[Bug rtl-optimization/102178] [12 Regression] SPECFP 2006 470.lbm regressions on AMD Zen CPUs after r12-897-gde56f95afaaa22

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102178

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|2021-09-03 00:00:00         |2022-4-11

--- Comment #31 from Richard Biener <rguenth at gcc dot gnu.org> ---
For the testcase I still see 58 vmovq GPR <-> XMM at -Ofast -march=znver2,
resulting from spilling of xmms.  And there's still cases like

        movq    .LC0(%rip), %rax
...
        vmovq   %rax, %xmm2
        vmovq   %rax, %xmm4

Honza - you said you wanted to have some look here.  As I understand Vlad
costing isn't taken into account when choosing alternatives for reloading.