[llvm-bugs] [Bug 47457] New: LoopLoadElim: assert failed "cast<Ty>() argument of incompatible type!"
via llvm-bugs
llvm-bugs at lists.llvm.org
Tue Sep 8 02:21:44 PDT 2020
https://bugs.llvm.org/show_bug.cgi?id=47457
Bug ID: 47457
Summary: LoopLoadElim: assert failed "cast<Ty>() argument of
incompatible type!"
Product: new-bugs
Version: trunk
Hardware: PC
OS: Windows NT
Status: NEW
Severity: enhancement
Priority: P
Component: new bugs
Assignee: unassignedbugs at nondot.org
Reporter: max.kazantsev at azul.com
CC: htmldeveloper at gmail.com, llvm-bugs at lists.llvm.org
Command: opt -loop-load-elim -S ./test.ll
test.ll:
source_filename = "./test.ll"
target datalayout =
"e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128-ni:1-p2:32:8:8:32-ni:2"
target triple = "x86_64-unknown-linux-gnu"
define void @pluto() {
bb:
br label %bb1
bb1: ; preds = %bb6, %bb1, %bb
%tmp = phi i32 [ undef, %bb ], [ 0, %bb1 ], [ %tmp3, %bb6 ]
br i1 undef, label %bb1, label %bb2
bb2: ; preds = %bb1
%tmp3 = add i32 %tmp, 1
%tmp4 = icmp ult i32 %tmp, undef
br i1 %tmp4, label %bb6, label %bb5
bb5: ; preds = %bb2
ret void
bb6: ; preds = %bb2
br i1 undef, label %bb7, label %bb1
bb7: ; preds = %bb7, %bb6
%tmp8 = phi i32 [ %tmp15, %bb7 ], [ %tmp3, %bb6 ]
%tmp9 = phi i32 [ %tmp8, %bb7 ], [ %tmp, %bb6 ]
%tmp10 = zext i32 %tmp9 to i64
%tmp11 = getelementptr inbounds float, float addrspace(1)* null, i64 %tmp10
%tmp12 = load float, float addrspace(1)* %tmp11, align 4
%tmp13 = zext i32 %tmp8 to i64
%tmp14 = getelementptr inbounds float, float addrspace(1)* null, i64 %tmp13
store float 1.000000e+00, float addrspace(1)* %tmp14, align 4
%tmp15 = add nuw nsw i32 %tmp8, 1
%tmp16 = icmp sgt i32 %tmp8, 78
br i1 %tmp16, label %bb17, label %bb7
bb17: ; preds = %bb7
unreachable
}
Failed attempt of unconditionlal cast of some SCEV to SCEVAddRec. Our
downstream test first showed this failure starting from patch:
commit 99166fd4fb422351f131fb1265cb85d5f6c5b8da
Author: Florian Hahn <flo at fhahn.com>
Date: Wed Jul 29 14:54:03 2020 +0100
[SCEVExpander] Add option to preserve LCSSA directly.
But I haven't yet confirmed that it's related to the root cause (it might have
just exposed an older bug).
--
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20200908/1a4bd76a/attachment.html>
More information about the llvm-bugs
mailing list