----- Taras Glek <[email protected]> wrote: > The two errors I looked at, seem to be false positives
Hi, yeah, the great majority of these reports are going to be false positives (I hope that was clear from the talk!). The trick here is that it should be clear from the report where the tool got confused, and it should be easy to add an annotation which the tool can use to check the assertion. For both of these reports, the tool picked the wrong postcondition for a called function, and in both the right postcondition is easy to specify. > http://sixgill.org/firefox/write_overflow/ResolveSymlink_1244447419593944471.html The first step the tool took for this access was pick the postcondition (__return < 4097) for PR_Read: PR_IMPLEMENT(PRInt32) PR_Read(PRFileDesc *fd, void *buf, PRInt32 amount); > > http://sixgill.org/firefox/write_overflow/EncodeString_3315072988729628109.html > > > > Taras _______________________________________________ dev-static-analysis mailing list [email protected] https://lists.mozilla.org/listinfo/dev-static-analysis
