699e1b395e
Rollup merge of #153931 - cyrgani:old-consts, r=Mark-Simulacrum
...
remove usages of to-be-deprecated numeric constants
Split out from rust-lang/rust#146882 .
2026-03-23 12:14:55 +01:00
3776c07842
Merge pull request #2050 from folkertdev/aarch64-rand
...
add ACLE random number generation intrinsics
2026-03-16 21:03:10 +00:00
a893257cea
remove usages of to-be-deprecated numeric constants
2026-03-15 22:14:11 +00:00
ccf7d4e978
inline assert_instr tests
2026-03-13 11:09:47 +01:00
30c24f4530
ci: update to actions/checkout@v6
...
ci is showing a lot of warnings (72) right now. apparently
actions/checkout@v4 uses Node.js 20, and all github actions are
scheduled to be force opted-in to Node.js 24 on 2026-06-02.
I don't anticipate bumping the checkout action to v6 / Node.js 24
to cause any issues (Node.js 24 drops support for ARM32 and macOS
versions <= 13.4, but this shouldn't matter because we use Docker
to test in those environments, not github runners natively) but if
it does cause issues it's probably better to find out now rather
than by surprise 3 months from now... :)
2026-03-12 16:05:20 -07:00
4f10c640d1
remove cfg_attr on aarch64/arm64ec in the aarch64 module
2026-03-12 10:37:52 +01:00
b245493113
Merge pull request #2061 from RalfJung/f64-tests
...
s390x: add f64 tests for vec_min
2026-03-12 09:15:10 +00:00
ebd8272229
Merge pull request #2053 from folkertdev/aarch64-int-simd-reduce-min-max
...
aarch64: use `simd_reduce_{min, max}` on integers
2026-03-12 09:13:08 +00:00
80b869876a
s390x: add f64 tests for vec_min
2026-03-11 18:33:14 +01:00
a6687175c9
go back to portable LLVM intrinsic to avoid fallback trouble
2026-03-11 09:05:25 +01:00
47d82e245a
add f32 min/max tests
2026-03-10 23:29:35 +01:00
752a87fb4b
s390x: use llvm.s390 intrinsics instead of simd_fmin/fmax
2026-03-10 22:43:59 +01:00
f326a429fa
Merge pull request #2051 from rust-lang/rustc-pull
...
Rustc pull update
2026-03-10 12:04:48 +00:00
46bada398f
aarch64: use simd_reduce_{min, max} on integers
2026-03-10 13:03:00 +01:00
f1bd647047
Revert "Merge pull request #1871 from folkertdev/aarch64-float-min-max"
...
This reverts commit 6a8a764262df5e65c06bc5d9180046a636a53ce9, reversing
changes made to a37563b5f8cec0c873864b786c33d00386189916.
2026-03-10 12:19:33 +01:00
5cc969050b
Merge ref 'eda4fc7733ee' from rust-lang/rust
...
Pull recent changes from https://github.com/rust-lang/rust via Josh.
Upstream ref: rust-lang/rust@eda4fc7733
Filtered ref: rust-lang/stdarch@3f3adc6031
Upstream diff: https://github.com/rust-lang/rust/compare/139651428df86cf88443295542c12ea617cbb587...eda4fc7733ee89e484d7120cafbd80dcb2fce66e
This merge was created using https://github.com/rust-lang/josh-sync .
2026-03-09 04:45:01 +00:00
279c43e727
Prepare for merging from rust-lang/rust
...
This updates the rust-version file to eda4fc7733
2026-03-09 04:42:45 +00:00
78157ddde9
Replace version placeholders with 1.95.0
...
(cherry picked from commit bad24ccbec
2026-03-07 10:42:01 -08:00
6c7d342e10
add ACLE random number generation intrinsics
2026-03-06 21:49:21 +01:00
34943378f6
Merge pull request #2048 from sayantn/movrs
...
Add `movrs` intrinsics
2026-03-06 16:50:55 +00:00
c61045d520
Merge pull request #2047 from sayantn/correct-bw-const
...
Correct const-stability attribute for avx512bw intrinsics
2026-03-06 16:50:39 +00:00
6abb95cc00
gate use of wasm_target_feature on wasm target arch
2026-03-06 14:09:59 +01:00
48d573f8e5
Add movrs intrinsics
2026-03-05 18:27:37 +05:30
3483792e0f
Merge pull request #2044 from sayantn/fix-signatures
...
Fix LLVM intrinsic signatures for AVX-VNNI
2026-03-04 21:34:44 +00:00
985cd2399a
Add immediate AMX intrinsics
2026-03-05 02:35:38 +05:30
35f7c8bc53
Correct stability attribute for avx512bw intrinsics
2026-03-05 00:43:31 +05:30
a24022ad4e
Fix LLVM intrinsic signatures for AVX-VNNI
2026-03-02 00:32:24 +05:30
e00790eb2c
Merge pull request #2042 from folkertdev/aarch64-non-power-of-two-reads-writes
...
aarch64: fix UB in non-power-of-two reads and writes
2026-02-27 21:30:39 +00:00
d06233df70
Merge pull request #2040 from ArunTamil21/add-test-mm-comige-ss
...
Add missing runtime test for _mm_comige_ss and fix _mm_comigt_ss test
2026-02-26 13:18:12 +00:00
2fa37e7b34
aarch64: fix UB in non-power-of-two reads and writes
2026-02-26 13:00:21 +01:00
f981ffe959
Remove _mm_comige_ss from skip list in x86-intel.rs
2026-02-26 09:19:29 +00:00
cb3046e5f2
Add missing runtime test for _mm_comige_ss and fix _mm_comigt_ss test
2026-02-25 23:56:42 +00:00
f01bcce3e2
update to resolver = 3
2026-02-25 15:00:04 +01:00
820ca48038
Update Intel SDE version to 10.5
2026-02-25 04:49:15 +05:30
f4cc7ee8de
Merge pull request #2035 from okaneco/const_packs
...
x86: Followup to add const for pack intrinsics and tests
2026-02-24 03:25:49 +00:00
47ef7f73de
aarch64: cleanup of some long array literals
2026-02-23 11:58:18 +01:00
ff558be420
Merge pull request #2032 from folkertdev/interleaving-store
...
use `intrinsics::simd` for interleaving store
2026-02-21 15:33:06 +00:00
37bd99e2e0
Add const to avx512bw intrinsics
2026-02-20 07:33:34 -05:00
4f881c4b89
Add const to avx2 intrinsics
2026-02-20 07:33:34 -05:00
17a4c8ea8f
Add const to sse41 intrinsics
2026-02-20 07:02:29 -05:00
22d10c516d
x86: Followup to add const for pack intrinsics
...
Add const to `sse2`, `sse41`, `avx2`, and `avx512bw` functions and tests
2026-02-20 06:53:07 -05:00
c7ecbd5e4b
Use intrinsics for avx512bw
2026-02-19 08:45:27 -05:00
639b07ec75
Use intrinsics for avx2
2026-02-19 08:45:27 -05:00
6fa33bea62
Use intrinsics for sse41
2026-02-19 08:45:27 -05:00
a5ba7941f9
x86: use simd::intrinsics for saturating packs
...
Use intrinsics for `sse2`, `sse41`, `avx2`, `avx512bw`
The majority of implementations make use of `simd_shuffle` since that
optimized through to the avx512 intrinsics that made use of the lower
target feature intrinsics. Combined with masked stores, instruction
tests would fail presumably due to the casting and clamping that
the compiler couldn't see through. This is a known weakness as seen
in the other masked stores like the truncating conversion stores.
2026-02-19 08:42:51 -05:00
2b656f99a5
use intrinsics::simd for aarch64 interleaving st
2026-02-18 22:27:23 +01:00
afb890e048
use intrinsics::simd for interleaving store of f16
2026-02-18 22:03:33 +01:00
e219383c8f
use intrinsics::simd for interleaving store of int64x1
2026-02-18 22:02:02 +01:00
c0c9d025d5
use intrinsics::simd for interleaving store
2026-02-18 22:02:02 +01:00
f0fd6f78d4
cleanup long shuffle mask literals
2026-02-18 21:31:48 +01:00
Jonathan Brouwer