Message ID | CALdTtnu=0zQ3GoBUMWrW-EPngJG=tYaxzJeTYP6cSfCXsOXQWA@mail.gmail.com |
---|---|
State | New |
Headers | show |
On 04/05/2016 09:31 PM, Dann Frazier wrote:
> git://kernel.ubuntu.com/dannf/ubuntu-xenial.git lp1566518
Provenance ? These don't look well formed to me.
On Tue, Apr 5, 2016 at 2:51 PM, Tim Gardner <tim.gardner@canonical.com> wrote: > On 04/05/2016 09:31 PM, Dann Frazier wrote: >> git://kernel.ubuntu.com/dannf/ubuntu-xenial.git lp1566518 > > Provenance ? These don't look well formed to me. Whups - you're right, sorry. Both are straight cherry-picks from upstream, but I accidentally cherry-picked the first from my test build tree, so it has the wrong upstream hash. I've fixed and repushed. -dann
On 04/05/2016 10:33 PM, Dann Frazier wrote:
> git://kernel.ubuntu.com/dannf/ubuntu-xenial.git lp1566518
That is quite a leap to go from an ext4 BUG to some page table patches.
Where is the intervening discussion ? I assume you have positive test
results for both of our arm64 platforms (e.g., Snapdragon).
I've applied this time because these appear to be destined for stable,
but proper SRU justification will require a bit more information.
rtg
On Wed, Apr 6, 2016 at 2:46 AM, Tim Gardner <tim.gardner@canonical.com> wrote: > On 04/05/2016 10:33 PM, Dann Frazier wrote: >> git://kernel.ubuntu.com/dannf/ubuntu-xenial.git lp1566518 > > That is quite a leap to go from an ext4 BUG to some page table patches. Sorry, I didn't mean to imply it was an ext4 bug, just that the symptom is an ext4 BUG(), caused by incorrect pte settings. I've now expanded on the root cause in the description, but see the upstream thread for the complete story. > Where is the intervening discussion ? I added a link to the upstream thread in the bug comments. I think you'll find that there is a clear root cause to an issue that is 100% reproducible on this platform. > I assume you have positive test > results for both of our arm64 platforms (e.g., Snapdragon). I do [now :)]. I tested w/ a rebase of these patches on top of Paulo's pull dragonboard request, and still have the blinken lights :) > I've applied this time because these appear to be destined for stable, > but proper SRU justification will require a bit more information. Understood. Thanks! -dann > rtg > -- > Tim Gardner tim.gardner@canonical.com