6502/llvm-6502
1
0
Fork 0
mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2024-12-15 20:29:48 +00:00
Code Issues Projects Releases Wiki Activity
65e1227b37
llvm-6502/test/Assembler/ConstantExprNoFold.ll

51 lines
2.2 KiB
LLVM
Raw Normal View History

Fix ConstantFold's folding of icmp instructions to recognize that, for example, a one-past-the-end pointer from one global variable may be equal to the base pointer of another global variable. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@173995 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-31 00:01:45 +00:00
; This test checks to make sure that constant exprs don't fold in some simple
; situations
; RUN: llvm-as < %s | llvm-dis | FileCheck %s
verify-uselistorder: Force -preserve-bc-use-list-order git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@216022 91177308-0d34-0410-b5e6-96231b3b80d8
2014-08-19 21:08:27 +00:00
; RUN: verify-uselistorder %s
Fix ConstantFold's folding of icmp instructions to recognize that, for example, a one-past-the-end pointer from one global variable may be equal to the base pointer of another global variable. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@173995 91177308-0d34-0410-b5e6-96231b3b80d8
2013-01-31 00:01:45 +00:00
; Even give it a datalayout, to tempt folding as much as possible.
target datalayout = "p:32:32"
@A = global i64 0
@B = global i64 0
; Don't fold this. @A might really be allocated next to @B, in which case the
; icmp should return true. It's not valid to *dereference* in @B from a pointer
; based on @A, but icmp isn't a dereference.
; CHECK: @C = global i1 icmp eq (i64* getelementptr inbounds (i64* @A, i64 1), i64* @B)
@C = global i1 icmp eq (i64* getelementptr inbounds (i64* @A, i64 1), i64* @B)
; Don't fold this completely away either. In theory this could be simplified
; to only use a gep on one side of the icmp though.
; CHECK: @D = global i1 icmp eq (i64* getelementptr inbounds (i64* @A, i64 1), i64* getelementptr inbounds (i64* @B, i64 2))
@D = global i1 icmp eq (i64* getelementptr inbounds (i64* @A, i64 1), i64* getelementptr inbounds (i64* @B, i64 2))
Add addrspacecast instruction. Patch by Michele Scandale! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194760 91177308-0d34-0410-b5e6-96231b3b80d8
2013-11-15 01:34:59 +00:00
; CHECK: @E = global i64 addrspace(1)* addrspacecast (i64* @A to i64 addrspace(1)*)
@E = global i64 addrspace(1)* addrspacecast(i64* @A to i64 addrspace(1)*)
IR: Don't add inbounds to GEPs of extern_weak variables Global variables that have `extern_weak` linkage may be null, so it's incorrect to add `inbounds` when constant folding. This also fixes a bug when parsing global aliases, whose forward reference placeholders are global variables with `extern_weak` linkage. If GEPs to these aliases are encountered before the alias itself, the GEPs would incorrectly gain the `inbounds` keyword as well. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@215803 91177308-0d34-0410-b5e6-96231b3b80d8
2014-08-16 01:54:32 +00:00
; Don't add an inbounds on @weak.gep, since @weak may be null.
; CHECK: @weak.gep = global i32* getelementptr (i32* @weak, i32 1)
@weak.gep = global i32* getelementptr (i32* @weak, i32 1)
@weak = extern_weak global i32
ConstantFold: Don't optimize comparisons with weak linkage objects Consider: void f() {} void __attribute__((weak)) g() {} bool b = &f != &g; It's possble for g to resolve to f if --defsym=g=f is passed on to the linker. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@223585 91177308-0d34-0410-b5e6-96231b3b80d8
2014-12-06 11:58:33 +00:00
; An object with weak linkage cannot have it's identity determined at compile time.
; CHECK: @F = global i1 icmp eq (i32* @weakany, i32* @glob)
@F = global i1 icmp eq (i32* @weakany, i32* @glob)
@weakany = weak global i32 0
ConstantFold: Zero-sized globals might land on top of another global A zero sized array is zero sized and might share its address with another global. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@223684 91177308-0d34-0410-b5e6-96231b3b80d8
2014-12-08 19:35:31 +00:00
; Empty globals might end up anywhere, even on top of another global.
; CHECK: @empty.cmp = global i1 icmp eq ([0 x i8]* @empty.1, [0 x i8]* @empty.2)
@empty.1 = external global [0 x i8], align 1
@empty.2 = external global [0 x i8], align 1
@empty.cmp = global i1 icmp eq ([0 x i8]* @empty.1, [0 x i8]* @empty.2)
IR: Don't add inbounds to GEPs of extern_weak variables Global variables that have `extern_weak` linkage may be null, so it's incorrect to add `inbounds` when constant folding. This also fixes a bug when parsing global aliases, whose forward reference placeholders are global variables with `extern_weak` linkage. If GEPs to these aliases are encountered before the alias itself, the GEPs would incorrectly gain the `inbounds` keyword as well. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@215803 91177308-0d34-0410-b5e6-96231b3b80d8
2014-08-16 01:54:32 +00:00
; Don't add an inbounds on @glob.a3, since it's not inbounds.
; CHECK: @glob.a3 = alias getelementptr (i32* @glob.a2, i32 1)
@glob = global i32 0
@glob.a3 = alias getelementptr (i32* @glob.a2, i32 1)
@glob.a2 = alias getelementptr (i32* @glob.a1, i32 1)
@glob.a1 = alias i32* @glob
Reference in New Issue Copy Permalink