[Bug rtl-optimization/85925] [7/8/9 regression] compilation of masking with 257 goes wrong in combine at -02

[ebotcazou at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85925

--- Comment #30 from Eric Botcazou <ebotcazou at gcc dot gnu.org> ---
> Don't have easy access to arm32 right now to verify if even with
> that change it still FAILs with the combine.c and rtlanal.c change reverted
> and passes with it and if it PASSes on big-endian targets.

Yes, it still fails on arm-eabi:

Running /home/eric/svn/gcc/gcc/testsuite/gcc.c-torture/execute/execute.exp ...
FAIL: gcc.c-torture/execute/20181120-1.c   -O2  execution test
FAIL: gcc.c-torture/execute/20181120-1.c   -O2 -flto -fno-use-linker-plugin
-flto-partition=none  execution test