This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
[Bug translator/11266] New: var_expanding_visitor barfs on non-assignment lvalues
- From: "jistone at redhat dot com" <sourceware-bugzilla at sourceware dot org>
- To: systemtap at sources dot redhat dot com
- Date: 10 Feb 2010 00:11:56 -0000
- Subject: [Bug translator/11266] New: var_expanding_visitor barfs on non-assignment lvalues
- Reply-to: sourceware-bugzilla at sourceware dot org
This family of visitors tries to deal with $target variables as lvalues, but
there are more way to be an lvalue than just assignment. "++", "--", and
"delete" also have lvalues, but aren't handled by var_expanding_visitor.
$ stap -e 'probe syscall.read { ++$fd }' -p2 -g
stap: ../tapsets.cxx:2522: virtual void
dwarf_var_expanding_visitor::visit_target_symbol(target_symbol*): Assertion
`!target_symbol_setter_functioncalls.empty()' failed.
Aborted (core dumped)
$ stap -e 'probe syscall.read { delete $fd }' -p2 -g
stap: ../tapsets.cxx:2522: virtual void
dwarf_var_expanding_visitor::visit_target_symbol(target_symbol*): Assertion
`!target_symbol_setter_functioncalls.empty()' failed.
Aborted (core dumped)
These should at least be thrown as semantic errors instead of crashing.
("Cannot increment target variable" / "Cannot delete target variable")
You can also fool it by mixing lvalue types, e.g.
$ stap -e 'probe syscall.read { ++$fd = 42 }' -p2 -g
stap: ../tapsets.cxx:1901: virtual void
var_expanding_visitor::visit_assignment(assignment*): Assertion `new_left ==
fcall' failed.
Aborted (core dumped)
--
Summary: var_expanding_visitor barfs on non-assignment lvalues
Product: systemtap
Version: unspecified
Status: NEW
Severity: normal
Priority: P2
Component: translator
AssignedTo: systemtap at sources dot redhat dot com
ReportedBy: jistone at redhat dot com
http://sourceware.org/bugzilla/show_bug.cgi?id=11266
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.