https://gcc.gnu.org/bugzilla/show_bug.cgi?id=124165

--- Comment #11 from Richard Biener <rguenth at gcc dot gnu.org> ---
(In reply to H.J. Lu from comment #10)
> Created attachment 63749 [details]
> Try this

Unoptimized with -fno-checking this gets us from

 thread pro- & epilogue             :  99.49 ( 86%)   236k (  0%)
 TOTAL                              : 115.04          439M
114.31user 0.34system 1:55.07elapsed 99%CPU (0avgtext+0avgdata
698488maxresident)k
0inputs+14720outputs (0major+143076minor)pagefaults 0swaps

to

 thread pro- & epilogue             :   0.12 (  1%)   236k (  0%)
 TOTAL                              :  13.33          439M
13.12user 0.22system 0:13.36elapsed 99%CPU (0avgtext+0avgdata
698988maxresident)k
0inputs+14816outputs (0major+127571minor)pagefaults 0swaps

Reply via email to