GNU bug report logs - #28432
dd: warn about partial read on final iread

Previous Next

To reply to this bug, email your comments to 28432 AT debbugs.gnu.org.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Barry Handelman <barry.handelman <at> gmail.com>
To: bug-coreutils <at> gnu.org
Subject: dd doesn't warn about partial read on final call to iread
Date: Tue, 12 Sep 2017 15:51:30 +0000

[Message part 1 (text/plain, inline)]

The iread function in dd.c issues a warning if the previous read was short,
suggesting that you use iflag=fullblock

However, the warning is issued when the previous call to iread was partial,
so if only the final read was short then no warning is issued.
Importantly, this will always be the case when there was only one read, and
it was partial.

I'd suggest that if the strategy of warning on subsequent reads is to be
used, the warning routine should be run once again after the final read.

[Message part 2 (text/html, inline)]

Message #8 received at 28432 <at> debbugs.gnu.org (full text, mbox):

From: Pádraig Brady <P <at> draigBrady.com>
To: Barry Handelman <barry.handelman <at> gmail.com>, 28432 <at> debbugs.gnu.org
Subject: Re: bug#28432: dd doesn't warn about partial read on final call to
 iread
Date: Wed, 13 Sep 2017 18:11:48 -0700

On 12/09/17 08:51, Barry Handelman wrote:
> The iread function in dd.c issues a warning if the previous read was short,
> suggesting that you use iflag=fullblock
> 
> However, the warning is issued when the previous call to iread was partial,
> so if only the final read was short then no warning is issued.
> Importantly, this will always be the case when there was only one read, and
> it was partial.
> 
> I'd suggest that if the strategy of warning on subsequent reads is to be
> used, the warning routine should be run once again after the final read.

The current approach was discussed at https://bugs.gnu.org/7362
A short read as the last read would be a common situation though?
You're specifically looking for the following to warn?

  (echo; sleep 1; echo) | dd count=1 bs=512

That may be a bit aggressive as that could be used to read
the current data available, up to a specified max.

cheers,
Pádraig.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.