From: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
To: Theodore Ts'o <tytso@mit.edu>
Cc: Michal Hocko <mhocko@suse.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
Dmitry Monakhov <dmonakhov@virtuozzo.com>,
Johannes Weiner <hannes@cmpxchg.org>,
linux-fsdevel@vger.kernel.org, linux-ext4@vger.kernel.org
Subject: [PATCH RFC] ext4: use __GFP_NOFAIL in ext4_free_blocks()
Date: Wed, 24 Feb 2016 20:09:12 +0300 [thread overview]
Message-ID: <20160224170912.2195.8153.stgit@buzz> (raw)
This might be unexpected but pages allocated for sbi->s_buddy_cache are
charged to current memory cgroup. So, GFP_NOFS allocation could fail if
current task has been killed by OOM or if current memory cgroup has no
free memory left. Block allocator cannot handle such failures here yet.
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
---
fs/ext4/mballoc.c | 47 ++++++++++++++++++++++++++++-------------------
1 file changed, 28 insertions(+), 19 deletions(-)
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 4424b7bf8ac6..8b7e573eaf97 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -815,7 +815,7 @@ static void mb_regenerate_buddy(struct ext4_buddy *e4b)
* for this page; do not hold this lock when calling this routine!
*/
-static int ext4_mb_init_cache(struct page *page, char *incore)
+static int ext4_mb_init_cache(struct page *page, char *incore, gfp_t gfp)
{
ext4_group_t ngroups;
int blocksize;
@@ -848,7 +848,7 @@ static int ext4_mb_init_cache(struct page *page, char *incore)
/* allocate buffer_heads to read bitmaps */
if (groups_per_page > 1) {
i = sizeof(struct buffer_head *) * groups_per_page;
- bh = kzalloc(i, GFP_NOFS);
+ bh = kzalloc(i, gfp);
if (bh == NULL) {
err = -ENOMEM;
goto out;
@@ -983,7 +983,7 @@ out:
* are on the same page e4b->bd_buddy_page is NULL and return value is 0.
*/
static int ext4_mb_get_buddy_page_lock(struct super_block *sb,
- ext4_group_t group, struct ext4_buddy *e4b)
+ ext4_group_t group, struct ext4_buddy *e4b, gfp_t gfp)
{
struct inode *inode = EXT4_SB(sb)->s_buddy_cache;
int block, pnum, poff;
@@ -1002,7 +1002,7 @@ static int ext4_mb_get_buddy_page_lock(struct super_block *sb,
block = group * 2;
pnum = block / blocks_per_page;
poff = block % blocks_per_page;
- page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS);
+ page = find_or_create_page(inode->i_mapping, pnum, gfp);
if (!page)
return -ENOMEM;
BUG_ON(page->mapping != inode->i_mapping);
@@ -1016,7 +1016,7 @@ static int ext4_mb_get_buddy_page_lock(struct super_block *sb,
block++;
pnum = block / blocks_per_page;
- page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS);
+ page = find_or_create_page(inode->i_mapping, pnum, gfp);
if (!page)
return -ENOMEM;
BUG_ON(page->mapping != inode->i_mapping);
@@ -1042,7 +1042,7 @@ static void ext4_mb_put_buddy_page_lock(struct ext4_buddy *e4b)
* calling this routine!
*/
static noinline_for_stack
-int ext4_mb_init_group(struct super_block *sb, ext4_group_t group)
+int ext4_mb_init_group(struct super_block *sb, ext4_group_t group, gfp_t gfp)
{
struct ext4_group_info *this_grp;
@@ -1062,7 +1062,7 @@ int ext4_mb_init_group(struct super_block *sb, ext4_group_t group)
* The call to ext4_mb_get_buddy_page_lock will mark the
* page accessed.
*/
- ret = ext4_mb_get_buddy_page_lock(sb, group, &e4b);
+ ret = ext4_mb_get_buddy_page_lock(sb, group, &e4b, gfp);
if (ret || !EXT4_MB_GRP_NEED_INIT(this_grp)) {
/*
* somebody initialized the group
@@ -1072,7 +1072,7 @@ int ext4_mb_init_group(struct super_block *sb, ext4_group_t group)
}
page = e4b.bd_bitmap_page;
- ret = ext4_mb_init_cache(page, NULL);
+ ret = ext4_mb_init_cache(page, NULL, gfp);
if (ret)
goto err;
if (!PageUptodate(page)) {
@@ -1091,7 +1091,7 @@ int ext4_mb_init_group(struct super_block *sb, ext4_group_t group)
}
/* init buddy cache */
page = e4b.bd_buddy_page;
- ret = ext4_mb_init_cache(page, e4b.bd_bitmap);
+ ret = ext4_mb_init_cache(page, e4b.bd_bitmap, gfp);
if (ret)
goto err;
if (!PageUptodate(page)) {
@@ -1109,8 +1109,8 @@ err:
* calling this routine!
*/
static noinline_for_stack int
-ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
- struct ext4_buddy *e4b)
+ext4_mb_load_buddy_gfp(struct super_block *sb, ext4_group_t group,
+ struct ext4_buddy *e4b, gfp_t gfp)
{
int blocks_per_page;
int block;
@@ -1140,7 +1140,7 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
* we need full data about the group
* to make a good selection
*/
- ret = ext4_mb_init_group(sb, group);
+ ret = ext4_mb_init_group(sb, group, gfp);
if (ret)
return ret;
}
@@ -1168,11 +1168,11 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
* wait for it to initialize.
*/
page_cache_release(page);
- page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS);
+ page = find_or_create_page(inode->i_mapping, pnum, gfp);
if (page) {
BUG_ON(page->mapping != inode->i_mapping);
if (!PageUptodate(page)) {
- ret = ext4_mb_init_cache(page, NULL);
+ ret = ext4_mb_init_cache(page, NULL, gfp);
if (ret) {
unlock_page(page);
goto err;
@@ -1204,11 +1204,12 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
if (page == NULL || !PageUptodate(page)) {
if (page)
page_cache_release(page);
- page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS);
+ page = find_or_create_page(inode->i_mapping, pnum, gfp);
if (page) {
BUG_ON(page->mapping != inode->i_mapping);
if (!PageUptodate(page)) {
- ret = ext4_mb_init_cache(page, e4b->bd_bitmap);
+ ret = ext4_mb_init_cache(page, e4b->bd_bitmap,
+ gfp);
if (ret) {
unlock_page(page);
goto err;
@@ -1247,6 +1248,12 @@ err:
return ret;
}
+static int ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
+ struct ext4_buddy *e4b)
+{
+ return ext4_mb_load_buddy_gfp(sb, group, e4b, GFP_NOFS);
+}
+
static void ext4_mb_unload_buddy(struct ext4_buddy *e4b)
{
if (e4b->bd_bitmap_page)
@@ -2045,7 +2052,7 @@ static int ext4_mb_good_group(struct ext4_allocation_context *ac,
/* We only do this if the grp has never been initialized */
if (unlikely(EXT4_MB_GRP_NEED_INIT(grp))) {
- int ret = ext4_mb_init_group(ac->ac_sb, group);
+ int ret = ext4_mb_init_group(ac->ac_sb, group, GFP_NOFS);
if (ret)
return ret;
}
@@ -4815,7 +4822,9 @@ do_more:
#endif
trace_ext4_mballoc_free(sb, inode, block_group, bit, count_clusters);
- err = ext4_mb_load_buddy(sb, block_group, &e4b);
+ /* __GFP_NOFAIL: retry infinitely, ignore TIF_MEMDIE and memcg limit. */
+ err = ext4_mb_load_buddy_gfp(sb, block_group, &e4b,
+ GFP_NOFS|__GFP_NOFAIL);
if (err)
goto error_return;
@@ -5217,7 +5226,7 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
grp = ext4_get_group_info(sb, group);
/* We only do this if the grp has never been initialized */
if (unlikely(EXT4_MB_GRP_NEED_INIT(grp))) {
- ret = ext4_mb_init_group(sb, group);
+ ret = ext4_mb_init_group(sb, group, GFP_NOFS);
if (ret)
break;
}
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next reply other threads:[~2016-02-24 17:09 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-02-24 17:09 Konstantin Khlebnikov [this message]
2016-02-25 9:01 ` [PATCH RFC] ext4: use __GFP_NOFAIL in ext4_free_blocks() Nikolay Borisov
2016-02-25 9:08 ` Michal Hocko
2016-02-25 9:12 ` Nikolay Borisov
2016-02-25 10:27 ` Michal Hocko
2016-03-13 21:30 ` Theodore Ts'o
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20160224170912.2195.8153.stgit@buzz \
--to=khlebnikov@yandex-team.ru \
--cc=dmonakhov@virtuozzo.com \
--cc=hannes@cmpxchg.org \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.com \
--cc=tytso@mit.edu \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).