2022-08-31 23:06:45

by Fangzhi Zuo

[permalink] [raw]
Subject: [PATCH] drm/dp: Avoid Reading DPCD_REV Before Native Aux Read

The attempt to read DPCD_REV before any native aux read breaks
majority of DP2 compliance.

The spec. requires DP_SINK_STATUS to be polled for the reset status
DP_INTRA_HOP_AUX_REPLY_INDICATION during the clear training stage.

Polling DP_SINK_STATUS each time gets DPCD_REV read first
that makes non link training regsiter DPCD_REV get read
during UHBR link training. It violates DP2 compliance.

Cc: Ville Syrjala <[email protected]>
Cc: Lyude Paul <[email protected]>
Signed-off-by: Fangzhi Zuo <[email protected]>
---
drivers/gpu/drm/display/drm_dp_helper.c | 18 ------------------
1 file changed, 18 deletions(-)

diff --git a/drivers/gpu/drm/display/drm_dp_helper.c b/drivers/gpu/drm/display/drm_dp_helper.c
index e7c22c2ca90c..c7aa5bafa667 100644
--- a/drivers/gpu/drm/display/drm_dp_helper.c
+++ b/drivers/gpu/drm/display/drm_dp_helper.c
@@ -571,24 +571,6 @@ ssize_t drm_dp_dpcd_read(struct drm_dp_aux *aux, unsigned int offset,
{
int ret;

- /*
- * HP ZR24w corrupts the first DPCD access after entering power save
- * mode. Eg. on a read, the entire buffer will be filled with the same
- * byte. Do a throw away read to avoid corrupting anything we care
- * about. Afterwards things will work correctly until the monitor
- * gets woken up and subsequently re-enters power save mode.
- *
- * The user pressing any button on the monitor is enough to wake it
- * up, so there is no particularly good place to do the workaround.
- * We just have to do it before any DPCD access and hope that the
- * monitor doesn't power down exactly after the throw away read.
- */
- if (!aux->is_remote) {
- ret = drm_dp_dpcd_probe(aux, DP_DPCD_REV);
- if (ret < 0)
- return ret;
- }
-
if (aux->is_remote)
ret = drm_dp_mst_dpcd_read(aux, offset, buffer, size);
else
--
2.34.1


2022-09-02 21:57:36

by Lyude Paul

[permalink] [raw]
Subject: Re: [PATCH] drm/dp: Avoid Reading DPCD_REV Before Native Aux Read

Reviewed-by: Lyude Paul <[email protected]>

On Wed, 2022-08-31 at 18:13 -0400, Fangzhi Zuo wrote:
> The attempt to read DPCD_REV before any native aux read breaks
> majority of DP2 compliance.
>
> The spec. requires DP_SINK_STATUS to be polled for the reset status
> DP_INTRA_HOP_AUX_REPLY_INDICATION during the clear training stage.
>
> Polling DP_SINK_STATUS each time gets DPCD_REV read first
> that makes non link training regsiter DPCD_REV get read
> during UHBR link training. It violates DP2 compliance.
>
> Cc: Ville Syrjala <[email protected]>
> Cc: Lyude Paul <[email protected]>
> Signed-off-by: Fangzhi Zuo <[email protected]>
> ---
> drivers/gpu/drm/display/drm_dp_helper.c | 18 ------------------
> 1 file changed, 18 deletions(-)
>
> diff --git a/drivers/gpu/drm/display/drm_dp_helper.c b/drivers/gpu/drm/display/drm_dp_helper.c
> index e7c22c2ca90c..c7aa5bafa667 100644
> --- a/drivers/gpu/drm/display/drm_dp_helper.c
> +++ b/drivers/gpu/drm/display/drm_dp_helper.c
> @@ -571,24 +571,6 @@ ssize_t drm_dp_dpcd_read(struct drm_dp_aux *aux, unsigned int offset,
> {
> int ret;
>
> - /*
> - * HP ZR24w corrupts the first DPCD access after entering power save
> - * mode. Eg. on a read, the entire buffer will be filled with the same
> - * byte. Do a throw away read to avoid corrupting anything we care
> - * about. Afterwards things will work correctly until the monitor
> - * gets woken up and subsequently re-enters power save mode.
> - *
> - * The user pressing any button on the monitor is enough to wake it
> - * up, so there is no particularly good place to do the workaround.
> - * We just have to do it before any DPCD access and hope that the
> - * monitor doesn't power down exactly after the throw away read.
> - */
> - if (!aux->is_remote) {
> - ret = drm_dp_dpcd_probe(aux, DP_DPCD_REV);
> - if (ret < 0)
> - return ret;
> - }
> -
> if (aux->is_remote)
> ret = drm_dp_mst_dpcd_read(aux, offset, buffer, size);
> else

--
Cheers,
Lyude Paul (she/her)
Software Engineer at Red Hat