diff options
56 files changed, 952 insertions, 344 deletions
diff --git a/Documentation/ABI/testing/sysfs-block b/Documentation/ABI/testing/sysfs-block index 4873c759d535..c1eb41cb9876 100644 --- a/Documentation/ABI/testing/sysfs-block +++ b/Documentation/ABI/testing/sysfs-block @@ -142,3 +142,67 @@ Description: with the previous I/O request are enabled. When set to 2, all merge tries are disabled. The default value is 0 - which enables all types of merge tries. + +What: /sys/block/<disk>/discard_alignment +Date: May 2011 +Contact: Martin K. Petersen <martin.petersen@oracle.com> +Description: + Devices that support discard functionality may + internally allocate space in units that are bigger than + the exported logical block size. The discard_alignment + parameter indicates how many bytes the beginning of the + device is offset from the internal allocation unit's + natural alignment. + +What: /sys/block/<disk>/<partition>/discard_alignment +Date: May 2011 +Contact: Martin K. Petersen <martin.petersen@oracle.com> +Description: + Devices that support discard functionality may + internally allocate space in units that are bigger than + the exported logical block size. The discard_alignment + parameter indicates how many bytes the beginning of the + partition is offset from the internal allocation unit's + natural alignment. + +What: /sys/block/<disk>/queue/discard_granularity +Date: May 2011 +Contact: Martin K. Petersen <martin.petersen@oracle.com> +Description: + Devices that support discard functionality may + internally allocate space using units that are bigger + than the logical block size. The discard_granularity + parameter indicates the size of the internal allocation + unit in bytes if reported by the device. Otherwise the + discard_granularity will be set to match the device's + physical block size. A discard_granularity of 0 means + that the device does not support discard functionality. + +What: /sys/block/<disk>/queue/discard_max_bytes +Date: May 2011 +Contact: Martin K. Petersen <martin.petersen@oracle.com> +Description: + Devices that support discard functionality may have + internal limits on the number of bytes that can be + trimmed or unmapped in a single operation. Some storage + protocols also have inherent limits on the number of + blocks that can be described in a single command. The + discard_max_bytes parameter is set by the device driver + to the maximum number of bytes that can be discarded in + a single operation. Discard requests issued to the + device must not exceed this limit. A discard_max_bytes + value of 0 means that the device does not support + discard functionality. + +What: /sys/block/<disk>/queue/discard_zeroes_data +Date: May 2011 +Contact: Martin K. Petersen <martin.petersen@oracle.com> +Description: + Devices that support discard functionality may return + stale or random data when a previously discarded block + is read back. This can cause problems if the filesystem + expects discarded blocks to be explicitly cleared. If a + device reports that it deterministically returns zeroes + when a discarded area is read the discard_zeroes_data + parameter will be set to one. Otherwise it will be 0 and + the result of reading a discarded area is undefined. @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 39 -EXTRAVERSION = -rc7 +EXTRAVERSION = NAME = Flesh-Eating Bats with Fangs # *DOCUMENTATION* diff --git a/arch/mips/ar7/gpio.c b/arch/mips/ar7/gpio.c index 425dfa5d6e12..bb571bcdb8f2 100644 --- a/arch/mips/ar7/gpio.c +++ b/arch/mips/ar7/gpio.c @@ -325,9 +325,7 @@ int __init ar7_gpio_init(void) size = 0x1f; } - gpch->regs = ioremap_nocache(AR7_REGS_GPIO, - AR7_REGS_GPIO + 0x10); - + gpch->regs = ioremap_nocache(AR7_REGS_GPIO, size); if (!gpch->regs) { printk(KERN_ERR "%s: failed to ioremap regs\n", gpch->chip.label); diff --git a/arch/mips/include/asm/dma-mapping.h b/arch/mips/include/asm/dma-mapping.h index 655f849bd08d..7aa37ddfca4b 100644 --- a/arch/mips/include/asm/dma-mapping.h +++ b/arch/mips/include/asm/dma-mapping.h @@ -5,7 +5,9 @@ #include <asm/cache.h> #include <asm-generic/dma-coherent.h> +#ifndef CONFIG_SGI_IP27 /* Kludge to fix 2.6.39 build for IP27 */ #include <dma-coherence.h> +#endif extern struct dma_map_ops *mips_dma_map_ops; diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c index 71350f7f2d88..e9b3af27d844 100644 --- a/arch/mips/kernel/traps.c +++ b/arch/mips/kernel/traps.c @@ -374,7 +374,8 @@ void __noreturn die(const char *str, struct pt_regs *regs) unsigned long dvpret = dvpe(); #endif /* CONFIG_MIPS_MT_SMTC */ - notify_die(DIE_OOPS, str, regs, 0, regs_to_trapnr(regs), SIGSEGV); + if (notify_die(DIE_OOPS, str, regs, 0, regs_to_trapnr(regs), SIGSEGV) == NOTIFY_STOP) + sig = 0; console_verbose(); spin_lock_irq(&die_lock); @@ -383,9 +384,6 @@ void __noreturn die(const char *str, struct pt_regs *regs) mips_mt_regdump(dvpret); #endif /* CONFIG_MIPS_MT_SMTC */ - if (notify_die(DIE_OOPS, str, regs, 0, regs_to_trapnr(regs), SIGSEGV) == NOTIFY_STOP) - sig = 0; - printk("%s[#%d]:\n", str, ++die_counter); show_registers(regs); add_taint(TAINT_DIE); diff --git a/arch/mips/rb532/gpio.c b/arch/mips/rb532/gpio.c index 37de05d595e7..6c47dfeb7be3 100644 --- a/arch/mips/rb532/gpio.c +++ b/arch/mips/rb532/gpio.c @@ -185,7 +185,7 @@ int __init rb532_gpio_init(void) struct resource *r; r = rb532_gpio_reg0_res; - rb532_gpio_chip->regbase = ioremap_nocache(r->start, r->end - r->start); + rb532_gpio_chip->regbase = ioremap_nocache(r->start, resource_size(r)); if (!rb532_gpio_chip->regbase) { printk(KERN_ERR "rb532: cannot remap GPIO register 0\n"); diff --git a/arch/powerpc/platforms/83xx/suspend.c b/arch/powerpc/platforms/83xx/suspend.c index 188272934cfb..104faa8aa23c 100644 --- a/arch/powerpc/platforms/83xx/suspend.c +++ b/arch/powerpc/platforms/83xx/suspend.c @@ -318,17 +318,20 @@ static const struct platform_suspend_ops mpc83xx_suspend_ops = { .end = mpc83xx_suspend_end, }; +static struct of_device_id pmc_match[]; static int pmc_probe(struct platform_device *ofdev) { + const struct of_device_id *match; struct device_node *np = ofdev->dev.of_node; struct resource res; struct pmc_type *type; int ret = 0; - if (!ofdev->dev.of_match) + match = of_match_device(pmc_match, &ofdev->dev); + if (!match) return -EINVAL; - type = ofdev->dev.of_match->data; + type = match->data; if (!of_device_is_available(np)) return -ENODEV; diff --git a/arch/powerpc/sysdev/fsl_msi.c b/arch/powerpc/sysdev/fsl_msi.c index d5679dc1e20f..01cd2f089512 100644 --- a/arch/powerpc/sysdev/fsl_msi.c +++ b/arch/powerpc/sysdev/fsl_msi.c @@ -304,8 +304,10 @@ static int __devinit fsl_msi_setup_hwirq(struct fsl_msi *msi, return 0; } +static const struct of_device_id fsl_of_msi_ids[]; static int __devinit fsl_of_msi_probe(struct platform_device *dev) { + const struct of_device_id *match; struct fsl_msi *msi; struct resource res; int err, i, j, irq_index, count; @@ -316,9 +318,10 @@ static int __devinit fsl_of_msi_probe(struct platform_device *dev) u32 offset; static const u32 all_avail[] = { 0, NR_MSI_IRQS }; - if (!dev->dev.of_match) + match = of_match_device(fsl_of_msi_ids, &dev->dev); + if (!match) return -EINVAL; - features = dev->dev.of_match->data; + features = match->data; printk(KERN_DEBUG "Setting up Freescale MSI support\n"); diff --git a/arch/sparc/kernel/pci_sabre.c b/arch/sparc/kernel/pci_sabre.c index 948068a083fc..d1840dbdaa2f 100644 --- a/arch/sparc/kernel/pci_sabre.c +++ b/arch/sparc/kernel/pci_sabre.c @@ -452,8 +452,10 @@ static void __devinit sabre_pbm_init(struct pci_pbm_info *pbm, sabre_scan_bus(pbm, &op->dev); } +static const struct of_device_id sabre_match[]; static int __devinit sabre_probe(struct platform_device *op) { + const struct of_device_id *match; const struct linux_prom64_registers *pr_regs; struct device_node *dp = op->dev.of_node; struct pci_pbm_info *pbm; @@ -463,7 +465,8 @@ static int __devinit sabre_probe(struct platform_device *op) const u32 *vdma; u64 clear_irq; - hummingbird_p = op->dev.of_match && (op->dev.of_match->data != NULL); + match = of_match_device(sabre_match, &op->dev); + hummingbird_p = match && (match->data != NULL); if (!hummingbird_p) { struct device_node *cpu_dp; diff --git a/arch/sparc/kernel/pci_schizo.c b/arch/sparc/kernel/pci_schizo.c index fecfcb2063c8..283fbc329a43 100644 --- a/arch/sparc/kernel/pci_schizo.c +++ b/arch/sparc/kernel/pci_schizo.c @@ -1458,11 +1458,15 @@ out_err: return err; } +static const struct of_device_id schizo_match[]; static int __devinit schizo_probe(struct platform_device *op) { - if (!op->dev.of_match) + const struct of_device_id *match; + + match = of_match_device(schizo_match, &op->dev); + if (!match) return -EINVAL; - return __schizo_init(op, (unsigned long) op->dev.of_match->data); + return __schizo_init(op, (unsigned long)match->data); } /* The ordering of this table is very important. Some Tomatillo diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index 471fdcc5df85..e41cc6f2ccc1 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -385,25 +385,40 @@ void blkiocg_update_timeslice_used(struct blkio_group *blkg, unsigned long time, spin_lock_irqsave(&blkg->stats_lock, flags); blkg->stats.time += time; +#ifdef CONFIG_DEBUG_BLK_CGROUP blkg->stats.unaccounted_time += unaccounted_time; +#endif spin_unlock_irqrestore(&blkg->stats_lock, flags); } EXPORT_SYMBOL_GPL(blkiocg_update_timeslice_used); +/* + * should be called under rcu read lock or queue lock to make sure blkg pointer + * is valid. + */ void blkiocg_update_dispatch_stats(struct blkio_group *blkg, uint64_t bytes, bool direction, bool sync) { - struct blkio_group_stats *stats; + struct blkio_group_stats_cpu *stats_cpu; unsigned long flags; - spin_lock_irqsave(&blkg->stats_lock, flags); - stats = &blkg->stats; - stats->sectors += bytes >> 9; - blkio_add_stat(stats->stat_arr[BLKIO_STAT_SERVICED], 1, direction, - sync); - blkio_add_stat(stats->stat_arr[BLKIO_STAT_SERVICE_BYTES], bytes, - direction, sync); - spin_unlock_irqrestore(&blkg->stats_lock, flags); + /* + * Disabling interrupts to provide mutual exclusion between two + * writes on same cpu. It probably is not needed for 64bit. Not + * optimizing that case yet. + */ + local_irq_save(flags); + + stats_cpu = this_cpu_ptr(blkg->stats_cpu); + + u64_stats_update_begin(&stats_cpu->syncp); + stats_cpu->sectors += bytes >> 9; + blkio_add_stat(stats_cpu->stat_arr_cpu[BLKIO_STAT_CPU_SERVICED], + 1, direction, sync); + blkio_add_stat(stats_cpu->stat_arr_cpu[BLKIO_STAT_CPU_SERVICE_BYTES], + bytes, direction, sync); + u64_stats_update_end(&stats_cpu->syncp); + local_irq_restore(flags); } EXPORT_SYMBOL_GPL(blkiocg_update_dispatch_stats); @@ -438,6 +453,20 @@ void blkiocg_update_io_merged_stats(struct blkio_group *blkg, bool direction, } EXPORT_SYMBOL_GPL(blkiocg_update_io_merged_stats); +/* + * This function allocates the per cpu stats for blkio_group. Should be called + * from sleepable context as alloc_per_cpu() requires that. + */ +int blkio_alloc_blkg_stats(struct blkio_group *blkg) +{ + /* Allocate memory for per cpu stats */ + blkg->stats_cpu = alloc_percpu(struct blkio_group_stats_cpu); + if (!blkg->stats_cpu) + return -ENOMEM; + return 0; +} +EXPORT_SYMBOL_GPL(blkio_alloc_blkg_stats); + void blkiocg_add_blkio_group(struct blkio_cgroup *blkcg, struct blkio_group *blkg, void *key, dev_t dev, enum blkio_policy_id plid) @@ -508,6 +537,30 @@ struct blkio_group *blkiocg_lookup_group(struct blkio_cgroup *blkcg, void *key) } EXPORT_SYMBOL_GPL(blkiocg_lookup_group); +static void blkio_reset_stats_cpu(struct blkio_group *blkg) +{ + struct blkio_group_stats_cpu *stats_cpu; + int i, j, k; + /* + * Note: On 64 bit arch this should not be an issue. This has the + * possibility of returning some inconsistent value on 32bit arch + * as 64bit update on 32bit is non atomic. Taking care of this + * corner case makes code very complicated, like sending IPIs to + * cpus, taking care of stats of offline cpus etc. + * + * reset stats is anyway more of a debug feature and this sounds a + * corner case. So I am not complicating the code yet until and + * unless this becomes a real issue. + */ + for_each_possible_cpu(i) { + stats_cpu = per_cpu_ptr(blkg->stats_cpu, i); + stats_cpu->sectors = 0; + for(j = 0; j < BLKIO_STAT_CPU_NR; j++) + for (k = 0; k < BLKIO_STAT_TOTAL; k++) + stats_cpu->stat_arr_cpu[j][k] = 0; + } +} + static int blkiocg_reset_stats(struct cgroup *cgroup, struct cftype *cftype, u64 val) { @@ -552,7 +605,11 @@ blkiocg_reset_stats(struct cgroup *cgroup, struct cftype *cftype, u64 val) } #endif spin_unlock(&blkg->stats_lock); + + /* Reset Per cpu stats which don't take blkg->stats_lock */ + blkio_reset_stats_cpu(blkg); } + spin_unlock_irq(&blkcg->lock); return 0; } @@ -598,6 +655,59 @@ static uint64_t blkio_fill_stat(char *str, int chars_left, uint64_t val, return val; } + +static uint64_t blkio_read_stat_cpu(struct blkio_group *blkg, + enum stat_type_cpu type, enum stat_sub_type sub_type) +{ + int cpu; + struct blkio_group_stats_cpu *stats_cpu; + u64 val = 0, tval; + + for_each_possible_cpu(cpu) { + unsigned int start; + stats_cpu = per_cpu_ptr(blkg->stats_cpu, cpu); + + do { + start = u64_stats_fetch_begin(&stats_cpu->syncp); + if (type == BLKIO_STAT_CPU_SECTORS) + tval = stats_cpu->sectors; + else + tval = stats_cpu->stat_arr_cpu[type][sub_type]; + } while(u64_stats_fetch_retry(&stats_cpu->syncp, start)); + + val += tval; + } + + return val; +} + +static uint64_t blkio_get_stat_cpu(struct blkio_group *blkg, + struct cgroup_map_cb *cb, dev_t dev, enum stat_type_cpu type) +{ + uint64_t disk_total, val; + char key_str[MAX_KEY_LEN]; + enum stat_sub_type sub_type; + + if (type == BLKIO_STAT_CPU_SECTORS) { + val = blkio_read_stat_cpu(blkg, type, 0); + return blkio_fill_stat(key_str, MAX_KEY_LEN - 1, val, cb, dev); + } + + for (sub_type = BLKIO_STAT_READ; sub_type < BLKIO_STAT_TOTAL; + sub_type++) { + blkio_get_key_name(sub_type, dev, key_str, MAX_KEY_LEN, false); + val = blkio_read_stat_cpu(blkg, type, sub_type); + cb->fill(cb, key_str, val); + } + + disk_total = blkio_read_stat_cpu(blkg, type, BLKIO_STAT_READ) + + blkio_read_stat_cpu(blkg, type, BLKIO_STAT_WRITE); + + blkio_get_key_name(BLKIO_STAT_TOTAL, dev, key_str, MAX_KEY_LEN, false); + cb->fill(cb, key_str, disk_total); + return disk_total; +} + /* This should be called with blkg->stats_lock held */ static uint64_t blkio_get_stat(struct blkio_group *blkg, struct cgroup_map_cb *cb, dev_t dev, enum stat_type type) @@ -609,9 +719,6 @@ static uint64_t blkio_get_stat(struct blkio_group *blkg, if (type == BLKIO_STAT_TIME) return blkio_fill_stat(key_str, MAX_KEY_LEN - 1, blkg->stats.time, cb, dev); - if (type == BLKIO_STAT_SECTORS) - return blkio_fill_stat(key_str, MAX_KEY_LEN - 1, - blkg->stats.sectors, cb, dev); #ifdef CONFIG_DEBUG_BLK_CGROUP if (type == BLKIO_STAT_UNACCOUNTED_TIME) return blkio_fill_stat(key_str, MAX_KEY_LEN - 1, @@ -1075,8 +1182,8 @@ static int blkiocg_file_read(struct cgroup *cgrp, struct cftype *cft, } static int blkio_read_blkg_stats(struct blkio_cgroup *blkcg, - struct cftype *cft, struct cgroup_map_cb *cb, enum stat_type type, - bool show_total) + struct cftype *cft, struct cgroup_map_cb *cb, + enum stat_type type, bool show_total, bool pcpu) { struct blkio_group *blkg; struct hlist_node *n; @@ -1087,10 +1194,15 @@ static int blkio_read_blkg_stats(struct blkio_cgroup *blkcg, if (blkg->dev) { if (!cftype_blkg_same_policy(cft, blkg)) continue; - spin_lock_irq(&blkg->stats_lock); - cgroup_total += blkio_get_stat(blkg, cb, blkg->dev, - type); - spin_unlock_irq(&blkg->stats_lock); + if (pcpu) + cgroup_total += blkio_get_stat_cpu(blkg, cb, + blkg->dev, type); + else { + spin_lock_irq(&blkg->stats_lock); + cgroup_total += blkio_get_stat(blkg, cb, + blkg->dev, type); + spin_unlock_irq(&blkg->stats_lock); + } } } if (show_total) @@ -1114,47 +1226,47 @@ static int blkiocg_file_read_map(struct cgroup *cgrp, struct cftype *cft, switch(name) { case BLKIO_PROP_time: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_TIME, 0); + BLKIO_STAT_TIME, 0, 0); case BLKIO_PROP_sectors: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_SECTORS, 0); + BLKIO_STAT_CPU_SECTORS, 0, 1); case BLKIO_PROP_io_service_bytes: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_SERVICE_BYTES, 1); + BLKIO_STAT_CPU_SERVICE_BYTES, 1, 1); case BLKIO_PROP_io_serviced: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_SERVICED, 1); + BLKIO_STAT_CPU_SERVICED, 1, 1); case BLKIO_PROP_io_service_time: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_SERVICE_TIME, 1); + BLKIO_STAT_SERVICE_TIME, 1, 0); case BLKIO_PROP_io_wait_time: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_WAIT_TIME, 1); + BLKIO_STAT_WAIT_TIME, 1, 0); case BLKIO_PROP_io_merged: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_MERGED, 1); + BLKIO_STAT_MERGED, 1, 0); case BLKIO_PROP_io_queued: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_QUEUED, 1); + BLKIO_STAT_QUEUED, 1, 0); #ifdef CONFIG_DEBUG_BLK_CGROUP case BLKIO_PROP_unaccounted_time: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_UNACCOUNTED_TIME, 0); + BLKIO_STAT_UNACCOUNTED_TIME, 0, 0); case BLKIO_PROP_dequeue: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_DEQUEUE, 0); + BLKIO_STAT_DEQUEUE, 0, 0); case BLKIO_PROP_avg_queue_size: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_AVG_QUEUE_SIZE, 0); + BLKIO_STAT_AVG_QUEUE_SIZE, 0, 0); case BLKIO_PROP_group_wait_time: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_GROUP_WAIT_TIME, 0); + BLKIO_STAT_GROUP_WAIT_TIME, 0, 0); case BLKIO_PROP_idle_time: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_IDLE_TIME, 0); + BLKIO_STAT_IDLE_TIME, 0, 0); case BLKIO_PROP_empty_time: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_EMPTY_TIME, 0); + BLKIO_STAT_EMPTY_TIME, 0, 0); #endif default: BUG(); @@ -1164,10 +1276,10 @@ static int blkiocg_file_read_map(struct cgroup *cgrp, struct cftype *cft, switch(name){ case BLKIO_THROTL_io_service_bytes: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_SERVICE_BYTES, 1); + BLKIO_STAT_CPU_SERVICE_BYTES, 1, 1); case BLKIO_THROTL_io_serviced: return blkio_read_blkg_stats(blkcg, cft, cb, - BLKIO_STAT_SERVICED, 1); + BLKIO_STAT_CPU_SERVICED, 1, 1); default: BUG(); } diff --git a/block/blk-cgroup.h b/block/blk-cgroup.h index c774930cc206..262226798093 100644 --- a/block/blk-cgroup.h +++ b/block/blk-cgroup.h @@ -14,6 +14,7 @@ */ #include <linux/cgroup.h> +#include <linux/u64_stats_sync.h> enum blkio_policy_id { BLKIO_POLICY_PROP = 0, /* Proportional Bandwidth division */ @@ -36,10 +37,6 @@ enum stat_type { * request completion for IOs doen by this cgroup. This may not be * accurate when NCQ is turned on. */ BLKIO_STAT_SERVICE_TIME = 0, - /* Total bytes transferred */ - BLKIO_STAT_SERVICE_BYTES, - /* Total IOs serviced, post merge */ - BLKIO_STAT_SERVICED, /* Total time spent waiting in scheduler queue in ns */ BLKIO_STAT_WAIT_TIME, /* Number of IOs merged */ @@ -48,10 +45,9 @@ enum stat_type { BLKIO_STAT_QUEUED, /* All the single valued stats go below this */ BLKIO_STAT_TIME, - BLKIO_STAT_SECTORS, +#ifdef CONFIG_DEBUG_BLK_CGROUP /* Time not charged to this cgroup */ BLKIO_STAT_UNACCOUNTED_TIME, -#ifdef CONFIG_DEBUG_BLK_CGROUP BLKIO_STAT_AVG_QUEUE_SIZE, BLKIO_STAT_IDLE_TIME, BLKIO_STAT_EMPTY_TIME, @@ -60,6 +56,16 @@ enum stat_type { #endif }; +/* Per cpu stats */ +enum stat_type_cpu { + BLKIO_STAT_CPU_SECTORS, + /* Total bytes transferred */ + BLKIO_STAT_CPU_SERVICE_BYTES, + /* Total IOs serviced, post merge */ + BLKIO_STAT_CPU_SERVICED, + BLKIO_STAT_CPU_NR +}; + enum stat_sub_type { BLKIO_STAT_READ = 0, BLKIO_STAT_WRITE, @@ -116,11 +122,11 @@ struct blkio_cgroup { struct blkio_group_stats { /* total disk time and nr sectors dispatched by this group */ uint64_t time; - uint64_t sectors; - /* Time not charged to this cgroup */ - uint64_t unaccounted_time; uint64_t stat_arr[BLKIO_STAT_QUEUED + 1][BLKIO_STAT_TOTAL]; #ifdef CONFIG_DEBUG_BLK_CGROUP + /* Time not charged to this cgroup */ + uint64_t unaccounted_time; + /* Sum of number of IOs queued across all samples */ uint64_t avg_queue_size_sum; /* Count of samples taken for average */ @@ -145,6 +151,13 @@ struct blkio_group_stats { #endif }; +/* Per cpu blkio group stats */ +struct blkio_group_stats_cpu { + uint64_t sectors; + uint64_t stat_arr_cpu[BLKIO_STAT_CPU_NR][BLKIO_STAT_TOTAL]; + struct u64_stats_sync syncp; +}; + struct blkio_group { /* An rcu protected unique identifier for the group */ void *key; @@ -160,6 +173,8 @@ struct blkio_group { /* Need to serialize the stats in the case of reset/update */ spinlock_t stats_lock; struct blkio_group_stats stats; + /* Per cpu stats pointer */ + struct blkio_group_stats_cpu __percpu *stats_cpu; }; struct blkio_policy_node { @@ -295,6 +310,7 @@ extern struct blkio_cgroup *task_blkio_cgroup(struct task_struct *tsk); extern void blkiocg_add_blkio_group(struct blkio_cgroup *blkcg, struct blkio_group *blkg, void *key, dev_t dev, enum blkio_policy_id plid); +extern int blkio_alloc_blkg_stats(struct blkio_group *blkg); extern int blkiocg_del_blkio_group(struct blkio_group *blkg); extern struct blkio_group *blkiocg_lookup_group(struct blkio_cgroup *blkcg, void *key); @@ -322,6 +338,8 @@ static inline void blkiocg_add_blkio_group(struct blkio_cgroup *blkcg, struct blkio_group *blkg, void *key, dev_t dev, enum blkio_policy_id plid) {} +static inline int blkio_alloc_blkg_stats(struct blkio_group *blkg) { return 0; } + static inline int blkiocg_del_blkio_group(struct blkio_group *blkg) { return 0; } diff --git a/block/blk-core.c b/block/blk-core.c index 3fe00a14822a..9e8e297374b9 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -1550,7 +1550,8 @@ static inline void __generic_make_request(struct bio *bio) goto end_io; } - blk_throtl_bio(q, &bio); + if (blk_throtl_bio(q, &bio)) + goto end_io; /* * If bio = NULL, bio has been throttled and will be submitted diff --git a/block/blk-exec.c b/block/blk-exec.c index 81e31819a597..8a0e7ec056e7 100644 --- a/block/blk-exec.c +++ b/block/blk-exec.c @@ -56,7 +56,7 @@ void blk_execute_rq_nowait(struct request_queue *q, struct gendisk *bd_disk, spin_lock_irq(q->queue_lock); __elv_add_request(q, rq, where); __blk_run_queue(q); - /* the queue is stopped so it won't be plugged+unplugged */ + /* the queue is stopped so it won't be run */ if (rq->cmd_type == REQ_TYPE_PM_RESUME) q->request_fn(q); spin_unlock_irq(q->queue_lock); diff --git a/block/blk-flush.c b/block/blk-flush.c index 6c9b5e189e62..bb21e4c36f70 100644 --- a/block/blk-flush.c +++ b/block/blk-flush.c @@ -212,13 +212,19 @@ static void flush_end_io(struct request *flush_rq, int error) } /* - * Moving a request silently to empty queue_head may stall the - * queue. Kick the queue in those cases. This function is called - * from request completion path and calling directly into - * request_fn may confuse the driver. Always use kblockd. + * Kick the queue to avoid stall for two cases: + * 1. Moving a request silently to empty queue_head may stall the + * queue. + * 2. When flush request is running in non-queueable queue, the + * queue is hold. Restart the queue after flush request is finished + * to avoid stall. + * This function is called from request completion path and calling + * directly into request_fn may confuse the driver. Always use + * kblockd. */ - if (queued) + if (queued || q->flush_queue_delayed) blk_run_queue_async(q); + q->flush_queue_delayed = 0; } /** diff --git a/block/blk-lib.c b/block/blk-lib.c index 25de73e4759b..78e627e2581d 100644 --- a/block/blk-lib.c +++ b/block/blk-lib.c @@ -9,17 +9,20 @@ #include "blk.h" -static void blkdev_discard_end_io(struct bio *bio, int err) -{ - if (err) { - if (err == -EOPNOTSUPP) - set_bit(BIO_EOPNOTSUPP, &bio->bi_flags); - clear_bit(BIO_UPTODATE, &bio->bi_flags); - } +struct bio_batch { + atomic_t done; + unsigned long flags; + struct completion *wait; +}; - if (bio->bi_private) - complete(bio->bi_private); +static void bio_batch_end_io(struct bio *bio, int err) +{ + struct bio_batch *bb = bio->bi_private; + if (err && (err != -EOPNOTSUPP)) + clear_bit(BIO_UPTODATE, &bb->flags); + if (atomic_dec_and_test(&bb->done)) + complete(bb->wait); bio_put(bio); } @@ -41,6 +44,7 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector, struct request_queue *q = bdev_get_queue(bdev); int type = REQ_WRITE | REQ_DISCARD; unsigned int max_discard_sectors; + struct bio_batch bb; struct bio *bio; int ret = 0; @@ -67,7 +71,11 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector, type |= REQ_SECURE; } - while (nr_sects && !ret) { + atomic_set(&bb.done, 1); + bb.flags = 1 << BIO_UPTODATE; + bb.wait = &wait; + + while (nr_sects) { bio = bio_alloc(gfp_mask, 1); if (!bio) { ret = -ENOMEM; @@ -75,9 +83,9 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector, } bio->bi_sector = sector; - bio->bi_end_io = blkdev_discard_end_io; + bio->bi_end_io = bio_batch_end_io; bio->bi_bdev = bdev; - bio->bi_private = &wait; + bio->bi_private = &bb; if (nr_sects > max_discard_sectors) { bio->bi_size = max_discard_sectors << 9; @@ -88,45 +96,21 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector, nr_sects = 0; } - bio_get(bio); + atomic_inc(&bb.done); submit_bio(type, bio); + } + /* Wait for bios in-flight */ + if (!atomic_dec_and_test(&bb.done)) wait_for_completion(&wait); - if (bio_flagged(bio, BIO_EOPNOTSUPP)) - ret = -EOPNOTSUPP; - else if (!bio_flagged(bio, BIO_UPTODATE)) - ret = -EIO; - bio_put(bio); - } + if (!test_bit(BIO_UPTODATE, &bb.flags)) + ret = -EIO; return ret; } EXPORT_SYMBOL(blkdev_issue_discard); -struct bio_batch -{ - atomic_t done; - unsigned long flags; - struct completion *wait; -}; - -static void bio_batch_end_io(struct bio *bio, int err) -{ - struct bio_batch *bb = bio->bi_private; - - if (err) { - if (err == -EOPNOTSUPP) - set_bit(BIO_EOPNOTSUPP, &bb->flags); - else - clear_bit(BIO_UPTODATE, &bb->flags); - } - if (bb) - if (atomic_dec_and_test(&bb->done)) - complete(bb->wait); - bio_put(bio); -} - /** * blkdev_issue_zeroout - generate number of zero filed write bios * @bdev: blockdev to issue @@ -151,7 +135,6 @@ int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector, bb.flags = 1 << BIO_UPTODATE; bb.wait = &wait; -submit: ret = 0; while (nr_sects != 0) { bio = bio_alloc(gfp_mask, @@ -168,9 +151,6 @@ submit: while (nr_sects != 0) { sz = min((sector_t) PAGE_SIZE >> 9 , nr_sects); - if (sz == 0) - /* bio has maximum size possible */ - break; ret = bio_add_page(bio, ZERO_PAGE(0), sz << 9, 0); nr_sects -= ret >> 9; sector += ret >> 9; @@ -190,16 +170,6 @@ submit: /* One of bios in the batch was completed with error.*/ ret = -EIO; - if (ret) - goto out; - - if (test_bit(BIO_EOPNOTSUPP, &bb.flags)) { - ret = -EOPNOTSUPP; - goto out; - } - if (nr_sects != 0) - goto submit; -out: return ret; } EXPORT_SYMBOL(blkdev_issue_zeroout); diff --git a/block/blk-settings.c b/block/blk-settings.c index 1fa769293597..fa1eb0449a05 100644 --- a/block/blk-settings.c +++ b/block/blk-settings.c @@ -120,7 +120,7 @@ void blk_set_default_limits(struct queue_limits *lim) lim->discard_granularity = 0; lim->discard_alignment = 0; lim->discard_misaligned = 0; - lim->discard_zeroes_data = -1; + lim->discard_zeroes_data = 1; lim->logical_block_size = lim->physical_block_size = lim->io_min = 512; lim->bounce_pfn = (unsigned long)(BLK_BOUNCE_ANY >> PAGE_SHIFT); lim->alignment_offset = 0; @@ -166,6 +166,7 @@ void blk_queue_make_request(struct request_queue *q, make_request_fn *mfn) blk_set_default_limits(&q->limits); blk_queue_max_hw_sectors(q, BLK_SAFE_MAX_SECTORS); + q->limits.discard_zeroes_data = 0; /* * by default assume old behaviour and bounce for any highmem page @@ -790,6 +791,12 @@ void blk_queue_flush(struct request_queue *q, unsigned int flush) } EXPORT_SYMBOL_GPL(blk_queue_flush); +void blk_queue_flush_queueable(struct request_queue *q, bool queueable) +{ + q->flush_not_queueable = !queueable; +} +EXPORT_SYMBOL_GPL(blk_queue_flush_queueable); + static int __init blk_settings_init(void) { blk_max_low_pfn = max_low_pfn - 1; diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index bd236313f35d..d935bd859c87 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -152,7 +152,8 @@ static ssize_t queue_discard_granularity_show(struct request_queue *q, char *pag static ssize_t queue_discard_max_show(struct request_queue *q, char *page) { - return queue_var_show(q->limits.max_discard_sectors << 9, page); + return sprintf(page, "%llu\n", + (unsigned long long)q->limits.max_discard_sectors << 9); } static ssize_t queue_discard_zeroes_data_show(struct request_queue *q, char *page) diff --git a/block/blk-throttle.c b/block/blk-throttle.c index 252a81a306f7..a62be8d0dc1b 100644 --- a/block/blk-throttle.c +++ b/block/blk-throttle.c @@ -78,6 +78,8 @@ struct throtl_grp { /* Some throttle limits got updated for the group */ int limits_changed; + + struct rcu_head rcu_head; }; struct throtl_data @@ -88,7 +90,7 @@ struct throtl_data /* service tree for active throtl groups */ struct throtl_rb_root tg_service_tree; - struct throtl_grp root_tg; + struct throtl_grp *root_tg; struct request_queue *queue; /* Total Number of queued bios on READ and WRITE lists */ @@ -151,56 +153,44 @@ static inline struct throtl_grp *throtl_ref_get_tg(struct throtl_grp *tg) return tg; } -static void throtl_put_tg(struct throtl_grp *tg) +static void throtl_free_tg(struct rcu_head *head) { - BUG_ON(atomic_read(&tg->ref) <= 0); - if (!atomic_dec_and_test(&tg->ref)) - return; + struct throtl_grp *tg; + + tg = container_of(head, struct throtl_grp, rcu_head); + free_percpu(tg->blkg.stats_cpu); kfree(tg); } -static struct throtl_grp * throtl_find_alloc_tg(struct throtl_data *td, - struct blkio_cgroup *blkcg) +static void throtl_put_tg(struct throtl_grp *tg) { - struct throtl_grp *tg = NULL; - void *key = td; - struct backing_dev_info *bdi = &td->queue->backing_dev_info; - unsigned int major, minor; + BUG_ON(atomic_read(&tg->ref) <= 0); + if (!atomic_dec_and_test(&tg->ref)) + return; /* - * TODO: Speed up blkiocg_lookup_group() by maintaining a radix - * tree of blkg (instead of traversing through hash list all - * the time. + * A group is freed in rcu manner. But having an rcu lock does not + * mean that one can access all the fields of blkg and assume these + * are valid. For example, don't try to follow throtl_data and + * request queue links. + * + * Having a reference to blkg under an rcu allows acess to only + * values local to groups like group stats and group rate limits */ + call_rcu(&tg->rcu_head, throtl_free_tg); +} - /* - * This is the common case when there are no blkio cgroups. - * Avoid lookup in this case - */ - if (blkcg == &blkio_root_cgroup) - tg = &td->root_tg; - else - tg = tg_of_blkg(blkiocg_lookup_group(blkcg, key)); - - /* Fill in device details for root group */ - if (tg && !tg->blkg.dev && bdi->dev && dev_name(bdi->dev)) { - sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor); - tg->blkg.dev = MKDEV(major, minor); - goto done; - } - - if (tg) - goto done; - - tg = kzalloc_node(sizeof(*tg), GFP_ATOMIC, td->queue->node); - if (!tg) - goto done; - +static void throtl_init_group(struct throtl_grp *tg) +{ INIT_HLIST_NODE(&tg->tg_node); RB_CLEAR_NODE(&tg->rb_node); bio_list_init(&tg->bio_lists[0]); bio_list_init(&tg->bio_lists[1]); - td->limits_changed = false; + tg->limits_changed = false; + + /* Practically unlimited BW */ + tg->bps[0] = tg->bps[1] = -1; + tg->iops[0] = tg->iops[1] = -1; /* * Take the initial reference that will be released on destroy @@ -209,33 +199,181 @@ static struct throtl_grp * throtl_find_alloc_tg(struct throtl_data *td, * exit or cgroup deletion path depending on who is exiting first. */ atomic_set(&tg->ref, 1); +} + +/* Should be called with rcu read lock held (needed for blkcg) */ +static void +throtl_add_group_to_td_list(struct throtl_data *td, struct throtl_grp *tg) +{ + hlist_add_head(&tg->tg_node, &td->tg_list); + td->nr_undestroyed_grps++; +} + +static void +__throtl_tg_fill_dev_details(struct throtl_data *td, struct throtl_grp *tg) +{ + struct backing_dev_info *bdi = &td->queue->backing_dev_info; + unsigned int major, minor; + + if (!tg || tg->blkg.dev) + return; + + /* + * Fill in device details for a group which might not have been + * filled at group creation time as queue was being instantiated + * and driver had not attached a device yet + */ + if (bdi->dev && dev_name(bdi->dev)) { + sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor); + tg->blkg.dev = MKDEV(major, minor); + } +} + +/* + * Should be called with without queue lock held. Here queue lock will be + * taken rarely. It will be taken only once during life time of a group + * if need be + */ +static void +throtl_tg_fill_dev_details(struct throtl_data *td, struct throtl_grp *tg) +{ + if (!tg || tg->blkg.dev) + return; + + spin_lock_irq(td->queue->queue_lock); + __throtl_tg_fill_dev_details(td, tg); + spin_unlock_irq(td->queue->queue_lock); +} + +static void throtl_init_add_tg_lists(struct throtl_data *td, + struct throtl_grp *tg, struct blkio_cgroup *blkcg) +{ + __throtl_tg_fill_dev_details(td, tg); /* Add group onto cgroup list */ - sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor); blkiocg_add_blkio_group(blkcg, &tg->blkg, (void *)td, - MKDEV(major, minor), BLKIO_POLICY_THROTL); + tg->blkg.dev, BLKIO_POLICY_THROTL); tg->bps[READ] = blkcg_get_read_bps(blkcg, tg->blkg.dev); tg->bps[WRITE] = blkcg_get_write_bps(blkcg, tg->blkg.dev); tg->iops[READ] = blkcg_get_read_iops(blkcg, tg->blkg.dev); tg->iops[WRITE] = blkcg_get_write_iops(blkcg, tg->blkg.dev); - hlist_add_head(&tg->tg_node, &td->tg_list); - td->nr_undestroyed_grps++; -done: + throtl_add_group_to_td_list(td, tg); +} + +/* Should be called without queue lock and outside of rcu period */ +static struct throtl_grp *throtl_alloc_tg(struct throtl_data *td) +{ + struct throtl_grp *tg = NULL; + int ret; + + tg = kzalloc_node(sizeof(*tg), GFP_ATOMIC, td->queue->node); + if (!tg) + return NULL; + + ret = blkio_alloc_blkg_stats(&tg->blkg); + + if (ret) { + kfree(tg); + return NULL; + } + + throtl_init_group(tg); return tg; } -static struct throtl_grp * throtl_get_tg(struct throtl_data *td) +static struct +throtl_grp *throtl_find_tg(struct throtl_data *td, struct blkio_cgroup *blkcg) { struct throtl_grp *tg = NULL; + void *key = td; + + /* + * This is the common case when there are no blkio cgroups. + * Avoid lookup in this case + */ + if (blkcg == &blkio_root_cgroup) + tg = td->root_tg; + else + tg = tg_of_blkg(blkiocg_lookup_group(blkcg, key)); + + __throtl_tg_fill_dev_details(td, tg); + return tg; +} + +/* + * This function returns with queue lock unlocked in case of error, like + * request queue is no more + */ +static struct throtl_grp * throtl_get_tg(struct throtl_data *td) +{ + struct throtl_grp *tg = NULL, *__tg = NULL; struct blkio_cgroup *blkcg; + struct request_queue *q = td->queue; rcu_read_lock(); blkcg = task_blkio_cgroup(current); - tg = throtl_find_alloc_tg(td, blkcg); - if (!tg) - tg = &td->root_tg; + tg = throtl_find_tg(td, blkcg); + if (tg) { + rcu_read_unlock(); + return tg; + } + + /* + * Need to allocate a group. Allocation of group also needs allocation + * of per cpu stats which in-turn takes a mutex() and can block. Hence + * we need to drop rcu lock and queue_lock before we call alloc + * + * Take the request queue reference to make sure queue does not + * go away once we return from allocation. + */ + blk_get_queue(q); + rcu_read_unlock(); + spin_unlock_irq(q->queue_lock); + + tg = throtl_alloc_tg(td); + /* + * We might have slept in group allocation. Make sure queue is not + * dead + */ + if (unlikely(test_bit(QUEUE_FLAG_DEAD, &q->queue_flags))) { + blk_put_queue(q); + if (tg) + kfree(tg); + + return ERR_PTR(-ENODEV); + } + blk_put_queue(q); + + /* Group allocated and queue is still alive. take the lock */ + spin_lock_irq(q->queue_lock); + + /* + * Initialize the new group. After sleeping, read the blkcg again. + */ + rcu_read_lock(); + blkcg = task_blkio_cgroup(current); + + /* + * If some other thread already allocated the group while we were + * not holding queue lock, free up the group + */ + __tg = throtl_find_tg(td, blkcg); + + if (__tg) { + kfree(tg); + rcu_read_unlock(); + return __tg; + } + + /* Group allocation failed. Account the IO to root group */ + if (!tg) { + tg = td->root_tg; + return tg; + } + + throtl_init_add_tg_lists(td, tg, blkcg); rcu_read_unlock(); return tg; } @@ -544,6 +682,12 @@ static bool tg_with_in_bps_limit(struct throtl_data *td, struct throtl_grp *tg, return 0; } +static bool tg_no_rule_group(struct throtl_grp *tg, bool rw) { + if (tg->bps[rw] == -1 && tg->iops[rw] == -1) + return 1; + return 0; +} + /* * Returns whether one can dispatch a bio or not. Also returns approx number * of jiffies to wait before this bio is with-in IO rate and can be dispatched @@ -608,10 +752,6 @@ static void throtl_charge_bio(struct throtl_grp *tg, struct bio *bio) tg->bytes_disp[rw] += bio->bi_size; tg->io_disp[rw]++; - /* - * TODO: This will take blkg->stats_lock. Figure out a way - * to avoid this cost. - */ blkiocg_update_dispatch_stats(&tg->blkg, bio->bi_size, rw, sync); } @@ -989,15 +1129,51 @@ int blk_throtl_bio(struct request_queue *q, struct bio **biop) struct throtl_grp *tg; struct bio *bio = *biop; bool rw = bio_data_dir(bio), update_disptime = true; + struct blkio_cgroup *blkcg; if (bio->bi_rw & REQ_THROTTLED) { bio->bi_rw &= ~REQ_THROTTLED; return 0; } + /* + * A throtl_grp pointer retrieved under rcu can be used to access + * basic fields like stats and io rates. If a group has no rules, + * just update the dispatch stats in lockless manner and return. + */ + + rcu_read_lock(); + blkcg = task_blkio_cgroup(current); + tg = throtl_find_tg(td, blkcg); + if (tg) { + throtl_tg_fill_dev_details(td, tg); + + if (tg_no_rule_group(tg, rw)) { + blkiocg_update_dispatch_stats(&tg->blkg, bio->bi_size, + rw, bio->bi_rw & REQ_SYNC); + rcu_read_unlock(); + return 0; + } + } + rcu_read_unlock(); + + /* + * Either group has not been allocated yet or it is not an unlimited + * IO group + */ + spin_lock_irq(q->queue_lock); tg = throtl_get_tg(td); + if (IS_ERR(tg)) { + if (PTR_ERR(tg) == -ENODEV) { + /* + * Queue is gone. No queue lock held here. + */ + return -ENODEV; + } + } + if (tg->nr_queued[rw]) { /* * There is already another bio queued in same dir. No @@ -1060,39 +1236,24 @@ int blk_throtl_init(struct request_queue *q) INIT_HLIST_HEAD(&td->tg_list); td->tg_service_tree = THROTL_RB_ROOT; td->limits_changed = false; + INIT_DELAYED_WORK(&td->throtl_work, blk_throtl_work); - /* Init root group */ - tg = &td->root_tg; - INIT_HLIST_NODE(&tg->tg_node); - RB_CLEAR_NODE(&tg->rb_node); - bio_list_init(&tg->bio_lists[0]); - bio_list_init(&tg->bio_lists[1]); - - /* Practically unlimited BW */ - tg->bps[0] = tg->bps[1] = -1; - tg->iops[0] = tg->iops[1] = -1; - td->limits_changed = false; + /* alloc and Init root group. */ + td->queue = q; + tg = throtl_alloc_tg(td); - /* - * Set root group reference to 2. One reference will be dropped when - * all groups on tg_list are being deleted during queue exit. Other - * reference will remain there as we don't want to delete this group - * as it is statically allocated and gets destroyed when throtl_data - * goes away. - */ - atomic_set(&tg->ref, 2); - hlist_add_head(&tg->tg_node, &td->tg_list); - td->nr_undestroyed_grps++; + if (!tg) { + kfree(td); + return -ENOMEM; + } - INIT_DELAYED_WORK(&td->throtl_work, blk_throtl_work); + td->root_tg = tg; rcu_read_lock(); - blkiocg_add_blkio_group(&blkio_root_cgroup, &tg->blkg, (void *)td, - 0, BLKIO_POLICY_THROTL); + throtl_init_add_tg_lists(td, tg, &blkio_root_cgroup); rcu_read_unlock(); /* Attach throtl data to request queue */ - td->queue = q; q->td = td; return 0; } diff --git a/block/blk.h b/block/blk.h index 4df474d37e6d..d6586287adc9 100644 --- a/block/blk.h +++ b/block/blk.h @@ -62,6 +62,26 @@ static inline struct request *__elv_next_request(struct request_queue *q) return rq; } + /* + * Flush request is running and flush request isn't queueable + * in the drive, we can hold the queue till flush request is + * finished. Even we don't do this, driver can't dispatch next + * requests and will requeue them. And this can improve + * throughput too. For example, we have request flush1, write1, + * flush 2. flush1 is dispatched, then queue is hold, write1 + * isn't inserted to queue. After flush1 is finished, flush2 + * will be dispatched. Since disk cache is already clean, + * flush2 will be finished very soon, so looks like flush2 is + * folded to flush1. + * Since the queue is hold, a flag is set to indicate the queue + * should be restarted later. Please see flush_end_io() for + * details. + */ + if (q->flush_pending_idx != q->flush_running_idx && + !queue_flush_queueable(q)) { + q->flush_queue_delayed = 1; + return NULL; + } if (test_bit(QUEUE_FLAG_DEAD, &q->queue_flags) || !q->elevator->ops->elevator_dispatch_fn(q, 0)) return NULL; diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index ab7a9e6a9b1c..d646b279c8bb 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -300,7 +300,9 @@ struct cfq_data { /* List of cfq groups being managed on this device*/ struct hlist_head cfqg_list; - struct rcu_head rcu; + + /* Number of groups which are on blkcg->blkg_list */ + unsigned int nr_blkcg_linked_grps; }; static struct cfq_group *cfq_get_next_cfqg(struct cfq_data *cfqd); @@ -1014,28 +1016,47 @@ void cfq_update_blkio_group_weight(void *key, struct blkio_group *blkg, cfqg->needs_update = true; } -static struct cfq_group * cfq_find_alloc_cfqg(struct cfq_data *cfqd, - struct blkio_cgroup *blkcg, int create) +static void cfq_init_add_cfqg_lists(struct cfq_data *cfqd, + struct cfq_group *cfqg, struct blkio_cgroup *blkcg) { - struct cfq_group *cfqg = NULL; - void *key = cfqd; - int i, j; - struct cfq_rb_root *st; struct backing_dev_info *bdi = &cfqd->queue->backing_dev_info; unsigned int major, minor; - cfqg = cfqg_of_blkg(blkiocg_lookup_group(blkcg, key)); - if (cfqg && !cfqg->blkg.dev && bdi->dev && dev_name(bdi->dev)) { + /* + * Add group onto cgroup list. It might happen that bdi->dev is + * not initialized yet. Initialize this new group without major + * and minor info and this info will be filled in once a new thread + * comes for IO. + */ + if (bdi->dev) { sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor); - cfqg->blkg.dev = MKDEV(major, minor); - goto done; - } - if (cfqg || !create) - goto done; + cfq_blkiocg_add_blkio_group(blkcg, &cfqg->blkg, + (void *)cfqd, MKDEV(major, minor)); + } else + cfq_blkiocg_add_blkio_group(blkcg, &cfqg->blkg, + (void *)cfqd, 0); + + cfqd->nr_blkcg_linked_grps++; + cfqg->weight = blkcg_get_weight(blkcg, cfqg->blkg.dev); + + /* Add group on cfqd list */ + hlist_add_head(&cfqg->cfqd_node, &cfqd->cfqg_list); +} + +/* + * Should be called from sleepable context. No request queue lock as per + * cpu stats are allocated dynamically and alloc_percpu needs to be called + * from sleepable context. + */ +static struct cfq_group * cfq_alloc_cfqg(struct cfq_data *cfqd) +{ + struct cfq_group *cfqg = NULL; + int i, j, ret; + struct cfq_rb_root *st; cfqg = kzalloc_node(sizeof(*cfqg), GFP_ATOMIC, cfqd->queue->node); if (!cfqg) - goto done; + return NULL; for_each_cfqg_st(cfqg, i, j, st) *st = CFQ_RB_ROOT; @@ -1049,43 +1070,94 @@ static struct cfq_group * cfq_find_alloc_cfqg(struct cfq_data *cfqd, */ cfqg->ref = 1; + ret = blkio_alloc_blkg_stats(&cfqg->blkg); + if (ret) { + kfree(cfqg); + return NULL; + } + + return cfqg; +} + +static struct cfq_group * +cfq_find_cfqg(struct cfq_data *cfqd, struct blkio_cgroup *blkcg) +{ + struct cfq_group *cfqg = NULL; + void *key = cfqd; + struct backing_dev_info *bdi = &cfqd->queue->backing_dev_info; + unsigned int major, minor; + /* - * Add group onto cgroup list. It might happen that bdi->dev is - * not initialized yet. Initialize this new group without major - * and minor info and this info will be filled in once a new thread - * comes for IO. See code above. + * This is the common case when there are no blkio cgroups. + * Avoid lookup in this case */ - if (bdi->dev) { - sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor); - cfq_blkiocg_add_blkio_group(blkcg, &cfqg->blkg, (void *)cfqd, - MKDEV(major, minor)); - } else - cfq_blkiocg_add_blkio_group(blkcg, &cfqg->blkg, (void *)cfqd, - 0); - - cfqg->weight = blkcg_get_weight(blkcg, cfqg->blkg.dev); + if (blkcg == &blkio_root_cgroup) + cfqg = &cfqd->root_group; + else + cfqg = cfqg_of_blkg(blkiocg_lookup_group(blkcg, key)); - /* Add group on cfqd list */ - hlist_add_head(&cfqg->cfqd_node, &cfqd->cfqg_list); + if (cfqg && !cfqg->blkg.dev && bdi->dev && dev_name(bdi->dev)) { + sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor); + cfqg->blkg.dev = MKDEV(major, minor); + } -done: return cfqg; } /* - * Search for the cfq group current task belongs to. If create = 1, then also - * create the cfq group if it does not exist. request_queue lock must be held. + * Search for the cfq group current task belongs to. request_queue lock must + * be held. */ -static struct cfq_group *cfq_get_cfqg(struct cfq_data *cfqd, int create) +static struct cfq_group *cfq_get_cfqg(struct cfq_data *cfqd) { struct blkio_cgroup *blkcg; - struct cfq_group *cfqg = NULL; + struct cfq_group *cfqg = NULL, *__cfqg = NULL; + struct request_queue *q = cfqd->queue; rcu_read_lock(); blkcg = task_blkio_cgroup(current); - cfqg = cfq_find_alloc_cfqg(cfqd, blkcg, create); - if (!cfqg && create) + cfqg = cfq_find_cfqg(cfqd, blkcg); + if (cfqg) { + rcu_read_unlock(); + return cfqg; + } + + /* + * Need to allocate a group. Allocation of group also needs allocation + * of per cpu stats which in-turn takes a mutex() and can block. Hence + * we need to drop rcu lock and queue_lock before we call alloc. + * + * Not taking any queue reference here and assuming that queue is + * around by the time we return. CFQ queue allocation code does + * the same. It might be racy though. + */ + + rcu_read_unlock(); + spin_unlock_irq(q->queue_lock); + + cfqg = cfq_alloc_cfqg(cfqd); + + spin_lock_irq(q->queue_lock); + + rcu_read_lock(); + blkcg = task_blkio_cgroup(current); + + /* + * If some other thread already allocated the group while we were + * not holding queue lock, free up the group + */ + __cfqg = cfq_find_cfqg(cfqd, blkcg); + + if (__cfqg) { + kfree(cfqg); + rcu_read_unlock(); + return __cfqg; + } + + if (!cfqg) cfqg = &cfqd->root_group; + + cfq_init_add_cfqg_lists(cfqd, cfqg, blkcg); rcu_read_unlock(); return cfqg; } @@ -1118,6 +1190,7 @@ static void cfq_put_cfqg(struct cfq_group *cfqg) return; for_each_cfqg_st(cfqg, i, j, st) BUG_ON(!RB_EMPTY_ROOT(&st->rb)); + free_percpu(cfqg->blkg.stats_cpu); kfree(cfqg); } @@ -1176,7 +1249,7 @@ void cfq_unlink_blkio_group(void *key, struct blkio_group *blkg) } #else /* GROUP_IOSCHED */ -static struct cfq_group *cfq_get_cfqg(struct cfq_data *cfqd, int create) +static struct cfq_group *cfq_get_cfqg(struct cfq_data *cfqd) { return &cfqd->root_group; } @@ -2911,7 +2984,7 @@ cfq_find_alloc_queue(struct cfq_data *cfqd, bool is_sync, struct cfq_group *cfqg; retry: - cfqg = cfq_get_cfqg(cfqd, 1); + cfqg = cfq_get_cfqg(cfqd); cic = cfq_cic_lookup(cfqd, ioc); /* cic always exists here */ cfqq = cic_to_cfqq(cic, is_sync); @@ -3815,15 +3888,11 @@ static void cfq_put_async_queues(struct cfq_data *cfqd) cfq_put_queue(cfqd->async_idle_cfqq); } -static void cfq_cfqd_free(struct rcu_head *head) -{ - kfree(container_of(head, struct cfq_data, rcu)); -} - static void cfq_exit_queue(struct elevator_queue *e) { struct cfq_data *cfqd = e->elevator_data; struct request_queue *q = cfqd->queue; + bool wait = false; cfq_shutdown_timer_wq(cfqd); @@ -3842,7 +3911,13 @@ static void cfq_exit_queue(struct elevator_queue *e) cfq_put_async_queues(cfqd); cfq_release_cfq_groups(cfqd); - cfq_blkiocg_del_blkio_group(&cfqd->root_group.blkg); + + /* + * If there are groups which we could not unlink from blkcg list, + * wait for a rcu period for them to be freed. + */ + if (cfqd->nr_blkcg_linked_grps) + wait = true; spin_unlock_irq(q->queue_lock); @@ -3852,8 +3927,20 @@ static void cfq_exit_queue(struct elevator_queue *e) ida_remove(&cic_index_ida, cfqd->cic_index); spin_unlock(&cic_index_lock); - /* Wait for cfqg->blkg->key accessors to exit their grace periods. */ - call_rcu(&cfqd->rcu, cfq_cfqd_free); + /* + * Wait for cfqg->blkg->key accessors to exit their grace periods. + * Do this wait only if there are other unlinked groups out + * there. This can happen if cgroup deletion path claimed the + * responsibility of cleaning up a group before queue cleanup code + * get to the group. + * + * Do not call synchronize_rcu() unconditionally as there are drivers + * which create/delete request queue hundreds of times during scan/boot + * and synchronize_rcu() can take significant time and slow down boot. + */ + if (wait) + synchronize_rcu(); + kfree(cfqd); } static int cfq_alloc_cic_index(void) @@ -3909,14 +3996,29 @@ static void *cfq_init_queue(struct request_queue *q) #ifdef CONFIG_CFQ_GROUP_IOSCHED /* - * Take a reference to root group which we never drop. This is just - * to make sure that cfq_put_cfqg() does not try to kfree root group + * Set root group reference to 2. One reference will be dropped when + * all groups on cfqd->cfqg_list are being deleted during queue exit. + * Other reference will remain there as we don't want to delete this + * group as it is statically allocated and gets destroyed when + * throtl_data goes away. */ - cfqg->ref = 1; + cfqg->ref = 2; + + if (blkio_alloc_blkg_stats(&cfqg->blkg)) { + kfree(cfqg); + kfree(cfqd); + return NULL; + } + rcu_read_lock(); + cfq_blkiocg_add_blkio_group(&blkio_root_cgroup, &cfqg->blkg, (void *)cfqd, 0); rcu_read_unlock(); + cfqd->nr_blkcg_linked_grps++; + + /* Add group on cfqd->cfqg_list */ + hlist_add_head(&cfqg->cfqd_node, &cfqd->cfqg_list); #endif /* * Not strictly needed (since RB_ROOT just clears the node and we diff --git a/block/elevator.c b/block/elevator.c index 45ca1e34f582..2a0b653c90fd 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -155,13 +155,8 @@ static struct elevator_type *elevator_get(const char *name) e = elevator_find(name); if (!e) { - char elv[ELV_NAME_MAX + strlen("-iosched")]; - spin_unlock(&elv_list_lock); - - snprintf(elv, sizeof(elv), "%s-iosched", name); - - request_module("%s", elv); + request_module("%s-iosched", name); spin_lock(&elv_list_lock); e = elevator_find(name); } diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index e2f57e9e12f0..dad2c952e0d2 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -1089,21 +1089,21 @@ static int atapi_drain_needed(struct request *rq) static int ata_scsi_dev_config(struct scsi_device *sdev, struct ata_device *dev) { + struct request_queue *q = sdev->request_queue; + if (!ata_id_has_unload(dev->id)) dev->flags |= ATA_DFLAG_NO_UNLOAD; /* configure max sectors */ - blk_queue_max_hw_sectors(sdev->request_queue, dev->max_sectors); + blk_queue_max_hw_sectors(q, dev->max_sectors); if (dev->class == ATA_DEV_ATAPI) { - struct request_queue *q = sdev->request_queue; void *buf; sdev->sector_size = ATA_SECT_SIZE; /* set DMA padding */ - blk_queue_update_dma_pad(sdev->request_queue, - ATA_DMA_PAD_SZ - 1); + blk_queue_update_dma_pad(q, ATA_DMA_PAD_SZ - 1); /* configure draining */ buf = kmalloc(ATAPI_MAX_DRAIN, q->bounce_gfp | GFP_KERNEL); @@ -1131,8 +1131,7 @@ static int ata_scsi_dev_config(struct scsi_device *sdev, "sector_size=%u > PAGE_SIZE, PIO may malfunction\n", sdev->sector_size); - blk_queue_update_dma_alignment(sdev->request_queue, - sdev->sector_size - 1); + blk_queue_update_dma_alignment(q, sdev->sector_size - 1); if (dev->flags & ATA_DFLAG_AN) set_bit(SDEV_EVT_MEDIA_CHANGE, sdev->supported_events); @@ -1145,6 +1144,8 @@ static int ata_scsi_dev_config(struct scsi_device *sdev, scsi_adjust_queue_depth(sdev, MSG_SIMPLE_TAG, depth); } + blk_queue_flush_queueable(q, false); + dev->sdev = sdev; return 0; } diff --git a/drivers/atm/fore200e.c b/drivers/atm/fore200e.c index bdd2719f3f68..bc9e702186dd 100644 --- a/drivers/atm/fore200e.c +++ b/drivers/atm/fore200e.c @@ -2643,16 +2643,19 @@ fore200e_init(struct fore200e* fore200e, struct device *parent) } #ifdef CONFIG_SBUS +static const struct of_device_id fore200e_sba_match[]; static int __devinit fore200e_sba_probe(struct platform_device *op) { + const struct of_device_id *match; const struct fore200e_bus *bus; struct fore200e *fore200e; static int index = 0; int err; - if (!op->dev.of_match) + match = of_match_device(fore200e_sba_match, &op->dev); + if (!match) return -EINVAL; - bus = op->dev.of_match->data; + bus = match->data; fore200e = kzalloc(sizeof(struct fore200e), GFP_KERNEL); if (!fore200e) diff --git a/drivers/block/paride/pcd.c b/drivers/block/paride/pcd.c index 8690e31d9932..a0aabd904a51 100644 --- a/drivers/block/paride/pcd.c +++ b/drivers/block/paride/pcd.c @@ -320,6 +320,8 @@ static void pcd_init_units(void) disk->first_minor = unit; strcpy(disk->disk_name, cd->name); /* umm... */ disk->fops = &pcd_bdops; + disk->flags = GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE; + disk->events = DISK_EVENT_MEDIA_CHANGE; } } diff --git a/drivers/cdrom/viocd.c b/drivers/cdrom/viocd.c index e427fbe45999..ae15a4ddaa9b 100644 --- a/drivers/cdrom/viocd.c +++ b/drivers/cdrom/viocd.c @@ -625,7 +625,9 @@ static int viocd_probe(struct vio_dev *vdev, const struct vio_device_id *id) blk_queue_max_hw_sectors(q, 4096 / 512); gendisk->queue = q; gendisk->fops = &viocd_fops; - gendisk->flags = GENHD_FL_CD|GENHD_FL_REMOVABLE; + gendisk->flags = GENHD_FL_CD | GENHD_FL_REMOVABLE | + GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE; + gendisk->events = DISK_EVENT_MEDIA_CHANGE; set_capacity(gendisk, 0); gendisk->private_data = d; d->viocd_disk = gendisk; diff --git a/drivers/char/hw_random/n2-drv.c b/drivers/char/hw_random/n2-drv.c index 43ac61978d8b..ac6739e085e3 100644 --- a/drivers/char/hw_random/n2-drv.c +++ b/drivers/char/hw_random/n2-drv.c @@ -619,15 +619,18 @@ static void __devinit n2rng_driver_version(void) pr_info("%s", version); } +static const struct of_device_id n2rng_match[]; static int __devinit n2rng_probe(struct platform_device *op) { + const struct of_device_id *match; int victoria_falls; int err = -ENOMEM; struct n2rng *np; - if (!op->dev.of_match) + match = of_match_device(n2rng_match, &op->dev); + if (!match) return -EINVAL; - victoria_falls = (op->dev.of_match->data != NULL); + victoria_falls = (match->data != NULL); n2rng_driver_version(); np = kzalloc(sizeof(*np), GFP_KERNEL); diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index cc6c9b2546a3..64c6b8530615 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -2554,9 +2554,11 @@ static struct pci_driver ipmi_pci_driver = { }; #endif /* CONFIG_PCI */ +static struct of_device_id ipmi_match[]; static int __devinit ipmi_probe(struct platform_device *dev) { #ifdef CONFIG_OF + const struct of_device_id *match; struct smi_info *info; struct resource resource; const __be32 *regsize, *regspacing, *regshift; @@ -2566,7 +2568,8 @@ static int __devinit ipmi_probe(struct platform_device *dev) dev_info(&dev->dev, "probing via device tree\n"); - if (!dev->dev.of_match) + match = of_match_device(ipmi_match, &dev->dev); + if (!match) return -EINVAL; ret = of_address_to_resource(np, 0, &resource); @@ -2601,7 +2604,7 @@ static int __devinit ipmi_probe(struct platform_device *dev) return -ENOMEM; } - info->si_type = (enum si_type) dev->dev.of_match->data; + info->si_type = (enum si_type) match->data; info->addr_source = SI_DEVICETREE; info->irq_setup = std_irq_setup; diff --git a/drivers/char/xilinx_hwicap/xilinx_hwicap.c b/drivers/char/xilinx_hwicap/xilinx_hwicap.c index d6412c16385f..39ccdeada791 100644 --- a/drivers/char/xilinx_hwicap/xilinx_hwicap.c +++ b/drivers/char/xilinx_hwicap/xilinx_hwicap.c @@ -715,13 +715,13 @@ static int __devexit hwicap_remove(struct device *dev) } #ifdef CONFIG_OF -static int __devinit hwicap_of_probe(struct platform_device *op) +static int __devinit hwicap_of_probe(struct platform_device *op, + const struct hwicap_driver_config *config) { struct resource res; const unsigned int *id; const char *family; int rc; - const struct hwicap_driver_config *config = op->dev.of_match->data; const struct config_registers *regs; @@ -751,20 +751,24 @@ static int __devinit hwicap_of_probe(struct platform_device *op) regs); } #else -static inline int hwicap_of_probe(struct platform_device *op) +static inline int hwicap_of_probe(struct platform_device *op, + const struct hwicap_driver_config *config) { return -EINVAL; } #endif /* CONFIG_OF */ +static const struct of_device_id __devinitconst hwicap_of_match[]; static int __devinit hwicap_drv_probe(struct platform_device *pdev) { + const struct of_device_id *match; struct resource *res; const struct config_registers *regs; const char *family; - if (pdev->dev.of_match) - return hwicap_of_probe(pdev); + match = of_match_device(hwicap_of_match, &pdev->dev); + if (match) + return hwicap_of_probe(pdev, match->data); res = platform_get_resource(pdev, IORESOURCE_MEM, 0); if (!res) diff --git a/drivers/edac/ppc4xx_edac.c b/drivers/edac/ppc4xx_edac.c index c1f0045ceb8e..af8e7b1aa290 100644 --- a/drivers/edac/ppc4xx_edac.c +++ b/drivers/edac/ppc4xx_edac.c @@ -1019,7 +1019,7 @@ ppc4xx_edac_mc_init(struct mem_ctl_info *mci, struct ppc4xx_edac_pdata *pdata = NULL; const struct device_node *np = op->dev.of_node; - if (op->dev.of_match == NULL) + if (of_match_device(ppc4xx_edac_match, &op->dev) == NULL) return -EINVAL; /* Initial driver pointers and private data */ diff --git a/drivers/i2c/busses/i2c-mpc.c b/drivers/i2c/busses/i2c-mpc.c index 75b984c519ac..107397a606b4 100644 --- a/drivers/i2c/busses/i2c-mpc.c +++ b/drivers/i2c/busses/i2c-mpc.c @@ -560,15 +560,18 @@ static struct i2c_adapter mpc_ops = { .timeout = HZ, }; +static const struct of_device_id mpc_i2c_of_match[]; static int __devinit fsl_i2c_probe(struct platform_device *op) { + const struct of_device_id *match; struct mpc_i2c *i2c; const u32 *prop; u32 clock = MPC_I2C_CLOCK_LEGACY; int result = 0; int plen; - if (!op->dev.of_match) + match = of_match_device(mpc_i2c_of_match, &op->dev); + if (!match) return -EINVAL; i2c = kzalloc(sizeof(*i2c), GFP_KERNEL); @@ -605,8 +608,8 @@ static int __devinit fsl_i2c_probe(struct platform_device *op) clock = *prop; } - if (op->dev.of_match->data) { - struct mpc_i2c_data *data = op->dev.of_match->data; + if (match->data) { + struct mpc_i2c_data *data = match->data; data->setup(op->dev.of_node, i2c, clock, data->prescaler); } else { /* Backwards compatibility */ diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c index a5ec5a7cb381..6e5123b1d341 100644 --- a/drivers/ide/ide-cd.c +++ b/drivers/ide/ide-cd.c @@ -1781,7 +1781,8 @@ static int ide_cd_probe(ide_drive_t *drive) ide_cd_read_toc(drive, &sense); g->fops = &idecd_ops; - g->flags |= GENHD_FL_REMOVABLE; + g->flags |= GENHD_FL_REMOVABLE | GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE; + g->events = DISK_EVENT_MEDIA_CHANGE; add_disk(g); return 0; diff --git a/drivers/mmc/host/sdhci-of-core.c b/drivers/mmc/host/sdhci-of-core.c index f9b611fc773e..60e4186a4345 100644 --- a/drivers/mmc/host/sdhci-of-core.c +++ b/drivers/mmc/host/sdhci-of-core.c @@ -124,8 +124,10 @@ static bool __devinit sdhci_of_wp_inverted(struct device_node *np) #endif } +static const struct of_device_id sdhci_of_match[]; static int __devinit sdhci_of_probe(struct platform_device *ofdev) { + const struct of_device_id *match; struct device_node *np = ofdev->dev.of_node; struct sdhci_of_data *sdhci_of_data; struct sdhci_host *host; @@ -134,9 +136,10 @@ static int __devinit sdhci_of_probe(struct platform_device *ofdev) int size; int ret; - if (!ofdev->dev.of_match) + match = of_match_device(sdhci_of_match, &ofdev->dev); + if (!match) return -EINVAL; - sdhci_of_data = ofdev->dev.of_match->data; + sdhci_of_data = match->data; if (!of_device_is_available(np)) return -ENODEV; diff --git a/drivers/mtd/maps/physmap_of.c b/drivers/mtd/maps/physmap_of.c index bd483f0c57e1..c1d33464aee8 100644 --- a/drivers/mtd/maps/physmap_of.c +++ b/drivers/mtd/maps/physmap_of.c @@ -214,11 +214,13 @@ static void __devinit of_free_probes(const char **probes) } #endif +static struct of_device_id of_flash_match[]; static int __devinit of_flash_probe(struct platform_device *dev) { #ifdef CONFIG_MTD_PARTITIONS const char **part_probe_types; #endif + const struct of_device_id *match; struct device_node *dp = dev->dev.of_node; struct resource res; struct of_flash *info; @@ -232,9 +234,10 @@ static int __devinit of_flash_probe(struct platform_device *dev) struct mtd_info **mtd_list = NULL; resource_size_t res_size; - if (!dev->dev.of_match) + match = of_match_device(of_flash_match, &dev->dev); + if (!match) return -EINVAL; - probe_type = dev->dev.of_match->data; + probe_type = match->data; reg_tuple_size = (of_n_addr_cells(dp) + of_n_size_cells(dp)) * sizeof(u32); diff --git a/drivers/net/can/mscan/mpc5xxx_can.c b/drivers/net/can/mscan/mpc5xxx_can.c index bd1d811c204f..5fedc3375562 100644 --- a/drivers/net/can/mscan/mpc5xxx_can.c +++ b/drivers/net/can/mscan/mpc5xxx_can.c @@ -247,8 +247,10 @@ static u32 __devinit mpc512x_can_get_clock(struct platform_device *ofdev, } #endif /* CONFIG_PPC_MPC512x */ +static struct of_device_id mpc5xxx_can_table[]; static int __devinit mpc5xxx_can_probe(struct platform_device *ofdev) { + const struct of_device_id *match; struct mpc5xxx_can_data *data; struct device_node *np = ofdev->dev.of_node; struct net_device *dev; @@ -258,9 +260,10 @@ static int __devinit mpc5xxx_can_probe(struct platform_device *ofdev) int irq, mscan_clksrc = 0; int err = -ENOMEM; - if (!ofdev->dev.of_match) + match = of_match_device(mpc5xxx_can_table, &ofdev->dev); + if (!match) return -EINVAL; - data = (struct mpc5xxx_can_data *)ofdev->dev.of_match->data; + data = match->data; base = of_iomap(np, 0); if (!base) { diff --git a/drivers/net/fs_enet/fs_enet-main.c b/drivers/net/fs_enet/fs_enet-main.c index 24cb953900dd..5131e61c358c 100644 --- a/drivers/net/fs_enet/fs_enet-main.c +++ b/drivers/net/fs_enet/fs_enet-main.c @@ -998,8 +998,10 @@ static const struct net_device_ops fs_enet_netdev_ops = { #endif }; +static struct of_device_id fs_enet_match[]; static int __devinit fs_enet_probe(struct platform_device *ofdev) { + const struct of_device_id *match; struct net_device *ndev; struct fs_enet_private *fep; struct fs_platform_info *fpi; @@ -1007,14 +1009,15 @@ static int __devinit fs_enet_probe(struct platform_device *ofdev) const u8 *mac_addr; int privsize, len, ret = -ENODEV; - if (!ofdev->dev.of_match) + match = of_match_device(fs_enet_match, &ofdev->dev); + if (!match) return -EINVAL; fpi = kzalloc(sizeof(*fpi), GFP_KERNEL); if (!fpi) return -ENOMEM; - if (!IS_FEC(ofdev->dev.of_match)) { + if (!IS_FEC(match)) { data = of_get_property(ofdev->dev.of_node, "fsl,cpm-command", &len); if (!data || len != 4) goto out_free_fpi; @@ -1049,7 +1052,7 @@ static int __devinit fs_enet_probe(struct platform_device *ofdev) fep->dev = &ofdev->dev; fep->ndev = ndev; fep->fpi = fpi; - fep->ops = ofdev->dev.of_match->data; + fep->ops = match->data; ret = fep->ops->setup_data(ndev); if (ret) diff --git a/drivers/net/fs_enet/mii-fec.c b/drivers/net/fs_enet/mii-fec.c index 7e840d373ab3..6a2e150e75bb 100644 --- a/drivers/net/fs_enet/mii-fec.c +++ b/drivers/net/fs_enet/mii-fec.c @@ -101,17 +101,20 @@ static int fs_enet_fec_mii_reset(struct mii_bus *bus) return 0; } +static struct of_device_id fs_enet_mdio_fec_match[]; static int __devinit fs_enet_mdio_probe(struct platform_device *ofdev) { + const struct of_device_id *match; struct resource res; struct mii_bus *new_bus; struct fec_info *fec; int (*get_bus_freq)(struct device_node *); int ret = -ENOMEM, clock, speed; - if (!ofdev->dev.of_match) + match = of_match_device(fs_enet_mdio_fec_match, &ofdev->dev); + if (!match) return -EINVAL; - get_bus_freq = ofdev->dev.of_match->data; + get_bus_freq = match->data; new_bus = mdiobus_alloc(); if (!new_bus) diff --git a/drivers/net/sunhme.c b/drivers/net/sunhme.c index eb4f59fb01e9..bff2f7999ff0 100644 --- a/drivers/net/sunhme.c +++ b/drivers/net/sunhme.c @@ -3237,15 +3237,18 @@ static void happy_meal_pci_exit(void) #endif #ifdef CONFIG_SBUS +static const struct of_device_id hme_sbus_match[]; static int __devinit hme_sbus_probe(struct platform_device *op) { + const struct of_device_id *match; struct device_node *dp = op->dev.of_node; const char *model = of_get_property(dp, "model", NULL); int is_qfe; - if (!op->dev.of_match) + match = of_match_device(hme_sbus_match, &op->dev); + if (!match) return -EINVAL; - is_qfe = (op->dev.of_match->data != NULL); + is_qfe = (match->data != NULL); if (!is_qfe && model && !strcmp(model, "SUNW,sbus-qfe")) is_qfe = 1; diff --git a/drivers/scsi/qlogicpti.c b/drivers/scsi/qlogicpti.c index e2d45c91b8e8..9689d41c7888 100644 --- a/drivers/scsi/qlogicpti.c +++ b/drivers/scsi/qlogicpti.c @@ -1292,8 +1292,10 @@ static struct scsi_host_template qpti_template = { .use_clustering = ENABLE_CLUSTERING, }; +static const struct of_device_id qpti_match[]; static int __devinit qpti_sbus_probe(struct platform_device *op) { + const struct of_device_id *match; struct scsi_host_template *tpnt; struct device_node *dp = op->dev.of_node; struct Scsi_Host *host; @@ -1301,9 +1303,10 @@ static int __devinit qpti_sbus_probe(struct platform_device *op) static int nqptis; const char *fcode; - if (!op->dev.of_match) + match = of_match_device(qpti_match, &op->dev); + if (!match) return -EINVAL; - tpnt = op->dev.of_match->data; + tpnt = match->data; /* Sometimes Antares cards come up not completely * setup, and we get a report of a zero IRQ. diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c index 95019c747cc1..4778e2707168 100644 --- a/drivers/scsi/sr.c +++ b/drivers/scsi/sr.c @@ -636,7 +636,7 @@ static int sr_probe(struct device *dev) disk->first_minor = minor; sprintf(disk->disk_name, "sr%d", minor); disk->fops = &sr_bdops; - disk->flags = GENHD_FL_CD; + disk->flags = GENHD_FL_CD | GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE; disk->events = DISK_EVENT_MEDIA_CHANGE | DISK_EVENT_EJECT_REQUEST; blk_queue_rq_timeout(sdev->request_queue, SR_TIMEOUT); diff --git a/drivers/tty/serial/of_serial.c b/drivers/tty/serial/of_serial.c index 0e8eec516df4..c911b2419abb 100644 --- a/drivers/tty/serial/of_serial.c +++ b/drivers/tty/serial/of_serial.c @@ -80,14 +80,17 @@ static int __devinit of_platform_serial_setup(struct platform_device *ofdev, /* * Try to register a serial port */ +static struct of_device_id of_platform_serial_table[]; static int __devinit of_platform_serial_probe(struct platform_device *ofdev) { + const struct of_device_id *match; struct of_serial_info *info; struct uart_port port; int port_type; int ret; - if (!ofdev->dev.of_match) + match = of_match_device(of_platform_serial_table, &ofdev->dev); + if (!match) return -EINVAL; if (of_find_property(ofdev->dev.of_node, "used-by-rtas", NULL)) @@ -97,7 +100,7 @@ static int __devinit of_platform_serial_probe(struct platform_device *ofdev) if (info == NULL) return -ENOMEM; - port_type = (unsigned long)ofdev->dev.of_match->data; + port_type = (unsigned long)match->data; ret = of_platform_serial_setup(ofdev, port_type, &port); if (ret) goto out; diff --git a/drivers/usb/gadget/fsl_qe_udc.c b/drivers/usb/gadget/fsl_qe_udc.c index 36613b37c504..3a68e09309f7 100644 --- a/drivers/usb/gadget/fsl_qe_udc.c +++ b/drivers/usb/gadget/fsl_qe_udc.c @@ -2539,15 +2539,18 @@ static void qe_udc_release(struct device *dev) } /* Driver probe functions */ +static const struct of_device_id qe_udc_match[]; static int __devinit qe_udc_probe(struct platform_device *ofdev) { + const struct of_device_id *match; struct device_node *np = ofdev->dev.of_node; struct qe_ep *ep; unsigned int ret = 0; unsigned int i; const void *prop; - if (!ofdev->dev.of_match) + match = of_match_device(qe_udc_match, &ofdev->dev); + if (!match) return -EINVAL; prop = of_get_property(np, "mode", NULL); @@ -2561,7 +2564,7 @@ static int __devinit qe_udc_probe(struct platform_device *ofdev) return -ENOMEM; } - udc_controller->soc_type = (unsigned long)ofdev->dev.of_match->data; + udc_controller->soc_type = (unsigned long)match->data; udc_controller->usb_regs = of_iomap(np, 0); if (!udc_controller->usb_regs) { ret = -ENOMEM; diff --git a/drivers/watchdog/mpc8xxx_wdt.c b/drivers/watchdog/mpc8xxx_wdt.c index 528bceb220fd..eed5436ffb51 100644 --- a/drivers/watchdog/mpc8xxx_wdt.c +++ b/drivers/watchdog/mpc8xxx_wdt.c @@ -185,17 +185,20 @@ static struct miscdevice mpc8xxx_wdt_miscdev = { .fops = &mpc8xxx_wdt_fops, }; +static const struct of_device_id mpc8xxx_wdt_match[]; static int __devinit mpc8xxx_wdt_probe(struct platform_device *ofdev) { int ret; + const struct of_device_id *match; struct device_node *np = ofdev->dev.of_node; struct mpc8xxx_wdt_type *wdt_type; u32 freq = fsl_get_sys_freq(); bool enabled; - if (!ofdev->dev.of_match) + match = of_match_device(mpc8xxx_wdt_match, &ofdev->dev); + if (!match) return -EINVAL; - wdt_type = ofdev->dev.of_match->data; + wdt_type = match->data; if (!freq || freq == -1) return -EINVAL; diff --git a/fs/block_dev.c b/fs/block_dev.c index 257b00e98428..d7c2e0fddc6f 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -1237,6 +1237,8 @@ int blkdev_get(struct block_device *bdev, fmode_t mode, void *holder) res = __blkdev_get(bdev, mode, 0); if (whole) { + struct gendisk *disk = whole->bd_disk; + /* finish claiming */ mutex_lock(&bdev->bd_mutex); spin_lock(&bdev_lock); @@ -1263,15 +1265,16 @@ int blkdev_get(struct block_device *bdev, fmode_t mode, void *holder) spin_unlock(&bdev_lock); /* - * Block event polling for write claims. Any write - * holder makes the write_holder state stick until all - * are released. This is good enough and tracking - * individual writeable reference is too fragile given - * the way @mode is used in blkdev_get/put(). + * Block event polling for write claims if requested. Any + * write holder makes the write_holder state stick until + * all are released. This is good enough and tracking + * individual writeable reference is too fragile given the + * way @mode is used in blkdev_get/put(). */ - if (!res && (mode & FMODE_WRITE) && !bdev->bd_write_holder) { + if ((disk->flags & GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE) && + !res && (mode & FMODE_WRITE) && !bdev->bd_write_holder) { bdev->bd_write_holder = true; - disk_block_events(bdev->bd_disk); + disk_block_events(disk); } mutex_unlock(&bdev->bd_mutex); diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c index 3313dd19f543..9a37a9b6de3a 100644 --- a/fs/configfs/dir.c +++ b/fs/configfs/dir.c @@ -53,11 +53,14 @@ DEFINE_SPINLOCK(configfs_dirent_lock); static void configfs_d_iput(struct dentry * dentry, struct inode * inode) { - struct configfs_dirent * sd = dentry->d_fsdata; + struct configfs_dirent *sd = dentry->d_fsdata; if (sd) { BUG_ON(sd->s_dentry != dentry); + /* Coordinate with configfs_readdir */ + spin_lock(&configfs_dirent_lock); sd->s_dentry = NULL; + spin_unlock(&configfs_dirent_lock); configfs_put(sd); } iput(inode); @@ -689,7 +692,8 @@ static int create_default_group(struct config_group *parent_group, sd = child->d_fsdata; sd->s_type |= CONFIGFS_USET_DEFAULT; } else { - d_delete(child); + BUG_ON(child->d_inode); + d_drop(child); dput(child); } } @@ -1545,7 +1549,7 @@ static int configfs_readdir(struct file * filp, void * dirent, filldir_t filldir struct configfs_dirent * parent_sd = dentry->d_fsdata; struct configfs_dirent *cursor = filp->private_data; struct list_head *p, *q = &cursor->s_sibling; - ino_t ino; + ino_t ino = 0; int i = filp->f_pos; switch (i) { @@ -1573,6 +1577,7 @@ static int configfs_readdir(struct file * filp, void * dirent, filldir_t filldir struct configfs_dirent *next; const char * name; int len; + struct inode *inode = NULL; next = list_entry(p, struct configfs_dirent, s_sibling); @@ -1581,9 +1586,28 @@ static int configfs_readdir(struct file * filp, void * dirent, filldir_t filldir name = configfs_get_name(next); len = strlen(name); - if (next->s_dentry) - ino = next->s_dentry->d_inode->i_ino; - else + + /* + * We'll have a dentry and an inode for + * PINNED items and for open attribute + * files. We lock here to prevent a race + * with configfs_d_iput() clearing + * s_dentry before calling iput(). + * + * Why do we go to the trouble? If + * someone has an attribute file open, + * the inode number should match until + * they close it. Beyond that, we don't + * care. + */ + spin_lock(&configfs_dirent_lock); + dentry = next->s_dentry; + if (dentry) + inode = dentry->d_inode; + if (inode) + ino = inode->i_ino; + spin_unlock(&configfs_dirent_lock); + if (!inode) ino = iunique(configfs_sb, 2); if (filldir(dirent, name, len, filp->f_pos, ino, @@ -1683,7 +1707,8 @@ int configfs_register_subsystem(struct configfs_subsystem *subsys) err = configfs_attach_group(sd->s_element, &group->cg_item, dentry); if (err) { - d_delete(dentry); + BUG_ON(dentry->d_inode); + d_drop(dentry); dput(dentry); } else { spin_lock(&configfs_dirent_lock); diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 643720209a98..9a3e6bbff27b 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -539,25 +539,41 @@ static int o2hb_verify_crc(struct o2hb_region *reg, /* We want to make sure that nobody is heartbeating on top of us -- * this will help detect an invalid configuration. */ -static int o2hb_check_last_timestamp(struct o2hb_region *reg) +static void o2hb_check_last_timestamp(struct o2hb_region *reg) { - int node_num, ret; struct o2hb_disk_slot *slot; struct o2hb_disk_heartbeat_block *hb_block; + char *errstr; - node_num = o2nm_this_node(); - - ret = 1; - slot = ®->hr_slots[node_num]; + slot = ®->hr_slots[o2nm_this_node()]; /* Don't check on our 1st timestamp */ - if (slot->ds_last_time) { - hb_block = slot->ds_raw_block; + if (!slot->ds_last_time) + return; - if (le64_to_cpu(hb_block->hb_seq) != slot->ds_last_time) - ret = 0; - } + hb_block = slot->ds_raw_block; + if (le64_to_cpu(hb_block->hb_seq) == slot->ds_last_time && + le64_to_cpu(hb_block->hb_generation) == slot->ds_last_generation && + hb_block->hb_node == slot->ds_node_num) + return; - return ret; +#define ERRSTR1 "Another node is heartbeating on device" +#define ERRSTR2 "Heartbeat generation mismatch on device" +#define ERRSTR3 "Heartbeat sequence mismatch on device" + + if (hb_block->hb_node != slot->ds_node_num) + errstr = ERRSTR1; + else if (le64_to_cpu(hb_block->hb_generation) != + slot->ds_last_generation) + errstr = ERRSTR2; + else + errstr = ERRSTR3; + + mlog(ML_ERROR, "%s (%s): expected(%u:0x%llx, 0x%llx), " + "ondisk(%u:0x%llx, 0x%llx)\n", errstr, reg->hr_dev_name, + slot->ds_node_num, (unsigned long long)slot->ds_last_generation, + (unsigned long long)slot->ds_last_time, hb_block->hb_node, + (unsigned long long)le64_to_cpu(hb_block->hb_generation), + (unsigned long long)le64_to_cpu(hb_block->hb_seq)); } static inline void o2hb_prepare_block(struct o2hb_region *reg, @@ -983,9 +999,7 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) /* With an up to date view of the slots, we can check that no * other node has been improperly configured to heartbeat in * our slot. */ - if (!o2hb_check_last_timestamp(reg)) - mlog(ML_ERROR, "Device \"%s\": another node is heartbeating " - "in our slot!\n", reg->hr_dev_name); + o2hb_check_last_timestamp(reg); /* fill in the proper info for our next heartbeat */ o2hb_prepare_block(reg, reg->hr_generation); @@ -999,8 +1013,8 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) } i = -1; - while((i = find_next_bit(configured_nodes, O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) { - + while((i = find_next_bit(configured_nodes, + O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) { change |= o2hb_check_slot(reg, ®->hr_slots[i]); } @@ -1690,6 +1704,7 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, struct file *filp = NULL; struct inode *inode = NULL; ssize_t ret = -EINVAL; + int live_threshold; if (reg->hr_bdev) goto out; @@ -1766,8 +1781,18 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, * A node is considered live after it has beat LIVE_THRESHOLD * times. We're not steady until we've given them a chance * _after_ our first read. + * The default threshold is bare minimum so as to limit the delay + * during mounts. For global heartbeat, the threshold doubled for the + * first region. */ - atomic_set(®->hr_steady_iterations, O2HB_LIVE_THRESHOLD + 1); + live_threshold = O2HB_LIVE_THRESHOLD; + if (o2hb_global_heartbeat_active()) { + spin_lock(&o2hb_live_lock); + if (o2hb_pop_count(&o2hb_region_bitmap, O2NM_MAX_REGIONS) == 1) + live_threshold <<= 1; + spin_unlock(&o2hb_live_lock); + } + atomic_set(®->hr_steady_iterations, live_threshold + 1); hb_task = kthread_run(o2hb_thread, reg, "o2hb-%s", reg->hr_item.ci_name); diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index 9fe5b8fd658f..8582e3f4f120 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -2868,7 +2868,7 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh, bytes = blocks_wanted << sb->s_blocksize_bits; struct ocfs2_super *osb = OCFS2_SB(dir->i_sb); struct ocfs2_inode_info *oi = OCFS2_I(dir); - struct ocfs2_alloc_context *data_ac; + struct ocfs2_alloc_context *data_ac = NULL; struct ocfs2_alloc_context *meta_ac = NULL; struct buffer_head *dirdata_bh = NULL; struct buffer_head *dx_root_bh = NULL; diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index 7540a492eaba..3b179d6cbde0 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -1614,7 +1614,8 @@ static int dlm_try_to_join_domain(struct dlm_ctxt *dlm) spin_unlock(&dlm->spinlock); /* Support for global heartbeat and node info was added in 1.1 */ - if (dlm_protocol.pv_major > 1 || dlm_protocol.pv_minor > 0) { + if (dlm->dlm_locking_proto.pv_major > 1 || + dlm->dlm_locking_proto.pv_minor > 0) { status = dlm_send_nodeinfo(dlm, ctxt->yes_resp_map); if (status) { mlog_errno(status); diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index fede57ed005f..84d166328cf7 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -2574,6 +2574,9 @@ fail: res->state &= ~DLM_LOCK_RES_MIGRATING; wake = 1; spin_unlock(&res->spinlock); + if (dlm_is_host_down(ret)) + dlm_wait_for_node_death(dlm, target, + DLM_NODE_DEATH_WAIT_MAX); goto leave; } diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 41565ae52856..89659d6dc206 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1607,6 +1607,9 @@ static void ocfs2_calc_trunc_pos(struct inode *inode, range = le32_to_cpu(rec->e_cpos) + ocfs2_rec_clusters(el, rec); if (le32_to_cpu(rec->e_cpos) >= trunc_start) { + /* + * remove an entire extent record. + */ *trunc_cpos = le32_to_cpu(rec->e_cpos); /* * Skip holes if any. @@ -1617,7 +1620,16 @@ static void ocfs2_calc_trunc_pos(struct inode *inode, *blkno = le64_to_cpu(rec->e_blkno); *trunc_end = le32_to_cpu(rec->e_cpos); } else if (range > trunc_start) { + /* + * remove a partial extent record, which means we're + * removing the last extent record. + */ *trunc_cpos = trunc_start; + /* + * skip hole if any. + */ + if (range < *trunc_end) + *trunc_end = range; *trunc_len = *trunc_end - trunc_start; coff = trunc_start - le32_to_cpu(rec->e_cpos); *blkno = le64_to_cpu(rec->e_blkno) + diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index b141a44605ca..295d56454e8b 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -1260,6 +1260,9 @@ void ocfs2_complete_mount_recovery(struct ocfs2_super *osb) { struct ocfs2_journal *journal = osb->journal; + if (ocfs2_is_hard_readonly(osb)) + return; + /* No need to queue up our truncate_log as regular cleanup will catch * that */ ocfs2_queue_recovery_completion(journal, osb->slot_num, diff --git a/fs/partitions/check.c b/fs/partitions/check.c index d545e97d99c3..8ed4d3433199 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -255,7 +255,11 @@ ssize_t part_discard_alignment_show(struct device *dev, struct device_attribute *attr, char *buf) { struct hd_struct *p = dev_to_part(dev); - return sprintf(buf, "%u\n", p->discard_alignment); + struct gendisk *disk = dev_to_disk(dev); + + return sprintf(buf, "%u\n", + queue_limit_discard_alignment(&disk->queue->limits, + p->start_sect)); } ssize_t part_stat_show(struct device *dev, @@ -449,8 +453,6 @@ struct hd_struct *add_partition(struct gendisk *disk, int partno, p->start_sect = start; p->alignment_offset = queue_limit_alignment_offset(&disk->queue->limits, start); - p->discard_alignment = - queue_limit_discard_alignment(&disk->queue->limits, start); p->nr_sects = len; p->partno = partno; p->policy = get_disk_ro(disk); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 2ad95fa1d130..ae9091a68480 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -257,7 +257,7 @@ struct queue_limits { unsigned char misaligned; unsigned char discard_misaligned; unsigned char cluster; - signed char discard_zeroes_data; + unsigned char discard_zeroes_data; }; struct request_queue @@ -364,6 +364,8 @@ struct request_queue * for flush operations */ unsigned int flush_flags; + unsigned int flush_not_queueable:1; + unsigned int flush_queue_delayed:1; unsigned int flush_pending_idx:1; unsigned int flush_running_idx:1; unsigned long flush_pending_since; @@ -843,6 +845,7 @@ extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *); extern void blk_queue_rq_timed_out(struct request_queue *, rq_timed_out_fn *); extern void blk_queue_rq_timeout(struct request_queue *, unsigned int); extern void blk_queue_flush(struct request_queue *q, unsigned int flush); +extern void blk_queue_flush_queueable(struct request_queue *q, bool queueable); extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev); extern int blk_rq_map_sg(struct request_queue *, struct request *, struct scatterlist *); @@ -1066,13 +1069,16 @@ static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector { unsigned int alignment = (sector << 9) & (lim->discard_granularity - 1); + if (!lim->max_discard_sectors) + return 0; + return (lim->discard_granularity + lim->discard_alignment - alignment) & (lim->discard_granularity - 1); } static inline unsigned int queue_discard_zeroes_data(struct request_queue *q) { - if (q->limits.discard_zeroes_data == 1) + if (q->limits.max_discard_sectors && q->limits.discard_zeroes_data == 1) return 1; return 0; @@ -1111,6 +1117,11 @@ static inline unsigned int block_size(struct block_device *bdev) return bdev->bd_block_size; } +static inline bool queue_flush_queueable(struct request_queue *q) +{ + return !q->flush_not_queueable; +} + typedef struct {struct page *v;} Sector; unsigned char *read_dev_sector(struct block_device *, sector_t, Sector *); diff --git a/include/linux/device.h b/include/linux/device.h index ab8dfc095709..d08399db6e2c 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -442,7 +442,6 @@ struct device { struct dev_archdata archdata; struct device_node *of_node; /* associated device tree node */ - const struct of_device_id *of_match; /* matching of_device_id from driver */ dev_t devt; /* dev_t, creates the sysfs "dev" */ diff --git a/include/linux/genhd.h b/include/linux/genhd.h index d764a426e9fd..b78956b3c2e7 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -100,7 +100,6 @@ struct hd_struct { sector_t start_sect; sector_t nr_sects; sector_t alignment_offset; - unsigned int discard_alignment; struct device __dev; struct kobject *holder_dir; int policy, partno; @@ -127,6 +126,7 @@ struct hd_struct { #define GENHD_FL_SUPPRESS_PARTITION_INFO 32 #define GENHD_FL_EXT_DEVT 64 /* allow extended devt */ #define GENHD_FL_NATIVE_CAPACITY 128 +#define GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE 256 enum { DISK_EVENT_MEDIA_CHANGE = 1 << 0, /* media changed */ diff --git a/include/linux/of_device.h b/include/linux/of_device.h index 8bfe6c1d4365..ae5638480ef2 100644 --- a/include/linux/of_device.h +++ b/include/linux/of_device.h @@ -21,8 +21,7 @@ extern void of_device_make_bus_id(struct device *dev); static inline int of_driver_match_device(struct device *dev, const struct device_driver *drv) { - dev->of_match = of_match_device(drv->of_match_table, dev); - return dev->of_match != NULL; + return of_match_device(drv->of_match_table, dev) != NULL; } extern struct platform_device *of_dev_get(struct platform_device *dev); @@ -58,6 +57,11 @@ static inline int of_device_uevent(struct device *dev, static inline void of_device_node_put(struct device *dev) { } +static inline const struct of_device_id *of_match_device( + const struct of_device_id *matches, const struct device *dev) +{ + return NULL; +} #endif /* CONFIG_OF_DEVICE */ #endif /* _LINUX_OF_DEVICE_H */ |