On Oct. 2, 2013, 2:15 p.m., Michal Minar wrote:
> src/logicalfile/file.c, lines 144-146
>
<
http://reviewboard-openlmi.rhcloud.com/r/916/diff/1/?file=5058#file5058li...
>
> Does this really solve any issue?
> IMHO CMNewObjectPath should ensure the result is NULL if status != CMPI_RC_OK.
I wrote it like that exactly because of this. I *wasn't sure* what happens if
CMNewObjectPath fails somehow, as it seems to be server dependent. The code is ugly, but
should be fail-proof, unless I got the logic wrong.
- Jan
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://reviewboard-openlmi.rhcloud.com/r/916/#review1279
-----------------------------------------------------------
On Oct. 2, 2013, 1:32 p.m., Jan Synacek wrote:
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://reviewboard-openlmi.rhcloud.com/r/916/
-----------------------------------------------------------
(Updated Oct. 2, 2013, 1:32 p.m.)
Review request for OpenLMI Developers.
Repository: openlmi-providers
Description
-------
logicalfile: fix possible null pointer dereference
Diffs
-----
src/logicalfile/file.c 3e9c834367f76e41588795f8ec1a3f1fa8f657c0
Diff:
http://reviewboard-openlmi.rhcloud.com/r/916/diff/
Testing
-------
Thanks,
Jan Synacek