On Sat, Feb 05, 2011 at 05:02:55PM +0000, Russell King - ARM Linux wrote: > On Wed, Jan 12, 2011 at 07:13:58PM +0100, Per Forlin wrote: > > Add support to prepare one MMC request while another is active on > > the host. This is done by making the issue_rw_rq() asynchronous. > > The increase in throughput is proportional to the time it takes to > > prepare a request and how fast the memory is. The faster the MMC/SD is > > the more significant the prepare request time becomes. Measurements on U5500 > > and U8500 on eMMC shows significant performance gain for DMA on MMC for large > > reads. In the PIO case there is some gain in performance for large reads too. > > There seems to be no or small performance gain for write, don't have a good > > explanation for this yet. > > It might be worth seeing what effect the following patch has. This > moves the dsb out of the cache operations into a separate function, > so we only do one dsb per DMA mapping/unmapping operation. That's > particularly significant for the scattergather code. > > I don't remember the reason why this was dropped as a candidate for > merging - could that be because the dsb needs to be before the outer > cache maintainence? Adding Catalin for comment on that. FWIW, trying this with MMC on OMAP4, I see no measurable difference in performance nor CPU usage. -- To unsubscribe from this list: send the line "unsubscribe linux-mmc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
