+ to: Linus + cc: Adrian On 2 January 2018 at 10:10, Benjamin Beckmeyer <beckmeyer.b@xxxxxxxxx> wrote: > Hey all, > > I have a problem with mmc flash on an i.mx 25 architecture. Im running it with Linux 4.14.x and the flash is really slow. > I digged a little bit deeper into this problem and it has nothing to do with the driver (sdhci-esdhci-imx.c), because all registers for the SDHCI have the same values in a kernel before 4.14 and in the current 4.14 kernel. > > I figured out that following commit "mmc: Delete bounce buffer handling" is the problem. I compiled my kernel without that patch and everything is running fine. There is a patch "mmc: sdhci: Avoid swiotlb buffer being full" and I hoped this will fix the problem but it doesn't.. But maybe it doesn't fixed it for the imx driver. When we applied the commit "mmc: Delete bounce buffer handling", is was simply because we more or less thought it was dead code. It turned out, that wasn't case and we got some error reported due to it. Those errors *should* be fixed, but perhaps not from a performance regression point view. I have looped in Linus (and Adrian), hopefully we can figure out a good of how to move forward on this topic. > > Hopefully anybody can help me with this problem. But I thought somebody with a better understanding of the mmc core components knows how to handle this problem (maybe somebody from linaro, because the two patches come from them). > > Kind regards, > Benjamin Thanks for reporting! Could you perhaps post some performance measurements done "before" and "after". A guess a simple "dd" test is sufficient, just to get an idea in what range the regression is about? Kind regards Uffe -- 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
