linux-kernel - Re: [PATCH] LoongArch: ptrace: Merge ptrace_hbp_set

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 for Android: free password hash cracker in your pocket

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

Message-ID: <429e94ee-ee77-3a5c-a2f7-c0d027fb7c4c@loongson.cn>
Date: Wed, 5 Nov 2025 11:12:35 +0800
From: Tiezhu Yang <yangtiezhu@...ngson.cn>
To: Huacai Chen <chenhuacai@...nel.org>, Bill Tsui <b10902118@....edu.tw>
Cc: oleg@...hat.com, kernel@...0n.name, loongarch@...ts.linux.dev,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH] LoongArch: ptrace: Merge ptrace_hbp_set_*()

Hi Bill,

On 2025/11/5 上午8:57, Huacai Chen wrote:
> Hi, Tiezhu,
> 
> On Wed, Oct 29, 2025 at 11:38 PM Bill Tsui <b10902118@....edu.tw> wrote:
>>
>> In hw_break_set(), those functions actually can be combined into one.
>> This eliminates intermediate calls to modify_user_hw_breakpoint() that
>> may leave hardware registers in a partially updated state.
>>
>> This redundancy was originally found in ARM ptrace, where it caused
>> non-4-byte address alignment checks to fail.
>>
>> Link:
>> https://lore.kernel.org/all/20251018133731.42505-2-b10902118@ntu.edu.tw/
>>
>> The LoongArch implementation appears to have been derived from ARM,
>> so this refactor helps avoid the same issue and simplifies future
>> maintenance.
> What do you think about it?

Can you describe directly what is wrong with the current code
on LoongArch? Please do not show the ARM's under-review patch
link in the commit message.

Does this patch fix real problem? If there is a test case to
reproduce the problem, please add it into the commit message.

If there is no functional change, just to refactor the code or
avoid the potential issues, please give a explicit description.
We will test GDB to see whether there is a regression.

Thanks,
Tiezhu