panic in TCP Limited Transmit after RTO

Bill Hacker wbh at conducive.org
Sat Mar 12 09:42:42 PST 2005


Hiten Pandya wrote:

Noritoshi Demizu wrote:

Hi,

 (If this is not the correct mailing list for this kind of report,
  please let me know.  Thank you.)
I am observing the TCP SACK behavior of DragonFlyBSD these days.
Today, I experienced "sent too much" panic in sys/netinet/tcp_input.c
    You are not alone, I've experienced this panic too, just don't
    have a way to reproduce it.  I will give your patch a try.
    Thanks Demizu-san. :-)

        -Hiten
Patch is good news. Had not had panics, but had experienced jamming
of the LAN a few times that blocked the 4.11 BSD and PowerBook from
the ADSL router.  At least DFLY was doing *something* with the packets...
Just dld'ed FreeBSD 6 snapshot.  Has device timeout on the same fxp0 that
5.3 couldn't find at all, but 4.X, Dragonfly, OS/2, WinWoes, Aos/Bluebottle,
and several Lin-variants are quite happy with. Hardly a bleeding-edge NIC.
S'pose FreeBSD 6.X can be counted on to not panic with that MB, PCI bus,
and 'plain vanilla' Intel fxp0 combo......since it can't even FIND it 
post-boot.

Snatching a fresh DragonFly 'LATEST' now.  'A journey of 10,000 li ....'

Thanks!

Bill






More information about the Bugs mailing list