[GIT PULL] NBD fixes

[GIT PULL] NBD fixes

[Markus Pargmann]

Hi Jens,

This is a pull request for one NBD patch. It fixes a memory leak which occures
when a module parameter has an invalid value.

It is based on v3.19-rc2. If you prefer a different base please let me know.

Best Regards,

Markus


The following changes since commit b7392d2247cfe6771f95d256374f1a8e6a6f48d6:

  Linux 3.19-rc2 (2014-12-28 16:49:37 -0800)

are available in the git repository at:

  git://git.pengutronix.de/git/mpa/linux-nbd.git tags/nbd_fixes_20150212

for you to fetch changes up to 473b3bfe169ca0b2192d7184786429bd75d359da:

  nbd: fix possible memory leak (2015-02-12 13:21:55 +0100)

----------------------------------------------------------------
NBD fixes based on v3.19-rc2

----------------------------------------------------------------
Sudip Mukherjee (1):
      nbd: fix possible memory leak

 drivers/block/nbd.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

-- 
2.1.4

[PATCH] nbd: fix possible memory leak

[Markus Pargmann]

From: Sudip Mukherjee <sudipm.mukherjee@gmail.com>

we have already allocated memory for nbd_dev, but we were not
releasing that memory and just returning the error value.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Acked-by: Paul Clements <Paul.Clements@SteelEye.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
---
 drivers/block/nbd.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index 4bc2a5cb9935..a98c41f72c63 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -803,10 +803,6 @@ static int __init nbd_init(void)
 		return -EINVAL;
 	}
 
-	nbd_dev = kcalloc(nbds_max, sizeof(*nbd_dev), GFP_KERNEL);
-	if (!nbd_dev)
-		return -ENOMEM;
-
 	part_shift = 0;
 	if (max_part > 0) {
 		part_shift = fls(max_part);
@@ -828,6 +824,10 @@ static int __init nbd_init(void)
 	if (nbds_max > 1UL << (MINORBITS - part_shift))
 		return -EINVAL;
 
+	nbd_dev = kcalloc(nbds_max, sizeof(*nbd_dev), GFP_KERNEL);
+	if (!nbd_dev)
+		return -ENOMEM;
+
 	for (i = 0; i < nbds_max; i++) {
 		struct gendisk *disk = alloc_disk(1 << part_shift);
 		if (!disk)
-- 
2.1.4

[PULL v2] NBD fixes

[Markus Pargmann]

Hi Jens,

This is essentially a resend of the pull request from last month
(nbd_fixes_20150212).

It includes one NBD patch that fixes a memory leak which occures when a module
parameter has an invalid value.

It is based now based on v4.0-rc1.

Best Regards,

Markus


The following changes since commit c517d838eb7d07bbe9507871fab3931deccff539:

  Linux 4.0-rc1 (2015-02-22 18:21:14 -0800)

are available in the git repository at:

  git://git.pengutronix.de/git/mpa/linux-nbd.git tags/nbd_fixes_20150305

for you to fetch changes up to ff6b8090e26ef7649ef0cc6b42389141ef48b0cf:

  nbd: fix possible memory leak (2015-03-05 08:51:03 +0100)

----------------------------------------------------------------
NBD fixes based on v4.0-rc1

----------------------------------------------------------------
Sudip Mukherjee (1):
      nbd: fix possible memory leak

 drivers/block/nbd.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

-- 
2.1.4

[PATCH] nbd: fix possible memory leak

[Markus Pargmann]

From: Sudip Mukherjee <sudipm.mukherjee@gmail.com>

we have already allocated memory for nbd_dev, but we were not
releasing that memory and just returning the error value.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Acked-by: Paul Clements <Paul.Clements@SteelEye.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Markus Pargmann <mpa@pengutronix.de>
---
 drivers/block/nbd.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index 4bc2a5cb9935..a98c41f72c63 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -803,10 +803,6 @@ static int __init nbd_init(void)
 		return -EINVAL;
 	}
 
-	nbd_dev = kcalloc(nbds_max, sizeof(*nbd_dev), GFP_KERNEL);
-	if (!nbd_dev)
-		return -ENOMEM;
-
 	part_shift = 0;
 	if (max_part > 0) {
 		part_shift = fls(max_part);
@@ -828,6 +824,10 @@ static int __init nbd_init(void)
 	if (nbds_max > 1UL << (MINORBITS - part_shift))
 		return -EINVAL;
 
+	nbd_dev = kcalloc(nbds_max, sizeof(*nbd_dev), GFP_KERNEL);
+	if (!nbd_dev)
+		return -ENOMEM;
+
 	for (i = 0; i < nbds_max; i++) {
 		struct gendisk *disk = alloc_disk(1 << part_shift);
 		if (!disk)
-- 
2.1.4

[PATCH] nbd: fix possible memory leak

[Sudip Mukherjee]

we have already allocated memory for nbd_dev, but we were not
releasing that memory and just returning the error value.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
---

Hi Paul,
after seeing your mail about maintainers, just started to see the code.
this patch is one way, another way can be to bring the kcalloc part below these checks. 

 drivers/block/nbd.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index 4bc2a5c..13ea239 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -822,11 +822,15 @@ static int __init nbd_init(void)
 		max_part = (1UL << part_shift) - 1;
 	}
 
-	if ((1UL << part_shift) > DISK_MAX_PARTS)
-		return -EINVAL;
+	if ((1UL << part_shift) > DISK_MAX_PARTS) {
+		err = -EINVAL;
+		goto out_free;
+	}
 
-	if (nbds_max > 1UL << (MINORBITS - part_shift))
-		return -EINVAL;
+	if (nbds_max > 1UL << (MINORBITS - part_shift)) {
+		err = -EINVAL;
+		goto out_free;
+	}
 
 	for (i = 0; i < nbds_max; i++) {
 		struct gendisk *disk = alloc_disk(1 << part_shift);
@@ -889,6 +893,7 @@ out:
 		blk_cleanup_queue(nbd_dev[i].disk->queue);
 		put_disk(nbd_dev[i].disk);
 	}
+out_free:
 	kfree(nbd_dev);
 	return err;
 }
-- 
1.8.1.2

Re: [PATCH] nbd: fix possible memory leak

[Paul Clements]

Sudip,

On Tue, Jan 13, 2015 at 12:19 PM, Sudip Mukherjee
<sudipm.mukherjee@gmail.com> wrote:
> we have already allocated memory for nbd_dev, but we were not
> releasing that memory and just returning the error value.
>
> Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> ---
>
> Hi Paul,
> after seeing your mail about maintainers, just started to see the code.
> this patch is one way, another way can be to bring the kcalloc part below these checks.

Probably better to move the kcalloc below, rather than adding more
branches, wouldn't you agree?

Thanks,
Paul