Re: [PATCH v2] Persistent grant maps for xen blk drivers

[Oliver Chick]

Hi Konrad,

I have applied my patch to your #linux-next, and compiled. I don't
experience the same problem that you do - both PV and HVM boot and work.

Do you have anything special in your setup that might help me reproduce
the problem?


On Fri, 2012-09-21 at 19:41 +0100, Konrad Rzeszutek Wilk wrote:
> On Fri, Sep 21, 2012 at 04:52:47PM +0100, Oliver Chick wrote:
> > This patch implements persistent grants for the xen-blk{front,back}
> > mechanism. The effect of this change is to reduce the number of unmap
> > operations performed, since they cause a (costly) TLB shootdown. This
> > allows the I/O performance to scale better when a large number of VMs
> > are performing I/O.
> 
> So I applied your patch to my #linux-next and I got this (the dom0 is the
> same as the domU)
> I get this when using an PVHVM (*)and PV guest:
> # fdisk /dev/xvda
> [   28.594049] BUG: unable to handle kernel NULL pointer dereference at 0000000000000008
> [   28.595028] IP: [<ffffffffa003fe9a>] blkif_interrupt+0x3aa/0x5d0 [xen_blkfront]
> [   28.595028] PGD dcaf0067 PUD 7fb5e067 PMD 0 
> [   28.595028] Oops: 0000 [#1] SMP 
> [   28.595028] Modules linked in: iscsi_boot_sysfs iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi libcrc32c crc32c sg sr_mod cdrom ata_generic ata_piix crc32c_intel libata scsi_mod xen_blkfront xen_netfront fb_sys_fops sysimgblt sysfillrect syscopyarea xen_kbdfront xenfs xen_privcmd [last unloaded: dump_dma]
> [   28.595028] CPU 0 
> [   28.595028] Pid: 0, comm: swapper/0 Tainted: G           O 3.6.0-rc6upstream-00196-ge45cba2 #1 Xen HVM domU
> [   28.595028] RIP: 0010:[<ffffffffa003fe9a>]  [<ffffffffa003fe9a>] blkif_interrupt+0x3aa/0x5d0 [xen_blkfront]
> [   28.595028] RSP: 0018:ffff88010fc03da8  EFLAGS: 00010002
> [   28.595028] RAX: 0000000000000000 RBX: ffffffff81a00000 RCX: ffff88010af5f140
> [   28.595028] RDX: ffff88010af5f080 RSI: ffff8800dbbf8210 RDI: ffff8800dc679c00
> [   28.595028] RBP: ffff88010fc03e28 R08: 0000000000000001 R09: 0000000000000011
> [   28.595028] R10: 0000000000000000 R11: 0000000000000000 R12: 6db6db6db6db6db7
> [   28.595028] R13: ffff88010af5f1a8 R14: 0000000000000004 R15: 0000000000000000
> [   28.595028] FS:  0000000000000000(0000) GS:ffff88010fc00000(0000) knlGS:0000000000000000
> [   28.595028] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> [   28.595028] CR2: 0000000000000008 CR3: 00000000d4ac0000 CR4: 00000000000006f0
> [   28.595028] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> [   28.595028] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> [   28.595028] Process swapper/0 (pid: 0, threadinfo ffffffff81a00000, task ffffffff81a13420)
> [   28.595028] Stack:
> [   28.595028]  0000000000000086 ffff8800d9951060 ffff8800d9953000 0000000881376d09
> [   28.595028]  0000000000000008 0000000a81376f2d 0000000000000000 ffff8800dbbf8210
> [   28.595028]  ffff8800dbbf8000 ffff88010af5f140 ffff88010fc03e08 ffff8800dcc4e3c0
> [   28.595028] Call Trace:
> [   28.595028]  <IRQ> 
> [   28.595028]  [<ffffffff8110dcdc>] handle_irq_event_percpu+0x7c/0x240
> [   28.595028]  [<ffffffff8110df02>] handle_irq_event+0x62/0x90
> [   28.595028]  [<ffffffff8111178e>] handle_edge_irq+0x8e/0x160
> [   28.595028]  [<ffffffff81376ac4>] __xen_evtchn_do_upcall+0x1c4/0x2a0
> [   28.595028]  [<ffffffff813775ba>] xen_evtchn_do_upcall+0x2a/0x40
> [   28.595028]  [<ffffffff816314ba>] xen_hvm_callback_vector+0x6a/0x70
> [   28.595028]  <EOI> 
> [   28.595028]  [<ffffffff8107b3a6>] ? native_safe_halt+0x6/0x10
> [   28.595028]  [<ffffffff810555da>] default_idle+0x4a/0x210
> [   28.595028]  [<ffffffff81054d79>] cpu_idle+0x99/0xe0
> [   28.595028]  [<ffffffff816015da>] rest_init+0x8a/0xa0
> [   28.595028]  [<ffffffff81abcd72>] start_kernel+0x383/0x390
> [   28.595028]  [<ffffffff81abc80d>] ? kernel_init+0x1e8/0x1e8
> [   28.595028]  [<ffffffff81abc356>] x86_64_start_reservations+0x131/0x136
> [   28.595028]  [<ffffffff81abc45e>] x86_64_start_kernel+0x103/0x112
> [   28.595028] Code: 00 49 83 c5 10 41 8b 45 08 41 03 45 0c 3d 00 10 00 00 0f 87 7e 01 00 00 48 8b 75 b8 49 63 c6 41 83 c6 01 48 8b 84 c6 d0 00 00 00 <48> 8b 40 08 83 43 1c 01 48 8d 14 c5 00 00 00 00 48 c1 e0 06 48 
> [   28.595028] RIP  [<ffffffffa003fe9a>] blkif_interrupt+0x3aa/0x5d0 [xen_blkfront]
> [   28.595028]  RSP <ffff88010fc03da8>
> [   28.595028] CR2: 0000000000000008
> [   28.595028] ---[ end trace 49f5630a34777570 ]---
> [   28.595028] Kernel panic - not syncing: Fatal exception in interrupt
> [   28.595028] Rebooting in 60 seconds..
> 
> *: With a PVHVM guest I get
> 
> [  261.927218] privcmd_fault: vma=ffff88002a31dce8 7f4edc095000-7f4edc195000, pgoff=c8, uv=00007f4edc15d000
> 
> thought if I applied your patch on top of v3.6-rc6 I didn't see the privcmd_fault but
> I did see the guest crash.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/