if an entire variable is being dumped into an xfb buffer, there's no need
to create an explicit xfb variable to copy the value into, and instead
the xfb attributes can just be set normally on the variable
this doesn't work for geometry shaders because outputs are per-vertex
fixes all KHR-GL46.enhanced_layouts xfb tests
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9271>
running nir_lower_io_arrays_to_elements_no_indirects for only some stages
breaks location-setting for the stages which don't run it when
e.g., dmat2x3 variables are sometimes split across locations and
sometimes jammed into a single location (TCS I'm looking at you)
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9271>
if we get some crazy matrix types in here then we need to ensure that
we accurately unwrap them and copy the components
fixes KHR-GL46.enhanced_layouts.xfb_stride
Fixes: 1b130c42b8 ("zink: implement streamout and xfb handling in ntv")
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9271>
this was added during review, but it was never correct and just crashes
valid cases like streamout from a mat3x4 type
Fixes: b6f8f3a3ba ("zink: fix streamout for clipdistance")
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9271>
This will take effect in future patches when we are able to query the
kernel to set device->vram.size to a non-zero size.
Builds on Sagar's ("anv: Query memory region info") patch, and
re-organizes things as recommended by Lionel (and Jason).
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9324>
Just treat the llc and non-llc paths as separate cases. This will also
help when adding the local memory setup.
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9324>
When tyding up this section in 1e5b09f42f ("spirv: Tidy some repeated
if checks by using a switch statement.") the break got lost. It is
not a real problem because the next case just break, but better to
have it explicitly here instead of a FALLTHROUGH.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9440>
In nouveau's PBO path with GS support and no VS layer export, we got:
intrinsic store_output (ssa_1, ssa_0) (0, 15, 0, 160, 128) /* base=0 */ /* wrmask=xyzw */ /* component=0 */ /* src_type=float32 */ /* location=0 slots=1 */ /* out_pos */
[...]
vec3 32 ssa_4 = mov ssa_3.xxx
intrinsic store_output (ssa_4, ssa_0) (0, 4, 0, 160, 128) /* base=0 */ /* wrmask=z */ /* component=0 */ /* src_type=float32 */ /* location=0 slots=1 *//* out_pos */
The mov's SSA value we would decide we could store directly to the output,
since nothing else used it. However, the store has a writemask, and the
ALU op was stomping over it instead of ANDing with the output decl's
existing writemask.
Fixes: f79f382c81 ("nir_to_tgsi: Store directly to TGSI outputs when possible.")
Closes: #4380
Tested-by: Ilia Mirkin <imirkin@alum.mit.edu>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9376>
This replaces the new_src parameter of nir_ssa_def_rewrite_uses_after()
with an SSA def, and rewrites all the users as needed.
Acked-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9383>
This commit replaces the new_src parameter of nir_ssa_def_rewrite_uses()
with an SSA def, removes nir_ssa_def_rewrite_uses_ssa(), and rewrites
all the users as needed.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Acked-by: Alyssa Rosenzweig <alyssa@collabora.com>
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9383>
This is currently an alias for nir_ssa_def_rewrite_uses but we move all
the instances which used it to write a non-SSA source to the newly named
helper.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Acked-by: Alyssa Rosenzweig <alyssa@collabora.com>
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9383>
While we're here, add __gen_get_batch_address declarations to more files
because we're about to start requiring it on all GFX 12.5+.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9445>
Add logical shift left and right operations support to mi_builder.
v1:
- Add GEN_GEN > 12 check (Jordan Justen)
- Add gen_mi_has_shift function (Jordan Justen)
- Fix commit title (Jordan Justen)
v2 (Jason Ekstrand):
- Add _imm versions of all of them
- Better handle corner-cases in _imm helpers
- Handle the power-of-two limitation for _imm versions
- Add tests
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9445>
The container is moved from before and hence returns size 0. To get the
correct value, the new instruction container must be used instead.
This was flagged by clang-tidy. The fixed call still triggers the
corresponding diagnostic, hence this change silences it by adding a
redundant clear() after move.
Fixes: 7f1b537304 ("aco: add new NOP insertion pass for GFX6-9")
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9432>
This pass needs to run on the last shader in a pipeline writing
gl_Position. In GLES2, that's always the vertex shader, but in ES3.2, it
can be a geometry or tessellation shader. The shared code works the same
in this case, just make the assert more generous.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Reviewed-by: Erico Nunes <nunes.erico@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9444>
Now the texture virtual memory usage is less of a problem,
we can use this workaround permanently.
In the spirit of the API it's certainly not the proper way
of implementing DYNAMIC textures (it seems they are ok
to have hidden copies in driver managed memory, but not have
virtual addressing space reduced), but it makes sense for us,
both performance wise, and to avoid bugs.
Signed-off-by: Axel Davy <davyaxel0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9377>
On 32 bits, virtual memory is sometimes too short for apps.
Textures can hold virtual memory 3 ways:
1) MANAGED textures have a RAM copy of any texture
2) SYSTEMMEM is used to have RAM copy of DEFAULT textures
(to upload them for example)
3) Textures being mapped.
Nine cannot do much for 3). It's up to driver to really unmap textures
when possible on 32 bits to reduce virtual memory usage.
It's not clear whether on Windows anything special is done for
1) and 2). However there is clear indication some efforts have
been done on 3) to really unmap when it makes sense.
My understanding is that other implementations reduce the usage
of 1) by deleting the RAM copy once the texture is uploaded
(Dxvk's behaviour is controlled by evictManagedOnUnlock).
The obvious issue with that approach is whether the texture is
read by the application after some time. In that case,
we have to recreate the RAM backing from the GPU buffer.
And apps DO that. Indeed I found that for example Mass Effect 2
with High Texture mods (one of the crash case fixed by this patch serie),
When the character gets close to an object, a high res texture and replaces
the low res one. The high res one simply has more levels, and the game seems
to optimize reading the high res texture by retrieving the small-resolution
levels from the original low res texture.
In other words during gameplay, the game will randomly read MANAGED textures.
This is expected to be fast as the data is supposed to be in RAM...
Instead of taking that RAM copy eviction approach, this patchset
proposes a different approach: storing in memfd and release the
virtual memory until needed.
Basically instead of using malloc(), we create a memfd file
and map it. When the data doesn't seem to be accessed anymore,
we can unmap the memfd file.
If the data is needed, the memfd file is mapped again.
This trick enables to allocate more than 4GB on 32 bits apps.
The advantage of this approach over the RAM eviction one,
is that the load is much faster and doesn't block the GPU.
Of course we have problems if there's not enough memory to map the
memfd file. But the problem is the same for the RAM eviction approach.
Naturally on 64 bits, we do not use memfd.
Signed-off-by: Axel Davy <davyaxel0@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9377>
We should only pass in a new indirect_info object if we actually set valid
values in it.
Fixes: abe8ef862f "gallium: make pipe_draw_indirect_info * a draw_vbo parameter"
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9425>
