X-Git-Url: https://git.whamcloud.com/?p=fs%2Flustre-release.git;a=blobdiff_plain;f=lustre%2Fosc%2Flproc_osc.c;h=414d918e9f402308049b8f86c3f6197ab1b27a9b;hp=7aa7b2cf885d0cee8fed7135d1e26708a57aab59;hb=3cce65712d94cffe8f1626545845b95b88aef672;hpb=9b3a6b6f5bd13e79c6a5509f6032211549df004f diff --git a/lustre/osc/lproc_osc.c b/lustre/osc/lproc_osc.c index 7aa7b2c..414d918 100644 --- a/lustre/osc/lproc_osc.c +++ b/lustre/osc/lproc_osc.c @@ -1,6 +1,4 @@ -/* -*- mode: c; c-basic-offset: 8; indent-tabs-mode: nil; -*- - * vim:expandtab:shiftwidth=8:tabstop=8: - * +/* * GPL HEADER START * * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. @@ -17,17 +15,15 @@ * * You should have received a copy of the GNU General Public License * version 2 along with this program; If not, see - * http://www.sun.com/software/products/lustre/docs/GPLv2.pdf - * - * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara, - * CA 95054 USA or visit www.sun.com if you need additional information or - * have any questions. + * http://www.gnu.org/licenses/gpl-2.0.html * * GPL HEADER END */ /* - * Copyright 2008 Sun Microsystems, Inc. All rights reserved + * Copyright (c) 2002, 2010, Oracle and/or its affiliates. All rights reserved. * Use is subject to license terms. + * + * Copyright (c) 2011, 2017, Intel Corporation. */ /* * This file is part of Lustre, http://www.lustre.org/ @@ -41,537 +37,808 @@ #include #include #include +#include + #include "osc_internal.h" -#ifdef LPROCFS -static int osc_rd_active(char *page, char **start, off_t off, - int count, int *eof, void *data) +static ssize_t active_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct obd_device *dev = data; - int rc; - - LPROCFS_CLIMP_CHECK(dev); - rc = snprintf(page, count, "%d\n", !dev->u.cli.cl_import->imp_deactive); - LPROCFS_CLIMP_EXIT(dev); - return rc; + struct obd_device *dev = container_of(kobj, struct obd_device, + obd_kset.kobj); + int rc; + + LPROCFS_CLIMP_CHECK(dev); + rc = sprintf(buf, "%d\n", !dev->u.cli.cl_import->imp_deactive); + LPROCFS_CLIMP_EXIT(dev); + return rc; } -static int osc_wr_active(struct file *file, const char *buffer, - unsigned long count, void *data) +static ssize_t active_store(struct kobject *kobj, struct attribute *attr, + const char *buffer, size_t count) { - struct obd_device *dev = data; - int val, rc; - - rc = lprocfs_write_helper(buffer, count, &val); - if (rc) - return rc; - if (val < 0 || val > 1) - return -ERANGE; + struct obd_device *dev = container_of(kobj, struct obd_device, + obd_kset.kobj); + bool val; + int rc; + + rc = kstrtobool(buffer, &val); + if (rc) + return rc; + + /* opposite senses */ + if (dev->u.cli.cl_import->imp_deactive == val) + rc = ptlrpc_set_import_active(dev->u.cli.cl_import, val); + else + CDEBUG(D_CONFIG, "activate %u: ignoring repeat request\n", + (unsigned int)val); + + return count; +} +LUSTRE_RW_ATTR(active); - LPROCFS_CLIMP_CHECK(dev); - /* opposite senses */ - if (dev->u.cli.cl_import->imp_deactive == val) - rc = ptlrpc_set_import_active(dev->u.cli.cl_import, val); - else - CDEBUG(D_CONFIG, "activate %d: ignoring repeat request\n", val); +static ssize_t max_rpcs_in_flight_show(struct kobject *kobj, + struct attribute *attr, + char *buf) +{ + struct obd_device *dev = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct client_obd *cli = &dev->u.cli; + ssize_t len; + + spin_lock(&cli->cl_loi_list_lock); + len = sprintf(buf, "%u\n", cli->cl_max_rpcs_in_flight); + spin_unlock(&cli->cl_loi_list_lock); + return len; +} - LPROCFS_CLIMP_EXIT(dev); - return count; +static ssize_t max_rpcs_in_flight_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) +{ + struct obd_device *dev = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct client_obd *cli = &dev->u.cli; + int adding, added, req_count; + unsigned int val; + int rc; + + rc = kstrtouint(buffer, 0, &val); + if (rc) + return rc; + + if (val == 0 || val > OSC_MAX_RIF_MAX) + return -ERANGE; + + LPROCFS_CLIMP_CHECK(dev); + + adding = (int)val - cli->cl_max_rpcs_in_flight; + req_count = atomic_read(&osc_pool_req_count); + if (adding > 0 && req_count < osc_reqpool_maxreqcount) { + /* + * There might be some race which will cause over-limit + * allocation, but it is fine. + */ + if (req_count + adding > osc_reqpool_maxreqcount) + adding = osc_reqpool_maxreqcount - req_count; + + added = osc_rq_pool->prp_populate(osc_rq_pool, adding); + atomic_add(added, &osc_pool_req_count); + } + + spin_lock(&cli->cl_loi_list_lock); + cli->cl_max_rpcs_in_flight = val; + client_adjust_max_dirty(cli); + spin_unlock(&cli->cl_loi_list_lock); + + LPROCFS_CLIMP_EXIT(dev); + return count; } +LUSTRE_RW_ATTR(max_rpcs_in_flight); -static int osc_rd_max_pages_per_rpc(char *page, char **start, off_t off, - int count, int *eof, void *data) +static ssize_t max_dirty_mb_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct obd_device *dev = data; - struct client_obd *cli = &dev->u.cli; - int rc; + struct obd_device *dev = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct client_obd *cli = &dev->u.cli; + unsigned long val; - client_obd_list_lock(&cli->cl_loi_list_lock); - rc = snprintf(page, count, "%d\n", cli->cl_max_pages_per_rpc); - client_obd_list_unlock(&cli->cl_loi_list_lock); - return rc; + spin_lock(&cli->cl_loi_list_lock); + val = PAGES_TO_MiB(cli->cl_dirty_max_pages); + spin_unlock(&cli->cl_loi_list_lock); + + return sprintf(buf, "%lu\n", val); } -static int osc_wr_max_pages_per_rpc(struct file *file, const char *buffer, - unsigned long count, void *data) +static ssize_t max_dirty_mb_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct obd_device *dev = data; - struct client_obd *cli = &dev->u.cli; - struct obd_connect_data *ocd = &cli->cl_import->imp_connect_data; - int val, rc; + struct obd_device *dev = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct client_obd *cli = &dev->u.cli; + unsigned long pages_number, max_dirty_mb; + int rc; - rc = lprocfs_write_helper(buffer, count, &val); - if (rc) - return rc; + rc = kstrtoul(buffer, 10, &max_dirty_mb); + if (rc) + return rc; - LPROCFS_CLIMP_CHECK(dev); - if (val < 1 || val > ocd->ocd_brw_size >> CFS_PAGE_SHIFT) { - LPROCFS_CLIMP_EXIT(dev); - return -ERANGE; - } - client_obd_list_lock(&cli->cl_loi_list_lock); - cli->cl_max_pages_per_rpc = val; - client_obd_list_unlock(&cli->cl_loi_list_lock); + pages_number = MiB_TO_PAGES(max_dirty_mb); + + if (pages_number >= MiB_TO_PAGES(OSC_MAX_DIRTY_MB_MAX) || + pages_number > cfs_totalram_pages() / 4) /* 1/4 of RAM */ + return -ERANGE; - LPROCFS_CLIMP_EXIT(dev); - return count; + spin_lock(&cli->cl_loi_list_lock); + cli->cl_dirty_max_pages = pages_number; + osc_wake_cache_waiters(cli); + spin_unlock(&cli->cl_loi_list_lock); + + return count; } +LUSTRE_RW_ATTR(max_dirty_mb); + +LUSTRE_ATTR(ost_conn_uuid, 0444, conn_uuid_show, NULL); +LUSTRE_RO_ATTR(conn_uuid); -static int osc_rd_max_rpcs_in_flight(char *page, char **start, off_t off, - int count, int *eof, void *data) +LUSTRE_RW_ATTR(ping); + +static int osc_cached_mb_seq_show(struct seq_file *m, void *v) { - struct obd_device *dev = data; - struct client_obd *cli = &dev->u.cli; - int rc; + struct obd_device *dev = m->private; + struct client_obd *cli = &dev->u.cli; + int shift = 20 - PAGE_SHIFT; + + seq_printf(m, "used_mb: %ld\n" + "busy_cnt: %ld\n" + "reclaim: %llu\n", + (atomic_long_read(&cli->cl_lru_in_list) + + atomic_long_read(&cli->cl_lru_busy)) >> shift, + atomic_long_read(&cli->cl_lru_busy), + cli->cl_lru_reclaim); + + return 0; +} - client_obd_list_lock(&cli->cl_loi_list_lock); - rc = snprintf(page, count, "%u\n", cli->cl_max_rpcs_in_flight); - client_obd_list_unlock(&cli->cl_loi_list_lock); - return rc; +/* shrink the number of caching pages to a specific number */ +static ssize_t osc_cached_mb_seq_write(struct file *file, + const char __user *buffer, + size_t count, loff_t *off) +{ + struct obd_device *dev = ((struct seq_file *)file->private_data)->private; + struct client_obd *cli = &dev->u.cli; + __s64 pages_number; + long rc; + char kernbuf[128]; + + if (count >= sizeof(kernbuf)) + return -EINVAL; + + if (copy_from_user(kernbuf, buffer, count)) + return -EFAULT; + kernbuf[count] = 0; + + buffer += lprocfs_find_named_value(kernbuf, "used_mb:", &count) - + kernbuf; + rc = lprocfs_str_with_units_to_s64(buffer, count, &pages_number, 'M'); + if (rc) + return rc; + + pages_number >>= PAGE_SHIFT; + + if (pages_number < 0) + return -ERANGE; + + rc = atomic_long_read(&cli->cl_lru_in_list) - pages_number; + if (rc > 0) { + struct lu_env *env; + __u16 refcheck; + + env = cl_env_get(&refcheck); + if (!IS_ERR(env)) { + (void)osc_lru_shrink(env, cli, rc, true); + cl_env_put(env, &refcheck); + } + } + + return count; } -static int osc_wr_max_rpcs_in_flight(struct file *file, const char *buffer, - unsigned long count, void *data) +LPROC_SEQ_FOPS(osc_cached_mb); + +static ssize_t cur_dirty_bytes_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct obd_device *dev = data; - struct client_obd *cli = &dev->u.cli; - struct ptlrpc_request_pool *pool = cli->cl_import->imp_rq_pool; - int val, rc; + struct obd_device *dev = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct client_obd *cli = &dev->u.cli; + ssize_t len; - rc = lprocfs_write_helper(buffer, count, &val); - if (rc) - return rc; + spin_lock(&cli->cl_loi_list_lock); + len = sprintf(buf, "%lu\n", cli->cl_dirty_pages << PAGE_SHIFT); + spin_unlock(&cli->cl_loi_list_lock); - if (val < 1 || val > OSC_MAX_RIF_MAX) - return -ERANGE; + return len; +} +LUSTRE_RO_ATTR(cur_dirty_bytes); - LPROCFS_CLIMP_CHECK(dev); - if (pool && val > cli->cl_max_rpcs_in_flight) - pool->prp_populate(pool, val-cli->cl_max_rpcs_in_flight); +static int osc_cur_grant_bytes_seq_show(struct seq_file *m, void *v) +{ + struct obd_device *dev = m->private; + struct client_obd *cli = &dev->u.cli; - client_obd_list_lock(&cli->cl_loi_list_lock); - cli->cl_max_rpcs_in_flight = val; - client_obd_list_unlock(&cli->cl_loi_list_lock); + spin_lock(&cli->cl_loi_list_lock); + seq_printf(m, "%lu\n", cli->cl_avail_grant); + spin_unlock(&cli->cl_loi_list_lock); + return 0; +} - LPROCFS_CLIMP_EXIT(dev); - return count; +static ssize_t osc_cur_grant_bytes_seq_write(struct file *file, + const char __user *buffer, + size_t count, loff_t *off) +{ + struct obd_device *obd = ((struct seq_file *)file->private_data)->private; + struct client_obd *cli = &obd->u.cli; + s64 val; + int rc; + + if (obd == NULL) + return 0; + + rc = lprocfs_str_with_units_to_s64(buffer, count, &val, '1'); + if (rc) + return rc; + if (val < 0) + return val; + + /* this is only for shrinking grant */ + spin_lock(&cli->cl_loi_list_lock); + if (val >= cli->cl_avail_grant) { + spin_unlock(&cli->cl_loi_list_lock); + return 0; + } + + spin_unlock(&cli->cl_loi_list_lock); + + LPROCFS_CLIMP_CHECK(obd); + if (cli->cl_import->imp_state == LUSTRE_IMP_FULL) + rc = osc_shrink_grant_to_target(cli, val); + LPROCFS_CLIMP_EXIT(obd); + + return rc ? rc : count; } +LPROC_SEQ_FOPS(osc_cur_grant_bytes); -static int osc_rd_max_dirty_mb(char *page, char **start, off_t off, int count, - int *eof, void *data) +static ssize_t cur_lost_grant_bytes_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct obd_device *dev = data; - struct client_obd *cli = &dev->u.cli; - long val; - int mult; + struct obd_device *dev = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct client_obd *cli = &dev->u.cli; + ssize_t len; + + spin_lock(&cli->cl_loi_list_lock); + len = sprintf(buf, "%lu\n", cli->cl_lost_grant); + spin_unlock(&cli->cl_loi_list_lock); + return len; +} +LUSTRE_RO_ATTR(cur_lost_grant_bytes); - client_obd_list_lock(&cli->cl_loi_list_lock); - val = cli->cl_dirty_max; - client_obd_list_unlock(&cli->cl_loi_list_lock); +static ssize_t grant_shrink_interval_show(struct kobject *kobj, + struct attribute *attr, + char *buf) +{ + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); - mult = 1 << 20; - return lprocfs_read_frac_helper(page, count, val, mult); + return sprintf(buf, "%lld\n", obd->u.cli.cl_grant_shrink_interval); } -static int osc_wr_max_dirty_mb(struct file *file, const char *buffer, - unsigned long count, void *data) +static ssize_t grant_shrink_interval_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct obd_device *dev = data; - struct client_obd *cli = &dev->u.cli; - int pages_number, mult, rc; + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); + unsigned int val; + int rc; - mult = 1 << (20 - CFS_PAGE_SHIFT); - rc = lprocfs_write_frac_helper(buffer, count, &pages_number, mult); - if (rc) - return rc; + rc = kstrtouint(buffer, 0, &val); + if (rc) + return rc; - if (pages_number < 0 || - pages_number > OSC_MAX_DIRTY_MB_MAX << (20 - CFS_PAGE_SHIFT) || - pages_number > num_physpages / 4) /* 1/4 of RAM */ - return -ERANGE; + if (val == 0) + return -ERANGE; - client_obd_list_lock(&cli->cl_loi_list_lock); - cli->cl_dirty_max = (obd_count)(pages_number << CFS_PAGE_SHIFT); - osc_wake_cache_waiters(cli); - client_obd_list_unlock(&cli->cl_loi_list_lock); + obd->u.cli.cl_grant_shrink_interval = val; + osc_update_next_shrink(&obd->u.cli); + osc_schedule_grant_work(); - return count; + return count; } +LUSTRE_RW_ATTR(grant_shrink_interval); -static int osc_rd_cur_dirty_bytes(char *page, char **start, off_t off, - int count, int *eof, void *data) +static ssize_t checksums_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct obd_device *dev = data; - struct client_obd *cli = &dev->u.cli; - int rc; + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); - client_obd_list_lock(&cli->cl_loi_list_lock); - rc = snprintf(page, count, "%lu\n", cli->cl_dirty); - client_obd_list_unlock(&cli->cl_loi_list_lock); - return rc; + return sprintf(buf, "%d\n", obd->u.cli.cl_checksum ? 1 : 0); } -static int osc_rd_cur_grant_bytes(char *page, char **start, off_t off, - int count, int *eof, void *data) +static ssize_t checksums_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct obd_device *dev = data; - struct client_obd *cli = &dev->u.cli; - int rc; + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); + bool val; + int rc; + + rc = kstrtobool(buffer, &val); + if (rc) + return rc; + + obd->u.cli.cl_checksum = val; - client_obd_list_lock(&cli->cl_loi_list_lock); - rc = snprintf(page, count, "%lu\n", cli->cl_avail_grant); - client_obd_list_unlock(&cli->cl_loi_list_lock); - return rc; + return count; } +LUSTRE_RW_ATTR(checksums); -static int osc_rd_create_count(char *page, char **start, off_t off, int count, - int *eof, void *data) +static int osc_checksum_type_seq_show(struct seq_file *m, void *v) { - struct obd_device *obd = data; + struct obd_device *obd = m->private; + int i; + DECLARE_CKSUM_NAME; if (obd == NULL) return 0; - return snprintf(page, count, "%d\n", - obd->u.cli.cl_oscc.oscc_grow_count); + for (i = 0; i < ARRAY_SIZE(cksum_name); i++) { + if (((1 << i) & obd->u.cli.cl_supp_cksum_types) == 0) + continue; + if (obd->u.cli.cl_cksum_type == (1 << i)) + seq_printf(m, "[%s] ", cksum_name[i]); + else + seq_printf(m, "%s ", cksum_name[i]); + } + seq_printf(m, "\n"); + return 0; } -/** - * Set OSC creator's osc_creator::oscc_grow_count - * - * \param file proc file - * \param buffer buffer containing the value - * \param count buffer size - * \param data obd device - * - * \retval \a count - */ -static int osc_wr_create_count(struct file *file, const char *buffer, - unsigned long count, void *data) +static ssize_t osc_checksum_type_seq_write(struct file *file, + const char __user *buffer, + size_t count, loff_t *off) { - struct obd_device *obd = data; - int val, rc, i; - - if (obd == NULL) - return 0; + struct obd_device *obd = ((struct seq_file *)file->private_data)->private; + int i; + DECLARE_CKSUM_NAME; + char kernbuf[10]; + int rc = -EINVAL; + + if (obd == NULL) + return 0; + + if (count > sizeof(kernbuf) - 1) + return -EINVAL; + if (copy_from_user(kernbuf, buffer, count)) + return -EFAULT; + + if (count > 0 && kernbuf[count - 1] == '\n') + kernbuf[count - 1] = '\0'; + else + kernbuf[count] = '\0'; + + for (i = 0; i < ARRAY_SIZE(cksum_name); i++) { + if (strcmp(kernbuf, cksum_name[i]) == 0) { + obd->u.cli.cl_preferred_cksum_type = BIT(i); + if (obd->u.cli.cl_supp_cksum_types & BIT(i)) { + obd->u.cli.cl_cksum_type = BIT(i); + rc = count; + } else { + rc = -ENOTSUPP; + } + break; + } + } + return rc; +} +LPROC_SEQ_FOPS(osc_checksum_type); - rc = lprocfs_write_helper(buffer, count, &val); - if (rc) - return rc; - - /* The MDT ALWAYS needs to limit the precreate count to - * OST_MAX_PRECREATE, and the constant cannot be changed - * because it is a value shared between the OSC and OST - * that is the maximum possible number of objects that will - * ever be handled by MDT->OST recovery processing. - * - * If the OST ever gets a request to delete more orphans, - * this implies that something has gone badly on the MDT - * and the OST will refuse to delete so much data from the - * filesystem as a safety measure. */ - if (val < OST_MIN_PRECREATE || val > OST_MAX_PRECREATE) - return -ERANGE; - if (val > obd->u.cli.cl_oscc.oscc_max_grow_count) - return -ERANGE; - - for (i = 1; (i << 1) <= val; i <<= 1) - ; - obd->u.cli.cl_oscc.oscc_grow_count = i; - - return count; -} - -/** - * Read OSC creator's osc_creator::oscc_max_grow_count - * - * \param page buffer to hold the returning string - * \param start - * \param off - * \param count - * \param eof - * proc read function parameters, please refer to kernel - * code fs/proc/generic.c proc_file_read() - * \param data obd device - * - * \retval number of characters printed. - */ -static int osc_rd_max_create_count(char *page, char **start, off_t off, - int count, int *eof, void *data) +static ssize_t resend_count_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct obd_device *obd = data; - - if (obd == NULL) - return 0; + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); - return snprintf(page, count, "%d\n", - obd->u.cli.cl_oscc.oscc_max_grow_count); + return sprintf(buf, "%u\n", atomic_read(&obd->u.cli.cl_resends)); } -/** - * Set OSC creator's osc_creator::oscc_max_grow_count - * - * \param file proc file - * \param buffer buffer containing the value - * \param count buffer size - * \param data obd device - * - * \retval \a count - */ -static int osc_wr_max_create_count(struct file *file, const char *buffer, - unsigned long count, void *data) +static ssize_t resend_count_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct obd_device *obd = data; - int val, rc; - - if (obd == NULL) - return 0; - - rc = lprocfs_write_helper(buffer, count, &val); - if (rc) - return rc; + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); + unsigned int val; + int rc; - if (val < 0) - return -ERANGE; - if (val > OST_MAX_PRECREATE) - return -ERANGE; + rc = kstrtouint(buffer, 10, &val); + if (rc) + return rc; - if (obd->u.cli.cl_oscc.oscc_grow_count > val) - obd->u.cli.cl_oscc.oscc_grow_count = val; + atomic_set(&obd->u.cli.cl_resends, val); - obd->u.cli.cl_oscc.oscc_max_grow_count = val; - - return count; + return count; } +LUSTRE_RW_ATTR(resend_count); -static int osc_rd_prealloc_next_id(char *page, char **start, off_t off, - int count, int *eof, void *data) +static ssize_t checksum_dump_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct obd_device *obd = data; - - if (obd == NULL) - return 0; + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); - return snprintf(page, count, LPU64"\n", - obd->u.cli.cl_oscc.oscc_next_id); + return sprintf(buf, "%d\n", obd->u.cli.cl_checksum_dump ? 1 : 0); } -static int osc_rd_prealloc_last_id(char *page, char **start, off_t off, - int count, int *eof, void *data) +static ssize_t checksum_dump_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct obd_device *obd = data; + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); + bool val; + int rc; - if (obd == NULL) - return 0; + rc = kstrtobool(buffer, &val); + if (rc) + return rc; - return snprintf(page, count, LPU64"\n", - obd->u.cli.cl_oscc.oscc_last_id); + obd->u.cli.cl_checksum_dump = val; + + return count; } +LUSTRE_RW_ATTR(checksum_dump); -static int osc_rd_checksum(char *page, char **start, off_t off, int count, - int *eof, void *data) +static ssize_t contention_seconds_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct obd_device *obd = data; - - if (obd == NULL) - return 0; + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct osc_device *od = obd2osc_dev(obd); - return snprintf(page, count, "%d\n", - obd->u.cli.cl_checksum ? 1 : 0); + return sprintf(buf, "%lld\n", od->od_contention_time); } -static int osc_wr_checksum(struct file *file, const char *buffer, - unsigned long count, void *data) +static ssize_t contention_seconds_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct obd_device *obd = data; - int val, rc; + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct osc_device *od = obd2osc_dev(obd); + unsigned int val; + int rc; - if (obd == NULL) - return 0; + rc = kstrtouint(buffer, 0, &val); + if (rc) + return rc; - rc = lprocfs_write_helper(buffer, count, &val); - if (rc) - return rc; + od->od_contention_time = val; - obd->u.cli.cl_checksum = (val ? 1 : 0); + return count; +} +LUSTRE_RW_ATTR(contention_seconds); + +static ssize_t lockless_truncate_show(struct kobject *kobj, + struct attribute *attr, + char *buf) +{ + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct osc_device *od = obd2osc_dev(obd); - return count; + return sprintf(buf, "%u\n", od->od_lockless_truncate); } -static int osc_rd_checksum_type(char *page, char **start, off_t off, int count, - int *eof, void *data) +static ssize_t lockless_truncate_store(struct kobject *kobj, + struct attribute *attr, + const char *buffer, + size_t count) { - struct obd_device *obd = data; - int i, len =0; - DECLARE_CKSUM_NAME; + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct osc_device *od = obd2osc_dev(obd); + bool val; + int rc; - if (obd == NULL) - return 0; + rc = kstrtobool(buffer, &val); + if (rc) + return rc; - for (i = 0; i < ARRAY_SIZE(cksum_name) && len < count; i++) { - if (((1 << i) & obd->u.cli.cl_supp_cksum_types) == 0) - continue; - if (obd->u.cli.cl_cksum_type == (1 << i)) - len += snprintf(page + len, count - len, "[%s] ", - cksum_name[i]); - else - len += snprintf(page + len, count - len, "%s ", - cksum_name[i]); - } - if (len < count) - len += sprintf(page + len, "\n"); - return len; + od->od_lockless_truncate = val; + + return count; } +LUSTRE_RW_ATTR(lockless_truncate); -static int osc_wd_checksum_type(struct file *file, const char *buffer, - unsigned long count, void *data) +static ssize_t destroys_in_flight_show(struct kobject *kobj, + struct attribute *attr, + char *buf) { - struct obd_device *obd = data; - int i; - DECLARE_CKSUM_NAME; - char kernbuf[10]; + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); - if (obd == NULL) - return 0; + return sprintf(buf, "%u\n", + atomic_read(&obd->u.cli.cl_destroy_in_flight)); +} +LUSTRE_RO_ATTR(destroys_in_flight); - if (count > sizeof(kernbuf) - 1) - return -EINVAL; - if (copy_from_user(kernbuf, buffer, count)) - return -EFAULT; - if (count > 0 && kernbuf[count - 1] == '\n') - kernbuf[count - 1] = '\0'; - else - kernbuf[count] = '\0'; - - for (i = 0; i < ARRAY_SIZE(cksum_name); i++) { - if (((1 << i) & obd->u.cli.cl_supp_cksum_types) == 0) - continue; - if (!strcmp(kernbuf, cksum_name[i])) { - obd->u.cli.cl_cksum_type = 1 << i; - return count; - } - } - return -EINVAL; -} - -static int osc_rd_resend_count(char *page, char **start, off_t off, int count, - int *eof, void *data) -{ - struct obd_device *obd = data; - - return snprintf(page, count, "%u\n", atomic_read(&obd->u.cli.cl_resends)); -} - -static int osc_wr_resend_count(struct file *file, const char *buffer, - unsigned long count, void *data) -{ - struct obd_device *obd = data; - int val, rc; - - rc = lprocfs_write_helper(buffer, count, &val); - if (rc) - return rc; - - if (val < 0) - return -EINVAL; - - atomic_set(&obd->u.cli.cl_resends, val); - - return count; -} - -static struct lprocfs_vars lprocfs_osc_obd_vars[] = { - { "uuid", lprocfs_rd_uuid, 0, 0 }, - { "ping", 0, lprocfs_wr_ping, 0, 0, 0222 }, - { "connect_flags", lprocfs_rd_connect_flags, 0, 0 }, - { "blocksize", lprocfs_rd_blksize, 0, 0 }, - { "kbytestotal", lprocfs_rd_kbytestotal, 0, 0 }, - { "kbytesfree", lprocfs_rd_kbytesfree, 0, 0 }, - { "kbytesavail", lprocfs_rd_kbytesavail, 0, 0 }, - { "filestotal", lprocfs_rd_filestotal, 0, 0 }, - { "filesfree", lprocfs_rd_filesfree, 0, 0 }, - //{ "filegroups", lprocfs_rd_filegroups, 0, 0 }, - { "ost_server_uuid", lprocfs_rd_server_uuid, 0, 0 }, - { "ost_conn_uuid", lprocfs_rd_conn_uuid, 0, 0 }, - { "active", osc_rd_active, - osc_wr_active, 0 }, - { "max_pages_per_rpc", osc_rd_max_pages_per_rpc, - osc_wr_max_pages_per_rpc, 0 }, - { "max_rpcs_in_flight", osc_rd_max_rpcs_in_flight, - osc_wr_max_rpcs_in_flight, 0 }, - { "max_dirty_mb", osc_rd_max_dirty_mb, osc_wr_max_dirty_mb, 0 }, - { "cur_dirty_bytes", osc_rd_cur_dirty_bytes, 0, 0 }, - { "cur_grant_bytes", osc_rd_cur_grant_bytes, 0, 0 }, - { "create_count", osc_rd_create_count, osc_wr_create_count, 0 }, - { "max_create_count", osc_rd_max_create_count, - osc_wr_max_create_count, 0}, - { "prealloc_next_id", osc_rd_prealloc_next_id, 0, 0 }, - { "prealloc_last_id", osc_rd_prealloc_last_id, 0, 0 }, - { "checksums", osc_rd_checksum, osc_wr_checksum, 0 }, - { "checksum_type", osc_rd_checksum_type, osc_wd_checksum_type, 0 }, - { "resend_count", osc_rd_resend_count, osc_wr_resend_count, 0}, - { "timeouts", lprocfs_rd_timeouts, 0, 0 }, - { 0 } -}; +LPROC_SEQ_FOPS_RW_TYPE(osc, obd_max_pages_per_rpc); -static struct lprocfs_vars lprocfs_osc_module_vars[] = { - { "num_refs", lprocfs_rd_numrefs, 0, 0 }, - { 0 } -}; +LUSTRE_RW_ATTR(short_io_bytes); -#define pct(a,b) (b ? a * 100 / b : 0) +#ifdef CONFIG_PROC_FS +static int osc_unstable_stats_seq_show(struct seq_file *m, void *v) +{ + struct obd_device *dev = m->private; + struct client_obd *cli = &dev->u.cli; + long pages; + int mb; + + pages = atomic_long_read(&cli->cl_unstable_count); + mb = (pages * PAGE_SIZE) >> 20; + + seq_printf(m, "unstable_pages: %20ld\n" + "unstable_mb: %10d\n", + pages, mb); + return 0; +} +LPROC_SEQ_FOPS_RO(osc_unstable_stats); -static int osc_rpc_stats_seq_show(struct seq_file *seq, void *v) +static ssize_t idle_timeout_show(struct kobject *kobj, struct attribute *attr, + char *buf) { - struct timeval now; - struct obd_device *dev = seq->private; - struct client_obd *cli = &dev->u.cli; - unsigned long read_tot = 0, write_tot = 0, read_cum, write_cum; - int i; + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct client_obd *cli = &obd->u.cli; + int ret; - do_gettimeofday(&now); + LPROCFS_CLIMP_CHECK(obd); + ret = sprintf(buf, "%u\n", cli->cl_import->imp_idle_timeout); + LPROCFS_CLIMP_EXIT(obd); - client_obd_list_lock(&cli->cl_loi_list_lock); + return ret; +} - seq_printf(seq, "snapshot_time: %lu.%lu (secs.usecs)\n", - now.tv_sec, now.tv_usec); - seq_printf(seq, "read RPCs in flight: %d\n", - cli->cl_r_in_flight); - seq_printf(seq, "write RPCs in flight: %d\n", - cli->cl_w_in_flight); - seq_printf(seq, "pending write pages: %d\n", - cli->cl_pending_w_pages); - seq_printf(seq, "pending read pages: %d\n", - cli->cl_pending_r_pages); +static ssize_t idle_timeout_store(struct kobject *kobj, struct attribute *attr, + const char *buffer, size_t count) +{ + struct obd_device *dev = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct client_obd *cli = &dev->u.cli; + struct ptlrpc_request *req; + unsigned int idle_debug = 0; + unsigned int val; + int rc; + + if (strncmp(buffer, "debug", 5) == 0) { + idle_debug = D_CONSOLE; + } else if (strncmp(buffer, "nodebug", 6) == 0) { + idle_debug = D_HA; + } else { + rc = kstrtouint(buffer, 10, &val); + if (rc) + return rc; + + if (val > CONNECTION_SWITCH_MAX) + return -ERANGE; + } + + LPROCFS_CLIMP_CHECK(dev); + if (idle_debug) { + cli->cl_import->imp_idle_debug = idle_debug; + } else { + if (!val) { + /* initiate the connection if it's in IDLE state */ + req = ptlrpc_request_alloc(cli->cl_import, + &RQF_OST_STATFS); + if (req != NULL) + ptlrpc_req_finished(req); + } + cli->cl_import->imp_idle_timeout = val; + } + LPROCFS_CLIMP_EXIT(dev); + + return count; +} +LUSTRE_RW_ATTR(idle_timeout); - seq_printf(seq, "\n\t\t\tread\t\t\twrite\n"); - seq_printf(seq, "pages per rpc rpcs %% cum %% |"); - seq_printf(seq, " rpcs %% cum %%\n"); +static ssize_t idle_connect_store(struct kobject *kobj, struct attribute *attr, + const char *buffer, size_t count) +{ + struct obd_device *dev = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct client_obd *cli = &dev->u.cli; + struct ptlrpc_request *req; + + LPROCFS_CLIMP_CHECK(dev); + /* to initiate the connection if it's in IDLE state */ + req = ptlrpc_request_alloc(cli->cl_import, &RQF_OST_STATFS); + if (req) + ptlrpc_req_finished(req); + ptlrpc_pinger_force(cli->cl_import); + LPROCFS_CLIMP_EXIT(dev); + + return count; +} +LUSTRE_WO_ATTR(idle_connect); - read_tot = lprocfs_oh_sum(&cli->cl_read_page_hist); - write_tot = lprocfs_oh_sum(&cli->cl_write_page_hist); +static ssize_t grant_shrink_show(struct kobject *kobj, struct attribute *attr, + char *buf) +{ + struct obd_device *obd = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct client_obd *cli = &obd->u.cli; + struct obd_connect_data *ocd; + ssize_t len; - read_cum = 0; - write_cum = 0; - for (i = 0; i < OBD_HIST_MAX; i++) { - unsigned long r = cli->cl_read_page_hist.oh_buckets[i]; - unsigned long w = cli->cl_write_page_hist.oh_buckets[i]; - read_cum += r; - write_cum += w; - seq_printf(seq, "%d:\t\t%10lu %3lu %3lu | %10lu %3lu %3lu\n", - 1 << i, r, pct(r, read_tot), - pct(read_cum, read_tot), w, - pct(w, write_tot), - pct(write_cum, write_tot)); - if (read_cum == read_tot && write_cum == write_tot) - break; - } + LPROCFS_CLIMP_CHECK(obd); + ocd = &cli->cl_import->imp_connect_data; - seq_printf(seq, "\n\t\t\tread\t\t\twrite\n"); - seq_printf(seq, "rpcs in flight rpcs %% cum %% |"); - seq_printf(seq, " rpcs %% cum %%\n"); + len = snprintf(buf, PAGE_SIZE, "%d\n", + !!OCD_HAS_FLAG(ocd, GRANT_SHRINK)); + LPROCFS_CLIMP_EXIT(obd); + + return len; +} + +static ssize_t grant_shrink_store(struct kobject *kobj, struct attribute *attr, + const char *buffer, size_t count) +{ + struct obd_device *dev = container_of(kobj, struct obd_device, + obd_kset.kobj); + struct client_obd *cli = &dev->u.cli; + struct obd_connect_data *ocd; + bool val; + int rc; + + if (dev == NULL) + return 0; + + rc = kstrtobool(buffer, &val); + if (rc) + return rc; + + LPROCFS_CLIMP_CHECK(dev); + ocd = &cli->cl_import->imp_connect_data; + + if (!val) { + if (OCD_HAS_FLAG(ocd, GRANT_SHRINK)) + ocd->ocd_connect_flags &= ~OBD_CONNECT_GRANT_SHRINK; + } else { + /** + * server replied obd_connect_data is always bigger, so + * client's imp_connect_flags_orig are always supported + * by the server + */ + if (!OCD_HAS_FLAG(ocd, GRANT_SHRINK) && + cli->cl_import->imp_connect_flags_orig & + OBD_CONNECT_GRANT_SHRINK) + ocd->ocd_connect_flags |= OBD_CONNECT_GRANT_SHRINK; + } + + LPROCFS_CLIMP_EXIT(dev); + + return count; +} +LUSTRE_RW_ATTR(grant_shrink); + +LPROC_SEQ_FOPS_RO_TYPE(osc, connect_flags); +LPROC_SEQ_FOPS_RO_TYPE(osc, server_uuid); +LPROC_SEQ_FOPS_RO_TYPE(osc, timeouts); +LPROC_SEQ_FOPS_RO_TYPE(osc, state); + +LPROC_SEQ_FOPS_RW_TYPE(osc, import); +LPROC_SEQ_FOPS_RW_TYPE(osc, pinger_recov); + +struct lprocfs_vars lprocfs_osc_obd_vars[] = { + { .name = "connect_flags", + .fops = &osc_connect_flags_fops }, + { .name = "ost_server_uuid", + .fops = &osc_server_uuid_fops }, + { .name = "max_pages_per_rpc", + .fops = &osc_obd_max_pages_per_rpc_fops }, + { .name = "osc_cached_mb", + .fops = &osc_cached_mb_fops }, + { .name = "cur_grant_bytes", + .fops = &osc_cur_grant_bytes_fops }, + { .name = "checksum_type", + .fops = &osc_checksum_type_fops }, + { .name = "timeouts", + .fops = &osc_timeouts_fops }, + { .name = "import", + .fops = &osc_import_fops }, + { .name = "state", + .fops = &osc_state_fops }, + { .name = "pinger_recov", + .fops = &osc_pinger_recov_fops }, + { .name = "unstable_stats", + .fops = &osc_unstable_stats_fops }, + { NULL } +}; - read_tot = lprocfs_oh_sum(&cli->cl_read_rpc_hist); - write_tot = lprocfs_oh_sum(&cli->cl_write_rpc_hist); +static int osc_rpc_stats_seq_show(struct seq_file *seq, void *v) +{ + struct timespec64 now; + struct obd_device *dev = seq->private; + struct client_obd *cli = &dev->u.cli; + unsigned long read_tot = 0, write_tot = 0, read_cum, write_cum; + int i; + + ktime_get_real_ts64(&now); + + spin_lock(&cli->cl_loi_list_lock); + + seq_printf(seq, "snapshot_time: %lld.%09lu (secs.nsecs)\n", + (s64)now.tv_sec, now.tv_nsec); + seq_printf(seq, "read RPCs in flight: %d\n", + cli->cl_r_in_flight); + seq_printf(seq, "write RPCs in flight: %d\n", + cli->cl_w_in_flight); + seq_printf(seq, "pending write pages: %d\n", + atomic_read(&cli->cl_pending_w_pages)); + seq_printf(seq, "pending read pages: %d\n", + atomic_read(&cli->cl_pending_r_pages)); + + seq_printf(seq, "\n\t\t\tread\t\t\twrite\n"); + seq_printf(seq, "pages per rpc rpcs %% cum %% |"); + seq_printf(seq, " rpcs %% cum %%\n"); + + read_tot = lprocfs_oh_sum(&cli->cl_read_page_hist); + write_tot = lprocfs_oh_sum(&cli->cl_write_page_hist); + + read_cum = 0; + write_cum = 0; + for (i = 0; i < OBD_HIST_MAX; i++) { + unsigned long r = cli->cl_read_page_hist.oh_buckets[i]; + unsigned long w = cli->cl_write_page_hist.oh_buckets[i]; + + read_cum += r; + write_cum += w; + seq_printf(seq, "%d:\t\t%10lu %3u %3u | %10lu %3u %3u\n", + 1 << i, r, pct(r, read_tot), + pct(read_cum, read_tot), w, + pct(w, write_tot), + pct(write_cum, write_tot)); + if (read_cum == read_tot && write_cum == write_tot) + break; + } + + seq_printf(seq, "\n\t\t\tread\t\t\twrite\n"); + seq_printf(seq, "rpcs in flight rpcs %% cum %% |"); + seq_printf(seq, " rpcs %% cum %%\n"); + + read_tot = lprocfs_oh_sum(&cli->cl_read_rpc_hist); + write_tot = lprocfs_oh_sum(&cli->cl_write_rpc_hist); read_cum = 0; write_cum = 0; @@ -580,11 +847,11 @@ static int osc_rpc_stats_seq_show(struct seq_file *seq, void *v) unsigned long w = cli->cl_write_rpc_hist.oh_buckets[i]; read_cum += r; write_cum += w; - seq_printf(seq, "%d:\t\t%10lu %3lu %3lu | %10lu %3lu %3lu\n", - i, r, pct(r, read_tot), - pct(read_cum, read_tot), w, - pct(w, write_tot), - pct(write_cum, write_tot)); + seq_printf(seq, "%d:\t\t%10lu %3u %3u | %10lu %3u %3u\n", + i, r, pct(r, read_tot), + pct(read_cum, read_tot), w, + pct(w, write_tot), + pct(write_cum, write_tot)); if (read_cum == read_tot && write_cum == write_tot) break; } @@ -603,21 +870,21 @@ static int osc_rpc_stats_seq_show(struct seq_file *seq, void *v) unsigned long w = cli->cl_write_offset_hist.oh_buckets[i]; read_cum += r; write_cum += w; - seq_printf(seq, "%d:\t\t%10lu %3lu %3lu | %10lu %3lu %3lu\n", - (i == 0) ? 0 : 1 << (i - 1), - r, pct(r, read_tot), pct(read_cum, read_tot), - w, pct(w, write_tot), pct(write_cum, write_tot)); + seq_printf(seq, "%d:\t\t%10lu %3u %3u | %10lu %3u %3u\n", + (i == 0) ? 0 : 1 << (i - 1), + r, pct(r, read_tot), pct(read_cum, read_tot), + w, pct(w, write_tot), pct(write_cum, write_tot)); if (read_cum == read_tot && write_cum == write_tot) break; } - client_obd_list_unlock(&cli->cl_loi_list_lock); + spin_unlock(&cli->cl_loi_list_lock); return 0; } -#undef pct -static ssize_t osc_rpc_stats_seq_write(struct file *file, const char *buf, +static ssize_t osc_rpc_stats_seq_write(struct file *file, + const char __user *buf, size_t len, loff_t *off) { struct seq_file *seq = file->private_data; @@ -633,18 +900,103 @@ static ssize_t osc_rpc_stats_seq_write(struct file *file, const char *buf, return len; } - LPROC_SEQ_FOPS(osc_rpc_stats); -int lproc_osc_attach_seqstat(struct obd_device *dev) +static int osc_stats_seq_show(struct seq_file *seq, void *v) +{ + struct timespec64 now; + struct obd_device *dev = seq->private; + struct osc_stats *stats = &obd2osc_dev(dev)->od_stats; + + ktime_get_real_ts64(&now); + + seq_printf(seq, "snapshot_time: %lld.%09lu (secs.nsecs)\n", + (s64)now.tv_sec, now.tv_nsec); + seq_printf(seq, "lockless_write_bytes\t\t%llu\n", + stats->os_lockless_writes); + seq_printf(seq, "lockless_read_bytes\t\t%llu\n", + stats->os_lockless_reads); + seq_printf(seq, "lockless_truncate\t\t%llu\n", + stats->os_lockless_truncates); + return 0; +} + +static ssize_t osc_stats_seq_write(struct file *file, + const char __user *buf, + size_t len, loff_t *off) +{ + struct seq_file *seq = file->private_data; + struct obd_device *dev = seq->private; + struct osc_stats *stats = &obd2osc_dev(dev)->od_stats; + + memset(stats, 0, sizeof(*stats)); + return len; +} + +LPROC_SEQ_FOPS(osc_stats); + +int lprocfs_osc_attach_seqstat(struct obd_device *dev) { - return lprocfs_obd_seq_create(dev, "rpc_stats", 0444, - &osc_rpc_stats_fops, dev); + int rc; + + rc = lprocfs_seq_create(dev->obd_proc_entry, "osc_stats", 0644, + &osc_stats_fops, dev); + if (rc == 0) + rc = lprocfs_obd_seq_create(dev, "rpc_stats", 0644, + &osc_rpc_stats_fops, dev); + + return rc; } +#endif /* CONFIG_PROC_FS */ + +static struct attribute *osc_attrs[] = { + &lustre_attr_active.attr, + &lustre_attr_checksums.attr, + &lustre_attr_checksum_dump.attr, + &lustre_attr_contention_seconds.attr, + &lustre_attr_cur_dirty_bytes.attr, + &lustre_attr_cur_lost_grant_bytes.attr, + &lustre_attr_destroys_in_flight.attr, + &lustre_attr_grant_shrink_interval.attr, + &lustre_attr_lockless_truncate.attr, + &lustre_attr_max_dirty_mb.attr, + &lustre_attr_max_rpcs_in_flight.attr, + &lustre_attr_short_io_bytes.attr, + &lustre_attr_resend_count.attr, + &lustre_attr_ost_conn_uuid.attr, + &lustre_attr_conn_uuid.attr, + &lustre_attr_ping.attr, + &lustre_attr_idle_timeout.attr, + &lustre_attr_idle_connect.attr, + &lustre_attr_grant_shrink.attr, + NULL, +}; -void lprocfs_osc_init_vars(struct lprocfs_static_vars *lvars) +int osc_tunables_init(struct obd_device *obd) { - lvars->module_vars = lprocfs_osc_module_vars; - lvars->obd_vars = lprocfs_osc_obd_vars; + int rc; + + obd->obd_vars = lprocfs_osc_obd_vars; + obd->obd_ktype.default_attrs = osc_attrs; + rc = lprocfs_obd_setup(obd, false); + if (rc) + return rc; +#ifdef CONFIG_PROC_FS + /* If the basic OSC proc tree construction succeeded then + * lets do the rest. + */ + rc = lprocfs_osc_attach_seqstat(obd); + if (rc) + goto obd_cleanup; + +#endif /* CONFIG_PROC_FS */ + rc = sptlrpc_lprocfs_cliobd_attach(obd); + if (rc) + goto obd_cleanup; + + ptlrpc_lprocfs_register_obd(obd); +obd_cleanup: + if (rc) + lprocfs_obd_cleanup(obd); + return rc; } -#endif /* LPROCFS */