[gcc r15-1726] vect: Fix shift-by-induction for single-lane slp

https://gcc.gnu.org/g:1ff5f8f8a05dd57620a1e2abbf87bd511b113cce commit r15-1726-g1ff5f8f8a05dd57620a1e2abbf87bd511b113cce Author: Feng Xue Date: Wed Jun 26 22:02:53 2024 +0800 vect: Fix shift-by-induction for single-lane slp Allow shift-by-induction for slp node, when it is single

[gcc r15-1727] vect: Determine input vectype for multiple lane-reducing operations

https://gcc.gnu.org/g:3aa004f1db327d5728a8fd0afcfed24e767f0499 commit r15-1727-g3aa004f1db327d5728a8fd0afcfed24e767f0499 Author: Feng Xue Date: Sun Jun 16 13:00:32 2024 +0800 vect: Determine input vectype for multiple lane-reducing operations The input vectype of reduction PHI sta

[gcc r15-2094] vect: Add a unified vect_get_num_copies for slp and non-slp

https://gcc.gnu.org/g:e7fbae834f8db2508d3161d88efe7ddbb702e437 commit r15-2094-ge7fbae834f8db2508d3161d88efe7ddbb702e437 Author: Feng Xue Date: Fri Jul 12 16:38:28 2024 +0800 vect: Add a unified vect_get_num_copies for slp and non-slp Extend original vect_get_num_copies (pure loop

[gcc r15-2095] vect: Refit lane-reducing to be normal operation

https://gcc.gnu.org/g:8b59fa9d8ca25bdf0792390a8bdeae151532a530 commit r15-2095-g8b59fa9d8ca25bdf0792390a8bdeae151532a530 Author: Feng Xue Date: Tue Jul 2 17:12:00 2024 +0800 vect: Refit lane-reducing to be normal operation Vector stmts number of an operation is calculated based on

[gcc r15-2096] vect: Support multiple lane-reducing operations for loop reduction [PR114440]

https://gcc.gnu.org/g:178cc419512f7e358f88dfe2336625aa99cd7438 commit r15-2096-g178cc419512f7e358f88dfe2336625aa99cd7438 Author: Feng Xue Date: Wed May 29 17:22:36 2024 +0800 vect: Support multiple lane-reducing operations for loop reduction [PR114440] For lane-reducing operation

[gcc r15-2097] vect: Optimize order of lane-reducing operations in loop def-use cycles

https://gcc.gnu.org/g:db3c8c9726d0bafbb9f85b6d7027fe83602643e7 commit r15-2097-gdb3c8c9726d0bafbb9f85b6d7027fe83602643e7 Author: Feng Xue Date: Wed May 29 17:28:14 2024 +0800 vect: Optimize order of lane-reducing operations in loop def-use cycles When transforming multiple lane-re

[gcc r15-2374] vect: Fix single_imm_use in tree_vect_patterns

https://gcc.gnu.org/g:49339d8b7e03a7ba0d4a5e118af993f175485b41 commit r15-2374-g49339d8b7e03a7ba0d4a5e118af993f175485b41 Author: Feng Xue Date: Fri Jun 14 15:49:23 2024 +0800 vect: Fix single_imm_use in tree_vect_patterns Since pattern statement coexists with normal statements in

[gcc r15-2719] vect: Fix dot-product slp testcases [PR116000]

https://gcc.gnu.org/g:8e2c9360c2df4b16582d3b9eb34e8c448798a1f3 commit r15-2719-g8e2c9360c2df4b16582d3b9eb34e8c448798a1f3 Author: Feng Xue Date: Mon Aug 5 18:13:55 2024 +0800 vect: Fix dot-product slp testcases [PR116000] These testcases were added by the patch of supporting multip

[gcc r15-2737] vect: Allow unsigned-to-signed promotion in vect_look_through_possible_promotion [PR115707]

https://gcc.gnu.org/g:3c089ee5d5a86cab0b27c69b96c4354c496520ac commit r15-2737-g3c089ee5d5a86cab0b27c69b96c4354c496520ac Author: Feng Xue Date: Mon Aug 5 15:23:56 2024 +0800 vect: Allow unsigned-to-signed promotion in vect_look_through_possible_promotion [PR115707] The function f

[gcc r15-2738] vect: Add missed opcodes in vect_get_smallest_scalar_type [PR115228]

https://gcc.gnu.org/g:95990db02b86282249396b06f65f4f9f582bab42 commit r15-2738-g95990db02b86282249396b06f65f4f9f582bab42 Author: Feng Xue Date: Mon Aug 5 15:53:19 2024 +0800 vect: Add missed opcodes in vect_get_smallest_scalar_type [PR115228] Some opcodes are missed when determini

[gcc r15-863] vect: Use vect representative statement instead of original in patch recog [PR115060]

https://gcc.gnu.org/g:a3aeff4ce95bd616a2108dc2363d9cbaba53b170 commit r15-863-ga3aeff4ce95bd616a2108dc2363d9cbaba53b170 Author: Feng Xue Date: Thu May 23 15:25:53 2024 +0800 vect: Use vect representative statement instead of original in patch recog [PR115060] Some utility functio

[gcc r15-903] vect: Unify bbs in loop_vec_info and bb_vec_info

https://gcc.gnu.org/g:9c747183efa555e45200523c162021e385511be5 commit r15-903-g9c747183efa555e45200523c162021e385511be5 Author: Feng Xue Date: Thu May 16 11:08:38 2024 +0800 vect: Unify bbs in loop_vec_info and bb_vec_info Both derived classes have their own "bbs" field, which hav

[gcc r15-904] Delete a file due to push error

https://gcc.gnu.org/g:b24b081113c696f4e523c8ae53fc3ab89c3b4e4d commit r15-904-gb24b081113c696f4e523c8ae53fc3ab89c3b4e4d Author: Feng Xue Date: Wed May 29 22:20:45 2024 +0800 Delete a file due to push error gcc/ * tree-vect-loop.c : Removed. Diff: --- gcc/tree-vect-lo

[gcc r15-961] vect: Add a function to check lane-reducing code

https://gcc.gnu.org/g:c0f31701556c4162463f28bc0f03007f40a6176e commit r15-961-gc0f31701556c4162463f28bc0f03007f40a6176e Author: Feng Xue Date: Wed May 29 13:12:12 2024 +0800 vect: Add a function to check lane-reducing code Check if an operation is lane-reducing requires comparison

[gcc r15-962] vect: Split out partial vect checking for reduction into a function

https://gcc.gnu.org/g:79c3547b8adfdfdb2a167c1b9c9428902510adab commit r15-962-g79c3547b8adfdfdb2a167c1b9c9428902510adab Author: Feng Xue Date: Wed May 29 13:45:09 2024 +0800 vect: Split out partial vect checking for reduction into a function Partial vectorization checking for vect

[gcc r15-963] vect: Bind input vectype to lane-reducing operation

https://gcc.gnu.org/g:d53f555edb95248dbf81347ba5e4136e9a491eca commit r15-963-gd53f555edb95248dbf81347ba5e4136e9a491eca Author: Feng Xue Date: Wed May 29 16:41:57 2024 +0800 vect: Bind input vectype to lane-reducing operation The input vectype is an attribute of lane-reducing oper

[gcc r15-1462] vect: Remove duplicated check on reduction operand

https://gcc.gnu.org/g:a944e57506fc64b8eede79c2405ba0b498461f0b commit r15-1462-ga944e57506fc64b8eede79c2405ba0b498461f0b Author: Feng Xue Date: Sun Jun 16 12:08:56 2024 +0800 vect: Remove duplicated check on reduction operand In vectorizable_reduction, one check on a reduction ope

[gcc r15-1461] vect: Add a function to check lane-reducing stmt

https://gcc.gnu.org/g:70466e6f9d9fb87f78ffe2e397ca876b380cb493 commit r15-1461-g70466e6f9d9fb87f78ffe2e397ca876b380cb493 Author: Feng Xue Date: Sat Jun 15 23:17:10 2024 +0800 vect: Add a function to check lane-reducing stmt Add a utility function to check if a statement is lane-re

[gcc r15-1463] vect: Use one reduction_type local variable

https://gcc.gnu.org/g:0726f1cde5459ccdbaa6af8c6904276a28d572ba commit r15-1463-g0726f1cde5459ccdbaa6af8c6904276a28d572ba Author: Feng Xue Date: Sun Jun 16 12:17:26 2024 +0800 vect: Use one reduction_type local variable Two local variables were defined to refer same STMT_VINFO_REDU

[gcc r15-1464] vect: Use an array to replace 3 relevant variables

https://gcc.gnu.org/g:b9c369d900ccfbd2271028611af3f08b5cf6f998 commit r15-1464-gb9c369d900ccfbd2271028611af3f08b5cf6f998 Author: Feng Xue Date: Sun Jun 16 13:21:13 2024 +0800 vect: Use an array to replace 3 relevant variables It's better to place 3 relevant independent variables i

[gcc r15-1465] vect: Tighten an assertion for lane-reducing in transform

https://gcc.gnu.org/g:ecbc96bb2873e453b0bd33d602ce34ad0d9d9cfd commit r15-1465-gecbc96bb2873e453b0bd33d602ce34ad0d9d9cfd Author: Feng Xue Date: Sun Jun 16 13:33:52 2024 +0800 vect: Tighten an assertion for lane-reducing in transform According to logic of code nearby the assertion,

[gcc r15-4294] vect: Fix inconsistency in fully-masked lane-reducing op generation [PR116985]

https://gcc.gnu.org/g:a9173a50e7e346a218323916e4d3add8552529ae commit r15-4294-ga9173a50e7e346a218323916e4d3add8552529ae Author: Feng Xue Date: Fri Oct 11 14:55:05 2024 +0800 vect: Fix inconsistency in fully-masked lane-reducing op generation [PR116985] To align vectorized def/us

[gcc r15-4283] MAINTAINERS: Add myself to write after approval

https://gcc.gnu.org/g:2a865ad31997a9c7ac6768f39ac87232dfda97d0 commit r15-4283-g2a865ad31997a9c7ac6768f39ac87232dfda97d0 Author: Feng Xue Date: Sat Oct 12 15:45:58 2024 +0800 MAINTAINERS: Add myself to write after approval ChangeLog: * MAINTAINERS: Add myself to w