RE: [PATCH 3/3] mmc: omap_hsmmc: enable SDIO card interrupts

["Madhusudhan" <madhu.cr@ti.com>]

> -----Original Message-----
> From: Paul Walmsley [mailto:paul@pwsan.com]
> Sent: Thursday, February 18, 2010 2:21 PM
> To: Madhusudhan
> Cc: 'David Vrabel'; sakoman@gmail.com; r-woodruff2@ti.com; sawant@ti.com;
> linux-mmc@vger.kernel.org; linux-omap@vger.kernel.org
> Subject: RE: [PATCH 3/3] mmc: omap_hsmmc: enable SDIO card interrupts
> 
> Hi,
> 
> (cc'ing Steve, Richard, Anand)
> 
> On Thu, 18 Feb 2010, Madhusudhan wrote:
> 
> > Not to add further confusion but as far as I understand SDIO should be
> > able to asynchronously wakeup OMAP regardless of MMC clocks being OFF .
> > As per the SDIO spec DAT1 line is used for signaling the card interrupt
> > to the host. The TRM clearly shows a GPIO connected to DAT1 line for
> > wakeup through the SDIO interrupt. Wouldn't this help?
> 
> Is the integration code in arch/arm/*omap* set to enable wakeup on the
> DAT1 ball on the OMAP?  I don't see any trace of that in the code.  That
> would surely prevent wakeup from succeeding on an off-mode enabled kernel.
> Sounds like there may be other problems.  Why not give it a try with an
> SDIO card and see if you can get it to work?
> 
> > Definitely these are good features that are being added. Just that there
> > are confusions around and not everyone has a setup to validate these
> > features.
> 
> Hopefully someone inside TI is validating SDIO wakeup on Linux?
> 
> > If I look in the mainline kernel today there is only one SDIO card
> > driver " sdio_uart.c".
> 
> [paul@twilight current]$ find . -name "*sdio*c" -print | fgrep -v /core/
> ./drivers/ssb/sdio.c
> ./drivers/bluetooth/btsdio.c
> ./drivers/bluetooth/btmrvl_sdio.c
> ./drivers/mmc/host/mvsdio.c
> ./drivers/net/wimax/i2400m/sdio-rx.c
> ./drivers/net/wimax/i2400m/sdio-tx.c
> ./drivers/net/wimax/i2400m/sdio.c
> ./drivers/net/wimax/i2400m/sdio-fw.c
> ./drivers/net/wireless/iwmc3200wifi/sdio.c
> ./drivers/net/wireless/libertas/if_sdio.c
> ./drivers/net/wireless/wl12xx/wl1251_sdio.c
> ./drivers/net/wireless/b43/sdio.c
> ./drivers/media/dvb/siano/smssdio.c
> [paul@twilight current]$
> 
> There appear to be least seven SDIO card drivers in 2.6.34-rc7. At least
> one of these is for a TI chip - the wl1251.  I think some of the
> Gumstix Overo OMAP3 boards are using an SDIO-connected Marvell Libertas:
> 
> http://www.gumstix.com/store/catalog/product_info.php?products_id=252
> 
> David is probably testing with a Bluetooth card - maybe he can comment
> further.
> 
> > And at least I don't see a way I can test any of these features myself.
> 
> Could you clarify?
> 
I meant that right now I am not equipped with a setup that has a SDIO device
that works to a level that this wakeup feature can be validated.

Regards,
Madhu
> 
> - Paul