| Age | Commit message (Collapse) | Author |
|---|
|
When purging the program hash table, the refcount _should_ be one since
the program is referenced by the hash table. Need to explicitly set to
zero before calling delete().
Also, purge high-level shader hash tables before low-level program hash tables.
|
|
Use _mesa_reference_vert/fragprog() wherever we assign program pointers.
Fixes a memory corruption bug found with glean/api2 test.
|
|
|
|
shader/shader program objects to avoid memory access error.
|
|
This reverts commit 3ffd11f71d021f672b9bc15b3c39c155a0e2fecb.
|
|
|
|
|
|
Currently a callback delete_shader_cb is used for deleting shader
and shader program objects. Mesa detaches all attached shaders in
_mesa_free_shader_program_data when deleting shader program objects. However
it is likely that these shaders have been freed in _mesa_free_shader,
which will result in unexpected behaviour. This fix uses a single callback for
shader program objects and deletes shader program objects before shader objects.
|
|
|
|
|
|
|
|
This fixes the X server crash reported by KeithP on Aug 29.
The old context's FBOs will be unreferenced during context destruction so
there's no memleak with this change.
|
|
group's objects (Shane Blackett)
|
|
|
|
previously did.
Basically, in update_framebuffer() (which should be called after an FBO is bound
with MakeCurrent or BindFramebuffer) we check if the FBO is a window-system FBO.
If it is, update the FBO's GL_READ/DRAW_BUFFER state according to the context state.
Old code still in place but disabled with #if 0 / #endif.
|
|
|
|
Free FBOs before textures since the later may be referenced by the former.
Need to bind the context we're destroying if there isn't a current context
so that ctx->DeleteTexture() etc can be used.
|
|
|
|
Use new _mesa_reference_texobj() function for referencing/unreferencing
textures. Add new assertions/tests to try to detect invalid usage of
deleted textures.
|
|
|
|
Make sure the relevant fields in window fbs get updated at appropriate time
(those are NOT the same as fbos!!!), and fix up related code accordingly.
This is a bit ugly, but there's a reason the issues section in EXT_fbo is
a couple hundred pages long...
Hopefully correct now.
|
|
|
|
the framebuffer objects attached to drawables can have invalidate state
associated with them, since for the window framebuffer this is per-context
state and not per-fbo state. Since drivers rely on that information
(otherwise would need to check if currently the window-framebuffer is
bound in a lot of places) fix it up in _mesa_make_current (ugly).
(Brought over from i915tex_privbuffers, where it fixes xdemos/wincopy
when switching to front buffer rendering.)
|
|
of -I flags.
|
|
an app issue, not a mesa bug.
|
|
|
|
Shadow sampling from texture arrays is still not implemented. Everything
else should be there, though.
|
|
|
|
|
|
constants.
Previously, this limited us to 12 temp regs for vertex programs. Many vertex
shaders could exceed that. This forces us to stop using t_vb_arbprogram.c
for now because of its particular register indexing scheme. Need to increase
bits allocated for register indexing, etc.
|
|
|
|
|
|
Was removed during glsl-compiler work. Still need to go back and revisit this
because of the interaction with fragment shaders...
|
|
|
|
Conflicts:
src/mesa/main/context.c
|
|
Use _mesa_reference_framebuffer() and _mesa_unreference_framebuffer() functions
to be sure reference counting is done correctly. Additional assertions are
done too. Note _mesa_dereference_framebuffer() renamed to "unreference" as
that's more accurate.
|
|
Also, added DeletePending field to gl_framebuffer used when a window has been
deleted, but there still may be rendering contexts attached to the
gl_framebuffer object.
|
|
Conflicts:
src/mesa/main/state.c
src/mesa/shader/program.c
src/mesa/shader/program.h
src/mesa/shader/programopt.c
src/mesa/shader/slang/slang_execute.c
src/mesa/sources
src/mesa/swrast/s_arbshader.c
src/mesa/swrast/s_context.c
src/mesa/swrast/s_span.c
src/mesa/swrast/s_zoom.c
src/mesa/tnl/t_context.c
src/mesa/tnl/t_save_api.c
src/mesa/tnl/t_vb_arbprogram.c
src/mesa/tnl/t_vp_build.c
src/mesa/tnl/t_vtx_eval.c
|
|
Conflicts:
src/mesa/main/texcompress_s3tc.c
src/mesa/tnl/t_array_api.c
|
|
This leaves one last XFree86Server ifdef in Mesa core.
---
Bug 9285: misc glcore, xmesa cleanups
ACKed by Ian Romanick.
|
|
The old GLcore interface was replaced in Xorg 7.1 with the addition of AIGLX,
it is only used by DDX's which are known not to work with the new DIX glx code.
---
Bug 9285: misc glcore, xmesa cleanups
ACKed by Ian Romanick.
|
|
into vbo-0.2
Conflicts:
src/mesa/array_cache/sources
src/mesa/drivers/dri/i965/brw_context.c
src/mesa/drivers/dri/i965/brw_draw.c
src/mesa/drivers/dri/i965/brw_fallback.c
src/mesa/drivers/dri/i965/brw_vs_emit.c
src/mesa/drivers/dri/i965/brw_vs_tnl.c
src/mesa/drivers/dri/mach64/mach64_context.c
src/mesa/main/extensions.c
src/mesa/main/getstring.c
src/mesa/tnl/sources
src/mesa/tnl/t_save_api.c
src/mesa/tnl/t_save_playback.c
src/mesa/tnl/t_vtx_api.c
src/mesa/tnl/t_vtx_exec.c
src/mesa/vbo/vbo_attrib.h
src/mesa/vbo/vbo_exec_api.c
src/mesa/vbo/vbo_save_api.c
src/mesa/vbo/vbo_save_draw.c
|
|
Initial fog value was being set to {0,0,0,0}. This results in vector
size 4, but isn't necessary. The regular {0,0,0,1} works fine.
|
|
|
|
|
|
Use the gl_shader struct as it should be.
Renamed gl_linked_program to gl_shader_program.
Store both shaders and programs in the same hash table and use the Type field
to distinguish them.
|
|
|
|
New datatypes, constants, variables.
|
|
|
|
|
