Re: [dpdk-dev] [PATCH v6 00/10] Register non-EAL threads as lcore

2020-07-08 Thread David Marchand
On Mon, Jul 6, 2020 at 10:53 PM David Marchand wrote: > > OVS and some other applications have been hacking into DPDK internals to > fake EAL threads and avoid performance penalty of only having non-EAL > threads. > > This series proposes to add a new type of lcores and maps those threads > to suc

Re: [dpdk-dev] [PATCH v6 00/10] Register non-EAL threads as lcore

2020-07-08 Thread David Marchand
On Tue, Jul 7, 2020 at 1:23 AM Ananyev, Konstantin wrote: > > OVS and some other applications have been hacking into DPDK internals to > > fake EAL threads and avoid performance penalty of only having non-EAL > > threads. > > > > This series proposes to add a new type of lcores and maps those thre

Re: [dpdk-dev] [PATCH v6 00/10] Register non-EAL threads as lcore

2020-07-06 Thread Ananyev, Konstantin
Hi David, > OVS and some other applications have been hacking into DPDK internals to > fake EAL threads and avoid performance penalty of only having non-EAL > threads. > > This series proposes to add a new type of lcores and maps those threads > to such lcores. > non-EAL threads won't run the

[dpdk-dev] [PATCH v6 00/10] Register non-EAL threads as lcore

2020-07-06 Thread David Marchand
OVS and some other applications have been hacking into DPDK internals to fake EAL threads and avoid performance penalty of only having non-EAL threads. This series proposes to add a new type of lcores and maps those threads to such lcores. non-EAL threads won't run the DPDK eal mainloop. As a cons