In that case, the mask is specified on 32bit lanes, so we need to shift
it if it's > 0x3. The expand modifier will take care of selecting the
right side of the 32bit vector.
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14885>
Even though 8-bit ALUs are not supported, we can have [un]pack_32_4x8
instructions which translate to IMOVs, and those operate on 8-bit
vectors. The problem is, the swizzling granularity is 16 bit, which
means we don't support
MOV.i8 R0.xyzw, TMP0.xxxx, R1.zyxw
and the compiler doesn't even complain, it just applies 8 bit
swizzling directly, which obviously doesn't work.
This is probably not the right way to fix that, but I thought I'd
raised the issue with a hack to fix, so we can get the discussion
started.
(Found while debugging FB store lowering on Midgard).
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14885>
Needed to link the disassembler separate from the rest of the compiler,
as in out-of-tree pandecode builds. Which I haven't done for Midgard in
well over a year, enough time for this to bit rot.
Signed-off-by: Alyssa Rosenzweig <alyssa@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/14185>
This commit makes mir_pack_swizzle set rep_low, rep_high and half in a
more explicit way, helping in the transition to a unified expand_mode
enum in the following commit.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9461>
This commits fixes a regression caused by commit
616394cf31.
Fixes dEQP-GLES31.functional.compute.shared_var.atomic.min.*,
dEQP-GLES31.functional.ssbo.atomic.min.*,
dEQP-GLES31.functional.compute.shared_var.atomic.max.* and
dEQP-GLES31.functional.ssbo.atomic.max.*.
Midgard's non-image and non-varying store operations have a unique
behavior for the load/store writemask, with each bit being responsible
for 1/4th of the size of the total write operation. This means that we
have to pack the writemask differently since we stopped using st_u128
for everything.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10067>
It needs to respect the existing swizzle, as well as the type size.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6321>
This saves power and time by skipping a roundtrip to the register file.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6128>
This refactor prepares emit_alu_bundle() for the next commit that
reconstructs branch instructions right before emission.
It also simplifies the code since the previous control flow was only
better when we had the prepacked fields in midgard_instruction.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5933>
This commit removes the `ins->alu` field from midgard_instruction,
simplifying the code by just recreating midgard_vector_alu later when we
have to emit it.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5933>
This commit moves the packing of registers and other things from
install_registers_instr() to midgard_emit.c, right before emitting the
binary.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5933>
This commit makes `ins->op` the correct field to use with load_store
instructions.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5933>
This commit makes the `ins->op` the correct field to use with texture
instructions.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5933>
In an effort to simplify MIR by not prepacking instructions, this commit
removes references to `ins->alu.outmod` so that we can later remove the
`ins->alu` field from midgard_instruction.
Every place that was using `ins->alu.outmod` was changed to now use the
generic `ins->outmod` field instead.
We then reconstruct the outmod field right before emission.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5933>
In an effort to simplify MIR by not prepacking instructions, this commit
removes references to `ins->alu.reg_mode` so that we can later remove
the `ins->alu` field from midgard_instruction.
Every place that was using reg_mode was changed to now use the generic
`ins->src_type` field instead.
We then reconstruct the reg_mode field right before emission.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5933>
In an effort to simplify MIR by not prepacking instructions, this commit
removes references to `ins->alu.op` so that we can later remove the
`ins->alu` field from midgard_instruction.
Every place that was using ins->op was changed to now use the generic
`ins->op` field instead.
We then reconstruct the `alu.op` field right before emission.
This new field is generic and can contain opcodes for ALU, texture or
load/store instructions. It should be used in conjunction with
`ins->type`, just like the current prepacked `op` field.
Signed-off-by: Italo Nicola <italonicola@collabora.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5933>
We don't make great use of it (due to the scheduler not being aware
yet), but we can pack for it regardless and maybe pick up some win.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5286>
We can pass it all off to emit time, and let the types in the IR do the
heavylifting in the meantime, which is a lot easier to get right.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5151>
Symmetric with vector mods, except for normal which is packed as
sign-extend. (flag 2 never seen in the wild)
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5151>
With this, we may remove all invert passes and simply look at the src
modifier on NIR->MIR and fixup at pack time. No shader-db changes.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5102>
Corresponds roughly to what we analyze. Note that "terminate AND
execute" is a contradiction (rather: it's equivalent to just
terminating), hence why there are only three possibilities for the
states of the flags:
.cont = continue, don't execute
.last = don't continue, don't execute
.cont.last = continue and execute
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5014>
Alyssa Rosenzweig
Boris Brezillon
Tomeu Vizoso
Italo Nicola
Alyssa Rosenzweig
Eric Anholt