[lock-free] Re: Scalable hash map

2018-12-28 Thread manuel
Hi, sorry for resurrecting this old thread! The link to the files (http://groups.google.com/group/lock-free/files) no longer works. I've downloaded the hash_map.zip file from your homepage, but this version of the code only supports pod-types of 4 or 8 bytes. I am particularly interested in yo

Re: [lock-free] Re: Scalable hash map

2018-12-28 Thread Dmitry Vyukov
On Fri, Dec 28, 2018 at 1:22 PM wrote: > > Hi, > > sorry for resurrecting this old thread! > > The link to the files (http://groups.google.com/group/lock-free/files) no > longer works. > I've downloaded the hash_map.zip file from your homepage, but this version of > the code only supports pod-ty

Re: [lock-free] Re: Scalable hash map

2018-12-28 Thread Manuel Pöter
Hi Dmitry, in one of your last post you wrote: I've uploaded modified hash map which supports strings (and any other > satellite data) as keys/values: > http://groups.google.com/group/lock-free/files > There is example included which uses strings as both keys and values. > That made me think

Re: [lock-free] Re: Scalable hash map

2018-12-29 Thread Manuel Pöter
BTW - I think a found a race condition that causes find to incorrectly return false. Consider the following scenario: Thread A removes an entry with key k1 which is stored directly in the cell. The cell's head points to an add_item with key k2, so thread A first updates the cell's state to reflec