If you want to access it, you should use the Drawable in xlib_drawable instead.

Otherwise we'll be unable to use our pbuffers.

This fixes a regression introduced with commit

"st-api: Rework how drawables are invalidated v3"

where the glx state tracker manager would invalidate a drawable each time it
checks the drawable dimensions, even during a validate call, which
resulted in an endless loop, since the state tracker would immediately
detect the new invalidation and rerun the validate...

This change marks the drawable invalid only if the drawable dimensions actually
changed during the validate, which will result in at most a single
unnecessary validate by the context running a validate during which the
dimensions changed.

To avoid unnecessary validates altogether, we need to implement yet another
st-api change: Returning the current time stamp from the validate function,
as suggested by Chia-I Wu. The glx state tracker manager could then return
the stamp resulting from the last drawable dimension check.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>

Fix compilation error due to commit
"Rework how drawables are invalidated v3"

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>

The drivers have been changed so that they behave as if all of the flags
were set. This is already implicit in most hardware drivers and required
for multiple contexts.

Some state trackers were also abusing the PIPE_FLUSH_RENDER_CACHE flag
to decide whether flush_frontbuffer should be called.
New flag ST_FLUSH_FRONT has been added to st_api.h as a replacement.

So that they don't have the driver-specific param and return type.

This is a follow-up to the ARB_sync patch for st/mesa and completes
the ARB_sync implementation.

xmesa_get_name returns the name of the st_api, which is the same as
MESA_VERSION_STRING.

The current context should be notified when the the front/back buffers
of the current drawable are swapped.  The notification was skipped when
xmesa_strict_invalidate is false (the default).

This fixes fdo bug #29774.

Add struct st_context_attribs to describe context profiles and
attributes.  Modify st_api::create_context to take the new struct
instead of an st_visual.

st_context_attribs can be used to support GLX_ARB_create_context_profile
and GLX_EXT_create_context_es2_profile in the future.  But the
motivation for doing it now is to be able to replace ST_API_OPENGL_ES1
and ST_API_OPENGL_ES2 by profiles.

Having 3 st_api's to provide OpenGL, OpenGL ES 1.1, and OpenGL ES 2.0 is
not a sane abstraction, since all of them share glapi for current
context/dispatch management.

Make st/glx include only main/core.h from core mesa.

Introduce a new configuration option XMESA_STRICT_INVALIDATE to switch
between swapbuffers-based and glViewport-based buffer invalidation.

Default strict invalidate to false, ie glViewport-based invalidation,
aka ST_MANAGER_BROKEN_INVALIDATE.

This means we will not call XGetGeometry after every swapbuffers,
which allows swapbuffers to remain asynchronous.  For apps running at
100fps with synchronous swapping, a 10% boost is typical.  For gears,
I see closer to 20% speedup.

Note that the work of copying data on swapbuffers doesn't disappear -
this change just allows the X server to execute the PutImage
asynchronously without us effectively blocked until its completion.

This applies even to llvmpipe's threaded rasterization as the
swapbuffers operation was a large part of the serial component of an
llvmpipe frame.

The downside of this is correctness - applications which don't call
glViewport on window resizes will get incorrect rendering, unless
XMESA_STRICT_INVALIDATE is set.

The ultimate solution would be to have per-frame but asynchronous
invalidation.  Xcb almost looks as if it could provide this, but the
API doesn't seem to quite be there.

With this change, xmesa_get_window_size still does one round trip, but
that's better than doing two.

Try to eliminate some unnecessary X server round trips.

commit 0189cb2fde
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sat Apr 10 12:48:43 2010 +0100

    gallium: don't use generic get_transfer func for textures

    It doesn't know and can't fill in the stride value.

commit 65bc6f88fd
Author: Chia-I Wu <olv@lunarg.com>
Date:   Sat Apr 10 13:49:34 2010 +0800

    i915g: Initialize screen surface function.

commit eb56e64986
Merge: f8b0a7f e7f1e5c
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sat Apr 10 00:38:43 2010 +0100

    Merge commit 'origin/master' into gallium-resources

    Conflicts:
    	src/gallium/drivers/r300/r300_texture.c

commit f8b0a7f6a3
Merge: a3c9980 f43c679
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sat Apr 10 00:35:09 2010 +0100

    Merge commit 'origin/master' into gallium-resources

    Conflicts:
    	src/gallium/drivers/r300/r300_texture.c

commit a3c99807de
Author: unknown <michal@.(none)>
Date:   Fri Apr 9 18:51:39 2010 +0200

    scons: Add missing sources.

commit 927cec79ce
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Fri Apr 9 18:07:56 2010 +0200

    gallium: fix another compile warning after merge. Hmpf.

commit 52953cd7b0
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Fri Apr 9 18:02:11 2010 +0200

    gallium: fix comment

commit 7c8763aa6c
Author: unknown <michal@.(none)>
Date:   Fri Apr 9 18:05:20 2010 +0200

    util: Fix type cast.

commit 9d0086411a
Author: unknown <michal@.(none)>
Date:   Fri Apr 9 18:04:33 2010 +0200

    libgl-gdi: Use proper unwrap functions for resources.

commit 251a5cdd18
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Fri Apr 9 17:51:23 2010 +0200

    gallium: more comments fixup

commit 8f3f9d5e1e
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Fri Apr 9 17:48:18 2010 +0200

    gallium: another fix after merge

commit 41f00a32ee
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Fri Apr 9 17:44:30 2010 +0200

    gallium: more pipe_texture/resource fixes after merge

commit faf53328d1
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Fri Apr 9 17:44:24 2010 +0200

    gallium: fix comments for changed USAGE flags

commit fdcb17bea4
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Fri Apr 9 16:40:07 2010 +0200

    gallium/pb: pb uses PB_USAGE_ flags, not PIPE_TRANSFER_ (same value anyway)

commit c95f7278ec
Author: Keith Whitwell <keithw@vmware.com>
Date:   Fri Apr 9 13:44:35 2010 +0100

    llvmpipe: fix merge glitches

commit 28f8b86831
Author: Keith Whitwell <keithw@vmware.com>
Date:   Fri Apr 9 13:41:39 2010 +0100

    r300g: update after merge for pipe_resources

commit 248c93cbc0
Author: Keith Whitwell <keithw@vmware.com>
Date:   Fri Apr 9 13:41:20 2010 +0100

    st/mesa: fix old pipe_texture usages

commit a563b1c5c2
Author: Keith Whitwell <keithw@vmware.com>
Date:   Fri Apr 9 13:40:56 2010 +0100

    r300g: remove unused variable

commit 734500131d
Author: Keith Whitwell <keithw@vmware.com>
Date:   Fri Apr 9 13:40:36 2010 +0100

    nv50: fix compiler warning

commit efd402e130
Merge: fec8a1d 5452615
Author: Keith Whitwell <keithw@vmware.com>
Date:   Fri Apr 9 13:33:57 2010 +0100

    Merge commit 'origin/master' into gallium-resources

    Conflicts:
    	src/gallium/drivers/cell/ppu/cell_screen.c
    	src/gallium/drivers/cell/ppu/cell_texture.c
    	src/gallium/drivers/llvmpipe/lp_screen.c
    	src/gallium/drivers/r300/r300_context.c
    	src/gallium/drivers/r300/r300_render.c
    	src/gallium/drivers/r300/r300_screen.c
    	src/gallium/drivers/r300/r300_state.c
    	src/gallium/drivers/r300/r300_texture.c
    	src/gallium/drivers/r300/r300_transfer.c
    	src/gallium/state_trackers/egl/common/egl_g3d.h
    	src/gallium/state_trackers/egl/kms/native_kms.c
    	src/gallium/state_trackers/egl/x11/native_dri2.c
    	src/gallium/state_trackers/egl/x11/native_ximage.c

commit fec8a1db13
Author: Marek Olšák <maraeo@gmail.com>
Date:   Sat Apr 3 07:58:34 2010 +0200

    util: fix assertion failures in pipe_buffer_flush_mapped_range

commit 1ff3984c2e
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Thu Apr 8 17:44:54 2010 +0200

    docs: fix transfer_map description

commit 20bf14be8a
Author: Keith Whitwell <keithw@vmware.com>
Date:   Thu Apr 8 14:39:13 2010 +0100

    util: fix up several uses of pipe_map_buffer_range

    This function used to return a pointer to where the start of the
    actual buffer would have been, even though only the requested range is
    being mapped.

    In the resources change, the function was modified to use a transfer
    internally, and started returning the pointer to the beginning of the
    transfer, ie the mapped range.

    Some users of the function were changed to reflect this new behaviour,
    some were not.  Since then the function has reverted to its original
    behaviour, matching master.

    This change restores some of the users of the map_buffer_range helper
    to expect the old/original behaviour.

commit 33179a8605
Author: Keith Whitwell <keithw@vmware.com>
Date:   Thu Apr 8 14:38:54 2010 +0100

    mesa/st: fix up several uses of pipe_map_buffer_range

    This function used to return a pointer to where the start of the
    actual buffer would have been, even though only the requested range is
    being mapped.

    In the resources change, the function was modified to use a transfer
    internally, and started returning the pointer to the beginning of the
    transfer, ie the mapped range.

    Some users of the function were changed to reflect this new behaviour,
    some were not.  Since then the function has reverted to its original
    behaviour, matching master.

    This change restores some of the users of the map_buffer_range helper
    to expect the old/original behaviour.

commit 3f5363d4dc
Author: Keith Whitwell <keithw@vmware.com>
Date:   Wed Apr 7 17:26:52 2010 +0100

    util: map_range and flush_range have offsets relative to start of buffer

commit 7eb1bfb97a
Author: Keith Whitwell <keithw@vmware.com>
Date:   Wed Apr 7 17:26:08 2010 +0100

    nv50: fix compiler warning

commit d040daff06
Author: Keith Whitwell <keithw@vmware.com>
Date:   Wed Apr 7 17:25:58 2010 +0100

    nvfx: fix compiler warning

commit 49ec01dffb
Author: Chia-I Wu <olv@lunarg.com>
Date:   Mon Apr 5 11:58:53 2010 +0800

    mesa/es: Fixes for gallium-resources.

commit 47c87ada45
Author: Marek Olšák <maraeo@gmail.com>
Date:   Sat Apr 3 05:19:20 2010 +0200

    r300g: fix transfers for textures created from winsys handles

commit 5f2701fdda
Author: Luca Barbieri <luca@luca-barbieri.com>
Date:   Sat Apr 3 03:52:38 2010 +0200

    nouveau: don't use the staging usage

    Maybe it could make sense, but for now dynamic is enough.

    None of these avoid uncached reads from GART on AGP cards.

commit 0db20fa49e
Author: Luca Barbieri <luca@luca-barbieri.com>
Date:   Sat Apr 3 03:27:19 2010 +0200

    i965: add brw_resource.c to Makefile

commit b94f3e7389
Author: Luca Barbieri <luca@luca-barbieri.com>
Date:   Sat Apr 3 01:48:33 2010 +0200

    nouveau: fix for gallium-resources

commit a01ff99a19
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Thu Apr 1 19:26:35 2010 +0200

    gallium: fix missing includes

commit 26aeded562
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Thu Apr 1 19:19:18 2010 +0200

    gallium: remove video interface and related stuff

    These interfaces weren't quite was needed, and building disabled for a while.
    Some code actually build since some branch merge, and were now not fully
    converted to gallium-resources.
    See http://www.mail-archive.com/mesa3d-dev@lists.sourceforge.net/msg09619.html
    for a discussion of this. Video related work is done in origin/pipe-video
    branch.

commit c64285aea4
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Thu Apr 1 18:45:54 2010 +0200

    python: fixes for resource changes

    doesn't look quite ok yet, but sort of compiles.

commit 03d4d5a41f
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Thu Apr 1 18:34:46 2010 +0200

    gallium: s/u_box_orgin_2d/u_box_origin_2d

commit 2444f02314
Author: Marek Olšák <maraeo@gmail.com>
Date:   Thu Apr 1 03:26:50 2010 +0200

    r300g: fix segfault when the transfers functions are used

    Still broken.

commit 6f09bf4066
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Thu Apr 1 00:05:12 2010 +0200

    r300g: compile fixes

commit 76711ff40d
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Thu Apr 1 00:04:47 2010 +0200

    nvfx: more compile fixes

commit c5d2e90c9c
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Wed Mar 31 23:18:50 2010 +0200

    gallium: more mostly merge fallout fixes...

commit fbc3722696
Merge: 86d9225 d97f696
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Wed Mar 31 22:09:35 2010 +0200

    Merge commit 'origin/master' into gallium-resources

    Conflicts:
    	src/gallium/drivers/cell/ppu/cell_screen.c
    	src/gallium/drivers/i915/i915_buffer.c
    	src/gallium/drivers/i915/i915_context.h
    	src/gallium/drivers/i915/i915_resource_texture.c
    	src/gallium/drivers/i915/i915_screen.c
    	src/gallium/drivers/i915/i915_state_emit.c
    	src/gallium/drivers/i965/brw_resource_texture.c
    	src/gallium/drivers/llvmpipe/lp_screen.c
    	src/gallium/drivers/llvmpipe/lp_setup.c
    	src/gallium/drivers/nvfx/nv30_fragtex.c
    	src/gallium/drivers/nvfx/nv40_fragtex.c
    	src/gallium/drivers/nvfx/nvfx_miptree.c
    	src/gallium/drivers/nvfx/nvfx_screen.c
    	src/gallium/drivers/nvfx/nvfx_transfer.c
    	src/gallium/drivers/r300/r300_state.c
    	src/gallium/drivers/svga/svga_screen_texture.c
    	src/gallium/state_trackers/dri/common/dri_drawable.c
    	src/gallium/state_trackers/dri/common/dri_screen.c
    	src/gallium/state_trackers/dri/common/dri_st_api.h
    	src/gallium/state_trackers/dri/drm/dri1.c
    	src/gallium/state_trackers/dri/drm/dri1.h
    	src/gallium/state_trackers/dri/drm/dri2.c
    	src/gallium/state_trackers/python/st_device.c
    	src/gallium/state_trackers/python/st_sample.c
    	src/mesa/state_tracker/st_cb_clear.c
    	src/mesa/state_tracker/st_cb_drawpixels.c
    	src/mesa/state_tracker/st_cb_readpixels.c
    	src/mesa/state_tracker/st_cb_texture.c
    	src/mesa/state_tracker/st_extensions.c

commit 86d9225d19
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Wed Mar 31 19:06:06 2010 +0200

    gallium: more fixes for bind changes

commit a215ef0606
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Wed Mar 31 18:48:36 2010 +0200

    gallium/docs: some updates for bind changes

commit c6c7e6746c
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Tue Mar 30 20:24:26 2010 +0200

    gallium: more bind change compile fixes

commit a83fa1504b
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Tue Mar 30 17:37:13 2010 +0200

    compile fixes

commit 30dc8afcd2
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Tue Mar 30 16:56:28 2010 +0200

    fix some merge issues

commit 30aa617fee
Merge: 1dde609 3a830bc
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Tue Mar 30 16:09:45 2010 +0200

    Merge commit 'origin/gallium-buffer-usage-cleanup' into gallium-resources

    Conflicts:
    	src/gallium/drivers/nouveau/nouveau_screen.c
    	src/gallium/drivers/nvfx/nvfx_transfer.c
    	src/gallium/winsys/drm/radeon/core/radeon_drm_buffer.c

commit 1dde609ad6
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Wed Mar 24 02:35:00 2010 +0100

    docs: some updates for pipe_resource

commit f236f9660d
Author: Luca Barbieri <luca@luca-barbieri.com>
Date:   Wed Mar 24 01:31:28 2010 +0100

    nvfx: fix for gallium-resources

    It seems to work with basic applications but almost surely needs more work.

    In particular, it probably shouldn't use PIPE_BUFFER_USAGE_* flags
    and should use PIPE_TRANSFER_* in several places.

    Also, we probably don't want the vtable indirect calls and that ought
    to be replaced with something better instead.

commit 5a136ad7b6
Author: Luca Barbieri <luca@luca-barbieri.com>
Date:   Wed Mar 24 01:31:19 2010 +0100

    nv50: fix build in gallium-resources

    Not actually tested.

    Also needs next patch tee to actually build, this is just the nv50 part
    split from the rest.

commit 3a830bc4a3
Author: Keith Whitwell <keithw@vmware.com>
Date:   Tue Mar 23 14:00:52 2010 -0700

    st/egl: fix up for binding flags

commit c6a80dc32e
Author: Keith Whitwell <keithw@vmware.com>
Date:   Tue Mar 23 13:52:15 2010 -0700

    r300: restore 4k alignment for oqbo buffers

commit e75a8d5ea9
Author: Keith Whitwell <keithw@vmware.com>
Date:   Tue Mar 23 13:00:07 2010 -0700

    gallium: bind flags

commit 1f5b509543
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Tue Mar 23 19:32:21 2010 +0100

    i965g: scons compile fixes

commit 2c385f8f90
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Tue Mar 23 19:20:33 2010 +0100

    nouveau: drm compile fix

commit b285086ebd
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Tue Mar 23 18:36:19 2010 +0100

    r300g: pipe_resource compile fixes

    bring back mistakenly deleted radeon_buffer.h
    plus some more

commit 7810606f42
Author: Michal Krol <michal@vmware.com>
Date:   Tue Mar 23 16:21:03 2010 +0100

    softpipe: Map GS constants, too.

commit 366f1176fb
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Tue Mar 23 15:51:52 2010 +0100

    failover: update for pipe_resources

commit 615f44d70d
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Tue Mar 23 15:51:02 2010 +0100

    identity: remove double is_resource_reference assignment

commit 7008586020
Author: Roland Scheidegger <sroland@vmware.com>
Date:   Tue Mar 23 15:50:32 2010 +0100

    trace: compile fix

commit 058c5697bd
Merge: dd7ba13 b33fd3c
Author: Keith Whitwell <keithw@vmware.com>
Date:   Tue Mar 23 06:40:39 2010 -0700

    Merge commit 'origin/gallium-resources' into gallium-buffer-usage-cleanup

    Conflicts:
    	src/gallium/state_trackers/vega/api_filters.c
    	src/mesa/state_tracker/st_cb_drawpixels.c

commit b33fd3ce3d
Author: Michal Krol <michal@vmware.com>
Date:   Mon Mar 22 21:03:26 2010 +0100

    gallium: Usage parameter of get_transfer/transfer_inline_write is a bitfield.

commit 9c1162d9d6
Author: Michal Krol <michal@vmware.com>
Date:   Mon Mar 22 20:50:49 2010 +0100

    scons: Update file lists after gallium-resources changes.

commit af9793ab9e
Author: Michal Krol <michal@vmware.com>
Date:   Mon Mar 22 20:04:39 2010 +0100

    gallium: Do not use `template` for formal parameter names.

commit dc2e12d714
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 22:41:34 2010 +0000

    ws/nouveau: remove pipe_texture reference

commit b94c72329f
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 22:40:41 2010 +0000

    ws/xlib: remove pipe_buffer reference in comment

commit 0a2af3eeae
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 22:39:34 2010 +0000

    st/vega: clean up reference to pipe_texture

commit 437ce98daa
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 22:39:02 2010 +0000

    st/python: begin conversion to pipe_resources, much more to do

commit 1b02e1ee3e
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 22:29:34 2010 +0000

    st/xorg: update for pipe_resources

commit eb39977fe7
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 22:23:51 2010 +0000

    st/dri: update for pipe_resources

commit e447aeff59
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 22:23:36 2010 +0000

    st/egl: update for pipe_resources

commit e4cc48da8f
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 22:08:44 2010 +0000

    r300: drop use of R300 DONT SYNC flag

commit 129a83ab4d
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 22:08:17 2010 +0000

    pipebuffer: use transfer flag

commit 575b35ee6b
Merge: f29ac73 9fc6c8b
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 22:03:25 2010 +0000

    Merge commit 'origin/master' into gallium-resources

    Conflicts:
    	src/gallium/drivers/llvmpipe/lp_texture.c
    	src/gallium/drivers/r300/r300_context.c
    	src/gallium/drivers/r300/r300_texture.c
    	src/gallium/winsys/drm/radeon/core/radeon_buffer.h

commit f29ac73f3f
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 18:37:25 2010 +0000

    cell: attempt to convert to pipe_resources

    Can't even compile test this driver.

commit 484b1947f4
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 17:25:50 2010 +0000

    nvfx: restore usage of pipe_winsys

    The interface that cannot be killed...

commit ac76ac6eb3
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 17:25:10 2010 +0000

    nv50: fix warning

commit 9683f44234
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 17:14:31 2010 +0000

    util: restore u_simple_screen, nouveau still relies on it

commit 961cbcb622
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 16:51:54 2010 +0000

    nouveau: convert nvfx and nv50 to pipe_resources

    Compile tested only.

    This was a deeper change than I was hoping for, due to the
    layering of the pipe_texture implementation in each driver on
    top of a shared pipe_buffer implementation in the shared code.

    Have modified the shared code to act as a set of convenience
    routines operating on nouveau_bo objects.

    Each driver now uses the u_resource_vtbl technique to split the
    implementation of pipe_resources between the existing miptree code
    for textures and a new, minimal buffer implementation in each
    driver.

    Eventually these should be combined, not least because APIs are now
    allowing things like binding buffer resources as textures and render
    targets.

commit 18ba74016d
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 10:02:54 2010 +0000

    r300: fix compilation after merge

    Also build r300 by default.

commit eb9c0175c8
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 09:59:49 2010 +0000

    st/vega: fix up after merge

commit ea8dd1d4ae
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 09:59:44 2010 +0000

    aux: remove unused piperesource helpers

commit be7af29d3a
Merge: d22c2c6 12deb9e
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 21 09:54:53 2010 +0000

    Merge commit 'origin/master' into gallium-resources

    Conflicts:
    	src/gallium/auxiliary/cso_cache/cso_context.c
    	src/gallium/auxiliary/cso_cache/cso_context.h
    	src/gallium/drivers/r300/r300_context.c
    	src/gallium/drivers/r300/r300_render.c
    	src/gallium/drivers/r300/r300_state.c
    	src/gallium/drivers/r300/r300_state_derived.c
    	src/gallium/state_trackers/vega/api_filters.c
    	src/gallium/state_trackers/vega/image.c
    	src/gallium/state_trackers/vega/image.h
    	src/gallium/state_trackers/vega/mask.c
    	src/gallium/state_trackers/vega/mask.h
    	src/gallium/state_trackers/vega/paint.c
    	src/gallium/state_trackers/vega/paint.h
    	src/gallium/state_trackers/vega/renderer.c
    	src/gallium/state_trackers/vega/renderer.h
    	src/gallium/state_trackers/vega/shader.c
    	src/gallium/state_trackers/vega/vg_context.h
    	src/gallium/state_trackers/vega/vg_tracker.c
    	src/mesa/state_tracker/st_manager.c

commit d22c2c6cb2
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sat Mar 20 11:48:54 2010 +0000

    drm/r300: update for r300g pipe_resources conversion

    Remove old files that related to pipe_buffers but weren't being
    built.  Hopefully this is correct.

commit f07b2c8369
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sat Mar 20 11:47:03 2010 +0000

    r300: convert to pipe_resources

    Do a very shallow conversion - basically keeping the existing
    buffer and texture code intact and using a vtbl struct
    inside our resource struct to select between the two implementations.

    The buffer and texture treatments could be further merged without
    much effort, but try to keep the existing code working at this point.

commit feca9c3ca6
Author: Keith Whitwell <keithw@vmware.com>
Date:   Thu Mar 18 06:00:34 2010 +0000

    gallium: update new merges to pipe_resource

commit 1cad983eac
Author: Keith Whitwell <keithw@vmware.com>
Date:   Thu Mar 18 06:00:19 2010 +0000

    drm/sw: update new merges to pipe_resource

commit 191d39490e
Author: Keith Whitwell <keithw@vmware.com>
Date:   Thu Mar 18 06:00:01 2010 +0000

    vg: update new merges to pipe_resource

commit b727c59bc4
Author: Keith Whitwell <keithw@vmware.com>
Date:   Thu Mar 18 05:59:38 2010 +0000

    llvmpipe: update new merges to pipe_resource

commit 5f4b64b37f
Author: Keith Whitwell <keithw@vmware.com>
Date:   Thu Mar 18 05:59:23 2010 +0000

    brw: pipe_resource fixes

commit d4aca209f5
Author: Keith Whitwell <keithw@vmware.com>
Date:   Thu Mar 18 05:59:06 2010 +0000

    util: update new merges to pipe_resource

commit cf6bef0afe
Merge: 1997231 6de8e56
Author: Keith Whitwell <keithw@vmware.com>
Date:   Thu Mar 18 05:38:50 2010 +0000

    Merge commit 'origin/master' into gallium-resources

commit 1997231916
Merge: ad88ac7 e1ee3ea
Author: Keith Whitwell <keithw@vmware.com>
Date:   Wed Mar 17 08:46:38 2010 +0000

    Merge commit 'origin/master' into gallium-resources

    Conflicts:
    	src/gallium/auxiliary/Makefile
    	src/gallium/auxiliary/util/u_blit.c
    	src/gallium/auxiliary/util/u_blit.h
    	src/gallium/auxiliary/util/u_gen_mipmap.c
    	src/gallium/auxiliary/util/u_gen_mipmap.h
    	src/mesa/state_tracker/st_cb_texture.c
    	src/mesa/state_tracker/st_gen_mipmap.c

commit ad88ac7903
Merge: 77bc770 8cdfd12
Author: Keith Whitwell <keithw@vmware.com>
Date:   Tue Mar 16 09:13:07 2010 +0000

    Merge branch 'gallium-sampler-view' into gallium-resources

    Conflicts:
    	src/gallium/auxiliary/cso_cache/cso_context.c
    	src/gallium/auxiliary/util/u_blit.c
    	src/gallium/drivers/llvmpipe/lp_texture.c
    	src/gallium/drivers/softpipe/sp_texture.c
    	src/mesa/state_tracker/st_cb_fbo.c
    	src/mesa/state_tracker/st_framebuffer.c
    	src/mesa/state_tracker/st_texture.c

commit 77bc770c99
Author: Keith Whitwell <keithw@vmware.com>
Date:   Mon Mar 15 22:21:48 2010 +0000

    util: missing file

commit f83c91db8a
Merge: c1d4774 42910eb
Author: Keith Whitwell <keithw@vmware.com>
Date:   Mon Mar 15 09:48:58 2010 +0000

    Merge commit 'origin/gallium-sampler-view' into gallium-resources

    Conflicts:
    	src/gallium/drivers/nv40/nv40_transfer.c
    	src/gallium/drivers/nvfx/nvfx_transfer.c
    	src/gallium/drivers/trace/tr_drm.c

commit dd7ba1378f
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 23:54:36 2010 +0000

    gallium: start a cleanup of buffer_usage

    Remove fairly meaningless CPU/GPU READ/WRITE flags and
    replace with proper usages.

commit c1d4774187
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 23:05:45 2010 +0000

    llvmpipe: don't FREE userbuffer data

commit 9bfa07afe1
Merge: 65757a1 08cddfe
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 22:54:51 2010 +0000

    Merge commit 'origin/master' into gallium-resources

    Conflicts:
    	src/gallium/drivers/llvmpipe/lp_rast.c
    	src/gallium/drivers/llvmpipe/lp_scene.c
    	src/gallium/drivers/llvmpipe/lp_texture.c
    	src/gallium/drivers/llvmpipe/lp_texture.h
    	src/gallium/drivers/softpipe/sp_texture.c
    	src/gallium/drivers/svga/svga_screen_texture.c

commit 65757a143f
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 22:41:17 2010 +0000

    svga: build fixes

commit 2f54352205
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 22:39:25 2010 +0000

    gallivm: build fix

commit 42642ec098
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 22:38:33 2010 +0000

    llvmpipe: convert to pipe_resources

commit 7bbcb21e20
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 22:19:30 2010 +0000

    gallivm: convert to pipe_resources

commit 88ae0d0461
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 21:01:22 2010 +0000

    configs: build svga

commit 0e112bc698
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 21:01:17 2010 +0000

    gallium: restore PIPE_BUFFER_USAGE_CUSTOM

commit 102aca688b
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 21:00:41 2010 +0000

    util: more transfer helpers

commit a79f6a4a08
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 20:59:36 2010 +0000

    svga: convert to use pipe_resrource

    As with others so far, a fairly shallow conversion.

commit 087fb54492
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 18:38:08 2010 +0000

    brw: fix function name

commit cfc9dd707d
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 18:19:06 2010 +0000

    gallium: enable brw compile

commit 8a5b86d76b
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 18:18:50 2010 +0000

    brw: compiles with pipe_resource

commit 563ca458b5
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 18:18:42 2010 +0000

    i915: hook up userbuffer create

commit b5095b4824
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 17:20:51 2010 +0000

    i915: use helpers for is_resource_referenced

commit d5392bdc6d
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 17:20:38 2010 +0000

    util: helpers for is_resource_referenced

commit 2f3492a5ae
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 17:08:50 2010 +0000

    i915: remove buffer.c again

commit 1373a35b65
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 17:08:34 2010 +0000

    i915: add new files to scons

commit 0251612d70
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 16:38:29 2010 +0000

    i915: compiling with pipe_resources

commit 9a02358642
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 13:51:16 2010 +0000

    gallium: remove inline_read transfer

commit a6ba315e25
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 13:50:32 2010 +0000

    trace: get running

    Some dumping will be incorrect or disabled, but it runs without
    crashing

commit 2133f1d90a
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 13:49:42 2010 +0000

    gallium: remove the inline_read transfer

    There aren't enough users of this to justify it.

commit bccaf1fa30
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 12:30:37 2010 +0000

    identity: hook up inline transfer operations

commit e4c152a344
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 12:21:54 2010 +0000

    gallium: build trace and identity

commit 0b5a311db7
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 12:21:36 2010 +0000

    gallium: add more of the transfer state to pipe_transfer

    Not really sure if recording all the arguments to the
    create_{transfer,texture,surface,etc} functions in the result of those
    calls is a great idea, but it seems we're fairly dependent on it
    throughout the code.

commit a23985c26e
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 12:19:46 2010 +0000

    identity: compiles with pipe_resources

commit d0d6309443
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 12:13:02 2010 +0000

    trace: compiles with pipe_resources

commit a4451ea459
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 11:39:50 2010 +0000

    softpipe: give userbuffers a format other than NONE

    Most mesa demos working

commit 32bb1bd4ba
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 11:39:21 2010 +0000

    util: correct argument order in pipe_buffer_map

commit 7e2696c064
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 11:32:55 2010 +0000

    softpipe: transfer flush

commit a0543b13c0
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 11:32:13 2010 +0000

    util: noop implementation of transfer_flush_region

commit ce418533be
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 11:26:07 2010 +0000

    gallium: softpipe runs gears with pipe_resources

commit bfda4f2eb3
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 11:25:48 2010 +0000

    util: get clip_tile working again

commit f5ef219c3b
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 09:43:20 2010 +0000

    softpipe: use u_transfer helpers

commit 072957aab2
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 09:42:46 2010 +0000

    util: helpers for inline transfers

commit 9c45561fb0
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 09:42:25 2010 +0000

    util: fix typo calculating transfer box

commit f3e98fd47f
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 09:25:46 2010 +0000

    st/vega: convert to pipe_resource

commit d1b7b00afc
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sun Mar 14 08:37:56 2010 +0000

    gallium: begin converting drivers to pipe_resource

    Work in progress...

commit 51c25117f5
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sat Mar 13 20:16:27 2010 +0000

    gallium: work in progress on layering resources on top of old textures

    Helper code in an aux module to avoid rewriting all the drivers.

commit fb6764d3ce
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sat Mar 13 19:19:09 2010 +0000

    heaps of wip

commit ee6b3bc730
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sat Mar 13 16:38:02 2010 +0000

    wip2

commit 1830880212
Merge: 90b4045 47bfbd4
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sat Mar 13 15:14:03 2010 +0000

    Merge branch 'gallium-sampler-view' into gallium-resources

    Conflicts:
    	src/gallium/include/pipe/p_context.h
    	src/mesa/state_tracker/st_atom_texture.c
    	src/mesa/state_tracker/st_cb_bitmap.c
    	src/mesa/state_tracker/st_cb_drawpixels.c
    	src/mesa/state_tracker/st_cb_texture.c
    	src/mesa/state_tracker/st_context.c
    	src/mesa/state_tracker/st_context.h
    	src/mesa/state_tracker/st_texture.h

commit 90b4045fbc
Author: Keith Whitwell <keithw@vmware.com>
Date:   Sat Mar 13 14:52:43 2010 +0000

    wip

There are X visuals that Gallium or the code does not support.  We could
not assert the color format to be supported.  Return PIPE_FORMAT_NONE in
such cases and let the caller handle it.

This is a quick & dirty solution, but it works.  See comments in
the code for other ideas.

Fixes regressions/breakage seen in progs/xdemos/glxheads, etc. from commit
6632915e95.

This is a back-port of commit ef2664da6c
from master.

This fixes a pretty big performance regression caused by commit
3475e88442.

When the user does not request a stencil buffer it's important that we
don't use a depth/stencil format (or at least make it our last choice).
If the user calls glClear(GL_DEPTH_BUFFER_BIT) when we have a combined
depth/stencil buffer, that causes us to hit the clear_with_quad() path
which can be much, much slower than calling pipe_context::clear().

Also, try to use a shallower depth format before a deeper one.

Create a per-display pipe_context as needed to copy the contents between
framebuffer attachments.  This allows us to support
GLX_MESA_copy_sub_buffer.

This basically adds a static xmesa_display to collect per-display static
variables in xm_api.c.  Multiple display support is still missing, but
this is a step forward.

Stupid typos again..

Instead of guessing the API in st/glx, let the target decide how to
create st_api.

Return a better format instead of an exact format in
choose_depth_stencil_format.  Also, prefer formats with stencil bits.

The state trackers need this value so that they can supply it
as the "void *" argument to flush_frontbuffer.

Fixes single-buffer rendering.

Several targets may want to reuse this code.  It's pretty simple, not
sure if this is really a win.

Finally we can inject layers for debug, trace and "other" with
relative ease between state-tracker and driver.

No need for the user of this winsys to supply/manage the x11 gc for
us.

Just use flush_frontbuffer directly.  The flush_frontbuffer routine has
been somewhat devalued recently, but it is actually just the right
interface for our needs.

It is in pipe_screen, meaning that any wrapping (eg trace module)
will get properly unwrapped before we try and use the pipe_surface
argument for real.

If a particular co-state-tracker needs to implement this itself, it
should organize a way to allow the winsys to call back up to its
level, rather than hijacking the driver-supplied implementation.

Introduce xlib_drawable struct, pass this down to winsys instead of
having it use the internal data structures from glx/x11

Remove the rgbMode and indexBits parameters from _mesa_create_visual
and _mesa_initialize_visual.  These values are now hardcoded to
GL_TRUE and 0.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>

Actually ask the gallium screen what Z/stencil format is supported.
This will let us remove some hacks in the llvmpipe driver.