When deleting a directory, the directory should not be removed if the
file in the subdirectory fails to be deleted.
Background information:
When I tested the kernel using LTP (linux-test-project).I found the
mv command have some issue. The LTP test case use the mv command to
move the directory t1 in the cgroup file system to the /tmp directory.
becase files in the cgroup file system are not allowed to be removed.
so the mv reported "Permission denied", but I used the ls command to
view the results and found that the directory t1 had been removed
from the cgroup file system. For the same test case, I used the mv
tool in the GNU coreutils, and the directory t1 will not be removed.
the following testcase use busybox mv:
/ # mount -t cgroup -o cpu cgroup /cpu
/ # cd /cpu
/cpu # mkdir -p t1
/cpu # ls
cgroup.clone_children cpu.cfs_period_us cpu.stat t1
cgroup.procs cpu.cfs_quota_us notify_on_release tasks
cgroup.sane_behavior cpu.shares release_agent
/cpu # mv t1 /tmp
mv: can't remove 't1/cgroup.procs': Operation not permitted
mv: can't remove 't1/cpu.cfs_period_us': Operation not permitted
mv: can't remove 't1/cpu.stat': Operation not permitted
mv: can't remove 't1/cpu.shares': Operation not permitted
mv: can't remove 't1/cpu.cfs_quota_us': Operation not permitted
mv: can't remove 't1/tasks': Operation not permitted
mv: can't remove 't1/notify_on_release': Operation not permitted
mv: can't remove 't1/cgroup.clone_children': Operation not permitted
/cpu # ls
cgroup.clone_children cpu.cfs_period_us cpu.stat cgroup.procs
cpu.cfs_quota_us notify_on_release tasks cgroup.sane_behavior
cpu.shares release_agent
/cpu #
This patch fixed it, don't call rmdir if remove_file return failure,
and under certain file systems, the mv could work normally.
Signed-off-by: Chen Yu <yu.chen3@zte.com.cn>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
If fgets() returns incomplete string, we replace NUL with
'\n', and then trim() runs on a non-NUL-terminated buffer.
Prevent that.
While at it, bump buffer from 1k to 2k.
function old new delta
query 519 524 +5
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
- recurse is not acting as expected (looks like broken old code)
- when not using verbose output, return value for retained files is 1,
which is not expected.
Signed-off-by: sagivd <sagivdev@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
function old new delta
archivefile 167 176 +9
Signed-off-by: Andre Goddard Rosa <andre.rosa@lge.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Some syslog producers provide inconsistent timestamps, so provide an option
to ignore the message timestamps and always locally timestamp. In order to
implement this, invert the valid-timestamp check, but only use the timestamp
if this option is not enabled.
This is in line with what what other syslogd implementations do:
From sysklogd syslogd.c:
* Sun Nov 7 12:28:47 CET 2004: Martin Schulze <joey@infodrom.org>
* Discard any timestamp information found in received syslog
* messages. This will affect local messages sent from a
* different timezone.
rsyslog's imuxsock module similary has an (enabled by default)
IgnoreTimestamp option:
https://www.rsyslog.com/doc/v8-stable/configuration/modules/imuxsock.html
function old new delta
packed_usage 32877 32912 +35
timestamp_and_log 363 376 +13
syslogd_main 1638 1641 +3
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 3/0 up/down: 51/0) Total: 51 bytes
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Wed, 28 Mar 2018 18:37:51 +0800
expand: Do not quote backslashes in unquoted parameter expansion
Here is a better example:
a="/*/\nullx" b="/*/\null"; printf "%s\n" $a $b
dash currently prints
/*/\nullx
/*/\null
bash prints
/*/\nullx
/dev/null
You may argue the bash behaviour is inconsistent but it actually
makes sense. What happens is that quote removal only applies to
the original token as seen by the shell. It is never applied to
the result of parameter expansion.
Now you may ask why on earth does the second line say "/dev/null"
instead of "/dev/\null". Well that's because it is not the quote
removal step that removed the backslash, but the pathname expansion.
The fact that the /de\v does not become /dev even though it exists
is just the result of the optimisation to avoid unnecessarily
calling stat(2). I have checked POSIX and I don't see anything
that forbids this behaviour.
So going back to dash yes I think we should adopt the bash behaviour
for pathname expansion and keep the existing case semantics.
This patch does exactly that. Note that this patch does not work
unless you have already applied
https://patchwork.kernel.org/patch/10306507/
because otherwise the optimisation mentioned above does not get
detected correctly and we will end up doing quote removal twice.
This patch also updates expmeta to handle naked backslashes at
the end of the pattern which is now possible.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
function old new delta
expmeta 618 653 +35
memtodest 146 147 +1
Tested to work with both ASH_INTERNAL_GLOB on and off.
hush does not handle this correctly.
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Sat, 19 May 2018 02:39:37 +0800
exec: Return 126 on most errors in shellexec
Currently when shellexec fails on most errors the shell will exit
with exit status 2. This patch changes it to 126 in order to avoid
ambiguities with the exit status from a successful exec.
The errors that result in 127 has also been expanded to include
ENOTDIR, ENAMETOOLONG and ELOOP.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
function old new delta
shellexec 245 254 +9
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Tue, 27 Mar 2018 00:39:35 +0800
eval: Restore input files in evalcommand
When evalcommand invokes a command that modifies parsefile and
then bails out without popping the file, we need to ensure the
input file is restored so that the shell can continue to execute.
Reported-by: Martijn Dekker <martijn@inlv.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
function old new delta
unwindfiles - 20 +20
evalcommand 1635 1653 +18
getoptscmd 584 595 +11
popallfiles 20 10 -10
------------------------------------------------------------------------------
(add/remove: 1/0 grow/shrink: 2/1 up/down: 49/-10) Total: 39 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Tue, 3 Apr 2018 00:40:25 +0800
parser: Fix parsing of ${}
dash -c 'echo ${}' should print "Bad subtitution" but instead
fails with "Syntax error: Missing '}'". This is caused by us
reading an extra character beyond the right brace. This patch
fixes it so that this construct only fails during expansion rather
than during parsing.
Fixes: 3df3edd13389 ("[PARSER] Report substition errors at...")
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
function old new delta
readtoken1 2907 2916 +9
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Fri, 23 Mar 2018 18:58:47 +0800
expand: Fix ghost fields with unquoted $@/$*
You're right. The proper fix to this is to ensure that nulonly
is not set in varvalue for $*. It should only be set for $@ when
it's inside double quotes.
In fact there is another bug while we're playing with $@/$*.
When IFS is set to a non-whitespace character such as :, $*
outside quotes won't remove empty fields as it should.
This patch fixes both problems.
Reported-by: Martijn Dekker <martijn@inlv.org>
Suggested-by: Harald van Dijk <harald@gigawatt.nl>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
function old new delta
argstr 1111 1113 +2
evalvar 571 569 -2
varvalue 579 576 -3
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 1/2 up/down: 2/-5) Total: -3 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Sat, 19 May 2018 02:39:43 +0800
var: Set IFS to fixed value at start time
This patch forces the IFS variable to always be set to its default
value, regardless of the environment.
It also removes the long unused IFS_BROKEN code.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Wed, 4 Apr 2018 17:54:01 +0800
eval: Variable assignments on functions are no longer persistent
Dirk Fieldhouse <fieldhouse@gmx.net> wrote:
> In POSIX.1-2017 ("simultaneously IEEE Std 1003.1™-2017 and The Open
> Group Technical Standard Base Specifications, Issue 7")
> <http://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_09>,
> we read under '2.9.1 Simple Commands'
>
> "Variable assignments shall be performed as follows:
> ...
> - If the command name is a standard utility implemented as a function
> (see XBD Utility), the effect of variable assignments shall be as if the
> utility was not implemented as a function.
> ...
> - If the command name is a function that is not a standard utility
> implemented as a function, variable assignments shall affect the current
> execution environment during the execution of the function. It is
> unspecified:
>
> * Whether or not the variable assignments persist after the
> completion of the function
>
> * Whether or not the variables gain the export attribute during
> the execution of the function
>
> * Whether or not export attributes gained as a result of the
> variable assignments persist after the completion of the function (if
> variable assignments persist after the completion of the function)"
POSIX used to require the current dash behaviour. However, you're
right that this is no longer the case.
This patch will remove the persistence of the variable assignment.
I have considered the exporting the variables during the function
execution but have decided against it because:
1) It makes the code bigger.
2) dash has never done this in the past.
3) You cannot use this portably anyway.
Reported-by: Dirk Fieldhouse <fieldhouse@gmx.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
function old new delta
evalcommand 1606 1635 +29
evalcase 313 317 +4
evalfun 280 268 -12
pushlocalvars 48 - -48
------------------------------------------------------------------------------
(add/remove: 0/1 grow/shrink: 2/1 up/down: 33/-60) Total: -27 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
On fast network, I've seen "delay:0.002000" shown for all packets,
thus completely losing information on what real delays are.
The new code is careful to not reject packets with tiny delays
if the delay "grows a lot" but is still tiny:
0.000009 is "much larger" than 0.000001 (nine times larger),
but is still very good small delay.
function old new delta
recv_and_process_peer_pkt 863 889 +26
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Implement the -am argument to allow choosing an AUTH method.
For now only PLAIN and LOGIN are supported, but others can be added
easily in the future.
AUTH PLAIN required adding a new variant of encode_base64() capable of
handling NUL characters in the input string; the old function is now a
wrapper for the newer one.
function old new delta
encode_n_base64 - 236 +236
sendmail_main 1199 1380 +181
packed_usage 32873 32877 +4
encode_base64 242 36 -206
------------------------------------------------------------------------------
(add/remove: 1/0 grow/shrink: 2/1 up/down: 421/-206) Total: 215 bytes
Signed-off-by: Raffaello D. Di Napoli <rafdev@dinapo.li>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Symlinks in a subdirectory that is to become target rootfs are sometimes
dangling because they link to canonical file names that are not present
on the host, but are present relative to the target rootfs root. Don't
copy over dangling symlinks when noclobber is enabled
The -e test treats dangling symlinks as non-existent files. Add -h test
that returns true for all symlinks.
Cc: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
