Re: [dpdk-dev] [PATCH v2] rcu/test: make gloabl variable per core

2019-06-09 Thread Thomas Monjalon
07/06/2019 15:30, Honnappa Nagarahalli: > > > > Bad start, there is a typo in the title :) > I guess, my apologies do not have any meaning anymore > > > > > 16/05/2019 03:14, Honnappa Nagarahalli: > > > Each hash entry has a pointer to one uint32 memory location. > > > However, all the readers i

Re: [dpdk-dev] [PATCH v2] rcu/test: make gloabl variable per core

2019-06-06 Thread Honnappa Nagarahalli
> > Bad start, there is a typo in the title :) I guess, my apologies do not have any meaning anymore > > 16/05/2019 03:14, Honnappa Nagarahalli: > > Each hash entry has a pointer to one uint32 memory location. > > However, all the readers increment the same location causing race > > conditions.

Re: [dpdk-dev] [PATCH v2] rcu/test: make gloabl variable per core

2019-06-05 Thread Thomas Monjalon
05/06/2019 12:50, Thomas Monjalon: > Bad start, there is a typo in the title :) > > 16/05/2019 03:14, Honnappa Nagarahalli: > > Each hash entry has a pointer to one uint32 memory location. > > However, all the readers increment the same location causing > > race conditions. Allocate memory for eac

Re: [dpdk-dev] [PATCH v2] rcu/test: make gloabl variable per core

2019-06-05 Thread Thomas Monjalon
Bad start, there is a typo in the title :) 16/05/2019 03:14, Honnappa Nagarahalli: > Each hash entry has a pointer to one uint32 memory location. > However, all the readers increment the same location causing > race conditions. Allocate memory for each thread so that each > thread will increment i

[dpdk-dev] [PATCH v2] rcu/test: make gloabl variable per core

2019-05-15 Thread Honnappa Nagarahalli
Each hash entry has a pointer to one uint32 memory location. However, all the readers increment the same location causing race conditions. Allocate memory for each thread so that each thread will increment its own memory location. Fixes: b87089b0bb19 ("test/rcu: add API and functional tests") Rep