linux-kernel - Re: [PATCH v2] MIPS: Fix MAX_REG_OFFSET and remove zero-length struct member

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

Message-Id: <B71034AC-B0FC-4C5F-8562-661D6AD11056@linux.dev>
Date: Fri, 18 Apr 2025 15:38:36 +0200
From: Thorsten Blum <thorsten.blum@...ux.dev>
To: "Maciej W. Rozycki" <macro@...am.me.uk>
Cc: Thomas Bogendoerfer <tsbogend@...ha.franken.de>,
 Oleg Nesterov <oleg@...hat.com>,
 linux-mips@...r.kernel.org,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] MIPS: Fix MAX_REG_OFFSET and remove zero-length struct
 member

On 18. Apr 2025, at 14:44, Maciej W. Rozycki wrote:
> On Fri, 18 Apr 2025, Thorsten Blum wrote:
>>> Though the fix is incorrect for CPU_CAVIUM_OCTEON, because it doesn't 
>>> allow one to access the second half of the last register, and I find it 
>>> exceedingly complex anyway.  Just:
>>> 
>>> #define MAX_REG_OFFSET \
>>> (offsetof(struct pt_regs, __last) - sizeof(unsigned long))
>>> 
>>> will do (as `regs_get_register' operates on `unsigned long' quantities).
>> 
>> Does regs_get_register() even work for CPU_CAVIUM_OCTEON when accessing
>> the last two registers because they're both ULL, not UL? (independent of
>> my patch)
> 
> Or rather two arrays of registers.  With 32-bit configurations their 
> contents have to be retrieved by pieces.  I don't know if it's handled by 
> the caller(s) though as I'm not familiar with this interface.

Ah, CPU_CAVIUM_OCTEON seems to be 64-bit only, so there's no difference
between UL and ULL. Then both my patch and your suggestion:

  #define MAX_REG_OFFSET (offsetof(struct pt_regs, __last) - sizeof(unsigned long))

should be fine.

I still prefer my approach without '__last[0]' because it also silences
the following false-positive Coccinelle warning, which is how I stumbled
upon this in the first place:

  ./ptrace.h:51:15-21: WARNING use flexible-array member instead

Would it make sense to also change the register arrays 'mpl' and 'mtp'
from ULL to UL? ULL seems unnecessarily confusing to me.

Thanks,
Thorsten