radeonsi: add a workaround for inexact SNORM8 blitting again

GFX9 is affected.

We only have tests for GL_x_SNORM where x is R8, RG8, RGB8, and RGBA8.

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>

radeonsi/gfx9: fix TC-compatible stencil compression

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>

radeonsi/gfx9: fix TXF_LZ with 1D textures

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>

radeonsi/gfx9: disable sparse buffers

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>

ac/sid.h: don't use parentheses in PKT3_RELEASE_MEM definition

The parses skips the line if it contains parentheses.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>

ac: parse EVENT_WRITE_EOP, RELEASE_MEM, WAIT_REG_MEM, NOWHERE

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>

st/mesa: simplify returning GL_VENDOR

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>

st/mesa: remove the "Gallium 0.4 on" prefix from GL_RENDERER

If you want to keep it for your driver, please raise your hand.
The prefix will probably have to be added into the driver instead of here.

I cringe when I look at my long renderer string:
  Gallium 0.4 on AMD Radeon R9 Fury Series (DRM 3.17.0 / 4.11.0-staging-01277-gab25a9e, LLVM 5.0.0)

I'm sincerely sorry for all apps that detect Mesa by expecting "Gallium"
in the string.

Reviewed-by: Eric Anholt <eric@anholt.net>

st/mesa: don't update MSAA states for GL_FRAMEBUFFER_SRGB

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>

i965: Ignore anisotropic filtering in nearest mode.

This fixes both Europa Universalis IV and Stellaris rendering on i965.
This was tested on SKL.

This fix was discovered by Jakub Szuppe at Stream HPC
(https://streamhpc.com/).

bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96958
bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=95530
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Tested-by: Dylan Baker <dylan@pnwbakers.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Cc: 17.1 <mesa-stable@lists.freedesktop.org>

glsl: gl_Max{Vertex,Fragment}UniformComponents exist in all desktop GL versions

The current implementation assumed that these were replaced in GLSL >= 4.10
by gl_Max{Vertex,Fragment}UniformVectors, however this is not true: both
built-ins should be produced from GLSL 4.10 onwards.

This was raised by new CTS tests that are in development.

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>

docs: update calendar, add news item and link release notes for 17.1.3

Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

docs: add sha256 checksums for 17.1.3

Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

docs: add release notes for 17.1.3

Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

st/glsl_to_tgsi: use correct writemask when converting generic intrinsics

This fixes a bug when lowering ballotARB: previously, using writemask 0xf,
emit_asm would create TGSI_OPCODE_BALLOT instructions that span two registers
to cover 4 64-bit channels. This could trample over other a neighbouring
temporary.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101360
Cc: 17.1 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>

gallium/radeon/gfx9: fix PBO texture uploads to compressed textures

st/mesa creates a surface that reinterprets the compressed blocks as
RGBA16UI or RGBA32UI. We have to adjust width0 & height0 accordingly to
avoid out-of-bounds memory accesses by CB.

Cc: 17.1 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>

r600: fix off-by-one in egd_tables.py

Port of the corresponding fix in sid_tables.py.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>

amd/common: fix off-by-one in sid_tables.py

The very last entry in the sid_strings_offsets table ended up missing,
leading to out-of-bounds reads and potential crashes.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>

i965: update MaxTextureRectSize to match PRMs and comply with OpenGL 4.1+

We were exposing 4096, but we can do up to 8192 in Gen4-6 and up to
16384 in gen7+. OpenGL 4.1+ requires at least 16384.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

mesa: add KHR_no_error support for gl*UniformHandleui64*ARB

Similar to _mesa_uniform() except that we have to call
validate_uniform_parameters() instead of validate_uniform().

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

mesa: add KHR_no_error support for glGetImageHandleARB()

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

mesa: add KHR_no_error support for glGetTexture*HandleARB()

It would be nice to have a no_error path for
_mesa_test_texobj_completeness() because this function doesn't
only test if the texture is complete.

Anyway, that seems enough for now and a bunch of checks are
skipped with this patch.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

mesa: add KHR_no_error support for glMake{Image,Texture}Handle*ResidentARB()

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

mesa: add KHR_no_error support for glIs{Image,Texture}HandleResidentARB()

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

radeonsi: reduce overhead for resident textures which need color decompression

This is done by introducing a separate list.

si_decompress_textures() is now 5x faster.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>

radeonsi: reduce overhead for resident textures which need depth decompression

This is done by introducing a separate list.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>

radeonsi: use util_dynarray_foreach for bindless resources

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>

mesa/util: add util_dynarray_clear() helper

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>

gallium/radeon: add a new HUD query for the number of resident handles

Useful for debugging performance issues when ARB_bindless_texture
is enabled. This query doesn't make a distinction between texture
and image handles.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>

i965/gen4: Refactor depth/stencil rebase

Effectively there is the same code twice, once for depth and
again for stencil.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Topi Pohjolainen <topi.pohjolainen@intel.com>

i965: Drop depth/stencil miptree pointers in alignment workaround

In brw_workaround_depthstencil_alignment() corresponding
renderbuffers are always set to refer to the same temp miptrees.
There is no need to carry them in context.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Topi Pohjolainen <topi.pohjolainen@intel.com>

i965/gen4: Simplify depth/stencil invalidate check

There is no separate stencil on gen < 6.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Topi Pohjolainen <topi.pohjolainen@intel.com>

i965/gen4: Remove redundant check for depth when rebasing stencil

In case of gen < 6 stencil (if present) is always combined with
depth. Both stencil and depth attachments point to the same
physical surface.
Alignment workaround starts by considering depth and updates
stencil accordingly. Current logic continues with stencil and
in vain considers the case where depth would refer to different
surface than stencil.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Topi Pohjolainen <topi.pohjolainen@intel.com>

i965/gen4: Remove non-existing stencil and hiz buffer setup

Separate stencil and hiz are only enabled for gen6+.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Topi Pohjolainen <topi.pohjolainen@intel.com>

android: ac: add missing libdrm_amdgpu shared dependency

Fixes building errors in amd/common:

target  C: libmesa_amd_common <= external/mesa/src/amd/common/ac_gpu_info.c
...
target  C: libmesa_amd_common <= external/mesa/src/amd/common/ac_surface.c
...

external/mesa/src/amd/common/ac_gpu_info.h:31:10: fatal error: 'amdgpu.h' file not found
         ^
2 errors

Fixes: 98a2492 ("ac_surface: use radeon_info from ac_gpu_info")
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>

r600: include libelf headers only as needed

Headers are required only when building with OpenCL. As we're building
w/o it libelf may be missing, hence we'll error out as below:

src/gallium/drivers/r600/evergreen_compute.c:27:10:
fatal error: 'gelf.h' file not found
         ^
1 error generated.

Fixes: d96a210842 ("r600g,compute: provide local copy of functions from
ac_binary.c")
Reviewed-by: Jan Vesely <jan.vesely@rutgers.edu>
Reported-by: Mauro Rossi <issor.oruam@gmail.com>
Tested-by: Mauro Rossi <issor.oruam@gmail.com>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

radeonsi: include ac_binary.h for struct ac_shader_binary

The header embeds the struct so it needs the header inclusion instead of
the dummy forward declaration.

Cc: Nicolai Hähnle <nicolai.haehnle@amd.com>
Cc: Marek Olšák <marek.olsak@amd.com>
Cc: Tom Stellard <tstellar@redhat.com>
Fixes: 32206c5e560 ("radeonsi: Add radeon_shader_binary member to struct
si_shader")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>

r600, radeon: move radeon_shader_binary_{init,clean} back to radeon

Those are used by r600 and radeonsi, so moving them within the former
was a bad idea.

Fixes: d96a210842b ("r600g,compute: provide local copy of functions
from ac_binary.c")
Cc: Jan Vesely <jan.vesely@rutgers.edu>
Cc: Aaron Watry <awatry@gmail.com>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>

ac: resolve conflicts introduced with "ac: remove amdgpu.h dependency"

The commit did not add the relevant includes - in particular
stdint.h and stdbool.h for the respective standard types.

At the same time, the amdgpu_device_handle typedef redeclaration was
off.

Fixes: 81945ded0dc ("ac: remove amdgpu.h dependency")
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101471
Cc: Mark Janes <mark.a.janes@intel.com>
Cc: Gregor Münch <gr.muench@gmail.com>
Reported-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reported-by: Mark Janes <mark.a.janes@intel.com>
Reported-by: Gregor Münch <gr.muench@gmail.com>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>

i965/gen4: Set depth offset when there is stencil attachment only

Current version fails to set depthstencil.depth_offset when there
is only stencil attachment (it does set the intra tile offsets
though). Fixes piglits:

g45,g965,ilk:   depthstencil-render-miplevels 1024 s=z24_s8
g45,ilk:        depthstencil-render-miplevels 273 s=z24_s8

CC: mesa-stable@lists.freedesktop.org
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Topi Pohjolainen <topi.pohjolainen@intel.com>

i965/gen6: Remove dead code in hiz surface setup

In intel_hiz_miptree_buf_create() the miptree is unconditionally
created with MIPTREE_LAYOUT_FORCE_ALL_SLICE_AT_LOD.

Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Topi Pohjolainen <topi.pohjolainen@intel.com>

intel/isl/gen6: Allow arrayed stencil

Nothing prevents arrayed stencil surfaces even though hardware
doesn't support mipmapping.

Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Signed-off-by: Topi Pohjolainen <topi.pohjolainen@intel.com>

svga: add new num-failed-allocations HUD query

This counter is incremented if we fail to allocate memory for
vertex/index/const buffers, textures, etc.

Reviewed-by: Neha Bhende <bhenden@vmware.com>

gallium/hud: support GALLIUM_HUD_DUMP_DIR feature on Windows

Use a dummy implementation of the access() function.  Use \ path separator.
Add a few comments.

Reviewed-by: Neha Bhende <bhenden@vmware.com>

svga: add a few minor comments

Trivial.

mesa: whitespace fixes in enable.c

Remove trailing whitespace, replace tabs w/ spaces, etc.  Trivial.

i965: Convert SF_STATE to genxml.

This patch finishes the work done by Ken of converting SF_STATE to genxml, and
merges it with gen6+ code for emitting that state.

Signed-off-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

genxml: The viewport state offset is actually an address.

This fixes code generation on gen45.

Signed-off-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

genxml: Rename fields to match gen6+.

"Anti-aliasing Enable" to "Anti-Aliasing Enable".

Signed-off-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

genxml: Rename SF_STATE field to match gen6+.

Rename "Use Point Width State" to "Point Width Source". It accepts the same
values and has the same meaning as gen6+, so lets keep them with the same name
to simplify the code.

Signed-off-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

i965: aa_line_distance_mode should be before the padding.

It seems that it was never set correctly.

Signed-off-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

swr/rast: Fix read-back of viewport array index

Binner/clipper read viewport array index from the vertex header as needed.
Move viewport state to BACKEND_STATE.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Refactor includes to limit simdintrin.h usage

Reduces the files rebuilt after modifying simdintrin.h from
84 to 64.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Fix read-back of render target array index

The last FE stage can emit render target array index. Currently we only
check to see if GS is emitting it. Moved the state to BACKEND_STATE and
plumbed the driver to set it.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Adjust cast for gcc warning

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Don't transition hottile resolved->dirty during store tiles

Fixes crash when dumping render targets and RT surface has been deleted.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: gen_llvm_types.py support for SIMD256/SIMD512

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Properly size GS stage scratch space

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Fix early z / query interaction

For certain cases, we perform early z for optimization. The GL_SAMPLES_PASSED
query was providing erroneous results because we were counting the number
of samples passed before the fragment shader, which did not work if the
fragment shader contained a discard.

Account properly for discard and early z, by anding the zpass mask with
the post fragment shader active mask, after the fragment shader.

Fixes the following piglit tests:
    - occlusion-query-discard
    - occlusion_query_meta_fragments

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Share vertex memory between VS input/output

Removes large simdvertex stack allocation.

Vertex shader must ensure reads happen before writes.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Add support for dynamic vertex size for VS output

Add support for dynamic vertex size for the vertex shader output.

Add new state in SWR_FRONTEND_STATE to specify the size.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: SIMD16 FE - improve calcDeterminantIntVertical

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Add support to PA for variable sized vertices

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Rework attribute layout

Move fixed attributes to the top and pack single component SGVs.
WIP to support dynamically allocated vertex size.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Remove explicit primitive id slot in the vertex layout

- Remove any special casing in the PS stage when primitive ID is input.
  Treat as a normal attribute that must be set up properly in the FE linkage.
- Remove primitive id from the PS_CONTEXT and TRI_FLAGS

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Fix invalid 16-bit format traits for A1R5G5B5

Correctly handle formats of <= 16 bits where the component bits don't
add up to the pixel size.

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

swr/rast: Implement JIT shader caching to disk

Disabled by default; currently doesn't cache shaders (fs,gs,vs).

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>

gallium/docs: improve docs for SAMPLE_POS, SAMPLE_INFO, TXQS, MSAA semantics

For the SAMPLE_POS and SAMPLE_INFO opcodes, clarify resource vs. render
target queries, range of postion values, swizzling, etc.  We basically
follow the DX10.1 conventions.

For the TXQS opcode and TGSI_SEMANTIC_SAMPLEID, clarify return value
and type.

For the TGSI_SEMANTIC_SAMPLEPOS system value, clarify the range of
positions returned.

v2: use 'undef' for unused vector components.  Use (0.5, 0.5, undef, undef)
for sample pos when MSAA not applicable.

v3: Add note that OPCODE_SAMPLE_INFO, OPCODE_SAMPLE_POS are not used yet
and the information is subject to change.

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>

svga: add some missing SVGA_STATS_* enum values, prefix strings

To fix the build when VMX86_STATS is defined.
Also, some minor whitespace changes to match upstream code.

Reviewed-by: Charmaine Lee <charmainel@vmware.com>

radeonsi: add new polaris12 pci id

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: 17.0 17.1 <mesa-stable@lists.freedesktop.org>

swr: Don't crash when encountering a VBO with stride = 0.

The swr driver uses vertex_buffer->stride to determine the number
of elements in a VBO. A recent change to the state-tracker made it
possible for VBO's with stride=0. This resulted in a divide by zero
crash in the driver. The solution is to use the pre-calculated vertex
element stream_pitch in this case.

This patch fixes the crash in a number of piglit and VTK tests introduced
by 17f776c27be266f2.

There are several VTK tests that still crash and need proper handling of
vertex_buffer_index.  This will come in a follow-on patch.

v2: Correctly update all parameters for VBO constants (stride = 0).
    Also fixes the remaining crashes/regressions that v1 did
    not address, without touching vertex_buffer_index.

Reviewed-by: Tim Rowley <timothy.o.rowley@intel.com>

intel/isl: Add the maximum surface size limit

V2: Use 2^31 bytes (2GB) surface size limit on pre-gen9 and
    2^38 bytes for gen9+.

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>

intel/isl: Use uint64_t to store total surface size

Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>

i965: Mark freshly allocate bo as idle

When created, buffers are idle, so mark them as such to save an early
ioctl or mistakenly assuming the fresh buffer is busy.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Kenneth Graunke <kenneth@whitecape.org>
Cc: Matt Turner <mattst88@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

etnaviv: add rs-operations sw query

It could be useful to get the number of emited resolve operations when
doing driver optimizations.

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>

etnaviv: advertise correct max LOD bias

The maximum LOD bias supported is the same as the max texture level
supported.

Fixes piglit: ext_texture_lod_bias

Fixes: c9e8b49b ("etnaviv: gallium driver for Vivante GPUs")
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>

etnaviv: mask correct channel for RB swapped rendertargets

Now that we support RB swapped targets by using a shader variant, we
must derive the color mask from both the blend state and the bound
framebuffer.

Fixes piglit: fbo-colormask-formats

Fixes: 7f62ffb68ad ("etnaviv: add support for rb swap")
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>

etnaviv: replace translate_clear_color with util_pack_color

This replaces the open coded etnaviv version of the color pack with the
common util_pack_color.

Fixes piglits:
arb_color_buffer_float-clear
fcc-front-buffer-distraction
fbo-clearmipmap

Fixes: c9e8b49b ("etnaviv: gallium driver for Vivante GPUs")
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>

etnaviv: remove bogus assert

etna_resource_copy_region handles resources with multiple samples
by falling back to the software path. There is no need to kill the
application there.

Fixes: c9e8b49b ("etnaviv: gallium driver for Vivante GPUs")
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>

etnaviv: use padded width/height for resource copies

When copying a resource fully we can just blit the whole level. This allows
to use the RS even for level sizes not aligned to the RS min alignment. This
is especially useful, as etna_copy_resource is part of the software fallback
paths (used in etna_transfer), that are used for doing unaligned copies.

Fixes: c9e8b49b ("etnaviv: gallium driver for Vivante GPUs")
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>

etnaviv: don't try RS blit if blit region is unaligned

If the blit region is not aligned to the RS min alignment don't try
to execute the blit, but fall back to the software path.

Fixes: c9e8b49b ("etnaviv: gallium driver for Vivante GPUs")
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Lucas Stach <dev@lynxeye.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>

Revert "amd/common: add missing libdrm include path"

This reverts commit 44b29dd7b6cdc1a3fde58c367b9de8081ac4167b.

Should no longer be required as of last patch.

Cc: Eric Engestrom <eric.engestrom@imgtec.com>

ac: remove amdgpu.h dependency

Add a couple of forward declarations and drop the amdgpu.h requirement.

With this we can build the r300 and r600 drivers without the need for
amdgpu.

v2:
 - Add amdgpu.h include in the C file (Marek)
 - Add a comment about pre C11 typedef redeclaration warning (Eric)

Cc: Nicolai Hähnle <nicolai.haehnle@amd.com>
Cc: Marek Olšák <marek.olsak@amd.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101189
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

r600g,compute: provide local copy of functions from ac_binary.c

This is a verbatim copy of the code. The functions can be cleaned up since
r600 does not use all the stuff that gcn does.
The symbol names have been changed since we still use ac_binary.h header
(for struct definition)

v2: Add ifdef guard around r600_binary_clean call (Aaron)
    Remove stray comment

Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Tested-By: Aaron Watry <awatry@gmail.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

r600: android: amdgpu_common is only required when building OpenCL

v2: split off Android changes

Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

egl/display: make platform detection thread-safe

Imagine there are 2 threads that both call _eglGetNativePlatform()
simultaneously:
- thread 1 completes the first "if (native_platform ==
  _EGL_INVALID_PLATFORM)" check and is preempted to do something else
- thread 2 executes the whole function, does "native_platform =
  _EGL_NATIVE_PLATFORM" and just before returning it's preempted
- thread 1 wakes up and calls _eglGetNativePlatformFromEnv() which
  returns _EGL_INVALID_PLATFORM because no env vars are set, updates
  native_platform and then gets preempted again
- thread 2 wakes up and returns wrong _EGL_INVALID_PLATFORM

Solve this by doing the detection in a local var and only overwriting
the global one at the end, if no other thread has updated it since.

This means the platform detected in the thread might not be the platform
returned by the function, but this is a different issue that will need
to be discussed when this becomes possible.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101252
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Grazvydas Ignotas <notasas@gmail.com>
Acked-by: Emil Velikov <emil.l.velikov@gmail.com>

egl/display: only detect the platform once

My refactor missed the fact that `native_platform` is static.
Add the proper guard around the detection code, as it might not be
necessary, and only print the debug message when a detection was
actually performed.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101252
Fixes: 7adb9b094894a512c019 ("egl/display: remove unnecessary code and
                              make it easier to read")
Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Grazvydas Ignotas <notasas@gmail.com>
Acked-by: Emil Velikov <emil.l.velikov@gmail.com>

svga: Relax the format checks for copy_region_vgpu10 somewhat

The new generic checks were actually more restrictive than the previous svga-
specific tests and not vice versa. So bypass the common format checks for
copy_region_vgpu10.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Neha Bhende <bhenden@vmware.com>

svga: Fix incorrect format conversion blit destination

The blit.dst.resource member that was used as destination was
modified earlier in the function, effectively making us try to blit
the content onto itself. Fix this and also add a debug printout when the
format conversion blits fail.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Neha Bhende <bhenden@vmware.com>

svga: Fix srgb copy_region regression

This fixes a tf2 srgb copy_region regression from
"svga: Rework the blit and resource_copy_region functionality v3"

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>

svga: Prefer accelerated blits over cpu copy region

This reduces the number of cpu copy_region fallbacks on a Nvidia system
running the piglit command

./publish/bin/piglit run  -1 -t copy -t blit tests/quick

from 64789 to 780

Previously this has caused a regression in piglit test
spec@!opengl 1.0@gl-1.0-scissor-copypixels, but I'm currently not able to
reproduce that regression.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Charmaine Lee <charmainel@vmware.com>

svga: Support accelerated conditional blitting

The blitter has functions to save and restore the conditional rendering state,
but we currently don't save the needed info.

Since also the copy_region_vgpu10 path supports conditional blitting,
we instead use the same function as the clearing routines and move
that function to svga_pipe_query.c

Note that we still haven't implemented conditional blitting with
the software fallbacks.

Fixes piglit nv_conditional_render::copyteximage

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Charmaine Lee <charmainel@vmware.com>

svga: Use utility functions to help determine whether we can use copy_region

It seems like the SVGA tests are in general more stringent than the utility
tests, but they also miss some blitter features like filters and window
rectangles, and if new blitter features are added in the future, it might
be possible that we forget adding tests for those.

So in addition to the SVGA tests, use the utility tests to restrict the
situations where we can use copy_region.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Charmaine Lee <charmainel@vmware.com>

svga: Rework the blit and resource_copy_region functionality v3

This work was initially trigged by the fact that imported surfaces may
be backed by other SVGA3D formats than the default. Therefore some fixes were
needed to avoid using the copy_region_vgpu10() functionality for incompatible
SVGA3D formats where the pipe formats were OK. This situation happens when
using dri3.

Also in some situations, for example where a R8G8_UNORM surface is backed by
an SVGA3D_NV12 format, we can't use the copy_region functionality at all and
thus need to fall back to the quad blitter also for the resource_copy_region
function. This situation doesn't happen currently, but will if we start using
video textures.

The patch makes the blit- and copy_region paths similar and the decision whether
to use a certain gpu command should now be easy to locate. Probably the
resource_copy_region path will suffer from a minor additional cpu overhead,
but on the other hand there are more cases now that we accelerate, since
we try harder before falling back to cpu copies / blits.

v2: Addressed review comments and fixed up piglit failures by sometimes
preferring cpu_copy_region() over blit().
v3: Removed a stray test statement. Updated commit message.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Charmaine Lee <charmainel@vmware.com>

i965: Improve conditional rendering in fallback paths.

We need to fall back in a couple of cases:
- Sandybridge (it just doesn't do this in hardware)
- Occlusion queries on Gen7-7.5 with command parser version < 2
- Transform feedback overflow queries on Gen7, or on Gen7.5 with
  command parser version < 7

In these cases, we printed a perf_debug message and fell back to
_mesa_check_conditional_render(), which stalls until the full
query result is available.  Additionally, the code to handle this
was a bit of a mess.

We can do better by using our normal conditional rendering code,
and setting a new state, BRW_PREDICATE_STATE_STALL_FOR_QUERY, when
we would have set BRW_PREDICATE_STATE_USE_BIT.  Only if that state
is set do we perf_debug and potentially stall.  This means we avoid
stalls when we have a partial query result (i.e. we know it's > 0,
but don't have the full value).  The perf_debug should trigger less
often as well.

Still, this is primarily intended as a cleanup.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>

configure.ac: remove manual AC_SUBST for pthread-stubs

Unneeded, since the PKG_CHECK_MODULES macro already does the
substitution of the package Cflags/Libs.

Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>

configure.ac: add -pthread to PTHREAD_LIBS

As described inline - follow what's written in the manual and what works
for all platforms that Mesa supports.

We want to untangle things leaving only -pthread, yet that has a
potential of causing regressions. Thus we'll do it as a follow-up patch.

As a nice side-effect this resolves issues, where the system lacks
libpthread.so, yet the linker does not warn about it and we and up with
unresolved symbols.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101071
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>

mesa: stop assigning unused storage for non-bindless opaque types

The storage was once used by get_sampler_uniform_value() but that
was fixed long ago to use the uniform storage assigned by the
linker.

By not assigning storage for images/samplers the constant buffer
for gallium drivers will be reduced which could result in small
perf improvements.

V2: rebase on ARB_bindless_texture

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>

egl/android: Fix typ-o

Signed-off-by: Robert Foss <robert.foss@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>

draw: check for line_width != 1.0f in validate_pipeline()

We shouldn't use the wide line stage if the line width is 1.
This check isn't strictly needed because all drivers are (now)
specifying a line wide threshold of at least 1.0 pixels, but
let's play it safe.

Reviewed-by: Charmaine Lee <charmainel@vmware.com>