On 12/07/2017 22:52, Peter Wemm wrote:
> On Tuesday, July 11, 2017 11:55:47 PM Peter Wemm wrote:
>> On Tuesday, July 11, 2017 11:08:15 PM Peter Wemm wrote:
>>> On Wednesday, June 28, 2017 01:59:20 PM Andriy Gapon wrote:
A side bonus of this change is that now a vdev zio has a pointer
On Tuesday, July 11, 2017 11:55:47 PM Peter Wemm wrote:
> On Tuesday, July 11, 2017 11:08:15 PM Peter Wemm wrote:
> > On Wednesday, June 28, 2017 01:59:20 PM Andriy Gapon wrote:
> > > A side bonus of this change is that now a vdev zio has a pointer
> > > to its corresponding bio while the zio
On Tuesday, July 11, 2017 11:08:15 PM Peter Wemm wrote:
> On Wednesday, June 28, 2017 01:59:20 PM Andriy Gapon wrote:
> > A side bonus of this change is that now a vdev zio has a pointer
> > to its corresponding bio while the zio is active.
> fault virtual address = 0x28
>
> db> where
>
On Wednesday, June 28, 2017 01:59:20 PM Andriy Gapon wrote:
> A side bonus of this change is that now a vdev zio has a pointer
> to its corresponding bio while the zio is active.
> @@ -1094,6 +1088,15 @@ sendreq:
> static void
> vdev_geom_io_done(zio_t *zio)
> {
> + struct bio *bp =
Author: avg
Date: Wed Jun 28 13:59:20 2017
New Revision: 320452
URL: https://svnweb.freebsd.org/changeset/base/320452
Log:
fix an architectural problem introduced in r320156, ZFS ABD import
The implementation of ZFS refcount_t uses the emulated illumos mutex
(the sx lock) and the waiting