Re: [Beignet] Anything from master that should be in (Debian) 1.3?

We also don't full test these two patches, but seems they will not cause regression. For patch 7e181af2, it is an improvement. For patch b70d65ba, it check the drive name, which is hard code in linux kernel i915 drm driver. > -Original Message- > From: Beignet

Re: [Beignet] [PATCH] Allow creating out-of-order queues with clCreateCommandQueue

Thanks for your patches, they looks good to me, have pushed. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Rebecca N. Palmer > Sent: Sunday, July 22, 2018 3:26 AM > To: beignet@lists.freedesktop.org > Subject: [Beignet] [PATCH] Allow

Re: [Beignet] warning, curbe size exceed limitation.

then hardware's restriction, beignet will print this warning and trunk the used curbe size. Thanks, Yang Rong From: Vaughan, Thomas E [mailto:tevau...@ball.com] Sent: Wednesday, January 31, 2018 11:59 PM To: Yang, Rong R <rong.r.y...@intel.com>; beignet@lists.freedesktop.org Subject: RE: w

Re: [Beignet] [PATCH 1/3] Ensure that DRM device uses the i915 driver

The patchset LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Mark Thompson > Sent: Thursday, February 1, 2018 3:58 AM > To: beignet@lists.freedesktop.org > Subject: [Beignet] [PATCH 1/3] Ensure that DRM device uses

Re: [Beignet] warning, curbe size exceed limitation.

Do you use this array as kernel's argument? If the argument's size exceeds the argument buffer's size, beignet will print this warning. And IVB's argument buffer size is larger than BayTrail. Thanks, Yang Rong From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of Vaughan,

Re: [Beignet] [PATCH 2/2] Enable Coffee Lake support

One inline comment. Can you also add other Coffee Lake pciids? 0x3E90, 12Eus, gt1 0x3E93, 12Eus, gt1 0x3E99, 12Eus, gt1 0x3EA1, 12Eus, gt1 0x3EA4, 12Eus, gt1 0x3E91, 24EUs, gt2 0x3E92, 24EUs, gt2 0x3E94, 24EUs, gt2 0x3E96, 24EUs, gt2 0x3E9A, 24EUs, gt2 0x3E9B, 24EUs, gt2 0x3EA0, 24EUs, gt2

Re: [Beignet] [PATCH 1/2] Ensure that DRM device uses the i915 driver

This patch LGTM, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Mark Thompson > Sent: Wednesday, January 24, 2018 6:52 AM > To: beignet@lists.freedesktop.org > Subject: [Beignet] [PATCH 1/2] Ensure that DRM device uses the i915

Re: [Beignet] Is "X server found. dri2 connection failed!" normal on Wayland?

Yes, this warning is legacy, I will apply your patch, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Rebecca N. Palmer > Sent: Tuesday, January 9, 2018 5:12 AM > To: beignet@lists.freedesktop.org > Subject: [Beignet] Is "X server

Re: [Beignet] [PATCH] Docs: OCL_STRICT_CONFORMANCE is default-on since 1.1

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Rebecca N. Palmer > Sent: Saturday, October 28, 2017 3:24 AM > To: beignet@lists.freedesktop.org > Subject: [Beignet] [PATCH] Docs: OCL_STRICT_CONFORMANCE is default-on

Re: [Beignet] [PATCH] Docs: Fix grammar

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Rebecca N. Palmer > Sent: Saturday, October 28, 2017 3:29 AM > To: beignet@lists.freedesktop.org > Subject: [Beignet] [PATCH] Docs: Fix grammar > > Signed-off-by:

Re: [Beignet] [PATCH] GBE: Remove TBAA.

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Song, Ruiling > Sent: Wednesday, October 18, 2017 3:36 PM > To: beignet@lists.freedesktop.org > Subject: Re: [Beignet] [PATCH] GBE: Remove TBAA. > > Please skip this

Re: [Beignet] [PATCH] metainfo: escape ampersand

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Igor Gnatenko > Sent: Wednesday, October 4, 2017 4:13 PM > To: beignet@lists.freedesktop.org > Cc: Igor Gnatenko > Subject: [Beignet]

Re: [Beignet] beignet and LLVM 4/5

It is under the release testing. If there is no issue, I will release a new version 1.3.2 that include llvm 4 and 5 support. > -Original Message- > From: Rebecca N. Palmer [mailto:rebecca_pal...@zoho.com] > Sent: Wednesday, October 4, 2017 3:10 PM > To: Yang, Rong

Re: [Beignet] [Patch V2 2/2] GBE: enable llvm5.0 support.

Pushed, thanks. > -Original Message- > From: Song, Ruiling > Sent: Wednesday, September 20, 2017 4:44 PM > To: Yang, Rong R <rong.r.y...@intel.com>; beignet@lists.freedesktop.org > Cc: Yang, Rong R <rong.r.y...@intel.com> > Subject: RE: [Beignet] [Patch V2 2/

Re: [Beignet] [PATCH] libocl: Add shuffle and shuffle2 builtins for half type

Pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Song, Ruiling > Sent: Wednesday, September 20, 2017 4:44 PM > To: Jan Vesely ; beignet@lists.freedesktop.org > Subject: Re: [Beignet] [PATCH] libocl:

Re: [Beignet] [PATCH] backend: use simd-1 for scalar dst in indirectMov.

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Ruiling Song > Sent: Wednesday, September 13, 2017 2:07 PM > To: beignet@lists.freedesktop.org > Cc: Song, Ruiling > Subject: [Beignet] [PATCH]

Re: [Beignet] [PATCH] libocl: Consider only bottom ilogb(2m-1)+1 bits

Pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Song, Ruiling > Sent: Wednesday, September 20, 2017 4:44 PM > To: Jan Vesely ; beignet@lists.freedesktop.org > Subject: Re: [Beignet] [PATCH] libocl:

Re: [Beignet] beignet and LLVM 4

Hi Rebecca, LLVM 5. 0 has been released, we are planning to release a minor release 1.3.2 to support LLVM 4.0 and LLVM 5.0 after beignet's LLVM5.0 patches are ready. Dose fedora applying 4.0 support is urgent? If not, I suggest you wait 1.3.2 release. Thanks, Yang Rong >

Re: [Beignet] Status of cl_khr_gl_sharing

The application is responsible for synchronizing access to shared objects. So I think only one thing clEnqueueAcquire/ReleaseGLObjects to do is set the parameter event's status correctly. I will implement them, and remove the other NOT_IMPLEMENTED functions. > -Original Message- > From:

Re: [Beignet] Status of cl_khr_gl_sharing

Commit a892148 has re-implement this extension, use the eglExportDMABUFImageMESA and glGetTexLevelParameteriv to get sharing buffer's informatin. Now it require the libEGL.so and libGL.so's version >= 13.0. But this extension is partially implemented, only support clCreateFromGLTexture2D. >

Re: [Beignet] Kaby lake perf worse than Skylake?

> -Original Message- > From: Michael Gratton [mailto:m...@vee.net] > Sent: Thursday, July 20, 2017 12:16 > To: Yang, Rong R <rong.r.y...@intel.com> > Cc: beignet@lists.freedesktop.org > Subject: Re: [Beignet] Kaby lake perf worse than Skylake? > > Hi, >

Re: [Beignet] Kaby lake perf worse than Skylake?

Hi, Is the quantity of infrared data they process same? Is it affected by fps? Has you try other method, such as the profiling time. Actually, in beignet, the Kabylake is almost same as Skylake. > -Original Message- > From: Beignet

Re: [Beignet] [PATCH] Fix GCC6 build bug

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Xiuli Pan > Sent: Tuesday, July 11, 2017 10:02 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli > Subject: [Beignet] [PATCH] Fix GCC6 build

Re: [Beignet] [PATCH] backend: refine global immediate optimization

Forgot to push this patch, pushed, thanks. > -Original Message- > From: Song, Ruiling > Sent: Thursday, July 20, 2017 9:13 > To: Wang, Rander <rander.w...@intel.com>; beig...@freedesktop.org > Cc: Wang, Rander <rander.w...@intel.com>; Yichao Yu > <

Re: [Beignet] [PATCH V5] backend: refine load store optimization

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Ruiling Song > Sent: Thursday, July 13, 2017 10:45 > To: beignet@lists.freedesktop.org > Cc: Song, Ruiling ; Wang, Rander >

Re: [Beignet] [PATCH] backend: refine math log function

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > rander.wang > Sent: Monday, June 19, 2017 13:21 > To: beig...@freedesktop.org > Cc: Wang, Rander > Subject: [Beignet] [PATCH] backend: refine

Re: [Beignet] [PATCH] backend: refine pow function

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > rander.wang > Sent: Thursday, June 22, 2017 17:41 > To: beig...@freedesktop.org > Cc: Wang, Rander > Subject: [Beignet] [PATCH] backend: refine

Re: [Beignet] [PATCH] Runtime: refine max group size for SKL & KBL

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > rander.wang > Sent: Friday, June 23, 2017 11:02 > To: beig...@freedesktop.org > Cc: Wang, Rander > Subject: [Beignet] [PATCH] Runtime: refine

Re: [Beignet] [PATCH V4] backend: add global immediate optimization

GEN is support mixed type instructions, mixed UD and UW. For example, UD * UW. How about handle the U/UD in one if branch? > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Wang, Rander > Sent: Monday, July 3, 2017 9:33 > To:

Re: [Beignet] [PATCH] backend: refine fdiv to rcp at some cases

One comment. Thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > rander.wang > Sent: Monday, June 19, 2017 13:34 > To: beig...@freedesktop.org > Cc: Wang, Rander > Subject: [Beignet] [PATCH] backend: refine

Re: [Beignet] [PATCH] backend: improve add zero pattern

After remove negation check, the function name doNegAddOptimization is not suitable. Can you also change function name and the comment? > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > rander.wang > Sent: Friday, June 23, 2017 15:37 > To:

Re: [Beignet] [PATCH] GBE: clean llvm module's clone and release.

Pushed, thanks. > -Original Message- > From: Pan, Xiuli > Sent: Thursday, June 22, 2017 14:54 > To: Yang, Rong R <rong.r.y...@intel.com>; beignet@lists.freedesktop.org > Cc: Yang, Rong R <rong.r.y...@intel.com> > Subject: RE: [Beignet] [PATCH] GBE: clean l

Re: [Beignet] [PATCH V4] backend: add global immediate optimization

Pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Song, Ruiling > Sent: Thursday, June 22, 2017 14:30 > To: Wang, Rander ; beig...@freedesktop.org > Cc: Wang, Rander > Subject:

Re: [Beignet] [PATCH V2] backend: refine load/store merging algorithm

Pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Song, Ruiling > Sent: Thursday, June 22, 2017 14:29 > To: Wang, Rander ; beig...@freedesktop.org > Cc: Wang, Rander > Subject:

Re: [Beignet] [PATCH v2] Add missed kernel names into built-in kernel list.

Rename "__cl_cpy_region_unalign_same_offset;" to "__cl_copy_region_unalign_same_offset;", and "__cl_copy_image_3d_to_2d;" is duplicated. I have modified them and pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of >

Re: [Beignet] [PATCH] Backend: Fix double free of the cloned_module

Because llvmToGen accept the filename argument, so it need to create and delete module. I think the module should not be deleted in llvmToGen, the caller decide to delete or not. I will send another patch to refine it. > -Original Message- > From: Beignet

Re: [Beignet] [PATCH] Runtime: Add missing SKL deivce ID

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Xiuli Pan > Sent: Wednesday, June 14, 2017 11:41 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli > Subject: [Beignet] [PATCH] Runtime: Add

Re: [Beignet] [PATCH] Add aligned copy kernels into built-in kernel list.

There are still some kernels missed, can you add them together? > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > yan.w...@linux.intel.com > Sent: Wednesday, June 21, 2017 11:26 > To: beignet@lists.freedesktop.org > Cc: Yan Wang

Re: [Beignet] [PATCH] Fix context leak with internal kernels

Rebase to master and add internal_ctx_refs when ctx->image_queue is not NULL. The patch LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Patrick Beaulieu > Sent: Friday, June 16, 2017 7:15 > To:

Re: [Beignet] [PATCH V2] backend: refine the local copy propagation.

Fix a build error "case SEL_OP_RHADD" by manual, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > rander.wang > Sent: Thursday, June 15, 2017 9:48 > To: beig...@freedesktop.org > Cc: Wang, Rander >

Re: [Beignet] [PATCH V2 2/3] Runtime: Add new API enums for cl_intel_required_subgroup_size extension

The patchset LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Xiuli Pan > Sent: Thursday, June 15, 2017 16:45 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli > Subject: [Beignet] [PATCH V2

Re: [Beignet] [PATCH] do constant folding for kernel struct args

Ok, It only handle very special case, and must all before postPhiCopyOptimization. I will pushed it, thanks. > -Original Message- > From: Guo, Yejun > Sent: Tuesday, June 13, 2017 20:51 > To: Yang, Rong R <rong.r.y...@intel.com>; Wang, Rander > <rander.w...@intel

Re: [Beignet] [PATCH] backend: refine the local copy propagation.

Addc also can't use src modifier. All instruction selection opcode used addc need also return false, such as hadd. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > rander.wang > Sent: Tuesday, June 13, 2017 15:08 > To:

Re: [Beignet] [PATCH 2/2] Use aligned16 and aligne4 kernel to copy for large 3D image with TILE_Y.

LGTM, except some format. I have run git clang-format by manual and pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > yan.w...@linux.intel.com > Sent: Tuesday, June 13, 2017 16:32 > To: beignet@lists.freedesktop.org > Cc:

Re: [Beignet] [PATCH 1/3] Backend: Add intel_reqd_sub_group_size support

Xiuli > Sent: Tuesday, June 13, 2017 17:04 > To: Yang, Rong R <rong.r.y...@intel.com>; beignet@lists.freedesktop.org > Subject: RE: [Beignet] [PATCH 1/3] Backend: Add intel_reqd_sub_group_size > support > > The spec has required the subgroup size to be 8 or 16, and I think w

Re: [Beignet] [PATCH] do constant folding for kernel struct args

Yejun > Sent: Tuesday, June 13, 2017 16:39 > To: Yang, Rong R <rong.r.y...@intel.com>; Wang, Rander > <rander.w...@intel.com>; Pan, Xiuli <xiuli@intel.com>; > beignet@lists.freedesktop.org > Subject: RE: [Beignet] [PATCH] do constant folding for kernel st

Re: [Beignet] [PATCH 1/3] Backend: Add intel_reqd_sub_group_size support

> -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Xiuli Pan > Sent: Monday, June 5, 2017 16:28 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli > Subject: [Beignet] [PATCH 1/3] Backend: Add

Re: [Beignet] [PATCH 2/3] Runtime: Add new API enums for cl_intel_required_subgroup_size extension

Also need add the extension define to backend/src/libocl/include/ocl.h file. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Xiuli Pan > Sent: Monday, June 5, 2017 16:28 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli

Re: [Beignet] [PATCH 3/3] Utset: Add test case for cl_intel_required_subgroup_size extension

> -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Xiuli Pan > Sent: Monday, June 5, 2017 16:28 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli > Subject: [Beignet] [PATCH 3/3] Utset: Add test case for >

Re: [Beignet] [PATCH v5 7/7] Optimize clEnqueueWriteImageByKernel and clEnqueuReadImageByKernel.

The patchset LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > yan.w...@linux.intel.com > Sent: Tuesday, June 13, 2017 15:46 > To: beignet@lists.freedesktop.org > Cc: Yan Wang > Subject:

Re: [Beignet] [PATCH] do constant folding for kernel struct args

Has you consider the value from two arguments case. For example: Struct s1{ int i, float4 f4; } Struct s2{ int i; short s; float4 f4; } __kernel void k(s1, s2, __global float *dst) { int gid = get_global_id(0); float4 *p; if (gid % 2) { p = } else {

Re: [Beignet] [PATCH] Runtime: Fix a mssing llvm version marco for LLVM40+

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Pan, Xiuli > Sent: Monday, June 5, 2017 16:25 > To: Pan, Xiuli ; beignet@lists.freedesktop.org > Subject: Re: [Beignet] [PATCH] Runtime: Fix a

Re: [Beignet] [PATCH] backend: refine exp function with float input

Pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Pan, Xiuli > Sent: Wednesday, June 7, 2017 15:52 > To: Wang, Rander ; beig...@freedesktop.org > Cc: Wang, Rander > Subject: Re:

Re: [Beignet] [PATCH] backend: refine hypot function

Pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Pan, Xiuli > Sent: Wednesday, June 7, 2017 15:53 > To: Wang, Rander ; beig...@freedesktop.org > Cc: Wang, Rander > Subject: Re:

Re: [Beignet] [PATCH] keep GEN IR as SSA style

LGTM, pushed, thanks. Please add Signed-off-by message next time. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Guo, Yejun > Sent: Wednesday, June 7, 2017 15:44 > To: beignet@lists.freedesktop.org > Cc: Guo, Yejun

Re: [Beignet] [PATCH v2 2/2] Fix bug of clEnqueueCopyBufferToImage and clEnqueueCopyImageToBuffer.

The patchset LGTM, pushed, thanks. BYW: should also support align2 later. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > yan.w...@linux.intel.com > Sent: Thursday, May 25, 2017 15:10 > To: beignet@lists.freedesktop.org > Cc: Yan Wang

Re: [Beignet] [PATCH] build: fix cmake code generation dependencies.

LGTM, thanks, will push it later. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Ismo Puustinen > Sent: Tuesday, May 23, 2017 16:27 > To: beignet@lists.freedesktop.org > Cc: Puustinen, Ismo > Subject: [Beignet]

Re: [Beignet] [PATCH 2/2] Fix bug of clEnqueueCopyBufferToImage and clEnqueueCopyImageToBuffer.

Can you write a function to decide the alignment? Then it could be reused by cl_mem_copy_buffer_to_image and cl_mem_copy_image_to_buffer. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > yan.w...@linux.intel.com > Sent: Thursday, May 25,

Re: [Beignet] [PATCH V2] Backend: Fix performance regression with sampler refine fro LLVM40

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Xiuli Pan > Sent: Wednesday, May 17, 2017 17:02 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli > Subject: [Beignet] [PATCH V2] Backend: Fix

Re: [Beignet] [PATCH v3 8/8] Implement TILE_Y large image in clEnqueueWriteImage.

The patchset LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > yan.w...@linux.intel.com > Sent: Tuesday, May 16, 2017 19:04 > To: beignet@lists.freedesktop.org > Cc: Yan Wang > Subject:

Re: [Beignet] [PATCH] Backend: Fix llvm40 assert about literal structs

Pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Guo, Yejun > Sent: Wednesday, May 17, 2017 20:57 > To: Pan, Xiuli ; beignet@lists.freedesktop.org > Subject: Re: [Beignet] [PATCH] Backend: Fix llvm40

Re: [Beignet] [PATCH 3/3] GLK: add geminilake runtime support.

Pushed, thanks. > -Original Message- > From: Pan, Xiuli > Sent: Monday, May 15, 2017 15:05 > To: Yang, Rong R <rong.r.y...@intel.com>; beignet@lists.freedesktop.org > Cc: Yang, Rong R <rong.r.y...@intel.com> > Subject: RE: [Beignet] [PATCH 3/3] GLK:

Re: [Beignet] How widespread is "Exec event...error...-5" (#98647 / #100639)?

As we know, this issue is introduce by commit https://cgit.freedesktop.org/beignet/commit/?id=ff57cee0519db1287053c7c05a2cb4e9700d3334. To clarify, this commit is not only for ocl 2.0, ocl 1.2 also need it for null point check in the opencl kernel. Suppose a corner case: __kernel test(__global

Re: [Beignet] Release or patch for LLVM 4.0

The LLVM 4.0 patchset has been merged to master. We could cherry-pick these patches (from 8570e5 to 919dec) to branch Release_1.3, there is no conflict. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Bruno Pagani > Sent: Tuesday, April

Re: [Beignet] Random error with very low prabability in Haswell platform

How long don’t your benchmark run? Does linux kernel reset the GPU? You could run `dmesg` to get this information. From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of Song, Ruiling Sent: Tuesday, April 11, 2017 9:22 To: Gao, Sanshan ;

Re: [Beignet] [PATCH newRT] Wrap all memory allocate functions.

Actually, you implement a hash table with insert/delete operations, does linux has these apis? > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > junyan...@inbox.com > Sent: Thursday, March 23, 2017 15:46 > To: beignet@lists.freedesktop.org

Re: [Beignet] Limit get_program_global_data() calls to OpenCL 2.0

The workaround LGTM, pushed, thanks. BTW, please add the signed-off-by information by `git format-patch -s` next time. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Jan Beich > Sent: Thursday, March 16, 2017 18:13 > To:

Re: [Beignet] [Patch V2 2/3] fix regression on pre-BDW platform.

The patchset LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > xionghu@intel.com > Sent: Monday, March 20, 2017 22:38 > To: beignet@lists.freedesktop.org > Cc: Luo, Xionghu > Subject:

Re: [Beignet] [PATCHv2] Properly check return value from __cxa_demangle

Pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Jan Beich > Sent: Friday, March 17, 2017 22:16 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli > Subject: [Beignet] [PATCHv2] Properly check return

Re: [Beignet] [PATCH V2] CMAKE: Refine builtin kernel bin generator

Pushed. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Xiuli Pan > Sent: Thursday, March 2, 2017 11:34 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli > Subject: [Beignet] [PATCH V2] CMAKE: Refine builtin

Re: [Beignet] [PATCH 1/4] Backend: Add missing Unaligned OWord Block Read disasm

This patch LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Xiuli Pan > Sent: Tuesday, March 7, 2017 12:35 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli > Subject: [Beignet] [PATCH 1/4]

Re: [Beignet] [PATCH 7/7] Backend: for BDW and after, According to BSpec no need to split CMP when src is DW DF

There is a build error: error: no ‘bool gbe::Gen8Encoder::needToSplitCmpBySrcType(gbe::GenEncoder*, gbe::GenRegister, gbe::GenRegister)’ member function declared in class ‘gbe::Gen8Encoder’ bool Gen8Encoder::needToSplitCmpBySrcType(GenEncoder *p, GenRegister src0, GenRegister src1) { Need

Re: [Beignet] [PATCH 2/2] Backend: refine the geometry function

Pushed. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Song, Ruiling > Sent: Tuesday, March 7, 2017 13:54 > To: Wang, Rander ; beignet@lists.freedesktop.org > Cc: Wang, Rander > Subject: Re:

Re: [Beignet] [PATCH RFC] Add AppStream metadata

metadata_license is just the license for this XML, right? If so, MIT is ok for me, what your opinion? I have no problem with id section. open(os.path.join(source_directory,"src/cl_device_data.h"),"r",encoding= 'utf-8') has an error in python2.x: TypeError: 'encoding' is an invalid keyword

Re: [Beignet] [PATCH] Backend: Fix a selection ir optimization bug

LGTM, will push it later, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Xiuli Pan > Sent: Monday, February 27, 2017 11:16 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli > Subject: [Beignet] [PATCH]

Re: [Beignet] [PATCH] MAD compact instrcution could not support "absolute" attribute.

Pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Pan, Xiuli > Sent: Thursday, February 23, 2017 16:34 > To: yan.w...@linux.intel.com; beignet@lists.freedesktop.org > Subject: Re: [Beignet] [PATCH] MAD compact instrcution

Re: [Beignet] [PATCH] Enable support for two-component 16-bit planes

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Mark Thompson > Sent: Saturday, February 11, 2017 21:50 > To: beignet@lists.freedesktop.org > Subject: [Beignet] [PATCH] Enable support for two-component 16-bit planes

Re: [Beignet] [PATCH V5] Enable OpenCL 2.0 only where supported

Pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Xiuli Pan > Sent: Monday, February 13, 2017 9:52 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli > Subject: [Beignet] [PATCH V5] Enable OpenCL 2.0

Re: [Beignet] [PATCH V4] Enable OpenCL 2.0 only where supported

> -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Rebecca N. Palmer > Sent: Saturday, February 11, 2017 7:02 > To: beignet@lists.freedesktop.org > Subject: Re: [Beignet] [PATCH V4] Enable OpenCL 2.0 only where supported

Re: [Beignet] [PATCH v2 2/2] Add document of using cl_khr_gl_sharing to do gl buffer sharing.

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Chuanbo Weng > Sent: Friday, February 10, 2017 15:48 > To: beignet@lists.freedesktop.org > Cc: Weng, Chuanbo > Subject: [Beignet] [PATCH v2

Re: [Beignet] [PATCH] API: Fix device type bugs

age- > From: Yang, Rong R > Sent: Friday, February 10, 2017 15:40 > To: 'Xiuli Pan' <xiuli@intel.com>; beignet@lists.freedesktop.org > Cc: giuseppe.bilo...@gmail.com; Pan, Xiuli <xiuli@intel.com> > Subject: RE: [Beignet] [PATCH] API: Fix device type bugs > > Th

Re: [Beignet] [PATCH] API: Fix device type bugs

There are two types of relative return error, one is CL_INVALID_DEVICE_TYPE, for invalid value, another is CL_DEVICE_NOT_FOUND, for no matched device. The original check for CL_INVALID_DEVICE_TYPE is right, but miss the CL_DEVICE_NOT_FOUND check. Giuseppe's patch will return

Re: [Beignet] [PATCH] Free context devices on context release

LGTM, thanks, pushed. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Giuseppe Bilotta > Sent: Friday, February 10, 2017 6:13 > To: Beignet ML > Cc: Giuseppe Bilotta > Subject:

Re: [Beignet] [PATCH] Fix obvious copy-paste

Yes, always NULL. LGTM, thanks, pushed. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Giuseppe Bilotta > Sent: Friday, February 10, 2017 6:13 > To: Beignet ML > Cc: Giuseppe Bilotta

Re: [Beignet] [PATCH 5/5] Enable OpenCL 2.0 only where supported

Because use -cl-std=CL1.2 by default when OpenCL 2.0 enabled, I prefer to always report address_bits = 32 now. OpenCL spec consider only one address bits in one device, but when GEN9 now support both 32 bits and 64 bits address, so there is no way to comply with spec. I think we could change

Re: [Beignet] [PATCH] Typo in error message

Pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > He Junyan > Sent: Friday, February 3, 2017 21:26 > To: beignet@lists.freedesktop.org > Subject: Re: [Beignet] [PATCH] Typo in error message > > Thanks for fixing it. > > On

Re: [Beignet] [PATCH] Add a CMake option for toggling OCL ICD Loader compatibility

LGTM, pushed, thanks. As Simon's suggestion, I think it is make sense to ask users to explicitly disable ICD if no ICD header, I will send a patch for it. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Marek Szuba > Sent: Wednesday,

Re: [Beignet] [PATCH] GBE: use shift for PowerOf2 size when lowering GEP.

This patch looks good to me. As discuss, we would better add the similar optimize in GEN selection, because GEP lower is a part of PowerOf 2 multiply. I will add it later. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Ruiling Song >

Re: [Beignet] [PATCH] Fix typo

Pushed. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Pan, Xiuli > Sent: Monday, January 23, 2017 10:38 > To: Rebecca N. Palmer ; > beignet@lists.freedesktop.org > Subject: Re: [Beignet] [PATCH] Fix typo > >

Re: [Beignet] [PATCH] Make CL-GL sharing available via ICD

Pushed. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Weng, Chuanbo > Sent: Monday, January 23, 2017 19:44 > To: Rebecca N. Palmer ; > beignet@lists.freedesktop.org > Subject: Re: [Beignet] [PATCH] Make CL-GL

[Beignet] [ANNOUNCE] Beignet 1.3.0

ong ctz fail. GBE: fix long work group fail. Runtime: Fix a event bug. GBE: if PointerFamily is FAMILY_QWORD, chv and bxt need special handle. GBE: fix legacy read64 mix pointer bug. GBE: fix a mix analyze bug. Add some pointer access check. Yang, Rong R

Re: [Beignet] [PATCH] Doc: add OpenCL 2.0 section to readme.

Thanks, refine and pushed. > -Original Message- > From: Song, Ruiling > Sent: Friday, January 20, 2017 16:33 > To: Yang, Rong R <rong.r.y...@intel.com>; beignet@lists.freedesktop.org > Cc: Yang, Rong R <rong.r.y...@intel.com> > Subject: RE: [Beignet] [PATC

Re: [Beignet] [PATCH] Fix two bugs about command queue destroy.

LGTM, pushed. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > junyan...@inbox.com > Sent: Wednesday, January 11, 2017 16:36 > To: beignet@lists.freedesktop.org > Cc: He, Junyan > Subject: [Beignet] [PATCH] Fix two

Re: [Beignet] [PATCH] GBE: Fix getTypesize bug with LLVM3.9

LLVM3.9 have changed pipes' meta, LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Xiuli Pan > Sent: Monday, January 9, 2017 15:00 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli > Subject:

Re: [Beignet] [PATCH] Fail, don't assert, if unable to create context

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Rebecca N. Palmer > Sent: Monday, January 9, 2017 5:33 > To: beignet@lists.freedesktop.org > Subject: [Beignet] [PATCH] Fail, don't assert, if unable to create context

Re: [Beignet] [Patch V2 1/2] GBE: fix legacy read64 mix pointer bug.

Pushed, thanks. > -Original Message- > From: Song, Ruiling > Sent: Monday, January 9, 2017 13:21 > To: Yang, Rong R <rong.r.y...@intel.com>; beignet@lists.freedesktop.org > Cc: Yang, Rong R <rong.r.y...@intel.com> > Subject: RE: [Beignet] [Patch V2 1/2] GBE:

Re: [Beignet] [PATCH] Runtime: return CL_INVALID_EVENT_WAIT_LIST if not event in the wait list.

Pushed. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > He Junyan > Sent: Wednesday, December 28, 2016 17:56 > To: beignet@lists.freedesktop.org > Subject: Re: [Beignet] [PATCH] Runtime: return > CL_INVALID_EVENT_WAIT_LIST if not event in

Re: [Beignet] [PATCH] Avoid possible invalid pointer by vector interator.

Pushed. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Song, Ruiling > Sent: Wednesday, December 28, 2016 16:33 > To: yan.w...@linux.intel.com; beignet@lists.freedesktop.org > Subject: Re: [Beignet] [PATCH] Avoid possible invalid pointer

Re: [Beignet] [PATCH 10/19] OCL20: enable -cl-std=CL2.0.

The first 10 patches pushed, thanks. > -Original Message- > From: Song, Ruiling > Sent: Monday, December 19, 2016 14:47 > To: Yang, Rong R <rong.r.y...@intel.com>; beignet@lists.freedesktop.org; > Pan, Xiuli <xiuli....@intel.com> > Cc: Yang, Rong R <ron

Re: [Beignet] [PATCH] Refine list related functions.

LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > junyan...@inbox.com > Sent: Wednesday, December 21, 2016 19:05 > To: beignet@lists.freedesktop.org > Subject: [Beignet] [PATCH] Refine list related functions. > >

  1   2   3   4   5   6   7   >