Show patches with: Submitter = Jiahao Xu       |    State = Action Required       |    Archived = No       |   44 patches
Patch Series A/F/R/T S/W/F Date Submitter Delegate State
LoongArch: Fix incorrect return type for frecipe/frsqrte intrinsic functions LoongArch: Fix incorrect return type for frecipe/frsqrte intrinsic functions - - - - --- 2024-01-24 Jiahao Xu New
Loongarch: Remove vec_concatz<mode> pattern Loongarch: Remove vec_concatz<mode> pattern - - - - --- 2024-01-24 Jiahao Xu New
[v3] LoongArch: Define LOGICAL_OP_NON_SHORT_CIRCUIT [v3] LoongArch: Define LOGICAL_OP_NON_SHORT_CIRCUIT - - - - --- 2024-01-16 Jiahao Xu New
LoongArch: Fix pattern vec_concatz<mode> LoongArch: Fix pattern vec_concatz<mode> - - - - --- 2024-01-16 Jiahao Xu New
LoongArch: Split vec_selects of bottom elements into simple move LoongArch: Split vec_selects of bottom elements into simple move - - - - --- 2024-01-16 Jiahao Xu New
LoongArch: Implenment vec_init<M><N> where N is a LSX vector mode LoongArch: Implenment vec_init<M><N> where N is a LSX vector mode - - - - --- 2024-01-05 Jiahao Xu New
LoongArch: Optimize zero_extendqisi2 and zero_extendqidi2 patterns LoongArch: Optimize zero_extendqisi2 and zero_extendqidi2 patterns - - - - --- 2024-01-05 Jiahao Xu New
LoongArch: Improve lasx_xvpermi_q_<LASX:mode> insn pattern LoongArch: Improve lasx_xvpermi_q_<LASX:mode> insn pattern - - - - --- 2024-01-05 Jiahao Xu New
[v2] LoongArch: Fix incorrect code generation for sad pattern [v2] LoongArch: Fix incorrect code generation for sad pattern - - - - --- 2023-12-14 Jiahao Xu New
LoongArch: Fix incorrect code generation for sad pattern LoongArch: Fix incorrect code generation for sad pattern - - - - --- 2023-12-14 Jiahao Xu New
[v2] LoongArch: Define LOGICAL_OP_NON_SHORT_CIRCUIT. [v2] LoongArch: Define LOGICAL_OP_NON_SHORT_CIRCUIT. - - - - --- 2023-12-12 Jiahao Xu New
LoongArch: Define LOGICAL_OP_NON_SHORT_CIRCUIT. LoongArch: Define LOGICAL_OP_NON_SHORT_CIRCUIT. - - - - --- 2023-12-12 Jiahao Xu New
[v3,5/5] LoongArch: Vectorized loop unrolling is disable for divf/sqrtf/rsqrtf when -mrecip is enab… Add support for approximate instructions and optimize divf/sqrtf/rsqrtf operations. - - - - --- 2023-12-06 Jiahao Xu New
[v3,4/5] LoongArch: New options -mrecip and -mrecip= with ffast-math. Add support for approximate instructions and optimize divf/sqrtf/rsqrtf operations. - - - - --- 2023-12-06 Jiahao Xu New
[v3,3/5] LoongArch: Redefine pattern for xvfrecip/vfrecip instructions. Add support for approximate instructions and optimize divf/sqrtf/rsqrtf operations. - - - - --- 2023-12-06 Jiahao Xu New
[v3,2/5] LoongArch: Use standard pattern name for xvfrsqrt/vfrsqrt instructions. Add support for approximate instructions and optimize divf/sqrtf/rsqrtf operations. - - - - --- 2023-12-06 Jiahao Xu New
[v3,1/5] LoongArch: Add support for LoongArch V1.1 approximate instructions. Add support for approximate instructions and optimize divf/sqrtf/rsqrtf operations. - - - - --- 2023-12-06 Jiahao Xu New
[v2,5/5] LoongArch: Vectorized loop unrolling is disable for divf/sqrtf/rsqrtf when -mrecip is enab… Add support for approximate instructions and optimize divf/sqrtf/rsqrt operations. - - - - --- 2023-12-05 Jiahao Xu New
[v2,4/5] LoongArch: New options -mrecip and -mrecip= with ffast-math. Add support for approximate instructions and optimize divf/sqrtf/rsqrt operations. - - - - --- 2023-12-05 Jiahao Xu New
[v2,3/5] LoongArch: Redefine pattern for xvfrecip/vfrecip instructions. Add support for approximate instructions and optimize divf/sqrtf/rsqrt operations. - - - - --- 2023-12-05 Jiahao Xu New
[v2,2/5] LoongArch: Use standard pattern name for xvfrsqrt/vfrsqrt instructions. Add support for approximate instructions and optimize divf/sqrtf/rsqrt operations. - - - - --- 2023-12-05 Jiahao Xu New
[v2,1/5] LoongArch: Add support for LoongArch V1.1 approximate instructions. Add support for approximate instructions and optimize divf/sqrtf/rsqrt operations. - - - - --- 2023-12-05 Jiahao Xu New
LoongArch: Fix ICE and use simplify_gen_subreg instead of gen_rtx_SUBREG directly. LoongArch: Fix ICE and use simplify_gen_subreg instead of gen_rtx_SUBREG directly. - - - - --- 2023-11-29 Jiahao Xu New
LoongArch: Fix lsx-vshuf.c and lasx-xvshuf_b.c tests fail on LA664 [PR112611] LoongArch: Fix lsx-vshuf.c and lasx-xvshuf_b.c tests fail on LA664 [PR112611] - - - - --- 2023-11-29 Jiahao Xu New
[5/5] LoongArch: Vectorized loop unrolling is not performed on divf/sqrtf/rsqrtf with turns on -mre… LoongArch: Add -mrecip option support - - - - --- 2023-11-28 Jiahao Xu New
[4/5] LoongArch: New options -mrecip and -mrecip= with ffast-math. LoongArch: Add -mrecip option support - - - - --- 2023-11-28 Jiahao Xu New
[3/5] LoongArch: Redefine pattern for xvfrecip/vfrecip instructions. LoongArch: Add -mrecip option support - - - - --- 2023-11-28 Jiahao Xu New
[2/5] LoongArch: Use standard pattern name for xvfrsqrt/vfrsqrt instructions. LoongArch: Add -mrecip option support - - - - --- 2023-11-28 Jiahao Xu New
[1/5] LoongArch: Add support for approximate instructions. LoongArch: Add -mrecip option support - - - - --- 2023-11-28 Jiahao Xu New
LoongArch: Add support for xorsign. LoongArch: Add support for xorsign. - - - - --- 2023-11-17 Jiahao Xu New
LoongArch: Add support for xorsign. LoongArch: Add support for xorsign. - - - - --- 2023-11-17 Jiahao Xu New
LoongArch: Add support for xorsign. LoongArch: Add support for xorsign. - - - - --- 2023-11-17 Jiahao Xu New
LoongArch: Fix scan-assembler-times of lasx/lsx test case. LoongArch: Fix scan-assembler-times of lasx/lsx test case. - - - - --- 2023-11-16 Jiahao Xu New
LoongArch: Fix scan-assembler-times of lasx/lsx test case. LoongArch: Fix scan-assembler-times of lasx/lsx test case. - - - - --- 2023-11-16 Jiahao Xu New
LoongArch: Increase cost of vector aligned store/load. LoongArch: Increase cost of vector aligned store/load. - - - - --- 2023-11-16 Jiahao Xu New
[1/2] LoongArch: Increase cost of vector aligned store/load. [1/2] LoongArch: Increase cost of vector aligned store/load. - - - - --- 2023-11-16 Jiahao Xu New
LoongArch: Increase cost of vector aligned store/load. LoongArch: Increase cost of vector aligned store/load. - - - - --- 2023-11-16 Jiahao Xu New
LoongArch:Enable vcond_mask_mn expanders for SF/DF modes. LoongArch:Enable vcond_mask_mn expanders for SF/DF modes. - - - - --- 2023-10-23 Jiahao Xu New
LoongArch:Enable vcond_mask_mn expanders for SF/DF modes. LoongArch:Enable vcond_mask_mn expanders for SF/DF modes. - - - - --- 2023-10-23 Jiahao Xu New
LoongArch:Enable vcond_mask_mn expanders for SF/DF modes. LoongArch:Enable vcond_mask_mn expanders for SF/DF modes. - - - - --- 2023-10-23 Jiahao Xu New
LoongArch:Enable vcond_mask_mn expanders for SF/DF modes. LoongArch:Enable vcond_mask_mn expanders for SF/DF modes. - - - - --- 2023-10-23 Jiahao Xu New
[3/3] LoongArch:Implement the new vector cost model framework. Optimize loongarch vector implementation. - - - - --- 2023-10-16 Jiahao Xu New
[2/3] LoongArch:Implement vec_widen standard names. Optimize loongarch vector implementation. - - - - --- 2023-10-16 Jiahao Xu New
[1/3] LoongArch:Implement avg and sad standard names. Optimize loongarch vector implementation. - - - - --- 2023-10-16 Jiahao Xu New