[openib-general] Re: PATCH] mthca - command interface - revised

Michael S. Tsirkin mst at mellanox.co.il
Wed Feb 15 13:42:19 PST 2006


Quoting r. Roland Dreier <rdreier at cisco.com>:
> Subject: Re: PATCH] mthca - command interface - revised
> 
>     Michael> AFAIK, which of the two options gives better performance
>     Michael> might depend on the application and the specific system.
>     Michael> For now, Eli made the simpler option the default.
> 
> Have you seen cases where using the HCR is faster?  It seems that in
> both cases we are doing posted writes to PCI memory, except that the
> HCR case has to do at least one (slow) read to check the go bit.  The
> doorbell case does use more write barriers since all the writes have
> to be ordered, but I have a hard time believing that the write
> barriers are anywhere near as expensive as the read of the go bit.

We did not yet figure it out why.
Possibly some systems slow down if you do a lot of PIO writes in a burst.

-- 
Michael S. Tsirkin
Staff Engineer, Mellanox Technologies



More information about the general mailing list