Re: [PATCH v3 5/5] riscv: vdso: Implement __vdso_futex_robust_try_unlock()

From: Nam Cao <namcao@linutronix.de>
To: "Thomas Weißschuh" <thomas.weissschuh@linutronix.de>
Cc: "Paul Walmsley" <pjw@kernel.org>,
	"Palmer Dabbelt" <palmer@dabbelt.com>,
	"Albert Ou" <aou@eecs.berkeley.edu>,
	"Alexandre Ghiti" <alex@ghiti.fr>,
	linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org,
	"André Almeida" <andrealmeid@igalia.com>,
	"Mathieu Desnoyers" <mathieu.desnoyers@efficios.com>,
	"Thomas Gleixner" <tglx@kernel.org>,
	"Peter Zijlstra" <peterz@infradead.org>,
	"Sebastian Andrzej Siewior" <bigeasy@linutronix.de>
Subject: Re: [PATCH v3 5/5] riscv: vdso: Implement __vdso_futex_robust_try_unlock()
Date: Mon, 20 Jul 2026 13:37:59 +0200	[thread overview]
Message-ID: <87ldb5evs8.fsf@yellow.woof> (raw)
In-Reply-To: <20260717142423-c5be3c7c-d4cb-4a68-8e85-3e3754e42666@linutronix.de>

Thomas Weißschuh <thomas.weissschuh@linutronix.de> writes:
> Can it ever happen that the toolchain has differing zacas support between
> 32-bit and 64-bit? Then cpu_supports_zacas() may not be sufficient.

No idea. Probably not.

But if it does differ, we will get a clear compile-time error message
(unrecognized instruction or something like that). We can worry about it then.

> If it is guaranteed to be in sync I would prefer to reuse
> CONFIG_TOOLCHAIN_HAS_ZACAS instead of the cc-option check in the Makefile
> to make that clear and error out early if it should ever change.

I intend to apply Vivian's .option suggestion. That won't be required then.
>
> Also this assumes that compat vDSO alternative patching works.
> But this is not the case. My series [0] to implement it has not yet been
> applied. This revision is missing the note about this dependency.
>
> [0] https://lore.kernel.org/all/20260630-riscv-vdso32-alternative-v1-0-a32fd89b7b1c@linutronix.de/

Right, thanks.

> Could be :
>
> vdso-syms += __vdso_futex_robust_list$(BITS)_try_unlock

Wasn't aware that BITS exists, nice!

Nam