From b0e98aa9c411585eb586b2fa98873c936735008e Mon Sep 17 00:00:00 2001 From: Gerald Schaefer Date: Tue, 20 Oct 2020 20:20:07 +0200 Subject: s390/mm: make pmd/pud_deref() large page aware pmd/pud_deref() assume that they will never operate on large pmd/pud entries, and therefore only use the non-large _xxx_ENTRY_ORIGIN mask. With commit 9ec8fa8dc331b ("s390/vmemmap: extend modify_pagetable() to handle vmemmap"), that assumption is no longer true, at least for pmd_deref(). In theory, we could end up with wrong addresses because some of the non-address bits of a large entry would not be masked out. In practice, this does not (yet) show any impact, because vmemmap_free() is currently never used for s390. Fix pmd/pud_deref() to check for the entry type and use the _xxx_ENTRY_ORIGIN_LARGE mask for large entries. While at it, also move pmd/pud_pfn() around, in order to avoid code duplication, because they do the same thing. Fixes: 9ec8fa8dc331b ("s390/vmemmap: extend modify_pagetable() to handle vmemmap") Cc: # 5.9 Signed-off-by: Gerald Schaefer Reviewed-by: Alexander Gordeev Signed-off-by: Heiko Carstens --- arch/s390/include/asm/pgtable.h | 52 ++++++++++++++++++++++++----------------- 1 file changed, 30 insertions(+), 22 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index 6b8d8c69b1a1..b5dbae78969b 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -692,16 +692,6 @@ static inline int pud_large(pud_t pud) return !!(pud_val(pud) & _REGION3_ENTRY_LARGE); } -static inline unsigned long pud_pfn(pud_t pud) -{ - unsigned long origin_mask; - - origin_mask = _REGION_ENTRY_ORIGIN; - if (pud_large(pud)) - origin_mask = _REGION3_ENTRY_ORIGIN_LARGE; - return (pud_val(pud) & origin_mask) >> PAGE_SHIFT; -} - #define pmd_leaf pmd_large static inline int pmd_large(pmd_t pmd) { @@ -747,16 +737,6 @@ static inline int pmd_none(pmd_t pmd) return pmd_val(pmd) == _SEGMENT_ENTRY_EMPTY; } -static inline unsigned long pmd_pfn(pmd_t pmd) -{ - unsigned long origin_mask; - - origin_mask = _SEGMENT_ENTRY_ORIGIN; - if (pmd_large(pmd)) - origin_mask = _SEGMENT_ENTRY_ORIGIN_LARGE; - return (pmd_val(pmd) & origin_mask) >> PAGE_SHIFT; -} - #define pmd_write pmd_write static inline int pmd_write(pmd_t pmd) { @@ -1238,11 +1218,39 @@ static inline pte_t mk_pte(struct page *page, pgprot_t pgprot) #define pud_index(address) (((address) >> PUD_SHIFT) & (PTRS_PER_PUD-1)) #define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1)) -#define pmd_deref(pmd) (pmd_val(pmd) & _SEGMENT_ENTRY_ORIGIN) -#define pud_deref(pud) (pud_val(pud) & _REGION_ENTRY_ORIGIN) #define p4d_deref(pud) (p4d_val(pud) & _REGION_ENTRY_ORIGIN) #define pgd_deref(pgd) (pgd_val(pgd) & _REGION_ENTRY_ORIGIN) +static inline unsigned long pmd_deref(pmd_t pmd) +{ + unsigned long origin_mask; + + origin_mask = _SEGMENT_ENTRY_ORIGIN; + if (pmd_large(pmd)) + origin_mask = _SEGMENT_ENTRY_ORIGIN_LARGE; + return pmd_val(pmd) & origin_mask; +} + +static inline unsigned long pmd_pfn(pmd_t pmd) +{ + return pmd_deref(pmd) >> PAGE_SHIFT; +} + +static inline unsigned long pud_deref(pud_t pud) +{ + unsigned long origin_mask; + + origin_mask = _REGION_ENTRY_ORIGIN; + if (pud_large(pud)) + origin_mask = _REGION3_ENTRY_ORIGIN_LARGE; + return pud_val(pud) & origin_mask; +} + +static inline unsigned long pud_pfn(pud_t pud) +{ + return pud_deref(pud) >> PAGE_SHIFT; +} + /* * The pgd_offset function *always* adds the index for the top-level * region/segment table. This is done to get a sequence like the -- cgit v1.2.3 From e99198661ecd02545b926ba40d1e91626bb29647 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 26 Oct 2020 16:15:22 +0100 Subject: s390/vdso: remove empty unused file Signed-off-by: Heiko Carstens --- arch/s390/include/asm/vdso/vdso.h | 0 1 file changed, 0 insertions(+), 0 deletions(-) delete mode 100644 arch/s390/include/asm/vdso/vdso.h (limited to 'arch/s390') diff --git a/arch/s390/include/asm/vdso/vdso.h b/arch/s390/include/asm/vdso/vdso.h deleted file mode 100644 index e69de29bb2d1..000000000000 -- cgit v1.2.3 From cfef9aa69a7382a205661a83e621114b37824474 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 26 Oct 2020 16:15:22 +0100 Subject: s390/vdso: remove unused constants Signed-off-by: Heiko Carstens --- arch/s390/kernel/asm-offsets.c | 8 -------- 1 file changed, 8 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c index ece58f2217cb..2012c1cf0853 100644 --- a/arch/s390/kernel/asm-offsets.c +++ b/arch/s390/kernel/asm-offsets.c @@ -61,14 +61,6 @@ int main(void) BLANK(); OFFSET(__VDSO_GETCPU_VAL, vdso_per_cpu_data, getcpu_val); BLANK(); - /* constants used by the vdso */ - DEFINE(__CLOCK_REALTIME, CLOCK_REALTIME); - DEFINE(__CLOCK_MONOTONIC, CLOCK_MONOTONIC); - DEFINE(__CLOCK_REALTIME_COARSE, CLOCK_REALTIME_COARSE); - DEFINE(__CLOCK_MONOTONIC_COARSE, CLOCK_MONOTONIC_COARSE); - DEFINE(__CLOCK_THREAD_CPUTIME_ID, CLOCK_THREAD_CPUTIME_ID); - DEFINE(__CLOCK_COARSE_RES, LOW_RES_NSEC); - BLANK(); /* idle data offsets */ OFFSET(__CLOCK_IDLE_ENTER, s390_idle_data, clock_idle_enter); OFFSET(__CLOCK_IDLE_EXIT, s390_idle_data, clock_idle_exit); -- cgit v1.2.3 From c3d9cdca73d0e49f01a71cdc477a09b04b1b30fc Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 26 Oct 2020 19:52:04 +0100 Subject: s390: update defconfigs Signed-off-by: Heiko Carstens --- arch/s390/configs/debug_defconfig | 10 ++++++---- arch/s390/configs/defconfig | 9 +++++---- arch/s390/configs/zfcpdump_defconfig | 2 +- 3 files changed, 12 insertions(+), 9 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/configs/debug_defconfig b/arch/s390/configs/debug_defconfig index 0784bf3caf43..a4d3c578fbd8 100644 --- a/arch/s390/configs/debug_defconfig +++ b/arch/s390/configs/debug_defconfig @@ -93,9 +93,10 @@ CONFIG_CLEANCACHE=y CONFIG_FRONTSWAP=y CONFIG_CMA_DEBUG=y CONFIG_CMA_DEBUGFS=y +CONFIG_CMA_AREAS=7 CONFIG_MEM_SOFT_DIRTY=y CONFIG_ZSWAP=y -CONFIG_ZSMALLOC=m +CONFIG_ZSMALLOC=y CONFIG_ZSMALLOC_STAT=y CONFIG_DEFERRED_STRUCT_PAGE_INIT=y CONFIG_IDLE_PAGE_TRACKING=y @@ -378,7 +379,6 @@ CONFIG_NETLINK_DIAG=m CONFIG_CGROUP_NET_PRIO=y CONFIG_BPF_JIT=y CONFIG_NET_PKTGEN=m -# CONFIG_NET_DROP_MONITOR is not set CONFIG_PCI=y # CONFIG_PCIEASPM is not set CONFIG_PCI_DEBUG=y @@ -386,7 +386,7 @@ CONFIG_HOTPLUG_PCI=y CONFIG_HOTPLUG_PCI_S390=y CONFIG_DEVTMPFS=y CONFIG_CONNECTOR=y -CONFIG_ZRAM=m +CONFIG_ZRAM=y CONFIG_BLK_DEV_LOOP=m CONFIG_BLK_DEV_CRYPTOLOOP=m CONFIG_BLK_DEV_DRBD=m @@ -689,6 +689,7 @@ CONFIG_CRYPTO_TEST=m CONFIG_CRYPTO_DH=m CONFIG_CRYPTO_ECDH=m CONFIG_CRYPTO_ECRDSA=m +CONFIG_CRYPTO_SM2=m CONFIG_CRYPTO_CURVE25519=m CONFIG_CRYPTO_GCM=y CONFIG_CRYPTO_CHACHA20POLY1305=m @@ -709,7 +710,6 @@ CONFIG_CRYPTO_RMD160=m CONFIG_CRYPTO_RMD256=m CONFIG_CRYPTO_RMD320=m CONFIG_CRYPTO_SHA3=m -CONFIG_CRYPTO_SM3=m CONFIG_CRYPTO_TGR192=m CONFIG_CRYPTO_WP512=m CONFIG_CRYPTO_AES_TI=m @@ -753,6 +753,7 @@ CONFIG_CRYPTO_DES_S390=m CONFIG_CRYPTO_AES_S390=m CONFIG_CRYPTO_GHASH_S390=m CONFIG_CRYPTO_CRC32_S390=y +CONFIG_CRYPTO_DEV_VIRTIO=m CONFIG_CORDIC=m CONFIG_CRC32_SELFTEST=y CONFIG_CRC4=m @@ -829,6 +830,7 @@ CONFIG_NETDEV_NOTIFIER_ERROR_INJECT=m CONFIG_FAULT_INJECTION=y CONFIG_FAILSLAB=y CONFIG_FAIL_PAGE_ALLOC=y +CONFIG_FAULT_INJECTION_USERCOPY=y CONFIG_FAIL_MAKE_REQUEST=y CONFIG_FAIL_IO_TIMEOUT=y CONFIG_FAIL_FUTEX=y diff --git a/arch/s390/configs/defconfig b/arch/s390/configs/defconfig index 905bc8c4cfaf..17d5df2c1eff 100644 --- a/arch/s390/configs/defconfig +++ b/arch/s390/configs/defconfig @@ -87,9 +87,10 @@ CONFIG_KSM=y CONFIG_TRANSPARENT_HUGEPAGE=y CONFIG_CLEANCACHE=y CONFIG_FRONTSWAP=y +CONFIG_CMA_AREAS=7 CONFIG_MEM_SOFT_DIRTY=y CONFIG_ZSWAP=y -CONFIG_ZSMALLOC=m +CONFIG_ZSMALLOC=y CONFIG_ZSMALLOC_STAT=y CONFIG_DEFERRED_STRUCT_PAGE_INIT=y CONFIG_IDLE_PAGE_TRACKING=y @@ -371,7 +372,6 @@ CONFIG_NETLINK_DIAG=m CONFIG_CGROUP_NET_PRIO=y CONFIG_BPF_JIT=y CONFIG_NET_PKTGEN=m -# CONFIG_NET_DROP_MONITOR is not set CONFIG_PCI=y # CONFIG_PCIEASPM is not set CONFIG_HOTPLUG_PCI=y @@ -379,7 +379,7 @@ CONFIG_HOTPLUG_PCI_S390=y CONFIG_UEVENT_HELPER=y CONFIG_DEVTMPFS=y CONFIG_CONNECTOR=y -CONFIG_ZRAM=m +CONFIG_ZRAM=y CONFIG_BLK_DEV_LOOP=m CONFIG_BLK_DEV_CRYPTOLOOP=m CONFIG_BLK_DEV_DRBD=m @@ -680,6 +680,7 @@ CONFIG_CRYPTO_TEST=m CONFIG_CRYPTO_DH=m CONFIG_CRYPTO_ECDH=m CONFIG_CRYPTO_ECRDSA=m +CONFIG_CRYPTO_SM2=m CONFIG_CRYPTO_CURVE25519=m CONFIG_CRYPTO_GCM=y CONFIG_CRYPTO_CHACHA20POLY1305=m @@ -701,7 +702,6 @@ CONFIG_CRYPTO_RMD160=m CONFIG_CRYPTO_RMD256=m CONFIG_CRYPTO_RMD320=m CONFIG_CRYPTO_SHA3=m -CONFIG_CRYPTO_SM3=m CONFIG_CRYPTO_TGR192=m CONFIG_CRYPTO_WP512=m CONFIG_CRYPTO_AES_TI=m @@ -745,6 +745,7 @@ CONFIG_CRYPTO_DES_S390=m CONFIG_CRYPTO_AES_S390=m CONFIG_CRYPTO_GHASH_S390=m CONFIG_CRYPTO_CRC32_S390=y +CONFIG_CRYPTO_DEV_VIRTIO=m CONFIG_CORDIC=m CONFIG_PRIME_NUMBERS=m CONFIG_CRC4=m diff --git a/arch/s390/configs/zfcpdump_defconfig b/arch/s390/configs/zfcpdump_defconfig index 8f67c55625f9..a302630341ef 100644 --- a/arch/s390/configs/zfcpdump_defconfig +++ b/arch/s390/configs/zfcpdump_defconfig @@ -17,11 +17,11 @@ CONFIG_HZ_100=y # CONFIG_CHSC_SCH is not set # CONFIG_SCM_BUS is not set CONFIG_CRASH_DUMP=y -# CONFIG_SECCOMP is not set # CONFIG_PFAULT is not set # CONFIG_S390_HYPFS_FS is not set # CONFIG_VIRTUALIZATION is not set # CONFIG_S390_GUEST is not set +# CONFIG_SECCOMP is not set CONFIG_PARTITION_ADVANCED=y CONFIG_IBM_PARTITION=y # CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set -- cgit v1.2.3 From de5d9dae150ca1c1b5c7676711a9ca139d1a8dec Mon Sep 17 00:00:00 2001 From: Qian Cai Date: Wed, 28 Oct 2020 14:27:42 -0400 Subject: s390/smp: move rcu_cpu_starting() earlier The call to rcu_cpu_starting() in smp_init_secondary() is not early enough in the CPU-hotplug onlining process, which results in lockdep splats as follows: WARNING: suspicious RCU usage ----------------------------- kernel/locking/lockdep.c:3497 RCU-list traversed in non-reader section!! other info that might help us debug this: RCU used illegally from offline CPU! rcu_scheduler_active = 1, debug_locks = 1 no locks held by swapper/1/0. Call Trace: show_stack+0x158/0x1f0 dump_stack+0x1f2/0x238 __lock_acquire+0x2640/0x4dd0 lock_acquire+0x3a8/0xd08 _raw_spin_lock_irqsave+0xc0/0xf0 clockevents_register_device+0xa8/0x528 init_cpu_timer+0x33e/0x468 smp_init_secondary+0x11a/0x328 smp_start_secondary+0x82/0x88 This is avoided by moving the call to rcu_cpu_starting up near the beginning of the smp_init_secondary() function. Note that the raw_smp_processor_id() is required in order to avoid calling into lockdep before RCU has declared the CPU to be watched for readers. Link: https://lore.kernel.org/lkml/160223032121.7002.1269740091547117869.tip-bot2@tip-bot2/ Signed-off-by: Qian Cai Acked-by: Paul E. McKenney Signed-off-by: Heiko Carstens --- arch/s390/kernel/smp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index ebfe86d097f0..390d97daa2b3 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -855,13 +855,14 @@ void __init smp_detect_cpus(void) static void smp_init_secondary(void) { - int cpu = smp_processor_id(); + int cpu = raw_smp_processor_id(); S390_lowcore.last_update_clock = get_tod_clock(); restore_access_regs(S390_lowcore.access_regs_save_area); set_cpu_flag(CIF_ASCE_PRIMARY); set_cpu_flag(CIF_ASCE_SECONDARY); cpu_init(); + rcu_cpu_starting(cpu); preempt_disable(); init_cpu_timer(); vtime_init(); -- cgit v1.2.3 From 0b2ca2c7d0c9e2731d01b6c862375d44a7e13923 Mon Sep 17 00:00:00 2001 From: Niklas Schnelle Date: Mon, 2 Nov 2020 11:33:04 +0100 Subject: s390/pci: fix hot-plug of PCI function missing bus Under some circumstances in particular with "Reconfigure I/O Path" a zPCI function may first appear in Standby through a PCI event with PEC 0x0302 which initially makes it visible to the zPCI subsystem, Only after that is it configured with a zPCI event with PEC 0x0301. If the zbus is still missing a PCI function zero (devfn == 0) when the PCI event 0x0301 is handled zdev->zbus->bus is still NULL and gets dereferenced in common code. Check for this case and enable but don't scan the zPCI function. This matches what would happen if we immediately got the 0x0301 configuration request or the function was included in CLP List PCI. In all cases the PCI functions with devfn != 0 will be scanned once function 0 appears. Fixes: 3047766bc6ec ("s390/pci: fix enabling a reserved PCI function") Cc: # 5.8 Signed-off-by: Niklas Schnelle Acked-by: Pierre Morel Signed-off-by: Heiko Carstens --- arch/s390/pci/pci_event.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'arch/s390') diff --git a/arch/s390/pci/pci_event.c b/arch/s390/pci/pci_event.c index d33f21545dfd..9a6bae503fe6 100644 --- a/arch/s390/pci/pci_event.c +++ b/arch/s390/pci/pci_event.c @@ -101,6 +101,10 @@ static void __zpci_event_availability(struct zpci_ccdf_avail *ccdf) if (ret) break; + /* the PCI function will be scanned once function 0 appears */ + if (!zdev->zbus->bus) + break; + pdev = pci_scan_single_device(zdev->zbus->bus, zdev->devfn); if (!pdev) break; -- cgit v1.2.3 From ce9dfafe29bed86fe3cda330ac6072ce84e1ff81 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Tue, 3 Nov 2020 16:55:43 +0100 Subject: s390: fix system call exit path The system call exit path is running with interrupts enabled while checking for TIF/PIF/CIF bits which require special handling. If all bits have been checked interrupts are disabled and the kernel exits to user space. The problem is that after checking all bits and before interrupts are disabled bits can be set already again, due to interrupt handling. This means that the kernel can exit to user space with some TIF/PIF/CIF bits set, which should never happen. E.g. TIF_NEED_RESCHED might be set, which might lead to additional latencies, since that bit will only be recognized with next exit to user space. Fix this by checking the corresponding bits only when interrupts are disabled. Fixes: 0b0ed657fe00 ("s390: remove critical section cleanup from entry.S") Cc: # 5.8 Acked-by: Sven Schnelle Signed-off-by: Heiko Carstens --- arch/s390/kernel/entry.S | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/s390') diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 86235919c2d1..5346545b9860 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -422,6 +422,7 @@ ENTRY(system_call) #endif LOCKDEP_SYS_EXIT .Lsysc_tif: + DISABLE_INTS TSTMSK __PT_FLAGS(%r11),_PIF_WORK jnz .Lsysc_work TSTMSK __TI_flags(%r12),_TIF_WORK @@ -444,6 +445,7 @@ ENTRY(system_call) # One of the work bits is on. Find out which one. # .Lsysc_work: + ENABLE_INTS TSTMSK __TI_flags(%r12),_TIF_NEED_RESCHED jo .Lsysc_reschedule TSTMSK __PT_FLAGS(%r11),_PIF_SYSCALL_RESTART -- cgit v1.2.3 From 267fb27352b6fc9fdbad753127a239f75618ecbc Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 30 Oct 2020 15:50:32 +0100 Subject: perf: Reduce stack usage of perf_output_begin() __perf_output_begin() has an on-stack struct perf_sample_data in the unlikely case it needs to generate a LOST record. However, every call to perf_output_begin() must already have a perf_sample_data on-stack. Reported-by: Thomas Gleixner Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20201030151954.985416146@infradead.org --- arch/powerpc/perf/imc-pmu.c | 2 +- arch/s390/kernel/perf_cpum_sf.c | 2 +- arch/x86/events/intel/ds.c | 4 ++-- include/linux/perf_event.h | 7 +++++-- kernel/events/core.c | 32 +++++++++++++++++--------------- kernel/events/ring_buffer.c | 20 +++++++++++--------- 6 files changed, 37 insertions(+), 30 deletions(-) (limited to 'arch/s390') diff --git a/arch/powerpc/perf/imc-pmu.c b/arch/powerpc/perf/imc-pmu.c index 9ed4fcccf8a9..7b25548ec42b 100644 --- a/arch/powerpc/perf/imc-pmu.c +++ b/arch/powerpc/perf/imc-pmu.c @@ -1336,7 +1336,7 @@ static void dump_trace_imc_data(struct perf_event *event) /* If this is a valid record, create the sample */ struct perf_output_handle handle; - if (perf_output_begin(&handle, event, header.size)) + if (perf_output_begin(&handle, &data, event, header.size)) return; perf_output_sample(&handle, &header, &data, event); diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c index 4f9e4626df55..00255ae3979d 100644 --- a/arch/s390/kernel/perf_cpum_sf.c +++ b/arch/s390/kernel/perf_cpum_sf.c @@ -672,7 +672,7 @@ static void cpumsf_output_event_pid(struct perf_event *event, rcu_read_lock(); perf_prepare_sample(&header, data, event, regs); - if (perf_output_begin(&handle, event, header.size)) + if (perf_output_begin(&handle, data, event, header.size)) goto out; /* Update the process ID (see also kernel/events/core.c) */ diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c index 404315df1e16..cd2ae14a0a98 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -642,8 +642,8 @@ int intel_pmu_drain_bts_buffer(void) rcu_read_lock(); perf_prepare_sample(&header, &data, event, ®s); - if (perf_output_begin(&handle, event, header.size * - (top - base - skip))) + if (perf_output_begin(&handle, &data, event, + header.size * (top - base - skip))) goto unlock; for (at = base; at < top; at++) { diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 0c19d279b97f..b775ae0a8c87 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -1400,11 +1400,14 @@ perf_event_addr_filters(struct perf_event *event) extern void perf_event_addr_filters_sync(struct perf_event *event); extern int perf_output_begin(struct perf_output_handle *handle, + struct perf_sample_data *data, struct perf_event *event, unsigned int size); extern int perf_output_begin_forward(struct perf_output_handle *handle, - struct perf_event *event, - unsigned int size); + struct perf_sample_data *data, + struct perf_event *event, + unsigned int size); extern int perf_output_begin_backward(struct perf_output_handle *handle, + struct perf_sample_data *data, struct perf_event *event, unsigned int size); diff --git a/kernel/events/core.c b/kernel/events/core.c index 5a29ab09e72d..fc681c7c1e03 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -7186,6 +7186,7 @@ __perf_event_output(struct perf_event *event, struct perf_sample_data *data, struct pt_regs *regs, int (*output_begin)(struct perf_output_handle *, + struct perf_sample_data *, struct perf_event *, unsigned int)) { @@ -7198,7 +7199,7 @@ __perf_event_output(struct perf_event *event, perf_prepare_sample(&header, data, event, regs); - err = output_begin(&handle, event, header.size); + err = output_begin(&handle, data, event, header.size); if (err) goto exit; @@ -7264,7 +7265,7 @@ perf_event_read_event(struct perf_event *event, int ret; perf_event_header__init_id(&read_event.header, &sample, event); - ret = perf_output_begin(&handle, event, read_event.header.size); + ret = perf_output_begin(&handle, &sample, event, read_event.header.size); if (ret) return; @@ -7533,7 +7534,7 @@ static void perf_event_task_output(struct perf_event *event, perf_event_header__init_id(&task_event->event_id.header, &sample, event); - ret = perf_output_begin(&handle, event, + ret = perf_output_begin(&handle, &sample, event, task_event->event_id.header.size); if (ret) goto out; @@ -7636,7 +7637,7 @@ static void perf_event_comm_output(struct perf_event *event, return; perf_event_header__init_id(&comm_event->event_id.header, &sample, event); - ret = perf_output_begin(&handle, event, + ret = perf_output_begin(&handle, &sample, event, comm_event->event_id.header.size); if (ret) @@ -7736,7 +7737,7 @@ static void perf_event_namespaces_output(struct perf_event *event, perf_event_header__init_id(&namespaces_event->event_id.header, &sample, event); - ret = perf_output_begin(&handle, event, + ret = perf_output_begin(&handle, &sample, event, namespaces_event->event_id.header.size); if (ret) goto out; @@ -7863,7 +7864,7 @@ static void perf_event_cgroup_output(struct perf_event *event, void *data) perf_event_header__init_id(&cgroup_event->event_id.header, &sample, event); - ret = perf_output_begin(&handle, event, + ret = perf_output_begin(&handle, &sample, event, cgroup_event->event_id.header.size); if (ret) goto out; @@ -7989,7 +7990,7 @@ static void perf_event_mmap_output(struct perf_event *event, } perf_event_header__init_id(&mmap_event->event_id.header, &sample, event); - ret = perf_output_begin(&handle, event, + ret = perf_output_begin(&handle, &sample, event, mmap_event->event_id.header.size); if (ret) goto out; @@ -8299,7 +8300,7 @@ void perf_event_aux_event(struct perf_event *event, unsigned long head, int ret; perf_event_header__init_id(&rec.header, &sample, event); - ret = perf_output_begin(&handle, event, rec.header.size); + ret = perf_output_begin(&handle, &sample, event, rec.header.size); if (ret) return; @@ -8333,7 +8334,7 @@ void perf_log_lost_samples(struct perf_event *event, u64 lost) perf_event_header__init_id(&lost_samples_event.header, &sample, event); - ret = perf_output_begin(&handle, event, + ret = perf_output_begin(&handle, &sample, event, lost_samples_event.header.size); if (ret) return; @@ -8388,7 +8389,7 @@ static void perf_event_switch_output(struct perf_event *event, void *data) perf_event_header__init_id(&se->event_id.header, &sample, event); - ret = perf_output_begin(&handle, event, se->event_id.header.size); + ret = perf_output_begin(&handle, &sample, event, se->event_id.header.size); if (ret) return; @@ -8463,7 +8464,7 @@ static void perf_log_throttle(struct perf_event *event, int enable) perf_event_header__init_id(&throttle_event.header, &sample, event); - ret = perf_output_begin(&handle, event, + ret = perf_output_begin(&handle, &sample, event, throttle_event.header.size); if (ret) return; @@ -8506,7 +8507,7 @@ static void perf_event_ksymbol_output(struct perf_event *event, void *data) perf_event_header__init_id(&ksymbol_event->event_id.header, &sample, event); - ret = perf_output_begin(&handle, event, + ret = perf_output_begin(&handle, &sample, event, ksymbol_event->event_id.header.size); if (ret) return; @@ -8596,7 +8597,7 @@ static void perf_event_bpf_output(struct perf_event *event, void *data) perf_event_header__init_id(&bpf_event->event_id.header, &sample, event); - ret = perf_output_begin(&handle, event, + ret = perf_output_begin(&handle, data, event, bpf_event->event_id.header.size); if (ret) return; @@ -8705,7 +8706,8 @@ static void perf_event_text_poke_output(struct perf_event *event, void *data) perf_event_header__init_id(&text_poke_event->event_id.header, &sample, event); - ret = perf_output_begin(&handle, event, text_poke_event->event_id.header.size); + ret = perf_output_begin(&handle, &sample, event, + text_poke_event->event_id.header.size); if (ret) return; @@ -8786,7 +8788,7 @@ static void perf_log_itrace_start(struct perf_event *event) rec.tid = perf_event_tid(event, current); perf_event_header__init_id(&rec.header, &sample, event); - ret = perf_output_begin(&handle, event, rec.header.size); + ret = perf_output_begin(&handle, &sample, event, rec.header.size); if (ret) return; diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index 192b8abc6330..ef91ae75ca56 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -147,6 +147,7 @@ ring_buffer_has_space(unsigned long head, unsigned long tail, static __always_inline int __perf_output_begin(struct perf_output_handle *handle, + struct perf_sample_data *data, struct perf_event *event, unsigned int size, bool backward) { @@ -237,18 +238,16 @@ __perf_output_begin(struct perf_output_handle *handle, handle->size = (1UL << page_shift) - offset; if (unlikely(have_lost)) { - struct perf_sample_data sample_data; - lost_event.header.size = sizeof(lost_event); lost_event.header.type = PERF_RECORD_LOST; lost_event.header.misc = 0; lost_event.id = event->id; lost_event.lost = local_xchg(&rb->lost, 0); - perf_event_header__init_id(&lost_event.header, - &sample_data, event); + /* XXX mostly redundant; @data is already fully initializes */ + perf_event_header__init_id(&lost_event.header, data, event); perf_output_put(handle, lost_event); - perf_event__output_id_sample(event, handle, &sample_data); + perf_event__output_id_sample(event, handle, data); } return 0; @@ -263,22 +262,25 @@ out: } int perf_output_begin_forward(struct perf_output_handle *handle, - struct perf_event *event, unsigned int size) + struct perf_sample_data *data, + struct perf_event *event, unsigned int size) { - return __perf_output_begin(handle, event, size, false); + return __perf_output_begin(handle, data, event, size, false); } int perf_output_begin_backward(struct perf_output_handle *handle, + struct perf_sample_data *data, struct perf_event *event, unsigned int size) { - return __perf_output_begin(handle, event, size, true); + return __perf_output_begin(handle, data, event, size, true); } int perf_output_begin(struct perf_output_handle *handle, + struct perf_sample_data *data, struct perf_event *event, unsigned int size) { - return __perf_output_begin(handle, event, size, + return __perf_output_begin(handle, data, event, size, unlikely(is_write_backward(event))); } -- cgit v1.2.3 From 76a4efa80900fc40e0fdf243b42aec9fb8c35d24 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 30 Oct 2020 12:14:21 +0100 Subject: perf/arch: Remove perf_sample_data::regs_user_copy struct perf_sample_data lives on-stack, we should be careful about it's size. Furthermore, the pt_regs copy in there is only because x86_64 is a trainwreck, solve it differently. Reported-by: Thomas Gleixner Signed-off-by: Peter Zijlstra (Intel) Tested-by: Steven Rostedt Link: https://lkml.kernel.org/r/20201030151955.258178461@infradead.org --- arch/arm/kernel/perf_regs.c | 3 +-- arch/arm64/kernel/perf_regs.c | 3 +-- arch/csky/kernel/perf_regs.c | 3 +-- arch/powerpc/perf/perf_regs.c | 3 +-- arch/riscv/kernel/perf_regs.c | 3 +-- arch/s390/kernel/perf_regs.c | 3 +-- arch/x86/kernel/perf_regs.c | 15 +++++++++++---- include/linux/perf_event.h | 6 ------ include/linux/perf_regs.h | 6 ++---- kernel/events/core.c | 8 +++----- 10 files changed, 22 insertions(+), 31 deletions(-) (limited to 'arch/s390') diff --git a/arch/arm/kernel/perf_regs.c b/arch/arm/kernel/perf_regs.c index 05fe92aa7d98..0529f90395c9 100644 --- a/arch/arm/kernel/perf_regs.c +++ b/arch/arm/kernel/perf_regs.c @@ -32,8 +32,7 @@ u64 perf_reg_abi(struct task_struct *task) } void perf_get_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy) + struct pt_regs *regs) { regs_user->regs = task_pt_regs(current); regs_user->abi = perf_reg_abi(current); diff --git a/arch/arm64/kernel/perf_regs.c b/arch/arm64/kernel/perf_regs.c index 94e8718e7229..f6f58e6265df 100644 --- a/arch/arm64/kernel/perf_regs.c +++ b/arch/arm64/kernel/perf_regs.c @@ -73,8 +73,7 @@ u64 perf_reg_abi(struct task_struct *task) } void perf_get_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy) + struct pt_regs *regs) { regs_user->regs = task_pt_regs(current); regs_user->abi = perf_reg_abi(current); diff --git a/arch/csky/kernel/perf_regs.c b/arch/csky/kernel/perf_regs.c index eb32838b8210..09b7f88a2d6a 100644 --- a/arch/csky/kernel/perf_regs.c +++ b/arch/csky/kernel/perf_regs.c @@ -32,8 +32,7 @@ u64 perf_reg_abi(struct task_struct *task) } void perf_get_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy) + struct pt_regs *regs) { regs_user->regs = task_pt_regs(current); regs_user->abi = perf_reg_abi(current); diff --git a/arch/powerpc/perf/perf_regs.c b/arch/powerpc/perf/perf_regs.c index 8e53f2fc3fe0..6f681b105eec 100644 --- a/arch/powerpc/perf/perf_regs.c +++ b/arch/powerpc/perf/perf_regs.c @@ -144,8 +144,7 @@ u64 perf_reg_abi(struct task_struct *task) } void perf_get_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy) + struct pt_regs *regs) { regs_user->regs = task_pt_regs(current); regs_user->abi = (regs_user->regs) ? perf_reg_abi(current) : diff --git a/arch/riscv/kernel/perf_regs.c b/arch/riscv/kernel/perf_regs.c index 04a38fbeb9c7..fd304a248de6 100644 --- a/arch/riscv/kernel/perf_regs.c +++ b/arch/riscv/kernel/perf_regs.c @@ -36,8 +36,7 @@ u64 perf_reg_abi(struct task_struct *task) } void perf_get_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy) + struct pt_regs *regs) { regs_user->regs = task_pt_regs(current); regs_user->abi = perf_reg_abi(current); diff --git a/arch/s390/kernel/perf_regs.c b/arch/s390/kernel/perf_regs.c index 4352a504f235..6e9e5d5e927e 100644 --- a/arch/s390/kernel/perf_regs.c +++ b/arch/s390/kernel/perf_regs.c @@ -53,8 +53,7 @@ u64 perf_reg_abi(struct task_struct *task) } void perf_get_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy) + struct pt_regs *regs) { /* * Use the regs from the first interruption and let diff --git a/arch/x86/kernel/perf_regs.c b/arch/x86/kernel/perf_regs.c index bb7e1132290b..f9e5352b3bef 100644 --- a/arch/x86/kernel/perf_regs.c +++ b/arch/x86/kernel/perf_regs.c @@ -101,8 +101,7 @@ u64 perf_reg_abi(struct task_struct *task) } void perf_get_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy) + struct pt_regs *regs) { regs_user->regs = task_pt_regs(current); regs_user->abi = perf_reg_abi(current); @@ -129,12 +128,20 @@ u64 perf_reg_abi(struct task_struct *task) return PERF_SAMPLE_REGS_ABI_64; } +static DEFINE_PER_CPU(struct pt_regs, nmi_user_regs); + void perf_get_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy) + struct pt_regs *regs) { + struct pt_regs *regs_user_copy = this_cpu_ptr(&nmi_user_regs); struct pt_regs *user_regs = task_pt_regs(current); + if (!in_nmi()) { + regs_user->regs = user_regs; + regs_user->abi = perf_reg_abi(current); + return; + } + /* * If we're in an NMI that interrupted task_pt_regs setup, then * we can't sample user regs at all. This check isn't really diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index b775ae0a8c87..96450f6fb1de 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -1022,13 +1022,7 @@ struct perf_sample_data { struct perf_callchain_entry *callchain; u64 aux_size; - /* - * regs_user may point to task_pt_regs or to regs_user_copy, depending - * on arch details. - */ struct perf_regs regs_user; - struct pt_regs regs_user_copy; - struct perf_regs regs_intr; u64 stack_user_size; diff --git a/include/linux/perf_regs.h b/include/linux/perf_regs.h index 2d12e97d5e7b..f632c5725f16 100644 --- a/include/linux/perf_regs.h +++ b/include/linux/perf_regs.h @@ -20,8 +20,7 @@ u64 perf_reg_value(struct pt_regs *regs, int idx); int perf_reg_validate(u64 mask); u64 perf_reg_abi(struct task_struct *task); void perf_get_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy); + struct pt_regs *regs); #else #define PERF_REG_EXTENDED_MASK 0 @@ -42,8 +41,7 @@ static inline u64 perf_reg_abi(struct task_struct *task) } static inline void perf_get_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy) + struct pt_regs *regs) { regs_user->regs = task_pt_regs(current); regs_user->abi = perf_reg_abi(current); diff --git a/kernel/events/core.c b/kernel/events/core.c index fc681c7c1e03..d67c9cbb0f6a 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -6374,14 +6374,13 @@ perf_output_sample_regs(struct perf_output_handle *handle, } static void perf_sample_regs_user(struct perf_regs *regs_user, - struct pt_regs *regs, - struct pt_regs *regs_user_copy) + struct pt_regs *regs) { if (user_mode(regs)) { regs_user->abi = perf_reg_abi(current); regs_user->regs = regs; } else if (!(current->flags & PF_KTHREAD)) { - perf_get_regs_user(regs_user, regs, regs_user_copy); + perf_get_regs_user(regs_user, regs); } else { regs_user->abi = PERF_SAMPLE_REGS_ABI_NONE; regs_user->regs = NULL; @@ -7083,8 +7082,7 @@ void perf_prepare_sample(struct perf_event_header *header, } if (sample_type & (PERF_SAMPLE_REGS_USER | PERF_SAMPLE_STACK_USER)) - perf_sample_regs_user(&data->regs_user, regs, - &data->regs_user_copy); + perf_sample_regs_user(&data->regs_user, regs); if (sample_type & PERF_SAMPLE_REGS_USER) { /* regs dump ABI info */ -- cgit v1.2.3 From 1ed576a20cd5c93295f57d6b7400357bd8d01b21 Mon Sep 17 00:00:00 2001 From: Janosch Frank Date: Tue, 20 Oct 2020 06:12:07 -0400 Subject: KVM: s390: pv: Mark mm as protected after the set secure parameters and improve cleanup We can only have protected guest pages after a successful set secure parameters call as only then the UV allows imports and unpacks. By moving the test we can now also check for it in s390_reset_acc() and do an early return if it is 0. Signed-off-by: Janosch Frank Fixes: 29b40f105ec8 ("KVM: s390: protvirt: Add initial vm and cpu lifecycle handling") Reviewed-by: Cornelia Huck Signed-off-by: Christian Borntraeger --- arch/s390/kvm/kvm-s390.c | 2 +- arch/s390/kvm/pv.c | 3 ++- arch/s390/mm/gmap.c | 2 ++ 3 files changed, 5 insertions(+), 2 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 6b74b92c1a58..08ea6c4735cd 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -2312,7 +2312,7 @@ static int kvm_s390_handle_pv(struct kvm *kvm, struct kvm_pv_cmd *cmd) struct kvm_s390_pv_unp unp = {}; r = -EINVAL; - if (!kvm_s390_pv_is_protected(kvm)) + if (!kvm_s390_pv_is_protected(kvm) || !mm_is_protected(kvm->mm)) break; r = -EFAULT; diff --git a/arch/s390/kvm/pv.c b/arch/s390/kvm/pv.c index eb99e2f95ebe..f5847f9dec7c 100644 --- a/arch/s390/kvm/pv.c +++ b/arch/s390/kvm/pv.c @@ -208,7 +208,6 @@ int kvm_s390_pv_init_vm(struct kvm *kvm, u16 *rc, u16 *rrc) return -EIO; } kvm->arch.gmap->guest_handle = uvcb.guest_handle; - atomic_set(&kvm->mm->context.is_protected, 1); return 0; } @@ -228,6 +227,8 @@ int kvm_s390_pv_set_sec_parms(struct kvm *kvm, void *hdr, u64 length, u16 *rc, *rrc = uvcb.header.rrc; KVM_UV_EVENT(kvm, 3, "PROTVIRT VM SET PARMS: rc %x rrc %x", *rc, *rrc); + if (!cc) + atomic_set(&kvm->mm->context.is_protected, 1); return cc ? -EINVAL : 0; } diff --git a/arch/s390/mm/gmap.c b/arch/s390/mm/gmap.c index cfb0017f33a7..64795d034926 100644 --- a/arch/s390/mm/gmap.c +++ b/arch/s390/mm/gmap.c @@ -2690,6 +2690,8 @@ static const struct mm_walk_ops reset_acc_walk_ops = { #include void s390_reset_acc(struct mm_struct *mm) { + if (!mm_is_protected(mm)) + return; /* * we might be called during * reset: we walk the pages and clear -- cgit v1.2.3 From 6cbf1e960fa52e4c63a6dfa4cda8736375b34ccc Mon Sep 17 00:00:00 2001 From: Collin Walling Date: Wed, 4 Nov 2020 13:10:32 -0500 Subject: KVM: s390: remove diag318 reset code The diag318 data must be set to 0 by VM-wide reset events triggered by diag308. As such, KVM should not handle resetting this data via the VCPU ioctls. Fixes: 23a60f834406 ("s390/kvm: diagnose 0x318 sync and reset") Signed-off-by: Collin Walling Reviewed-by: Christian Borntraeger Reviewed-by: Janosch Frank Acked-by: Cornelia Huck Signed-off-by: Christian Borntraeger Link: https://lore.kernel.org/r/20201104181032.109800-1-walling@linux.ibm.com --- arch/s390/kvm/kvm-s390.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 08ea6c4735cd..425d3d75320b 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -3564,7 +3564,6 @@ static void kvm_arch_vcpu_ioctl_initial_reset(struct kvm_vcpu *vcpu) vcpu->arch.sie_block->pp = 0; vcpu->arch.sie_block->fpf &= ~FPF_BPBC; vcpu->arch.sie_block->todpr = 0; - vcpu->arch.sie_block->cpnc = 0; } } @@ -3582,7 +3581,6 @@ static void kvm_arch_vcpu_ioctl_clear_reset(struct kvm_vcpu *vcpu) regs->etoken = 0; regs->etoken_extension = 0; - regs->diag318 = 0; } int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) -- cgit v1.2.3 From 966e7ea434484a006700c144bca629a14f93530c Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Tue, 10 Nov 2020 14:46:55 +0100 Subject: s390: update defconfigs Signed-off-by: Heiko Carstens --- arch/s390/configs/debug_defconfig | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/s390') diff --git a/arch/s390/configs/debug_defconfig b/arch/s390/configs/debug_defconfig index a4d3c578fbd8..fe6f529ac82c 100644 --- a/arch/s390/configs/debug_defconfig +++ b/arch/s390/configs/debug_defconfig @@ -1,3 +1,4 @@ +CONFIG_UAPI_HEADER_TEST=y CONFIG_SYSVIPC=y CONFIG_POSIX_MQUEUE=y CONFIG_WATCH_QUEUE=y -- cgit v1.2.3 From 78d732e1f326f74f240d416af9484928303d9951 Mon Sep 17 00:00:00 2001 From: Thomas Richter Date: Wed, 11 Nov 2020 16:26:25 +0100 Subject: s390/cpum_sf.c: fix file permission for cpum_sfb_size This file is installed by the s390 CPU Measurement sampling facility device driver to export supported minimum and maximum sample buffer sizes. This file is read by lscpumf tool to display the details of the device driver capabilities. The lscpumf tool might be invoked by a non-root user. In this case it does not print anything because the file contents can not be read. Fix this by allowing read access for all users. Reading the file contents is ok, changing the file contents is left to the root user only. For further reference and details see: [1] https://github.com/ibm-s390-tools/s390-tools/issues/97 Fixes: 69f239ed335a ("s390/cpum_sf: Dynamically extend the sampling buffer if overflows occur") Cc: # 3.14 Signed-off-by: Thomas Richter Acked-by: Sumanth Korikkar Signed-off-by: Heiko Carstens --- arch/s390/kernel/perf_cpum_sf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c index 4f9e4626df55..f100c9209743 100644 --- a/arch/s390/kernel/perf_cpum_sf.c +++ b/arch/s390/kernel/perf_cpum_sf.c @@ -2228,4 +2228,4 @@ out: } arch_initcall(init_cpum_sampling_pmu); -core_param(cpum_sfb_size, CPUM_SF_MAX_SDB, sfb_size, 0640); +core_param(cpum_sfb_size, CPUM_SF_MAX_SDB, sfb_size, 0644); -- cgit v1.2.3 From 4c80d05714d347405865802b7098f1c97362cbef Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Tue, 17 Nov 2020 20:00:33 +0100 Subject: s390/uv: handle destroy page legacy interface Older firmware can return rc=0x107 rrc=0xd for destroy page if the page is already non-secure. This should be handled like a success as already done by newer firmware. Signed-off-by: Christian Borntraeger Fixes: 1a80b54d1ce1 ("s390/uv: add destroy page call") Reviewed-by: David Hildenbrand Acked-by: Cornelia Huck Reviewed-by: Janosch Frank --- arch/s390/kernel/uv.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/uv.c b/arch/s390/kernel/uv.c index 14bd9d58edc9..883bfed9f5c2 100644 --- a/arch/s390/kernel/uv.c +++ b/arch/s390/kernel/uv.c @@ -129,8 +129,15 @@ int uv_destroy_page(unsigned long paddr) .paddr = paddr }; - if (uv_call(0, (u64)&uvcb)) + if (uv_call(0, (u64)&uvcb)) { + /* + * Older firmware uses 107/d as an indication of a non secure + * page. Let us emulate the newer variant (no-op). + */ + if (uvcb.header.rc == 0x107 && uvcb.header.rrc == 0xd) + return 0; return -EINVAL; + } return 0; } -- cgit v1.2.3 From 1179f170b6f0af7bb0b3b7628136eaac450ddf31 Mon Sep 17 00:00:00 2001 From: Sven Schnelle Date: Fri, 20 Nov 2020 14:17:52 +0100 Subject: s390: fix fpu restore in entry.S We need to disable interrupts in load_fpu_regs(). Otherwise an interrupt might come in after the registers are loaded, but before CIF_FPU is cleared in load_fpu_regs(). When the interrupt returns, CIF_FPU will be cleared and the registers will never be restored. The entry.S code usually saves the interrupt state in __SF_EMPTY on the stack when disabling/restoring interrupts. sie64a however saves the pointer to the sie control block in __SF_SIE_CONTROL, which references the same location. This is non-obvious to the reader. To avoid thrashing the sie control block pointer in load_fpu_regs(), move the __SIE_* offsets eight bytes after __SF_EMPTY on the stack. Cc: # 5.8 Fixes: 0b0ed657fe00 ("s390: remove critical section cleanup from entry.S") Reported-by: Pierre Morel Signed-off-by: Sven Schnelle Acked-by: Christian Borntraeger Reviewed-by: Heiko Carstens Signed-off-by: Heiko Carstens --- arch/s390/kernel/asm-offsets.c | 10 +++++----- arch/s390/kernel/entry.S | 2 ++ 2 files changed, 7 insertions(+), 5 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c index 2012c1cf0853..483051e10db3 100644 --- a/arch/s390/kernel/asm-offsets.c +++ b/arch/s390/kernel/asm-offsets.c @@ -53,11 +53,11 @@ int main(void) /* stack_frame offsets */ OFFSET(__SF_BACKCHAIN, stack_frame, back_chain); OFFSET(__SF_GPRS, stack_frame, gprs); - OFFSET(__SF_EMPTY, stack_frame, empty1); - OFFSET(__SF_SIE_CONTROL, stack_frame, empty1[0]); - OFFSET(__SF_SIE_SAVEAREA, stack_frame, empty1[1]); - OFFSET(__SF_SIE_REASON, stack_frame, empty1[2]); - OFFSET(__SF_SIE_FLAGS, stack_frame, empty1[3]); + OFFSET(__SF_EMPTY, stack_frame, empty1[0]); + OFFSET(__SF_SIE_CONTROL, stack_frame, empty1[1]); + OFFSET(__SF_SIE_SAVEAREA, stack_frame, empty1[2]); + OFFSET(__SF_SIE_REASON, stack_frame, empty1[3]); + OFFSET(__SF_SIE_FLAGS, stack_frame, empty1[4]); BLANK(); OFFSET(__VDSO_GETCPU_VAL, vdso_per_cpu_data, getcpu_val); BLANK(); diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 5346545b9860..26bb0603c5a1 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -1068,6 +1068,7 @@ EXPORT_SYMBOL(save_fpu_regs) * %r4 */ load_fpu_regs: + stnsm __SF_EMPTY(%r15),0xfc lg %r4,__LC_CURRENT aghi %r4,__TASK_thread TSTMSK __LC_CPU_FLAGS,_CIF_FPU @@ -1099,6 +1100,7 @@ load_fpu_regs: .Lload_fpu_regs_done: ni __LC_CPU_FLAGS+7,255-_CIF_FPU .Lload_fpu_regs_exit: + ssm __SF_EMPTY(%r15) BR_EX %r14 .Lload_fpu_regs_end: ENDPROC(load_fpu_regs) -- cgit v1.2.3 From 58c644ba512cfbc2e39b758dd979edd1d6d00e27 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 20 Nov 2020 11:50:35 +0100 Subject: sched/idle: Fix arch_cpu_idle() vs tracing We call arch_cpu_idle() with RCU disabled, but then use local_irq_{en,dis}able(), which invokes tracing, which relies on RCU. Switch all arch_cpu_idle() implementations to use raw_local_irq_{en,dis}able() and carefully manage the lockdep,rcu,tracing state like we do in entry. (XXX: we really should change arch_cpu_idle() to not return with interrupts enabled) Reported-by: Sven Schnelle Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Mark Rutland Tested-by: Mark Rutland Link: https://lkml.kernel.org/r/20201120114925.594122626@infradead.org --- arch/alpha/kernel/process.c | 2 +- arch/arm/kernel/process.c | 2 +- arch/arm64/kernel/process.c | 2 +- arch/csky/kernel/process.c | 2 +- arch/h8300/kernel/process.c | 2 +- arch/hexagon/kernel/process.c | 2 +- arch/ia64/kernel/process.c | 2 +- arch/microblaze/kernel/process.c | 2 +- arch/mips/kernel/idle.c | 12 ++++++------ arch/nios2/kernel/process.c | 2 +- arch/openrisc/kernel/process.c | 2 +- arch/parisc/kernel/process.c | 2 +- arch/powerpc/kernel/idle.c | 4 ++-- arch/riscv/kernel/process.c | 2 +- arch/s390/kernel/idle.c | 6 +++--- arch/sh/kernel/idle.c | 2 +- arch/sparc/kernel/leon_pmc.c | 4 ++-- arch/sparc/kernel/process_32.c | 2 +- arch/sparc/kernel/process_64.c | 4 ++-- arch/um/kernel/process.c | 2 +- arch/x86/include/asm/mwait.h | 2 -- arch/x86/kernel/process.c | 12 +++++++----- kernel/sched/idle.c | 28 +++++++++++++++++++++++++++- 23 files changed, 64 insertions(+), 38 deletions(-) (limited to 'arch/s390') diff --git a/arch/alpha/kernel/process.c b/arch/alpha/kernel/process.c index 7462a7911002..4c7b0414a3ff 100644 --- a/arch/alpha/kernel/process.c +++ b/arch/alpha/kernel/process.c @@ -57,7 +57,7 @@ EXPORT_SYMBOL(pm_power_off); void arch_cpu_idle(void) { wtint(0); - local_irq_enable(); + raw_local_irq_enable(); } void arch_cpu_idle_dead(void) diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 8e6ace03e960..9f199b1e8383 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -71,7 +71,7 @@ void arch_cpu_idle(void) arm_pm_idle(); else cpu_do_idle(); - local_irq_enable(); + raw_local_irq_enable(); } void arch_cpu_idle_prepare(void) diff --git a/arch/arm64/kernel/process.c b/arch/arm64/kernel/process.c index 4784011cecac..9ebe02574127 100644 --- a/arch/arm64/kernel/process.c +++ b/arch/arm64/kernel/process.c @@ -126,7 +126,7 @@ void arch_cpu_idle(void) * tricks */ cpu_do_idle(); - local_irq_enable(); + raw_local_irq_enable(); } #ifdef CONFIG_HOTPLUG_CPU diff --git a/arch/csky/kernel/process.c b/arch/csky/kernel/process.c index f730869e21ee..69af6bc87e64 100644 --- a/arch/csky/kernel/process.c +++ b/arch/csky/kernel/process.c @@ -102,6 +102,6 @@ void arch_cpu_idle(void) #ifdef CONFIG_CPU_PM_STOP asm volatile("stop\n"); #endif - local_irq_enable(); + raw_local_irq_enable(); } #endif diff --git a/arch/h8300/kernel/process.c b/arch/h8300/kernel/process.c index aea0a40b77a9..bc1364db58fe 100644 --- a/arch/h8300/kernel/process.c +++ b/arch/h8300/kernel/process.c @@ -57,7 +57,7 @@ asmlinkage void ret_from_kernel_thread(void); */ void arch_cpu_idle(void) { - local_irq_enable(); + raw_local_irq_enable(); __asm__("sleep"); } diff --git a/arch/hexagon/kernel/process.c b/arch/hexagon/kernel/process.c index 5a0a95d93ddb..67767c5ed98c 100644 --- a/arch/hexagon/kernel/process.c +++ b/arch/hexagon/kernel/process.c @@ -44,7 +44,7 @@ void arch_cpu_idle(void) { __vmwait(); /* interrupts wake us up, but irqs are still disabled */ - local_irq_enable(); + raw_local_irq_enable(); } /* diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c index 6b61a703bcf5..c9ff8796b509 100644 --- a/arch/ia64/kernel/process.c +++ b/arch/ia64/kernel/process.c @@ -239,7 +239,7 @@ void arch_cpu_idle(void) if (mark_idle) (*mark_idle)(1); - safe_halt(); + raw_safe_halt(); if (mark_idle) (*mark_idle)(0); diff --git a/arch/microblaze/kernel/process.c b/arch/microblaze/kernel/process.c index a9e46e525cd0..f99860771ff4 100644 --- a/arch/microblaze/kernel/process.c +++ b/arch/microblaze/kernel/process.c @@ -149,5 +149,5 @@ int dump_fpu(struct pt_regs *regs, elf_fpregset_t *fpregs) void arch_cpu_idle(void) { - local_irq_enable(); + raw_local_irq_enable(); } diff --git a/arch/mips/kernel/idle.c b/arch/mips/kernel/idle.c index 5bc3b04693c7..18e69ebf5691 100644 --- a/arch/mips/kernel/idle.c +++ b/arch/mips/kernel/idle.c @@ -33,19 +33,19 @@ static void __cpuidle r3081_wait(void) { unsigned long cfg = read_c0_conf(); write_c0_conf(cfg | R30XX_CONF_HALT); - local_irq_enable(); + raw_local_irq_enable(); } static void __cpuidle r39xx_wait(void) { if (!need_resched()) write_c0_conf(read_c0_conf() | TX39_CONF_HALT); - local_irq_enable(); + raw_local_irq_enable(); } void __cpuidle r4k_wait(void) { - local_irq_enable(); + raw_local_irq_enable(); __r4k_wait(); } @@ -64,7 +64,7 @@ void __cpuidle r4k_wait_irqoff(void) " .set arch=r4000 \n" " wait \n" " .set pop \n"); - local_irq_enable(); + raw_local_irq_enable(); } /* @@ -84,7 +84,7 @@ static void __cpuidle rm7k_wait_irqoff(void) " wait \n" " mtc0 $1, $12 # stalls until W stage \n" " .set pop \n"); - local_irq_enable(); + raw_local_irq_enable(); } /* @@ -257,7 +257,7 @@ void arch_cpu_idle(void) if (cpu_wait) cpu_wait(); else - local_irq_enable(); + raw_local_irq_enable(); } #ifdef CONFIG_CPU_IDLE diff --git a/arch/nios2/kernel/process.c b/arch/nios2/kernel/process.c index 4ffe857e6ada..50b4eb19a6cc 100644 --- a/arch/nios2/kernel/process.c +++ b/arch/nios2/kernel/process.c @@ -33,7 +33,7 @@ EXPORT_SYMBOL(pm_power_off); void arch_cpu_idle(void) { - local_irq_enable(); + raw_local_irq_enable(); } /* diff --git a/arch/openrisc/kernel/process.c b/arch/openrisc/kernel/process.c index 0ff391f00334..3c98728cce24 100644 --- a/arch/openrisc/kernel/process.c +++ b/arch/openrisc/kernel/process.c @@ -79,7 +79,7 @@ void machine_power_off(void) */ void arch_cpu_idle(void) { - local_irq_enable(); + raw_local_irq_enable(); if (mfspr(SPR_UPR) & SPR_UPR_PMP) mtspr(SPR_PMR, mfspr(SPR_PMR) | SPR_PMR_DME); } diff --git a/arch/parisc/kernel/process.c b/arch/parisc/kernel/process.c index f196d96e2f9f..a92a23d6acd9 100644 --- a/arch/parisc/kernel/process.c +++ b/arch/parisc/kernel/process.c @@ -169,7 +169,7 @@ void __cpuidle arch_cpu_idle_dead(void) void __cpuidle arch_cpu_idle(void) { - local_irq_enable(); + raw_local_irq_enable(); /* nop on real hardware, qemu will idle sleep. */ asm volatile("or %%r10,%%r10,%%r10\n":::); diff --git a/arch/powerpc/kernel/idle.c b/arch/powerpc/kernel/idle.c index ae0e2632393d..1f835539fda4 100644 --- a/arch/powerpc/kernel/idle.c +++ b/arch/powerpc/kernel/idle.c @@ -52,9 +52,9 @@ void arch_cpu_idle(void) * interrupts enabled, some don't. */ if (irqs_disabled()) - local_irq_enable(); + raw_local_irq_enable(); } else { - local_irq_enable(); + raw_local_irq_enable(); /* * Go into low thread priority and possibly * low power mode. diff --git a/arch/riscv/kernel/process.c b/arch/riscv/kernel/process.c index 19225ec65db6..dd5f985b1f40 100644 --- a/arch/riscv/kernel/process.c +++ b/arch/riscv/kernel/process.c @@ -36,7 +36,7 @@ extern asmlinkage void ret_from_kernel_thread(void); void arch_cpu_idle(void) { wait_for_interrupt(); - local_irq_enable(); + raw_local_irq_enable(); } void show_regs(struct pt_regs *regs) diff --git a/arch/s390/kernel/idle.c b/arch/s390/kernel/idle.c index f7f1e64e0d98..2b85096964f8 100644 --- a/arch/s390/kernel/idle.c +++ b/arch/s390/kernel/idle.c @@ -33,10 +33,10 @@ void enabled_wait(void) PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK; clear_cpu_flag(CIF_NOHZ_DELAY); - local_irq_save(flags); + raw_local_irq_save(flags); /* Call the assembler magic in entry.S */ psw_idle(idle, psw_mask); - local_irq_restore(flags); + raw_local_irq_restore(flags); /* Account time spent with enabled wait psw loaded as idle time. */ raw_write_seqcount_begin(&idle->seqcount); @@ -123,7 +123,7 @@ void arch_cpu_idle_enter(void) void arch_cpu_idle(void) { enabled_wait(); - local_irq_enable(); + raw_local_irq_enable(); } void arch_cpu_idle_exit(void) diff --git a/arch/sh/kernel/idle.c b/arch/sh/kernel/idle.c index 0dc0f52f9bb8..f59814983bd5 100644 --- a/arch/sh/kernel/idle.c +++ b/arch/sh/kernel/idle.c @@ -22,7 +22,7 @@ static void (*sh_idle)(void); void default_idle(void) { set_bl_bit(); - local_irq_enable(); + raw_local_irq_enable(); /* Isn't this racy ? */ cpu_sleep(); clear_bl_bit(); diff --git a/arch/sparc/kernel/leon_pmc.c b/arch/sparc/kernel/leon_pmc.c index 065e2d4b7290..396f46bca52e 100644 --- a/arch/sparc/kernel/leon_pmc.c +++ b/arch/sparc/kernel/leon_pmc.c @@ -50,7 +50,7 @@ static void pmc_leon_idle_fixup(void) register unsigned int address = (unsigned int)leon3_irqctrl_regs; /* Interrupts need to be enabled to not hang the CPU */ - local_irq_enable(); + raw_local_irq_enable(); __asm__ __volatile__ ( "wr %%g0, %%asr19\n" @@ -66,7 +66,7 @@ static void pmc_leon_idle_fixup(void) static void pmc_leon_idle(void) { /* Interrupts need to be enabled to not hang the CPU */ - local_irq_enable(); + raw_local_irq_enable(); /* For systems without power-down, this will be no-op */ __asm__ __volatile__ ("wr %g0, %asr19\n\t"); diff --git a/arch/sparc/kernel/process_32.c b/arch/sparc/kernel/process_32.c index adfcaeab3ddc..a02363735915 100644 --- a/arch/sparc/kernel/process_32.c +++ b/arch/sparc/kernel/process_32.c @@ -74,7 +74,7 @@ void arch_cpu_idle(void) { if (sparc_idle) (*sparc_idle)(); - local_irq_enable(); + raw_local_irq_enable(); } /* XXX cli/sti -> local_irq_xxx here, check this works once SMP is fixed. */ diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c index a75093b993f9..6f8c7822fc06 100644 --- a/arch/sparc/kernel/process_64.c +++ b/arch/sparc/kernel/process_64.c @@ -62,11 +62,11 @@ void arch_cpu_idle(void) { if (tlb_type != hypervisor) { touch_nmi_watchdog(); - local_irq_enable(); + raw_local_irq_enable(); } else { unsigned long pstate; - local_irq_enable(); + raw_local_irq_enable(); /* The sun4v sleeping code requires that we have PSTATE.IE cleared over * the cpu sleep hypervisor call. diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c index 3bed09538dd9..9505a7e87396 100644 --- a/arch/um/kernel/process.c +++ b/arch/um/kernel/process.c @@ -217,7 +217,7 @@ void arch_cpu_idle(void) { cpu_tasks[current_thread_info()->cpu].pid = os_getpid(); um_idle_sleep(); - local_irq_enable(); + raw_local_irq_enable(); } int __cant_sleep(void) { diff --git a/arch/x86/include/asm/mwait.h b/arch/x86/include/asm/mwait.h index e039a933aca3..29dd27b5a339 100644 --- a/arch/x86/include/asm/mwait.h +++ b/arch/x86/include/asm/mwait.h @@ -88,8 +88,6 @@ static inline void __mwaitx(unsigned long eax, unsigned long ebx, static inline void __sti_mwait(unsigned long eax, unsigned long ecx) { - trace_hardirqs_on(); - mds_idle_clear_cpu_buffers(); /* "mwait %eax, %ecx;" */ asm volatile("sti; .byte 0x0f, 0x01, 0xc9;" diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index ba4593a913fa..145a7ac0c19a 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -685,7 +685,7 @@ void arch_cpu_idle(void) */ void __cpuidle default_idle(void) { - safe_halt(); + raw_safe_halt(); } #if defined(CONFIG_APM_MODULE) || defined(CONFIG_HALTPOLL_CPUIDLE_MODULE) EXPORT_SYMBOL(default_idle); @@ -736,6 +736,8 @@ void stop_this_cpu(void *dummy) /* * AMD Erratum 400 aware idle routine. We handle it the same way as C3 power * states (local apic timer and TSC stop). + * + * XXX this function is completely buggered vs RCU and tracing. */ static void amd_e400_idle(void) { @@ -757,9 +759,9 @@ static void amd_e400_idle(void) * The switch back from broadcast mode needs to be called with * interrupts disabled. */ - local_irq_disable(); + raw_local_irq_disable(); tick_broadcast_exit(); - local_irq_enable(); + raw_local_irq_enable(); } /* @@ -801,9 +803,9 @@ static __cpuidle void mwait_idle(void) if (!need_resched()) __sti_mwait(0, 0); else - local_irq_enable(); + raw_local_irq_enable(); } else { - local_irq_enable(); + raw_local_irq_enable(); } __current_clr_polling(); } diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c index 24d0ee26377d..c6932b8f4467 100644 --- a/kernel/sched/idle.c +++ b/kernel/sched/idle.c @@ -78,7 +78,7 @@ void __weak arch_cpu_idle_dead(void) { } void __weak arch_cpu_idle(void) { cpu_idle_force_poll = 1; - local_irq_enable(); + raw_local_irq_enable(); } /** @@ -94,9 +94,35 @@ void __cpuidle default_idle_call(void) trace_cpu_idle(1, smp_processor_id()); stop_critical_timings(); + + /* + * arch_cpu_idle() is supposed to enable IRQs, however + * we can't do that because of RCU and tracing. + * + * Trace IRQs enable here, then switch off RCU, and have + * arch_cpu_idle() use raw_local_irq_enable(). Note that + * rcu_idle_enter() relies on lockdep IRQ state, so switch that + * last -- this is very similar to the entry code. + */ + trace_hardirqs_on_prepare(); + lockdep_hardirqs_on_prepare(_THIS_IP_); rcu_idle_enter(); + lockdep_hardirqs_on(_THIS_IP_); + arch_cpu_idle(); + + /* + * OK, so IRQs are enabled here, but RCU needs them disabled to + * turn itself back on.. funny thing is that disabling IRQs + * will cause tracing, which needs RCU. Jump through hoops to + * make it 'work'. + */ + raw_local_irq_disable(); + lockdep_hardirqs_off(_THIS_IP_); rcu_idle_exit(); + lockdep_hardirqs_on(_THIS_IP_); + raw_local_irq_enable(); + start_critical_timings(); trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id()); } -- cgit v1.2.3