If the driver said it can't do the shader, then listen to it and don't ask
it to create the shaders anyway. Fixes a bunch of spam on i915/r300 (with
!16878) about unsupported opcodes during dEQP runs.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16895>
When lower_wpos_pntc is used, the state tracker inserts code to
transform gl_PointCoord.y according to a uniform, to account for
API-requested point coordinate origin and framebuffer orientation. With
the transformation, driver-supplied point coordinates are expected to
have an upper left origin.
If the hardware point coordinate supports (only) a lower left origin,
the backend has to use lower_wpos_pntc and then lower *again* to flip
back. This ends up transforming twice, which is wasteful:
a = load point coord Y with lower left origin
a' = 1.0 - a
a'' = uniform_transform(a')
However, lower_wpos_pntc is quite capable of transforming for a lower
left origin too, it just needs to flip the transformation. Add a CAP
specifying the point coordinate origin convention, rather than assuming
upper-left. This simplifies the Asahi code greatly.
Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16829>
to allow exposing 4G - 1. The "SIZE" was also a misnomer because it meant
elements. This no longer clamps the size to INT_MAX in st/mesa.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16881>
so that values greater than INT_MAX report INT_MAX with glGetInteger and
the full value with glGetInteger64. This allows exposing compute grid sizes
and max buffer sizes greater than INT_MAX.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16881>
nir_lower_int_to_float() does this at the end of compilation, no need to
do it up front.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16823>
NIR has fdiv, and all the NIR backends have to have lower_fdiv set
appropriately already since various passes (format conversions,
tgsi_to_nir, nir_fast_normalize(), etc.) might generate one.
This causes softpipe and llvmpipe to now do actual divides, since
lower_fdiv is not set there. Note that llvmpipe's rcp implementation is a
divide of 1.0 by x, so now we're going to be just doing div(x, y) instead
of mul(x, div(1.0, y)).
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16823>
It's way more concise to write as nir_builder calls.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16823>
The drivers not setting it were:
- nv30, which gets lowering using NIR's lower_fsat flag.
- r300, which gets lowering using NIR's lower_fsat flag.
- a2xx, which has was getting it optimized back to fsat anyway.
This drops the check for the cap from gallium nine. While nine does have
a non-nir path, I think it's safe to assume that if you have SM3
texturing, you can do fsat.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16823>
Use an internal geometry shader to handle input primitives. Do full
accurate culling and clipping in the shader and output hit result and
min/max depth to a SSBO for final being written to select buffer.
With multiple result slots in SSBO we can left multiple draws on the
fly and wait them done when buffer is full or exit GL_SELECT mode.
This provides quicker selection response compared to software based
solution. Tested on Discovery Studio 2020: some complex model needs
1~2s selection response time originally, now it's almost selected
immidiately.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15765>
When glthread not enabled, CurrentClientDispatch and CurrentServerDispatch
should be same. This does not cause problems before because OutsideBeginEnd
and BeginEnd have same BeginEnd entries, so when
CurrentServerDispatch==OutsideBeginEnd
CurrentClientDispatch==BeginEnd
will call into same BeginEnd _mesa_* functions.
But we'll add another dispatch table to replace BeginEnd when HW GL_SELECT
mode, so this needs to be fixed. Otherwise some function like _mesa_Rectf
which always call with CurrentServerDispatch will go into wrong entries.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15765>
For constructing dispatch table used in GL_SELECT mode. Every vertex
inserted need to also insert a name stack offset attribute.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15765>
HW code path will not flush vertex whenever name stack change.
It will save the current name stack and write to select buffer
only when no space left or exit select mode.
This let us submit multi draws from different name stack at
once instead of submit draws for a single name stack then
wait it finish before submit next one.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15765>
No functional change, just pack existing software based implementation into
the HardwareAcceleratedSelect switch, will add hardware implementation in
next commit.
ctx->Select.NameStackDepth is sure to be <=MAX_NAME_STACK_DEPTH, so removed
the overflow check in _mesa_LoadName.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Sgined-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15765>
outputs_written is uint64_t, should count max reg number
by util_last_bit64(). Otherwise the following access will
overflow the allocated array with a smaller size.
cc: mesa-stable
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15765>
according to the spec, atomic counters can be bound at any offset divisible by 4,
which means that any driver that uses the ssbo lowering pass and doesn't have
a min offset align of 4 is potentially broken
to handle this, use a statevar to inject the misaligned remainder of the offset
into the shader as a uniform. for well-aligned counter binds, the uniform offset
will be 0
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16749>
this function should be called late to allow for other passes potentially
making changes which affect the states in use by shaders
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16749>
some hardware can't do a ssbo offset=4, as required by the atomic->ssbo
lowering pass, so for these cases an offset can be passed for the counter
as a uniform, and the shaders can be adjusted accordingly
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16749>
This is used for the old, buggy and slow GLSL IR loop unrolling
code. All drivers have now switched to the NIR unrolling code so
here we remove the CAP.
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16366>
Calling this directly in the linker code allows us to place it between
the varying linker and uniform linker calls which allows for better
optimisation/removal of uniforms.
Also in a later patch it allows us to insert a new nir based
lower_const_arrays_to_uniforms() call after the gl_nir_link_opts()
call. This is important because it allows the linking opts to
move constant arrays to later stages if possible before
lower_const_arrays_to_uniforms() turns them into uniforms.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6541
Acked-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16770>
st_texture_release_all_sampler_views uses the validate_mutex,
but st_get_texture_sampler_view_from_stobj didn't.
Since they both modify stObj->view we could have threadA in
st_get_texture_sampler_view_from_stobj with a non-NULL sv,
so expecting sv->view to be non-NULL, while threadB was in
st_texture_release_all_sampler_views clearing sv->view.
It's also needed to protect st_sampler_view::private_refcount,
which is supposed to be used from the owning context thread,
but can also be used by any context in st_texture_release_all_sampler_views.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6088
Fixes: ef5d427413 ("st/mesa: add a mechanism to bypass atomics when binding sampler views")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16779>
this generally means an app is broken in some way, so throw a log
message to be helpful
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16821>
Since we now depend on C11, we know that we have support for the C99
math functionality. So let's drop the c99_math.h compatibility wrapper,
and just include <math.h> directly.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16812>
The callback allows to request different vectorization factors
per instruction depending on e.g. bitsize or opcode.
This patch also removes using the vectorize_vec2_16bit option
from nir_opt_vectorize().
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13080>
NIR already has the necessary lowering, and the GLSL lowering violates
GLSL IR validation rules. Once quadop lowering was turned off, the IR
validation at the end of the compile path on DEBUG builds caught the
problem.
In order to move the lowering to NIR, though, we need to make sure that
drivers supporting these functions actually have the lowering flag set.
xfails added for t860, where apparently this tickles a variety of existing
64-bit bugs in the backend.
Fixes: #6461
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Mykhailo Skorokhodov <mykhailo.skorokhodov@globallogic.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16437>
lower_int64 doesn't need it, and the scalarizing ended up tickling some
bug in virgl once I started using lower_int64.
Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16437>
Emma Anholt
Alyssa Rosenzweig
Marek Olšák
Mike Blumenkrantz
Qiang Yu
Timothy Arceri
Pierre-Eric Pelloux-Prayer
Erik Faye-Lund
Daniel Schürmann