b1d47467e2
This commit removes the artificial types <512 x i1> and <1024 x i1>
from HVX intrinsics, and makes v512i1 and v1024i1 no longer legal on
Hexagon.
It may cause existing bitcode files to become invalid.
* Converting between vector predicates and vector registers must be
done explicitly via vandvrt/vandqrt instructions (their intrinsics),
i.e. (for 64-byte mode):
%Q = call <64 x i1> @llvm.hexagon.V6.vandvrt(<16 x i32> %V, i32 -1)
%V = call <16 x i32> @llvm.hexagon.V6.vandqrt(<64 x i1> %Q, i32 -1)
The conversion intrinsics are:
declare <64 x i1> @llvm.hexagon.V6.vandvrt(<16 x i32>, i32)
declare <128 x i1> @llvm.hexagon.V6.vandvrt.128B(<32 x i32>, i32)
declare <16 x i32> @llvm.hexagon.V6.vandqrt(<64 x i1>, i32)
declare <32 x i32> @llvm.hexagon.V6.vandqrt.128B(<128 x i1>, i32)
They are all pure.
* Vector predicate values cannot be loaded/stored directly. This directly
reflects the architecture restriction. Loading and storing or vector
predicates must be done indirectly via vector registers and explicit
conversions via vandvrt/vandqrt instructions.
16 lines
481 B
LLVM
16 lines
481 B
LLVM
; RUN: llc -march=hexagon < %s | FileCheck %s
|
|
; REQUIRES: asserts
|
|
|
|
; Make sure we can handle the 'q' constraint in the 128-byte mode.
|
|
|
|
target triple = "hexagon"
|
|
|
|
; CHECK-LABEL: fred
|
|
; CHECK: if (q{{[0-3]}}) vmem
|
|
define void @fred() #0 {
|
|
tail call void asm sideeffect "if ($0) vmem($1) = $2;", "q,r,v,~{memory}"(<128 x i1> undef, <32 x i32>* undef, <32 x i32> undef) #0
|
|
ret void
|
|
}
|
|
|
|
attributes #0 = { nounwind "target-cpu"="hexagonv60" "target-features"="+hvx,+hvx-length128b" }
|