Now that higher levels are enforcing decoration sanity, we don't need
the vtn_asserts here. This function *should* be safe but we still want
a few well-placed regular asserts in case something goes awry.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
This reworks the error checking on our generic handling of decorations.
The objective is to validate all of the SPIR-V assumptions we make
up-front and convert redundant checks to compiled-out asserts. The most
important part of this is to ensure that member decorations only occur
on OpTypeStruct and that the member is never out-of-bounds. This way
later code can assume that the member is sane and not have to worry
about OOB array access due to a misplaced OpMemberDecorate.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
This is a bit simpler since we have fewer enum values in the case. It's
also a bit more efficient because we're making fewer glsl_get_* calls.
While we're at it, add better type validation.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Now that vtn_base_type is a real and full base type, we can switch on
that instead of the GLSL base type which is a lot fewer cases in our
switch.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
VGPR1 = InstanceID / StepRate0; // StepRate0 can be set to 1
Ported from RadeonSI.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
For Vega10 and Raven that need a special workaround for the
scissor bug.
This seems to give a minor boost for Talos and Dota 2, at least.
To reduce the cost of memcmp, the driver checks if it's
really useful to do the comparison.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
VGPR1 is only needed for topology that needs 3 offsets like
triangles or quads.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
../../src/util/.libs/libmesautil.a(libmesautil_la-u_queue.o): In function `u_thread_setname':
/builddir/build/BUILD/mesa-17.3.1/src/util/../../src/util/u_thread.h:66: undefined reference to `pthread_setname_np'
../../src/util/.libs/libmesautil.a(libmesautil_la-u_queue.o): In function `thrd_join':
/builddir/build/BUILD/mesa-17.3.1/src/util/../../include/c11/threads_posix.h:336: undefined reference to `pthread_join'
../../src/util/.libs/libmesautil.a(libmesautil_la-u_queue.o): In function `u_thread_create':
/builddir/build/BUILD/mesa-17.3.1/src/util/../../src/util/u_thread.h:48: undefined reference to `pthread_sigmask'
../../src/util/.libs/libmesautil.a(libmesautil_la-u_queue.o): In function `thrd_create':
/builddir/build/BUILD/mesa-17.3.1/src/util/../../include/c11/threads_posix.h:296: undefined reference to `pthread_create'
../../src/util/.libs/libmesautil.a(libmesautil_la-u_queue.o): In function `u_thread_create':
/builddir/build/BUILD/mesa-17.3.1/src/util/../../src/util/u_thread.h:50: undefined reference to `pthread_sigmask'
/builddir/build/BUILD/mesa-17.3.1/src/util/../../src/util/u_thread.h:50: undefined reference to `pthread_sigmask'
../../src/util/.libs/libmesautil.a(libmesautil_la-u_queue.o): In function `call_once':
/builddir/build/BUILD/mesa-17.3.1/src/util/../../include/c11/threads_posix.h:96: undefined reference to `pthread_once'
../../src/util/.libs/libmesautil.a(libmesautil_la-u_queue.o): In function `u_thread_get_time_nano':
/builddir/build/BUILD/mesa-17.3.1/src/util/../../src/util/u_thread.h:84: undefined reference to `pthread_getcpuclockid'
collect2: error: ld returned 1 exit status
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Igor Gnatenko <ignatenko@redhat.com>
This was never enabled in secondary buffers because hiz_enabled was
never set to true for those.
If the app provides a framebuffer in the inheritance info when beginning
a secondary buffer, we can determine if HiZ is enabled and therefore
allow the PMA optimization to be enabled within the command buffer.
This improves performance by ~13% on an internal benchmark on Skylake.
v2: Use anv_cmd_buffer_get_depth_stencil_view().
Signed-off-by: Alex Smith <asmith@feralinteractive.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Respect the std430 rules for determining offset and size of struct
members when using a std430 buffer. std140 rules lead to wrong buffer
offsets in that case.
Fixes my test case attached in Bugzilla. No piglit changes.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104492
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
A part of the driver constbuf area is allocated for bindless images. Any
update requires uploading to all driver constbufs. This also extends the
driver constbuf to 64KB, up from 2KB.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
This keeps a list of resident textures (per context), and dumps that
list into the active buffer list when submitting. We also treat bindless
texture fetches slightly differently, wrt the meaning of indirect, and
not requiring the SAMPLER file to be used.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
In preparation for bindless images, we have to retrieve the
target/format info from the instruction directly, as there will be no
declaration. Furthermore, for bound images, this information is still
available in the instruction, so we can drop the declaration-based
mechanism entirely.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
I'm fairly sure both of the changed sites are OK as-is, but they're
fragile, so this is just safening them up. Since this is happening
pre-ssa, we don't want to be overwriting values that may potentially get
used later on.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
If we free the bo, then the PTE may get deallocated immediately. We have
to make sure that the submission includes a ref to the old bo so that it
remains mapped for the duration of the command execution.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
intel_batchbuffer_emit_float is dead code, it should go.
intel_batchbuffer_emit_dword only had one user, which had bungled using
them by forgetting to call intel_batchbuffer_require_space first. So it
seems wise to delete these unsafe helpers.
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
intel_batchbuffer_emit_dword doesn't reserve space for the DWord it
emits. In the past, we had some reserved batch space to ensure this
worked. With the switch to growing batches, we need to actually request
space so that we grow if necessary.
Fixes: 2c46a67b41 (i965: Delete BATCH_RESERVED handling.)
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
The enums are moved to the top and indented like the rest of the file.
Comments are added to split up the function aliases by corresponding
extension. This should make no functional difference.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
If we have a color attachment, but its writes are masked, this would
have still returned true. This is inconsistent with how HasWriteableRT
in 3DSTATE_PS_BLEND is set, which does take the mask into account.
This could lead to PixelShaderHasUAV not being set in 3DSTATE_PS_EXTRA
if the fragment shader does use UAVs, meaning the fragment shader may
not be invoked because HasWriteableRT is false. Specifically, this was
seen to occur when the shader also enables early fragment tests: the
fragment shader was not invoked despite passing depth/stencil.
Fix by taking the color write mask into account in this function. This
is consistent with how things are done on i965.
Signed-off-by: Alex Smith <asmith@feralinteractive.com>
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Similar to RadeonSI.
This fixes:
dEQP-VK.image.texel_view_compatible.graphic.basic.attachment_read.bc*r16g16b16a16_sfloat
dEQP-VK.image.extended_usage_bit.attachment_write.r16_sfloat
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
When allocate_user_sgprs() was called, ctx->stage was actually
unset and 0 is for the vertex shader. This doesn't change
anything for now because of the spill support thing.
Though, the number of user SGPRs has to be fixed for merged
shaders on GFX9. It was broken before anyway.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This might decrease VGPR spilling, because we no longer
have to use v4i32 for 2D fetches when level == 0. We now
use v2i32 for those cases.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Using 4, as it is the default value on mesa. See mesa/main/config.h
and the following commit that introduced the value:
15ac66e331
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
ARB_transform_feedback3 sets a minimum of 1, ARB_gpu_shader5 a minimum
of 4. It shouldn't matter too much, so choosing the later.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Used to handle how many ubo you can define on the context. Minimimum
defined as 36 on ARB_uniform_buffer_object spec, up to 84 on OpenGL
4.6 (12 per stage at each moment).
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Every now and then I execute the standalone compiler, get the
non-version error, and need to remember what I'm doing wrong
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Jason Ekstrand
Caio Marcelo de Oliveira Filho
Samuel Pitoiset
Igor Gnatenko
Alex Smith
Florian Will
Karol Herbst
Ilia Mirkin
Kenneth Graunke
Marek Olšák
Neil Roberts
Alejandro Piñeiro