linux-kernel - Re: [PATCH] kexec: Fix uninitialized struct kimage *image pointer

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <4F91083B8C9DC587+11236f66-0903-4fec-93e6-b628f9767934@uniontech.com>
Date: Mon, 15 Dec 2025 17:18:41 +0800
From: Qiang Ma <maqianga@...ontech.com>
To: Baoquan He <bhe@...hat.com>
Cc: akpm@...ux-foundation.org, kexec@...ts.infradead.org,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH] kexec: Fix uninitialized struct kimage *image pointer


在 2025/12/15 16:50, Baoquan He 写道:
> On 12/15/25 at 12:41pm, Qiang Ma wrote:
>> 在 2025/12/15 09:18, Baoquan He 写道:
>>> On 12/14/25 at 07:35pm, Qiang Ma wrote:
>>>> 在 2025/12/12 18:53, Baoquan He 写道:
>>>>> On 12/12/25 at 03:16pm, Qiang Ma wrote:
>>>>>> The image is initialized to NULL. Then, after calling kimage_alloc_init,
>>>>>> we can directly goto 'out' because at this time, the kimage_free will
>>>>>> determine whether image is a NULL pointer.
>>>>> Rechecked the code flow, in kimage_alloc_init(), if anything wrong, the
>>>>> allocated memory are all freed via out_free_control_pages and
>>>>> out_free_image accordingly, any place missed? If no, I think the current
>>>>> code is correctly handled.
>>>> I rechecked the code and found no omissions.
>>> Hmm, my bad, I didn't say my question clearly. I checked code, didn't
>>> find anything wrong in the current code. In kimage_alloc_init(), the
>>> allocated memory are all freed on failure, no memory leaked. Means you
>>> are fixing correct code.
>> Oh, I see. I recalled that this fix was in preparation for patch
>> "kexec: add kexec flag to control debug printing" for kexec_dbg_print
>> to be reset to false in kimage_free.
>>
>> In that case, I don't think this patch should be posted separately.
> If it's prepared for later patch, it should not be saying it's fixing
> issue. People may be confused and try to add it to stable kernel.
That makes sense. I renamed it: "kexec: Replace the goto out_unlock with 
out" in v2.
>>>>>> Signed-off-by: Qiang Ma <maqianga@...ontech.com>
>>>>>> ---
>>>>>>     kernel/kexec.c | 4 +++-
>>>>>>     1 file changed, 3 insertions(+), 1 deletion(-)
>>>>> Acked-by: Baoquan He <bhe@...hat.com>
>>>>>
>>>>>> diff --git a/kernel/kexec.c b/kernel/kexec.c
>>>>>> index 28008e3d462e..9bb1f2b6b268 100644
>>>>>> --- a/kernel/kexec.c
>>>>>> +++ b/kernel/kexec.c
>>>>>> @@ -95,6 +95,8 @@ static int do_kexec_load(unsigned long entry, unsigned long nr_segments,
>>>>>>     	unsigned long i;
>>>>>>     	int ret;
>>>>>> +	image = NULL;
>>>>>> +
>>>>>>     	/*
>>>>>>     	 * Because we write directly to the reserved memory region when loading
>>>>>>     	 * crash kernels we need a serialization here to prevent multiple crash
>>>>>> @@ -129,7 +131,7 @@ static int do_kexec_load(unsigned long entry, unsigned long nr_segments,
>>>>>>     	ret = kimage_alloc_init(&image, entry, nr_segments, segments, flags);
>>>>>>     	if (ret)
>>>>>> -		goto out_unlock;
>>>>>> +		goto out;
>>>>>>     	if (flags & KEXEC_PRESERVE_CONTEXT)
>>>>>>     		image->preserve_context = 1;
>>>>>> -- 
>>>>>> 2.20.1
>>>>>>
>