Hi,
On Wed, Oct 31, 2012 at 06:17:02AM +0100, Hebbar, Gururaja wrote:
> On Mon, Oct 29, 2012 at 21:47:19, Balbi, Felipe wrote:
> > Hi,
> >
> > On Mon, Oct 29, 2012 at 06:26:48PM +0530, Hebbar, Gururaja wrote:
> > > HSMMC IP on AM33xx need a special setting to handle High-speed cards.
> > > Other platforms like TI81xx, OMAP4 may need this as-well. This depends
> > > on the HSMMC IP timing closure done for the high speed cards.
> > >
> > > From AM335x TRM (SPRUH73F - 18.3.12 Output Signals Generation)
> > >
> > > The MMC/SD/SDIO output signals can be driven on either falling edge or
> > > rising edge depending on the SD_HCTL[2] HSPE bit. This feature allows
> > > to reach better timing performance, and thus to increase data transfer
> > > frequency.
> > >
> > > There are few pre-requisites for enabling the HSPE bit
> > > - Controller should support High-Speed-Enable Bit and
> > > - Controller should not be using DDR Mode and
> > > - Controller should advertise that it supports High Speed in
> > > capabilities register and
> > > - MMC/SD clock coming out of controller > 25MHz
> > >
> > > Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@xxxxxx>
> > > ---
> > > Rebased on mmc-next (v3.7.0-rc1)
> > > Only Build tested since EDMA support for AM335x is not yet added
> > >
> > > :100644 100644 be76a23... ed271fc... M Documentation/devicetree/bindings/mmc/ti-omap-hsmmc.txt
> > > :100644 100644 8b4e4f2... 346af5b... M arch/arm/plat-omap/include/plat/mmc.h
> > > :100644 100644 54bfd0c... 3fd95cb... M drivers/mmc/host/omap_hsmmc.c
> > > .../devicetree/bindings/mmc/ti-omap-hsmmc.txt | 1 +
> > > arch/arm/plat-omap/include/plat/mmc.h | 1 +
> > > drivers/mmc/host/omap_hsmmc.c | 30 +++++++++++++++++++-
> > > 3 files changed, 31 insertions(+), 1 deletions(-)
> > >
> > > diff --git a/Documentation/devicetree/bindings/mmc/ti-omap-hsmmc.txt b/Documentation/devicetree/bindings/mmc/ti-omap-hsmmc.txt
> > > index be76a23..ed271fc 100644
> > > --- a/Documentation/devicetree/bindings/mmc/ti-omap-hsmmc.txt
> > > +++ b/Documentation/devicetree/bindings/mmc/ti-omap-hsmmc.txt
> > > @@ -19,6 +19,7 @@ ti,dual-volt: boolean, supports dual voltage cards
> > > "supply-name" examples are "vmmc", "vmmc_aux" etc
> > > ti,non-removable: non-removable slot (like eMMC)
> > > ti,needs-special-reset: Requires a special softreset sequence
> > > +ti,needs-special-hs-handling: HSMMC IP needs special setting for handling High Speed
> >
> > can't you do runtime revision detection for this ?
>
> Platforms like AM335x & TI814x both use the same IP.
> But only AM335x needed this bit to be set in order to sustain high
> speed transfers. This was concluded after AM335x HSMMC IP timing
> closure was completed for High speed Cards.
fair enough.
> > > Example:
> > > mmc1: mmc@0x4809c000 {
> > > diff --git a/arch/arm/plat-omap/include/plat/mmc.h b/arch/arm/plat-omap/include/plat/mmc.h
> > > index 8b4e4f2..346af5b 100644
> > > --- a/arch/arm/plat-omap/include/plat/mmc.h
> > > +++ b/arch/arm/plat-omap/include/plat/mmc.h
> > > @@ -126,6 +126,7 @@ struct omap_mmc_platform_data {
> > > /* we can put the features above into this variable */
> > > #define HSMMC_HAS_PBIAS (1 << 0)
> > > #define HSMMC_HAS_UPDATED_RESET (1 << 1)
> > > +#define HSMMC_HAS_HSPE_SUPPORT (1 << 2)
> > > unsigned features;
> > >
> > > int switch_pin; /* gpio (card detect) */
> > > diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c
> > > index 54bfd0c..3fd95cb 100644
> > > --- a/drivers/mmc/host/omap_hsmmc.c
> > > +++ b/drivers/mmc/host/omap_hsmmc.c
> > > @@ -62,6 +62,7 @@
> > >
> > > #define VS18 (1 << 26)
> > > #define VS30 (1 << 25)
> > > +#define HSS (1 << 21)
> > > #define SDVS18 (0x5 << 9)
> > > #define SDVS30 (0x6 << 9)
> > > #define SDVS33 (0x7 << 9)
> > > @@ -89,6 +90,7 @@
> > > #define MSBS (1 << 5)
> > > #define BCE (1 << 1)
> > > #define FOUR_BIT (1 << 1)
> > > +#define HSPE (1 << 2)
> > > #define DDR (1 << 19)
> > > #define DW8 (1 << 5)
> > > #define CC 0x1
> > > @@ -489,6 +491,7 @@ static void omap_hsmmc_set_clock(struct omap_hsmmc_host *host)
> > > struct mmc_ios *ios = &host->mmc->ios;
> > > unsigned long regval;
> > > unsigned long timeout;
> > > + unsigned long clkdiv;
> > >
> > > dev_vdbg(mmc_dev(host->mmc), "Set clock to %uHz\n", ios->clock);
> > >
> > > @@ -496,7 +499,8 @@ static void omap_hsmmc_set_clock(struct omap_hsmmc_host *host)
> > >
> > > regval = OMAP_HSMMC_READ(host->base, SYSCTL);
> > > regval = regval & ~(CLKD_MASK | DTO_MASK);
> > > - regval = regval | (calc_divisor(host, ios) << 6) | (DTO << 16);
> > > + clkdiv = calc_divisor(host, ios);
> > > + regval = regval | (clkdiv << 6) | (DTO << 16);
> >
> > not part of $SUBJECT
>
> The output of the calc_divisor() is reused in the below HSPE checking. So
> instead of calling it twice I have taken the output of that function into
> a variable and reused it.
looks like this could/should go into commit log ;-)
> > > OMAP_HSMMC_WRITE(host->base, SYSCTL, regval);
> > > OMAP_HSMMC_WRITE(host->base, SYSCTL,
> > > OMAP_HSMMC_READ(host->base, SYSCTL) | ICE);
> > > @@ -507,6 +511,27 @@ static void omap_hsmmc_set_clock(struct omap_hsmmc_host *host)
> > > && time_before(jiffies, timeout))
> > > cpu_relax();
> > >
> > > + /*
> > > + * Enable High-Speed Support
> > > + * Pre-Requisites
> > > + * - Controller should support High-Speed-Enable Bit
> > > + * - Controller should not be using DDR Mode
> > > + * - Controller should advertise that it supports High Speed
> > > + * in capabilities register
> > > + * - MMC/SD clock coming out of controller > 25MHz
> > > + */
> > > + if ((mmc_slot(host).features & HSMMC_HAS_HSPE_SUPPORT) &&
> >
> > is this really needed...
>
> As explained above, this is needed only on platforms which requires the
> Bit to be set (currently AM335x). Such platforms will set this
> parameter in it .dts file
>
> >
> > > + (ios->timing != MMC_TIMING_UHS_DDR50) &&
> > > + ((OMAP_HSMMC_READ(host->base, CAPA) & HSS) == HSS)) {
> >
> > ... provided you check device's register ?
>
> As explained above, both TI814x & AM335x support HSS but only AM335x
> require HSPE bit set.
ok, understood. Just one small simplification:
if (a & BIT(n)) will have the same effect as if ((a & BIT(n) == BIT(n))),
so you can simplify that a little bit.
--
balbi
Attachment:
signature.asc
Description: Digital signature
