Patchwork GNU Compiler Collection

login
register
mail settings

Incoming patches

« Previous 1 2 3 4 5 6 7 8 9 10 ... 366 367 Next »
Filters: State = Action Required remove filter    |    Archived = No remove filter
Patch A/R/T Date Submitter Delegate State
Correct cost model for strided loads 0 0 0 2012-06-10 William J. Schmidt New
Hoist adjacent loads 0 0 0 2012-06-11 William J. Schmidt New
[RFC] First cut at using vec_construct for strided loads 0 0 0 2012-06-13 William J. Schmidt New
[committed] Fix PR53647 0 0 0 2012-06-13 William J. Schmidt New
Some vector cost model cleanup 0 0 0 2012-06-13 William J. Schmidt New
Fix PR53703 0 0 0 2012-06-18 William J. Schmidt New
Strength reduction preliminaries 0 0 0 2012-06-21 William J. Schmidt New
Strength reduction 0 0 0 2012-06-25 William J. Schmidt New
Fix PR46556 (straight-line strength reduction, part 2) 0 0 0 2012-06-28 William J. Schmidt New
[RFC] New target interface for vectorizer cost model 0 0 0 2012-07-03 William J. Schmidt New
[RFC] New target interface for vectorizer cost model 0 0 0 2012-07-03 William J. Schmidt New
[RFC] New target interface for vectorizer cost model 0 0 0 2012-07-06 William J. Schmidt New
[committed] Fix PR53955 0 0 0 2012-07-13 William J. Schmidt New
Enable vectorizer cost model by default at -O3 0 0 0 2012-07-15 William J. Schmidt New
Add flag to control straight-line strength reduction 0 0 0 2012-07-17 William J. Schmidt New
Add flag to control straight-line strength reduction 0 0 0 2012-07-18 William J. Schmidt New
Vectorizer cost model outside-cost changes 0 0 0 2012-07-24 William J. Schmidt New
Change IVOPTS and strength reduction to use expmed cost model 0 0 0 2012-07-25 William J. Schmidt New
Change IVOPTS and strength reduction to use expmed cost model 0 0 0 2012-07-25 William J. Schmidt New
Fix PR53733 0 0 0 2012-07-30 William J. Schmidt New
[rs6000] Vectorizer heuristic 0 0 0 2012-07-31 William J. Schmidt New
[committed] Strength reduction clean-up (base name => base expr) 0 0 0 2012-08-01 William J. Schmidt New
Strength reduction part 3 of 4: candidates with unknown strides 0 0 0 2012-08-01 William J. Schmidt New
[committed] Fix PR53773 0 0 0 2012-08-03 William J. Schmidt New
Fix PR54211 0 0 0 2012-08-10 William J. Schmidt New
[testsuite] New effective target long_neq_int 0 0 0 2012-08-10 William J. Schmidt New
Fix PR54240 0 0 0 2012-08-14 William J. Schmidt New
Fix PR54245 0 0 0 2012-08-14 William J. Schmidt New
Straight-line strength reduction, 4/4: Conditional candidates 0 0 0 2012-08-23 William J. Schmidt New
Fix PR54492 0 0 0 2012-09-10 William J. Schmidt New
Fix PR54492 0 0 0 2012-09-10 William J. Schmidt New
Fix PR54674 0 0 0 2012-09-24 William J. Schmidt New
[committed] Fix PR55008 0 0 0 2012-10-22 William J. Schmidt New
Fix PR56321 0 0 0 2013-02-15 William J. Schmidt New
Fix PR56605 0 0 0 2013-03-13 William J. Schmidt New
SLSR for conditional candidates 0 0 0 2013-03-29 William J. Schmidt New
[PowerPC] Fix PR 56843 0 0 0 2013-04-05 William J. Schmidt New
[committed] Avoid bootstrap failure for i686-pc-linux-gnu due to SLSR 0 0 0 2013-05-04 William J. Schmidt New
Fix thinko in SLSR that caused x86 bootstrap failure 0 0 0 2013-05-05 William J. Schmidt New
Use anonymous SSA name support in SLSR 0 0 0 2013-05-06 William J. Schmidt New
Fix SLSR conditional candidate bug with commuted operands 0 0 0 2013-05-06 William J. Schmidt New
Limit SLSR increment vector size to avoid quadratic behavior 0 0 0 2013-05-07 William J. Schmidt New
Restrict conditional SLSR candidates to CAND_MULTs 0 0 0 2013-05-07 William J. Schmidt New
[committed] Correct signedness warning in gimple-ssa-strength-reduction.c 0 0 0 2013-05-08 William J. Schmidt New
[rs6000] Increase MALLOC_ABI_ALIGNMENT for 32-bit PowerPC 0 0 0 2013-05-17 William J. Schmidt New
Fix PR57441 (memory error in SLSR) 0 0 0 2013-05-29 William J. Schmidt New
[PowerPC] rs6000_expand_vector_init selects wrong field for splat in LE mode 0 0 0 2013-07-22 William J. Schmidt New
[PowerPC] altivec_expand_vec_perm_const reverses pack pattern arguments in little endian mode 0 0 0 2013-07-22 William J. Schmidt New
[PowerPC] altivec_expand_vec_perm_const reverses pack pattern arguments in little endian mode 0 0 0 2013-07-23 William J. Schmidt New
[PowerPC] altivec_expand_vec_perm_const selects wrong field for splat in LE mode 0 0 0 2013-07-23 William J. Schmidt New
[PowerPC] Fix unaligned Altivec load/stores in LE mode 0 0 0 2013-07-24 William J. Schmidt New
[PowerPC] Fix vector pack insns for little endian 0 0 0 2013-07-24 William J. Schmidt New
Fix PR57993 (ICE in slsr) 0 0 0 2013-07-28 William J. Schmidt New
[committed] Update SLSR candidate table for update_stmt replacements 0 0 0 2013-08-03 William J. Schmidt New
[PowerPC] Fix PR57949 (ABI alignment issue) 0 0 0 2013-08-14 William J. Schmidt New
[PowerPC] Change code generation for VSX loads and stores for little endian 0 0 0 2013-10-01 William J. Schmidt New
[rs6000] Correct vector permute for little endian 0 0 0 2013-10-06 William J. Schmidt New
[rs6000] Fix variable permute control vectors for little endian 0 0 0 2013-10-09 William J. Schmidt New
[rs6000] Handle missing permute splits for V2DF/V4SF in little endian 0 0 0 2013-10-11 William J. Schmidt New
[rs6000] Fix vsx_concat_<mode> insns for little endian 0 0 0 2013-10-16 William J. Schmidt New
[rs6000] Fix vsx_unpack expansions for endianness 0 0 0 2013-10-16 William J. Schmidt New
[rs6000] Adjust vec_unpacku patterns for little endian 0 0 0 2013-10-20 William J. Schmidt New
[rs6000] Be careful with special permute masks for little endian 0 0 0 2013-10-21 William J. Schmidt New
[rs6000] Be careful with special permute masks for little endian 0 0 0 2013-10-21 William J. Schmidt New
[rs6000] Be careful with special permute masks for little endian, take 2 0 0 0 2013-10-22 William J. Schmidt New
[rs6000] Fix mulv8hi3 pattern for little endian 0 0 0 2013-10-23 William J. Schmidt New
[testsuite/rs6000] Adjust two VMX tests for little endian 0 0 0 2013-10-28 William J. Schmidt New
[rs6000] Correct handling of multiply high-part for little endian 0 0 0 2013-10-30 William J. Schmidt New
[rs6000] Fix rs6000_expand_vector_set for little endian 0 0 0 2013-11-01 William J. Schmidt New
[rs6000] Re-permute source register for postreload splits of VSX LE stores 0 0 0 2013-11-02 William J. Schmidt New
[rs6000] (1/3) Reverse meanings of multiply even/odd for little endian 0 0 0 2013-11-04 William J. Schmidt New
[rs6000] (2/3) Fix widening multiply high/low operations for little endian 0 0 0 2013-11-04 William J. Schmidt New
[rs6000] (3/3) Fix mulv4si3 and mulv8hi3 patterns for little endian 0 0 0 2013-11-04 William J. Schmidt New
[rs6000] Fix vect_pack_trunc_v2df pattern for little endian 0 0 0 2013-11-04 William J. Schmidt New
[rs6000] (1/3) Reverse meanings of multiply even/odd for little endian 0 0 0 2013-11-05 William J. Schmidt New
[rs6000] (3/3) Fix mulv4si3 and mulv8hi3 patterns for little endian 0 0 0 2013-11-05 William J. Schmidt New
[rs6000] Fix vect_pack_trunc_v2df pattern for little endian 0 0 0 2013-11-05 William J. Schmidt New
[rs6000] Enable VSX code generation in little endian mode 0 0 0 2013-11-05 William J. Schmidt New
[rs6000] Correct programmer access to vperm for little endian 0 0 0 2013-11-15 William J. Schmidt New
[rs6000] Emit correct note for DWARF CFI information on LE prolog VSX stores 0 0 0 2013-11-17 William J. Schmidt New
[rs6000] Fix libcpp/lex.c Altivec code to be correct for little endian 0 0 0 2013-11-18 William J. Schmidt New
[rs6000] Fix V16QI vector splat for little endian 0 0 0 2013-11-19 William J. Schmidt New
[rs6000] Correct vsx_set and vsx_extract patterns for little endian 0 0 0 2013-11-20 William J. Schmidt New
[rs6000] More vector LE cleanups 0 0 0 2013-11-22 William J. Schmidt New
[rs6000] Generate correct constant permutes using xxpermdi 0 0 0 2013-11-22 William J. Schmidt New
[rs6000] Skip test gfortran.dg/nan_7.f90 for little endian PowerPC 0 0 0 2013-11-27 William J. Schmidt New
[rs6000] Skip another test case for little endian 0 0 0 2013-12-02 William J. Schmidt New
[rs6000] Add -maltivec={le,be} options 0 0 0 2014-01-07 William J. Schmidt New
[rs6000,committed] Remove duplicates from altivec_overloaded_builtins 0 0 0 2014-01-08 William J. Schmidt New
[rs6000] Add -maltivec={le,be} options 0 0 0 2014-01-09 William J. Schmidt New
[rs6000,committed] Fix pasto in rs6000-builtin.def 0 0 0 2014-01-10 William J. Schmidt New
[rs6000] Implement -maltivec=be for vec_insert and vec_extract Altivec intrinsics 0 0 0 2014-01-13 William J. Schmidt New
[rs6000] Implement -maltivec=be for vec_mule and vec_mulo Altivec intrinsics 0 0 0 2014-01-13 William J. Schmidt New
[rs6000] Add tests for vec_insert and vec_extract implemented with VSX instructions 0 0 0 2014-01-21 William J. Schmidt New
[rs6000] Avoid optimization problem for VEC_PERM_EXPR 0 0 0 2014-01-23 William J. Schmidt New
[rs6000] Implement -maltivec=be for vec_mergeh and vec_mergel Altivec builtins 0 0 0 2014-01-24 William J. Schmidt New
[rs6000] Implement -maltivec=be for vec_mergeh and vec_mergel Altivec builtins 0 0 0 2014-01-29 William J. Schmidt New
[rs6000] Implement -maltivec=be for vec_splat builtins 0 0 0 2014-01-30 William J. Schmidt New
[rs6000] Clean up mergeh/mergel patterns to avoid missed optimizations 0 0 0 2014-01-30 William J. Schmidt New
[rs6000] Handle -maltivec=be on little endian for vec_sums 0 0 0 2014-01-31 William J. Schmidt New