3932a9c074
Previously, chatty had logic that would skip a certain number of log entries to satify pruning, but otherwise keep the reader connected. This was a best-effort attempt at helping pruning and had additional logic that handled further disruptions, if logd's memory was 2x the allotted memory. The new logic has two components: 1) memcpy() each individual log message in FlushTo() such that there are no references to the underlying log data without a lock held. Note, that this memcpy is completely negligible for performance. 2) In Prune(), immediately delete all log chunks required to reduce memory to the allotted amount, which is now safe given 1). If readers will lose logs, continue to print a warning. This additionally makes the Clear() logic deterministic. It was previously best effort in chatty, but will immediately and always clear all logs for SerializedLogBuffer. Bug: 163617910 Test: logging unit tests Test: Prune() immediately frees buffers during high log pressure Test: Clear() immediately frees buffers during high log pressure Change-Id: I40fe9b791312af3dc256b166e5c34425f4ca51ac |
||
---|---|---|
.. | ||
doc_images | ||
fuzz | ||
.clang-format | ||
Android.bp | ||
AndroidTest.xml | ||
auditctl.cpp | ||
ChattyLogBuffer.cpp | ||
ChattyLogBuffer.h | ||
ChattyLogBufferTest.cpp | ||
CommandListener.cpp | ||
CommandListener.h | ||
CompressionEngine.cpp | ||
CompressionEngine.h | ||
event.logtags | ||
libaudit.cpp | ||
libaudit.h | ||
LogAudit.cpp | ||
LogAudit.h | ||
LogBuffer.h | ||
LogBufferElement.cpp | ||
LogBufferElement.h | ||
LogBufferTest.cpp | ||
LogBufferTest.h | ||
logd.rc | ||
logd_test.cpp | ||
LogKlog.cpp | ||
LogKlog.h | ||
LogListener.cpp | ||
LogListener.h | ||
LogPermissions.cpp | ||
LogPermissions.h | ||
LogReader.cpp | ||
LogReader.h | ||
LogReaderList.cpp | ||
LogReaderList.h | ||
LogReaderThread.cpp | ||
LogReaderThread.h | ||
LogSize.cpp | ||
LogSize.h | ||
LogStatistics.cpp | ||
LogStatistics.h | ||
logtagd.rc | ||
LogTags.cpp | ||
LogTags.h | ||
LogUtils.h | ||
LogWriter.h | ||
main.cpp | ||
OWNERS | ||
PruneList.cpp | ||
PruneList.h | ||
README.auditd | ||
README.compression.md | ||
README.property | ||
README.replay.md | ||
RecordedLogMessage.h | ||
RecordingLogBuffer.cpp | ||
RecordingLogBuffer.h | ||
ReplayMessages.cpp | ||
rwlock.h | ||
SerializedData.h | ||
SerializedFlushToState.cpp | ||
SerializedFlushToState.h | ||
SerializedFlushToStateTest.cpp | ||
SerializedLogBuffer.cpp | ||
SerializedLogBuffer.h | ||
SerializedLogChunk.cpp | ||
SerializedLogChunk.h | ||
SerializedLogChunkTest.cpp | ||
SerializedLogEntry.h | ||
SimpleLogBuffer.cpp | ||
SimpleLogBuffer.h |
logd can record and replay log messages for offline analysis.
Recording Messages
logd has a RecordingLogBuffer
buffer that records messages to /data/misc/logd/recorded-messages.
It stores messages in memory until that file is accessible, in order to capture all messages since
the beginning of boot. It is only meant for logging developers to use and must be manually enabled
in by adding RecordingLogBuffer.cpp
to Android.bp
and setting
log_buffer = new SimpleLogBuffer(&reader_list, &log_tags, &log_statistics);
in main.cpp
.
Recording messages may delay the Log() function from completing and it is highly recommended to make
the logd socket in liblog
blocking, by removing SOCK_NONBLOCK
from the socket()
call in
liblog/logd_writer.cpp
.
Replaying Messages
Recorded messages can be replayed offline with the replay_messages
tool. It runs on host and
device and supports the following options:
interesting
- this prints 'interesting' statistics for each of the log buffer types (simple, chatty, serialized). The statistics are:- Log Entry Count
- Size (the uncompressed size of the log messages in bytes)
- Overhead (the total cost of the log messages in memory in bytes)
- Range (the range of time that the logs cover in seconds)
memory_usage BUFFER_TYPE
- this prints the memory usage (sum of private dirty pages of thereplay_messages
process). Note that the input file is mmap()'ed as RO/Shared so it does not appear in these dirty pages, and a baseline is taken before allocating the log buffers, so only their contributions are measured. The tool outputs the memory usage every 100,000 messages.latency BUFFER_TYPE
- this prints statistics of the latency of the Log() function for the given buffer type. It specifically prints the 1st, 2nd, and 3rd quartiles; the 95th, 99th, and 99.99th percentiles; and the maximum latency.print_logs BUFFER_TYPE [buffers] [print_point]
- this prints the logs as processed by the given buffer_type from the buffers specified bybuffers
starting after the number of logs specified byprint_point
have been logged. This acts as if a user calledlogcat
immediately after the specified logs have been logged, which is particularly useful since it will show the chatty pruning messages at that point. It additionally prints the statistics fromlogcat -S
after the logs.buffers
is a comma separated list of the numeric buffer id values from<android/log.h>
. For example,0,1,3
represents the main, radio, and system buffers. It can can also beall
.print_point
is an positive integer. If it is unspecified, logs are printed after the entire input file is consumed.nothing BUFFER_TYPE
- this does nothing other than read the input file and call Log() for the given buffer type. This is used for profiling CPU usage of strictly the log buffer.