max_waves is just for shader-db stats for now, but threadsize will
replace the various mechanisms used to determine threadsize across the
different gen's. Calculating these correctly entails adding a bunch of
details about the sizes of various things to ir3. In the future we will
use the guts of the max_waves calculation to inform RA decisions as
well, which is why the max_waves calculation is broken up into register
dependent/independent pieces.
Something should be said about the units of reg_size_vec4. These units
were chosen for two reasons:
1. As said in the comment, it makes some calculations easier.
2. For a4xx/a5xx, where we don't know as much because we haven't done
the same sorts of experiments to probe for the HW configuration, it
corresponds more directly to things that are known. The existing code
switches to the smaller threadsize when r24.x or higher is used,
which translates directly to a reg_size_vec4 of 48. If we chose
different units (e.g. multiplying by wave_granularity and/or
threadsize_base), then to match the same behavior we'd have to set
reg_size_vec4 based on some other parameters that aren't 100% known.
If someone comes along and updates them, they might inadvertantly
break it.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9498>
We want to use the local_size when available to calculate the threadsize
in ir3, and we need it to work with e.g. computerator where we don't
have a nir shader. Add a local_size field and use that in computerator
instead of of a separate structure that's inaccessable to core ir3.
Also set a dummy local_size in the tests to avoid a divide-by-zero.
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9498>
Quoting Jason's commit message (afa8f5892), that also applies here:
"The Vulkan API provides a mechanism for applications to cache their
own shaders and manage on-disk pipeline caching themselves.
Generally, this is what I would recommend to application developers
and I've resisted implementing driver-side transparent caching in the
Vulkan driver for a long time. However, not all applications do this
and, for some use-cases, it's just not practical."
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9403>
Also return the proper Vulkan result for this case, that is somewhat
tricky. Technically Create[Graphics/Compute]Pipeline only allow OOM
errors. So for this case, there is only the alternative of the generic
VK_ERROR_UNKNOWN, even if we known the cause of the error. From spec:
"VK_ERROR_UNKNOWN will be returned by an implementation when an
unexpected error occurs that cannot be attributed to valid behavior
of the application and implementation. Under these conditions, it
may be returned from any command returning a VkResult"
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9403>
Until now we were always doing a two-step cache lookup, as we were
using the NIR shaders to fill up the key to lookup for the compiled
shaders. But since we were already generating the sha1 key with the
original SPIR-V shader (or its internal NIR representation) any info
we were collecting from from NIR is already implicit in the original
shader, so we can avoid using the NIR in most cases.
Because the v3d_key that is used to compile a shader is populated with
data coming directly from the NIR shader or produced during NIR
lowerings, we can't use it directly as part of the pipeline cache
entry. We could split them, but that would be confusing, so we add a
new struct, v3dv_pipeline_key used specifically to search for the
compiled shaders on the pipeline cache. v3d_key would be still used to
compile the shaders.
As we are using the same sha1 key for all compiled shaders in a
pipeline, we can also group all of them in the same cache entry, so we
don't need a lookup for each stage. This also allows to cache pipeline
data shared by all the stages (like the descriptor maps).
While we are here, we also create a single BO to store the assembly
for all the pipeline stages.
Finally, we remove the link to the variant on the pipeline stage
struct, to avoid the confusion of having two links to the same
data. This mostly means that we stop to use the pipeline stage
structures after the pipeline is created, so we can freed them.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9403>
Mostly the same that main mesa gl_shader_stage, but including the
coordinate shader. This would allow to loop over all the available
stages (for example if we need to free them, compute the max spill
size, etc).
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9403>
We stopped to re-use them after pippeline creation long ago, so let's
reduce the size of both structs, and avoid serialize/deserialize for
the variant case.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9403>
If we were able to get a shader variant from the pipeline cache, we
will not have the nir shader available.
Note that this is what we were doing on the driver before the nir io
helpers were available.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9403>
This maps the nir shader data.location to its final
data.driver_location. In general we are using the driver location as
index (like vattr_sizes on the same struct), so having this map is
useful if what we have is the data.location, and we don't have
available the original nir shader.
v2: use memset instead of for loop, and nir_foreach_shader_in_variable
instead of nir_foreach_variable_with_modes (Iago)
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9403>
As we plan to try to get directly the compiled variant from the cache,
it would be possible to not have available the nir shaders, so we add
this info on prog data.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9403>
Right now we were not pre-generating several variants, but we decided
to let this method, just in case we need that idea back. This ended
being a bad idea. Several months have passed without that need, so
having that method just adds confusion. Also, if we need to add a
multiple-variant in the future, perhaps we would need to do it
different, so let's not template in advance.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9403>
We tweak a little some of the individual messages, and add a new
option to dump the stats when the pipeline destroy.
As we are here we also we also tweak the names of the global options
to make it more clear.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9403>
We haven't added the sm/metric info post-GM200. So make sure we don't
report available queries there. Additionally, fix the group retrieval to
take into account that either driver stats or SM stats may not be
available.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9740>
There may be up to 3 pushbuf data's submitted in this function, make
sure that we allocate enough room for these.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9740>
nouveau_pushbuf_data does not check that there are available pushes in
the current batch. So add a space check for this. Avoids crashes with
the HUD and glxspheres.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9740>
Before texture views were supported, this was needed in some cases.
However with texture views (added eons ago), there is now a target
attribute in the template. Use it and remove the explicit argument.
Should help prevent bugs with inconsistencies between the stored
template target and the one that's used to create the view.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9740>
This target is used to determine whether MS scaling factors are to be
uploaded.
Fixes: 1288ac7632 ("nv50: don't try to upload MSAA settings for BUFFER textures")
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9740>
* fsin and fcos require normalization of the input
* bitfield_insert requires an additional shift of the insert value
v2: drop bitfield_insert lowering code, it is already avaibable as
compiler option (Rhys Perry)
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9452>
r600 expect the input values to be normalited by divinding by 2 *PI, so
add an opcode to be able to lower this in nir.
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9452>
Some backends, like r600 support a fused version of int and float compare
against zero and and csel. Adding these opcodes here makes it possible to
optimize this in nir.
v2: Add rules for float compare + csel
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9452>
Some hardware supports a version of find_msb where the bits are counted
starting at the high bit, and this needs some lowering to obtain the
value that is expected by *find_msb
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Reviewed-by: Kristian H. Kristensen <hoegsberg@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9452>
this uses the same mechanics as surface caching, but it
also requires that surfaces keep refs of the framebuffers they're
attached to so that they can invalidate the fb object upon destruction,
as, similar to program objects, the fb objects are "owned" by their attachments
loosely based on patches by Antonio Caggiano
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9665>
pipe_surface_reference uses surface->context, which is not reliable when sharing
surfaces between contexts. since a surface will never be destroyed outside of
zink if its context is dead, forcing surfaces to go directly to the screen object
prevents accessing dead contexts
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9665>
Danylo Piliaiev
Connor Abbott
Alejandro Piñeiro
Mike Blumenkrantz
Dylan Baker
Michel Zou
Ilia Mirkin
Gert Wollny
Rhys Perry
Iago Toral Quiroga