It is preferable to always cc'ing the fsdevel list. The latest version of diskdev_cmds is on opensource.apple.com; there is a good porting on git://cavan.codon.org.uk/hfsplus-tools.git The problem with the unused inode message - at least for case-sensitive journalled hfs+, is found, and there is a patch sitting in a hard disk. Vyacheslav Dubeyko reviewed it and pointed out a problem - basically the patch isn't safe on SMP systems under load. So it needed to be reworked/retested. mount -o force does not do journalling, full stop. OTOH, I also see some problems with netgear's changes, and it does not do journalling properly, I think, under some usage. in a nutshell, it is a work-in-progress. I hope to put a tar ball out anyway soon. In any case, please have fsdevel in the cc for any correspondence on this matter. --- On Sun, 27/1/13, Kyle Laracey <kalaracey@xxxxxxxxx> wrote: > From: Kyle Laracey <kalaracey@xxxxxxxxx> > Subject: linux hfsplus support question > To: htl10@xxxxxxxxxxxxxxxxxxxxx > Date: Sunday, 27 January, 2013, 1:43 > Hi Hin-Tak, > > My name is Kyle Laracey, and I've been poking around the > hfsplus kernel driver, as well as the fsdevel mailing list, > in hope that I might be able to do something about the > current situation regarding reading journalled hfsplus > partitions from linux (I dual boot on a Mac). I was > wondering if you could answer a couple questions? > > > 1) I applied the Netgear patches you mentioned (located at > your Sourceforge page) to my 3.2(.35) kernel and began doing > some testing. As you mentioned in this thread (http://marc.info/?l=linux-fsdevel&m=134290542528362&w=2), > I deleted some files from Linux that were placed there in OS > X on a journalled hfsplus partition created from OS X. I > then ran fsck.hfsplus on linux with the most current > fsck.hfsplus package I could find (the Ubuntu "hfsprogs" > package) but that seems to be ported from an older version > of diskdev_cmds; from the output of your fsck.hfsplus: > > `Executing fsck_hfs (version > 540.1-Linux).` > > it looked like you were using a much more up-to-date > version. Would that be Naohiro Aota's attempt at a port (his > github page for his diskdev_cmds port: https://github.com/naota/diskdev_cmds)? I tried > compiling but had a bit of difficulty. > > > 2) I received similar output from the errors that you were > getting, like `Unused node is not erased (node = 18)`. I was > wondering, has there been any progress on this front, i.e., > is the Netgear-patched 3.2/3.4 hfsplus driver the most > up-to-date version of the driver? Or would you say the > vanilla hfsplus driver, when used with `mount -f` to force > read-write, not read-only, is more advanced? > > > Thank you, > Kyle Laracey -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
