| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
Reviewed-by: Michel Dänzer <[email protected]>
|
|
|
|
|
|
| |
The sample index is always in W.
Reviewed-by: Michel Dänzer <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This makes things a bit nicer, and more importantly it fixes an issue
where a "downgraded" array texture (due to view reduced to 1 layer and
addressed with (non-array) samplec instruction) would use the wrong
coord as shadow reference value. (This could also be fixed by passing
target through the sampler interface much the same way as is done for
size queries, might do this eventually anyway.)
And if we'd ever want to support (shadow) cube map arrays, we'd need
5 coords in any case.
v2: fix bugs (texel fetch using wrong layer coord for 1d, shadow tex
using wrong shadow coord for 2d...). Plus need to project the shadow
coord, and just for fun keep projecting the layer coord too.
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Instead of passing s,t,r coordinates pass a coord array - the reason is that
I need to pass more coords (in particular for shadow "coord", future will also
need another one for cube map arrays) so just pass them as an array.
Also, to simplify things, use fixed location for the shadow reference value I
want to get rid of the silly "where is the right coord value" game.
Keep old-style however for aos sampling (which is not going to need shadow
coord, though for cube map arrays it still would need fixing).
(Next patch will pass those through using the new arrangement directly from
sampler interface.)
v2: fix up soa split path (unreachable currently but still...)
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We need to put border color into texture format color space which
essentially means clamping for non-float, normalized formats (not entirely
sure if we're also meant to quantize the float but it's probably ok not to
do it thankfully).
For OpenGL we could do this easily outside generated code due to the
1:1 sampler/texture correspondence but not for d3d10 which is terrible
(as we recalculate a constant over and over again per shader invocation).
Fortunately border color should be rare enough that we don't care THAT much.
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Calling the prepare outputs cleans up the slot assignments
for outputs, unfortunately aapoint and aaline didn't have
code to reset their slots after the initial setup, this
was messing up our slot assignments. The unfilled stage
was just missing the initial assignment of the face slot.
This fixes all of the reported piglit failures.
Signed-off-by: Zack Rusin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
|
|
|
|
| |
This patch adds the level query support to the video decoders
and uses some more reasonable defaults.
v2: (ck) add commit message
Reviewed-by: Christian König <[email protected]>
|
|
|
|
|
|
|
|
|
| |
FSEQ/FSGE/FSLT/FSNE work just the same as SEQ/SGE/SLT/SNE except skip the
select.
And just for consistency use the same appropriate ordered/unordered comparisons
for the old opcodes as well.
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
|
| |
Also while here add a bunch of other forgotten (integer) instructions to
tgsi_util_get_inst_usage_mask() (which isn't used for much except optimizing
away unused input components), though it may still be incomplete.
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Because we must maintain an exec_mask even if there's currently nothing
on the mask stack, we can still have an exec_mask at the end of the program.
Effectively, this mask should be set back to default when returning from main.
Without relying on END/RET opcode (I think it's valid to have neither) it is
actually difficult to do this, as there doesn't seem any reasonable place to
do it, so instead let's just say the exec_mask is invalid outside main (which
it really is effectively).
The problem is that geometry shader called end_primitive outside the shader
(in the epilogue), and as a result used a bogus mask, leading to bugs if we
had to set the (somewhat misnamed) ret_in_main bit anywhere. So just avoid
the mask combining function when called from outside the shader.
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
|
|
| |
The code was quite weird, the second comparison was in fact a complete no-op
and we can also do the comparison with the vector directly instead of scalar,
which should not also be faster but it is way more obvious how that mask
is actually going to look like.
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Instead of reducing masks to 0/1 simply use the mask directly as -1.
Also use some signed comparison instead of unsigned (as far as I understand
these values have to be (very) small and signed means llvm doesn't have to
apply additional logic to do the unsigned comparisons the cpu can't do).
Saves a couple of instructions in some test geometry shader here.
v2: that was a bit to much optimization, don't skip combining the masks...
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
| |
And dump the variant key too (same as vs does).
Just so I can stop wondering why I see the tgsi dump for fs and vs but not
gs...
|
|
|
|
| |
Was meant to match the name used elsewhere, spotted by Anthony.
|
|
|
|
|
|
|
|
|
|
|
| |
My previous attempt at doing so double-failed miserably (minification of
zero still gives one, and even if it would not the value was never written
anyway).
While here also rename the confusingly named int_vec bld as we have int vecs
of different sizes, and rename need_nr_mips (as this also changes out-of-bounds
behavior) to is_sviewinfo too.
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
d3d10 has no notion of distinct array resources neither at the resource nor
sampler view level. However, shader dcl of resources certainly has, and
d3d10 expects resinfo to return the values according to that - in particular
a resource might have been a 1d texture with some array layers, then the
sampler view might have only used 1 layer so it can be accessed both as 1d
or 1d array texture (I think - the former definitely works). resinfo of a
resource decleared as array needs to return number of array layers but
non-array resource needs to return 0 (and not 1). Hence fix this by passing
the target from the shader decl to emit_size_query and use that (in case of
OpenGL the target will come from the instruction itself).
Could probably do the same for actual sampling, though it may not matter there
(as the bogus components will essentially get clamped away), possibly could
wreak havoc though if it REALLY doesn't match (which is of course an error
but still).
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
| |
Specifically, must return 0 for non-existent mip levels (and non-existent
textures which is an unsolved problem) for everything but total mip count.
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
| |
Otherwise gcc might do very unsafe optimizations, spotted by Uros Bizjak.
Hopefully this time it's finally right?
|
|
|
|
|
|
|
|
|
|
|
|
| |
We can't be injecting the primitive id's in the pipeline because
by that time the primitives have already been decomposed. To
properly number the primitives we need to handle the adjacency
primitives by hand. This patch moves the prim id injection into
the original primitive assembler and completely removes the
useless pipeline stage.
Signed-off-by: Zack Rusin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Without reseting the vertex id, with primitives where the same
vertex is used with different primitives (e.g. tri/lines strips)
our vbuf module won't re-emit those vertices with the changed
primitive id. So lets reset the vertex id whenever injecting
new primitive id to make sure that the vertex data is correctly
emitted.
Signed-off-by: Zack Rusin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
|
|
|
|
|
| |
Before inserting new front face and prim id outputs cleanup
the old extra outputs, otherwise our cache will use previous
output slots which will break as soon as outputs of the current
shader don't match the last.
Signed-off-by: Zack Rusin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
| |
Oops. Should fix https://bugs.freedesktop.org/show_bug.cgi?id=67921
|
| |
|
|
|
|
|
|
|
|
| |
d3d10 specifies ordered comparisons for everything but not_equal which is
unordered (http://msdn.microsoft.com/en-us/library/windows/desktop/cc308050.aspx).
OpenGL probably doesn't care.
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
|
| |
This is wrong both for OpenGL and d3d. (In fact clamping is a side effect
of converting to depth format, so this should really do quantization too
at least in d3d10 for the comparisons to be truly correct.)
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
| |
Clearly the returned values need to be per-element if the lod is per element.
Does not actually change behavior yet.
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
| |
For d3d10 and ARB_robust_buffer_access_behavior, we are required to return
0 for out-of-bounds coordinates (for which we can just enable the code already
there was just disabled). Additionally, also need to return 0 for
out-of-bounds mip level and out-of-bounds layer. This changes the logic
so instead of clamping the level/layer, an out-of-bound mask is computed
instead in this case (actual clamping then can be omitted just like with
coordinates, since we set the fetch offset to zero if that happens anyway).
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
While so far this only causes some harmless test failures, there's lots more
cpus with DAZ. All 64bit capable ones can do it (particularly relevant for
AMD cpus as they supported sse3 very very late) but if really necessary we
can check support for that for real with some more magic.
(In fact just about ANY cpu with sse2 can support DAZ, I believe the only
exception are first gen P4 (Willamette) and from those only early steppings
which can't do it it's almost like intel forgot to add it... - a real pity
though docs say you can't just try to set it as they will throw a GPF.)
While this was meant to address https://bugs.freedesktop.org/show_bug.cgi?id=67672
it does not fix it. Most likely the tests need fixing as I don't think
there's any guarantee about denorm handling in the reference math library
functions if the flags aren't set to standard values. Nevertheless enabling
DAZ on all cpus which can do it should be the right thing to do.
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Should be much faster, seems to work in softpipe.
While here (also it's now disabled) fix up the pow factor - the former value
is what is in GL core it is however not actually accurate to fp32 standard
(as it is 1.0/2.4), and if someone would do all the accurate math there's no
reason to waste 8 mantissa bits or so...
v2: use real table generating function instead of just printing the values
(might take a bit longer as it does calculations on some 3+ million floats
but much more descriptive obviously).
Also fix up another inaccurate pow factor (this time in the python code) -
wondering where the couple one bit errors came from :-(.
Reviewed-by: Jose Fonseca <[email protected]>
Reviewed-by: Zack Rusin <[email protected]>
|
|
|
|
| |
I think it's actually not good enough now...
|
|
|
|
|
|
|
|
|
|
| |
Since llvm -3.4svn r187618, TargetOptions doesn't provide
RealignStack, so only enable it with llvm<3.4
This option must now be specified using function attributes, see LLVM
commit r187618
Reviewed-by: Tom Stellard <[email protected]>
|
|
|
|
|
|
|
|
|
| |
unfilled_stage::face_slot is of type int.
Fixes "Unsigned compared against 0" defect reported by Coverity.
Signed-off-by: Vinson Lee <[email protected]>
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
|
|
|
|
|
| |
pp_free_bos dereferences ppq without a null check.
Fixes "Dereference before null check" defect reported by Coverity.
Signed-off-by: Vinson Lee <[email protected]>
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
| |
the issue is that stream output is run before the pipeline, which
means that unless we decompose the primitives before the so
then things crash. we could convert the entire stream output
code into a pipeline stage but it will take a bit, so for now
fix the crashes by simply re-adding the old input assembler
which is run before the SO.
Signed-off-by: Zack Rusin <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
we used to have a face primitive assembler that we ran after if
the gs was missing but we had adjacency primitives in the pipeline,
lets convert it to a pipeline stage, which allows us to use it
to inject outputs (primitive id) into the vertices. it's also
a lot cleaner because the decomposition is already handled for us.
Signed-off-by: Zack Rusin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Inject front face only if the fragment shader uses it and
propagate through all channels because otherwise we'll
need to figure out the exact swizzle that the fs expects and
it's just simpler to make sure all the components within
the front face register are correctly set.
Signed-off-by: Zack Rusin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
| |
We're already in an "if (File == TGSI_FILE_INPUT)" block at that point.
|
|
|
|
|
|
|
|
|
| |
Replace "fulldecl->Semantic.Name/Index" with semName/semIndex.
Simplify if/else logic for TGSI_FILE_OUTPUT code.
Remove old comment.
Fix indentation.
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
clipping would drop the extra outputs because it always
used the number of standard vertex shader outputs, without
geometry shader or extra outputs. The commit makes sure
that clipping with geometry shaders which have more outputs
than the current vertex shader and with extra outputs correctly
propagates the entire vertex.
Signed-off-by: Zack Rusin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Draw module can decompose primitives into wireframe models, which
is a fancy word for 'lines', unfortunately that decomposition means
that we weren't able to preserve the original front-face info which
could be derived from the original primitives (lines don't have a
'face'). To fix it allow draw module to inject a fake face semantic
into outputs from which the backends can figure out the original
frontfacing info of the primitives.
Signed-off-by: Zack Rusin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Draw sometimes injects extra shader outputs (aa points, lines or
front face), unfortunately most of the pipeline and llvm code
didn't handle them at all. It only worked if number of inputs
happened to be bigger or equal to the number of shader outputs
plus the extra injected outputs. In particular when running
the pipeline which depends on the vertex_id in the vertex_header
things were completely broken. The patch adjust the code to
correctly use the total number of shader outputs (the standard
ones plus the injected ones) to make it all stop crashing and
work.
Signed-off-by: Zack Rusin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
|
|
|
|
|
|
| |
Instead of using the magical 4 use the above computed
vertex size. Doesn't change the behavior, just makes the code
a bit cleaner.
Signed-off-by: Zack Rusin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
| |
when dumping shader outputs it's nice to have the integer
values of the outputs, in particular because some values
are integers.
Signed-off-by: Zack Rusin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
| |
Adding code to detect the usage of prim id and front face
semantics in fragment shaders.
Signed-off-by: Zack Rusin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
| |
we forgot to add ucmp to the list of opcodes, so it was never
generated for ureg.
Signed-off-by: Zack Rusin <[email protected]>
Reviewed-by: Roland Scheidegger <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
llvm shifts are undefined for shift counts exceeding (or matching) bit width,
so need to apply a mask for the tgsi shift instructions.
v2: only use mask for the tgsi shift instructions, not for the build shift
helpers. None of the internal callers need this behavior, and while llvm can
optimize away the masking for constants there are legitimate cases where it
might not be able to do so even if we know that shift count must be smaller
than type width (currently all such callers do not use the build shift
helpers).
Reviewed-by: Zack Rusin <[email protected]>
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
| |
c shifts are undefined for shift counts exceeding (or matching) bit width,
so need to apply a mask (on x86 it actually would usually probably work as
shifts do masking on int domain shifts - unless some auto-vectorizer would
come along at last as simd domain does not mask the shift count).
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously we were using truncation, which gives the correct result
only for numbers in [0.5-1.0] range (because there's no mantissa bits
to do any rounding there).
This is frequently hit (and probably only used there) when converting
fragment depth to depth format (d24s8 etc.) or otherwise dealing with
depth format.
v2: as spotted by Jose, get rid of extra type (src_type is already unsigned).
Reviewed-by: Jose Fonseca <[email protected]>
|
|
|
|
| |
Reviewed-by: Roland Scheidegger <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
| |
The variable 'usage' was being used for two different things.
Sometimes for PIPE_USAGE_x and other times for PIPE_TRANSFER_x.
This renames usage to access when we're talking about PIPE_TRANSFER_x
flags. Plus, add a bunch of comments to remind us what's going on.
Also, use unsigned for PIPE_TRANSFER_x bitmask to be consistent with
other places. And add a missing const qualifier.
Reviewed-by: Roland Scheidegger <[email protected]>
|