mesa.git
15 years agomesa: use elementBuf local var instead of ctx->Array.ElementArrayBufferObj
Brian Paul [Wed, 6 May 2009 18:23:28 +0000 (12:23 -0600)]
mesa: use elementBuf local var instead of ctx->Array.ElementArrayBufferObj

Makes no real difference, but more consistant.

15 years agoi965: Remove the forced lack of caching for renderbuffer surface state.
Eric Anholt [Wed, 6 May 2009 17:51:58 +0000 (10:51 -0700)]
i965: Remove the forced lack of caching for renderbuffer surface state.

This snuck in with the multi-draw-buffers commit, and is a major penalty
to performance.  It doesn't appear to be required, as the only dependency
the surface BO has is on the state key (and if there's some other dependency,
it should just be in the key).

This brings openarena performance up to almost 2% faster than Mesa 7.4.

15 years agoi965: Remove _NEW_PROGRAM from brw_wm_surfaces setup dependencies.
Eric Anholt [Wed, 6 May 2009 14:58:40 +0000 (07:58 -0700)]
i965: Remove _NEW_PROGRAM from brw_wm_surfaces setup dependencies.

This was a leftover from the brw_wm_constant_buffer change.

15 years agomesa: be smarter about allocating vert/frag program constants
Brian Paul [Wed, 6 May 2009 14:45:55 +0000 (08:45 -0600)]
mesa: be smarter about allocating vert/frag program constants

Try to re-use constants/literals more often to make best use of the
constant buffer space.

See bug 21354.

15 years agoi965: Split WM constant buffer update from other WM surfaces.
Eric Anholt [Wed, 6 May 2009 00:16:15 +0000 (17:16 -0700)]
i965: Split WM constant buffer update from other WM surfaces.

This can avoid re-uploading constant data when it isn't necessary, and is
a step towards not updating other surfaces just because constants change.
It also brings the upload of the constant buffer next to the creation.

This brings openarena performance up another 4%, to 91% of the Mesa 7.4 branch.

15 years agoi965: Disentangle VS constant surface state from WM surface state.
Eric Anholt [Tue, 5 May 2009 21:05:54 +0000 (14:05 -0700)]
i965: Disentangle VS constant surface state from WM surface state.

Also, only create VS surface state if there's a VS constant buffer to be
uploaded, and set the contents of the buffer at the same time as creation.

15 years agoi965: Don't create constant buffers if they won't be used.
Eric Anholt [Tue, 5 May 2009 20:07:33 +0000 (13:07 -0700)]
i965: Don't create constant buffers if they won't be used.

Really, the creation and upload of constants should be in the same place,
since they should only happen together, and a state flag should be
triggered by them so that we don't thrash state around so much for just
updating constants.  But this still recovers openarena performance by
another 19%, leaving us 16% behind Mesa 7.4 branch.

15 years agomesa: only use fallback texture when using shaders, not fixed-function (take two)
Brian Paul [Tue, 5 May 2009 22:29:04 +0000 (16:29 -0600)]
mesa: only use fallback texture when using shaders, not fixed-function (take two)

The semantics are a little different for shaders vs. fixed-function when
trying to use an incomplete texture.  The fallback texture returning
(0,0,0,1) should only be used with shaders.  For fixed function, the texture
unit is truly disabled/ignored.

Fixes glean fbo test regression.

15 years agoRevert "mesa: only use fallback texture when using shaders, not fixed-function"
Brian Paul [Tue, 5 May 2009 22:30:30 +0000 (16:30 -0600)]
Revert "mesa: only use fallback texture when using shaders, not fixed-function"

This reverts commit a0edbfb28fb2e670c657d52190a7e8b1ccf4f46e.

This patch didn't completely fix the problem.  The next patch will.

15 years agomesa: only use fallback texture when using shaders, not fixed-function
Brian Paul [Tue, 5 May 2009 22:29:04 +0000 (16:29 -0600)]
mesa: only use fallback texture when using shaders, not fixed-function

The semantics are a little different for shaders vs. fixed-function when
trying to use an incomplete texture.  The fallback texture returning
(0,0,0,1) should only be used with shaders.

Fixes glean fbo test regression.

15 years agomesa: only use fallback texture when using shaders, not fixed-function
Brian Paul [Tue, 5 May 2009 22:21:20 +0000 (16:21 -0600)]
mesa: only use fallback texture when using shaders, not fixed-function

The semantics are a little different for shaders vs. fixed-function when
trying to use an incomplete texture.  The fallback texture returning
(0,0,0,1) should only be used with shaders.

Fixes glean fbo test regression.

15 years agomesa: minor simplification in enable_texture(), updated comments
Brian Paul [Tue, 5 May 2009 22:17:55 +0000 (16:17 -0600)]
mesa: minor simplification in enable_texture(), updated comments

15 years agoMerge branch 'mesa_7_5_branch'
Brian Paul [Tue, 5 May 2009 16:13:13 +0000 (10:13 -0600)]
Merge branch 'mesa_7_5_branch'

15 years agogallium: remove redundant cube face #defines
Brian Paul [Tue, 5 May 2009 15:39:39 +0000 (09:39 -0600)]
gallium: remove redundant cube face #defines

15 years agogallium: fix texcoords for cubemap mipmap generation
Brian Paul [Tue, 5 May 2009 16:11:51 +0000 (10:11 -0600)]
gallium: fix texcoords for cubemap mipmap generation

We need to use (s,t,r) cubemap coords when sampling from the cubemap texture
so the rendered quad gets texture samples from the correct cube face.

15 years agoprogs/trivial: add test for vertex program invarient transform
Keith Whitwell [Tue, 5 May 2009 12:00:44 +0000 (13:00 +0100)]
progs/trivial: add test for vertex program invarient transform

15 years agomesa: more complete fix for transform_invarient glitches
Keith Whitwell [Tue, 5 May 2009 11:12:28 +0000 (12:12 +0100)]
mesa: more complete fix for transform_invarient glitches

Add a new flag mvp_with_dp4 in the context, and use that to switch
both ffvertex.c and programopt.c vertex transformation code to
either DP4 or MUL/MAD implementations.

15 years agomesa/main: set PREFER_DP4 to match position_invarient code
Keith Whitwell [Fri, 1 May 2009 17:20:42 +0000 (18:20 +0100)]
mesa/main: set PREFER_DP4 to match position_invarient code

This is a quick fix for z fighting in quake4 caused by the mismatch
between vertex transformation here and in the position_invarient code.
Full fix would be to make this driver-tunable and adjust both
position_invarient and ffvertex_prog.c code to respect driver
preferences.

15 years agowgl: Implemente SwapLayerBuffers.
José Fonseca [Fri, 1 May 2009 15:30:08 +0000 (16:30 +0100)]
wgl: Implemente SwapLayerBuffers.

15 years agoscons: Don't use deprecated Options.
José Fonseca [Fri, 1 May 2009 15:12:17 +0000 (16:12 +0100)]
scons: Don't use deprecated Options.

15 years agoutil: Limit the stack walk to avoid referencing undefined memory.
José Fonseca [Thu, 30 Apr 2009 12:10:58 +0000 (13:10 +0100)]
util: Limit the stack walk to avoid referencing undefined memory.

15 years agogallium/tgsi: hack around linker/archiver breakage
Keith Whitwell [Thu, 30 Apr 2009 12:09:34 +0000 (13:09 +0100)]
gallium/tgsi: hack around linker/archiver breakage

Add a dummy function which exists only so that tgsi_text_translate()
doesn't get magic-ed out of the libtgsi.a archive by the build system.
Don't remove unless you know this has been fixed - check on
mingw/scons builds as well.

15 years agoprogs/trivial: add vbo-noninterleaved test
Keith Whitwell [Thu, 30 Apr 2009 11:35:59 +0000 (12:35 +0100)]
progs/trivial: add vbo-noninterleaved test

15 years agowgl: Include alpha bits in pixel format's cColorBits field.
José Fonseca [Thu, 30 Apr 2009 11:24:08 +0000 (12:24 +0100)]
wgl: Include alpha bits in pixel format's cColorBits field.

15 years agogallium/draw: cope with unused vertex_elements
Keith Whitwell [Thu, 30 Apr 2009 09:59:19 +0000 (10:59 +0100)]
gallium/draw: cope with unused vertex_elements

15 years agotrivial: add line-flat.c
Keith Whitwell [Wed, 29 Apr 2009 13:21:41 +0000 (14:21 +0100)]
trivial: add line-flat.c

15 years agowgl: UINT_PTR null value is an integral type, so return 0 instead of NULL.
José Fonseca [Tue, 28 Apr 2009 18:46:56 +0000 (19:46 +0100)]
wgl: UINT_PTR null value is an integral type, so return 0 instead of NULL.

15 years agopb: Save the stack backtrace when creating/mapping a debug buffer.
José Fonseca [Tue, 28 Apr 2009 17:53:52 +0000 (18:53 +0100)]
pb: Save the stack backtrace when creating/mapping a debug buffer.

15 years agopb: Dump the fenced buffer sizes.
José Fonseca [Tue, 28 Apr 2009 16:07:01 +0000 (17:07 +0100)]
pb: Dump the fenced buffer sizes.

15 years agonv04: small fix again
Stephane Marchesin [Tue, 5 May 2009 10:03:38 +0000 (12:03 +0200)]
nv04: small fix again

15 years agoFix an nv04 bug (thanks Mhopf).
Stephane Marchesin [Tue, 5 May 2009 09:29:58 +0000 (11:29 +0200)]
Fix an nv04 bug (thanks Mhopf).

15 years agomesa: also print program params/constants when dumping shaders to disk
Brian Paul [Mon, 4 May 2009 17:14:35 +0000 (11:14 -0600)]
mesa: also print program params/constants when dumping shaders to disk

15 years agomesa: remove some unfinished/devel code
Brian Paul [Mon, 4 May 2009 17:13:35 +0000 (11:13 -0600)]
mesa: remove some unfinished/devel code

15 years agoglx: replace Xmalloc() calls with Xcalloc()
Brian Paul [Mon, 4 May 2009 13:52:55 +0000 (07:52 -0600)]
glx: replace Xmalloc() calls with Xcalloc()

Fixes a bug where psp->WaitX was uninitialized.  Reported by Chris Clayton.

15 years agodraw: added some inf/nan debug code (disabled)
Brian Paul [Fri, 1 May 2009 23:52:23 +0000 (17:52 -0600)]
draw: added some inf/nan debug code (disabled)

15 years agogallium/glx: say we're direct rendering
Brian Paul [Thu, 23 Apr 2009 16:24:19 +0000 (10:24 -0600)]
gallium/glx: say we're direct rendering

15 years agomesa: in glReadBufer() set _NEW_BUFFERS, not _NEW_PIXEL
Brian Paul [Sat, 2 May 2009 00:31:04 +0000 (18:31 -0600)]
mesa: in glReadBufer() set _NEW_BUFFERS, not _NEW_PIXEL

Since GL_READ_BUFFER is historically part of the gl_pixel_attrib group
it made sense to signal changes with _NEW_PIXEL.  But now with FBOs it's
also part of the framebuffer state.

Now _NEW_PIXEL strictly indicates pixels transfer state changes.

This change avoids framebuffer state validation when any random bit of
pixel-transfer state is set.

DRI drivers updated too:  don't check _NEW_COLOR when updating framebuffer
state.  I think that was just copied from the Xlib driver because we care
about dither enable/disable state there.

15 years agoMerge branch 'mesa_7_5_branch'
Brian Paul [Sat, 2 May 2009 00:21:02 +0000 (18:21 -0600)]
Merge branch 'mesa_7_5_branch'

15 years agost: create renderbuffer's pipe_surface in st_render_texture()
Brian Paul [Sat, 2 May 2009 00:17:34 +0000 (18:17 -0600)]
st: create renderbuffer's pipe_surface in st_render_texture()

Previously we created the pipe_surface during framebuffer validation.
But if we did a glCopyTex[Sub]Image() before anything else we wouldn't yet
have the surface.  This fixes that.

15 years agomesa: fix state validation bug for glCopyTex[Sub]Image()
Brian Paul [Sat, 2 May 2009 00:04:07 +0000 (18:04 -0600)]
mesa: fix state validation bug for glCopyTex[Sub]Image()

We need to make sure the framebuffer state is up to date to make sure we
read pixels from the right buffer when doing a texture image copy.

15 years agost: if st_swapbuffers() is called for single-buffered visual don't crash
Brian Paul [Fri, 1 May 2009 22:44:04 +0000 (16:44 -0600)]
st: if st_swapbuffers() is called for single-buffered visual don't crash

Furthermore, return pointer(s) to the front color buffer(s).

15 years agoxlib winsys: use new st_swapbuffers() function
Brian Paul [Fri, 1 May 2009 18:28:57 +0000 (12:28 -0600)]
xlib winsys: use new st_swapbuffers() function

The front/back buffer pointers are truly swapped (when there is an actual
front buffer).

This fixes some issues seen with apps/tests that draw to both the front
and back color buffers.  The true swap allows us to avoid the (potentially)
slow surface_copy() call in update_framebuffer_state() and is cleaner
overall.

15 years agost: when creating an on-demand front color buffer, init to back buffer image
Brian Paul [Fri, 1 May 2009 18:48:46 +0000 (12:48 -0600)]
st: when creating an on-demand front color buffer, init to back buffer image

When we create a new front color buffer (user called glDrawBuffer(GL_FRONT))
initialize it to the contents of the back buffer.  Any previous call to
SwapBuffers() would have done that in effect, so make it reality.

15 years agost: added st_renderbuffer::defined flag
Brian Paul [Fri, 1 May 2009 18:25:42 +0000 (12:25 -0600)]
st: added st_renderbuffer::defined flag

Indicates whether there's defined image contents, or garbage/don't care.
This is set when we draw into a renderbuffer and cleared when we resize/
reallocate a renderbuffer or do a buffer swap (back buffer becomes undefined).

We use this to determine whether the front color buffer has been drawn to,
and whether to display its contents upon glFlush/Finish(), when the new
st_swapbuffers() function is used.

15 years agomesa: fix comment
Brian Paul [Fri, 1 May 2009 18:12:43 +0000 (12:12 -0600)]
mesa: fix comment

15 years agoMerge branch 'mesa_7_5_branch'
Brian Paul [Fri, 1 May 2009 17:57:00 +0000 (11:57 -0600)]
Merge branch 'mesa_7_5_branch'

15 years agoFix symbol list for mangled Mesa on Darwin.
Tom Fogal [Wed, 29 Apr 2009 04:03:12 +0000 (21:03 -0700)]
Fix symbol list for mangled Mesa on Darwin.

When building mangled Mesa on Darwin, the exported symbols are
named `_mgluWhatever' instead of simply `_gluWhatever'.  When
using a list of exported symbols via the system ld's
`-exported_symbols_list' command line option (as done by mklib),
this resulted in error messages about exporting symbols which do
not exist.

Fortunately the file format accepts simple wildcards.  This throws
a wildcard so that the symbol list will match both the mangled and
non-mangled names, preventing the warning and actually exporting
the correct symbols in one shot.

15 years agoset: new st_swapbuffers() which does a true front/back buffer swap
Brian Paul [Fri, 1 May 2009 16:50:04 +0000 (10:50 -0600)]
set: new st_swapbuffers() which does a true front/back buffer swap

The pointers to the front/back renderbuffers are exchanged.
This new function isn't actually used yet...

15 years agoactually add debug to the debug config
Zack Rusin [Fri, 1 May 2009 16:50:38 +0000 (12:50 -0400)]
actually add debug to the debug config

15 years agoegl_xlib: Pass RTLD_LAZY to dlopen.
José Fonseca [Tue, 24 Mar 2009 19:18:06 +0000 (19:18 +0000)]
egl_xlib: Pass RTLD_LAZY to dlopen.

dlopen manpage mandates that either RTLD_LAZY or RTLD_NOW flags must be
passed. Not doing so was causing a NULL return on debian unstable x86-64.

15 years agoTest either GL_FRONT_LEFT or GL_FRONT for front-buffer rendering
Ian Romanick [Tue, 28 Apr 2009 17:08:57 +0000 (10:08 -0700)]
Test either GL_FRONT_LEFT or GL_FRONT for front-buffer rendering

For non-stereo visuals, which is all we support, we treat
GL_FRONT_LEFT as GL_FRONT.  However, they are technically different,
and they have different enum values.  Test for either one to determine
if we're in front-buffer rendering mode.

This fix was suggested by Pierre Willenbrock.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
15 years agost: when double buffered, only create front color buffer on demand
Brian Paul [Fri, 1 May 2009 16:15:21 +0000 (10:15 -0600)]
st: when double buffered, only create front color buffer on demand

Before we always created the front color buffer, even if was never used.
This can save some memory.

15 years agoMerge branch 'const-buffer-changes'
Brian Paul [Fri, 1 May 2009 15:37:14 +0000 (09:37 -0600)]
Merge branch 'const-buffer-changes'

Conflicts:

src/mesa/drivers/dri/i965/brw_curbe.c
src/mesa/drivers/dri/i965/brw_vs_emit.c
src/mesa/drivers/dri/i965/brw_wm_glsl.c

15 years agomesa: create/use a fallback texture when bound texture is incomplete
Brian Paul [Fri, 1 May 2009 15:30:32 +0000 (09:30 -0600)]
mesa: create/use a fallback texture when bound texture is incomplete

When a GLSL sampler reads from an incomplete texture it should
return (0,0,0,1).  Instead of jumping through hoops in all the drivers
to make this happen, just create/install a fallback texture with those
texel values.

Fixes piglit/fp-incomplete-tex on i965 and more importantly, fixes some
GPU lockups when trying to sample from missing surfaces.  If a binding
table entry is NULL, it seems that sampling sometimes works, but not
always (lockup).

Todo: create a fallback texture for each type of texture target?

15 years agodocs: initial 7.6 release notes page
Brian Paul [Fri, 1 May 2009 15:24:19 +0000 (09:24 -0600)]
docs: initial 7.6 release notes page

15 years agomesa: bump version to 7.6 (devel)
Brian Paul [Fri, 1 May 2009 15:22:20 +0000 (09:22 -0600)]
mesa: bump version to 7.6 (devel)

15 years agomesa: remove -devel suffix from version
Brian Paul [Fri, 1 May 2009 15:19:59 +0000 (09:19 -0600)]
mesa: remove -devel suffix from version

15 years agoconfigs: bump MESA_MINOR to 5
Brian Paul [Fri, 1 May 2009 15:15:14 +0000 (09:15 -0600)]
configs: bump MESA_MINOR to 5

15 years agodocs: fix typo
Brian Paul [Thu, 30 Apr 2009 23:27:51 +0000 (17:27 -0600)]
docs: fix typo

15 years agoradeon: Don't even bother with things too big to fit into our card.
Corbin Simpson [Fri, 1 May 2009 14:29:14 +0000 (07:29 -0700)]
radeon: Don't even bother with things too big to fit into our card.

15 years agor300-gallium, radeon-winsys: Hide radeon_cs from r300 pipe.
Corbin Simpson [Fri, 1 May 2009 13:01:52 +0000 (06:01 -0700)]
r300-gallium, radeon-winsys: Hide radeon_cs from r300 pipe.

15 years agor300-gallium, radeon-winsys: Space accounting.
Corbin Simpson [Fri, 1 May 2009 12:54:53 +0000 (05:54 -0700)]
r300-gallium, radeon-winsys: Space accounting.

It is no longer optional in current libdrm, so it was time to actually
start counting our BOs.

15 years agor300-gallium, radeon-winsys: Reorganize r300_winsys header, break ABI.
Corbin Simpson [Fri, 1 May 2009 12:03:56 +0000 (05:03 -0700)]
r300-gallium, radeon-winsys: Reorganize r300_winsys header, break ABI.

Make things more consistent, prepare for more function hooks.

15 years agor300-gallium: Don't bother with conditional double define.
Corbin Simpson [Fri, 1 May 2009 11:47:50 +0000 (04:47 -0700)]
r300-gallium: Don't bother with conditional double define.

We'll just forever leave it in r300_winsys.h since it's needed for
whichever winsys is hosting the pipe.

15 years agoradeon: Fix cast and comment.
Corbin Simpson [Tue, 28 Apr 2009 11:37:56 +0000 (04:37 -0700)]
radeon: Fix cast and comment.

15 years agodocs: asst. updates for 7.5 release
Brian Paul [Thu, 30 Apr 2009 23:17:31 +0000 (17:17 -0600)]
docs: asst. updates for 7.5 release

15 years agodocs: bring in 7.4 doc updates from mesa_7_4_branch
Brian Paul [Thu, 30 Apr 2009 23:13:22 +0000 (17:13 -0600)]
docs: bring in 7.4 doc updates from mesa_7_4_branch

15 years agodocs: notes about new version number meaning in release notes
Brian Paul [Thu, 30 Apr 2009 23:11:20 +0000 (17:11 -0600)]
docs: notes about new version number meaning in release notes

15 years agomesa: update linux-static, linux-x86-static configs
Brian Paul [Thu, 30 Apr 2009 23:04:01 +0000 (17:04 -0600)]
mesa: update linux-static, linux-x86-static configs

15 years agodemos: silence warning
Brian Paul [Thu, 30 Apr 2009 23:03:54 +0000 (17:03 -0600)]
demos: silence warning

15 years agomklib: replace if/expr with case
Tormod Volden [Thu, 30 Apr 2009 22:55:54 +0000 (16:55 -0600)]
mklib: replace if/expr with case

Saves forking an expr for every object.

15 years agomesa: Prepend "-Wl," to linking options
Tormod Volden [Thu, 30 Apr 2009 22:52:56 +0000 (16:52 -0600)]
mesa: Prepend "-Wl," to linking options

Let mklib ignore -Wl options inside the object list when building
static libraries

15 years agoi915simple: remove duplicated i915_context.c in C_SOURCES
Brian Paul [Thu, 30 Apr 2009 22:52:02 +0000 (16:52 -0600)]
i915simple: remove duplicated i915_context.c in C_SOURCES

15 years agor300: Increase reference count of texture objects referenced by current state.
Michel Dänzer [Thu, 30 Apr 2009 11:21:08 +0000 (13:21 +0200)]
r300: Increase reference count of texture objects referenced by current state.

Fixes a use-after-free reported in
http://bugs.freedesktop.org/show_bug.cgi?id=20539, so this possibly fixes that
bug. It has been confirmed to fix
http://bugs.freedesktop.org/show_bug.cgi?id=17895 .

15 years agogallium: Add SCons build support for the DRI state tracker.
Michel Dänzer [Thu, 30 Apr 2009 07:39:29 +0000 (09:39 +0200)]
gallium: Add SCons build support for the DRI state tracker.

15 years agoegl: Don't install demodriver
Dan Nicholson [Thu, 30 Apr 2009 03:33:50 +0000 (20:33 -0700)]
egl: Don't install demodriver

I don't think anyone besides a developer would ever want to use the demo
egl driver. Furthermore, egl would only ever load demodriver if it was
set via EGL_DRIVER in the environment. In that case, I think you can
point it to your mesa source directory.

15 years agoprogs/tests: Update ignores
Jakob Bornecrantz [Wed, 29 Apr 2009 19:44:03 +0000 (20:44 +0100)]
progs/tests: Update ignores

15 years agoprogs/tests: Add mipmap_comp for mipmap testing with compressed textures
Jakob Bornecrantz [Wed, 29 Apr 2009 19:17:21 +0000 (20:17 +0100)]
progs/tests: Add mipmap_comp for mipmap testing with compressed textures

15 years agoUse variable library name in pkg-config output.
Tom Fogal [Wed, 29 Apr 2009 16:32:46 +0000 (10:32 -0600)]
Use variable library name in pkg-config output.

Previously the pkg-config output files would contain e.g. `-lGL'
and `-lGLU', even if the user modified their configuration to
build libraries with different names.  This modifies the
pkg-config inputs, and corresponding makery, so that modifying the
output library name will cause the appropriate updated name to
appear in the pkg-config `-l' option.

Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
15 years agoautoconf: Add switch for optional EGL
Dan Nicholson [Wed, 29 Apr 2009 19:11:43 +0000 (12:11 -0700)]
autoconf: Add switch for optional EGL

EGL doesn't build on all platforms, so allow people to opt out.

15 years agoautoconf: Clean up some m4 usage
Dan Nicholson [Wed, 29 Apr 2009 13:49:27 +0000 (06:49 -0700)]
autoconf: Clean up some m4 usage

m4_fatal is equivalent to m4_errprint + m4_exit.

15 years agomesa: added _mesa_check_soa_dependencies() function
Brian Paul [Wed, 29 Apr 2009 17:52:06 +0000 (11:52 -0600)]
mesa: added _mesa_check_soa_dependencies() function

This function will check an instruction to see if there's data dependencies
between the dst and src registers if executed in an SOA manner.

15 years agotgsi: added tgsi_check_soa_dependencies() and related debug code (disabled)
Brian Paul [Wed, 29 Apr 2009 17:56:57 +0000 (11:56 -0600)]
tgsi: added tgsi_check_soa_dependencies() and related debug code (disabled)

The TGSI interpeter operates in SOA style.  We need to check for data
dependencies in instructions which read from and write to the same register.
For now just adding some debug code to detect that condition.  Actual fixes
to follow.

15 years agomesa: added _mesa_check_soa_dependencies() function
Brian Paul [Wed, 29 Apr 2009 17:52:06 +0000 (11:52 -0600)]
mesa: added _mesa_check_soa_dependencies() function

This function will check an instruction to see if there's data dependencies
between the dst and src registers if executed in an SOA manner.

15 years agosoftpipe: return PIPE_UNREFERENCED in softpipe_is_buffer/texture_referenced()
Brian Paul [Tue, 28 Apr 2009 20:29:27 +0000 (14:29 -0600)]
softpipe: return PIPE_UNREFERENCED in softpipe_is_buffer/texture_referenced()

This allows the engine demo to run again (avoid crash in VBO code).
This stuff still needs to be revisited someday though...

15 years agomesa/st: protect internal flushes with FLUSH_CURRENT
Keith Whitwell [Tue, 28 Apr 2009 16:50:19 +0000 (17:50 +0100)]
mesa/st: protect internal flushes with FLUSH_CURRENT

Already doing this for driver.flush()

15 years agomesa/main: protect driver.finish with FLUSH_CURRENT
Keith Whitwell [Tue, 28 Apr 2009 16:49:50 +0000 (17:49 +0100)]
mesa/main: protect driver.finish with FLUSH_CURRENT

Already doing this for driver.flush()

15 years agomesa/st: remove duplicate offset calculation
Keith Whitwell [Tue, 28 Apr 2009 13:51:11 +0000 (14:51 +0100)]
mesa/st: remove duplicate offset calculation

15 years agogallium/draw: add ability to print out active pipeline stages
Keith Whitwell [Tue, 28 Apr 2009 13:50:05 +0000 (14:50 +0100)]
gallium/draw: add ability to print out active pipeline stages

15 years agomesa/st: translate VERT_ATTRIB_GENERIC8..15 in st_translate_vertex_program
Keith Whitwell [Mon, 27 Apr 2009 14:33:44 +0000 (15:33 +0100)]
mesa/st: translate VERT_ATTRIB_GENERIC8..15 in st_translate_vertex_program

It seems quake4 can hit these attributes sometimes.

15 years agoprogs: add fflushes for cygwin
Keith Whitwell [Tue, 28 Apr 2009 10:49:55 +0000 (11:49 +0100)]
progs: add fflushes for cygwin

15 years agoutil/indices: remove debug prints
Keith Whitwell [Mon, 27 Apr 2009 18:36:42 +0000 (19:36 +0100)]
util/indices: remove debug prints

15 years agomesa/st: workaround for crashes in st_copy_texsubimage
Keith Whitwell [Mon, 27 Apr 2009 17:56:26 +0000 (18:56 +0100)]
mesa/st: workaround for crashes in st_copy_texsubimage

Proper fix for this hasn't been identified, but avoid crashing.

15 years agowgl: Store current HDC/HGLRC in stw_context.
José Fonseca [Mon, 27 Apr 2009 19:24:55 +0000 (20:24 +0100)]
wgl: Store current HDC/HGLRC in stw_context.

Less TLS lookups.

15 years agoR300: add quadpipe overrides
Alex Deucher [Tue, 28 Apr 2009 16:37:29 +0000 (12:37 -0400)]
R300: add quadpipe overrides

RV410 SE chips only have 1 quadpipe.
Also, handle other R300 chip with quadpipe override

15 years agoi965: avoid segfault in intel_update_renderbuffers() if using DRI1
Brian Paul [Tue, 28 Apr 2009 15:58:44 +0000 (09:58 -0600)]
i965: avoid segfault in intel_update_renderbuffers() if using DRI1

15 years agoswrast: add missing break in clamp_rect_coord_linear()
Brian Paul [Tue, 28 Apr 2009 15:05:19 +0000 (09:05 -0600)]
swrast: add missing break in clamp_rect_coord_linear()

See bug 21461.

15 years agodemos: asst. updates, clean-ups
Brian Paul [Mon, 27 Apr 2009 23:01:59 +0000 (17:01 -0600)]
demos: asst. updates, clean-ups

15 years agogallium dri st: Propagate the drawable info when we bind to new drawables.
Thomas Hellstrom [Tue, 28 Apr 2009 12:43:41 +0000 (14:43 +0200)]
gallium dri st: Propagate the drawable info when we bind to new drawables.

Signed-off-by: Thomas Hellstrom <thellstrom-at-vmware-dot-com>
15 years agogallium dri st: Use st_get_current() instead of GET_CURRENT_CONTEXT()
Thomas Hellstrom [Tue, 28 Apr 2009 12:20:45 +0000 (14:20 +0200)]
gallium dri st: Use st_get_current() instead of GET_CURRENT_CONTEXT()

Signed-off-by: Thomas Hellstrom <thellstrom-at-vmware-dot-com>