Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 12 May 2005 11:38:59 +0200
From:      "Daniel Eriksson" <>
To:        <>
Cc:        'Paul Saab' <>
Subject:   RE: cvs commit: src/sys/netinet tcp_input.c tcp_output.c tcp_sack.c tcp_var.h
In-Reply-To: <>

Next in thread | Previous in thread | Raw E-Mail | Index | Archive | Help
Paul Saab wrote:

> ps          2005-05-11 21:37:43 UTC
>   FreeBSD src repository
>   Modified files:
>     sys/netinet          tcp_input.c tcp_output.c tcp_sack.c 
>                          tcp_var.h 
>   Log:
>   When looking for the next hole to retransmit from the scoreboard,
>   or to compute the total retransmitted bytes in this sack recovery
>   episode, the scoreboard is traversed. While in sack recovery, this
>   traversal occurs on every call to tcp_output(), every dupack and
>   every partial ack. The scoreboard could potentially get quite large,
>   making this traversal expensive.
>   This change optimizes this by storing hints (for the next hole to
>   retransmit and the total retransmitted bytes in this sack recovery
>   episode) reducing the complexity to find these values from O(n) to
>   constant time.
>   The debug code that sanity checks the hints against the computed
>   value will be removed eventually.

After upgrading one of my servers this morning (dual Athlon MP) I'm getting
a fair amount of these on the console:

tcp_sack_output: Computed sack hole not the same as cached value

So far I have ~50 such lines in the log after ~3 hours of uptime. No users
have reported any problems though.

/Daniel Eriksson