On Tue, 25 Feb 2025, "Manna, Animesh" <animesh.manna@xxxxxxxxx> wrote:
>> -----Original Message-----
>> From: Nikula, Jani <jani.nikula@xxxxxxxxx>
>> Sent: Monday, February 24, 2025 4:26 PM
>> To: Manna, Animesh <animesh.manna@xxxxxxxxx>; intel-
>> gfx@xxxxxxxxxxxxxxxxxxxxx; intel-xe@xxxxxxxxxxxxxxxxxxxxx
>> Cc: Hogander, Jouni <jouni.hogander@xxxxxxxxx>; B, Jeevan
>> <jeevan.b@xxxxxxxxx>; Manna, Animesh <animesh.manna@xxxxxxxxx>
>> Subject: Re: [PATCH v4 5/8] drm/i915/lobf: Add debug interface for lobf
>>
>> On Mon, 24 Feb 2025, Animesh Manna <animesh.manna@xxxxxxxxx> wrote:
>> > Add an interface in debugfs which will help in debugging LOBF feature.
>> >
>> > v1: Initial version.
>> > v2:
>> > - Remove FORCE_EN flag. [Jouni]
>> > - Change prefix from I915 to INTEL. [Jani]
>> > - Use u8 instead of bool for lobf-debug flag. [Jani]
>> >
>> > Signed-off-by: Animesh Manna <animesh.manna@xxxxxxxxx>
>> > ---
>> > drivers/gpu/drm/i915/display/intel_alpm.c | 51 +++++++++++++++++++
>> > .../drm/i915/display/intel_display_types.h | 5 ++
>> > 2 files changed, 56 insertions(+)
>> >
>> > diff --git a/drivers/gpu/drm/i915/display/intel_alpm.c
>> > b/drivers/gpu/drm/i915/display/intel_alpm.c
>> > index 83719ee1721c..5c70677ac3c0 100644
>> > --- a/drivers/gpu/drm/i915/display/intel_alpm.c
>> > +++ b/drivers/gpu/drm/i915/display/intel_alpm.c
>> > @@ -276,6 +276,9 @@ void intel_alpm_lobf_compute_config(struct
>> intel_dp *intel_dp,
>> > int waketime_in_lines, first_sdp_position;
>> > int context_latency, guardband;
>> >
>> > + if (intel_dp->alpm_parameters.lobf_debug &
>> INTEL_LOBF_DEBUG_DISABLE)
>> > + return;
>> > +
>> > if (!intel_dp_is_edp(intel_dp))
>> > return;
>> >
>> > @@ -449,6 +452,51 @@ static int i915_edp_lobf_info_show(struct
>> > seq_file *m, void *data)
>> >
>> > DEFINE_SHOW_ATTRIBUTE(i915_edp_lobf_info);
>> >
>> > +static int
>> > +i915_edp_lobf_debug_get(void *data, u64 *val) {
>> > + struct intel_display *display = data;
>>
>> Where do you pass display as data? It's connector?
>
> I should pass display, thanks for catching.
It's a connector based debugfs file, so you should probably keep passing
the connector, and just assign to connector here.
>
>>
>> > + struct intel_encoder *encoder;
>> > + int ret = -ENODEV;
>> > +
>> > + for_each_intel_dp(display->drm, encoder) {
>> > + struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
>> > +
>> > + if (!intel_dp_is_edp(intel_dp))
>> > + return ret;
>>
>> What is this supposed to do?
>
> Other than EDP connector lobg_debug is not available, so will remove the above condition check.
This debugfs file is connector based. Why are we looping all dp anyway?
I don't understand.
>>
>> > +
>> > + // TODO: split to each edp transcoder.
>>
>> What does this mean?
>
> Currently lobf debug option only supported with one EDP, TODO added to support second EDP.
Again, this debugfs file is connector based. I don't understand the TODO
at all.
>
>>
>> > + *val = READ_ONCE(intel_dp->alpm_parameters.lobf_debug);
>>
>> You read this from all intel_dp and combine into one? What?
>
> Will return from here, will take care in next version.
>
>> > + ret = 0;
>> > + }
>> > +
>> > + return ret;
>> > +}
>> > +
>> > +static int
>> > +i915_edp_lobf_debug_set(void *data, u64 val) {
>> > + struct intel_display *display = data;
>> > + struct intel_encoder *encoder;
>> > + int ret = -ENODEV;
>> > +
>> > + for_each_intel_dp(display->drm, encoder) {
>> > + struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
>> > +
>> > + if (!intel_dp_is_edp(intel_dp))
>> > + return ret;
>> > +
>> > + // TODO: split to each edp transcoder.
>> > + intel_dp->alpm_parameters.lobf_debug = val;
>> > + }
>> > +
>> > + return ret;
>>
>> So this always returns failure?
>
> Missed to set the ret variable, will take care in next version.
> Thanks for review.
>
> Regards,
> Animesh
>>
>> > +}
>> > +
>> > +DEFINE_SIMPLE_ATTRIBUTE(i915_edp_lobf_debug_fops,
>> > + i915_edp_lobf_debug_get,
>> i915_edp_lobf_debug_set,
>> > + "%llu\n");
>> > +
>> > void intel_alpm_lobf_debugfs_add(struct intel_connector *connector)
>> > {
>> > struct intel_display *display = to_intel_display(connector); @@
>> > -458,6 +506,9 @@ void intel_alpm_lobf_debugfs_add(struct
>> intel_connector *connector)
>> > connector->base.connector_type !=
>> DRM_MODE_CONNECTOR_eDP)
>> > return;
You'll only have these debugfs files for eDP connectors anyway. There's
no need for eDP checks later.
BR,
Jani.
>> >
>> > + debugfs_create_file("i915_edp_lobf_debug", 0644, root,
>> > + connector, &i915_edp_lobf_debug_fops);
>> > +
>> > debugfs_create_file("i915_edp_lobf_info", 0444, root,
>> > connector, &i915_edp_lobf_info_fops); } diff --git
>> > a/drivers/gpu/drm/i915/display/intel_display_types.h
>> > b/drivers/gpu/drm/i915/display/intel_display_types.h
>> > index b78721c451b8..b6ec9a8fadd9 100644
>> > --- a/drivers/gpu/drm/i915/display/intel_display_types.h
>> > +++ b/drivers/gpu/drm/i915/display/intel_display_types.h
>> > @@ -1808,6 +1808,11 @@ struct intel_dp {
>> > u8 aux_less_wake_lines;
>> > u8 silence_period_sym_clocks;
>> > u8 lfps_half_cycle_num_of_syms;
>> > +
>> > +#define INTEL_LOBF_DEBUG_MODE_MASK 0x0f
>> > +#define INTEL_LOBF_DEBUG_DEFAULT 0x00
>> > +#define INTEL_LOBF_DEBUG_DISABLE 0x01
>> > + u8 lobf_debug;
>>
>> Just overly complex still.
>>
>> > } alpm_parameters;
>> >
>> > u8 alpm_dpcd;
>>
>> --
>> Jani Nikula, Intel
--
Jani Nikula, Intel
