mmc broken?

mmc broken?

[Steve Sakoman]

I attempted an Overo build this afternoon with
fb3d15c023ff08c879155db630895f38526b95f6.

I set bootargs for rootfs on mmc.  The boot progresses normally and
then hangs waiting for the rootfs to mount.

Where I previously got:

  Waiting for root device /dev/mmcblk0p2...
  mmc0: host does not support reading read-only switch. assuming write-enable.
  mmc0: new SD card at address ee21
  mmcblk0: mmc0:ee21 SU02G 1.89 GiB
   mmcblk0: p1 p2

I now get:

  Waiting for root device /dev/mmcblk0p2...

Has anyone else seen mmc issues with rc7?

Steve

Re: mmc broken?

[Tony Lindgren]

[-- Attachment #1: Type: text/plain, Size: 762 bytes --]

* Steve Sakoman <sakoman@gmail.com> [081205 16:31]:
> I attempted an Overo build this afternoon with
> fb3d15c023ff08c879155db630895f38526b95f6.
> 
> I set bootargs for rootfs on mmc.  The boot progresses normally and
> then hangs waiting for the rootfs to mount.
> 
> Where I previously got:
> 
>   Waiting for root device /dev/mmcblk0p2...
>   mmc0: host does not support reading read-only switch. assuming write-enable.
>   mmc0: new SD card at address ee21
>   mmcblk0: mmc0:ee21 SU02G 1.89 GiB
>    mmcblk0: p1 p2
> 
> I now get:
> 
>   Waiting for root device /dev/mmcblk0p2...
> 
> Has anyone else seen mmc issues with rc7?

I think I did it again while cleaning up.. Can you try this patch?
The name was conflicting with the other MMC omap driver.

Tony

[-- Attachment #2: mmc-fix.patch --]
[-- Type: text/x-diff, Size: 453 bytes --]

diff --git a/arch/arm/plat-omap/devices.c b/arch/arm/plat-omap/devices.c
index 25c6d10..70108f8 100644
--- a/arch/arm/plat-omap/devices.c
+++ b/arch/arm/plat-omap/devices.c
@@ -206,7 +206,7 @@ int __init omap_mmc_add(int id, unsigned long base, unsigned long size,
 	struct resource res[OMAP_MMC_NR_RES];
 	int ret;
 
-	pdev = platform_device_alloc("mmci-omap", id);
+	pdev = platform_device_alloc("mmci-omap-hs", id);
 	if (!pdev)
 		return -ENOMEM;
 

Re: mmc broken?

[Tony Lindgren]

[-- Attachment #1: Type: text/plain, Size: 946 bytes --]

* Tony Lindgren <tony@atomide.com> [081205 16:34]:
> * Steve Sakoman <sakoman@gmail.com> [081205 16:31]:
> > I attempted an Overo build this afternoon with
> > fb3d15c023ff08c879155db630895f38526b95f6.
> > 
> > I set bootargs for rootfs on mmc.  The boot progresses normally and
> > then hangs waiting for the rootfs to mount.
> > 
> > Where I previously got:
> > 
> >   Waiting for root device /dev/mmcblk0p2...
> >   mmc0: host does not support reading read-only switch. assuming write-enable.
> >   mmc0: new SD card at address ee21
> >   mmcblk0: mmc0:ee21 SU02G 1.89 GiB
> >    mmcblk0: p1 p2
> > 
> > I now get:
> > 
> >   Waiting for root device /dev/mmcblk0p2...
> > 
> > Has anyone else seen mmc issues with rc7?
> 
> I think I did it again while cleaning up.. Can you try this patch?
> The name was conflicting with the other MMC omap driver.

Actually now it breaks for earlier omaps, it  needs to be like this
patch instead.

> Tony


[-- Attachment #2: mmc-fix.patch --]
[-- Type: text/x-diff, Size: 600 bytes --]

diff --git a/arch/arm/plat-omap/devices.c b/arch/arm/plat-omap/devices.c
index 25c6d10..2c3c72f 100644
--- a/arch/arm/plat-omap/devices.c
+++ b/arch/arm/plat-omap/devices.c
@@ -204,9 +204,15 @@ int __init omap_mmc_add(int id, unsigned long base, unsigned long size,
 {
 	struct platform_device *pdev;
 	struct resource res[OMAP_MMC_NR_RES];
+	char *name;
 	int ret;
 
-	pdev = platform_device_alloc("mmci-omap", id);
+	if (cpu_class_is_omap1() || cpu_is_omap242x())
+		name = "mmci-omap";
+	else
+		name = "mmci-omap-hs";
+
+	pdev = platform_device_alloc(name, id);
 	if (!pdev)
 		return -ENOMEM;
 

Re: mmc broken?

[Steve Sakoman]

On Fri, Dec 5, 2008 at 4:49 PM, Tony Lindgren <tony@atomide.com> wrote:
> * Tony Lindgren <tony@atomide.com> [081205 16:34]:
>> * Steve Sakoman <sakoman@gmail.com> [081205 16:31]:
>> > I attempted an Overo build this afternoon with
>> > fb3d15c023ff08c879155db630895f38526b95f6.
>> >
>> > I set bootargs for rootfs on mmc.  The boot progresses normally and
>> > then hangs waiting for the rootfs to mount.
>> >
>> > Where I previously got:
>> >
>> >   Waiting for root device /dev/mmcblk0p2...
>> >   mmc0: host does not support reading read-only switch. assuming write-enable.
>> >   mmc0: new SD card at address ee21
>> >   mmcblk0: mmc0:ee21 SU02G 1.89 GiB
>> >    mmcblk0: p1 p2
>> >
>> > I now get:
>> >
>> >   Waiting for root device /dev/mmcblk0p2...
>> >
>> > Has anyone else seen mmc issues with rc7?
>>
>> I think I did it again while cleaning up.. Can you try this patch?
>> The name was conflicting with the other MMC omap driver.
>
> Actually now it breaks for earlier omaps, it  needs to be like this
> patch instead.

Heh, isn't that how it always goes!  Two steps forward, one step back :-)

I tested your first patch and it did indeed solve the issue.  Looks
like your second patch will also work.  I'll verify this later this
evening.

Steve

Re: mmc broken?

[Tony Lindgren]

* Steve Sakoman <sakoman@gmail.com> [081205 17:01]:
> On Fri, Dec 5, 2008 at 4:49 PM, Tony Lindgren <tony@atomide.com> wrote:
> > * Tony Lindgren <tony@atomide.com> [081205 16:34]:
> >> * Steve Sakoman <sakoman@gmail.com> [081205 16:31]:
> >> > I attempted an Overo build this afternoon with
> >> > fb3d15c023ff08c879155db630895f38526b95f6.
> >> >
> >> > I set bootargs for rootfs on mmc.  The boot progresses normally and
> >> > then hangs waiting for the rootfs to mount.
> >> >
> >> > Where I previously got:
> >> >
> >> >   Waiting for root device /dev/mmcblk0p2...
> >> >   mmc0: host does not support reading read-only switch. assuming write-enable.
> >> >   mmc0: new SD card at address ee21
> >> >   mmcblk0: mmc0:ee21 SU02G 1.89 GiB
> >> >    mmcblk0: p1 p2
> >> >
> >> > I now get:
> >> >
> >> >   Waiting for root device /dev/mmcblk0p2...
> >> >
> >> > Has anyone else seen mmc issues with rc7?
> >>
> >> I think I did it again while cleaning up.. Can you try this patch?
> >> The name was conflicting with the other MMC omap driver.
> >
> > Actually now it breaks for earlier omaps, it  needs to be like this
> > patch instead.
> 
> Heh, isn't that how it always goes!  Two steps forward, one step back :-)

:)

> I tested your first patch and it did indeed solve the issue.  Looks
> like your second patch will also work.  I'll verify this later this
> evening.

Pushed it already, seems to work.

Tony