Basically, to "DCE" blits triggered by resource shadowing, in cases
where the levels are immediately completely overwritten. For example,
mid-frame texture upload to level zero triggers shadowing and back-blits
to the remaining levels, which are immediately overwritten by
glGenerateMipmap().
Signed-off-by: Rob Clark <robdclark@gmail.com>
To make batch re-ordering useful, we need to be able to create shadow
resources to avoid a flush/stall in transfer_map(). For example,
uploading new texture contents or updating a UBO mid-batch. In these
cases, we want to clone the buffer, and update the new buffer, leaving
the old buffer (whose reference is held by cmdstream) as a shadow.
This is done by blitting the remaining other levels (and whatever part
of current level that is not discarded) from the old/shadow buffer to
the new one.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Note that I originally also had a entry-point that would construct a key
and do lookup from a pipe_surface. I ended up not needing that (yet?)
but it is easy-enough to re-introduce later if we need it for the blit
path.
For now, not enabled by default, but can be enabled (on a3xx/a4xx) with
FD_MESA_DEBUG=reorder.
Signed-off-by: Rob Clark <robdclark@gmail.com>
To flush batches out of order, the gmem code needs to not depend on
state from fd_context (since that may apply to a more recent batch).
So this all moves into batch.
The one exception is the gmem/pipe/tile state itself. But this is
only used from gmem code (and batches are flushed serially). The
alternative would be having to re-calculate GMEM layout on every
batch, even if the dimensions of the render targets are the same.
Note: This opens up the possibility of pushing gmem/submit into a
helper thread.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Introduce the batch object, to track a batch/submit's worth of
ringbuffers and other bookkeeping. In this first step, just move
the ringbuffers into batch, since that is mostly uninteresting
churn.
For now there is just a single batch at a time. Note that one
outcome of this change is that rb's are allocated/freed on each
use. But the expectation is that the bo pool in libdrm_freedreno
will save us the GEM bo alloc/free which was the initial reason
to implement a rb pool in gallium.
The purpose of the batch is to eventually facilitate out-of-order
rendering, with batches associated to framebuffer state, and
tracking the dependencies on other batches.
Signed-off-by: Rob Clark <robdclark@gmail.com>
This bug seems to have always been there. Applications changing shaders
but not textures between draw calls would have gotten undefined behavior.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
This just needs to be done by st_validate_state.
v2: add "shaders_may_be_dirty" flags for not skipping st_validate_state
on _NEW_PROGRAM to detect real shader changes
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
The goal is to do this in st_validate_state:
while (dirty)
atoms[u_bit_scan(&dirty)]->update(st);
That implies that atoms can't specify which flags they consume.
There is exactly one ST_NEW_* flag for each atom. (58 flags in total)
There are macros that combine multiple flags into one for easier use.
All _NEW_* flags are translated into ST_NEW_* flags in st_invalidate_state.
st/mesa doesn't keep the _NEW_* flags after that.
torcs is 2% faster between the previous patch and the end of this series.
v2: - add st_atom_list.h to Makefile.sources
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
The pass I introduced in commit a2dc11a781
was entirely broken. A missing "break" made the load_interpolated_input
case always fall through to "default" and hit a "continue", making it
not actually move any load_interpolated_input intrinsics at all.
It would only move the simple load_barycentric_* intrinsics, which
don't emit any code anyway, making it basically useless.
The initial version I sent of the pass worked, but I apparently
failed to verify that the simplified version in v2 actually worked.
With the obvious fix applied (so we actually tried to move
load_interpolated_input intrinsics), I discovered a second bug: we
weren't moving the offset SSA def to the top, breaking SSA validation.
The new version of the pass actually moves load_interpolated_input
intrinsics and all their dependencies, as intended.
Papers over GPU hangs on Ivybridge and Baytrail caused by the
recent NIR FS input rework by restoring the old behavior.
(I'm not honestly sure why they hang with PLN not at the top.)
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97083
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Valgrind detected that variable ir_copy_propagation_visitor::killed_all
is uninitialized.
Signed-off-by: Jan Ziak (http://atom-symbol.net) <0xe2.0x9a.0x9b@gmail.com>
Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Jan Ziak (http://atom-symbol.net) <0xe2.0x9a.0x9b@gmail.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-and-Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Exported dmabufs can get imported by the same process, but the handle was
not getting added to the hash table on export. Add the handle to the hash
table on export.
Signed-off-by: Rob Herring <robh@kernel.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
We should use the persample_dispatch variable in prog_data.
Fixes all (~60) the DEQP sample shading tests. Many tests exited with
VK_ERROR_OUT_OF_DEVICE_MEMORY without this patch.
V2: Use the shader key bits set in brw_compile_fs (Jason)
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
android.opengl.cts.WrapperTest#testGetIntegerv1 CTS test calls
eglTerminate, followed by eglReleaseThread. A similar case is
observed in this bug: https://bugs.freedesktop.org/show_bug.cgi?id=69622,
where the test calls eglTerminate, then eglMakeCurrent(dpy, NULL, NULL, NULL).
With the current code, dri2_dpy structure is freed on eglTerminate
call, so the display is not initialized when eglReleaseThread calls
MakeCurrent with NULL parameters, to unbind the context, which
causes a a segfault in drv->API.MakeCurrent (dri2_make_current),
either in glFlush or in a latter call.
eglTerminate specifies that "If contexts or surfaces associated
with display is current to any thread, they are not released until
they are no longer current as a result of eglMakeCurrent."
However, to properly free the current context/surface (i.e., call
glFlush, unbindContext, driDestroyContext), we still need the
display vtbl (and possibly an active dri dpy connection). Therefore,
we add some reference counter to dri2_egl_display, to make sure
the structure is kept allocated as long as it is required.
One drawback of this is that eglInitialize may not completely reinitialize
the display (if eglTerminate was called with a current context), however,
this seems to meet the EGL spec quite well, and does not permanently
leak any context/display even for incorrectly written apps.
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Nicolas Boichat <drinkcat@chromium.org>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
The file was removed with earlier commit breaking 'make dist'.
Drop it from Makefile.sources since it's no longer around.
Fixes: 16985eb308 ("vc4: Switch to using the libdrm-provided
vc4_drm.h.")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
The order of optimizations can lead to the conditional discard optimization
being applied twice to the same discard statement. In this case, we must
ensure that both conditions are applied.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96762
Cc: mesa-stable@lists.freedesktop.org
Tested-by: Kai Wasserbäch <kai@dev.carbon-project.org>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
When an application declares varying arrays but does not actually do any
indirect indexing, some array indices may end up unused in the consuming
shader, so the number of input slots that correspond to the array ends
up less than the array_size.
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Without this GCC 4.8.x throws below error:
error: invalid initialization of non-const reference of type
'clover::llvm::compat::raw_ostream_to_emit_file {aka llvm::raw_svector_ostream&}'
from an rvalue of type '<brace-enclosed initializer list>'
v2: change commit title and add error message like Eric Engestrom requested
Signed-off-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97019
[ Francisco Jerez: Trivial formatting fix. ]
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
These are only used by get_matching_input() which has been call
at this point so free the hash tables.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
This exposes OpenGL 4.1 on Maxwell (tested on GM107 and GM206).
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
PFETCH, actually ISBERD on GM107+ ISA only accepts a GPR for src0.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
The number of outputs patch (limited to 255) has moved in the TCP
header, but blob seems to also set the old position. Also, the high
8-bits are now located inbetween the min/max parallel output read
address at position 20.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
In case of split primitives we need to restore
the original setting of the vtx.attrsz array to make
immediate mode attribute array tracking work.
v2: Use bool instead of boolean.
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Reviewed-by: Brian Paul <brianp@vmware.com>
Tested-by: Brian Paul <brianp@vmware.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96950
This cuts down the overhead of si_dump_shader when ddebug is capturing
shader logs, which is done for every draw call unconditionally (that's
quite a lot of work for a draw call).
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
For good performance while being able to generate decent hang reports.
The report doesn't contain the parsed IB and the buffer list, but it
isolates the draw call and dumps shaders while not having to flush
the context.
This is for GPU hangs that are harder to reproduce and require interactive
playing for minutes or even hours.
dd_pipe.h explains some implementation details. Initializing, copying
(recording) and clearing states is most of the code.
The performance should be at least 50% of the normal performance depending
on the circumstances. (i.e. 50% is expected to be the worst case scenario,
not the best case) The majority of time is spent in
dump_debug_state(PIPE_DUMP_CURRENT_SHADERS) and that's after all
the optimizations in later patches. There is no obvious way to optimize
that further.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Rob Clark
Rob Clark
Marek Olšák
Kenneth Graunke
Jan Ziak
Rob Herring
Anuj Phogat
Nicolas Boichat
Emil Velikov
Nicolai Hähnle
Dieter Nützel
Timothy Arceri
Samuel Pitoiset
Mathias Fröhlich