[00:03:57] <Simon Wilkinson> sbrk() will have stopped providing useful information years ago - most mallocs use mmap to obtain their heap now.
[04:31:54] --- Simon Wilkinson has left
[04:43:26] --- Simon Wilkinson has become available
[04:54:30] --- Simon Wilkinson has left
[06:47:17] --- mvita has become available
[06:47:58] --- meffie has become available
[07:40:25] --- deason has become available
[08:33:12] <kaduk@jabber.openafs.org/barnowl> Why ... is src/util/sys.c still in the tree?
[08:36:57] <jaltman/FrogsLeap> clearly because you have not removed it yet
[08:37:21] <kaduk@jabber.openafs.org/barnowl> :)
[08:37:54] <jaltman/FrogsLeap> a better question is why was it ever in the tree?
[08:39:54] <kaduk@jabber.openafs.org/barnowl> Also true.
[08:40:38] <kaduk@jabber.openafs.org/barnowl> A file of that name seems like it might be a place to put a function
that gets the "size" of the current process, is most of why I noticed.
[08:44:42] <meffie> looking in src/util always feels like rummaging in an old basement...
[08:45:15] <jaltman/FrogsLeap> apparently it is a command that returns the SYS_NAME of the current afs install.   A useful command.
[08:45:47] <jaltman/FrogsLeap> and its installed
[08:46:03] <jaltman/FrogsLeap> not sure who uses it
[08:47:37] <meffie> my guess was it was the predecessor to src/venus/livesys
[08:48:39] <jaltman/FrogsLeap> they do different things.   sys returns the userspace version and livesys the kernel version
[08:49:10] <meffie> ah
[08:49:31] <jaltman/FrogsLeap> and since sysname can be changed "fs sysname" livesys may not even report the actual kernel sysname
[09:10:27] --- meffie has left
[09:59:24] --- meffie has become available
[10:37:40] --- Simon Wilkinson has become available
[10:43:52] <kaduk@jabber.openafs.org/barnowl> Not fatal, but still worrisome:
/Users/kaduk/openafs/openafs/src/afs/VNOPS/afs_vnop_lookup.c:864:38: warning: in
compatible pointer types passing 'struct vcache *' to parameter of type 'vnode_t
' (aka 'struct vnode *') [-Wincompatible-pointer-types]
                    afs_darwin_finalizevnode(tvcp, VTOAFS(adp), NULL, 0, 1);
                                                   ^~~~~~~~~~~
[10:51:58] --- Simon Wilkinson has left
[10:53:11] --- meffie has left
[10:53:11] --- meffie has become available
[11:47:16] <mvita> http://wiki.openafs.org/CodingStyle <http://wiki.openafs.org/CodingStyle:> states "All new code must compile cleanly when configured with --enable-warnings".   Is this still sufficient, or should we now be using —enable-checking instead?   
[11:48:44] <mvita> I've only recently started using —enable-checking
[12:00:45] --- Simon Wilkinson has become available
[12:18:46] <kaduk@jabber.openafs.org/barnowl> I think that snipped in lookup should just be using AFSTOV instead of
VTOAFS.
[13:37:13] --- meffie has left
[15:05:19] --- mvita has left
[15:16:07] --- deason has left
[15:16:08] --- deason has become available
[17:01:03] --- deason has left
[23:59:35] --- Simon Wilkinson has left
[23:59:46] --- Simon Wilkinson has become available