Anybody with more insights? Why is the interval between two CMD24
more or less in the range of 6ms and the CMD13 follows in varying
intervals between 1ms to 4ms? What's keeping the driver from bursting
CMD24 and successive CMD13 requests?
A 4*128KB dd run shows the following patterns of latency between
various CMDs issued:
http://i43.tinypic.com/2bcmsw.png
The latency distribution is plotted here:
http://i41.tinypic.com/9gvbqo.png
bin limits freq count pct
<0ms 0 0.0%
<1ms 922 43.6%
<6ms 491 23.2%
<10ms 411 19.4%
<20ms 113 5.3%
<100ms 163 7.7%
<300ms 9 0.4%
<1000ms 7 0.3%
It took 25 seconds to write 4*128KB to the SD-card using the current
drivers. More than 50% of the CMD requests issued take more than 1ms
from issuing to response between the host and the card. Anybody else
with this issue on the i.MX25? How can I pinpoint this down to the
culprits in the kernel source?
Regards
--
Joan C. Abelaira
--
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
