T0b1/wasmtime
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

x64: Improve codegen for splats (#6025)

Browse Source 
This commit goes through the lowerings for the CLIF `splat` instruction
and improves the support for each operator. Many of these lowerings are
mirrored from v8/SpiderMonkey and there are a number of improvements:

* AVX2 `v{p,}broadcast*` instructions are added and used when available.
* Float-based splats are much simpler and always a single-instruction
* Integer-based splats don't insert into an uninit xmm value and instead
  start out with a `movd` to move into an `xmm` register. This
  thoeretically breaks dependencies with prior instructions since `movd`
  creates a fresh new value in the destination register.
* Loads are now sunk into all of the instructions. A new extractor,
  `sinkable_load_exact`, was added to sink the i8/i16 loads.
This commit is contained in:
Alex Crichton
2023-03-15 16:33:56 -05:00
committed by GitHub
parent a10c50afe9
commit d76f7ee52e
12 changed files with 1216 additions and 82 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
cranelift/filetests/filetests/runtests/simd-splat.clif
View File

@@ -4,6 +4,8 @@ target aarch64
target s390x
set enable_simd
target x86_64 has_sse3 has_ssse3 has_sse41
target x86_64 has_sse3 has_ssse3 has_sse41 has_avx
target x86_64 has_sse3 has_ssse3 has_sse41 has_avx has_avx2
function %splat_i8x16(i8) -> i8x16 {
block0(v0: i8):