| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
readpixels_can_use_memcpy will later call _mesa_format_matches_format_and_type
which does much tighter checks than these to decide if we can use
memcpy for readpixels.
Also, the checks do not seem to be extensive enough anyway, since we are
checking for signed/unsigned conversion only when the framebuffer has integers,
but the same checks could be done for other types anyway, since as long as
there is a signed/unsigned conversion we can't memcpy.
No regressions observed on i965/llvmpipe.
Reviewed-by: Anuj Phogat <[email protected]>
Reviewed-by: Marek Olšák <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
From Muchnick's Advanced Compiler Design and Implementation:
"To determine which variables are live at each point in a flowgraph, we
perform a backward data-flow analysis"
Previously, we were walking the blocks forwards and updating the livein and
then the liveout. However, the livein calculation depends on the liveout
and the liveout depends on the successor blocks. The net result is that it
takes one full iteration to go from liveout to livein and then another
full iteration to propagate to the predecessors. This works out to an
O(n^2) computation where n is the number of blocks. If we run things in
the other order, it's O(nl) where l is the maximum loop depth which is
practically bounded by 3.
In b2c6ba0c4b21391dc35018e1c8c4f7f7d8952bea, we made this same change in
the FS backend to great effect. Might as well keep it consistent and make
the same change for vec4. Also, this took the time to run the test:
ES31-CTS.arrays_of_arrays.InteractionFunctionCalls1
from 6:49.62 to 3:31.40 on Timothy Arceri's machine.
Reviewed-by: Matt Turner <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
gen8 had some special restrictions which don't seem to carry over to gen9.
Quoting the spec for SKL:
"The Z_Height and Z_Width values must equal those present in
3DSTATE_DEPTH_BUFFER incremented by one."
This fixes nothing in piglit (and regresses nothing).
Signed-off-by: Ben Widawsky <[email protected]>
Reviewed-by: Anuj Phogat <[email protected]>
Reviewed-by: Jordan Justen <[email protected]>
|
|
|
|
| |
Trivial. Picked from my tessellation branch.
|
|
|
|
|
|
|
|
|
| |
This is always 0 - only brw_workaround_depthstencil_alignment ever sets
it, and that doesn't run on Gen6+. My initial Broadwell depth state
commit had this mistake.
Signed-off-by: Kenneth Graunke <[email protected]>
Reviewed-by: Anuj Phogat <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We already recognize min(max(a, 0.0), 1.0) as a saturate, but neglected
this variant (which is also handled by the GLSL IR pass).
shader-db results on Broadwell:
total instructions in shared programs: 7363046 -> 7362788 (-0.00%)
instructions in affected programs: 11928 -> 11670 (-2.16%)
helped: 64
HURT: 0
Signed-off-by: Kenneth Graunke <[email protected]>
Reviewed-by: Iago Toral Quiroga <[email protected]>
|
|
|
|
|
|
|
|
|
|
| |
PIPE_MAX_SHADER_INPUTS was recently bumped to 80 because of tessellation.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91099
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91101
Reviewed-by: Brian Paul <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The thread counts and URB information are all speculative numbers that were
based on some CHV numbers at the time.
v2:
Originally this patch had PCI IDs. I've moved that to a new patch at the end of
the series.
Remove is_cherryview hack.
Add PCI ids. These match the ones defined in the kernel. The only one tested by
us is 0x0a84.
Capitalize the hex string (Mark)
Signed-off-by: Ben Widawsky <[email protected]>
Tested-by: "Lecluse, Philippe" <[email protected]>
Reviewed-by: Mark Janes <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Commit b765119c changed the default value of all the counter bits to
64. However, older hardware only has 32 counter bits.
This has only been build-tested. We don't have any tests that verify
the advertised value against implementation behavior, so I don't know
what additional testing could be done.
NOTE: It appears that many Gallium drivers (at least r300 and i915g)
have the same problem, but I don't see a way for the state-tracker to
determine the counter size. Marek says, "For Gallium, a new PIPE_CAP or
new get_xxx_param function will be needed."
Signed-off-by: Ian Romanick <[email protected]>
Reviewed-by: Marek Olšák <[email protected]>
Cc: Alex Deucher <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
From Muchnick's Advanced Compiler Design and Implementation:
"To determine which variables are live at each point in a flowgraph, we
perform a backward data-flow analysis"
Previously, we were walking the blocks forwards and updating the livein and
then the liveout. However, the livein calculation depends on the liveout
and the liveout depends on the successor blocks. The net result is that it
takes one full iteration to go from liveout to livein and then another
full iteration to propagate to the predecessors. This works out to an
O(n^2) computation where n is the number of blocks. If we run things in
the other order, it's O(nl) where l is the maximum loop depth which is
practically bounded by 3.
On my HSW desktop, one particular shadertoy test gets a 20% improvement in
compile times:
N Min Max Median Avg Stddev
x 10 15.965 16.884 16.026 16.1822 0.34736846
+ 10 12.813 13.052 12.876 12.8891 0.06913666
Difference at 95.0% confidence
-3.2931 +/- 0.235316
-20.3501% +/- 1.45417%
(Student's t, pooled s = 0.250444)
Reviewed-by: Matt Turner <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
| |
This is based on Kenneth's patch to delete 'most of the IR'. Due to
linker changes to clone variables, we can now free all of IR.
Saves 58MB of memory when replaying a Dota 2 trace on Broadwell.
Signed-off-by: Tapani Pälli <[email protected]>
Reviewed-by: Kenneth Graunke <[email protected]>
Cc: [email protected]
|
|
|
|
|
|
|
|
|
|
|
| |
This increases memory pressure during linking but makes it easier
for backend to free IR after it is not needed anymore.
v2: use resource list as ralloc context in case of relink (Kenneth)
Signed-off-by: Tapani Pälli <[email protected]>
Reviewed-by: Kenneth Graunke <[email protected]>
Cc: [email protected]
|
|
|
|
|
| |
Signed-off-by: Chris Wilson <[email protected]>
Reviewed-by: Kenneth Graunke <[email protected]>
|
|
|
|
|
|
|
|
|
| |
Start trimming the fat from intel_batchbuffer.c. First by moving the set
of routines for emitting PIPE_CONTROLS (along with the lore concerning
hardware workarounds) to a separate brw_pipe_control.c
Signed-off-by: Chris Wilson <[email protected]>
Reviewed-by: Kenneth Graunke <[email protected]>
|
|
|
|
|
|
|
| |
Suggested by Jason Ekstrand.
Signed-off-by: Kenneth Graunke <[email protected]>
Reviewed-by: Samuel Iglesias Gonsálvez <[email protected]>
|
|
|
|
| |
Trivial.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This caused us to always free the pipe_surface for the renderbuffer.
The subsequent call to st_update_renderbuffer_surface() would typically
just recreate it. Remove the call to pipe_surface_release() and let
st_update_renderbuffer_surface() take care of freeing the old surface
if it needs to be replaced (because of change to mipmap level, etc).
This can save quite a few calls to pipe_context::create_surface() and
surface_destroy().
Reviewed-by: Marek Olšák <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
| |
Signed-off-by: Emil Velikov <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fixes the following build issue, when building without libudev.
CCLD libGL.la
./.libs/libglx.a(dri2_glx.o): In function `dri2CreateScreen':
src/glx/dri2_glx.c:1186: undefined reference to `loader_open_device'
collect2: ld returned 1 exit status
CCLD libEGL.la
Undefined symbols for architecture x86_64:
"_loader_open_device", referenced from:
_dri2_initialize_x11_dri2 in libegl_dri2.a(platform_x11.o)
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91077
Signed-off-by: Julien Isorce <[email protected]>
Reviewed-by: Emil Velikov <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
| |
1000 ms is an extreme value for typical interactive loads. A large
cache has some disadvantages. Search for reusable BOs can take a long
time and memory might get exhausted.
Let's be rather conservative and use half of the old value,
500ms. This is beneficial to some loads on my test system and there
are no regressions.
Reviewed-by: Marek Olšák <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is the basic granularity for BO allocations. The alignment also
helps with BO reuse by the cached bufmgr.
This results in a huge 45% speedup in Metro 2033 Redux on my test
system. The game relies on buffer orphaning with very small buffers
(hundreds of bytes in size) and that did not work efficiently
before. This change may also affect other applications and games.
Reviewed-by: Marek Olšák <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Tapani Pälli <[email protected]>
Reviewed-by: Timothy Arceri <[email protected]>
|
|
|
|
| |
Trivial.
|
|
|
|
|
|
|
| |
Matt, Jason, and I haven't found this useful in a long time.
Signed-off-by: Kenneth Graunke <[email protected]>
Reviewed-by: Matt Turner <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
But only when doing so is safe according to the
RADEON_INFO_VA_UNMAP_WORKING kernel query.
This avoids kernel GPU VM address range conflicts when the BO has other
references than the GEM handle being closed, e.g. when the BO is shared.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90537
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90873
Cc: "10.5 10.6" <[email protected]>
Reviewed-by: Christian König <[email protected]>
|
| |
|
| |
|
|
|
|
|
|
| |
It returns a new value for each sample in the TLB. We've already avoided
trying to get the same index's color multiple times at the vc4_program.c
level, so we're not losing anything by doing this.
|
|
|
|
| |
We've done so for all the other QIR instruction generation in this file.
|
|
|
|
|
| |
It's fairly separate from the rest of the TLB operations at frag end time,
and we'll need to run it multiple times to support MSAA blending.
|
|
|
|
|
| |
It's the same value for loads and stores, because they're basically the
same packet.
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
Noticed while trying to add GL_ARB_texture_multisample support to vc4.
Reviewed-by: Ilia Mirkin <[email protected]>
|
|
|
|
| |
Reviewed-by: Jordan Justen <[email protected]>
|
|
|
|
|
|
|
| |
As of this commit, nothing actually needs the brw_context.
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Chris Forbes <[email protected]>
|
|
|
|
|
|
|
| |
This way we can stop doing is_gles3 checks inside of the compiler.
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Chris Forbes <[email protected]>
|
|
|
|
|
| |
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Chris Forbes <[email protected]>
|
|
|
|
|
|
|
|
|
| |
Previously, these were pulled out of the GL context conditionally based on
whether we were running ff/ARB or a GLSL program. Now, we just pass them
in so that the visitor doesn't have to grab them itself.
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Chris Forbes <[email protected]>
|
|
|
|
|
|
|
| |
Previously, we were pulling it from brw->do_rep_send
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Chris Forbes <[email protected]>
|
|
|
|
|
| |
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Chris Forbes <[email protected]>
|
|
|
|
|
|
|
|
|
|
| |
Previously, each shader took 3 shader time indices which were potentially
at arbirary points in the shader time buffer. Now, each shader gets a
single index which refers to 3 consecutive locations in the buffer. This
simplifies some of the logic at the cost of having a magic 3 a few places.
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Chris Forbes <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This creates the options at screen cration time and then we just copy them
into the context at context creation time. We also move is_scalar to the
brw_compiler structure.
We also end up manually setting some values that the core would have set by
default for us. Fortunately, there are only two non-zero shader compiler
option defaults that we aren't overriding anyway so this isn't a big deal.
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Chris Forbes <[email protected]>
|
|
|
|
|
| |
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Chris Forbes <[email protected]>
|
|
|
|
|
|
|
| |
While we're at it, we'll drop the note about 10-20% performance loss.
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Chris Forbes <[email protected]>
|
|
|
|
|
|
|
| |
We never used the fact that it was variadic anyway.
Reviewed-by: Chris Forbes <[email protected]>
Reviewed-by: Kenneth Graunke <[email protected]>
|
|
|
|
|
|
| |
v2: Do bufmgr set_debug and set_aub_dump at screen time as well.
Reviewed-by: Kenneth Graunke <[email protected]>
|
|
|
|
|
| |
Reviewed-by: Chris Forbes <[email protected]>
Reviewed-by: Kenneth Graunke <[email protected]>
|
|
|
|
|
|
|
|
| |
v2 (Ken): Make shader_debug_log a printf-like function.
v3 (Jason): Add a void * to pass the brw_context through
Reviewed-by: Chris Forbes <[email protected]>
Reviewed-by: Kenneth Graunke <[email protected]>
|