On 2016-06-08, Christian Weisgerber wrote:
>> That's still testing server side for the contents of the ports tree,
>> isn't it? Not as heavily stressed as putting it on the server would
>> be, but it still gives it a bit of a workout.
>
> I have now put the patch on the central server in the pack
On 2016-06-08, Stuart Henderson wrote:
> That's still testing server side for the contents of the ports tree,
> isn't it? Not as heavily stressed as putting it on the server would
> be, but it still gives it a bit of a workout.
I have now put the patch on the central server in the package
buildi
On 2016/06/08 14:48, Christian Weisgerber wrote:
> On 2016-06-07, Theo de Raadt wrote:
>
> > Did I miss a report about nfs server and client? I know I have not tested
> > this diff.
>
> I put it on the amd64 package building machines and ran two bulk
> builds with it. That qualifies as a succe
On 2016-06-07, Theo de Raadt wrote:
> Did I miss a report about nfs server and client? I know I have not tested
> this diff.
I put it on the amd64 package building machines and ran two bulk
builds with it. That qualifies as a successful client test.
I haven't gotten around yet to testing the
> Martin Natano wrote:
> > It is time for the lockmgr() api to die. The api is only used by
> > filesystems, where it is a trivial change to use rrw locks instead. All
> > it needs is LK_* defines for the RW_* flags. (See the sys/lock.h hunk in
> > the diff below.)
> >
> > The ffs regress tests di
Martin Natano wrote:
> It is time for the lockmgr() api to die. The api is only used by
> filesystems, where it is a trivial change to use rrw locks instead. All
> it needs is LK_* defines for the RW_* flags. (See the sys/lock.h hunk in
> the diff below.)
>
> The ffs regress tests display the same
> Martin Natano wrote:
> > It is time for the lockmgr() api to die. The api is only used by
> > filesystems, where it is a trivial change to use rrw locks instead. All
> > it needs is LK_* defines for the RW_* flags. (See the sys/lock.h hunk in
> > the diff below.)
> >
> > The ffs regress tests di
Martin Natano wrote:
> It is time for the lockmgr() api to die. The api is only used by
> filesystems, where it is a trivial change to use rrw locks instead. All
> it needs is LK_* defines for the RW_* flags. (See the sys/lock.h hunk in
> the diff below.)
>
> The ffs regress tests display the same
It is time for the lockmgr() api to die. The api is only used by
filesystems, where it is a trivial change to use rrw locks instead. All
it needs is LK_* defines for the RW_* flags. (See the sys/lock.h hunk in
the diff below.)
The ffs regress tests display the same number of fail/ok results before
