projects / platform / upstream / glibc.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: a0bf289) | patch
S390: Allow "v" constraint for long double math_opt_barrier and math_force_eval with...
authorStefan Liebler <stli@linux.ibm.com>
Wed, 31 Mar 2021 14:17:01 +0000 (16:17 +0200)
committerStefan Liebler <stli@linux.ibm.com>
Thu, 1 Apr 2021 07:14:20 +0000 (09:14 +0200)
commit01e045117520200a1285ed112576c1be002a34d6
tree0ae1140eb60c8977aa385909fc660cf2603b9cebtree | snapshot
parenta0bf2897ce1afb30296e7b886e344dad0c788c0fcommit | diff
S390: Allow "v" constraint for long double math_opt_barrier and math_force_eval with GCC 11.

Starting with GCC 11, long double values can also be processed in vector
registers if build with -march >= z14.  Then GCC defines the
__LONG_DOUBLE_VX__ macro.

FYI: GCC commit "IBM Z: Introduce __LONG_DOUBLE_VX__ macro"
https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=f47df2af313d2ce7f9149149010a142c2237beda
sysdeps/s390/fpu/math-barriers.h diff | blob | history
Domain: System / Toolchain;