Packet corruption in re0 (and em0 too)
raul at pop.isdefe.es
Fri Feb 22 22:18:08 UTC 2008
Quoting myself after some more testing ...
> the same behavior happens. I left the box by night compiling CURRENT and
> this morning I've run my 'stress tests' (once again with re0) ... the
> box break to gdb as far I remember (too sleepy, sorry). Well, at least
> no garbage over the link BD.
Really sleepy, it was not gdb, just a LOR.
If is anybody interested on it, I'll can reproduce it after a bit more
tests with RELENG_7_0.
> The fastest method I found to reproduce the problem is receive a
> compresed tar (with good compression ratio) over the network (using nc,
> 100 Mbps fdx), pipe through tar that decompress and write it on disk (a
> lot of MB/s and tps looking at 'systat -vm'). The source tar file is
> about 5Gb and link always fails before EOF :|.
That's still true at least for RELENG_7_0 and re0. Nothing new I guess.
I can't reproduce it with em0. Sorry for the noise.
I'm going to try the proposed if_re.c and if_rlreg.h as Christoph is
More information about the freebsd-current