If an instruction didn't already use SDWA convert_to_SDWA in apply_extract
will add ubyte0/uword0 selections for v1b/v2b operands. This loses information
that the instruction doesn't care about the high bits and makes the next
apply_extract_twice fail.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Fixes: 6cb9d39bc2 ("aco: combine extracts with sub-dword definitions")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/32803>
v_add can be dual issued on gfx11, v_med3 cannot.
Don't use v_add directly to still optimize omod(fsat(x)).
Foz-DB GFX1100:
Totals from 32702 (24.24% of 134913) affected shaders:
Latency: 475008203 -> 474928037 (-0.02%); split: -0.02%, +0.00%
InvThroughput: 59226198 -> 59140787 (-0.14%); split: -0.14%, +0.00%
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/21402>
This doesn't make sense, opsel preserves the not selected half of the register,
p_insert zeros it.
No Foz-DB changes.
Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Fixes: 54292e99c7 ("aco: optimize 32-bit extracts and inserts using SDWA")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19253>
Instead of SDWA v_mov_b32/v_xor_b32, we can use a combination of
v_add_u16/v_sub_u16 (add/sub swap, similar to xor swap) and v_perm_b32
with a literal.
I don't know yet if GFX11 adds any new instructions which makes this
easier, but this approach should have full functionality.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16595>
Instead, we can rely on the fact that subdword definitions
must preserve the unused bits while dword definitions either
pad or sign-extend.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12640>
This commit introduces a new struct SubdwordSel
in order to ease and clean up the usage of SDWA
selections. This includes removing the distinction
between register-allocated and fixed SDWA selections.
Instead, SDWA selections can now also access the high
bits of subdword variables. Alignment and sizes are
validated accordingly. Size, offset and sign_extend
can be evaluated via helper methods.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12640>
This migration was done with libclang-based automatic tooling, which
performed these replacements:
* Operand(uint8_t) -> Operand::c8
* Operand(uint16_t) -> Operand::c16
* Operand(uint32_t, false) -> Operand::c32
* Operand(uint32_t, bool) -> Operand::c32_or_c64
* Operand(uint64_t) -> Operand::c64
* Operand(0) -> Operand::zero(num_bytes)
Casts that were previously used for constructor selection have automatically
been removed (e.g. Operand((uint16_t)1) -> Operand::c16(1)).
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11653>
