On Tue, 2012-05-01 at 16:10 -0700, Alexander Duyck wrote:

On 05/01/2012 03:58 PM, Alexander Duyck wrote:

quoted

Eric,

I think I have found a bug, although it is not specific to this patch
but it is related.  It looks like the TCP coalesce code is causing
tcpdump to fail when using frags.  Based on the comments in the patch I
am assuming you have an ixgbe adapter to test with as that is what I
reproduced this on.  To reproduce the issue all you need to do is run
"tcpdump -i ethX > /dev/null" on one console, and on a second console
run a netperf TCP_MAERTS test to some other server.  Tcpdump will exit
out with a message about bad address like this:

tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on eth1, link-type EN10MB (Ethernet), capture size 96 bytes
tcpdump: pcap_loop: recvfrom: Bad address
682 packets captured
2357 packets received by filter
1543 packets dropped by kernel


A bisect of the issue tracked it down to:

1402d366019fedaa2b024f2bac06b7cc9a8782e1 is first bad commit
commit 1402d366019fedaa2b024f2bac06b7cc9a8782e1
Author: Eric Dumazet [off-list ref]
Date:   Mon Apr 23 07:11:42 2012 +0000

    tcp: introduce tcp_try_coalesce
    
    commit c8628155ece3 (tcp: reduce out_of_order memory use) took care of
    coalescing tcp segments provided by legacy devices (linear skbs)
    
    We extend this idea to fragged skbs, as their truesize can be heavy.
    
    ixgbe for example uses 256+1024+PAGE_SIZE/2 = 3328 bytes per segment.
    
    Use this coalescing strategy for receive queue too.
    
    This contributes to reduce number of tcp collapses, at minimal cost, and
    reduces memory overhead and packets drops.
    
    Signed-off-by: Eric Dumazet [off-list ref]
    Cc: Neal Cardwell [off-list ref]
    Cc: Tom Herbert [off-list ref]
    Cc: Maciej Żenczykowski [off-list ref]
    Cc: Ilpo Järvinen [off-list ref]
    Acked-by: Neal Cardwell [off-list ref]
    Signed-off-by: David S. Miller [off-list ref]

:040000 040000 8ca3e0b4e6c6a8f375fd800069d24203880623f3 2576d34c5c9cfc717a11e2ebe054143956716b93 M	net

I suspect we are dealing with either a shared or cloned skb in this
case, though I haven't verified which it is yet.

Thanks,

Alex

One additional note.  It looks like LRO and GRO need to be disabled to
trigger the bug.  If either of them are enabled it doesn't seem to
occur.  Likely due to the fact that they are doing the coalescing before
it gets up to the tcp_try_coalesce call.

Thanks Alex, I'll take a look.

It seems my tcpdump is different than yours.