On Fri, 11 Jun 2021, "Navare, Manasi" <manasi.d.navare@xxxxxxxxx> wrote:
> On Thu, Jun 10, 2021 at 12:05:28PM +0300, Jani Nikula wrote:
>> Add a single point of truth for figuring out the primary/secondary crtc
>> for bigjoiner instead of duplicating the magic pipe +/- 1 in multiple
>> places.
>>
>> Also fix the pipe validity checks to properly take non-contiguous pipes
>> into account. The current checks may theoretically overflow
>> i915->pipe_to_crtc_mapping[pipe], albeit with a warning, due to fused
>> off pipes, as INTEL_NUM_PIPES() returns the actual number of pipes on
>> the platform, and the check is for INTEL_NUM_PIPES() == pipe + 1.
>>
>> Prefer primary/secondary terminology going forward.
>>
>> v2:
>> - Improved abstractions for pipe validity etc.
>>
>> Fixes: 8a029c113b17 ("drm/i915/dp: Modify VDSC helpers to configure DSC for Bigjoiner slave")
>> Fixes: d961eb20adb6 ("drm/i915/bigjoiner: atomic commit changes for uncompressed joiner")
>> Cc: Animesh Manna <animesh.manna@xxxxxxxxx>
>> Cc: Manasi Navare <manasi.d.navare@xxxxxxxxx>
>> Cc: Vandita Kulkarni <vandita.kulkarni@xxxxxxxxx>
>> Signed-off-by: Jani Nikula <jani.nikula@xxxxxxxxx>
>>
>> ---
>>
>> Dropped patch 2/2 [1], as the pipes need to be adjacent for big joiner,
>> even if pipes have been fused off.
>>
>> [1] https://patchwork.freedesktop.org/patch/msgid/20210603122842.22496-2-jani.nikula@xxxxxxxxx
>> ---
>> drivers/gpu/drm/i915/display/intel_display.c | 7 ++--
>> .../drm/i915/display/intel_display_types.h | 8 ++++
>> drivers/gpu/drm/i915/display/intel_vdsc.c | 40 +++++++++++++------
>> drivers/gpu/drm/i915/display/intel_vdsc.h | 1 +
>> 4 files changed, 40 insertions(+), 16 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
>> index 362bff9beb5c..3bad4e00f7be 100644
>> --- a/drivers/gpu/drm/i915/display/intel_display.c
>> +++ b/drivers/gpu/drm/i915/display/intel_display.c
>> @@ -9618,7 +9618,6 @@ static int intel_atomic_check_bigjoiner(struct intel_atomic_state *state,
>> struct intel_crtc_state *old_crtc_state,
>> struct intel_crtc_state *new_crtc_state)
>> {
>> - struct drm_i915_private *dev_priv = to_i915(state->base.dev);
>> struct intel_crtc_state *slave_crtc_state, *master_crtc_state;
>> struct intel_crtc *slave, *master;
>>
>> @@ -9634,15 +9633,15 @@ static int intel_atomic_check_bigjoiner(struct intel_atomic_state *state,
>> if (!new_crtc_state->bigjoiner)
>> return 0;
>>
>> - if (1 + crtc->pipe >= INTEL_NUM_PIPES(dev_priv)) {
>> + slave = intel_dsc_get_bigjoiner_secondary(crtc);
>> + if (!slave) {
>> DRM_DEBUG_KMS("[CRTC:%d:%s] Big joiner configuration requires "
>> "CRTC + 1 to be used, doesn't exist\n",
>> crtc->base.base.id, crtc->base.name);
>> return -EINVAL;
>> }
>>
>> - slave = new_crtc_state->bigjoiner_linked_crtc =
>> - intel_get_crtc_for_pipe(dev_priv, crtc->pipe + 1);
>> + new_crtc_state->bigjoiner_linked_crtc = slave;
>> slave_crtc_state = intel_atomic_get_crtc_state(&state->base, slave);
>> master = crtc;
>> if (IS_ERR(slave_crtc_state))
>> diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
>> index 15e91a99c8b9..7d64d8487fbe 100644
>> --- a/drivers/gpu/drm/i915/display/intel_display_types.h
>> +++ b/drivers/gpu/drm/i915/display/intel_display_types.h
>> @@ -1723,6 +1723,14 @@ vlv_pipe_to_channel(enum pipe pipe)
>> }
>> }
>>
>> +static inline bool intel_pipe_valid(struct drm_i915_private *i915, enum pipe pipe)
>> +{
>> + return (pipe >= 0 &&
>> + pipe < ARRAY_SIZE(i915->pipe_to_crtc_mapping) &&
>> + INTEL_INFO(i915)->pipe_mask & BIT(pipe) &&
>> + i915->pipe_to_crtc_mapping[pipe]);
>
> So no need to check INTEL_NUM_PIPES as long as the index exists in pipe_to_crtc_mappings correct?
Checking against INTEL_NUM_PIPES() is just plain wrong, because pipe may
be >= INTEL_NUM_PIPES() if pipes have been fused off.
>
>> +}
>> +
>> static inline struct intel_crtc *
>> intel_get_first_crtc(struct drm_i915_private *dev_priv)
>> {
>> diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.c b/drivers/gpu/drm/i915/display/intel_vdsc.c
>> index 7121b66bf96d..85749370508c 100644
>> --- a/drivers/gpu/drm/i915/display/intel_vdsc.c
>> +++ b/drivers/gpu/drm/i915/display/intel_vdsc.c
>> @@ -1106,6 +1106,27 @@ static i915_reg_t dss_ctl2_reg(const struct intel_crtc_state *crtc_state)
>> return is_pipe_dsc(crtc_state) ? ICL_PIPE_DSS_CTL2(pipe) : DSS_CTL2;
>> }
>>
>> +static struct intel_crtc *
>> +_get_crtc_for_pipe(struct drm_i915_private *i915, enum pipe pipe)
>> +{
>> + if (!intel_pipe_valid(i915, pipe))
>> + return NULL;
>> +
>> + return intel_get_crtc_for_pipe(i915, pipe);
>> +}
>
> Can we not include intel_pipe_valid() check in intel_get_crtc_for_pipe() so that any other function that calls intel_get_crtc_for_pipe() also
> validates the pipe ?
There's a use case for intel_dsc_get_bigjoiner_secondary() that requires
graceful error handling withouth a WARN. intel_get_crtc_for_pipe()
always WARNs on invalid pipe.
We should probably use intel_pipe_valid() in intel_get_crtc_for_pipe()
too, and WARN, but that's for another patch another time.
BR,
Jani.
>
> Manasi
>
>
>> +
>> +struct intel_crtc *
>> +intel_dsc_get_bigjoiner_secondary(const struct intel_crtc *primary_crtc)
>> +{
>> + return _get_crtc_for_pipe(to_i915(primary_crtc->base.dev), primary_crtc->pipe + 1);
>> +}
>> +
>> +static struct intel_crtc *
>> +intel_dsc_get_bigjoiner_primary(const struct intel_crtc *secondary_crtc)
>> +{
>> + return _get_crtc_for_pipe(to_i915(secondary_crtc->base.dev), secondary_crtc->pipe - 1);
>> +}
>> +
>> void intel_uncompressed_joiner_enable(const struct intel_crtc_state *crtc_state)
>> {
>> struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
>> @@ -1178,15 +1199,13 @@ void intel_uncompressed_joiner_get_config(struct intel_crtc_state *crtc_state)
>> dss_ctl1 = intel_de_read(dev_priv, dss_ctl1_reg(crtc_state));
>> if (dss_ctl1 & UNCOMPRESSED_JOINER_MASTER) {
>> crtc_state->bigjoiner = true;
>> - if (!WARN_ON(INTEL_NUM_PIPES(dev_priv) == crtc->pipe + 1))
>> - crtc_state->bigjoiner_linked_crtc =
>> - intel_get_crtc_for_pipe(dev_priv, crtc->pipe + 1);
>> + crtc_state->bigjoiner_linked_crtc = intel_dsc_get_bigjoiner_secondary(crtc);
>> + drm_WARN_ON(&dev_priv->drm, !crtc_state->bigjoiner_linked_crtc);
>> } else if (dss_ctl1 & UNCOMPRESSED_JOINER_SLAVE) {
>> crtc_state->bigjoiner = true;
>> crtc_state->bigjoiner_slave = true;
>> - if (!WARN_ON(crtc->pipe == PIPE_A))
>> - crtc_state->bigjoiner_linked_crtc =
>> - intel_get_crtc_for_pipe(dev_priv, crtc->pipe - 1);
>> + crtc_state->bigjoiner_linked_crtc = intel_dsc_get_bigjoiner_primary(crtc);
>> + drm_WARN_ON(&dev_priv->drm, !crtc_state->bigjoiner_linked_crtc);
>> }
>> }
>>
>> @@ -1224,14 +1243,11 @@ void intel_dsc_get_config(struct intel_crtc_state *crtc_state)
>>
>> if (!(dss_ctl1 & MASTER_BIG_JOINER_ENABLE)) {
>> crtc_state->bigjoiner_slave = true;
>> - if (!WARN_ON(crtc->pipe == PIPE_A))
>> - crtc_state->bigjoiner_linked_crtc =
>> - intel_get_crtc_for_pipe(dev_priv, crtc->pipe - 1);
>> + crtc_state->bigjoiner_linked_crtc = intel_dsc_get_bigjoiner_primary(crtc);
>> } else {
>> - if (!WARN_ON(INTEL_NUM_PIPES(dev_priv) == crtc->pipe + 1))
>> - crtc_state->bigjoiner_linked_crtc =
>> - intel_get_crtc_for_pipe(dev_priv, crtc->pipe + 1);
>> + crtc_state->bigjoiner_linked_crtc = intel_dsc_get_bigjoiner_secondary(crtc);
>> }
>> + drm_WARN_ON(&dev_priv->drm, !crtc_state->bigjoiner_linked_crtc);
>> }
>>
>> /* FIXME: add more state readout as needed */
>> diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.h b/drivers/gpu/drm/i915/display/intel_vdsc.h
>> index fe4d45561253..dfb1fd38deb4 100644
>> --- a/drivers/gpu/drm/i915/display/intel_vdsc.h
>> +++ b/drivers/gpu/drm/i915/display/intel_vdsc.h
>> @@ -22,5 +22,6 @@ void intel_uncompressed_joiner_get_config(struct intel_crtc_state *crtc_state);
>> void intel_dsc_get_config(struct intel_crtc_state *crtc_state);
>> enum intel_display_power_domain
>> intel_dsc_power_domain(const struct intel_crtc_state *crtc_state);
>> +struct intel_crtc *intel_dsc_get_bigjoiner_secondary(const struct intel_crtc *primary_crtc);
>>
>> #endif /* __INTEL_VDSC_H__ */
>> --
>> 2.20.1
>>
--
Jani Nikula, Intel Open Source Graphics Center
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
