[PATCH] pcf50606: Fix wrong indexing in show_chgstate

Andy Green andy at openmoko.com
Fri Jan 25 15:40:34 CET 2008

Hash: SHA1

Somebody in the thread at some point said:
> Andy Green wrote:
>> It's wasting like 80+ bytes like that in the sparsely populated array of
>> pointers to names.
> Because of the same problem, we still have a wastage of about 4MB in
> the GTA01 SD/MMC driver, so who cares about 80 bytes ? ;-)

80 bytes here, 4MB there... pretty soon you're talking real memory ;-)
Well the Glamo MMC driver doesn't pull anything like that.

> I think I've missed your answers to my questions about the SD/MMC
> voltage calculation. Is there a reason why, in your patch, the
> sequence of values seems to be 100mV too low in general, and have a
> discontinuity at the upper end ?

The tolerances on these devices are typ +/- 10%, 100mV down saves power
(==VI) and doesn't make problems until we find a card that asks for 1.0V
Vdd, which I didn't hear about.  The micro SDs I have all ask for 3.3V
and get 3.2V.  No card asks for > 3.3V anyway.  If it really irks you
I'll patch it.

> If the algorithm you have there is good, we can share it with GTA01
> as well, nicely solving the above problem.

I think GTA01 MMC on the S3C bus is what we can call "legacy", GTA02 has
the MMC via Glamo.  But I guess the fact is we carry that 4MB around
with our GTA02 kernel too...

- -Andy
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org


More information about the openmoko-kernel mailing list