linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Cedric Le Goater <clg@fr.ibm.com>
To: "Ilpo Järvinen" <ilpo.jarvinen@helsinki.fi>
Cc: David Miller <davem@davemloft.net>,
	Andrew Morton <akpm@linux-foundation.org>,
	LKML <linux-kernel@vger.kernel.org>,
	Netdev <netdev@vger.kernel.org>
Subject: Re: 2.6.24-rc4-mm1 - BUG in tcp_fragment
Date: Fri, 14 Dec 2007 07:52:10 +0100	[thread overview]
Message-ID: <4762281A.8000600@fr.ibm.com> (raw)
In-Reply-To: <Pine.LNX.4.64.0712140024360.12376@kivilampi-30.cs.helsinki.fi>

Ilpo Järvinen wrote:
> On Thu, 13 Dec 2007, Cedric Le Goater wrote:
> 
>> I got this one while compiling on NFS.
>>
>> C.
>>
>> kernel BUG at /home/legoater/linux/2.6.24-rc4-mm1/include/net/tcp.h:1480!
> 
> I'm not exactly sure what patches you have applied and which patches are 
> not, with rc4-mm1 there are two patches (first one was incomplete, I 
> assume you had at least that one based on your other mail) to really fix 
> the issues in (__|)tcp_reset_fack_counts(...). 

Yes I only have the first patch you sent on lkml on top of 2.6.24-rc4-mm1.
attached below. I didn't see the second one on lkml ?  

> However, there seems to be so much breakage that I have a bit trouble to 
> decide where to start... The situation seems bit scary :-).

my n/w environment seems to reproduce these issues quite easily. if you
need some testing, just ping me.

Cheers,

C. 

> So, I might soon prepare a revert patch for most of the questionable 
> TCP parts and ask Dave to apply it (and drop them fully during next 
> rebase) unless I suddently figure something out soon which explains 
> all/most of the problems, then return to drawing board. ...As it seems 
> that the cumulative ACK processing problem discovered later on (having 
> rather cumbersome solution with skbs only) will make part of the work 
> that's currently in net-2.6.25 quite useless/duplicate effort. But thanks 
> anyway for reporting these.
> 
> 

Subject: [PATCH] [TCP]: Fix fack_count miscountings (multiple places)

1) Fack_count is set incorrectly if the highest sent skb is
already sacked (the skb->prev won't return it because it's on
the other list already). These manifest as fackets_out counting
error later on, the second-order effects are very hard to track,
so it may fix all out-standing TCP bug reports.

2) Prev == NULL check was wrong way around

3) Last skb's fack count was incorrectly skipped while() {} loop

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
---
 include/net/tcp.h |   22 ++++++++++++++++------
 1 files changed, 16 insertions(+), 6 deletions(-)

diff --git a/include/net/tcp.h b/include/net/tcp.h
index 9dbed0b..11a7e3e 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1337,10 +1337,20 @@ static inline struct sk_buff *tcp_send_head(struct sock *sk)
 static inline void tcp_advance_send_head(struct sock *sk, struct sk_buff *skb)
 {
        struct sk_buff *prev = tcp_write_queue_prev(sk, skb);
+       unsigned int fc = 0;
+
+       if (prev == (struct sk_buff *)&sk->sk_write_queue)
+               prev = NULL;
+       else if (!tcp_skb_adjacent(sk, prev, skb))
+               prev = NULL;
 
-       if (prev != (struct sk_buff *)&sk->sk_write_queue)
-               TCP_SKB_CB(skb)->fack_count = TCP_SKB_CB(prev)->fack_count +
-                                             tcp_skb_pcount(prev);
+       if ((prev == NULL) && !__tcp_write_queue_empty(sk, TCP_WQ_SACKED))
+               prev = __tcp_write_queue_tail(sk, TCP_WQ_SACKED);
+
+       if (prev != NULL)
+               fc = TCP_SKB_CB(prev)->fack_count + tcp_skb_pcount(prev);
+
+       TCP_SKB_CB(skb)->fack_count = fc;
 
        sk->sk_send_head = tcp_write_queue_next(sk, skb);
        if (sk->sk_send_head == (struct sk_buff *)&sk->sk_write_queue)
@@ -1464,7 +1474,7 @@ static inline struct sk_buff *__tcp_reset_fack_counts(struct sock *sk,
 {
        unsigned int fc = 0;
 
-       if (prev == NULL)
+       if (prev != NULL)
                fc = TCP_SKB_CB(*prev)->fack_count + tcp_skb_pcount(*prev);
 
        BUG_ON((*prev != NULL) && !tcp_skb_adjacent(sk, *prev, skb));
@@ -1521,7 +1531,7 @@ static inline void tcp_reset_fack_counts(struct sock *sk, struct sk_buff *inskb)
                skb[otherq] = prev->next;
        }
 
-       while (skb[queue] != __tcp_write_queue_tail(sk, queue)) {
+       do {
                /* Lazy find for the other queue */
                if (skb[queue] == NULL) {
                        skb[queue] = tcp_write_queue_find(sk, TCP_SKB_CB(prev)->seq,
@@ -1535,7 +1545,7 @@ static inline void tcp_reset_fack_counts(struct sock *sk, struct sk_buff *inskb)
                        break;
 
                queue ^= TCP_WQ_SACKED;
-       }
+       } while (skb[queue] != __tcp_write_queue_tail(sk, queue));
 }
 
 static inline void __tcp_insert_write_queue_after(struct sk_buff *skb,
-- 1.5.0.6

  reply	other threads:[~2007-12-14  6:52 UTC|newest]

Thread overview: 112+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-12-05  5:17 2.6.24-rc4-mm1 Andrew Morton
2007-12-05  9:15 ` 2.6.24-rc4-mm1: kobj changes fallout on powerpc Olof Johansson
2007-12-05 13:11   ` Kamalesh Babulal
2007-12-05 15:46     ` Greg KH
2007-12-05 14:12 ` 2.6.24-rc4-mm1 kobject changes broken with hvcs driver " Kamalesh Babulal
2007-12-05 15:47   ` Greg KH
2007-12-06 18:19   ` Balbir Singh
2007-12-06 18:50     ` Greg KH
2007-12-06 18:49       ` Kamalesh Babulal
2007-12-06 18:58       ` Balbir Singh
2007-12-06 19:21         ` Badari Pulavarty
2007-12-07  1:29           ` Balbir Singh
2007-12-06 20:31         ` Greg KH
2007-12-06 23:54           ` Badari Pulavarty
2007-12-07  0:32             ` Greg KH
2007-12-07  3:02               ` Kamalesh Babulal
2007-12-07  5:14                 ` Greg KH
2007-12-07 22:01                   ` Balbir Singh
2007-12-05 23:41 ` 2.6.24-rc4-mm1: hostbyte=0x01 driverbyte=0x00 (now bisected) Alexey Dobriyan
2007-12-06  7:52   ` Hannes Reinecke
2007-12-06 12:08     ` Jens Axboe
2007-12-06 19:19     ` Alexey Dobriyan
2007-12-06  3:15 ` 2.6.24-rc4-mm1 Kernel build fails on S390x Kamalesh Babulal
2007-12-06  7:19   ` Andrew Morton
2007-12-06  6:59 ` 2.6.24-rc4-mm1 Reuben Farrelly
2007-12-06  7:09   ` 2.6.24-rc4-mm1 David Miller
2007-12-07 13:16     ` 2.6.24-rc4-mm1 Ilpo Järvinen
2007-12-12 17:57       ` 2.6.24-rc4-mm1 Cedric Le Goater
2007-12-06  7:35   ` 2.6.24-rc4-mm1 Andrew Morton
2007-12-10 12:24     ` 2.6.24-rc4-mm1 Ilpo Järvinen
2007-12-10 20:05       ` 2.6.24-rc4-mm1 Ilpo Järvinen
2007-12-12 19:21       ` 2.6.24-rc4-mm1 Cedric Le Goater
2007-12-13 17:38         ` tcp_sacktag_one() WARNING (was Re: 2.6.24-rc4-mm1) Cedric Le Goater
2007-12-06 11:49 ` 2.6.24-rc4-mm1 Valdis.Kletnieks
2007-12-06 12:04   ` 2.6.24-rc4-mm1 Andrew Morton
2007-12-06 19:18     ` 2.6.24-rc4-mm1 Valdis.Kletnieks
2007-12-06 19:38       ` 2.6.24-rc4-mm1 Greg KH
2007-12-06 20:04         ` 2.6.24-rc4-mm1 Valdis.Kletnieks
2007-12-06 22:04         ` [dm-devel] 2.6.24-rc4-mm1 Kay Sievers
2007-12-06 22:12           ` Alasdair G Kergon
2007-12-06 23:12           ` Valdis.Kletnieks
2007-12-06 23:24             ` Kay Sievers
2007-12-07 18:20               ` Valdis.Kletnieks
2007-12-07 18:44                 ` Kay Sievers
2007-12-07 20:28                   ` Valdis.Kletnieks
2007-12-07 20:49                     ` Kay Sievers
2007-12-06 22:28 ` 2.6.24-rc4-mm1: VDSOSYM build error Laurent Riffard
2007-12-06 22:37   ` Andrew Morton
2007-12-06 23:28     ` Miles Lane
2007-12-06 23:34       ` Andrew Morton
2007-12-06 23:47         ` Miles Lane
2007-12-07 10:36         ` Ingo Molnar
2007-12-07  1:14     ` [PATCH x86/mm] x86 vDSO: canonicalize sysenter .eh_frame Roland McGrath
2007-12-07  1:27       ` Harvey Harrison
2007-12-07  3:27         ` Miles Lane
2007-12-07  9:44       ` Ingo Molnar
2007-12-07  2:12 ` 2.6.24-rc4-mm1 Dave Young
2007-12-07 22:22   ` 2.6.24-rc4-mm1 Luis R. Rodriguez
2007-12-10  1:07     ` 2.6.24-rc4-mm1 Dave Young
2007-12-09 17:55   ` 2.6.24-rc4-mm1 Nick Kossifidis
     [not found] ` <E1J0Yfy-0001V7-Rr@localhost>
2007-12-07  8:35   ` [PATCH BUGFIX] hid: the `bit' in hidinput_mapping_quirks() is an out parameter Fengguang Wu
2007-12-10 10:03     ` Jiri Kosina
2007-12-07 14:34 ` broken suspend (sched related) [Was: 2.6.24-rc4-mm1] Jiri Slaby
2007-12-07 15:11   ` Ingo Molnar
2007-12-07 17:51     ` Ingo Molnar
2007-12-08  8:10       ` Jiri Slaby
2007-12-08  8:39         ` Ingo Molnar
2007-12-08  9:23           ` Jiri Slaby
2007-12-08 15:24             ` Ingo Molnar
2007-12-08 17:34               ` Jiri Slaby
2007-12-08 17:43               ` Jiri Slaby
2007-12-09  8:06                 ` Ingo Molnar
2007-12-08 23:12               ` Jiri Slaby
2007-12-09  7:46                 ` Ingo Molnar
2007-12-09  9:09                   ` Jiri Slaby
2007-12-10  8:19                   ` Gautham R Shenoy
2007-12-10  8:55                     ` Jiri Slaby
2007-12-10  9:10                       ` Ingo Molnar
2007-12-10 10:15                         ` Gautham R Shenoy
2007-12-10 10:21                           ` Ingo Molnar
2007-12-10 11:08                             ` Gautham R Shenoy
2007-12-10 11:28                               ` Ingo Molnar
2007-12-10 11:49                                 ` Gautham R Shenoy
2007-12-10  9:29                     ` Ingo Molnar
2007-12-07 18:20 ` [PATCH] md: balance braces in raid5 debug code Mariusz Kozlowski
2007-12-08  0:04 ` 2.6.24-rc4-mm1: undefined reference to `compat_sys_timerfd' on sparc64 Mariusz Kozlowski
2007-12-08  0:08   ` Andrew Morton
2007-12-08  9:17     ` Mariusz Kozlowski
2007-12-11 10:15     ` David Miller
2007-12-08 18:20 ` 2.6.24-rc4-mm1: some issues " Mariusz Kozlowski
2007-12-08 18:22   ` Andrew Morton
2007-12-09  8:45     ` David Miller
2007-12-09  9:03       ` Andrew Morton
2007-12-10 14:48 ` 2.6.24-rc4-mm1 Reuben Farrelly
2007-12-10 21:11   ` 2.6.24-rc4-mm1 Andrew Morton
2007-12-11 14:12     ` 2.6.24-rc4-mm1 Reuben Farrelly
2007-12-11 16:20 ` 2.6.24-rc4-mm1 Martin Bligh
2007-12-11 16:59   ` 2.6.24-rc4-mm1 Randy Dunlap
2007-12-11 17:50     ` 2.6.24-rc4-mm1 Martin Bligh
     [not found] ` <33307c790712110813h23def95dvd068b7226e9fcd36@mail.gmail.com>
2007-12-11 20:37   ` 2.6.24-rc4-mm1 Andrew Morton
2007-12-11 21:20     ` 2.6.24-rc4-mm1 Ingo Molnar
2007-12-11 21:26     ` 2.6.24-rc4-mm1 Kok, Auke
2007-12-11 21:59       ` 2.6.24-rc4-mm1 Kok, Auke
2007-12-11 22:10       ` 2.6.24-rc4-mm1 Andrew Morton
2007-12-11 22:17         ` 2.6.24-rc4-mm1 Kok, Auke
2007-12-11 23:15           ` 2.6.24-rc4-mm1 Randy Dunlap
2007-12-12  4:16 ` 2.6.24-rc4-mm1 Rik van Riel
2007-12-13 17:45 ` 2.6.24-rc4-mm1 - BUG in tcp_fragment Cedric Le Goater
2007-12-13 23:00   ` Ilpo Järvinen
2007-12-14  6:52     ` Cedric Le Goater [this message]
2007-12-14 20:14     ` [PATCH net-2.6.25] Revert recent TCP work Ilpo Järvinen
2007-12-16 22:21       ` David Miller

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4762281A.8000600@fr.ibm.com \
    --to=clg@fr.ibm.com \
    --cc=akpm@linux-foundation.org \
    --cc=davem@davemloft.net \
    --cc=ilpo.jarvinen@helsinki.fi \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).