On Wed, Aug 30, 2017 at 06:50:34PM +0800, Kinglong Mee wrote: > On 8/30/2017 18:38, Niels de Vos wrote: > > Hi Kinglong, > > > > You have two patches under review for which I would like to know your > > plans: > > > > - mem-pool: New macro GF_CALLOC2/GF_MALLOC2 supports assign release function > > https://review.gluster.org/18092 > > > > - mem-pool: always record mem_header for memory allocated by GF_XALLOC > > https://review.gluster.org/18056 > > > > > > As discussed in the review comments, it would be better to move this > > functionality to api/src/... files. The allocations done by gfapi can > > then add their own (minimal) memory header, that have a pointer to a > > destructor/release function. Not having this in the 'struct mem_header' > > has its benefits. > > > > Do you plan to continue updating this change, or shall I send a new > > version of the patch(es) so that you can review them? > > Yes, I will update this change as adding a separate memory header for gfapi. > I will push the new version this day or tomorrow. Great, thanks! > Maybe we should abandon those two patches, and start a new review for > the new patch? If you only need one patch for the gfapi approach, please re-use 18092 as that one has most of the discussion. You can abandon 18056 in that case. Cheers, Niels _______________________________________________ Gluster-devel mailing list Gluster-devel@xxxxxxxxxxx http://lists.gluster.org/mailman/listinfo/gluster-devel
