Skip to contentSkip to conversation

venus: helper: do not set constrained format for UBWC

PATCH Superseded Mansur Alisha Shaik 5 participants 6 messages May 28, 2021 – Jun 19, 2021

Subsystems (4)

v1 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 venus: helper: do not set constrained format for UBWC

Do not set constrained format explicitly for UBWC

---
 drivers/media/platform/qcom/venus/helpers.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/media/platform/qcom/venus/helpers.c +4 -0
index b813d6dba..e4b8a2a 100644
@@ -1138,8 +1138,12 @@ int venus_helper_set_format_constraints(struct venus_inst *inst)
 	if (!IS_V6(inst->core))
 		return 0;
 
+	if (inst->opb_fmt == HFI_COLOR_FORMAT_NV12_UBWC)
+		return 0;
+
 	pconstraint.buffer_type = HFI_BUFFER_OUTPUT2;
 	pconstraint.num_planes = 2;
+
 	pconstraint.plane_format[0].stride_multiples = 128;
 	pconstraint.plane_format[0].max_stride = 8192;
 	pconstraint.plane_format[0].min_plane_buffer_height_multiple = 32;
Signature · 2 lines
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member 
of Code Aurora Forum, hosted by The Linux Foundation
#1 lore raw
Quoted 3 lines · Mansur Alisha Shaik

Do not set constrained format explicitly for UBWC

Could you give a little bit more detail on why, what the side effects are ?

Should this be a Fixes: ?

Quoted 9 lines
---
  drivers/media/platform/qcom/venus/helpers.c | 4 ++++
  1 file changed, 4 insertions(+)
drivers/media/platform/qcom/venus/helpers.c +0 -0
index b813d6dba..e4b8a2a 100644

drivers/media/platform/qcom/venus/helpers.c int venus_helper_set_format_constraints(struct venus_inst *inst) · line 1146

Quoted 9 lines
drivers/media/platform/qcom/venus/helpers.c +4 -0
@@ -1138,8 +1138,12 @@ int venus_helper_set_format_constraints(struct venus_inst *inst)
  	if (!IS_V6(inst->core))
  		return 0;
 
+	if (inst->opb_fmt == HFI_COLOR_FORMAT_NV12_UBWC)
+		return 0;
+
  	pconstraint.buffer_type = HFI_BUFFER_OUTPUT2;
  	pconstraint.num_planes = 2;
+

Additional newline should be removed.

drivers/media/platform/qcom/venus/helpers.c int venus_helper_set_format_constraints(struct venus_inst *inst) · line 1149

Quoted 4 lines
drivers/media/platform/qcom/venus/helpers.c +0 -0
  	pconstraint.plane_format[0].stride_multiples = 128;
  	pconstraint.plane_format[0].max_stride = 8192;
  	pconstraint.plane_format[0].min_plane_buffer_height_multiple = 32;

---
bod

#2 lore raw

Mansur, could you answer to Bryan's comments?

Quoted 38 lines · Bryan O'Donoghue
Quoted 3 lines · Mansur Alisha Shaik

Do not set constrained format explicitly for UBWC

Could you give a little bit more detail on why, what the side effects are ?

Should this be a Fixes: ?

Quoted 10 lines

---
  drivers/media/platform/qcom/venus/helpers.c | 4 ++++

  1 file changed, 4 insertions(+)
drivers/media/platform/qcom/venus/helpers.c +0 -0
index b813d6dba..e4b8a2a 100644

drivers/media/platform/qcom/venus/helpers.c int venus_helper_set_format_constraints(struct · line 1146

Quoted 9 lines
drivers/media/platform/qcom/venus/helpers.c +3 -0
@@ -1138,8 +1138,12 @@ int venus_helper_set_format_constraints(struct
 venus_inst *inst)
       if (!IS_V6(inst->core))
           return 0;
   +    if (inst->opb_fmt == HFI_COLOR_FORMAT_NV12_UBWC)
+        return 0;
+
       pconstraint.buffer_type = HFI_BUFFER_OUTPUT2;
       pconstraint.num_planes = 2;
+

Additional newline should be removed.

Quoted 4 lines
drivers/media/platform/qcom/venus/helpers.c +0 -0
       pconstraint.plane_format[0].stride_multiples = 128;
       pconstraint.plane_format[0].max_stride = 8192;

      pconstraint.plane_format[0].min_plane_buffer_height_multiple = 32;

---
bod

Signature · 2 lines
regards,
Stan
#3 lore raw
Quoted 10 lines · Stanimir Varbanov

Mansur, could you answer to Bryan's comments?

Quoted 8 lines · Bryan O'Donoghue
Quoted 7 lines · Mansur Alisha Shaik
Quoted 3 lines

Do not set constrained format explicitly for UBWC

Could you give a little bit more detail on why, what the side effects are ?

Sorry for late response, by default for NV12_UBWC is 128x32

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Should this be a Fixes: ?

without this fix on V6, firmware throws below SFR
qcom-venus aa00000.video-codec: SFR message from FW:
QC_IMAGE_VERSION_STRING=video-firmware.1.0-df9cb37cf8e507a4468265658702247652351a49
Err_Fatal -
/local/mnt/workspace/pkg/builds/dynamic_timely/tree2/vendor/qcom/proprietary/video-firmware-noship/venus_proc/venus/decoders/common/src/video_decoder.c:6644:2fef3
which result in playback happens through Software codec.

Quoted 30 lines
Quoted 30 lines
Quoted 30 lines
Quoted 12 lines

---
  drivers/media/platform/qcom/venus/helpers.c | 4 ++++

  1 file changed, 4 insertions(+)
drivers/media/platform/qcom/venus/helpers.c +0 -0
index b813d6dba..e4b8a2a 100644
@@ -1138,8 +1138,12 @@ int 

venus_helper_set_format_constraints(struct
venus_inst *inst)

Quoted 8 lines

      if (!IS_V6(inst->core))
          return 0;
  +    if (inst->opb_fmt == HFI_COLOR_FORMAT_NV12_UBWC) +        return 0;
+
      pconstraint.buffer_type = HFI_BUFFER_OUTPUT2; pconstraint.num_planes = 2;
+

Additional newline should be removed.

Quoted 4 lines

      pconstraint.plane_format[0].stride_multiples = 128; pconstraint.plane_format[0].max_stride = 8192; pconstraint.plane_format[0].min_plane_buffer_height_multiple = 32;

---
bod

#4 lore raw
Quoted 4 lines · On 14/06/2021 07:25, [email protected] wrote:
Quoted 3 lines · Stanimir Varbanov

Mansur, could you answer to Bryan's comments?

Quoted 1 line · Bryan O'Donoghue
Quoted 9 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines · Mansur Alisha Shaik
Quoted 3 lines

Do not set constrained format explicitly for UBWC

Could you give a little bit more detail on why, what the side effects are ?

    Sorry for late response, by default for NV12_UBWC is 128x32

Right so we have

pconstraint.plane_format[0].stride_multiples = 128; pconstraint.plane_format[0].min_plane_buffer_height_multiple = 32;

and

pconstraint.plane_format[1].stride_multiples = 128; pconstraint.plane_format[1].min_plane_buffer_height_multiple = 16;

and your patch says if opb_fmt == HFI_COLOR_FORMAT_NV12_UBWC then the we shouldn't do hfi_session_set_property()

I'm sure that's a fix that works but, I wonder would it be possible to fix this routine to continue to do hfi_session_set_property() with updated parameters for opb_fmt == HFI_COLOR_FORMAT_NV12_UBWC ?

Could you drill down into the detail in the commit log a little bit maybe giving a description of why returning for opb_fmt == HFI_COLOR_FORMAT_NV12_UBWC is the right thing to-do instead of adding a new case to the routine for HFI_COLOR_FORMAT_NV12_UBWC and calling hfi_session_set_property() ?

Its more for my own education on this topic :) but, also helps somebody else reading the log to understand what the fix is, why it is done this way.

Quoted 10 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Should this be a Fixes: ?

    without this fix on V6, firmware throws below SFR
    qcom-venus aa00000.video-codec: SFR message from FW:
QC_IMAGE_VERSION_STRING=video-firmware.1.0-df9cb37cf8e507a4468265658702247652351a49

    Err_Fatal -
/local/mnt/workspace/pkg/builds/dynamic_timely/tree2/vendor/qcom/proprietary/video-firmware-noship/venus_proc/venus/decoders/common/src/video_decoder.c:6644:2fef3

    which result in playback happens through Software codec.

OK, I think I can answer my own question here.

Technically no since 6XX isn't in the long-term-support kernel but, I'd suggest adding a "Fixes" anyway, so that other users know to apply this patch to their trees.

Fixes bc28936bbba9 ("media: venus: helpers, hfi, vdec: Set actual plane constraints to FW")

---
bod

#5 lore raw

Hi Bryan,

Quoted 25 lines · Bryan O'Donoghue
Quoted 4 lines · On 14/06/2021 07:25, [email protected] wrote:
Quoted 3 lines · Stanimir Varbanov

Mansur, could you answer to Bryan's comments?

Quoted 1 line · Bryan O'Donoghue
Quoted 9 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines · Mansur Alisha Shaik
Quoted 3 lines

Do not set constrained format explicitly for UBWC

Could you give a little bit more detail on why, what the side effects are ?

    Sorry for late response, by default for NV12_UBWC is 128x32

Right so we have

pconstraint.plane_format[0].stride_multiples = 128; pconstraint.plane_format[0].min_plane_buffer_height_multiple = 32;

and

pconstraint.plane_format[1].stride_multiples = 128; pconstraint.plane_format[1].min_plane_buffer_height_multiple = 16;

Quoted 6 lines

and your patch says if opb_fmt == HFI_COLOR_FORMAT_NV12_UBWC then the we shouldn't do hfi_session_set_property()

I'm sure that's a fix that works but, I wonder would it be possible to fix this routine to continue to do hfi_session_set_property() with updated parameters for opb_fmt == HFI_COLOR_FORMAT_NV12_UBWC ?

Venus hardware would always go with alignments as 128x32 for WxH irrespective of color formats. It happened so (historically) that for NV12 (linear) format, usecase started demanding for alignments as 256(interlace) and 512(HEIF), and hence NV12 was defaulted to align as 512x512 in firmware. This was done to avoid carrying multiple alignments for different usecases, since aligning with 512x512 would also align it with 128x32 and 256x256 as well.
For UBWC, there is no need to override the default alignment of 128x32, hence the api was added to override default alignments for applicable formats, in this case NV12(Linear).

Quoted 5 lines

Could you drill down into the detail in the commit log a little bit maybe giving a description of why returning for opb_fmt == HFI_COLOR_FORMAT_NV12_UBWC is the right thing to-do instead of adding a new case to the routine for HFI_COLOR_FORMAT_NV12_UBWC and calling hfi_session_set_property() ?

Above details should provide info on the need to set this only for NV12 (linear) and skip for NV12 (UBWC).

Quoted 12 lines

Its more for my own education on this topic :) but, also helps somebody else reading the log to understand what the fix is, why it is done this way.

Quoted 8 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Should this be a Fixes: ?

    without this fix on V6, firmware throws below SFR
    qcom-venus aa00000.video-codec: SFR message from FW:
QC_IMAGE_VERSION_STRING=video-firmware.1.0-df9cb37cf8e507a4468265658702247652351a49
    Err_Fatal -
/local/mnt/workspace/pkg/builds/dynamic_timely/tree2/vendor/qcom/proprietary/video-firmware-noship/venus_proc/venus/decoders/common/src/video_decoder.c:6644:2fef3
    which result in playback happens through Software codec.

Quoted 8 lines

OK, I think I can answer my own question here.

Technically no since 6XX isn't in the long-term-support kernel but, I'd suggest adding a "Fixes" anyway, so that other users know to apply this patch to their trees.

Fixes bc28936bbba9 ("media: venus: helpers, hfi, vdec: Set actual

plane constraints to FW")

Yes, it should go as "Fixes" tag to above patch which we made for 6xx recently.

Thanks,
Vikash