Skip to contentSkip to conversation

Expand amdgpu documentation

PATCH Superseded Rodrigo Siqueira 6 participants 18 messages Jun 13, 2022 – Jun 18, 2022

Subsystems (4)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

cover Expand amdgpu documentation

This patchset introduces some new AMDGPU documentation. You will find:

1. A CSV table that maps which component version is part of some ASIC families. This can be useful to narrow down bugs;
2. Some explanation about AMD Display Pipeline;
3. An explanation of Multiple Plane Overlay, which can be useful for userspace applications.

Since we have some diagrams, I recommend applying these patches and building the documentation to have a better experience.

Thanks
Siqueira

Rodrigo Siqueira (3):
Documentation/gpu: Add info table for ASICs Documentation/gpu: Add an explanation about the DCN pipeline Documentation/gpu: Add Multiplane Overlay doc

 .../gpu/amdgpu/apu-asic-info-table.csv        |    8 +
 .../gpu/amdgpu/dgpu-asic-info-table.csv       |   24 +
 .../gpu/amdgpu/display/dcn-overview.rst       |   58 +
 Documentation/gpu/amdgpu/display/index.rst    |    1 +
 .../gpu/amdgpu/display/mpo-cursor.svg         |  435 +++++++
 .../gpu/amdgpu/display/mpo-overview.rst       |  242 ++++
 .../multi-display-hdcp-mpo-less-pipe-ex.svg   |  220 ++++
 .../amdgpu/display/multi-display-hdcp-mpo.svg |  171 +++
 .../amdgpu/display/pipeline_4k_no_split.svg   |  958 +++++++++++++++
 .../gpu/amdgpu/display/pipeline_4k_split.svg  | 1062 +++++++++++++++++
 .../single-display-mpo-multi-video.svg        |  339 ++++++
 .../gpu/amdgpu/display/single-display-mpo.svg |  266 +++++
 Documentation/gpu/amdgpu/driver-misc.rst      |   17 +
 13 files changed, 3801 insertions(+)
 create mode 100644 Documentation/gpu/amdgpu/apu-asic-info-table.csv
 create mode 100644 Documentation/gpu/amdgpu/dgpu-asic-info-table.csv
 create mode 100644 Documentation/gpu/amdgpu/display/mpo-cursor.svg
 create mode 100644 Documentation/gpu/amdgpu/display/mpo-overview.rst
 create mode 100644 Documentation/gpu/amdgpu/display/multi-display-hdcp-mpo-less-pipe-ex.svg
 create mode 100644 Documentation/gpu/amdgpu/display/multi-display-hdcp-mpo.svg
 create mode 100644 Documentation/gpu/amdgpu/display/pipeline_4k_no_split.svg
 create mode 100644 Documentation/gpu/amdgpu/display/pipeline_4k_split.svg
 create mode 100644 Documentation/gpu/amdgpu/display/single-display-mpo-multi-video.svg
 create mode 100644 Documentation/gpu/amdgpu/display/single-display-mpo.svg
Signature · 1 line
2.25.1
#1 lore raw

PATCH 1/3 Documentation/gpu: Add info table for ASICs

Amdgpu driver is used in an extensive range of devices, and each ASIC has some specific configuration. As a result of this variety, sometimes it is hard to identify the correct block that might cause the issue. This commit expands the amdgpu kernel-doc to alleviate this issue by introducing one ASIC table that describes dGPU and another one that shares the APU info.

Cc Harry Wentland <[email protected]>
Cc Nicholas Kazlauskas <[email protected]>
Cc Bhawanpreet Lakha <[email protected]>
Cc Hersen Wu <[email protected]>
Cc Alex Hung <[email protected]>
Cc Pierre-Eric Pelloux-Prayer <[email protected]>
Cc Leo Li <[email protected]>
Cc Simon Ser <[email protected]>
Cc Pekka Paalanen <[email protected]>
Cc Sean Paul <[email protected]>
Cc Mark Yacoub <[email protected]>
Cc Pierre-Loup <[email protected]>
Cc Michel Dänzer <[email protected]>
---
 .../gpu/amdgpu/apu-asic-info-table.csv        |  8 +++++++
 .../gpu/amdgpu/dgpu-asic-info-table.csv       | 24 +++++++++++++++++++
 Documentation/gpu/amdgpu/driver-misc.rst      | 17 +++++++++++++
 3 files changed, 49 insertions(+)
 create mode 100644 Documentation/gpu/amdgpu/apu-asic-info-table.csv
 create mode 100644 Documentation/gpu/amdgpu/dgpu-asic-info-table.csv
Documentation/gpu/amdgpu/apu-asic-info-table.csv new file +8 -0
new file mode 100644
index 000000000000..98c6988e424e
@@ -0,0 +1,8 @@
+Product Name, Code Reference, DCN/DCE version, GC version, VCE/UVD/VCN version, SDMA version
+Radeon R* Graphics, CARRIZO/STONEY, DCE 11, 8, VCE 3 / UVD 6, 3
+Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN/PICASSO, DCN 1.0, 9.1.0, VCN 1.0, 4.1.0
+Ryzen 4000 series, RENOIR, DCN 2.1, 9.3, VCN 2.2, 4.1.2
+Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN2, DCN 1.0, 9.2.2, VCN 1.0.1, 4.1.1
+SteamDeck, VANGOGH, DCN 3.0.1, 10.3.1, VCN 3.1.0, 5.2.1
+Ryzen 5000 series, GREEN SARDINE, DCN 2.1, 9.3, VCN 2.2, 4.1.1
+Ryzen 6000 Zen, YELLOW CARP, 3.1.2, 10.3.3, VCN 3.1.1, 5.2.3
Documentation/gpu/amdgpu/dgpu-asic-info-table.csv new file +24 -0
new file mode 100644
index 000000000000..84617aa35dab
@@ -0,0 +1,24 @@
+Product Name, Code Reference, DCN/DCE version, GC version, VCN version, SDMA version
+AMD Radeon (TM) HD 8500M/ 8600M /M200 /M320 /M330 /M335 Series, HAINAN, --,  6, --, --
+AMD Radeon HD 7800 /7900 /FireGL Series, TAHITI, DCE 6, 6, VCE 1 / UVD 3, --
+AMD Radeon R7 (TM|HD) M265 /M370 /8500M /8600 /8700 /8700M, OLAND, DCE 6, 6, VCE 1 / UVD 3, --
+AMD Radeon (TM) (HD|R7) 7800 /7970 /8800 /8970 /370/ Series, PITCAIRN, DCE 6, 6, VCE 1 / UVD 3, --
+AMD Radeon (TM|R7|R9|HD) E8860 /M360 /7700 /7800 /8800 /9000(M) /W4100 Series, VERDE, DCE 6, 6, VCE 1 / UVD 3, --
+AMD Radeon HD M280X /M380 /7700 /8950 /W5100, BONAIRE, DCE 8, 7, VCE 2 / UVD 4.2, 1
+AMD Radeon (R9|TM) 200 /390 /W8100 /W9100 Series, HAWAII, DCE 8, 7, VCE 2 / UVD 4.2, 1
+AMD Radeon (TM) R(5|7) M315 /M340 /M360, TOPAZ, *, 8, --, 2
+AMD Radeon (TM) R9 200 /380 /W7100 /S7150 /M390 /M395 Series, TONGA, DCE 10, 8, VCE 3 / UVD 5, 3
+AMD Radeon (FirePro) (TM) R9 Fury Series, FIJI, DCE 10, 8, VCE 3 / UVD 6, 3
+Radeon RX 470 /480 /570 /580 /590 Series - AMD Radeon (TM) (Pro WX) 5100 /E9390 /E9560 /E9565 /V7350 /7100 /P30PH, POLARIS10, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3
+Radeon (TM) (RX|Pro WX) E9260 /460 /V5300X /550 /560(X) Series, POLARIS11, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3
+Radeon (RX/Pro) 500 /540(X) /550 /640 /WX2100 /WX3100 /WX200 Series, POLARIS12, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3
+Radeon (RX|TM) (PRO|WX) Vega /MI25 /V320 /V340L /8200 /9100 /SSG MxGPU, VEGA10, DCE 12, 9.0.1, VCE 4.0.0 / UVD 7.0.0, 4.0.0
+AMD Radeon (Pro) VII /MI50 /MI60, VEGA20, DCE 12, 9.4.0, VCE 4.1.0 / UVD 7.2.0, 4.2.0
+MI100, ARCTURUS, *, 9.4.1, VCN 2.5.0, 4.2.2
+MI200, ALDEBARAN, *, 9.4.2, VCN 2.6.0, 4.4.0
+AMD Radeon (RX|Pro) 5600(M|XT) /5700 (M|XT|XTB) /W5700, NAVI10, DCN 2.0.0, 10.1.10, VCN 2.0.0, 5.0.0
+AMD Radeon (Pro) 5300 /5500XTB/5500(XT|M) /W5500M /W5500, NAVI14, DCN 2.0.0, 10.1.1, VCN 2.0.2, 5.0.2
+AMD Radeon RX 6800(XT) /6900(XT) /W6800, SIENNA_CICHLID, DCN 3.0.0, 10.3.0, VCN 3.0.0, 5.2.0
+AMD Radeon RX 6700 XT / 6800M / 6700M, NAVY_FLOUNDER, DCN 3.0.0, 10.3.2, VCN 3.0.0, 5.2.2
+AMD Radeon RX 6600(XT) /6600M /W6600 /W6600M, DIMGREY_CAVEFISH, DCN 3.0.2, 10.3.4, VCN 3.0.16, 5.2.4
+AMD Radeon RX 6500M /6300M /W6500M /W6300M, BEIGE_GOBY, DCN 3.0.3, 10.3.5, VCN 3.0.33, 5.2.5
Documentation/gpu/amdgpu/driver-misc.rst +17 -0
index e3d6b2fa2493..1800543d45f7 100644
@@ -32,6 +32,23 @@ unique_id
 .. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c
    :doc: unique_id
 
+Accelerated Processing Units (APU) Info
+---------------------------------------
+
+.. csv-table::
+   :header-rows: 1
+   :widths: 3, 2, 2, 1, 1, 1
+   :file: ./apu-asic-info-table.csv
+
+Discrete GPU Info
+-----------------
+
+.. csv-table::
+   :header-rows: 1
+   :widths: 3, 2, 2, 1, 1, 1
+   :file: ./dgpu-asic-info-table.csv
+
+
 GPU Memory Usage Information
 ============================
 
Signature · 1 line
2.25.1
#2 lore raw

Re: [PATCH 1/3] Documentation/gpu: Add info table for ASICs

Quoted 6 lines · Rodrigo Siqueira

Amdgpu driver is used in an extensive range of devices, and each ASIC has some specific configuration. As a result of this variety, sometimes it is hard to identify the correct block that might cause the issue. This commit expands the amdgpu kernel-doc to alleviate this issue by introducing one ASIC table that describes dGPU and another one that shares the APU info.

Nice!

Are there plans to keep this list up-to-date?

FWIW, my go-to reference tables are [1], updated by Alex.

Would it make sense to add a "GFX Core" column?

[1]: https://www.x.org/wiki/RadeonFeature/

#3 lore raw

Re: [PATCH 1/3] Documentation/gpu: Add info table for ASICs

Quoted 10 lines · Simon Ser
Quoted 6 lines · Rodrigo Siqueira

Amdgpu driver is used in an extensive range of devices, and each ASIC has some specific configuration. As a result of this variety, sometimes it is hard to identify the correct block that might cause the issue. This commit expands the amdgpu kernel-doc to alleviate this issue by introducing one ASIC table that describes dGPU and another one that shares the APU info.

Quoted 7 lines

Nice!

Are there plans to keep this list up-to-date?

FWIW, my go-to reference tables are [1], updated by Alex.

Would it make sense to add a "GFX Core" column?

That's what the GC column is for.

Alex

Quoted 2 lines

[1]: https://www.x.org/wiki/RadeonFeature/

#4 lore raw

Re: [PATCH 1/3] Documentation/gpu: Add info table for ASICs

Quoted 13 lines · Alex Deucher
Quoted 11 lines · On Tue, Jun 14, 2022 at 2:16 PM Simon Ser [email protected] wrote:
Quoted 6 lines · On Monday, June 13th, 2022 at 22:01, Rodrigo Siqueira [email protected]…

Amdgpu driver is used in an extensive range of devices, and each ASIC has some specific configuration. As a result of this variety, sometimes it is hard to identify the correct block that might cause the issue. This commit expands the amdgpu kernel-doc to alleviate this issue by introducing one ASIC table that describes dGPU and another one that shares the APU info.

Nice!

Quoted 7 lines
Quoted 5 lines

Are there plans to keep this list up-to-date?

FWIW, my go-to reference tables are 1, updated by Alex.

Would it make sense to add a "GFX Core" column?

That's what the GC column is for.

Oh! Does this stand for "GFX Core", or for "Graphics and Compute"? The glossary documents GC as the latter. If there is a name conflict, maybe we can keep using the long name in the table, or document what "GC" means in a sentence.

#5 lore raw

Re: [PATCH 1/3] Documentation/gpu: Add info table for ASICs

Quoted 20 lines · Simon Ser
Quoted 17 lines · Alex Deucher
Quoted 15 lines · On Tue, Jun 14, 2022 at 2:16 PM Simon Ser [email protected] wrote:
Quoted 6 lines · On Monday, June 13th, 2022 at 22:01, Rodrigo Siqueira [email protected]…

Amdgpu driver is used in an extensive range of devices, and each ASIC has some specific configuration. As a result of this variety, sometimes it is hard to identify the correct block that might cause the issue. This commit expands the amdgpu kernel-doc to alleviate this issue by introducing one ASIC table that describes dGPU and another one that shares the APU info.

Nice!

Are there plans to keep this list up-to-date?

FWIW, my go-to reference tables are 1, updated by Alex.

Quoted 8 lines
Quoted 3 lines
Quoted 1 line

Would it make sense to add a "GFX Core" column?

That's what the GC column is for.

Oh! Does this stand for "GFX Core", or for "Graphics and Compute"? The glossary documents GC as the latter. If there is a name conflict, maybe we can keep using the long name in the table, or document what "GC" means in a sentence.

Both? The hardware block is called GC, but I think different teams expand it to Graphics Core (as in GCN) or Graphics and Compute. Same hardware block either way. I guess we should include both in the glossary.

Alex

#6 lore raw

Re: [PATCH 1/3] Documentation/gpu: Add info table for ASICs

Quoted 30 lines · Alex Deucher
Quoted 20 lines · Simon Ser
Quoted 17 lines · Alex Deucher
Quoted 15 lines · On Tue, Jun 14, 2022 at 2:16 PM Simon Ser [email protected] wrote:
Quoted 6 lines · On Monday, June 13th, 2022 at 22:01, Rodrigo Siqueira [email protected]…

Amdgpu driver is used in an extensive range of devices, and each ASIC has some specific configuration. As a result of this variety, sometimes it is hard to identify the correct block that might cause the issue. This commit expands the amdgpu kernel-doc to alleviate this issue by introducing one ASIC table that describes dGPU and another one that shares the APU info.

Nice!

Are there plans to keep this list up-to-date?

FWIW, my go-to reference tables are 1, updated by Alex.

Quoted 8 lines
Quoted 3 lines
Quoted 1 line

Would it make sense to add a "GFX Core" column?

That's what the GC column is for.

Oh! Does this stand for "GFX Core", or for "Graphics and Compute"? The glossary documents GC as the latter. If there is a name conflict, maybe we can keep using the long name in the table, or document what "GC" means in a sentence.

Quoted 4 lines

Both? The hardware block is called GC, but I think different teams expand it to Graphics Core (as in GCN) or Graphics and Compute. Same hardware block either way. I guess we should include both in the glossary.

How should I add it? Something like this:

GC/GFX:
Graphics and Compute/Core

Thanks
Siqueira

Quoted 1 line

Alex

#7 lore raw

Re: [PATCH 1/3] Documentation/gpu: Add info table for ASICs

Quoted 40 lines · Rodrigo Siqueira Jordao
Quoted 30 lines · Alex Deucher
Quoted 20 lines · Simon Ser
Quoted 17 lines · Alex Deucher
Quoted 15 lines · On Tue, Jun 14, 2022 at 2:16 PM Simon Ser [email protected] wrote:
Quoted 6 lines · On Monday, June 13th, 2022 at 22:01, Rodrigo Siqueira [email protected]…

Amdgpu driver is used in an extensive range of devices, and each ASIC has some specific configuration. As a result of this variety, sometimes it is hard to identify the correct block that might cause the issue. This commit expands the amdgpu kernel-doc to alleviate this issue by introducing one ASIC table that describes dGPU and another one that shares the APU info.

Nice!

Are there plans to keep this list up-to-date?

FWIW, my go-to reference tables are 1, updated by Alex.

Quoted 8 lines
Quoted 3 lines
Quoted 1 line

Would it make sense to add a "GFX Core" column?

That's what the GC column is for.

Oh! Does this stand for "GFX Core", or for "Graphics and Compute"? The glossary documents GC as the latter. If there is a name conflict, maybe we can keep using the long name in the table, or document what "GC" means in a sentence.

Quoted 4 lines

Both? The hardware block is called GC, but I think different teams expand it to Graphics Core (as in GCN) or Graphics and Compute. Same hardware block either way. I guess we should include both in the glossary.

Quoted 4 lines

How should I add it? Something like this:

GC/GFX:
Graphics and Compute/Core

maybe something like

GC:
Graphics and Compute or Graphics Core. Refers to the graphics
hardware block on the GPU.

Quoted 6 lines

Thanks
Siqueira

Quoted 1 line

Alex

#8 lore raw

Re: [PATCH 1/3] Documentation/gpu: Add info table for ASICs

In reply to Rodrigo Siqueira

Nice!

Quoted 2 lines
Documentation/gpu/amdgpu/apu-asic-info-table.csv +0 -0

...

Quoted 2 lines

+Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN/PICASSO, DCN 1.0, 9.1.0, VCN 1.0, 4.1.0

...

Quoted 2 lines

+Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN2, DCN 1.0, 9.2.2, VCN 1.0.1, 4.1.1

Those similar-but-not-quite lines tend to confuse me :)

Best regards,

Signature · 1 line
Yann
#9 lore raw

Re: [PATCH 1/3] Documentation/gpu: Add info table for ASICs

Quoted 5 lines · Yann Dirson

Nice!

Quoted 2 lines
Documentation/gpu/amdgpu/apu-asic-info-table.csv +0 -0

...

Quoted 7 lines
Quoted 2 lines

+Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN/PICASSO, DCN 1.0, 9.1.0, VCN 1.0, 4.1.0

...

Quoted 2 lines

+Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN2, DCN 1.0, 9.2.2, VCN 1.0.1, 4.1.1

Those similar-but-not-quite lines tend to confuse me :)

I guess it will be easier to visualize this data after build the kernel doc since the csv file will became an html table.

Thanks
Siqueira

Quoted 1 line

Best regards,

#10 lore raw

Re: [PATCH 1/3] Documentation/gpu: Add info table for ASICs

Quoted 12 lines · Yann Dirson
Quoted 9 lines

Nice!

Quoted 2 lines
Documentation/gpu/amdgpu/apu-asic-info-table.csv +0 -0

...

Quoted 3 lines

+Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx,
RAVEN/PICASSO, DCN 1.0, 9.1.0, VCN 1.0, 4.1.0

...

Quoted 8 lines
Quoted 5 lines
Quoted 3 lines

+Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx,
RAVEN2, DCN 1.0, 9.2.2, VCN 1.0.1, 4.1.1

Those similar-but-not-quite lines tend to confuse me :)

I guess it will be easier to visualize this data after build the kernel doc since the csv file will became an html table.

Possibly, though my surprise is rather that 2 different lines in the table have the same label in their first cell. Do RAVEN/PICASSO and RAVEN2 really share the same product name despite different hardware blocks ?

#11 lore raw

Re: [PATCH 1/3] Documentation/gpu: Add info table for ASICs

Quoted 19 lines · Yann Dirson
Quoted 19 lines · Yann Dirson
Quoted 14 lines

Nice!

Quoted 2 lines
Documentation/gpu/amdgpu/apu-asic-info-table.csv +0 -0

...

Quoted 3 lines

+Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx,
RAVEN/PICASSO, DCN 1.0, 9.1.0, VCN 1.0, 4.1.0

...

Quoted 3 lines

+Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx,
RAVEN2, DCN 1.0, 9.2.2, VCN 1.0.1, 4.1.1

Those similar-but-not-quite lines tend to confuse me :)

Quoted 7 lines
Quoted 2 lines

I guess it will be easier to visualize this data after build the kernel doc since the csv file will became an html table.

Possibly, though my surprise is rather that 2 different lines in the table have the same label in their first cell. Do RAVEN/PICASSO and RAVEN2 really share the same product name despite different hardware blocks ?

Sort of, the first column is marketing names. Various Ryzen 3000 variants are Raven or Picasso or Raven2. There are tons of parts in that series: Ryzen 3 3100, Ryzen 5 3500, Ryzen 7 3700X, etc. Same with the embedded parts: AMD Ryzen Embedded V1202B, AMD Ryzen Embedded V1807B, etc. I'm not sure how feasible it is to put every SKU in the table.

Alex

#12 lore raw

PATCH 2/3 Documentation/gpu: Add an explanation about the DCN pipeline

In reply to Rodrigo Siqueira

In the DCN code, we constantly talk about hardware pipeline, pipeline, or even just pipes, which is a concept that is not obvious to everyone. For this reason, this commit expands the DCN overview explanation by adding a new section that describes what a pipeline is from the DCN perspective.

Cc Harry Wentland <[email protected]>
Cc Nicholas Kazlauskas <[email protected]>
Cc Bhawanpreet Lakha <[email protected]>
Cc Hersen Wu <[email protected]>
Cc Alex Hung <[email protected]>
Cc Pierre-Eric Pelloux-Prayer <[email protected]>
Cc Leo Li <[email protected]>
Cc Simon Ser <[email protected]>
Cc Pekka Paalanen <[email protected]>
Cc Sean Paul <[email protected]>
Cc Mark Yacoub <[email protected]>
Cc Pierre-Loup <[email protected]>
Cc Michel Dänzer <[email protected]>
---
 .../gpu/amdgpu/display/dcn-overview.rst       |   58 +
 .../amdgpu/display/pipeline_4k_no_split.svg   |  958 +++++++++++++++
 .../gpu/amdgpu/display/pipeline_4k_split.svg  | 1062 +++++++++++++++++
 3 files changed, 2078 insertions(+)
 create mode 100644 Documentation/gpu/amdgpu/display/pipeline_4k_no_split.svg
 create mode 100644 Documentation/gpu/amdgpu/display/pipeline_4k_split.svg
Documentation/gpu/amdgpu/display/dcn-overview.rst +58 -0
index f98624d7828e..430e363fccae 100644
@@ -124,6 +124,64 @@ depth format), bit-depth reduction/dithering would kick in. In OPP, we would
 also apply a regamma function to introduce the gamma removed earlier back.
 Eventually, we output data in integer format at DIO.
 
+AMD Hardware Pipeline
+---------------------
+
+When discussing graphics on Linux, the **pipeline** term is often overloaded
+with multiple meanings, and DCN adds another context to this word. In the DCN
+driver, we use the term **hardware pipeline** or  **pipeline** or just **pipe**
+to indicate how many DCN blocks we are instantiating to address some specific
+configuration. DCN composes those pipelines together when the data reach the
+end of the OPTC block before sending it to the display. We have this pipeline
+concept for trying to save energy and optimize bandwidth utilization while
+providing the maximum performance per watt. It is easier to discuss it if we
+dive into this topic by taking the example of a 4k display in the below
+example:
+
+.. kernel-figure:: pipeline_4k_no_split.svg
+
+Additionally, let's take a look at parts of the DTN log (see
+'Documentation/gpu/amdgpu/display/dc-debug.rst' for more information) since
+this log can help us to see part of this pipeline behavior in real-time::
+
+ HUBP:  format  addr_hi  width  height ...
+ [ 0]:      8h      81h   3840    2160
+ [ 1]:      0h       0h      0       0
+ [ 2]:      0h       0h      0       0
+ [ 3]:      0h       0h      0       0
+ [ 4]:      0h       0h      0       0
+ ...
+ MPCC:  OPP  DPP ...
+ [ 0]:   0h   0h ...
+
+The first thing to notice from the diagram and DTN log it is the fact that we
+have different clock domains for each part of the DCN blocks. In this example,
+we have just a single **pipeline** where the data flows from DCHUB to DIO, as
+we intuitively expect. Nonetheless, DCN is flexible, as mentioned before, and
+we can split this single pipe differently, as described in the below diagram:
+
+.. kernel-figure:: pipeline_4k_split.svg
+
+Now, if we inspect the DTN log again we can see some interesting changes::
+
+ HUBP:  format  addr_hi  width  height ...
+ [ 0]:      8h      81h   1920    2160 ...
+ ...
+ [ 4]:      0h       0h      0       0 ...
+ [ 5]:      8h      81h   1920    2160 ...
+ ...
+ MPCC:  OPP  DPP ...
+ [ 0]:   0h   0h ...
+ [ 5]:   0h   5h ...
+
+From the above example, we now split the display pipeline into two vertical
+parts of 1920x2160 (i.e., 3440x2160), and as a result, we could reduce the
+clock frequency in the DPP part. This is not only useful for saving power but
+also to better handle the required throughput. The idea to keep in mind here is
+that the pipe configuration can vary a lot according to the display
+configuration, and it is the DML's responsibility to set up all required
+configuration parameters for multiple scenarios supported by our hardware.
+
 Global Sync
 -----------
 
Documentation/gpu/amdgpu/display/pipeline_4k_no_split.svg new file +958 -0

958 lines not on this page · raw message

Documentation/gpu/amdgpu/display/pipeline_4k_split.svg new file +1062 -0

1,062 lines not on this page · raw message

Signature · 1 line
2.25.1
#13 lore raw

Re: [PATCH 2/3] Documentation/gpu: Add an explanation about the DCN pipeline

Quoted 33 lines · On Mon, 13 Jun 2022 16:01:11 -0400 Rodrigo Siqueira <[email protected]>…

In the DCN code, we constantly talk about hardware pipeline, pipeline, or even just pipes, which is a concept that is not obvious to everyone. For this reason, this commit expands the DCN overview explanation by adding a new section that describes what a pipeline is from the DCN perspective.

Cc Harry Wentland <[email protected]>
Cc Nicholas Kazlauskas <[email protected]>
Cc Bhawanpreet Lakha <[email protected]>
Cc Hersen Wu <[email protected]>
Cc Alex Hung <[email protected]>
Cc Pierre-Eric Pelloux-Prayer <[email protected]>
Cc Leo Li <[email protected]>
Cc Simon Ser <[email protected]>
Cc Pekka Paalanen <[email protected]>
Cc Sean Paul <[email protected]>
Cc Mark Yacoub <[email protected]>
Cc Pierre-Loup <[email protected]>
Cc Michel Dänzer <[email protected]>
---
 .../gpu/amdgpu/display/dcn-overview.rst       |   58 +
 .../amdgpu/display/pipeline_4k_no_split.svg   |  958 +++++++++++++++
 .../gpu/amdgpu/display/pipeline_4k_split.svg  | 1062 +++++++++++++++++
 3 files changed, 2078 insertions(+)
 create mode 100644 Documentation/gpu/amdgpu/display/pipeline_4k_no_split.svg
 create mode 100644 Documentation/gpu/amdgpu/display/pipeline_4k_split.svg
Documentation/gpu/amdgpu/display/dcn-overview.rst +0 -0
index f98624d7828e..430e363fccae 100644

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 126

Quoted 3 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +0 -0
@@ -124,6 +124,64 @@ depth format), bit-depth reduction/dithering would kick in. In OPP, we would
 also apply a regamma function to introduce the gamma removed earlier back.
 Eventually, we output data in integer format at DIO.
 

Hi,

always nice to see more docs!

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 133

Quoted 7 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +7 -0
+AMD Hardware Pipeline
+---------------------
+
+When discussing graphics on Linux, the **pipeline** term is often overloaded
+with multiple meanings, and DCN adds another context to this word. In the DCN
+driver, we use the term **hardware pipeline** or  **pipeline** or just **pipe**
+to indicate how many DCN blocks we are instantiating to address some specific

This seems to say that a "pipeline" is a number, a count (how many).

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 134

Quoted 1 line
Documentation/gpu/amdgpu/display/dcn-overview.rst +1 -0
+configuration. DCN composes those pipelines together when the data reach the

This seems to imply that each "DCN block" is "a pipeline".

Is either true? Is a pipeline not a sequence of selected DCN blocks instead?

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 141

Quoted 7 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +7 -0
+end of the OPTC block before sending it to the display. We have this pipeline
+concept for trying to save energy and optimize bandwidth utilization while
+providing the maximum performance per watt. It is easier to discuss it if we
+dive into this topic by taking the example of a 4k display in the below
+example:
+
+.. kernel-figure:: pipeline_4k_no_split.svg

A bit too hard for me to look at SVGs from email patches, sorry.

You wouldn't happen to have a git branch available in Gitlab or Github or even cgit somewhere I could visit with my browser?

Thanks,
pq

Quoted 46 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +43 -0
+
+Additionally, let's take a look at parts of the DTN log (see
+'Documentation/gpu/amdgpu/display/dc-debug.rst' for more information) since
+this log can help us to see part of this pipeline behavior in real-time::
+
+ HUBP:  format  addr_hi  width  height ...
+ [ 0]:      8h      81h   3840    2160
+ [ 1]:      0h       0h      0       0
+ [ 2]:      0h       0h      0       0
+ [ 3]:      0h       0h      0       0
+ [ 4]:      0h       0h      0       0
+ ...
+ MPCC:  OPP  DPP ...
+ [ 0]:   0h   0h ...
+
+The first thing to notice from the diagram and DTN log it is the fact that we
+have different clock domains for each part of the DCN blocks. In this example,
+we have just a single **pipeline** where the data flows from DCHUB to DIO, as
+we intuitively expect. Nonetheless, DCN is flexible, as mentioned before, and
+we can split this single pipe differently, as described in the below diagram:
+
+.. kernel-figure:: pipeline_4k_split.svg
+
+Now, if we inspect the DTN log again we can see some interesting changes::
+
+ HUBP:  format  addr_hi  width  height ...
+ [ 0]:      8h      81h   1920    2160 ...
+ ...
+ [ 4]:      0h       0h      0       0 ...
+ [ 5]:      8h      81h   1920    2160 ...
+ ...
+ MPCC:  OPP  DPP ...
+ [ 0]:   0h   0h ...
+ [ 5]:   0h   5h ...
+
+From the above example, we now split the display pipeline into two vertical
+parts of 1920x2160 (i.e., 3440x2160), and as a result, we could reduce the
+clock frequency in the DPP part. This is not only useful for saving power but
+also to better handle the required throughput. The idea to keep in mind here is
+that the pipe configuration can vary a lot according to the display
+configuration, and it is the DML's responsibility to set up all required
+configuration parameters for multiple scenarios supported by our hardware.
+
 Global Sync
 -----------
 
#14 lore raw

Re: [PATCH 2/3] Documentation/gpu: Add an explanation about the DCN pipeline

Quoted 42 lines · Pekka Paalanen
Quoted 33 lines · On Mon, 13 Jun 2022 16:01:11 -0400 Rodrigo Siqueira <[email protected]>…

In the DCN code, we constantly talk about hardware pipeline, pipeline, or even just pipes, which is a concept that is not obvious to everyone. For this reason, this commit expands the DCN overview explanation by adding a new section that describes what a pipeline is from the DCN perspective.

Cc Harry Wentland <[email protected]>
Cc Nicholas Kazlauskas <[email protected]>
Cc Bhawanpreet Lakha <[email protected]>
Cc Hersen Wu <[email protected]>
Cc Alex Hung <[email protected]>
Cc Pierre-Eric Pelloux-Prayer <[email protected]>
Cc Leo Li <[email protected]>
Cc Simon Ser <[email protected]>
Cc Pekka Paalanen <[email protected]>
Cc Sean Paul <[email protected]>
Cc Mark Yacoub <[email protected]>
Cc Pierre-Loup <[email protected]>
Cc Michel Dänzer <[email protected]>
---
  .../gpu/amdgpu/display/dcn-overview.rst       |   58 +
  .../amdgpu/display/pipeline_4k_no_split.svg   |  958 +++++++++++++++
  .../gpu/amdgpu/display/pipeline_4k_split.svg  | 1062 +++++++++++++++++
  3 files changed, 2078 insertions(+)

create mode 100644 Documentation/gpu/amdgpu/display/pipeline_4k_no_split.svg create mode 100644 Documentation/gpu/amdgpu/display/pipeline_4k_split.svg

Documentation/gpu/amdgpu/display/dcn-overview.rst +0 -0
index f98624d7828e..430e363fccae 100644

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 126

Quoted 3 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +0 -0
@@ -124,6 +124,64 @@ depth format), bit-depth reduction/dithering would kick in. In OPP, we would
  also apply a regamma function to introduce the gamma removed earlier back.
  Eventually, we output data in integer format at DIO.
 

Hi,

Quoted 1 line

always nice to see more docs!

Hi Pekka,

First of all, thanks for your review.

Quoted 11 lines

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 133

Quoted 8 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +7 -0
+AMD Hardware Pipeline
+---------------------
+
+When discussing graphics on Linux, the **pipeline** term is often overloaded
+with multiple meanings, and DCN adds another context to this word. In the DCN
+driver, we use the term **hardware pipeline** or  **pipeline** or just **pipe**
+to indicate how many DCN blocks we are instantiating to address some specific

This seems to say that a "pipeline" is a number, a count (how many).

Quoted 6 lines

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 134

Quoted 1 line
Documentation/gpu/amdgpu/display/dcn-overview.rst +1 -0
+configuration. DCN composes those pipelines together when the data reach the

This seems to imply that each "DCN block" is "a pipeline".

Is either true? Is a pipeline not a sequence of selected DCN blocks instead?

In the DCN we can have different pipeline configurations. For example, see the below picture:

https://www.kernel.org/doc/html/latest/gpu/amdgpu/display/dcn-overview.html

Let's say that our pipeline policy does not allow pipe split, in that case we will have just one instance of each component. However, if we decided to split our pipe, we will have more instance of the other components (e.g., 2 DPP). I added some examples about it a little bit below in this patch.

Try to apply my series in your local kernel [1]. That way you will be able to open the svg files at Documentation/gpu/amdgpu/display/.

Maybe I can rephrase it, do you have some suggestions?

1. https://patchwork.freedesktop.org/series/105071/

Thanks
Siqueira

Quoted 63 lines

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 141

Quoted 7 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +7 -0
+end of the OPTC block before sending it to the display. We have this pipeline
+concept for trying to save energy and optimize bandwidth utilization while
+providing the maximum performance per watt. It is easier to discuss it if we
+dive into this topic by taking the example of a 4k display in the below
+example:
+
+.. kernel-figure:: pipeline_4k_no_split.svg

A bit too hard for me to look at SVGs from email patches, sorry.

You wouldn't happen to have a git branch available in Gitlab or Github or even cgit somewhere I could visit with my browser?

Thanks,
pq

Quoted 46 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +43 -0
+
+Additionally, let's take a look at parts of the DTN log (see
+'Documentation/gpu/amdgpu/display/dc-debug.rst' for more information) since
+this log can help us to see part of this pipeline behavior in real-time::
+
+ HUBP:  format  addr_hi  width  height ...
+ [ 0]:      8h      81h   3840    2160
+ [ 1]:      0h       0h      0       0
+ [ 2]:      0h       0h      0       0
+ [ 3]:      0h       0h      0       0
+ [ 4]:      0h       0h      0       0
+ ...
+ MPCC:  OPP  DPP ...
+ [ 0]:   0h   0h ...
+
+The first thing to notice from the diagram and DTN log it is the fact that we
+have different clock domains for each part of the DCN blocks. In this example,
+we have just a single **pipeline** where the data flows from DCHUB to DIO, as
+we intuitively expect. Nonetheless, DCN is flexible, as mentioned before, and
+we can split this single pipe differently, as described in the below diagram:
+
+.. kernel-figure:: pipeline_4k_split.svg
+
+Now, if we inspect the DTN log again we can see some interesting changes::
+
+ HUBP:  format  addr_hi  width  height ...
+ [ 0]:      8h      81h   1920    2160 ...
+ ...
+ [ 4]:      0h       0h      0       0 ...
+ [ 5]:      8h      81h   1920    2160 ...
+ ...
+ MPCC:  OPP  DPP ...
+ [ 0]:   0h   0h ...
+ [ 5]:   0h   5h ...
+
+From the above example, we now split the display pipeline into two vertical
+parts of 1920x2160 (i.e., 3440x2160), and as a result, we could reduce the
+clock frequency in the DPP part. This is not only useful for saving power but
+also to better handle the required throughput. The idea to keep in mind here is
+that the pipe configuration can vary a lot according to the display
+configuration, and it is the DML's responsibility to set up all required
+configuration parameters for multiple scenarios supported by our hardware.
+
  Global Sync
  -----------
 
#15 lore raw

Re: [PATCH 2/3] Documentation/gpu: Add an explanation about the DCN pipeline

Quoted 73 lines · On Thu, 16 Jun 2022 17:59:21 -0400 Rodrigo Siqueira Jordao…
Quoted 42 lines · Pekka Paalanen
Quoted 33 lines · On Mon, 13 Jun 2022 16:01:11 -0400 Rodrigo Siqueira <[email protected]>…

In the DCN code, we constantly talk about hardware pipeline, pipeline, or even just pipes, which is a concept that is not obvious to everyone. For this reason, this commit expands the DCN overview explanation by adding a new section that describes what a pipeline is from the DCN perspective.

Cc Harry Wentland <[email protected]>
Cc Nicholas Kazlauskas <[email protected]>
Cc Bhawanpreet Lakha <[email protected]>
Cc Hersen Wu <[email protected]>
Cc Alex Hung <[email protected]>
Cc Pierre-Eric Pelloux-Prayer <[email protected]>
Cc Leo Li <[email protected]>
Cc Simon Ser <[email protected]>
Cc Pekka Paalanen <[email protected]>
Cc Sean Paul <[email protected]>
Cc Mark Yacoub <[email protected]>
Cc Pierre-Loup <[email protected]>
Cc Michel Dänzer <[email protected]>
---
  .../gpu/amdgpu/display/dcn-overview.rst       |   58 +
  .../amdgpu/display/pipeline_4k_no_split.svg   |  958 +++++++++++++++
  .../gpu/amdgpu/display/pipeline_4k_split.svg  | 1062 +++++++++++++++++
  3 files changed, 2078 insertions(+)

create mode 100644 Documentation/gpu/amdgpu/display/pipeline_4k_no_split.svg create mode 100644 Documentation/gpu/amdgpu/display/pipeline_4k_split.svg

Documentation/gpu/amdgpu/display/dcn-overview.rst +0 -0
index f98624d7828e..430e363fccae 100644

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 126

Quoted 3 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +0 -0
@@ -124,6 +124,64 @@ depth format), bit-depth reduction/dithering would kick in. In OPP, we would
  also apply a regamma function to introduce the gamma removed earlier back.
  Eventually, we output data in integer format at DIO.
 

Hi,

Quoted 1 line

always nice to see more docs!

Hi Pekka,

First of all, thanks for your review.

Quoted 11 lines

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 133

Quoted 8 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +7 -0
+AMD Hardware Pipeline
+---------------------
+
+When discussing graphics on Linux, the **pipeline** term is often overloaded
+with multiple meanings, and DCN adds another context to this word. In the DCN
+driver, we use the term **hardware pipeline** or  **pipeline** or just **pipe**
+to indicate how many DCN blocks we are instantiating to address some specific  

This seems to say that a "pipeline" is a number, a count (how many).

Quoted 6 lines

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 134

Quoted 1 line
Documentation/gpu/amdgpu/display/dcn-overview.rst +1 -0
+configuration. DCN composes those pipelines together when the data reach the  

This seems to imply that each "DCN block" is "a pipeline".

Is either true? Is a pipeline not a sequence of selected DCN blocks instead?

In the DCN we can have different pipeline configurations. For example, see the below picture:

https://www.kernel.org/doc/html/latest/gpu/amdgpu/display/dcn-overview.html

Quoted 8 lines

Let's say that our pipeline policy does not allow pipe split, in that case we will have just one instance of each component. However, if we decided to split our pipe, we will have more instance of the other components (e.g., 2 DPP). I added some examples about it a little bit below in this patch.

Try to apply my series in your local kernel [1]. That way you will be able to open the svg files at Documentation/gpu/amdgpu/display/.

Sorry, that's too much to ask for patch review on a project I don't work on.

Quoted 1 line

Maybe I can rephrase it, do you have some suggestions?

I'm not sure what your definition of a pipeline is. If you have split blocks, does that mean you have one pipeline or multiple pipelines? Does block splitting allow to create multiple pipelines in the sense that they do not all contain the exact same blocks?

The only thing I could suggest is to not say that a pipeline is a count. That would be done by replacing "to indicate how many DCN blocks" by "to indicate which DCN blocks".

Thanks,
pq

Quoted 21 lines

1. https://patchwork.freedesktop.org/series/105071/

Thanks
Siqueira

Quoted 16 lines

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 141

Quoted 7 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +7 -0
+end of the OPTC block before sending it to the display. We have this pipeline
+concept for trying to save energy and optimize bandwidth utilization while
+providing the maximum performance per watt. It is easier to discuss it if we
+dive into this topic by taking the example of a 4k display in the below
+example:
+
+.. kernel-figure:: pipeline_4k_no_split.svg  

A bit too hard for me to look at SVGs from email patches, sorry.

You wouldn't happen to have a git branch available in Gitlab or Github or even cgit somewhere I could visit with my browser?

Thanks,
pq

#16 lore raw

Re: [PATCH 2/3] Documentation/gpu: Add an explanation about the DCN pipeline

Quoted 63 lines · Pekka Paalanen
Quoted 60 lines · On Thu, 16 Jun 2022 17:59:21 -0400 Rodrigo Siqueira Jordao…
Quoted 42 lines · Pekka Paalanen
Quoted 33 lines · On Mon, 13 Jun 2022 16:01:11 -0400 Rodrigo Siqueira <[email protected]>…

In the DCN code, we constantly talk about hardware pipeline, pipeline, or even just pipes, which is a concept that is not obvious to everyone. For this reason, this commit expands the DCN overview explanation by adding a new section that describes what a pipeline is from the DCN perspective.

Cc Harry Wentland <[email protected]>
Cc Nicholas Kazlauskas <[email protected]>
Cc Bhawanpreet Lakha <[email protected]>
Cc Hersen Wu <[email protected]>
Cc Alex Hung <[email protected]>
Cc Pierre-Eric Pelloux-Prayer <[email protected]>
Cc Leo Li <[email protected]>
Cc Simon Ser <[email protected]>
Cc Pekka Paalanen <[email protected]>
Cc Sean Paul <[email protected]>
Cc Mark Yacoub <[email protected]>
Cc Pierre-Loup <[email protected]>
Cc Michel Dänzer <[email protected]>
---
  .../gpu/amdgpu/display/dcn-overview.rst       |   58 +
  .../amdgpu/display/pipeline_4k_no_split.svg   |  958 +++++++++++++++
  .../gpu/amdgpu/display/pipeline_4k_split.svg  | 1062 +++++++++++++++++
  3 files changed, 2078 insertions(+)

create mode 100644 Documentation/gpu/amdgpu/display/pipeline_4k_no_split.svg create mode 100644 Documentation/gpu/amdgpu/display/pipeline_4k_split.svg

Documentation/gpu/amdgpu/display/dcn-overview.rst +0 -0
index f98624d7828e..430e363fccae 100644

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 126

Quoted 3 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +0 -0
@@ -124,6 +124,64 @@ depth format), bit-depth reduction/dithering would kick in. In OPP, we would
  also apply a regamma function to introduce the gamma removed earlier back.
  Eventually, we output data in integer format at DIO.
 

Hi,

Quoted 1 line

always nice to see more docs!

Hi Pekka,

First of all, thanks for your review.

Quoted 11 lines

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 133

Quoted 8 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +7 -0
+AMD Hardware Pipeline
+---------------------
+
+When discussing graphics on Linux, the **pipeline** term is often overloaded
+with multiple meanings, and DCN adds another context to this word. In the DCN
+driver, we use the term **hardware pipeline** or  **pipeline** or just **pipe**
+to indicate how many DCN blocks we are instantiating to address some specific  

This seems to say that a "pipeline" is a number, a count (how many).

Quoted 7 lines
Quoted 7 lines
Quoted 6 lines

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 134

Quoted 1 line
Documentation/gpu/amdgpu/display/dcn-overview.rst +1 -0
+configuration. DCN composes those pipelines together when the data reach the  

This seems to imply that each "DCN block" is "a pipeline".

Is either true? Is a pipeline not a sequence of selected DCN blocks instead?

I think the confusion stems from unfortunate phrasing.

Siqueira, you seem to be saying that a **pipeline** is indicating how many DCN blocks we're instantiating. This is incorrect and confusing. We have a number of pipelines for each SoC, e.g. 4 for many APUs and 6 for many dGPUs.

A pipeline is a sequence of HW blocks:
HUBP > DPP > MPC > OPP > etc.

DC treats those as individual resources, so we build a pipeline conceptually by taking resources for all individual HW blocks to compose one pipeline but in actuality (in HW) these blocks are not arbitrarily configurable. What we can do is combine pipes at the MPC as such

HUBP > DPP > MPC > OPP > output
HUBP > DPP > MPC ^

In addition the DSC block can be assigned to pipelines as needed.

So, in short we have a number of pipelines per SoC. If we have 4 pipes we have 4 HUBPs, 4 DPPs, 4 MPCc, etc. There are few exceptions to that, like the DSC block.

On the output side we have a number of different encoders that are attached to our output connectors. Each pipe is routed to the appropriate encoder for the display output.

Harry

Quoted 54 lines
Quoted 6 lines

In the DCN we can have different pipeline configurations. For example, see the below picture:

https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.kernel.org%2Fdoc%2Fhtml%2Flatest%2Fgpu%2Famdgpu%2Fdisplay%2Fdcn-overview.html&amp;data=05%7C01%7Charry.wentland%40amd.com%7C5c60cfb7c2e5483e1b0d08da5033f013%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637910481695853125%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000%7C%7C%7C&amp;sdata=TX6JmSkn2ybJSqtOz7DRl4q745qzsCLYSkQ0RvAXsrk%3D&amp;reserved=0

Quoted 8 lines

Let's say that our pipeline policy does not allow pipe split, in that case we will have just one instance of each component. However, if we decided to split our pipe, we will have more instance of the other components (e.g., 2 DPP). I added some examples about it a little bit below in this patch.

Try to apply my series in your local kernel [1]. That way you will be able to open the svg files at Documentation/gpu/amdgpu/display/.

Sorry, that's too much to ask for patch review on a project I don't work on.

Quoted 1 line

Maybe I can rephrase it, do you have some suggestions?

I'm not sure what your definition of a pipeline is. If you have split blocks, does that mean you have one pipeline or multiple pipelines? Does block splitting allow to create multiple pipelines in the sense that they do not all contain the exact same blocks?

The only thing I could suggest is to not say that a pipeline is a count. That would be done by replacing "to indicate how many DCN blocks" by "to indicate which DCN blocks".

Thanks,
pq

Quoted 21 lines

1. https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpatchwork.freedesktop.org%2Fseries%2F105071%2F&amp;data=05%7C01%7Charry.wentland%40amd.com%7C5c60cfb7c2e5483e1b0d08da5033f013%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637910481695853125%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000%7C%7C%7C&amp;sdata=MYx6R8tIvFu6f46Rj7GWH4MIfa7bjYjdYitFnc6wNZs%3D&amp;reserved=0

Thanks
Siqueira

Quoted 16 lines

Documentation/gpu/amdgpu/display/dcn-overview.rst depth format), bit-depth reduction/dithering would kick in. In OPP, we would · line 141

Quoted 7 lines
Documentation/gpu/amdgpu/display/dcn-overview.rst +7 -0
+end of the OPTC block before sending it to the display. We have this pipeline
+concept for trying to save energy and optimize bandwidth utilization while
+providing the maximum performance per watt. It is easier to discuss it if we
+dive into this topic by taking the example of a 4k display in the below
+example:
+
+.. kernel-figure:: pipeline_4k_no_split.svg  

A bit too hard for me to look at SVGs from email patches, sorry.

You wouldn't happen to have a git branch available in Gitlab or Github or even cgit somewhere I could visit with my browser?

Thanks,
pq

#17 lore raw

PATCH 3/3 Documentation/gpu: Add Multiplane Overlay doc

In reply to Rodrigo Siqueira

Multiple plane overlay is a feature supported by AMD hardware, but it has specific details that deserve proper documentation. This commit introduces a documentation that describes some of the features, limitations, and use cases for this feature. Part of this documentation came from some discussion in the public upstream [1][2].

[1]. https://lore.kernel.org/amd-gfx/3qY-QeukF_Q_MJeIXAuBjO4szbS4jRtqkTifXnbnN3bp88SxVodFQRpah3mIIVJq24DUkF6g0rOGdCmSqTvVxx9LCGEItmzLw8uWU44jtXE=@emersion.fr/
[2]. https://lore.kernel.org/amd-gfx/[email protected]/

Cc Harry Wentland <[email protected]>
Cc Nicholas Kazlauskas <[email protected]>
Cc Bhawanpreet Lakha <[email protected]>
Cc Hersen Wu <[email protected]>
Cc Alex Hung <[email protected]>
Cc Pierre-Eric Pelloux-Prayer <[email protected]>
Cc Simon Ser <[email protected]>
Cc Pekka Paalanen <[email protected]>
Cc Sean Paul <[email protected]>
Cc Mark Yacoub <[email protected]>
Cc Leo Li <[email protected]>
Cc Pierre-Loup <[email protected]>
Cc Michel Dänzer <[email protected]>
---
 Documentation/gpu/amdgpu/display/index.rst    |   1 +
 .../gpu/amdgpu/display/mpo-cursor.svg         | 435 ++++++++++++++++++
 .../gpu/amdgpu/display/mpo-overview.rst       | 242 ++++++++++
 .../multi-display-hdcp-mpo-less-pipe-ex.svg   | 220 +++++++++
 .../amdgpu/display/multi-display-hdcp-mpo.svg | 171 +++++++
 .../single-display-mpo-multi-video.svg        | 339 ++++++++++++++
 .../gpu/amdgpu/display/single-display-mpo.svg | 266 +++++++++++
 7 files changed, 1674 insertions(+)
 create mode 100644 Documentation/gpu/amdgpu/display/mpo-cursor.svg
 create mode 100644 Documentation/gpu/amdgpu/display/mpo-overview.rst
 create mode 100644 Documentation/gpu/amdgpu/display/multi-display-hdcp-mpo-less-pipe-ex.svg
 create mode 100644 Documentation/gpu/amdgpu/display/multi-display-hdcp-mpo.svg
 create mode 100644 Documentation/gpu/amdgpu/display/single-display-mpo-multi-video.svg
 create mode 100644 Documentation/gpu/amdgpu/display/single-display-mpo.svg
Documentation/gpu/amdgpu/display/index.rst +1 -0
index c1fb2fb3c710..f8a4f53d70d8 100644
@@ -28,4 +28,5 @@ table of content:
    display-manager.rst
    dc-debug.rst
    dcn-overview.rst
+   mpo-overview.rst
    dc-glossary.rst
Documentation/gpu/amdgpu/display/mpo-cursor.svg new file +435 -0

435 lines not on this page · raw message

Documentation/gpu/amdgpu/display/mpo-overview.rst new file +242 -0

242 lines not on this page · raw message

Documentation/gpu/amdgpu/display/multi-display-hdcp-mpo-less-pipe-ex.svg new file +220 -0

220 lines not on this page · raw message

Documentation/gpu/amdgpu/display/multi-display-hdcp-mpo.svg new file +171 -0

171 lines not on this page · raw message

Documentation/gpu/amdgpu/display/single-display-mpo-multi-video.svg new file +339 -0

339 lines not on this page · raw message

Documentation/gpu/amdgpu/display/single-display-mpo.svg new file +266 -0

266 lines not on this page · raw message

Signature · 1 line
2.25.1