From: "Rémi Denis-Courmont" <remi@remlab.net> To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org> Subject: Re: [FFmpeg-devel] [PATCH 1/4] configure: aarch64: Support assembling the dotprod and i8mm arch extensions Date: Wed, 31 May 2023 19:29:24 +0300 Message-ID: <1762536.A2Pd2Nc0NH@basile.remlab.net> (raw) In-Reply-To: <c872991-cafc-a014-1874-12d3b82e3d0@martin.st> Le tiistaina 30. toukokuuta 2023, 15.25.25 EEST Martin Storsjö a écrit : > On Sun, 28 May 2023, Rémi Denis-Courmont wrote: > > Le sunnuntaina 28. toukokuuta 2023, 0.34.15 EEST Martin Storsjö a écrit : > >> I guess the alternative would be to just try to set .arch > >> <highest-supported-that-we-care-about>. I was worried that support for > >> e.g. armv8.6-a appeared later in toolchains than support for the > >> individual extension i8mm, but at least from a quick browse in binutils > >> history, they seem to have been added at the same time, so there's > >> probably no such drawback. > >> > >> Or what's the situation with e.g. SVE2 - was ".arch_extension sve2" > >> supported significantly earlier than ".arch armv9-a"? > > > > I have not tested SVE on LLVM. AFAIK, SVE and SVE2 are optional from 8.2 > > and 9.0 onward respectively, and not mandatory in any version, so if your > > toolchain supports neither .arch with plus sign, nor .arch_extension, it > > is game over. > > I didn't meant specifically whether LLVM supports it here, just in general > wrt binutils and how to enable the feature. > > FWIW it seems like SVE2 is a mandatory part of 9.0 Yes and no. SVE requires ARMv8.2, and SVE2 requires ARMv9.0, but neither are ever required by any existing ARM version. DDI0487 is abundantly clear that SVE2 is OPTIONAL: "[The Scalable Vector Extension version 2 (SVE2)] feature is supported in AArch64 state only. This feature is OPTIONAL in an Armv9.0 implementation," However it adds that "standard Armv9-A software platforms support FEAT_SVE2." > - assembling SVE2 instructions can be done with ".arch armv9-a". Presumably binutils targets "standard Armv9-A software platforms" by default, as opposed to just any "Armv9.0 implementation". I guess that means that any Cortex, Neoverse or other proper ARM design will include SVE2, but big-ass architecture licensees such as APPL and NVDA are not required to include SVE2 in their own designs. > But there are about 2 years > worth of deployed binutils based toolchains that do recognize ".arch > armv8.2-a; .arch_extension sve2" but don't recognize ".arch armv9-a". This is not entirely surprising, since SVE is much older than ARMv9, even if SVE2 requires ARMv9. -- レミ・デニ-クールモン http://www.remlab.net/ _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
next prev parent reply other threads:[~2023-05-31 16:29 UTC|newest] Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top 2023-05-26 8:03 Martin Storsjö 2023-05-26 8:03 ` [FFmpeg-devel] [PATCH 2/4] aarch64: Add cpu flags for the dotprod and i8mm extensions Martin Storsjö 2023-05-26 8:03 ` [FFmpeg-devel] [PATCH 3/4] aarch64: Add linux runtime cpu feature detection using getauxval(AT_HWCAP) Martin Storsjö 2023-05-27 9:04 ` Rémi Denis-Courmont 2023-05-27 20:35 ` Martin Storsjö 2023-05-28 5:58 ` Rémi Denis-Courmont 2023-05-26 8:03 ` [FFmpeg-devel] [PATCH 4/4] aarch64: Add Apple runtime detection of dotprod and i8mm using sysctl Martin Storsjö 2023-05-27 8:44 ` [FFmpeg-devel] [PATCH 1/4] configure: aarch64: Support assembling the dotprod and i8mm arch extensions Rémi Denis-Courmont 2023-05-27 21:34 ` Martin Storsjö 2023-05-28 5:50 ` Rémi Denis-Courmont 2023-05-30 12:25 ` Martin Storsjö 2023-05-31 16:29 ` Rémi Denis-Courmont [this message] 2023-06-01 11:04 ` Martin Storsjö
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=1762536.A2Pd2Nc0NH@basile.remlab.net \ --to=remi@remlab.net \ --cc=ffmpeg-devel@ffmpeg.org \ /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
Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel This inbox may be cloned and mirrored by anyone: git clone --mirror https://master.gitmailbox.com/ffmpegdev/0 ffmpegdev/git/0.git # If you have public-inbox 1.1+ installed, you may # initialize and index your mirror using the following commands: public-inbox-init -V2 ffmpegdev ffmpegdev/ https://master.gitmailbox.com/ffmpegdev \ ffmpegdev@gitmailbox.com public-inbox-index ffmpegdev Example config snippet for mirrors. AGPL code for this site: git clone https://public-inbox.org/public-inbox.git