Re: [PATCH v2 0/2] x86/vdso: Replace deprecated simple_strtoul() with kstrtouint()

From: Thorsten Blum

Date: Wed Oct 07 2026 - 04:32:03 EST


On Tue, Oct 06, 2026 at 08:12:59PM +0000, Habil Eren Türker wrote:
> > I submitted a very similar series a while ago [1], but haven't received
> > feedback yet (besides a Reviewed-by from Randy on the documentation
> > change).
>
> Hi Thorsten,
>
> You are right, and thank you for pointing this out. My series is essentially
> a duplicate of yours. Your series came first.
>
> I will withdraw my v2 series and let yours be the one to land.
>
> In the meantime, I took the time to review and test your series on top
> of v7.3-rc5. Both the code and the runtime behaviour look correct to me:
>
> Code review:
> - kstrtouint() preserves the original semantics of simple_strtoul()
> while adding proper overflow checking.
> - Error paths are consistent with the existing vdso_setup() logic.
> - The documentation removal for vdso32=2 is accurate.
>
> Runtime testing:
> - x86_64 QEMU/q35 guest, TCG acceleration, qemu64 CPU model
> - KASAN profile (KASAN + KFENCE + DEBUG_OBJECTS)
> - memory suite at 256M/512M/1G, 3 iterations per case
> - All cases pass, no KASAN/KCSAN/lockdep findings, CRIT_ERRORS=0
>
> Reviewed-by: Habil Eren Türker <habilerenturker@xxxxxxxxxxx>
> Tested-by: Habil Eren Türker <habilerenturker@xxxxxxxxxxx>

Thank you for reviewing and testing my series, but I believe you must
add your Reviewed-by: and Tested-by: to my series [1], not as a reply to
your own series.

> Sorry for the duplicate noise.

No worries.

Thanks,
Thorsten

[1] https://lore.kernel.org/lkml/20260831101657.2947-5-blum@xxxxxxxxxx/