beignet  

Messages by Thread

• [Beignet] [PATCH] GBE: Optimize extraLiveOut register info. Ruiling Song
  • Re: [Beignet] [PATCH] GBE: Optimize extraLiveOut register info. Zhigang Gong
  • Re: [Beignet] [PATCH] GBE: Optimize extraLiveOut register info. Yang, Rong R
• [Beignet] Fix definition of FLT_MIN/FLT_MAX for clang-3.8 compilation Andrew Lavin
  • Re: [Beignet] Fix definition of FLT_MIN/FLT_MAX for clang-3.8 compilation Andrew Lavin
  • Re: [Beignet] Fix definition of FLT_MIN/FLT_MAX for clang-3.8 compilation Pan, Xiuli
  • Re: [Beignet] Fix definition of FLT_MIN/FLT_MAX for clang-3.8 compilation Yang, Rong R
• [Beignet] [PATCH 01/12] Runtime: Fix thread id calculation. Xiuli Pan
  • [Beignet] [PATCH 07/12] Backend: Add sub_group built-in functions for intel extension Xiuli Pan
  • [Beignet] [PATCH 09/12] Backend: Add intel_sub_group_block_read/write form buffer Xiuli Pan
  • [Beignet] [PATCH 04/12] Utest: Add subgroup work item test cases Xiuli Pan
  • [Beignet] [PATCH 10/12] Utest: Add tset case for block read/write buffer Xiuli Pan
  • [Beignet] [PATCH 11/12] Backend: Add intel_sub_group_block_read/write form image Xiuli Pan
  • [Beignet] [PATCH 03/12] Backend: Add subgroup work item builtin functions Xiuli Pan
  • [Beignet] [PATCH 06/12] Utest: Remove old sub_group_all/any utest Xiuli Pan
  • [Beignet] [PATCH 02/12] Runtime: Add API clGetKernelSubGroupInfoKHR for subgroup extension Xiuli Pan
  • [Beignet] [PATCH 08/12] Utest: Add test case for sub_group functions Xiuli Pan
  • [Beignet] [PATCH 05/12] Backend: Refine return value of sub_group_all/any to 1 Xiuli Pan
  • [Beignet] [PATCH 12/12] Utest: Add tset case for block read/write image Xiuli Pan
• [Beignet] [PATCH] GBE: fix a patch JMPI assert. Yang Rong
  • Re: [Beignet] [PATCH] GBE: fix a patch JMPI assert. Song, Ruiling
• [Beignet] [PATCH] Runtime: Use uanme to get kernel architecture. Yang Rong
  • Re: [Beignet] [PATCH] Runtime: Use uanme to get kernel architecture. Song, Ruiling
• [Beignet] [Patch V2 07/10] GBE: disable the read byte as DW. Yang Rong
  • Re: [Beignet] [Patch V2 07/10] GBE: disable the read byte as DW. Zhigang Gong
  • Re: [Beignet] [Patch V2 07/10] GBE: disable the read byte as DW. Yang, Rong R
• [Beignet] [PATCH 2/5] Backend: Optimization internal math, lower polynomials Grigore Lupescu
  • [Beignet] [PATCH 3/5] Backend: Optimization internal math, use native Grigore Lupescu
  • [Beignet] [PATCH 4/5] Backend: Optimization internal math, use mad Grigore Lupescu
  • Re: [Beignet] [PATCH 4/5] Backend: Optimization internal math, use mad Song, Ruiling
  • Re: [Beignet] [PATCH 4/5] Backend: Optimization internal math, use mad Yang, Rong R
  • Re: [Beignet] [PATCH 4/5] Backend: Optimization internal math, use mad Lupescu, Grigore
  • Re: [Beignet] [PATCH 4/5] Backend: Optimization internal math, use mad Yang, Rong R
  • Re: [Beignet] [PATCH 4/5] Backend: Optimization internal math, use mad Lupescu, Grigore
  • Re: [Beignet] [PATCH 4/5] Backend: Optimization internal math, use mad Song, Ruiling
  • Re: [Beignet] [PATCH 4/5] Backend: Optimization internal math, use mad Lupescu, Grigore
  • [Beignet] [PATCH 4/4] Backend: Optimization internal math, use mad Grigore Lupescu
  • Re: [Beignet] [PATCH 4/4] Backend: Optimization internal math, use mad Song, Ruiling
  • [Beignet] [PATCH 1/4] Benchmark: Evaluate math performance on intervals Grigore Lupescu
  • [Beignet] [PATCH 2/4] Backend: Optimization internal math, lower polynomials Grigore Lupescu
  • Re: [Beignet] [PATCH 2/4] Backend: Optimization internal math, lower polynomials Song, Ruiling
  • Re: [Beignet] [PATCH 2/4] Backend: Optimization internal math, lower polynomials Lupescu, Grigore
  • [Beignet] [PATCH 1/4] Benchmark: Evaluate math performance on intervals grigore . lupescu
  • [Beignet] [PATCH 2/4] Backend: Optimization internal math, lower polynomials grigore . lupescu
  • [Beignet] [PATCH 3/4] Backend: Optimization internal math, use native grigore . lupescu
  • [Beignet] [PATCH 4/4] Backend: Optimization internal math, use mad grigore . lupescu
  • Re: [Beignet] [PATCH 4/4] Backend: Optimization internal math, use mad Yang, Rong R
  • Re: [Beignet] [PATCH 2/4] Backend: Optimization internal math, lower polynomials Lupescu, Grigore
  • [Beignet] [PATCH 3/4] Backend: Optimization internal math, use native Grigore Lupescu
  • [Beignet] [PATCH 4/4] Backend: Optimization internal math, use mad Grigore Lupescu
  • [Beignet] [PATCH 5/5] Backend: Optimization internal math, experiment clc AMD Grigore Lupescu
• [Beignet] [PATCH 1/5] Benchmark: Evaluate math performance on intervals Grigore Lupescu
• [Beignet] [Patch V2 05/10] Android: fix __thread keyword issue in android. Yang Rong
• [Beignet] [PATCH 1/2] CMakeLists: respect existing CMAKE_C/CXX_FLAGS Ross Burton
  • [Beignet] [PATCH 2/2] Ensure paths to beignet.bc and beignet.pch include a / before the filename Ross Burton
  • Re: [Beignet] [PATCH 2/2] Ensure paths to beignet.bc and beignet.pch include a / before the filename Guo, Yejun
  • Re: [Beignet] [PATCH 2/2] Ensure paths to beignet.bc and beignet.pch include a / before the filename Yang, Rong R
• [Beignet] [PATCH 1/8] OCL20: add device enqueue helper functions in backend. Yang Rong
  • [Beignet] [PATCH 4/8] OCL20: handle device enqueue helper functions in the backend. Yang Rong
  • Re: [Beignet] [PATCH 4/8] OCL20: handle device enqueue helper functions in the backend. Pan, Xiuli
  • Re: [Beignet] [PATCH 4/8] OCL20: handle device enqueue helper functions in the backend. Song, Ruiling
  • [Beignet] [PATCH 2/8] OCL20: add device enqueue builtins. Yang Rong
  • Re: [Beignet] [PATCH 2/8] OCL20: add device enqueue builtins. Pan, Xiuli
  • [Beignet] [PATCH 3/8] OCL20: add ir register enqueuebufptr for enqueue global buffer. Yang Rong
  • [Beignet] [PATCH 7/8] OCL20: handle device enqueue in runtime. Yang Rong
  • [Beignet] [PATCH 8/8] OCL20: add device enqueue test case. Yang Rong
  • [Beignet] [PATCH 5/8] OCL20: Add runtime functions to get the device enqueue info. Yang Rong
  • [Beignet] [PATCH 6/8] OCL20: add a cl_kernel pointer to gpgpu. Yang Rong
• [Beignet] [PATCH 1/2] Backend: Add intel_sub_group_block_read/write form buffer Xiuli Pan
  • [Beignet] [PATCH 2/2] Utest: Add tset case for block read/write buffer Xiuli Pan
  • Re: [Beignet] [PATCH 1/2] Backend: Add intel_sub_group_block_read/write form buffer Andrew Lavin
• [Beignet] [PATCH 01/10] Android: change the saved file path. Yang Rong
  • [Beignet] [PATCH 04/10] Android: add android mk files. Yang Rong
  • [Beignet] [PATCH 02/10] Android: erase the stl in iterator loop, must update the iterator. Yang Rong
  • [Beignet] [PATCH 05/10] Android: fix __thread keyword issue in android. Yang Rong
  • [Beignet] [PATCH 09/10] Android: workaround libdrm flags. Yang Rong
  • [Beignet] [PATCH 08/10] Android: work around a LLVM 3.5 unrolling bug. Yang Rong
  • [Beignet] [PATCH 03/10] Android: derived the Allocator from std::allocator. Yang Rong
  • [Beignet] [PATCH 06/10] Android: disable pch in android. Yang Rong
  • [Beignet] [PATCH 10/10] Android: fix vector16 error in android. Yang Rong
  • [Beignet] [PATCH 07/10] Android: disable the read byte as DW. Yang Rong
  • Re: [Beignet] [PATCH 07/10] Android: disable the read byte as DW. Zhigang Gong
  • Re: [Beignet] [PATCH 07/10] Android: disable the read byte as DW. Yang, Rong R
• [Beignet] [PATCH] 3 op math functions dst need 16 byte align when allocate register. xionghu . luo
• [Beignet] Issue: Event callbacks not being called on event completion Андрей Глебов
  • Re: [Beignet] Issue: Event callbacks not being called on event completion Yang, Rong R
• [Beignet] [PATCH V2 5/5] Runtime: Fix memleak in build program for bin Xiuli Pan
• [Beignet] [PATCH 1/4] Backend: Refine returen value of sub_group_all/any to 1 Xiuli Pan
  • [Beignet] [PATCH 2/4] Utest: Remove old sub_group_all/any utest Xiuli Pan
  • [Beignet] [PATCH 4/4] Utest: Add test case for sub_group functions Xiuli Pan
  • [Beignet] [PATCH 3/4] Backend: Add sub_group built-in functions for intel extension Xiuli Pan
• [Beignet] cl_intel_subgroups support Andrew Lavin
  • Re: [Beignet] cl_intel_subgroups support Pan, Xiuli
  • Re: [Beignet] cl_intel_subgroups support Andrew Lavin
  • Re: [Beignet] cl_intel_subgroups support Pan, Xiuli
• [Beignet] utests difficulty on Halo GT4 Andrew Lavin
  • Re: [Beignet] utests difficulty on Halo GT4 Pan, Xiuli
  • Re: [Beignet] utests difficulty on Halo GT4 Andrew Lavin
• [Beignet] [PATCH V2 2/2] Utest: Remove some unsuport work group tests Xiuli Pan
• [Beignet] [PATCH 1/4] Runtime: Fix thread id calculation. Xiuli Pan
  • [Beignet] [PATCH 2/4] Runtime: Add API clGetKernelSubGroupInfoKHR for subgroup extension Xiuli Pan
  • [Beignet] [PATCH 4/4] Utest: Add subgroup work item test cases Xiuli Pan
  • [Beignet] [PATCH 3/4] Backend: Add subgroup work item builtin functions Xiuli Pan
• [Beignet] [PATCH 1/2] Backend: Copy workgroup emit function to gen8 Xiuli Pan
  • [Beignet] [PATCH 2/2] Utest: Remove some unsuport work group tests Xiuli Pan
• [Beignet] [PATCH 01/12] gbe/llvm: fix potential null pointer dereference. xionghu . luo
  • [Beignet] [PATCH 07/12] gbe/ir: defaultData is not initialized by default. xionghu . luo
  • [Beignet] [PATCH 10/12] gbe: ts array out of boundary. xionghu . luo
  • [Beignet] [PATCH 04/12] utest: error handling to avoid null pointer dereference. xionghu . luo
  • [Beignet] [PATCH 11/12] fix kernel build warnings. xionghu . luo
  • [Beignet] [PATCH 06/12] utest: init uninitialized local variables. xionghu . luo
  • [Beignet] [PATCH 09/12] runtime: thread_ids not initialized after created. xionghu . luo
  • [Beignet] [PATCH 02/12] no return value for non-void function. xionghu . luo
  • [Beignet] [PATCH 03/12] runtime: error handling to avoid null pointer dereference. xionghu . luo
  • [Beignet] [PATCH 08/12] gbe/ir: initialize the InstructionBase by a straightforward way. xionghu . luo
  • [Beignet] [PATCH 05/12] file name length overflow check. xionghu . luo
  • Re: [Beignet] [PATCH 01/12] gbe/llvm: fix potential null pointer dereference. Yang, Rong R
• [Beignet] [PATCH] add howto for stand alone utest. xionghu . luo
  • Re: [Beignet] [PATCH] add howto for stand alone utest. Yang, Rong R
• [Beignet] [PATCH 1/2] Benchmark: Fix benchmark bugs with image map Xiuli Pan
  • [Beignet] [PATCH 2/2] Benchmark: Fix Benchmark heap use after free problem Xiuli Pan
• [Beignet] [PATCH] Backend: Fix memleak in serialize_program Xiuli Pan
• [Beignet] [PATCH 1/5] Backend: Fix printfs mem leak Xiuli Pan
  • [Beignet] [PATCH 4/5] Runtime: Fix memleak of barrier evnets Xiuli Pan
  • [Beignet] [PATCH 2/5] Backend: Fix memleak form abi::__cxa_demangle Xiuli Pan
  • [Beignet] [PATCH 3/5] Utest: Fix utest memleaks Xiuli Pan
  • [Beignet] [PATCH 5/5] Runtime: Fix memleak in build program for bin Xiuli Pan
  • Re: [Beignet] [PATCH 1/5] Backend: Fix printfs mem leak Luo, Xionghu
  • Re: [Beignet] [PATCH 1/5] Backend: Fix printfs mem leak Pan, Xiuli
• [Beignet] [PATCH] Add support for gcc 6 Xiuli Pan
  • Re: [Beignet] [PATCH] Add support for gcc 6 Yang, Rong R
• [Beignet] [PATCH] Remove unncessary assertion in printf processing. yan . wang
  • Re: [Beignet] [PATCH] Remove unncessary assertion in printf processing. Pan, Xiuli
  • Re: [Beignet] [PATCH] Remove unncessary assertion in printf processing. Yang, Rong R
• [Beignet] [PATCH 2/3] Backend: Optimization use more native math Grigore Lupescu
  • [Beignet] [PATCH 3/3] Backend: Optimization of internal math functions Grigore Lupescu
  • Re: [Beignet] [PATCH 3/3] Backend: Optimization of internal math functions Song, Ruiling
  • Re: [Beignet] [PATCH 3/3] Backend: Optimization of internal math functions Lupescu, Grigore
• [Beignet] [PATCH 1/3] Benchmark: Evaluate math performance on intervals Grigore Lupescu
  • Re: [Beignet] [PATCH 1/3] Benchmark: Evaluate math performance on intervals Song, Ruiling
  • Re: [Beignet] [PATCH 1/3] Benchmark: Evaluate math performance on intervals Lupescu, Grigore
  • Re: [Beignet] [PATCH 1/3] Benchmark: Evaluate math performance on intervals Lupescu, Grigore
  • Re: [Beignet] [PATCH 1/3] Benchmark: Evaluate math performance on intervals Song, Ruiling
  • Re: [Beignet] [PATCH 1/3] Benchmark: Evaluate math performance on intervals Lupescu, Grigore
  • Re: [Beignet] [PATCH 1/3] Benchmark: Evaluate math performance on intervals Song, Ruiling
  • [Beignet] [PATCH 3/3] Backend: Optimization experiment by using AMD clc Grigore Lupescu
  • [Beignet] [PATCH 1/3] Benchmark: Evaluate math performance on intervals Grigore Lupescu
  • [Beignet] [PATCH 2/3] Backend: Optimization of internal math functions Grigore Lupescu
• [Beignet] [PATCH] Backend: Fix bug build with clang Xiuli Pan
  • Re: [Beignet] [PATCH] Backend: Fix bug build with clang Yang, Rong R
• [Beignet] [PATCH OCL20 V2 1/4] Backend: Chang scan limit for GVN pass 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 OCL20 V2 2/4] Backend: Refine script for math.h Xiuli Pan
  • Re: [Beignet] [PATCH OCL20 V2 1/4] Backend: Chang scan limit for GVN pass Yang, Rong R
• Re: [Beignet] drm/i915/bxt: Pooled EU support in beignet Yang, Rong R
• [Beignet] [PATCH] cl_api; Check image origin and region for NULL Jan Vesely
  • Re: [Beignet] [PATCH] cl_api; Check image origin and region for NULL Yang, Rong R
• [Beignet] [PATCH V3] GBE: Do more aggressive load/store merging. Ruiling Song
• [Beignet] [PATCH] only release cmrt device when it is already created Guo Yejun
  • Re: [Beignet] [PATCH] only release cmrt device when it is already created Yang, Rong R
• [Beignet] [PATCH V2 1/2] assert equation issue. xionghu . luo
  • [Beignet] [PATCH V2 2/2] write mask in disassembly not parse correctly. xionghu . luo
  • Re: [Beignet] [PATCH V2 2/2] write mask in disassembly not parse correctly. Guo, Yejun
  • Re: [Beignet] [PATCH V2 2/2] write mask in disassembly not parse correctly. Yang, Rong R
  • Re: [Beignet] [PATCH V2 1/2] assert equation issue. Guo, Yejun
• [Beignet] [PATCH] fix typo. xionghu . luo
  • Re: [Beignet] [PATCH] fix typo. Guo, Yejun
• [Beignet] [PATCH V2] GBE: try to avoid bank conflict in register allocator. Ruiling Song
  • Re: [Beignet] [PATCH V2] GBE: try to avoid bank conflict in register allocator. Matt Turner
• [Beignet] [PATCH V2 0/2] patch dependency. xionghu . luo
  • [Beignet] [PATCH V2 2/2] enable utest compiler_math_3op for mad test. xionghu . luo
  • Re: [Beignet] [PATCH V2 2/2] enable utest compiler_math_3op for mad test. Guo, Yejun
  • Re: [Beignet] [PATCH V2 2/2] enable utest compiler_math_3op for mad test. Yang, Rong R
  • [Beignet] [PATCH V2 1/2] handle simd8 and simd16 accrordingly for alu3. xionghu . luo
• [Beignet] [PATCH] runtime: we can accept large program global variable. Ruiling Song
• [Beignet] [PATCH] Backend: Chang scan limit for GVN pass Xiuli Pan
• [Beignet] [PATCH] GBE: try to avoid bank conflict in register allocator. Ruiling Song
  • Re: [Beignet] [PATCH] GBE: try to avoid bank conflict in register allocator. Luo, Xionghu
  • Re: [Beignet] [PATCH] GBE: try to avoid bank conflict in register allocator. Song, Ruiling
• [Beignet] [PATCH 1/2] handle simd8 and simd16 accrordingly for alu3. xionghu . luo
  • [Beignet] [PATCH 2/2] enable utest compiler_math_3op for mad test. xionghu . luo
  • Re: [Beignet] [PATCH 1/2] handle simd8 and simd16 accrordingly for alu3. Guo, Yejun
• [Beignet] [PATCH OCL20 V3] Backend: Refine workgroup all with SIMD_ALL algorithm 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] set SIMD width as 1 for mad when the dst is uniform Guo Yejun
  • Re: [Beignet] [PATCH] set SIMD width as 1 for mad when the dst is uniform Luo, Xionghu
• [Beignet] [PATCH OCL20] Backend: Refine script for math.h Xiuli Pan
• [Beignet] [PATCH OCL20 V2] Backend: Refine workgroup all with SIMD_ALL algorithm Xiuli Pan
• [Beignet] [PATCH] FindLLVM: allow LLVM/Clang 3.7 Rebecca N. Palmer
  • Re: [Beignet] [PATCH] FindLLVM: allow LLVM/Clang 3.7 Yang, Rong R
• [Beignet] [PATCH v4 3/5] standalone utest for unified OpenCL implementation. xionghu . luo
  • Re: [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. Ruiling Song
  • Re: [Beignet] [PATCH] GBE: Fix destination grf register type for cmp instruction. Pan, Xiuli
  • Re: [Beignet] [PATCH] GBE: Fix destination grf register type for cmp instruction. Yang, Rong R
• [Beignet] [PATCH 1/5] add sanity check for Image Region in runtime. xionghu . luo
  • [Beignet] [PATCH 4/5] fix failed cases for stand alone utest; xionghu . luo
  • [Beignet] [PATCH 2/5] buf[0] destroyed twice in event cases. xionghu . luo
  • [Beignet] [PATCH 5/5] add functions cl_check_beignet. xionghu . luo

• Earlier messages
• Later messages