From: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
To: Josef Bacik <josef@redhat.com>
Cc: linux-btrfs@vger.kernel.org, chris.mason@oracle.com
Subject: Re: [PATCH] btrfs: fix return value check of btrfs_start_transaction()
Date: Fri, 21 Jan 2011 15:06:29 +0900 [thread overview]
Message-ID: <4D392265.1020003@jp.fujitsu.com> (raw)
In-Reply-To: <4D38C980.3020001@jp.fujitsu.com>
(2011/01/21 8:47), Tsutomu Itoh wrote:
> (2011/01/21 1:09), Josef Bacik wrote:
>> I'd rather we go through and have these things return an error than do a
>> BUG_ON(). We're moving towards a more stable BTRFS, not one that panics more
>> often :).
>
> Yes, I also think so.
> This patch is my first step.
>
> My modification policy is as follows:
>
> 1. short term
> - To more stable BTRFS, the part that should be corrected is clarified.
> - The panic is not done by the NULL pointer reference etc.
This means, even if temporary increase BUG_ON()...
In addition, I think that an important memory allocation should retry several times.
So, I propose the following patches as this sample.
>
> 2. long term
> - BUG_ON() is decreased by using the forced-readonly framework(already posted by Liu Bo),
> etc.
This patch retries kmem_cache_alloc() in start_transaction() several times.
Signed-off-by: Tsutomu Itoh <t-itoh@jp.fujitsu.com>
---
fs/btrfs/transaction.c | 22 +++++++++++++++++++++-
1 file changed, 21 insertions(+), 1 deletion(-)
diff -urNp linux-2.6.38-rc1/fs/btrfs/transaction.c linux-2.6.38-rc1.new/fs/btrfs/transaction.c
--- linux-2.6.38-rc1/fs/btrfs/transaction.c 2011-01-19 08:14:02.000000000 +0900
+++ linux-2.6.38-rc1.new/fs/btrfs/transaction.c 2011-01-21 14:08:02.000000000 +0900
@@ -22,6 +22,7 @@
#include <linux/writeback.h>
#include <linux/pagemap.h>
#include <linux/blkdev.h>
+#include <linux/ratelimit.h>
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
@@ -175,6 +176,25 @@ static int may_wait_transaction(struct b
return 0;
}
+#define MAX_ITERATIONS 10
+
+static struct btrfs_trans_handle *alloc_trans_handle(void)
+{
+ struct btrfs_trans_handle *ret;
+ int i = 0;
+
+ ret = kmem_cache_alloc(btrfs_trans_handle_cachep, GFP_NOFS);
+ if (!ret) {
+ pr_notice_ratelimited("ENOMEM in %s, retrying.\n", __func__);
+ do {
+ yield();
+ ret = kmem_cache_alloc(btrfs_trans_handle_cachep,
+ GFP_NOFS);
+ } while (!ret && i++ < MAX_ITERATIONS);
+ }
+ return ret;
+}
+
static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root,
u64 num_items, int type)
{
@@ -185,7 +205,7 @@ static struct btrfs_trans_handle *start_
if (root->fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR)
return ERR_PTR(-EROFS);
again:
- h = kmem_cache_alloc(btrfs_trans_handle_cachep, GFP_NOFS);
+ h = alloc_trans_handle();
if (!h)
return ERR_PTR(-ENOMEM);
next prev parent reply other threads:[~2011-01-21 6:06 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-01-20 6:19 [PATCH] btrfs: fix return value check of btrfs_start_transaction() Tsutomu Itoh
2011-01-20 16:09 ` Josef Bacik
2011-01-20 23:47 ` Tsutomu Itoh
2011-01-21 6:06 ` Tsutomu Itoh [this message]
2011-01-28 21:53 ` Chris Mason
2011-01-31 0:03 ` Tsutomu Itoh
2011-02-01 2:15 ` Tsutomu Itoh
2011-02-01 12:38 ` Chris Mason
2011-01-21 1:59 ` liubo
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=4D392265.1020003@jp.fujitsu.com \
--to=t-itoh@jp.fujitsu.com \
--cc=chris.mason@oracle.com \
--cc=josef@redhat.com \
--cc=linux-btrfs@vger.kernel.org \
/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).