X-Git-Url: https://git.whamcloud.com/?p=fs%2Flustre-release.git;a=blobdiff_plain;f=lustre%2Fllite%2Flproc_llite.c;h=697c4c110292244cad179780fd30e8e92bde4528;hp=c84394ba83c93adc7acd636c587e79ec30e890be;hb=3e0d994fbf4c4f2e5c51b2be5669ad97aa02f840;hpb=d79ffa3ff7461d8dcfb831f0024ed093a3f6f104 diff --git a/lustre/llite/lproc_llite.c b/lustre/llite/lproc_llite.c index c84394ba..697c4c1 100644 --- a/lustre/llite/lproc_llite.c +++ b/lustre/llite/lproc_llite.c @@ -27,15 +27,13 @@ */ /* * This file is part of Lustre, http://www.lustre.org/ - * Lustre is a trademark of Sun Microsystems, Inc. */ #define DEBUG_SUBSYSTEM S_LLITE #include #include -#ifdef HAVE_UIDGID_HEADER -# include -#endif +#include + #include #include #include @@ -43,48 +41,130 @@ #include "llite_internal.h" #include "vvp_internal.h" -struct proc_dir_entry *proc_lustre_fs_root; +static struct kobject *llite_kobj; +static struct dentry *llite_root; + +static void llite_kobj_release(struct kobject *kobj) +{ + if (!IS_ERR_OR_NULL(llite_root)) { + debugfs_remove(llite_root); + llite_root = NULL; + } + + kfree(kobj); +} + +static struct kobj_type llite_kobj_ktype = { + .release = llite_kobj_release, + .sysfs_ops = &lustre_sysfs_ops, +}; + +int llite_tunables_register(void) +{ + int rc; + + llite_kobj = kzalloc(sizeof(*llite_kobj), GFP_KERNEL); + if (!llite_kobj) + return -ENOMEM; + + llite_kobj->kset = lustre_kset; + rc = kobject_init_and_add(llite_kobj, &llite_kobj_ktype, + &lustre_kset->kobj, "%s", "llite"); + if (rc) + goto free_kobj; + + llite_root = debugfs_create_dir("llite", debugfs_lustre_root); + return 0; + +free_kobj: + kobject_put(llite_kobj); + llite_kobj = NULL; + + return rc; +} + +void llite_tunables_unregister(void) +{ + kobject_put(llite_kobj); + llite_kobj = NULL; +} -#ifdef CONFIG_PROC_FS -/* /proc/lustre/llite mount point registration */ +/* /lustre/llite mount point registration */ static const struct file_operations ll_rw_extents_stats_fops; static const struct file_operations ll_rw_extents_stats_pp_fops; static const struct file_operations ll_rw_offset_stats_fops; -static __s64 ll_stats_pid_write(const char __user *buf, size_t len); -static int ll_blksize_seq_show(struct seq_file *m, void *v) +/** + * ll_stats_pid_write() - Determine if stats collection should be enabled + * @buf: Buffer containing the data written + * @len: Number of bytes in the buffer + * + * Several proc files begin collecting stats when a value is written, and stop + * collecting when either '0' or 'disable' is written. This function checks the + * written value to see if collection should be enabled or disabled. + * + * Return: If '0' or 'disable' is provided, 0 is returned. If the text + * equivalent of a number is written, that number is returned. Otherwise, + * 1 is returned. Non-zero return values indicate collection should be enabled. + */ +static s64 ll_stats_pid_write(const char __user *buf, size_t len) { - struct super_block *sb = m->private; - struct obd_statfs osfs; + unsigned long long value = 1; + char kernbuf[16]; int rc; - LASSERT(sb != NULL); - rc = ll_statfs_internal(sb, &osfs, OBD_STATFS_NODELAY); - if (!rc) - seq_printf(m, "%u\n", osfs.os_bsize); - return rc; + rc = kstrtoull_from_user(buf, len, 0, &value); + if (rc < 0 && len < sizeof(kernbuf)) { + if (copy_from_user(kernbuf, buf, len)) + return -EFAULT; + kernbuf[len] = 0; + + if (kernbuf[len - 1] == '\n') + kernbuf[len - 1] = 0; + + if (strncasecmp(kernbuf, "disable", 7) == 0) + value = 0; + } + + return value; } -LPROC_SEQ_FOPS_RO(ll_blksize); -static int ll_stat_blksize_seq_show(struct seq_file *m, void *v) +static ssize_t blocksize_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct ll_sb_info *sbi = ll_s2sbi((struct super_block *)m->private); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + struct obd_statfs osfs; + int rc; - seq_printf(m, "%u\n", sbi->ll_stat_blksize); + rc = ll_statfs_internal(sbi, &osfs, OBD_STATFS_NODELAY); + if (rc) + return rc; - return 0; + return sprintf(buf, "%u\n", osfs.os_bsize); } +LUSTRE_RO_ATTR(blocksize); -static ssize_t ll_stat_blksize_seq_write(struct file *file, - const char __user *buffer, - size_t count, loff_t *off) +static ssize_t stat_blocksize_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct seq_file *m = file->private_data; - struct ll_sb_info *sbi = ll_s2sbi((struct super_block *)m->private); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + + return sprintf(buf, "%u\n", sbi->ll_stat_blksize); +} + +static ssize_t stat_blocksize_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); unsigned int val; int rc; - rc = kstrtouint_from_user(buffer, count, 0, &val); + rc = kstrtouint(buffer, 10, &val); if (rc) return rc; @@ -95,160 +175,143 @@ static ssize_t ll_stat_blksize_seq_write(struct file *file, return count; } -LPROC_SEQ_FOPS(ll_stat_blksize); +LUSTRE_RW_ATTR(stat_blocksize); -static int ll_kbytestotal_seq_show(struct seq_file *m, void *v) +static ssize_t kbytestotal_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); struct obd_statfs osfs; + u32 blk_size; + u64 result; int rc; - LASSERT(sb != NULL); - rc = ll_statfs_internal(sb, &osfs, OBD_STATFS_NODELAY); - if (!rc) { - __u32 blk_size = osfs.os_bsize >> 10; - __u64 result = osfs.os_blocks; + rc = ll_statfs_internal(sbi, &osfs, OBD_STATFS_NODELAY); + if (rc) + return rc; + + blk_size = osfs.os_bsize >> 10; + result = osfs.os_blocks; - while (blk_size >>= 1) - result <<= 1; + while (blk_size >>= 1) + result <<= 1; - seq_printf(m, "%llu\n", result); - } - return rc; + return sprintf(buf, "%llu\n", result); } -LPROC_SEQ_FOPS_RO(ll_kbytestotal); +LUSTRE_RO_ATTR(kbytestotal); -static int ll_kbytesfree_seq_show(struct seq_file *m, void *v) +static ssize_t kbytesfree_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); struct obd_statfs osfs; + u32 blk_size; + u64 result; int rc; - LASSERT(sb != NULL); - rc = ll_statfs_internal(sb, &osfs, OBD_STATFS_NODELAY); - if (!rc) { - __u32 blk_size = osfs.os_bsize >> 10; - __u64 result = osfs.os_bfree; + rc = ll_statfs_internal(sbi, &osfs, OBD_STATFS_NODELAY); + if (rc) + return rc; - while (blk_size >>= 1) - result <<= 1; + blk_size = osfs.os_bsize >> 10; + result = osfs.os_bfree; - seq_printf(m, "%llu\n", result); - } - return rc; + while (blk_size >>= 1) + result <<= 1; + + return sprintf(buf, "%llu\n", result); } -LPROC_SEQ_FOPS_RO(ll_kbytesfree); +LUSTRE_RO_ATTR(kbytesfree); -static int ll_kbytesavail_seq_show(struct seq_file *m, void *v) +static ssize_t kbytesavail_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); struct obd_statfs osfs; + u32 blk_size; + u64 result; int rc; - LASSERT(sb != NULL); - rc = ll_statfs_internal(sb, &osfs, OBD_STATFS_NODELAY); - if (!rc) { - __u32 blk_size = osfs.os_bsize >> 10; - __u64 result = osfs.os_bavail; + rc = ll_statfs_internal(sbi, &osfs, OBD_STATFS_NODELAY); + if (rc) + return rc; - while (blk_size >>= 1) - result <<= 1; + blk_size = osfs.os_bsize >> 10; + result = osfs.os_bavail; - seq_printf(m, "%llu\n", result); - } - return rc; + while (blk_size >>= 1) + result <<= 1; + + return sprintf(buf, "%llu\n", result); } -LPROC_SEQ_FOPS_RO(ll_kbytesavail); +LUSTRE_RO_ATTR(kbytesavail); -static int ll_filestotal_seq_show(struct seq_file *m, void *v) +static ssize_t filestotal_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); struct obd_statfs osfs; int rc; - LASSERT(sb != NULL); - rc = ll_statfs_internal(sb, &osfs, OBD_STATFS_NODELAY); - if (!rc) - seq_printf(m, "%llu\n", osfs.os_files); - return rc; + rc = ll_statfs_internal(sbi, &osfs, OBD_STATFS_NODELAY); + if (rc) + return rc; + + return sprintf(buf, "%llu\n", osfs.os_files); } -LPROC_SEQ_FOPS_RO(ll_filestotal); +LUSTRE_RO_ATTR(filestotal); -static int ll_filesfree_seq_show(struct seq_file *m, void *v) +static ssize_t filesfree_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); struct obd_statfs osfs; int rc; - LASSERT(sb != NULL); - rc = ll_statfs_internal(sb, &osfs, OBD_STATFS_NODELAY); - if (!rc) - seq_printf(m, "%llu\n", osfs.os_ffree); - return rc; + rc = ll_statfs_internal(sbi, &osfs, OBD_STATFS_NODELAY); + if (rc) + return rc; + + return sprintf(buf, "%llu\n", osfs.os_ffree); } -LPROC_SEQ_FOPS_RO(ll_filesfree); +LUSTRE_RO_ATTR(filesfree); -static int ll_client_type_seq_show(struct seq_file *m, void *v) +static ssize_t client_type_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct ll_sb_info *sbi = ll_s2sbi((struct super_block *)m->private); - - LASSERT(sbi != NULL); - - seq_puts(m, "local client\n"); - return 0; + return sprintf(buf, "local client\n"); } -LPROC_SEQ_FOPS_RO(ll_client_type); +LUSTRE_RO_ATTR(client_type); -static int ll_fstype_seq_show(struct seq_file *m, void *v) -{ - struct super_block *sb = m->private; +LUSTRE_RW_ATTR(foreign_symlink_enable); - LASSERT(sb != NULL); - seq_printf(m, "%s\n", sb->s_type->name); - return 0; -} -LPROC_SEQ_FOPS_RO(ll_fstype); +LUSTRE_RW_ATTR(foreign_symlink_prefix); -static int ll_sb_uuid_seq_show(struct seq_file *m, void *v) -{ - struct super_block *sb = m->private; +LUSTRE_RW_ATTR(foreign_symlink_upcall); - LASSERT(sb != NULL); - seq_printf(m, "%s\n", ll_s2sbi(sb)->ll_sb_uuid.uuid); - return 0; -} -LPROC_SEQ_FOPS_RO(ll_sb_uuid); +LUSTRE_WO_ATTR(foreign_symlink_upcall_info); -static int ll_xattr_cache_seq_show(struct seq_file *m, void *v) +static ssize_t fstype_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct ll_sb_info *sbi = ll_s2sbi((struct super_block *)m->private); - - seq_printf(m, "%u\n", sbi->ll_xattr_cache_enabled); - return 0; + return sprintf(buf, "lustre\n"); } +LUSTRE_RO_ATTR(fstype); -static ssize_t ll_xattr_cache_seq_write(struct file *file, - const char __user *buffer, - size_t count, loff_t *off) +static ssize_t uuid_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct seq_file *m = file->private_data; - struct ll_sb_info *sbi = ll_s2sbi((struct super_block *)m->private); - bool val; - int rc; - - rc = kstrtobool_from_user(buffer, count, &val); - if (rc) - return rc; - - if (val && !(sbi->ll_flags & LL_SBI_XATTR_CACHE)) - return -ENOTSUPP; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); - sbi->ll_xattr_cache_enabled = val; - sbi->ll_xattr_cache_set = 1; - - return count; + return sprintf(buf, "%s\n", sbi->ll_sb_uuid.uuid); } -LPROC_SEQ_FOPS(ll_xattr_cache); +LUSTRE_RO_ATTR(uuid); static int ll_site_stats_seq_show(struct seq_file *m, void *v) { @@ -260,181 +323,169 @@ static int ll_site_stats_seq_show(struct seq_file *m, void *v) */ return cl_site_stats_print(lu2cl_site(ll_s2sbi(sb)->ll_site), m); } -LPROC_SEQ_FOPS_RO(ll_site_stats); -static int ll_max_readahead_mb_seq_show(struct seq_file *m, void *v) -{ - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); - long pages_number; - int mult; +LDEBUGFS_SEQ_FOPS_RO(ll_site_stats); - spin_lock(&sbi->ll_lock); - pages_number = sbi->ll_ra_info.ra_max_pages; - spin_unlock(&sbi->ll_lock); +static ssize_t max_read_ahead_mb_show(struct kobject *kobj, + struct attribute *attr, char *buf) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); - mult = 1 << (20 - PAGE_SHIFT); - return lprocfs_seq_read_frac_helper(m, pages_number, mult); + return scnprintf(buf, PAGE_SIZE, "%lu\n", + PAGES_TO_MiB(sbi->ll_ra_info.ra_max_pages)); } -static ssize_t -ll_max_readahead_mb_seq_write(struct file *file, const char __user *buffer, - size_t count, loff_t *off) +static ssize_t max_read_ahead_mb_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, size_t count) { - struct seq_file *m = file->private_data; - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); - __s64 pages_number; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + u64 ra_max_mb, pages_number; int rc; - rc = lprocfs_str_with_units_to_s64(buffer, count, &pages_number, 'M'); + rc = sysfs_memparse(buffer, count, &ra_max_mb, "MiB"); if (rc) return rc; - pages_number >>= PAGE_SHIFT; - - if (pages_number < 0 || pages_number > totalram_pages / 2) { + pages_number = round_up(ra_max_mb, 1024 * 1024) >> PAGE_SHIFT; + CDEBUG(D_INFO, "%s: set max_read_ahead_mb=%llu (%llu pages)\n", + sbi->ll_fsname, PAGES_TO_MiB(pages_number), pages_number); + if (pages_number > cfs_totalram_pages() / 2) { /* 1/2 of RAM */ - CERROR("%s: can't set max_readahead_mb=%lu > %luMB\n", - ll_get_fsname(sb, NULL, 0), - (unsigned long)pages_number >> (20 - PAGE_SHIFT), - totalram_pages >> (20 - PAGE_SHIFT + 1)); + CERROR("%s: cannot set max_read_ahead_mb=%llu > totalram/2=%luMB\n", + sbi->ll_fsname, PAGES_TO_MiB(pages_number), + PAGES_TO_MiB(cfs_totalram_pages() / 2)); return -ERANGE; } spin_lock(&sbi->ll_lock); sbi->ll_ra_info.ra_max_pages = pages_number; spin_unlock(&sbi->ll_lock); + return count; } -LPROC_SEQ_FOPS(ll_max_readahead_mb); +LUSTRE_RW_ATTR(max_read_ahead_mb); -static int ll_max_readahead_per_file_mb_seq_show(struct seq_file *m, void *v) +static ssize_t max_read_ahead_per_file_mb_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); - long pages_number; - int mult; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); - spin_lock(&sbi->ll_lock); - pages_number = sbi->ll_ra_info.ra_max_pages_per_file; - spin_unlock(&sbi->ll_lock); - - mult = 1 << (20 - PAGE_SHIFT); - return lprocfs_seq_read_frac_helper(m, pages_number, mult); + return scnprintf(buf, PAGE_SIZE, "%lu\n", + PAGES_TO_MiB(sbi->ll_ra_info.ra_max_pages_per_file)); } -static ssize_t -ll_max_readahead_per_file_mb_seq_write(struct file *file, - const char __user *buffer, - size_t count, loff_t *off) +static ssize_t max_read_ahead_per_file_mb_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct seq_file *m = file->private_data; - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + u64 ra_max_file_mb, pages_number; int rc; - __s64 pages_number; - rc = lprocfs_str_with_units_to_s64(buffer, count, &pages_number, 'M'); + rc = sysfs_memparse(buffer, count, &ra_max_file_mb, "MiB"); if (rc) return rc; - pages_number >>= PAGE_SHIFT; - - if (pages_number < 0 || pages_number > sbi->ll_ra_info.ra_max_pages) { - CERROR("%s: can't set max_readahead_per_file_mb=%lu > " - "max_read_ahead_mb=%lu\n", ll_get_fsname(sb, NULL, 0), - (unsigned long)pages_number >> (20 - PAGE_SHIFT), - sbi->ll_ra_info.ra_max_pages >> (20 - PAGE_SHIFT)); + pages_number = round_up(ra_max_file_mb, 1024 * 1024) >> PAGE_SHIFT; + if (pages_number > sbi->ll_ra_info.ra_max_pages) { + CERROR("%s: cannot set max_read_ahead_per_file_mb=%llu > max_read_ahead_mb=%lu\n", + sbi->ll_fsname, PAGES_TO_MiB(pages_number), + PAGES_TO_MiB(sbi->ll_ra_info.ra_max_pages)); return -ERANGE; } spin_lock(&sbi->ll_lock); sbi->ll_ra_info.ra_max_pages_per_file = pages_number; spin_unlock(&sbi->ll_lock); + return count; } -LPROC_SEQ_FOPS(ll_max_readahead_per_file_mb); +LUSTRE_RW_ATTR(max_read_ahead_per_file_mb); -static int ll_max_read_ahead_whole_mb_seq_show(struct seq_file *m, void *v) +static ssize_t max_read_ahead_whole_mb_show(struct kobject *kobj, + struct attribute *attr, char *buf) { - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); - long pages_number; - int mult; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); - spin_lock(&sbi->ll_lock); - pages_number = sbi->ll_ra_info.ra_max_read_ahead_whole_pages; - spin_unlock(&sbi->ll_lock); - - mult = 1 << (20 - PAGE_SHIFT); - return lprocfs_seq_read_frac_helper(m, pages_number, mult); + return scnprintf(buf, PAGE_SIZE, "%lu\n", + PAGES_TO_MiB(sbi->ll_ra_info.ra_max_read_ahead_whole_pages)); } -static ssize_t -ll_max_read_ahead_whole_mb_seq_write(struct file *file, - const char __user *buffer, - size_t count, loff_t *off) +static ssize_t max_read_ahead_whole_mb_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, size_t count) { - struct seq_file *m = file->private_data; - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + u64 ra_max_whole_mb, pages_number; int rc; - __s64 pages_number; - rc = lprocfs_str_with_units_to_s64(buffer, count, &pages_number, 'M'); + rc = sysfs_memparse(buffer, count, &ra_max_whole_mb, "MiB"); if (rc) return rc; - pages_number >>= PAGE_SHIFT; - + pages_number = round_up(ra_max_whole_mb, 1024 * 1024) >> PAGE_SHIFT; /* Cap this at the current max readahead window size, the readahead - * algorithm does this anyway so it's pointless to set it larger. */ - if (pages_number < 0 || - pages_number > sbi->ll_ra_info.ra_max_pages_per_file) { - int pages_shift = 20 - PAGE_SHIFT; - CERROR("%s: can't set max_read_ahead_whole_mb=%lu > " - "max_read_ahead_per_file_mb=%lu\n", - ll_get_fsname(sb, NULL, 0), - (unsigned long)pages_number >> pages_shift, - sbi->ll_ra_info.ra_max_pages_per_file >> pages_shift); + * algorithm does this anyway so it's pointless to set it larger. + */ + if (pages_number > sbi->ll_ra_info.ra_max_pages_per_file) { + CERROR("%s: cannot set max_read_ahead_whole_mb=%llu > max_read_ahead_per_file_mb=%lu\n", + sbi->ll_fsname, PAGES_TO_MiB(pages_number), + PAGES_TO_MiB(sbi->ll_ra_info.ra_max_pages_per_file)); + return -ERANGE; } spin_lock(&sbi->ll_lock); sbi->ll_ra_info.ra_max_read_ahead_whole_pages = pages_number; spin_unlock(&sbi->ll_lock); + return count; } -LPROC_SEQ_FOPS(ll_max_read_ahead_whole_mb); +LUSTRE_RW_ATTR(max_read_ahead_whole_mb); static int ll_max_cached_mb_seq_show(struct seq_file *m, void *v) { struct super_block *sb = m->private; struct ll_sb_info *sbi = ll_s2sbi(sb); struct cl_client_cache *cache = sbi->ll_cache; - int shift = 20 - PAGE_SHIFT; + struct ll_ra_info *ra = &sbi->ll_ra_info; long max_cached_mb; long unused_mb; - max_cached_mb = cache->ccc_lru_max >> shift; - unused_mb = atomic_long_read(&cache->ccc_lru_left) >> shift; + mutex_lock(&cache->ccc_max_cache_mb_lock); + max_cached_mb = PAGES_TO_MiB(cache->ccc_lru_max); + unused_mb = PAGES_TO_MiB(atomic_long_read(&cache->ccc_lru_left)); + mutex_unlock(&cache->ccc_max_cache_mb_lock); + seq_printf(m, "users: %d\n" - "max_cached_mb: %ld\n" - "used_mb: %ld\n" - "unused_mb: %ld\n" - "reclaim_count: %u\n", + "max_cached_mb: %ld\n" + "used_mb: %ld\n" + "unused_mb: %ld\n" + "reclaim_count: %u\n" + "max_read_ahead_mb: %lu\n" + "used_read_ahead_mb: %d\n", atomic_read(&cache->ccc_users), max_cached_mb, max_cached_mb - unused_mb, unused_mb, - cache->ccc_lru_shrinkers); + cache->ccc_lru_shrinkers, + PAGES_TO_MiB(ra->ra_max_pages), + PAGES_TO_MiB(atomic_read(&ra->ra_cur_pages))); return 0; } -static ssize_t -ll_max_cached_mb_seq_write(struct file *file, const char __user *buffer, - size_t count, loff_t *off) +static ssize_t ll_max_cached_mb_seq_write(struct file *file, + const char __user *buffer, + size_t count, loff_t *off) { struct seq_file *m = file->private_data; struct super_block *sb = m->private; @@ -444,38 +495,36 @@ ll_max_cached_mb_seq_write(struct file *file, const char __user *buffer, long diff = 0; long nrpages = 0; __u16 refcheck; - __s64 pages_number; - long rc; - char kernbuf[128]; - ENTRY; + u64 pages_number; + int rc; + char kernbuf[128], *ptr; + ENTRY; if (count >= sizeof(kernbuf)) RETURN(-EINVAL); if (copy_from_user(kernbuf, buffer, count)) RETURN(-EFAULT); - kernbuf[count] = 0; + kernbuf[count] = '\0'; - buffer += lprocfs_find_named_value(kernbuf, "max_cached_mb:", &count) - - kernbuf; - rc = lprocfs_str_with_units_to_s64(buffer, count, &pages_number, 'M'); + ptr = lprocfs_find_named_value(kernbuf, "max_cached_mb:", &count); + rc = sysfs_memparse(ptr, count, &pages_number, "MiB"); if (rc) RETURN(rc); pages_number >>= PAGE_SHIFT; - if (pages_number < 0 || pages_number > totalram_pages) { + if (pages_number < 0 || pages_number > cfs_totalram_pages()) { CERROR("%s: can't set max cache more than %lu MB\n", - ll_get_fsname(sb, NULL, 0), - totalram_pages >> (20 - PAGE_SHIFT)); + sbi->ll_fsname, + PAGES_TO_MiB(cfs_totalram_pages())); RETURN(-ERANGE); } /* Allow enough cache so clients can make well-formed RPCs */ pages_number = max_t(long, pages_number, PTLRPC_MAX_BRW_PAGES); - spin_lock(&sbi->ll_lock); + mutex_lock(&cache->ccc_max_cache_mb_lock); diff = pages_number - cache->ccc_lru_max; - spin_unlock(&sbi->ll_lock); /* easy - add more LRU slots. */ if (diff >= 0) { @@ -485,7 +534,7 @@ ll_max_cached_mb_seq_write(struct file *file, const char __user *buffer, env = cl_env_get(&refcheck); if (IS_ERR(env)) - RETURN(rc); + GOTO(out_unlock, rc = PTR_ERR(env)); diff = -diff; while (diff > 0) { @@ -493,17 +542,21 @@ ll_max_cached_mb_seq_write(struct file *file, const char __user *buffer, /* reduce LRU budget from free slots. */ do { - long ov, nv; + long lru_left_old, lru_left_new, lru_left_ret; - ov = atomic_long_read(&cache->ccc_lru_left); - if (ov == 0) + lru_left_old = atomic_long_read(&cache->ccc_lru_left); + if (lru_left_old == 0) break; - nv = ov > diff ? ov - diff : 0; - rc = atomic_long_cmpxchg(&cache->ccc_lru_left, ov, nv); - if (likely(ov == rc)) { - diff -= ov - nv; - nrpages += ov - nv; + lru_left_new = lru_left_old > diff ? + lru_left_old - diff : 0; + lru_left_ret = + atomic_long_cmpxchg(&cache->ccc_lru_left, + lru_left_old, + lru_left_new); + if (likely(lru_left_old == lru_left_ret)) { + diff -= lru_left_old - lru_left_new; + nrpages += lru_left_old - lru_left_new; break; } } while (1); @@ -516,8 +569,11 @@ ll_max_cached_mb_seq_write(struct file *file, const char __user *buffer, break; } + /* Request extra free slots to avoid them all being used + * by other processes before this can continue shrinking. + */ + tmp = diff + min_t(long, diff, MiB_TO_PAGES(1024)); /* difficult - have to ask OSCs to drop LRU slots. */ - tmp = diff << 1; rc = obd_set_info_async(env, sbi->ll_dt_exp, sizeof(KEY_CACHE_LRU_SHRINK), KEY_CACHE_LRU_SHRINK, @@ -529,32 +585,31 @@ ll_max_cached_mb_seq_write(struct file *file, const char __user *buffer, out: if (rc >= 0) { - spin_lock(&sbi->ll_lock); cache->ccc_lru_max = pages_number; - spin_unlock(&sbi->ll_lock); rc = count; } else { atomic_long_add(nrpages, &cache->ccc_lru_left); } +out_unlock: + mutex_unlock(&cache->ccc_max_cache_mb_lock); return rc; } -LPROC_SEQ_FOPS(ll_max_cached_mb); +LDEBUGFS_SEQ_FOPS(ll_max_cached_mb); -static int ll_checksum_seq_show(struct seq_file *m, void *v) +static ssize_t checksums_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); - seq_printf(m, "%u\n", (sbi->ll_flags & LL_SBI_CHECKSUM) ? 1 : 0); - return 0; + return sprintf(buf, "%u\n", (sbi->ll_flags & LL_SBI_CHECKSUM) ? 1 : 0); } -static ssize_t ll_checksum_seq_write(struct file *file, - const char __user *buffer, - size_t count, loff_t *off) +static ssize_t checksums_store(struct kobject *kobj, struct attribute *attr, + const char *buffer, size_t count) { - struct seq_file *m = file->private_data; - struct ll_sb_info *sbi = ll_s2sbi((struct super_block *)m->private); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); bool val; int tmp; int rc; @@ -563,7 +618,7 @@ static ssize_t ll_checksum_seq_write(struct file *file, /* Not set up yet */ return -EAGAIN; - rc = kstrtobool_from_user(buffer, count, &val); + rc = kstrtobool(buffer, &val); if (rc) return rc; if (val) @@ -579,137 +634,184 @@ static ssize_t ll_checksum_seq_write(struct file *file, return count; } -LPROC_SEQ_FOPS(ll_checksum); +LUSTRE_RW_ATTR(checksums); -static int ll_rd_track_id(struct seq_file *m, enum stats_track_type type) +LUSTRE_ATTR(checksum_pages, 0644, checksums_show, checksums_store); + +static ssize_t ll_rd_track_id(struct kobject *kobj, char *buf, + enum stats_track_type type) { - struct super_block *sb = m->private; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); - if (ll_s2sbi(sb)->ll_stats_track_type == type) { - seq_printf(m, "%d\n", - ll_s2sbi(sb)->ll_stats_track_id); - } else if (ll_s2sbi(sb)->ll_stats_track_type == STATS_TRACK_ALL) { - seq_puts(m, "0 (all)\n"); - } else { - seq_puts(m, "untracked\n"); - } - return 0; + if (sbi->ll_stats_track_type == type) + return sprintf(buf, "%d\n", sbi->ll_stats_track_id); + else if (sbi->ll_stats_track_type == STATS_TRACK_ALL) + return sprintf(buf, "0 (all)\n"); + + return sprintf(buf, "untracked\n"); } -static int ll_wr_track_id(const char __user *buffer, unsigned long count, - void *data, enum stats_track_type type) +static ssize_t ll_wr_track_id(struct kobject *kobj, const char *buffer, + size_t count, enum stats_track_type type) { - struct super_block *sb = data; - unsigned int pid; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + unsigned long pid; int rc; - rc = kstrtouint_from_user(buffer, count, 0, &pid); + rc = kstrtoul(buffer, 10, &pid); if (rc) return rc; - ll_s2sbi(sb)->ll_stats_track_id = pid; + sbi->ll_stats_track_id = pid; if (pid == 0) - ll_s2sbi(sb)->ll_stats_track_type = STATS_TRACK_ALL; + sbi->ll_stats_track_type = STATS_TRACK_ALL; else - ll_s2sbi(sb)->ll_stats_track_type = type; - lprocfs_clear_stats(ll_s2sbi(sb)->ll_stats); + sbi->ll_stats_track_type = type; + lprocfs_clear_stats(sbi->ll_stats); return count; } -static int ll_track_pid_seq_show(struct seq_file *m, void *v) +static ssize_t stats_track_pid_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - return ll_rd_track_id(m, STATS_TRACK_PID); + return ll_rd_track_id(kobj, buf, STATS_TRACK_PID); } -static ssize_t ll_track_pid_seq_write(struct file *file, - const char __user *buffer, - size_t count, loff_t *off) +static ssize_t stats_track_pid_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct seq_file *seq = file->private_data; - return ll_wr_track_id(buffer, count, seq->private, STATS_TRACK_PID); + return ll_wr_track_id(kobj, buffer, count, STATS_TRACK_PID); } -LPROC_SEQ_FOPS(ll_track_pid); +LUSTRE_RW_ATTR(stats_track_pid); -static int ll_track_ppid_seq_show(struct seq_file *m, void *v) +static ssize_t stats_track_ppid_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - return ll_rd_track_id(m, STATS_TRACK_PPID); + return ll_rd_track_id(kobj, buf, STATS_TRACK_PPID); } -static ssize_t ll_track_ppid_seq_write(struct file *file, - const char __user *buffer, - size_t count, loff_t *off) +static ssize_t stats_track_ppid_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct seq_file *seq = file->private_data; - return ll_wr_track_id(buffer, count, seq->private, STATS_TRACK_PPID); + return ll_wr_track_id(kobj, buffer, count, STATS_TRACK_PPID); } -LPROC_SEQ_FOPS(ll_track_ppid); +LUSTRE_RW_ATTR(stats_track_ppid); -static int ll_track_gid_seq_show(struct seq_file *m, void *v) +static ssize_t stats_track_gid_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - return ll_rd_track_id(m, STATS_TRACK_GID); + return ll_rd_track_id(kobj, buf, STATS_TRACK_GID); } -static ssize_t ll_track_gid_seq_write(struct file *file, - const char __user *buffer, - size_t count, loff_t *off) +static ssize_t stats_track_gid_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct seq_file *seq = file->private_data; - return ll_wr_track_id(buffer, count, seq->private, STATS_TRACK_GID); + return ll_wr_track_id(kobj, buffer, count, STATS_TRACK_GID); } -LPROC_SEQ_FOPS(ll_track_gid); +LUSTRE_RW_ATTR(stats_track_gid); -static int ll_statahead_max_seq_show(struct seq_file *m, void *v) +static ssize_t statahead_running_max_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); - seq_printf(m, "%u\n", sbi->ll_sa_max); - return 0; + return scnprintf(buf, PAGE_SIZE, "%u\n", sbi->ll_sa_running_max); } -static ssize_t ll_statahead_max_seq_write(struct file *file, - const char __user *buffer, - size_t count, loff_t *off) +static ssize_t statahead_running_max_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct seq_file *m = file->private_data; - struct ll_sb_info *sbi = ll_s2sbi((struct super_block *)m->private); - unsigned int val; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + unsigned long val; int rc; - rc = kstrtouint_from_user(buffer, count, 0, &val); + rc = kstrtoul(buffer, 0, &val); + if (rc) + return rc; + + if (val <= LL_SA_RUNNING_MAX) { + sbi->ll_sa_running_max = val; + return count; + } + + CERROR("Bad statahead_running_max value %lu. Valid values " + "are in the range [0, %d]\n", val, LL_SA_RUNNING_MAX); + + return -ERANGE; +} +LUSTRE_RW_ATTR(statahead_running_max); + +static ssize_t statahead_max_show(struct kobject *kobj, + struct attribute *attr, + char *buf) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + + return sprintf(buf, "%u\n", sbi->ll_sa_max); +} + +static ssize_t statahead_max_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + unsigned long val; + int rc; + + rc = kstrtoul(buffer, 0, &val); if (rc) return rc; if (val <= LL_SA_RPC_MAX) sbi->ll_sa_max = val; else - CERROR("Bad statahead_max value %u. Valid values are in " - "the range [0, %d]\n", val, LL_SA_RPC_MAX); + CERROR("Bad statahead_max value %lu. Valid values are in the range [0, %d]\n", + val, LL_SA_RPC_MAX); return count; } -LPROC_SEQ_FOPS(ll_statahead_max); +LUSTRE_RW_ATTR(statahead_max); -static int ll_statahead_agl_seq_show(struct seq_file *m, void *v) +static ssize_t statahead_agl_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); - seq_printf(m, "%u\n", - sbi->ll_flags & LL_SBI_AGL_ENABLED ? 1 : 0); - return 0; + return sprintf(buf, "%u\n", sbi->ll_flags & LL_SBI_AGL_ENABLED ? 1 : 0); } -static ssize_t ll_statahead_agl_seq_write(struct file *file, - const char __user *buffer, - size_t count, loff_t *off) +static ssize_t statahead_agl_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct seq_file *m = file->private_data; - struct ll_sb_info *sbi = ll_s2sbi((struct super_block *)m->private); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); bool val; int rc; - rc = kstrtobool_from_user(buffer, count, &val); + rc = kstrtobool(buffer, &val); if (rc) return rc; @@ -720,7 +822,7 @@ static ssize_t ll_statahead_agl_seq_write(struct file *file, return count; } -LPROC_SEQ_FOPS(ll_statahead_agl); +LUSTRE_RW_ATTR(statahead_agl); static int ll_statahead_stats_seq_show(struct seq_file *m, void *v) { @@ -728,35 +830,37 @@ static int ll_statahead_stats_seq_show(struct seq_file *m, void *v) struct ll_sb_info *sbi = ll_s2sbi(sb); seq_printf(m, "statahead total: %u\n" - "statahead wrong: %u\n" - "agl total: %u\n", - atomic_read(&sbi->ll_sa_total), - atomic_read(&sbi->ll_sa_wrong), - atomic_read(&sbi->ll_agl_total)); + "statahead wrong: %u\n" + "agl total: %u\n", + atomic_read(&sbi->ll_sa_total), + atomic_read(&sbi->ll_sa_wrong), + atomic_read(&sbi->ll_agl_total)); return 0; } -LPROC_SEQ_FOPS_RO(ll_statahead_stats); -static int ll_lazystatfs_seq_show(struct seq_file *m, void *v) +LDEBUGFS_SEQ_FOPS_RO(ll_statahead_stats); + +static ssize_t lazystatfs_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); - seq_printf(m, "%u\n", - (sbi->ll_flags & LL_SBI_LAZYSTATFS) ? 1 : 0); - return 0; + return sprintf(buf, "%u\n", (sbi->ll_flags & LL_SBI_LAZYSTATFS) ? 1 : 0); } -static ssize_t ll_lazystatfs_seq_write(struct file *file, - const char __user *buffer, - size_t count, loff_t *off) +static ssize_t lazystatfs_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct seq_file *m = file->private_data; - struct ll_sb_info *sbi = ll_s2sbi((struct super_block *)m->private); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); bool val; int rc; - rc = kstrtobool_from_user(buffer, count, &val); + rc = kstrtobool(buffer, &val); if (rc) return rc; @@ -767,12 +871,44 @@ static ssize_t ll_lazystatfs_seq_write(struct file *file, return count; } -LPROC_SEQ_FOPS(ll_lazystatfs); +LUSTRE_RW_ATTR(lazystatfs); -static int ll_max_easize_seq_show(struct seq_file *m, void *v) +static ssize_t statfs_max_age_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + + return scnprintf(buf, PAGE_SIZE, "%u\n", sbi->ll_statfs_max_age); +} + +static ssize_t statfs_max_age_store(struct kobject *kobj, + struct attribute *attr, const char *buffer, + size_t count) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + unsigned int val; + int rc; + + rc = kstrtouint(buffer, 10, &val); + if (rc) + return rc; + if (val > OBD_STATFS_CACHE_MAX_AGE) + return -EINVAL; + + sbi->ll_statfs_max_age = val; + + return count; +} +LUSTRE_RW_ATTR(statfs_max_age); + +static ssize_t max_easize_show(struct kobject *kobj, + struct attribute *attr, + char *buf) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); unsigned int ealen; int rc; @@ -780,10 +916,11 @@ static int ll_max_easize_seq_show(struct seq_file *m, void *v) if (rc) return rc; - seq_printf(m, "%u\n", ealen); - return 0; + /* Limit xattr size returned to userspace based on kernel maximum */ + return scnprintf(buf, PAGE_SIZE, "%u\n", + ealen > XATTR_SIZE_MAX ? XATTR_SIZE_MAX : ealen); } -LPROC_SEQ_FOPS_RO(ll_max_easize); +LUSTRE_RO_ATTR(max_easize); /** * Get default_easize. @@ -796,10 +933,12 @@ LPROC_SEQ_FOPS_RO(ll_max_easize); * \retval 0 on success * \retval negative negated errno on failure */ -static int ll_default_easize_seq_show(struct seq_file *m, void *v) +static ssize_t default_easize_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); unsigned int ealen; int rc; @@ -807,8 +946,9 @@ static int ll_default_easize_seq_show(struct seq_file *m, void *v) if (rc) return rc; - seq_printf(m, "%u\n", ealen); - return 0; + /* Limit xattr size returned to userspace based on kernel maximum */ + return scnprintf(buf, PAGE_SIZE, "%u\n", + ealen > XATTR_SIZE_MAX ? XATTR_SIZE_MAX : ealen); } /** @@ -827,20 +967,20 @@ static int ll_default_easize_seq_show(struct seq_file *m, void *v) * \retval positive \a count on success * \retval negative negated errno on failure */ -static ssize_t ll_default_easize_seq_write(struct file *file, - const char __user *buffer, - size_t count, loff_t *unused) +static ssize_t default_easize_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct seq_file *seq = file->private_data; - struct super_block *sb = (struct super_block *)seq->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); unsigned int val; int rc; if (count == 0) return 0; - rc = kstrtouint_from_user(buffer, count, 0, &val); + rc = kstrtouint(buffer, 10, &val); if (rc) return rc; @@ -850,11 +990,11 @@ static ssize_t ll_default_easize_seq_write(struct file *file, return count; } -LPROC_SEQ_FOPS(ll_default_easize); +LUSTRE_RW_ATTR(default_easize); static int ll_sbi_flags_seq_show(struct seq_file *m, void *v) { - const char *str[] = LL_SBI_FLAGS; + const char *const str[] = LL_SBI_FLAGS; struct super_block *sb = m->private; int flags = ll_s2sbi(sb)->ll_flags; int i = 0; @@ -862,7 +1002,7 @@ static int ll_sbi_flags_seq_show(struct seq_file *m, void *v) while (flags != 0) { if (ARRAY_SIZE(str) <= i) { CERROR("%s: Revise array LL_SBI_FLAGS to match sbi " - "flags please.\n", ll_get_fsname(sb, NULL, 0)); + "flags please.\n", ll_s2sbi(sb)->ll_fsname); return -EINVAL; } @@ -874,27 +1014,64 @@ static int ll_sbi_flags_seq_show(struct seq_file *m, void *v) seq_printf(m, "\b\n"); return 0; } -LPROC_SEQ_FOPS_RO(ll_sbi_flags); -static int ll_tiny_write_seq_show(struct seq_file *m, void *v) +LDEBUGFS_SEQ_FOPS_RO(ll_sbi_flags); + +static ssize_t xattr_cache_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); - seq_printf(m, "%u\n", !!(sbi->ll_flags & LL_SBI_TINY_WRITE)); - return 0; + return sprintf(buf, "%u\n", sbi->ll_xattr_cache_enabled); } -static ssize_t ll_tiny_write_seq_write( - struct file *file, const char __user *buffer, size_t count, loff_t *off) +static ssize_t xattr_cache_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct seq_file *m = file->private_data; - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); bool val; int rc; - rc = kstrtobool_from_user(buffer, count, &val); + rc = kstrtobool(buffer, &val); + if (rc) + return rc; + + if (val && !(sbi->ll_flags & LL_SBI_XATTR_CACHE)) + return -ENOTSUPP; + + sbi->ll_xattr_cache_enabled = val; + sbi->ll_xattr_cache_set = 1; + + return count; +} +LUSTRE_RW_ATTR(xattr_cache); + +static ssize_t tiny_write_show(struct kobject *kobj, + struct attribute *attr, + char *buf) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + + return sprintf(buf, "%u\n", !!(sbi->ll_flags & LL_SBI_TINY_WRITE)); +} + +static ssize_t tiny_write_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + bool val; + int rc; + + rc = kstrtobool(buffer, &val); if (rc) return rc; @@ -907,28 +1084,160 @@ static ssize_t ll_tiny_write_seq_write( return count; } -LPROC_SEQ_FOPS(ll_tiny_write); +LUSTRE_RW_ATTR(tiny_write); -static int ll_fast_read_seq_show(struct seq_file *m, void *v) +static ssize_t max_read_ahead_async_active_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); - seq_printf(m, "%u\n", !!(sbi->ll_flags & LL_SBI_FAST_READ)); - return 0; + return scnprintf(buf, PAGE_SIZE, "%u\n", + sbi->ll_ra_info.ra_async_max_active); +} + +static ssize_t max_read_ahead_async_active_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + unsigned int val; + int rc; + + rc = kstrtouint(buffer, 10, &val); + if (rc) + return rc; + + /** + * It doesn't make any sense to make it exceed what + * workqueue could acutally support. This can easily + * over subscripe the cores but Lustre internally + * throttles to avoid those impacts. + */ + if (val > WQ_UNBOUND_MAX_ACTIVE) { + CERROR("%s: cannot set max_read_ahead_async_active=%u larger than %u\n", + sbi->ll_fsname, val, WQ_UNBOUND_MAX_ACTIVE); + return -ERANGE; + } + + spin_lock(&sbi->ll_lock); + sbi->ll_ra_info.ra_async_max_active = val; + spin_unlock(&sbi->ll_lock); + + return count; +} +LUSTRE_RW_ATTR(max_read_ahead_async_active); + +static ssize_t read_ahead_async_file_threshold_mb_show(struct kobject *kobj, + struct attribute *attr, + char *buf) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + + return scnprintf(buf, PAGE_SIZE, "%lu\n", PAGES_TO_MiB( + sbi->ll_ra_info.ra_async_pages_per_file_threshold)); } static ssize_t -ll_fast_read_seq_write(struct file *file, const char __user *buffer, - size_t count, loff_t *off) +read_ahead_async_file_threshold_mb_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, size_t count) +{ + unsigned long pages_number; + unsigned long max_ra_per_file; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + int rc; + + rc = kstrtoul(buffer, 10, &pages_number); + if (rc) + return rc; + + pages_number = MiB_TO_PAGES(pages_number); + max_ra_per_file = sbi->ll_ra_info.ra_max_pages_per_file; + if (pages_number < 0 || pages_number > max_ra_per_file) { + CERROR("%s: can't set read_ahead_async_file_threshold_mb=%lu > " + "max_read_readahead_per_file_mb=%lu\n", sbi->ll_fsname, + PAGES_TO_MiB(pages_number), + PAGES_TO_MiB(max_ra_per_file)); + return -ERANGE; + } + sbi->ll_ra_info.ra_async_pages_per_file_threshold = pages_number; + + return count; +} +LUSTRE_RW_ATTR(read_ahead_async_file_threshold_mb); + +static ssize_t read_ahead_range_kb_show(struct kobject *kobj, + struct attribute *attr,char *buf) { - struct seq_file *m = file->private_data; - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + + return snprintf(buf, PAGE_SIZE, "%lu\n", + sbi->ll_ra_info.ra_range_pages << (PAGE_SHIFT - 10)); +} + +static ssize_t +read_ahead_range_kb_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, size_t count) +{ + unsigned long pages_number; + unsigned long max_ra_per_file; + u64 val; + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + int rc; + + rc = sysfs_memparse(buffer, count, &val, "KiB"); + if (rc < 0) + return rc; + + pages_number = val >> PAGE_SHIFT; + /* Disable mmap range read */ + if (pages_number == 0) + goto out; + + max_ra_per_file = sbi->ll_ra_info.ra_max_pages_per_file; + if (pages_number > max_ra_per_file || + pages_number < RA_MIN_MMAP_RANGE_PAGES) + return -ERANGE; + +out: + spin_lock(&sbi->ll_lock); + sbi->ll_ra_info.ra_range_pages = pages_number; + spin_unlock(&sbi->ll_lock); + + return count; +} +LUSTRE_RW_ATTR(read_ahead_range_kb); + +static ssize_t fast_read_show(struct kobject *kobj, + struct attribute *attr, + char *buf) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + + return sprintf(buf, "%u\n", !!(sbi->ll_flags & LL_SBI_FAST_READ)); +} + +static ssize_t fast_read_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); bool val; int rc; - rc = kstrtobool_from_user(buffer, count, &val); + rc = kstrtobool(buffer, &val); if (rc) return rc; @@ -941,40 +1250,110 @@ ll_fast_read_seq_write(struct file *file, const char __user *buffer, return count; } -LPROC_SEQ_FOPS(ll_fast_read); +LUSTRE_RW_ATTR(fast_read); -static int ll_pio_seq_show(struct seq_file *m, void *v) +static ssize_t file_heat_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); - seq_printf(m, "%u\n", !!(sbi->ll_flags & LL_SBI_PIO)); - return 0; + return scnprintf(buf, PAGE_SIZE, "%u\n", + !!(sbi->ll_flags & LL_SBI_FILE_HEAT)); } -static ssize_t ll_pio_seq_write(struct file *file, const char __user *buffer, - size_t count, loff_t *off) +static ssize_t file_heat_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct seq_file *m = file->private_data; - struct super_block *sb = m->private; - struct ll_sb_info *sbi = ll_s2sbi(sb); + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); bool val; int rc; - rc = kstrtobool_from_user(buffer, count, &val); + rc = kstrtobool(buffer, &val); if (rc) return rc; spin_lock(&sbi->ll_lock); if (val) - sbi->ll_flags |= LL_SBI_PIO; + sbi->ll_flags |= LL_SBI_FILE_HEAT; else - sbi->ll_flags &= ~LL_SBI_PIO; + sbi->ll_flags &= ~LL_SBI_FILE_HEAT; spin_unlock(&sbi->ll_lock); return count; } -LPROC_SEQ_FOPS(ll_pio); +LUSTRE_RW_ATTR(file_heat); + +static ssize_t heat_decay_percentage_show(struct kobject *kobj, + struct attribute *attr, + char *buf) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + + return scnprintf(buf, PAGE_SIZE, "%u\n", + (sbi->ll_heat_decay_weight * 100 + 128) / 256); +} + +static ssize_t heat_decay_percentage_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + unsigned long val; + int rc; + + rc = kstrtoul(buffer, 10, &val); + if (rc) + return rc; + + if (val < 0 || val > 100) + return -ERANGE; + + sbi->ll_heat_decay_weight = (val * 256 + 50) / 100; + + return count; +} +LUSTRE_RW_ATTR(heat_decay_percentage); + +static ssize_t heat_period_second_show(struct kobject *kobj, + struct attribute *attr, + char *buf) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + + return scnprintf(buf, PAGE_SIZE, "%u\n", sbi->ll_heat_period_second); +} + +static ssize_t heat_period_second_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + unsigned long val; + int rc; + + rc = kstrtoul(buffer, 10, &val); + if (rc) + return rc; + + if (val <= 0) + return -ERANGE; + + sbi->ll_heat_period_second = val; + + return count; +} +LUSTRE_RW_ATTR(heat_period_second); static int ll_unstable_stats_seq_show(struct seq_file *m, void *v) { @@ -988,8 +1367,8 @@ static int ll_unstable_stats_seq_show(struct seq_file *m, void *v) mb = (pages * PAGE_SIZE) >> 20; seq_printf(m, "unstable_check: %8d\n" - "unstable_pages: %12ld\n" - "unstable_mb: %8d\n", + "unstable_pages: %12ld\n" + "unstable_mb: %8d\n", cache->ccc_unstable_check, pages, mb); return 0; } @@ -1026,7 +1405,8 @@ static ssize_t ll_unstable_stats_seq_write(struct file *file, return count; } -LPROC_SEQ_FOPS(ll_unstable_stats); + +LDEBUGFS_SEQ_FOPS(ll_unstable_stats); static int ll_root_squash_seq_show(struct seq_file *m, void *v) { @@ -1047,10 +1427,10 @@ static ssize_t ll_root_squash_seq_write(struct file *file, struct ll_sb_info *sbi = ll_s2sbi(sb); struct root_squash_info *squash = &sbi->ll_squash; - return lprocfs_wr_root_squash(buffer, count, squash, - ll_get_fsname(sb, NULL, 0)); + return lprocfs_wr_root_squash(buffer, count, squash, sbi->ll_fsname); } -LPROC_SEQ_FOPS(ll_root_squash); + +LDEBUGFS_SEQ_FOPS(ll_root_squash); static int ll_nosquash_nids_seq_show(struct seq_file *m, void *v) { @@ -1059,7 +1439,7 @@ static int ll_nosquash_nids_seq_show(struct seq_file *m, void *v) struct root_squash_info *squash = &sbi->ll_squash; int len; - down_read(&squash->rsi_sem); + spin_lock(&squash->rsi_lock); if (!list_empty(&squash->rsi_nosquash_nids)) { len = cfs_print_nidlist(m->buf + m->count, m->size - m->count, &squash->rsi_nosquash_nids); @@ -1068,7 +1448,7 @@ static int ll_nosquash_nids_seq_show(struct seq_file *m, void *v) } else { seq_puts(m, "NONE\n"); } - up_read(&squash->rsi_sem); + spin_unlock(&squash->rsi_lock); return 0; } @@ -1083,8 +1463,7 @@ static ssize_t ll_nosquash_nids_seq_write(struct file *file, struct root_squash_info *squash = &sbi->ll_squash; int rc; - rc = lprocfs_wr_nosquash_nids(buffer, count, squash, - ll_get_fsname(sb, NULL, 0)); + rc = lprocfs_wr_nosquash_nids(buffer, count, squash, sbi->ll_fsname); if (rc < 0) return rc; @@ -1092,142 +1471,180 @@ static ssize_t ll_nosquash_nids_seq_write(struct file *file, return rc; } -LPROC_SEQ_FOPS(ll_nosquash_nids); -struct lprocfs_vars lprocfs_llite_obd_vars[] = { - { .name = "uuid", - .fops = &ll_sb_uuid_fops }, - { .name = "fstype", - .fops = &ll_fstype_fops }, +LDEBUGFS_SEQ_FOPS(ll_nosquash_nids); + +static int ll_pcc_seq_show(struct seq_file *m, void *v) +{ + struct super_block *sb = m->private; + struct ll_sb_info *sbi = ll_s2sbi(sb); + + return pcc_super_dump(&sbi->ll_pcc_super, m); +} + +static ssize_t ll_pcc_seq_write(struct file *file, const char __user *buffer, + size_t count, loff_t *off) +{ + struct seq_file *m = file->private_data; + struct super_block *sb = m->private; + struct ll_sb_info *sbi = ll_s2sbi(sb); + int rc; + char *kernbuf; + + if (count >= LPROCFS_WR_PCC_MAX_CMD) + return -EINVAL; + + if (!(exp_connect_flags2(sbi->ll_md_exp) & OBD_CONNECT2_PCC)) + return -EOPNOTSUPP; + + OBD_ALLOC(kernbuf, count + 1); + if (kernbuf == NULL) + return -ENOMEM; + + if (copy_from_user(kernbuf, buffer, count)) + GOTO(out_free_kernbuff, rc = -EFAULT); + + rc = pcc_cmd_handle(kernbuf, count, &sbi->ll_pcc_super); +out_free_kernbuff: + OBD_FREE(kernbuf, count + 1); + return rc ? rc : count; +} +LDEBUGFS_SEQ_FOPS(ll_pcc); + +struct ldebugfs_vars lprocfs_llite_obd_vars[] = { { .name = "site", .fops = &ll_site_stats_fops }, - { .name = "blocksize", - .fops = &ll_blksize_fops }, - { .name = "stat_blocksize", - .fops = &ll_stat_blksize_fops }, - { .name = "kbytestotal", - .fops = &ll_kbytestotal_fops }, - { .name = "kbytesfree", - .fops = &ll_kbytesfree_fops }, - { .name = "kbytesavail", - .fops = &ll_kbytesavail_fops }, - { .name = "filestotal", - .fops = &ll_filestotal_fops }, - { .name = "filesfree", - .fops = &ll_filesfree_fops }, - { .name = "client_type", - .fops = &ll_client_type_fops }, - { .name = "max_read_ahead_mb", - .fops = &ll_max_readahead_mb_fops }, - { .name = "max_read_ahead_per_file_mb", - .fops = &ll_max_readahead_per_file_mb_fops }, - { .name = "max_read_ahead_whole_mb", - .fops = &ll_max_read_ahead_whole_mb_fops }, { .name = "max_cached_mb", .fops = &ll_max_cached_mb_fops }, - { .name = "checksum_pages", - .fops = &ll_checksum_fops }, - { .name = "stats_track_pid", - .fops = &ll_track_pid_fops }, - { .name = "stats_track_ppid", - .fops = &ll_track_ppid_fops }, - { .name = "stats_track_gid", - .fops = &ll_track_gid_fops }, - { .name = "statahead_max", - .fops = &ll_statahead_max_fops }, - { .name = "statahead_agl", - .fops = &ll_statahead_agl_fops }, { .name = "statahead_stats", .fops = &ll_statahead_stats_fops }, - { .name = "lazystatfs", - .fops = &ll_lazystatfs_fops }, - { .name = "max_easize", - .fops = &ll_max_easize_fops }, - { .name = "default_easize", - .fops = &ll_default_easize_fops }, - { .name = "sbi_flags", - .fops = &ll_sbi_flags_fops }, - { .name = "xattr_cache", - .fops = &ll_xattr_cache_fops }, { .name = "unstable_stats", .fops = &ll_unstable_stats_fops }, + { .name = "sbi_flags", + .fops = &ll_sbi_flags_fops }, { .name = "root_squash", .fops = &ll_root_squash_fops }, { .name = "nosquash_nids", .fops = &ll_nosquash_nids_fops }, - { .name = "fast_read", - .fops = &ll_fast_read_fops, }, - { .name = "pio", - .fops = &ll_pio_fops, }, - { .name = "tiny_write", - .fops = &ll_tiny_write_fops, }, + { .name = "pcc", + .fops = &ll_pcc_fops, }, { NULL } }; #define MAX_STRING_SIZE 128 +static struct attribute *llite_attrs[] = { + &lustre_attr_blocksize.attr, + &lustre_attr_stat_blocksize.attr, + &lustre_attr_kbytestotal.attr, + &lustre_attr_kbytesfree.attr, + &lustre_attr_kbytesavail.attr, + &lustre_attr_filestotal.attr, + &lustre_attr_filesfree.attr, + &lustre_attr_client_type.attr, + &lustre_attr_foreign_symlink_enable.attr, + &lustre_attr_foreign_symlink_prefix.attr, + &lustre_attr_foreign_symlink_upcall.attr, + &lustre_attr_foreign_symlink_upcall_info.attr, + &lustre_attr_fstype.attr, + &lustre_attr_uuid.attr, + &lustre_attr_checksums.attr, + &lustre_attr_checksum_pages.attr, + &lustre_attr_max_read_ahead_mb.attr, + &lustre_attr_max_read_ahead_per_file_mb.attr, + &lustre_attr_max_read_ahead_whole_mb.attr, + &lustre_attr_max_read_ahead_async_active.attr, + &lustre_attr_read_ahead_async_file_threshold_mb.attr, + &lustre_attr_read_ahead_range_kb.attr, + &lustre_attr_stats_track_pid.attr, + &lustre_attr_stats_track_ppid.attr, + &lustre_attr_stats_track_gid.attr, + &lustre_attr_statahead_running_max.attr, + &lustre_attr_statahead_max.attr, + &lustre_attr_statahead_agl.attr, + &lustre_attr_lazystatfs.attr, + &lustre_attr_statfs_max_age.attr, + &lustre_attr_max_easize.attr, + &lustre_attr_default_easize.attr, + &lustre_attr_xattr_cache.attr, + &lustre_attr_fast_read.attr, + &lustre_attr_tiny_write.attr, + &lustre_attr_file_heat.attr, + &lustre_attr_heat_decay_percentage.attr, + &lustre_attr_heat_period_second.attr, + NULL, +}; + +static void sbi_kobj_release(struct kobject *kobj) +{ + struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info, + ll_kset.kobj); + complete(&sbi->ll_kobj_unregister); +} + +static struct kobj_type sbi_ktype = { + .default_attrs = llite_attrs, + .sysfs_ops = &lustre_sysfs_ops, + .release = sbi_kobj_release, +}; + static const struct llite_file_opcode { - __u32 opcode; - __u32 type; - const char *opname; + __u32 opcode; + __u32 type; + const char *opname; } llite_opcode_table[LPROC_LL_FILE_OPCODES] = { - /* file operation */ - { LPROC_LL_DIRTY_HITS, LPROCFS_TYPE_REGS, "dirty_pages_hits" }, - { LPROC_LL_DIRTY_MISSES, LPROCFS_TYPE_REGS, "dirty_pages_misses" }, - { LPROC_LL_READ_BYTES, LPROCFS_CNTR_AVGMINMAX|LPROCFS_TYPE_BYTES, - "read_bytes" }, - { LPROC_LL_WRITE_BYTES, LPROCFS_CNTR_AVGMINMAX|LPROCFS_TYPE_BYTES, - "write_bytes" }, - { LPROC_LL_BRW_READ, LPROCFS_CNTR_AVGMINMAX|LPROCFS_TYPE_PAGES, - "brw_read" }, - { LPROC_LL_BRW_WRITE, LPROCFS_CNTR_AVGMINMAX|LPROCFS_TYPE_PAGES, - "brw_write" }, - { LPROC_LL_IOCTL, LPROCFS_TYPE_REGS, "ioctl" }, - { LPROC_LL_OPEN, LPROCFS_TYPE_REGS, "open" }, - { LPROC_LL_RELEASE, LPROCFS_TYPE_REGS, "close" }, - { LPROC_LL_MAP, LPROCFS_TYPE_REGS, "mmap" }, - { LPROC_LL_FAULT, LPROCFS_TYPE_REGS, "page_fault" }, - { LPROC_LL_MKWRITE, LPROCFS_TYPE_REGS, "page_mkwrite" }, - { LPROC_LL_LLSEEK, LPROCFS_TYPE_REGS, "seek" }, - { LPROC_LL_FSYNC, LPROCFS_TYPE_REGS, "fsync" }, - { LPROC_LL_READDIR, LPROCFS_TYPE_REGS, "readdir" }, - /* inode operation */ - { LPROC_LL_SETATTR, LPROCFS_TYPE_REGS, "setattr" }, - { LPROC_LL_TRUNC, LPROCFS_TYPE_REGS, "truncate" }, - { LPROC_LL_FLOCK, LPROCFS_TYPE_REGS, "flock" }, - { LPROC_LL_GETATTR, LPROCFS_TYPE_REGS, "getattr" }, - /* dir inode operation */ - { LPROC_LL_CREATE, LPROCFS_TYPE_REGS, "create" }, - { LPROC_LL_LINK, LPROCFS_TYPE_REGS, "link" }, - { LPROC_LL_UNLINK, LPROCFS_TYPE_REGS, "unlink" }, - { LPROC_LL_SYMLINK, LPROCFS_TYPE_REGS, "symlink" }, - { LPROC_LL_MKDIR, LPROCFS_TYPE_REGS, "mkdir" }, - { LPROC_LL_RMDIR, LPROCFS_TYPE_REGS, "rmdir" }, - { LPROC_LL_MKNOD, LPROCFS_TYPE_REGS, "mknod" }, - { LPROC_LL_RENAME, LPROCFS_TYPE_REGS, "rename" }, - /* special inode operation */ - { LPROC_LL_STAFS, LPROCFS_TYPE_REGS, "statfs" }, - { LPROC_LL_ALLOC_INODE, LPROCFS_TYPE_REGS, "alloc_inode" }, - { LPROC_LL_SETXATTR, LPROCFS_TYPE_REGS, "setxattr" }, - { LPROC_LL_GETXATTR, LPROCFS_TYPE_REGS, "getxattr" }, - { LPROC_LL_GETXATTR_HITS, LPROCFS_TYPE_REGS, "getxattr_hits" }, - { LPROC_LL_LISTXATTR, LPROCFS_TYPE_REGS, "listxattr" }, - { LPROC_LL_REMOVEXATTR, LPROCFS_TYPE_REGS, "removexattr" }, - { LPROC_LL_INODE_PERM, LPROCFS_TYPE_REGS, "inode_permission" }, + /* file operation */ + { LPROC_LL_READ_BYTES, LPROCFS_TYPE_BYTES_FULL, "read_bytes" }, + { LPROC_LL_WRITE_BYTES, LPROCFS_TYPE_BYTES_FULL, "write_bytes" }, + { LPROC_LL_READ, LPROCFS_TYPE_LATENCY, "read" }, + { LPROC_LL_WRITE, LPROCFS_TYPE_LATENCY, "write" }, + { LPROC_LL_IOCTL, LPROCFS_TYPE_REQS, "ioctl" }, + { LPROC_LL_OPEN, LPROCFS_TYPE_LATENCY, "open" }, + { LPROC_LL_RELEASE, LPROCFS_TYPE_LATENCY, "close" }, + { LPROC_LL_MMAP, LPROCFS_TYPE_LATENCY, "mmap" }, + { LPROC_LL_FAULT, LPROCFS_TYPE_LATENCY, "page_fault" }, + { LPROC_LL_MKWRITE, LPROCFS_TYPE_LATENCY, "page_mkwrite" }, + { LPROC_LL_LLSEEK, LPROCFS_TYPE_LATENCY, "seek" }, + { LPROC_LL_FSYNC, LPROCFS_TYPE_LATENCY, "fsync" }, + { LPROC_LL_READDIR, LPROCFS_TYPE_LATENCY, "readdir" }, + /* inode operation */ + { LPROC_LL_SETATTR, LPROCFS_TYPE_LATENCY, "setattr" }, + { LPROC_LL_TRUNC, LPROCFS_TYPE_LATENCY, "truncate" }, + { LPROC_LL_FLOCK, LPROCFS_TYPE_LATENCY, "flock" }, + { LPROC_LL_GETATTR, LPROCFS_TYPE_LATENCY, "getattr" }, + { LPROC_LL_FALLOCATE, LPROCFS_TYPE_LATENCY, "fallocate"}, + /* dir inode operation */ + { LPROC_LL_CREATE, LPROCFS_TYPE_LATENCY, "create" }, + { LPROC_LL_LINK, LPROCFS_TYPE_LATENCY, "link" }, + { LPROC_LL_UNLINK, LPROCFS_TYPE_LATENCY, "unlink" }, + { LPROC_LL_SYMLINK, LPROCFS_TYPE_LATENCY, "symlink" }, + { LPROC_LL_MKDIR, LPROCFS_TYPE_LATENCY, "mkdir" }, + { LPROC_LL_RMDIR, LPROCFS_TYPE_LATENCY, "rmdir" }, + { LPROC_LL_MKNOD, LPROCFS_TYPE_LATENCY, "mknod" }, + { LPROC_LL_RENAME, LPROCFS_TYPE_LATENCY, "rename" }, + /* special inode operation */ + { LPROC_LL_STATFS, LPROCFS_TYPE_LATENCY, "statfs" }, + { LPROC_LL_SETXATTR, LPROCFS_TYPE_LATENCY, "setxattr" }, + { LPROC_LL_GETXATTR, LPROCFS_TYPE_LATENCY, "getxattr" }, + { LPROC_LL_GETXATTR_HITS, LPROCFS_TYPE_REQS, "getxattr_hits" }, + { LPROC_LL_LISTXATTR, LPROCFS_TYPE_LATENCY, "listxattr" }, + { LPROC_LL_REMOVEXATTR, LPROCFS_TYPE_LATENCY, "removexattr" }, + { LPROC_LL_INODE_PERM, LPROCFS_TYPE_LATENCY, "inode_permission" }, }; -void ll_stats_ops_tally(struct ll_sb_info *sbi, int op, int count) +void ll_stats_ops_tally(struct ll_sb_info *sbi, int op, long count) { - if (!sbi->ll_stats) - return; - if (sbi->ll_stats_track_type == STATS_TRACK_ALL) - lprocfs_counter_add(sbi->ll_stats, op, count); - else if (sbi->ll_stats_track_type == STATS_TRACK_PID && - sbi->ll_stats_track_id == current->pid) - lprocfs_counter_add(sbi->ll_stats, op, count); - else if (sbi->ll_stats_track_type == STATS_TRACK_PPID && - sbi->ll_stats_track_id == current->parent->pid) - lprocfs_counter_add(sbi->ll_stats, op, count); + if (!sbi->ll_stats) + return; + + if (sbi->ll_stats_track_type == STATS_TRACK_ALL) + lprocfs_counter_add(sbi->ll_stats, op, count); + else if (sbi->ll_stats_track_type == STATS_TRACK_PID && + sbi->ll_stats_track_id == current->pid) + lprocfs_counter_add(sbi->ll_stats, op, count); + else if (sbi->ll_stats_track_type == STATS_TRACK_PPID && + sbi->ll_stats_track_id == current->parent->pid) + lprocfs_counter_add(sbi->ll_stats, op, count); else if (sbi->ll_stats_track_type == STATS_TRACK_GID && sbi->ll_stats_track_id == from_kgid(&init_user_ns, current_gid())) @@ -1235,191 +1652,141 @@ void ll_stats_ops_tally(struct ll_sb_info *sbi, int op, int count) } EXPORT_SYMBOL(ll_stats_ops_tally); -static const char *ra_stat_string[] = { - [RA_STAT_HIT] = "hits", - [RA_STAT_MISS] = "misses", - [RA_STAT_DISTANT_READPAGE] = "readpage not consecutive", - [RA_STAT_MISS_IN_WINDOW] = "miss inside window", - [RA_STAT_FAILED_GRAB_PAGE] = "failed grab_cache_page", - [RA_STAT_FAILED_MATCH] = "failed lock match", - [RA_STAT_DISCARDED] = "read but discarded", - [RA_STAT_ZERO_LEN] = "zero length file", - [RA_STAT_ZERO_WINDOW] = "zero size window", - [RA_STAT_EOF] = "read-ahead to EOF", - [RA_STAT_MAX_IN_FLIGHT] = "hit max r-a issue", - [RA_STAT_WRONG_GRAB_PAGE] = "wrong page from grab_cache_page", - [RA_STAT_FAILED_REACH_END] = "failed to reach end" +static const char *const ra_stat_string[] = { + [RA_STAT_HIT] = "hits", + [RA_STAT_MISS] = "misses", + [RA_STAT_DISTANT_READPAGE] = "readpage_not_consecutive", + [RA_STAT_MISS_IN_WINDOW] = "miss_inside_window", + [RA_STAT_FAILED_GRAB_PAGE] = "failed_grab_cache_page", + [RA_STAT_FAILED_MATCH] = "failed_lock_match", + [RA_STAT_DISCARDED] = "read_but_discarded", + [RA_STAT_ZERO_LEN] = "zero_length_file", + [RA_STAT_ZERO_WINDOW] = "zero_size_window", + [RA_STAT_EOF] = "readahead_to_eof", + [RA_STAT_MAX_IN_FLIGHT] = "hit_max_readahead_issue", + [RA_STAT_WRONG_GRAB_PAGE] = "wrong_page_from_grab_cache_page", + [RA_STAT_FAILED_REACH_END] = "failed_to_reach_end", + [RA_STAT_ASYNC] = "async_readahead", + [RA_STAT_FAILED_FAST_READ] = "failed_to_fast_read", + [RA_STAT_MMAP_RANGE_READ] = "mmap_range_read", }; -LPROC_SEQ_FOPS_RO_TYPE(llite, name); -LPROC_SEQ_FOPS_RO_TYPE(llite, uuid); - -int lprocfs_ll_register_mountpoint(struct proc_dir_entry *parent, - struct super_block *sb) +int ll_debugfs_register_super(struct super_block *sb, const char *name) { - struct lprocfs_vars lvars[2]; struct lustre_sb_info *lsi = s2lsi(sb); struct ll_sb_info *sbi = ll_s2sbi(sb); - char name[MAX_STRING_SIZE + 1], *ptr; - int err, id, len, rc; - ENTRY; + int err, id; - memset(lvars, 0, sizeof(lvars)); + ENTRY; + LASSERT(sbi); - name[MAX_STRING_SIZE] = '\0'; - lvars[0].name = name; + if (IS_ERR_OR_NULL(llite_root)) + goto out_ll_kset; - LASSERT(sbi != NULL); + sbi->ll_debugfs_entry = debugfs_create_dir(name, llite_root); + ldebugfs_add_vars(sbi->ll_debugfs_entry, lprocfs_llite_obd_vars, sb); - /* Get fsname */ - len = strlen(lsi->lsi_lmd->lmd_profile); - ptr = strrchr(lsi->lsi_lmd->lmd_profile, '-'); - if (ptr && (strcmp(ptr, "-client") == 0)) - len -= 7; + debugfs_create_file("dump_page_cache", 0444, sbi->ll_debugfs_entry, sbi, + &vvp_dump_pgcache_file_ops); - /* Mount info */ - snprintf(name, MAX_STRING_SIZE, "%.*s-%p", len, - lsi->lsi_lmd->lmd_profile, sb); + debugfs_create_file("extents_stats", 0644, sbi->ll_debugfs_entry, sbi, + &ll_rw_extents_stats_fops); - sbi->ll_proc_root = lprocfs_register(name, parent, NULL, NULL); - if (IS_ERR(sbi->ll_proc_root)) { - err = PTR_ERR(sbi->ll_proc_root); - sbi->ll_proc_root = NULL; - RETURN(err); - } + debugfs_create_file("extents_stats_per_process", 0644, + sbi->ll_debugfs_entry, sbi, + &ll_rw_extents_stats_pp_fops); - rc = lprocfs_seq_create(sbi->ll_proc_root, "dump_page_cache", 0444, - &vvp_dump_pgcache_file_ops, sbi); - if (rc) - CWARN("Error adding the dump_page_cache file\n"); - - rc = lprocfs_seq_create(sbi->ll_proc_root, "extents_stats", 0644, - &ll_rw_extents_stats_fops, sbi); - if (rc) - CWARN("Error adding the extent_stats file\n"); - - rc = lprocfs_seq_create(sbi->ll_proc_root, "extents_stats_per_process", - 0644, &ll_rw_extents_stats_pp_fops, sbi); - if (rc) - CWARN("Error adding the extents_stats_per_process file\n"); - - rc = lprocfs_seq_create(sbi->ll_proc_root, "offset_stats", 0644, - &ll_rw_offset_stats_fops, sbi); - if (rc) - CWARN("Error adding the offset_stats file\n"); + debugfs_create_file("offset_stats", 0644, sbi->ll_debugfs_entry, sbi, + &ll_rw_offset_stats_fops); /* File operations stats */ sbi->ll_stats = lprocfs_alloc_stats(LPROC_LL_FILE_OPCODES, LPROCFS_STATS_FLAG_NONE); if (sbi->ll_stats == NULL) - GOTO(out, err = -ENOMEM); + GOTO(out_debugfs, err = -ENOMEM); + /* do counter init */ for (id = 0; id < LPROC_LL_FILE_OPCODES; id++) { - __u32 type = llite_opcode_table[id].type; - void *ptr = NULL; - if (type & LPROCFS_TYPE_REGS) - ptr = "regs"; + u32 type = llite_opcode_table[id].type; + void *ptr = "unknown"; + + if (type & LPROCFS_TYPE_REQS) + ptr = "reqs"; else if (type & LPROCFS_TYPE_BYTES) ptr = "bytes"; - else if (type & LPROCFS_TYPE_PAGES) - ptr = "pages"; + else if (type & LPROCFS_TYPE_USEC) + ptr = "usec"; lprocfs_counter_init(sbi->ll_stats, - llite_opcode_table[id].opcode, - (type & LPROCFS_CNTR_AVGMINMAX), + llite_opcode_table[id].opcode, type, llite_opcode_table[id].opname, ptr); } - err = lprocfs_register_stats(sbi->ll_proc_root, "stats", sbi->ll_stats); - if (err) - GOTO(out, err); + + debugfs_create_file("stats", 0644, sbi->ll_debugfs_entry, + sbi->ll_stats, &ldebugfs_stats_seq_fops); sbi->ll_ra_stats = lprocfs_alloc_stats(ARRAY_SIZE(ra_stat_string), LPROCFS_STATS_FLAG_NONE); if (sbi->ll_ra_stats == NULL) - GOTO(out, err = -ENOMEM); + GOTO(out_stats, err = -ENOMEM); for (id = 0; id < ARRAY_SIZE(ra_stat_string); id++) lprocfs_counter_init(sbi->ll_ra_stats, id, 0, ra_stat_string[id], "pages"); - err = lprocfs_register_stats(sbi->ll_proc_root, "read_ahead_stats", - sbi->ll_ra_stats); - if (err) - GOTO(out, err); + debugfs_create_file("read_ahead_stats", 0644, sbi->ll_debugfs_entry, + sbi->ll_ra_stats, &ldebugfs_stats_seq_fops); - err = lprocfs_add_vars(sbi->ll_proc_root, lprocfs_llite_obd_vars, sb); +out_ll_kset: + /* Yes we also register sysfs mount kset here as well */ + sbi->ll_kset.kobj.parent = llite_kobj; + sbi->ll_kset.kobj.ktype = &sbi_ktype; + init_completion(&sbi->ll_kobj_unregister); + err = kobject_set_name(&sbi->ll_kset.kobj, "%s", name); if (err) - GOTO(out, err); + GOTO(out_ra_stats, err); + + err = kset_register(&sbi->ll_kset); + if (err) + GOTO(out_ra_stats, err); + + lsi->lsi_kobj = kobject_get(&sbi->ll_kset.kobj); + + RETURN(0); +out_ra_stats: + lprocfs_free_stats(&sbi->ll_ra_stats); +out_stats: + lprocfs_free_stats(&sbi->ll_stats); +out_debugfs: + debugfs_remove_recursive(sbi->ll_debugfs_entry); -out: - if (err) { - lprocfs_remove(&sbi->ll_proc_root); - lprocfs_free_stats(&sbi->ll_ra_stats); - lprocfs_free_stats(&sbi->ll_stats); - } RETURN(err); } -int lprocfs_ll_register_obd(struct super_block *sb, const char *obdname) +void ll_debugfs_unregister_super(struct super_block *sb) { - struct lprocfs_vars lvars[2]; + struct lustre_sb_info *lsi = s2lsi(sb); struct ll_sb_info *sbi = ll_s2sbi(sb); - struct obd_device *obd; - struct proc_dir_entry *dir; - char name[MAX_STRING_SIZE + 1]; - int err; - ENTRY; - - memset(lvars, 0, sizeof(lvars)); - name[MAX_STRING_SIZE] = '\0'; - lvars[0].name = name; + debugfs_remove_recursive(sbi->ll_debugfs_entry); - LASSERT(sbi != NULL); - LASSERT(obdname != NULL); + if (sbi->ll_dt_obd) + sysfs_remove_link(&sbi->ll_kset.kobj, + sbi->ll_dt_obd->obd_type->typ_name); - obd = class_name2obd(obdname); - - LASSERT(obd != NULL); - LASSERT(obd->obd_magic == OBD_DEVICE_MAGIC); - LASSERT(obd->obd_type->typ_name != NULL); - - dir = proc_mkdir(obd->obd_type->typ_name, sbi->ll_proc_root); - if (dir == NULL) - GOTO(out, err = -ENOMEM); - - snprintf(name, MAX_STRING_SIZE, "common_name"); - lvars[0].fops = &llite_name_fops; - err = lprocfs_add_vars(dir, lvars, obd); - if (err) - GOTO(out, err); + if (sbi->ll_md_obd) + sysfs_remove_link(&sbi->ll_kset.kobj, + sbi->ll_md_obd->obd_type->typ_name); - snprintf(name, MAX_STRING_SIZE, "uuid"); - lvars[0].fops = &llite_uuid_fops; - err = lprocfs_add_vars(dir, lvars, obd); - if (err) - GOTO(out, err); + kobject_put(lsi->lsi_kobj); -out: - if (err) { - lprocfs_remove(&sbi->ll_proc_root); - lprocfs_free_stats(&sbi->ll_ra_stats); - lprocfs_free_stats(&sbi->ll_stats); - } - RETURN(err); -} + kset_unregister(&sbi->ll_kset); + wait_for_completion(&sbi->ll_kobj_unregister); -void lprocfs_ll_unregister_mountpoint(struct ll_sb_info *sbi) -{ - if (sbi->ll_proc_root) { - lprocfs_remove(&sbi->ll_proc_root); - lprocfs_free_stats(&sbi->ll_ra_stats); - lprocfs_free_stats(&sbi->ll_stats); - } + lprocfs_free_stats(&sbi->ll_ra_stats); + lprocfs_free_stats(&sbi->ll_stats); } #undef MAX_STRING_SIZE -#define pct(a,b) (b ? a * 100 / b : 0) - static void ll_display_extents_info(struct ll_rw_extents_info *io_extents, struct seq_file *seq, int which) { @@ -1438,26 +1805,26 @@ static void ll_display_extents_info(struct ll_rw_extents_info *io_extents, write_tot += pp_info->pp_w_hist.oh_buckets[i]; } - for(i = 0; i < LL_HIST_MAX; i++) { - r = pp_info->pp_r_hist.oh_buckets[i]; - w = pp_info->pp_w_hist.oh_buckets[i]; - read_cum += r; - write_cum += w; - end = 1 << (i + LL_HIST_START - units); - seq_printf(seq, "%4lu%c - %4lu%c%c: %14lu %4lu %4lu | " - "%14lu %4lu %4lu\n", start, *unitp, end, *unitp, - (i == LL_HIST_MAX - 1) ? '+' : ' ', - r, pct(r, read_tot), pct(read_cum, read_tot), - w, pct(w, write_tot), pct(write_cum, write_tot)); - start = end; - if (start == 1<<10) { - start = 1; - units += 10; - unitp++; - } - if (read_cum == read_tot && write_cum == write_tot) - break; - } + for(i = 0; i < LL_HIST_MAX; i++) { + r = pp_info->pp_r_hist.oh_buckets[i]; + w = pp_info->pp_w_hist.oh_buckets[i]; + read_cum += r; + write_cum += w; + end = 1 << (i + LL_HIST_START - units); + seq_printf(seq, "%4lu%c - %4lu%c%c: %14lu %4u %4u | " + "%14lu %4u %4u\n", start, *unitp, end, *unitp, + (i == LL_HIST_MAX - 1) ? '+' : ' ', + r, pct(r, read_tot), pct(read_cum, read_tot), + w, pct(w, write_tot), pct(write_cum, write_tot)); + start = end; + if (start == (1 << 10)) { + start = 1; + units += 10; + unitp++; + } + if (read_cum == read_tot && write_cum == write_tot) + break; + } } static int ll_rw_extents_stats_pp_seq_show(struct seq_file *seq, void *v) @@ -1522,7 +1889,7 @@ static ssize_t ll_rw_extents_stats_pp_seq_write(struct file *file, return len; } -LPROC_SEQ_FOPS(ll_rw_extents_stats_pp); +LDEBUGFS_SEQ_FOPS(ll_rw_extents_stats_pp); static int ll_rw_extents_stats_seq_show(struct seq_file *seq, void *v) { @@ -1580,7 +1947,8 @@ static ssize_t ll_rw_extents_stats_seq_write(struct file *file, return len; } -LPROC_SEQ_FOPS(ll_rw_extents_stats); + +LDEBUGFS_SEQ_FOPS(ll_rw_extents_stats); void ll_rw_stats_tally(struct ll_sb_info *sbi, pid_t pid, struct ll_file_data *file, loff_t pos, @@ -1617,15 +1985,15 @@ void ll_rw_stats_tally(struct ll_sb_info *sbi, pid_t pid, lprocfs_oh_clear(&io_extents->pp_extents[cur].pp_w_hist); } - for(i = 0; (count >= (1 << LL_HIST_START << i)) && - (i < (LL_HIST_MAX - 1)); i++); - if (rw == 0) { - io_extents->pp_extents[cur].pp_r_hist.oh_buckets[i]++; - io_extents->pp_extents[LL_PROCESS_HIST_MAX].pp_r_hist.oh_buckets[i]++; - } else { - io_extents->pp_extents[cur].pp_w_hist.oh_buckets[i]++; - io_extents->pp_extents[LL_PROCESS_HIST_MAX].pp_w_hist.oh_buckets[i]++; - } + for (i = 0; (count >= 1 << (LL_HIST_START + i)) && + (i < (LL_HIST_MAX - 1)); i++); + if (rw == 0) { + io_extents->pp_extents[cur].pp_r_hist.oh_buckets[i]++; + io_extents->pp_extents[LL_PROCESS_HIST_MAX].pp_r_hist.oh_buckets[i]++; + } else { + io_extents->pp_extents[cur].pp_w_hist.oh_buckets[i]++; + io_extents->pp_extents[LL_PROCESS_HIST_MAX].pp_w_hist.oh_buckets[i]++; + } spin_unlock(&sbi->ll_pp_extent_lock); spin_lock(&sbi->ll_process_lock); @@ -1711,7 +2079,7 @@ static int ll_rw_offset_stats_seq_show(struct seq_file *seq, void *v) for (i = 0; i < LL_OFFSET_HIST_MAX; i++) { if (offset[i].rw_pid != 0) seq_printf(seq, - "%3c %10d %14Lu %14Lu %17lu %17lu %14Lu", + "%3c %10d %14llu %14llu %17lu %17lu %14lld\n", offset[i].rw_op == READ ? 'R' : 'W', offset[i].rw_pid, offset[i].rw_range_start, @@ -1725,7 +2093,7 @@ static int ll_rw_offset_stats_seq_show(struct seq_file *seq, void *v) for (i = 0; i < LL_PROCESS_HIST_MAX; i++) { if (process[i].rw_pid != 0) seq_printf(seq, - "%3c %10d %14Lu %14Lu %17lu %17lu %14Lu", + "%3c %10d %14llu %14llu %17lu %17lu %14lld\n", process[i].rw_op == READ ? 'R' : 'W', process[i].rw_pid, process[i].rw_range_start, @@ -1771,41 +2139,4 @@ static ssize_t ll_rw_offset_stats_seq_write(struct file *file, return len; } -/** - * ll_stats_pid_write() - Determine if stats collection should be enabled - * @buf: Buffer containing the data written - * @len: Number of bytes in the buffer - * - * Several proc files begin collecting stats when a value is written, and stop - * collecting when either '0' or 'disable' is written. This function checks the - * written value to see if collection should be enabled or disabled. - * - * Return: If '0' or 'disable' is provided, 0 is returned. If the text - * equivalent of a number is written, that number is returned. Otherwise, - * 1 is returned. Non-zero return values indicate collection should be enabled. - */ -static __s64 ll_stats_pid_write(const char __user *buf, size_t len) -{ - unsigned long long value = 1; - int rc; - char kernbuf[16]; - - rc = kstrtoull_from_user(buf, len, 0, &value); - if (rc < 0 && len < sizeof(kernbuf)) { - - if (copy_from_user(kernbuf, buf, len)) - return -EFAULT; - kernbuf[len] = 0; - - if (kernbuf[len - 1] == '\n') - kernbuf[len - 1] = 0; - - if (strncasecmp(kernbuf, "disable", 7) == 0) - value = 0; - } - - return value; -} - -LPROC_SEQ_FOPS(ll_rw_offset_stats); -#endif /* CONFIG_PROC_FS */ +LDEBUGFS_SEQ_FOPS(ll_rw_offset_stats);