rust-lang/rust
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
307,470 Commits 1 Branch 0 Tags
359bfa901b9d8eddbed7b88b6716dbb77b7bfb95
Commit Graph

2629 Commits

Author SHA1 Message Date
Martin Liska
63603fd41f Add space in order to fix rendering of Rounding options 2024-09-22 17:57:31 -04:00
Martin Liska
f319d8c5d9 Make documentation about __mm_$op_ps more precise 2024-09-22 11:55:20 -04:00
Martin Liska
8376fbe34f Provide link for rounding parameters - SSE 2024-09-22 11:32:58 -04:00
Martin Liska
f8813384e7 Provide link for rounding parameters 2024-09-22 11:32:58 -04:00
Martin Liska
fcf6e95c90 Change markdown for ROUNDING params in f16c.rs 2024-09-22 11:32:58 -04:00
Martin Liska
04f4ca9680 Ignore XSTATE_BV when comparing XTEST XsaveArea 2024-09-22 07:56:46 -04:00
Sayantan Chakraborty
d6f40ec495 Fixes the shrd bug 2024-09-22 07:33:16 -04:00
Kajetan Puchalski
485ded6369 std_detect: Add sme-b16b16 as an explicit aarch64 target feature 
LLVM 20 split out what used to be called b16b16 and correspond to aarch64
FEAT_SVE_B16B16 into sve-b16b16 and sme-b16b16.
Add sme-b16b16 as an explicit feature and update the detection accordingly.
 2024-09-18 12:48:54 -04:00
Taiki Endo
24b5e3c99e Use C string literal 2024-09-18 05:58:40 -04:00
Taiki Endo
9e5b3f5c8c std_detect: Fix link in mips.rs 2024-09-18 05:30:10 -04:00
Taiki Endo
720973b1b4 std_detect: Use elf_aux_info on FreeBSD 2024-09-18 05:07:32 -04:00
Mads Marquart
8a511191a0 Enable feature detection on all Apple/Darwin targets 
Tested in the simulator and on the device I had lying around, a 1st
generation iPad Mini (which isn't Aarch64, but shows that the
`sysctlbyname` calls still work even there, even if they return false).

`sysctlbyname` _should_ be safe to use without causing rejections from
the app store, as its usage is documented in:
https://developer.apple.com/documentation/kernel/1387446-sysctlbyname/determining_instruction_set_characteristics

Also, the standard library will use these soon anyhow, so this shouldn't
affect the situation:
https://github.com/rust-lang/rust/pull/129019
 2024-09-14 04:25:01 +01:00
Mads Marquart
0fb034979a Add Mac Catalyst to CI 2024-09-14 04:01:52 +01:00
Mads Marquart
e7ccd5763f Remove unnecessary SDKROOT/MACOS_DEPLOYMENT_TARGET setup in CI 
rustc will figure these out by itself.
 2024-09-14 04:01:52 +01:00
Ralf Jung
4eed3c8a23 simd_extract, simd_insert: use absolute path in macro 2024-08-28 14:55:57 +01:00
Ralf Jung
7ccb13c178 simd_shuffle: pass the idx argument as a vector 2024-08-28 11:28:07 +01:00
dheaton-arm
03f144596f Fix test verification and allow for generic types in intrinsics 2024-08-27 15:14:28 +01:00
dheaton-arm
92b6a3abae Add MTE intrinsics 
Adds intrinsic functions to `core_arch::aarch64` for MTE, as per the
ACLE:
- __arm_mte_create_random_tag
- __arm_mte_increment_tag
- __arm_mte_exclude_tag
- __arm_mte_set_tag
- __arm_mte_get_tag
- __arm_mte_ptrdiff

These are unavailable without the `mte` target feature.
 2024-08-27 15:14:28 +01:00
James Barford-Evans
4f6820bd1b fix rebase, remove the const_vector attribure 
formatting

remove comma

put back square brackets
 2024-08-21 01:18:59 +01:00
George Wort
2caf80145b Use #[rustc_intrinsic_const_vector_arg] for Neon intrinsics 
This allows the constant vector to be passed directly
to LLVM which fixes a few issues where the LLVM intrinsic
expects a constant vector.
 2024-08-21 01:18:59 +01:00
Kjetil Kjeka
869ddbeb66 NVPTX: Add f16 SIMD intrinsics 2024-08-19 20:16:10 +01:00
Eduardo Sánchez Muñoz
f23656fbb6 Avoid legacy numeric constants 2024-08-19 17:01:20 +01:00
Eduardo Sánchez Muñoz
c103c015a8 Add loongarch64-unknown-linux-gnu to CI 2024-08-19 16:38:59 +01:00
Eduardo Sánchez Muñoz
76d71ff37e Use full path to refer simd_shuffle intrinsic from simd_shuffle! macro 
Avoids needing to import `crate::intrinsics::simd::simd_shuffle` in each file where `simd_shuffle!` is used and fixes loongarch64
 2024-08-17 18:07:04 +01:00
Eduardo Sánchez Muñoz
d9f0480fae Remove #![feature(asm_const)] 2024-08-17 18:07:04 +01:00
Luca Versari
87567faed1 Clarify the layout documentation for x86 SIMD types. 2024-08-16 14:33:04 +01:00
Luca Versari
d54fe45ffb Document the layout of x86 SIMD types. 2024-08-14 14:57:25 +01:00
Scott McMurray
4f02e2f8a9 Keep Debug the same 
This ended up way more annoying than expected, because I needed to refactor the `types!` macro to separate out the stability attributes to put those on the `Debug` impl without also copying all the `#[doc]`s.  But I like how it came out in the end, reducing duplication in the macro invocations.

But without it all the C-consistency tests fail.
 2024-08-08 23:47:25 +01:00
Scott McMurray
1bf1eff5cc Move entirely to array-based SIMD 
See MCP#621

This tries to make as few changes as possible -- it keeps the `new` functions taking all the parameters, for example.
 2024-08-08 23:47:25 +01:00
Sayantan Chakraborty
0c304072bc Remove the dummy function 2024-08-04 12:10:17 +01:00
Sayantan Chakraborty
200905e0e9 Fix _mm_stream_si64 2024-08-03 22:58:47 +01:00
Jonas Fierlings
47068b1a06 Fix markdown list in docs 2024-08-03 19:24:37 +01:00
ziyizhang-1
50cd4ef0c5 initial commit to enable amx 
AMX Intrinsics:

amx-tile:
  - _tile_loadconfig
  - _tile_storeconfig
  - _tile_loadd
  - _tile_release
  - _tile_stored
  - _tile_stream_loadd
  - _tile_zero
amx-int8:
  - _tile_dpbssd
  - _tile_dpbsud
  - _tile_dpbusd
  - _tile_dpbuud
amx-bf16:
  - _tile_dpbf16ps
amx-fp16
  - _tile_dpfp16ps
amx-complex
  - _tile_cmmimfp16ps
  - _tile_cmmrlfp16ps
 2024-08-03 19:02:09 +01:00
sayantn
4a13560ede Update Intrinsics List to v3.6.9 
Add `#[inline]` to avx512ifma intrinsics
Fix the test equality.
Remove the stability attributes in simd types and test functions
 2024-07-26 12:20:06 +01:00
sayantn
3cf2b7d74f AVX512FP16 Part 9: Remaining avx512fp16 and avxneconvert 2024-07-26 12:20:06 +01:00
sayantn
318e9ec7e7 AVX512FP16 Part 8: Convert from f16 2024-07-26 12:20:06 +01:00
sayantn
cea6530177 AVX512FP16 Part 7: Convert to f16 2024-07-26 12:20:06 +01:00
sayantn
734355993e AVX512FP16 Part 6: Remaining 
`cmpph`, `fpclass`, reduce, `blend`, `permutex`
 2024-07-26 12:20:06 +01:00
sayantn
debe317dcf AVX512FP16 Part 5: FP-Support 
`getexp`, `getmant`, `roundscale`, `scalef`, `reduce`
 2024-07-26 12:20:06 +01:00
sayantn
c024ef206f AVX512FP16 Part 4: Math functions 
Reciprocal, RSqrt, Sqrt, Max, Min
 2024-07-26 12:20:06 +01:00
sayantn
b88dfd6c03 AVX512FP16 Part 3: FMA 2024-07-26 12:20:06 +01:00
sayantn
7be9f610e3 AVX512_FP16 Part 2: Complex Multiplication 2024-07-26 12:20:06 +01:00
sayantn
60dfe5f264 AVX512FP16 Part 1 
Add-Sub-Mul-Div, Load-Store-Move, `comi`, `set`
 2024-07-26 12:20:06 +01:00
sayantn
c878b773d5 AVX512FP16 Part 0: Types 2024-07-26 12:20:06 +01:00
daxpedda
a1ad6bf8be Move Wasm's relaxed SIMD to Rust v1.82 2024-07-25 16:38:08 +01:00
sayantn
74f53212a0 Stabilize simd_x86_updates 2024-07-25 16:07:35 +01:00
Yuri Astrakhan
dd87060bf3 Minor lints for stdarch-gen-arm/src/main.rs 
Just a few minor cleanups
 2024-07-25 15:41:21 +01:00
Kajetan Puchalski
351ec5744c std_detect: Update aarch64 feature dependencies to LLVM upstream 
Feature dependencies for newer aarch64 fetaures differ between LLVM 18
in the Rust tree and upstream LLVM 19.
This commit updates those dependencies to reflect new LLVM upstream
changes.
 2024-07-25 15:18:37 +01:00
Kajetan Puchalski
41dc17d3e5 std_detect: Sort aarch64 features 
Alphabetically sort the list of aarch64 features.
The list was getting a bit too chaotic so it was worth properly
sorting.
 2024-07-25 15:18:37 +01:00
Kajetan Puchalski
ef538bc614 std_detect: Add aarch64/linux/LLVM SME features 
Add detection for SME features supported by LLVM and the Linux Kernel.
Include commented-out hwcap fields for features supported by Linux but not by LLVM.

This commit adds feature detection for the following features:

- FEAT_SME
- FEAT_SME_F16F16
- FEAT_SME_F64F64
- FEAT_SME_F8F16
- FEAT_SME_F8F32
- FEAT_SME_FA64
- FEAT_SME_I16I64
- FEAT_SME_LUTv2
- FEAT_SME2
- FEAT_SME2p1
- FEAT_SSVE_FP8DOT2
- FEAT_SSVE_FP8DOT4
- FEAT_SSVE_FP8FMA

Linux features: https://github.com/torvalds/linux/blob/master/arch/arm64/include/uapi/asm/hwcap.h
LLVM features: llvm-project/llvm/lib/Target/AArch64/AArch64.td
 2024-07-25 15:18:37 +01:00