RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-01 16:33:37 +01:00
Code Issues Projects Releases Wiki Activity
fcde15ab12
llvm-mirror/test/Transforms/InstCombine/align-external.ll

30 lines
646 B
LLVM
Raw Normal View History

Change tests from "opt %s" to "opt < %s" so that opt doesn't see the input filename so that opt doesn't print the input filename in the output so that grep lines in the tests don't unintentionally match strings in the input filename. llvm-svn: 81537
2009-09-11 20:01:28 +02:00
; RUN: opt < %s -instcombine -S | FileCheck %s
Don't assume that external global variables are aligned at their preferred alignment. Only the minimum alignment guaranteed by the ABI may be assumed. llvm-svn: 78668
2009-08-11 17:50:03 +02:00
Fix a theoretical problem (not seen in the wild): if different instances of a weak variable are compiled by different compilers, such as GCC and LLVM, while LLVM may increase the alignment to the preferred alignment there is no reason to think that GCC will use anything more than the ABI alignment. Since it is the GCC version that might end up in the final program (as the linkage is weak), it is wrong to increase the alignment of loads from the global up to the preferred alignment as the alignment might only be the ABI alignment. Increasing alignment up to the ABI alignment might be OK, but I'm not totally convinced that it is. It seems better to just leave the alignment of weak globals alone. llvm-svn: 145413
2011-11-29 19:26:38 +01:00
; Don't assume that external global variables or those with weak linkage have
; their preferred alignment. They may only have the ABI minimum alignment.
Don't assume that external global variables are aligned at their preferred alignment. Only the minimum alignment guaranteed by the ABI may be assumed. llvm-svn: 78668
2009-08-11 17:50:03 +02:00
; CHECK: %s = shl i64 %a, 3
; CHECK: %r = or i64 %s, ptrtoint (i32* @A to i64)
; CHECK: %q = add i64 %r, 1
; CHECK: ret i64 %q
Disallow the undocumented practice of starting the datalayout string with '-'. Update some test cases accordingly. llvm-svn: 168516
2012-11-23 15:51:42 +01:00
target datalayout = "i32:8:32"
Don't assume that external global variables are aligned at their preferred alignment. Only the minimum alignment guaranteed by the ABI may be assumed. llvm-svn: 78668
2009-08-11 17:50:03 +02:00
@A = external global i32
Fix a theoretical problem (not seen in the wild): if different instances of a weak variable are compiled by different compilers, such as GCC and LLVM, while LLVM may increase the alignment to the preferred alignment there is no reason to think that GCC will use anything more than the ABI alignment. Since it is the GCC version that might end up in the final program (as the linkage is weak), it is wrong to increase the alignment of loads from the global up to the preferred alignment as the alignment might only be the ABI alignment. Increasing alignment up to the ABI alignment might be OK, but I'm not totally convinced that it is. It seems better to just leave the alignment of weak globals alone. llvm-svn: 145413
2011-11-29 19:26:38 +01:00
@B = weak_odr global i32 0
Don't assume that external global variables are aligned at their preferred alignment. Only the minimum alignment guaranteed by the ABI may be assumed. llvm-svn: 78668
2009-08-11 17:50:03 +02:00
define i64 @foo(i64 %a) {
%t = ptrtoint i32* @A to i64
%s = shl i64 %a, 3
%r = or i64 %t, %s
%q = add i64 %r, 1
ret i64 %q
}
Fix a theoretical problem (not seen in the wild): if different instances of a weak variable are compiled by different compilers, such as GCC and LLVM, while LLVM may increase the alignment to the preferred alignment there is no reason to think that GCC will use anything more than the ABI alignment. Since it is the GCC version that might end up in the final program (as the linkage is weak), it is wrong to increase the alignment of loads from the global up to the preferred alignment as the alignment might only be the ABI alignment. Increasing alignment up to the ABI alignment might be OK, but I'm not totally convinced that it is. It seems better to just leave the alignment of weak globals alone. llvm-svn: 145413
2011-11-29 19:26:38 +01:00
define i32 @bar() {
; CHECK: @bar
%r = load i32* @B, align 1
; CHECK: align 1
ret i32 %r
}
Reference in New Issue Copy Permalink