From af756d78ac042a2eed2417c5250d4b675d43bf93 Mon Sep 17 00:00:00 2001
From: Adrian Kummerlaender
Date: Wed, 17 Feb 2016 15:02:48 +0100
Subject: Implement static allocator for initialization

The previous interposition logic based on plain usage of `dlsym` analogously to various online examples led to a deadlock during _neovim_ startup. This deadlock was caused by _neovim_'s custom memory allocation library _jemalloc_ because it calls `mmap` during its initialization phase. The problem with calling `mmap` during initialization is that this already leads to executing `libChangeLog`'s `mmap` version whoes static `actual_mmap` function pointer is not initialized at this point in time. This is detected and leads to a call to `dlsym` to remedy this situation. Sadly `dlsym` in turn requires memory allocation using `calloc` which leads us back to initializing _jemalloc_ and as such to a deadlock.

I first saw this as a bug in _jemalloc_ which seemed to be confirmed by a short search in my search engine of choice. This prompted me to create an appropriate [bug report](https://github.com/jemalloc/jemalloc/issues/329) which was dismissed as a problem in the way `mmap` was interposed and not as a bug in the library. Thus it seems to be accepted practice that it is not the responsibility of a custom memory allocator to cater to the initialization needs of other libraries relying on function interposition. This is of course a valid position as the whole issue is a kind of _chicken and egg_ problem where both sides can be argued.

To cut to the chase I was left with the only option of working around this deadlock by adapting `libChangeLog` to call `dlsym` without relying on the wrapped application's memory allocator of choice. The most straight forward way to do this is to provide another custom memory allocator alongside the _payload_ function interpositions of `mmap` and friends.

`init/alloc.cc` implements such a selectively transparent memory allocator that offers a small static buffer for usage in the context of executing `dlsym`.The choice between forwarding memory allocation requests to the wrapped application's allocator and using the static buffer is governed by `init::dlsymContext`. This tiny helper class maintains an `dlsym_level` counter by posing as a scope guard.

The end result of this extension to `libChangeLog` is that it now also works with applications using _jemalloc_ such as _neovim_ and should overall be much more robust during its initialization phase.
---
 src/init/alloc.cc | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++
 src/init/alloc.h  | 22 ++++++++++++++++++++++
 2 files changed, 72 insertions(+)
 create mode 100644 src/init/alloc.cc
 create mode 100644 src/init/alloc.h

(limited to 'src/init')

diff --git a/src/init/alloc.cc b/src/init/alloc.cc
new file mode 100644
index 0000000..d18917b
--- /dev/null
+++ b/src/init/alloc.cc
@@ -0,0 +1,50 @@
+#include "alloc.h"
+
+#include <cstdio>
+#include <cstdint>
+
+namespace init {
+
+std::uint8_t dlsym_level   = 0;
+size_t       buffer_offset = 0;
+char         buffer[4096];
+
+void* static_malloc(size_t size) {
+	if ( buffer_offset + size >= sizeof(buffer) ) {
+		std::fprintf(stderr, "static allocator out of memory: %lu\n", buffer_offset);
+		exit(1);
+	}
+
+	buffer_offset += size;
+
+	return buffer + buffer_offset;
+}
+
+void* static_calloc(size_t block, size_t size) {
+	void* const mem = static_malloc(block * size);
+
+	for ( size_t i = 0; i < (block * size); ++i ) {
+		*(static_cast<char* const>(mem) + i) = '\0';
+	}
+
+	return mem;
+}
+
+bool from_static_buffer(void* ptr) {
+	return static_cast<char*>(ptr) >= buffer
+	    && static_cast<char*>(ptr) <= buffer + sizeof(buffer);
+}
+
+dlsymContext::dlsymContext() {
+	++dlsym_level;
+}
+
+dlsymContext::~dlsymContext() {
+	--dlsym_level;
+}
+
+bool dlsymContext::is_active() {
+	return dlsym_level > 0;
+}
+
+}
diff --git a/src/init/alloc.h b/src/init/alloc.h
new file mode 100644
index 0000000..cd1228d
--- /dev/null
+++ b/src/init/alloc.h
@@ -0,0 +1,22 @@
+#ifndef CHANGE_SRC_INIT_ALLOC_H_
+#define CHANGE_SRC_INIT_ALLOC_H_
+
+#include <cstdlib>
+
+namespace init {
+
+void* static_malloc(size_t size);
+void* static_calloc(size_t block, size_t size);
+
+bool from_static_buffer(void* ptr);
+
+struct dlsymContext {
+	dlsymContext();
+	~dlsymContext();
+
+	static bool is_active();
+};
+
+}
+
+#endif  // CHANGE_SRC_INIT_ALLOC_H_
-- 
cgit v1.2.3