maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/Transforms/TailCallElim/reorder_load.ll

148 lines
4.5 KiB
LLVM
Raw Normal View History

This commit is a 4x squash commit consisting of 4x functions converted to use FileCheck instead of grep. Messages: Converted test case trivial_codegen_tailcall.ll to use FileCheck. Converted test return_constant.ll to use FileCheck instead of grep. Converted test reorder_load.ll to use FileCheck instead of grep. Converted test intervening-inst.ll to use FileCheck instead of grep. llvm-svn: 172171
2013-01-11 12:12:53 +08:00
; RUN: opt < %s -tailcallelim -S | FileCheck %s
Improve tail call elim to move loads above readonly calls when it allows forming a tail call. Patch by Frits van Bommel. This implements PR4323. llvm-svn: 73752
2009-06-19 12:22:16 +08:00
; PR4323
Fix a long-standing miscompile in the load analysis that was uncovered by my refactoring of this code. The method isSafeToLoadUnconditionally assumes that the load will proceed with the preferred type alignment. Given that, it has to ensure that the alloca or global is at least that aligned. It has always done this historically when a datalayout is present, but has never checked it when the datalayout is absent. When I refactored the code in r220156, I exposed this path when datalayout was present and that turned the latent bug into a patent bug. This fixes the issue by just removing the special case which allows folding things without datalayout. This isn't worth the complexity of trying to tease apart when it is or isn't safe without actually knowing the preferred alignment. llvm-svn: 220161
2014-10-19 16:17:50 +08:00
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
Improve tail call elim to move loads above readonly calls when it allows forming a tail call. Patch by Frits van Bommel. This implements PR4323. llvm-svn: 73752
2009-06-19 12:22:16 +08:00
; Several cases where tail call elimination should move the load above the call,
; then eliminate the tail recursion.
@global = external global i32 ; <i32*> [#uses=1]
@extern_weak_global = extern_weak global i32 ; <i32*> [#uses=1]
; This load can be moved above the call because the function won't write to it
; and the call has no side effects.
define fastcc i32 @raise_load_1(i32* %a_arg, i32 %a_len_arg, i32 %start_arg) nounwind readonly {
Fix a long-standing miscompile in the load analysis that was uncovered by my refactoring of this code. The method isSafeToLoadUnconditionally assumes that the load will proceed with the preferred type alignment. Given that, it has to ensure that the alloca or global is at least that aligned. It has always done this historically when a datalayout is present, but has never checked it when the datalayout is absent. When I refactored the code in r220156, I exposed this path when datalayout was present and that turned the latent bug into a patent bug. This fixes the issue by just removing the special case which allows folding things without datalayout. This isn't worth the complexity of trying to tease apart when it is or isn't safe without actually knowing the preferred alignment. llvm-svn: 220161
2014-10-19 16:17:50 +08:00
; CHECK-LABEL: @raise_load_1(
; CHECK-NOT: call
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
; CHECK: load i32, i32*
Fix a long-standing miscompile in the load analysis that was uncovered by my refactoring of this code. The method isSafeToLoadUnconditionally assumes that the load will proceed with the preferred type alignment. Given that, it has to ensure that the alloca or global is at least that aligned. It has always done this historically when a datalayout is present, but has never checked it when the datalayout is absent. When I refactored the code in r220156, I exposed this path when datalayout was present and that turned the latent bug into a patent bug. This fixes the issue by just removing the special case which allows folding things without datalayout. This isn't worth the complexity of trying to tease apart when it is or isn't safe without actually knowing the preferred alignment. llvm-svn: 220161
2014-10-19 16:17:50 +08:00
; CHECK-NOT: call
; CHECK: }
Improve tail call elim to move loads above readonly calls when it allows forming a tail call. Patch by Frits van Bommel. This implements PR4323. llvm-svn: 73752
2009-06-19 12:22:16 +08:00
entry:
%tmp2 = icmp sge i32 %start_arg, %a_len_arg ; <i1> [#uses=1]
br i1 %tmp2, label %if, label %else
if: ; preds = %entry
ret i32 0
else: ; preds = %entry
%tmp7 = add i32 %start_arg, 1 ; <i32> [#uses=1]
%tmp8 = call fastcc i32 @raise_load_1(i32* %a_arg, i32 %a_len_arg, i32 %tmp7) ; <i32> [#uses=1]
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%tmp9 = load i32, i32* %a_arg ; <i32> [#uses=1]
Improve tail call elim to move loads above readonly calls when it allows forming a tail call. Patch by Frits van Bommel. This implements PR4323. llvm-svn: 73752
2009-06-19 12:22:16 +08:00
%tmp10 = add i32 %tmp9, %tmp8 ; <i32> [#uses=1]
ret i32 %tmp10
}
; This load can be moved above the call because the function won't write to it
; and the load provably can't trap.
define fastcc i32 @raise_load_2(i32* %a_arg, i32 %a_len_arg, i32 %start_arg) readonly {
Fix a long-standing miscompile in the load analysis that was uncovered by my refactoring of this code. The method isSafeToLoadUnconditionally assumes that the load will proceed with the preferred type alignment. Given that, it has to ensure that the alloca or global is at least that aligned. It has always done this historically when a datalayout is present, but has never checked it when the datalayout is absent. When I refactored the code in r220156, I exposed this path when datalayout was present and that turned the latent bug into a patent bug. This fixes the issue by just removing the special case which allows folding things without datalayout. This isn't worth the complexity of trying to tease apart when it is or isn't safe without actually knowing the preferred alignment. llvm-svn: 220161
2014-10-19 16:17:50 +08:00
; CHECK-LABEL: @raise_load_2(
; CHECK-NOT: call
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
; CHECK: load i32, i32*
Fix a long-standing miscompile in the load analysis that was uncovered by my refactoring of this code. The method isSafeToLoadUnconditionally assumes that the load will proceed with the preferred type alignment. Given that, it has to ensure that the alloca or global is at least that aligned. It has always done this historically when a datalayout is present, but has never checked it when the datalayout is absent. When I refactored the code in r220156, I exposed this path when datalayout was present and that turned the latent bug into a patent bug. This fixes the issue by just removing the special case which allows folding things without datalayout. This isn't worth the complexity of trying to tease apart when it is or isn't safe without actually knowing the preferred alignment. llvm-svn: 220161
2014-10-19 16:17:50 +08:00
; CHECK-NOT: call
; CHECK: }
Improve tail call elim to move loads above readonly calls when it allows forming a tail call. Patch by Frits van Bommel. This implements PR4323. llvm-svn: 73752
2009-06-19 12:22:16 +08:00
entry:
%tmp2 = icmp sge i32 %start_arg, %a_len_arg ; <i1> [#uses=1]
br i1 %tmp2, label %if, label %else
if: ; preds = %entry
ret i32 0
else: ; preds = %entry
%nullcheck = icmp eq i32* %a_arg, null ; <i1> [#uses=1]
br i1 %nullcheck, label %unwind, label %recurse
unwind: ; preds = %else
Replace uses of unwind with unreachable for the same effect. llvm-svn: 140077
2011-09-20 06:16:15 +08:00
unreachable
Improve tail call elim to move loads above readonly calls when it allows forming a tail call. Patch by Frits van Bommel. This implements PR4323. llvm-svn: 73752
2009-06-19 12:22:16 +08:00
recurse: ; preds = %else
%tmp7 = add i32 %start_arg, 1 ; <i32> [#uses=1]
%tmp8 = call fastcc i32 @raise_load_2(i32* %a_arg, i32 %a_len_arg, i32 %tmp7) ; <i32> [#uses=1]
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%tmp9 = load i32, i32* @global ; <i32> [#uses=1]
Improve tail call elim to move loads above readonly calls when it allows forming a tail call. Patch by Frits van Bommel. This implements PR4323. llvm-svn: 73752
2009-06-19 12:22:16 +08:00
%tmp10 = add i32 %tmp9, %tmp8 ; <i32> [#uses=1]
ret i32 %tmp10
}
; This load can be safely moved above the call (even though it's from an
; extern_weak global) because the call has no side effects.
define fastcc i32 @raise_load_3(i32* %a_arg, i32 %a_len_arg, i32 %start_arg) nounwind readonly {
Fix a long-standing miscompile in the load analysis that was uncovered by my refactoring of this code. The method isSafeToLoadUnconditionally assumes that the load will proceed with the preferred type alignment. Given that, it has to ensure that the alloca or global is at least that aligned. It has always done this historically when a datalayout is present, but has never checked it when the datalayout is absent. When I refactored the code in r220156, I exposed this path when datalayout was present and that turned the latent bug into a patent bug. This fixes the issue by just removing the special case which allows folding things without datalayout. This isn't worth the complexity of trying to tease apart when it is or isn't safe without actually knowing the preferred alignment. llvm-svn: 220161
2014-10-19 16:17:50 +08:00
; CHECK-LABEL: @raise_load_3(
; CHECK-NOT: call
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
; CHECK: load i32, i32*
Fix a long-standing miscompile in the load analysis that was uncovered by my refactoring of this code. The method isSafeToLoadUnconditionally assumes that the load will proceed with the preferred type alignment. Given that, it has to ensure that the alloca or global is at least that aligned. It has always done this historically when a datalayout is present, but has never checked it when the datalayout is absent. When I refactored the code in r220156, I exposed this path when datalayout was present and that turned the latent bug into a patent bug. This fixes the issue by just removing the special case which allows folding things without datalayout. This isn't worth the complexity of trying to tease apart when it is or isn't safe without actually knowing the preferred alignment. llvm-svn: 220161
2014-10-19 16:17:50 +08:00
; CHECK-NOT: call
; CHECK: }
Improve tail call elim to move loads above readonly calls when it allows forming a tail call. Patch by Frits van Bommel. This implements PR4323. llvm-svn: 73752
2009-06-19 12:22:16 +08:00
entry:
%tmp2 = icmp sge i32 %start_arg, %a_len_arg ; <i1> [#uses=1]
br i1 %tmp2, label %if, label %else
if: ; preds = %entry
ret i32 0
else: ; preds = %entry
%tmp7 = add i32 %start_arg, 1 ; <i32> [#uses=1]
%tmp8 = call fastcc i32 @raise_load_3(i32* %a_arg, i32 %a_len_arg, i32 %tmp7) ; <i32> [#uses=1]
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%tmp9 = load i32, i32* @extern_weak_global ; <i32> [#uses=1]
Improve tail call elim to move loads above readonly calls when it allows forming a tail call. Patch by Frits van Bommel. This implements PR4323. llvm-svn: 73752
2009-06-19 12:22:16 +08:00
%tmp10 = add i32 %tmp9, %tmp8 ; <i32> [#uses=1]
ret i32 %tmp10
}
; The second load can be safely moved above the call even though it's from an
; unknown pointer (which normally means it might trap) because the first load
; proves it doesn't trap.
define fastcc i32 @raise_load_4(i32* %a_arg, i32 %a_len_arg, i32 %start_arg) readonly {
Fix a long-standing miscompile in the load analysis that was uncovered by my refactoring of this code. The method isSafeToLoadUnconditionally assumes that the load will proceed with the preferred type alignment. Given that, it has to ensure that the alloca or global is at least that aligned. It has always done this historically when a datalayout is present, but has never checked it when the datalayout is absent. When I refactored the code in r220156, I exposed this path when datalayout was present and that turned the latent bug into a patent bug. This fixes the issue by just removing the special case which allows folding things without datalayout. This isn't worth the complexity of trying to tease apart when it is or isn't safe without actually knowing the preferred alignment. llvm-svn: 220161
2014-10-19 16:17:50 +08:00
; CHECK-LABEL: @raise_load_4(
; CHECK-NOT: call
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
; CHECK: load i32, i32*
; CHECK-NEXT: load i32, i32*
Fix a long-standing miscompile in the load analysis that was uncovered by my refactoring of this code. The method isSafeToLoadUnconditionally assumes that the load will proceed with the preferred type alignment. Given that, it has to ensure that the alloca or global is at least that aligned. It has always done this historically when a datalayout is present, but has never checked it when the datalayout is absent. When I refactored the code in r220156, I exposed this path when datalayout was present and that turned the latent bug into a patent bug. This fixes the issue by just removing the special case which allows folding things without datalayout. This isn't worth the complexity of trying to tease apart when it is or isn't safe without actually knowing the preferred alignment. llvm-svn: 220161
2014-10-19 16:17:50 +08:00
; CHECK-NOT: call
; CHECK: }
Improve tail call elim to move loads above readonly calls when it allows forming a tail call. Patch by Frits van Bommel. This implements PR4323. llvm-svn: 73752
2009-06-19 12:22:16 +08:00
entry:
%tmp2 = icmp sge i32 %start_arg, %a_len_arg ; <i1> [#uses=1]
br i1 %tmp2, label %if, label %else
if: ; preds = %entry
ret i32 0
else: ; preds = %entry
%nullcheck = icmp eq i32* %a_arg, null ; <i1> [#uses=1]
br i1 %nullcheck, label %unwind, label %recurse
unwind: ; preds = %else
Replace uses of unwind with unreachable for the same effect. llvm-svn: 140077
2011-09-20 06:16:15 +08:00
unreachable
Improve tail call elim to move loads above readonly calls when it allows forming a tail call. Patch by Frits van Bommel. This implements PR4323. llvm-svn: 73752
2009-06-19 12:22:16 +08:00
recurse: ; preds = %else
%tmp7 = add i32 %start_arg, 1 ; <i32> [#uses=1]
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%first = load i32, i32* %a_arg ; <i32> [#uses=1]
Improve tail call elim to move loads above readonly calls when it allows forming a tail call. Patch by Frits van Bommel. This implements PR4323. llvm-svn: 73752
2009-06-19 12:22:16 +08:00
%tmp8 = call fastcc i32 @raise_load_4(i32* %a_arg, i32 %first, i32 %tmp7) ; <i32> [#uses=1]
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%second = load i32, i32* %a_arg ; <i32> [#uses=1]
Improve tail call elim to move loads above readonly calls when it allows forming a tail call. Patch by Frits van Bommel. This implements PR4323. llvm-svn: 73752
2009-06-19 12:22:16 +08:00
%tmp10 = add i32 %second, %tmp8 ; <i32> [#uses=1]
ret i32 %tmp10
}
Push isDereferenceableAndAlignedPointer down into isSafeToLoadUnconditionally Reviewed By: reames Differential Revision: http://reviews.llvm.org/D16226 llvm-svn: 258010
2016-01-17 20:35:29 +08:00
; This load can be moved above the call because the function won't write to it
; and the a_arg is dereferenceable.
define fastcc i32 @raise_load_5(i32* dereferenceable(4) %a_arg, i32 %a_len_arg, i32 %start_arg) readonly {
; CHECK-LABEL: @raise_load_5(
; CHECK-NOT: call
; CHECK: load i32, i32*
; CHECK-NOT: call
; CHECK: }
entry:
%tmp2 = icmp sge i32 %start_arg, %a_len_arg ; <i1> [#uses=1]
br i1 %tmp2, label %if, label %else
if: ; preds = %entry
ret i32 0
else: ; preds = %entry
%tmp7 = add i32 %start_arg, 1 ; <i32> [#uses=1]
%tmp8 = call fastcc i32 @raise_load_5(i32* %a_arg, i32 %a_len_arg, i32 %tmp7) ; <i32> [#uses=1]
%tmp9 = load i32, i32* %a_arg ; <i32> [#uses=1]
%tmp10 = add i32 %tmp9, %tmp8 ; <i32> [#uses=1]
ret i32 %tmp10
}