[Bug tree-optimization/92645] Hand written vector code is 450 times slower when compiled with GCC compared to Clang

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92645

--- Comment #18 from Richard Biener <rguenth at gcc dot gnu.org> ---
Author: rguenth
Date: Tue Dec  3 11:59:13 2019
New Revision: 278934

URL: https://gcc.gnu.org/viewcvs?rev=278934&root=gcc&view=rev
Log:
2019-12-03  Richard Biener  <rguenther@suse.de>

        PR tree-optimization/92645
        * gimple-fold.c (gimple_fold_builtin_memory_op): Fold memcpy
        from or to a properly aligned register variable.

        * gcc.target/i386/pr92645-5.c: New testcase.

Added:
    trunk/gcc/testsuite/gcc.target/i386/pr92645-5.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/gimple-fold.c
    trunk/gcc/testsuite/ChangeLog