[PATCH 5/5] file_ioctl.c: fix FIDEDUPERANGE output

[Jeff Mahoney]

On 5/27/16 4:36 AM, Dmitry V. Levin wrote:
> On Thu, May 26, 2016 at 11:33:07PM -0400, jeffm at suse.com wrote:
>>
>> * file_ioctl.c (file_ioctl): initialize rc to true in FIDEDUPERANGE.
>>   Otherwise we'll never show the output data in abbrev mode.
> 
> Are we going to show something more interesting than " => {info=...}"
> on exiting FIDEDUPERANGE in abbrev mode?

For other unbounded ranges, I don't print them unless we have -v and
then I bound them via max_strlen.  It just happens to be that in most
other cases, we have other values to display in addition to the
unbounded part, so it looks more interesting.

I'll ask Mark Fasheh, the author of duperemove, how many extents are
typically deduped in a single ioctl call.  If it's like 1 or 2, I can
make that the default display and then ... after that.

-Jeff

-- 
Jeff Mahoney
SUSE Labs

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 881 bytes
Desc: OpenPGP digital signature
URL: <http://lists.strace.io/pipermail/strace-devel/attachments/20160527/823c4c4c/attachment.bin>