public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@kernel.org>
To: "Thomas Weißschuh" <thomas.weissschuh@linutronix.de>
Cc: LKML <linux-kernel@vger.kernel.org>,
	"Mathieu Desnoyers" <mathieu.desnoyers@efficios.com>,
	"André Almeida" <andrealmeid@igalia.com>,
	"Sebastian Andrzej Siewior" <bigeasy@linutronix.de>,
	"Carlos O'Donell" <carlos@redhat.com>,
	"Peter Zijlstra" <peterz@infradead.org>,
	"Florian Weimer" <fweimer@redhat.com>,
	"Rich Felker" <dalias@aerifal.cx>,
	"Torvald Riegel" <triegel@redhat.com>,
	"Darren Hart" <dvhart@infradead.org>,
	"Ingo Molnar" <mingo@kernel.org>,
	"Davidlohr Bueso" <dave@stgolabs.net>,
	"Arnd Bergmann" <arnd@arndb.de>,
	"Liam R . Howlett" <Liam.Howlett@oracle.com>
Subject: Re: [patch 8/8] x86/vdso: Implement __vdso_futex_robust_try_unlock()
Date: Tue, 17 Mar 2026 10:51:47 +0100	[thread overview]
Message-ID: <87bjgmokb0.ffs@tglx> (raw)
In-Reply-To: <20260317081606-6b19f6a3-34e5-4056-9b8b-e9b90028b960@linutronix.de>

On Tue, Mar 17 2026 at 08:25, Thomas Weißschuh wrote:
> On Mon, Mar 16, 2026 at 06:13:34PM +0100, Thomas Gleixner wrote:
>> +/*
>> + * Compat enabled kernels have to take the size bit into account to support the
>> + * mixed size use case of gaming emulators. Contrary to the kernel robust unlock
>> + * mechanism all of this does not test for the 32-bit modifier in 32-bit VDSOs
>> + * and in compat disabled kernels. User space can keep the pieces.
>> + */
>> +#if defined(CONFIG_X86_64) && !defined(BUILD_VDSO32_64)
>
> #ifndef __x86_64__ ?

#ifdef :)

Just had to double check and convince myself that __x86_64__ is set when
building for the X86_X32 ABI. Seems to work.

>> +uint32_t __vdso_futex_robust_try_unlock(uint32_t *lock, uint32_t tid, void *pop)
>
> While uint32_t is originally a userspace type, in the kernel it also pulls in
> other internal types which are problematic in the vDSO. __u32 from
> uapi/linux/types.h avoid this issue.

Sure.

>> --- a/arch/x86/entry/vdso/vdso64/vdsox32.lds.S
>> +++ b/arch/x86/entry/vdso/vdso64/vdsox32.lds.S
>> @@ -22,6 +22,12 @@ VERSION {
>>  		__vdso_getcpu;
>>  		__vdso_time;
>>  		__vdso_clock_getres;
>> +#ifdef CONFIG_FUTEX_ROBUST_UNLOCK
>> +		__vdso_futex_robust_try_unlock;
>
>> +		__vdso_futex_robust_try_unlock_cs_start;
>> +		__vdso_futex_robust_try_unlock_cs_success;
>> +		__vdso_futex_robust_try_unlock_cs_end;
>
> These three symbols are not meant to be used from outside the vDSO
> implementation, so they don't need to be exported by the linkerscripts.

You are right in principle and I had it differently in the first place
until I realized that exposing them is useful for debugging and
validation purposes.

As I pointed out in the cover letter you can use GDB to actually verify
the fixup magic. That makes it obviously possible to write a user space
selftest without requiring to decode the internals of the VDSO.

Due to my pretty limited userspace DSO knowledge that was the best I
came up with. If you have a better idea, please let me know.

Thanks,

        tglx

  reply	other threads:[~2026-03-17  9:51 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-16 17:12 [patch 0/8] futex: Address the robust futex unlock race for real Thomas Gleixner
2026-03-16 17:12 ` [patch 1/8] futex: Move futex task related data into a struct Thomas Gleixner
2026-03-16 17:55   ` Mathieu Desnoyers
2026-03-17  2:24   ` André Almeida
2026-03-17  9:52     ` Thomas Gleixner
2026-03-16 17:13 ` [patch 2/8] futex: Move futex related mm_struct " Thomas Gleixner
2026-03-16 18:00   ` Mathieu Desnoyers
2026-03-16 17:13 ` [patch 3/8] futex: Provide UABI defines for robust list entry modifiers Thomas Gleixner
2026-03-16 18:02   ` Mathieu Desnoyers
2026-03-17  2:38   ` André Almeida
2026-03-17  9:53     ` Thomas Gleixner
2026-03-16 17:13 ` [patch 4/8] futex: Add support for unlocking robust futexes Thomas Gleixner
2026-03-16 18:24   ` Mathieu Desnoyers
2026-03-17 16:17   ` André Almeida
2026-03-17 20:46     ` Peter Zijlstra
2026-03-17 22:40       ` Thomas Gleixner
2026-03-18  8:02         ` Peter Zijlstra
2026-03-18  8:06           ` Florian Weimer
2026-03-18 14:47           ` Peter Zijlstra
2026-03-18 16:03             ` Thomas Gleixner
2026-03-16 17:13 ` [patch 5/8] futex: Add robust futex unlock IP range Thomas Gleixner
2026-03-16 18:36   ` Mathieu Desnoyers
2026-03-17 19:19   ` André Almeida
2026-03-16 17:13 ` [patch 6/8] futex: Provide infrastructure to plug the non contended robust futex unlock race Thomas Gleixner
2026-03-16 18:35   ` Mathieu Desnoyers
2026-03-16 20:29     ` Thomas Gleixner
2026-03-16 20:52       ` Mathieu Desnoyers
2026-03-16 17:13 ` [patch 7/8] x86/vdso: Prepare for robust futex unlock support Thomas Gleixner
2026-03-16 17:13 ` [patch 8/8] x86/vdso: Implement __vdso_futex_robust_try_unlock() Thomas Gleixner
2026-03-16 19:19   ` Mathieu Desnoyers
2026-03-16 21:02     ` Thomas Gleixner
2026-03-16 22:35       ` Mathieu Desnoyers
2026-03-16 21:14     ` Thomas Gleixner
2026-03-16 21:29     ` Thomas Gleixner
2026-03-17  7:25   ` Thomas Weißschuh
2026-03-17  9:51     ` Thomas Gleixner [this message]
2026-03-17 11:17       ` Thomas Weißschuh
2026-03-18 16:17         ` Thomas Gleixner
2026-03-19  7:41           ` Thomas Weißschuh
2026-03-19  8:53             ` Florian Weimer
2026-03-19  9:04               ` Thomas Weißschuh
2026-03-19  9:08               ` Peter Zijlstra
2026-03-19 23:31                 ` Thomas Gleixner
2026-03-19 10:36             ` Sebastian Andrzej Siewior
2026-03-19 10:49               ` Thomas Weißschuh
2026-03-19 10:55                 ` Sebastian Andrzej Siewior
2026-03-17  8:28   ` Florian Weimer
2026-03-17  9:36     ` Thomas Gleixner
2026-03-17 10:37       ` Florian Weimer
2026-03-17 22:32         ` Thomas Gleixner
2026-03-18 22:08           ` Thomas Gleixner
2026-03-18 22:10             ` Peter Zijlstra
2026-03-19  2:05             ` André Almeida
2026-03-19  7:10               ` Thomas Gleixner
2026-03-17 15:33   ` Uros Bizjak
2026-03-18  8:21     ` Thomas Gleixner
2026-03-18  8:32       ` Uros Bizjak

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87bjgmokb0.ffs@tglx \
    --to=tglx@kernel.org \
    --cc=Liam.Howlett@oracle.com \
    --cc=andrealmeid@igalia.com \
    --cc=arnd@arndb.de \
    --cc=bigeasy@linutronix.de \
    --cc=carlos@redhat.com \
    --cc=dalias@aerifal.cx \
    --cc=dave@stgolabs.net \
    --cc=dvhart@infradead.org \
    --cc=fweimer@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.org \
    --cc=thomas.weissschuh@linutronix.de \
    --cc=triegel@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox