linux-kernel - Re: [PATCH] riscv: kgdb: Ensure that BUFMAX

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: <2f6dd420-2bfd-794d-182c-74934577272d@kernel.org>
Date: Wed, 8 Oct 2025 11:28:02 -0600 (MDT)
From: Paul Walmsley <pjw@...nel.org>
To: Miquel Sabaté Solà <mikisabate@...il.com>
cc: linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org, 
    pjw@...nel.org, palmer@...belt.com, alex@...ti.fr, wangyuli@...ontech.com
Subject: Re: [PATCH] riscv: kgdb: Ensure that BUFMAX > NUMREGBYTES

On Mon, 29 Sep 2025, Miquel Sabaté Solà wrote:

> Miquel Sabaté Solà @ 2025-09-15 16:32 +02:
> 
> > The current value of BUFMAX is similar as in other architectures, but as
> > per documentation on KGDB (see
> > 'Documentation/process/debugging/kgdb.rst'), BUFMAX has to be larger
> > than NUMREGBYTES.
> >
> > Some NUMREGBYTES architectures (e.g. powerpc or hexagon) actually define
> > BUFMAX in relation to NUMREGBYTES, and thus this condition is always
> > guaranteed. Since 2048 is a value that is generally accepted on all
> > architectures, and that is larger than the current value of NUMREGBYTES,
> > we can keep this value in arch/riscv, but we can at least add an
> > 'static_assert' as an extra measure just in case NUMREGBYTES changes in
> > the future for some unforseen reason.
> >
> > Signed-off-by: Miquel Sabaté Solà <mikisabate@...il.com>

[ ... ]

> Gentle ping :)

Thanks, queued for early v6.18-rc.


- Paul