This was something that came up in the slop MR. Not sure it's actually a
good idea or not but kind of curious what people think, given we have a
sound tool (Coccinelle) to do the transform. Saves a redundant branch
but means extra noninlined function calls.. likely no actual perf impact
but saves some code.
Via Coccinelle patches:
@@
expression ptr;
@@
-if (ptr) {
-free(ptr);
-}
+free(ptr);
@@
expression ptr;
@@
-if (ptr) {
-FREE(ptr);
-}
+FREE(ptr);
@@
expression ptr;
@@
-if (ptr) {
-ralloc_free(ptr);
-}
+ralloc_free(ptr);
@@
expression ptr;
@@
-if (ptr != NULL) {
-free(ptr);
-}
-
+free(ptr);
@@
expression ptr;
@@
-if (ptr != NULL) {
-FREE(ptr);
-}
-
+FREE(ptr);
@@
expression ptr;
@@
-if (ptr != NULL) {
-ralloc_free(ptr);
-}
-
+ralloc_free(ptr);
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com> [v3d]
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org> [venus]
Reviewed-by: Frank Binns <frank.binns@imgtec.com> [powervr]
Reviewed-by: Janne Grunau <j@jannau.net> [asahi]
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com> [radv]
Reviewed-by: Job Noorman <jnoorman@igalia.com> [ir3]
Acked-by: Marek Olšák <maraeo@gmail.com>
Acked-by: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Acked-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Job Noorman <jnoorman@igalia.com>
Acked-by: Yonggang Luo <luoyonggang@gmail.com>
Acked-by: Christian Gmeiner <cgmeiner@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37892>
Here we switch to using a sorted arrays for the binary search which
significantly speeds up the lookups. For a shader with ~8000
uniforms its up to 10x faster and the godot-tps-gles3-high.trace
in issue #13894 returns to its original runtime length before we
switched to using range remap in e052254066
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37754>
If the ptr value is NULL and we match an existing entry during an
insert call we now just return the existing value. This allows us
to drop an extra lookup, this will become important as the
following patches change `util_range_remap()` lookups to use a
sorted array that is not created until after we have added all
entries to our linked list.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37754>
WSL version 1 has a known limitation that the kernel does not support
the memfd_create() syscall, and it always returns -1 (see
https://github.com/microsoft/WSL/issues/3542). This results in
lavapipe/llvmpipe failing to create the anonymous file it needs for
allocations from the pipe_screen object, which in turn results in
results in failed application calls to Vulkan APIs (in the case I
observed, vkMapMemory returned an invalid (0xffffffff) pointer along
with VK_SUCCESS, leading to the application segfaulting).
This issue can be reproduced by simply running `vkcube` (or, presumably
any other simple Vulkan application) inside WSL 1, e.g.:
xvfb-run -s '-screen 0 1024x768x24' vkcube --c 300
This patch addresses the issue with several changes:
1. llvmpipe_create_screen() now checks for errors from the
os_create_anonymous_file function and errors out early, making it
easier to track down issues similar to this. Previously, the invalid
fd of -1 would be stored in the pipe_screen struct and the problems
would only appear later.
2. os_create_anonymous_file() now attempts to handle the case where
memfd_create() fails, by falling back to creating a file in a
temporary directory. This fallback is the same as is already done for
builds that don't have memfd_create available at build time - note
that the difference here is that this is a _runtime_ fallback, as is
needed for the case of WSL 1.
3. The fallback logic previously relied on the XDG_RUNTIME_DIR
environment variable, which might not be set when running inside WSL
because there is unlikely to be a desktop environment configured.
This patch adds a fallback of creating a new directory inside /tmp in
this case.
Reviewed-by: Lucas Fryzek <lfryzek@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37679>
Xwayland seems to mix implicit and explicit sync, depending on client
app. This trips up the heuristic that disables implicit sync once it
starts seeing app using explicit sync. This is not typical behavior,
so add a driconf override to disable the heuristic.
Fixes: 137cd3b0fa ("freedreno/drm: Move no_implicit_sync accounting")
Cc: mesa-stable
Signed-off-by: Rob Clark <rob.clark@oss.qualcomm.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37703>
Implement retrieving color buffer info via gralloc0. This is a simpler
alternative to imapper4/5 which requires a dependency on libui that
would require a heavy effort to import headers and stub to be able to
build out of tree.
Since VNDK no longer releases headers since API Level 35 and they are
now only auto-generated, copy over the neccessary defines.
Signed-off-by: Juston Li <justonli@google.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37185>
Many instruction sets (Intel, Apple, NVIDIA) implement bitwise operations with a
single general instruction that takes a lookup table in sum-of-products
(minterms) form. Working with these tables manually is a bit gnarly, and
multiple backends need to do this. This adds common code for representing such
lookup tables with 2- or 3-sources, with a rich set of helpers for building,
inspecting, and manipulating the LUTs.
Eventually, we may want to introduce a nir_op_bitop3 instruction to NIR to build
common code for fusing boolean expression trees into lookup tables. That NIR
pass will presumably use the helpers here.
NAK already has this abstraction internally (in Rust). Possibly NAK will be able
to drop (or cut down) that code once we handle this in NIR.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/37200>
This is helpful for the driver to have the option to provide a custom
threshold for the PQ test performed by the graph coloring algorithm.
A threshold lower than the physical number of registers is helpful on
platforms where the number of registers used can impose a limit on the
thread parallelism of the program. In such platforms even though a
passing PQ test guarantees that the node can be pushed onto the stack
and neglected while coloring the remaining nodes, the ordering in
which this happen can have a dramatic effect in the register pressure
of the resulting shader and therefore also on the thread parallelism
of the program.
Setting a P value threshold lower than the real P value will cause
nodes with Q value above the threshold to use the existing optimistic
coloring heuristic that takes the effort of ordering nodes in the
stack by Q value, in order to do a better job at minimizing the total
register requirement of the program. Even though this causes us to
hit the optimistic codepaths for trivially colorable nodes the
interference graph is still guaranteed to be trivially colorable if it
was trivially colorable without the override.
The use of a threshold lower than the real P value will come at a
compile-time performance cost, the specific trade-off between
compile-time and run-time can be adjusted by the driver based on the
number of registers available to each thread without causing a hit to
thread parallelism.
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/36618>
José Roberto de Souza