Hi,
resend:
# Some discussion is needed on buffering/pacing. I'd like data that
# the function call overhead from thread_whatever to ospf_write is a
# big difference, when ospf_write itself can have a non-trivial
# function call tree below it. That doesn't seem obviously true?
And why '3'? It seems as a consequence of this, the next patch is:
"ospfd: Limit the number of interfaces serviced at one time"
So then we've added 2 mystic tunables to control the number of packets
sent per invocation of ospf_write and the number of interfaces.
Does this buy a /tonne/ of performance? Cause if not, is it worth the
issues buffering can cause under load?
regards,
Paul
On Fri, 11 Dec 2015, Donald Sharp wrote:
From: Ayan Banerjee <[email protected]>
In the "write" direction added a command to ensure that Quagga
is able to send out K (=3 by default) packets per thread-write.
regards,
--
Paul Jakma | [email protected] | @pjakma | Key ID: 0xD86BF79464A2FF6A
Fortune:
Do you know the difference between a yankee and a damyankee?
A yankee comes south to *_____visit*.
_______________________________________________
Quagga-dev mailing list
[email protected]
https://lists.quagga.net/mailman/listinfo/quagga-dev
