Patchwork network traffic stop with 2.6.29 after ftp put

login
register
mail settings
Submitter Eric Dumazet
Date March 24, 2009, 2:02 p.m.
Message ID <49C8E7FF.7020906@cosmosbay.com>
Download mbox | patch
Permalink /patch/24972/
State Not Applicable
Delegated to: David Miller
Headers show

Comments

Eric Dumazet - March 24, 2009, 2:02 p.m.
Marco Berizzi a écrit :
> Eric Dumazet wrote:
> 
>> Could you please send us :
> 
> just an update: this bug is not related to
> ftp: it appears to me that after some
> traffic (imap, http) linux stop receiving
> any packets.
> 
>> iptables -nvL
> 
> there isn't any iptables rules
> (there isn't even the iptables package
> installed on my box)
>  
>> netstat -s
> 
> Ip:
>     4808 total packets received
>     28 with invalid addresses
>     0 forwarded
>     0 incoming packets discarded
>     4685 incoming packets delivered
>     1226 requests sent out
> Icmp:
>     0 ICMP messages received
>     0 input ICMP message failed.
>     ICMP input histogram:
>     68 ICMP messages sent
>     0 ICMP messages failed
>     ICMP output histogram:
>         destination unreachable: 47
>         echo request: 21
> IcmpMsg:
>         OutType3: 47
>         OutType8: 21
> Tcp:
>     21 active connections openings
>     0 passive connection openings
>     0 failed connection attempts
>     0 connection resets received
>     0 connections established
>     1490 segments received
>     1104 segments send out
>     30 segments retransmited
>     0 bad segments received.
>     9 resets sent
> Udp:
>     4 packets received
>     0 packets to unknown port received.
>     0 packet receive errors
>     24 packets sent
> UdpLite:
> TcpExt:
>     10 TCP sockets finished time wait in fast timer
>     27 delayed acks sent
>     1240 packet headers predicted
>     58 acknowledgments not containing data payload received
>     5 predicted acknowledgments
>     1 congestion windows recovered without slow start after partial ack
>     5 other TCP timeouts
>     4 connections aborted due to timeout
> IpExt:
>     InBcastPkts: 3191
> 
> 
>> cat /proc/slabinfo
> 
> slabinfo - version: 2.1
> # name            <active_objs> <num_objs> <objsize> <objperslab> <pagesperslab> : tunables <limit> <batchcount> <sharedfactor> : slabdata <active_slabs> <num_slabs> <sharedavail>
> isofs_inode_cache      0      0    320   12    1 : tunables    0    0    0 : slabdata      0      0      0
> fat_inode_cache       23     23    344   23    2 : tunables    0    0    0 : slabdata      1      1      0
> fat_cache            170    170     24  170    1 : tunables    0    0    0 : slabdata      1      1      0
> sgpool-128            12     12   2560   12    8 : tunables    0    0    0 : slabdata      1      1      0
> sgpool-64             12     12   1280   12    4 : tunables    0    0    0 : slabdata      1      1      0
> sgpool-32             12     12    640   12    2 : tunables    0    0    0 : slabdata      1      1      0
> sgpool-16             12     12    320   12    1 : tunables    0    0    0 : slabdata      1      1      0
> flow_cache             0      0     80   51    1 : tunables    0    0    0 : slabdata      0      0      0
> ext2_inode_cache    8700   8700    408   10    1 : tunables    0    0    0 : slabdata    870    870      0
> kiocb                 25     25    160   25    1 : tunables    0    0    0 : slabdata      1      1      0
> shmem_inode_cache     10     10    384   10    1 : tunables    0    0    0 : slabdata      1      1      0
> posix_timers_cache      0      0    104   39    1 : tunables    0    0    0 : slabdata      0      0      0
> UNIX                 108    120    384   10    1 : tunables    0    0    0 : slabdata     12     12      0
> UDP-Lite               0      0    480    8    1 : tunables    0    0    0 : slabdata      0      0      0
> xfrm_dst_cache         0      0    288   14    1 : tunables    0    0    0 : slabdata      0      0      0
> RAW                    9      9    448    9    1 : tunables    0    0    0 : slabdata      1      1      0
> UDP                    8      8    480    8    1 : tunables    0    0    0 : slabdata      1      1      0
> tw_sock_TCP           32     32    128   32    1 : tunables    0    0    0 : slabdata      1      1      0
> TCP                   15     15   1056   15    4 : tunables    0    0    0 : slabdata      1      1      0
> blkdev_queue          25     26   1200   13    4 : tunables    0    0    0 : slabdata      2      2      0
> blkdev_requests       38     54    216   18    1 : tunables    0    0    0 : slabdata      3      3      0
> biovec-256            10     10   3072   10    8 : tunables    0    0    0 : slabdata      1      1      0
> biovec-128             0      0   1536   10    4 : tunables    0    0    0 : slabdata      0      0      0
> biovec-64             10     10    768   10    2 : tunables    0    0    0 : slabdata      1      1      0
> sock_inode_cache     124    132    352   11    1 : tunables    0    0    0 : slabdata     12     12      0
> skbuff_fclone_cache     11     11    352   11    1 : tunables    0    0    0 : slabdata      1      1      0
> file_lock_cache       39     39    104   39    1 : tunables    0    0    0 : slabdata      1      1      0
> Acpi-Operand         675    714     40  102    1 : tunables    0    0    0 : slabdata      7      7      0
> Acpi-Namespace       510    510     24  170    1 : tunables    0    0    0 : slabdata      3      3      0
> proc_inode_cache     130    130    312   13    1 : tunables    0    0    0 : slabdata     10     10      0
> sigqueue              28     28    144   28    1 : tunables    0    0    0 : slabdata      1      1      0
> radix_tree_node     3626   3640    296   13    1 : tunables    0    0    0 : slabdata    280    280      0
> bdev_cache            19     19    416   19    2 : tunables    0    0    0 : slabdata      1      1      0
> sysfs_dir_cache     5864   5865     48   85    1 : tunables    0    0    0 : slabdata     69     69      0
> inode_cache          113    126    288   14    1 : tunables    0    0    0 : slabdata      9      9      0
> dentry             13469  13472    128   32    1 : tunables    0    0    0 : slabdata    421    421      0
> buffer_head        13067  13067     56   73    1 : tunables    0    0    0 : slabdata    179    179      0
> vm_area_struct      3893   4462     88   46    1 : tunables    0    0    0 : slabdata     97     97      0
> mm_struct             54     57    416   19    2 : tunables    0    0    0 : slabdata      3      3      0
> signal_cache          60     64    480    8    1 : tunables    0    0    0 : slabdata      8      8      0
> sighand_cache         65     72   1312   12    4 : tunables    0    0    0 : slabdata      6      6      0
> task_struct           67     90    800   10    2 : tunables    0    0    0 : slabdata      9      9      0
> anon_vma             951   1280     16  256    1 : tunables    0    0    0 : slabdata      5      5      0
> idr_layer_cache      177    182    152   26    1 : tunables    0    0    0 : slabdata      7      7      0
> kmalloc-4096          17     24   4096    8    8 : tunables    0    0    0 : slabdata      3      3      0
> kmalloc-2048         126    136   2048    8    4 : tunables    0    0    0 : slabdata     17     17      0
> kmalloc-1024         178    192   1024    8    2 : tunables    0    0    0 : slabdata     24     24      0
> kmalloc-512         1134   1152    512    8    1 : tunables    0    0    0 : slabdata    144    144      0
> kmalloc-256          175    208    256   16    1 : tunables    0    0    0 : slabdata     13     13      0
> kmalloc-128         1171   1344    128   32    1 : tunables    0    0    0 : slabdata     42     42      0
> kmalloc-64          2011   2048     64   64    1 : tunables    0    0    0 : slabdata     32     32      0
> kmalloc-32           974   1024     32  128    1 : tunables    0    0    0 : slabdata      8      8      0
> kmalloc-16          1249   1280     16  256    1 : tunables    0    0    0 : slabdata      5      5      0
> kmalloc-8           2400   2560      8  512    1 : tunables    0    0    0 : slabdata      5      5      0
> kmalloc-192         1222   1260    192   21    1 : tunables    0    0    0 : slabdata     60     60      0
> kmalloc-96           398    630     96   42    1 : tunables    0    0    0 : slabdata     15     15      0
> 
> 
> 

Thanks Marco

You probably have the problem Ingo reported on lkml, please try Herbert fix :

http://marc.info/?l=linux-kernel&m=123790184128396&w=2

net: Fix netpoll lockup in legacy receive path

When I fixed the GRO crash in the legacy receive path I used
napi_complete to replace __napi_complete.  Unfortunately they're
not the same when NETPOLL is enabled, which may result in us
not calling __napi_complete at all.

While this is fishy in itself, let's make the obvious fix right
now of reverting to the previous state where we always called
__napi_complete.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>



--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Marco Berizzi - March 24, 2009, 3:41 p.m.
Eric Dumazet wrote:

> You probably have the problem Ingo reported on lkml, please try Herbert
> fix :
> 
> http://marc.info/?l=linux-kernel&m=123790184128396&w=2

yes this patch fix the problem.
Thanks Eric and thanks Herbert.

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Marco Berizzi - March 25, 2009, 8:10 a.m.
Marco Berizzi wrote:
> Eric Dumazet wrote:
> 
> > You probably have the problem Ingo reported on lkml, please try Herbert
> > fix :
> > 
> > http://marc.info/?l=linux-kernel&m=123790184128396&w=2
> 
> yes this patch fix the problem.
> Thanks Eric and thanks Herbert.

no, it doesn't fix the problem.
This morning I got the same problem
(I was reading the kernel mailing list
on lkml.indiana.edu with firefox)
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Herbert Xu - March 25, 2009, 12:09 p.m.
On Wed, Mar 25, 2009 at 09:10:29AM +0100, Marco Berizzi wrote:
>
> no, it doesn't fix the problem.
> This morning I got the same problem
> (I was reading the kernel mailing list
> on lkml.indiana.edu with firefox)

Please try the full GRO revert for netif_rx that I just posted.

Thanks,
Marco Berizzi - March 26, 2009, 8:52 a.m.
Herbert Xu wrote:

> On Wed, Mar 25, 2009 at 09:10:29AM +0100, Marco Berizzi wrote:
> >
> > no, it doesn't fix the problem.
> > This morning I got the same problem
> > (I was reading the kernel mailing list
> > on lkml.indiana.edu with firefox)
> 
> Please try the full GRO revert for netif_rx that I just posted.

ok, I have reverted the GRO and I have not
got the problem.

Thanks.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch

diff --git a/net/core/dev.c b/net/core/dev.c
index e3fe5c7..523f53e 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2580,24 +2580,26 @@  static int process_backlog(struct napi_struct *napi, int quota)
 	int work = 0;
 	struct softnet_data *queue = &__get_cpu_var(softnet_data);
 	unsigned long start_time = jiffies;
+	struct sk_buff *skb;
 
 	napi->weight = weight_p;
 	do {
-		struct sk_buff *skb;
-
 		local_irq_disable();
 		skb = __skb_dequeue(&queue->input_pkt_queue);
-		if (!skb) {
-			local_irq_enable();
-			napi_complete(napi);
-			goto out;
-		}
 		local_irq_enable();
+		if (!skb)
+			break;
 
 		napi_gro_receive(napi, skb);
 	} while (++work < quota && jiffies == start_time);
 
 	napi_gro_flush(napi);
+	if (skb)
+		goto out;
+
+	local_irq_disable();
+	__napi_complete(napi);
+	local_irq_enable();
 
 out:
 	return work;