mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-23 03:02:36 +01:00
3f23d4b8c3
tryLatency compares two sched candidates. For the top zone it prefers the one with lesser depth, but only if that depth is greater than the total latency of the instructions we've already scheduled -- otherwise its latency would be hidden and there would be no stall. Unfortunately it only tests the depth of one of the candidates. This can lead to situations where the TopDepthReduce heuristic does not kick in, but a lower priority heuristic chooses the other candidate, whose depth *is* greater than the already scheduled latency, which causes a stall. The fix is to apply the heuristic if the depth of *either* candidate is greater than the already scheduled latency. All this also applies to the BotHeightReduce heuristic in the bottom zone. Differential Revision: https://reviews.llvm.org/D72392
103 lines
4.4 KiB
LLVM
103 lines
4.4 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc -verify-machineinstrs < %s -mcpu=pwr9 -mtriple=powerpc64le-ibm-linux| FileCheck %s
|
|
|
|
; This file verifies that for a given floating point load / store pair,
|
|
; if the load value isn't used by any other operations,
|
|
; then consider transforming the pair to integer load / store operations
|
|
|
|
@a1 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a2 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a3 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a4 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a5 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a6 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a7 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a8 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a9 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a10 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a11 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a12 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a13 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a14 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a15 = local_unnamed_addr global double 0.000000e+00, align 8
|
|
@a16 = local_unnamed_addr global ppc_fp128 0xM00000000000000000000000000000000, align 16
|
|
@a17 = local_unnamed_addr global fp128 0xL00000000000000000000000000000000, align 16
|
|
|
|
; Because this test function is trying to pass float argument by stack,
|
|
; so the fpr is only used to load/store float argument
|
|
define signext i32 @test() nounwind {
|
|
; CHECK-LABEL: test:
|
|
; CHECK: # %bb.0:
|
|
; CHECK-NEXT: mflr 0
|
|
; CHECK-NEXT: std 0, 16(1)
|
|
; CHECK-NEXT: stdu 1, -192(1)
|
|
; CHECK-NEXT: addis 3, 2, a1@toc@ha
|
|
; CHECK-NEXT: addis 5, 2, a16@toc@ha
|
|
; CHECK-NEXT: addis 6, 2, a17@toc@ha
|
|
; CHECK-NEXT: addis 4, 2, a15@toc@ha
|
|
; CHECK-NEXT: lfd 1, a1@toc@l(3)
|
|
; CHECK-NEXT: addis 3, 2, a2@toc@ha
|
|
; CHECK-NEXT: addi 5, 5, a16@toc@l
|
|
; CHECK-NEXT: addi 6, 6, a17@toc@l
|
|
; CHECK-NEXT: ld 4, a15@toc@l(4)
|
|
; CHECK-NEXT: lfd 2, a2@toc@l(3)
|
|
; CHECK-NEXT: addis 3, 2, a3@toc@ha
|
|
; CHECK-NEXT: lxvx 34, 0, 6
|
|
; CHECK-NEXT: lxvx 0, 0, 5
|
|
; CHECK-NEXT: li 5, 152
|
|
; CHECK-NEXT: lfd 3, a3@toc@l(3)
|
|
; CHECK-NEXT: addis 3, 2, a4@toc@ha
|
|
; CHECK-NEXT: lfd 4, a4@toc@l(3)
|
|
; CHECK-NEXT: addis 3, 2, a5@toc@ha
|
|
; CHECK-NEXT: lfd 5, a5@toc@l(3)
|
|
; CHECK-NEXT: addis 3, 2, a6@toc@ha
|
|
; CHECK-NEXT: lfd 6, a6@toc@l(3)
|
|
; CHECK-NEXT: addis 3, 2, a7@toc@ha
|
|
; CHECK-NEXT: lfd 7, a7@toc@l(3)
|
|
; CHECK-NEXT: addis 3, 2, a8@toc@ha
|
|
; CHECK-NEXT: lfd 8, a8@toc@l(3)
|
|
; CHECK-NEXT: addis 3, 2, a9@toc@ha
|
|
; CHECK-NEXT: lfd 9, a9@toc@l(3)
|
|
; CHECK-NEXT: addis 3, 2, a10@toc@ha
|
|
; CHECK-NEXT: lfd 10, a10@toc@l(3)
|
|
; CHECK-NEXT: addis 3, 2, a11@toc@ha
|
|
; CHECK-NEXT: lfd 11, a11@toc@l(3)
|
|
; CHECK-NEXT: addis 3, 2, a12@toc@ha
|
|
; CHECK-NEXT: lfd 12, a12@toc@l(3)
|
|
; CHECK-NEXT: addis 3, 2, a13@toc@ha
|
|
; CHECK-NEXT: lfd 13, a13@toc@l(3)
|
|
; CHECK-NEXT: addis 3, 2, a14@toc@ha
|
|
; CHECK-NEXT: ld 3, a14@toc@l(3)
|
|
; CHECK-NEXT: stxvx 0, 1, 5
|
|
; CHECK-NEXT: std 4, 144(1)
|
|
; CHECK-NEXT: std 3, 136(1)
|
|
; CHECK-NEXT: bl _Z3fooddddddddddddddd
|
|
; CHECK-NEXT: nop
|
|
; CHECK-NEXT: li 3, 0
|
|
; CHECK-NEXT: addi 1, 1, 192
|
|
; CHECK-NEXT: ld 0, 16(1)
|
|
; CHECK-NEXT: mtlr 0
|
|
; CHECK-NEXT: blr
|
|
%1 = load double, double* @a1, align 8
|
|
%2 = load double, double* @a2, align 8
|
|
%3 = load double, double* @a3, align 8
|
|
%4 = load double, double* @a4, align 8
|
|
%5 = load double, double* @a5, align 8
|
|
%6 = load double, double* @a6, align 8
|
|
%7 = load double, double* @a7, align 8
|
|
%8 = load double, double* @a8, align 8
|
|
%9 = load double, double* @a9, align 8
|
|
%10 = load double, double* @a10, align 8
|
|
%11 = load double, double* @a11, align 8
|
|
%12 = load double, double* @a12, align 8
|
|
%13 = load double, double* @a13, align 8
|
|
%14 = load double, double* @a14, align 8
|
|
%15 = load double, double* @a15, align 8
|
|
%16 = load ppc_fp128, ppc_fp128* @a16, align 16
|
|
%17 = load fp128, fp128* @a17, align 16
|
|
tail call void @_Z3fooddddddddddddddd(double %1, double %2, double %3, double %4, double %5, double %6, double %7, double %8, double %9, double %10, double %11, double %12, double %13, double %14, double %15, ppc_fp128 %16, fp128 %17)
|
|
ret i32 0
|
|
}
|
|
|
|
declare void @_Z3fooddddddddddddddd(double, double, double, double, double, double, double, double, double, double, double, double, double, double, double, ppc_fp128, fp128)
|