Bug 5423 - stap segfaults when running specific script
Summary: stap segfaults when running specific script
Status: RESOLVED FIXED
Alias: None
Product: systemtap
Classification: Unclassified
Component: translator (show other bugs)
Version: unspecified
: P2 normal
Target Milestone: ---
Assignee: Unassigned
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2007-11-28 02:15 UTC by Anderson Lizardo
Modified: 2007-11-28 02:31 UTC (History)
0 users

See Also:
Host:
Target:
Build:
Last reconfirmed:


Attachments
stap segfault testcase script (134 bytes, text/plain)
2007-11-28 02:16 UTC, Anderson Lizardo
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Anderson Lizardo 2007-11-28 02:15:53 UTC
The attached stap script causes stap to segfault when running with:

# stap -p2 -g test.stp

version info:
- systemtap-snapshot-20071124
- elfutils-0.131

Note: the segfault could only be reproduced with the exact script attached,
including comments and whitespaces.
Comment 1 Anderson Lizardo 2007-11-28 02:16:48 UTC
Created attachment 2113 [details]
stap segfault testcase script
Comment 2 Anderson Lizardo 2007-11-28 02:19:17 UTC
Forgot to attach the gdb backtrace:

#0  require<expression*> (v=0xbfd61978, dst=0xbfd619b0, src=0x61767465) at
staptree.h:845
#1  0x0806cf68 in deep_copy_visitor::deep_copy (s=0x61767465) at staptree.cxx:2389
#2  0x08075c32 in derived_probe (this=0x85827c0, p=0x8119c18, l=0x8119b90) at
elaborate.cxx:56
#3  0x080d18fd in be_builder::build (this=0x8535860, base=0x8119c18,
location=0x8119b90, parameters=@0xbfd61b80, finished_results=@0xbfd61fdc) at
tapsets.cxx:85
#4  0x08076612 in match_node::find_and_build (this=0x8535888, s=@0xbfd62114,
p=0x8119c18, loc=0x8119b90, pos=1, results=@0xbfd61fdc) at elaborate.cxx:343
#5  0x08076663 in match_node::find_and_build (this=0x8117168, s=@0xbfd62114,
p=0x8119c18, loc=0x8119b90, pos=0, results=@0xbfd61fdc) at elaborate.cxx:402
#6  0x0807a6f8 in derive_probes (s=@0xbfd62114, p=0x8119c18, dps=@0xbfd61fdc,
optional=<value optimized out>) at elaborate.cxx:593
#7  0x0807af6b in semantic_pass_symbols (s=@0xbfd62114) at elaborate.cxx:1000
#8  0x0807efd4 in semantic_pass (s=@0xbfd62114) at elaborate.cxx:1041
#9  0x08051a9e in main (argc=4, argv=0xbfd629d4) at main.cxx:690
Comment 3 Frank Ch. Eigler 2007-11-28 02:31:31 UTC
Thanks for the report; this bug was fixed yesterday in CVS.