Set of 26 potential bugs

[Maxime Villard]

Hi,
I've improved my code scanner, and it has found 26 potential bugs under sys/.
Instead of flooding tech-kern@/tech-net@/etc with questions and fixes, I put
here a comprehensive list:

http://m00nbsd.net/ae123a9bae03f7dde5c6d654412daf5a.html#Report-2

Some of them will probably need to be pulled up, like the leak in netinet6
(#03-0x02) which seems to be triggerable from root.

Maxime

[Rich Neswold]

Post by Maxime Villard
I've improved my code scanner, and it has found 26 potential bugs under sys/.
http://m00nbsd.net/ae123a9bae03f7dde5c6d654412daf5a.html#Report-2

Is (#02-0x03) really a problem? The code looks correct to me.

Near the end of this same file, however, I see:

rl->cprl_count++;

followed a few lines later by:

if (rl == NULL) {
sc->error = ENOMEM;
return;
}

which looks like a problem.

[Maxime Villard]

Post by Maxime Villard
I've improved my code scanner, and it has found 26 potential bugs under sys/.
http://m00nbsd.net/ae123a9bae03f7dde5c6d654412daf5a.html#Report-2
Is (#02-0x03) really a problem? The code looks correct to me.
rl->cprl_count++;
if (rl == NULL) {
sc->error = ENOMEM;
return;
}
which looks like a problem.
--
Rich

Hum hum, you're right. After debugging a bit, it seems that it took the wrong
line, and indeed l.363 is the problem.