[parted-devel] Bug: Removal of BLKPG causes regression of ability to manipulate disks with other partitions in use

[Jim Meyering]

Phillip Susi wrote:
> On 3/29/2010 12:44 PM, Jim Meyering wrote:
>> If you can use an existing FD, then please do.
>> That would be far better.
>
> Will do.
>
>> If you must use open, then you must handle failure,
>> and must not call ioctl or close on a negative "fd".
>
> I'll use the existing fd so this goes away, but out of curiosity, why
> not?  Won't it just fail with EBADF?

Right, but those syscall failures would also pollute the results of strace
and valgrind runs, not to mention those from any decent static-analyzer.
We already have far too many false positives on those fronts.

>> Other than this bit of quoted code, I haven't reviewed
>> your change at all -- was waiting for the newer version
>> you said would be coming soon.  Did I miss it?
>
> No, you didn't.  I'm going to make some of the changes that have already
> been suggested and test it tonight.  Should send it to you either
> tonight or tomorrow morning.