I have a GBe Via Velocity NIC on an Abit AV8 motherboard, with
1GB DDR and a 3700+ CPU, running kernel 2.6.19. When I push a
lot of data via TCP-mounted NFS, I get a lot of these messages
in the system log and the machine is briefly slow to respond:
eth0: excessive work at interrupt.
I am guessing the system becomes slow when it is 'catching up'
with all the interrupts.
There are a lot of parameters associated with the via-velocity
driver. Is there a reference online detailing how to adjust
them, or a more general network-card kernel parameter tuning
FAQ?
I have txcsum_offload enabled.
rx_copybreak
Copy breakpoint for copy-only-tiny-frames (int)
int_works
Number of packets per interrupt services (array of int)
txcsum_offload
Enable transmit packet checksum offload (array of int)
IP_byte_align
Enable IP header dword aligned (array of int)
DMA_length
DMA length (array of int)
rx_thresh
Receive fifo threshold (array of int)
TxDescriptors
Number of transmit descriptors (array of int)
parm
RxDescriptors
Number of receive descriptors (array of int)
--
Brian Hall
http://pcisys.net/~brihall
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
