llvm-project/llvm/test/Transforms/SampleProfile/fnptr.ll

158 lines
6.5 KiB
LLVM
Raw Normal View History

; The two profiles used in this test are the same but encoded in different
; formats. This checks that we produce the same profile annotations regardless
; of the profile format.
;
; RUN: opt < %s -sample-profile -sample-profile-file=%S/Inputs/fnptr.prof | opt -analyze -branch-prob | FileCheck %s
; RUN: opt < %s -sample-profile -sample-profile-file=%S/Inputs/fnptr.binprof | opt -analyze -branch-prob | FileCheck %s
; RUN: opt < %s -passes=sample-profile -sample-profile-file=%S/Inputs/fnptr.prof | opt -analyze -branch-prob | FileCheck %s
; RUN: opt < %s -passes=sample-profile -sample-profile-file=%S/Inputs/fnptr.binprof | opt -analyze -branch-prob | FileCheck %s
; CHECK: edge for.body3 -> if.then probability is 0x1a56a56a / 0x80000000 = 20.58%
; CHECK: edge for.body3 -> if.else probability is 0x65a95a96 / 0x80000000 = 79.42%
; CHECK: edge for.inc -> for.inc12 probability is 0x000fbd1c / 0x80000000 = 0.05%
; CHECK: edge for.inc -> for.body3 probability is 0x7ff042e4 / 0x80000000 = 99.95%
; CHECK: edge for.inc12 -> for.end14 probability is 0x04000000 / 0x80000000 = 3.12%
; CHECK: edge for.inc12 -> for.cond1.preheader probability is 0x7c000000 / 0x80000000 = 96.88%
; Original C++ test case.
;
; #include <stdlib.h>
; #include <math.h>
; #include <stdio.h>
;
; #define N 10000
; #define M 6000
;
; double foo(int x) {
; return x * sin((double)x);
; }
;
; double bar(int x) {
; return x - cos((double)x);
; }
;
; int main() {
; double (*fptr)(int);
; double S = 0;
; for (int i = 0; i < N; i++)
; for (int j = 0; j < M; j++) {
; fptr = (rand() % 100 < 30) ? foo : bar;
; if (rand() % 100 < 10)
; S += (*fptr)(i + j * 300);
; else
; S += (*fptr)(i - j / 840);
; }
; printf("S = %lf\n", S);
; return 0;
; }
@.str = private unnamed_addr constant [9 x i8] c"S = %lf\0A\00", align 1
define double @_Z3fooi(i32 %x) #0 !dbg !3 {
entry:
%conv = sitofp i32 %x to double, !dbg !2
%call = tail call double @sin(double %conv) #3, !dbg !8
%mul = fmul double %conv, %call, !dbg !8
ret double %mul, !dbg !8
}
declare double @sin(double) #1
define double @_Z3bari(i32 %x) #0 !dbg !10 {
entry:
%conv = sitofp i32 %x to double, !dbg !9
%call = tail call double @cos(double %conv) #3, !dbg !11
%sub = fsub double %conv, %call, !dbg !11
ret double %sub, !dbg !11
}
declare double @cos(double) #1
define i32 @main() #2 !dbg !13 {
entry:
br label %for.cond1.preheader, !dbg !12
for.cond1.preheader: ; preds = %for.inc12, %entry
%i.025 = phi i32 [ 0, %entry ], [ %inc13, %for.inc12 ]
%S.024 = phi double [ 0.000000e+00, %entry ], [ %S.2.lcssa, %for.inc12 ]
br label %for.body3, !dbg !14
for.body3: ; preds = %for.inc, %for.cond1.preheader
%j.023 = phi i32 [ 0, %for.cond1.preheader ], [ %inc, %for.inc ]
%S.122 = phi double [ %S.024, %for.cond1.preheader ], [ %S.2, %for.inc ]
%call = tail call i32 @rand() #3, !dbg !15
%rem = srem i32 %call, 100, !dbg !15
%cmp4 = icmp slt i32 %rem, 30, !dbg !15
%_Z3fooi._Z3bari = select i1 %cmp4, double (i32)* @_Z3fooi, double (i32)* @_Z3bari, !dbg !15
%call5 = tail call i32 @rand() #3, !dbg !16
%rem6 = srem i32 %call5, 100, !dbg !16
%cmp7 = icmp slt i32 %rem6, 10, !dbg !16
br i1 %cmp7, label %if.then, label %if.else, !dbg !16
if.then: ; preds = %for.body3
%mul = mul nsw i32 %j.023, 300, !dbg !18
%add = add nsw i32 %mul, %i.025, !dbg !18
%call8 = tail call double %_Z3fooi._Z3bari(i32 %add), !dbg !18
br label %for.inc, !dbg !18
if.else: ; preds = %for.body3
%div = sdiv i32 %j.023, 840, !dbg !19
%sub = sub nsw i32 %i.025, %div, !dbg !19
%call10 = tail call double %_Z3fooi._Z3bari(i32 %sub), !dbg !19
br label %for.inc
for.inc: ; preds = %if.then, %if.else
%call8.pn = phi double [ %call8, %if.then ], [ %call10, %if.else ]
%S.2 = fadd double %S.122, %call8.pn, !dbg !18
%inc = add nsw i32 %j.023, 1, !dbg !20
%exitcond = icmp eq i32 %j.023, 5999, !dbg !14
br i1 %exitcond, label %for.inc12, label %for.body3, !dbg !14
for.inc12: ; preds = %for.inc
%S.2.lcssa = phi double [ %S.2, %for.inc ]
%inc13 = add nsw i32 %i.025, 1, !dbg !22
%exitcond26 = icmp eq i32 %i.025, 9999, !dbg !12
br i1 %exitcond26, label %for.end14, label %for.cond1.preheader, !dbg !12
for.end14: ; preds = %for.inc12
%S.2.lcssa.lcssa = phi double [ %S.2.lcssa, %for.inc12 ]
[opaque pointer type] Add textual IR support for explicit type parameter to the call instruction See r230786 and r230794 for similar changes to gep and load respectively. Call is a bit different because it often doesn't have a single explicit type - usually the type is deduced from the arguments, and just the return type is explicit. In those cases there's no need to change the IR. When that's not the case, the IR usually contains the pointer type of the first operand - but since typed pointers are going away, that representation is insufficient so I'm just stripping the "pointerness" of the explicit type away. This does make the IR a bit weird - it /sort of/ reads like the type of the first operand: "call void () %x(" but %x is actually of type "void ()*" and will eventually be just of type "ptr". But this seems not too bad and I don't think it would benefit from repeating the type ("void (), void () * %x(" and then eventually "void (), ptr %x(") as has been done with gep and load. This also has a side benefit: since the explicit type is no longer a pointer, there's no ambiguity between an explicit type and a function that returns a function pointer. Previously this case needed an explicit type (eg: a function returning a void() function was written as "call void () () * @x(" rather than "call void () * @x(" because of the ambiguity between a function returning a pointer to a void() function and a function returning void). No ambiguity means even function pointer return types can just be written alone, without writing the whole function's type. This leaves /only/ the varargs case where the explicit type is required. Given the special type syntax in call instructions, the regex-fu used for migration was a bit more involved in its own unique way (as every one of these is) so here it is. Use it in conjunction with the apply.sh script and associated find/xargs commands I've provided in rr230786 to migrate your out of tree tests. Do let me know if any of this doesn't cover your cases & we can iterate on a more general script/regexes to help others with out of tree tests. About 9 test cases couldn't be automatically migrated - half of those were functions returning function pointers, where I just had to manually delete the function argument types now that we didn't need an explicit function type there. The other half were typedefs of function types used in calls - just had to manually drop the * from those. import fileinput import sys import re pat = re.compile(r'((?:=|:|^|\s)call\s(?:[^@]*?))(\s*$|\s*(?:(?:\[\[[a-zA-Z0-9_]+\]\]|[@%](?:(")?[\\\?@a-zA-Z0-9_.]*?(?(3)"|)|{{.*}}))(?:\(|$)|undef|inttoptr|bitcast|null|asm).*$)') addrspace_end = re.compile(r"addrspace\(\d+\)\s*\*$") func_end = re.compile("(?:void.*|\)\s*)\*$") def conv(match, line): if not match or re.search(addrspace_end, match.group(1)) or not re.search(func_end, match.group(1)): return line return line[:match.start()] + match.group(1)[:match.group(1).rfind('*')].rstrip() + match.group(2) + line[match.end():] for line in sys.stdin: sys.stdout.write(conv(re.search(pat, line), line)) llvm-svn: 235145
2015-04-17 07:24:18 +08:00
%call15 = tail call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([9 x i8], [9 x i8]* @.str, i64 0, i64 0), double %S.2.lcssa.lcssa), !dbg !24
ret i32 0, !dbg !25
}
; Function Attrs: nounwind
declare i32 @rand() #1
; Function Attrs: nounwind
declare i32 @printf(i8* nocapture readonly, ...) #1
!llvm.module.flags = !{!0}
!llvm.ident = !{!1}
!llvm.dbg.cu = !{!26}
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!0 = !{i32 2, !"Debug Info Version", i32 3}
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!1 = !{!"clang version 3.6.0 "}
!2 = !DILocation(line: 9, column: 3, scope: !3)
!3 = distinct !DISubprogram(name: "foo", line: 8, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: true, unit: !26, scopeLine: 8, file: !4, scope: !5, type: !6, variables: !7)
!4 = !DIFile(filename: "fnptr.cc", directory: ".")
!5 = !DIFile(filename: "fnptr.cc", directory: ".")
!6 = !DISubroutineType(types: !7)
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!7 = !{}
!8 = !DILocation(line: 9, column: 14, scope: !3)
!9 = !DILocation(line: 13, column: 3, scope: !10)
!10 = distinct !DISubprogram(name: "bar", line: 12, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: true, unit: !26, scopeLine: 12, file: !4, scope: !5, type: !6, variables: !7)
!11 = !DILocation(line: 13, column: 14, scope: !10)
!12 = !DILocation(line: 19, column: 3, scope: !13)
!13 = distinct !DISubprogram(name: "main", line: 16, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: true, unit: !26, scopeLine: 16, file: !4, scope: !5, type: !6, variables: !7)
!14 = !DILocation(line: 20, column: 5, scope: !13)
!15 = !DILocation(line: 21, column: 15, scope: !13)
!16 = !DILocation(line: 22, column: 11, scope: !13)
!18 = !DILocation(line: 23, column: 14, scope: !13)
!19 = !DILocation(line: 25, column: 14, scope: !13)
!20 = !DILocation(line: 20, column: 28, scope: !13)
!22 = !DILocation(line: 19, column: 26, scope: !13)
!24 = !DILocation(line: 27, column: 3, scope: !13)
!25 = !DILocation(line: 28, column: 3, scope: !13)
!26 = distinct !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.5 ", isOptimized: false, emissionKind: FullDebug, file: !4)