llvm-6502/test/CodeGen/ARM/avoid-cpsr-rmw.ll
Andrew Trick 6ea2b9608a Allocate local registers in order for optimal coloring.
Also avoid locals evicting locals just because they want a cheaper register.

Problem: MI Sched knows exactly how many registers we have and assumes
they can be colored. In cases where we have large blocks, usually from
unrolled loops, greedy coloring fails. This is a source of
"regressions" from the MI Scheduler on x86. I noticed this issue on
x86 where we have long chains of two-address defs in the same live
range. It's easy to see this in matrix multiplication benchmarks like
IRSmk and even the unit test misched-matmul.ll.

A fundamental difference between the LLVM register allocator and
conventional graph coloring is that in our model a live range can't
discover its neighbors, it can only verify its neighbors. That's why
we initially went for greedy coloring and added eviction to deal with
the hard cases. However, for singly defined and two-address live
ranges, we can optimally color without visiting neighbors simply by
processing the live ranges in instruction order.

Other beneficial side effects:

It is much easier to understand and debug regalloc for large blocks
when the live ranges are allocated in order. Yes, global allocation is
still very confusing, but it's nice to be able to comprehend what
happened locally.

Heuristics could be added to bias register assignment based on
instruction locality (think late register pairing, banks...).

Intuituvely this will make some test cases that are on the threshold
of register pressure more stable.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@187139 91177308-0d34-0410-b5e6-96231b3b80d8
2013-07-25 18:35:14 +00:00

119 lines
3.9 KiB
LLVM

; RUN: true
; Disabled for a single commit only.
; disabled: llc < %s -mtriple=thumbv7-apple-darwin -mcpu=cortex-a9 | FileCheck %s
; disabled: llc < %s -mtriple=thumbv7-apple-darwin -mcpu=swift | FileCheck %s
; Avoid some 's' 16-bit instruction which partially update CPSR (and add false
; dependency) when it isn't dependent on last CPSR defining instruction.
; rdar://8928208
define i32 @t1(i32 %a, i32 %b, i32 %c, i32 %d) nounwind readnone {
entry:
; CHECK-LABEL: t1:
; CHECK: muls [[REG:(r[0-9]+)]], r3, r2
; CHECK-NEXT: mul [[REG2:(r[0-9]+)]], r1, r0
; CHECK-NEXT: muls r0, [[REG]], [[REG2]]
%0 = mul nsw i32 %a, %b
%1 = mul nsw i32 %c, %d
%2 = mul nsw i32 %0, %1
ret i32 %2
}
; Avoid partial CPSR dependency via loop backedge.
; rdar://10357570
define void @t2(i32* nocapture %ptr1, i32* %ptr2, i32 %c) nounwind {
entry:
; CHECK-LABEL: t2:
%tobool7 = icmp eq i32* %ptr2, null
br i1 %tobool7, label %while.end, label %while.body
while.body:
; CHECK: while.body
; CHECK: mul r{{[0-9]+}}
; CHECK-NOT: muls
%ptr1.addr.09 = phi i32* [ %add.ptr, %while.body ], [ %ptr1, %entry ]
%ptr2.addr.08 = phi i32* [ %incdec.ptr, %while.body ], [ %ptr2, %entry ]
%0 = load i32* %ptr1.addr.09, align 4
%arrayidx1 = getelementptr inbounds i32* %ptr1.addr.09, i32 1
%1 = load i32* %arrayidx1, align 4
%arrayidx3 = getelementptr inbounds i32* %ptr1.addr.09, i32 2
%2 = load i32* %arrayidx3, align 4
%arrayidx4 = getelementptr inbounds i32* %ptr1.addr.09, i32 3
%3 = load i32* %arrayidx4, align 4
%add.ptr = getelementptr inbounds i32* %ptr1.addr.09, i32 4
%mul = mul i32 %1, %0
%mul5 = mul i32 %mul, %2
%mul6 = mul i32 %mul5, %3
store i32 %mul6, i32* %ptr2.addr.08, align 4
%incdec.ptr = getelementptr inbounds i32* %ptr2.addr.08, i32 -1
%tobool = icmp eq i32* %incdec.ptr, null
br i1 %tobool, label %while.end, label %while.body
while.end:
ret void
}
; Allow partial CPSR dependency when code size is the priority.
; rdar://12878928
define void @t3(i32* nocapture %ptr1, i32* %ptr2, i32 %c) nounwind minsize {
entry:
; CHECK-LABEL: t3:
%tobool7 = icmp eq i32* %ptr2, null
br i1 %tobool7, label %while.end, label %while.body
while.body:
; CHECK: while.body
; CHECK: mul r{{[0-9]+}}
; CHECK: muls
%ptr1.addr.09 = phi i32* [ %add.ptr, %while.body ], [ %ptr1, %entry ]
%ptr2.addr.08 = phi i32* [ %incdec.ptr, %while.body ], [ %ptr2, %entry ]
%0 = load i32* %ptr1.addr.09, align 4
%arrayidx1 = getelementptr inbounds i32* %ptr1.addr.09, i32 1
%1 = load i32* %arrayidx1, align 4
%arrayidx3 = getelementptr inbounds i32* %ptr1.addr.09, i32 2
%2 = load i32* %arrayidx3, align 4
%arrayidx4 = getelementptr inbounds i32* %ptr1.addr.09, i32 3
%3 = load i32* %arrayidx4, align 4
%add.ptr = getelementptr inbounds i32* %ptr1.addr.09, i32 4
%mul = mul i32 %1, %0
%mul5 = mul i32 %mul, %2
%mul6 = mul i32 %mul5, %3
store i32 %mul6, i32* %ptr2.addr.08, align 4
%incdec.ptr = getelementptr inbounds i32* %ptr2.addr.08, i32 -1
%tobool = icmp eq i32* %incdec.ptr, null
br i1 %tobool, label %while.end, label %while.body
while.end:
ret void
}
; Avoid producing tMOVi8 after a high-latency flag-setting operation.
; <rdar://problem/13468102>
define void @t4(i32* nocapture %p, double* nocapture %q) {
entry:
; CHECK: t4
; CHECK: vmrs APSR_nzcv, fpscr
; CHECK: if.then
; CHECK-NOT: movs
%0 = load double* %q, align 4
%cmp = fcmp olt double %0, 1.000000e+01
%incdec.ptr1 = getelementptr inbounds i32* %p, i32 1
br i1 %cmp, label %if.then, label %if.else
if.then:
store i32 7, i32* %p, align 4
%incdec.ptr2 = getelementptr inbounds i32* %p, i32 2
store i32 8, i32* %incdec.ptr1, align 4
store i32 9, i32* %incdec.ptr2, align 4
br label %if.end
if.else:
store i32 3, i32* %p, align 4
%incdec.ptr5 = getelementptr inbounds i32* %p, i32 2
store i32 5, i32* %incdec.ptr1, align 4
store i32 6, i32* %incdec.ptr5, align 4
br label %if.end
if.end:
ret void
}