Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION
My 2c, I don't care about doxygen markup anymore -- as there is no critical mass behind the idea of automatic generated docs from source here --, but I can't relate with this notion that too many comments in a header is pollution. Who would want to see just the enumeration names or the structures without annotations, and to what purpose? I really can't understand... For a developer using or modifying the headers having the documentation right there seems the best: one can quickly lookup the documentation of a symbol by jumping to its definition in his favorite editor or IDE. It doesn't need to open another file. For me it would make sense to have a high level manual / guide in a separate document, but reference documentation is better in the source, revisioned in the same way. The best example of a well documented header that I've come across is http://cgit.freedesktop.org/mesa/mesa/tree/src/gallium/drivers/svga/include/svga3d_reg.h (written by Micah Dowty and others at VMware). It has no special markup, it omits the obvious stuff, and anybody that reads that file understands both the semantics and the syntax of the interface. My ideal for pipe/p_*.h would be something like that. All these ascii tools look nice -- I've used asciidoc many times and I'm quite happy with it --, but it is yet another markup to learn -- it appears easy but it is like every other apparently intuitive and non-intrusive markup language out there in wikis and so on, all slightly different from each other, and if if you forget a space, or use '*'' instead of '-' or some minor detail like that, you get crap in the output. You find yourself looking up the tool documentation more often that you'd expect! Anyway, having more documentation, wherever it goes, is a very good thing IMO. So whatever works is fine by be. Jose From: Keith Whitwell [kei...@vmware.com] Sent: Saturday, December 19, 2009 16:42 To: tom fogal Cc: mesa3d-dev@lists.sourceforge.net Subject: Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION Tom, These both look great! The markup language is almost invisible and the raw documents look a lot like the sort of ascii stuff I tend to write before figuring out how to turn it into something fancier... Now just to pick one... Keith From: tom fogal [tfo...@alumni.unh.edu] Sent: Saturday, December 19, 2009 8:32 AM To: Keith Whitwell Cc: Corbin Simpson; Christoph Bumiller; mesa3d-dev@lists.sourceforge.net Subject: Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION Keith Whitwell kei...@vmware.com writes: Ideally that would mean being able to produce a single patch or patch series that contains the changes to the headers *and* the changes to the spec, so th at all is being considered together. That means in turn having the spec in the git tree. Ideally we'd have an online version as well, in a wiki style. I'm interested in suggestions for how to make this happen. Ideally we *wouldn't* pollute the header files with so much text and markup that they become unreadable - ie. I don't really like the idea of doing the spec with doxygen - I'd prefer the headers to be human-readable. Thoughts? I'm partial to / been considering both asciidoc and sphinx in my own projects. http://www.methods.co.nz/asciidoc/ http://sphinx.pocoo.org/ Both are text-based formats that are intended to be readable `raw', but are processable into more presentable formats. The git documentation: http://www.kernel.org/pub/software/scm/git/docs/user-manual.html is done with asciidoc. Python documentation: http://docs.python.org/library/index.html is done with sphinx. -tom From: Corbin Simpson [mostawesomed...@gmail.com] Sent: Friday, December 18, 2009 7:11 PM To: Christoph Bumiller Cc: mesa3d-dev@lists.sourceforge.net Subject: Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHA DE_FIRST_CONVENTION NAK to this series. Keith hasn't responded, although I expect that he would also NAK this. I would much rather have quads just never respect flatshade_first as part of the spec, than jump through these weird param hoops. Should somebody be documenting the API? I keep on having these kinds of stupid edge questions come up; r300 apparently is the quirkiest hardware of that generation. ~ C. On Fri, Dec 18, 2009 at 2:15 AM, Christoph Bumiller e0425...@student.tuwien.ac.at wrote: Marek Olšák schrieb: Hi, GL_ARB_provoking_vertex states that quads are not required to abide the provoking vertex convention, and the actual hardware and/or driver behavior can be queried with GL_QUADS_FOLLOW_PROVOKING_VERTEX_CONVENTION. I'd like to add a new PIPE_CAP_* to query for this capability in Gallium, as it appears R3xx-R5xx hardware doesn't support the first vertex
Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION
I've started documenting things, using rest (since I already know it). I'll push a branch in a bit once I've got enough of it down. ~ C. On Sat, Dec 19, 2009 at 8:42 AM, Keith Whitwell kei...@vmware.com wrote: Tom, These both look great! The markup language is almost invisible and the raw documents look a lot like the sort of ascii stuff I tend to write before figuring out how to turn it into something fancier... Now just to pick one... Keith From: tom fogal [tfo...@alumni.unh.edu] Sent: Saturday, December 19, 2009 8:32 AM To: Keith Whitwell Cc: Corbin Simpson; Christoph Bumiller; mesa3d-dev@lists.sourceforge.net Subject: Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION Keith Whitwell kei...@vmware.com writes: Ideally that would mean being able to produce a single patch or patch series that contains the changes to the headers *and* the changes to the spec, so th at all is being considered together. That means in turn having the spec in the git tree. Ideally we'd have an online version as well, in a wiki style. I'm interested in suggestions for how to make this happen. Ideally we *wouldn't* pollute the header files with so much text and markup that they become unreadable - ie. I don't really like the idea of doing the spec with doxygen - I'd prefer the headers to be human-readable. Thoughts? I'm partial to / been considering both asciidoc and sphinx in my own projects. http://www.methods.co.nz/asciidoc/ http://sphinx.pocoo.org/ Both are text-based formats that are intended to be readable `raw', but are processable into more presentable formats. The git documentation: http://www.kernel.org/pub/software/scm/git/docs/user-manual.html is done with asciidoc. Python documentation: http://docs.python.org/library/index.html is done with sphinx. -tom From: Corbin Simpson [mostawesomed...@gmail.com] Sent: Friday, December 18, 2009 7:11 PM To: Christoph Bumiller Cc: mesa3d-dev@lists.sourceforge.net Subject: Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHA DE_FIRST_CONVENTION NAK to this series. Keith hasn't responded, although I expect that he would also NAK this. I would much rather have quads just never respect flatshade_first as part of the spec, than jump through these weird param hoops. Should somebody be documenting the API? I keep on having these kinds of stupid edge questions come up; r300 apparently is the quirkiest hardware of that generation. ~ C. On Fri, Dec 18, 2009 at 2:15 AM, Christoph Bumiller e0425...@student.tuwien.ac.at wrote: Marek Olšák schrieb: Hi, GL_ARB_provoking_vertex states that quads are not required to abide the provoking vertex convention, and the actual hardware and/or driver behavior can be queried with GL_QUADS_FOLLOW_PROVOKING_VERTEX_CONVENTION. I'd like to add a new PIPE_CAP_* to query for this capability in Gallium, as it appears R3xx-R5xx hardware doesn't support the first vertex convention for quads and I'd like the driver to behave correctly. Fortunately, other primitive types are supported. I decided to use the name quads follow flatshade_first convention instead of provoking vertex convention because the actual state variable in pipe_rasterizer_state is called flatshade_first. The attached patch: - adds PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION - adds the query in the Mesa state tracker - and updates softpipe and llvmpipe to return 1 when this cap is queried, and r300g to explicitly return 0 You can add a return 1 for nv50, too, in case you do push this patch. I just tested and for quads I can also make them use either the first or the last vertex's colour, i.e. flatshade convention is respected. Thanks, Christoph. Please review/push. Cheers. Marek -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and ea sy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev --- --- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and eas y Join now
Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION
On 12/19/2009 04:11 AM, Corbin Simpson wrote: NAK to this series. Keith hasn't responded, although I expect that he would also NAK this. I would much rather have quads just never respect flatshade_first as part of the spec, than jump through these weird param hoops. Should somebody be documenting the API? I keep on having these kinds of stupid edge questions come up; r300 apparently is the quirkiest hardware of that generation. ~ C. How about we kick out quads altogether ? Having to flip the first/last bit depending on *what's being drawn* is quirky. I realize nv50 don't follow the convention for quads (at least that's what the spec says), so I'm outnumbered, but I don't see the harm in following what OpenGL does here, i.e. just have the query. On Fri, Dec 18, 2009 at 2:15 AM, Christoph Bumiller e0425...@student.tuwien.ac.at wrote: Marek Olšák schrieb: Hi, GL_ARB_provoking_vertex states that quads are not required to abide the provoking vertex convention, and the actual hardware and/or driver behavior can be queried with GL_QUADS_FOLLOW_PROVOKING_VERTEX_CONVENTION. I'd like to add a new PIPE_CAP_* to query for this capability in Gallium, as it appears R3xx-R5xx hardware doesn't support the first vertex convention for quads and I'd like the driver to behave correctly. Fortunately, other primitive types are supported. I decided to use the name quads follow flatshade_first convention instead of provoking vertex convention because the actual state variable in pipe_rasterizer_state is called flatshade_first. The attached patch: - adds PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION - adds the query in the Mesa state tracker - and updates softpipe and llvmpipe to return 1 when this cap is queried, and r300g to explicitly return 0 You can add a return 1 for nv50, too, in case you do push this patch. I just tested and for quads I can also make them use either the first or the last vertex's colour, i.e. flatshade convention is respected. Thanks, Christoph. Please review/push. Cheers. Marek -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION
Keith Whitwell kei...@vmware.com writes: Ideally that would mean being able to produce a single patch or patch series that contains the changes to the headers *and* the changes to the spec, so th at all is being considered together. That means in turn having the spec in the git tree. Ideally we'd have an online version as well, in a wiki style. I'm interested in suggestions for how to make this happen. Ideally we *wouldn't* pollute the header files with so much text and markup that they become unreadable - ie. I don't really like the idea of doing the spec with doxygen - I'd prefer the headers to be human-readable. Thoughts? I'm partial to / been considering both asciidoc and sphinx in my own projects. http://www.methods.co.nz/asciidoc/ http://sphinx.pocoo.org/ Both are text-based formats that are intended to be readable `raw', but are processable into more presentable formats. The git documentation: http://www.kernel.org/pub/software/scm/git/docs/user-manual.html is done with asciidoc. Python documentation: http://docs.python.org/library/index.html is done with sphinx. -tom From: Corbin Simpson [mostawesomed...@gmail.com] Sent: Friday, December 18, 2009 7:11 PM To: Christoph Bumiller Cc: mesa3d-dev@lists.sourceforge.net Subject: Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHA DE_FIRST_CONVENTION NAK to this series. Keith hasn't responded, although I expect that he would also NAK this. I would much rather have quads just never respect flatshade_first as part of the spec, than jump through these weird param hoops. Should somebody be documenting the API? I keep on having these kinds of stupid edge questions come up; r300 apparently is the quirkiest hardware of that generation. ~ C. On Fri, Dec 18, 2009 at 2:15 AM, Christoph Bumiller e0425...@student.tuwien.ac.at wrote: Marek Olák schrieb: Hi, GL_ARB_provoking_vertex states that quads are not required to abide the provoking vertex convention, and the actual hardware and/or driver behavior can be queried with GL_QUADS_FOLLOW_PROVOKING_VERTEX_CONVENTION. I'd like to add a new PIPE_CAP_* to query for this capability in Gallium, as it appears R3xx-R5xx hardware doesn't support the first vertex convention for quads and I'd like the driver to behave correctly. Fortunately, other primitive types are supported. I decided to use the name quads follow flatshade_first convention instead of provoking vertex convention because the actual state variable in pipe_rasterizer_state is called flatshade_first. The attached patch: - adds PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION - adds the query in the Mesa state tracker - and updates softpipe and llvmpipe to return 1 when this cap is queried, and r300g to explicitly return 0 You can add a return 1 for nv50, too, in case you do push this patch. I just tested and for quads I can also make them use either the first or the last vertex's colour, i.e. flatshade convention is respected. Thanks, Christoph. Please review/push. Cheers. Marek -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and ea sy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev --- --- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and eas y Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev -- Only fools are easily impressed by what is only barely beyond their reach. ~ Unknown Corbin Simpson mostawesomed...@gmail.com - - This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers
Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION
On Thu, 2009-12-17 at 20:41 -0800, Marek Olšák wrote: Hi, GL_ARB_provoking_vertex states that quads are not required to abide the provoking vertex convention, and the actual hardware and/or driver behavior can be queried with GL_QUADS_FOLLOW_PROVOKING_VERTEX_CONVENTION. I'd like to add a new PIPE_CAP_* to query for this capability in Gallium, as it appears R3xx-R5xx hardware doesn't support the first vertex convention for quads and I'd like the driver to behave correctly. Fortunately, other primitive types are supported. I decided to use the name quads follow flatshade_first convention instead of provoking vertex convention because the actual state variable in pipe_rasterizer_state is called flatshade_first. The attached patch: - adds PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION - adds the query in the Mesa state tracker - and updates softpipe and llvmpipe to return 1 when this cap is queried, and r300g to explicitly return 0 Please review/push. Cheers. Marek Hi Marek, One problem I have with this patch and many of past get_param changes for that matter, is that it changes default behavior and forces all pipe drivers to be updated. It is not the first time that a get_param changes broke drivers because it. I happened with PIPE_CAP_BLEND_EQUATION_SEPARATE. As the number of drivers is increases this is a no-no. It also complicates writing drivers since they have to answer a large number of queries, most of which are specific to one or two particular hardware devices. IMO, there are two ways to overcome this: a) when introducing new PIPE_CAP_xxx have its semantics such that 0 will yield the previous behavior b) change get_param/paramf prototype so that it is passed the default value as an argument. That said, reading http://www.opengl.org/registry/specs/EXT/provoking_vertex.txt , issue #2 (How should quads be handled?) it seems that 0 is actually a better default -- provoking vertex of quads does not apply to D3D and is only relevant for that extension. So I don't oppose for this to go in as is. But, independent of this change, lets fix the get_param/paramf calls. Keith, does b) above sound good to you? Jose -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION
Marek Olšák schrieb: Hi, GL_ARB_provoking_vertex states that quads are not required to abide the provoking vertex convention, and the actual hardware and/or driver behavior can be queried with GL_QUADS_FOLLOW_PROVOKING_VERTEX_CONVENTION. I'd like to add a new PIPE_CAP_* to query for this capability in Gallium, as it appears R3xx-R5xx hardware doesn't support the first vertex convention for quads and I'd like the driver to behave correctly. Fortunately, other primitive types are supported. I decided to use the name quads follow flatshade_first convention instead of provoking vertex convention because the actual state variable in pipe_rasterizer_state is called flatshade_first. The attached patch: - adds PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION - adds the query in the Mesa state tracker - and updates softpipe and llvmpipe to return 1 when this cap is queried, and r300g to explicitly return 0 You can add a return 1 for nv50, too, in case you do push this patch. I just tested and for quads I can also make them use either the first or the last vertex's colour, i.e. flatshade convention is respected. Thanks, Christoph. Please review/push. Cheers. Marek -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION
NAK to this series. Keith hasn't responded, although I expect that he would also NAK this. I would much rather have quads just never respect flatshade_first as part of the spec, than jump through these weird param hoops. Should somebody be documenting the API? I keep on having these kinds of stupid edge questions come up; r300 apparently is the quirkiest hardware of that generation. ~ C. On Fri, Dec 18, 2009 at 2:15 AM, Christoph Bumiller e0425...@student.tuwien.ac.at wrote: Marek Olšák schrieb: Hi, GL_ARB_provoking_vertex states that quads are not required to abide the provoking vertex convention, and the actual hardware and/or driver behavior can be queried with GL_QUADS_FOLLOW_PROVOKING_VERTEX_CONVENTION. I'd like to add a new PIPE_CAP_* to query for this capability in Gallium, as it appears R3xx-R5xx hardware doesn't support the first vertex convention for quads and I'd like the driver to behave correctly. Fortunately, other primitive types are supported. I decided to use the name quads follow flatshade_first convention instead of provoking vertex convention because the actual state variable in pipe_rasterizer_state is called flatshade_first. The attached patch: - adds PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION - adds the query in the Mesa state tracker - and updates softpipe and llvmpipe to return 1 when this cap is queried, and r300g to explicitly return 0 You can add a return 1 for nv50, too, in case you do push this patch. I just tested and for quads I can also make them use either the first or the last vertex's colour, i.e. flatshade convention is respected. Thanks, Christoph. Please review/push. Cheers. Marek -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev -- Only fools are easily impressed by what is only barely beyond their reach. ~ Unknown Corbin Simpson mostawesomed...@gmail.com -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
Re: [Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION
Okay, after experimenting a bit, I've found that only quads don't work, see 759fd1f25f3 for details. Thankfully, this is directly addressed in the GL spec; could we maybe redefine the Gallium API to say that quads and quadstrips shouldn't respect the flag? ~ C. On Fri, Dec 18, 2009 at 7:11 PM, Corbin Simpson mostawesomed...@gmail.com wrote: NAK to this series. Keith hasn't responded, although I expect that he would also NAK this. I would much rather have quads just never respect flatshade_first as part of the spec, than jump through these weird param hoops. Should somebody be documenting the API? I keep on having these kinds of stupid edge questions come up; r300 apparently is the quirkiest hardware of that generation. ~ C. On Fri, Dec 18, 2009 at 2:15 AM, Christoph Bumiller e0425...@student.tuwien.ac.at wrote: Marek Olšák schrieb: Hi, GL_ARB_provoking_vertex states that quads are not required to abide the provoking vertex convention, and the actual hardware and/or driver behavior can be queried with GL_QUADS_FOLLOW_PROVOKING_VERTEX_CONVENTION. I'd like to add a new PIPE_CAP_* to query for this capability in Gallium, as it appears R3xx-R5xx hardware doesn't support the first vertex convention for quads and I'd like the driver to behave correctly. Fortunately, other primitive types are supported. I decided to use the name quads follow flatshade_first convention instead of provoking vertex convention because the actual state variable in pipe_rasterizer_state is called flatshade_first. The attached patch: - adds PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION - adds the query in the Mesa state tracker - and updates softpipe and llvmpipe to return 1 when this cap is queried, and r300g to explicitly return 0 You can add a return 1 for nv50, too, in case you do push this patch. I just tested and for quads I can also make them use either the first or the last vertex's colour, i.e. flatshade convention is respected. Thanks, Christoph. Please review/push. Cheers. Marek -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev -- Only fools are easily impressed by what is only barely beyond their reach. ~ Unknown Corbin Simpson mostawesomed...@gmail.com -- Only fools are easily impressed by what is only barely beyond their reach. ~ Unknown Corbin Simpson mostawesomed...@gmail.com -- This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
[Mesa3d-dev] [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION
Hi,
GL_ARB_provoking_vertex states that quads are not required to abide
the provoking vertex convention, and the actual hardware and/or driver
behavior can be queried with
GL_QUADS_FOLLOW_PROVOKING_VERTEX_CONVENTION.
I'd like to add a new PIPE_CAP_* to query for this capability in
Gallium, as it appears R3xx-R5xx hardware doesn't support the first
vertex convention for quads and I'd like the driver to behave
correctly. Fortunately, other primitive types are supported.
I decided to use the name quads follow flatshade_first convention
instead of provoking vertex convention because the actual state
variable in pipe_rasterizer_state is called flatshade_first.
The attached patch:
- adds PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION
- adds the query in the Mesa state tracker
- and updates softpipe and llvmpipe to return 1 when this cap is
queried, and r300g to explicitly return 0
Please review/push.
Cheers.
Marek
From bbf279f81e03a6872507cbf533b382748c4ae015 Mon Sep 17 00:00:00 2001
From: =?utf-8?q?Marek=20Ol=C5=A1=C3=A1k?= mar...@gmail.com
Date: Fri, 18 Dec 2009 05:10:27 +0100
Subject: [PATCH] gallium: add PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION
---
src/gallium/drivers/llvmpipe/lp_screen.c |2 ++
src/gallium/drivers/r300/r300_screen.c |2 ++
src/gallium/drivers/softpipe/sp_screen.c |2 ++
src/gallium/include/pipe/p_defines.h |1 +
src/mesa/state_tracker/st_extensions.c |4
5 files changed, 11 insertions(+), 0 deletions(-)
diff --git a/src/gallium/drivers/llvmpipe/lp_screen.c b/src/gallium/drivers/llvmpipe/lp_screen.c
index 9b47415..fa786f9 100644
--- a/src/gallium/drivers/llvmpipe/lp_screen.c
+++ b/src/gallium/drivers/llvmpipe/lp_screen.c
@@ -110,6 +110,8 @@ llvmpipe_get_param(struct pipe_screen *screen, int param)
return 1;
case PIPE_CAP_BLEND_EQUATION_SEPARATE:
return 1;
+ case PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION:
+ return 1;
default:
return 0;
}
diff --git a/src/gallium/drivers/r300/r300_screen.c b/src/gallium/drivers/r300/r300_screen.c
index c2ea9b6..0b75ecf 100644
--- a/src/gallium/drivers/r300/r300_screen.c
+++ b/src/gallium/drivers/r300/r300_screen.c
@@ -138,6 +138,8 @@ static int r300_get_param(struct pipe_screen* pscreen, int param)
return 0;
case PIPE_CAP_BLEND_EQUATION_SEPARATE:
return 1;
+case PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION:
+return 0;
default:
debug_printf(r300: Implementation error: Bad param %d\n,
param);
diff --git a/src/gallium/drivers/softpipe/sp_screen.c b/src/gallium/drivers/softpipe/sp_screen.c
index bd3532d..510150b 100644
--- a/src/gallium/drivers/softpipe/sp_screen.c
+++ b/src/gallium/drivers/softpipe/sp_screen.c
@@ -91,6 +91,8 @@ softpipe_get_param(struct pipe_screen *screen, int param)
return 1;
case PIPE_CAP_BLEND_EQUATION_SEPARATE:
return 1;
+ case PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION:
+ return 1;
default:
return 0;
}
diff --git a/src/gallium/include/pipe/p_defines.h b/src/gallium/include/pipe/p_defines.h
index fe1390d..6a37893 100644
--- a/src/gallium/include/pipe/p_defines.h
+++ b/src/gallium/include/pipe/p_defines.h
@@ -393,6 +393,7 @@ enum pipe_transfer_usage {
#define PIPE_CAP_MAX_PREDICATE_REGISTERS 30
#define PIPE_CAP_MAX_COMBINED_SAMPLERS 31 /* Maximum texture image units accessible from vertex
and fragment shaders combined */
+#define PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION 32
/**
diff --git a/src/mesa/state_tracker/st_extensions.c b/src/mesa/state_tracker/st_extensions.c
index ef3cbc5..db3380b 100644
--- a/src/mesa/state_tracker/st_extensions.c
+++ b/src/mesa/state_tracker/st_extensions.c
@@ -125,6 +125,10 @@ void st_init_limits(struct st_context *st)
= CLAMP(screen-get_param(screen, PIPE_CAP_MAX_RENDER_TARGETS),
1, MAX_DRAW_BUFFERS);
+ c-QuadsFollowProvokingVertexConvention
+ = CLAMP(screen-get_param(screen, PIPE_CAP_QUADS_FOLLOW_FLATSHADE_FIRST_CONVENTION),
+ 0, 1);
+
/* Is TGSI_OPCODE_CONT supported? */
/* XXX separate query for early function return? */
st-ctx-Shader.EmitContReturn =
--
1.6.3.3
--
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support
A streamlined, 14 day to market process makes app distribution fast and easy
Join now and get one step closer to millions of Verizon customers
http://p.sf.net/sfu/verizon-dev2dev ___
Mesa3d-dev mailing list
Mesa3d-dev@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
