| Message ID | 1330004063.2986.2789.camel@triegel.csb |
|---|---|
| State | New |
| Headers | show |
On 02/23/12 05:34, Torvald Riegel wrote: > libitm: Use ml_wt as default TM methods for >1 thread. > > libitm/ > * retry.cc (GTM::gtm_thread::number_of_threads_changed): Change > default dispatch for more than 1 thread to ml_wt. Ok. r~
diff --git a/libitm/retry.cc b/libitm/retry.cc index 2c1483e..660bf52 100644 --- a/libitm/retry.cc +++ b/libitm/retry.cc @@ -314,7 +314,7 @@ GTM::gtm_thread::number_of_threads_changed(unsigned previous, unsigned now) set_default_dispatch(default_dispatch_user); else { - abi_dispatch* a = dispatch_serialirr_onwrite(); + abi_dispatch* a = dispatch_ml_wt(); if (a->supports(now)) set_default_dispatch(a); else
This patch changes the default TM method that is used if more than one thread is registered to ml_wt (from serialirr_onwrite previously). For one registered thread, it's still serialirr. Right now, this will not be a performance advantage for every workload and number of threads. However, of the TM methods we have implemented so far, ml_wt is the only one that allows concurrent update transactions. gl_wt update txns essentially abort all other txns that have read something. serialirr_onwrite serializes all update txns. ml_wt can execute disjoint-access-parallel txns in parallel, unless we get false sharing via the address-to-orec mapping. Therefore, even though ml_wt performance isn't great currently, it's the only option that doesn't suffer from a principal scalability bottleneck. In the long-term, this should get improved by a less trivial choice regarding which TM method to use (e.g., based on actual abort rate and other factors). Likewise, the ml_wt lock-mapping can hopefully be improved, so that we can at least get better scalability even if the single-thread overheads are higher. We might also want to add other TM methods. But all of that will require some time to get done, so until then, I think the better choice right now is still to go with ml_wt as-is in the meantime. OK? PS: I also noticed that libitm/testsuite/libitm.c/memcpy-1.c takes surprisingly long when using ml_wt compared to serialirr. I will investigate this later. commit ce7f7cd1797cb3c4136c53ff038d2f8f7f0bfad7 Author: Torvald Riegel <triegel@redhat.com> Date: Thu Feb 23 14:16:14 2012 +0100 libitm: Use ml_wt as default TM methods for >1 thread. libitm/ * retry.cc (GTM::gtm_thread::number_of_threads_changed): Change default dispatch for more than 1 thread to ml_wt.