Zack Rusin [Tue, 2 Mar 2010 15:18:29 +0000 (10:18 -0500)]
llvmpipe: improve based on review from Jose and fix else clauses
else was broken in the outter most else statemants, plus the code
didn't need an inverted mask to compute the inverse of the current
condition.
Keith Whitwell [Tue, 2 Mar 2010 15:02:29 +0000 (15:02 +0000)]
llvmpipe: remove some ifdefs
No longer any need to protect expensive asserts with #ifdef DEBUG.
Keith Whitwell [Tue, 2 Mar 2010 14:41:35 +0000 (14:41 +0000)]
llvmpipe: reorganize transfer usage
Move transfer creation and mapping to the "scene" object, and out of
the rasterizer. The rasterizer operates on already-mapped
framebuffers only, and no longer needs a screen or context pointer.
The scene object has access to a pipe_context, and this reorg prepares
for moving transfer functionality from the screen to the context.
Brian Paul [Tue, 2 Mar 2010 14:50:15 +0000 (07:50 -0700)]
intel: passs correct context type to intel_prepare_render() call
Brian Paul [Tue, 2 Mar 2010 14:47:56 +0000 (07:47 -0700)]
i965: include macros.h to get CLAMP macro
Brian Paul [Tue, 2 Mar 2010 14:34:29 +0000 (07:34 -0700)]
glx: replace assertion with conditional
See fd.o bug 26832.
Michal Krol [Tue, 2 Mar 2010 12:35:30 +0000 (13:35 +0100)]
Merge branch 'gallium-no-rhw-position'
Conflicts:
src/gallium/drivers/r300/r300_blit.c
src/gallium/drivers/r300/r300_emit.c
src/gallium/drivers/r300/r300_state_derived.c
Marek Olšák [Tue, 2 Mar 2010 01:37:45 +0000 (02:37 +0100)]
util/blitter: Fix the viewport transformation for Z coordinates
When clearing buffers, the depth is specified in the range [0, 1]
and should be passed through blitter "as is".
Vinson Lee [Tue, 2 Mar 2010 09:15:02 +0000 (01:15 -0800)]
i915g: Add assert to check for null pointer dereference.
Vinson Lee [Tue, 2 Mar 2010 09:10:29 +0000 (01:10 -0800)]
glsl: Add assert to check input to strcmp.
Vinson Lee [Tue, 2 Mar 2010 09:01:23 +0000 (01:01 -0800)]
mesa: Add asserts to check inputs to memcpy.
Dave Airlie [Tue, 2 Mar 2010 04:05:21 +0000 (14:05 +1000)]
r600: fix blender picking.
This fixes the sw fallback for GL_SELECT picking modes.
Fixes object picking blender + depthpick test
http://bugs.freedesktop.org/show_bug.cgi?id=26419
Signed-off-by: Dave Airlie <airlied@redhat.com>
Zack Rusin [Tue, 2 Mar 2010 03:11:08 +0000 (22:11 -0500)]
llvmpipe: implement some control-flow
implements if/else/endif constructs and lays down the code for looping
and others. we create a conditional execution mask which decides which
of the four inputs are enabled for any store. it's used only if an
execution mask is present, otherwise we go through a direct store.
George Sapountzis [Tue, 2 Mar 2010 01:31:16 +0000 (03:31 +0200)]
glapi: fix compile with ES
José Fonseca [Mon, 1 Mar 2010 23:18:29 +0000 (23:18 +0000)]
glapi: Fix syntax.
Pauli Nieminen [Mon, 1 Mar 2010 21:47:35 +0000 (23:47 +0200)]
radeon: Reduce number of radeon_bo_map calls that require kernel entry.
DMA buffers are often remapped many times for singel cs. To reduce number of
mapping calls ending to kernel dma buffers are mapped when allocated and unmapped
when full or in cs flush. This makes indifual mapping calls in other parts of
code simple increment/decrement reference count which is faster than entering
kernel.
This improves Anholt's openarena benchmark from 36 fps to 44 fps.
Karl Schultz [Mon, 1 Mar 2010 19:21:13 +0000 (12:21 -0700)]
Remove now obsolete mesa_bzero symbol.
George Sapountzis [Mon, 1 Mar 2010 15:28:55 +0000 (17:28 +0200)]
glapi: drop SOLARIS_THREADS
It seems that SOLARIS_THREADS is not used and does not work.
George Sapountzis [Mon, 1 Mar 2010 02:44:02 +0000 (04:44 +0200)]
glapi.c: misc coscmetic for FreeTSD
- move out of of the dispatch/context block to after corresponding init functions
- use more consistent naming with corresponding init functions
- XXX use _glthread_InitTSD() vs (void)_glthread_GetTSD() in _glapi_check_multithread() XXX
George Sapountzis [Sat, 27 Feb 2010 13:19:44 +0000 (15:19 +0200)]
glapi.c: misc cosmetic
mainly, move the multithread check mutex out of the dispatch/context block,
closer to where it is used.
George Sapountzis [Fri, 26 Feb 2010 21:43:46 +0000 (23:43 +0200)]
glapi.c: mv init_glapi_relocs to glapi_getproc.c
George Sapountzis [Fri, 26 Feb 2010 18:01:20 +0000 (20:01 +0200)]
glapi.c: mv check_table to glapi_getproc.c
George Sapountzis [Fri, 26 Feb 2010 22:24:58 +0000 (00:24 +0200)]
glapi.c: split check_table to not_null and get_proc
George Sapountzis [Fri, 26 Feb 2010 17:09:38 +0000 (19:09 +0200)]
glapi.h: drop dispatch_override prototypes
cannot find them in mesa tree
George Sapountzis [Sat, 27 Feb 2010 00:48:27 +0000 (02:48 +0200)]
glapi.h: misc cosmetic
- revert context/dispatch order, similar to glapi.c
- stray GL_GLEXT_PROTOTYPES
George Sapountzis [Fri, 26 Feb 2010 18:58:24 +0000 (20:58 +0200)]
glapi.h: consolidate GET_DISPATCH() and GET_CURRENT_CONTEXT() macros
Use likely() macro, as this is what most projects use.
Drops GL_CALL define, cannot find it in mesa tree.
Also, whitespace cleaunps in glthread.h
Michal Krol [Mon, 1 Mar 2010 16:57:18 +0000 (17:57 +0100)]
tgsi: Cleanup exec code for CLAMP, CMP and CND.
Michal Krol [Mon, 1 Mar 2010 16:07:03 +0000 (17:07 +0100)]
r300: Save viewport and clip states before invoking blitter.
Michal Krol [Mon, 1 Mar 2010 16:06:20 +0000 (17:06 +0100)]
util: Fix u_blitter setup code after bypass_vs_clip_and_viewport removal.
Not tested.
José Fonseca [Mon, 1 Mar 2010 14:56:05 +0000 (14:56 +0000)]
util: A few more todo marks.
José Fonseca [Mon, 1 Mar 2010 14:49:18 +0000 (14:49 +0000)]
util: More documentation for u_format.csv
José Fonseca [Mon, 1 Mar 2010 14:24:41 +0000 (14:24 +0000)]
util: Tweak comments regarding R8G8B8A8/X8_UNORM.
José Fonseca [Mon, 1 Mar 2010 14:22:30 +0000 (14:22 +0000)]
python: Don't mention deprecated drivers=trace option.
Now trace always built.
José Fonseca [Sat, 27 Feb 2010 23:50:09 +0000 (23:50 +0000)]
python: Don't mention deprecated drivers=trace option.
Now trace always built.
José Fonseca [Sat, 27 Feb 2010 23:49:58 +0000 (23:49 +0000)]
llvmpipe: Don't mention deprecated drivers=trace option.
Now trace always built.
José Fonseca [Sat, 27 Feb 2010 23:46:32 +0000 (23:46 +0000)]
progs/gallium: Make a phony target "unit" to run unit tests.
Francisco Jerez [Mon, 1 Mar 2010 00:56:38 +0000 (01:56 +0100)]
dri/nouveau: Combine the dummy texture more deterministically on nv0x.
Francisco Jerez [Mon, 1 Mar 2010 00:53:22 +0000 (01:53 +0100)]
dri/nouveau: Fix up the ADD texture environment on nv0x.
Andrew Randrianasulu [Mon, 1 Mar 2010 00:48:29 +0000 (01:48 +0100)]
dri/nouveau: Add two stencil operation cases in nv04_state_raster.c.
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
Andrew Randrianasulu [Mon, 1 Mar 2010 00:46:32 +0000 (01:46 +0100)]
dri/nouveau: Allow to render to XRGB8888 textures.
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
Marek Olšák [Sun, 28 Feb 2010 18:28:31 +0000 (19:28 +0100)]
r300g: atomize texture and sampler states
Marek Olšák [Sun, 28 Feb 2010 06:45:56 +0000 (07:45 +0100)]
r300g: decouple vertex stream setup (PSC) and VS output mapping (VAP_OUT)
Formerly known as vertex_format_state. These two are completely
unrelated when using HWTCL and decoupling them makes the design
less SWTCL-centric.
When bypass_vs_clip_and_viewport gets removed, the PSC setup will
no longer be a derived state.
This change shouldn't make unbreaking SWTCL harder.
Jakob Bornecrantz [Sun, 28 Feb 2010 16:28:44 +0000 (16:28 +0000)]
i965g: Fix texture creation
Alex Weiss [Sat, 27 Feb 2010 19:47:43 +0000 (14:47 -0500)]
Fixed mklib to properly merge static libraries on darwin.
Signed-off-by: Dan Nicholson <dbn.lists@gmail.com>
Vinson Lee [Sun, 28 Feb 2010 05:57:38 +0000 (21:57 -0800)]
dri: Remove dead code.
Vinson Lee [Sun, 28 Feb 2010 05:19:07 +0000 (21:19 -0800)]
glut: Add asserts to check for null pointer dereferences.
Vinson Lee [Sun, 28 Feb 2010 04:23:17 +0000 (20:23 -0800)]
glx: Add assert to check input to memcpy.
Vinson Lee [Sun, 28 Feb 2010 04:11:31 +0000 (20:11 -0800)]
mesa: Add asserts to check inputs to memcpy.
Vinson Lee [Sun, 28 Feb 2010 01:19:31 +0000 (17:19 -0800)]
mesa: Add assert to check input to memcpy is not null.
Vinson Lee [Sun, 28 Feb 2010 01:01:25 +0000 (17:01 -0800)]
glsl: Move assert from outside to inside of _slang_gen_swizzle function.
Vinson Lee [Sun, 28 Feb 2010 00:51:17 +0000 (16:51 -0800)]
dri: Add assert to check for null pointer dereference.
Vinson Lee [Sun, 28 Feb 2010 00:41:49 +0000 (16:41 -0800)]
i810: Add assert to check for null pointer dereference.
Vinson Lee [Sun, 28 Feb 2010 00:38:09 +0000 (16:38 -0800)]
mach64: Add asserts to check for null pointer dereferences.
Vinson Lee [Sun, 28 Feb 2010 00:30:49 +0000 (16:30 -0800)]
mga: Add assert to check for null pointer dereference.
Vinson Lee [Sun, 28 Feb 2010 00:28:30 +0000 (16:28 -0800)]
r128: Add assert to check for null pointer dereference.
Marek Olšák [Sat, 27 Feb 2010 02:20:58 +0000 (03:20 +0100)]
st/mesa: do not advertise S3TC if the external lib is not available
Marek Olšák [Sat, 27 Feb 2010 16:00:03 +0000 (17:00 +0100)]
r300g: put validating buffers after flushing
Also cleaning up the nasty validation process.
Marek Olšák [Sat, 27 Feb 2010 22:12:46 +0000 (23:12 +0100)]
r300g: use the atom size directly during emission
Marek Olšák [Sat, 27 Feb 2010 22:37:01 +0000 (23:37 +0100)]
r300g: atomize invalidation of texture caches
Marek Olšák [Sat, 27 Feb 2010 22:26:19 +0000 (23:26 +0100)]
r300g: atomize PVS flush
The first non-state atom. It's better and cleaner to have it.
Marek Olšák [Sat, 27 Feb 2010 20:54:16 +0000 (21:54 +0100)]
r300g: add size parameter to the atom emit functions
Maintaining a closer relationship between the atom size and what's passed
in BEGIN_CS.
Marcin Slusarz [Sat, 27 Feb 2010 21:55:35 +0000 (13:55 -0800)]
mesa: remove libmesagallium.a on make clean
Signed-off-by: José Fonseca <jfonseca@vmware.com>
Vinson Lee [Sat, 27 Feb 2010 23:36:11 +0000 (15:36 -0800)]
r300: Add asserts to check for null pointer dereferences.
Vinson Lee [Sat, 27 Feb 2010 23:29:44 +0000 (15:29 -0800)]
savage: Add assert to check for null pointer dereference.
Vinson Lee [Sat, 27 Feb 2010 23:13:38 +0000 (15:13 -0800)]
sis: Add assert to check function pointer is not null.
Vinson Lee [Sat, 27 Feb 2010 23:07:30 +0000 (15:07 -0800)]
tdfx: Add asserts to check for null pointer dereferences.
Vinson Lee [Sat, 27 Feb 2010 22:56:27 +0000 (14:56 -0800)]
unichrome: Assert that pointer is not null before dereferencing.
José Fonseca [Sat, 27 Feb 2010 20:28:53 +0000 (20:28 +0000)]
util: Fix X8/A8B8G8R8_SNORM's swizzle.
José Fonseca [Sat, 27 Feb 2010 20:28:16 +0000 (20:28 +0000)]
progs/gallium: Add unit tests for pixel format packing/unpacking.
Marek Olšák [Sat, 27 Feb 2010 19:05:00 +0000 (20:05 +0100)]
r300g: move the emission of GA_POINT_MINMAX into emit_fb_state
The only practical limits are the ones derived from the currently-set
framebuffer state.
Marek Olšák [Sat, 27 Feb 2010 19:03:34 +0000 (20:03 +0100)]
r300g: put the emission of R300_US_OUT_FMT_UNUSED back
It wasn't such a good idea to remove it. :/
Marek Olšák [Sat, 27 Feb 2010 19:00:38 +0000 (20:00 +0100)]
r300g: mark rasterizer_state as dirty only when it's not NULL
Marek Olšák [Sat, 27 Feb 2010 01:46:20 +0000 (02:46 +0100)]
radeong: clean up includes
Marek Olšák [Fri, 26 Feb 2010 23:37:01 +0000 (00:37 +0100)]
r300g: always emit the correct max vertex index to avoid DRM errors
Fixing bizarre reports that a vertex buffer is not large enough.
Marek Olšák [Fri, 26 Feb 2010 21:34:09 +0000 (22:34 +0100)]
r300g: remove pointless "while"
Vinson Lee [Sat, 27 Feb 2010 10:28:00 +0000 (02:28 -0800)]
radeon: Assert pointer is not null before dereferencing.
Add back an assert that was removed in commit
cd5f167353f16fb4f5b349002625b704f3e23778.
Vinson Lee [Sat, 27 Feb 2010 10:18:06 +0000 (02:18 -0800)]
dri: Assert pointer is not null before dereferencing.
Vinson Lee [Sat, 27 Feb 2010 10:01:56 +0000 (02:01 -0800)]
st/dri: Assert pointer is not null before dereferencing.
Vinson Lee [Sat, 27 Feb 2010 09:52:46 +0000 (01:52 -0800)]
r600: Assert pointer is not null before dereferencing.
Vinson Lee [Sat, 27 Feb 2010 08:31:14 +0000 (00:31 -0800)]
glsl: Assert input to strcmp is not null.
Vinson Lee [Sat, 27 Feb 2010 08:08:14 +0000 (00:08 -0800)]
glsl: Assert pointer is not null before dereferencing.
Vinson Lee [Sat, 27 Feb 2010 07:58:59 +0000 (23:58 -0800)]
mesa: Remove unnecessary header.
Jakob Bornecrantz [Thu, 25 Feb 2010 15:51:30 +0000 (16:51 +0100)]
st/xorg: Re-enable crtc on resize
Jakob Bornecrantz [Thu, 25 Feb 2010 15:51:15 +0000 (16:51 +0100)]
st/xorg: Fix copy-topy
Jakob Bornecrantz [Mon, 22 Feb 2010 19:06:23 +0000 (20:06 +0100)]
vmware/xorg: Bump to match vmware driver
Jakob Bornecrantz [Mon, 22 Feb 2010 13:47:15 +0000 (14:47 +0100)]
st/xorg: Make resize fail gracefully
Jakob Bornecrantz [Mon, 22 Feb 2010 13:46:10 +0000 (14:46 +0100)]
st/xorg: Obey max {width|height} from kernel and Gallium
Thomas Hellstrom [Mon, 15 Feb 2010 13:10:19 +0000 (14:10 +0100)]
st/xorg: Encapsulate all customizable stuff in a separate class.
This avoids exposing the ms driver structure to the winsys,
and nicely encapsulates driver customizable stuff.
In the future more things might be customizable by the winsys, like
throttling, 3D readback etc.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Jakob Bornecrantz [Thu, 18 Feb 2010 15:02:03 +0000 (16:02 +0100)]
vmware/xorg: Export modinfo just as the other vmware X drivers does
Jakob Bornecrantz [Thu, 18 Feb 2010 14:53:42 +0000 (15:53 +0100)]
st/xorg: A bit more debugging info
Jakob Bornecrantz [Thu, 18 Feb 2010 14:51:59 +0000 (15:51 +0100)]
st/xorg: Make default of debugging fallbacks smart
Set it to the same as 2d acceleration
Jakob Bornecrantz [Thu, 11 Feb 2010 13:41:40 +0000 (14:41 +0100)]
drm/vmware: Latest interface changes are backwards incopatible
Since the execbuffer change actually changed size off the ioctl
struct and not just a reuse of padded bits, we can't support
old kernels as easily as the scanout change was.
Thomas Hellstrom [Mon, 15 Feb 2010 13:24:51 +0000 (14:24 +0100)]
st/xorg, vmware/xorg: Fix xnfcalloc arguments.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Thomas Hellstrom [Tue, 9 Feb 2010 21:54:15 +0000 (22:54 +0100)]
vmware: Update vmwgfx_drm.h to kernel version.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Alex Deucher [Fri, 26 Feb 2010 20:24:12 +0000 (15:24 -0500)]
r600: enable OQ on rv740 on drms with working pipe config
Klaus Schnass [Fri, 26 Feb 2010 20:09:50 +0000 (15:09 -0500)]
r600: fail to validate unsupported texture formats
Christoph Bumiller [Fri, 26 Feb 2010 20:11:07 +0000 (21:11 +0100)]
nv40: fix include of nv04_surface_2d.h
Kristian Høgsberg [Fri, 26 Feb 2010 20:11:29 +0000 (15:11 -0500)]
eglinfo: Wrap extension string
Kristian Høgsberg [Fri, 26 Feb 2010 19:51:32 +0000 (14:51 -0500)]
xeglgears: Add an case that exercises EGL_KHR_gl_renderbuffer_image
Kristian Høgsberg [Fri, 26 Feb 2010 19:51:11 +0000 (14:51 -0500)]
egl_dri2: Implement EGL_KHR_gl_renderbuffer_image