Hello,
On Xen, when booting a guest with a system disk and an additional swap
disk I'm getting a calltrace.
xen hypervisor: 4.1.2; linux dom0: v3.3.6; linux guest: v3.2.17
When booting without a swap disk, I don't have the issue.
I also tested a guest with v3.3.6: same problem. But from v3.4-rc2,
the issue is fixed.
I cherry-picked:
052b198 swap: don't do discard if no discard option added
Applied and tested on top of v3.2.17 and v3.3.6, it fixes the issue.
Pid: 0, comm: swapper/0 Not tainted 3.2.17-x86_64 #12
Call Trace:
<IRQ>
[<ffffffff810919da>] ? handle_irq_event_percpu+0x3a/0x140
[<ffffffff81091b29>] ? handle_irq_event+0x49/0x80
[<ffffffff81094e7d>] ? handle_edge_irq+0x6d/0x120
[<ffffffff81229088>] ? __xen_evtchn_do_upcall+0x1b8/0x280
[<ffffffff8122a442>] ? xen_evtchn_do_upcall+0x22/0x40
[<ffffffff8133f4fe>] ? xen_do_hypervisor_callback+0x1e/0x30
<EOI>
[<ffffffff810013aa>] ? hypercall_page+0x3aa/0x1000
[<ffffffff810013aa>] ? hypercall_page+0x3aa/0x1000
[<ffffffff8100768c>] ? xen_safe_halt+0xc/0x20
[<ffffffff81013563>] ? default_idle+0x23/0x40
[<ffffffff8100b073>] ? cpu_idle+0x63/0xb0
[<ffffffff81654c43>] ? start_kernel+0x362/0x36d
[<ffffffff81657491>] ? xen_start_kernel+0x558/0x55e
Code: 39 ed 0f 84 1c 02 00 00 44 8b 7b 48 4c 8b 73 50 41 83 ef 01 41
21 ef 49 6b c7 70 4d 8b 64 06 40 49 69 c4 d0 00 00 00 48 8d 14 03 <48>
8b 8a 78 02 00 00 48 89 4c 24 10 80 ba 09 02 00 00 00 74 6d
RIP [<ffffffff8125ed66>] blkif_interrupt+0x66/0x320
RSP <ffff88001fc03e18>
---[ end trace dfd4e5623eb06620 ]---
Regards,
--
William
On Mon, May 21, 2012 at 12:30:45AM +0200, William Dauchy wrote:
> Hello,
>
> On Xen, when booting a guest with a system disk and an additional swap
> disk I'm getting a calltrace.
> xen hypervisor: 4.1.2; linux dom0: v3.3.6; linux guest: v3.2.17
> When booting without a swap disk, I don't have the issue.
> I also tested a guest with v3.3.6: same problem. But from v3.4-rc2,
> the issue is fixed.
> I cherry-picked:
> 052b198 swap: don't do discard if no discard option added
So you are asking for 052b198 to be back-ported.
I am OK with that but I think Shaohua needs to Ack that and
ask Greg to put it on [email protected]
> Applied and tested on top of v3.2.17 and v3.3.6, it fixes the issue.
>
> Pid: 0, comm: swapper/0 Not tainted 3.2.17-x86_64 #12
> Call Trace:
> <IRQ>
> [<ffffffff810919da>] ? handle_irq_event_percpu+0x3a/0x140
> [<ffffffff81091b29>] ? handle_irq_event+0x49/0x80
> [<ffffffff81094e7d>] ? handle_edge_irq+0x6d/0x120
> [<ffffffff81229088>] ? __xen_evtchn_do_upcall+0x1b8/0x280
> [<ffffffff8122a442>] ? xen_evtchn_do_upcall+0x22/0x40
> [<ffffffff8133f4fe>] ? xen_do_hypervisor_callback+0x1e/0x30
> <EOI>
> [<ffffffff810013aa>] ? hypercall_page+0x3aa/0x1000
> [<ffffffff810013aa>] ? hypercall_page+0x3aa/0x1000
> [<ffffffff8100768c>] ? xen_safe_halt+0xc/0x20
> [<ffffffff81013563>] ? default_idle+0x23/0x40
> [<ffffffff8100b073>] ? cpu_idle+0x63/0xb0
> [<ffffffff81654c43>] ? start_kernel+0x362/0x36d
> [<ffffffff81657491>] ? xen_start_kernel+0x558/0x55e
> Code: 39 ed 0f 84 1c 02 00 00 44 8b 7b 48 4c 8b 73 50 41 83 ef 01 41
> 21 ef 49 6b c7 70 4d 8b 64 06 40 49 69 c4 d0 00 00 00 48 8d 14 03 <48>
> 8b 8a 78 02 00 00 48 89 4c 24 10 80 ba 09 02 00 00 00 74 6d
> RIP [<ffffffff8125ed66>] blkif_interrupt+0x66/0x320
> RSP <ffff88001fc03e18>
> ---[ end trace dfd4e5623eb06620 ]---
>
> Regards,
> --
> William
>
> _______________________________________________
> Xen-devel mailing list
> [email protected]
> http://lists.xen.org/xen-devel
Hello,
On Mon, May 21, 2012 at 8:15 PM, Konrad Rzeszutek Wilk
<[email protected]> wrote:
> So you are asking for 052b198 to be back-ported.
> I am OK with that but I think Shaohua needs to Ack that and
> ask Greg to put it on [email protected]
Yes, since I didn't find the official process to propose an
already-in-tree commit to stable@
(http://kernel.org/doc/Documentation/stable_kernel_rules.txt), I was
just asking around, maybe to get Shaohua feedback.
I guess it meets the requirements to be integrated in stable; tested
on my side in 3.2.x and 3.3.x and fixing a precise issue.
Regards,
--
William
On Mon, May 21, 2012 at 11:02:26PM +0200, William Dauchy wrote:
> Hello,
>
> On Mon, May 21, 2012 at 8:15 PM, Konrad Rzeszutek Wilk
> <[email protected]> wrote:
> > So you are asking for 052b198 to be back-ported.
> > I am OK with that but I think Shaohua needs to Ack that and
> > ask Greg to put it on [email protected]
>
> Yes, since I didn't find the official process to propose an
> already-in-tree commit to stable@
> (http://kernel.org/doc/Documentation/stable_kernel_rules.txt), I was
> just asking around, maybe to get Shaohua feedback.
>
> I guess it meets the requirements to be integrated in stable; tested
> on my side in 3.2.x and 3.3.x and fixing a precise issue.
Now applied to the 3.3.x tree, thanks.
greg k-h
On Thu, May 24, 2012 at 8:50 PM, Greg Kroah-Hartman
<[email protected]> wrote:
> Now applied to the 3.3.x tree, thanks.
Thanks.
Ben, do you plan to apply it on top of the 3.2.x tree?
Regards,
--
William
On Fri, 2012-05-25 at 23:19 +0200, William Dauchy wrote:
> On Thu, May 24, 2012 at 8:50 PM, Greg Kroah-Hartman
> <[email protected]> wrote:
> > Now applied to the 3.3.x tree, thanks.
>
> Thanks.
>
> Ben, do you plan to apply it on top of the 3.2.x tree?
Just added it to the queue, thanks.
Ben.
--
Ben Hutchings
You can't have everything. Where would you put it?
On Mon, 21 May 2012, Konrad Rzeszutek Wilk wrote:
> On Mon, May 21, 2012 at 12:30:45AM +0200, William Dauchy wrote:
> > Hello,
> >
> > On Xen, when booting a guest with a system disk and an additional swap
> > disk I'm getting a calltrace.
> > xen hypervisor: 4.1.2; linux dom0: v3.3.6; linux guest: v3.2.17
> > When booting without a swap disk, I don't have the issue.
> > I also tested a guest with v3.3.6: same problem. But from v3.4-rc2,
> > the issue is fixed.
> > I cherry-picked:
>
> > 052b198 swap: don't do discard if no discard option added
>
> So you are asking for 052b198 to be back-ported.
>
> I am OK with that but I think Shaohua needs to Ack that and
> ask Greg to put it on [email protected]
Since that commit did indeed go into v3.4, I won't quarrel with it
now going to stable.
But the commit went in to work around the slow discard implementation
on OCZ Vertex II SSDs.
Please, could someone explain to me the meaning of the stacktrace
below (which is missing a WARNING or BUG line?), and how disabling
swap discard fixes it?
At present I see no connection (beyond the fact that the patch fixes
the symptom): in the absence of understanding, I have to beware that
the underlying issue may remain unfixed.
Hugh
>
>
>
> > Applied and tested on top of v3.2.17 and v3.3.6, it fixes the issue.
> >
> > Pid: 0, comm: swapper/0 Not tainted 3.2.17-x86_64 #12
> > Call Trace:
> > <IRQ>
> > [<ffffffff810919da>] ? handle_irq_event_percpu+0x3a/0x140
> > [<ffffffff81091b29>] ? handle_irq_event+0x49/0x80
> > [<ffffffff81094e7d>] ? handle_edge_irq+0x6d/0x120
> > [<ffffffff81229088>] ? __xen_evtchn_do_upcall+0x1b8/0x280
> > [<ffffffff8122a442>] ? xen_evtchn_do_upcall+0x22/0x40
> > [<ffffffff8133f4fe>] ? xen_do_hypervisor_callback+0x1e/0x30
> > <EOI>
> > [<ffffffff810013aa>] ? hypercall_page+0x3aa/0x1000
> > [<ffffffff810013aa>] ? hypercall_page+0x3aa/0x1000
> > [<ffffffff8100768c>] ? xen_safe_halt+0xc/0x20
> > [<ffffffff81013563>] ? default_idle+0x23/0x40
> > [<ffffffff8100b073>] ? cpu_idle+0x63/0xb0
> > [<ffffffff81654c43>] ? start_kernel+0x362/0x36d
> > [<ffffffff81657491>] ? xen_start_kernel+0x558/0x55e
> > Code: 39 ed 0f 84 1c 02 00 00 44 8b 7b 48 4c 8b 73 50 41 83 ef 01 41
> > 21 ef 49 6b c7 70 4d 8b 64 06 40 49 69 c4 d0 00 00 00 48 8d 14 03 <48>
> > 8b 8a 78 02 00 00 48 89 4c 24 10 80 ba 09 02 00 00 00 74 6d
> > RIP [<ffffffff8125ed66>] blkif_interrupt+0x66/0x320
> > RSP <ffff88001fc03e18>
> > ---[ end trace dfd4e5623eb06620 ]---
On Sat, May 26, 2012 at 11:29:37AM -0700, Hugh Dickins wrote:
> On Mon, 21 May 2012, Konrad Rzeszutek Wilk wrote:
> > On Mon, May 21, 2012 at 12:30:45AM +0200, William Dauchy wrote:
> > > Hello,
> > >
> > > On Xen, when booting a guest with a system disk and an additional swap
> > > disk I'm getting a calltrace.
> > > xen hypervisor: 4.1.2; linux dom0: v3.3.6; linux guest: v3.2.17
> > > When booting without a swap disk, I don't have the issue.
> > > I also tested a guest with v3.3.6: same problem. But from v3.4-rc2,
> > > the issue is fixed.
> > > I cherry-picked:
> >
> > > 052b198 swap: don't do discard if no discard option added
> >
> > So you are asking for 052b198 to be back-ported.
> >
> > I am OK with that but I think Shaohua needs to Ack that and
> > ask Greg to put it on [email protected]
>
> Since that commit did indeed go into v3.4, I won't quarrel with it
> now going to stable.
>
> But the commit went in to work around the slow discard implementation
> on OCZ Vertex II SSDs.
>
> Please, could someone explain to me the meaning of the stacktrace
> below (which is missing a WARNING or BUG line?), and how disabling
> swap discard fixes it?
I think I know and just narrowed down the issue this Friday.
William, could you please apply the patch outlined in
https://bugzilla.redhat.com/show_bug.cgi?id=824641
to your dom0 and see if that (so do not have 052b198 in your branch)
>
> At present I see no connection (beyond the fact that the patch fixes
> the symptom): in the absence of understanding, I have to beware that
> the underlying issue may remain unfixed.
<nods>
>
> Hugh
>
> >
> >
> >
> > > Applied and tested on top of v3.2.17 and v3.3.6, it fixes the issue.
> > >
> > > Pid: 0, comm: swapper/0 Not tainted 3.2.17-x86_64 #12
> > > Call Trace:
> > > <IRQ>
> > > [<ffffffff810919da>] ? handle_irq_event_percpu+0x3a/0x140
> > > [<ffffffff81091b29>] ? handle_irq_event+0x49/0x80
> > > [<ffffffff81094e7d>] ? handle_edge_irq+0x6d/0x120
> > > [<ffffffff81229088>] ? __xen_evtchn_do_upcall+0x1b8/0x280
> > > [<ffffffff8122a442>] ? xen_evtchn_do_upcall+0x22/0x40
> > > [<ffffffff8133f4fe>] ? xen_do_hypervisor_callback+0x1e/0x30
> > > <EOI>
> > > [<ffffffff810013aa>] ? hypercall_page+0x3aa/0x1000
> > > [<ffffffff810013aa>] ? hypercall_page+0x3aa/0x1000
> > > [<ffffffff8100768c>] ? xen_safe_halt+0xc/0x20
> > > [<ffffffff81013563>] ? default_idle+0x23/0x40
> > > [<ffffffff8100b073>] ? cpu_idle+0x63/0xb0
> > > [<ffffffff81654c43>] ? start_kernel+0x362/0x36d
> > > [<ffffffff81657491>] ? xen_start_kernel+0x558/0x55e
> > > Code: 39 ed 0f 84 1c 02 00 00 44 8b 7b 48 4c 8b 73 50 41 83 ef 01 41
> > > 21 ef 49 6b c7 70 4d 8b 64 06 40 49 69 c4 d0 00 00 00 48 8d 14 03 <48>
> > > 8b 8a 78 02 00 00 48 89 4c 24 10 80 ba 09 02 00 00 00 74 6d
> > > RIP [<ffffffff8125ed66>] blkif_interrupt+0x66/0x320
> > > RSP <ffff88001fc03e18>
> > > ---[ end trace dfd4e5623eb06620 ]---
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
Hello Konrad,
On Tue, May 29, 2012 at 4:47 PM, Konrad Rzeszutek Wilk
<[email protected]> wrote:
> I think I know and just narrowed down the issue this Friday.
> William, could you please apply the patch outlined in
> https://bugzilla.redhat.com/show_bug.cgi?id=824641
> to your dom0 and see if that (so do not have 052b198 in your branch)
I applied the patch on dom0 and removed 052b198 from my virtual
machine and it worked.
Regards,
--
William
On Wed, May 30, 2012 at 10:29:16PM +0200, William Dauchy wrote:
> Hello Konrad,
>
> On Tue, May 29, 2012 at 4:47 PM, Konrad Rzeszutek Wilk
> <[email protected]> wrote:
> > I think I know and just narrowed down the issue this Friday.
> > William, could you please apply the patch outlined in
> > https://bugzilla.redhat.com/show_bug.cgi?id=824641
> > to your dom0 and see if that (so do not have 052b198 in your branch)
>
> I applied the patch on dom0 and removed 052b198 from my virtual
> machine and it worked.
Great. Is it OK to attach a Tested-by tag to the patch with your name?
>
> Regards,
> --
> William
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
On Wed, May 30, 2012 at 11:16 PM, Konrad Rzeszutek Wilk
<[email protected]> wrote:
> Great. Is it OK to attach a Tested-by tag to the patch with your name?
Sure.
--
William
On Tue, 29 May 2012, Konrad Rzeszutek Wilk wrote:
>
> I think I know and just narrowed down the issue this Friday.
>
> William, could you please apply the patch outlined in
> https://bugzilla.redhat.com/show_bug.cgi?id=824641
> to your dom0 and see if that (so do not have 052b198 in your branch)
>
> >
> > At present I see no connection (beyond the fact that the patch fixes
> > the symptom): in the absence of understanding, I have to beware that
> > the underlying issue may remain unfixed.
>
> <nods>
Thanks a lot for pursuing that to a much more satisfying conclusion.
Hugh