In a situation where double-register values are used, the phi nodes can
still end up being u32 values. They all get merged into one RA node
though. When fixing up the merge (which comes after the phi node), the
phi node's def would get fixed, but not its sources which would remain
at the low register value.
This maintains the invariant that a phi node's defs and sources are
allocated the same register.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Since intel is always going to be little-endian,
GL_UNSIGNED_INT_8_8_8_8_REV is the same as GL_UNSIGNED_BYTE for RGBA and
BGRA textures, so the same acceleration code will work. We might as well
use it.
Signed-off-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
In Piglit's EXT_framebuffer_multisample/alpha-to-coverage-dual-src-blend
test, key->nr_color_regions == 2, but the dual source blend FB write has
ir->target set to 0. So we failed to set "Last Render Target Select" on
any FB write message.
We only emit one FB write per render target, so my comment about setting
LastRT on every FB write directed at the last color region is a bit...
misinformed. According to the documentation, depth buffer writes and
scoreboard updates happen on the FB write with LastRT set, so I believe
we want to set it only once.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Cc: "10.2" <mesa-stable@lists.freedesktop.org>
This will be useful for INTEL_DEBUG=optimizer in the vec4 backend, which
needs to know whether it's currently processing a VS or GS. It isn't
worth adding virtual methods for this case.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Dropping this helps most lines fit in an 80 column terminal. The
absence of WE_normal also helps call attention to WE_all, where
something unusual is going on.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Don't assert (debug builds) or assign random uninitialized value for
predicate register (p0).. that screws up kill, etc.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Accuracy of some operations was recently improved in the R600 backend,
at the cost of slower code. This is required for compute shaders,
but not for graphics shaders. Add unsafe-fp-math hint to make LLVM
generate faster but possibly less accurate code.
Piglit didn't indicate any regressions.
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Now that we know that the pool is defragmented, we positively know
that allocated + unallocated will be the total size of the
current pool plus all the items that will be promoted. So we only
need to grow the pool once.
This will allow us to just add the new items to the end of the
item_list without the need of looking for a place to the new item.
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
This way we can avoid defragmenting the pool, even if it is needed
to defragment it, and looping again through the list of unallocated
items.
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
This patch adds a new member to the pool to track its status.
For now it is used only for the 'fragmented' status, but if
needed it could be used for more statuses.
The pool will be considered fragmented if: An item that isn't
the last is freed or demoted.
This 'strategy' has a problem, although it shouldn't cause any bug.
If for example we have two items, A and B. We choose to free A first,
now the pool will have the 'fragmented' status. If we now free B,
the pool will retain its 'fragmented' status even if it isn't
fragmented.
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
This new function will move items forward in the pool, so that
there's no gap between them, effectively defragmenting the pool.
For now this function is a bit dumb as it just moves items
forward without trying to see if other items in the pool could
fit in the gaps.
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
This function will be used in the future by compute_memory_defrag
to move items forward in the pool.
It does so by first checking for overlaping ranges, if the ranges
don't overlap it will copy the contents directly. If they overlap
it will try first to make a temporary buffer, if this buffer fails
to allocate, it will finally fall back to a mapping.
Note that it will only be needed to move items forward, it only
checks for overlapping ranges in that case. If needed, it can
easily be added by changing the first if.
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
Actually what we currently handle is just the SCALED versions, and not
the int versions. The difference probably matters more when we actually
support integer in the compiler.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Teach new compiler scheduling and register assignment how to deal with
relative addressing. This gets us what we need to avoid falling back to
old compiler for CONST[ADDR[0].x+n]. It is also a prerequisite for temp
file relative addressing, although that is going to also need some
cleverness in register assignment to keep arrays grouped together.
NOTE: doing address calculation in full precision and then narrowing to
s16 in the mov to addr reg seems to sometimes cause lockups (and
sometimes work?!). It seems more reliable to do the address calculation
in s16, like the blob does. Which means teaching RA how to deal with
mixed half and full precision allocation. Fortunately that didn't turn
out to be too hard, so that is a nice bonus which we could probably take
better advantage of elsewhere.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Technically we should not need these. CP_LOAD_STATE can be pipelined.
But removing them broke a few piglit tests, like fbo-depth-
GL_DEPTH_COMPONENT24-readpixels. I expect these are just masking a
problem elsewhere, or perhaps they are only needed under some more
specific circumstances. But until that is understood properly, give
back a bit of the perf boost we got from c63450e8.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Adds an implementation of the ClearTexSubImage driver entry point that tries
to set up an FBO to render to the texture and then calls glClearBuffer with a
scissor to perform the actual clear. If an FBO can't be created for the
texture then it will fall back to using _mesa_store_ClearTexSubImage.
When used in combination with _mesa_store_ClearTexSubImage this should provide
an implementation that works for all DRI-based drivers. However as this has
only been tested with the i965 driver it is currently only enabled there.
v2: Only enable the extension for the i965 driver instead of all DRI drivers.
Remove an unnecessary goto. Don't require GL_ARB_framebuffer_object. Add
some more comments.
v3: Use glClearBuffer* to avoid having to modify glClearColor and friends.
Handle sRGB textures. Explicitly disable dithering.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen at intel.com>
The Meta implementation of glClearTexSubImage is going to want to ensure that
dithering is disabled so that it can get a consistent color across the whole
texture when clearing. This adds a state flag to easily save it and set it to
the default value when performing meta operations.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Adds an implmentation of the ClearTexSubImage driver entry point that just
maps the texture and writes the values in. The extension is not yet enabled by
default because it doesn't work with multisample textures as they don't have a
simple linear layout.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
This adds the driver entry point for glClearTexSubImage and fills in the
_mesa_ClearTexImage and _mesa_ClearTexSubImage functions that call it.
v2: Don't clear some of the images if only one of them makes an error
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
In texture_error_check() there was a snippet of code to check whether the
given format and internal format are basically compatible. This has been split
out into its own static helper function so that it can be used by an
implementation of glClearTexImage too.
Should reduce overhead because the caching buffer manager doesn't need to
consider buffers of the wrong type.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
because float depth texture data needs clamping to [0.0, 1.0]. Let the
_mesa_texstore() fallback to slower path.
Fixes Khronos GLES3 CTS tests:
shadow_execution_vert
shadow_execution_frag
V2: Move the check to _mesa_texstore_can_use_memcpy() function.
Add check for floating point data types.
Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
We might be able to do this without an extra program key field, but this
is non-invasive and fixes the bug, for now.
This fixes the following Piglit tests on Broadwell:
- ARB_sample_shading/builtin-gl-sample-id 2
- ARB_sample_shading/builtin-gl-sample-position 2
- EXT_framebuffer_multisample/multisample-blit 2 color
- EXT_framebuffer_multisample/multisample-blit 2 color linear
- EXT_framebuffer_multisample/multisample-blit 2 depth
- EXT_framebuffer_multisample/no-color 2 depth combined
- EXT_framebuffer_multisample/no-color 2 depth separate
- EXT_framebuffer_multisample/no-color 2 depth single
- EXT_framebuffer_multisample/no-color 2 depth-computed combined
- EXT_framebuffer_multisample/no-color 2 depth-computed separate
- EXT_framebuffer_multisample/no-color 2 depth-computed single
- EXT_framebuffer_multisample/unaligned-blit 2 color msaa
- EXT_framebuffer_multisample/unaligned-blit 2 depth msaa
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=80991
Reviewed-by: Matt Turner <mattst88@gmail.com>
Cc: "10.2" <mesa-stable@lists.freedesktop.org>
Otherwise, the performance warning for shader recompiles will just say
"something else".
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
It doesn't exist, so attempting to read it will trigger generation
assertions in the brw_inst API.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Printing the hex offsets makes it basically impossible to diff assembly:
if you add even a single instruction, the entire shader shows up as a
difference. So, every time I want to compare assembly, I have to strip
this out.
The hex offsets might be useful when debugging compaction, or when
inspecting the program cache buffer. Since it's occasionally useful,
but uncommon, this patch disables it by default, but makes it easy to
re-enable it temporarily when the need arises.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Ilia Mirkin