logd: iterator corruption paranoia
Add checking for impossible(tm) scenarios within LogBuffer::flushTo:
1) When iterating through the log entries, check if the iterator
returns two identical element references and break out of the loop.
2) Cap the maximum number of log entries we will skip while holding
the iterator lock at 4194304, break out of the loop.
We print a message to the kernel logs if we hit these cases.
ToDo: Remove this paranoia at some future date.
Test: gTest liblog-unit-tests logcat-unit-tests and logd-unit-tests
Bug: 37378309
Change-Id: I789594649db14093238828b9f6d1daeca8b780c2
diff --git a/logd/LogBuffer.cpp b/logd/LogBuffer.cpp
index 4a790c9..a6bea0c 100644
--- a/logd/LogBuffer.cpp
+++ b/logd/LogBuffer.cpp
@@ -1118,9 +1118,22 @@
log_time max = start;
+ LogBufferElement* lastElement = nullptr; // iterator corruption paranoia
+ static const size_t maxSkip = 4194304; // maximum entries to skip
+ size_t skip = maxSkip;
for (; it != mLogElements.end(); ++it) {
LogBufferElement* element = *it;
+ if (!--skip) {
+ android::prdebug("reader.per: too many elements skipped");
+ break;
+ }
+ if (element == lastElement) {
+ android::prdebug("reader.per: identical elements");
+ break;
+ }
+ lastElement = element;
+
if (!privileged && (element->getUid() != uid)) {
continue;
}
@@ -1165,6 +1178,7 @@
return max;
}
+ skip = maxSkip;
pthread_mutex_lock(&mLogElementsLock);
}
pthread_mutex_unlock(&mLogElementsLock);