Re: [Beignet] [PATCH] Runtime: because double's built-ins haven't completely support, so disable it by default.

2015-12-22 Thread Zhigang Gong
This patch LGTM. Thanks, Zhigang Gong. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Yang Rong > Sent: Tuesday, December 22, 2015 4:37 PM > To: beignet@lists.freedesktop.org > Cc: Yang Rong > Subject: [Beignet]

Re: [Beignet] [PATCH] Make FindLLVM.cmake work with empty "system-libs"

2015-12-22 Thread Yang, Rong R
Yes, if everything is built from source, LLVM don't need static libraries. The patch looks good to me, thanks for your contribute. Could you post the e-mail submission patch? Use git format-patch and git send-email. > -Original Message- > From: Beignet

Re: [Beignet] [PATCH] Driver: Fix GPGPU delete bug

2015-12-22 Thread Yang, Rong R
LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Pan Xiuli > Sent: Tuesday, December 22, 2015 15:27 > To: beignet@lists.freedesktop.org > Cc: Pan, Xiuli > Subject: [Beignet] [PATCH] Driver: Fix GPGPU delete bug > >

Re: [Beignet] [PATCH] output warning message if do not find a good local_work_size

2015-12-22 Thread Yang, Rong R
LGTM, pushed, thanks. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Guo Yejun > Sent: Tuesday, December 22, 2015 7:49 > To: beignet@lists.freedesktop.org > Cc: Guo, Yejun > Subject: [Beignet] [PATCH] output warning message if do not

Re: [Beignet] [PATCH] fix a bug when the first operand of intel_sub_group_shuffle is uniform

2015-12-22 Thread Guo, Yejun
ping for review, thanks. -Original Message- From: Guo, Yejun Sent: Thursday, December 17, 2015 6:43 AM To: beignet@lists.freedesktop.org Cc: Guo, Yejun Subject: [PATCH] fix a bug when the first operand of intel_sub_group_shuffle is uniform actually, if the first operand is uniform, no

Re: [Beignet] [PATCH 2/3] ocl20/GBE: support generic load/store

2015-12-22 Thread Yang, Rong R
The patcheset LGTM, pushed. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Ruiling Song > Sent: Friday, November 27, 2015 10:09 > To: beignet@lists.freedesktop.org > Cc: Song, Ruiling > Subject: [Beignet] [PATCH 2/3] ocl20/GBE: support

[Beignet] [PATCH] Runtime: because double's built-ins haven't completely support, so disable it by default.

2015-12-22 Thread Yang Rong
Add a cmake option for it, cmake with option -DEXPERIMENTAL_DOUBLE=true to enable it. --- CMakeLists.txt | 6 ++ src/cl_device_id.c | 21 + 2 files changed, 27 insertions(+) diff --git a/CMakeLists.txt b/CMakeLists.txt index 8762f7c..97725ca 100644 ---

Re: [Beignet] [PATCH V3] ocl20/runtime: take the first 64KB page table entries.

2015-12-22 Thread Luo, Xionghu
This patch LGTM. Thanks. Luo Xionghu Best Regards -Original Message- From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of Ruiling Song Sent: Tuesday, December 22, 2015 1:47 PM To: beignet@lists.freedesktop.org Cc: Song, Ruiling Subject: [Beignet] [PATCH V3]

[Beignet] [PATCH 2/2] change built-in function name from get_sub_group_id to get_sub_group_local_id

2015-12-22 Thread Guo Yejun
Fix bug at https://bugs.freedesktop.org/show_bug.cgi?id=93469 The fucntion is mapped to OP_SIMD_ID which returns the SIMD lane ID. However, the SIMD lane ID is the equivalent of get_sub_group_local_id(). contributor: Georg Kolling Signed-off-by: Guo Yejun