[syslinux] [PATCH 001/001] Add UFS1/2 support to Extlinux installer.

[Raphael S Carvalho]

On Fri, Jul 19, 2013 at 12:48 PM, Matt Fleming <matt at console-pimps.org>wrote:

> On Fri, 12 Jul, at 08:31:27PM, Raphael S.Carvalho wrote:
> > @@ -575,7 +594,7 @@ bail:
> >      return 1;
> >  }
> >
> > -/* btrfs has to install the ldlinux.sys in the first 64K blank area,
> which
> > +/* btrfs/ufs2 has to install the ldlinux.sys in the first 64K blank
> area, which
> >     is not managered by btrfs tree, so actually this is not installed as
> files.
> >     since the cow feature of btrfs will move the ldlinux.sys every where
> */
> >  int btrfs_install_file(const char *path, int devfd, struct stat *rst)
>
> The ufs2 portion of this could do with more explanation, since the btrfs
> reasons don't apply to ufs2, right (ufs2 has no cow feature)?
>
> > @@ -955,9 +974,10 @@ static char * get_default_subvol(char * rootdir,
> char * subvol)
> >
> >  static int install_file(const char *path, int devfd, struct stat *rst)
> >  {
> > -    if (fs_type == EXT2 || fs_type == VFAT || fs_type == NTFS)
> > +    if (fs_type == EXT2 || fs_type == VFAT || fs_type == NTFS
> > +     || fs_type == UFS1)
> >       return ext2_fat_install_file(path, devfd, rst);
> > -    else if (fs_type == BTRFS)
> > +    else if (fs_type == BTRFS || fs_type == UFS2)
> >       return btrfs_install_file(path, devfd, rst);
> >      else if (fs_type == XFS)
> >       return xfs_install_file(path, devfd, rst);
> > @@ -983,7 +1003,7 @@ static int validate_device(const char *path, int
> devfd)
> >      struct statfs sfs;
> >      int pfd;
> >      int rv = -1;
> > -
> > +
> >      pfd = open(path, O_RDONLY|O_DIRECTORY);
> >      if (pfd < 0)
> >       goto err;
> > @@ -1063,6 +1083,17 @@ static const char *find_device(const char
> *mtab_file, dev_t dev)
> >               done = true;
> >               break;
> >           }
> > +
> > +         break;
> > +     case UFS1:
> > +     case UFS2:
> > +         if (!strcmp(mnt->mnt_type, "ufs") && !stat(mnt->mnt_fsname,
> &dst) &&
> > +             dst.st_rdev == dev) {
> > +             done = true;
> > +             break;
>
> Don't need this break statement.
>
> --
> Matt Fleming, Intel Open Source Technology Center
>

Thanks for the review.

I will work on these last issues, and resent the patch.

Regards,
Raphael S. Carvalho