Messages in this thread | | | Date | Tue, 22 Oct 2013 18:18:14 -0700 | From | Jon Mason <> | Subject | Re: [PATCH v2 11/13] NTB: convert to dmaengine_unmap_data |
| |
On Tue, Oct 22, 2013 at 06:05:31PM -0700, Dan Williams wrote: > On Tue, Oct 22, 2013 at 4:12 PM, Jon Mason <jon.mason@intel.com> wrote: > > On Tue, Oct 22, 2013 at 02:29:36PM -0700, Dan Williams wrote: > >> On Fri, Oct 18, 2013 at 6:06 PM, Jon Mason <jon.mason@intel.com> wrote: > >> > On Fri, Oct 18, 2013 at 07:35:31PM +0200, Bartlomiej Zolnierkiewicz wrote: > >> >> Use the generic unmap object to unmap dma buffers. > >> >> > >> >> As NTB can be compiled without DMA_ENGINE support add > >> > > >> > Seems like the stubs should be added outside of this patch. > >> > >> I think they are ok here as this is the only driver that uses them. > >> The alternative is a new api patch without a user. > >> > >> > Also, the > >> > comment implies that NTB could not be compiled without DMA_ENGINE > >> > support before, which it could be. > >> > >> Hmm, I read it as "since NTB *can* be compiled without dmaengine here > >> are some stubs". > > > > This poses an overall question of whether it would simply be better to > > abstract all of the with/without DMA_ENGINE part and simply remap it > > to memcpy if DMA_ENGINE is not set (or if the DMA engine is > > hotplugged). Of course, this is outside the scope of this patch. > > That's at least the promise of async_memcpy() it does not care if a > channel is there or not, but I think it is better if the client has a > strict dma and non-dma path. Hiding the dma details from the client > seems to have been the wrong choice at least for raid. > > > That is fine. It can be like this in the short term. > > > > Thanks, > > Jon > > I'll take that as: > > Acked-by: Jon Mason <jon.mason@intel.com>
Begrudgingly-Acked-by: Jon Mason <jon.mason@intel.com>
> > ...but holler if not.
| |