On Saturday 10 January 2009 02:45, Khem Raj wrote:
> Hi Denys
> 
> Shouldn't this check be like the one in the patch

@@ -41,7 +41,7 @@
         * Try to catch this problem at uclibc build time:  */
        struct BUG_sigset_size {
                int BUG_sigset_size
-                       [sizeof(act->sa_mask) != _NSIG / 8 ? 1 : -1];
+                       [sizeof(act->sa_mask) == (_NSIG - 1) / 8 ? 1 : -1];
        };
        return __syscall_rt_sigaction(sig, act, oact, sizeof(act->sa_mask));
 }

Yes, It should be ==. (_NSIG - 1) is correct but not necessary,
since it is always 2^N + 1 and "+1" will be lost in / 8.

Thanks. Fixed in svn.
--
vda
_______________________________________________
uClibc mailing list
[email protected]
http://lists.busybox.net/mailman/listinfo/uclibc

Reply via email to