[PATCH 15/19 v3] mkfs: don't treat files as though they are block devices

[Jan Tulak]

On Fri, Apr 29, 2016 at 9:11 PM, Eric Sandeen <sandeen at sandeen.net> wrote:

>
>
> On 4/29/16 9:47 AM, Jan Tulak wrote:
> > From: Dave Chinner <dchinner at redhat.com>
> >
> > If the device is actually a file, and "-d file" is not specified,
> > mkfs will try to treat it as a block device and get stuff wrong.
> > Image files don't necessarily have the same sector sizes as the
> > block device or filesystem underlying the image file, nor should we
> > be issuing discard ioctls on image files.
> >
> > To fix this sanely, only require "-d file" if the device name is
> > invalid to trigger creation of the file. Otherwise, use stat() to
> > determine if the device is a file or block device and deal with that
> > appropriately by setting the "isfile" variables and turning off
> > direct IO. Then ensure that we check the "isfile" options before
> > doing things that are specific to block devices.
> >
> > Other file/blockdev issues fixed:
> >       - use getstr to detect specifying the data device name
> >         twice.
> >       - check file/size/name parameters before anything else.
> >       - overwrite checks need to be done before the image file is
> >         opened and potentially truncated.
> >       - blkid_get_topology() should not be called for image files,
> >         so warn when it is called that way.
> >       - zero_old_xfs_structures() emits a spurious error:
> >               "existing superblock read failed: Success"
> >         when it is run on a truncated image file. Don't warn if we
> >         see this problem on an image file.
> >       - Don't issue discards on image files.
> >       - Use fsync() for image files, not BLKFLSBUF in
> >         platform_flush_device() for Linux.
> >
> > Signed-off-by: Dave Chinner <dchinner at redhat.com>
> > Signed-off-by: Jan Tulak <jtulak at redhat.com>
> >
> > ---
> > CHANGES:
> > * read image file size in advance of O_TRUNC in case of dfile&&dcreat
> > Signed-off-by: Jan Tulak <jtulak at redhat.com>
> > ---
> >  libxfs/init.c   |  21 ++++++-
> >  libxfs/linux.c  |  11 +++-
> >  mkfs/xfs_mkfs.c | 182
> ++++++++++++++++++++++++++++++++++++++------------------
> >  3 files changed, 154 insertions(+), 60 deletions(-)
> >
> > diff --git a/libxfs/init.c b/libxfs/init.c
> > index 8d747e8..c7ae00d 100644
> > --- a/libxfs/init.c
> > +++ b/libxfs/init.c
> > @@ -253,8 +253,15 @@ libxfs_init(libxfs_init_t *a)
> >       rtname = a->rtname;
> >       a->dfd = a->logfd = a->rtfd = -1;
> >       a->ddev = a->logdev = a->rtdev = 0;
> > -     a->dbsize = a->lbsize = a->rtbsize = 0;
> > -     a->dsize = a->logBBsize = a->logBBstart = a->rtsize = 0;
> > +     a->lbsize = a->rtbsize = 0;
> > +     a->logBBsize = a->logBBstart = a->rtsize = 0;
> > +
> > +     // We can reset dbsize only when it is not a file, or we won't
> > +     // truncate it. Otherwise, we loose the size of the file forever.
>
> please don't use c++ comments in xfsprogs, we use /* comments */
>

Sorry. ​:-[ ​



>
> ...
>
> >  static void
> > +check_device_type(
> > +     const char      *name,
> > +     int             *isfile,
> > +     bool            no_size,
> > +     bool            no_name,
> > +     int             *create,
> > +     bool            force_overwrite,
> > +     const char      *optname)
> > +{
> > +     struct stat64 statbuf;
> > +     /*
> > +     if (*isfile && (no_size || no_name)) {
> > +             fprintf(stderr,
> > +     _("if -%s file then -%s name and -%s size are required\n"),
> > +                     optname, optname, optname);
> > +             usage();
> > +     }*/
>
> What is this?
>
> ​Again :-[
I commented it out during the development and then forgot to really delete
it.​



> Jan, I'm just going to go back to the original patches posted in your V2
> series,
> and either give them Reviewed-by's, or send followup fix-up patches with a
>
> Eric Sandeen <sandeen at redhat.com>: fixed up foo, bar, baz
>
> tag and a Reviewed-by to go with it, I think that might be the fastest
> path to
> finally getting this stuff merged.
>
>
​OK, whatever you think is the best. :-)​

Thanks,
Jan




> Thanks,
> -Eric
>
> _______________________________________________
> xfs mailing list
> xfs at oss.sgi.com
> http://oss.sgi.com/mailman/listinfo/xfs
>



-- 
Jan Tulak
jtulak at redhat.com / jan at tulak.me
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://oss.sgi.com/pipermail/xfs/attachments/20160503/e9c59dbb/attachment-0001.html>