On Tue, Oct 11, 2011 at 12:27:17PM +0200, David Täht wrote:
> On my list already would be "an analysis of the effects of broken sack
> processing on linux 2.4.24-3.1", of which I *think* I've captured
> multiple examples of in the raw traces I've been collecting for
> months... (so if anyone is interested in the raw data, I can provide)
Do you have any more information? The only thing I could find online was that
there were SACK issues that were supposed to be fixed by 2.6.16; nothing
about a fix in 3.1 or post-3.1.

This was a typo on my part - 2.6.24-3.1, not 2.4


The specific commit that concerned me was this one. I had seen multiple cases of odd behavior (basically tcp sacking like crazy until the cwr collapses or a tcp reset) that I *think* this fix explains.



commit f779b2d60ab95c17f1e025778ed0df3ec2f05d75
Author: Zheng Yan <zheng.z.yan@intel.com>
Date:   Sun Sep 18 22:37:34 2011 -0400

    tcp: fix validation of D-SACK
   
    D-SACK is allowed to reside below snd_una. But the corresponding check
    in tcp_is_sackblock_valid() is the exact opposite. It looks like a typo.
   
    Signed-off-by: Zheng Yan <zheng.z.yan@intel.com>
    Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
    Signed-off-by: David S. Miller <davem@davemloft.net>



-- 
Dave Täht