Mitch:
I have a VM using a 82576 based VF. Running:
$ ethtool -C eth2 rx-usecs 0
generates the following trace on console:
[ 894.683322] divide error: 0000 [#1] SMP
[ 894.684020] CPU 1
[ 894.684020] Modules linked in: sunrpc virtio_net igbvf virtio_blk
virtio_pci virtio_ring virtio
[ 894.684020]
[ 894.684020] Pid: 7310, comm: ethtool Not tainted 3.5.0-rc1 #0 Bochs Bochs
[ 894.684020] RIP: 0010:[<ffffffffa00259ec>] [<ffffffffa00259ec>]
igbvf_set_coalesce+0x5b/0x8b [igbvf]
[ 894.684020] RSP: 0018:ffff88003cd51c38 EFLAGS: 00010246
[ 894.684020] RAX: 000000003b9aca00 RBX: ffff88003aa42000 RCX:
0000000000000000
[ 894.684020] RDX: 0000000000000000 RSI: ffff88003cd51c48 RDI:
ffff88003aa42780
[ 894.684020] RBP: ffff88003cd51c38 R08: 0000000000000000 R09:
0000000000000000
[ 894.684020] R10: 0000000000000000 R11: ffff88003cd51ec8 R12:
ffff88003cd51c48
[ 894.684020] R13: 000000000000000f R14: 00000000ffffffff R15:
00000000fffffff2
[ 894.684020] FS: 00007f8142ccb720(0000) GS:ffff88003fd00000(0000)
knlGS:0000000000000000
[ 894.684020] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 894.684020] CR2: 000000000040466c CR3: 000000003c754000 CR4:
00000000000007e0
[ 894.684020] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[ 894.684020] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
0000000000000400
[ 894.684020] Process ethtool (pid: 7310, threadinfo ffff88003cd50000,
task ffff88003b984470)
[ 894.684020] Stack:
[ 894.684020] ffff88003cd51cb8 ffffffff8132bcf5 000000000000000f
0000000000000000
[ 894.684020] 0000000000000000 0000000000000000 0000000000000000
0000000000000000
[ 894.684020] 0000000000000000 0000000000000000 0000000000000000
0000000000000000
[ 894.684020] Call Trace:
[ 894.684020] [<ffffffff8132bcf5>] ethtool_set_coalesce+0x54/0x5d
[ 894.684020] [<ffffffff8132d0c9>] dev_ethtool+0x5dc/0x175b
[ 894.684020] [<ffffffff81057921>] ? need_resched+0x1e/0x28
[ 894.684020] [<ffffffff81057934>] ? should_resched+0x9/0x29
[ 894.684020] [<ffffffff813be822>] ? _cond_resched+0xe/0x22
[ 894.684020] [<ffffffff8132b227>] dev_ioctl+0x517/0x684
[ 894.684020] [<ffffffff810ec91f>] ? pmd_offset+0x14/0x3b
[ 894.684020] [<ffffffff81315045>] sock_do_ioctl+0x3d/0x48
[ 894.684020] [<ffffffff8131545f>] sock_ioctl+0x1f8/0x207
[ 894.684020] [<ffffffff8111f6fa>] do_vfs_ioctl+0x475/0x4b6
[ 894.684020] [<ffffffff811f8e70>] ?
inode_has_perm.clone.19.clone.27+0x33/0x35
[ 894.684020] [<ffffffff811f9298>] ? file_has_perm+0x73/0x7e
[ 894.684020] [<ffffffff8110f6cf>] ? fd_install+0x57/0x60
[ 894.684020] [<ffffffff8111f791>] sys_ioctl+0x56/0x79
[ 894.684020] [<ffffffff813c5de9>] system_call_fastpath+0x16/0x1b
[ 894.684020] Code: 83 f8 02 77 0f c7 87 0c 03 00 00 e8 01 00 00 8b 46
04 eb 19 8d 04 8d 00 00 00 00 31 d2 89 87 0c 03 00 00 c1 e1 0a b8 00 ca
9a 3b <f7> f1 89 87 08 03 00 00 8b 97 0c 03 00 00 48 8b 87 80 03 00 00
[ 894.684020] RIP [<ffffffffa00259ec>] igbvf_set_coalesce+0x5b/0x8b
[igbvf]
[ 894.684020] RSP <ffff88003cd51c38>
[ 894.779474] ---[ end trace 162bed6b66df758d ]---
This commit introduced the problem:
commit ab50a2a430693b0961dc7b7d9fe2a4bd77d11ea6
Author: Mitch A Williams [off-list ref]
Date: Sat Jan 14 08:10:50 2012 +0000
igbvf: refactor Interrupt Throttle Rate code
David
-----Original Message-----
From: David Ahern [mailto:daahern@cisco.com]
Sent: Monday, June 18, 2012 11:16 AM
To: Williams, Mitch A
Cc: netdev@vger.kernel.org
Subject: divide by 0 error in igbvf_set_coalesce - ab50a2a
Mitch:
I have a VM using a 82576 based VF. Running:
$ ethtool -C eth2 rx-usecs 0
generates the following trace on console:
[ 894.683322] divide error: 0000 [#1] SMP [ 894.684020] CPU 1 [
Thanks for letting me know, David. I'll look into it and get a patch out soon. Shouldn't be that big of a deal to fix.
In the meantime, my advice to you is, "Don't do that."
-Mitch
894.684020] Modules linked in: sunrpc virtio_net igbvf virtio_blk
virtio_pci virtio_ring virtio [ 894.684020] [ 894.684020] Pid: 7310,
comm: ethtool Not tainted 3.5.0-rc1 #0 Bochs Bochs [ 894.684020] RIP:
0010:[<ffffffffa00259ec>] [<ffffffffa00259ec>]
igbvf_set_coalesce+0x5b/0x8b [igbvf] [ 894.684020] RSP:
0018:ffff88003cd51c38 EFLAGS: 00010246 [ 894.684020] RAX:
000000003b9aca00 RBX: ffff88003aa42000 RCX:
0000000000000000
[ 894.684020] RDX: 0000000000000000 RSI: ffff88003cd51c48 RDI:
ffff88003aa42780
[ 894.684020] RBP: ffff88003cd51c38 R08: 0000000000000000 R09:
0000000000000000
[ 894.684020] R10: 0000000000000000 R11: ffff88003cd51ec8 R12:
ffff88003cd51c48
[ 894.684020] R13: 000000000000000f R14: 00000000ffffffff R15:
00000000fffffff2
[ 894.684020] FS: 00007f8142ccb720(0000) GS:ffff88003fd00000(0000)
knlGS:0000000000000000
[ 894.684020] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [
894.684020] CR2: 000000000040466c CR3: 000000003c754000 CR4:
00000000000007e0
[ 894.684020] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[ 894.684020] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
0000000000000400
[ 894.684020] Process ethtool (pid: 7310, threadinfo ffff88003cd50000,
task ffff88003b984470) [ 894.684020] Stack:
[ 894.684020] ffff88003cd51cb8 ffffffff8132bcf5 000000000000000f
0000000000000000
[ 894.684020] 0000000000000000 0000000000000000 0000000000000000
0000000000000000
[ 894.684020] 0000000000000000 0000000000000000 0000000000000000
0000000000000000
[ 894.684020] Call Trace:
[ 894.684020] [<ffffffff8132bcf5>] ethtool_set_coalesce+0x54/0x5d [
894.684020] [<ffffffff8132d0c9>] dev_ethtool+0x5dc/0x175b [
894.684020] [<ffffffff81057921>] ? need_resched+0x1e/0x28 [
894.684020] [<ffffffff81057934>] ? should_resched+0x9/0x29 [
894.684020] [<ffffffff813be822>] ? _cond_resched+0xe/0x22 [
894.684020] [<ffffffff8132b227>] dev_ioctl+0x517/0x684 [ 894.684020]
[<ffffffff810ec91f>] ? pmd_offset+0x14/0x3b [ 894.684020]
[<ffffffff81315045>] sock_do_ioctl+0x3d/0x48 [ 894.684020]
[<ffffffff8131545f>] sock_ioctl+0x1f8/0x207 [ 894.684020]
[<ffffffff8111f6fa>] do_vfs_ioctl+0x475/0x4b6 [ 894.684020]
[<ffffffff811f8e70>] ?
inode_has_perm.clone.19.clone.27+0x33/0x35
[ 894.684020] [<ffffffff811f9298>] ? file_has_perm+0x73/0x7e [
894.684020] [<ffffffff8110f6cf>] ? fd_install+0x57/0x60 [ 894.684020]
[<ffffffff8111f791>] sys_ioctl+0x56/0x79 [ 894.684020]
[<ffffffff813c5de9>] system_call_fastpath+0x16/0x1b [ 894.684020] Code:
83 f8 02 77 0f c7 87 0c 03 00 00 e8 01 00 00 8b 46
04 eb 19 8d 04 8d 00 00 00 00 31 d2 89 87 0c 03 00 00 c1 e1 0a b8 00 ca
9a 3b <f7> f1 89 87 08 03 00 00 8b 97 0c 03 00 00 48 8b 87 80 03 00 00 [
894.684020] RIP [<ffffffffa00259ec>] igbvf_set_coalesce+0x5b/0x8b
[igbvf] [ 894.684020] RSP <ffff88003cd51c38> [ 894.779474] ---[ end
trace 162bed6b66df758d ]---
This commit introduced the problem:
commit ab50a2a430693b0961dc7b7d9fe2a4bd77d11ea6
Author: Mitch A Williams [off-list ref]
Date: Sat Jan 14 08:10:50 2012 +0000
igbvf: refactor Interrupt Throttle Rate code
David
On 6/18/12 2:45 PM, Williams, Mitch A wrote:
Thanks for letting me know, David. I'll look into it and get a patch out soon. Shouldn't be that big of a deal to fix.
Could you CC me on the patch so I know when it's fixed? I have enough
events to poll.
In the meantime, my advice to you is, "Don't do that."
Uh, yea. Figured that part out. ;-)
Thanks,
David
On 06/20/2012 03:21 PM, David Ahern wrote:
On 6/18/12 2:45 PM, Williams, Mitch A wrote:
quoted
Thanks for letting me know, David. I'll look into it and get a patch
out soon. Shouldn't be that big of a deal to fix.
Could you CC me on the patch so I know when it's fixed? I have enough
events to poll.
I will CC you when I push Mitch's patch upstream.