All of lore.kernel.org
 help / color / mirror / Atom feed
* [intel-gfx] Displayport compliance testing
@ 2014-10-09 15:38 Todd Previte
  2014-10-09 15:38 ` [PATCH 02/10] drm/i915: Add counters in the drm_dp_aux struct for I2C NACKs and DEFERs Todd Previte
  0 siblings, 1 reply; 6+ messages in thread
From: Todd Previte @ 2014-10-09 15:38 UTC (permalink / raw)
  To: intel-gfx

The kernel side is responsible for the acknowledgement of the test requests and 
setup of the required parameters. It also handles the necessary AUX transactions 
for reading the EDID and DPCD as well as writing response codes or checksums as 
necessary. Performing these operations in userspace would add unnecessary delays 
and complicate the interface more than necessary. The userspace application then 
handles the large motions - frame buffer management, mode sets and link 
configuration. The following is an overview of the basic event handling for 
compliance testing inside the kernel:
        - The test device  signals the DUT with an HPD pulse. This can  be a 
          short or long pulse, depending on circumstances.
        - The interrupt generated by the HPD pulse invokes the test handler, 
          which reads the test device DPCD to determine what actions are necessary.
        - Once the test handler determines which test has been requested, it 
          invokes the kernel-side handler function and then signals the userspace 
          app. If no userspace app has been registered, the signal is ignored.
        - Test responses and status is written out / reported as necessary and 
          normal operation is resumed.

The userspace support application and additional documentation will be posted 
to the list for review soon.

^ permalink raw reply	[flat|nested] 6+ messages in thread
[parent not found: <1412868486-47759-1-git-send-email-tprevite@gmail.com>]
* Re: [PATCH 03/12] drm/i915: Add counters in the drm_dp_aux struct for I2C NACKs and DEFERs
@ 2014-07-21 22:37 Paulo Zanoni
  2014-11-04 22:17 ` [PATCH 02/10] " Todd Previte
  0 siblings, 1 reply; 6+ messages in thread
From: Paulo Zanoni @ 2014-07-21 22:37 UTC (permalink / raw)
  To: Todd Previte; +Cc: Intel Graphics Development

2014-07-14 16:10 GMT-03:00 Todd Previte <tprevite@gmail.com>:
> These counters are used for Displayort complinace testing to detect error conditions
> when executing certain compliance tests. Currently these are used in the EDID tests
> to determine if the video mode needs to be set to the preferred mode or the failsafe
> mode.
>

It would be nice if you could cite on the commit message the name of
the specification and the name of the test(s) that use it.

Usually when I have patches that touch things outside
drivers/gpu/drm/i915, I add a Cc tag so I don't forget to email the
appropriate list:

Cc: dri-devel@lists.freedesktop.org
> Signed-off-by: Todd Previte <tprevite@gmail.com>
> ---
>  drivers/gpu/drm/drm_dp_helper.c | 2 ++
>  include/drm/drm_dp_helper.h     | 1 +
>  2 files changed, 3 insertions(+)
>
> diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c
> index 08e33b8..8353051 100644
> --- a/drivers/gpu/drm/drm_dp_helper.c
> +++ b/drivers/gpu/drm/drm_dp_helper.c
> @@ -654,10 +654,12 @@ static int drm_dp_i2c_do_msg(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
>
>                 case DP_AUX_I2C_REPLY_NACK:
>                         DRM_DEBUG_KMS("I2C nack\n");
> +                       aux->i2c_nack_count++;
>                         return -EREMOTEIO;
>
>                 case DP_AUX_I2C_REPLY_DEFER:
>                         DRM_DEBUG_KMS("I2C defer\n");
> +                       aux->i2c_defer_count++;
>                         usleep_range(400, 500);
>                         continue;
>
> diff --git a/include/drm/drm_dp_helper.h b/include/drm/drm_dp_helper.h
> index a21568b..3749cb4 100644
> --- a/include/drm/drm_dp_helper.h
> +++ b/include/drm/drm_dp_helper.h
> @@ -550,6 +550,7 @@ struct drm_dp_aux {
>         struct mutex hw_mutex;
>         ssize_t (*transfer)(struct drm_dp_aux *aux,
>                             struct drm_dp_aux_msg *msg);
> +       uint8_t i2c_nack_count, i2c_defer_count;

Does it really need to be uint8_t? I see on patch 7 that you don't
really write this value to a place that only accepts uint8_t-sized
arguments, so I fear that if we get 256 NACKs or DEFERs we may end up
doing the wrong thing.

Also, why don't we need to count the native NACKs and DEFERs?

>  };
>
>  ssize_t drm_dp_dpcd_read(struct drm_dp_aux *aux, unsigned int offset,
> --
> 1.9.1
>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx



-- 
Paulo Zanoni

^ permalink raw reply	[flat|nested] 6+ messages in thread
end of thread, other threads:[~2014-11-04 22:26 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
     [not found] <1412868814-47925-1-git-send-email-tprevite@gmail.com>
2014-10-09 15:33 ` [PATCH 02/10] drm/i915: Add counters in the drm_dp_aux struct for I2C NACKs and DEFERs Todd Previte
2014-10-09 15:36   ` Todd Previte
2014-10-09 15:38 [intel-gfx] Displayport compliance testing Todd Previte
2014-10-09 15:38 ` [PATCH 02/10] drm/i915: Add counters in the drm_dp_aux struct for I2C NACKs and DEFERs Todd Previte
     [not found] <1412868486-47759-1-git-send-email-tprevite@gmail.com>
2014-10-09 15:27 ` Todd Previte
  -- strict thread matches above, loose matches on Subject: below --
2014-07-21 22:37 [PATCH 03/12] " Paulo Zanoni
2014-11-04 22:17 ` [PATCH 02/10] " Todd Previte
2014-11-04 22:26   ` Daniel Vetter

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.