https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122456
Andrew Macleod <amacleod at redhat dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |hubicka at ucw dot cz,
| |mliska at suse dot cz
--- Comment #4 from Andrew Macleod <amacleod at redhat dot com> ---
(In reply to Sam James from comment #3)
> Can you try the same recipe, but `make profiledbootstrap` after configure?
Ugg. Confirmed to be that patch. Given the nature of the patch, I expect it
has uncovered some latent issue that a more precise range has caused an errant
optimization to happen. A number of values that use to be VARYING now have a
range.
This may take a while. Are there any LTO/bootstrapping tools to help figure
out which file has been miscompiled in which stage?
The link failure is in stage 3, but the problem likely dates back to a
miscompiled stgae 1 file in the stage 2 compiler which which then does
something very wrong in stage 3 :-P
