$ echo $$
18448
$ echo $? <wait here, run "kill -INT 18448" in other shell><press enter>
<=== NOTHING??
$
That empty line does not look right. After this patch:
$ echo $$
18448
$ echo $? <wait here, run "kill -INT 18448" in other shell><press enter>
^C
$
function old new delta
fgetc_interactive 245 246 +1
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
"rm -i FILE" and "yes" can now be interrupted by ^C in hush.
This also now works:
$ usleep 19999999
^C
$ echo $?
130
function old new delta
run_pipe 1668 1711 +43
pseudo_exec_argv 312 321 +9
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 2/0 up/down: 52/0) Total: 52 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Sun, 13 Jul 2008 22:34:50 +0800
[OPTIONS] Added support for -l
This patch adds support for the -l option (login shell) as required
by the LSB.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
It's a bit bigger, but gets rid of one global variable
function old new delta
options 554 576 +22
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
The fact that shell has open fds to tty and/or scripts should be
unobservable, if possible. In particular, if redirect tries to dup
one of them via ">&script_fd", it's better to pretend that script_fd
is closed, and thus redirect fails with EBADF.
Fixes these two testcase failures:
ash-redir/redir_to_bad_fd.tests
hush-redir/redir_to_bad_fd3.tests
function old new delta
redirect 1018 1129 +111
setup_redirects 250 359 +109
readtoken1 2651 2655 +4
cmdloop 185 187 +2
changepath 194 195 +1
save_fd_on_redirect 203 194 -9
evaltree 501 484 -17
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 5/2 up/down: 227/-26) Total: 201 bytes
text data bss dec hex filename
914553 485 6848 921886 e111e busybox_old
914754 485 6848 922087 e11e7 busybox_unstripped
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
New code is similar to what hush is doing.
Make CLOSED to -1: same as dash.
popredir() loses "restore" parameter: same as dash.
COPYFD_RESTORE bit is no longer necessary.
This change fixes this interactive bug:
$ ls -l /proc/$$/fd 10>&-
ash: can't set tty process group: Bad file descriptor
ash: can't set tty process group: Bad file descriptor
[1]+ Done(2) ls -l /proc/${\$}/fd 10>&4294967295
function old new delta
unwindredir 29 27 -2
tryexec 154 152 -2
evaltree 503 501 -2
evalcommand 1369 1367 -2
cmdloop 187 185 -2
redirect 1029 1018 -11
popredir 153 123 -30
need_to_remember 36 - -36
is_hidden_fd 68 - -68
------------------------------------------------------------------------------
(add/remove: 0/2 grow/shrink: 0/7 up/down: 0/-155) Total: -155 bytes
text data bss dec hex filename
914572 485 6848 921905 e1131 busybox_old
914553 485 6848 921886 e111e busybox_unstripped
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Thu, 27 May 2010 15:03:46 +0800
[REDIR] Fix popredir on abnormal exit from built-in
Just like the poplocalvar problem recently fixed, redirections
can also be leaked in case of an abnormal exit. This patch fixes
it using the same method as poplocalvar, by storing the previous
redirection state and restoring to that point.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Thu, 27 Dec 2007 13:54:16 +1100
[PARSER] Add FAKEEOFMARK for expandstr
Previously expandstr used the string "" to indicate that it needs to be
treated just like a here-doc except that there is no terminator. However,
the string "" is in fact a valid here-doc terminator so now that we deal
with it correctly expandstr no longer works in the presence of new-lines
in the prompt.
This patch introduces the FAKEEOFMARK macro which does not equal any
real EOF marker but is distinct from the NULL pointer which is used to
indicate non-here-doc contexts.
Thanks to Markus Triska for reporting this regression.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Unfortunately, I did not find the failing example for this old fix.
I also tweaked the code which was added by this commit:
"
Date: Mon Sep 24 18:30:02 2007 +0000
ash: fix prompt expansion (Natanael Copa <natanael.copa@gmail.com>)
"
since other parts of code do expect expandstr() to use DQSYNTAX, not PSSYNTAX.
function old new delta
parse_stream 2609 2634 +25
setprompt_if 128 133 +5
read_profile 32 37 +5
evalcommand 1334 1339 +5
expandstr 122 120 -2
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 4/1 up/down: 40/-2) Total: 38 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream comment:
Date: Sun, 11 Nov 2007 14:21:23 +0800
[PARSER] Removed noexpand/length check on eofmark
On Tue, Oct 30, 2007 at 04:23:35AM +0000, Oleg Verych wrote:
>
> } 8<<""
> ======================
Actually this (the empty delim) only works with dash by accident.
I've tried bash and pdksh and they both terminate on the first
empty line which is what you would expect rather than EOF. The
real Korn shell does something completely different.
I've fixed this in dash to conform to bash/pdksh.
> In [0] it's stated, that delimiter isn't evaluated (expanded), only
> quoiting must be checked. That if() seems to be completely bogus.
OK I agree. The reason it was there is because the parser would
have already replaced the dollar sign by an internal representation.
I've fixed it properly with this patch.
Test case:
cat <<- $a
OK
$a
cat <<- ""
OK
echo OK
Old result:
dash: Syntax error: Illegal eof marker for << redirection
OK
echo OK
New result:
OK
OK
OK
function old new delta
parsefname 227 152 -75
readtoken1 2819 2651 -168
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 0/2 up/down: 0/-243) Total: -243 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
The '%m' conversion specifier prints an error message based on the
current value of 'errno'. It is available in the GNU C library,
Cygwin (since 2012), uClibc and musl.
It is not available in various BSDs, BSD-derived systems (MacOS,
Android) or Microsoft Windows.
Use a symbol defined in platform.h to control how error messages
can be formatted to display the 'errno' message. On platforms that
support it use '%m'; on other platforms use '%s' and strerror().
On platforms that have '%m' there is essentially no change in the
size of the binary. Otherwise:
function old new delta
redirect 1287 1310 +23
xtcsetpgrp 27 44 +17
dup2_or_raise 34 51 +17
setinputfile 267 275 +8
.rodata 163379 163371 -8
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 4/1 up/down: 65/-8) Total: 57 bytes
Signed-off-by: Ron Yorston <rmy@pobox.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Defining a function alias with __attribute__ ((alias("evaltree"),__noreturn__))
is not that usual, and clang had a bug which made it misunderstand
this construct.
Switch to:
ALWAYS_INLINE NORETURN evaltreenr() { evaltree(); unreachable(); }
Older gcc's do not know unreachable(), on them we pay the price of having
a few extra calls to abort():
function old new delta
evalsubshell 151 156 +5
evalpipe 357 362 +5
argstr 1141 1144 +3
On newer gcc, code size does not change.
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Tue, 6 Jul 2010 17:40:53 +0800
[VAR] Fix loss of variables when hash collides
Brian Koropoff reported that the new var patches broke the following
script:
#!/bin/dash
GDM_LANG="bar"
OPTION="foo"
unset GDM_LANG
# OPTION has mysteriously become unset
echo "$OPTION"
He correctly diagnosed this as a result of removing all variables
in the hash chain preceding the one that should be removed in
setvareq.
He also provided a patch to fix this.
This patch is based on his but without keeping the original vpp.
As a result, we now store new variables at the end of the hash
chain instead of the beginning.
To make this work, setvareq/setvar now returns the vp pointer
modified. In case they're used to unset a variable the pointer
returned is undefined. This is because mklocal needs it and
used to get it by assuming that the new variable always appear
at the beginning of the chain.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Thu, 27 May 2010 11:50:19 +0800
[VAR] Do not poplocalvars prematurely on regular utilities
The recent cmdenviron removal broke regular utilities by calling
poplocalvars too early. This patch fixes that by postponing the
poplocalvars for regular utilities until they have completed.
In order to ensure that local still works, it is now a special
built-in.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Thu, 27 May 2010 11:32:55 +0800
[VAR] Fix poplocalvar on abnormal exit from function
The new localvar code broke the abnormal exit from functions
and built-ins by not restoring the original localvar state.
This patch fixes this by storing the previous localvar state so
that we always unwind correctly in case of an abnormal exit.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Wed, 26 May 2010 18:54:19 +0800
[VAR] Replace cmdenviron with localvars
This patch replaces the cmdenviron mechanism for temporary command
variables with the localvars mechanism used by functions.
This reduces code size, and more importantly, makes the variable
assignment take effect immediately as required by POSIX.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Tue, 25 May 2010 18:14:32 +0800
[VAR] Fix poplocalvar leak
When a variable is marked as local, we set VSTRFIXED on its vp
recored. However, poplocalvar never clears this flag for variables
that were unset to begin with. Thus if you ever made an unset
variable local, it would get the VSTRFIXED bit and stick around
forever.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Mon, 24 May 2010 15:31:27 +0800
[VAR] Add localvars nesting
This patch adds localvars nesting infrastructure so we can reuse
the localvars mechanism for command evaluation.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
If STANDALONE and we run a NOEXEC applet, saved copies of redirected fds
were visible for the child. They have CLOEXEC bit, yes, but we do not exec
in this case.
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Thu, 7 Jul 2011 13:58:48 +0800
[SHELL] Optimize dash -c "command" to avoid a fork
On Sun, Apr 10, 2011 at 07:36:49AM +0000, Jonathan Nieder wrote:
> From: Jilles Tjoelker <jilles@stack.nl>
> Date: Sat, 13 Jun 2009 16:17:45 -0500
>
> This change only affects strings passed to -c, when the -s option is
> not used.
>
> Use the EV_EXIT flag to inform the eval machinery that the string
> being passed is the entirety of input. This way, a fork may be
> omitted in many special cases.
>
> If there are empty lines after the last command, the evalcmd will not
> see the end early enough and forks will not be omitted. The same thing
> seems to happen in bash.
>
> Example:
> sh -c 'ps lT'
> No longer shows a shell process waiting for ps to finish.
>
> [jn: ported from FreeBSD SVN r194128. Bugs are mine.]
>
> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Instead of detecting EOF using the input layer, I'm going to
use the parser instead. In either case, we always have to read
ahead in order to complete the parsing of the previous node.
Therefore we always know whether there is more to come, except
in the case where we see a newline/semicolon or similar.
For the purposes of sh -c, this should be sufficient.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
function old new delta
evalstring 190 224 +34
ash_main 1014 1022 +8
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 2/0 up/down: 42/0) Total: 42 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Thu, 10 Mar 2011 16:52:13 +0800
[REDIR] Replace GPL noclobberopen code with the FreeBSD version
Replace noclobberopen() from bash with the FreeBSD code for noclobber
opens.
This also reduces code size by eliminating an unnecessary check.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
function old new delta
changepath 192 194 +2
localcmd 366 364 -2
expmeta 521 517 -4
redirect 1210 1135 -75
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 1/3 up/down: 2/-81) Total: -79 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
We fixed the problem differently than they. Let's not deviate.
Upstream commit:
Date: Thu, 27 May 2010 20:07:29 +1000
[EXPAND] Fix corruption of redirections with byte 0x81
In other ash variants, a partial implementation of ksh-like cmd >file*
adds and removes CTLESC bytes ('\x81') in redirection filenames,
preserving 8-bit transparency. Long ago, dash removed the code to add
the CTLESC bytes, but not the code to remove them, causing corruption of
filenames containing CTLESC. This commit removes the code to remove the
CTLESC bytes.
The CTLESC byte occurs frequently in UTF-8 encoded non-Latin text.
This bug has been reported various times to Ubuntu and Debian (e.g.
Launchpad Ubuntu #422298). This patch is the same as the one submitted
by Alexander Korolkov in Ubuntu #422298.
Signed-off-by: Jilles Tjoelker <jilles@stack.nl>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
function old new delta
changepath 194 192 -2
expandarg 1000 984 -16
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 0/2 up/down: 0/-18) Total: -18 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Tue, 25 May 2010 20:55:05 +0800
[VAR] Move unsetvar functionality into setvareq
This patch moves the unsetvar code into setvareq so that we can
no have a pathological case of an unset variable hanging around
unless it has a bit pinning it like VEXPORT.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
function old new delta
setvareq 227 303 +76
expmeta 517 521 +4
localcmd 364 366 +2
unsetcmd 96 76 -20
unsetvar 129 7 -122
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 3/2 up/down: 82/-142) Total: -60 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
This change fixes the build in setups where there are
no headers defining WIFSTOPPED and WSTOPSIG (where JOBS has to be
set to 0).
This partially reverts 4700fb5be (ash: make dowait() a bit more
readable. Logic is unchanged, 2015-10-09).
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
$ PATH=/extra/path:/usr/sbin:/usr/bin:/sbin:/bin \
> busybox sh -xc 'command -V ls; command -V ls; command -Vp ls; command -vp ls'
+ command -V ls
ls is /bin/ls
+ command -V ls
ls is a tracked alias for /bin/ls
+ command -Vp ls
ls is a tracked alias for (null)
+ command -vp ls
Segmentation fault
describe_command should respect `path' argument. Looking up in the hash table
may gives incorrect index in entry.u.index and finally causes incorrect output
or SIGSEGV.
function old new delta
describe_command 386 313 -73
Signed-off-by: Youfu Zhang <zhangyoufu@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
When using musl libc glob() a very long string can cause glob() to fail,
which leads to an out of memory error being raised by ash.
This can happen easily if a very long quoted string contains *, even
though no glob expansion should ever be performed on it (since it's
quoted).
Fix this by properly parsing control characters and escaping and only
accept unquoted metacharacters. While we're at it, unify this check for
libc and built-in glob expansion
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
bash has a feature: it restores termios after a successful wait for
a foreground job which had at least one stopped or sigkilled member.
The probable rationale is that SIGSTOP and SIGKILL can preclude task from
properly restoring tty state. Should we do this too?
A reproducer: ^Z an interactive python:
$ python
Python 2.7.12 (...)
>>> ^Z
{ python leaves tty in -icanon -echo state. We do survive that... }
[1]+ Stopped python
{ ...however, next program (python no.2) does not survive it well: }
$ python
Python 2.7.12 (...)
>>> Traceback (most recent call last):
{ above, I typed "qwerty<CR>", but -echo state is still in effect }
File "<stdin>", line 1, in <module>
NameError: name 'qwerty' is not defined
The implementation is modeled on bash code and seems to work.
However, I'm not sure we should do this. For one: what if I'd fg
the stopped python instead? It'll be confused by "restored" tty state.
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Mike deleted it:
commit 39456a18a1
Author: Mike Frysinger <vapier@gentoo.org>
Date: Sat Mar 28 12:21:57 2009 +0000
stop lying about [[ test support
probably because it was not properly ifdefed around, and was enabled
even when bash compat is off.
I just tested it - it works:
$ [ *.diff = z.diff ]; echo $?
0
$ [[ *.diff = z.diff ]]; echo $?
1
Of course, not all numerous bash tricks of [[ ]] are implemented...
function old new delta
bltins2 60 72 +12
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Splitting these options makes it self-documenting about what
bash-compatible features we have.
Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Splitting these options makes it self-documenting about what
bash-compatible features we have.
Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Redundant help texts (one which only repeats the description)
are deleted.
Descriptions and help texts are trimmed.
Some config options are moved, even across menus.
No config option _names_ are changed.
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
It's a bit overkill (who would want it off?) but ash already has it
configurable. Let's be symmetric.
Also tweak kbuild logic to use ASH_BUILTIN_ECHO to select echo.o,
not ASH.
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Also made it and printf, type and wait builtins optional.
function old new delta
builtin_kill - 323 +323
bltins1 336 348 +12
builtin_type 114 116 +2
------------------------------------------------------------------------------
(add/remove: 1/0 grow/shrink: 2/0 up/down: 337/0) Total: 337 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
commit e19923f665 deleted clearredir()
call in shellexec():
ash: [REDIR] Remove redundant CLOEXEC calls
Upstream commit:
Now that we're marking file descriptors as CLOEXEC in savefd, we no longer
need to close them on exec or in setinputfd.
but it missed one place where we don't set CLOEXEC. Fixing this.
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
This would makes all ash options indented inside "ash" in menuconfig.
It appears that menuconfig has a limit at tracking multiple dependency
lines like this (it looks like a "diamond problem" but I'm not sure if
it is):
---ASH <----------
/ \ ASH_OPTIMIZE_FOR_SIZE
!NOMMU <-*----SH_IS_ASH <----[OR] <--ASH_INTERNAL_GLOB
\ / ASH_RANDOM_SUPPORT
---BASH_IS_ASH <-- [...]
The kconfig-language document [1] states that:
> If a menu entry somehow depends on the previous entry, it can be
> made a submenu of it. First, the previous (parent) symbol must be
> part of the dependency list and then one of these two conditions
> must be true:
> - the child entry must become invisible, if the parent is set to 'n'
[BusyBox ash used to satisfy this, but no longer does]
> - the child entry must only be visible, if the parent is visible
[BusyBox ash configs actually satisfy this, but because of
"diamond" above this might not be easily detected]
So I found out a direct workaround: by making ash options explicitly
depend on !NOMMU, we can tell menuconfig that rule 2 above is satisfied
without any more tracking.
---------------------
/ \
!NOMMU <-*-----ASH <-------- \
\ \ \ ASH_OPTIMIZE_FOR_SIZE
*---SH_IS_ASH <---[OR]-[AND] <--ASH_INTERNAL_GLOB
\ / ASH_RANDOM_SUPPORT
--BASH_IS_ASH <- [...]
So all ash options would now be indented under "ash".
[1] "Documentation/kbuild/kconfig-language.txt" in Linux kernel source
Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Mention the behavior if user selects CONFIG_SH_IS_ASH but not
CONFIG_ASH. We will be explicit that invocations like "busybox ash"
will not work for such configuration.
Also clarify help text of CONFIG_BASH_IS_* that bash compatibility in
ash is not complete. (It shouldn't be anyway - ash can't support every
bash quirk out there.)
Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
The commit 'ash,hush: set exit code 127 in "sh /does/not/exist" case'
only partly implemented the dash commit '[ERROR] Allow the originator
of EXERROR to set the exit status'. This resulted in incorrect error
codes for a syntax error:
$ )
$ echo $?
0
or a redirection error for a special builtin:
$ rm -f xxx
$ eval cat <xxx
$ echo $?
0
Signed-off-by: Ron Yorston <rmy@pobox.com>
Reported-by: Martijn Dekker <martijn@inlv.org>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
The same can be done for msh, but we are probably better off just deleting it
in a next versio or two.
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
PIPE_SEQ is used most often, having it zero makes code smaller:
function old new delta
done_word 719 707 -12
parse_stream 2546 2531 -15
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
I thought gcc can detect this itself. It doesn't.
function old new delta
run_list 1030 1021 -9
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Was playing with "sleep 3 | exit 3 & wait %1" and noticed that often
SIGCHLD arrives even before I get to signal masking. Can avoid it in this
case.
function old new delta
wait_for_child_or_signal 228 265 +37
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Also add tests. wait5.tests so far fails (but works for ash and dash).
function old new delta
builtin_wait 305 283 -22
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
It would be nice to provide bash-like "remember las exitcode"
thingy, but it's a bit complex. For now, match ash and dash.
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
In this example:
ash -c 'readonly x; echo $(command eval x=2)'
evalstring() is called after forkchild(), which calls popallfiles().
On exception, evalstring() will popfile().
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
It was not properly interruptible, and did not update job status
(the exited processes were still thought of as running).
function old new delta
process_wait_result - 453 +453
wait_for_child_or_signal - 199 +199
run_list 996 1002 +6
checkjobs_and_fg_shell 41 43 +2
builtin_wait 328 215 -113
checkjobs 516 142 -374
------------------------------------------------------------------------------
(add/remove: 2/0 grow/shrink: 2/2 up/down: 660/-487) Total: 173 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit:
Date: Sun, 22 Feb 2009 18:10:01 +0800
[JOBS] Fix dowait signal race
This test program by Alexey Gladkov can cause dash to enter an
infinite loop in waitcmd.
#!/bin/dash
trap "echo TRAP" USR1
stub() {
echo ">>> STUB $1" >&2
sleep $1
echo "<<< STUB $1" >&2
kill -USR1 $$
}
stub 3 &
stub 2 &
until { echo "###"; wait; } do
echo "*** $?"
done
The problem is that if we get a signal after the wait3 system
call has returned but before we get to INTON in dowait, then
we can jump back up to the top and lose the exit status. So
if we then wait for the job that has just exited, then it'll
stay there forever.
I made the original change that caused this bug to fix pretty
much the same bug but in the opposite direction. That is, if
we get a signal after we enter wait3 but before we hit the kernel
then it too can cause the wait to go on forever (assuming the
child doesn't exit).
In fact this is pretty much exactly the scenario that you'll
find in glibc's documentation on pause(). The solution is given
there too, in the form of sigsuspend, which is the only way to
do the check and wait atomically.
So this patch fixes Alexey's race without reintroducing the old
bug by converting the blocking wait3 to a sigsuspend.
In order to do this we need to set a signal handler for SIGCHLD,
so the code has been modified to always do that.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
I failed to reproduce the bug (it requires precise timing), but it seems real.
function old new delta
dowait 284 463 +179
setsignal 301 326 +25
signal_handler 59 76 +17
ash_main 1481 1487 +6
localcmd 350 348 -2
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 4/1 up/down: 227/-2) Total: 225 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Upstream commit 1:
Date: Sun, 22 Feb 2009 18:16:13 +0800
[SIGNAL] Remove EXSIG
Now that waitcmd no longer uses EXSIG we can remove it.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Upstream commit 2:
Date: Thu, 2 Oct 2014 21:07:55 +0800
[ERROR] Set exitstatus in onint
Currently the exit status when we receive SIGINT is set in evalcommand
which means that it doesn't always get set. For example, if you press
CTRL-C at the prompt of an interactive dash, the exit status is not
set to 130 as it is in many other Bourne shells.
This patch fixes this by moving the setting of the exit status into
onint which also simplifies evalcommand.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Upstream commit 3:
Date: Fri, 3 Oct 2014 14:07:07 +0800
[EVAL] Do not clobber exitstatus in evalcommand
All originators of EXERROR have been setting the exitstatus for
a while now. So it is no longer appropriate to set it explicitly
in evalcommand.
In fact doing so may cause the original exitstatus to be lost.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Last three coomits:
function old new delta
waitcmd 186 224 +38
dowait 276 284 +8
waitforjob 104 107 +3
localcmd 348 350 +2
showjobs 64 61 -3
forkshell 263 260 -3
raise_interrupt 93 67 -26
blocking_wait_with_raise_on_sig 40 - -40
evalcommand 1264 1208 -56
evaltree 809 498 -311
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
