[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: a couple of bugs

From: Benoit Sigoure
Subject: Re: a couple of bugs
Date: Mon, 17 Dec 2007 16:36:58 +0100

On Dec 17, 2007, at 2:03 PM, Eric Blake wrote:

According to Benoit Sigoure on 12/17/2007 1:04 AM:

./ ./config.status --file=-:input </dev/null >/dev/ full
/tmp/autoconf/tests/testsuite.dir/at-test-source: line 893: /dev/ full:
Operation not supported

This looks like a tpyo but I checked in the source and it's not, why not
use something like /nonexistent/whatever?

There's a different between a shell failure in being able to create the file (the tool under test isn't even run, because the shell errored out beforehand), and a write file when writing to an existing device that runs
out of space.  We really do want the latter, not the former.  So the
correct fix is to first check whether /dev/full exists and is writable,
and if not, skip the test, before proceeding on with the attempt to
redirect tool output into the full device.  Coreutils and m4 head both
have tests along those lines, which can be used as an example.

Wait a minute, /dev/full exists somewhere? I've never heard of such a device. Hmm I checked on a Linux box and there is indeed a /dev/full (character special (1/7)), but this seems very Linux specific.

Benoit Sigoure aka Tsuna
EPITA Research and Development Laboratory

reply via email to

[Prev in Thread] Current Thread [Next in Thread]