add proper fuxed-based locking for stdio

previously, stdio used spinlocks, which would be unacceptable if we
ever add support for thread priorities, and which yielded
pathologically bad performance if an application attempted to use
flockfile on a key file as a major/primary locking mechanism.

i had held off on making this change for fear that it would hurt
performance in the non-threaded case, but actually support for
recursive locking had already inflicted that cost. by having the
internal locking functions store a flag indicating whether they need
to perform unlocking, rather than using the actual recursive lock
counter, i was able to combine the conditionals at unlock time,
eliminating any additional cost, and also avoid a nasty corner case
where a huge number of calls to ftrylockfile could cause deadlock
later at the point of internal locking.

this commit also fixes some issues with usage of pthread_self
conflicting with __attribute__((const)) which resulted in crashes with
some compiler versions/optimizations, mainly in flockfile prior to
pthread_create.

3 files changed, 8 insertions, 6 deletions

diff --git a/src/internal/libc.h b/src/internal/libc.h
index 906de2ca..929ff97a 100644
--- a/src/internal/libc.h
+++ b/src/internal/libc.h
@@ -38,7 +38,8 @@ extern struct __libc *__libc_loc(void) __attribute__((const));
 
 /* Designed to avoid any overhead in non-threaded processes */
 void __lock(volatile int *);
-void __lockfile(FILE *);
+int __lockfile(FILE *);
+void __unlockfile(FILE *);
 #define LOCK(x) (libc.threads_minus_1 ? (__lock(x),1) : ((void)(x),1))
 #define UNLOCK(x) (*(volatile int *)(x)=0)
 
diff --git a/src/internal/pthread_impl.h b/src/internal/pthread_impl.h
index e6089f02..03af4c12 100644
--- a/src/internal/pthread_impl.h
+++ b/src/internal/pthread_impl.h
@@ -87,6 +87,8 @@ struct __timer {
 #define SIGTIMER_SET ((sigset_t *)(unsigned long [1+(sizeof(long)==4)]){ \
 	 0x80000000 })
 
+pthread_t __pthread_self_init(void);
+
 int __set_thread_area(void *);
 int __libc_sigaction(int, const struct sigaction *, struct sigaction *);
 int __libc_sigprocmask(int, const sigset_t *, sigset_t *);
diff --git a/src/internal/stdio_impl.h b/src/internal/stdio_impl.h
index 76b58be9..c5f45eb1 100644
--- a/src/internal/stdio_impl.h
+++ b/src/internal/stdio_impl.h
@@ -24,8 +24,8 @@
 
 #define UNGET 8
 
-#define FLOCK(f) ((libc.threads_minus_1 && (f)->lock>=0) ? (__lockfile((f)),0) : 0)
-#define FUNLOCK(f) ((f)->lockcount && (--(f)->lockcount || ((f)->lock=0)))
+#define FLOCK(f) int __need_unlock = ((f)->lock>=0 ? __lockfile((f)) : 0)
+#define FUNLOCK(f) if (__need_unlock) __unlockfile((f)); else
 
 #define F_PERM 1
 #define F_NORD 4
@@ -49,12 +49,12 @@ struct __FILE_s {
 	FILE *prev, *next;
 	int fd;
 	int pipe_pid;
-	long dummy2;
+	long lockcount;
 	short dummy3;
 	signed char mode;
 	signed char lbf;
 	int lock;
-	int lockcount;
+	int waiters;
 	void *cookie;
 	off_t off;
 	int (*flush)(FILE *);
@@ -86,7 +86,6 @@ FILE *__fdopen(int, const char *);
 
 #define OFLLOCK() LOCK(&libc.ofl_lock)
 #define OFLUNLOCK() UNLOCK(&libc.ofl_lock)
-#define ofl_head (libc.ofl_head)
 
 #define feof(f) ((f)->flags & F_EOF)
 #define ferror(f) ((f)->flags & F_ERR)