On Fri, Aug 23, 2019 at 02:57:35AM -0400, Jan Stancek wrote:
----- Original Message -----
On Thu, Aug 22, 2019 at 06:48:50PM -0400, CKI Project wrote:
>
> Hello,
>
> We ran automated tests on a patchset that was proposed for merging into
> this
> kernel tree. The patches were applied to:
>
> Kernel repo:
> https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git
> Commit: aad39e30fb9e - Linux 5.2.9
>
> The results of these automated tests are provided below.
>
> Overall result: FAILED (see details below)
> Merge: OK
> Compile: OK
> Tests: FAILED
>
> All kernel binaries, config files, and logs are available for download
> here:
>
> https://artifacts.cki-project.org/pipelines/116984
>
>
>
> One or more kernel tests failed:
>
> aarch64:
> ❌ LTP lite
> ❌ Loopdev Sanity
Odd, what suddenly broke? No new patches had been added to the queue
since Tuesday, except I removed a single patch. Removing a patch
shouldn't break anything, right?
It's a race:
[ 1289.578972] LTP: starting mtest06 ( mmap1)
...
[ 1455.794564] kernel BUG at mm/filemap.c:171!
Here's a verbose description:
https://lore.kernel.org/lkml/50b8914e20d1d62bb2dee42d342836c2c16ebee7.1563438048.git.jstancek@xxxxxxxxxx/
Final (less verbose) patch:
e1b98fa31664 ("locking/rwsem: Add missing ACQUIRE to read_slowpath exit when queue is empty")
Code review found also this issue, which is theoretical and very unlikely, but it's a small patch:
99143f82a255 ("lcoking/rwsem: Add missing ACQUIRE to read_slowpath sleep loop")
Okay, so this is not a new regression in 5.2 but rather something that
we just ended up hitting now.
We can queue up the fixes for the next release.
--
Thanks,
Sasha
