Re: FIX ME in oxu210p-hcd.c

Re: FIX ME in oxu210p-hcd.c

[nick]

I found a unfixed FIX ME in the file stated in my above message. I am wondering what to set hcd->self.comtroller->dma_mask 
to as it's now been defined to NULL and clearly even as a newbie this seem incorrect.
Regards Nick  

RE: FIX ME in oxu210p-hcd.c

[Peter Chen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="utf-8", Size: 544 bytes --]


 
> Subject: Re: FIX ME in oxu210p-hcd.c
> 
> 
> I found a unfixed FIX ME in the file stated in my above message. I am
> wondering what to set hcd->self.comtroller->dma_mask to as it's now been
> defined to NULL and clearly even as a newbie this seem incorrect.
> Regards Nick

Usually, it is set at its controller driver or pass through through device tree or
platform data.

Peter
ÿôèº{.nÇ+‰·Ÿ®‰­†+%ŠËÿ±éݶ\x17¥Šwÿº{.nÇ+‰·¥Š{±þG«éÿŠ{ayº\x1dʇڙë,j\a­¢f£¢·hšïêÿ‘êçz_è®\x03(­éšŽŠÝ¢j"ú\x1a¶^[m§ÿÿ¾\a«þG«éÿ¢¸?™¨è­Ú&£ø§~á¶iO•æ¬z·švØ^\x14\x04\x1a¶^[m§ÿÿÃ\fÿ¶ìÿ¢¸?–I¥

Re: FIX ME in oxu210p-hcd.c

[nick]

On 14-09-21 07:53 PM, Peter Chen wrote:
> 
>  
>> Subject: Re: FIX ME in oxu210p-hcd.c
>>
>>
>> I found a unfixed FIX ME in the file stated in my above message. I am
>> wondering what to set hcd->self.comtroller->dma_mask to as it's now been
>> defined to NULL and clearly even as a newbie this seem incorrect.
>> Regards Nick
> 
> Usually, it is set at its controller driver or pass through through device tree or
> platform data.
> 
> Peter
> 
Sorry Peter,
I apologize for asking for more help here but I will paste the function below and with my changes.
Please let me known if I am wrong and how to fix it as I new here.
Sorry for Wasting Your Time,
Nick 
static int oxu_reset(struct usb_hcd *hcd)
{
 	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
 	int ret;

 	spin_lock_init(&oxu->mem_lock);
 	INIT_LIST_HEAD(&oxu->urb_list);
	oxu->urb_len = 0;

-	/* FIMXE */
+ 	hcd->self.controller->dma_mask = hcd->regs;

 	if (oxu->is_otg) {
		oxu->caps = hcd->regs + OXU_OTG_CAP_OFFSET;
		oxu->regs = hcd->regs + OXU_OTG_CAP_OFFSET + \
			HC_LENGTH(readl(&oxu->caps->hc_capbase));
 
		oxu->mem = hcd->regs + OXU_SPH_MEM;
	} else {
		oxu->caps = hcd->regs + OXU_SPH_CAP_OFFSET;
 		oxu->regs = hcd->regs + OXU_SPH_CAP_OFFSET + \
 			HC_LENGTH(readl(&oxu->caps->hc_capbase));

 		oxu->mem = hcd->regs + OXU_OTG_MEM;
	}

 	oxu->hcs_params = readl(&oxu->caps->hcs_params);
 	oxu->sbrn = 0x20;

	ret = oxu_hcd_init(hcd);
	if (ret)
 		return ret;
 
 	return 0;
}

RE: FIX ME in oxu210p-hcd.c

[Peter Chen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="utf-8", Size: 2568 bytes --]


 
> >> Subject: Re: FIX ME in oxu210p-hcd.c
> >>
> >>
> >> I found a unfixed FIX ME in the file stated in my above message. I am
> >> wondering what to set hcd->self.comtroller->dma_mask to as it's now
> >> been defined to NULL and clearly even as a newbie this seem incorrect.
> >> Regards Nick
> >
> > Usually, it is set at its controller driver or pass through through
> > device tree or platform data.
> >
> > Peter
> >
> Sorry Peter,
> I apologize for asking for more help here but I will paste the function below and
> with my changes.
> Please let me known if I am wrong and how to fix it as I new here.
> Sorry for Wasting Your Time,

You are welcome

> Nick
> static int oxu_reset(struct usb_hcd *hcd) {
>  	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
>  	int ret;
> 
>  	spin_lock_init(&oxu->mem_lock);
>  	INIT_LIST_HEAD(&oxu->urb_list);
> 	oxu->urb_len = 0;
> 
> -	/* FIMXE */
> + 	hcd->self.controller->dma_mask = hcd->regs;
> 

It is the dma mask, not the register.

Try below patch to see if it works for you:

diff --git a/drivers/usb/host/oxu210hp-hcd.c b/drivers/usb/host/oxu210hp-hcd.c
index da5fb0e..5549851 100644
--- a/drivers/usb/host/oxu210hp-hcd.c
+++ b/drivers/usb/host/oxu210hp-hcd.c
@@ -37,6 +37,7 @@
 #include <linux/moduleparam.h>
 #include <linux/dma-mapping.h>
 #include <linux/io.h>
+#include <linux/dma-mapping.h>
 
 #include <asm/irq.h>
 #include <asm/unaligned.h>
@@ -3841,6 +3842,12 @@ static int oxu_drv_probe(struct platform_device *pdev)
                goto error;
        }
 
+       ret = dma_coerce_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+       if (ret) {
+               dev_err(&pdev->dev, "set dma mask error\n");
+               goto error;
+       }
+
        /* Allocate a driver data struct to hold useful info for both
         * SPH & OTG devices
         */

Peter


>  	if (oxu->is_otg) {
> 		oxu->caps = hcd->regs + OXU_OTG_CAP_OFFSET;
> 		oxu->regs = hcd->regs + OXU_OTG_CAP_OFFSET + \
> 			HC_LENGTH(readl(&oxu->caps->hc_capbase));
> 
> 		oxu->mem = hcd->regs + OXU_SPH_MEM;
> 	} else {
> 		oxu->caps = hcd->regs + OXU_SPH_CAP_OFFSET;
>  		oxu->regs = hcd->regs + OXU_SPH_CAP_OFFSET + \
>  			HC_LENGTH(readl(&oxu->caps->hc_capbase));
> 
>  		oxu->mem = hcd->regs + OXU_OTG_MEM;
> 	}
> 
>  	oxu->hcs_params = readl(&oxu->caps->hcs_params);
>  	oxu->sbrn = 0x20;
> 
> 	ret = oxu_hcd_init(hcd);
> 	if (ret)
>  		return ret;
> 
>  	return 0;
> }
ÿôèº{.nÇ+‰·Ÿ®‰­†+%ŠËÿ±éݶ\x17¥Šwÿº{.nÇ+‰·¥Š{±þG«éÿŠ{ayº\x1dʇڙë,j\a­¢f£¢·hšïêÿ‘êçz_è®\x03(­éšŽŠÝ¢j"ú\x1a¶^[m§ÿÿ¾\a«þG«éÿ¢¸?™¨è­Ú&£ø§~á¶iO•æ¬z·švØ^\x14\x04\x1a¶^[m§ÿÿÃ\fÿ¶ìÿ¢¸?–I¥

Re: FIX ME in oxu210p-hcd.c

[nick]

[-- Attachment #1: Type: text/plain, Size: 3063 bytes --]



On 14-09-21 10:11 PM, Peter Chen wrote:
> 
>  
>>>> Subject: Re: FIX ME in oxu210p-hcd.c
>>>>
>>>>
>>>> I found a unfixed FIX ME in the file stated in my above message. I am
>>>> wondering what to set hcd->self.comtroller->dma_mask to as it's now
>>>> been defined to NULL and clearly even as a newbie this seem incorrect.
>>>> Regards Nick
>>>
>>> Usually, it is set at its controller driver or pass through through
>>> device tree or platform data.
>>>
>>> Peter
>>>
>> Sorry Peter,
>> I apologize for asking for more help here but I will paste the function below and
>> with my changes.
>> Please let me known if I am wrong and how to fix it as I new here.
>> Sorry for Wasting Your Time,
> 
> You are welcome
> 
>> Nick
>> static int oxu_reset(struct usb_hcd *hcd) {
>>  	struct oxu_hcd *oxu = hcd_to_oxu(hcd);
>>  	int ret;
>>
>>  	spin_lock_init(&oxu->mem_lock);
>>  	INIT_LIST_HEAD(&oxu->urb_list);
>> 	oxu->urb_len = 0;
>>
>> -	/* FIMXE */
>> + 	hcd->self.controller->dma_mask = hcd->regs;
>>
> 
> It is the dma mask, not the register.
> 
> Try below patch to see if it works for you:
> 
> diff --git a/drivers/usb/host/oxu210hp-hcd.c b/drivers/usb/host/oxu210hp-hcd.c
> index da5fb0e..5549851 100644
> --- a/drivers/usb/host/oxu210hp-hcd.c
> +++ b/drivers/usb/host/oxu210hp-hcd.c
> @@ -37,6 +37,7 @@
>  #include <linux/moduleparam.h>
>  #include <linux/dma-mapping.h>
>  #include <linux/io.h>
> +#include <linux/dma-mapping.h>
>  
>  #include <asm/irq.h>
>  #include <asm/unaligned.h>
> @@ -3841,6 +3842,12 @@ static int oxu_drv_probe(struct platform_device *pdev)
>                 goto error;
>         }
>  
> +       ret = dma_coerce_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
> +       if (ret) {
> +               dev_err(&pdev->dev, "set dma mask error\n");
> +               goto error;
> +       }
> +
>         /* Allocate a driver data struct to hold useful info for both
>          * SPH & OTG devices
>          */
> 
> Peter
> 
> 
>>  	if (oxu->is_otg) {
>> 		oxu->caps = hcd->regs + OXU_OTG_CAP_OFFSET;
>> 		oxu->regs = hcd->regs + OXU_OTG_CAP_OFFSET + \
>> 			HC_LENGTH(readl(&oxu->caps->hc_capbase));
>>
>> 		oxu->mem = hcd->regs + OXU_SPH_MEM;
>> 	} else {
>> 		oxu->caps = hcd->regs + OXU_SPH_CAP_OFFSET;
>>  		oxu->regs = hcd->regs + OXU_SPH_CAP_OFFSET + \
>>  			HC_LENGTH(readl(&oxu->caps->hc_capbase));
>>
>>  		oxu->mem = hcd->regs + OXU_OTG_MEM;
>> 	}
>>
>>  	oxu->hcs_params = readl(&oxu->caps->hcs_params);
>>  	oxu->sbrn = 0x20;
>>
>> 	ret = oxu_hcd_init(hcd);
>> 	if (ret)
>>  		return ret;
>>
>>  	return 0;
>> }
Unfortunately I can't get it to apply with git apply and get the following message,
 Checking patch drivers/usb/host/oxu210hp-hcd.c...
error: while searching for:
               goto error;
       }

       /* Allocate a driver data struct to hold useful info for both
        * SPH & OTG devices
        */

error: patch failed: drivers/usb/host/oxu210hp-hcd.c:3841
error: drivers/usb/host/oxu210hp-hcd.c: patch does not apply
I will paste the version of my patch file for your troubles.
Regards Nick 

[-- Attachment #2: oxu210hp-hcd.patch --]
[-- Type: text/x-patch, Size: 797 bytes --]

diff --git a/drivers/usb/host/oxu210hp-hcd.c b/drivers/usb/host/oxu210hp-hcd.c
index da5fb0e..5549851 100644
--- a/drivers/usb/host/oxu210hp-hcd.c
+++ b/drivers/usb/host/oxu210hp-hcd.c
@@ -37,6 +37,7 @@
 #include <linux/moduleparam.h>
 #include <linux/dma-mapping.h>
 #include <linux/io.h>
+#include <linux/dma-mapping.h>
 
 #include <asm/irq.h>
 #include <asm/unaligned.h>
@@ -3841,6 +3842,12 @@ static int oxu_drv_probe(struct platform_device *pdev)
                goto error;
        }
 
+       ret = dma_coerce_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+       if (ret) {
+               dev_err(&pdev->dev, "set dma mask error\n");
+               goto error;
+       }
+
        /* Allocate a driver data struct to hold useful info for both
         * SPH & OTG devices
         */

Re: FIX ME in oxu210p-hcd.c

[gregkh]

On Sun, Sep 21, 2014 at 10:03:28PM -0400, nick wrote:
> 
> 
> On 14-09-21 07:53 PM, Peter Chen wrote:
> > 
> >  
> >> Subject: Re: FIX ME in oxu210p-hcd.c
> >>
> >>
> >> I found a unfixed FIX ME in the file stated in my above message. I am
> >> wondering what to set hcd->self.comtroller->dma_mask to as it's now been
> >> defined to NULL and clearly even as a newbie this seem incorrect.
> >> Regards Nick
> > 
> > Usually, it is set at its controller driver or pass through through device tree or
> > platform data.
> > 
> > Peter
> > 
> Sorry Peter,
> I apologize for asking for more help here but I will paste the function below and with my changes.
> Please let me known if I am wrong and how to fix it as I new here.
> Sorry for Wasting Your Time,

Then please do not.  Just because your other email addresses were banned
from lkml, don't keep popping up and bothering people, it's rude, and
will cause this address to be banned as well.

Re: FIX ME in oxu210p-hcd.c

[nick]

On 14-09-21 10:11 PM, gregkh@linuxfoundation.org wrote:
> On Sun, Sep 21, 2014 at 10:03:28PM -0400, nick wrote:
>>
>>
>> On 14-09-21 07:53 PM, Peter Chen wrote:
>>>
>>>  
>>>> Subject: Re: FIX ME in oxu210p-hcd.c
>>>>
>>>>
>>>> I found a unfixed FIX ME in the file stated in my above message. I am
>>>> wondering what to set hcd->self.comtroller->dma_mask to as it's now been
>>>> defined to NULL and clearly even as a newbie this seem incorrect.
>>>> Regards Nick
>>>
>>> Usually, it is set at its controller driver or pass through through device tree or
>>> platform data.
>>>
>>> Peter
>>>
>> Sorry Peter,
>> I apologize for asking for more help here but I will paste the function below and with my changes.
>> Please let me known if I am wrong and how to fix it as I new here.
>> Sorry for Wasting Your Time,
> 
> Then please do not.  Just because your other email addresses were banned
> from lkml, don't keep popping up and bothering people, it's rude, and
> will cause this address to be banned as well.
> 
Sorry Greg,
I don't want to get banned again. I was trying to help out and learn, I was apologizing not
for wasting time as much as for making sure I wasn't wasting maintainer time again. I also 
am coming to the conclusion that my terrible patches were a waste of time and I am trying 
to get back into helping out.
Cheers,
Nick  

Re: FIX ME in oxu210p-hcd.c

[gregkh]

On Sun, Sep 21, 2014 at 10:17:38PM -0400, nick wrote:
> 
> 
> On 14-09-21 10:11 PM, gregkh@linuxfoundation.org wrote:
> > On Sun, Sep 21, 2014 at 10:03:28PM -0400, nick wrote:
> >>
> >>
> >> On 14-09-21 07:53 PM, Peter Chen wrote:
> >>>
> >>>  
> >>>> Subject: Re: FIX ME in oxu210p-hcd.c
> >>>>
> >>>>
> >>>> I found a unfixed FIX ME in the file stated in my above message. I am
> >>>> wondering what to set hcd->self.comtroller->dma_mask to as it's now been
> >>>> defined to NULL and clearly even as a newbie this seem incorrect.
> >>>> Regards Nick
> >>>
> >>> Usually, it is set at its controller driver or pass through through device tree or
> >>> platform data.
> >>>
> >>> Peter
> >>>
> >> Sorry Peter,
> >> I apologize for asking for more help here but I will paste the function below and with my changes.
> >> Please let me known if I am wrong and how to fix it as I new here.
> >> Sorry for Wasting Your Time,
> > 
> > Then please do not.  Just because your other email addresses were banned
> > from lkml, don't keep popping up and bothering people, it's rude, and
> > will cause this address to be banned as well.
> > 
> Sorry Greg,
> I don't want to get banned again. I was trying to help out and learn, I was apologizing not
> for wasting time as much as for making sure I wasn't wasting maintainer time again. I also 
> am coming to the conclusion that my terrible patches were a waste of time and I am trying 
> to get back into helping out.

You were asked to finish the Eudyptula challenge before bothering any
other kernel developers with questions / patches.  Until that happens,
you are in my killfile, now with yet-another-email-address.

greg k-h

Re: FIX ME in oxu210p-hcd.c

[nick]

On 14-09-21 10:24 PM, gregkh@linuxfoundation.org wrote:
> On Sun, Sep 21, 2014 at 10:17:38PM -0400, nick wrote:
>>
>>
>> On 14-09-21 10:11 PM, gregkh@linuxfoundation.org wrote:
>>> On Sun, Sep 21, 2014 at 10:03:28PM -0400, nick wrote:
>>>>
>>>>
>>>> On 14-09-21 07:53 PM, Peter Chen wrote:
>>>>>
>>>>>  
>>>>>> Subject: Re: FIX ME in oxu210p-hcd.c
>>>>>>
>>>>>>
>>>>>> I found a unfixed FIX ME in the file stated in my above message. I am
>>>>>> wondering what to set hcd->self.comtroller->dma_mask to as it's now been
>>>>>> defined to NULL and clearly even as a newbie this seem incorrect.
>>>>>> Regards Nick
>>>>>
>>>>> Usually, it is set at its controller driver or pass through through device tree or
>>>>> platform data.
>>>>>
>>>>> Peter
>>>>>
>>>> Sorry Peter,
>>>> I apologize for asking for more help here but I will paste the function below and with my changes.
>>>> Please let me known if I am wrong and how to fix it as I new here.
>>>> Sorry for Wasting Your Time,
>>>
>>> Then please do not.  Just because your other email addresses were banned
>>> from lkml, don't keep popping up and bothering people, it's rude, and
>>> will cause this address to be banned as well.
>>>
>> Sorry Greg,
>> I don't want to get banned again. I was trying to help out and learn, I was apologizing not
>> for wasting time as much as for making sure I wasn't wasting maintainer time again. I also 
>> am coming to the conclusion that my terrible patches were a waste of time and I am trying 
>> to get back into helping out.
> 
> You were asked to finish the Eudyptula challenge before bothering any
> other kernel developers with questions / patches.  Until that happens,
> you are in my killfile, now with yet-another-email-address.
> 
> greg k-h
> 
Greg K-H,
I really don't want you to get any more upset with me then you already are. The reason I gave up on the
challenge was it was mostly driver based and I wanted to learn more in other areas. If you would like to
discuss with me your concerns about my work on this list and how I can aid more please let me known.
Sorry ,
Nick 

Re: FIX ME in oxu210p-hcd.c

[gregkh]

Damm, didn't configure my kill-file correctly, and this snuck in, so
might as well respond...

On Sun, Sep 21, 2014 at 10:36:18PM -0400, nick wrote:
> On 14-09-21 10:24 PM, gregkh@linuxfoundation.org wrote:
> >> Sorry Greg,
> >> I don't want to get banned again. I was trying to help out and learn, I was apologizing not
> >> for wasting time as much as for making sure I wasn't wasting maintainer time again. I also 
> >> am coming to the conclusion that my terrible patches were a waste of time and I am trying 
> >> to get back into helping out.
> > 
> > You were asked to finish the Eudyptula challenge before bothering any
> > other kernel developers with questions / patches.  Until that happens,
> > you are in my killfile, now with yet-another-email-address.
> > 
> Greg K-H,
> I really don't want you to get any more upset with me then you already are. The reason I gave up on the
> challenge was it was mostly driver based and I wanted to learn more in other areas. If you would like to
> discuss with me your concerns about my work on this list and how I can aid more please let me known.

You stopped so early in the challenge, you really have no idea what it
is "mostly" about, so don't make rash statements like that.

Again, unless you finish it, you will be ignored by me, and probably by
everyone else as well.

greg k-h