aboutsummaryrefslogtreecommitdiff
path: root/gcc/match.pd
AgeCommit message (Expand)AuthorFilesLines
2025-02-14tree-optimization/90579 - avoid STLF fail by better optimizingRichard Biener1-0/+56
2025-01-31[committed][PR tree-optimization/114277] Fix missed optimization for multipli...Jeff Law1-0/+17
2025-01-27match.pd: Canonicalize unsigned division by power of two into right shift [PR...Jakub Jelinek1-0/+9
2025-01-27match.pd: Fix indefinite recursion during exp-log transformations [PR118490]Soumya AR1-2/+2
2025-01-21match: Improve the `x ==/!= ~x` pattern [PR118483]Andrew Pinski1-2/+5
2025-01-17match.pd: Fix (FTYPE) N CMP (FTYPE) M optimization for GENERIC [PR118522]Jakub Jelinek1-1/+1
2025-01-15match: Simplify `1 >> x` into `x == 0` [PR102705]Andrew Pinski1-8/+8
2025-01-09match.pd: Avoid introducing UB in the a r<< (32-b) -> a r>> b optimization [P...Jakub Jelinek1-5/+23
2025-01-08Match: Update the comments for indicating SAT_* patternPan Li1-0/+6
2025-01-08Match: Refactor the signed SAT_* match for saturated value [NFC]Pan Li1-21/+17
2025-01-08Match: Refactor the signed SAT_TRUNC match patterns [NFC]Pan Li1-33/+32
2025-01-08Match: Refactor the signed SAT_SUB match patterns [NFC]Pan Li1-58/+40
2025-01-02Update copyright years.Jakub Jelinek1-1/+1
2024-12-21Match: Refactor the signed SAT_ADD match patterns [NFC]Pan Li1-82/+58
2024-12-14[PATCH v3] match.pd: Add pattern to simplify `(a - 1) & -a` to `0`Jovan Vukic1-0/+16
2024-12-12match.pd: Defer some CTZ/CLZ foldings until after ubsan pass for -fsanitize=b...Jakub Jelinek1-10/+33
2024-12-06More duplicates reported by genmatchRichard Biener1-8/+8
2024-12-06Remove some duplicates reported by genmatchRichard Biener1-5/+5
2024-12-05Match: Refactor the unsigned SAT_TRUNC match patterns [NFC]Pan Li1-60/+52
2024-12-03tree-ssanames, match.pd: get_nonzero_bits/with_*_nonzero_bits* cleanups and i...Jakub Jelinek1-16/+29
2024-12-03Match: Refactor the unsigned SAT_SUB match patterns [NFC]Pan Li1-114/+89
2024-11-27match: Improve handling of double convert [PR117776]Andrew Pinski1-2/+2
2024-11-27Match: Refactor the unsigned SAT_ADD match ADD_OVERFLOW pattern [NFC]Pan Li1-53/+30
2024-11-27match.pd: Avoid introducing UB in the ((X /[ex] C1) +- C2) * (C1 * C3) simpli...Jakub Jelinek1-1/+9
2024-11-22match.pd: Fix up the new simpliofiers using with_possible_nonzero_bits2 [PR11...Jakub Jelinek1-3/+3
2024-11-21Match: Refactor the unsigned SAT_ADD match pattern [NFC]Pan Li1-64/+45
2024-11-18match: Fix the `max<a,b>==0` pattern for pointers [PR117646]Andrew Pinski1-2/+4
2024-11-16match: Optimize `max(a,b) == 0` to `(a|b) == 0` for unsigned [PR115275]Andrew Pinski1-0/+8
2024-11-16MATCH: Simplify `min(a, b) op max(a, b)` to `a op b` [PR109401]Eikansh Gupta1-0/+7
2024-11-15match.pd: Fold vec_perm with view_convertJennifer Schmitz1-5/+8
2024-11-14MATCH: Simplify `a rrotate (32-b) -> a lrotate b` [PR109906]Eikansh Gupta1-0/+9
2024-11-13Match: Fold pow calls to ldexp when possible [PR57492]Soumya AR1-0/+25
2024-11-13Revert "Match: Simplify branch form 3 of unsigned SAT_ADD into branchless"Pan Li1-7/+4
2024-11-12Match: Optimize log (x) CMP CST and exp (x) CMP CST operationsSoumya AR1-1/+14
2024-11-07Match:Support signed imm SAT_ADD form1xuli1-0/+13
2024-11-05match: Fix comment for `X != 0 ? X + ~0 : 0` transformationAndrew Pinski1-1/+1
2024-11-03match: Fix `a != 0 ? a - 1 : 0` pattern [PR117363]Andrew Pinski1-2/+3
2024-10-30Match: Simplify (x != 0 ? x + ~0 : 0) to (x - x != 0).xuli1-0/+10
2024-10-25match.pd: Add std::pow folding optimizations.Jennifer Schmitz1-0/+28
2024-10-25Match: Simplify branch form 3 of unsigned SAT_ADD into branchlessPan Li1-4/+7
2024-10-25Default expand_vec_cond_expr_p code to ERROR_MARKRichard Biener1-10/+7
2024-10-25Restrict :c to commutative ops as intendedRichard Biener1-1/+1
2024-10-24Try to simplify (X >> C1) * (C2 << C1) -> X * C2Richard Sandiford1-0/+13
2024-10-24Handle POLY_INT_CSTs in get_nonzero_bitsRichard Sandiford1-0/+2
2024-10-24Try to simplify (X >> C1) << (C1 + C2) -> X << C2Richard Sandiford1-0/+13
2024-10-24Generalise ((X /[ex] A) +- B) * A -> X +- A * B ruleRichard Sandiford1-14/+24
2024-10-24Simplify (X /[ex] C1) * (C1 * C2) -> X * C2Richard Sandiford1-0/+8
2024-10-24Use get_nonzero_bits to simplify trunc_div to exact_divRichard Sandiford1-0/+9
2024-10-24Make more places handle exact_div like trunc_divRichard Sandiford1-29/+31
2024-10-22Match: Support IMM=1 for unsigned scalar .SAT_SUB IMM form 1xuli1-0/+7