[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <46e494ba-a1c5-4576-adbb-b63c77a73365@wanadoo.fr>
Date: Thu, 14 Nov 2024 02:47:38 +0900
From: Vincent Mailhol <mailhol.vincent@...adoo.fr>
To: Rasmus Villemoes <linux@...musvillemoes.dk>
Cc: Linus Torvalds <torvalds@...ux-foundation.org>,
Yury Norov <yury.norov@...il.com>,
Luc Van Oostenryck <luc.vanoostenryck@...il.com>,
linux-kernel@...r.kernel.org, linux-sparse@...r.kernel.org,
Rikard Falkeborn <rikard.falkeborn@...il.com>
Subject: Re: [RESEND PATCH v3 1/2] compiler.h: add _static_assert()
On 13/11/2024 at 05:03, Rasmus Villemoes wrote:
> On Wed, Nov 13 2024, Vincent Mailhol <mailhol.vincent@...adoo.fr> wrote:
>
>> Declare a new _statically_true() macro which, by making use of the
>> __builtin_choose_expr() and __is_constexpr(x) combo, always produces a
>> constant expression.
> Looks sane, but $subject needs s/_static_assert/_statically_true/. And
> to be completely pedantic, macros are defined, not declared.
Sorry for the silly mistake in the subject. And agreed that "define" is
more appropriate than "declare" when speaking of macros.
>> - above examples, and a bit more:
>>
>> https://godbolt.org/z/zzqM1ajPj
>>
>> - a proof that statically_true() does better constant folding than _statically_true()
>>
>> https://godbolt.org/z/vK6KK4hMG
> At least the second of these might be good to refer to in the commit
> message itself.
Ack. I rewrote the set of examples in the v4 and added a few samples to
the macro comment.
Yours sincerely,
Vincent Mailhol
Powered by blists - more mailing lists