This fills in the delay slots of thread end as much as we can (other than
being cautious about potential TLBZ writes).
In the process, I moved the thread end THRSW instruction creation to the
scheduler. Once we start emitting THRSWs in the shader, we need to
schedule the thread-end one differently from other THRSWs, so having it in
there makes that easy.
Now, instead of a magic write register for VPM stores we have an
instruction to do them (which means no packing of other ALU ops into it),
with the ability to reorder the VPM stores due to the offset being baked
into the instruction.
VPM loads also gain the ability to be reordered by packing the row into
the A argument. They also no longer write to the r3 accumulator, and
instead must be stored to a physical register.
The WRTMUC replaces the implicit uniform loads in the first two texture
instructions. LDVPM disappears in favor of an ALU op. LDVARY, LDTMU,
LDTLB, and LDUNIF*RF now write to arbitrary registers, which required
passing the devinfo through to a few more functions.
This creates two new internal dependencies, idep_nir_headers and
idep_nir. The former encapsulates the generation of nir_opcodes.h and
nir_builder_opcodes.h and adding src/compiler/nir as an include path.
This ensures that any target that needs nir headers will have the
includes and that the generated headers will be generated before the
target is build. The second, idep_nir, includes the first and
additionally links to libnir.
This is intended to make it easier to avoid race conditions in the build
when using nir, since the number of consumers for libnir and it's
headers are quite high.
Acked-by: Eric Engestrom <eric.engestrom@imgtec.com>
Signed-off-by: Dylan Baker <dylan.c.baker@intel.com>
I want to do the SETMSF.IFA to discard only if execute == 0 and cond, so
our dest of the PUSHZ needs to be nonzero if execute or !cond are nonzero.
Fixes dEQP-GLES3.functional.shaders.discard.dynamic_loop_dynamic.
This means that with no flatshading we'll emit the single-byte
ZERO_ALL_FLAT_SHADE_FLAGS, and otherwise emit a set of FLAT_SHADE_FLAGS to
get all the bits we need set.
There's a _SET enum in the packet we could use to possibly set entire
ranges of the bitfield without using another packet, but this at least
fixes the conformance failure.
In updating the simulator, behavior changed slightly so that our old code
wasn't getting glxgears's flatshading interpolated right. Emit flat
shading code just like we would for a normal flat-shaded varying, by
passing a flag in the shader key for glShadeModel(GL_FLAT) state and
customizing the color inputs based on that.
The compiler decides how many LDTMUs we're going to emit, and that must
match the P1 flags. This brings the return channel counting to a single
place (so all that's passed into the compiler is "how many return channels
you may request from this texture's format), and was a necessary step for
shadow samplers once we stop using OVRTMUOUT=0.
Fixes almost all of piglit's arb_shader_texture_lod grad tests, except for
the base -texgrad/texgradcube ones which fail on what appear to be
precision problems.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
The HW doesn't add the base level anywhere (the min/max lod clamping is
what does base level), so we need to add it manually in this case.
Fixes piglit tex-miplevel-selection *Lod 2D.
Unused since original import of VC5.
Fixes: ade416d023 ("broadcom: Add VC5 NIR compiler.")
Signed-off-by: Andreas Boll <andreas.boll.dev@gmail.com>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
The v3d_qpu_writes_r*() were only checking for fixed-function accumulator
writes, not normal ALU writes to those regs.
Fixes fs-discard-exit-2 on simulation (but not HW).
In the case of fneg(0.0), we were getting back 0.0 instead of -0.0. We
were also needing an immediate 0 value for ineg, when there's an opcode to
do the job properly.
Fixes fs-floatBitsToInt-neg.shader_test.
The HW has no native sampler support for multisample textures, but since
we only need to support txf_ms and the layout is UIF, we just need to
scale up the texcoords and then add in the sample.
This drops the old TEXTURE_MSAA_ADDR special uniform, since we're treating
MSAA textures as textures, rather than basically texbos like VC4 had to.
We only have 2x16 unpacking in our ALUs. To enable this, we also need
lower_fdiv for its new instructions, which had been handled at a higher
level previously.
I already had the texture's wrapping set up to use different behavior for
nearest or linear, so we just needed to saturate the coordinates in linear
mode to get the "proper" blend between the edge and border values.
A bit of spec text suggested that (like vc4) condition codes should be
used for discards, and the simulator was fine with it, but the 7268
disagrees and you have to use SETMSF instead or the color comes through.
Fixes glsl-fs-discard-01 and many of the interpolation-with-clipping
tests.
We don't have native instructions for them, so set up the lowering. Once
we support the bfi instructions that get generated, they should start
actually working.
v2: Default vc5 to off, since it requires the simulator currently. Add
missing dep on the XML generation from libbroadcom_vc5.
Reviewed-by: Dylan Baker <dylan@pnwbakers.com> (v1)
In order to keep early-Z from writing early in a discard shader, you need
to set the "modifies Z" bit in the shader state (which the new
prog_data.discards will indicate). Then, in the shader we do a TLB write
to make Z passthrough happen (the QPU result is ignored, so we use a NULL
source).
This is a pretty straightforward fork of VC4's NIR compiler to VC5. The
condition codes, registers, and I/O have all changed, making the backend
hard to share, though their heritage is still recognizable.
v2: Move to src/broadcom/compiler to match intel's layout, rename more
"vc5" to "v3d", rename QIR to VIR ("V3D IR") to avoid symbol conflicts
with vc4, use new v3d_debug header, add compiler init/free functions,
do texture swizzling in NIR to allow optimization.
