Libguestfs - May 2015 - sfdisk: No more Cylinder / Head / Sector support in util-linux 2.26 and later

Hi,

it was brought to my attention that sfdisk has lost the ability to deal
with C/H/S addressing as of util-linux 2.26, so the -C ,-H, -S command
line parameters are no longer supported. Should there be a compile-time
check based on "sfdisk -h" output that would cause daemon/sfdisk.c to
be
compiled only if something like the following is present?

,----
| Override the detected geometry using:
|  -C, --cylinders <number>  set the number of cylinders to use
|  -H, --heads <number>      set the number of heads to use
|  -S, --sectors <number>    set the number of sectors to use
`----

Cheers,
-Hilko

On Sun, May 17, 2015 at 05:06:15PM +0200, Hilko Bengen
wrote:
> Hi,
> 
> it was brought to my attention that sfdisk has lost the ability to deal
> with C/H/S addressing as of util-linux 2.26, so the -C ,-H, -S command
> line parameters are no longer supported. Should there be a compile-time
> check based on "sfdisk -h" output that would cause
daemon/sfdisk.c to be
> compiled only if something like the following is present?
> 
> ,----
> | Override the detected geometry using:
> |  -C, --cylinders <number>  set the number of cylinders to use
> |  -H, --heads <number>      set the number of heads to use
> |  -S, --sectors <number>    set the number of sectors to use
> `----
Yeah, unfortunately sfdisk was rewritten upstream, causing several
problems:

https://bugzilla.redhat.com/show_bug.cgi?id=1183236
https://bugzilla.redhat.com/show_bug.cgi?id=1212507

Ideally I'd like to remove our dependency on sfdisk, but it's rather a
lot of work.

I think a compile-time check would just cause the API to fail in a
different way, so I don't think it would help users very much, unless
I'm misunderstanding what you mean.

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
Fedora Windows cross-compiler. Compile Windows programs, test, and
build Windows installers. Over 100 libraries supported.
http://fedoraproject.org/wiki/MinGW

* Richard W.M. Jones:
> I think a compile-time check would just cause the API to fail in a
> different way, so I don't think it would help users very much, unless
> I'm misunderstanding what you mean.
I think you are right.

Cheers,
-Hilko