Commit Graph

162 Commits

Author SHA1 Message Date
Anuj Mittal
cd17442923 onednn: turn on PACKAGECONFIG for GPU engine
Build OCL GPU engine by default now that compute runtime works with
latest LLVM.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-08-12 12:37:57 +08:00
Anuj Mittal
ecca05175c intel-compute-runtime: upgrade 22.23.23405 -> 22.31.23852
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-08-12 10:19:24 +08:00
Anuj Mittal
cfef5f829e intel-graphics-compiler: upgrade 1.0.11378 -> 1.0.11702.1
Update vc-intrinsics to v0.5.0 tag.

This release includes support for LLVM14.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-08-10 14:58:47 +08:00
Yogesh Tyagi
2914c71cc8 ispc : upgrade 1.17.0 -> 1.18.0
* Don't build with clang specifically and let user decide.
* Drop already merged patch: 0001-Enable-LLVM-15.0-support.patch
* ispc generated headers include a comment will full path to the header. Patch the code to drop this comment.
* Tweak bison and flex invocation to make sure they don't include #line directives.
* Add flex-native to DEPENDS.
* Release notes:
  https://github.com/ispc/ispc/releases/tag/v1.18.0

Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-07-27 19:02:59 +08:00
Yogesh Tyagi
97ac7b761e onednn : Upgrade 2.6 -> 2.6.1
Release Notes:
https://github.com/oneapi-src/oneDNN/releases/tag/v2.6.1

Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-07-21 07:56:45 +08:00
Anuj Mittal
5ffac66996 intel-compute-runtime: upgrade 22.22.23355 -> 22.23.23405
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-06-17 16:30:06 +08:00
Anuj Mittal
2f913cdb36 intel-graphics-compiler: upgrade 1.0.11279 -> 1.0.11378
Remove the backported patch.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-06-17 16:30:06 +08:00
Anuj Mittal
66f4ff0083 intel-compute-runtime: upgrade 22.11.22682 -> 22.22.23355
Remove upstreamed patch and revert to default for built-ins compilation.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-06-10 16:26:06 +08:00
Anuj Mittal
7cd06a4a71 intel-graphics-compiler: upgrade 1.0.10395 -> 1.0.11279
* Remove upstreamed patches and refresh others.
* Include the LICENSE.md file and remove the cpp file which just points to
MIT license.
* Update the license to MIT and Apache-2.0
* Install vcb tool from -native recipe.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-06-10 16:25:57 +08:00
Anuj Mittal
ebb8c1c26e intel-compute-runtime: fix failures with gcc12
Backport patch to fix build failures like:

| recipe-sysroot/usr/include/c++/12.1.0/bits/unique_ptr.h:95:9: error: 'void operator delete(void*, std::size_t)' called on pointer returned from a mismatched allocation function [-Werror=mismatched-new-delete]
|    95 |         delete __ptr;
|       |         ^~~~~~~~~~~~
| git/shared/source/os_interface/windows/wddm/adapter_info.cpp: In function 'std::wstring NEO::queryAdapterDriverStorePath(const Gdi&, D3DKMT_HANDLE)':
| git/shared/source/os_interface/windows/wddm/adapter_info.cpp:31:117: note: returned from 'void* operator new [](std::size_t)'
|    31 |     std::unique_ptr<uint64_t> storage{new uint64_t[(privateDataSizeNeeded + sizeof(uint64_t) - 1) / sizeof(uint64_t)]};
|       |                                                                                                                     ^
| In member function 'void std::default_delete<_Tp>::operator()(_Tp*) const [with _Tp = long unsigned int]',
|     inlined from 'std::unique_ptr<_Tp, _Dp>::~unique_ptr() [with _Tp = long unsigned int; _Dp = std::default_delete<long unsigned int>]' at recipe-sysroot/usr/include/c++/12.1.0/bits/unique_ptr.h:396:17,
|     inlined from 'std::wstring NEO::queryAdapterDriverStorePath(const Gdi&, D3DKMT_HANDLE)' at git/shared/source/os_interface/windows/wddm/adapter_info.cpp:46:1:
| recipe-sysroot/usr/include/c++/12.1.0/bits/unique_ptr.h:95:9: error: 'void operator delete(void*, std::size_t)' called on pointer returned from a mismatched allocation function [-Werror=mismatched-new-delete]
|    95 |         delete __ptr;
|       |         ^~~~~~~~~~~~
| git/shared/source/os_interface/windows/wddm/adapter_info.cpp: In function 'std::wstring NEO::queryAdapterDriverStorePath(const Gdi&, D3DKMT_HANDLE)':
| git/shared/source/os_interface/windows/wddm/adapter_info.cpp:31:117: note: returned from 'void* operator new [](std::size_t)'
|    31 |     std::unique_ptr<uint64_t> storage{new uint64_t[(privateDataSizeNeeded + sizeof(uint64_t) - 1) / sizeof(uint64_t)]};
|       |                                                                                                                     ^

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-05-20 00:25:29 +08:00
Anuj Mittal
a678464d4d intel-graphics-compiler: define SRCREV_FORMAT
Since this fetches multiple repositories, define SRCREV_FORMAT so SRCPV
is expanded properly.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-05-10 15:29:14 +08:00
Naveen Saini
40814b4385 onednn: upgrade 2.5.3 -> 2.6
Signed-off-by: Naveen Saini <naveen.kumar.saini@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-05-06 18:57:54 +08:00
Naveen Saini
c49915bdbe ispc: disable build for Windows, android and other targets
Signed-off-by: Naveen Saini <naveen.kumar.saini@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-04-22 22:33:57 +08:00
Anuj Mittal
b9da851fa5 llvm-project-source: refresh patches
Refresh the patches as recommended by the opencl-clang and include
support for OpenCL 3.0.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-04-19 20:56:14 +08:00
Anuj Mittal
7980023e6d vc-intrinsics: update to latest
Use the revision just beyond newly created 0.1.0 tag and bump PE.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-04-13 19:09:02 +08:00
Anuj Mittal
68e00896f2 opencl-clang: upgrade 13.0.0 -> 14.0.0
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-04-05 15:59:12 +08:00
Anuj Mittal
c993e8e815 Remove support for building with LLVM 10
We no longer support building with older branches of OE-Core/meta-clang
so remove LLVM 10 specific configurations and patches.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-04-05 15:59:12 +08:00
Dongwon Kim
c2c0f2ade5 compute-runtime: version update from 22.08.22549 to 22.11.22682
Signed-off-by: Dongwon Kim <dongwon.kim@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-03-24 18:34:50 +08:00
Anuj Mittal
311b432394 intel-graphics-compiler: fix compile issues
Backport patches to fix compile problems like:

| /build/build/tmp/work/corei7-64-poky-linux/intel-graphics-compiler/1.0.10395-r0/git/visa/LocalScheduler/SWSB_G4IR.cpp:1459:49: error: expected primary-expression before 'int'
|  1459 |         int maxTokenDelay = std::numeric_limits<int>::min(); //The delay may cause if reuse
|       |                                                 ^~~
| /build/build/tmp/work/corei7-64-poky-linux/intel-graphics-compiler/1.0.10395-r0/git/visa/LocalScheduler/SWSB_G4IR.cpp:1460:37: error: 'numeric_limits' is not a member of 'std'
|  1460 |         int minTokenDistance = std::numeric_limits<int>::max(); //The distance from the reused node

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-03-22 18:13:47 +08:00
Naveen Saini
cd94a7af54 onednn: upgrade 2.5.2 -> 2.5.3
Signed-off-by: Naveen Saini <naveen.kumar.saini@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-03-07 16:14:46 +08:00
Anuj Mittal
2bde8e7514 intel-compute-runtime: fix patch fuzz
Fix warnings:

| Applying patch allow-to-find-cpp-generation-tool.patch
| patching file shared/source/built_ins/kernels/CMakeLists.txt
| Hunk #1 succeeded at 94 with fuzz 2 (offset -6 lines).
| Hunk #2 succeeded at 147 with fuzz 2 (offset 3 lines).
|
| The context lines in the patches can be updated with devtool:
|
|     devtool modify intel-compute-runtime
|     devtool finish --force-patch-refresh intel-compute-runtime <layer_path>
|
| Don't forget to review changes done by devtool!

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-03-07 14:16:04 +08:00
Dongwon Kim
6fba58adb3 compute-runtime: version update from 21.40.21182 to 22.08.22549
Signed-off-by: Dongwon Kim <dongwon.kim@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-03-04 15:06:34 +08:00
Dongwon Kim
4a0450a4fa intel-graphics-compiler: version update from 1.0.8744 to 1.0.10395
Signed-off-by: Dongwon Kim <dongwon.kim@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-03-04 15:03:21 +08:00
Ezhilarasan
f25cc6b707 ipsc: upgrade 1.16.1 -> 1.17.0
Release notes:
https://github.com/ispc/ispc/releases/tag/v1.17.0

Signed-off-by: Ezhilarasan <ezhilarasanx.s@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-03-02 17:39:01 +08:00
ezhilarasan s
2c6d52453c onednn: upgrade 2.4.4 -> 2.5.2
*Fixed performance regression in binary primitive with broadcast (b972174, ff75122).
 *Fixed issue with SYCL device properties initialization (cabc5ca, 095f13e).
 *Fixed issue in matmul primitive with zero points (3157354).
 *Fixed segmentation fault in depthwise convolution primitive for shapes with huge spatial size for processors with Intel AVX-512 support (6834764, 1d2addc).
 *Fixed issue in forward convolution primitive for processors with Intel AVX2 support (d691137).
 *Fixed performance regression on GPUs with SYCL runtime (d8364e5).

Signed-off-by: ezhilarasan s <ezhilarasanx.s@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2022-02-12 00:37:48 +08:00
Anuj Mittal
5ebbe2ae21 onednn: upgrade 2.4 -> 2.4.4
Release notes:
https://github.com/oneapi-src/oneDNN/releases/tag/v2.4.4

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-11-24 22:34:33 +08:00
Anuj Mittal
e4d0c52cba meta: add explicit branch and protocol to SRC_URI
Add branch name explicitly to SRC_URI where it's not defined and switch
to using https protocol for Github projects.

The change was made using convert_srcuri script for OE-Core.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-11-03 12:35:28 +08:00
Naveen Saini
d8584ea9d7 intel-graphics-compiler: add PACKAGECONFIG for VectorCompiler
Allow users to disable it but enable by default.

Signed-off-by: Naveen Saini <naveen.kumar.saini@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-10-25 22:02:18 +08:00
Anuj Mittal
3b386a9dc8 llvm-project-source/10.0.1: backport fixes from LLVM 11
Include a fix and two supporting patches from LLVM 11 to fix problems
with double registration while both clang-cpp and llvm lib are linked
(like in case of opencl-clang).

| CommandLine Error: Option 'mc-relax-all' registered more than once!
| LLVM ERROR: inconsistency in registered CommandLine options

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-10-21 23:42:39 +08:00
Anuj Mittal
a809b8c531 Remove support for gatesgarth
Building with oe-core gatesgarth is no longer supported. Remove from
LAYERSERIES_COMPAT and remove the LLVM 11 patches as well.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-10-13 11:23:41 +08:00
Anuj Mittal
467d15d57a intel-graphics-compiler: enable VectorCompiler
Use the option to use Khronos translator instead of the
pre-built SPIRVDLL with VC. Also have the native recipe
install a binary needed for the target builds.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-10-13 11:23:34 +08:00
Anuj Mittal
3cb5e468fb intel-compute-runtime: upgrade 21.37.20939 -> 21.40.21182
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-10-12 09:22:04 +08:00
Anuj Mittal
afb4218a3e onednn: upgrade 2.3.2 -> 2.4
Release notes:
https://github.com/oneapi-src/oneDNN/releases/tag/v2.4

Remove reference to Mathjax license as it's no longer used.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-10-12 00:40:55 +08:00
Anuj Mittal
cd0ba51efd intel-compute-runtime: upgrade 21.33.20678 -> 21.37.20939
Disable built-ins compilation when building for native.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-10-07 22:51:41 +08:00
Anuj Mittal
2d445772a7 intel-graphics-compiler: upgrade 1.0.8365 -> 1.0.8744
Build vc-intrintics as part of this project as igc expects it to. Remove
the LLVM fixes as they are all available upstream now. Also depend on
-native for invoking elf_packager.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-10-07 22:51:26 +08:00
Anuj Mittal
e332560497 Revert "intel-graphics-compiler: upgrade 1.0.8365 -> 1.0.8708"
This reverts commit 8a1fb876b0.

The upgrade doesn't build properly and needs re-work.
2021-10-04 11:22:47 +08:00
Anuj Mittal
8a1fb876b0 intel-graphics-compiler: upgrade 1.0.8365 -> 1.0.8708
Remove patches that are not required now.

IGC_BUILD__USE_KHRONOS_SPIRV_TRANSLATOR has been renamed so it wasn't
taking effect. Remove the option for now.

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-09-30 15:52:23 +08:00
Naveen Saini
de377897d6 vc-intrinsics: update to include latest commits
It includes:
a2f2f10 Fix Cmake Error
6221091 Fix llvm verifier assertion
2bbb573 Add missing header
fd9bf4b Wrap AttributeList::hasFnAttr
90113a5 add intrinsic to represent visa madw instr
4e3870d Revert "Introduce new vc intrinsic to access timestamp register"
0bf761e Mark lit tests as XFAIL for llvm 14 (tot)
05d3f3d Add global variables support to GenXSingleElementVector pass
43b1af8 Fix typo in GenXIntrinsics/CMakeLists.txt
185f382 update copyright headers update Python copyright to PEP8 style
63cbfe0 update MIT copyright headers
f45e04c Introduce new vc intrinsic to access timestamp register
d9ffe1f (origin/dpcpp_staging) Add intrinsics for bindless buffers support
6d29de3 update comment format of rst copyright headers
b976b83 introduce bf_cvt intrinsic

Udated LIC_FILES_CHKSUM, due to reformating and restructuing of
copyright notice header.

Signed-off-by: Naveen Saini <naveen.kumar.saini@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-09-03 23:43:52 +08:00
Naveen Saini
8be678fdef opencl-clang/13.0.0: add recipe
Build opencl-clang with llvm-13 and point to latest from ocl-open-130
branch.

Signed-off-by: Naveen Saini <naveen.kumar.saini@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-09-03 09:25:01 +08:00
Naveen Saini
7cd64cc018 opencl-clang: restructure
Restructured recipes, having common code in .inc

Signed-off-by: Naveen Saini <naveen.kumar.saini@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-09-03 09:25:01 +08:00
Naveen Saini
a6dadb2499 opencl-clang/12.0.0: build with llvm-12 only
Recipe for opencl-clang/13.0.0 is being added to build against
LLVM-13 and higer.

Signed-off-by: Naveen Saini <naveen.kumar.saini@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-09-03 09:25:01 +08:00
Zoltán Böszörményi
ddb3127f24 intel-graphics-compiler: LLVM 13 fixes
Also add CXXFLAGS=-Wno-error=deprecated-declarations, to turn
some deprecated LLVM API usage from warnings to errors.

Signed-off-by: Zoltán Böszörményi <zboszor@gmail.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-08-27 23:28:04 +08:00
Naveen Saini
23d702e38f llvm/10.0.0: apply ispc recommended patches
ISPC recommends building LLVM 10 with some additional patches to work
around some bugs in this version. Add those patches to our build as well.

https://github.com/ispc/ispc/tree/v1.16.1/llvm_patches

Signed-off-by: Naveen Saini <naveen.kumar.saini@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-08-27 23:28:04 +08:00
Naveen Saini
e69b38c4b7 llvm/11.0.0: apply ispc recommended patches
ISPC recommends building LLVM 11 with some additional patches to work
around some bugs in this version. Add those patches to our build as well.

https://github.com/ispc/ispc/tree/v1.16.1/llvm_patches

Signed-off-by: Naveen Saini <naveen.kumar.saini@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-08-27 23:28:04 +08:00
Naveen Saini
61efde5c41 llvm/12.0.0: apply ispc recommended patches
ISPC recommends building LLVM 12 with some additional patches to work
around some bugs in this version. Add those patches to our build as well.

https://github.com/ispc/ispc/tree/v1.16.1/llvm_patches

Signed-off-by: Naveen Saini <naveen.kumar.saini@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-08-27 23:28:04 +08:00
Dongwon Kim
a41f5a188a compute-runtime: move up to the commit for ADL-P support
new compute-runtime code base including ADL-P support and
one more ADL-S sku with dev id of 0x4690

Signed-off-by: Dongwon Kim <dongwon.kim@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-08-26 12:10:09 +08:00
Anuj Mittal
f4522647f2 onednn: upgrade 2.2.3 -> 2.3.2
Release notes:
https://github.com/oneapi-src/oneDNN/releases

License-Update: Copyright years updated

Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-08-25 01:12:01 +08:00
Dongwon Kim
5b1776caeb intel-compute-runtime: upgrade 21.04.18912 -> 21.33.20678
Source move up to the latest that contains ADL-P support.

Also, this includes an extra patch to revert following commit
to avoid unit-test failure during compute-runtime build:

ab52c7023 Change builtin kernels compilation process

License-Update: LICENSE file has been removed and replaced by
LICENSE.md. License is still the same.

Signed-off-by: Dongwon Kim <dongwon.kim@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-08-24 16:30:22 +08:00
Dongwon Kim
6f8e742aab intel-graphics-compiler: update IGC to 1.0.8365
Updating IGC to add ADL-P support.
All patches under files folder are rebased as well.

Signed-off-by: Dongwon Kim <dongwon.kim@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-08-24 16:05:35 +08:00
Naveen Saini
6a628fa0b5 llvm/12.0.0: apply opencl-clang recommend patches
https://github.com/intel/opencl-clang/tree/ocl-open-120/patches

Signed-off-by: Naveen Saini <naveen.kumar.saini@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
2021-08-24 10:41:24 +08:00