netdev - Re: [PATCH] Fix possible double-ack w/ user dma

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Sun, 28 Sep 2008 22:45:51 -0500
From:	Ali Saidi <saidi@...in.umich.edu>
To:	netdev@...r.kernel.org
Subject: Re: [PATCH] Fix possible double-ack w/ user dma

I attached a broken patch. This one is correct, although there are  
probably other ways to fix the issue.

Ali



Download attachment "fix_double_ack.diff" of type "application/octet-stream" (373 bytes)


On Sep 28, 2008, at 9:13 PM, Ali Saidi wrote:

> When TCP receive copy offload is enabled it's possible that   
> tcp_rcv_established() will cause two acks to be sent for a single  
> packet. In the case that a tcp_dma_early_copy() is successful,  
> copied_early is set to true which causes tcp_cleanup_rbuf() to be  
> called early which can send an ack. Further along in  
> tcp_rcv_established(),  __tcp_ack_snd_check() is called and will  
> schedule a delayed ACK. If no packets are processed before the  
> delayed ack timer expires the packet will be acked twice.
>
> Attached is a patch that addresses the problem.
>
> Ali
>
>
>
> <fix_double_ack.diff>