By searching in the Chromium gerrit for the git commit history of the addition of the ymm register, it's found that Intel had been working for 256bit wide vector instructions since 2021. ( https://chromium-review.googlesource.com/q/owner:jie....@intel.com) .
>From the CL https://chromium-review.googlesource.com/c/v8/v8/+/3411357 you can see re-vectorization phase is added to the wasm pipeline so that for Intel platforms 256bit-wide vector can be used to squeeze hardware computation power. 'Qiaowen Yang' via v8-users <v8-users@googlegroups.com> 于2023年2月15日周三 17:36写道： > I found that x86 assembler includes ymm registers, which are 256-bits in > length. I am wondering what this will be used in V8. > V8 does not have auto-vectorization, and the simd instructions in V8 all > come from webassembly, which only has a `v128` data type. So, how will ymm > registers be used in V8? > > -- > -- > v8-users mailing list > v8-users@googlegroups.com > http://groups.google.com/group/v8-users > --- > You received this message because you are subscribed to the Google Groups > "v8-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to v8-users+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/v8-users/2f54d449-d6e6-4184-9465-b90c189a629cn%40googlegroups.com > <https://groups.google.com/d/msgid/v8-users/2f54d449-d6e6-4184-9465-b90c189a629cn%40googlegroups.com?utm_medium=email&utm_source=footer> > . > -- -- v8-users mailing list v8-users@googlegroups.com http://groups.google.com/group/v8-users --- You received this message because you are subscribed to the Google Groups "v8-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to v8-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/v8-users/CA%2BZLJ-EAxy2R6JXB8vKc_rcVKQJH5WwNhcROAQn7xK5Nutwyeg%40mail.gmail.com.
