Re: CONFIG
On Sat, Jul 19, 2008 at 11:31:45AM +0200, Denys Vlasenko wrote:
On Saturday 19 July 2008 08:58, [EMAIL PROTECTED] wrote:
Added verbose comments on config_read().
Added testsuite (runtest parse).
Let us collect interesting cases and thoroughly test them.
Fixed inittab parsing. Bernhard, please test.
- while (config_read(parser, token, -4, 0, :, '#') = 0) {
+ while (config_read(parser, token, 4, 4, #:,
PARSE_DONT_TRIM|PARSE_DONT_REDUCE|PARSE_LAST_IS_GREEDY) 0) {
I'm sceptical about this new interface. How can i state that i
-) accept _no_ comment chars but e.g. ;, as delimiter?
-) accept ^[[:space:]]*[#;] comments and ;[[:space:]] as delim?
The latter is probably not used -- or at least not widely used -- in bb.
Thanks,
___
busybox mailing list
busybox@busybox.net
http://busybox.net/cgi-bin/mailman/listinfo/busybox
Re: CONFIG
Hi! I'm sceptical about this new interface. How can i state that i -) accept _no_ comment chars but e.g. ;, as delimiter? pass \0;, as delims We can go farther and pass comment-chars-if-any\0delimiter-chars as delims; this will require a trivial change in config_read() -) accept ^[[:space:]]*[#;] comments and ;[[:space:]] as delim? The latter is probably not used -- or at least not widely used -- in bb. You answered your question by yourself, Bernhard. It is not required by any existing bb applet by the moment. However, regexps would be nice FEATURE. Just let us work on it... Thanks for discussion, -- Vladimir ___ busybox mailing list busybox@busybox.net http://busybox.net/cgi-bin/mailman/listinfo/busybox
new applets lpr,lpq (update)
Hello Walter, hello Denys, i just tried your busybox/printutils/lpr on my Blackfin/uClinux board. I want to print via network on an external lpd. Printing small files ( a few bytes) works. Printing longer files ( some kBytes) doesnt. /var/log/syslog on the serverside says: Jul 21 09:41:00 PC535li cups-lpd[7691]: Connection from blackfin (IPv4 192.168.70.120) Jul 21 09:41:00 PC535li cups-lpd[7691]: Receive print job for psprint Jul 21 09:41:00 PC535li cups-lpd[7691]: Error while reading file - Connection reset by peer Jul 21 09:41:00 PC535li cups-lpd[7691]: Closing connection I am using BusyBox 1.11.1 (stable) I can provide exported snapshots from wireshark, if you like. Can you help me? PS: Printing from an XScale board using LPRng works. - Mit freundlichen Gruessen / Best regards Klaus Hachmeister Research Development ___ busybox mailing list busybox@busybox.net http://busybox.net/cgi-bin/mailman/listinfo/busybox
Re: new applets lpr,lpq (update)
hello klaus, thanks for you bug report, i am very busy this week; Short: no chance for a quick fix but you may check at what size lpr stops working (my guess 4096 bytes) that may give us a hint where the problem is. perhaps vladimir ([EMAIL PROTECTED]) has some ideas. re, wh [EMAIL PROTECTED] wrote: Hello Walter, hello Denys, i just tried your busybox/printutils/lpr on my Blackfin/uClinux board. I want to print via network on an external lpd. Printing small files ( a few bytes) works. Printing longer files ( some kBytes) doesnt. /var/log/syslog on the serverside says: Jul 21 09:41:00 PC535li cups-lpd[7691]: Connection from blackfin (IPv4 192.168.70.120) Jul 21 09:41:00 PC535li cups-lpd[7691]: Receive print job for psprint Jul 21 09:41:00 PC535li cups-lpd[7691]: Error while reading file - Connection reset by peer Jul 21 09:41:00 PC535li cups-lpd[7691]: Closing connection I am using BusyBox 1.11.1 (stable) I can provide exported snapshots from wireshark, if you like. Can you help me? PS: Printing from an XScale board using LPRng works. - Mit freundlichen Gruessen / Best regards Klaus Hachmeister Research Development ___ busybox mailing list busybox@busybox.net http://busybox.net/cgi-bin/mailman/listinfo/busybox ___ busybox mailing list busybox@busybox.net http://busybox.net/cgi-bin/mailman/listinfo/busybox
Re: new applets lpr,lpq (update)
Klaus, can you send log of strace -f -v -s 1024 lpr your-parameters-here ? -- Vladimir ___ busybox mailing list busybox@busybox.net http://busybox.net/cgi-bin/mailman/listinfo/busybox
Re: new applets lpr,lpq (update)
Another example. (another postscriptfile)
--
/etc/syslog
Jul 21 13:44:29 PC535li cups-lpd[11476]: Connection from blackfin (IPv4
192.168.70.120)
Jul 21 13:44:29 PC535li cups-lpd[11476]: Receive print job for psprint
Jul 21 13:44:29 PC535li cups-lpd[11476]: Error while reading file - Connection
reset by peer
Jul 21 13:44:29 PC535li cups-lpd[11476]: Closing connection
Jul 21 13:44:57 PC535li cups-lpd[11479]: Connection from blackfin (IPv4
192.168.70.120)
Jul 21 13:44:57 PC535li cups-lpd[11479]: Receive print job for psprint
Jul 21 13:44:58 PC535li cups-lpd[11479]: Error while reading trailing nul -
Connection reset by peer
Jul 21 13:44:58 PC535li cups-lpd[11479]: Closing connection
--
/ # lpr [EMAIL PROTECTED] /mnt/nfs/testprint.ps
/ # strace -f -v -s 1024 lpr [EMAIL PROTECTED] /mnt/nfs/testprint.ps
ioctl(0, SNDCTL_TMR_TIMEBASE or TCGETS, {c_iflags=0x500, c_oflags=0x5,
c_cflags=0x1cb1, c_lflags=0x8a3b, c_line=0,
c_cc=\x03\x1c\x08\x15\x04\x00\x01\x00\x0
ioctl(1, SNDCTL_TMR_TIMEBASE or TCGETS, {c_iflags=0x500, c_oflags=0x5,
c_cflags=0x1cb1, c_lflags=0x8a3b, c_line=0,
c_cc=\x03\x1c\x08\x15\x04\x00\x01\x00\x0
getuid32() = 0
getuid32() = 0
open(/etc/passwd, O_RDONLY) = 3
ioctl(3, SNDCTL_TMR_TIMEBASE or TCGETS, 0x462c94) = -1 ENOTTY (Inappropriate
ioctl for device)
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_SHARED|MAP_ANONYMOUS|0x400, 0,
0) = 0xe7
read(3, root:C7O0oWtNuGHNE:0:0:root:/root:/bin/sh\nnobody:x:99:99:Nobody::\n,
256) = 66
close(3)= 0
open(/etc/resolv.conf, O_RDONLY) = -1 ENOENT (No such file or directory)
open(/etc/config/resolv.conf, O_RDONLY) = -1 ENOENT (No such file or
directory)
open(/etc/hosts, O_RDONLY)= 3
ioctl(3, SNDCTL_TMR_TIMEBASE or TCGETS, 0x4625dc) = -1 ENOTTY (Inappropriate
ioctl for device)
read(3, # Do not remove the following line, or various programs\n# that
require network functionality will
fail.\n127.0.0.1\tlocalhost.localdomain\tlocalh9
close(3)= 0
socket(PF_INET, SOCK_STREAM, IPPROTO_IP) = 3
setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
connect(3, {sa_family=AF_INET, sin_port=htons(515),
sin_addr=inet_addr(192.168.70.12)}, 16) = 0
getpid()= 336
uname({sysname=Linux, nodename=blackfin,
release=2.6.22.18-ADI-2008R1-svn, version=#60 Fri Jul 18 13:39:32 CEST
2008, machine=blackfin}) = 0
write(3, \2psprint\n, 9) = 9
read(3, \0, 1)= 1
open(/mnt/nfs/testprint.ps, O_RDONLY) = 4
write(3, \00255 cfA336blackfin\n, 19) = 19
write(3, Hblackfin\nProot\nJ/mnt/nfs/testprint.ps\nldfA336blackfin\n, 55) =
55
Re: new applets lpr,lpq (update)
successful printing
-
/var/log/syslog
Jul 21 13:57:20 PC535li cups-lpd[11552]: Receive print job for psprint
Jul 21 13:57:20 PC535li cups-lpd[11552]: Print file - job ID = 94
Jul 21 13:57:20 PC535li cups-lpd[11552]: Closing connection
Jul 21 13:57:20 PC535li kernel: [19437.375611] audit(1216641440.551:38):
type=1503 operation=inode_permission requested_mask=::r
denied_mask=::r name=/var/run/samba/gencache.tdb pid=11558
profile=/usr/sbin/cupsd namespace=default
Jul 21 13:57:20 PC535li kernel: [19437.375642] audit(1216641440.551:39):
type=1503 operation=inode_permission requested_mask=::r
denied_mask=::r name=/var/run/samba/gencache.tdb pid=11558
profile=/usr/sbin/cupsd namespace=default
Jul 21 13:57:20 PC535li kernel: [19437.385953] audit(1216641440.563:40):
type=1503 operation=inode_permission requested_mask=::r
denied_mask=::r name=/var/run/samba/gencache.tdb pid=11558
profile=/usr/sbin/cupsd namespace=default
Jul 21 13:57:50 PC535li cups-lpd[11561]: Connection from blackfin (IPv4
192.168.70.120)
Jul 21 13:57:50 PC535li cups-lpd[11561]: Receive print job for psprint
Jul 21 13:57:50 PC535li cups-lpd[11561]: Print file - job ID = 95
Jul 21 13:57:50 PC535li cups-lpd[11561]: Closing connection
Jul 21 13:57:50 PC535li kernel: [19467.618088] audit(1216641470.839:41):
type=1503 operation=inode_permission requested_mask=::r
denied_mask=::r name=/var/run/samba/gencache.tdb pid=11567
profile=/usr/sbin/cupsd namespace=default
Jul 21 13:57:50 PC535li kernel: [19467.618823] audit(1216641470.843:42):
type=1503 operation=inode_permission requested_mask=::r
denied_mask=::r name=/var/run/samba/gencache.tdb pid=11567
profile=/usr/sbin/cupsd namespace=default
Jul 21 13:57:50 PC535li kernel: [19467.626842] audit(1216641470.851:43):
type=1503 operation=inode_permission requested_mask=::r
denied_mask=::r name=/var/run/samba/gencache.tdb pid=11567
profile=/usr/sbin/cupsd namespace=default
-
/ # lpr
[EMAIL PROTECTED] /etc/hosts
/ # strace -f -v -s 1024 lpr
[EMAIL PROTECTED] /etc/hosts
ioctl(0, SNDCTL_TMR_TIMEBASE or TCGETS, {c_iflags=0x500, c_oflags=0x5,
c_cflags=0x1cb1, c_lflags=0x8a3b, c_line=0, c_cc=\x03\x1c\x08\x15\x04
\x00\x01\x00\x0
ioctl(1, SNDCTL_TMR_TIMEBASE or TCGETS, {c_iflags=0x500, c_oflags=0x5,
c_cflags=0x1cb1, c_lflags=0x8a3b, c_line=0, c_cc=\x03\x1c\x08\x15\x04
\x00\x01\x00\x0
getuid32() =
0
getuid32() =
0
open(/etc/passwd, O_RDONLY) =
3
ioctl(3, SNDCTL_TMR_TIMEBASE or TCGETS, 0x462c9c) = -1 ENOTTY
(Inappropriate ioctl for
device)
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_SHARED|MAP_ANONYMOUS|
0x400, 0, 0) =
0xfa6000
read(3, root:C7O0oWtNuGHNE:0:0:root:/root:/bin/sh
\nnobody:x:99:99:Nobody::\n, 256) =
66
close(3)=
0
open(/etc/resolv.conf, O_RDONLY) = -1 ENOENT (No such file or
directory)
open(/etc/config/resolv.conf, O_RDONLY) = -1 ENOENT (No such file or
directory)
open(/etc/hosts, O_RDONLY)=
3
ioctl(3, SNDCTL_TMR_TIMEBASE or TCGETS, 0x4625e4) = -1 ENOTTY
(Inappropriate ioctl for
device)
read(3, # Do not remove the following line, or various programs\n# that
require network functionality will fail.\n127.0.0.1
\tlocalhost.localdomain\tlocalh9
close(3)=
0
socket(PF_INET, SOCK_STREAM, IPPROTO_IP) =
3
setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) =
0
Re: CONFIG
On Mon, Jul 21, 2008 at 02:09:59PM +0400, Vladimir Dronnikov wrote: Among the few remaining applets there are inetd, httpd, something in ip (rt_table_*?) and vanilla modprobe.*/modules.dep. Can you enumerate more? Let's build a list (perhaps in TODO) so we can check if all potential users are cought. + ifupdown - init + dumpleases ? Does that make sense? + inetd + bb__pgsreader() + httpd + rtnl_tab_initialize() /* number[[:space:]]entry\n#comment */ + modules.dep, modules.alias + mount + expand/unexpand (?, probably no) + procps_scan() for ENABLE_FEATURE_TOPMEM + parse_config_file() /* our main /etc/busybox.conf */ + check_securetty() + lsmod (/proc/modules parsing) + do_sethostname() /* if benefical */ + interface.c etc.. I didn't update the list, so please mark the ones that are already converted or missing. TIA ___ busybox mailing list busybox@busybox.net http://busybox.net/cgi-bin/mailman/listinfo/busybox
Re: CONFIG
+ check_securetty() + do_sethostname() /* if benefical */ These I attacked yesterday. Pending... -- Vladimir ___ busybox mailing list busybox@busybox.net http://busybox.net/cgi-bin/mailman/listinfo/busybox
Re: 'test' behavior.
Denys Vlasenko wrote: On Friday 18 July 2008 22:30, Chris Larson wrote: It occurs with current busybox svn. Original test case: -f = a -o b Minimal test case: -f = a w/ explicit 3 argument handling disabled in the code Try attached patch. That fixed it, thanks a lot :) -- Chris Larson Dedicated Engineer Montavista, Inc. Email: clarson AT mvista DOT com Email: clarson AT kergoth DOT com ___ busybox mailing list busybox@busybox.net http://busybox.net/cgi-bin/mailman/listinfo/busybox
Fun w/ test :)
[EMAIL PROTECTED] busybox% ./busybox test -f = [EMAIL PROTECTED] busybox% ./busybox test \( -f = \) test: closing paren expected Hehe. I highly doubt anyone would actually name a file =, but what the heck, figured you'd want to know about any cases that confuse the parser. It's interesting to note that -f = works, but ( -f = ) does not. No customer issue on this of course, just made myself try lots of combinations during the testing. I found a case where the unknown operand error is seen. If it sees an operand after an expression, which is not -o or -a, it gives that error: [EMAIL PROTECTED] busybox% ./busybox test -f = -a [EMAIL PROTECTED] busybox% ./busybox test -f = -a -e foo test: -e: unknown operand Due to the precedence, that's seen as two expressions without a -a or -o to join them. The error message could be confusing in this case. -e is, of course, a valid operator, and -e foo is a valid expression, but -e is not valid -there-, only -a or -o are. -- Chris Larson Dedicated Engineer Montavista, Inc. Email: clarson AT mvista DOT com Email: clarson AT kergoth DOT com ___ busybox mailing list busybox@busybox.net http://busybox.net/cgi-bin/mailman/listinfo/busybox
Re: new applets lpr,lpq (update)
On Monday 21 July 2008 12:57, [EMAIL PROTECTED] wrote:
Shure,
now the output of /etc/syslog (serverside) is differrent. (but still no
printing)
--
/etc/syslog
Jul 21 12:44:36 PC535li cups-lpd[11071]: Connection from blackfin (IPv4
192.168.70.120)
Jul 21 12:44:36 PC535li cups-lpd[11071]: Receive print job for psprint
Jul 21 12:44:36 PC535li cups-lpd[11071]: Error while reading file -
Connection reset by peer
Jul 21 12:44:36 PC535li cups-lpd[11071]: Closing connection
Jul 21 12:45:24 PC535li cups-lpd[11072]: Connection from blackfin (IPv4
192.168.70.120)
Jul 21 12:45:24 PC535li cups-lpd[11072]: Receive print job for psprint
Jul 21 12:45:48 PC535li cups-lpd[11072]: Print file - job ID = 91
Jul 21 12:45:48 PC535li cups-lpd[11072]: Closing connection
One printing was ok (second one), another failed.
/ # lpr
[EMAIL PROTECTED] /mnt/nfs/testpage.ps
/ # strace -f -v -s 1024 lpr
[EMAIL PROTECTED] /mnt/nfs/testpage.ps
ioctl(0, SNDCTL_TMR_TIMEBASE or TCGETS, {c_iflags=0x500, c_oflags=0x5,
c_cflags=0x1cb1, c_lflags=0x8a3b, c_line=0, c_cc=\x03\x1c\x08\x15\x04
\x00\x01\x00\x0
...
...
I find it very pleasant to spend 15 minutes unwrapping the strace,
removing extra spaces and the like. However note that some other
developers may have less perverted methods of having fun, and may
resort to simple entertainments like beating the email's author
with big stick. :)
Let's take a look.
# strace -f -v -s 1024 lpr [EMAIL PROTECTED] /mnt/nfs/testpage.ps
ioctl(0, SNDCTL_TMR_TIMEBASE or TCGETS, {c_iflags=0x500, c_oflags=0x5,
c_cflags=0x1cb1, c_lflags=0x8a3b, c_line=0,
c_cc=\x03\x1c\x08\x15\x04\x00\x01\x00\x0
ioctl(1, SNDCTL_TMR_TIMEBASE or TCGETS, {c_iflags=0x500, c_oflags=0x5,
c_cflags=0x1cb1, c_lflags=0x8a3b, c_line=0,
c_cc=\x03\x1c\x08\x15\x04\x00\x01\x00\x0
...
close(3)= 0
nothing interesting above.
socket(PF_INET, SOCK_STREAM, IPPROTO_IP) = 3
setsockopt(3, SOL_SOCKET, SO_REUSEADDR, [1], 4) = 0
connect(3, {sa_family=AF_INET, sin_port=htons(515),
sin_addr=inet_addr(192.168.70.12)}, 16) = 0
We opened a TCP connection to 192.168.70.12:515.
getpid()= 260
uname({sysname=Linux, nodename=blackfin,
release=2.6.22.18-ADI-2008R1-svn, version=#60 Fri Jul 18 13:39:32 CEST
2008, machine=blackfin}) = 0
write(3, \2psprint\n, 9) = 9
We say I want to print to queue 'psprint'.
read(3, \0, 1)= 1
Server says ok
open(/mnt/nfs/testpage.ps, O_RDONLY) = 4
write(3, \00254 cfA260blackfin\n, 19) = 19
We say here goes the controlfile (\002)
of 54 bytes (54) named cfA260blackfin
write(3, Hblackfin\nProot\nJ/mnt/nfs/testpage.ps\nldfA260blackfin\n, 54) = 54
We send controlfile contents:
Hhost\n, Puser\n, Jjobname\n ldatafilename\n
write(3, \0, 1) = 1
We say is it ok?
read(3, \0, 1)= 1
server says ok
fstat(4, {st_dev=makedev(0, 15), st_ino=696322, st_mode=S_IFREG|0644,
st_nlink=1, st_uid=0, st_gid=0, st_blksize=8192, st_blocks=520, st_size=260183,
st_at0
write(3, \003260183 dfA260blackfin\n, 23) = 23
We say here is a datafile, 260183 bytes, named dfA260blackfin
read(4, %!PS-Adobe-2.0\n%%Title: testpage.fig\n%%Creator: fig2dev Version 3.2
Patchlevel 3d\n%%CreationDate: Mon Dec 3 11:54:07 2001\n%%For: [EMAIL
PROTECTED]
write(3, %!PS-Adobe-2.0\n%%Title: testpage.fig\n%%Creator: fig2dev Version 3.2
Patchlevel 3d\n%%CreationDate: Mon Dec 3 11:54:07 2001\n%%For: [EMAIL
PROTECTED]
read(4, Bold ff 150.00 scf sf\n5700 7800 m\ngs 1 -1 sc (Gold) dup swpop 2 div
neg 0 rm col-1 sh gr\n/Helvetica-Bold ff 150.00 scf sf\n69007800 m\ngs 1 6
write(3, Bold ff 150.00 scf sf\n5700 7800 m\ngs 1 -1 sc (Gold) dup swpop 2 div
neg 0 rm col-1 sh gr\n/Helvetica-Bold ff 150.00 scf sf\n69007800 m\ngs 16
...
read(4, 00dfff4e4a4e\n009b004e003d024e003d0e4e029b034e005e009b354e\nrestore
grestore\n%\n% End Imported PIC File: LPRngT-L.gif\n%\n/Times-Roman ff 180.00 5
write(3, 00dfff4e4a4e\n009b004e003d024e003d0e4e029b034e005e009b354e\nrestore
grestore\n%\n% End Imported PIC File: LPRngT-L.gif\n%\n/Times-Roman ff 180.005
we sent the file
write(3, \0, 1) = 1
We say is it ok?
read(3, \0, 1)= 1
server says ok
close(3)= 0
_exit(0)
we exit.
What confuses me here is that it looks like server was totally happy
and actually accepted the job. I suspect you sent a strace of
successful lpr.
--
vda
___
busybox mailing list
busybox@busybox.net
http://busybox.net/cgi-bin/mailman/listinfo/busybox
Re: CONFIG
On Sunday 20 July 2008 22:00, [EMAIL PROTECTED] wrote:
*_for_{read,write} introduced and used. Wonder what is the size change?
Please consider applying.
- fp = fopen(/etc/securetty, r);
- if (!fp) {
- /* A missing securetty file is not an error. */
- return 1;
- }
- while (fgets(buf, sizeof(buf)-1, fp)) {
- for (i = strlen(buf)-1; i = 0; --i) {
- if (!isspace(buf[i]))
+ char *buf;
+ int ret = 1;
+ parser_t *parser = config_open(/etc/securetty);
Should have used config_open2(..., fopen_for_read) here,
as it shouldn't warn.
I hope you actually tested that securetty code still works.
Size:
--
(add/remove: 5/1 grow/shrink: 8/74 up/down: 174/-737)Total: -563 bytes
textdata bss dec hex filename
788506 6016696 795803 c249b busybox_old
787943 6016696 795240 c2268 busybox_unstripped
Good work!!!
Applied.
--
vda
___
busybox mailing list
busybox@busybox.net
http://busybox.net/cgi-bin/mailman/listinfo/busybox
Re: Fun w/ test :)
Denys Vlasenko wrote: Paul, hi. :) On Monday 21 July 2008 20:35, Chris Larson wrote: [EMAIL PROTECTED] busybox% ./busybox test -f = [EMAIL PROTECTED] busybox% ./busybox test \( -f = \) test: closing paren expected Hehe. I highly doubt anyone would actually name a file =, but what the heck, figured you'd want to know about any cases that confuse the parser. It's interesting to note that -f = works, but ( -f = ) does not. No customer issue on this of course, just made myself try lots of combinations during the testing. I found a case where the unknown operand error is seen. If it sees an operand after an expression, which is not -o or -a, it gives that error: [EMAIL PROTECTED] busybox% ./busybox test -f = -a [EMAIL PROTECTED] busybox% ./busybox test -f = -a -e foo test: -e: unknown operand Due to the precedence, that's seen as two expressions without a -a or -o to join them. The error message could be confusing in this case. -e is, of course, a valid operator, and -e foo is a valid expression, but -e is not valid -there-, only -a or -o are. So, now, do you see why xstr trick is actually not so silly? I now recall reading about this in the shell portability section of the autoconf manual. Excerpt: Posix also says that `test ! string', `test -n string' and `test -z string' work with any string, but many shells (such as Solaris, AIX 3.2, unicos 10.0.0.6, Digital Unix 4, etc.) get confused if string looks like an operator: $ test -n = test: argument expected $ test ! -n test: argument expected Similarly, Posix says that both `test string1 = string2' and `test string1 != string2' work for any pairs of strings, but in practice this is not true for troublesome strings that look like operators or parentheses, or that begin with `-'. It is best to protect such strings with a leading `X', e.g., `test Xstring != X' rather than `test -n string' or `test ! string'. -- Chris Larson Dedicated Engineer Montavista, Inc. Email: clarson AT mvista DOT com Email: clarson AT kergoth DOT com ___ busybox mailing list busybox@busybox.net http://busybox.net/cgi-bin/mailman/listinfo/busybox
