[openib-general] Re: [PATCH] remove in_atomic

Michael S. Tsirkin mst at mellanox.co.il
Sun Aug 7 03:35:58 PDT 2005


Quoting r. Libor Michalek <limichal at cisco.com>:
> Subject: Re: [PATCH] remove in_atomic
> 
> On Mon, Aug 01, 2005 at 09:35:29AM +0300, Michael S. Tsirkin wrote:
> > in_atomic isnt a reliable way to check that we are in an atomic context.
> > Just schedule work, always, since most cq polling is currently done under
> > a spinlock, anyway.
> 
>   I agree this patch is correct, but did you see any performance change?

No, I didnt notice a big change on my hardware.

> With it applied I'm seeing a lot more variance in performance from run
> to run. I believe the differences look to do with timing and how
> frequently we are using source rdma advertisement vs. the sink. I believe
> the slow start algorithm for src_avails needs improvment.
> 
> -Libor

I have some simpler ideas for performance improvement: for example,
avoid deferring send iocb, or poll cq outside the spinlock in thread
context.


-- 
MST



More information about the general mailing list