On Wed, 2012-11-14 at 10:43 +0100, Karel Zak wrote:
> On Wed, Nov 14, 2012 at 12:04:52AM -0800, Davidlohr Bueso wrote:
> > There's no harm when writing the MBR, changed or not. This also
>
Sorry about the long delay, life has been crazy busy - trying to squeeze
in some util-linux time :-)
> it's harm -- kernel has to parse the new PT, udev has to scan all the
> stuff etc. ... many event and every event is painful if you have
> complicated setup (DM, multipath, udisks, ...).
good point.
>
> > allows us to get rid of the 'MBRbuffer_changed' global variable.
>
> It would be nice to add to fdisk_context
>
> void *labeldata;
>
> and define in fdiskdos.h
>
> struct fdisk_dosdata {
> int write_status; /* FDISKDOS_WRITE_* flags */
> ...
> }
>
> enum {
> FDISKDOS_WRITE_HDR = (1 << 1) /* header buffer modified (e.g. ID change) */
> FDISKDOS_WRITE_PT = (1 << 2) /* partition table modified */
> FDISKDOS_WRITE_EPT = (1 << 3) /* any extended partition table modified */
> };
>
> (maybe you found better names for the macros and variables :-)
>
>
> than we can add fdisk_is_modified() and is_modified() to fdisk_label
> driver and remove all label specific junk from fdisk.c.
Yeah, a generic pointer to label-specific data is something that we have
to do. I will take a look at this option for a v2.
Thanks,
Davidlohr
--
To unsubscribe from this list: send the line "unsubscribe util-linux" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
