PERFORCE change 114786 for review

Roman Divacky rdivacky at FreeBSD.org
Wed Feb 21 09:03:50 UTC 2007


http://perforce.freebsd.org/chv.cgi?CH=114786

Change 114786 by rdivacky at rdivacky_witten on 2007/02/21 09:02:58

	In the FUTEX_WAKE_OP the *timeout param is being abused for number
	of wakes up. Change the code accordingly.

Affected files ...

.. //depot/projects/linuxolator/src/sys/compat/linux/linux_futex.c#10 edit

Differences ...

==== //depot/projects/linuxolator/src/sys/compat/linux/linux_futex.c#10 (text+ko) ====

@@ -304,12 +304,10 @@
 		if (op_ret > 0) {
 			op_ret = 0;
 			/*
-			 * Linux uses the address of the timespec parameter
-			 * as the number of retries, so any large number will
-			 * be ok.
+			 * Linux abuses the address of the timespec parameter
+			 * as the number of retries.
 			 */
 			op_ret += futex_wake(f2, (int) (unsigned long) args->timeout, NULL);
-//			op_ret += futex_wake(f2, 0xffffffff, NULL);
 			ret += op_ret;
 		}
 		futex_put(f2);


More information about the p4-projects mailing list