Messages in this thread | | | Date | Sun, 12 Oct 2008 17:58:53 +0200 | From | Pierre Ossman <> | Subject | Re: [PATCH v2 4/7] atmel-mci: support multiple mmc slots |
| |
On Sun, 12 Oct 2008 15:13:44 +0200 Haavard Skinnemoen <haavard.skinnemoen@atmel.com> wrote:
> Pierre Ossman <drzeus-list@drzeus.cx> wrote: > > On Sun, 5 Oct 2008 18:21:27 +0200 > > Haavard Skinnemoen <haavard.skinnemoen@atmel.com> wrote: > > > > > + /* Calculate clock divider */ > > > + clkdiv = DIV_ROUND_UP(host->bus_hz, 2 * clock_min) - 1; > > > if (clkdiv > 255) { > > > dev_warn(&mmc->class_dev, > > > "clock %u too slow; using %lu\n", > > > - ios->clock, host->bus_hz / (2 * 256)); > > > + clock_min, host->bus_hz / (2 * 256)); > > > clkdiv = 255; > > > } > > > > > > > Has this ever happened, or is it just a bit defensive? :) > > > > (not that there is something wrong with that) > > It can happen if the controller is hooked up to a fast bus and fOD is > slow for whatever reason. And if it does, clkdiv will wrap around and > end up at a rate much faster than requested, so saturating it at the > maximum divider is the safe thing to do. >
Isn't this calculated for f_min though so that the MMC core will never request such a frequency?
Rgds -- -- Pierre Ossman
Linux kernel, MMC maintainer http://www.kernel.org rdesktop, core developer http://www.rdesktop.org
WARNING: This correspondence is being monitored by the Swedish government. Make sure your server uses encryption for SMTP traffic and consider using PGP for end-to-end encryption. [unhandled content-type:application/pgp-signature] | |