[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: choice of file system build-dir can change how coreutils works
From: |
Paul Eggert |
Subject: |
Re: choice of file system build-dir can change how coreutils works |
Date: |
Wed, 21 Feb 2007 13:59:36 -0800 |
User-agent: |
Gnus/5.1008 (Gnus v5.10.8) Emacs/21.4 (gnu/linux) |
Jim Meyering <address@hidden> writes:
> However, note that for O_NOFOLLOW, it'd be a whole different kettle
> of fish, (since it is used in security-related areas) if configure
> detected support for it on the build file system, but an application
> used it on a different file system, without support.
This should be a rarer problem, I think, since O_NOFOLLOW is typically
implemented by the operating system, not by the file system, so it
typically either works on all file systems, or works on no file
systems.
> In this case, perhaps it's ok to let dd support the notaime flag
> as long as the system defines O_NOATIME.
I think so too. Here's a patch.
2007-02-21 Paul Eggert <address@hidden>
* doc/coreutils.texi (dd invocation): Warn that noatime might not be
reliable.
* src/dd.c (flags, usage): Look at O_NOATIME, not
HAVE_WORKING_O_NOATIME, to decide whether to support the noatime
flag, so that dd attempts O_NOATIME even if the build file system
does not support it. Problem reported by Jim Meyering today in
bug-coreutils.
* tests/dd/misc: Generate a warning, not a failure, if noatime
exists but fails.
diff --git a/doc/coreutils.texi b/doc/coreutils.texi
index 1a2dba4..04c1b4e 100644
--- a/doc/coreutils.texi
+++ b/doc/coreutils.texi
@@ -7442,6 +7442,8 @@ Use non-blocking I/O.
@opindex noatime
@cindex access time
Do not update the file's access time.
+Some older file systems silently ignore this flag, so it is a good
+idea to test it on your files before relying on it.
@item noctty
@opindex noctty
diff --git a/src/dd.c b/src/dd.c
index d699955..27a4a08 100644
--- a/src/dd.c
+++ b/src/dd.c
@@ -265,7 +265,7 @@ static struct symbol_value const flags[] =
{"direct", O_DIRECT},
{"directory", O_DIRECTORY},
{"dsync", O_DSYNC},
- {"noatime", HAVE_WORKING_O_NOATIME ? O_NOATIME : 0},
+ {"noatime", O_NOATIME},
{"noctty", O_NOCTTY},
{"nofollow", HAVE_WORKING_O_NOFOLLOW ? O_NOFOLLOW : 0},
{"nolinks", O_NOLINKS},
@@ -472,7 +472,7 @@ Each FLAG symbol may be:\n\
fputs (_(" sync likewise, but also for metadata\n"), stdout);
if (O_NONBLOCK)
fputs (_(" nonblock use non-blocking I/O\n"), stdout);
- if (HAVE_WORKING_O_NOATIME)
+ if (O_NOATIME)
fputs (_(" noatime do not update access time\n"), stdout);
if (O_NOCTTY)
fputs (_(" noctty do not assign controlling terminal from file\n"),
diff --git a/tests/dd/misc b/tests/dd/misc
index 513221b..fa5cfbe 100755
--- a/tests/dd/misc
+++ b/tests/dd/misc
@@ -29,6 +29,8 @@ if test "$VERBOSE" = yes; then
dd --version
fi
+fail=0
+warn=0
test_failure=0
echo data > $tmp_in || test_failure=1
ln $tmp_in $tmp_in2 || test_failure=1
@@ -63,8 +65,15 @@ sleep 1
if dd iflag=noatime if=$tmp_in of=$tmp_out 2> /dev/null; then
new_ls=`ls -u --full-time $tmp_in`
if test "x$old_ls" != "x$new_ls"; then
- echo "dd iflag=noatime updated atime; O_NOATIME bug in your kernel?" >&2
- fail=1
+ cat >&2 <<EOF
+=================================================================
+$0: WARNING!!!
+This operating system has the O_NOATIME file status flag,
+but it is silently ignored in some cases.
+Therefore, dd options like iflag=noatime may be silently ignored.
+=================================================================
+EOF
+ warn=77
fi
fi
@@ -77,6 +86,7 @@ fi
outbytes=`echo x | dd bs=3 ibs=10 obs=10 conv=sync 2>/dev/null | wc -c`
test "$outbytes" -eq 3 || fail=1
-rm -f $tmp_in $tmp_in2 $tmp_sym $tmp_out
+rm -f $tmp_in $tmp_in2 $tmp_sym $tmp_out || fail=1
+test $fail -eq 0 && fail=$warn
exit $fail
M ChangeLog
M doc/ChangeLog
M doc/coreutils.texi
M src/dd.c
M tests/dd/misc
Committed as e67725a333183beb9531d02645bfb942fb170a7d