aboutsummaryrefslogtreecommitdiff
path: root/gcc/tree-vect-slp.cc
AgeCommit message (Expand)AuthorFilesLines
2024-12-16vect: Do not try to duplicate_and_interleave one-element mode.Robin Dapp1-0/+3
2024-12-04tree-optimization/116083 - SLP discovery slownessRichard Biener1-28/+29
2024-12-03tree-optimization/117874 - optimize SLP discovery budget useRichard Biener1-10/+29
2024-12-02tree-optimization/116352 - SLP scheduling and stmt orderRichard Biener1-6/+23
2024-11-22build: Remove INCLUDE_MEMORY [PR117737]Andrew Pinski1-1/+0
2024-11-18vect: Add maskload else value support.Robin Dapp1-5/+17
2024-11-14tree-optimization/117567 - make SLP reassoc resilent against NULL lanesRichard Biener1-15/+40
2024-11-14tree-optimization/117559 - avoid hybrid SLP for masked load/store lanesRichard Biener1-7/+30
2024-11-12tree-optimization/116973 - SLP permute lower heuristic and single-lane SLPRichard Biener1-6/+11
2024-11-12tree-optimization/117484 - issue with SLP discovery of permuted .MASK_LOADRichard Biener1-9/+13
2024-11-11Add missing SLP discovery for CFN[_MASK][_LEN]_SCATTER_STORERichard Biener1-1/+16
2024-11-10Do not cost the permute node that are part of SLP load-lanesRichard Biener1-3/+5
2024-10-31middle-end: Lower all gconds during vector pattern matching [PR117176]Tamar Christina1-30/+9
2024-10-30Remove vectorizer finish_cost wrapperRichard Biener1-5/+6
2024-10-29tree-optimization/117343 - decide_masked_load_lanes and stale graphRichard Biener1-0/+4
2024-10-26Assert finished vectorizer pattern COND_EXPR transitionRichard Biener1-1/+1
2024-10-25tree-optimization/116575 - SLP masked load-lanes discoveryRichard Biener1-3/+138
2024-10-18Adjust assert in vect_build_slp_tree_2Richard Biener1-5/+1
2024-10-18middle-end: Fix GSI for gcond root [PR117140]Tamar Christina1-1/+1
2024-10-17Fix gcc.dg/vect/vect-early-break_39.c FAIL with forced SLPRichard Biener1-0/+1
2024-10-16tree-optimization/117050 - fix ICE with non-grouped .MASK_LOAD SLPRichard Biener1-9/+9
2024-10-16Remove SLP_INSTANCE_UNROLLING_FACTOR, compute VF in vect_make_slp_decisionRichard Biener1-17/+55
2024-10-15tree-optimization/117147 - bogus re-use of previous ldst_pRichard Biener1-5/+8
2024-10-14middle-end: support SLP early breakTamar Christina1-5/+111
2024-10-11tree-optimization/117080 - Add SLP_TREE_MEMORY_ACCESS_TYPERichard Biener1-0/+1
2024-10-11tree-optimization/117050 - fix ICE with non-grouped .MASK_LOAD SLPRichard Biener1-1/+2
2024-10-11Fix possible wrong-code with masked store-lanesRichard Biener1-10/+20
2024-10-10vect: Avoid divide by zero for permutes of extern VLA vectorsRichard Sandiford1-3/+12
2024-10-10tree-optimization/117060 - fix oversight in vect_build_slp_tree_1Richard Biener1-2/+3
2024-10-10Allow SLP store of mixed external and constantRichard Biener1-17/+7
2024-10-09tree-optimization/116575 - handle SLP of permuted masked loadsRichard Biener1-4/+54
2024-10-09Fix memory leak in vect_cse_slp_nodesRichard Biener1-1/+1
2024-10-07vect: Add more dump messages for VLA SLP permutation [PR116583]Richard Sandiford1-2/+14
2024-10-07vect: Support more VLA SLP permutations [PR116583]Richard Sandiford1-27/+80
2024-10-07vect: Restructure repeating_p case for SLP permutations [PR116583]Richard Sandiford1-34/+41
2024-10-07vect: Variable lane indices in vectorizable_slp_permutation_1 [PR116583]Richard Sandiford1-7/+9
2024-10-06Add single-lane SLP support to .GOMP_SIMD_LANE vectorizationRichard Biener1-0/+11
2024-10-05Improve load permutation loweringRichard Biener1-14/+17
2024-09-24build: enable C++11 narrowing warningsJason Merrill1-2/+2
2024-09-24tree-optimization/115372 - failed store-lanes in some casesRichard Biener1-0/+18
2024-09-23tree-optimization/116810 - out-of-bound access to matches[]Richard Biener1-1/+1
2024-09-23middle-end: Insert invariant instructions before the gsi [PR116812]Tamar Christina1-4/+2
2024-09-22middle-end: lower COND_EXPR into gimple form in vect_recog_bool_patternTamar Christina1-0/+14
2024-09-20Handle patterns as SLP roots of only live stmtsRichard Biener1-0/+1
2024-09-19Fall back to single-lane SLP before falling back to no SLPRichard Biener1-18/+25
2024-09-12Abort loop SLP analysis quickerRichard Biener1-1/+9
2024-09-12Better recover from SLP reassociation fails during discoveryRichard Biener1-15/+14
2024-09-10tree-optimization/116658 - latent issue in vect_is_slp_load_nodeRichard Biener1-3/+4
2024-09-06Fix SLP double-reduction supportRichard Biener1-1/+3
2024-09-06Handle non-grouped stores as single-lane SLPRichard Biener1-9/+36
generated by cgit v1.1 at 2025-01-12 09:28:26 +0000