https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86007
Andrew Pinski changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86007
--- Comment #5 from Andrew Pinski ---
(In reply to Ștefan Talpalaru from comment #4)
> Mystery solved: half the CPU cores report LWP as disabled, the other half as
> enabled.
That has to be a kernel issue and not a GCC issue
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86007
--- Comment #4 from Ștefan Talpalaru ---
Mystery solved: half the CPU cores report LWP as disabled, the other half as
enabled.
Both at boot, as seen with "grep -E '(core id|lwp)' /proc/cpuinfo" and
afterwards, using the CPUID instruction:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86007
--- Comment #3 from Ștefan Talpalaru ---
The problem persists in 8.2.0 and in current trunk. You need to run that
command a hundred of times or so, on a Piledriver, to replicate it:
for i in `seq 1 100`; do gcc -S t.c -march=native -v 2>&1 |
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86007
Richard Biener changed:
What|Removed |Added
Target||x86_64-*-*, i?86-*-*