Manfred Spraul wrote:
> I assume you twisted the netif_xy_queue() calls:
> if(tx_full)
> stop();
> else
> wake();
yes :)
> Did you check why? Was the tx ring full, but the queue not stopped, or
> something like this? Perhaps the timeout function is called to early?
I did not check why, it simply seemed obvious and logical to stop the
transmit queue at that point. And the fact that adding this check fixes
the problems some drivers have been having does lend support to this.
Alexey tells me I am not getting the softnet stuff quite right, so there
may be more changes. But -- so far testing seemed to be 100% ok for all
the drivers I've fixed in this manner. :)
Jeff
-- Jeff Garzik | Writing software is Building 1024 | more fun than working. MandrakeSoft, Inc. | -Anon- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Feb 15 2000 - 21:00:25 EST