The use of 'minimum = 1' is definitely correct for GL_LINES.

It sounds like what's actually happening is that the stride of the vertex arrays isn't being taken correctly into account when chopping the large arrays into bite-sized chunks...

Keith
--- Begin Message ---
Bugs item #677007, was opened at 2003-01-29 20:23
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100003&aid=677007&group_id=3

Category: mesa-core
Group: Rendering Error
Status: Open
Resolution: None
Priority: 5
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: glDrawArrays with GL_LINES type

Initial Comment:
Seems like using glDrawArrays to draw GL_LINES is 
not working well. The endpoints get mixed up and the 
lines do not appear were they shoud be. Does the same 
thing in feedback mode.

Just check the attached image.

I've noticed this effect since version 3.4.2. Everything was
fine in 3.4.1 and earlier.

This effect can be seen on Linux ans SGI systems at 
least since I tested those two.

----------------------------------------------------------------------

Comment By: Heath Feather (heath30)
Date: 2003-03-28 17:00

Message:
Logged In: YES 
user_id=744523

I too have the same problem, going through the Mesa source
code revealed that the vertices are dispatched in 256 vertex
batches. So I changed my code to submit my vertices to
gldrawarrays in groups of 256 or less. at this point the
crazy lines disappeared, and all worked fine. Vertex arrays
of less that 256 vertices work fine too. I then went and
took a closer look at "t_array_api.c"  _tnl_DrawArrays
function. In this function a variable called "minimum" was
being initialized to 1 for my "GL_LINES objects", the effect
that this was having on the array was that instead of
getting x,y...xy...etc. I get yx,yx etc...The first x is
never seen, and allsubsequent vertices are now half of one
and half of the other. Setting this variable to 0 ,
rebuilding mesa and running my application again fixed the
problem. 

I took a slightly longer look at the code and think that
there is a signifcantly less convoluted way of batching the
vertices, "If that's what its trying to do". However I don't
have the time to fix and test it adequately.

Hope this helps in identifying the issue. 

P.S.

The code in 4.1 and 5.0 appeared to be the same, and have
the same problem.


Heath.

----------------------------------------------------------------------

Comment By: Brian Paul (brianp)
Date: 2003-02-08 15:46

Message:
Logged In: YES 
user_id=983

Can you provide a simple test program?


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100003&aid=677007&group_id=3


-------------------------------------------------------
This SF.net email is sponsored by:
The Definitive IT and Networking Event. Be There!
NetWorld+Interop Las Vegas 2003 -- Register today!
http://ads.sourceforge.net/cgi-bin/redirect.pl?keyn0001en
_______________________________________________
Mesa3d-dev mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/mesa3d-dev



--- End Message ---

Reply via email to