1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-25 20:23:11 +01:00
llvm-mirror/include/llvm/MC/MCPseudoProbe.h
Hongtao Yu 85e4f6f241 [CSSPGO] Pseudo probe encoding and emission.
This change implements pseudo probe encoding and emission for CSSPGO. Please see RFC here for more context: https://groups.google.com/g/llvm-dev/c/1p1rdYbL93s

Pseudo probes are in the form of intrinsic calls on IR/MIR but they do not turn into any machine instructions. Instead they are emitted into the binary as a piece of data in standalone sections.  The probe-specific sections are not needed to be loaded into memory at execution time, thus they do not incur a runtime overhead. 

**ELF object emission**

The binary data to emit are organized as two ELF sections, i.e, the `.pseudo_probe_desc` section and the `.pseudo_probe` section. The `.pseudo_probe_desc` section stores a function descriptor for each function and the `.pseudo_probe` section stores the actual probes, each fo which corresponds to an IR basic block or an IR function callsite. A function descriptor is stored as a module-level metadata during the compilation and is serialized into the object file during object emission.

Both the probe descriptors and pseudo probes can be emitted into a separate ELF section per function to leverage the linker for deduplication.  A `.pseudo_probe` section shares the same COMDAT group with the function code so that when the function is dead, the probes are dead and disposed too. On the contrary, a `.pseudo_probe_desc` section has its own COMDAT group. This is because even if a function is dead, its probes may be inlined into other functions and its descriptor is still needed by the profile generation tool.

The format of `.pseudo_probe_desc` section looks like:

```
.section   .pseudo_probe_desc,"",@progbits
.quad   6309742469962978389  // Func GUID
.quad   4294967295           // Func Hash
.byte   9                    // Length of func name
.ascii  "_Z5funcAi"          // Func name
.quad   7102633082150537521
.quad   138828622701
.byte   12
.ascii  "_Z8funcLeafi"
.quad   446061515086924981
.quad   4294967295
.byte   9
.ascii  "_Z5funcBi"
.quad   -2016976694713209516
.quad   72617220756
.byte   7
.ascii  "_Z3fibi"
```

For each `.pseudoprobe` section, the encoded binary data consists of a single function record corresponding to an outlined function (i.e, a function with a code entry in the `.text` section). A function record has the following format :

```
FUNCTION BODY (one for each outlined function present in the text section)
    GUID (uint64)
        GUID of the function
    NPROBES (ULEB128)
        Number of probes originating from this function.
    NUM_INLINED_FUNCTIONS (ULEB128)
        Number of callees inlined into this function, aka number of
        first-level inlinees
    PROBE RECORDS
        A list of NPROBES entries. Each entry contains:
          INDEX (ULEB128)
          TYPE (uint4)
            0 - block probe, 1 - indirect call, 2 - direct call
          ATTRIBUTE (uint3)
            reserved
          ADDRESS_TYPE (uint1)
            0 - code address, 1 - address delta
          CODE_ADDRESS (uint64 or ULEB128)
            code address or address delta, depending on ADDRESS_TYPE
    INLINED FUNCTION RECORDS
        A list of NUM_INLINED_FUNCTIONS entries describing each of the inlined
        callees.  Each record contains:
          INLINE SITE
            GUID of the inlinee (uint64)
            ID of the callsite probe (ULEB128)
          FUNCTION BODY
            A FUNCTION BODY entry describing the inlined function.
```

To support building a context-sensitive profile, probes from inlinees are grouped by their inline contexts. An inline context is logically a call path through which a callee function lands in a caller function. The probe emitter builds an inline tree based on the debug metadata for each outlined function in the form of a trie tree. A tree root is the outlined function. Each tree edge stands for a callsite where inlining happens. Pseudo probes originating from an inlinee function are stored in a tree node and the tree path starting from the root all the way down to the tree node is the inline context of the probes. The emission happens on the whole tree top-down recursively. Probes of a tree node will be emitted altogether with their direct parent edge. Since a pseudo probe corresponds to a real code address, for size savings, the address is encoded as a delta from the previous probe except for the first probe. Variant-sized integer encoding, aka LEB128, is used for address delta and probe index.

**Assembling**

Pseudo probes can be printed as assembly directives alternatively. This allows for good assembly code readability and also provides a view of how optimizations and pseudo probes affect each other, especially helpful for diff time assembly analysis.

A pseudo probe directive has the following operands in order: function GUID, probe index, probe type, probe attributes and inline context. The directive is generated by the compiler and can be parsed by the assembler to form an encoded `.pseudoprobe` section in the object file.

A example assembly looks like:

```
foo2: # @foo2
# %bb.0: # %bb0
pushq %rax
testl %edi, %edi
.pseudoprobe 837061429793323041 1 0 0
je .LBB1_1
# %bb.2: # %bb2
.pseudoprobe 837061429793323041 6 2 0
callq foo
.pseudoprobe 837061429793323041 3 0 0
.pseudoprobe 837061429793323041 4 0 0
popq %rax
retq
.LBB1_1: # %bb1
.pseudoprobe 837061429793323041 5 1 0
callq *%rsi
.pseudoprobe 837061429793323041 2 0 0
.pseudoprobe 837061429793323041 4 0 0
popq %rax
retq
# -- End function
.section .pseudo_probe_desc,"",@progbits
.quad 6699318081062747564
.quad 72617220756
.byte 3
.ascii "foo"
.quad 837061429793323041
.quad 281547593931412
.byte 4
.ascii "foo2"
```

With inlining turned on, the assembly may look different around %bb2 with an inlined probe:

```
# %bb.2:                                # %bb2
.pseudoprobe    837061429793323041 3 0
.pseudoprobe    6699318081062747564 1 0 @ 837061429793323041:6
.pseudoprobe    837061429793323041 4 0
popq    %rax
retq
```

**Disassembling**

We have a disassembling tool (llvm-profgen) that can display disassembly alongside with pseudo probes. So far it only supports ELF executable file.

An example disassembly looks like:

```
00000000002011a0 <foo2>:
  2011a0: 50                    push   rax
  2011a1: 85 ff                 test   edi,edi
  [Probe]:  FUNC: foo2  Index: 1  Type: Block
  2011a3: 74 02                 je     2011a7 <foo2+0x7>
  [Probe]:  FUNC: foo2  Index: 3  Type: Block
  [Probe]:  FUNC: foo2  Index: 4  Type: Block
  [Probe]:  FUNC: foo   Index: 1  Type: Block  Inlined: @ foo2:6
  2011a5: 58                    pop    rax
  2011a6: c3                    ret
  [Probe]:  FUNC: foo2  Index: 2  Type: Block
  2011a7: bf 01 00 00 00        mov    edi,0x1
  [Probe]:  FUNC: foo2  Index: 5  Type: IndirectCall
  2011ac: ff d6                 call   rsi
  [Probe]:  FUNC: foo2  Index: 4  Type: Block
  2011ae: 58                    pop    rax
  2011af: c3                    ret
```

Reviewed By: wmi

Differential Revision: https://reviews.llvm.org/D91878
2020-12-10 17:29:28 -08:00

179 lines
6.5 KiB
C++

//===- MCPseudoProbe.h - Pseudo probe encoding support ---------*- C++ -*-===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
// This file contains the declaration of the MCPseudoProbe to support the pseudo
// probe encoding for AutoFDO. Pseudo probes together with their inline context
// are encoded in a DFS recursive way in the .pseudoprobe sections. For each
// .pseudoprobe section, the encoded binary data consist of a single or mutiple
// function records each for one outlined function. A function record has the
// following format :
//
// FUNCTION BODY (one for each outlined function present in the text section)
// GUID (uint64)
// GUID of the function
// NPROBES (ULEB128)
// Number of probes originating from this function.
// NUM_INLINED_FUNCTIONS (ULEB128)
// Number of callees inlined into this function, aka number of
// first-level inlinees
// PROBE RECORDS
// A list of NPROBES entries. Each entry contains:
// INDEX (ULEB128)
// TYPE (uint4)
// 0 - block probe, 1 - indirect call, 2 - direct call
// ATTRIBUTE (uint3)
// reserved
// ADDRESS_TYPE (uint1)
// 0 - code address, 1 - address delta
// CODE_ADDRESS (uint64 or ULEB128)
// code address or address delta, depending on ADDRESS_TYPE
// INLINED FUNCTION RECORDS
// A list of NUM_INLINED_FUNCTIONS entries describing each of the inlined
// callees. Each record contains:
// INLINE SITE
// GUID of the inlinee (uint64)
// ID of the callsite probe (ULEB128)
// FUNCTION BODY
// A FUNCTION BODY entry describing the inlined function.
//===----------------------------------------------------------------------===//
#ifndef LLVM_MC_MCPSEUDOPROBE_H
#define LLVM_MC_MCPSEUDOPROBE_H
#include "llvm/ADT/MapVector.h"
#include "llvm/MC/MCSection.h"
#include <functional>
#include <map>
#include <vector>
namespace llvm {
class MCStreamer;
class MCSymbol;
class MCObjectStreamer;
enum class MCPseudoProbeFlag {
// If set, indicates that the probe is encoded as an address delta
// instead of a real code address.
AddressDelta = 0x1,
};
/// Instances of this class represent a pseudo probe instance for a pseudo probe
/// table entry, which is created during a machine instruction is assembled and
/// uses an address from a temporary label created at the current address in the
/// current section.
class MCPseudoProbe {
MCSymbol *Label;
uint64_t Guid;
uint64_t Index;
uint8_t Type;
uint8_t Attributes;
public:
MCPseudoProbe(MCSymbol *Label, uint64_t Guid, uint64_t Index, uint64_t Type,
uint64_t Attributes)
: Label(Label), Guid(Guid), Index(Index), Type(Type),
Attributes(Attributes) {
assert(Type <= 0xFF && "Probe type too big to encode, exceeding 2^8");
assert(Attributes <= 0xFF &&
"Probe attributes too big to encode, exceeding 2^16");
}
MCSymbol *getLabel() const { return Label; }
uint64_t getGuid() const { return Guid; }
uint64_t getIndex() const { return Index; }
uint8_t getType() const { return Type; }
uint8_t getAttributes() const { return Attributes; }
void emit(MCObjectStreamer *MCOS, const MCPseudoProbe *LastProbe) const;
};
// An inline frame has the form <Guid, ProbeID>
using InlineSite = std::tuple<uint64_t, uint32_t>;
using MCPseudoProbeInlineStack = SmallVector<InlineSite, 8>;
// A Tri-tree based data structure to group probes by inline stack.
// A tree is allocated for a standalone .text section. A fake
// instance is created as the root of a tree.
// A real instance of this class is created for each function, either an
// unlined function that has code in .text section or an inlined function.
class MCPseudoProbeInlineTree {
uint64_t Guid;
// Set of probes that come with the function.
std::vector<MCPseudoProbe> Probes;
// Use std::map for a deterministic output.
std::map<InlineSite, MCPseudoProbeInlineTree *> Inlinees;
// Root node has a GUID 0.
bool isRoot() { return Guid == 0; }
MCPseudoProbeInlineTree *getOrAddNode(InlineSite Site);
public:
MCPseudoProbeInlineTree() = default;
MCPseudoProbeInlineTree(uint64_t Guid) : Guid(Guid) {}
~MCPseudoProbeInlineTree();
void addPseudoProbe(const MCPseudoProbe &Probe,
const MCPseudoProbeInlineStack &InlineStack);
void emit(MCObjectStreamer *MCOS, const MCPseudoProbe *&LastProbe);
};
/// Instances of this class represent the pseudo probes inserted into a compile
/// unit.
class MCPseudoProbeSection {
public:
void addPseudoProbe(MCSection *Sec, const MCPseudoProbe &Probe,
const MCPseudoProbeInlineStack &InlineStack) {
MCProbeDivisions[Sec].addPseudoProbe(Probe, InlineStack);
}
// TODO: Sort by getOrdinal to ensure a determinstic section order
using MCProbeDivisionMap = std::map<MCSection *, MCPseudoProbeInlineTree>;
private:
// A collection of MCPseudoProbe for each text section. The MCPseudoProbes
// are grouped by GUID of the functions where they are from and will be
// encoded by groups. In the comdat scenario where a text section really only
// contains the code of a function solely, the probes associated with a comdat
// function are still grouped by GUIDs due to inlining that can bring probes
// from different functions into one function.
MCProbeDivisionMap MCProbeDivisions;
public:
const MCProbeDivisionMap &getMCProbes() const { return MCProbeDivisions; }
bool empty() const { return MCProbeDivisions.empty(); }
void emit(MCObjectStreamer *MCOS);
};
class MCPseudoProbeTable {
// A collection of MCPseudoProbe in the current module grouped by text
// sections. MCPseudoProbes will be encoded into a corresponding
// .pseudoprobe section. With functions emitted as separate comdats,
// a text section really only contains the code of a function solely, and the
// probes associated with the text section will be emitted into a standalone
// .pseudoprobe section that shares the same comdat group with the function.
MCPseudoProbeSection MCProbeSections;
public:
static void emit(MCObjectStreamer *MCOS);
MCPseudoProbeSection &getProbeSections() { return MCProbeSections; }
#ifndef NDEBUG
static int DdgPrintIndent;
#endif
};
} // end namespace llvm
#endif // LLVM_MC_MCPSEUDOPROBE_H