Courtesy of clang:
src/gallium/auxiliary/gallivm/lp_bld_sample.c:1483:10: warning: array index of '2' indexes past the end of an array (that contains 2 elements) [-Warray-bounds]
tmp[2] = lp_build_swizzle_aos(coord_bld, ddx_ddy[1], swizzle02);
^ ~
src/gallium/auxiliary/gallivm/lp_bld_sample.c:1430:10: note: array 'tmp' declared here
LLVMValueRef ddx_ddy[2], tmp[2], rho_vec;
^
src/gallium/auxiliary/gallivm/lp_bld_sample.c:1487:56: warning: array index of '2' indexes past the end of an array (that contains 2 elements) [-Warray-bounds]
rho_vec = lp_build_add(coord_bld, rho_vec, tmp[2]);
^ ~
src/gallium/auxiliary/gallivm/lp_bld_sample.c:1430:10: note: array 'tmp' declared here
LLVMValueRef ddx_ddy[2], tmp[2], rho_vec;
^
src/gallium/auxiliary/gallivm/lp_bld_sample.c:1491:56: warning: array index of '2' indexes past the end of an array (that contains 2 elements) [-Warray-bounds]
rho_vec = lp_build_max(coord_bld, rho_vec, tmp[2]);
^ ~
src/gallium/auxiliary/gallivm/lp_bld_sample.c:1430:10: note: array 'tmp' declared here
LLVMValueRef ddx_ddy[2], tmp[2], rho_vec;
^
This move the tracing timeout and printing into winsys and add
an debug environement variable for it (R600_DEBUG=trace_cs).
Lot of file touched because of winsys API changes.
v2: Do not write lockup file if ib uniq id does not match last one
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
There was a lot of code in evergreen_compute_internal.c that was not
being used at all and most of it was duplicating code from other parts
of the driver.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
v2:
- Fix usage of set_constant_buffer()
- Fix typo in comment
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
New textures or vertex buffers don't always require patching and
re-emitting the shaders. So do a better job of figuring out when we
actually have to patch the shader.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
If a bug in an app/stater-tacker causes vertex buffer to fetch vertex
elements that are not bound, simply return zeros instead of crashing.
Reviewed-by: Brian Paul <brianp@vmware.com>
clang is supports most gcc options / extensions, with a some exceptions.
The biggest advantage of using clang is that compilation times are much
short.
One can tell scons to use clang when building by invoking it as
CC=clang CXX=clang++ scons libgl-xlib
We were assigning incorrect const register for immediates, and
potentially writing immediate const to the wrong location. This fixes
an incorrect-rendering bug with xonotic.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Set a few extra registers to make sure we are in proper state for
clearing. And also add some debug options to mark all state dirty in
clear and gmem operations to aid in debugging.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
There is a bit we need to set for 2D vs 3D fetch, to tell the hw whether
there are two or there valid input components.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
The previous approach of using the dst register as an intermediate
temporary doesn't work in a lot of cases. For example, if the dst
register is the same as one of the src registers.
For now, just simplify it and always allocate a new register to use as
an intermediate. In some cases this will result in more registers used
than required. I think the best solution would be to implement an
optimization pass to reduce the number of registers used, which would
also solve the problem we have now of not being able to use GPRs that
are assigned for TGSI_FILE_INPUT.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Opps, didn't notice that I had left it stubbed out.
Also, make things fail a bit more gracefully when things go wrong.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Really this should be set based on buffer format, not on color vs
depth/stencil. Probably there should be more formats that set the bit
as we add support for more render target formats.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
The libelf implementation that is distributed here:
http://www.mr511.de/software/english.html
requires calling elf_version() prior to calling elf_memory()
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Lighter weight then using streamout. Only evergreen
and newer asics support embedded data as src with
CP DMA.
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Running piglit with this was causing all sort of weird stuff happening
to my desktop (Chromium webpages become blank, Qt Creator flickered,
etc). I tracked this down to shared memory segment leakage when GL is
not shutdown properly. The segments can be seen running `ipcs` and
looking for nattch==0.
This changes fixes this by calling shmctl(IPC_RMID) soon after creation
(which does not remove the segment immediately, but simply marks it for
removal when no more processes are attached).
This matches src/mesa/drivers/x11/xm_buffer.c behaviour.
v2:
- move shmctl(IPC_RMID) after XShmAttach() for *BSD, per Chris Wilson
- remove stray debug printfs, spotted by Ian Romanick
NOTE: This is a candidate for stable branches.
Reviewed-by: Brian Paul <brianp@vmware.com>
We need to handle the leading vertex information when
assembling primitives for the geometry shader otherwise
the resulting triangles will have vertices at incorrect
input locations.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
Should fix: https://bugs.freedesktop.org/show_bug.cgi?id=63702
v2: add a comment that this is just a workaround
v3: fix typo in comment
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
The previous commit introduced extra words, breaking the formatting.
This text transformation was done automatically via the following shell
command:
$ git grep 'THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY' | sed 's/:.*$//' | xargs -I {} sh -c 'vim -e -s {} < vimscript2
where 'vimscript2' is a file containing:
/THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY/;/^ *$/ !fmt -w 78 -p '// '
:wq
Reviewed-by: Brian Paul <brianp@vmware.com>
The previous commit introduced extra words, breaking the formatting.
This text transformation was done automatically via the following shell
command:
$ git grep 'THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY' | sed 's/:.*$//' | xargs -I {} sh -c 'vim -e -s {} < vimscript
where 'vimscript' is a file containing:
/THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY/;/\*\// !fmt -w 78 -p ' * '
:wq
Reviewed-by: Brian Paul <brianp@vmware.com>
This brings the license text in line with the MIT License as published
on the Open Source Initiative website:
http://opensource.org/licenses/mit-license.php
Generated automatically be the following shell command:
$ git grep 'THE AUTHORS BE LIABLE' | sed 's/:.*$//g' | xargs -I '{}' \
sed -i 's/THE AUTHORS/THE AUTHORS OR COPYRIGHT HOLDERS/' {}
This introduces some wrapping issues, to be fixed in the next commit.
Reviewed-by: Brian Paul <brianp@vmware.com>
Generated automatically be the following shell command:
$ git grep 'BRIAN PAUL BE LIABLE' | sed 's/:.*$//g' | xargs -I '{}' \
sed -i 's/BRIAN PAUL/THE AUTHORS/' {}
The intention here is to protect all authors, not just Brian Paul. I
believe that was already the sensible interpretation, but spelling it
out is probably better.
More practically, it also prevents people from accidentally copy &
pasting the license into a new file which says Brian is not liable when
he isn't even one of the authors.
Reviewed-by: Brian Paul <brianp@vmware.com>
Squashed commit of the following:
commit 04c5fa2cbb8e89d6f2fa5a75af1cca03b1f6b852
Author: José Fonseca <jfonseca@vmware.com>
Date: Tue Apr 23 17:37:18 2013 +0100
gallium: s/lower_left_origin/bottom_edge_rule/
commit 4dff4f64fa83b9737def136fffd161d55e4f1722
Author: José Fonseca <jfonseca@vmware.com>
Date: Tue Apr 23 17:35:04 2013 +0100
gallium: Move diagram to docs.
commit 442a63012c8c3c3797f45e03f2ca20ad5f399832
Author: James Benton <jbenton@vmware.com>
Date: Fri May 11 17:50:55 2012 +0100
gallium: Replace gl_rasterization_rules with lower_left_origin and half_pixel_center.
This change is necessary to achieve correct results when using OpenGL
FBOs.
Reviewed-by: Marek Olšák <maraeo@gmail.com>
This fixes a crash when a resource cannot be mapped to the CPU's address space
because it's too big.
This puts a global pipe_context in r600_screen, which is guarded by a mutex,
so that we can use pipe_context when there isn't one around.
Hopefully our multi-context support is solid.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
NOTE: This is a candidate for the 9.1 branch.
Although this might be useful for ARB_clear_buffer_object,
I need it for initializating resources in r600g.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
v2: comment cleanups
NOTE: This is a candidate for the 9.1 branch.
Number of vertices to fetch doesn't always equal the number of input
vertices. To correctly compute the number if IA primitives we need
to use the total number of input vertices, not only those that
need to be fetched.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
TGSI geometry shader input declerations are of the IN[][2] format
and the dimensions of the array have to be deduced from the input
primitive property.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
We want to be able to reset certain parts of the pipeline,
in particular the input primitive index, but only either with
seperate invocations of the draw_vbo or new instances. In all
other cases (e.g. new invocations due to primitive restart)
that data needs to be preserved. Add a function through which
we can reset instance dependent data.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
Same approach as in the llvmpipe, if the geometry shader is
null and we have stream output then attach it to the vertex
shader right before executing the draw pipeline.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
TEMP is not the only register file that accept unsigned. OUT too.
Actually, what determines the appropriate type of the destination value is
not the opcode, but rather the register.
Also cleanup/simplify code. Add a few more asserts, but also make
code more robust by handling graceful if assert fails.
This fixes segfault / assertion in the included vert-uadd.sh graw shader.
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
2.7 was a particularly trouble ridden release.
Furthermore, the bug no longer can be reproduced ever since the
first_level state was taken in account.
Reviewed-by: Brian Paul <brianp@vmware.com>
They are supported on LLVM 3.1, at least on x86. (I haven't tested on PPC
though.)
Actually lp_build_linear_mip_levels() already has been emitting them for
some time.
This avoids intrinsics, which tend to be an obstacle for certain
optimization passes.
Reviewed-by: Brian Paul <brianp@vmware.com>
There will be a new IR for a3xx, which has a very different shader ISA
(more scalar oriented). So rename to avoid conflicts later when I start
adding a3xx support to the gallium driver.
Signed-off-by: Rob Clark <Rob Clark robdclark@freedesktop.org>
The standalone shader assembler needed some meta-data to know about
attributes/varyings/etc, to do the shader linkage. We don't need these
parts with gallium/tgsi, so just get rid of it.
Signed-off-by: Rob Clark <Rob Clark robdclark@freedesktop.org>
Should be able to handle all things which make this tricky to implement.
Fallthroughs, including most notably into/out of default, should be handled
correctly but are quite a mess.
If we see largely unoptimized switches in the wild should probably think
about some "real" switch optimization pass, e.g. things like this:
switch
case1
someinst
brk
case2
default
case3
someinst
brk
case4
someinst
endswitch
are legal, but the pointless case2/case3 statements not only cause condition
evaluation but will turn this into a "fake" fallthrough case (because
mask and defaultmask are already updated for case2 when default is
encountered) requiring executing code twice.
If default is at the end though, there's never any code re-execution, and
if that's not the case if there's no fallthrough in (not even a fake one)
and out of default there's no code re-execution neither.
v2: add comments, and use enum for break type instead of magic boolean.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
It seems there was a typo in gallivm breakc handling (I am actually still
not sure it is really needed but otherwise that statement really should go
away). Also fix the wrong src argument type, even though they weren't really
used.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
While initially that opcode probably was meant for something along the
lines of sm3 break_comp it has never worked that way (not even the
argument count was right) and now the opcode has quite different
semantics so just remove it. (Discovered by Jose Fonseca)
This is still not really correct, since at least for sm 4.0
the nesting limit is 64 per subroutine, and subroutine nesting itself
has a limit of 32, so since we have a flat stack we'd need 32*64.
But this should probably be better fixed with per-subroutine stacks,
since otherwise these structures get really big (like 100kB for the
lp_exec_mask).
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Input assembler needs to be able to decompose adjacency primitives
into something that can be understood by the rest of the pipeline.
The specs say that the adjacency primitives are *only* visible
in the geometry shader, for everything else they need to be
decomposed. Which in most of the cases is not an issue, because
the geometry shader always decomposes them for us, but without
geometry shader we were passing unchanged adjacency primitives
to the rest of the pipeline and causing crashes everywhere. This
commit introduces a primitive assembler which, if geometry
shader is missing and the input primitive is one of the
adjacency primitives, decomposes them into something
that the rest of the pipeline can understand.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
pre_clip_pos is a float[4] we just used (*float)[4] to be able to
jump within the array of vertex_headers with it. So if the idx
happened to be anything but 0, we'd actually read from some garbage
in memory. Change it to just be a simple pointer instead of casting
it to something that it's not. As suggested by Jose.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Since stdbool.h's "true" and "false" are #defines, they got expanded when
used as macro arguments, and that expanded value was stored in the
XML string, producing XML that driconf would then fail to parse.
Currently no drivers included stdbool along with driconf, but I keep
accidentally doing so on intel as we move towards using normal C.
v2: rebase on master.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> (v1)
Most test pass, issue are with border color and swizzle.
Based on ircnick<maelcum> patch.
v2: Restaged commit hunk
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
v2: Remove left over code
v3: Restage properly the commit so hunk of first one are not in
second one.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
NOTE: This is a candidate for the 9.1 branch
Signed-off-by: Vadim Girlin <vadimgirlin@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Not all are supported as render targets.
The state tracker fallback of using RGBA instead of RGBX currently
fails for blending, we could work around this by clearing their alpha
to 1 and modifying the color mask to disable writing alpha.
This is the only sane solution for nv50 and nvc0 (really, trust me),
but since on other hardware the border colour is tightly coupled with
texture state they'd have to undo the swizzle, so I've added a cap.
The dependency of update_sampler on the texture updates was
introduced to avoid doing the apply_depthmode to the swizzle twice.
v2: Moved swizzling helper to u_format.c, extended the CAP to
provide more accurate information.
Turns out the previous "fix" for handling per-pixel face selection and
derivatives didn't work out that well - the derivatives were wrong by
quite a bit, in theory transformation of the derivatives into cube space
should work, but would be _a lot_ more work than the "simplified" transform
used.
So, for explicit derivatives, I'm just giving up and go back to not honoring
them.
For implicit derivatives (and the fake explicit ones) however we try
something a little different, we just calculate rho as we would for a 3d
texture, that is after scaling the coords by the inverse major axis.
This gives the same results as calculating the derivs after projection of
the coords to the same face as long as all pixels hit the same face (and
only without rho_no_opt, otherwise it should be a bit worse). And when
not all pixels are hitting the same face, the results aren't so hot but
not catastrophically bad (I believe not off by more than a factor of 2 without
no_rho_approx and not more than sqrt(2) with no_rho_approx). I think this is
better than just picking the wrong face but who knows...
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
This will calculate rho correctly as
sqrt(max((ds/dx)^2 + (dt/dx)^2 + (dr/dx)^2), (ds/dx)^2 + (dt/dx)^2 + (dr/dx)^2))
instead of max(|ds/dx|,|dt/dx|,|dr/dx|,|ds/dy|,|dt/dy,|dr/dy|)
(for 3 coords - 2 coords work analogous, for 1 coord there's no point doing
the exact version), for both implicit and explicit derivatives.
While such approximation seems to be allowed in OpenGL some APIs may be less
forgiving, and the error can be quite large (sqrt(2) for 2 coords, sqrt(3) for
3 coords so wrong by nearly one mip level in the latter case).
This also helps to single out "real" bugs from "expected" ones, so it is debug
only (though at least combined with no_brilinear I didn't really see much of a
performance difference but only tested with a debug build - at least with
implicit mipmaps the instruction count is almost exactly the same though the
instructions are more complex (1 sqrt and mul/adds instead of and/max mostly).
The code when the option isn't set stays exactly the same.
v2: rename no_rho_opt to no_rho_approx.
Reviewed-by: Brian Paul <brianp@vmware.com>
We were trying to use a destroy method from a deleted context.
This fix is based on what's in the svga driver.
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>
we were ignoring leading/provoking vertex settings which was
breaking decomposition of some strips.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
we were using the wrong vars, reporting incorrect stream output
statistics.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
We were always treating the vertex index as a scalar but when the
shader is using indirect addressing it will be a vector of indices
for each channel. This was causing some nasty crashes insides
LLVM.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
If a window's minimized we get a zero-size window. Skip the SwapBuffers
in that case to avoid some warning messages with the VMware svga driver.
Internal bug #996695
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
The specification says that the geometry shader should exit if the
number of emitted vertices is bigger or equal to max_output_vertices and
we can't do that because we're running in the SoA mode, which means that
our storing routines will keep getting called on channels that have
overflown (even though they will be masked out, but we just can't skip
them).
So we need some scratch area where we can keep writing the overflown
vertices without overwriting anything important or crashing.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Can happen if we were using stream output without geometry
shader, by returning early we avoid a crash.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
This is a basic implementation of the pipeline statistics in the
draw module. The interface is similar to the stream output statistics
and also requires that the callers explicitly enable it.
Included is the implementation of the interface in llvmpipe and
softpipe. Only softpipe enables the pipeline statistics capability
though because llvmpipe is lacking gathering of the fragment shading
and rasterization statistics.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
The issue with SOA execution and end_primitive opcode is that it
can be executed both when we haven't emitted any vertices, in
which case we don't want to emit an empty primitive, and when
the execution mask is zero and the execution should be skipped. We
handled only the latter of those conditions. Now we're combining the
execution mask with a mask created from emitted vertices to handle
both cases. As a result we don't need the pending_end_primitive
flag which was broken because it was static and could be affected
by both above mentioned conditions at run-time.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
which means that our execution mask in GS is equal to 1 not 0xf.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Same as with llvmpipe: we can't be divind/moding by zero and we
need to make sure that dividing/moding by zero produces 0xffffffff.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
TGSI_OPCODE_IF condition had two possible interpretations:
- src.x != 0.0f
- Mesa statetracker when PIPE_SHADER_CAP_INTEGERS was false either for
vertex and fragment shaders
- gallivm/llvmpipe
- postprocess
- vl state tracker
- vega state tracker
- most old drivers
- old internal state trackers
- many graw examples
- src.x != 0U
- Mesa statetracker when PIPE_SHADER_CAP_INTEGERS was true for both
vertex and fragment shaders
- tgsi_exec/softpipe
- r600
- radeonsi
- nv50
And drivers that use draw module also were a mess (because Mesa would
emit float IFs, but draw module supports native integers so it would
interpret IF arg as integers...)
This sort of works if the source argument is limited to float +0.0f or
+1.0f, integer 0, but would fail if source is float -0.0f, or integer in
the float NaN range. It could also fail if source is integer 1, and
hardware flushes denormalized numbers to zero.
But with this change there are now two opcodes, IF and UIF, with clear
meaning.
Drivers that do not support native integers do not need to worry about
UIF. However, for backwards compatibility with old state trackers and
examples, it is advisable that native integer capable drivers also
support the float IF opcode.
I tried to implement this for r600 and radeonsi based on the surrounding
code. I couldn't do this for nouveau, so I just shunted IF/UIF
together, which matches the current behavior.
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
v2:
- Incorporate Roland's feedback.
- Fix r600_shader.c merge conflict.
- Fix typo in radeon, spotted by Michel Dänzer.
- Incorporte Christoph Bumiller's patch to handle TGSI_OPCODE_IF(float)
properly in nv50/ir.
There is a hardware bug on Cayman where a BREAK/CONTINUE followed by
LOOP_STARTxxx for nested loops may put the branch stack into a state
such that ALU_PUSH_BEFORE doesn't work as expected. Workaround this
by replacing the ALU_PUSH_BEFORE with a PUSH + ALU
Fixes piglit tests EXT_transform_feedback/order*
v2: Use existing loop count and improve comment
v3: [Vadim Girlin] Set jump address for PUSH instructions
NOTE: This is a candidate for the 9.1 branch
Signed-off-by: Vadim Girlin <vadimgirlin@gmail.com>
configure still uses it to print the enabled winsys.
Tested-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-and-Tested-by: Andreas Boll <andreas.boll.dev@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
configure still uses it to print the enabled targets.
Tested-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-and-Tested-by: Andreas Boll <andreas.boll.dev@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
And don't build it from other Makefiles. That's awful, and breaks
distclean.
Tested-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-and-Tested-by: Andreas Boll <andreas.boll.dev@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Tested-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-and-Tested-by: Andreas Boll <andreas.boll.dev@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
configure still uses it to print the enabled state trackers.
Tested-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-and-Tested-by: Andreas Boll <andreas.boll.dev@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Tested-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-and-Tested-by: Andreas Boll <andreas.boll.dev@gmail.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Instead of emitting configuration values (e.g. number of gprs used) in a
predefined order, the LLVM backend now emits these values in
register/value pairs. The first dword contains the register address and
the second dword contians the value to write.
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Inserting the value for the second quad in the wrong place for the
following shuffle. This meant the row or image stride was undefined which is
quite catastrophic, can lead to bogus texels fetched or just segfault.
This code is only hit for SoA path currently, still surprising it
didn't crash more or caused more visible issues (I think llvm used a
broadcast shuffle for the undefined parts of the vector, hence the undefined
value for the second quad was just the same as that from the first quad,
so as long as both quads hit the same mip level everything was fine, and since
lower mips always have the same large stride it made it less likely to
hit out-of-bound memory in case of differing lods).
Note: this is a candidate for stable branches.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Null platform IDs are OK according to the spec, but some applications have
been reported to get paranoid and assume that our NULL platform is unusable.
As it doesn't hurt to have device enumeration separate from the rest of the
device code (quite the opposite, it makes the code cleaner), make the API use
an actual platform object that keeps track of the available devices instead of
the former NULL pointer.
Reported-and-reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
Pointed out by gcc
nve4_compute.c: In function 'nve4_launch_grid':
nve4_compute.c:511:7: warning: 'ret' may be used uninitialized in
this function [-Wmaybe-uninitialized]
if (ret)
^
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Edit by Christoph Bumiller:
Set it to -1 to indicate failure and only when it's actually required.
As otherwise it is unused - pointed out by gcc
nve4_compute.c:586:20: warning: 'nve4_cache_split_name' defined but not used [-Wunused-function]
static const char *nve4_cache_split_name(unsigned value)
^
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
For debug build we'll hit the assert, for release we are going to emit random data
as subOp is used uninitilised. Spotted by gcc
codegen/nv50_ir_emit_nv50.cpp: In member function 'void nv50_ir::CodeEmitterNV50::emitATOM(const nv50_ir::Instruction*)':
codegen/nv50_ir_emit_nv50.cpp:1554:12: warning: 'subOp' may be used uninitialized in this function [-Wmaybe-uninitialized]
uint8_t subOp;
^
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
v2: I rewrote this to use the sample positions properly.
v3: rewrite properly to use bitfield to cast back to signed ints
Signed-off-by: Dave Airlie <airlied@redhat.com>
I've no idea when sample_chan would ever be 4 here, but 4 is most
definitely wrong, array textures have it as 3 as well.
Also the cayman code though unused is obviously wrong.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Just everything you need for UVD with r600g and radeonsi.
v2: move UVD code to radeon subdir, clean up build system additions,
remove an unused SI function, disable tiling on SI for now.
v3: some minor indentation fix and rebased
v4: dpb size calculation fixed
v5: implement proper fall-back in case the kernel doesn't support UVD,
based on patches from Andreas Boll but cleaned up a bit more.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Separated from UVD patch for clarity.
v2: sync with next tree for 3.10
v3: as pointed out by Andreas Bool check for drm minor >= 32
http://cgit.freedesktop.org/~agd5f/linux/log/?h=drm-next-3.10-wip
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Andreas Boll <andreas.boll.dev@gmail.com>
both mov and ucmp can be used to move variables of any type.
correctly note that about ucmp in the tgsi_info and make
sure gallivm can handle that by correctly casting the untyped
moves.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
We were using simple temporaries, without using alloca or phi
nodes which meant that on every iteration of the loop our
temporaries, which were holding the number of vertices and
primitives which were emitted, were being reset to zero. Now
we're using alloca to allocate those variables to preserve
them across conditionals.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
We were missing the implementation of PIPE_QUERY_SO_STATISTICS
query, this change implements it on top of the existing
facilities.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
We want to both make sure we never divide by zero to not generate
sigfpe and that divide by zero is guaranteed to return 0xffffffff.
Based on José idea.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
we break when the mask values are 0 not, 1, plus it's bit comparison
not a floating point comparison. This fixes both.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
v2: fix instrinsic name as well
v3: LLVM revision incremented as well
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Drawing subtitles didn't increased the dirty area of the surface.
Reported and tested by freeedrich on irc.
v2: don't clear the surface
Signed-off-by: Christian König <christian.koenig@amd.com>
Fixes uninitialized scalar variable defect reported by Coverity.
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
The multiplication part of tgsi_umad did not work on Cayman, because it did
not populate the correct vector slots.
This fixed hardlocks in the EXT_transform_feedback/order tests.
NOTE: This is a candidate for the stable branches.
(might not be easy to cherry-pick though)
Signed-off-by: Marek Olšák <maraeo@gmail.com>
v2:
- Only dump shaders when env variable is set.
v3:
- Don't emit VGT registers
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com
This function is a holdover from r600g and is identical to
si_pm4_inval_texture_cache(), so it is not needed.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com
This target string now contains four values instead of three. The old
processor field (which was really being interpreted as arch) has been split
into two fields: processor and arch. This allows drivers to pass a
more a more detailed description of the hardware to compiler frontends.
v2:
- Adapt to libclc changes
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Add UTIL_FORMAT_LAYOUT_ETC to util_format_is_compressed. It was missing.
Signed-off-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Don't check if there's sampler support for stencil if we're not
going to actually blit/copy stencil values. Fixes the case where
we mistakenly said we can't support a blit of depth values from
S8Z24 to X8Z24.
Also, rename the is_stencil variable to dst_has_stencil to improve
readability.
NOTE: This is a candidate for the stable branches.
Reviewed-by: Marek Olšák <maraeo@gmail.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
Switch to use the envytools generated headers for register/bitfield
definitions. This is the first step in preparing to add a3xx support,
since it avoids having conflicting names for a3xx and a2xx registers.
And since I'm using envytools for a3xx it is simpler to just use it for
everything.
This shouldn't cause any functional change, it is really just a lot of
renaming.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Otherwise we will not receive destroy windows events, causing framebuffers
to leak.
This happens particularly with java and jogl.
Tested with java + jogl, MATLAB.
VMware Internal Bug Number: 1013086.
Reviewed-by: Brian Paul <brianp@vmware.com>
At least on llvm 3.2 this appears to work fine. Tested on an Athlon XP
2600+, which has sse and 3dnow but not sse2.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Build time option, set RADEON_CS_DUMP_ON_LOCKUP to 1 in radeon_drm_cs.h to
enable it.
When enabled after each cs submission the code will try to detect lockup by
waiting on one of the buffer of the cs to become idle, after a timeout it
will consider that the cs triggered a lockup and will write a radeon_lockup.c
file in current directory that have all information for replaying the cs.
To build this file :
gcc -O0 -g radeon_lockup.c -ldrm -o radeon_lockup -I/usr/include/libdrm
v2: Add radeon_ctx.h file to mesa git tree
v3: Slightly improve dumped file for easier editing, only dump first faulty cs
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
The default wrap mode (PIPE_TEX_WRAP_REPEAT) is incompatible with
unnormalized texcoords (at least for softpipe).
v2: use PIPE_TEX_WRAP_CLAMP_TO_EDGE
Reviewed-by: Marek Olšák <maraeo@gmail.com>
The ar_ge_as_at variable was just very very confusing since the condition
was actually the other way around (as_at_ge_ar). So change the condition
(and the selects depending on it) to match the variable name.
And also change the chosen major axis in case the coord values are the
same. OpenGL doesn't care one bit which one is chosen in this case but
it looks like dx10 would require z chosen over y, and y chosen over x
(previously did x chosen over y, y chosen over z). Since it's all the
same effort just honor dx10's wishes. (Though actually, for some prefered
orderings, we could save one (or two with derivatives) selects since the
tnewx and tnewz (and the corresponding dmax values) are the same.)
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
GCC 4.8 now warns about typedefs that are local to a scope and not
used anywhere within that scope. This produced spurious warnings with
the STATIC_ASSERT() macro (which used a typedef to provoke a compile
error in the event of an assertion failure).
This patch switches to a simpler technique that avoids the warning.
v2: Avoid GCC-specific syntax. Also update p_compiler.h.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
The former just checks that the given block is valid by checking
the header and footer.
The later sets the memory block's tag. With extra debug code, we
can use that for monitoring/checking particular allocations.
Reviewed-by: José Fonseca <jfonseca@vmware.com>
This is trivial now, though need to make sure we pass all the necessary
derivative values (which is 3 each for ddx/ddy not 2).
Passes piglit arb_shader_texture_lod-texgradcube test.
v2: add the forgotten abs() for all incoming derivatives (discovered
by new piglit arb_shader_texture_lod-texgradcube test, though more by
luck as it was failing only for exactly one pixel...).
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
This proved to be tricky, the problem is that after selection/mirroring
we cannot calculate reasonable derivatives (if not all pixels in a quad
end up on the same face the derivatives could get "randomly" exceedingly
large).
However, it is actually quite easy to simply calculate the derivatives
before selection/mirroring and then transform them similar to
the cube coordinates (they only need selection/projection, but not
mirroring as we're not interested in the sign bit, of course). While
there is a tiny bit more work to do (need to calculate derivs for 3
coords instead of 2, and additional selects) it also simplifies things
somewhat for the coord selection itself (as we save some broadcast aos
shuffles, and we don't need to calculate the average vector) - hence if
derivatives aren't needed this should actually be faster.
Also, this has the benefit that this will (trivially) work for explicit
derivatives too, which we completely ignored before that (will be in a
separate commit for better trackability).
Note that while the way for getting rho looks very different, it should
result in "nearly" the same values as before (the "nearly" is only because
before the code would choose the face based on an "average" vector and hence
the derivatives calculated according to this face, where now (for implicit
derivatives) the derivatives are projected on the face selected for the
first (top-left) pixel in a quad, so not necessarly the same face).
The transformation done might not quite be state-of-the-art, calculating
length(dx,dy) as max(dx,dy) certainly isn't neither but this stays the
same as before (that is I think a better transform would _somehow_ take
the "derivative major axis" into account so that derivative changes in
the major axis wouldn't get ignored).
Should solve some accuracy problems with cubemaps (can easily be seen with
the cubemap demo when switching wrapping/filtering), though we still don't
do seamless filtering to fix it completely (so not per-sample but per-pixel
is certainly better than per-quad and already sufficient for accurate
results with nearest tex filter).
As for performance, it seems to be a tiny bit faster too (maybe 3% or so
with cubemap demo). Which I'd have expected with nearest/nearest filtering
where this will be less instructions, but the difference seems to actually
be larger with linear/linear_mipmap_linear where it is slightly more
instructions, probably the code appears less serialized allowing better
scheduling (on a sandy bridge cpu). It actually seems to be now at least
as fast as the old path using a conditional when using 128bit vectors too
(that is probably more a result of testing with a newer cpu though), for now
that old path is still there but unused.
No piglit regressions.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Using a different packing for the single coord case should save a shuffle.
Plus some minor style fixes.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Should be way faster of course on cpus supporting this (includes AMD
Bulldozer and Jaguar cores, Intel Ivy Bridge and up (except budget models)).
Passes piglit fbo-blending-formats GL_ARB_texture_float -auto on Ivy Bridge.
Reviewed-by: Brian Paul <brianp@vmware.com>
When geometry shaders are present, one needs to be able to create
an empty geometry shader with stream output that needs to be
resolved later and attached to the currently bound vertex shader.
Lets add support for it to llvmpipe and draw. draw allows attaching
independent stream output info to any vertex shader and llvmpipe
resolves at draw time which vertex shader the given empty geometry
shader should be linked to.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
We need to reset the internal state of the so buffers or we'll
keep appending even though we're not supposed to.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
we use draw_set_mapped_so_targets nowadays
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
I think this was there before and got accidently
removed during a merge. Same code as for the GS
context, which is also using an enum instead of
hardcoded numbers.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
It's quite helpful during the rendering when we know
exactly the count of the vertices available in the
buffer.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
We were flushing with incorrect number of primitives. TGSI exec
can only work with a single primitive at a time. Plus the fetching
with multiple primitives on llvm paths wasn't copying the last
element.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
Instead of void pointers use a base interface.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
The functions are prototyped in u_transfer.h and are related to the
other functions in u_transfer.c.
The next patch will re-use the u_resource.c file for new code.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Ported from r600g commit:
8891b2f9c9
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
NOTE: This is a candidate for the 9.1 branch.
The fallbacks count is the number of drawing calls that use a "draw"
module fallback, such as polygon stipple.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
This still isn't optimal, since the fence will signal a bit late,
but better than checking on the bo, which may never be ready if it
is shared (which is likely).
Also, renamed "pixels-rendered" to "samples-passed" because the
occlusion counter increments even if colour and depth writes are
disabled, or (on some implementations) for killed fragments that
passed the depth test when PS early_fragment_tests is set.
Conceptually the same as previously done in float_to_half.
Should cut down number of instructions from 14 to 10 or so, but
will promote some NaNs to Infs, so it's disabled.
It gets a bit tricky though handling all the cases correctly...
Passes basic tests either way (though there are no tests testing special
cases, but some manual tests injecting them seemed promising).
v2: style and comment fixes suggested by Jose
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
This replaces the existing float-to-half implementation.
There are definitely a couple of differences - the old implementation
had unspecified(?) rounding behavior, and could at least in theory
construct Inf values out of NaNs. NaNs and Infs should now always be
properly propagated, and rounding behavior is now towards zero
(note this means too large but non-Infinity values get propagated to max
representable value, not Infinity).
The implementation will definitely not match util code, however (which
does nearest rounding, which also means too large values will get
propagated to Infinity).
Also fix a bogus round mask probably leading to rounding bugs...
v2: fix a logic bug in handling infs/nans.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Reduced stack size allows to run more threads in some cases,
improving performance for the shaders that use stack (that is, for the
shaders with control flow instructions). E.g. with unigine-based apps.
v4: implement exact computation taking into account wavefront size
v5: add cases for RV620, RS880
Signed-off-by: Vadim Girlin <vadimgirlin@gmail.com>
v2: reduce key size, don't copy key around to much.
v3: remove key size reduction
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
This works different than on R600, we need to add the start instance manually.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
This should be used by both SI and R600.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Fixes mplayer -vo vdpau OSD.
NOTE: This is a candidate for the 9.1 branch.
Reported-by: Igor Vagulin <igor.vagulin@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Tested-by: Christian König <christian.koenig@amd.com>
If we're drawing to a surface that's 2048 x 2048 pixels or larger there's
danger of fixed-point overflow in the triangle rasterization code. That
leads to various rendering glitches.
Rather than implement some intricate changes to the rasterization code,
simply subdivide triangles into smaller subtriangles to avoid the issue.
Only do this when the drawing surface is larger than 2048 by 2048.
Reviewed-by: José Fonseca <jfonseca@vmware.com>
Similar enough that we can try to use shared code.
v2: fix a stupid bug using wrong variable causing mayhem with Inf and NaNs.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com
There's more, but this only adds (most) of the counters that are
handled directly by the shader processors.
The other counter domains are not handled on the multiprocessor and
there are no FIFO object methods for configuring them.
Instead, they have to be programmed by the kernel via PCOUNTER, and
the interface for this isn't in place yet.
We weren't correctly propagating the samplers and sampler views
when they were related to geometry shaders.
Signed-off-by: Zack Rusin <zackr@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>