llvm-project/compiler-rt/lib/xray/xray_buffer_queue.h

160 lines
4.8 KiB
C
Raw Normal View History

//===-- xray_buffer_queue.h ------------------------------------*- C++ -*-===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file is a part of XRay, a dynamic runtime instrumentation system.
//
// Defines the interface for a buffer queue implementation.
//
//===----------------------------------------------------------------------===//
#ifndef XRAY_BUFFER_QUEUE_H
#define XRAY_BUFFER_QUEUE_H
#include <cstddef>
#include "sanitizer_common/sanitizer_atomic.h"
#include "sanitizer_common/sanitizer_mutex.h"
namespace __xray {
/// BufferQueue implements a circular queue of fixed sized buffers (much like a
/// freelist) but is concerned mostly with making it really quick to initialise,
/// finalise, and get/return buffers to the queue. This is one key component of
/// the "flight data recorder" (FDR) mode to support ongoing XRay function call
/// trace collection.
class BufferQueue {
public:
[XRay] Use optimistic logging model for FDR mode Summary: Before this change, the FDR mode implementation relied on at thread-exit handling to return buffers back to the (global) buffer queue. This introduces issues with the initialisation of the thread_local objects which, even through the use of pthread_setspecific(...) may eventually call into an allocation function. Similar to previous changes in this line, we're finding that there is a huge potential for deadlocks when initialising these thread-locals when the memory allocation implementation is also xray-instrumented. In this change, we limit the call to pthread_setspecific(...) to provide a non-null value to associate to the key created with pthread_key_create(...). While this doesn't completely eliminate the potential for the deadlock(s), it does allow us to still clean up at thread exit when we need to. The change is that we don't need to do more work when starting and ending a thread's lifetime. We also have a test to make sure that we actually can safely recycle the buffers in case we end up re-using the buffer(s) available from the queue on multiple thread entry/exits. This change cuts across both LLVM and compiler-rt to allow us to update both the XRay runtime implementation as well as the library support for loading these new versions of the FDR mode logging. Version 2 of the FDR logging implementation makes the following changes: * Introduction of a new 'BufferExtents' metadata record that's outside of the buffer's contents but are written before the actual buffer. This data is associated to the Buffer handed out by the BufferQueue rather than a record that occupies bytes in the actual buffer. * Removal of the "end of buffer" records. This is in-line with the changes we described above, to allow for optimistic logging without explicit record writing at thread exit. The optimistic logging model operates under the following assumptions: * Threads writing to the buffers will potentially race with the thread attempting to flush the log. To avoid this situation from occuring, we make sure that when we've finalized the logging implementation, that threads will see this finalization state on the next write, and either choose to not write records the thread would have written or write the record(s) in two phases -- first write the record(s), then update the extents metadata. * We change the buffer queue implementation so that once it's handed out a buffer to a thread, that we assume that buffer is marked "used" to be able to capture partial writes. None of this will be safe to handle if threads are racing to write the extents records and the reader thread is attempting to flush the log. The optimism comes from the finalization routine being required to complete before we attempt to flush the log. This is a fairly significant semantics change for the FDR implementation. This is why we've decided to update the version number for FDR mode logs. The tools, however, still need to be able to support older versions of the log until we finally deprecate those earlier versions. Reviewers: dblaikie, pelikan, kpw Subscribers: llvm-commits, hiraditya Differential Revision: https://reviews.llvm.org/D39526 llvm-svn: 318733
2017-11-21 15:16:57 +08:00
struct alignas(64) BufferExtents {
__sanitizer::atomic_uint64_t Size;
};
struct Buffer {
void *Buffer = nullptr;
size_t Size = 0;
[XRay] Use optimistic logging model for FDR mode Summary: Before this change, the FDR mode implementation relied on at thread-exit handling to return buffers back to the (global) buffer queue. This introduces issues with the initialisation of the thread_local objects which, even through the use of pthread_setspecific(...) may eventually call into an allocation function. Similar to previous changes in this line, we're finding that there is a huge potential for deadlocks when initialising these thread-locals when the memory allocation implementation is also xray-instrumented. In this change, we limit the call to pthread_setspecific(...) to provide a non-null value to associate to the key created with pthread_key_create(...). While this doesn't completely eliminate the potential for the deadlock(s), it does allow us to still clean up at thread exit when we need to. The change is that we don't need to do more work when starting and ending a thread's lifetime. We also have a test to make sure that we actually can safely recycle the buffers in case we end up re-using the buffer(s) available from the queue on multiple thread entry/exits. This change cuts across both LLVM and compiler-rt to allow us to update both the XRay runtime implementation as well as the library support for loading these new versions of the FDR mode logging. Version 2 of the FDR logging implementation makes the following changes: * Introduction of a new 'BufferExtents' metadata record that's outside of the buffer's contents but are written before the actual buffer. This data is associated to the Buffer handed out by the BufferQueue rather than a record that occupies bytes in the actual buffer. * Removal of the "end of buffer" records. This is in-line with the changes we described above, to allow for optimistic logging without explicit record writing at thread exit. The optimistic logging model operates under the following assumptions: * Threads writing to the buffers will potentially race with the thread attempting to flush the log. To avoid this situation from occuring, we make sure that when we've finalized the logging implementation, that threads will see this finalization state on the next write, and either choose to not write records the thread would have written or write the record(s) in two phases -- first write the record(s), then update the extents metadata. * We change the buffer queue implementation so that once it's handed out a buffer to a thread, that we assume that buffer is marked "used" to be able to capture partial writes. None of this will be safe to handle if threads are racing to write the extents records and the reader thread is attempting to flush the log. The optimism comes from the finalization routine being required to complete before we attempt to flush the log. This is a fairly significant semantics change for the FDR implementation. This is why we've decided to update the version number for FDR mode logs. The tools, however, still need to be able to support older versions of the log until we finally deprecate those earlier versions. Reviewers: dblaikie, pelikan, kpw Subscribers: llvm-commits, hiraditya Differential Revision: https://reviews.llvm.org/D39526 llvm-svn: 318733
2017-11-21 15:16:57 +08:00
BufferExtents* Extents;
};
private:
struct BufferRep {
// The managed buffer.
Buffer Buff;
// This is true if the buffer has been returned to the available queue, and
// is considered "used" by another thread.
bool Used = false;
};
// Size of each individual Buffer.
size_t BufferSize;
[XRay][compiler-rt] XRay Flight Data Recorder Mode Summary: In this change we introduce the notion of a "flight data recorder" mode for XRay logging, where XRay logs in-memory first, and write out data on-demand as required (as opposed to the naive implementation that keeps logging while tracing is "on"). This depends on D26232 where we implement the core data structure for holding the buffers that threads will be using to write out records of operation. This implementation only currently works on x86_64 and depends heavily on the TSC math to write out smaller records to the inmemory buffers. Also, this implementation defines two different kinds of records with different sizes (compared to the current naive implementation): a MetadataRecord (16 bytes) and a FunctionRecord (8 bytes). MetadataRecord entries are meant to write out information like the thread ID for which the metadata record is defined for, whether the execution of a thread moved to a different CPU, etc. while a FunctionRecord represents the different kinds of function call entry/exit records we might encounter in the course of a thread's execution along with a delta from the last time the logging handler was called. While this implementation is not exactly what is described in the original XRay whitepaper, this one gives us an initial implementation that we can iterate and build upon. Reviewers: echristo, rSerge, majnemer Subscribers: mehdi_amini, llvm-commits, mgorny Differential Revision: https://reviews.llvm.org/D27038 llvm-svn: 293015
2017-01-25 11:50:46 +08:00
BufferRep *Buffers;
size_t BufferCount;
__sanitizer::SpinMutex Mutex;
__sanitizer::atomic_uint8_t Finalizing;
// Pointers to buffers managed/owned by the BufferQueue.
void **OwnedBuffers;
// Pointer to the next buffer to be handed out.
BufferRep *Next;
// Pointer to the entry in the array where the next released buffer will be
// placed.
BufferRep *First;
// Count of buffers that have been handed out through 'getBuffer'.
size_t LiveBuffers;
public:
enum class ErrorCode : unsigned {
Ok,
NotEnoughMemory,
QueueFinalizing,
UnrecognizedBuffer,
AlreadyFinalized,
};
static const char *getErrorString(ErrorCode E) {
switch (E) {
case ErrorCode::Ok:
return "(none)";
case ErrorCode::NotEnoughMemory:
return "no available buffers in the queue";
case ErrorCode::QueueFinalizing:
return "queue already finalizing";
case ErrorCode::UnrecognizedBuffer:
return "buffer being returned not owned by buffer queue";
case ErrorCode::AlreadyFinalized:
return "queue already finalized";
}
return "unknown error";
}
[XRay][compiler-rt] XRay Flight Data Recorder Mode Summary: In this change we introduce the notion of a "flight data recorder" mode for XRay logging, where XRay logs in-memory first, and write out data on-demand as required (as opposed to the naive implementation that keeps logging while tracing is "on"). This depends on D26232 where we implement the core data structure for holding the buffers that threads will be using to write out records of operation. This implementation only currently works on x86_64 and depends heavily on the TSC math to write out smaller records to the inmemory buffers. Also, this implementation defines two different kinds of records with different sizes (compared to the current naive implementation): a MetadataRecord (16 bytes) and a FunctionRecord (8 bytes). MetadataRecord entries are meant to write out information like the thread ID for which the metadata record is defined for, whether the execution of a thread moved to a different CPU, etc. while a FunctionRecord represents the different kinds of function call entry/exit records we might encounter in the course of a thread's execution along with a delta from the last time the logging handler was called. While this implementation is not exactly what is described in the original XRay whitepaper, this one gives us an initial implementation that we can iterate and build upon. Reviewers: echristo, rSerge, majnemer Subscribers: mehdi_amini, llvm-commits, mgorny Differential Revision: https://reviews.llvm.org/D27038 llvm-svn: 293015
2017-01-25 11:50:46 +08:00
/// Initialise a queue of size |N| with buffers of size |B|. We report success
/// through |Success|.
BufferQueue(size_t B, size_t N, bool &Success);
/// Updates |Buf| to contain the pointer to an appropriate buffer. Returns an
/// error in case there are no available buffers to return when we will run
/// over the upper bound for the total buffers.
///
/// Requirements:
/// - BufferQueue is not finalising.
///
/// Returns:
/// - ErrorCode::NotEnoughMemory on exceeding MaxSize.
/// - ErrorCode::Ok when we find a Buffer.
/// - ErrorCode::QueueFinalizing or ErrorCode::AlreadyFinalized on
/// a finalizing/finalized BufferQueue.
ErrorCode getBuffer(Buffer &Buf);
/// Updates |Buf| to point to nullptr, with size 0.
///
/// Returns:
/// - ErrorCode::Ok when we successfully release the buffer.
/// - ErrorCode::UnrecognizedBuffer for when this BufferQueue does not own
/// the buffer being released.
ErrorCode releaseBuffer(Buffer &Buf);
bool finalizing() const {
return __sanitizer::atomic_load(&Finalizing,
__sanitizer::memory_order_acquire);
}
/// Returns the configured size of the buffers in the buffer queue.
size_t ConfiguredBufferSize() const { return BufferSize; }
[XRay][compiler-rt] XRay Flight Data Recorder Mode Summary: In this change we introduce the notion of a "flight data recorder" mode for XRay logging, where XRay logs in-memory first, and write out data on-demand as required (as opposed to the naive implementation that keeps logging while tracing is "on"). This depends on D26232 where we implement the core data structure for holding the buffers that threads will be using to write out records of operation. This implementation only currently works on x86_64 and depends heavily on the TSC math to write out smaller records to the inmemory buffers. Also, this implementation defines two different kinds of records with different sizes (compared to the current naive implementation): a MetadataRecord (16 bytes) and a FunctionRecord (8 bytes). MetadataRecord entries are meant to write out information like the thread ID for which the metadata record is defined for, whether the execution of a thread moved to a different CPU, etc. while a FunctionRecord represents the different kinds of function call entry/exit records we might encounter in the course of a thread's execution along with a delta from the last time the logging handler was called. While this implementation is not exactly what is described in the original XRay whitepaper, this one gives us an initial implementation that we can iterate and build upon. Reviewers: echristo, rSerge, majnemer Subscribers: mehdi_amini, llvm-commits, mgorny Differential Revision: https://reviews.llvm.org/D27038 llvm-svn: 293015
2017-01-25 11:50:46 +08:00
/// Sets the state of the BufferQueue to finalizing, which ensures that:
///
/// - All subsequent attempts to retrieve a Buffer will fail.
/// - All releaseBuffer operations will not fail.
///
/// After a call to finalize succeeds, all subsequent calls to finalize will
/// fail with ErrorCode::QueueFinalizing.
ErrorCode finalize();
[XRay][compiler-rt] XRay Flight Data Recorder Mode Summary: In this change we introduce the notion of a "flight data recorder" mode for XRay logging, where XRay logs in-memory first, and write out data on-demand as required (as opposed to the naive implementation that keeps logging while tracing is "on"). This depends on D26232 where we implement the core data structure for holding the buffers that threads will be using to write out records of operation. This implementation only currently works on x86_64 and depends heavily on the TSC math to write out smaller records to the inmemory buffers. Also, this implementation defines two different kinds of records with different sizes (compared to the current naive implementation): a MetadataRecord (16 bytes) and a FunctionRecord (8 bytes). MetadataRecord entries are meant to write out information like the thread ID for which the metadata record is defined for, whether the execution of a thread moved to a different CPU, etc. while a FunctionRecord represents the different kinds of function call entry/exit records we might encounter in the course of a thread's execution along with a delta from the last time the logging handler was called. While this implementation is not exactly what is described in the original XRay whitepaper, this one gives us an initial implementation that we can iterate and build upon. Reviewers: echristo, rSerge, majnemer Subscribers: mehdi_amini, llvm-commits, mgorny Differential Revision: https://reviews.llvm.org/D27038 llvm-svn: 293015
2017-01-25 11:50:46 +08:00
/// Applies the provided function F to each Buffer in the queue, only if the
/// Buffer is marked 'used' (i.e. has been the result of getBuffer(...) and a
/// releaseBuffer(...) operation).
template <class F>
void apply(F Fn) {
__sanitizer::SpinMutexLock G(&Mutex);
for (auto I = Buffers, E = Buffers + BufferCount; I != E; ++I) {
const auto &T = *I;
if (T.Used) Fn(T.Buff);
[XRay][compiler-rt] XRay Flight Data Recorder Mode Summary: In this change we introduce the notion of a "flight data recorder" mode for XRay logging, where XRay logs in-memory first, and write out data on-demand as required (as opposed to the naive implementation that keeps logging while tracing is "on"). This depends on D26232 where we implement the core data structure for holding the buffers that threads will be using to write out records of operation. This implementation only currently works on x86_64 and depends heavily on the TSC math to write out smaller records to the inmemory buffers. Also, this implementation defines two different kinds of records with different sizes (compared to the current naive implementation): a MetadataRecord (16 bytes) and a FunctionRecord (8 bytes). MetadataRecord entries are meant to write out information like the thread ID for which the metadata record is defined for, whether the execution of a thread moved to a different CPU, etc. while a FunctionRecord represents the different kinds of function call entry/exit records we might encounter in the course of a thread's execution along with a delta from the last time the logging handler was called. While this implementation is not exactly what is described in the original XRay whitepaper, this one gives us an initial implementation that we can iterate and build upon. Reviewers: echristo, rSerge, majnemer Subscribers: mehdi_amini, llvm-commits, mgorny Differential Revision: https://reviews.llvm.org/D27038 llvm-svn: 293015
2017-01-25 11:50:46 +08:00
}
}
// Cleans up allocated buffers.
~BufferQueue();
};
} // namespace __xray
#endif // XRAY_BUFFER_QUEUE_H