* Fw: [2.5.66-mm3+debug] traces.
@ 2003-04-05 1:29 Andrew Morton
2003-04-05 1:35 ` David S. Miller
0 siblings, 1 reply; 4+ messages in thread
From: Andrew Morton @ 2003-04-05 1:29 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev
[-- Attachment #1: Type: text/plain, Size: 11443 bytes --]
Dave,
Ian was hitting the BUG in local_bh_enable(). The call trace is
below.
This patch fixes it up:
diff -puN drivers/net/tulip/dmfe.c~dmfe-kfree_skb-fix drivers/net/tulip/dmfe.c
--- 25/drivers/net/tulip/dmfe.c~dmfe-kfree_skb-fix Fri Apr 4 17:26:33 2003
+++ 25-akpm/drivers/net/tulip/dmfe.c Fri Apr 4 17:26:41 2003
@@ -668,13 +668,13 @@ static int dmfe_start_xmit(struct sk_buf
if ( db->tx_queue_cnt < TX_FREE_DESC_CNT )
netif_wake_queue(dev);
- /* free this SKB */
- dev_kfree_skb(skb);
-
/* Restore CR7 to enable interrupt */
spin_unlock_irqrestore(&db->lock, flags);
outl(db->cr7_data, dev->base_addr + DCR7);
+ /* free this SKB */
+ dev_kfree_skb(skb);
+
return 0;
}
_
It is a bit worrisome that dev_kfree_skb() is now illegal if local interrupts
are disabled. I'd expect there to be more instances.
Begin forwarded message:
Date: 05 Apr 2003 02:28:30 +0200
From: Ian Kumlien <pomac@vapor.com>
To: Andrew Morton <akpm@digeo.com>
Subject: [2.5.66-mm3+debug] traces.
Hi,
I'm currently to tired to proof read these to see if they are the same
or not... i think they are... And i don't trust 2.5.x's makesystem to
keep track any more =)
Anyways, I added em all to this mail, i HOPE that you have no
limitations on your mail account and so on. I also hope that your on a
dsl or simular connection =).
btw, it seems to happen on every nfs access. I'll explore the
possibilities of using the pure tulip network driver tomorrow...
irqs are enabled
Call Trace:
[<c011f424>] local_bh_enable+0x74/0x80
[<c038ba8a>] svc_write_space+0x2a/0x90
[<c030d309>] __kfree_skb+0x69/0xf0
[<c030c3b9>] sock_wfree+0x49/0x50
[<c030c370>] sock_wfree+0x0/0x50
[<c030d2fd>] __kfree_skb+0x5d/0xf0
[<c0244010>] dmfe_start_xmit+0x150/0x240
[<c0310c2d>] dev_queue_xmit_nit+0x9d/0xf0
[<c0318195>] qdisc_restart+0x65/0x140
[<c0310f10>] dev_queue_xmit+0x200/0x2b0
[<c0321ee9>] ip_output+0x109/0x220
[<c0323ae6>] ip_push_pending_frames+0x296/0x380
[<c034020d>] udp_push_pending_frames+0xfd/0x210
[<c03405ad>] udp_sendmsg+0x24d/0x880
[<c0323384>] ip_append_data+0x664/0x750
[<c031dd6d>] __ip_route_output_key+0x2d/0xf0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c03097de>] sock_sendmsg+0x9e/0xd0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c030caf4>] sock_no_sendpage+0xa4/0xb0
[<c0340d81>] udp_sendpage+0x1a1/0x2c0
[<c0348ed7>] inet_sendpage+0x67/0xd0
[<c038b6a1>] svc_sendto+0xb1/0x270
[<c038be81>] svc_udp_sendto+0x21/0x40
[<c038d288>] svc_send+0x78/0xf0
[<c038efd8>] svcauth_unix_release+0x38/0x40
[<c01ac180>] nfsd_dispatch+0x0/0x1f0
[<c038ac0c>] svc_process+0x1fc/0x680
[<c01abfe9>] nfsd+0x1b9/0x350
[<c01abe30>] nfsd+0x0/0x350
[<c010706d>] kernel_thread_helper+0x5/0x18
irqs are enabled
Call Trace:
[<c011f424>] local_bh_enable+0x74/0x80
[<c038ba8a>] svc_write_space+0x2a/0x90
[<c030d309>] __kfree_skb+0x69/0xf0
[<c030c3b9>] sock_wfree+0x49/0x50
[<c030c370>] sock_wfree+0x0/0x50
[<c030d2fd>] __kfree_skb+0x5d/0xf0
[<c0244010>] dmfe_start_xmit+0x150/0x240
[<c0310c2d>] dev_queue_xmit_nit+0x9d/0xf0
[<c0318195>] qdisc_restart+0x65/0x140
[<c0310f10>] dev_queue_xmit+0x200/0x2b0
[<c0321ee9>] ip_output+0x109/0x220
[<c0323ae6>] ip_push_pending_frames+0x296/0x380
[<c034020d>] udp_push_pending_frames+0xfd/0x210
[<c03405ad>] udp_sendmsg+0x24d/0x880
[<c0323384>] ip_append_data+0x664/0x750
[<c031dd6d>] __ip_route_output_key+0x2d/0xf0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c03097de>] sock_sendmsg+0x9e/0xd0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c030caf4>] sock_no_sendpage+0xa4/0xb0
[<c0340d81>] udp_sendpage+0x1a1/0x2c0
[<c0348ed7>] inet_sendpage+0x67/0xd0
[<c038b6a1>] svc_sendto+0xb1/0x270
[<c038be81>] svc_udp_sendto+0x21/0x40
[<c038d288>] svc_send+0x78/0xf0
[<c038efd8>] svcauth_unix_release+0x38/0x40
[<c01ac180>] nfsd_dispatch+0x0/0x1f0
[<c038ac0c>] svc_process+0x1fc/0x680
[<c01abfe9>] nfsd+0x1b9/0x350
[<c01abe30>] nfsd+0x0/0x350
[<c010706d>] kernel_thread_helper+0x5/0x18
irqs are enabled
Call Trace:
[<c011f424>] local_bh_enable+0x74/0x80
[<c038ba8a>] svc_write_space+0x2a/0x90
[<c030d309>] __kfree_skb+0x69/0xf0
[<c030c3b9>] sock_wfree+0x49/0x50
[<c030c370>] sock_wfree+0x0/0x50
[<c030d2fd>] __kfree_skb+0x5d/0xf0
[<c0244010>] dmfe_start_xmit+0x150/0x240
[<c0310c2d>] dev_queue_xmit_nit+0x9d/0xf0
[<c0318195>] qdisc_restart+0x65/0x140
[<c0310f10>] dev_queue_xmit+0x200/0x2b0
[<c0321ee9>] ip_output+0x109/0x220
[<c0323ae6>] ip_push_pending_frames+0x296/0x380
[<c034020d>] udp_push_pending_frames+0xfd/0x210
[<c03405ad>] udp_sendmsg+0x24d/0x880
[<c0323384>] ip_append_data+0x664/0x750
[<c031dd6d>] __ip_route_output_key+0x2d/0xf0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c03097de>] sock_sendmsg+0x9e/0xd0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c030caf4>] sock_no_sendpage+0xa4/0xb0
[<c0340d81>] udp_sendpage+0x1a1/0x2c0
[<c0348ed7>] inet_sendpage+0x67/0xd0
[<c038b6a1>] svc_sendto+0xb1/0x270
[<c038be81>] svc_udp_sendto+0x21/0x40
[<c038d288>] svc_send+0x78/0xf0
[<c038efd8>] svcauth_unix_release+0x38/0x40
[<c01ac180>] nfsd_dispatch+0x0/0x1f0
[<c038ac0c>] svc_process+0x1fc/0x680
[<c01abfe9>] nfsd+0x1b9/0x350
[<c01abe30>] nfsd+0x0/0x350
[<c010706d>] kernel_thread_helper+0x5/0x18
irqs are enabled
Call Trace:
[<c011f424>] local_bh_enable+0x74/0x80
[<c038ba8a>] svc_write_space+0x2a/0x90
[<c030d309>] __kfree_skb+0x69/0xf0
[<c030c3b9>] sock_wfree+0x49/0x50
[<c030c370>] sock_wfree+0x0/0x50
[<c030d2fd>] __kfree_skb+0x5d/0xf0
[<c0244010>] dmfe_start_xmit+0x150/0x240
[<c0310c2d>] dev_queue_xmit_nit+0x9d/0xf0
[<c0318195>] qdisc_restart+0x65/0x140
[<c0310f10>] dev_queue_xmit+0x200/0x2b0
[<c0321ee9>] ip_output+0x109/0x220
[<c0323ae6>] ip_push_pending_frames+0x296/0x380
[<c034020d>] udp_push_pending_frames+0xfd/0x210
[<c03405ad>] udp_sendmsg+0x24d/0x880
[<c0323384>] ip_append_data+0x664/0x750
[<c031dd6d>] __ip_route_output_key+0x2d/0xf0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c03097de>] sock_sendmsg+0x9e/0xd0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c030caf4>] sock_no_sendpage+0xa4/0xb0
[<c0340d81>] udp_sendpage+0x1a1/0x2c0
[<c0348ed7>] inet_sendpage+0x67/0xd0
[<c038b6a1>] svc_sendto+0xb1/0x270
[<c038be81>] svc_udp_sendto+0x21/0x40
[<c038d288>] svc_send+0x78/0xf0
[<c038efd8>] svcauth_unix_release+0x38/0x40
[<c01ac180>] nfsd_dispatch+0x0/0x1f0
[<c038ac0c>] svc_process+0x1fc/0x680
[<c01abfe9>] nfsd+0x1b9/0x350
[<c01abe30>] nfsd+0x0/0x350
[<c010706d>] kernel_thread_helper+0x5/0x18
irqs are enabled
Call Trace:
[<c011f424>] local_bh_enable+0x74/0x80
[<c038ba8a>] svc_write_space+0x2a/0x90
[<c030d309>] __kfree_skb+0x69/0xf0
[<c030c3b9>] sock_wfree+0x49/0x50
[<c030c370>] sock_wfree+0x0/0x50
[<c030d2fd>] __kfree_skb+0x5d/0xf0
[<c0244010>] dmfe_start_xmit+0x150/0x240
[<c0310c2d>] dev_queue_xmit_nit+0x9d/0xf0
[<c0318195>] qdisc_restart+0x65/0x140
[<c0310f10>] dev_queue_xmit+0x200/0x2b0
[<c0321ee9>] ip_output+0x109/0x220
[<c0323ae6>] ip_push_pending_frames+0x296/0x380
[<c034020d>] udp_push_pending_frames+0xfd/0x210
[<c03405ad>] udp_sendmsg+0x24d/0x880
[<c0323384>] ip_append_data+0x664/0x750
[<c031dd6d>] __ip_route_output_key+0x2d/0xf0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c03097de>] sock_sendmsg+0x9e/0xd0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c030caf4>] sock_no_sendpage+0xa4/0xb0
[<c0340d81>] udp_sendpage+0x1a1/0x2c0
[<c0348ed7>] inet_sendpage+0x67/0xd0
[<c038b6a1>] svc_sendto+0xb1/0x270
[<c038be81>] svc_udp_sendto+0x21/0x40
[<c038d288>] svc_send+0x78/0xf0
[<c038efd8>] svcauth_unix_release+0x38/0x40
[<c01ac180>] nfsd_dispatch+0x0/0x1f0
[<c038ac0c>] svc_process+0x1fc/0x680
[<c01abfe9>] nfsd+0x1b9/0x350
[<c01abe30>] nfsd+0x0/0x350
[<c010706d>] kernel_thread_helper+0x5/0x18
irqs are enabled
Call Trace:
[<c011f424>] local_bh_enable+0x74/0x80
[<c038ba8a>] svc_write_space+0x2a/0x90
[<c030d309>] __kfree_skb+0x69/0xf0
[<c030c3b9>] sock_wfree+0x49/0x50
[<c030c370>] sock_wfree+0x0/0x50
[<c030d2fd>] __kfree_skb+0x5d/0xf0
[<c0244010>] dmfe_start_xmit+0x150/0x240
[<c0310c2d>] dev_queue_xmit_nit+0x9d/0xf0
[<c0318195>] qdisc_restart+0x65/0x140
[<c0310f10>] dev_queue_xmit+0x200/0x2b0
[<c0321ee9>] ip_output+0x109/0x220
[<c0323ae6>] ip_push_pending_frames+0x296/0x380
[<c034020d>] udp_push_pending_frames+0xfd/0x210
[<c03405ad>] udp_sendmsg+0x24d/0x880
[<c0323384>] ip_append_data+0x664/0x750
[<c031dd6d>] __ip_route_output_key+0x2d/0xf0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c03097de>] sock_sendmsg+0x9e/0xd0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c030caf4>] sock_no_sendpage+0xa4/0xb0
[<c0340d81>] udp_sendpage+0x1a1/0x2c0
[<c0348ed7>] inet_sendpage+0x67/0xd0
[<c038b6a1>] svc_sendto+0xb1/0x270
[<c038be81>] svc_udp_sendto+0x21/0x40
[<c038d288>] svc_send+0x78/0xf0
[<c038efd8>] svcauth_unix_release+0x38/0x40
[<c01ac180>] nfsd_dispatch+0x0/0x1f0
[<c038ac0c>] svc_process+0x1fc/0x680
[<c01abfe9>] nfsd+0x1b9/0x350
[<c01abe30>] nfsd+0x0/0x350
[<c010706d>] kernel_thread_helper+0x5/0x18
irqs are enabled
Call Trace:
[<c011f424>] local_bh_enable+0x74/0x80
[<c038ba8a>] svc_write_space+0x2a/0x90
[<c030d309>] __kfree_skb+0x69/0xf0
[<c012dab0>] nanosleep_wake_up+0x0/0x10
[<c030c3b9>] sock_wfree+0x49/0x50
[<c030c370>] sock_wfree+0x0/0x50
[<c030d2fd>] __kfree_skb+0x5d/0xf0
[<c0244010>] dmfe_start_xmit+0x150/0x240
[<c0310c2d>] dev_queue_xmit_nit+0x9d/0xf0
[<c0318195>] qdisc_restart+0x65/0x140
[<c0310f10>] dev_queue_xmit+0x200/0x2b0
[<c0321ee9>] ip_output+0x109/0x220
[<c0323ae6>] ip_push_pending_frames+0x296/0x380
[<c034020d>] udp_push_pending_frames+0xfd/0x210
[<c03405ad>] udp_sendmsg+0x24d/0x880
[<c0323384>] ip_append_data+0x664/0x750
[<c031dd6d>] __ip_route_output_key+0x2d/0xf0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c03097de>] sock_sendmsg+0x9e/0xd0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c030caf4>] sock_no_sendpage+0xa4/0xb0
[<c0340d81>] udp_sendpage+0x1a1/0x2c0
[<c0348ed7>] inet_sendpage+0x67/0xd0
[<c038b6a1>] svc_sendto+0xb1/0x270
[<c038be81>] svc_udp_sendto+0x21/0x40
[<c038d288>] svc_send+0x78/0xf0
[<c038efd8>] svcauth_unix_release+0x38/0x40
[<c01ac180>] nfsd_dispatch+0x0/0x1f0
[<c038ac0c>] svc_process+0x1fc/0x680
[<c01abfe9>] nfsd+0x1b9/0x350
[<c01abe30>] nfsd+0x0/0x350
[<c010706d>] kernel_thread_helper+0x5/0x18
irqs are enabled
Call Trace:
[<c011f424>] local_bh_enable+0x74/0x80
[<c038ba8a>] svc_write_space+0x2a/0x90
[<c030d309>] __kfree_skb+0x69/0xf0
[<c030c3b9>] sock_wfree+0x49/0x50
[<c030c370>] sock_wfree+0x0/0x50
[<c030d2fd>] __kfree_skb+0x5d/0xf0
[<c0244010>] dmfe_start_xmit+0x150/0x240
[<c0310c2d>] dev_queue_xmit_nit+0x9d/0xf0
[<c0318195>] qdisc_restart+0x65/0x140
[<c0310f10>] dev_queue_xmit+0x200/0x2b0
[<c0321ee9>] ip_output+0x109/0x220
[<c0323ae6>] ip_push_pending_frames+0x296/0x380
[<c034020d>] udp_push_pending_frames+0xfd/0x210
[<c03405ad>] udp_sendmsg+0x24d/0x880
[<c0323384>] ip_append_data+0x664/0x750
[<c031dd6d>] __ip_route_output_key+0x2d/0xf0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c03097de>] sock_sendmsg+0x9e/0xd0
[<c0348e5d>] inet_sendmsg+0x4d/0x60
[<c030caf4>] sock_no_sendpage+0xa4/0xb0
[<c0340d81>] udp_sendpage+0x1a1/0x2c0
[<c0348ed7>] inet_sendpage+0x67/0xd0
[<c038b6a1>] svc_sendto+0xb1/0x270
[<c038be81>] svc_udp_sendto+0x21/0x40
[<c038d288>] svc_send+0x78/0xf0
[<c038efd8>] svcauth_unix_release+0x38/0x40
[<c01ac180>] nfsd_dispatch+0x0/0x1f0
[<c038ac0c>] svc_process+0x1fc/0x680
[<c01abfe9>] nfsd+0x1b9/0x350
[<c01abe30>] nfsd+0x0/0x350
[<c010706d>] kernel_thread_helper+0x5/0x18
--
Ian Kumlien <pomac@vapor.com>
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 190 bytes --]
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [2.5.66-mm3+debug] traces.
2003-04-05 1:29 Fw: [2.5.66-mm3+debug] traces Andrew Morton
@ 2003-04-05 1:35 ` David S. Miller
2003-04-05 2:14 ` Andrew Morton
0 siblings, 1 reply; 4+ messages in thread
From: David S. Miller @ 2003-04-05 1:35 UTC (permalink / raw)
To: akpm; +Cc: netdev
From: Andrew Morton <akpm@digeo.com>
Date: Fri, 4 Apr 2003 17:29:27 -0800
It is a bit worrisome that dev_kfree_skb() is now illegal if local interrupts
are disabled. I'd expect there to be more instances.
Why worrisome? It's an error, and now at least we'll know
about it. This isn't a "is now illegal", it's always been
illegal. You simply cannot take BH locks inside of IRQ disabling
ones, this is true 2.4.x too it just doesn't BUG() on you there
(you get a potential deadlock instead).
This patch should go through the Tulip maintainer (Jeff I guess)
especially since it's a fix that belongs in 2.4.x as well.
Thanks Andrew.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [2.5.66-mm3+debug] traces.
2003-04-05 1:35 ` David S. Miller
@ 2003-04-05 2:14 ` Andrew Morton
2003-04-05 2:27 ` David S. Miller
0 siblings, 1 reply; 4+ messages in thread
From: Andrew Morton @ 2003-04-05 2:14 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev
"David S. Miller" <davem@redhat.com> wrote:
>
> You simply cannot take BH locks inside of IRQ disabling
> ones, this is true 2.4.x too it just doesn't BUG() on you there
> (you get a potential deadlock instead).
I don't recall seeing a description or discussion of this deadlock. Any
pointers?
> This patch should go through the Tulip maintainer (Jeff I guess)
> especially since it's a fix that belongs in 2.4.x as well.
Yup, I'll spam Jeff.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [2.5.66-mm3+debug] traces.
2003-04-05 2:14 ` Andrew Morton
@ 2003-04-05 2:27 ` David S. Miller
0 siblings, 0 replies; 4+ messages in thread
From: David S. Miller @ 2003-04-05 2:27 UTC (permalink / raw)
To: akpm; +Cc: netdev
From: Andrew Morton <akpm@digeo.com>
Date: Fri, 4 Apr 2003 18:14:27 -0800
"David S. Miller" <davem@redhat.com> wrote:
>
> You simply cannot take BH locks inside of IRQ disabling
> ones, this is true 2.4.x too it just doesn't BUG() on you there
> (you get a potential deadlock instead).
I don't recall seeing a description or discussion of this deadlock.
Any pointers?
Search the archives for the subject "BUG or not? GFP_KERNEL with
interrupts disabled." unfortunately it appears a lot of that
thread got privatized and I don't have any logs.
The long and short of it is:
spin_lock_irq();
anything which does local_bh_{enable,disable}()
spin_unlock_irq();
and similar are illegal.
local_bh_enable() invokes do_softirq() if softints are pending,
do_softirq() enables cpu interrupts, which would result in a deadlock
in the above case (because an interrupt could arrive and try to take
the IRQ spinlock taken by spin_lock_irq()).
Therefore it's a bug, and it's a bug everywhere BUG() check or not. :-)
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2003-04-05 2:27 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-04-05 1:29 Fw: [2.5.66-mm3+debug] traces Andrew Morton
2003-04-05 1:35 ` David S. Miller
2003-04-05 2:14 ` Andrew Morton
2003-04-05 2:27 ` David S. Miller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).