New: 3c59x.c: Regression since 2.6.36 (incl.), for e. g., TCP stalls on receive
Subsystems (2)
Quoted 26 lines · On Mon, 7 Feb 2011 18:55:29 GMT [email protected] wrote:
https://bugzilla.kernel.org/show_bug.cgi?id=28542
Summary: 3c59x.c: Regression since 2.6.36 (incl.), for e. g., TCP stalls on receive Product: Drivers Version: 2.5 Kernel Version: 2.6.36—2.6.37 Platform: All OS/Version: Linux Tree: Mainline Status: NEW Severity: high Priority: P1 Component: Network AssignedTo: [email protected] ReportedBy: [email protected] Regression: YesDump of kernel source bz2 download being stalled:
http://poige.livejournal.com/475372.html#cutid1 (see the bunch of ACKs #5555523).As a workaround I've just replaced
drivers/net/3c59x.cwith its former (2.6.35.11) version, the connectivity problems are gone now.
Quoted 2 lines · Andrew Morton
On Mon, 7 Feb 2011 18:55:29 GMT
[email protected] wrote:
[...]
Quoted 3 lines
Quoted 3 lines
Dump of kernel source bz2 download being stalled:
http://poige.livejournal.com/475372.html#cutid1 (see the bunch of ACKs #5555523).
That's not a stall, that's selective-ACKing (see the ranges listed on the right of each ACK line).
Quoted 2 lines
Quoted 2 lines
As a workaround I've just replaced
drivers/net/3c59x.cwith its former (2.6.35.11) version, the connectivity problems are gone now.
This could just be a coincidence. But it could be that packet loss is now more likely for some reason. Does the rx_fifo_errors statistic increase quickly with the later version of the driver? (This seems to be the only receive error type reported.)
Ben.
Signature · 2 lines
Ben Hutchings Once a job is fouled up, anything done to improve it makes it worse.