On Thu, Jun 04, 2015 at 03:27:04PM +0200, Maxime Ripard wrote: > Hi Vinod, > > On Mon, May 18, 2015 at 01:46:14PM +0200, Maxime Ripard wrote: > > Hi Vinod, > > > > I know that the memset capabilities have been removed a while back > > because no-one was actually using it. > > > > On the Atmel SAMA5D4, we would really benefit from such a feature in > > the kernel. Indeed, that SoC lacks a GPU, even though having a > > display, so we can actually use the DMA controller to accelerate the > > framebuffer manipulations instead of the GPU or some hardware > > controller like usually done, involving using memset. > > > > The current way that it's done is through an userspace API for > > dmaengine. Maybe that could be moved to the DRM driver for this SoC, I > > don't know and we can argue about this, but the bottom line is that > > for our use case, we do need to have memset back :) > > > > I understand that you might not like it, especially since we're not > > really introducing a new user and even the current code we have based > > on this is far from upstreamable, so we won't even introduce any user > > for that code short term. > > > > Apart from that, that serie is based on the interleaved support for > > that driver I sent earlier today. > > > > Thanks, > > Maxime > > Any comments on that serie? It was on top of queue round the time you pined, but this mail pushed it back more, so now done with this, it looks good to me so have applied it now -- ~Vinod -- To unsubscribe from this list: send the line "unsubscribe dmaengine" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
