Thierry Reding [Fri, 23 Feb 2018 13:13:27 +0000 (14:13 +0100)]
disk cache: Link with -latomic if necessary
The disk cache implementation uses 64-bit atomic operations. For some
architectures, such as 32-bit ARM, GCC will not be able to translate
these operations into atomic, lock-free instructions and will instead
rely on the external atomics library to provide these operations.
Check at configuration time whether or not linking against libatomic
is necessary and if so, create a dependency that can be used while
linking the mesautil library.
This is the meson equivalent of
2ef7f23820a6 ("configure: check if
-latomic is needed for __atomic_*").
For some background information on this, see:
https://gcc.gnu.org/wiki/Atomic/GCCMM
Changes in v2:
- clarify meaning of lock-free in commit message
- fix build if -latomic is not necessary
Acked-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Samuel Pitoiset [Thu, 1 Mar 2018 09:53:49 +0000 (10:53 +0100)]
radv: do not set pending_reset_query in BeginCommandBuffer()
This is just useless for two reasons:
1) flush_bits is not set accordingly, so nothing will be flushed
in BeginQuery().
2) we always flush caches in EndCommandBuffer(), so if a reset
is done in a previous command buffer we are safe.
Cc: "18.0" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Alex Smith <asmith@feralinteractive.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Dave Airlie [Thu, 1 Mar 2018 03:38:32 +0000 (03:38 +0000)]
r600/cayman: fix fragcood loading recip generation.
This fixes some hangs seen where the recip_ieee opcodes would
end up split across the wrong slots.
Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Kenneth Graunke [Mon, 12 Feb 2018 15:18:29 +0000 (07:18 -0800)]
i965: Allow 48-bit addressing on Gen8+.
This allows most GPU objects to use the full 48-bit address space
offered by Gen8+ platforms, rather than being stuck with 32-bit.
This expands the available GPU memory from 4G to 256TB or so.
A few objects - instruction, scratch, and vertex buffers - need to
remain pinned in the low 4GB of the address space for various reasons.
We default everything to 48-bit but disable it in those cases.
Thanks to Jason Ekstrand for blazing this trail in anv first and
finding the nasty undocumented hardware issues. This patch simply
rips off all of his findings.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Kenneth Graunke [Mon, 26 Feb 2018 23:51:04 +0000 (15:51 -0800)]
i965: Shorten the name of the workaround BO.
This makes the name shorter in debug printouts. If "workaround_bo"
is good enough for the code, it's probably good enough for debugging.
Kenneth Graunke [Tue, 28 Nov 2017 18:07:43 +0000 (10:07 -0800)]
i965: Add debugging code to dump the validation list.
When anything goes wrong with this code, dumping the validation list
is a useful way to figure out what's happening.
Jason Ekstrand [Thu, 1 Mar 2018 03:57:44 +0000 (19:57 -0800)]
intel/fs: Set up sampler message headers in the visitor on gen7+
This gives the scheduler visibility into the headers which should
improve scheduling. More importantly, however, it lets the scheduler
know that the header gets written. As-is, the scheduler thinks that a
texture instruction only reads it's payload and is unaware that it may
write to the first register so it may reorder it with respect to a read
from that register. This is causing issues in a couple of Dota 2 vertex
shaders.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104923
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Timothy Arceri [Thu, 1 Mar 2018 09:17:38 +0000 (20:17 +1100)]
ac: fix nir_intrinsic_shared_atomic_comp_swap handling
Following on from
49879f377870 this makes sure we use the correct
src index.
Fixes cts test:
KHR-GL46.compute_shader.atomic-case3
Reviewed-by: Dave Airlie <airlied@redhat.com>
Timothy Arceri [Thu, 1 Mar 2018 02:39:20 +0000 (13:39 +1100)]
st/glsl_to_nir: simplify st_nir_assign_var_locations() and fix for fs outputs
We only need to check for previously processed location on user
defined varyings as they are the only ones that support component
packing. Therefore a single instance of processed_locs can be
shared by regular varyings and patches.
For simplicity we make processed_locs an array in order to handle
dual source bleanding.
Fixes the follow piglit test on radeonsi:
tests/spec/arb_enhanced_layouts/execution/component-layout/fs-output.shader_test
Reviewed-by: Dave Airlie <airlied@redhat.com>
Jason Ekstrand [Sat, 24 Feb 2018 05:12:50 +0000 (21:12 -0800)]
anv: Enable MSAA fast-clears
This speeds up the Sascha Willems multisampling demo by around 25% when
using 8x or 16x MSAA.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Jason Ekstrand [Sat, 24 Feb 2018 05:12:35 +0000 (21:12 -0800)]
anv/cmd_buffer: Add support for MCS fast-clears and resolves
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Jason Ekstrand [Sat, 24 Feb 2018 05:00:52 +0000 (21:00 -0800)]
anv/cmd_buffer: Add helpers for computing resolve predicates
We'll want to re-use the complex resolve predicate computations for MCS
resolves so it's nice to have them as helper functions.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Jason Ekstrand [Sat, 24 Feb 2018 04:45:26 +0000 (20:45 -0800)]
anv/cmd_buffer: Handle MCS identical to CCS_E in compute_aux_usage
This doesn't actually do anything because att_state->fast_clear is
determined based on the return value of anv_layout_to_fast_clear_type
which currently returns NONE for multisampled images.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Jason Ekstrand [Sat, 24 Feb 2018 05:11:58 +0000 (21:11 -0800)]
anv/blorp: Pass the clear address to blorp for subpass MSAA resolves
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Jason Ekstrand [Sat, 24 Feb 2018 06:05:39 +0000 (22:05 -0800)]
anv/blorp: Allow indirect clear colors on blorp sources on gen7
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Jason Ekstrand [Sat, 11 Nov 2017 22:32:21 +0000 (14:32 -0800)]
anv/blorp: Add partial clear support to anv_image_mcs_op
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Jason Ekstrand [Sat, 11 Nov 2017 22:28:17 +0000 (14:28 -0800)]
intel/blorp: Add indirect clear color support to mcs_partial_resolve
This is a bit complicated because we have to get the indirect clear
color in there somehow. In order to not do any more work in the shader
than needed, we set it up as it's own vertex binding which points
directly at the clear color address specified by the client.
Acked-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Jason Ekstrand [Sat, 11 Nov 2017 21:40:03 +0000 (13:40 -0800)]
intel/blorp: Add a helper for filling out VERTEX_BUFFER_STATE
There are enough #ifs in there that it's kind-of pointless to duplicate
it for each buffer.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Andriy Khulap [Thu, 1 Mar 2018 08:44:28 +0000 (10:44 +0200)]
i965: Fix RELOC_WRITE typo in brw_store_data_imm64()
Fixes: 6c530ad11605
("i965: Reduce passing 2x32b of reloc_domains to 2 bits")
Signed-off-by: Andriy Khulap <andriy.khulap@globallogic.com>
Signed-off-by: Vadym Shovkoplias <vadym.shovkoplias@globallogic.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Jonathan Gray [Wed, 28 Feb 2018 10:21:14 +0000 (21:21 +1100)]
gallium/util: use sockets on PIPE_OS_UNIX in u_network
Instead of listing all the UNIX PIPE_OS platforms just use
PIPE_OS_UNIX. Makes BSD sockets available on PIPE_OS_BSD.
Signed-off-by: Jonathan Gray <jsg@jsg.id.au>
Reviewed-by: Brian Paul <brianp@vmware.com>
Jonathan Gray [Wed, 28 Feb 2018 10:19:19 +0000 (21:19 +1100)]
util: use clock_gettime() on PIPE_OS_BSD
OpenBSD, FreeBSD, NetBSD and DragonFlyBSD all have clock_gettime()
so use it when PIPE_OS_BSD is defined.
Signed-off-by: Jonathan Gray <jsg@jsg.id.au>
Reviewed-by: Brian Paul <brianp@vmware.com>
Jose Maria Casanova Crespo [Thu, 1 Mar 2018 17:06:52 +0000 (18:06 +0100)]
nir/search: Include 8 and 16-bit support in construct_value
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Jason Ekstrand [Wed, 28 Feb 2018 21:15:04 +0000 (13:15 -0800)]
nir/search: Support 8 and 16-bit constants in match_value
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova@igalia.com>
Andres Gomez [Wed, 28 Feb 2018 21:18:59 +0000 (23:18 +0200)]
travis: make Meson find the proper llvm-config
Travis CI has moved to LLVM 5.0, and meson is detecting automatically
the available version in /usr/local/bin based on the PATH env variable
order preference.
As for 0.44.x, Meson cannot receive the path to the llvm-config binary
as a configuration parameter. See
https://github.com/mesonbuild/meson/issues/2887 and
https://github.com/dcbaker/meson/commit/
7c8b6ee3fa42f43c9ac7dcacc61a77eca3f1bcef
We want to use the custom (APT) installed version. Therefore, let's
make Meson find our wanted version sooner than the one at
/usr/local/bin
Once this is corrected, we would still need a patch similar to:
https://lists.freedesktop.org/archives/mesa-dev/2017-December/180217.html
v2: Create the link only to the specificly wanted LLVM version (Gert).
Cc: Eric Engestrom <eric.engestrom@imgtec.com>
Cc: Dylan Baker <dylan@pnwbakers.com>
Cc: Emil Velikov <emil.velikov@collabora.com>
Cc: Juan A. Suarez Romero <jasuarez@igalia.com>
Cc: Gert Wollny <gw.fossdev@gmail.com>
Cc: Jon Turney <jon.turney@dronecode.org.uk>
Signed-off-by: Andres Gomez <agomez@igalia.com>
Reviewed-and-Tested-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Reviewed-by: Juan A. Suarez <jasuarez@igalia.com>
Reviewed-By: Gert Wollny <gw.fossdev@gmail.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Andres Gomez [Wed, 28 Feb 2018 21:15:07 +0000 (23:15 +0200)]
meson: fix LLVM version detection when <= 3.4
3 digits versions in LLVM only started from 3.4.1 on.
Hence, even if you can perfectly build with an old LLVM (< 3.4.1) in
the system while not needing LLVM at all (auto), when passing through
the LLVM version detection code, meson will fail when accessing
"_llvm_version[2]" due to:
"Index 2 out of bounds of array of size 2."
v2: Properly compare LLVM version and set patch version to 0
if < 3.4.1 (Eric).
v3: Improve the commit log explanation (Eric).
Cc: Dylan Baker <dylan@pnwbakers.com>
Cc: Eric Engestrom <eric.engestrom@imgtec.com>
Signed-off-by: Andres Gomez <agomez@igalia.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Iago Toral Quiroga [Thu, 1 Mar 2018 06:59:42 +0000 (07:59 +0100)]
i965/sbe: fix number of inputs for active components
In
16631ca30ea6 we fixed gen9 active components to account for padded
inputs in the URB, which we can have with SSO programs. To do that,
instead of going through the bitfield of inputs (which doesn't include
padding information), we compute the number of inputs from the size
of the URB entry.
Unfortunately, there are some special inputs that are not stored in
the URB and that we also need to account for. These special inputs
are identified and handled during calculate_attr_overrides().
Instead of keeping track of the exact number of inputs, we just
program active components for all possible inputs like we do in
anvil.
This fixes a regression in a WebGL program that uses Point Sprite
functionality (specifically, VARYING_SLOT_PNTC).
v2:
- Add 'Fixes' tag (Mark Janes)
- make no_vue_inputs int instead of uint32_t, and add const qualifier
to num_inputs variable (Ian)
v3:
- Do not try to count inputs correctly, just program all input
slots like we do in anvil (Ken)
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105224
Fixes: 16631ca30ea6 (i965/sbe: fix active components for SSO programs with over 16 inputs)
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Samuel Pitoiset [Wed, 28 Feb 2018 19:28:53 +0000 (20:28 +0100)]
radv: only emit cache flushes when the pool size is large enough
This is an optimization which reduces the number of flushes for
small pool buffers.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Samuel Pitoiset [Wed, 28 Feb 2018 19:22:29 +0000 (20:22 +0100)]
radv: keep track of the query pool size
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Samuel Pitoiset [Wed, 28 Feb 2018 20:47:11 +0000 (21:47 +0100)]
radv: make sure to emit cache flushes before starting a query
If the query pool has been previously resetted using the compute
shader path.
Fixes: a41e2e9cf5 ("radv: allow to use a compute shader for resetting the query pool")
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105292
Cc: "18.0" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Alejandro Piñeiro [Wed, 28 Feb 2018 12:01:56 +0000 (13:01 +0100)]
nir/serialize: handle var->name being NULL
var->name could be NULL under ARB_gl_spirv for example. And in any
case, the code is already handing var name being NULL when reading a
variable, so it is consistent to do it writing a variable too.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Jose Maria Casanova Crespo [Fri, 23 Feb 2018 00:15:13 +0000 (01:15 +0100)]
anv: Enable VK_KHR_16bit_storage for PushConstant
Enables storagePushConstant16 features of VK_KHR_16bit_storage for Gen8+.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Jose Maria Casanova Crespo [Tue, 20 Feb 2018 09:28:41 +0000 (10:28 +0100)]
spirv/i965/anv: Relax push constant offset assertions being 32-bit aligned
The introduction of 16-bit types with VK_KHR_16bit_storages implies that
push constant offsets could be multiple of 2-bytes. Some assertions are
updated so offsets should be just multiple of size of the base type but
in some cases we can not assume it as doubles aren't aligned to 8 bytes
in some cases.
For 16-bit types, the push constant offset takes into account the
internal offset in the 32-bit uniform bucket adding 2-bytes when we access
not 32-bit aligned elements. In all 32-bit aligned cases it just becomes 0.
v2: Assert offsets to be aligned to the dest type size. (Jason Ekstrand)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Jose Maria Casanova Crespo [Thu, 22 Feb 2018 16:36:37 +0000 (17:36 +0100)]
spirv: Calculate properly 16-bit vector sizes
Range in 16-bit push constants load was being calculated
wrongly using 4-bytes per element instead of 2-bytes as it
should be.
v2: Use glsl_get_bit_size instead of if statement
(Jason Ekstrand)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Jose Maria Casanova Crespo [Mon, 20 Nov 2017 22:28:45 +0000 (23:28 +0100)]
anv: Enable VK_KHR_16bit_storage for SSBO and UBO
Enables storageBuffer16BitAccess and uniformAndStorageBuffer16BitAccesss
features of VK_KHR_16bit_storage for Gen8+.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Jose Maria Casanova Crespo [Wed, 31 Jan 2018 23:26:04 +0000 (00:26 +0100)]
i965/fs: Support 16-bit store_ssbo with VK_KHR_relaxed_block_layout
Restrict the use of untyped_surface_write with 16-bit pairs in
ssbo to the cases where we can guarantee that offset is multiple
of 4.
Taking into account that VK_KHR_relaxed_block_layout is available
in ANV we can only guarantee that when we have a constant offset
that is multiple of 4. For non constant offsets we will always use
byte_scattered_write.
v2: (Jason Ekstrand)
- Assert offset_reg to be multiple of 4 if it is immediate.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Jose Maria Casanova Crespo [Wed, 31 Jan 2018 23:05:11 +0000 (00:05 +0100)]
i965/fs: Support 16-bit do_read_vector with VK_KHR_relaxed_block_layout
16-bit load_ubo/ssbo operations that call do_untyped_read_vector don't
guarantee that offsets are multiple of 4-bytes as required by untyped_read
message. This happens for example in the case of f16mat3x3 when then
VK_KHR_relaxed_block_layout is enabled.
Vectors reads when we have non-constant offsets are implemented with
multiple byte_scattered_read messages that not require 32-bit aligned offsets.
Now for all constant offsets we can use the untyped_read_surface message.
In the case of constant offsets not aligned to 32-bits, we calculate a
start offset 32-bit aligned and use the shuffle_32bit_load_result_to_16bit_data
function and the first_component parameter to skip the copy of the unneeded
component.
v2: (Jason Ekstrand)
Use untyped_read_surface messages always we have constant offsets.
v3: (Jason Ekstrand)
Simplify loop for reads with non constant offsets.
Use end - start to calculate the number of 32-bit components to read with
constant offsets.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Jose Maria Casanova Crespo [Mon, 26 Feb 2018 19:28:34 +0000 (20:28 +0100)]
i965/fs: shuffle_32bit_load_result_to_16bit_data now skips components
This helper used to load 16bit components from 32-bits read now allows
skipping components with the new parameter first_component. The semantics
now skip components until we reach the first_component, and then reads the
number of components passed to the function.
All previous uses of the helper are updated to use 0 as first_component.
This will allow read 16-bit components when the first one is not aligned
32-bit. Enabling more usages of untyped_reads with 16-bit types.
v2: (Jason Ektrand)
Change parameters order to first_component, num_components
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Jose Maria Casanova Crespo [Tue, 30 Jan 2018 08:59:34 +0000 (09:59 +0100)]
isl/i965/fs: SSBO/UBO buffers need size padding if not multiple of 32-bit
The surfaces that backup the GPU buffers have a boundary check that
considers that access to partial dwords are considered out-of-bounds.
For example, buffers with 1,3 16-bit elements has size 2 or 6 and the
last two bytes would always be read as 0 or its writting ignored.
The introduction of 16-bit types implies that we need to align the size
to 4-bytew multiples so that partial dwords could be read/written.
Adding an inconditional +2 size to buffers not being multiple of 2
solves this issue for the general cases of UBO or SSBO.
But, when unsized arrays of 16-bit elements are used it is not possible
to know if the size was padded or not. To solve this issue the
implementation calculates the needed size of the buffer surfaces,
as suggested by Jason:
surface_size = isl_align(buffer_size, 4) +
(isl_align(buffer_size, 4) - buffer_size)
So when we calculate backwards the buffer_size in the backend we
update the resinfo return value with:
buffer_size = (surface_size & ~3) - (surface_size & 3)
It is also exposed this buffer requirements when robust buffer access
is enabled so these buffer sizes recommend being multiple of 4.
v2: (Jason Ekstrand)
Move padding logic fron anv to isl_surface_state.
Move calculus of original size from spirv to driver backend.
v3: (Jason Ekstrand)
Rename some variables and use a similar expresion when calculating.
padding than when obtaining the original buffer size.
Avoid use of unnecesary component call at brw_fs_nir.
v4: (Jason Ekstrand)
Complete comment with buffer size calculus explanation in brw_fs_nir.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Mathias Fröhlich [Sun, 25 Feb 2018 17:01:07 +0000 (18:01 +0100)]
vbo: Remove vbo_save_vertex_list::vertex_size.
Like before use local variables from compile_vertex_list instead.
Remove vertex_size from struct vbo_save_vertex_list.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Sun, 25 Feb 2018 17:01:07 +0000 (18:01 +0100)]
vbo: Remove vbo_save_vertex_list::buffer_offset.
The buffer_offset is used in aligned_vertex_buffer_offset.
But now that most of these decisions are done in compile_vertex_list
we can work on local variables instead of struct members in the
display list code. Clean that up and remove buffer_offset.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Sun, 25 Feb 2018 17:01:07 +0000 (18:01 +0100)]
vbo: Remove vbo_save_vertex_list::start_vertex.
Replace last use on replay with _vbo_save_get_{min,max}_index. Appart from
that it is not used anymore.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Sun, 25 Feb 2018 17:01:07 +0000 (18:01 +0100)]
vbo: Remove vbo_save_vertex_list::attrsz.
Is not used anymore on replay, move the last use in display list
compilation to the original array in the display list compiler.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Sun, 25 Feb 2018 17:01:07 +0000 (18:01 +0100)]
vbo: Remove vbo_save_vertex_list::attrtype.
Is not used anymore on replay, move the last use in display list
compilation to the original array in the display list compiler.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Sun, 25 Feb 2018 17:01:07 +0000 (18:01 +0100)]
vbo: Remove vbo_save_vertex_list::enabled.
Is not used anymore on replay.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Sun, 25 Feb 2018 17:01:07 +0000 (18:01 +0100)]
vbo: Remove reference to the vertex_store from the dlist node.
Since we now store a set of VAOs in the display list, use these object
to get the reference to the VBO in several places.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Sun, 25 Feb 2018 17:01:07 +0000 (18:01 +0100)]
vbo: Implement current values update in terms of the VAO.
Use the information already present in the VAO to update the current values
after display list replay. Set GL_OUT_OF_MEMORY on allocation failure
for the current value update storage.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Sun, 25 Feb 2018 17:01:07 +0000 (18:01 +0100)]
vbo: Implement vbo_loopback_vertex_list in terms of the VAO.
Use the information already present in the VAO to replay a display list
node using immediate mode draw commands. Use a hand full of helper methods
that will be useful for the next patches also.
v2: Insert asserts, constify local variables.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Sun, 25 Feb 2018 17:01:07 +0000 (18:01 +0100)]
vbo: Use a local variable for the dlist offsets.
The master value is now stored inside the VAO already present in
struct vbo_save_vertex_list. Remove the unneeded copy from dlist storage.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Sun, 25 Feb 2018 17:01:07 +0000 (18:01 +0100)]
vbo: Remove unused vbo_save_context::wrap_count.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Mathias Fröhlich [Sun, 25 Feb 2018 17:01:07 +0000 (18:01 +0100)]
vbo: Remove unused vbo_save_vertex_list::dangling_attr_ref.
Reviewed-by: Brian Paul <brianp@vmware.com>
Signed-off-by: Mathias Fröhlich <Mathias.Froehlich@web.de>
Jason Ekstrand [Wed, 28 Feb 2018 23:25:48 +0000 (15:25 -0800)]
anv: Always set has_context_priority
We don't zalloc the physical device so we need to unconditionally set
everything. Crucible helpfully initializes all allocations to 139 so it
was getting true regardless of whether or not the kernel actually
supports context priorities.
Fixes: 6d8ab53303331 "anv: implement VK_EXT_global_priority extension"
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Mark Janes [Thu, 1 Mar 2018 01:26:08 +0000 (17:26 -0800)]
Revert "i965: Only emit 3DSTATE_DRAWING_RECTANGLE once on gen8+"
This reverts commit
a2c1e48f15995a826dc759e064c2603882a37e0c.
On BDWGT3e and KBLGT3e systems, this commit regressed the following
tests:
piglit.spec.ext_framebuffer_multisample.accuracy 2 stencil_resolve small depthstencil
piglit.spec.ext_framebuffer_multisample.accuracy 4 stencil_resolve small depthstencil
piglit.spec.ext_framebuffer_multisample.accuracy 6 stencil_resolve small depthstencil
piglit.spec.ext_framebuffer_multisample.accuracy 8 stencil_resolve small depthstencil
piglit.spec.ext_framebuffer_multisample.accuracy all_samples stencil_resolve small depthstencil
Dave Airlie [Thu, 1 Mar 2018 00:01:33 +0000 (10:01 +1000)]
radeonsi/nir: increase values to 8 for gs fetch.
This stops a crash when running (still fails):
tests/spec/arb_gpu_shader_fp64/execution/explicit-location-gs-fs-vs.shader_test
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Mon, 26 Feb 2018 20:52:49 +0000 (21:52 +0100)]
radv: Use the syncobj wait ioctl to wait on fences if possible.
Handles the !waitAll and signal after the start of the wait cases correctly.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Mon, 26 Feb 2018 22:48:27 +0000 (23:48 +0100)]
radv: Implement more efficient !waitAll fence waiting.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Mon, 26 Feb 2018 21:54:06 +0000 (22:54 +0100)]
radv: Implement waiting on non-submitted fences.
Fixes: f4e499ec79 "radv: add initial non-conformant radv vulkan driver"
Reviewed-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Mon, 26 Feb 2018 21:50:41 +0000 (22:50 +0100)]
radv: Implement WaitForFences with !waitAll.
Nothing to do except using a busy wait loop. At least for old kernels.
A better implementation for newer kernels to come later.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105255
Fixes: f4e499ec79 "radv: add initial non-conformant radv vulkan driver"
Reviewed-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Wed, 28 Feb 2018 23:38:19 +0000 (09:38 +1000)]
ac/nir: fix shared atomic operations.
The nir->llvm conversion was using the wrong srcs.
Fixes:
tests/spec/arb_compute_shader/execution/shared-atomics.shader_test
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Wed, 28 Feb 2018 23:24:01 +0000 (09:24 +1000)]
ac/nir: don't apply slice rounding on txf_ms
This matches the tgsi code.
Fixes arb_texture_multisample texelFetch piglit tests.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Fixes: f4e499ec7914 (radv: add initial non-conformant radv vulkan driver)
Signed-off-by: Dave Airlie <airlied@redhat.com>
Timothy Arceri [Tue, 13 Feb 2018 02:06:51 +0000 (13:06 +1100)]
radeonsi: set some context vars for nir path
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Timothy Arceri [Thu, 1 Feb 2018 21:50:09 +0000 (08:50 +1100)]
gallium: remove llvm from ir struct
This was added in
425dc4c4b366 but never used. Also since
100796c15c3a native has superseded llvm.
Acked-by: Dave Airlie <airlied@redhat.com>
Kenneth Graunke [Wed, 28 Feb 2018 21:22:22 +0000 (13:22 -0800)]
i965: Don't emit MOVs with undefined registers for Gen4 point clipping.
Gen4 point clipping calls brw_clip_tri_alloc_regs with nr_verts == 0,
which means that c->reg.vertex[] isn't initialized. It then emits MOVs
to stomp components of those uninitialized registers to 0.
This started causing assertions after Matt's recent series, when those
uninitialized registers started getting BRW_REGISTER_TYPE_NF, which
definitely doesn't exist on Gen4-5.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Eric Anholt [Fri, 23 Feb 2018 23:35:25 +0000 (15:35 -0800)]
broadcom/vc5: Fix regression in the page-cache slice size alignment.
We need to align the size of the slice, not the offset of the next slice.
Fixes KHR-GLES3.texture_repeat_mode.rgba32ui_11x131_2_clamp_to_edge.
Fixes: b4b4ada7616d ("broadcom/vc5: Fix layout of 3D textures.")
Jason Ekstrand [Fri, 3 Nov 2017 17:36:32 +0000 (10:36 -0700)]
i965: Only emit 3DSTATE_DRAWING_RECTANGLE once on gen8+
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Jason Ekstrand [Fri, 3 Nov 2017 21:13:08 +0000 (14:13 -0700)]
i965: Be more clever about setting up our viewport clip
Before, we were trusting in the hardware to take the intersection
of the viewport clip with the drawing rectangle. Unfortunately,
3DSTATE_DRAWING_RECTANGLE is fairly expensive because it implicitly
does a full pipeline stall. If we're a bit more careful with our
viewport clipping, we can just re-emit it once at context creation
time.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Wed, 28 Feb 2018 21:25:21 +0000 (13:25 -0800)]
intel/compiler: Re-add .vs_inputs_dual_locations = true
Looks like a rebase mistake.
Fixes: 89fe5190a256 ("intel/compiler: Lower flrp32 on Gen11+")
Dave Airlie [Wed, 28 Feb 2018 06:42:53 +0000 (06:42 +0000)]
r600/shader: when using images always load thread id gpr at start (v2)
The delayed loading code was fail if we had control flow.
This fixes:
tests/spec/arb_shader_image_load_store/execution/image_checkerboard.shader_test
v2: don't use temp_reg before setting temp_reg up.
Tested-by: Gert Wollny <gw.fossdev@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Wed, 28 Feb 2018 20:15:30 +0000 (20:15 +0000)]
r600: fix whitespace in recent 1d texture commit.
trivial fix.
Matt Turner [Mon, 29 Jan 2018 23:52:39 +0000 (15:52 -0800)]
intel/compiler: Add ICL to test_eu_validate.cpp
With the Align16 tests now disabled, we can run the rest of the tests in
ICL mode (and see them pass!)
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Thu, 8 Feb 2018 18:23:11 +0000 (10:23 -0800)]
intel/compiler: Disable Align16 tests on Gen11+
Align16 is no more.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Wed, 14 Jun 2017 23:43:05 +0000 (16:43 -0700)]
intel/compiler: Add instruction compaction support on Gen11
Gen11 only differs from SKL+ in that it uses a new datatype index table.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Wed, 14 Jun 2017 23:14:11 +0000 (16:14 -0700)]
intel/compiler: Mark line, pln, and lrp as removed on Gen11+
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Wed, 14 Jun 2017 23:20:41 +0000 (16:20 -0700)]
intel/compiler: Lower flrp32 on Gen11+
The LRP instruction is no more.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Fri, 16 Jun 2017 00:29:16 +0000 (17:29 -0700)]
intel/compiler/fs: Implement ddy without using align16 for Gen11+
Align16 is no more. We previously generated an align16 ADD instruction
to calculate DDY:
add(16) g25<1>F -g23<4>.xyxyF g23<4>.zwzwF { align16 1H };
Without align16, we now implement it as:
add(4) g25<1>F -g23<0,2,1>F g23.2<0,2,1>F { align1 1N };
add(4) g25.4<1>F -g23.4<0,2,1>F g23.6<0,2,1>F { align1 1N };
add(4) g26<1>F -g24<0,2,1>F g24.2<0,2,1>F { align1 1N };
add(4) g26.4<1>F -g24.4<0,2,1>F g24.6<0,2,1>F { align1 1N };
where only the first two instructions are needed in SIMD8 mode.
Note: an earlier version of the patch implemented this in two
instructions in SIMD16:
add(8) g25<2>F -g23<4,2,0>F g23.2<4,2,0>F { align1 1N };
add(8) g25.1<2>F -g23.1<4,2,0>F g23.3<4,2,0>F { align1 1N };
but I realized that the channel enable bits will not be correct. If we
knew we were under uniform control flow, we could emit only those two
instructions however.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Fri, 16 Jun 2017 00:20:29 +0000 (17:20 -0700)]
intel/compiler/fs: Simplify ddx/ddy code generation
The brw_reg() constructor just obfuscates things here, in my opinion.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Thu, 15 Jun 2017 22:41:40 +0000 (15:41 -0700)]
intel/compiler/fs: Pass fs_inst to generate_ddx/ddy instead of opcode
In a future patch, generate_ddy will want to inspect inst->exec_size.
Change generate_ddx as well for consistency.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Mon, 23 Oct 2017 17:44:39 +0000 (10:44 -0700)]
intel/compiler/fs: Don't generate integer DWord multiply on Gen11
Like CHV et al., Gen11 does not support 32x32 -> 32/64-bit integer
multiplies.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Wed, 14 Jun 2017 21:47:19 +0000 (14:47 -0700)]
intel/compiler/fs: Implement FS_OPCODE_LINTERP with MADs on Gen11+
The PLN instruction is no more. Its functionality is now implemented
using two MAD instructions with the new native-float type. Instead of
pln(16) r20.0<1>:F r10.4<0;1,0>:F r4.0<8;8,1>:F
we now have
mad(8) acc0<1>:NF r10.7<0;1,0>:F r4.0<8;8,1>:F r10.4<0;1,0>:F
mad(8) r20.0<1>:F acc0<8;8,1>:NF r5.0<8;8,1>:F r10.5<0;1,0>:F
mad(8) acc0<1>:NF r10.7<0;1,0>:F r6.0<8;8,1>:F r10.4<0;1,0>:F
mad(8) r21.0<1>:F acc0<8;8,1>:NF r7.0<8;8,1>:F r10.5<0;1,0>:F
... and in the case of SIMD8 only the first pair of MAD instructions is
used.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Wed, 14 Jun 2017 18:06:45 +0000 (11:06 -0700)]
intel/compiler/fs: Return multiple_instructions_emitted from generate_linterp
If multiple instructions are emitted, special handling of things like
conditional mod and NoDDClr/NoDDChk need to be performed.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Wed, 14 Jun 2017 21:47:19 +0000 (14:47 -0700)]
intel/compiler/fs: Fix application of cmod and saturate to LINE/MAC pair
This isn't technically broken, but the next patch will make this
function report whether it generated multiple instructions, and that
information will be used to disable the application of conditional mod
by the generic code.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Wed, 14 Jun 2017 18:03:19 +0000 (11:03 -0700)]
intel/compiler: Add Gen11+ native float type
This new type exposes the additional precision offered by the
accumulator register and will be used in the next patch to implement the
functionality of the PLN instruction using a pair of MAD instructions.
One weird thing to note: align1 ternary instructions may only have an
accumulator in the dst or src1 normally, but when src0's type is :NF
the accumulator is read.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Fri, 25 Aug 2017 16:50:29 +0000 (09:50 -0700)]
intel/compiler: Add Gen11 register types
The hardware register types' encodings have changed on Gen11. Good thing
we have that superfluous looking brw_reg_type abstraction lying around!
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Mon, 11 Dec 2017 21:59:13 +0000 (13:59 -0800)]
intel: Disable 64-bit extensions on platforms without 64-bit types
Gen11 does not support DF, Q, UQ types in hardware. As a result, we have
to disable some GL extensions until they can be reimplemented.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Anuj Phogat [Wed, 10 May 2017 22:26:51 +0000 (15:26 -0700)]
intel: Add icl pci id for INTEL_DEVID_OVERRIDE
Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Matt Turner [Mon, 26 Feb 2018 22:25:17 +0000 (14:25 -0800)]
i965: Warn about preliminary support for Gen11
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Anuj Phogat [Tue, 14 Mar 2017 21:43:34 +0000 (14:43 -0700)]
intel: Add a preliminary device for Ice Lake
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Anuj Phogat <anuj.phogat@intel.com>
Tapani Pälli [Wed, 28 Feb 2018 16:54:24 +0000 (18:54 +0200)]
anv: remove anv_gem_set_context_priority helper
anv_gem_set_context_param is to be used directly instead!
Fixes: 6d8ab53303 "anv: implement VK_EXT_global_priority extension"
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
George Kyriazis [Tue, 20 Feb 2018 06:07:57 +0000 (00:07 -0600)]
swr/rast: revert clip distance precision
Fixes piglit tests that broke with
8a64593bde
Reviewed-By: Bruce Cherniak <bruce.cherniak@intel.com>
George Kyriazis [Wed, 7 Feb 2018 18:24:23 +0000 (12:24 -0600)]
swr/rast: Faster frustum prim culling
Fix clipper validMask setting. We don't need to run frustum rejected
primitives through the clipper. Perform frustum culling with only
frustum clip codes. Guardband clip codes cannot be used because they
overlap frustum codes.
Reviewed-By: Bruce Cherniak <bruce.cherniak@intel.com>
George Kyriazis [Wed, 14 Feb 2018 07:13:13 +0000 (01:13 -0600)]
swr/rast: Consolidate TRANSLATE_ADDRESS
Translate is now part of an overloaded LOAD call which required a change to
the code gen to skip the load functions in order to handle them manually
to make them virtual.
Reviewed-By: Bruce Cherniak <bruce.cherniak@intel.com>
George Kyriazis [Wed, 14 Feb 2018 01:22:03 +0000 (19:22 -0600)]
swr/rast: Code generation cleanup
Generate more compact code from gen_llvm.hpp.
Reviewed-By: Bruce Cherniak <bruce.cherniak@intel.com>
George Kyriazis [Tue, 13 Feb 2018 23:38:55 +0000 (17:38 -0600)]
swr/rast: Remove draw type from event definitions
- Have the draw type sent to DrawInfoEvent in handlers created in
archrast.cpp. The draw type no longer needs to be sent during during
AR_API_EVENT() call in api.cpp.
- Remove draw type from event defintions in events_private.proto, no
longer needed
Reviewed-By: Bruce Cherniak <bruce.cherniak@intel.com>
George Kyriazis [Tue, 27 Feb 2018 17:34:45 +0000 (11:34 -0600)]
swr/rast: whitespace change
Reviewed-By: Bruce Cherniak <bruce.cherniak@intel.com>
George Kyriazis [Fri, 16 Feb 2018 17:14:50 +0000 (11:14 -0600)]
swr/rast: Fix index buffer overfetch issue for non-indexed draws
Populate pLastIndex, even for the non-indexed case. An zero pLastIndex
can cause the index offsets inside the fetcher to have non-sensical values
that can be either very large positive or very large negative numbers.
Reviewed-By: Bruce Cherniak <bruce.cherniak@intel.com>
Roland Scheidegger [Wed, 28 Feb 2018 03:28:29 +0000 (04:28 +0100)]
softpipe: don't iterate through PIPE_MAX_SHADER_SAMPLER_VIEWS
We were setting view to NULL if the iteration was larger than i.
But in fact if the view is NULL the code did nothing anyway...
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Roland Scheidegger [Wed, 28 Feb 2018 02:01:23 +0000 (03:01 +0100)]
cso: don't cycle through PIPE_MAX_SHADER_SAMPLER_VIEWS on context destroy
There's no point, we know the highest non-null one.
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Roland Scheidegger [Sun, 25 Feb 2018 03:26:37 +0000 (04:26 +0100)]
draw: don't needlessly iterate through all sampler view slots
We already stored the highest (potentially) used number.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Tapani Pälli [Tue, 23 Jan 2018 12:01:00 +0000 (14:01 +0200)]
anv: implement VK_EXT_global_priority extension
v2: add ANV_CONTEXT_REALTIME_PRIORITY (Chris)
use unreachable with unknown priority (Samuel)
v3: add stubs in gem_stubs.c (Emil)
use priority defines from gen_defines.h
v4: cleanup, add anv_gem_set_context_param (Jason)
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com> (v2)
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> (v2)
Reviewed-by: Emil Velikov <emil.velikov@collabora.com> (v3)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Tapani Pälli [Mon, 22 Jan 2018 06:22:53 +0000 (08:22 +0200)]
i965: use context priority definitions from gen_defines.h
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Tapani Pälli [Mon, 22 Jan 2018 06:17:50 +0000 (08:17 +0200)]
intel: add new common header gen_defines.h
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>