On Sun, Mar 30, 2008 at 1:44 PM, ali <[EMAIL PROTECTED]> wrote: > > > Just a guess...but it may be due to byte ordering issues; you did this > > on PPC, which is big-endian, and ext2 was originally developed on > > Linux/x86, which is little-endian. > That could be it, but ext3 works in PPC linux distros.(I have not > tested ext2fuse on PPC linux distros though)
ext2 works on PPC as well, but you're not using ext2. You're using ext2fuse. > > If you care about the filesystem, I wouldn't touch let what you've > > compiled write to it. fsck cannot perform miracles, and if your hosed > > driver corrupts your filesystem, you may never be able to recover it. > You are right, but I did not know how to use ext2fuse as readonly mount -o ro > and I don't have any important data on my linux partition.(I'm just > testing different linux distros on my PPC mac to see how much it has > progressed, I use Leopard for serious works.) > > > > A more correct thing to do would be to remove the declaration from imager.c > Thats what I have tried the first time I got the error, but it caused > several other errors The problem appears to be that configure is looking for the definition of ssize_t in sys/types.h and darwin has it elsewhere. The correct thing to do is fix configure so that it finds it in the correct place, but barring that, I tweaked config.h to define HAVE_TYPE_SSIZE_T. > and after about one day of trial and error I > decided to temporarily modify string.h . Then the next error was: > ext2fs.c:52: error: 'struct stat' has no member named 'st_atim' > ext2fs.c:53: error: 'struct stat' has no member named 'st_mtim' > ext2fs.c:54: error: 'struct stat' has no member named 'st_ctim' man 2 stat The structure members have a different name on darwin (e.g., st_atimespec). > After googling a bit I preferred to comment those lines of ext2fs.c . > The las error was: > /usr/local/include/fuse/fuse_lowlevel_compat.h:77:24: error: sys/ > statfs.h: No such file or directory > I have commented sys/statfs.h from fuse_lowlevel_compat.h too. Then it > compiled successfully. Mmmm...that looks like an issue with FUSE. Amit, where does that file come from? Is that something you get straight from the linux fuse dist, or is it something you create? Regardless, on leopard, at least, that needs to be #include <sys/param.h> #include <sys/mount.h> rather than #include <sys/statfs.h> FWIW, I got a successful compile. Haven't tested it yet, though. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "macfuse-devel" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/macfuse-devel?hl=en -~----------~----~----~----~------~----~------~--~---
