according to spec, this is supposed to handle fragment shader fetch
from previous draw output, not color output readback from previous
color output write
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10706>
this is much easier to read and is going to greatly simplify the eventual
multidraw implementation which will be dropped in
also it allows moving conditionals outside of loops to very slightly improve
drawoverhead performance (with multidraw)
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10701>
instead of manually creating the ivci, there's already a util function for
that which will handle everything
also only set layer info if there are multiple layers
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10703>
Make progType a constructor argument.
Fix defect reported by Coverity Scan.
Uninitialized scalar field (UNINIT_CTOR)
uninit_member: Non-static class member progType is not initialized
in this constructor nor in any functions that it calls.
Suggested-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10562>
Fix defect reported by Coverity Scan.
Uninitialized scalar field (UNINIT_CTOR)
uninit_member: Non-static class member tail is not initialized in
this constructor nor in any functions that it calls.
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10561>
Fix defect reported by Coverity Scan.
Dereference before null check (REVERSE_INULL)
check_after_deref: Null-checking bsp_bo suggests that it may be
null, but it has already been dereferenced on all paths leading
to the check.
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10564>
Fix defect reported by Coverity Scan.
Uninitialized scalar field (UNINIT_CTOR)
uninit_member: Non-static class member tag is not initialized in
this constructor nor in any functions that it calls.
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10563>
Fixes the following error on Ubuntu 18.04 with GCC 7.
src/gallium/drivers/nouveau/codegen/nv50_ir_from_tgsi.cpp:1448:72:
sorry, unimplemented: non-trivial designated initializers not supported
info_out->prop.cp.gmem[gmemSlot++] = {.valid = 1, .slot = i};
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Karol Herbst <kherbst@redhat.com>
Fixes: f451854f3 ("nv50: add remapping of buffers/images into unified space")
Signed-off-by: Ernst Sjöstrand <ernstp@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10668>
Blending in SP_BLEND_CNTL is not a binary flag but the same
mask as in RB_BLEND_CNTL. It is a per-mrt enable bit for blending.
Copied form a6xx, on a5xx it should be have the same since it seems
to have the same structure layout.
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10682>
Blending in SP_BLEND_CNTL is not a binary flag but the same
mask as in RB_BLEND_CNTL. It is a per-mrt enable bit for blending.
Example SP_BLEND_CNTL produced by blob on a630 and different MRT
blendings:
SP_BLEND_CNTL: { UNK8 | 0x6 }
SP_BLEND_CNTL: { ENABLED | UNK8 | 0xe }
(Decoded before this commit)
Fixes mis-rendering with D3D11 game "Spelunky 2".
Signed-off-by: Danylo Piliaiev <dpiliaiev@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10682>
We can do stuff like this:
vec1 32 ssa_207 = fsum3 ssa_209.xxx
In this case, we'd end up not swizzling in get_alu_src, and reading
components out-of-bounds, which LLVM isn't very happy about, and thus
takes punitive actions, in the form of a segfault.
We don't want that, and we already know from the opcode what the
component counts should be here.
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10690>
Hook-up support for native blits. We keep using u_blitter by default
for now.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10548>
In our simplistic model of assigning types to pointer, we treat
return values specially because their "type" can't be known
easily before their first use. Improve the "ret" handling by
removing one from their count when we reassign the type to
something else.
Signed-off-by: Matti Hamalainen <ccr@tnsp.org>
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10648>
As per the suggestion in #4609, implement mode/option -M/--method-only
which only prints call method names, for quick overview of what is
happening in the trace. The same option can be used with both
dump.py and tracediff.sh.
Signed-off-by: Matti Hamalainen <ccr@tnsp.org>
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10648>
Implement better argument parsing/handling in tracediff.sh
so that the options passed to dump.py and sdiff are not required
to be positional.
Signed-off-by: Matti Hamalainen <ccr@tnsp.org>
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10648>
Enabling --left-column option for sdiff inside tracediff.sh we
get output that only shows the differing lines on the right column.
This, when used with the -N/--named option should make many
trace diffs more understandable.
Signed-off-by: Matti Hamalainen <ccr@tnsp.org>
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10648>
As suggested by zmike in issue #4609, raw pointer values in
are problematic for diffing state dumps. In attempt to remedy
this, we implement an option --named/-N in dump.py that
generates symbolic names for all pointers based on the context.
Signed-off-by: Matti Hamalainen <ccr@tnsp.org>
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10648>
Previously some variants (such as execution time and call number
were suppressed in tracediff.sh via a sed script. It makes sense
to implement an option to leave out such variants to begin with
in dump.py, so let's do so and use it.
Signed-off-by: Matti Hamalainen <ccr@tnsp.org>
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10648>
Previously we inherited some options from parse.py, but
that made no sense for some of the options that are not
needed for dump_state.py (such as --plain, as we output
only JSON format text.)
So, remove the inherit and implement filename argument
here independantly.
Signed-off-by: Matti Hamalainen <ccr@tnsp.org>
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10648>
Clean up tracediff.sh shellscript slightly, partially in preparation
for upcoming changes. Also add a signal handler for removing the
created temporary directory and files if we are interrupted.
Also, use the new --plain option of dump.py to suppress output
of ANSI color/formatting codes, so we do not have to post-filter
them later.
Signed-off-by: Matti Hamalainen <ccr@tnsp.org>
Acked-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10648>
Disabling it that made us start to fail register allocation for a
few tests. Re-enable it until we figure out what is causing that.
Fixes: ca9e0871fb ('v3d: enable NIR loop unrolling')
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10684>
We have to unwind the lod into the scalar path correctly.
Fixes a crash with renderdoc demo
Fixes: e168d148d7 ("gallivm/nir: handle non-uniform texture offsets")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10676>
