Re: [Xen-devel] [PATCH 0/4] xen: events: improve event channel IRQ allocation strategy.

[Ian Campbell <Ian.Campbell@xxxxxxxxxxxxx>]

On Tue, 2011-01-11 at 20:40 +0000, Konrad Rzeszutek Wilk wrote:
> On Tue, Jan 11, 2011 at 07:25:18PM +0000, Ian Campbell wrote:
> > On Tue, 2011-01-11 at 18:34 +0000, Konrad Rzeszutek Wilk wrote:
> > > 
> > > > The series has been tested as:
> > > >       * PV guest with a PCI passthrough device.
> > > 
> > > Which type of domain0? A 2.6.37 + Xen PCI backend + your patches?
> > > Or the older 2.6.32?
> > 
> > The 2.6.32 variant. Same for the HVM test. Dom 0 was 2.6.37 + patches.
> 
> Ok. Could you test the PV guest with a PCI passthrough device
> on a PV domain 0 that is 2.6.37 based? It _should_ work
> but it never hurts to try.
> 
> For your convience I've put up a merge tree with stable/* patches,
> devel/irq.rework (has mine and these patches you have posted),
> devel/xen-pciback, devel/gntdev (Stefano's last posting)
> 
> It is devel/next-2.6.37

I tested using this kernel as dom0 and my previous 2.6.37+patches as PV
domU and PCI passthrough seemed to work ok. In particular:
[    0.380253] uhci_hcd 0000:00:00.0: enabling device (0000 -> 0001)
[    0.380606] uhci_hcd 0000:00:00.0: Xen PCI mapped GSI20 to IRQ44
[    0.380669] uhci_hcd 0000:00:00.0: enabling bus mastering
[    0.380794] uhci_hcd 0000:00:00.0: setting latency timer to 64
[    0.380844] uhci_hcd 0000:00:00.0: UHCI Host Controller
[    0.381369] uhci_hcd 0000:00:00.0: new USB bus registered, assigned bus 
number 1

It didn't stop me testing but I got a bunch of spew in the form of ten
score of these earlish in the boot of devel/next-2.6.37 as dom0:

WARNING: at 
/local/scratch/ianc/devel/kernels/linux-2.6/arch/x86/xen/multicalls.c:182 
xen_mc_flush+0x293/0x2a0()
Hardware name: PowerEdge 860
Modules linked in:
Pid: 0, comm: swapper Tainted: G        W   2.6.37-x86_32p-xen0-00105-g4abcf5c 
#99
Call Trace:
 [<c1003c33>] ? xen_mc_flush+0x293/0x2a0
 [<c1003c33>] ? xen_mc_flush+0x293/0x2a0
 [<c103ef7c>] warn_slowpath_common+0x6c/0xa0
 [<c1003c33>] ? xen_mc_flush+0x293/0x2a0
 [<c103efcd>] warn_slowpath_null+0x1d/0x20
 [<c1003c33>] xen_mc_flush+0x293/0x2a0
 [<c1006597>] ? xen_set_domain_pte+0x57/0x100
 [<c10065df>] xen_set_domain_pte+0x9f/0x100
 [<c1003e00>] ? __raw_callee_save_xen_pte_val+0x0/0x8
 [<c1006716>] xen_set_pte+0x86/0x90
 [<c1003e00>] ? __raw_callee_save_xen_pte_val+0x0/0x8
 [<c1562e8b>] xen_set_pte_init+0x8a/0x96
 [<c1571825>] kernel_physical_mapping_init+0x2d3/0x3de
 [<c138117e>] init_memory_mapping+0x27e/0x4f0
 [<c156461e>] setup_arch+0x74d/0xcc4
 [<c139fa7f>] ? _raw_spin_unlock_irqrestore+0x3f/0x70
 [<c103fd2d>] ? vprintk+0x2ad/0x470
 [<c1069c28>] ? trace_hardirqs_off_caller+0xa8/0x140
 [<c1006990>] ? __raw_callee_save_xen_save_fl+0x0/0x8
 [<c1006998>] ? __raw_callee_save_xen_restore_fl+0x0/0x8
 [<c1069c28>] ? trace_hardirqs_off_caller+0xa8/0x140
 [<c1175712>] ? __raw_spin_lock_init+0x32/0x60
 [<c1002640>] ? xen_cpuid+0x0/0xa0
 [<c1002640>] ? xen_cpuid+0x0/0xa0
 [<c155e72c>] start_kernel+0x8b/0x381
 [<c155e0b3>] i386_start_kernel+0xa2/0xde
 [<c156175a>] xen_start_kernel+0x5fa/0x6b0
---[ end trace 4eaa2a86a8e4bf7c ]---

Full bootlog is attached.

Attachment: bootlog.gz
Description: GNU Zip compressed data

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