dev->flags &= ~(IFF_UP | IFF_RUNNING); /* disable rx and tx */ config = bmread(dev, RXCFG); diff --git a/drivers/net/sk98lin/skge.c b/drivers/net/sk98lin/skge.c index 05b827f79f54..1ccb2989001c 100644 --- a/drivers/net/sk98lin/skge.c +++ b/drivers/net/sk98lin/skge.c @@ -4212,7 +4212,7 @@ SK_BOOL DualNet; Flags); SkGeStopPort(pAC, IoC, FromPort, SK_STOP_ALL, SK_HARD_RST); - pAC->dev[Param.Para32[0]]->flags &= ~IFF_RUNNING; + netif_carrier_off(pAC->dev[Param.Para32[0]]); spin_unlock_irqrestore( &pAC->TxPort[FromPort][TX_PRIO_LOW].TxDesRingLock, Flags); @@ -4355,7 +4355,7 @@ SK_BOOL DualNet; } /* Inform the world that link protocol is up. */ - pAC->dev[Param.Para32[0]]->flags |= IFF_RUNNING; + netif_carrier_on(pAC->dev[Param.Para32[0]]); break; case SK_DRV_NET_DOWN: /* SK_U32 Reason */ @@ -4368,7 +4368,7 @@ SK_BOOL DualNet; } else { DoPrintInterfaceChange = SK_TRUE; } - pAC->dev[Param.Para32[1]]->flags &= ~IFF_RUNNING; + netif_carrier_off(pAC->dev[Param.Para32[1]]); break; case SK_DRV_SWITCH_HARD: /* SK_U32 FromPortIdx SK_U32 ToPortIdx */ SK_DBG_MSG(NULL, SK_DBGMOD_DRV, SK_DBGCAT_DRV_EVENT, @@ -4961,7 +4961,6 @@ static int __devinit skge_probe_one(struct pci_dev *pdev, #ifdef CONFIG_NET_POLL_CONTROLLER dev->poll_controller = &SkGePollController; #endif - dev->flags &= ~IFF_RUNNING; SET_NETDEV_DEV(dev, &pdev->dev); SET_ETHTOOL_OPS(dev, &SkGeEthtoolOps); @@ -5035,7 +5034,6 @@ static int __devinit skge_probe_one(struct pci_dev *pdev, dev->set_mac_address = &SkGeSetMacAddr; dev->do_ioctl = &SkGeIoctl; dev->change_mtu = &SkGeChangeMtu; - dev->flags &= ~IFF_RUNNING; SET_NETDEV_DEV(dev, &pdev->dev); SET_ETHTOOL_OPS(dev, &SkGeEthtoolOps); diff --git a/drivers/net/tlan.c b/drivers/net/tlan.c index a7ffa64502dd..ce116624f136 100644 --- a/drivers/net/tlan.c +++ b/drivers/net/tlan.c @@ -2807,7 +2807,7 @@ void TLan_PhyMonitor( struct net_device *dev ) if (priv->link) { priv->link = 0; printk(KERN_DEBUG "TLAN: %s has lost link\n", dev->name); - dev->flags &= ~IFF_RUNNING; + netif_carrier_off(dev); TLan_SetTimer( dev, (2*HZ), TLAN_TIMER_LINK_BEAT ); return; } @@ -2817,7 +2817,7 @@ void TLan_PhyMonitor( struct net_device *dev ) if ((phy_status & MII_GS_LINK) && !priv->link) { priv->link = 1; printk(KERN_DEBUG "TLAN: %s has reestablished link\n", dev->name); - dev->flags |= IFF_RUNNING; + netif_carrier_on(dev); } /* Setup a new monitor */ diff --git a/drivers/net/tokenring/ibmtr.c b/drivers/net/tokenring/ibmtr.c index c098863bdd9d..3873917a9c22 100644 --- a/drivers/net/tokenring/ibmtr.c +++ b/drivers/net/tokenring/ibmtr.c @@ -888,11 +888,6 @@ static int tok_open(struct net_device *dev) ti->sap_status = CLOSED; /* CLOSED or OPEN */ ti->open_failure = NO; /* NO or YES */ ti->open_mode = MANUAL; /* MANUAL or AUTOMATIC */ - /* 12/2000 not typical Linux, but we can use RUNNING to let us know when - the network has crapped out or cables are disconnected. Useful because - the IFF_UP flag stays up the whole time, until ifconfig tr0 down. - */ - dev->flags &= ~IFF_RUNNING; ti->sram_phys &= ~1; /* to reverse what we do in tok_close */ /* init the spinlock */ @@ -1242,7 +1237,7 @@ irqreturn_t tok_interrupt(int irq, void *dev_id, struct pt_regs *regs) ti->open_status = CLOSED; ti->sap_status = CLOSED; ti->open_mode = AUTOMATIC; - dev->flags &= ~IFF_RUNNING; + netif_carrier_off(dev); netif_stop_queue(dev); ti->open_action = RESTART; outb(0, dev->base_addr + ADAPTRESET); @@ -1323,7 +1318,7 @@ irqreturn_t tok_interrupt(int irq, void *dev_id, struct pt_regs *regs) break; } netif_wake_queue(dev); - dev->flags |= IFF_RUNNING;/*BMS 12/2000*/ + netif_carrier_on(dev); break; case DIR_INTERRUPT: case DIR_MOD_OPEN_PARAMS: @@ -1427,7 +1422,7 @@ irqreturn_t tok_interrupt(int irq, void *dev_id, struct pt_regs *regs) ring_status); if(ring_status& (REMOVE_RECV|AUTO_REMOVAL|LOBE_FAULT)){ netif_stop_queue(dev); - dev->flags &= ~IFF_RUNNING;/*not typical Linux*/ + netif_carrier_off(dev); DPRINTK("Remove received, or Auto-removal error" ", or Lobe fault\n"); DPRINTK("We'll try to reopen the closed adapter" diff --git a/drivers/net/wan/lmc/lmc_main.c b/drivers/net/wan/lmc/lmc_main.c index 15e545f66cd7..2b948ea397d5 100644 --- a/drivers/net/wan/lmc/lmc_main.c +++ b/drivers/net/wan/lmc/lmc_main.c @@ -723,7 +723,7 @@ static void lmc_watchdog (unsigned long data) /*fold00*/ /* lmc_reset (sc); Why reset??? The link can go down ok */ /* Inform the world that link has been lost */ - dev->flags &= ~IFF_RUNNING; + netif_carrier_off(dev); } /* @@ -736,7 +736,7 @@ static void lmc_watchdog (unsigned long data) /*fold00*/ /* lmc_reset (sc); Again why reset??? */ /* Inform the world that link protocol is back up. */ - dev->flags |= IFF_RUNNING; + netif_carrier_on(dev); /* Now we have to tell the syncppp that we had an outage * and that it should deal. Calling sppp_reopen here @@ -1168,8 +1168,6 @@ static void lmc_running_reset (struct net_device *dev) /*fold00*/ sc->lmc_media->set_link_status (sc, 1); sc->lmc_media->set_status (sc, NULL); - //dev->flags |= IFF_RUNNING; - netif_wake_queue(dev); sc->lmc_txfull = 0; @@ -1233,8 +1231,6 @@ static int lmc_ifdown (struct net_device *dev) /*fold00*/ csr6 &= ~LMC_DEC_SR; /* Turn off the Receive bit */ LMC_CSR_WRITE (sc, csr_command, csr6); - dev->flags &= ~IFF_RUNNING; - sc->stats.rx_missed_errors += LMC_CSR_READ (sc, csr_missed_frames) & 0xffff; diff --git a/include/linux/if.h b/include/linux/if.h index d73a9d62f208..ce627d9092ef 100644 --- a/include/linux/if.h +++ b/include/linux/if.h @@ -33,7 +33,7 @@ #define IFF_LOOPBACK 0x8 /* is a loopback net */ #define IFF_POINTOPOINT 0x10 /* interface is has p-p link */ #define IFF_NOTRAILERS 0x20 /* avoid use of trailers */ -#define IFF_RUNNING 0x40 /* resources allocated */ +#define IFF_RUNNING 0x40 /* interface running and carrier ok */ #define IFF_NOARP 0x80 /* no ARP protocol */ #define IFF_PROMISC 0x100 /* receive all packets */ #define IFF_ALLMULTI 0x200 /* receive all multicast packets*/ -- cgit v1.2.3 From fff9cfd99c0f88645c3f50d7476d6c8cef99f140 Mon Sep 17 00:00:00 2001 From: Date: Thu, 12 May 2005 20:24:19 -0400 Subject: [PATCH] Wireless Extensions 18 (aka WPA) This is version 18 of the Wireless Extensions. The main change is that it adds all the necessary APIs for WPA and WPA2 support. This work was entirely done by Jouni Malinen, so let's thank him for both his hard work and deep expertise on the subject ;-) This APIs obviously doesn't do much by itself and works in concert with driver support (Jouni already sent you the HostAP changes) and userspace (Jouni is updating wpa_supplicant). This is also orthogonal with the ongoing work on in-kernel IEEE support (but potentially useful). The patch is attached, tested with 2.6.11. Normally, I would ask you to push that directly in the kernel (99% of the patch has been on my web page for ages and it does not affect non-WPA stuff), but Jouni convinced me that it should bake a few weeks in wireless-2.6 first, so that other driver maintainers can get up to speed with it. Signed-off-by: Jeff Garzik --- include/linux/wireless.h | 283 ++++++++++++++++++++++++++++++++++++++++++++++- net/core/wireless.c | 74 ++++++++++++- 2 files changed, 352 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/wireless.h b/include/linux/wireless.h index 2f51f2b6562e..ae485f9c916e 100644 --- a/include/linux/wireless.h +++ b/include/linux/wireless.h @@ -1,10 +1,10 @@ /* * This file define a set of standard wireless extensions * - * Version : 17 21.6.04 + * Version : 18 12.3.05 * * Authors : Jean Tourrilhes - HPL - - * Copyright (c) 1997-2004 Jean Tourrilhes, All Rights Reserved. + * Copyright (c) 1997-2005 Jean Tourrilhes, All Rights Reserved. */ #ifndef _LINUX_WIRELESS_H @@ -82,7 +82,7 @@ * (there is some stuff that will be added in the future...) * I just plan to increment with each new version. */ -#define WIRELESS_EXT 17 +#define WIRELESS_EXT 18 /* * Changes : @@ -182,6 +182,21 @@ * - Document (struct iw_quality *)->updated, add new flags (INVALID) * - Wireless Event capability in struct iw_range * - Add support for relative TxPower (yick !) + * + * V17 to V18 (From Jouni Malinen ) + * ---------- + * - Add support for WPA/WPA2 + * - Add extended encoding configuration (SIOCSIWENCODEEXT and + * SIOCGIWENCODEEXT) + * - Add SIOCSIWGENIE/SIOCGIWGENIE + * - Add SIOCSIWMLME + * - Add SIOCSIWPMKSA + * - Add struct iw_range bit field for supported encoding capabilities + * - Add optional scan request parameters for SIOCSIWSCAN + * - Add SIOCSIWAUTH/SIOCGIWAUTH for setting authentication and WPA + * related parameters (extensible up to 4096 parameter values) + * - Add wireless events: IWEVGENIE, IWEVMICHAELMICFAILURE, + * IWEVASSOCREQIE, IWEVASSOCRESPIE, IWEVPMKIDCAND */ /**************************** CONSTANTS ****************************/ @@ -256,6 +271,30 @@ #define SIOCSIWPOWER 0x8B2C /* set Power Management settings */ #define SIOCGIWPOWER 0x8B2D /* get Power Management settings */ +/* WPA : Generic IEEE 802.11 informatiom element (e.g., for WPA/RSN/WMM). + * This ioctl uses struct iw_point and data buffer that includes IE id and len + * fields. More than one IE may be included in the request. Setting the generic + * IE to empty buffer (len=0) removes the generic IE from the driver. Drivers + * are allowed to generate their own WPA/RSN IEs, but in these cases, drivers + * are required to report the used IE as a wireless event, e.g., when + * associating with an AP. */ +#define SIOCSIWGENIE 0x8B30 /* set generic IE */ +#define SIOCGIWGENIE 0x8B31 /* get generic IE */ + +/* WPA : IEEE 802.11 MLME requests */ +#define SIOCSIWMLME 0x8B16 /* request MLME operation; uses + * struct iw_mlme */ +/* WPA : Authentication mode parameters */ +#define SIOCSIWAUTH 0x8B32 /* set authentication mode params */ +#define SIOCGIWAUTH 0x8B33 /* get authentication mode params */ + +/* WPA : Extended version of encoding configuration */ +#define SIOCSIWENCODEEXT 0x8B34 /* set encoding token & mode */ +#define SIOCGIWENCODEEXT 0x8B35 /* get encoding token & mode */ + +/* WPA2 : PMKSA cache management */ +#define SIOCSIWPMKSA 0x8B36 /* PMKSA cache operation */ + /* -------------------- DEV PRIVATE IOCTL LIST -------------------- */ /* These 32 ioctl are wireless device private, for 16 commands. @@ -297,6 +336,34 @@ #define IWEVCUSTOM 0x8C02 /* Driver specific ascii string */ #define IWEVREGISTERED 0x8C03 /* Discovered a new node (AP mode) */ #define IWEVEXPIRED 0x8C04 /* Expired a node (AP mode) */ +#define IWEVGENIE 0x8C05 /* Generic IE (WPA, RSN, WMM, ..) + * (scan results); This includes id and + * length fields. One IWEVGENIE may + * contain more than one IE. Scan + * results may contain one or more + * IWEVGENIE events. */ +#define IWEVMICHAELMICFAILURE 0x8C06 /* Michael MIC failure + * (struct iw_michaelmicfailure) + */ +#define IWEVASSOCREQIE 0x8C07 /* IEs used in (Re)Association Request. + * The data includes id and length + * fields and may contain more than one + * IE. This event is required in + * Managed mode if the driver + * generates its own WPA/RSN IE. This + * should be sent just before + * IWEVREGISTERED event for the + * association. */ +#define IWEVASSOCRESPIE 0x8C08 /* IEs used in (Re)Association + * Response. The data includes id and + * length fields and may contain more + * than one IE. This may be sent + * between IWEVASSOCREQIE and + * IWEVREGISTERED events for the + * association. */ +#define IWEVPMKIDCAND 0x8C09 /* PMKID candidate for RSN + * pre-authentication + * (struct iw_pmkid_cand) */ #define IWEVFIRST 0x8C00 @@ -432,12 +499,94 @@ #define IW_SCAN_THIS_MODE 0x0020 /* Scan only this Mode */ #define IW_SCAN_ALL_RATE 0x0040 /* Scan all Bit-Rates */ #define IW_SCAN_THIS_RATE 0x0080 /* Scan only this Bit-Rate */ +/* struct iw_scan_req scan_type */ +#define IW_SCAN_TYPE_ACTIVE 0 +#define IW_SCAN_TYPE_PASSIVE 1 /* Maximum size of returned data */ #define IW_SCAN_MAX_DATA 4096 /* In bytes */ /* Max number of char in custom event - use multiple of them if needed */ #define IW_CUSTOM_MAX 256 /* In bytes */ +/* Generic information element */ +#define IW_GENERIC_IE_MAX 1024 + +/* MLME requests (SIOCSIWMLME / struct iw_mlme) */ +#define IW_MLME_DEAUTH 0 +#define IW_MLME_DISASSOC 1 + +/* SIOCSIWAUTH/SIOCGIWAUTH struct iw_param flags */ +#define IW_AUTH_INDEX 0x0FFF +#define IW_AUTH_FLAGS 0xF000 +/* SIOCSIWAUTH/SIOCGIWAUTH parameters (0 .. 4095) + * (IW_AUTH_INDEX mask in struct iw_param flags; this is the index of the + * parameter that is being set/get to; value will be read/written to + * struct iw_param value field) */ +#define IW_AUTH_WPA_VERSION 0 +#define IW_AUTH_CIPHER_PAIRWISE 1 +#define IW_AUTH_CIPHER_GROUP 2 +#define IW_AUTH_KEY_MGMT 3 +#define IW_AUTH_TKIP_COUNTERMEASURES 4 +#define IW_AUTH_DROP_UNENCRYPTED 5 +#define IW_AUTH_80211_AUTH_ALG 6 +#define IW_AUTH_WPA_ENABLED 7 +#define IW_AUTH_RX_UNENCRYPTED_EAPOL 8 +#define IW_AUTH_ROAMING_CONTROL 9 +#define IW_AUTH_PRIVACY_INVOKED 10 + +/* IW_AUTH_WPA_VERSION values (bit field) */ +#define IW_AUTH_WPA_VERSION_DISABLED 0x00000001 +#define IW_AUTH_WPA_VERSION_WPA 0x00000002 +#define IW_AUTH_WPA_VERSION_WPA2 0x00000004 + +/* IW_AUTH_PAIRWISE_CIPHER and IW_AUTH_GROUP_CIPHER values (bit field) */ +#define IW_AUTH_CIPHER_NONE 0x00000001 +#define IW_AUTH_CIPHER_WEP40 0x00000002 +#define IW_AUTH_CIPHER_TKIP 0x00000004 +#define IW_AUTH_CIPHER_CCMP 0x00000008 +#define IW_AUTH_CIPHER_WEP104 0x00000010 + +/* IW_AUTH_KEY_MGMT values (bit field) */ +#define IW_AUTH_KEY_MGMT_802_1X 1 +#define IW_AUTH_KEY_MGMT_PSK 2 + +/* IW_AUTH_80211_AUTH_ALG values (bit field) */ +#define IW_AUTH_ALG_OPEN_SYSTEM 0x00000001 +#define IW_AUTH_ALG_SHARED_KEY 0x00000002 +#define IW_AUTH_ALG_LEAP 0x00000004 + +/* IW_AUTH_ROAMING_CONTROL values */ +#define IW_AUTH_ROAMING_ENABLE 0 /* driver/firmware based roaming */ +#define IW_AUTH_ROAMING_DISABLE 1 /* user space program used for roaming + * control */ + +/* SIOCSIWENCODEEXT definitions */ +#define IW_ENCODE_SEQ_MAX_SIZE 8 +/* struct iw_encode_ext ->alg */ +#define IW_ENCODE_ALG_NONE 0 +#define IW_ENCODE_ALG_WEP 1 +#define IW_ENCODE_ALG_TKIP 2 +#define IW_ENCODE_ALG_CCMP 3 +/* struct iw_encode_ext ->ext_flags */ +#define IW_ENCODE_EXT_TX_SEQ_VALID 0x00000001 +#define IW_ENCODE_EXT_RX_SEQ_VALID 0x00000002 +#define IW_ENCODE_EXT_GROUP_KEY 0x00000004 +#define IW_ENCODE_EXT_SET_TX_KEY 0x00000008 + +/* IWEVMICHAELMICFAILURE : struct iw_michaelmicfailure ->flags */ +#define IW_MICFAILURE_KEY_ID 0x00000003 /* Key ID 0..3 */ +#define IW_MICFAILURE_GROUP 0x00000004 +#define IW_MICFAILURE_PAIRWISE 0x00000008 +#define IW_MICFAILURE_STAKEY 0x00000010 +#define IW_MICFAILURE_COUNT 0x00000060 /* 1 or 2 (0 = count not supported) + */ + +/* Bit field values for enc_capa in struct iw_range */ +#define IW_ENC_CAPA_WPA 0x00000001 +#define IW_ENC_CAPA_WPA2 0x00000002 +#define IW_ENC_CAPA_CIPHER_TKIP 0x00000004 +#define IW_ENC_CAPA_CIPHER_CCMP 0x00000008 + /* Event capability macros - in (struct iw_range *)->event_capa * Because we have more than 32 possible events, we use an array of * 32 bit bitmasks. Note : 32 bits = 0x20 = 2^5. */ @@ -546,6 +695,132 @@ struct iw_thrspy struct iw_quality high; /* High threshold */ }; +/* + * Optional data for scan request + * + * Note: these optional parameters are controlling parameters for the + * scanning behavior, these do not apply to getting scan results + * (SIOCGIWSCAN). Drivers are expected to keep a local BSS table and + * provide a merged results with all BSSes even if the previous scan + * request limited scanning to a subset, e.g., by specifying an SSID. + * Especially, scan results are required to include an entry for the + * current BSS if the driver is in Managed mode and associated with an AP. + */ +struct iw_scan_req +{ + __u8 scan_type; /* IW_SCAN_TYPE_{ACTIVE,PASSIVE} */ + __u8 essid_len; + __u8 num_channels; /* num entries in channel_list; + * 0 = scan all allowed channels */ + __u8 flags; /* reserved as padding; use zero, this may + * be used in the future for adding flags + * to request different scan behavior */ + struct sockaddr bssid; /* ff:ff:ff:ff:ff:ff for broadcast BSSID or + * individual address of a specific BSS */ + + /* + * Use this ESSID if IW_SCAN_THIS_ESSID flag is used instead of using + * the current ESSID. This allows scan requests for specific ESSID + * without having to change the current ESSID and potentially breaking + * the current association. + */ + __u8 essid[IW_ESSID_MAX_SIZE]; + + /* + * Optional parameters for changing the default scanning behavior. + * These are based on the MLME-SCAN.request from IEEE Std 802.11. + * TU is 1.024 ms. If these are set to 0, driver is expected to use + * reasonable default values. min_channel_time defines the time that + * will be used to wait for the first reply on each channel. If no + * replies are received, next channel will be scanned after this. If + * replies are received, total time waited on the channel is defined by + * max_channel_time. + */ + __u32 min_channel_time; /* in TU */ + __u32 max_channel_time; /* in TU */ + + struct iw_freq channel_list[IW_MAX_FREQUENCIES]; +}; + +/* ------------------------- WPA SUPPORT ------------------------- */ + +/* + * Extended data structure for get/set encoding (this is used with + * SIOCSIWENCODEEXT/SIOCGIWENCODEEXT. struct iw_point and IW_ENCODE_* + * flags are used in the same way as with SIOCSIWENCODE/SIOCGIWENCODE and + * only the data contents changes (key data -> this structure, including + * key data). + * + * If the new key is the first group key, it will be set as the default + * TX key. Otherwise, default TX key index is only changed if + * IW_ENCODE_EXT_SET_TX_KEY flag is set. + * + * Key will be changed with SIOCSIWENCODEEXT in all cases except for + * special "change TX key index" operation which is indicated by setting + * key_len = 0 and ext_flags |= IW_ENCODE_EXT_SET_TX_KEY. + * + * tx_seq/rx_seq are only used when respective + * IW_ENCODE_EXT_{TX,RX}_SEQ_VALID flag is set in ext_flags. Normal + * TKIP/CCMP operation is to set RX seq with SIOCSIWENCODEEXT and start + * TX seq from zero whenever key is changed. SIOCGIWENCODEEXT is normally + * used only by an Authenticator (AP or an IBSS station) to get the + * current TX sequence number. Using TX_SEQ_VALID for SIOCSIWENCODEEXT and + * RX_SEQ_VALID for SIOCGIWENCODEEXT are optional, but can be useful for + * debugging/testing. + */ +struct iw_encode_ext +{ + __u32 ext_flags; /* IW_ENCODE_EXT_* */ + __u8 tx_seq[IW_ENCODE_SEQ_MAX_SIZE]; /* LSB first */ + __u8 rx_seq[IW_ENCODE_SEQ_MAX_SIZE]; /* LSB first */ + struct sockaddr addr; /* ff:ff:ff:ff:ff:ff for broadcast/multicast + * (group) keys or unicast address for + * individual keys */ + __u16 alg; /* IW_ENCODE_ALG_* */ + __u16 key_len; + __u8 key[0]; +}; + +/* SIOCSIWMLME data */ +struct iw_mlme +{ + __u16 cmd; /* IW_MLME_* */ + __u16 reason_code; + struct sockaddr addr; +}; + +/* SIOCSIWPMKSA data */ +#define IW_PMKSA_ADD 1 +#define IW_PMKSA_REMOVE 2 +#define IW_PMKSA_FLUSH 3 + +#define IW_PMKID_LEN 16 + +struct iw_pmksa +{ + __u32 cmd; /* IW_PMKSA_* */ + struct sockaddr bssid; + __u8 pmkid[IW_PMKID_LEN]; +}; + +/* IWEVMICHAELMICFAILURE data */ +struct iw_michaelmicfailure +{ + __u32 flags; + struct sockaddr src_addr; + __u8 tsc[IW_ENCODE_SEQ_MAX_SIZE]; /* LSB first */ +}; + +/* IWEVPMKIDCAND data */ +#define IW_PMKID_CAND_PREAUTH 0x00000001 /* RNS pre-authentication enabled */ +struct iw_pmkid_cand +{ + __u32 flags; /* IW_PMKID_CAND_* */ + __u32 index; /* the smaller the index, the higher the + * priority */ + struct sockaddr bssid; +}; + /* ------------------------ WIRELESS STATS ------------------------ */ /* * Wireless statistics (used for /proc/net/wireless) @@ -725,6 +1000,8 @@ struct iw_range struct iw_freq freq[IW_MAX_FREQUENCIES]; /* list */ /* Note : this frequency list doesn't need to fit channel numbers, * because each entry contain its channel index */ + + __u32 enc_capa; /* IW_ENC_CAPA_* bit field */ }; /* diff --git a/net/core/wireless.c b/net/core/wireless.c index 750cc5daeb03..b2fe378dfbf8 100644 --- a/net/core/wireless.c +++ b/net/core/wireless.c @@ -2,7 +2,7 @@ * This file implement the Wireless Extensions APIs. * * Authors : Jean Tourrilhes - HPL - - * Copyright (c) 1997-2004 Jean Tourrilhes, All Rights Reserved. + * Copyright (c) 1997-2005 Jean Tourrilhes, All Rights Reserved. * * (As all part of the Linux kernel, this file is GPL) */ @@ -187,6 +187,12 @@ static const struct iw_ioctl_description standard_ioctl[] = { .header_type = IW_HEADER_TYPE_ADDR, .flags = IW_DESCR_FLAG_DUMP, }, + [SIOCSIWMLME - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = sizeof(struct iw_mlme), + .max_tokens = sizeof(struct iw_mlme), + }, [SIOCGIWAPLIST - SIOCIWFIRST] = { .header_type = IW_HEADER_TYPE_POINT, .token_size = sizeof(struct sockaddr) + @@ -195,7 +201,10 @@ static const struct iw_ioctl_description standard_ioctl[] = { .flags = IW_DESCR_FLAG_NOMAX, }, [SIOCSIWSCAN - SIOCIWFIRST] = { - .header_type = IW_HEADER_TYPE_PARAM, + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = 0, + .max_tokens = sizeof(struct iw_scan_req), }, [SIOCGIWSCAN - SIOCIWFIRST] = { .header_type = IW_HEADER_TYPE_POINT, @@ -273,6 +282,42 @@ static const struct iw_ioctl_description standard_ioctl[] = { [SIOCGIWPOWER - SIOCIWFIRST] = { .header_type = IW_HEADER_TYPE_PARAM, }, + [SIOCSIWGENIE - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [SIOCGIWGENIE - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [SIOCSIWAUTH - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_PARAM, + }, + [SIOCGIWAUTH - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_PARAM, + }, + [SIOCSIWENCODEEXT - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = sizeof(struct iw_encode_ext), + .max_tokens = sizeof(struct iw_encode_ext) + + IW_ENCODING_TOKEN_MAX, + }, + [SIOCGIWENCODEEXT - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = sizeof(struct iw_encode_ext), + .max_tokens = sizeof(struct iw_encode_ext) + + IW_ENCODING_TOKEN_MAX, + }, + [SIOCSIWPMKSA - SIOCIWFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .min_tokens = sizeof(struct iw_pmksa), + .max_tokens = sizeof(struct iw_pmksa), + }, }; static const int standard_ioctl_num = (sizeof(standard_ioctl) / sizeof(struct iw_ioctl_description)); @@ -299,6 +344,31 @@ static const struct iw_ioctl_description standard_event[] = { [IWEVEXPIRED - IWEVFIRST] = { .header_type = IW_HEADER_TYPE_ADDR, }, + [IWEVGENIE - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [IWEVMICHAELMICFAILURE - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = sizeof(struct iw_michaelmicfailure), + }, + [IWEVASSOCREQIE - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [IWEVASSOCRESPIE - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = IW_GENERIC_IE_MAX, + }, + [IWEVPMKIDCAND - IWEVFIRST] = { + .header_type = IW_HEADER_TYPE_POINT, + .token_size = 1, + .max_tokens = sizeof(struct iw_pmkid_cand), + }, }; static const int standard_event_num = (sizeof(standard_event) / sizeof(struct iw_ioctl_description)); -- cgit v1.2.3 From c04049939f88b29e235d2da217bce6e8ead44f32 Mon Sep 17 00:00:00 2001 From: Steve Grubb Date: Fri, 13 May 2005 18:17:42 +0100 Subject: AUDIT: Add message types to audit records This patch adds more messages types to the audit subsystem so that audit analysis is quicker, intuitive, and more useful. Signed-off-by: Steve Grubb --- I forgot one type in the big patch. I need to add one for user space originating SE Linux avc messages. This is used by dbus and nscd. -Steve --- Updated to 2.6.12-rc4-mm1. -dwmw2 Signed-off-by: David Woodhouse --- include/linux/audit.h | 66 ++++++++++++++++++++++++++--------- kernel/audit.c | 78 +++++++++++++++++++++++++++++------------- kernel/auditsc.c | 42 ++++++++++++++--------- security/selinux/avc.c | 4 +-- security/selinux/hooks.c | 2 +- security/selinux/nlmsgtab.c | 8 +++++ security/selinux/ss/services.c | 4 +-- 7 files changed, 143 insertions(+), 61 deletions(-) (limited to 'include') diff --git a/include/linux/audit.h b/include/linux/audit.h index 405332ebf3c6..1a15ba38c660 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -27,15 +27,53 @@ #include #include -/* Request and reply types */ +/* The netlink messages for the audit system is divided into blocks: + * 1000 - 1099 are for commanding the audit system + * 1100 - 1199 user space trusted application messages + * 1200 - 1299 messages internal to the audit daemon + * 1300 - 1399 audit event messages + * 1400 - 1499 SE Linux use + * 1500 - 1999 future use + * 2000 is for otherwise unclassified kernel audit messages + * + * Messages from 1000-1199 are bi-directional. 1200-1299 are exclusively user + * space. Anything over that is kernel --> user space communication. + */ #define AUDIT_GET 1000 /* Get status */ #define AUDIT_SET 1001 /* Set status (enable/disable/auditd) */ -#define AUDIT_LIST 1002 /* List filtering rules */ -#define AUDIT_ADD 1003 /* Add filtering rule */ -#define AUDIT_DEL 1004 /* Delete filtering rule */ -#define AUDIT_USER 1005 /* Send a message from user-space */ +#define AUDIT_LIST 1002 /* List syscall filtering rules */ +#define AUDIT_ADD 1003 /* Add syscall filtering rule */ +#define AUDIT_DEL 1004 /* Delete syscall filtering rule */ +#define AUDIT_USER 1005 /* Message from userspace -- deprecated */ #define AUDIT_LOGIN 1006 /* Define the login id and information */ -#define AUDIT_SIGNAL_INFO 1010 /* Get information about sender of signal*/ +#define AUDIT_WATCH_INS 1007 /* Insert file/dir watch entry */ +#define AUDIT_WATCH_REM 1008 /* Remove file/dir watch entry */ +#define AUDIT_WATCH_LIST 1009 /* List all file/dir watches */ +#define AUDIT_SIGNAL_INFO 1010 /* Get info about sender of signal to auditd */ + +#define AUDIT_USER_AUTH 1100 /* User space authentication */ +#define AUDIT_USER_ACCT 1101 /* User space acct change */ +#define AUDIT_USER_MGMT 1102 /* User space acct management */ +#define AUDIT_CRED_ACQ 1103 /* User space credential acquired */ +#define AUDIT_CRED_DISP 1104 /* User space credential disposed */ +#define AUDIT_USER_START 1105 /* User space session start */ +#define AUDIT_USER_END 1106 /* User space session end */ +#define AUDIT_USER_AVC 1107 /* User space avc message */ + +#define AUDIT_DAEMON_START 1200 /* Daemon startup record */ +#define AUDIT_DAEMON_END 1201 /* Daemon normal stop record */ +#define AUDIT_DAEMON_ABORT 1202 /* Daemon error stop record */ +#define AUDIT_DAEMON_CONFIG 1203 /* Daemon config change */ + +#define AUDIT_SYSCALL 1300 /* Syscall event */ +#define AUDIT_FS_WATCH 1301 /* Filesystem watch event */ +#define AUDIT_PATH 1302 /* Filname path information */ +#define AUDIT_IPC 1303 /* IPC record */ +#define AUDIT_SOCKET 1304 /* Socket record */ +#define AUDIT_CONFIG_CHANGE 1305 /* Audit system configuration change */ + +#define AUDIT_AVC 1400 /* SE Linux avc denial or grant */ +#define AUDIT_SELINUX_ERR 1401 /* Internal SE Linux Errors */ #define AUDIT_KERNEL 2000 /* Asynchronous audit record. NOT A REQUEST. */ @@ -216,14 +254,11 @@ extern void audit_signal_info(int sig, struct task_struct *t); #ifdef CONFIG_AUDIT /* These are defined in audit.c */ /* Public API */ -#define audit_log(ctx, fmt, args...) \ - audit_log_type(ctx, AUDIT_KERNEL, 0, fmt, ##args) -extern void audit_log_type(struct audit_context *ctx, int type, - int pid, const char *fmt, ...) - __attribute__((format(printf,4,5))); +extern void audit_log(struct audit_context *ctx, int type, + const char *fmt, ...) + __attribute__((format(printf,3,4))); -extern struct audit_buffer *audit_log_start(struct audit_context *ctx, int type, - int pid); +extern struct audit_buffer *audit_log_start(struct audit_context *ctx,int type); extern void audit_log_format(struct audit_buffer *ab, const char *fmt, ...) __attribute__((format(printf,2,3))); @@ -243,9 +278,8 @@ extern void audit_send_reply(int pid, int seq, int type, void *payload, int size); extern void audit_log_lost(const char *message); #else -#define audit_log(c,f,...) do { ; } while (0) -#define audit_log_type(c,t,p,f,...) do { ; } while (0) -#define audit_log_start(c,t,p) ({ NULL; }) +#define audit_log(c,t,f,...) do { ; } while (0) +#define audit_log_start(c,t) ({ NULL; }) #define audit_log_vformat(b,f,a) do { ; } while (0) #define audit_log_format(b,f,...) do { ; } while (0) #define audit_log_end(b) do { ; } while (0) diff --git a/kernel/audit.c b/kernel/audit.c index 187164572bd0..4e940c05ede8 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -140,6 +140,12 @@ struct audit_buffer { struct audit_context *ctx; /* NULL or associated context */ }; +static void audit_set_pid(struct audit_buffer *ab, pid_t pid) +{ + struct nlmsghdr *nlh = (struct nlmsghdr *)ab->skb->data; + nlh->nlmsg_pid = pid; +} + struct audit_entry { struct list_head list; struct audit_rule rule; @@ -233,7 +239,8 @@ static int audit_set_rate_limit(int limit, uid_t loginuid) { int old = audit_rate_limit; audit_rate_limit = limit; - audit_log(NULL, "audit_rate_limit=%d old=%d by auid %u", + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "audit_rate_limit=%d old=%d by auid %u", audit_rate_limit, old, loginuid); return old; } @@ -242,7 +249,8 @@ static int audit_set_backlog_limit(int limit, uid_t loginuid) { int old = audit_backlog_limit; audit_backlog_limit = limit; - audit_log(NULL, "audit_backlog_limit=%d old=%d by auid %u", + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "audit_backlog_limit=%d old=%d by auid %u", audit_backlog_limit, old, loginuid); return old; } @@ -253,8 +261,9 @@ static int audit_set_enabled(int state, uid_t loginuid) if (state != 0 && state != 1) return -EINVAL; audit_enabled = state; - audit_log(NULL, "audit_enabled=%d old=%d by auid %u", - audit_enabled, old, loginuid); + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "audit_enabled=%d old=%d by auid %u", + audit_enabled, old, loginuid); return old; } @@ -266,8 +275,9 @@ static int audit_set_failure(int state, uid_t loginuid) && state != AUDIT_FAIL_PANIC) return -EINVAL; audit_failure = state; - audit_log(NULL, "audit_failure=%d old=%d by auid %u", - audit_failure, old, loginuid); + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "audit_failure=%d old=%d by auid %u", + audit_failure, old, loginuid); return old; } @@ -316,6 +326,14 @@ static int audit_netlink_ok(kernel_cap_t eff_cap, u16 msg_type) err = -EPERM; break; case AUDIT_USER: + case AUDIT_USER_AUTH: + case AUDIT_USER_ACCT: + case AUDIT_USER_MGMT: + case AUDIT_CRED_ACQ: + case AUDIT_CRED_DISP: + case AUDIT_USER_START: + case AUDIT_USER_END: + case AUDIT_USER_AVC: if (!cap_raised(eff_cap, CAP_AUDIT_WRITE)) err = -EPERM; break; @@ -332,6 +350,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) void *data; struct audit_status *status_get, status_set; int err; + struct audit_buffer *ab; u16 msg_type = nlh->nlmsg_type; uid_t loginuid; /* loginuid of sender */ struct audit_sig_info sig_data; @@ -373,7 +392,8 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) if (status_get->mask & AUDIT_STATUS_PID) { int old = audit_pid; audit_pid = status_get->pid; - audit_log(NULL, "audit_pid=%d old=%d by auid %u", + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "audit_pid=%d old=%d by auid %u", audit_pid, old, loginuid); } if (status_get->mask & AUDIT_STATUS_RATE_LIMIT) @@ -383,13 +403,26 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) loginuid); break; case AUDIT_USER: - audit_log_type(NULL, AUDIT_USER, pid, + case AUDIT_USER_AUTH: + case AUDIT_USER_ACCT: + case AUDIT_USER_MGMT: + case AUDIT_CRED_ACQ: + case AUDIT_CRED_DISP: + case AUDIT_USER_START: + case AUDIT_USER_END: + case AUDIT_USER_AVC: + ab = audit_log_start(NULL, msg_type); + if (!ab) + break; /* audit_panic has been called */ + audit_log_format(ab, "user pid=%d uid=%d length=%d loginuid=%u" " msg='%.1024s'", pid, uid, (int)(nlh->nlmsg_len - ((char *)data - (char *)nlh)), loginuid, (char *)data); + audit_set_pid(ab, pid); + audit_log_end(ab); break; case AUDIT_ADD: case AUDIT_DEL: @@ -504,7 +537,7 @@ static int __init audit_init(void) audit_initialized = 1; audit_enabled = audit_default; - audit_log(NULL, "initialized"); + audit_log(NULL, AUDIT_KERNEL, "initialized"); return 0; } __initcall(audit_init); @@ -541,10 +574,12 @@ static void audit_buffer_free(struct audit_buffer *ab) spin_unlock_irqrestore(&audit_freelist_lock, flags); } -static struct audit_buffer * audit_buffer_alloc(int gfp_mask) +static struct audit_buffer * audit_buffer_alloc(struct audit_context *ctx, + int gfp_mask, int type) { unsigned long flags; struct audit_buffer *ab = NULL; + struct nlmsghdr *nlh; spin_lock_irqsave(&audit_freelist_lock, flags); if (!list_empty(&audit_freelist)) { @@ -566,6 +601,12 @@ static struct audit_buffer * audit_buffer_alloc(int gfp_mask) if (!ab->skb) goto err; + ab->ctx = ctx; + nlh = (struct nlmsghdr *)skb_put(ab->skb, NLMSG_SPACE(0)); + nlh->nlmsg_type = type; + nlh->nlmsg_flags = 0; + nlh->nlmsg_pid = 0; + nlh->nlmsg_seq = 0; return ab; err: audit_buffer_free(ab); @@ -578,12 +619,11 @@ err: * syscall, then the syscall is marked as auditable and an audit record * will be written at syscall exit. If there is no associated task, tsk * should be NULL. */ -struct audit_buffer *audit_log_start(struct audit_context *ctx, int type, int pid) +struct audit_buffer *audit_log_start(struct audit_context *ctx, int type) { struct audit_buffer *ab = NULL; struct timespec t; unsigned int serial; - struct nlmsghdr *nlh; if (!audit_initialized) return NULL; @@ -600,19 +640,12 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, int type, int pi return NULL; } - ab = audit_buffer_alloc(GFP_ATOMIC); + ab = audit_buffer_alloc(ctx, GFP_ATOMIC, type); if (!ab) { audit_log_lost("out of memory in audit_log_start"); return NULL; } - ab->ctx = ctx; - nlh = (struct nlmsghdr *)skb_put(ab->skb, NLMSG_SPACE(0)); - nlh->nlmsg_type = type; - nlh->nlmsg_flags = 0; - nlh->nlmsg_pid = pid; - nlh->nlmsg_seq = 0; - if (!audit_get_stamp(ab->ctx, &t, &serial)) { t = CURRENT_TIME; serial = 0; @@ -809,13 +842,12 @@ void audit_log_end(struct audit_buffer *ab) /* Log an audit record. This is a convenience function that calls * audit_log_start, audit_log_vformat, and audit_log_end. It may be * called in any context. */ -void audit_log_type(struct audit_context *ctx, int type, int pid, - const char *fmt, ...) +void audit_log(struct audit_context *ctx, int type, const char *fmt, ...) { struct audit_buffer *ab; va_list args; - ab = audit_log_start(ctx, type, pid); + ab = audit_log_start(ctx, type); if (ab) { va_start(args, fmt); audit_log_vformat(ab, fmt, args); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index d089263253a7..1b7c91f9d5ff 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -286,7 +286,8 @@ int audit_receive_filter(int type, int pid, int uid, int seq, void *data, err = audit_add_rule(entry, &audit_entlist); if (!err && (flags & AUDIT_AT_EXIT)) err = audit_add_rule(entry, &audit_extlist); - audit_log(NULL, "auid %u added an audit rule\n", loginuid); + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "auid %u added an audit rule\n", loginuid); break; case AUDIT_DEL: flags =((struct audit_rule *)data)->flags; @@ -296,7 +297,8 @@ int audit_receive_filter(int type, int pid, int uid, int seq, void *data, err = audit_del_rule(data, &audit_entlist); if (!err && (flags & AUDIT_AT_EXIT)) err = audit_del_rule(data, &audit_extlist); - audit_log(NULL, "auid %u removed an audit rule\n", loginuid); + audit_log(NULL, AUDIT_CONFIG_CHANGE, + "auid %u removed an audit rule\n", loginuid); break; default: return -EINVAL; @@ -648,7 +650,7 @@ static void audit_log_exit(struct audit_context *context) int i; struct audit_buffer *ab; - ab = audit_log_start(context, AUDIT_KERNEL, 0); + ab = audit_log_start(context, AUDIT_SYSCALL); if (!ab) return; /* audit_panic has been called */ audit_log_format(ab, "syscall=%d", context->major); @@ -680,28 +682,28 @@ static void audit_log_exit(struct audit_context *context) while (context->aux) { struct audit_aux_data *aux; - ab = audit_log_start(context, AUDIT_KERNEL, 0); + aux = context->aux; + + ab = audit_log_start(context, aux->type); if (!ab) continue; /* audit_panic has been called */ - aux = context->aux; - context->aux = aux->next; - - audit_log_format(ab, "auxitem=%d", aux->type); switch (aux->type) { - case AUDIT_AUX_IPCPERM: { + case AUDIT_IPC: { struct audit_aux_data_ipcctl *axi = (void *)aux; audit_log_format(ab, - " qbytes=%lx uid=%d gid=%d mode=%x", + " qbytes=%lx iuid=%d igid=%d mode=%x", axi->qbytes, axi->uid, axi->gid, axi->mode); } } audit_log_end(ab); + + context->aux = aux->next; kfree(aux); } for (i = 0; i < context->name_count; i++) { - ab = audit_log_start(context, AUDIT_KERNEL, 0); + ab = audit_log_start(context, AUDIT_PATH); if (!ab) continue; /* audit_panic has been called */ audit_log_format(ab, "item=%d", i); @@ -711,7 +713,7 @@ static void audit_log_exit(struct audit_context *context) } if (context->names[i].ino != (unsigned long)-1) audit_log_format(ab, " inode=%lu dev=%02x:%02x mode=%#o" - " uid=%d gid=%d rdev=%02x:%02x", + " ouid=%d ogid=%d rdev=%02x:%02x", context->names[i].ino, MAJOR(context->names[i].dev), MINOR(context->names[i].dev), @@ -1008,10 +1010,16 @@ int audit_get_stamp(struct audit_context *ctx, int audit_set_loginuid(struct task_struct *task, uid_t loginuid) { if (task->audit_context) { - audit_log_type(NULL, AUDIT_LOGIN, 0, - "login pid=%d uid=%u old loginuid=%u new loginuid=%u", - task->pid, task->uid, task->audit_context->loginuid, - loginuid); + struct audit_buffer *ab; + + ab = audit_log_start(NULL, AUDIT_LOGIN); + if (ab) { + audit_log_format(ab, "login pid=%d uid=%u " + "old loginuid=%u new loginuid=%u", + task->pid, task->uid, + task->audit_context->loginuid, loginuid); + audit_log_end(ab); + } task->audit_context->loginuid = loginuid; } return 0; @@ -1039,7 +1047,7 @@ int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode) ax->gid = gid; ax->mode = mode; - ax->d.type = AUDIT_AUX_IPCPERM; + ax->d.type = AUDIT_IPC; ax->d.next = context->aux; context->aux = (void *)ax; return 0; diff --git a/security/selinux/avc.c b/security/selinux/avc.c index 9e71a1bbe011..042f91e9f9d2 100644 --- a/security/selinux/avc.c +++ b/security/selinux/avc.c @@ -242,7 +242,7 @@ void __init avc_init(void) avc_node_cachep = kmem_cache_create("avc_node", sizeof(struct avc_node), 0, SLAB_PANIC, NULL, NULL); - audit_log(current->audit_context, "AVC INITIALIZED\n"); + audit_log(current->audit_context, AUDIT_KERNEL, "AVC INITIALIZED\n"); } int avc_get_hash_stats(char *page) @@ -549,7 +549,7 @@ void avc_audit(u32 ssid, u32 tsid, return; } - ab = audit_log_start(current->audit_context, AUDIT_KERNEL, 0); + ab = audit_log_start(current->audit_context, AUDIT_AVC); if (!ab) return; /* audit_panic has been called */ audit_log_format(ab, "avc: %s ", denied ? "denied" : "granted"); diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index aae1e794fe48..db845cbd5841 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -3419,7 +3419,7 @@ static int selinux_nlmsg_perm(struct sock *sk, struct sk_buff *skb) err = selinux_nlmsg_lookup(isec->sclass, nlh->nlmsg_type, &perm); if (err) { if (err == -EINVAL) { - audit_log(current->audit_context, + audit_log(current->audit_context, AUDIT_SELINUX_ERR, "SELinux: unrecognized netlink message" " type=%hu for sclass=%hu\n", nlh->nlmsg_type, isec->sclass); diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c index deac14367d43..67e77acc4795 100644 --- a/security/selinux/nlmsgtab.c +++ b/security/selinux/nlmsgtab.c @@ -98,6 +98,14 @@ static struct nlmsg_perm nlmsg_audit_perms[] = { AUDIT_DEL, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, { AUDIT_USER, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, { AUDIT_SIGNAL_INFO, NETLINK_AUDIT_SOCKET__NLMSG_READ }, + { AUDIT_USER_AUTH, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_USER_ACCT, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_USER_MGMT, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_CRED_ACQ, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_CRED_DISP, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_USER_START, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_USER_END, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, + { AUDIT_USER_AVC, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, }; diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index 5a820cf88c9c..07fdf6ee6148 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -365,7 +365,7 @@ static int security_validtrans_handle_fail(struct context *ocontext, goto out; if (context_struct_to_string(tcontext, &t, &tlen) < 0) goto out; - audit_log(current->audit_context, + audit_log(current->audit_context, AUDIT_SELINUX_ERR, "security_validate_transition: denied for" " oldcontext=%s newcontext=%s taskcontext=%s tclass=%s", o, n, t, policydb.p_class_val_to_name[tclass-1]); @@ -742,7 +742,7 @@ static int compute_sid_handle_invalid_context( goto out; if (context_struct_to_string(newcontext, &n, &nlen) < 0) goto out; - audit_log(current->audit_context, + audit_log(current->audit_context, AUDIT_SELINUX_ERR, "security_compute_sid: invalid context %s" " for scontext=%s" " tcontext=%s" -- cgit v1.2.3 From 23f32d18aa589e228c5a9e12e0d0c67c9b5bcdce Mon Sep 17 00:00:00 2001 From: Steve Grubb Date: Fri, 13 May 2005 18:35:15 +0100 Subject: AUDIT: Fix some spelling errors I'm going through the kernel code and have a patch that corrects several spelling errors in comments. From: Steve Grubb Signed-off-by: David Woodhouse --- include/linux/audit.h | 2 +- kernel/audit.c | 4 ++-- kernel/auditsc.c | 4 ++-- 3 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/audit.h b/include/linux/audit.h index 1a15ba38c660..51e5879af7fc 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -181,7 +181,7 @@ struct audit_message { struct audit_status { __u32 mask; /* Bit mask for valid entries */ - __u32 enabled; /* 1 = enabled, 0 = disbaled */ + __u32 enabled; /* 1 = enabled, 0 = disabled */ __u32 failure; /* Failure-to-log action */ __u32 pid; /* pid of auditd process */ __u32 rate_limit; /* messages rate limit (per second) */ diff --git a/kernel/audit.c b/kernel/audit.c index 4e940c05ede8..74779d3769fa 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -116,7 +116,7 @@ static LIST_HEAD(audit_entlist); static LIST_HEAD(audit_extlist); /* The netlink socket is only to be read by 1 CPU, which lets us assume - * that list additions and deletions never happen simultaneiously in + * that list additions and deletions never happen simultaneously in * auditsc.c */ static DECLARE_MUTEX(audit_netlink_sem); @@ -775,7 +775,7 @@ void audit_log_d_path(struct audit_buffer *ab, const char *prefix, } } -/* Remove queued messages from the audit_txlist and send them to userspace. */ +/* Remove queued messages from the audit_txlist and send them to user space. */ static void audit_tasklet_handler(unsigned long arg) { LIST_HEAD(list); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 1b7c91f9d5ff..773d28a3f701 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -444,7 +444,7 @@ static enum audit_state audit_filter_task(struct task_struct *tsk) /* At syscall entry and exit time, this filter is called if the * audit_state is not low enough that auditing cannot take place, but is - * also not high enough that we already know we have to write and audit + * also not high enough that we already know we have to write an audit * record (i.e., the state is AUDIT_SETUP_CONTEXT or AUDIT_BUILD_CONTEXT). */ static enum audit_state audit_filter_syscall(struct task_struct *tsk, @@ -750,7 +750,7 @@ void audit_free(struct task_struct *tsk) /* Compute a serial number for the audit record. Audit records are * written to user-space as soon as they are generated, so a complete * audit record may be written in several pieces. The timestamp of the - * record and this serial number are used by the user-space daemon to + * record and this serial number are used by the user-space tools to * determine which pieces belong to the same audit record. The * (timestamp,serial) tuple is unique for each syscall and is live from * syscall entry to syscall exit. -- cgit v1.2.3 From a1365275e745bb0a173c918a52bcdfa6ce122f7e Mon Sep 17 00:00:00 2001 From: Sascha Hauer Date: Thu, 5 May 2005 15:14:15 -0700 Subject: [PATCH] DM9000 network driver This patch adds support for the davicom dm9000 network driver. The dm9000 is found on some embedded arm boards such as the pimx1 or the scb9328. Signed-off-by: Sascha Hauer Signed-off-by: Ben Dooks Signed-off-by: Andrew Morton diff -puN /dev/null drivers/net/dm9000.c --- drivers/net/Kconfig | 12 + drivers/net/Makefile | 1 + drivers/net/dm9000.c | 1219 ++++++++++++++++++++++++++++++++++++++++++++++++ drivers/net/dm9000.h | 135 ++++++ include/linux/dm9000.h | 36 ++ 5 files changed, 1403 insertions(+) create mode 100644 drivers/net/dm9000.c create mode 100644 drivers/net/dm9000.h create mode 100644 include/linux/dm9000.h (limited to 'include') diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 3a0a55b62aaf..4d0c1fb15c7c 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -824,6 +824,18 @@ config SMC9194 . The module will be called smc9194. +config DM9000 + tristate "DM9000 support" + depends on ARM && NET_ETHERNET + select CRC32 + select MII + ---help--- + Support for DM9000 chipset. + + To compile this driver as a module, choose M here and read + . The module will be + called dm9000. + config NET_VENDOR_RACAL bool "Racal-Interlan (Micom) NI cards" depends on NET_ETHERNET && ISA diff --git a/drivers/net/Makefile b/drivers/net/Makefile index 6202b10dbb4d..cbd11688608e 100644 --- a/drivers/net/Makefile +++ b/drivers/net/Makefile @@ -179,6 +179,7 @@ obj-$(CONFIG_AMD8111_ETH) += amd8111e.o obj-$(CONFIG_IBMVETH) += ibmveth.o obj-$(CONFIG_S2IO) += s2io.o obj-$(CONFIG_SMC91X) += smc91x.o +obj-$(CONFIG_DM9000) += dm9000.o obj-$(CONFIG_FEC_8XX) += fec_8xx/ obj-$(CONFIG_ARM) += arm/ diff --git a/drivers/net/dm9000.c b/drivers/net/dm9000.c new file mode 100644 index 000000000000..f4ba0ffb8637 --- /dev/null +++ b/drivers/net/dm9000.c @@ -0,0 +1,1219 @@ +/* + * dm9000.c: Version 1.2 03/18/2003 + * + * A Davicom DM9000 ISA NIC fast Ethernet driver for Linux. + * Copyright (C) 1997 Sten Wang + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * (C)Copyright 1997-1998 DAVICOM Semiconductor,Inc. All Rights Reserved. + * + * V0.11 06/20/2001 REG_0A bit3=1, default enable BP with DA match + * 06/22/2001 Support DM9801 progrmming + * E3: R25 = ((R24 + NF) & 0x00ff) | 0xf000 + * E4: R25 = ((R24 + NF) & 0x00ff) | 0xc200 + * R17 = (R17 & 0xfff0) | NF + 3 + * E5: R25 = ((R24 + NF - 3) & 0x00ff) | 0xc200 + * R17 = (R17 & 0xfff0) | NF + * + * v1.00 modify by simon 2001.9.5 + * change for kernel 2.4.x + * + * v1.1 11/09/2001 fix force mode bug + * + * v1.2 03/18/2003 Weilun Huang : + * Fixed phy reset. + * Added tx/rx 32 bit mode. + * Cleaned up for kernel merge. + * + * 03/03/2004 Sascha Hauer + * Port to 2.6 kernel + * + * 24-Sep-2004 Ben Dooks + * Cleanup of code to remove ifdefs + * Allowed platform device data to influence access width + * Reformatting areas of code + * + * 17-Mar-2005 Sascha Hauer + * * removed 2.4 style module parameters + * * removed removed unused stat counter and fixed + * net_device_stats + * * introduced tx_timeout function + * * reworked locking + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "dm9000.h" + +/* Board/System/Debug information/definition ---------------- */ + +#define DM9000_PHY 0x40 /* PHY address 0x01 */ + +#define TRUE 1 +#define FALSE 0 + +#define CARDNAME "dm9000" +#define PFX CARDNAME ": " + +#define DM9000_TIMER_WUT jiffies+(HZ*2) /* timer wakeup time : 2 second */ + +#define DM9000_DEBUG 0 + +#if DM9000_DEBUG > 2 +#define PRINTK3(args...) printk(CARDNAME ": " args) +#else +#define PRINTK3(args...) do { } while(0) +#endif + +#if DM9000_DEBUG > 1 +#define PRINTK2(args...) printk(CARDNAME ": " args) +#else +#define PRINTK2(args...) do { } while(0) +#endif + +#if DM9000_DEBUG > 0 +#define PRINTK1(args...) printk(CARDNAME ": " args) +#define PRINTK(args...) printk(CARDNAME ": " args) +#else +#define PRINTK1(args...) do { } while(0) +#define PRINTK(args...) printk(KERN_DEBUG args) +#endif + +/* + * Transmit timeout, default 5 seconds. + */ +static int watchdog = 5000; +module_param(watchdog, int, 0400); +MODULE_PARM_DESC(watchdog, "transmit timeout in milliseconds"); + +/* Structure/enum declaration ------------------------------- */ +typedef struct board_info { + + void __iomem *io_addr; /* Register I/O base address */ + void __iomem *io_data; /* Data I/O address */ + u16 irq; /* IRQ */ + + u16 tx_pkt_cnt; + u16 queue_pkt_len; + u16 queue_start_addr; + u16 dbug_cnt; + u8 io_mode; /* 0:word, 2:byte */ + u8 phy_addr; + + void (*inblk)(void __iomem *port, void *data, int length); + void (*outblk)(void __iomem *port, void *data, int length); + void (*dumpblk)(void __iomem *port, int length); + + struct resource *addr_res; /* resources found */ + struct resource *data_res; + struct resource *addr_req; /* resources requested */ + struct resource *data_req; + struct resource *irq_res; + + struct timer_list timer; + struct net_device_stats stats; + unsigned char srom[128]; + spinlock_t lock; + + struct mii_if_info mii; + u32 msg_enable; +} board_info_t; + +/* function declaration ------------------------------------- */ +static int dm9000_probe(struct device *); +static int dm9000_open(struct net_device *); +static int dm9000_start_xmit(struct sk_buff *, struct net_device *); +static int dm9000_stop(struct net_device *); +static int dm9000_do_ioctl(struct net_device *, struct ifreq *, int); + + +static void dm9000_timer(unsigned long); +static void dm9000_init_dm9000(struct net_device *); + +static struct net_device_stats *dm9000_get_stats(struct net_device *); + +static irqreturn_t dm9000_interrupt(int, void *, struct pt_regs *); + +static int dm9000_phy_read(struct net_device *dev, int phyaddr_unsused, int reg); +static void dm9000_phy_write(struct net_device *dev, int phyaddr_unused, int reg, + int value); +static u16 read_srom_word(board_info_t *, int); +static void dm9000_rx(struct net_device *); +static void dm9000_hash_table(struct net_device *); + +//#define DM9000_PROGRAM_EEPROM +#ifdef DM9000_PROGRAM_EEPROM +static void program_eeprom(board_info_t * db); +#endif +/* DM9000 network board routine ---------------------------- */ + +static void +dm9000_reset(board_info_t * db) +{ + PRINTK1("dm9000x: resetting\n"); + /* RESET device */ + writeb(DM9000_NCR, db->io_addr); + udelay(200); + writeb(NCR_RST, db->io_data); + udelay(200); +} + +/* + * Read a byte from I/O port + */ +static u8 +ior(board_info_t * db, int reg) +{ + writeb(reg, db->io_addr); + return readb(db->io_data); +} + +/* + * Write a byte to I/O port + */ + +static void +iow(board_info_t * db, int reg, int value) +{ + writeb(reg, db->io_addr); + writeb(value, db->io_data); +} + +/* routines for sending block to chip */ + +static void dm9000_outblk_8bit(void __iomem *reg, void *data, int count) +{ + writesb(reg, data, count); +} + +static void dm9000_outblk_16bit(void __iomem *reg, void *data, int count) +{ + writesw(reg, data, (count+1) >> 1); +} + +static void dm9000_outblk_32bit(void __iomem *reg, void *data, int count) +{ + writesl(reg, data, (count+3) >> 2); +} + +/* input block from chip to memory */ + +static void dm9000_inblk_8bit(void __iomem *reg, void *data, int count) +{ + readsb(reg, data, count+1); +} + + +static void dm9000_inblk_16bit(void __iomem *reg, void *data, int count) +{ + readsw(reg, data, (count+1) >> 1); +} + +static void dm9000_inblk_32bit(void __iomem *reg, void *data, int count) +{ + readsl(reg, data, (count+3) >> 2); +} + +/* dump block from chip to null */ + +static void dm9000_dumpblk_8bit(void __iomem *reg, int count) +{ + int i; + int tmp; + + for (i = 0; i < count; i++) + tmp = readb(reg); +} + +static void dm9000_dumpblk_16bit(void __iomem *reg, int count) +{ + int i; + int tmp; + + count = (count + 1) >> 1; + + for (i = 0; i < count; i++) + tmp = readw(reg); +} + +static void dm9000_dumpblk_32bit(void __iomem *reg, int count) +{ + int i; + int tmp; + + count = (count + 3) >> 2; + + for (i = 0; i < count; i++) + tmp = readl(reg); +} + +/* dm9000_set_io + * + * select the specified set of io routines to use with the + * device + */ + +static void dm9000_set_io(struct board_info *db, int byte_width) +{ + /* use the size of the data resource to work out what IO + * routines we want to use + */ + + switch (byte_width) { + case 1: + db->dumpblk = dm9000_dumpblk_8bit; + db->outblk = dm9000_outblk_8bit; + db->inblk = dm9000_inblk_8bit; + break; + + case 2: + db->dumpblk = dm9000_dumpblk_16bit; + db->outblk = dm9000_outblk_16bit; + db->inblk = dm9000_inblk_16bit; + break; + + case 3: + printk(KERN_ERR PFX ": 3 byte IO, falling back to 16bit\n"); + db->dumpblk = dm9000_dumpblk_16bit; + db->outblk = dm9000_outblk_16bit; + db->inblk = dm9000_inblk_16bit; + break; + + case 4: + default: + db->dumpblk = dm9000_dumpblk_32bit; + db->outblk = dm9000_outblk_32bit; + db->inblk = dm9000_inblk_32bit; + break; + } +} + + +/* Our watchdog timed out. Called by the networking layer */ +static void dm9000_timeout(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + u8 reg_save; + unsigned long flags; + + /* Save previous register address */ + reg_save = readb(db->io_addr); + spin_lock_irqsave(db->lock,flags); + + netif_stop_queue(dev); + dm9000_reset(db); + dm9000_init_dm9000(dev); + /* We can accept TX packets again */ + dev->trans_start = jiffies; + netif_wake_queue(dev); + + /* Restore previous register address */ + writeb(reg_save, db->io_addr); + spin_unlock_irqrestore(db->lock,flags); +} + + +/* dm9000_release_board + * + * release a board, and any mapped resources + */ + +static void +dm9000_release_board(struct platform_device *pdev, struct board_info *db) +{ + if (db->data_res == NULL) { + if (db->addr_res != NULL) + release_mem_region((unsigned long)db->io_addr, 4); + return; + } + + /* unmap our resources */ + + iounmap(db->io_addr); + iounmap(db->io_data); + + /* release the resources */ + + if (db->data_req != NULL) { + release_resource(db->data_req); + kfree(db->data_req); + } + + if (db->addr_res != NULL) { + release_resource(db->data_req); + kfree(db->addr_req); + } +} + +#define res_size(_r) (((_r)->end - (_r)->start) + 1) + +/* + * Search DM9000 board, allocate space and register it + */ +static int +dm9000_probe(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct dm9000_plat_data *pdata = pdev->dev.platform_data; + struct board_info *db; /* Point a board information structure */ + struct net_device *ndev; + unsigned long base; + int ret = 0; + int iosize; + int i; + u32 id_val; + + printk(KERN_INFO "%s Ethernet Driver\n", CARDNAME); + + /* Init network device */ + ndev = alloc_etherdev(sizeof (struct board_info)); + if (!ndev) { + printk("%s: could not allocate device.\n", CARDNAME); + return -ENOMEM; + } + + SET_MODULE_OWNER(ndev); + SET_NETDEV_DEV(ndev, dev); + + PRINTK2("dm9000_probe()"); + + /* setup board info structure */ + db = (struct board_info *) ndev->priv; + memset(db, 0, sizeof (*db)); + + if (pdev->num_resources < 2) { + ret = -ENODEV; + goto out; + } + + switch (pdev->num_resources) { + case 2: + base = pdev->resource[0].start; + + if (!request_mem_region(base, 4, ndev->name)) { + ret = -EBUSY; + goto out; + } + + ndev->base_addr = base; + ndev->irq = pdev->resource[1].start; + db->io_addr = (void *)base; + db->io_data = (void *)(base + 4); + + break; + + case 3: + db->addr_res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + db->data_res = platform_get_resource(pdev, IORESOURCE_MEM, 1); + db->irq_res = platform_get_resource(pdev, IORESOURCE_IRQ, 0); + + if (db->addr_res == NULL || db->data_res == NULL) { + printk(KERN_ERR PFX "insufficient resources\n"); + ret = -ENOENT; + goto out; + } + + i = res_size(db->addr_res); + db->addr_req = request_mem_region(db->addr_res->start, i, + pdev->name); + + if (db->addr_req == NULL) { + printk(KERN_ERR PFX "cannot claim address reg area\n"); + ret = -EIO; + goto out; + } + + db->io_addr = ioremap(db->addr_res->start, i); + + if (db->io_addr == NULL) { + printk(KERN_ERR "failed to ioremap address reg\n"); + ret = -EINVAL; + goto out; + } + + iosize = res_size(db->data_res); + db->data_req = request_mem_region(db->data_res->start, iosize, + pdev->name); + + if (db->data_req == NULL) { + printk(KERN_ERR PFX "cannot claim data reg area\n"); + ret = -EIO; + goto out; + } + + db->io_data = ioremap(db->data_res->start, iosize); + + if (db->io_data == NULL) { + printk(KERN_ERR "failed to ioremap data reg\n"); + ret = -EINVAL; + goto out; + } + + /* fill in parameters for net-dev structure */ + + ndev->base_addr = (unsigned long)db->io_addr; + ndev->irq = db->irq_res->start; + + /* ensure at least we have a default set of IO routines */ + dm9000_set_io(db, iosize); + + } + + /* check to see if anything is being over-ridden */ + if (pdata != NULL) { + /* check to see if the driver wants to over-ride the + * default IO width */ + + if (pdata->flags & DM9000_PLATF_8BITONLY) + dm9000_set_io(db, 1); + + if (pdata->flags & DM9000_PLATF_16BITONLY) + dm9000_set_io(db, 2); + + if (pdata->flags & DM9000_PLATF_32BITONLY) + dm9000_set_io(db, 4); + + /* check to see if there are any IO routine + * over-rides */ + + if (pdata->inblk != NULL) + db->inblk = pdata->inblk; + + if (pdata->outblk != NULL) + db->outblk = pdata->outblk; + + if (pdata->dumpblk != NULL) + db->dumpblk = pdata->dumpblk; + } + + dm9000_reset(db); + + /* try two times, DM9000 sometimes gets the first read wrong */ + for (i = 0; i < 2; i++) { + id_val = ior(db, DM9000_VIDL); + id_val |= (u32)ior(db, DM9000_VIDH) << 8; + id_val |= (u32)ior(db, DM9000_PIDL) << 16; + id_val |= (u32)ior(db, DM9000_PIDH) << 24; + + if (id_val == DM9000_ID) + break; + printk("%s: read wrong id 0x%08x\n", CARDNAME, id_val); + } + + if (id_val != DM9000_ID) { + printk("%s: wrong id: 0x%08x\n", CARDNAME, id_val); + goto release; + } + + /* from this point we assume that we have found a DM9000 */ + + /* driver system function */ + ether_setup(ndev); + + ndev->open = &dm9000_open; + ndev->hard_start_xmit = &dm9000_start_xmit; + ndev->tx_timeout = &dm9000_timeout; + ndev->watchdog_timeo = msecs_to_jiffies(watchdog); + ndev->stop = &dm9000_stop; + ndev->get_stats = &dm9000_get_stats; + ndev->set_multicast_list = &dm9000_hash_table; + ndev->do_ioctl = &dm9000_do_ioctl; + +#ifdef DM9000_PROGRAM_EEPROM + program_eeprom(db); +#endif + db->msg_enable = NETIF_MSG_LINK; + db->mii.phy_id_mask = 0x1f; + db->mii.reg_num_mask = 0x1f; + db->mii.force_media = 0; + db->mii.full_duplex = 0; + db->mii.dev = ndev; + db->mii.mdio_read = dm9000_phy_read; + db->mii.mdio_write = dm9000_phy_write; + + /* Read SROM content */ + for (i = 0; i < 64; i++) + ((u16 *) db->srom)[i] = read_srom_word(db, i); + + /* Set Node Address */ + for (i = 0; i < 6; i++) + ndev->dev_addr[i] = db->srom[i]; + + if (!is_valid_ether_addr(ndev->dev_addr)) + printk("%s: Invalid ethernet MAC address. Please " + "set using ifconfig\n", ndev->name); + + dev_set_drvdata(dev, ndev); + ret = register_netdev(ndev); + + if (ret == 0) { + printk("%s: dm9000 at %p,%p IRQ %d MAC: ", + ndev->name, db->io_addr, db->io_data, ndev->irq); + for (i = 0; i < 5; i++) + printk("%02x:", ndev->dev_addr[i]); + printk("%02x\n", ndev->dev_addr[5]); + } + return 0; + + release: + out: + printk("%s: not found (%d).\n", CARDNAME, ret); + + dm9000_release_board(pdev, db); + kfree(ndev); + + return ret; +} + +/* + * Open the interface. + * The interface is opened whenever "ifconfig" actives it. + */ +static int +dm9000_open(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + + PRINTK2("entering dm9000_open\n"); + + if (request_irq(dev->irq, &dm9000_interrupt, SA_SHIRQ, dev->name, dev)) + return -EAGAIN; + + /* Initialize DM9000 board */ + dm9000_reset(db); + dm9000_init_dm9000(dev); + + /* Init driver variable */ + db->dbug_cnt = 0; + + /* set and active a timer process */ + init_timer(&db->timer); + db->timer.expires = DM9000_TIMER_WUT * 2; + db->timer.data = (unsigned long) dev; + db->timer.function = &dm9000_timer; + add_timer(&db->timer); + + mii_check_media(&db->mii, netif_msg_link(db), 1); + netif_start_queue(dev); + + return 0; +} + +/* + * Initilize dm9000 board + */ +static void +dm9000_init_dm9000(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + + PRINTK1("entering %s\n",__FUNCTION__); + + /* I/O mode */ + db->io_mode = ior(db, DM9000_ISR) >> 6; /* ISR bit7:6 keeps I/O mode */ + + /* GPIO0 on pre-activate PHY */ + iow(db, DM9000_GPR, 0); /* REG_1F bit0 activate phyxcer */ + iow(db, DM9000_GPCR, GPCR_GEP_CNTL); /* Let GPIO0 output */ + iow(db, DM9000_GPR, 0); /* Enable PHY */ + + /* Program operating register */ + iow(db, DM9000_TCR, 0); /* TX Polling clear */ + iow(db, DM9000_BPTR, 0x3f); /* Less 3Kb, 200us */ + iow(db, DM9000_FCR, 0xff); /* Flow Control */ + iow(db, DM9000_SMCR, 0); /* Special Mode */ + /* clear TX status */ + iow(db, DM9000_NSR, NSR_WAKEST | NSR_TX2END | NSR_TX1END); + iow(db, DM9000_ISR, ISR_CLR_STATUS); /* Clear interrupt status */ + + /* Set address filter table */ + dm9000_hash_table(dev); + + /* Activate DM9000 */ + iow(db, DM9000_RCR, RCR_DIS_LONG | RCR_DIS_CRC | RCR_RXEN); + /* Enable TX/RX interrupt mask */ + iow(db, DM9000_IMR, IMR_PAR | IMR_PTM | IMR_PRM); + + /* Init Driver variable */ + db->tx_pkt_cnt = 0; + db->queue_pkt_len = 0; + dev->trans_start = 0; + spin_lock_init(&db->lock); +} + +/* + * Hardware start transmission. + * Send a packet to media from the upper layer. + */ +static int +dm9000_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + + PRINTK3("dm9000_start_xmit\n"); + + if (db->tx_pkt_cnt > 1) + return 1; + + netif_stop_queue(dev); + + /* Disable all interrupts */ + iow(db, DM9000_IMR, IMR_PAR); + + /* Move data to DM9000 TX RAM */ + writeb(DM9000_MWCMD, db->io_addr); + + (db->outblk)(db->io_data, skb->data, skb->len); + db->stats.tx_bytes += skb->len; + + /* TX control: First packet immediately send, second packet queue */ + if (db->tx_pkt_cnt == 0) { + + /* First Packet */ + db->tx_pkt_cnt++; + + /* Set TX length to DM9000 */ + iow(db, DM9000_TXPLL, skb->len & 0xff); + iow(db, DM9000_TXPLH, (skb->len >> 8) & 0xff); + + /* Issue TX polling command */ + iow(db, DM9000_TCR, TCR_TXREQ); /* Cleared after TX complete */ + + dev->trans_start = jiffies; /* save the time stamp */ + + } else { + /* Second packet */ + db->tx_pkt_cnt++; + db->queue_pkt_len = skb->len; + } + + /* free this SKB */ + dev_kfree_skb(skb); + + /* Re-enable resource check */ + if (db->tx_pkt_cnt == 1) + netif_wake_queue(dev); + + /* Re-enable interrupt */ + iow(db, DM9000_IMR, IMR_PAR | IMR_PTM | IMR_PRM); + + return 0; +} + +static void +dm9000_shutdown(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + + /* RESET device */ + dm9000_phy_write(dev, 0, MII_BMCR, BMCR_RESET); /* PHY RESET */ + iow(db, DM9000_GPR, 0x01); /* Power-Down PHY */ + iow(db, DM9000_IMR, IMR_PAR); /* Disable all interrupt */ + iow(db, DM9000_RCR, 0x00); /* Disable RX */ +} + +/* + * Stop the interface. + * The interface is stopped when it is brought. + */ +static int +dm9000_stop(struct net_device *ndev) +{ + board_info_t *db = (board_info_t *) ndev->priv; + + PRINTK1("entering %s\n",__FUNCTION__); + + /* deleted timer */ + del_timer(&db->timer); + + netif_stop_queue(ndev); + netif_carrier_off(ndev); + + /* free interrupt */ + free_irq(ndev->irq, ndev); + + dm9000_shutdown(ndev); + + return 0; +} + +/* + * DM9000 interrupt handler + * receive the packet to upper layer, free the transmitted packet + */ + +void +dm9000_tx_done(struct net_device *dev, board_info_t * db) +{ + int tx_status = ior(db, DM9000_NSR); /* Got TX status */ + + if (tx_status & (NSR_TX2END | NSR_TX1END)) { + /* One packet sent complete */ + db->tx_pkt_cnt--; + db->stats.tx_packets++; + + /* Queue packet check & send */ + if (db->tx_pkt_cnt > 0) { + iow(db, DM9000_TXPLL, db->queue_pkt_len & 0xff); + iow(db, DM9000_TXPLH, (db->queue_pkt_len >> 8) & 0xff); + iow(db, DM9000_TCR, TCR_TXREQ); + dev->trans_start = jiffies; + } + netif_wake_queue(dev); + } +} + +static irqreturn_t +dm9000_interrupt(int irq, void *dev_id, struct pt_regs *regs) +{ + struct net_device *dev = dev_id; + board_info_t *db; + int int_status; + u8 reg_save; + + PRINTK3("entering %s\n",__FUNCTION__); + + if (!dev) { + PRINTK1("dm9000_interrupt() without DEVICE arg\n"); + return IRQ_HANDLED; + } + + /* A real interrupt coming */ + db = (board_info_t *) dev->priv; + spin_lock(&db->lock); + + /* Save previous register address */ + reg_save = readb(db->io_addr); + + /* Disable all interrupts */ + iow(db, DM9000_IMR, IMR_PAR); + + /* Got DM9000 interrupt status */ + int_status = ior(db, DM9000_ISR); /* Got ISR */ + iow(db, DM9000_ISR, int_status); /* Clear ISR status */ + + /* Received the coming packet */ + if (int_status & ISR_PRS) + dm9000_rx(dev); + + /* Trnasmit Interrupt check */ + if (int_status & ISR_PTS) + dm9000_tx_done(dev, db); + + /* Re-enable interrupt mask */ + iow(db, DM9000_IMR, IMR_PAR | IMR_PTM | IMR_PRM); + + /* Restore previous register address */ + writeb(reg_save, db->io_addr); + + spin_unlock(&db->lock); + + return IRQ_HANDLED; +} + +/* + * Get statistics from driver. + */ +static struct net_device_stats * +dm9000_get_stats(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + return &db->stats; +} + +/* + * Process the upper socket ioctl command + */ +static int +dm9000_do_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) +{ + PRINTK1("entering %s\n",__FUNCTION__); + return 0; +} + +/* + * A periodic timer routine + * Dynamic media sense, allocated Rx buffer... + */ +static void +dm9000_timer(unsigned long data) +{ + struct net_device *dev = (struct net_device *) data; + board_info_t *db = (board_info_t *) dev->priv; + u8 reg_save; + unsigned long flags; + + PRINTK3("dm9000_timer()\n"); + + spin_lock_irqsave(db->lock,flags); + /* Save previous register address */ + reg_save = readb(db->io_addr); + + mii_check_media(&db->mii, netif_msg_link(db), 0); + + /* Restore previous register address */ + writeb(reg_save, db->io_addr); + spin_unlock_irqrestore(db->lock,flags); + + /* Set timer again */ + db->timer.expires = DM9000_TIMER_WUT; + add_timer(&db->timer); +} + +struct dm9000_rxhdr { + u16 RxStatus; + u16 RxLen; +} __attribute__((__packed__)); + +/* + * Received a packet and pass to upper layer + */ +static void +dm9000_rx(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + struct dm9000_rxhdr rxhdr; + struct sk_buff *skb; + u8 rxbyte, *rdptr; + int GoodPacket; + int RxLen; + + /* Check packet ready or not */ + do { + ior(db, DM9000_MRCMDX); /* Dummy read */ + + /* Get most updated data */ + rxbyte = readb(db->io_data); + + /* Status check: this byte must be 0 or 1 */ + if (rxbyte > DM9000_PKT_RDY) { + printk("status check failed: %d\n", rxbyte); + iow(db, DM9000_RCR, 0x00); /* Stop Device */ + iow(db, DM9000_ISR, IMR_PAR); /* Stop INT request */ + return; + } + + if (rxbyte != DM9000_PKT_RDY) + return; + + /* A packet ready now & Get status/length */ + GoodPacket = TRUE; + writeb(DM9000_MRCMD, db->io_addr); + + (db->inblk)(db->io_data, &rxhdr, sizeof(rxhdr)); + + RxLen = rxhdr.RxLen; + + /* Packet Status check */ + if (RxLen < 0x40) { + GoodPacket = FALSE; + PRINTK1("Bad Packet received (runt)\n"); + } + + if (RxLen > DM9000_PKT_MAX) { + PRINTK1("RST: RX Len:%x\n", RxLen); + } + + if (rxhdr.RxStatus & 0xbf00) { + GoodPacket = FALSE; + if (rxhdr.RxStatus & 0x100) { + PRINTK1("fifo error\n"); + db->stats.rx_fifo_errors++; + } + if (rxhdr.RxStatus & 0x200) { + PRINTK1("crc error\n"); + db->stats.rx_crc_errors++; + } + if (rxhdr.RxStatus & 0x8000) { + PRINTK1("length error\n"); + db->stats.rx_length_errors++; + } + } + + /* Move data from DM9000 */ + if (GoodPacket + && ((skb = dev_alloc_skb(RxLen + 4)) != NULL)) { + skb->dev = dev; + skb_reserve(skb, 2); + rdptr = (u8 *) skb_put(skb, RxLen - 4); + + /* Read received packet from RX SRAM */ + + (db->inblk)(db->io_data, rdptr, RxLen); + db->stats.rx_bytes += RxLen; + + /* Pass to upper layer */ + skb->protocol = eth_type_trans(skb, dev); + netif_rx(skb); + db->stats.rx_packets++; + + } else { + /* need to dump the packet's data */ + + (db->dumpblk)(db->io_data, RxLen); + } + } while (rxbyte == DM9000_PKT_RDY); +} + +/* + * Read a word data from SROM + */ +static u16 +read_srom_word(board_info_t * db, int offset) +{ + iow(db, DM9000_EPAR, offset); + iow(db, DM9000_EPCR, EPCR_ERPRR); + mdelay(8); /* according to the datasheet 200us should be enough, + but it doesn't work */ + iow(db, DM9000_EPCR, 0x0); + return (ior(db, DM9000_EPDRL) + (ior(db, DM9000_EPDRH) << 8)); +} + +#ifdef DM9000_PROGRAM_EEPROM +/* + * Write a word data to SROM + */ +static void +write_srom_word(board_info_t * db, int offset, u16 val) +{ + iow(db, DM9000_EPAR, offset); + iow(db, DM9000_EPDRH, ((val >> 8) & 0xff)); + iow(db, DM9000_EPDRL, (val & 0xff)); + iow(db, DM9000_EPCR, EPCR_WEP | EPCR_ERPRW); + mdelay(8); /* same shit */ + iow(db, DM9000_EPCR, 0); +} + +/* + * Only for development: + * Here we write static data to the eeprom in case + * we don't have valid content on a new board + */ +static void +program_eeprom(board_info_t * db) +{ + u16 eeprom[] = { 0x0c00, 0x007f, 0x1300, /* MAC Address */ + 0x0000, /* Autoload: accept nothing */ + 0x0a46, 0x9000, /* Vendor / Product ID */ + 0x0000, /* pin control */ + 0x0000, + }; /* Wake-up mode control */ + int i; + for (i = 0; i < 8; i++) + write_srom_word(db, i, eeprom[i]); +} +#endif + + +/* + * Calculate the CRC valude of the Rx packet + * flag = 1 : return the reverse CRC (for the received packet CRC) + * 0 : return the normal CRC (for Hash Table index) + */ + +static unsigned long +cal_CRC(unsigned char *Data, unsigned int Len, u8 flag) +{ + + u32 crc = ether_crc_le(Len, Data); + + if (flag) + return ~crc; + + return crc; +} + +/* + * Set DM9000 multicast address + */ +static void +dm9000_hash_table(struct net_device *dev) +{ + board_info_t *db = (board_info_t *) dev->priv; + struct dev_mc_list *mcptr = dev->mc_list; + int mc_cnt = dev->mc_count; + u32 hash_val; + u16 i, oft, hash_table[4]; + unsigned long flags; + + PRINTK2("dm9000_hash_table()\n"); + + spin_lock_irqsave(&db->lock,flags); + + for (i = 0, oft = 0x10; i < 6; i++, oft++) + iow(db, oft, dev->dev_addr[i]); + + /* Clear Hash Table */ + for (i = 0; i < 4; i++) + hash_table[i] = 0x0; + + /* broadcast address */ + hash_table[3] = 0x8000; + + /* the multicast address in Hash Table : 64 bits */ + for (i = 0; i < mc_cnt; i++, mcptr = mcptr->next) { + hash_val = cal_CRC((char *) mcptr->dmi_addr, 6, 0) & 0x3f; + hash_table[hash_val / 16] |= (u16) 1 << (hash_val % 16); + } + + /* Write the hash table to MAC MD table */ + for (i = 0, oft = 0x16; i < 4; i++) { + iow(db, oft++, hash_table[i] & 0xff); + iow(db, oft++, (hash_table[i] >> 8) & 0xff); + } + + spin_unlock_irqrestore(&db->lock,flags); +} + + +/* + * Read a word from phyxcer + */ +static int +dm9000_phy_read(struct net_device *dev, int phy_reg_unused, int reg) +{ + board_info_t *db = (board_info_t *) dev->priv; + unsigned long flags; + int ret; + + spin_lock_irqsave(&db->lock,flags); + /* Fill the phyxcer register into REG_0C */ + iow(db, DM9000_EPAR, DM9000_PHY | reg); + + iow(db, DM9000_EPCR, 0xc); /* Issue phyxcer read command */ + udelay(100); /* Wait read complete */ + iow(db, DM9000_EPCR, 0x0); /* Clear phyxcer read command */ + + /* The read data keeps on REG_0D & REG_0E */ + ret = (ior(db, DM9000_EPDRH) << 8) | ior(db, DM9000_EPDRL); + + spin_unlock_irqrestore(&db->lock,flags); + + return ret; +} + +/* + * Write a word to phyxcer + */ +static void +dm9000_phy_write(struct net_device *dev, int phyaddr_unused, int reg, int value) +{ + board_info_t *db = (board_info_t *) dev->priv; + unsigned long flags; + + spin_lock_irqsave(&db->lock,flags); + + /* Fill the phyxcer register into REG_0C */ + iow(db, DM9000_EPAR, DM9000_PHY | reg); + + /* Fill the written data into REG_0D & REG_0E */ + iow(db, DM9000_EPDRL, (value & 0xff)); + iow(db, DM9000_EPDRH, ((value >> 8) & 0xff)); + + iow(db, DM9000_EPCR, 0xa); /* Issue phyxcer write command */ + udelay(500); /* Wait write complete */ + iow(db, DM9000_EPCR, 0x0); /* Clear phyxcer write command */ + + spin_unlock_irqrestore(&db->lock,flags); +} + +static int +dm9000_drv_suspend(struct device *dev, u32 state, u32 level) +{ + struct net_device *ndev = dev_get_drvdata(dev); + + if (ndev && level == SUSPEND_DISABLE) { + if (netif_running(ndev)) { + netif_device_detach(ndev); + dm9000_shutdown(ndev); + } + } + return 0; +} + +static int +dm9000_drv_resume(struct device *dev, u32 level) +{ + struct net_device *ndev = dev_get_drvdata(dev); + board_info_t *db = (board_info_t *) ndev->priv; + + if (ndev && level == RESUME_ENABLE) { + + if (netif_running(ndev)) { + dm9000_reset(db); + dm9000_init_dm9000(ndev); + + netif_device_attach(ndev); + } + } + return 0; +} + +static int +dm9000_drv_remove(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct net_device *ndev = dev_get_drvdata(dev); + + dev_set_drvdata(dev, NULL); + + unregister_netdev(ndev); + dm9000_release_board(pdev, (board_info_t *) ndev->priv); + kfree(ndev); /* free device structure */ + + PRINTK1("clean_module() exit\n"); + + return 0; +} + +static struct device_driver dm9000_driver = { + .name = "dm9000", + .bus = &platform_bus_type, + .probe = dm9000_probe, + .remove = dm9000_drv_remove, + .suspend = dm9000_drv_suspend, + .resume = dm9000_drv_resume, +}; + +static int __init +dm9000_init(void) +{ + return driver_register(&dm9000_driver); /* search board and register */ +} + +static void __exit +dm9000_cleanup(void) +{ + driver_unregister(&dm9000_driver); +} + +module_init(dm9000_init); +module_exit(dm9000_cleanup); + +MODULE_AUTHOR("Sascha Hauer, Ben Dooks"); +MODULE_DESCRIPTION("Davicom DM9000 network driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/dm9000.h b/drivers/net/dm9000.h new file mode 100644 index 000000000000..82cad360bafc --- /dev/null +++ b/drivers/net/dm9000.h @@ -0,0 +1,135 @@ +/* + * dm9000 Ethernet + */ + +#ifndef _DM9000X_H_ +#define _DM9000X_H_ + +#define DM9000_ID 0x90000A46 + +/* although the registers are 16 bit, they are 32-bit aligned. + */ + +#define DM9000_NCR 0x00 +#define DM9000_NSR 0x01 +#define DM9000_TCR 0x02 +#define DM9000_TSR1 0x03 +#define DM9000_TSR2 0x04 +#define DM9000_RCR 0x05 +#define DM9000_RSR 0x06 +#define DM9000_ROCR 0x07 +#define DM9000_BPTR 0x08 +#define DM9000_FCTR 0x09 +#define DM9000_FCR 0x0A +#define DM9000_EPCR 0x0B +#define DM9000_EPAR 0x0C +#define DM9000_EPDRL 0x0D +#define DM9000_EPDRH 0x0E +#define DM9000_WCR 0x0F + +#define DM9000_PAR 0x10 +#define DM9000_MAR 0x16 + +#define DM9000_GPCR 0x1e +#define DM9000_GPR 0x1f +#define DM9000_TRPAL 0x22 +#define DM9000_TRPAH 0x23 +#define DM9000_RWPAL 0x24 +#define DM9000_RWPAH 0x25 + +#define DM9000_VIDL 0x28 +#define DM9000_VIDH 0x29 +#define DM9000_PIDL 0x2A +#define DM9000_PIDH 0x2B + +#define DM9000_CHIPR 0x2C +#define DM9000_SMCR 0x2F + +#define DM9000_MRCMDX 0xF0 +#define DM9000_MRCMD 0xF2 +#define DM9000_MRRL 0xF4 +#define DM9000_MRRH 0xF5 +#define DM9000_MWCMDX 0xF6 +#define DM9000_MWCMD 0xF8 +#define DM9000_MWRL 0xFA +#define DM9000_MWRH 0xFB +#define DM9000_TXPLL 0xFC +#define DM9000_TXPLH 0xFD +#define DM9000_ISR 0xFE +#define DM9000_IMR 0xFF + +#define NCR_EXT_PHY (1<<7) +#define NCR_WAKEEN (1<<6) +#define NCR_FCOL (1<<4) +#define NCR_FDX (1<<3) +#define NCR_LBK (3<<1) +#define NCR_RST (1<<0) + +#define NSR_SPEED (1<<7) +#define NSR_LINKST (1<<6) +#define NSR_WAKEST (1<<5) +#define NSR_TX2END (1<<3) +#define NSR_TX1END (1<<2) +#define NSR_RXOV (1<<1) + +#define TCR_TJDIS (1<<6) +#define TCR_EXCECM (1<<5) +#define TCR_PAD_DIS2 (1<<4) +#define TCR_CRC_DIS2 (1<<3) +#define TCR_PAD_DIS1 (1<<2) +#define TCR_CRC_DIS1 (1<<1) +#define TCR_TXREQ (1<<0) + +#define TSR_TJTO (1<<7) +#define TSR_LC (1<<6) +#define TSR_NC (1<<5) +#define TSR_LCOL (1<<4) +#define TSR_COL (1<<3) +#define TSR_EC (1<<2) + +#define RCR_WTDIS (1<<6) +#define RCR_DIS_LONG (1<<5) +#define RCR_DIS_CRC (1<<4) +#define RCR_ALL (1<<3) +#define RCR_RUNT (1<<2) +#define RCR_PRMSC (1<<1) +#define RCR_RXEN (1<<0) + +#define RSR_RF (1<<7) +#define RSR_MF (1<<6) +#define RSR_LCS (1<<5) +#define RSR_RWTO (1<<4) +#define RSR_PLE (1<<3) +#define RSR_AE (1<<2) +#define RSR_CE (1<<1) +#define RSR_FOE (1<<0) + +#define FCTR_HWOT(ot) (( ot & 0xf ) << 4 ) +#define FCTR_LWOT(ot) ( ot & 0xf ) + +#define IMR_PAR (1<<7) +#define IMR_ROOM (1<<3) +#define IMR_ROM (1<<2) +#define IMR_PTM (1<<1) +#define IMR_PRM (1<<0) + +#define ISR_ROOS (1<<3) +#define ISR_ROS (1<<2) +#define ISR_PTS (1<<1) +#define ISR_PRS (1<<0) +#define ISR_CLR_STATUS (ISR_ROOS | ISR_ROS | ISR_PTS | ISR_PRS) + +#define EPCR_REEP (1<<5) +#define EPCR_WEP (1<<4) +#define EPCR_EPOS (1<<3) +#define EPCR_ERPRR (1<<2) +#define EPCR_ERPRW (1<<1) +#define EPCR_ERRE (1<<0) + +#define GPCR_GEP_CNTL (1<<0) + +#define DM9000_PKT_RDY 0x01 /* Packet ready to receive */ +#define DM9000_PKT_MAX 1536 /* Received packet max size */ + +#endif /* _DM9000X_H_ */ + diff --git a/include/linux/dm9000.h b/include/linux/dm9000.h new file mode 100644 index 000000000000..0008e2ad0c9f --- /dev/null +++ b/include/linux/dm9000.h @@ -0,0 +1,36 @@ +/* include/linux/dm9000.h + * + * Copyright (c) 2004 Simtec Electronics + * Ben Dooks + * + * Header file for dm9000 platform data + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * +*/ + +#ifndef __DM9000_PLATFORM_DATA +#define __DM9000_PLATFORM_DATA __FILE__ + +/* IO control flags */ + +#define DM9000_PLATF_8BITONLY (0x0001) +#define DM9000_PLATF_16BITONLY (0x0002) +#define DM9000_PLATF_32BITONLY (0x0004) + +/* platfrom data for platfrom device structure's platfrom_data field */ + +struct dm9000_plat_data { + unsigned int flags; + + /* allow replacement IO routines */ + + void (*inblk)(void __iomem *reg, void *data, int len); + void (*outblk)(void __iomem *reg, void *data, int len); + void (*dumpblk)(void __iomem *reg, int len); +}; + +#endif /* __DM9000_PLATFORM_DATA */ + -- cgit v1.2.3 From b3dd65f958354226275522b5a64157834bdc5415 Mon Sep 17 00:00:00 2001 From: Krzysztof Halasa Date: Thu, 21 Apr 2005 15:57:25 +0200 Subject: [PATCH] Generic HDLC update The attached patch updates generic HDLC to version 1.18. FR Cisco LMI production-tested. Please apply to Linux 2.6. Thanks. Changes: - doc updates - added Cisco LMI support to Frame-Relay code - cleaned hdlc_fr.c a bit, removed some orphaned #defines etc. - fixed a problem with non-functional LMI in FR DCE mode. - changed diagnostic messages to better conform to FR standards - all protocols: information about carrier changes (DCD line) is now printed to kernel logs. Signed-Off-By: Krzysztof Halasa --- Documentation/networking/generic-hdlc.txt | 51 ++--- drivers/net/wan/hdlc_fr.c | 320 ++++++++++++++++-------------- drivers/net/wan/hdlc_generic.c | 16 +- include/linux/hdlc.h | 4 +- 4 files changed, 216 insertions(+), 175 deletions(-) (limited to 'include') diff --git a/Documentation/networking/generic-hdlc.txt b/Documentation/networking/generic-hdlc.txt index 7d1dc6b884f3..31bc8b759b75 100644 --- a/Documentation/networking/generic-hdlc.txt +++ b/Documentation/networking/generic-hdlc.txt @@ -1,21 +1,21 @@ Generic HDLC layer Krzysztof Halasa -January, 2003 Generic HDLC layer currently supports: -- Frame Relay (ANSI, CCITT and no LMI), with ARP support (no InARP). - Normal (routed) and Ethernet-bridged (Ethernet device emulation) - interfaces can share a single PVC. -- raw HDLC - either IP (IPv4) interface or Ethernet device emulation. -- Cisco HDLC, -- PPP (uses syncppp.c), -- X.25 (uses X.25 routines). - -There are hardware drivers for the following cards: -- C101 by Moxa Technologies Co., Ltd. -- RISCom/N2 by SDL Communications Inc. -- and others, some not in the official kernel. +1. Frame Relay (ANSI, CCITT, Cisco and no LMI). + - Normal (routed) and Ethernet-bridged (Ethernet device emulation) + interfaces can share a single PVC. + - ARP support (no InARP support in the kernel - there is an + experimental InARP user-space daemon available on: + http://www.kernel.org/pub/linux/utils/net/hdlc/). +2. raw HDLC - either IP (IPv4) interface or Ethernet device emulation. +3. Cisco HDLC. +4. PPP (uses syncppp.c). +5. X.25 (uses X.25 routines). + +Generic HDLC is a protocol driver only - it needs a low-level driver +for your particular hardware. Ethernet device emulation (using HDLC or Frame-Relay PVC) is compatible with IEEE 802.1Q (VLANs) and 802.1D (Ethernet bridging). @@ -24,7 +24,7 @@ with IEEE 802.1Q (VLANs) and 802.1D (Ethernet bridging). Make sure the hdlc.o and the hardware driver are loaded. It should create a number of "hdlc" (hdlc0 etc) network devices, one for each WAN port. You'll need the "sethdlc" utility, get it from: - http://hq.pm.waw.pl/hdlc/ + http://www.kernel.org/pub/linux/utils/net/hdlc/ Compile sethdlc.c utility: gcc -O2 -Wall -o sethdlc sethdlc.c @@ -52,12 +52,12 @@ Setting interface: * v35 | rs232 | x21 | t1 | e1 - sets physical interface for a given port if the card has software-selectable interfaces loopback - activate hardware loopback (for testing only) -* clock ext - external clock (uses DTE RX and TX clock) -* clock int - internal clock (provides clock signal on DCE clock output) -* clock txint - TX internal, RX external (provides TX clock on DCE output) -* clock txfromrx - TX clock derived from RX clock (TX clock on DCE output) -* rate - sets clock rate in bps (not required for external clock or - for txfromrx) +* clock ext - both RX clock and TX clock external +* clock int - both RX clock and TX clock internal +* clock txint - RX clock external, TX clock internal +* clock txfromrx - RX clock external, TX clock derived from RX clock +* rate - sets clock rate in bps (for "int" or "txint" clock only) + Setting protocol: @@ -79,7 +79,7 @@ Setting protocol: * x25 - sets X.25 mode * fr - Frame Relay mode - lmi ansi / ccitt / none - LMI (link management) type + lmi ansi / ccitt / cisco / none - LMI (link management) type dce - Frame Relay DCE (network) side LMI instead of default DTE (user). It has nothing to do with clocks! t391 - link integrity verification polling timer (in seconds) - user @@ -119,13 +119,14 @@ or -If you have a problem with N2 or C101 card, you can issue the "private" -command to see port's packet descriptor rings (in kernel logs): +If you have a problem with N2, C101 or PLX200SYN card, you can issue the +"private" command to see port's packet descriptor rings (in kernel logs): sethdlc hdlc0 private -The hardware driver has to be build with CONFIG_HDLC_DEBUG_RINGS. +The hardware driver has to be build with #define DEBUG_RINGS. Attaching this info to bug reports would be helpful. Anyway, let me know if you have problems using this. -For patches and other info look at http://hq.pm.waw.pl/hdlc/ +For patches and other info look at: +. diff --git a/drivers/net/wan/hdlc_fr.c b/drivers/net/wan/hdlc_fr.c index 7f450b51a6cb..a5d6891c9d4c 100644 --- a/drivers/net/wan/hdlc_fr.c +++ b/drivers/net/wan/hdlc_fr.c @@ -2,7 +2,7 @@ * Generic HDLC support routines for Linux * Frame Relay support * - * Copyright (C) 1999 - 2003 Krzysztof Halasa + * Copyright (C) 1999 - 2005 Krzysztof Halasa * * This program is free software; you can redistribute it and/or modify it * under the terms of version 2 of the GNU General Public License @@ -27,6 +27,10 @@ active = open and "link reliable" exist = new = not used + CCITT LMI: ITU-T Q.933 Annex A + ANSI LMI: ANSI T1.617 Annex D + CISCO LMI: the original, aka "Gang of Four" LMI + */ #include @@ -49,45 +53,41 @@ #undef DEBUG_ECN #undef DEBUG_LINK -#define MAXLEN_LMISTAT 20 /* max size of status enquiry frame */ - -#define PVC_STATE_NEW 0x01 -#define PVC_STATE_ACTIVE 0x02 -#define PVC_STATE_FECN 0x08 /* FECN condition */ -#define PVC_STATE_BECN 0x10 /* BECN condition */ - - -#define FR_UI 0x03 -#define FR_PAD 0x00 - -#define NLPID_IP 0xCC -#define NLPID_IPV6 0x8E -#define NLPID_SNAP 0x80 -#define NLPID_PAD 0x00 -#define NLPID_Q933 0x08 - - -#define LMI_DLCI 0 /* LMI DLCI */ -#define LMI_PROTO 0x08 -#define LMI_CALLREF 0x00 /* Call Reference */ -#define LMI_ANSI_LOCKSHIFT 0x95 /* ANSI lockshift */ -#define LMI_REPTYPE 1 /* report type */ -#define LMI_CCITT_REPTYPE 0x51 -#define LMI_ALIVE 3 /* keep alive */ -#define LMI_CCITT_ALIVE 0x53 -#define LMI_PVCSTAT 7 /* pvc status */ -#define LMI_CCITT_PVCSTAT 0x57 -#define LMI_FULLREP 0 /* full report */ -#define LMI_INTEGRITY 1 /* link integrity report */ -#define LMI_SINGLE 2 /* single pvc report */ +#define FR_UI 0x03 +#define FR_PAD 0x00 + +#define NLPID_IP 0xCC +#define NLPID_IPV6 0x8E +#define NLPID_SNAP 0x80 +#define NLPID_PAD 0x00 +#define NLPID_CCITT_ANSI_LMI 0x08 +#define NLPID_CISCO_LMI 0x09 + + +#define LMI_CCITT_ANSI_DLCI 0 /* LMI DLCI */ +#define LMI_CISCO_DLCI 1023 + +#define LMI_CALLREF 0x00 /* Call Reference */ +#define LMI_ANSI_LOCKSHIFT 0x95 /* ANSI locking shift */ +#define LMI_ANSI_CISCO_REPTYPE 0x01 /* report type */ +#define LMI_CCITT_REPTYPE 0x51 +#define LMI_ANSI_CISCO_ALIVE 0x03 /* keep alive */ +#define LMI_CCITT_ALIVE 0x53 +#define LMI_ANSI_CISCO_PVCSTAT 0x07 /* PVC status */ +#define LMI_CCITT_PVCSTAT 0x57 + +#define LMI_FULLREP 0x00 /* full report */ +#define LMI_INTEGRITY 0x01 /* link integrity report */ +#define LMI_SINGLE 0x02 /* single PVC report */ + #define LMI_STATUS_ENQUIRY 0x75 #define LMI_STATUS 0x7D /* reply */ #define LMI_REPT_LEN 1 /* report type element length */ #define LMI_INTEG_LEN 2 /* link integrity element length */ -#define LMI_LENGTH 13 /* standard LMI frame length */ -#define LMI_ANSI_LENGTH 14 +#define LMI_CCITT_CISCO_LENGTH 13 /* LMI frame lengths */ +#define LMI_ANSI_LENGTH 14 typedef struct { @@ -223,51 +223,34 @@ static inline struct net_device** get_dev_p(pvc_device *pvc, int type) } -static inline u16 status_to_dlci(u8 *status, int *active, int *new) -{ - *new = (status[2] & 0x08) ? 1 : 0; - *active = (status[2] & 0x02) ? 1 : 0; - - return ((status[0] & 0x3F) << 4) | ((status[1] & 0x78) >> 3); -} - - -static inline void dlci_to_status(u16 dlci, u8 *status, int active, int new) -{ - status[0] = (dlci >> 4) & 0x3F; - status[1] = ((dlci << 3) & 0x78) | 0x80; - status[2] = 0x80; - - if (new) - status[2] |= 0x08; - else if (active) - status[2] |= 0x02; -} - - - static int fr_hard_header(struct sk_buff **skb_p, u16 dlci) { u16 head_len; struct sk_buff *skb = *skb_p; switch (skb->protocol) { - case __constant_ntohs(ETH_P_IP): + case __constant_ntohs(NLPID_CCITT_ANSI_LMI): head_len = 4; skb_push(skb, head_len); - skb->data[3] = NLPID_IP; + skb->data[3] = NLPID_CCITT_ANSI_LMI; break; - case __constant_ntohs(ETH_P_IPV6): + case __constant_ntohs(NLPID_CISCO_LMI): head_len = 4; skb_push(skb, head_len); - skb->data[3] = NLPID_IPV6; + skb->data[3] = NLPID_CISCO_LMI; break; - case __constant_ntohs(LMI_PROTO): + case __constant_ntohs(ETH_P_IP): + head_len = 4; + skb_push(skb, head_len); + skb->data[3] = NLPID_IP; + break; + + case __constant_ntohs(ETH_P_IPV6): head_len = 4; skb_push(skb, head_len); - skb->data[3] = LMI_PROTO; + skb->data[3] = NLPID_IPV6; break; case __constant_ntohs(ETH_P_802_3): @@ -461,13 +444,14 @@ static void fr_lmi_send(struct net_device *dev, int fullrep) hdlc_device *hdlc = dev_to_hdlc(dev); struct sk_buff *skb; pvc_device *pvc = hdlc->state.fr.first_pvc; - int len = (hdlc->state.fr.settings.lmi == LMI_ANSI) ? LMI_ANSI_LENGTH - : LMI_LENGTH; - int stat_len = 3; + int lmi = hdlc->state.fr.settings.lmi; + int dce = hdlc->state.fr.settings.dce; + int len = lmi == LMI_ANSI ? LMI_ANSI_LENGTH : LMI_CCITT_CISCO_LENGTH; + int stat_len = (lmi == LMI_CISCO) ? 6 : 3; u8 *data; int i = 0; - if (hdlc->state.fr.settings.dce && fullrep) { + if (dce && fullrep) { len += hdlc->state.fr.dce_pvc_count * (2 + stat_len); if (len > HDLC_MAX_MRU) { printk(KERN_WARNING "%s: Too many PVCs while sending " @@ -484,29 +468,31 @@ static void fr_lmi_send(struct net_device *dev, int fullrep) } memset(skb->data, 0, len); skb_reserve(skb, 4); - skb->protocol = __constant_htons(LMI_PROTO); - fr_hard_header(&skb, LMI_DLCI); + if (lmi == LMI_CISCO) { + skb->protocol = __constant_htons(NLPID_CISCO_LMI); + fr_hard_header(&skb, LMI_CISCO_DLCI); + } else { + skb->protocol = __constant_htons(NLPID_CCITT_ANSI_LMI); + fr_hard_header(&skb, LMI_CCITT_ANSI_DLCI); + } data = skb->tail; data[i++] = LMI_CALLREF; - data[i++] = hdlc->state.fr.settings.dce - ? LMI_STATUS : LMI_STATUS_ENQUIRY; - if (hdlc->state.fr.settings.lmi == LMI_ANSI) + data[i++] = dce ? LMI_STATUS : LMI_STATUS_ENQUIRY; + if (lmi == LMI_ANSI) data[i++] = LMI_ANSI_LOCKSHIFT; - data[i++] = (hdlc->state.fr.settings.lmi == LMI_CCITT) - ? LMI_CCITT_REPTYPE : LMI_REPTYPE; + data[i++] = lmi == LMI_CCITT ? LMI_CCITT_REPTYPE : + LMI_ANSI_CISCO_REPTYPE; data[i++] = LMI_REPT_LEN; data[i++] = fullrep ? LMI_FULLREP : LMI_INTEGRITY; - - data[i++] = (hdlc->state.fr.settings.lmi == LMI_CCITT) - ? LMI_CCITT_ALIVE : LMI_ALIVE; + data[i++] = lmi == LMI_CCITT ? LMI_CCITT_ALIVE : LMI_ANSI_CISCO_ALIVE; data[i++] = LMI_INTEG_LEN; data[i++] = hdlc->state.fr.txseq =fr_lmi_nextseq(hdlc->state.fr.txseq); data[i++] = hdlc->state.fr.rxseq; - if (hdlc->state.fr.settings.dce && fullrep) { + if (dce && fullrep) { while (pvc) { - data[i++] = (hdlc->state.fr.settings.lmi == LMI_CCITT) - ? LMI_CCITT_PVCSTAT : LMI_PVCSTAT; + data[i++] = lmi == LMI_CCITT ? LMI_CCITT_PVCSTAT : + LMI_ANSI_CISCO_PVCSTAT; data[i++] = stat_len; /* LMI start/restart */ @@ -523,8 +509,20 @@ static void fr_lmi_send(struct net_device *dev, int fullrep) fr_log_dlci_active(pvc); } - dlci_to_status(pvc->dlci, data + i, - pvc->state.active, pvc->state.new); + if (lmi == LMI_CISCO) { + data[i] = pvc->dlci >> 8; + data[i + 1] = pvc->dlci & 0xFF; + } else { + data[i] = (pvc->dlci >> 4) & 0x3F; + data[i + 1] = ((pvc->dlci << 3) & 0x78) | 0x80; + data[i + 2] = 0x80; + } + + if (pvc->state.new) + data[i + 2] |= 0x08; + else if (pvc->state.active) + data[i + 2] |= 0x02; + i += stat_len; pvc = pvc->next; } @@ -569,6 +567,8 @@ static void fr_set_link_state(int reliable, struct net_device *dev) pvc_carrier(0, pvc); pvc->state.exist = pvc->state.active = 0; pvc->state.new = 0; + if (!hdlc->state.fr.settings.dce) + pvc->state.bandwidth = 0; pvc = pvc->next; } } @@ -583,11 +583,12 @@ static void fr_timer(unsigned long arg) int i, cnt = 0, reliable; u32 list; - if (hdlc->state.fr.settings.dce) + if (hdlc->state.fr.settings.dce) { reliable = hdlc->state.fr.request && time_before(jiffies, hdlc->state.fr.last_poll + hdlc->state.fr.settings.t392 * HZ); - else { + hdlc->state.fr.request = 0; + } else { hdlc->state.fr.last_errors <<= 1; /* Shift the list */ if (hdlc->state.fr.request) { if (hdlc->state.fr.reliable) @@ -634,65 +635,88 @@ static void fr_timer(unsigned long arg) static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) { hdlc_device *hdlc = dev_to_hdlc(dev); - int stat_len; pvc_device *pvc; - int reptype = -1, error, no_ram; u8 rxseq, txseq; - int i; + int lmi = hdlc->state.fr.settings.lmi; + int dce = hdlc->state.fr.settings.dce; + int stat_len = (lmi == LMI_CISCO) ? 6 : 3, reptype, error, no_ram, i; - if (skb->len < ((hdlc->state.fr.settings.lmi == LMI_ANSI) - ? LMI_ANSI_LENGTH : LMI_LENGTH)) { + if (skb->len < (lmi == LMI_ANSI ? LMI_ANSI_LENGTH : + LMI_CCITT_CISCO_LENGTH)) { printk(KERN_INFO "%s: Short LMI frame\n", dev->name); return 1; } - if (skb->data[5] != (!hdlc->state.fr.settings.dce ? - LMI_STATUS : LMI_STATUS_ENQUIRY)) { - printk(KERN_INFO "%s: LMI msgtype=%x, Not LMI status %s\n", - dev->name, skb->data[2], - hdlc->state.fr.settings.dce ? "enquiry" : "reply"); + if (skb->data[3] != (lmi == LMI_CISCO ? NLPID_CISCO_LMI : + NLPID_CCITT_ANSI_LMI)) { + printk(KERN_INFO "%s: Received non-LMI frame with LMI" + " DLCI\n", dev->name); + return 1; + } + + if (skb->data[4] != LMI_CALLREF) { + printk(KERN_INFO "%s: Invalid LMI Call reference (0x%02X)\n", + dev->name, skb->data[4]); + return 1; + } + + if (skb->data[5] != (dce ? LMI_STATUS_ENQUIRY : LMI_STATUS)) { + printk(KERN_INFO "%s: Invalid LMI Message type (0x%02X)\n", + dev->name, skb->data[5]); return 1; } - i = (hdlc->state.fr.settings.lmi == LMI_ANSI) ? 7 : 6; + if (lmi == LMI_ANSI) { + if (skb->data[6] != LMI_ANSI_LOCKSHIFT) { + printk(KERN_INFO "%s: Not ANSI locking shift in LMI" + " message (0x%02X)\n", dev->name, skb->data[6]); + return 1; + } + i = 7; + } else + i = 6; - if (skb->data[i] != - ((hdlc->state.fr.settings.lmi == LMI_CCITT) - ? LMI_CCITT_REPTYPE : LMI_REPTYPE)) { - printk(KERN_INFO "%s: Not a report type=%x\n", + if (skb->data[i] != (lmi == LMI_CCITT ? LMI_CCITT_REPTYPE : + LMI_ANSI_CISCO_REPTYPE)) { + printk(KERN_INFO "%s: Not an LMI Report type IE (0x%02X)\n", dev->name, skb->data[i]); return 1; } - i++; - i++; /* Skip length field */ + if (skb->data[++i] != LMI_REPT_LEN) { + printk(KERN_INFO "%s: Invalid LMI Report type IE length" + " (%u)\n", dev->name, skb->data[i]); + return 1; + } - reptype = skb->data[i++]; + reptype = skb->data[++i]; + if (reptype != LMI_INTEGRITY && reptype != LMI_FULLREP) { + printk(KERN_INFO "%s: Unsupported LMI Report type (0x%02X)\n", + dev->name, reptype); + return 1; + } - if (skb->data[i]!= - ((hdlc->state.fr.settings.lmi == LMI_CCITT) - ? LMI_CCITT_ALIVE : LMI_ALIVE)) { - printk(KERN_INFO "%s: Unsupported status element=%x\n", - dev->name, skb->data[i]); + if (skb->data[++i] != (lmi == LMI_CCITT ? LMI_CCITT_ALIVE : + LMI_ANSI_CISCO_ALIVE)) { + printk(KERN_INFO "%s: Not an LMI Link integrity verification" + " IE (0x%02X)\n", dev->name, skb->data[i]); return 1; } - i++; - i++; /* Skip length field */ + if (skb->data[++i] != LMI_INTEG_LEN) { + printk(KERN_INFO "%s: Invalid LMI Link integrity verification" + " IE length (%u)\n", dev->name, skb->data[i]); + return 1; + } + i++; hdlc->state.fr.rxseq = skb->data[i++]; /* TX sequence from peer */ rxseq = skb->data[i++]; /* Should confirm our sequence */ txseq = hdlc->state.fr.txseq; - if (hdlc->state.fr.settings.dce) { - if (reptype != LMI_FULLREP && reptype != LMI_INTEGRITY) { - printk(KERN_INFO "%s: Unsupported report type=%x\n", - dev->name, reptype); - return 1; - } + if (dce) hdlc->state.fr.last_poll = jiffies; - } error = 0; if (!hdlc->state.fr.reliable) @@ -703,7 +727,7 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) error = 1; } - if (hdlc->state.fr.settings.dce) { + if (dce) { if (hdlc->state.fr.fullrep_sent && !error) { /* Stop sending full report - the last one has been confirmed by DTE */ hdlc->state.fr.fullrep_sent = 0; @@ -725,6 +749,7 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) hdlc->state.fr.dce_changed = 0; } + hdlc->state.fr.request = 1; /* got request */ fr_lmi_send(dev, reptype == LMI_FULLREP ? 1 : 0); return 0; } @@ -739,7 +764,6 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) if (reptype != LMI_FULLREP) return 0; - stat_len = 3; pvc = hdlc->state.fr.first_pvc; while (pvc) { @@ -750,24 +774,35 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) no_ram = 0; while (skb->len >= i + 2 + stat_len) { u16 dlci; + u32 bw; unsigned int active, new; - if (skb->data[i] != ((hdlc->state.fr.settings.lmi == LMI_CCITT) - ? LMI_CCITT_PVCSTAT : LMI_PVCSTAT)) { - printk(KERN_WARNING "%s: Invalid PVCSTAT ID: %x\n", - dev->name, skb->data[i]); + if (skb->data[i] != (lmi == LMI_CCITT ? LMI_CCITT_PVCSTAT : + LMI_ANSI_CISCO_PVCSTAT)) { + printk(KERN_INFO "%s: Not an LMI PVC status IE" + " (0x%02X)\n", dev->name, skb->data[i]); return 1; } - i++; - if (skb->data[i] != stat_len) { - printk(KERN_WARNING "%s: Invalid PVCSTAT length: %x\n", - dev->name, skb->data[i]); + if (skb->data[++i] != stat_len) { + printk(KERN_INFO "%s: Invalid LMI PVC status IE length" + " (%u)\n", dev->name, skb->data[i]); return 1; } i++; - dlci = status_to_dlci(skb->data + i, &active, &new); + new = !! (skb->data[i + 2] & 0x08); + active = !! (skb->data[i + 2] & 0x02); + if (lmi == LMI_CISCO) { + dlci = (skb->data[i] << 8) | skb->data[i + 1]; + bw = (skb->data[i + 3] << 16) | + (skb->data[i + 4] << 8) | + (skb->data[i + 5]); + } else { + dlci = ((skb->data[i] & 0x3F) << 4) | + ((skb->data[i + 1] & 0x78) >> 3); + bw = 0; + } pvc = add_pvc(dev, dlci); @@ -783,9 +818,11 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) pvc->state.deleted = 0; if (active != pvc->state.active || new != pvc->state.new || + bw != pvc->state.bandwidth || !pvc->state.exist) { pvc->state.new = new; pvc->state.active = active; + pvc->state.bandwidth = bw; pvc_carrier(active, pvc); fr_log_dlci_active(pvc); } @@ -801,6 +838,7 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb) pvc_carrier(0, pvc); pvc->state.active = pvc->state.new = 0; pvc->state.exist = 0; + pvc->state.bandwidth = 0; fr_log_dlci_active(pvc); } pvc = pvc->next; @@ -829,22 +867,15 @@ static int fr_rx(struct sk_buff *skb) dlci = q922_to_dlci(skb->data); - if (dlci == LMI_DLCI) { - if (hdlc->state.fr.settings.lmi == LMI_NONE) - goto rx_error; /* LMI packet with no LMI? */ - - if (data[3] == LMI_PROTO) { - if (fr_lmi_recv(ndev, skb)) - goto rx_error; - else { - dev_kfree_skb_any(skb); - return NET_RX_SUCCESS; - } - } - - printk(KERN_INFO "%s: Received non-LMI frame with LMI DLCI\n", - ndev->name); - goto rx_error; + if ((dlci == LMI_CCITT_ANSI_DLCI && + (hdlc->state.fr.settings.lmi == LMI_ANSI || + hdlc->state.fr.settings.lmi == LMI_CCITT)) || + (dlci == LMI_CISCO_DLCI && + hdlc->state.fr.settings.lmi == LMI_CISCO)) { + if (fr_lmi_recv(ndev, skb)) + goto rx_error; + dev_kfree_skb_any(skb); + return NET_RX_SUCCESS; } pvc = find_pvc(hdlc, dlci); @@ -1170,7 +1201,8 @@ int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr) if ((new_settings.lmi != LMI_NONE && new_settings.lmi != LMI_ANSI && - new_settings.lmi != LMI_CCITT) || + new_settings.lmi != LMI_CCITT && + new_settings.lmi != LMI_CISCO) || new_settings.t391 < 1 || new_settings.t392 < 2 || new_settings.n391 < 1 || diff --git a/drivers/net/wan/hdlc_generic.c b/drivers/net/wan/hdlc_generic.c index 6ed064cb4469..a63f6a2cc4f7 100644 --- a/drivers/net/wan/hdlc_generic.c +++ b/drivers/net/wan/hdlc_generic.c @@ -1,7 +1,7 @@ /* * Generic HDLC support routines for Linux * - * Copyright (C) 1999 - 2003 Krzysztof Halasa + * Copyright (C) 1999 - 2005 Krzysztof Halasa * * This program is free software; you can redistribute it and/or modify it * under the terms of version 2 of the GNU General Public License @@ -38,7 +38,7 @@ #include -static const char* version = "HDLC support module revision 1.17"; +static const char* version = "HDLC support module revision 1.18"; #undef DEBUG_LINK @@ -126,10 +126,13 @@ void hdlc_set_carrier(int on, struct net_device *dev) if (!hdlc->open) goto carrier_exit; - if (hdlc->carrier) + if (hdlc->carrier) { + printk(KERN_INFO "%s: Carrier detected\n", dev->name); __hdlc_set_carrier_on(dev); - else + } else { + printk(KERN_INFO "%s: Carrier lost\n", dev->name); __hdlc_set_carrier_off(dev); + } carrier_exit: spin_unlock_irqrestore(&hdlc->state_lock, flags); @@ -157,8 +160,11 @@ int hdlc_open(struct net_device *dev) spin_lock_irq(&hdlc->state_lock); - if (hdlc->carrier) + if (hdlc->carrier) { + printk(KERN_INFO "%s: Carrier detected\n", dev->name); __hdlc_set_carrier_on(dev); + } else + printk(KERN_INFO "%s: No carrier\n", dev->name); hdlc->open = 1; diff --git a/include/linux/hdlc.h b/include/linux/hdlc.h index 503194e62fe1..ed2927ef1ff7 100644 --- a/include/linux/hdlc.h +++ b/include/linux/hdlc.h @@ -1,7 +1,7 @@ /* * Generic HDLC support routines for Linux * - * Copyright (C) 1999-2003 Krzysztof Halasa + * Copyright (C) 1999-2005 Krzysztof Halasa * * This program is free software; you can redistribute it and/or modify it * under the terms of version 2 of the GNU General Public License @@ -41,6 +41,7 @@ #define LMI_NONE 1 /* No LMI, all PVCs are static */ #define LMI_ANSI 2 /* ANSI Annex D */ #define LMI_CCITT 3 /* ITU-T Annex A */ +#define LMI_CISCO 4 /* The "original" LMI, aka Gang of Four */ #define HDLC_MAX_MTU 1500 /* Ethernet 1500 bytes */ #define HDLC_MAX_MRU (HDLC_MAX_MTU + 10 + 14 + 4) /* for ETH+VLAN over FR */ @@ -89,6 +90,7 @@ typedef struct pvc_device_struct { unsigned int deleted: 1; unsigned int fecn: 1; unsigned int becn: 1; + unsigned int bandwidth; /* Cisco LMI reporting only */ }state; }pvc_device; -- cgit v1.2.3 From 3ec3b2fba526ead2fa3f3d7c91924f39a0733749 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Tue, 17 May 2005 12:08:48 +0100 Subject: AUDIT: Capture sys_socketcall arguments and sockaddrs Signed-off-by: David Woodhouse --- include/linux/audit.h | 7 ++++- kernel/auditsc.c | 73 +++++++++++++++++++++++++++++++++++++++++++++++++-- net/socket.c | 9 +++++-- 3 files changed, 84 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/audit.h b/include/linux/audit.h index 51e5879af7fc..2f5dc60f8bbd 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -69,8 +69,9 @@ #define AUDIT_FS_WATCH 1301 /* Filesystem watch event */ #define AUDIT_PATH 1302 /* Filname path information */ #define AUDIT_IPC 1303 /* IPC record */ -#define AUDIT_SOCKET 1304 /* Socket record */ +#define AUDIT_SOCKETCALL 1304 /* sys_socketcall arguments */ #define AUDIT_CONFIG_CHANGE 1305 /* Audit system configuration change */ +#define AUDIT_SOCKADDR 1306 /* sockaddr copied as syscall arg */ #define AUDIT_AVC 1400 /* SE Linux avc denial or grant */ #define AUDIT_SELINUX_ERR 1401 /* Internal SE Linux Errors */ @@ -235,6 +236,8 @@ extern int audit_get_stamp(struct audit_context *ctx, extern int audit_set_loginuid(struct task_struct *task, uid_t loginuid); extern uid_t audit_get_loginuid(struct audit_context *ctx); extern int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode); +extern int audit_socketcall(int nargs, unsigned long *args); +extern int audit_sockaddr(int len, void *addr); extern void audit_signal_info(int sig, struct task_struct *t); #else #define audit_alloc(t) ({ 0; }) @@ -248,6 +251,8 @@ extern void audit_signal_info(int sig, struct task_struct *t); #define audit_get_stamp(c,t,s) ({ 0; }) #define audit_get_loginuid(c) ({ -1; }) #define audit_ipc_perms(q,u,g,m) ({ 0; }) +#define audit_socketcall(n,a) ({ 0; }) +#define audit_sockaddr(len, addr) ({ 0; }) #define audit_signal_info(s,t) do { ; } while (0) #endif diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 773d28a3f701..818778d5b6ad 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -34,7 +34,7 @@ #include #include #include - +#include #include #include #include @@ -112,6 +112,18 @@ struct audit_aux_data_ipcctl { mode_t mode; }; +struct audit_aux_data_socketcall { + struct audit_aux_data d; + int nargs; + unsigned long args[0]; +}; + +struct audit_aux_data_sockaddr { + struct audit_aux_data d; + int len; + char a[0]; +}; + /* The per-task audit context. */ struct audit_context { @@ -694,7 +706,22 @@ static void audit_log_exit(struct audit_context *context) audit_log_format(ab, " qbytes=%lx iuid=%d igid=%d mode=%x", axi->qbytes, axi->uid, axi->gid, axi->mode); - } + break; } + + case AUDIT_SOCKETCALL: { + int i; + struct audit_aux_data_socketcall *axs = (void *)aux; + audit_log_format(ab, "nargs=%d", axs->nargs); + for (i=0; inargs; i++) + audit_log_format(ab, " a%d=%lx", i, axs->args[i]); + break; } + + case AUDIT_SOCKADDR: { + struct audit_aux_data_sockaddr *axs = (void *)aux; + + audit_log_format(ab, "saddr="); + audit_log_hex(ab, axs->a, axs->len); + break; } } audit_log_end(ab); @@ -1053,6 +1080,48 @@ int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode) return 0; } +int audit_socketcall(int nargs, unsigned long *args) +{ + struct audit_aux_data_socketcall *ax; + struct audit_context *context = current->audit_context; + + if (likely(!context)) + return 0; + + ax = kmalloc(sizeof(*ax) + nargs * sizeof(unsigned long), GFP_KERNEL); + if (!ax) + return -ENOMEM; + + ax->nargs = nargs; + memcpy(ax->args, args, nargs * sizeof(unsigned long)); + + ax->d.type = AUDIT_SOCKETCALL; + ax->d.next = context->aux; + context->aux = (void *)ax; + return 0; +} + +int audit_sockaddr(int len, void *a) +{ + struct audit_aux_data_sockaddr *ax; + struct audit_context *context = current->audit_context; + + if (likely(!context)) + return 0; + + ax = kmalloc(sizeof(*ax) + len, GFP_KERNEL); + if (!ax) + return -ENOMEM; + + ax->len = len; + memcpy(ax->a, a, len); + + ax->d.type = AUDIT_SOCKADDR; + ax->d.next = context->aux; + context->aux = (void *)ax; + return 0; +} + void audit_signal_info(int sig, struct task_struct *t) { extern pid_t audit_sig_pid; diff --git a/net/socket.c b/net/socket.c index cec0cb38b9ce..6b7c3b51a7c1 100644 --- a/net/socket.c +++ b/net/socket.c @@ -81,6 +81,7 @@ #include #include #include +#include #ifdef CONFIG_NET_RADIO #include /* Note : will define WIRELESS_EXT */ @@ -226,7 +227,7 @@ int move_addr_to_kernel(void __user *uaddr, int ulen, void *kaddr) return 0; if(copy_from_user(kaddr,uaddr,ulen)) return -EFAULT; - return 0; + return audit_sockaddr(ulen, kaddr); } /** @@ -1906,7 +1907,11 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args) /* copy_from_user should be SMP safe. */ if (copy_from_user(a, args, nargs[call])) return -EFAULT; - + + err = audit_socketcall(nargs[call]/sizeof(unsigned long), args); + if (err) + return err; + a0=a[0]; a1=a[1]; -- cgit v1.2.3 From 8eac3757158ccd9c6b43f44f228a5762fec33781 Mon Sep 17 00:00:00 2001 From: Russ Anderson <(rja@sgi.com)> Date: Mon, 16 May 2005 15:19:00 -0700 Subject: [IA64-SGI] Make Altix SAL call to POD reentrant Change the SAL call for POD mode to be reentrant. This change is SN specific. Signed-off-by: Russ Anderson (rja@sgi.com) Signed-off-by: Tony Luck --- include/asm-ia64/sn/sn_sal.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-ia64/sn/sn_sal.h b/include/asm-ia64/sn/sn_sal.h index 56d74ca76b5d..94cc5392533a 100644 --- a/include/asm-ia64/sn/sn_sal.h +++ b/include/asm-ia64/sn/sn_sal.h @@ -472,7 +472,7 @@ static inline u64 ia64_sn_pod_mode(void) { struct ia64_sal_retval isrv; - SAL_CALL(isrv, SN_SAL_POD_MODE, 0, 0, 0, 0, 0, 0, 0); + SAL_CALL_REENTRANT(isrv, SN_SAL_POD_MODE, 0, 0, 0, 0, 0, 0, 0); if (isrv.status) return 0; return isrv.v0; -- cgit v1.2.3 From 6872ec548970e9fb3ccd61013f84f9bb8b30fa9a Mon Sep 17 00:00:00 2001 From: Russ Anderson <(rja@sgi.com)> Date: Mon, 16 May 2005 15:30:00 -0700 Subject: [IS64-SGI] Set Altix error handling features The 2.6 kernel has CPE error thresholding. This patch lets SAL know of this error handling feature. The changes are SN specific. Signed-off-by: Russ Anderson (rja@sgi.com) Signed-off-by: Tony Luck --- arch/ia64/sn/kernel/setup.c | 2 ++ include/asm-ia64/sn/sn_sal.h | 26 ++++++++++++++++++++++++++ 2 files changed, 28 insertions(+) (limited to 'include') diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c index 4fb44984afe6..e64cb8175f7a 100644 --- a/arch/ia64/sn/kernel/setup.c +++ b/arch/ia64/sn/kernel/setup.c @@ -271,6 +271,8 @@ void __init sn_setup(char **cmdline_p) int major = sn_sal_rev_major(), minor = sn_sal_rev_minor(); extern void sn_cpu_init(void); + ia64_sn_plat_set_error_handling_features(); + /* * If the generic code has enabled vga console support - lets * get rid of it again. This is a kludge for the fact that ACPI diff --git a/include/asm-ia64/sn/sn_sal.h b/include/asm-ia64/sn/sn_sal.h index 94cc5392533a..eb0395ad0d6a 100644 --- a/include/asm-ia64/sn/sn_sal.h +++ b/include/asm-ia64/sn/sn_sal.h @@ -115,6 +115,13 @@ #define SAL_IROUTER_INTR_XMIT SAL_CONSOLE_INTR_XMIT #define SAL_IROUTER_INTR_RECV SAL_CONSOLE_INTR_RECV +/* + * Error Handling Features + */ +#define SAL_ERR_FEAT_MCA_SLV_TO_OS_INIT_SLV 0x1 +#define SAL_ERR_FEAT_LOG_SBES 0x2 +#define SAL_ERR_FEAT_MFR_OVERRIDE 0x4 +#define SAL_ERR_FEAT_SBE_THRESHOLD 0xffff0000 /* * SAL Error Codes @@ -341,6 +348,25 @@ ia64_sn_plat_cpei_handler(void) return ret_stuff.status; } +/* + * Set Error Handling Features + */ +static inline u64 +ia64_sn_plat_set_error_handling_features(void) +{ + struct ia64_sal_retval ret_stuff; + + ret_stuff.status = 0; + ret_stuff.v0 = 0; + ret_stuff.v1 = 0; + ret_stuff.v2 = 0; + SAL_CALL_REENTRANT(ret_stuff, SN_SAL_SET_ERROR_HANDLING_FEATURES, + (SAL_ERR_FEAT_MCA_SLV_TO_OS_INIT_SLV | SAL_ERR_FEAT_LOG_SBES), + 0, 0, 0, 0, 0, 0); + + return ret_stuff.status; +} + /* * Checks for console input. */ -- cgit v1.2.3 From 209aba03243ee42a22f8df8d08aa9963f62aec64 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 18 May 2005 10:21:07 +0100 Subject: AUDIT: Treat all user messages identically. It's silly to have to add explicit entries for new userspace messages as we invent them. Just treat all messages in the user range the same. Signed-off-by: David Woodhouse --- include/linux/audit.h | 17 ++--------------- kernel/audit.c | 20 ++------------------ security/selinux/nlmsgtab.c | 17 +++++++---------- 3 files changed, 11 insertions(+), 43 deletions(-) (limited to 'include') diff --git a/include/linux/audit.h b/include/linux/audit.h index 2f5dc60f8bbd..17ea5d522d81 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -51,14 +51,8 @@ #define AUDIT_WATCH_LIST 1009 /* List all file/dir watches */ #define AUDIT_SIGNAL_INFO 1010 /* Get info about sender of signal to auditd */ -#define AUDIT_USER_AUTH 1100 /* User space authentication */ -#define AUDIT_USER_ACCT 1101 /* User space acct change */ -#define AUDIT_USER_MGMT 1102 /* User space acct management */ -#define AUDIT_CRED_ACQ 1103 /* User space credential acquired */ -#define AUDIT_CRED_DISP 1104 /* User space credential disposed */ -#define AUDIT_USER_START 1105 /* User space session start */ -#define AUDIT_USER_END 1106 /* User space session end */ -#define AUDIT_USER_AVC 1107 /* User space avc message */ +#define AUDIT_FIRST_USER_MSG 1100 /* Userspace messages uninteresting to kernel */ +#define AUDIT_LAST_USER_MSG 1199 #define AUDIT_DAEMON_START 1200 /* Daemon startup record */ #define AUDIT_DAEMON_END 1201 /* Daemon normal stop record */ @@ -173,13 +167,6 @@ #define AUDIT_ARCH_V850 (EM_V850|__AUDIT_ARCH_LE) #define AUDIT_ARCH_X86_64 (EM_X86_64|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) -#ifndef __KERNEL__ -struct audit_message { - struct nlmsghdr nlh; - char data[1200]; -}; -#endif - struct audit_status { __u32 mask; /* Bit mask for valid entries */ __u32 enabled; /* 1 = enabled, 0 = disabled */ diff --git a/kernel/audit.c b/kernel/audit.c index a0e33b6897d7..e6d88635032c 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -325,15 +325,7 @@ static int audit_netlink_ok(kernel_cap_t eff_cap, u16 msg_type) if (!cap_raised(eff_cap, CAP_AUDIT_CONTROL)) err = -EPERM; break; - case AUDIT_USER: - case AUDIT_USER_AUTH: - case AUDIT_USER_ACCT: - case AUDIT_USER_MGMT: - case AUDIT_CRED_ACQ: - case AUDIT_CRED_DISP: - case AUDIT_USER_START: - case AUDIT_USER_END: - case AUDIT_USER_AVC: + case AUDIT_FIRST_USER_MSG...AUDIT_LAST_USER_MSG: if (!cap_raised(eff_cap, CAP_AUDIT_WRITE)) err = -EPERM; break; @@ -402,15 +394,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) audit_set_backlog_limit(status_get->backlog_limit, loginuid); break; - case AUDIT_USER: - case AUDIT_USER_AUTH: - case AUDIT_USER_ACCT: - case AUDIT_USER_MGMT: - case AUDIT_CRED_ACQ: - case AUDIT_CRED_DISP: - case AUDIT_USER_START: - case AUDIT_USER_END: - case AUDIT_USER_AVC: + case AUDIT_FIRST_USER_MSG...AUDIT_LAST_USER_MSG: ab = audit_log_start(NULL, msg_type); if (!ab) break; /* audit_panic has been called */ diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c index 67e77acc4795..f0fb6d76f7c5 100644 --- a/security/selinux/nlmsgtab.c +++ b/security/selinux/nlmsgtab.c @@ -98,14 +98,6 @@ static struct nlmsg_perm nlmsg_audit_perms[] = { AUDIT_DEL, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, { AUDIT_USER, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, { AUDIT_SIGNAL_INFO, NETLINK_AUDIT_SOCKET__NLMSG_READ }, - { AUDIT_USER_AUTH, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_USER_ACCT, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_USER_MGMT, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_CRED_ACQ, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_CRED_DISP, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_USER_START, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_USER_END, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, - { AUDIT_USER_AVC, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, }; @@ -150,8 +142,13 @@ int selinux_nlmsg_lookup(u16 sclass, u16 nlmsg_type, u32 *perm) break; case SECCLASS_NETLINK_AUDIT_SOCKET: - err = nlmsg_perm(nlmsg_type, perm, nlmsg_audit_perms, - sizeof(nlmsg_audit_perms)); + if (nlmsg_type >= AUDIT_FIRST_USER_MSG && + nlmsg_type <= AUDIT_LAST_USER_MSG) { + *perm = NETLINK_AUDIT_SOCKET__NLMSG_RELAY; + } else { + err = nlmsg_perm(nlmsg_type, perm, nlmsg_audit_perms, + sizeof(nlmsg_audit_perms)); + } break; /* No messaging from userspace, or class unknown/unhandled */ -- cgit v1.2.3 From a1ecf7f6e65637ba4470405ad39794710dbf85d4 Mon Sep 17 00:00:00 2001 From: Tony Luck Date: Wed, 18 May 2005 16:06:00 -0700 Subject: [IA64] alternate perfmon handler Patch from Charles Spirakis Some linux customers want to optimize their applications on the latest hardware but are not yet willing to upgrade to the latest kernel. This patch provides a way to plug in an alternate, basic, and GPL'ed PMU subsystem to help with their monitoring needs or for specialty work. It can also be used in case of serious unexpected bugs in perfmon. Mutual exclusion between the two subsystems is guaranteed, hence no conflict can arise from both subsystem being present. Acked-by: Stephane Eranian Signed-off-by: Tony Luck --- arch/ia64/kernel/perfmon.c | 175 +++++++++++++++++++++++++++++++++++++++++---- include/asm-ia64/perfmon.h | 8 +++ 2 files changed, 168 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index 71c101601e3e..ab9682f8b044 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -11,7 +11,7 @@ * Version Perfmon-2.x is a rewrite of perfmon-1.x * by Stephane Eranian, Hewlett Packard Co. * - * Copyright (C) 1999-2003, 2005 Hewlett Packard Co + * Copyright (C) 1999-2005 Hewlett Packard Co * Stephane Eranian * David Mosberger-Tang * @@ -497,6 +497,9 @@ typedef struct { static pfm_stats_t pfm_stats[NR_CPUS]; static pfm_session_t pfm_sessions; /* global sessions information */ +static spinlock_t pfm_alt_install_check; +static pfm_intr_handler_desc_t *pfm_alt_intr_handler; + static struct proc_dir_entry *perfmon_dir; static pfm_uuid_t pfm_null_uuid = {0,}; @@ -606,6 +609,7 @@ DEFINE_PER_CPU(unsigned long, pfm_syst_info); DEFINE_PER_CPU(struct task_struct *, pmu_owner); DEFINE_PER_CPU(pfm_context_t *, pmu_ctx); DEFINE_PER_CPU(unsigned long, pmu_activation_number); +EXPORT_SYMBOL_GPL(per_cpu__pfm_syst_info); /* forward declaration */ @@ -1325,7 +1329,7 @@ pfm_reserve_session(struct task_struct *task, int is_syswide, unsigned int cpu) error_conflict: DPRINT(("system wide not possible, conflicting session [%d] on CPU%d\n", pfm_sessions.pfs_sys_session[cpu]->pid, - smp_processor_id())); + cpu)); abort: UNLOCK_PFS(flags); @@ -5555,26 +5559,32 @@ pfm_interrupt_handler(int irq, void *arg, struct pt_regs *regs) int ret; this_cpu = get_cpu(); - min = pfm_stats[this_cpu].pfm_ovfl_intr_cycles_min; - max = pfm_stats[this_cpu].pfm_ovfl_intr_cycles_max; + if (likely(!pfm_alt_intr_handler)) { + min = pfm_stats[this_cpu].pfm_ovfl_intr_cycles_min; + max = pfm_stats[this_cpu].pfm_ovfl_intr_cycles_max; - start_cycles = ia64_get_itc(); + start_cycles = ia64_get_itc(); - ret = pfm_do_interrupt_handler(irq, arg, regs); + ret = pfm_do_interrupt_handler(irq, arg, regs); - total_cycles = ia64_get_itc(); + total_cycles = ia64_get_itc(); - /* - * don't measure spurious interrupts - */ - if (likely(ret == 0)) { - total_cycles -= start_cycles; + /* + * don't measure spurious interrupts + */ + if (likely(ret == 0)) { + total_cycles -= start_cycles; - if (total_cycles < min) pfm_stats[this_cpu].pfm_ovfl_intr_cycles_min = total_cycles; - if (total_cycles > max) pfm_stats[this_cpu].pfm_ovfl_intr_cycles_max = total_cycles; + if (total_cycles < min) pfm_stats[this_cpu].pfm_ovfl_intr_cycles_min = total_cycles; + if (total_cycles > max) pfm_stats[this_cpu].pfm_ovfl_intr_cycles_max = total_cycles; - pfm_stats[this_cpu].pfm_ovfl_intr_cycles += total_cycles; + pfm_stats[this_cpu].pfm_ovfl_intr_cycles += total_cycles; + } + } + else { + (*pfm_alt_intr_handler->handler)(irq, arg, regs); } + put_cpu_no_resched(); return IRQ_HANDLED; } @@ -6425,6 +6435,141 @@ static struct irqaction perfmon_irqaction = { .name = "perfmon" }; +static void +pfm_alt_save_pmu_state(void *data) +{ + struct pt_regs *regs; + + regs = ia64_task_regs(current); + + DPRINT(("called\n")); + + /* + * should not be necessary but + * let's take not risk + */ + pfm_clear_psr_up(); + pfm_clear_psr_pp(); + ia64_psr(regs)->pp = 0; + + /* + * This call is required + * May cause a spurious interrupt on some processors + */ + pfm_freeze_pmu(); + + ia64_srlz_d(); +} + +void +pfm_alt_restore_pmu_state(void *data) +{ + struct pt_regs *regs; + + regs = ia64_task_regs(current); + + DPRINT(("called\n")); + + /* + * put PMU back in state expected + * by perfmon + */ + pfm_clear_psr_up(); + pfm_clear_psr_pp(); + ia64_psr(regs)->pp = 0; + + /* + * perfmon runs with PMU unfrozen at all times + */ + pfm_unfreeze_pmu(); + + ia64_srlz_d(); +} + +int +pfm_install_alt_pmu_interrupt(pfm_intr_handler_desc_t *hdl) +{ + int ret, i; + int reserve_cpu; + + /* some sanity checks */ + if (hdl == NULL || hdl->handler == NULL) return -EINVAL; + + /* do the easy test first */ + if (pfm_alt_intr_handler) return -EBUSY; + + /* one at a time in the install or remove, just fail the others */ + if (!spin_trylock(&pfm_alt_install_check)) { + return -EBUSY; + } + + /* reserve our session */ + for_each_online_cpu(reserve_cpu) { + ret = pfm_reserve_session(NULL, 1, reserve_cpu); + if (ret) goto cleanup_reserve; + } + + /* save the current system wide pmu states */ + ret = on_each_cpu(pfm_alt_save_pmu_state, NULL, 0, 1); + if (ret) { + DPRINT(("on_each_cpu() failed: %d\n", ret)); + goto cleanup_reserve; + } + + /* officially change to the alternate interrupt handler */ + pfm_alt_intr_handler = hdl; + + spin_unlock(&pfm_alt_install_check); + + return 0; + +cleanup_reserve: + for_each_online_cpu(i) { + /* don't unreserve more than we reserved */ + if (i >= reserve_cpu) break; + + pfm_unreserve_session(NULL, 1, i); + } + + spin_unlock(&pfm_alt_install_check); + + return ret; +} +EXPORT_SYMBOL_GPL(pfm_install_alt_pmu_interrupt); + +int +pfm_remove_alt_pmu_interrupt(pfm_intr_handler_desc_t *hdl) +{ + int i; + int ret; + + if (hdl == NULL) return -EINVAL; + + /* cannot remove someone else's handler! */ + if (pfm_alt_intr_handler != hdl) return -EINVAL; + + /* one at a time in the install or remove, just fail the others */ + if (!spin_trylock(&pfm_alt_install_check)) { + return -EBUSY; + } + + pfm_alt_intr_handler = NULL; + + ret = on_each_cpu(pfm_alt_restore_pmu_state, NULL, 0, 1); + if (ret) { + DPRINT(("on_each_cpu() failed: %d\n", ret)); + } + + for_each_online_cpu(i) { + pfm_unreserve_session(NULL, 1, i); + } + + spin_unlock(&pfm_alt_install_check); + + return 0; +} +EXPORT_SYMBOL_GPL(pfm_remove_alt_pmu_interrupt); + /* * perfmon initialization routine, called from the initcall() table */ diff --git a/include/asm-ia64/perfmon.h b/include/asm-ia64/perfmon.h index ed5416c5b1ac..7f3333dd00e4 100644 --- a/include/asm-ia64/perfmon.h +++ b/include/asm-ia64/perfmon.h @@ -177,6 +177,10 @@ typedef union { extern long perfmonctl(int fd, int cmd, void *arg, int narg); +typedef struct { + void (*handler)(int irq, void *arg, struct pt_regs *regs); +} pfm_intr_handler_desc_t; + extern void pfm_save_regs (struct task_struct *); extern void pfm_load_regs (struct task_struct *); @@ -187,6 +191,10 @@ extern void pfm_syst_wide_update_task(struct task_struct *, unsigned long info, extern void pfm_inherit(struct task_struct *task, struct pt_regs *regs); extern void pfm_init_percpu(void); extern void pfm_handle_work(void); +extern int pfm_install_alt_pmu_interrupt(pfm_intr_handler_desc_t *h); +extern int pfm_remove_alt_pmu_interrupt(pfm_intr_handler_desc_t *h); + + /* * Reset PMD register flags -- cgit v1.2.3 From 867d1191fca388a79e4bb500dd85a9e871c96b99 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sun, 24 Apr 2005 02:06:05 -0500 Subject: [SCSI] remove requeue feature from blk_insert_request() blk_insert_request() has a unobivous feature of requeuing a request setting REQ_SPECIAL|REQ_SOFTBARRIER. SCSI midlayer was the only user and as previous patches removed the usage, remove the feature from blk_insert_request(). Only special requests should be queued with blk_insert_request(). All requeueing should go through blk_requeue_request(). Signed-off-by: Tejun Heo Signed-off-by: James Bottomley --- drivers/block/ll_rw_blk.c | 20 ++++++-------------- drivers/block/paride/pd.c | 2 +- drivers/block/sx8.c | 4 ++-- drivers/scsi/scsi_lib.c | 2 +- include/linux/blkdev.h | 2 +- 5 files changed, 11 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/drivers/block/ll_rw_blk.c b/drivers/block/ll_rw_blk.c index 11ef9d9ea139..f20eba22b14b 100644 --- a/drivers/block/ll_rw_blk.c +++ b/drivers/block/ll_rw_blk.c @@ -2038,7 +2038,6 @@ EXPORT_SYMBOL(blk_requeue_request); * @rq: request to be inserted * @at_head: insert request at head or tail of queue * @data: private data - * @reinsert: true if request it a reinsertion of previously processed one * * Description: * Many block devices need to execute commands asynchronously, so they don't @@ -2053,8 +2052,9 @@ EXPORT_SYMBOL(blk_requeue_request); * host that is unable to accept a particular command. */ void blk_insert_request(request_queue_t *q, struct request *rq, - int at_head, void *data, int reinsert) + int at_head, void *data) { + int where = at_head ? ELEVATOR_INSERT_FRONT : ELEVATOR_INSERT_BACK; unsigned long flags; /* @@ -2071,20 +2071,12 @@ void blk_insert_request(request_queue_t *q, struct request *rq, /* * If command is tagged, release the tag */ - if (reinsert) - blk_requeue_request(q, rq); - else { - int where = ELEVATOR_INSERT_BACK; - - if (at_head) - where = ELEVATOR_INSERT_FRONT; + if (blk_rq_tagged(rq)) + blk_queue_end_tag(q, rq); - if (blk_rq_tagged(rq)) - blk_queue_end_tag(q, rq); + drive_stat_acct(rq, rq->nr_sectors, 1); + __elv_add_request(q, rq, where, 0); - drive_stat_acct(rq, rq->nr_sectors, 1); - __elv_add_request(q, rq, where, 0); - } if (blk_queue_plugged(q)) __generic_unplug_device(q); else diff --git a/drivers/block/paride/pd.c b/drivers/block/paride/pd.c index 202a5a74ad37..fa49d62626ba 100644 --- a/drivers/block/paride/pd.c +++ b/drivers/block/paride/pd.c @@ -723,7 +723,7 @@ static int pd_special_command(struct pd_unit *disk, rq.ref_count = 1; rq.waiting = &wait; rq.end_io = blk_end_sync_rq; - blk_insert_request(disk->gd->queue, &rq, 0, func, 0); + blk_insert_request(disk->gd->queue, &rq, 0, func); wait_for_completion(&wait); rq.waiting = NULL; if (rq.errors) diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c index 797f5988c2b5..5ed3a6379452 100644 --- a/drivers/block/sx8.c +++ b/drivers/block/sx8.c @@ -614,7 +614,7 @@ static int carm_array_info (struct carm_host *host, unsigned int array_idx) spin_unlock_irq(&host->lock); DPRINTK("blk_insert_request, tag == %u\n", idx); - blk_insert_request(host->oob_q, crq->rq, 1, crq, 0); + blk_insert_request(host->oob_q, crq->rq, 1, crq); return 0; @@ -653,7 +653,7 @@ static int carm_send_special (struct carm_host *host, carm_sspc_t func) crq->msg_bucket = (u32) rc; DPRINTK("blk_insert_request, tag == %u\n", idx); - blk_insert_request(host->oob_q, crq->rq, 1, crq, 0); + blk_insert_request(host->oob_q, crq->rq, 1, crq); return 0; } diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 861d5f5c9722..47a4ad40bf4e 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -92,7 +92,7 @@ int scsi_insert_special_req(struct scsi_request *sreq, int at_head) */ sreq->sr_request->flags &= ~REQ_DONTPREP; blk_insert_request(sreq->sr_device->request_queue, sreq->sr_request, - at_head, sreq, 0); + at_head, sreq); return 0; } diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index ef1afc178c0a..4a99b76c5a33 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -544,7 +544,7 @@ extern void blk_end_sync_rq(struct request *rq); extern void blk_attempt_remerge(request_queue_t *, struct request *); extern void __blk_attempt_remerge(request_queue_t *, struct request *); extern struct request *blk_get_request(request_queue_t *, int, int); -extern void blk_insert_request(request_queue_t *, struct request *, int, void *, int); +extern void blk_insert_request(request_queue_t *, struct request *, int, void *); extern void blk_requeue_request(request_queue_t *, struct request *); extern void blk_plug_device(request_queue_t *); extern int blk_remove_plug(request_queue_t *); -- cgit v1.2.3 From daa6eda65a53e5addf86c6bc829129ff51b08bda Mon Sep 17 00:00:00 2001 From: Gerd Knorr Date: Tue, 10 May 2005 10:59:13 +0200 Subject: [SCSI] add scsi changer driver This patch adds a device driver for scsi media changer devices. Signed-off-by: Gerd Knorr Signed-off-by: James Bottomley --- Documentation/scsi/scsi-changer.txt | 180 ++++++ drivers/scsi/Kconfig | 18 + drivers/scsi/Makefile | 1 + drivers/scsi/ch.c | 1025 +++++++++++++++++++++++++++++++++++ include/linux/chio.h | 168 ++++++ include/linux/major.h | 1 + include/scsi/scsi.h | 3 + 7 files changed, 1396 insertions(+) create mode 100644 Documentation/scsi/scsi-changer.txt create mode 100644 drivers/scsi/ch.c create mode 100644 include/linux/chio.h (limited to 'include') diff --git a/Documentation/scsi/scsi-changer.txt b/Documentation/scsi/scsi-changer.txt new file mode 100644 index 000000000000..c132687b017a --- /dev/null +++ b/Documentation/scsi/scsi-changer.txt @@ -0,0 +1,180 @@ + +README for the SCSI media changer driver +======================================== + +This is a driver for SCSI Medium Changer devices, which are listed +with "Type: Medium Changer" in /proc/scsi/scsi. + +This is for *real* Jukeboxes. It is *not* supported to work with +common small CD-ROM changers, neither one-lun-per-slot SCSI changers +nor IDE drives. + +Userland tools available from here: + http://linux.bytesex.org/misc/changer.html + + +General Information +------------------- + +First some words about how changers work: A changer has 2 (possibly +more) SCSI ID's. One for the changer device which controls the robot, +and one for the device which actually reads and writes the data. The +later may be anything, a MOD, a CD-ROM, a tape or whatever. For the +changer device this is a "don't care", he *only* shuffles around the +media, nothing else. + + +The SCSI changer model is complex, compared to - for example - IDE-CD +changers. But it allows to handle nearly all possible cases. It knows +4 different types of changer elements: + + media transport - this one shuffles around the media, i.e. the + transport arm. Also known as "picker". + storage - a slot which can hold a media. + import/export - the same as above, but is accessable from outside, + i.e. there the operator (you !) can use this to + fill in and remove media from the changer. + Sometimes named "mailslot". + data transfer - this is the device which reads/writes, i.e. the + CD-ROM / Tape / whatever drive. + +None of these is limited to one: A huge Jukebox could have slots for +123 CD-ROM's, 5 CD-ROM readers (and therefore 6 SCSI ID's: the changer +and each CD-ROM) and 2 transport arms. No problem to handle. + + +How it is implemented +--------------------- + +I implemented the driver as character device driver with a NetBSD-like +ioctl interface. Just grabbed NetBSD's header file and one of the +other linux SCSI device drivers as starting point. The interface +should be source code compatible with NetBSD. So if there is any +software (anybody knows ???) which supports a BSDish changer driver, +it should work with this driver too. + +Over time a few more ioctls where added, volume tag support for example +wasn't covered by the NetBSD ioctl API. + + +Current State +------------- + +Support for more than one transport arm is not implemented yet (and +nobody asked for it so far...). + +I test and use the driver myself with a 35 slot cdrom jukebox from +Grundig. I got some reports telling it works ok with tape autoloaders +(Exabyte, HP and DEC). Some People use this driver with amanda. It +works fine with small (11 slots) and a huge (4 MOs, 88 slots) +magneto-optical Jukebox. Probably with lots of other changers too, most +(but not all :-) people mail me only if it does *not* work... + +I don't have any device lists, neither black-list nor white-list. Thus +it is quite useless to ask me whenever a specific device is supported or +not. In theory every changer device which supports the SCSI-2 media +changer command set should work out-of-the-box with this driver. If it +doesn't, it is a bug. Either within the driver or within the firmware +of the changer device. + + +Using it +-------- + +This is a character device with major number is 86, so use +"mknod /dev/sch0 c 86 0" to create the special file for the driver. + +If the module finds the changer, it prints some messages about the +device [ try "dmesg" if you don't see anything ] and should show up in +/proc/devices. If not.... some changers use ID ? / LUN 0 for the +device and ID ? / LUN 1 for the robot mechanism. But Linux does *not* +look for LUN's other than 0 as default, becauce there are to many +broken devices. So you can try: + + 1) echo "scsi add-single-device 0 0 ID 1" > /proc/scsi/scsi + (replace ID with the SCSI-ID of the device) + 2) boot the kernel with "max_scsi_luns=1" on the command line + (append="max_scsi_luns=1" in lilo.conf should do the trick) + + +Trouble? +-------- + +If you insmod the driver with "insmod debug=1", it will be verbose and +prints a lot of stuff to the syslog. Compiling the kernel with +CONFIG_SCSI_CONSTANTS=y improves the quality of the error messages alot +because the kernel will translate the error codes into human-readable +strings then. + +You can display these messages with the dmesg command (or check the +logfiles). If you email me some question becauce of a problem with the +driver, please include these messages. + + +Insmod options +-------------- + +debug=0/1 + Enable debug messages (see above, default: 0). + +verbose=0/1 + Be verbose (default: 1). + +init=0/1 + Send INITIALIZE ELEMENT STATUS command to the changer + at insmod time (default: 1). + +timeout_init= + timeout for the INITIALIZE ELEMENT STATUS command + (default: 3600). + +timeout_move= + timeout for all other commands (default: 120). + +dt_id=,,... +dt_lun=,,... + These two allow to specify the SCSI ID and LUN for the data + transfer elements. You likely don't need this as the jukebox + should provide this information. But some devices don't ... + +vendor_firsts= +vendor_counts= +vendor_labels= + These insmod options can be used to tell the driver that there + are some vendor-specific element types. Grundig for example + does this. Some jukeboxes have a printer to label fresh burned + CDs, which is addressed as element 0xc000 (type 5). To tell the + driver about this vendor-specific element, use this: + $ insmod ch \ + vendor_firsts=0xc000 \ + vendor_counts=1 \ + vendor_labels=printer + All three insmod options accept up to four comma-separated + values, this way you can configure the element types 5-8. + You likely need the SCSI specs for the device in question to + find the correct values as they are not covered by the SCSI-2 + standard. + + +Credits +------- + +I wrote this driver using the famous mailing-patches-around-the-world +method. With (more or less) help from: + + Daniel Moehwald + Dane Jasper + R. Scott Bailey + Jonathan Corbet + +Special thanks go to + Martin Kuehne +for a old, second-hand (but full functional) cdrom jukebox which I use +to develop/test driver and tools now. + +Have fun, + + Gerd + +-- +Gerd Knorr diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig index 2ef5aee86b29..ba88be399a59 100644 --- a/drivers/scsi/Kconfig +++ b/drivers/scsi/Kconfig @@ -137,6 +137,24 @@ config CHR_DEV_SG If unsure, say N. +config CHR_DEV_SCH + tristate "SCSI media changer support" + depends on SCSI + ---help--- + This is a driver for SCSI media changers. Most common devices are + tape libraries and MOD/CDROM jukeboxes. *Real* jukeboxes, you + don't need this for those tiny 6-slot cdrom changers. Media + changers are listed as "Type: Medium Changer" in /proc/scsi/scsi. + If you have such hardware and want to use it with linux, say Y + here. Check for details. + + If you want to compile this as a module ( = code which can be + inserted in and removed from the running kernel whenever you want), + say M here and read and + . The module will be called ch.o. + If unsure, say N. + + comment "Some SCSI devices (e.g. CD jukebox) support multiple LUNs" depends on SCSI diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile index 51d9c1e1884b..3746fb9fa2f5 100644 --- a/drivers/scsi/Makefile +++ b/drivers/scsi/Makefile @@ -140,6 +140,7 @@ obj-$(CONFIG_CHR_DEV_OSST) += osst.o obj-$(CONFIG_BLK_DEV_SD) += sd_mod.o obj-$(CONFIG_BLK_DEV_SR) += sr_mod.o obj-$(CONFIG_CHR_DEV_SG) += sg.o +obj-$(CONFIG_CHR_DEV_SCH) += ch.o scsi_mod-y += scsi.o hosts.o scsi_ioctl.o constants.o \ scsicam.o scsi_error.o scsi_lib.o \ diff --git a/drivers/scsi/ch.c b/drivers/scsi/ch.c new file mode 100644 index 000000000000..44f5a71ec34a --- /dev/null +++ b/drivers/scsi/ch.c @@ -0,0 +1,1025 @@ +/* + * SCSI Media Changer device driver for Linux 2.6 + * + * (c) 1996-2003 Gerd Knorr + * + */ + +#define VERSION "0.25" + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include /* here are all the ioctls */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#define CH_DT_MAX 16 +#define CH_TYPES 8 + +MODULE_DESCRIPTION("device driver for scsi media changer devices"); +MODULE_AUTHOR("Gerd Knorr "); +MODULE_LICENSE("GPL"); + +static int init = 1; +module_param(init, int, 0444); +MODULE_PARM_DESC(init, \ + "initialize element status on driver load (default: on)"); + +static int timeout_move = 300; +module_param(timeout_move, int, 0644); +MODULE_PARM_DESC(timeout_move,"timeout for move commands " + "(default: 300 seconds)"); + +static int timeout_init = 3600; +module_param(timeout_init, int, 0644); +MODULE_PARM_DESC(timeout_init,"timeout for INITIALIZE ELEMENT STATUS " + "(default: 3600 seconds)"); + +static int verbose = 1; +module_param(verbose, int, 0644); +MODULE_PARM_DESC(verbose,"be verbose (default: on)"); + +static int debug = 0; +module_param(debug, int, 0644); +MODULE_PARM_DESC(debug,"enable/disable debug messages, also prints more " + "detailed sense codes on scsi errors (default: off)"); + +static int dt_id[CH_DT_MAX] = { [ 0 ... (CH_DT_MAX-1) ] = -1 }; +static int dt_lun[CH_DT_MAX]; +module_param_array(dt_id, int, NULL, 0444); +module_param_array(dt_lun, int, NULL, 0444); + +/* tell the driver about vendor-specific slots */ +static int vendor_firsts[CH_TYPES-4]; +static int vendor_counts[CH_TYPES-4]; +module_param_array(vendor_firsts, int, NULL, 0444); +module_param_array(vendor_counts, int, NULL, 0444); + +static char *vendor_labels[CH_TYPES-4] = { + "v0", "v1", "v2", "v3" +}; +// module_param_string_array(vendor_labels, NULL, 0444); + +#define dprintk(fmt, arg...) if (debug) \ + printk(KERN_DEBUG "%s: " fmt, ch->name , ## arg) +#define vprintk(fmt, arg...) if (verbose) \ + printk(KERN_INFO "%s: " fmt, ch->name , ## arg) + +/* ------------------------------------------------------------------- */ + +#define MAX_RETRIES 1 + +static int ch_probe(struct device *); +static int ch_remove(struct device *); +static int ch_open(struct inode * inode, struct file * filp); +static int ch_release(struct inode * inode, struct file * filp); +static int ch_ioctl(struct inode * inode, struct file * filp, + unsigned int cmd, unsigned long arg); +#ifdef CONFIG_COMPAT +static long ch_ioctl_compat(struct file * filp, + unsigned int cmd, unsigned long arg); +#endif + +static struct class_simple * ch_sysfs_class; + +typedef struct { + struct list_head list; + int minor; + char name[8]; + struct scsi_device *device; + struct scsi_device **dt; /* ptrs to data transfer elements */ + u_int firsts[CH_TYPES]; + u_int counts[CH_TYPES]; + u_int unit_attention; + u_int voltags; + struct semaphore lock; +} scsi_changer; + +static LIST_HEAD(ch_devlist); +static spinlock_t ch_devlist_lock = SPIN_LOCK_UNLOCKED; +static int ch_devcount; + +static struct scsi_driver ch_template = +{ + .owner = THIS_MODULE, + .gendrv = { + .name = "ch", + .probe = ch_probe, + .remove = ch_remove, + }, +}; + +static struct file_operations changer_fops = +{ + .owner = THIS_MODULE, + .open = ch_open, + .release = ch_release, + .ioctl = ch_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = ch_ioctl_compat, +#endif +}; + +static struct { + unsigned char sense; + unsigned char asc; + unsigned char ascq; + int errno; +} err[] = { +/* Just filled in what looks right. Hav'nt checked any standard paper for + these errno assignments, so they may be wrong... */ + { + .sense = ILLEGAL_REQUEST, + .asc = 0x21, + .ascq = 0x01, + .errno = EBADSLT, /* Invalid element address */ + },{ + .sense = ILLEGAL_REQUEST, + .asc = 0x28, + .ascq = 0x01, + .errno = EBADE, /* Import or export element accessed */ + },{ + .sense = ILLEGAL_REQUEST, + .asc = 0x3B, + .ascq = 0x0D, + .errno = EXFULL, /* Medium destination element full */ + },{ + .sense = ILLEGAL_REQUEST, + .asc = 0x3B, + .ascq = 0x0E, + .errno = EBADE, /* Medium source element empty */ + },{ + .sense = ILLEGAL_REQUEST, + .asc = 0x20, + .ascq = 0x00, + .errno = EBADRQC, /* Invalid command operation code */ + },{ + /* end of list */ + } +}; + +/* ------------------------------------------------------------------- */ + +static int ch_find_errno(unsigned char *sense_buffer) +{ + int i,errno = 0; + + /* Check to see if additional sense information is available */ + if (sense_buffer[7] > 5 && + sense_buffer[12] != 0) { + for (i = 0; err[i].errno != 0; i++) { + if (err[i].sense == sense_buffer[ 2] && + err[i].asc == sense_buffer[12] && + err[i].ascq == sense_buffer[13]) { + errno = -err[i].errno; + break; + } + } + } + if (errno == 0) + errno = -EIO; + return errno; +} + +static int +ch_do_scsi(scsi_changer *ch, unsigned char *cmd, + void *buffer, unsigned buflength, + enum dma_data_direction direction) +{ + int errno, retries = 0, timeout; + struct scsi_request *sr; + + sr = scsi_allocate_request(ch->device, GFP_KERNEL); + if (NULL == sr) + return -ENOMEM; + + timeout = (cmd[0] == INITIALIZE_ELEMENT_STATUS) + ? timeout_init : timeout_move; + + retry: + errno = 0; + if (debug) { + dprintk("command: "); + __scsi_print_command(cmd); + } + + scsi_wait_req(sr, cmd, buffer, buflength, + timeout * HZ, MAX_RETRIES); + + dprintk("result: 0x%x\n",sr->sr_result); + if (driver_byte(sr->sr_result) & DRIVER_SENSE) { + if (debug) + scsi_print_req_sense(ch->name, sr); + errno = ch_find_errno(sr->sr_sense_buffer); + + switch(sr->sr_sense_buffer[2] & 0xf) { + case UNIT_ATTENTION: + ch->unit_attention = 1; + if (retries++ < 3) + goto retry; + break; + } + } + scsi_release_request(sr); + return errno; +} + +/* ------------------------------------------------------------------------ */ + +static int +ch_elem_to_typecode(scsi_changer *ch, u_int elem) +{ + int i; + + for (i = 0; i < CH_TYPES; i++) { + if (elem >= ch->firsts[i] && + elem < ch->firsts[i] + + ch->counts[i]) + return i+1; + } + return 0; +} + +static int +ch_read_element_status(scsi_changer *ch, u_int elem, char *data) +{ + u_char cmd[12]; + u_char *buffer; + int result; + + buffer = kmalloc(512, GFP_KERNEL | GFP_DMA); + if(!buffer) + return -ENOMEM; + + retry: + memset(cmd,0,sizeof(cmd)); + cmd[0] = READ_ELEMENT_STATUS; + cmd[1] = (ch->device->lun << 5) | + (ch->voltags ? 0x10 : 0) | + ch_elem_to_typecode(ch,elem); + cmd[2] = (elem >> 8) & 0xff; + cmd[3] = elem & 0xff; + cmd[5] = 1; + cmd[9] = 255; + if (0 == (result = ch_do_scsi(ch, cmd, buffer, 256, DMA_FROM_DEVICE))) { + if (((buffer[16] << 8) | buffer[17]) != elem) { + dprintk("asked for element 0x%02x, got 0x%02x\n", + elem,(buffer[16] << 8) | buffer[17]); + kfree(buffer); + return -EIO; + } + memcpy(data,buffer+16,16); + } else { + if (ch->voltags) { + ch->voltags = 0; + vprintk("device has no volume tag support\n"); + goto retry; + } + dprintk("READ ELEMENT STATUS for element 0x%x failed\n",elem); + } + kfree(buffer); + return result; +} + +static int +ch_init_elem(scsi_changer *ch) +{ + int err; + u_char cmd[6]; + + vprintk("INITIALIZE ELEMENT STATUS, may take some time ...\n"); + memset(cmd,0,sizeof(cmd)); + cmd[0] = INITIALIZE_ELEMENT_STATUS; + cmd[1] = ch->device->lun << 5; + err = ch_do_scsi(ch, cmd, NULL, 0, DMA_NONE); + vprintk("... finished\n"); + return err; +} + +static int +ch_readconfig(scsi_changer *ch) +{ + u_char cmd[10], data[16]; + u_char *buffer; + int result,id,lun,i; + u_int elem; + + buffer = kmalloc(512, GFP_KERNEL | GFP_DMA); + if (!buffer) + return -ENOMEM; + memset(buffer,0,512); + + memset(cmd,0,sizeof(cmd)); + cmd[0] = MODE_SENSE; + cmd[1] = ch->device->lun << 5; + cmd[2] = 0x1d; + cmd[4] = 255; + result = ch_do_scsi(ch, cmd, buffer, 255, DMA_FROM_DEVICE); + if (0 != result) { + cmd[1] |= (1<<3); + result = ch_do_scsi(ch, cmd, buffer, 255, DMA_FROM_DEVICE); + } + if (0 == result) { + ch->firsts[CHET_MT] = + (buffer[buffer[3]+ 6] << 8) | buffer[buffer[3]+ 7]; + ch->counts[CHET_MT] = + (buffer[buffer[3]+ 8] << 8) | buffer[buffer[3]+ 9]; + ch->firsts[CHET_ST] = + (buffer[buffer[3]+10] << 8) | buffer[buffer[3]+11]; + ch->counts[CHET_ST] = + (buffer[buffer[3]+12] << 8) | buffer[buffer[3]+13]; + ch->firsts[CHET_IE] = + (buffer[buffer[3]+14] << 8) | buffer[buffer[3]+15]; + ch->counts[CHET_IE] = + (buffer[buffer[3]+16] << 8) | buffer[buffer[3]+17]; + ch->firsts[CHET_DT] = + (buffer[buffer[3]+18] << 8) | buffer[buffer[3]+19]; + ch->counts[CHET_DT] = + (buffer[buffer[3]+20] << 8) | buffer[buffer[3]+21]; + vprintk("type #1 (mt): 0x%x+%d [medium transport]\n", + ch->firsts[CHET_MT], + ch->counts[CHET_MT]); + vprintk("type #2 (st): 0x%x+%d [storage]\n", + ch->firsts[CHET_ST], + ch->counts[CHET_ST]); + vprintk("type #3 (ie): 0x%x+%d [import/export]\n", + ch->firsts[CHET_IE], + ch->counts[CHET_IE]); + vprintk("type #4 (dt): 0x%x+%d [data transfer]\n", + ch->firsts[CHET_DT], + ch->counts[CHET_DT]); + } else { + vprintk("reading element address assigment page failed!\n"); + } + + /* vendor specific element types */ + for (i = 0; i < 4; i++) { + if (0 == vendor_counts[i]) + continue; + if (NULL == vendor_labels[i]) + continue; + ch->firsts[CHET_V1+i] = vendor_firsts[i]; + ch->counts[CHET_V1+i] = vendor_counts[i]; + vprintk("type #%d (v%d): 0x%x+%d [%s, vendor specific]\n", + i+5,i+1,vendor_firsts[i],vendor_counts[i], + vendor_labels[i]); + } + + /* look up the devices of the data transfer elements */ + ch->dt = kmalloc(ch->counts[CHET_DT]*sizeof(struct scsi_device), + GFP_KERNEL); + for (elem = 0; elem < ch->counts[CHET_DT]; elem++) { + id = -1; + lun = 0; + if (elem < CH_DT_MAX && -1 != dt_id[elem]) { + id = dt_id[elem]; + lun = dt_lun[elem]; + vprintk("dt 0x%x: [insmod option] ", + elem+ch->firsts[CHET_DT]); + } else if (0 != ch_read_element_status + (ch,elem+ch->firsts[CHET_DT],data)) { + vprintk("dt 0x%x: READ ELEMENT STATUS failed\n", + elem+ch->firsts[CHET_DT]); + } else { + vprintk("dt 0x%x: ",elem+ch->firsts[CHET_DT]); + if (data[6] & 0x80) { + if (verbose) + printk("not this SCSI bus\n"); + ch->dt[elem] = NULL; + } else if (0 == (data[6] & 0x30)) { + if (verbose) + printk("ID/LUN unknown\n"); + ch->dt[elem] = NULL; + } else { + id = ch->device->id; + lun = 0; + if (data[6] & 0x20) id = data[7]; + if (data[6] & 0x10) lun = data[6] & 7; + } + } + if (-1 != id) { + if (verbose) + printk("ID %i, LUN %i, ",id,lun); + ch->dt[elem] = + scsi_device_lookup(ch->device->host, + ch->device->channel, + id,lun); + if (!ch->dt[elem]) { + /* should not happen */ + if (verbose) + printk("Huh? device not found!\n"); + } else { + if (verbose) + printk("name: %8.8s %16.16s %4.4s\n", + ch->dt[elem]->vendor, + ch->dt[elem]->model, + ch->dt[elem]->rev); + } + } + } + ch->voltags = 1; + kfree(buffer); + + return 0; +} + +/* ------------------------------------------------------------------------ */ + +static int +ch_position(scsi_changer *ch, u_int trans, u_int elem, int rotate) +{ + u_char cmd[10]; + + dprintk("position: 0x%x\n",elem); + if (0 == trans) + trans = ch->firsts[CHET_MT]; + memset(cmd,0,sizeof(cmd)); + cmd[0] = POSITION_TO_ELEMENT; + cmd[1] = ch->device->lun << 5; + cmd[2] = (trans >> 8) & 0xff; + cmd[3] = trans & 0xff; + cmd[4] = (elem >> 8) & 0xff; + cmd[5] = elem & 0xff; + cmd[8] = rotate ? 1 : 0; + return ch_do_scsi(ch, cmd, NULL, 0, DMA_NONE); +} + +static int +ch_move(scsi_changer *ch, u_int trans, u_int src, u_int dest, int rotate) +{ + u_char cmd[12]; + + dprintk("move: 0x%x => 0x%x\n",src,dest); + if (0 == trans) + trans = ch->firsts[CHET_MT]; + memset(cmd,0,sizeof(cmd)); + cmd[0] = MOVE_MEDIUM; + cmd[1] = ch->device->lun << 5; + cmd[2] = (trans >> 8) & 0xff; + cmd[3] = trans & 0xff; + cmd[4] = (src >> 8) & 0xff; + cmd[5] = src & 0xff; + cmd[6] = (dest >> 8) & 0xff; + cmd[7] = dest & 0xff; + cmd[10] = rotate ? 1 : 0; + return ch_do_scsi(ch, cmd, NULL,0, DMA_NONE); +} + +static int +ch_exchange(scsi_changer *ch, u_int trans, u_int src, + u_int dest1, u_int dest2, int rotate1, int rotate2) +{ + u_char cmd[12]; + + dprintk("exchange: 0x%x => 0x%x => 0x%x\n", + src,dest1,dest2); + if (0 == trans) + trans = ch->firsts[CHET_MT]; + memset(cmd,0,sizeof(cmd)); + cmd[0] = EXCHANGE_MEDIUM; + cmd[1] = ch->device->lun << 5; + cmd[2] = (trans >> 8) & 0xff; + cmd[3] = trans & 0xff; + cmd[4] = (src >> 8) & 0xff; + cmd[5] = src & 0xff; + cmd[6] = (dest1 >> 8) & 0xff; + cmd[7] = dest1 & 0xff; + cmd[8] = (dest2 >> 8) & 0xff; + cmd[9] = dest2 & 0xff; + cmd[10] = (rotate1 ? 1 : 0) | (rotate2 ? 2 : 0); + + return ch_do_scsi(ch, cmd, NULL,0, DMA_NONE); +} + +static void +ch_check_voltag(char *tag) +{ + int i; + + for (i = 0; i < 32; i++) { + /* restrict to ascii */ + if (tag[i] >= 0x7f || tag[i] < 0x20) + tag[i] = ' '; + /* don't allow search wildcards */ + if (tag[i] == '?' || + tag[i] == '*') + tag[i] = ' '; + } +} + +static int +ch_set_voltag(scsi_changer *ch, u_int elem, + int alternate, int clear, u_char *tag) +{ + u_char cmd[12]; + u_char *buffer; + int result; + + buffer = kmalloc(512, GFP_KERNEL); + if (!buffer) + return -ENOMEM; + memset(buffer,0,512); + + dprintk("%s %s voltag: 0x%x => \"%s\"\n", + clear ? "clear" : "set", + alternate ? "alternate" : "primary", + elem, tag); + memset(cmd,0,sizeof(cmd)); + cmd[0] = SEND_VOLUME_TAG; + cmd[1] = (ch->device->lun << 5) | + ch_elem_to_typecode(ch,elem); + cmd[2] = (elem >> 8) & 0xff; + cmd[3] = elem & 0xff; + cmd[5] = clear + ? (alternate ? 0x0d : 0x0c) + : (alternate ? 0x0b : 0x0a); + + cmd[9] = 255; + + memcpy(buffer,tag,32); + ch_check_voltag(buffer); + + result = ch_do_scsi(ch, cmd, buffer, 256, DMA_TO_DEVICE); + kfree(buffer); + return result; +} + +static int ch_gstatus(scsi_changer *ch, int type, unsigned char *dest) +{ + int retval = 0; + u_char data[16]; + unsigned int i; + + down(&ch->lock); + for (i = 0; i < ch->counts[type]; i++) { + if (0 != ch_read_element_status + (ch, ch->firsts[type]+i,data)) { + retval = -EIO; + break; + } + put_user(data[2], dest+i); + if (data[2] & CESTATUS_EXCEPT) + vprintk("element 0x%x: asc=0x%x, ascq=0x%x\n", + ch->firsts[type]+i, + (int)data[4],(int)data[5]); + retval = ch_read_element_status + (ch, ch->firsts[type]+i,data); + if (0 != retval) + break; + } + up(&ch->lock); + return retval; +} + +/* ------------------------------------------------------------------------ */ + +static int +ch_release(struct inode *inode, struct file *file) +{ + scsi_changer *ch = file->private_data; + + scsi_device_put(ch->device); + file->private_data = NULL; + return 0; +} + +static int +ch_open(struct inode *inode, struct file *file) +{ + scsi_changer *tmp, *ch; + int minor = iminor(inode); + + spin_lock(&ch_devlist_lock); + ch = NULL; + list_for_each_entry(tmp,&ch_devlist,list) { + if (tmp->minor == minor) + ch = tmp; + } + if (NULL == ch || scsi_device_get(ch->device)) { + spin_unlock(&ch_devlist_lock); + return -ENXIO; + } + spin_unlock(&ch_devlist_lock); + + file->private_data = ch; + return 0; +} + +static int +ch_checkrange(scsi_changer *ch, unsigned int type, unsigned int unit) +{ + if (type >= CH_TYPES || unit >= ch->counts[type]) + return -1; + return 0; +} + +static int ch_ioctl(struct inode * inode, struct file * file, + unsigned int cmd, unsigned long arg) +{ + scsi_changer *ch = file->private_data; + int retval; + + switch (cmd) { + case CHIOGPARAMS: + { + struct changer_params params; + + params.cp_curpicker = 0; + params.cp_npickers = ch->counts[CHET_MT]; + params.cp_nslots = ch->counts[CHET_ST]; + params.cp_nportals = ch->counts[CHET_IE]; + params.cp_ndrives = ch->counts[CHET_DT]; + + if (copy_to_user((void *) arg, ¶ms, sizeof(params))) + return -EFAULT; + return 0; + } + case CHIOGVPARAMS: + { + struct changer_vendor_params vparams; + + memset(&vparams,0,sizeof(vparams)); + if (ch->counts[CHET_V1]) { + vparams.cvp_n1 = ch->counts[CHET_V1]; + strncpy(vparams.cvp_label1,vendor_labels[0],16); + } + if (ch->counts[CHET_V2]) { + vparams.cvp_n2 = ch->counts[CHET_V2]; + strncpy(vparams.cvp_label2,vendor_labels[1],16); + } + if (ch->counts[CHET_V3]) { + vparams.cvp_n3 = ch->counts[CHET_V3]; + strncpy(vparams.cvp_label3,vendor_labels[2],16); + } + if (ch->counts[CHET_V4]) { + vparams.cvp_n4 = ch->counts[CHET_V4]; + strncpy(vparams.cvp_label4,vendor_labels[3],16); + } + if (copy_to_user((void *) arg, &vparams, sizeof(vparams))) + return -EFAULT; + return 0; + } + + case CHIOPOSITION: + { + struct changer_position pos; + + if (copy_from_user(&pos, (void*)arg, sizeof (pos))) + return -EFAULT; + + if (0 != ch_checkrange(ch, pos.cp_type, pos.cp_unit)) { + dprintk("CHIOPOSITION: invalid parameter\n"); + return -EBADSLT; + } + down(&ch->lock); + retval = ch_position(ch,0, + ch->firsts[pos.cp_type] + pos.cp_unit, + pos.cp_flags & CP_INVERT); + up(&ch->lock); + return retval; + } + + case CHIOMOVE: + { + struct changer_move mv; + + if (copy_from_user(&mv, (void*)arg, sizeof (mv))) + return -EFAULT; + + if (0 != ch_checkrange(ch, mv.cm_fromtype, mv.cm_fromunit) || + 0 != ch_checkrange(ch, mv.cm_totype, mv.cm_tounit )) { + dprintk("CHIOMOVE: invalid parameter\n"); + return -EBADSLT; + } + + down(&ch->lock); + retval = ch_move(ch,0, + ch->firsts[mv.cm_fromtype] + mv.cm_fromunit, + ch->firsts[mv.cm_totype] + mv.cm_tounit, + mv.cm_flags & CM_INVERT); + up(&ch->lock); + return retval; + } + + case CHIOEXCHANGE: + { + struct changer_exchange mv; + + if (copy_from_user(&mv, (void*)arg, sizeof (mv))) + return -EFAULT; + + if (0 != ch_checkrange(ch, mv.ce_srctype, mv.ce_srcunit ) || + 0 != ch_checkrange(ch, mv.ce_fdsttype, mv.ce_fdstunit) || + 0 != ch_checkrange(ch, mv.ce_sdsttype, mv.ce_sdstunit)) { + dprintk("CHIOEXCHANGE: invalid parameter\n"); + return -EBADSLT; + } + + down(&ch->lock); + retval = ch_exchange + (ch,0, + ch->firsts[mv.ce_srctype] + mv.ce_srcunit, + ch->firsts[mv.ce_fdsttype] + mv.ce_fdstunit, + ch->firsts[mv.ce_sdsttype] + mv.ce_sdstunit, + mv.ce_flags & CE_INVERT1, mv.ce_flags & CE_INVERT2); + up(&ch->lock); + return retval; + } + + case CHIOGSTATUS: + { + struct changer_element_status ces; + + if (copy_from_user(&ces, (void*)arg, sizeof (ces))) + return -EFAULT; + if (ces.ces_type < 0 || ces.ces_type >= CH_TYPES) + return -EINVAL; + + return ch_gstatus(ch, ces.ces_type, ces.ces_data); + } + + case CHIOGELEM: + { + struct changer_get_element cge; + u_char cmd[12]; + u_char *buffer; + unsigned int elem; + int result,i; + + if (copy_from_user(&cge, (void*)arg, sizeof (cge))) + return -EFAULT; + + if (0 != ch_checkrange(ch, cge.cge_type, cge.cge_unit)) + return -EINVAL; + elem = ch->firsts[cge.cge_type] + cge.cge_unit; + + buffer = kmalloc(512, GFP_KERNEL | GFP_DMA); + if (!buffer) + return -ENOMEM; + down(&ch->lock); + + voltag_retry: + memset(cmd,0,sizeof(cmd)); + cmd[0] = READ_ELEMENT_STATUS; + cmd[1] = (ch->device->lun << 5) | + (ch->voltags ? 0x10 : 0) | + ch_elem_to_typecode(ch,elem); + cmd[2] = (elem >> 8) & 0xff; + cmd[3] = elem & 0xff; + cmd[5] = 1; + cmd[9] = 255; + + if (0 == (result = ch_do_scsi(ch, cmd, buffer, 256, DMA_FROM_DEVICE))) { + cge.cge_status = buffer[18]; + cge.cge_flags = 0; + if (buffer[18] & CESTATUS_EXCEPT) { + cge.cge_errno = EIO; + } + if (buffer[25] & 0x80) { + cge.cge_flags |= CGE_SRC; + if (buffer[25] & 0x40) + cge.cge_flags |= CGE_INVERT; + elem = (buffer[26]<<8) | buffer[27]; + for (i = 0; i < 4; i++) { + if (elem >= ch->firsts[i] && + elem < ch->firsts[i] + ch->counts[i]) { + cge.cge_srctype = i; + cge.cge_srcunit = elem-ch->firsts[i]; + } + } + } + if ((buffer[22] & 0x30) == 0x30) { + cge.cge_flags |= CGE_IDLUN; + cge.cge_id = buffer[23]; + cge.cge_lun = buffer[22] & 7; + } + if (buffer[9] & 0x80) { + cge.cge_flags |= CGE_PVOLTAG; + memcpy(cge.cge_pvoltag,buffer+28,36); + } + if (buffer[9] & 0x40) { + cge.cge_flags |= CGE_AVOLTAG; + memcpy(cge.cge_avoltag,buffer+64,36); + } + } else if (ch->voltags) { + ch->voltags = 0; + vprintk("device has no volume tag support\n"); + goto voltag_retry; + } + kfree(buffer); + up(&ch->lock); + + if (copy_to_user((void*)arg, &cge, sizeof (cge))) + return -EFAULT; + return result; + } + + case CHIOINITELEM: + { + down(&ch->lock); + retval = ch_init_elem(ch); + up(&ch->lock); + return retval; + } + + case CHIOSVOLTAG: + { + struct changer_set_voltag csv; + int elem; + + if (copy_from_user(&csv, (void*)arg, sizeof(csv))) + return -EFAULT; + + if (0 != ch_checkrange(ch, csv.csv_type, csv.csv_unit)) { + dprintk("CHIOSVOLTAG: invalid parameter\n"); + return -EBADSLT; + } + elem = ch->firsts[csv.csv_type] + csv.csv_unit; + down(&ch->lock); + retval = ch_set_voltag(ch, elem, + csv.csv_flags & CSV_AVOLTAG, + csv.csv_flags & CSV_CLEARTAG, + csv.csv_voltag); + up(&ch->lock); + return retval; + } + + default: + return scsi_ioctl(ch->device, cmd, (void*)arg); + + } +} + +#ifdef CONFIG_COMPAT + +struct changer_element_status32 { + int ces_type; + compat_uptr_t ces_data; +}; +#define CHIOGSTATUS32 _IOW('c', 8,struct changer_element_status32) + +static long ch_ioctl_compat(struct file * file, + unsigned int cmd, unsigned long arg) +{ + scsi_changer *ch = file->private_data; + + switch (cmd) { + case CHIOGPARAMS: + case CHIOGVPARAMS: + case CHIOPOSITION: + case CHIOMOVE: + case CHIOEXCHANGE: + case CHIOGELEM: + case CHIOINITELEM: + case CHIOSVOLTAG: + /* compatible */ + return ch_ioctl(NULL /* inode, unused */, + file, cmd, arg); + case CHIOGSTATUS32: + { + struct changer_element_status32 ces32; + unsigned char *data; + + if (copy_from_user(&ces32, (void*)arg, sizeof (ces32))) + return -EFAULT; + if (ces32.ces_type < 0 || ces32.ces_type >= CH_TYPES) + return -EINVAL; + + data = compat_ptr(ces32.ces_data); + return ch_gstatus(ch, ces32.ces_type, data); + } + default: + // return scsi_ioctl_compat(ch->device, cmd, (void*)arg); + return -ENOIOCTLCMD; + + } +} +#endif + +/* ------------------------------------------------------------------------ */ + +static int ch_probe(struct device *dev) +{ + struct scsi_device *sd = to_scsi_device(dev); + scsi_changer *ch; + + if (sd->type != TYPE_MEDIUM_CHANGER) + return -ENODEV; + + ch = kmalloc(sizeof(*ch), GFP_KERNEL); + if (NULL == ch) + return -ENOMEM; + + memset(ch,0,sizeof(*ch)); + ch->minor = ch_devcount; + sprintf(ch->name,"ch%d",ch->minor); + init_MUTEX(&ch->lock); + ch->device = sd; + ch_readconfig(ch); + if (init) + ch_init_elem(ch); + + devfs_mk_cdev(MKDEV(SCSI_CHANGER_MAJOR,ch->minor), + S_IFCHR | S_IRUGO | S_IWUGO, ch->name); + class_simple_device_add(ch_sysfs_class, + MKDEV(SCSI_CHANGER_MAJOR,ch->minor), + dev, "s%s", ch->name); + + printk(KERN_INFO "Attached scsi changer %s " + "at scsi%d, channel %d, id %d, lun %d\n", + ch->name, sd->host->host_no, sd->channel, sd->id, sd->lun); + + spin_lock(&ch_devlist_lock); + list_add_tail(&ch->list,&ch_devlist); + ch_devcount++; + spin_unlock(&ch_devlist_lock); + return 0; +} + +static int ch_remove(struct device *dev) +{ + struct scsi_device *sd = to_scsi_device(dev); + scsi_changer *tmp, *ch; + + spin_lock(&ch_devlist_lock); + ch = NULL; + list_for_each_entry(tmp,&ch_devlist,list) { + if (tmp->device == sd) + ch = tmp; + } + BUG_ON(NULL == ch); + list_del(&ch->list); + spin_unlock(&ch_devlist_lock); + + class_simple_device_remove(MKDEV(SCSI_CHANGER_MAJOR,ch->minor)); + devfs_remove(ch->name); + kfree(ch->dt); + kfree(ch); + ch_devcount--; + return 0; +} + +static int __init init_ch_module(void) +{ + int rc; + + printk(KERN_INFO "SCSI Media Changer driver v" VERSION " \n"); + ch_sysfs_class = class_simple_create(THIS_MODULE, "scsi_changer"); + if (IS_ERR(ch_sysfs_class)) { + rc = PTR_ERR(ch_sysfs_class); + return rc; + } + rc = register_chrdev(SCSI_CHANGER_MAJOR,"ch",&changer_fops); + if (rc < 0) { + printk("Unable to get major %d for SCSI-Changer\n", + SCSI_CHANGER_MAJOR); + goto fail1; + } + rc = scsi_register_driver(&ch_template.gendrv); + if (rc < 0) + goto fail2; + return 0; + + fail2: + unregister_chrdev(SCSI_CHANGER_MAJOR, "ch"); + fail1: + class_simple_destroy(ch_sysfs_class); + return rc; +} + +static void __exit exit_ch_module(void) +{ + scsi_unregister_driver(&ch_template.gendrv); + unregister_chrdev(SCSI_CHANGER_MAJOR, "ch"); + class_simple_destroy(ch_sysfs_class); +} + +module_init(init_ch_module); +module_exit(exit_ch_module); + +/* + * Local variables: + * c-basic-offset: 8 + * End: + */ diff --git a/include/linux/chio.h b/include/linux/chio.h new file mode 100644 index 000000000000..63035ae67e63 --- /dev/null +++ b/include/linux/chio.h @@ -0,0 +1,168 @@ +/* + * ioctl interface for the scsi media changer driver + */ + +/* changer element types */ +#define CHET_MT 0 /* media transport element (robot) */ +#define CHET_ST 1 /* storage element (media slots) */ +#define CHET_IE 2 /* import/export element */ +#define CHET_DT 3 /* data transfer element (tape/cdrom/whatever) */ +#define CHET_V1 4 /* vendor specific #1 */ +#define CHET_V2 5 /* vendor specific #2 */ +#define CHET_V3 6 /* vendor specific #3 */ +#define CHET_V4 7 /* vendor specific #4 */ + + +/* + * CHIOGPARAMS + * query changer properties + * + * CHIOVGPARAMS + * query vendor-specific element types + * + * accessing elements works by specifing type and unit of the element. + * for eample, storage elements are addressed with type = CHET_ST and + * unit = 0 .. cp_nslots-1 + * + */ +struct changer_params { + int cp_curpicker; /* current transport element */ + int cp_npickers; /* number of transport elements (CHET_MT) */ + int cp_nslots; /* number of storage elements (CHET_ST) */ + int cp_nportals; /* number of import/export elements (CHET_IE) */ + int cp_ndrives; /* number of data transfer elements (CHET_DT) */ +}; +struct changer_vendor_params { + int cvp_n1; /* number of vendor specific elems (CHET_V1) */ + char cvp_label1[16]; + int cvp_n2; /* number of vendor specific elems (CHET_V2) */ + char cvp_label2[16]; + int cvp_n3; /* number of vendor specific elems (CHET_V3) */ + char cvp_label3[16]; + int cvp_n4; /* number of vendor specific elems (CHET_V4) */ + char cvp_label4[16]; + int reserved[8]; +}; + + +/* + * CHIOMOVE + * move a medium from one element to another + */ +struct changer_move { + int cm_fromtype; /* type/unit of source element */ + int cm_fromunit; + int cm_totype; /* type/unit of destination element */ + int cm_tounit; + int cm_flags; +}; +#define CM_INVERT 1 /* flag: rotate media (for double-sided like MOD) */ + + +/* + * CHIOEXCHANGE + * move one medium from element #1 to element #2, + * and another one from element #2 to element #3. + * element #1 and #3 are allowed to be identical. + */ +struct changer_exchange { + int ce_srctype; /* type/unit of element #1 */ + int ce_srcunit; + int ce_fdsttype; /* type/unit of element #2 */ + int ce_fdstunit; + int ce_sdsttype; /* type/unit of element #3 */ + int ce_sdstunit; + int ce_flags; +}; +#define CE_INVERT1 1 +#define CE_INVERT2 2 + + +/* + * CHIOPOSITION + * move the transport element (robot arm) to a specific element. + */ +struct changer_position { + int cp_type; + int cp_unit; + int cp_flags; +}; +#define CP_INVERT 1 + + +/* + * CHIOGSTATUS + * get element status for all elements of a specific type + */ +struct changer_element_status { + int ces_type; + unsigned char *ces_data; +}; +#define CESTATUS_FULL 0x01 /* full */ +#define CESTATUS_IMPEXP 0x02 /* media was imported (inserted by sysop) */ +#define CESTATUS_EXCEPT 0x04 /* error condition */ +#define CESTATUS_ACCESS 0x08 /* access allowed */ +#define CESTATUS_EXENAB 0x10 /* element can export media */ +#define CESTATUS_INENAB 0x20 /* element can import media */ + + +/* + * CHIOGELEM + * get more detailed status informtion for a single element + */ +struct changer_get_element { + int cge_type; /* type/unit */ + int cge_unit; + int cge_status; /* status */ + int cge_errno; /* errno */ + int cge_srctype; /* source element of the last move/exchange */ + int cge_srcunit; + int cge_id; /* scsi id (for data transfer elements) */ + int cge_lun; /* scsi lun (for data transfer elements) */ + char cge_pvoltag[36]; /* primary volume tag */ + char cge_avoltag[36]; /* alternate volume tag */ + int cge_flags; +}; +/* flags */ +#define CGE_ERRNO 0x01 /* errno available */ +#define CGE_INVERT 0x02 /* media inverted */ +#define CGE_SRC 0x04 /* media src available */ +#define CGE_IDLUN 0x08 /* ID+LUN available */ +#define CGE_PVOLTAG 0x10 /* primary volume tag available */ +#define CGE_AVOLTAG 0x20 /* alternate volume tag available */ + + +/* + * CHIOSVOLTAG + * set volume tag + */ +struct changer_set_voltag { + int csv_type; /* type/unit */ + int csv_unit; + char csv_voltag[36]; /* volume tag */ + int csv_flags; +}; +#define CSV_PVOLTAG 0x01 /* primary volume tag */ +#define CSV_AVOLTAG 0x02 /* alternate volume tag */ +#define CSV_CLEARTAG 0x04 /* clear volume tag */ + +/* ioctls */ +#define CHIOMOVE _IOW('c', 1,struct changer_move) +#define CHIOEXCHANGE _IOW('c', 2,struct changer_exchange) +#define CHIOPOSITION _IOW('c', 3,struct changer_position) +#define CHIOGPICKER _IOR('c', 4,int) /* not impl. */ +#define CHIOSPICKER _IOW('c', 5,int) /* not impl. */ +#define CHIOGPARAMS _IOR('c', 6,struct changer_params) +#define CHIOGSTATUS _IOW('c', 8,struct changer_element_status) +#define CHIOGELEM _IOW('c',16,struct changer_get_element) +#define CHIOINITELEM _IO('c',17) +#define CHIOSVOLTAG _IOW('c',18,struct changer_set_voltag) +#define CHIOGVPARAMS _IOR('c',19,struct changer_vendor_params) + +/* ---------------------------------------------------------------------- */ + +/* + * Local variables: + * c-basic-offset: 8 + * End: + */ diff --git a/include/linux/major.h b/include/linux/major.h index 4b62c42b842c..e36a46702d94 100644 --- a/include/linux/major.h +++ b/include/linux/major.h @@ -100,6 +100,7 @@ #define I2O_MAJOR 80 /* 80->87 */ #define SHMIQ_MAJOR 85 /* Linux/mips, SGI /dev/shmiq */ +#define SCSI_CHANGER_MAJOR 86 #define IDE6_MAJOR 88 #define IDE7_MAJOR 89 diff --git a/include/scsi/scsi.h b/include/scsi/scsi.h index 659ecf48fb4a..ca1e3b4a3183 100644 --- a/include/scsi/scsi.h +++ b/include/scsi/scsi.h @@ -41,6 +41,7 @@ extern const char *const scsi_device_types[MAX_SCSI_DEVICE_CODE]; #define FORMAT_UNIT 0x04 #define READ_BLOCK_LIMITS 0x05 #define REASSIGN_BLOCKS 0x07 +#define INITIALIZE_ELEMENT_STATUS 0x07 #define READ_6 0x08 #define WRITE_6 0x0a #define SEEK_6 0x0b @@ -65,6 +66,7 @@ extern const char *const scsi_device_types[MAX_SCSI_DEVICE_CODE]; #define READ_10 0x28 #define WRITE_10 0x2a #define SEEK_10 0x2b +#define POSITION_TO_ELEMENT 0x2b #define WRITE_VERIFY 0x2e #define VERIFY 0x2f #define SEARCH_HIGH 0x30 @@ -97,6 +99,7 @@ extern const char *const scsi_device_types[MAX_SCSI_DEVICE_CODE]; #define PERSISTENT_RESERVE_OUT 0x5f #define REPORT_LUNS 0xa0 #define MOVE_MEDIUM 0xa5 +#define EXCHANGE_MEDIUM 0xa6 #define READ_12 0xa8 #define WRITE_12 0xaa #define WRITE_VERIFY_12 0xae -- cgit v1.2.3 From 011161051bbc25f7f8b7df059dbd934c534443f0 Mon Sep 17 00:00:00 2001 From: Stephen Smalley Date: Sat, 21 May 2005 00:15:52 +0100 Subject: AUDIT: Avoid sleeping function in SElinux AVC audit. This patch changes the SELinux AVC to defer logging of paths to the audit framework upon syscall exit, by saving a reference to the (dentry,vfsmount) pair in an auxiliary audit item on the current audit context for processing by audit_log_exit. Signed-off-by: Stephen Smalley Signed-off-by: David Woodhouse --- include/linux/audit.h | 3 +++ kernel/auditsc.c | 40 ++++++++++++++++++++++++++++++++++++++++ security/selinux/avc.c | 17 ++++++++--------- 3 files changed, 51 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/linux/audit.h b/include/linux/audit.h index 17ea5d522d81..4b7caf0c6e10 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -69,6 +69,7 @@ #define AUDIT_AVC 1400 /* SE Linux avc denial or grant */ #define AUDIT_SELINUX_ERR 1401 /* Internal SE Linux Errors */ +#define AUDIT_AVC_PATH 1402 /* dentry, vfsmount pair from avc */ #define AUDIT_KERNEL 2000 /* Asynchronous audit record. NOT A REQUEST. */ @@ -225,6 +226,7 @@ extern uid_t audit_get_loginuid(struct audit_context *ctx); extern int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode); extern int audit_socketcall(int nargs, unsigned long *args); extern int audit_sockaddr(int len, void *addr); +extern int audit_avc_path(struct dentry *dentry, struct vfsmount *mnt); extern void audit_signal_info(int sig, struct task_struct *t); #else #define audit_alloc(t) ({ 0; }) @@ -240,6 +242,7 @@ extern void audit_signal_info(int sig, struct task_struct *t); #define audit_ipc_perms(q,u,g,m) ({ 0; }) #define audit_socketcall(n,a) ({ 0; }) #define audit_sockaddr(len, addr) ({ 0; }) +#define audit_avc_path(dentry, mnt) ({ 0; }) #define audit_signal_info(s,t) do { ; } while (0) #endif diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 78d7a13fc86f..8dc5b2767145 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -34,6 +34,7 @@ #include #include #include +#include #include #include #include @@ -124,6 +125,11 @@ struct audit_aux_data_sockaddr { char a[0]; }; +struct audit_aux_data_path { + struct audit_aux_data d; + struct dentry *dentry; + struct vfsmount *mnt; +}; /* The per-task audit context. */ struct audit_context { @@ -553,6 +559,11 @@ static inline void audit_free_aux(struct audit_context *context) struct audit_aux_data *aux; while ((aux = context->aux)) { + if (aux->type == AUDIT_AVC_PATH) { + struct audit_aux_data_path *axi = (void *)aux; + dput(axi->dentry); + mntput(axi->mnt); + } context->aux = aux->next; kfree(aux); } @@ -724,6 +735,14 @@ static void audit_log_exit(struct audit_context *context) audit_log_format(ab, "saddr="); audit_log_hex(ab, axs->a, axs->len); break; } + + case AUDIT_AVC_PATH: { + struct audit_aux_data_path *axi = (void *)aux; + audit_log_d_path(ab, "path=", axi->dentry, axi->mnt); + dput(axi->dentry); + mntput(axi->mnt); + break; } + } audit_log_end(ab); @@ -1124,6 +1143,27 @@ int audit_sockaddr(int len, void *a) return 0; } +int audit_avc_path(struct dentry *dentry, struct vfsmount *mnt) +{ + struct audit_aux_data_path *ax; + struct audit_context *context = current->audit_context; + + if (likely(!context)) + return 0; + + ax = kmalloc(sizeof(*ax), GFP_ATOMIC); + if (!ax) + return -ENOMEM; + + ax->dentry = dget(dentry); + ax->mnt = mntget(mnt); + + ax->d.type = AUDIT_AVC_PATH; + ax->d.next = context->aux; + context->aux = (void *)ax; + return 0; +} + void audit_signal_info(int sig, struct task_struct *t) { extern pid_t audit_sig_pid; diff --git a/security/selinux/avc.c b/security/selinux/avc.c index 62b963aca275..0fbc3e98c5ea 100644 --- a/security/selinux/avc.c +++ b/security/selinux/avc.c @@ -573,13 +573,10 @@ void avc_audit(u32 ssid, u32 tsid, case AVC_AUDIT_DATA_FS: if (a->u.fs.dentry) { struct dentry *dentry = a->u.fs.dentry; - if (a->u.fs.mnt) { - audit_log_d_path(ab, "path=", dentry, - a->u.fs.mnt); - } else { - audit_log_format(ab, " name=%s", - dentry->d_name.name); - } + if (a->u.fs.mnt) + audit_avc_path(dentry, a->u.fs.mnt); + audit_log_format(ab, " name=%s", + dentry->d_name.name); inode = dentry->d_inode; } else if (a->u.fs.inode) { struct dentry *dentry; @@ -630,8 +627,10 @@ void avc_audit(u32 ssid, u32 tsid, case AF_UNIX: u = unix_sk(sk); if (u->dentry) { - audit_log_d_path(ab, "path=", - u->dentry, u->mnt); + audit_avc_path(u->dentry, u->mnt); + audit_log_format(ab, " name=%s", + u->dentry->d_name.name); + break; } if (!u->addr) -- cgit v1.2.3 From bfb4496e7239c9132d732a65cdcf3d6a7431ad1a Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Sat, 21 May 2005 21:08:09 +0100 Subject: AUDIT: Assign serial number to non-syscall messages Move audit_serial() into audit.c and use it to generate serial numbers on messages even when there is no audit context from syscall auditing. This allows us to disambiguate audit records when more than one is generated in the same millisecond. Based on a patch by Steve Grubb after he observed the problem. Signed-off-by: David Woodhouse --- include/linux/audit.h | 7 ++++--- kernel/audit.c | 46 ++++++++++++++++++++++++++++++++++++++++++---- kernel/auditsc.c | 46 ++++++---------------------------------------- 3 files changed, 52 insertions(+), 47 deletions(-) (limited to 'include') diff --git a/include/linux/audit.h b/include/linux/audit.h index 4b7caf0c6e10..3278ddf41ce6 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -219,8 +219,9 @@ extern void audit_inode(const char *name, const struct inode *inode); /* Private API (for audit.c only) */ extern int audit_receive_filter(int type, int pid, int uid, int seq, void *data, uid_t loginuid); -extern int audit_get_stamp(struct audit_context *ctx, - struct timespec *t, unsigned int *serial); +extern unsigned int audit_serial(void); +extern void auditsc_get_stamp(struct audit_context *ctx, + struct timespec *t, unsigned int *serial); extern int audit_set_loginuid(struct task_struct *task, uid_t loginuid); extern uid_t audit_get_loginuid(struct audit_context *ctx); extern int audit_ipc_perms(unsigned long qbytes, uid_t uid, gid_t gid, mode_t mode); @@ -237,7 +238,7 @@ extern void audit_signal_info(int sig, struct task_struct *t); #define audit_putname(n) do { ; } while (0) #define audit_inode(n,i) do { ; } while (0) #define audit_receive_filter(t,p,u,s,d,l) ({ -EOPNOTSUPP; }) -#define audit_get_stamp(c,t,s) ({ 0; }) +#define auditsc_get_stamp(c,t,s) do { BUG(); } while (0) #define audit_get_loginuid(c) ({ -1; }) #define audit_ipc_perms(q,u,g,m) ({ 0; }) #define audit_socketcall(n,a) ({ 0; }) diff --git a/kernel/audit.c b/kernel/audit.c index f0a003acf621..35306f4369e7 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -597,6 +597,47 @@ err: return NULL; } +/* Compute a serial number for the audit record. Audit records are + * written to user-space as soon as they are generated, so a complete + * audit record may be written in several pieces. The timestamp of the + * record and this serial number are used by the user-space tools to + * determine which pieces belong to the same audit record. The + * (timestamp,serial) tuple is unique for each syscall and is live from + * syscall entry to syscall exit. + * + * Atomic values are only guaranteed to be 24-bit, so we count down. + * + * NOTE: Another possibility is to store the formatted records off the + * audit context (for those records that have a context), and emit them + * all at syscall exit. However, this could delay the reporting of + * significant errors until syscall exit (or never, if the system + * halts). */ +unsigned int audit_serial(void) +{ + static atomic_t serial = ATOMIC_INIT(0xffffff); + unsigned int a, b; + + do { + a = atomic_read(&serial); + if (atomic_dec_and_test(&serial)) + atomic_set(&serial, 0xffffff); + b = atomic_read(&serial); + } while (b != a - 1); + + return 0xffffff - b; +} + +static inline void audit_get_stamp(struct audit_context *ctx, + struct timespec *t, unsigned int *serial) +{ + if (ctx) + auditsc_get_stamp(ctx, t, serial); + else { + *t = CURRENT_TIME; + *serial = audit_serial(); + } +} + /* Obtain an audit buffer. This routine does locking to obtain the * audit buffer, but then no locking is required for calls to * audit_log_*format. If the tsk is a task that is currently in a @@ -630,10 +671,7 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, int type) return NULL; } - if (!audit_get_stamp(ab->ctx, &t, &serial)) { - t = CURRENT_TIME; - serial = 0; - } + audit_get_stamp(ab->ctx, &t, &serial); audit_log_format(ab, "audit(%lu.%03lu:%u): ", t.tv_sec, t.tv_nsec/1000000, serial); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 4193811d4fe1..74c2ae804ca8 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -795,36 +795,6 @@ void audit_free(struct task_struct *tsk) audit_free_context(context); } -/* Compute a serial number for the audit record. Audit records are - * written to user-space as soon as they are generated, so a complete - * audit record may be written in several pieces. The timestamp of the - * record and this serial number are used by the user-space tools to - * determine which pieces belong to the same audit record. The - * (timestamp,serial) tuple is unique for each syscall and is live from - * syscall entry to syscall exit. - * - * Atomic values are only guaranteed to be 24-bit, so we count down. - * - * NOTE: Another possibility is to store the formatted records off the - * audit context (for those records that have a context), and emit them - * all at syscall exit. However, this could delay the reporting of - * significant errors until syscall exit (or never, if the system - * halts). */ -static inline unsigned int audit_serial(void) -{ - static atomic_t serial = ATOMIC_INIT(0xffffff); - unsigned int a, b; - - do { - a = atomic_read(&serial); - if (atomic_dec_and_test(&serial)) - atomic_set(&serial, 0xffffff); - b = atomic_read(&serial); - } while (b != a - 1); - - return 0xffffff - b; -} - /* Fill in audit context at syscall entry. This only happens if the * audit context was created when the task was created and the state or * filters demand the audit context be built. If the state from the @@ -1042,17 +1012,13 @@ void audit_inode(const char *name, const struct inode *inode) context->names[idx].rdev = inode->i_rdev; } -int audit_get_stamp(struct audit_context *ctx, - struct timespec *t, unsigned int *serial) +void auditsc_get_stamp(struct audit_context *ctx, + struct timespec *t, unsigned int *serial) { - if (ctx) { - t->tv_sec = ctx->ctime.tv_sec; - t->tv_nsec = ctx->ctime.tv_nsec; - *serial = ctx->serial; - ctx->auditable = 1; - return 1; - } - return 0; + t->tv_sec = ctx->ctime.tv_sec; + t->tv_nsec = ctx->ctime.tv_nsec; + *serial = ctx->serial; + ctx->auditable = 1; } int audit_set_loginuid(struct task_struct *task, uid_t loginuid) -- cgit v1.2.3 From 631e8a1398ce4cfef8b30678d51daf0c64313a09 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 16 May 2005 01:59:55 +0100 Subject: [SCSI] TYPE_RBC cache fixes (sbp2.c affected) a) TYPE_SDAD renamed to TYPE_RBC and taken to scsi.h b) in sbp2.c remapping of TYPE_RPB to TYPE_DISK turned off c) relevant places in midlayer and sd.c taught to accept TYPE_RBC d) sd.c::sd_read_cache_type() looks into page 6 when dealing with TYPE_RBC - these guys have writeback cache flag there and are not guaranteed to have page 8 at all. e) sd_read_cache_type() got an extra sanity check - it checks that it got the page it asked for before using its contents. And screams if mismatch had happened. Rationale: there are broken devices out there that are "helpful" enough to go for "I don't have a page you've asked for, here, have another one". For example, PL3507 had been caught doing just that... f) sbp2 sets sdev->use_10_for_rw and sdev->use_10_for_ms instead of bothering to remap READ6/WRITE6/MOD_SENSE, so most of the conversions in there are gone now. Incidentally, I wonder if USB storage devices that have no mode page 8 are simply RBC ones. I haven't touched that, but it might be interesting to check... Signed-off-by: Al Viro Signed-off-by: James Bottomley --- drivers/ieee1394/sbp2.c | 130 ++--------------------------------------------- drivers/ieee1394/sbp2.h | 4 -- drivers/scsi/scsi_scan.c | 1 + drivers/scsi/sd.c | 36 +++++++++---- include/scsi/scsi.h | 1 + 5 files changed, 32 insertions(+), 140 deletions(-) (limited to 'include') diff --git a/drivers/ieee1394/sbp2.c b/drivers/ieee1394/sbp2.c index 00c7b958361a..094e646ed4de 100644 --- a/drivers/ieee1394/sbp2.c +++ b/drivers/ieee1394/sbp2.c @@ -1070,7 +1070,7 @@ static int sbp2_handle_physdma_read(struct hpsb_host *host, int nodeid, quadlet_ static __inline__ int sbp2_command_conversion_device_type(u8 device_type) { return (((device_type == TYPE_DISK) || - (device_type == TYPE_SDAD) || + (device_type == TYPE_RBC) || (device_type == TYPE_ROM)) ? 1:0); } @@ -2111,102 +2111,6 @@ static int sbp2_send_command(struct scsi_id_instance_data *scsi_id, */ static void sbp2_check_sbp2_command(struct scsi_id_instance_data *scsi_id, unchar *cmd) { - unchar new_cmd[16]; - u8 device_type = SBP2_DEVICE_TYPE (scsi_id->sbp2_device_type_and_lun); - - SBP2_DEBUG("sbp2_check_sbp2_command"); - - switch (*cmd) { - - case READ_6: - - if (sbp2_command_conversion_device_type(device_type)) { - - SBP2_DEBUG("Convert READ_6 to READ_10"); - - /* - * Need to turn read_6 into read_10 - */ - new_cmd[0] = 0x28; - new_cmd[1] = (cmd[1] & 0xe0); - new_cmd[2] = 0x0; - new_cmd[3] = (cmd[1] & 0x1f); - new_cmd[4] = cmd[2]; - new_cmd[5] = cmd[3]; - new_cmd[6] = 0x0; - new_cmd[7] = 0x0; - new_cmd[8] = cmd[4]; - new_cmd[9] = cmd[5]; - - memcpy(cmd, new_cmd, 10); - - } - - break; - - case WRITE_6: - - if (sbp2_command_conversion_device_type(device_type)) { - - SBP2_DEBUG("Convert WRITE_6 to WRITE_10"); - - /* - * Need to turn write_6 into write_10 - */ - new_cmd[0] = 0x2a; - new_cmd[1] = (cmd[1] & 0xe0); - new_cmd[2] = 0x0; - new_cmd[3] = (cmd[1] & 0x1f); - new_cmd[4] = cmd[2]; - new_cmd[5] = cmd[3]; - new_cmd[6] = 0x0; - new_cmd[7] = 0x0; - new_cmd[8] = cmd[4]; - new_cmd[9] = cmd[5]; - - memcpy(cmd, new_cmd, 10); - - } - - break; - - case MODE_SENSE: - - if (sbp2_command_conversion_device_type(device_type)) { - - SBP2_DEBUG("Convert MODE_SENSE_6 to MODE_SENSE_10"); - - /* - * Need to turn mode_sense_6 into mode_sense_10 - */ - new_cmd[0] = 0x5a; - new_cmd[1] = cmd[1]; - new_cmd[2] = cmd[2]; - new_cmd[3] = 0x0; - new_cmd[4] = 0x0; - new_cmd[5] = 0x0; - new_cmd[6] = 0x0; - new_cmd[7] = 0x0; - new_cmd[8] = cmd[4]; - new_cmd[9] = cmd[5]; - - memcpy(cmd, new_cmd, 10); - - } - - break; - - case MODE_SELECT: - - /* - * TODO. Probably need to change mode select to 10 byte version - */ - - default: - break; - } - - return; } /* @@ -2271,14 +2175,6 @@ static void sbp2_check_sbp2_response(struct scsi_id_instance_data *scsi_id, scsi_buf[4] = 36 - 5; } - /* - * Check for Simple Direct Access Device and change it to TYPE_DISK - */ - if ((scsi_buf[0] & 0x1f) == TYPE_SDAD) { - SBP2_DEBUG("Changing TYPE_SDAD to TYPE_DISK"); - scsi_buf[0] &= 0xe0; - } - /* * Fix ansi revision and response data format */ @@ -2287,27 +2183,6 @@ static void sbp2_check_sbp2_response(struct scsi_id_instance_data *scsi_id, break; - case MODE_SENSE: - - if (sbp2_command_conversion_device_type(device_type)) { - - SBP2_DEBUG("Modify mode sense response (10 byte version)"); - - scsi_buf[0] = scsi_buf[1]; /* Mode data length */ - scsi_buf[1] = scsi_buf[2]; /* Medium type */ - scsi_buf[2] = scsi_buf[3]; /* Device specific parameter */ - scsi_buf[3] = scsi_buf[7]; /* Block descriptor length */ - memcpy(scsi_buf + 4, scsi_buf + 8, scsi_buf[0]); - } - - break; - - case MODE_SELECT: - - /* - * TODO. Probably need to change mode select to 10 byte version - */ - default: break; } @@ -2690,7 +2565,8 @@ static void sbp2scsi_complete_command(struct scsi_id_instance_data *scsi_id, static int sbp2scsi_slave_configure (struct scsi_device *sdev) { blk_queue_dma_alignment(sdev->request_queue, (512 - 1)); - + sdev->use_10_for_rw = 1; + sdev->use_10_for_ms = 1; return 0; } diff --git a/drivers/ieee1394/sbp2.h b/drivers/ieee1394/sbp2.h index a84b039a05b9..cd425be74841 100644 --- a/drivers/ieee1394/sbp2.h +++ b/drivers/ieee1394/sbp2.h @@ -266,10 +266,6 @@ struct sbp2_status_block { #define SBP2_MAX_UDS_PER_NODE 16 /* Maximum scsi devices per node */ #define SBP2_MAX_SECTORS 255 /* Max sectors supported */ -#ifndef TYPE_SDAD -#define TYPE_SDAD 0x0e /* simplified direct access device */ -#endif - /* * SCSI direction table... * (now used as a back-up in case the direction passed down from above is "unknown") diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index cca772624ae7..588611568d14 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -625,6 +625,7 @@ static int scsi_add_lun(struct scsi_device *sdev, char *inq_result, int *bflags) case TYPE_MEDIUM_CHANGER: case TYPE_ENCLOSURE: case TYPE_COMM: + case TYPE_RBC: sdev->writeable = 1; break; case TYPE_WORM: diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 19afb25e44d3..bb8235598787 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -1368,17 +1368,26 @@ sd_read_write_protect_flag(struct scsi_disk *sdkp, char *diskname, */ static void sd_read_cache_type(struct scsi_disk *sdkp, char *diskname, - struct scsi_request *SRpnt, unsigned char *buffer) { + struct scsi_request *SRpnt, unsigned char *buffer) +{ int len = 0, res; - const int dbd = 0; /* DBD */ - const int modepage = 0x08; /* current values, cache page */ + int dbd; + int modepage; struct scsi_mode_data data; struct scsi_sense_hdr sshdr; if (sdkp->device->skip_ms_page_8) goto defaults; + if (sdkp->device->type == TYPE_RBC) { + modepage = 6; + dbd = 8; + } else { + modepage = 8; + dbd = 0; + } + /* cautiously ask */ res = sd_do_mode_sense(SRpnt, dbd, modepage, buffer, 4, &data); @@ -1409,11 +1418,20 @@ sd_read_cache_type(struct scsi_disk *sdkp, char *diskname, "write back, no read (daft)" }; int ct = 0; - int offset = data.header_length + - data.block_descriptor_length + 2; + int offset = data.header_length + data.block_descriptor_length; - sdkp->WCE = ((buffer[offset] & 0x04) != 0); - sdkp->RCD = ((buffer[offset] & 0x01) != 0); + if ((buffer[offset] & 0x3f) != modepage) { + printk(KERN_ERR "%s: got wrong page\n", diskname); + goto defaults; + } + + if (modepage == 8) { + sdkp->WCE = ((buffer[offset + 2] & 0x04) != 0); + sdkp->RCD = ((buffer[offset + 2] & 0x01) != 0); + } else { + sdkp->WCE = ((buffer[offset + 2] & 0x01) == 0); + sdkp->RCD = 0; + } ct = sdkp->RCD + 2*sdkp->WCE; @@ -1533,7 +1551,7 @@ static int sd_probe(struct device *dev) int error; error = -ENODEV; - if ((sdp->type != TYPE_DISK) && (sdp->type != TYPE_MOD)) + if (sdp->type != TYPE_DISK && sdp->type != TYPE_MOD && sdp->type != TYPE_RBC) goto out; SCSI_LOG_HLQUEUE(3, printk("sd_attach: scsi device: <%d,%d,%d,%d>\n", @@ -1570,7 +1588,7 @@ static int sd_probe(struct device *dev) sdkp->openers = 0; if (!sdp->timeout) { - if (sdp->type == TYPE_DISK) + if (sdp->type != TYPE_MOD) sdp->timeout = SD_TIMEOUT; else sdp->timeout = SD_MOD_TIMEOUT; diff --git a/include/scsi/scsi.h b/include/scsi/scsi.h index ca1e3b4a3183..1fb233741513 100644 --- a/include/scsi/scsi.h +++ b/include/scsi/scsi.h @@ -213,6 +213,7 @@ static inline int scsi_status_is_good(int status) #define TYPE_COMM 0x09 /* Communications device */ #define TYPE_ENCLOSURE 0x0d /* Enclosure Services Device */ #define TYPE_RAID 0x0c +#define TYPE_RBC 0x0e #define TYPE_NO_LUN 0x7f /* -- cgit v1.2.3 From a283bd37d00e92e8874ca6325ae071691d4db388 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Tue, 24 May 2005 12:06:38 -0500 Subject: [SCSI] Add target alloc/destroy callbacks to the host template This gives the HBA driver notice when a target is created and destroyed to allow it to manage its own target based allocations accordingly. This is a much reduced verson of the original patch sent in by James.Smart@Emulex.com Signed-off-by: James Bottomley --- drivers/scsi/scsi_scan.c | 24 +++++++++++++++++++++--- include/scsi/scsi_device.h | 4 +++- include/scsi/scsi_host.h | 25 +++++++++++++++++++++++++ 3 files changed, 49 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index 588611568d14..4d273ceb1d09 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -293,6 +293,10 @@ static void scsi_target_dev_release(struct device *dev) { struct device *parent = dev->parent; struct scsi_target *starget = to_scsi_target(dev); + struct Scsi_Host *shost = dev_to_shost(parent); + + if (shost->hostt->target_destroy) + shost->hostt->target_destroy(starget); kfree(starget); put_device(parent); } @@ -360,9 +364,23 @@ static struct scsi_target *scsi_alloc_target(struct device *parent, list_add_tail(&starget->siblings, &shost->__targets); spin_unlock_irqrestore(shost->host_lock, flags); /* allocate and add */ - transport_setup_device(&starget->dev); - device_add(&starget->dev); - transport_add_device(&starget->dev); + transport_setup_device(dev); + device_add(dev); + transport_add_device(dev); + if (shost->hostt->target_alloc) { + int error = shost->hostt->target_alloc(starget); + + if(error) { + dev_printk(KERN_ERR, dev, "target allocation failed, error %d\n", error); + /* don't want scsi_target_reap to do the final + * put because it will be under the host lock */ + get_device(dev); + scsi_target_reap(starget); + put_device(dev); + return NULL; + } + } + return starget; found: diff --git a/include/scsi/scsi_device.h b/include/scsi/scsi_device.h index c018020d9160..63c91dd85ca1 100644 --- a/include/scsi/scsi_device.h +++ b/include/scsi/scsi_device.h @@ -154,7 +154,9 @@ struct scsi_target { unsigned int id; /* target id ... replace * scsi_device.id eventually */ unsigned long create:1; /* signal that it needs to be added */ - unsigned long starget_data[0]; + void *hostdata; /* available to low-level driver */ + unsigned long starget_data[0]; /* for the transport */ + /* starget_data must be the last element!!!! */ } __attribute__((aligned(sizeof(unsigned long)))); #define to_scsi_target(d) container_of(d, struct scsi_target, dev) diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h index 1cee1e100943..db9914adeac9 100644 --- a/include/scsi/scsi_host.h +++ b/include/scsi/scsi_host.h @@ -10,6 +10,7 @@ struct block_device; struct module; struct scsi_cmnd; struct scsi_device; +struct scsi_target; struct Scsi_Host; struct scsi_host_cmd_pool; struct scsi_transport_template; @@ -227,6 +228,30 @@ struct scsi_host_template { */ void (* slave_destroy)(struct scsi_device *); + /* + * Before the mid layer attempts to scan for a new device attached + * to a target where no target currently exists, it will call this + * entry in your driver. Should your driver need to allocate any + * structs or perform any other init items in order to send commands + * to a currently unused target, then this is where you can perform + * those allocations. + * + * Return values: 0 on success, non-0 on failure + * + * Status: OPTIONAL + */ + int (* target_alloc)(struct scsi_target *); + + /* + * Immediately prior to deallocating the target structure, and + * after all activity to attached scsi devices has ceased, the + * midlayer calls this point so that the driver may deallocate + * and terminate any references to the target. + * + * Status: OPTIONAL + */ + void (* target_destroy)(struct scsi_target *); + /* * fill in this function to allow the queue depth of this host * to be changeable (on a per device basis). returns either -- cgit v1.2.3 From c3e9dda4f5702ee5b346f4770de53f79e8ad1d8d Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Tue, 24 May 2005 16:57:31 -0500 Subject: [SCSI] allow the HBA to reserve target and device private areas This patch basically allows any HBA attached to the SPI transport class to declare an extra area which the mid-layer will allocate as part of its device and target allocations. Signed-off-by: James Bottomley --- include/scsi/scsi_transport.h | 38 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 38 insertions(+) (limited to 'include') diff --git a/include/scsi/scsi_transport.h b/include/scsi/scsi_transport.h index 2dcee7a84752..a4f1837a33b1 100644 --- a/include/scsi/scsi_transport.h +++ b/include/scsi/scsi_transport.h @@ -21,6 +21,7 @@ #define SCSI_TRANSPORT_H #include +#include struct scsi_transport_template { /* the attribute containers */ @@ -32,8 +33,11 @@ struct scsi_transport_template { * space of this size will be left at the end of the * scsi_* structure */ int device_size; + int device_private_offset; int target_size; + int target_private_offset; int host_size; + /* no private offset for the host; there's an alternative mechanism */ /* * True if the transport wants to use a host-based work-queue @@ -45,4 +49,38 @@ struct scsi_transport_template { dev_to_shost((tc)->dev) +/* Private area maintenance. The driver requested allocations come + * directly after the transport class allocations (if any). The idea + * is that you *must* call these only once. The code assumes that the + * initial values are the ones the transport specific code requires */ +static inline void +scsi_transport_reserve_target(struct scsi_transport_template * t, int space) +{ + BUG_ON(t->target_private_offset != 0); + t->target_private_offset = ALIGN(t->target_size, sizeof(void *)); + t->target_size = t->target_private_offset + space; +} +static inline void +scsi_transport_reserve_device(struct scsi_transport_template * t, int space) +{ + BUG_ON(t->device_private_offset != 0); + t->device_private_offset = ALIGN(t->device_size, sizeof(void *)); + t->device_size = t->device_private_offset + space; +} +static inline void * +scsi_transport_target_data(struct scsi_target *starget) +{ + struct Scsi_Host *shost = dev_to_shost(&starget->dev); + return (u8 *)starget->starget_data + + shost->transportt->target_private_offset; + +} +static inline void * +scsi_transport_device_data(struct scsi_device *sdev) +{ + struct Scsi_Host *shost = sdev->host; + return (u8 *)sdev->sdev_data + + shost->transportt->device_private_offset; +} + #endif /* SCSI_TRANSPORT_H */ -- cgit v1.2.3 From aa8f0dc6c3dbf1cf3ff58f3e945c981be134814d Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Thu, 26 May 2005 21:54:27 -0400 Subject: libata: Fix use-after-iounmap Jens Axboe pointed out that the iounmap() call in libata was occurring too early, and some drivers (ahci, probably others) were using ioremap'd memory after it had been unmapped. The patch should address that problem by way of improving the libata driver API: * move ->host_stop() call after all ->port_stop() calls have occurred. * create default helper function ata_host_stop(), and move iounmap() call there. * add ->host_stop_prewalk() hook, use it in sata_qstor.c (hi Mark). sata_qstor appears to require the host-stop-before-port-stop ordering that existed prior to applying the attached patch. --- drivers/scsi/ahci.c | 2 ++ drivers/scsi/ata_piix.c | 2 ++ drivers/scsi/libata-core.c | 15 +++++++++++---- drivers/scsi/sata_nv.c | 2 ++ drivers/scsi/sata_promise.c | 1 + drivers/scsi/sata_qstor.c | 2 ++ drivers/scsi/sata_sil.c | 1 + drivers/scsi/sata_sis.c | 1 + drivers/scsi/sata_svw.c | 1 + drivers/scsi/sata_sx4.c | 2 ++ drivers/scsi/sata_uli.c | 1 + drivers/scsi/sata_via.c | 1 + drivers/scsi/sata_vsc.c | 1 + include/linux/libata.h | 1 + 14 files changed, 29 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/drivers/scsi/ahci.c b/drivers/scsi/ahci.c index da5bd33d982d..8b468a628700 100644 --- a/drivers/scsi/ahci.c +++ b/drivers/scsi/ahci.c @@ -289,6 +289,8 @@ static void ahci_host_stop(struct ata_host_set *host_set) { struct ahci_host_priv *hpriv = host_set->private_data; kfree(hpriv); + + ata_host_stop(host_set); } static int ahci_port_start(struct ata_port *ap) diff --git a/drivers/scsi/ata_piix.c b/drivers/scsi/ata_piix.c index 3867f91ef8c7..54c52349adc5 100644 --- a/drivers/scsi/ata_piix.c +++ b/drivers/scsi/ata_piix.c @@ -153,6 +153,7 @@ static struct ata_port_operations piix_pata_ops = { .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_operations piix_sata_ops = { @@ -180,6 +181,7 @@ static struct ata_port_operations piix_sata_ops = { .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_info piix_port_info[] = { diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c index ee9b96da841e..2b41cd3a8ec6 100644 --- a/drivers/scsi/libata-core.c +++ b/drivers/scsi/libata-core.c @@ -3321,6 +3321,13 @@ void ata_port_stop (struct ata_port *ap) dma_free_coherent(dev, ATA_PRD_TBL_SZ, ap->prd, ap->prd_dma); } +void ata_host_stop (struct ata_host_set *host_set) +{ + if (host_set->mmio_base) + iounmap(host_set->mmio_base); +} + + /** * ata_host_remove - Unregister SCSI host structure with upper layers * @ap: Port to unregister @@ -3877,10 +3884,6 @@ void ata_pci_remove_one (struct pci_dev *pdev) } free_irq(host_set->irq, host_set); - if (host_set->ops->host_stop) - host_set->ops->host_stop(host_set); - if (host_set->mmio_base) - iounmap(host_set->mmio_base); for (i = 0; i < host_set->n_ports; i++) { ap = host_set->ports[i]; @@ -3899,6 +3902,9 @@ void ata_pci_remove_one (struct pci_dev *pdev) scsi_host_put(ap->host); } + if (host_set->ops->host_stop) + host_set->ops->host_stop(host_set); + kfree(host_set); pci_release_regions(pdev); @@ -3996,6 +4002,7 @@ EXPORT_SYMBOL_GPL(ata_chk_err); EXPORT_SYMBOL_GPL(ata_exec_command); EXPORT_SYMBOL_GPL(ata_port_start); EXPORT_SYMBOL_GPL(ata_port_stop); +EXPORT_SYMBOL_GPL(ata_host_stop); EXPORT_SYMBOL_GPL(ata_interrupt); EXPORT_SYMBOL_GPL(ata_qc_prep); EXPORT_SYMBOL_GPL(ata_bmdma_setup); diff --git a/drivers/scsi/sata_nv.c b/drivers/scsi/sata_nv.c index 69009f853a49..b0403ccd8a25 100644 --- a/drivers/scsi/sata_nv.c +++ b/drivers/scsi/sata_nv.c @@ -329,6 +329,8 @@ static void nv_host_stop (struct ata_host_set *host_set) host->host_desc->disable_hotplug(host_set); kfree(host); + + ata_host_stop(host_set); } static int nv_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) diff --git a/drivers/scsi/sata_promise.c b/drivers/scsi/sata_promise.c index c4e9e0298122..b18c90582e67 100644 --- a/drivers/scsi/sata_promise.c +++ b/drivers/scsi/sata_promise.c @@ -122,6 +122,7 @@ static struct ata_port_operations pdc_ata_ops = { .scr_write = pdc_sata_scr_write, .port_start = pdc_port_start, .port_stop = pdc_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_info pdc_port_info[] = { diff --git a/drivers/scsi/sata_qstor.c b/drivers/scsi/sata_qstor.c index dfd362104717..1383e8a28d72 100644 --- a/drivers/scsi/sata_qstor.c +++ b/drivers/scsi/sata_qstor.c @@ -536,6 +536,8 @@ static void qs_host_stop(struct ata_host_set *host_set) writeb(0, mmio_base + QS_HCT_CTRL); /* disable host interrupts */ writeb(QS_CNFG3_GSRST, mmio_base + QS_HCF_CNFG3); /* global reset */ + + ata_host_stop(host_set); } static void qs_host_init(unsigned int chip_id, struct ata_probe_ent *pe) diff --git a/drivers/scsi/sata_sil.c b/drivers/scsi/sata_sil.c index 2b2ff48be396..238580d244e6 100644 --- a/drivers/scsi/sata_sil.c +++ b/drivers/scsi/sata_sil.c @@ -161,6 +161,7 @@ static struct ata_port_operations sil_ops = { .scr_write = sil_scr_write, .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_info sil_port_info[] = { diff --git a/drivers/scsi/sata_sis.c b/drivers/scsi/sata_sis.c index 5105ddd08447..e418b89c6b9d 100644 --- a/drivers/scsi/sata_sis.c +++ b/drivers/scsi/sata_sis.c @@ -114,6 +114,7 @@ static struct ata_port_operations sis_ops = { .scr_write = sis_scr_write, .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_info sis_port_info = { diff --git a/drivers/scsi/sata_svw.c b/drivers/scsi/sata_svw.c index 05075bd3a893..edef1fa969fc 100644 --- a/drivers/scsi/sata_svw.c +++ b/drivers/scsi/sata_svw.c @@ -313,6 +313,7 @@ static struct ata_port_operations k2_sata_ops = { .scr_write = k2_sata_scr_write, .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static void k2_sata_setup_port(struct ata_ioports *port, unsigned long base) diff --git a/drivers/scsi/sata_sx4.c b/drivers/scsi/sata_sx4.c index 70118650c461..140cea05de3f 100644 --- a/drivers/scsi/sata_sx4.c +++ b/drivers/scsi/sata_sx4.c @@ -245,6 +245,8 @@ static void pdc20621_host_stop(struct ata_host_set *host_set) iounmap(dimm_mmio); kfree(hpriv); + + ata_host_stop(host_set); } static int pdc_port_start(struct ata_port *ap) diff --git a/drivers/scsi/sata_uli.c b/drivers/scsi/sata_uli.c index 0bff4f475f26..a71fb54eebd3 100644 --- a/drivers/scsi/sata_uli.c +++ b/drivers/scsi/sata_uli.c @@ -113,6 +113,7 @@ static struct ata_port_operations uli_ops = { .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_info uli_port_info = { diff --git a/drivers/scsi/sata_via.c b/drivers/scsi/sata_via.c index 3a7830667277..f43183c19a12 100644 --- a/drivers/scsi/sata_via.c +++ b/drivers/scsi/sata_via.c @@ -134,6 +134,7 @@ static struct ata_port_operations svia_sata_ops = { .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static struct ata_port_info svia_port_info = { diff --git a/drivers/scsi/sata_vsc.c b/drivers/scsi/sata_vsc.c index 2c28f0ad73c2..f67c34330ae9 100644 --- a/drivers/scsi/sata_vsc.c +++ b/drivers/scsi/sata_vsc.c @@ -230,6 +230,7 @@ static struct ata_port_operations vsc_sata_ops = { .scr_write = vsc_sata_scr_write, .port_start = ata_port_start, .port_stop = ata_port_stop, + .host_stop = ata_host_stop, }; static void __devinit vsc_sata_setup_port(struct ata_ioports *port, unsigned long base) diff --git a/include/linux/libata.h b/include/linux/libata.h index 1f7e2039a04e..e74f301e9bae 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -410,6 +410,7 @@ extern u8 ata_chk_err(struct ata_port *ap); extern void ata_exec_command(struct ata_port *ap, struct ata_taskfile *tf); extern int ata_port_start (struct ata_port *ap); extern void ata_port_stop (struct ata_port *ap); +extern void ata_host_stop (struct ata_host_set *host_set); extern irqreturn_t ata_interrupt (int irq, void *dev_instance, struct pt_regs *regs); extern void ata_qc_prep(struct ata_queued_cmd *qc); extern int ata_qc_issue_prot(struct ata_queued_cmd *qc); -- cgit v1.2.3 From 8f37d47c9bf74cb48692691086b482e315d07f40 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Fri, 27 May 2005 12:17:28 +0100 Subject: AUDIT: Record working directory when syscall arguments are pathnames Signed-off-by: David Woodhouse --- include/linux/audit.h | 3 ++- kernel/auditsc.c | 23 +++++++++++++++++++++++ 2 files changed, 25 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/audit.h b/include/linux/audit.h index 3278ddf41ce6..bf2ad3ba72eb 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -61,11 +61,12 @@ #define AUDIT_SYSCALL 1300 /* Syscall event */ #define AUDIT_FS_WATCH 1301 /* Filesystem watch event */ -#define AUDIT_PATH 1302 /* Filname path information */ +#define AUDIT_PATH 1302 /* Filename path information */ #define AUDIT_IPC 1303 /* IPC record */ #define AUDIT_SOCKETCALL 1304 /* sys_socketcall arguments */ #define AUDIT_CONFIG_CHANGE 1305 /* Audit system configuration change */ #define AUDIT_SOCKADDR 1306 /* sockaddr copied as syscall arg */ +#define AUDIT_CWD 1307 /* Current working directory */ #define AUDIT_AVC 1400 /* SE Linux avc denial or grant */ #define AUDIT_SELINUX_ERR 1401 /* Internal SE Linux Errors */ diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 7556c479d5af..e75f84e1a1a0 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -145,6 +145,8 @@ struct audit_context { int auditable; /* 1 if record should be written */ int name_count; struct audit_names names[AUDIT_NAMES]; + struct dentry * pwd; + struct vfsmount * pwdmnt; struct audit_context *previous; /* For nested syscalls */ struct audit_aux_data *aux; @@ -552,6 +554,12 @@ static inline void audit_free_names(struct audit_context *context) if (context->names[i].name) __putname(context->names[i].name); context->name_count = 0; + if (context->pwd) + dput(context->pwd); + if (context->pwdmnt) + mntput(context->pwdmnt); + context->pwd = NULL; + context->pwdmnt = NULL; } static inline void audit_free_aux(struct audit_context *context) @@ -745,10 +753,18 @@ static void audit_log_exit(struct audit_context *context) audit_log_end(ab); } + if (context->pwd && context->pwdmnt) { + ab = audit_log_start(context, AUDIT_CWD); + if (ab) { + audit_log_d_path(ab, "cwd=", context->pwd, context->pwdmnt); + audit_log_end(ab); + } + } for (i = 0; i < context->name_count; i++) { ab = audit_log_start(context, AUDIT_PATH); if (!ab) continue; /* audit_panic has been called */ + audit_log_format(ab, "item=%d", i); if (context->names[i].name) { audit_log_format(ab, " name="); @@ -929,6 +945,13 @@ void audit_getname(const char *name) context->names[context->name_count].name = name; context->names[context->name_count].ino = (unsigned long)-1; ++context->name_count; + if (!context->pwd) { + read_lock(¤t->fs->lock); + context->pwd = dget(current->fs->pwd); + context->pwdmnt = mntget(current->fs->pwdmnt); + read_unlock(¤t->fs->lock); + } + } /* Intercept a putname request. Called from -- cgit v1.2.3 From 668d1e6093110f7534e661e2ff43d54c74659b6d Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Sat, 28 May 2005 02:11:12 -0500 Subject: Input: This patch adds dummy gameport_register_port, gameport_unregister_port and gameport_set_phys functions to gameport.h for the case when a driver can't use gameport. This fixes the compilation of some OSS drivers with GAMEPORT=n without the need to #if inside every single driver. This patch also removes the non-working and now obsolete SOUND_GAMEPORT. This patch is also an alternative solution for ALSA drivers with similar problems (but #if's inside the drivers might have the advantage of saving some more bytes of gameport is not available). The only user-visible change is that for GAMEPORT=m the affected OSS drivers are now allowed to be built statically (but they won't have gameport support). Signed-off-by: Adrian Bunk Signed-off-by: Vojtech Pavlik Signed-off-by: Dmitry Torokhov --- drivers/input/gameport/Kconfig | 20 -------------------- include/linux/gameport.h | 28 +++++++++++++++++++++++++--- sound/oss/Kconfig | 12 ++++++------ 3 files changed, 31 insertions(+), 29 deletions(-) (limited to 'include') diff --git a/drivers/input/gameport/Kconfig b/drivers/input/gameport/Kconfig index 6282f460aba0..1d93f5092904 100644 --- a/drivers/input/gameport/Kconfig +++ b/drivers/input/gameport/Kconfig @@ -68,23 +68,3 @@ config GAMEPORT_CS461X depends on PCI endif - -# Yes, SOUND_GAMEPORT looks a bit odd. Yes, it ends up being turned on -# in every .config. Please don't touch it. It is here to handle an -# unusual dependency between GAMEPORT and sound drivers. -# -# Some sound drivers call gameport functions. If GAMEPORT is -# not selected, empty stubs are provided for the functions and all is -# well. -# If GAMEPORT is built in, everything is fine. -# If GAMEPORT is a module, however, it would need to be loaded for the -# sound driver to be able to link properly. Therefore, the sound -# driver must be a module as well in that case. Since there's no way -# to express that directly in Kconfig, we use SOUND_GAMEPORT to -# express it. SOUND_GAMEPORT boils down to "if GAMEPORT is 'm', -# anything that depends on SOUND_GAMEPORT must be 'm' as well. if -# GAMEPORT is 'y' or 'n', it can be anything". -config SOUND_GAMEPORT - tristate - default m if GAMEPORT=m - default y diff --git a/include/linux/gameport.h b/include/linux/gameport.h index b1272f822cfa..cd623eccdbea 100644 --- a/include/linux/gameport.h +++ b/include/linux/gameport.h @@ -67,6 +67,8 @@ int gameport_open(struct gameport *gameport, struct gameport_driver *drv, int mo void gameport_close(struct gameport *gameport); void gameport_rescan(struct gameport *gameport); +#if defined(CONFIG_GAMEPORT) || (defined(MODULE) && defined(CONFIG_GAMEPORT_MODULE)) + void __gameport_register_port(struct gameport *gameport, struct module *owner); static inline void gameport_register_port(struct gameport *gameport) { @@ -75,6 +77,29 @@ static inline void gameport_register_port(struct gameport *gameport) void gameport_unregister_port(struct gameport *gameport); +void gameport_set_phys(struct gameport *gameport, const char *fmt, ...) + __attribute__ ((format (printf, 2, 3))); + +#else + +static inline void gameport_register_port(struct gameport *gameport) +{ + return; +} + +static inline void gameport_unregister_port(struct gameport *gameport) +{ + return; +} + +static inline void gameport_set_phys(struct gameport *gameport, + const char *fmt, ...) +{ + return; +} + +#endif + static inline struct gameport *gameport_allocate_port(void) { struct gameport *gameport = kcalloc(1, sizeof(struct gameport), GFP_KERNEL); @@ -92,9 +117,6 @@ static inline void gameport_set_name(struct gameport *gameport, const char *name strlcpy(gameport->name, name, sizeof(gameport->name)); } -void gameport_set_phys(struct gameport *gameport, const char *fmt, ...) - __attribute__ ((format (printf, 2, 3))); - /* * Use the following fucntions to manipulate gameport's per-port * driver-specific data. diff --git a/sound/oss/Kconfig b/sound/oss/Kconfig index a9602f89d6b1..e537bd66a707 100644 --- a/sound/oss/Kconfig +++ b/sound/oss/Kconfig @@ -112,7 +112,7 @@ config SOUND_BCM_CS4297A config SOUND_ES1370 tristate "Ensoniq AudioPCI (ES1370)" - depends on SOUND_PRIME!=n && SOUND && PCI && SOUND_GAMEPORT + depends on SOUND_PRIME!=n && SOUND && PCI help Say Y or M if you have a PCI sound card utilizing the Ensoniq ES1370 chipset, such as Ensoniq's AudioPCI (non-97). To find @@ -125,7 +125,7 @@ config SOUND_ES1370 config SOUND_ES1371 tristate "Creative Ensoniq AudioPCI 97 (ES1371)" - depends on SOUND_PRIME!=n && SOUND && PCI && SOUND_GAMEPORT + depends on SOUND_PRIME!=n && SOUND && PCI help Say Y or M if you have a PCI sound card utilizing the Ensoniq ES1371 chipset, such as Ensoniq's AudioPCI97. To find out if @@ -138,7 +138,7 @@ config SOUND_ES1371 config SOUND_ESSSOLO1 tristate "ESS Technology Solo1" - depends on SOUND_PRIME!=n && SOUND && SOUND_GAMEPORT && PCI + depends on SOUND_PRIME!=n && SOUND && PCI help Say Y or M if you have a PCI sound card utilizing the ESS Technology Solo1 chip. To find out if your sound card uses a @@ -179,7 +179,7 @@ config SOUND_HARMONY config SOUND_SONICVIBES tristate "S3 SonicVibes" - depends on SOUND_PRIME!=n && SOUND && SOUND_GAMEPORT + depends on SOUND_PRIME!=n && SOUND help Say Y or M if you have a PCI sound card utilizing the S3 SonicVibes chipset. To find out if your sound card uses a @@ -226,7 +226,7 @@ config SOUND_AU1550_AC97 config SOUND_TRIDENT tristate "Trident 4DWave DX/NX, SiS 7018 or ALi 5451 PCI Audio Core" - depends on SOUND_PRIME!=n && SOUND && SOUND_GAMEPORT + depends on SOUND_PRIME!=n && SOUND ---help--- Say Y or M if you have a PCI sound card utilizing the Trident 4DWave-DX/NX chipset or your mother board chipset has SiS 7018 @@ -739,7 +739,7 @@ config SOUND_NM256 config SOUND_MAD16 tristate "OPTi MAD16 and/or Mozart based cards" - depends on SOUND_OSS && SOUND_GAMEPORT + depends on SOUND_OSS ---help--- Answer Y if your card has a Mozart (OAK OTI-601) or MAD16 (OPTi 82C928 or 82C929 or 82C931) audio interface chip. These chips are -- cgit v1.2.3 From b9e0d0696c9e86d4e40b23039feefb8c377cb524 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Sat, 28 May 2005 15:51:53 -0700 Subject: [PATCH] uml: fix a couple of warnings Eliminate an unused variable warning in ptrace.c and a size mismatch warning by adding a cast to __pa. Signed-off-by: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/kernel/ptrace.c | 19 +++---------------- include/asm-um/page.h | 8 +++++++- 2 files changed, 10 insertions(+), 17 deletions(-) (limited to 'include') diff --git a/arch/um/kernel/ptrace.c b/arch/um/kernel/ptrace.c index 2925e15324de..71af4d503899 100644 --- a/arch/um/kernel/ptrace.c +++ b/arch/um/kernel/ptrace.c @@ -322,11 +322,9 @@ void syscall_trace(union uml_pt_regs *regs, int entryexit) UPT_SYSCALL_ARG2(regs), UPT_SYSCALL_ARG3(regs), UPT_SYSCALL_ARG4(regs)); - else { - int res = UPT_SYSCALL_RET(regs); - audit_syscall_exit(current, AUDITSC_RESULT(res), - res); - } + else audit_syscall_exit(current, + AUDITSC_RESULT(UPT_SYSCALL_RET(regs)), + UPT_SYSCALL_RET(regs)); } /* Fake a debug trap */ @@ -356,14 +354,3 @@ void syscall_trace(union uml_pt_regs *regs, int entryexit) current->exit_code = 0; } } - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/include/asm-um/page.h b/include/asm-um/page.h index 504ea8e486b0..5afee8a8cdf3 100644 --- a/include/asm-um/page.h +++ b/include/asm-um/page.h @@ -98,7 +98,13 @@ extern unsigned long uml_physmem; extern unsigned long to_phys(void *virt); extern void *to_virt(unsigned long phys); -#define __pa(virt) to_phys((void *) virt) + +/* Cast to unsigned long before casting to void * to avoid a warning from + * mmap_kmem about cutting a long long down to a void *. Not sure that + * casting is the right thing, but 32-bit UML can't have 64-bit virtual + * addresses + */ +#define __pa(virt) to_phys((void *) (unsigned long) virt) #define __va(phys) to_virt((unsigned long) phys) #define page_to_pfn(page) ((page) - mem_map) -- cgit v1.2.3 From b3461034d7d46455060c8476910be22b6b0fc313 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sat, 28 May 2005 15:52:00 -0700 Subject: [PATCH] uml: stack dump fix Copy (and adapt) to UML the stack code dumper used in i386 when CONFIG_FRAME_POINTER is enabled. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/include/sysrq.h | 3 +- arch/um/kernel/sysrq.c | 21 ++++++++---- arch/um/sys-i386/sysrq.c | 80 +++++++++++++++++++++++++++++++++++++++++++- arch/um/sys-ppc/sysrq.c | 14 +------- arch/um/sys-x86_64/sysrq.c | 11 +----- include/asm-um/thread_info.h | 7 ++-- 6 files changed, 100 insertions(+), 36 deletions(-) (limited to 'include') diff --git a/arch/um/include/sysrq.h b/arch/um/include/sysrq.h index 2ce9423460b3..c8d332b56b98 100644 --- a/arch/um/include/sysrq.h +++ b/arch/um/include/sysrq.h @@ -1,6 +1,7 @@ #ifndef __UM_SYSRQ_H #define __UM_SYSRQ_H -extern void show_trace(unsigned long *stack); +struct task_struct; +extern void show_trace(struct task_struct* task, unsigned long *stack); #endif diff --git a/arch/um/kernel/sysrq.c b/arch/um/kernel/sysrq.c index e630438f9e73..f80850091e79 100644 --- a/arch/um/kernel/sysrq.c +++ b/arch/um/kernel/sysrq.c @@ -3,6 +3,7 @@ * Licensed under the GPL */ +#include "linux/config.h" #include "linux/sched.h" #include "linux/kernel.h" #include "linux/module.h" @@ -12,14 +13,14 @@ #include "sysrq.h" #include "user_util.h" -void show_trace(unsigned long * stack) +/* Catch non-i386 SUBARCH's. */ +#if !defined(CONFIG_UML_X86) || defined(CONFIG_64BIT) +void show_trace(struct task_struct *task, unsigned long * stack) { - /* XXX: Copy the CONFIG_FRAME_POINTER stack-walking backtrace from - * arch/i386/kernel/traps.c, and then move this to sys-i386/sysrq.c.*/ unsigned long addr; if (!stack) { - stack = (unsigned long*) &stack; + stack = (unsigned long*) &stack; WARN_ON(1); } @@ -35,6 +36,7 @@ void show_trace(unsigned long * stack) } printk("\n"); } +#endif /* * stack dumps generator - this is used by arch-independent code. @@ -44,7 +46,7 @@ void dump_stack(void) { unsigned long stack; - show_trace(&stack); + show_trace(current, &stack); } EXPORT_SYMBOL(dump_stack); @@ -59,7 +61,11 @@ void show_stack(struct task_struct *task, unsigned long *esp) int i; if (esp == NULL) { - if (task != current) { + if (task != current && task != NULL) { + /* XXX: Isn't this bogus? I.e. isn't this the + * *userspace* stack of this task? If not so, use this + * even when task == current (as in i386). + */ esp = (unsigned long *) KSTK_ESP(task); /* Which one? No actual difference - just coding style.*/ //esp = (unsigned long *) PT_REGS_IP(&task->thread.regs); @@ -77,5 +83,6 @@ void show_stack(struct task_struct *task, unsigned long *esp) printk("%08lx ", *stack++); } - show_trace(esp); + printk("Call Trace: \n"); + show_trace(current, esp); } diff --git a/arch/um/sys-i386/sysrq.c b/arch/um/sys-i386/sysrq.c index 281fc7b8ca00..e3706d15c4f5 100644 --- a/arch/um/sys-i386/sysrq.c +++ b/arch/um/sys-i386/sysrq.c @@ -3,12 +3,15 @@ * Licensed under the GPL */ +#include "linux/config.h" #include "linux/kernel.h" #include "linux/smp.h" #include "linux/sched.h" +#include "linux/kallsyms.h" #include "asm/ptrace.h" #include "sysrq.h" +/* This is declared by */ void show_regs(struct pt_regs *regs) { printk("\n"); @@ -31,5 +34,80 @@ void show_regs(struct pt_regs *regs) 0xffff & PT_REGS_DS(regs), 0xffff & PT_REGS_ES(regs)); - show_trace((unsigned long *) ®s); + show_trace(NULL, (unsigned long *) ®s); } + +/* Copied from i386. */ +static inline int valid_stack_ptr(struct thread_info *tinfo, void *p) +{ + return p > (void *)tinfo && + p < (void *)tinfo + THREAD_SIZE - 3; +} + +/* Adapted from i386 (we also print the address we read from). */ +static inline unsigned long print_context_stack(struct thread_info *tinfo, + unsigned long *stack, unsigned long ebp) +{ + unsigned long addr; + +#ifdef CONFIG_FRAME_POINTER + while (valid_stack_ptr(tinfo, (void *)ebp)) { + addr = *(unsigned long *)(ebp + 4); + printk("%08lx: [<%08lx>]", ebp + 4, addr); + print_symbol(" %s", addr); + printk("\n"); + ebp = *(unsigned long *)ebp; + } +#else + while (valid_stack_ptr(tinfo, stack)) { + addr = *stack; + if (__kernel_text_address(addr)) { + printk("%08lx: [<%08lx>]", (unsigned long) stack, addr); + print_symbol(" %s", addr); + printk("\n"); + } + stack++; + } +#endif + return ebp; +} + +void show_trace(struct task_struct* task, unsigned long * stack) +{ + unsigned long ebp; + struct thread_info *context; + + /* Turn this into BUG_ON if possible. */ + if (!stack) { + stack = (unsigned long*) &stack; + printk("show_trace: got NULL stack, implicit assumption task == current"); + WARN_ON(1); + } + + if (!task) + task = current; + + if (task != current) { + //ebp = (unsigned long) KSTK_EBP(task); + /* Which one? No actual difference - just coding style.*/ + ebp = (unsigned long) PT_REGS_EBP(&task->thread.regs); + } else { + asm ("movl %%ebp, %0" : "=r" (ebp) : ); + } + + context = (struct thread_info *) + ((unsigned long)stack & (~(THREAD_SIZE - 1))); + print_context_stack(context, stack, ebp); + + /*while (((long) stack & (THREAD_SIZE-1)) != 0) { + addr = *stack; + if (__kernel_text_address(addr)) { + printk("%08lx: [<%08lx>]", (unsigned long) stack, addr); + print_symbol(" %s", addr); + printk("\n"); + } + stack++; + }*/ + printk("\n"); +} + diff --git a/arch/um/sys-ppc/sysrq.c b/arch/um/sys-ppc/sysrq.c index 82d6e9335bb6..2f816f1a0ff4 100644 --- a/arch/um/sys-ppc/sysrq.c +++ b/arch/um/sys-ppc/sysrq.c @@ -27,17 +27,5 @@ void show_regs(struct pt_regs_subarch *regs) 0xffff & regs->xds, 0xffff & regs->xes); #endif - show_trace(®s->gpr[1]); + show_trace(current, ®s->gpr[1]); } - - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/sys-x86_64/sysrq.c b/arch/um/sys-x86_64/sysrq.c index ddf74691a610..d0a25af19a5b 100644 --- a/arch/um/sys-x86_64/sysrq.c +++ b/arch/um/sys-x86_64/sysrq.c @@ -36,14 +36,5 @@ void __show_regs(struct pt_regs * regs) void show_regs(struct pt_regs *regs) { __show_regs(regs); - show_trace((unsigned long *) ®s); + show_trace(current, (unsigned long *) ®s); } - -/* Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/include/asm-um/thread_info.h b/include/asm-um/thread_info.h index a10ea155907e..e4f0198240c5 100644 --- a/include/asm-um/thread_info.h +++ b/include/asm-um/thread_info.h @@ -41,18 +41,17 @@ struct thread_info { #define init_thread_info (init_thread_union.thread_info) #define init_stack (init_thread_union.stack) +#define THREAD_SIZE ((1 << CONFIG_KERNEL_STACK_ORDER) * PAGE_SIZE) /* how to get the thread information struct from C */ static inline struct thread_info *current_thread_info(void) { struct thread_info *ti; - unsigned long mask = PAGE_SIZE * - (1 << CONFIG_KERNEL_STACK_ORDER) - 1; - ti = (struct thread_info *) (((unsigned long) &ti) & ~mask); + unsigned long mask = THREAD_SIZE - 1; + ti = (struct thread_info *) (((unsigned long) &ti) & ~mask); return ti; } /* thread information allocation */ -#define THREAD_SIZE ((1 << CONFIG_KERNEL_STACK_ORDER) * PAGE_SIZE) #define alloc_thread_info(tsk) \ ((struct thread_info *) kmalloc(THREAD_SIZE, GFP_KERNEL)) #define free_thread_info(ti) kfree(ti) -- cgit v1.2.3 From affac4bcbba617d7329b04c7882bc49a9826e12f Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sat, 28 May 2005 15:52:00 -0700 Subject: [PATCH] uml: fix PREEMPT_ACTIVE This is a continuation for UML of: http://linux.bkbits.net:8080/linux-2.5/cset@41791ab52lfMuF2i3V-eTIGRBbDYKQ Signed-off-by: Paolo 'Blaisorblade' Giarrusso Cc: Paul Mackerras Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-um/thread_info.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-um/thread_info.h b/include/asm-um/thread_info.h index e4f0198240c5..1feaaf148ef1 100644 --- a/include/asm-um/thread_info.h +++ b/include/asm-um/thread_info.h @@ -61,7 +61,7 @@ static inline struct thread_info *current_thread_info(void) #endif -#define PREEMPT_ACTIVE 0x4000000 +#define PREEMPT_ACTIVE 0x10000000 #define TIF_SYSCALL_TRACE 0 /* syscall trace active */ #define TIF_SIGPENDING 1 /* signal pending */ -- cgit v1.2.3 From 37053cc84d51a790a6e5c3987665902cdab122a5 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Sat, 28 May 2005 15:52:01 -0700 Subject: [PATCH] sh: PREEMPT_ACTIVE fix Port Paulus's ppc64 fix to sh: When the generic IRQ stuff went in, it seems that HARDIRQ_BITS got bumped from 9 (for ppc64) up to 12. Consequently, the PREEMPT_ACTIVE bit is now within HARDIRQ_MASK, and I get in_interrupt() falsely returning true when PREEMPT_ACTIVE is set, and thus a BUG_ON tripping in arch/ppc64/mm/tlb.c. Signed-off-by: Paul Mundt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-sh/thread_info.h | 2 +- include/asm-sh64/thread_info.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h index d82f883d8e6d..4bbbd9f3c37e 100644 --- a/include/asm-sh/thread_info.h +++ b/include/asm-sh/thread_info.h @@ -27,7 +27,7 @@ struct thread_info { #endif -#define PREEMPT_ACTIVE 0x4000000 +#define PREEMPT_ACTIVE 0x10000000 /* * macros/functions for gaining access to the thread information structure diff --git a/include/asm-sh64/thread_info.h b/include/asm-sh64/thread_info.h index e65f394da472..8a32d6bd0b79 100644 --- a/include/asm-sh64/thread_info.h +++ b/include/asm-sh64/thread_info.h @@ -73,7 +73,7 @@ static inline struct thread_info *current_thread_info(void) #define THREAD_SIZE 8192 -#define PREEMPT_ACTIVE 0x4000000 +#define PREEMPT_ACTIVE 0x10000000 /* thread information flags */ #define TIF_SYSCALL_TRACE 0 /* syscall trace active */ -- cgit v1.2.3 From 8f28e8fa46625310102aea06fac61ba04c8b5b88 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sat, 28 May 2005 15:52:02 -0700 Subject: [PATCH] irq code: Add coherence test for PREEMPT_ACTIVE After porting this fixlet to UML: http://linux.bkbits.net:8080/linux-2.5/cset@41791ab52lfMuF2i3V-eTIGRBbDYKQ , I've also added a warning which should refuse compilation with insane values for PREEMPT_ACTIVE... maybe we should simply move PREEMPT_ACTIVE out of architectures using GENERIC_IRQS. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/hardirq.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h index ebc712e91066..8336dba18971 100644 --- a/include/linux/hardirq.h +++ b/include/linux/hardirq.h @@ -43,13 +43,17 @@ #define __IRQ_MASK(x) ((1UL << (x))-1) #define PREEMPT_MASK (__IRQ_MASK(PREEMPT_BITS) << PREEMPT_SHIFT) -#define HARDIRQ_MASK (__IRQ_MASK(HARDIRQ_BITS) << HARDIRQ_SHIFT) #define SOFTIRQ_MASK (__IRQ_MASK(SOFTIRQ_BITS) << SOFTIRQ_SHIFT) +#define HARDIRQ_MASK (__IRQ_MASK(HARDIRQ_BITS) << HARDIRQ_SHIFT) #define PREEMPT_OFFSET (1UL << PREEMPT_SHIFT) #define SOFTIRQ_OFFSET (1UL << SOFTIRQ_SHIFT) #define HARDIRQ_OFFSET (1UL << HARDIRQ_SHIFT) +#if PREEMPT_ACTIVE < (1 << (HARDIRQ_SHIFT + HARDIRQ_BITS)) +#error PREEMPT_ACTIVE is too low! +#endif + #define hardirq_count() (preempt_count() & HARDIRQ_MASK) #define softirq_count() (preempt_count() & SOFTIRQ_MASK) #define irq_count() (preempt_count() & (HARDIRQ_MASK | SOFTIRQ_MASK)) -- cgit v1.2.3 From 8e21683bb458c73c6fe87844202a64eac730995e Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sat, 28 May 2005 15:52:04 -0700 Subject: [PATCH] uml: remove jail mode + other leftovers This var is currently useless, as it's apparent from reading the code. Until 2.6.11 it was used in some code related to jail mode, in the same proc.: if(jail){ while(!reading) sched_yield(); } jail mode has been dropped, together with that use, so let's finish dropping this. Also, remove some other useless definitions I met. Acked-by: Jeff Dike Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/kernel/tt/process_kern.c | 8 +------- arch/um/kernel/um_arch.c | 6 ------ include/asm-um/pgtable.h | 8 -------- 3 files changed, 1 insertion(+), 21 deletions(-) (limited to 'include') diff --git a/arch/um/kernel/tt/process_kern.c b/arch/um/kernel/tt/process_kern.c index df810ca8fc12..776310fd5b8b 100644 --- a/arch/um/kernel/tt/process_kern.c +++ b/arch/um/kernel/tt/process_kern.c @@ -32,10 +32,6 @@ void *switch_to_tt(void *prev, void *next, void *last) unsigned long flags; int err, vtalrm, alrm, prof, cpu; char c; - /* jailing and SMP are incompatible, so this doesn't need to be - * made per-cpu - */ - static int reading; from = prev; to = next; @@ -59,13 +55,11 @@ void *switch_to_tt(void *prev, void *next, void *last) c = 0; set_current(to); - reading = 0; err = os_write_file(to->thread.mode.tt.switch_pipe[1], &c, sizeof(c)); if(err != sizeof(c)) panic("write of switch_pipe failed, err = %d", -err); - reading = 1; - if(from->thread.mode.tt.switch_pipe[0] == -1) + if(from->thread.mode.tt.switch_pipe[0] == -1) os_kill_process(os_getpid(), 0); err = os_read_file(from->thread.mode.tt.switch_pipe[0], &c, sizeof(c)); diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c index 4d10ec372a67..418427107b29 100644 --- a/arch/um/kernel/um_arch.c +++ b/arch/um/kernel/um_arch.c @@ -111,12 +111,6 @@ struct seq_operations cpuinfo_op = { .show = show_cpuinfo, }; -pte_t * __bad_pagetable(void) -{ - panic("Someone should implement __bad_pagetable"); - return(NULL); -} - /* Set in linux_main */ unsigned long host_task_size; unsigned long task_size; diff --git a/include/asm-um/pgtable.h b/include/asm-um/pgtable.h index 510e513c7f88..a88040920311 100644 --- a/include/asm-um/pgtable.h +++ b/include/asm-um/pgtable.h @@ -114,17 +114,9 @@ extern unsigned long end_iomem; extern unsigned long pg0[1024]; /* - * BAD_PAGETABLE is used when we need a bogus page-table, while - * BAD_PAGE is used for a bogus page. - * * ZERO_PAGE is a global shared page that is always zero: used * for zero-mapped memory areas etc.. */ -extern pte_t __bad_page(void); -extern pte_t * __bad_pagetable(void); - -#define BAD_PAGETABLE __bad_pagetable() -#define BAD_PAGE __bad_page() #define ZERO_PAGE(vaddr) virt_to_page(empty_zero_page) -- cgit v1.2.3 From a6dbba77a9d4b47c60d60c4f07fa79b3ca93a1d5 Mon Sep 17 00:00:00 2001 From: Vitaly Bordug Date: Sat, 28 May 2005 15:52:09 -0700 Subject: [PATCH] ppc32: Support for 82xx PQII on-chip PCI bridge This patch adds on-chip PCI bridge support for the PQ2 family. The incomplete existent code is updated with interrupt handling stuff and board-specific bits for 8272ADS and PQ2FADS; the related files were renamed (from m8260_pci to m82xx_pci) to be of more generic fashion. This is tested with 8266ADS and 8272ADS, should work on PQ2FADS as well. Signed-off-by: Vitaly Bordug Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ppc/Kconfig | 4 +- arch/ppc/platforms/pq2ads.h | 41 ++-- arch/ppc/syslib/Makefile | 2 +- arch/ppc/syslib/m8260_pci.c | 193 ------------------ arch/ppc/syslib/m8260_pci.h | 76 ------- arch/ppc/syslib/m8260_pci_erratum9.c | 10 +- arch/ppc/syslib/m8260_setup.c | 11 +- arch/ppc/syslib/m82xx_pci.c | 383 +++++++++++++++++++++++++++++++++++ arch/ppc/syslib/m82xx_pci.h | 92 +++++++++ include/asm-ppc/cpm2.h | 46 +++++ include/asm-ppc/m8260_pci.h | 1 + include/asm-ppc/mpc8260.h | 2 +- 12 files changed, 558 insertions(+), 303 deletions(-) delete mode 100644 arch/ppc/syslib/m8260_pci.c delete mode 100644 arch/ppc/syslib/m8260_pci.h create mode 100644 arch/ppc/syslib/m82xx_pci.c create mode 100644 arch/ppc/syslib/m82xx_pci.h (limited to 'include') diff --git a/arch/ppc/Kconfig b/arch/ppc/Kconfig index 600f23d7fd33..6e6377a69d5b 100644 --- a/arch/ppc/Kconfig +++ b/arch/ppc/Kconfig @@ -1143,12 +1143,12 @@ config PCI_QSPAN config PCI_8260 bool - depends on PCI && 8260 && !8272 + depends on PCI && 8260 default y config 8260_PCI9 bool " Enable workaround for MPC826x erratum PCI 9" - depends on PCI_8260 + depends on PCI_8260 && !ADS8272 default y choice diff --git a/arch/ppc/platforms/pq2ads.h b/arch/ppc/platforms/pq2ads.h index cf5e5dd06d63..067d9a5aebc1 100644 --- a/arch/ppc/platforms/pq2ads.h +++ b/arch/ppc/platforms/pq2ads.h @@ -49,10 +49,10 @@ /* PCI interrupt controller */ #define PCI_INT_STAT_REG 0xF8200000 #define PCI_INT_MASK_REG 0xF8200004 -#define PIRQA (NR_SIU_INTS + 0) -#define PIRQB (NR_SIU_INTS + 1) -#define PIRQC (NR_SIU_INTS + 2) -#define PIRQD (NR_SIU_INTS + 3) +#define PIRQA (NR_CPM_INTS + 0) +#define PIRQB (NR_CPM_INTS + 1) +#define PIRQC (NR_CPM_INTS + 2) +#define PIRQD (NR_CPM_INTS + 3) /* * PCI memory map definitions for MPC8266ADS-PCI. @@ -68,28 +68,23 @@ * 0x00000000-0x1FFFFFFF 0x00000000-0x1FFFFFFF MPC8266 local memory */ -/* window for a PCI master to access MPC8266 memory */ -#define PCI_SLV_MEM_LOCAL 0x00000000 /* Local base */ -#define PCI_SLV_MEM_BUS 0x00000000 /* PCI base */ +/* All the other PCI memory map definitions reside at syslib/m82xx_pci.h + Here we should redefine what is unique for this board */ +#define M82xx_PCI_SLAVE_MEM_LOCAL 0x00000000 /* Local base */ +#define M82xx_PCI_SLAVE_MEM_BUS 0x00000000 /* PCI base */ +#define M82xx_PCI_SLAVE_MEM_SIZE 0x10000000 /* 256 Mb */ -/* window for the processor to access PCI memory with prefetching */ -#define PCI_MSTR_MEM_LOCAL 0x80000000 /* Local base */ -#define PCI_MSTR_MEM_BUS 0x80000000 /* PCI base */ -#define PCI_MSTR_MEM_SIZE 0x20000000 /* 512MB */ +#define M82xx_PCI_SLAVE_SEC_WND_SIZE ~(0x40000000 - 1U) /* 2 x 512Mb */ +#define M82xx_PCI_SLAVE_SEC_WND_BASE 0x80000000 /* PCI Memory base */ -/* window for the processor to access PCI memory without prefetching */ -#define PCI_MSTR_MEMIO_LOCAL 0xA0000000 /* Local base */ -#define PCI_MSTR_MEMIO_BUS 0xA0000000 /* PCI base */ -#define PCI_MSTR_MEMIO_SIZE 0x20000000 /* 512MB */ +#if defined(CONFIG_ADS8272) +#define PCI_INT_TO_SIU SIU_INT_IRQ2 +#elif defined(CONFIG_PQ2FADS) +#define PCI_INT_TO_SIU SIU_INT_IRQ6 +#else +#warning PCI Bridge will be without interrupts support +#endif -/* window for the processor to access PCI I/O */ -#define PCI_MSTR_IO_LOCAL 0xF4000000 /* Local base */ -#define PCI_MSTR_IO_BUS 0x00000000 /* PCI base */ -#define PCI_MSTR_IO_SIZE 0x04000000 /* 64MB */ - -#define _IO_BASE PCI_MSTR_IO_LOCAL -#define _ISA_MEM_BASE PCI_MSTR_MEMIO_LOCAL -#define PCI_DRAM_OFFSET PCI_SLV_MEM_BUS #endif /* CONFIG_PCI */ #endif /* __MACH_ADS8260_DEFS */ diff --git a/arch/ppc/syslib/Makefile b/arch/ppc/syslib/Makefile index 46cbb3e39d4c..96acf85800d4 100644 --- a/arch/ppc/syslib/Makefile +++ b/arch/ppc/syslib/Makefile @@ -81,7 +81,7 @@ obj-$(CONFIG_SBC82xx) += todc_time.o obj-$(CONFIG_SPRUCE) += cpc700_pic.o indirect_pci.o pci_auto.o \ todc_time.o obj-$(CONFIG_8260) += m8260_setup.o -obj-$(CONFIG_PCI_8260) += m8260_pci.o indirect_pci.o +obj-$(CONFIG_PCI_8260) += m82xx_pci.o indirect_pci.o pci_auto.o obj-$(CONFIG_8260_PCI9) += m8260_pci_erratum9.o obj-$(CONFIG_CPM2) += cpm2_common.o cpm2_pic.o ifeq ($(CONFIG_PPC_GEN550),y) diff --git a/arch/ppc/syslib/m8260_pci.c b/arch/ppc/syslib/m8260_pci.c deleted file mode 100644 index 057cc3f8ff37..000000000000 --- a/arch/ppc/syslib/m8260_pci.c +++ /dev/null @@ -1,193 +0,0 @@ -/* - * (C) Copyright 2003 - * Wolfgang Denk, DENX Software Engineering, wd@denx.de. - * - * (C) Copyright 2004 Red Hat, Inc. - * - * See file CREDITS for list of people who contributed to this - * project. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation; either version 2 of - * the License, or (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, - * MA 02111-1307 USA - */ - -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include -#include -#include - -#include "m8260_pci.h" - - -/* PCI bus configuration registers. - */ - -static void __init m8260_setup_pci(struct pci_controller *hose) -{ - volatile cpm2_map_t *immap = cpm2_immr; - unsigned long pocmr; - u16 tempShort; - -#ifndef CONFIG_ATC /* already done in U-Boot */ - /* - * Setting required to enable IRQ1-IRQ7 (SIUMCR [DPPC]), - * and local bus for PCI (SIUMCR [LBPC]). - */ - immap->im_siu_conf.siu_82xx.sc_siumcr = 0x00640000; -#endif - - /* Make PCI lowest priority */ - /* Each 4 bits is a device bus request and the MS 4bits - is highest priority */ - /* Bus 4bit value - --- ---------- - CPM high 0b0000 - CPM middle 0b0001 - CPM low 0b0010 - PCI reguest 0b0011 - Reserved 0b0100 - Reserved 0b0101 - Internal Core 0b0110 - External Master 1 0b0111 - External Master 2 0b1000 - External Master 3 0b1001 - The rest are reserved */ - immap->im_siu_conf.siu_82xx.sc_ppc_alrh = 0x61207893; - - /* Park bus on core while modifying PCI Bus accesses */ - immap->im_siu_conf.siu_82xx.sc_ppc_acr = 0x6; - - /* - * Set up master window that allows the CPU to access PCI space. This - * window is set up using the first SIU PCIBR registers. - */ - immap->im_memctl.memc_pcimsk0 = MPC826x_PCI_MASK; - immap->im_memctl.memc_pcibr0 = MPC826x_PCI_BASE | PCIBR_ENABLE; - - /* Disable machine check on no response or target abort */ - immap->im_pci.pci_emr = cpu_to_le32(0x1fe7); - /* Release PCI RST (by default the PCI RST signal is held low) */ - immap->im_pci.pci_gcr = cpu_to_le32(PCIGCR_PCI_BUS_EN); - - /* give it some time */ - mdelay(1); - - /* - * Set up master window that allows the CPU to access PCI Memory (prefetch) - * space. This window is set up using the first set of Outbound ATU registers. - */ - immap->im_pci.pci_potar0 = cpu_to_le32(MPC826x_PCI_LOWER_MEM >> 12); - immap->im_pci.pci_pobar0 = cpu_to_le32((MPC826x_PCI_LOWER_MEM - MPC826x_PCI_MEM_OFFSET) >> 12); - pocmr = ((MPC826x_PCI_UPPER_MEM - MPC826x_PCI_LOWER_MEM) >> 12) ^ 0xfffff; - immap->im_pci.pci_pocmr0 = cpu_to_le32(pocmr | POCMR_ENABLE | POCMR_PREFETCH_EN); - - /* - * Set up master window that allows the CPU to access PCI Memory (non-prefetch) - * space. This window is set up using the second set of Outbound ATU registers. - */ - immap->im_pci.pci_potar1 = cpu_to_le32(MPC826x_PCI_LOWER_MMIO >> 12); - immap->im_pci.pci_pobar1 = cpu_to_le32((MPC826x_PCI_LOWER_MMIO - MPC826x_PCI_MMIO_OFFSET) >> 12); - pocmr = ((MPC826x_PCI_UPPER_MMIO - MPC826x_PCI_LOWER_MMIO) >> 12) ^ 0xfffff; - immap->im_pci.pci_pocmr1 = cpu_to_le32(pocmr | POCMR_ENABLE); - - /* - * Set up master window that allows the CPU to access PCI IO space. This window - * is set up using the third set of Outbound ATU registers. - */ - immap->im_pci.pci_potar2 = cpu_to_le32(MPC826x_PCI_IO_BASE >> 12); - immap->im_pci.pci_pobar2 = cpu_to_le32(MPC826x_PCI_LOWER_IO >> 12); - pocmr = ((MPC826x_PCI_UPPER_IO - MPC826x_PCI_LOWER_IO) >> 12) ^ 0xfffff; - immap->im_pci.pci_pocmr2 = cpu_to_le32(pocmr | POCMR_ENABLE | POCMR_PCI_IO); - - /* - * Set up slave window that allows PCI masters to access MPC826x local memory. - * This window is set up using the first set of Inbound ATU registers - */ - - immap->im_pci.pci_pitar0 = cpu_to_le32(MPC826x_PCI_SLAVE_MEM_LOCAL >> 12); - immap->im_pci.pci_pibar0 = cpu_to_le32(MPC826x_PCI_SLAVE_MEM_BUS >> 12); - pocmr = ((MPC826x_PCI_SLAVE_MEM_SIZE-1) >> 12) ^ 0xfffff; - immap->im_pci.pci_picmr0 = cpu_to_le32(pocmr | PICMR_ENABLE | PICMR_PREFETCH_EN); - - /* See above for description - puts PCI request as highest priority */ - immap->im_siu_conf.siu_82xx.sc_ppc_alrh = 0x03124567; - - /* Park the bus on the PCI */ - immap->im_siu_conf.siu_82xx.sc_ppc_acr = PPC_ACR_BUS_PARK_PCI; - - /* Host mode - specify the bridge as a host-PCI bridge */ - early_write_config_word(hose, 0, 0, PCI_CLASS_DEVICE, PCI_CLASS_BRIDGE_HOST); - - /* Enable the host bridge to be a master on the PCI bus, and to act as a PCI memory target */ - early_read_config_word(hose, 0, 0, PCI_COMMAND, &tempShort); - early_write_config_word(hose, 0, 0, PCI_COMMAND, - tempShort | PCI_COMMAND_MASTER | PCI_COMMAND_MEMORY); -} - -void __init m8260_find_bridges(void) -{ - extern int pci_assign_all_busses; - struct pci_controller * hose; - - pci_assign_all_busses = 1; - - hose = pcibios_alloc_controller(); - - if (!hose) - return; - - ppc_md.pci_swizzle = common_swizzle; - - hose->first_busno = 0; - hose->bus_offset = 0; - hose->last_busno = 0xff; - - setup_m8260_indirect_pci(hose, - (unsigned long)&cpm2_immr->im_pci.pci_cfg_addr, - (unsigned long)&cpm2_immr->im_pci.pci_cfg_data); - - m8260_setup_pci(hose); - hose->pci_mem_offset = MPC826x_PCI_MEM_OFFSET; - - hose->io_base_virt = ioremap(MPC826x_PCI_IO_BASE, - MPC826x_PCI_IO_SIZE); - isa_io_base = (unsigned long) hose->io_base_virt; - - /* setup resources */ - pci_init_resource(&hose->mem_resources[0], - MPC826x_PCI_LOWER_MEM, - MPC826x_PCI_UPPER_MEM, - IORESOURCE_MEM|IORESOURCE_PREFETCH, "PCI prefetchable memory"); - - pci_init_resource(&hose->mem_resources[1], - MPC826x_PCI_LOWER_MMIO, - MPC826x_PCI_UPPER_MMIO, - IORESOURCE_MEM, "PCI memory"); - - pci_init_resource(&hose->io_resource, - MPC826x_PCI_LOWER_IO, - MPC826x_PCI_UPPER_IO, - IORESOURCE_IO, "PCI I/O"); -} diff --git a/arch/ppc/syslib/m8260_pci.h b/arch/ppc/syslib/m8260_pci.h deleted file mode 100644 index d1352120acd7..000000000000 --- a/arch/ppc/syslib/m8260_pci.h +++ /dev/null @@ -1,76 +0,0 @@ - -#ifndef _PPC_KERNEL_M8260_PCI_H -#define _PPC_KERNEL_M8260_PCI_H - -#include - -/* - * Local->PCI map (from CPU) controlled by - * MPC826x master window - * - * 0x80000000 - 0xBFFFFFFF Total CPU2PCI space PCIBR0 - * - * 0x80000000 - 0x9FFFFFFF PCI Mem with prefetch (Outbound ATU #1) - * 0xA0000000 - 0xAFFFFFFF PCI Mem w/o prefetch (Outbound ATU #2) - * 0xB0000000 - 0xB0FFFFFF 32-bit PCI IO (Outbound ATU #3) - * - * PCI->Local map (from PCI) - * MPC826x slave window controlled by - * - * 0x00000000 - 0x07FFFFFF MPC826x local memory (Inbound ATU #1) - */ - -/* - * Slave window that allows PCI masters to access MPC826x local memory. - * This window is set up using the first set of Inbound ATU registers - */ - -#ifndef MPC826x_PCI_SLAVE_MEM_LOCAL -#define MPC826x_PCI_SLAVE_MEM_LOCAL (((struct bd_info *)__res)->bi_memstart) -#define MPC826x_PCI_SLAVE_MEM_BUS (((struct bd_info *)__res)->bi_memstart) -#define MPC826x_PCI_SLAVE_MEM_SIZE (((struct bd_info *)__res)->bi_memsize) -#endif - -/* - * This is the window that allows the CPU to access PCI address space. - * It will be setup with the SIU PCIBR0 register. All three PCI master - * windows, which allow the CPU to access PCI prefetch, non prefetch, - * and IO space (see below), must all fit within this window. - */ -#ifndef MPC826x_PCI_BASE -#define MPC826x_PCI_BASE 0x80000000 -#define MPC826x_PCI_MASK 0xc0000000 -#endif - -#ifndef MPC826x_PCI_LOWER_MEM -#define MPC826x_PCI_LOWER_MEM 0x80000000 -#define MPC826x_PCI_UPPER_MEM 0x9fffffff -#define MPC826x_PCI_MEM_OFFSET 0x00000000 -#endif - -#ifndef MPC826x_PCI_LOWER_MMIO -#define MPC826x_PCI_LOWER_MMIO 0xa0000000 -#define MPC826x_PCI_UPPER_MMIO 0xafffffff -#define MPC826x_PCI_MMIO_OFFSET 0x00000000 -#endif - -#ifndef MPC826x_PCI_LOWER_IO -#define MPC826x_PCI_LOWER_IO 0x00000000 -#define MPC826x_PCI_UPPER_IO 0x00ffffff -#define MPC826x_PCI_IO_BASE 0xb0000000 -#define MPC826x_PCI_IO_SIZE 0x01000000 -#endif - -#ifndef _IO_BASE -#define _IO_BASE isa_io_base -#endif - -#ifdef CONFIG_8260_PCI9 -struct pci_controller; -extern void setup_m8260_indirect_pci(struct pci_controller* hose, - u32 cfg_addr, u32 cfg_data); -#else -#define setup_m8260_indirect_pci setup_indirect_pci -#endif - -#endif /* _PPC_KERNEL_M8260_PCI_H */ diff --git a/arch/ppc/syslib/m8260_pci_erratum9.c b/arch/ppc/syslib/m8260_pci_erratum9.c index 9c0582d639e0..1dc7e4e1d491 100644 --- a/arch/ppc/syslib/m8260_pci_erratum9.c +++ b/arch/ppc/syslib/m8260_pci_erratum9.c @@ -31,7 +31,7 @@ #include #include -#include "m8260_pci.h" +#include "m82xx_pci.h" #ifdef CONFIG_8260_PCI9 /*#include */ /* included in asm/io.h */ @@ -248,11 +248,11 @@ EXPORT_SYMBOL(idma_pci9_read_le); static inline int is_pci_mem(unsigned long addr) { - if (addr >= MPC826x_PCI_LOWER_MMIO && - addr <= MPC826x_PCI_UPPER_MMIO) + if (addr >= M82xx_PCI_LOWER_MMIO && + addr <= M82xx_PCI_UPPER_MMIO) return 1; - if (addr >= MPC826x_PCI_LOWER_MEM && - addr <= MPC826x_PCI_UPPER_MEM) + if (addr >= M82xx_PCI_LOWER_MEM && + addr <= M82xx_PCI_UPPER_MEM) return 1; return 0; } diff --git a/arch/ppc/syslib/m8260_setup.c b/arch/ppc/syslib/m8260_setup.c index 23ea3f694de2..fda75d79050c 100644 --- a/arch/ppc/syslib/m8260_setup.c +++ b/arch/ppc/syslib/m8260_setup.c @@ -34,7 +34,8 @@ unsigned char __res[sizeof(bd_t)]; extern void cpm2_reset(void); -extern void m8260_find_bridges(void); +extern void pq2_find_bridges(void); +extern void pq2pci_init_irq(void); extern void idma_pci9_init(void); /* Place-holder for board-specific init */ @@ -56,7 +57,7 @@ m8260_setup_arch(void) idma_pci9_init(); #endif #ifdef CONFIG_PCI_8260 - m8260_find_bridges(); + pq2_find_bridges(); #endif #ifdef CONFIG_BLK_DEV_INITRD if (initrd_start) @@ -173,6 +174,12 @@ m8260_init_IRQ(void) * in case the boot rom changed something on us. */ cpm2_immr->im_intctl.ic_siprr = 0x05309770; + +#if defined(CONFIG_PCI) && (defined(CONFIG_ADS8272) || defined(CONFIG_PQ2FADS)) + /* Initialize stuff for the 82xx CPLD IC and install demux */ + pq2pci_init_irq(); +#endif + } /* diff --git a/arch/ppc/syslib/m82xx_pci.c b/arch/ppc/syslib/m82xx_pci.c new file mode 100644 index 000000000000..5e7a7edcea74 --- /dev/null +++ b/arch/ppc/syslib/m82xx_pci.c @@ -0,0 +1,383 @@ +/* + * + * (C) Copyright 2003 + * Wolfgang Denk, DENX Software Engineering, wd@denx.de. + * + * (C) Copyright 2004 Red Hat, Inc. + * + * 2005 (c) MontaVista Software, Inc. + * Vitaly Bordug + * + * See file CREDITS for list of people who contributed to this + * project. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation; either version 2 of + * the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, + * MA 02111-1307 USA + */ + +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "m82xx_pci.h" + +/* + * Interrupt routing + */ + +static inline int +pq2pci_map_irq(struct pci_dev *dev, unsigned char idsel, unsigned char pin) +{ + static char pci_irq_table[][4] = + /* + * PCI IDSEL/INTPIN->INTLINE + * A B C D + */ + { + { PIRQA, PIRQB, PIRQC, PIRQD }, /* IDSEL 22 - PCI slot 0 */ + { PIRQD, PIRQA, PIRQB, PIRQC }, /* IDSEL 23 - PCI slot 1 */ + { PIRQC, PIRQD, PIRQA, PIRQB }, /* IDSEL 24 - PCI slot 2 */ + }; + + const long min_idsel = 22, max_idsel = 24, irqs_per_slot = 4; + return PCI_IRQ_TABLE_LOOKUP; +} + +static void +pq2pci_mask_irq(unsigned int irq) +{ + int bit = irq - NR_CPM_INTS; + + *(volatile unsigned long *) PCI_INT_MASK_REG |= (1 << (31 - bit)); + return; +} + +static void +pq2pci_unmask_irq(unsigned int irq) +{ + int bit = irq - NR_CPM_INTS; + + *(volatile unsigned long *) PCI_INT_MASK_REG &= ~(1 << (31 - bit)); + return; +} + +static void +pq2pci_mask_and_ack(unsigned int irq) +{ + int bit = irq - NR_CPM_INTS; + + *(volatile unsigned long *) PCI_INT_MASK_REG |= (1 << (31 - bit)); + return; +} + +static void +pq2pci_end_irq(unsigned int irq) +{ + int bit = irq - NR_CPM_INTS; + + *(volatile unsigned long *) PCI_INT_MASK_REG &= ~(1 << (31 - bit)); + return; +} + +struct hw_interrupt_type pq2pci_ic = { + "PQ2 PCI", + NULL, + NULL, + pq2pci_unmask_irq, + pq2pci_mask_irq, + pq2pci_mask_and_ack, + pq2pci_end_irq, + 0 +}; + +static irqreturn_t +pq2pci_irq_demux(int irq, void *dev_id, struct pt_regs *regs) +{ + unsigned long stat, mask, pend; + int bit; + + for(;;) { + stat = *(volatile unsigned long *) PCI_INT_STAT_REG; + mask = *(volatile unsigned long *) PCI_INT_MASK_REG; + pend = stat & ~mask & 0xf0000000; + if (!pend) + break; + for (bit = 0; pend != 0; ++bit, pend <<= 1) { + if (pend & 0x80000000) + __do_IRQ(NR_CPM_INTS + bit, regs); + } + } + + return IRQ_HANDLED; +} + +static struct irqaction pq2pci_irqaction = { + .handler = pq2pci_irq_demux, + .flags = SA_INTERRUPT, + .mask = CPU_MASK_NONE, + .name = "PQ2 PCI cascade", +}; + + +void +pq2pci_init_irq(void) +{ + int irq; + volatile cpm2_map_t *immap = cpm2_immr; +#if defined CONFIG_ADS8272 + /* configure chip select for PCI interrupt controller */ + immap->im_memctl.memc_br3 = PCI_INT_STAT_REG | 0x00001801; + immap->im_memctl.memc_or3 = 0xffff8010; +#elif defined CONFIG_PQ2FADS + immap->im_memctl.memc_br8 = PCI_INT_STAT_REG | 0x00001801; + immap->im_memctl.memc_or8 = 0xffff8010; +#endif + for (irq = NR_CPM_INTS; irq < NR_CPM_INTS + 4; irq++) + irq_desc[irq].handler = &pq2pci_ic; + + /* make PCI IRQ level sensitive */ + immap->im_intctl.ic_siexr &= + ~(1 << (14 - (PCI_INT_TO_SIU - SIU_INT_IRQ1))); + + /* mask all PCI interrupts */ + *(volatile unsigned long *) PCI_INT_MASK_REG |= 0xfff00000; + + /* install the demultiplexer for the PCI cascade interrupt */ + setup_irq(PCI_INT_TO_SIU, &pq2pci_irqaction); + return; +} + +static int +pq2pci_exclude_device(u_char bus, u_char devfn) +{ + return PCIBIOS_SUCCESSFUL; +} + +/* PCI bus configuration registers. + */ +static void +pq2ads_setup_pci(struct pci_controller *hose) +{ + __u32 val; + volatile cpm2_map_t *immap = cpm2_immr; + bd_t* binfo = (bd_t*) __res; + u32 sccr = immap->im_clkrst.car_sccr; + uint pci_div,freq,time; + /* PCI int lowest prio */ + /* Each 4 bits is a device bus request and the MS 4bits + is highest priority */ + /* Bus 4bit value + --- ---------- + CPM high 0b0000 + CPM middle 0b0001 + CPM low 0b0010 + PCI reguest 0b0011 + Reserved 0b0100 + Reserved 0b0101 + Internal Core 0b0110 + External Master 1 0b0111 + External Master 2 0b1000 + External Master 3 0b1001 + The rest are reserved + */ + immap->im_siu_conf.siu_82xx.sc_ppc_alrh = 0x61207893; + /* park bus on core */ + immap->im_siu_conf.siu_82xx.sc_ppc_acr = PPC_ACR_BUS_PARK_CORE; + /* + * Set up master windows that allow the CPU to access PCI space. These + * windows are set up using the two SIU PCIBR registers. + */ + + immap->im_memctl.memc_pcimsk0 = M82xx_PCI_PRIM_WND_SIZE; + immap->im_memctl.memc_pcibr0 = M82xx_PCI_PRIM_WND_BASE | PCIBR_ENABLE; + +#ifdef M82xx_PCI_SEC_WND_SIZE + immap->im_memctl.memc_pcimsk1 = M82xx_PCI_SEC_WND_SIZE; + immap->im_memctl.memc_pcibr1 = M82xx_PCI_SEC_WND_BASE | PCIBR_ENABLE; +#endif + +#if defined CONFIG_ADS8272 + immap->im_siu_conf.siu_82xx.sc_siumcr = + (immap->im_siu_conf.siu_82xx.sc_siumcr & + ~(SIUMCR_BBD | SIUMCR_ESE | SIUMCR_PBSE | + SIUMCR_CDIS | SIUMCR_DPPC11 | SIUMCR_L2CPC11 | + SIUMCR_LBPC11 | SIUMCR_APPC11 | + SIUMCR_CS10PC11 | SIUMCR_BCTLC11 | SIUMCR_MMR11)) | + SIUMCR_DPPC11 | SIUMCR_L2CPC01 | SIUMCR_LBPC00 | + SIUMCR_APPC10 | SIUMCR_CS10PC00 | + SIUMCR_BCTLC00 | SIUMCR_MMR11 ; + +#elif defined CONFIG_PQ2FADS + /* + * Setting required to enable IRQ1-IRQ7 (SIUMCR [DPPC]), + * and local bus for PCI (SIUMCR [LBPC]). + */ + immap->im_siu_conf.siu_82xx.sc_siumcr = (immap->im_siu_conf.sc_siumcr & + ~(SIUMCR_L2PC11 | SIUMCR_LBPC11 | SIUMCR_CS10PC11 | SIUMCR_APPC11) | + SIUMCR_BBD | SIUMCR_LBPC01 | SIUMCR_DPPC11 | SIUMCR_APPC10; +#endif + /* Enable PCI */ + immap->im_pci.pci_gcr = cpu_to_le32(PCIGCR_PCI_BUS_EN); + + pci_div = ( (sccr & SCCR_PCI_MODCK) ? 2 : 1) * + ( ( (sccr & SCCR_PCIDF_MSK) >> SCCR_PCIDF_SHIFT) + 1); + freq = (uint)((2*binfo->bi_cpmfreq)/(pci_div)); + time = (int)666666/freq; + /* due to PCI Local Bus spec, some devices needs to wait such a long + time after RST deassertion. More specifically, 0.508s for 66MHz & twice more for 33 */ + printk("%s: The PCI bus is %d Mhz.\nWaiting %s after deasserting RST...\n",__FILE__,freq, + (time==1) ? "0.5 seconds":"1 second" ); + + { + int i; + for(i=0;i<(500*time);i++) + udelay(1000); + } + + /* setup ATU registers */ + immap->im_pci.pci_pocmr0 = cpu_to_le32(POCMR_ENABLE | POCMR_PCI_IO | + ((~(M82xx_PCI_IO_SIZE - 1U)) >> POTA_ADDR_SHIFT)); + immap->im_pci.pci_potar0 = cpu_to_le32(M82xx_PCI_LOWER_IO >> POTA_ADDR_SHIFT); + immap->im_pci.pci_pobar0 = cpu_to_le32(M82xx_PCI_IO_BASE >> POTA_ADDR_SHIFT); + + /* Set-up non-prefetchable window */ + immap->im_pci.pci_pocmr1 = cpu_to_le32(POCMR_ENABLE | ((~(M82xx_PCI_MMIO_SIZE-1U)) >> POTA_ADDR_SHIFT)); + immap->im_pci.pci_potar1 = cpu_to_le32(M82xx_PCI_LOWER_MMIO >> POTA_ADDR_SHIFT); + immap->im_pci.pci_pobar1 = cpu_to_le32((M82xx_PCI_LOWER_MMIO - M82xx_PCI_MMIO_OFFSET) >> POTA_ADDR_SHIFT); + + /* Set-up prefetchable window */ + immap->im_pci.pci_pocmr2 = cpu_to_le32(POCMR_ENABLE |POCMR_PREFETCH_EN | + (~(M82xx_PCI_MEM_SIZE-1U) >> POTA_ADDR_SHIFT)); + immap->im_pci.pci_potar2 = cpu_to_le32(M82xx_PCI_LOWER_MEM >> POTA_ADDR_SHIFT); + immap->im_pci.pci_pobar2 = cpu_to_le32((M82xx_PCI_LOWER_MEM - M82xx_PCI_MEM_OFFSET) >> POTA_ADDR_SHIFT); + + /* Inbound transactions from PCI memory space */ + immap->im_pci.pci_picmr0 = cpu_to_le32(PICMR_ENABLE | PICMR_PREFETCH_EN | + ((~(M82xx_PCI_SLAVE_MEM_SIZE-1U)) >> PITA_ADDR_SHIFT)); + immap->im_pci.pci_pibar0 = cpu_to_le32(M82xx_PCI_SLAVE_MEM_BUS >> PITA_ADDR_SHIFT); + immap->im_pci.pci_pitar0 = cpu_to_le32(M82xx_PCI_SLAVE_MEM_LOCAL>> PITA_ADDR_SHIFT); + +#if defined CONFIG_ADS8272 + /* PCI int highest prio */ + immap->im_siu_conf.siu_82xx.sc_ppc_alrh = 0x01236745; +#elif defined CONFIG_PQ2FADS + immap->im_siu_conf.siu_82xx.sc_ppc_alrh = 0x03124567; +#endif + /* park bus on PCI */ + immap->im_siu_conf.siu_82xx.sc_ppc_acr = PPC_ACR_BUS_PARK_PCI; + + /* Enable bus mastering and inbound memory transactions */ + early_read_config_dword(hose, hose->first_busno, 0, PCI_COMMAND, &val); + val &= 0xffff0000; + val |= PCI_COMMAND_MEMORY|PCI_COMMAND_MASTER; + early_write_config_dword(hose, hose->first_busno, 0, PCI_COMMAND, val); + +} + +void __init pq2_find_bridges(void) +{ + extern int pci_assign_all_busses; + struct pci_controller * hose; + int host_bridge; + + pci_assign_all_busses = 1; + + hose = pcibios_alloc_controller(); + + if (!hose) + return; + + ppc_md.pci_swizzle = common_swizzle; + + hose->first_busno = 0; + hose->bus_offset = 0; + hose->last_busno = 0xff; + +#ifdef CONFIG_ADS8272 + hose->set_cfg_type = 1; +#endif + + setup_m8260_indirect_pci(hose, + (unsigned long)&cpm2_immr->im_pci.pci_cfg_addr, + (unsigned long)&cpm2_immr->im_pci.pci_cfg_data); + + /* Make sure it is a supported bridge */ + early_read_config_dword(hose, + 0, + PCI_DEVFN(0,0), + PCI_VENDOR_ID, + &host_bridge); + switch (host_bridge) { + case PCI_DEVICE_ID_MPC8265: + break; + case PCI_DEVICE_ID_MPC8272: + break; + default: + printk("Attempting to use unrecognized host bridge ID" + " 0x%08x.\n", host_bridge); + break; + } + + pq2ads_setup_pci(hose); + + hose->io_space.start = M82xx_PCI_LOWER_IO; + hose->io_space.end = M82xx_PCI_UPPER_IO; + hose->mem_space.start = M82xx_PCI_LOWER_MEM; + hose->mem_space.end = M82xx_PCI_UPPER_MMIO; + hose->pci_mem_offset = M82xx_PCI_MEM_OFFSET; + + isa_io_base = + (unsigned long) ioremap(M82xx_PCI_IO_BASE, + M82xx_PCI_IO_SIZE); + hose->io_base_virt = (void *) isa_io_base; + + /* setup resources */ + pci_init_resource(&hose->mem_resources[0], + M82xx_PCI_LOWER_MEM, + M82xx_PCI_UPPER_MEM, + IORESOURCE_MEM|IORESOURCE_PREFETCH, "PCI prefetchable memory"); + + pci_init_resource(&hose->mem_resources[1], + M82xx_PCI_LOWER_MMIO, + M82xx_PCI_UPPER_MMIO, + IORESOURCE_MEM, "PCI memory"); + + pci_init_resource(&hose->io_resource, + M82xx_PCI_LOWER_IO, + M82xx_PCI_UPPER_IO, + IORESOURCE_IO | 1, "PCI I/O"); + + ppc_md.pci_exclude_device = pq2pci_exclude_device; + hose->last_busno = pciauto_bus_scan(hose, hose->first_busno); + + ppc_md.pci_map_irq = pq2pci_map_irq; + ppc_md.pcibios_fixup = NULL; + ppc_md.pcibios_fixup_bus = NULL; + +} diff --git a/arch/ppc/syslib/m82xx_pci.h b/arch/ppc/syslib/m82xx_pci.h new file mode 100644 index 000000000000..924f73f8e595 --- /dev/null +++ b/arch/ppc/syslib/m82xx_pci.h @@ -0,0 +1,92 @@ + +#ifndef _PPC_KERNEL_M82XX_PCI_H +#define _PPC_KERNEL_M82XX_PCI_H + +#include +/* + * Local->PCI map (from CPU) controlled by + * MPC826x master window + * + * 0xF6000000 - 0xF7FFFFFF IO space + * 0x80000000 - 0xBFFFFFFF CPU2PCI memory space PCIBR0 + * + * 0x80000000 - 0x9FFFFFFF PCI Mem with prefetch (Outbound ATU #1) + * 0xA0000000 - 0xBFFFFFFF PCI Mem w/o prefetch (Outbound ATU #2) + * 0xF6000000 - 0xF7FFFFFF 32-bit PCI IO (Outbound ATU #3) + * + * PCI->Local map (from PCI) + * MPC826x slave window controlled by + * + * 0x00000000 - 0x07FFFFFF MPC826x local memory (Inbound ATU #1) + */ + +/* + * Slave window that allows PCI masters to access MPC826x local memory. + * This window is set up using the first set of Inbound ATU registers + */ + +#ifndef M82xx_PCI_SLAVE_MEM_LOCAL +#define M82xx_PCI_SLAVE_MEM_LOCAL (((struct bd_info *)__res)->bi_memstart) +#define M82xx_PCI_SLAVE_MEM_BUS (((struct bd_info *)__res)->bi_memstart) +#define M82xx_PCI_SLAVE_MEM_SIZE (((struct bd_info *)__res)->bi_memsize) +#endif + +/* + * This is the window that allows the CPU to access PCI address space. + * It will be setup with the SIU PCIBR0 register. All three PCI master + * windows, which allow the CPU to access PCI prefetch, non prefetch, + * and IO space (see below), must all fit within this window. + */ + +#ifndef M82xx_PCI_LOWER_MEM +#define M82xx_PCI_LOWER_MEM 0x80000000 +#define M82xx_PCI_UPPER_MEM 0x9fffffff +#define M82xx_PCI_MEM_OFFSET 0x00000000 +#define M82xx_PCI_MEM_SIZE 0x20000000 +#endif + +#ifndef M82xx_PCI_LOWER_MMIO +#define M82xx_PCI_LOWER_MMIO 0xa0000000 +#define M82xx_PCI_UPPER_MMIO 0xafffffff +#define M82xx_PCI_MMIO_OFFSET 0x00000000 +#define M82xx_PCI_MMIO_SIZE 0x20000000 +#endif + +#ifndef M82xx_PCI_LOWER_IO +#define M82xx_PCI_LOWER_IO 0x00000000 +#define M82xx_PCI_UPPER_IO 0x01ffffff +#define M82xx_PCI_IO_BASE 0xf6000000 +#define M82xx_PCI_IO_SIZE 0x02000000 +#endif + +#ifndef M82xx_PCI_PRIM_WND_SIZE +#define M82xx_PCI_PRIM_WND_SIZE ~(M82xx_PCI_IO_SIZE - 1U) +#define M82xx_PCI_PRIM_WND_BASE (M82xx_PCI_IO_BASE) +#endif + +#ifndef M82xx_PCI_SEC_WND_SIZE +#define M82xx_PCI_SEC_WND_SIZE ~(M82xx_PCI_MEM_SIZE + M82xx_PCI_MMIO_SIZE - 1U) +#define M82xx_PCI_SEC_WND_BASE (M82xx_PCI_LOWER_MEM) +#endif + +#ifndef POTA_ADDR_SHIFT +#define POTA_ADDR_SHIFT 12 +#endif + +#ifndef PITA_ADDR_SHIFT +#define PITA_ADDR_SHIFT 12 +#endif + +#ifndef _IO_BASE +#define _IO_BASE isa_io_base +#endif + +#ifdef CONFIG_8260_PCI9 +struct pci_controller; +extern void setup_m8260_indirect_pci(struct pci_controller* hose, + u32 cfg_addr, u32 cfg_data); +#else +#define setup_m8260_indirect_pci setup_indirect_pci +#endif + +#endif /* _PPC_KERNEL_M8260_PCI_H */ diff --git a/include/asm-ppc/cpm2.h b/include/asm-ppc/cpm2.h index 42fd1068cf2a..c5883dbed63f 100644 --- a/include/asm-ppc/cpm2.h +++ b/include/asm-ppc/cpm2.h @@ -1039,6 +1039,52 @@ typedef struct im_idma { #define CMXSCR_TS4CS_CLK7 0x00000006 /* SCC4 Tx Clock Source is CLK7 */ #define CMXSCR_TS4CS_CLK8 0x00000007 /* SCC4 Tx Clock Source is CLK8 */ +/*----------------------------------------------------------------------- + * SIUMCR - SIU Module Configuration Register 4-31 + */ +#define SIUMCR_BBD 0x80000000 /* Bus Busy Disable */ +#define SIUMCR_ESE 0x40000000 /* External Snoop Enable */ +#define SIUMCR_PBSE 0x20000000 /* Parity Byte Select Enable */ +#define SIUMCR_CDIS 0x10000000 /* Core Disable */ +#define SIUMCR_DPPC00 0x00000000 /* Data Parity Pins Configuration*/ +#define SIUMCR_DPPC01 0x04000000 /* - " - */ +#define SIUMCR_DPPC10 0x08000000 /* - " - */ +#define SIUMCR_DPPC11 0x0c000000 /* - " - */ +#define SIUMCR_L2CPC00 0x00000000 /* L2 Cache Pins Configuration */ +#define SIUMCR_L2CPC01 0x01000000 /* - " - */ +#define SIUMCR_L2CPC10 0x02000000 /* - " - */ +#define SIUMCR_L2CPC11 0x03000000 /* - " - */ +#define SIUMCR_LBPC00 0x00000000 /* Local Bus Pins Configuration */ +#define SIUMCR_LBPC01 0x00400000 /* - " - */ +#define SIUMCR_LBPC10 0x00800000 /* - " - */ +#define SIUMCR_LBPC11 0x00c00000 /* - " - */ +#define SIUMCR_APPC00 0x00000000 /* Address Parity Pins Configuration*/ +#define SIUMCR_APPC01 0x00100000 /* - " - */ +#define SIUMCR_APPC10 0x00200000 /* - " - */ +#define SIUMCR_APPC11 0x00300000 /* - " - */ +#define SIUMCR_CS10PC00 0x00000000 /* CS10 Pin Configuration */ +#define SIUMCR_CS10PC01 0x00040000 /* - " - */ +#define SIUMCR_CS10PC10 0x00080000 /* - " - */ +#define SIUMCR_CS10PC11 0x000c0000 /* - " - */ +#define SIUMCR_BCTLC00 0x00000000 /* Buffer Control Configuration */ +#define SIUMCR_BCTLC01 0x00010000 /* - " - */ +#define SIUMCR_BCTLC10 0x00020000 /* - " - */ +#define SIUMCR_BCTLC11 0x00030000 /* - " - */ +#define SIUMCR_MMR00 0x00000000 /* Mask Masters Requests */ +#define SIUMCR_MMR01 0x00004000 /* - " - */ +#define SIUMCR_MMR10 0x00008000 /* - " - */ +#define SIUMCR_MMR11 0x0000c000 /* - " - */ +#define SIUMCR_LPBSE 0x00002000 /* LocalBus Parity Byte Select Enable*/ + +/*----------------------------------------------------------------------- + * SCCR - System Clock Control Register 9-8 +*/ +#define SCCR_PCI_MODE 0x00000100 /* PCI Mode */ +#define SCCR_PCI_MODCK 0x00000080 /* Value of PCI_MODCK pin */ +#define SCCR_PCIDF_MSK 0x00000078 /* PCI division factor */ +#define SCCR_PCIDF_SHIFT 3 + + #endif /* __CPM2__ */ #endif /* __KERNEL__ */ diff --git a/include/asm-ppc/m8260_pci.h b/include/asm-ppc/m8260_pci.h index 163a6b91d5b2..bf9e05dd54b5 100644 --- a/include/asm-ppc/m8260_pci.h +++ b/include/asm-ppc/m8260_pci.h @@ -19,6 +19,7 @@ * Define the vendor/device ID for the MPC8265. */ #define PCI_DEVICE_ID_MPC8265 ((0x18C0 << 16) | PCI_VENDOR_ID_MOTOROLA) +#define PCI_DEVICE_ID_MPC8272 ((0x18C1 << 16) | PCI_VENDOR_ID_MOTOROLA) #define M8265_PCIBR0 0x101ac #define M8265_PCIBR1 0x101b0 diff --git a/include/asm-ppc/mpc8260.h b/include/asm-ppc/mpc8260.h index d820894e5991..89eb8a2ac693 100644 --- a/include/asm-ppc/mpc8260.h +++ b/include/asm-ppc/mpc8260.h @@ -41,7 +41,7 @@ #endif #ifdef CONFIG_PCI_8260 -#include +#include #endif /* Make sure the memory translation stuff is there if PCI not used. -- cgit v1.2.3 From d8a33ac435c43a1a404b2ec560ef1d1536710c36 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Sun, 29 May 2005 14:13:47 -0700 Subject: [BRIDGE]: features change notification Resend of earlier patch (no changes) from Catalin used to provide device feature change notification. Signed-off-by: Catalin BOIE Acked-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/netdevice.h | 1 + include/linux/notifier.h | 1 + net/core/dev.c | 12 ++++++++++++ net/core/ethtool.c | 8 +++++++- 4 files changed, 21 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index b25bd02720d3..d8c65ecef9d9 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -913,6 +913,7 @@ extern void dev_mc_discard(struct net_device *dev); extern void dev_set_promiscuity(struct net_device *dev, int inc); extern void dev_set_allmulti(struct net_device *dev, int inc); extern void netdev_state_change(struct net_device *dev); +extern void netdev_features_change(struct net_device *dev); /* Load a device via the kmod */ extern void dev_load(const char *name); extern void dev_mcast_init(void); diff --git a/include/linux/notifier.h b/include/linux/notifier.h index 9303a003e9ab..5937dd6053c3 100644 --- a/include/linux/notifier.h +++ b/include/linux/notifier.h @@ -56,6 +56,7 @@ extern int notifier_call_chain(struct notifier_block **n, unsigned long val, voi #define NETDEV_CHANGEADDR 0x0008 #define NETDEV_GOING_DOWN 0x0009 #define NETDEV_CHANGENAME 0x000A +#define NETDEV_FEAT_CHANGE 0x000B #define SYS_DOWN 0x0001 /* Notify of system down */ #define SYS_RESTART SYS_DOWN diff --git a/net/core/dev.c b/net/core/dev.c index d4d9e2680adb..f15a3ffff635 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -760,6 +760,18 @@ int dev_change_name(struct net_device *dev, char *newname) return err; } +/** + * netdev_features_change - device changes fatures + * @dev: device to cause notification + * + * Called to indicate a device has changed features. + */ +void netdev_features_change(struct net_device *dev) +{ + notifier_call_chain(&netdev_chain, NETDEV_FEAT_CHANGE, dev); +} +EXPORT_SYMBOL(netdev_features_change); + /** * netdev_state_change - device changes state * @dev: device to cause notification diff --git a/net/core/ethtool.c b/net/core/ethtool.c index f05fde97c43d..252bfc6f03f4 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -682,6 +682,7 @@ int dev_ethtool(struct ifreq *ifr) void __user *useraddr = ifr->ifr_data; u32 ethcmd; int rc; + int old_features; /* * XXX: This can be pushed down into the ethtool_* handlers that @@ -703,6 +704,8 @@ int dev_ethtool(struct ifreq *ifr) if ((rc = dev->ethtool_ops->begin(dev)) < 0) return rc; + old_features = dev->features; + switch (ethcmd) { case ETHTOOL_GSET: rc = ethtool_get_settings(dev, useraddr); @@ -712,7 +715,6 @@ int dev_ethtool(struct ifreq *ifr) break; case ETHTOOL_GDRVINFO: rc = ethtool_get_drvinfo(dev, useraddr); - break; case ETHTOOL_GREGS: rc = ethtool_get_regs(dev, useraddr); @@ -801,6 +803,10 @@ int dev_ethtool(struct ifreq *ifr) if(dev->ethtool_ops->complete) dev->ethtool_ops->complete(dev); + + if (old_features != dev->features) + netdev_features_change(dev); + return rc; ioctl: -- cgit v1.2.3 From 81e8157583c559c27aac75c708d40a35f563d734 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Sun, 29 May 2005 14:14:35 -0700 Subject: [BRIDGE]: make dev->features unsigned The features field in netdevice is really a bitmask, and bitmask's should be unsigned. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- net/core/ethtool.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index d8c65ecef9d9..470af8c1a4a0 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -401,7 +401,7 @@ struct net_device } reg_state; /* Net device features */ - int features; + unsigned long features; #define NETIF_F_SG 1 /* Scatter/gather IO. */ #define NETIF_F_IP_CSUM 2 /* Can checksum only TCP/UDP over IPv4. */ #define NETIF_F_NO_CSUM 4 /* Does not require checksum. F.e. loopack. */ diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 252bfc6f03f4..2a56a5218363 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -682,7 +682,7 @@ int dev_ethtool(struct ifreq *ifr) void __user *useraddr = ifr->ifr_data; u32 ethcmd; int rc; - int old_features; + unsigned long old_features; /* * XXX: This can be pushed down into the ethtool_* handlers that -- cgit v1.2.3 From 8f937c6099858eee15fae14009dcbd05177fa91d Mon Sep 17 00:00:00 2001 From: Harald Welte Date: Sun, 29 May 2005 20:23:46 -0700 Subject: [IPV4]: Primary and secondary addresses Add an option to make secondary IP addresses get promoted when primary IP addresses are removed from the device. It defaults to off to preserve existing behavior. Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- include/linux/inetdevice.h | 2 ++ include/linux/sysctl.h | 1 + net/ipv4/devinet.c | 34 +++++++++++++++++++++++++++++----- 3 files changed, 32 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h index 6fafb27877a7..7e1e15f934f3 100644 --- a/include/linux/inetdevice.h +++ b/include/linux/inetdevice.h @@ -29,6 +29,7 @@ struct ipv4_devconf int no_xfrm; int no_policy; int force_igmp_version; + int promote_secondaries; void *sysctl; }; @@ -71,6 +72,7 @@ struct in_device #define IN_DEV_SEC_REDIRECTS(in_dev) (ipv4_devconf.secure_redirects || (in_dev)->cnf.secure_redirects) #define IN_DEV_IDTAG(in_dev) ((in_dev)->cnf.tag) #define IN_DEV_MEDIUM_ID(in_dev) ((in_dev)->cnf.medium_id) +#define IN_DEV_PROMOTE_SECONDARIES(in_dev) (ipv4_devconf.promote_secondaries || (in_dev)->cnf.promote_secondaries) #define IN_DEV_RX_REDIRECTS(in_dev) \ ((IN_DEV_FORWARD(in_dev) && \ diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 772998147e3e..23032d9d6071 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -399,6 +399,7 @@ enum NET_IPV4_CONF_FORCE_IGMP_VERSION=17, NET_IPV4_CONF_ARP_ANNOUNCE=18, NET_IPV4_CONF_ARP_IGNORE=19, + NET_IPV4_CONF_PROMOTE_SECONDARIES=20, __NET_IPV4_CONF_MAX }; diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 3cc96730c4ed..478a30179a52 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -233,11 +233,14 @@ int inet_addr_onlink(struct in_device *in_dev, u32 a, u32 b) static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, int destroy) { + struct in_ifaddr *promote = NULL; struct in_ifaddr *ifa1 = *ifap; ASSERT_RTNL(); - /* 1. Deleting primary ifaddr forces deletion all secondaries */ + /* 1. Deleting primary ifaddr forces deletion all secondaries + * unless alias promotion is set + **/ if (!(ifa1->ifa_flags & IFA_F_SECONDARY)) { struct in_ifaddr *ifa; @@ -251,11 +254,16 @@ static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, continue; } - *ifap1 = ifa->ifa_next; + if (!IN_DEV_PROMOTE_SECONDARIES(in_dev)) { + *ifap1 = ifa->ifa_next; - rtmsg_ifa(RTM_DELADDR, ifa); - notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa); - inet_free_ifa(ifa); + rtmsg_ifa(RTM_DELADDR, ifa); + notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa); + inet_free_ifa(ifa); + } else { + promote = ifa; + break; + } } } @@ -281,6 +289,13 @@ static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, if (!in_dev->ifa_list) inetdev_destroy(in_dev); } + + if (promote && IN_DEV_PROMOTE_SECONDARIES(in_dev)) { + /* not sure if we should send a delete notify first? */ + promote->ifa_flags &= ~IFA_F_SECONDARY; + rtmsg_ifa(RTM_NEWADDR, promote); + notifier_call_chain(&inetaddr_chain, NETDEV_UP, promote); + } } static int inet_insert_ifa(struct in_ifaddr *ifa) @@ -1384,6 +1399,15 @@ static struct devinet_sysctl_table { .proc_handler = &ipv4_doint_and_flush, .strategy = &ipv4_doint_and_flush_strategy, }, + { + .ctl_name = NET_IPV4_CONF_PROMOTE_SECONDARIES, + .procname = "promote_secondaries", + .data = &ipv4_devconf.promote_secondaries, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &ipv4_doint_and_flush, + .strategy = &ipv4_doint_and_flush_strategy, + }, }, .devinet_dev = { { -- cgit v1.2.3 From 79165121bc09c209451487d977df910c4ff6fc94 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Sun, 29 May 2005 20:24:30 -0700 Subject: [NET]: Add is_multicast_ether_addr() in include/linux/etherdevice.h This patch adds is_multicast_ether_addr() to go along with is_valid_ether_addr() and friends. It then changes is_valid_ether_addr() to use the new macro, and fixes up the comment on that function to move implementation details out of the API doco. Signed-off-by: Michael Ellerman Signed-off-by: David S. Miller --- include/linux/etherdevice.h | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index 220748b7abea..a1478258d002 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -55,19 +55,33 @@ static inline int is_zero_ether_addr(const u8 *addr) return !(addr[0] | addr[1] | addr[2] | addr[3] | addr[4] | addr[5]); } +/** + * is_multicast_ether_addr - Determine if the given Ethernet address is a + * multicast address. + * + * @addr: Pointer to a six-byte array containing the Ethernet address + * + * Return true if the address is a multicast address. + */ +static inline int is_multicast_ether_addr(const u8 *addr) +{ + return addr[0] & 0x01; +} + /** * is_valid_ether_addr - Determine if the given Ethernet address is valid * @addr: Pointer to a six-byte array containing the Ethernet address * * Check that the Ethernet address (MAC) is not 00:00:00:00:00:00, is not - * a multicast address, and is not FF:FF:FF:FF:FF:FF. The multicast - * and FF:FF:... tests are combined into the single test "!(addr[0]&1)". + * a multicast address, and is not FF:FF:FF:FF:FF:FF. * * Return true if the address is valid. */ static inline int is_valid_ether_addr(const u8 *addr) { - return !(addr[0]&1) && !is_zero_ether_addr(addr); + /* FF:FF:FF:FF:FF:FF is a multicast address so we don't need to + * explicitly check for it here. */ + return !is_multicast_ether_addr(addr) && !is_zero_ether_addr(addr); } /** @@ -83,6 +97,6 @@ static inline void random_ether_addr(u8 *addr) addr [0] &= 0xfe; /* clear multicast bit */ addr [0] |= 0x02; /* set local assignment bit (IEEE802) */ } -#endif +#endif /* __KERNEL__ */ #endif /* _LINUX_ETHERDEVICE_H */ -- cgit v1.2.3 From 37e20a66db02eff9adbeee043af990cca85d0034 Mon Sep 17 00:00:00 2001 From: "Pravin B. Shelar" Date: Sun, 29 May 2005 20:26:44 -0700 Subject: [IPV4]: Kill MULTIPATHHOLDROUTE flag. It cannot work properly, so just ignore it in drr and rr multipath algorithms just like the random multipath algorithm does. Suggested by Herbert Xu. Signed-off by: Pravin B. Shelar Signed-off-by: David S. Miller --- include/net/route.h | 3 --- net/ipv4/multipath_drr.c | 18 +----------------- net/ipv4/multipath_rr.c | 20 -------------------- 3 files changed, 1 insertion(+), 40 deletions(-) (limited to 'include') diff --git a/include/net/route.h b/include/net/route.h index efe92b239ef1..d34ca8fc6756 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -181,9 +181,6 @@ static inline int ip_route_newports(struct rtable **rp, u16 sport, u16 dport, memcpy(&fl, &(*rp)->fl, sizeof(fl)); fl.fl_ip_sport = sport; fl.fl_ip_dport = dport; -#if defined(CONFIG_IP_ROUTE_MULTIPATH_CACHED) - fl.flags |= FLOWI_FLAG_MULTIPATHOLDROUTE; -#endif ip_rt_put(*rp); *rp = NULL; return ip_route_output_flow(rp, &fl, sk, 0); diff --git a/net/ipv4/multipath_drr.c b/net/ipv4/multipath_drr.c index 9349686131fc..cf2e6bcf7973 100644 --- a/net/ipv4/multipath_drr.c +++ b/net/ipv4/multipath_drr.c @@ -57,7 +57,6 @@ struct multipath_device { static struct multipath_device state[MULTIPATH_MAX_DEVICECANDIDATES]; static DEFINE_SPINLOCK(state_lock); -static struct rtable *last_selection = NULL; static int inline __multipath_findslot(void) { @@ -111,11 +110,6 @@ struct notifier_block drr_dev_notifier = { .notifier_call = drr_dev_event, }; -static void drr_remove(struct rtable *rt) -{ - if (last_selection == rt) - last_selection = NULL; -} static void drr_safe_inc(atomic_t *usecount) { @@ -144,14 +138,6 @@ static void drr_select_route(const struct flowi *flp, int devidx = -1; int cur_min_devidx = -1; - /* if necessary and possible utilize the old alternative */ - if ((flp->flags & FLOWI_FLAG_MULTIPATHOLDROUTE) != 0 && - last_selection != NULL) { - result = last_selection; - *rp = result; - return; - } - /* 1. make sure all alt. nexthops have the same GC related data */ /* 2. determine the new candidate to be returned */ result = NULL; @@ -229,12 +215,10 @@ static void drr_select_route(const struct flowi *flp, } *rp = result; - last_selection = result; } static struct ip_mp_alg_ops drr_ops = { .mp_alg_select_route = drr_select_route, - .mp_alg_remove = drr_remove, }; static int __init drr_init(void) @@ -244,7 +228,7 @@ static int __init drr_init(void) if (err) return err; - err = multipath_alg_register(&drr_ops, IP_MP_ALG_RR); + err = multipath_alg_register(&drr_ops, IP_MP_ALG_DRR); if (err) goto fail; diff --git a/net/ipv4/multipath_rr.c b/net/ipv4/multipath_rr.c index 554a82568160..061b6b253982 100644 --- a/net/ipv4/multipath_rr.c +++ b/net/ipv4/multipath_rr.c @@ -47,29 +47,12 @@ #include #include -#define MULTIPATH_MAX_CANDIDATES 40 - -static struct rtable* last_used = NULL; - -static void rr_remove(struct rtable *rt) -{ - if (last_used == rt) - last_used = NULL; -} - static void rr_select_route(const struct flowi *flp, struct rtable *first, struct rtable **rp) { struct rtable *nh, *result, *min_use_cand = NULL; int min_use = -1; - /* if necessary and possible utilize the old alternative */ - if ((flp->flags & FLOWI_FLAG_MULTIPATHOLDROUTE) != 0 && - last_used != NULL) { - result = last_used; - goto out; - } - /* 1. make sure all alt. nexthops have the same GC related data * 2. determine the new candidate to be returned */ @@ -90,15 +73,12 @@ static void rr_select_route(const struct flowi *flp, if (!result) result = first; -out: - last_used = result; result->u.dst.__use++; *rp = result; } static struct ip_mp_alg_ops rr_ops = { .mp_alg_select_route = rr_select_route, - .mp_alg_remove = rr_remove, }; static int __init rr_init(void) -- cgit v1.2.3 From 69f6a0fafcdf0bfe85af182695d6d38f80f9d549 Mon Sep 17 00:00:00 2001 From: Jon Mason Date: Sun, 29 May 2005 20:27:24 -0700 Subject: [NET]: Add ethtool support for NETIF_F_HW_CSUM. Signed-off-by: Jon Mason Signed-off-by: David S. Miller --- include/linux/ethtool.h | 1 + net/core/ethtool.c | 12 +++++++++++- 2 files changed, 12 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index c85b210490ea..a0ab26aab450 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -256,6 +256,7 @@ struct net_device; u32 ethtool_op_get_link(struct net_device *dev); u32 ethtool_op_get_tx_csum(struct net_device *dev); int ethtool_op_set_tx_csum(struct net_device *dev, u32 data); +int ethtool_op_set_tx_hw_csum(struct net_device *dev, u32 data); u32 ethtool_op_get_sg(struct net_device *dev); int ethtool_op_set_sg(struct net_device *dev, u32 data); u32 ethtool_op_get_tso(struct net_device *dev); diff --git a/net/core/ethtool.c b/net/core/ethtool.c index 2a56a5218363..8ec484894d68 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -29,7 +29,7 @@ u32 ethtool_op_get_link(struct net_device *dev) u32 ethtool_op_get_tx_csum(struct net_device *dev) { - return (dev->features & NETIF_F_IP_CSUM) != 0; + return (dev->features & (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM)) != 0; } int ethtool_op_set_tx_csum(struct net_device *dev, u32 data) @@ -42,6 +42,15 @@ int ethtool_op_set_tx_csum(struct net_device *dev, u32 data) return 0; } +int ethtool_op_set_tx_hw_csum(struct net_device *dev, u32 data) +{ + if (data) + dev->features |= NETIF_F_HW_CSUM; + else + dev->features &= ~NETIF_F_HW_CSUM; + + return 0; +} u32 ethtool_op_get_sg(struct net_device *dev) { return (dev->features & NETIF_F_SG) != 0; @@ -823,3 +832,4 @@ EXPORT_SYMBOL(ethtool_op_get_tx_csum); EXPORT_SYMBOL(ethtool_op_set_sg); EXPORT_SYMBOL(ethtool_op_set_tso); EXPORT_SYMBOL(ethtool_op_set_tx_csum); +EXPORT_SYMBOL(ethtool_op_set_tx_hw_csum); -- cgit v1.2.3 From af00f9811e0ccbd3db84ddc4cffb0da942653393 Mon Sep 17 00:00:00 2001 From: Andy Currid Date: Mon, 23 May 2005 08:55:45 -0700 Subject: [PATCH] PCI: amd74xx patch for new NVIDIA device IDs Here's the 2.6 amd74xx patch for NVIDIA MCP51. Signed-off-by: Andy Currid Signed-off-by: Greg Kroah-Hartman --- drivers/ide/pci/amd74xx.c | 3 +++ include/linux/pci_ids.h | 6 ++++++ 2 files changed, 9 insertions(+) (limited to 'include') diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c index 47225e324356..4e0f13d1d060 100644 --- a/drivers/ide/pci/amd74xx.c +++ b/drivers/ide/pci/amd74xx.c @@ -72,6 +72,7 @@ static struct amd_ide_chip { { PCI_DEVICE_ID_NVIDIA_NFORCE3S_SATA2, 0x50, AMD_UDMA_133 }, { PCI_DEVICE_ID_NVIDIA_NFORCE_CK804_IDE, 0x50, AMD_UDMA_133 }, { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP04_IDE, 0x50, AMD_UDMA_133 }, + { PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE, 0x50, AMD_UDMA_133 }, { 0 } }; @@ -487,6 +488,7 @@ static ide_pci_device_t amd74xx_chipsets[] __devinitdata = { /* 12 */ DECLARE_NV_DEV("NFORCE3-250-SATA2"), /* 13 */ DECLARE_NV_DEV("NFORCE-CK804"), /* 14 */ DECLARE_NV_DEV("NFORCE-MCP04"), + /* 15 */ DECLARE_NV_DEV("NFORCE-MCP51"), }; static int __devinit amd74xx_probe(struct pci_dev *dev, const struct pci_device_id *id) @@ -521,6 +523,7 @@ static struct pci_device_id amd74xx_pci_tbl[] = { #endif { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_CK804_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 13 }, { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP04_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 14 }, + { PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 15 }, { 0, }, }; MODULE_DEVICE_TABLE(pci, amd74xx_pci_tbl); diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 7ccbc2e4272c..b0d6134e1ee6 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1230,6 +1230,12 @@ #define PCI_DEVICE_ID_NVIDIA_QUADRO4_900XGL 0x0258 #define PCI_DEVICE_ID_NVIDIA_QUADRO4_750XGL 0x0259 #define PCI_DEVICE_ID_NVIDIA_QUADRO4_700XGL 0x025B +#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE 0x0265 +#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_SATA 0x0266 +#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_SATA2 0x0267 +#define PCI_DEVICE_ID_NVIDIA_NVENET_12 0x0268 +#define PCI_DEVICE_ID_NVIDIA_NVENET_13 0x0269 +#define PCI_DEVICE_ID_NVIDIA_MCP51_AUDIO 0x026B #define PCI_DEVICE_ID_NVIDIA_GEFORCE4_TI_4800 0x0280 #define PCI_DEVICE_ID_NVIDIA_GEFORCE4_TI_4800_8X 0x0281 #define PCI_DEVICE_ID_NVIDIA_GEFORCE4_TI_4800SE 0x0282 -- cgit v1.2.3 From 7c963ad1d113790a8c723a178988b675868f3abe Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 31 May 2005 16:57:59 -0700 Subject: [SPARC64]: Fix streaming buffer flushing on PCI and SBUS. Firstly, if the direction is TODEVICE, then dirty data in the streaming cache is impossible so we can elide the flush-flag synchronization in that case. Next, the context allocator is broken. It is highly likely that contexts get used multiple times for different dma mappings, which confuses the strbuf flushing code and makes it run inefficiently. Signed-off-by: David S. Miller --- arch/sparc64/kernel/pci_iommu.c | 90 +++++++++++++++++++++++++++++++--------- arch/sparc64/kernel/pci_psycho.c | 2 +- arch/sparc64/kernel/pci_sabre.c | 2 +- arch/sparc64/kernel/pci_schizo.c | 2 +- arch/sparc64/kernel/sbus.c | 20 ++++++--- include/asm-sparc64/iommu.h | 2 + include/asm-sparc64/pbm.h | 8 ++-- 7 files changed, 94 insertions(+), 32 deletions(-) (limited to 'include') diff --git a/arch/sparc64/kernel/pci_iommu.c b/arch/sparc64/kernel/pci_iommu.c index 33ca56c90da2..1807876f8c36 100644 --- a/arch/sparc64/kernel/pci_iommu.c +++ b/arch/sparc64/kernel/pci_iommu.c @@ -196,6 +196,34 @@ static iopte_t *alloc_consistent_cluster(struct pci_iommu *iommu, unsigned long return NULL; } +static int iommu_alloc_ctx(struct pci_iommu *iommu) +{ + int lowest = iommu->ctx_lowest_free; + int sz = IOMMU_NUM_CTXS - lowest; + int n = find_next_zero_bit(iommu->ctx_bitmap, sz, lowest); + + if (unlikely(n == sz)) { + n = find_next_zero_bit(iommu->ctx_bitmap, lowest, 1); + if (unlikely(n == lowest)) { + printk(KERN_WARNING "IOMMU: Ran out of contexts.\n"); + n = 0; + } + } + if (n) + __set_bit(n, iommu->ctx_bitmap); + + return n; +} + +static inline void iommu_free_ctx(struct pci_iommu *iommu, int ctx) +{ + if (likely(ctx)) { + __clear_bit(ctx, iommu->ctx_bitmap); + if (ctx < iommu->ctx_lowest_free) + iommu->ctx_lowest_free = ctx; + } +} + /* Allocate and map kernel buffer of size SIZE using consistent mode * DMA for PCI device PDEV. Return non-NULL cpu-side address if * successful and set *DMA_ADDRP to the PCI side dma address. @@ -236,7 +264,7 @@ void *pci_alloc_consistent(struct pci_dev *pdev, size_t size, dma_addr_t *dma_ad npages = size >> IO_PAGE_SHIFT; ctx = 0; if (iommu->iommu_ctxflush) - ctx = iommu->iommu_cur_ctx++; + ctx = iommu_alloc_ctx(iommu); first_page = __pa(first_page); while (npages--) { iopte_val(*iopte) = (IOPTE_CONSISTENT(ctx) | @@ -317,6 +345,8 @@ void pci_free_consistent(struct pci_dev *pdev, size_t size, void *cpu, dma_addr_ } } + iommu_free_ctx(iommu, ctx); + spin_unlock_irqrestore(&iommu->lock, flags); order = get_order(size); @@ -360,7 +390,7 @@ dma_addr_t pci_map_single(struct pci_dev *pdev, void *ptr, size_t sz, int direct base_paddr = __pa(oaddr & IO_PAGE_MASK); ctx = 0; if (iommu->iommu_ctxflush) - ctx = iommu->iommu_cur_ctx++; + ctx = iommu_alloc_ctx(iommu); if (strbuf->strbuf_enabled) iopte_protection = IOPTE_STREAMING(ctx); else @@ -380,39 +410,55 @@ bad: return PCI_DMA_ERROR_CODE; } -static void pci_strbuf_flush(struct pci_strbuf *strbuf, struct pci_iommu *iommu, u32 vaddr, unsigned long ctx, unsigned long npages) +static void pci_strbuf_flush(struct pci_strbuf *strbuf, struct pci_iommu *iommu, u32 vaddr, unsigned long ctx, unsigned long npages, int direction) { int limit; - PCI_STC_FLUSHFLAG_INIT(strbuf); if (strbuf->strbuf_ctxflush && iommu->iommu_ctxflush) { unsigned long matchreg, flushreg; + u64 val; flushreg = strbuf->strbuf_ctxflush; matchreg = PCI_STC_CTXMATCH_ADDR(strbuf, ctx); - limit = 100000; + if (pci_iommu_read(matchreg) == 0) + goto do_flush_sync; + pci_iommu_write(flushreg, ctx); - for(;;) { - if (((long)pci_iommu_read(matchreg)) >= 0L) - break; - limit--; - if (!limit) - break; - udelay(1); + if ((val = pci_iommu_read(matchreg)) == 0) + goto do_flush_sync; + + val &= 0xffff; + while (val) { + if (val & 0x1) + pci_iommu_write(flushreg, ctx); + val >>= 1; } - if (!limit) + val = pci_iommu_read(matchreg); + if (unlikely(val)) { printk(KERN_WARNING "pci_strbuf_flush: ctx flush " - "timeout vaddr[%08x] ctx[%lx]\n", - vaddr, ctx); + "timeout matchreg[%lx] ctx[%lx]\n", + val, ctx); + goto do_page_flush; + } } else { unsigned long i; + do_page_flush: for (i = 0; i < npages; i++, vaddr += IO_PAGE_SIZE) pci_iommu_write(strbuf->strbuf_pflush, vaddr); } +do_flush_sync: + /* If the device could not have possibly put dirty data into + * the streaming cache, no flush-flag synchronization needs + * to be performed. + */ + if (direction == PCI_DMA_TODEVICE) + return; + + PCI_STC_FLUSHFLAG_INIT(strbuf); pci_iommu_write(strbuf->strbuf_fsync, strbuf->strbuf_flushflag_pa); (void) pci_iommu_read(iommu->write_complete_reg); @@ -466,7 +512,7 @@ void pci_unmap_single(struct pci_dev *pdev, dma_addr_t bus_addr, size_t sz, int /* Step 1: Kick data out of streaming buffers if necessary. */ if (strbuf->strbuf_enabled) - pci_strbuf_flush(strbuf, iommu, bus_addr, ctx, npages); + pci_strbuf_flush(strbuf, iommu, bus_addr, ctx, npages, direction); /* Step 2: Clear out first TSB entry. */ iopte_make_dummy(iommu, base); @@ -474,6 +520,8 @@ void pci_unmap_single(struct pci_dev *pdev, dma_addr_t bus_addr, size_t sz, int free_streaming_cluster(iommu, bus_addr - iommu->page_table_map_base, npages, ctx); + iommu_free_ctx(iommu, ctx); + spin_unlock_irqrestore(&iommu->lock, flags); } @@ -613,7 +661,7 @@ int pci_map_sg(struct pci_dev *pdev, struct scatterlist *sglist, int nelems, int /* Step 4: Choose a context if necessary. */ ctx = 0; if (iommu->iommu_ctxflush) - ctx = iommu->iommu_cur_ctx++; + ctx = iommu_alloc_ctx(iommu); /* Step 5: Create the mappings. */ if (strbuf->strbuf_enabled) @@ -678,7 +726,7 @@ void pci_unmap_sg(struct pci_dev *pdev, struct scatterlist *sglist, int nelems, /* Step 1: Kick data out of streaming buffers if necessary. */ if (strbuf->strbuf_enabled) - pci_strbuf_flush(strbuf, iommu, bus_addr, ctx, npages); + pci_strbuf_flush(strbuf, iommu, bus_addr, ctx, npages, direction); /* Step 2: Clear out first TSB entry. */ iopte_make_dummy(iommu, base); @@ -686,6 +734,8 @@ void pci_unmap_sg(struct pci_dev *pdev, struct scatterlist *sglist, int nelems, free_streaming_cluster(iommu, bus_addr - iommu->page_table_map_base, npages, ctx); + iommu_free_ctx(iommu, ctx); + spin_unlock_irqrestore(&iommu->lock, flags); } @@ -724,7 +774,7 @@ void pci_dma_sync_single_for_cpu(struct pci_dev *pdev, dma_addr_t bus_addr, size } /* Step 2: Kick data out of streaming buffers. */ - pci_strbuf_flush(strbuf, iommu, bus_addr, ctx, npages); + pci_strbuf_flush(strbuf, iommu, bus_addr, ctx, npages, direction); spin_unlock_irqrestore(&iommu->lock, flags); } @@ -768,7 +818,7 @@ void pci_dma_sync_sg_for_cpu(struct pci_dev *pdev, struct scatterlist *sglist, i i--; npages = (IO_PAGE_ALIGN(sglist[i].dma_address + sglist[i].dma_length) - bus_addr) >> IO_PAGE_SHIFT; - pci_strbuf_flush(strbuf, iommu, bus_addr, ctx, npages); + pci_strbuf_flush(strbuf, iommu, bus_addr, ctx, npages, direction); spin_unlock_irqrestore(&iommu->lock, flags); } diff --git a/arch/sparc64/kernel/pci_psycho.c b/arch/sparc64/kernel/pci_psycho.c index 3567fa879e1f..534320ef0db2 100644 --- a/arch/sparc64/kernel/pci_psycho.c +++ b/arch/sparc64/kernel/pci_psycho.c @@ -1212,7 +1212,7 @@ static void __init psycho_iommu_init(struct pci_controller_info *p) /* Setup initial software IOMMU state. */ spin_lock_init(&iommu->lock); - iommu->iommu_cur_ctx = 0; + iommu->ctx_lowest_free = 1; /* Register addresses. */ iommu->iommu_control = p->pbm_A.controller_regs + PSYCHO_IOMMU_CONTROL; diff --git a/arch/sparc64/kernel/pci_sabre.c b/arch/sparc64/kernel/pci_sabre.c index 5525d1ec4af8..53d333b4a4e8 100644 --- a/arch/sparc64/kernel/pci_sabre.c +++ b/arch/sparc64/kernel/pci_sabre.c @@ -1265,7 +1265,7 @@ static void __init sabre_iommu_init(struct pci_controller_info *p, /* Setup initial software IOMMU state. */ spin_lock_init(&iommu->lock); - iommu->iommu_cur_ctx = 0; + iommu->ctx_lowest_free = 1; /* Register addresses. */ iommu->iommu_control = p->pbm_A.controller_regs + SABRE_IOMMU_CONTROL; diff --git a/arch/sparc64/kernel/pci_schizo.c b/arch/sparc64/kernel/pci_schizo.c index e93fcadc3722..5753175b94e6 100644 --- a/arch/sparc64/kernel/pci_schizo.c +++ b/arch/sparc64/kernel/pci_schizo.c @@ -1753,7 +1753,7 @@ static void schizo_pbm_iommu_init(struct pci_pbm_info *pbm) /* Setup initial software IOMMU state. */ spin_lock_init(&iommu->lock); - iommu->iommu_cur_ctx = 0; + iommu->ctx_lowest_free = 1; /* Register addresses, SCHIZO has iommu ctx flushing. */ iommu->iommu_control = pbm->pbm_regs + SCHIZO_IOMMU_CONTROL; diff --git a/arch/sparc64/kernel/sbus.c b/arch/sparc64/kernel/sbus.c index 76ea6455433f..89f5e019f24c 100644 --- a/arch/sparc64/kernel/sbus.c +++ b/arch/sparc64/kernel/sbus.c @@ -117,17 +117,25 @@ static void iommu_flush(struct sbus_iommu *iommu, u32 base, unsigned long npages #define STRBUF_TAG_VALID 0x02UL -static void sbus_strbuf_flush(struct sbus_iommu *iommu, u32 base, unsigned long npages) +static void sbus_strbuf_flush(struct sbus_iommu *iommu, u32 base, unsigned long npages, int direction) { unsigned long n; int limit; - iommu->strbuf_flushflag = 0UL; n = npages; while (n--) upa_writeq(base + (n << IO_PAGE_SHIFT), iommu->strbuf_regs + STRBUF_PFLUSH); + /* If the device could not have possibly put dirty data into + * the streaming cache, no flush-flag synchronization needs + * to be performed. + */ + if (direction == SBUS_DMA_TODEVICE) + return; + + iommu->strbuf_flushflag = 0UL; + /* Whoopee cushion! */ upa_writeq(__pa(&iommu->strbuf_flushflag), iommu->strbuf_regs + STRBUF_FSYNC); @@ -421,7 +429,7 @@ void sbus_unmap_single(struct sbus_dev *sdev, dma_addr_t dma_addr, size_t size, spin_lock_irqsave(&iommu->lock, flags); free_streaming_cluster(iommu, dma_base, size >> IO_PAGE_SHIFT); - sbus_strbuf_flush(iommu, dma_base, size >> IO_PAGE_SHIFT); + sbus_strbuf_flush(iommu, dma_base, size >> IO_PAGE_SHIFT, direction); spin_unlock_irqrestore(&iommu->lock, flags); } @@ -584,7 +592,7 @@ void sbus_unmap_sg(struct sbus_dev *sdev, struct scatterlist *sg, int nents, int iommu = sdev->bus->iommu; spin_lock_irqsave(&iommu->lock, flags); free_streaming_cluster(iommu, dvma_base, size >> IO_PAGE_SHIFT); - sbus_strbuf_flush(iommu, dvma_base, size >> IO_PAGE_SHIFT); + sbus_strbuf_flush(iommu, dvma_base, size >> IO_PAGE_SHIFT, direction); spin_unlock_irqrestore(&iommu->lock, flags); } @@ -596,7 +604,7 @@ void sbus_dma_sync_single_for_cpu(struct sbus_dev *sdev, dma_addr_t base, size_t size = (IO_PAGE_ALIGN(base + size) - (base & IO_PAGE_MASK)); spin_lock_irqsave(&iommu->lock, flags); - sbus_strbuf_flush(iommu, base & IO_PAGE_MASK, size >> IO_PAGE_SHIFT); + sbus_strbuf_flush(iommu, base & IO_PAGE_MASK, size >> IO_PAGE_SHIFT, direction); spin_unlock_irqrestore(&iommu->lock, flags); } @@ -620,7 +628,7 @@ void sbus_dma_sync_sg_for_cpu(struct sbus_dev *sdev, struct scatterlist *sg, int size = IO_PAGE_ALIGN(sg[i].dma_address + sg[i].dma_length) - base; spin_lock_irqsave(&iommu->lock, flags); - sbus_strbuf_flush(iommu, base, size >> IO_PAGE_SHIFT); + sbus_strbuf_flush(iommu, base, size >> IO_PAGE_SHIFT, direction); spin_unlock_irqrestore(&iommu->lock, flags); } diff --git a/include/asm-sparc64/iommu.h b/include/asm-sparc64/iommu.h index 5fd16e42a045..0de7a3da79cd 100644 --- a/include/asm-sparc64/iommu.h +++ b/include/asm-sparc64/iommu.h @@ -16,4 +16,6 @@ #define IOPTE_CACHE 0x0000000000000010UL /* Cached (in UPA E-cache) */ #define IOPTE_WRITE 0x0000000000000002UL /* Writeable */ +#define IOMMU_NUM_CTXS 4096 + #endif /* !(_SPARC_IOMMU_H) */ diff --git a/include/asm-sparc64/pbm.h b/include/asm-sparc64/pbm.h index 92999631c819..4c15610a2bac 100644 --- a/include/asm-sparc64/pbm.h +++ b/include/asm-sparc64/pbm.h @@ -15,6 +15,7 @@ #include #include #include +#include /* The abstraction used here is that there are PCI controllers, * each with one (Sabre) or two (PSYCHO/SCHIZO) PCI bus modules @@ -40,9 +41,6 @@ struct pci_iommu { */ spinlock_t lock; - /* Context allocator. */ - unsigned int iommu_cur_ctx; - /* IOMMU page table, a linear array of ioptes. */ iopte_t *page_table; /* The page table itself. */ int page_table_sz_bits; /* log2 of ow many pages does it map? */ @@ -87,6 +85,10 @@ struct pci_iommu { u16 flush; } alloc_info[PBM_NCLUSTERS]; + /* CTX allocation. */ + unsigned long ctx_lowest_free; + unsigned long ctx_bitmap[IOMMU_NUM_CTXS / (sizeof(unsigned long) * 8)]; + /* Here a PCI controller driver describes the areas of * PCI memory space where DMA to/from physical memory * are addressed. Drivers interrogate the PCI layer -- cgit v1.2.3 From c5d28fb297efaa97c4b90e36f9dff3066e7f2778 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Tue, 31 May 2005 19:03:46 -0700 Subject: [CPUFREQ] Recalibrate cpu_khz [2/2] Some cpufreq drivers (at that time, only powernow-k7) need to recalibrate the cpu_khz at runtime. Signed-off-by: Bruno Ducrot Signed-off-by: Andrew Morton Signed-off-by: Dave Jones --- arch/i386/kernel/timers/common.c | 6 ++++-- arch/i386/kernel/timers/timer_tsc.c | 20 ++++++++++++++++++++ include/asm-i386/timer.h | 1 + 3 files changed, 25 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/arch/i386/kernel/timers/common.c b/arch/i386/kernel/timers/common.c index f7f90005e22e..8e201219f525 100644 --- a/arch/i386/kernel/timers/common.c +++ b/arch/i386/kernel/timers/common.c @@ -6,6 +6,7 @@ #include #include #include +#include #include #include @@ -24,7 +25,7 @@ #define CALIBRATE_TIME (5 * 1000020/HZ) -unsigned long __init calibrate_tsc(void) +unsigned long calibrate_tsc(void) { mach_prepare_counter(); @@ -139,7 +140,7 @@ bad_calibration: #endif /* calculate cpu_khz */ -void __init init_cpu_khz(void) +void init_cpu_khz(void) { if (cpu_has_tsc) { unsigned long tsc_quotient = calibrate_tsc(); @@ -158,3 +159,4 @@ void __init init_cpu_khz(void) } } } + diff --git a/arch/i386/kernel/timers/timer_tsc.c b/arch/i386/kernel/timers/timer_tsc.c index 7926d967be00..180444d87824 100644 --- a/arch/i386/kernel/timers/timer_tsc.c +++ b/arch/i386/kernel/timers/timer_tsc.c @@ -320,6 +320,26 @@ core_initcall(cpufreq_tsc); static inline void cpufreq_delayed_get(void) { return; } #endif +int recalibrate_cpu_khz(void) +{ +#ifndef CONFIG_SMP + unsigned long cpu_khz_old = cpu_khz; + + if (cpu_has_tsc) { + init_cpu_khz(); + cpu_data[0].loops_per_jiffy = + cpufreq_scale(cpu_data[0].loops_per_jiffy, + cpu_khz_old, + cpu_khz); + return 0; + } else + return -ENODEV; +#else + return -ENODEV; +#endif +} +EXPORT_SYMBOL(recalibrate_cpu_khz); + static void mark_offset_tsc(void) { unsigned long lost,delay; diff --git a/include/asm-i386/timer.h b/include/asm-i386/timer.h index 40c54f69780e..c34709849839 100644 --- a/include/asm-i386/timer.h +++ b/include/asm-i386/timer.h @@ -53,6 +53,7 @@ extern struct init_timer_opts timer_cyclone_init; extern unsigned long calibrate_tsc(void); extern void init_cpu_khz(void); +extern int recalibrate_cpu_khz(void); #ifdef CONFIG_HPET_TIMER extern struct init_timer_opts timer_hpet_init; extern unsigned long calibrate_tsc_hpet(unsigned long *tsc_hpet_quotient_ptr); -- cgit v1.2.3 From b53cc6ead046093477ec7a3354d620337101ea5b Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Tue, 31 May 2005 19:03:47 -0700 Subject: [CPUFREQ] fix up comment in cpufreq.h Fix up comment in cpufreq.h stating transition latency should be passed in microseconds -- it was decided long ago to switch to nanoseconds. Signed-off-by: Dominik Brodowski Signed-off-by: Dave Jones --- include/linux/cpufreq.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index f21af067d015..927daa86c9b3 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -49,7 +49,7 @@ int cpufreq_unregister_notifier(struct notifier_block *nb, unsigned int list); /* Frequency values here are CPU kHz so that hardware which doesn't run * with some frequencies can complain without having to guess what per * cent / per mille means. - * Maximum transition latency is in microseconds - if it's unknown, + * Maximum transition latency is in nanoseconds - if it's unknown, * CPUFREQ_ETERNAL shall be used. */ -- cgit v1.2.3 From 5f64f73957f6cae3222f97f2599199ee562f7f3f Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 1 Jun 2005 17:07:27 +1000 Subject: [PATCH] ppc32/ppc64: cleanup /proc/device-tree This cleans up the /proc/device-tree representation of the Open Firmware device-tree on ppc and ppc64. It does the following things: - Workaround an issue in some Apple device-trees where a property may exist with the same name as a child node of the parent. We now simply "drop" the property instead of creating duplicate entries in /proc with random result... - Do not try to chop off the "@0" at the end of a node name whose unit address is 0. This is not useful, inconsistent, and the code was buggy and didn't always work anyway. - Do not create symlinks for the short name and unit address parts of a node. These were never really used, bloated the memory footprint of the device-tree with useless struct proc_dir_entry and their matching dentry and inode cache bloat. This results in smaller code, smaller memory footprint, and a more accurate view of the tree presented to userland. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Linus Torvalds --- arch/ppc64/kernel/pSeries_reconfig.c | 8 --- fs/proc/proc_devtree.c | 105 +++++++++++++---------------------- include/asm-ppc64/prom.h | 13 +---- 3 files changed, 39 insertions(+), 87 deletions(-) (limited to 'include') diff --git a/arch/ppc64/kernel/pSeries_reconfig.c b/arch/ppc64/kernel/pSeries_reconfig.c index cb5443f2e49b..dc2a69d412a2 100644 --- a/arch/ppc64/kernel/pSeries_reconfig.c +++ b/arch/ppc64/kernel/pSeries_reconfig.c @@ -47,14 +47,6 @@ static void remove_node_proc_entries(struct device_node *np) remove_proc_entry(pp->name, np->pde); pp = pp->next; } - - /* Assuming that symlinks have the same parent directory as - * np->pde. - */ - if (np->name_link) - remove_proc_entry(np->name_link->name, parent->pde); - if (np->addr_link) - remove_proc_entry(np->addr_link->name, parent->pde); if (np->pde) remove_proc_entry(np->pde->name, parent->pde); } diff --git a/fs/proc/proc_devtree.c b/fs/proc/proc_devtree.c index 67423c696c0a..6fd57f154197 100644 --- a/fs/proc/proc_devtree.c +++ b/fs/proc/proc_devtree.c @@ -12,15 +12,8 @@ #include #ifndef HAVE_ARCH_DEVTREE_FIXUPS -static inline void set_node_proc_entry(struct device_node *np, struct proc_dir_entry *de) -{ -} - -static void inline set_node_name_link(struct device_node *np, struct proc_dir_entry *de) -{ -} - -static void inline set_node_addr_link(struct device_node *np, struct proc_dir_entry *de) +static inline void set_node_proc_entry(struct device_node *np, + struct proc_dir_entry *de) { } #endif @@ -58,89 +51,67 @@ static int property_read_proc(char *page, char **start, off_t off, /* * Process a node, adding entries for its children and its properties. */ -void proc_device_tree_add_node(struct device_node *np, struct proc_dir_entry *de) +void proc_device_tree_add_node(struct device_node *np, + struct proc_dir_entry *de) { struct property *pp; struct proc_dir_entry *ent; - struct device_node *child, *sib; - const char *p, *at; - int l; - struct proc_dir_entry *list, **lastp, *al; + struct device_node *child; + struct proc_dir_entry *list = NULL, **lastp; + const char *p; set_node_proc_entry(np, de); lastp = &list; - for (pp = np->properties; pp != 0; pp = pp->next) { - /* - * Unfortunately proc_register puts each new entry - * at the beginning of the list. So we rearrange them. - */ - ent = create_proc_read_entry(pp->name, strncmp(pp->name, "security-", 9) ? - S_IRUGO : S_IRUSR, de, property_read_proc, pp); - if (ent == 0) - break; - if (!strncmp(pp->name, "security-", 9)) - ent->size = 0; /* don't leak number of password chars */ - else - ent->size = pp->length; - *lastp = ent; - lastp = &ent->next; - } - child = NULL; - while ((child = of_get_next_child(np, child))) { + for (child = NULL; (child = of_get_next_child(np, child));) { p = strrchr(child->full_name, '/'); if (!p) p = child->full_name; else ++p; - /* chop off '@0' if the name ends with that */ - l = strlen(p); - if (l > 2 && p[l-2] == '@' && p[l-1] == '0') - l -= 2; ent = proc_mkdir(p, de); if (ent == 0) break; *lastp = ent; + ent->next = NULL; lastp = &ent->next; proc_device_tree_add_node(child, ent); - - /* - * If we left the address part on the name, consider - * adding symlinks from the name and address parts. - */ - if (p[l] != 0 || (at = strchr(p, '@')) == 0) - continue; - + } + of_node_put(child); + for (pp = np->properties; pp != 0; pp = pp->next) { /* - * If this is the first node with a given name property, - * add a symlink with the name property as its name. + * Yet another Apple device-tree bogosity: on some machines, + * they have properties & nodes with the same name. Those + * properties are quite unimportant for us though, thus we + * simply "skip" them here, but we do have to check. */ - sib = NULL; - while ((sib = of_get_next_child(np, sib)) && sib != child) - if (sib->name && strcmp(sib->name, child->name) == 0) - break; - if (sib == child && strncmp(p, child->name, l) != 0) { - al = proc_symlink(child->name, de, ent->name); - if (al == 0) { - of_node_put(sib); + for (ent = list; ent != NULL; ent = ent->next) + if (!strcmp(ent->name, pp->name)) break; - } - set_node_name_link(child, al); - *lastp = al; - lastp = &al->next; + if (ent != NULL) { + printk(KERN_WARNING "device-tree: property \"%s\" name" + " conflicts with node in %s\n", pp->name, + np->full_name); + continue; } - of_node_put(sib); + /* - * Add another directory with the @address part as its name. + * Unfortunately proc_register puts each new entry + * at the beginning of the list. So we rearrange them. */ - al = proc_symlink(at, de, ent->name); - if (al == 0) + ent = create_proc_read_entry(pp->name, + strncmp(pp->name, "security-", 9) + ? S_IRUGO : S_IRUSR, de, + property_read_proc, pp); + if (ent == 0) break; - set_node_addr_link(child, al); - *lastp = al; - lastp = &al->next; + if (!strncmp(pp->name, "security-", 9)) + ent->size = 0; /* don't leak number of password chars */ + else + ent->size = pp->length; + ent->next = NULL; + *lastp = ent; + lastp = &ent->next; } - of_node_put(child); - *lastp = NULL; de->subdir = list; } diff --git a/include/asm-ppc64/prom.h b/include/asm-ppc64/prom.h index 2440a2c90ae9..04b1a84f7ca3 100644 --- a/include/asm-ppc64/prom.h +++ b/include/asm-ppc64/prom.h @@ -147,9 +147,7 @@ struct device_node { struct device_node *sibling; struct device_node *next; /* next device of same type */ struct device_node *allnext; /* next in list of all nodes */ - struct proc_dir_entry *pde; /* this node's proc directory */ - struct proc_dir_entry *name_link; /* name symlink */ - struct proc_dir_entry *addr_link; /* addr symlink */ + struct proc_dir_entry *pde; /* this node's proc directory */ struct kref kref; unsigned long _flags; }; @@ -174,15 +172,6 @@ static inline void set_node_proc_entry(struct device_node *dn, struct proc_dir_e dn->pde = de; } -static void inline set_node_name_link(struct device_node *dn, struct proc_dir_entry *de) -{ - dn->name_link = de; -} - -static void inline set_node_addr_link(struct device_node *dn, struct proc_dir_entry *de) -{ - dn->addr_link = de; -} /* OBSOLETE: Old stlye node lookup */ extern struct device_node *find_devices(const char *name); -- cgit v1.2.3 From 79f1248962cfa1e11a5610e0349bc3515687516d Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 2 Jun 2005 14:02:02 -0700 Subject: [PATCH] ppc64: cleanup SPR definitions There are a bunch of irrelevant SPR definitions in asm/processer.h. Cut them down a bit, also add a DABR_TRANSLATION define which will be used shortly. Signed-off-by: Anton Blanchard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-ppc64/processor.h | 159 +----------------------------------------- 1 file changed, 1 insertion(+), 158 deletions(-) (limited to 'include') diff --git a/include/asm-ppc64/processor.h b/include/asm-ppc64/processor.h index 0035efe2db2b..5ba6aa45e1e2 100644 --- a/include/asm-ppc64/processor.h +++ b/include/asm-ppc64/processor.h @@ -120,103 +120,18 @@ /* Special Purpose Registers (SPRNs)*/ -#define SPRN_CDBCR 0x3D7 /* Cache Debug Control Register */ #define SPRN_CTR 0x009 /* Count Register */ #define SPRN_DABR 0x3F5 /* Data Address Breakpoint Register */ -#define SPRN_DAC1 0x3F6 /* Data Address Compare 1 */ -#define SPRN_DAC2 0x3F7 /* Data Address Compare 2 */ +#define DABR_TRANSLATION (1UL << 2) #define SPRN_DAR 0x013 /* Data Address Register */ -#define SPRN_DBCR 0x3F2 /* Debug Control Regsiter */ -#define DBCR_EDM 0x80000000 -#define DBCR_IDM 0x40000000 -#define DBCR_RST(x) (((x) & 0x3) << 28) -#define DBCR_RST_NONE 0 -#define DBCR_RST_CORE 1 -#define DBCR_RST_CHIP 2 -#define DBCR_RST_SYSTEM 3 -#define DBCR_IC 0x08000000 /* Instruction Completion Debug Evnt */ -#define DBCR_BT 0x04000000 /* Branch Taken Debug Event */ -#define DBCR_EDE 0x02000000 /* Exception Debug Event */ -#define DBCR_TDE 0x01000000 /* TRAP Debug Event */ -#define DBCR_FER 0x00F80000 /* First Events Remaining Mask */ -#define DBCR_FT 0x00040000 /* Freeze Timers on Debug Event */ -#define DBCR_IA1 0x00020000 /* Instr. Addr. Compare 1 Enable */ -#define DBCR_IA2 0x00010000 /* Instr. Addr. Compare 2 Enable */ -#define DBCR_D1R 0x00008000 /* Data Addr. Compare 1 Read Enable */ -#define DBCR_D1W 0x00004000 /* Data Addr. Compare 1 Write Enable */ -#define DBCR_D1S(x) (((x) & 0x3) << 12) /* Data Adrr. Compare 1 Size */ -#define DAC_BYTE 0 -#define DAC_HALF 1 -#define DAC_WORD 2 -#define DAC_QUAD 3 -#define DBCR_D2R 0x00000800 /* Data Addr. Compare 2 Read Enable */ -#define DBCR_D2W 0x00000400 /* Data Addr. Compare 2 Write Enable */ -#define DBCR_D2S(x) (((x) & 0x3) << 8) /* Data Addr. Compare 2 Size */ -#define DBCR_SBT 0x00000040 /* Second Branch Taken Debug Event */ -#define DBCR_SED 0x00000020 /* Second Exception Debug Event */ -#define DBCR_STD 0x00000010 /* Second Trap Debug Event */ -#define DBCR_SIA 0x00000008 /* Second IAC Enable */ -#define DBCR_SDA 0x00000004 /* Second DAC Enable */ -#define DBCR_JOI 0x00000002 /* JTAG Serial Outbound Int. Enable */ -#define DBCR_JII 0x00000001 /* JTAG Serial Inbound Int. Enable */ -#define SPRN_DBCR0 0x3F2 /* Debug Control Register 0 */ -#define SPRN_DBCR1 0x3BD /* Debug Control Register 1 */ -#define SPRN_DBSR 0x3F0 /* Debug Status Register */ -#define SPRN_DCCR 0x3FA /* Data Cache Cacheability Register */ -#define DCCR_NOCACHE 0 /* Noncacheable */ -#define DCCR_CACHE 1 /* Cacheable */ -#define SPRN_DCMP 0x3D1 /* Data TLB Compare Register */ -#define SPRN_DCWR 0x3BA /* Data Cache Write-thru Register */ -#define DCWR_COPY 0 /* Copy-back */ -#define DCWR_WRITE 1 /* Write-through */ -#define SPRN_DEAR 0x3D5 /* Data Error Address Register */ #define SPRN_DEC 0x016 /* Decrement Register */ -#define SPRN_DMISS 0x3D0 /* Data TLB Miss Register */ #define SPRN_DSISR 0x012 /* Data Storage Interrupt Status Register */ #define DSISR_NOHPTE 0x40000000 /* no translation found */ #define DSISR_PROTFAULT 0x08000000 /* protection fault */ #define DSISR_ISSTORE 0x02000000 /* access was a store */ #define DSISR_DABRMATCH 0x00400000 /* hit data breakpoint */ #define DSISR_NOSEGMENT 0x00200000 /* STAB/SLB miss */ -#define SPRN_EAR 0x11A /* External Address Register */ -#define SPRN_ESR 0x3D4 /* Exception Syndrome Register */ -#define ESR_IMCP 0x80000000 /* Instr. Machine Check - Protection */ -#define ESR_IMCN 0x40000000 /* Instr. Machine Check - Non-config */ -#define ESR_IMCB 0x20000000 /* Instr. Machine Check - Bus error */ -#define ESR_IMCT 0x10000000 /* Instr. Machine Check - Timeout */ -#define ESR_PIL 0x08000000 /* Program Exception - Illegal */ -#define ESR_PPR 0x04000000 /* Program Exception - Priveleged */ -#define ESR_PTR 0x02000000 /* Program Exception - Trap */ -#define ESR_DST 0x00800000 /* Storage Exception - Data miss */ -#define ESR_DIZ 0x00400000 /* Storage Exception - Zone fault */ -#define SPRN_EVPR 0x3D6 /* Exception Vector Prefix Register */ -#define SPRN_HASH1 0x3D2 /* Primary Hash Address Register */ -#define SPRN_HASH2 0x3D3 /* Secondary Hash Address Resgister */ #define SPRN_HID0 0x3F0 /* Hardware Implementation Register 0 */ -#define HID0_EMCP (1<<31) /* Enable Machine Check pin */ -#define HID0_EBA (1<<29) /* Enable Bus Address Parity */ -#define HID0_EBD (1<<28) /* Enable Bus Data Parity */ -#define HID0_SBCLK (1<<27) -#define HID0_EICE (1<<26) -#define HID0_ECLK (1<<25) -#define HID0_PAR (1<<24) -#define HID0_DOZE (1<<23) -#define HID0_NAP (1<<22) -#define HID0_SLEEP (1<<21) -#define HID0_DPM (1<<20) -#define HID0_ICE (1<<15) /* Instruction Cache Enable */ -#define HID0_DCE (1<<14) /* Data Cache Enable */ -#define HID0_ILOCK (1<<13) /* Instruction Cache Lock */ -#define HID0_DLOCK (1<<12) /* Data Cache Lock */ -#define HID0_ICFI (1<<11) /* Instr. Cache Flash Invalidate */ -#define HID0_DCI (1<<10) /* Data Cache Invalidate */ -#define HID0_SPD (1<<9) /* Speculative disable */ -#define HID0_SGE (1<<7) /* Store Gathering Enable */ -#define HID0_SIED (1<<7) /* Serial Instr. Execution [Disable] */ -#define HID0_BTIC (1<<5) /* Branch Target Instruction Cache Enable */ -#define HID0_ABE (1<<3) /* Address Broadcast Enable */ -#define HID0_BHTE (1<<2) /* Branch History Table Enable */ -#define HID0_BTCD (1<<1) /* Branch target cache disable */ #define SPRN_MSRDORM 0x3F1 /* Hardware Implementation Register 1 */ #define SPRN_HID1 0x3F1 /* Hardware Implementation Register 1 */ #define SPRN_IABR 0x3F2 /* Instruction Address Breakpoint Register */ @@ -225,23 +140,8 @@ #define SPRN_HID5 0x3F6 /* 970 HID5 */ #define SPRN_TSC 0x3FD /* Thread switch control */ #define SPRN_TST 0x3FC /* Thread switch timeout */ -#define SPRN_IAC1 0x3F4 /* Instruction Address Compare 1 */ -#define SPRN_IAC2 0x3F5 /* Instruction Address Compare 2 */ -#define SPRN_ICCR 0x3FB /* Instruction Cache Cacheability Register */ -#define ICCR_NOCACHE 0 /* Noncacheable */ -#define ICCR_CACHE 1 /* Cacheable */ -#define SPRN_ICDBDR 0x3D3 /* Instruction Cache Debug Data Register */ -#define SPRN_ICMP 0x3D5 /* Instruction TLB Compare Register */ -#define SPRN_ICTC 0x3FB /* Instruction Cache Throttling Control Reg */ -#define SPRN_IMISS 0x3D4 /* Instruction TLB Miss Register */ -#define SPRN_IMMR 0x27E /* Internal Memory Map Register */ #define SPRN_L2CR 0x3F9 /* Level 2 Cache Control Regsiter */ #define SPRN_LR 0x008 /* Link Register */ -#define SPRN_PBL1 0x3FC /* Protection Bound Lower 1 */ -#define SPRN_PBL2 0x3FE /* Protection Bound Lower 2 */ -#define SPRN_PBU1 0x3FD /* Protection Bound Upper 1 */ -#define SPRN_PBU2 0x3FF /* Protection Bound Upper 2 */ -#define SPRN_PID 0x3B1 /* Process ID */ #define SPRN_PIR 0x3FF /* Processor Identification Register */ #define SPRN_PIT 0x3DB /* Programmable Interval Timer */ #define SPRN_PURR 0x135 /* Processor Utilization of Resources Register */ @@ -249,9 +149,6 @@ #define SPRN_RPA 0x3D6 /* Required Physical Address Register */ #define SPRN_SDA 0x3BF /* Sampled Data Address Register */ #define SPRN_SDR1 0x019 /* MMU Hash Base Register */ -#define SPRN_SGR 0x3B9 /* Storage Guarded Register */ -#define SGR_NORMAL 0 -#define SGR_GUARDED 1 #define SPRN_SIA 0x3BB /* Sampled Instruction Address Register */ #define SPRN_SPRG0 0x110 /* Special Purpose Register General 0 */ #define SPRN_SPRG1 0x111 /* Special Purpose Register General 1 */ @@ -264,49 +161,8 @@ #define SPRN_TBWL 0x11C /* Time Base Lower Register (super, W/O) */ #define SPRN_TBWU 0x11D /* Time Base Write Upper Register (super, W/O) */ #define SPRN_HIOR 0x137 /* 970 Hypervisor interrupt offset */ -#define SPRN_TCR 0x3DA /* Timer Control Register */ -#define TCR_WP(x) (((x)&0x3)<<30) /* WDT Period */ -#define WP_2_17 0 /* 2^17 clocks */ -#define WP_2_21 1 /* 2^21 clocks */ -#define WP_2_25 2 /* 2^25 clocks */ -#define WP_2_29 3 /* 2^29 clocks */ -#define TCR_WRC(x) (((x)&0x3)<<28) /* WDT Reset Control */ -#define WRC_NONE 0 /* No reset will occur */ -#define WRC_CORE 1 /* Core reset will occur */ -#define WRC_CHIP 2 /* Chip reset will occur */ -#define WRC_SYSTEM 3 /* System reset will occur */ -#define TCR_WIE 0x08000000 /* WDT Interrupt Enable */ -#define TCR_PIE 0x04000000 /* PIT Interrupt Enable */ -#define TCR_FP(x) (((x)&0x3)<<24) /* FIT Period */ -#define FP_2_9 0 /* 2^9 clocks */ -#define FP_2_13 1 /* 2^13 clocks */ -#define FP_2_17 2 /* 2^17 clocks */ -#define FP_2_21 3 /* 2^21 clocks */ -#define TCR_FIE 0x00800000 /* FIT Interrupt Enable */ -#define TCR_ARE 0x00400000 /* Auto Reload Enable */ -#define SPRN_THRM1 0x3FC /* Thermal Management Register 1 */ -#define THRM1_TIN (1<<0) -#define THRM1_TIV (1<<1) -#define THRM1_THRES (0x7f<<2) -#define THRM1_TID (1<<29) -#define THRM1_TIE (1<<30) -#define THRM1_V (1<<31) -#define SPRN_THRM2 0x3FD /* Thermal Management Register 2 */ -#define SPRN_THRM3 0x3FE /* Thermal Management Register 3 */ -#define THRM3_E (1<<31) -#define SPRN_TSR 0x3D8 /* Timer Status Register */ -#define TSR_ENW 0x80000000 /* Enable Next Watchdog */ -#define TSR_WIS 0x40000000 /* WDT Interrupt Status */ -#define TSR_WRS(x) (((x)&0x3)<<28) /* WDT Reset Status */ -#define WRS_NONE 0 /* No WDT reset occurred */ -#define WRS_CORE 1 /* WDT forced core reset */ -#define WRS_CHIP 2 /* WDT forced chip reset */ -#define WRS_SYSTEM 3 /* WDT forced system reset */ -#define TSR_PIS 0x08000000 /* PIT Interrupt Status */ -#define TSR_FIS 0x04000000 /* FIT Interrupt Status */ #define SPRN_USIA 0x3AB /* User Sampled Instruction Address Register */ #define SPRN_XER 0x001 /* Fixed Point Exception Register */ -#define SPRN_ZPR 0x3B0 /* Zone Protection Register */ #define SPRN_VRSAVE 0x100 /* Vector save */ /* Performance monitor SPRs */ @@ -352,28 +208,19 @@ #define CTR SPRN_CTR /* Counter Register */ #define DAR SPRN_DAR /* Data Address Register */ #define DABR SPRN_DABR /* Data Address Breakpoint Register */ -#define DCMP SPRN_DCMP /* Data TLB Compare Register */ #define DEC SPRN_DEC /* Decrement Register */ -#define DMISS SPRN_DMISS /* Data TLB Miss Register */ #define DSISR SPRN_DSISR /* Data Storage Interrupt Status Register */ -#define EAR SPRN_EAR /* External Address Register */ -#define HASH1 SPRN_HASH1 /* Primary Hash Address Register */ -#define HASH2 SPRN_HASH2 /* Secondary Hash Address Register */ #define HID0 SPRN_HID0 /* Hardware Implementation Register 0 */ #define MSRDORM SPRN_MSRDORM /* MSR Dormant Register */ #define NIADORM SPRN_NIADORM /* NIA Dormant Register */ #define TSC SPRN_TSC /* Thread switch control */ #define TST SPRN_TST /* Thread switch timeout */ #define IABR SPRN_IABR /* Instruction Address Breakpoint Register */ -#define ICMP SPRN_ICMP /* Instruction TLB Compare Register */ -#define IMISS SPRN_IMISS /* Instruction TLB Miss Register */ -#define IMMR SPRN_IMMR /* PPC 860/821 Internal Memory Map Register */ #define L2CR SPRN_L2CR /* PPC 750 L2 control register */ #define __LR SPRN_LR #define PVR SPRN_PVR /* Processor Version */ #define PIR SPRN_PIR /* Processor ID */ #define PURR SPRN_PURR /* Processor Utilization of Resource Register */ -//#define RPA SPRN_RPA /* Required Physical Address Register */ #define SDR1 SPRN_SDR1 /* MMU hash base register */ #define SPR0 SPRN_SPRG0 /* Supervisor Private Registers */ #define SPR1 SPRN_SPRG1 @@ -389,10 +236,6 @@ #define TBRU SPRN_TBRU /* Time Base Read Upper Register */ #define TBWL SPRN_TBWL /* Time Base Write Lower Register */ #define TBWU SPRN_TBWU /* Time Base Write Upper Register */ -#define ICTC 1019 -#define THRM1 SPRN_THRM1 /* Thermal Management Register 1 */ -#define THRM2 SPRN_THRM2 /* Thermal Management Register 2 */ -#define THRM3 SPRN_THRM3 /* Thermal Management Register 3 */ #define XER SPRN_XER /* Processor Version Register (PVR) field extraction */ -- cgit v1.2.3 From 6dc2f0c7df6cefda5932ac8bcd9ca5ef45de36ee Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 2 Jun 2005 14:02:02 -0700 Subject: [PATCH] ppc64: cleanup iseries runlight support The iseries has a bar graph on the front panel that shows how busy it is. The operating system sets and clears a bit in the CTRL register to control it. Instead of going to the complexity of using a thread info bit, just set and clear it in the idle loop. Also create two helper functions, ppc64_runlatch_on and ppc64_runlatch_off. Finally don't use the short form of the SPR defines. Signed-off-by: Anton Blanchard Acked-by: Stephen Rothwell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ppc64/kernel/entry.S | 9 --------- arch/ppc64/kernel/head.S | 10 +++++----- arch/ppc64/kernel/idle.c | 8 +++----- arch/ppc64/kernel/process.c | 3 --- arch/ppc64/kernel/sysfs.c | 8 ++------ include/asm-ppc64/processor.h | 27 +++++++++++++++++++++------ include/asm-ppc64/thread_info.h | 4 ++-- 7 files changed, 33 insertions(+), 36 deletions(-) (limited to 'include') diff --git a/arch/ppc64/kernel/entry.S b/arch/ppc64/kernel/entry.S index d3604056e1a9..b61572eb2a71 100644 --- a/arch/ppc64/kernel/entry.S +++ b/arch/ppc64/kernel/entry.S @@ -436,15 +436,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) REST_8GPRS(14, r1) REST_10GPRS(22, r1) -#ifdef CONFIG_PPC_ISERIES - clrrdi r7,r1,THREAD_SHIFT /* get current_thread_info() */ - ld r7,TI_FLAGS(r7) /* Get run light flag */ - mfspr r9,CTRLF - srdi r7,r7,TIF_RUN_LIGHT - insrdi r9,r7,1,63 /* Insert run light into CTRL */ - mtspr CTRLT,r9 -#endif - /* convert old thread to its task_struct for return value */ addi r3,r3,-THREAD ld r7,_NIP(r1) /* Return to _switch caller in new task */ diff --git a/arch/ppc64/kernel/head.S b/arch/ppc64/kernel/head.S index 92a744c31ab1..346dbf606b5d 100644 --- a/arch/ppc64/kernel/head.S +++ b/arch/ppc64/kernel/head.S @@ -626,10 +626,10 @@ system_reset_iSeries: lhz r24,PACAPACAINDEX(r13) /* Get processor # */ cmpwi 0,r24,0 /* Are we processor 0? */ beq .__start_initialization_iSeries /* Start up the first processor */ - mfspr r4,CTRLF - li r5,RUNLATCH /* Turn off the run light */ + mfspr r4,SPRN_CTRLF + li r5,CTRL_RUNLATCH /* Turn off the run light */ andc r4,r4,r5 - mtspr CTRLT,r4 + mtspr SPRN_CTRLT,r4 1: HMT_LOW @@ -2082,9 +2082,9 @@ _GLOBAL(hmt_start_secondary) mfspr r4, HID0 ori r4, r4, 0x1 mtspr HID0, r4 - mfspr r4, CTRLF + mfspr r4, SPRN_CTRLF oris r4, r4, 0x40 - mtspr CTRLT, r4 + mtspr SPRN_CTRLT, r4 blr #endif diff --git a/arch/ppc64/kernel/idle.c b/arch/ppc64/kernel/idle.c index 6abc621d3ba0..f24ce2b87200 100644 --- a/arch/ppc64/kernel/idle.c +++ b/arch/ppc64/kernel/idle.c @@ -75,13 +75,9 @@ static int iSeries_idle(void) { struct paca_struct *lpaca; long oldval; - unsigned long CTRL; /* ensure iSeries run light will be out when idle */ - clear_thread_flag(TIF_RUN_LIGHT); - CTRL = mfspr(CTRLF); - CTRL &= ~RUNLATCH; - mtspr(CTRLT, CTRL); + ppc64_runlatch_off(); lpaca = get_paca(); @@ -111,7 +107,9 @@ static int iSeries_idle(void) } } + ppc64_runlatch_on(); schedule(); + ppc64_runlatch_off(); } return 0; diff --git a/arch/ppc64/kernel/process.c b/arch/ppc64/kernel/process.c index 8b0686122738..cdfecbeb331f 100644 --- a/arch/ppc64/kernel/process.c +++ b/arch/ppc64/kernel/process.c @@ -378,9 +378,6 @@ copy_thread(int nr, unsigned long clone_flags, unsigned long usp, childregs->gpr[1] = sp + sizeof(struct pt_regs); p->thread.regs = NULL; /* no user register state */ clear_ti_thread_flag(p->thread_info, TIF_32BIT); -#ifdef CONFIG_PPC_ISERIES - set_ti_thread_flag(p->thread_info, TIF_RUN_LIGHT); -#endif } else { childregs->gpr[1] = usp; p->thread.regs = childregs; diff --git a/arch/ppc64/kernel/sysfs.c b/arch/ppc64/kernel/sysfs.c index 0925694c3ce5..c8fa6569b2fd 100644 --- a/arch/ppc64/kernel/sysfs.c +++ b/arch/ppc64/kernel/sysfs.c @@ -113,7 +113,6 @@ void ppc64_enable_pmcs(void) #ifdef CONFIG_PPC_PSERIES unsigned long set, reset; int ret; - unsigned int ctrl; #endif /* CONFIG_PPC_PSERIES */ /* Only need to enable them once */ @@ -167,11 +166,8 @@ void ppc64_enable_pmcs(void) * On SMT machines we have to set the run latch in the ctrl register * in order to make PMC6 spin. */ - if (cpu_has_feature(CPU_FTR_SMT)) { - ctrl = mfspr(CTRLF); - ctrl |= RUNLATCH; - mtspr(CTRLT, ctrl); - } + if (cpu_has_feature(CPU_FTR_SMT)) + ppc64_runlatch_on(); #endif /* CONFIG_PPC_PSERIES */ } diff --git a/include/asm-ppc64/processor.h b/include/asm-ppc64/processor.h index 5ba6aa45e1e2..809c634ba1df 100644 --- a/include/asm-ppc64/processor.h +++ b/include/asm-ppc64/processor.h @@ -164,6 +164,9 @@ #define SPRN_USIA 0x3AB /* User Sampled Instruction Address Register */ #define SPRN_XER 0x001 /* Fixed Point Exception Register */ #define SPRN_VRSAVE 0x100 /* Vector save */ +#define SPRN_CTRLF 0x088 +#define SPRN_CTRLT 0x098 +#define CTRL_RUNLATCH 0x1 /* Performance monitor SPRs */ #define SPRN_SIAR 780 @@ -279,12 +282,6 @@ #define XGLUE(a,b) a##b #define GLUE(a,b) XGLUE(a,b) -/* iSeries CTRL register (for runlatch) */ - -#define CTRLT 0x098 -#define CTRLF 0x088 -#define RUNLATCH 0x0001 - #ifdef __ASSEMBLY__ #define _GLOBAL(name) \ @@ -499,6 +496,24 @@ static inline void prefetchw(const void *x) #define HAVE_ARCH_PICK_MMAP_LAYOUT +static inline void ppc64_runlatch_on(void) +{ + unsigned long ctrl; + + ctrl = mfspr(SPRN_CTRLF); + ctrl |= CTRL_RUNLATCH; + mtspr(SPRN_CTRLT, ctrl); +} + +static inline void ppc64_runlatch_off(void) +{ + unsigned long ctrl; + + ctrl = mfspr(SPRN_CTRLF); + ctrl &= ~CTRL_RUNLATCH; + mtspr(SPRN_CTRLT, ctrl); +} + #endif /* __KERNEL__ */ #endif /* __ASSEMBLY__ */ diff --git a/include/asm-ppc64/thread_info.h b/include/asm-ppc64/thread_info.h index 037b5e06083c..48b7900e90ec 100644 --- a/include/asm-ppc64/thread_info.h +++ b/include/asm-ppc64/thread_info.h @@ -96,7 +96,7 @@ static inline struct thread_info *current_thread_info(void) #define TIF_POLLING_NRFLAG 4 /* true if poll_idle() is polling TIF_NEED_RESCHED */ #define TIF_32BIT 5 /* 32 bit binary */ -#define TIF_RUN_LIGHT 6 /* iSeries run light */ +/* #define SPARE 6 */ #define TIF_ABI_PENDING 7 /* 32/64 bit switch needed */ #define TIF_SYSCALL_AUDIT 8 /* syscall auditing active */ #define TIF_SINGLESTEP 9 /* singlestepping active */ @@ -110,7 +110,7 @@ static inline struct thread_info *current_thread_info(void) #define _TIF_NEED_RESCHED (1< Date: Thu, 2 Jun 2005 18:17:13 -0400 Subject: libata: update inline source docs --- drivers/scsi/libata-core.c | 283 +++++++++++++++++++++++++++++++++++++++++++-- include/linux/libata.h | 58 ++++++++++ 2 files changed, 334 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c index 31b2984f379a..86e84ed87bee 100644 --- a/drivers/scsi/libata-core.c +++ b/drivers/scsi/libata-core.c @@ -186,6 +186,28 @@ static void ata_tf_load_mmio(struct ata_port *ap, struct ata_taskfile *tf) ata_wait_idle(ap); } + +/** + * ata_tf_load - send taskfile registers to host controller + * @ap: Port to which output is sent + * @tf: ATA taskfile register set + * + * Outputs ATA taskfile to standard ATA host controller using MMIO + * or PIO as indicated by the ATA_FLAG_MMIO flag. + * Writes the control, feature, nsect, lbal, lbam, and lbah registers. + * Optionally (ATA_TFLAG_LBA48) writes hob_feature, hob_nsect, + * hob_lbal, hob_lbam, and hob_lbah. + * + * This function waits for idle (!BUSY and !DRQ) after writing + * registers. If the control register has a new value, this + * function also waits for idle after writing control and before + * writing the remaining registers. + * + * May be used as the tf_load() entry in ata_port_operations. + * + * LOCKING: + * Inherited from caller. + */ void ata_tf_load(struct ata_port *ap, struct ata_taskfile *tf) { if (ap->flags & ATA_FLAG_MMIO) @@ -195,11 +217,11 @@ void ata_tf_load(struct ata_port *ap, struct ata_taskfile *tf) } /** - * ata_exec_command - issue ATA command to host controller + * ata_exec_command_pio - issue ATA command to host controller * @ap: port to which command is being issued * @tf: ATA taskfile register set * - * Issues PIO/MMIO write to ATA command register, with proper + * Issues PIO write to ATA command register, with proper * synchronization with interrupt handler / other threads. * * LOCKING: @@ -235,6 +257,18 @@ static void ata_exec_command_mmio(struct ata_port *ap, struct ata_taskfile *tf) ata_pause(ap); } + +/** + * ata_exec_command - issue ATA command to host controller + * @ap: port to which command is being issued + * @tf: ATA taskfile register set + * + * Issues PIO/MMIO write to ATA command register, with proper + * synchronization with interrupt handler / other threads. + * + * LOCKING: + * spin_lock_irqsave(host_set lock) + */ void ata_exec_command(struct ata_port *ap, struct ata_taskfile *tf) { if (ap->flags & ATA_FLAG_MMIO) @@ -305,7 +339,7 @@ void ata_tf_to_host_nolock(struct ata_port *ap, struct ata_taskfile *tf) } /** - * ata_tf_read - input device's ATA taskfile shadow registers + * ata_tf_read_pio - input device's ATA taskfile shadow registers * @ap: Port from which input is read * @tf: ATA taskfile register set for storing input * @@ -368,6 +402,23 @@ static void ata_tf_read_mmio(struct ata_port *ap, struct ata_taskfile *tf) } } + +/** + * ata_tf_read - input device's ATA taskfile shadow registers + * @ap: Port from which input is read + * @tf: ATA taskfile register set for storing input + * + * Reads ATA taskfile registers for currently-selected device + * into @tf. + * + * Reads nsect, lbal, lbam, lbah, and device. If ATA_TFLAG_LBA48 + * is set, also reads the hob registers. + * + * May be used as the tf_read() entry in ata_port_operations. + * + * LOCKING: + * Inherited from caller. + */ void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf) { if (ap->flags & ATA_FLAG_MMIO) @@ -381,7 +432,7 @@ void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf) * @ap: port where the device is * * Reads ATA taskfile status register for currently-selected device - * and return it's value. This also clears pending interrupts + * and return its value. This also clears pending interrupts * from this device * * LOCKING: @@ -397,7 +448,7 @@ static u8 ata_check_status_pio(struct ata_port *ap) * @ap: port where the device is * * Reads ATA taskfile status register for currently-selected device - * via MMIO and return it's value. This also clears pending interrupts + * via MMIO and return its value. This also clears pending interrupts * from this device * * LOCKING: @@ -408,6 +459,20 @@ static u8 ata_check_status_mmio(struct ata_port *ap) return readb((void __iomem *) ap->ioaddr.status_addr); } + +/** + * ata_check_status - Read device status reg & clear interrupt + * @ap: port where the device is + * + * Reads ATA taskfile status register for currently-selected device + * and return its value. This also clears pending interrupts + * from this device + * + * May be used as the check_status() entry in ata_port_operations. + * + * LOCKING: + * Inherited from caller. + */ u8 ata_check_status(struct ata_port *ap) { if (ap->flags & ATA_FLAG_MMIO) @@ -415,6 +480,20 @@ u8 ata_check_status(struct ata_port *ap) return ata_check_status_pio(ap); } + +/** + * ata_altstatus - Read device alternate status reg + * @ap: port where the device is + * + * Reads ATA taskfile alternate status register for + * currently-selected device and return its value. + * + * Note: may NOT be used as the check_altstatus() entry in + * ata_port_operations. + * + * LOCKING: + * Inherited from caller. + */ u8 ata_altstatus(struct ata_port *ap) { if (ap->ops->check_altstatus) @@ -425,6 +504,20 @@ u8 ata_altstatus(struct ata_port *ap) return inb(ap->ioaddr.altstatus_addr); } + +/** + * ata_chk_err - Read device error reg + * @ap: port where the device is + * + * Reads ATA taskfile error register for + * currently-selected device and return its value. + * + * Note: may NOT be used as the check_err() entry in + * ata_port_operations. + * + * LOCKING: + * Inherited from caller. + */ u8 ata_chk_err(struct ata_port *ap) { if (ap->ops->check_err) @@ -873,10 +966,24 @@ void ata_dev_id_string(u16 *id, unsigned char *s, } } + +/** + * ata_noop_dev_select - Select device 0/1 on ATA bus + * @ap: ATA channel to manipulate + * @device: ATA device (numbered from zero) to select + * + * This function performs no actual function. + * + * May be used as the dev_select() entry in ata_port_operations. + * + * LOCKING: + * caller. + */ void ata_noop_dev_select (struct ata_port *ap, unsigned int device) { } + /** * ata_std_dev_select - Select device 0/1 on ATA bus * @ap: ATA channel to manipulate @@ -884,7 +991,9 @@ void ata_noop_dev_select (struct ata_port *ap, unsigned int device) * * Use the method defined in the ATA specification to * make either device 0, or device 1, active on the - * ATA channel. + * ATA channel. Works with both PIO and MMIO. + * + * May be used as the dev_select() entry in ata_port_operations. * * LOCKING: * caller. @@ -2127,6 +2236,19 @@ void ata_qc_prep(struct ata_queued_cmd *qc) * spin_lock_irqsave(host_set lock) */ + + +/** + * ata_sg_init_one - Prepare a one-entry scatter-gather list. + * @qc: Queued command + * @buf: transfer buffer + * @buflen: length of buf + * + * Builds a single-entry scatter-gather list to initiate a + * transfer utilizing the specified buffer. + * + * LOCKING: + */ void ata_sg_init_one(struct ata_queued_cmd *qc, void *buf, unsigned int buflen) { struct scatterlist *sg; @@ -2158,6 +2280,18 @@ void ata_sg_init_one(struct ata_queued_cmd *qc, void *buf, unsigned int buflen) * spin_lock_irqsave(host_set lock) */ + +/** + * ata_sg_init - Assign a scatter gather list to a queued command + * @qc: Queued command + * @sg: Scatter-gather list + * @n_elem: length of sg list + * + * Attaches a scatter-gather list to a queued command. + * + * LOCKING: + */ + void ata_sg_init(struct ata_queued_cmd *qc, struct scatterlist *sg, unsigned int n_elem) { @@ -2331,6 +2465,18 @@ static void ata_pio_complete (struct ata_port *ap) ata_qc_complete(qc, drv_stat); } + +/** + * swap_buf_le16 - + * @buf: Buffer to swap + * @buf_words: Number of 16-bit words in buffer. + * + * Swap halves of 16-bit words if needed to convert from + * little-endian byte order to native cpu byte order, or + * vice-versa. + * + * LOCKING: + */ void swap_buf_le16(u16 *buf, unsigned int buf_words) { #ifdef __BIG_ENDIAN @@ -2992,6 +3138,7 @@ err_out: return -1; } + /** * ata_qc_issue_prot - issue taskfile to device in proto-dependent manner * @qc: command to issue to device @@ -3001,6 +3148,8 @@ err_out: * classes called "protocols", and issuing each type of protocol * is slightly different. * + * May be used as the qc_issue() entry in ata_port_operations. + * * LOCKING: * spin_lock_irqsave(host_set lock) * @@ -3058,7 +3207,7 @@ int ata_qc_issue_prot(struct ata_queued_cmd *qc) } /** - * ata_bmdma_setup - Set up PCI IDE BMDMA transaction + * ata_bmdma_setup_mmio - Set up PCI IDE BMDMA transaction * @qc: Info associated with this ATA transaction. * * LOCKING: @@ -3165,6 +3314,18 @@ static void ata_bmdma_start_pio (struct ata_queued_cmd *qc) ap->ioaddr.bmdma_addr + ATA_DMA_CMD); } + +/** + * ata_bmdma_start - Start a PCI IDE BMDMA transaction + * @qc: Info associated with this ATA transaction. + * + * Writes the ATA_DMA_START flag to the DMA command register. + * + * May be used as the bmdma_start() entry in ata_port_operations. + * + * LOCKING: + * spin_lock_irqsave(host_set lock) + */ void ata_bmdma_start(struct ata_queued_cmd *qc) { if (qc->ap->flags & ATA_FLAG_MMIO) @@ -3173,6 +3334,20 @@ void ata_bmdma_start(struct ata_queued_cmd *qc) ata_bmdma_start_pio(qc); } + +/** + * ata_bmdma_setup - Set up PCI IDE BMDMA transaction + * @qc: Info associated with this ATA transaction. + * + * Writes address of PRD table to device's PRD Table Address + * register, sets the DMA control register, and calls + * ops->exec_command() to start the transfer. + * + * May be used as the bmdma_setup() entry in ata_port_operations. + * + * LOCKING: + * spin_lock_irqsave(host_set lock) + */ void ata_bmdma_setup(struct ata_queued_cmd *qc) { if (qc->ap->flags & ATA_FLAG_MMIO) @@ -3181,6 +3356,19 @@ void ata_bmdma_setup(struct ata_queued_cmd *qc) ata_bmdma_setup_pio(qc); } + +/** + * ata_bmdma_irq_clear - Clear PCI IDE BMDMA interrupt. + * @qc: Info associated with this ATA transaction. + * + * Clear interrupt and error flags in DMA status register. + * + * May be used as the irq_clear() entry in ata_port_operations. + * + * LOCKING: + * spin_lock_irqsave(host_set lock) + */ + void ata_bmdma_irq_clear(struct ata_port *ap) { if (ap->flags & ATA_FLAG_MMIO) { @@ -3193,6 +3381,19 @@ void ata_bmdma_irq_clear(struct ata_port *ap) } + +/** + * ata_bmdma_status - Read PCI IDE BMDMA status + * @qc: Info associated with this ATA transaction. + * + * Read and return BMDMA status register. + * + * May be used as the bmdma_status() entry in ata_port_operations. + * + * LOCKING: + * spin_lock_irqsave(host_set lock) + */ + u8 ata_bmdma_status(struct ata_port *ap) { u8 host_stat; @@ -3204,6 +3405,19 @@ u8 ata_bmdma_status(struct ata_port *ap) return host_stat; } + +/** + * ata_bmdma_stop - Stop PCI IDE BMDMA transfer + * @qc: Info associated with this ATA transaction. + * + * Clears the ATA_DMA_START flag in the dma control register + * + * May be used as the bmdma_stop() entry in ata_port_operations. + * + * LOCKING: + * spin_lock_irqsave(host_set lock) + */ + void ata_bmdma_stop(struct ata_port *ap) { if (ap->flags & ATA_FLAG_MMIO) { @@ -3407,6 +3621,19 @@ err_out: ata_qc_complete(qc, ATA_ERR); } + +/** + * ata_port_start - Set port up for dma. + * @ap: Port to initialize + * + * Called just after data structures for each port are + * initialized. Allocates space for PRD table. + * + * May be used as the port_start() entry in ata_port_operations. + * + * LOCKING: + */ + int ata_port_start (struct ata_port *ap) { struct device *dev = ap->host_set->dev; @@ -3420,6 +3647,18 @@ int ata_port_start (struct ata_port *ap) return 0; } + +/** + * ata_port_stop - Undo ata_port_start() + * @ap: Port to shut down + * + * Frees the PRD table. + * + * May be used as the port_stop() entry in ata_port_operations. + * + * LOCKING: + */ + void ata_port_stop (struct ata_port *ap) { struct device *dev = ap->host_set->dev; @@ -3720,7 +3959,15 @@ int ata_scsi_release(struct Scsi_Host *host) /** * ata_std_ports - initialize ioaddr with standard port offsets. * @ioaddr: IO address structure to be initialized + * + * Utility function which initializes data_addr, error_addr, + * feature_addr, nsect_addr, lbal_addr, lbam_addr, lbah_addr, + * device_addr, status_addr, and command_addr to standard offsets + * relative to cmd_addr. + * + * Does not set ctl_addr, altstatus_addr, bmdma_addr, or scr_addr. */ + void ata_std_ports(struct ata_ioports *ioaddr) { ioaddr->data_addr = ioaddr->cmd_addr + ATA_REG_DATA; @@ -3762,6 +4009,20 @@ ata_probe_ent_alloc(struct device *dev, struct ata_port_info *port) return probe_ent; } + + +/** + * ata_pci_init_native_mode - Initialize native-mode driver + * @pdev: pci device to be initialized + * @port: array[2] of pointers to port info structures. + * + * Utility function which allocates and initializes an + * ata_probe_ent structure for a standard dual-port + * PIO-based IDE controller. The returned ata_probe_ent + * structure can be passed to ata_device_add(). The returned + * ata_probe_ent structure should then be freed with kfree(). + */ + #ifdef CONFIG_PCI struct ata_probe_ent * ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port) @@ -3843,6 +4104,14 @@ ata_pci_init_legacy_mode(struct pci_dev *pdev, struct ata_port_info **port, * @port_info: Information from low-level host driver * @n_ports: Number of ports attached to host controller * + * This is a helper function which can be called from a driver's + * xxx_init_one() probe function if the hardware uses traditional + * IDE taskfile registers. + * + * This function calls pci_enable_device(), reserves its register + * regions, sets the dma mask, enables bus master mode, and calls + * ata_device_add() + * * LOCKING: * Inherited from PCI layer (may sleep). * diff --git a/include/linux/libata.h b/include/linux/libata.h index 1f7e2039a04e..ad410590664f 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -466,12 +466,34 @@ static inline u8 ata_chk_status(struct ata_port *ap) return ap->ops->check_status(ap); } + +/** + * ata_pause - Flush writes and pause 400 nanoseconds. + * @ap: Port to wait for. + * + * LOCKING: + * Inherited from caller. + */ + static inline void ata_pause(struct ata_port *ap) { ata_altstatus(ap); ndelay(400); } + +/** + * ata_busy_wait - Wait for a port status register + * @ap: Port to wait for. + * + * Waits up to max*10 microseconds for the selected bits in the port's + * status register to be cleared. + * Returns final value of status register. + * + * LOCKING: + * Inherited from caller. + */ + static inline u8 ata_busy_wait(struct ata_port *ap, unsigned int bits, unsigned int max) { @@ -486,6 +508,18 @@ static inline u8 ata_busy_wait(struct ata_port *ap, unsigned int bits, return status; } + +/** + * ata_wait_idle - Wait for a port to be idle. + * @ap: Port to wait for. + * + * Waits up to 10ms for port's BUSY and DRQ signals to clear. + * Returns final value of status register. + * + * LOCKING: + * Inherited from caller. + */ + static inline u8 ata_wait_idle(struct ata_port *ap) { u8 status = ata_busy_wait(ap, ATA_BUSY | ATA_DRQ, 1000); @@ -524,6 +558,18 @@ static inline void ata_tf_init(struct ata_port *ap, struct ata_taskfile *tf, uns tf->device = ATA_DEVICE_OBS | ATA_DEV1; } + +/** + * ata_irq_on - Enable interrupts on a port. + * @ap: Port on which interrupts are enabled. + * + * Enable interrupts on a legacy IDE device using MMIO or PIO, + * wait for idle, clear any pending interrupts. + * + * LOCKING: + * Inherited from caller. + */ + static inline u8 ata_irq_on(struct ata_port *ap) { struct ata_ioports *ioaddr = &ap->ioaddr; @@ -543,6 +589,18 @@ static inline u8 ata_irq_on(struct ata_port *ap) return tmp; } + +/** + * ata_irq_ack - Acknowledge a device interrupt. + * @ap: Port on which interrupts are enabled. + * + * Wait up to 10 ms for legacy IDE device to become idle (BUSY + * or BUSY+DRQ clear). Obtain dma status and port status from + * device. Clear the interrupt. Return port status. + * + * LOCKING: + */ + static inline u8 ata_irq_ack(struct ata_port *ap, unsigned int chk_drq) { unsigned int bits = chk_drq ? ATA_BUSY | ATA_DRQ : ATA_BUSY; -- cgit v1.2.3 From b597ef4712c05c962640a655386a7f06cc1a1fbc Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 2 Jun 2005 16:36:00 -0700 Subject: [NET]: Fix locking in shaper driver. o use a semaphore instead of an opencoded and racy lock o move locking out of shaper_kick and into the callers - most just released the lock before calling shaper_kick o remove in_interrupt() tests. from ->close we can always block, from ->hard_start_xmit and timer context never Signed-off-by: Christoph Hellwig Signed-off-by: David S. Miller --- drivers/net/shaper.c | 86 +++++++++++------------------------------------ include/linux/if_shaper.h | 3 +- 2 files changed, 20 insertions(+), 69 deletions(-) (limited to 'include') diff --git a/drivers/net/shaper.c b/drivers/net/shaper.c index e68cf5fb4920..20edeb345792 100644 --- a/drivers/net/shaper.c +++ b/drivers/net/shaper.c @@ -100,35 +100,8 @@ static int sh_debug; /* Debug flag */ #define SHAPER_BANNER "CymruNet Traffic Shaper BETA 0.04 for Linux 2.1\n" -/* - * Locking - */ - -static int shaper_lock(struct shaper *sh) -{ - /* - * Lock in an interrupt must fail - */ - while (test_and_set_bit(0, &sh->locked)) - { - if (!in_interrupt()) - sleep_on(&sh->wait_queue); - else - return 0; - - } - return 1; -} - static void shaper_kick(struct shaper *sh); -static void shaper_unlock(struct shaper *sh) -{ - clear_bit(0, &sh->locked); - wake_up(&sh->wait_queue); - shaper_kick(sh); -} - /* * Compute clocks on a buffer */ @@ -157,17 +130,15 @@ static void shaper_setspeed(struct shaper *shaper, int bitspersec) * Throw a frame at a shaper. */ -static int shaper_qframe(struct shaper *shaper, struct sk_buff *skb) + +static int shaper_start_xmit(struct sk_buff *skb, struct net_device *dev) { + struct shaper *shaper = dev->priv; struct sk_buff *ptr; - /* - * Get ready to work on this shaper. Lock may fail if its - * an interrupt and locked. - */ - - if(!shaper_lock(shaper)) - return -1; + if (down_trylock(&shaper->sem)) + return -1; + ptr=shaper->sendq.prev; /* @@ -260,7 +231,8 @@ static int shaper_qframe(struct shaper *shaper, struct sk_buff *skb) dev_kfree_skb(ptr); shaper->stats.collisions++; } - shaper_unlock(shaper); + shaper_kick(shaper); + up(&shaper->sem); return 0; } @@ -297,8 +269,13 @@ static void shaper_queue_xmit(struct shaper *shaper, struct sk_buff *skb) static void shaper_timer(unsigned long data) { - struct shaper *sh=(struct shaper *)data; - shaper_kick(sh); + struct shaper *shaper = (struct shaper *)data; + + if (!down_trylock(&shaper->sem)) { + shaper_kick(shaper); + up(&shaper->sem); + } else + mod_timer(&shaper->timer, jiffies); } /* @@ -310,19 +287,6 @@ static void shaper_kick(struct shaper *shaper) { struct sk_buff *skb; - /* - * Shaper unlock will kick - */ - - if (test_and_set_bit(0, &shaper->locked)) - { - if(sh_debug) - printk("Shaper locked.\n"); - mod_timer(&shaper->timer, jiffies); - return; - } - - /* * Walk the list (may be empty) */ @@ -364,8 +328,6 @@ static void shaper_kick(struct shaper *shaper) if(skb!=NULL) mod_timer(&shaper->timer, SHAPERCB(skb)->shapeclock); - - clear_bit(0, &shaper->locked); } @@ -376,14 +338,12 @@ static void shaper_kick(struct shaper *shaper) static void shaper_flush(struct shaper *shaper) { struct sk_buff *skb; - if(!shaper_lock(shaper)) - { - printk(KERN_ERR "shaper: shaper_flush() called by an irq!\n"); - return; - } + + down(&shaper->sem); while((skb=skb_dequeue(&shaper->sendq))!=NULL) dev_kfree_skb(skb); - shaper_unlock(shaper); + shaper_kick(shaper); + up(&shaper->sem); } /* @@ -426,13 +386,6 @@ static int shaper_close(struct net_device *dev) * ARP and other resolutions and not before. */ - -static int shaper_start_xmit(struct sk_buff *skb, struct net_device *dev) -{ - struct shaper *sh=dev->priv; - return shaper_qframe(sh, skb); -} - static struct net_device_stats *shaper_get_stats(struct net_device *dev) { struct shaper *sh=dev->priv; @@ -623,7 +576,6 @@ static void shaper_init_priv(struct net_device *dev) init_timer(&sh->timer); sh->timer.function=shaper_timer; sh->timer.data=(unsigned long)sh; - init_waitqueue_head(&sh->wait_queue); } /* diff --git a/include/linux/if_shaper.h b/include/linux/if_shaper.h index 0485b256d043..004e6f09a6e2 100644 --- a/include/linux/if_shaper.h +++ b/include/linux/if_shaper.h @@ -23,7 +23,7 @@ struct shaper __u32 shapeclock; unsigned long recovery; /* Time we can next clock a packet out on an empty queue */ - unsigned long locked; + struct semaphore sem; struct net_device_stats stats; struct net_device *dev; int (*hard_start_xmit) (struct sk_buff *skb, @@ -38,7 +38,6 @@ struct shaper int (*hard_header_cache)(struct neighbour *neigh, struct hh_cache *hh); void (*header_cache_update)(struct hh_cache *hh, struct net_device *dev, unsigned char * haddr); struct net_device_stats* (*get_stats)(struct net_device *dev); - wait_queue_head_t wait_queue; struct timer_list timer; }; -- cgit v1.2.3 From 5ba0eac6e0b7e2889649a1105d97c600595e2bb1 Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Thu, 2 Jun 2005 16:48:05 -0700 Subject: [NET]: Fix HH_DATA_OFF. When the hardware header size is a multiple of HH_DATA_MOD, HH_DATA_OFF() incorrectly returns HH_DATA_MOD (instead of 0). This affects ieee80211 layer as 802.11 header is 32 bytes long. Signed-off-by: Jiri Benc Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 470af8c1a4a0..ba5d1236aa17 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -204,7 +204,7 @@ struct hh_cache /* cached hardware header; allow for machine alignment needs. */ #define HH_DATA_MOD 16 #define HH_DATA_OFF(__len) \ - (HH_DATA_MOD - ((__len) & (HH_DATA_MOD - 1))) + (HH_DATA_MOD - (((__len - 1) & (HH_DATA_MOD - 1)) + 1)) #define HH_DATA_ALIGN(__len) \ (((__len)+(HH_DATA_MOD-1))&~(HH_DATA_MOD - 1)) unsigned long hh_data[HH_DATA_ALIGN(LL_MAX_HEADER) / sizeof(long)]; -- cgit v1.2.3 From 719df469cb51199316ae2a11c75a8046be34b899 Mon Sep 17 00:00:00 2001 From: Roman Kagan Date: Fri, 6 May 2005 00:55:56 +0400 Subject: [PATCH] USB: update urb documentation On Wed, May 04, 2005 at 01:37:30PM -0700, David Brownell wrote: > On Wednesday 04 May 2005 12:19 pm, Roman Kagan wrote: > > struct urb { > > /* private, usb core and host controller only fields in the urb */ > > ... > > struct list_head urb_list; /* list pointer to all active urbs */ > > ... > > }; > > > > Is it safe to use it for driver's purposes when the driver owns the urb, > > that is, starting from the completion routine until the urb is submitted > > with usb_submit_urb()? > > Right now, it should be. Great! FWIW I've briefly tested a modified version of usbatm using the list head in struct urb instead of creating a wrapper struct, and I haven't seen any failures yet. So I tend to believe that your "should be" actually means "is" :) > > If it is, can it be guaranteed in future, e.g. > > by moving the list head into the public section of struct urb? > > In fact I'm not sure why it ever got called "private" to usbcore/hcds. > I thought the idea was that it should be like urb->status, reserved for > whoever controls the URB. OK then how about the following (essentially documentation) patch? Signed-off-by: Roman Kagan Acked-by: David Brownell Signed-off-by: Greg Kroah-Hartman --- include/linux/usb.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/usb.h b/include/linux/usb.h index 41d1a644c9d4..2d1ac5058534 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -796,6 +796,10 @@ typedef void (*usb_complete_t)(struct urb *, struct pt_regs *); * of the iso_frame_desc array, and the number of errors is reported in * error_count. Completion callbacks for ISO transfers will normally * (re)submit URBs to ensure a constant transfer rate. + * + * Note that even fields marked "public" should not be touched by the driver + * when the urb is owned by the hcd, that is, since the call to + * usb_submit_urb() till the entry into the completion routine. */ struct urb { @@ -803,12 +807,12 @@ struct urb struct kref kref; /* reference count of the URB */ spinlock_t lock; /* lock for the URB */ void *hcpriv; /* private data for host controller */ - struct list_head urb_list; /* list pointer to all active urbs */ int bandwidth; /* bandwidth for INT/ISO request */ atomic_t use_count; /* concurrent submissions counter */ u8 reject; /* submissions will fail */ /* public, documented fields in the urb that can be used by drivers */ + struct list_head urb_list; /* list head for use by the urb owner */ struct usb_device *dev; /* (in) pointer to associated device */ unsigned int pipe; /* (in) pipe information */ int status; /* (return) non-ISO status */ -- cgit v1.2.3 From 4ab5c01c7cc28f29d5cf5f2a591cb6f7bbece48e Mon Sep 17 00:00:00 2001 From: Deepak Saxena Date: Fri, 3 Jun 2005 20:52:25 +0100 Subject: [PATCH] ARM: 2692/1: Fix compile warnings in include/asm-arm/arch-ixp2000/io.h Patch from Deepak Saxena This patch fixes the following warnings: include/asm/arch/io.h: In function `insw': include/asm/arch/io.h:78: warning: comparison of distinct pointer types lacks acast include/asm/arch/io.h:79: warning: comparison of distinct pointer types lacks acast include/asm/arch/io.h: In function `outsw': include/asm/arch/io.h:103: warning: comparison of distinct pointer types lacks a cast include/asm/arch/io.h:104: warning: comparison of distinct pointer types lacks a cast include/asm/arch/io.h: In function `inw': include/asm/arch/io.h:127: warning: comparison of distinct pointer types lacks a cast Signed-off-by: Deepak Saxena Signed-off-by: Russell King --- include/asm-arm/arch-ixp2000/io.h | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/asm-arm/arch-ixp2000/io.h b/include/asm-arm/arch-ixp2000/io.h index a8e3c2daefd6..083462668e18 100644 --- a/include/asm-arm/arch-ixp2000/io.h +++ b/include/asm-arm/arch-ixp2000/io.h @@ -75,8 +75,8 @@ static inline void insw(u32 ptr, void *buf, int length) * Is this cycle meant for the CS8900? */ if ((machine_is_ixdp2401() || machine_is_ixdp2801()) && - ((port >= IXDP2X01_CS8900_VIRT_BASE) && - (port <= IXDP2X01_CS8900_VIRT_END))) { + (((u32)port >= (u32)IXDP2X01_CS8900_VIRT_BASE) && + ((u32)port <= (u32)IXDP2X01_CS8900_VIRT_END))) { u8 *buf8 = (u8*)buf; register u32 tmp32; @@ -100,8 +100,8 @@ static inline void outsw(u32 ptr, void *buf, int length) * Is this cycle meant for the CS8900? */ if ((machine_is_ixdp2401() || machine_is_ixdp2801()) && - ((port >= IXDP2X01_CS8900_VIRT_BASE) && - (port <= IXDP2X01_CS8900_VIRT_END))) { + (((u32)port >= (u32)IXDP2X01_CS8900_VIRT_BASE) && + ((u32)port <= (u32)IXDP2X01_CS8900_VIRT_END))) { register u32 tmp32; u8 *buf8 = (u8*)buf; do { @@ -124,8 +124,8 @@ static inline u16 inw(u32 ptr) * Is this cycle meant for the CS8900? */ if ((machine_is_ixdp2401() || machine_is_ixdp2801()) && - ((port >= IXDP2X01_CS8900_VIRT_BASE) && - (port <= IXDP2X01_CS8900_VIRT_END))) { + (((u32)port >= (u32)IXDP2X01_CS8900_VIRT_BASE) && + ((u32)port <= (u32)IXDP2X01_CS8900_VIRT_END))) { return (u16)(*port); } @@ -137,8 +137,8 @@ static inline void outw(u16 value, u32 ptr) register volatile u32 *port = (volatile u32 *)ptr; if ((machine_is_ixdp2401() || machine_is_ixdp2801()) && - ((port >= IXDP2X01_CS8900_VIRT_BASE) && - (port <= IXDP2X01_CS8900_VIRT_END))) { + (((u32)port >= (u32)IXDP2X01_CS8900_VIRT_BASE) && + ((u32)port <= (u32)IXDP2X01_CS8900_VIRT_END))) { *port = value; return; } -- cgit v1.2.3 From 718a30a5cf0a9142f716a49853bd4b4a25a8da1b Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Fri, 3 Jun 2005 20:52:26 +0100 Subject: [PATCH] ARM: 2696/1: remove ';' in ELF_DATA define in asm-arm{,26}/elf.h Patch from Mike Frysinger the ELF_DATA define in both arm asm subdirs of linux/include/ contain a semicolon at the end. this of course will cause any code that tries to use ELF_DATA in assignment or comparison to fail. no other arch has a semicolon in their ELF_DATA defines. Signed-off-by: Mike Frysinger Signed-off-by: Russell King --- include/asm-arm/elf.h | 4 ++-- include/asm-arm26/elf.h | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/asm-arm/elf.h b/include/asm-arm/elf.h index cbceacbe5afa..a1696ba238d3 100644 --- a/include/asm-arm/elf.h +++ b/include/asm-arm/elf.h @@ -38,9 +38,9 @@ typedef struct user_fp elf_fpregset_t; */ #define ELF_CLASS ELFCLASS32 #ifdef __ARMEB__ -#define ELF_DATA ELFDATA2MSB; +#define ELF_DATA ELFDATA2MSB #else -#define ELF_DATA ELFDATA2LSB; +#define ELF_DATA ELFDATA2LSB #endif #define ELF_ARCH EM_ARM diff --git a/include/asm-arm26/elf.h b/include/asm-arm26/elf.h index 8b149474db24..5a47fdb3015d 100644 --- a/include/asm-arm26/elf.h +++ b/include/asm-arm26/elf.h @@ -36,7 +36,7 @@ typedef struct { void *null; } elf_fpregset_t; * These are used to set parameters in the core dumps. */ #define ELF_CLASS ELFCLASS32 -#define ELF_DATA ELFDATA2LSB; +#define ELF_DATA ELFDATA2LSB #define ELF_ARCH EM_ARM #define USE_ELF_CORE_DUMP -- cgit v1.2.3 From c5c3a6d8fe923b8780b9cd10e72344b8cf8518b5 Mon Sep 17 00:00:00 2001 From: Bodo Stroesser Date: Sat, 4 Jun 2005 15:43:32 -0700 Subject: [PATCH] s390: uml ptrace fixes To make UML build and run on s390, I needed to do these two little changes: 1) UML includes some of the subarch's (s390) headers. I had to change one of them with the following one-liner, to make this compile. AFAICS, this change doesn't break compilation of s390 itself. 2) UML needs to intercept syscalls via ptrace to invalidate the syscall, read syscall's parameters and write the result with the result of UML's syscall processing. Also, UML needs to make sure, that the host does no syscall restart processing. On i386 for example, this can be done by writing -1 to orig_eax on the 2nd syscall interception (orig_eax is the syscall number, which after the interception is used as a "interrupt was a syscall" flag only. Unfortunately, s390 holds syscall number and syscall result in gpr2 and its "interrupt was a syscall" flag (trap) is unreachable via ptrace. So I changed the host to set trap to -1, if the syscall number is changed to an invalid value on the first syscall interception. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/s390/kernel/ptrace.c | 7 +++++++ include/asm-s390/user.h | 2 +- 2 files changed, 8 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index 329d9391c83d..06afa3103ace 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -760,6 +760,13 @@ syscall_trace(struct pt_regs *regs, int entryexit) ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD) ? 0x80 : 0)); + /* + * If the debuffer has set an invalid system call number, + * we prepare to skip the system call restart handling. + */ + if (!entryexit && regs->gprs[2] >= NR_syscalls) + regs->trap = -1; + /* * this isn't the same as continuing with a signal, but it will do * for normal use. strace only continues with a signal if the diff --git a/include/asm-s390/user.h b/include/asm-s390/user.h index c64f8c181df3..1dc74baf03c4 100644 --- a/include/asm-s390/user.h +++ b/include/asm-s390/user.h @@ -10,7 +10,7 @@ #define _S390_USER_H #include -#include +#include /* Core file format: The core file is written in such a way that gdb can understand it and provide useful information to the user (under linux we use the 'trad-core' bfd). There are quite a number of -- cgit v1.2.3 From 8e2894e51d6407e47226a60c0d19bf384642c55a Mon Sep 17 00:00:00 2001 From: Yoshinori Sato Date: Mon, 6 Jun 2005 13:35:56 -0700 Subject: [PATCH] h8300 build error fix h8300 was missing a few definitions. Signed-off-by: Yoshinori Sato Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-h8300/kmap_types.h | 6 ++++-- include/asm-h8300/mman.h | 3 +++ 2 files changed, 7 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-h8300/kmap_types.h b/include/asm-h8300/kmap_types.h index 82431edeb2a1..1ec8a3427120 100644 --- a/include/asm-h8300/kmap_types.h +++ b/include/asm-h8300/kmap_types.h @@ -1,5 +1,5 @@ -#ifndef _ASM_KMAP_TYPES_H -#define _ASM_KMAP_TYPES_H +#ifndef _ASM_H8300_KMAP_TYPES_H +#define _ASM_H8300_KMAP_TYPES_H enum km_type { KM_BOUNCE_READ, @@ -13,6 +13,8 @@ enum km_type { KM_PTE1, KM_IRQ0, KM_IRQ1, + KM_SOFTIRQ0, + KM_SOFTIRQ1, KM_TYPE_NR }; diff --git a/include/asm-h8300/mman.h b/include/asm-h8300/mman.h index abe08856c84f..63f727a59850 100644 --- a/include/asm-h8300/mman.h +++ b/include/asm-h8300/mman.h @@ -4,6 +4,7 @@ #define PROT_READ 0x1 /* page can be read */ #define PROT_WRITE 0x2 /* page can be written */ #define PROT_EXEC 0x4 /* page can be executed */ +#define PROT_SEM 0x8 /* page may be used for atomic ops */ #define PROT_NONE 0x0 /* page can not be accessed */ #define PROT_GROWSDOWN 0x01000000 /* mprotect flag: extend change to start of growsdown vma */ #define PROT_GROWSUP 0x02000000 /* mprotect flag: extend change to end of growsup vma */ @@ -19,6 +20,8 @@ #define MAP_EXECUTABLE 0x1000 /* mark it as an executable */ #define MAP_LOCKED 0x2000 /* pages are locked */ #define MAP_NORESERVE 0x4000 /* don't check for reservations */ +#define MAP_POPULATE 0x8000 /* populate (prefault) pagetables */ +#define MAP_NONBLOCK 0x10000 /* do not block on IO */ #define MS_ASYNC 1 /* sync memory asynchronously */ #define MS_INVALIDATE 2 /* invalidate the caches */ -- cgit v1.2.3 From 74262de5d15fe0cab9e6fe75655a502d3c90acca Mon Sep 17 00:00:00 2001 From: Tom Rini Date: Mon, 6 Jun 2005 15:50:08 -0700 Subject: [PATCH] ppc32: add to On ppc32, uses __user, but doesn't directly include . This adds that in. Without this, glibc will not compile. Signed-off-by: Tom Rini Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-ppc/sigcontext.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-ppc/sigcontext.h b/include/asm-ppc/sigcontext.h index f82dcccdee1e..b7a417e0a921 100644 --- a/include/asm-ppc/sigcontext.h +++ b/include/asm-ppc/sigcontext.h @@ -2,7 +2,7 @@ #define _ASM_PPC_SIGCONTEXT_H #include - +#include struct sigcontext { unsigned long _unused[4]; -- cgit v1.2.3 From 3f5948fa2cbbda1261eec9a39ef3004b3caf73fb Mon Sep 17 00:00:00 2001 From: David Mosberger Date: Mon, 6 Jun 2005 15:50:09 -0700 Subject: [PATCH] Include before testing CONFIG_ACPI I'm not sure why this issue is suddenly showing, but without this patchlet, the zx1 config won't compile anymore (e.g., to see the compilation-error, look for "***" in [1]). [1] http://www.gelato.unsw.edu.au/kerncomp/results//2005-06-06-17-00/zx1_defconfig-log.html Signed-off-by: David Mosberger-Tang Cc: "Brown, Len" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/acpi.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/acpi.h b/include/linux/acpi.h index d5a55bdb9c3c..b123cc08773d 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -25,6 +25,8 @@ #ifndef _LINUX_ACPI_H #define _LINUX_ACPI_H +#include + #ifdef CONFIG_ACPI #ifndef _LINUX -- cgit v1.2.3 From d0de98fa16169562bd74913c6c9b3857f9065c79 Mon Sep 17 00:00:00 2001 From: Alan Hourihane Date: Tue, 31 May 2005 19:50:49 +0100 Subject: [PATCH] i945G patch for agpgart Attached is a small patch for i945G support against 2.6.11.11. From: Alan Hourihane Signed-off-by: Dave Jones --- drivers/char/agp/intel-agp.c | 15 +++++++++++++-- include/linux/pci_ids.h | 2 ++ 2 files changed, 15 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c index 8c7d727432bb..6a5047e0d333 100644 --- a/drivers/char/agp/intel-agp.c +++ b/drivers/char/agp/intel-agp.c @@ -418,7 +418,8 @@ static void intel_i830_init_gtt_entries(void) case I915_GMCH_GMS_STOLEN_48M: /* Check it's really I915G */ if (agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82915G_HB || - agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82915GM_HB) + agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82915GM_HB || + agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82945G_HB) gtt_entries = MB(48) - KB(size); else gtt_entries = 0; @@ -426,7 +427,8 @@ static void intel_i830_init_gtt_entries(void) case I915_GMCH_GMS_STOLEN_64M: /* Check it's really I915G */ if (agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82915G_HB || - agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82915GM_HB) + agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82915GM_HB || + agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_82945G_HB) gtt_entries = MB(64) - KB(size); else gtt_entries = 0; @@ -1662,6 +1664,14 @@ static int __devinit agp_intel_probe(struct pci_dev *pdev, } name = "915GM"; break; + case PCI_DEVICE_ID_INTEL_82945G_HB: + if (find_i830(PCI_DEVICE_ID_INTEL_82945G_IG)) { + bridge->driver = &intel_915_driver; + } else { + bridge->driver = &intel_845_driver; + } + name = "945G"; + break; case PCI_DEVICE_ID_INTEL_7505_0: bridge->driver = &intel_7505_driver; name = "E7505"; @@ -1801,6 +1811,7 @@ static struct pci_device_id agp_intel_pci_table[] = { ID(PCI_DEVICE_ID_INTEL_7205_0), ID(PCI_DEVICE_ID_INTEL_82915G_HB), ID(PCI_DEVICE_ID_INTEL_82915GM_HB), + ID(PCI_DEVICE_ID_INTEL_82945G_HB), { } }; diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index b0d6134e1ee6..18f734ec9181 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2382,6 +2382,8 @@ #define PCI_DEVICE_ID_INTEL_82915G_IG 0x2582 #define PCI_DEVICE_ID_INTEL_82915GM_HB 0x2590 #define PCI_DEVICE_ID_INTEL_82915GM_IG 0x2592 +#define PCI_DEVICE_ID_INTEL_82945G_HB 0x2770 +#define PCI_DEVICE_ID_INTEL_82945G_IG 0x2772 #define PCI_DEVICE_ID_INTEL_ICH6_0 0x2640 #define PCI_DEVICE_ID_INTEL_ICH6_1 0x2641 #define PCI_DEVICE_ID_INTEL_ICH6_2 0x2642 -- cgit v1.2.3 From 07eee78ea8ba2d0b7b20551c35a3e7dd158d50bb Mon Sep 17 00:00:00 2001 From: Keir Fraser Date: Wed, 30 Mar 2005 13:17:04 -0800 Subject: [PATCH] AGP fix for Xen VMM When Linux is running on the Xen virtual machine monitor, physical addresses are virtualised and cannot be directly referenced by the AGP GART. This patch fixes the GART driver for Xen by adding a layer of abstraction between physical addresses and 'GART addresses'. Architecture-specific functions are also defined for allocating and freeing the GATT. Xen requires this to ensure that table really is contiguous from the point of view of the GART. These extra interface functions are defined as 'no-ops' for all existing architectures that use the GART driver. Signed-off-by: Keir Fraser Signed-off-by: Andrew Morton Signed-off-by: Dave Jones --- drivers/char/agp/agp.h | 2 ++ drivers/char/agp/ali-agp.c | 4 ++-- drivers/char/agp/amd-k7-agp.c | 6 +++--- drivers/char/agp/amd64-agp.c | 4 ++-- drivers/char/agp/ati-agp.c | 6 +++--- drivers/char/agp/backend.c | 6 +++--- drivers/char/agp/efficeon-agp.c | 2 +- drivers/char/agp/generic.c | 17 ++++++++--------- drivers/char/agp/hp-agp.c | 4 ++-- drivers/char/agp/i460-agp.c | 4 ++-- drivers/char/agp/intel-agp.c | 6 +++--- drivers/char/agp/sworks-agp.c | 8 ++++---- drivers/char/agp/uninorth-agp.c | 2 +- include/asm-alpha/agp.h | 10 ++++++++++ include/asm-i386/agp.h | 10 ++++++++++ include/asm-ia64/agp.h | 10 ++++++++++ include/asm-ppc/agp.h | 10 ++++++++++ include/asm-ppc64/agp.h | 10 ++++++++++ include/asm-sparc64/agp.h | 10 ++++++++++ include/asm-x86_64/agp.h | 10 ++++++++++ 20 files changed, 106 insertions(+), 35 deletions(-) (limited to 'include') diff --git a/drivers/char/agp/agp.h b/drivers/char/agp/agp.h index ad9c11391d81..c1fe013c64f3 100644 --- a/drivers/char/agp/agp.h +++ b/drivers/char/agp/agp.h @@ -278,6 +278,8 @@ void agp3_generic_cleanup(void); #define AGP_GENERIC_SIZES_ENTRIES 11 extern struct aper_size_info_16 agp3_generic_sizes[]; +#define virt_to_gart(x) (phys_to_gart(virt_to_phys(x))) +#define gart_to_virt(x) (phys_to_virt(gart_to_phys(x))) extern int agp_off; extern int agp_try_unsupported_boot; diff --git a/drivers/char/agp/ali-agp.c b/drivers/char/agp/ali-agp.c index 0212febda654..9c9c9c2247ce 100644 --- a/drivers/char/agp/ali-agp.c +++ b/drivers/char/agp/ali-agp.c @@ -150,7 +150,7 @@ static void *m1541_alloc_page(struct agp_bridge_data *bridge) pci_read_config_dword(agp_bridge->dev, ALI_CACHE_FLUSH_CTRL, &temp); pci_write_config_dword(agp_bridge->dev, ALI_CACHE_FLUSH_CTRL, (((temp & ALI_CACHE_FLUSH_ADDR_MASK) | - virt_to_phys(addr)) | ALI_CACHE_FLUSH_EN )); + virt_to_gart(addr)) | ALI_CACHE_FLUSH_EN )); return addr; } @@ -174,7 +174,7 @@ static void m1541_destroy_page(void * addr) pci_read_config_dword(agp_bridge->dev, ALI_CACHE_FLUSH_CTRL, &temp); pci_write_config_dword(agp_bridge->dev, ALI_CACHE_FLUSH_CTRL, (((temp & ALI_CACHE_FLUSH_ADDR_MASK) | - virt_to_phys(addr)) | ALI_CACHE_FLUSH_EN)); + virt_to_gart(addr)) | ALI_CACHE_FLUSH_EN)); agp_generic_destroy_page(addr); } diff --git a/drivers/char/agp/amd-k7-agp.c b/drivers/char/agp/amd-k7-agp.c index e62a3c2c44a9..3a41672e4d66 100644 --- a/drivers/char/agp/amd-k7-agp.c +++ b/drivers/char/agp/amd-k7-agp.c @@ -43,7 +43,7 @@ static int amd_create_page_map(struct amd_page_map *page_map) SetPageReserved(virt_to_page(page_map->real)); global_cache_flush(); - page_map->remapped = ioremap_nocache(virt_to_phys(page_map->real), + page_map->remapped = ioremap_nocache(virt_to_gart(page_map->real), PAGE_SIZE); if (page_map->remapped == NULL) { ClearPageReserved(virt_to_page(page_map->real)); @@ -154,7 +154,7 @@ static int amd_create_gatt_table(struct agp_bridge_data *bridge) agp_bridge->gatt_table_real = (u32 *)page_dir.real; agp_bridge->gatt_table = (u32 __iomem *)page_dir.remapped; - agp_bridge->gatt_bus_addr = virt_to_phys(page_dir.real); + agp_bridge->gatt_bus_addr = virt_to_gart(page_dir.real); /* Get the address for the gart region. * This is a bus address even on the alpha, b/c its @@ -167,7 +167,7 @@ static int amd_create_gatt_table(struct agp_bridge_data *bridge) /* Calculate the agp offset */ for (i = 0; i < value->num_entries / 1024; i++, addr += 0x00400000) { - writel(virt_to_phys(amd_irongate_private.gatt_pages[i]->real) | 1, + writel(virt_to_gart(amd_irongate_private.gatt_pages[i]->real) | 1, page_dir.remapped+GET_PAGE_DIR_OFF(addr)); readl(page_dir.remapped+GET_PAGE_DIR_OFF(addr)); /* PCI Posting. */ } diff --git a/drivers/char/agp/amd64-agp.c b/drivers/char/agp/amd64-agp.c index 399c042f68f0..1407945a5892 100644 --- a/drivers/char/agp/amd64-agp.c +++ b/drivers/char/agp/amd64-agp.c @@ -219,7 +219,7 @@ static struct aper_size_info_32 amd_8151_sizes[7] = static int amd_8151_configure(void) { - unsigned long gatt_bus = virt_to_phys(agp_bridge->gatt_table_real); + unsigned long gatt_bus = virt_to_gart(agp_bridge->gatt_table_real); /* Configure AGP regs in each x86-64 host bridge. */ for_each_nb() { @@ -591,7 +591,7 @@ static void __devexit agp_amd64_remove(struct pci_dev *pdev) { struct agp_bridge_data *bridge = pci_get_drvdata(pdev); - release_mem_region(virt_to_phys(bridge->gatt_table_real), + release_mem_region(virt_to_gart(bridge->gatt_table_real), amd64_aperture_sizes[bridge->aperture_size_idx].size); agp_remove_bridge(bridge); agp_put_bridge(bridge); diff --git a/drivers/char/agp/ati-agp.c b/drivers/char/agp/ati-agp.c index a65f8827c283..e572ced9100a 100644 --- a/drivers/char/agp/ati-agp.c +++ b/drivers/char/agp/ati-agp.c @@ -61,7 +61,7 @@ static int ati_create_page_map(ati_page_map *page_map) SetPageReserved(virt_to_page(page_map->real)); err = map_page_into_agp(virt_to_page(page_map->real)); - page_map->remapped = ioremap_nocache(virt_to_phys(page_map->real), + page_map->remapped = ioremap_nocache(virt_to_gart(page_map->real), PAGE_SIZE); if (page_map->remapped == NULL || err) { ClearPageReserved(virt_to_page(page_map->real)); @@ -343,7 +343,7 @@ static int ati_create_gatt_table(struct agp_bridge_data *bridge) agp_bridge->gatt_table_real = (u32 *)page_dir.real; agp_bridge->gatt_table = (u32 __iomem *) page_dir.remapped; - agp_bridge->gatt_bus_addr = virt_to_bus(page_dir.real); + agp_bridge->gatt_bus_addr = virt_to_gart(page_dir.real); /* Write out the size register */ current_size = A_SIZE_LVL2(agp_bridge->current_size); @@ -373,7 +373,7 @@ static int ati_create_gatt_table(struct agp_bridge_data *bridge) /* Calculate the agp offset */ for(i = 0; i < value->num_entries / 1024; i++, addr += 0x00400000) { - writel(virt_to_bus(ati_generic_private.gatt_pages[i]->real) | 1, + writel(virt_to_gart(ati_generic_private.gatt_pages[i]->real) | 1, page_dir.remapped+GET_PAGE_DIR_OFF(addr)); readl(page_dir.remapped+GET_PAGE_DIR_OFF(addr)); /* PCI Posting. */ } diff --git a/drivers/char/agp/backend.c b/drivers/char/agp/backend.c index 2f3dfb63bdc6..4d4e602fdc7e 100644 --- a/drivers/char/agp/backend.c +++ b/drivers/char/agp/backend.c @@ -148,7 +148,7 @@ static int agp_backend_initialize(struct agp_bridge_data *bridge) return -ENOMEM; } - bridge->scratch_page_real = virt_to_phys(addr); + bridge->scratch_page_real = virt_to_gart(addr); bridge->scratch_page = bridge->driver->mask_memory(bridge, bridge->scratch_page_real, 0); } @@ -189,7 +189,7 @@ static int agp_backend_initialize(struct agp_bridge_data *bridge) err_out: if (bridge->driver->needs_scratch_page) bridge->driver->agp_destroy_page( - phys_to_virt(bridge->scratch_page_real)); + gart_to_virt(bridge->scratch_page_real)); if (got_gatt) bridge->driver->free_gatt_table(bridge); if (got_keylist) { @@ -214,7 +214,7 @@ static void agp_backend_cleanup(struct agp_bridge_data *bridge) if (bridge->driver->agp_destroy_page && bridge->driver->needs_scratch_page) bridge->driver->agp_destroy_page( - phys_to_virt(bridge->scratch_page_real)); + gart_to_virt(bridge->scratch_page_real)); } /* When we remove the global variable agp_bridge from all drivers diff --git a/drivers/char/agp/efficeon-agp.c b/drivers/char/agp/efficeon-agp.c index 1383c3165ea1..ac19fdcd21c1 100644 --- a/drivers/char/agp/efficeon-agp.c +++ b/drivers/char/agp/efficeon-agp.c @@ -219,7 +219,7 @@ static int efficeon_create_gatt_table(struct agp_bridge_data *bridge) efficeon_private.l1_table[index] = page; - value = __pa(page) | pati | present | index; + value = virt_to_gart(page) | pati | present | index; pci_write_config_dword(agp_bridge->dev, EFFICEON_ATTPAGE, value); diff --git a/drivers/char/agp/generic.c b/drivers/char/agp/generic.c index c321a924e38a..d62505b5d25a 100644 --- a/drivers/char/agp/generic.c +++ b/drivers/char/agp/generic.c @@ -153,7 +153,7 @@ void agp_free_memory(struct agp_memory *curr) } if (curr->page_count != 0) { for (i = 0; i < curr->page_count; i++) { - curr->bridge->driver->agp_destroy_page(phys_to_virt(curr->memory[i])); + curr->bridge->driver->agp_destroy_page(gart_to_virt(curr->memory[i])); } } agp_free_key(curr->key); @@ -209,7 +209,7 @@ struct agp_memory *agp_allocate_memory(struct agp_bridge_data *bridge, agp_free_memory(new); return NULL; } - new->memory[i] = virt_to_phys(addr); + new->memory[i] = virt_to_gart(addr); new->page_count++; } new->bridge = bridge; @@ -806,8 +806,7 @@ int agp_generic_create_gatt_table(struct agp_bridge_data *bridge) break; } - table = (char *) __get_free_pages(GFP_KERNEL, - page_order); + table = alloc_gatt_pages(page_order); if (table == NULL) { i++; @@ -838,7 +837,7 @@ int agp_generic_create_gatt_table(struct agp_bridge_data *bridge) size = ((struct aper_size_info_fixed *) temp)->size; page_order = ((struct aper_size_info_fixed *) temp)->page_order; num_entries = ((struct aper_size_info_fixed *) temp)->num_entries; - table = (char *) __get_free_pages(GFP_KERNEL, page_order); + table = alloc_gatt_pages(page_order); } if (table == NULL) @@ -853,7 +852,7 @@ int agp_generic_create_gatt_table(struct agp_bridge_data *bridge) agp_gatt_table = (void *)table; bridge->driver->cache_flush(); - bridge->gatt_table = ioremap_nocache(virt_to_phys(table), + bridge->gatt_table = ioremap_nocache(virt_to_gart(table), (PAGE_SIZE * (1 << page_order))); bridge->driver->cache_flush(); @@ -861,11 +860,11 @@ int agp_generic_create_gatt_table(struct agp_bridge_data *bridge) for (page = virt_to_page(table); page <= virt_to_page(table_end); page++) ClearPageReserved(page); - free_pages((unsigned long) table, page_order); + free_gatt_pages(table, page_order); return -ENOMEM; } - bridge->gatt_bus_addr = virt_to_phys(bridge->gatt_table_real); + bridge->gatt_bus_addr = virt_to_gart(bridge->gatt_table_real); /* AK: bogus, should encode addresses > 4GB */ for (i = 0; i < num_entries; i++) { @@ -919,7 +918,7 @@ int agp_generic_free_gatt_table(struct agp_bridge_data *bridge) for (page = virt_to_page(table); page <= virt_to_page(table_end); page++) ClearPageReserved(page); - free_pages((unsigned long) bridge->gatt_table_real, page_order); + free_gatt_pages(bridge->gatt_table_real, page_order); agp_gatt_table = NULL; bridge->gatt_table = NULL; diff --git a/drivers/char/agp/hp-agp.c b/drivers/char/agp/hp-agp.c index 6052bfa04c72..99762b6c19ae 100644 --- a/drivers/char/agp/hp-agp.c +++ b/drivers/char/agp/hp-agp.c @@ -110,7 +110,7 @@ static int __init hp_zx1_ioc_shared(void) hp->gart_size = HP_ZX1_GART_SIZE; hp->gatt_entries = hp->gart_size / hp->io_page_size; - hp->io_pdir = phys_to_virt(readq(hp->ioc_regs+HP_ZX1_PDIR_BASE)); + hp->io_pdir = gart_to_virt(readq(hp->ioc_regs+HP_ZX1_PDIR_BASE)); hp->gatt = &hp->io_pdir[HP_ZX1_IOVA_TO_PDIR(hp->gart_base)]; if (hp->gatt[0] != HP_ZX1_SBA_IOMMU_COOKIE) { @@ -248,7 +248,7 @@ hp_zx1_configure (void) agp_bridge->mode = readl(hp->lba_regs+hp->lba_cap_offset+PCI_AGP_STATUS); if (hp->io_pdir_owner) { - writel(virt_to_phys(hp->io_pdir), hp->ioc_regs+HP_ZX1_PDIR_BASE); + writel(virt_to_gart(hp->io_pdir), hp->ioc_regs+HP_ZX1_PDIR_BASE); readl(hp->ioc_regs+HP_ZX1_PDIR_BASE); writel(hp->io_tlb_ps, hp->ioc_regs+HP_ZX1_TCNFG); readl(hp->ioc_regs+HP_ZX1_TCNFG); diff --git a/drivers/char/agp/i460-agp.c b/drivers/char/agp/i460-agp.c index adbea896c0d2..94943298c03e 100644 --- a/drivers/char/agp/i460-agp.c +++ b/drivers/char/agp/i460-agp.c @@ -372,7 +372,7 @@ static int i460_alloc_large_page (struct lp_desc *lp) } memset(lp->alloced_map, 0, map_size); - lp->paddr = virt_to_phys(lpage); + lp->paddr = virt_to_gart(lpage); lp->refcount = 0; atomic_add(I460_KPAGES_PER_IOPAGE, &agp_bridge->current_memory_agp); return 0; @@ -383,7 +383,7 @@ static void i460_free_large_page (struct lp_desc *lp) kfree(lp->alloced_map); lp->alloced_map = NULL; - free_pages((unsigned long) phys_to_virt(lp->paddr), I460_IO_PAGE_SHIFT - PAGE_SHIFT); + free_pages((unsigned long) gart_to_virt(lp->paddr), I460_IO_PAGE_SHIFT - PAGE_SHIFT); atomic_sub(I460_KPAGES_PER_IOPAGE, &agp_bridge->current_memory_agp); } diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c index 6a5047e0d333..51266d6b4d78 100644 --- a/drivers/char/agp/intel-agp.c +++ b/drivers/char/agp/intel-agp.c @@ -286,7 +286,7 @@ static struct agp_memory *alloc_agpphysmem_i8xx(size_t pg_count, int type) if (new == NULL) return NULL; - new->memory[0] = virt_to_phys(addr); + new->memory[0] = virt_to_gart(addr); if (pg_count == 4) { /* kludge to get 4 physical pages for ARGB cursor */ new->memory[1] = new->memory[0] + PAGE_SIZE; @@ -329,10 +329,10 @@ static void intel_i810_free_by_type(struct agp_memory *curr) agp_free_key(curr->key); if(curr->type == AGP_PHYS_MEMORY) { if (curr->page_count == 4) - i8xx_destroy_pages(phys_to_virt(curr->memory[0])); + i8xx_destroy_pages(gart_to_virt(curr->memory[0])); else agp_bridge->driver->agp_destroy_page( - phys_to_virt(curr->memory[0])); + gart_to_virt(curr->memory[0])); vfree(curr->memory); } kfree(curr); diff --git a/drivers/char/agp/sworks-agp.c b/drivers/char/agp/sworks-agp.c index 10c23302dd84..a9fb12c20eb7 100644 --- a/drivers/char/agp/sworks-agp.c +++ b/drivers/char/agp/sworks-agp.c @@ -51,7 +51,7 @@ static int serverworks_create_page_map(struct serverworks_page_map *page_map) } SetPageReserved(virt_to_page(page_map->real)); global_cache_flush(); - page_map->remapped = ioremap_nocache(virt_to_phys(page_map->real), + page_map->remapped = ioremap_nocache(virt_to_gart(page_map->real), PAGE_SIZE); if (page_map->remapped == NULL) { ClearPageReserved(virt_to_page(page_map->real)); @@ -162,7 +162,7 @@ static int serverworks_create_gatt_table(struct agp_bridge_data *bridge) /* Create a fake scratch directory */ for(i = 0; i < 1024; i++) { writel(agp_bridge->scratch_page, serverworks_private.scratch_dir.remapped+i); - writel(virt_to_phys(serverworks_private.scratch_dir.real) | 1, page_dir.remapped+i); + writel(virt_to_gart(serverworks_private.scratch_dir.real) | 1, page_dir.remapped+i); } retval = serverworks_create_gatt_pages(value->num_entries / 1024); @@ -174,7 +174,7 @@ static int serverworks_create_gatt_table(struct agp_bridge_data *bridge) agp_bridge->gatt_table_real = (u32 *)page_dir.real; agp_bridge->gatt_table = (u32 __iomem *)page_dir.remapped; - agp_bridge->gatt_bus_addr = virt_to_phys(page_dir.real); + agp_bridge->gatt_bus_addr = virt_to_gart(page_dir.real); /* Get the address for the gart region. * This is a bus address even on the alpha, b/c its @@ -187,7 +187,7 @@ static int serverworks_create_gatt_table(struct agp_bridge_data *bridge) /* Calculate the agp offset */ for(i = 0; i < value->num_entries / 1024; i++) - writel(virt_to_phys(serverworks_private.gatt_pages[i]->real)|1, page_dir.remapped+i); + writel(virt_to_gart(serverworks_private.gatt_pages[i]->real)|1, page_dir.remapped+i); return 0; } diff --git a/drivers/char/agp/uninorth-agp.c b/drivers/char/agp/uninorth-agp.c index a673971f2a90..c8255312b8c1 100644 --- a/drivers/char/agp/uninorth-agp.c +++ b/drivers/char/agp/uninorth-agp.c @@ -407,7 +407,7 @@ static int uninorth_create_gatt_table(struct agp_bridge_data *bridge) bridge->gatt_table_real = (u32 *) table; bridge->gatt_table = (u32 *)table; - bridge->gatt_bus_addr = virt_to_phys(table); + bridge->gatt_bus_addr = virt_to_gart(table); for (i = 0; i < num_entries; i++) bridge->gatt_table[i] = 0; diff --git a/include/asm-alpha/agp.h b/include/asm-alpha/agp.h index c99dbbb5bcb5..ef855a3bc0f5 100644 --- a/include/asm-alpha/agp.h +++ b/include/asm-alpha/agp.h @@ -10,4 +10,14 @@ #define flush_agp_mappings() #define flush_agp_cache() mb() +/* Convert a physical address to an address suitable for the GART. */ +#define phys_to_gart(x) (x) +#define gart_to_phys(x) (x) + +/* GATT allocation. Returns/accepts GATT kernel virtual address. */ +#define alloc_gatt_pages(order) \ + ((char *)__get_free_pages(GFP_KERNEL, (order))) +#define free_gatt_pages(table, order) \ + free_pages((unsigned long)(table), (order)) + #endif diff --git a/include/asm-i386/agp.h b/include/asm-i386/agp.h index a917ff50354f..b82f5f3ab887 100644 --- a/include/asm-i386/agp.h +++ b/include/asm-i386/agp.h @@ -21,4 +21,14 @@ int unmap_page_from_agp(struct page *page); worth it. Would need a page for it. */ #define flush_agp_cache() asm volatile("wbinvd":::"memory") +/* Convert a physical address to an address suitable for the GART. */ +#define phys_to_gart(x) (x) +#define gart_to_phys(x) (x) + +/* GATT allocation. Returns/accepts GATT kernel virtual address. */ +#define alloc_gatt_pages(order) \ + ((char *)__get_free_pages(GFP_KERNEL, (order))) +#define free_gatt_pages(table, order) \ + free_pages((unsigned long)(table), (order)) + #endif diff --git a/include/asm-ia64/agp.h b/include/asm-ia64/agp.h index d1316f1e6ee1..4e517f0e6afa 100644 --- a/include/asm-ia64/agp.h +++ b/include/asm-ia64/agp.h @@ -18,4 +18,14 @@ #define flush_agp_mappings() /* nothing */ #define flush_agp_cache() mb() +/* Convert a physical address to an address suitable for the GART. */ +#define phys_to_gart(x) (x) +#define gart_to_phys(x) (x) + +/* GATT allocation. Returns/accepts GATT kernel virtual address. */ +#define alloc_gatt_pages(order) \ + ((char *)__get_free_pages(GFP_KERNEL, (order))) +#define free_gatt_pages(table, order) \ + free_pages((unsigned long)(table), (order)) + #endif /* _ASM_IA64_AGP_H */ diff --git a/include/asm-ppc/agp.h b/include/asm-ppc/agp.h index be27cfa8c5b0..ca9e423307f4 100644 --- a/include/asm-ppc/agp.h +++ b/include/asm-ppc/agp.h @@ -10,4 +10,14 @@ #define flush_agp_mappings() #define flush_agp_cache() mb() +/* Convert a physical address to an address suitable for the GART. */ +#define phys_to_gart(x) (x) +#define gart_to_phys(x) (x) + +/* GATT allocation. Returns/accepts GATT kernel virtual address. */ +#define alloc_gatt_pages(order) \ + ((char *)__get_free_pages(GFP_KERNEL, (order))) +#define free_gatt_pages(table, order) \ + free_pages((unsigned long)(table), (order)) + #endif diff --git a/include/asm-ppc64/agp.h b/include/asm-ppc64/agp.h index be27cfa8c5b0..ca9e423307f4 100644 --- a/include/asm-ppc64/agp.h +++ b/include/asm-ppc64/agp.h @@ -10,4 +10,14 @@ #define flush_agp_mappings() #define flush_agp_cache() mb() +/* Convert a physical address to an address suitable for the GART. */ +#define phys_to_gart(x) (x) +#define gart_to_phys(x) (x) + +/* GATT allocation. Returns/accepts GATT kernel virtual address. */ +#define alloc_gatt_pages(order) \ + ((char *)__get_free_pages(GFP_KERNEL, (order))) +#define free_gatt_pages(table, order) \ + free_pages((unsigned long)(table), (order)) + #endif diff --git a/include/asm-sparc64/agp.h b/include/asm-sparc64/agp.h index ba05bdf9a211..58f8cb6ae767 100644 --- a/include/asm-sparc64/agp.h +++ b/include/asm-sparc64/agp.h @@ -8,4 +8,14 @@ #define flush_agp_mappings() #define flush_agp_cache() mb() +/* Convert a physical address to an address suitable for the GART. */ +#define phys_to_gart(x) (x) +#define gart_to_phys(x) (x) + +/* GATT allocation. Returns/accepts GATT kernel virtual address. */ +#define alloc_gatt_pages(order) \ + ((char *)__get_free_pages(GFP_KERNEL, (order))) +#define free_gatt_pages(table, order) \ + free_pages((unsigned long)(table), (order)) + #endif diff --git a/include/asm-x86_64/agp.h b/include/asm-x86_64/agp.h index 0bb9019d58aa..06c52ee9c06b 100644 --- a/include/asm-x86_64/agp.h +++ b/include/asm-x86_64/agp.h @@ -19,4 +19,14 @@ int unmap_page_from_agp(struct page *page); worth it. Would need a page for it. */ #define flush_agp_cache() asm volatile("wbinvd":::"memory") +/* Convert a physical address to an address suitable for the GART. */ +#define phys_to_gart(x) (x) +#define gart_to_phys(x) (x) + +/* GATT allocation. Returns/accepts GATT kernel virtual address. */ +#define alloc_gatt_pages(order) \ + ((char *)__get_free_pages(GFP_KERNEL, (order))) +#define free_gatt_pages(table, order) \ + free_pages((unsigned long)(table), (order)) + #endif -- cgit v1.2.3 From eda9937656e0b9879ca521140fe61cbc9788c398 Mon Sep 17 00:00:00 2001 From: Matthew Dobson Date: Tue, 7 Jun 2005 13:22:05 -0700 Subject: [PATCH] send_IPI_mask_sequence() warning fix In file included from arch/i386/kernel/smp.c:235: include/asm-i386/mach-numaq/mach_ipi.h:4: warning: `send_IPI_mask_sequence' declared inline after its definition Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-i386/mach-numaq/mach_ipi.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-i386/mach-numaq/mach_ipi.h b/include/asm-i386/mach-numaq/mach_ipi.h index 1b46fd3f2ae3..c6044488e9e6 100644 --- a/include/asm-i386/mach-numaq/mach_ipi.h +++ b/include/asm-i386/mach-numaq/mach_ipi.h @@ -1,7 +1,7 @@ #ifndef __ASM_MACH_IPI_H #define __ASM_MACH_IPI_H -inline void send_IPI_mask_sequence(cpumask_t, int vector); +void send_IPI_mask_sequence(cpumask_t, int vector); static inline void send_IPI_mask(cpumask_t mask, int vector) { -- cgit v1.2.3 From ad597bd518559f59ede8d01262cdf4467e13282e Mon Sep 17 00:00:00 2001 From: David Mosberger-Tang Date: Wed, 8 Jun 2005 10:45:00 -0700 Subject: [IA64] Fill holes in FIXADDR_USER space with zero pages. This fixes an oops reported by Jason Baron. Signed-off-by: David Mosberger-Tang Signed-off-by: Tony Luck --- arch/ia64/mm/init.c | 19 +++++++++++++++++-- include/asm-ia64/pgtable.h | 8 ++++++-- 2 files changed, 23 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index 547785e3cba2..4eb2f52b87a1 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -305,8 +305,9 @@ setup_gate (void) struct page *page; /* - * Map the gate page twice: once read-only to export the ELF headers etc. and once - * execute-only page to enable privilege-promotion via "epc": + * Map the gate page twice: once read-only to export the ELF + * headers etc. and once execute-only page to enable + * privilege-promotion via "epc": */ page = virt_to_page(ia64_imva(__start_gate_section)); put_kernel_page(page, GATE_ADDR, PAGE_READONLY); @@ -315,6 +316,20 @@ setup_gate (void) put_kernel_page(page, GATE_ADDR + PAGE_SIZE, PAGE_GATE); #else put_kernel_page(page, GATE_ADDR + PERCPU_PAGE_SIZE, PAGE_GATE); + /* Fill in the holes (if any) with read-only zero pages: */ + { + unsigned long addr; + + for (addr = GATE_ADDR + PAGE_SIZE; + addr < GATE_ADDR + PERCPU_PAGE_SIZE; + addr += PAGE_SIZE) + { + put_kernel_page(ZERO_PAGE(0), addr, + PAGE_READONLY); + put_kernel_page(ZERO_PAGE(0), addr + PERCPU_PAGE_SIZE, + PAGE_READONLY); + } + } #endif ia64_patch_gate(); } diff --git a/include/asm-ia64/pgtable.h b/include/asm-ia64/pgtable.h index ea121a002309..fcc9c3344ab4 100644 --- a/include/asm-ia64/pgtable.h +++ b/include/asm-ia64/pgtable.h @@ -8,7 +8,7 @@ * This hopefully works with any (fixed) IA-64 page-size, as defined * in . * - * Copyright (C) 1998-2004 Hewlett-Packard Co + * Copyright (C) 1998-2005 Hewlett-Packard Co * David Mosberger-Tang */ @@ -551,7 +551,11 @@ do { \ /* These tell get_user_pages() that the first gate page is accessible from user-level. */ #define FIXADDR_USER_START GATE_ADDR -#define FIXADDR_USER_END (GATE_ADDR + 2*PERCPU_PAGE_SIZE) +#ifdef HAVE_BUGGY_SEGREL +# define FIXADDR_USER_END (GATE_ADDR + 2*PAGE_SIZE) +#else +# define FIXADDR_USER_END (GATE_ADDR + 2*PERCPU_PAGE_SIZE) +#endif #define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG #define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY -- cgit v1.2.3 From 57cfa5e97ff60d909a7e0a5f6e4d0a49aa3e2aec Mon Sep 17 00:00:00 2001 From: Giorgio Padrin Date: Wed, 8 Jun 2005 19:00:15 +0100 Subject: [PATCH] ARM: 2703/1: pxa-regs.h: complete I2S GPIO alternate functions for PXA27x Patch from Giorgio Padrin The patch completes I2S GPIO alternate functions for PXA27x, adding I2S_SYSCLK. File: pxa-regs.h . Signed-off-by: Giorgio Padrin Signed-off-by: Russell King --- include/asm-arm/arch-pxa/pxa-regs.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/asm-arm/arch-pxa/pxa-regs.h b/include/asm-arm/arch-pxa/pxa-regs.h index 39741d3c9a34..b5e54a9e9fa7 100644 --- a/include/asm-arm/arch-pxa/pxa-regs.h +++ b/include/asm-arm/arch-pxa/pxa-regs.h @@ -1296,6 +1296,7 @@ #define GPIO111_MMCDAT3 111 /* MMC DAT3 (PXA27x) */ #define GPIO111_MMCCS1 111 /* MMC Chip Select 1 (PXA27x) */ #define GPIO112_MMCCMD 112 /* MMC CMD (PXA27x) */ +#define GPIO113_I2S_SYSCLK 113 /* I2S System Clock (PXA27x) */ #define GPIO113_AC97_RESET_N 113 /* AC97 NRESET on (PXA27x) */ /* GPIO alternate function mode & direction */ @@ -1428,6 +1429,7 @@ #define GPIO111_MMCDAT3_MD (111 | GPIO_ALT_FN_1_OUT) #define GPIO110_MMCCS1_MD (111 | GPIO_ALT_FN_1_OUT) #define GPIO112_MMCCMD_MD (112 | GPIO_ALT_FN_1_OUT) +#define GPIO113_I2S_SYSCLK_MD (113 | GPIO_ALT_FN_1_OUT) #define GPIO113_AC97_RESET_N_MD (113 | GPIO_ALT_FN_2_OUT) #define GPIO117_I2CSCL_MD (117 | GPIO_ALT_FN_1_OUT) #define GPIO118_I2CSDA_MD (118 | GPIO_ALT_FN_1_IN) -- cgit v1.2.3 From 6d1cfbab4de64f2d0c5b0f81177ade0d75b69288 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Wed, 8 Jun 2005 14:13:14 -0700 Subject: [TG3]: Fix 5700/5701 DMA corruption on Apple G4. Fix 5700/5701 DMA write corruption on Apple G4 by detecting the Apple UniNorth PCI 1.5 chipset and adjusting the DMA write boundary to 16. DMA test fails to detect the problem with this chipset. Thanks to Manuel Perez Ayala for reporting the problem and helping to debug it. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/tg3.c | 18 ++++++++++++++++-- include/linux/pci_ids.h | 1 + 2 files changed, 17 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index e944aac258e3..77337c3b4285 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -9695,10 +9695,24 @@ static int __devinit tg3_test_dma(struct tg3 *tp) } if ((tp->dma_rwctrl & DMA_RWCTRL_WRITE_BNDRY_MASK) != DMA_RWCTRL_WRITE_BNDRY_16) { + static struct pci_device_id dma_wait_state_chipsets[] = { + { PCI_DEVICE(PCI_VENDOR_ID_APPLE, + PCI_DEVICE_ID_APPLE_UNI_N_PCI15) }, + { }, + }; + /* DMA test passed without adjusting DMA boundary, - * just restore the calculated DMA boundary + * now look for chipsets that are known to expose the + * DMA bug without failing the test. */ - tp->dma_rwctrl = saved_dma_rwctrl; + if (pci_dev_present(dma_wait_state_chipsets)) { + tp->dma_rwctrl &= ~DMA_RWCTRL_WRITE_BNDRY_MASK; + tp->dma_rwctrl |= DMA_RWCTRL_WRITE_BNDRY_16; + } + else + /* Safe to use the calculated DMA boundary. */ + tp->dma_rwctrl = saved_dma_rwctrl; + tw32(TG3PCI_DMA_RW_CTRL, tp->dma_rwctrl); } diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 18f734ec9181..b8b4ebf9abf1 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -874,6 +874,7 @@ #define PCI_DEVICE_ID_APPLE_KL_USB_P 0x0026 #define PCI_DEVICE_ID_APPLE_UNI_N_AGP_P 0x0027 #define PCI_DEVICE_ID_APPLE_UNI_N_AGP15 0x002d +#define PCI_DEVICE_ID_APPLE_UNI_N_PCI15 0x002e #define PCI_DEVICE_ID_APPLE_UNI_N_FW2 0x0030 #define PCI_DEVICE_ID_APPLE_UNI_N_GMAC2 0x0032 #define PCI_DEVIEC_ID_APPLE_UNI_N_ATA 0x0033 -- cgit v1.2.3 From 4890062960cbc4d3cebdbd8261a68bc85efcf5d4 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Wed, 8 Jun 2005 15:10:48 -0700 Subject: [PKT_SCHED]: Allow socket attributes to be matched on via meta ematch Adds meta collectors for all socket attributes that make sense to be filtered upon. Some of them are only useful for debugging but having them doesn't hurt. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/tc_ematch/tc_em_meta.h | 30 ++++ net/sched/em_meta.c | 291 ++++++++++++++++++++++++++++++++--- 2 files changed, 297 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/include/linux/tc_ematch/tc_em_meta.h b/include/linux/tc_ematch/tc_em_meta.h index aa6b48bb4dcd..a6b2cc530af5 100644 --- a/include/linux/tc_ematch/tc_em_meta.h +++ b/include/linux/tc_ematch/tc_em_meta.h @@ -56,6 +56,36 @@ enum TCF_META_ID_TCCLASSID, TCF_META_ID_RTCLASSID, TCF_META_ID_RTIIF, + TCF_META_ID_SK_FAMILY, + TCF_META_ID_SK_STATE, + TCF_META_ID_SK_REUSE, + TCF_META_ID_SK_BOUND_IF, + TCF_META_ID_SK_REFCNT, + TCF_META_ID_SK_SHUTDOWN, + TCF_META_ID_SK_PROTO, + TCF_META_ID_SK_TYPE, + TCF_META_ID_SK_RCVBUF, + TCF_META_ID_SK_RMEM_ALLOC, + TCF_META_ID_SK_WMEM_ALLOC, + TCF_META_ID_SK_OMEM_ALLOC, + TCF_META_ID_SK_WMEM_QUEUED, + TCF_META_ID_SK_RCV_QLEN, + TCF_META_ID_SK_SND_QLEN, + TCF_META_ID_SK_ERR_QLEN, + TCF_META_ID_SK_FORWARD_ALLOCS, + TCF_META_ID_SK_SNDBUF, + TCF_META_ID_SK_ALLOCS, + TCF_META_ID_SK_ROUTE_CAPS, + TCF_META_ID_SK_HASHENT, + TCF_META_ID_SK_LINGERTIME, + TCF_META_ID_SK_ACK_BACKLOG, + TCF_META_ID_SK_MAX_ACK_BACKLOG, + TCF_META_ID_SK_PRIO, + TCF_META_ID_SK_RCVLOWAT, + TCF_META_ID_SK_RCVTIMEO, + TCF_META_ID_SK_SNDTIMEO, + TCF_META_ID_SK_SENDMSG_OFF, + TCF_META_ID_SK_WRITE_PENDING, __TCF_META_ID_MAX }; #define TCF_META_ID_MAX (__TCF_META_ID_MAX - 1) diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index f1eeaf65cee5..ed2a46cbb67f 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -32,7 +32,7 @@ * +-----------+ +-----------+ * | | * ---> meta_ops[INT][INDEV](...) | - * | | + * | | * ----------- | * V V * +-----------+ +-----------+ @@ -70,6 +70,7 @@ #include #include #include +#include struct meta_obj { @@ -283,6 +284,214 @@ META_COLLECTOR(int_rtiif) dst->value = ((struct rtable*) skb->dst)->fl.iif; } +/************************************************************************** + * Socket Attributes + **************************************************************************/ + +#define SKIP_NONLOCAL(skb) \ + if (unlikely(skb->sk == NULL)) { \ + *err = -1; \ + return; \ + } + +META_COLLECTOR(int_sk_family) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_family; +} + +META_COLLECTOR(int_sk_state) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_state; +} + +META_COLLECTOR(int_sk_reuse) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_reuse; +} + +META_COLLECTOR(int_sk_bound_if) +{ + SKIP_NONLOCAL(skb); + /* No error if bound_dev_if is 0, legal userspace check */ + dst->value = skb->sk->sk_bound_dev_if; +} + +META_COLLECTOR(var_sk_bound_if) +{ + SKIP_NONLOCAL(skb); + + if (skb->sk->sk_bound_dev_if == 0) { + dst->value = (unsigned long) "any"; + dst->len = 3; + } else { + struct net_device *dev; + + dev = dev_get_by_index(skb->sk->sk_bound_dev_if); + *err = var_dev(dev, dst); + if (dev) + dev_put(dev); + } +} + +META_COLLECTOR(int_sk_refcnt) +{ + SKIP_NONLOCAL(skb); + dst->value = atomic_read(&skb->sk->sk_refcnt); +} + +META_COLLECTOR(int_sk_rcvbuf) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_rcvbuf; +} + +META_COLLECTOR(int_sk_shutdown) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_shutdown; +} + +META_COLLECTOR(int_sk_proto) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_protocol; +} + +META_COLLECTOR(int_sk_type) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_type; +} + +META_COLLECTOR(int_sk_rmem_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = atomic_read(&skb->sk->sk_rmem_alloc); +} + +META_COLLECTOR(int_sk_wmem_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = atomic_read(&skb->sk->sk_wmem_alloc); +} + +META_COLLECTOR(int_sk_omem_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = atomic_read(&skb->sk->sk_omem_alloc); +} + +META_COLLECTOR(int_sk_rcv_qlen) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_receive_queue.qlen; +} + +META_COLLECTOR(int_sk_snd_qlen) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_write_queue.qlen; +} + +META_COLLECTOR(int_sk_wmem_queued) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_wmem_queued; +} + +META_COLLECTOR(int_sk_fwd_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_forward_alloc; +} + +META_COLLECTOR(int_sk_sndbuf) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_sndbuf; +} + +META_COLLECTOR(int_sk_alloc) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_allocation; +} + +META_COLLECTOR(int_sk_route_caps) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_route_caps; +} + +META_COLLECTOR(int_sk_hashent) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_hashent; +} + +META_COLLECTOR(int_sk_lingertime) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_lingertime / HZ; +} + +META_COLLECTOR(int_sk_err_qlen) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_error_queue.qlen; +} + +META_COLLECTOR(int_sk_ack_bl) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_ack_backlog; +} + +META_COLLECTOR(int_sk_max_ack_bl) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_max_ack_backlog; +} + +META_COLLECTOR(int_sk_prio) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_priority; +} + +META_COLLECTOR(int_sk_rcvlowat) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_rcvlowat; +} + +META_COLLECTOR(int_sk_rcvtimeo) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_rcvtimeo / HZ; +} + +META_COLLECTOR(int_sk_sndtimeo) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_sndtimeo / HZ; +} + +META_COLLECTOR(int_sk_sendmsg_off) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_sndmsg_off; +} + +META_COLLECTOR(int_sk_write_pend) +{ + SKIP_NONLOCAL(skb); + dst->value = skb->sk->sk_write_pending; +} + /************************************************************************** * Meta value collectors assignment table **************************************************************************/ @@ -293,41 +502,75 @@ struct meta_ops struct meta_value *, struct meta_obj *, int *); }; +#define META_ID(name) TCF_META_ID_##name +#define META_FUNC(name) { .get = meta_##name } + /* Meta value operations table listing all meta value collectors and * assigns them to a type and meta id. */ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = { [TCF_META_TYPE_VAR] = { - [TCF_META_ID_DEV] = { .get = meta_var_dev }, - [TCF_META_ID_INDEV] = { .get = meta_var_indev }, - [TCF_META_ID_REALDEV] = { .get = meta_var_realdev } + [META_ID(DEV)] = META_FUNC(var_dev), + [META_ID(INDEV)] = META_FUNC(var_indev), + [META_ID(REALDEV)] = META_FUNC(var_realdev), + [META_ID(SK_BOUND_IF)] = META_FUNC(var_sk_bound_if), }, [TCF_META_TYPE_INT] = { - [TCF_META_ID_RANDOM] = { .get = meta_int_random }, - [TCF_META_ID_LOADAVG_0] = { .get = meta_int_loadavg_0 }, - [TCF_META_ID_LOADAVG_1] = { .get = meta_int_loadavg_1 }, - [TCF_META_ID_LOADAVG_2] = { .get = meta_int_loadavg_2 }, - [TCF_META_ID_DEV] = { .get = meta_int_dev }, - [TCF_META_ID_INDEV] = { .get = meta_int_indev }, - [TCF_META_ID_REALDEV] = { .get = meta_int_realdev }, - [TCF_META_ID_PRIORITY] = { .get = meta_int_priority }, - [TCF_META_ID_PROTOCOL] = { .get = meta_int_protocol }, - [TCF_META_ID_SECURITY] = { .get = meta_int_security }, - [TCF_META_ID_PKTTYPE] = { .get = meta_int_pkttype }, - [TCF_META_ID_PKTLEN] = { .get = meta_int_pktlen }, - [TCF_META_ID_DATALEN] = { .get = meta_int_datalen }, - [TCF_META_ID_MACLEN] = { .get = meta_int_maclen }, + [META_ID(RANDOM)] = META_FUNC(int_random), + [META_ID(LOADAVG_0)] = META_FUNC(int_loadavg_0), + [META_ID(LOADAVG_1)] = META_FUNC(int_loadavg_1), + [META_ID(LOADAVG_2)] = META_FUNC(int_loadavg_2), + [META_ID(DEV)] = META_FUNC(int_dev), + [META_ID(INDEV)] = META_FUNC(int_indev), + [META_ID(REALDEV)] = META_FUNC(int_realdev), + [META_ID(PRIORITY)] = META_FUNC(int_priority), + [META_ID(PROTOCOL)] = META_FUNC(int_protocol), + [META_ID(SECURITY)] = META_FUNC(int_security), + [META_ID(PKTTYPE)] = META_FUNC(int_pkttype), + [META_ID(PKTLEN)] = META_FUNC(int_pktlen), + [META_ID(DATALEN)] = META_FUNC(int_datalen), + [META_ID(MACLEN)] = META_FUNC(int_maclen), #ifdef CONFIG_NETFILTER - [TCF_META_ID_NFMARK] = { .get = meta_int_nfmark }, + [META_ID(NFMARK)] = META_FUNC(int_nfmark), #endif - [TCF_META_ID_TCINDEX] = { .get = meta_int_tcindex }, + [META_ID(TCINDEX)] = META_FUNC(int_tcindex), #ifdef CONFIG_NET_CLS_ACT - [TCF_META_ID_TCVERDICT] = { .get = meta_int_tcverd }, - [TCF_META_ID_TCCLASSID] = { .get = meta_int_tcclassid }, + [META_ID(TCVERDICT)] = META_FUNC(int_tcverd), + [META_ID(TCCLASSID)] = META_FUNC(int_tcclassid), #endif #ifdef CONFIG_NET_CLS_ROUTE - [TCF_META_ID_RTCLASSID] = { .get = meta_int_rtclassid }, + [META_ID(RTCLASSID)] = META_FUNC(int_rtclassid), #endif - [TCF_META_ID_RTIIF] = { .get = meta_int_rtiif } + [META_ID(RTIIF)] = META_FUNC(int_rtiif), + [META_ID(SK_FAMILY)] = META_FUNC(int_sk_family), + [META_ID(SK_STATE)] = META_FUNC(int_sk_state), + [META_ID(SK_REUSE)] = META_FUNC(int_sk_reuse), + [META_ID(SK_BOUND_IF)] = META_FUNC(int_sk_bound_if), + [META_ID(SK_REFCNT)] = META_FUNC(int_sk_refcnt), + [META_ID(SK_RCVBUF)] = META_FUNC(int_sk_rcvbuf), + [META_ID(SK_SNDBUF)] = META_FUNC(int_sk_sndbuf), + [META_ID(SK_SHUTDOWN)] = META_FUNC(int_sk_shutdown), + [META_ID(SK_PROTO)] = META_FUNC(int_sk_proto), + [META_ID(SK_TYPE)] = META_FUNC(int_sk_type), + [META_ID(SK_RMEM_ALLOC)] = META_FUNC(int_sk_rmem_alloc), + [META_ID(SK_WMEM_ALLOC)] = META_FUNC(int_sk_wmem_alloc), + [META_ID(SK_OMEM_ALLOC)] = META_FUNC(int_sk_omem_alloc), + [META_ID(SK_WMEM_QUEUED)] = META_FUNC(int_sk_wmem_queued), + [META_ID(SK_RCV_QLEN)] = META_FUNC(int_sk_rcv_qlen), + [META_ID(SK_SND_QLEN)] = META_FUNC(int_sk_snd_qlen), + [META_ID(SK_ERR_QLEN)] = META_FUNC(int_sk_err_qlen), + [META_ID(SK_FORWARD_ALLOCS)] = META_FUNC(int_sk_fwd_alloc), + [META_ID(SK_ALLOCS)] = META_FUNC(int_sk_alloc), + [META_ID(SK_ROUTE_CAPS)] = META_FUNC(int_sk_route_caps), + [META_ID(SK_HASHENT)] = META_FUNC(int_sk_hashent), + [META_ID(SK_LINGERTIME)] = META_FUNC(int_sk_lingertime), + [META_ID(SK_ACK_BACKLOG)] = META_FUNC(int_sk_ack_bl), + [META_ID(SK_MAX_ACK_BACKLOG)] = META_FUNC(int_sk_max_ack_bl), + [META_ID(SK_PRIO)] = META_FUNC(int_sk_prio), + [META_ID(SK_RCVLOWAT)] = META_FUNC(int_sk_rcvlowat), + [META_ID(SK_RCVTIMEO)] = META_FUNC(int_sk_rcvtimeo), + [META_ID(SK_SNDTIMEO)] = META_FUNC(int_sk_sndtimeo), + [META_ID(SK_SENDMSG_OFF)] = META_FUNC(int_sk_sendmsg_off), + [META_ID(SK_WRITE_PENDING)] = META_FUNC(int_sk_write_pend), } }; -- cgit v1.2.3 From f8acd944ea511af02485b1709c07ac7aac12aa48 Mon Sep 17 00:00:00 2001 From: William Lee Irwin III Date: Wed, 8 Jun 2005 15:48:52 -0700 Subject: [PATCH] sparc32: silence access_ok() warnings The fact that access_ok() doesn't use some of its arguments trips some unused variable warnings. This patch silences them permanently. Signed-off-by: William Irwin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-sparc/uaccess.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-sparc/uaccess.h b/include/asm-sparc/uaccess.h index f461144067ee..0a780e84a12b 100644 --- a/include/asm-sparc/uaccess.h +++ b/include/asm-sparc/uaccess.h @@ -41,10 +41,11 @@ * No one can read/write anything from userland in the kernel space by setting * large size and address near to PAGE_OFFSET - a fault will break his intentions. */ -#define __user_ok(addr,size) ((addr) < STACK_TOP) +#define __user_ok(addr, size) ({ (void)(size); (addr) < STACK_TOP; }) #define __kernel_ok (segment_eq(get_fs(), KERNEL_DS)) #define __access_ok(addr,size) (__user_ok((addr) & get_fs().seg,(size))) -#define access_ok(type,addr,size) __access_ok((unsigned long)(addr),(size)) +#define access_ok(type, addr, size) \ + ({ (void)(type); __access_ok((unsigned long)(addr), size); }) /* this function will go away soon - use access_ok() instead */ static inline int __deprecated verify_area(int type, const void __user * addr, unsigned long size) -- cgit v1.2.3 From 05062d96a23ec0959ee5ea969f40813170c73c0e Mon Sep 17 00:00:00 2001 From: Peter Chubb Date: Wed, 8 Jun 2005 15:50:20 -0700 Subject: [PATCH] ia64: fix floating-point preemption problem There've been reports of problems with CONFIG_PREEMPT=y and the high floating point partition. This is caused by the possibility of preemption and rescheduling on a different processor while saving or restioirng the high partition. The only places where the FPU state is touched are in ptrace, in switch_to(), and where handling a floating-point exception. In switch_to() preemption is off. So it's only in trap.c and ptrace.c that we need to prevent preemption. Here is a patch that adds commentary to make the conditions clear, and adds appropriate preempt_{en,dis}able() calls to make it so. In trap.c I use preempt_enable_no_resched(), as we're about to return to user space where the preemption flag will be checked anyway. Signed-off-by: Peter Chubb Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/kernel/ptrace.c | 6 ++++++ arch/ia64/kernel/traps.c | 11 ++++++++++- include/asm-ia64/processor.h | 10 ++++++++-- 3 files changed, 24 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c index 08c8a5eb25ab..575a8f657b31 100644 --- a/arch/ia64/kernel/ptrace.c +++ b/arch/ia64/kernel/ptrace.c @@ -635,11 +635,17 @@ ia64_flush_fph (struct task_struct *task) { struct ia64_psr *psr = ia64_psr(ia64_task_regs(task)); + /* + * Prevent migrating this task while + * we're fiddling with the FPU state + */ + preempt_disable(); if (ia64_is_local_fpu_owner(task) && psr->mfh) { psr->mfh = 0; task->thread.flags |= IA64_THREAD_FPH_VALID; ia64_save_fpu(&task->thread.fph[0]); } + preempt_enable(); } /* diff --git a/arch/ia64/kernel/traps.c b/arch/ia64/kernel/traps.c index 9bad6652d531..1861173bd4f6 100644 --- a/arch/ia64/kernel/traps.c +++ b/arch/ia64/kernel/traps.c @@ -220,13 +220,21 @@ disabled_fph_fault (struct pt_regs *regs) /* first, grant user-level access to fph partition: */ psr->dfh = 0; + + /* + * Make sure that no other task gets in on this processor + * while we're claiming the FPU + */ + preempt_disable(); #ifndef CONFIG_SMP { struct task_struct *fpu_owner = (struct task_struct *)ia64_get_kr(IA64_KR_FPU_OWNER); - if (ia64_is_local_fpu_owner(current)) + if (ia64_is_local_fpu_owner(current)) { + preempt_enable_no_resched(); return; + } if (fpu_owner) ia64_flush_fph(fpu_owner); @@ -244,6 +252,7 @@ disabled_fph_fault (struct pt_regs *regs) */ psr->mfh = 1; } + preempt_enable_no_resched(); } static inline int diff --git a/include/asm-ia64/processor.h b/include/asm-ia64/processor.h index 9e1ba8b7fb68..91bbd1f22461 100644 --- a/include/asm-ia64/processor.h +++ b/include/asm-ia64/processor.h @@ -403,7 +403,10 @@ extern void ia64_setreg_unknown_kr (void); * task_struct at this point. */ -/* Return TRUE if task T owns the fph partition of the CPU we're running on. */ +/* + * Return TRUE if task T owns the fph partition of the CPU we're running on. + * Must be called from code that has preemption disabled. + */ #define ia64_is_local_fpu_owner(t) \ ({ \ struct task_struct *__ia64_islfo_task = (t); \ @@ -411,7 +414,10 @@ extern void ia64_setreg_unknown_kr (void); && __ia64_islfo_task == (struct task_struct *) ia64_get_kr(IA64_KR_FPU_OWNER)); \ }) -/* Mark task T as owning the fph partition of the CPU we're running on. */ +/* + * Mark task T as owning the fph partition of the CPU we're running on. + * Must be called from code that has preemption disabled. + */ #define ia64_set_local_fpu_owner(t) do { \ struct task_struct *__ia64_slfo_task = (t); \ __ia64_slfo_task->thread.last_fph_cpu = smp_processor_id(); \ -- cgit v1.2.3 From ce10d979053379553757c3b178a138facaddff82 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Wed, 8 Jun 2005 21:59:15 +1000 Subject: [PATCH] ppc64: Fix PER_LINUX32 behaviour This patch fixes some bugs in the ppc64 PER_LINUX32 implementation, noted by Juergen Kreileder: * uname(2) doesn't respect PER_LINUX32, it returns 'ppc64' instead of 'ppc' * Child processes of a PER_LINUX32 process don't inherit PER_LINUX32 Along the way I took the opportunity to move things around so that sys_ppc32.c only has 32-bit syscall emulation functions and to remove the obsolete "fakeppc" command line option. Signed-off-by: Paul Mackerras Signed-off-by: Linus Torvalds --- arch/ppc64/kernel/misc.S | 2 +- arch/ppc64/kernel/sys_ppc32.c | 70 ++++++++++++++++++++----------------------- arch/ppc64/kernel/syscalls.c | 33 ++++++++++++-------- include/asm-ppc64/elf.h | 4 +-- 4 files changed, 56 insertions(+), 53 deletions(-) (limited to 'include') diff --git a/arch/ppc64/kernel/misc.S b/arch/ppc64/kernel/misc.S index b944717c1dbd..e3c73b3425dc 100644 --- a/arch/ppc64/kernel/misc.S +++ b/arch/ppc64/kernel/misc.S @@ -792,7 +792,7 @@ _GLOBAL(sys_call_table32) .llong .compat_sys_newstat .llong .compat_sys_newlstat .llong .compat_sys_newfstat - .llong .sys_uname + .llong .sys32_uname .llong .sys_ni_syscall /* 110 old iopl syscall */ .llong .sys_vhangup .llong .sys_ni_syscall /* old idle syscall */ diff --git a/arch/ppc64/kernel/sys_ppc32.c b/arch/ppc64/kernel/sys_ppc32.c index 7cf7a9600025..9c8e317c598d 100644 --- a/arch/ppc64/kernel/sys_ppc32.c +++ b/arch/ppc64/kernel/sys_ppc32.c @@ -791,31 +791,6 @@ asmlinkage int sys32_pciconfig_iobase(u32 which, u32 in_bus, u32 in_devfn) } -asmlinkage int ppc64_newuname(struct new_utsname __user * name) -{ - int errno = sys_newuname(name); - - if (current->personality == PER_LINUX32 && !errno) { - if(copy_to_user(name->machine, "ppc\0\0", 8)) { - errno = -EFAULT; - } - } - return errno; -} - -asmlinkage int ppc64_personality(unsigned long personality) -{ - int ret; - if (current->personality == PER_LINUX32 && personality == PER_LINUX) - personality = PER_LINUX32; - ret = sys_personality(personality); - if (ret == PER_LINUX32) - ret = PER_LINUX; - return ret; -} - - - /* Note: it is necessary to treat mode as an unsigned int, * with the corresponding cast to a signed int to insure that the * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) @@ -1158,26 +1133,47 @@ asmlinkage long sys32_sysctl(struct __sysctl_args32 __user *args) } #endif +asmlinkage int sys32_uname(struct old_utsname __user * name) +{ + int err = 0; + + down_read(&uts_sem); + if (copy_to_user(name, &system_utsname, sizeof(*name))) + err = -EFAULT; + up_read(&uts_sem); + if (!err && personality(current->personality) == PER_LINUX32) { + /* change "ppc64" to "ppc" */ + if (__put_user(0, name->machine + 3) + || __put_user(0, name->machine + 4)) + err = -EFAULT; + } + return err; +} + asmlinkage int sys32_olduname(struct oldold_utsname __user * name) { int error; - - if (!name) - return -EFAULT; + if (!access_ok(VERIFY_WRITE,name,sizeof(struct oldold_utsname))) return -EFAULT; down_read(&uts_sem); error = __copy_to_user(&name->sysname,&system_utsname.sysname,__OLD_UTS_LEN); - error -= __put_user(0,name->sysname+__OLD_UTS_LEN); - error -= __copy_to_user(&name->nodename,&system_utsname.nodename,__OLD_UTS_LEN); - error -= __put_user(0,name->nodename+__OLD_UTS_LEN); - error -= __copy_to_user(&name->release,&system_utsname.release,__OLD_UTS_LEN); - error -= __put_user(0,name->release+__OLD_UTS_LEN); - error -= __copy_to_user(&name->version,&system_utsname.version,__OLD_UTS_LEN); - error -= __put_user(0,name->version+__OLD_UTS_LEN); - error -= __copy_to_user(&name->machine,&system_utsname.machine,__OLD_UTS_LEN); - error = __put_user(0,name->machine+__OLD_UTS_LEN); + error |= __put_user(0,name->sysname+__OLD_UTS_LEN); + error |= __copy_to_user(&name->nodename,&system_utsname.nodename,__OLD_UTS_LEN); + error |= __put_user(0,name->nodename+__OLD_UTS_LEN); + error |= __copy_to_user(&name->release,&system_utsname.release,__OLD_UTS_LEN); + error |= __put_user(0,name->release+__OLD_UTS_LEN); + error |= __copy_to_user(&name->version,&system_utsname.version,__OLD_UTS_LEN); + error |= __put_user(0,name->version+__OLD_UTS_LEN); + error |= __copy_to_user(&name->machine,&system_utsname.machine,__OLD_UTS_LEN); + error |= __put_user(0,name->machine+__OLD_UTS_LEN); + if (personality(current->personality) == PER_LINUX32) { + /* change "ppc64" to "ppc" */ + error |= __put_user(0, name->machine + 3); + error |= __put_user(0, name->machine + 4); + } + up_read(&uts_sem); error = error ? -EFAULT : 0; diff --git a/arch/ppc64/kernel/syscalls.c b/arch/ppc64/kernel/syscalls.c index f2865ff8d2f9..a8cbb202b8cd 100644 --- a/arch/ppc64/kernel/syscalls.c +++ b/arch/ppc64/kernel/syscalls.c @@ -199,24 +199,33 @@ out: return ret; } -static int __init set_fakeppc(char *str) +long ppc64_personality(unsigned long personality) { - if (*str) - return 0; - init_task.personality = PER_LINUX32; - return 1; + long ret; + + if (personality(current->personality) == PER_LINUX32 + && personality == PER_LINUX) + personality = PER_LINUX32; + ret = sys_personality(personality); + if (ret == PER_LINUX32) + ret = PER_LINUX; + return ret; } -__setup("fakeppc", set_fakeppc); -asmlinkage int sys_uname(struct old_utsname __user * name) +long ppc64_newuname(struct new_utsname __user * name) { - int err = -EFAULT; - + int err = 0; + down_read(&uts_sem); - if (name && !copy_to_user(name, &system_utsname, sizeof (*name))) - err = 0; + if (copy_to_user(name, &system_utsname, sizeof(*name))) + err = -EFAULT; up_read(&uts_sem); - + if (!err && personality(current->personality) == PER_LINUX32) { + /* change ppc64 to ppc */ + if (__put_user(0, name->machine + 3) + || __put_user(0, name->machine + 4)) + err = -EFAULT; + } return err; } diff --git a/include/asm-ppc64/elf.h b/include/asm-ppc64/elf.h index 6c42d61bedd1..085eedb956fe 100644 --- a/include/asm-ppc64/elf.h +++ b/include/asm-ppc64/elf.h @@ -221,9 +221,7 @@ do { \ set_thread_flag(TIF_ABI_PENDING); \ else \ clear_thread_flag(TIF_ABI_PENDING); \ - if (ibcs2) \ - set_personality(PER_SVR4); \ - else if (current->personality != PER_LINUX32) \ + if (personality(current->personality) != PER_LINUX32) \ set_personality(PER_LINUX); \ } while (0) -- cgit v1.2.3 From a58e76f25432dc5e3e84d04c27bec03347ca365b Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Sun, 12 Jun 2005 10:56:26 +0200 Subject: [PATCH] Remove obsolete HAVE_ARCH_GET_SIGNAL_TO_DELIVER? Now m68k no longer sets HAVE_ARCH_GET_SIGNAL_TO_DELIVER, can it be removed completely? Or may ARM26 still need it? Note that its usage was removed from kernel/signal.c about 2 months ago. Signed-off-by: Geert Uytterhoeven Signed-off-by: Linus Torvalds --- include/asm-arm26/signal.h | 3 --- include/linux/signal.h | 2 -- 2 files changed, 5 deletions(-) (limited to 'include') diff --git a/include/asm-arm26/signal.h b/include/asm-arm26/signal.h index dedb29280303..37ad25355591 100644 --- a/include/asm-arm26/signal.h +++ b/include/asm-arm26/signal.h @@ -166,9 +166,6 @@ typedef struct sigaltstack { #include #define sigmask(sig) (1UL << ((sig) - 1)) -//FIXME!!! -//#define HAVE_ARCH_GET_SIGNAL_TO_DELIVER - #endif diff --git a/include/linux/signal.h b/include/linux/signal.h index 0a98f5ec5cae..7be18b5e2fb4 100644 --- a/include/linux/signal.h +++ b/include/linux/signal.h @@ -231,10 +231,8 @@ extern int __group_send_sig_info(int, struct siginfo *, struct task_struct *); extern long do_sigpending(void __user *, unsigned long); extern int sigprocmask(int, sigset_t *, sigset_t *); -#ifndef HAVE_ARCH_GET_SIGNAL_TO_DELIVER struct pt_regs; extern int get_signal_to_deliver(siginfo_t *info, struct k_sigaction *return_ka, struct pt_regs *regs, void *cookie); -#endif #endif /* __KERNEL__ */ -- cgit v1.2.3 From 03722adce90a248d0bea77d390decbd05991e2d2 Mon Sep 17 00:00:00 2001 From: Tom Rini Date: Mon, 13 Jun 2005 13:57:10 -0700 Subject: [NET]: linux/if_tr.h needs asm/byteorder.h uses __be16, but does not directly include . Add this in, so that dhcp/net-tools token ring code can compile again. Signed-off-by: Tom Rini Signed-off-by: David S. Miller --- include/linux/if_tr.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/if_tr.h b/include/linux/if_tr.h index 6688b414c529..3fba9e2f5427 100644 --- a/include/linux/if_tr.h +++ b/include/linux/if_tr.h @@ -19,6 +19,8 @@ #ifndef _LINUX_IF_TR_H #define _LINUX_IF_TR_H +#include /* For __be16 */ + /* IEEE 802.5 Token-Ring magic constants. The frame sizes omit the preamble and FCS/CRC (frame check sequence). */ #define TR_ALEN 6 /* Octets in one token-ring addr */ -- cgit v1.2.3 From cdac4e07748934e37e415437055ed591aed9eb21 Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Mon, 13 Jun 2005 15:12:33 -0700 Subject: [SCTP] Add support for ip_nonlocal_bind sysctl & IP_FREEBIND socket option Signed-off-by: Neil Horman Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- include/net/ip.h | 1 + net/ipv4/af_inet.c | 1 + net/sctp/protocol.c | 7 +++++-- 3 files changed, 7 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/ip.h b/include/net/ip.h index 3f63992eb712..32360bbe143f 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -163,6 +163,7 @@ DECLARE_SNMP_STAT(struct linux_mib, net_statistics); extern int sysctl_local_port_range[2]; extern int sysctl_ip_default_ttl; +extern int sysctl_ip_nonlocal_bind; #ifdef CONFIG_INET /* The function in 2.2 was invalid, producing wrong result for diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index b3cb49ce5fad..03942f133944 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1181,6 +1181,7 @@ EXPORT_SYMBOL(inet_stream_connect); EXPORT_SYMBOL(inet_stream_ops); EXPORT_SYMBOL(inet_unregister_protosw); EXPORT_SYMBOL(net_statistics); +EXPORT_SYMBOL(sysctl_ip_nonlocal_bind); #ifdef INET_REFCNT_DEBUG EXPORT_SYMBOL(inet_sock_nr); diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 2e1f9c3556f5..5135e1a25d25 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -378,10 +378,13 @@ static int sctp_v4_available(union sctp_addr *addr, struct sctp_sock *sp) { int ret = inet_addr_type(addr->v4.sin_addr.s_addr); - /* FIXME: ip_nonlocal_bind sysctl support. */ - if (addr->v4.sin_addr.s_addr != INADDR_ANY && ret != RTN_LOCAL) + if (addr->v4.sin_addr.s_addr != INADDR_ANY && + ret != RTN_LOCAL && + !sp->inet.freebind && + !sysctl_ip_nonlocal_bind) return 0; + return 1; } -- cgit v1.2.3 From 1c2fb7f93cb20621772bf304f3dba0849942e5db Mon Sep 17 00:00:00 2001 From: "J. Simonetti" Date: Mon, 13 Jun 2005 15:19:03 -0700 Subject: [IPV4]: Sysctl configurable icmp error source address. This patch alows you to change the source address of icmp error messages. It applies cleanly to 2.6.11.11 and retains the default behaviour. In the old (default) behaviour icmp error messages are sent with the ip of the exiting interface. The new behaviour (when the sysctl variable is toggled on), it will send the message with the ip of the interface that received the packet that caused the icmp error. This is the behaviour network administrators will expect from a router. It makes debugging complicated network layouts much easier. Also, all 'vendor routers' I know of have the later behaviour. Signed-off-by: David S. Miller --- include/linux/sysctl.h | 1 + net/ipv4/icmp.c | 9 +++++++-- net/ipv4/sysctl_net_ipv4.c | 9 +++++++++ 3 files changed, 17 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 23032d9d6071..a17745c80a91 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -346,6 +346,7 @@ enum NET_TCP_MODERATE_RCVBUF=106, NET_TCP_TSO_WIN_DIVISOR=107, NET_TCP_BIC_BETA=108, + NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR=109, }; enum { diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 85bf0d3e294b..cb759484979d 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -207,6 +207,7 @@ int sysctl_icmp_ignore_bogus_error_responses; int sysctl_icmp_ratelimit = 1 * HZ; int sysctl_icmp_ratemask = 0x1818; +int sysctl_icmp_errors_use_inbound_ifaddr; /* * ICMP control array. This specifies what to do with each ICMP. @@ -511,8 +512,12 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, u32 info) */ saddr = iph->daddr; - if (!(rt->rt_flags & RTCF_LOCAL)) - saddr = 0; + if (!(rt->rt_flags & RTCF_LOCAL)) { + if (sysctl_icmp_errors_use_inbound_ifaddr) + saddr = inet_select_addr(skb_in->dev, 0, RT_SCOPE_LINK); + else + saddr = 0; + } tos = icmp_pointers[type].error ? ((iph->tos & IPTOS_TOS_MASK) | IPTOS_PREC_INTERNETCONTROL) : diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 3aafb298c1c1..23068bddbf0b 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -23,6 +23,7 @@ extern int sysctl_ip_nonlocal_bind; extern int sysctl_icmp_echo_ignore_all; extern int sysctl_icmp_echo_ignore_broadcasts; extern int sysctl_icmp_ignore_bogus_error_responses; +extern int sysctl_icmp_errors_use_inbound_ifaddr; /* From ip_fragment.c */ extern int sysctl_ipfrag_low_thresh; @@ -395,6 +396,14 @@ ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = &proc_dointvec }, + { + .ctl_name = NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR, + .procname = "icmp_errors_use_inbound_ifaddr", + .data = &sysctl_icmp_errors_use_inbound_ifaddr, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, { .ctl_name = NET_IPV4_ROUTE, .procname = "route", -- cgit v1.2.3 From fea7722fd7b45c6957caed84251d95269319fe16 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Thu, 16 Jun 2005 18:01:11 +0100 Subject: [PATCH] ARM: 2713/1: Fix the GPIO base for Integrator/CP Patch from Catalin Marinas The GPIO base for Integrator/CP is different from the Integrator/AP. This patch sets the correct value for INTEGRATOR_GPIO_BASE. Signed-off-by: Catalin Marinas Signed-off-by: Russell King --- arch/arm/mach-integrator/integrator_cp.c | 1 - include/asm-arm/arch-integrator/platform.h | 4 ++++ 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/arch/arm/mach-integrator/integrator_cp.c b/arch/arm/mach-integrator/integrator_cp.c index 3b948e8c2751..e0a01eef0993 100644 --- a/arch/arm/mach-integrator/integrator_cp.c +++ b/arch/arm/mach-integrator/integrator_cp.c @@ -83,7 +83,6 @@ static struct map_desc intcp_io_desc[] __initdata = { { IO_ADDRESS(INTEGRATOR_UART1_BASE), INTEGRATOR_UART1_BASE, SZ_4K, MT_DEVICE }, { IO_ADDRESS(INTEGRATOR_DBG_BASE), INTEGRATOR_DBG_BASE, SZ_4K, MT_DEVICE }, { IO_ADDRESS(INTEGRATOR_GPIO_BASE), INTEGRATOR_GPIO_BASE, SZ_4K, MT_DEVICE }, - { 0xfc900000, 0xc9000000, SZ_4K, MT_DEVICE }, { 0xfca00000, 0xca000000, SZ_4K, MT_DEVICE }, { 0xfcb00000, 0xcb000000, SZ_4K, MT_DEVICE }, }; diff --git a/include/asm-arm/arch-integrator/platform.h b/include/asm-arm/arch-integrator/platform.h index bd364f5a99bc..96ad3d2a66d1 100644 --- a/include/asm-arm/arch-integrator/platform.h +++ b/include/asm-arm/arch-integrator/platform.h @@ -293,7 +293,11 @@ #define INTEGRATOR_DBG_SWITCH (INTEGRATOR_DBG_BASE + INTEGRATOR_DBG_SWITCH_OFFSET) +#if defined(CONFIG_ARCH_INTEGRATOR_AP) #define INTEGRATOR_GPIO_BASE 0x1B000000 /* GPIO */ +#elif defined(CONFIG_ARCH_INTEGRATOR_CP) +#define INTEGRATOR_GPIO_BASE 0xC9000000 /* GPIO */ +#endif /* ------------------------------------------------------------------------ * KMI keyboard/mouse definitions -- cgit v1.2.3 From 95220a2ea334b5ff2168cc9bf31c0e08b29bae21 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Thu, 16 Jun 2005 18:01:12 +0100 Subject: [PATCH] ARM: 2714/1: Fix the IB2 definitions for the Versatile platform Patch from Catalin Marinas The initial IB2 addresses did not depend on the IB2 base. This patch defines them as (VERSATILE_IB2_BASE + offset). Signed-off-by: Catalin Marinas Signed-off-by: Russell King --- include/asm-arm/arch-versatile/platform.h | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/asm-arm/arch-versatile/platform.h b/include/asm-arm/arch-versatile/platform.h index 2598d1f08548..a71093e44c58 100644 --- a/include/asm-arm/arch-versatile/platform.h +++ b/include/asm-arm/arch-versatile/platform.h @@ -498,11 +498,17 @@ /* * IB2 Versatile/AB expansion board definitions */ -#define VERSATILE_IB2_CAMERA_BANK 0x24000000 -#define VERSATILE_IB2_KBD_DATAREG 0x25000000 -#define VERSATILE_IB2_IER 0x26000000 /* for VICINTSOURCE27 */ -#define VERSATILE_IB2_CTRL 0x27000000 -#define VERSATILE_IB2_STAT 0x27000004 +#define VERSATILE_IB2_CAMERA_BANK VERSATILE_IB2_BASE +#define VERSATILE_IB2_KBD_DATAREG (VERSATILE_IB2_BASE + 0x01000000) + +/* VICINTSOURCE27 */ +#define VERSATILE_IB2_INT_BASE (VERSATILE_IB2_BASE + 0x02000000) +#define VERSATILE_IB2_IER (VERSATILE_IB2_INT_BASE + 0) +#define VERSATILE_IB2_ISR (VERSATILE_IB2_INT_BASE + 4) + +#define VERSATILE_IB2_CTL_BASE (VERSATILE_IB2_BASE + 0x03000000) +#define VERSATILE_IB2_CTRL (VERSATILE_IB2_CTL_BASE + 0) +#define VERSATILE_IB2_STAT (VERSATILE_IB2_CTL_BASE + 4) #endif #endif -- cgit v1.2.3 From b8112df71cae7d6a86158caeb19d215f56c4f9ab Mon Sep 17 00:00:00 2001 From: Lee Revell Date: Wed, 15 Jun 2005 14:19:03 -0400 Subject: [SCSI] Add DMA mask constants other than 32 and 64 bit Signed-Off-By: Lee Revell Signed-off-by: James Bottomley --- include/linux/dma-mapping.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include') diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h index 806c305332c1..2d80cc761a15 100644 --- a/include/linux/dma-mapping.h +++ b/include/linux/dma-mapping.h @@ -14,7 +14,12 @@ enum dma_data_direction { }; #define DMA_64BIT_MASK 0xffffffffffffffffULL +#define DMA_40BIT_MASK 0x000000ffffffffffULL +#define DMA_39BIT_MASK 0x0000007fffffffffULL #define DMA_32BIT_MASK 0x00000000ffffffffULL +#define DMA_31BIT_MASK 0x000000007fffffffULL +#define DMA_30BIT_MASK 0x000000003fffffffULL +#define DMA_29BIT_MASK 0x000000001fffffffULL #include -- cgit v1.2.3 From e65f38ed0bb7af367ff919c573cf29643fc5f9e8 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 18 Jun 2005 09:33:31 +0100 Subject: [PATCH] ARM SMP: Add support for startup of secondary processors Create a temporary page table to startup secondary processors. This page table must have a 1:1 virtual/physical mapping for the kernel in addition to the standard mappings to ensure that the secondary CPU can enable its MMU safely. Signed-off-by: Russell King --- arch/arm/kernel/head.S | 44 ++++++++++++++++++++ arch/arm/kernel/smp.c | 107 +++++++++++++++++++++++++++++++++++++++++++++++++ include/asm-arm/smp.h | 14 +++++++ 3 files changed, 165 insertions(+) (limited to 'include') diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S index 4733877296d4..bd4823c74645 100644 --- a/arch/arm/kernel/head.S +++ b/arch/arm/kernel/head.S @@ -2,6 +2,8 @@ * linux/arch/arm/kernel/head.S * * Copyright (C) 1994-2002 Russell King + * Copyright (c) 2003 ARM Limited + * All Rights Reserved * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -165,6 +167,48 @@ __mmap_switched: stmia r6, {r0, r4} @ Save control register values b start_kernel +#if defined(CONFIG_SMP) + .type secondary_startup, #function +ENTRY(secondary_startup) + /* + * Common entry point for secondary CPUs. + * + * Ensure that we're in SVC mode, and IRQs are disabled. Lookup + * the processor type - there is no need to check the machine type + * as it has already been validated by the primary processor. + */ + msr cpsr_c, #PSR_F_BIT | PSR_I_BIT | MODE_SVC + bl __lookup_processor_type + movs r10, r5 @ invalid processor? + moveq r0, #'p' @ yes, error 'p' + beq __error + + /* + * Use the page tables supplied from __cpu_up. + */ + adr r4, __secondary_data + ldmia r4, {r5, r6, r13} @ address to jump to after + sub r4, r4, r5 @ mmu has been enabled + ldr r4, [r6, r4] @ get secondary_data.pgdir + adr lr, __enable_mmu @ return address + add pc, r10, #12 @ initialise processor + @ (return control reg) + + /* + * r6 = &secondary_data + */ +ENTRY(__secondary_switched) + ldr sp, [r6, #4] @ get secondary_data.stack + mov fp, #0 + b secondary_start_kernel + + .type __secondary_data, %object +__secondary_data: + .long . + .long secondary_data + .long __secondary_switched +#endif /* defined(CONFIG_SMP) */ + /* diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c index ecc8c3332408..45ed036336e0 100644 --- a/arch/arm/kernel/smp.c +++ b/arch/arm/kernel/smp.c @@ -24,6 +24,9 @@ #include #include #include +#include +#include +#include #include #include #include @@ -36,6 +39,13 @@ cpumask_t cpu_present_mask; cpumask_t cpu_online_map; +/* + * as from 2.5, kernels no longer have an init_tasks structure + * so we need some other way of telling a new secondary core + * where to place its SVC stack + */ +struct secondary_data secondary_data; + /* * structures for inter-processor calls * - A collection of single bit ipi messages. @@ -71,6 +81,8 @@ static DEFINE_SPINLOCK(smp_call_function_lock); int __init __cpu_up(unsigned int cpu) { struct task_struct *idle; + pgd_t *pgd; + pmd_t *pmd; int ret; /* @@ -83,10 +95,55 @@ int __init __cpu_up(unsigned int cpu) return PTR_ERR(idle); } + /* + * Allocate initial page tables to allow the new CPU to + * enable the MMU safely. This essentially means a set + * of our "standard" page tables, with the addition of + * a 1:1 mapping for the physical address of the kernel. + */ + pgd = pgd_alloc(&init_mm); + pmd = pmd_offset(pgd, PHYS_OFFSET); + *pmd = __pmd((PHYS_OFFSET & PGDIR_MASK) | + PMD_TYPE_SECT | PMD_SECT_AP_WRITE); + + /* + * We need to tell the secondary core where to find + * its stack and the page tables. + */ + secondary_data.stack = (void *)idle->thread_info + THREAD_SIZE - 8; + secondary_data.pgdir = virt_to_phys(pgd); + wmb(); + /* * Now bring the CPU into our world. */ ret = boot_secondary(cpu, idle); + if (ret == 0) { + unsigned long timeout; + + /* + * CPU was successfully started, wait for it + * to come online or time out. + */ + timeout = jiffies + HZ; + while (time_before(jiffies, timeout)) { + if (cpu_online(cpu)) + break; + + udelay(10); + barrier(); + } + + if (!cpu_online(cpu)) + ret = -EIO; + } + + secondary_data.stack = 0; + secondary_data.pgdir = 0; + + *pmd_offset(pgd, PHYS_OFFSET) = __pmd(0); + pgd_free(pgd); + if (ret) { printk(KERN_CRIT "cpu_up: processor %d failed to boot\n", cpu); /* @@ -97,6 +154,56 @@ int __init __cpu_up(unsigned int cpu) return ret; } +/* + * This is the secondary CPU boot entry. We're using this CPUs + * idle thread stack, but a set of temporary page tables. + */ +asmlinkage void __init secondary_start_kernel(void) +{ + struct mm_struct *mm = &init_mm; + unsigned int cpu = smp_processor_id(); + + printk("CPU%u: Booted secondary processor\n", cpu); + + /* + * All kernel threads share the same mm context; grab a + * reference and switch to it. + */ + atomic_inc(&mm->mm_users); + atomic_inc(&mm->mm_count); + current->active_mm = mm; + cpu_set(cpu, mm->cpu_vm_mask); + cpu_switch_mm(mm->pgd, mm); + enter_lazy_tlb(mm, current); + + cpu_init(); + + /* + * Give the platform a chance to do its own initialisation. + */ + platform_secondary_init(cpu); + + /* + * Enable local interrupts. + */ + local_irq_enable(); + local_fiq_enable(); + + calibrate_delay(); + + smp_store_cpu_info(cpu); + + /* + * OK, now it's safe to let the boot CPU continue + */ + cpu_set(cpu, cpu_online_map); + + /* + * OK, it's off to the idle thread for us + */ + cpu_idle(); +} + /* * Called by both boot and secondaries to move global data into * per-processor storage. diff --git a/include/asm-arm/smp.h b/include/asm-arm/smp.h index f21fd8f6bcdd..bd44f894690f 100644 --- a/include/asm-arm/smp.h +++ b/include/asm-arm/smp.h @@ -55,4 +55,18 @@ extern void smp_cross_call(cpumask_t callmap); */ extern int boot_secondary(unsigned int cpu, struct task_struct *); +/* + * Perform platform specific initialisation of the specified CPU. + */ +extern void platform_secondary_init(unsigned int cpu); + +/* + * Initial data for bringing up a secondary CPU. + */ +struct secondary_data { + unsigned long pgdir; + void *stack; +}; +extern struct secondary_data secondary_data; + #endif /* ifndef __ASM_ARM_SMP_H */ -- cgit v1.2.3 From 26b15dad9f1c19d6d4f7b999b07eaa6d98e4b375 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sat, 18 Jun 2005 22:42:13 -0700 Subject: [IPSEC] Add complete xfrm event notification Heres the final patch. What this patch provides - netlink xfrm events - ability to have events generated by netlink propagated to pfkey and vice versa. - fixes the acquire lets-be-happy-with-one-success issue Signed-off-by: Jamal Hadi Salim Signed-off-by: Herbert Xu --- include/linux/xfrm.h | 2 + include/net/xfrm.h | 29 +++- net/key/af_key.c | 357 +++++++++++++++++++++++++++++++++++++------------- net/xfrm/xfrm_state.c | 74 ++++++++--- net/xfrm/xfrm_user.c | 272 +++++++++++++++++++++++++++++++++++++- 5 files changed, 616 insertions(+), 118 deletions(-) (limited to 'include') diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index fd2ef742a9fd..03bc600516ea 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -257,5 +257,7 @@ struct xfrm_usersa_flush { #define XFRMGRP_ACQUIRE 1 #define XFRMGRP_EXPIRE 2 +#define XFRMGRP_SA 4 +#define XFRMGRP_POLICY 8 #endif /* _LINUX_XFRM_H */ diff --git a/include/net/xfrm.h b/include/net/xfrm.h index d675836ba6c3..a159655ebede 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -158,6 +158,27 @@ enum { XFRM_STATE_DEAD }; +/* events that could be sent by kernel */ +enum { + XFRM_SAP_INVALID, + XFRM_SAP_EXPIRED, + XFRM_SAP_ADDED, + XFRM_SAP_UPDATED, + XFRM_SAP_DELETED, + XFRM_SAP_FLUSHED, + __XFRM_SAP_MAX +}; +#define XFRM_SAP_MAX (__XFRM_SAP_MAX - 1) + +/* callback structure passed from either netlink or pfkey */ +struct km_event +{ + u32 data; + u32 seq; + u32 pid; + u32 event; +}; + struct xfrm_type; struct xfrm_dst; struct xfrm_policy_afinfo { @@ -179,6 +200,8 @@ struct xfrm_policy_afinfo { extern int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo); extern int xfrm_policy_unregister_afinfo(struct xfrm_policy_afinfo *afinfo); +extern void km_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c); +extern void km_state_notify(struct xfrm_state *x, struct km_event *c); #define XFRM_ACQ_EXPIRES 30 @@ -290,11 +313,11 @@ struct xfrm_mgr { struct list_head list; char *id; - int (*notify)(struct xfrm_state *x, int event); + int (*notify)(struct xfrm_state *x, struct km_event *c); int (*acquire)(struct xfrm_state *x, struct xfrm_tmpl *, struct xfrm_policy *xp, int dir); struct xfrm_policy *(*compile_policy)(u16 family, int opt, u8 *data, int len, int *dir); int (*new_mapping)(struct xfrm_state *x, xfrm_address_t *ipaddr, u16 sport); - int (*notify_policy)(struct xfrm_policy *x, int dir, int event); + int (*notify_policy)(struct xfrm_policy *x, int dir, struct km_event *c); }; extern int xfrm_register_km(struct xfrm_mgr *km); @@ -817,7 +840,7 @@ extern int xfrm_state_add(struct xfrm_state *x); extern int xfrm_state_update(struct xfrm_state *x); extern struct xfrm_state *xfrm_state_lookup(xfrm_address_t *daddr, u32 spi, u8 proto, unsigned short family); extern struct xfrm_state *xfrm_find_acq_byseq(u32 seq); -extern void xfrm_state_delete(struct xfrm_state *x); +extern int xfrm_state_delete(struct xfrm_state *x); extern void xfrm_state_flush(u8 proto); extern int xfrm_replay_check(struct xfrm_state *x, u32 seq); extern void xfrm_replay_advance(struct xfrm_state *x, u32 seq); diff --git a/net/key/af_key.c b/net/key/af_key.c index ce980aa94ed8..d086c117f5f0 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -1240,13 +1240,85 @@ static int pfkey_acquire(struct sock *sk, struct sk_buff *skb, struct sadb_msg * return 0; } +static inline int event2poltype(int event) +{ + switch (event) { + case XFRM_SAP_DELETED: + return SADB_X_SPDDELETE; + case XFRM_SAP_ADDED: + return SADB_X_SPDADD; + case XFRM_SAP_UPDATED: + return SADB_X_SPDUPDATE; + case XFRM_SAP_EXPIRED: + // return SADB_X_SPDEXPIRE; + default: + printk("pfkey: Unknown policy event %d\n", event); + break; + } + + return 0; +} + +static inline int event2keytype(int event) +{ + switch (event) { + case XFRM_SAP_DELETED: + return SADB_DELETE; + case XFRM_SAP_ADDED: + return SADB_ADD; + case XFRM_SAP_UPDATED: + return SADB_UPDATE; + case XFRM_SAP_EXPIRED: + return SADB_EXPIRE; + default: + printk("pfkey: Unknown SA event %d\n", event); + break; + } + + return 0; +} + +/* ADD/UPD/DEL */ +static int key_notify_sa(struct xfrm_state *x, struct km_event *c) +{ + struct sk_buff *skb; + struct sadb_msg *hdr; + int hsc = 3; + + if (c->event == XFRM_SAP_DELETED) + hsc = 0; + + if (c->event == XFRM_SAP_EXPIRED) { + if (c->data) + hsc = 2; + else + hsc = 1; + } + + skb = pfkey_xfrm_state2msg(x, 0, hsc); + + if (IS_ERR(skb)) + return PTR_ERR(skb); + + hdr = (struct sadb_msg *) skb->data; + hdr->sadb_msg_version = PF_KEY_V2; + hdr->sadb_msg_type = event2keytype(c->event); + hdr->sadb_msg_satype = pfkey_proto2satype(x->id.proto); + hdr->sadb_msg_errno = 0; + hdr->sadb_msg_reserved = 0; + hdr->sadb_msg_seq = c->seq; + hdr->sadb_msg_pid = c->pid; + + pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL); + + return 0; +} static int pfkey_add(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { - struct sk_buff *out_skb; - struct sadb_msg *out_hdr; struct xfrm_state *x; int err; + struct km_event c; xfrm_probe_algs(); @@ -1254,6 +1326,7 @@ static int pfkey_add(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, if (IS_ERR(x)) return PTR_ERR(x); + xfrm_state_hold(x); if (hdr->sadb_msg_type == SADB_ADD) err = xfrm_state_add(x); else @@ -1265,27 +1338,23 @@ static int pfkey_add(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, return err; } - out_skb = pfkey_xfrm_state2msg(x, 0, 3); - if (IS_ERR(out_skb)) - return PTR_ERR(out_skb); /* XXX Should we return 0 here ? */ - - out_hdr = (struct sadb_msg *) out_skb->data; - out_hdr->sadb_msg_version = hdr->sadb_msg_version; - out_hdr->sadb_msg_type = hdr->sadb_msg_type; - out_hdr->sadb_msg_satype = pfkey_proto2satype(x->id.proto); - out_hdr->sadb_msg_errno = 0; - out_hdr->sadb_msg_reserved = 0; - out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; - out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - - pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, sk); + if (hdr->sadb_msg_type == SADB_ADD) + c.event = XFRM_SAP_ADDED; + else + c.event = XFRM_SAP_UPDATED; + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + km_state_notify(x, &c); + xfrm_state_put(x); - return 0; + return err; } static int pfkey_delete(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { struct xfrm_state *x; + struct km_event c; + int err; if (!ext_hdrs[SADB_EXT_SA-1] || !present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1], @@ -1301,13 +1370,19 @@ static int pfkey_delete(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h return -EPERM; } - xfrm_state_delete(x); - xfrm_state_put(x); + err = xfrm_state_delete(x); + if (err < 0) { + xfrm_state_put(x); + return err; + } - pfkey_broadcast(skb_clone(skb, GFP_KERNEL), GFP_KERNEL, - BROADCAST_ALL, sk); + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + c.event = XFRM_SAP_DELETED; + km_state_notify(x, &c); + xfrm_state_put(x); - return 0; + return err; } static int pfkey_get(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) @@ -1445,28 +1520,42 @@ static int pfkey_register(struct sock *sk, struct sk_buff *skb, struct sadb_msg return 0; } +static int key_notify_sa_flush(struct km_event *c) +{ + struct sk_buff *skb; + struct sadb_msg *hdr; + + skb = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_ATOMIC); + if (!skb) + return -ENOBUFS; + hdr = (struct sadb_msg *) skb_put(skb, sizeof(struct sadb_msg)); + hdr->sadb_msg_satype = pfkey_proto2satype(c->data); + hdr->sadb_msg_seq = c->seq; + hdr->sadb_msg_pid = c->pid; + hdr->sadb_msg_version = PF_KEY_V2; + hdr->sadb_msg_errno = (uint8_t) 0; + hdr->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); + + pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL); + + return 0; +} + static int pfkey_flush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { unsigned proto; - struct sk_buff *skb_out; - struct sadb_msg *hdr_out; + struct km_event c; proto = pfkey_satype2proto(hdr->sadb_msg_satype); if (proto == 0) return -EINVAL; - skb_out = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_KERNEL); - if (!skb_out) - return -ENOBUFS; - xfrm_state_flush(proto); - - hdr_out = (struct sadb_msg *) skb_put(skb_out, sizeof(struct sadb_msg)); - pfkey_hdr_dup(hdr_out, hdr); - hdr_out->sadb_msg_errno = (uint8_t) 0; - hdr_out->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); - - pfkey_broadcast(skb_out, GFP_KERNEL, BROADCAST_ALL, NULL); + c.data = proto; + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + c.event = XFRM_SAP_FLUSHED; + km_state_notify(NULL, &c); return 0; } @@ -1859,6 +1948,35 @@ static void pfkey_xfrm_policy2msg(struct sk_buff *skb, struct xfrm_policy *xp, i hdr->sadb_msg_reserved = atomic_read(&xp->refcnt); } +static int key_notify_policy(struct xfrm_policy *xp, int dir, struct km_event *c) +{ + struct sk_buff *out_skb; + struct sadb_msg *out_hdr; + int err; + + out_skb = pfkey_xfrm_policy2msg_prep(xp); + if (IS_ERR(out_skb)) { + err = PTR_ERR(out_skb); + goto out; + } + pfkey_xfrm_policy2msg(out_skb, xp, dir); + + out_hdr = (struct sadb_msg *) out_skb->data; + out_hdr->sadb_msg_version = PF_KEY_V2; + + if (c->data && c->event == XFRM_SAP_DELETED) + out_hdr->sadb_msg_type = SADB_X_SPDDELETE2; + else + out_hdr->sadb_msg_type = event2poltype(c->event); + out_hdr->sadb_msg_errno = 0; + out_hdr->sadb_msg_seq = c->seq; + out_hdr->sadb_msg_pid = c->pid; + pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, NULL); +out: + return 0; + +} + static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { int err; @@ -1866,8 +1984,7 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h struct sadb_address *sa; struct sadb_x_policy *pol; struct xfrm_policy *xp; - struct sk_buff *out_skb; - struct sadb_msg *out_hdr; + struct km_event c; if (!present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1], ext_hdrs[SADB_EXT_ADDRESS_DST-1]) || @@ -1935,31 +2052,23 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h (err = parse_ipsecrequests(xp, pol)) < 0) goto out; - out_skb = pfkey_xfrm_policy2msg_prep(xp); - if (IS_ERR(out_skb)) { - err = PTR_ERR(out_skb); - goto out; - } - err = xfrm_policy_insert(pol->sadb_x_policy_dir-1, xp, hdr->sadb_msg_type != SADB_X_SPDUPDATE); if (err) { - kfree_skb(out_skb); - goto out; + kfree(xp); + return err; } - pfkey_xfrm_policy2msg(out_skb, xp, pol->sadb_x_policy_dir-1); + if (hdr->sadb_msg_type == SADB_X_SPDUPDATE) + c.event = XFRM_SAP_UPDATED; + else + c.event = XFRM_SAP_ADDED; - xfrm_pol_put(xp); + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; - out_hdr = (struct sadb_msg *) out_skb->data; - out_hdr->sadb_msg_version = hdr->sadb_msg_version; - out_hdr->sadb_msg_type = hdr->sadb_msg_type; - out_hdr->sadb_msg_satype = 0; - out_hdr->sadb_msg_errno = 0; - out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; - out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, sk); + km_policy_notify(xp, pol->sadb_x_policy_dir-1, &c); + xfrm_pol_put(xp); return 0; out: @@ -1973,9 +2082,8 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg struct sadb_address *sa; struct sadb_x_policy *pol; struct xfrm_policy *xp; - struct sk_buff *out_skb; - struct sadb_msg *out_hdr; struct xfrm_selector sel; + struct km_event c; if (!present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1], ext_hdrs[SADB_EXT_ADDRESS_DST-1]) || @@ -2010,25 +2118,40 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg err = 0; + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + c.event = XFRM_SAP_DELETED; + km_policy_notify(xp, pol->sadb_x_policy_dir-1, &c); + + xfrm_pol_put(xp); + return err; +} + +static int key_pol_get_resp(struct sock *sk, struct xfrm_policy *xp, struct sadb_msg *hdr, int dir) +{ + int err; + struct sk_buff *out_skb; + struct sadb_msg *out_hdr; + err = 0; + out_skb = pfkey_xfrm_policy2msg_prep(xp); if (IS_ERR(out_skb)) { err = PTR_ERR(out_skb); goto out; } - pfkey_xfrm_policy2msg(out_skb, xp, pol->sadb_x_policy_dir-1); + pfkey_xfrm_policy2msg(out_skb, xp, dir); out_hdr = (struct sadb_msg *) out_skb->data; out_hdr->sadb_msg_version = hdr->sadb_msg_version; - out_hdr->sadb_msg_type = SADB_X_SPDDELETE; + out_hdr->sadb_msg_type = hdr->sadb_msg_type; out_hdr->sadb_msg_satype = 0; out_hdr->sadb_msg_errno = 0; out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, sk); + pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ONE, sk); err = 0; out: - xfrm_pol_put(xp); return err; } @@ -2037,8 +2160,7 @@ static int pfkey_spdget(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h int err; struct sadb_x_policy *pol; struct xfrm_policy *xp; - struct sk_buff *out_skb; - struct sadb_msg *out_hdr; + struct km_event c; if ((pol = ext_hdrs[SADB_X_EXT_POLICY-1]) == NULL) return -EINVAL; @@ -2050,24 +2172,16 @@ static int pfkey_spdget(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h err = 0; - out_skb = pfkey_xfrm_policy2msg_prep(xp); - if (IS_ERR(out_skb)) { - err = PTR_ERR(out_skb); - goto out; + c.seq = hdr->sadb_msg_seq; + c.pid = hdr->sadb_msg_pid; + if (hdr->sadb_msg_type == SADB_X_SPDDELETE2) { + c.data = 1; // to signal pfkey of SADB_X_SPDDELETE2 + c.event = XFRM_SAP_DELETED; + km_policy_notify(xp, pol->sadb_x_policy_dir-1, &c); + } else { + err = key_pol_get_resp(sk, xp, hdr, pol->sadb_x_policy_dir-1); } - pfkey_xfrm_policy2msg(out_skb, xp, pol->sadb_x_policy_dir-1); - out_hdr = (struct sadb_msg *) out_skb->data; - out_hdr->sadb_msg_version = hdr->sadb_msg_version; - out_hdr->sadb_msg_type = hdr->sadb_msg_type; - out_hdr->sadb_msg_satype = 0; - out_hdr->sadb_msg_errno = 0; - out_hdr->sadb_msg_seq = hdr->sadb_msg_seq; - out_hdr->sadb_msg_pid = hdr->sadb_msg_pid; - pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ALL, sk); - err = 0; - -out: xfrm_pol_put(xp); return err; } @@ -2102,22 +2216,34 @@ static int pfkey_spddump(struct sock *sk, struct sk_buff *skb, struct sadb_msg * return xfrm_policy_walk(dump_sp, &data); } -static int pfkey_spdflush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) +static int key_notify_policy_flush(struct km_event *c) { struct sk_buff *skb_out; - struct sadb_msg *hdr_out; + struct sadb_msg *hdr; - skb_out = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_KERNEL); + skb_out = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_ATOMIC); if (!skb_out) return -ENOBUFS; + hdr = (struct sadb_msg *) skb_put(skb_out, sizeof(struct sadb_msg)); + hdr->sadb_msg_seq = c->seq; + hdr->sadb_msg_pid = c->pid; + hdr->sadb_msg_version = PF_KEY_V2; + hdr->sadb_msg_errno = (uint8_t) 0; + hdr->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); + pfkey_broadcast(skb_out, GFP_ATOMIC, BROADCAST_ALL, NULL); + return 0; - xfrm_policy_flush(); +} + +static int pfkey_spdflush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) +{ + struct km_event c; - hdr_out = (struct sadb_msg *) skb_put(skb_out, sizeof(struct sadb_msg)); - pfkey_hdr_dup(hdr_out, hdr); - hdr_out->sadb_msg_errno = (uint8_t) 0; - hdr_out->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t)); - pfkey_broadcast(skb_out, GFP_KERNEL, BROADCAST_ALL, NULL); + xfrm_policy_flush(); + c.event = XFRM_SAP_FLUSHED; + c.pid = hdr->sadb_msg_pid; + c.seq = hdr->sadb_msg_seq; + km_policy_notify(NULL, 0, &c); return 0; } @@ -2317,11 +2443,23 @@ static void dump_esp_combs(struct sk_buff *skb, struct xfrm_tmpl *t) } } -static int pfkey_send_notify(struct xfrm_state *x, int hard) +static int key_notify_policy_expire(struct xfrm_policy *xp, struct km_event *c) +{ + return 0; +} + +static int key_notify_sa_expire(struct xfrm_state *x, struct km_event *c) { struct sk_buff *out_skb; struct sadb_msg *out_hdr; - int hsc = (hard ? 2 : 1); + int hard; + int hsc; + + hard = c->data; + if (hard) + hsc = 2; + else + hsc = 1; out_skb = pfkey_xfrm_state2msg(x, 0, hsc); if (IS_ERR(out_skb)) @@ -2340,6 +2478,44 @@ static int pfkey_send_notify(struct xfrm_state *x, int hard) return 0; } +static int pfkey_send_notify(struct xfrm_state *x, struct km_event *c) +{ + switch (c->event) { + case XFRM_SAP_EXPIRED: + return key_notify_sa_expire(x, c); + case XFRM_SAP_DELETED: + case XFRM_SAP_ADDED: + case XFRM_SAP_UPDATED: + return key_notify_sa(x, c); + case XFRM_SAP_FLUSHED: + return key_notify_sa_flush(c); + default: + printk("pfkey: Unknown SA event %d\n", c->event); + break; + } + + return 0; +} + +static int pfkey_send_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c) +{ + switch (c->event) { + case XFRM_SAP_EXPIRED: + return key_notify_policy_expire(xp, c); + case XFRM_SAP_DELETED: + case XFRM_SAP_ADDED: + case XFRM_SAP_UPDATED: + return key_notify_policy(xp, dir, c); + case XFRM_SAP_FLUSHED: + return key_notify_policy_flush(c); + default: + printk("pfkey: Unknown policy event %d\n", c->event); + break; + } + + return 0; +} + static u32 get_acqseq(void) { u32 res; @@ -2856,6 +3032,7 @@ static struct xfrm_mgr pfkeyv2_mgr = .acquire = pfkey_send_acquire, .compile_policy = pfkey_compile_policy, .new_mapping = pfkey_send_new_mapping, + .notify_policy = pfkey_send_policy_notify, }; static void __exit ipsec_pfkey_exit(void) diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index d11747c2a763..918a94c552a5 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -50,7 +50,7 @@ static DEFINE_SPINLOCK(xfrm_state_gc_lock); static int xfrm_state_gc_flush_bundles; -static void __xfrm_state_delete(struct xfrm_state *x); +static int __xfrm_state_delete(struct xfrm_state *x); static struct xfrm_state_afinfo *xfrm_state_get_afinfo(unsigned short family); static void xfrm_state_put_afinfo(struct xfrm_state_afinfo *afinfo); @@ -215,8 +215,10 @@ void __xfrm_state_destroy(struct xfrm_state *x) } EXPORT_SYMBOL(__xfrm_state_destroy); -static void __xfrm_state_delete(struct xfrm_state *x) +static int __xfrm_state_delete(struct xfrm_state *x) { + int err = -ESRCH; + if (x->km.state != XFRM_STATE_DEAD) { x->km.state = XFRM_STATE_DEAD; spin_lock(&xfrm_state_lock); @@ -245,14 +247,21 @@ static void __xfrm_state_delete(struct xfrm_state *x) * is what we are dropping here. */ atomic_dec(&x->refcnt); + err = 0; } + + return err; } -void xfrm_state_delete(struct xfrm_state *x) +int xfrm_state_delete(struct xfrm_state *x) { + int err; + spin_lock_bh(&x->lock); - __xfrm_state_delete(x); + err = __xfrm_state_delete(x); spin_unlock_bh(&x->lock); + + return err; } EXPORT_SYMBOL(xfrm_state_delete); @@ -796,34 +805,60 @@ EXPORT_SYMBOL(xfrm_replay_advance); static struct list_head xfrm_km_list = LIST_HEAD_INIT(xfrm_km_list); static DEFINE_RWLOCK(xfrm_km_lock); -static void km_state_expired(struct xfrm_state *x, int hard) +void km_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c) { struct xfrm_mgr *km; - if (hard) - x->km.state = XFRM_STATE_EXPIRED; - else - x->km.dying = 1; + read_lock(&xfrm_km_lock); + list_for_each_entry(km, &xfrm_km_list, list) + if (km->notify_policy) + km->notify_policy(xp, dir, c); + read_unlock(&xfrm_km_lock); +} +void km_state_notify(struct xfrm_state *x, struct km_event *c) +{ + struct xfrm_mgr *km; read_lock(&xfrm_km_lock); list_for_each_entry(km, &xfrm_km_list, list) - km->notify(x, hard); + if (km->notify) + km->notify(x, c); read_unlock(&xfrm_km_lock); +} + +EXPORT_SYMBOL(km_policy_notify); +EXPORT_SYMBOL(km_state_notify); + +static void km_state_expired(struct xfrm_state *x, int hard) +{ + struct km_event c; + + if (hard) + x->km.state = XFRM_STATE_EXPIRED; + else + x->km.dying = 1; + c.data = hard; + c.event = XFRM_SAP_EXPIRED; + km_state_notify(x, &c); if (hard) wake_up(&km_waitq); } +/* + * We send to all registered managers regardless of failure + * We are happy with one success +*/ static int km_query(struct xfrm_state *x, struct xfrm_tmpl *t, struct xfrm_policy *pol) { - int err = -EINVAL; + int err = -EINVAL, acqret; struct xfrm_mgr *km; read_lock(&xfrm_km_lock); list_for_each_entry(km, &xfrm_km_list, list) { - err = km->acquire(x, t, pol, XFRM_POLICY_OUT); - if (!err) - break; + acqret = km->acquire(x, t, pol, XFRM_POLICY_OUT); + if (!acqret) + err = acqret; } read_unlock(&xfrm_km_lock); return err; @@ -848,13 +883,12 @@ EXPORT_SYMBOL(km_new_mapping); void km_policy_expired(struct xfrm_policy *pol, int dir, int hard) { - struct xfrm_mgr *km; + struct km_event c; - read_lock(&xfrm_km_lock); - list_for_each_entry(km, &xfrm_km_list, list) - if (km->notify_policy) - km->notify_policy(pol, dir, hard); - read_unlock(&xfrm_km_lock); + c.data = hard; + c.data = hard; + c.event = XFRM_SAP_EXPIRED; + km_policy_notify(pol, dir, &c); if (hard) wake_up(&km_waitq); diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 97509011c274..bd8e6882c083 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -277,6 +277,7 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) struct xfrm_usersa_info *p = NLMSG_DATA(nlh); struct xfrm_state *x; int err; + struct km_event c; err = verify_newsa_info(p, (struct rtattr **) xfrma); if (err) @@ -286,6 +287,7 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) if (!x) return err; + xfrm_state_hold(x); if (nlh->nlmsg_type == XFRM_MSG_NEWSA) err = xfrm_state_add(x); else @@ -294,14 +296,27 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) if (err < 0) { x->km.state = XFRM_STATE_DEAD; xfrm_state_put(x); + return err; } + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + if (nlh->nlmsg_type == XFRM_MSG_NEWSA) + c.event = XFRM_SAP_ADDED; + else + c.event = XFRM_SAP_UPDATED; + + km_state_notify(x, &c); + xfrm_state_put(x); + return err; } static int xfrm_del_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) { struct xfrm_state *x; + int err; + struct km_event c; struct xfrm_usersa_id *p = NLMSG_DATA(nlh); x = xfrm_state_lookup(&p->daddr, p->spi, p->proto, p->family); @@ -313,10 +328,19 @@ static int xfrm_del_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) return -EPERM; } - xfrm_state_delete(x); + err = xfrm_state_delete(x); + if (err < 0) { + xfrm_state_put(x); + return err; + } + + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + c.event = XFRM_SAP_DELETED; + km_state_notify(x, &c); xfrm_state_put(x); - return 0; + return err; } static void copy_to_user_state(struct xfrm_state *x, struct xfrm_usersa_info *p) @@ -681,6 +705,7 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr { struct xfrm_userpolicy_info *p = NLMSG_DATA(nlh); struct xfrm_policy *xp; + struct km_event c; int err; int excl; @@ -692,6 +717,10 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr if (!xp) return err; + /* shouldnt excl be based on nlh flags?? + * Aha! this is anti-netlink really i.e more pfkey derived + * in netlink excl is a flag and you wouldnt need + * a type XFRM_MSG_UPDPOLICY - JHS */ excl = nlh->nlmsg_type == XFRM_MSG_NEWPOLICY; err = xfrm_policy_insert(p->dir, xp, excl); if (err) { @@ -699,6 +728,15 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr return err; } + if (!excl) + c.event = XFRM_SAP_UPDATED; + else + c.event = XFRM_SAP_ADDED; + + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + km_policy_notify(xp, p->dir, &c); + xfrm_pol_put(xp); return 0; @@ -816,6 +854,7 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr struct xfrm_policy *xp; struct xfrm_userpolicy_id *p; int err; + struct km_event c; int delete; p = NLMSG_DATA(nlh); @@ -843,6 +882,11 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr NETLINK_CB(skb).pid, MSG_DONTWAIT); } + } else { + c.event = XFRM_SAP_DELETED; + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + km_policy_notify(xp, p->dir, &c); } xfrm_pol_put(xp); @@ -852,15 +896,28 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr static int xfrm_flush_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) { + struct km_event c; struct xfrm_usersa_flush *p = NLMSG_DATA(nlh); xfrm_state_flush(p->proto); + c.data = p->proto; + c.event = XFRM_SAP_FLUSHED; + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + km_state_notify(NULL, &c); + return 0; } static int xfrm_flush_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) { + struct km_event c; + xfrm_policy_flush(); + c.event = XFRM_SAP_FLUSHED; + c.seq = nlh->nlmsg_seq; + c.pid = nlh->nlmsg_pid; + km_policy_notify(NULL, 0, &c); return 0; } @@ -1069,10 +1126,12 @@ nlmsg_failure: return -1; } -static int xfrm_send_state_notify(struct xfrm_state *x, int hard) +static int xfrm_exp_state_notify(struct xfrm_state *x, struct km_event *c) { struct sk_buff *skb; + int hard = c ->data; + /* fix to do alloc using NLM macros */ skb = alloc_skb(sizeof(struct xfrm_user_expire) + 16, GFP_ATOMIC); if (skb == NULL) return -ENOMEM; @@ -1085,6 +1144,122 @@ static int xfrm_send_state_notify(struct xfrm_state *x, int hard) return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_EXPIRE, GFP_ATOMIC); } +static int xfrm_notify_sa_flush(struct km_event *c) +{ + struct xfrm_usersa_flush *p; + struct nlmsghdr *nlh; + struct sk_buff *skb; + unsigned char *b; + int len = NLMSG_LENGTH(sizeof(struct xfrm_usersa_flush)); + + skb = alloc_skb(len, GFP_ATOMIC); + if (skb == NULL) + return -ENOMEM; + b = skb->tail; + + nlh = NLMSG_PUT(skb, c->pid, c->seq, + XFRM_MSG_FLUSHSA, sizeof(*p)); + nlh->nlmsg_flags = 0; + + p = NLMSG_DATA(nlh); + p->proto = c->data; + + nlh->nlmsg_len = skb->tail - b; + + return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_SA, GFP_ATOMIC); + +nlmsg_failure: + kfree_skb(skb); + return -1; +} + +static int inline xfrm_sa_len(struct xfrm_state *x) +{ + int l = NLMSG_LENGTH(sizeof(struct xfrm_usersa_info)); + if (x->aalg) + l += RTA_SPACE(sizeof(*x->aalg) + (x->aalg->alg_key_len+7)/8); + if (x->ealg) + l += RTA_SPACE(sizeof(*x->ealg) + (x->ealg->alg_key_len+7)/8); + if (x->calg) + l += RTA_SPACE(sizeof(*x->calg)); + if (x->encap) + l += RTA_SPACE(sizeof(*x->encap)); + + return l; +} + +static int xfrm_notify_sa(struct xfrm_state *x, struct km_event *c) +{ + struct xfrm_usersa_info *p; + struct nlmsghdr *nlh; + struct sk_buff *skb; + u32 nlt; + unsigned char *b; + int len = xfrm_sa_len(x); + + skb = alloc_skb(len, GFP_ATOMIC); + if (skb == NULL) + return -ENOMEM; + b = skb->tail; + + if (c->event == XFRM_SAP_ADDED) + nlt = XFRM_MSG_NEWSA; + else if (c->event == XFRM_SAP_UPDATED) + nlt = XFRM_MSG_UPDSA; + else if (c->event == XFRM_SAP_DELETED) + nlt = XFRM_MSG_DELSA; + else + goto nlmsg_failure; + + nlh = NLMSG_PUT(skb, c->pid, c->seq, nlt, sizeof(*p)); + nlh->nlmsg_flags = 0; + + p = NLMSG_DATA(nlh); + copy_to_user_state(x, p); + + if (x->aalg) + RTA_PUT(skb, XFRMA_ALG_AUTH, + sizeof(*(x->aalg))+(x->aalg->alg_key_len+7)/8, x->aalg); + if (x->ealg) + RTA_PUT(skb, XFRMA_ALG_CRYPT, + sizeof(*(x->ealg))+(x->ealg->alg_key_len+7)/8, x->ealg); + if (x->calg) + RTA_PUT(skb, XFRMA_ALG_COMP, sizeof(*(x->calg)), x->calg); + + if (x->encap) + RTA_PUT(skb, XFRMA_ENCAP, sizeof(*x->encap), x->encap); + + nlh->nlmsg_len = skb->tail - b; + + return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_SA, GFP_ATOMIC); + +nlmsg_failure: +rtattr_failure: + kfree_skb(skb); + return -1; +} + +static int xfrm_send_state_notify(struct xfrm_state *x, struct km_event *c) +{ + + switch (c->event) { + case XFRM_SAP_EXPIRED: + return xfrm_exp_state_notify(x, c); + case XFRM_SAP_DELETED: + case XFRM_SAP_UPDATED: + case XFRM_SAP_ADDED: + return xfrm_notify_sa(x, c); + case XFRM_SAP_FLUSHED: + return xfrm_notify_sa_flush(c); + default: + printk("xfrm_user: Unknown SA event %d\n", c->event); + break; + } + + return 0; + +} + static int build_acquire(struct sk_buff *skb, struct xfrm_state *x, struct xfrm_tmpl *xt, struct xfrm_policy *xp, int dir) @@ -1218,7 +1393,7 @@ nlmsg_failure: return -1; } -static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, int hard) +static int xfrm_exp_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c) { struct sk_buff *skb; size_t len; @@ -1229,7 +1404,7 @@ static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, int hard) if (skb == NULL) return -ENOMEM; - if (build_polexpire(skb, xp, dir, hard) < 0) + if (build_polexpire(skb, xp, dir, c->data) < 0) BUG(); NETLINK_CB(skb).dst_groups = XFRMGRP_EXPIRE; @@ -1237,6 +1412,93 @@ static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, int hard) return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_EXPIRE, GFP_ATOMIC); } +static int xfrm_notify_policy(struct xfrm_policy *xp, int dir, struct km_event *c) +{ + struct xfrm_userpolicy_info *p; + struct nlmsghdr *nlh; + struct sk_buff *skb; + u32 nlt = 0 ; + unsigned char *b; + int len = RTA_SPACE(sizeof(struct xfrm_user_tmpl) * xp->xfrm_nr); + len += NLMSG_SPACE(sizeof(struct xfrm_userpolicy_info)); + + skb = alloc_skb(len, GFP_ATOMIC); + if (skb == NULL) + return -ENOMEM; + b = skb->tail; + + if (c->event == XFRM_SAP_ADDED) + nlt = XFRM_MSG_NEWPOLICY; + else if (c->event == XFRM_SAP_UPDATED) + nlt = XFRM_MSG_UPDPOLICY; + else if (c->event == XFRM_SAP_DELETED) + nlt = XFRM_MSG_DELPOLICY; + else + goto nlmsg_failure; + + nlh = NLMSG_PUT(skb, c->pid, c->seq, nlt, sizeof(*p)); + + p = NLMSG_DATA(nlh); + + nlh->nlmsg_flags = 0; + + copy_to_user_policy(xp, p, dir); + if (copy_to_user_tmpl(xp, skb) < 0) + goto nlmsg_failure; + + nlh->nlmsg_len = skb->tail - b; + + return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_POLICY, GFP_ATOMIC); + +nlmsg_failure: + kfree_skb(skb); + return -1; +} + +static int xfrm_notify_policy_flush(struct km_event *c) +{ + struct nlmsghdr *nlh; + struct sk_buff *skb; + unsigned char *b; + int len = NLMSG_LENGTH(0); + + skb = alloc_skb(len, GFP_ATOMIC); + if (skb == NULL) + return -ENOMEM; + b = skb->tail; + + + nlh = NLMSG_PUT(skb, c->pid, c->seq, XFRM_MSG_FLUSHPOLICY, 0); + + nlh->nlmsg_len = skb->tail - b; + + return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_POLICY, GFP_ATOMIC); + +nlmsg_failure: + kfree_skb(skb); + return -1; +} + +static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c) +{ + + switch (c->event) { + case XFRM_SAP_ADDED: + case XFRM_SAP_UPDATED: + case XFRM_SAP_DELETED: + return xfrm_notify_policy(xp, dir, c); + case XFRM_SAP_FLUSHED: + return xfrm_notify_policy_flush(c); + case XFRM_SAP_EXPIRED: + return xfrm_exp_policy_notify(xp, dir, c); + default: + printk("xfrm_user: Unknown Policy event %d\n", c->event); + } + + return 0; + +} + static struct xfrm_mgr netlink_mgr = { .id = "netlink", .notify = xfrm_send_state_notify, -- cgit v1.2.3 From 4666faab095230ec8aa62da6c33391287f281154 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sat, 18 Jun 2005 22:43:22 -0700 Subject: [IPSEC] Kill spurious hard expire messages This patch ensures that the hard state/policy expire notifications are only sent when the state/policy is successfully removed from their respective tables. As it is, it's possible for a state/policy to both expire through reaching a hard limit, as well as being deleted by the user. Note that this behaviour isn't actually forbidden by RFC 2367. However, it is a quality of implementation issue. As an added bonus, the restructuring in this patch will help eventually in moving the expire notifications from softirq context into process context, thus improving their reliability. One important side-effect from this change is that SAs reaching their hard byte/packet limits are now deleted immediately, just like SAs that have reached their hard time limits. Previously they were announced immediately but only deleted after 30 seconds. This is bad because it prevents the system from issuing an ACQUIRE command until the existing state was deleted by the user or expires after the time is up. In the scenario where the expire notification was lost this introduces a 30 second delay into the system for no good reason. Signed-off-by: Herbert Xu --- include/net/xfrm.h | 2 +- net/xfrm/xfrm_policy.c | 8 +++++--- net/xfrm/xfrm_state.c | 16 +++++++--------- 3 files changed, 13 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index a159655ebede..aaa0f5f330e2 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -679,7 +679,7 @@ static inline int xfrm_sk_clone_policy(struct sock *sk) return 0; } -extern void xfrm_policy_delete(struct xfrm_policy *pol, int dir); +extern int xfrm_policy_delete(struct xfrm_policy *pol, int dir); static inline void xfrm_sk_free_policy(struct sock *sk) { diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index d07f5ce31824..0a4260719a12 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -216,8 +216,8 @@ out: expired: read_unlock(&xp->lock); - km_policy_expired(xp, dir, 1); - xfrm_policy_delete(xp, dir); + if (!xfrm_policy_delete(xp, dir)) + km_policy_expired(xp, dir, 1); xfrm_pol_put(xp); } @@ -555,7 +555,7 @@ static struct xfrm_policy *__xfrm_policy_unlink(struct xfrm_policy *pol, return NULL; } -void xfrm_policy_delete(struct xfrm_policy *pol, int dir) +int xfrm_policy_delete(struct xfrm_policy *pol, int dir) { write_lock_bh(&xfrm_policy_lock); pol = __xfrm_policy_unlink(pol, dir); @@ -564,7 +564,9 @@ void xfrm_policy_delete(struct xfrm_policy *pol, int dir) if (dir < XFRM_POLICY_MAX) atomic_inc(&flow_cache_genid); xfrm_policy_kill(pol); + return 0; } + return -ENOENT; } int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol) diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 918a94c552a5..94f7416a4ab0 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -154,6 +154,7 @@ static void xfrm_timer_handler(unsigned long data) next = tmo; } + x->km.dying = warn; if (warn) km_state_expired(x, 0); resched: @@ -169,9 +170,8 @@ expired: next = 2; goto resched; } - if (x->id.spi != 0) + if (!__xfrm_state_delete(x) && x->id.spi) km_state_expired(x, 1); - __xfrm_state_delete(x); out: spin_unlock(&x->lock); @@ -566,16 +566,18 @@ int xfrm_state_check_expire(struct xfrm_state *x) if (x->curlft.bytes >= x->lft.hard_byte_limit || x->curlft.packets >= x->lft.hard_packet_limit) { - km_state_expired(x, 1); - if (!mod_timer(&x->timer, jiffies + XFRM_ACQ_EXPIRES*HZ)) + x->km.state = XFRM_STATE_EXPIRED; + if (!mod_timer(&x->timer, jiffies)) xfrm_state_hold(x); return -EINVAL; } if (!x->km.dying && (x->curlft.bytes >= x->lft.soft_byte_limit || - x->curlft.packets >= x->lft.soft_packet_limit)) + x->curlft.packets >= x->lft.soft_packet_limit)) { + x->km.dying = 1; km_state_expired(x, 0); + } return 0; } EXPORT_SYMBOL(xfrm_state_check_expire); @@ -833,10 +835,6 @@ static void km_state_expired(struct xfrm_state *x, int hard) { struct km_event c; - if (hard) - x->km.state = XFRM_STATE_EXPIRED; - else - x->km.dying = 1; c.data = hard; c.event = XFRM_SAP_EXPIRED; km_state_notify(x, &c); -- cgit v1.2.3 From bf08867f91a43aa3ba2e4598c06c4769a6cdddf6 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sat, 18 Jun 2005 22:44:00 -0700 Subject: [IPSEC] Turn km_event.data into a union This patch turns km_event.data into a union. This makes code that uses it clearer. Signed-off-by: Herbert Xu --- include/net/xfrm.h | 7 ++++++- net/key/af_key.c | 17 +++++------------ net/xfrm/xfrm_state.c | 5 ++--- net/xfrm/xfrm_user.c | 9 ++++----- 4 files changed, 17 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index aaa0f5f330e2..fda356e81014 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -173,7 +173,12 @@ enum { /* callback structure passed from either netlink or pfkey */ struct km_event { - u32 data; + union { + u32 hard; + u32 proto; + u32 byid; + } data; + u32 seq; u32 pid; u32 event; diff --git a/net/key/af_key.c b/net/key/af_key.c index 560c93c10890..3fae5c4b48f6 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -1293,13 +1293,6 @@ static int key_notify_sa(struct xfrm_state *x, struct km_event *c) if (c->event == XFRM_SAP_DELETED) hsc = 0; - if (c->event == XFRM_SAP_EXPIRED) { - if (c->data) - hsc = 2; - else - hsc = 1; - } - skb = pfkey_xfrm_state2msg(x, 0, hsc); if (IS_ERR(skb)) @@ -1534,7 +1527,7 @@ static int key_notify_sa_flush(struct km_event *c) if (!skb) return -ENOBUFS; hdr = (struct sadb_msg *) skb_put(skb, sizeof(struct sadb_msg)); - hdr->sadb_msg_satype = pfkey_proto2satype(c->data); + hdr->sadb_msg_satype = pfkey_proto2satype(c->data.proto); hdr->sadb_msg_seq = c->seq; hdr->sadb_msg_pid = c->pid; hdr->sadb_msg_version = PF_KEY_V2; @@ -1556,7 +1549,7 @@ static int pfkey_flush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hd return -EINVAL; xfrm_state_flush(proto); - c.data = proto; + c.data.proto = proto; c.seq = hdr->sadb_msg_seq; c.pid = hdr->sadb_msg_pid; c.event = XFRM_SAP_FLUSHED; @@ -1969,7 +1962,7 @@ static int key_notify_policy(struct xfrm_policy *xp, int dir, struct km_event *c out_hdr = (struct sadb_msg *) out_skb->data; out_hdr->sadb_msg_version = PF_KEY_V2; - if (c->data && c->event == XFRM_SAP_DELETED) + if (c->data.byid && c->event == XFRM_SAP_DELETED) out_hdr->sadb_msg_type = SADB_X_SPDDELETE2; else out_hdr->sadb_msg_type = event2poltype(c->event); @@ -2180,7 +2173,7 @@ static int pfkey_spdget(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h c.seq = hdr->sadb_msg_seq; c.pid = hdr->sadb_msg_pid; if (hdr->sadb_msg_type == SADB_X_SPDDELETE2) { - c.data = 1; // to signal pfkey of SADB_X_SPDDELETE2 + c.data.byid = 1; c.event = XFRM_SAP_DELETED; km_policy_notify(xp, pol->sadb_x_policy_dir-1, &c); } else { @@ -2460,7 +2453,7 @@ static int key_notify_sa_expire(struct xfrm_state *x, struct km_event *c) int hard; int hsc; - hard = c->data; + hard = c->data.hard; if (hard) hsc = 2; else diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 94f7416a4ab0..e068bd721050 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -835,7 +835,7 @@ static void km_state_expired(struct xfrm_state *x, int hard) { struct km_event c; - c.data = hard; + c.data.hard = hard; c.event = XFRM_SAP_EXPIRED; km_state_notify(x, &c); @@ -883,8 +883,7 @@ void km_policy_expired(struct xfrm_policy *pol, int dir, int hard) { struct km_event c; - c.data = hard; - c.data = hard; + c.data.hard = hard; c.event = XFRM_SAP_EXPIRED; km_policy_notify(pol, dir, &c); diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index bd8e6882c083..4d3237d08ffb 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -900,7 +900,7 @@ static int xfrm_flush_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma struct xfrm_usersa_flush *p = NLMSG_DATA(nlh); xfrm_state_flush(p->proto); - c.data = p->proto; + c.data.proto = p->proto; c.event = XFRM_SAP_FLUSHED; c.seq = nlh->nlmsg_seq; c.pid = nlh->nlmsg_pid; @@ -1129,14 +1129,13 @@ nlmsg_failure: static int xfrm_exp_state_notify(struct xfrm_state *x, struct km_event *c) { struct sk_buff *skb; - int hard = c ->data; /* fix to do alloc using NLM macros */ skb = alloc_skb(sizeof(struct xfrm_user_expire) + 16, GFP_ATOMIC); if (skb == NULL) return -ENOMEM; - if (build_expire(skb, x, hard) < 0) + if (build_expire(skb, x, c->data.hard) < 0) BUG(); NETLINK_CB(skb).dst_groups = XFRMGRP_EXPIRE; @@ -1162,7 +1161,7 @@ static int xfrm_notify_sa_flush(struct km_event *c) nlh->nlmsg_flags = 0; p = NLMSG_DATA(nlh); - p->proto = c->data; + p->proto = c->data.proto; nlh->nlmsg_len = skb->tail - b; @@ -1404,7 +1403,7 @@ static int xfrm_exp_policy_notify(struct xfrm_policy *xp, int dir, struct km_eve if (skb == NULL) return -ENOMEM; - if (build_polexpire(skb, xp, dir, c->data) < 0) + if (build_polexpire(skb, xp, dir, c->data.hard) < 0) BUG(); NETLINK_CB(skb).dst_groups = XFRMGRP_EXPIRE; -- cgit v1.2.3 From f60f6b8f70c756fc786d68f02ec17a1e84db645f Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sat, 18 Jun 2005 22:44:37 -0700 Subject: [IPSEC] Use XFRM_MSG_* instead of XFRM_SAP_* This patch removes XFRM_SAP_* and converts them over to XFRM_MSG_*. The netlink interface is meant to map directly onto the underlying xfrm subsystem. Therefore rather than using a new independent representation for the events we can simply use the existing ones from xfrm_user. Signed-off-by: Herbert Xu --- include/net/xfrm.h | 12 ---------- net/key/af_key.c | 60 ++++++++++++++++++++++++------------------------ net/xfrm/xfrm_state.c | 4 ++-- net/xfrm/xfrm_user.c | 63 +++++++++++++++------------------------------------ 4 files changed, 50 insertions(+), 89 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index fda356e81014..0e65e02b7a1d 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -158,18 +158,6 @@ enum { XFRM_STATE_DEAD }; -/* events that could be sent by kernel */ -enum { - XFRM_SAP_INVALID, - XFRM_SAP_EXPIRED, - XFRM_SAP_ADDED, - XFRM_SAP_UPDATED, - XFRM_SAP_DELETED, - XFRM_SAP_FLUSHED, - __XFRM_SAP_MAX -}; -#define XFRM_SAP_MAX (__XFRM_SAP_MAX - 1) - /* callback structure passed from either netlink or pfkey */ struct km_event { diff --git a/net/key/af_key.c b/net/key/af_key.c index 3fae5c4b48f6..577f0bb5bb31 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -1248,13 +1248,13 @@ static int pfkey_acquire(struct sock *sk, struct sk_buff *skb, struct sadb_msg * static inline int event2poltype(int event) { switch (event) { - case XFRM_SAP_DELETED: + case XFRM_MSG_DELPOLICY: return SADB_X_SPDDELETE; - case XFRM_SAP_ADDED: + case XFRM_MSG_NEWPOLICY: return SADB_X_SPDADD; - case XFRM_SAP_UPDATED: + case XFRM_MSG_UPDPOLICY: return SADB_X_SPDUPDATE; - case XFRM_SAP_EXPIRED: + case XFRM_MSG_POLEXPIRE: // return SADB_X_SPDEXPIRE; default: printk("pfkey: Unknown policy event %d\n", event); @@ -1267,13 +1267,13 @@ static inline int event2poltype(int event) static inline int event2keytype(int event) { switch (event) { - case XFRM_SAP_DELETED: + case XFRM_MSG_DELSA: return SADB_DELETE; - case XFRM_SAP_ADDED: + case XFRM_MSG_NEWSA: return SADB_ADD; - case XFRM_SAP_UPDATED: + case XFRM_MSG_UPDSA: return SADB_UPDATE; - case XFRM_SAP_EXPIRED: + case XFRM_MSG_EXPIRE: return SADB_EXPIRE; default: printk("pfkey: Unknown SA event %d\n", event); @@ -1290,7 +1290,7 @@ static int key_notify_sa(struct xfrm_state *x, struct km_event *c) struct sadb_msg *hdr; int hsc = 3; - if (c->event == XFRM_SAP_DELETED) + if (c->event == XFRM_MSG_DELSA) hsc = 0; skb = pfkey_xfrm_state2msg(x, 0, hsc); @@ -1337,9 +1337,9 @@ static int pfkey_add(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, } if (hdr->sadb_msg_type == SADB_ADD) - c.event = XFRM_SAP_ADDED; + c.event = XFRM_MSG_NEWSA; else - c.event = XFRM_SAP_UPDATED; + c.event = XFRM_MSG_UPDSA; c.seq = hdr->sadb_msg_seq; c.pid = hdr->sadb_msg_pid; km_state_notify(x, &c); @@ -1376,7 +1376,7 @@ static int pfkey_delete(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h c.seq = hdr->sadb_msg_seq; c.pid = hdr->sadb_msg_pid; - c.event = XFRM_SAP_DELETED; + c.event = XFRM_MSG_DELSA; km_state_notify(x, &c); xfrm_state_put(x); @@ -1552,7 +1552,7 @@ static int pfkey_flush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hd c.data.proto = proto; c.seq = hdr->sadb_msg_seq; c.pid = hdr->sadb_msg_pid; - c.event = XFRM_SAP_FLUSHED; + c.event = XFRM_MSG_FLUSHSA; km_state_notify(NULL, &c); return 0; @@ -1962,7 +1962,7 @@ static int key_notify_policy(struct xfrm_policy *xp, int dir, struct km_event *c out_hdr = (struct sadb_msg *) out_skb->data; out_hdr->sadb_msg_version = PF_KEY_V2; - if (c->data.byid && c->event == XFRM_SAP_DELETED) + if (c->data.byid && c->event == XFRM_MSG_DELPOLICY) out_hdr->sadb_msg_type = SADB_X_SPDDELETE2; else out_hdr->sadb_msg_type = event2poltype(c->event); @@ -2058,9 +2058,9 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h } if (hdr->sadb_msg_type == SADB_X_SPDUPDATE) - c.event = XFRM_SAP_UPDATED; - else - c.event = XFRM_SAP_ADDED; + c.event = XFRM_MSG_UPDPOLICY; + else + c.event = XFRM_MSG_NEWPOLICY; c.seq = hdr->sadb_msg_seq; c.pid = hdr->sadb_msg_pid; @@ -2118,7 +2118,7 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg c.seq = hdr->sadb_msg_seq; c.pid = hdr->sadb_msg_pid; - c.event = XFRM_SAP_DELETED; + c.event = XFRM_MSG_DELPOLICY; km_policy_notify(xp, pol->sadb_x_policy_dir-1, &c); xfrm_pol_put(xp); @@ -2174,7 +2174,7 @@ static int pfkey_spdget(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h c.pid = hdr->sadb_msg_pid; if (hdr->sadb_msg_type == SADB_X_SPDDELETE2) { c.data.byid = 1; - c.event = XFRM_SAP_DELETED; + c.event = XFRM_MSG_DELPOLICY; km_policy_notify(xp, pol->sadb_x_policy_dir-1, &c); } else { err = key_pol_get_resp(sk, xp, hdr, pol->sadb_x_policy_dir-1); @@ -2238,7 +2238,7 @@ static int pfkey_spdflush(struct sock *sk, struct sk_buff *skb, struct sadb_msg struct km_event c; xfrm_policy_flush(); - c.event = XFRM_SAP_FLUSHED; + c.event = XFRM_MSG_FLUSHPOLICY; c.pid = hdr->sadb_msg_pid; c.seq = hdr->sadb_msg_seq; km_policy_notify(NULL, 0, &c); @@ -2479,13 +2479,13 @@ static int key_notify_sa_expire(struct xfrm_state *x, struct km_event *c) static int pfkey_send_notify(struct xfrm_state *x, struct km_event *c) { switch (c->event) { - case XFRM_SAP_EXPIRED: + case XFRM_MSG_EXPIRE: return key_notify_sa_expire(x, c); - case XFRM_SAP_DELETED: - case XFRM_SAP_ADDED: - case XFRM_SAP_UPDATED: + case XFRM_MSG_DELSA: + case XFRM_MSG_NEWSA: + case XFRM_MSG_UPDSA: return key_notify_sa(x, c); - case XFRM_SAP_FLUSHED: + case XFRM_MSG_FLUSHSA: return key_notify_sa_flush(c); default: printk("pfkey: Unknown SA event %d\n", c->event); @@ -2498,13 +2498,13 @@ static int pfkey_send_notify(struct xfrm_state *x, struct km_event *c) static int pfkey_send_policy_notify(struct xfrm_policy *xp, int dir, struct km_event *c) { switch (c->event) { - case XFRM_SAP_EXPIRED: + case XFRM_MSG_POLEXPIRE: return key_notify_policy_expire(xp, c); - case XFRM_SAP_DELETED: - case XFRM_SAP_ADDED: - case XFRM_SAP_UPDATED: + case XFRM_MSG_DELPOLICY: + case XFRM_MSG_NEWPOLICY: + case XFRM_MSG_UPDPOLICY: return key_notify_policy(xp, dir, c); - case XFRM_SAP_FLUSHED: + case XFRM_MSG_FLUSHPOLICY: return key_notify_policy_flush(c); default: printk("pfkey: Unknown policy event %d\n", c->event); diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index e068bd721050..2537f26f097c 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -836,7 +836,7 @@ static void km_state_expired(struct xfrm_state *x, int hard) struct km_event c; c.data.hard = hard; - c.event = XFRM_SAP_EXPIRED; + c.event = XFRM_MSG_EXPIRE; km_state_notify(x, &c); if (hard) @@ -884,7 +884,7 @@ void km_policy_expired(struct xfrm_policy *pol, int dir, int hard) struct km_event c; c.data.hard = hard; - c.event = XFRM_SAP_EXPIRED; + c.event = XFRM_MSG_POLEXPIRE; km_policy_notify(pol, dir, &c); if (hard) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index ff6fc610730e..bc4fd7456873 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -301,10 +301,7 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) c.seq = nlh->nlmsg_seq; c.pid = nlh->nlmsg_pid; - if (nlh->nlmsg_type == XFRM_MSG_NEWSA) - c.event = XFRM_SAP_ADDED; - else - c.event = XFRM_SAP_UPDATED; + c.event = nlh->nlmsg_type; km_state_notify(x, &c); xfrm_state_put(x); @@ -336,7 +333,7 @@ static int xfrm_del_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma) c.seq = nlh->nlmsg_seq; c.pid = nlh->nlmsg_pid; - c.event = XFRM_SAP_DELETED; + c.event = nlh->nlmsg_type; km_state_notify(x, &c); xfrm_state_put(x); @@ -728,11 +725,7 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr return err; } - if (!excl) - c.event = XFRM_SAP_UPDATED; - else - c.event = XFRM_SAP_ADDED; - + c.event = nlh->nlmsg_type; c.seq = nlh->nlmsg_seq; c.pid = nlh->nlmsg_pid; km_policy_notify(xp, p->dir, &c); @@ -884,7 +877,7 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfr } } else { c.data.byid = p->index; - c.event = XFRM_SAP_DELETED; + c.event = nlh->nlmsg_type; c.seq = nlh->nlmsg_seq; c.pid = nlh->nlmsg_pid; km_policy_notify(xp, p->dir, &c); @@ -902,7 +895,7 @@ static int xfrm_flush_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma xfrm_state_flush(p->proto); c.data.proto = p->proto; - c.event = XFRM_SAP_FLUSHED; + c.event = nlh->nlmsg_type; c.seq = nlh->nlmsg_seq; c.pid = nlh->nlmsg_pid; km_state_notify(NULL, &c); @@ -915,7 +908,7 @@ static int xfrm_flush_policy(struct sk_buff *skb, struct nlmsghdr *nlh, void **x struct km_event c; xfrm_policy_flush(); - c.event = XFRM_SAP_FLUSHED; + c.event = nlh->nlmsg_type; c.seq = nlh->nlmsg_seq; c.pid = nlh->nlmsg_pid; km_policy_notify(NULL, 0, &c); @@ -1193,7 +1186,6 @@ static int xfrm_notify_sa(struct xfrm_state *x, struct km_event *c) struct xfrm_usersa_info *p; struct nlmsghdr *nlh; struct sk_buff *skb; - u32 nlt; unsigned char *b; int len = xfrm_sa_len(x); @@ -1202,16 +1194,7 @@ static int xfrm_notify_sa(struct xfrm_state *x, struct km_event *c) return -ENOMEM; b = skb->tail; - if (c->event == XFRM_SAP_ADDED) - nlt = XFRM_MSG_NEWSA; - else if (c->event == XFRM_SAP_UPDATED) - nlt = XFRM_MSG_UPDSA; - else if (c->event == XFRM_SAP_DELETED) - nlt = XFRM_MSG_DELSA; - else - goto nlmsg_failure; - - nlh = NLMSG_PUT(skb, c->pid, c->seq, nlt, sizeof(*p)); + nlh = NLMSG_PUT(skb, c->pid, c->seq, c->event, sizeof(*p)); nlh->nlmsg_flags = 0; p = NLMSG_DATA(nlh); @@ -1243,13 +1226,13 @@ static int xfrm_send_state_notify(struct xfrm_state *x, struct km_event *c) { switch (c->event) { - case XFRM_SAP_EXPIRED: + case XFRM_MSG_EXPIRE: return xfrm_exp_state_notify(x, c); - case XFRM_SAP_DELETED: - case XFRM_SAP_UPDATED: - case XFRM_SAP_ADDED: + case XFRM_MSG_DELSA: + case XFRM_MSG_UPDSA: + case XFRM_MSG_NEWSA: return xfrm_notify_sa(x, c); - case XFRM_SAP_FLUSHED: + case XFRM_MSG_FLUSHSA: return xfrm_notify_sa_flush(c); default: printk("xfrm_user: Unknown SA event %d\n", c->event); @@ -1417,7 +1400,6 @@ static int xfrm_notify_policy(struct xfrm_policy *xp, int dir, struct km_event * struct xfrm_userpolicy_info *p; struct nlmsghdr *nlh; struct sk_buff *skb; - u32 nlt = 0 ; unsigned char *b; int len = RTA_SPACE(sizeof(struct xfrm_user_tmpl) * xp->xfrm_nr); len += NLMSG_SPACE(sizeof(struct xfrm_userpolicy_info)); @@ -1427,16 +1409,7 @@ static int xfrm_notify_policy(struct xfrm_policy *xp, int dir, struct km_event * return -ENOMEM; b = skb->tail; - if (c->event == XFRM_SAP_ADDED) - nlt = XFRM_MSG_NEWPOLICY; - else if (c->event == XFRM_SAP_UPDATED) - nlt = XFRM_MSG_UPDPOLICY; - else if (c->event == XFRM_SAP_DELETED) - nlt = XFRM_MSG_DELPOLICY; - else - goto nlmsg_failure; - - nlh = NLMSG_PUT(skb, c->pid, c->seq, nlt, sizeof(*p)); + nlh = NLMSG_PUT(skb, c->pid, c->seq, c->event, sizeof(*p)); p = NLMSG_DATA(nlh); @@ -1483,13 +1456,13 @@ static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, struct km_ev { switch (c->event) { - case XFRM_SAP_ADDED: - case XFRM_SAP_UPDATED: - case XFRM_SAP_DELETED: + case XFRM_MSG_NEWPOLICY: + case XFRM_MSG_UPDPOLICY: + case XFRM_MSG_DELPOLICY: return xfrm_notify_policy(xp, dir, c); - case XFRM_SAP_FLUSHED: + case XFRM_MSG_FLUSHPOLICY: return xfrm_notify_policy_flush(c); - case XFRM_SAP_EXPIRED: + case XFRM_MSG_POLEXPIRE: return xfrm_exp_policy_notify(xp, dir, c); default: printk("xfrm_user: Unknown Policy event %d\n", c->event); -- cgit v1.2.3 From 1944972d3bb651474a5021c9da8d0166ae19f1eb Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sat, 18 Jun 2005 22:46:19 -0700 Subject: [SLAB] Introduce kmem_cache_name This is for use with slab users that pass a dynamically allocated slab name in kmem_cache_create, so that before destroying the slab one can retrieve the name and free its memory. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/slab.h | 1 + mm/slab.c | 6 ++++++ 2 files changed, 7 insertions(+) (limited to 'include') diff --git a/include/linux/slab.h b/include/linux/slab.h index 7d66385ae750..76cf7e60216c 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -64,6 +64,7 @@ extern int kmem_cache_shrink(kmem_cache_t *); extern void *kmem_cache_alloc(kmem_cache_t *, unsigned int __nocast); extern void kmem_cache_free(kmem_cache_t *, void *); extern unsigned int kmem_cache_size(kmem_cache_t *); +extern const char *kmem_cache_name(kmem_cache_t *); extern kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags); /* Size description struct for general caches. */ diff --git a/mm/slab.c b/mm/slab.c index 840742641152..c78d343b3c5f 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -2620,6 +2620,12 @@ unsigned int kmem_cache_size(kmem_cache_t *cachep) } EXPORT_SYMBOL(kmem_cache_size); +const char *kmem_cache_name(kmem_cache_t *cachep) +{ + return cachep->name; +} +EXPORT_SYMBOL_GPL(kmem_cache_name); + struct ccupdate_struct { kmem_cache_t *cachep; struct array_cache *new[NR_CPUS]; -- cgit v1.2.3 From 2e6599cb899ba4b133f42cbf9d2b1883d2dc583a Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sat, 18 Jun 2005 22:46:52 -0700 Subject: [NET] Generalise TCP's struct open_request minisock infrastructure Kept this first changeset minimal, without changing existing names to ease peer review. Basicaly tcp_openreq_alloc now receives the or_calltable, that in turn has two new members: ->slab, that replaces tcp_openreq_cachep ->obj_size, to inform the size of the openreq descendant for a specific protocol The protocol specific fields in struct open_request were moved to a class hierarchy, with the things that are common to all connection oriented PF_INET protocols in struct inet_request_sock, the TCP ones in tcp_request_sock, that is an inet_request_sock, that is an open_request. I.e. this uses the same approach used for the struct sock class hierarchy, with sk_prot indicating if the protocol wants to use the open_request infrastructure by filling in sk_prot->rsk_prot with an or_calltable. Results? Performance is improved and TCP v4 now uses only 64 bytes per open request minisock, down from 96 without this patch :-) Next changeset will rename some of the structs, fields and functions mentioned above, struct or_calltable is way unclear, better name it struct request_sock_ops, s/struct open_request/struct request_sock/g, etc. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/ip.h | 21 ++++++++++ include/linux/ipv6.h | 13 +++++++ include/linux/tcp.h | 11 ++++++ include/net/request_sock.h | 77 +++++++++++++++++++++++++++++++++++++ include/net/sock.h | 4 ++ include/net/tcp.h | 87 ++++++----------------------------------- include/net/tcp_ecn.h | 7 ++-- net/core/sock.c | 35 +++++++++++++++++ net/ipv4/syncookies.c | 39 ++++++++++--------- net/ipv4/tcp.c | 9 ----- net/ipv4/tcp_diag.c | 25 ++++++------ net/ipv4/tcp_ipv4.c | 72 +++++++++++++++++++--------------- net/ipv4/tcp_minisocks.c | 48 ++++++++++++----------- net/ipv4/tcp_output.c | 25 ++++++------ net/ipv4/tcp_timer.c | 2 +- net/ipv6/tcp_ipv6.c | 96 ++++++++++++++++++++++++---------------------- 16 files changed, 341 insertions(+), 230 deletions(-) create mode 100644 include/net/request_sock.h (limited to 'include') diff --git a/include/linux/ip.h b/include/linux/ip.h index 8438c68591f9..d5b7c907204e 100644 --- a/include/linux/ip.h +++ b/include/linux/ip.h @@ -81,6 +81,7 @@ #ifdef __KERNEL__ #include #include +#include #include #include #include @@ -107,6 +108,26 @@ struct ip_options { #define optlength(opt) (sizeof(struct ip_options) + opt->optlen) +struct inet_request_sock { + struct open_request req; + u32 loc_addr; + u32 rmt_addr; + u16 rmt_port; + u16 snd_wscale : 4, + rcv_wscale : 4, + tstamp_ok : 1, + sack_ok : 1, + wscale_ok : 1, + ecn_ok : 1, + acked : 1; + struct ip_options *opt; +}; + +static inline struct inet_request_sock *inet_rsk(const struct open_request *sk) +{ + return (struct inet_request_sock *)sk; +} + struct ipv6_pinfo; struct inet_sock { diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index ab0d0efbf240..98acdbf3d446 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -193,6 +193,19 @@ struct inet6_skb_parm { #define IP6CB(skb) ((struct inet6_skb_parm*)((skb)->cb)) +struct tcp6_request_sock { + struct tcp_request_sock req; + struct in6_addr loc_addr; + struct in6_addr rmt_addr; + struct sk_buff *pktopts; + int iif; +}; + +static inline struct tcp6_request_sock *tcp6_rsk(const struct open_request *sk) +{ + return (struct tcp6_request_sock *)sk; +} + /** * struct ipv6_pinfo - ipv6 private area * diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 14a55e3e3a50..86771b37b80d 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -230,6 +230,17 @@ struct tcp_options_received { __u16 mss_clamp; /* Maximal mss, negotiated at connection setup */ }; +struct tcp_request_sock { + struct inet_request_sock req; + __u32 rcv_isn; + __u32 snt_isn; +}; + +static inline struct tcp_request_sock *tcp_rsk(const struct open_request *req) +{ + return (struct tcp_request_sock *)req; +} + struct tcp_sock { /* inet_sock has to be the first member of tcp_sock */ struct inet_sock inet; diff --git a/include/net/request_sock.h b/include/net/request_sock.h new file mode 100644 index 000000000000..9502f5587931 --- /dev/null +++ b/include/net/request_sock.h @@ -0,0 +1,77 @@ +/* + * NET Generic infrastructure for Network protocols. + * + * Definitions for request_sock + * + * Authors: Arnaldo Carvalho de Melo + * + * From code originally in include/net/tcp.h + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#ifndef _REQUEST_SOCK_H +#define _REQUEST_SOCK_H + +#include +#include +#include + +struct open_request; +struct sk_buff; +struct dst_entry; +struct proto; + +struct or_calltable { + int family; + kmem_cache_t *slab; + int obj_size; + int (*rtx_syn_ack)(struct sock *sk, + struct open_request *req, + struct dst_entry *dst); + void (*send_ack)(struct sk_buff *skb, + struct open_request *req); + void (*send_reset)(struct sk_buff *skb); + void (*destructor)(struct open_request *req); +}; + +/* struct open_request - mini sock to represent a connection request + */ +struct open_request { + struct open_request *dl_next; /* Must be first member! */ + u16 mss; + u8 retrans; + u8 __pad; + /* The following two fields can be easily recomputed I think -AK */ + u32 window_clamp; /* window clamp at creation time */ + u32 rcv_wnd; /* rcv_wnd offered first time */ + u32 ts_recent; + unsigned long expires; + struct or_calltable *class; + struct sock *sk; +}; + +static inline struct open_request *tcp_openreq_alloc(struct or_calltable *class) +{ + struct open_request *req = kmem_cache_alloc(class->slab, SLAB_ATOMIC); + + if (req != NULL) + req->class = class; + + return req; +} + +static inline void tcp_openreq_fastfree(struct open_request *req) +{ + kmem_cache_free(req->class->slab, req); +} + +static inline void tcp_openreq_free(struct open_request *req) +{ + req->class->destructor(req); + tcp_openreq_fastfree(req); +} + +#endif /* _REQUEST_SOCK_H */ diff --git a/include/net/sock.h b/include/net/sock.h index a9ef3a6a13f3..6919276af8af 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -484,6 +484,8 @@ extern void sk_stream_kill_queues(struct sock *sk); extern int sk_wait_data(struct sock *sk, long *timeo); +struct or_calltable; + /* Networking protocol blocks we attach to sockets. * socket layer -> transport layer interface * transport -> network interface is defined by struct inet_proto @@ -547,6 +549,8 @@ struct proto { kmem_cache_t *slab; unsigned int obj_size; + struct or_calltable *rsk_prot; + struct module *owner; char name[32]; diff --git a/include/net/tcp.h b/include/net/tcp.h index e71f8ba3e101..d438ba566b89 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -31,6 +31,7 @@ #include #include #include +#include #include #include #include @@ -613,74 +614,6 @@ extern atomic_t tcp_memory_allocated; extern atomic_t tcp_sockets_allocated; extern int tcp_memory_pressure; -struct open_request; - -struct or_calltable { - int family; - int (*rtx_syn_ack) (struct sock *sk, struct open_request *req, struct dst_entry*); - void (*send_ack) (struct sk_buff *skb, struct open_request *req); - void (*destructor) (struct open_request *req); - void (*send_reset) (struct sk_buff *skb); -}; - -struct tcp_v4_open_req { - __u32 loc_addr; - __u32 rmt_addr; - struct ip_options *opt; -}; - -#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) -struct tcp_v6_open_req { - struct in6_addr loc_addr; - struct in6_addr rmt_addr; - struct sk_buff *pktopts; - int iif; -}; -#endif - -/* this structure is too big */ -struct open_request { - struct open_request *dl_next; /* Must be first member! */ - __u32 rcv_isn; - __u32 snt_isn; - __u16 rmt_port; - __u16 mss; - __u8 retrans; - __u8 __pad; - __u16 snd_wscale : 4, - rcv_wscale : 4, - tstamp_ok : 1, - sack_ok : 1, - wscale_ok : 1, - ecn_ok : 1, - acked : 1; - /* The following two fields can be easily recomputed I think -AK */ - __u32 window_clamp; /* window clamp at creation time */ - __u32 rcv_wnd; /* rcv_wnd offered first time */ - __u32 ts_recent; - unsigned long expires; - struct or_calltable *class; - struct sock *sk; - union { - struct tcp_v4_open_req v4_req; -#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) - struct tcp_v6_open_req v6_req; -#endif - } af; -}; - -/* SLAB cache for open requests. */ -extern kmem_cache_t *tcp_openreq_cachep; - -#define tcp_openreq_alloc() kmem_cache_alloc(tcp_openreq_cachep, SLAB_ATOMIC) -#define tcp_openreq_fastfree(req) kmem_cache_free(tcp_openreq_cachep, req) - -static inline void tcp_openreq_free(struct open_request *req) -{ - req->class->destructor(req); - tcp_openreq_fastfree(req); -} - #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) #define TCP_INET_FAMILY(fam) ((fam) == AF_INET) #else @@ -1832,17 +1765,19 @@ static __inline__ void tcp_openreq_init(struct open_request *req, struct tcp_options_received *rx_opt, struct sk_buff *skb) { + struct inet_request_sock *ireq = inet_rsk(req); + req->rcv_wnd = 0; /* So that tcp_send_synack() knows! */ - req->rcv_isn = TCP_SKB_CB(skb)->seq; + tcp_rsk(req)->rcv_isn = TCP_SKB_CB(skb)->seq; req->mss = rx_opt->mss_clamp; req->ts_recent = rx_opt->saw_tstamp ? rx_opt->rcv_tsval : 0; - req->tstamp_ok = rx_opt->tstamp_ok; - req->sack_ok = rx_opt->sack_ok; - req->snd_wscale = rx_opt->snd_wscale; - req->wscale_ok = rx_opt->wscale_ok; - req->acked = 0; - req->ecn_ok = 0; - req->rmt_port = skb->h.th->source; + ireq->tstamp_ok = rx_opt->tstamp_ok; + ireq->sack_ok = rx_opt->sack_ok; + ireq->snd_wscale = rx_opt->snd_wscale; + ireq->wscale_ok = rx_opt->wscale_ok; + ireq->acked = 0; + ireq->ecn_ok = 0; + ireq->rmt_port = skb->h.th->source; } extern void tcp_enter_memory_pressure(void); diff --git a/include/net/tcp_ecn.h b/include/net/tcp_ecn.h index dc1456389a97..94ad970e844a 100644 --- a/include/net/tcp_ecn.h +++ b/include/net/tcp_ecn.h @@ -2,6 +2,7 @@ #define _NET_TCP_ECN_H_ 1 #include +#include #define TCP_HP_BITS (~(TCP_RESERVED_BITS|TCP_FLAG_PSH)) @@ -40,7 +41,7 @@ static inline void TCP_ECN_send_syn(struct sock *sk, struct tcp_sock *tp, static __inline__ void TCP_ECN_make_synack(struct open_request *req, struct tcphdr *th) { - if (req->ecn_ok) + if (inet_rsk(req)->ecn_ok) th->ece = 1; } @@ -113,14 +114,14 @@ static inline int TCP_ECN_rcv_ecn_echo(struct tcp_sock *tp, struct tcphdr *th) static inline void TCP_ECN_openreq_child(struct tcp_sock *tp, struct open_request *req) { - tp->ecn_flags = req->ecn_ok ? TCP_ECN_OK : 0; + tp->ecn_flags = inet_rsk(req)->ecn_ok ? TCP_ECN_OK : 0; } static __inline__ void TCP_ECN_create_request(struct open_request *req, struct tcphdr *th) { if (sysctl_tcp_ecn && th->ece && th->cwr) - req->ecn_ok = 1; + inet_rsk(req)->ecn_ok = 1; } #endif diff --git a/net/core/sock.c b/net/core/sock.c index 96e00b08698f..a6ec3ada7f9e 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -118,6 +118,7 @@ #include #include #include +#include #include #include #include @@ -1363,6 +1364,7 @@ static LIST_HEAD(proto_list); int proto_register(struct proto *prot, int alloc_slab) { + char *request_sock_slab_name; int rc = -ENOBUFS; if (alloc_slab) { @@ -1374,6 +1376,25 @@ int proto_register(struct proto *prot, int alloc_slab) prot->name); goto out; } + + if (prot->rsk_prot != NULL) { + static const char mask[] = "request_sock_%s"; + + request_sock_slab_name = kmalloc(strlen(prot->name) + sizeof(mask) - 1, GFP_KERNEL); + if (request_sock_slab_name == NULL) + goto out_free_sock_slab; + + sprintf(request_sock_slab_name, mask, prot->name); + prot->rsk_prot->slab = kmem_cache_create(request_sock_slab_name, + prot->rsk_prot->obj_size, 0, + SLAB_HWCACHE_ALIGN, NULL, NULL); + + if (prot->rsk_prot->slab == NULL) { + printk(KERN_CRIT "%s: Can't create request sock SLAB cache!\n", + prot->name); + goto out_free_request_sock_slab_name; + } + } } write_lock(&proto_list_lock); @@ -1382,6 +1403,12 @@ int proto_register(struct proto *prot, int alloc_slab) rc = 0; out: return rc; +out_free_request_sock_slab_name: + kfree(request_sock_slab_name); +out_free_sock_slab: + kmem_cache_destroy(prot->slab); + prot->slab = NULL; + goto out; } EXPORT_SYMBOL(proto_register); @@ -1395,6 +1422,14 @@ void proto_unregister(struct proto *prot) prot->slab = NULL; } + if (prot->rsk_prot != NULL && prot->rsk_prot->slab != NULL) { + const char *name = kmem_cache_name(prot->rsk_prot->slab); + + kmem_cache_destroy(prot->rsk_prot->slab); + kfree(name); + prot->rsk_prot->slab = NULL; + } + list_del(&prot->node); write_unlock(&proto_list_lock); } diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index e923d2f021aa..dd47e6da6fb3 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -190,6 +190,8 @@ static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb, struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, struct ip_options *opt) { + struct inet_request_sock *ireq; + struct tcp_request_sock *treq; struct tcp_sock *tp = tcp_sk(sk); __u32 cookie = ntohl(skb->h.th->ack_seq) - 1; struct sock *ret = sk; @@ -209,19 +211,20 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESRECV); - req = tcp_openreq_alloc(); ret = NULL; + req = tcp_openreq_alloc(&or_ipv4); /* for safety */ if (!req) goto out; - req->rcv_isn = htonl(skb->h.th->seq) - 1; - req->snt_isn = cookie; + ireq = inet_rsk(req); + treq = tcp_rsk(req); + treq->rcv_isn = htonl(skb->h.th->seq) - 1; + treq->snt_isn = cookie; req->mss = mss; - req->rmt_port = skb->h.th->source; - req->af.v4_req.loc_addr = skb->nh.iph->daddr; - req->af.v4_req.rmt_addr = skb->nh.iph->saddr; - req->class = &or_ipv4; /* for savety */ - req->af.v4_req.opt = NULL; + ireq->rmt_port = skb->h.th->source; + ireq->loc_addr = skb->nh.iph->daddr; + ireq->rmt_addr = skb->nh.iph->saddr; + ireq->opt = NULL; /* We throwed the options of the initial SYN away, so we hope * the ACK carries the same options again (see RFC1122 4.2.3.8) @@ -229,17 +232,15 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, if (opt && opt->optlen) { int opt_size = sizeof(struct ip_options) + opt->optlen; - req->af.v4_req.opt = kmalloc(opt_size, GFP_ATOMIC); - if (req->af.v4_req.opt) { - if (ip_options_echo(req->af.v4_req.opt, skb)) { - kfree(req->af.v4_req.opt); - req->af.v4_req.opt = NULL; - } + ireq->opt = kmalloc(opt_size, GFP_ATOMIC); + if (ireq->opt != NULL && ip_options_echo(ireq->opt, skb)) { + kfree(ireq->opt); + ireq->opt = NULL; } } - req->snd_wscale = req->rcv_wscale = req->tstamp_ok = 0; - req->wscale_ok = req->sack_ok = 0; + ireq->snd_wscale = ireq->rcv_wscale = ireq->tstamp_ok = 0; + ireq->wscale_ok = ireq->sack_ok = 0; req->expires = 0UL; req->retrans = 0; @@ -253,8 +254,8 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, struct flowi fl = { .nl_u = { .ip4_u = { .daddr = ((opt && opt->srr) ? opt->faddr : - req->af.v4_req.rmt_addr), - .saddr = req->af.v4_req.loc_addr, + ireq->rmt_addr), + .saddr = ireq->loc_addr, .tos = RT_CONN_FLAGS(sk) } }, .proto = IPPROTO_TCP, .uli_u = { .ports = @@ -272,7 +273,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, &req->rcv_wnd, &req->window_clamp, 0, &rcv_wscale); /* BTW win scale with syncookies is 0 by definition */ - req->rcv_wscale = rcv_wscale; + ireq->rcv_wscale = rcv_wscale; ret = get_cookie_sock(sk, skb, req, &rt->u.dst); out: return ret; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 0d9a4fd5f1a4..a3cabfa2022a 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -271,7 +271,6 @@ int sysctl_tcp_fin_timeout = TCP_FIN_TIMEOUT; DEFINE_SNMP_STAT(struct tcp_mib, tcp_statistics); -kmem_cache_t *tcp_openreq_cachep; kmem_cache_t *tcp_bucket_cachep; kmem_cache_t *tcp_timewait_cachep; @@ -2271,13 +2270,6 @@ void __init tcp_init(void) __skb_cb_too_small_for_tcp(sizeof(struct tcp_skb_cb), sizeof(skb->cb)); - tcp_openreq_cachep = kmem_cache_create("tcp_open_request", - sizeof(struct open_request), - 0, SLAB_HWCACHE_ALIGN, - NULL, NULL); - if (!tcp_openreq_cachep) - panic("tcp_init: Cannot alloc open_request cache."); - tcp_bucket_cachep = kmem_cache_create("tcp_bind_bucket", sizeof(struct tcp_bind_bucket), 0, SLAB_HWCACHE_ALIGN, @@ -2374,7 +2366,6 @@ EXPORT_SYMBOL(tcp_destroy_sock); EXPORT_SYMBOL(tcp_disconnect); EXPORT_SYMBOL(tcp_getsockopt); EXPORT_SYMBOL(tcp_ioctl); -EXPORT_SYMBOL(tcp_openreq_cachep); EXPORT_SYMBOL(tcp_poll); EXPORT_SYMBOL(tcp_read_sock); EXPORT_SYMBOL(tcp_recvmsg); diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index 8faa8948f75c..700ff2413588 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -458,6 +458,7 @@ static int tcpdiag_fill_req(struct sk_buff *skb, struct sock *sk, struct open_request *req, u32 pid, u32 seq) { + const struct inet_request_sock *ireq = inet_rsk(req); struct inet_sock *inet = inet_sk(sk); unsigned char *b = skb->tail; struct tcpdiagmsg *r; @@ -482,9 +483,9 @@ static int tcpdiag_fill_req(struct sk_buff *skb, struct sock *sk, tmo = 0; r->id.tcpdiag_sport = inet->sport; - r->id.tcpdiag_dport = req->rmt_port; - r->id.tcpdiag_src[0] = req->af.v4_req.loc_addr; - r->id.tcpdiag_dst[0] = req->af.v4_req.rmt_addr; + r->id.tcpdiag_dport = ireq->rmt_port; + r->id.tcpdiag_src[0] = ireq->loc_addr; + r->id.tcpdiag_dst[0] = ireq->rmt_addr; r->tcpdiag_expires = jiffies_to_msecs(tmo), r->tcpdiag_rqueue = 0; r->tcpdiag_wqueue = 0; @@ -493,9 +494,9 @@ static int tcpdiag_fill_req(struct sk_buff *skb, struct sock *sk, #ifdef CONFIG_IP_TCPDIAG_IPV6 if (r->tcpdiag_family == AF_INET6) { ipv6_addr_copy((struct in6_addr *)r->id.tcpdiag_src, - &req->af.v6_req.loc_addr); + &tcp6_rsk(req)->loc_addr); ipv6_addr_copy((struct in6_addr *)r->id.tcpdiag_dst, - &req->af.v6_req.rmt_addr); + &tcp6_rsk(req)->rmt_addr); } #endif nlh->nlmsg_len = skb->tail - b; @@ -545,9 +546,11 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, reqnum = 0; for (req = head; req; reqnum++, req = req->dl_next) { + struct inet_request_sock *ireq = inet_rsk(req); + if (reqnum < s_reqnum) continue; - if (r->id.tcpdiag_dport != req->rmt_port && + if (r->id.tcpdiag_dport != ireq->rmt_port && r->id.tcpdiag_dport) continue; @@ -555,16 +558,16 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, entry.saddr = #ifdef CONFIG_IP_TCPDIAG_IPV6 (entry.family == AF_INET6) ? - req->af.v6_req.loc_addr.s6_addr32 : + tcp6_rsk(req)->loc_addr.s6_addr32 : #endif - &req->af.v4_req.loc_addr; + &ireq->loc_addr; entry.daddr = #ifdef CONFIG_IP_TCPDIAG_IPV6 (entry.family == AF_INET6) ? - req->af.v6_req.rmt_addr.s6_addr32 : + tcp6_rsk(req)->rmt_addr.s6_addr32 : #endif - &req->af.v4_req.rmt_addr; - entry.dport = ntohs(req->rmt_port); + &ireq->rmt_addr; + entry.dport = ntohs(ireq->rmt_port); if (!tcpdiag_bc_run(RTA_DATA(bc), RTA_PAYLOAD(bc), &entry)) diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index dad98e4a5043..e156be90df14 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -880,9 +880,11 @@ static struct open_request *tcp_v4_search_req(struct tcp_sock *tp, for (prev = &lopt->syn_table[tcp_v4_synq_hash(raddr, rport, lopt->hash_rnd)]; (req = *prev) != NULL; prev = &req->dl_next) { - if (req->rmt_port == rport && - req->af.v4_req.rmt_addr == raddr && - req->af.v4_req.loc_addr == laddr && + const struct inet_request_sock *ireq = inet_rsk(req); + + if (ireq->rmt_port == rport && + ireq->rmt_addr == raddr && + ireq->loc_addr == laddr && TCP_INET_FAMILY(req->class->family)) { BUG_TRAP(!req->sk); *prevp = prev; @@ -897,7 +899,7 @@ static void tcp_v4_synq_add(struct sock *sk, struct open_request *req) { struct tcp_sock *tp = tcp_sk(sk); struct tcp_listen_opt *lopt = tp->listen_opt; - u32 h = tcp_v4_synq_hash(req->af.v4_req.rmt_addr, req->rmt_port, lopt->hash_rnd); + u32 h = tcp_v4_synq_hash(inet_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port, lopt->hash_rnd); req->expires = jiffies + TCP_TIMEOUT_INIT; req->retrans = 0; @@ -1065,7 +1067,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info) */ BUG_TRAP(!req->sk); - if (seq != req->snt_isn) { + if (seq != tcp_rsk(req)->snt_isn) { NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS); goto out; } @@ -1256,7 +1258,7 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb) static void tcp_v4_or_send_ack(struct sk_buff *skb, struct open_request *req) { - tcp_v4_send_ack(skb, req->snt_isn + 1, req->rcv_isn + 1, req->rcv_wnd, + tcp_v4_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } @@ -1264,18 +1266,19 @@ static struct dst_entry* tcp_v4_route_req(struct sock *sk, struct open_request *req) { struct rtable *rt; - struct ip_options *opt = req->af.v4_req.opt; + const struct inet_request_sock *ireq = inet_rsk(req); + struct ip_options *opt = inet_rsk(req)->opt; struct flowi fl = { .oif = sk->sk_bound_dev_if, .nl_u = { .ip4_u = { .daddr = ((opt && opt->srr) ? opt->faddr : - req->af.v4_req.rmt_addr), - .saddr = req->af.v4_req.loc_addr, + ireq->rmt_addr), + .saddr = ireq->loc_addr, .tos = RT_CONN_FLAGS(sk) } }, .proto = IPPROTO_TCP, .uli_u = { .ports = { .sport = inet_sk(sk)->sport, - .dport = req->rmt_port } } }; + .dport = ireq->rmt_port } } }; if (ip_route_output_flow(&rt, &fl, sk, 0)) { IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES); @@ -1297,6 +1300,7 @@ static struct dst_entry* tcp_v4_route_req(struct sock *sk, static int tcp_v4_send_synack(struct sock *sk, struct open_request *req, struct dst_entry *dst) { + const struct inet_request_sock *ireq = inet_rsk(req); int err = -1; struct sk_buff * skb; @@ -1310,14 +1314,14 @@ static int tcp_v4_send_synack(struct sock *sk, struct open_request *req, struct tcphdr *th = skb->h.th; th->check = tcp_v4_check(th, skb->len, - req->af.v4_req.loc_addr, - req->af.v4_req.rmt_addr, + ireq->loc_addr, + ireq->rmt_addr, csum_partial((char *)th, skb->len, skb->csum)); - err = ip_build_and_send_pkt(skb, sk, req->af.v4_req.loc_addr, - req->af.v4_req.rmt_addr, - req->af.v4_req.opt); + err = ip_build_and_send_pkt(skb, sk, ireq->loc_addr, + ireq->rmt_addr, + ireq->opt); if (err == NET_XMIT_CN) err = 0; } @@ -1332,8 +1336,8 @@ out: */ static void tcp_v4_or_free(struct open_request *req) { - if (req->af.v4_req.opt) - kfree(req->af.v4_req.opt); + if (inet_rsk(req)->opt) + kfree(inet_rsk(req)->opt); } static inline void syn_flood_warning(struct sk_buff *skb) @@ -1387,6 +1391,7 @@ int sysctl_max_syn_backlog = 256; struct or_calltable or_ipv4 = { .family = PF_INET, + .obj_size = sizeof(struct tcp_request_sock), .rtx_syn_ack = tcp_v4_send_synack, .send_ack = tcp_v4_or_send_ack, .destructor = tcp_v4_or_free, @@ -1395,6 +1400,7 @@ struct or_calltable or_ipv4 = { int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) { + struct inet_request_sock *ireq; struct tcp_options_received tmp_opt; struct open_request *req; __u32 saddr = skb->nh.iph->saddr; @@ -1433,7 +1439,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) if (sk_acceptq_is_full(sk) && tcp_synq_young(sk) > 1) goto drop; - req = tcp_openreq_alloc(); + req = tcp_openreq_alloc(&or_ipv4); if (!req) goto drop; @@ -1461,10 +1467,10 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) tcp_openreq_init(req, &tmp_opt, skb); - req->af.v4_req.loc_addr = daddr; - req->af.v4_req.rmt_addr = saddr; - req->af.v4_req.opt = tcp_v4_save_options(sk, skb); - req->class = &or_ipv4; + ireq = inet_rsk(req); + ireq->loc_addr = daddr; + ireq->rmt_addr = saddr; + ireq->opt = tcp_v4_save_options(sk, skb); if (!want_cookie) TCP_ECN_create_request(req, skb->h.th); @@ -1523,7 +1529,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) isn = tcp_v4_init_sequence(sk, skb); } - req->snt_isn = isn; + tcp_rsk(req)->snt_isn = isn; if (tcp_v4_send_synack(sk, req, dst)) goto drop_and_free; @@ -1551,6 +1557,7 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, struct open_request *req, struct dst_entry *dst) { + struct inet_request_sock *ireq; struct inet_sock *newinet; struct tcp_sock *newtp; struct sock *newsk; @@ -1570,11 +1577,12 @@ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, newtp = tcp_sk(newsk); newinet = inet_sk(newsk); - newinet->daddr = req->af.v4_req.rmt_addr; - newinet->rcv_saddr = req->af.v4_req.loc_addr; - newinet->saddr = req->af.v4_req.loc_addr; - newinet->opt = req->af.v4_req.opt; - req->af.v4_req.opt = NULL; + ireq = inet_rsk(req); + newinet->daddr = ireq->rmt_addr; + newinet->rcv_saddr = ireq->loc_addr; + newinet->saddr = ireq->loc_addr; + newinet->opt = ireq->opt; + ireq->opt = NULL; newinet->mc_index = tcp_v4_iif(skb); newinet->mc_ttl = skb->nh.iph->ttl; newtp->ext_header_len = 0; @@ -2454,15 +2462,16 @@ void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo) static void get_openreq4(struct sock *sk, struct open_request *req, char *tmpbuf, int i, int uid) { + const struct inet_request_sock *ireq = inet_rsk(req); int ttd = req->expires - jiffies; sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X" " %02X %08X:%08X %02X:%08lX %08X %5d %8d %u %d %p", i, - req->af.v4_req.loc_addr, + ireq->loc_addr, ntohs(inet_sk(sk)->sport), - req->af.v4_req.rmt_addr, - ntohs(req->rmt_port), + ireq->rmt_addr, + ntohs(ireq->rmt_port), TCP_SYN_RECV, 0, 0, /* could print option size, but that is af dependent. */ 1, /* timers active (only the expire timer) */ @@ -2618,6 +2627,7 @@ struct proto tcp_prot = { .sysctl_rmem = sysctl_tcp_rmem, .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp_sock), + .rsk_prot = &or_ipv4, }; diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index eea1a17a9ac2..1037401c7cc8 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -692,6 +692,8 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, struct sock *newsk = sk_alloc(PF_INET, GFP_ATOMIC, sk->sk_prot, 0); if(newsk != NULL) { + struct inet_request_sock *ireq = inet_rsk(req); + struct tcp_request_sock *treq = tcp_rsk(req); struct tcp_sock *newtp; struct sk_filter *filter; @@ -703,7 +705,7 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, tcp_sk(newsk)->bind_hash = NULL; /* Clone the TCP header template */ - inet_sk(newsk)->dport = req->rmt_port; + inet_sk(newsk)->dport = ireq->rmt_port; sock_lock_init(newsk); bh_lock_sock(newsk); @@ -739,14 +741,14 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, /* Now setup tcp_sock */ newtp = tcp_sk(newsk); newtp->pred_flags = 0; - newtp->rcv_nxt = req->rcv_isn + 1; - newtp->snd_nxt = req->snt_isn + 1; - newtp->snd_una = req->snt_isn + 1; - newtp->snd_sml = req->snt_isn + 1; + newtp->rcv_nxt = treq->rcv_isn + 1; + newtp->snd_nxt = treq->snt_isn + 1; + newtp->snd_una = treq->snt_isn + 1; + newtp->snd_sml = treq->snt_isn + 1; tcp_prequeue_init(newtp); - tcp_init_wl(newtp, req->snt_isn, req->rcv_isn); + tcp_init_wl(newtp, treq->snt_isn, treq->rcv_isn); newtp->retransmits = 0; newtp->backoff = 0; @@ -775,10 +777,10 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, tcp_set_ca_state(newtp, TCP_CA_Open); tcp_init_xmit_timers(newsk); skb_queue_head_init(&newtp->out_of_order_queue); - newtp->rcv_wup = req->rcv_isn + 1; - newtp->write_seq = req->snt_isn + 1; + newtp->rcv_wup = treq->rcv_isn + 1; + newtp->write_seq = treq->snt_isn + 1; newtp->pushed_seq = newtp->write_seq; - newtp->copied_seq = req->rcv_isn + 1; + newtp->copied_seq = treq->rcv_isn + 1; newtp->rx_opt.saw_tstamp = 0; @@ -808,18 +810,18 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, newsk->sk_socket = NULL; newsk->sk_sleep = NULL; - newtp->rx_opt.tstamp_ok = req->tstamp_ok; - if((newtp->rx_opt.sack_ok = req->sack_ok) != 0) { + newtp->rx_opt.tstamp_ok = ireq->tstamp_ok; + if((newtp->rx_opt.sack_ok = ireq->sack_ok) != 0) { if (sysctl_tcp_fack) newtp->rx_opt.sack_ok |= 2; } newtp->window_clamp = req->window_clamp; newtp->rcv_ssthresh = req->rcv_wnd; newtp->rcv_wnd = req->rcv_wnd; - newtp->rx_opt.wscale_ok = req->wscale_ok; + newtp->rx_opt.wscale_ok = ireq->wscale_ok; if (newtp->rx_opt.wscale_ok) { - newtp->rx_opt.snd_wscale = req->snd_wscale; - newtp->rx_opt.rcv_wscale = req->rcv_wscale; + newtp->rx_opt.snd_wscale = ireq->snd_wscale; + newtp->rx_opt.rcv_wscale = ireq->rcv_wscale; } else { newtp->rx_opt.snd_wscale = newtp->rx_opt.rcv_wscale = 0; newtp->window_clamp = min(newtp->window_clamp, 65535U); @@ -881,7 +883,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, } /* Check for pure retransmitted SYN. */ - if (TCP_SKB_CB(skb)->seq == req->rcv_isn && + if (TCP_SKB_CB(skb)->seq == tcp_rsk(req)->rcv_isn && flg == TCP_FLAG_SYN && !paws_reject) { /* @@ -959,7 +961,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, * Invalid ACK: reset will be sent by listening socket */ if ((flg & TCP_FLAG_ACK) && - (TCP_SKB_CB(skb)->ack_seq != req->snt_isn+1)) + (TCP_SKB_CB(skb)->ack_seq != tcp_rsk(req)->snt_isn + 1)) return sk; /* Also, it would be not so bad idea to check rcv_tsecr, which @@ -970,7 +972,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, /* RFC793: "first check sequence number". */ if (paws_reject || !tcp_in_window(TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq, - req->rcv_isn+1, req->rcv_isn+1+req->rcv_wnd)) { + tcp_rsk(req)->rcv_isn + 1, tcp_rsk(req)->rcv_isn + 1 + req->rcv_wnd)) { /* Out of window: send ACK and drop. */ if (!(flg & TCP_FLAG_RST)) req->class->send_ack(skb, req); @@ -981,12 +983,12 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, /* In sequence, PAWS is OK. */ - if (tmp_opt.saw_tstamp && !after(TCP_SKB_CB(skb)->seq, req->rcv_isn+1)) + if (tmp_opt.saw_tstamp && !after(TCP_SKB_CB(skb)->seq, tcp_rsk(req)->rcv_isn + 1)) req->ts_recent = tmp_opt.rcv_tsval; - if (TCP_SKB_CB(skb)->seq == req->rcv_isn) { + if (TCP_SKB_CB(skb)->seq == tcp_rsk(req)->rcv_isn) { /* Truncate SYN, it is out of window starting - at req->rcv_isn+1. */ + at tcp_rsk(req)->rcv_isn + 1. */ flg &= ~TCP_FLAG_SYN; } @@ -1003,8 +1005,8 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, return NULL; /* If TCP_DEFER_ACCEPT is set, drop bare ACK. */ - if (tp->defer_accept && TCP_SKB_CB(skb)->end_seq == req->rcv_isn+1) { - req->acked = 1; + if (tp->defer_accept && TCP_SKB_CB(skb)->end_seq == tcp_rsk(req)->rcv_isn + 1) { + inet_rsk(req)->acked = 1; return NULL; } @@ -1026,7 +1028,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, listen_overflow: if (!sysctl_tcp_abort_on_overflow) { - req->acked = 1; + inet_rsk(req)->acked = 1; return NULL; } diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index fa24e7ae1f40..f3c8747caf91 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1358,6 +1358,7 @@ int tcp_send_synack(struct sock *sk) struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, struct open_request *req) { + struct inet_request_sock *ireq = inet_rsk(req); struct tcp_sock *tp = tcp_sk(sk); struct tcphdr *th; int tcp_header_size; @@ -1373,47 +1374,47 @@ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, skb->dst = dst_clone(dst); tcp_header_size = (sizeof(struct tcphdr) + TCPOLEN_MSS + - (req->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0) + - (req->wscale_ok ? TCPOLEN_WSCALE_ALIGNED : 0) + + (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0) + + (ireq->wscale_ok ? TCPOLEN_WSCALE_ALIGNED : 0) + /* SACK_PERM is in the place of NOP NOP of TS */ - ((req->sack_ok && !req->tstamp_ok) ? TCPOLEN_SACKPERM_ALIGNED : 0)); + ((ireq->sack_ok && !ireq->tstamp_ok) ? TCPOLEN_SACKPERM_ALIGNED : 0)); skb->h.th = th = (struct tcphdr *) skb_push(skb, tcp_header_size); memset(th, 0, sizeof(struct tcphdr)); th->syn = 1; th->ack = 1; if (dst->dev->features&NETIF_F_TSO) - req->ecn_ok = 0; + ireq->ecn_ok = 0; TCP_ECN_make_synack(req, th); th->source = inet_sk(sk)->sport; - th->dest = req->rmt_port; - TCP_SKB_CB(skb)->seq = req->snt_isn; + th->dest = ireq->rmt_port; + TCP_SKB_CB(skb)->seq = tcp_rsk(req)->snt_isn; TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(skb)->seq + 1; TCP_SKB_CB(skb)->sacked = 0; skb_shinfo(skb)->tso_segs = 1; skb_shinfo(skb)->tso_size = 0; th->seq = htonl(TCP_SKB_CB(skb)->seq); - th->ack_seq = htonl(req->rcv_isn + 1); + th->ack_seq = htonl(tcp_rsk(req)->rcv_isn + 1); if (req->rcv_wnd == 0) { /* ignored for retransmitted syns */ __u8 rcv_wscale; /* Set this up on the first call only */ req->window_clamp = tp->window_clamp ? : dst_metric(dst, RTAX_WINDOW); /* tcp_full_space because it is guaranteed to be the first packet */ tcp_select_initial_window(tcp_full_space(sk), - dst_metric(dst, RTAX_ADVMSS) - (req->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0), + dst_metric(dst, RTAX_ADVMSS) - (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0), &req->rcv_wnd, &req->window_clamp, - req->wscale_ok, + ireq->wscale_ok, &rcv_wscale); - req->rcv_wscale = rcv_wscale; + ireq->rcv_wscale = rcv_wscale; } /* RFC1323: The window in SYN & SYN/ACK segments is never scaled. */ th->window = htons(req->rcv_wnd); TCP_SKB_CB(skb)->when = tcp_time_stamp; - tcp_syn_build_options((__u32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), req->tstamp_ok, - req->sack_ok, req->wscale_ok, req->rcv_wscale, + tcp_syn_build_options((__u32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), ireq->tstamp_ok, + ireq->sack_ok, ireq->wscale_ok, ireq->rcv_wscale, TCP_SKB_CB(skb)->when, req->ts_recent); diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 799ebe061e2c..ba30ca0aa6a3 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -513,7 +513,7 @@ static void tcp_synack_timer(struct sock *sk) while ((req = *reqp) != NULL) { if (time_after_eq(now, req->expires)) { if ((req->retrans < thresh || - (req->acked && req->retrans < max_retries)) + (inet_rsk(req)->acked && req->retrans < max_retries)) && !req->class->rtx_syn_ack(sk, req, NULL)) { unsigned long timeo; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 0f69e800a0ad..9199ad2fde0d 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -407,11 +407,13 @@ static struct open_request *tcp_v6_search_req(struct tcp_sock *tp, for (prev = &lopt->syn_table[tcp_v6_synq_hash(raddr, rport, lopt->hash_rnd)]; (req = *prev) != NULL; prev = &req->dl_next) { - if (req->rmt_port == rport && + const struct tcp6_request_sock *treq = tcp6_rsk(req); + + if (inet_rsk(req)->rmt_port == rport && req->class->family == AF_INET6 && - ipv6_addr_equal(&req->af.v6_req.rmt_addr, raddr) && - ipv6_addr_equal(&req->af.v6_req.loc_addr, laddr) && - (!req->af.v6_req.iif || req->af.v6_req.iif == iif)) { + ipv6_addr_equal(&treq->rmt_addr, raddr) && + ipv6_addr_equal(&treq->loc_addr, laddr) && + (!treq->iif || treq->iif == iif)) { BUG_TRAP(req->sk == NULL); *prevp = prev; return req; @@ -923,7 +925,7 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, */ BUG_TRAP(req->sk == NULL); - if (seq != req->snt_isn) { + if (seq != tcp_rsk(req)->snt_isn) { NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS); goto out; } @@ -960,6 +962,7 @@ out: static int tcp_v6_send_synack(struct sock *sk, struct open_request *req, struct dst_entry *dst) { + struct tcp6_request_sock *treq = tcp6_rsk(req); struct ipv6_pinfo *np = inet6_sk(sk); struct sk_buff * skb; struct ipv6_txoptions *opt = NULL; @@ -969,19 +972,19 @@ static int tcp_v6_send_synack(struct sock *sk, struct open_request *req, memset(&fl, 0, sizeof(fl)); fl.proto = IPPROTO_TCP; - ipv6_addr_copy(&fl.fl6_dst, &req->af.v6_req.rmt_addr); - ipv6_addr_copy(&fl.fl6_src, &req->af.v6_req.loc_addr); + ipv6_addr_copy(&fl.fl6_dst, &treq->rmt_addr); + ipv6_addr_copy(&fl.fl6_src, &treq->loc_addr); fl.fl6_flowlabel = 0; - fl.oif = req->af.v6_req.iif; - fl.fl_ip_dport = req->rmt_port; + fl.oif = treq->iif; + fl.fl_ip_dport = inet_rsk(req)->rmt_port; fl.fl_ip_sport = inet_sk(sk)->sport; if (dst == NULL) { opt = np->opt; if (opt == NULL && np->rxopt.bits.srcrt == 2 && - req->af.v6_req.pktopts) { - struct sk_buff *pktopts = req->af.v6_req.pktopts; + treq->pktopts) { + struct sk_buff *pktopts = treq->pktopts; struct inet6_skb_parm *rxopt = IP6CB(pktopts); if (rxopt->srcrt) opt = ipv6_invert_rthdr(sk, (struct ipv6_rt_hdr*)(pktopts->nh.raw + rxopt->srcrt)); @@ -1008,10 +1011,10 @@ static int tcp_v6_send_synack(struct sock *sk, struct open_request *req, struct tcphdr *th = skb->h.th; th->check = tcp_v6_check(th, skb->len, - &req->af.v6_req.loc_addr, &req->af.v6_req.rmt_addr, + &treq->loc_addr, &treq->rmt_addr, csum_partial((char *)th, skb->len, skb->csum)); - ipv6_addr_copy(&fl.fl6_dst, &req->af.v6_req.rmt_addr); + ipv6_addr_copy(&fl.fl6_dst, &treq->rmt_addr); err = ip6_xmit(sk, skb, &fl, opt, 0); if (err == NET_XMIT_CN) err = 0; @@ -1026,12 +1029,13 @@ done: static void tcp_v6_or_free(struct open_request *req) { - if (req->af.v6_req.pktopts) - kfree_skb(req->af.v6_req.pktopts); + if (tcp6_rsk(req)->pktopts) + kfree_skb(tcp6_rsk(req)->pktopts); } static struct or_calltable or_ipv6 = { .family = AF_INET6, + .obj_size = sizeof(struct tcp6_request_sock), .rtx_syn_ack = tcp_v6_send_synack, .send_ack = tcp_v6_or_send_ack, .destructor = tcp_v6_or_free, @@ -1221,7 +1225,7 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb) static void tcp_v6_or_send_ack(struct sk_buff *skb, struct open_request *req) { - tcp_v6_send_ack(skb, req->snt_isn+1, req->rcv_isn+1, req->rcv_wnd, req->ts_recent); + tcp_v6_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } @@ -1264,7 +1268,7 @@ static void tcp_v6_synq_add(struct sock *sk, struct open_request *req) { struct tcp_sock *tp = tcp_sk(sk); struct tcp_listen_opt *lopt = tp->listen_opt; - u32 h = tcp_v6_synq_hash(&req->af.v6_req.rmt_addr, req->rmt_port, lopt->hash_rnd); + u32 h = tcp_v6_synq_hash(&tcp6_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port, lopt->hash_rnd); req->sk = NULL; req->expires = jiffies + TCP_TIMEOUT_INIT; @@ -1284,6 +1288,7 @@ static void tcp_v6_synq_add(struct sock *sk, struct open_request *req) */ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) { + struct tcp6_request_sock *treq; struct ipv6_pinfo *np = inet6_sk(sk); struct tcp_options_received tmp_opt; struct tcp_sock *tp = tcp_sk(sk); @@ -1308,7 +1313,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (sk_acceptq_is_full(sk) && tcp_synq_young(sk) > 1) goto drop; - req = tcp_openreq_alloc(); + req = tcp_openreq_alloc(&or_ipv6); if (req == NULL) goto drop; @@ -1321,28 +1326,28 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) tmp_opt.tstamp_ok = tmp_opt.saw_tstamp; tcp_openreq_init(req, &tmp_opt, skb); - req->class = &or_ipv6; - ipv6_addr_copy(&req->af.v6_req.rmt_addr, &skb->nh.ipv6h->saddr); - ipv6_addr_copy(&req->af.v6_req.loc_addr, &skb->nh.ipv6h->daddr); + treq = tcp6_rsk(req); + ipv6_addr_copy(&treq->rmt_addr, &skb->nh.ipv6h->saddr); + ipv6_addr_copy(&treq->loc_addr, &skb->nh.ipv6h->daddr); TCP_ECN_create_request(req, skb->h.th); - req->af.v6_req.pktopts = NULL; + treq->pktopts = NULL; if (ipv6_opt_accepted(sk, skb) || np->rxopt.bits.rxinfo || np->rxopt.bits.rxhlim) { atomic_inc(&skb->users); - req->af.v6_req.pktopts = skb; + treq->pktopts = skb; } - req->af.v6_req.iif = sk->sk_bound_dev_if; + treq->iif = sk->sk_bound_dev_if; /* So that link locals have meaning */ if (!sk->sk_bound_dev_if && - ipv6_addr_type(&req->af.v6_req.rmt_addr) & IPV6_ADDR_LINKLOCAL) - req->af.v6_req.iif = tcp_v6_iif(skb); + ipv6_addr_type(&treq->rmt_addr) & IPV6_ADDR_LINKLOCAL) + treq->iif = tcp_v6_iif(skb); if (isn == 0) isn = tcp_v6_init_sequence(sk,skb); - req->snt_isn = isn; + tcp_rsk(req)->snt_isn = isn; if (tcp_v6_send_synack(sk, req, NULL)) goto drop; @@ -1363,6 +1368,7 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, struct open_request *req, struct dst_entry *dst) { + struct tcp6_request_sock *treq = tcp6_rsk(req); struct ipv6_pinfo *newnp, *np = inet6_sk(sk); struct tcp6_sock *newtcp6sk; struct inet_sock *newinet; @@ -1426,10 +1432,10 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, goto out_overflow; if (np->rxopt.bits.srcrt == 2 && - opt == NULL && req->af.v6_req.pktopts) { - struct inet6_skb_parm *rxopt = IP6CB(req->af.v6_req.pktopts); + opt == NULL && treq->pktopts) { + struct inet6_skb_parm *rxopt = IP6CB(treq->pktopts); if (rxopt->srcrt) - opt = ipv6_invert_rthdr(sk, (struct ipv6_rt_hdr*)(req->af.v6_req.pktopts->nh.raw+rxopt->srcrt)); + opt = ipv6_invert_rthdr(sk, (struct ipv6_rt_hdr *)(treq->pktopts->nh.raw + rxopt->srcrt)); } if (dst == NULL) { @@ -1438,16 +1444,16 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, memset(&fl, 0, sizeof(fl)); fl.proto = IPPROTO_TCP; - ipv6_addr_copy(&fl.fl6_dst, &req->af.v6_req.rmt_addr); + ipv6_addr_copy(&fl.fl6_dst, &treq->rmt_addr); if (opt && opt->srcrt) { struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt; ipv6_addr_copy(&final, &fl.fl6_dst); ipv6_addr_copy(&fl.fl6_dst, rt0->addr); final_p = &final; } - ipv6_addr_copy(&fl.fl6_src, &req->af.v6_req.loc_addr); + ipv6_addr_copy(&fl.fl6_src, &treq->loc_addr); fl.oif = sk->sk_bound_dev_if; - fl.fl_ip_dport = req->rmt_port; + fl.fl_ip_dport = inet_rsk(req)->rmt_port; fl.fl_ip_sport = inet_sk(sk)->sport; if (ip6_dst_lookup(sk, &dst, &fl)) @@ -1482,10 +1488,10 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, memcpy(newnp, np, sizeof(struct ipv6_pinfo)); - ipv6_addr_copy(&newnp->daddr, &req->af.v6_req.rmt_addr); - ipv6_addr_copy(&newnp->saddr, &req->af.v6_req.loc_addr); - ipv6_addr_copy(&newnp->rcv_saddr, &req->af.v6_req.loc_addr); - newsk->sk_bound_dev_if = req->af.v6_req.iif; + ipv6_addr_copy(&newnp->daddr, &treq->rmt_addr); + ipv6_addr_copy(&newnp->saddr, &treq->loc_addr); + ipv6_addr_copy(&newnp->rcv_saddr, &treq->loc_addr); + newsk->sk_bound_dev_if = treq->iif; /* Now IPv6 options... @@ -1498,11 +1504,10 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, /* Clone pktoptions received with SYN */ newnp->pktoptions = NULL; - if (req->af.v6_req.pktopts) { - newnp->pktoptions = skb_clone(req->af.v6_req.pktopts, - GFP_ATOMIC); - kfree_skb(req->af.v6_req.pktopts); - req->af.v6_req.pktopts = NULL; + if (treq->pktopts != NULL) { + newnp->pktoptions = skb_clone(treq->pktopts, GFP_ATOMIC); + kfree_skb(treq->pktopts); + treq->pktopts = NULL; if (newnp->pktoptions) skb_set_owner_r(newnp->pktoptions, newsk); } @@ -2058,8 +2063,8 @@ static void get_openreq6(struct seq_file *seq, if (ttd < 0) ttd = 0; - src = &req->af.v6_req.loc_addr; - dest = &req->af.v6_req.rmt_addr; + src = &tcp6_rsk(req)->loc_addr; + dest = &tcp6_rsk(req)->rmt_addr; seq_printf(seq, "%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X " "%02X %08X:%08X %02X:%08lX %08X %5d %8d %d %d %p\n", @@ -2069,7 +2074,7 @@ static void get_openreq6(struct seq_file *seq, ntohs(inet_sk(sk)->sport), dest->s6_addr32[0], dest->s6_addr32[1], dest->s6_addr32[2], dest->s6_addr32[3], - ntohs(req->rmt_port), + ntohs(inet_rsk(req)->rmt_port), TCP_SYN_RECV, 0,0, /* could print option size, but that is af dependent. */ 1, /* timers active (only the expire timer) */ @@ -2239,6 +2244,7 @@ struct proto tcpv6_prot = { .sysctl_rmem = sysctl_tcp_rmem, .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp6_sock), + .rsk_prot = &or_ipv6, }; static struct inet6_protocol tcpv6_protocol = { -- cgit v1.2.3 From 60236fdd08b2169045a3bbfc5ffe1576e6c3c17b Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sat, 18 Jun 2005 22:47:21 -0700 Subject: [NET] Rename open_request to request_sock Ok, this one just renames some stuff to have a better namespace and to dissassociate it from TCP: struct open_request -> struct request_sock tcp_openreq_alloc -> reqsk_alloc tcp_openreq_free -> reqsk_free tcp_openreq_fastfree -> __reqsk_free With this most of the infrastructure closely resembles a struct sock methods subset. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/ip.h | 4 ++-- include/linux/ipv6.h | 2 +- include/linux/tcp.h | 6 ++--- include/net/request_sock.h | 34 ++++++++++++++-------------- include/net/sock.h | 4 ++-- include/net/tcp.h | 30 ++++++++++++------------- include/net/tcp_ecn.h | 6 ++--- net/ipv4/syncookies.c | 12 +++++----- net/ipv4/tcp.c | 12 +++++----- net/ipv4/tcp_diag.c | 4 ++-- net/ipv4/tcp_ipv4.c | 56 +++++++++++++++++++++++----------------------- net/ipv4/tcp_minisocks.c | 14 ++++++------ net/ipv4/tcp_output.c | 2 +- net/ipv4/tcp_timer.c | 6 ++--- net/ipv6/tcp_ipv6.c | 42 +++++++++++++++++----------------- 15 files changed, 117 insertions(+), 117 deletions(-) (limited to 'include') diff --git a/include/linux/ip.h b/include/linux/ip.h index d5b7c907204e..31e7cedd9f84 100644 --- a/include/linux/ip.h +++ b/include/linux/ip.h @@ -109,7 +109,7 @@ struct ip_options { #define optlength(opt) (sizeof(struct ip_options) + opt->optlen) struct inet_request_sock { - struct open_request req; + struct request_sock req; u32 loc_addr; u32 rmt_addr; u16 rmt_port; @@ -123,7 +123,7 @@ struct inet_request_sock { struct ip_options *opt; }; -static inline struct inet_request_sock *inet_rsk(const struct open_request *sk) +static inline struct inet_request_sock *inet_rsk(const struct request_sock *sk) { return (struct inet_request_sock *)sk; } diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 98acdbf3d446..6fcd6a0ade24 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -201,7 +201,7 @@ struct tcp6_request_sock { int iif; }; -static inline struct tcp6_request_sock *tcp6_rsk(const struct open_request *sk) +static inline struct tcp6_request_sock *tcp6_rsk(const struct request_sock *sk) { return (struct tcp6_request_sock *)sk; } diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 86771b37b80d..fb54292a15aa 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -236,7 +236,7 @@ struct tcp_request_sock { __u32 snt_isn; }; -static inline struct tcp_request_sock *tcp_rsk(const struct open_request *req) +static inline struct tcp_request_sock *tcp_rsk(const struct request_sock *req) { return (struct tcp_request_sock *)req; } @@ -393,8 +393,8 @@ struct tcp_sock { struct tcp_listen_opt *listen_opt; /* FIFO of established children */ - struct open_request *accept_queue; - struct open_request *accept_queue_tail; + struct request_sock *accept_queue; + struct request_sock *accept_queue_tail; unsigned int keepalive_time; /* time before keep alive takes place */ unsigned int keepalive_intvl; /* time interval between keep alive probes */ diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 9502f5587931..08a8fd1d1610 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -19,28 +19,28 @@ #include #include -struct open_request; +struct request_sock; struct sk_buff; struct dst_entry; struct proto; -struct or_calltable { +struct request_sock_ops { int family; kmem_cache_t *slab; int obj_size; int (*rtx_syn_ack)(struct sock *sk, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst); void (*send_ack)(struct sk_buff *skb, - struct open_request *req); + struct request_sock *req); void (*send_reset)(struct sk_buff *skb); - void (*destructor)(struct open_request *req); + void (*destructor)(struct request_sock *req); }; -/* struct open_request - mini sock to represent a connection request +/* struct request_sock - mini sock to represent a connection request */ -struct open_request { - struct open_request *dl_next; /* Must be first member! */ +struct request_sock { + struct request_sock *dl_next; /* Must be first member! */ u16 mss; u8 retrans; u8 __pad; @@ -49,29 +49,29 @@ struct open_request { u32 rcv_wnd; /* rcv_wnd offered first time */ u32 ts_recent; unsigned long expires; - struct or_calltable *class; + struct request_sock_ops *rsk_ops; struct sock *sk; }; -static inline struct open_request *tcp_openreq_alloc(struct or_calltable *class) +static inline struct request_sock *reqsk_alloc(struct request_sock_ops *ops) { - struct open_request *req = kmem_cache_alloc(class->slab, SLAB_ATOMIC); + struct request_sock *req = kmem_cache_alloc(ops->slab, SLAB_ATOMIC); if (req != NULL) - req->class = class; + req->rsk_ops = ops; return req; } -static inline void tcp_openreq_fastfree(struct open_request *req) +static inline void __reqsk_free(struct request_sock *req) { - kmem_cache_free(req->class->slab, req); + kmem_cache_free(req->rsk_ops->slab, req); } -static inline void tcp_openreq_free(struct open_request *req) +static inline void reqsk_free(struct request_sock *req) { - req->class->destructor(req); - tcp_openreq_fastfree(req); + req->rsk_ops->destructor(req); + __reqsk_free(req); } #endif /* _REQUEST_SOCK_H */ diff --git a/include/net/sock.h b/include/net/sock.h index 6919276af8af..e593af5b1ecc 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -484,7 +484,7 @@ extern void sk_stream_kill_queues(struct sock *sk); extern int sk_wait_data(struct sock *sk, long *timeo); -struct or_calltable; +struct request_sock_ops; /* Networking protocol blocks we attach to sockets. * socket layer -> transport layer interface @@ -549,7 +549,7 @@ struct proto { kmem_cache_t *slab; unsigned int obj_size; - struct or_calltable *rsk_prot; + struct request_sock_ops *rsk_prot; struct module *owner; diff --git a/include/net/tcp.h b/include/net/tcp.h index d438ba566b89..6663086a5e35 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -641,7 +641,7 @@ struct tcp_func { struct sock * (*syn_recv_sock) (struct sock *sk, struct sk_buff *skb, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst); int (*remember_stamp) (struct sock *sk); @@ -785,8 +785,8 @@ extern enum tcp_tw_status tcp_timewait_state_process(struct tcp_tw_bucket *tw, unsigned len); extern struct sock * tcp_check_req(struct sock *sk,struct sk_buff *skb, - struct open_request *req, - struct open_request **prev); + struct request_sock *req, + struct request_sock **prev); extern int tcp_child_process(struct sock *parent, struct sock *child, struct sk_buff *skb); @@ -836,12 +836,12 @@ extern int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb); extern struct sock * tcp_create_openreq_child(struct sock *sk, - struct open_request *req, + struct request_sock *req, struct sk_buff *skb); extern struct sock * tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst); extern int tcp_v4_do_rcv(struct sock *sk, @@ -855,7 +855,7 @@ extern int tcp_connect(struct sock *sk); extern struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, - struct open_request *req); + struct request_sock *req); extern int tcp_disconnect(struct sock *sk, int flags); @@ -1683,7 +1683,7 @@ static inline int tcp_full_space(const struct sock *sk) return tcp_win_from_space(sk->sk_rcvbuf); } -static inline void tcp_acceptq_queue(struct sock *sk, struct open_request *req, +static inline void tcp_acceptq_queue(struct sock *sk, struct request_sock *req, struct sock *child) { struct tcp_sock *tp = tcp_sk(sk); @@ -1707,11 +1707,11 @@ struct tcp_listen_opt int qlen_young; int clock_hand; u32 hash_rnd; - struct open_request *syn_table[TCP_SYNQ_HSIZE]; + struct request_sock *syn_table[TCP_SYNQ_HSIZE]; }; static inline void -tcp_synq_removed(struct sock *sk, struct open_request *req) +tcp_synq_removed(struct sock *sk, struct request_sock *req) { struct tcp_listen_opt *lopt = tcp_sk(sk)->listen_opt; @@ -1745,23 +1745,23 @@ static inline int tcp_synq_is_full(struct sock *sk) return tcp_synq_len(sk) >> tcp_sk(sk)->listen_opt->max_qlen_log; } -static inline void tcp_synq_unlink(struct tcp_sock *tp, struct open_request *req, - struct open_request **prev) +static inline void tcp_synq_unlink(struct tcp_sock *tp, struct request_sock *req, + struct request_sock **prev) { write_lock(&tp->syn_wait_lock); *prev = req->dl_next; write_unlock(&tp->syn_wait_lock); } -static inline void tcp_synq_drop(struct sock *sk, struct open_request *req, - struct open_request **prev) +static inline void tcp_synq_drop(struct sock *sk, struct request_sock *req, + struct request_sock **prev) { tcp_synq_unlink(tcp_sk(sk), req, prev); tcp_synq_removed(sk, req); - tcp_openreq_free(req); + reqsk_free(req); } -static __inline__ void tcp_openreq_init(struct open_request *req, +static __inline__ void tcp_openreq_init(struct request_sock *req, struct tcp_options_received *rx_opt, struct sk_buff *skb) { diff --git a/include/net/tcp_ecn.h b/include/net/tcp_ecn.h index 94ad970e844a..64980ee8c92a 100644 --- a/include/net/tcp_ecn.h +++ b/include/net/tcp_ecn.h @@ -39,7 +39,7 @@ static inline void TCP_ECN_send_syn(struct sock *sk, struct tcp_sock *tp, } static __inline__ void -TCP_ECN_make_synack(struct open_request *req, struct tcphdr *th) +TCP_ECN_make_synack(struct request_sock *req, struct tcphdr *th) { if (inet_rsk(req)->ecn_ok) th->ece = 1; @@ -112,13 +112,13 @@ static inline int TCP_ECN_rcv_ecn_echo(struct tcp_sock *tp, struct tcphdr *th) } static inline void TCP_ECN_openreq_child(struct tcp_sock *tp, - struct open_request *req) + struct request_sock *req) { tp->ecn_flags = inet_rsk(req)->ecn_ok ? TCP_ECN_OK : 0; } static __inline__ void -TCP_ECN_create_request(struct open_request *req, struct tcphdr *th) +TCP_ECN_create_request(struct request_sock *req, struct tcphdr *th) { if (sysctl_tcp_ecn && th->ece && th->cwr) inet_rsk(req)->ecn_ok = 1; diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index dd47e6da6fb3..72d014442185 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -169,10 +169,10 @@ static inline int cookie_check(struct sk_buff *skb, __u32 cookie) return mssind < NUM_MSS ? msstab[mssind] + 1 : 0; } -extern struct or_calltable or_ipv4; +extern struct request_sock_ops tcp_request_sock_ops; static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst) { struct tcp_sock *tp = tcp_sk(sk); @@ -182,7 +182,7 @@ static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb, if (child) tcp_acceptq_queue(sk, req, child); else - tcp_openreq_free(req); + reqsk_free(req); return child; } @@ -195,7 +195,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, struct tcp_sock *tp = tcp_sk(sk); __u32 cookie = ntohl(skb->h.th->ack_seq) - 1; struct sock *ret = sk; - struct open_request *req; + struct request_sock *req; int mss; struct rtable *rt; __u8 rcv_wscale; @@ -212,7 +212,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESRECV); ret = NULL; - req = tcp_openreq_alloc(&or_ipv4); /* for safety */ + req = reqsk_alloc(&tcp_request_sock_ops); /* for safety */ if (!req) goto out; @@ -262,7 +262,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, { .sport = skb->h.th->dest, .dport = skb->h.th->source } } }; if (ip_route_output_key(&rt, &fl)) { - tcp_openreq_free(req); + reqsk_free(req); goto out; } } diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index a3cabfa2022a..1c29feb6b35f 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -516,8 +516,8 @@ static void tcp_listen_stop (struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); struct tcp_listen_opt *lopt = tp->listen_opt; - struct open_request *acc_req = tp->accept_queue; - struct open_request *req; + struct request_sock *acc_req = tp->accept_queue; + struct request_sock *req; int i; tcp_delete_keepalive_timer(sk); @@ -533,7 +533,7 @@ static void tcp_listen_stop (struct sock *sk) while ((req = lopt->syn_table[i]) != NULL) { lopt->syn_table[i] = req->dl_next; lopt->qlen--; - tcp_openreq_free(req); + reqsk_free(req); /* Following specs, it would be better either to send FIN * (and enter FIN-WAIT-1, it is normal close) @@ -573,7 +573,7 @@ static void tcp_listen_stop (struct sock *sk) sock_put(child); sk_acceptq_removed(sk); - tcp_openreq_fastfree(req); + __reqsk_free(req); } BUG_TRAP(!sk->sk_ack_backlog); } @@ -1894,7 +1894,7 @@ static int wait_for_connect(struct sock *sk, long timeo) struct sock *tcp_accept(struct sock *sk, int flags, int *err) { struct tcp_sock *tp = tcp_sk(sk); - struct open_request *req; + struct request_sock *req; struct sock *newsk; int error; @@ -1927,7 +1927,7 @@ struct sock *tcp_accept(struct sock *sk, int flags, int *err) newsk = req->sk; sk_acceptq_removed(sk); - tcp_openreq_fastfree(req); + __reqsk_free(req); BUG_TRAP(newsk->sk_state != TCP_SYN_RECV); release_sock(sk); return newsk; diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index 700ff2413588..67277800d0c1 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -455,7 +455,7 @@ static int tcpdiag_dump_sock(struct sk_buff *skb, struct sock *sk, } static int tcpdiag_fill_req(struct sk_buff *skb, struct sock *sk, - struct open_request *req, + struct request_sock *req, u32 pid, u32 seq) { const struct inet_request_sock *ireq = inet_rsk(req); @@ -542,7 +542,7 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, } for (j = s_j; j < TCP_SYNQ_HSIZE; j++) { - struct open_request *req, *head = lopt->syn_table[j]; + struct request_sock *req, *head = lopt->syn_table[j]; reqnum = 0; for (req = head; req; reqnum++, req = req->dl_next) { diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index e156be90df14..95528a75a63d 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -36,7 +36,7 @@ * ACK bit. * Andi Kleen : Implemented fast path mtu discovery. * Fixed many serious bugs in the - * open_request handling and moved + * request_sock handling and moved * most of it into the af independent code. * Added tail drop and some other bugfixes. * Added new listen sematics. @@ -869,13 +869,13 @@ static __inline__ u32 tcp_v4_synq_hash(u32 raddr, u16 rport, u32 rnd) return (jhash_2words(raddr, (u32) rport, rnd) & (TCP_SYNQ_HSIZE - 1)); } -static struct open_request *tcp_v4_search_req(struct tcp_sock *tp, - struct open_request ***prevp, +static struct request_sock *tcp_v4_search_req(struct tcp_sock *tp, + struct request_sock ***prevp, __u16 rport, __u32 raddr, __u32 laddr) { struct tcp_listen_opt *lopt = tp->listen_opt; - struct open_request *req, **prev; + struct request_sock *req, **prev; for (prev = &lopt->syn_table[tcp_v4_synq_hash(raddr, rport, lopt->hash_rnd)]; (req = *prev) != NULL; @@ -885,7 +885,7 @@ static struct open_request *tcp_v4_search_req(struct tcp_sock *tp, if (ireq->rmt_port == rport && ireq->rmt_addr == raddr && ireq->loc_addr == laddr && - TCP_INET_FAMILY(req->class->family)) { + TCP_INET_FAMILY(req->rsk_ops->family)) { BUG_TRAP(!req->sk); *prevp = prev; break; @@ -895,7 +895,7 @@ static struct open_request *tcp_v4_search_req(struct tcp_sock *tp, return req; } -static void tcp_v4_synq_add(struct sock *sk, struct open_request *req) +static void tcp_v4_synq_add(struct sock *sk, struct request_sock *req) { struct tcp_sock *tp = tcp_sk(sk); struct tcp_listen_opt *lopt = tp->listen_opt; @@ -1052,7 +1052,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info) } switch (sk->sk_state) { - struct open_request *req, **prev; + struct request_sock *req, **prev; case TCP_LISTEN: if (sock_owned_by_user(sk)) goto out; @@ -1256,14 +1256,14 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb) tcp_tw_put(tw); } -static void tcp_v4_or_send_ack(struct sk_buff *skb, struct open_request *req) +static void tcp_v4_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req) { tcp_v4_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } static struct dst_entry* tcp_v4_route_req(struct sock *sk, - struct open_request *req) + struct request_sock *req) { struct rtable *rt; const struct inet_request_sock *ireq = inet_rsk(req); @@ -1294,10 +1294,10 @@ static struct dst_entry* tcp_v4_route_req(struct sock *sk, /* * Send a SYN-ACK after having received an ACK. - * This still operates on a open_request only, not on a big + * This still operates on a request_sock only, not on a big * socket. */ -static int tcp_v4_send_synack(struct sock *sk, struct open_request *req, +static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req, struct dst_entry *dst) { const struct inet_request_sock *ireq = inet_rsk(req); @@ -1332,9 +1332,9 @@ out: } /* - * IPv4 open_request destructor. + * IPv4 request_sock destructor. */ -static void tcp_v4_or_free(struct open_request *req) +static void tcp_v4_reqsk_destructor(struct request_sock *req) { if (inet_rsk(req)->opt) kfree(inet_rsk(req)->opt); @@ -1353,7 +1353,7 @@ static inline void syn_flood_warning(struct sk_buff *skb) } /* - * Save and compile IPv4 options into the open_request if needed. + * Save and compile IPv4 options into the request_sock if needed. */ static inline struct ip_options *tcp_v4_save_options(struct sock *sk, struct sk_buff *skb) @@ -1389,12 +1389,12 @@ static inline struct ip_options *tcp_v4_save_options(struct sock *sk, */ int sysctl_max_syn_backlog = 256; -struct or_calltable or_ipv4 = { +struct request_sock_ops tcp_request_sock_ops = { .family = PF_INET, .obj_size = sizeof(struct tcp_request_sock), .rtx_syn_ack = tcp_v4_send_synack, - .send_ack = tcp_v4_or_send_ack, - .destructor = tcp_v4_or_free, + .send_ack = tcp_v4_reqsk_send_ack, + .destructor = tcp_v4_reqsk_destructor, .send_reset = tcp_v4_send_reset, }; @@ -1402,7 +1402,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) { struct inet_request_sock *ireq; struct tcp_options_received tmp_opt; - struct open_request *req; + struct request_sock *req; __u32 saddr = skb->nh.iph->saddr; __u32 daddr = skb->nh.iph->daddr; __u32 isn = TCP_SKB_CB(skb)->when; @@ -1439,7 +1439,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) if (sk_acceptq_is_full(sk) && tcp_synq_young(sk) > 1) goto drop; - req = tcp_openreq_alloc(&or_ipv4); + req = reqsk_alloc(&tcp_request_sock_ops); if (!req) goto drop; @@ -1535,14 +1535,14 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) goto drop_and_free; if (want_cookie) { - tcp_openreq_free(req); + reqsk_free(req); } else { tcp_v4_synq_add(sk, req); } return 0; drop_and_free: - tcp_openreq_free(req); + reqsk_free(req); drop: TCP_INC_STATS_BH(TCP_MIB_ATTEMPTFAILS); return 0; @@ -1554,7 +1554,7 @@ drop: * now create the new socket. */ struct sock *tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst) { struct inet_request_sock *ireq; @@ -1613,9 +1613,9 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) struct iphdr *iph = skb->nh.iph; struct tcp_sock *tp = tcp_sk(sk); struct sock *nsk; - struct open_request **prev; + struct request_sock **prev; /* Find possible connection requests. */ - struct open_request *req = tcp_v4_search_req(tp, &prev, th->source, + struct request_sock *req = tcp_v4_search_req(tp, &prev, th->source, iph->saddr, iph->daddr); if (req) return tcp_check_req(sk, skb, req, prev); @@ -2152,13 +2152,13 @@ static void *listening_get_next(struct seq_file *seq, void *cur) ++st->num; if (st->state == TCP_SEQ_STATE_OPENREQ) { - struct open_request *req = cur; + struct request_sock *req = cur; tp = tcp_sk(st->syn_wait_sk); req = req->dl_next; while (1) { while (req) { - if (req->class->family == st->family) { + if (req->rsk_ops->family == st->family) { cur = req; goto out; } @@ -2459,7 +2459,7 @@ void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo) memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops)); } -static void get_openreq4(struct sock *sk, struct open_request *req, +static void get_openreq4(struct sock *sk, struct request_sock *req, char *tmpbuf, int i, int uid) { const struct inet_request_sock *ireq = inet_rsk(req); @@ -2627,7 +2627,7 @@ struct proto tcp_prot = { .sysctl_rmem = sysctl_tcp_rmem, .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp_sock), - .rsk_prot = &or_ipv4, + .rsk_prot = &tcp_request_sock_ops, }; diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 1037401c7cc8..0e6d525a8341 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -684,7 +684,7 @@ out: * Actually, we could lots of memory writes here. tp of listening * socket contains all necessary default parameters. */ -struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, struct sk_buff *skb) +struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, struct sk_buff *skb) { /* allocate the newsk from the same slab of the master sock, * if not, at sk_free time we'll try to free it from the wrong @@ -853,12 +853,12 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct open_request *req, /* * Process an incoming packet for SYN_RECV sockets represented - * as an open_request. + * as a request_sock. */ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, - struct open_request *req, - struct open_request **prev) + struct request_sock *req, + struct request_sock **prev) { struct tcphdr *th = skb->h.th; struct tcp_sock *tp = tcp_sk(sk); @@ -903,7 +903,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, * Enforce "SYN-ACK" according to figure 8, figure 6 * of RFC793, fixed by RFC1122. */ - req->class->rtx_syn_ack(sk, req, NULL); + req->rsk_ops->rtx_syn_ack(sk, req, NULL); return NULL; } @@ -975,7 +975,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, tcp_rsk(req)->rcv_isn + 1, tcp_rsk(req)->rcv_isn + 1 + req->rcv_wnd)) { /* Out of window: send ACK and drop. */ if (!(flg & TCP_FLAG_RST)) - req->class->send_ack(skb, req); + req->rsk_ops->send_ack(skb, req); if (paws_reject) NET_INC_STATS_BH(LINUX_MIB_PAWSESTABREJECTED); return NULL; @@ -1035,7 +1035,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, embryonic_reset: NET_INC_STATS_BH(LINUX_MIB_EMBRYONICRSTS); if (!(flg & TCP_FLAG_RST)) - req->class->send_reset(skb); + req->rsk_ops->send_reset(skb); tcp_synq_drop(sk, req, prev); return NULL; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index f3c8747caf91..f17c6577e337 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1356,7 +1356,7 @@ int tcp_send_synack(struct sock *sk) * Prepare a SYN-ACK. */ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, - struct open_request *req) + struct request_sock *req) { struct inet_request_sock *ireq = inet_rsk(req); struct tcp_sock *tp = tcp_sk(sk); diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index ba30ca0aa6a3..f03efe5fb76a 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -468,7 +468,7 @@ static void tcp_synack_timer(struct sock *sk) int max_retries = tp->syn_retries ? : sysctl_tcp_synack_retries; int thresh = max_retries; unsigned long now = jiffies; - struct open_request **reqp, *req; + struct request_sock **reqp, *req; int i, budget; if (lopt == NULL || lopt->qlen == 0) @@ -514,7 +514,7 @@ static void tcp_synack_timer(struct sock *sk) if (time_after_eq(now, req->expires)) { if ((req->retrans < thresh || (inet_rsk(req)->acked && req->retrans < max_retries)) - && !req->class->rtx_syn_ack(sk, req, NULL)) { + && !req->rsk_ops->rtx_syn_ack(sk, req, NULL)) { unsigned long timeo; if (req->retrans++ == 0) @@ -533,7 +533,7 @@ static void tcp_synack_timer(struct sock *sk) lopt->qlen--; if (req->retrans == 0) lopt->qlen_young--; - tcp_openreq_free(req); + reqsk_free(req); continue; } reqp = &req->dl_next; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 9199ad2fde0d..068cd4a8c292 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -65,7 +65,7 @@ #include static void tcp_v6_send_reset(struct sk_buff *skb); -static void tcp_v6_or_send_ack(struct sk_buff *skb, struct open_request *req); +static void tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req); static void tcp_v6_send_check(struct sock *sk, struct tcphdr *th, int len, struct sk_buff *skb); @@ -394,15 +394,15 @@ static u32 tcp_v6_synq_hash(struct in6_addr *raddr, u16 rport, u32 rnd) return c & (TCP_SYNQ_HSIZE - 1); } -static struct open_request *tcp_v6_search_req(struct tcp_sock *tp, - struct open_request ***prevp, +static struct request_sock *tcp_v6_search_req(struct tcp_sock *tp, + struct request_sock ***prevp, __u16 rport, struct in6_addr *raddr, struct in6_addr *laddr, int iif) { struct tcp_listen_opt *lopt = tp->listen_opt; - struct open_request *req, **prev; + struct request_sock *req, **prev; for (prev = &lopt->syn_table[tcp_v6_synq_hash(raddr, rport, lopt->hash_rnd)]; (req = *prev) != NULL; @@ -410,7 +410,7 @@ static struct open_request *tcp_v6_search_req(struct tcp_sock *tp, const struct tcp6_request_sock *treq = tcp6_rsk(req); if (inet_rsk(req)->rmt_port == rport && - req->class->family == AF_INET6 && + req->rsk_ops->family == AF_INET6 && ipv6_addr_equal(&treq->rmt_addr, raddr) && ipv6_addr_equal(&treq->loc_addr, laddr) && (!treq->iif || treq->iif == iif)) { @@ -908,9 +908,9 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, icmpv6_err_convert(type, code, &err); - /* Might be for an open_request */ + /* Might be for an request_sock */ switch (sk->sk_state) { - struct open_request *req, **prev; + struct request_sock *req, **prev; case TCP_LISTEN: if (sock_owned_by_user(sk)) goto out; @@ -959,7 +959,7 @@ out: } -static int tcp_v6_send_synack(struct sock *sk, struct open_request *req, +static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req, struct dst_entry *dst) { struct tcp6_request_sock *treq = tcp6_rsk(req); @@ -1027,18 +1027,18 @@ done: return err; } -static void tcp_v6_or_free(struct open_request *req) +static void tcp_v6_reqsk_destructor(struct request_sock *req) { if (tcp6_rsk(req)->pktopts) kfree_skb(tcp6_rsk(req)->pktopts); } -static struct or_calltable or_ipv6 = { +static struct request_sock_ops tcp6_request_sock_ops = { .family = AF_INET6, .obj_size = sizeof(struct tcp6_request_sock), .rtx_syn_ack = tcp_v6_send_synack, - .send_ack = tcp_v6_or_send_ack, - .destructor = tcp_v6_or_free, + .send_ack = tcp_v6_reqsk_send_ack, + .destructor = tcp_v6_reqsk_destructor, .send_reset = tcp_v6_send_reset }; @@ -1223,7 +1223,7 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb) tcp_tw_put(tw); } -static void tcp_v6_or_send_ack(struct sk_buff *skb, struct open_request *req) +static void tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req) { tcp_v6_send_ack(skb, tcp_rsk(req)->snt_isn + 1, tcp_rsk(req)->rcv_isn + 1, req->rcv_wnd, req->ts_recent); } @@ -1231,7 +1231,7 @@ static void tcp_v6_or_send_ack(struct sk_buff *skb, struct open_request *req) static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) { - struct open_request *req, **prev; + struct request_sock *req, **prev; struct tcphdr *th = skb->h.th; struct tcp_sock *tp = tcp_sk(sk); struct sock *nsk; @@ -1264,7 +1264,7 @@ static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) return sk; } -static void tcp_v6_synq_add(struct sock *sk, struct open_request *req) +static void tcp_v6_synq_add(struct sock *sk, struct request_sock *req) { struct tcp_sock *tp = tcp_sk(sk); struct tcp_listen_opt *lopt = tp->listen_opt; @@ -1292,7 +1292,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) struct ipv6_pinfo *np = inet6_sk(sk); struct tcp_options_received tmp_opt; struct tcp_sock *tp = tcp_sk(sk); - struct open_request *req = NULL; + struct request_sock *req = NULL; __u32 isn = TCP_SKB_CB(skb)->when; if (skb->protocol == htons(ETH_P_IP)) @@ -1313,7 +1313,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (sk_acceptq_is_full(sk) && tcp_synq_young(sk) > 1) goto drop; - req = tcp_openreq_alloc(&or_ipv6); + req = reqsk_alloc(&tcp6_request_sock_ops); if (req == NULL) goto drop; @@ -1358,14 +1358,14 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) drop: if (req) - tcp_openreq_free(req); + reqsk_free(req); TCP_INC_STATS_BH(TCP_MIB_ATTEMPTFAILS); return 0; /* don't send reset */ } static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, - struct open_request *req, + struct request_sock *req, struct dst_entry *dst) { struct tcp6_request_sock *treq = tcp6_rsk(req); @@ -2055,7 +2055,7 @@ static int tcp_v6_destroy_sock(struct sock *sk) /* Proc filesystem TCPv6 sock list dumping. */ static void get_openreq6(struct seq_file *seq, - struct sock *sk, struct open_request *req, int i, int uid) + struct sock *sk, struct request_sock *req, int i, int uid) { struct in6_addr *dest, *src; int ttd = req->expires - jiffies; @@ -2244,7 +2244,7 @@ struct proto tcpv6_prot = { .sysctl_rmem = sysctl_tcp_rmem, .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp6_sock), - .rsk_prot = &or_ipv6, + .rsk_prot = &tcp6_request_sock_ops, }; static struct inet6_protocol tcpv6_protocol = { -- cgit v1.2.3 From 0e87506fcc734647c7b2497eee4eb81e785c857a Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sat, 18 Jun 2005 22:47:59 -0700 Subject: [NET] Generalise tcp_listen_opt This chunks out the accept_queue and tcp_listen_opt code and moves them to net/core/request_sock.c and include/net/request_sock.h, to make it useful for other transport protocols, DCCP being the first one to use it. Next patches will rename tcp_listen_opt to accept_sock and remove the inline tcp functions that just call a reqsk_queue_ function. Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/linux/tcp.h | 17 +---- include/net/request_sock.h | 178 +++++++++++++++++++++++++++++++++++++++++++++ include/net/tcp.h | 46 ++---------- net/core/Makefile | 3 +- net/core/request_sock.c | 48 ++++++++++++ net/ipv4/tcp.c | 67 ++++++----------- net/ipv4/tcp_diag.c | 6 +- net/ipv4/tcp_ipv4.c | 32 +++----- net/ipv4/tcp_minisocks.c | 6 +- net/ipv4/tcp_timer.c | 10 +-- net/ipv6/tcp_ipv6.c | 14 +--- 11 files changed, 281 insertions(+), 146 deletions(-) create mode 100644 net/core/request_sock.c (limited to 'include') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index fb54292a15aa..97a7c9e03df5 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -379,22 +379,7 @@ struct tcp_sock { __u32 total_retrans; /* Total retransmits for entire connection */ - /* The syn_wait_lock is necessary only to avoid proc interface having - * to grab the main lock sock while browsing the listening hash - * (otherwise it's deadlock prone). - * This lock is acquired in read mode only from listening_get_next() - * and it's acquired in write mode _only_ from code that is actively - * changing the syn_wait_queue. All readers that are holding - * the master sock lock don't need to grab this lock in read mode - * too as the syn_wait_queue writes are always protected from - * the main sock lock. - */ - rwlock_t syn_wait_lock; - struct tcp_listen_opt *listen_opt; - - /* FIFO of established children */ - struct request_sock *accept_queue; - struct request_sock *accept_queue_tail; + struct request_sock_queue accept_queue; /* FIFO of established children */ unsigned int keepalive_time; /* time before keep alive takes place */ unsigned int keepalive_intvl; /* time interval between keep alive probes */ diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 08a8fd1d1610..38943ed04e73 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -16,7 +16,9 @@ #define _REQUEST_SOCK_H #include +#include #include + #include struct request_sock; @@ -74,4 +76,180 @@ static inline void reqsk_free(struct request_sock *req) __reqsk_free(req); } +extern int sysctl_max_syn_backlog; + +/** struct tcp_listen_opt - listen state + * + * @max_qlen_log - log_2 of maximal queued SYNs/REQUESTs + */ +struct tcp_listen_opt { + u8 max_qlen_log; + /* 3 bytes hole, try to use */ + int qlen; + int qlen_young; + int clock_hand; + u32 hash_rnd; + struct request_sock *syn_table[0]; +}; + +/** struct request_sock_queue - queue of request_socks + * + * @rskq_accept_head - FIFO head of established children + * @rskq_accept_tail - FIFO tail of established children + * @syn_wait_lock - serializer + * + * %syn_wait_lock is necessary only to avoid proc interface having to grab the main + * lock sock while browsing the listening hash (otherwise it's deadlock prone). + * + * This lock is acquired in read mode only from listening_get_next() seq_file + * op and it's acquired in write mode _only_ from code that is actively + * changing rskq_accept_head. All readers that are holding the master sock lock + * don't need to grab this lock in read mode too as rskq_accept_head. writes + * are always protected from the main sock lock. + */ +struct request_sock_queue { + struct request_sock *rskq_accept_head; + struct request_sock *rskq_accept_tail; + rwlock_t syn_wait_lock; + struct tcp_listen_opt *listen_opt; +}; + +extern int reqsk_queue_alloc(struct request_sock_queue *queue, + const int nr_table_entries); + +static inline struct tcp_listen_opt *reqsk_queue_yank_listen_sk(struct request_sock_queue *queue) +{ + struct tcp_listen_opt *lopt; + + write_lock_bh(&queue->syn_wait_lock); + lopt = queue->listen_opt; + queue->listen_opt = NULL; + write_unlock_bh(&queue->syn_wait_lock); + + return lopt; +} + +static inline void reqsk_queue_destroy(struct request_sock_queue *queue) +{ + kfree(reqsk_queue_yank_listen_sk(queue)); +} + +static inline struct request_sock * + reqsk_queue_yank_acceptq(struct request_sock_queue *queue) +{ + struct request_sock *req = queue->rskq_accept_head; + + queue->rskq_accept_head = queue->rskq_accept_head = NULL; + return req; +} + +static inline int reqsk_queue_empty(struct request_sock_queue *queue) +{ + return queue->rskq_accept_head == NULL; +} + +static inline void reqsk_queue_unlink(struct request_sock_queue *queue, + struct request_sock *req, + struct request_sock **prev_req) +{ + write_lock(&queue->syn_wait_lock); + *prev_req = req->dl_next; + write_unlock(&queue->syn_wait_lock); +} + +static inline void reqsk_queue_add(struct request_sock_queue *queue, + struct request_sock *req, + struct sock *parent, + struct sock *child) +{ + req->sk = child; + sk_acceptq_added(parent); + + if (queue->rskq_accept_head == NULL) + queue->rskq_accept_head = req; + else + queue->rskq_accept_tail->dl_next = req; + + queue->rskq_accept_tail = req; + req->dl_next = NULL; +} + +static inline struct request_sock *reqsk_queue_remove(struct request_sock_queue *queue) +{ + struct request_sock *req = queue->rskq_accept_head; + + BUG_TRAP(req != NULL); + + queue->rskq_accept_head = req->dl_next; + if (queue->rskq_accept_head == NULL) + queue->rskq_accept_tail = NULL; + + return req; +} + +static inline struct sock *reqsk_queue_get_child(struct request_sock_queue *queue, + struct sock *parent) +{ + struct request_sock *req = reqsk_queue_remove(queue); + struct sock *child = req->sk; + + BUG_TRAP(child != NULL); + + sk_acceptq_removed(parent); + __reqsk_free(req); + return child; +} + +static inline int reqsk_queue_removed(struct request_sock_queue *queue, + struct request_sock *req) +{ + struct tcp_listen_opt *lopt = queue->listen_opt; + + if (req->retrans == 0) + --lopt->qlen_young; + + return --lopt->qlen; +} + +static inline int reqsk_queue_added(struct request_sock_queue *queue) +{ + struct tcp_listen_opt *lopt = queue->listen_opt; + const int prev_qlen = lopt->qlen; + + lopt->qlen_young++; + lopt->qlen++; + return prev_qlen; +} + +static inline int reqsk_queue_len(struct request_sock_queue *queue) +{ + return queue->listen_opt != NULL ? queue->listen_opt->qlen : 0; +} + +static inline int reqsk_queue_len_young(struct request_sock_queue *queue) +{ + return queue->listen_opt->qlen_young; +} + +static inline int reqsk_queue_is_full(struct request_sock_queue *queue) +{ + return queue->listen_opt->qlen >> queue->listen_opt->max_qlen_log; +} + +static inline void reqsk_queue_hash_req(struct request_sock_queue *queue, + u32 hash, struct request_sock *req, + unsigned timeout) +{ + struct tcp_listen_opt *lopt = queue->listen_opt; + + req->expires = jiffies + timeout; + req->retrans = 0; + req->sk = NULL; + req->dl_next = lopt->syn_table[hash]; + + write_lock(&queue->syn_wait_lock); + lopt->syn_table[hash] = req; + write_unlock(&queue->syn_wait_lock); +} + #endif /* _REQUEST_SOCK_H */ diff --git a/include/net/tcp.h b/include/net/tcp.h index 6663086a5e35..a2e323c54457 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1686,71 +1686,41 @@ static inline int tcp_full_space(const struct sock *sk) static inline void tcp_acceptq_queue(struct sock *sk, struct request_sock *req, struct sock *child) { - struct tcp_sock *tp = tcp_sk(sk); - - req->sk = child; - sk_acceptq_added(sk); - - if (!tp->accept_queue_tail) { - tp->accept_queue = req; - } else { - tp->accept_queue_tail->dl_next = req; - } - tp->accept_queue_tail = req; - req->dl_next = NULL; + reqsk_queue_add(&tcp_sk(sk)->accept_queue, req, sk, child); } -struct tcp_listen_opt -{ - u8 max_qlen_log; /* log_2 of maximal queued SYNs */ - int qlen; - int qlen_young; - int clock_hand; - u32 hash_rnd; - struct request_sock *syn_table[TCP_SYNQ_HSIZE]; -}; - static inline void tcp_synq_removed(struct sock *sk, struct request_sock *req) { - struct tcp_listen_opt *lopt = tcp_sk(sk)->listen_opt; - - if (--lopt->qlen == 0) + if (reqsk_queue_removed(&tcp_sk(sk)->accept_queue, req) == 0) tcp_delete_keepalive_timer(sk); - if (req->retrans == 0) - lopt->qlen_young--; } static inline void tcp_synq_added(struct sock *sk) { - struct tcp_listen_opt *lopt = tcp_sk(sk)->listen_opt; - - if (lopt->qlen++ == 0) + if (reqsk_queue_added(&tcp_sk(sk)->accept_queue) == 0) tcp_reset_keepalive_timer(sk, TCP_TIMEOUT_INIT); - lopt->qlen_young++; } static inline int tcp_synq_len(struct sock *sk) { - return tcp_sk(sk)->listen_opt->qlen; + return reqsk_queue_len(&tcp_sk(sk)->accept_queue); } static inline int tcp_synq_young(struct sock *sk) { - return tcp_sk(sk)->listen_opt->qlen_young; + return reqsk_queue_len_young(&tcp_sk(sk)->accept_queue); } static inline int tcp_synq_is_full(struct sock *sk) { - return tcp_synq_len(sk) >> tcp_sk(sk)->listen_opt->max_qlen_log; + return reqsk_queue_is_full(&tcp_sk(sk)->accept_queue); } static inline void tcp_synq_unlink(struct tcp_sock *tp, struct request_sock *req, - struct request_sock **prev) + struct request_sock **prev) { - write_lock(&tp->syn_wait_lock); - *prev = req->dl_next; - write_unlock(&tp->syn_wait_lock); + reqsk_queue_unlink(&tp->accept_queue, req, prev); } static inline void tcp_synq_drop(struct sock *sk, struct request_sock *req, diff --git a/net/core/Makefile b/net/core/Makefile index 81f03243fe2f..5e0c56b7f607 100644 --- a/net/core/Makefile +++ b/net/core/Makefile @@ -2,7 +2,8 @@ # Makefile for the Linux networking core. # -obj-y := sock.o skbuff.o iovec.o datagram.o stream.o scm.o gen_stats.o gen_estimator.o +obj-y := sock.o request_sock.o skbuff.o iovec.o datagram.o stream.o scm.o \ + gen_stats.o gen_estimator.o obj-$(CONFIG_SYSCTL) += sysctl_net_core.o diff --git a/net/core/request_sock.c b/net/core/request_sock.c new file mode 100644 index 000000000000..1258333ca007 --- /dev/null +++ b/net/core/request_sock.c @@ -0,0 +1,48 @@ +/* + * NET Generic infrastructure for Network protocols. + * + * Authors: Arnaldo Carvalho de Melo + * + * From code originally in include/net/tcp.h + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include + +#include + +int reqsk_queue_alloc(struct request_sock_queue *queue, + const int nr_table_entries) +{ + const int lopt_size = sizeof(struct tcp_listen_opt) + + nr_table_entries * sizeof(struct request_sock *); + struct tcp_listen_opt *lopt = kmalloc(lopt_size, GFP_KERNEL); + + if (lopt == NULL) + return -ENOMEM; + + memset(lopt, 0, lopt_size); + + for (lopt->max_qlen_log = 6; + (1 << lopt->max_qlen_log) < sysctl_max_syn_backlog; + lopt->max_qlen_log++); + + get_random_bytes(&lopt->hash_rnd, sizeof(lopt->hash_rnd)); + rwlock_init(&queue->syn_wait_lock); + queue->rskq_accept_head = queue->rskq_accept_head = NULL; + + write_lock_bh(&queue->syn_wait_lock); + queue->listen_opt = lopt; + write_unlock_bh(&queue->syn_wait_lock); + + return 0; +} + +EXPORT_SYMBOL(reqsk_queue_alloc); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 1c29feb6b35f..b85a46dd40a0 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -316,7 +316,7 @@ EXPORT_SYMBOL(tcp_enter_memory_pressure); static __inline__ unsigned int tcp_listen_poll(struct sock *sk, poll_table *wait) { - return tcp_sk(sk)->accept_queue ? (POLLIN | POLLRDNORM) : 0; + return !reqsk_queue_empty(&tcp_sk(sk)->accept_queue) ? (POLLIN | POLLRDNORM) : 0; } /* @@ -462,28 +462,15 @@ int tcp_listen_start(struct sock *sk) { struct inet_sock *inet = inet_sk(sk); struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt; + int rc = reqsk_queue_alloc(&tp->accept_queue, TCP_SYNQ_HSIZE); + + if (rc != 0) + return rc; sk->sk_max_ack_backlog = 0; sk->sk_ack_backlog = 0; - tp->accept_queue = tp->accept_queue_tail = NULL; - rwlock_init(&tp->syn_wait_lock); tcp_delack_init(tp); - lopt = kmalloc(sizeof(struct tcp_listen_opt), GFP_KERNEL); - if (!lopt) - return -ENOMEM; - - memset(lopt, 0, sizeof(struct tcp_listen_opt)); - for (lopt->max_qlen_log = 6; ; lopt->max_qlen_log++) - if ((1 << lopt->max_qlen_log) >= sysctl_max_syn_backlog) - break; - get_random_bytes(&lopt->hash_rnd, 4); - - write_lock_bh(&tp->syn_wait_lock); - tp->listen_opt = lopt; - write_unlock_bh(&tp->syn_wait_lock); - /* There is race window here: we announce ourselves listening, * but this transition is still not validated by get_port(). * It is OK, because this socket enters to hash table only @@ -500,10 +487,7 @@ int tcp_listen_start(struct sock *sk) } sk->sk_state = TCP_CLOSE; - write_lock_bh(&tp->syn_wait_lock); - tp->listen_opt = NULL; - write_unlock_bh(&tp->syn_wait_lock); - kfree(lopt); + reqsk_queue_destroy(&tp->accept_queue); return -EADDRINUSE; } @@ -515,18 +499,16 @@ int tcp_listen_start(struct sock *sk) static void tcp_listen_stop (struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->listen_opt; - struct request_sock *acc_req = tp->accept_queue; + struct tcp_listen_opt *lopt; + struct request_sock *acc_req; struct request_sock *req; int i; tcp_delete_keepalive_timer(sk); /* make all the listen_opt local to us */ - write_lock_bh(&tp->syn_wait_lock); - tp->listen_opt = NULL; - write_unlock_bh(&tp->syn_wait_lock); - tp->accept_queue = tp->accept_queue_tail = NULL; + lopt = reqsk_queue_yank_listen_sk(&tp->accept_queue); + acc_req = reqsk_queue_yank_acceptq(&tp->accept_queue); if (lopt->qlen) { for (i = 0; i < TCP_SYNQ_HSIZE; i++) { @@ -1867,11 +1849,11 @@ static int wait_for_connect(struct sock *sk, long timeo) prepare_to_wait_exclusive(sk->sk_sleep, &wait, TASK_INTERRUPTIBLE); release_sock(sk); - if (!tp->accept_queue) + if (reqsk_queue_empty(&tp->accept_queue)) timeo = schedule_timeout(timeo); lock_sock(sk); err = 0; - if (tp->accept_queue) + if (!reqsk_queue_empty(&tp->accept_queue)) break; err = -EINVAL; if (sk->sk_state != TCP_LISTEN) @@ -1894,7 +1876,6 @@ static int wait_for_connect(struct sock *sk, long timeo) struct sock *tcp_accept(struct sock *sk, int flags, int *err) { struct tcp_sock *tp = tcp_sk(sk); - struct request_sock *req; struct sock *newsk; int error; @@ -1905,37 +1886,31 @@ struct sock *tcp_accept(struct sock *sk, int flags, int *err) */ error = -EINVAL; if (sk->sk_state != TCP_LISTEN) - goto out; + goto out_err; /* Find already established connection */ - if (!tp->accept_queue) { + if (reqsk_queue_empty(&tp->accept_queue)) { long timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK); /* If this is a non blocking socket don't sleep */ error = -EAGAIN; if (!timeo) - goto out; + goto out_err; error = wait_for_connect(sk, timeo); if (error) - goto out; + goto out_err; } - req = tp->accept_queue; - if ((tp->accept_queue = req->dl_next) == NULL) - tp->accept_queue_tail = NULL; - - newsk = req->sk; - sk_acceptq_removed(sk); - __reqsk_free(req); + newsk = reqsk_queue_get_child(&tp->accept_queue, sk); BUG_TRAP(newsk->sk_state != TCP_SYN_RECV); - release_sock(sk); - return newsk; - out: release_sock(sk); + return newsk; +out_err: + newsk = NULL; *err = error; - return NULL; + goto out; } /* diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index 67277800d0c1..c3328fa48837 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -529,9 +529,9 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, entry.family = sk->sk_family; - read_lock_bh(&tp->syn_wait_lock); + read_lock_bh(&tp->accept_queue.syn_wait_lock); - lopt = tp->listen_opt; + lopt = tp->accept_queue.listen_opt; if (!lopt || !lopt->qlen) goto out; @@ -588,7 +588,7 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, } out: - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); return err; } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 95528a75a63d..1745dc8d25e6 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -874,7 +874,7 @@ static struct request_sock *tcp_v4_search_req(struct tcp_sock *tp, __u16 rport, __u32 raddr, __u32 laddr) { - struct tcp_listen_opt *lopt = tp->listen_opt; + struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; struct request_sock *req, **prev; for (prev = &lopt->syn_table[tcp_v4_synq_hash(raddr, rport, lopt->hash_rnd)]; @@ -898,18 +898,10 @@ static struct request_sock *tcp_v4_search_req(struct tcp_sock *tp, static void tcp_v4_synq_add(struct sock *sk, struct request_sock *req) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->listen_opt; + struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; u32 h = tcp_v4_synq_hash(inet_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port, lopt->hash_rnd); - req->expires = jiffies + TCP_TIMEOUT_INIT; - req->retrans = 0; - req->sk = NULL; - req->dl_next = lopt->syn_table[h]; - - write_lock(&tp->syn_wait_lock); - lopt->syn_table[h] = req; - write_unlock(&tp->syn_wait_lock); - + reqsk_queue_hash_req(&tp->accept_queue, h, req, TCP_TIMEOUT_INIT); tcp_synq_added(sk); } @@ -2167,17 +2159,17 @@ static void *listening_get_next(struct seq_file *seq, void *cur) if (++st->sbucket >= TCP_SYNQ_HSIZE) break; get_req: - req = tp->listen_opt->syn_table[st->sbucket]; + req = tp->accept_queue.listen_opt->syn_table[st->sbucket]; } sk = sk_next(st->syn_wait_sk); st->state = TCP_SEQ_STATE_LISTENING; - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); } else { tp = tcp_sk(sk); - read_lock_bh(&tp->syn_wait_lock); - if (tp->listen_opt && tp->listen_opt->qlen) + read_lock_bh(&tp->accept_queue.syn_wait_lock); + if (reqsk_queue_len(&tp->accept_queue)) goto start_req; - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); sk = sk_next(sk); } get_sk: @@ -2187,8 +2179,8 @@ get_sk: goto out; } tp = tcp_sk(sk); - read_lock_bh(&tp->syn_wait_lock); - if (tp->listen_opt && tp->listen_opt->qlen) { + read_lock_bh(&tp->accept_queue.syn_wait_lock); + if (reqsk_queue_len(&tp->accept_queue)) { start_req: st->uid = sock_i_uid(sk); st->syn_wait_sk = sk; @@ -2196,7 +2188,7 @@ start_req: st->sbucket = 0; goto get_req; } - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); } if (++st->bucket < TCP_LHTABLE_SIZE) { sk = sk_head(&tcp_listening_hash[st->bucket]); @@ -2383,7 +2375,7 @@ static void tcp_seq_stop(struct seq_file *seq, void *v) case TCP_SEQ_STATE_OPENREQ: if (v) { struct tcp_sock *tp = tcp_sk(st->syn_wait_sk); - read_unlock_bh(&tp->syn_wait_lock); + read_unlock_bh(&tp->accept_queue.syn_wait_lock); } case TCP_SEQ_STATE_LISTENING: if (v != SEQ_START_TOKEN) diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 0e6d525a8341..b3943e7562f3 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -790,10 +790,8 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, newtp->probes_out = 0; newtp->rx_opt.num_sacks = 0; newtp->urg_data = 0; - newtp->listen_opt = NULL; - newtp->accept_queue = newtp->accept_queue_tail = NULL; - /* Deinitialize syn_wait_lock to trap illegal accesses. */ - memset(&newtp->syn_wait_lock, 0, sizeof(newtp->syn_wait_lock)); + /* Deinitialize accept_queue to trap illegal accesses. */ + memset(&newtp->accept_queue, 0, sizeof(newtp->accept_queue)); /* Back to base struct sock members. */ newsk->sk_err = 0; diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index f03efe5fb76a..d97d191149c1 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -464,7 +464,7 @@ out_unlock: static void tcp_synack_timer(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->listen_opt; + struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; int max_retries = tp->syn_retries ? : sysctl_tcp_synack_retries; int thresh = max_retries; unsigned long now = jiffies; @@ -527,12 +527,8 @@ static void tcp_synack_timer(struct sock *sk) } /* Drop this request */ - write_lock(&tp->syn_wait_lock); - *reqp = req->dl_next; - write_unlock(&tp->syn_wait_lock); - lopt->qlen--; - if (req->retrans == 0) - lopt->qlen_young--; + tcp_synq_unlink(tp, req, reqp); + reqsk_queue_removed(&tp->accept_queue, req); reqsk_free(req); continue; } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 068cd4a8c292..84091daad6b5 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -401,7 +401,7 @@ static struct request_sock *tcp_v6_search_req(struct tcp_sock *tp, struct in6_addr *laddr, int iif) { - struct tcp_listen_opt *lopt = tp->listen_opt; + struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; struct request_sock *req, **prev; for (prev = &lopt->syn_table[tcp_v6_synq_hash(raddr, rport, lopt->hash_rnd)]; @@ -1267,18 +1267,10 @@ static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) static void tcp_v6_synq_add(struct sock *sk, struct request_sock *req) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->listen_opt; + struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; u32 h = tcp_v6_synq_hash(&tcp6_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port, lopt->hash_rnd); - req->sk = NULL; - req->expires = jiffies + TCP_TIMEOUT_INIT; - req->retrans = 0; - req->dl_next = lopt->syn_table[h]; - - write_lock(&tp->syn_wait_lock); - lopt->syn_table[h] = req; - write_unlock(&tp->syn_wait_lock); - + reqsk_queue_hash_req(&tp->accept_queue, h, req, TCP_TIMEOUT_INIT); tcp_synq_added(sk); } -- cgit v1.2.3 From 2ad69c55a282315e6119cf7fd744f26a925bdfd2 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sat, 18 Jun 2005 22:48:55 -0700 Subject: [NET] rename struct tcp_listen_opt to struct listen_sock Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/net/request_sock.h | 16 ++++++++-------- net/core/request_sock.c | 4 ++-- net/ipv4/tcp.c | 2 +- net/ipv4/tcp_diag.c | 2 +- net/ipv4/tcp_ipv4.c | 4 ++-- net/ipv4/tcp_timer.c | 2 +- net/ipv6/tcp_ipv6.c | 4 ++-- 7 files changed, 17 insertions(+), 17 deletions(-) (limited to 'include') diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 38943ed04e73..72fd6f5e86b1 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -78,11 +78,11 @@ static inline void reqsk_free(struct request_sock *req) extern int sysctl_max_syn_backlog; -/** struct tcp_listen_opt - listen state +/** struct listen_sock - listen state * * @max_qlen_log - log_2 of maximal queued SYNs/REQUESTs */ -struct tcp_listen_opt { +struct listen_sock { u8 max_qlen_log; /* 3 bytes hole, try to use */ int qlen; @@ -111,15 +111,15 @@ struct request_sock_queue { struct request_sock *rskq_accept_head; struct request_sock *rskq_accept_tail; rwlock_t syn_wait_lock; - struct tcp_listen_opt *listen_opt; + struct listen_sock *listen_opt; }; extern int reqsk_queue_alloc(struct request_sock_queue *queue, const int nr_table_entries); -static inline struct tcp_listen_opt *reqsk_queue_yank_listen_sk(struct request_sock_queue *queue) +static inline struct listen_sock *reqsk_queue_yank_listen_sk(struct request_sock_queue *queue) { - struct tcp_listen_opt *lopt; + struct listen_sock *lopt; write_lock_bh(&queue->syn_wait_lock); lopt = queue->listen_opt; @@ -203,7 +203,7 @@ static inline struct sock *reqsk_queue_get_child(struct request_sock_queue *queu static inline int reqsk_queue_removed(struct request_sock_queue *queue, struct request_sock *req) { - struct tcp_listen_opt *lopt = queue->listen_opt; + struct listen_sock *lopt = queue->listen_opt; if (req->retrans == 0) --lopt->qlen_young; @@ -213,7 +213,7 @@ static inline int reqsk_queue_removed(struct request_sock_queue *queue, static inline int reqsk_queue_added(struct request_sock_queue *queue) { - struct tcp_listen_opt *lopt = queue->listen_opt; + struct listen_sock *lopt = queue->listen_opt; const int prev_qlen = lopt->qlen; lopt->qlen_young++; @@ -240,7 +240,7 @@ static inline void reqsk_queue_hash_req(struct request_sock_queue *queue, u32 hash, struct request_sock *req, unsigned timeout) { - struct tcp_listen_opt *lopt = queue->listen_opt; + struct listen_sock *lopt = queue->listen_opt; req->expires = jiffies + timeout; req->retrans = 0; diff --git a/net/core/request_sock.c b/net/core/request_sock.c index 1258333ca007..78fd60a46bf2 100644 --- a/net/core/request_sock.c +++ b/net/core/request_sock.c @@ -21,9 +21,9 @@ int reqsk_queue_alloc(struct request_sock_queue *queue, const int nr_table_entries) { - const int lopt_size = sizeof(struct tcp_listen_opt) + + const int lopt_size = sizeof(struct listen_sock) + nr_table_entries * sizeof(struct request_sock *); - struct tcp_listen_opt *lopt = kmalloc(lopt_size, GFP_KERNEL); + struct listen_sock *lopt = kmalloc(lopt_size, GFP_KERNEL); if (lopt == NULL) return -ENOMEM; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index b85a46dd40a0..3a4c52e77e01 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -499,7 +499,7 @@ int tcp_listen_start(struct sock *sk) static void tcp_listen_stop (struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt; + struct listen_sock *lopt; struct request_sock *acc_req; struct request_sock *req; int i; diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c index c3328fa48837..634befc07921 100644 --- a/net/ipv4/tcp_diag.c +++ b/net/ipv4/tcp_diag.c @@ -514,7 +514,7 @@ static int tcpdiag_dump_reqs(struct sk_buff *skb, struct sock *sk, struct tcpdiag_entry entry; struct tcpdiagreq *r = NLMSG_DATA(cb->nlh); struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt; + struct listen_sock *lopt; struct rtattr *bc = NULL; struct inet_sock *inet = inet_sk(sk); int j, s_j; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 1745dc8d25e6..485ca9cb1707 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -874,7 +874,7 @@ static struct request_sock *tcp_v4_search_req(struct tcp_sock *tp, __u16 rport, __u32 raddr, __u32 laddr) { - struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; + struct listen_sock *lopt = tp->accept_queue.listen_opt; struct request_sock *req, **prev; for (prev = &lopt->syn_table[tcp_v4_synq_hash(raddr, rport, lopt->hash_rnd)]; @@ -898,7 +898,7 @@ static struct request_sock *tcp_v4_search_req(struct tcp_sock *tp, static void tcp_v4_synq_add(struct sock *sk, struct request_sock *req) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; + struct listen_sock *lopt = tp->accept_queue.listen_opt; u32 h = tcp_v4_synq_hash(inet_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port, lopt->hash_rnd); reqsk_queue_hash_req(&tp->accept_queue, h, req, TCP_TIMEOUT_INIT); diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index d97d191149c1..b127b4498565 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -464,7 +464,7 @@ out_unlock: static void tcp_synack_timer(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; + struct listen_sock *lopt = tp->accept_queue.listen_opt; int max_retries = tp->syn_retries ? : sysctl_tcp_synack_retries; int thresh = max_retries; unsigned long now = jiffies; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 84091daad6b5..2414937f2a83 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -401,7 +401,7 @@ static struct request_sock *tcp_v6_search_req(struct tcp_sock *tp, struct in6_addr *laddr, int iif) { - struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; + struct listen_sock *lopt = tp->accept_queue.listen_opt; struct request_sock *req, **prev; for (prev = &lopt->syn_table[tcp_v6_synq_hash(raddr, rport, lopt->hash_rnd)]; @@ -1267,7 +1267,7 @@ static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) static void tcp_v6_synq_add(struct sock *sk, struct request_sock *req) { struct tcp_sock *tp = tcp_sk(sk); - struct tcp_listen_opt *lopt = tp->accept_queue.listen_opt; + struct listen_sock *lopt = tp->accept_queue.listen_opt; u32 h = tcp_v6_synq_hash(&tcp6_rsk(req)->rmt_addr, inet_rsk(req)->rmt_port, lopt->hash_rnd); reqsk_queue_hash_req(&tp->accept_queue, h, req, TCP_TIMEOUT_INIT); -- cgit v1.2.3 From e52c1f17e4ea8e61bd26eb25f1a184202693c2b9 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sat, 18 Jun 2005 22:49:40 -0700 Subject: [NET]: Move sysctl_max_syn_backlog into request_sock.c This fixes the CONFIG_INET=n build failure noticed by Andrew Morton. Signed-off-by: David S. Miller --- include/net/tcp.h | 1 - net/core/request_sock.c | 16 ++++++++++++++++ net/ipv4/tcp_ipv4.c | 16 ---------------- 3 files changed, 16 insertions(+), 17 deletions(-) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index a2e323c54457..f730935b824a 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -564,7 +564,6 @@ static __inline__ int tcp_sk_listen_hashfn(struct sock *sk) #define TCP_NAGLE_PUSH 4 /* Cork is overriden for already queued data */ /* sysctl variables for tcp */ -extern int sysctl_max_syn_backlog; extern int sysctl_tcp_timestamps; extern int sysctl_tcp_window_scaling; extern int sysctl_tcp_sack; diff --git a/net/core/request_sock.c b/net/core/request_sock.c index 78fd60a46bf2..bb55675f0685 100644 --- a/net/core/request_sock.c +++ b/net/core/request_sock.c @@ -18,6 +18,22 @@ #include +/* + * Maximum number of SYN_RECV sockets in queue per LISTEN socket. + * One SYN_RECV socket costs about 80bytes on a 32bit machine. + * It would be better to replace it with a global counter for all sockets + * but then some measure against one socket starving all other sockets + * would be needed. + * + * It was 128 by default. Experiments with real servers show, that + * it is absolutely not enough even at 100conn/sec. 256 cures most + * of problems. This value is adjusted to 128 for very small machines + * (<=32Mb of memory) and to 1024 on normal or better ones (>=256Mb). + * Further increasing requires to change hash table size. + */ +int sysctl_max_syn_backlog = 256; +EXPORT_SYMBOL(sysctl_max_syn_backlog); + int reqsk_queue_alloc(struct request_sock_queue *queue, const int nr_table_entries) { diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 485ca9cb1707..2d41d5d6ad19 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1366,21 +1366,6 @@ static inline struct ip_options *tcp_v4_save_options(struct sock *sk, return dopt; } -/* - * Maximum number of SYN_RECV sockets in queue per LISTEN socket. - * One SYN_RECV socket costs about 80bytes on a 32bit machine. - * It would be better to replace it with a global counter for all sockets - * but then some measure against one socket starving all other sockets - * would be needed. - * - * It was 128 by default. Experiments with real servers show, that - * it is absolutely not enough even at 100conn/sec. 256 cures most - * of problems. This value is adjusted to 128 for very small machines - * (<=32Mb of memory) and to 1024 on normal or better ones (>=256Mb). - * Further increasing requires to change hash table size. - */ -int sysctl_max_syn_backlog = 256; - struct request_sock_ops tcp_request_sock_ops = { .family = PF_INET, .obj_size = sizeof(struct tcp_request_sock), @@ -2662,7 +2647,6 @@ EXPORT_SYMBOL(tcp_proc_register); EXPORT_SYMBOL(tcp_proc_unregister); #endif EXPORT_SYMBOL(sysctl_local_port_range); -EXPORT_SYMBOL(sysctl_max_syn_backlog); EXPORT_SYMBOL(sysctl_tcp_low_latency); EXPORT_SYMBOL(sysctl_tcp_tw_reuse); -- cgit v1.2.3 From f88a10d65620d97b6d0a7e352a3493c1b7e7409b Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:50:12 -0700 Subject: [NETLINK]: New message building macros NLMSG_PUT_ANSWER(skb, nlcb, type, length) Start a new netlink message as answer to a request, returns the message header. NLMSG_END(skb, nlh) End a netlink message, fixes total message length, returns skb->len. NLMSG_CANCEL(skb, nlh) Cancel the building process and trim whole message from skb again, returns -1. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/netlink.h | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index b2738ac8bc99..8d1cb419a930 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -171,8 +171,21 @@ __nlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, int type, int len) } #define NLMSG_PUT(skb, pid, seq, type, len) \ -({ if (skb_tailroom(skb) < (int)NLMSG_SPACE(len)) goto nlmsg_failure; \ - __nlmsg_put(skb, pid, seq, type, len); }) +({ if (skb_tailroom(skb) < (int)NLMSG_SPACE(len)) \ + goto nlmsg_failure; \ + __nlmsg_put(skb, pid, seq, type, len); }) + +#define NLMSG_PUT_ANSWER(skb, cb, type, len) \ + NLMSG_PUT(skb, NETLINK_CB((cb)->skb).pid, \ + (cb)->nlh->nlmsg_seq, type, len) + +#define NLMSG_END(skb, nlh) \ +({ (nlh)->nlmsg_len = (skb)->tail - (unsigned char *) (nlh); \ + (skb)->len; }) + +#define NLMSG_CANCEL(skb, nlh) \ +({ skb_trim(skb, (unsigned char *) (nlh) - (skb)->data); \ + -1; }) extern int netlink_dump_start(struct sock *ssk, struct sk_buff *skb, struct nlmsghdr *nlh, -- cgit v1.2.3 From 00768244923f66801958a8d2d103f7b65608c9b6 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:50:38 -0700 Subject: [NETLINK] Routing attribute related shortcuts RTA_GET_U(32|64)(tlv) Assumes TLV is a u32/u64 field and returns its value. RTA_GET_[M]SECS(tlv) Assumes TLV is a u64 and transports jiffies converted to seconds or milliseconds and returns its value. RTA_PUT_U(32|64)(skb, type, value) Appends %value as fixed u32/u64 to %skb as TLV %type. RTA_PUT_[M]SECS(skb, type, jiffies) Converts %jiffies to secs/msecs and appends it as u64 to %skb as TLV %type. RTA_PUT_STRING(skb, type, string) Appends %NUL terminated %string to %skb as TLV %type. RTA_NEST(skb, type) Starts a nested TLV %type and returns the nesting handle. RTA_NEST_END(skb, nesting_handle) Finishes the nested TLV %nesting_handle, must be called symmetric to RTA_NEST(). Returns skb->len RTA_NEST_CANCEL(skb, nesting_handle) Cancel the nested TLV %nesting_handle and trim nested TLV from skb again, returns -1. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 45 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 45 insertions(+) (limited to 'include') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 91ac97c20777..a09b5d42babf 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -789,6 +789,51 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi ({ if (unlikely(skb_tailroom(skb) < (int)(attrlen))) \ goto rtattr_failure; \ memcpy(skb_put(skb, RTA_ALIGN(attrlen)), data, attrlen); }) + +#define RTA_PUT_U32(skb, attrtype, value) \ +({ u32 _tmp = (value); \ + RTA_PUT(skb, attrtype, sizeof(u32), &_tmp); }) + +#define RTA_PUT_U64(skb, attrtype, value) \ +({ u64 _tmp = (value); \ + RTA_PUT(skb, attrtype, sizeof(u64), &_tmp); }) + +#define RTA_PUT_SECS(skb, attrtype, value) \ + RTA_PUT_U64(skb, attrtype, (value) / HZ) + +#define RTA_PUT_MSECS(skb, attrtype, value) \ + RTA_PUT_U64(skb, attrtype, jiffies_to_msecs(value)) + +#define RTA_PUT_STRING(skb, attrtype, value) \ + RTA_PUT(skb, attrtype, strlen(value) + 1, value) + +#define RTA_NEST(skb, type) \ +({ struct rtattr *__start = (struct rtattr *) (skb)->tail; \ + RTA_PUT(skb, type, 0, NULL); \ + __start; }) + +#define RTA_NEST_END(skb, start) \ +({ (start)->rta_len = ((skb)->tail - (unsigned char *) (start)); \ + (skb)->len; }) + +#define RTA_NEST_CANCEL(skb, start) \ +({ skb_trim(skb, (unsigned char *) (start) - (skb)->data); \ + -1; }) + +#define RTA_GET_U32(rta) \ +({ if (!rta || RTA_PAYLOAD(rta) < sizeof(u32)) \ + goto rtattr_failure; \ + *(u32 *) RTA_DATA(rta); }) + +#define RTA_GET_U64(rta) \ +({ u64 _tmp; \ + if (!rta || RTA_PAYLOAD(rta) < sizeof(u64)) \ + goto rtattr_failure; \ + memcpy(&_tmp, RTA_DATA(rta), sizeof(_tmp)); \ + _tmp; }) + +#define RTA_GET_SECS(rta) ((unsigned long) RTA_GET_U64(rta) * HZ) +#define RTA_GET_MSECS(rta) (msecs_to_jiffies((unsigned long) RTA_GET_U64(rta))) static inline struct rtattr * __rta_reserve(struct sk_buff *skb, int attrtype, int attrlen) -- cgit v1.2.3 From c7fb64db001f83ece669c76a02d8ec2fdb1dd307 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:50:55 -0700 Subject: [NETLINK]: Neighbour table configuration and statistics via rtnetlink To retrieve the neighbour tables send RTM_GETNEIGHTBL with the NLM_F_DUMP flag set. Every neighbour table configuration is spread over multiple messages to avoid running into message size limits on systems with many interfaces. The first message in the sequence transports all not device specific data such as statistics, configuration, and the default parameter set. This message is followed by 0..n messages carrying device specific parameter sets. Although the ordering should be sufficient, NDTA_NAME can be used to identify sequences. The initial message can be identified by checking for NDTA_CONFIG. The device specific messages do not contain this TLV but have NDTPA_IFINDEX set to the corresponding interface index. To change neighbour table attributes, send RTM_SETNEIGHTBL with NDTA_NAME set. Changeable attribute include NDTA_THRESH[1-3], NDTA_GC_INTERVAL, and all TLVs in NDTA_PARMS unless marked otherwise. Device specific parameter sets can be changed by setting NDTPA_IFINDEX to the interface index of the corresponding device. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 107 +++++++++++++++ include/net/neighbour.h | 4 + net/core/neighbour.c | 317 +++++++++++++++++++++++++++++++++++++++++++- net/core/rtnetlink.c | 20 +-- security/selinux/nlmsgtab.c | 2 + 5 files changed, 439 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index a09b5d42babf..5a5cda160267 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -89,6 +89,13 @@ enum { RTM_GETANYCAST = 62, #define RTM_GETANYCAST RTM_GETANYCAST + RTM_NEWNEIGHTBL = 64, +#define RTM_NEWNEIGHTBL RTM_NEWNEIGHTBL + RTM_GETNEIGHTBL = 66, +#define RTM_GETNEIGHTBL RTM_GETNEIGHTBL + RTM_SETNEIGHTBL, +#define RTM_SETNEIGHTBL RTM_SETNEIGHTBL + __RTM_MAX, #define RTM_MAX (((__RTM_MAX + 3) & ~3) - 1) }; @@ -493,6 +500,106 @@ struct nda_cacheinfo __u32 ndm_refcnt; }; + +/***************************************************************** + * Neighbour tables specific messages. + * + * To retrieve the neighbour tables send RTM_GETNEIGHTBL with the + * NLM_F_DUMP flag set. Every neighbour table configuration is + * spread over multiple messages to avoid running into message + * size limits on systems with many interfaces. The first message + * in the sequence transports all not device specific data such as + * statistics, configuration, and the default parameter set. + * This message is followed by 0..n messages carrying device + * specific parameter sets. + * Although the ordering should be sufficient, NDTA_NAME can be + * used to identify sequences. The initial message can be identified + * by checking for NDTA_CONFIG. The device specific messages do + * not contain this TLV but have NDTPA_IFINDEX set to the + * corresponding interface index. + * + * To change neighbour table attributes, send RTM_SETNEIGHTBL + * with NDTA_NAME set. Changeable attribute include NDTA_THRESH[1-3], + * NDTA_GC_INTERVAL, and all TLVs in NDTA_PARMS unless marked + * otherwise. Device specific parameter sets can be changed by + * setting NDTPA_IFINDEX to the interface index of the corresponding + * device. + ****/ + +struct ndt_stats +{ + __u64 ndts_allocs; + __u64 ndts_destroys; + __u64 ndts_hash_grows; + __u64 ndts_res_failed; + __u64 ndts_lookups; + __u64 ndts_hits; + __u64 ndts_rcv_probes_mcast; + __u64 ndts_rcv_probes_ucast; + __u64 ndts_periodic_gc_runs; + __u64 ndts_forced_gc_runs; +}; + +enum { + NDTPA_UNSPEC, + NDTPA_IFINDEX, /* u32, unchangeable */ + NDTPA_REFCNT, /* u32, read-only */ + NDTPA_REACHABLE_TIME, /* u64, read-only, msecs */ + NDTPA_BASE_REACHABLE_TIME, /* u64, msecs */ + NDTPA_RETRANS_TIME, /* u64, msecs */ + NDTPA_GC_STALETIME, /* u64, msecs */ + NDTPA_DELAY_PROBE_TIME, /* u64, msecs */ + NDTPA_QUEUE_LEN, /* u32 */ + NDTPA_APP_PROBES, /* u32 */ + NDTPA_UCAST_PROBES, /* u32 */ + NDTPA_MCAST_PROBES, /* u32 */ + NDTPA_ANYCAST_DELAY, /* u64, msecs */ + NDTPA_PROXY_DELAY, /* u64, msecs */ + NDTPA_PROXY_QLEN, /* u32 */ + NDTPA_LOCKTIME, /* u64, msecs */ + __NDTPA_MAX +}; +#define NDTPA_MAX (__NDTPA_MAX - 1) + +struct ndtmsg +{ + __u8 ndtm_family; + __u8 ndtm_pad1; + __u16 ndtm_pad2; +}; + +struct ndt_config +{ + __u16 ndtc_key_len; + __u16 ndtc_entry_size; + __u32 ndtc_entries; + __u32 ndtc_last_flush; /* delta to now in msecs */ + __u32 ndtc_last_rand; /* delta to now in msecs */ + __u32 ndtc_hash_rnd; + __u32 ndtc_hash_mask; + __u32 ndtc_hash_chain_gc; + __u32 ndtc_proxy_qlen; +}; + +enum { + NDTA_UNSPEC, + NDTA_NAME, /* char *, unchangeable */ + NDTA_THRESH1, /* u32 */ + NDTA_THRESH2, /* u32 */ + NDTA_THRESH3, /* u32 */ + NDTA_CONFIG, /* struct ndt_config, read-only */ + NDTA_PARMS, /* nested TLV NDTPA_* */ + NDTA_STATS, /* struct ndt_stats, read-only */ + NDTA_GC_INTERVAL, /* u64, msecs */ + __NDTA_MAX +}; +#define NDTA_MAX (__NDTA_MAX - 1) + +#define NDTA_RTA(r) ((struct rtattr*)(((char*)(r)) + \ + NLMSG_ALIGN(sizeof(struct ndtmsg)))) +#define NDTA_PAYLOAD(n) NLMSG_PAYLOAD(n,sizeof(struct ndtmsg)) + + /**** * General form of address family dependent message. ****/ diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 4f33bbc21e7f..17191ac9be70 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -65,6 +65,7 @@ struct neighbour; struct neigh_parms { + struct net_device *dev; struct neigh_parms *next; int (*neigh_setup)(struct neighbour *); struct neigh_table *tbl; @@ -252,6 +253,9 @@ extern int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg); extern int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg); extern void neigh_app_ns(struct neighbour *n); +extern int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb); +extern int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg); + extern void neigh_for_each(struct neigh_table *tbl, void (*cb)(struct neighbour *, void *), void *cookie); extern void __neigh_for_each_release(struct neigh_table *tbl, int (*cb)(struct neighbour *)); extern void pneigh_for_each(struct neigh_table *tbl, void (*cb)(struct pneigh_entry *)); diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 43bdc521e20d..0841ac78c67d 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1276,9 +1276,14 @@ struct neigh_parms *neigh_parms_alloc(struct net_device *dev, INIT_RCU_HEAD(&p->rcu_head); p->reachable_time = neigh_rand_reach_time(p->base_reachable_time); - if (dev && dev->neigh_setup && dev->neigh_setup(dev, p)) { - kfree(p); - return NULL; + if (dev) { + if (dev->neigh_setup && dev->neigh_setup(dev, p)) { + kfree(p); + return NULL; + } + + dev_hold(dev); + p->dev = dev; } p->sysctl_table = NULL; write_lock_bh(&tbl->lock); @@ -1309,6 +1314,8 @@ void neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms) *p = parms->next; parms->dead = 1; write_unlock_bh(&tbl->lock); + if (parms->dev) + dev_put(parms->dev); call_rcu(&parms->rcu_head, neigh_rcu_free_parms); return; } @@ -1546,6 +1553,308 @@ out: return err; } +static int neightbl_fill_parms(struct sk_buff *skb, struct neigh_parms *parms) +{ + struct rtattr *nest = RTA_NEST(skb, NDTA_PARMS); + + if (parms->dev) + RTA_PUT_U32(skb, NDTPA_IFINDEX, parms->dev->ifindex); + + RTA_PUT_U32(skb, NDTPA_REFCNT, atomic_read(&parms->refcnt)); + RTA_PUT_U32(skb, NDTPA_QUEUE_LEN, parms->queue_len); + RTA_PUT_U32(skb, NDTPA_PROXY_QLEN, parms->proxy_qlen); + RTA_PUT_U32(skb, NDTPA_APP_PROBES, parms->app_probes); + RTA_PUT_U32(skb, NDTPA_UCAST_PROBES, parms->ucast_probes); + RTA_PUT_U32(skb, NDTPA_MCAST_PROBES, parms->mcast_probes); + RTA_PUT_MSECS(skb, NDTPA_REACHABLE_TIME, parms->reachable_time); + RTA_PUT_MSECS(skb, NDTPA_BASE_REACHABLE_TIME, + parms->base_reachable_time); + RTA_PUT_MSECS(skb, NDTPA_GC_STALETIME, parms->gc_staletime); + RTA_PUT_MSECS(skb, NDTPA_DELAY_PROBE_TIME, parms->delay_probe_time); + RTA_PUT_MSECS(skb, NDTPA_RETRANS_TIME, parms->retrans_time); + RTA_PUT_MSECS(skb, NDTPA_ANYCAST_DELAY, parms->anycast_delay); + RTA_PUT_MSECS(skb, NDTPA_PROXY_DELAY, parms->proxy_delay); + RTA_PUT_MSECS(skb, NDTPA_LOCKTIME, parms->locktime); + + return RTA_NEST_END(skb, nest); + +rtattr_failure: + return RTA_NEST_CANCEL(skb, nest); +} + +static int neightbl_fill_info(struct neigh_table *tbl, struct sk_buff *skb, + struct netlink_callback *cb) +{ + struct nlmsghdr *nlh; + struct ndtmsg *ndtmsg; + + nlh = NLMSG_PUT_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg)); + ndtmsg = NLMSG_DATA(nlh); + + NLMSG_SET_MULTIPART(nlh); + + read_lock_bh(&tbl->lock); + ndtmsg->ndtm_family = tbl->family; + + RTA_PUT_STRING(skb, NDTA_NAME, tbl->id); + RTA_PUT_MSECS(skb, NDTA_GC_INTERVAL, tbl->gc_interval); + RTA_PUT_U32(skb, NDTA_THRESH1, tbl->gc_thresh1); + RTA_PUT_U32(skb, NDTA_THRESH2, tbl->gc_thresh2); + RTA_PUT_U32(skb, NDTA_THRESH3, tbl->gc_thresh3); + + { + unsigned long now = jiffies; + unsigned int flush_delta = now - tbl->last_flush; + unsigned int rand_delta = now - tbl->last_rand; + + struct ndt_config ndc = { + .ndtc_key_len = tbl->key_len, + .ndtc_entry_size = tbl->entry_size, + .ndtc_entries = atomic_read(&tbl->entries), + .ndtc_last_flush = jiffies_to_msecs(flush_delta), + .ndtc_last_rand = jiffies_to_msecs(rand_delta), + .ndtc_hash_rnd = tbl->hash_rnd, + .ndtc_hash_mask = tbl->hash_mask, + .ndtc_hash_chain_gc = tbl->hash_chain_gc, + .ndtc_proxy_qlen = tbl->proxy_queue.qlen, + }; + + RTA_PUT(skb, NDTA_CONFIG, sizeof(ndc), &ndc); + } + + { + int cpu; + struct ndt_stats ndst; + + memset(&ndst, 0, sizeof(ndst)); + + for (cpu = 0; cpu < NR_CPUS; cpu++) { + struct neigh_statistics *st; + + if (!cpu_possible(cpu)) + continue; + + st = per_cpu_ptr(tbl->stats, cpu); + ndst.ndts_allocs += st->allocs; + ndst.ndts_destroys += st->destroys; + ndst.ndts_hash_grows += st->hash_grows; + ndst.ndts_res_failed += st->res_failed; + ndst.ndts_lookups += st->lookups; + ndst.ndts_hits += st->hits; + ndst.ndts_rcv_probes_mcast += st->rcv_probes_mcast; + ndst.ndts_rcv_probes_ucast += st->rcv_probes_ucast; + ndst.ndts_periodic_gc_runs += st->periodic_gc_runs; + ndst.ndts_forced_gc_runs += st->forced_gc_runs; + } + + RTA_PUT(skb, NDTA_STATS, sizeof(ndst), &ndst); + } + + BUG_ON(tbl->parms.dev); + if (neightbl_fill_parms(skb, &tbl->parms) < 0) + goto rtattr_failure; + + read_unlock_bh(&tbl->lock); + return NLMSG_END(skb, nlh); + +rtattr_failure: + read_unlock_bh(&tbl->lock); + return NLMSG_CANCEL(skb, nlh); + +nlmsg_failure: + return -1; +} + +static int neightbl_fill_param_info(struct neigh_table *tbl, + struct neigh_parms *parms, + struct sk_buff *skb, + struct netlink_callback *cb) +{ + struct ndtmsg *ndtmsg; + struct nlmsghdr *nlh; + + nlh = NLMSG_PUT_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg)); + ndtmsg = NLMSG_DATA(nlh); + + NLMSG_SET_MULTIPART(nlh); + + read_lock_bh(&tbl->lock); + ndtmsg->ndtm_family = tbl->family; + RTA_PUT_STRING(skb, NDTA_NAME, tbl->id); + + if (neightbl_fill_parms(skb, parms) < 0) + goto rtattr_failure; + + read_unlock_bh(&tbl->lock); + return NLMSG_END(skb, nlh); + +rtattr_failure: + read_unlock_bh(&tbl->lock); + return NLMSG_CANCEL(skb, nlh); + +nlmsg_failure: + return -1; +} + +static inline struct neigh_parms *lookup_neigh_params(struct neigh_table *tbl, + int ifindex) +{ + struct neigh_parms *p; + + for (p = &tbl->parms; p; p = p->next) + if ((p->dev && p->dev->ifindex == ifindex) || + (!p->dev && !ifindex)) + return p; + + return NULL; +} + +int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) +{ + struct neigh_table *tbl; + struct ndtmsg *ndtmsg = NLMSG_DATA(nlh); + struct rtattr **tb = arg; + int err = -EINVAL; + + if (!tb[NDTA_NAME - 1] || !RTA_PAYLOAD(tb[NDTA_NAME - 1])) + return -EINVAL; + + read_lock(&neigh_tbl_lock); + for (tbl = neigh_tables; tbl; tbl = tbl->next) { + if (ndtmsg->ndtm_family && tbl->family != ndtmsg->ndtm_family) + continue; + + if (!rtattr_strcmp(tb[NDTA_NAME - 1], tbl->id)) + break; + } + + if (tbl == NULL) { + err = -ENOENT; + goto errout; + } + + /* + * We acquire tbl->lock to be nice to the periodic timers and + * make sure they always see a consistent set of values. + */ + write_lock_bh(&tbl->lock); + + if (tb[NDTA_THRESH1 - 1]) + tbl->gc_thresh1 = RTA_GET_U32(tb[NDTA_THRESH1 - 1]); + + if (tb[NDTA_THRESH2 - 1]) + tbl->gc_thresh2 = RTA_GET_U32(tb[NDTA_THRESH2 - 1]); + + if (tb[NDTA_THRESH3 - 1]) + tbl->gc_thresh3 = RTA_GET_U32(tb[NDTA_THRESH3 - 1]); + + if (tb[NDTA_GC_INTERVAL - 1]) + tbl->gc_interval = RTA_GET_MSECS(tb[NDTA_GC_INTERVAL - 1]); + + if (tb[NDTA_PARMS - 1]) { + struct rtattr *tbp[NDTPA_MAX]; + struct neigh_parms *p; + u32 ifindex = 0; + + if (rtattr_parse_nested(tbp, NDTPA_MAX, tb[NDTA_PARMS - 1]) < 0) + goto rtattr_failure; + + if (tbp[NDTPA_IFINDEX - 1]) + ifindex = RTA_GET_U32(tbp[NDTPA_IFINDEX - 1]); + + p = lookup_neigh_params(tbl, ifindex); + if (p == NULL) { + err = -ENOENT; + goto rtattr_failure; + } + + if (tbp[NDTPA_QUEUE_LEN - 1]) + p->queue_len = RTA_GET_U32(tbp[NDTPA_QUEUE_LEN - 1]); + + if (tbp[NDTPA_PROXY_QLEN - 1]) + p->proxy_qlen = RTA_GET_U32(tbp[NDTPA_PROXY_QLEN - 1]); + + if (tbp[NDTPA_APP_PROBES - 1]) + p->app_probes = RTA_GET_U32(tbp[NDTPA_APP_PROBES - 1]); + + if (tbp[NDTPA_UCAST_PROBES - 1]) + p->ucast_probes = + RTA_GET_U32(tbp[NDTPA_UCAST_PROBES - 1]); + + if (tbp[NDTPA_MCAST_PROBES - 1]) + p->mcast_probes = + RTA_GET_U32(tbp[NDTPA_MCAST_PROBES - 1]); + + if (tbp[NDTPA_BASE_REACHABLE_TIME - 1]) + p->base_reachable_time = + RTA_GET_MSECS(tbp[NDTPA_BASE_REACHABLE_TIME - 1]); + + if (tbp[NDTPA_GC_STALETIME - 1]) + p->gc_staletime = + RTA_GET_MSECS(tbp[NDTPA_GC_STALETIME - 1]); + + if (tbp[NDTPA_DELAY_PROBE_TIME - 1]) + p->delay_probe_time = + RTA_GET_MSECS(tbp[NDTPA_DELAY_PROBE_TIME - 1]); + + if (tbp[NDTPA_RETRANS_TIME - 1]) + p->retrans_time = + RTA_GET_MSECS(tbp[NDTPA_RETRANS_TIME - 1]); + + if (tbp[NDTPA_ANYCAST_DELAY - 1]) + p->anycast_delay = + RTA_GET_MSECS(tbp[NDTPA_ANYCAST_DELAY - 1]); + + if (tbp[NDTPA_PROXY_DELAY - 1]) + p->proxy_delay = + RTA_GET_MSECS(tbp[NDTPA_PROXY_DELAY - 1]); + + if (tbp[NDTPA_LOCKTIME - 1]) + p->locktime = RTA_GET_MSECS(tbp[NDTPA_LOCKTIME - 1]); + } + + err = 0; + +rtattr_failure: + write_unlock_bh(&tbl->lock); +errout: + read_unlock(&neigh_tbl_lock); + return err; +} + +int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb) +{ + int idx, family; + int s_idx = cb->args[0]; + struct neigh_table *tbl; + + family = ((struct rtgenmsg *)NLMSG_DATA(cb->nlh))->rtgen_family; + + read_lock(&neigh_tbl_lock); + for (tbl = neigh_tables, idx = 0; tbl; tbl = tbl->next) { + struct neigh_parms *p; + + if (idx < s_idx || (family && tbl->family != family)) + continue; + + if (neightbl_fill_info(tbl, skb, cb) <= 0) + break; + + for (++idx, p = tbl->parms.next; p; p = p->next, idx++) { + if (idx < s_idx) + continue; + + if (neightbl_fill_param_info(tbl, p, skb, cb) <= 0) + goto out; + } + + } +out: + read_unlock(&neigh_tbl_lock); + cb->args[0] = idx; + + return skb->len; +} static int neigh_fill_info(struct sk_buff *skb, struct neighbour *n, u32 pid, u32 seq, int event) @@ -2352,6 +2661,8 @@ EXPORT_SYMBOL(neigh_update); EXPORT_SYMBOL(neigh_update_hhs); EXPORT_SYMBOL(pneigh_enqueue); EXPORT_SYMBOL(pneigh_lookup); +EXPORT_SYMBOL(neightbl_dump_info); +EXPORT_SYMBOL(neightbl_set); #ifdef CONFIG_ARPD EXPORT_SYMBOL(neigh_app_ns); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 00caf4b318b2..56a20f014b8a 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -100,6 +100,7 @@ static const int rtm_min[RTM_NR_FAMILIES] = [RTM_FAM(RTM_NEWPREFIX)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), [RTM_FAM(RTM_GETMULTICAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), [RTM_FAM(RTM_GETANYCAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)), + [RTM_FAM(RTM_NEWNEIGHTBL)] = NLMSG_LENGTH(sizeof(struct ndtmsg)), }; static const int rta_max[RTM_NR_FAMILIES] = @@ -113,6 +114,7 @@ static const int rta_max[RTM_NR_FAMILIES] = [RTM_FAM(RTM_NEWTCLASS)] = TCA_MAX, [RTM_FAM(RTM_NEWTFILTER)] = TCA_MAX, [RTM_FAM(RTM_NEWACTION)] = TCAA_MAX, + [RTM_FAM(RTM_NEWNEIGHTBL)] = NDTA_MAX, }; void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const void *data) @@ -649,14 +651,16 @@ static void rtnetlink_rcv(struct sock *sk, int len) static struct rtnetlink_link link_rtnetlink_table[RTM_NR_MSGTYPES] = { - [RTM_GETLINK - RTM_BASE] = { .dumpit = rtnetlink_dump_ifinfo }, - [RTM_SETLINK - RTM_BASE] = { .doit = do_setlink }, - [RTM_GETADDR - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, - [RTM_GETROUTE - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, - [RTM_NEWNEIGH - RTM_BASE] = { .doit = neigh_add }, - [RTM_DELNEIGH - RTM_BASE] = { .doit = neigh_delete }, - [RTM_GETNEIGH - RTM_BASE] = { .dumpit = neigh_dump_info }, - [RTM_GETRULE - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, + [RTM_GETLINK - RTM_BASE] = { .dumpit = rtnetlink_dump_ifinfo }, + [RTM_SETLINK - RTM_BASE] = { .doit = do_setlink }, + [RTM_GETADDR - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, + [RTM_GETROUTE - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, + [RTM_NEWNEIGH - RTM_BASE] = { .doit = neigh_add }, + [RTM_DELNEIGH - RTM_BASE] = { .doit = neigh_delete }, + [RTM_GETNEIGH - RTM_BASE] = { .dumpit = neigh_dump_info }, + [RTM_GETRULE - RTM_BASE] = { .dumpit = rtnetlink_dump_all }, + [RTM_GETNEIGHTBL - RTM_BASE] = { .dumpit = neightbl_dump_info }, + [RTM_SETNEIGHTBL - RTM_BASE] = { .doit = neightbl_set }, }; static int rtnetlink_event(struct notifier_block *this, unsigned long event, void *ptr) diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c index f0fb6d76f7c5..92b057becb4b 100644 --- a/security/selinux/nlmsgtab.c +++ b/security/selinux/nlmsgtab.c @@ -63,6 +63,8 @@ static struct nlmsg_perm nlmsg_route_perms[] = { RTM_GETPREFIX, NETLINK_ROUTE_SOCKET__NLMSG_READ }, { RTM_GETMULTICAST, NETLINK_ROUTE_SOCKET__NLMSG_READ }, { RTM_GETANYCAST, NETLINK_ROUTE_SOCKET__NLMSG_READ }, + { RTM_GETNEIGHTBL, NETLINK_ROUTE_SOCKET__NLMSG_READ }, + { RTM_SETNEIGHTBL, NETLINK_ROUTE_SOCKET__NLMSG_WRITE }, }; static struct nlmsg_perm nlmsg_firewall_perms[] = -- cgit v1.2.3 From 88121aea7bdb5fdc527388e262381829c4e1db16 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:51:12 -0700 Subject: [NEIGHBOUR]: Remove unused fields in struct neigh_parms and neigh_table Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/net/neighbour.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include') diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 17191ac9be70..89809891e5ab 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -69,8 +69,6 @@ struct neigh_parms struct neigh_parms *next; int (*neigh_setup)(struct neighbour *); struct neigh_table *tbl; - int entries; - void *priv; void *sysctl_table; @@ -193,7 +191,6 @@ struct neigh_table atomic_t entries; rwlock_t lock; unsigned long last_rand; - struct neigh_parms *parms_list; kmem_cache_t *kmem_cachep; struct neigh_statistics *stats; struct neighbour **hash_buckets; -- cgit v1.2.3 From c52a3f89f882b84fc422000655c023fe73e701cf Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:51:26 -0700 Subject: [NETLINK]: Fix RTA_NEST_CANCEL(). Only skb_trim() if 'start' is non-NULL. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 5a5cda160267..6fff8c4c99c7 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -924,7 +924,8 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi (skb)->len; }) #define RTA_NEST_CANCEL(skb, start) \ -({ skb_trim(skb, (unsigned char *) (start) - (skb)->data); \ +({ if (start) \ + skb_trim(skb, (unsigned char *) (start) - (skb)->data); \ -1; }) #define RTA_GET_U32(rta) \ -- cgit v1.2.3 From 8f48bcd4ef11a69add178fc3111a77e7ee95bacd Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:52:36 -0700 Subject: [RTNETLINK]: Add RTA_(PUT|GET) shortcuts for u8, u16, and flag Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) (limited to 'include') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index 6fff8c4c99c7..e68dbf0bf579 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -897,6 +897,14 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi goto rtattr_failure; \ memcpy(skb_put(skb, RTA_ALIGN(attrlen)), data, attrlen); }) +#define RTA_PUT_U8(skb, attrtype, value) \ +({ u8 _tmp = (value); \ + RTA_PUT(skb, attrtype, sizeof(u8), &_tmp); }) + +#define RTA_PUT_U16(skb, attrtype, value) \ +({ u16 _tmp = (value); \ + RTA_PUT(skb, attrtype, sizeof(u16), &_tmp); }) + #define RTA_PUT_U32(skb, attrtype, value) \ ({ u32 _tmp = (value); \ RTA_PUT(skb, attrtype, sizeof(u32), &_tmp); }) @@ -914,6 +922,9 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi #define RTA_PUT_STRING(skb, attrtype, value) \ RTA_PUT(skb, attrtype, strlen(value) + 1, value) +#define RTA_PUT_FLAG(skb, attrtype) \ + RTA_PUT(skb, attrtype, 0, NULL); + #define RTA_NEST(skb, type) \ ({ struct rtattr *__start = (struct rtattr *) (skb)->tail; \ RTA_PUT(skb, type, 0, NULL); \ @@ -928,6 +939,16 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi skb_trim(skb, (unsigned char *) (start) - (skb)->data); \ -1; }) +#define RTA_GET_U8(rta) \ +({ if (!rta || RTA_PAYLOAD(rta) < sizeof(u8)) \ + goto rtattr_failure; \ + *(u8 *) RTA_DATA(rta); }) + +#define RTA_GET_U16(rta) \ +({ if (!rta || RTA_PAYLOAD(rta) < sizeof(u16)) \ + goto rtattr_failure; \ + *(u16 *) RTA_DATA(rta); }) + #define RTA_GET_U32(rta) \ ({ if (!rta || RTA_PAYLOAD(rta) < sizeof(u32)) \ goto rtattr_failure; \ @@ -940,6 +961,8 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi memcpy(&_tmp, RTA_DATA(rta), sizeof(_tmp)); \ _tmp; }) +#define RTA_GET_FLAG(rta) (!!(rta)) + #define RTA_GET_SECS(rta) ((unsigned long) RTA_GET_U64(rta) * HZ) #define RTA_GET_MSECS(rta) (msecs_to_jiffies((unsigned long) RTA_GET_U64(rta))) -- cgit v1.2.3 From 1797754ea7ee5e0d859b0a32506ff999f8d5fb71 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:53:48 -0700 Subject: [NETLINK]: Introduce NLMSG_NEW macro to better handle netlink flags Introduces a new macro NLMSG_NEW which extends NLMSG_PUT but takes a flags argument. NLMSG_PUT stays there for compatibility but now calls NLMSG_NEW with flags == 0. NLMSG_PUT_ANSWER is renamed to NLMSG_NEW_ANSWER which now also takes a flags argument. Also converts the users of NLMSG_PUT_ANSWER to use NLMSG_NEW_ANSWER and fixes the two direct users of __nlmsg_put to either provide the flags or use NLMSG_NEW(_ANSWER). Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/netlink.h | 17 ++++++++++------- net/core/neighbour.c | 8 ++++---- net/netlink/af_netlink.c | 8 +++++--- 3 files changed, 19 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 8d1cb419a930..e38407a23d04 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -156,7 +156,7 @@ struct netlink_notify }; static __inline__ struct nlmsghdr * -__nlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, int type, int len) +__nlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, int type, int len, int flags) { struct nlmsghdr *nlh; int size = NLMSG_LENGTH(len); @@ -164,20 +164,23 @@ __nlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, int type, int len) nlh = (struct nlmsghdr*)skb_put(skb, NLMSG_ALIGN(size)); nlh->nlmsg_type = type; nlh->nlmsg_len = size; - nlh->nlmsg_flags = 0; + nlh->nlmsg_flags = flags; nlh->nlmsg_pid = pid; nlh->nlmsg_seq = seq; return nlh; } -#define NLMSG_PUT(skb, pid, seq, type, len) \ +#define NLMSG_NEW(skb, pid, seq, type, len, flags) \ ({ if (skb_tailroom(skb) < (int)NLMSG_SPACE(len)) \ goto nlmsg_failure; \ - __nlmsg_put(skb, pid, seq, type, len); }) + __nlmsg_put(skb, pid, seq, type, len, flags); }) + +#define NLMSG_PUT(skb, pid, seq, type, len) \ + NLMSG_NEW(skb, pid, seq, type, len, 0) -#define NLMSG_PUT_ANSWER(skb, cb, type, len) \ - NLMSG_PUT(skb, NETLINK_CB((cb)->skb).pid, \ - (cb)->nlh->nlmsg_seq, type, len) +#define NLMSG_NEW_ANSWER(skb, cb, type, len, flags) \ + NLMSG_NEW(skb, NETLINK_CB((cb)->skb).pid, \ + (cb)->nlh->nlmsg_seq, type, len, flags) #define NLMSG_END(skb, nlh) \ ({ (nlh)->nlmsg_len = (skb)->tail - (unsigned char *) (nlh); \ diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 2296a145fb78..0fb742e228cc 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1590,8 +1590,8 @@ static int neightbl_fill_info(struct neigh_table *tbl, struct sk_buff *skb, struct nlmsghdr *nlh; struct ndtmsg *ndtmsg; - nlh = NLMSG_PUT_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg)); - nlh->nlmsg_flags |= NLM_F_MULTI; + nlh = NLMSG_NEW_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg), + NLM_F_MULTI); ndtmsg = NLMSG_DATA(nlh); @@ -1675,8 +1675,8 @@ static int neightbl_fill_param_info(struct neigh_table *tbl, struct ndtmsg *ndtmsg; struct nlmsghdr *nlh; - nlh = NLMSG_PUT_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg)); - nlh->nlmsg_flags |= NLM_F_MULTI; + nlh = NLMSG_NEW_ANSWER(skb, cb, RTM_NEWNEIGHTBL, sizeof(struct ndtmsg), + NLM_F_MULTI); ndtmsg = NLMSG_DATA(nlh); diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index e41ce458c2a9..70bcd4744d93 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1095,8 +1095,7 @@ static int netlink_dump(struct sock *sk) return 0; } - nlh = __nlmsg_put(skb, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, NLMSG_DONE, sizeof(int)); - nlh->nlmsg_flags |= NLM_F_MULTI; + nlh = NLMSG_NEW_ANSWER(skb, cb, NLMSG_DONE, sizeof(len), NLM_F_MULTI); memcpy(NLMSG_DATA(nlh), &len, sizeof(len)); skb_queue_tail(&sk->sk_receive_queue, skb); sk->sk_data_ready(sk, skb->len); @@ -1107,6 +1106,9 @@ static int netlink_dump(struct sock *sk) netlink_destroy_callback(cb); return 0; + +nlmsg_failure: + return -ENOBUFS; } int netlink_dump_start(struct sock *ssk, struct sk_buff *skb, @@ -1178,7 +1180,7 @@ void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err) } rep = __nlmsg_put(skb, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq, - NLMSG_ERROR, sizeof(struct nlmsgerr)); + NLMSG_ERROR, sizeof(struct nlmsgerr), 0); errmsg = NLMSG_DATA(rep); errmsg->error = err; memcpy(&errmsg->msg, nlh, err ? nlh->nlmsg_len : sizeof(struct nlmsghdr)); -- cgit v1.2.3 From 0603eac0d6b77acac5924a2734228cbaf072f993 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Sat, 18 Jun 2005 22:54:36 -0700 Subject: [IPSEC]: Add XFRMA_SA/XFRMA_POLICY for delete notification This patch changes the format of the XFRM_MSG_DELSA and XFRM_MSG_DELPOLICY notification so that the main message sent is of the same format as that received by the kernel if the original message was via netlink. This also means that we won't lose the byid information carried in km_event. Since this user interface is introduced by Jamal's patch we can still afford to change it. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/xfrm.h | 2 ++ net/xfrm/xfrm_user.c | 47 +++++++++++++++++++++++++++++++++++++++++++---- 2 files changed, 45 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index 03bc600516ea..d68391a9b9f3 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -174,6 +174,8 @@ enum xfrm_attr_type_t { XFRMA_ALG_COMP, /* struct xfrm_algo */ XFRMA_ENCAP, /* struct xfrm_algo + struct xfrm_encap_tmpl */ XFRMA_TMPL, /* 1 or more struct xfrm_user_tmpl */ + XFRMA_SA, + XFRMA_POLICY, __XFRMA_MAX #define XFRMA_MAX (__XFRMA_MAX - 1) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index ffe1b217347c..5ce8558eac91 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1168,7 +1168,7 @@ nlmsg_failure: static int inline xfrm_sa_len(struct xfrm_state *x) { - int l = NLMSG_LENGTH(sizeof(struct xfrm_usersa_info)); + int l = 0; if (x->aalg) l += RTA_SPACE(sizeof(*x->aalg) + (x->aalg->alg_key_len+7)/8); if (x->ealg) @@ -1184,20 +1184,39 @@ static int inline xfrm_sa_len(struct xfrm_state *x) static int xfrm_notify_sa(struct xfrm_state *x, struct km_event *c) { struct xfrm_usersa_info *p; + struct xfrm_usersa_id *id; struct nlmsghdr *nlh; struct sk_buff *skb; unsigned char *b; int len = xfrm_sa_len(x); + int headlen; + + headlen = sizeof(*p); + if (c->event == XFRM_MSG_DELSA) { + len += RTA_SPACE(headlen); + headlen = sizeof(*id); + } + len += NLMSG_SPACE(headlen); skb = alloc_skb(len, GFP_ATOMIC); if (skb == NULL) return -ENOMEM; b = skb->tail; - nlh = NLMSG_PUT(skb, c->pid, c->seq, c->event, sizeof(*p)); + nlh = NLMSG_PUT(skb, c->pid, c->seq, c->event, headlen); nlh->nlmsg_flags = 0; p = NLMSG_DATA(nlh); + if (c->event == XFRM_MSG_DELSA) { + id = NLMSG_DATA(nlh); + memcpy(&id->daddr, &x->id.daddr, sizeof(id->daddr)); + id->spi = x->id.spi; + id->family = x->props.family; + id->proto = x->id.proto; + + p = RTA_DATA(__RTA_PUT(skb, XFRMA_SA, sizeof(*p))); + } + copy_to_user_state(x, p); if (x->aalg) @@ -1398,20 +1417,39 @@ static int xfrm_exp_policy_notify(struct xfrm_policy *xp, int dir, struct km_eve static int xfrm_notify_policy(struct xfrm_policy *xp, int dir, struct km_event *c) { struct xfrm_userpolicy_info *p; + struct xfrm_userpolicy_id *id; struct nlmsghdr *nlh; struct sk_buff *skb; unsigned char *b; int len = RTA_SPACE(sizeof(struct xfrm_user_tmpl) * xp->xfrm_nr); - len += NLMSG_SPACE(sizeof(struct xfrm_userpolicy_info)); + int headlen; + + headlen = sizeof(*p); + if (c->event == XFRM_MSG_DELPOLICY) { + len += RTA_SPACE(headlen); + headlen = sizeof(*id); + } + len += NLMSG_SPACE(headlen); skb = alloc_skb(len, GFP_ATOMIC); if (skb == NULL) return -ENOMEM; b = skb->tail; - nlh = NLMSG_PUT(skb, c->pid, c->seq, c->event, sizeof(*p)); + nlh = NLMSG_PUT(skb, c->pid, c->seq, c->event, headlen); p = NLMSG_DATA(nlh); + if (c->event == XFRM_MSG_DELPOLICY) { + id = NLMSG_DATA(nlh); + memset(id, 0, sizeof(*id)); + id->dir = dir; + if (c->data.byid) + id->index = xp->index; + else + memcpy(&id->sel, &xp->selector, sizeof(id->sel)); + + p = RTA_DATA(__RTA_PUT(skb, XFRMA_POLICY, sizeof(*p))); + } nlh->nlmsg_flags = 0; @@ -1424,6 +1462,7 @@ static int xfrm_notify_policy(struct xfrm_policy *xp, int dir, struct km_event * return netlink_broadcast(xfrm_nl, skb, 0, XFRMGRP_POLICY, GFP_ATOMIC); nlmsg_failure: +rtattr_failure: kfree_skb(skb); return -1; } -- cgit v1.2.3 From 9972b25d0c6e7f8f893eb3444dea37b42b1201de Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Sat, 18 Jun 2005 22:57:26 -0700 Subject: [PKT_SCHED]: Generic queue management interface for qdiscs using internal skb queues Implements an interface to be used by leaf qdiscs maintaining an internal skb queue. The interface maintains a backlog in bytes additionaly to the skb_queue_len() maintained by the queue itself. Relevant statistics get incremented automatically. Every function comes in two variants, one assuming Qdisc->q is used as queue and the second taking a sk_buff_head as argument. Be aware that, if you use multiple queues, you still have to maintain the Qdisc->q.qlen counter yourself. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/net/sch_generic.h | 122 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 122 insertions(+) (limited to 'include') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index c57504b3b518..7b97405e2dbf 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -172,4 +172,126 @@ tcf_destroy(struct tcf_proto *tp) kfree(tp); } +static inline int __qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff_head *list) +{ + __skb_queue_tail(list, skb); + sch->qstats.backlog += skb->len; + sch->bstats.bytes += skb->len; + sch->bstats.packets++; + + return NET_XMIT_SUCCESS; +} + +static inline int qdisc_enqueue_tail(struct sk_buff *skb, struct Qdisc *sch) +{ + return __qdisc_enqueue_tail(skb, sch, &sch->q); +} + +static inline struct sk_buff *__qdisc_dequeue_head(struct Qdisc *sch, + struct sk_buff_head *list) +{ + struct sk_buff *skb = __skb_dequeue(list); + + if (likely(skb != NULL)) + sch->qstats.backlog -= skb->len; + + return skb; +} + +static inline struct sk_buff *qdisc_dequeue_head(struct Qdisc *sch) +{ + return __qdisc_dequeue_head(sch, &sch->q); +} + +static inline struct sk_buff *__qdisc_dequeue_tail(struct Qdisc *sch, + struct sk_buff_head *list) +{ + struct sk_buff *skb = __skb_dequeue_tail(list); + + if (likely(skb != NULL)) + sch->qstats.backlog -= skb->len; + + return skb; +} + +static inline struct sk_buff *qdisc_dequeue_tail(struct Qdisc *sch) +{ + return __qdisc_dequeue_tail(sch, &sch->q); +} + +static inline int __qdisc_requeue(struct sk_buff *skb, struct Qdisc *sch, + struct sk_buff_head *list) +{ + __skb_queue_head(list, skb); + sch->qstats.backlog += skb->len; + sch->qstats.requeues++; + + return NET_XMIT_SUCCESS; +} + +static inline int qdisc_requeue(struct sk_buff *skb, struct Qdisc *sch) +{ + return __qdisc_requeue(skb, sch, &sch->q); +} + +static inline void __qdisc_reset_queue(struct Qdisc *sch, + struct sk_buff_head *list) +{ + /* + * We do not know the backlog in bytes of this list, it + * is up to the caller to correct it + */ + skb_queue_purge(list); +} + +static inline void qdisc_reset_queue(struct Qdisc *sch) +{ + __qdisc_reset_queue(sch, &sch->q); + sch->qstats.backlog = 0; +} + +static inline unsigned int __qdisc_queue_drop(struct Qdisc *sch, + struct sk_buff_head *list) +{ + struct sk_buff *skb = __qdisc_dequeue_tail(sch, list); + + if (likely(skb != NULL)) { + unsigned int len = skb->len; + kfree_skb(skb); + return len; + } + + return 0; +} + +static inline unsigned int qdisc_queue_drop(struct Qdisc *sch) +{ + return __qdisc_queue_drop(sch, &sch->q); +} + +static inline int qdisc_drop(struct sk_buff *skb, struct Qdisc *sch) +{ + kfree_skb(skb); + sch->qstats.drops++; + + return NET_XMIT_DROP; +} + +static inline int qdisc_reshape_fail(struct sk_buff *skb, struct Qdisc *sch) +{ + sch->qstats.drops++; + +#ifdef CONFIG_NET_CLS_POLICE + if (sch->reshape_fail == NULL || sch->reshape_fail(skb, sch)) + goto drop; + + return NET_XMIT_SUCCESS; + +drop: +#endif + kfree_skb(skb); + return NET_XMIT_DROP; +} + #endif -- cgit v1.2.3 From fe6ef2daa29df8fef1a95446faccd18ab163993e Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 19 Jun 2005 09:52:07 +0100 Subject: [PATCH] ARM SMP: Add missed files from Integrator/CP platform Add missed new files from basic SMP support for the Integrator/CP platform. Signed-off-by: Russell King --- arch/arm/mach-integrator/headsmp.S | 37 +++++++ arch/arm/mach-integrator/platsmp.c | 192 ++++++++++++++++++++++++++++++++++ include/asm-arm/arch-integrator/smp.h | 19 ++++ 3 files changed, 248 insertions(+) create mode 100644 arch/arm/mach-integrator/headsmp.S create mode 100644 arch/arm/mach-integrator/platsmp.c create mode 100644 include/asm-arm/arch-integrator/smp.h (limited to 'include') diff --git a/arch/arm/mach-integrator/headsmp.S b/arch/arm/mach-integrator/headsmp.S new file mode 100644 index 000000000000..ceaa88e30d70 --- /dev/null +++ b/arch/arm/mach-integrator/headsmp.S @@ -0,0 +1,37 @@ +/* + * linux/arch/arm/mach-integrator/headsmp.S + * + * Copyright (c) 2003 ARM Limited + * All Rights Reserved + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include + + __INIT + +/* + * Integrator specific entry point for secondary CPUs. This provides + * a "holding pen" into which all secondary cores are held until we're + * ready for them to initialise. + */ +ENTRY(integrator_secondary_startup) + adr r4, 1f + ldmia r4, {r5, r6} + sub r4, r4, r5 + ldr r6, [r6, r4] +pen: ldr r7, [r6] + cmp r7, r0 + bne pen + + /* + * we've been released from the holding pen: secondary_stack + * should now contain the SVC stack for this core + */ + b secondary_startup + +1: .long . + .long phys_pen_release diff --git a/arch/arm/mach-integrator/platsmp.c b/arch/arm/mach-integrator/platsmp.c new file mode 100644 index 000000000000..ead15dfcb53d --- /dev/null +++ b/arch/arm/mach-integrator/platsmp.c @@ -0,0 +1,192 @@ +/* + * linux/arch/arm/mach-cintegrator/platsmp.c + * + * Copyright (C) 2002 ARM Ltd. + * All Rights Reserved + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +extern void integrator_secondary_startup(void); + +/* + * control for which core is the next to come out of the secondary + * boot "holding pen" + */ +volatile int __initdata pen_release = -1; +unsigned long __initdata phys_pen_release = 0; + +static DEFINE_SPINLOCK(boot_lock); + +void __init platform_secondary_init(unsigned int cpu) +{ + /* + * the primary core may have used a "cross call" soft interrupt + * to get this processor out of WFI in the BootMonitor - make + * sure that we are no longer being sent this soft interrupt + */ + smp_cross_call_done(cpumask_of_cpu(cpu)); + + /* + * if any interrupts are already enabled for the primary + * core (e.g. timer irq), then they will not have been enabled + * for us: do so + */ + secondary_scan_irqs(); + + /* + * let the primary processor know we're out of the + * pen, then head off into the C entry point + */ + pen_release = -1; + + /* + * Synchronise with the boot thread. + */ + spin_lock(&boot_lock); + spin_unlock(&boot_lock); +} + +int __init boot_secondary(unsigned int cpu, struct task_struct *idle) +{ + unsigned long timeout; + + /* + * set synchronisation state between this boot processor + * and the secondary one + */ + spin_lock(&boot_lock); + + /* + * The secondary processor is waiting to be released from + * the holding pen - release it, then wait for it to flag + * that it has been released by resetting pen_release. + * + * Note that "pen_release" is the hardware CPU ID, whereas + * "cpu" is Linux's internal ID. + */ + pen_release = cpu; + + /* + * XXX + * + * This is a later addition to the booting protocol: the + * bootMonitor now puts secondary cores into WFI, so + * poke_milo() no longer gets the cores moving; we need + * to send a soft interrupt to wake the secondary core. + * Use smp_cross_call() for this, since there's little + * point duplicating the code here + */ + smp_cross_call(cpumask_of_cpu(cpu)); + + timeout = jiffies + (1 * HZ); + while (time_before(jiffies, timeout)) { + if (pen_release == -1) + break; + + udelay(10); + } + + /* + * now the secondary core is starting up let it run its + * calibrations, then wait for it to finish + */ + spin_unlock(&boot_lock); + + return pen_release != -1 ? -ENOSYS : 0; +} + +static void __init poke_milo(void) +{ + extern void secondary_startup(void); + + /* nobody is to be released from the pen yet */ + pen_release = -1; + + phys_pen_release = virt_to_phys(&pen_release); + + /* + * write the address of secondary startup into the system-wide + * flags register, then clear the bottom two bits, which is what + * BootMonitor is waiting for + */ +#if 1 +#define CINTEGRATOR_HDR_FLAGSS_OFFSET 0x30 + __raw_writel(virt_to_phys(integrator_secondary_startup), + (IO_ADDRESS(INTEGRATOR_HDR_BASE) + + CINTEGRATOR_HDR_FLAGSS_OFFSET)); +#define CINTEGRATOR_HDR_FLAGSC_OFFSET 0x34 + __raw_writel(3, + (IO_ADDRESS(INTEGRATOR_HDR_BASE) + + CINTEGRATOR_HDR_FLAGSC_OFFSET)); +#endif + + mb(); +} + +void __init smp_prepare_cpus(unsigned int max_cpus) +{ + unsigned int ncores = get_core_count(); + unsigned int cpu = smp_processor_id(); + int i; + + /* sanity check */ + if (ncores == 0) { + printk(KERN_ERR + "Integrator/CP: strange CM count of 0? Default to 1\n"); + + ncores = 1; + } + + if (ncores > NR_CPUS) { + printk(KERN_WARNING + "Integrator/CP: no. of cores (%d) greater than configured " + "maximum of %d - clipping\n", + ncores, NR_CPUS); + ncores = NR_CPUS; + } + + /* + * start with some more config for the Boot CPU, now that + * the world is a bit more alive (which was not the case + * when smp_prepare_boot_cpu() was called) + */ + smp_store_cpu_info(cpu); + + /* + * are we trying to boot more cores than exist? + */ + if (max_cpus > ncores) + max_cpus = ncores; + + /* + * Initialise the present mask - this tells us which CPUs should + * be present. + */ + for (i = 0; i < max_cpus; i++) { + cpu_set(i, cpu_present_mask); + } + + /* + * Do we need any more CPUs? If so, then let them know where + * to start. Note that, on modern versions of MILO, the "poke" + * doesn't actually do anything until each individual core is + * sent a soft interrupt to get it out of WFI + */ + if (max_cpus > 1) + poke_milo(); +} diff --git a/include/asm-arm/arch-integrator/smp.h b/include/asm-arm/arch-integrator/smp.h new file mode 100644 index 000000000000..0ec7093f7c37 --- /dev/null +++ b/include/asm-arm/arch-integrator/smp.h @@ -0,0 +1,19 @@ +#ifndef ASMARM_ARCH_SMP_H +#define ASMARM_ARCH_SMP_H + +#include + +#include +#include + +#define hard_smp_processor_id() \ + ({ \ + unsigned int cpunum; \ + __asm__("mrc p15, 0, %0, c0, c0, 5" \ + : "=r" (cpunum)); \ + cpunum &= 0x0F; \ + }) + +extern void secondary_scan_irqs(void); + +#endif -- cgit v1.2.3 From 36c5ed23b9f535d1c79986efb45f9c1f115e0997 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 19 Jun 2005 18:39:33 +0100 Subject: [PATCH] ARM SMP: Fix PXA/SA11x0 suspend resume crash We need to re-initialise the stack pointers for undefined, IRQ and abort mode handlers whenever we resume. Signed-off-by: Russell King --- arch/arm/kernel/setup.c | 2 +- arch/arm/mach-pxa/pm.c | 2 ++ arch/arm/mach-sa1100/pm.c | 2 ++ include/asm-arm/system.h | 1 + 4 files changed, 6 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 7ecdda3f1253..9fed5fa194d9 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -328,7 +328,7 @@ static void __init setup_processor(void) * cpu_init dumps the cache information, initialises SMP specific * information, and sets up the per-CPU stacks. */ -void __init cpu_init(void) +void cpu_init(void) { unsigned int cpu = smp_processor_id(); struct stack *stk = &stacks[cpu]; diff --git a/arch/arm/mach-pxa/pm.c b/arch/arm/mach-pxa/pm.c index 9799fe80df23..ac4dd4336160 100644 --- a/arch/arm/mach-pxa/pm.c +++ b/arch/arm/mach-pxa/pm.c @@ -133,6 +133,8 @@ static int pxa_pm_enter(suspend_state_t state) /* *** go zzz *** */ pxa_cpu_pm_enter(state); + cpu_init(); + /* after sleeping, validate the checksum */ checksum = 0; for (i = 0; i < SLEEP_SAVE_SIZE - 1; i++) diff --git a/arch/arm/mach-sa1100/pm.c b/arch/arm/mach-sa1100/pm.c index 379ea5e3950f..59c7964cfe11 100644 --- a/arch/arm/mach-sa1100/pm.c +++ b/arch/arm/mach-sa1100/pm.c @@ -88,6 +88,8 @@ static int sa11x0_pm_enter(suspend_state_t state) /* go zzz */ sa1100_cpu_suspend(); + cpu_init(); + /* * Ensure not to come back here if it wasn't intended */ diff --git a/include/asm-arm/system.h b/include/asm-arm/system.h index b13a8da4847b..8405eb6558ed 100644 --- a/include/asm-arm/system.h +++ b/include/asm-arm/system.h @@ -104,6 +104,7 @@ extern void show_pte(struct mm_struct *mm, unsigned long addr); extern void __show_regs(struct pt_regs *); extern int cpu_architecture(void); +extern void cpu_init(void); #define set_cr(x) \ __asm__ __volatile__( \ -- cgit v1.2.3 From b8a9b66fbee09d0cc71c272b5c1d1f3320afbbf0 Mon Sep 17 00:00:00 2001 From: Russell King Date: Mon, 20 Jun 2005 11:31:09 +0100 Subject: [PATCH] ARM: Add common CACHE_COLOUR macro Signed-off-by: Russell King --- arch/arm/mm/copypage-v6.c | 6 ++---- include/asm-arm/cacheflush.h | 3 +++ 2 files changed, 5 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/arch/arm/mm/copypage-v6.c b/arch/arm/mm/copypage-v6.c index a8c00236bd3d..27d041574ea7 100644 --- a/arch/arm/mm/copypage-v6.c +++ b/arch/arm/mm/copypage-v6.c @@ -30,8 +30,6 @@ static DEFINE_SPINLOCK(v6_lock); -#define DCACHE_COLOUR(vaddr) ((vaddr & (SHMLBA - 1)) >> PAGE_SHIFT) - /* * Copy the user page. No aliasing to deal with so we can just * attack the kernel's existing mapping of these pages. @@ -55,7 +53,7 @@ void v6_clear_user_page_nonaliasing(void *kaddr, unsigned long vaddr) */ void v6_copy_user_page_aliasing(void *kto, const void *kfrom, unsigned long vaddr) { - unsigned int offset = DCACHE_COLOUR(vaddr); + unsigned int offset = CACHE_COLOUR(vaddr); unsigned long from, to; /* @@ -95,7 +93,7 @@ void v6_copy_user_page_aliasing(void *kto, const void *kfrom, unsigned long vadd */ void v6_clear_user_page_aliasing(void *kaddr, unsigned long vaddr) { - unsigned int offset = DCACHE_COLOUR(vaddr); + unsigned int offset = CACHE_COLOUR(vaddr); unsigned long to = to_address + (offset << PAGE_SHIFT); /* diff --git a/include/asm-arm/cacheflush.h b/include/asm-arm/cacheflush.h index 09ffeed507c2..035cdcff43d2 100644 --- a/include/asm-arm/cacheflush.h +++ b/include/asm-arm/cacheflush.h @@ -16,6 +16,9 @@ #include #include +#include + +#define CACHE_COLOUR(vaddr) ((vaddr & (SHMLBA - 1)) >> PAGE_SHIFT) /* * Cache Model -- cgit v1.2.3 From 09f0551d20ddf6d22c333adcc59f2b1148734273 Mon Sep 17 00:00:00 2001 From: Russell King Date: Mon, 20 Jun 2005 18:44:37 +0100 Subject: [PATCH] ARM: Add iomap support for ARM Signed-off-by: Russell King --- arch/arm/Kconfig | 4 ---- arch/arm/mm/ioremap.c | 47 +++++++++++++++++++++++++++++++++++++++++++++++ include/asm-arm/io.h | 27 +++++++++++++++++++++++++++ 3 files changed, 74 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 475950c8a831..1a1773f81393 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -67,10 +67,6 @@ config GENERIC_BUST_SPINLOCK config GENERIC_ISA_DMA bool -config GENERIC_IOMAP - bool - default y - config FIQ bool diff --git a/arch/arm/mm/ioremap.c b/arch/arm/mm/ioremap.c index 00bb8fd37a59..7110e54182b1 100644 --- a/arch/arm/mm/ioremap.c +++ b/arch/arm/mm/ioremap.c @@ -170,3 +170,50 @@ void __iounmap(void __iomem *addr) vfree((void *) (PAGE_MASK & (unsigned long) addr)); } EXPORT_SYMBOL(__iounmap); + +#ifdef __io +void __iomem *ioport_map(unsigned long port, unsigned int nr) +{ + return __io(port); +} +EXPORT_SYMBOL(ioport_map); + +void ioport_unmap(void __iomem *addr) +{ +} +EXPORT_SYMBOL(ioport_unmap); +#endif + +#ifdef CONFIG_PCI +#include +#include + +void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen) +{ + unsigned long start = pci_resource_start(dev, bar); + unsigned long len = pci_resource_len(dev, bar); + unsigned long flags = pci_resource_flags(dev, bar); + + if (!len || !start) + return NULL; + if (maxlen && len > maxlen) + len = maxlen; + if (flags & IORESOURCE_IO) + return ioport_map(start, len); + if (flags & IORESOURCE_MEM) { + if (flags & IORESOURCE_CACHEABLE) + return ioremap(start, len); + return ioremap_nocache(start, len); + } + return NULL; +} +EXPORT_SYMBOL(pci_iomap); + +void pci_iounmap(struct pci_dev *dev, void __iomem *addr) +{ + if ((unsigned long)addr >= VMALLOC_START && + (unsigned long)addr < VMALLOC_END) + iounmap(addr); +} +EXPORT_SYMBOL(pci_iounmap); +#endif diff --git a/include/asm-arm/io.h b/include/asm-arm/io.h index 658ffa384fda..08a46302d265 100644 --- a/include/asm-arm/io.h +++ b/include/asm-arm/io.h @@ -272,6 +272,33 @@ extern void __iounmap(void __iomem *addr); #define iounmap(cookie) __arch_iounmap(cookie) #endif +/* + * io{read,write}{8,16,32} macros + */ +#define ioread8(p) ({ unsigned int __v = __raw_readb(p); __v; }) +#define ioread16(p) ({ unsigned int __v = le16_to_cpu(__raw_readw(p)); __v; }) +#define ioread32(p) ({ unsigned int __v = le32_to_cpu(__raw_readl(p)); __v; }) + +#define iowrite8(v,p) __raw_writeb(v, p) +#define iowrite16(v,p) __raw_writew(cpu_to_le16(v), p) +#define iowrite32(v,p) __raw_writel(cpu_to_le32(v), p) + +#define ioread8_rep(p,d,c) __raw_readsb(p,d,c) +#define ioread16_rep(p,d,c) __raw_readsw(p,d,c) +#define ioread32_rep(p,d,c) __raw_readsl(p,d,c) + +#define iowrite8_rep(p,s,c) __raw_writesb(p,s,c) +#define iowrite16_rep(p,s,c) __raw_writesw(p,s,c) +#define iowrite32_rep(p,s,c) __raw_writesl(p,s,c) + +extern void __iomem *ioport_map(unsigned long port, unsigned int nr); +extern void ioport_unmap(void __iomem *addr); + +struct pci_dev; + +extern void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen); +extern void pci_iounmap(struct pci_dev *dev, void __iomem *addr); + /* * can the hardware map this into one segment or not, given no other * constraints. -- cgit v1.2.3 From 038c5b602524b33447008492e932cdd0a1e806c9 Mon Sep 17 00:00:00 2001 From: Bellido Nicolas Date: Mon, 20 Jun 2005 18:51:05 +0100 Subject: [PATCH] ARM: 2686/2: AAEC-2000 Core support Patch from Bellido Nicolas Core support for AAEC-2000 based platforms. This is an updated version of the previous patch, and takes into account Russell's comments. AAED-2000 default configuration will follow as soon as some problems with the bootloader are sorted out... Signed-off-by: Nicolas Bellido Signed-off-by: Russell King --- arch/arm/Kconfig | 7 ++ arch/arm/Makefile | 1 + arch/arm/mach-aaec2000/Kconfig | 11 ++ arch/arm/mach-aaec2000/Makefile | 9 ++ arch/arm/mach-aaec2000/aaed2000.c | 48 +++++++++ arch/arm/mach-aaec2000/core.c | 157 ++++++++++++++++++++++++++++ arch/arm/mach-aaec2000/core.h | 16 +++ arch/arm/mm/Kconfig | 2 +- include/asm-arm/arch-aaec2000/aaec2000.h | 151 ++++++++++++++++++++++++++ include/asm-arm/arch-aaec2000/debug-macro.S | 36 +++++++ include/asm-arm/arch-aaec2000/dma.h | 17 +++ include/asm-arm/arch-aaec2000/entry-macro.S | 33 ++++++ include/asm-arm/arch-aaec2000/hardware.h | 49 +++++++++ include/asm-arm/arch-aaec2000/io.h | 19 ++++ include/asm-arm/arch-aaec2000/irqs.h | 46 ++++++++ include/asm-arm/arch-aaec2000/memory.h | 73 +++++++++++++ include/asm-arm/arch-aaec2000/param.h | 15 +++ include/asm-arm/arch-aaec2000/system.h | 24 +++++ include/asm-arm/arch-aaec2000/timex.h | 18 ++++ include/asm-arm/arch-aaec2000/uncompress.h | 47 +++++++++ include/asm-arm/arch-aaec2000/vmalloc.h | 16 +++ 21 files changed, 794 insertions(+), 1 deletion(-) create mode 100644 arch/arm/mach-aaec2000/Kconfig create mode 100644 arch/arm/mach-aaec2000/Makefile create mode 100644 arch/arm/mach-aaec2000/aaed2000.c create mode 100644 arch/arm/mach-aaec2000/core.c create mode 100644 arch/arm/mach-aaec2000/core.h create mode 100644 include/asm-arm/arch-aaec2000/aaec2000.h create mode 100644 include/asm-arm/arch-aaec2000/debug-macro.S create mode 100644 include/asm-arm/arch-aaec2000/dma.h create mode 100644 include/asm-arm/arch-aaec2000/entry-macro.S create mode 100644 include/asm-arm/arch-aaec2000/hardware.h create mode 100644 include/asm-arm/arch-aaec2000/io.h create mode 100644 include/asm-arm/arch-aaec2000/irqs.h create mode 100644 include/asm-arm/arch-aaec2000/memory.h create mode 100644 include/asm-arm/arch-aaec2000/param.h create mode 100644 include/asm-arm/arch-aaec2000/system.h create mode 100644 include/asm-arm/arch-aaec2000/timex.h create mode 100644 include/asm-arm/arch-aaec2000/uncompress.h create mode 100644 include/asm-arm/arch-aaec2000/vmalloc.h (limited to 'include') diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 1a1773f81393..efdb12d73566 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -198,6 +198,11 @@ config ARCH_H720X help This enables support for systems based on the Hynix HMS720x +config ARCH_AAEC2000 + bool "Agilent AAEC-2000 based" + help + This enables support for systems based on the Agilent AAEC-2000 + endchoice source "arch/arm/mach-clps711x/Kconfig" @@ -230,6 +235,8 @@ source "arch/arm/mach-h720x/Kconfig" source "arch/arm/mach-versatile/Kconfig" +source "arch/arm/mach-aaec2000/Kconfig" + # Definitions to make life easier config ARCH_ACORN bool diff --git a/arch/arm/Makefile b/arch/arm/Makefile index 2277e3d179cc..8330495e2448 100644 --- a/arch/arm/Makefile +++ b/arch/arm/Makefile @@ -97,6 +97,7 @@ textaddr-$(CONFIG_ARCH_FORTUNET) := 0xc0008000 machine-$(CONFIG_ARCH_VERSATILE) := versatile machine-$(CONFIG_ARCH_IMX) := imx machine-$(CONFIG_ARCH_H720X) := h720x + machine-$(CONFIG_ARCH_AAEC2000) := aaec2000 ifeq ($(CONFIG_ARCH_EBSA110),y) # This is what happens if you forget the IOCS16 line. diff --git a/arch/arm/mach-aaec2000/Kconfig b/arch/arm/mach-aaec2000/Kconfig new file mode 100644 index 000000000000..5e4bef93754c --- /dev/null +++ b/arch/arm/mach-aaec2000/Kconfig @@ -0,0 +1,11 @@ +if ARCH_AAEC2000 + +menu "Agilent AAEC-2000 Implementations" + +config MACH_AAED2000 + bool "Agilent AAED-2000 Development Platform" + select CPU_ARM920T + +endmenu + +endif diff --git a/arch/arm/mach-aaec2000/Makefile b/arch/arm/mach-aaec2000/Makefile new file mode 100644 index 000000000000..20ec83896c37 --- /dev/null +++ b/arch/arm/mach-aaec2000/Makefile @@ -0,0 +1,9 @@ +# +# Makefile for the linux kernel. +# + +# Common support (must be linked before board specific support) +obj-y += core.o + +# Specific board support +obj-$(CONFIG_MACH_AAED2000) += aaed2000.o diff --git a/arch/arm/mach-aaec2000/aaed2000.c b/arch/arm/mach-aaec2000/aaed2000.c new file mode 100644 index 000000000000..5417ca3f4621 --- /dev/null +++ b/arch/arm/mach-aaec2000/aaed2000.c @@ -0,0 +1,48 @@ +/* + * linux/arch/arm/mach-aaec2000/aaed2000.c + * + * Support for the Agilent AAED-2000 Development Platform. + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "core.h" + +static void __init aaed2000_init_irq(void) +{ + aaec2000_init_irq(); +} + +static void __init aaed2000_map_io(void) +{ + aaec2000_map_io(); +} + +MACHINE_START(AAED2000, "Agilent AAED-2000 Development Platform") + MAINTAINER("Nicolas Bellido Y Ortega") + BOOT_MEM(0xf0000000, PIO_BASE, VIO_BASE) + MAPIO(aaed2000_map_io) + INITIRQ(aaed2000_init_irq) + .timer = &aaec2000_timer, +MACHINE_END diff --git a/arch/arm/mach-aaec2000/core.c b/arch/arm/mach-aaec2000/core.c new file mode 100644 index 000000000000..fc145b3768fa --- /dev/null +++ b/arch/arm/mach-aaec2000/core.c @@ -0,0 +1,157 @@ +/* + * linux/arch/arm/mach-aaec2000/core.c + * + * Code common to all AAEC-2000 machines + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#include +#include +#include + +/* + * Common I/O mapping: + * + * Static virtual address mappings are as follow: + * + * 0xf8000000-0xf8001ffff: Devices connected to APB bus + * 0xf8002000-0xf8003ffff: Devices connected to AHB bus + * + * Below 0xe8000000 is reserved for vm allocation. + * + * The machine specific code must provide the extra mapping beside the + * default mapping provided here. + */ +static struct map_desc standard_io_desc[] __initdata = { + /* virtual physical length type */ + { VIO_APB_BASE, PIO_APB_BASE, IO_APB_LENGTH, MT_DEVICE }, + { VIO_AHB_BASE, PIO_AHB_BASE, IO_AHB_LENGTH, MT_DEVICE } +}; + +void __init aaec2000_map_io(void) +{ + iotable_init(standard_io_desc, ARRAY_SIZE(standard_io_desc)); +} + +/* + * Interrupt handling routines + */ +static void aaec2000_int_ack(unsigned int irq) +{ + IRQ_INTSR = 1 << irq; +} + +static void aaec2000_int_mask(unsigned int irq) +{ + IRQ_INTENC |= (1 << irq); +} + +static void aaec2000_int_unmask(unsigned int irq) +{ + IRQ_INTENS |= (1 << irq); +} + +static struct irqchip aaec2000_irq_chip = { + .ack = aaec2000_int_ack, + .mask = aaec2000_int_mask, + .unmask = aaec2000_int_unmask, +}; + +void __init aaec2000_init_irq(void) +{ + unsigned int i; + + for (i = 0; i < NR_IRQS; i++) { + set_irq_handler(i, do_level_IRQ); + set_irq_chip(i, &aaec2000_irq_chip); + set_irq_flags(i, IRQF_VALID); + } + + /* Disable all interrupts */ + IRQ_INTENC = 0xffffffff; + + /* Clear any pending interrupts */ + IRQ_INTSR = IRQ_INTSR; +} + +/* + * Time keeping + */ +/* IRQs are disabled before entering here from do_gettimeofday() */ +static unsigned long aaec2000_gettimeoffset(void) +{ + unsigned long ticks_to_match, elapsed, usec; + + /* Get ticks before next timer match */ + ticks_to_match = TIMER1_LOAD - TIMER1_VAL; + + /* We need elapsed ticks since last match */ + elapsed = LATCH - ticks_to_match; + + /* Now, convert them to usec */ + usec = (unsigned long)(elapsed * (tick_nsec / 1000))/LATCH; + + return usec; +} + +/* We enter here with IRQs enabled */ +static irqreturn_t +aaec2000_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs) +{ + /* TODO: Check timer accuracy */ + write_seqlock(&xtime_lock); + + timer_tick(regs); + TIMER1_CLEAR = 1; + + write_sequnlock(&xtime_lock); + + return IRQ_HANDLED; +} + +static struct irqaction aaec2000_timer_irq = { + .name = "AAEC-2000 Timer Tick", + .flags = SA_INTERRUPT, + .handler = aaec2000_timer_interrupt +}; + +static void __init aaec2000_timer_init(void) +{ + /* Disable timer 1 */ + TIMER1_CTRL = 0; + + /* We have somehow to generate a 100Hz clock. + * We then use the 508KHz timer in periodic mode. + */ + TIMER1_LOAD = LATCH; + TIMER1_CLEAR = 1; /* Clear interrupt */ + + setup_irq(INT_TMR1_OFL, &aaec2000_timer_irq); + + TIMER1_CTRL = TIMER_CTRL_ENABLE | + TIMER_CTRL_PERIODIC | + TIMER_CTRL_CLKSEL_508K; +} + +struct sys_timer aaec2000_timer = { + .init = aaec2000_timer_init, + .offset = aaec2000_gettimeoffset, +}; + diff --git a/arch/arm/mach-aaec2000/core.h b/arch/arm/mach-aaec2000/core.h new file mode 100644 index 000000000000..91893d848c16 --- /dev/null +++ b/arch/arm/mach-aaec2000/core.h @@ -0,0 +1,16 @@ +/* + * linux/arch/arm/mach-aaec2000/core.h + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + +struct sys_timer; + +extern struct sys_timer aaec2000_timer; +extern void __init aaec2000_map_io(void); +extern void __init aaec2000_init_irq(void); diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig index 3fefb43c67f7..95606b4a3ba6 100644 --- a/arch/arm/mm/Kconfig +++ b/arch/arm/mm/Kconfig @@ -62,7 +62,7 @@ config CPU_ARM720T # ARM920T config CPU_ARM920T bool "Support ARM920T processor" if !ARCH_S3C2410 - depends on ARCH_INTEGRATOR || ARCH_S3C2410 || ARCH_IMX + depends on ARCH_INTEGRATOR || ARCH_S3C2410 || ARCH_IMX || ARCH_AAEC2000 default y if ARCH_S3C2410 select CPU_32v4 select CPU_ABRT_EV4T diff --git a/include/asm-arm/arch-aaec2000/aaec2000.h b/include/asm-arm/arch-aaec2000/aaec2000.h new file mode 100644 index 000000000000..0e9b7e18af05 --- /dev/null +++ b/include/asm-arm/arch-aaec2000/aaec2000.h @@ -0,0 +1,151 @@ +/* + * linux/include/asm-arm/arch-aaec2000/aaec2000.h + * + * AAEC-2000 registers definition + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ASM_ARCH_AAEC2000_H +#define __ASM_ARCH_AAEC2000_H + +#ifndef __ASM_ARCH_HARDWARE_H +#error You must include hardware.h not this file +#endif /* __ASM_ARCH_HARDWARE_H */ + +/* Interrupt controller */ +#define IRQ_BASE __REG(0x80000500) +#define IRQ_INTSR __REG(0x80000500) /* Int Status Register */ +#define IRQ_INTRSR __REG(0x80000504) /* Int Raw (unmasked) Status */ +#define IRQ_INTENS __REG(0x80000508) /* Int Enable Set */ +#define IRQ_INTENC __REG(0x8000050c) /* Int Enable Clear */ + +/* UART 1 */ +#define UART1_BASE __REG(0x80000600) +#define UART1_DR __REG(0x80000600) /* Data/FIFO Register */ +#define UART1_LCR __REG(0x80000604) /* Link Control Register */ +#define UART1_BRCR __REG(0x80000608) /* Baud Rate Control Register */ +#define UART1_CR __REG(0x8000060c) /* Control Register */ +#define UART1_SR __REG(0x80000610) /* Status Register */ +#define UART1_INT __REG(0x80000614) /* Interrupt Status Register */ +#define UART1_INTM __REG(0x80000618) /* Interrupt Mask Register */ +#define UART1_INTRES __REG(0x8000061c) /* Int Result (masked status) Register */ + +/* UART 2 */ +#define UART2_BASE __REG(0x80000700) +#define UART2_DR __REG(0x80000700) /* Data/FIFO Register */ +#define UART2_LCR __REG(0x80000704) /* Link Control Register */ +#define UART2_BRCR __REG(0x80000708) /* Baud Rate Control Register */ +#define UART2_CR __REG(0x8000070c) /* Control Register */ +#define UART2_SR __REG(0x80000710) /* Status Register */ +#define UART2_INT __REG(0x80000714) /* Interrupt Status Register */ +#define UART2_INTM __REG(0x80000718) /* Interrupt Mask Register */ +#define UART2_INTRES __REG(0x8000071c) /* Int Result (masked status) Register */ + +/* UART 3 */ +#define UART3_BASE __REG(0x80000800) +#define UART3_DR __REG(0x80000800) /* Data/FIFO Register */ +#define UART3_LCR __REG(0x80000804) /* Link Control Register */ +#define UART3_BRCR __REG(0x80000808) /* Baud Rate Control Register */ +#define UART3_CR __REG(0x8000080c) /* Control Register */ +#define UART3_SR __REG(0x80000810) /* Status Register */ +#define UART3_INT __REG(0x80000814) /* Interrupt Status Register */ +#define UART3_INTM __REG(0x80000818) /* Interrupt Mask Register */ +#define UART3_INTRES __REG(0x8000081c) /* Int Result (masked status) Register */ + +/* These are used in some places */ +#define _UART1_BASE __PREG(UART1_BASE) +#define _UART2_BASE __PREG(UART2_BASE) +#define _UART3_BASE __PREG(UART3_BASE) + +/* UART Registers Offsets */ +#define UART_DR 0x00 +#define UART_LCR 0x04 +#define UART_BRCR 0x08 +#define UART_CR 0x0c +#define UART_SR 0x10 +#define UART_INT 0x14 +#define UART_INTM 0x18 +#define UART_INTRES 0x1c + +/* UART_LCR Bitmask */ +#define UART_LCR_BRK (1 << 0) /* Send Break */ +#define UART_LCR_PEN (1 << 1) /* Parity Enable */ +#define UART_LCR_EP (1 << 2) /* Even/Odd Parity */ +#define UART_LCR_S2 (1 << 3) /* One/Two Stop bits */ +#define UART_LCR_FIFO (1 << 4) /* FIFO Enable */ +#define UART_LCR_WL5 (0 << 5) /* Word Length - 5 bits */ +#define UART_LCR_WL6 (1 << 5) /* Word Length - 6 bits */ +#define UART_LCR_WL7 (1 << 6) /* Word Length - 7 bits */ +#define UART_LCR_WL8 (1 << 7) /* Word Length - 8 bits */ + +/* UART_CR Bitmask */ +#define UART_CR_EN (1 << 0) /* UART Enable */ +#define UART_CR_SIR (1 << 1) /* IrDA SIR Enable */ +#define UART_CR_SIRLP (1 << 2) /* Low Power IrDA Enable */ +#define UART_CR_RXP (1 << 3) /* Receive Pin Polarity */ +#define UART_CR_TXP (1 << 4) /* Transmit Pin Polarity */ +#define UART_CR_MXP (1 << 5) /* Modem Pin Polarity */ +#define UART_CR_LOOP (1 << 6) /* Loopback Mode */ + +/* UART_SR Bitmask */ +#define UART_SR_CTS (1 << 0) /* Clear To Send Status */ +#define UART_SR_DSR (1 << 1) /* Data Set Ready Status */ +#define UART_SR_DCD (1 << 2) /* Data Carrier Detect Status */ +#define UART_SR_TxBSY (1 << 3) /* Transmitter Busy Status */ +#define UART_SR_RxFE (1 << 4) /* Receive FIFO Empty Status */ +#define UART_SR_TxFF (1 << 5) /* Transmit FIFO Full Status */ +#define UART_SR_RxFF (1 << 6) /* Receive FIFO Full Status */ +#define UART_SR_TxFE (1 << 7) /* Transmit FIFO Empty Status */ + +/* UART_INT Bitmask */ +#define UART_INT_RIS (1 << 0) /* Rx Interrupt */ +#define UART_INT_TIS (1 << 1) /* Tx Interrupt */ +#define UART_INT_MIS (1 << 2) /* Modem Interrupt */ +#define UART_INT_RTIS (1 << 3) /* Receive Timeout Interrupt */ + +/* Timer 1 */ +#define TIMER1_BASE __REG(0x80000c00) +#define TIMER1_LOAD __REG(0x80000c00) /* Timer 1 Load Register */ +#define TIMER1_VAL __REG(0x80000c04) /* Timer 1 Value Register */ +#define TIMER1_CTRL __REG(0x80000c08) /* Timer 1 Control Register */ +#define TIMER1_CLEAR __REG(0x80000c0c) /* Timer 1 Clear Register */ + +/* Timer 2 */ +#define TIMER2_BASE __REG(0x80000d00) +#define TIMER2_LOAD __REG(0x80000d00) /* Timer 2 Load Register */ +#define TIMER2_VAL __REG(0x80000d04) /* Timer 2 Value Register */ +#define TIMER2_CTRL __REG(0x80000d08) /* Timer 2 Control Register */ +#define TIMER2_CLEAR __REG(0x80000d0c) /* Timer 2 Clear Register */ + +/* Timer 3 */ +#define TIMER3_BASE __REG(0x80000e00) +#define TIMER3_LOAD __REG(0x80000e00) /* Timer 3 Load Register */ +#define TIMER3_VAL __REG(0x80000e04) /* Timer 3 Value Register */ +#define TIMER3_CTRL __REG(0x80000e08) /* Timer 3 Control Register */ +#define TIMER3_CLEAR __REG(0x80000e0c) /* Timer 3 Clear Register */ + +/* Timer Control register bits */ +#define TIMER_CTRL_ENABLE (1 << 7) /* Enable (Start° Timer */ +#define TIMER_CTRL_PERIODIC (1 << 6) /* Periodic Running Mode */ +#define TIMER_CTRL_FREE_RUNNING (0 << 6) /* Normal Running Mode */ +#define TIMER_CTRL_CLKSEL_508K (1 << 3) /* 508KHz Clock select (Timer 1, 2) */ +#define TIMER_CTRL_CLKSEL_2K (0 << 3) /* 2KHz Clock Select (Timer 1, 2)*/ + +/* Power and State Control */ +#define POWER_BASE __REG(0x80000400) +#define POWER_PWRSR __REG(0x80000400) /* Power Status Register */ +#define POWER_PWRCNT __REG(0x80000404) /* Power/Clock control */ +#define POWER_HALT __REG(0x80000408) /* Power Idle Mode */ +#define POWER_STDBY __REG(0x8000040c) /* Power Standby Mode */ +#define POWER_BLEOI __REG(0x80000410) /* Battery Low End of Interrupt */ +#define POWER_MCEOI __REG(0x80000414) /* Media Changed EoI */ +#define POWER_TEOI __REG(0x80000418) /* Tick EoI */ +#define POWER_STFCLR __REG(0x8000041c) /* NbFlg, RSTFlg, PFFlg, CLDFlg Clear */ +#define POWER_CLKSET __REG(0x80000420) /* Clock Speed Control */ + +#endif /* __ARM_ARCH_AAEC2000_H */ diff --git a/include/asm-arm/arch-aaec2000/debug-macro.S b/include/asm-arm/arch-aaec2000/debug-macro.S new file mode 100644 index 000000000000..e4f1fa539a74 --- /dev/null +++ b/include/asm-arm/arch-aaec2000/debug-macro.S @@ -0,0 +1,36 @@ +/* linux/include/asm-arm/arch-aaec2000/debug-macro.S + * + * Debugging macro include header + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + + .macro addruart,rx + mrc p15, 0, \rx, c1, c0 + tst \rx, #1 @ MMU enabled? + moveq \rx, #0x80000000 @ physical + movne \rx, #io_p2v(0x80000000) @ virtual + orr \rx, \rx, #0x00000800 + .endm + + .macro senduart,rd,rx + str \rd, [\rx, #0] + .endm + + .macro busyuart,rd,rx +1002: ldr \rd, [\rx, #0x10] + tst \rd, #(1 << 7) + beq 1002b + .endm + + .macro waituart,rd,rx +#if 0 +1001: ldr \rd, [\rx, #0x10] + tst \rd, #(1 << 5) + beq 1001b +#endif + .endm diff --git a/include/asm-arm/arch-aaec2000/dma.h b/include/asm-arm/arch-aaec2000/dma.h new file mode 100644 index 000000000000..28c890b4a1d3 --- /dev/null +++ b/include/asm-arm/arch-aaec2000/dma.h @@ -0,0 +1,17 @@ +/* + * linux/include/asm-arm/arch-aaec2000/dma.h + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ASM_ARCH_DMA_H +#define __ASM_ARCH_DMA_H + +#define MAX_DMA_ADDRESS 0xffffffff +#define MAX_DMA_CHANNELS 0 + +#endif diff --git a/include/asm-arm/arch-aaec2000/entry-macro.S b/include/asm-arm/arch-aaec2000/entry-macro.S new file mode 100644 index 000000000000..df31313ab07e --- /dev/null +++ b/include/asm-arm/arch-aaec2000/entry-macro.S @@ -0,0 +1,33 @@ +/* + * linux/include/asm-arm/arch-aaec2000/entry-macro.S + * + * Low-level IRQ helper for aaec-2000 based platforms + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + + .macro disable_fiq + .endm + + .macro get_irqnr_and_base, irqnr, irqstat, base, tmp + mov r4, #0xf8000000 + add r4, r4, #0x00000500 + mov \base, r4 + ldr \irqstat, [\base, #0] + cmp \irqstat, #0 + bne 1001f + ldr \irqnr, =NR_IRQS+1 + b 1003f +1001: mov \irqnr, #0 +1002: ands \tmp, \irqstat, #1 + mov \irqstat, \irqstat, LSR #1 + add \irqnr, \irqnr, #1 + beq 1002b + sub \irqnr, \irqnr, #1 +1003: + .endm diff --git a/include/asm-arm/arch-aaec2000/hardware.h b/include/asm-arm/arch-aaec2000/hardware.h new file mode 100644 index 000000000000..4c37219e030e --- /dev/null +++ b/include/asm-arm/arch-aaec2000/hardware.h @@ -0,0 +1,49 @@ +/* + * linux/include/asm-arm/arch-aaec2000/hardware.h + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ASM_ARCH_HARDWARE_H +#define __ASM_ARCH_HARDWARE_H + +#include + +/* The kernel is loaded at physical address 0xf8000000. + * We map the IO space a bit after + */ +#define PIO_APB_BASE 0x80000000 +#define VIO_APB_BASE 0xf8000000 +#define IO_APB_LENGTH 0x2000 +#define PIO_AHB_BASE 0x80002000 +#define VIO_AHB_BASE 0xf8002000 +#define IO_AHB_LENGTH 0x2000 + +#define VIO_BASE VIO_APB_BASE +#define PIO_BASE PIO_APB_BASE + +#define io_p2v(x) ( (x) - PIO_BASE + VIO_BASE ) +#define io_v2p(x) ( (x) + PIO_BASE - VIO_BASE ) + +#ifndef __ASSEMBLY__ + +#include + +/* FIXME: Is it needed to optimize this a la pxa ?? */ +#define __REG(x) (*((volatile u32 *)io_p2v(x))) +#define __PREG(x) (io_v2p((u32)&(x))) + +#else /* __ASSEMBLY__ */ + +#define __REG(x) io_p2v(x) +#define __PREG(x) io_v2p(x) + +#endif + +#include "aaec2000.h" + +#endif /* __ASM_ARCH_HARDWARE_H */ diff --git a/include/asm-arm/arch-aaec2000/io.h b/include/asm-arm/arch-aaec2000/io.h new file mode 100644 index 000000000000..c58a8d10425a --- /dev/null +++ b/include/asm-arm/arch-aaec2000/io.h @@ -0,0 +1,19 @@ +/* + * linux/include/asm-arm/arch-aaec2000/io.h + * + * Copied from asm/arch/sa1100/io.h + */ +#ifndef __ASM_ARM_ARCH_IO_H +#define __ASM_ARM_ARCH_IO_H + +#define IO_SPACE_LIMIT 0xffffffff + +/* + * We don't actually have real ISA nor PCI buses, but there is so many + * drivers out there that might just work if we fake them... + */ +#define __io(a) ((void __iomem *)(a)) +#define __mem_pci(a) (a) +#define __mem_isa(a) (a) + +#endif diff --git a/include/asm-arm/arch-aaec2000/irqs.h b/include/asm-arm/arch-aaec2000/irqs.h new file mode 100644 index 000000000000..de252220e806 --- /dev/null +++ b/include/asm-arm/arch-aaec2000/irqs.h @@ -0,0 +1,46 @@ +/* + * linux/include/asm-arm/arch-aaec2000/irqs.h + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ASM_ARCH_IRQS_H +#define __ASM_ARCH_IRQS_H + + +#define INT_GPIOF0_FIQ 0 /* External GPIO Port F O Fast Interrupt Input */ +#define INT_BL_FIQ 1 /* Battery Low Fast Interrupt */ +#define INT_WE_FIQ 2 /* Watchdog Expired Fast Interrupt */ +#define INT_MV_FIQ 3 /* Media Changed Interrupt */ +#define INT_SC 4 /* Sound Codec Interrupt */ +#define INT_GPIO1 5 /* GPIO Port F Configurable Int 1 */ +#define INT_GPIO2 6 /* GPIO Port F Configurable Int 2 */ +#define INT_GPIO3 7 /* GPIO Port F Configurable Int 3 */ +#define INT_TMR1_OFL 8 /* Timer 1 Overflow Interrupt */ +#define INT_TMR2_OFL 9 /* Timer 2 Overflow Interrupt */ +#define INT_RTC_CM 10 /* RTC Compare Match Interrupt */ +#define INT_TICK 11 /* 64Hz Tick Interrupt */ +#define INT_UART1 12 /* UART1 Interrupt */ +#define INT_UART2 13 /* UART2 & Modem State Changed Interrupt */ +#define INT_LCD 14 /* LCD Interrupt */ +#define INT_SSI 15 /* SSI End of Transfer Interrupt */ +#define INT_UART3 16 /* UART3 Interrupt */ +#define INT_SCI 17 /* SCI Interrupt */ +#define INT_AAC 18 /* Advanced Audio Codec Interrupt */ +#define INT_MMC 19 /* MMC Interrupt */ +#define INT_USB 20 /* USB Interrupt */ +#define INT_DMA 21 /* DMA Interrupt */ +#define INT_TMR3_UOFL 22 /* Timer 3 Underflow Interrupt */ +#define INT_GPIO4 23 /* GPIO Port F Configurable Int 4 */ +#define INT_GPIO5 24 /* GPIO Port F Configurable Int 4 */ +#define INT_GPIO6 25 /* GPIO Port F Configurable Int 4 */ +#define INT_GPIO7 26 /* GPIO Port F Configurable Int 4 */ +#define INT_BMI 27 /* BMI Interrupt */ + +#define NR_IRQS (INT_BMI + 1) + +#endif /* __ASM_ARCH_IRQS_H */ diff --git a/include/asm-arm/arch-aaec2000/memory.h b/include/asm-arm/arch-aaec2000/memory.h new file mode 100644 index 000000000000..681b6a6171a1 --- /dev/null +++ b/include/asm-arm/arch-aaec2000/memory.h @@ -0,0 +1,73 @@ +/* + * linux/include/asm-arm/arch-aaec2000/memory.h + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ASM_ARCH_MEMORY_H +#define __ASM_ARCH_MEMORY_H + +#include + +#define PHYS_OFFSET (0xf0000000UL) + +#define __virt_to_bus(x) __virt_to_phys(x) +#define __bus_to_virt(x) __phys_to_virt(x) + +#ifdef CONFIG_DISCONTIGMEM + +/* + * The nodes are the followings: + * + * node 0: 0xf000.0000 - 0xf3ff.ffff + * node 1: 0xf400.0000 - 0xf7ff.ffff + * node 2: 0xf800.0000 - 0xfbff.ffff + * node 3: 0xfc00.0000 - 0xffff.ffff + */ + +/* + * Given a kernel address, find the home node of the underlying memory. + */ +#define KVADDR_TO_NID(addr) \ + (((unsigned long)(addr) - PAGE_OFFSET) >> NODE_MAX_MEM_SHIFT) + +/* + * Given a page frame number, convert it to a node id. + */ +#define PFN_TO_NID(pfn) \ + (((pfn) - PHYS_PFN_OFFSET) >> (NODE_MAX_MEM_SHIFT - PAGE_SHIFT)) + +/* + * Given a kaddr, ADDR_TO_MAPBASE finds the owning node of the memory + * and return the mem_map of that node. + */ +#define ADDR_TO_MAPBASE(kaddr) NODE_MEM_MAP(KVADDR_TO_NID(kaddr)) + +/* + * Given a page frame number, find the owning node of the memory + * and return the mem_map of that node. + */ +#define PFN_TO_MAPBASE(pfn) NODE_MEM_MAP(PFN_TO_NID(pfn)) + +/* + * Given a kaddr, LOCAL_MEM_MAP finds the owning node of the memory + * and returns the index corresponding to the appropriate page in the + * node's mem_map. + */ +#define LOCAL_MAP_NR(addr) \ + (((unsigned long)(addr) & (NODE_MAX_MEM_SIZE - 1)) >> PAGE_SHIFT) + +#define NODE_MAX_MEM_SHIFT 26 +#define NODE_MAX_MEM_SIZE (1 << NODE_MAX_MEM_SHIFT) + +#else + +#define PFN_TO_NID(addr) (0) + +#endif /* CONFIG_DISCONTIGMEM */ + +#endif /* __ASM_ARCH_MEMORY_H */ diff --git a/include/asm-arm/arch-aaec2000/param.h b/include/asm-arm/arch-aaec2000/param.h new file mode 100644 index 000000000000..139936c2faf2 --- /dev/null +++ b/include/asm-arm/arch-aaec2000/param.h @@ -0,0 +1,15 @@ +/* + * linux/include/asm-arm/arch-aaec2000/param.h + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ASM_ARCH_PARAM_H +#define __ASM_ARCH_PARAM_H + +#endif /* __ASM_ARCH_PARAM_H */ + diff --git a/include/asm-arm/arch-aaec2000/system.h b/include/asm-arm/arch-aaec2000/system.h new file mode 100644 index 000000000000..08de97b407a8 --- /dev/null +++ b/include/asm-arm/arch-aaec2000/system.h @@ -0,0 +1,24 @@ +/* + * linux/include/asm-arm/arch-aaed2000/system.h + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ASM_ARCH_SYSTEM_H +#define __ASM_ARCH_SYSTEM_H + +static inline void arch_idle(void) +{ + cpu_do_idle(); +} + +static inline void arch_reset(char mode) +{ + cpu_reset(0); +} + +#endif /* __ASM_ARCH_SYSTEM_H */ diff --git a/include/asm-arm/arch-aaec2000/timex.h b/include/asm-arm/arch-aaec2000/timex.h new file mode 100644 index 000000000000..f5708b38fb7f --- /dev/null +++ b/include/asm-arm/arch-aaec2000/timex.h @@ -0,0 +1,18 @@ +/* + * linux/include/asm-arm/arch-aaec2000/timex.h + * + * AAEC-2000 Architecture timex specification + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ASM_ARCH_TIMEX_H +#define __ASM_ARCH_TIMEX_H + +#define CLOCK_TICK_RATE 508000 + +#endif /* __ASM_ARCH_TIMEX_H */ diff --git a/include/asm-arm/arch-aaec2000/uncompress.h b/include/asm-arm/arch-aaec2000/uncompress.h new file mode 100644 index 000000000000..fff0c94b75c4 --- /dev/null +++ b/include/asm-arm/arch-aaec2000/uncompress.h @@ -0,0 +1,47 @@ +/* + * linux/include/asm-arm/arch-aaec2000/uncompress.h + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ASM_ARCH_UNCOMPRESS_H +#define __ASM_ARCH_UNCOMPRESS_H + +#include "hardware.h" + +#define UART(x) (*(volatile unsigned long *)(serial_port + (x))) + +static void putstr( const char *s ) +{ + unsigned long serial_port; + do { + serial_port = _UART3_BASE; + if (UART(UART_CR) & UART_CR_EN) break; + serial_port = _UART1_BASE; + if (UART(UART_CR) & UART_CR_EN) break; + serial_port = _UART2_BASE; + if (UART(UART_CR) & UART_CR_EN) break; + return; + } while (0); + + for (; *s; s++) { + /* wait for space in the UART's transmitter */ + while ((UART(UART_SR) & UART_SR_TxFF)); + /* send the character out. */ + UART(UART_DR) = *s; + /* if a LF, also do CR... */ + if (*s == 10) { + while ((UART(UART_SR) & UART_SR_TxFF)); + UART(UART_DR) = 13; + } + } +} + +#define arch_decomp_setup() +#define arch_decomp_wdog() + +#endif /* __ASM_ARCH_UNCOMPRESS_H */ diff --git a/include/asm-arm/arch-aaec2000/vmalloc.h b/include/asm-arm/arch-aaec2000/vmalloc.h new file mode 100644 index 000000000000..ecb991e2e4ff --- /dev/null +++ b/include/asm-arm/arch-aaec2000/vmalloc.h @@ -0,0 +1,16 @@ +/* + * linux/include/asm-arm/arch-aaec2000/vmalloc.h + * + * Copyright (c) 2005 Nicolas Bellido Y Ortega + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#ifndef __ASM_ARCH_VMALLOC_H +#define __ASM_ARCH_VMALLOC_H + +#define VMALLOC_END (PAGE_OFFSET + 0x10000000) + +#endif /* __ASM_ARCH_VMALLOC_H */ -- cgit v1.2.3 From c0da085ad2e6b1419b8a7439538f7f15eb5c4777 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Mon, 20 Jun 2005 18:51:06 +0100 Subject: [PATCH] ARM: 2693/1: Add PCI support for Versatile/PB Patch from Catalin Marinas This patch adds PCI support for the Versatile PB926 platform. Signed-off-by: Colin King Signed-off-by: Catalin Marinas Signed-off-by: Russell King --- arch/arm/Kconfig | 2 +- arch/arm/mach-versatile/Makefile | 1 + arch/arm/mach-versatile/core.c | 14 +- arch/arm/mach-versatile/pci.c | 360 ++++++++++++++++++++++++++++++ include/asm-arm/arch-versatile/hardware.h | 27 ++- include/asm-arm/arch-versatile/io.h | 2 +- include/asm-arm/arch-versatile/platform.h | 17 +- 7 files changed, 404 insertions(+), 19 deletions(-) create mode 100644 arch/arm/mach-versatile/pci.c (limited to 'include') diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index efdb12d73566..ee8a9ad7bbd9 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -280,7 +280,7 @@ config ISA_DMA_API default y config PCI - bool "PCI support" if ARCH_INTEGRATOR_AP + bool "PCI support" if ARCH_INTEGRATOR_AP || ARCH_VERSATILE_PB help Find out whether you have a PCI motherboard. PCI is the name of a bus system, i.e. the way the CPU talks to the other stuff inside diff --git a/arch/arm/mach-versatile/Makefile b/arch/arm/mach-versatile/Makefile index 5d608837757a..ba81e70ed813 100644 --- a/arch/arm/mach-versatile/Makefile +++ b/arch/arm/mach-versatile/Makefile @@ -5,3 +5,4 @@ obj-y := core.o clock.o obj-$(CONFIG_ARCH_VERSATILE_PB) += versatile_pb.o obj-$(CONFIG_MACH_VERSATILE_AB) += versatile_ab.o +obj-$(CONFIG_PCI) += pci.o diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c index 302c2a7b9b63..6a7cbea5e098 100644 --- a/arch/arm/mach-versatile/core.c +++ b/arch/arm/mach-versatile/core.c @@ -196,11 +196,15 @@ static struct map_desc versatile_io_desc[] __initdata = { #ifdef CONFIG_DEBUG_LL { IO_ADDRESS(VERSATILE_UART0_BASE), VERSATILE_UART0_BASE, SZ_4K, MT_DEVICE }, #endif -#ifdef FIXME - { PCI_MEMORY_VADDR, PHYS_PCI_MEM_BASE, SZ_16M, MT_DEVICE }, - { PCI_CONFIG_VADDR, PHYS_PCI_CONFIG_BASE, SZ_16M, MT_DEVICE }, - { PCI_V3_VADDR, PHYS_PCI_V3_BASE, SZ_512K, MT_DEVICE }, - { PCI_IO_VADDR, PHYS_PCI_IO_BASE, SZ_64K, MT_DEVICE }, +#ifdef CONFIG_PCI + { IO_ADDRESS(VERSATILE_PCI_CORE_BASE), VERSATILE_PCI_CORE_BASE, SZ_4K, MT_DEVICE }, + { VERSATILE_PCI_VIRT_BASE, VERSATILE_PCI_BASE, VERSATILE_PCI_BASE_SIZE, MT_DEVICE }, + { VERSATILE_PCI_CFG_VIRT_BASE, VERSATILE_PCI_CFG_BASE, VERSATILE_PCI_CFG_BASE_SIZE, MT_DEVICE }, +#if 0 + { VERSATILE_PCI_VIRT_MEM_BASE0, VERSATILE_PCI_MEM_BASE0, SZ_16M, MT_DEVICE }, + { VERSATILE_PCI_VIRT_MEM_BASE1, VERSATILE_PCI_MEM_BASE1, SZ_16M, MT_DEVICE }, + { VERSATILE_PCI_VIRT_MEM_BASE2, VERSATILE_PCI_MEM_BASE2, SZ_16M, MT_DEVICE }, +#endif #endif }; diff --git a/arch/arm/mach-versatile/pci.c b/arch/arm/mach-versatile/pci.c new file mode 100644 index 000000000000..d1565e851f0e --- /dev/null +++ b/arch/arm/mach-versatile/pci.c @@ -0,0 +1,360 @@ +/* + * linux/arch/arm/mach-versatile/pci.c + * + * (C) Copyright Koninklijke Philips Electronics NV 2004. All rights reserved. + * You can redistribute and/or modify this software under the terms of version 2 + * of the GNU General Public License as published by the Free Software Foundation. + * THIS SOFTWARE IS PROVIDED "AS IS" WITHOUT ANY WARRANTY; WITHOUT EVEN THE IMPLIED + * WARRANTY OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * Koninklijke Philips Electronics nor its subsidiaries is obligated to provide any support for this software. + * + * ARM Versatile PCI driver. + * + * 14/04/2005 Initial version, colin.king@philips.com + * + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +/* + * these spaces are mapped using the following base registers: + * + * Usage Local Bus Memory Base/Map registers used + * + * Mem 50000000 - 5FFFFFFF LB_BASE0/LB_MAP0, non prefetch + * Mem 60000000 - 6FFFFFFF LB_BASE1/LB_MAP1, prefetch + * IO 44000000 - 4FFFFFFF LB_BASE2/LB_MAP2, IO + * Cfg 42000000 - 42FFFFFF PCI config + * + */ +#define SYS_PCICTL IO_ADDRESS(VERSATILE_SYS_PCICTL) +#define PCI_IMAP0 IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x0) +#define PCI_IMAP1 IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x4) +#define PCI_IMAP2 IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x8) +#define PCI_SMAP0 IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x10) +#define PCI_SMAP1 IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x14) +#define PCI_SMAP2 IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0x18) +#define PCI_SELFID IO_ADDRESS(VERSATILE_PCI_CORE_BASE+0xc) + +#define DEVICE_ID_OFFSET 0x00 +#define CSR_OFFSET 0x04 +#define CLASS_ID_OFFSET 0x08 + +#define VP_PCI_DEVICE_ID 0x030010ee +#define VP_PCI_CLASS_ID 0x0b400000 + +static unsigned long pci_slot_ignore = 0; + +static int __init versatile_pci_slot_ignore(char *str) +{ + int retval; + int slot; + + while ((retval = get_option(&str,&slot))) { + if ((slot < 0) || (slot > 31)) { + printk("Illegal slot value: %d\n",slot); + } else { + pci_slot_ignore |= (1 << slot); + } + } + return 1; +} + +__setup("pci_slot_ignore=", versatile_pci_slot_ignore); + + +static unsigned long __pci_addr(struct pci_bus *bus, + unsigned int devfn, int offset) +{ + unsigned int busnr = bus->number; + + /* + * Trap out illegal values + */ + if (offset > 255) + BUG(); + if (busnr > 255) + BUG(); + if (devfn > 255) + BUG(); + + return (VERSATILE_PCI_CFG_VIRT_BASE | (busnr << 16) | + (PCI_SLOT(devfn) << 11) | (PCI_FUNC(devfn) << 8) | offset); +} + +static int versatile_read_config(struct pci_bus *bus, unsigned int devfn, int where, + int size, u32 *val) +{ + unsigned long addr = __pci_addr(bus, devfn, where); + u32 v; + int slot = PCI_SLOT(devfn); + + if (pci_slot_ignore & (1 << slot)) { + /* Ignore this slot */ + switch (size) { + case 1: + v = 0xff; + break; + case 2: + v = 0xffff; + break; + default: + v = 0xffffffff; + } + } else { + switch (size) { + case 1: + addr &= ~3; + v = __raw_readb(addr); + break; + + case 2: + v = __raw_readl(addr & ~3); + if (addr & 2) v >>= 16; + v &= 0xffff; + break; + + default: + addr &= ~3; + v = __raw_readl(addr); + break; + } + } + + *val = v; + return PCIBIOS_SUCCESSFUL; +} + +static int versatile_write_config(struct pci_bus *bus, unsigned int devfn, int where, + int size, u32 val) +{ + unsigned long addr = __pci_addr(bus, devfn, where); + int slot = PCI_SLOT(devfn); + + if (pci_slot_ignore & (1 << slot)) { + return PCIBIOS_SUCCESSFUL; + } + + switch (size) { + case 1: + __raw_writeb((u8)val, addr); + break; + + case 2: + __raw_writew((u16)val, addr); + break; + + case 4: + __raw_writel(val, addr); + break; + } + + return PCIBIOS_SUCCESSFUL; +} + +static struct pci_ops pci_versatile_ops = { + .read = versatile_read_config, + .write = versatile_write_config, +}; + +static struct resource io_mem = { + .name = "PCI I/O space", + .start = VERSATILE_PCI_MEM_BASE0, + .end = VERSATILE_PCI_MEM_BASE0+VERSATILE_PCI_MEM_BASE0_SIZE-1, + .flags = IORESOURCE_IO, +}; + +static struct resource non_mem = { + .name = "PCI non-prefetchable", + .start = VERSATILE_PCI_MEM_BASE1, + .end = VERSATILE_PCI_MEM_BASE1+VERSATILE_PCI_MEM_BASE1_SIZE-1, + .flags = IORESOURCE_MEM, +}; + +static struct resource pre_mem = { + .name = "PCI prefetchable", + .start = VERSATILE_PCI_MEM_BASE2, + .end = VERSATILE_PCI_MEM_BASE2+VERSATILE_PCI_MEM_BASE2_SIZE-1, + .flags = IORESOURCE_MEM | IORESOURCE_PREFETCH, +}; + +static int __init pci_versatile_setup_resources(struct resource **resource) +{ + int ret = 0; + + ret = request_resource(&iomem_resource, &io_mem); + if (ret) { + printk(KERN_ERR "PCI: unable to allocate I/O " + "memory region (%d)\n", ret); + goto out; + } + ret = request_resource(&iomem_resource, &non_mem); + if (ret) { + printk(KERN_ERR "PCI: unable to allocate non-prefetchable " + "memory region (%d)\n", ret); + goto release_io_mem; + } + ret = request_resource(&iomem_resource, &pre_mem); + if (ret) { + printk(KERN_ERR "PCI: unable to allocate prefetchable " + "memory region (%d)\n", ret); + goto release_non_mem; + } + + /* + * bus->resource[0] is the IO resource for this bus + * bus->resource[1] is the mem resource for this bus + * bus->resource[2] is the prefetch mem resource for this bus + */ + resource[0] = &io_mem; + resource[1] = &non_mem; + resource[2] = &pre_mem; + + goto out; + + release_non_mem: + release_resource(&non_mem); + release_io_mem: + release_resource(&io_mem); + out: + return ret; +} + +int __init pci_versatile_setup(int nr, struct pci_sys_data *sys) +{ + int ret = 0; + int i; + int myslot = -1; + unsigned long val; + + if (nr == 0) { + sys->mem_offset = 0; + ret = pci_versatile_setup_resources(sys->resource); + if (ret < 0) { + printk("pci_versatile_setup: resources... oops?\n"); + goto out; + } + } else { + printk("pci_versatile_setup: resources... nr == 0??\n"); + goto out; + } + + __raw_writel(VERSATILE_PCI_MEM_BASE0 >> 28,PCI_IMAP0); + __raw_writel(VERSATILE_PCI_MEM_BASE1 >> 28,PCI_IMAP1); + __raw_writel(VERSATILE_PCI_MEM_BASE2 >> 28,PCI_IMAP2); + + __raw_writel(1, SYS_PCICTL); + + val = __raw_readl(SYS_PCICTL); + if (!(val & 1)) { + printk("Not plugged into PCI backplane!\n"); + ret = -EIO; + goto out; + } + + /* + * We need to discover the PCI core first to configure itself + * before the main PCI probing is performed + */ + for (i=0; i<32; i++) { + if ((__raw_readl(VERSATILE_PCI_VIRT_BASE+(i<<11)+DEVICE_ID_OFFSET) == VP_PCI_DEVICE_ID) && + (__raw_readl(VERSATILE_PCI_VIRT_BASE+(i<<11)+CLASS_ID_OFFSET) == VP_PCI_CLASS_ID)) { + myslot = i; + + __raw_writel(myslot, PCI_SELFID); + val = __raw_readl(VERSATILE_PCI_CFG_VIRT_BASE+(myslot<<11)+CSR_OFFSET); + val |= (1<<2); + __raw_writel(val, VERSATILE_PCI_CFG_VIRT_BASE+(myslot<<11)+CSR_OFFSET); + break; + } + } + + if (myslot == -1) { + printk("Cannot find PCI core!\n"); + ret = -EIO; + } else { + printk("PCI core found (slot %d)\n",myslot); + /* Do not to map Versatile FPGA PCI device + into memory space as we are short of + mappable memory */ + pci_slot_ignore |= (1 << myslot); + ret = 1; + } + + out: + return ret; +} + + +struct pci_bus *pci_versatile_scan_bus(int nr, struct pci_sys_data *sys) +{ + return pci_scan_bus(sys->busnr, &pci_versatile_ops, sys); +} + +/* + * V3_LB_BASE? - local bus address + * V3_LB_MAP? - pci bus address + */ +void __init pci_versatile_preinit(void) +{ +} + +void __init pci_versatile_postinit(void) +{ +} + + +/* + * map the specified device/slot/pin to an IRQ. Different backplanes may need to modify this. + */ +static int __init versatile_map_irq(struct pci_dev *dev, u8 slot, u8 pin) +{ + int irq; + int devslot = PCI_SLOT(dev->devfn); + + /* slot, pin, irq + 24 1 27 + 25 1 28 untested + 26 1 29 + 27 1 30 untested + */ + + irq = 27 + ((slot + pin + 2) % 3); /* Fudged */ + + printk("map irq: slot %d, pin %d, devslot %d, irq: %d\n",slot,pin,devslot,irq); + + return irq; +} + +static struct hw_pci versatile_pci __initdata = { + .swizzle = NULL, + .map_irq = versatile_map_irq, + .nr_controllers = 1, + .setup = pci_versatile_setup, + .scan = pci_versatile_scan_bus, + .preinit = pci_versatile_preinit, + .postinit = pci_versatile_postinit, +}; + +static int __init versatile_pci_init(void) +{ + pci_common_init(&versatile_pci); + return 0; +} + +subsys_initcall(versatile_pci_init); diff --git a/include/asm-arm/arch-versatile/hardware.h b/include/asm-arm/arch-versatile/hardware.h index d5fb4a251e7f..41c1bee342ad 100644 --- a/include/asm-arm/arch-versatile/hardware.h +++ b/include/asm-arm/arch-versatile/hardware.h @@ -25,19 +25,26 @@ #include #include -// FIXME = PCI settings need to be fixed!!!!! - /* - * Similar to above, but for PCI addresses (memory, IO, Config and the - * V3 chip itself). WARNING: this has to mirror definitions in platform.h + * PCI space virtual addresses */ -#define PCI_MEMORY_VADDR 0xe8000000 -#define PCI_CONFIG_VADDR 0xec000000 -#define PCI_V3_VADDR 0xed000000 -#define PCI_IO_VADDR 0xee000000 +#define VERSATILE_PCI_VIRT_BASE 0xe8000000 +#define VERSATILE_PCI_CFG_VIRT_BASE 0xe9000000 + +#if 0 +#define VERSATILE_PCI_VIRT_MEM_BASE0 0xf4000000 +#define VERSATILE_PCI_VIRT_MEM_BASE1 0xf5000000 +#define VERSATILE_PCI_VIRT_MEM_BASE2 0xf6000000 + +#define PCIO_BASE VERSATILE_PCI_VIRT_MEM_BASE0 +#define PCIMEM_BASE VERSATILE_PCI_VIRT_MEM_BASE1 +#endif + +/* CIK guesswork */ +#define PCIBIOS_MIN_IO 0x44000000 +#define PCIBIOS_MIN_MEM 0x50000000 -#define PCIO_BASE PCI_IO_VADDR -#define PCIMEM_BASE PCI_MEMORY_VADDR +#define pcibios_assign_all_busses() 1 /* macro to get at IO space when running virtually */ #define IO_ADDRESS(x) (((x) & 0x0fffffff) + (((x) >> 4) & 0x0f000000) + 0xf0000000) diff --git a/include/asm-arm/arch-versatile/io.h b/include/asm-arm/arch-versatile/io.h index dbb7158788fc..9f895bf61494 100644 --- a/include/asm-arm/arch-versatile/io.h +++ b/include/asm-arm/arch-versatile/io.h @@ -20,7 +20,7 @@ #ifndef __ASM_ARM_ARCH_IO_H #define __ASM_ARM_ARCH_IO_H -#define IO_SPACE_LIMIT 0xffff +#define IO_SPACE_LIMIT 0xffffffff #define __io(a) ((void __iomem *)(a)) #define __mem_pci(a) (a) diff --git a/include/asm-arm/arch-versatile/platform.h b/include/asm-arm/arch-versatile/platform.h index a71093e44c58..cbdd9fb96332 100644 --- a/include/asm-arm/arch-versatile/platform.h +++ b/include/asm-arm/arch-versatile/platform.h @@ -76,7 +76,7 @@ #define VERSATILE_SYS_NVFLAGSSET_OFFSET 0x38 #define VERSATILE_SYS_NVFLAGSCLR_OFFSET 0x3C #define VERSATILE_SYS_RESETCTL_OFFSET 0x40 -#define VERSATILE_SYS_PICCTL_OFFSET 0x44 +#define VERSATILE_SYS_PCICTL_OFFSET 0x44 #define VERSATILE_SYS_MCI_OFFSET 0x48 #define VERSATILE_SYS_FLASH_OFFSET 0x4C #define VERSATILE_SYS_CLCD_OFFSET 0x50 @@ -114,7 +114,7 @@ #define VERSATILE_SYS_NVFLAGSSET (VERSATILE_SYS_BASE + VERSATILE_SYS_NVFLAGSSET_OFFSET) #define VERSATILE_SYS_NVFLAGSCLR (VERSATILE_SYS_BASE + VERSATILE_SYS_NVFLAGSCLR_OFFSET) #define VERSATILE_SYS_RESETCTL (VERSATILE_SYS_BASE + VERSATILE_SYS_RESETCTL_OFFSET) -#define VERSATILE_SYS_PICCTL (VERSATILE_SYS_BASE + VERSATILE_SYS_PICCTL_OFFSET) +#define VERSATILE_SYS_PCICTL (VERSATILE_SYS_BASE + VERSATILE_SYS_PCICTL_OFFSET) #define VERSATILE_SYS_MCI (VERSATILE_SYS_BASE + VERSATILE_SYS_MCI_OFFSET) #define VERSATILE_SYS_FLASH (VERSATILE_SYS_BASE + VERSATILE_SYS_FLASH_OFFSET) #define VERSATILE_SYS_CLCD (VERSATILE_SYS_BASE + VERSATILE_SYS_CLCD_OFFSET) @@ -225,7 +225,20 @@ #define VERSATILE_SSMC_BASE 0x20000000 /* SSMC */ #define VERSATILE_IB2_BASE 0x24000000 /* IB2 module */ #define VERSATILE_MBX_BASE 0x40000000 /* MBX */ + +/* PCI space */ #define VERSATILE_PCI_BASE 0x41000000 /* PCI Interface */ +#define VERSATILE_PCI_CFG_BASE 0x42000000 +#define VERSATILE_PCI_MEM_BASE0 0x44000000 +#define VERSATILE_PCI_MEM_BASE1 0x50000000 +#define VERSATILE_PCI_MEM_BASE2 0x60000000 +/* Sizes of above maps */ +#define VERSATILE_PCI_BASE_SIZE 0x01000000 +#define VERSATILE_PCI_CFG_BASE_SIZE 0x02000000 +#define VERSATILE_PCI_MEM_BASE0_SIZE 0x0c000000 /* 32Mb */ +#define VERSATILE_PCI_MEM_BASE1_SIZE 0x10000000 /* 256Mb */ +#define VERSATILE_PCI_MEM_BASE2_SIZE 0x10000000 /* 256Mb */ + #define VERSATILE_SDRAM67_BASE 0x70000000 /* SDRAM banks 6 and 7 */ #define VERSATILE_LT_BASE 0x80000000 /* Logic Tile expansion */ -- cgit v1.2.3 From e4fe19819ef32950541503042f32e71b67edffc7 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Mon, 20 Jun 2005 18:51:07 +0100 Subject: [PATCH] ARM: 2701/1: free up ixp2000 timer 4 for the watchdog Patch from Lennert Buytenhek The IXP2000 has four timers, but if we're on an A-step IXP2800, timer 2 and 3 don't work. We need two timers for timekeeping (one for the timer interrupt and one for tracking missed jiffies), so on early IXP2800s we have no other choice but to use timer 1 and 4 for that, but on all other IXP2000s we'd rather leave timer 4 free since that's the only timer we can use for the watchdog. So, on buggy IXP2000s (i.e. the A-step IXP2800) we use timer 4 for tracking missed jiffies, and on all all non-buggy IXP2000s (i.e. everything but the A-step IXP2800) we use timer 2. On a pre-production IXP2800, this patch should print these messages on boot: Enabling IXP2800 erratum #25 workaround Unable to use IXP2000 watchdog due to IXP2800 erratum #25 On any non-buggy IXP2800 (as well as on IXP2400s) you shouldn't see anything at all, and the watchdog should be usable again. Signed-off-by: Lennert Buytenhek Signed-off-by: Deepak Saxena Signed-off-by: Russell King --- arch/arm/mach-ixp2000/core.c | 34 ++++++++++++++++++++++------- drivers/char/watchdog/ixp2000_wdt.c | 7 +++++- include/asm-arm/arch-ixp2000/ixp2000-regs.h | 1 + 3 files changed, 33 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/arch/arm/mach-ixp2000/core.c b/arch/arm/mach-ixp2000/core.c index 4f3c3d5c781c..fc0555596d6d 100644 --- a/arch/arm/mach-ixp2000/core.c +++ b/arch/arm/mach-ixp2000/core.c @@ -162,12 +162,13 @@ void __init ixp2000_map_io(void) static unsigned ticks_per_jiffy; static unsigned ticks_per_usec; static unsigned next_jiffy_time; +static volatile unsigned long *missing_jiffy_timer_csr; unsigned long ixp2000_gettimeoffset (void) { unsigned long offset; - offset = next_jiffy_time - *IXP2000_T4_CSR; + offset = next_jiffy_time - *missing_jiffy_timer_csr; return offset / ticks_per_usec; } @@ -179,7 +180,7 @@ static int ixp2000_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs) /* clear timer 1 */ ixp2000_reg_write(IXP2000_T1_CLR, 1); - while ((next_jiffy_time - *IXP2000_T4_CSR) > ticks_per_jiffy) { + while ((next_jiffy_time - *missing_jiffy_timer_csr) > ticks_per_jiffy) { timer_tick(regs); next_jiffy_time -= ticks_per_jiffy; } @@ -197,20 +198,37 @@ static struct irqaction ixp2000_timer_irq = { void __init ixp2000_init_time(unsigned long tick_rate) { - ixp2000_reg_write(IXP2000_T1_CLR, 0); - ixp2000_reg_write(IXP2000_T4_CLR, 0); - ticks_per_jiffy = (tick_rate + HZ/2) / HZ; ticks_per_usec = tick_rate / 1000000; + /* + * We use timer 1 as our timer interrupt. + */ + ixp2000_reg_write(IXP2000_T1_CLR, 0); ixp2000_reg_write(IXP2000_T1_CLD, ticks_per_jiffy - 1); ixp2000_reg_write(IXP2000_T1_CTL, (1 << 7)); /* - * We use T4 as a monotonic counter to track missed jiffies + * We use a second timer as a monotonic counter for tracking + * missed jiffies. The IXP2000 has four timers, but if we're + * on an A-step IXP2800, timer 2 and 3 don't work, so on those + * chips we use timer 4. Timer 4 is the only timer that can + * be used for the watchdog, so we use timer 2 if we're on a + * non-buggy chip. */ - ixp2000_reg_write(IXP2000_T4_CLD, -1); - ixp2000_reg_write(IXP2000_T4_CTL, (1 << 7)); + if ((*IXP2000_PRODUCT_ID & 0x001ffef0) == 0x00000000) { + printk(KERN_INFO "Enabling IXP2800 erratum #25 workaround\n"); + + ixp2000_reg_write(IXP2000_T4_CLR, 0); + ixp2000_reg_write(IXP2000_T4_CLD, -1); + ixp2000_reg_write(IXP2000_T4_CTL, (1 << 7)); + missing_jiffy_timer_csr = IXP2000_T4_CSR; + } else { + ixp2000_reg_write(IXP2000_T2_CLR, 0); + ixp2000_reg_write(IXP2000_T2_CLD, -1); + ixp2000_reg_write(IXP2000_T2_CTL, (1 << 7)); + missing_jiffy_timer_csr = IXP2000_T2_CSR; + } next_jiffy_time = 0xffffffff; /* register for interrupt */ diff --git a/drivers/char/watchdog/ixp2000_wdt.c b/drivers/char/watchdog/ixp2000_wdt.c index ab659d37b4d2..4e98c215e5b1 100644 --- a/drivers/char/watchdog/ixp2000_wdt.c +++ b/drivers/char/watchdog/ixp2000_wdt.c @@ -192,7 +192,12 @@ static struct miscdevice ixp2000_wdt_miscdev = static int __init ixp2000_wdt_init(void) { - wdt_tick_rate = (*IXP2000_T1_CLD * HZ)/ 256;; + if ((*IXP2000_PRODUCT_ID & 0x001ffef0) == 0x00000000) { + printk(KERN_INFO "Unable to use IXP2000 watchdog due to IXP2800 erratum #25.\n"); + return -EIO; + } + + wdt_tick_rate = (*IXP2000_T1_CLD * HZ) / 256; return misc_register(&ixp2000_wdt_miscdev); } diff --git a/include/asm-arm/arch-ixp2000/ixp2000-regs.h b/include/asm-arm/arch-ixp2000/ixp2000-regs.h index 6c56708d0ff0..a1d9e181b10f 100644 --- a/include/asm-arm/arch-ixp2000/ixp2000-regs.h +++ b/include/asm-arm/arch-ixp2000/ixp2000-regs.h @@ -363,6 +363,7 @@ #define IXP2000_MIN_REV_MASK 0x0000000F #define IXP2000_PROD_ID_MASK 0xFFFFFFFF +#define IXP2000_PRODUCT_ID GLOBAL_REG(0x00) #define IXP2000_MISC_CONTROL GLOBAL_REG(0x04) #define IXP2000_MSF_CLK_CNTRL GLOBAL_REG(0x08) #define IXP2000_RESET0 GLOBAL_REG(0x0c) -- cgit v1.2.3 From 3f7a87d2fa9b42f7aade43914f060df68cc89cc7 Mon Sep 17 00:00:00 2001 From: Frank Filz Date: Mon, 20 Jun 2005 13:14:57 -0700 Subject: [SCTP] sctp_connectx() API support Implements sctp_connectx() as defined in the SCTP sockets API draft by tunneling the request through a setsockopt(). Signed-off-by: Frank Filz Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- include/net/sctp/command.h | 8 +- include/net/sctp/constants.h | 7 - include/net/sctp/sctp.h | 17 ++ include/net/sctp/sm.h | 8 +- include/net/sctp/structs.h | 41 ++++- include/net/sctp/user.h | 3 + net/sctp/associola.c | 151 ++++++++++++---- net/sctp/endpointola.c | 1 - net/sctp/input.c | 2 +- net/sctp/outqueue.c | 11 +- net/sctp/sm_make_chunk.c | 20 ++- net/sctp/sm_sideeffect.c | 105 ++++++++--- net/sctp/sm_statefuns.c | 148 ++++++++++------ net/sctp/sm_statetable.c | 6 +- net/sctp/socket.c | 405 ++++++++++++++++++++++++++++++------------- net/sctp/transport.c | 4 +- 16 files changed, 677 insertions(+), 260 deletions(-) (limited to 'include') diff --git a/include/net/sctp/command.h b/include/net/sctp/command.h index ebc5282e6d58..dc107ffad483 100644 --- a/include/net/sctp/command.h +++ b/include/net/sctp/command.h @@ -65,9 +65,11 @@ typedef enum { SCTP_CMD_TIMER_START, /* Start a timer. */ SCTP_CMD_TIMER_RESTART, /* Restart a timer. */ SCTP_CMD_TIMER_STOP, /* Stop a timer. */ - SCTP_CMD_COUNTER_RESET, /* Reset a counter. */ - SCTP_CMD_COUNTER_INC, /* Increment a counter. */ + SCTP_CMD_INIT_CHOOSE_TRANSPORT, /* Choose transport for an INIT. */ + SCTP_CMD_INIT_COUNTER_RESET, /* Reset init counter. */ + SCTP_CMD_INIT_COUNTER_INC, /* Increment init counter. */ SCTP_CMD_INIT_RESTART, /* High level, do init timer work. */ + SCTP_CMD_COOKIEECHO_RESTART, /* High level, do cookie-echo timer work. */ SCTP_CMD_INIT_FAILED, /* High level, do init failure work. */ SCTP_CMD_REPORT_DUP, /* Report a duplicate TSN. */ SCTP_CMD_STRIKE, /* Mark a strike against a transport. */ @@ -118,7 +120,6 @@ typedef union { int error; sctp_state_t state; sctp_event_timeout_t to; - sctp_counter_t counter; void *ptr; struct sctp_chunk *chunk; struct sctp_association *asoc; @@ -165,7 +166,6 @@ SCTP_ARG_CONSTRUCTOR(U16, __u16, u16) SCTP_ARG_CONSTRUCTOR(U8, __u8, u8) SCTP_ARG_CONSTRUCTOR(ERROR, int, error) SCTP_ARG_CONSTRUCTOR(STATE, sctp_state_t, state) -SCTP_ARG_CONSTRUCTOR(COUNTER, sctp_counter_t, counter) SCTP_ARG_CONSTRUCTOR(TO, sctp_event_timeout_t, to) SCTP_ARG_CONSTRUCTOR(PTR, void *, ptr) SCTP_ARG_CONSTRUCTOR(CHUNK, struct sctp_chunk *, chunk) diff --git a/include/net/sctp/constants.h b/include/net/sctp/constants.h index 2b76c0f4babc..4868c7f7749d 100644 --- a/include/net/sctp/constants.h +++ b/include/net/sctp/constants.h @@ -263,13 +263,6 @@ enum { SCTP_MIN_PMTU = 576 }; enum { SCTP_MAX_DUP_TSNS = 16 }; enum { SCTP_MAX_GABS = 16 }; -typedef enum { - SCTP_COUNTER_INIT_ERROR, -} sctp_counter_t; - -/* How many counters does an association need? */ -#define SCTP_NUMBER_COUNTERS 5 - /* Here we define the default timers. */ /* cookie timer def = ? seconds */ diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 960abfa48d68..ef2738159ab3 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -223,6 +223,22 @@ DECLARE_SNMP_STAT(struct sctp_mib, sctp_statistics); extern int sctp_debug_flag; #define SCTP_DEBUG_PRINTK(whatever...) \ ((void) (sctp_debug_flag && printk(KERN_DEBUG whatever))) +#define SCTP_DEBUG_PRINTK_IPADDR(lead, trail, leadparm, saddr, otherparms...) \ + if (sctp_debug_flag) { \ + if (saddr->sa.sa_family == AF_INET6) { \ + printk(KERN_DEBUG \ + lead "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x" trail, \ + leadparm, \ + NIP6(saddr->v6.sin6_addr), \ + otherparms); \ + } else { \ + printk(KERN_DEBUG \ + lead "%u.%u.%u.%u" trail, \ + leadparm, \ + NIPQUAD(saddr->v4.sin_addr.s_addr), \ + otherparms); \ + } \ + } #define SCTP_ENABLE_DEBUG { sctp_debug_flag = 1; } #define SCTP_DISABLE_DEBUG { sctp_debug_flag = 0; } @@ -236,6 +252,7 @@ extern int sctp_debug_flag; #else /* SCTP_DEBUG */ #define SCTP_DEBUG_PRINTK(whatever...) +#define SCTP_DEBUG_PRINTK_IPADDR(whatever...) #define SCTP_ENABLE_DEBUG #define SCTP_DISABLE_DEBUG #define SCTP_ASSERT(expr, str, func) diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index f4fcee104707..a53e08a45e32 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -116,7 +116,8 @@ sctp_state_fn_t sctp_sf_eat_data_fast_4_4; sctp_state_fn_t sctp_sf_eat_sack_6_2; sctp_state_fn_t sctp_sf_tabort_8_4_8; sctp_state_fn_t sctp_sf_operr_notify; -sctp_state_fn_t sctp_sf_t1_timer_expire; +sctp_state_fn_t sctp_sf_t1_init_timer_expire; +sctp_state_fn_t sctp_sf_t1_cookie_timer_expire; sctp_state_fn_t sctp_sf_t2_timer_expire; sctp_state_fn_t sctp_sf_t4_timer_expire; sctp_state_fn_t sctp_sf_t5_timer_expire; @@ -258,7 +259,10 @@ struct sctp_chunk *sctp_make_fwdtsn(const struct sctp_association *asoc, void sctp_chunk_assign_tsn(struct sctp_chunk *); void sctp_chunk_assign_ssn(struct sctp_chunk *); -void sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, __u16 error); +sctp_disposition_t sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, + __u16 error, + const struct sctp_association *asoc, + struct sctp_transport *transport); /* Prototypes for statetable processing. */ diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 6c24d9cd3d66..dfad4d3c581c 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -867,10 +867,13 @@ struct sctp_transport { */ unsigned long last_time_ecne_reduced; - /* active : The current active state of this destination, - * : i.e. DOWN, UP, etc. + /* The number of times INIT has been sent on this transport. */ + int init_sent_count; + + /* state : The current state of this destination, + * : i.e. SCTP_ACTIVE, SCTP_INACTIVE, SCTP_UNKOWN. */ - int active; + int state; /* hb_allowed : The current heartbeat state of this destination, * : i.e. ALLOW-HB, NO-HEARTBEAT, etc. @@ -1222,9 +1225,6 @@ struct sctp_endpoint { /* sendbuf acct. policy. */ __u32 sndbuf_policy; - - /* Name for debugging output... */ - char *debug_name; }; /* Recover the outter endpoint structure. */ @@ -1314,11 +1314,23 @@ struct sctp_association { * : association. Normally this information is * : hashed or keyed for quick lookup and access * : of the TCB. + * : The list is also initialized with the list + * : of addresses passed with the sctp_connectx() + * : call. * * It is a list of SCTP_transport's. */ struct list_head transport_addr_list; + /* transport_count + * + * Peer : A count of the number of peer addresses + * Transport : in the Peer Transport Address List. + * Address : + * Count : + */ + __u16 transport_count; + /* port * The transport layer port number. */ @@ -1486,6 +1498,9 @@ struct sctp_association { /* Transport to which SHUTDOWN chunk was last sent. */ struct sctp_transport *shutdown_last_sent_to; + /* Transport to which INIT chunk was last sent. */ + struct sctp_transport *init_last_sent_to; + /* Next TSN : The next TSN number to be assigned to a new * : DATA chunk. This is sent in the INIT or INIT * : ACK chunk to the peer and incremented each @@ -1549,8 +1564,11 @@ struct sctp_association { /* The message size at which SCTP fragmentation will occur. */ __u32 frag_point; - /* Currently only one counter is used to count INIT errors. */ - int counters[SCTP_NUMBER_COUNTERS]; + /* Counter used to count INIT errors. */ + int init_err_counter; + + /* Count the number of INIT cycles (for doubling timeout). */ + int init_cycle; /* Default send parameters. */ __u16 default_stream; @@ -1708,6 +1726,8 @@ void sctp_association_free(struct sctp_association *); void sctp_association_put(struct sctp_association *); void sctp_association_hold(struct sctp_association *); +struct sctp_transport *sctp_assoc_choose_init_transport( + struct sctp_association *); struct sctp_transport *sctp_assoc_choose_shutdown_transport( struct sctp_association *); void sctp_assoc_update_retran_path(struct sctp_association *); @@ -1717,9 +1737,12 @@ int sctp_assoc_lookup_laddr(struct sctp_association *asoc, const union sctp_addr *laddr); struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *, const union sctp_addr *address, - const int gfp); + const int gfp, + const int peer_state); void sctp_assoc_del_peer(struct sctp_association *asoc, const union sctp_addr *addr); +void sctp_assoc_rm_peer(struct sctp_association *asoc, + struct sctp_transport *peer); void sctp_assoc_control_transport(struct sctp_association *, struct sctp_transport *, sctp_transport_cmd_t, sctp_sn_error_t); diff --git a/include/net/sctp/user.h b/include/net/sctp/user.h index 2758e8ce4f25..f6328aeddcce 100644 --- a/include/net/sctp/user.h +++ b/include/net/sctp/user.h @@ -111,6 +111,8 @@ enum sctp_optname { #define SCTP_GET_LOCAL_ADDRS_NUM SCTP_GET_LOCAL_ADDRS_NUM SCTP_GET_LOCAL_ADDRS, /* Get all local addresss. */ #define SCTP_GET_LOCAL_ADDRS SCTP_GET_LOCAL_ADDRS + SCTP_SOCKOPT_CONNECTX, /* CONNECTX requests. */ +#define SCTP_SOCKOPT_CONNECTX SCTP_SOCKOPT_CONNECTX }; /* @@ -527,6 +529,7 @@ struct sctp_paddrinfo { enum sctp_spinfo_state { SCTP_INACTIVE, SCTP_ACTIVE, + SCTP_UNKNOWN = 0xffff /* Value used for transport state unknown */ }; /* diff --git a/net/sctp/associola.c b/net/sctp/associola.c index 663843d97a92..7ae6aa772dab 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -191,10 +191,6 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a asoc->last_cwr_tsn = asoc->ctsn_ack_point; asoc->unack_data = 0; - SCTP_DEBUG_PRINTK("myctsnap for %s INIT as 0x%x.\n", - asoc->ep->debug_name, - asoc->ctsn_ack_point); - /* ADDIP Section 4.1 Asconf Chunk Procedures * * When an endpoint has an ASCONF signaled change to be sent to the @@ -211,6 +207,7 @@ static struct sctp_association *sctp_association_init(struct sctp_association *a /* Make an empty list of remote transport addresses. */ INIT_LIST_HEAD(&asoc->peer.transport_addr_list); + asoc->peer.transport_count = 0; /* RFC 2960 5.1 Normal Establishment of an Association * @@ -288,6 +285,7 @@ struct sctp_association *sctp_association_new(const struct sctp_endpoint *ep, asoc->base.malloced = 1; SCTP_DBG_OBJCNT_INC(assoc); + SCTP_DEBUG_PRINTK("Created asoc %p\n", asoc); return asoc; @@ -356,6 +354,8 @@ void sctp_association_free(struct sctp_association *asoc) sctp_transport_free(transport); } + asoc->peer.transport_count = 0; + /* Free any cached ASCONF_ACK chunk. */ if (asoc->addip_last_asconf_ack) sctp_chunk_free(asoc->addip_last_asconf_ack); @@ -400,7 +400,7 @@ void sctp_assoc_set_primary(struct sctp_association *asoc, /* If the primary path is changing, assume that the * user wants to use this new path. */ - if (transport->active) + if (transport->state != SCTP_INACTIVE) asoc->peer.active_path = transport; /* @@ -428,10 +428,58 @@ void sctp_assoc_set_primary(struct sctp_association *asoc, transport->cacc.next_tsn_at_change = asoc->next_tsn; } +/* Remove a transport from an association. */ +void sctp_assoc_rm_peer(struct sctp_association *asoc, + struct sctp_transport *peer) +{ + struct list_head *pos; + struct sctp_transport *transport; + + SCTP_DEBUG_PRINTK_IPADDR("sctp_assoc_rm_peer:association %p addr: ", + " port: %d\n", + asoc, + (&peer->ipaddr), + peer->ipaddr.v4.sin_port); + + /* If we are to remove the current retran_path, update it + * to the next peer before removing this peer from the list. + */ + if (asoc->peer.retran_path == peer) + sctp_assoc_update_retran_path(asoc); + + /* Remove this peer from the list. */ + list_del(&peer->transports); + + /* Get the first transport of asoc. */ + pos = asoc->peer.transport_addr_list.next; + transport = list_entry(pos, struct sctp_transport, transports); + + /* Update any entries that match the peer to be deleted. */ + if (asoc->peer.primary_path == peer) + sctp_assoc_set_primary(asoc, transport); + if (asoc->peer.active_path == peer) + asoc->peer.active_path = transport; + if (asoc->peer.last_data_from == peer) + asoc->peer.last_data_from = transport; + + /* If we remove the transport an INIT was last sent to, set it to + * NULL. Combined with the update of the retran path above, this + * will cause the next INIT to be sent to the next available + * transport, maintaining the cycle. + */ + if (asoc->init_last_sent_to == peer) + asoc->init_last_sent_to = NULL; + + asoc->peer.transport_count--; + + sctp_transport_free(peer); +} + /* Add a transport address to an association. */ struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc, const union sctp_addr *addr, - int gfp) + const int gfp, + const int peer_state) { struct sctp_transport *peer; struct sctp_sock *sp; @@ -442,14 +490,25 @@ struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc, /* AF_INET and AF_INET6 share common port field. */ port = addr->v4.sin_port; + SCTP_DEBUG_PRINTK_IPADDR("sctp_assoc_add_peer:association %p addr: ", + " port: %d state:%s\n", + asoc, + addr, + addr->v4.sin_port, + peer_state == SCTP_UNKNOWN?"UNKNOWN":"ACTIVE"); + /* Set the port if it has not been set yet. */ if (0 == asoc->peer.port) asoc->peer.port = port; /* Check to see if this is a duplicate. */ peer = sctp_assoc_lookup_paddr(asoc, addr); - if (peer) + if (peer) { + if (peer_state == SCTP_ACTIVE && + peer->state == SCTP_UNKNOWN) + peer->state = SCTP_ACTIVE; return peer; + } peer = sctp_transport_new(addr, gfp); if (!peer) @@ -516,8 +575,12 @@ struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc, /* Set the transport's RTO.initial value */ peer->rto = asoc->rto_initial; + /* Set the peer's active state. */ + peer->state = peer_state; + /* Attach the remote transport to our asoc. */ list_add_tail(&peer->transports, &asoc->peer.transport_addr_list); + asoc->peer.transport_count++; /* If we do not yet have a primary path, set one. */ if (!asoc->peer.primary_path) { @@ -525,8 +588,9 @@ struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc, asoc->peer.retran_path = peer; } - if (asoc->peer.active_path == asoc->peer.retran_path) + if (asoc->peer.active_path == asoc->peer.retran_path) { asoc->peer.retran_path = peer; + } return peer; } @@ -537,37 +601,16 @@ void sctp_assoc_del_peer(struct sctp_association *asoc, { struct list_head *pos; struct list_head *temp; - struct sctp_transport *peer = NULL; struct sctp_transport *transport; list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) { transport = list_entry(pos, struct sctp_transport, transports); if (sctp_cmp_addr_exact(addr, &transport->ipaddr)) { - peer = transport; - list_del(pos); + /* Do book keeping for removing the peer and free it. */ + sctp_assoc_rm_peer(asoc, transport); break; } } - - /* The address we want delete is not in the association. */ - if (!peer) - return; - - /* Get the first transport of asoc. */ - pos = asoc->peer.transport_addr_list.next; - transport = list_entry(pos, struct sctp_transport, transports); - - /* Update any entries that match the peer to be deleted. */ - if (asoc->peer.primary_path == peer) - sctp_assoc_set_primary(asoc, transport); - if (asoc->peer.active_path == peer) - asoc->peer.active_path = transport; - if (asoc->peer.retran_path == peer) - asoc->peer.retran_path = transport; - if (asoc->peer.last_data_from == peer) - asoc->peer.last_data_from = transport; - - sctp_transport_free(peer); } /* Lookup a transport by address. */ @@ -608,12 +651,12 @@ void sctp_assoc_control_transport(struct sctp_association *asoc, /* Record the transition on the transport. */ switch (command) { case SCTP_TRANSPORT_UP: - transport->active = SCTP_ACTIVE; + transport->state = SCTP_ACTIVE; spc_state = SCTP_ADDR_AVAILABLE; break; case SCTP_TRANSPORT_DOWN: - transport->active = SCTP_INACTIVE; + transport->state = SCTP_INACTIVE; spc_state = SCTP_ADDR_UNREACHABLE; break; @@ -643,7 +686,7 @@ void sctp_assoc_control_transport(struct sctp_association *asoc, list_for_each(pos, &asoc->peer.transport_addr_list) { t = list_entry(pos, struct sctp_transport, transports); - if (!t->active) + if (t->state == SCTP_INACTIVE) continue; if (!first || t->last_time_heard > first->last_time_heard) { second = first; @@ -663,7 +706,7 @@ void sctp_assoc_control_transport(struct sctp_association *asoc, * [If the primary is active but not most recent, bump the most * recently used transport.] */ - if (asoc->peer.primary_path->active && + if (asoc->peer.primary_path->state != SCTP_INACTIVE && first != asoc->peer.primary_path) { second = first; first = asoc->peer.primary_path; @@ -958,7 +1001,7 @@ void sctp_assoc_update(struct sctp_association *asoc, transports); if (!sctp_assoc_lookup_paddr(asoc, &trans->ipaddr)) sctp_assoc_add_peer(asoc, &trans->ipaddr, - GFP_ATOMIC); + GFP_ATOMIC, SCTP_ACTIVE); } asoc->ctsn_ack_point = asoc->next_tsn - 1; @@ -998,7 +1041,7 @@ void sctp_assoc_update_retran_path(struct sctp_association *asoc) /* Try to find an active transport. */ - if (t->active) { + if (t->state != SCTP_INACTIVE) { break; } else { /* Keep track of the next transport in case @@ -1019,6 +1062,40 @@ void sctp_assoc_update_retran_path(struct sctp_association *asoc) } asoc->peer.retran_path = t; + + SCTP_DEBUG_PRINTK_IPADDR("sctp_assoc_update_retran_path:association" + " %p addr: ", + " port: %d\n", + asoc, + (&t->ipaddr), + t->ipaddr.v4.sin_port); +} + +/* Choose the transport for sending a INIT packet. */ +struct sctp_transport *sctp_assoc_choose_init_transport( + struct sctp_association *asoc) +{ + struct sctp_transport *t; + + /* Use the retran path. If the last INIT was sent over the + * retran path, update the retran path and use it. + */ + if (!asoc->init_last_sent_to) { + t = asoc->peer.active_path; + } else { + if (asoc->init_last_sent_to == asoc->peer.retran_path) + sctp_assoc_update_retran_path(asoc); + t = asoc->peer.retran_path; + } + + SCTP_DEBUG_PRINTK_IPADDR("sctp_assoc_update_retran_path:association" + " %p addr: ", + " port: %d\n", + asoc, + (&t->ipaddr), + t->ipaddr.v4.sin_port); + + return t; } /* Choose the transport for sending a SHUTDOWN packet. */ diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c index 334f61773e6d..2ec0320fac3b 100644 --- a/net/sctp/endpointola.c +++ b/net/sctp/endpointola.c @@ -134,7 +134,6 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, ep->last_key = ep->current_key = 0; ep->key_changed_at = jiffies; - ep->debug_name = "unnamedEndpoint"; return ep; } diff --git a/net/sctp/input.c b/net/sctp/input.c index fffc880a646d..339f7acfdb64 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -353,7 +353,7 @@ void sctp_icmp_proto_unreachable(struct sock *sk, sctp_do_sm(SCTP_EVENT_T_OTHER, SCTP_ST_OTHER(SCTP_EVENT_ICMP_PROTO_UNREACH), - asoc->state, asoc->ep, asoc, NULL, + asoc->state, asoc->ep, asoc, t, GFP_ATOMIC); } diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c index 1b2d4adc4ddb..4eb81a1407b7 100644 --- a/net/sctp/outqueue.c +++ b/net/sctp/outqueue.c @@ -682,9 +682,9 @@ int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout) if (!new_transport) { new_transport = asoc->peer.active_path; - } else if (!new_transport->active) { - /* If the chunk is Heartbeat or Heartbeat Ack, - * send it to chunk->transport, even if it's + } else if (new_transport->state == SCTP_INACTIVE) { + /* If the chunk is Heartbeat or Heartbeat Ack, + * send it to chunk->transport, even if it's * inactive. * * 3.3.6 Heartbeat Acknowledgement: @@ -840,7 +840,8 @@ int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout) * Otherwise, we want to use the active path. */ new_transport = chunk->transport; - if (!new_transport || !new_transport->active) + if (!new_transport || + new_transport->state == SCTP_INACTIVE) new_transport = asoc->peer.active_path; /* Change packets if necessary. */ @@ -1454,7 +1455,7 @@ static void sctp_check_transmitted(struct sctp_outq *q, /* Mark the destination transport address as * active if it is not so marked. */ - if (!transport->active) { + if (transport->state == SCTP_INACTIVE) { sctp_assoc_control_transport( transport->asoc, transport, diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 33ac8bf47b0e..5baed9bb7de5 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -1830,7 +1830,7 @@ int sctp_process_init(struct sctp_association *asoc, sctp_cid_t cid, * be a a better choice than any of the embedded addresses. */ if (peer_addr) - if(!sctp_assoc_add_peer(asoc, peer_addr, gfp)) + if(!sctp_assoc_add_peer(asoc, peer_addr, gfp, SCTP_ACTIVE)) goto nomem; /* Process the initialization parameters. */ @@ -1841,6 +1841,14 @@ int sctp_process_init(struct sctp_association *asoc, sctp_cid_t cid, goto clean_up; } + /* Walk list of transports, removing transports in the UNKNOWN state. */ + list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) { + transport = list_entry(pos, struct sctp_transport, transports); + if (transport->state == SCTP_UNKNOWN) { + sctp_assoc_rm_peer(asoc, transport); + } + } + /* The fixed INIT headers are always in network byte * order. */ @@ -1906,7 +1914,8 @@ int sctp_process_init(struct sctp_association *asoc, sctp_cid_t cid, * stream sequence number shall be set to 0. */ - /* Allocate storage for the negotiated streams if it is not a temporary * association. + /* Allocate storage for the negotiated streams if it is not a temporary + * association. */ if (!asoc->temp) { int assoc_id; @@ -1952,6 +1961,9 @@ clean_up: list_del_init(pos); sctp_transport_free(transport); } + + asoc->peer.transport_count = 0; + nomem: return 0; } @@ -1995,7 +2007,7 @@ static int sctp_process_param(struct sctp_association *asoc, af->from_addr_param(&addr, param.addr, asoc->peer.port, 0); scope = sctp_scope(peer_addr); if (sctp_in_scope(&addr, scope)) - if (!sctp_assoc_add_peer(asoc, &addr, gfp)) + if (!sctp_assoc_add_peer(asoc, &addr, gfp, SCTP_ACTIVE)) return 0; break; @@ -2396,7 +2408,7 @@ static __u16 sctp_process_asconf_param(struct sctp_association *asoc, * Due to Resource Shortage'. */ - peer = sctp_assoc_add_peer(asoc, &addr, GFP_ATOMIC); + peer = sctp_assoc_add_peer(asoc, &addr, GFP_ATOMIC, SCTP_ACTIVE); if (!peer) return SCTP_ERROR_RSRC_LOW; diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index f65fa441952f..778639db125a 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -414,11 +414,13 @@ static void sctp_do_8_2_transport_strike(struct sctp_association *asoc, */ asoc->overall_error_count++; - if (transport->active && + if (transport->state != SCTP_INACTIVE && (transport->error_count++ >= transport->max_retrans)) { - SCTP_DEBUG_PRINTK("transport_strike: transport " - "IP:%d.%d.%d.%d failed.\n", - NIPQUAD(transport->ipaddr.v4.sin_addr)); + SCTP_DEBUG_PRINTK_IPADDR("transport_strike:association %p", + " transport IP: port:%d failed.\n", + asoc, + (&transport->ipaddr), + transport->ipaddr.v4.sin_port); sctp_assoc_control_transport(asoc, transport, SCTP_TRANSPORT_DOWN, SCTP_FAILED_THRESHOLD); @@ -593,7 +595,7 @@ static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds, /* Mark the destination transport address as active if it is not so * marked. */ - if (!t->active) + if (t->state == SCTP_INACTIVE) sctp_assoc_control_transport(asoc, t, SCTP_TRANSPORT_UP, SCTP_HEARTBEAT_SUCCESS); @@ -665,8 +667,11 @@ static void sctp_cmd_new_state(sctp_cmd_seq_t *cmds, asoc->state = state; + SCTP_DEBUG_PRINTK("sctp_cmd_new_state: asoc %p[%s]\n", + asoc, sctp_state_tbl[state]); + if (sctp_style(sk, TCP)) { - /* Change the sk->sk_state of a TCP-style socket that has + /* Change the sk->sk_state of a TCP-style socket that has * sucessfully completed a connect() call. */ if (sctp_state(asoc, ESTABLISHED) && sctp_sstate(sk, CLOSED)) @@ -678,6 +683,16 @@ static void sctp_cmd_new_state(sctp_cmd_seq_t *cmds, sk->sk_shutdown |= RCV_SHUTDOWN; } + if (sctp_state(asoc, COOKIE_WAIT)) { + /* Reset init timeouts since they may have been + * increased due to timer expirations. + */ + asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] = + asoc->ep->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT]; + asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] = + asoc->ep->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE]; + } + if (sctp_state(asoc, ESTABLISHED) || sctp_state(asoc, CLOSED) || sctp_state(asoc, SHUTDOWN_RECEIVED)) { @@ -1120,10 +1135,10 @@ static int sctp_cmd_interpreter(sctp_event_t event_type, * to be executed only during failed attempts of * association establishment. */ - if ((asoc->peer.retran_path != - asoc->peer.primary_path) && - (asoc->counters[SCTP_COUNTER_INIT_ERROR] > 0)) { - sctp_add_cmd_sf(commands, + if ((asoc->peer.retran_path != + asoc->peer.primary_path) && + (asoc->init_err_counter > 0)) { + sctp_add_cmd_sf(commands, SCTP_CMD_FORCE_PRIM_RETRAN, SCTP_NULL()); } @@ -1237,18 +1252,67 @@ static int sctp_cmd_interpreter(sctp_event_t event_type, sctp_association_put(asoc); break; + case SCTP_CMD_INIT_CHOOSE_TRANSPORT: + chunk = cmd->obj.ptr; + t = sctp_assoc_choose_init_transport(asoc); + asoc->init_last_sent_to = t; + chunk->transport = t; + t->init_sent_count++; + break; + case SCTP_CMD_INIT_RESTART: /* Do the needed accounting and updates * associated with restarting an initialization - * timer. + * timer. Only multiply the timeout by two if + * all transports have been tried at the current + * timeout. + */ + t = asoc->init_last_sent_to; + asoc->init_err_counter++; + + if (t->init_sent_count > (asoc->init_cycle + 1)) { + asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] *= 2; + if (asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] > + asoc->max_init_timeo) { + asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] = + asoc->max_init_timeo; + } + asoc->init_cycle++; + SCTP_DEBUG_PRINTK( + "T1 INIT Timeout adjustment" + " init_err_counter: %d" + " cycle: %d" + " timeout: %d\n", + asoc->init_err_counter, + asoc->init_cycle, + asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT]); + } + + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT)); + break; + + case SCTP_CMD_COOKIEECHO_RESTART: + /* Do the needed accounting and updates + * associated with restarting an initialization + * timer. Only multiply the timeout by two if + * all transports have been tried at the current + * timeout. */ - asoc->counters[SCTP_COUNTER_INIT_ERROR]++; - asoc->timeouts[cmd->obj.to] *= 2; - if (asoc->timeouts[cmd->obj.to] > + asoc->init_err_counter++; + + asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] *= 2; + if (asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] > asoc->max_init_timeo) { - asoc->timeouts[cmd->obj.to] = + asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] = asoc->max_init_timeo; } + SCTP_DEBUG_PRINTK( + "T1 COOKIE Timeout adjustment" + " init_err_counter: %d" + " timeout: %d\n", + asoc->init_err_counter, + asoc->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE]); /* If we've sent any data bundled with * COOKIE-ECHO we need to resend. @@ -1261,7 +1325,7 @@ static int sctp_cmd_interpreter(sctp_event_t event_type, sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, - SCTP_TO(cmd->obj.to)); + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_COOKIE)); break; case SCTP_CMD_INIT_FAILED: @@ -1273,12 +1337,13 @@ static int sctp_cmd_interpreter(sctp_event_t event_type, subtype, chunk, cmd->obj.u32); break; - case SCTP_CMD_COUNTER_INC: - asoc->counters[cmd->obj.counter]++; + case SCTP_CMD_INIT_COUNTER_INC: + asoc->init_err_counter++; break; - case SCTP_CMD_COUNTER_RESET: - asoc->counters[cmd->obj.counter] = 0; + case SCTP_CMD_INIT_COUNTER_RESET: + asoc->init_err_counter = 0; + asoc->init_cycle = 0; break; case SCTP_CMD_REPORT_DUP: diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 8e01b8f09ac2..058189684c7c 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -533,6 +533,9 @@ sctp_disposition_t sctp_sf_do_5_1C_ack(const struct sctp_endpoint *ep, sctp_add_cmd_sf(commands, SCTP_CMD_PEER_INIT, SCTP_PEER_INIT(initchunk)); + /* Reset init error count upon receipt of INIT-ACK. */ + sctp_add_cmd_sf(commands, SCTP_CMD_INIT_COUNTER_RESET, SCTP_NULL()); + /* 5.1 C) "A" shall stop the T1-init timer and leave * COOKIE-WAIT state. "A" shall then ... start the T1-cookie * timer, and enter the COOKIE-ECHOED state. @@ -775,8 +778,7 @@ sctp_disposition_t sctp_sf_do_5_1E_ca(const struct sctp_endpoint *ep, * from the COOKIE-ECHOED state to the COOKIE-WAIT * state is performed. */ - sctp_add_cmd_sf(commands, SCTP_CMD_COUNTER_RESET, - SCTP_COUNTER(SCTP_COUNTER_INIT_ERROR)); + sctp_add_cmd_sf(commands, SCTP_CMD_INIT_COUNTER_RESET, SCTP_NULL()); /* RFC 2960 5.1 Normal Establishment of an Association * @@ -1019,10 +1021,22 @@ sctp_disposition_t sctp_sf_backbeat_8_3(const struct sctp_endpoint *ep, link = sctp_assoc_lookup_paddr(asoc, &from_addr); /* This should never happen, but lets log it if so. */ - if (!link) { - printk(KERN_WARNING - "%s: Could not find address %d.%d.%d.%d\n", - __FUNCTION__, NIPQUAD(from_addr.v4.sin_addr)); + if (unlikely(!link)) { + if (from_addr.sa.sa_family == AF_INET6) { + printk(KERN_WARNING + "%s association %p could not find address " + "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + __FUNCTION__, + asoc, + NIP6(from_addr.v6.sin6_addr)); + } else { + printk(KERN_WARNING + "%s association %p could not find address " + "%u.%u.%u.%u\n", + __FUNCTION__, + asoc, + NIPQUAD(from_addr.v4.sin_addr.s_addr)); + } return SCTP_DISPOSITION_DISCARD; } @@ -2095,9 +2109,7 @@ static sctp_disposition_t sctp_sf_do_5_2_6_stale(const struct sctp_endpoint *ep, sctp_errhdr_t *err; struct sctp_chunk *reply; struct sctp_bind_addr *bp; - int attempts; - - attempts = asoc->counters[SCTP_COUNTER_INIT_ERROR] + 1; + int attempts = asoc->init_err_counter + 1; if (attempts >= asoc->max_init_attempts) { sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, @@ -2157,8 +2169,7 @@ static sctp_disposition_t sctp_sf_do_5_2_6_stale(const struct sctp_endpoint *ep, /* Cast away the const modifier, as we want to just * rerun it through as a sideffect. */ - sctp_add_cmd_sf(commands, SCTP_CMD_COUNTER_INC, - SCTP_COUNTER(SCTP_COUNTER_INIT_ERROR)); + sctp_add_cmd_sf(commands, SCTP_CMD_INIT_COUNTER_INC, SCTP_NULL()); sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, SCTP_TO(SCTP_EVENT_TIMEOUT_T1_COOKIE)); @@ -2281,8 +2292,7 @@ sctp_disposition_t sctp_sf_cookie_wait_abort(const struct sctp_endpoint *ep, if (len >= sizeof(struct sctp_chunkhdr) + sizeof(struct sctp_errhdr)) error = ((sctp_errhdr_t *)chunk->skb->data)->cause; - sctp_stop_t1_and_abort(commands, error); - return SCTP_DISPOSITION_ABORT; + return sctp_stop_t1_and_abort(commands, error, asoc, chunk->transport); } /* @@ -2294,8 +2304,8 @@ sctp_disposition_t sctp_sf_cookie_wait_icmp_abort(const struct sctp_endpoint *ep void *arg, sctp_cmd_seq_t *commands) { - sctp_stop_t1_and_abort(commands, SCTP_ERROR_NO_ERROR); - return SCTP_DISPOSITION_ABORT; + return sctp_stop_t1_and_abort(commands, SCTP_ERROR_NO_ERROR, asoc, + (struct sctp_transport *)arg); } /* @@ -2318,8 +2328,12 @@ sctp_disposition_t sctp_sf_cookie_echoed_abort(const struct sctp_endpoint *ep, * * This is common code called by several sctp_sf_*_abort() functions above. */ -void sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, __u16 error) +sctp_disposition_t sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, + __u16 error, + const struct sctp_association *asoc, + struct sctp_transport *transport) { + SCTP_DEBUG_PRINTK("ABORT received (INIT).\n"); sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, SCTP_STATE(SCTP_STATE_CLOSED)); SCTP_INC_STATS(SCTP_MIB_ABORTEDS); @@ -2328,6 +2342,7 @@ void sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, __u16 error) /* CMD_INIT_FAILED will DELETE_TCB. */ sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, SCTP_U32(error)); + return SCTP_DISPOSITION_ABORT; } /* @@ -3805,6 +3820,10 @@ sctp_disposition_t sctp_sf_do_prm_asoc(const struct sctp_endpoint *ep, sctp_add_cmd_sf(commands, SCTP_CMD_NEW_ASOC, SCTP_ASOC((struct sctp_association *) asoc)); + /* Choose transport for INIT. */ + sctp_add_cmd_sf(commands, SCTP_CMD_INIT_CHOOSE_TRANSPORT, + SCTP_CHUNK(repl)); + /* After sending the INIT, "A" starts the T1-init timer and * enters the COOKIE-WAIT state. */ @@ -4589,7 +4608,7 @@ sctp_disposition_t sctp_sf_do_6_2_sack(const struct sctp_endpoint *ep, } /* - * sctp_sf_t1_timer_expire + * sctp_sf_t1_init_timer_expire * * Section: 4 Note: 2 * Verification Tag: @@ -4603,7 +4622,59 @@ sctp_disposition_t sctp_sf_do_6_2_sack(const struct sctp_endpoint *ep, * endpoint MUST abort the initialization process and report the * error to SCTP user. * - * 3) If the T1-cookie timer expires, the endpoint MUST retransmit + * Outputs + * (timers, events) + * + */ +sctp_disposition_t sctp_sf_t1_init_timer_expire(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *repl = NULL; + struct sctp_bind_addr *bp; + int attempts = asoc->init_err_counter + 1; + + SCTP_DEBUG_PRINTK("Timer T1 expired (INIT).\n"); + + if (attempts < asoc->max_init_attempts) { + bp = (struct sctp_bind_addr *) &asoc->base.bind_addr; + repl = sctp_make_init(asoc, bp, GFP_ATOMIC, 0); + if (!repl) + return SCTP_DISPOSITION_NOMEM; + + /* Choose transport for INIT. */ + sctp_add_cmd_sf(commands, SCTP_CMD_INIT_CHOOSE_TRANSPORT, + SCTP_CHUNK(repl)); + + /* Issue a sideeffect to do the needed accounting. */ + sctp_add_cmd_sf(commands, SCTP_CMD_INIT_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT)); + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl)); + } else { + SCTP_DEBUG_PRINTK("Giving up on INIT, attempts: %d" + " max_init_attempts: %d\n", + attempts, asoc->max_init_attempts); + sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, + SCTP_U32(SCTP_ERROR_NO_ERROR)); + return SCTP_DISPOSITION_DELETE_TCB; + } + + return SCTP_DISPOSITION_CONSUME; +} + +/* + * sctp_sf_t1_cookie_timer_expire + * + * Section: 4 Note: 2 + * Verification Tag: + * Inputs + * (endpoint, asoc) + * + * RFC 2960 Section 4 Notes + * 3) If the T1-cookie timer expires, the endpoint MUST retransmit * COOKIE ECHO and re-start the T1-cookie timer without changing * state. This MUST be repeated up to 'Max.Init.Retransmits' times. * After that, the endpoint MUST abort the initialization process and @@ -4613,46 +4684,26 @@ sctp_disposition_t sctp_sf_do_6_2_sack(const struct sctp_endpoint *ep, * (timers, events) * */ -sctp_disposition_t sctp_sf_t1_timer_expire(const struct sctp_endpoint *ep, +sctp_disposition_t sctp_sf_t1_cookie_timer_expire(const struct sctp_endpoint *ep, const struct sctp_association *asoc, const sctp_subtype_t type, void *arg, sctp_cmd_seq_t *commands) { - struct sctp_chunk *repl; - struct sctp_bind_addr *bp; - sctp_event_timeout_t timer = (sctp_event_timeout_t) arg; - int timeout; - int attempts; - - timeout = asoc->timeouts[timer]; - attempts = asoc->counters[SCTP_COUNTER_INIT_ERROR] + 1; - repl = NULL; + struct sctp_chunk *repl = NULL; + int attempts = asoc->init_err_counter + 1; - SCTP_DEBUG_PRINTK("Timer T1 expired.\n"); + SCTP_DEBUG_PRINTK("Timer T1 expired (COOKIE-ECHO).\n"); if (attempts < asoc->max_init_attempts) { - switch (timer) { - case SCTP_EVENT_TIMEOUT_T1_INIT: - bp = (struct sctp_bind_addr *) &asoc->base.bind_addr; - repl = sctp_make_init(asoc, bp, GFP_ATOMIC, 0); - break; - - case SCTP_EVENT_TIMEOUT_T1_COOKIE: - repl = sctp_make_cookie_echo(asoc, NULL); - break; - - default: - BUG(); - break; - }; - + repl = sctp_make_cookie_echo(asoc, NULL); if (!repl) - goto nomem; + return SCTP_DISPOSITION_NOMEM; /* Issue a sideeffect to do the needed accounting. */ - sctp_add_cmd_sf(commands, SCTP_CMD_INIT_RESTART, - SCTP_TO(timer)); + sctp_add_cmd_sf(commands, SCTP_CMD_COOKIEECHO_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_COOKIE)); + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl)); } else { sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, @@ -4661,9 +4712,6 @@ sctp_disposition_t sctp_sf_t1_timer_expire(const struct sctp_endpoint *ep, } return SCTP_DISPOSITION_CONSUME; - -nomem: - return SCTP_DISPOSITION_NOMEM; } /* RFC2960 9.2 If the timer expires, the endpoint must re-send the SHUTDOWN diff --git a/net/sctp/sm_statetable.c b/net/sctp/sm_statetable.c index 8967846f69e8..75ef10408764 100644 --- a/net/sctp/sm_statetable.c +++ b/net/sctp/sm_statetable.c @@ -783,7 +783,8 @@ static const sctp_sm_table_entry_t other_event_table[SCTP_NUM_OTHER_TYPES][SCTP_ /* SCTP_STATE_COOKIE_WAIT */ \ {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ /* SCTP_STATE_COOKIE_ECHOED */ \ - {.fn = sctp_sf_t1_timer_expire, .name = "sctp_sf_t1_timer_expire"}, \ + {.fn = sctp_sf_t1_cookie_timer_expire, \ + .name = "sctp_sf_t1_cookie_timer_expire"}, \ /* SCTP_STATE_ESTABLISHED */ \ {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ /* SCTP_STATE_SHUTDOWN_PENDING */ \ @@ -802,7 +803,8 @@ static const sctp_sm_table_entry_t other_event_table[SCTP_NUM_OTHER_TYPES][SCTP_ /* SCTP_STATE_CLOSED */ \ {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ /* SCTP_STATE_COOKIE_WAIT */ \ - {.fn = sctp_sf_t1_timer_expire, .name = "sctp_sf_t1_timer_expire"}, \ + {.fn = sctp_sf_t1_init_timer_expire, \ + .name = "sctp_sf_t1_init_timer_expire"}, \ /* SCTP_STATE_COOKIE_ECHOED */ \ {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ /* SCTP_STATE_ESTABLISHED */ \ diff --git a/net/sctp/socket.c b/net/sctp/socket.c index e6926cb19420..aad55dc3792b 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -262,18 +262,18 @@ static struct sctp_transport *sctp_addr_id2transport(struct sock *sk, * sockaddr_in6 [RFC 2553]), * addr_len - the size of the address structure. */ -SCTP_STATIC int sctp_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) +SCTP_STATIC int sctp_bind(struct sock *sk, struct sockaddr *addr, int addr_len) { int retval = 0; sctp_lock_sock(sk); - SCTP_DEBUG_PRINTK("sctp_bind(sk: %p, uaddr: %p, addr_len: %d)\n", - sk, uaddr, addr_len); + SCTP_DEBUG_PRINTK("sctp_bind(sk: %p, addr: %p, addr_len: %d)\n", + sk, addr, addr_len); /* Disallow binding twice. */ if (!sctp_sk(sk)->ep->base.bind_addr.port) - retval = sctp_do_bind(sk, (union sctp_addr *)uaddr, + retval = sctp_do_bind(sk, (union sctp_addr *)addr, addr_len); else retval = -EINVAL; @@ -318,23 +318,27 @@ SCTP_STATIC int sctp_do_bind(struct sock *sk, union sctp_addr *addr, int len) unsigned short snum; int ret = 0; - SCTP_DEBUG_PRINTK("sctp_do_bind(sk: %p, newaddr: %p, len: %d)\n", - sk, addr, len); - /* Common sockaddr verification. */ af = sctp_sockaddr_af(sp, addr, len); - if (!af) + if (!af) { + SCTP_DEBUG_PRINTK("sctp_do_bind(sk: %p, newaddr: %p, len: %d) EINVAL\n", + sk, addr, len); return -EINVAL; + } + + snum = ntohs(addr->v4.sin_port); + + SCTP_DEBUG_PRINTK_IPADDR("sctp_do_bind(sk: %p, new addr: ", + ", port: %d, new port: %d, len: %d)\n", + sk, + addr, + bp->port, snum, + len); /* PF specific bind() address verification. */ if (!sp->pf->bind_verify(sp, addr)) return -EADDRNOTAVAIL; - snum= ntohs(addr->v4.sin_port); - - SCTP_DEBUG_PRINTK("sctp_do_bind: port: %d, new port: %d\n", - bp->port, snum); - /* We must either be unbound, or bind to the same port. */ if (bp->port && (snum != bp->port)) { SCTP_DEBUG_PRINTK("sctp_do_bind:" @@ -816,7 +820,8 @@ out: * * Basically do nothing but copying the addresses from user to kernel * land and invoking either sctp_bindx_add() or sctp_bindx_rem() on the sk. - * This is used for tunneling the sctp_bindx() request through sctp_setsockopt() * from userspace. + * This is used for tunneling the sctp_bindx() request through sctp_setsockopt() + * from userspace. * * We don't use copy_from_user() for optimization: we first do the * sanity checks (buffer size -fast- and access check-healthy @@ -913,6 +918,243 @@ out: return err; } +/* __sctp_connect(struct sock* sk, struct sockaddr *kaddrs, int addrs_size) + * + * Common routine for handling connect() and sctp_connectx(). + * Connect will come in with just a single address. + */ +static int __sctp_connect(struct sock* sk, + struct sockaddr *kaddrs, + int addrs_size) +{ + struct sctp_sock *sp; + struct sctp_endpoint *ep; + struct sctp_association *asoc = NULL; + struct sctp_association *asoc2; + struct sctp_transport *transport; + union sctp_addr to; + struct sctp_af *af; + sctp_scope_t scope; + long timeo; + int err = 0; + int addrcnt = 0; + int walk_size = 0; + struct sockaddr *sa_addr; + void *addr_buf; + + sp = sctp_sk(sk); + ep = sp->ep; + + /* connect() cannot be done on a socket that is already in ESTABLISHED + * state - UDP-style peeled off socket or a TCP-style socket that + * is already connected. + * It cannot be done even on a TCP-style listening socket. + */ + if (sctp_sstate(sk, ESTABLISHED) || + (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING))) { + err = -EISCONN; + goto out_free; + } + + /* Walk through the addrs buffer and count the number of addresses. */ + addr_buf = kaddrs; + while (walk_size < addrs_size) { + sa_addr = (struct sockaddr *)addr_buf; + af = sctp_get_af_specific(sa_addr->sa_family); + + /* If the address family is not supported or if this address + * causes the address buffer to overflow return EINVAL. + */ + if (!af || (walk_size + af->sockaddr_len) > addrs_size) { + err = -EINVAL; + goto out_free; + } + + err = sctp_verify_addr(sk, (union sctp_addr *)sa_addr, + af->sockaddr_len); + if (err) + goto out_free; + + memcpy(&to, sa_addr, af->sockaddr_len); + to.v4.sin_port = ntohs(to.v4.sin_port); + + /* Check if there already is a matching association on the + * endpoint (other than the one created here). + */ + asoc2 = sctp_endpoint_lookup_assoc(ep, &to, &transport); + if (asoc2 && asoc2 != asoc) { + if (asoc2->state >= SCTP_STATE_ESTABLISHED) + err = -EISCONN; + else + err = -EALREADY; + goto out_free; + } + + /* If we could not find a matching association on the endpoint, + * make sure that there is no peeled-off association matching + * the peer address even on another socket. + */ + if (sctp_endpoint_is_peeled_off(ep, &to)) { + err = -EADDRNOTAVAIL; + goto out_free; + } + + if (!asoc) { + /* If a bind() or sctp_bindx() is not called prior to + * an sctp_connectx() call, the system picks an + * ephemeral port and will choose an address set + * equivalent to binding with a wildcard address. + */ + if (!ep->base.bind_addr.port) { + if (sctp_autobind(sk)) { + err = -EAGAIN; + goto out_free; + } + } + + scope = sctp_scope(&to); + asoc = sctp_association_new(ep, sk, scope, GFP_KERNEL); + if (!asoc) { + err = -ENOMEM; + goto out_free; + } + } + + /* Prime the peer's transport structures. */ + transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL, + SCTP_UNKNOWN); + if (!transport) { + err = -ENOMEM; + goto out_free; + } + + addrcnt++; + addr_buf += af->sockaddr_len; + walk_size += af->sockaddr_len; + } + + err = sctp_assoc_set_bind_addr_from_ep(asoc, GFP_KERNEL); + if (err < 0) { + goto out_free; + } + + err = sctp_primitive_ASSOCIATE(asoc, NULL); + if (err < 0) { + goto out_free; + } + + /* Initialize sk's dport and daddr for getpeername() */ + inet_sk(sk)->dport = htons(asoc->peer.port); + af = sctp_get_af_specific(to.sa.sa_family); + af->to_sk_daddr(&to, sk); + + timeo = sock_sndtimeo(sk, sk->sk_socket->file->f_flags & O_NONBLOCK); + err = sctp_wait_for_connect(asoc, &timeo); + + /* Don't free association on exit. */ + asoc = NULL; + +out_free: + + SCTP_DEBUG_PRINTK("About to exit __sctp_connect() free asoc: %p" + " kaddrs: %p err: %d\n", + asoc, kaddrs, err); + if (asoc) + sctp_association_free(asoc); + return err; +} + +/* Helper for tunneling sctp_connectx() requests through sctp_setsockopt() + * + * API 8.9 + * int sctp_connectx(int sd, struct sockaddr *addrs, int addrcnt); + * + * If sd is an IPv4 socket, the addresses passed must be IPv4 addresses. + * If the sd is an IPv6 socket, the addresses passed can either be IPv4 + * or IPv6 addresses. + * + * A single address may be specified as INADDR_ANY or IN6ADDR_ANY, see + * Section 3.1.2 for this usage. + * + * addrs is a pointer to an array of one or more socket addresses. Each + * address is contained in its appropriate structure (i.e. struct + * sockaddr_in or struct sockaddr_in6) the family of the address type + * must be used to distengish the address length (note that this + * representation is termed a "packed array" of addresses). The caller + * specifies the number of addresses in the array with addrcnt. + * + * On success, sctp_connectx() returns 0. On failure, sctp_connectx() returns + * -1, and sets errno to the appropriate error code. + * + * For SCTP, the port given in each socket address must be the same, or + * sctp_connectx() will fail, setting errno to EINVAL. + * + * An application can use sctp_connectx to initiate an association with + * an endpoint that is multi-homed. Much like sctp_bindx() this call + * allows a caller to specify multiple addresses at which a peer can be + * reached. The way the SCTP stack uses the list of addresses to set up + * the association is implementation dependant. This function only + * specifies that the stack will try to make use of all the addresses in + * the list when needed. + * + * Note that the list of addresses passed in is only used for setting up + * the association. It does not necessarily equal the set of addresses + * the peer uses for the resulting association. If the caller wants to + * find out the set of peer addresses, it must use sctp_getpaddrs() to + * retrieve them after the association has been set up. + * + * Basically do nothing but copying the addresses from user to kernel + * land and invoking either sctp_connectx(). This is used for tunneling + * the sctp_connectx() request through sctp_setsockopt() from userspace. + * + * We don't use copy_from_user() for optimization: we first do the + * sanity checks (buffer size -fast- and access check-healthy + * pointer); if all of those succeed, then we can alloc the memory + * (expensive operation) needed to copy the data to kernel. Then we do + * the copying without checking the user space area + * (__copy_from_user()). + * + * On exit there is no need to do sockfd_put(), sys_setsockopt() does + * it. + * + * sk The sk of the socket + * addrs The pointer to the addresses in user land + * addrssize Size of the addrs buffer + * + * Returns 0 if ok, <0 errno code on error. + */ +SCTP_STATIC int sctp_setsockopt_connectx(struct sock* sk, + struct sockaddr __user *addrs, + int addrs_size) +{ + int err = 0; + struct sockaddr *kaddrs; + + SCTP_DEBUG_PRINTK("%s - sk %p addrs %p addrs_size %d\n", + __FUNCTION__, sk, addrs, addrs_size); + + if (unlikely(addrs_size <= 0)) + return -EINVAL; + + /* Check the user passed a healthy pointer. */ + if (unlikely(!access_ok(VERIFY_READ, addrs, addrs_size))) + return -EFAULT; + + /* Alloc space for the address array in kernel memory. */ + kaddrs = (struct sockaddr *)kmalloc(addrs_size, GFP_KERNEL); + if (unlikely(!kaddrs)) + return -ENOMEM; + + if (__copy_from_user(kaddrs, addrs, addrs_size)) { + err = -EFAULT; + } else { + err = __sctp_connect(sk, kaddrs, addrs_size); + } + + kfree(kaddrs); + return err; +} + /* API 3.1.4 close() - UDP Style Syntax * Applications use close() to perform graceful shutdown (as described in * Section 10.1 of [SCTP]) on ALL the associations currently represented @@ -1095,7 +1337,7 @@ SCTP_STATIC int sctp_sendmsg(struct kiocb *iocb, struct sock *sk, sp = sctp_sk(sk); ep = sp->ep; - SCTP_DEBUG_PRINTK("Using endpoint: %s.\n", ep->debug_name); + SCTP_DEBUG_PRINTK("Using endpoint: %p.\n", ep); /* We cannot send a message over a TCP-style listening socket. */ if (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING)) { @@ -1306,7 +1548,7 @@ SCTP_STATIC int sctp_sendmsg(struct kiocb *iocb, struct sock *sk, } /* Prime the peer's transport structures. */ - transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL); + transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL, SCTP_UNKNOWN); if (!transport) { err = -ENOMEM; goto out_free; @@ -2208,6 +2450,12 @@ SCTP_STATIC int sctp_setsockopt(struct sock *sk, int level, int optname, optlen, SCTP_BINDX_REM_ADDR); break; + case SCTP_SOCKOPT_CONNECTX: + /* 'optlen' is the size of the addresses buffer. */ + retval = sctp_setsockopt_connectx(sk, (struct sockaddr __user *)optval, + optlen); + break; + case SCTP_DISABLE_FRAGMENTS: retval = sctp_setsockopt_disable_fragments(sk, optval, optlen); break; @@ -2283,112 +2531,29 @@ out_nounlock: * * len: the size of the address. */ -SCTP_STATIC int sctp_connect(struct sock *sk, struct sockaddr *uaddr, +SCTP_STATIC int sctp_connect(struct sock *sk, struct sockaddr *addr, int addr_len) { - struct sctp_sock *sp; - struct sctp_endpoint *ep; - struct sctp_association *asoc; - struct sctp_transport *transport; - union sctp_addr to; - struct sctp_af *af; - sctp_scope_t scope; - long timeo; int err = 0; + struct sctp_af *af; sctp_lock_sock(sk); - SCTP_DEBUG_PRINTK("%s - sk: %p, sockaddr: %p, addr_len: %d)\n", - __FUNCTION__, sk, uaddr, addr_len); - - sp = sctp_sk(sk); - ep = sp->ep; - - /* connect() cannot be done on a socket that is already in ESTABLISHED - * state - UDP-style peeled off socket or a TCP-style socket that - * is already connected. - * It cannot be done even on a TCP-style listening socket. - */ - if (sctp_sstate(sk, ESTABLISHED) || - (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING))) { - err = -EISCONN; - goto out_unlock; - } - - err = sctp_verify_addr(sk, (union sctp_addr *)uaddr, addr_len); - if (err) - goto out_unlock; - - if (addr_len > sizeof(to)) - addr_len = sizeof(to); - memcpy(&to, uaddr, addr_len); - to.v4.sin_port = ntohs(to.v4.sin_port); - - asoc = sctp_endpoint_lookup_assoc(ep, &to, &transport); - if (asoc) { - if (asoc->state >= SCTP_STATE_ESTABLISHED) - err = -EISCONN; - else - err = -EALREADY; - goto out_unlock; - } - - /* If we could not find a matching association on the endpoint, - * make sure that there is no peeled-off association matching the - * peer address even on another socket. - */ - if (sctp_endpoint_is_peeled_off(ep, &to)) { - err = -EADDRNOTAVAIL; - goto out_unlock; - } - - /* If a bind() or sctp_bindx() is not called prior to a connect() - * call, the system picks an ephemeral port and will choose an address - * set equivalent to binding with a wildcard address. - */ - if (!ep->base.bind_addr.port) { - if (sctp_autobind(sk)) { - err = -EAGAIN; - goto out_unlock; - } - } - - scope = sctp_scope(&to); - asoc = sctp_association_new(ep, sk, scope, GFP_KERNEL); - if (!asoc) { - err = -ENOMEM; - goto out_unlock; - } + SCTP_DEBUG_PRINTK("%s - sk: %p, sockaddr: %p, addr_len: %d\n", + __FUNCTION__, sk, addr, addr_len); - /* Prime the peer's transport structures. */ - transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL); - if (!transport) { - sctp_association_free(asoc); - goto out_unlock; - } - err = sctp_assoc_set_bind_addr_from_ep(asoc, GFP_KERNEL); - if (err < 0) { - sctp_association_free(asoc); - goto out_unlock; - } - - err = sctp_primitive_ASSOCIATE(asoc, NULL); - if (err < 0) { - sctp_association_free(asoc); - goto out_unlock; + /* Validate addr_len before calling common connect/connectx routine. */ + af = sctp_get_af_specific(addr->sa_family); + if (!af || addr_len < af->sockaddr_len) { + err = -EINVAL; + } else { + /* Pass correct addr len to common routine (so it knows there + * is only one address being passed. + */ + err = __sctp_connect(sk, addr, af->sockaddr_len); } - /* Initialize sk's dport and daddr for getpeername() */ - inet_sk(sk)->dport = htons(asoc->peer.port); - af = sctp_get_af_specific(to.sa.sa_family); - af->to_sk_daddr(&to, sk); - - timeo = sock_sndtimeo(sk, sk->sk_socket->file->f_flags & O_NONBLOCK); - err = sctp_wait_for_connect(asoc, &timeo); - -out_unlock: sctp_release_sock(sk); - return err; } @@ -2677,12 +2842,15 @@ static int sctp_getsockopt_sctp_status(struct sock *sk, int len, /* Map ipv4 address into v4-mapped-on-v6 address. */ sctp_get_pf_specific(sk->sk_family)->addr_v4map(sctp_sk(sk), (union sctp_addr *)&status.sstat_primary.spinfo_address); - status.sstat_primary.spinfo_state = transport->active; + status.sstat_primary.spinfo_state = transport->state; status.sstat_primary.spinfo_cwnd = transport->cwnd; status.sstat_primary.spinfo_srtt = transport->srtt; status.sstat_primary.spinfo_rto = jiffies_to_msecs(transport->rto); status.sstat_primary.spinfo_mtu = transport->pmtu; + if (status.sstat_primary.spinfo_state == SCTP_UNKNOWN) + status.sstat_primary.spinfo_state = SCTP_ACTIVE; + if (put_user(len, optlen)) { retval = -EFAULT; goto out; @@ -2733,12 +2901,15 @@ static int sctp_getsockopt_peer_addr_info(struct sock *sk, int len, return -EINVAL; pinfo.spinfo_assoc_id = sctp_assoc2id(transport->asoc); - pinfo.spinfo_state = transport->active; + pinfo.spinfo_state = transport->state; pinfo.spinfo_cwnd = transport->cwnd; pinfo.spinfo_srtt = transport->srtt; pinfo.spinfo_rto = jiffies_to_msecs(transport->rto); pinfo.spinfo_mtu = transport->pmtu; + if (pinfo.spinfo_state == SCTP_UNKNOWN) + pinfo.spinfo_state = SCTP_ACTIVE; + if (put_user(len, optlen)) { retval = -EFAULT; goto out; @@ -3591,7 +3762,8 @@ SCTP_STATIC int sctp_getsockopt(struct sock *sk, int level, int optname, int retval = 0; int len; - SCTP_DEBUG_PRINTK("sctp_getsockopt(sk: %p, ...)\n", sk); + SCTP_DEBUG_PRINTK("sctp_getsockopt(sk: %p... optname: %d)\n", + sk, optname); /* I can hardly begin to describe how wrong this is. This is * so broken as to be worse than useless. The API draft @@ -4596,8 +4768,7 @@ out: return err; do_error: - if (asoc->counters[SCTP_COUNTER_INIT_ERROR] + 1 >= - asoc->max_init_attempts) + if (asoc->init_err_counter + 1 >= asoc->max_init_attempts) err = -ETIMEDOUT; else err = -ECONNREFUSED; diff --git a/net/sctp/transport.c b/net/sctp/transport.c index f30882e1e96a..0ec0fde6e6c5 100644 --- a/net/sctp/transport.c +++ b/net/sctp/transport.c @@ -83,7 +83,9 @@ static struct sctp_transport *sctp_transport_init(struct sctp_transport *peer, peer->last_time_used = jiffies; peer->last_time_ecne_reduced = jiffies; - peer->active = SCTP_ACTIVE; + peer->init_sent_count = 0; + + peer->state = SCTP_ACTIVE; peer->hb_allowed = 0; /* Initialize the default path max_retrans. */ -- cgit v1.2.3 From 72cb6962a91f2af9eef69a06198e1949c10259ae Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Mon, 20 Jun 2005 13:18:08 -0700 Subject: [IPSEC]: Add xfrm_init_state This patch adds xfrm_init_state which is simply a wrapper that calls xfrm_get_type and subsequently x->type->init_state. It also gets rid of the unused args argument. Abstracting it out allows us to add common initialisation code, e.g., to set family-specific flags. The add_time setting in xfrm_user.c was deleted because it's already set by xfrm_state_alloc. Signed-off-by: Herbert Xu Acked-by: James Morris Signed-off-by: David S. Miller --- include/net/xfrm.h | 3 ++- net/ipv4/ah4.c | 2 +- net/ipv4/esp4.c | 2 +- net/ipv4/ipcomp.c | 11 +++-------- net/ipv4/xfrm4_tunnel.c | 2 +- net/ipv6/ah6.c | 2 +- net/ipv6/esp6.c | 2 +- net/ipv6/ipcomp6.c | 9 ++------- net/ipv6/xfrm6_tunnel.c | 2 +- net/key/af_key.c | 12 +++--------- net/xfrm/xfrm_policy.c | 1 - net/xfrm/xfrm_state.c | 21 +++++++++++++++++++++ net/xfrm/xfrm_user.c | 9 +-------- 13 files changed, 38 insertions(+), 40 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 0e65e02b7a1d..77bfdde440f8 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -225,7 +225,7 @@ struct xfrm_type struct module *owner; __u8 proto; - int (*init_state)(struct xfrm_state *x, void *args); + int (*init_state)(struct xfrm_state *x); void (*destructor)(struct xfrm_state *); int (*input)(struct xfrm_state *, struct xfrm_decap_state *, struct sk_buff *skb); int (*post_input)(struct xfrm_state *, struct xfrm_decap_state *, struct sk_buff *skb); @@ -839,6 +839,7 @@ extern int xfrm_replay_check(struct xfrm_state *x, u32 seq); extern void xfrm_replay_advance(struct xfrm_state *x, u32 seq); extern int xfrm_state_check(struct xfrm_state *x, struct sk_buff *skb); extern int xfrm_state_mtu(struct xfrm_state *x, int mtu); +extern int xfrm_init_state(struct xfrm_state *x); extern int xfrm4_rcv(struct sk_buff *skb); extern int xfrm4_output(struct sk_buff *skb); extern int xfrm4_tunnel_register(struct xfrm_tunnel *handler); diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c index 0e98f2235b6e..514c85b2631a 100644 --- a/net/ipv4/ah4.c +++ b/net/ipv4/ah4.c @@ -200,7 +200,7 @@ static void ah4_err(struct sk_buff *skb, u32 info) xfrm_state_put(x); } -static int ah_init_state(struct xfrm_state *x, void *args) +static int ah_init_state(struct xfrm_state *x) { struct ah_data *ahp = NULL; struct xfrm_algo_desc *aalg_desc; diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index eae84cc39d3f..ba57446d5d1f 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -362,7 +362,7 @@ static void esp_destroy(struct xfrm_state *x) kfree(esp); } -static int esp_init_state(struct xfrm_state *x, void *args) +static int esp_init_state(struct xfrm_state *x) { struct esp_data *esp = NULL; diff --git a/net/ipv4/ipcomp.c b/net/ipv4/ipcomp.c index 1a23c5263b99..2065944fd9e5 100644 --- a/net/ipv4/ipcomp.c +++ b/net/ipv4/ipcomp.c @@ -236,15 +236,10 @@ static struct xfrm_state *ipcomp_tunnel_create(struct xfrm_state *x) t->props.mode = 1; t->props.saddr.a4 = x->props.saddr.a4; t->props.flags = x->props.flags; - - t->type = xfrm_get_type(IPPROTO_IPIP, t->props.family); - if (t->type == NULL) - goto error; - - if (t->type->init_state(t, NULL)) + + if (xfrm_init_state(t)) goto error; - t->km.state = XFRM_STATE_VALID; atomic_set(&t->tunnel_users, 1); out: return t; @@ -422,7 +417,7 @@ static void ipcomp_destroy(struct xfrm_state *x) kfree(ipcd); } -static int ipcomp_init_state(struct xfrm_state *x, void *args) +static int ipcomp_init_state(struct xfrm_state *x) { int err; struct ipcomp_data *ipcd; diff --git a/net/ipv4/xfrm4_tunnel.c b/net/ipv4/xfrm4_tunnel.c index 413191f585f6..e1fe360ed27a 100644 --- a/net/ipv4/xfrm4_tunnel.c +++ b/net/ipv4/xfrm4_tunnel.c @@ -84,7 +84,7 @@ static void ipip_err(struct sk_buff *skb, u32 info) handler->err_handler(skb, &arg); } -static int ipip_init_state(struct xfrm_state *x, void *args) +static int ipip_init_state(struct xfrm_state *x) { if (!x->props.mode) return -EINVAL; diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c index e3ecf626cbf7..986fdfdccbcd 100644 --- a/net/ipv6/ah6.c +++ b/net/ipv6/ah6.c @@ -339,7 +339,7 @@ static void ah6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, xfrm_state_put(x); } -static int ah6_init_state(struct xfrm_state *x, void *args) +static int ah6_init_state(struct xfrm_state *x) { struct ah_data *ahp = NULL; struct xfrm_algo_desc *aalg_desc; diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index be7095d6babe..324db62515a2 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -296,7 +296,7 @@ static void esp6_destroy(struct xfrm_state *x) kfree(esp); } -static int esp6_init_state(struct xfrm_state *x, void *args) +static int esp6_init_state(struct xfrm_state *x) { struct esp_data *esp = NULL; diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c index 6cde5310cd76..423feb46ccc0 100644 --- a/net/ipv6/ipcomp6.c +++ b/net/ipv6/ipcomp6.c @@ -234,14 +234,9 @@ static struct xfrm_state *ipcomp6_tunnel_create(struct xfrm_state *x) t->props.mode = 1; memcpy(t->props.saddr.a6, x->props.saddr.a6, sizeof(struct in6_addr)); - t->type = xfrm_get_type(IPPROTO_IPV6, t->props.family); - if (t->type == NULL) + if (xfrm_init_state(t)) goto error; - if (t->type->init_state(t, NULL)) - goto error; - - t->km.state = XFRM_STATE_VALID; atomic_set(&t->tunnel_users, 1); out: @@ -420,7 +415,7 @@ static void ipcomp6_destroy(struct xfrm_state *x) xfrm6_tunnel_free_spi((xfrm_address_t *)&x->props.saddr); } -static int ipcomp6_init_state(struct xfrm_state *x, void *args) +static int ipcomp6_init_state(struct xfrm_state *x) { int err; struct ipcomp_data *ipcd; diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c index ffcadd68b951..60c26c87277e 100644 --- a/net/ipv6/xfrm6_tunnel.c +++ b/net/ipv6/xfrm6_tunnel.c @@ -466,7 +466,7 @@ static void xfrm6_tunnel_err(struct sk_buff *skb, struct inet6_skb_parm *opt, return; } -static int xfrm6_tunnel_init_state(struct xfrm_state *x, void *args) +static int xfrm6_tunnel_init_state(struct xfrm_state *x) { if (!x->props.mode) return -EINVAL; diff --git a/net/key/af_key.c b/net/key/af_key.c index 98b72f2024ff..652dd09ccd3a 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -1096,17 +1096,11 @@ static struct xfrm_state * pfkey_msg2xfrm_state(struct sadb_msg *hdr, } } - x->type = xfrm_get_type(proto, x->props.family); - if (x->type == NULL) { - err = -ENOPROTOOPT; - goto out; - } - if (x->type->init_state(x, NULL)) { - err = -EINVAL; + err = xfrm_init_state(x); + if (err) goto out; - } + x->km.seq = hdr->sadb_msg_seq; - x->km.state = XFRM_STATE_VALID; return x; out: diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 0a4260719a12..d65ed8684fc1 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -118,7 +118,6 @@ retry: xfrm_policy_put_afinfo(afinfo); return type; } -EXPORT_SYMBOL(xfrm_get_type); int xfrm_dst_lookup(struct xfrm_dst **dst, struct flowi *fl, unsigned short family) diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 2537f26f097c..1845b73d69f9 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -1055,6 +1055,27 @@ int xfrm_state_mtu(struct xfrm_state *x, int mtu) } EXPORT_SYMBOL(xfrm_state_mtu); + +int xfrm_init_state(struct xfrm_state *x) +{ + int err; + + err = -ENOENT; + x->type = xfrm_get_type(x->id.proto, x->props.family); + if (x->type == NULL) + goto error; + + err = x->type->init_state(x); + if (err) + goto error; + + x->km.state = XFRM_STATE_VALID; + +error: + return err; +} + +EXPORT_SYMBOL(xfrm_init_state); void __init xfrm_state_init(void) { diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 5ce8558eac91..ecade4893a13 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -249,17 +249,10 @@ static struct xfrm_state *xfrm_state_construct(struct xfrm_usersa_info *p, if ((err = attach_encap_tmpl(&x->encap, xfrma[XFRMA_ENCAP-1]))) goto error; - err = -ENOENT; - x->type = xfrm_get_type(x->id.proto, x->props.family); - if (x->type == NULL) - goto error; - - err = x->type->init_state(x, NULL); + err = xfrm_init_state(x); if (err) goto error; - x->curlft.add_time = (unsigned long) xtime.tv_sec; - x->km.state = XFRM_STATE_VALID; x->km.seq = p->seq; return x; -- cgit v1.2.3 From d094cd83c06e06e01d8edb540555f3f64e4081c2 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Mon, 20 Jun 2005 13:19:41 -0700 Subject: [IPSEC]: Add xfrm_state_afinfo->init_flags This patch adds the xfrm_state_afinfo->init_flags hook which allows each address family to perform any common initialisation that does not require a corresponding destructor call. It will be used subsequently to set the XFRM_STATE_NOPMTUDISC flag in IPv4. It also fixes up the error codes returned by xfrm_init_state. Signed-off-by: Herbert Xu Acked-by: James Morris Signed-off-by: David S. Miller --- include/net/xfrm.h | 1 + net/xfrm/xfrm_state.c | 20 ++++++++++++++++++-- 2 files changed, 19 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 77bfdde440f8..029522a4ceda 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -204,6 +204,7 @@ struct xfrm_state_afinfo { rwlock_t lock; struct list_head *state_bydst; struct list_head *state_byspi; + int (*init_flags)(struct xfrm_state *x); void (*init_tempsel)(struct xfrm_state *x, struct flowi *fl, struct xfrm_tmpl *tmpl, xfrm_address_t *daddr, xfrm_address_t *saddr); diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 1845b73d69f9..9d206c282cf1 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -1058,10 +1058,26 @@ EXPORT_SYMBOL(xfrm_state_mtu); int xfrm_init_state(struct xfrm_state *x) { + struct xfrm_state_afinfo *afinfo; + int family = x->props.family; int err; - err = -ENOENT; - x->type = xfrm_get_type(x->id.proto, x->props.family); + err = -EAFNOSUPPORT; + afinfo = xfrm_state_get_afinfo(family); + if (!afinfo) + goto error; + + err = 0; + if (afinfo->init_flags) + err = afinfo->init_flags(x); + + xfrm_state_put_afinfo(afinfo); + + if (err) + goto error; + + err = -EPROTONOSUPPORT; + x->type = xfrm_get_type(x->id.proto, family); if (x->type == NULL) goto error; -- cgit v1.2.3 From dd87147eed934eaff92869f3d158697c7239d1d2 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Mon, 20 Jun 2005 13:21:43 -0700 Subject: [IPSEC]: Add XFRM_STATE_NOPMTUDISC flag This patch adds the flag XFRM_STATE_NOPMTUDISC for xfrm states. It is similar to the nopmtudisc on IPIP/GRE tunnels. It only has an effect on IPv4 tunnel mode states. For these states, it will ensure that the DF flag is always cleared. This is primarily useful to work around ICMP blackholes. In future this flag could also allow a larger MTU to be set within the tunnel just like IPIP/GRE tunnels. This could be useful for short haul tunnels where temporary fragmentation outside the tunnel is desired over smaller fragments inside the tunnel. Signed-off-by: Herbert Xu Acked-by: James Morris Signed-off-by: David S. Miller --- include/linux/pfkeyv2.h | 1 + include/linux/xfrm.h | 1 + net/ipv4/xfrm4_output.c | 8 ++++++-- net/ipv4/xfrm4_state.c | 9 +++++++++ net/key/af_key.c | 4 ++++ 5 files changed, 21 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/pfkeyv2.h b/include/linux/pfkeyv2.h index e6b519220245..724066778aff 100644 --- a/include/linux/pfkeyv2.h +++ b/include/linux/pfkeyv2.h @@ -245,6 +245,7 @@ struct sadb_x_nat_t_port { /* Security Association flags */ #define SADB_SAFLAGS_PFS 1 +#define SADB_SAFLAGS_NOPMTUDISC 0x20000000 #define SADB_SAFLAGS_DECAP_DSCP 0x40000000 #define SADB_SAFLAGS_NOECN 0x80000000 diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index d68391a9b9f3..f0d423300d84 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -196,6 +196,7 @@ struct xfrm_usersa_info { __u8 flags; #define XFRM_STATE_NOECN 1 #define XFRM_STATE_DECAP_DSCP 2 +#define XFRM_STATE_NOPMTUDISC 4 }; struct xfrm_usersa_id { diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c index af2392ae5769..66620a95942a 100644 --- a/net/ipv4/xfrm4_output.c +++ b/net/ipv4/xfrm4_output.c @@ -33,6 +33,7 @@ static void xfrm4_encap(struct sk_buff *skb) struct dst_entry *dst = skb->dst; struct xfrm_state *x = dst->xfrm; struct iphdr *iph, *top_iph; + int flags; iph = skb->nh.iph; skb->h.ipiph = iph; @@ -51,10 +52,13 @@ static void xfrm4_encap(struct sk_buff *skb) /* DS disclosed */ top_iph->tos = INET_ECN_encapsulate(iph->tos, iph->tos); - if (x->props.flags & XFRM_STATE_NOECN) + + flags = x->props.flags; + if (flags & XFRM_STATE_NOECN) IP_ECN_clear(top_iph); - top_iph->frag_off = iph->frag_off & htons(IP_DF); + top_iph->frag_off = (flags & XFRM_STATE_NOPMTUDISC) ? + 0 : (iph->frag_off & htons(IP_DF)); if (!top_iph->frag_off) __ip_select_ident(top_iph, dst, 0); diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c index 223a2e83853f..050611d7a967 100644 --- a/net/ipv4/xfrm4_state.c +++ b/net/ipv4/xfrm4_state.c @@ -7,12 +7,20 @@ * */ +#include #include #include #include static struct xfrm_state_afinfo xfrm4_state_afinfo; +static int xfrm4_init_flags(struct xfrm_state *x) +{ + if (ipv4_config.no_pmtu_disc) + x->props.flags |= XFRM_STATE_NOPMTUDISC; + return 0; +} + static void __xfrm4_init_tempsel(struct xfrm_state *x, struct flowi *fl, struct xfrm_tmpl *tmpl, @@ -109,6 +117,7 @@ __xfrm4_find_acq(u8 mode, u32 reqid, u8 proto, static struct xfrm_state_afinfo xfrm4_state_afinfo = { .family = AF_INET, .lock = RW_LOCK_UNLOCKED, + .init_flags = xfrm4_init_flags, .init_tempsel = __xfrm4_init_tempsel, .state_lookup = __xfrm4_state_lookup, .find_acq = __xfrm4_find_acq, diff --git a/net/key/af_key.c b/net/key/af_key.c index 652dd09ccd3a..4879743b945a 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -690,6 +690,8 @@ static struct sk_buff * pfkey_xfrm_state2msg(struct xfrm_state *x, int add_keys, sa->sadb_sa_flags |= SADB_SAFLAGS_NOECN; if (x->props.flags & XFRM_STATE_DECAP_DSCP) sa->sadb_sa_flags |= SADB_SAFLAGS_DECAP_DSCP; + if (x->props.flags & XFRM_STATE_NOPMTUDISC) + sa->sadb_sa_flags |= SADB_SAFLAGS_NOPMTUDISC; /* hard time */ if (hsc & 2) { @@ -974,6 +976,8 @@ static struct xfrm_state * pfkey_msg2xfrm_state(struct sadb_msg *hdr, x->props.flags |= XFRM_STATE_NOECN; if (sa->sadb_sa_flags & SADB_SAFLAGS_DECAP_DSCP) x->props.flags |= XFRM_STATE_DECAP_DSCP; + if (sa->sadb_sa_flags & SADB_SAFLAGS_NOPMTUDISC) + x->props.flags |= XFRM_STATE_NOPMTUDISC; lifetime = (struct sadb_lifetime*) ext_hdrs[SADB_EXT_LIFETIME_HARD-1]; if (lifetime != NULL) { -- cgit v1.2.3 From f852640e74f71e6dd38146e1149ec1fe6da2fb07 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 20 Jun 2005 13:31:11 -0700 Subject: [AX25]: endian-annotate ax25_type_trans() Signed-off-by: Alexey Dobriyan Acked-by: Ralf Baechle Signed-off-by: David S. Miller --- include/net/ax25.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/ax25.h b/include/net/ax25.h index 9e6368a54547..828a3a93dda1 100644 --- a/include/net/ax25.h +++ b/include/net/ax25.h @@ -220,7 +220,7 @@ static __inline__ void ax25_cb_put(ax25_cb *ax25) } } -static inline unsigned short ax25_type_trans(struct sk_buff *skb, struct net_device *dev) +static inline __be16 ax25_type_trans(struct sk_buff *skb, struct net_device *dev) { skb->dev = dev; skb->pkt_type = PACKET_HOST; -- cgit v1.2.3 From f6e276ee67c0ac9efafd24bc6f7a84aa359656df Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 20 Jun 2005 13:32:05 -0700 Subject: [ATALK]: endian annotations Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/linux/atalk.h | 26 +++++++++++++------------- net/appletalk/aarp.c | 2 +- net/appletalk/ddp.c | 2 +- 3 files changed, 15 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/linux/atalk.h b/include/linux/atalk.h index 31d3fc25ccbd..09a1451c1159 100644 --- a/include/linux/atalk.h +++ b/include/linux/atalk.h @@ -20,7 +20,7 @@ #define SIOCATALKDIFADDR (SIOCPROTOPRIVATE + 0) struct atalk_addr { - __u16 s_net; + __be16 s_net; __u8 s_node; }; @@ -33,8 +33,8 @@ struct sockaddr_at { struct atalk_netrange { __u8 nr_phase; - __u16 nr_firstnet; - __u16 nr_lastnet; + __be16 nr_firstnet; + __be16 nr_lastnet; }; #ifdef __KERNEL__ @@ -70,8 +70,8 @@ struct atalk_iface { struct atalk_sock { /* struct sock has to be the first member of atalk_sock */ struct sock sk; - unsigned short dest_net; - unsigned short src_net; + __be16 dest_net; + __be16 src_net; unsigned char dest_node; unsigned char src_node; unsigned char dest_port; @@ -95,9 +95,9 @@ struct ddpehdr { deh_hops:4, deh_len:10; #endif - __u16 deh_sum; - __u16 deh_dnet; - __u16 deh_snet; + __be16 deh_sum; + __be16 deh_dnet; + __be16 deh_snet; __u8 deh_dnode; __u8 deh_snode; __u8 deh_dport; @@ -142,24 +142,24 @@ struct ddpshdr { /* AppleTalk AARP headers */ struct elapaarp { - __u16 hw_type; + __be16 hw_type; #define AARP_HW_TYPE_ETHERNET 1 #define AARP_HW_TYPE_TOKENRING 2 - __u16 pa_type; + __be16 pa_type; __u8 hw_len; __u8 pa_len; #define AARP_PA_ALEN 4 - __u16 function; + __be16 function; #define AARP_REQUEST 1 #define AARP_REPLY 2 #define AARP_PROBE 3 __u8 hw_src[ETH_ALEN] __attribute__ ((packed)); __u8 pa_src_zero __attribute__ ((packed)); - __u16 pa_src_net __attribute__ ((packed)); + __be16 pa_src_net __attribute__ ((packed)); __u8 pa_src_node __attribute__ ((packed)); __u8 hw_dst[ETH_ALEN] __attribute__ ((packed)); __u8 pa_dst_zero __attribute__ ((packed)); - __u16 pa_dst_net __attribute__ ((packed)); + __be16 pa_dst_net __attribute__ ((packed)); __u8 pa_dst_node __attribute__ ((packed)); }; diff --git a/net/appletalk/aarp.c b/net/appletalk/aarp.c index 54640c01b50c..10d040461021 100644 --- a/net/appletalk/aarp.c +++ b/net/appletalk/aarp.c @@ -565,7 +565,7 @@ int aarp_send_ddp(struct net_device *dev, struct sk_buff *skb, * numbers we just happen to need. Now put the * length in the lower two. */ - *((__u16 *)skb->data) = htons(skb->len); + *((__be16 *)skb->data) = htons(skb->len); ft = 1; } /* diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c index 876dbac71060..192b529f86a4 100644 --- a/net/appletalk/ddp.c +++ b/net/appletalk/ddp.c @@ -401,7 +401,7 @@ out_err: } /* Find a match for a specific network:node pair */ -static struct atalk_iface *atalk_find_interface(int net, int node) +static struct atalk_iface *atalk_find_interface(__be16 net, int node) { struct atalk_iface *iface; -- cgit v1.2.3 From 246955fe4c38bd706ae30e37c64892c94213775d Mon Sep 17 00:00:00 2001 From: Robert Olsson Date: Mon, 20 Jun 2005 13:36:39 -0700 Subject: [NETLINK]: fib_lookup() via netlink Below is a more generic patch to do fib_lookup via netlink. For others we should say that we discussed this as a way to verify route selection. It's also possible there are others uses for this. In short the fist half of struct fib_result_nl is filled in by caller and netlink call fills in the other half and returns it. In case anyone is interested there is a corresponding user app to compare the full routing table this was used to test implementation of the LC-trie. Signed-off-by: David S. Miller --- include/linux/netlink.h | 1 + include/net/ip_fib.h | 14 +++++++++++++ net/ipv4/fib_frontend.c | 55 +++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 70 insertions(+) (limited to 'include') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index e38407a23d04..561d4dc75836 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -14,6 +14,7 @@ #define NETLINK_SELINUX 7 /* SELinux event notifications */ #define NETLINK_ARPD 8 #define NETLINK_AUDIT 9 /* auditing */ +#define NETLINK_FIB_LOOKUP 10 #define NETLINK_ROUTE6 11 /* af_inet6 route comm channel */ #define NETLINK_IP6_FW 13 #define NETLINK_DNRTMSG 14 /* DECnet routing messages */ diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index e5a5f6b62f88..a4208a336ac0 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -109,6 +109,20 @@ struct fib_result { #endif }; +struct fib_result_nl { + u32 fl_addr; /* To be looked up*/ + u32 fl_fwmark; + unsigned char fl_tos; + unsigned char fl_scope; + unsigned char tb_id_in; + + unsigned char tb_id; /* Results */ + unsigned char prefixlen; + unsigned char nh_sel; + unsigned char type; + unsigned char scope; + int err; +}; #ifdef CONFIG_IP_ROUTE_MULTIPATH diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 563e7d612706..cd8e45ab9580 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -516,6 +516,60 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa) #undef BRD1_OK } +static void nl_fib_lookup(struct fib_result_nl *frn, struct fib_table *tb ) +{ + + struct fib_result res; + struct flowi fl = { .nl_u = { .ip4_u = { .daddr = frn->fl_addr, + .fwmark = frn->fl_fwmark, + .tos = frn->fl_tos, + .scope = frn->fl_scope } } }; + if (tb) { + local_bh_disable(); + + frn->tb_id = tb->tb_id; + frn->err = tb->tb_lookup(tb, &fl, &res); + + if (!frn->err) { + frn->prefixlen = res.prefixlen; + frn->nh_sel = res.nh_sel; + frn->type = res.type; + frn->scope = res.scope; + } + local_bh_enable(); + } +} + +static void nl_fib_input(struct sock *sk, int len) +{ + struct sk_buff *skb = NULL; + struct nlmsghdr *nlh = NULL; + struct fib_result_nl *frn; + int err; + u32 pid; + struct fib_table *tb; + + skb = skb_recv_datagram(sk, 0, 0, &err); + nlh = (struct nlmsghdr *)skb->data; + + frn = (struct fib_result_nl *) NLMSG_DATA(nlh); + tb = fib_get_table(frn->tb_id_in); + + nl_fib_lookup(frn, tb); + + pid = nlh->nlmsg_pid; /*pid of sending process */ + NETLINK_CB(skb).groups = 0; /* not in mcast group */ + NETLINK_CB(skb).pid = 0; /* from kernel */ + NETLINK_CB(skb).dst_pid = pid; + NETLINK_CB(skb).dst_groups = 0; /* unicast */ + netlink_unicast(sk, skb, pid, MSG_DONTWAIT); +} + +static void nl_fib_lookup_init(void) +{ + netlink_kernel_create(NETLINK_FIB_LOOKUP, nl_fib_input); +} + static void fib_disable_ip(struct net_device *dev, int force) { if (fib_sync_down(0, dev, force)) @@ -604,6 +658,7 @@ void __init ip_fib_init(void) register_netdevice_notifier(&fib_netdev_notifier); register_inetaddr_notifier(&fib_inetaddr_notifier); + nl_fib_lookup_init(); } EXPORT_SYMBOL(inet_addr_type); -- cgit v1.2.3 From e3a15db2415579d5136b9ba9b52fe27c66da8780 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Tue, 26 Apr 2005 02:31:08 -0500 Subject: [PATCH] sysfs_{create|remove}_link should take const char * sysfs: make sysfs_{create|remove}_link to take const char * name. Signed-off-by: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- fs/sysfs/symlink.c | 8 ++++---- include/linux/sysfs.h | 10 +++++----- 2 files changed, 9 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/fs/sysfs/symlink.c b/fs/sysfs/symlink.c index dfdf70174354..fae57c83a722 100644 --- a/fs/sysfs/symlink.c +++ b/fs/sysfs/symlink.c @@ -43,7 +43,7 @@ static void fill_object_path(struct kobject * kobj, char * buffer, int length) } } -static int sysfs_add_link(struct dentry * parent, char * name, struct kobject * target) +static int sysfs_add_link(struct dentry * parent, const char * name, struct kobject * target) { struct sysfs_dirent * parent_sd = parent->d_fsdata; struct sysfs_symlink * sl; @@ -79,7 +79,7 @@ exit1: * @target: object we're pointing to. * @name: name of the symlink. */ -int sysfs_create_link(struct kobject * kobj, struct kobject * target, char * name) +int sysfs_create_link(struct kobject * kobj, struct kobject * target, const char * name) { struct dentry * dentry = kobj->dentry; int error = 0; @@ -99,13 +99,13 @@ int sysfs_create_link(struct kobject * kobj, struct kobject * target, char * nam * @name: name of the symlink to remove. */ -void sysfs_remove_link(struct kobject * kobj, char * name) +void sysfs_remove_link(struct kobject * kobj, const char * name) { sysfs_hash_and_remove(kobj->dentry,name); } static int sysfs_get_target_path(struct kobject * kobj, struct kobject * target, - char *path) + char *path) { char * s; int depth, size; diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h index 38b58b30814a..931b5aaaf380 100644 --- a/include/linux/sysfs.h +++ b/include/linux/sysfs.h @@ -105,11 +105,11 @@ sysfs_chmod_file(struct kobject *kobj, struct attribute *attr, mode_t mode); extern void sysfs_remove_file(struct kobject *, const struct attribute *); -extern int -sysfs_create_link(struct kobject * kobj, struct kobject * target, char * name); +extern int +sysfs_create_link(struct kobject * kobj, struct kobject * target, const char * name); extern void -sysfs_remove_link(struct kobject *, char * name); +sysfs_remove_link(struct kobject *, const char * name); int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr); int sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr); @@ -153,12 +153,12 @@ static inline void sysfs_remove_file(struct kobject * k, const struct attribute ; } -static inline int sysfs_create_link(struct kobject * k, struct kobject * t, char * n) +static inline int sysfs_create_link(struct kobject * k, struct kobject * t, const char * n) { return 0; } -static inline void sysfs_remove_link(struct kobject * k, char * name) +static inline void sysfs_remove_link(struct kobject * k, const char * name) { ; } -- cgit v1.2.3 From f3b4f3c6dec04c6c8261fe22645f07b39976595a Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Tue, 26 Apr 2005 02:32:00 -0500 Subject: [PATCH] Make kobject's name be const char * kobject: make kobject's name const char * since users should not attempt to change it (except by calling kobject_rename). Signed-off-by: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- include/linux/kobject.h | 6 +++--- lib/kobject.c | 2 +- lib/kobject_uevent.c | 4 ++-- 3 files changed, 6 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/kobject.h b/include/linux/kobject.h index 765d660d3bea..76dc67245c0c 100644 --- a/include/linux/kobject.h +++ b/include/linux/kobject.h @@ -33,7 +33,7 @@ extern u64 hotplug_seqnum; struct kobject { - char * k_name; + const char * k_name; char name[KOBJ_NAME_LEN]; struct kref kref; struct list_head entry; @@ -46,7 +46,7 @@ struct kobject { extern int kobject_set_name(struct kobject *, const char *, ...) __attribute__((format(printf,2,3))); -static inline char * kobject_name(struct kobject * kobj) +static inline const char * kobject_name(const struct kobject * kobj) { return kobj->k_name; } @@ -57,7 +57,7 @@ extern void kobject_cleanup(struct kobject *); extern int kobject_add(struct kobject *); extern void kobject_del(struct kobject *); -extern int kobject_rename(struct kobject *, char *new_name); +extern int kobject_rename(struct kobject *, const char *new_name); extern int kobject_register(struct kobject *); extern void kobject_unregister(struct kobject *); diff --git a/lib/kobject.c b/lib/kobject.c index 94048826624c..dd0917dd9fa9 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -279,7 +279,7 @@ EXPORT_SYMBOL(kobject_set_name); * @new_name: object's new name */ -int kobject_rename(struct kobject * kobj, char *new_name) +int kobject_rename(struct kobject * kobj, const char *new_name) { int error = 0; diff --git a/lib/kobject_uevent.c b/lib/kobject_uevent.c index 1ede5aa33376..8e49d21057e4 100644 --- a/lib/kobject_uevent.c +++ b/lib/kobject_uevent.c @@ -197,7 +197,7 @@ void kobject_hotplug(struct kobject *kobj, enum kobject_action action) int i = 0; int retval; char *kobj_path = NULL; - char *name = NULL; + const char *name = NULL; char *action_string; u64 seq; struct kobject *top_kobj = kobj; @@ -249,7 +249,7 @@ void kobject_hotplug(struct kobject *kobj, enum kobject_action action) name = kobject_name(&kset->kobj); argv [0] = hotplug_path; - argv [1] = name; + argv [1] = (char *)name; /* won't be changed but 'const' has to go */ argv [2] = NULL; /* minimal command environment */ -- cgit v1.2.3 From 419cab3fc69588ebe35b845cc3a584ae172463de Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Tue, 26 Apr 2005 02:32:54 -0500 Subject: [PATCH] kset_hotplug_ops->name shoudl return const char * kobject: change name() method in kset_hotplug_ops return const char * since users shoudl not try to modify returned data. Signed-off-by: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- drivers/base/class.c | 2 +- drivers/base/core.c | 2 +- include/linux/kobject.h | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/drivers/base/class.c b/drivers/base/class.c index d2a2f8f2b4ed..ff33ebbf1dc7 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -262,7 +262,7 @@ static int class_hotplug_filter(struct kset *kset, struct kobject *kobj) return 0; } -static char *class_hotplug_name(struct kset *kset, struct kobject *kobj) +static const char *class_hotplug_name(struct kset *kset, struct kobject *kobj) { struct class_device *class_dev = to_class_dev(kobj); diff --git a/drivers/base/core.c b/drivers/base/core.c index fbc223486f81..32f3f09a42e0 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -102,7 +102,7 @@ static int dev_hotplug_filter(struct kset *kset, struct kobject *kobj) return 0; } -static char *dev_hotplug_name(struct kset *kset, struct kobject *kobj) +static const char *dev_hotplug_name(struct kset *kset, struct kobject *kobj) { struct device *dev = to_dev(kobj); diff --git a/include/linux/kobject.h b/include/linux/kobject.h index 76dc67245c0c..3b22304f12fd 100644 --- a/include/linux/kobject.h +++ b/include/linux/kobject.h @@ -94,7 +94,7 @@ struct kobj_type { */ struct kset_hotplug_ops { int (*filter)(struct kset *kset, struct kobject *kobj); - char *(*name)(struct kset *kset, struct kobject *kobj); + const char *(*name)(struct kset *kset, struct kobject *kobj); int (*hotplug)(struct kset *kset, struct kobject *kobj, char **envp, int num_envp, char *buffer, int buffer_size); }; -- cgit v1.2.3 From 8d790d74085833ba2a3e84b5bcd683be4981c29a Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Tue, 26 Apr 2005 02:34:05 -0500 Subject: [PATCH] make driver's name be const char * Driver core: change driver's, bus's, class's and platform device's names to be const char * so one can use const char *drv_name = "asdfg"; when initializing structures. Also kill couple of whitespaces. Signed-off-by: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/devices.c | 2 +- include/linux/device.h | 12 ++++++------ 2 files changed, 7 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/drivers/usb/core/devices.c b/drivers/usb/core/devices.c index ef0b35731ff0..83e815d3cd52 100644 --- a/drivers/usb/core/devices.c +++ b/drivers/usb/core/devices.c @@ -239,7 +239,7 @@ static char *usb_dump_interface_descriptor(char *start, char *end, int setno) { const struct usb_interface_descriptor *desc = &intfc->altsetting[setno].desc; - char *driver_name = ""; + const char *driver_name = ""; if (start > end) return start; diff --git a/include/linux/device.h b/include/linux/device.h index df94c0de53f2..fa9e6ca08f5a 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -47,7 +47,7 @@ struct class_device; struct class_simple; struct bus_type { - char * name; + const char * name; struct subsystem subsys; struct kset drivers; @@ -98,17 +98,17 @@ extern int bus_create_file(struct bus_type *, struct bus_attribute *); extern void bus_remove_file(struct bus_type *, struct bus_attribute *); struct device_driver { - char * name; + const char * name; struct bus_type * bus; struct completion unloaded; struct kobject kobj; struct list_head devices; - struct module * owner; + struct module * owner; int (*probe) (struct device * dev); - int (*remove) (struct device * dev); + int (*remove) (struct device * dev); void (*shutdown) (struct device * dev); int (*suspend) (struct device * dev, pm_message_t state, u32 level); int (*resume) (struct device * dev, u32 level); @@ -142,7 +142,7 @@ extern void driver_remove_file(struct device_driver *, struct driver_attribute * * device classes */ struct class { - char * name; + const char * name; struct subsystem subsys; struct list_head children; @@ -366,7 +366,7 @@ extern struct device *device_find(const char *name, struct bus_type *bus); /* drivers/base/platform.c */ struct platform_device { - char * name; + const char * name; u32 id; struct device dev; u32 num_resources; -- cgit v1.2.3 From d48593bf208e0d046c35fb0707ae5b23fef8c4ff Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Fri, 29 Apr 2005 00:58:46 -0500 Subject: [PATCH] Make attributes names const char * sysfs: make attributes and attribute_group's names const char * Signed-off-by: Dmitry Torokhov Signed-off-by: Greg Kroah-Hartman --- drivers/infiniband/core/sysfs.c | 120 ++++++++++++++++------------------------ drivers/pci/pci-sysfs.c | 13 +++-- include/linux/sysfs.h | 4 +- 3 files changed, 58 insertions(+), 79 deletions(-) (limited to 'include') diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c index 3a413f72ff6d..5febd6d8b885 100644 --- a/drivers/infiniband/core/sysfs.c +++ b/drivers/infiniband/core/sysfs.c @@ -40,9 +40,7 @@ struct ib_port { struct kobject kobj; struct ib_device *ibdev; struct attribute_group gid_group; - struct attribute **gid_attr; struct attribute_group pkey_group; - struct attribute **pkey_attr; u8 port_num; }; @@ -60,8 +58,9 @@ struct port_attribute port_attr_##_name = __ATTR(_name, _mode, _show, _store) struct port_attribute port_attr_##_name = __ATTR_RO(_name) struct port_table_attribute { - struct port_attribute attr; - int index; + struct port_attribute attr; + char name[8]; + int index; }; static ssize_t port_attr_show(struct kobject *kobj, @@ -398,17 +397,16 @@ static void ib_port_release(struct kobject *kobj) struct attribute *a; int i; - for (i = 0; (a = p->gid_attr[i]); ++i) { - kfree(a->name); + for (i = 0; (a = p->gid_group.attrs[i]); ++i) kfree(a); - } - for (i = 0; (a = p->pkey_attr[i]); ++i) { - kfree(a->name); + kfree(p->gid_group.attrs); + + for (i = 0; (a = p->pkey_group.attrs[i]); ++i) kfree(a); - } - kfree(p->gid_attr); + kfree(p->pkey_group.attrs); + kfree(p); } @@ -449,58 +447,45 @@ static int ib_device_hotplug(struct class_device *cdev, char **envp, return 0; } -static int alloc_group(struct attribute ***attr, - ssize_t (*show)(struct ib_port *, - struct port_attribute *, char *buf), - int len) +static struct attribute ** +alloc_group_attrs(ssize_t (*show)(struct ib_port *, + struct port_attribute *, char *buf), + int len) { - struct port_table_attribute ***tab_attr = - (struct port_table_attribute ***) attr; + struct attribute **tab_attr; + struct port_table_attribute *element; int i; - int ret; - - *tab_attr = kmalloc((1 + len) * sizeof *tab_attr, GFP_KERNEL); - if (!*tab_attr) - return -ENOMEM; - memset(*tab_attr, 0, (1 + len) * sizeof *tab_attr); + tab_attr = kcalloc(1 + len, sizeof(struct attribute *), GFP_KERNEL); + if (!tab_attr) + return NULL; - for (i = 0; i < len; ++i) { - (*tab_attr)[i] = kmalloc(sizeof *(*tab_attr)[i], GFP_KERNEL); - if (!(*tab_attr)[i]) { - ret = -ENOMEM; + for (i = 0; i < len; i++) { + element = kcalloc(1, sizeof(struct port_table_attribute), + GFP_KERNEL); + if (!element) goto err; - } - memset((*tab_attr)[i], 0, sizeof *(*tab_attr)[i]); - (*tab_attr)[i]->attr.attr.name = kmalloc(8, GFP_KERNEL); - if (!(*tab_attr)[i]->attr.attr.name) { - ret = -ENOMEM; - goto err; - } - if (snprintf((*tab_attr)[i]->attr.attr.name, 8, "%d", i) >= 8) { - ret = -ENOMEM; + if (snprintf(element->name, sizeof(element->name), + "%d", i) >= sizeof(element->name)) goto err; - } - (*tab_attr)[i]->attr.attr.mode = S_IRUGO; - (*tab_attr)[i]->attr.attr.owner = THIS_MODULE; - (*tab_attr)[i]->attr.show = show; - (*tab_attr)[i]->index = i; - } - - return 0; + element->attr.attr.name = element->name; + element->attr.attr.mode = S_IRUGO; + element->attr.attr.owner = THIS_MODULE; + element->attr.show = show; + element->index = i; -err: - for (i = 0; i < len; ++i) { - if ((*tab_attr)[i]) - kfree((*tab_attr)[i]->attr.attr.name); - kfree((*tab_attr)[i]); + tab_attr[i] = &element->attr.attr; } - kfree(*tab_attr); + return tab_attr; - return ret; +err: + while (--i >= 0) + kfree(tab_attr[i]); + kfree(tab_attr); + return NULL; } static int add_port(struct ib_device *device, int port_num) @@ -541,23 +526,20 @@ static int add_port(struct ib_device *device, int port_num) if (ret) goto err_put; - ret = alloc_group(&p->gid_attr, show_port_gid, attr.gid_tbl_len); - if (ret) - goto err_remove_pma; - p->gid_group.name = "gids"; - p->gid_group.attrs = p->gid_attr; + p->gid_group.attrs = alloc_group_attrs(show_port_gid, attr.gid_tbl_len); + if (!p->gid_group.attrs) + goto err_remove_pma; ret = sysfs_create_group(&p->kobj, &p->gid_group); if (ret) goto err_free_gid; - ret = alloc_group(&p->pkey_attr, show_port_pkey, attr.pkey_tbl_len); - if (ret) - goto err_remove_gid; - p->pkey_group.name = "pkeys"; - p->pkey_group.attrs = p->pkey_attr; + p->pkey_group.attrs = alloc_group_attrs(show_port_pkey, + attr.pkey_tbl_len); + if (!p->pkey_group.attrs) + goto err_remove_gid; ret = sysfs_create_group(&p->kobj, &p->pkey_group); if (ret) @@ -568,23 +550,19 @@ static int add_port(struct ib_device *device, int port_num) return 0; err_free_pkey: - for (i = 0; i < attr.pkey_tbl_len; ++i) { - kfree(p->pkey_attr[i]->name); - kfree(p->pkey_attr[i]); - } + for (i = 0; i < attr.pkey_tbl_len; ++i) + kfree(p->pkey_group.attrs[i]); - kfree(p->pkey_attr); + kfree(p->pkey_group.attrs); err_remove_gid: sysfs_remove_group(&p->kobj, &p->gid_group); err_free_gid: - for (i = 0; i < attr.gid_tbl_len; ++i) { - kfree(p->gid_attr[i]->name); - kfree(p->gid_attr[i]); - } + for (i = 0; i < attr.gid_tbl_len; ++i) + kfree(p->gid_group.attrs[i]); - kfree(p->gid_attr); + kfree(p->gid_group.attrs); err_remove_pma: sysfs_remove_group(&p->kobj, &pma_group); diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index 6ca0061137a6..e8aad151175f 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -339,16 +339,17 @@ pci_create_resource_files(struct pci_dev *pdev) if (!pci_resource_len(pdev, i)) continue; - res_attr = kmalloc(sizeof(*res_attr) + 10, GFP_ATOMIC); + /* allocate attribute structure, piggyback attribute name */ + res_attr = kcalloc(1, sizeof(*res_attr) + 10, GFP_ATOMIC); if (res_attr) { - memset(res_attr, 0, sizeof(*res_attr) + 10); + char *res_attr_name = (char *)(res_attr + 1); + pdev->res_attr[i] = res_attr; - /* Allocated above after the res_attr struct */ - res_attr->attr.name = (char *)(res_attr + 1); - sprintf(res_attr->attr.name, "resource%d", i); - res_attr->size = pci_resource_len(pdev, i); + sprintf(res_attr_name, "resource%d", i); + res_attr->attr.name = res_attr_name; res_attr->attr.mode = S_IRUSR | S_IWUSR; res_attr->attr.owner = THIS_MODULE; + res_attr->size = pci_resource_len(pdev, i); res_attr->mmap = pci_mmap_resource; res_attr->private = &pdev->resource[i]; sysfs_create_bin_file(&pdev->dev.kobj, res_attr); diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h index 931b5aaaf380..d9cd2d31d377 100644 --- a/include/linux/sysfs.h +++ b/include/linux/sysfs.h @@ -16,13 +16,13 @@ struct kobject; struct module; struct attribute { - char * name; + const char * name; struct module * owner; mode_t mode; }; struct attribute_group { - char * name; + const char * name; struct attribute ** attrs; }; -- cgit v1.2.3 From e9ba6365fd4f0d9e7d022c883bd044fbaa48257f Mon Sep 17 00:00:00 2001 From: "gregkh@suse.de" Date: Tue, 15 Mar 2005 11:54:21 -0800 Subject: [PATCH] CLASS: move a "simple" class logic into the class core. One step on improving the class api so that it can not be used incorrectly. This also fixes the module owner issue with the dev files that happened when the devt logic moved to the class core. Based on a patch originally written by Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/class.c | 145 +++++++++++++++++++++++++++++++++++++++++++++---- include/linux/device.h | 9 +++ 2 files changed, 143 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/drivers/base/class.c b/drivers/base/class.c index 344b8cd73901..a3b006b6f2ba 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -16,6 +16,7 @@ #include #include #include +#include #include "base.h" #define to_class_attr(_attr) container_of(_attr, struct class_attribute, attr) @@ -162,6 +163,51 @@ void class_unregister(struct class * cls) subsystem_unregister(&cls->subsys); } +static void class_create_release(struct class *cls) +{ + kfree(cls); +} + +static void class_device_create_release(struct class_device *class_dev) +{ + kfree(class_dev); +} + +struct class *class_create(struct module *owner, char *name) +{ + struct class *cls; + int retval; + + cls = kmalloc(sizeof(struct class), GFP_KERNEL); + if (!cls) { + retval = -ENOMEM; + goto error; + } + memset(cls, 0x00, sizeof(struct class)); + + cls->name = name; + cls->owner = owner; + cls->class_release = class_create_release; + cls->release = class_device_create_release; + + retval = class_register(cls); + if (retval) + goto error; + + return cls; + +error: + kfree(cls); + return ERR_PTR(retval); +} + +void class_destroy(struct class *cls) +{ + if ((cls == NULL) || (IS_ERR(cls))) + return; + + class_unregister(cls); +} /* Class Device Stuff */ @@ -375,7 +421,6 @@ static ssize_t show_dev(struct class_device *class_dev, char *buf) { return print_dev_t(buf, class_dev->devt); } -static CLASS_DEVICE_ATTR(dev, S_IRUGO, show_dev, NULL); void class_device_initialize(struct class_device *class_dev) { @@ -412,7 +457,31 @@ int class_device_add(struct class_device *class_dev) if ((error = kobject_add(&class_dev->kobj))) goto register_done; - /* now take care of our own registration */ + /* add the needed attributes to this device */ + if (MAJOR(class_dev->devt)) { + struct class_device_attribute *attr; + attr = kmalloc(sizeof(*attr), GFP_KERNEL); + if (!attr) { + error = -ENOMEM; + kobject_del(&class_dev->kobj); + goto register_done; + } + memset(attr, sizeof(*attr), 0x00); + attr->attr.name = "dev"; + attr->attr.mode = S_IRUGO; + attr->attr.owner = parent->owner; + attr->show = show_dev; + attr->store = NULL; + class_device_create_file(class_dev, attr); + class_dev->devt_attr = attr; + } + + class_device_add_attrs(class_dev); + if (class_dev->dev) + sysfs_create_link(&class_dev->kobj, + &class_dev->dev->kobj, "device"); + + /* notify any interfaces this device is now here */ if (parent) { down(&parent->sem); list_add_tail(&class_dev->node, &parent->children); @@ -421,16 +490,8 @@ int class_device_add(struct class_device *class_dev) class_intf->add(class_dev); up(&parent->sem); } - - if (MAJOR(class_dev->devt)) - class_device_create_file(class_dev, &class_device_attr_dev); - - class_device_add_attrs(class_dev); - if (class_dev->dev) - sysfs_create_link(&class_dev->kobj, - &class_dev->dev->kobj, "device"); - kobject_hotplug(&class_dev->kobj, KOBJ_ADD); + register_done: if (error && parent) class_put(parent); @@ -444,6 +505,41 @@ int class_device_register(struct class_device *class_dev) return class_device_add(class_dev); } +struct class_device *class_device_create(struct class *cls, dev_t devt, + struct device *device, char *fmt, ...) +{ + va_list args; + struct class_device *class_dev = NULL; + int retval = -ENODEV; + + if (cls == NULL || IS_ERR(cls)) + goto error; + + class_dev = kmalloc(sizeof(struct class_device), GFP_KERNEL); + if (!class_dev) { + retval = -ENOMEM; + goto error; + } + memset(class_dev, 0x00, sizeof(struct class_device)); + + class_dev->devt = devt; + class_dev->dev = device; + class_dev->class = cls; + + va_start(args, fmt); + vsnprintf(class_dev->class_id, BUS_ID_SIZE, fmt, args); + va_end(args); + retval = class_device_register(class_dev); + if (retval) + goto error; + + return class_dev; + +error: + kfree(class_dev); + return ERR_PTR(retval); +} + void class_device_del(struct class_device *class_dev) { struct class * parent = class_dev->class; @@ -460,6 +556,11 @@ void class_device_del(struct class_device *class_dev) if (class_dev->dev) sysfs_remove_link(&class_dev->kobj, "device"); + if (class_dev->devt_attr) { + class_device_remove_file(class_dev, class_dev->devt_attr); + kfree(class_dev->devt_attr); + class_dev->devt_attr = NULL; + } class_device_remove_attrs(class_dev); kobject_hotplug(&class_dev->kobj, KOBJ_REMOVE); @@ -477,6 +578,24 @@ void class_device_unregister(struct class_device *class_dev) class_device_put(class_dev); } +void class_device_destroy(struct class *cls, dev_t devt) +{ + struct class_device *class_dev = NULL; + struct class_device *class_dev_tmp; + + down(&cls->sem); + list_for_each_entry(class_dev_tmp, &cls->children, node) { + if (class_dev_tmp->devt == devt) { + class_dev = class_dev_tmp; + break; + } + } + up(&cls->sem); + + if (class_dev) + class_device_unregister(class_dev); +} + int class_device_rename(struct class_device *class_dev, char *new_name) { int error = 0; @@ -576,6 +695,8 @@ EXPORT_SYMBOL_GPL(class_register); EXPORT_SYMBOL_GPL(class_unregister); EXPORT_SYMBOL_GPL(class_get); EXPORT_SYMBOL_GPL(class_put); +EXPORT_SYMBOL_GPL(class_create); +EXPORT_SYMBOL_GPL(class_destroy); EXPORT_SYMBOL_GPL(class_device_register); EXPORT_SYMBOL_GPL(class_device_unregister); @@ -584,6 +705,8 @@ EXPORT_SYMBOL_GPL(class_device_add); EXPORT_SYMBOL_GPL(class_device_del); EXPORT_SYMBOL_GPL(class_device_get); EXPORT_SYMBOL_GPL(class_device_put); +EXPORT_SYMBOL_GPL(class_device_create); +EXPORT_SYMBOL_GPL(class_device_destroy); EXPORT_SYMBOL_GPL(class_device_create_file); EXPORT_SYMBOL_GPL(class_device_remove_file); EXPORT_SYMBOL_GPL(class_device_create_bin_file); diff --git a/include/linux/device.h b/include/linux/device.h index fa9e6ca08f5a..73250d01c01f 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -143,6 +143,7 @@ extern void driver_remove_file(struct device_driver *, struct driver_attribute * */ struct class { const char * name; + struct module * owner; struct subsystem subsys; struct list_head children; @@ -185,6 +186,7 @@ struct class_device { struct kobject kobj; struct class * class; /* required */ dev_t devt; /* dev_t, creates the sysfs "dev" */ + struct class_device_attribute *devt_attr; struct device * dev; /* not necessary, but nice to have */ void * class_data; /* class-specific data */ @@ -245,6 +247,13 @@ struct class_interface { extern int class_interface_register(struct class_interface *); extern void class_interface_unregister(struct class_interface *); +extern struct class *class_create(struct module *owner, char *name); +extern void class_destroy(struct class *cls); +extern struct class_device *class_device_create(struct class *cls, dev_t devt, + struct device *device, char *fmt, ...) + __attribute__((format(printf,4,5))); +extern void class_device_destroy(struct class *cls, dev_t devt); + /* interface for class simple stuff */ extern struct class_simple *class_simple_create(struct module *owner, char *name); extern void class_simple_destroy(struct class_simple *cs); -- cgit v1.2.3 From 1235686f6e67cf30c460eb77d90a6cb4be57b92f Mon Sep 17 00:00:00 2001 From: "gregkh@suse.de" Date: Tue, 15 Mar 2005 14:26:30 -0800 Subject: [PATCH] INPUT: move to use the new class code, instead of class_simple Signed-off-by: Greg Kroah-Hartman --- drivers/input/evdev.c | 9 +++++---- drivers/input/input.c | 10 +++++----- drivers/input/joydev.c | 8 ++++---- drivers/input/mousedev.c | 16 +++++++++------- drivers/input/tsdev.c | 9 +++++---- include/linux/input.h | 2 +- 6 files changed, 29 insertions(+), 25 deletions(-) (limited to 'include') diff --git a/drivers/input/evdev.c b/drivers/input/evdev.c index 17552a29978b..556264b43425 100644 --- a/drivers/input/evdev.c +++ b/drivers/input/evdev.c @@ -431,9 +431,9 @@ static struct input_handle *evdev_connect(struct input_handler *handler, struct devfs_mk_cdev(MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + minor), S_IFCHR|S_IRUGO|S_IWUSR, "input/event%d", minor); - class_simple_device_add(input_class, - MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + minor), - dev->dev, "event%d", minor); + class_device_create(input_class, + MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + minor), + dev->dev, "event%d", minor); return &evdev->handle; } @@ -443,7 +443,8 @@ static void evdev_disconnect(struct input_handle *handle) struct evdev *evdev = handle->private; struct evdev_list *list; - class_simple_device_remove(MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + evdev->minor)); + class_device_destroy(input_class, + MKDEV(INPUT_MAJOR, EVDEV_MINOR_BASE + evdev->minor)); devfs_remove("input/event%d", evdev->minor); evdev->exist = 0; diff --git a/drivers/input/input.c b/drivers/input/input.c index 3385dd03abfc..83c77c990dda 100644 --- a/drivers/input/input.c +++ b/drivers/input/input.c @@ -702,13 +702,13 @@ static int __init input_proc_init(void) static inline int input_proc_init(void) { return 0; } #endif -struct class_simple *input_class; +struct class *input_class; static int __init input_init(void) { int retval = -ENOMEM; - input_class = class_simple_create(THIS_MODULE, "input"); + input_class = class_create(THIS_MODULE, "input"); if (IS_ERR(input_class)) return PTR_ERR(input_class); input_proc_init(); @@ -718,7 +718,7 @@ static int __init input_init(void) remove_proc_entry("devices", proc_bus_input_dir); remove_proc_entry("handlers", proc_bus_input_dir); remove_proc_entry("input", proc_bus); - class_simple_destroy(input_class); + class_destroy(input_class); return retval; } @@ -728,7 +728,7 @@ static int __init input_init(void) remove_proc_entry("handlers", proc_bus_input_dir); remove_proc_entry("input", proc_bus); unregister_chrdev(INPUT_MAJOR, "input"); - class_simple_destroy(input_class); + class_destroy(input_class); } return retval; } @@ -741,7 +741,7 @@ static void __exit input_exit(void) devfs_remove("input"); unregister_chrdev(INPUT_MAJOR, "input"); - class_simple_destroy(input_class); + class_destroy(input_class); } subsys_initcall(input_init); diff --git a/drivers/input/joydev.c b/drivers/input/joydev.c index 627d343dfba1..39775fc380c7 100644 --- a/drivers/input/joydev.c +++ b/drivers/input/joydev.c @@ -452,9 +452,9 @@ static struct input_handle *joydev_connect(struct input_handler *handler, struct devfs_mk_cdev(MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + minor), S_IFCHR|S_IRUGO|S_IWUSR, "input/js%d", minor); - class_simple_device_add(input_class, - MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + minor), - dev->dev, "js%d", minor); + class_device_create(input_class, + MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + minor), + dev->dev, "js%d", minor); return &joydev->handle; } @@ -464,7 +464,7 @@ static void joydev_disconnect(struct input_handle *handle) struct joydev *joydev = handle->private; struct joydev_list *list; - class_simple_device_remove(MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + joydev->minor)); + class_device_destroy(input_class, MKDEV(INPUT_MAJOR, JOYDEV_MINOR_BASE + joydev->minor)); devfs_remove("input/js%d", joydev->minor); joydev->exist = 0; diff --git a/drivers/input/mousedev.c b/drivers/input/mousedev.c index 96fb9870834a..062848ac7e6b 100644 --- a/drivers/input/mousedev.c +++ b/drivers/input/mousedev.c @@ -647,9 +647,9 @@ static struct input_handle *mousedev_connect(struct input_handler *handler, stru devfs_mk_cdev(MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + minor), S_IFCHR|S_IRUGO|S_IWUSR, "input/mouse%d", minor); - class_simple_device_add(input_class, - MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + minor), - dev->dev, "mouse%d", minor); + class_device_create(input_class, + MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + minor), + dev->dev, "mouse%d", minor); return &mousedev->handle; } @@ -659,7 +659,8 @@ static void mousedev_disconnect(struct input_handle *handle) struct mousedev *mousedev = handle->private; struct mousedev_list *list; - class_simple_device_remove(MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + mousedev->minor)); + class_device_destroy(input_class, + MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + mousedev->minor)); devfs_remove("input/mouse%d", mousedev->minor); mousedev->exist = 0; @@ -735,8 +736,8 @@ static int __init mousedev_init(void) devfs_mk_cdev(MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX), S_IFCHR|S_IRUGO|S_IWUSR, "input/mice"); - class_simple_device_add(input_class, MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX), - NULL, "mice"); + class_device_create(input_class, + MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX), NULL, "mice"); #ifdef CONFIG_INPUT_MOUSEDEV_PSAUX if (!(psaux_registered = !misc_register(&psaux_mouse))) @@ -755,7 +756,8 @@ static void __exit mousedev_exit(void) misc_deregister(&psaux_mouse); #endif devfs_remove("input/mice"); - class_simple_device_remove(MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX)); + class_device_destroy(input_class, + MKDEV(INPUT_MAJOR, MOUSEDEV_MINOR_BASE + MOUSEDEV_MIX)); input_unregister_handler(&mousedev_handler); } diff --git a/drivers/input/tsdev.c b/drivers/input/tsdev.c index d0afba85720b..50c63a155156 100644 --- a/drivers/input/tsdev.c +++ b/drivers/input/tsdev.c @@ -414,9 +414,9 @@ static struct input_handle *tsdev_connect(struct input_handler *handler, S_IFCHR|S_IRUGO|S_IWUSR, "input/ts%d", minor); devfs_mk_cdev(MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + minor + TSDEV_MINORS/2), S_IFCHR|S_IRUGO|S_IWUSR, "input/tsraw%d", minor); - class_simple_device_add(input_class, - MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + minor), - dev->dev, "ts%d", minor); + class_device_create(input_class, + MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + minor), + dev->dev, "ts%d", minor); return &tsdev->handle; } @@ -426,7 +426,8 @@ static void tsdev_disconnect(struct input_handle *handle) struct tsdev *tsdev = handle->private; struct tsdev_list *list; - class_simple_device_remove(MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + tsdev->minor)); + class_device_destroy(input_class, + MKDEV(INPUT_MAJOR, TSDEV_MINOR_BASE + tsdev->minor)); devfs_remove("input/ts%d", tsdev->minor); devfs_remove("input/tsraw%d", tsdev->minor); tsdev->exist = 0; diff --git a/include/linux/input.h b/include/linux/input.h index 72731d7d189e..9d9598ed760d 100644 --- a/include/linux/input.h +++ b/include/linux/input.h @@ -1015,7 +1015,7 @@ static inline void input_set_abs_params(struct input_dev *dev, int axis, int min dev->absbit[LONG(axis)] |= BIT(axis); } -extern struct class_simple *input_class; +extern struct class *input_class; #endif #endif -- cgit v1.2.3 From 8561b10f6e7ef0a085709ffc844f74130a067abe Mon Sep 17 00:00:00 2001 From: "gregkh@suse.de" Date: Tue, 15 Mar 2005 15:10:13 -0800 Subject: [PATCH] USB: move the usb hcd code to use the new class code. This moves a kref into the main hcd structure, which detaches it from the class device structure. Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/hcd.c | 61 ++++++++++++++++++++------------------------- drivers/usb/host/ehci-dbg.c | 10 ++++---- drivers/usb/host/ohci-dbg.c | 10 ++++---- include/linux/usb.h | 5 ++-- 4 files changed, 39 insertions(+), 47 deletions(-) (limited to 'include') diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c index 266e9e06a9f5..d041782e0c8b 100644 --- a/drivers/usb/core/hcd.c +++ b/drivers/usb/core/hcd.c @@ -651,50 +651,45 @@ static int usb_rh_urb_dequeue (struct usb_hcd *hcd, struct urb *urb) /*-------------------------------------------------------------------------*/ /* exported only within usbcore */ -struct usb_bus *usb_bus_get (struct usb_bus *bus) +struct usb_bus *usb_bus_get(struct usb_bus *bus) { - struct class_device *tmp; + if (bus) + kref_get(&bus->kref); + return bus; +} - if (!bus) - return NULL; +static void usb_host_release(struct kref *kref) +{ + struct usb_bus *bus = container_of(kref, struct usb_bus, kref); - tmp = class_device_get(&bus->class_dev); - if (tmp) - return to_usb_bus(tmp); - else - return NULL; + if (bus->release) + bus->release(bus); } /* exported only within usbcore */ -void usb_bus_put (struct usb_bus *bus) +void usb_bus_put(struct usb_bus *bus) { if (bus) - class_device_put(&bus->class_dev); + kref_put(&bus->kref, usb_host_release); } /*-------------------------------------------------------------------------*/ -static void usb_host_release(struct class_device *class_dev) -{ - struct usb_bus *bus = to_usb_bus(class_dev); - - if (bus->release) - bus->release(bus); -} - -static struct class usb_host_class = { - .name = "usb_host", - .release = &usb_host_release, -}; +static struct class *usb_host_class; int usb_host_init(void) { - return class_register(&usb_host_class); + int retval = 0; + + usb_host_class = class_create(THIS_MODULE, "usb_host"); + if (IS_ERR(usb_host_class)) + retval = PTR_ERR(usb_host_class); + return retval; } void usb_host_cleanup(void) { - class_unregister(&usb_host_class); + class_destroy(usb_host_class); } /** @@ -719,8 +714,7 @@ static void usb_bus_init (struct usb_bus *bus) INIT_LIST_HEAD (&bus->bus_list); - class_device_initialize(&bus->class_dev); - bus->class_dev.class = &usb_host_class; + kref_init(&bus->kref); } /** @@ -761,7 +755,6 @@ struct usb_bus *usb_alloc_bus (struct usb_operations *op) static int usb_register_bus(struct usb_bus *bus) { int busnum; - int retval; down (&usb_bus_list_lock); busnum = find_next_zero_bit (busmap.busmap, USB_MAXBUS, 1); @@ -774,15 +767,15 @@ static int usb_register_bus(struct usb_bus *bus) return -E2BIG; } - snprintf(bus->class_dev.class_id, BUS_ID_SIZE, "usb%d", busnum); - bus->class_dev.dev = bus->controller; - retval = class_device_add(&bus->class_dev); - if (retval) { + bus->class_dev = class_device_create(usb_host_class, MKDEV(0,0), bus->controller, "usb%d", busnum); + if (IS_ERR(bus->class_dev)) { clear_bit(busnum, busmap.busmap); up(&usb_bus_list_lock); - return retval; + return PTR_ERR(bus->class_dev); } + class_set_devdata(bus->class_dev, bus); + /* Add it to the local list of buses */ list_add (&bus->bus_list, &usb_bus_list); up (&usb_bus_list_lock); @@ -820,7 +813,7 @@ static void usb_deregister_bus (struct usb_bus *bus) clear_bit (bus->busnum, busmap.busmap); - class_device_del(&bus->class_dev); + class_device_unregister(bus->class_dev); } /** diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c index 9b347d765383..2ff11d53567b 100644 --- a/drivers/usb/host/ehci-dbg.c +++ b/drivers/usb/host/ehci-dbg.c @@ -450,7 +450,7 @@ show_async (struct class_device *class_dev, char *buf) *buf = 0; - bus = to_usb_bus(class_dev); + bus = class_get_devdata(class_dev); hcd = bus->hcpriv; ehci = hcd_to_ehci (hcd); next = buf; @@ -496,7 +496,7 @@ show_periodic (struct class_device *class_dev, char *buf) return 0; seen_count = 0; - bus = to_usb_bus(class_dev); + bus = class_get_devdata(class_dev); hcd = bus->hcpriv; ehci = hcd_to_ehci (hcd); next = buf; @@ -633,7 +633,7 @@ show_registers (struct class_device *class_dev, char *buf) static char fmt [] = "%*s\n"; static char label [] = ""; - bus = to_usb_bus(class_dev); + bus = class_get_devdata(class_dev); hcd = bus->hcpriv; ehci = hcd_to_ehci (hcd); next = buf; @@ -735,7 +735,7 @@ static CLASS_DEVICE_ATTR (registers, S_IRUGO, show_registers, NULL); static inline void create_debug_files (struct ehci_hcd *ehci) { - struct class_device *cldev = &ehci_to_hcd(ehci)->self.class_dev; + struct class_device *cldev = ehci_to_hcd(ehci)->self.class_dev; class_device_create_file(cldev, &class_device_attr_async); class_device_create_file(cldev, &class_device_attr_periodic); @@ -744,7 +744,7 @@ static inline void create_debug_files (struct ehci_hcd *ehci) static inline void remove_debug_files (struct ehci_hcd *ehci) { - struct class_device *cldev = &ehci_to_hcd(ehci)->self.class_dev; + struct class_device *cldev = ehci_to_hcd(ehci)->self.class_dev; class_device_remove_file(cldev, &class_device_attr_async); class_device_remove_file(cldev, &class_device_attr_periodic); diff --git a/drivers/usb/host/ohci-dbg.c b/drivers/usb/host/ohci-dbg.c index 62f53a213808..c58408c95c3d 100644 --- a/drivers/usb/host/ohci-dbg.c +++ b/drivers/usb/host/ohci-dbg.c @@ -481,7 +481,7 @@ show_async (struct class_device *class_dev, char *buf) size_t temp; unsigned long flags; - bus = to_usb_bus(class_dev); + bus = class_get_devdata(class_dev); hcd = bus->hcpriv; ohci = hcd_to_ohci(hcd); @@ -514,7 +514,7 @@ show_periodic (struct class_device *class_dev, char *buf) return 0; seen_count = 0; - bus = to_usb_bus(class_dev); + bus = class_get_devdata(class_dev); hcd = bus->hcpriv; ohci = hcd_to_ohci(hcd); next = buf; @@ -611,7 +611,7 @@ show_registers (struct class_device *class_dev, char *buf) char *next; u32 rdata; - bus = to_usb_bus(class_dev); + bus = class_get_devdata(class_dev); hcd = bus->hcpriv; ohci = hcd_to_ohci(hcd); regs = ohci->regs; @@ -684,7 +684,7 @@ static CLASS_DEVICE_ATTR (registers, S_IRUGO, show_registers, NULL); static inline void create_debug_files (struct ohci_hcd *ohci) { - struct class_device *cldev = &ohci_to_hcd(ohci)->self.class_dev; + struct class_device *cldev = ohci_to_hcd(ohci)->self.class_dev; class_device_create_file(cldev, &class_device_attr_async); class_device_create_file(cldev, &class_device_attr_periodic); @@ -694,7 +694,7 @@ static inline void create_debug_files (struct ohci_hcd *ohci) static inline void remove_debug_files (struct ohci_hcd *ohci) { - struct class_device *cldev = &ohci_to_hcd(ohci)->self.class_dev; + struct class_device *cldev = ohci_to_hcd(ohci)->self.class_dev; class_device_remove_file(cldev, &class_device_attr_async); class_device_remove_file(cldev, &class_device_attr_periodic); diff --git a/include/linux/usb.h b/include/linux/usb.h index 2d1ac5058534..3d508bf08402 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -287,15 +287,14 @@ struct usb_bus { struct dentry *usbfs_dentry; /* usbfs dentry entry for the bus */ - struct class_device class_dev; /* class device for this bus */ + struct class_device *class_dev; /* class device for this bus */ + struct kref kref; /* handles reference counting this bus */ void (*release)(struct usb_bus *bus); /* function to destroy this bus's memory */ #if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE) struct mon_bus *mon_bus; /* non-null when associated */ int monitored; /* non-zero when monitored */ #endif }; -#define to_usb_bus(d) container_of(d, struct usb_bus, class_dev) - /* -------------------------------------------------------------------------- */ -- cgit v1.2.3 From cd987d38cc59053e0bab8150ffaca33b109067f3 Mon Sep 17 00:00:00 2001 From: "gregkh@suse.de" Date: Wed, 23 Mar 2005 11:12:38 -0800 Subject: [PATCH] class: remove class_simple code, as no one in the tree is using it anymore. Signed-off-by: Greg Kroah-Hartman --- drivers/base/Makefile | 2 +- drivers/base/class_simple.c | 199 -------------------------------------------- include/linux/device.h | 10 --- 3 files changed, 1 insertion(+), 210 deletions(-) delete mode 100644 drivers/base/class_simple.c (limited to 'include') diff --git a/drivers/base/Makefile b/drivers/base/Makefile index a47928a2e575..771f6c80e824 100644 --- a/drivers/base/Makefile +++ b/drivers/base/Makefile @@ -1,7 +1,7 @@ # Makefile for the Linux device tree obj-y := core.o sys.o bus.o \ - driver.o class.o class_simple.o platform.o \ + driver.o class.o platform.o \ cpu.o firmware.o init.o map.o dmapool.o \ attribute_container.o transport_class.o obj-y += power/ diff --git a/drivers/base/class_simple.c b/drivers/base/class_simple.c deleted file mode 100644 index 27699eb20a37..000000000000 --- a/drivers/base/class_simple.c +++ /dev/null @@ -1,199 +0,0 @@ -/* - * class_simple.c - a "simple" interface for classes for simple char devices. - * - * Copyright (c) 2003-2004 Greg Kroah-Hartman - * Copyright (c) 2003-2004 IBM Corp. - * - * This file is released under the GPLv2 - * - */ - -#include -#include -#include - -struct class_simple { - struct class class; -}; -#define to_class_simple(d) container_of(d, struct class_simple, class) - -struct simple_dev { - struct list_head node; - struct class_device class_dev; -}; -#define to_simple_dev(d) container_of(d, struct simple_dev, class_dev) - -static LIST_HEAD(simple_dev_list); -static DEFINE_SPINLOCK(simple_dev_list_lock); - -static void release_simple_dev(struct class_device *class_dev) -{ - struct simple_dev *s_dev = to_simple_dev(class_dev); - kfree(s_dev); -} - -static void class_simple_release(struct class *class) -{ - struct class_simple *cs = to_class_simple(class); - kfree(cs); -} - -/** - * class_simple_create - create a struct class_simple structure - * @owner: pointer to the module that is to "own" this struct class_simple - * @name: pointer to a string for the name of this class. - * - * This is used to create a struct class_simple pointer that can then be used - * in calls to class_simple_device_add(). This is used when you do not wish to - * create a full blown class support for a type of char devices. - * - * Note, the pointer created here is to be destroyed when finished by making a - * call to class_simple_destroy(). - */ -struct class_simple *class_simple_create(struct module *owner, char *name) -{ - struct class_simple *cs; - int retval; - - cs = kmalloc(sizeof(*cs), GFP_KERNEL); - if (!cs) { - retval = -ENOMEM; - goto error; - } - memset(cs, 0x00, sizeof(*cs)); - - cs->class.name = name; - cs->class.class_release = class_simple_release; - cs->class.release = release_simple_dev; - - retval = class_register(&cs->class); - if (retval) - goto error; - - return cs; - -error: - kfree(cs); - return ERR_PTR(retval); -} -EXPORT_SYMBOL(class_simple_create); - -/** - * class_simple_destroy - destroys a struct class_simple structure - * @cs: pointer to the struct class_simple that is to be destroyed - * - * Note, the pointer to be destroyed must have been created with a call to - * class_simple_create(). - */ -void class_simple_destroy(struct class_simple *cs) -{ - if ((cs == NULL) || (IS_ERR(cs))) - return; - - class_unregister(&cs->class); -} -EXPORT_SYMBOL(class_simple_destroy); - -/** - * class_simple_device_add - adds a class device to sysfs for a character driver - * @cs: pointer to the struct class_simple that this device should be registered to. - * @dev: the dev_t for the device to be added. - * @device: a pointer to a struct device that is assiociated with this class device. - * @fmt: string for the class device's name - * - * This function can be used by simple char device classes that do not - * implement their own class device registration. A struct class_device will - * be created in sysfs, registered to the specified class. A "dev" file will - * be created, showing the dev_t for the device. The pointer to the struct - * class_device will be returned from the call. Any further sysfs files that - * might be required can be created using this pointer. - * Note: the struct class_simple passed to this function must have previously been - * created with a call to class_simple_create(). - */ -struct class_device *class_simple_device_add(struct class_simple *cs, dev_t dev, struct device *device, const char *fmt, ...) -{ - va_list args; - struct simple_dev *s_dev = NULL; - int retval; - - if ((cs == NULL) || (IS_ERR(cs))) { - retval = -ENODEV; - goto error; - } - - s_dev = kmalloc(sizeof(*s_dev), GFP_KERNEL); - if (!s_dev) { - retval = -ENOMEM; - goto error; - } - memset(s_dev, 0x00, sizeof(*s_dev)); - - s_dev->class_dev.devt = dev; - s_dev->class_dev.dev = device; - s_dev->class_dev.class = &cs->class; - - va_start(args, fmt); - vsnprintf(s_dev->class_dev.class_id, BUS_ID_SIZE, fmt, args); - va_end(args); - retval = class_device_register(&s_dev->class_dev); - if (retval) - goto error; - - spin_lock(&simple_dev_list_lock); - list_add(&s_dev->node, &simple_dev_list); - spin_unlock(&simple_dev_list_lock); - - return &s_dev->class_dev; - -error: - kfree(s_dev); - return ERR_PTR(retval); -} -EXPORT_SYMBOL(class_simple_device_add); - -/** - * class_simple_set_hotplug - set the hotplug callback in the embedded struct class - * @cs: pointer to the struct class_simple to hold the pointer - * @hotplug: function pointer to the hotplug function - * - * Implement and set a hotplug function to add environment variables specific to this - * class on the hotplug event. - */ -int class_simple_set_hotplug(struct class_simple *cs, - int (*hotplug)(struct class_device *dev, char **envp, int num_envp, char *buffer, int buffer_size)) -{ - if ((cs == NULL) || (IS_ERR(cs))) - return -ENODEV; - cs->class.hotplug = hotplug; - return 0; -} -EXPORT_SYMBOL(class_simple_set_hotplug); - -/** - * class_simple_device_remove - removes a class device that was created with class_simple_device_add() - * @dev: the dev_t of the device that was previously registered. - * - * This call unregisters and cleans up a class device that was created with a - * call to class_device_simple_add() - */ -void class_simple_device_remove(dev_t dev) -{ - struct simple_dev *s_dev = NULL; - int found = 0; - - spin_lock(&simple_dev_list_lock); - list_for_each_entry(s_dev, &simple_dev_list, node) { - if (s_dev->class_dev.devt == dev) { - found = 1; - break; - } - } - if (found) { - list_del(&s_dev->node); - spin_unlock(&simple_dev_list_lock); - class_device_unregister(&s_dev->class_dev); - } else { - spin_unlock(&simple_dev_list_lock); - } -} -EXPORT_SYMBOL(class_simple_device_remove); diff --git a/include/linux/device.h b/include/linux/device.h index 73250d01c01f..68a95358013d 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -44,7 +44,6 @@ struct device; struct device_driver; struct class; struct class_device; -struct class_simple; struct bus_type { const char * name; @@ -254,15 +253,6 @@ extern struct class_device *class_device_create(struct class *cls, dev_t devt, __attribute__((format(printf,4,5))); extern void class_device_destroy(struct class *cls, dev_t devt); -/* interface for class simple stuff */ -extern struct class_simple *class_simple_create(struct module *owner, char *name); -extern void class_simple_destroy(struct class_simple *cs); -extern struct class_device *class_simple_device_add(struct class_simple *cs, dev_t dev, struct device *device, const char *fmt, ...) - __attribute__((format(printf,4,5))); -extern int class_simple_set_hotplug(struct class_simple *, - int (*hotplug)(struct class_device *dev, char **envp, int num_envp, char *buffer, int buffer_size)); -extern void class_simple_device_remove(dev_t dev); - struct device { struct list_head node; /* node in sibling list */ -- cgit v1.2.3 From af70316af182f4716cc5eec7e0d27fc731d164bd Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Mon, 21 Mar 2005 10:41:04 -0800 Subject: [PATCH] Add a semaphore to struct device to synchronize calls to its driver. This adds a per-device semaphore that is taken before every call from the core to a driver method. This prevents e.g. simultaneous calls to the ->suspend() or ->resume() and ->probe() or ->release(), potentially saving a whole lot of headaches. It also moves us a step closer to removing the bus rwsem, since it protects the fields in struct device that are modified by the core. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/bus.c | 14 +++++++++++--- drivers/base/core.c | 1 + drivers/base/power/resume.c | 8 ++++++-- drivers/base/power/suspend.c | 3 ++- include/linux/device.h | 4 ++++ 5 files changed, 24 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/drivers/base/bus.c b/drivers/base/bus.c index 80ce88de56fa..aa27f76d28cd 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -283,18 +283,22 @@ void device_bind_driver(struct device * dev) */ int driver_probe_device(struct device_driver * drv, struct device * dev) { + int error = 0; + if (drv->bus->match && !drv->bus->match(dev, drv)) return -ENODEV; + down(&dev->sem); dev->driver = drv; if (drv->probe) { - int error = drv->probe(dev); + error = drv->probe(dev); if (error) { dev->driver = NULL; + up(&dev->sem); return error; } } - + up(&dev->sem); device_bind_driver(dev); return 0; } @@ -385,7 +389,10 @@ void driver_attach(struct device_driver * drv) void device_release_driver(struct device * dev) { - struct device_driver * drv = dev->driver; + struct device_driver * drv; + + down(&dev->sem); + drv = dev->driver; if (drv) { sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj)); sysfs_remove_link(&dev->kobj, "driver"); @@ -394,6 +401,7 @@ void device_release_driver(struct device * dev) drv->remove(dev); dev->driver = NULL; } + up(&dev->sem); } diff --git a/drivers/base/core.c b/drivers/base/core.c index a293a788abd4..93440824b800 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -212,6 +212,7 @@ void device_initialize(struct device *dev) INIT_LIST_HEAD(&dev->driver_list); INIT_LIST_HEAD(&dev->bus_list); INIT_LIST_HEAD(&dev->dma_pools); + init_MUTEX(&dev->sem); } /** diff --git a/drivers/base/power/resume.c b/drivers/base/power/resume.c index 26468971ef5a..bdd96b03b885 100644 --- a/drivers/base/power/resume.c +++ b/drivers/base/power/resume.c @@ -22,6 +22,9 @@ extern int sysdev_resume(void); int resume_device(struct device * dev) { + int error = 0; + + down(&dev->sem); if (dev->power.pm_parent && dev->power.pm_parent->power.power_state) { dev_err(dev, "PM: resume from %d, parent %s still %d\n", @@ -31,9 +34,10 @@ int resume_device(struct device * dev) } if (dev->bus && dev->bus->resume) { dev_dbg(dev,"resuming\n"); - return dev->bus->resume(dev); + error = dev->bus->resume(dev); } - return 0; + up(&dev->sem); + return error; } diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c index 0ec44ef840be..807e13fb205b 100644 --- a/drivers/base/power/suspend.c +++ b/drivers/base/power/suspend.c @@ -39,6 +39,7 @@ int suspend_device(struct device * dev, pm_message_t state) { int error = 0; + down(&dev->sem); if (dev->power.power_state) { dev_dbg(dev, "PM: suspend %d-->%d\n", dev->power.power_state, state); @@ -58,7 +59,7 @@ int suspend_device(struct device * dev, pm_message_t state) dev_dbg(dev, "suspending\n"); error = dev->bus->suspend(dev, state); } - + up(&dev->sem); return error; } diff --git a/include/linux/device.h b/include/linux/device.h index 68a95358013d..195b327f3e19 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -264,6 +264,10 @@ struct device { struct kobject kobj; char bus_id[BUS_ID_SIZE]; /* position on parent bus */ + struct semaphore sem; /* semaphore to synchronize calls to + * its driver. + */ + struct bus_type * bus; /* type of bus device is on */ struct device_driver *driver; /* which driver has allocated this device */ -- cgit v1.2.3 From fae3cd00255e3e51ffd59fedb1bdb91ec96be395 Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Mon, 21 Mar 2005 10:59:56 -0800 Subject: [PATCH] Add driver_for_each_device(). Now there's an iterator for accessing each device bound to a driver. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman Index: linux-2.6.12-rc2/drivers/base/driver.c =================================================================== --- drivers/base/driver.c | 35 +++++++++++++++++++++++++++++++++++ include/linux/device.h | 3 +++ 2 files changed, 38 insertions(+) (limited to 'include') diff --git a/drivers/base/driver.c b/drivers/base/driver.c index 3b269f7e5213..484fed1985aa 100644 --- a/drivers/base/driver.c +++ b/drivers/base/driver.c @@ -18,6 +18,41 @@ #define to_dev(node) container_of(node, struct device, driver_list) #define to_drv(obj) container_of(obj, struct device_driver, kobj) + +/** + * driver_for_each_device - Iterator for devices bound to a driver. + * @drv: Driver we're iterating. + * @data: Data to pass to the callback. + * @fn: Function to call for each device. + * + * Take the bus's rwsem and iterate over the @drv's list of devices, + * calling @fn for each one. + */ + +int driver_for_each_device(struct device_driver * drv, struct device * start, + void * data, int (*fn)(struct device *, void *)) +{ + struct list_head * head; + struct device * dev; + int error = 0; + + down_read(&drv->bus->subsys.rwsem); + head = &drv->devices; + dev = list_prepare_entry(start, head, driver_list); + list_for_each_entry_continue(dev, head, driver_list) { + get_device(dev); + error = fn(dev, data); + put_device(dev); + if (error) + break; + } + up_read(&drv->bus->subsys.rwsem); + return error; +} + +EXPORT_SYMBOL(driver_for_each_device); + + /** * driver_create_file - create sysfs file for driver. * @drv: driver. diff --git a/include/linux/device.h b/include/linux/device.h index 195b327f3e19..10f5aa20e041 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -136,6 +136,9 @@ struct driver_attribute driver_attr_##_name = __ATTR(_name,_mode,_show,_store) extern int driver_create_file(struct device_driver *, struct driver_attribute *); extern void driver_remove_file(struct device_driver *, struct driver_attribute *); +extern int driver_for_each_device(struct device_driver * drv, struct device * start, + void * data, int (*fn)(struct device *, void *)); + /* * device classes -- cgit v1.2.3 From 9a19fea43616066561e221359596ce532e631395 Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Mon, 21 Mar 2005 11:45:16 -0800 Subject: [PATCH] Add initial implementation of klist helpers. This klist interface provides a couple of structures that wrap around struct list_head to provide explicit list "head" (struct klist) and list "node" (struct klist_node) objects. For struct klist, a spinlock is included that protects access to the actual list itself. struct klist_node provides a pointer to the klist that owns it and a kref reference count that indicates the number of current users of that node in the list. The entire point is to provide an interface for iterating over a list that is safe and allows for modification of the list during the iteration (e.g. insertion and removal), including modification of the current node on the list. It works using a 3rd object type - struct klist_iter - that is declared and initialized before an iteration. klist_next() is used to acquire the next element in the list. It returns NULL if there are no more items. This klist interface provides a couple of structures that wrap around struct list_head to provide explicit list "head" (struct klist) and list "node" (struct klist_node) objects. For struct klist, a spinlock is included that protects access to the actual list itself. struct klist_node provides a pointer to the klist that owns it and a kref reference count that indicates the number of current users of that node in the list. The entire point is to provide an interface for iterating over a list that is safe and allows for modification of the list during the iteration (e.g. insertion and removal), including modification of the current node on the list. It works using a 3rd object type - struct klist_iter - that is declared and initialized before an iteration. klist_next() is used to acquire the next element in the list. It returns NULL if there are no more items. Internally, that routine takes the klist's lock, decrements the reference count of the previous klist_node and increments the count of the next klist_node. It then drops the lock and returns. There are primitives for adding and removing nodes to/from a klist. When deleting, klist_del() will simply decrement the reference count. Only when the count goes to 0 is the node removed from the list. klist_remove() will try to delete the node from the list and block until it is actually removed. This is useful for objects (like devices) that have been removed from the system and must be freed (but must wait until all accessors have finished). Internally, that routine takes the klist's lock, decrements the reference count of the previous klist_node and increments the count of the next klist_node. It then drops the lock and returns. There are primitives for adding and removing nodes to/from a klist. When deleting, klist_del() will simply decrement the reference count. Only when the count goes to 0 is the node removed from the list. klist_remove() will try to delete the node from the list and block until it is actually removed. This is useful for objects (like devices) that have been removed from the system and must be freed (but must wait until all accessors have finished). Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman diff -Nru a/include/linux/klist.h b/include/linux/klist.h --- include/linux/klist.h | 53 +++++++++++ lib/Makefile | 7 +- lib/klist.c | 248 ++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 305 insertions(+), 3 deletions(-) create mode 100644 include/linux/klist.h create mode 100644 lib/klist.c (limited to 'include') diff --git a/include/linux/klist.h b/include/linux/klist.h new file mode 100644 index 000000000000..fb52f9d9d611 --- /dev/null +++ b/include/linux/klist.h @@ -0,0 +1,53 @@ +/* + * klist.h - Some generic list helpers, extending struct list_head a bit. + * + * Implementations are found in lib/klist.c + * + * + * Copyright (C) 2005 Patrick Mochel + * + * This file is rleased under the GPL v2. + */ + +#include +#include +#include +#include + + +struct klist { + spinlock_t k_lock; + struct list_head k_list; +}; + + +extern void klist_init(struct klist * k); + + +struct klist_node { + struct klist * n_klist; + struct list_head n_node; + struct kref n_ref; + struct completion n_removed; +}; + +extern void klist_add_tail(struct klist * k, struct klist_node * n); +extern void klist_add_head(struct klist * k, struct klist_node * n); + +extern void klist_del(struct klist_node * n); +extern void klist_remove(struct klist_node * n); + + +struct klist_iter { + struct klist * i_klist; + struct list_head * i_head; + struct klist_node * i_cur; +}; + + +extern void klist_iter_init(struct klist * k, struct klist_iter * i); +extern void klist_iter_init_node(struct klist * k, struct klist_iter * i, + struct klist_node * n); +extern void klist_iter_exit(struct klist_iter * i); +extern struct klist_node * klist_next(struct klist_iter * i); + diff --git a/lib/Makefile b/lib/Makefile index 7c70db79c0e0..9eccea9429a7 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -4,9 +4,10 @@ lib-y := errno.o ctype.o string.o vsprintf.o cmdline.o \ bust_spinlocks.o rbtree.o radix-tree.o dump_stack.o \ - kobject.o kref.o idr.o div64.o int_sqrt.o \ - bitmap.o extable.o kobject_uevent.o prio_tree.o sha1.o \ - halfmd4.o + idr.o div64.o int_sqrt.o bitmap.o extable.o prio_tree.o \ + sha1.o halfmd4.o + +lib-y += kobject.o kref.o kobject_uevent.o klist.o obj-y += sort.o parser.o diff --git a/lib/klist.c b/lib/klist.c new file mode 100644 index 000000000000..6f760424648b --- /dev/null +++ b/lib/klist.c @@ -0,0 +1,248 @@ +/* + * klist.c - Routines for manipulating klists. + * + * + * This klist interface provides a couple of structures that wrap around + * struct list_head to provide explicit list "head" (struct klist) and + * list "node" (struct klist_node) objects. For struct klist, a spinlock + * is included that protects access to the actual list itself. struct + * klist_node provides a pointer to the klist that owns it and a kref + * reference count that indicates the number of current users of that node + * in the list. + * + * The entire point is to provide an interface for iterating over a list + * that is safe and allows for modification of the list during the + * iteration (e.g. insertion and removal), including modification of the + * current node on the list. + * + * It works using a 3rd object type - struct klist_iter - that is declared + * and initialized before an iteration. klist_next() is used to acquire the + * next element in the list. It returns NULL if there are no more items. + * Internally, that routine takes the klist's lock, decrements the reference + * count of the previous klist_node and increments the count of the next + * klist_node. It then drops the lock and returns. + * + * There are primitives for adding and removing nodes to/from a klist. + * When deleting, klist_del() will simply decrement the reference count. + * Only when the count goes to 0 is the node removed from the list. + * klist_remove() will try to delete the node from the list and block + * until it is actually removed. This is useful for objects (like devices) + * that have been removed from the system and must be freed (but must wait + * until all accessors have finished). + * + * Copyright (C) 2005 Patrick Mochel + * + * This file is released under the GPL v2. + */ + +#include +#include + + +/** + * klist_init - Initialize a klist structure. + * @k: The klist we're initializing. + */ + +void klist_init(struct klist * k) +{ + INIT_LIST_HEAD(&k->k_list); + spin_lock_init(&k->k_lock); +} + +EXPORT_SYMBOL_GPL(klist_init); + + +static void add_head(struct klist * k, struct klist_node * n) +{ + spin_lock(&k->k_lock); + list_add(&n->n_node, &k->k_list); + spin_unlock(&k->k_lock); +} + +static void add_tail(struct klist * k, struct klist_node * n) +{ + spin_lock(&k->k_lock); + list_add_tail(&n->n_node, &k->k_list); + spin_unlock(&k->k_lock); +} + + +static void klist_node_init(struct klist * k, struct klist_node * n) +{ + INIT_LIST_HEAD(&n->n_node); + init_completion(&n->n_removed); + kref_init(&n->n_ref); + n->n_klist = k; +} + + +/** + * klist_add_head - Initialize a klist_node and add it to front. + * @k: klist it's going on. + * @n: node we're adding. + */ + +void klist_add_head(struct klist * k, struct klist_node * n) +{ + klist_node_init(k, n); + add_head(k, n); +} + +EXPORT_SYMBOL_GPL(klist_add_head); + + +/** + * klist_add_tail - Initialize a klist_node and add it to back. + * @k: klist it's going on. + * @n: node we're adding. + */ + +void klist_add_tail(struct klist * k, struct klist_node * n) +{ + klist_node_init(k, n); + add_tail(k, n); +} + +EXPORT_SYMBOL_GPL(klist_add_tail); + + +static void klist_release(struct kref * kref) +{ + struct klist_node * n = container_of(kref, struct klist_node, n_ref); + list_del(&n->n_node); + complete(&n->n_removed); +} + +static int klist_dec_and_del(struct klist_node * n) +{ + return kref_put(&n->n_ref, klist_release); +} + + +/** + * klist_del - Decrement the reference count of node and try to remove. + * @n: node we're deleting. + */ + +void klist_del(struct klist_node * n) +{ + struct klist * k = n->n_klist; + + spin_lock(&k->k_lock); + klist_dec_and_del(n); + spin_unlock(&k->k_lock); +} + +EXPORT_SYMBOL_GPL(klist_del); + + +/** + * klist_remove - Decrement the refcount of node and wait for it to go away. + * @n: node we're removing. + */ + +void klist_remove(struct klist_node * n) +{ + spin_lock(&n->n_klist->k_lock); + klist_dec_and_del(n); + spin_unlock(&n->n_klist->k_lock); + wait_for_completion(&n->n_removed); +} + +EXPORT_SYMBOL_GPL(klist_remove); + + +/** + * klist_iter_init_node - Initialize a klist_iter structure. + * @k: klist we're iterating. + * @i: klist_iter we're filling. + * @n: node to start with. + * + * Similar to klist_iter_init(), but starts the action off with @n, + * instead of with the list head. + */ + +void klist_iter_init_node(struct klist * k, struct klist_iter * i, struct klist_node * n) +{ + i->i_klist = k; + i->i_head = &k->k_list; + i->i_cur = n; +} + +EXPORT_SYMBOL_GPL(klist_iter_init_node); + + +/** + * klist_iter_init - Iniitalize a klist_iter structure. + * @k: klist we're iterating. + * @i: klist_iter structure we're filling. + * + * Similar to klist_iter_init_node(), but start with the list head. + */ + +void klist_iter_init(struct klist * k, struct klist_iter * i) +{ + klist_iter_init_node(k, i, NULL); +} + +EXPORT_SYMBOL_GPL(klist_iter_init); + + +/** + * klist_iter_exit - Finish a list iteration. + * @i: Iterator structure. + * + * Must be called when done iterating over list, as it decrements the + * refcount of the current node. Necessary in case iteration exited before + * the end of the list was reached, and always good form. + */ + +void klist_iter_exit(struct klist_iter * i) +{ + if (i->i_cur) { + klist_del(i->i_cur); + i->i_cur = NULL; + } +} + +EXPORT_SYMBOL_GPL(klist_iter_exit); + + +static struct klist_node * to_klist_node(struct list_head * n) +{ + return container_of(n, struct klist_node, n_node); +} + + +/** + * klist_next - Ante up next node in list. + * @i: Iterator structure. + * + * First grab list lock. Decrement the reference count of the previous + * node, if there was one. Grab the next node, increment its reference + * count, drop the lock, and return that next node. + */ + +struct klist_node * klist_next(struct klist_iter * i) +{ + struct list_head * next; + struct klist_node * knode = NULL; + + spin_lock(&i->i_klist->k_lock); + if (i->i_cur) { + next = i->i_cur->n_node.next; + klist_dec_and_del(i->i_cur); + } else + next = i->i_head->next; + + if (next != i->i_head) { + knode = to_klist_node(next); + kref_get(&knode->n_ref); + } + i->i_cur = knode; + spin_unlock(&i->i_klist->k_lock); + return knode; +} + +EXPORT_SYMBOL_GPL(klist_next); -- cgit v1.2.3 From 465c7a3a3a5aabcedd2e43612cac5a12f23da19a Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Mon, 21 Mar 2005 11:49:14 -0800 Subject: [PATCH] Add a klist to struct bus_type for its devices. - Use it for bus_for_each_dev(). - Use the klist spinlock instead of the bus rwsem. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/bus.c | 54 ++++++++++++++++++++++---------------------------- include/linux/device.h | 3 +++ 2 files changed, 27 insertions(+), 30 deletions(-) (limited to 'include') diff --git a/drivers/base/bus.c b/drivers/base/bus.c index 1b05c1399e37..c8da37230b3e 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -134,28 +134,6 @@ static struct kobj_type ktype_bus = { decl_subsys(bus, &ktype_bus, NULL); -static int __bus_for_each_dev(struct bus_type *bus, struct device *start, - void *data, int (*fn)(struct device *, void *)) -{ - struct list_head *head; - struct device *dev; - int error = 0; - - if (!(bus = get_bus(bus))) - return -EINVAL; - - head = &bus->devices.list; - dev = list_prepare_entry(start, head, bus_list); - list_for_each_entry_continue(dev, head, bus_list) { - get_device(dev); - error = fn(dev, data); - put_device(dev); - if (error) - break; - } - put_bus(bus); - return error; -} static int __bus_for_each_drv(struct bus_type *bus, struct device_driver *start, void * data, int (*fn)(struct device_driver *, void *)) @@ -180,6 +158,13 @@ static int __bus_for_each_drv(struct bus_type *bus, struct device_driver *start, return error; } + +static struct device * next_device(struct klist_iter * i) +{ + struct klist_node * n = klist_next(i); + return n ? container_of(n, struct device, knode_bus) : NULL; +} + /** * bus_for_each_dev - device iterator. * @bus: bus type. @@ -203,12 +188,19 @@ static int __bus_for_each_drv(struct bus_type *bus, struct device_driver *start, int bus_for_each_dev(struct bus_type * bus, struct device * start, void * data, int (*fn)(struct device *, void *)) { - int ret; + struct klist_iter i; + struct device * dev; + int error = 0; - down_read(&bus->subsys.rwsem); - ret = __bus_for_each_dev(bus, start, data, fn); - up_read(&bus->subsys.rwsem); - return ret; + if (!bus) + return -EINVAL; + + klist_iter_init_node(&bus->klist_devices, &i, + (start ? &start->knode_bus : NULL)); + while ((dev = next_device(&i)) && !error) + error = fn(dev, data); + klist_iter_exit(&i); + return error; } /** @@ -293,6 +285,7 @@ int bus_add_device(struct device * dev) list_add_tail(&dev->bus_list, &dev->bus->devices.list); device_attach(dev); up_write(&dev->bus->subsys.rwsem); + klist_add_tail(&bus->klist_devices, &dev->knode_bus); device_add_attrs(bus, dev); sysfs_create_link(&bus->devices.kobj, &dev->kobj, dev->bus_id); sysfs_create_link(&dev->kobj, &dev->bus->subsys.kset.kobj, "bus"); @@ -315,6 +308,7 @@ void bus_remove_device(struct device * dev) sysfs_remove_link(&dev->kobj, "bus"); sysfs_remove_link(&dev->bus->devices.kobj, dev->bus_id); device_remove_attrs(dev->bus, dev); + klist_remove(&dev->knode_bus); down_write(&dev->bus->subsys.rwsem); pr_debug("bus %s: remove device %s\n", dev->bus->name, dev->bus_id); device_release_driver(dev); @@ -439,9 +433,7 @@ int bus_rescan_devices(struct bus_type * bus) { int count = 0; - down_write(&bus->subsys.rwsem); - __bus_for_each_dev(bus, NULL, &count, bus_rescan_devices_helper); - up_write(&bus->subsys.rwsem); + bus_for_each_dev(bus, NULL, &count, bus_rescan_devices_helper); return count; } @@ -542,6 +534,8 @@ int bus_register(struct bus_type * bus) retval = kset_register(&bus->drivers); if (retval) goto bus_drivers_fail; + + klist_init(&bus->klist_devices); bus_add_attrs(bus); pr_debug("bus type '%s' registered\n", bus->name); diff --git a/include/linux/device.h b/include/linux/device.h index 10f5aa20e041..e36953cf7f14 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -51,6 +52,7 @@ struct bus_type { struct subsystem subsys; struct kset drivers; struct kset devices; + struct klist klist_devices; struct bus_attribute * bus_attrs; struct device_attribute * dev_attrs; @@ -262,6 +264,7 @@ struct device { struct list_head bus_list; /* node in bus's list */ struct list_head driver_list; struct list_head children; + struct klist_node knode_bus; struct device * parent; struct kobject kobj; -- cgit v1.2.3 From 38fdac3cdce276554b4484a41f8ec2daf81cb2ff Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Mon, 21 Mar 2005 12:00:18 -0800 Subject: [PATCH] Add a klist to struct bus_type for its drivers. - Use it in bus_for_each_drv(). - Use the klist spinlock instead of the bus rwsem. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/bus.c | 52 ++++++++++++++++++++++---------------------------- include/linux/device.h | 2 ++ 2 files changed, 25 insertions(+), 29 deletions(-) (limited to 'include') diff --git a/drivers/base/bus.c b/drivers/base/bus.c index c8da37230b3e..736dc1f5a316 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -135,30 +135,6 @@ static struct kobj_type ktype_bus = { decl_subsys(bus, &ktype_bus, NULL); -static int __bus_for_each_drv(struct bus_type *bus, struct device_driver *start, - void * data, int (*fn)(struct device_driver *, void *)) -{ - struct list_head *head; - struct device_driver *drv; - int error = 0; - - if (!(bus = get_bus(bus))) - return -EINVAL; - - head = &bus->drivers.list; - drv = list_prepare_entry(start, head, kobj.entry); - list_for_each_entry_continue(drv, head, kobj.entry) { - get_driver(drv); - error = fn(drv, data); - put_driver(drv); - if (error) - break; - } - put_bus(bus); - return error; -} - - static struct device * next_device(struct klist_iter * i) { struct klist_node * n = klist_next(i); @@ -203,6 +179,14 @@ int bus_for_each_dev(struct bus_type * bus, struct device * start, return error; } + + +static struct device_driver * next_driver(struct klist_iter * i) +{ + struct klist_node * n = klist_next(i); + return n ? container_of(n, struct device_driver, knode_bus) : NULL; +} + /** * bus_for_each_drv - driver iterator * @bus: bus we're dealing with. @@ -226,12 +210,19 @@ int bus_for_each_dev(struct bus_type * bus, struct device * start, int bus_for_each_drv(struct bus_type * bus, struct device_driver * start, void * data, int (*fn)(struct device_driver *, void *)) { - int ret; + struct klist_iter i; + struct device_driver * drv; + int error = 0; - down_read(&bus->subsys.rwsem); - ret = __bus_for_each_drv(bus, start, data, fn); - up_read(&bus->subsys.rwsem); - return ret; + if (!bus) + return -EINVAL; + + klist_iter_init_node(&bus->klist_drivers, &i, + start ? &start->knode_bus : NULL); + while ((drv = next_driver(&i)) && !error) + error = fn(drv, data); + klist_iter_exit(&i); + return error; } static int device_add_attrs(struct bus_type * bus, struct device * dev) @@ -376,6 +367,7 @@ int bus_add_driver(struct device_driver * drv) down_write(&bus->subsys.rwsem); driver_attach(drv); up_write(&bus->subsys.rwsem); + klist_add_tail(&bus->klist_drivers, &drv->knode_bus); module_add_driver(drv->owner, drv); driver_add_attrs(bus, drv); @@ -397,6 +389,7 @@ void bus_remove_driver(struct device_driver * drv) { if (drv->bus) { driver_remove_attrs(drv->bus, drv); + klist_remove(&drv->knode_bus); down_write(&drv->bus->subsys.rwsem); pr_debug("bus %s: remove driver %s\n", drv->bus->name, drv->name); driver_detach(drv); @@ -536,6 +529,7 @@ int bus_register(struct bus_type * bus) goto bus_drivers_fail; klist_init(&bus->klist_devices); + klist_init(&bus->klist_drivers); bus_add_attrs(bus); pr_debug("bus type '%s' registered\n", bus->name); diff --git a/include/linux/device.h b/include/linux/device.h index e36953cf7f14..ea9ab33dfe71 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -53,6 +53,7 @@ struct bus_type { struct kset drivers; struct kset devices; struct klist klist_devices; + struct klist klist_drivers; struct bus_attribute * bus_attrs; struct device_attribute * dev_attrs; @@ -105,6 +106,7 @@ struct device_driver { struct completion unloaded; struct kobject kobj; struct list_head devices; + struct klist_node knode_bus; struct module * owner; -- cgit v1.2.3 From 94e7b1c5ff2055571703e38b059afffe17658432 Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Mon, 21 Mar 2005 12:25:36 -0800 Subject: [PATCH] Add a klist to struct device_driver for the devices bound to it. - Use it in driver_for_each_device() instead of the regular list_head and stop using the bus's rwsem for protection. - Use driver_for_each_device() in driver_detach() so we don't deadlock on the bus's rwsem. - Remove ->devices. - Move klist access and sysfs link access out from under device's semaphore, since they're synchronized through other means. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 2 +- drivers/base/dd.c | 34 +++++++++++++++++++--------------- drivers/base/driver.c | 27 +++++++++++++++------------ include/linux/device.h | 3 ++- 4 files changed, 37 insertions(+), 29 deletions(-) (limited to 'include') diff --git a/drivers/base/core.c b/drivers/base/core.c index 93440824b800..bc5bec61a01a 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -209,8 +209,8 @@ void device_initialize(struct device *dev) kobject_init(&dev->kobj); INIT_LIST_HEAD(&dev->node); INIT_LIST_HEAD(&dev->children); - INIT_LIST_HEAD(&dev->driver_list); INIT_LIST_HEAD(&dev->bus_list); + INIT_LIST_HEAD(&dev->driver_list); INIT_LIST_HEAD(&dev->dma_pools); init_MUTEX(&dev->sem); } diff --git a/drivers/base/dd.c b/drivers/base/dd.c index b709b1e0cb2a..47cbb5641235 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -40,7 +40,7 @@ void device_bind_driver(struct device * dev) { pr_debug("bound device '%s' to driver '%s'\n", dev->bus_id, dev->driver->name); - list_add_tail(&dev->driver_list, &dev->driver->devices); + klist_add_tail(&dev->driver->klist_devices, &dev->knode_driver); sysfs_create_link(&dev->driver->kobj, &dev->kobj, kobject_name(&dev->kobj)); sysfs_create_link(&dev->kobj, &dev->driver->kobj, "driver"); @@ -164,31 +164,35 @@ void driver_attach(struct device_driver * drv) */ void device_release_driver(struct device * dev) { - struct device_driver * drv; + struct device_driver * drv = dev->driver; + + if (!drv) + return; + + sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj)); + sysfs_remove_link(&dev->kobj, "driver"); + klist_remove(&dev->knode_driver); down(&dev->sem); - drv = dev->driver; - if (drv) { - sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj)); - sysfs_remove_link(&dev->kobj, "driver"); - list_del_init(&dev->driver_list); - if (drv->remove) - drv->remove(dev); - dev->driver = NULL; - } + if (drv->remove) + drv->remove(dev); + dev->driver = NULL; up(&dev->sem); } +static int __remove_driver(struct device * dev, void * unused) +{ + device_release_driver(dev); + return 0; +} + /** * driver_detach - detach driver from all devices it controls. * @drv: driver. */ void driver_detach(struct device_driver * drv) { - while (!list_empty(&drv->devices)) { - struct device * dev = container_of(drv->devices.next, struct device, driver_list); - device_release_driver(dev); - } + driver_for_each_device(drv, NULL, NULL, __remove_driver); } diff --git a/drivers/base/driver.c b/drivers/base/driver.c index 484fed1985aa..34bd38aa7eb8 100644 --- a/drivers/base/driver.c +++ b/drivers/base/driver.c @@ -19,6 +19,12 @@ #define to_drv(obj) container_of(obj, struct device_driver, kobj) +static struct device * next_device(struct klist_iter * i) +{ + struct klist_node * n = klist_next(i); + return n ? container_of(n, struct device, knode_driver) : NULL; +} + /** * driver_for_each_device - Iterator for devices bound to a driver. * @drv: Driver we're iterating. @@ -32,21 +38,18 @@ int driver_for_each_device(struct device_driver * drv, struct device * start, void * data, int (*fn)(struct device *, void *)) { - struct list_head * head; + struct klist_iter i; struct device * dev; int error = 0; - down_read(&drv->bus->subsys.rwsem); - head = &drv->devices; - dev = list_prepare_entry(start, head, driver_list); - list_for_each_entry_continue(dev, head, driver_list) { - get_device(dev); + if (!drv) + return -EINVAL; + + klist_iter_init_node(&drv->klist_devices, &i, + start ? &start->knode_driver : NULL); + while ((dev = next_device(&i)) && !error) error = fn(dev, data); - put_device(dev); - if (error) - break; - } - up_read(&drv->bus->subsys.rwsem); + klist_iter_exit(&i); return error; } @@ -120,7 +123,7 @@ void put_driver(struct device_driver * drv) */ int driver_register(struct device_driver * drv) { - INIT_LIST_HEAD(&drv->devices); + klist_init(&drv->klist_devices); init_completion(&drv->unloaded); return bus_add_driver(drv); } diff --git a/include/linux/device.h b/include/linux/device.h index ea9ab33dfe71..96c71b59fdef 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -105,7 +105,7 @@ struct device_driver { struct completion unloaded; struct kobject kobj; - struct list_head devices; + struct klist klist_devices; struct klist_node knode_bus; struct module * owner; @@ -266,6 +266,7 @@ struct device { struct list_head bus_list; /* node in bus's list */ struct list_head driver_list; struct list_head children; + struct klist_node knode_driver; struct klist_node knode_bus; struct device * parent; -- cgit v1.2.3 From cb85b6f1cc811ecb9ed4b950206d8941ba710e68 Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Thu, 24 Mar 2005 10:48:35 -0800 Subject: [PATCH] Remove the unused device_find(). Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 19 ------------------- include/linux/device.h | 1 - 2 files changed, 20 deletions(-) (limited to 'include') diff --git a/drivers/base/core.c b/drivers/base/core.c index bc5bec61a01a..0eb1d424bd1d 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -401,24 +401,6 @@ int device_for_each_child(struct device * dev, void * data, return error; } -/** - * device_find - locate device on a bus by name. - * @name: name of the device. - * @bus: bus to scan for the device. - * - * Call kset_find_obj() to iterate over list of devices on - * a bus to find device by name. Return device if found. - * - * Note that kset_find_obj increments device's reference count. - */ -struct device *device_find(const char *name, struct bus_type *bus) -{ - struct kobject *k = kset_find_obj(&bus->devices, name); - if (k) - return to_dev(k); - return NULL; -} - int __init devices_init(void) { return subsystem_register(&devices_subsys); @@ -434,7 +416,6 @@ EXPORT_SYMBOL_GPL(device_del); EXPORT_SYMBOL_GPL(device_unregister); EXPORT_SYMBOL_GPL(get_device); EXPORT_SYMBOL_GPL(put_device); -EXPORT_SYMBOL_GPL(device_find); EXPORT_SYMBOL_GPL(device_create_file); EXPORT_SYMBOL_GPL(device_remove_file); diff --git a/include/linux/device.h b/include/linux/device.h index 96c71b59fdef..f1c38d869ac9 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -372,7 +372,6 @@ extern int (*platform_notify_remove)(struct device * dev); */ extern struct device * get_device(struct device * dev); extern void put_device(struct device * dev); -extern struct device *device_find(const char *name, struct bus_type *bus); /* drivers/base/platform.c */ -- cgit v1.2.3 From 8b0c250be489dcbf1a3a33bb4ec4c7f33735a365 Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Thu, 24 Mar 2005 12:58:57 -0800 Subject: [PATCH] add klist_node_attached() to determine if a node is on a list or not. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman diff -Nru a/include/linux/klist.h b/include/linux/klist.h --- include/linux/klist.h | 2 ++ lib/klist.c | 16 ++++++++++++++++ 2 files changed, 18 insertions(+) (limited to 'include') diff --git a/include/linux/klist.h b/include/linux/klist.h index fb52f9d9d611..eebf5e5696ec 100644 --- a/include/linux/klist.h +++ b/include/linux/klist.h @@ -37,6 +37,8 @@ extern void klist_add_head(struct klist * k, struct klist_node * n); extern void klist_del(struct klist_node * n); extern void klist_remove(struct klist_node * n); +extern int klist_node_attached(struct klist_node * n); + struct klist_iter { struct klist * i_klist; diff --git a/lib/klist.c b/lib/klist.c index 6f760424648b..02177d72dc89 100644 --- a/lib/klist.c +++ b/lib/klist.c @@ -112,6 +112,7 @@ static void klist_release(struct kref * kref) struct klist_node * n = container_of(kref, struct klist_node, n_ref); list_del(&n->n_node); complete(&n->n_removed); + n->n_klist = NULL; } static int klist_dec_and_del(struct klist_node * n) @@ -153,6 +154,19 @@ void klist_remove(struct klist_node * n) EXPORT_SYMBOL_GPL(klist_remove); +/** + * klist_node_attached - Say whether a node is bound to a list or not. + * @n: Node that we're testing. + */ + +int klist_node_attached(struct klist_node * n) +{ + return (n->n_klist != NULL); +} + +EXPORT_SYMBOL_GPL(klist_node_attached); + + /** * klist_iter_init_node - Initialize a klist_iter structure. * @k: klist we're iterating. @@ -246,3 +260,5 @@ struct klist_node * klist_next(struct klist_iter * i) } EXPORT_SYMBOL_GPL(klist_next); + + -- cgit v1.2.3 From 7dc35cdf69149a7f2b5216ada9bafe359746ac1c Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Thu, 24 Mar 2005 13:03:35 -0800 Subject: [PATCH] Remove struct device::bus_list. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 1 - include/linux/device.h | 1 - 2 files changed, 2 deletions(-) (limited to 'include') diff --git a/drivers/base/core.c b/drivers/base/core.c index 0eb1d424bd1d..f258a21ed78a 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -209,7 +209,6 @@ void device_initialize(struct device *dev) kobject_init(&dev->kobj); INIT_LIST_HEAD(&dev->node); INIT_LIST_HEAD(&dev->children); - INIT_LIST_HEAD(&dev->bus_list); INIT_LIST_HEAD(&dev->driver_list); INIT_LIST_HEAD(&dev->dma_pools); init_MUTEX(&dev->sem); diff --git a/include/linux/device.h b/include/linux/device.h index f1c38d869ac9..13f65853e583 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -263,7 +263,6 @@ extern void class_device_destroy(struct class *cls, dev_t devt); struct device { struct list_head node; /* node in sibling list */ - struct list_head bus_list; /* node in bus's list */ struct list_head driver_list; struct list_head children; struct klist_node knode_driver; -- cgit v1.2.3 From 63c4f204ffc8219696bda88eb20c9873d007a2fc Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Thu, 24 Mar 2005 13:08:05 -0800 Subject: [PATCH] Remove struct device::driver_list. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 1 - include/linux/device.h | 1 - 2 files changed, 2 deletions(-) (limited to 'include') diff --git a/drivers/base/core.c b/drivers/base/core.c index f258a21ed78a..bfa4268289ea 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -209,7 +209,6 @@ void device_initialize(struct device *dev) kobject_init(&dev->kobj); INIT_LIST_HEAD(&dev->node); INIT_LIST_HEAD(&dev->children); - INIT_LIST_HEAD(&dev->driver_list); INIT_LIST_HEAD(&dev->dma_pools); init_MUTEX(&dev->sem); } diff --git a/include/linux/device.h b/include/linux/device.h index 13f65853e583..d2434934d091 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -263,7 +263,6 @@ extern void class_device_destroy(struct class *cls, dev_t devt); struct device { struct list_head node; /* node in sibling list */ - struct list_head driver_list; struct list_head children; struct klist_node knode_driver; struct klist_node knode_bus; -- cgit v1.2.3 From 36239577cfb6b9a7c111209536b54200b0252ebf Mon Sep 17 00:00:00 2001 From: "mochel@digitalimplant.org" Date: Thu, 24 Mar 2005 19:08:30 -0800 Subject: [PATCH] Use a klist for device child lists. - Use klist iterator in device_for_each_child(), making it safe to use for removing devices. - Remove unused list_to_dev() function. - Kills all usage of devices_subsys.rwsem. Signed-off-by: Patrick Mochel Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 30 +++++++++++++++--------------- include/linux/device.h | 10 ++-------- 2 files changed, 17 insertions(+), 23 deletions(-) (limited to 'include') diff --git a/drivers/base/core.c b/drivers/base/core.c index bfa4268289ea..1d8c7790b55a 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -207,8 +207,7 @@ void device_initialize(struct device *dev) { kobj_set_kset_s(dev, devices_subsys); kobject_init(&dev->kobj); - INIT_LIST_HEAD(&dev->node); - INIT_LIST_HEAD(&dev->children); + klist_init(&dev->klist_children); INIT_LIST_HEAD(&dev->dma_pools); init_MUTEX(&dev->sem); } @@ -249,10 +248,8 @@ int device_add(struct device *dev) goto PMError; if ((error = bus_add_device(dev))) goto BusError; - down_write(&devices_subsys.rwsem); if (parent) - list_add_tail(&dev->node, &parent->children); - up_write(&devices_subsys.rwsem); + klist_add_tail(&parent->klist_children, &dev->knode_parent); /* notify platform of device entry */ if (platform_notify) @@ -335,10 +332,8 @@ void device_del(struct device * dev) { struct device * parent = dev->parent; - down_write(&devices_subsys.rwsem); if (parent) - list_del_init(&dev->node); - up_write(&devices_subsys.rwsem); + klist_remove(&dev->knode_parent); /* Notify the platform of the removal, in case they * need to do anything... @@ -372,6 +367,12 @@ void device_unregister(struct device * dev) } +static struct device * next_device(struct klist_iter * i) +{ + struct klist_node * n = klist_next(i); + return n ? container_of(n, struct device, knode_parent) : NULL; +} + /** * device_for_each_child - device child iterator. * @dev: parent struct device. @@ -384,18 +385,17 @@ void device_unregister(struct device * dev) * We check the return of @fn each time. If it returns anything * other than 0, we break out and return that value. */ -int device_for_each_child(struct device * dev, void * data, +int device_for_each_child(struct device * parent, void * data, int (*fn)(struct device *, void *)) { + struct klist_iter i; struct device * child; int error = 0; - down_read(&devices_subsys.rwsem); - list_for_each_entry(child, &dev->children, node) { - if((error = fn(child, data))) - break; - } - up_read(&devices_subsys.rwsem); + klist_iter_init(&parent->klist_children, &i); + while ((child = next_device(&i)) && !error) + error = fn(child, data); + klist_iter_exit(&i); return error; } diff --git a/include/linux/device.h b/include/linux/device.h index d2434934d091..43249260cd1c 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -262,8 +262,8 @@ extern void class_device_destroy(struct class *cls, dev_t devt); struct device { - struct list_head node; /* node in sibling list */ - struct list_head children; + struct klist klist_children; + struct klist_node knode_parent; /* node in sibling list */ struct klist_node knode_driver; struct klist_node knode_bus; struct device * parent; @@ -298,12 +298,6 @@ struct device { void (*release)(struct device * dev); }; -static inline struct device * -list_to_dev(struct list_head *node) -{ - return list_entry(node, struct device, node); -} - static inline void * dev_get_drvdata (struct device *dev) { -- cgit v1.2.3 From 0d3e5a2e39b6ba2974e9e7c2a429018c45de8e76 Mon Sep 17 00:00:00 2001 From: Patrick Mochel Date: Tue, 5 Apr 2005 23:46:33 -0700 Subject: [PATCH] Driver Core: fix bk-driver-core kills ppc64 There's no check to see if the device is already bound to a driver, which could do bad things. The first thing to go wrong is that it will try to match a driver with a device already bound to one. In some cases (it appears with USB with drivers/usb/core/usb.c::usb_match_id()), some drivers will match a device based on the class type, so it would be common (especially for HID devices) to match a device that is already bound. The fun comes when ->probe() is called, it fails, then driver_probe_device() does this: dev->driver = NULL; Later on, that pointer could be be dereferenced without checking and cause hell to break loose. This problem could be nasty. It's very hardware dependent, since some devices could have a different set of matching qualifiers than others. Now, I don't quite see exactly where/how you were getting that crash. You're dereferencing bad memory, but I'm not sure which pointer was bad and where it came from, but it could have come from a couple of different places. The patch below will hopefully fix it all up for you. It's against 2.6.12-rc2-mm1, and does the following: - Move logic to driver_probe_device() and comments uncommon returns: 1 - If device is bound 0 - If device not bound, and no error error - If there was an error. - Move locking to caller of that function, since we want to lock a device for the entire time we're trying to bind it to a driver (to prevent against a driver being loaded at the same time). - Update __device_attach() and __driver_attach() to do that locking. - Check if device is already bound in __driver_attach() - Update the converse device_release_driver() so it locks the device around all of the operations. - Mark driver_probe_device() as static and remove export. It's an internal function, it should stay that way, and there are no other callers. If there is ever a need to export it, we can audit it as necessary. Signed-off-by: Andrew Morton --- drivers/base/dd.c | 142 ++++++++++++++++++++++++++----------------------- include/linux/device.h | 1 - 2 files changed, 74 insertions(+), 69 deletions(-) (limited to 'include') diff --git a/drivers/base/dd.c b/drivers/base/dd.c index dd2a8a79c121..8510918109e0 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -35,6 +35,8 @@ * nor take the bus's rwsem. Please verify those are accounted * for before calling this. (It is ok to call with no other effort * from a driver's probe() method.) + * + * This function must be called with @dev->sem held. */ void device_bind_driver(struct device * dev) { @@ -57,54 +59,56 @@ void device_bind_driver(struct device * dev) * because we don't know the format of the ID structures, nor what * is to be considered a match and what is not. * - * If we find a match, we call @drv->probe(@dev) if it exists, and - * call device_bind_driver() above. + * + * This function returns 1 if a match is found, an error if one + * occurs (that is not -ENODEV or -ENXIO), and 0 otherwise. + * + * This function must be called with @dev->sem held. */ -int driver_probe_device(struct device_driver * drv, struct device * dev) +static int driver_probe_device(struct device_driver * drv, struct device * dev) { - int error = 0; + int ret = 0; if (drv->bus->match && !drv->bus->match(dev, drv)) - return -ENODEV; + goto Done; - down(&dev->sem); + pr_debug("%s: Matched Device %s with Driver %s\n", + drv->bus->name, dev->bus_id, drv->name); dev->driver = drv; if (drv->probe) { - error = drv->probe(dev); - if (error) { + ret = drv->probe(dev); + if (ret) { dev->driver = NULL; - up(&dev->sem); - return error; + goto ProbeFailed; } } - up(&dev->sem); device_bind_driver(dev); - return 0; + ret = 1; + pr_debug("%s: Bound Device %s to Driver %s\n", + drv->bus->name, dev->bus_id, drv->name); + goto Done; + + ProbeFailed: + if (ret == -ENODEV || ret == -ENXIO) { + /* Driver matched, but didn't support device + * or device not found. + * Not an error; keep going. + */ + ret = 0; + } else { + /* driver matched but the probe failed */ + printk(KERN_WARNING + "%s: probe of %s failed with error %d\n", + drv->name, dev->bus_id, ret); + } + Done: + return ret; } static int __device_attach(struct device_driver * drv, void * data) { struct device * dev = data; - int error; - - error = driver_probe_device(drv, dev); - if (error) { - if ((error == -ENODEV) || (error == -ENXIO)) { - /* Driver matched, but didn't support device - * or device not found. - * Not an error; keep going. - */ - error = 0; - } else { - /* driver matched but the probe failed */ - printk(KERN_WARNING - "%s: probe of %s failed with error %d\n", - drv->name, dev->bus_id, error); - } - return error; - } - /* stop looking, this device is attached */ - return 1; + return driver_probe_device(drv, dev); } /** @@ -114,37 +118,43 @@ static int __device_attach(struct device_driver * drv, void * data) * Walk the list of drivers that the bus has and call * driver_probe_device() for each pair. If a compatible * pair is found, break out and return. + * + * Returns 1 if the device was bound to a driver; 0 otherwise. */ int device_attach(struct device * dev) { + int ret = 0; + + down(&dev->sem); if (dev->driver) { device_bind_driver(dev); - return 1; - } - - return bus_for_each_drv(dev->bus, NULL, dev, __device_attach); + ret = 1; + } else + ret = bus_for_each_drv(dev->bus, NULL, dev, __device_attach); + up(&dev->sem); + return ret; } static int __driver_attach(struct device * dev, void * data) { struct device_driver * drv = data; - int error = 0; - - if (!dev->driver) { - error = driver_probe_device(drv, dev); - if (error) { - if (error != -ENODEV) { - /* driver matched but the probe failed */ - printk(KERN_WARNING - "%s: probe of %s failed with error %d\n", - drv->name, dev->bus_id, error); - } else - error = 0; - return error; - } - /* stop looking, this driver is attached */ - return 1; - } + + /* + * Lock device and try to bind to it. We drop the error + * here and always return 0, because we need to keep trying + * to bind to devices and some drivers will return an error + * simply if it didn't support the device. + * + * driver_probe_device() will spit a warning if there + * is an error. + */ + + down(&dev->sem); + if (!dev->driver) + driver_probe_device(drv, dev); + up(&dev->sem); + + return 0; } @@ -156,9 +166,6 @@ static int __driver_attach(struct device * dev, void * data) * match the driver with each one. If driver_probe_device() * returns 0 and the @dev->driver is set, we've found a * compatible pair. - * - * Note that we ignore the -ENODEV error from driver_probe_device(), - * since it's perfectly valid for a driver not to bind to any devices. */ void driver_attach(struct device_driver * drv) { @@ -176,19 +183,19 @@ void driver_attach(struct device_driver * drv) */ void device_release_driver(struct device * dev) { - struct device_driver * drv = dev->driver; - - if (!drv) - return; - - sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj)); - sysfs_remove_link(&dev->kobj, "driver"); - klist_del(&dev->knode_driver); + struct device_driver * drv; down(&dev->sem); - if (drv->remove) - drv->remove(dev); - dev->driver = NULL; + if (dev->driver) { + drv = dev->driver; + sysfs_remove_link(&drv->kobj, kobject_name(&dev->kobj)); + sysfs_remove_link(&dev->kobj, "driver"); + klist_del(&dev->knode_driver); + + if (drv->remove) + drv->remove(dev); + dev->driver = NULL; + } up(&dev->sem); } @@ -208,7 +215,6 @@ void driver_detach(struct device_driver * drv) } -EXPORT_SYMBOL_GPL(driver_probe_device); EXPORT_SYMBOL_GPL(device_bind_driver); EXPORT_SYMBOL_GPL(device_release_driver); EXPORT_SYMBOL_GPL(device_attach); diff --git a/include/linux/device.h b/include/linux/device.h index 43249260cd1c..91aac349b9a7 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -325,7 +325,6 @@ extern int device_for_each_child(struct device *, void *, * Manual binding of a device to driver. See drivers/base/bus.c * for information on use. */ -extern int driver_probe_device(struct device_driver * drv, struct device * dev); extern void device_bind_driver(struct device * dev); extern void device_release_driver(struct device * dev); extern int device_attach(struct device * dev); -- cgit v1.2.3 From 4b45099b75832434c5113b9aed1499f8a69d13d5 Mon Sep 17 00:00:00 2001 From: Keiichiro Tokunaga Date: Sun, 8 May 2005 21:28:53 +0900 Subject: [PATCH] Driver core: unregister_node() for hotplug use This adds a generic function 'unregister_node()'. It is used to remove objects of a node going away for hotplug. All the devices on the node must be unregistered before calling this function. Signed-off-by: Keiichiro Tokunaga Signed-off-by: Greg Kroah-Hartman diff -puN drivers/base/node.c~numa_hp_base drivers/base/node.c --- drivers/base/node.c | 20 ++++++++++++++++++-- include/linux/node.h | 1 + 2 files changed, 19 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/base/node.c b/drivers/base/node.c index 583d57ec49a8..5d4517ccc422 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -136,7 +136,7 @@ static SYSDEV_ATTR(distance, S_IRUGO, node_read_distance, NULL); * * Initialize and register the node device. */ -int __init register_node(struct node *node, int num, struct node *parent) +int register_node(struct node *node, int num, struct node *parent) { int error; @@ -153,8 +153,24 @@ int __init register_node(struct node *node, int num, struct node *parent) return error; } +/** + * unregister_node - unregister a node device + * @node: node going away + * + * Unregisters a node device @node. All the devices on the node must be + * unregistered before calling this function. + */ +void unregister_node(struct node *node) +{ + sysdev_remove_file(&node->sysdev, &attr_cpumap); + sysdev_remove_file(&node->sysdev, &attr_meminfo); + sysdev_remove_file(&node->sysdev, &attr_numastat); + sysdev_remove_file(&node->sysdev, &attr_distance); + + sysdev_unregister(&node->sysdev); +} -int __init register_node_type(void) +static int __init register_node_type(void) { return sysdev_class_register(&node_class); } diff --git a/include/linux/node.h b/include/linux/node.h index 6e0a697e594e..254dc3de650b 100644 --- a/include/linux/node.h +++ b/include/linux/node.h @@ -27,6 +27,7 @@ struct node { }; extern int register_node(struct node *, int, struct node *); +extern void unregister_node(struct node *node); #define to_node(sys_device) container_of(sys_device, struct node, sysdev) -- cgit v1.2.3 From acaefc25d21f850e47ecc5098d1e0bc442c526be Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 18 May 2005 14:40:59 +0200 Subject: [PATCH] libfs: add simple attribute files Based on the discussion about spufs attributes, this is my suggestion for a more generic attribute file support that can be used by both debugfs and spufs. Simple attribute files behave similarly to sequential files from a kernel programmers perspective in that a standard set of file operations is provided and only an open operation needs to be written that registers file specific get() and set() functions. These operations are defined as void foo_set(void *data, u64 val); and u64 foo_get(void *data); where data is the inode->u.generic_ip pointer of the file and the operations just need to make send of that pointer. The infrastructure makes sure this works correctly with concurrent access and partial read calls. A macro named DEFINE_SIMPLE_ATTRIBUTE is provided to further simplify using the attributes. This patch already contains the changes for debugfs to use attributes for its internal file operations. Signed-off-by: Arnd Bergmann Signed-off-by: Greg Kroah-Hartman --- fs/debugfs/file.c | 67 ++++++++++++++++------------------- fs/libfs.c | 100 +++++++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/fs.h | 46 ++++++++++++++++++++++++ 3 files changed, 175 insertions(+), 38 deletions(-) (limited to 'include') diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index 548556ff2506..efc97d9b7860 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -45,44 +45,15 @@ struct file_operations debugfs_file_operations = { .open = default_open, }; -#define simple_type(type, format, temptype, strtolfn) \ -static ssize_t read_file_##type(struct file *file, char __user *user_buf, \ - size_t count, loff_t *ppos) \ -{ \ - char buf[32]; \ - type *val = file->private_data; \ - \ - snprintf(buf, sizeof(buf), format "\n", *val); \ - return simple_read_from_buffer(user_buf, count, ppos, buf, strlen(buf));\ -} \ -static ssize_t write_file_##type(struct file *file, const char __user *user_buf,\ - size_t count, loff_t *ppos) \ -{ \ - char *endp; \ - char buf[32]; \ - int buf_size; \ - type *val = file->private_data; \ - temptype tmp; \ - \ - memset(buf, 0x00, sizeof(buf)); \ - buf_size = min(count, (sizeof(buf)-1)); \ - if (copy_from_user(buf, user_buf, buf_size)) \ - return -EFAULT; \ - \ - tmp = strtolfn(buf, &endp, 0); \ - if ((endp == buf) || ((type)tmp != tmp)) \ - return -EINVAL; \ - *val = tmp; \ - return count; \ -} \ -static struct file_operations fops_##type = { \ - .read = read_file_##type, \ - .write = write_file_##type, \ - .open = default_open, \ -}; -simple_type(u8, "%c", unsigned long, simple_strtoul); -simple_type(u16, "%hi", unsigned long, simple_strtoul); -simple_type(u32, "%i", unsigned long, simple_strtoul); +static void debugfs_u8_set(void *data, u64 val) +{ + *(u8 *)data = val; +} +static u64 debugfs_u8_get(void *data) +{ + return *(u8 *)data; +} +DEFINE_SIMPLE_ATTRIBUTE(fops_u8, debugfs_u8_get, debugfs_u8_set, "%llu\n"); /** * debugfs_create_u8 - create a file in the debugfs filesystem that is used to read and write a unsigned 8 bit value. @@ -116,6 +87,16 @@ struct dentry *debugfs_create_u8(const char *name, mode_t mode, } EXPORT_SYMBOL_GPL(debugfs_create_u8); +static void debugfs_u16_set(void *data, u64 val) +{ + *(u16 *)data = val; +} +static u64 debugfs_u16_get(void *data) +{ + return *(u16 *)data; +} +DEFINE_SIMPLE_ATTRIBUTE(fops_u16, debugfs_u16_get, debugfs_u16_set, "%llu\n"); + /** * debugfs_create_u16 - create a file in the debugfs filesystem that is used to read and write a unsigned 8 bit value. * @@ -148,6 +129,16 @@ struct dentry *debugfs_create_u16(const char *name, mode_t mode, } EXPORT_SYMBOL_GPL(debugfs_create_u16); +static void debugfs_u32_set(void *data, u64 val) +{ + *(u32 *)data = val; +} +static u64 debugfs_u32_get(void *data) +{ + return *(u32 *)data; +} +DEFINE_SIMPLE_ATTRIBUTE(fops_u32, debugfs_u32_get, debugfs_u32_set, "%llu\n"); + /** * debugfs_create_u32 - create a file in the debugfs filesystem that is used to read and write a unsigned 8 bit value. * diff --git a/fs/libfs.c b/fs/libfs.c index f90b29595927..5025563e7379 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -519,6 +519,102 @@ int simple_transaction_release(struct inode *inode, struct file *file) return 0; } +/* Simple attribute files */ + +struct simple_attr { + u64 (*get)(void *); + void (*set)(void *, u64); + char get_buf[24]; /* enough to store a u64 and "\n\0" */ + char set_buf[24]; + void *data; + const char *fmt; /* format for read operation */ + struct semaphore sem; /* protects access to these buffers */ +}; + +/* simple_attr_open is called by an actual attribute open file operation + * to set the attribute specific access operations. */ +int simple_attr_open(struct inode *inode, struct file *file, + u64 (*get)(void *), void (*set)(void *, u64), + const char *fmt) +{ + struct simple_attr *attr; + + attr = kmalloc(sizeof(*attr), GFP_KERNEL); + if (!attr) + return -ENOMEM; + + attr->get = get; + attr->set = set; + attr->data = inode->u.generic_ip; + attr->fmt = fmt; + init_MUTEX(&attr->sem); + + file->private_data = attr; + + return nonseekable_open(inode, file); +} + +int simple_attr_close(struct inode *inode, struct file *file) +{ + kfree(file->private_data); + return 0; +} + +/* read from the buffer that is filled with the get function */ +ssize_t simple_attr_read(struct file *file, char __user *buf, + size_t len, loff_t *ppos) +{ + struct simple_attr *attr; + size_t size; + ssize_t ret; + + attr = file->private_data; + + if (!attr->get) + return -EACCES; + + down(&attr->sem); + if (*ppos) /* continued read */ + size = strlen(attr->get_buf); + else /* first read */ + size = scnprintf(attr->get_buf, sizeof(attr->get_buf), + attr->fmt, + (unsigned long long)attr->get(attr->data)); + + ret = simple_read_from_buffer(buf, len, ppos, attr->get_buf, size); + up(&attr->sem); + return ret; +} + +/* interpret the buffer as a number to call the set function with */ +ssize_t simple_attr_write(struct file *file, const char __user *buf, + size_t len, loff_t *ppos) +{ + struct simple_attr *attr; + u64 val; + size_t size; + ssize_t ret; + + attr = file->private_data; + + if (!attr->set) + return -EACCES; + + down(&attr->sem); + ret = -EFAULT; + size = min(sizeof(attr->set_buf) - 1, len); + if (copy_from_user(attr->set_buf, buf, size)) + goto out; + + ret = len; /* claim we got the whole input */ + attr->set_buf[size] = '\0'; + val = simple_strtol(attr->set_buf, NULL, 0); + attr->set(attr->data, val); +out: + up(&attr->sem); + return ret; +} + EXPORT_SYMBOL(dcache_dir_close); EXPORT_SYMBOL(dcache_dir_lseek); EXPORT_SYMBOL(dcache_dir_open); @@ -547,3 +643,7 @@ EXPORT_SYMBOL(simple_read_from_buffer); EXPORT_SYMBOL(simple_transaction_get); EXPORT_SYMBOL(simple_transaction_read); EXPORT_SYMBOL(simple_transaction_release); +EXPORT_SYMBOL_GPL(simple_attr_open); +EXPORT_SYMBOL_GPL(simple_attr_close); +EXPORT_SYMBOL_GPL(simple_attr_read); +EXPORT_SYMBOL_GPL(simple_attr_write); diff --git a/include/linux/fs.h b/include/linux/fs.h index 0180102dace1..9b8b696d4f15 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1657,6 +1657,52 @@ static inline void simple_transaction_set(struct file *file, size_t n) ar->size = n; } +/* + * simple attribute files + * + * These attributes behave similar to those in sysfs: + * + * Writing to an attribute immediately sets a value, an open file can be + * written to multiple times. + * + * Reading from an attribute creates a buffer from the value that might get + * read with multiple read calls. When the attribute has been read + * completely, no further read calls are possible until the file is opened + * again. + * + * All attributes contain a text representation of a numeric value + * that are accessed with the get() and set() functions. + */ +#define DEFINE_SIMPLE_ATTRIBUTE(__fops, __get, __set, __fmt) \ +static int __fops ## _open(struct inode *inode, struct file *file) \ +{ \ + __simple_attr_check_format(__fmt, 0ull); \ + return simple_attr_open(inode, file, __get, __set, __fmt); \ +} \ +static struct file_operations __fops = { \ + .owner = THIS_MODULE, \ + .open = __fops ## _open, \ + .release = simple_attr_close, \ + .read = simple_attr_read, \ + .write = simple_attr_write, \ +}; + +static inline void __attribute__((format(printf, 1, 2))) +__simple_attr_check_format(const char *fmt, ...) +{ + /* don't do anything, just let the compiler check the arguments; */ +} + +int simple_attr_open(struct inode *inode, struct file *file, + u64 (*get)(void *), void (*set)(void *, u64), + const char *fmt); +int simple_attr_close(struct inode *inode, struct file *file); +ssize_t simple_attr_read(struct file *file, char __user *buf, + size_t len, loff_t *ppos); +ssize_t simple_attr_write(struct file *file, const char __user *buf, + size_t len, loff_t *ppos); + + #ifdef CONFIG_SECURITY static inline char *alloc_secdata(void) { -- cgit v1.2.3 From 54b6f35c99974e99e64c05c2895718355123c55f Mon Sep 17 00:00:00 2001 From: Yani Ioannou Date: Tue, 17 May 2005 06:39:34 -0400 Subject: [PATCH] Driver core: change device_attribute callbacks This patch adds the device_attribute paramerter to the device_attribute store and show sysfs callback functions, and passes a reference to the attribute when the callbacks are called. Signed-off-by: Yani Ioannou Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 4 ++-- include/linux/device.h | 6 ++++-- 2 files changed, 6 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/drivers/base/core.c b/drivers/base/core.c index 1d8c7790b55a..86d79755fbfb 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -39,7 +39,7 @@ dev_attr_show(struct kobject * kobj, struct attribute * attr, char * buf) ssize_t ret = -EIO; if (dev_attr->show) - ret = dev_attr->show(dev, buf); + ret = dev_attr->show(dev, dev_attr, buf); return ret; } @@ -52,7 +52,7 @@ dev_attr_store(struct kobject * kobj, struct attribute * attr, ssize_t ret = -EIO; if (dev_attr->store) - ret = dev_attr->store(dev, buf, count); + ret = dev_attr->store(dev, dev_attr, buf, count); return ret; } diff --git a/include/linux/device.h b/include/linux/device.h index 91aac349b9a7..7b781a72b293 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -335,8 +335,10 @@ extern void driver_attach(struct device_driver * drv); struct device_attribute { struct attribute attr; - ssize_t (*show)(struct device * dev, char * buf); - ssize_t (*store)(struct device * dev, const char * buf, size_t count); + ssize_t (*show)(struct device *dev, struct device_attribute *attr, + char *buf); + ssize_t (*store)(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count); }; #define DEVICE_ATTR(_name,_mode,_show,_store) \ -- cgit v1.2.3 From f2d03e1b3f00f1c5971463ab0101bef0c521ad3b Mon Sep 17 00:00:00 2001 From: Yani Ioannou Date: Tue, 17 May 2005 06:44:59 -0400 Subject: [PATCH] Driver Core: include: update device attribute callbacks Signed-off-by: Yani Ioannou Signed-off-by: Greg Kroah-Hartman --- include/asm-ppc/ocp.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-ppc/ocp.h b/include/asm-ppc/ocp.h index b98db3cdae83..c726f1845190 100644 --- a/include/asm-ppc/ocp.h +++ b/include/asm-ppc/ocp.h @@ -189,7 +189,7 @@ extern void ocp_for_each_device(void(*callback)(struct ocp_device *, void *arg), /* Sysfs support */ #define OCP_SYSFS_ADDTL(type, format, name, field) \ static ssize_t \ -show_##name##_##field(struct device *dev, char *buf) \ +show_##name##_##field(struct device *dev, struct device_attribute *attr, char *buf) \ { \ struct ocp_device *odev = to_ocp_dev(dev); \ type *add = odev->def->additions; \ -- cgit v1.2.3 From 0a3e7eeabc9f76b7496488aad2d11626ff6a2a4f Mon Sep 17 00:00:00 2001 From: Yani Ioannou Date: Tue, 17 May 2005 22:59:05 -0400 Subject: [PATCH] I2C: add i2c sensor_device_attribute and macros This patch creates a new header with a potential standard i2c sensor attribute type (which simply includes an int representing the sensor number/index) and the associated macros, SENSOR_DEVICE_ATTR to define a static attribute and to_sensor_dev_attr to get a sensor_device_attribute reference from an embedded device_attribute reference. Signed-off-by: Yani Ioannou --- include/linux/i2c-sysfs.h | 36 ++++++++++++++++++++++++++++++++++++ 1 file changed, 36 insertions(+) create mode 100644 include/linux/i2c-sysfs.h (limited to 'include') diff --git a/include/linux/i2c-sysfs.h b/include/linux/i2c-sysfs.h new file mode 100644 index 000000000000..d7bf6ce11679 --- /dev/null +++ b/include/linux/i2c-sysfs.h @@ -0,0 +1,36 @@ +/* + * i2c-sysfs.h - i2c chip driver sysfs defines + * + * Copyright (C) 2005 Yani Ioannou + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ +#ifndef _LINUX_I2C_SYSFS_H +#define _LINUX_I2C_SYSFS_H + +struct sensor_device_attribute{ + struct device_attribute dev_attr; + int index; +}; +#define to_sensor_dev_attr(_dev_attr) \ + container_of(_dev_attr, struct sensor_device_attribute, dev_attr) + +#define SENSOR_DEVICE_ATTR(_name,_mode,_show,_store,_index) \ +struct sensor_device_attribute sensor_dev_attr_##_name = { \ + .dev_attr = __ATTR(_name,_mode,_show,_store), \ + .index = _index, \ +} + +#endif /* _LINUX_I2C_SYSFS_H */ -- cgit v1.2.3 From 988d186de5b6966a71a8cc52e6cb4895fd2f7799 Mon Sep 17 00:00:00 2001 From: Maneesh Soni Date: Tue, 31 May 2005 10:39:14 +0530 Subject: [PATCH] sysfs-iattr: add sysfs_setattr o This adds ->i_op->setattr VFS method for sysfs inodes. The changed attribues are saved in the persistent sysfs_dirent structure as a pointer to struct iattr. The struct iattr is allocated only for those sysfs_dirent's for which default attributes are getting changed. Thanks to Jon Smirl for this suggestion. Signed-off-by: Maneesh Soni Signed-off-by: Greg Kroah-Hartman --- fs/sysfs/dir.c | 1 + fs/sysfs/inode.c | 65 +++++++++++++++++++++++++++++++++++++++++++++++++++ fs/sysfs/sysfs.h | 2 ++ include/linux/sysfs.h | 1 + 4 files changed, 69 insertions(+) (limited to 'include') diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index 5cf3270014c0..37d7a6875d86 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -233,6 +233,7 @@ static struct dentry * sysfs_lookup(struct inode *dir, struct dentry *dentry, struct inode_operations sysfs_dir_inode_operations = { .lookup = sysfs_lookup, + .setattr = sysfs_setattr, }; static void remove_dir(struct dentry * d) diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index aff7b2dfa8ee..400c90be5685 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -26,6 +26,71 @@ static struct backing_dev_info sysfs_backing_dev_info = { .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK, }; +static struct inode_operations sysfs_inode_operations ={ + .setattr = sysfs_setattr, +}; + +int sysfs_setattr(struct dentry * dentry, struct iattr * iattr) +{ + struct inode * inode = dentry->d_inode; + struct sysfs_dirent * sd = dentry->d_fsdata; + struct iattr * sd_iattr; + unsigned int ia_valid = iattr->ia_valid; + int error; + + if (!sd) + return -EINVAL; + + sd_iattr = sd->s_iattr; + + error = inode_change_ok(inode, iattr); + if (error) + return error; + + error = inode_setattr(inode, iattr); + if (error) + return error; + + if (!sd_iattr) { + /* setting attributes for the first time, allocate now */ + sd_iattr = kmalloc(sizeof(struct iattr), GFP_KERNEL); + if (!sd_iattr) + return -ENOMEM; + /* assign default attributes */ + memset(sd_iattr, 0, sizeof(struct iattr)); + sd_iattr->ia_mode = sd->s_mode; + sd_iattr->ia_uid = 0; + sd_iattr->ia_gid = 0; + sd_iattr->ia_atime = sd_iattr->ia_mtime = sd_iattr->ia_ctime = CURRENT_TIME; + sd->s_iattr = sd_iattr; + } + + /* attributes were changed atleast once in past */ + + if (ia_valid & ATTR_UID) + sd_iattr->ia_uid = iattr->ia_uid; + if (ia_valid & ATTR_GID) + sd_iattr->ia_gid = iattr->ia_gid; + if (ia_valid & ATTR_ATIME) + sd_iattr->ia_atime = timespec_trunc(iattr->ia_atime, + inode->i_sb->s_time_gran); + if (ia_valid & ATTR_MTIME) + sd_iattr->ia_mtime = timespec_trunc(iattr->ia_mtime, + inode->i_sb->s_time_gran); + if (ia_valid & ATTR_CTIME) + sd_iattr->ia_ctime = timespec_trunc(iattr->ia_ctime, + inode->i_sb->s_time_gran); + if (ia_valid & ATTR_MODE) { + umode_t mode = iattr->ia_mode; + + if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID)) + mode &= ~S_ISGID; + sd_iattr->ia_mode = mode; + } + + return error; +} + struct inode * sysfs_new_inode(mode_t mode) { struct inode * inode = new_inode(sysfs_sb); diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h index a8a24a0c0b3b..e8e9f0c2573e 100644 --- a/fs/sysfs/sysfs.h +++ b/fs/sysfs/sysfs.h @@ -17,6 +17,7 @@ extern void sysfs_remove_subdir(struct dentry *); extern const unsigned char * sysfs_get_name(struct sysfs_dirent *sd); extern void sysfs_drop_dentry(struct sysfs_dirent *sd, struct dentry *parent); +extern int sysfs_setattr(struct dentry *dentry, struct iattr *iattr); extern struct rw_semaphore sysfs_rename_sem; extern struct super_block * sysfs_sb; @@ -75,6 +76,7 @@ static inline void release_sysfs_dirent(struct sysfs_dirent * sd) kobject_put(sl->target_kobj); kfree(sl); } + kfree(sd->s_iattr); kmem_cache_free(sysfs_dir_cachep, sd); } diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h index d9cd2d31d377..392da5a6dacb 100644 --- a/include/linux/sysfs.h +++ b/include/linux/sysfs.h @@ -73,6 +73,7 @@ struct sysfs_dirent { int s_type; umode_t s_mode; struct dentry * s_dentry; + struct iattr * s_iattr; }; #define SYSFS_ROOT 0x0001 -- cgit v1.2.3