This structure contains two fields, binding and index, that store the
binding in the descriptor set and the index inside the binding.
These structures are defined as uint8_t, but the types in Vulkan
specification are uint32_t, so big values are clamp.
This fixes dEQP-VK.binding_model.shader_access.*.multiple_arbitrary_descriptors.*
v2: use UINT32_MAX for index when having no render targets (Tapani)
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Fixes warnings like
warning: implicit conversion from enumeration type 'enum isl_format' to
different enumeration type 'enum GEN10_SURFACE_FORMAT'
[-Wenum-conversion]
.SourceElementFormat = ISL_FORMAT_R32_UINT,
^~~~~~~~~~~~~~~~~~~
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
The functions we're marking as UNUSED in isl_surface_state.c are used
only when compiling for particular generations.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Fixes warnings like
warning: implicit conversion from enumeration type 'enum isl_format' to
different enumeration type 'enum GEN10_SURFACE_FORMAT'
[-Wenum-conversion]
.SourceElementFormat = ISL_FORMAT_R32_UINT,
^~~~~~~~~~~~~~~~~~~
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Unless you have data, the compiler knows better than you whether a
function should be inlined.
Unlike all other cases in this series, the removal of the inline keyword
from isl_format_has_channel_type actually changes the resulting binary
with gcc-6.3.0:
text data bss dec hex filename
7831116 346384 420648 8598148 833284 i965_dri.so before
7830716 346384 420648 8597748 8330f4 i965_dri.so after
I think this is likely an improvement. No difference in the resulting
binary with clang-4.0.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
The functions we're marking as UNUSED in genX_pipeline.c are used only
when compiling for particular generations.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Fixes warnings like
warning: implicit conversion from enumeration type 'enum isl_format' to
different enumeration type 'enum GEN10_SURFACE_FORMAT'
[-Wenum-conversion]
.SourceElementFormat = ISL_FORMAT_R32_UINT,
^~~~~~~~~~~~~~~~~~~
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Unless you have data, the compiler knows better than you whether a
function should be inlined.
No difference in the resulting binary with gcc-6.3.0 or clang-4.0.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Avoids Clang's warning about the current code:
warning: suggest braces around initialization of subobject
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Clang doesn't realize that 0 and 1 are the only possibilities, a thinks
lots of variables might be uninitialized.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Clang has "-Wno-initializer-overrides", while gcc has
"-Wno-override-init". Quiets a lot of warnings with clang.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
In order to implement VK_KHR_external_fence, we need to back our fences
with something that's shareable. Since the kernel wait interface for
sync objects already supports waiting for multiple fences in one go, it
makes anv_WaitForFences much simpler if we only have one type of fence.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
This is just a refactor, similar to what we did for semaphores, in
preparation for handling VK_KHR_external_fence.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
This commit changes fences to work a bit more like BO semaphores.
Instead of the fence being a batch, it's simply a BO that gets added
to the validation list for the last execbuf call in the QueueSubmit
operation. It's a bit annoying finding the last submit in the execbuf
but this allows us to avoid the dummy execbuf.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
We didn't allow them before because it didn't look like the spec allowed
it. It certainly doesn't make much sense. However, there are CTS tests
that apparently hit this. What the spec actually says is:
"Importing a payload using handle types with copy transference
creates a duplicate copy of the payload at the time of import, but
makes no further reference to it. Fence signaling, waiting, and
resetting operations performed on the target of copy imports must
not affect any other fence or payload."
A SYNC_FD has copy transference but the import may be temporary or
permanent. If you do a permanent import of something with copy
transference, I guess it's supposed to work and end up resetting the
permanent state. In any case, there seems to be no real harm in
allowing it, so why not.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
The PRM SKL-Vol 2b-05.16 says:
"Within a VERTEX_ELEMENT_STATE structure, if a Component Control
field is set to something other than VFCOMP_STORE_SRC, no
higher-numbered Component Control fields may be set to
VFCOMP_STORE_SRC. In other words, only trailing components can be set
to something other than VFCOMP_STORE_SRC."
Since we set the component 1 to VFCOMP_STORE_0 on gen8+, and
VFCOMP_STORE_IID on gen5+, and we are not using components 2 and 3,
let's also set them to VFCOMP_STORE_0.
Signed-off-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Right now, OpenGL uses the GLSL lowering for shared variables and anv
uses NIR to lower them. For a long time, we've done this weird thing
where we do the NIR lowering unconditionally and then add the SLM sizes
from the two together. This works because one of them will always be 0
but it's a bit sketchy. Let's just move the NIR-based lowering into
anv_pipeline and get rid of the sketch.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Render target surfaces always start at binding table index 0.
This is required for us to use headerless FB writes, which we
really want to do. So, we'll never change that.
Given that, it's not necessary to look up a wm_prog_data field
which we already know contains 0. We can drop the dependency in
brw_renderbuffer_surfaces (Gen4-5)...which was already confusingly
missing from gen6_renderbuffer_surfaces.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Probably harmless, but will overwrite errno with a failure status
code. Reported by coverity.
CID 1416600: Argument cannot be negative (NEGATIVE_RETURNS)
Fixes: 5c4e4932e0 (anv: Implement support for exporting semaphores as FENCE_FD)
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
The anv_execbuf_add_bo() call can actually fail in practice, which
should cause the QueueSubmit operation to fail. Reported by Coverity.
CID: 1416606: Unchecked return value (CHECKED_RETURN)
Fixes: 017cdb10cf (anv: Submit a dummy batch when only semaphores are provided.)
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
brw_hw_type_to_reg_type() needs to know only whether the file is
BRW_IMMEDIATE_VALUE or not, which is not a valid file for the
destination. gcc and clang will evaluate __builtin_strcmp() at compile
time, so we can use it to pass a constant file for the destination.
text data bss dec hex filename
7816214 346248 420496 8582958 82f72e i965_dri.so before
7816070 346248 420496 8582814 82f69e i965_dri.so after
Reviewed-by: Scott D Phillips <scott.d.phillips@intel.com>
text data bss dec hex filename
7816886 346248 420496 8583630 82f9ce i965_dri.so before
7816214 346248 420496 8582958 82f72e i965_dri.so after
Reviewed-by: Scott D Phillips <scott.d.phillips@intel.com>
Previously the brw_inst{,_set}_{dst,src0,src1}_reg_type() functions
provided access to the hardware encodings for the register types. We
often mixed these with the logical BRW_REGISTER_TYPE_* enums (which
themselves used to be the hardware format!) with bad results.
With that functionality now available with the hw_ versions (see
previous commit), we now add functions that take the logical
BRW_REGISTER_TYPE_* enums and convert into the hardware format and vice
versa. To do the conversion we also have to provide the file.
Note the asymmetry between the two functions: the new getter reads the
file from the instruction word, and to ensure that is always set the
setter writes both the file and the type.
Reviewed-by: Scott D Phillips <scott.d.phillips@intel.com>
I'm going to encapsulate all of the logic dealing with register types in
this file.
Rename the parameters for the hardware encodings from type -> hw_type at
the same time.
Reviewed-by: Scott D Phillips <scott.d.phillips@intel.com>
After the last patch converted things into enums, I helpfully got a
compiler warning about these missing from the switch statement.
Reviewed-by: Scott D Phillips <scott.d.phillips@intel.com>
The hardware encodings often mean different things depending on whether
the source is an immediate.
Reviewed-by: Scott D Phillips <scott.d.phillips@intel.com>
These vaguely corresponded to the hardware encodings, but that is purely
historical at this point. Reorder them so we stop making things "almost
work" when mixing enums.
The ordering has been closen so that no enum value is the same as a
compatible hardware encoding.
Reviewed-by: Scott D Phillips <scott.d.phillips@intel.com>
UB and B type encodings are the same as UV and VF. Noticed when writing
the following patch.
Reviewed-by: Scott D Phillips <scott.d.phillips@intel.com>
The destination stride must be equivalent to a dword if VF is used.
Also, since the only compaction table entires with "i:vf" have the
destination as "r:f" specifically check that the destination is of type
float.
Reviewed-by: Scott D Phillips <scott.d.phillips@intel.com>
Note that there's no point in testing on G45, since its compaction is
the same as Gen5. Same logic applies to Gen7 variants and low-power
parts.
Reviewed-by: Scott D Phillips <scott.d.phillips@intel.com>
ISL already offers functions to fill out most kinds of SURFACE_STATE,
so why not handle null surfaces too?
Null surfaces are simple, so we can just take the dimensions, rather
than an entirte fill structure.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Vulkan allows you to do a submit whose only job is to wait on and
trigger semaphores. The easiest way for us to support that right
now is to insert a dummy execbuf.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
This patch adds an implementation based on DRM BOs. We don't actually
advertise the extension yet because we want to add a couple more paths
first.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
BLEND_STATE packing was modified to be variable-length in:
9670124e31 genxml: Make BLEND_STATE command support variable length array.
The initial gen10.xml still had the old, fixed-length style
definition for BLEND_STATE. So gen10_upload_blend_state would
overwrite the packed BLEND_STATE_ENTRYs with its own fixed array
of all-zero entries when packing BLEND_STATE. This caused
BLEND_STATE upload to not work at all.
Fixes: aa416f515a ("i965/genxml: Add gen10.xml")
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
v2: move is_aux into if block. (Jason)
Use else block instead of goto (Jason)
v3: Fix up logic for is_aux (Ben)
Fix up size calculations and add FIXME (Ben)
v4 (Jason Ekstrand):
Use the aux_pitch in the image instead of calculating it
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Acked-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Reviewed-by: Chad Versace <chadversary@chromium.org>
Mesa will map user defined vertex input attributes to slots
starting at VERT_ATTRIB_GENERIC0 which gives us room for only 16
slots (up to GL_VERT_ATTRIB_MAX). This sufficient for GL, where
we expose exactly 16 vertex attributes for user defined inputs, but
in Vulkan we can expose up to 28 (which are also mapped from
VERT_ATTRIB_GENERIC0 onwards) so we need to account for this when
we scope the size of the array of attribute workaround flags
that is used during the brw_vertex_workarounds NIR pass. This
prevents out-of-bounds accesses in that array for NIR shaders
that use more than 16 vertex input attributes.
Fixes:
dEQP-VK.pipeline.vertex_input.max_attributes.*
Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Also, silence an obnoxious finishme that started occurring for all
GL applications which use stencil after the i965 ISL conversion.
v2: Check against 3DSTATE_STENCIL_BUFFER's pitch bits when using
separate stencil, and 3DSTATE_DEPTH_BUFFER's bits when using
combined depth-stencil.
Cc: "17.2" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
We were calculating the total height of 2D surfaces by multiplying the
row pitch by the number of slices. This means that we actually request
slightly more space than actually needed since the padding on the last
slice is unnecessary. For tiled surfaces this is not likely to make a
difference. For linear surfaces, on the other hand, this means we may
require additional memory. In particular, this makes the i965 driver
reject EGL imports of buffers which do not have this extra padding.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Cc: "17.2" <mesa-stable@lists.freedesktop.org>
The docs contain a bunch of commentary about the need to pad various
surfaces out to multiples of something or other. However, all of those
requirements are about avoiding GTT errors due to missing pages when the
data port or sampler accesses slightly out-of-bounds. However, because
the kernel already fills all the empty space in our GTT with the scratch
page, we never have to worry about faulting due to OOB reads. There are
two caveats to this:
1) There is some potential for issues with caches here if extra data
ends up in a cache we don't expect due to OOB reads. However,
because we always trash the entire cache whenever we need to move
anything between cache domains, this shouldn't be an issue.
2) There is a potential issue if a surface gets placed at the very top
of the GTT by the kernel. In this case, the hardware could
potentially end up trying to read past the top of the GTT. If it
nicely wraps around at the 48-bit (or 32-bit) boundary, then this
shouldn't be an issue thanks to the scratch page. If it doesn't,
then we need to come up with something to handle it.
Up until some of the GL move to ISL, having the padding code in there
just caused us to harmlessly use a bit more memory in Vulkan. However,
now that we're using ISL sizes to validate external dma-buf images,
these padding requirements are causing us to reject otherwise valid
images due to the size of the BO being too small.
Acked-by: Kenneth Graunke <kenneth@whitecape.org>
Tested-by: Tapani Pälli <tapani.palli@intel.com>
Tested-by: Tomasz Figa <tfiga@chromium.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Cc: "17.2" <mesa-stable@lists.freedesktop.org>
The kernel only cares about whether the object is to be written to or
not, only reduces (reloc.read_domains, reloc.write_domain) down to just
!!reloc.write_domain. When we use NO_RELOC, the kernel doesn't even read
those relocs and instead userspace has to pass that information in the
execobject.flags. We can simplify our reloc api by also removing the
unused read/write domains and only pass the resultant flags.
The caveat to the above are when we need to make the kernel aware that
certain objects need to take into account different work arounds.
Previously, this was done using the magic (INSTRUCTION, INSTRUCTION)
reloc domains. NO_RELOC requires this to be passed in the execobject
flags as well, and now we push that up the callstack.
The API is more compact, more expressive of what happens underneath, but
unfortunately requires more knowledge of the system at the point of use.
Conversely it also means that knowledge is specific and not generally
applied and so not overused.
text data bss dec hex filename
8502991 356912 424944 9284847 8dacef lib/i965_dri.so (before)
8500455 356912 424944 9282311 8da307 lib/i965_dri.so (after)
v2: (by Ken) Rebase.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Commit 0ab04ba979 (anv: Use python to generate ICD json files) changed
the way ICD json files are created.
Remove the old .in files from extra dist, and add the python script.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Fixes build error with anv_extensions.c not found for
libmesa_anv_entrypoints.
Fixes: d62063c "anv: Autogenerate extension query and lookup"
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
If dual object compile fails (as seems to happen with virgl a
fair bit, and does piglit even have any tests for it?), we end up
not restarting the pull params, so we call
vec4_visitor::move_uniform_array_access_to_pull_constant
a second time and it runs over the ends of the alloc.
Fixes: tests/spec/glsl-1.50/execution/geometry/max-input-components.shader_test
running inside virgl on ivybridge.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
This is more lines of code but the python is far easier to read than the
sed expressions we were using before. Also, this allows us to pull the
API version from anv_entrypoints.py so it never gets out-of-sync.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
The VkVersion class is probably overkill but it makes it really easy to
compare versions in a way that's safe without the caller having to think
about patch vs. no patch.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
This way we can use "from anv_extensions import *" in the entrypoint
generator without worrying too much about pollution
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
this change reverts commit 4f695731, we want to be able to build
with -DDEBUG and gen_decoder on Android.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Makes it possible to build Mesa on Android with -DDEBUG with
the next patch that reverts 4f695731.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
As time goes on, extension advertising is going to get more complex.
Today, we either implement an extension or we don't. However, in the
future, whether or not we advertise an extension will depend on kernel
or hardware features. This commit introduces a python codegen framework
that generates the anv_EnumerateFooExtensionProperties functions as well
as a pair of anv_foo_extension_supported functions for querying for the
support of a given extension string. Each extension has an "enable"
predicate that is any valid C expression. For device extensions, the
physical device is available as "device" so the expression could be
something such as "device->has_kernel_feature". For instance
extensions, the only option is VK_USE_PLATFORM defines.
This mechanism also means that we have a single one-line-per-entry table
for all extension declarations instead of the two tables we had in
anv_device.c and the one we had in anv_entrypoints_gen.py. The Python
code is smart and uses the XML to determine whether an extension is an
instance extension or device extension.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
This will allow us to keep everything in one place when it comes to
declaring what extensions are supported.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
The EU limit of 128 GRFs should allow 32 vertex elements of 4 GRFs.
However, the maximum allowed value of "Vertex URB Entry Read Length"
in SIMD8 is 15. And 15 * 8 = 120 gives us a limit of 30 vertex elements.
Because we also need to reserve a vertex buffer to upload
VertexIndex/InstanceIndex and another to upload DrawID when needed,
we can only expose 28.
Cc: "17.2" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
We already have a helper for doing this in BLORP, this just moves the
logic into ISL where we can share it with other components.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
This makes it much easier to edit the template and doesn't really dirty
the python all that much.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Image layouts only let us know that an image *may* be fast-cleared. For
this reason we can end up with redundant resolves. Testing has shown
that such resolves can measurably hurt performance and that predicating
them can avoid the penalty.
v2:
- Introduce additional resolve state management function (Jason Ekstrand).
- Enable easy retrieval of fast clear state fields.
v3: Use more descriptive field enums (Jason)
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
With an earlier patch from this series, resolves are additionally
performed on layout transitions. Remove the now unnecessary implicit
resolves within render passes.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
v2: Expound on comment for the pipe controls (Jason Ekstrand).
v3:
- Cast base_layer to uint64_t to avoid overflow.
- Remove "seems" from the pipe control comment.
- Fix clamp of layer_count (Jason Ekstrand).
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Use the performance warning infrastructure to provide helpful
information when testing applications.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
For readability, bring the assignment of CCS closer to the assignment of
NONE and MCS.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
The lifespan of the fast-clear data will surpass the render pass scope.
We need CCS_D to be enabled in order to invalidate blocks previously
marked as cleared and to sample cleared data correctly.
v2: Avoid refactoring.
v3: Allow CCS_D for subpass resolves.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
The next patch enables the use of CCS_D even when the color attachment
will not be fast-cleared. Catch the gen7 case early to simplify the
changes required.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
We'll be performing a GPU memcpy in more places to copy small amounts of
data. Add an alternate function that thrashes less state.
v2:
- Make a new function (Jason Ekstrand).
- Move the #define into the function.
v3:
- Update the function name (Jason).
- Update comments.
v4: Use an indirect drawing register as TEMP_REG (Jason Ekstrand).
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
v2: Remove ::first_subpass_layout assertion (Jason Ekstrand).
v3: Allow some fast clears in the GENERAL layout.
v4: Remove extra '||' and adjust line break (Jason Ekstrand).
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
v2: Don't pass in the command buffer (Jason Ekstrand).
v3: Remove an incorrect assertion and an if condition for gen7.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
This will be used to load and store clear values from surface state
objects.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
It may technically be possible to enable some sort of fast-clear support
for at least the base slice of a 2D array texture on gen7. However,
it's not documented to work, we've never tried to do it in GL, and we
have no idea what the hardware does if you turn on CCS_D with arrayed
rendering. Let's just play it safe and disallow it for now. If someone
really cares that much about gen7 performance, they can come along and
try to get it working later.
We have a very specific row pitch that we want and we don't want ISL to
be changing it on us so just be explicit about it.
Fixes: a40f043034
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
There is the same constraintg later on as assert in
isl_gen7_choose_image_alignment_el() so catch it earlier in order
to return error instead of crash.
Needed to avoid crashes with piglits on IVB and HSW:
arb_internalformat_query2.image_format_compatibility_type pname checks
arb_internalformat_query2.all internalformat_<x>_type pname checks
arb_internalformat_query2.max dimensions related pname checks
arb_copy_image.arb_copy_image-formats --samples=2/4/6/8
arb_texture_float.multisample-fast-clear gl_arb_texture_float
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
These formats are already allowed by the i965 GL driver, and the
feature seems to work just fine.
There are tests for multisampled rendering in piglit:
tests/spec/ext_framebuffer_multisample which can be patched to
try 16I/32I in addition to GL_RGBA8I.
IvyBridge passed all tests with all sample numbers.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
These formats are already allowed by the i965 GL driver, and the
feature seems to work just fine.
There are tests for multisampled rendering in piglit:
tests/spec/ext_framebuffer_multisample which can be patched to
try GL_RGBA16F/32F/16I/16UI/32I/32UI in addition to GL_RGBA/8I.
IvyBridge passed all tests with all sample numbers and even
with 128-bit formats.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Some hardware, like i965, doesn't support group sizes greater than 32.
In that case, we can reduce the destination size of the ballot
intrinsic, which will simplify our code generation.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
The implementation of ballotARB() will start by zeroing the flags
register. So, a doing something like
if (gl_SubGroupInvocationARB % 2u == 0u) {
... = ballotARB(true);
[...]
} else {
... = ballotARB(true);
[...]
}
(like fs-ballot-if-else.shader_test does) would generate identical MOVs
to the same destination (the flag register!), and we definitely do not
want to pull that out of the control flow.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
The implementations of the ARB_shader_ballot intrinsics will explicitly
read the flag as a source register.
Reviewed-by: Matt Turner <mattst88@gmail.com>
We already had a channel_num system value, which I'm renaming to
subgroup_invocation to match the rest of the new system values.
Note that while ballotARB(true) will return zeros in the high 32-bits on
systems where gl_SubGroupSizeARB <= 32, the gl_SubGroup??MaskARB
variables do not consider whether channels are enabled. See issue (1) of
ARB_shader_ballot.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
The implementations of the ARB_shader_group_vote intrinsics will
explicitly write the flag as the destination register.
Reviewed-by: Matt Turner <mattst88@gmail.com>
I don't expect anyone is going to care about using this in vec4 programs
(vertex/tessellation/geometry on Gen6/7), no one has come up with a good
way to implement it much less test it.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Specifically, constant fold intrinsics from ARB_shader_group_vote, but I
suspect it'll be useful for other things in the future.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>