usamoi
|
a4a7184370
|
mark gfni, vaes, vpclmulqdq intrinsics as safe
|
2025-05-30 17:35:18 +00:00 |
|
Ralf Jung
|
9561e8e0a3
|
cmpxchg16b: use atomic_compare_exchange from libcore
|
2025-05-30 17:14:51 +00:00 |
|
Ralf Jung
|
b58ce625a2
|
allow aarch64_softfloat_neon for backwards compatibility
|
2025-05-21 10:39:23 +00:00 |
|
Folkert de Vries
|
7b995254d4
|
use a tuple to return the condition code
|
2025-05-20 16:39:24 +00:00 |
|
Folkert de Vries
|
42e6ad3ee8
|
use the right load instruction
|
2025-05-20 14:26:46 +00:00 |
|
Folkert de Vries
|
a850e8738a
|
avx512_target_feature is now stable on nightly
|
2025-05-20 14:26:46 +00:00 |
|
WANG Rui
|
63235a6ba5
|
Partially stabilize LoongArch target features
|
2025-05-12 19:47:49 +00:00 |
|
sayantn
|
546e26518f
|
Run aarch64-pc-windows-msvc runs on the new windows-11-arm runners
|
2025-05-06 16:10:50 +00:00 |
|
James Barford-Evans
|
c65583c2b3
|
Pr feedback for instruction & hookup CI for aarch64_be
|
2025-05-03 05:06:15 +00:00 |
|
James Barford-Evans
|
ef92b59242
|
fix - aarch64_be tests
|
2025-05-03 05:06:15 +00:00 |
|
sayantn
|
f8782f778b
|
Fix errors in incorrect SAE and ROUNDING parameters
|
2025-05-03 05:03:31 +00:00 |
|
sayantn
|
ed25a9bde4
|
Implement sha512 , sm3 and sm4 intrinsics
|
2025-05-01 11:30:47 +00:00 |
|
sayantn
|
400dab3df7
|
Fix stdarch-verify
|
2025-05-01 11:30:47 +00:00 |
|
sayantn
|
a21396870c
|
Add avx512vl requirement to testsuite for avx512fp16 128 and 256 bit
|
2025-04-29 11:59:15 +00:00 |
|
sayantn
|
288ae22354
|
Fix errors in decoupling avx512vl and avx512dq from avx512fp16
|
2025-04-29 11:59:15 +00:00 |
|
sayantn
|
caf463ae41
|
Fix CI errors due to alignment issues in msvc
|
2025-04-29 11:59:15 +00:00 |
|
sayantn
|
e61df091c1
|
Remove workarounds for llvm/llvm-project#98306
|
2025-04-23 10:40:09 +00:00 |
|
Folkert de Vries
|
f77386cf2b
|
run powerpc64le assert_instr on CI
|
2025-04-23 10:38:19 +00:00 |
|
Ralf Jung
|
1e3c5e3c99
|
use 'unadjusted' ABI for wasm LLVM intrinsics
|
2025-04-21 17:59:41 +00:00 |
|
sayantn
|
bc9e29d9be
|
Replace cfg(stdarch_intel_sde) with STDARCH_TEST_SKIP_FUNCTION
|
2025-04-20 21:39:39 +00:00 |
|
sayantn
|
6cc17153d9
|
Re-enable all conditionally-disabled x86 assert_instr tests
|
2025-04-20 21:39:39 +00:00 |
|
bendn
|
3dd82b698f
|
allow unnecessary transmutes
|
2025-04-17 18:27:37 +00:00 |
|
Folkert de Vries
|
c899659c44
|
add vec_extract , vec_insert , vec_promote and vec_insert_and_zero
|
2025-04-17 17:06:09 +00:00 |
|
sayantn
|
e804386587
|
Change void* type for gather /scatter intrinsics
|
2025-04-17 13:59:07 +00:00 |
|
sayantn
|
d6b0426127
|
Change void* type for cvt_storeu intrinsics
|
2025-04-17 13:59:07 +00:00 |
|
sayantn
|
376c3f33c5
|
Change void* type for compressstore intrinsics
|
2025-04-17 13:59:07 +00:00 |
|
sayantn
|
b10a6817d9
|
Change void* type for 3 intrinsics
- `_mm512_load_si512`
- `_mm512_loadu_si512`
- `_mm512_stream_si512`
|
2025-04-17 13:59:07 +00:00 |
|
bendn
|
b5046525d5
|
fix broken intra doc links
|
2025-04-11 11:30:32 +00:00 |
|
bendn
|
3ea2e5600f
|
allow unnecessary transmutes
|
2025-03-27 21:36:07 +00:00 |
|
Folkert de Vries
|
9c8f736849
|
make documentation headers consistent
this now always uses the name as specified by the official docs
|
2025-03-26 13:50:38 +00:00 |
|
Folkert de Vries
|
69a7cb8be3
|
add s390x to the module docs
|
2025-03-26 13:50:38 +00:00 |
|
James Barford-Evans
|
2099eccb2e
|
pr feedback - remove the commented out vcombine_f16
|
2025-03-25 10:53:54 +00:00 |
|
James Barford-Evans
|
08b9752ee1
|
refactor - arm_shared intrinsics are now YAML, where possible use anchor
tags
|
2025-03-25 10:53:54 +00:00 |
|
Vadim Petrochenkov
|
ebe8804680
|
sse42: Add unsafe blocks around unsafe function calls
to fix the `unsafe_op_in_unsafe_fn` lint
|
2025-03-24 23:26:29 +00:00 |
|
David Pathakjee
|
a083e2a6c8
|
Minor correction to __m512d documentation.
A 512-bit register is f64x8, not f64x4. Likely a copy-paste error from the _m256d documentation,
which seems correct.
|
2025-03-24 23:26:12 +00:00 |
|
Ralf Jung
|
a03fdf1441
|
use consistent wording around the 'undefined' intrinsics, and slightly expand their docs
|
2025-03-20 22:51:52 +00:00 |
|
WANG Rui
|
14ddd423fb
|
Incldue loongarch64 in the list of other architectures
|
2025-03-20 21:54:32 +00:00 |
|
Folkert de Vries
|
81f6100463
|
move unsafe pointer writes to the surface
|
2025-03-16 20:02:22 +00:00 |
|
Folkert de Vries
|
1f202e7fe7
|
shink the size of type signatures
|
2025-03-16 20:02:22 +00:00 |
|
Folkert de Vries
|
3b8973d216
|
add vec_meadd , vec_moadd , vec_mhadd and vec_mladd
|
2025-03-16 20:02:22 +00:00 |
|
Folkert de Vries
|
44cda131be
|
add vec_mulh
|
2025-03-16 20:02:22 +00:00 |
|
Folkert de Vries
|
f7dc654108
|
add vec_mulo
|
2025-03-16 20:02:22 +00:00 |
|
Folkert de Vries
|
5c61b664e7
|
add vec_any_* and vec_all_*
|
2025-03-16 20:02:22 +00:00 |
|
Folkert de Vries
|
e30905984c
|
add vec_all_nan , vec_any_nan , vec_all_numeric and vec_any_numeric
|
2025-03-16 20:02:22 +00:00 |
|
Folkert de Vries
|
2667fd06ff
|
add vec_cmpeq_idx and variations
|
2025-03-16 20:02:22 +00:00 |
|
Folkert de Vries
|
b866e43594
|
add vec_cmpeq and vec_cmpne
|
2025-03-16 20:02:22 +00:00 |
|
Folkert de Vries
|
8affaabfaf
|
add vec_cmpgt , vec_cmplt , vec_cmpge , vec_cmple
|
2025-03-16 20:02:22 +00:00 |
|
Folkert de Vries
|
81a2841164
|
let's not use &mut until we get confirmation it's OK
|
2025-03-16 20:02:22 +00:00 |
|
Folkert de Vries
|
d218bd1624
|
add vec_cmprg_or_0_idx_cc and vec_cmpnrg_or_0_idx_cc
|
2025-03-16 20:02:22 +00:00 |
|
Folkert de Vries
|
790d77e4e6
|
add vec_cmprg_or_0_idx and vec_cmpnrg_or_0_idx
|
2025-03-16 20:02:22 +00:00 |
|