On Thu 07-03-19 12:38:15, Dave Hansen wrote:
> On 3/6/19 12:53 AM, Richard Biener wrote:
> > When running the gcc.target/i386/mpx/memmove-1.c testcase
> > from the GCC 8 branch on MPX capable hardware the testcase
> > faults and the kernel log reports the following:
> 
> While I don't doubt that we have some MPX bugs around, I wasn't able to
> reproduce this one with that binary.  Is there anything else that would
> help us track this down?

I have simply executed the binary without any special preparation and I
could see the leak information in dmesg

[112423.206497] BUG: Bad rss-counter state mm:000000007aa9c8a7 idx:1 val:25593

I do not see the fault/signal with
$ uname -r
5.0.0-00510-gcd2a3bf02625

though. My config is attached.
-- 
Michal Hocko
SUSE Labs