Hello Sergei,
added
Sekhar Nori <nsekhar@xxxxxx>
Kevin Hilman <khilman@xxxxxx>
to cc ...
Sergei Shtylyov wrote:
> Hello.
>
> On 11.11.2011 11:19, Felipe Balbi wrote:
>
>>> I try to bring up usb 2.0 support on an am1808 based
>>> board and Linux version 3.1.0-rc10 and I am facing
>>> some issues:
>>>
>>> - in arch/arm/mach-davinci/usb.c in
>>> static struct musb_hdrc_platform_data usb_data
>>> ".mode" var get only initialized if some of the
>>> CONFIG_USB_MUSB_* config options are set, but your
>>>
>>> commit 622859634a663c5e55d0e2a2cdbb55ac058d97b3
>>> Author: Felipe Balbi<balbi@xxxxxx>
>>> Date: Wed Jun 22 17:28:09 2011 +0300
>>>
>>> usb: musb: drop a gigantic amount of ifdeferry
>>>
>>> dropped a lot of ifdefs, also the USB_MUSB_HOST,
>>> USB_MUSB_PERIPHERAL and USB_MUSB_OTG defines, which
>>> are used in arch/arm/mach-davinci/usb.c!
>>>
>>> So I think, this ".mode" var is used uninitialized,
>>> or miss I something?
>
> In that case it's initialized to 0 by default (as this is a static
> variable) -- which corresponds to MUSB_UNDEFINED. That's not a good
> value indeed.
Ok, so we must do here something ...
>>> I made a patch, which sets this var to MUSB_OTG
>>> which should be the right value ... should I post
>>> it? (BTW: in arch/arm/mach-davinci/board-da830-evm.c
>>> is also an CONFIG_USB_MUSB_HOST define, which
>>> should be cleaned up)
>> sure, that needs to be applied.
>
> No, that's not a correct patch, I'm afraid...
... yes, I see, because the DM644x EVM board has OTG incapable
host only connector ...
>>
>>> - If I set ".mode = MUSB_OTG" and set in the syscfg2
>>> soc register ~CFGCHIP2_OTGMODE a memstick is not
>>> detected.
>
> Have you loaded the gadget driver? You have to in the OTG mode --
> else the USB host won't initialize.
Yes. It seems to work, if I made a timeout in the code (see
below) ... but have no idea why I need this timeout...
>>> But if I add in drivers/usb/musb/da8xx.c
>>> da8xx_musb_interrupt() a 10ms delay befor the
>>> "if (status& (DA8XX_INTR_DRVVBUS<< DA8XX_INTR_USB_SHIFT)) {"
>>> line usb works fine!
>>> It also works without this timeout if I change the
>>> "dev_dbg(musb->controller, "USB IRQ %08x\n", status);"
>>> to
>>> "printk(musb->controller, "USB IRQ %08x\n", status);"
>>>
>>> -> some timing issue, but I have no idea why!
>>> (Maybe you have an idea?)
>> unfortunately I don't know any details about DaVinci.
>
> AM1808 is not exactly DaVinci, to be precise...
>
>> Sergei, are you able to help on this question ?
>
> Maybe. :-)
I hope it ;-)
>>
>>> - We use on the board host only mode, so i tried the
>>> following:
>>>
>>> set syscfg2 |= CFGCHIP2_FORCE_HOST in board code,
>>> and in arch/arm/mach-davinci/usb.c
>>>
>>> static struct musb_hdrc_platform_data usb_data = {
>>> - .mode = MUSB_OTG,
>>> + .mode = MUSB_HOST,
>>>
>>> -> USB works (without the above timeout) fine!
>
> Well, I'm not surprised...
... Oh, why?
>>> Do you have any idea?
>>>
>>> I think, the issue with the CONFIG_USB_MUSB_* and
>>> the not setup var ".mode" in musb_hdrc_platform_data
>>> must be fixed ... and I think MUSB_OTG should be the right
>>> value ... or?
>> yes, the MUSB IP is always OTG-capable and if that's not working on
>> DaVinci, the root cause needs to be found and fixed.
>
> IIRC, DaVinci DM644x EVM board has OTG incapable host only
> connector, hence that code.
Ok, so we need something like the
commit 622859634a663c5e55d0e2a2cdbb55ac058d97b3
Author: Felipe Balbi <balbi@xxxxxx>
Date: Wed Jun 22 17:28:09 2011 +0300
usb: musb: drop a gigantic amount of ifdeferry
deleted here?
Hmm... just tried this on the am1808 eval board TMDXEXP1808L...
same issue with kernel 3.1.0-rc10! I need this 10ms timeout and
force host mode to get my memstick working with 3.1.0-rc10 ...
tried the kernel from:
$ git clone git://arago-project.org/git/projects/linux-omapl1.git
$ cd linux-omapl1/
$ make da850_omapl138_defconfig
$ make -s uImage
->
Uncompressing Linux... done, booting the kernel.
Linux version 2.6.33-rc4 (hs@xxxxxxxxxxxxxx) (gcc version 4.5.1 (GCC) ) #1 PREEMPT Mon Nov 14 12:00:50 CET 2011
CPU: ARM926EJ-S [41069265] revision 5 (ARMv5TEJ), cr=00053177
CPU: VIVT data cache, VIVT instruction cache
Machine: DaVinci DA850/OMAP-L138/AM18xx EVM
Memory policy: ECC disabled, Data cache writeback
DaVinci da850/omap-l138 variant 0x0
and kernel from:
http://arago-project.org/git/projects/?p=linux-davinci.git;a=summary
Uncompressing Linux... done, booting the kernel.
Linux version 2.6.37+ (hs@xxxxxxxxxxxxxx) (gcc version 4.5.1 (GCC) ) #1 PREEMPT Mon Oct 10 07:56:07 CEST 2011
CPU: ARM926EJ-S [41069265] revision 5 (ARMv5TEJ), cr=00053177
CPU: VIVT data cache, VIVT instruction cache
Machine: DaVinci DA850/OMAP-L138/AM18x EVM
Memory policy: ECC disabled, Data cache writeback
DaVinci da850/omap-l138 variant 0x0
With both kernel images a memstick gets detected on the
am1808 eval board ... so I think, there is no hw problem.
I dig into this ... ideas are welcome!
bye,
Heiko
--
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
