[Bug 37142] Too much vertex buffers uploads

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

Marek Ol??k  changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution|--- |WONTFIX

--- Comment #8 from Marek Ol??k  ---
Closing. Ian's comment explains it:

> EXT_compiled_vertex_array is dead.  It has been dead for almost 10 years. 
> No new applications should use it, and old applications that use it should
> be fast enough on modern hardware.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-- next part --
An HTML attachment was scrubbed...
URL: 

[Bug 37142] Too much vertex buffers uploads

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

--- Comment #7 from Pierre-Eric Pelloux-Prayer  
2011-05-17 05:33:51 PDT ---
Created an attachment (id=46804)
 View: https://bugs.freedesktop.org/attachment.cgi?id=46804
 Review: https://bugs.freedesktop.org/review?bug=37142=46804

vertex buffer upload caching

Here's a patch adding vertex buffer upload caching. 
It's not extensively tested (only with custom app and openarena), and is a bit
hackish (mainly regarding : ctx->Array.NewState field, but it seems unused
elsewhere).
Anyway, it allowed me to do some testing and confirm the initial guess : it
brings no performance improvements to openarena :-/ (but that was expected, as
profiling the game tells that u_upload_data < 1% cpu time)

So I guess the handling of EXT_compiled_vertex_array is useless - maybe with 1
exception : lower end video cards with lower bandwidth ? I can't test this
scenario here, as I only have AMD HD4850.

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

[Bug 37142] Too much vertex buffers uploads

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

--- Comment #7 from Pierre-Eric Pelloux-Prayer pell...@gmail.com 2011-05-17 
05:33:51 PDT ---
Created an attachment (id=46804)
 View: https://bugs.freedesktop.org/attachment.cgi?id=46804
 Review: https://bugs.freedesktop.org/review?bug=37142attachment=46804

vertex buffer upload caching

Here's a patch adding vertex buffer upload caching. 
It's not extensively tested (only with custom app and openarena), and is a bit
hackish (mainly regarding : ctx-Array.NewState field, but it seems unused
elsewhere).
Anyway, it allowed me to do some testing and confirm the initial guess : it
brings no performance improvements to openarena :-/ (but that was expected, as
profiling the game tells that u_upload_data  1% cpu time)

So I guess the handling of EXT_compiled_vertex_array is useless - maybe with 1
exception : lower end video cards with lower bandwidth ? I can't test this
scenario here, as I only have AMD HD4850.

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 37142] Too much vertex buffers uploads

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

--- Comment #6 from Sven Arvidsson  2011-05-13 09:22:41 PDT ---
(In reply to comment #3)
> That's why I asked for external opinions. I still feel that IF it can brings
> significant enough performance improvement, it could be useful - mainly 
> because
> openarena and all games using quake3 based engine are used in every 
> open-source
> drivers benchmarks.

Hey! A bug that can be solved by ignoring Phoronix!

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

[Bug 37142] Too much vertex buffers uploads

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

--- Comment #5 from Fabio Pedretti  2011-05-13 00:11:50 
PDT ---
Note that ioquake3 will probably support VBOs soon:
https://bugzilla.icculus.org/show_bug.cgi?id=4358

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

[Bug 37142] Too much vertex buffers uploads

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

--- Comment #5 from Fabio Pedretti fabio@libero.it 2011-05-13 00:11:50 
PDT ---
Note that ioquake3 will probably support VBOs soon:
https://bugzilla.icculus.org/show_bug.cgi?id=4358

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 37142] Too much vertex buffers uploads

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

--- Comment #6 from Sven Arvidsson s...@whiz.se 2011-05-13 09:22:41 PDT ---
(In reply to comment #3)
 That's why I asked for external opinions. I still feel that IF it can brings
 significant enough performance improvement, it could be useful - mainly 
 because
 openarena and all games using quake3 based engine are used in every 
 open-source
 drivers benchmarks.

Hey! A bug that can be solved by ignoring Phoronix!

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 37142] Too much vertex buffers uploads

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

--- Comment #4 from Pierre-Eric Pelloux-Prayer  
2011-05-12 13:27:09 PDT ---
(In reply to comment #2)
> (In reply to comment #0)
> > What do you think ?
> 
> You can't optimize the uploads because the sizes of all buffers are not known
> when glLockArraysEXT is called. The only things you've got are pointers to the
> buffers.

Sure, but I was thinking to something more like : 
- glLockArraysEXT does nothing more than curently
- uploading a vertex buffer is done as before except in one case : when trying
to upload the same buffer AND glLock is enabled => the uploading is skipped as
data is already in GPU memory (which is the case shown in the attached trace
file)

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

[Bug 37142] Too much vertex buffers uploads

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

--- Comment #3 from Pierre-Eric Pelloux-Prayer  
2011-05-12 13:19:13 PDT ---
> EXT_compiled_vertex_array is dead.  It has been dead for almost 10 years.  No
> new applications should use it, and old applications that use it should be 
> fast
> enough on modern hardware.  I really don't think we should add any code to 
> Mesa
> to make this case fast.  We'd be better off submitting patches to openarena to
> use VBOs instead.

That's why I asked for external opinions. I still feel that IF it can brings
significant enough performance improvement, it could be useful - mainly because
openarena and all games using quake3 based engine are used in every open-source
drivers benchmarks.

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

[Bug 37142] Too much vertex buffers uploads

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

--- Comment #2 from Marek Ol??k  2011-05-12 12:55:38 PDT 
---
(In reply to comment #0)
> What do you think ?

You can't optimize the uploads because the sizes of all buffers are not known
when glLockArraysEXT is called. The only things you've got are pointers to the
buffers.

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

[Bug 37142] Too much vertex buffers uploads

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

--- Comment #1 from Ian Romanick  2011-05-12 10:55:30 
PDT ---
(In reply to comment #0)
> Created an attachment (id=46638)
 --> (https://bugs.freedesktop.org/attachment.cgi?id=46638)
> openarena + verbose vertex buffer upload logs
> 
> Test case : openarena + anholt benchmark + r600g
> 
> I added several fprintf debug trace to mesa code around Vertex buffer
> uploading.
> Basically, openarena render things using :
> glVertexPointeer()
> glLockArraysEXT()
> several glDrawElements()
> glUnlockArraysEXT()
> 
> It seems that each call to glDrawElements implies a reupload of the vertex
> buffers, even if it has not changed (as glLock/Unlock calls tell, at least for
> part of the buffer).

EXT_compiled_vertex_array is dead.  It has been dead for almost 10 years.  No
new applications should use it, and old applications that use it should be fast
enough on modern hardware.  I really don't think we should add any code to Mesa
to make this case fast.  We'd be better off submitting patches to openarena to
use VBOs instead.

> Another related bug (it seems) is in : cso_set_vertex_buffers() which does a
> test before calling : util_copy_vertex_buffers and pipe->set_vertex_buffers
> This test always return true, thus the 2 above functions are always called. 
> The
> test is always true because it memcmp all pipe_vertex_buffer, which contains a
> 'buffer' pointer, which changes at each frame (see st_draw.c:349).
> 
> I'm trying to build a patch which fix the cso_set_vertex_buffers problem and
> then, taking advantage of glLock/Unlock calls to fix the upload issue.
> 
> What do you think ?

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

[Bug 37142] Too much vertex buffers uploads

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

Ian Romanick  changed:

   What|Removed |Added

  Attachment #46638|application/octet-stream|text/plain
  mime type||

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

[Bug 37142] Too much vertex buffers uploads

[bugzilla-dae...@freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

Michel D?nzer  changed:

   What|Removed |Added

Product|DRI |Mesa
Version|DRI CVS |git
  Component|DRM/Radeon  |Drivers/Gallium/r600

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.

[Bug 37142] Too much vertex buffers uploads

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

Ian Romanick i...@freedesktop.org changed:

   What|Removed |Added

  Attachment #46638|application/octet-stream|text/plain
  mime type||

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 37142] Too much vertex buffers uploads

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

--- Comment #1 from Ian Romanick i...@freedesktop.org 2011-05-12 10:55:30 PDT 
---
(In reply to comment #0)
 Created an attachment (id=46638)
 -- (https://bugs.freedesktop.org/attachment.cgi?id=46638)
 openarena + verbose vertex buffer upload logs
 
 Test case : openarena + anholt benchmark + r600g
 
 I added several fprintf debug trace to mesa code around Vertex buffer
 uploading.
 Basically, openarena render things using :
 glVertexPointeer()
 glLockArraysEXT()
 several glDrawElements()
 glUnlockArraysEXT()
 
 It seems that each call to glDrawElements implies a reupload of the vertex
 buffers, even if it has not changed (as glLock/Unlock calls tell, at least for
 part of the buffer).

EXT_compiled_vertex_array is dead.  It has been dead for almost 10 years.  No
new applications should use it, and old applications that use it should be fast
enough on modern hardware.  I really don't think we should add any code to Mesa
to make this case fast.  We'd be better off submitting patches to openarena to
use VBOs instead.

 Another related bug (it seems) is in : cso_set_vertex_buffers() which does a
 test before calling : util_copy_vertex_buffers and pipe-set_vertex_buffers
 This test always return true, thus the 2 above functions are always called. 
 The
 test is always true because it memcmp all pipe_vertex_buffer, which contains a
 'buffer' pointer, which changes at each frame (see st_draw.c:349).
 
 I'm trying to build a patch which fix the cso_set_vertex_buffers problem and
 then, taking advantage of glLock/Unlock calls to fix the upload issue.
 
 What do you think ?

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 37142] Too much vertex buffers uploads

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=37142

--- Comment #2 from Marek Olšák mar...@gmail.com 2011-05-12 12:55:38 PDT ---
(In reply to comment #0)
 What do you think ?

You can't optimize the uploads because the sizes of all buffers are not known
when glLockArraysEXT is called. The only things you've got are pointers to the
buffers.

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are the assignee for the bug.
___
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel