llvm-project/compiler-rt/lib/xray/xray_powerpc64.cpp
Sebastian Kreutzer e738a5d8e3
Reapply " [XRay] Add support for instrumentation of DSOs on x86_64 (#90959)" (#113548)
This fixes remaining issues in my previous PR #90959.

Changes:
- Removed dependency on LLVM header in `xray_interface.cpp`
- Fixed XRay patching for some targets due to missing changes in
architecture-specific patching functions
- Addressed some remaining compiler warnings that I missed in the
previous patch
- Formatting

I have tested these changes on `x86_64` (natively), as well as
`ppc64le`, `aarch64` and `arm32` (cross-compiled and emulated using
qemu).

**Original description:**

This PR introduces shared library (DSO) support for XRay based on a
revised version of the implementation outlined in [this
RFC](https://discourse.llvm.org/t/rfc-upstreaming-dso-instrumentation-support-for-xray/73000).
The feature enables the patching and handling of events from DSOs,
supporting both libraries linked at startup or explicitly loaded, e.g.
via `dlopen`.
This patch adds the following:
- The `-fxray-shared` flag to enable the feature (turned off by default)
- A small runtime library that is linked into every instrumented DSO,
providing position-independent trampolines and code to register with the
main XRay runtime
- Changes to the XRay runtime to support management and patching of
multiple objects

These changes are fully backward compatible, i.e. running without
instrumented DSOs will produce identical traces (in terms of recorded
function IDs) to the previous implementation.

Due to my limited ability to test on other architectures, this feature
is only implemented and tested with x86_64. Extending support to other
architectures is fairly straightforward, requiring only a
position-independent implementation of the architecture-specific
trampoline implementation (see
`compiler-rt/lib/xray/xray_trampoline_x86_64.S` for reference).

This patch does not include any functionality to resolve function IDs
from DSOs for the provided logging/tracing modes. These modes still work
and will record calls from DSOs, but symbol resolution for these
functions in not available. Getting this to work properly requires
recording information about the loaded DSOs and should IMO be discussed
in a separate RFC, as there are mulitple feasible approaches.

---------

Co-authored-by: Sebastian Kreutzer <sebastian.kreutzer@tu-darmstadt.de>
2024-10-25 10:15:25 +02:00

132 lines
4.6 KiB
C++

//===-- xray_powerpc64.cpp --------------------------------------*- 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 is a part of XRay, a dynamic runtime instrumentation system.
//
// Implementation of powerpc64 and powerpc64le routines.
//
//===----------------------------------------------------------------------===//
#include "sanitizer_common/sanitizer_common.h"
#include "xray_defs.h"
#include "xray_interface_internal.h"
#include "xray_utils.h"
#include <atomic>
#include <cassert>
#include <cstring>
#ifndef __LITTLE_ENDIAN__
#error powerpc64 big endian is not supported for now.
#endif
namespace {
constexpr unsigned long long JumpOverInstNum = 7;
void clearCache(void *Addr, size_t Len) {
const size_t LineSize = 32;
const intptr_t Mask = ~(LineSize - 1);
const intptr_t StartLine = ((intptr_t)Addr) & Mask;
const intptr_t EndLine = ((intptr_t)Addr + Len + LineSize - 1) & Mask;
for (intptr_t Line = StartLine; Line < EndLine; Line += LineSize)
asm volatile("dcbf 0, %0" : : "r"(Line));
asm volatile("sync");
for (intptr_t Line = StartLine; Line < EndLine; Line += LineSize)
asm volatile("icbi 0, %0" : : "r"(Line));
asm volatile("isync");
}
} // namespace
extern "C" void __clear_cache(void *start, void *end);
namespace __xray {
bool patchFunctionEntry(const bool Enable, uint32_t FuncId,
const XRaySledEntry &Sled,
const XRayTrampolines &Trampolines,
bool LogArgs) XRAY_NEVER_INSTRUMENT {
// TODO: Trampoline addresses are currently inserted at compile-time, using
// __xray_FunctionEntry and __xray_FunctionExit only.
// To support DSO instrumentation, trampolines have to be written during
// patching (see implementation on X86_64, e.g.).
const uint64_t Address = Sled.address();
if (Enable) {
// lis 0, FuncId[16..32]
// li 0, FuncId[0..15]
*reinterpret_cast<uint64_t *>(Address) =
(0x3c000000ull + (FuncId >> 16)) +
((0x60000000ull + (FuncId & 0xffff)) << 32);
} else {
// b +JumpOverInstNum instructions.
*reinterpret_cast<uint32_t *>(Address) =
0x48000000ull + (JumpOverInstNum << 2);
}
clearCache(reinterpret_cast<void *>(Address), 8);
return true;
}
bool patchFunctionExit(
const bool Enable, uint32_t FuncId, const XRaySledEntry &Sled,
const XRayTrampolines &Trampolines) XRAY_NEVER_INSTRUMENT {
// TODO: Trampoline addresses are currently inserted at compile-time, using
// __xray_FunctionEntry and __xray_FunctionExit only.
// To support DSO instrumentation, trampolines have to be written during
// patching (see implementation on X86_64, e.g.).
const uint64_t Address = Sled.address();
if (Enable) {
// lis 0, FuncId[16..32]
// li 0, FuncId[0..15]
*reinterpret_cast<uint64_t *>(Address) =
(0x3c000000ull + (FuncId >> 16)) +
((0x60000000ull + (FuncId & 0xffff)) << 32);
} else {
// Copy the blr/b instruction after JumpOverInstNum instructions.
*reinterpret_cast<uint32_t *>(Address) =
*(reinterpret_cast<uint32_t *>(Address) + JumpOverInstNum);
}
clearCache(reinterpret_cast<void *>(Address), 8);
return true;
}
bool patchFunctionTailExit(
const bool Enable, const uint32_t FuncId, const XRaySledEntry &Sled,
const XRayTrampolines &Trampolines) XRAY_NEVER_INSTRUMENT {
return patchFunctionExit(Enable, FuncId, Sled, Trampolines);
}
// FIXME: Maybe implement this better?
bool probeRequiredCPUFeatures() XRAY_NEVER_INSTRUMENT { return true; }
bool patchCustomEvent(const bool Enable, const uint32_t FuncId,
const XRaySledEntry &Sled) XRAY_NEVER_INSTRUMENT {
// FIXME: Implement in powerpc64?
return false;
}
bool patchTypedEvent(const bool Enable, const uint32_t FuncId,
const XRaySledEntry &Sled) XRAY_NEVER_INSTRUMENT {
// FIXME: Implement in powerpc64?
return false;
}
} // namespace __xray
extern "C" void __xray_ArgLoggerEntry() XRAY_NEVER_INSTRUMENT {
// FIXME: this will have to be implemented in the trampoline assembly file
}
extern "C" void __xray_FunctionTailExit() XRAY_NEVER_INSTRUMENT {
// For PowerPC, calls to __xray_FunctionEntry and __xray_FunctionExit
// are statically inserted into the sled. Tail exits are handled like normal
// function exits. This trampoline is therefore not implemented.
// This stub is placed here to avoid linking issues.
}