On Mon, Feb 13, 2012 at 1:16 PM, Jeff Chua <jeff.chua.linux@xxxxxxxxx> wrote: > On Mon, Feb 13, 2012 at 12:41 PM, Jeff Chua <jeff.chua.linux@xxxxxxxxx> wrote: >> I ran tests for frequency that works dd out 1G of data from 8MHz to >> 16MHz, and it seems frequency less than 8.4MHz can only transfer at >> rate of 14.3MB/s and above 8.4MHz, speed goes up to 18.3 MB/s. Above >> 17MHz, errors encounters. >> >> hz1 8000000 73.5029 s 14.3 MB/s >> hz2 8000000 73.4740 s 14.3 MB/s >> hz3 8000000 73.3982 s 14.3 MB/s >> hz1 8200000 71.9677 s 14.6 MB/s >> hz2 8200000 73.4936 s 14.3 MB/s >> hz3 8200000 73.3462 s 14.3 MB/s >> hz1 8400000 56.6296 s 18.5 MB/s >> hz2 8400000 56.7616 s 18.5 MB/s >> hz3 8400000 56.7926 s 18.5 MB/s >> hz1 8600000 56.6622 s 18.5 MB/s >> hz2 8600000 56.7201 s 18.5 MB/s >> hz3 8600000 56.6476 s 18.5 MB/s >> hz1 8800000 56.8846 s 18.4 MB/s >> hz2 8800000 56.8226 s 18.5 MB/s >> hz3 8800000 56.7070 s 18.5 MB/s > > I've turned on write, and there's no error writing. But speed is > slower than accessing it via USB2 SD card reader. Same SD. > Via SD PCI 18.779 sec 4.04 %cpu Via USB2.0 Card Reader 12.205 sec 2.04 %cpu Via USB3.0 Card Reader 11.427 sec 2.01 %cpu Shouldn't PCI SD be faster? Same SD used for all tests. THanks, Jeff Thanks, Jeff -- 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
