OPTIM: tools: optimize my_ffsl() for x86_64

This call is now used quite a bit in the fd cache, to decide which cache
to add/remove the fd to/from, when waking up a task for a single thread
in __task_wakeup(), in fd_cant_recv() and in fd_process_cached_events(),
and we can replace it with a single instruction, removing ~30 instructions
and ~80 bytes from the inner loop of some of these functions.

In addition the test for zero value was replaced with a comment saying
that it is illegal and leads to an undefined behaviour. The code does
not make use of this useless case today.

diff --git a/include/common/standard.h b/include/common/standard.h
index 5c7d15229e529375986b41516fc3671fbb0b66d1..fabe97276bba2d8a8f25cdba6eb398f3a23832db 100644 (file)
--- a/include/common/standard.h
+++ b/include/common/standard.h
@@ -802,13 +802,16 @@ static inline unsigned int my_popcountl(unsigned long a)
 }
 
 /* Simple ffs implementation. It returns the position of the lowest bit set to
- * one. */
+ * one. It is illegal to call it with a==0 (undefined result).
+ */
 static inline unsigned int my_ffsl(unsigned long a)
 {
-       unsigned int cnt;
+       unsigned long cnt;
 
-       if (!a)
-               return 0;
+#if defined(__x86_64__)
+       __asm__("bsr %1,%0\n" : "=r" (cnt) : "rm" (a));
+       cnt++;
+#else
 
        cnt = 1;
 #if LONG_MAX > 0x7FFFFFFFL /* 64bits */
@@ -837,6 +840,7 @@ static inline unsigned int my_ffsl(unsigned long a)
                a >>= 1;
                cnt += 1;
        }
+#endif /* x86_64 */
 
        return cnt;
 }