RE: [PATCH] Adapt to new evtchn interface (RE:[Xen-ia64-devel]Changesto virq/irq interface in xen-unstable)

["Magenheimer, Dan (HP Labs Fort Collins)" <dan.magenheimer@xxxxxx>]

Sorry for the email overload on this topic...

I also successfully booted CONFIG_XEN_SMP with two processors
and ran a second domain.  It's not clear how to tell whether
each domain is running on a separate processor (xm vcpu-list
doesn't seem to work), but I have anecdotal evidence:

# xm list; sleep 10; xm list

shows each domain getting 5 seconds of time when booted
non-SMP and 10 seconds of time when booted SMP.

I haven't been able to successfully launch a third domain
however...


> -----Original Message-----
> From: xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx 
> [mailto:xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf 
> Of Magenheimer, Dan (HP Labs Fort Collins)
> Sent: Thursday, November 10, 2005 3:00 PM
> To: Tian, Kevin; Matt Chapman
> Cc: xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
> Subject: RE: [PATCH] Adapt to new evtchn interface 
> (RE:[Xen-ia64-devel]Changesto virq/irq interface in xen-unstable)
> 
> Rebuilding once again from a fresh clone, I am able
> to boot successfully to a login prompt.  So, I am
> going to request that Keir pull this into xen-unstable.
> 
> > -----Original Message-----
> > From: xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx 
> > [mailto:xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf 
> > Of Magenheimer, Dan (HP Labs Fort Collins)
> > Sent: Thursday, November 10, 2005 1:08 PM
> > To: Tian, Kevin; Matt Chapman
> > Cc: xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
> > Subject: RE: [PATCH] Adapt to new evtchn interface 
> > (RE:[Xen-ia64-devel]Changes to virq/irq interface in xen-unstable)
> > 
> > OK, I think I found the problem... blkfront no longer
> > gets config'd by default due to a recent change in
> > xen-unstable that allows blkfront to be a module.
> > 
> > When I fixed this, I was able to boot a second domain...
> > as far as the root disk getting mounted and kernel memory
> > being freed, but... I forgot to disable CONFIG_VT.
> > When I rebuilt with CONFIG_VT off, the second domain
> > failed with an error about no vbd backend!  So I need
> > more investigation but thought I would report progress
> > and I have committed the fix (to config in blkfront).
> > 
> > Dan
> > 
> > > -----Original Message-----
> > > From: xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx 
> > > [mailto:xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf 
> > > Of Magenheimer, Dan (HP Labs Fort Collins)
> > > Sent: Thursday, November 10, 2005 10:19 AM
> > > To: Tian, Kevin; Matt Chapman
> > > Cc: xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
> > > Subject: RE: [PATCH] Adapt to new evtchn interface (RE: 
> > > [Xen-ia64-devel]Changes to virq/irq interface in xen-unstable)
> > > 
> > > OK, I have merged this with the latest xen-unstable and
> > > updated xen-ia64-unstable with the merge.
> > > 
> > > I wasn't able to boot multiple domains with the result,
> > > but it was probably pilot error (I will start over from
> > > scratch and retry).  But it is also possible that another
> > > xen-unstable change in the last two days has broken
> > > xen/ia64 multiple domain support again. :-(
> > > 
> > > Could you test with xen-ia64-unstable tip?   If it works
> > > for you (or if I get it to work on my retry), I will ask
> > > Keir to pull.  If not, could you look through the xen-unstable
> > > csets since the last working version?
> > > 
> > > Thanks,
> > > Dan
> > > 
> > > > -----Original Message-----
> > > > From: Tian, Kevin [mailto:kevin.tian@xxxxxxxxx] 
> > > > Sent: Thursday, November 10, 2005 2:17 AM
> > > > To: Magenheimer, Dan (HP Labs Fort Collins); Matt Chapman
> > > > Cc: xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
> > > > Subject: [PATCH] Adapt to new evtchn interface (RE: 
> > > > [Xen-ia64-devel] Changes to virq/irq interface in xen-unstable)
> > > > 
> > > > Hi, Dan,
> > > >         Attached is the necessary change on ia64 side to adapt 
> > > > new evtchn interface from Keir. What I did is to pull Rev 
> > > > 7699 ("Rationalise the kernel event-channel binding 
> > > > interfaces. The") into xen-ia64-unstable.hg. (Before 
> > > > vcpu_translate patch).
> > > > 
> > > >         Now I've just confirmed domU and domVTI (with 
> > > > yesterday's patch) still working for me. And you can make a 
> > > try then.
> > > > 
> > > >         BTW, because the Rev 7699 from xen-unstable.hg also 
> > > > modifies evtchn_ia64.c which may conflict with Rev7664 from 
> > > > xen-ia64-unstable.hg (for notify_remote_via_irq). So I 
> > > > attached both patch and modified evtchn_ia64.c, in case it 
> > > > can't be applied cleanly.
> > > > 
> > > > Thanks,
> > > > Kevin
> > > > 
> > > > >-----Original Message-----
> > > > >From: Magenheimer, Dan (HP Labs Fort Collins) 
> > > > [mailto:dan.magenheimer@xxxxxx]
> > > > >Sent: 2005å11æ10æ 7:48
> > > > >To: Tian, Kevin; Matt Chapman
> > > > >Cc: xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
> > > > >Subject: RE: [Xen-ia64-devel] Changes to virq/irq interface 
> > > > in xen-unstable
> > > > >
> > > > >Kevin, Matt --
> > > > >
> > > > >I started to look at these changes but I am not as familiar
> > > > >with this code as you and, since multiple domains is just
> > > > >now working again, I am hesitant to mess with it.  Could
> > > > >one or both of you provide a patch to fix up ia64 to
> > > > >match Keir's interface change?
> > > > >
> > > > >This is relatively URGENT as I believe this is the only
> > > > >fix still needed to get ia64 multiple domains working in
> > > > >the xen-unstable tree.  (Hopefully, no recent or near-future
> > > > >changes in xen-unstable will break it before we are able
> > > > >to merge again.)  Once we have it working in xen-unstable,
> > > > >it will be much easier to find problems if something breaks.
> > > > >And it is very likely that the Xen 3.0 freeze is coming soon
> > > > >and it would be nice to have it in-tree before the freeze.
> > > > >
> > > > >You may want to clone the xen-unstable tree as a base and
> > > > >apply Kevin's recent patch (see xen-ia64-unstable cset 7664).
> > > > >This should work unless another recent xen-unstable cset
> > > > >has broken us again, which we would like to know now rather
> > > > >than later anyway.
> > > > >
> > > > >It appears we need to implement bind_virq_to_irqhandler
> > > > >and unbind_from_irqhandler.  I think bind_ipi_to_irqhandler is
> > > > >probably just a panic stub for now until we have SMP guest
> > > > >support.
> > > > >
> > > > >Thanks,
> > > > >Dan
> > > > >
> > > > >> -----Original Message-----
> > > > >> From: Tian, Kevin [mailto:kevin.tian@xxxxxxxxx]
> > > > >> Sent: Tuesday, November 08, 2005 7:35 PM
> > > > >> To: Magenheimer, Dan (HP Labs Fort Collins);
> > > > >> xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
> > > > >> Subject: RE: [Xen-ia64-devel] Changes to virq/irq interface
> > > > >> in xen-unstable
> > > > >>
> > > > >> Yes, not many changes required for ia64 and some simple stubs
> > > > >> are enough. Actually that's a good way to go. However
> > > > >> currently I'm still stuck on the environmental issue related
> > > > >> possibly to udev or hotplug, which preventing "xm create"
> > > > >> executed correctly. Hope other guys interested in this issue
> > > > >> can also make a try to see how quick we can solve it easily.
> > > > >> ;-) Once it's solved, we can adapt to Keir's changes and 
> > > then test.
> > > > >>
> > > > >> Thanks,
> > > > >> Kevin
> > > > >> >-----Original Message-----
> > > > >> >From: xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx
> > > > >> >[mailto:xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx] On
> > > > >> Behalf Of Magenheimer,
> > > > >> >Dan (HP Labs Fort Collins)
> > > > >> >Sent: 2005å11æ9æ 1:33
> > > > >> >To: xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
> > > > >> >Subject: [Xen-ia64-devel] Changes to virq/irq interface in
> > > > >> xen-unstable
> > > > >> >
> > > > >> >Kevin, would you take a look at this?  The changes are
> > > > >> >probably not hard, but until multiple domains is up on
> > > > >> >tip, I can't test and don't want to break your work.
> > > > >> >
> > > > >> >Keir's changes are in xen-unstable cset 7699.
> > > > >> >
> > > > >> >Dan
> > > > >> >
> > > > >> >> -----Original Message-----
> > > > >> >> From: Keir Fraser [mailto:Keir.Fraser@xxxxxxxxxxxx]
> > > > >> >> Sent: Tuesday, November 08, 2005 8:08 AM
> > > > >> >> To: Magenheimer, Dan (HP Labs Fort Collins)
> > > > >> >> Cc: xen-devel@xxxxxxxxxxxxxxxxxxx
> > > > >> >> Subject: Re: Please pull xen-ia64-unstable
> > > > >> >>
> > > > >> >>
> > > > >> >> On 7 Nov 2005, at 18:45, Magenheimer, Dan (HP Labs Fort
> > > > >> >> Collins) wrote:
> > > > >> >>
> > > > >> >> > Please pull
> > > > >> >> >
> > > > >> >> > http://xenbits.xensource.com/ext/xen-ia64-unstable.hg
> > > > >> >> >
> > > > >> >> > Merged up thru xen-unstable cset 7642.
> > > > >> >>
> > > > >> >> Okay, pulled through. I noticed that the event-channel
> > > > >> >> interfaces still
> > > > >> >> cause problems (bind_virq_to_irq in particular). I've had
> > > > >> >> another go at
> > > > >> >> rationalising the binding interface in light of this, so
> > > > >> you now need
> > > > >> >> to implement:
> > > > >> >>   bind_evtchn_to_irq --- same as it ever was
> > > > >> >>   bind_virq_to_irqhandler --- replaces bind_virq_to_irq
> > > > >> >>   bind_ipi_to_irqhandler --- replaces bind_ipi_to_irq
> > > > >> >>   unbind_from_irqhandler --- common unbind function 
> > > > that replaces
> > > > >> >> unbind_evtchn_from_irqhandler, unbind_virq_from_irq and
> > > > >> >> unbind_ipi_from_irq
> > > > >> >>
> > > > >> >> This interface eliminates all evtchn-related ia64 
> ifdefs in
> > > > >> >> common code.
> > > > >> >>
> > > > >> >>   -- Keir
> > > > >> >>
> > > > >> >
> > > > >> >
> > > > >> >> -----Original Message-----
> > > > >> >> From: Keir Fraser [mailto:Keir.Fraser@xxxxxxxxxxxx]
> > > > >> >> Sent: Tuesday, November 08, 2005 8:48 AM
> > > > >> >> To: Magenheimer, Dan (HP Labs Fort Collins)
> > > > >> >> Cc: xen-devel@xxxxxxxxxxxxxxxxxxx
> > > > >> >> Subject: Re: Please pull xen-ia64-unstable
> > > > >> >>
> > > > >> >>
> > > > >> >> On 8 Nov 2005, at 15:41, Magenheimer, Dan (HP Labs Fort
> > > > >> >> Collins) wrote:
> > > > >> >>
> > > > >> >> > Great, thanks!  Are these new interfaces checked into
> > > > >> >> > xen-unstable yet?  I can't find the changeset.
> > > > >> >>
> > > > >> >> They have to get through our automatic regression tests.
> > > > >> Assuming all
> > > > >> >> goes well the changes should be pushed in less 
> than an hour,
> > > > >> >>
> > > > >> >>   -- Keir
> > > > >> >>
> > > > >> >>
> > > > >> >
> > > > >> >_______________________________________________
> > > > >> >Xen-ia64-devel mailing list
> > > > >> >Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
> > > > >> >http://lists.xensource.com/xen-ia64-devel
> > > > >>
> > > > 
> > > 
> > 
> 

_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel