[gcc r16-3737] s390: Implement clz and ctz for SI mode

https://gcc.gnu.org/g:f50cff9766c5e42e1f5d20fe3e6c135601341663 commit r16-3737-gf50cff9766c5e42e1f5d20fe3e6c135601341663 Author: Juergen Christ Date: Thu Aug 28 15:48:31 2025 +0200 s390: Implement clz and ctz for SI mode To properly implement __builtin_ffs for SI mode, implement c

[gcc r16-3748] Fix load/store bias handling for extractlast.

https://gcc.gnu.org/g:64abb12aacbb64eb17e906edc24a788575f4103d commit r16-3748-g64abb12aacbb64eb17e906edc24a788575f4103d Author: Juergen Christ Date: Mon Sep 1 11:29:45 2025 +0200 Fix load/store bias handling for extractlast. The length returned by vect_get_loop_len is REALLEN + B

[gcc r15-10108] testsuite: Adjust s390x params for vector tests.

https://gcc.gnu.org/g:50c91681558e4029757077030aeaedfd4ef1e7a2 commit r15-10108-g50c91681558e4029757077030aeaedfd4ef1e7a2 Author: Juergen Christ Date: Tue Jul 29 16:23:24 2025 +0200 testsuite: Adjust s390x params for vector tests. Loop peeling and minimal loop vectorization thresh

[gcc r16-2616] testsuite: Adjust s390x params for vector tests.

https://gcc.gnu.org/g:9e91a99b2bb7d7f2724c638b38fff8b96c54ccd4 commit r16-2616-g9e91a99b2bb7d7f2724c638b38fff8b96c54ccd4 Author: Juergen Christ Date: Tue Jul 29 16:23:24 2025 +0200 testsuite: Adjust s390x params for vector tests. Loop peeling and minimal loop vectorization thresho

[gcc r16-2239] s390: Implement reduction optabs

https://gcc.gnu.org/g:e91b8e0449165be0b285be8ce6cdf1e32a2052ee commit r16-2239-ge91b8e0449165be0b285be8ce6cdf1e32a2052ee Author: Juergen Christ Date: Tue Jul 8 19:02:42 2025 +0200 s390: Implement reduction optabs Implementation and tests for the standard reduction optabs.

[gcc r16-2238] s390: Remove min-vect-loop-bound override

https://gcc.gnu.org/g:383ec62349c1645386c335904f5b51649ee1619d commit r16-2238-g383ec62349c1645386c335904f5b51649ee1619d Author: Juergen Christ Date: Tue Jul 8 18:59:26 2025 +0200 s390: Remove min-vect-loop-bound override The default setting of s390 for the parameter min-vect-loop

[gcc r16-2127] s390: Fix vector pattern tests for -m31.

https://gcc.gnu.org/g:d6161f9e0d39670afca5afff0a5c56c258f43200 commit r16-2127-gd6161f9e0d39670afca5afff0a5c56c258f43200 Author: Juergen Christ Date: Wed Jul 9 11:19:50 2025 +0200 s390: Fix vector pattern tests for -m31. Vectorization of int patterns requires 64bit long type (at l

[gcc r16-2090] s390: Split tests for 31bit support

https://gcc.gnu.org/g:32d41517c7276399e57b2b3f29e9790ae32d4883 commit r16-2090-g32d41517c7276399e57b2b3f29e9790ae32d4883 Author: Juergen Christ Date: Tue Jul 8 11:26:38 2025 +0200 s390: Split tests for 31bit support The new vector pattern tests used int128 without guard. This cau

[gcc r16-2068] s390: Optimize fmin/fmax.

https://gcc.gnu.org/g:c476f554e3f52086181d5c85701db34f6f390e3c commit r16-2068-gc476f554e3f52086181d5c85701db34f6f390e3c Author: Juergen Christ Date: Fri Jun 20 16:08:34 2025 +0200 s390: Optimize fmin/fmax. On VXE targets, we can directly use the fp min/max instruction instead of

[gcc r16-2055] s390: Add some missing vector patterns.

https://gcc.gnu.org/g:188acc9e8bacdbba56ed2b32d09f191da759500a commit r16-2055-g188acc9e8bacdbba56ed2b32d09f191da759500a Author: Juergen Christ Date: Mon Jun 23 12:04:20 2025 +0200 s390: Add some missing vector patterns. Some patterns that are detected by the autovectorizer can be

[gcc r16-1988] s390: More vec-perm-const cases.

https://gcc.gnu.org/g:680ee33e1dfc24345f0a881eb96cf4431db335a9 commit r16-1988-g680ee33e1dfc24345f0a881eb96cf4431db335a9 Author: Juergen Christ Date: Fri Jun 27 12:20:04 2025 +0200 s390: More vec-perm-const cases. s390 missed constant vector permutation cases based on the vector p

[gcc r16-1651] s390: Fix float vector extract for pre-z13

https://gcc.gnu.org/g:bd9cac12373aecaa0ebee808f805c617f2c15375 commit r16-1651-gbd9cac12373aecaa0ebee808f805c617f2c15375 Author: Juergen Christ Date: Wed Jun 18 15:16:28 2025 +0200 s390: Fix float vector extract for pre-z13 Also provide the vec_extract patterns for floats on pre-z

[gcc r16-907] s390x: Fix bootstrap.

https://gcc.gnu.org/g:6790558b4da2ffce55e507807b8f46e5dd5ad2b4 commit r16-907-g6790558b4da2ffce55e507807b8f46e5dd5ad2b4 Author: Juergen Christ Date: Tue May 27 15:42:10 2025 +0200 s390x: Fix bootstrap. A typo in the mnemonic attribute caused a failed bootstrap. Not sure how t

[gcc r16-895] s390: Floating point vector lane handling

https://gcc.gnu.org/g:e9e6a7c9f59c1a304f66ab81ead8f58838cf81b6 commit r16-895-ge9e6a7c9f59c1a304f66ab81ead8f58838cf81b6 Author: Juergen Christ Date: Thu May 8 14:11:00 2025 +0200 s390: Floating point vector lane handling Since floating point and vector registers overlap on s390, m

[gcc r15-7962] s390: fix delegitimization of addresses

https://gcc.gnu.org/g:4001281d69e9f8e01d2183f0e931e15ec4a522eb commit r15-7962-g4001281d69e9f8e01d2183f0e931e15ec4a522eb Author: Juergen Christ Date: Mon Mar 10 10:03:36 2025 +0100 s390: fix delegitimization of addresses In legitimize_pic_address we create a (const (unspec ...

[gcc r14-11076] s390: Fix UNSPEC_CC_TO_INT canonicalization

https://gcc.gnu.org/g:7b58a7e27afac52ca630d5db523503b75c732305 commit r14-11076-g7b58a7e27afac52ca630d5db523503b75c732305 Author: Juergen Christ Date: Mon Dec 9 15:26:54 2024 +0100 s390: Fix UNSPEC_CC_TO_INT canonicalization Canonicalization of comparisons for UNSPEC_CC_TO_INT mis

[gcc r15-6043] s390: Fix UNSPEC_CC_TO_INT canonicalization

https://gcc.gnu.org/g:bdc572f9a42b6a68dec1a5593d5311f45bd29cc9 commit r15-6043-gbdc572f9a42b6a68dec1a5593d5311f45bd29cc9 Author: Juergen Christ Date: Fri Dec 6 18:52:36 2024 +0100 s390: Fix UNSPEC_CC_TO_INT canonicalization Canonicalization of comparisons for UNSPEC_CC_TO_INT miss

[gcc r14-9877] s390x: Optimize vector permute with constant indexes

https://gcc.gnu.org/g:1f719aa7c0d35b47f3d15a54f074718b78e92bb0 commit r14-9877-g1f719aa7c0d35b47f3d15a54f074718b78e92bb0 Author: Juergen Christ Date: Mon Apr 1 13:34:37 2024 +0200 s390x: Optimize vector permute with constant indexes Loop vectorizer can generate vector permutes wit

[gcc r14-9468] s390: fix htm-builtins test cases

https://gcc.gnu.org/g:075104eef6d0ff215c394b6eaadcb525fa7c4975 commit r14-9468-g075104eef6d0ff215c394b6eaadcb525fa7c4975 Author: Juergen Christ Date: Wed Oct 25 14:57:03 2023 +0200 s390: fix htm-builtins test cases Transactional and non-transactional stores to the same cache line