In some cases we can end up trying to add a write dependency on ourself,
which shouldn't trigger a flush.
Avoids an extra couple flushes per from in stk.
Signed-off-by: Rob Clark <robdclark@gmail.com>
ctx->last_fence isn't such a terribly clever idea, if batches can be
flushed out of order. Instead, each batch now holds a fence, which is
created before the batch is flushed (useful for next patch), that later
gets populated after the batch is actually flushed.
Signed-off-by: Rob Clark <robdclark@gmail.com>
In transfer_map(), when we need to flush batches that read from a
resource, we should be holding screen->lock to guard against race
conditions. Somehow deferred flush seems to make this existing
race more obvious.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Since almost all BOs will be in one CL at a time, this cache will almost
always hit except for the first usage of the BO in each CL.
This didn't show up as statistically significant on the minetest trace
(n=340), but if I lop off the throttled lobe of the bimodal distribution,
it very clearly does (0.74731% +/- 0.162093%, n=269).
No significant difference in the minetest replay, but it should reduce
overhead by not requiring that we write quad indices to index buffers that
we repeatedly re-upload (and making the draw packet smaller, as well).
Over the course of the series the actual game seems to be up by 1-2 fps.
Now that there's only one user of it, it's pretty obvious how to avoid
emitting redundant ones. This should save a bunch of kernel validation
overhead.
No statistically sigificant difference on the minetest trace I was looking
at (n=169), but the maximum FPS is up by .3%
Originally there was CL code for handling various relocations back when I
had relocs for the TSDA/TA buffers. Now that the kernel handles those
entirely on its own, I can inline that code into the one place using it.
We failed to take the start into account for how many vertices to draw in
this round, so we would end up decrementing count below 0, which as an
unsigned number meant we would loop until the CLs soon ran out of space.
When I wrote the code I was thinking about how to use the previously
emitted shader state (no index bias baked into the elements) by emitting
up to 65535 and then only re-emitting with bias for the second wround, but
that doesn't work if the start is over 65535. Instead, just delay
emitting shader state until we get into the drawarrays GFXH-515 loop and
always bake the bias in when we're doing the workaround.
They are the same thing, but this is more consistent with the rest of
the project.
Signed-off-by: Dylan Baker <dylanx.c.baker@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
v2: - Make -Dlmsensors=false work
- Simplify auto and true cases
Signed-off-by: Dylan Baker <dylanx.c.baker@intel.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
This just builds on the image support. Evergreen only has ssbo
for fragment and compute no other stages.
v2: handle images and ssbo in the same shader properly (Ilia)
v3: fix RESQ on buffers,
fix missing atom emit
fix first element offset
use R32 format
write separate buffer rat store path.
(from running deqp gles3.1 tests)
Signed-off-by: Dave Airlie <airlied@redhat.com>
On cayman it appears the cmp component is now in Z.
Fixes:
arb_shader_image_load_store-dead-fragments on cayman.
Signed-off-by: Dave Airlie <airlied@redhat.com>
These didn't handle the TGSI at all properly, this fixes
them to use the common path for 64->32 then adds the 32->int
on at the end.
Fixes:
generated_tests/spec/arb_gpu_shader_fp64/execution/conversion/*
Signed-off-by: Dave Airlie <airlied@redhat.com>
This fixes issues seen with certain versions of Unreal Engine 4 editor
and games built with that using GLSL 4.30.
v2: add driinfo_gallium change (Emil Velikov)
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97852
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103801
Acked-by: Andres Gomez <agomez@igalia.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Prepare for two texture handling paths, the descriptor-based
path will be added in a future commit. These are structured
so that the texture implementation handles its own state
emission.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
This needs to be shared between texture_plain and texture_desc.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
This will be shared with the texture descriptor path.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Need this to efficiently emit texture descriptor invalidations.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Track varying component offset of the point size output, as well as
provide the offset of the point coord input.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Update state objects to add new state, and emit function to emit new
state.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
- This core must load shaders from memory (AFAIK)
- Yet another new location for UNIFORMS
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Update context reset for HALTI3..HALTI5, sorting states for the HALTI
version that has them.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
RS align is not necessary and might even be harmful when using the BLT
engine for blitting.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Add an implemenation of key clear_blit functions using the BLT engine
that replaced the RS on GC7000.
Also set level->size correctly for imported resources. This is important
for the BLT resolve-in-place path to work for them.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Prepare for BLT-based blitting path by moving RS-based
blitting to the RS implementation file, making this
self-contained.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Want to be able to emit state from the texture implementation,
and the blitter implementation.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
When the BLT is involved as source or target, add an extra BLT
enable/disable sequence around the sync sequence.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
The blob does this, as DRAW_INSTANCED can replace fully all the other
draw commands. It is also required to handle integer vertex formats.
The other path is only there for compatibility and might go away (or at
least rot to become buggy due to dis-use) in newer hardware.
As a by-effect this changes the behavior for GC3000-, by no longer using
the index offset for DRAW_INDEXED but instead adding it to INDEX_ADDR.
This should make no difference.
Preparation for GC7000 support.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
This is used by HALTI2+ (GC3000+) when drawing with DRAW_INSTANCED.
It is also necessary when switching between integer and floating point
vertex element formats.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Most files in gallium/radeon now include si_pipe.h.
chip_class and family are now here:
sscreen->info.family
sscreen->info.chip_class
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
