Re: [Mesa3d-dev] Additional vertex program instructions
Keith Whitwell wrote: Ian Romanick wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 This patch adds the missing NV_vertex_program2 and NV_vertex_program3 instructions to the various internal tables. It does not add support for them to the parser nor does it add support for them to the program interpreter. It will probably a bit before I can add support to the parser. I'm still trying to understand how src/mesa/shader/grammar works (and why that was used instead of flex byacc). I've not yet committed this patch because it breaks the instruction union in t_vb_arbprogram.h. The opcode field in that union was only 6 bits. When there were only 59+3 instructions that worked fine. However, this patch increases the base instruction count to 69. This requires 7 bits. The problem is that increasing the opcode width to 7 bits increases the size of instruction::rsw to 33 bits. :( For the purposes of the integration work, I wouldn't be unhappy to see the t_vb_arbprogram.* code turned off temporarily and let the nvvertexec.c executor do the job in the interim. I say this for a few reasons - one that I'm expecting great things from your college-project optimizer, and second that if I were to continue with t_vb_arbprogram.*, there would be a signficiant rework based on what I've learned so far, and thirdly, it will make the ongoing reorg easier in the meantime. Oh, and while we're talking about archiving code, all the TCC stuff for fragment programs can definitely be put in that category. Keith --- SF.Net email is sponsored by: Tame your development challenges with Apache's Geronimo App Server. Download it for free - -and be entered to win a 42 plasma tv or your very own Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
Re: [Mesa3d-dev] Additional vertex program instructions
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Brian Paul wrote:
Ian Romanick wrote:
This patch adds the missing NV_vertex_program2 and NV_vertex_program3
instructions to the various internal tables. It does not add support
for them to the parser nor does it add support for them to the program
interpreter.
It will probably a bit before I can add support to the parser. I'm
still trying to understand how src/mesa/shader/grammar works (and why
that was used instead of flex byacc).
The ARB program parser is a bit of a mystery to me too.
:(
If I'd rewrite the parsers toay I'd probably use lex/yacc. I originally
implemented my own recursive descent parsers for the NV extensions
because the language was pretty simple. Judging from LOC and
understandability, I don't think I did too bad.
I think initially the PSU students are going to develop their code
outside Mesa. This is just to keep the learing curve down. Once
they're able to generate something (and we can verify that it's a
correct something), we'll work it into Mesa CVS (probably a branch).
We (read: I) were looking at building a standalone test framework for
that. I still hadn't decided how I wanted to get programs, especially
programs that use the added instructions, in. I had considered writing
a simple parser in lex byacc, but I'd like to avoid doing too much
extra work. However, it's starting to sound like a better idea. :(
Index: src/mesa/shader/program.c
===
RCS file: /cvs/mesa/Mesa/src/mesa/shader/program.c,v
retrieving revision 1.35
diff -u -d -r1.35 program.c
--- src/mesa/shader/program.c5 Nov 2005 20:18:18 -1.35
+++ src/mesa/shader/program.c7 Nov 2005 16:57:34 -
[...]
- { OPCODE_TXD,TXD, 3 },
+ { OPCODE_TXD,TXD, 2 },
Why that change? TXD takes 3 vector arguments, the texcoord, the
partial derivates wrt X, and the partial derivatives wrt Y.
It appears that you are correct. One of the various
NV_fragment_program* specs that I looked at listed it as taking only 2
vector inputs. I made the change based on that.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.6 (GNU/Linux)
iD8DBQFDb6SgX1gOwKyEAw8RAjrxAJoDji8mu966QQ0Cpx2ozCMPuTerTQCfTC4i
d3UQuMBsiAL13w+iLH9jcpI=
=I47A
-END PGP SIGNATURE-
---
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. Download
it for free - -and be entered to win a 42 plasma tv or your very own
Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php
___
Mesa3d-dev mailing list
Mesa3d-dev@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
Re: [Mesa3d-dev] Additional vertex program instructions
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Keith Whitwell wrote: Oh, and while we're talking about archiving code, all the TCC stuff for fragment programs can definitely be put in that category. You're saying to gut support for s_fragprog_to_c and s_tcc? I think I can handle that. -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iD8DBQFDb7xjX1gOwKyEAw8RAn30AKCB61TXIn4q2TfBgadXyw8McsLwrACeJy78 GiJcmZAsui6D+XH2Ql/Gias= =HgQt -END PGP SIGNATURE- --- SF.Net email is sponsored by: Tame your development challenges with Apache's Geronimo App Server. Download it for free - -and be entered to win a 42 plasma tv or your very own Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
Re: [Mesa3d-dev] Additional vertex program instructions
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Brian Paul wrote: Ian Romanick wrote: It will probably a bit before I can add support to the parser. I'm still trying to understand how src/mesa/shader/grammar works (and why that was used instead of flex byacc). The ARB program parser is a bit of a mystery to me too. If I'd rewrite the parsers toay I'd probably use lex/yacc. I originally implemented my own recursive descent parsers for the NV extensions because the language was pretty simple. Judging from LOC and understandability, I don't think I did too bad. How would using lex yacc impact non-Unix builds? Meaning, will that cause problems on Windows and Mac OS? -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iD8DBQFDb792X1gOwKyEAw8RAoYiAJ93YlQyIcvcdeFbuyofLDDJINzh2QCeNN7O 51FrMDbRFQ0uEocRULN5BGg= =dc3e -END PGP SIGNATURE- --- SF.Net email is sponsored by: Tame your development challenges with Apache's Geronimo App Server. Download it for free - -and be entered to win a 42 plasma tv or your very own Sony(tm)PSP. Click here to play: http://sourceforge.net/geronimo.php ___ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
