Otherwise, in presence of p_exit_early_if the main FS will always
jump to the PS epilog regardless the exact mask.
This fixes dEQP-VK.draw.renderpass.shader_invocation.helper_invocation
and few vkd3d-proton regressions when PS epilogs are forced.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17617>
Because the data register of atomic VMEM instructions
is shared between src and dst, it might be necessary
to create live-range splits during RA.
Make the live-range splits explicit in WQM mode.
Totals from 7 (0.01% of 134913) affected shaders: (GFX10.3)
Latency: 17209 -> 17210 (+0.01%)
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15347>
When the application uses depth values outside of the [0.0,1.0] range
with VK_EXT_depth_range_unrestricted, or when explicitly disabled.
Otherwise, the driver can clamp to [0.0,1.0] internally for optimal
performance.
From the Vulkan spec "in 28.10.1. Depth Clamping and Range Adjustment":
"If depth clamping is not enabled and zf is not in the range [0, 1]
and either VK_EXT_depth_range_unrestricted is not enabled, or the
depth attachment has a fixed-point format, then zf is undefined
following this step."
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16349>
This feature allows shaders to use pointers to buffers which may
not be bound via descriptor sets. Access to these buffers is done
via global intrinsics.
Because the buffers are not accessed through descriptor sets, any
live buffer flagged with VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT_KHR
can be accessed by any shader using global intrinsics, so the driver
needs to make sure all these buffers are mapped by the kernel when
it submits the job for execution.
We handle this by tracking if any draw call or compute dispatch in
a job uses a pipeline that has any such shaders. If so, the job is
flagged as using buffer device address and the kernel submission
for that job will add all live BOs bound to buffers flagged with the
buffer device address usage flag.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17275>
This adds support for global 64-bit GPU addresses as a pair of
32-bit values. This is useful for platforms with 32-bit GPUs
that want to support VK_KHR_buffer_device_address, which makes
GPU addresses explicitly 64-bit.
With the new format we also add new global intrinsics with 2x32
suffix that consume the new address format.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17275>
If we have 2 pipelines that consume the same push constant data
but where one of them only uses direct access and the other has
indirect access, a draw with the first pipeline would clear the
dirty flag without updating the UBO and by the time we bind and
draw with the second pipeline we won't upload the constants either
because the first draw cleared the dirty flag.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17536>
Since we allocate this ourselves we can immediately add it to the
job at the time we allocate it.
This also fixes a bug we introduced when we implemented inline
uniforms because since that commit, if we had an inline uniform
buffer at index 1 which happend to have indirect access we would
track it in slot 0 instead of slot 1, potentially overwriting
the push constant buffer reference.
Fixes: ea3223e7a4 ('v3dv: implement VK_EXT_inline_uniform_block')
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17536>
We have code in there to allocate various segments of MAX_PUSH_CONSTANTS_SIZE
to handle the case of various draw calls in the same command buffer requiring
different push constants, so we are implicitly expecting it to be larger than
this. In fact, this only works now because when we allocate a BO we are always
at least allocating a full page, so the least we ever allocate is 4096 bytes,
so be explicit about it to avoid confusion.
Also, since we were always mapping MAX_PUSH_CONSTANTS_SIZE and the mapping
always starts at the beginning of the BO, it looks like after the first copy
when the resource offset is not zero, we would be writing outside the mapped
range. Always map the full size of the BO instead to ensure this doesn't
happen.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17536>
We have been always uploading MAX_PUSH_CONSTANTS_SIZE but now that
we track the actual size of the push constant buffer we can use
this instead.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17536>
If the command buffer didn't have any push constants or the meta
operation didn't write any new constants we don't need to restore
the state.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17536>
This dynamic state uses the pipeline information so it needs to be
reemitted whenever the pipeline changes.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17601>
Those fields have confusing names. They hold non dynamic state,
specified at pipeline creation that we copy into the dynamic state of
the command buffer whenever we bind a pipeline. This non dynamic state
might get picked up in the dynamically emitted instructions of the 3D
pipeline because our HW packets are not exactly splitted like the
Vulkan API.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17601>
Even though this doesn't change anything, it's not good to use an
ARRAY_SIZE for one array to iterate over another.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17601>
anv_descriptor_set_layout already has the information we're gather
here.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17601>
This bit mask is already computed in
anv_graphics_pipeline::dynamic_states in anv_graphics_pipeline_init().
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17601>
This implements the "tortoise and the hare" algorithm for detecting
cycles in graphs. We use the caller's iterator as the hare and our own
internal copy as the tortoise. Conveniently, VkBaseOutStructure (and
VkBaseInStructure which is identical except the pointer type on pNext)
have a pointer we can use for the tortoise and an sType which we can use
for a counter to ensure we only increment the tortose every other loop
iteration.
There are more efficient algorithms than tortoise and hare but they
require allocating memory for something like a hash set of seen nodes.
Since this for debug purposes only, it's ok for it to be a bit
inefficient in the case where it hits the assert. In the usual case of
no loops, it's the same runtime efficiency as the unchecked version
except that it does a tiny bit of math and 50% more pointer chases.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17596>
[anholt: changed to make all drivers do the right thing by moving the
payload barycentric check into the compiler]
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17381>
In addition to hopefully generating shorter code, this optimizes out a
comparison of a mediump-cast value in
dEQP-GLES2.functional.shaders.algorithm.rgb_to_hsl_fragment passed
through ANGLE, and allows the test to pass. We believe it to be a
test bug, but emitting better code like apparently everyone else does
is also a fine result.
No change on GLES gfxbench shaders.
Fixes: #6585
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17546>
RADV and PowerVR use the same implementation.
Turnip does use a slightly modified version but the helper only has one
use -> just inline it and get rid of turnip's vk_format.h.
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17515>
dri3_glx.c includes xshmfence and glxcmds.c includes xf86vm, neither of
which are listed as dependencies of the glx lib in the meson.build file.
Consequently, those files would fail to compile on machines that did not
have xshmfence and xf86vm installed globally. This commit rectifies the
issue by adding the missing dependencies
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17585>
GL drivers have an implicit default of "in bounds" for unwritten clipdistance
values, but some (layered) drivers have to deal with api mismatch which
prevents that implicit value from being used after the shader
gets mangled by various compiler passes
to avoid issues here, write out all members of the clipdistance array
with zeroes at the very start of the shader such that any values the
shader actually writes will naturally overwrite the implicit zero and
any unwritten values will now be written as zero
fixes#6845
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17498>
this can get called from multiple threads with the recent llvmpipe
overlapping rendering changes, so make sure to lock around the
map/unmapping so they can't race.
This should fixes some crashes seen with kwin.
Reviewed-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Tested-by: Adam Williamson (Fedora)
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17531>
If the fragment shader exports MRTZ and the epilog some color exports,
DONE/VM should be added to the last export.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17485>
The main fragment shader can only export MRTZ (if present) and the
epilog will export colors.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17485>
The first operand of this new pseudo-instruction is a 64-bit SGPR for
the continue PC, followed by a variable list of fixed VGPRS for the
color exports which are the PS epilog inputs.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17485>
The PS epilog would be a "normal" compiled shader using RA, etc. It
will declare up to 8x4 VGPRs for all color exports.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17485>
The main FS would have to jump to the PC of the PS epilog. Given that
shaders are allocated in the 32-bit addr space, one user SGPR is fine.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17485>
To pass the necessary pipeline information for compiling PS epilogs.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17485>
This is currently disabled but it will be used for testing first,
and then for graphics pipeline libraries.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17485>
When doing 10bit encoding in ffmpeg it uses
VaDeriveImage, and that could result in missing
mapping the chroma buffer of the input frame.
This WA to disallow ffmpeg using VaDeriveImage
function, so that VaCreateImage and VaPutImage can
be used and WA the chroma buffer mapping issue.
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Ruijing Dong <ruijing.dong@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17472>
We're not quite conformant with the extension, because we don't have
a valid conformance version.
That's not a quick-fix, so we should probably just accept some failures
for now.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16998>
This is just a bag of misc properties that we should fill in.
Not all of them are filled out super accurately, but this is the best we
can do for now.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16998>
This should be possible to support, but we don't support minmax blending
at all yet, so let's leave these as unsupported for now.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16998>
Before enabling Vulkan 1.2 support, we need to fix the TODO in here.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16998>
These might not be exactly right, but they are good enough for now.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16998>
We can do better here in the future, but this is what's supported right
now.
Reviewed-by: Jesse Natalie <jenatali@microsoft.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16998>
When using a static callable stack, the required scratch has already
been allocated.
Dynamic stacks are located at the end of scratch memory
and are allocated on demand using radv_set_rt_stack_size.
Static stacks live at the start of scratch memory and are allocated in
create_rt_shader by setting scratch_size.
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-By: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17579>
When culling enabled, it will use LDS space, which overlap with
the prim id export.
Fixes: e97f0463a8 ("ac/nir: Implement NGG deferred attribute culling in NIR.")
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17593>
This function causes a crash with RADV_DEBUG=llvm and this commit
works around that crash.
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17580>
this ensures the alpha component is full if it must be read for fbfetch
fixes (RGBX swapchain config):
KHR-GL46.blend_equation_advanced*
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17366>
these are only ever going to hurt tiler perf, so remove the footgun
this also means there's no more srgb format conversion needed, so delete
all of that too
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17366>
in the scenario where:
* at least 1 color buffer was bound and a depth buffer was bound
* no color clear was enabled
* a zs clear was enabled
* the zs clear was never flushed
* the zs clear needs a renderpass
* the fb state changes
the color buffer(s) would be unbound, following which the depth buffer unbind
would trigger a renderpass, which would utilize the just-unbound color buffers,
which have no batch tracking, thus creating a case where the surface was destroyed
while it was still in use
cc: mesa-stable
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17366>
formats like GL_RGB10_A2UI can be cleared with out of range values,
so to ensure consistent driver behavior, pre-clamp to the valid range
affects:
KHR-GL46.direct_state_access.renderbuffers_storage_multisample
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17366>
it's unknown whether there may be clears to the depth attachment at the start of
a renderpass, so always assume there will be
Fixes: c132a28745 ("zink: use store op NONE when necessary for depth usage")
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17366>
this was really stupid: instead of just binding a new fb and firing off
a clear, the code was calling u_blitter to bind a new fb and do actual
draws
Acked-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17366>
These are reachable, and
dEQP-VK.api.smoke.triangle_ext_structs,Crash is why.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17586>
LLVM-15 enables opaque pointers by default. We temporarilly request
non-opaque pointers while we migrate our code to support non-opaque pointers.
This workaround needs to be removed before LLVM-16.
See #6615
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Acked-by: Kai Wasserbäch <kai@dev.carbon-project.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17514>
Similar motivation as c426e21ff1 ("turnip: Reverse the order of walking
pipes or tiles on odd rows."), but instead we just swap the order of
alternate rows of fd_tile the the gmem stateobj.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17303>
When point smoothing is enabled then this lowering pass will
modifies the alpha component of every write to fragment output.
Anti-aliased points get rounded with respect to their radius instead
of square.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15117>
When poly_line smoothing is enabled then this lowering pass will
modify the alpha component of every write to fragment output
using sample coverage mask.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16245>
You'll get all this and more anyway once you're in NIR. This lets us GC a
bunch more ARB program transformation code.
No effect in shader-db on softpipe.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17528>
This replaces our mesa_remove_output_reads(), which in turn GCs some other
ARB program transformation code.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17528>
We don't need to go grubbing around in the ARB program when we can use the
right variable type at prog_to_nir time. This does leave
fp->system_values_read/inputs_read as they were, but I don't see anywhere
that that matters (the NIR will have its info gathered appropriately, and
other lowering may also cause mismatch between the gl_program and the
NIR).
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17528>
This uses APIs that are not available on Win7. Since this is a build-time
configuration, and since we can't use the SDK version as an indicator
(since you can support Win7 via new SDKs), a new option is added to allow
disabling it, to maintain Win7 support if desired.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17431>
This change adds the tracepoints that can help understand app behavior
for debugging and performance optimization purposes.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17497>
Previously we suballocate only for host visible memory type to reduce
the kvm mem slot usage. That is no longer an issue given the limit has
been raised. However, we should still suballocate to make layering
clients performant. So we just suballocate regardless of mem type.
This change also increases the allowed suballocation size request from
64K to 128K, which makes layering clients happier.
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17497>
binding/unbinding a texture affects the previously specified parameters,
so ensure the driver flag for clamp emulation is also set to perform
updates as needed
Fixes: e8f71f6ac4 ("mesa/st: add PIPE_CAP_GL_CLAMP")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17459>
nir->info.has_transform_feedback_varyings is set for all stages in the
pipeline when xfb is present, so it can't be used for this
harmless, but this is more correct
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17404>
this splits all the members of a struct into separate variables to
improve xfb inlining and reduce the number of locations consumed by
xfb outputs, reducing the chances of running out of shader outputs
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17404>
get_slot_components() returns the total number of output components
for arrays for initial evaluation phase, but during the packed->inlined
conversion the arrayed size must be normalized to the slot's component count
in order to effectively catch and inline the array
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17404>
these are not necessarily the same case even if in glsl they are the same,
and by splitting it out a bunch of redundant array[scalar] code can be deleted
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17404>
We used to do it for every queue, which was duplicate work as pstate is
per-device. It could also cause trouble when multiple hw_ctx are created as
the call will succeed for only one of them and the rest will return -EBUSY.
Simplify and fix this by only setting for the first non-null hw_ctx.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17541>
In the direct path we already skipped draws, but in DGC I noticed
that just emitting these packets can cause issues ...
Cc: mesa-stable
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17269>
Got to put the commandbuffers & uploadbuffers there. With DGC
those can be allocated by the application.
Excluding it from all other buffers/images to avoid using the
precious 32bit address space.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17269>
while some (tg4) sample ops can use different bit sizes in spirv, most
cannot, and all the shader variables are always emitted as 32bit, so
ensure the 32bit type is always what's being used for sampling
cc: mesa-stable
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17427>
It's everyone's favourite day, infrastructure maintenance Friday.
This includes manual disables for a618-vk and zink-anv-tgl, because
apparently the disable-on-variable rules don't carry through to those
jobs for ... some reason.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17553>
When the shader state is destroyed before the async shader compile is
done, we get a use after free in the async thread, as the shader state
it is operating on is gone. Fix this by dropping any pending job from
the async queue or wait for it to finish before destroying the state
by calling util_queue_drop_job.
Also call util_queue_fence_destroy, which would have caught this issue
by asserting that the queue_fence is in signalled state when the
shader state is destroyed.
Fixes: 1141ed5859 ("etnaviv: async shader compile")
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17534>
The two macros introduced here form a (hopefully) unobjectionable
subset of those added in !17203.
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17488>
This type is useful beyond the scope of winsys.
It can now be used without being lumbered with a dependency on
pvr_winsys.h. Since pvr_winsys.h is used by pvr_private.h, this can be
a common cause for circular dependencies during development.
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17488>
It's still 1.17 but the version is changed due to the fixed size
fw struct update.
Signed-off-by: Karmjit Mahil <Karmjit.Mahil@imgtec.com>
Reviewed-by: Rajnesh Kanwal <rajnesh.kanwal@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17491>
Instead of trying to compact the surface state table to get rid of any
unused render targets, emit MAX(1, colorAttachmentCount) surface states
always. This ensures that secondaries will always match with primaries
when we go to do the copy since there's no rule requiring the secondary
to have VK_FORMAT_UNDEFINED when the primary has a NULL image view.
Fixes: 3501a3f9ed ("anv: Convert to 100% dynamic rendering")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17543>
On Linux, the static glapi path sees libGL.so implementing the static
glapi, and the drivers (libgallium_dri.so) updating/reading the TLS
vars.
On Windows, to allow libgallium_wgl.dll to be a full ICD, it's
responsible for implementing the actual static glapi. However, before
this change, OpenGL32.dll was also implementing the static glapi,
meaning that GL API calls from OpenGL32.dll didn't route to the driver
correctly because the TLS vars were never actually set - the driver set
its copy, and OpenGL32.dll read its own copy.
Now, always build a bridge and static version of glapi when not using
shared. The bridge version is linked into OpenGL32.dll, and the static
version is linked into the driver on Windows. GLES only builds with
shared glapi - but after this, shared glapi is not really needed on
Windows for GLES, since the driver has all of the data.
Fixes: f36921ef ("wgl: Refactor drivers to a libgallium_wgl.dll")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6560
Acked-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16713>
The sampler views array needs to be dimensioned by
PIPE_MAX_SHADER_SAMPLER_VIEWS, not PIPE_MAX_SAMPLERS.
This fixes out-of-bounds array writes when using more than 32
textures in a shader.
Also add some assertions to check array indexing elsewhere.
And change loop limits to be based on ARRAY_SIZE().
Signed-off-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17062>
This fixes another VMware test (dx9-stretch-formats-copy-a8r8g8b8-x8r8g8b8).
Signed-off-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17062>
The logicop_enable and independent_blend_enable vars need to always
be assigned, otherwise, once turned on, they could never be disabled.
This fixes a number of failures in VMware's test suite.
Signed-off-by: Brian Paul <brianp@vmware.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17062>
This allows a VMware test to pass. The comments in lp_bld_limits.h
mention SM 3.0 requirements, but we're in the SM 5.0 era...
Signed-off-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17062>
For the linear rendering fast path, we need to know whether the
texcoord argument to tex instructions comes directly from an FS input
(swizzling OK, but no arithmetic, etc). Use the input register info
to fill in the tex_info object.
This is part of a fix for some linear rendering bugs.
Signed-off-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17062>
We were saving the address of the constants[] and nir_constants[]
arrays in the jit structure. But those arrays went out of scope
before use.
This patch moves the constants[] array to the function scope and
consolidates the TGSI/NIR paths.
Signed-off-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17062>
The texture_unit field needs to be set like the sampler_unit field.
Also, add a swizzle initialization and some comments.
Signed-off-by: Brian Paul <brianp@vmare.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17062>
And GET_DADX(), GET_DADY(), GET_PLANES(). This is a bit more
readable, making the expected parameter types explicit.
Signed-off-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17062>
The code did not handle more than 32 textures. We have a test that
exercises 128 textures (views) and crashed w/ memory corruption down
in the llvm code generator because of this.
Signed-off-by: Brian Paul <brianp@vmware.com>
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17062>
To get the size (in bits) of a bitset.
And minor clean-up in __bitset_ffs().
Signed-off-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17062>
Use the existing 'set_state' field. Some code was using 'state'
and other code was using 'set_state'. It didn't really matter
since lp_rast_cmd_arg is a union, but this removes some potential
confusion.
Signed-off-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17062>
Streamout is not L3 coherent so previous writes to the same address
might be pending and overwrite the SO writes later when they get
flushed from L3, even though the SO write happened later in the batch.
v2: Use the right flag (not COUNTER)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6680
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17503>
Unlike image writes, buffer writes may access the same memory in
different ways, which we've seen in the past can cause problems. Use an
incoherent access to force flush/invalidate between accesses to the same
buffer, unless we know the barrier applies to images only.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17193>
Required for indirect(2) ray tracing to work.
Fixes the following tests:
dEQP-VK.ray_tracing_pipeline.trace_rays_indirect2.indirect_*
dEQP-VK.ray_tracing_pipeline.trace_rays_cmds_maintenance_1.indirect2_*
Fixes: 16585664 ("radv: vkCmdTraceRaysIndirect2KHR")
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17316>
Required for indirect(2) ray tracing to work.
Fixes: b30f96dd ("radv,aco: Use ray_launch_size_addr")
Signed-off-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17316>
Certain player has hard coded max_transform_hierarchy_depth_inter and
max_transform_hierarchy_depth_intra values set through VA-API, which
doesn't work on radeon HW. Until properly fixing it on player side,
temporarily adding this workaround to use calculated values instead.
Signed-off-by: Boyuan Zhang <boyuan.zhang@amd.com>
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17489>
the si_pm4_clear_state() initializes the variable in struct si_pm4_state which
anyway gets freed in si_pm4_free_state(). Hence no need to call
si_pm4_clear_state() in si_pm4_free_state().
Signed-off-by: Yogesh Mohan Marimuthu <yogesh.mohanmarimuthu@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17504>
Fixes dEQP-VK.ray_query.advanced.using_wrapper_function.comp.*
An empty struct is causing problems because when passing it as
argument the spirv parser will just drop the argument, considering it
does not hold any data.
v2: update radv CI
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 4c703686db ("spirv: handle ray query intrinsics")
Reviewed-by: Konstantin Seurer <konstantin.seurer@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17420>
Format on buffers is irrelevant and bind flag validation needs to be disabled.
Reviewed-by: Bill Kristiansen <billkris@microsoft.com>
Reviewed-by: Giancarlo Devich <gdevich@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17446>
If the memobj wraps a resource, then we only succeed the
mapping operation if the gallium desc matches the D3D12
resource desc.
If the memobj wraps a heap, then we can place whatever
gallium is describing on the heap.
Reviewed-by: Bill Kristiansen <billkris@microsoft.com>
Reviewed-by: Giancarlo Devich <gdevich@microsoft.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17446>
Win32 memory objects can be imported by name (const void *
that will be interpreted as const wchar_t *)
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Acked-by: Marek Olák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17446>
Remove the previous compile-time early-ZS implementation and replace it with the
decoupled early-ZS implementation. This uses more efficient settings in some
cases (depth/stencil tests always passes or do not write), and fixes the
settings used in another case (alpha-to-coverage enabled with an otherwise
early-ZS shader.)
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Closes: #6206
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17428>
If we know ahead-of-time that depth/stencil testing will always pass, it's
better to use weak_early than force_early. However, if depth/stencil testing
could fail (discarding pixels), we'd rather use force_early. Determine which
case we're in at CSO create time.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17428>
The new early-ZS helpers are pure functions, leaf nodes of the call graph, and
implemented with a different algorithm from the "oracle" table of correct values
for various combinations of states. Further, incorrect settings often still pass
CTS while causing game bugs or inefficiencies. That combination makes the
helpers an excellent candidate for unit tests. Add some.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17428>
Bifrost (and Valhall) separate early-ZS configuration into two fields: when does
the depth/stencil buffer update happen? and when are pixels killed by the
depth/stencil tests? The driver separately configures these to occur early
(before the shader executes) or late (after the ATEST instruction executes at
the end of the shader). Early tests are generally more efficient, but various
combinations of API state and fragment shader properties can require late
updates and/or late kills for correctness. Determining how to configure these
fields is nontrivial.
Our current implementation (on Bifrost) configures these fields at fragment
shader compile time and bakes the settings into the RSD. This is both wrong
(using early testing when late testing is required) and suboptimal (using late
testing when early testing would suffice). We need to defer this configuration
until draw time, when we know rasterizer and Z/S state.
Reclassifying at draw time (as we currently do on Valhall) would be expensive,
especially with the extra terms added in here. To cope, decouple the shader
classification from the draw-time configuration. Since there are only a few bits
of draw state involved, this implementation just calculates all possible states.
Then the draw time classification is just indexing into a lookup table.
The actual algorithm used to classify is written with correctness and clarity in
mind. Unlike the current classification algorithm (which tries to match what the
DDK does, poorly), this algorithm embeds its proofs of correctness.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17428>
In theory this wait is required for correct behaviour of discarded threads with
ATEST. Mesa usually waits before the instruction after ATEST, so this wait will
get optimized out by va_merge_flow, but as our scheduler gets more sophisticated
this could become an issue.
Let's stay on the safe side and insert the recommended wait.
No shader-db changes.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17428>
In theory, ATEST can take any combination of registers for inputs.
Experimentally, however, ATEST requires the coverage mask in R60. This avoids
regressing the following dEQP tests, which write their coverage mask with
pixel-frequency-shading but without writing to the depth/stencil buffer.
dEQP-GLES31.functional.shaders.sample_variables.sample_mask.discard_half_per_pixel.*
This issue is known to affect both Mali-G52 (v7) and Mali-G57 (v9). I am unsure
if this is a silicon bug or just an obscure implementation detail.
No shader-db changes.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17428>
Nos that glsl_to_nir is setting sample_shading_enable whenever FB fetch
is used, we don't need to duplicate it here.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14020>
Now that this information is accurately gathered by spirv_to_nir, we no
longer need the hack. We just need to fix up the way we handle some of
the key bits.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14020>
Thanks to the previous commit, we no longer need this check.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14020>
If framebuffer fetch is used, we have to enable sample shading because
the fetched framebuffer value is per-sample.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14020>
We don't really want to base this on a late nir_gather_info for two
reasons:
1) The Vulkan spec says that if a sample-qualified input, SampleID, or
SamplePosition are in the entry-point's interface, you get
per-sample dispatch. This means we really should gather this
information before dead-code has a chance to delete anything.
2) We want to be able to add nir_intrinsic_load_sample_pos intrinsics
as part of lowering passes without causing per-sample interpolation.
This means nir_gather_info needs to stop gathering it.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14020>
We don't really want to base this on a late nir_gather_info for two
reasons:
1) The GL spec says that any static use of a sample-qualified input,
gl_SampleID, or gl_SamplePosition causes per-sample dispatch. This
means we really should gather this information before dead-code has
a chance to delete anything.
2) We want to be able to add nir_intrinsic_load_sample_pos intrinsics
as part of lowering passes without causing per-sample interpolation.
This means nir_gather_info needs to stop gathering it.
For 1, this doesn't actually get us quite there as GLSL IR may have
deleted something already. However, it does get us closer.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14020>
This lets us drop demote_sample_qualifiers as well as a back-end check
for key->multisample_fbo.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14020>
On Intel, we have to do this because we can't ask for the per-sample
barycentrics without setting the per-sample dispatch bit or the GPU will
hang. However, nothing we're doing in this pass is Intel-specific and
it may be a useful optimization for someone else so we may as well make
it a generic NIR pass. This version actually does a bit more than the
current brw_nir_demote_sample_qualifiers() pass as it also handles
pre-nir_lower_io interp_dref_at* as well as a couple system values which
we can easily constant-fold.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14020>
NIR constructs this information for us as part of nir_gather_info these
days so we can simplify our logic a bit. This will also let us be more
correct once we move uses_sample_shading scraping earlier.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14020>
Pandecode is not thread-safe (for a large number of reasons) and does not even
try to be. This is a problem when tracing (or just using PAN_MESA_DEBUG=sync)
multithreaded applications. The most common symptom of the problem are assertion
failures deep in the red-black tree implementation, which is not thread-safe.
Just protect the whole thing by a "in pandecode?" mutex, since this is not
performance sensitive code and we don't really care about the extra
serialization incurred. As pandecode does not recurse into itself, we may simply
lock at the beginning and unlock at the end of each entrypoint in pandecode,
which is thread-safe regardless of how pandecode is used. A few entrypoints are
refactored to avoid early returns to keep the lock/unlock calls in obvious
visual pairs.
Fixes flakes when running the CL CTS with PAN_MESA_DEBUG=sync like we would in
CI (e.g: events.event_flush)
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Tested-by: Icecream95 <ixn@disroot.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17409>
The physical tile buffer size (and hence the maximum available tilebuffer size)
are implementation-defined. Track this information on the device so we can
correctly select tile sizes, instead of hardcoding the value for Midgard.
Implementation values are pulled from the "Tile bits/pixel" row of the public
Mali data sheet [1]. That row lists the maximum number of bits available for a
pixel given the maximum tile size and pipelining. For currently supported
hardware (v9 and older), that maximum tile size is 16x16. So those values should
be multiplied by (16 * 16 * 2) / 8 to get the physical size in bytes.
This may improve Bifrost/Valhall performance on workloads using multiple render
targets. It also gets us ready for the dazzling array of tile sizes available
with v10.
[1] https://developer.arm.com/documentation/102849/latest/
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17432>
Separate out "calculating the size of each pixel", "selecting a tile size", and
"calculating the colour buffer allocation". Then implement the middle (selecting
a tile size) with a simple constant time expression, rather than a loop. There's
a bit of related clean up in here.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17432>
The if statement we insert would insert a new block before the end block
(and remove the old pre-end-block). If the new block ended up later in
the HT due to its pointer's hash value, you'd emit another copy of the if
statement after the last one. I saw this happen up to 4 times in testing.
The worst case would be if all those additions and removals ended up
reallocating the HT, at which point we might use-after-free.
Fixes inconsistent shader-db results with geometry shaders.
Cc: mesa-stable.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17501>
va_pack asserts a large number of invariants about the instruction being packed.
If one of these fails (due to an invalid instruction), it's helpful to inspect
the failing instruction, as it may not be apparent in a large shader. Pass the
instruction through with all the assertions in va_pack for easier debugging.
Now assertion failures when packing are easier to debug:
Invalid invariant lo.type == hi.type:
= STORE.i32.flow2.wls br5, br2, wls_ptr[1], byte_offset:0
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17224>
When we assert out due to certain invalid encoding, it's helpful to know what
instruction is causing the failure, since it may not be obvious from the
assembly for large shaders. Now we get nice errors when failing:
Invalid opcode:
br0 = VAR_TEX.f32.flow8.store.skip.lod_mode.center , texture_index:0, varying_index:0
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17224>
Apparently, the point coord origin within a batch can change with Gallium
(seemingly even with GLES? where that's impossible at an API level?) but that
doesn't matter if we're not drawing points. This might have to do with internal
Gallium CSOs (e.g. u_blitter). Issue noticed in SuperTuxKart, which was getting
state change flushes. Performance on one level on a Valhall GPU improved from
26fps to 29fps.
Fixes: 3641dfe436 ("panfrost: Flip point coords in hardware")
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17430>
b6a30b72ab ("panfrost: Implement provoking vertices on Valhall") added an
assertion that every draw selects a particular provoking vertex. The intent was
to ensure provoking vertex selection actually happened. Unfortunately, the
assertion is too strong, as the provoking vertex is irrelevant for some (most)
draws. For those, we don't *want* to commit to a particular provoking vertex for
those to avoid flushing.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17430>
driFetchDrawable is only ever called from the MakeCurrent path, which
means it has to handle the case of pre-GLX-1.3 Windows being named as
the drawable. When it finds the drawable in the hash, it increments its
refcount before returning it, so for a GLXWindow it would be 2 on first
return, one from glXCreateWindow and one from glXMakeCurrent. But when
it does not find the drawable and creates one for the naked Window, the
reference count on first return would only be 1. As a result, if this
context was then ever bound to a different drawable, the old Window's
DRI drawable state (like the back buffer) would be destroyed.
Fixes piglit's glx-multi-window-single-context and glx-make-current for
a variety of drivers.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6713
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17479>
There is nothing more TODO here. With softpin, which is available on all
GPUs using texture descriptors, there is no need for the kernel to patch
the descriptor, as the proper GPU virtual address is filled in by userspace.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17448>
There already is a error handling label to free the sampler view
struct and return failure. Consistently use this label to make
error handling more uniform.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17448>
Texture descriptors currently waste a massive ammount of memory, as every
one is allocated via a separate BO. As the allocation granularity of the
kernel is 4KB and the descriptor is only 256B, 93.75% of the allocated
memory is wasted.
Add a simple suballocator for the texture descriptors, to allocate multiple
ones out of a single kernel BO. This isn't perfect, as freed slots in the
suballocated resource are not reused, but worst-case we end up with the
same waste as we had before. This also potentially improves efficiency at
the kernel side, as this reduces the number of BOs needed for the sampler
views in each submit.
As the BO is now used by multiple descriptors, avoid syncing with the GPU
via the cpu_prep/fini calls, as to not introduce stalls between pending
rendering and new descriptors being filled. This is safe, as each
suballocation slot is only used once, so newly filled slots are certainly
not in use by the GPU.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17448>
The dummy texture descriptor and the dummy render target relocs are not ever
changed by a context operation, so we can save some space by moving them to
the screen and potentially share them and the BOs backing them between
multiple contexts.
Also don't hold two pointers to the same BO, one in the reloc and one raw,
but always just use the reloc one.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17448>
libc++ does not yet implement the c++17 monotonic_buffer_resource,
so emulate it by doing normal allocations that are cleaned up
when the resource is destroyed.
v2: - Use C include and version without namespace aligned_alloc
- add include for sstream needed with clang++ (maurossi)
Closes: #6836
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Filip Gawin <filip@gawin.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17452>
With that most piglits for compatibility contexts are
passing, so enable higher compatibility profile support.
v2: fix formatting ahd fallthrough tag (Filip)
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Filip Gawin <filip@gawin.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17484>
Another hot-fix: when a local register is written to, it is
actually unlikely that the value is never used, so just make
sure that this is never done.
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Filip Gawin <filip@gawin.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17484>
On Cayman bank_swizzle[4] is never counted up, so add an
additional condition to make sure the loop is finished
at one point. This is a hot-fix, the logic below should be
improved.
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Filip Gawin <filip@gawin.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17484>
Now that we have separate C types for the different sub-command types,
we can require a pointer to that type to be passed into functions
which expect the current sub-command to be of a specific type.
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Rajnesh Kanwal <rajnesh.kanwal@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17458>
This is a simple optimization to make type-specific uses of struct
pvr_sub_cmd slightly less verbose.
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Rajnesh Kanwal <rajnesh.kanwal@imgtec.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17458>
The compiler has improved significantly since we found this issue
and this is no longer required.
Notice that because we are increasing the number of samplers
supported beyond what we can loop unroll (currently capped at 16),
some piglit tests that test the maximum number of samplers supported
start to fail because they use indirect indexing on a sampler array
and we don't support that (previously the indirect indexing was
removed by loop unrolling). This is a bug in tests which the
GLSL linker detects, failing to compile the shaders.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17509>
Also, remove the FIXME to pre-compute this in images. We only use
this helper from copy/clear operations where we may be working
with a compatible framebuffer format instead of the original image.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17509>
Correct the max thread number for DG2+ platforms according
to below bspec.
Ref: Bspec: 47202
Cc: mesa-stable
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Chuansheng Liu <chuansheng.liu@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17506>
Before, if the ssbo is too large this would always return 0.
Also, this code is easier to optimize, so the common case of offset 0
and pot stride results in one ushr instead of 5+ instructions.
Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Reviewed-by: Jason Ekstrand <jason.ekstrand@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17468>
SIGSEGV is used by Vulkan API trace layers to track user changes in
device memory mapped to user space. Now with drivers such as Zink, GLES
applications are translated into Vulkan API calls and therefore it is
possible to be tracked by Vulkan api trace layers.
Blocking SIGSEGV hinders one of the memory tracking mechanisms used by
such layers.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17273>
With dynamic vertex bindings the vertex format lookups are a lot
more frequent (vs being baked in the pipeline). Add a simple lookup
cache using a dynamic array to keep track of the hw values, and
avoid repeated translation.
This also reduces the memset to just the bitfields since all
the others will be overwritten.
Seen in perf traces gputest gimark with zink on radv.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15846>
Ensure that if we have swizzle on the initial format, that the
component bits are identical with the lowered format.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17385>
For some formats like VK_FORMAT_B5G6R5_UNORM_PACK16, we have no direct
matching HW format. We can support it by swizzling.
We already apply those swizzles for image views. We just forgot to
deal with buffer views.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Cc: mesa-stable
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6235
Reviewed-by: Nanley Chery <nanley.g.chery@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/17385>