Messages in this thread | | | Date | Thu, 23 Feb 2012 15:59:09 -0800 | From | "H. Peter Anvin" <> | Subject | Re: [PATCH 0/5] i387: stable kernel backport |
| |
On 02/23/2012 03:54 PM, Greg Kroah-Hartman wrote: > On Thu, Feb 23, 2012 at 03:18:03PM -0800, H. Peter Anvin wrote: >> On 02/23/2012 03:16 PM, Greg Kroah-Hartman wrote: >>> >>> And a "clean" 3.2.7 doesn't have that problem? >>> >>> Does the same thing happen in Linus's tree at the moment? >>> >> >> Clean 3.2.7 I don't think has that problem, although of course when I >> run the stress test it fails on other ways so it's hard to 100% rule out. >> >> Linus' tree doesn't have that problems. > > Ugh. > > So, we need more testers, so let me release a -rc kernel with just these > changes in it, to get a wider testing range, and make an easy "this > kernel worked, but this one didn't" for people to work with. > > I don't suppose that 'git bisect' would work for this, as it's really a > "all or nothing" type thing with this series from what I can tell, > right? >
Yes. At Suresh' suggestion I am running a reference run without AES-NI, just to make sure we're not chasing a wireless driver bug.
-hpa
-- H. Peter Anvin, Intel Open Source Technology Center I work for Intel. I don't speak on their behalf.
| |