Re: Intel Moorestown EHCI and the HOSTPCx registers

[Matthieu CASTET <matthieu.castet@xxxxxxxxxx>]

Greg KH a écrit :
> On Fri, Jun 22, 2012 at 02:39:19AM +0000, Pan, Jacob jun wrote:
>>
>>> -----Original Message-----
>>> From: Greg KH [mailto:gregkh@xxxxxxxxxxxxxxxxxxx]
>>> Sent: Thursday, June 21, 2012 3:51 PM
>>> To: Pan, Jacob jun
>>> Cc: Alan Stern; Du, Alek; Brown, Len; USB list; Wu, Hao
>>> Subject: Re: Intel Moorestown EHCI and the HOSTPCx registers
>>>
>>> On Thu, Jun 21, 2012 at 09:21:44PM +0000, Pan, Jacob jun wrote:
>>>>
>>>>> -----Original Message-----
>>>>> From: Alan Stern [mailto:stern@xxxxxxxxxxxxxxxxxxx]
>>>>> Sent: Thursday, June 21, 2012 10:25 AM
>>>>> To: Du, Alek; Pan, Jacob jun; Brown, Len
>>>>> Cc: USB list
>>>>> Subject: Intel Moorestown EHCI and the HOSTPCx registers
>>>>>
>>>>> Alek and Jacob added support for the Intel Moorestown EHCI
>>>>> controller, with its special HOSTPCx registers, way back in 2009
>>>>> (commit 331ac6b288d9f3689514ced1878041fb0df7e13c).  That support
>>>>> included a new flag, ehci->has_hostpc, which gets tested in lots of places.
>>>>>
>>>>> But it never gets set anywhere!  All that material is dead code.
>>>>>
>>>>> Was it ever tested?  Should it be removed?  Or should something be
>>> added?
>>>>> What's the story here?
>>>>>
>>>> CC Hao, I am no longer working on this. The registers are valid.
>>> Really?  I thought Moorestown support was removed from the kernel a while
>>> ago, shouldn't this also be cleaned up and removed?
>> The same USB IP block is used in Medfield (the successor of Moorestown).
>> But I am not sure about the latest effort of upstreaming.
> 
> As no in-kernel users are using the code, I recommend deleting it, any
> objection to that?
> 
In this case to be consistent you should remove it from usb/chipidea driver.
hostpc stuff match with lpm stuff in usb/chipidea driver. [1]

Alexander what do you think of this ?

Also some code "PHY low power", could be reused for existing chipidea core, but
of course the code could be removed and added later



[1]
drivers/usb/chipidea/ci.h: * @lpm: set if the device is LPM capable
drivers/usb/chipidea/ci.h:      unsigned        lpm;
drivers/usb/chipidea/core.c:static uintptr_t ci_regs_nolpm[] = {
drivers/usb/chipidea/core.c:static uintptr_t ci_regs_lpm[] = {
drivers/usb/chipidea/core.c:static int hw_alloc_regmap(struct ci13xxx *ci, bool
is_lpm)
drivers/usb/chipidea/core.c:                    (is_lpm ? ci_regs_lpm[i] :
ci_regs_nolpm[i]);
drivers/usb/chipidea/core.c:                    (is_lpm
drivers/usb/chipidea/core.c:                     ? ci_regs_lpm[OP_ENDPTCTRL]
drivers/usb/chipidea/core.c:                     : ci_regs_nolpm[OP_ENDPTCTRL]);
drivers/usb/chipidea/core.c:    ci->hw_bank.lpm  = reg;
drivers/usb/chipidea/core.c:    dev_dbg(ci->dev, "ChipIdea HDRC found, lpm: %d;
cap: %p op: %p\n",
drivers/usb/chipidea/core.c:            ci->hw_bank.lpm, ci->hw_bank.cap,
ci->hw_bank.op);
drivers/usb/chipidea/core.c:            pr_err("lpm = %i", ci->hw_bank.lpm);
drivers/usb/chipidea/host.c:    ehci->has_hostpc = ci->hw_bank.lpm;
drivers/usb/chipidea/udc.c:     return udc->hw_bank.lpm ? hw_read(udc, OP_DEVLC,
DEVLC_PSPD) :
--
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