Konstantinos Margaritis
9a53b193e1
add missing findLSB for ppc64le
2023-12-21 23:25:37 +00:00
Konstantinos Margaritis
de66c745ad
fix debug prints for z on ppc64le
2023-12-21 23:25:37 +00:00
Konstantinos Margaritis
f866b72de9
fix debug formats for z on arm
2023-12-21 23:25:20 +00:00
Konstantinos Margaritis
db3b0e9474
comparemask_type is u64a on Arm, use single load_mask
2023-12-21 23:25:20 +00:00
Konstantinos Margaritis
5814d3298f
remove unneeded shifts
2023-12-21 23:25:20 +00:00
Konstantinos Margaritis
0e2f6c1540
refactor Noodle Single/Double to use masked loads
2023-12-21 23:24:45 +00:00
Konstantinos Margaritis
5f65b9f271
fix types of z in debug prints
2023-12-21 23:24:35 +00:00
Konstantinos Margaritis
476cefb8e7
fix loadu_maskz, remove old defines
2023-12-21 23:24:31 +00:00
Konstantinos Margaritis
9f66822599
define HAVE_MASKED_LOADS for AVX512
2023-12-21 23:24:22 +00:00
Konstantinos Margaritis
d4fde85897
refactor Noodle to use the same loop as Shufti/Truffle, now it's at least 2x as fast
2023-12-21 23:24:15 +00:00
Konstantinos Margaritis
17fb9f41f6
Merge pull request #215 from VectorCamp/feature/use-ccache
...
use ccache if available
2023-12-22 01:18:22 +02:00
Konstantinos Margaritis
ad70693999
use ccache if available
2023-12-21 12:59:56 +00:00
Konstantinos Margaritis
3113d1ca30
Merge pull request #212 from VectorCamp/bugfix/fix-simde-build
...
SIMDe on Clang needs SIMDE_NO_CHECK_IMMEDIATE_CONSTANT defined and other SIMDe related fixes now that SIMDe is part of the CI pipeline.
Some issue with SIMDe on x86 still remains because of an upstream bug:
https://github.com/simd-everywhere/simde/issues/1119
Similarly SIMDe native with clang on Arm also poses a non-high priority build failure:
https://buildbot-ci.vectorcamp.gr/#/builders/129/builds/11
Possibly a SIMDe issue as well, need to investigate but will merge this PR as these are non-blockers.
2023-12-21 11:04:32 +02:00
Konstantinos Margaritis
10d957477a
fix typo in baseline x86 arch definition
2023-12-20 22:21:00 +02:00
Konstantinos Margaritis
ef37e6015a
native CPU on SIMDe will enable all sorts of features in an unpredicted manner, set sane defaults
2023-12-20 16:43:38 +00:00
Konstantinos Margaritis
306e8612be
GREATER_EQUAL
2023-12-20 15:27:56 +00:00
Konstantinos Margaritis
a7a12844e7
reorganize OS detection
2023-12-20 17:16:45 +02:00
Konstantinos Margaritis
44f19c1006
fix submodule headers detection
2023-12-20 17:16:23 +02:00
Konstantinos Margaritis
2aa5e1c710
fix arch=native on arm+clang
2023-12-20 15:15:38 +00:00
Konstantinos Margaritis
1b915cfb93
add fallback pdep64 for x86 if no HAVE_BMI2
2023-12-20 08:25:30 +02:00
Konstantinos Margaritis
49e6fe15a2
add missing pdep64 for x86 bitutils
2023-12-20 00:12:15 +02:00
Konstantinos Margaritis
8cba258e7f
add missing pdep64 for arm and ppc64le
2023-12-19 23:15:27 +02:00
Konstantinos Margaritis
c8ba7fa1d3
add missing pdep64 for common bitutils
2023-12-19 23:09:03 +02:00
Konstantinos Margaritis
e15ad9308a
SIMDe on Clang needs SIMDE_NO_CHECK_IMMEDIATE_CONSTANT defined
2023-12-19 17:31:43 +02:00
Konstantinos Margaritis
a26bed96bc
Merge pull request #203 from VectorCamp/feature/enable-simde-backend
...
Feature/enable simde backend
2023-11-29 11:22:08 +02:00
Konstantinos Margaritis
519bd64c65
fix failing allbits test for ppc64le on clang15
2023-11-29 01:39:05 +02:00
Konstantinos Margaritis
d3f6d2ad06
updates to the Readme
2023-11-28 18:27:08 +02:00
Konstantinos Margaritis
9fd0ce5d44
search for SIMDE sse4.2.h header
2023-11-28 17:39:55 +02:00
Konstantinos Margaritis
6332cb91f5
separate ARCH_FLAG logic
2023-11-28 17:28:48 +02:00
Konstantinos Margaritis
3beda7e5e0
add missing else
2023-11-28 14:09:26 +02:00
Konstantinos Margaritis
be9ce68767
make diffrich384 available on all arches
2023-11-28 12:06:46 +00:00
Konstantnos Margaritis
f5e508b13f
fix compilation for SIMDe
2023-11-27 20:52:52 +00:00
Konstantnos Margaritis
23aeaecf53
use pkg-config for SIMDe
2023-11-27 20:51:47 +00:00
Konstantnos Margaritis
8c7b503ac4
fix TUNE_FLAG for SIMDE_BACKEND
2023-11-27 20:51:29 +00:00
Konstantinos Margaritis
f57928ea08
fix SIMDe emulation builds on Arm, add native translation from x86 for comparison
2023-11-27 12:21:58 +00:00
Konstantnos Margaritis
dfacf75855
existing scalar implementations were incorrect -but never tested, ported from arm/ppc64le
2023-11-23 16:09:10 +00:00
Konstantnos Margaritis
20f4f542a5
add missing intrinsics for SIMDe backend
2023-11-23 16:08:26 +00:00
Konstantnos Margaritis
62cb8d6c2d
fix test for SIMDe
2023-11-23 16:07:58 +00:00
Konstantnos Margaritis
b32ca719d9
SIMDE is a valid platform
2023-11-23 13:07:28 +00:00
Konstantnos Margaritis
7c53b4e608
add include dirs
2023-11-21 17:14:21 +00:00
Konstantnos Margaritis
14c9222a48
add generic tune flags
2023-11-21 17:13:54 +00:00
Konstantnos Margaritis
a8e9b9069e
enable SIMDe backend
2023-11-21 17:13:33 +00:00
Konstantnos Margaritis
b068087240
add SIMDe ports of simd_utils and supervector
2023-11-21 17:12:04 +00:00
Konstantnos Margaritis
b5cde5ebf7
mofidied .gitmodules
2023-11-21 17:11:09 +00:00
Konstantnos Margaritis
8455cba03d
add SIMDe cmake file
2023-11-21 17:09:48 +00:00
Konstantnos Margaritis
129015afc6
add SIMDe git submodule
2023-11-21 17:09:24 +00:00
Konstantnos Margaritis
d24d67c28b
Add SIMDe backend to CMake
2023-11-21 17:06:22 +00:00
Konstantinos Margaritis
44b893abfc
Merge pull request #200 from VectorCamp/bugfix/install-static-libs
...
fix missing installation of static libs
2023-11-21 11:46:29 +02:00
Konstantinos Margaritis
c3a6bb3cb3
Merge pull request #199 from gliwka/fix-missing-hs-version-header
...
Fix missing hs_version.h header (closes #198 )
2023-11-21 11:46:00 +02:00
Konstantinos Margaritis
d611fcbaa8
fix missing installation of static libs
2023-11-20 22:39:12 +02:00