[PATCH AUTOSEL 6.1 13/22] proc/vmcore: fix potential memory leak in vmcore_init()

[PATCH AUTOSEL 6.1 13/22] proc/vmcore: fix potential memory leak in vmcore_init()

[Sasha Levin]

From: Jianglei Nie <niejianglei2021@163.com>

[ Upstream commit 12b9d301ff73122aebd78548fa4c04ca69ed78fe ]

Patch series "Some minor cleanup patches resent".

The first three patches trivial clean up patches.

And for the patch "kexec: replace crash_mem_range with range", I got a
ibm-p9wr ppc64le system to test, it works well.

This patch (of 4):

elfcorehdr_alloc() allocates a memory chunk for elfcorehdr_addr with
kzalloc().  If is_vmcore_usable() returns false, elfcorehdr_addr is a
predefined value.  If parse_crash_elf_headers() gets some error and
returns a negetive value, the elfcorehdr_addr should be released with
elfcorehdr_free().

Fix it by calling elfcorehdr_free() when parse_crash_elf_headers() fails.

Link: https://lkml.kernel.org/r/20220929042936.22012-1-bhe@redhat.com
Link: https://lkml.kernel.org/r/20220929042936.22012-2-bhe@redhat.com
Signed-off-by: Jianglei Nie <niejianglei2021@163.com>
Signed-off-by: Baoquan He <bhe@redhat.com>
Acked-by: Baoquan He <bhe@redhat.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Chen Lifu <chenlifu@huawei.com>
Cc: "Eric W . Biederman" <ebiederm@xmission.com>
Cc: Li Chen <lchen@ambarella.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Petr Mladek <pmladek@suse.com>
Cc: Russell King <linux@armlinux.org.uk>
Cc: ye xingchen <ye.xingchen@zte.com.cn>
Cc: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/proc/vmcore.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c
index f2aa86c421f2..74747571d58e 100644
--- a/fs/proc/vmcore.c
+++ b/fs/proc/vmcore.c
@@ -1567,6 +1567,7 @@ static int __init vmcore_init(void)
 		return rc;
 	rc = parse_crash_elf_headers();
 	if (rc) {
+		elfcorehdr_free(elfcorehdr_addr);
 		pr_warn("Kdump: vmcore not initialized\n");
 		return rc;
 	}
-- 
2.35.1


_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

Re: [PATCH AUTOSEL 6.1 13/22] proc/vmcore: fix potential memory leak in vmcore_init()

[Andrew Morton]

On Sat, 17 Dec 2022 10:27:14 -0500 Sasha Levin <sashal@kernel.org> wrote:

> From: Jianglei Nie <niejianglei2021@163.com>
> 
> [ Upstream commit 12b9d301ff73122aebd78548fa4c04ca69ed78fe ]
> 
> Patch series "Some minor cleanup patches resent".
> 
> The first three patches trivial clean up patches.
>
> And for the patch "kexec: replace crash_mem_range with range", I got a
> ibm-p9wr ppc64le system to test, it works well.
> 
> This patch (of 4):
> 
> elfcorehdr_alloc() allocates a memory chunk for elfcorehdr_addr with
> kzalloc().  If is_vmcore_usable() returns false, elfcorehdr_addr is a
> predefined value.  If parse_crash_elf_headers() gets some error and
> returns a negetive value, the elfcorehdr_addr should be released with
> elfcorehdr_free().

This is exceedingly minor - a single memory leak per boot, under very
rare circumstances.


With every patch I merge I consider -stable.  Often I'll discuss the
desirability of a backport with the author and with reviewers.  Every
single patch.  And then some damn script comes along and overrides that
quite careful decision.  argh.

Can we please do something like

	if (akpm && !cc:stable)
		dont_backport()

And even go further - if your script thinks it might be something we
should backport and if it didn't have cc:stable then contact the
author, reviewers and committers and ask them to reconsider before we
go and backport it.  This approach will have the advantage of training
people to consider the backport more consistently.


I'd (still) like to have a new patch tag like Not-For-Stable: or
cc:not-stable or something to tell your scripts "yes, we thought about
it and we decided no".

_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

Re: [PATCH AUTOSEL 6.1 13/22] proc/vmcore: fix potential memory leak in vmcore_init()

[Sasha Levin]

On Sat, Dec 17, 2022 at 04:32:28PM -0800, Andrew Morton wrote:
>On Sat, 17 Dec 2022 10:27:14 -0500 Sasha Levin <sashal@kernel.org> wrote:
>
>> From: Jianglei Nie <niejianglei2021@163.com>
>>
>> [ Upstream commit 12b9d301ff73122aebd78548fa4c04ca69ed78fe ]
>>
>> Patch series "Some minor cleanup patches resent".
>>
>> The first three patches trivial clean up patches.
>>
>> And for the patch "kexec: replace crash_mem_range with range", I got a
>> ibm-p9wr ppc64le system to test, it works well.
>>
>> This patch (of 4):
>>
>> elfcorehdr_alloc() allocates a memory chunk for elfcorehdr_addr with
>> kzalloc().  If is_vmcore_usable() returns false, elfcorehdr_addr is a
>> predefined value.  If parse_crash_elf_headers() gets some error and
>> returns a negetive value, the elfcorehdr_addr should be released with
>> elfcorehdr_free().
>
>This is exceedingly minor - a single memory leak per boot, under very
>rare circumstances.
>
>
>With every patch I merge I consider -stable.  Often I'll discuss the
>desirability of a backport with the author and with reviewers.  Every
>single patch.  And then some damn script comes along and overrides that
>quite careful decision.  argh.
>
>Can we please do something like
>
>	if (akpm && !cc:stable)
>		dont_backport()

Yup, I already had it set for 'akpm && mm/ && !cc:stable', happy to
remove the 'mm/' restriction if you're doing the same for the rest of
the patches you review.

>And even go further - if your script thinks it might be something we
>should backport and if it didn't have cc:stable then contact the
>author, reviewers and committers and ask them to reconsider before we
>go and backport it.  This approach will have the advantage of training
>people to consider the backport more consistently.

This is what this mail is all about: I haven't queued up the patch yet,
it gives folks week+ to review, and all it takes is a simple "no" for me
to drop it.

>I'd (still) like to have a new patch tag like Not-For-Stable: or
>cc:not-stable or something to tell your scripts "yes, we thought about
>it and we decided no".

No objections on my part.

-- 
Thanks,
Sasha

_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec