diff mbox series

usb: typec: altmodes/displayport: correct pin assignment for UFP receptacles

Message ID 20220727110503.5260-1-macpaul.lin@mediatek.com (mailing list archive)
State New, archived
Headers show
Series usb: typec: altmodes/displayport: correct pin assignment for UFP receptacles | expand

Commit Message

Macpaul Lin July 27, 2022, 11:05 a.m. UTC
From: Pablo Sun <pablo.sun@mediatek.com>

From: Pablo Sun <pablo.sun@mediatek.com>

Fix incorrect pin assignment values when connecting to a monitor with
Type-C receptacle instead of a plug.

According to specification, an UFP_D receptacle's pin assignment
should came from the UFP_D pin assignments field (bit 23:16), while
an UFP_D plug's assignments are described in the DFP_D pin assignments
(bit 15:8) during Mode Discovery.

For example the LG 27 UL850-W is a monitor with Type-C receptacle.
The monitor responds to MODE DISCOVERY command with following
DisplayPort Capability flag:

        dp->alt->vdo=0x140045

The existing logic only take cares of UPF_D plug case,
and would take the bit 15:8 for this 0x140045 case.

This results in an non-existing pin assignment 0x0 in
dp_altmode_configure.

To fix this problem a new set of macros are introduced
to take plug/receptacle differences into consideration.

Co-developed-by: Pablo Sun <pablo.sun@mediatek.com>
Signed-off-by: Pablo Sun <pablo.sun@mediatek.com>
Co-developed-by: Macpaul Lin <macpaul.lin@mediatek.com>
Signed-off-by: Macpaul Lin <macpaul.lin@mediatek.com>
Reviewed-by: Guillaume Ranquet <granquet@baylibre.com>
Cc: stable@vger.kernel.org
---
 drivers/usb/typec/altmodes/displayport.c | 4 ++--
 include/linux/usb/typec_dp.h             | 5 +++++
 2 files changed, 7 insertions(+), 2 deletions(-)

Comments

Heikki Krogerus Aug. 2, 2022, 7:28 a.m. UTC | #1
Hi guys,

On Wed, Jul 27, 2022 at 07:05:03PM +0800, Macpaul Lin wrote:
> From: Pablo Sun <pablo.sun@mediatek.com>
> 
> From: Pablo Sun <pablo.sun@mediatek.com>

Looks like there's something wrong with the formating of the patch
here?

> Fix incorrect pin assignment values when connecting to a monitor with
> Type-C receptacle instead of a plug.
> 
> According to specification, an UFP_D receptacle's pin assignment
> should came from the UFP_D pin assignments field (bit 23:16), while
> an UFP_D plug's assignments are described in the DFP_D pin assignments
> (bit 15:8) during Mode Discovery.
> 
> For example the LG 27 UL850-W is a monitor with Type-C receptacle.
> The monitor responds to MODE DISCOVERY command with following
> DisplayPort Capability flag:
> 
>         dp->alt->vdo=0x140045
> 
> The existing logic only take cares of UPF_D plug case,
> and would take the bit 15:8 for this 0x140045 case.
> 
> This results in an non-existing pin assignment 0x0 in
> dp_altmode_configure.
> 
> To fix this problem a new set of macros are introduced
> to take plug/receptacle differences into consideration.
> 
> Co-developed-by: Pablo Sun <pablo.sun@mediatek.com>
> Signed-off-by: Pablo Sun <pablo.sun@mediatek.com>
> Co-developed-by: Macpaul Lin <macpaul.lin@mediatek.com>
> Signed-off-by: Macpaul Lin <macpaul.lin@mediatek.com>
> Reviewed-by: Guillaume Ranquet <granquet@baylibre.com>
> Cc: stable@vger.kernel.org

If this is a fix, then you need to have the "Fixes" tag that tells
which commit the patch is fixing.

> ---
>  drivers/usb/typec/altmodes/displayport.c | 4 ++--
>  include/linux/usb/typec_dp.h             | 5 +++++
>  2 files changed, 7 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/usb/typec/altmodes/displayport.c b/drivers/usb/typec/altmodes/displayport.c
> index 9360ca177c7d..8dd0e505ef99 100644
> --- a/drivers/usb/typec/altmodes/displayport.c
> +++ b/drivers/usb/typec/altmodes/displayport.c
> @@ -98,8 +98,8 @@ static int dp_altmode_configure(struct dp_altmode *dp, u8 con)
>  	case DP_STATUS_CON_UFP_D:
>  	case DP_STATUS_CON_BOTH: /* NOTE: First acting as DP source */
>  		conf |= DP_CONF_UFP_U_AS_UFP_D;
> -		pin_assign = DP_CAP_DFP_D_PIN_ASSIGN(dp->alt->vdo) &
> -			     DP_CAP_UFP_D_PIN_ASSIGN(dp->port->vdo);
> +		pin_assign = DP_CAP_PIN_ASSIGN_UFP_D(dp->alt->vdo) &
> +				 DP_CAP_PIN_ASSIGN_DFP_D(dp->port->vdo);
>  		break;
>  	default:
>  		break;
> diff --git a/include/linux/usb/typec_dp.h b/include/linux/usb/typec_dp.h
> index cfb916cccd31..8d09c2f0a9b8 100644
> --- a/include/linux/usb/typec_dp.h
> +++ b/include/linux/usb/typec_dp.h
> @@ -73,6 +73,11 @@ enum {
>  #define DP_CAP_USB			BIT(7)
>  #define DP_CAP_DFP_D_PIN_ASSIGN(_cap_)	(((_cap_) & GENMASK(15, 8)) >> 8)
>  #define DP_CAP_UFP_D_PIN_ASSIGN(_cap_)	(((_cap_) & GENMASK(23, 16)) >> 16)
> +/* Get pin assignment taking plug & receptacle into consideration */
> +#define DP_CAP_PIN_ASSIGN_UFP_D(_cap_) ((_cap_ & DP_CAP_RECEPTACLE) ? \
> +			DP_CAP_UFP_D_PIN_ASSIGN(_cap_) : DP_CAP_DFP_D_PIN_ASSIGN(_cap_))
> +#define DP_CAP_PIN_ASSIGN_DFP_D(_cap_) ((_cap_ & DP_CAP_RECEPTACLE) ? \
> +			DP_CAP_DFP_D_PIN_ASSIGN(_cap_) : DP_CAP_UFP_D_PIN_ASSIGN(_cap_))
>  
>  /* DisplayPort Status Update VDO bits */
>  #define DP_STATUS_CONNECTION(_status_)	((_status_) & 3)
> -- 
> 2.18.0

thanks,
Macpaul Lin Aug. 4, 2022, 2:42 a.m. UTC | #2
On 8/2/22 3:28 PM, Heikki Krogerus wrote:
> Hi guys,
> 
> On Wed, Jul 27, 2022 at 07:05:03PM +0800, Macpaul Lin wrote:
>> From: Pablo Sun <pablo.sun@mediatek.com>
>>
>> From: Pablo Sun <pablo.sun@mediatek.com>
> 
> Looks like there's something wrong with the formating of the patch
> here?

The duplicated line will be removed in patch v2.

>> Fix incorrect pin assignment values when connecting to a monitor with
>> Type-C receptacle instead of a plug.
>>
>> According to specification, an UFP_D receptacle's pin assignment
>> should came from the UFP_D pin assignments field (bit 23:16), while
>> an UFP_D plug's assignments are described in the DFP_D pin assignments
>> (bit 15:8) during Mode Discovery.
>>
>> For example the LG 27 UL850-W is a monitor with Type-C receptacle.
>> The monitor responds to MODE DISCOVERY command with following
>> DisplayPort Capability flag:
>>
>>          dp->alt->vdo=0x140045
>>
>> The existing logic only take cares of UPF_D plug case,
>> and would take the bit 15:8 for this 0x140045 case.
>>
>> This results in an non-existing pin assignment 0x0 in
>> dp_altmode_configure.
>>
>> To fix this problem a new set of macros are introduced
>> to take plug/receptacle differences into consideration.
>>
>> Co-developed-by: Pablo Sun <pablo.sun@mediatek.com>
>> Signed-off-by: Pablo Sun <pablo.sun@mediatek.com>
>> Co-developed-by: Macpaul Lin <macpaul.lin@mediatek.com>
>> Signed-off-by: Macpaul Lin <macpaul.lin@mediatek.com>
>> Reviewed-by: Guillaume Ranquet <granquet@baylibre.com>
>> Cc: stable@vger.kernel.org
> 
> If this is a fix, then you need to have the "Fixes" tag that tells
> which commit the patch is fixing.

Thanks for the review.
Will add "Fixes" tag and "Reviewed-by" tags in patch v2.

>> ---
>>   drivers/usb/typec/altmodes/displayport.c | 4 ++--
>>   include/linux/usb/typec_dp.h             | 5 +++++
>>   2 files changed, 7 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/usb/typec/altmodes/displayport.c b/drivers/usb/typec/altmodes/displayport.c
>> index 9360ca177c7d..8dd0e505ef99 100644
>> --- a/drivers/usb/typec/altmodes/displayport.c
>> +++ b/drivers/usb/typec/altmodes/displayport.c
>> @@ -98,8 +98,8 @@ static int dp_altmode_configure(struct dp_altmode *dp, u8 con)
>>   	case DP_STATUS_CON_UFP_D:
>>   	case DP_STATUS_CON_BOTH: /* NOTE: First acting as DP source */
>>   		conf |= DP_CONF_UFP_U_AS_UFP_D;
>> -		pin_assign = DP_CAP_DFP_D_PIN_ASSIGN(dp->alt->vdo) &
>> -			     DP_CAP_UFP_D_PIN_ASSIGN(dp->port->vdo);
>> +		pin_assign = DP_CAP_PIN_ASSIGN_UFP_D(dp->alt->vdo) &
>> +				 DP_CAP_PIN_ASSIGN_DFP_D(dp->port->vdo);
>>   		break;
>>   	default:
>>   		break;
>> diff --git a/include/linux/usb/typec_dp.h b/include/linux/usb/typec_dp.h
>> index cfb916cccd31..8d09c2f0a9b8 100644
>> --- a/include/linux/usb/typec_dp.h
>> +++ b/include/linux/usb/typec_dp.h
>> @@ -73,6 +73,11 @@ enum {
>>   #define DP_CAP_USB			BIT(7)
>>   #define DP_CAP_DFP_D_PIN_ASSIGN(_cap_)	(((_cap_) & GENMASK(15, 8)) >> 8)
>>   #define DP_CAP_UFP_D_PIN_ASSIGN(_cap_)	(((_cap_) & GENMASK(23, 16)) >> 16)
>> +/* Get pin assignment taking plug & receptacle into consideration */
>> +#define DP_CAP_PIN_ASSIGN_UFP_D(_cap_) ((_cap_ & DP_CAP_RECEPTACLE) ? \
>> +			DP_CAP_UFP_D_PIN_ASSIGN(_cap_) : DP_CAP_DFP_D_PIN_ASSIGN(_cap_))
>> +#define DP_CAP_PIN_ASSIGN_DFP_D(_cap_) ((_cap_ & DP_CAP_RECEPTACLE) ? \
>> +			DP_CAP_DFP_D_PIN_ASSIGN(_cap_) : DP_CAP_UFP_D_PIN_ASSIGN(_cap_))
>>   
>>   /* DisplayPort Status Update VDO bits */
>>   #define DP_STATUS_CONNECTION(_status_)	((_status_) & 3)
>> -- 
>> 2.18.0
> 
> thanks,
> 

Thanks.
Macpaul Lin
diff mbox series

Patch

diff --git a/drivers/usb/typec/altmodes/displayport.c b/drivers/usb/typec/altmodes/displayport.c
index 9360ca177c7d..8dd0e505ef99 100644
--- a/drivers/usb/typec/altmodes/displayport.c
+++ b/drivers/usb/typec/altmodes/displayport.c
@@ -98,8 +98,8 @@  static int dp_altmode_configure(struct dp_altmode *dp, u8 con)
 	case DP_STATUS_CON_UFP_D:
 	case DP_STATUS_CON_BOTH: /* NOTE: First acting as DP source */
 		conf |= DP_CONF_UFP_U_AS_UFP_D;
-		pin_assign = DP_CAP_DFP_D_PIN_ASSIGN(dp->alt->vdo) &
-			     DP_CAP_UFP_D_PIN_ASSIGN(dp->port->vdo);
+		pin_assign = DP_CAP_PIN_ASSIGN_UFP_D(dp->alt->vdo) &
+				 DP_CAP_PIN_ASSIGN_DFP_D(dp->port->vdo);
 		break;
 	default:
 		break;
diff --git a/include/linux/usb/typec_dp.h b/include/linux/usb/typec_dp.h
index cfb916cccd31..8d09c2f0a9b8 100644
--- a/include/linux/usb/typec_dp.h
+++ b/include/linux/usb/typec_dp.h
@@ -73,6 +73,11 @@  enum {
 #define DP_CAP_USB			BIT(7)
 #define DP_CAP_DFP_D_PIN_ASSIGN(_cap_)	(((_cap_) & GENMASK(15, 8)) >> 8)
 #define DP_CAP_UFP_D_PIN_ASSIGN(_cap_)	(((_cap_) & GENMASK(23, 16)) >> 16)
+/* Get pin assignment taking plug & receptacle into consideration */
+#define DP_CAP_PIN_ASSIGN_UFP_D(_cap_) ((_cap_ & DP_CAP_RECEPTACLE) ? \
+			DP_CAP_UFP_D_PIN_ASSIGN(_cap_) : DP_CAP_DFP_D_PIN_ASSIGN(_cap_))
+#define DP_CAP_PIN_ASSIGN_DFP_D(_cap_) ((_cap_ & DP_CAP_RECEPTACLE) ? \
+			DP_CAP_DFP_D_PIN_ASSIGN(_cap_) : DP_CAP_UFP_D_PIN_ASSIGN(_cap_))
 
 /* DisplayPort Status Update VDO bits */
 #define DP_STATUS_CONNECTION(_status_)	((_status_) & 3)