On 08/27/2015 07:33 PM, Jason Gunthorpe wrote: > On Wed, Aug 26, 2015 at 12:43:15PM -0400, Doug Ledford wrote: > >> That still takes us back to the fact that the locking changes are >> unneeded. I'm not opposed to them, but as you mentioned in your first >> email, they should go with the changes that require them, and none of >> the changes in the first patch require them. Which means that if we >> want to keep them, it might be worth splitting them out and giving them >> their own patch with an explanation of why they are a benefit (lightly >> contended code, saves a release/reacquire on the failure path). > > Lets just drop them, the cost for restructing was an added empty lock > grab on a non-error path. I've reworked the patch to not perform any locking changes and applied the result. -- Doug Ledford <dledford@xxxxxxxxxx> GPG KeyID: 0E572FDD
Attachment:
signature.asc
Description: OpenPGP digital signature