Ralf Jung
a34883b5d3
manually const-ify shuffle arguments ( #1160 )
2021-05-11 21:11:52 +01:00
SparrowLii
7516a80c31
Add vset neon instructions
2021-05-11 13:38:16 +01:00
Alex Crichton
2d9b71bca6
Add doc aliases for all wasm intrinsics
...
Recommended in #74372
2021-05-11 01:30:44 +01:00
Thom Chiovoloni
21c01768b7
Avoid using simd_f(min|max) in _mm256_(min|max)_p[sd]
2021-05-09 13:36:39 +01:00
Amanieu d'Antras
e9f73d0dc8
Fix asm! in bit-test intrinsics on x32
2021-05-08 19:40:07 +01:00
Amanieu d'Antras
994a4250a9
Use AT&T syntax to support LLVM 10
2021-05-07 23:19:18 +01:00
SparrowLii
8a2936b9a2
Completion of vcvt neon instruction
2021-05-07 23:02:39 +01:00
Ralf Jung
ed761b261c
remvoe const_fn leftovers
2021-05-07 17:51:27 +01:00
Pietro Albini
1d6ff635c0
remove cfg(not(bootstrap)) for 1.54
2021-05-07 00:31:04 +01:00
SparrowLii
911ace84b2
Add vqrdmulh, vqrdmlah, vqrdmlsh neon instructions
2021-05-06 15:44:54 +01:00
Alex Crichton
128aa9a7e5
Update docs for v128_any_true
2021-05-03 15:56:41 +01:00
Alex Crichton
1d92c1d8b2
Another round of wasm SIMD updates
...
This round is dependant on
https://github.com/rust-lang/llvm-project/pull/101 landing first in
rust-lang/rust and won't pass CI until that does. That PR, however, will
also break wasm CI because it's changing how the wasm target works. My
goal here is to open this early to get it out there so that when that PR
lands in rust-lang/rust and CI breaks in stdarch then this can be merged
to make CI green again.
The changes here are mostly around the codegen for various intrinsics.
Some wasm-specific intrinsics have been removed in favor of more general
LLVM intrinsics, and other intrinsics have been removed in favor of
pattern-matching codegen.
The only new instruction supported as part of this chagne is
`v128.any_true`. This leaves only one instruction unsupported in LLVM
which is `i64x2.abs`. I think the codegen for the instruction is correct
in stdsimd, though, and LLVM just needs to update with a pattern-match
to actually emit the opcode. That'll happen in a future LLVM update.
2021-05-03 15:56:41 +01:00
Sparrow Li
fd29f9602c
Add vmul_n, vmul_lane, vmulx neon instructions ( #1147 )
2021-04-30 21:09:41 +01:00
Sparrow Li
07f1d0cae3
Add vmla_n, vmla_lane, vmls_n, vmls_lane neon instructions ( #1145 )
2021-04-28 22:59:41 +01:00
scottmcm
54a2d8b82a
Remove #![feature(try_trait)] from a test ( #1142 )
...
I'm working on `try_trait_v2` which will break this, so I'm going
around removing uses from the rustc tree where I can.
2021-04-26 00:45:20 +01:00
Amanieu d'Antras
63daa088fd
Move cfg!(target_feature) directly into is_*_feature_detected!() ( #1141 )
...
Fixes #1135
2021-04-24 08:02:24 +01:00
Sparrow Li
8852d07441
add vcopy neon instructions ( #1139 )
2021-04-24 01:49:11 +01:00
Ralf Jung
03e109a2f3
remove unused const_fn feature ( #1140 )
2021-04-23 16:46:38 +01:00
Christopher Serr
a43f92a181
Add vrndn neon instructions ( #1086 )
...
This adds the neon instructions for lane-wise rounding without actually
converting the lanes to integers.
2021-04-22 06:08:40 +01:00
Sparrow Li
de3e8f72c5
Add vqdmul* neon instructions ( #1130 )
2021-04-21 15:27:08 +01:00
surechen
20c0120362
add neon instruction vaddlv_* ( #1129 )
2021-04-20 15:19:04 +01:00
Sparrow Li
6354de5993
Add vrshl, vrshr, vrshrn, vrsra, vsra neon instructions ( #1127 )
2021-04-19 17:49:44 +01:00
surechen
d46e0086e4
add neon instruction vfma_n_* ( #1122 )
2021-04-17 17:45:54 +01:00
Sebastian Thiel
43126c3f65
[DRAFT] intrinsics for all architectures appear in rustdoc ( #1104 )
2021-04-17 13:46:33 +01:00
Sparrow Li
e792dfd02c
add vqshl, vqshrn, vqshrun neon instructions ( #1120 )
2021-04-16 13:22:39 +01:00
Amanieu d'Antras
0073d6ace3
Avoid using ebx as an asm! operand ( #1121 )
...
It is sometimes reserved by LLVM.
2021-04-16 13:20:00 +01:00
Sparrow Li
23f45cc955
Add vqrsh* neon instructions ( #1119 )
2021-04-15 12:29:04 +01:00
liushuyu
33afae1df7
aarch64: add uzp1, uzp2 instructions ( #1118 )
2021-04-15 12:21:31 +01:00
scottmcm
4e5d5c2e56
Typo fix: fist -> first ( #1117 )
2021-04-14 21:13:14 +01:00
surechen
aaaa9335eb
add neon instruction vfma ( #1116 )
2021-04-14 15:34:53 +01:00
Amanieu d'Antras
6405058a6f
Don't import from core ( #1115 )
...
This breaks when core_arch is included as a module in core.
2021-04-13 04:31:37 +01:00
Sparrow Li
88a5de08cb
Allow primitive types in the code generator and add vdup instructions ( #1114 )
2021-04-12 14:08:26 +01:00
Joshua Nelson
b411a5c375
Convert all crates to 2018 edition ( #1109 )
2021-04-11 15:26:35 +01:00
Amanieu d'Antras
1bce95c732
Remove remaining uses of #[rustc_args_required_const] ( #1113 )
2021-04-11 14:22:52 +01:00
Sparrow Li
5ee0274cef
add vshl and vshr neon instructions ( #1111 )
2021-04-09 08:49:23 +01:00
surechen
a3b5ef2b0c
add neon instruction vsubw_* and vsubl_* ( #1112 )
2021-04-08 15:22:35 +01:00
Alex Crichton
c6356546c0
Updates for wasm simd support ( #1110 )
...
* Uncomment some i64-related instruction assertions now that LLVM
supports the opcodes.
* Fix the codegen for `{i,u}32x4_trunc_sat_f32x4`. This was originally
introduced using `simd_cast` but that inherits LLVM's UB related to
float-to-integer casts out of bounds. Since the original inception of
these intrinsics in LLVM dedicated intrinsics for the wasm
instructions have been added, so this swithces the implementation to
using those.
* Uncomment `f64x2_convert_low_i32x4` instruction assertion and add a
test now that this is implemented in Wasmtime.
2021-04-07 17:05:34 +01:00
Joshua Nelson
7bab2c0695
Deny 2018 idiom lints ( #1108 )
...
This lint is allow by default, which is why this wasn't spotted earlier.
It's denied by rust-lang/rust, so it's good to warn about it here so it
can be fixed more quickly.
2021-04-07 05:46:39 +01:00
surechen
e6a81b7566
add neon instruction vmaxnm_f* vpmaxnm_f* vminnm_f* vpminnm_f* ( #1105 )
2021-04-06 06:57:05 +01:00
Sparrow Li
6201670dd2
Enable constant parameters in the code generator and add vext instructions ( #1106 )
2021-04-02 22:02:22 +01:00
Sparrow Li
7b21d85a41
add vmovn_high, vrbit, vrnd, vsubhn neon instructions ( #1103 )
2021-03-31 15:48:58 +01:00
surechen
ef9ec33482
add support for neon instruction vqabs_* ( #1102 )
2021-03-30 12:22:38 +01:00
Sparrow Li
64f84788c9
add vreinterpret neon instructions ( #1101 )
2021-03-30 06:51:01 +01:00
surechen
f9e5dfdd66
support neon instruction vabdl_* and vabdl_high_* ( #1100 )
2021-03-29 11:17:36 +01:00
Sparrow Li
e83d05a3c0
add simd_neg platform intrinsic and vneg, vqneg neon instructions ( #1099 )
2021-03-29 03:43:36 +01:00
surechen
182d7593f4
support for neon instructions vabal_* and vabal_high_* ( #1097 )
2021-03-27 16:28:36 +00:00
Amanieu d'Antras
3769d5df36
Remove unused constify macros ( #1098 )
2021-03-26 14:34:32 +00:00
Alex Crichton
3bbb381ac1
Tweak names of wasm SIMD intrinsics ( #1096 )
2021-03-25 15:11:30 +00:00
SparrowLii
770ced9a69
add vzip1, vzip2 instructions
2021-03-23 05:50:03 +00:00
SparrowLii
91bed9bd0c
correct instruction names
2021-03-23 05:50:03 +00:00
