Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1532132 > unrolled thread

[PATCH 1/2] usb: dwc3: gadget: Fix full speed mode

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  [PATCH 1/2] usb: dwc3: gadget: Fix full speed mode Roger Quadros <rogerq@ti.com> - 2016-11-29 12:00 +0100
    Re: [PATCH 1/2] usb: dwc3: gadget: Fix full speed mode Felipe Balbi <balbi@kernel.org> - 2016-11-29 13:00 +0100
      Re: [PATCH 1/2] usb: dwc3: gadget: Fix full speed mode Roger Quadros <rogerq@ti.com> - 2016-11-29 13:30 +0100
        Re: [PATCH 1/2] usb: dwc3: gadget: Fix full speed mode Roger Quadros <rogerq@ti.com> - 2016-12-07 13:20 +0100
          Re: [PATCH 1/2] usb: dwc3: gadget: Fix full speed mode Felipe Balbi <balbi@kernel.org> - 2016-12-07 13:50 +0100
            Re: [PATCH 1/2] usb: dwc3: gadget: Fix full speed mode John Youn <John.Youn@synopsys.com> - 2016-12-08 04:10 +0100

#1532132 — [PATCH 1/2] usb: dwc3: gadget: Fix full speed mode

DCFG.DEVSPD == 0x3 is not valid and we need to set
DCFG.DEVSPD to 0x1 for full speed mode.

Signed-off-by: Roger Quadros <rogerq@ti.com>
---
 drivers/usb/dwc3/gadget.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c
index 1dfa56a5f..797e013 100644
--- a/drivers/usb/dwc3/gadget.c
+++ b/drivers/usb/dwc3/gadget.c
@@ -1597,7 +1597,7 @@ static int __dwc3_gadget_start(struct dwc3 *dwc)
 			reg |= DWC3_DCFG_LOWSPEED;
 			break;
 		case USB_SPEED_FULL:
-			reg |= DWC3_DCFG_FULLSPEED1;
+			reg |= DWC3_DCFG_FULLSPEED2;
 			break;
 		case USB_SPEED_HIGH:
 			reg |= DWC3_DCFG_HIGHSPEED;
-- 
2.7.4

[toc] | [next] | [standalone]

#1532217

[Multipart message — attachments visible in raw view] — view raw

Hi,

Roger Quadros <rogerq@ti.com> writes:
> DCFG.DEVSPD == 0x3 is not valid and we need to set
> DCFG.DEVSPD to 0x1 for full speed mode.

seems like it has been made invalid somewhere between 1.73a and
2.60a. Can you figure it out from Documentation why and when it was made
invalid? We might need revision checks here.

-- 
balbi

[toc] | [prev] | [next] | [standalone]

#1532240

Hi,

On 29/11/16 13:51, Felipe Balbi wrote:
> 
> Hi,
> 
> Roger Quadros <rogerq@ti.com> writes:
>> DCFG.DEVSPD == 0x3 is not valid and we need to set
>> DCFG.DEVSPD to 0x1 for full speed mode.
> 
> seems like it has been made invalid somewhere between 1.73a and
> 2.60a. Can you figure it out from Documentation why and when it was made
> invalid? We might need revision checks here.
> 

I'll try to dig out more.
For now from TI DRA7 TRM, 0x3 seems to be FS for serial PHY. (see below)
Do you know if any platform uses that mode?
If we need to support both Full speed modes, how do we specify which
mode we want to set? Some DT parameter?

"0x0: High Speed (HS): 480 Mbit/s - Supported from all
USB controllers
0x1: Full Speed (FS): 12 Mbit/s - Supported from all USB
controllers
0x3: Full Speed (FS): 12 Mbit/s on serial PHY: NOT
SUPPORTED
0x4: Super Speed (SS): 5 Gbit/s - Supported only from
USB1 controller
0x2: Low Speed (LS): 1.5 Mbit/s on serial PHY: NOT
SUPPORTED"

cheers,
-roger

[toc] | [prev] | [next] | [standalone]

#1537711

Hi,

On 29/11/16 14:28, Roger Quadros wrote:
> Hi,
> 
> On 29/11/16 13:51, Felipe Balbi wrote:
>>
>> Hi,
>>
>> Roger Quadros <rogerq@ti.com> writes:
>>> DCFG.DEVSPD == 0x3 is not valid and we need to set
>>> DCFG.DEVSPD to 0x1 for full speed mode.
>>
>> seems like it has been made invalid somewhere between 1.73a and
>> 2.60a. Can you figure it out from Documentation why and when it was made
>> invalid? We might need revision checks here.
>>
> 
> I'll try to dig out more.

I couldn't figure out more information on this. The changelogs in the TRMs
don't capture this change and I don't have access to all TRM versions
so I can't say which version it got changed and why.

Can we please involve someone from Synopsis to provide this information?
Thanks.

cheers,
-roger

[toc] | [prev] | [next] | [standalone]

#1537727

[Multipart message — attachments visible in raw view] — view raw

Hi,

Roger Quadros <rogerq@ti.com> writes:
>>> Roger Quadros <rogerq@ti.com> writes:
>>>> DCFG.DEVSPD == 0x3 is not valid and we need to set
>>>> DCFG.DEVSPD to 0x1 for full speed mode.
>>>
>>> seems like it has been made invalid somewhere between 1.73a and
>>> 2.60a. Can you figure it out from Documentation why and when it was made
>>> invalid? We might need revision checks here.
>>>
>> 
>> I'll try to dig out more.
>
> I couldn't figure out more information on this. The changelogs in the TRMs
> don't capture this change and I don't have access to all TRM versions
> so I can't say which version it got changed and why.
>
> Can we please involve someone from Synopsis to provide this information?
> Thanks.

John, could you help us with this query? We'd like to understand why one
of the FULLSPEED modes got removed. Do we need a revision check or can
we assume that the other mode was never supposed to be used?

thanks

-- 
balbi

[toc] | [prev] | [next] | [standalone]

#1538287

On 12/7/2016 4:44 AM, Felipe Balbi wrote:
> 
> Hi,
> 
> Roger Quadros <rogerq@ti.com> writes:
>>>> Roger Quadros <rogerq@ti.com> writes:
>>>>> DCFG.DEVSPD == 0x3 is not valid and we need to set
>>>>> DCFG.DEVSPD to 0x1 for full speed mode.
>>>>
>>>> seems like it has been made invalid somewhere between 1.73a and
>>>> 2.60a. Can you figure it out from Documentation why and when it was made
>>>> invalid? We might need revision checks here.
>>>>
>>>
>>> I'll try to dig out more.
>>
>> I couldn't figure out more information on this. The changelogs in the TRMs
>> don't capture this change and I don't have access to all TRM versions
>> so I can't say which version it got changed and why.
>>
>> Can we please involve someone from Synopsis to provide this information?
>> Thanks.
> 
> John, could you help us with this query? We'd like to understand why one
> of the FULLSPEED modes got removed. Do we need a revision check or can
> we assume that the other mode was never supposed to be used?
> 

Full speed is 0x1. 0x3 may still work due to how the bits are
checked. But it definitely should be 0x1.

I'm not sure if it was 0x3 before. I still need to confirm whether
that was the case or not and if so why.

Regards,
John

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web