[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <87f74eab-ae51-08a4-5b7e-261f437146f4@basealt.ru>
Date: Mon, 29 Jan 2024 21:33:59 +0300
From: kovalev@...linux.org
To: Greg KH <greg@...ah.com>
Cc: stable@...r.kernel.org, linux-kernel@...r.kernel.org,
bpf@...r.kernel.org, netdev@...r.kernel.org, kpsingh@...nel.org,
john.fastabend@...il.com, yhs@...com, songliubraving@...com, kafai@...com,
andrii@...nel.org, daniel@...earbox.net, ast@...nel.org,
nickel@...linux.org, oficerovas@...linux.org, dutyrok@...linux.org
Subject: Re: [PATCH 5.10.y 0/1] bpf: fix warning ftrace_verify_code
29.01.2024 19:20, Greg KH wrote:
> On Mon, Jan 29, 2024 at 12:17:45PM +0300, kovalev@...linux.org wrote:
>> Syzkaller hit 'WARNING in ftrace_verify_code' bug.
>>
>> This bug is not a vulnerability and is reproduced only when running
>> with root privileges on stable 5.10 kernel.
> What about 5.15.y? We can't take a patch for older kernels and not for
> newer ones, right?
Indeed, this patch was not backported at 5.15.
I fixed it, successfully tested it on the 5.15.148 kernel and sent the
patch [1]
However, at the time of the kernel build, I noticed a compiler warning
about an unused "*tmp" variable,
that I fixed in the patch for 5.15, but remained in this patch 5.10:
...
+ void *new, *tmp;
...
need to fix on
...
+ void *new;
...
Therefore, I resend the patch with a fix (deleted the unused tmp
pointer). [2]
[1]
https://lore.kernel.org/bpf/20240129180108.284057-1-kovalev@altlinux.org/T/#u
[2]
https://lore.kernel.org/bpf/20240129183120.284801-1-kovalev@altlinux.org/T/#u
--
Regards,
Vasiliy Kovalev
Powered by blists - more mailing lists