aaron/ispc
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
1,554 Commits 6 Branches 36 Tags
2951cad365f7dbaf4bc3f73b0a822ddd4a40357b
Commit Graph

157 Commits

Author SHA1 Message Date
evghenii
2951cad365 added description for multi-dimensional tasking 2013-12-09 13:10:26 +01:00
Dmitry Babokin
660f14245b Merge pull request #611 from dbabokin/typos 
Typo fix and copyright update
 2013-09-30 08:40:38 -07:00
Dmitry Babokin
7942bdb728 Typo fix and copyright update 2013-09-30 18:09:59 +04:00
Dmitry Babokin
8a39af8f72 Release 1.5.0 2013-09-27 23:27:05 +04:00
Dmitry Babokin
f45f6cb32a Test, documentation and vim support for double precision constants 2013-09-19 12:49:45 +04:00
james.brodman
8db378b265 Revert "Remove support for using SVML for math lib routines." 
This reverts commit d9c38b5c1f.
 2013-09-04 16:01:58 -04:00
Dmitry Babokin
501a23ad20 Typos fixes in docs 2013-08-29 14:48:09 +04:00
Dmitry Babokin
443987f536 fixing ispc.rst file properties (should not be executable) 2013-08-27 15:33:44 +04:00
Matt Pharr
0c5742b6f8 Implement new naming scheme for --target. 
Now targets are named like "<isa>-i<mask size>x<gang size>", e.g.
"sse4-i8x16", or "avx2-i32x16".

The old target names are still supported.
 2013-08-08 19:23:44 -07:00
Matt Pharr
5b20b06bd9 Add avg_{up,down}_int{8,16} routines to stdlib 
These compute the average of two given values, rounding up and down,
respectively, if the result isn't exact.  When possible, these are
mapped to target-specific intrinsics (PADD[BW] on IA and VH[R]ADD[US]
on NEON.)

A subsequent commit will add pattern-matching to generate calls to
these intrinsincs when the corresponding patterns are detected in the
IR.)
 2013-08-06 08:41:12 -07:00
Matt Pharr
4f48d3258a Documentation updates for NEON 2013-07-31 20:06:04 -07:00
Matt Pharr
d9c38b5c1f Remove support for using SVML for math lib routines. 
This path was poorly maintained and wasn't actually available on most
targets.
 2013-07-31 06:56:48 -07:00
Matt Pharr
b6df447b55 Add reduce_add() for int8 and int16 types. 
This maps to specialized instructions (e.g. PSADBW) when available.
 2013-07-25 09:46:01 -07:00
Dmitry Babokin
28f0bce9f2 Release 1.4.4 2013-07-19 16:22:10 -07:00
Dmitry Babokin
594485c38c Release 1.4.3 2013-06-25 18:38:21 +04:00
Dmitry Babokin
cf9ceb6bf9 Release 1.4.2, 11 June 2013 2013-06-11 17:18:54 +04:00
Dmitry Babokin
6c392ee4a1 Changes for 1.4.1 release 2013-05-28 19:46:30 +04:00
Dmitry Babokin
481bcc732b Changes for 1.4.0 release 2013-05-27 16:48:41 +04:00
Dmitry Babokin
de7ba7a55b Bringing docs/ispc.rst in sync with ispc.html at web site (some changes were done there directly) 2013-05-21 16:44:46 +04:00
Dmitry Babokin
3cb827ac56 Fix for some typos in User's Guide 2013-03-04 03:04:46 +04:00
Jean-Luc Duprat
24087ff3cc Expose none() in the ISPC standard library. 
On KNC: all(), any() and none() do not generate a redundant movmsk instruction.
 2012-11-27 13:38:28 -08:00
Matt Pharr
360cc8044e Improve RNG documentation. 
Issue #390.
 2012-10-03 08:33:43 -07:00
Matt Pharr
ec2e9b5e79 Fix typo in assert() documentation. 
Issue #388.
 2012-10-03 08:26:38 -07:00
Jean-Luc Duprat
f928bbb53c Updated usage of Initial Many Core Instructions (Intel® IMCI) instructions. 2012-09-04 10:57:25 -07:00
Jean-Luc Duprat
1ab7500dbb Updated user's guide to comply with Intel® Xeon Phi™ brand usage guidelines 2012-09-04 10:53:01 -07:00
Jean-Luc Duprat
52836aae87 Minor documentation clarrification on the impact of ICC -fp-model except option. 2012-08-01 10:24:35 -07:00
Jean-Luc Duprat
63ed90b0fd docs/build.sh runs rst2html rather than rst2html.py 
Explicitly documented that fact that ICC needs the -mmic flag to compile for KNC.
Updated ISPC User Guide with details on ICC compiler options that impact FP performance in generated code.
 2012-07-30 11:47:25 -07:00
Matt Pharr
f6989cce38 Disallow native output with generic targets, C++ output with non-generic targets. 
Also wrote FAQs about why this is the way it is.
Issue #334.
 2012-07-20 09:55:50 -07:00
Matt Pharr
2c640f7e52 Add support for RDRAND in IvyBridge. 
The standard library now provides a variety of rdrand() functions
that call out to RDRAND, when available.

Issue #263.
 2012-07-12 06:07:07 -07:00
Matt Pharr
2bacebb1fb Doc fixes (Crystal Lemire). 2012-07-11 19:51:28 -07:00
Matt Pharr
57f0cf30c0 Fix small typos in documentation. 2012-07-07 11:19:57 -07:00
Matt Pharr
b8503b9255 News and doxygen version number bump for 1.3.0 2012-06-29 08:38:38 -07:00
Matt Pharr
b7bc76d3cc Documentation updates for 1.3.0. 2012-06-29 08:35:29 -07:00
Matt Pharr
2c7b650240 Add FAQ to explain how to launch per-instance tasks with foreach_active and unmasked. 
Issue #227.
 2012-06-22 14:32:05 -07:00
Matt Pharr
54459255d4 Add unmasked { } statement. 
This reestablishes an "all on" execution mask for the gang, which can
be useful for nested parallelism..
 2012-06-22 14:30:58 -07:00
Matt Pharr
b4a078e2f6 Add foreach_active iteration statement. 
Issue #298.
 2012-06-22 10:35:43 -07:00
Matt Pharr
46716aada3 Switch to unordered floating point compares. 
In particular, this gives us desired behavior for NaNs (all compares
involving a NaN evaluate to true).  This in turn allows writing the
canonical isnan() function as "v != v".

Added isnan() to the standard library as well.
 2012-06-20 13:25:53 -07:00
Matt Pharr
3bc66136b2 Add foreach_unique iteration construct. 
Idea via Ingo Wald / IVL compiler.
 2012-06-20 10:04:24 -07:00
Matt Pharr
4945dc3682 Add contributors link to docs HTML templates 2012-06-13 06:11:08 -07:00
Matt Pharr
8fd9b84a80 Update seed_rng() in stdlib to take a varying seed. 
Previously, we were trying to take a uniform seed and then shuffle that
around to initialize the state for each of the program instances.  This
was becoming increasingly untenable and brittle.

Now a varying seed is expected and used.
 2012-05-30 10:35:41 -07:00
Matt Pharr
90db01d038 Represent MOVMSK'ed masks with int64s rather than int32s. 
This allows us to scale up to 64-wide execution.
 2012-05-25 11:57:23 -07:00
Matt Pharr
0c5d7ff8f2 Add rygorous's float->srgb8 conversion routine to the stdlib. 
Issue #230
 2012-04-27 10:03:19 -10:00
Matt Pharr
0e2b315ded Add FAQ about foreach code generation. 
(i.e. "why's there that extra stuff at the end and what can I do
about it if it's not necessary?)

Issue #231.
 2012-04-27 09:35:37 -10:00
Matt Pharr
3e74d1c544 Fix documentation bug with typedef. 2012-04-25 17:15:20 -10:00
Matt Pharr
016b502d46 Update release notes for 1.2.2, bump version number in doxygen 2012-04-20 14:26:00 -07:00
Matt Pharr
491fa239bd Add atomic swap and cmpxchg for void * as well. 
Issue #232.
 2012-04-11 06:12:31 -07:00
Matt Pharr
70a5348f43 Add size_t, ptrdiff_t, and [u]intptr_t types. 2012-04-11 05:32:53 -07:00
Matt Pharr
2aa61007c6 Remove memory_barrier() calls from atomics. 
This was unnecessary overhead to impose on all callers; the user
should handle these as needed on their own.

Also added some explanatory text to the documentation that highlights
that memory_barrier() is only needed across HW threads/cores, not
across program instances in a gang.
 2012-04-10 19:37:03 -07:00
Matt Pharr
d0e583b29c Release notes and doxygen version nubmer bump for 1.2.1 2012-04-06 16:02:19 -07:00
Matt Pharr
c7dc8862a5 Add FAQs about various language details. 
One of these finishes off issue #225.
 2012-04-05 15:24:26 -07:00