Dmitry Babokin
30270584aa
Merge pull request #749 from egaburov/nvptx_clean
...
Experimental support for PTX with examples
2014-10-16 15:56:02 +04:00
evghenii
92377426bd
changed progress bar implementation
2014-10-16 11:53:05 +02:00
Anton Mitrokhin
3d71932ca6
knc.h: added [] overloads to type defenitions
2014-10-16 12:57:55 +04:00
Dmitry Babokin
ccf4d00385
Merge pull request #874 from dbabokin/alloy-gnu-toolchain
...
alloy.py script upgrade
2014-10-15 23:09:50 +04:00
Dmitry Babokin
f7390aaec9
Adding default sysroot detection for MacOS 10.9 (and newer), which
...
allows working out of the box. Otherwise C standard headers are not
found.
2014-10-15 23:02:59 +04:00
Dmitry Babokin
d79529cada
Adding --with-gcc-toolchain alloy switch, which is passed to configure.
...
This is the only way to build functional clang on old Linux system,
which has old system gcc and new gcc (4.7+) installed in non-system
location.
Also adding switch to build only x86 target in clang.
2014-10-15 20:40:37 +04:00
Dmitry Babokin
8f9a935132
Merge pull request #873 from dbabokin/knc
...
KNC prefetch:
2014-10-14 20:59:48 +04:00
Dmitry Babokin
3f24c8dedc
KNC prefetch:
...
- make L3 prefetch to hit L2$, instead of being nop.
- fix vector prefetch to use single intrinsic, instead of two (this is
caused by bug in Composer 14.0 documentation).
2014-10-14 20:24:16 +04:00
Dmitry Babokin
741df13eb8
Merge pull request #871 from ncos/knc-backend-merge
...
Replaced vec16_i1 type definition in knc.h from typedef to struct version, adding a constructor
2014-10-14 19:24:53 +04:00
Dmitry Babokin
8e31d7e99c
Merge pull request #872 from ncos/alloy_modifications
...
Changed the 'newest_LLVM' variable to 3.5 and fixed an error in help message
2014-10-14 19:24:38 +04:00
Anton Mitrokhin
7ac8a9ea04
changed vec16_i1 type definition in knc.h from typedef to struct
2014-10-14 18:52:03 +04:00
Anton Mitrokhin
dcf70d887b
Changed the 'newest_LLVM' variable to 3.5 and fixed an error in help message
2014-10-14 18:23:47 +04:00
evghenii
558d7ee1d3
fix compilation with gcc481
2014-10-14 15:49:50 +02:00
Dmitry Babokin
2e49ecd56f
Merge pull request #870 from aguskov/master
...
Added a patch for LLVM 3.5 to enable proper debugging under Windows
2014-10-14 17:31:55 +04:00
evghenii
47b1f2182f
fix documentation typo
2014-10-14 15:24:30 +02:00
aguskov
efee86fa2e
Update 3_5_win_coff_debug_info.patch
2014-10-14 17:17:49 +04:00
evghenii
4e7ae5269b
added pseudo_prefetch definitions
2014-10-14 14:48:02 +02:00
evghenii
9238c72e08
Merge branch 'master' into nvptx_clean_master
2014-10-14 14:27:00 +02:00
evghenii
83a863ea83
fix compilation for llvm 3.5
2014-10-14 14:17:36 +02:00
Andrey Guskov
75759254c3
Added a patch for LLVM 3.5 to enable proper debugging under Windows
2014-10-14 15:58:23 +04:00
Dmitry Babokin
44ed10b6f2
Merge pull request #869 from dbabokin/typo
...
Typo fix
2014-10-14 14:54:05 +04:00
Dmitry Babokin
82ed2ce416
Typo fix
2014-10-14 14:51:12 +04:00
Dmitry Babokin
29c0f75306
Merge pull request #868 from dbabokin/win_build
...
Adding missing dependency for LLVM 3.6 (Windows specific problem) and fi...
2014-10-13 19:36:37 +04:00
Dmitry Babokin
481224bbcd
Adding missing dependency for LLVM 3.6 (Windows specific problem) and fixing end-of-lines in ispc.vcxproj
2014-10-13 19:31:11 +04:00
Dmitry Babokin
743f89f93f
Merge pull request #867 from jbrodman/master
...
Fix warnings in knc.h with appropriate casting
2014-10-09 11:52:10 +04:00
James Brodman
3aa2cce504
Fix warnings in knc.h with appropriate casting
2014-10-08 17:21:36 -07:00
Dmitry Babokin
a83fec3dd0
Merge pull request #865 from ncos/ispc_build_fails
...
Build fail fix for ISPC with LLVM 3.6 (current trunk)
2014-10-02 18:29:20 +04:00
Dmitry Babokin
b8a9139f8e
Merge pull request #864 from Vsevolod-Livinskij/opt_prefetch
...
Optimized prefetch
2014-10-02 17:16:51 +04:00
Vsevolod Livinskiy
eb61d5df72
Support for cache 2/3 and all targets
2014-10-02 16:25:23 +04:00
Anton Mitrokhin
57fb2a75ec
new interface for 'DIBuilder::insertDeclare' and 'DIBuilder::createGlobalVariable' in LLVM 3.6 (ids: 076fd5dfc1f0600183bbc7db974dc7b39086136d and bc88cfc3512326d6c8f8dbf3c15d880c0e21feb0 correspondingly
2014-10-02 15:35:19 +04:00
Vsevolod Livinskiy
0a6eb61ad0
Extend gather-scatter optimization with prefetch optimization
2014-10-02 15:21:43 +04:00
Dmitry Babokin
df38b862c2
Merge pull request #863 from ncos/knc-fails
...
Changed the default knc header file from 'knc-i1x16.h' to 'knc.h'
2014-10-02 13:57:59 +04:00
Dmitry Babokin
ddc6e33bc0
Merge pull request #862 from ncos/knc-backend-merge
...
Modification of 'knc.h'
2014-10-02 13:57:45 +04:00
Anton Mitrokhin
700fe244e7
removed debug macros
2014-10-02 13:08:47 +04:00
Anton Mitrokhin
c934a68bc4
changed 'const int' to 'const int8_t' in '__vec16_i8' constructor
2014-10-02 13:00:55 +04:00
Anton Mitrokhin
8295df5a1e
fixed header in 'knc.h'
2014-09-25 22:54:29 +04:00
Anton Mitrokhin
3b16cd8c56
fixed 'knc-i1x16.h' to compile with icc v.15 beta
2014-09-25 21:29:49 +04:00
Anton Mitrokhin
4560df284b
changed codestyle to 2 spaces in 'knc.h'
2014-09-25 21:29:49 +04:00
Anton Mitrokhin
832aff5d76
commented out '__vec16_i1' type as there is no '__cast_uitofp' which produces compfails
2014-09-25 21:29:49 +04:00
Anton Mitrokhin
46bd353027
fixed 'INT_MIN' bug in '__gather64_i64'
2014-09-25 21:29:49 +04:00
Anton Mitrokhin
1a2979aa7f
start fixing gather/scatter functions (INT_MIN fix)
2014-09-25 21:29:49 +04:00
Anton Mitrokhin
8b8e313dc6
starded work on gather/scatter instructions
2014-09-25 21:29:48 +04:00
Anton Mitrokhin
0881463d69
changed '__vec16_i1 __equal_i64(const __vec16_i64 &a, const __vec16_i6_i64 &b, __vec16_i1 mask)' function
2014-09-25 21:29:48 +04:00
Anton Mitrokhin
2e92989101
changed '__vec16_i64 __load(const __vec16_i64 *p)'
2014-09-25 21:29:48 +04:00
Anton Mitrokhin
1d69b954bd
added several memory functions for i8
2014-09-25 21:29:48 +04:00
Anton Mitrokhin
90843b3bff
changed a constructor of '__vec16_i8'
2014-09-25 21:29:48 +04:00
Anton Mitrokhin
78a7ef9fc5
added several math functions, new runfails: gather-int8-2/4; ldexp-double (previously compfailed)
2014-09-25 21:29:47 +04:00
Anton Mitrokhin
efa0ea01f3
add 'void __masked_store_i64(void *p, const __vec16_i64 &v, __vec16_i1 mask)' function. now 450 compfails and 12 runfails
2014-09-25 21:29:47 +04:00
Anton Mitrokhin
ddf5df6193
made 'int64_t __extract_element(const __vec16_i64 &v, uint32_t index)' function
2014-09-25 21:29:47 +04:00
Anton Mitrokhin
4fff0ab571
added __vec16_i32 __cast_trunc(__vec16_i32, const __vec16_i64 i64) function. ptr-diff-3/5/6 tests are apparently runfailing because of it
2014-09-25 21:29:47 +04:00
