At 08:57 AM 1/26/2007, Carlo Palma wrote:
>It works for my SD card!
>ok maybe it's not _the_ solution.. but it let me to use the h1940 with the
>suspend support!
>many many thanks Tom!
Ok, so far so good. I experimented switching between rtp's "new_mmc"
driver and the old one, and it seems clear there are two bugs.
The good news is that the new driver fixes the speed issue, using this
driver, my slower card is working fine at 16.875MHz.
The bad news is that the mmc issue is the core bug. The second SD_SWITCH
command sends 0x80fffff1 (switch to highspeed mode) to any card that
reports it supports highspeed mode. Problem is, the h1940's slot doesn't
cope with this, and the card instantly is taken offline. Oops.
I'll keep poking around, but I suspect we need to talk to the mmc
maintainers. As a workaround, I guess we could hack the s3cmci driver
to suppress the switch request. It's only the second one that causes
the problem, the first one (0x00fffff1) is just testing the capability. (the
high bit 0x80 is the "set" request, the low nibble 0x01 is highspeed mode.)
Anyway, the bottom line is the new s3cmci driver is working well.
I recommend it because it doesn't limit the speed as severely. Is
16.875MHz really the top speed of the slot though? I guess that
might be a limitation of the h1940, this is 1/4 of the 67.5MHz pclk.
Tom.
Received on Sat Jan 27 2007 - 11:28:11 EST
This archive was generated by hypermail 2.2.0 : Sat Jan 27 2007 - 11:28:37 EST