linux-kernel - Re: [RESEND PATCH v4 1/4] dma-mapping: benchmark: Add padding to ensure uABI remained consistent

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: <ed07e57d-00e0-44b3-b53e-b5295aee8561@samsung.com>
Date: Mon, 23 Jun 2025 14:01:23 +0200
From: Marek Szyprowski <m.szyprowski@...sung.com>
To: Barry Song <21cnbao@...il.com>, Jonathan Cameron
	<Jonathan.Cameron@...wei.com>
Cc: Qinxin Xia <xiaqinxin@...wei.com>, robin.murphy@....com,
	yangyicong@...wei.com, hch@....de, iommu@...ts.linux.dev,
	prime.zeng@...wei.com, fanghao11@...wei.com, linux-kernel@...r.kernel.org,
	linuxarm@...wei.com
Subject: Re: [RESEND PATCH v4 1/4] dma-mapping: benchmark: Add padding to
 ensure uABI remained consistent

On 16.06.2025 12:40, Barry Song wrote:
> On Mon, Jun 16, 2025 at 9:53 PM Jonathan Cameron
> <Jonathan.Cameron@...wei.com> wrote:
>> On Sat, 14 Jun 2025 22:34:51 +0800
>> Qinxin Xia <xiaqinxin@...wei.com> wrote:
>>> The padding field in the structure was previously reserved to
>>> maintain a stable interface for potential new fields, ensuring
>>> compatibility with user-space shared data structures.
>>> However,it was accidentally removed by tiantao in a prior commit,
>>> which may lead to incompatibility between user space and the kernel.
>>>
>>> This patch reinstates the padding to restore the original structure
>>> layout and preserve compatibility.
>>>
>>> Fixes: 8ddde07a3d28 ("dma-mapping: benchmark: extract a common header file for map_benchmark definition")
>>> Cc: stable@...r.kernel.org
>>> Acked-by: Barry Song <baohua@...nel.org>
>>> Signed-off-by: Qinxin Xia <xiaqinxin@...wei.com>
>> FWIW I checked the patch above indeed accidentally dropped the padding and the structure
>> is copied to userspace so this fix is correct.  Given it's not in a uapi header this
>> only really affects the selftest I think this is mostly a case of there possibly being
>> out of tree tools with a local copy of this structure definition.
> Somehow, I feel we have placed
> tools/testing/selftests/dma/dma_map_benchmark.c in the wrong location.
> As a selftest, it should have a mechanism to check kernel dependencies,
> start properly and automatically, and report pass or fail.
>
> dma_map_benchmark.c seems more like a tool that belongs in tools/dma,
> rather than a test.

Indeed imho it would be better to move it out of selftests directory.

Best regards
-- 
Marek Szyprowski, PhD
Samsung R&D Institute Poland