On Thursday 20 November 2008 15:11, OGAWA Hirofumi wrote: > Hi, > > Those are to get closer to kernel. And probably, we wanted inode.c in > userland, so this patches modify some part of inode.c to prepare. And > some random cleanup is also there. > > static-http://userweb.kernel.org/~hirofumi/tux3/ > > Please review and pull it. > > With above, we can also copy user/inode.c to fs/tux3/ with one line change. > > --- tux3fs/fs/tux3/inode.c.orig 2008-11-21 08:01:31.000000000 +0900 > +++ tux3fs/fs/tux3/inode.c 2008-11-21 08:08:22.000000000 +0900 > @@ -194,7 +194,7 @@ int save_inode(struct inode *inode) > int tuxio(struct file *file, char *data, unsigned len, int write) > { > int err = 0; > - struct inode *inode = file->f_inode; > + struct inode *inode = file->f_dentry->d_inode; > loff_t pos = file->f_pos; > trace("%s %u bytes at %Lu, isize = 0x%Lx", write ? "write" : "read", > len, (L)pos, (L)inode->i_size); > if (write && pos + len > MAX_FILESIZE) > > This functions may rewrite in kernel completely, I'm not sure. Well, > this difference wouldn't matter at least for now.
Tuxio can go inside #ifndef __KERNEL__. Tuxio is the userspace version of generic file read/write in kernel, but we are going to use the generic_* functions for now, and maybe forever as they evolve and develop better interfaces. Is there are reason to change tux_path[] to *tux_path? Daniel _______________________________________________ Tux3 mailing list Tux3@tux3.org http://tux3.org/cgi-bin/mailman/listinfo/tux3
