Dmitry Babokin
|
4c7fb35f57
|
Fixing off by 1 access to local array.
|
2016-08-31 19:38:33 +03:00 |
|
Andrey Shishpanov
|
ab4cd1e8a2
|
Fixed the last useless step in foreach_tiled in some cases
|
2016-04-12 00:29:17 +03:00 |
|
Vsevolod Livinskiy
|
6ea03bb3af
|
Fix for struct with undound members
|
2016-02-06 17:08:17 +03:00 |
|
Vsevolod Livinskiy
|
0ef3d3b429
|
Define dereferencing varying pointer to uniform struct with 'bound uniform' member.
|
2016-02-05 14:49:42 +03:00 |
|
Vsevolod Livinskiy
|
243d6c2625
|
Fix cast from uniform struct to varying struct.
|
2016-02-05 08:32:35 +03:00 |
|
Anton Mitrokhin
|
9bfc3850c7
|
new test file for double sin/cos/sincos
|
2015-08-02 15:46:15 +03:00 |
|
Andrey Shishpanov
|
3bc99c1da1
|
solution for asin trouble from report
|
2015-07-22 14:26:51 +03:00 |
|
Anton Mitrokhin
|
016781027c
|
Add a test file for ispc's isnan() function
|
2015-07-14 09:58:07 +03:00 |
|
Vsevolod Livinskiy
|
12ff721ec1
|
Add roundings for varying double in knc.h and knl.h
|
2015-05-14 16:26:26 +03:00 |
|
Andrey Guskov
|
ef9315200c
|
Added structure alignment in headers; extended the test system to support alignment tests
|
2015-02-17 17:58:34 +03:00 |
|
Dmitry Babokin
|
da83196996
|
Adding missing AddressOfExpr::GetLValueType()
|
2014-12-30 18:11:39 +03:00 |
|
evghenii
|
9238c72e08
|
Merge branch 'master' into nvptx_clean_master
|
2014-10-14 14:27:00 +02:00 |
|
Vsevolod Livinskiy
|
eb61d5df72
|
Support for cache 2/3 and all targets
|
2014-10-02 16:25:23 +04:00 |
|
Evghenii
|
fb581818c5
|
Merge branch 'master' into nvptx
|
2014-04-01 09:10:10 +02:00 |
|
Ilia Filippov
|
ecdc695b22
|
Changing overload rules to match C++ behavior: Emit a warning when the best overload match has some number of no-best matching parameters.
|
2014-03-25 12:41:09 +04:00 |
|
Ilia Filippov
|
6f44d5b55f
|
correction of overload issues
|
2014-03-24 15:47:21 +04:00 |
|
Evghenii
|
4641a15287
|
Merge branch 'master' into nvptx
|
2014-03-19 10:53:07 +01:00 |
|
Ilia Filippov
|
e1524891fc
|
Merge pull request #751 from Vsevolod-Livinskij/master
Saturating multiplication for int64 was added.
|
2014-03-12 00:12:34 -07:00 |
|
Ilia Filippov
|
6738af0a0c
|
changing uniform_min and uniform_max implementations for avx targets
|
2014-03-06 12:05:24 +04:00 |
|
Vsevolod Livinskij
|
af836cda27
|
Saturating multiplication for int64 was added.
|
2014-02-23 19:48:03 +04:00 |
|
Evghenii
|
42c4d3246c
|
Merge branch 'master' into nvptx_clean
|
2014-02-21 12:45:01 +01:00 |
|
Dmitry Babokin
|
e8680760bf
|
Merge pull request #741 from Vsevolod-Livinskij/master
Saturation arithmetic.
|
2014-02-21 12:30:58 +03:00 |
|
Dmitry Babokin
|
17d8047a93
|
Merge pull request #746 from ifilippov/master
Adding cases of 'cast' instructions in optimizations
|
2014-02-21 12:25:58 +03:00 |
|
Ilia Filippov
|
42e00ebb24
|
adding cases of 'cast' instructions in optimizations
|
2014-02-21 13:00:16 +04:00 |
|
Evghenii
|
ac05de6835
|
merged with master
|
2014-02-21 08:25:28 +01:00 |
|
Vsevolod Livinskij
|
7dd7020c5f
|
Dec constants was changed with hex constants.
|
2014-02-20 22:57:24 +04:00 |
|
Dmitry Babokin
|
f280b32fa4
|
Merge pull request #736 from egaburov/native_trigonometry
Native trigonometry
|
2014-02-20 19:18:35 +03:00 |
|
Evghenii
|
690a8acb30
|
merged with master
|
2014-02-20 15:22:09 +01:00 |
|
Evghenii
|
4196c723eb
|
merged with nvptx
|
2014-02-20 11:01:58 +01:00 |
|
Vsevolod Livinskij
|
735e6a8ab3
|
Saturation arithmetic mul and div for int8/int16/int32 and div for int64 was added
|
2014-02-18 02:07:13 +04:00 |
|
Vsevolod Livinskij
|
f5508db24f
|
Saturation arithmetic (sub and add) was added for int32/int64.
|
2014-02-17 18:55:40 +04:00 |
|
evghenii
|
193bba77b0
|
accuracy fix
|
2014-02-11 11:49:03 +01:00 |
|
Evghenii
|
f0779f95a3
|
added double precision tests
|
2014-02-11 11:40:40 +01:00 |
|
Vsevolod Livinskij
|
cef5b2eb04
|
Some changes in saturation arithmetic
|
2014-02-10 12:40:53 +04:00 |
|
Evghenii
|
fe98fe8cdc
|
added fast approximate rcp(double) accurate to 15 digits
|
2014-02-04 15:23:34 +01:00 |
|
Evghenii
|
eb1a495a7a
|
added support for fast approximate rsqrt(double). Provide 16 digit accurancy but is over 3x faster than 1/sqrt(double)
|
2014-02-04 14:44:54 +01:00 |
|
Evghenii
|
4515dd5c89
|
added tests for rcp/rsqrt double
|
2014-02-02 18:19:56 +01:00 |
|
Evghenii
|
b7b5c9ad1d
|
it is illegal to pass varying parapamter to a task function with nvptx target
|
2014-01-27 10:30:09 +01:00 |
|
Evghenii
|
1c2dbd6a27
|
a fix for .b0 ptx and some other code improvements
|
2014-01-27 08:51:05 +01:00 |
|
Evghenii
|
4ecf30530a
|
fixed for operator2 with nvptx target
|
2014-01-26 15:08:25 +01:00 |
|
Evghenii
|
fcbdd93043
|
half/scan for 64 bit/clock/num_cores and other additions
|
2014-01-25 16:43:33 +01:00 |
|
Evghenii
|
9090d8b128
|
added support for assert
|
2014-01-24 12:18:20 +01:00 |
|
Evghenii
|
5a8351d7ea
|
added varying new/delete
|
2014-01-24 09:22:55 +01:00 |
|
Evghenii
|
da7a2c0c7f
|
added emulation of "soa" data types via shared-memory
|
2014-01-23 16:17:06 +01:00 |
|
Evghenii
|
2e7609156a
|
fixes for exclclusive_scan_and/or_i32 and shuffle2 and __movmsk
|
2014-01-23 10:24:44 +01:00 |
|
Evghenii
|
06313e0ec3
|
exclusive_scan_and is supported, but must be called outside if-statements. in pricniple other must do the same
|
2014-01-22 22:12:51 +01:00 |
|
Evghenii
|
5376743281
|
added "const" before "static unfiform" in constant folding tests
|
2014-01-21 14:59:25 +01:00 |
|
Evghenii
|
215abab544
|
bugfix
|
2014-01-21 14:55:41 +01:00 |
|
Evghenii
|
bc99897fbb
|
+fixed some example, found some bugs, and bugs in ptxas/cuda
|
2014-01-21 14:51:27 +01:00 |
|
Evghenii
|
5a773ed62a
|
some cfor tests fixes for > 16 lanes
|
2014-01-20 16:42:33 +01:00 |
|