Re: [dpdk-dev] [PATCH v1 1/3] net/i40e: relax barrier in the Tx fastpath of vPMD

2020-03-06 Thread Jerin Jacob
On Fri, Mar 6, 2020 at 10:35 AM Gavin Hu wrote: > > To keep ordering of mixed accesses, rte_cio is sufficient. > The rte_io barrier inside the I40E_PCI_REG_WRITE is overkill.[1] > > This patch fixes by replacing with just sufficient barriers in the > normal PMD and vPMD. > > It showed 7% performan

[dpdk-dev] [PATCH v1 1/3] net/i40e: relax barrier in the Tx fastpath of vPMD

2020-03-05 Thread Gavin Hu
To keep ordering of mixed accesses, rte_cio is sufficient. The rte_io barrier inside the I40E_PCI_REG_WRITE is overkill.[1] This patch fixes by replacing with just sufficient barriers in the normal PMD and vPMD. It showed 7% performance uplift on ThunderX2 and 4% on Arm N1SDP. The test case is th