HP/Compaq (Winbond) SD/MMC reader supported

HP/Compaq (Winbond) SD/MMC reader supported

[Pierre Ossman]

Since most people will search the LKML for drivers I'll post this little 
notice so people can find it.

The Winbond W83L519D SD/MMC card reader found in many of HP/Compaq's 
laptops (x1000, nx7000, nx7010, etc.) can now be used in Linux. The 
driver is still in its early stages and not ready for inclusion in the 
main kernel. But for those brave enough to fiddle around the driver can 
do both reads and writes. The only known problem is that it doesn't like 
some cards (they do not get past the init. so no risk of data loss).

The project web page is at:

http://projects.drzeus.cx/wbsd/

Please post your results on the development mailing list there.

Rgds
Pierre Ossman

Re: HP/Compaq (Winbond) SD/MMC reader supported

[Pavel Machek]

Hi!

> Since most people will search the LKML for drivers I'll post this little 
> notice so people can find it.
> 
> The Winbond W83L519D SD/MMC card reader found in many of HP/Compaq's 
> laptops (x1000, nx7000, nx7010, etc.) can now be used in Linux. The 
> driver is still in its early stages and not ready for inclusion in the 
> main kernel. But for those brave enough to fiddle around the driver can 
> do both reads and writes. The only known problem is that it doesn't like 
> some cards (they do not get past the init. so no risk of data loss).
> 
> The project web page is at:
> 
> http://projects.drzeus.cx/wbsd/
> 
> Please post your results on the development mailing list there.

Hmm, it does something here on my nx5000... It causes 2 "bad parity
from KBD" errors and then freezes boot. But the chip is detected and I
see 

mmc0: W83L51xD id f00 at 0x248 irq 1 dma 0

message. (How do I guess right values for irq? I thought that
interference with keyboard means it uses irq 1, but it is probably not
the case, and default values did not work, too).

I'll try turning off ALSA because it actually freezes machine only
after alsa is loaded.
								Pavel 



-- 
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!

Re: HP/Compaq (Winbond) SD/MMC reader supported

[Pierre Ossman]

Pavel Machek wrote:

>
>
>Hmm, it does something here on my nx5000... It causes 2 "bad parity
>from KBD" errors and then freezes boot. But the chip is detected and I
>see 
>
>mmc0: W83L51xD id f00 at 0x248 irq 1 dma 0
>
>message. (How do I guess right values for irq? I thought that
>interference with keyboard means it uses irq 1, but it is probably not
>the case, and default values did not work, too).
>
>I'll try turning off ALSA because it actually freezes machine only
>after alsa is loaded.
>  
>
You seem to be running an old version of the driver. The ports for the 
driver are also often populated by SuperIO chip. The detection routine 
was a bit optimistic in the earlier versions so it started resetting the 
wrong hardware.

As of writing the latest version is 0.7 and a patch can be downloaded at:

http://projects.drzeus.cx/wbsd/download.php?get=files/wbsd-0.7.patch

Rgds
Pierre

Re: HP/Compaq (Winbond) SD/MMC reader supported

[Pavel Machek]

Hi!

> >Hmm, it does something here on my nx5000... It causes 2 "bad parity
> >from KBD" errors and then freezes boot. But the chip is detected and I
> >see 
> >
> >mmc0: W83L51xD id f00 at 0x248 irq 1 dma 0
> >
> >message. (How do I guess right values for irq? I thought that
> >interference with keyboard means it uses irq 1, but it is probably not
> >the case, and default values did not work, too).
> >
> >I'll try turning off ALSA because it actually freezes machine only
> >after alsa is loaded.
> > 
> >
> You seem to be running an old version of the driver. The ports for the 
> driver are also often populated by SuperIO chip. The detection routine 
> was a bit optimistic in the earlier versions so it started resetting the 
> wrong hardware.
> 
> As of writing the latest version is 0.7 and a patch can be downloaded at:
> 
> http://projects.drzeus.cx/wbsd/download.php?get=files/wbsd-0.7.patch

Hmm, now I get "Unknown hardware (id f00) found at 2e". I'll try to
investigate further.

									Pavel

-- 
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!

Re: HP/Compaq (Winbond) SD/MMC reader supported

[Pavel Machek]

Hi!

> >Hmm, it does something here on my nx5000... It causes 2 "bad parity
> >from KBD" errors and then freezes boot. But the chip is detected and I
> >see 
> >
> >mmc0: W83L51xD id f00 at 0x248 irq 1 dma 0
> >
> >message. (How do I guess right values for irq? I thought that
> >interference with keyboard means it uses irq 1, but it is probably not
> >the case, and default values did not work, too).
> >
> >I'll try turning off ALSA because it actually freezes machine only
> >after alsa is loaded.
> > 
> >
> You seem to be running an old version of the driver. The ports for the 
> driver are also often populated by SuperIO chip. The detection routine 
> was a bit optimistic in the earlier versions so it started resetting the 
> wrong hardware.
> 
> As of writing the latest version is 0.7 and a patch can be downloaded at:
> 
> http://projects.drzeus.cx/wbsd/download.php?get=files/wbsd-0.7.patch

Hmm, if I disable the check (and make id 0xf00 valid), it will freeze
my machine during boot :-(. Where did you get documentation for this
beast?

								Pavel
-- 
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!

Re: HP/Compaq (Winbond) SD/MMC reader supported

[Pierre Ossman]

Pavel Machek wrote:

>
>
>Hmm, if I disable the check (and make id 0xf00 valid), it will freeze
>my machine during boot :-(. Where did you get documentation for this
>beast?
>
>	
>
Is the 0xf00 id the only one you get? If it is a SuperIO chip then 
resetting it will probably cause all sorts of funky problems.
Do you know what SuperIO is used in the machine? And have you tried 
confirming that the card reader is indeed winbond? The easiest way to do 
that is to see if the Windows driver is wbsd.sys.

The documentation was given to me by one of Winbond's resellers (after a 
lot of begging and whining). It only includes documentation for one chip 
and the ID specified in it is not even the same as reported by the chip 
the spec is supposed to be for. Winbond is definitely not on my 
recommended vendors list.

Rgds
Pierre

Re: HP/Compaq (Winbond) SD/MMC reader supported

[Pavel Machek]

Hi!

> >Hmm, if I disable the check (and make id 0xf00 valid), it will freeze
> >my machine during boot :-(. Where did you get documentation for this
> >beast?
> >
> Is the 0xf00 id the only one you get? If it is a SuperIO chip then 

How can I (try to) get other id's? Yes, it seems to be stable across reboots.

> resetting it will probably cause all sorts of funky problems.
> Do you know what SuperIO is used in the machine? And have you tried 
> confirming that the card reader is indeed winbond? The easiest way to do 
> that is to see if the Windows driver is wbsd.sys.

Stefan, could you take a look? I rm -rf'ed my copy of windows :-(.

SuperIO is behind ISA bridge so it can not be deduced from lspci? Or I
may be completely off mark; there's Unknown mass storage controller:
Texas Instruments PCI7420 Flash Media Controller.... Hmm, that seems
like flash.
									Pavel
-- 
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!

Re: HP/Compaq (Winbond) SD/MMC reader supported

[Stefan Seyfried]

Pavel Machek wrote:
> Hi!

> Stefan, could you take a look? I rm -rf'ed my copy of windows :-(.

Me too, but fortunately i exchanged the nx5k with somebody else who 
hasn't :-)

Texas Instruments PCIxx20 Integrated FlashMedia controller
Driver: tifm.sys

Memory range: 90280000-90280fff
IRQ: 18

	Stefan

Re: HP/Compaq (Winbond) SD/MMC reader supported

[Pavel Machek]

Hi!

> >Stefan, could you take a look? I rm -rf'ed my copy of windows :-(.
> 
> Me too, but fortunately i exchanged the nx5k with somebody else who 
> hasn't :-)
> 
> Texas Instruments PCIxx20 Integrated FlashMedia controller
> Driver: tifm.sys
> 
> Memory range: 90280000-90280fff
> IRQ: 18

Hmm, okay, this does not look compatible :-(. Sorry for the noise.

								Pavel
-- 
People were complaining that M$ turns users into beta-testers...
...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl!

Re: HP/Compaq (Winbond) SD/MMC reader supported

[Pierre Ossman]

Pavel Machek wrote:

>Hi!
>
>  
>
>>Is the 0xf00 id the only one you get? If it is a SuperIO chip then 
>>    
>>
>
>How can I (try to) get other id's? Yes, it seems to be stable across reboots.
>
>  
>
You don't. The driver will list all it finds, so if there's only one 
line then that's it.

>>resetting it will probably cause all sorts of funky problems.
>>Do you know what SuperIO is used in the machine? And have you tried 
>>confirming that the card reader is indeed winbond? The easiest way to do 
>>that is to see if the Windows driver is wbsd.sys.
>>    
>>
>
>Stefan, could you take a look? I rm -rf'ed my copy of windows :-(.
>
>SuperIO is behind ISA bridge so it can not be deduced from lspci? Or I
>may be completely off mark; there's Unknown mass storage controller:
>Texas Instruments PCI7420 Flash Media Controller.... Hmm, that seems
>like flash.
>  
>
PCI7420 does indeed support SD/MMC so there's a high probability that 
this is the wrong driver for you. See if TI will release a complete 
spec. for the SD/MMC interface. Then you can start writing your own 
driver ;)

Rgds
Pierre