summaryrefslogtreecommitdiff
path: root/kernel
AgeCommit message (Expand)AuthorFilesLines
2022-07-28LoongArch64: Add DYNAMIC_ARCH supportgxw4-28/+57
2022-07-25Merge pull request #3691 from martin-frbg/issue3679-sparcMartin Kroeker1-0/+8
2022-07-24Fix DNRM2 returning INF instead of zero due to intermediate overflowMartin Kroeker1-4/+29
2022-07-19Merge pull request #3690 from RajalakshmiSR/cdotp10Martin Kroeker2-0/+2
2022-07-19Merge pull request #3689 from RajalakshmiSR/dgemvgcc10Martin Kroeker1-0/+3
2022-07-19fix DNRM2 returning INF instead of zero due to intermediate overflowMartin Kroeker1-0/+8
2022-07-18POWER: Fix complex dot function failuresRajalakshmi Srinivasaraghavan2-0/+2
2022-07-18POWER10: dgemv builtin renameRajalakshmi Srinivasaraghavan1-0/+3
2022-07-15LoongArch64: Fix dnrm2_tiny testcase failuregxw1-0/+10
2022-07-11MIPS64: Fix dnrm2_tiny testcase failuregxw1-0/+9
2022-07-05Eliminate uses of CREAL on left-hand side of assignmentsMartin Kroeker1-9/+6
2022-07-02workaround fault with ssq=inf,scale=0Martin Kroeker1-0/+1
2022-06-29Neoverse N2 sbgemm:Honglin Zhu4-385/+749
2022-06-29format codeHonglin Zhu3-50/+60
2022-06-29neoverse n2 sbgemm:Honglin Zhu5-35/+481
2022-06-29neoverse n2 sbgemm: init fileHonglin Zhu5-0/+194
2022-06-28Merge pull request #3669 from VFerrari/fix_small_matrix_kernelMartin Kroeker1-0/+4
2022-06-28Merge pull request #3642 from nursik/developMartin Kroeker1-0/+173
2022-06-25POWER10: Fix multithreading check when USE_THREAD=0VFerrari1-0/+4
2022-06-18Merge pull request #3655 from RajalakshmiSR/zgemmasmp10Martin Kroeker1-26/+26
2022-06-17POWER10: Fix ZGEMM testcase failuresRajalakshmi Srinivasaraghavan1-26/+26
2022-06-09POWER10: convert dgemv inline assemblyRajalakshmi Srinivasaraghavan1-320/+65
2022-06-06Merge branch 'develop' into risc-vXianyi Zhang31-319/+1242
2022-06-06Update RISC-V Intrinsic API.Xianyi Zhang29-245/+312
2022-06-02Fix MSVC ARM64 build. Add generic kernel for ARM64Nursultan Zarlyk1-0/+173
2022-05-20Revert "roll back DGEMM kernel ... for DYNAMIC_ARCH"Martin Kroeker1-6/+1
2022-05-12POWER10: Changing store instructions for Level1 functionsRajalakshmi Srinivasaraghavan16-274/+541
2022-04-30Fix generator rules for ?laswp_ncopy and ?neg_tcopyMartin Kroeker1-15/+15
2022-04-16fix undefined prefetchsizesMartin Kroeker1-0/+5
2022-04-16fix undefined prefetchsizeMartin Kroeker1-0/+4
2022-03-28CortexX1 is ARMV8 like A7xMartin Kroeker1-216/+1
2022-03-27Add initial support for Phytium FT2000 series and ARMV9 Cortex 510/710/X1/X2Martin Kroeker5-0/+867
2022-03-23Remove extraneous (and wrong) definition of sbgemm_r on x86_64Martin Kroeker1-1/+0
2022-03-11fix unsafe read of Y in assembly kernelCaroline Newcombe1-15/+16
2022-02-28Merge branch 'develop' into risc-vXianyi Zhang367-2272/+83937
2022-02-28Small Matrix: use proper inline asm input constraint for AVX512 maskWangyang Guo4-8/+8
2022-02-25really fix definition of SHUFFLE_MAGIC_NOMartin Kroeker1-4/+2
2022-02-25Remove stray $Martin Kroeker1-1/+1
2022-02-25Declare SHUFFLE_MAGIC_NO as const to placate clangMartin Kroeker1-1/+1
2022-02-25Define sbgemm_r to fix DYNAMIC_ARCH buildsMartin Kroeker1-0/+7
2022-02-24Merge pull request #3542 from martin-frbg/issue3540Martin Kroeker2-7/+7
2022-02-23Fix compilation of Skylake AVX512 kernels with GCC 6Mosè Giordano4-4/+4
2022-02-23Prevent compiler attempts to use k0 as mask registerMartin Kroeker1-6/+6
2022-02-23Fix non-portable u_int64_tMartin Kroeker1-1/+1
2022-02-23Guard uses of _mm512_reduce_add_p?Martin Kroeker4-0/+20
2022-02-06Merge pull request #3493 from martin-frbg/casts+cleanupMartin Kroeker8-8/+8
2022-01-27Add proper defaults for IMIN/IMAXMartin Kroeker1-2/+10
2022-01-22Add default KERNEL file for Elbrus E2K archMartin Kroeker1-0/+149
2022-01-22Create MakefileMartin Kroeker1-0/+1
2022-01-22Add Elbrus e2k architecture supportMartin Kroeker1-0/+4
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-22 01:39:23 +0000