aboutsummaryrefslogtreecommitdiff
path: root/gcc/config/aarch64
AgeCommit message (Expand)AuthorFilesLines
2022-02-09[COMMITTED] Fix PR aarch64/104474: ICE with vector float initializers and non...Andrew Pinski1-1/+1
2022-02-09aarch64: Extend vec_concat patterns to 8-byte vectorsRichard Sandiford3-42/+84
2022-02-09aarch64: Remove move_lo/hi_quad expandersRichard Sandiford1-93/+18
2022-02-09aarch64: Add a general vec_concat expanderRichard Sandiford3-76/+57
2022-02-09aarch64: Add more vec_combine patternsRichard Sandiford2-0/+66
2022-02-09aarch64: Remove redundant vec_concat patternsRichard Sandiford1-35/+2
2022-02-09aarch64: Generalise adjacency check for load_pair_lanesRichard Sandiford3-24/+38
2022-02-09aarch64: Generalise vec_set predicateRichard Sandiford1-1/+1
2022-02-09aarch64: Tighten general_operand predicatesRichard Sandiford1-3/+3
2022-02-03aarch64: Add missing movmisalign patternsRichard Sandiford1-2/+2
2022-02-03aarch64: Remove VALL_F16MOV iteratorRichard Sandiford2-7/+2
2022-02-02AArch64: use canonical ordering for complex mul, fma and fmsTamar Christina2-10/+10
2022-01-23[aarch64/64821]: Simplify __builtin_aarch64_sqrt* into internal function .SQRT.Andrew Pinski1-0/+7
2022-01-18Fix -Wformat-diag in various targets.Martin Liska1-1/+1
2022-01-18aarch64: Fix overly optimistic LDP/STP matching [PR104005]Richard Sandiford1-0/+1
2022-01-17Change references of .c files to .cc filesMartin Liska10-31/+31
2022-01-17Rename .c files to .cc files.Martin Liska9-0/+0
2022-01-12Fix -Wformat-diag for aarch64 target.Martin Liska1-13/+13
2022-01-03Update copyright years.Jakub Jelinek75-75/+75
2021-12-30c-family: Use BULTINS_LOCATION for predefined macros changed upon optimize or...Jakub Jelinek1-0/+2
2021-12-16aarch64: fix: ls64 tests fail on aarch64_be [PR103729]Przemyslaw Wirkus1-3/+2
2021-12-15aarch64: Don't classify vector pairs as short vectors [PR103094]Richard Sandiford1-2/+17
2021-12-14aarch64: Add LS64 extension and intrinsicsPrzemyslaw Wirkus7-0/+277
2021-12-13aarch64: Add command-line support for Armv8.8-aKyrylo Tkachov2-0/+6
2021-12-13aarch64: Use +mops to inline memset operationsKyrylo Tkachov4-25/+96
2021-12-13aarch64: Add memmove expansion for +mopsKyrylo Tkachov2-0/+51
2021-12-13aarch64: Add support for Armv8.8-a memory operations and memcpy expansionKyrylo Tkachov5-17/+89
2021-12-13pch: Small cleanupJakub Jelinek1-1/+1
2021-12-09pch: Fix aarch64 build [PR71934]Jakub Jelinek1-1/+1
2021-12-09Leverage VX_CPU_PREFIX in aarch64-vxworks.hOlivier Hainque1-1/+2
2021-12-02AArch64: Optimize right shift rounding narrowingTamar Christina3-20/+58
2021-11-30vect: Add support for fmax and fmin reductionsRichard Sandiford3-2/+28
2021-11-17aarch64: Add new vector mode V8DIPrzemyslaw Wirkus3-1/+10
2021-11-17Add IFN_COND_FMIN/FMAX functionsRichard Sandiford1-1/+18
2021-11-12aarch64: Remove redundant costing codeRichard Sandiford1-112/+30
2021-11-12aarch64: Use new hooks for vector comparisonsRichard Sandiford1-146/+145
2021-11-12aarch64: Add vf_factor to aarch64_vec_op_countRichard Sandiford1-6/+24
2021-11-12aarch64: Move cycle estimation into aarch64_vec_op_countRichard Sandiford1-98/+105
2021-11-12aarch64: Use an array of aarch64_vec_op_countsRichard Sandiford1-55/+60
2021-11-12aarch64: Use real scalar op countsRichard Sandiford1-94/+88
2021-11-12aarch64: Get floatness from stmt_infoRichard Sandiford1-2/+12
2021-11-12aarch64: Remove vectype from latency testsRichard Sandiford1-20/+13
2021-11-12aarch64: Fold aarch64_sve_op_count into aarch64_vec_op_countRichard Sandiford1-57/+96
2021-11-12aarch64: Detect more consecutive MEMsRichard Sandiford1-52/+104
2021-11-11aarch64: Use type-qualified builtins for vcombine_* Neon intrinsicsJonathan Wright4-30/+21
2021-11-11aarch64: Use type-qualified builtins for LD1/ST1 Neon intrinsicsJonathan Wright4-79/+115
2021-11-11aarch64: Use type-qualified builtins for ADDV Neon intrinsicsJonathan Wright2-7/+8
2021-11-11aarch64: Use type-qualified builtins for ADDP Neon intrinsicsJonathan Wright2-15/+10
2021-11-11aarch64: Use type-qualified builtins for [R]SUBHN[2] Neon intrinsicsJonathan Wright2-42/+28
2021-11-11aarch64: Use type-qualified builtins for [R]ADDHN[2] Neon intrinsicsJonathan Wright2-42/+28
generated by cgit v1.1 at 2024-08-18 04:26:58 +0000