diff options
42 files changed, 509 insertions, 542 deletions
diff --git a/Documentation/ABI/testing/sysfs-bus-rbd b/Documentation/ABI/testing/sysfs-bus-rbd index fa72ccb2282e..dbedafb095e2 100644 --- a/Documentation/ABI/testing/sysfs-bus-rbd +++ b/Documentation/ABI/testing/sysfs-bus-rbd @@ -57,13 +57,6 @@ create_snap $ echo <snap-name> > /sys/bus/rbd/devices/<dev-id>/snap_create -rollback_snap - - Rolls back data to the specified snapshot. This goes over the entire - list of rados blocks and sends a rollback command to each. - - $ echo <snap-name> > /sys/bus/rbd/devices/<dev-id>/snap_rollback - snap_* A directory per each snapshot diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index e084b7e981e8..776d76b8cb69 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -220,8 +220,9 @@ config NEED_MACH_MEMORY_H be avoided when possible. config PHYS_OFFSET - hex "Physical address of main memory" + hex "Physical address of main memory" if MMU depends on !ARM_PATCH_PHYS_VIRT && !NEED_MACH_MEMORY_H + default DRAM_BASE if !MMU help Please provide the physical address corresponding to the location of main memory in your system. diff --git a/arch/arm/include/asm/unwind.h b/arch/arm/include/asm/unwind.h index a5edf421005c..d1c3f3a71c94 100644 --- a/arch/arm/include/asm/unwind.h +++ b/arch/arm/include/asm/unwind.h @@ -30,14 +30,15 @@ enum unwind_reason_code { }; struct unwind_idx { - unsigned long addr; + unsigned long addr_offset; unsigned long insn; }; struct unwind_table { struct list_head list; - struct unwind_idx *start; - struct unwind_idx *stop; + const struct unwind_idx *start; + const struct unwind_idx *origin; + const struct unwind_idx *stop; unsigned long begin_addr; unsigned long end_addr; }; @@ -49,15 +50,6 @@ extern struct unwind_table *unwind_table_add(unsigned long start, extern void unwind_table_del(struct unwind_table *tab); extern void unwind_backtrace(struct pt_regs *regs, struct task_struct *tsk); -#ifdef CONFIG_ARM_UNWIND -extern int __init unwind_init(void); -#else -static inline int __init unwind_init(void) -{ - return 0; -} -#endif - #endif /* !__ASSEMBLY__ */ #ifdef CONFIG_ARM_UNWIND diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 8e9c98edc068..88b0941ce51e 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -640,6 +640,9 @@ static struct platform_device_id armpmu_plat_device_ids[] = { static int __devinit armpmu_device_probe(struct platform_device *pdev) { + if (!cpu_pmu) + return -ENODEV; + cpu_pmu->plat_device = pdev; return 0; } diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 3448a3f9cc8c..8fc2c8fcbdc6 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -895,8 +895,6 @@ void __init setup_arch(char **cmdline_p) { struct machine_desc *mdesc; - unwind_init(); - setup_processor(); mdesc = setup_machine_fdt(__atags_pointer); if (!mdesc) @@ -904,6 +902,12 @@ void __init setup_arch(char **cmdline_p) machine_desc = mdesc; machine_name = mdesc->name; +#ifdef CONFIG_ZONE_DMA + if (mdesc->dma_zone_size) { + extern unsigned long arm_dma_zone_size; + arm_dma_zone_size = mdesc->dma_zone_size; + } +#endif if (mdesc->soft_reboot) reboot_setup("s"); @@ -934,12 +938,6 @@ void __init setup_arch(char **cmdline_p) tcm_init(); -#ifdef CONFIG_ZONE_DMA - if (mdesc->dma_zone_size) { - extern unsigned long arm_dma_zone_size; - arm_dma_zone_size = mdesc->dma_zone_size; - } -#endif #ifdef CONFIG_MULTI_IRQ_HANDLER handle_arch_irq = mdesc->handle_irq; #endif diff --git a/arch/arm/kernel/unwind.c b/arch/arm/kernel/unwind.c index e7e8365795c3..3f03fe0c3269 100644 --- a/arch/arm/kernel/unwind.c +++ b/arch/arm/kernel/unwind.c @@ -67,7 +67,7 @@ EXPORT_SYMBOL(__aeabi_unwind_cpp_pr2); struct unwind_ctrl_block { unsigned long vrs[16]; /* virtual register set */ - unsigned long *insn; /* pointer to the current instructions word */ + const unsigned long *insn; /* pointer to the current instructions word */ int entries; /* number of entries left to interpret */ int byte; /* current byte number in the instructions word */ }; @@ -83,8 +83,9 @@ enum regs { PC = 15 }; -extern struct unwind_idx __start_unwind_idx[]; -extern struct unwind_idx __stop_unwind_idx[]; +extern const struct unwind_idx __start_unwind_idx[]; +static const struct unwind_idx *__origin_unwind_idx; +extern const struct unwind_idx __stop_unwind_idx[]; static DEFINE_SPINLOCK(unwind_lock); static LIST_HEAD(unwind_tables); @@ -98,45 +99,99 @@ static LIST_HEAD(unwind_tables); }) /* - * Binary search in the unwind index. The entries entries are + * Binary search in the unwind index. The entries are * guaranteed to be sorted in ascending order by the linker. + * + * start = first entry + * origin = first entry with positive offset (or stop if there is no such entry) + * stop - 1 = last entry */ -static struct unwind_idx *search_index(unsigned long addr, - struct unwind_idx *first, - struct unwind_idx *last) +static const struct unwind_idx *search_index(unsigned long addr, + const struct unwind_idx *start, + const struct unwind_idx *origin, + const struct unwind_idx *stop) { - pr_debug("%s(%08lx, %p, %p)\n", __func__, addr, first, last); + unsigned long addr_prel31; + + pr_debug("%s(%08lx, %p, %p, %p)\n", + __func__, addr, start, origin, stop); + + /* + * only search in the section with the matching sign. This way the + * prel31 numbers can be compared as unsigned longs. + */ + if (addr < (unsigned long)start) + /* negative offsets: [start; origin) */ + stop = origin; + else + /* positive offsets: [origin; stop) */ + start = origin; + + /* prel31 for address relavive to start */ + addr_prel31 = (addr - (unsigned long)start) & 0x7fffffff; - if (addr < first->addr) { + while (start < stop - 1) { + const struct unwind_idx *mid = start + ((stop - start) >> 1); + + /* + * As addr_prel31 is relative to start an offset is needed to + * make it relative to mid. + */ + if (addr_prel31 - ((unsigned long)mid - (unsigned long)start) < + mid->addr_offset) + stop = mid; + else { + /* keep addr_prel31 relative to start */ + addr_prel31 -= ((unsigned long)mid - + (unsigned long)start); + start = mid; + } + } + + if (likely(start->addr_offset <= addr_prel31)) + return start; + else { pr_warning("unwind: Unknown symbol address %08lx\n", addr); return NULL; - } else if (addr >= last->addr) - return last; + } +} - while (first < last - 1) { - struct unwind_idx *mid = first + ((last - first + 1) >> 1); +static const struct unwind_idx *unwind_find_origin( + const struct unwind_idx *start, const struct unwind_idx *stop) +{ + pr_debug("%s(%p, %p)\n", __func__, start, stop); + while (start < stop - 1) { + const struct unwind_idx *mid = start + ((stop - start) >> 1); - if (addr < mid->addr) - last = mid; + if (mid->addr_offset >= 0x40000000) + /* negative offset */ + start = mid; else - first = mid; + /* positive offset */ + stop = mid; } - - return first; + pr_debug("%s -> %p\n", __func__, stop); + return stop; } -static struct unwind_idx *unwind_find_idx(unsigned long addr) +static const struct unwind_idx *unwind_find_idx(unsigned long addr) { - struct unwind_idx *idx = NULL; + const struct unwind_idx *idx = NULL; unsigned long flags; pr_debug("%s(%08lx)\n", __func__, addr); - if (core_kernel_text(addr)) + if (core_kernel_text(addr)) { + if (unlikely(!__origin_unwind_idx)) + __origin_unwind_idx = + unwind_find_origin(__start_unwind_idx, + __stop_unwind_idx); + /* main unwind table */ idx = search_index(addr, __start_unwind_idx, - __stop_unwind_idx - 1); - else { + __origin_unwind_idx, + __stop_unwind_idx); + } else { /* module unwind tables */ struct unwind_table *table; @@ -145,7 +200,8 @@ static struct unwind_idx *unwind_find_idx(unsigned long addr) if (addr >= table->begin_addr && addr < table->end_addr) { idx = search_index(addr, table->start, - table->stop - 1); + table->origin, + table->stop); /* Move-to-front to exploit common traces */ list_move(&table->list, &unwind_tables); break; @@ -274,7 +330,7 @@ static int unwind_exec_insn(struct unwind_ctrl_block *ctrl) int unwind_frame(struct stackframe *frame) { unsigned long high, low; - struct unwind_idx *idx; + const struct unwind_idx *idx; struct unwind_ctrl_block ctrl; /* only go to a higher address on the stack */ @@ -399,7 +455,6 @@ struct unwind_table *unwind_table_add(unsigned long start, unsigned long size, unsigned long text_size) { unsigned long flags; - struct unwind_idx *idx; struct unwind_table *tab = kmalloc(sizeof(*tab), GFP_KERNEL); pr_debug("%s(%08lx, %08lx, %08lx, %08lx)\n", __func__, start, size, @@ -408,15 +463,12 @@ struct unwind_table *unwind_table_add(unsigned long start, unsigned long size, if (!tab) return tab; - tab->start = (struct unwind_idx *)start; - tab->stop = (struct unwind_idx *)(start + size); + tab->start = (const struct unwind_idx *)start; + tab->stop = (const struct unwind_idx *)(start + size); + tab->origin = unwind_find_origin(tab->start, tab->stop); tab->begin_addr = text_addr; tab->end_addr = text_addr + text_size; - /* Convert the symbol addresses to absolute values */ - for (idx = tab->start; idx < tab->stop; idx++) - idx->addr = prel31_to_addr(&idx->addr); - spin_lock_irqsave(&unwind_lock, flags); list_add_tail(&tab->list, &unwind_tables); spin_unlock_irqrestore(&unwind_lock, flags); @@ -437,16 +489,3 @@ void unwind_table_del(struct unwind_table *tab) kfree(tab); } - -int __init unwind_init(void) -{ - struct unwind_idx *idx; - - /* Convert the symbol addresses to absolute values */ - for (idx = __start_unwind_idx; idx < __stop_unwind_idx; idx++) - idx->addr = prel31_to_addr(&idx->addr); - - pr_debug("unwind: ARM stack unwinding initialised\n"); - - return 0; -} diff --git a/arch/x86/include/asm/e820.h b/arch/x86/include/asm/e820.h index c9547033e38e..908b96957d88 100644 --- a/arch/x86/include/asm/e820.h +++ b/arch/x86/include/asm/e820.h @@ -53,13 +53,6 @@ */ #define E820_RESERVED_KERN 128 -/* - * Address ranges that need to be mapped by the kernel direct - * mapping. This is used to make sure regions such as - * EFI_RUNTIME_SERVICES_DATA are directly mapped. See setup_arch(). - */ -#define E820_RESERVED_EFI 129 - #ifndef __ASSEMBLY__ #include <linux/types.h> struct e820entry { @@ -122,7 +115,6 @@ static inline void early_memtest(unsigned long start, unsigned long end) } #endif -extern unsigned long e820_end_pfn(unsigned long limit_pfn, unsigned type); extern unsigned long e820_end_of_ram_pfn(void); extern unsigned long e820_end_of_low_ram_pfn(void); extern u64 early_reserve_e820(u64 startt, u64 sizet, u64 align); diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index b8d8bfcd44a9..7093e4a6a0bc 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -33,6 +33,8 @@ extern unsigned long asmlinkage efi_call_phys(void *, ...); #define efi_call_virt6(f, a1, a2, a3, a4, a5, a6) \ efi_call_virt(f, a1, a2, a3, a4, a5, a6) +#define efi_ioremap(addr, size, type) ioremap_cache(addr, size) + #else /* !CONFIG_X86_32 */ extern u64 efi_call0(void *fp); @@ -82,6 +84,9 @@ extern u64 efi_call6(void *fp, u64 arg1, u64 arg2, u64 arg3, efi_call6((void *)(efi.systab->runtime->f), (u64)(a1), (u64)(a2), \ (u64)(a3), (u64)(a4), (u64)(a5), (u64)(a6)) +extern void __iomem *efi_ioremap(unsigned long addr, unsigned long size, + u32 type); + #endif /* CONFIG_X86_32 */ extern int add_efi_memmap; diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 65ffd110a81b..303a0e48f076 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -135,7 +135,6 @@ static void __init e820_print_type(u32 type) printk(KERN_CONT "(usable)"); break; case E820_RESERVED: - case E820_RESERVED_EFI: printk(KERN_CONT "(reserved)"); break; case E820_ACPI: @@ -784,7 +783,7 @@ u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align) /* * Find the highest page frame number we have available */ -unsigned long __init e820_end_pfn(unsigned long limit_pfn, unsigned type) +static unsigned long __init e820_end_pfn(unsigned long limit_pfn, unsigned type) { int i; unsigned long last_pfn = 0; diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 9a9e40fb091c..cf0ef986cb6d 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -691,8 +691,6 @@ early_param("reservelow", parse_reservelow); void __init setup_arch(char **cmdline_p) { - unsigned long end_pfn; - #ifdef CONFIG_X86_32 memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data)); visws_early_detect(); @@ -934,24 +932,7 @@ void __init setup_arch(char **cmdline_p) init_gbpages(); /* max_pfn_mapped is updated here */ - end_pfn = max_low_pfn; - -#ifdef CONFIG_X86_64 - /* - * There may be regions after the last E820_RAM region that we - * want to include in the kernel direct mapping, such as - * EFI_RUNTIME_SERVICES_DATA. - */ - if (efi_enabled) { - unsigned long efi_end; - - efi_end = e820_end_pfn(MAXMEM>>PAGE_SHIFT, E820_RESERVED_EFI); - if (efi_end > max_low_pfn) - end_pfn = efi_end; - } -#endif - - max_low_pfn_mapped = init_memory_mapping(0, end_pfn << PAGE_SHIFT); + max_low_pfn_mapped = init_memory_mapping(0, max_low_pfn<<PAGE_SHIFT); max_pfn_mapped = max_low_pfn_mapped; #ifdef CONFIG_X86_64 diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index c9718a16be15..37718f0f053d 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -323,13 +323,10 @@ static void __init do_add_efi_memmap(void) case EFI_UNUSABLE_MEMORY: e820_type = E820_UNUSABLE; break; - case EFI_RUNTIME_SERVICES_DATA: - e820_type = E820_RESERVED_EFI; - break; default: /* * EFI_RESERVED_TYPE EFI_RUNTIME_SERVICES_CODE - * EFI_MEMORY_MAPPED_IO + * EFI_RUNTIME_SERVICES_DATA EFI_MEMORY_MAPPED_IO * EFI_MEMORY_MAPPED_IO_PORT_SPACE EFI_PAL_CODE */ e820_type = E820_RESERVED; @@ -674,21 +671,10 @@ void __init efi_enter_virtual_mode(void) end_pfn = PFN_UP(end); if (end_pfn <= max_low_pfn_mapped || (end_pfn > (1UL << (32 - PAGE_SHIFT)) - && end_pfn <= max_pfn_mapped)) { + && end_pfn <= max_pfn_mapped)) va = __va(md->phys_addr); - - if (!(md->attribute & EFI_MEMORY_WB)) { - addr = (u64) (unsigned long)va; - npages = md->num_pages; - memrange_efi_to_native(&addr, &npages); - set_memory_uc(addr, npages); - } - } else { - if (!(md->attribute & EFI_MEMORY_WB)) - va = ioremap_nocache(md->phys_addr, size); - else - va = ioremap_cache(md->phys_addr, size); - } + else + va = efi_ioremap(md->phys_addr, size, md->type); md->virt_addr = (u64) (unsigned long) va; @@ -698,6 +684,13 @@ void __init efi_enter_virtual_mode(void) continue; } + if (!(md->attribute & EFI_MEMORY_WB)) { + addr = md->virt_addr; + npages = md->num_pages; + memrange_efi_to_native(&addr, &npages); + set_memory_uc(addr, npages); + } + systab = (u64) (unsigned long) efi_phys.systab; if (md->phys_addr <= systab && systab < end) { systab += md->virt_addr - md->phys_addr; diff --git a/arch/x86/platform/efi/efi_32.c b/arch/x86/platform/efi/efi_32.c index e36bf714cb77..40e446941dd7 100644 --- a/arch/x86/platform/efi/efi_32.c +++ b/arch/x86/platform/efi/efi_32.c @@ -39,43 +39,14 @@ */ static unsigned long efi_rt_eflags; -static pgd_t efi_bak_pg_dir_pointer[2]; void efi_call_phys_prelog(void) { - unsigned long cr4; - unsigned long temp; struct desc_ptr gdt_descr; local_irq_save(efi_rt_eflags); - /* - * If I don't have PAE, I should just duplicate two entries in page - * directory. If I have PAE, I just need to duplicate one entry in - * page directory. - */ - cr4 = read_cr4_safe(); - - if (cr4 & X86_CR4_PAE) { - efi_bak_pg_dir_pointer[0].pgd = - swapper_pg_dir[pgd_index(0)].pgd; - swapper_pg_dir[0].pgd = - swapper_pg_dir[pgd_index(PAGE_OFFSET)].pgd; - } else { - efi_bak_pg_dir_pointer[0].pgd = - swapper_pg_dir[pgd_index(0)].pgd; - efi_bak_pg_dir_pointer[1].pgd = - swapper_pg_dir[pgd_index(0x400000)].pgd; - swapper_pg_dir[pgd_index(0)].pgd = - swapper_pg_dir[pgd_index(PAGE_OFFSET)].pgd; - temp = PAGE_OFFSET + 0x400000; - swapper_pg_dir[pgd_index(0x400000)].pgd = - swapper_pg_dir[pgd_index(temp)].pgd; - } - - /* - * After the lock is released, the original page table is restored. - */ + load_cr3(initial_page_table); __flush_tlb_all(); gdt_descr.address = __pa(get_cpu_gdt_table(0)); @@ -85,28 +56,13 @@ void efi_call_phys_prelog(void) void efi_call_phys_epilog(void) { - unsigned long cr4; struct desc_ptr gdt_descr; gdt_descr.address = (unsigned long)get_cpu_gdt_table(0); gdt_descr.size = GDT_SIZE - 1; load_gdt(&gdt_descr); - cr4 = read_cr4_safe(); - - if (cr4 & X86_CR4_PAE) { - swapper_pg_dir[pgd_index(0)].pgd = - efi_bak_pg_dir_pointer[0].pgd; - } else { - swapper_pg_dir[pgd_index(0)].pgd = - efi_bak_pg_dir_pointer[0].pgd; - swapper_pg_dir[pgd_index(0x400000)].pgd = - efi_bak_pg_dir_pointer[1].pgd; - } - - /* - * After the lock is released, the original page table is restored. - */ + load_cr3(swapper_pg_dir); __flush_tlb_all(); local_irq_restore(efi_rt_eflags); diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index 312250c6b2de..ac3aa54e2654 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -80,3 +80,20 @@ void __init efi_call_phys_epilog(void) local_irq_restore(efi_flags); early_code_mapping_set_exec(0); } + +void __iomem *__init efi_ioremap(unsigned long phys_addr, unsigned long size, + u32 type) +{ + unsigned long last_map_pfn; + + if (type == EFI_MEMORY_MAPPED_IO) + return ioremap(phys_addr, size); + + last_map_pfn = init_memory_mapping(phys_addr, phys_addr + size); + if ((last_map_pfn << PAGE_SHIFT) < phys_addr + size) { + unsigned long top = last_map_pfn << PAGE_SHIFT; + efi_ioremap(top, size - (top - phys_addr), type); + } + + return (void __iomem *)__va(phys_addr); +} diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c index 65cc424359b0..148ab944378d 100644 --- a/drivers/block/rbd.c +++ b/drivers/block/rbd.c @@ -183,10 +183,6 @@ static LIST_HEAD(rbd_client_list); /* clients */ static int __rbd_init_snaps_header(struct rbd_device *rbd_dev); static void rbd_dev_release(struct device *dev); -static ssize_t rbd_snap_rollback(struct device *dev, - struct device_attribute *attr, - const char *buf, - size_t size); static ssize_t rbd_snap_add(struct device *dev, struct device_attribute *attr, const char *buf, @@ -461,6 +457,10 @@ static int rbd_header_from_disk(struct rbd_image_header *header, u32 snap_count = le32_to_cpu(ondisk->snap_count); int ret = -ENOMEM; + if (memcmp(ondisk, RBD_HEADER_TEXT, sizeof(RBD_HEADER_TEXT))) { + return -ENXIO; + } + init_rwsem(&header->snap_rwsem); header->snap_names_len = le64_to_cpu(ondisk->snap_names_len); header->snapc = kmalloc(sizeof(struct ceph_snap_context) + @@ -1356,32 +1356,6 @@ fail: } /* - * Request sync osd rollback - */ -static int rbd_req_sync_rollback_obj(struct rbd_device *dev, - u64 snapid, - const char *obj) -{ - struct ceph_osd_req_op *ops; - int ret = rbd_create_rw_ops(&ops, 1, CEPH_OSD_OP_ROLLBACK, 0); - if (ret < 0) - return ret; - - ops[0].snap.snapid = snapid; - - ret = rbd_req_sync_op(dev, NULL, - CEPH_NOSNAP, - 0, - CEPH_OSD_FLAG_WRITE | CEPH_OSD_FLAG_ONDISK, - ops, - 1, obj, 0, 0, NULL, NULL, NULL); - - rbd_destroy_ops(ops); - - return ret; -} - -/* * Request sync osd read */ static int rbd_req_sync_exec(struct rbd_device *dev, @@ -1610,8 +1584,13 @@ static int rbd_read_header(struct rbd_device *rbd_dev, goto out_dh; rc = rbd_header_from_disk(header, dh, snap_count, GFP_KERNEL); - if (rc < 0) + if (rc < 0) { + if (rc == -ENXIO) { + pr_warning("unrecognized header format" + " for image %s", rbd_dev->obj); + } goto out_dh; + } if (snap_count != header->total_snaps) { snap_count = header->total_snaps; @@ -1882,7 +1861,6 @@ static DEVICE_ATTR(name, S_IRUGO, rbd_name_show, NULL); static DEVICE_ATTR(refresh, S_IWUSR, NULL, rbd_image_refresh); static DEVICE_ATTR(current_snap, S_IRUGO, rbd_snap_show, NULL); static DEVICE_ATTR(create_snap, S_IWUSR, NULL, rbd_snap_add); -static DEVICE_ATTR(rollback_snap, S_IWUSR, NULL, rbd_snap_rollback); static struct attribute *rbd_attrs[] = { &dev_attr_size.attr, @@ -1893,7 +1871,6 @@ static struct attribute *rbd_attrs[] = { &dev_attr_current_snap.attr, &dev_attr_refresh.attr, &dev_attr_create_snap.attr, - &dev_attr_rollback_snap.attr, NULL }; @@ -2424,64 +2401,6 @@ err_unlock: return ret; } -static ssize_t rbd_snap_rollback(struct device *dev, - struct device_attribute *attr, - const char *buf, - size_t count) -{ - struct rbd_device *rbd_dev = dev_to_rbd(dev); - int ret; - u64 snapid; - u64 cur_ofs; - char *seg_name = NULL; - char *snap_name = kmalloc(count + 1, GFP_KERNEL); - ret = -ENOMEM; - if (!snap_name) - return ret; - - /* parse snaps add command */ - snprintf(snap_name, count, "%s", buf); - seg_name = kmalloc(RBD_MAX_SEG_NAME_LEN + 1, GFP_NOIO); - if (!seg_name) - goto done; - - mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING); - - ret = snap_by_name(&rbd_dev->header, snap_name, &snapid, NULL); - if (ret < 0) - goto done_unlock; - - dout("snapid=%lld\n", snapid); - - cur_ofs = 0; - while (cur_ofs < rbd_dev->header.image_size) { - cur_ofs += rbd_get_segment(&rbd_dev->header, - rbd_dev->obj, - cur_ofs, (u64)-1, - seg_name, NULL); - dout("seg_name=%s\n", seg_name); - - ret = rbd_req_sync_rollback_obj(rbd_dev, snapid, seg_name); - if (ret < 0) - pr_warning("could not roll back obj %s err=%d\n", - seg_name, ret); - } - - ret = __rbd_update_snaps(rbd_dev); - if (ret < 0) - goto done_unlock; - - ret = count; - -done_unlock: - mutex_unlock(&ctl_mutex); -done: - kfree(seg_name); - kfree(snap_name); - - return ret; -} - static struct bus_attribute rbd_bus_attrs[] = { __ATTR(add, S_IWUSR, NULL, rbd_add), __ATTR(remove, S_IWUSR, NULL, rbd_remove), diff --git a/drivers/hwmon/jz4740-hwmon.c b/drivers/hwmon/jz4740-hwmon.c index 7a48b1eb4233..5253d23361d9 100644 --- a/drivers/hwmon/jz4740-hwmon.c +++ b/drivers/hwmon/jz4740-hwmon.c @@ -59,7 +59,7 @@ static ssize_t jz4740_hwmon_read_adcin(struct device *dev, { struct jz4740_hwmon *hwmon = dev_get_drvdata(dev); struct completion *completion = &hwmon->read_completion; - unsigned long t; + long t; unsigned long val; int ret; @@ -203,7 +203,7 @@ static int __devexit jz4740_hwmon_remove(struct platform_device *pdev) return 0; } -struct platform_driver jz4740_hwmon_driver = { +static struct platform_driver jz4740_hwmon_driver = { .probe = jz4740_hwmon_probe, .remove = __devexit_p(jz4740_hwmon_remove), .driver = { diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c index a1cb21f95302..1e0e27cbe987 100644 --- a/drivers/mmc/card/block.c +++ b/drivers/mmc/card/block.c @@ -1606,6 +1606,14 @@ static const struct mmc_fixup blk_fixups[] = MMC_QUIRK_BLK_NO_CMD23), MMC_FIXUP("MMC32G", 0x11, CID_OEMID_ANY, add_quirk_mmc, MMC_QUIRK_BLK_NO_CMD23), + + /* + * Some Micron MMC cards needs longer data read timeout than + * indicated in CSD. + */ + MMC_FIXUP(CID_NAME_ANY, 0x13, 0x200, add_quirk_mmc, + MMC_QUIRK_LONG_READ_TIME), + END_FIXUP }; diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c index 5278ffb20e74..950b97d7412a 100644 --- a/drivers/mmc/core/core.c +++ b/drivers/mmc/core/core.c @@ -529,6 +529,18 @@ void mmc_set_data_timeout(struct mmc_data *data, const struct mmc_card *card) data->timeout_clks = 0; } } + + /* + * Some cards require longer data read timeout than indicated in CSD. + * Address this by setting the read timeout to a "reasonably high" + * value. For the cards tested, 300ms has proven enough. If necessary, + * this value can be increased if other problematic cards require this. + */ + if (mmc_card_long_read_time(card) && data->flags & MMC_DATA_READ) { + data->timeout_ns = 300000000; + data->timeout_clks = 0; + } + /* * Some cards need very high timeouts if driven in SPI mode. * The worst observed timeout was 900ms after writing a @@ -1213,6 +1225,46 @@ void mmc_set_driver_type(struct mmc_host *host, unsigned int drv_type) mmc_host_clk_release(host); } +static void mmc_poweroff_notify(struct mmc_host *host) +{ + struct mmc_card *card; + unsigned int timeout; + unsigned int notify_type = EXT_CSD_NO_POWER_NOTIFICATION; + int err = 0; + + card = host->card; + + /* + * Send power notify command only if card + * is mmc and notify state is powered ON + */ + if (card && mmc_card_mmc(card) && + (card->poweroff_notify_state == MMC_POWERED_ON)) { + + if (host->power_notify_type == MMC_HOST_PW_NOTIFY_SHORT) { + notify_type = EXT_CSD_POWER_OFF_SHORT; + timeout = card->ext_csd.generic_cmd6_time; + card->poweroff_notify_state = MMC_POWEROFF_SHORT; + } else { + notify_type = EXT_CSD_POWER_OFF_LONG; + timeout = card->ext_csd.power_off_longtime; + card->poweroff_notify_state = MMC_POWEROFF_LONG; + } + + err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL, + EXT_CSD_POWER_OFF_NOTIFICATION, + notify_type, timeout); + + if (err && err != -EBADMSG) + pr_err("Device failed to respond within %d poweroff " + "time. Forcefully powering down the device\n", + timeout); + + /* Set the card state to no notification after the poweroff */ + card->poweroff_notify_state = MMC_NO_POWER_NOTIFICATION; + } +} + /* * Apply power to the MMC stack. This is a two-stage process. * First, we enable power to the card without the clock running. @@ -1269,42 +1321,12 @@ static void mmc_power_up(struct mmc_host *host) void mmc_power_off(struct mmc_host *host) { - struct mmc_card *card; - unsigned int notify_type; - unsigned int timeout; - int err; - mmc_host_clk_hold(host); - card = host->card; host->ios.clock = 0; host->ios.vdd = 0; - if (card && mmc_card_mmc(card) && - (card->poweroff_notify_state == MMC_POWERED_ON)) { - - if (host->power_notify_type == MMC_HOST_PW_NOTIFY_SHORT) { - notify_type = EXT_CSD_POWER_OFF_SHORT; - timeout = card->ext_csd.generic_cmd6_time; - card->poweroff_notify_state = MMC_POWEROFF_SHORT; - } else { - notify_type = EXT_CSD_POWER_OFF_LONG; - timeout = card->ext_csd.power_off_longtime; - card->poweroff_notify_state = MMC_POWEROFF_LONG; - } - - err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL, - EXT_CSD_POWER_OFF_NOTIFICATION, - notify_type, timeout); - - if (err && err != -EBADMSG) - pr_err("Device failed to respond within %d poweroff " - "time. Forcefully powering down the device\n", - timeout); - - /* Set the card state to no notification after the poweroff */ - card->poweroff_notify_state = MMC_NO_POWER_NOTIFICATION; - } + mmc_poweroff_notify(host); /* * Reset ocr mask to be the highest possible voltage supported for @@ -2196,7 +2218,7 @@ int mmc_card_sleep(struct mmc_host *host) mmc_bus_get(host); - if (host->bus_ops && !host->bus_dead && host->bus_ops->awake) + if (host->bus_ops && !host->bus_dead && host->bus_ops->sleep) err = host->bus_ops->sleep(host); mmc_bus_put(host); @@ -2302,8 +2324,17 @@ int mmc_suspend_host(struct mmc_host *host) * pre-claim the host. */ if (mmc_try_claim_host(host)) { - if (host->bus_ops->suspend) + if (host->bus_ops->suspend) { + /* + * For eMMC 4.5 device send notify command + * before sleep, because in sleep state eMMC 4.5 + * devices respond to only RESET and AWAKE cmd + */ + mmc_poweroff_notify(host); err = host->bus_ops->suspend(host); + } + mmc_do_release_host(host); + if (err == -ENOSYS || !host->bus_ops->resume) { /* * We simply "remove" the card in this case. @@ -2318,7 +2349,6 @@ int mmc_suspend_host(struct mmc_host *host) host->pm_flags = 0; err = 0; } - mmc_do_release_host(host); } else { err = -EBUSY; } diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c index dbf421a6279c..d240427c1246 100644 --- a/drivers/mmc/core/mmc.c +++ b/drivers/mmc/core/mmc.c @@ -876,17 +876,21 @@ static int mmc_init_card(struct mmc_host *host, u32 ocr, * set the notification byte in the ext_csd register of device */ if ((host->caps2 & MMC_CAP2_POWEROFF_NOTIFY) && - (card->poweroff_notify_state == MMC_NO_POWER_NOTIFICATION)) { + (card->ext_csd.rev >= 6)) { err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL, EXT_CSD_POWER_OFF_NOTIFICATION, EXT_CSD_POWER_ON, card->ext_csd.generic_cmd6_time); if (err && err != -EBADMSG) goto free_card; - } - if (!err) - card->poweroff_notify_state = MMC_POWERED_ON; + /* + * The err can be -EBADMSG or 0, + * so check for success and update the flag + */ + if (!err) + card->poweroff_notify_state = MMC_POWERED_ON; + } /* * Activate high speed (if supported) diff --git a/drivers/mmc/host/mxcmmc.c b/drivers/mmc/host/mxcmmc.c index 325ea61e12d3..8e0fbe994047 100644 --- a/drivers/mmc/host/mxcmmc.c +++ b/drivers/mmc/host/mxcmmc.c @@ -732,6 +732,7 @@ static void mxcmci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios) "failed to config DMA channel. Falling back to PIO\n"); dma_release_channel(host->dma); host->do_dma = 0; + host->dma = NULL; } } diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c index 101cd31c8220..d5fe43d53c51 100644 --- a/drivers/mmc/host/omap_hsmmc.c +++ b/drivers/mmc/host/omap_hsmmc.c @@ -1010,6 +1010,7 @@ static void omap_hsmmc_dma_cleanup(struct omap_hsmmc_host *host, int errno) host->data->sg_len, omap_hsmmc_get_dma_dir(host, host->data)); omap_free_dma(dma_ch); + host->data->host_cookie = 0; } host->data = NULL; } @@ -1575,8 +1576,10 @@ static void omap_hsmmc_post_req(struct mmc_host *mmc, struct mmc_request *mrq, struct mmc_data *data = mrq->data; if (host->use_dma) { - dma_unmap_sg(mmc_dev(host->mmc), data->sg, data->sg_len, - omap_hsmmc_get_dma_dir(host, data)); + if (data->host_cookie) + dma_unmap_sg(mmc_dev(host->mmc), data->sg, + data->sg_len, + omap_hsmmc_get_dma_dir(host, data)); data->host_cookie = 0; } } diff --git a/drivers/mmc/host/sdhci-cns3xxx.c b/drivers/mmc/host/sdhci-cns3xxx.c index 4b920b7621cf..87b6f079b6e0 100644 --- a/drivers/mmc/host/sdhci-cns3xxx.c +++ b/drivers/mmc/host/sdhci-cns3xxx.c @@ -15,6 +15,7 @@ #include <linux/delay.h> #include <linux/device.h> #include <linux/mmc/host.h> +#include <linux/module.h> #include <mach/cns3xxx.h> #include "sdhci-pltfm.h" diff --git a/drivers/mmc/host/sdhci-s3c.c b/drivers/mmc/host/sdhci-s3c.c index 3d00e722efc9..cb60c4197e0a 100644 --- a/drivers/mmc/host/sdhci-s3c.c +++ b/drivers/mmc/host/sdhci-s3c.c @@ -644,8 +644,6 @@ static int sdhci_s3c_resume(struct platform_device *dev) static struct platform_driver sdhci_s3c_driver = { .probe = sdhci_s3c_probe, .remove = __devexit_p(sdhci_s3c_remove), - .suspend = sdhci_s3c_suspend, - .resume = sdhci_s3c_resume, .driver = { .owner = THIS_MODULE, .name = "s3c-sdhci", diff --git a/drivers/mmc/host/sh_mmcif.c b/drivers/mmc/host/sh_mmcif.c index 369366c8e205..d5505f3fe2a1 100644 --- a/drivers/mmc/host/sh_mmcif.c +++ b/drivers/mmc/host/sh_mmcif.c @@ -908,7 +908,7 @@ static void sh_mmcif_set_ios(struct mmc_host *mmc, struct mmc_ios *ios) if (host->power) { pm_runtime_put(&host->pd->dev); host->power = false; - if (p->down_pwr) + if (p->down_pwr && ios->power_mode == MMC_POWER_OFF) p->down_pwr(host->pd); } host->state = STATE_IDLE; diff --git a/drivers/mmc/host/tmio_mmc_pio.c b/drivers/mmc/host/tmio_mmc_pio.c index d85a60cda167..4208b3958069 100644 --- a/drivers/mmc/host/tmio_mmc_pio.c +++ b/drivers/mmc/host/tmio_mmc_pio.c @@ -798,7 +798,7 @@ static void tmio_mmc_set_ios(struct mmc_host *mmc, struct mmc_ios *ios) /* start bus clock */ tmio_mmc_clk_start(host); } else if (ios->power_mode != MMC_POWER_UP) { - if (host->set_pwr) + if (host->set_pwr && ios->power_mode == MMC_POWER_OFF) host->set_pwr(host->pdev, 0); if ((pdata->flags & TMIO_MMC_HAS_COLD_CD) && pdata->power) { diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 4144caf2f9d3..173b1d22e59b 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -87,7 +87,7 @@ static int ceph_set_page_dirty(struct page *page) snapc = ceph_get_snap_context(ci->i_snap_realm->cached_context); /* dirty the head */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_head_snapc == NULL) ci->i_head_snapc = ceph_get_snap_context(snapc); ++ci->i_wrbuffer_ref_head; @@ -100,7 +100,7 @@ static int ceph_set_page_dirty(struct page *page) ci->i_wrbuffer_ref-1, ci->i_wrbuffer_ref_head-1, ci->i_wrbuffer_ref, ci->i_wrbuffer_ref_head, snapc, snapc->seq, snapc->num_snaps); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); /* now adjust page */ spin_lock_irq(&mapping->tree_lock); @@ -391,7 +391,7 @@ static struct ceph_snap_context *get_oldest_context(struct inode *inode, struct ceph_snap_context *snapc = NULL; struct ceph_cap_snap *capsnap = NULL; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) { dout(" cap_snap %p snapc %p has %d dirty pages\n", capsnap, capsnap->context, capsnap->dirty_pages); @@ -407,7 +407,7 @@ static struct ceph_snap_context *get_oldest_context(struct inode *inode, dout(" head snapc %p has %d dirty pages\n", snapc, ci->i_wrbuffer_ref_head); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return snapc; } diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 0f327c6c9679..8b53193e4f7c 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -309,7 +309,7 @@ void ceph_reservation_status(struct ceph_fs_client *fsc, /* * Find ceph_cap for given mds, if any. * - * Called with i_lock held. + * Called with i_ceph_lock held. */ static struct ceph_cap *__get_cap_for_mds(struct ceph_inode_info *ci, int mds) { @@ -332,9 +332,9 @@ struct ceph_cap *ceph_get_cap_for_mds(struct ceph_inode_info *ci, int mds) { struct ceph_cap *cap; - spin_lock(&ci->vfs_inode.i_lock); + spin_lock(&ci->i_ceph_lock); cap = __get_cap_for_mds(ci, mds); - spin_unlock(&ci->vfs_inode.i_lock); + spin_unlock(&ci->i_ceph_lock); return cap; } @@ -361,15 +361,16 @@ static int __ceph_get_cap_mds(struct ceph_inode_info *ci) int ceph_get_cap_mds(struct inode *inode) { + struct ceph_inode_info *ci = ceph_inode(inode); int mds; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); mds = __ceph_get_cap_mds(ceph_inode(inode)); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return mds; } /* - * Called under i_lock. + * Called under i_ceph_lock. */ static void __insert_cap_node(struct ceph_inode_info *ci, struct ceph_cap *new) @@ -415,7 +416,7 @@ static void __cap_set_timeouts(struct ceph_mds_client *mdsc, * * If I_FLUSH is set, leave the inode at the front of the list. * - * Caller holds i_lock + * Caller holds i_ceph_lock * -> we take mdsc->cap_delay_lock */ static void __cap_delay_requeue(struct ceph_mds_client *mdsc, @@ -457,7 +458,7 @@ static void __cap_delay_requeue_front(struct ceph_mds_client *mdsc, /* * Cancel delayed work on cap. * - * Caller must hold i_lock. + * Caller must hold i_ceph_lock. */ static void __cap_delay_cancel(struct ceph_mds_client *mdsc, struct ceph_inode_info *ci) @@ -532,14 +533,14 @@ int ceph_add_cap(struct inode *inode, wanted |= ceph_caps_for_mode(fmode); retry: - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap = __get_cap_for_mds(ci, mds); if (!cap) { if (new_cap) { cap = new_cap; new_cap = NULL; } else { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); new_cap = get_cap(mdsc, caps_reservation); if (new_cap == NULL) return -ENOMEM; @@ -625,7 +626,7 @@ retry: if (fmode >= 0) __ceph_get_fmode(ci, fmode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); wake_up_all(&ci->i_cap_wq); return 0; } @@ -792,7 +793,7 @@ int ceph_caps_revoking(struct ceph_inode_info *ci, int mask) struct rb_node *p; int ret = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); for (p = rb_first(&ci->i_caps); p; p = rb_next(p)) { cap = rb_entry(p, struct ceph_cap, ci_node); if (__cap_is_valid(cap) && @@ -801,7 +802,7 @@ int ceph_caps_revoking(struct ceph_inode_info *ci, int mask) break; } } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("ceph_caps_revoking %p %s = %d\n", inode, ceph_cap_string(mask), ret); return ret; @@ -855,7 +856,7 @@ int __ceph_caps_mds_wanted(struct ceph_inode_info *ci) } /* - * called under i_lock + * called under i_ceph_lock */ static int __ceph_is_any_caps(struct ceph_inode_info *ci) { @@ -865,7 +866,7 @@ static int __ceph_is_any_caps(struct ceph_inode_info *ci) /* * Remove a cap. Take steps to deal with a racing iterate_session_caps. * - * caller should hold i_lock. + * caller should hold i_ceph_lock. * caller will not hold session s_mutex if called from destroy_inode. */ void __ceph_remove_cap(struct ceph_cap *cap) @@ -1028,7 +1029,7 @@ static void __queue_cap_release(struct ceph_mds_session *session, /* * Queue cap releases when an inode is dropped from our cache. Since - * inode is about to be destroyed, there is no need for i_lock. + * inode is about to be destroyed, there is no need for i_ceph_lock. */ void ceph_queue_caps_release(struct inode *inode) { @@ -1049,7 +1050,7 @@ void ceph_queue_caps_release(struct inode *inode) /* * Send a cap msg on the given inode. Update our caps state, then - * drop i_lock and send the message. + * drop i_ceph_lock and send the message. * * Make note of max_size reported/requested from mds, revoked caps * that have now been implemented. @@ -1061,13 +1062,13 @@ void ceph_queue_caps_release(struct inode *inode) * Return non-zero if delayed release, or we experienced an error * such that the caller should requeue + retry later. * - * called with i_lock, then drops it. + * called with i_ceph_lock, then drops it. * caller should hold snap_rwsem (read), s_mutex. */ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap, int op, int used, int want, int retain, int flushing, unsigned *pflush_tid) - __releases(cap->ci->vfs_inode->i_lock) + __releases(cap->ci->i_ceph_lock) { struct ceph_inode_info *ci = cap->ci; struct inode *inode = &ci->vfs_inode; @@ -1170,7 +1171,7 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap, xattr_version = ci->i_xattrs.version; } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); ret = send_cap_msg(session, ceph_vino(inode).ino, cap_id, op, keep, want, flushing, seq, flush_tid, issue_seq, mseq, @@ -1198,13 +1199,13 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap, * Unless @again is true, skip cap_snaps that were already sent to * the MDS (i.e., during this session). * - * Called under i_lock. Takes s_mutex as needed. + * Called under i_ceph_lock. Takes s_mutex as needed. */ void __ceph_flush_snaps(struct ceph_inode_info *ci, struct ceph_mds_session **psession, int again) - __releases(ci->vfs_inode->i_lock) - __acquires(ci->vfs_inode->i_lock) + __releases(ci->i_ceph_lock) + __acquires(ci->i_ceph_lock) { struct inode *inode = &ci->vfs_inode; int mds; @@ -1261,7 +1262,7 @@ retry: session = NULL; } if (!session) { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); mutex_lock(&mdsc->mutex); session = __ceph_lookup_mds_session(mdsc, mds); mutex_unlock(&mdsc->mutex); @@ -1275,7 +1276,7 @@ retry: * deletion or migration. retry, and we'll * get a better @mds value next time. */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); goto retry; } @@ -1285,7 +1286,7 @@ retry: list_del_init(&capsnap->flushing_item); list_add_tail(&capsnap->flushing_item, &session->s_cap_snaps_flushing); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("flush_snaps %p cap_snap %p follows %lld tid %llu\n", inode, capsnap, capsnap->follows, capsnap->flush_tid); @@ -1302,7 +1303,7 @@ retry: next_follows = capsnap->follows + 1; ceph_put_cap_snap(capsnap); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); goto retry; } @@ -1322,11 +1323,9 @@ out: static void ceph_flush_snaps(struct ceph_inode_info *ci) { - struct inode *inode = &ci->vfs_inode; - - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __ceph_flush_snaps(ci, NULL, 0); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } /* @@ -1373,7 +1372,7 @@ int __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask) * Add dirty inode to the flushing list. Assigned a seq number so we * can wait for caps to flush without starving. * - * Called under i_lock. + * Called under i_ceph_lock. */ static int __mark_caps_flushing(struct inode *inode, struct ceph_mds_session *session) @@ -1421,9 +1420,9 @@ static int try_nonblocking_invalidate(struct inode *inode) struct ceph_inode_info *ci = ceph_inode(inode); u32 invalidating_gen = ci->i_rdcache_gen; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); invalidate_mapping_pages(&inode->i_data, 0, -1); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (inode->i_data.nrpages == 0 && invalidating_gen == ci->i_rdcache_gen) { @@ -1470,7 +1469,7 @@ void ceph_check_caps(struct ceph_inode_info *ci, int flags, if (mdsc->stopping) is_delayed = 1; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_ceph_flags & CEPH_I_FLUSH) flags |= CHECK_CAPS_FLUSH; @@ -1480,7 +1479,7 @@ void ceph_check_caps(struct ceph_inode_info *ci, int flags, __ceph_flush_snaps(ci, &session, 0); goto retry_locked; retry: - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); retry_locked: file_wanted = __ceph_caps_file_wanted(ci); used = __ceph_caps_used(ci); @@ -1634,7 +1633,7 @@ ack: if (mutex_trylock(&session->s_mutex) == 0) { dout("inverting session/ino locks on %p\n", session); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (took_snap_rwsem) { up_read(&mdsc->snap_rwsem); took_snap_rwsem = 0; @@ -1648,7 +1647,7 @@ ack: if (down_read_trylock(&mdsc->snap_rwsem) == 0) { dout("inverting snap/in locks on %p\n", inode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); down_read(&mdsc->snap_rwsem); took_snap_rwsem = 1; goto retry; @@ -1664,10 +1663,10 @@ ack: mds = cap->mds; /* remember mds, so we don't repeat */ sent++; - /* __send_cap drops i_lock */ + /* __send_cap drops i_ceph_lock */ delayed += __send_cap(mdsc, cap, CEPH_CAP_OP_UPDATE, used, want, retain, flushing, NULL); - goto retry; /* retake i_lock and restart our cap scan. */ + goto retry; /* retake i_ceph_lock and restart our cap scan. */ } /* @@ -1681,7 +1680,7 @@ ack: else if (!is_delayed || force_requeue) __cap_delay_requeue(mdsc, ci); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (queue_invalidate) ceph_queue_invalidate(inode); @@ -1704,7 +1703,7 @@ static int try_flush_caps(struct inode *inode, struct ceph_mds_session *session, int flushing = 0; retry: - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_ceph_flags & CEPH_I_NOFLUSH) { dout("try_flush_caps skipping %p I_NOFLUSH set\n", inode); goto out; @@ -1716,7 +1715,7 @@ retry: int delayed; if (!session) { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); session = cap->session; mutex_lock(&session->s_mutex); goto retry; @@ -1727,18 +1726,18 @@ retry: flushing = __mark_caps_flushing(inode, session); - /* __send_cap drops i_lock */ + /* __send_cap drops i_ceph_lock */ delayed = __send_cap(mdsc, cap, CEPH_CAP_OP_FLUSH, used, want, cap->issued | cap->implemented, flushing, flush_tid); if (!delayed) goto out_unlocked; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __cap_delay_requeue(mdsc, ci); } out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); out_unlocked: if (session && unlock_session) mutex_unlock(&session->s_mutex); @@ -1753,7 +1752,7 @@ static int caps_are_flushed(struct inode *inode, unsigned tid) struct ceph_inode_info *ci = ceph_inode(inode); int i, ret = 1; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); for (i = 0; i < CEPH_CAP_BITS; i++) if ((ci->i_flushing_caps & (1 << i)) && ci->i_cap_flush_tid[i] <= tid) { @@ -1761,7 +1760,7 @@ static int caps_are_flushed(struct inode *inode, unsigned tid) ret = 0; break; } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return ret; } @@ -1868,10 +1867,10 @@ int ceph_write_inode(struct inode *inode, struct writeback_control *wbc) struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (__ceph_caps_dirty(ci)) __cap_delay_requeue_front(mdsc, ci); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } return err; } @@ -1894,7 +1893,7 @@ static void kick_flushing_capsnaps(struct ceph_mds_client *mdsc, struct inode *inode = &ci->vfs_inode; struct ceph_cap *cap; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap = ci->i_auth_cap; if (cap && cap->session == session) { dout("kick_flushing_caps %p cap %p capsnap %p\n", inode, @@ -1904,7 +1903,7 @@ static void kick_flushing_capsnaps(struct ceph_mds_client *mdsc, pr_err("%p auth cap %p not mds%d ???\n", inode, cap, session->s_mds); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } } @@ -1921,7 +1920,7 @@ void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc, struct ceph_cap *cap; int delayed = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap = ci->i_auth_cap; if (cap && cap->session == session) { dout("kick_flushing_caps %p cap %p %s\n", inode, @@ -1932,14 +1931,14 @@ void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc, cap->issued | cap->implemented, ci->i_flushing_caps, NULL); if (delayed) { - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __cap_delay_requeue(mdsc, ci); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } } else { pr_err("%p auth cap %p not mds%d ???\n", inode, cap, session->s_mds); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } } } @@ -1952,7 +1951,7 @@ static void kick_flushing_inode_caps(struct ceph_mds_client *mdsc, struct ceph_cap *cap; int delayed = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap = ci->i_auth_cap; dout("kick_flushing_inode_caps %p flushing %s flush_seq %lld\n", inode, ceph_cap_string(ci->i_flushing_caps), ci->i_cap_flush_seq); @@ -1964,12 +1963,12 @@ static void kick_flushing_inode_caps(struct ceph_mds_client *mdsc, cap->issued | cap->implemented, ci->i_flushing_caps, NULL); if (delayed) { - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __cap_delay_requeue(mdsc, ci); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } } else { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } } @@ -1978,7 +1977,7 @@ static void kick_flushing_inode_caps(struct ceph_mds_client *mdsc, * Take references to capabilities we hold, so that we don't release * them to the MDS prematurely. * - * Protected by i_lock. + * Protected by i_ceph_lock. */ static void __take_cap_refs(struct ceph_inode_info *ci, int got) { @@ -2016,7 +2015,7 @@ static int try_get_cap_refs(struct ceph_inode_info *ci, int need, int want, dout("get_cap_refs %p need %s want %s\n", inode, ceph_cap_string(need), ceph_cap_string(want)); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); /* make sure file is actually open */ file_wanted = __ceph_caps_file_wanted(ci); @@ -2077,7 +2076,7 @@ static int try_get_cap_refs(struct ceph_inode_info *ci, int need, int want, ceph_cap_string(have), ceph_cap_string(need)); } out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("get_cap_refs %p ret %d got %s\n", inode, ret, ceph_cap_string(*got)); return ret; @@ -2094,7 +2093,7 @@ static void check_max_size(struct inode *inode, loff_t endoff) int check = 0; /* do we need to explicitly request a larger max_size? */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if ((endoff >= ci->i_max_size || endoff > (inode->i_size << 1)) && endoff > ci->i_wanted_max_size) { @@ -2103,7 +2102,7 @@ static void check_max_size(struct inode *inode, loff_t endoff) ci->i_wanted_max_size = endoff; check = 1; } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (check) ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL); } @@ -2140,9 +2139,9 @@ retry: */ void ceph_get_cap_refs(struct ceph_inode_info *ci, int caps) { - spin_lock(&ci->vfs_inode.i_lock); + spin_lock(&ci->i_ceph_lock); __take_cap_refs(ci, caps); - spin_unlock(&ci->vfs_inode.i_lock); + spin_unlock(&ci->i_ceph_lock); } /* @@ -2160,7 +2159,7 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had) int last = 0, put = 0, flushsnaps = 0, wake = 0; struct ceph_cap_snap *capsnap; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (had & CEPH_CAP_PIN) --ci->i_pin_ref; if (had & CEPH_CAP_FILE_RD) @@ -2193,7 +2192,7 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had) } } } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("put_cap_refs %p had %s%s%s\n", inode, ceph_cap_string(had), last ? " last" : "", put ? " put" : ""); @@ -2225,7 +2224,7 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr, int found = 0; struct ceph_cap_snap *capsnap = NULL; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_wrbuffer_ref -= nr; last = !ci->i_wrbuffer_ref; @@ -2274,7 +2273,7 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr, } } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (last) { ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL); @@ -2291,7 +2290,7 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr, * Handle a cap GRANT message from the MDS. (Note that a GRANT may * actually be a revocation if it specifies a smaller cap set.) * - * caller holds s_mutex and i_lock, we drop both. + * caller holds s_mutex and i_ceph_lock, we drop both. * * return value: * 0 - ok @@ -2302,7 +2301,7 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, struct ceph_mds_session *session, struct ceph_cap *cap, struct ceph_buffer *xattr_buf) - __releases(inode->i_lock) + __releases(ci->i_ceph_lock) { struct ceph_inode_info *ci = ceph_inode(inode); int mds = session->s_mds; @@ -2453,7 +2452,7 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, } BUG_ON(cap->issued & ~cap->implemented); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (writeback) /* * queue inode for writeback: we can't actually call @@ -2483,7 +2482,7 @@ static void handle_cap_flush_ack(struct inode *inode, u64 flush_tid, struct ceph_mds_caps *m, struct ceph_mds_session *session, struct ceph_cap *cap) - __releases(inode->i_lock) + __releases(ci->i_ceph_lock) { struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc; @@ -2539,7 +2538,7 @@ static void handle_cap_flush_ack(struct inode *inode, u64 flush_tid, wake_up_all(&ci->i_cap_wq); out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (drop) iput(inode); } @@ -2562,7 +2561,7 @@ static void handle_cap_flushsnap_ack(struct inode *inode, u64 flush_tid, dout("handle_cap_flushsnap_ack inode %p ci %p mds%d follows %lld\n", inode, ci, session->s_mds, follows); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) { if (capsnap->follows == follows) { if (capsnap->flush_tid != flush_tid) { @@ -2585,7 +2584,7 @@ static void handle_cap_flushsnap_ack(struct inode *inode, u64 flush_tid, capsnap, capsnap->follows); } } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (drop) iput(inode); } @@ -2598,7 +2597,7 @@ static void handle_cap_flushsnap_ack(struct inode *inode, u64 flush_tid, static void handle_cap_trunc(struct inode *inode, struct ceph_mds_caps *trunc, struct ceph_mds_session *session) - __releases(inode->i_lock) + __releases(ci->i_ceph_lock) { struct ceph_inode_info *ci = ceph_inode(inode); int mds = session->s_mds; @@ -2617,7 +2616,7 @@ static void handle_cap_trunc(struct inode *inode, inode, mds, seq, truncate_size, truncate_seq); queue_trunc = ceph_fill_file_size(inode, issued, truncate_seq, truncate_size, size); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (queue_trunc) ceph_queue_vmtruncate(inode); @@ -2646,7 +2645,7 @@ static void handle_cap_export(struct inode *inode, struct ceph_mds_caps *ex, dout("handle_cap_export inode %p ci %p mds%d mseq %d\n", inode, ci, mds, mseq); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); /* make sure we haven't seen a higher mseq */ for (p = rb_first(&ci->i_caps); p; p = rb_next(p)) { @@ -2690,7 +2689,7 @@ static void handle_cap_export(struct inode *inode, struct ceph_mds_caps *ex, } /* else, we already released it */ - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } /* @@ -2745,9 +2744,9 @@ static void handle_cap_import(struct ceph_mds_client *mdsc, up_read(&mdsc->snap_rwsem); /* make sure we re-request max_size, if necessary */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_requested_max_size = 0; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } /* @@ -2762,6 +2761,7 @@ void ceph_handle_caps(struct ceph_mds_session *session, struct ceph_mds_client *mdsc = session->s_mdsc; struct super_block *sb = mdsc->fsc->sb; struct inode *inode; + struct ceph_inode_info *ci; struct ceph_cap *cap; struct ceph_mds_caps *h; int mds = session->s_mds; @@ -2815,6 +2815,7 @@ void ceph_handle_caps(struct ceph_mds_session *session, /* lookup ino */ inode = ceph_find_inode(sb, vino); + ci = ceph_inode(inode); dout(" op %s ino %llx.%llx inode %p\n", ceph_cap_op_name(op), vino.ino, vino.snap, inode); if (!inode) { @@ -2844,16 +2845,16 @@ void ceph_handle_caps(struct ceph_mds_session *session, } /* the rest require a cap */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap = __get_cap_for_mds(ceph_inode(inode), mds); if (!cap) { dout(" no cap on %p ino %llx.%llx from mds%d\n", inode, ceph_ino(inode), ceph_snap(inode), mds); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); goto flush_cap_releases; } - /* note that each of these drops i_lock for us */ + /* note that each of these drops i_ceph_lock for us */ switch (op) { case CEPH_CAP_OP_REVOKE: case CEPH_CAP_OP_GRANT: @@ -2869,7 +2870,7 @@ void ceph_handle_caps(struct ceph_mds_session *session, break; default: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); pr_err("ceph_handle_caps: unknown cap op %d %s\n", op, ceph_cap_op_name(op)); } @@ -2962,13 +2963,13 @@ void ceph_put_fmode(struct ceph_inode_info *ci, int fmode) struct inode *inode = &ci->vfs_inode; int last = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dout("put_fmode %p fmode %d %d -> %d\n", inode, fmode, ci->i_nr_by_mode[fmode], ci->i_nr_by_mode[fmode]-1); BUG_ON(ci->i_nr_by_mode[fmode] == 0); if (--ci->i_nr_by_mode[fmode] == 0) last++; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (last && ci->i_vino.snap == CEPH_NOSNAP) ceph_check_caps(ci, 0, NULL); @@ -2991,7 +2992,7 @@ int ceph_encode_inode_release(void **p, struct inode *inode, int used, dirty; int ret = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); used = __ceph_caps_used(ci); dirty = __ceph_caps_dirty(ci); @@ -3046,7 +3047,7 @@ int ceph_encode_inode_release(void **p, struct inode *inode, inode, cap, ceph_cap_string(cap->issued)); } } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return ret; } @@ -3061,7 +3062,7 @@ int ceph_encode_dentry_release(void **p, struct dentry *dentry, /* * force an record for the directory caps if we have a dentry lease. - * this is racy (can't take i_lock and d_lock together), but it + * this is racy (can't take i_ceph_lock and d_lock together), but it * doesn't have to be perfect; the mds will revoke anything we don't * release. */ diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index bca3948e9dbf..3eeb97661262 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -281,18 +281,18 @@ static int ceph_readdir(struct file *filp, void *dirent, filldir_t filldir) } /* can we use the dcache? */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if ((filp->f_pos == 2 || fi->dentry) && !ceph_test_mount_opt(fsc, NOASYNCREADDIR) && ceph_snap(inode) != CEPH_SNAPDIR && ceph_dir_test_complete(inode) && __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1)) { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); err = __dcache_readdir(filp, dirent, filldir); if (err != -EAGAIN) return err; } else { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } if (fi->dentry) { err = note_last_dentry(fi, fi->dentry->d_name.name, @@ -428,12 +428,12 @@ more: * were released during the whole readdir, and we should have * the complete dir contents in our cache. */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_release_count == fi->dir_release_count) { ceph_dir_set_complete(inode); ci->i_max_offset = filp->f_pos; } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("readdir %p filp %p done.\n", inode, filp); return 0; @@ -607,7 +607,7 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry, struct ceph_inode_info *ci = ceph_inode(dir); struct ceph_dentry_info *di = ceph_dentry(dentry); - spin_lock(&dir->i_lock); + spin_lock(&ci->i_ceph_lock); dout(" dir %p flags are %d\n", dir, ci->i_ceph_flags); if (strncmp(dentry->d_name.name, fsc->mount_options->snapdir_name, @@ -615,13 +615,13 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry, !is_root_ceph_dentry(dir, dentry) && ceph_dir_test_complete(dir) && (__ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1))) { - spin_unlock(&dir->i_lock); + spin_unlock(&ci->i_ceph_lock); dout(" dir %p complete, -ENOENT\n", dir); d_add(dentry, NULL); di->lease_shared_gen = ci->i_shared_gen; return NULL; } - spin_unlock(&dir->i_lock); + spin_unlock(&ci->i_ceph_lock); } op = ceph_snap(dir) == CEPH_SNAPDIR ? @@ -841,12 +841,12 @@ static int drop_caps_for_unlink(struct inode *inode) struct ceph_inode_info *ci = ceph_inode(inode); int drop = CEPH_CAP_LINK_SHARED | CEPH_CAP_LINK_EXCL; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (inode->i_nlink == 1) { drop |= ~(__ceph_caps_wanted(ci) | CEPH_CAP_PIN); ci->i_ceph_flags |= CEPH_I_NODELAY; } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return drop; } @@ -1015,10 +1015,10 @@ static int dir_lease_is_valid(struct inode *dir, struct dentry *dentry) struct ceph_dentry_info *di = ceph_dentry(dentry); int valid = 0; - spin_lock(&dir->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_shared_gen == di->lease_shared_gen) valid = __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1); - spin_unlock(&dir->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("dir_lease_is_valid dir %p v%u dentry %p v%u = %d\n", dir, (unsigned)ci->i_shared_gen, dentry, (unsigned)di->lease_shared_gen, valid); diff --git a/fs/ceph/file.c b/fs/ceph/file.c index ce549d31eeb7..ed72428d9c75 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -147,9 +147,9 @@ int ceph_open(struct inode *inode, struct file *file) /* trivially open snapdir */ if (ceph_snap(inode) == CEPH_SNAPDIR) { - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __ceph_get_fmode(ci, fmode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return ceph_init_file(inode, file, fmode); } @@ -158,7 +158,7 @@ int ceph_open(struct inode *inode, struct file *file) * write) or any MDS (for read). Update wanted set * asynchronously. */ - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (__ceph_is_any_real_caps(ci) && (((fmode & CEPH_FILE_MODE_WR) == 0) || ci->i_auth_cap)) { int mds_wanted = __ceph_caps_mds_wanted(ci); @@ -168,7 +168,7 @@ int ceph_open(struct inode *inode, struct file *file) inode, fmode, ceph_cap_string(wanted), ceph_cap_string(issued)); __ceph_get_fmode(ci, fmode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); /* adjust wanted? */ if ((issued & wanted) != wanted && @@ -180,10 +180,10 @@ int ceph_open(struct inode *inode, struct file *file) } else if (ceph_snap(inode) != CEPH_NOSNAP && (ci->i_snap_caps & wanted) == wanted) { __ceph_get_fmode(ci, fmode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return ceph_init_file(inode, file, fmode); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("open fmode %d wants %s\n", fmode, ceph_cap_string(wanted)); req = prepare_open_request(inode->i_sb, flags, 0); @@ -743,9 +743,9 @@ retry_snap: */ int dirty; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); ceph_put_cap_refs(ci, got); ret = generic_file_aio_write(iocb, iov, nr_segs, pos); @@ -764,9 +764,9 @@ retry_snap: if (ret >= 0) { int dirty; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (dirty) __mark_inode_dirty(inode, dirty); } @@ -797,7 +797,8 @@ static loff_t ceph_llseek(struct file *file, loff_t offset, int origin) mutex_lock(&inode->i_mutex); __ceph_do_pending_vmtruncate(inode); - if (origin != SEEK_CUR || origin != SEEK_SET) { + + if (origin == SEEK_END || origin == SEEK_DATA || origin == SEEK_HOLE) { ret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE); if (ret < 0) { offset = ret; diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index 116f36502f17..87fb132fb330 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -297,6 +297,8 @@ struct inode *ceph_alloc_inode(struct super_block *sb) dout("alloc_inode %p\n", &ci->vfs_inode); + spin_lock_init(&ci->i_ceph_lock); + ci->i_version = 0; ci->i_time_warp_seq = 0; ci->i_ceph_flags = 0; @@ -583,7 +585,7 @@ static int fill_inode(struct inode *inode, iinfo->xattr_len); } - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); /* * provided version will be odd if inode value is projected, @@ -680,7 +682,7 @@ static int fill_inode(struct inode *inode, char *sym; BUG_ON(symlen != inode->i_size); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); err = -ENOMEM; sym = kmalloc(symlen+1, GFP_NOFS); @@ -689,7 +691,7 @@ static int fill_inode(struct inode *inode, memcpy(sym, iinfo->symlink, symlen); sym[symlen] = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (!ci->i_symlink) ci->i_symlink = sym; else @@ -715,7 +717,7 @@ static int fill_inode(struct inode *inode, } no_change: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); /* queue truncate if we saw i_size decrease */ if (queue_trunc) @@ -750,13 +752,13 @@ no_change: info->cap.flags, caps_reservation); } else { - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dout(" %p got snap_caps %s\n", inode, ceph_cap_string(le32_to_cpu(info->cap.caps))); ci->i_snap_caps |= le32_to_cpu(info->cap.caps); if (cap_fmode >= 0) __ceph_get_fmode(ci, cap_fmode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } } else if (cap_fmode >= 0) { pr_warning("mds issued no caps on %llx.%llx\n", @@ -849,19 +851,20 @@ static void ceph_set_dentry_offset(struct dentry *dn) { struct dentry *dir = dn->d_parent; struct inode *inode = dir->d_inode; + struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_dentry_info *di; BUG_ON(!inode); di = ceph_dentry(dn); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (!ceph_dir_test_complete(inode)) { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return; } di->offset = ceph_inode(inode)->i_max_offset++; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); spin_lock(&dir->d_lock); spin_lock_nested(&dn->d_lock, DENTRY_D_LOCK_NESTED); @@ -1308,7 +1311,7 @@ int ceph_inode_set_size(struct inode *inode, loff_t size) struct ceph_inode_info *ci = ceph_inode(inode); int ret = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dout("set_size %p %llu -> %llu\n", inode, inode->i_size, size); inode->i_size = size; inode->i_blocks = (size + (1 << 9) - 1) >> 9; @@ -1318,7 +1321,7 @@ int ceph_inode_set_size(struct inode *inode, loff_t size) (ci->i_reported_size << 1) < ci->i_max_size) ret = 1; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return ret; } @@ -1376,20 +1379,20 @@ static void ceph_invalidate_work(struct work_struct *work) u32 orig_gen; int check = 0; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dout("invalidate_pages %p gen %d revoking %d\n", inode, ci->i_rdcache_gen, ci->i_rdcache_revoking); if (ci->i_rdcache_revoking != ci->i_rdcache_gen) { /* nevermind! */ - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); goto out; } orig_gen = ci->i_rdcache_gen; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); truncate_inode_pages(&inode->i_data, 0); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (orig_gen == ci->i_rdcache_gen && orig_gen == ci->i_rdcache_revoking) { dout("invalidate_pages %p gen %d successful\n", inode, @@ -1401,7 +1404,7 @@ static void ceph_invalidate_work(struct work_struct *work) inode, orig_gen, ci->i_rdcache_gen, ci->i_rdcache_revoking); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (check) ceph_check_caps(ci, 0, NULL); @@ -1460,10 +1463,10 @@ void __ceph_do_pending_vmtruncate(struct inode *inode) int wrbuffer_refs, wake = 0; retry: - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_truncate_pending == 0) { dout("__do_pending_vmtruncate %p none pending\n", inode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return; } @@ -1474,7 +1477,7 @@ retry: if (ci->i_wrbuffer_ref_head < ci->i_wrbuffer_ref) { dout("__do_pending_vmtruncate %p flushing snaps first\n", inode); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); filemap_write_and_wait_range(&inode->i_data, 0, inode->i_sb->s_maxbytes); goto retry; @@ -1484,15 +1487,15 @@ retry: wrbuffer_refs = ci->i_wrbuffer_ref; dout("__do_pending_vmtruncate %p (%d) to %lld\n", inode, ci->i_truncate_pending, to); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); truncate_inode_pages(inode->i_mapping, to); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_truncate_pending--; if (ci->i_truncate_pending == 0) wake = 1; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (wrbuffer_refs == 0) ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL); @@ -1547,7 +1550,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) if (IS_ERR(req)) return PTR_ERR(req); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); issued = __ceph_caps_issued(ci, NULL); dout("setattr %p issued %s\n", inode, ceph_cap_string(issued)); @@ -1695,7 +1698,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) } release &= issued; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (inode_dirty_flags) __mark_inode_dirty(inode, inode_dirty_flags); @@ -1717,7 +1720,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) __ceph_do_pending_vmtruncate(inode); return err; out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); ceph_mdsc_put_request(req); return err; } diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c index 5a14c29cbba6..790914a598dd 100644 --- a/fs/ceph/ioctl.c +++ b/fs/ceph/ioctl.c @@ -241,11 +241,11 @@ static long ceph_ioctl_lazyio(struct file *file) struct ceph_inode_info *ci = ceph_inode(inode); if ((fi->fmode & CEPH_FILE_MODE_LAZY) == 0) { - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_nr_by_mode[fi->fmode]--; fi->fmode |= CEPH_FILE_MODE_LAZY; ci->i_nr_by_mode[fi->fmode]++; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout("ioctl_layzio: file %p marked lazy\n", file); ceph_check_caps(ci, 0, NULL); diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 264ab701154f..6203d805eb45 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -732,21 +732,21 @@ static int __choose_mds(struct ceph_mds_client *mdsc, } } - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap = NULL; if (mode == USE_AUTH_MDS) cap = ci->i_auth_cap; if (!cap && !RB_EMPTY_ROOT(&ci->i_caps)) cap = rb_entry(rb_first(&ci->i_caps), struct ceph_cap, ci_node); if (!cap) { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); goto random; } mds = cap->session->s_mds; dout("choose_mds %p %llx.%llx mds%d (%scap %p)\n", inode, ceph_vinop(inode), mds, cap == ci->i_auth_cap ? "auth " : "", cap); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return mds; random: @@ -951,7 +951,7 @@ static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap, dout("removing cap %p, ci is %p, inode is %p\n", cap, ci, &ci->vfs_inode); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __ceph_remove_cap(cap); if (!__ceph_is_any_real_caps(ci)) { struct ceph_mds_client *mdsc = @@ -984,7 +984,7 @@ static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap, } spin_unlock(&mdsc->cap_dirty_lock); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); while (drop--) iput(inode); return 0; @@ -1015,10 +1015,10 @@ static int wake_up_session_cb(struct inode *inode, struct ceph_cap *cap, wake_up_all(&ci->i_cap_wq); if (arg) { - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_wanted_max_size = 0; ci->i_requested_max_size = 0; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } return 0; } @@ -1151,7 +1151,7 @@ static int trim_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg) if (session->s_trim_caps <= 0) return -1; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); mine = cap->issued | cap->implemented; used = __ceph_caps_used(ci); oissued = __ceph_caps_issued_other(ci, cap); @@ -1170,7 +1170,7 @@ static int trim_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg) __ceph_remove_cap(cap); } else { /* try to drop referring dentries */ - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); d_prune_aliases(inode); dout("trim_caps_cb %p cap %p pruned, count now %d\n", inode, cap, atomic_read(&inode->i_count)); @@ -1178,7 +1178,7 @@ static int trim_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg) } out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return 0; } @@ -1296,7 +1296,7 @@ static int check_cap_flush(struct ceph_mds_client *mdsc, u64 want_flush_seq) i_flushing_item); struct inode *inode = &ci->vfs_inode; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_cap_flush_seq <= want_flush_seq) { dout("check_cap_flush still flushing %p " "seq %lld <= %lld to mds%d\n", inode, @@ -1304,7 +1304,7 @@ static int check_cap_flush(struct ceph_mds_client *mdsc, u64 want_flush_seq) session->s_mds); ret = 0; } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } mutex_unlock(&session->s_mutex); ceph_put_mds_session(session); @@ -1495,6 +1495,7 @@ retry: pos, temp); } else if (stop_on_nosnap && inode && ceph_snap(inode) == CEPH_NOSNAP) { + spin_unlock(&temp->d_lock); break; } else { pos -= temp->d_name.len; @@ -2011,10 +2012,10 @@ void ceph_invalidate_dir_request(struct ceph_mds_request *req) struct ceph_inode_info *ci = ceph_inode(inode); dout("invalidate_dir_request %p (D_COMPLETE, lease(s))\n", inode); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ceph_dir_clear_complete(inode); ci->i_release_count++; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (req->r_dentry) ceph_invalidate_dentry_lease(req->r_dentry); @@ -2422,7 +2423,7 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap, if (err) goto out_free; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); cap->seq = 0; /* reset cap seq */ cap->issue_seq = 0; /* and issue_seq */ @@ -2445,7 +2446,7 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap, rec.v1.pathbase = cpu_to_le64(pathbase); reclen = sizeof(rec.v1); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (recon_state->flock) { int num_fcntl_locks, num_flock_locks; diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index 4bb239921dbd..a50ca0e39475 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -20,7 +20,7 @@ * * mdsc->snap_rwsem * - * inode->i_lock + * ci->i_ceph_lock * mdsc->snap_flush_lock * mdsc->cap_delay_lock * diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c index e26437191333..a559c80f127a 100644 --- a/fs/ceph/snap.c +++ b/fs/ceph/snap.c @@ -446,7 +446,7 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci) return; } - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); used = __ceph_caps_used(ci); dirty = __ceph_caps_dirty(ci); @@ -528,7 +528,7 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci) kfree(capsnap); } - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } /* @@ -537,7 +537,7 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci) * * If capsnap can now be flushed, add to snap_flush list, and return 1. * - * Caller must hold i_lock. + * Caller must hold i_ceph_lock. */ int __ceph_finish_cap_snap(struct ceph_inode_info *ci, struct ceph_cap_snap *capsnap) @@ -739,9 +739,9 @@ static void flush_snaps(struct ceph_mds_client *mdsc) inode = &ci->vfs_inode; ihold(inode); spin_unlock(&mdsc->snap_flush_lock); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __ceph_flush_snaps(ci, &session, 0); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); iput(inode); spin_lock(&mdsc->snap_flush_lock); } @@ -847,7 +847,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc, continue; ci = ceph_inode(inode); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (!ci->i_snap_realm) goto skip_inode; /* @@ -876,7 +876,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc, oldrealm = ci->i_snap_realm; ci->i_snap_realm = realm; spin_unlock(&realm->inodes_with_caps_lock); - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); ceph_get_snap_realm(mdsc, realm); ceph_put_snap_realm(mdsc, oldrealm); @@ -885,7 +885,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc, continue; skip_inode: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); iput(inode); } diff --git a/fs/ceph/super.c b/fs/ceph/super.c index 8dc73a594a90..b48f15f101a0 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -383,7 +383,7 @@ static int ceph_show_options(struct seq_file *m, struct vfsmount *mnt) if (fsopt->rsize != CEPH_RSIZE_DEFAULT) seq_printf(m, ",rsize=%d", fsopt->rsize); if (fsopt->rasize != CEPH_RASIZE_DEFAULT) - seq_printf(m, ",rasize=%d", fsopt->rsize); + seq_printf(m, ",rasize=%d", fsopt->rasize); if (fsopt->congestion_kb != default_congestion_kb()) seq_printf(m, ",write_congestion_kb=%d", fsopt->congestion_kb); if (fsopt->caps_wanted_delay_min != CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT) diff --git a/fs/ceph/super.h b/fs/ceph/super.h index 01bf189e08a9..edcbf3774a56 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -220,7 +220,7 @@ struct ceph_dentry_info { * The locking for D_COMPLETE is a bit odd: * - we can clear it at almost any time (see ceph_d_prune) * - it is only meaningful if: - * - we hold dir inode i_lock + * - we hold dir inode i_ceph_lock * - we hold dir FILE_SHARED caps * - the dentry D_COMPLETE is set */ @@ -250,6 +250,8 @@ struct ceph_inode_xattrs_info { struct ceph_inode_info { struct ceph_vino i_vino; /* ceph ino + snap */ + spinlock_t i_ceph_lock; + u64 i_version; u32 i_time_warp_seq; @@ -271,7 +273,7 @@ struct ceph_inode_info { struct ceph_inode_xattrs_info i_xattrs; - /* capabilities. protected _both_ by i_lock and cap->session's + /* capabilities. protected _both_ by i_ceph_lock and cap->session's * s_mutex. */ struct rb_root i_caps; /* cap list */ struct ceph_cap *i_auth_cap; /* authoritative cap, if any */ @@ -437,18 +439,18 @@ static inline void ceph_i_clear(struct inode *inode, unsigned mask) { struct ceph_inode_info *ci = ceph_inode(inode); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_ceph_flags &= ~mask; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } static inline void ceph_i_set(struct inode *inode, unsigned mask) { struct ceph_inode_info *ci = ceph_inode(inode); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); ci->i_ceph_flags |= mask; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); } static inline bool ceph_i_test(struct inode *inode, unsigned mask) @@ -456,9 +458,9 @@ static inline bool ceph_i_test(struct inode *inode, unsigned mask) struct ceph_inode_info *ci = ceph_inode(inode); bool r; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); r = (ci->i_ceph_flags & mask) == mask; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return r; } @@ -508,9 +510,9 @@ extern int __ceph_caps_issued_other(struct ceph_inode_info *ci, static inline int ceph_caps_issued(struct ceph_inode_info *ci) { int issued; - spin_lock(&ci->vfs_inode.i_lock); + spin_lock(&ci->i_ceph_lock); issued = __ceph_caps_issued(ci, NULL); - spin_unlock(&ci->vfs_inode.i_lock); + spin_unlock(&ci->i_ceph_lock); return issued; } @@ -518,9 +520,9 @@ static inline int ceph_caps_issued_mask(struct ceph_inode_info *ci, int mask, int touch) { int r; - spin_lock(&ci->vfs_inode.i_lock); + spin_lock(&ci->i_ceph_lock); r = __ceph_caps_issued_mask(ci, mask, touch); - spin_unlock(&ci->vfs_inode.i_lock); + spin_unlock(&ci->i_ceph_lock); return r; } @@ -743,10 +745,9 @@ extern int ceph_add_cap(struct inode *inode, extern void __ceph_remove_cap(struct ceph_cap *cap); static inline void ceph_remove_cap(struct ceph_cap *cap) { - struct inode *inode = &cap->ci->vfs_inode; - spin_lock(&inode->i_lock); + spin_lock(&cap->ci->i_ceph_lock); __ceph_remove_cap(cap); - spin_unlock(&inode->i_lock); + spin_unlock(&cap->ci->i_ceph_lock); } extern void ceph_put_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap); diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c index 96c6739a0280..a5e36e4488a7 100644 --- a/fs/ceph/xattr.c +++ b/fs/ceph/xattr.c @@ -343,8 +343,8 @@ void __ceph_destroy_xattrs(struct ceph_inode_info *ci) } static int __build_xattrs(struct inode *inode) - __releases(inode->i_lock) - __acquires(inode->i_lock) + __releases(ci->i_ceph_lock) + __acquires(ci->i_ceph_lock) { u32 namelen; u32 numattr = 0; @@ -372,7 +372,7 @@ start: end = p + ci->i_xattrs.blob->vec.iov_len; ceph_decode_32_safe(&p, end, numattr, bad); xattr_version = ci->i_xattrs.version; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); xattrs = kcalloc(numattr, sizeof(struct ceph_xattr *), GFP_NOFS); @@ -387,7 +387,7 @@ start: goto bad_lock; } - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_xattrs.version != xattr_version) { /* lost a race, retry */ for (i = 0; i < numattr; i++) @@ -418,7 +418,7 @@ start: return err; bad_lock: - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); bad: if (xattrs) { for (i = 0; i < numattr; i++) @@ -512,7 +512,7 @@ ssize_t ceph_getxattr(struct dentry *dentry, const char *name, void *value, if (vxattrs) vxattr = ceph_match_vxattr(vxattrs, name); - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dout("getxattr %p ver=%lld index_ver=%lld\n", inode, ci->i_xattrs.version, ci->i_xattrs.index_version); @@ -520,14 +520,14 @@ ssize_t ceph_getxattr(struct dentry *dentry, const char *name, void *value, (ci->i_xattrs.index_version >= ci->i_xattrs.version)) { goto get_xattr; } else { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); /* get xattrs from mds (if we don't already have them) */ err = ceph_do_getattr(inode, CEPH_STAT_CAP_XATTR); if (err) return err; } - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (vxattr && vxattr->readonly) { err = vxattr->getxattr_cb(ci, value, size); @@ -558,7 +558,7 @@ get_xattr: memcpy(value, xattr->val, xattr->val_len); out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return err; } @@ -573,7 +573,7 @@ ssize_t ceph_listxattr(struct dentry *dentry, char *names, size_t size) u32 len; int i; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); dout("listxattr %p ver=%lld index_ver=%lld\n", inode, ci->i_xattrs.version, ci->i_xattrs.index_version); @@ -581,13 +581,13 @@ ssize_t ceph_listxattr(struct dentry *dentry, char *names, size_t size) (ci->i_xattrs.index_version >= ci->i_xattrs.version)) { goto list_xattr; } else { - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); err = ceph_do_getattr(inode, CEPH_STAT_CAP_XATTR); if (err) return err; } - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); err = __build_xattrs(inode); if (err < 0) @@ -619,7 +619,7 @@ list_xattr: } out: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); return err; } @@ -739,7 +739,7 @@ int ceph_setxattr(struct dentry *dentry, const char *name, if (!xattr) goto out; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); retry: issued = __ceph_caps_issued(ci, NULL); if (!(issued & CEPH_CAP_XATTR_EXCL)) @@ -752,12 +752,12 @@ retry: required_blob_size > ci->i_xattrs.prealloc_blob->alloc_len) { struct ceph_buffer *blob = NULL; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); dout(" preaallocating new blob size=%d\n", required_blob_size); blob = ceph_buffer_new(required_blob_size, GFP_NOFS); if (!blob) goto out; - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); if (ci->i_xattrs.prealloc_blob) ceph_buffer_put(ci->i_xattrs.prealloc_blob); ci->i_xattrs.prealloc_blob = blob; @@ -770,13 +770,13 @@ retry: dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL); ci->i_xattrs.dirty = true; inode->i_ctime = CURRENT_TIME; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (dirty) __mark_inode_dirty(inode, dirty); return err; do_sync: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); err = ceph_sync_setxattr(dentry, name, value, size, flags); out: kfree(newname); @@ -833,7 +833,7 @@ int ceph_removexattr(struct dentry *dentry, const char *name) return -EOPNOTSUPP; } - spin_lock(&inode->i_lock); + spin_lock(&ci->i_ceph_lock); __build_xattrs(inode); issued = __ceph_caps_issued(ci, NULL); dout("removexattr %p issued %s\n", inode, ceph_cap_string(issued)); @@ -846,12 +846,12 @@ int ceph_removexattr(struct dentry *dentry, const char *name) ci->i_xattrs.dirty = true; inode->i_ctime = CURRENT_TIME; - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); if (dirty) __mark_inode_dirty(inode, dirty); return err; do_sync: - spin_unlock(&inode->i_lock); + spin_unlock(&ci->i_ceph_lock); err = ceph_send_removexattr(dentry, name); return err; } diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 73c3992b2bb4..ac86f8b3e3cb 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -156,6 +156,7 @@ __bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, * bdi_start_writeback - start writeback * @bdi: the backing device to write from * @nr_pages: the number of pages to write + * @reason: reason why some writeback work was initiated * * Description: * This does WB_SYNC_NONE opportunistic writeback. The IO is only @@ -1223,6 +1224,7 @@ static void wait_sb_inodes(struct super_block *sb) * writeback_inodes_sb_nr - writeback dirty inodes from given super_block * @sb: the superblock * @nr: the number of pages to write + * @reason: reason why some writeback work initiated * * Start writeback on some inodes on this super_block. No guarantees are made * on how many (if any) will be written, and this function does not wait @@ -1251,6 +1253,7 @@ EXPORT_SYMBOL(writeback_inodes_sb_nr); /** * writeback_inodes_sb - writeback dirty inodes from given super_block * @sb: the superblock + * @reason: reason why some writeback work was initiated * * Start writeback on some inodes on this super_block. No guarantees are made * on how many (if any) will be written, and this function does not wait @@ -1265,6 +1268,7 @@ EXPORT_SYMBOL(writeback_inodes_sb); /** * writeback_inodes_sb_if_idle - start writeback if none underway * @sb: the superblock + * @reason: reason why some writeback work was initiated * * Invoke writeback_inodes_sb if no writeback is currently underway. * Returns 1 if writeback was started, 0 if not. @@ -1285,6 +1289,7 @@ EXPORT_SYMBOL(writeback_inodes_sb_if_idle); * writeback_inodes_sb_if_idle - start writeback if none underway * @sb: the superblock * @nr: the number of pages to write + * @reason: reason why some writeback work was initiated * * Invoke writeback_inodes_sb if no writeback is currently underway. * Returns 1 if writeback was started, 0 if not. diff --git a/include/linux/log2.h b/include/linux/log2.h index 25b808631cd9..fd7ff3d91e6a 100644 --- a/include/linux/log2.h +++ b/include/linux/log2.h @@ -185,7 +185,6 @@ unsigned long __rounddown_pow_of_two(unsigned long n) #define rounddown_pow_of_two(n) \ ( \ __builtin_constant_p(n) ? ( \ - (n == 1) ? 0 : \ (1UL << ilog2(n))) : \ __rounddown_pow_of_two(n) \ ) diff --git a/include/linux/mmc/card.h b/include/linux/mmc/card.h index 415f2db414e1..c8ef9bc54d50 100644 --- a/include/linux/mmc/card.h +++ b/include/linux/mmc/card.h @@ -218,6 +218,7 @@ struct mmc_card { #define MMC_QUIRK_INAND_CMD38 (1<<6) /* iNAND devices have broken CMD38 */ #define MMC_QUIRK_BLK_NO_CMD23 (1<<7) /* Avoid CMD23 for regular multiblock */ #define MMC_QUIRK_BROKEN_BYTE_MODE_512 (1<<8) /* Avoid sending 512 bytes in */ +#define MMC_QUIRK_LONG_READ_TIME (1<<9) /* Data read time > CSD says */ /* byte mode */ unsigned int poweroff_notify_state; /* eMMC4.5 notify feature */ #define MMC_NO_POWER_NOTIFICATION 0 @@ -433,6 +434,11 @@ static inline int mmc_card_broken_byte_mode_512(const struct mmc_card *c) return c->quirks & MMC_QUIRK_BROKEN_BYTE_MODE_512; } +static inline int mmc_card_long_read_time(const struct mmc_card *c) +{ + return c->quirks & MMC_QUIRK_LONG_READ_TIME; +} + #define mmc_card_name(c) ((c)->cid.prod_name) #define mmc_card_id(c) (dev_name(&(c)->dev)) diff --git a/mm/filemap.c b/mm/filemap.c index c0018f2d50e0..c106d3b3cc64 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -2407,7 +2407,6 @@ static ssize_t generic_perform_write(struct file *file, iov_iter_count(i)); again: - /* * Bring in the user page that we will copy from _first_. * Otherwise there's a nasty deadlock on copying from the @@ -2463,7 +2462,10 @@ again: written += copied; balance_dirty_pages_ratelimited(mapping); - + if (fatal_signal_pending(current)) { + status = -EINTR; + break; + } } while (iov_iter_count(i)); return written ? written : status; diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 71252486bc6f..50f08241f981 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -411,8 +411,13 @@ void global_dirty_limits(unsigned long *pbackground, unsigned long *pdirty) * * Returns @bdi's dirty limit in pages. The term "dirty" in the context of * dirty balancing includes all PG_dirty, PG_writeback and NFS unstable pages. - * And the "limit" in the name is not seriously taken as hard limit in - * balance_dirty_pages(). + * + * Note that balance_dirty_pages() will only seriously take it as a hard limit + * when sleeping max_pause per page is not enough to keep the dirty pages under + * control. For example, when the device is completely stalled due to some error + * conditions, or when there are 1000 dd tasks writing to a slow 10MB/s USB key. + * In the other normal situations, it acts more gently by throttling the tasks + * more (rather than completely block them) when the bdi dirty pages go high. * * It allocates high/low dirty limits to fast/slow devices, in order to prevent * - starving fast devices @@ -594,6 +599,13 @@ static unsigned long bdi_position_ratio(struct backing_dev_info *bdi, */ if (unlikely(bdi_thresh > thresh)) bdi_thresh = thresh; + /* + * It's very possible that bdi_thresh is close to 0 not because the + * device is slow, but that it has remained inactive for long time. + * Honour such devices a reasonable good (hopefully IO efficient) + * threshold, so that the occasional writes won't be blocked and active + * writes can rampup the threshold quickly. + */ bdi_thresh = max(bdi_thresh, (limit - dirty) / 8); /* * scale global setpoint to bdi's: @@ -977,8 +989,7 @@ static unsigned long bdi_max_pause(struct backing_dev_info *bdi, * * 8 serves as the safety ratio. */ - if (bdi_dirty) - t = min(t, bdi_dirty * HZ / (8 * bw + 1)); + t = min(t, bdi_dirty * HZ / (8 * bw + 1)); /* * The pause time will be settled within range (max_pause/4, max_pause). @@ -1136,6 +1147,19 @@ pause: if (task_ratelimit) break; + /* + * In the case of an unresponding NFS server and the NFS dirty + * pages exceeds dirty_thresh, give the other good bdi's a pipe + * to go through, so that tasks on them still remain responsive. + * + * In theory 1 page is enough to keep the comsumer-producer + * pipe going: the flusher cleans 1 page => the task dirties 1 + * more page. However bdi_dirty has accounting errors. So use + * the larger and more IO friendly bdi_stat_error. + */ + if (bdi_dirty <= bdi_stat_error(bdi)) + break; + if (fatal_signal_pending(current)) break; } diff --git a/net/ceph/crush/mapper.c b/net/ceph/crush/mapper.c index 42599e31dcad..3a94eae7abe9 100644 --- a/net/ceph/crush/mapper.c +++ b/net/ceph/crush/mapper.c @@ -477,7 +477,6 @@ int crush_do_rule(struct crush_map *map, int i, j; int numrep; int firstn; - int rc = -1; BUG_ON(ruleno >= map->max_rules); @@ -491,23 +490,18 @@ int crush_do_rule(struct crush_map *map, * that this may or may not correspond to the specific types * referenced by the crush rule. */ - if (force >= 0) { - if (force >= map->max_devices || - map->device_parents[force] == 0) { - /*dprintk("CRUSH: forcefed device dne\n");*/ - rc = -1; /* force fed device dne */ - goto out; - } - if (!is_out(map, weight, force, x)) { - while (1) { - force_context[++force_pos] = force; - if (force >= 0) - force = map->device_parents[force]; - else - force = map->bucket_parents[-1-force]; - if (force == 0) - break; - } + if (force >= 0 && + force < map->max_devices && + map->device_parents[force] != 0 && + !is_out(map, weight, force, x)) { + while (1) { + force_context[++force_pos] = force; + if (force >= 0) + force = map->device_parents[force]; + else + force = map->bucket_parents[-1-force]; + if (force == 0) + break; } } @@ -600,10 +594,7 @@ int crush_do_rule(struct crush_map *map, BUG_ON(1); } } - rc = result_len; - -out: - return rc; + return result_len; } |