[PATCH] Btrfs: handle kmalloc() failure in btrfs_file_aio_write()

[PATCH] Btrfs: handle kmalloc() failure in btrfs_file_aio_write()

[JieJing.Zhang]

set errno to -ENOMEM and goto out

Signed-off-by: JieJing.Zhang <kzjeef@gmail.com>
---
 fs/btrfs/file.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index e354c33..e0fcb6c 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -926,6 +926,10 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
 		     PAGE_CACHE_SIZE, PAGE_CACHE_SIZE /
 		     (sizeof(struct page *)));
 	pages = kmalloc(nrptrs * sizeof(struct page *), GFP_KERNEL);
+	if (!pages) {
+		ret = -ENOMEM;
+		goto out;
+	}
 
 	/* generic_write_checks can change our pos */
 	start_pos = pos;
-- 
1.7.0.4

Re: [PATCH] Btrfs: handle kmalloc() failure in btrfs_file_aio_write()

[Zhang JieJing]

2010/6/20 JieJing.Zhang <kzjeef@gmail.com>
>
> set errno to -ENOMEM and goto out
>
> Signed-off-by: JieJing.Zhang <kzjeef@gmail.com>
> ---
> =A0fs/btrfs/file.c | =A0 =A04 ++++
> =A01 files changed, 4 insertions(+), 0 deletions(-)
>
> diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
> index e354c33..e0fcb6c 100644
> --- a/fs/btrfs/file.c
> +++ b/fs/btrfs/file.c
> @@ -926,6 +926,10 @@ static ssize_t btrfs_file_aio_write(struct kiocb=
 *iocb,
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 PAGE_CACHE_SIZE, PAGE_CACHE_S=
IZE /
> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 (sizeof(struct page *)));
> =A0 =A0 =A0 =A0pages =3D kmalloc(nrptrs * sizeof(struct page *), GFP_=
KERNEL);
I think this gfp should changed to GFP_NOFS to avoid deadlock, right ?
like this commit: 33818bbb84cd371b63ed8849cc5264d24c8b3aa2
> + =A0 =A0 =A0 if (!pages) {
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 ret =3D -ENOMEM;
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 goto out;
> + =A0 =A0 =A0 }
>
> =A0 =A0 =A0 =A0/* generic_write_checks can change our pos */
> =A0 =A0 =A0 =A0start_pos =3D pos;
> --
> 1.7.0.4
>
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" =
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] Btrfs: handle kmalloc() failure in btrfs_file_aio_write()

[Zhang JieJing]

This is the improved of the former patch.

>From d9c4cf7dd0e4489a31610e2fb2565ee29f0e4179 Mon Sep 17 00:00:00 2001
From: JieJing.Zhang <kzjeef@gmail.com>
Date: Sun, 4 Jul 2010 01:17:54 +0800
Subject: [PATCH] Btrfs: handle memory alloc failure in btrfs_file_aio_write()

1. kmalloc() should add a check of return valule.

2. In later code it memset zero to the allocated memory,
   so use kzalloc() instead of kmalloc() when allocation is better.

Signed-off-by: JieJing.Zhang <kzjeef@gmail.com>
---
 fs/btrfs/file.c |    7 +++++--
 1 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index e354c33..961612c 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -925,7 +925,11 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
 	nrptrs = min((iov_iter_count(&i) + PAGE_CACHE_SIZE - 1) /
 		     PAGE_CACHE_SIZE, PAGE_CACHE_SIZE /
 		     (sizeof(struct page *)));
-	pages = kmalloc(nrptrs * sizeof(struct page *), GFP_KERNEL);
+	pages = kzalloc(nrptrs * sizeof(struct page *), GFP_KERNEL);
+	if (!pages) {
+		ret = -ENOMEM;
+		goto out;
+	}

 	/* generic_write_checks can change our pos */
 	start_pos = pos;
@@ -968,7 +972,6 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
 					PAGE_CACHE_SHIFT;

 		WARN_ON(num_pages > nrptrs);
-		memset(pages, 0, sizeof(struct page *) * nrptrs);

 		ret = btrfs_delalloc_reserve_space(inode, write_bytes);
 		if (ret)
-- 
1.7.0.4