On Tue, Jun 11, 2024 at 06:28:30PM +0300, Rémi Denis-Courmont wrote:
> Le tiistaina 11. kesäkuuta 2024, 16.15.19 EEST Michael Niedermayer a écrit :
> > On Fri, Jun 07, 2024 at 09:19:46PM +0300, Rémi Denis-Courmont wrote:
> > > C code or compiler built-ins are preferable over inline assembler for
> > > byte-swaps as it allows for better optimisations (e.g. instruction
> > > scheduling) which would otherwise be impossible.
> > > 
> > > As with f64c2e710fa1a7b59753224e717f57c48462076f for x86 and Arm,
> > > this removes the inline assembler on GCC (and Clang) since we now
> > > require recent enough compiler versions (this indeed seems to work on
> > > AArch64).
> > > ---
> > > 
> > >  libavutil/aarch64/bswap.h | 56 ---------------------------------------
> > >  libavutil/avr32/bswap.h   | 44 ------------------------------
> > >  libavutil/bswap.h         |  8 +-----
> > >  libavutil/sh4/bswap.h     | 48 ---------------------------------
> > 
> > As you are writing that this preferrable for better optimisations
> > Please provide benchmarks (for sh4, avr32)
> 
> How would someone benchmark an architecture like AVR32 that is not just dead 
> but barely even commercially existed at all, and for which there exist no 
> known C11 compiler and thus cannot even compile FFmpeg?

then simply remove avr32 with that explanation (no C11 compiler, and any other
reason)

but if a commit message says the code is removed because that "allows for better optimisations"
then yes i ask for benchmarks

thx

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

The worst form of inequality is to try to make unequal things equal.
-- Aristotle