Re: HashDoS

On 22 Sep 2016 00:43, "Stanislav Malyshev" <smalyshev@gmail.com> wrote: > > Hi! > > > As linked in my first reply, there was a previous discussion on the > > topic: http://markmail.org/message/ttbgcvdu4f7uymfb > > The collision-counting approach that Yasuo references is linked at the > > start of the first mail: > > https://github.com/php/php-src/pull/1565 > > > > Collisions are counted during insertion operations. While we perform a > > hashtable insertion, we first need to iterate over the existing buckets > > for a certain hash to see if the insert is actually an update. The > > implementation simply keeps track of how many buckets we inspect while > > doing that. If the number is above a certain threshold, an error is > > generated. > > So, count is per-lookup, then I think it's fine then. Sorry for > misunderstanding it initially. Anything prevents us from merging it? If > not, let's do it. > I don't like the initial version of the patch that was causing fatal error for json_decode. That's not how json_decode should work. I think that Bob came up later with a better version that was using json recursion error. It might require a bit more work for 7.1 as I changed a json parser since then. Cheers Jakub