Eric Anholt [Fri, 26 Feb 2010 00:25:17 +0000 (16:25 -0800)]
i965: Don't set a nonexistent enable bit in several SNB state pointers.
The modify bit is now usually in the instruction header. The
exception is CC state pointers.
Eric Anholt [Thu, 25 Feb 2010 23:23:13 +0000 (15:23 -0800)]
i965: Lump SNB in with Ironlake for bigger VUEs.
This gets the VS to the point of accepting vertices. \o/
Eric Anholt [Thu, 25 Feb 2010 23:22:20 +0000 (15:22 -0800)]
i965: Fix the SNB VE valid bit.
So, when the docs say that 0 means enabled now? That's a lie.
Eric Anholt [Thu, 25 Feb 2010 22:27:07 +0000 (14:27 -0800)]
intel: Fix up INTEL_NO_HW support.
This was accidentally (it seems) deleted in
5203b7227ccb6b618fa42f08434d4a3cf123dca2
Eric Anholt [Thu, 25 Feb 2010 22:24:36 +0000 (14:24 -0800)]
i965: Don't include SNB in has_negative_rhw_bug.
Eric Anholt [Thu, 25 Feb 2010 22:05:27 +0000 (14:05 -0800)]
i965: Fix the SNB clip near VP API bit.
Eric Anholt [Thu, 25 Feb 2010 21:46:45 +0000 (13:46 -0800)]
i965: Fix the SNB URB entry count setup.
George Sapountzis [Fri, 26 Feb 2010 02:02:41 +0000 (04:02 +0200)]
drop header files and demos for dropped window systems
George Sapountzis [Fri, 26 Feb 2010 02:02:41 +0000 (04:02 +0200)]
remove hack for miniglx from dri/utils.c
George Sapountzis [Fri, 26 Feb 2010 02:02:41 +0000 (04:02 +0200)]
drop linux-solo configs
George Sapountzis [Fri, 26 Feb 2010 02:02:41 +0000 (04:02 +0200)]
minor cleanup of dri/Makefile.template's after miniglx drop
also makes the WINDOW_SYSTEM configure var useless ...
George Sapountzis [Fri, 26 Feb 2010 02:02:40 +0000 (04:02 +0200)]
drop stray src/mesa/main/sources
Brian Paul [Fri, 26 Feb 2010 01:59:56 +0000 (18:59 -0700)]
mesa: use simplified _BaseFormat value in render-to-texture code
Fixes fd.o bug 26762.
Brian Paul [Thu, 25 Feb 2010 23:20:01 +0000 (16:20 -0700)]
util: remove a trailing semicolon in format table code
Török Edvin [Thu, 25 Feb 2010 16:02:43 +0000 (09:02 -0700)]
progs/glsl: add ARCH_FLAGS to LDLIBS
Allows building 32-bit demos on 64-bit Debian host.
Signed-off-by: Brian Paul <brianp@vmware.com>
Vinson Lee [Fri, 26 Feb 2010 01:52:48 +0000 (17:52 -0800)]
util: Remove import of unused sys module.
Kristian Høgsberg [Thu, 25 Feb 2010 21:20:03 +0000 (16:20 -0500)]
Remove unused mini and mesa glu subdirectories
Kristian Høgsberg [Thu, 25 Feb 2010 21:19:20 +0000 (16:19 -0500)]
Drop unused glut subdirectories
Kristian Høgsberg [Thu, 25 Feb 2010 21:17:40 +0000 (16:17 -0500)]
Remove windml progs
Kristian Høgsberg [Thu, 25 Feb 2010 21:17:04 +0000 (16:17 -0500)]
Remove remaining miniglx references
Kristian Høgsberg [Thu, 25 Feb 2010 21:12:58 +0000 (16:12 -0500)]
Drop fb, ffb and gamma dri drivers
Kristian Høgsberg [Thu, 25 Feb 2010 21:10:09 +0000 (16:10 -0500)]
Drop classic mesa svga driver
Kristian Høgsberg [Thu, 25 Feb 2010 21:08:59 +0000 (16:08 -0500)]
Drop glide driver
Kristian Høgsberg [Thu, 25 Feb 2010 21:00:45 +0000 (16:00 -0500)]
Remove GGI support
Kristian Høgsberg [Thu, 25 Feb 2010 20:55:00 +0000 (15:55 -0500)]
Remove dos driver
Kristian Høgsberg [Thu, 25 Feb 2010 20:52:32 +0000 (15:52 -0500)]
Remove d3d driver
Kristian Høgsberg [Thu, 25 Feb 2010 20:49:18 +0000 (15:49 -0500)]
Remove directfb support
Kristian Høgsberg [Thu, 25 Feb 2010 20:50:52 +0000 (15:50 -0500)]
Remove allegro driver
Vinson Lee [Thu, 25 Feb 2010 22:26:25 +0000 (14:26 -0800)]
Revert "glu/sgi: Set freed pointer to NULL."
This commit breaks flightgear fgfs. A null pointer is
dereferenced in the function MakeVertex.
This reverts commit
9be414f4b93c763e1aee8f634a13721ae5d95fcd.
Eric Anholt [Thu, 25 Feb 2010 18:56:09 +0000 (10:56 -0800)]
Merge branch 'sandybridge'
This brings in the initial code I've been working on for bringing up
OpenGL on Sandybridge in the i965 driver. It doesn't work yet.
Eric Anholt [Tue, 23 Feb 2010 19:18:25 +0000 (11:18 -0800)]
docs: addition of ARB_fcc to i965.
Eric Anholt [Wed, 24 Feb 2010 17:16:19 +0000 (09:16 -0800)]
i965: Add SNB math opcode support.
This is untested at this point.
Eric Anholt [Wed, 24 Feb 2010 01:42:20 +0000 (17:42 -0800)]
i965: Emit SNB FF unit state after the unit's push constants.
There's a BUN for the WM unit that says WM_STATE must immediately
follow PS_CONSTANTS, which this addresses. Presumably other units are
roughly the same, too.
Eric Anholt [Mon, 1 Feb 2010 17:55:18 +0000 (09:55 -0800)]
i965: Set up the SNB sampler state pointers.
Eric Anholt [Mon, 1 Feb 2010 17:41:26 +0000 (09:41 -0800)]
i965: Set GS rendering enabled even when the GS stage is disabled.
This is copying the behavior of other drivers.
Eric Anholt [Mon, 1 Feb 2010 17:04:12 +0000 (09:04 -0800)]
i965: Add a couple SNB state packets I saw in other batchbuffer dumps.
Eric Anholt [Mon, 25 Jan 2010 22:48:06 +0000 (14:48 -0800)]
i965: Re-disable the VS.
There's stuff that needs to happen in the ISA before we can play with
actually executing anything in the VS.
Eric Anholt [Mon, 21 Dec 2009 22:28:18 +0000 (14:28 -0800)]
i965: Fix up some Sandybridge define checks for the structure rebase.
Eric Anholt [Mon, 21 Dec 2009 21:08:41 +0000 (13:08 -0800)]
i965: Move PIPELINE_SELECT to the top of gen6 3d pipeline setup.
Eric Anholt [Mon, 21 Dec 2009 19:48:32 +0000 (11:48 -0800)]
i965: Untested Sandybridge WM packets.
Eric Anholt [Mon, 21 Dec 2009 19:22:14 +0000 (11:22 -0800)]
i965: Hook up remaining Sandybridge state packets besides WM.
Eric Anholt [Fri, 29 Jan 2010 19:03:04 +0000 (11:03 -0800)]
i965: Untested Sandybridge SF setup.
Eric Anholt [Fri, 18 Dec 2009 02:41:18 +0000 (18:41 -0800)]
i965: Fixup Sandybridge GS packet from doc updates.
Eric Anholt [Thu, 17 Dec 2009 23:52:17 +0000 (15:52 -0800)]
i965: Fix up Sandybridge GS reg definitions
Eric Anholt [Thu, 17 Dec 2009 23:51:43 +0000 (15:51 -0800)]
i965: fix typo in SNB VS defines comment
Eric Anholt [Thu, 17 Dec 2009 22:51:42 +0000 (14:51 -0800)]
i965: fix typo in SNB GS register defines
Eric Anholt [Thu, 17 Dec 2009 17:47:53 +0000 (09:47 -0800)]
i965: Enable clip stats.
Eric Anholt [Thu, 17 Dec 2009 17:38:50 +0000 (09:38 -0800)]
i965: Add Sandybridge viewport setup.
Eric Anholt [Thu, 17 Dec 2009 16:23:37 +0000 (08:23 -0800)]
i965: Enable DRAWING_RECTANGLE emit on Sandybridge.
Eric Anholt [Thu, 17 Dec 2009 16:20:32 +0000 (08:20 -0800)]
i965: Add Sandybridge scissor state.
Eric Anholt [Mon, 14 Dec 2009 22:17:00 +0000 (14:17 -0800)]
i965: Fix up the VF stats packet header.
Eric Anholt [Mon, 14 Dec 2009 22:11:14 +0000 (14:11 -0800)]
i965: Set the state base address on Sandybridge.
Eric Anholt [Fri, 4 Dec 2009 02:52:35 +0000 (18:52 -0800)]
i965: Free gen6 state BOs on context destroy.
Eric Anholt [Fri, 29 Jan 2010 18:42:17 +0000 (10:42 -0800)]
i965: Giant pile of flushing to track down SNB bringup issues.
This should go away before we push the code.
Eric Anholt [Fri, 4 Dec 2009 02:31:32 +0000 (18:31 -0800)]
i965: Fix some defines of gen6 regs from docs comparison.
Eric Anholt [Fri, 4 Dec 2009 01:25:42 +0000 (17:25 -0800)]
i965: Add basic decode of new gen6 packet boundaries.
Eric Anholt [Fri, 4 Dec 2009 00:53:07 +0000 (16:53 -0800)]
i965: Reconnect the index/vertex setup.
Eric Anholt [Fri, 4 Dec 2009 01:08:32 +0000 (17:08 -0800)]
i965: Set up the SNB URB.
even with vs disabled, still doesn't work.
Eric Anholt [Thu, 3 Dec 2009 23:44:29 +0000 (15:44 -0800)]
i965: Try sending prims down the pipeline.
Now things catch on fire.
Eric Anholt [Thu, 3 Dec 2009 23:20:06 +0000 (15:20 -0800)]
i965: Try uploading SNB VS constants.
Eric Anholt [Thu, 3 Dec 2009 23:06:08 +0000 (15:06 -0800)]
i965: Try turning on the VS.
Eric Anholt [Thu, 3 Dec 2009 23:04:37 +0000 (15:04 -0800)]
i965: Get vp-tri batchbuffers running (no rendering).
Eric Anholt [Fri, 29 Jan 2010 19:12:18 +0000 (11:12 -0800)]
i965: Add untested REJECT_ALL clip state.
Eric Anholt [Fri, 27 Nov 2009 06:03:43 +0000 (22:03 -0800)]
i965: Add untested passthrough GS setup.
Eric Anholt [Mon, 16 Nov 2009 22:50:23 +0000 (14:50 -0800)]
i965: Add untested Sandybridge passthrough VS setup.
Eric Anholt [Fri, 29 Jan 2010 19:18:26 +0000 (11:18 -0800)]
i965: Start adding support for the Sandybridge CC unit.
Eric Anholt [Tue, 3 Nov 2009 22:02:24 +0000 (14:02 -0800)]
i965: Set up sandybridge binding table pointers but don't enable it yet.
It hangs the GPU at the clipper stage, presumably because we're lacking
other setup.
Eric Anholt [Tue, 3 Nov 2009 21:41:22 +0000 (13:41 -0800)]
i965: Update WM surface state setup for sandybridge's new BLEND_STATE.
Eric Anholt [Tue, 3 Nov 2009 21:19:30 +0000 (13:19 -0800)]
i965: Set up sandybridge depthbuffer.
Eric Anholt [Tue, 3 Nov 2009 19:28:47 +0000 (11:28 -0800)]
intel: Start adding defines and some bits for sandybridge bringup.
Francisco Jerez [Thu, 25 Feb 2010 00:55:13 +0000 (01:55 +0100)]
dri/nouveau: Use the XRGB8888 hardware texture format.
Francisco Jerez [Tue, 23 Feb 2010 15:39:26 +0000 (16:39 +0100)]
dri/nouveau: Use the hardware I8 format for intensity textures.
Francisco Jerez [Sun, 21 Feb 2010 13:23:40 +0000 (14:23 +0100)]
dri/nouveau: Implement EXT_texture_from_pixmap.
Francisco Jerez [Thu, 25 Feb 2010 01:15:54 +0000 (02:15 +0100)]
dri/nouveau: Support rectangle textures.
Francisco Jerez [Tue, 23 Feb 2010 15:31:42 +0000 (16:31 +0100)]
dri/nouveau: Don't try to map a non-existent teximage.
Francisco Jerez [Tue, 23 Feb 2010 15:30:48 +0000 (16:30 +0100)]
dri/nouveau: Add some RGB888 span functions.
Francisco Jerez [Tue, 23 Feb 2010 15:57:06 +0000 (16:57 +0100)]
dri/nouveau: Fake A8 and L8 texture support on nv04.
Francisco Jerez [Tue, 23 Feb 2010 15:15:14 +0000 (16:15 +0100)]
dri/nouveau: Restructure the nv[12]0 regcombiner code, and fake A8/L8 support.
Francisco Jerez [Mon, 22 Feb 2010 01:03:42 +0000 (02:03 +0100)]
dri/nouveau: Try to validate textures earlier.
Francisco Jerez [Mon, 22 Feb 2010 00:53:23 +0000 (01:53 +0100)]
dri/nouveau: Swizzle using the CPU when we hit a limitation of SIFM.
Francisco Jerez [Mon, 22 Feb 2010 00:50:04 +0000 (01:50 +0100)]
dri/nouveau: Clean up the nv04 surface code a bit.
Francisco Jerez [Sun, 21 Feb 2010 13:35:56 +0000 (14:35 +0100)]
dri/nouveau: Avoid mask overflow on nv04_surface_fill.
Francisco Jerez [Sun, 21 Feb 2010 13:30:31 +0000 (14:30 +0100)]
dri/nouveau: Set _BaseFormat correctly for z24s8 renderbuffers.
Francisco Jerez [Sun, 21 Feb 2010 13:28:17 +0000 (14:28 +0100)]
dri/nouveau: Fix stencil mask handling on glClear().
José Fonseca [Thu, 25 Feb 2010 17:02:52 +0000 (17:02 +0000)]
scons: Always build trace driver
There are so many state trackers and winsys that rely on it for debug
building, and trace can really build anywhere and is thin so there's no
point is making it an option.
Based on Xavier Chantry <chantry.xavier@gmail.com>'s patch:
"I also removed the trace check in xlib SConscript which is now useless, and
that script seems to deal with the cell driver, so I added cell in the check
for supported drivers."
Michal Krol [Thu, 25 Feb 2010 13:21:50 +0000 (14:21 +0100)]
python: Fix typo.
George Sapountzis [Thu, 25 Feb 2010 15:52:05 +0000 (17:52 +0200)]
glapi/gen: just copy in glx/glapi
George Sapountzis [Thu, 25 Feb 2010 15:52:05 +0000 (17:52 +0200)]
glapi: EXTRA_DEBUG is never defined, make this explicit
George Sapountzis [Thu, 25 Feb 2010 15:52:04 +0000 (17:52 +0200)]
glapi: minor ugliness to be able to copy to xorg
Brian Paul [Thu, 25 Feb 2010 15:36:31 +0000 (08:36 -0700)]
docs: remove demos.html page which contained one dead link
Michal Krol [Thu, 25 Feb 2010 14:01:10 +0000 (15:01 +0100)]
tgsi: Remove dead code.
Chia-I Wu [Wed, 24 Feb 2010 04:01:14 +0000 (12:01 +0800)]
mesa: Move src/mesa/glapi/dispatch.h to mesa.
glapi/dispatch.h is a core Mesa header file. Move the header file to
main/ to make this clear. It also becomes clear after this change that
IN_DRI_DRIVER is only used in core Mesa to enable the remap table.
Chia-I Wu [Wed, 24 Feb 2010 03:20:14 +0000 (11:20 +0800)]
glapi: Move src/mesa/main/dispatch.c to glapi and rename.
main/dispatch.c is a glapi source file. It is part of GLAPI_SOURCES in
sources.mak and part of glapi_sources in SConscript. This commit moves
it to glapi/ and renames it to glapi_dispatch.c.
Chia-I Wu [Wed, 24 Feb 2010 04:04:58 +0000 (12:04 +0800)]
glapi: Generated GLX sources should use glapidispatch.h.
Fix glX_proto_recv.py and glX_proto_send.py, and regenerate the sources.
Chia-I Wu [Wed, 24 Feb 2010 03:17:16 +0000 (11:17 +0800)]
mesa: Remove unused Makefile.{ugl,win}.
Chia-I Wu [Thu, 25 Feb 2010 13:18:20 +0000 (21:18 +0800)]
st/egl: Reduce validation round-trips in DRI2 backend.
This is similar to the changes to reduce the round-trips in ximage
backend.
When the server supports DRI_InvalidateBuffers, the event is used to
update the server stamp (this path is not tested). Otherwise, the
server stamp is updated when flush_frontbuffer or swap_buffers is
called. This makes sure at least the client APIs get the new buffers
when a new frame is started while skipping all round-trips during the
drawing.
Chia-I Wu [Thu, 25 Feb 2010 13:16:56 +0000 (21:16 +0800)]
st/egl: Reduce validation round-trips in ximage backend.
ximage_surface_validate is called several times per frame. This commit
adds the client and server stamps to reduce the round-trips to the
server. The idea is to bump the server stamp when flush_frontbuffer or
swap_buffers is called, and to skip the round-trip when the client stamp
is equal to the server stamp. This makes sure the client APIs get the
new buffers when a new frame is started while skipping all round-trips
during the drawing. To make this work, egl_g3d_validate_context is no
longer called after swap_buffers.
Chia-I Wu [Wed, 24 Feb 2010 07:34:16 +0000 (15:34 +0800)]
st/egl: Refactor dri2_surface_get_buffers.
Add dri2_surface_process_drawable_buffers to process the buffers
returned by the server.
Chia-I Wu [Wed, 17 Feb 2010 15:11:31 +0000 (23:11 +0800)]
st/egl: Add dri2InvalidateBuffers.
src/glx/dri2.c will call dri2InvalidateBuffers when
DRI2_InvalidateBuffers event is received. This fixes a missing symbol
error, and paves the way for event-based validation.
Vinson Lee [Thu, 25 Feb 2010 10:42:41 +0000 (02:42 -0800)]
util: Assert that pointer is not null before dereferencing.