https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=223059
--- Comment #1 from Alan Cox <[email protected]> --- Testing "(sp->sw_nblks - sp->sw_used >= npages)" before every call to blist_alloc() is of no real benefit. In the common case, where there is available swap space, it is added overhead, albeit a tiny amount of overhead. Only in the rare case, where there is no swap space available will it be of any benefit. But, in that case, the gain is minimal. The hinting mechanism in the blist code will quickly conclude that there is no available space, and return failure. In summary, the tiny amount of overhead in the common case is going to outweigh the gain in the rare case. -- You are receiving this mail because: You are the assignee for the bug. _______________________________________________ [email protected] mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-bugs To unsubscribe, send any mail to "[email protected]"
