Commit graph

3738 commits

Author SHA1 Message Date
Samuel Pitoiset
c66021069e radv/gfx10: implement a GE bug workaround
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-31 12:14:29 +02:00
Samuel Pitoiset
9a3fc7b6fa radv/gfx10: remove an obsolete VGT_REUSE_OFF workaround
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-31 12:14:29 +02:00
Samuel Pitoiset
bb8f25233a radv/gfx10: disable LATE_ALLOC_GS on Navi14
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-31 12:14:29 +02:00
Samuel Pitoiset
e041a74588 radv/gfx10: implement a bug workaround for GE_PC_ALLOC
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-31 12:14:29 +02:00
Samuel Pitoiset
0e1724af61 radv/gfx10: implement a bug workaround for NGG -> legacy transitions
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-31 12:14:29 +02:00
Samuel Pitoiset
29cca5f381 radv: skip draw calls with 0-sized index buffers
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-31 12:14:29 +02:00
Eric Engestrom
abc226cf41 tree-wide: replace MAYBE_UNUSED with ASSERTED
Suggested-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
2019-07-31 09:41:05 +01:00
Eric Engestrom
aed15fa799 radv: drop incorrect MAYBE_UNUSED
`compressed` is clearly always used on the line right after.

Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
2019-07-31 09:41:05 +01:00
Samuel Pitoiset
ea38565011 radv/gfx10: add Wave32 support for compute shaders
It can be enabled with RADV_PERFTEST=cswave32.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-31 09:35:04 +02:00
Marek Olšák
033c39a660 ac/nir: fix incorrect Phis if callbacks use control flow inside control flow 2019-07-30 22:06:23 -04:00
Marek Olšák
d3c80733cd ac/nir: handle abs modifier 2019-07-30 22:06:23 -04:00
Marek Olšák
efe2d8c5f9 ac: fix a memory leak in the error path of ac_build_type_name_for_intr 2019-07-30 22:06:23 -04:00
Marek Olšák
f6eca14f1b ac: allow control flow statements in NIR callbacks
This fixes a crash when compiling geometry shaders on radeonsi.
2019-07-30 22:06:23 -04:00
Marek Olšák
bfea7e4d29 ac/nir: handle negate modifier 2019-07-30 22:06:23 -04:00
Marek Olšák
9234275320 radeonsi/nir: implement FBFETCH for KHR_blend_equation_advanced 2019-07-30 22:06:23 -04:00
Marek Olšák
17021efc74 radeonsi: adjust RB+ blend optimization settings
based on PAL
2019-07-30 22:06:23 -04:00
Marek Olšák
27ac9a3326 ac/surface: allow linear swizzle mode automatic selection on gfx9 & 10
let addrlib make the decision to get the same result as PAL.
2019-07-30 22:06:23 -04:00
Bas Nieuwenhuizen
2b53c49d2f radv: Fix descriptor set allocation failure.
Set all the handles to VK_NULL_HANDLE:

"If the creation of any of those descriptor sets fails, then the implementation
must destroy all successfully created descriptor set objects from this command,
set all entries of the pDescriptorSets array to VK_NULL_HANDLE and return the
error."

(Vulkan 1.1.117 Spec, section 13.2)

CC: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Dave Airlie <airlied@redhat.com>
2019-07-30 22:33:24 +02:00
Andres Rodriguez
2b71b4e793 radv: fix queries with WAIT_BIT returning VK_NOT_READY
When vkGetQueryPoolResults() is called with VK_QUERY_RESULT_WAIT_BIT
set, the driver is supposed to wait for the query to become available
before returning.

Currently, radv returns once the query is indeed ready, but it returns
VK_NOT_READY. It also fails to populate the results.

The problem is a missing volatile in the secondary check for query
availability. This patch removes the secondary check altogether since it
is redundant with the preceding loop.

This bug was found with an unreleased version of SteamVR.

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-27 10:19:19 -04:00
Samuel Pitoiset
9d7ead6f9b radv/gfx10: only compile the GS copy shader on-demand
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-30 16:51:30 +02:00
Daniel Schürmann
45638e14fb radv: Don't include radv_private.h from radv_shader.h
This patch decouples radv_shader.h from any LLVM dependency.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-30 10:29:11 +02:00
Marek Olšák
7708540363 amd: add support for Arcturus
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
2019-07-29 17:52:54 -04:00
Marek Olšák
19d04191c4 radeonsi: add support for compute-only chips
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
2019-07-29 17:52:51 -04:00
Samuel Pitoiset
372c3dcfdb radv: implement VK_EXT_index_type_uint8
Natively supported on VI+.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-29 23:36:53 +02:00
Bas Nieuwenhuizen
aac492901a radv: Take variable descriptor counts into account for buffer entries.
Fixes: b5e04e9217 "radv: Support allocating variable size descriptor sets."
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111019
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
2019-07-29 20:42:53 +02:00
Samuel Pitoiset
58ee973e87 radv/gfx10: do not use the fast depth or stencil clear bytes path
It causes issues on GFX10.

This fixes rendering issues with vkmark and Wreckfest at least.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl
2019-07-29 14:47:13 +02:00
Samuel Pitoiset
4aa450193b ac: do not crash when the buffer data format is invalid
This might happen when a pipeline doesn't define the vertex input
state, so the buffer data format is 0 (aka INVALID).

This fixes crashes when compiling some shaders on GFX10.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-29 13:19:32 +02:00
Rhys Perry
a9f58af454 ac/nir: fix txf_ms with an offset
Seems to fix some hair artifacts in Max Payne 3:
https://github.com/daniel-schuermann/mesa/issues/76

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Fixes: f4e499ec79 ('radv: add initial non-conformant radv vulkan driver')
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
2019-07-29 11:50:13 +01:00
Connor Abbott
a69ab1b7d2 radv: Delete unused local variables in optimization loop
Totals from affected shaders:
SGPRS: 376 -> 376 (0.00 %)
VGPRS: 620 -> 560 (-9.68 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 292 -> 292 (0.00 %) dwords per thread
Code Size: 20024 -> 20144 (0.60 %) bytes
LDS: 0 -> 0 (0.00 %) blocks
Max Waves: 25 -> 25 (0.00 %)
Wait states: 0 -> 0 (0.00 %)

Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-29 11:37:46 +02:00
Eric Engestrom
d2de5b6ba2 anv+tu+radv: delete unusable dev_icd.json
As per previous commit, Meson doesn't support using uninstalled libs,
they're simply not ready until `ninja install` is ran, so delete them.

Suggested-by: Jason Ekstrand <jason@jlekstrand.net>
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net> # for anv
Reviewed-by: Eric Anholt <eric@anholt.net> # for tu
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl> # for radv
2019-07-26 14:47:53 +00:00
Bas Nieuwenhuizen
7e1fe81f56 radv: Set correct metadata size for GFX9+.
Without correct size, radeonsi assumes the metadata is incorrect,
which can and will cause issues.

Since the metadata is really incorrect without the size, let us
fix that.

Fixes: e43cc3e3af "radv/gfx9: handle GFX9 opaque metadata"
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
2019-07-25 17:07:53 +02:00
Samuel Pitoiset
7d11bf2155 radv/gfx10: fix intensity formats by setting ALPHA_IS_ON_MSB
This fixes
dEQP-VK.rasterization.primitive_size.points.point_size_*

This also fixes some black squares with the Sascha SSAO demo.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-25 15:48:24 +02:00
Samuel Pitoiset
6a504ab473 radv/gfx10: use L2 for DMA copy/fill operations
It's coherent and faster. GFX7-GFX9 should also support this but
for now only uses L2 for GFX10 because it's untested on previous gens.

This fixes dEQP-VK.memory.pipeline_barrier.transfer_*

This also fixes some missing geometry in Dawn Of War III because
VBOs weren't updated correctly.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-25 15:48:21 +02:00
Samuel Pitoiset
4389e85dc9 radv/gfx10: enable VK_EXT_transform_feedback
When a pipeline uses transform feedback, the driver fallbacks to
the legacy path because NGG support for streamout is a non-trivial
amount of work.

AMDVLK also uses the legacy path for streamout, while RadeonSI
uses the new NGG path.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-24 08:23:37 +02:00
Samuel Pitoiset
a3a4fa1860 radv/gfx10: do not enable NGG if a pipeline uses XFB
NGG GS for streamout requires a bunch of work, so enable it with
the legacy path only for now.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-24 08:23:34 +02:00
Samuel Pitoiset
09abe571a2 radv/gfx10: emit streamout shader config
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-24 08:23:32 +02:00
Samuel Pitoiset
383c2e625a radv/gfx10: declare streamout user SGPRs
Required for legacy streamout.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-24 08:23:30 +02:00
Samuel Pitoiset
fd195d8085 radv/gfx10: update streamout descriptors
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-24 08:23:27 +02:00
Samuel Pitoiset
ea337c8b7e radv/gfx10: fix VS input VGPRs with the legacy path
For some reasons, InstanceID is VGPR3 although StepRate0 is set to 1.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-24 08:23:21 +02:00
Bas Nieuwenhuizen
e5b3f0a867 radv/gfx10: Enable binning.
Numbers for Talos:

gfx10 without binning: 77.0 77.7 77.2 77.6
gfx10 with binning: 82.3 82.0 82.7 82.4

Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
2019-07-23 21:26:59 +02:00
Bas Nieuwenhuizen
3268c806fb radv/gfx10: Implement bin size calculation.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
2019-07-23 21:26:59 +02:00
Bas Nieuwenhuizen
4b757697e9 radv/gfx9: Select between depth/color bins based on area.
Mirrors radeonsi.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
2019-07-23 21:26:59 +02:00
Bas Nieuwenhuizen
22f2f76789 radv: Generalize binning settings.
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
2019-07-23 21:26:59 +02:00
Bas Nieuwenhuizen
793cbf6161 radv/gfx10: Use new scan converter.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
2019-07-23 21:26:59 +02:00
Bas Nieuwenhuizen
4058b354c5 radv: Set FLUSH_ON_BINNING_TRANSITION.
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
2019-07-23 21:26:59 +02:00
Bas Nieuwenhuizen
906fcfccfd radv: Use pbb_allow for framebuffer BREAK_BATCH.
Ported from radeonsi.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
2019-07-23 21:26:59 +02:00
Marek Olšák
6ac2146a98 ac/nir: implement nir_op_pack_{us}norm_2x16
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
2019-07-23 15:03:44 -04:00
Samuel Pitoiset
d36af71f44 radv/gfx10: enable CLEAR_state
It actually works.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-23 14:15:55 +02:00
Samuel Pitoiset
9343c93e34 radv: fix dumping disassembly with RADV_DEBUG=shaders
Fixes: a20a9d0c5e ("radv: dont store disasm string unless keep_shader_info flag set")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-23 10:22:29 +02:00
Samuel Pitoiset
b5116d3cb7 radv: fix crash in vkCmdClearAttachments with unused attachment
depth_stencil_attachment and/or ds_resolve attachment can be NULL.

This fixes crashes with
dEQP-VK.renderpass.suballocation.unused_clear_attachments.*

Cc: 19.1 <mesa-stable@lists.freedesktop.org>
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
2019-07-22 14:25:54 +02:00