236 Commits

Author SHA1 Message Date
gtsoul-tech
adda613f51 shiftTooManyBitsSigned 2024-04-24 11:13:28 +03:00
gtsoul-tech
e6c884358e uninitvar 2024-04-24 11:13:02 +03:00
gtsoul-tech
9b9df1b397 invalidPrintfArgType_sint 2024-04-24 11:07:23 +03:00
Konstantinos Margaritis
8cd365121d Revert "fix more unused-variable warnings"
This reverts commit afb1a1705f8073ba43b38845d3aa1329634083ed.
2024-01-20 17:46:29 +02:00
Konstantinos Margaritis
68dab83799 Revert "fix unused-variable warning"
This reverts commit ac02b589beebad99820a8b42e6b96e598e7da929.
2024-01-20 17:46:29 +02:00
Konstantinos Margaritis
528a165c20 Revert "don't demand 32/64-byte alignment if there is no 256/512-bit SIMD engine"
This reverts commit 719e1c9be6fd6fd316889ac7625253d0ad9c5fd5.
2024-01-19 17:41:40 +02:00
Konstantinos Margaritis
4fb8cee35f -Wno-pass-failed is only for ppc64le 2024-01-19 11:23:17 +02:00
Konstantinos Margaritis
5d38d0d0a5 add needed deprecation warning exceptions for SIMDe on Power VSX 2024-01-18 23:37:59 +02:00
Konstantinos Margaritis
719e1c9be6 don't demand 32/64-byte alignment if there is no 256/512-bit SIMD engine 2024-01-18 18:37:27 +02:00
Konstantinos Margaritis
afb1a1705f fix more unused-variable warnings 2024-01-17 17:03:19 +02:00
Konstantinos Margaritis
4d2bcff7b4 remove unused variable 2024-01-17 17:02:32 +02:00
Konstantinos Margaritis
ac02b589be fix unused-variable warning 2024-01-17 17:02:08 +02:00
Konstantinos Margaritis
1988ff5a6d Do not assume unit-internal is built for unit target 2024-01-11 13:23:37 +02:00
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
62cb8d6c2d fix test for SIMDe 2023-11-23 16:07:58 +00:00
Konstantinos Margaritis
24ae1670d6 WIP: Refactor CMake build system to more modular 2023-10-08 23:27:24 +03:00
Konstantinos Margaritis
a26661c849 remove extra print 2023-10-06 12:08:36 +03:00
Konstantinos Margaritis
22a24f12ea Reduce debug unit tests runtime even more
In single.cpp featuremask with AVX512 features is not relevant to non-x86 platforms,
and just extends the runtime for no reason.
2023-10-05 19:12:58 +03:00
Konstantinos Margaritis
e369681ce2 Don't run regression UE_2595 on debug, it times out CI 2023-10-05 10:40:30 +03:00
Konstantinos Margaritis
35c0711689 use the right type of cast 2023-10-04 23:35:10 +03:00
Konstantinos Margaritis
72afe16452 clang 16 as well 2023-10-04 22:07:34 +03:00
Konstantinos Margaritis
2e88df1a89 use the conditional in the right way 2023-10-04 20:35:58 +03:00
Konstantinos Margaritis
354fda48fb add conditional for __clang__ 2023-10-04 20:28:35 +03:00
Konstantinos Margaritis
b7d1bc0298 clang 15 (but not 16) fails on ppc64le with -Wdeprecate-lax-vec-conv-all 2023-10-04 20:09:45 +03:00
Konstantinos Margaritis
9aa61440ea Reduce unit test runtimes dramatically for debug builds 2023-10-04 19:21:30 +03:00
Hong, Yang A
4fb3a48dfd bugfix: add vbmi case for test in database.cpp 2023-09-05 13:52:10 +03:00
Hong, Yang A
6765b35d48 bugfix: add vbmi platform parameter for tests in single.cpp 2023-09-05 13:52:03 +03:00
Hong, Yang A
b7ee9102ee update year 2022 2023-09-05 13:49:52 +03:00
Hong, Yang A
684f0ce2cb UTF-8 validation: fix one cotec check corner issue
fix github issue #362
2023-09-05 13:49:41 +03:00
Konstantinos Margaritis
94fe406f0c [VSX] correct lshiftbyte_m128/rshiftbyte_m128, variable_byte_shift 2022-09-06 23:59:51 +03:00
Konstantinos Margaritis
74ab41897c Add missing <memory> header 2022-08-30 20:40:23 +03:00
Danila Kutenin
db52ce6f08 Fix avx512 movemask call 2022-07-20 09:03:50 +01:00
Danila Kutenin
49eb18ee4f Optimize vectorscan for aarch64 by using shrn instruction
This optimization is based on the thread
https://twitter.com/Danlark1/status/1539344279268691970 and uses
shift right and narrow by 4 instruction https://developer.arm.com/documentation/ddi0596/2020-12/SIMD-FP-Instructions/SHRN--SHRN2--Shift-Right-Narrow--immediate--

To achieve that, I needed to redesign a little movemask into comparemask
and have an additional step towards mask iteration. Our benchmarks
showed 10-15% improvement on average for long matches.
2022-06-26 22:55:45 +00:00
Danila Kutenin
9af996b936 Fix all ASAN issues in vectorscan 2022-02-18 17:14:51 +00:00
Konstantinos Margaritis
7cad514366 clang is more strict 2021-12-02 23:09:53 +02:00
Konstantinos Margaritis
00384c9e37 nit 2021-11-25 06:21:07 +00:00
Apostolos Tapsas
725a8d8f1a Removed duplicates 2021-11-24 15:09:53 +00:00
Apostolos Tapsas
35e5369c70 *fix palignr implementation for VSX Release mode
*add unit test for palignr
*enable unit test building for Release mode
2021-11-24 15:03:49 +00:00
Apostolos Tapsas
54158a1746 vermicelli and match implementations for ppc64el added 2021-11-13 19:36:46 +00:00
apostolos
e09d8674b4 resolving conficts after merging 2021-11-13 18:58:22 +02:00
apostolos
4114b8a480 SuperVector opandnot test enriched 2021-11-10 15:12:25 +02:00
apostolos
942deb7d80 test for load m128 from u64a function added 2021-11-10 09:01:28 +02:00
Konstantinos Margaritis
210295a702 remove vermicelli.h and replace it with vermicelli.hpp 2021-11-02 22:30:53 +02:00
Konstantinos Margaritis
bc1a1127cf add new include file 2021-11-01 16:28:50 +00:00
Konstantinos Margaritis
7b65b298c1 add arm vector types in union, avoid -flax-conversions, fix castings 2021-11-01 16:52:17 +02:00
apostolos
d9d39d48c5 prints commants and formating fixes 2021-11-01 10:09:15 +02:00
apostolos
3f17750a27 nits 2021-10-26 11:55:02 +03:00
apostolos
bf54aae779 Special case for Shuffle test added as well as comments for respectives implementations 2021-10-26 11:48:33 +03:00
Apostolos Tapsas
1eb3b19f63 Shuffle simd and SuperVector implementetions as well as their test realy fixed 2021-10-25 09:19:30 +03:00
Apostolos Tapsas
d43d6733b6 SuperVector shuffle implementation and test function optimized 2021-10-22 11:55:39 +00:00