| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
| |
Some rasterization code relies (for sse) on the first and third planes
(but not the second for now) being 128bit aligned, and we didn't get that
on 32bit - I mistakenly thought the 64bit number in the struct would get
the thing aligned to 64bit even on 32bit archs.
Stephane Marchesin really figured this out.
Reviewed-by: Jose Fonseca <[email protected]>
CC: <[email protected]>
|
|
|
|
| |
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
|
|
|
|
|
| |
This will allow drivers to make better decisions about texture sharing
for DRI2, DRI3, Wayland, and OpenCL.
v2: add read/write flags, take advantage of __DRI_IMAGE_USE_BACKBUFFER
Reviewed-by: Axel Davy <[email protected]>
|
|
|
|
| |
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
|
|
| |
Cc: [email protected]
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94088
Signed-off-by: Vinson Lee <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
| |
Reviewed-by: Brian Paul <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We no longer need to build any part of Mesa with Windows SDK 7.0.7600 or
MSVC 2008. MSVC 2013 will be the oldest we support.
In practice this means people are now free to declare variables in the
middle of blocks, on the whole Mesa tree.
Care should still be taken with variable length arrays and void pointer
arithmetic.
Reviewed-by: Brian Paul <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
Hella-acked-by: Ian Romanick <[email protected]>
|
|
|
|
| |
Reviewed-by: Marek Olšák <[email protected]>
|
|
|
|
|
|
|
|
|
|
| |
If you're worried about the duplication of some CAPs, we can remove them
later.
v2: add fields for memory eviction stats
Reviewed-by: Ilia Mirkin <[email protected]>
Reviewed-by: Alex Deucher <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Ilia Mirkin <[email protected]>
Reviewed-by: Marek Olšák <[email protected]>
|
|
|
|
|
|
|
|
|
|
| |
This cap indicates whether pipe->create_surface can reinterpret a texture
as a surface with a format of different block width/height (but equal
block size).
v2: fix whitespace
Reviewed-by: Edward O'Callaghan <[email protected]>
|
|
|
|
|
|
|
|
|
| |
This cap indicates that the driver only supports R, RG, RGB and RGBA
formats for PIPE_BUFFER sampler views.
v2: move into "unsupported features" section for nouveau (Ilia Mirkin)
Reviewed-by: Edward O'Callaghan <[email protected]>
|
|
|
|
| |
So it doesn't get out of sync in multiple places.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If the tri is fully inside a scissor edge (or rather, we just use the
bounding box of the tri for the comparison), then we can drop these
additional scissor "planes" early. We do not even need to allocate
space for them in the tri.
The math actually appears to be slightly iffy due to bounding boxes
being rounded, but it doesn't matter in the end.
Those scissor rects are costly - the 4 planes from the scissor are
already more expensive to calculate than the 3 planes from the tri itself,
and it also prevents us from using the specialized raster code for small
tris.
This helps openarena performance by about 8% or so. Of course, it helps
there that while openarena often enables scissoring (and even moves the
scissor rect around) I have not seen a single tri actually hit the
scissor rect, ever.
v2: drop individual scissor edges, and do it earlier, not even allocating
space for them.
v3: help the compiler a bit with simpler code, suggested by Brian.
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
|
|
| |
Just slightly simpler assembly.
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When we switched to 64bit rasterization, we could no longer use straight
aligned loads for loading the plane data. However, what the code actually
does for loading 3 planes, is 12 scalar loads + 9 unpacks, and then there's
another 8 unpacks for the transpose we need (!).
It would be possible to do the (scalar) loads of course already transposed
(at least saving the additional unpacks), however instead just use
(un)aligned vector loads, and recalculate the eo values, which is much less
instructions (note in case of the triangle_32_3_4 case, the eo values are
not even used, making the scalar loads + unpacks for them all the more
pointless).
This drops execution time of the triangle_32_3_4 function considerably,
albeit it doesn't really make a measurable difference (for small tris we're
essentially limited by vertex throughput in any case), for triangle_32_3_16
it's essentially noise (the loop is more costly than the initial code there).
(I'm thinking about just ditching storing the eo values in the plane data,
so could switch back to using aligned planes, however right now they are
still used in the other raster functions dealing with planes with scalar
code. Also not touching the ppc code, might not be that bad there in any
case.)
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
|
|
|
|
|
|
| |
Since the GREMEDY extensions are normally only exposed by the gremedy
debugger (and could possibly trigger debug paths in the app), we don't
expose the extension by default, but instead only with
ST_DEBUG=gremedy.
Signed-off-by: Rob Clark <[email protected]>
Reviewed-by: Ilia Mirkin <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
| |
Doing that is clearly a bug. We can't quite assert as st/mesa may hit this,
but increase at least visibility of it a bit.
(For the non-refcounted objects it would be illegal too, but we can't detect
that unless we'd store the context ourselves. Plus, those don't tend to cause
random crashes at context or object destruction time... So just sampler views,
surfaces and so targets for now.)
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
I removed this mistakenly in 2dbc20e45689e09766552517a74e2270e49817b5. I
actually thought it should not be necessary and a piglit run didn't show
any differences, but this shouldn't have been in there.
draw_prepare_shader_outputs() is in fact dependent on NEW_RASTERIZER.
The new polygon-mode-facing test indeed shows why this is necessary, there's
lots of invalid reads and writes with valgrind (also crashes without
valgrind), because the pre-pipeline vertex size doesn't match the
post-pipeline vertex size (note this won't help much with stages which don't
have the prepare hook which can grow the vertex size, in particular the wide
point stage, but this isn't used by llvmpipe). The test still won't pass, of
course, but it is only usage of uninitialized values now, which is much
less dangerous...
(Albeit I'm pretty sure for i915 it really is not needed anymore as it
doesn't care about the extra outputs and doesn't call
draw_prepare_shader_outputs().)
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
| |
If we have a d24x8 format, there is no stencil. Therefore, we can always
clear these bits too, which means this will be some kind of memset rather
than read-modify-write.
This is good for some 7% increase or so in gears with huge window size -
seems to have a bigger effect if things aren't in caches. Of course, any
real app won't spend nearly as much time comparatively in clearing
depth buffer in the first place, so the speedup will be much lower.
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch fixes a classic "confuse the enemy" bug.
_mm_andnot_si128 (SSE) and vec_andc (VMX) do the same operation, but the
arguments are opposite.
_mm_andnot_si128 performs "r = (~a) & b" while
vec_andc performs "r = a & (~b)"
To make sure this error won't return in another place, I added a wrapper
function, vec_andnot_si128, in u_pwr8.h, which makes the swap inside.
Signed-off-by: Oded Gabbay <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The cleaning up was quite a performance hog (making pipe_resource_reference
the number two in profilers on the vertex path, and 3rd overall, with its
cousin pipe_reference_described not far behind) if there were lots
of tiny draw calls (ipers). Now the reason was really that it was blindly
calling this for all potential shader views (so 32 each for vs and gs) even
though the app never touched a single one which could have been fixed,
however I can't come up with a good reason why we refcount these. We've got
references, of course, in the sampler views, which should be quite sufficient
as we do all vertex and geometry shader execution fully synchronous.
(Calling prepare_shader_sampling for all draw calls even if there were no
changes looks quite suboptimal too, but generally we don't really expect vs/gs
shader sampling to be used much with llvmpipe, and there's even an early exit
if there aren't any views to avoid the "null loop" albeit it's now no longer
always trying to loop through all 32 slots. Maybe improve another time...).
Of course, if we manage to make vertex loads run asynchronously some day,
we need references again, but adding that back would be the least of the
problems...
Also only set LP_NEW_SAMPLER_VIEW for fragment sampler views. Nothing on the
vertex side depends on it (I suppose we'd really wanted a separate flag in
any case).
(Good for a 3% improvement or so in ipers under the right conditions.)
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This was not really a leak per se, but we were referencing the textures for
longer than intended. If textures were set via llvmpipe_set_sampler_views()
(for fs) and then picked up by lp_setup_set_fragment_sampler_views(), they
were referenced in the setup state. However, the only way to unreference them
was by replacing them with another texture, and not when the texture slot
was replaced with a NULL sampler view. (They were then further also referenced
by the scene too which might have additional minor side effects as we limit
the memory size which is allowed to be referenced by a scene in a rather crude
way.) Only setup destruction (at context destruction time) then finally would
get rid of the references.
Fix this by noting the number of textures the last time, and unreference
things if the new view is NULL (avoiding having to unreference things
always up to PIPE_MAX_SHADER_SAMPLER_VIEWS which would also have worked).
Found by code inspection, no test...
v2: rename var
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch adds a new interface to support hardware mipmap generation.
PIPE_CAP_GENERATE_MIPMAP is added to allow a driver to specify
if this new interface is supported; if not supported, the state tracker will
fallback to mipmap generation by rendering/texturing.
v2: add PIPE_CAP_GENERATE_MIPMAP to the disabled section for all drivers
v3: add format to the generate_mipmap interface to allow mipmap generation
using a format other than the resource format
v4: fix return type of trace_context_generate_mipmap()
Reviewed-by: Brian Paul <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
| |
It makes sense to re-use pipe->invalidate_resource for the purpose of
glInvalidateBufferData, but this function is already implemented in vc4
where it doesn't have the expected behavior. So add a capability flag
to indicate that the driver supports the expected behavior.
Reviewed-by: Marek Olšák <[email protected]>
|
|
|
|
| |
some compiler was unhappy.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The trick here is to recognize that in the c + n * dcdx calculations,
not only can the lower FIXED_ORDER bits not change (as the dcdx values
have those all zero) but that this means the sign bit of the calculations
cannot be different as well, that is
sign(c + n*dcdx) == sign((c >> FIXED_ORDER) + n*(dcdx >> FIXED_ORDER)).
That shaves off more than enough bits to never require 64bit masks.
A shifted plane c value could still easily exceed 32 bits, however since we
throw out planes which are trivial accept even before binning (and similarly
don't even get to see tris for which there was a trivial reject plane)) this
is never a problem.
The idea isnt't all that revolutionary, in fact something similar was tried
ages ago (9773722c2b09d5f0615a47cecf4347859474dc56) back when the values were
only 32 bit anyway. I believe now it didn't quite work then because the
adjustment needed for testing trivial reject / partial masks wasn't handled
correctly.
This still keeps the separate 32/64 bit paths for now, as the 32 bit one still
looks minimally simpler (and also because if we'd pass in dcdx/dcdy/eo unscaled
from setup which would be a good reason to ditch the 32 bit path, we'd need to
change the special-purpose rasterization functions for small tris).
This passes piglit triangle-rasterization (-fbo -auto -max_size
-subpixelbits 8) and triangle-rasterization-overdraw (with some hacks
to make it work correctly with large sizes) easily (full piglit as
well of course, but most tests wouldn't use triangles large enough to
be affected, that is tris with a bounding box over 128x128).
The profiler says indeed time spent in rast_tri functions is reduced
substantially, BUT of course only if the tris are large. I measured a 3%
improvement in mesa gloss demo when supersized to twice the screen size...
Reviewed-by: Brian Paul <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
| |
Otherwise some planes we get in rasterization have subpixel precision, others
not. Doesn't matter so far, but will soon. (OpenGL actually supports viewports
with subpixel accuracy, so could even do bounding box calcs with that).
Reviewed-by: Brian Paul <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is quite a few less instructions, albeit still do the 2 64bit muls
with scalar c code (they'd need way more shuffles, plus fixup for the signed
mul so it totally doesn't seem worth it - x86 can do 32x32->64bit signed
scalar muls natively just fine after all (even on 32bit).
(This still doesn't have a very measurable performance impact in reality,
although profiler seems to say time spent in setup indeed has gone down by
10% or so overall. Maybe good for a 3% or so improvement in openarena.)
Reviewed-by: Brian Paul <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Ilia Mirkin <[email protected]>
Reviewed-by: Marek Olšák <[email protected]>
|
|
|
|
|
|
|
| |
v2: document the integer behavior
Reviewed-by: Edward O'Callaghan <[email protected]
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Ilia Mirkin <[email protected]>
Reviewed-by: Marek Olšák <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
| |
The sse path was pretty much disabled for practical purposes because the
largest allowed fb size was 128x128. So, adapt it for 64bit plane calculations.
This is actually not that difficult, though a problem is that we can't do
a signed 32x32->64bit mul, only unsigned, so need to fix that up. Overall,
the code still looks reasonable, though it's not like changes there in
setup really make much of a difference in the end...
Reviewed-by: Jose Fonseca <[email protected]>
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
| |
eo, just like dcdx and dcdy, cannot overflow 32bit.
Store it as unsigned though just in case (it cannot be negative, but
in theory twice as big as dcdx or dcdy so this gives it one more bit).
This doesn't really change anything, albeit it might help minimally on
32bit archs.
Reviewed-by: Jose Fonseca <[email protected]>
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
|
|
|
|
|
| |
Back in the day (before 24678700edaf5bb9da9be93a1367f1a24cfaa471) the values
were not actually in a struct but even then I can't see why we didn't simply
align the values. Especially since it's trivial to do so.
(Not that it actually matters since the code is pretty much unused for now.)
Reviewed-by: Oded Gabbay <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
They can't actually be 0 (as position is there) but should avoid confusion.
This was supposed to have been done by af7ba989fb5a39925a0a1261ed281fe7f48a16cf
but I accidentally pushed an older version of the patch in the end...
Also prettify slightly. And make some notes about the confusing and useless
fs input "map".
Reviewed-by: Jose Fonseca <[email protected]>
Reviewed-by: Brian Paul <[email protected]>
Reviewed-by: Edward O'Callaghan <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
draw emit couldn't care less what the interpolation mode is...
This somehow looked like it would matter, all drivers more or less
dutifully filled that in correctly. But this is only used for emit,
if draw needs to know about interpolation mode (for clipping for instance)
it will get that information from the vs anyway.
softpipe actually used to depend on that interpolation parameter, as it
abused that structure quite a bit but no longer.
Reviewed-by: Brian Paul <[email protected]>
Reviewed-by: Edward O'Callaghan <[email protected]>
|
|
|
|
|
|
|
|
|
| |
It was actually slightly buggy (missing initialization / setup not dependent
on new vs albeit I didn't see issues), but the case of non-existing attributes
is now handled by draw emit code so don't need that anymore.
Reviewed-by: Brian Paul <[email protected]>
Reviewed-by: Edward O'Callaghan <[email protected]>
|
|
|
|
|
|
| |
Found-by: Coccinelle
Signed-off-by: Edward O'Callaghan <[email protected]>
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch converts the SSE-optimized lp_rast_triangle_32_3_16()
to VMX/VSX.
I measured the results on POWER8 machine with 32 cores at 3.4GHz and
16GB of RAM.
FPS/Score
Name Before After Delta
------------------------------------------------
openarena 16.35 16.7 2.14%
xonotic 4.707 4.97 5.57%
glmark2 didn't show a significant (more than 1%) difference.
v2: Make sure code is build only on POWER8 LE machine
Signed-off-by: Oded Gabbay <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch converts the SSE-optimized build_mask_32() and
build_mask_linear_32() to VMX/VSX.
I measured the results on POWER8 machine with 32 cores at 3.4GHz and
16GB of RAM.
FPS/Score
Name Before After Delta
------------------------------------------------
glmark2 (score) 139.8 142.7 2.07%
openarena and xonotic didn't show a significant (more than 1%)
difference.
v2: Make sure code is build only on POWER8 LE machine
Signed-off-by: Oded Gabbay <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch converts the SSE optimization done in do_triangle_ccw to
VMX/VSX.
I measured the results on POWER8 machine with 32 cores at 3.4GHz and
16GB of RAM.
FPS/Score
Name Before After Delta
------------------------------------------------
glmark2 (score) 136.6 139.8 2.34%
openarena 16.14 16.35 1.30%
xonotic 4.655 4.707 1.11%
v2:
- Convert loads to use aligned loads
- Make sure code is build only on POWER8 LE machine
Signed-off-by: Oded Gabbay <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
|
| |
Signed-off-by: Ilia Mirkin <[email protected]>
Reviewed-by: Edward O'Callaghan <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
|
|
|
| |
This allows the state tracker to know that the various draw parameters
are available in vertex shaders.
Signed-off-by: Ilia Mirkin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
ARB_fragment_layer_viewport requires that if a fs reads layer or viewport
index but it wasn't output by gs (or vs with other extensions), then it reads
0. This never worked for llvmpipe, and is surprisingly non-trivial to fix.
The problem is the mechanism to handle non-existing outputs in draw is rather
crude, it will simply redirect them to whatever is at output 0, thus later
stages will just get garbage. So, rather than trying to fix this up (which
looks non-trivial), fix this up in llvmpipe setup by detecting this case there
and output a fixed zero directly.
While here, also optimize the hw vertex layout a bit - previously if the gs
outputted layer (or vp) and the fs read those inputs, we'd add them twice
to the vertex layout, which is unnecessary.
And do some minor cleanup, slots don't require that many bits, there was some
bogus (but harmless) float/int mixup for psize slot too, make the slots all
unsigned (we always put pos at pos zero thus everything else has to be positive
if it exists), and make sure they are properly initialized (layer and vp index
slot were not which looked fishy as they might not have got set back to zero
when changing from a gs which outputs them to one which does not).
This fixes the failures in piglit's arb_fragment_layer_viewport group
(3 each for layer and vp).
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
| |
Use NULL tests of the form `if (ptr)' or `if (!ptr)'.
They do not depend on the definition of the symbol NULL.
Further, they provide the opportunity for the accidental
assignment, are clear and succinct.
Signed-off-by: Edward O'Callaghan <[email protected]>
Signed-off-by: Marek Olšák <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Edward O'Callaghan <[email protected]>
Signed-off-by: Marek Olšák <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Edward O'Callaghan <[email protected]>
Signed-off-by: Marek Olšák <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
d3d10 actually requires using provoking (first) vertex. GL is happy with
any vertex (as long as we say it's undefined in the corresponding queries).
Up to now we actually used vertex 0 for viewport index, and vertex 1 for
layer (for tris), which really didn't make sense (probably a typo). Also,$
since we reorder vertices of clockwise triangle, that actually meant we used
a different vertex depending if the traingle was cw or ccw (still ok by gl).
However, it should be consistent with what draw (clip) does, and using
provoking vertex seems like the sensible choice (draw clip will be fixed
next as it is totally broken there).
While here, also use the correct viewport always even when not needed
in setup (we pass it down to jit fragment shader it might be needed there
for getting correct near/far depth values).
No piglit changes.
Reviewed-by: Brian Paul <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
| |
33339775565154040e0c4ea2e196217dccc08cdf added support for ASTC textures to
gallium. They don't have any helpers hooked up for software decoding, however,
so cannot support them in drivers relying on util code for decoding.
Reviewed-by: Jose Fonseca <[email protected]>
Reviewed-by: Ilia Mirkin <[email protected]>
|