linux-kernel - Re: [PATCH] bitops/32: Convert variable_ffs() and fls() zero-case handling to C

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: <CAHk-=wjpUbDYw5O+zXWMq8yq91zaZFLUgZJ0Y1PU1g-OCiftrQ@mail.gmail.com>
Date: Tue, 29 Apr 2025 15:34:30 -0700
From: Linus Torvalds <torvalds@...ux-foundation.org>
To: Andrew Cooper <andrew.cooper3@...rix.com>
Cc: "H. Peter Anvin" <hpa@...or.com>, Ingo Molnar <mingo@...nel.org>, Arnd Bergmann <arnd@...db.de>, 
	Arnd Bergmann <arnd@...nel.org>, Thomas Gleixner <tglx@...utronix.de>, Ingo Molnar <mingo@...hat.com>, 
	Borislav Petkov <bp@...en8.de>, Dave Hansen <dave.hansen@...ux.intel.com>, x86@...nel.org, 
	Juergen Gross <jgross@...e.com>, Boris Ostrovsky <boris.ostrovsky@...cle.com>, 
	Alexander Usyskin <alexander.usyskin@...el.com>, 
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>, Mateusz Jończyk <mat.jonczyk@...pl>, 
	Mike Rapoport <rppt@...nel.org>, Ard Biesheuvel <ardb@...nel.org>, Peter Zijlstra <peterz@...radead.org>, 
	linux-kernel@...r.kernel.org, xen-devel@...ts.xenproject.org
Subject: Re: [PATCH] bitops/32: Convert variable_ffs() and fls() zero-case
 handling to C

On Tue, 29 Apr 2025 at 15:22, Andrew Cooper <andrew.cooper3@...rix.com> wrote:
>
> Oh, I didn't realise there was also a perf difference too, but Agner Fog
> agrees.

The perf difference is exactly because of the issue where the non-rep
one acts as a cmov, and has basically two inputs (the bits to test in
the source, and the old value of the result register)

I guess it's not "fundamental", but lzcnt is basically a bit simpler
for hardware to implement, and the non-rep legacy bsf instruction
basically has a dependency on the previous value of the result
register.

So even when it's a single uop for both cases, that single uop can be
slower for the bsf because of the (typically false) dependency and
extra pressure on the rename registers.

       Linus