Re: patch saves some cycles by extending pfr_walktree() a bit

2015-11-09 Thread Alexandr Nedvedicky
Hello, > > For now the code gets more as we have two ways to iterate over the > tree. When you remove the additional work queues I expect many - > diffs. So if this code dupliation is temporary, this aproach is > fine for me. yes code duplication is temporary, I'd like to kill work queues. >

Re: patch saves some cycles by extending pfr_walktree() a bit

2015-11-07 Thread Alexander Bluhm
On Wed, Oct 28, 2015 at 06:08:00PM +0100, Alexandr Nedvedicky wrote: > make implementation of reference handling and further MP stuff bit easier. I agree that filling the temporary list makes transition to MP harder. > Also the patch is part of my effort to kill work queues in radix tables. For

patch saves some cycles by extending pfr_walktree() a bit

2015-10-28 Thread Alexandr Nedvedicky
Hello, This is yet another patch, which 'scratches surface', this time in pf_table.c. As briefly discussed in Varazdin the plan is to clean up pf_table.c a bit, to make implementation of reference handling and further MP stuff bit easier. I've noticed sub-optimal implementation table entries at