anv: Determine the type of mapping based on type metadata

Before, we were just comparing the type index to 0.  Now we actually
look the type up in the table and check its properties to determine what
kind of mapping we want to do.

Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Cc: "17.1" <mesa-stable@lists.freedesktop.org>

anv: Set up memory types and heaps during physical device init

Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Cc: "17.1" <mesa-stable@lists.freedesktop.org>

anv: Predicate 48bit support on gen >= 8

This doesn't matter right now since it only affects whether or not we
set the kernel bit but, if we ever do anything else based on it, we'll
want it to be correct per-gen.

Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Cc: "17.1" <mesa-stable@lists.freedesktop.org>

anv/image: Get rid of the memset(aux, 0, sizeof(aux)) hack

Up until now, we've been memsetting the auxiliary surface to 0 at
BindImageMemory time to ensure that it is properly initialized.
However, this isn't correct because apps are allowed to freely alias
memory between different images and buffers so long as they properly
track whether or not a particular image is valid and, if it isn't,
transition from UNINITIALIZED to something else before using it.  We
now implement those transitions so we can drop the hack.

Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Cc: "17.1" <mesa-stable@lists.freedesktop.org>

anv: Handle transitioning depth from UNDEFINED to other layouts

Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Cc: "17.1" <mesa-stable@lists.freedesktop.org>

anv: Handle color layout transitions from the UNINITIALIZED layout

This causes dEQP-VK.api.copy_and_blit.resolve_image.partial.* to start
failing due to test bugs.  See CL 1031 for a test fix.

Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Cc: "17.1" <mesa-stable@lists.freedesktop.org>

st/nine: Fix a regression and syntax cleanup

A few cleanups and in particular initializing properly
the new pipe_draw_info fields.
This should fix the regression caused by
330d0607ed60fd3edca192e54b4246310f06652f

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101088

Signed-off-by: Axel Davy <axel.davy@ens.fr>
Tested-by: Edmondo Tommasina <edmondo.tommasina@gmail.com>
Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>

mesa: Remove GL_APPLE_vertex_array_object stubs

Mark the functions 'exec="skip"' in the XML instead.  libGL will still
have the functions, but the driver won't try to use them.  I verified
that this commit works with piglit's 'object-namespace-pollution glClear
vertex-array' on x64 with a driver built from mesa-12.0.3 tag.

In fairness, this test also works with a libGL built from 7927d03.  I
believe it continues to work because on non-Windows platforms we
generate some extra, dummy dispatch functions that can be used when a
driver requests a function unknown to libGL.  This was done to provide
some "forward" compatibility with drivers that need more functions.
This doesn't work on Windows because the Windows calling convention is
for the callee to clean up the stack.  That's the theory anyway.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

gallium/radeon: pipe AMDGPU_INFO_NUM_VRAM_CPU_PAGE_FAULTS into gallium HUD

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

freedreno/ir3: switch to NIR by default

Now that we lower vars to regs, we no longer regress for anything that
does complex dereferences.  (With tgsi, derefers are already lowered
before tgsi_to_nir, but not with glsl_to_nir.)  In fact it actually
fixes a few things to bypass tgsi.

So make NIR the default (finally!)

Signed-off-by: Rob Clark <robdclark@gmail.com>

freedreno/ir3: lower arrays to regs

Instead of using load/store_var intrinsics, which can have complex
derefs in the case of multi-dimensional arrays, lower these to regs
and handle the direct/indirect loads in get_src() and stores in
put_dst().

This should let us switch to using nir by default.

Signed-off-by: Rob Clark <robdclark@gmail.com>

freedreno/ir3: add put_dst()

Signed-off-by: Rob Clark <robdclark@gmail.com>

freedreno/ir3: code-motion

Signed-off-by: Rob Clark <robdclark@gmail.com>

freedreno/ir3: fix cmdline compiler

standalone_compiler_cleanup() frees the glsl types, among other things,
so it needs to come after nir->ir3.  But since we exit after dumping the
disassembly, it is easier to just not call it at all.

Signed-off-by: Rob Clark <robdclark@gmail.com>

freedreno/ir3: add missing nir_opt_copy_prop_vars() pass

Signed-off-by: Rob Clark <robdclark@gmail.com>

freedreno/ir3: need different compiler options for a5xx

vertex_id_zero_based differs..

Signed-off-by: Rob Clark <robdclark@gmail.com>

freedreno/a5xx: remove copapasta from a4xx

Won't ever hit this w/ a420 gpu, so this is dead code.  Need to get astc
working to know whether to rip this out entirely or not.

Signed-off-by: Rob Clark <robdclark@gmail.com>

freedreno: only support SSBOs with nir

tgsi_to_nir does not support them.  Note that compute shaders already
force nir.

Signed-off-by: Rob Clark <robdclark@gmail.com>

freedreno/a5xx: add some missing texture formats

Signed-off-by: Rob Clark <robdclark@gmail.com>

freedreno/a5xx: provoking vertex

Signed-off-by: Rob Clark <robdclark@gmail.com>

freedreno: update generated headers

Signed-off-by: Rob Clark <robdclark@gmail.com>

nir/lower-atomics-to-ssbo: remove atomic_uint arrays too

Maybe there is a better way to do this.  But by the time we get to
assigning uniform locs, we want the atomic_uint's to all be gone,
otherwise we assert in st_glsl_attrib_type_size().

Signed-off-by: Rob Clark <robdclark@gmail.com>

nir/lower-atomics-to-ssbo: fix num_components

Fixes some piglits like arb_shader_atomic_counters-active-counters

Signed-off-by: Rob Clark <robdclark@gmail.com>

radeon: pass flags that can change shaders to disk_cache_create()

I wasn't sure if I should filter the flags so that we only use
flags that actually change the shader output. To avoid manual
updates we just pass in everything for now.

Reviewed-by: Eduardo Lima Mitev <elima@igalia.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>

util/disk_cache: add new driver_flags param to cache keys

This will be used for things such as adding driver specific environment
variables to the key. Allowing us to set environment vars that change
the shader and not have the driver ignore them if it finds existing
shaders in the cache.

Reviewed-by: Eduardo Lima Mitev <elima@igalia.com>

u_format_test: Ignore S3TC errors.

This prevents spurious failures when libtxc-dxtn-s2tc is installed.

Note: lp_test_format doesn't need any change since we were already
ignoring S3TC failures there.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Reviewed-by: Rhys Kidd <rhyskidd@gmail.com>

docs: Document ASTC extension support for SKL and BXT

v2: Remove the '+' after bxt

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

i965: Enable ASTC HDR for Broxton

This platform passes the following GLES3 tests:
ES3-CTS.functional.texture.compressed.astc.endpoint_value_hdr_cem_*

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

intel/isl: Add ASTC HDR to format lists and helpers

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

radv: Add compute HTILE fast clear.

Not really what the fast depth clear does, no matter whether you use
EXPCLEAR or not. Seems the fast clear using the DB HW always touches
the main buffer.

Signed-off-by: Bas Nieuwenhuizen <basni@google.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>

radv: Use correct clear words for HTILE.

Did some RE'ing what several HTILE words give when read from a descriptor
with HTILE compression enabled.

Seems to align with -pro usage for D16 too.

Signed-off-by: Bas Nieuwenhuizen <basni@google.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>

radv: Add queue masks for htile usage determination.

Signed-off-by: Bas Nieuwenhuizen <basni@google.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>

radv: Specify semantics of HTILE layout helpers.

And correct implementation to specify only what we support.

Signed-off-by: Bas Nieuwenhuizen <basni@google.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>

radv: Don't use a separate can_expclear.

We never use EXPCLEAR clears.

Signed-off-by: Bas Nieuwenhuizen <basni@google.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>

mesa: GL_ARB_shader_subroutine is not optional in core profile

   text    data     bss     dec     hex filename
7038459  235248   37280 7310987  6f8e8b 32-bit i965_dri.so before
7038227  235248   37280 7310755  6f8da3 32-bit i965_dri.so after
6681438  303400   50608 7035446  6b5a36 64-bit i965_dri.so before
6681254  303400   50608 7035262  6b597e 64-bit i965_dri.so after

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>

drirc: Add allow_glsl_builtin_variable_redeclaration for Dead Island Riptide Definitive Edition

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

gallium/radeon: add a query for monitoring Gallium thread load

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

radeonsi/gfx9: compile shaders with +xnack

so that LLVM doesn't allocate SGPRs where XNACK is.

Cc: 17.1 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>

vc4: Remove dead code in vc4_dump_surface_msaa()

Coverity caught the use of dead code copy-paste for
found_colors[] and num_found_colors.

CID: 1341850
Signed-off-by: Rhys Kidd <rhyskidd@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>

egl/wayland: verify event queue was allocated

We're already verified that 'window' wasn't NULL, I'm guessing this
allocation error is about the newly created queue.

CID: 1409754
Fixes: 03dd9a88b0b ("egl/wayland: Use per-surface event queues")
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>

mesa: add APPLE_vertex_array_object stubs

APPLE_vertex_array_object support was removed in 7927d0378fc7.
However it turns out we can't remove the functions because this
can cause issues when libglapi is used together with DRI
drivers built prior to said commit

Fixes: 7927d0378fc ("mesa: drop APPLE_vertex_array_object support")
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

glsl: set mask via initialisation list rather than in constructor body

Potentially more efficient as it may avoid the struct being initialised
twice.

Also add var to the initialisation list while we are here.

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

ralloc: Use strnlen() inside of strncat()

If the str is long or isn't null-terminated, strlen() could take a lot
of time or even crash. I don't know why was it used in the first place,
maybe for platforms without strnlen(), but strnlen() is already used
inside of ralloc_strndup(), so this change should not additionally
break anything.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

glcpp: Skip unnecessary line continuations removal

Overwhelming majority of shaders don't use line continuations. In my
shader-db only shaders from the Talos Principle and Serious Sam used
them, less than 1% out of all shaders. Optimize for this case, don't
do any copying if no line continuation was found.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

glcpp: Avoid unnecessary strcmp()

strcmp() is slow. Initiate comparison with "__LINE__" or "__FILE__"
only if the identifier starts with '_', which is rare.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

main: Move hashLockMutex/hashUnlockMutex to header and inline

Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>

main: Use _mesa_HashLock/UnlockMutex consistently

This is shorter and easier on the eyes. At the same time this
also ensures that we are always asserting that the table pointer
is not NULL. Currently that was not done for all situations.

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

util: Change the pointer hashing function

Use our knowledge that pointers are at least 4 byte aligned to remove
the useless digits. Then shift by 6, 10, and 14 bits and add this to
the original pointer, effectively folding in the entropy of the higher
bits of the pointer into a 4-bit section. Stopping at 14 means we can
add the entropy from 18 bits, or at least a 600Kbyte section of memory.
Assuming that ralloc allocates from a linearly allocated heap less than
this we can make a very efficient pointer hashing function for our usecase.
Even if we are not on an architecture that is 4 byte aligned, there is
still a high big chance that the thing we are allocating is at least
8 bytes in size, so even then we will have entropy into the third bit.

The 4 bit increment on the shifts is chosen rather arbitrarily; if we
had chosen a 3 bit increment we would need to add another xor to
cover a decently sized memorypool. Increasing it to 5 bits would
spread our entropy more, possibly hurting us with more collisions on
hash tables of size less than 32. With a hash table of size 16 there
are a max of 11 entries, and we can assume that with such a small table
collisions are not that painfull.

This allows us to hash the whole 32 or 64 bit pointer at once,
instead of running FNV1a, looping through each byte and doing
increments, decrements, muls, and xors on every byte. This cuts
_mesa_hash_data from 1.5 % on profiles, to making _mesa_hash_pointer
show up with a 0.09% share. Collisions on insertion actually seems to be
ever so slightly lower with this hash function, as found by printing
a loop counter and sorting the data.

perf stat shows a 1.5% reduction in instruction count,
and a 5% reduction in stalled cycles. Shader-db runtime goes
from 225 to 220 seconds.

No instruction-count changes in shader-db, but there are some minor
changes in cycle-count that is likely caused by nir walking a set
in some of its passes, and this causing a different ordering.
That might eventually lead to a difference in register allocation.
However, the effect is a net positive;

total cycles in shared programs: 24739550 -> 24738482 (-0.00%)
cycles in affected programs: 374468 -> 373400 (-0.29%)
helped: 178
HURT: 49

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Eric Anholt <eric@anholt.net>

vulkan/wsi/wayland: Fix proxy wrappers for swapchain recreation

Before the swapchain event queue is destroyed, all proxy objects that reference
it must be dropped. Otherwise we risk a use-after-free if a frame callback event
or buffer release events are received afterwards.
This happens when an application destroys and recreates a swapchain in FIFO
mode between two frames without using the VkSwapchainCreateInfoKHR::oldSwapchain
mechanism to keep the old swapchain until after the next redraw.

Fixes: 5034c615582a ("vulkan/wsi/wayland: Use proxy wrappers for swapchain")
Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
Cc: mesa-stable@lists.freedesktop.org

drirc: Add allow_glsl_builtin_variable_redeclaration for Dying Light and Dead Island Definitive Edition

This fixes the long-standing problem with Dying Light where the game would
produce a black screen when running under Mesa. This happened because the
game's vertex shaders redeclare gl_VertexID, which is a GLSL builtin.
Mesa's GLSL compiler is a little more strict than others, and would not
compile them:

    error: `gl_VertexID' redeclared

The allow_glsl_builtin_variable_redeclaration directive allows the shaders
to compile and the game to render. The game also requires OpenGL 4.4+ (GLSL
440), but does not request it explicitly. It must be forced with an
override, such as MESA_GL_VERSION_OVERRIDE=4.5 and
MESA_GLSL_VERSION_OVERRIDE=450. A compatibility context is *not* required
and forcing one with 4.5COMPAT or allow_higher_compat_version results in
graphical artifacts.

Dead Island Definitive Edition is another Techland port on the same engine
with the same problems, so we set the
allow_glsl_builtin_variable_redeclaration option for that game as well.

v2 (Samuel Pitoiset):
    - Rename allow_glsl_builtin_redeclaration ->
      allow_glsl_builtin_variable_redeclaration

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96449
Signed-off-by: John Brooks <john@fastquake.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>

glsl: Conditionally allow redeclaration of built-in variables

Conditional on allow_glsl_builtin_variable_redeclaration driconf option.

v2 (Samuel Pitoiset):
    - Rename allow_glsl_builtin_redeclaration ->
      allow_glsl_builtin_variable_redeclaration
    - style: put spaces after 'if'

Signed-off-by: John Brooks <john@fastquake.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>

driconf: Add allow_glsl_builtin_variable_redeclaration option

This option will allow GLSL builtins to be redeclared verbatim (e.g.
redeclaring "in int gl_VertexID" in a vertex shader). This is not strictly
valid and would normally fail to compile, but some applications (such as
newer Techland ports) do it and need more leniency.

v2 (Samuel Pitoiset):
    - Rename allow_glsl_builtin_redeclaration ->
      allow_glsl_builtin_variable_redeclaration

Signed-off-by: John Brooks <john@fastquake.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>

nv50,nvc0: clear index buffer bufctx bin unconditionally

The previous condition was to clear it out if it had previously been
set, not what's in the current draw. That information is gone now, so
just clear it unconditionally.

Fixes: 330d0607e ("gallium: remove pipe_index_buffer and set_index_buffer")
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>

nv50: fix vtxbuf cleanup

Use a user-buffer-aware cleanup function.

Fixes: c24c3b94ed ("gallium: decrease the size of pipe_vertex_buffer - 24 -> 16 bytes")
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>

i965: Use the upload BO for push constants on Gen7.5-Gen8.

We can easily use the upload BO for push constants on Gen7.5/Gen8 too,
at the cost of a relocation when emitting 3DSTATE_CONSTANT_XS.  We can
simply switch to using constant buffer pointer 2 instead of pointer 0,
like we do on Gen9+.

Ivybridge and Baytrail can't do this trick because they require the
constant buffers to be enabled in order, starting with 0.  We'd have
to set the INSTPM bit to make the constant buffer pointer not relative
to dynamic state base address, which would need kernel command parser
support.

Improves performance in GLBenchmark 2.7/TRex Offscreen by:
- Broadwell GT2: 0.305608% +/- 0.19877% (n = 68)
- Braswell: No difference proven (n = 742)
- Haswell GT3e: 0.180755% +/- 0.0237505% (n = 30)

Reviewed-by: Chris Forbes <chrisforbes@google.com>

i965: Use the upload BO for push constants on Gen9+.

Shaders can use quite a bit of uniform data.  Better to put it in the
upload buffers, like we do for client vertex data, rather than the
batch buffer state area, which is primarly used for indirect state.

This should free up batch space, allowing us to emit more commands in a
batch before flushing.  Because BRW_NEW_BATCH also causes a lot of state
to be re-emitted, it may also reduce CPU overhead a little bit.

We took this approach on Gen4-5, but switched to using the batch area
on Gen6+ because buffer 0 is relative to Dynamic State Base Address by
default, which is set to the start of the batch.

On Gen9+, we already use a relocation due to a workaround, so this is
trivial to change and has basically no downside.

Unfortunately we can't change compute shader push constants because
MEDIA_CURBE_LOAD always uses an offset from dynamic state base address.

Improves performance in GLBenchmark 2.7/TRex Offscreen by:
- Skylake GT4e: 0.52821% +/- 0.113402% (n = 190)
- Apollolake: 0.510225% +/- 0.273064% (n = 70)

Reviewed-by: Chris Forbes <chrisforbes@google.com>

i965: Drop BRW_NEW_PUSH_CONSTANT_ALLOCATION from CS packets.

I don't think CS push constant uploading uses the section of L3
controlled by 3DSTATE_PUSH_CONSTANT_ALLOC_XS.  So I don't think
it needs to be re-emitted when that space is reallocated.

The programming note in gen7_allocate_push_constants doesn't
indicate this is necessary, at least.

Reviewed-by: Chris Forbes <chrisforbes@google.com>

nvc0/ir: SHLADD's middle source must be an immediate

The instruction encodings only allow for immediates. Don't try to
replace a zero (which is dumb to have in that op in any case) with RZ.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org

android: add -Wl,--build-id=sha1 to LDFLAGS for libvulkan_intel

Just like is done on desktop and what is expected by the build-id code.

Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>

configure.ac: s/xcb-fixes/xcb-xfixes/

Former is not a thing, even if I have a hacked xcb-fixes.pc on my system.
Thanks for spotting it Mark!

Fixes: 9a90d6a9d4e ("configure.ac: add xcb-fixes to the XCB DRI3 list")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

configure.ac: add xcb-fixes to the XCB DRI3 list

The XCB module is used by the VL targets. Thus omitting it can lead to
link-time errors due to unresolved symbols.

Other DRI3 users such as the Vulkan WSI and the dri3 loader helper do
not use an update region in their xcb_present_pixmap() call. We will
look into that at a later stage.

Fixes: acf3d2afab0 ("configure: check once for DRI3 dependencies")
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101110
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

automake: add SWR LLVM gen_builder.hpp workaround

As gen_builder.hpp file is generated, it contains information that is
specific to the LLVM version it originates from.

As suggested by Tim, the file seems to be forwards compatible. So in
order to produce ship a file which will work everywhere we should be
using earlies supported LLVM - 3.9.

With this we're back on track and can build all of mesa without
python/mako/flex and friends.

In the long term we might want to see if the python generators can be
updated to produce LLVM version agnostic files. At least within the
range supported by SWR.

Cc: <mesa-stable@lists.freedesktop.org>
Cc: Chuck Atkins <chuck.atkins@kitware.com>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Tim Rowley <timothy.o.rowley@intel.com>

st/mesa: don't mark the program as in cache_fallback when there is cache miss

When we fallback currently the gl_program objects are re-allocated.

This is likely to change when the i965 cache lands, but for now
this fixes a crash when using MESA_GLSL=cache_fb. This env var
simulates the fallback path taken when a tgsi cache item doesn't
exist due to being evicted previously or some kind of error.

Unlike i965 we are always falling back at link time so it's safe to
just re-allocate everything. We will be unnecessarily freeing and
re-allocate a bunch of things here but it's probably not a huge deal,
and can be changed when the i965 code lands.

Fixes: 0e9991f957e2 ("glsl: don't reference shader prog data during cache fallback")
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>

mesa: add an env var to force cache fallback

For the gallium state tracker a tgsi binary may have been evicted
from the cache to make space. In this case we would take the
fallback path and recompile/link the shader.

On i965 there are a number of reasons we can get to the program
upload stage and have neither IR nor a valid cached binary.
For example the binary may have been evicted from the cache or
we need a variant that wasn't previously cached.

This environment variable enables us to force the fallback path that
would be taken in these cases and makes it easier to debug these
otherwise hard to reproduce scenarios.

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

st/mesa: improve shader cache debug info

This will explicitly state that we are following the fallback
path when we find invalid/corrupt cache items. It will also
output the fallback message when the fallback path is forced
via an environment variable, the following patches will allow
this.

Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>

travis: remove workarounds for the Vulkan target

Previously we required --enable-egl for the platform selection to work.
Additionally due to the broken DRI3 dependency tracking we needed
--enable-glx.

Since both of these are now sorted now we no longer need the
workarounds.

While we're here, explicitly enable dri3.

Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>

configure: trivial whitespace cleanup

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

configure: error out if building XVMC w/o supported platform

Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>

configure: error out if building VDPAU w/o supported platform

Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>

configure: error out if building OMX w/o supported platform

Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>

configure: error out if building VA w/o supported platform

A bit pedantic patch to fool proof should someone start thinkering
without knowing what they do.

Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>

st/xvmc: add DRI3 support

Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Tested-by: Ilia Mirkin <imirkin@alum.mit.edu>

st/omx: add DRI3 support

Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-and-Tested-by: Leo Liu <leo.liu@amd.com>

gallium/targets: link against XCB only as needed

OMX and VA can optionally use the X11 DRI2/DRI3, thus we should link
only as required.

Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

st/omx: fix building against X11-less setups

The vl_*_screen_create API properly falls back to a NOP when we're
building without specific platforms. So the only thing we need is to
handle the lack of X11/Xlib.h and provide a dummy Display define.

Cc: <mesa-stable@lists.freedesktop.org>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

st/omx: remove unneeded X11 include

En route to a X11-less builds

Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Nayan Deshmukh <nayan26deshmukh@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

st/omx: remove unused drm_driver.h includes

Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Nayan Deshmukh <nayan26deshmukh@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

st/va: check if vl_*_screen_create has failed only once

Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Nayan Deshmukh <nayan26deshmukh@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

st/va: fix misplaced closing bracket

It's been like this since the code was introduced.

Fixes: 86eb4131a90 (st/va: add headless support, i.e. VA_DISPLAY_DRM)
Cc: <mesa-stable@lists.freedesktop.org>
Cc: Julien Isorce <julien.isorce@gmail.com>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Nayan Deshmukh <nayan26deshmukh@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

st/va: move variable declaration to where its used

... and make it const, since we shouldn't tinker with it.

Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Nayan Deshmukh <nayan26deshmukh@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

auxiliary/vl: use vl_*_screen_create stubs when building w/o platform

Provide a dummy stub when the user has opted w/o said platform, thus
we can build the binaries without unnecessarily requiring X11/other
headers.

In order to avoid build and link-time issues, we remove the HAVE_DRI3
guards in the VA and VDPAU state-trackers.

With this change st/va will return VA_STATUS_ERROR_ALLOCATION_FAILED
instead of VA_STATUS_ERROR_UNIMPLEMENTED. That is fine since upstream
users of libva such as vlc and mpv do little error checking, let
alone distinguish between the two.

Cc: Leo Liu <leo.liu@amd.com>
Cc: Guttula, Suresh <Suresh.Guttula@amd.com>
Cc: mesa-stable@lists.freedesktop.org
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

configure: error out when building X11 Vulkan without DRI3

Vulkan supports only DRI3 enabled X11 platforms. Make it obvious,
should one consider building without it.

Cc: Jason Ekstrand <jason@jlekstrand.net>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>

loader: build libloader_dri3_helper.la only with HAVE_PLATFORM_X11

Pretty much every other place does the same.

Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>

vulkan: automake: remove unused VULKAN_LIB_DEPS variable

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

configure: check once for DRI3 dependencies

Currently we are having the XCB_DRI3 dependencies duplicated,
partially.

Just do a once-off check and add all of the respective CFLAGS/LIBS
where needed.

As a nice side effect this helps us solve a couple of FIXMEs.

DRI3 is not a thing w/o X11 so disable it in such cases.

Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>

configure: error out when building GLX w/o the X11 platform

Building EGL/Vulkan/other without X11, while GLX is enabled is confusing
and misleading. In practise anyone aiming at the former will also
disable GLX.

The inverse (some examples below) should still work:
 ./configure --disable-glx --with-platforms=x11 --with-vulkan-drivers=intel
 ./configure --disable-glx --with-platforms=x11 --enable-egl

Keep in mind that the X11 platform is enabled, by default.

Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>

configure: set HAVE_foo_PLATFORM as applicable

Rather than having multiple places that define the macros, do it just
once in configure. Makes existing code a bit shorter and easier to
manage as we fix the VL targets with follow-up commits.

Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>

configure: enable the surfaceless platform by default

A simple platform that you want to use in a many usecases. See the
spec file details.

It has no special requirements plus it takes less than a second to
build.

Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>

configure: loosen --with-platforms heuristics

Remove the enable-egl pre-requirement. Platform selection does not
depend on EGL.

Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>

configure: update remaining --with-egl-platforms references

Rename the remaining references to omit the egl part.

Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>

configure: rename remaining HAVE_EGL_PLATFORM_* guards

Analogous to others earlier, these will be used to control the platform
for more than the EGL driver.

Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>

configure: move platform handling further up

We'll need it for the Vulkan drivers and the VL targets.

Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>

Android: r600: fix build when LLVM is disabled

There's still an error after my recent clean-up if LLVM is not patched to
enable AMDGPU target:

external/mesa3d/src/amd/common/ac_llvm_util.c:38:2: error: implicit declaration of function 'LLVMInitializeAMDGPUTargetInfo' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
        LLVMInitializeAMDGPUTargetInfo();
        ^
external/mesa3d/src/amd/common/ac_llvm_util.c:39:2: error: implicit declaration of function 'LLVMInitializeAMDGPUTarget' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
        LLVMInitializeAMDGPUTarget();
        ^
external/mesa3d/src/amd/common/ac_llvm_util.c:40:2: error: implicit declaration of function 'LLVMInitializeAMDGPUTargetMC' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
        LLVMInitializeAMDGPUTargetMC();
        ^
external/mesa3d/src/amd/common/ac_llvm_util.c:41:2: error: implicit declaration of function 'LLVMInitializeAMDGPUAsmPrinter' is invalid in C99 [-Werror,-Wimplicit-function-declaration]
        LLVMInitializeAMDGPUAsmPrinter();
        ^

We need to drop libmesa_amd_common when LLVM is disabled, however there's
still a dependency on include paths for ac_binary.h. So explicitly add the
include path when LLVM is disabled.

Signed-off-by: Rob Herring <robh@kernel.org>
Acked-by: Nicolai Hähnle <nicolai.haehnle@amd.com>

virgl: fix virgl_bo_transfer_{put, get} box struct copy

Commit 3dfe61ed6ec6 ("gallium: decrease the size of pipe_box - 24 -> 16
bytes") changed the size of pipe_box, but the virgl code was relying on
pipe_box and drm_virtgpu_3d_box structs having the same size/layout doing
a struct copy. Copy the fields one by one instead.

Cc: Marek Olšák <marek.olsak@amd.com>
Cc: Dave Airlie <airlied@redhat.com>
Fixes: 3dfe61ed6ec ("gallium: decrease the size of pipe_box - 24 -> 16 bytes")
Signed-off-by: Rob Herring <robh@kernel.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>

egl: add g_egldispatchstubs.h to the release tarball

Fixes: ce562f9e3fa ("EGL: Implement the libglvnd interface for EGL (v3)")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

egl/android: fix segfault within swap_buffers

Function droid_swap_buffers may get called without dri2_surf->buffer set,
in these cases we don't have a back buffer set either. Patch fixes segfault
seen with 3DMark that uses android.opengl.GLSurfaceView for rendering it's UI.

backtrace:
   #00 pc 00013f88  /system/lib/egl/libGLES_mesa.so (droid_swap_buffers+104)
   #01 pc 000117b2  /system/lib/egl/libGLES_mesa.so (dri2_swap_buffers+50)
   #02 pc 000058b2  /system/lib/egl/libGLES_mesa.so (eglSwapBuffers+386)
   #03 pc 00011329  /system/lib/libEGL.so (eglSwapBuffersWithDamageKHR+553)
   #04 pc 000118e7  /system/lib/libEGL.so (eglSwapBuffers+55)
   #05 pc 000754dc  /system/lib/libandroid_runtime.so

Note, this is v1 as v2 caused dEQP regressions.

Fixes: 2acc69d ("EGL/Android: Add EGL_EXT_buffer_age extension")
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Acked-by: Emil Velikov <emil.velikov@collabora.com>
Cc: "17.1" <mesa-stable@lists.freedesktop.org>

egl/wayland: Ensure we get a back buffer

Commit 9ca6711faa03 changed the Wayland winsys to only block for the
frame callback inside SwapBuffers, rather than get_back_bo. get_back_bo
would perform a single non-blocking Wayland event dispatch, to try to
find any release events which we had pulled off the wire but not
actually processed. The blocking dispatch was moved to SwapBuffers.

This removed a guarantee that we would've processed all events inside
get_back_bo(), and introduced a failure whereby the server could've sent
a buffer release event, but we wouldn't have read it. In clients
unconstrained by SwapInterval (rendering ~as fast as possible), which
were being displayed directly without composition (buffer release delayed),
this could lead to get_back_bo() failing because there were no free
buffers available to it.

The drawing rightly failed, but this was papered over because of the
path in eglSwapBuffers() which attempts to guarantee a BO, in order to
support calling SwapBuffers twice in a row with no rendering actually
having been performed.

Since eglSwapBuffers will perform a blocking dispatch of Wayland
events, a buffer release would have arrived by that point, and we
could then choose a buffer to post to the server. The effect was that
frames were displayed out-of-order, since we grabbed a frame with random
past content to display to the compositor.

Ideally get_back_bo() failing should store a failure flag inside the
surface and cause the next SwapBuffers to fail, but for the meantime,
restore the correct behaviour such that get_back_bo() no longer fails.

Signed-off-by: Daniel Stone <daniels@collabora.com>
Reported-by: Eero Tamminen <eero.t.tamminen@intel.com>
Acked-by: Pekka Paalanen <pekka.paalanen@collabora.co.uk>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98833
Fixes: 9ca6711faa03 ("Revert "wayland: Block for the frame callback in get_back_bo not dri2_swap_buffers"")

egl/wayland: Use per-surface event queues

During display initialisation, we need a separate event queue to handle
the registry events, which is correctly handled. But we also need
separate per-surface event queues to handle swapchain-related events,
such as surface frame events and buffer release events. This avoids two
surfaces from the same EGLDisplay, both current on separate threads,
dispatching each other's events.

Create separate per-surface event queues, create wl_surface and wl_drm
proxy wrapper objects per surface, so we eliminate the race around
sending events to the wrong queue. swrast buffers do not need a
dedicated proxy wrapper, as the wl_shm_pool used to create the
wl_buffers, being transient, can itself be assigned to a queue.

Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 36b9976e1f99 ("egl/wayland: Avoid race conditions when on non-main thread")
Cc: mesa-stable@lists.freedesktop.org

egl/wayland: Don't open-code roundtrip

wl_display_roundtrip_queue() exists and can replace roundtrip(). The
API was introduced with wayland 1.6, while we currently require 1.11.

Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable@lists.freedesktop.org

vulkan/wsi/wayland: Use proxy wrappers for swapchain

Though most swapchain operations used a queue, they were racy in that
the object was created with the queue only set later, meaning that its
event could potentially be dispatched from the default queue in between
these two steps.

Use proxy wrappers to avoid this race, also assigning wl_buffers created
for the swapchain to the event queue.

Signed-off-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable@lists.freedesktop.org