[PATCH v2 01/22] staging/rdma/hfi1: Fix regression in send performance

ira.weiny ira.weiny at intel.com
Mon Oct 26 02:10:05 UTC 2015


On Wed, Oct 21, 2015 at 04:18:06PM +0300, Dan Carpenter wrote:
> On Mon, Oct 19, 2015 at 10:11:16PM -0400, ira.weiny at intel.com wrote:
> > From: Mike Marciniszyn <mike.marciniszyn at intel.com>
> > 
> > This additional call is a regression from qib.  For small messages the progress
> > routine always builds one and clears out the ahg state when the queue has gone
> > to empty which is the predominant case for small messages.
> > 
> > Inline the routine to mitigate the call and move the routine to qp.h for scope
> > reasons.
> > 
> > Reviewed-by: Dennis Dalessandro <dennis.dalessandro at intel.com>
> > Signed-off-by: Mike Marciniszyn <mike.marciniszyn at intel.com>
> > Signed-off-by: Ira Weiny <ira.weiny at intel.com>
> > ---
> 
> The whole point of this patch is to do:
> 
> > -	if (qp->s_sde)
> > +	if (qp->s_sde && qp->s_ahgidx >= 0)
> 
> It was not at all clear from the changelog and it was difficult to tell
> because we moved code around as well.

I've cleaned up the commit message.

v3 should be on its way after I make sure I have addressed all your comments on
all the patches.

Ira

> 
> regards,
> dan carpenter


More information about the devel mailing list