delta/beignet.git - gitlab.freedesktop.org: beignet/beignet.git

	Commit message (Expand)	Author	Age	Files	Lines
...
*	backend: refine the local copy propagation.	rander.wang	2017-06-16	1	-0/+34
*	Utset: Add test case for cl_intel_required_subgroup_size extension	Pan Xiuli	2017-06-16	5	-0/+75
*	Runtime: Add new API enums for cl_intel_required_subgroup_size extension	Pan Xiuli	2017-06-16	7	-0/+50
*	Backend: Add intel_reqd_sub_group_size support	Pan Xiuli	2017-06-16	3	-13/+45
*	do constant folding for kernel struct args	Guo Yejun	2017-06-16	6	-0/+213
*	Use aligned16 and aligne4 kernel to copy for large 3D image with TILE_Y.	Yan Wang	2017-06-14	7	-37/+149
*	Add test case for large 3D image with TILE_Y.	Yan Wang	2017-06-14	1	-0/+98
*	Optimize clEnqueueWriteImageByKernel and clEnqueuReadImageByKernel.	Yan Wang	2017-06-13	1	-7/+18
*	Fix bug of clEnqueueUnmapMemObjectForKernel and clEnqueueMapImageByKernel.	Yan Wang	2017-06-13	1	-34/+113
*	Add clFinish for guarantee the kernel copying is finished when create TILE_Y ...	Yan Wang	2017-06-13	1	-0/+7
*	Add cl_mem_record_map_mem_for_kernel() for record map adress for TILE_Y image...	Yan Wang	2017-06-13	2	-26/+88
*	Add utest to test writing data into large image (TILE_Y) by map/unmap and USE...	Yan Wang	2017-06-13	1	-0/+115
*	Add utest to test writing data into large image (TILE_Y) by map/unmap mode.	Yan Wang	2017-06-13	1	-0/+198
*	Add utest case for filling image by small region.	Yan Wang	2017-06-13	1	-0/+50
*	utests: added for optimization negativeAdd	rander	2017-06-09	3	-1/+46
*	Backend: Add optimization for negative modifier	rander	2017-06-09	1	-4/+28
*	utests: add utest for sqrt-div optimization	rander	2017-06-09	3	-1/+71
*	backend: add sqrt-div pattern to instruction select	rander	2017-06-09	1	-0/+69
*	Runtime: Fix a mssing llvm version marco for LLVM40+	Pan Xiuli	2017-06-09	1	-1/+1
*	keep GEN IR as SSA style	Guo Yejun	2017-06-09	1	-3/+5
*	backend: refine exp function with float input	rander	2017-06-09	1	-2/+56
*	backend: refine hypot function	rander	2017-06-09	1	-14/+60
*	Fix bug of clEnqueueCopyBufferToImage and clEnqueueCopyImageToBuffer.	Yan Wang	2017-05-25	5	-28/+89
*	Add utest to reproduce the bug of imagedim_non_pow_2 cases of conformance test.	Yan Wang	2017-05-25	1	-0/+46
*	build: fix cmake code generation dependencies.	Ismo Puustinen	2017-05-25	1	-2/+2
*	refresh DAG when an arg has both direct and indirect read	Guo Yejun	2017-05-23	1	-1/+16
*	Backend: Add sel ir output for MATH function	Pan Xiuli	2017-05-23	1	-0/+42
*	backend: fix tgamma error after restructure	rander	2017-05-23	1	-25/+31
*	Implement TILE_Y large image in clEnqueueWriteImage.	Yan Wang	2017-05-18	1	-0/+46
*	Implement TILE_Y large image in clEnqueueReadImage.	Yan Wang	2017-05-18	1	-0/+55
*	Implement TILE_Y large image in clEnqueueMapImage and clEnqueueUnmapMemObject.	Yan Wang	2017-05-18	1	-0/+111
*	Create image with TILE_Y mode still when image size>128MB for performance.	Yan Wang	2017-05-18	4	-6/+111
*	Add image use_hostptr benchmark case for testing large image operations.	Yan Wang	2017-05-18	2	-0/+85
*	Add image use_hostptr case for testing large image operations.	Yan Wang	2017-05-18	2	-0/+76
*	Add image filling case for testing large image operations.	Yan Wang	2017-05-18	2	-0/+121
*	Add image copying case for testing large image operations.	Yan Wang	2017-05-18	2	-0/+122
*	Backend: Fix performance regression with sampler refine fro LLVM40	Pan Xiuli	2017-05-18	2	-9/+41
*	Backend: Fix llvm40 assert about literal structs	Pan Xiuli	2017-05-18	1	-1/+2
*	backend: refine asin function	rander.wang	2017-05-17	1	-21/+7
*	backend: refine atan	rander.wang	2017-05-17	1	-53/+58
*	backend: refine acos	rander.wang	2017-05-17	1	-4/+9
*	backend: refine sincos	rander.wang	2017-05-17	2	-13/+277
*	backend: refine tan function	rander.wang	2017-05-17	1	-16/+45
*	backend: refine cos function	rander.wang	2017-05-17	1	-26/+25
*	backend: refine sin function	rander.wang	2017-05-17	1	-20/+22
*	backend: refine the argue reduce	rander.wang	2017-05-17	1	-24/+14
*	backend: refine pow function	rander.wang	2017-05-17	1	-134/+141
*	backend: refine the code structure of math	rander.wang	2017-05-17	9	-7538/+4073
*	GLK: add geminilake runtime support.	Yang Rong	2017-05-15	2	-2/+47
*	GLK: add geminilake backend support.	Yang Rong	2017-05-15	5	-2/+47