Add support for LinuxThreads.

When using LinuxThreads pthread_setspecific triggers recursive
allocation on all threads. Work around this by creating a global linked
list of in-progress tsd initializations.

This modifies the _tsd_get_wrapper macro-generated function. When it has
to initialize an TSD object it will push the item to the linked list
first. If this causes a recursive allocation then the _get_wrapper
request is satisfied from the list. When pthread_setspecific returns the
item is removed from the list.

This effectively adds a very poor substitute for real TLS used only
during pthread_setspecific allocation recursion.

Signed-off-by: Crestez Dan Leonard <lcrestez@ixiacom.com>

include/jemalloc/internal/tsd.h

@@ -6,6 +6,12 @@
 
 typedef bool (*malloc_tsd_cleanup_t)(void);
 
+#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \
+    !defined(_WIN32))
+typedef struct tsd_init_block_s tsd_init_block_t;
+typedef struct tsd_init_head_s tsd_init_head_t;
+#endif
+
 /*
  * TLS/TSD-agnostic macro-based implementation of thread-specific data.  There
  * are four macros that support (at least) three use cases: file-private,
@@ -81,6 +87,7 @@ extern bool		a_name##_booted;
 #else
 #define	malloc_tsd_externs(a_name, a_type)				\
 extern pthread_key_t	a_name##_tsd;					\
+extern tsd_init_head_t	a_name##_tsd_init_head;				\
 extern bool		a_name##_booted;
 #endif
 
@@ -105,6 +112,10 @@ a_attr bool		a_name##_booted = false;
 #else
 #define	malloc_tsd_data(a_attr, a_name, a_type, a_initializer)		\
 a_attr pthread_key_t	a_name##_tsd;					\
+a_attr tsd_init_head_t	a_name##_tsd_init_head = {			\
+	ql_head_initializer(blocks),					\
+	MALLOC_MUTEX_INITIALIZER					\
+};									\
 a_attr bool		a_name##_booted = false;
 #endif
 
@@ -333,8 +344,14 @@ a_name##_tsd_get_wrapper(void)						\
 	    pthread_getspecific(a_name##_tsd);				\
 									\
 	if (wrapper == NULL) {						\
+		tsd_init_block_t block;					\
+		wrapper = tsd_init_check_recursion(			\
+		    &a_name##_tsd_init_head, &block);			\
+		if (wrapper)						\
+		    return wrapper;					\
 		wrapper = (a_name##_tsd_wrapper_t *)			\
 		    malloc_tsd_malloc(sizeof(a_name##_tsd_wrapper_t));	\
+		block.data = wrapper;					\
 		if (wrapper == NULL) {					\
 			malloc_write("<jemalloc>: Error allocating"	\
 			    " TSD for "#a_name"\n");			\
@@ -350,6 +367,7 @@ a_name##_tsd_get_wrapper(void)						\
 			    " TSD for "#a_name"\n");			\
 			abort();					\
 		}							\
+		tsd_init_finish(&a_name##_tsd_init_head, &block);	\
 	}								\
 	return (wrapper);						\
 }									\
@@ -379,6 +397,19 @@ a_name##_tsd_set(a_type *val)						\
 /******************************************************************************/
 #ifdef JEMALLOC_H_STRUCTS
 
+#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \
+    !defined(_WIN32))
+struct tsd_init_block_s {
+	ql_elm(tsd_init_block_t)	link;
+	pthread_t			thread;
+	void				*data;
+};
+struct tsd_init_head_s {
+	ql_head(tsd_init_block_t)	blocks;
+	malloc_mutex_t			lock;
+};
+#endif
+
 #endif /* JEMALLOC_H_STRUCTS */
 /******************************************************************************/
 #ifdef JEMALLOC_H_EXTERNS
@@ -388,6 +419,12 @@ void	malloc_tsd_dalloc(void *wrapper);
 void	malloc_tsd_no_cleanup(void *);
 void	malloc_tsd_cleanup_register(bool (*f)(void));
 void	malloc_tsd_boot(void);
+#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \
+    !defined(_WIN32))
+void	*tsd_init_check_recursion(tsd_init_head_t *head,
+    tsd_init_block_t *block);
+void	tsd_init_finish(tsd_init_head_t *head, tsd_init_block_t *block);
+#endif
 
 #endif /* JEMALLOC_H_EXTERNS */
 /******************************************************************************/

src/tsd.c

@@ -105,3 +105,37 @@ JEMALLOC_SECTION(".CRT$XLY") JEMALLOC_ATTR(used)
 static const BOOL	(WINAPI *tls_callback)(HINSTANCE hinstDLL,
     DWORD fdwReason, LPVOID lpvReserved) = _tls_callback;
 #endif
+
+#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \
+    !defined(_WIN32))
+void *
+tsd_init_check_recursion(tsd_init_head_t *head, tsd_init_block_t *block)
+{
+	pthread_t self = pthread_self();
+	tsd_init_block_t *iter;
+
+	/* Check whether this thread has already inserted into the list. */
+	malloc_mutex_lock(&head->lock);
+	ql_foreach(iter, &head->blocks, link) {
+		if (iter->thread == self) {
+			malloc_mutex_unlock(&head->lock);
+			return (iter->data);
+		}
+	}
+	/* Insert block into list. */
+	ql_elm_new(block, link);
+	block->thread = self;
+	ql_tail_insert(&head->blocks, block, link);
+	malloc_mutex_unlock(&head->lock);
+	return (NULL);
+}
+
+void
+tsd_init_finish(tsd_init_head_t *head, tsd_init_block_t *block)
+{
+
+	malloc_mutex_lock(&head->lock);
+	ql_remove(&head->blocks, block, link);
+	malloc_mutex_unlock(&head->lock);
+}
+#endif