The introduction of block::reconvergence_point happened at the same time
as scan_clusters.macro so we forgot to set it correctly.
Fixes: 60413e11c2 ("ir3: optimize subgroup operations using brcst.active")
Signed-off-by: Job Noorman <jnoorman@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27463>
Besides switching to "ctx.active_mask" as the condition, this is basically
the same.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/23367>
On windows when external resources are imported
there is no information about them. And in such cases
resource_from_hanlde templ argument is equal NULL.
To support such case on virgl, virgl winsys can now
fill in template for resource, that will be used if
templ=NULL. Additionally helper functions were
added to convert virgl encoded enums to pipe.
Reviewed-by: Feng Jiang <jiangfeng@kylinos.cn>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27308>
While OpenGL requires that VE must be bound,
other mesa frontends, f.e. d3d10umd, can emit draw
without any VE bound. Which led to vctx->vertex_elements
to be null, which lead to null derefence. Add check
for ve not being null to avoid that.
Supported by virglrenderer@b8ac10db
Reviewed-by: Feng Jiang <jiangfeng@kylinos.cn>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27308>
* Cast to uint8_t* before doing pointer arithmetics
* Add zero to initializer list to initialize zeroed structs
* Don't include linux sepcific headers on WIN32
* Don't use build_id when it isn't available
Reviewed-by: Feng Jiang <jiangfeng@kylinos.cn>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27308>
The kernel checks for UVD msgs and if there aren't any gets upset,
so don't submit 0 length on UVD rings either to avoid that.
Cc: mesa-stable
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27186>
The CTS image allocation sometimes doesn't try to allocate a complete
DPB, but the amdgpu kernel module checks for this, so always make
the DPB max sized on uvd instances.
Fixes part of video decode on Fiji/Polaris
Cc: mesa-stable
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27186>
The submit_sub_index field is used by RGP to determine the number of
submits. Previously, it was incorrectly reporting the same number of
submits than command buffers.
Fixes: 88cbe32048 ("radv: add support for RGP queue events")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27439>
r600 can't handle indirect access to kcache 14 and 15, so
if the shader has more than 14 UBOs and there is indirect UBO
access load the values from kcache 14 (and 15) directly and do
a bcsel based on the buffer id to return the right value.
v2: - replace superfluous check with an assert (Triang3l)
- change the lowering pass to work on load_ubo
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10112
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27372>
This only dumps the begin tokens. Tokens are written to a buffer
containing a 12 byte header at the beginning.
We use an intermediate format for the ray history tokens because the RRA
format is very inefficient.
Reviewed-by: Friedrich Vock <friedrich.vock@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25548>
DRM modifiers are a BSD/Linux phenomenon.
We can also remove a bunch of these checks too. No Linux specific
symbol or header is **actually** used, and the DRM modifier is
just represented as uint64_t. But kept the style of the file
as is.
Reviewed-by: Serdar Kocdemir <kocdemir@google.com>
Reviewed-by: Yonggang Luo <luoyonggang@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27425>
The braces resulted in the never demoting the heap type
which resulted in an infinite loop if this become a necessity.
Fixes: eb394f5316
zink: redesign the allocation try loop to test all heaps
Signed-off-by: Gert Wollny <gert.wollny@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27446>
We need to unref the device *after* submit cleanup, so that
fd_submit_sp_destory() can still reference the device.
Fixes: d558cb664a ("freedreno/drm: Submit should hold ref to device")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27445>
Follow the blob and optimize subgroup operation using brcst.active and
getlast when supported.
The transformation consists of two parts. First, a NIR transform
replaces subgroup operations with a sequence of new brcst_active_ir3
intrinsics followed by a new [type]_clusters_ir3 intrinsic (where type
can be reduce, inclusive_scan, or exclusive_scan).
The brcst_active_ir3 intrinsic is lowered directly to a brcst.active
instruction. The other intrinsics get lowered to a new macro
(OPC_SCAN_CLUSTERS_MACRO) which later gets emitted as a loop (using
getlast/getone) that iterates all clusters and produces the requested
scan result.
OPC_SCAN_CLUSTERS_MACRO has a number of optional arguments. First, since
the exclusive scan result is not a natural by-product of the loop but
has to be calculated explicitly, its destination is optional. This is
necessary since adding it unconditionally will produce unused
instructions that won't be DCE'd anymore at this point. Second, when
performing 32b MUL_U reductions (that expand to multiple instructions),
an extra scratch register is necessary.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6387
Signed-off-by: Job Noorman <jnoorman@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26950>
Moves to shared registers keep the value written by the lowest active
fiber. This means that read_first.macro can simply be turned into a mov
and doesn't need a getone block.
Signed-off-by: Job Noorman <jnoorman@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26950>
