Re: [dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core functions non-inline

2019-07-30 Thread Stephen Hemminger
On Tue, 30 Jul 2019 17:11:31 +0100 Bruce Richardson wrote: > On Tue, Jul 30, 2019 at 08:54:13AM -0700, Stephen Hemminger wrote: > > On Tue, 30 Jul 2019 16:33:55 +0100 > > Bruce Richardson wrote: > > > > > On Tue, Jul 30, 2019 at 08:25:34AM -0700, Stephen Hemminger wrote: > > > > On Tue, 30

Re: [dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core functions non-inline

2019-07-30 Thread Bruce Richardson
On Tue, Jul 30, 2019 at 08:54:13AM -0700, Stephen Hemminger wrote: > On Tue, 30 Jul 2019 16:33:55 +0100 > Bruce Richardson wrote: > > > On Tue, Jul 30, 2019 at 08:25:34AM -0700, Stephen Hemminger wrote: > > > On Tue, 30 Jul 2019 14:49:49 +0200 > > > Marcin Zapolski wrote: > > > > > > > Make r

Re: [dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core functions non-inline

2019-07-30 Thread Bruce Richardson
Original Message- From: dev On > > > > Behalf Of Marcin Zapolski Sent: Tuesday, July 30, 2019 6:20 PM To: > > > > dev@dpdk.org Cc: Marcin Zapolski > > > > Subject: [dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core functions > > > > non- in

Re: [dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core functions non-inline

2019-07-30 Thread Wiles, Keith
> On Jul 30, 2019, at 10:54 AM, Stephen Hemminger > wrote: > > On Tue, 30 Jul 2019 16:33:55 +0100 > Bruce Richardson wrote: > >> On Tue, Jul 30, 2019 at 08:25:34AM -0700, Stephen Hemminger wrote: >>> On Tue, 30 Jul 2019 14:49:49 +0200 >>> Marcin Zapolski wrote: >>> Make rte_eth_rx_bu

Re: [dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core functions non-inline

2019-07-30 Thread Stephen Hemminger
On Tue, 30 Jul 2019 16:33:55 +0100 Bruce Richardson wrote: > On Tue, Jul 30, 2019 at 08:25:34AM -0700, Stephen Hemminger wrote: > > On Tue, 30 Jul 2019 14:49:49 +0200 > > Marcin Zapolski wrote: > > > > > Make rte_eth_rx_burst, rte_eth_tx_burst and other static inline ethdev > > > functions no

Re: [dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core functions non-inline

2019-07-30 Thread Jerin Jacob Kollanukkaran
> -Original Message- > From: Bruce Richardson > Sent: Tuesday, July 30, 2019 9:02 PM > To: Jerin Jacob Kollanukkaran > Cc: Marcin Zapolski ; dev@dpdk.org > Subject: [EXT] Re: [dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core func

Re: [dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core functions non-inline

2019-07-30 Thread Bruce Richardson
On Tue, Jul 30, 2019 at 08:25:34AM -0700, Stephen Hemminger wrote: > On Tue, 30 Jul 2019 14:49:49 +0200 > Marcin Zapolski wrote: > > > Make rte_eth_rx_burst, rte_eth_tx_burst and other static inline ethdev > > functions not inline. They are referencing DPDK internal structures and > > inlining fo

Re: [dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core functions non-inline

2019-07-30 Thread Bruce Richardson
On Tue, Jul 30, 2019 at 03:01:00PM +, Jerin Jacob Kollanukkaran wrote: > > -Original Message- From: dev On Behalf > > Of Marcin Zapolski Sent: Tuesday, July 30, 2019 6:20 PM To: > > dev@dpdk.org Cc: Marcin Zapolski > > Subject: [dpdk-dev] [RFC 19.11 1/2

Re: [dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core functions non-inline

2019-07-30 Thread Stephen Hemminger
On Tue, 30 Jul 2019 14:49:49 +0200 Marcin Zapolski wrote: > Make rte_eth_rx_burst, rte_eth_tx_burst and other static inline ethdev > functions not inline. They are referencing DPDK internal structures and > inlining forces those structures to be exposed to user applications. > > In internal test

Re: [dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core functions non-inline

2019-07-30 Thread Jerin Jacob Kollanukkaran
> -Original Message- > From: dev On Behalf Of Marcin Zapolski > Sent: Tuesday, July 30, 2019 6:20 PM > To: dev@dpdk.org > Cc: Marcin Zapolski > Subject: [dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core functions non- > inline > > Make rte_eth_rx_burst, rte_et

[dpdk-dev] [RFC 19.11 1/2] ethdev: make DPDK core functions non-inline

2019-07-30 Thread Marcin Zapolski
Make rte_eth_rx_burst, rte_eth_tx_burst and other static inline ethdev functions not inline. They are referencing DPDK internal structures and inlining forces those structures to be exposed to user applications. In internal testing with i40e NICs a performance drop of about 2% was observed with te