Thanks for that. I totally appreciate the fact that this is not a quick ticket and that it involves understanding of locking synchronizations (both are the reasons why I picked this ticket). I also poked about function calls a lot hence have specific questions around the code.
Specifically: My understanding now says that the main call that actually handles putAll operation is IgniteTxLocalAdapter.putAllAsync0 (coming down from GridCacheAdapter). Is that correct? In IgniteTxLocalAdapter.putAllAsync(), if I understand correctly, txLockAsync() is the call that actually locks the entries, so logic around this method needs to be modified for ATOMIC cache case,right? On Fri, Jul 3, 2015 at 8:25 PM, Dmitriy Setrakyan <dsetrak...@apache.org> wrote: > On Fri, Jul 3, 2015 at 5:39 AM, Atri Sharma <atri.j...@gmail.com> wrote: > > > Folks, > > > > Can someone please explain a bit around GridCacheAdapter#putAll's logic > > please? I am trying to understand which part works with locking. > > > > Atri, I think the best way to understand how this logic works is to set a > breakpoint there and trace it all the way through. This way you will get an > idea of which pieces are involved. > > Having said that, IGNITE-514 is a very delicate ticket and requires deep > understanding of how locking and synchronization work for atomic operations > in Ignite. If you want to tackle this ticket, you will need to spend enough > time tracing through the execution path and making sure you understand all > the intricate details. This is not a quick ticket. > > > > Regards, > > > > Atri > > > > -- > > Regards, > > > > Atri > > *l'apprenant* > > > -- Regards, Atri *l'apprenant*
