April 2016 Archives by author

• Messages sorted by: [ thread ] [ subject ] [ date ]
• More info on this list...

Starting: Fri Apr 1 04:39:24 UTC 2016
Ending: Fri Apr 29 06:32:04 UTC 2016
Messages: 138

• [Beignet] [PATCH 1/2] handle simd8 and simd16 accrordingly for alu3.   Guo, Yejun
• [Beignet] [PATCH V2 2/2] enable utest compiler_math_3op for mad test.   Guo, Yejun
• [Beignet] [PATCH] fix typo.   Guo, Yejun
• [Beignet] [PATCH V2 2/2] write mask in disassembly not parse correctly.   Guo, Yejun
• [Beignet] [PATCH V2 1/2] assert equation issue.   Guo, Yejun
• [Beignet] [PATCH] GBE: try to avoid bank conflict in register allocator.   Luo, Xionghu
• [Beignet] [PATCH] set SIMD width as 1 for mad when the dst is uniform   Luo, Xionghu
• [Beignet] [PATCH 1/2] Utest: Add workgroup broadcast tests   Grigore Lupescu
• [Beignet] [PATCH 2/2] Backend: Fix workgroup broadcast, add initial support qword   Grigore Lupescu
• [Beignet] [PATCH] Backend: Fix workgroup broadcast on QWORD, 1D works   Grigore Lupescu
• [Beignet] [PATCH] Backend: Fix workgroup all operation by selective mask   Grigore Lupescu
• [Beignet] [PATCH] Backend: Fix workgroup op masked elements by proper init value   Grigore Lupescu
• [Beignet] [PATCH V2 01/17] Backend: Workgroup reduce redesign using shared local memory   Grigore Lupescu
• [Beignet] [PATCH] Debug: Correct function type is being lost att LLVM backend   Grigore Lupescu
• [Beignet] [PATCH V2 02/17] Backend: Support for workgroup op predicate, scan inclusive/exclusive   Grigore Lupescu
• [Beignet] [PATCH V2 03/17] Backend: Initial support for long/ulong types in workgroup ops   Grigore Lupescu
• [Beignet] [PATCH V2 04/17] Backend: Fix barrier placement in workgroup functions   Grigore Lupescu
• [Beignet] [PATCH V2 05/17] Backend: Workgroup scan cmp qword, workaround for execution width 16   Grigore Lupescu
• [Beignet] [PATCH V2 06/17] Backend: Fix workgroup broadcast, support for qword   Grigore Lupescu
• [Beignet] [PATCH V2 07/17] Backend: Fix workgroup op masked elements by proper init value   Grigore Lupescu
• [Beignet] [PATCH V2 08/17] Backend: Code refactoring, additional comments on implementation   Grigore Lupescu
• [Beignet] [PATCH V2 09/17] Backend: Fix workgroup barrier call in gen context   Grigore Lupescu
• [Beignet] [PATCH V2 10/17] Backend: Fix workgroup float max init value   Grigore Lupescu
• [Beignet] [PATCH V2 11/17] Utest: Add workgroup reduce tests   Grigore Lupescu
• [Beignet] [PATCH V2 12/17] Utest: Add workgroup scan exclusive tests   Grigore Lupescu
• [Beignet] [PATCH V2 13/17] Utest: Add workgroup scan inclusive tests   Grigore Lupescu
• [Beignet] [PATCH V2 14/17] Utest: Add workgroup all/any tests   Grigore Lupescu
• [Beignet] [PATCH V2 15/17] Utest: Add workgroup broadcast tests   Grigore Lupescu
• [Beignet] [PATCH V2 16/17] Benchmark: Add performance tests for workgroup reduce/scan functions   Grigore Lupescu
• [Beignet] [PATCH V2 17/17] Benchmark: Add performance tests for workgroup broadcast   Grigore Lupescu
• [Beignet] [PATCH] FindLLVM: allow LLVM/Clang 3.7   Rebecca N. Palmer
• [Beignet] Fw: Skylake support ...   Pan, Xiuli
• [Beignet] GROMACS on beignet   Pan, Xiuli
• [Beignet] [PATCH] Utest: Add utest load spir for spir64   Xiuli Pan
• [Beignet] [PATCH 3/7] GBE: add ocl 2.0 work_group_barrier support.   Xiuli Pan
• [Beignet] [PATCH OCL20] Runtime: Add new param_name to clGetProgramBuildInfo   Xiuli Pan
• [Beignet] GROMACS on beignet   Pan, Xiuli
• [Beignet] [PATCH] Prepend std namespace to isnan and isinf calls.   Pan, Xiuli
• [Beignet] [PATCH] runtime: handle PROGRAM_BUILD_GLOBAL_VARIABLE_TOTAL_SIZE   Pan, Xiuli
• [Beignet] [PATCH 09/10] Backend: Full support workgroup reduce, scan inc/exc on DWORD and bellow datatypes   Xiuli Pan
• [Beignet] [PATCH V2] GBE: add ocl 2.0 work_group_barrier support.   Pan, Xiuli
• [Beignet] GROMACS on beignet   Pan, Xiuli
• [Beignet] [PATCH V2 15/17] Utest: Add workgroup broadcast tests   Xiuli Pan
• [Beignet] [PATCH V2 12/17] Utest: Add workgroup scan exclusive tests   Xiuli Pan
• [Beignet] [PATCH V2 12/17] Utest: Add workgroup scan exclusive tests   Xiuli Pan
• [Beignet] [PATCH OCL20] Backend: Refine workgroup all with SIMD_ALL algorithm   Xiuli Pan
• [Beignet] [PATCH OCL20 V2] Backend: Refine workgroup all with SIMD_ALL algorithm   Xiuli Pan
• [Beignet] [PATCH OCL20] Backend: Refine script for math.h   Xiuli Pan
• [Beignet] [PATCH OCL20 V2] GBE: imm64 should not be in src1 per hardware spec. V2: Refine assert to check if the value of imm can not be a imm32   Xiuli Pan
• [Beignet] [PATCH OCL20 V3] Backend: Refine workgroup all with SIMD_ALL algorithm   Xiuli Pan
• [Beignet] [PATCH] GBE: Fix destination grf register type for cmp instruction.   Pan, Xiuli
• [Beignet] [PATCH 3/5] GBE: imm64 should not be in src1 per hardware spec.   Pan, Xiuli
• [Beignet] [PATCH] runtime: handle PROGRAM_BUILD_GLOBAL_VARIABLE_TOTAL_SIZE   Pan, Xiuli
• [Beignet] [PATCH] Backend: Chang scan limit for GVN pass   Xiuli Pan
• [Beignet] [PATCH OCL20 V2 1/4] Backend: Chang scan limit for GVN pass   Xiuli Pan
• [Beignet] [PATCH OCL20 V2 2/4] Backend: Refine script for math.h   Xiuli Pan
• [Beignet] [PATCH OCL20 V2 3/4] Backend: Add workaround for instcombine will optimize fabs   Xiuli Pan
• [Beignet] [PATCH OCL20 V2 4/4] Runtime: Add support for queue size and fix error handling   Xiuli Pan
• [Beignet] [PATCH] Backend: Fix bug build with clang   Xiuli Pan
• [Beignet] GROMACS on beignet   Szilárd Páll
• [Beignet] GROMACS on beignet   Szilárd Páll
• [Beignet] GROMACS on beignet   Szilárd Páll
• [Beignet] [PATCH 1/7] GBE: Fix type mismatch bug.   Ruiling Song
• [Beignet] [PATCH 2/7] GBE: Fix SEL.bool issue.   Ruiling Song
• [Beignet] [PATCH 3/7] GBE: add ocl 2.0 work_group_barrier support.   Ruiling Song
• [Beignet] [PATCH 4/7] GBE: Fix bug when unspill a long type value from scratch.   Ruiling Song
• [Beignet] [PATCH 5/7] GBE: don't try to erase a llvm:Constant.   Ruiling Song
• [Beignet] [PATCH 6/7] GBE: the dst grf should use same width as source register   Ruiling Song
• [Beignet] [PATCH 7/7] GBE: retype double register to long type when do spilling.   Ruiling Song
• [Beignet] [PATCH 1/7] GBE: Fix type mismatch bug.   Song, Ruiling
• [Beignet] [PATCH 3/7] GBE: add ocl 2.0 work_group_barrier support.   Song, Ruiling
• [Beignet] [PATCH V2] GBE: add ocl 2.0 work_group_barrier support.   Ruiling Song
• [Beignet] [PATCH OCL20 v2] Backend: Refine typedef of ptrint_t   Song, Ruiling
• [Beignet] [PATCH] runtime: handle PROGRAM_BUILD_GLOBAL_VARIABLE_TOTAL_SIZE   Ruiling Song
• [Beignet] [PATCH OCL20] Runtime: Add new param_name to clGetProgramBuildInfo   Song, Ruiling
• [Beignet] [PATCH 1/5] GBE: Fix a typo in RelocTable   Ruiling Song
• [Beignet] [PATCH 2/5] runtime: prog->global_data may get 64bit address   Ruiling Song
• [Beignet] [PATCH 3/5] GBE: imm64 should not be in src1 per hardware spec.   Ruiling Song
• [Beignet] [PATCH 4/5] GBE: handle ConstantExpr in program-scope variable handling.   Ruiling Song
• [Beignet] [PATCH 5/5] GBE: Refine program scope variable logic.   Ruiling Song
• [Beignet] [PATCH] GBE: Fix destination grf register type for cmp instruction.   Ruiling Song
• [Beignet] [PATCH] GBE: try to avoid bank conflict in register allocator.   Ruiling Song
• [Beignet] [PATCH] runtime: we can accept large program global variable.   Ruiling Song
• [Beignet] [PATCH V2] GBE: try to avoid bank conflict in register allocator.   Ruiling Song
• [Beignet] [PATCH] GBE: try to avoid bank conflict in register allocator.   Song, Ruiling
• [Beignet] [PATCH V3] GBE: Do more aggressive load/store merging.   Ruiling Song
• [Beignet] [PATCH V2] GBE: try to avoid bank conflict in register allocator.   Matt Turner
• [Beignet] [PATCH] cl_api; Check image origin and region for NULL   Jan Vesely
• [Beignet] [PATCH] Flush kernel source dump stream   Zhenyu Wang
• [Beignet] [PATCH] Add condition checking of residuals because it may be NULL.   Yang, Rong R
• [Beignet] [PATCH] Libocl: change prototype of math built-in for OCL2.0 spec   Yang, Rong R
• [Beignet] [PATCH] [OCL20] atomic bug fix.   Yang, Rong R
• [Beignet] [PATCH] Prepend std namespace to isnan and isinf calls.   Yang, Rong R
• [Beignet] [PATCH] utest: do not check MV near image border   Yang, Rong R
• [Beignet] [PATCH] libocl: Refine return type of workitem built-in functions   Yang, Rong R
• [Beignet] [PATCH] Runtime: Add SKL device id for new SKL device   Yang, Rong R
• [Beignet] [PATCH] Increase size for compile log output   Yang, Rong R
• [Beignet] [PATCH V2] GBE: add ocl 2.0 work_group_barrier support.   Yang, Rong R
• [Beignet] [PATCH] Flush kernel source dump stream   Yang, Rong R
• [Beignet] [PATCH V2] Utest: Add result error tolerance for float data types   Yang, Rong R
• [Beignet] [PATCH v4 3/5] standalone utest for unified OpenCL implementation.   Yang, Rong R
• [Beignet] [PATCH] GBE: Fix destination grf register type for cmp instruction.   Yang, Rong R
• [Beignet] [PATCH] runtime: handle PROGRAM_BUILD_GLOBAL_VARIABLE_TOTAL_SIZE   Yang, Rong R
• [Beignet] [PATCH] only release cmrt device when it is already created   Yang, Rong R
• [Beignet] [PATCH V2 2/2] write mask in disassembly not parse correctly.   Yang, Rong R
• [Beignet] [PATCH V2 2/2] enable utest compiler_math_3op for mad test.   Yang, Rong R
• [Beignet] drm/i915/bxt: Pooled EU support in beignet   Yang, Rong R
• [Beignet] [PATCH OCL20 V2 1/4] Backend: Chang scan limit for GVN pass   Yang, Rong R
• [Beignet] [PATCH] set SIMD width as 1 for mad when the dst is uniform   Guo Yejun
• [Beignet] [PATCH] only release cmrt device when it is already created   Guo Yejun
• [Beignet] [PATCH] Backend: Fix barrier placement in workgroup reduce/scan   grigore.lupescu at intel.com
• [Beignet] [PATCH 1/3] Benchmark: Fix benchmark kernel of workgroup, function input   grigore.lupescu at intel.com
• [Beignet] [PATCH 2/3] Backend: Workgroup functions no check for src until fix on wellFormed   grigore.lupescu at intel.com
• [Beignet] [PATCH 3/3] Backend: Workgroup scan cmp on QW, workaround for width 16   grigore.lupescu at intel.com
• [Beignet] [PATCH V2] Utest: Fix broadcast long input type   grigore.lupescu at intel.com
• [Beignet] [PATCH V2] Utest: Fix scan exclusive limits in float types   grigore.lupescu at intel.com
• [Beignet] [PATCH V2] Utest: Remove scan exclusive tests of char/short   grigore.lupescu at intel.com
• [Beignet] [PATCH V2] Utest: Add result error tolerance for float data types   grigore.lupescu at intel.com
• [Beignet] [PATCH 1/3] add sanity check for Image Region in runtime.   xionghu.luo at intel.com
• [Beignet] [PATCH 2/3] standalone utest for unified OpenCL implementation.   xionghu.luo at intel.com
• [Beignet] [PATCH 3/3] fix utest failures on VPG code.   xionghu.luo at intel.com
• [Beignet] [PATCH v2 1/3] add sanity check for Image Region in runtime.   xionghu.luo at intel.com
• [Beignet] [PATCH v2 2/3] standalone utest for unified OpenCL implementation.   xionghu.luo at intel.com
• [Beignet] [PATCH v2 3/3] fix utest failures on VPG code.   xionghu.luo at intel.com
• [Beignet] [PATCH 1/5] add sanity check for Image Region in runtime.   xionghu.luo at intel.com
• [Beignet] [PATCH 2/5] buf[0] destroyed twice in event cases.   xionghu.luo at intel.com
• [Beignet] [PATCH 3/5] standalone utest for unified OpenCL implementation.   xionghu.luo at intel.com
• [Beignet] [PATCH 4/5] fix failed cases for stand alone utest;   xionghu.luo at intel.com
• [Beignet] [PATCH 5/5] add functions cl_check_beignet.   xionghu.luo at intel.com
• [Beignet] [PATCH v4 3/5] standalone utest for unified OpenCL implementation.   xionghu.luo at intel.com
• [Beignet] [PATCH 1/2] handle simd8 and simd16 accrordingly for alu3.   xionghu.luo at intel.com
• [Beignet] [PATCH 2/2] enable utest compiler_math_3op for mad test.   xionghu.luo at intel.com
• [Beignet] [PATCH V2 0/2] patch dependency.   xionghu.luo at intel.com
• [Beignet] [PATCH V2 1/2] handle simd8 and simd16 accrordingly for alu3.   xionghu.luo at intel.com
• [Beignet] [PATCH V2 2/2] enable utest compiler_math_3op for mad test.   xionghu.luo at intel.com
• [Beignet] [PATCH] fix typo.   xionghu.luo at intel.com
• [Beignet] [PATCH V2 1/2] assert equation issue.   xionghu.luo at intel.com
• [Beignet] [PATCH V2 2/2] write mask in disassembly not parse correctly.   xionghu.luo at intel.com

Last message date: Fri Apr 29 06:32:04 UTC 2016
Archived on: Fri Apr 29 06:32:15 UTC 2016

• Messages sorted by: [ thread ] [ subject ] [ date ]
• More info on this list...