goto free_kobj;
llite_root = debugfs_create_dir("llite", debugfs_lustre_root);
- if (IS_ERR_OR_NULL(llite_root)) {
- rc = llite_root ? PTR_ERR(llite_root) : -ENOMEM;
- llite_root = NULL;
+ return 0;
+
free_kobj:
- kobject_put(llite_kobj);
- llite_kobj = NULL;
- }
+ kobject_put(llite_kobj);
+ llite_kobj = NULL;
return rc;
}
{
struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info,
ll_kset.kobj);
- unsigned long ra_max_mb;
-
- spin_lock(&sbi->ll_lock);
- ra_max_mb = PAGES_TO_MiB(sbi->ll_ra_info.ra_max_pages);
- spin_unlock(&sbi->ll_lock);
- return snprintf(buf, PAGE_SIZE, "%lu\n", ra_max_mb);
+ return scnprintf(buf, PAGE_SIZE, "%lu\n",
+ PAGES_TO_MiB(sbi->ll_ra_info.ra_max_pages));
}
static ssize_t max_read_ahead_mb_store(struct kobject *kobj,
{
struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info,
ll_kset.kobj);
- unsigned long ra_max_file_mb;
- spin_lock(&sbi->ll_lock);
- ra_max_file_mb = PAGES_TO_MiB(sbi->ll_ra_info.ra_max_pages_per_file);
- spin_unlock(&sbi->ll_lock);
-
- return snprintf(buf, PAGE_SIZE, "%lu\n", ra_max_file_mb);
+ return scnprintf(buf, PAGE_SIZE, "%lu\n",
+ PAGES_TO_MiB(sbi->ll_ra_info.ra_max_pages_per_file));
}
static ssize_t max_read_ahead_per_file_mb_store(struct kobject *kobj,
{
struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info,
ll_kset.kobj);
- unsigned long ra_max_whole_mb;
-
- spin_lock(&sbi->ll_lock);
- ra_max_whole_mb = PAGES_TO_MiB(sbi->ll_ra_info.ra_max_read_ahead_whole_pages);
- spin_unlock(&sbi->ll_lock);
- return snprintf(buf, PAGE_SIZE, "%lu\n", ra_max_whole_mb);
+ return scnprintf(buf, PAGE_SIZE, "%lu\n",
+ PAGES_TO_MiB(sbi->ll_ra_info.ra_max_read_ahead_whole_pages));
}
static ssize_t max_read_ahead_whole_mb_store(struct kobject *kobj,
long max_cached_mb;
long unused_mb;
+ 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"
/* 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) {
env = cl_env_get(&refcheck);
if (IS_ERR(env))
- RETURN(PTR_ERR(env));
+ GOTO(out_unlock, rc = PTR_ERR(env));
diff = -diff;
while (diff > 0) {
/* reduce LRU budget from free slots. */
do {
- long ov, nv, retv;
+ 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;
- retv = atomic_long_cmpxchg(&cache->ccc_lru_left,
- ov, nv);
- if (likely(ov == retv)) {
- 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);
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,
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;
}
LDEBUGFS_SEQ_FOPS(ll_max_cached_mb);
if (rc)
return rc;
- if (val < 1 || val > WQ_UNBOUND_MAX_ACTIVE) {
- CERROR("%s: cannot set max_read_ahead_async_active=%u %s than %u\n",
- sbi->ll_fsname, val,
- val < 1 ? "smaller" : "larger",
- val < 1 ? 1 : WQ_UNBOUND_MAX_ACTIVE);
+ /**
+ * 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);
- workqueue_set_max_active(sbi->ll_ra_info.ll_readahead_wq, val);
return count;
}
OBD_FREE(kernbuf, count + 1);
return rc ? rc : count;
}
-LPROC_SEQ_FOPS(ll_pcc);
+LDEBUGFS_SEQ_FOPS(ll_pcc);
-struct lprocfs_vars lprocfs_llite_obd_vars[] = {
+struct ldebugfs_vars lprocfs_llite_obd_vars[] = {
{ .name = "site",
.fops = &ll_site_stats_fops },
{ .name = "max_cached_mb",
.release = sbi_kobj_release,
};
-#define LPROCFS_TYPE_LATENCY \
- (LPROCFS_TYPE_USEC | LPROCFS_CNTR_AVGMINMAX | LPROCFS_CNTR_STDDEV)
static const struct llite_file_opcode {
__u32 opcode;
__u32 type;
const char *opname;
} llite_opcode_table[LPROC_LL_FILE_OPCODES] = {
/* file operation */
- { 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_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_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" },
{
struct lustre_sb_info *lsi = s2lsi(sb);
struct ll_sb_info *sbi = ll_s2sbi(sb);
- int err, id, rc;
+ int err, id;
ENTRY;
LASSERT(sbi);
if (IS_ERR_OR_NULL(llite_root))
goto out_ll_kset;
- sbi->ll_debugfs_entry = ldebugfs_register(name, llite_root,
- lprocfs_llite_obd_vars, sb);
- if (IS_ERR_OR_NULL(sbi->ll_debugfs_entry)) {
- err = sbi->ll_debugfs_entry ? PTR_ERR(sbi->ll_debugfs_entry) :
- -ENOMEM;
- sbi->ll_debugfs_entry = NULL;
- RETURN(err);
- }
+ sbi->ll_debugfs_entry = debugfs_create_dir(name, llite_root);
+ ldebugfs_add_vars(sbi->ll_debugfs_entry, lprocfs_llite_obd_vars, sb);
- rc = ldebugfs_seq_create(sbi->ll_debugfs_entry, "dump_page_cache",0444,
- &vvp_dump_pgcache_file_ops, sbi);
- if (rc)
- CWARN("Error adding the dump_page_cache file\n");
+ debugfs_create_file("dump_page_cache", 0444, sbi->ll_debugfs_entry, sbi,
+ &vvp_dump_pgcache_file_ops);
- rc = ldebugfs_seq_create(sbi->ll_debugfs_entry, "extents_stats", 0644,
- &ll_rw_extents_stats_fops, sbi);
- if (rc)
- CWARN("Error adding the extent_stats file\n");
+ debugfs_create_file("extents_stats", 0644, sbi->ll_debugfs_entry, sbi,
+ &ll_rw_extents_stats_fops);
- rc = ldebugfs_seq_create(sbi->ll_debugfs_entry,
- "extents_stats_per_process", 0644,
- &ll_rw_extents_stats_pp_fops, sbi);
- if (rc)
- CWARN("Error adding the extents_stats_per_process file\n");
+ debugfs_create_file("extents_stats_per_process", 0644,
+ sbi->ll_debugfs_entry, sbi,
+ &ll_rw_extents_stats_pp_fops);
- rc = ldebugfs_seq_create(sbi->ll_debugfs_entry, "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,
/* do counter init */
for (id = 0; id < LPROC_LL_FILE_OPCODES; id++) {
u32 type = llite_opcode_table[id].type;
- void *ptr = NULL;
+ 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 = ldebugfs_register_stats(sbi->ll_debugfs_entry, "stats",
- sbi->ll_stats);
- if (err)
- GOTO(out_stats, 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);
lprocfs_counter_init(sbi->ll_ra_stats, id, 0,
ra_stat_string[id], "pages");
- err = ldebugfs_register_stats(sbi->ll_debugfs_entry, "read_ahead_stats",
- sbi->ll_ra_stats);
- if (err)
- GOTO(out_ra_stats, err);
+ debugfs_create_file("read_ahead_stats", 0644, sbi->ll_debugfs_entry,
+ sbi->ll_ra_stats, &ldebugfs_stats_seq_fops);
out_ll_kset:
/* Yes we also register sysfs mount kset here as well */
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 = BIT(i + LL_HIST_START - units);
+ 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 == BIT(10)) {
- start = 1;
- units += 10;
- unitp++;
- }
- if (read_cum == read_tot && write_cum == write_tot)
- break;
- }
+ (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)
lprocfs_oh_clear(&io_extents->pp_extents[cur].pp_w_hist);
}
- for (i = 0; (count >= BIT(LL_HIST_START + 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]++;