0000796: Applet name wrong in applets struct - BusyBox Bug and Patch Tracking

BusyBox Bug and Patch Tracking

Viewing Issue Simple Details [ Jump to Notes ]

[ View Advanced ] [ Issue History ] [ Print ]

Category

Severity

Reproducibility

Date Submitted

Last Update

0000796

[BusyBox] Documentation

minor

always

03-24-06 04:42

11-29-06 06:42

Reporter

View Status

public

Assigned To

BusyBox

Priority

normal

Resolution

unable to reproduce

Status

closed

Product Version

1.00

Summary

0000796: Applet name wrong in applets struct

Description

I'm experiencing a strange problem with busybox (latest version from CVS) on uClinux running on Microblaze. The problem is as follows:

I enable the vi applet in the BB config in uClinux, compiles it and downloads to target. When I try to run "vi", busybox just gives me the same printout as I get when I run "busybox". In the list of currently defined functions, the vi string is missing like this:

"tar, test, time, touch, true, umount, uname, uptime, , wget, yes"
^vi should go here

If I modifiy busybox.c main() and adds a for-loop printing out all applet names from the applet-struct it works! The for-loop is addded immediately before the run_applet_by_name() call. applets[XX]->name (XX eq. vi position) then points to a string among all the other strings for the names of the other applets.

If I instead removes the printout from the main() and does the same printout first thing in the run_applet_by_name() function in applets.c, I get another result! Now applets[XX]->name points to a different area of the memory containing the value 0 -- the string seems empty. This result in the bsearch failing and the vi applet never executes!

If I don't enable the vi applet in the config some other application in end of the applets struct gets it's name crippled. I haven't managed to see any pattern in what position in applets struct gets crippled.

Because of my limited debug tools, I really can't get that much more information.

The for-loop used for the printout looks like this:

for (i = 50; i < 61; i++) printf("%d: 0x%.8x,0x%.8x -> %s\n", i, applets[i].name, &applets[i].name, applets[i].name);

and a printout like this:

50: 0x25ad0638,0x25ad08dc -> tar
51: 0x25ad063c,0x25ad08e8 -> test
52: 0x25ad0644,0x25ad08f4 -> time
53: 0x25ad064c,0x25ad0900 -> touch
54: 0x25ad0654,0x25ad090c -> true
55: 0x25ad065c,0x25ad0918 -> umount
56: 0x25ad0664,0x25ad0924 -> uname
57: 0x25ad066c,0x25ad0930 -> uptime
58: 0x25ad04cc,0x25ad093c ->
59: 0x25ad0678,0x25ad0948 -> wget
60: 0x25ad0680,0x25ad0954 -> yes

I hope I can get back with more information soon.

Additional Information

Attached Files

Relationships

Notes
(0001414) bernhardf 06-11-06 10:40	Please provide more information on this one. If you can reproduce it with trunk, then please send us an update about this issue. If you don't see it with trunk (and compressed usage turned off) then i'll close this issue assuming that it was a problem with your toolchain. Thanks in advance.

(0001644) bernhardf 09-20-06 09:01	ping; Please retry with trunk and let us know what versions of toolchain on what arch you used when this failed. TIA

(0001753) vda 11-21-06 04:02	Hello? Anybody alive here?

Issue History
Date Modified	Username	Field	Change
03-24-06 04:42	ph	New Issue
03-24-06 04:42	ph	Status	new => assigned
03-24-06 04:42	ph	Assigned To	=> BusyBox
03-26-06 22:34	ph	Issue Monitored: ph
06-11-06 10:40	bernhardf	Note Added: 0001414
09-20-06 09:01	bernhardf	Note Added: 0001644
11-21-06 04:02	vda	Note Added: 0001753
11-21-06 04:02	vda	Status	assigned => feedback
11-29-06 06:42	vda	Status	feedback => closed
11-29-06 06:42	vda	Resolution	open => unable to reproduce
11-29-06 06:42	vda	Fixed in Version	=> 1.2.x