sac2c issueshttps://gitlab.sac-home.org/sac-group/sac2c/-/issues2017-11-19T21:36:32Zhttps://gitlab.sac-home.org/sac-group/sac2c/-/issues/1888CHK of FUNDEF_RETURN code faulty2017-11-19T21:36:32ZRobert BerneckyCHK of FUNDEF_RETURN code faulty| | |
| --- | --- |
| Bugzilla Link | [1018](http://bugs.sac-home.org/show_bug.cgi?id=1018) |
| Created on | Aug 29, 2012 16:47 |
| Resolution | FIXED |
| Resolved on | Aug 31, 2012 22:59 |
| Version | svn |
| OS | Linux |
| Architectu...| | |
| --- | --- |
| Bugzilla Link | [1018](http://bugs.sac-home.org/show_bug.cgi?id=1018) |
| Created on | Aug 29, 2012 16:47 |
| Resolution | FIXED |
| Resolved on | Aug 31, 2012 22:59 |
| Version | svn |
| OS | Linux |
| Architecture | PC |
## Extended Description
<pre>I recently introduced CHK code to vet FUNDEF_RETURN, but it has a flaw:
* 17: Converting from static single assignment form ...
**** Converting from SSA form ...
-> Running syntax tree checks
**** Reintroducing loops and conditionals ...
-> Running syntax tree checks
TRAVERSE ERROR: node of type 59810128:!invalid! found where 26:N_ap was expected!
The problem here is that FUNDEF_RETURN is no longer valid
once we no longer have loops as LACFUNS.
Similarly, it should not be valid before we have LACFUNS.
I am going to update ast.xml to reflect this situation and
will amend check_lib.c as well.
~/sac/testsuite/optimizations/constantfolding$ sac2c -V
sac2c v1.00-beta (Haggis And Apple)
developer rev 18189:MODIFIED linux-gnu_x86_64
(Wed Aug 29 11:33:27 EDT 2012 by sac)
sac2c bug897.sac -d treecheck -chkfreq 4 -v4
It is too bad we do not have any automated mechanism for vetting
AST attributes...</pre>BugZillaBugZillahttps://gitlab.sac-home.org/sac-group/sac2c/-/issues/1827-d treecheck fails to detect multiple pointers to same son2017-11-19T21:30:06ZRobert Bernecky-d treecheck fails to detect multiple pointers to same son| | |
| --- | --- |
| Bugzilla Link | [713](http://bugs.sac-home.org/show_bug.cgi?id=713) |
| Created on | May 24, 2010 13:48 |
| Resolution | FIXED |
| Resolved on | Jun 08, 2010 15:10 |
| Version | svn |
| OS | Linux |
| Architecture...| | |
| --- | --- |
| Bugzilla Link | [713](http://bugs.sac-home.org/show_bug.cgi?id=713) |
| Created on | May 24, 2010 13:48 |
| Resolution | FIXED |
| Resolved on | Jun 08, 2010 15:10 |
| Version | svn |
| OS | Linux |
| Architecture | PC |
## Extended Description
<pre>I have been scratching my head, and possibly other places, for
several days chasing a crash in FREElet(), which turned out to
be, I believe, a failure of -d treecheck to detect the case
when more than one node points to the same son node.
In my case on my local code, I ran into the problem when generator nodes
became N_array nodes. Although the generated IL LOOKS ok,
it is definitely damaged, because when either node tree
is freed, it takes the (illegally shared) son with it.
I think a fix might work along these lines, in check.c and
friends:
- introduce a HasParent flag in the common part of each node.
I don't know, offhand, where that part is defined, but I
suppose a grep of NODE_TYPE would be a good start.
- anonymous traversal over ALL nodes, at the front end of CHK,
that does:
HasParent(node) = FALSE;
- the regular CHK traversal following would then do:
complain if HasParent(son).
HasParent( son) = TRUE;
How do I code an anonymous traversal over all nodes?
I thought, based on conversations in the dim past, that this was supposed
to work already, but apparently, that is not the case.</pre>BugZillaBugZillahttps://gitlab.sac-home.org/sac-group/sac2c/-/issues/1822-d treecheck fails to detect bad AVIS_SSAASSIGN2017-11-19T21:29:25ZRobert Bernecky-d treecheck fails to detect bad AVIS_SSAASSIGN| | |
| --- | --- |
| Bugzilla Link | [702](http://bugs.sac-home.org/show_bug.cgi?id=702) |
| Created on | Apr 29, 2010 22:30 |
| Resolution | INVALID |
| Resolved on | May 01, 2010 20:12 |
| Version | svn |
| OS | Linux |
| Architectu...| | |
| --- | --- |
| Bugzilla Link | [702](http://bugs.sac-home.org/show_bug.cgi?id=702) |
| Created on | Apr 29, 2010 22:30 |
| Resolution | INVALID |
| Resolved on | May 01, 2010 20:12 |
| Version | svn |
| OS | Linux |
| Architecture | PC |
## Extended Description
<pre>I had a problem in PM with SSAASSIGN chain corruption, so
ran with -d treecheck to isolate the fault. No problems reported,
yet the crash still happens.
Then, I manually entered CHKdoTreeCheck(arg_node) calls
at IVEXIfundef entry and exit: Bingo! The bad SSAASSIGN
nodes were detected properly.
This is in my local code, so no idea if this hits the
live system:
rev 16795:16812:MODIFIED
I got it to fail with sac/testsuite/optimizations/awlf/duplhs.sac.</pre>BugZillaBugZillahttps://gitlab.sac-home.org/sac-group/sac2c/-/issues/1821format string attack on dbug.c2017-11-19T21:29:18ZClemens Grelckformat string attack on dbug.c| | |
| --- | --- |
| Bugzilla Link | [694](http://bugs.sac-home.org/show_bug.cgi?id=694) |
| Created on | Apr 07, 2010 21:33 |
| Resolution | FIXED |
| Resolved on | Apr 09, 2010 13:56 |
| Version | svn |
| OS | Linux |
| Architecture...| | |
| --- | --- |
| Bugzilla Link | [694](http://bugs.sac-home.org/show_bug.cgi?id=694) |
| Created on | Apr 07, 2010 21:33 |
| Resolution | FIXED |
| Resolved on | Apr 09, 2010 13:56 |
| Version | svn |
| OS | Linux |
| Architecture | PC |
## Extended Description
<pre>Some of my students from the compiler course found a flaw in the current
implementation of the dbug package (formerly known as Fred Fish).
http://en.wikipedia.org/wiki/Format_string_attack
In essence, function arguments are directly used as printf format strings,
which is a security hole and at least bad programming practice.</pre>BugZillaBugZilla