[ewg] Re: the default CQ moderation parameters patch in OFED 1.3

[Or Gerlitz]

Eli Cohen wrote:
> On Sun, 2008-01-06 at 10:47 +0200, Or Gerlitz wrote:

>> The patch below in OFED 1.3 does not check the device capabilities and hence
>> always fail on non connectx systems. Can you fix it such that we will not
>> get all those "why I we mthca0: failed to modify CQ params prints in the logs".

> I chose to use KERN_INFO to indicate that it is not a sever situation.
> Perhaps I should just remove the message.

I see, however, is there any problem to check the device capabilities 
before issuing the call or to print error message only if the return 
value is not ENOSYS?

>> Other then that and maybe even more important... I understand that it hard codes
>> ipoib to ask for delivery of interrupt only after MAX (16 packets received, 10 us
>> elapsed since first packet received), correct? so every simple ping-pong test that
>> measures IPoIB latency under small packet rate will have now 10us added to its latency?

> I did not notice that this deteriorates ping pong latency. Could you
> check if it does?

Yes, I run a simple datagram sockets ping-pong test on a pair of 
connectx (HW device 25418 FW 2.3.0 SW OFED-1.3-rc1) systems we have here 
and you can see that there's a ~10us different in the latency between 
the case of waiting for 16 frame vs 1 frame to deliver an interrupt.

> # ethtool -C ib1 adaptive-rx on rx-usecs 10 rx-frames 16
> # /tmp/udp_lat -c -i 193.168.80.11
> client_sig_handler:client_counter=26162 in 1 sec, latency=19.111 [usec]

> # ethtool -C ib1 adaptive-rx on rx-usecs 10 rx-frames 1
> # /tmp/udp_lat -c -i 193.168.80.11
> client_sig_handler:client_counter=50667 in 1 sec, latency=9.878 [usec]

Or.