| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
|
|
| |
The Vulkan driver doesn't use libdrm so we don't want to bake that in.
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
| |
This commit switches all of blorp from taking a brw_context to taking a
blorp_context and, where useful, a void *batch. In the GL driver, we only
have one active batch at a time so the brw_context *is* the batch but in
Vulkan, batch will point to the anv_cmd_buffer in which we are building
instructions.
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
|
|
|
|
| |
This gets rid of brw_context throughout the core of the state setup code.
Instead, it is replaced with blorp_batch which contains a pointer to the
blorp_context and a void* that the driver can use for its own blorp data.
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
| |
Previously, we passed the buffer address (as per the latest offset from the
kernel) to ISL to use when it filled out the surface state. We then called
drm_intel_bo_emit_reloc() to add the relocation to the list. The newly
added blorp_surface_reloc helper adds the relocation to the list and then
writes the buffer address directly into the surface state.
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
|
|
| |
This splits it into two functions very similar to gen7_upload_urb.
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
|
|
|
| |
This pulls most of the brw-specific bits into helpers with generic names.
Later, those will become the driver hooks for generic code.
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This sanitizes blorp's access to the i965 driver's shader cache by patching
it through the blorp_context. When we start using blorp in Vulkan, we will
simply have to implement such a caching interface in the Vulkan driver.
Note: In my first attempt at this, I simplified it down to a single
upload_shader entrypoint and implemented the caching inside of blorp. This
doesn't work, however, because the i965 driver will, on occation, dump its
entire cache and start over. When this happens, blorp needs to be able to
recompile its shaders and re-upload them. It's easiest to just expose the
caching interface.
Signed-off-by: Jason Ekstrand <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jason Ekstrand <[email protected]>
Reviewed-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
| |
Android porting of commit bebc1a1 "intel: Flatten the makefile structure"
Automake approach was followed, by moving makefiles a level up,
naming them Android.genxml.mk and Android.isl.mk,
performing the necessary adjustments to the paths,
adding src/intel/Android.mk and fixing mesa top level makefile.
Acked-by: Jason Ekstrand <[email protected]>
|
|
|
|
|
|
| |
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97513
Signed-off-by: Jan Vesely <[email protected]>
Reviewed-by: Francisco Jerez <[email protected]>
|
|
|
|
|
|
|
|
|
| |
v2: document the new cap
v3: fix 80 char limit in screen.rst
Signed-off-by: Jan Vesely <[email protected]>
Reviewed-by: Marek Olšák <[email protected]>
Acked-by: Ilia Mirkin <[email protected]>
|
|
|
|
| |
Reviewed-by: Charmaine Lee <[email protected]>
|
|
|
|
|
|
| |
Signed-off-by: Jordan Justen <[email protected]>
Acked-by: Kenneth Graunke <[email protected]>
Reviewed-by: Samuel Iglesias Gonsálvez <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Small code clean up that removes magic numbers where a TGSI
opcode has been defined.
No functional change expected as each opcode is unsupported on
the respective hardware.
Signed-off-by: Rhys Kidd <[email protected]>
Reviewed-by: Marek Olšák <[email protected]>
Tested-by: James Harvey <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
As reported by Clang, TGSI_OPCODE_DFMA (defined magic number 118) is
currently initialized twice for Cayman and Evergreen.
When Jan Vesely added double precision FMA opcode it did make sense
to locate it immediately after TGSI_OPCODE_DMAD, although this is
out of order.
This change cleans up the prior magic number definition and ensures
any later reordering of this struct will not create problems.
Prior change was:
commit 015e2e0fce3eea7884f8df275c2fadc35143a324
Author: Jan Vesely <[email protected]>
Date: Sat Jul 2 16:14:54 2016 -0400
r600g: Add double precision FMA ops
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96782
Fixes: 54c4d525da7c7fc1e103d7a3e6db015abb132d5d ("r600g: Enable FMA on chips that support it")
Signed-off-by: Jan Vesely <[email protected]>
Tested-by: James Harvey <[email protected]>
Signed-off-by: Marek Olšák <[email protected]>
Signed-off-by: Rhys Kidd <[email protected]>
Reviewed-by: Jan Vesely <[email protected]>
Reviewed-by: Marek Olšák <[email protected]>
Tested-by: James Harvey <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
| |
Noticed this error in a debug message whilst reviewing
https://bugs.freedesktop.org/show_bug.cgi?id=97477
This patch doesn't go towards fixing that bug, but at
least may clarify future debug output.
Signed-off-by: Rhys Kidd <[email protected]>
Reviewed-by: Eric Anholt <[email protected]>
|
|
|
|
|
| |
I missed this while adding loop support because the discard test inside a
loop was crashing before, anyway. Fixes piglit glsl-fs-discard-04.
|
| |
|
|
|
|
|
|
|
| |
vc4 is about to start using the shader info field to set up discard
handling.
Reviewed-by: Jason Ekstrand <[email protected]>
|
|
|
|
|
|
|
| |
Only rasterize scissor edges if one or more scissor/viewport
rects are not hottile aligned.
Signed-off-by: Tim Rowley <[email protected]>
|
|
|
|
| |
Signed-off-by: Tim Rowley <[email protected]>
|
|
|
|
|
|
|
| |
Implement SCATTERPS as a dynamic loop based on mask set bits
instead of a static compile time loop.
Signed-off-by: Tim Rowley <[email protected]>
|
|
|
|
|
|
| |
Fixes upper left rule for scissors and viewport/scissor macrotile alignment.
Signed-off-by: Tim Rowley <[email protected]>
|
|
|
|
| |
Signed-off-by: Tim Rowley <[email protected]>
|
|
|
|
| |
Signed-off-by: Tim Rowley <[email protected]>
|
|
|
|
|
|
| |
Use dynamic memory allocation for per-thread data
Signed-off-by: Tim Rowley <[email protected]>
|
|
|
|
| |
Signed-off-by: Tim Rowley <[email protected]>
|
|
|
|
|
|
|
|
| |
- use per-primitive viewports throughout the pipeline.
- track whether all available scissor rects are tile aligned.
Causes failures, so not taken into account when choosing rasterizer yet.
Signed-off-by: Tim Rowley <[email protected]>
|
|
|
|
|
|
|
|
|
| |
We were allocating global variables for the maximum LDS size
which made the compiler think we were using all of LDS, which
isn't the case.
Reviewed-By: Edward O'Callaghan <[email protected]>
Reviewed-by: Marek Olšák <[email protected]>
|
| |
|
| |
|
|
|
|
|
| |
Signed-off-by: Kai Wasserbäch <[email protected]>
Reviewed-by: Brian Paul <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Kai Wasserbäch <[email protected]>
Reviewed-by: Brian Paul <[email protected]>
|