> -----Original Message----- > From: Jerin Jacob [mailto:jerin.jacob at caviumnetworks.com] > Sent: Tuesday, February 16, 2016 5:44 PM > To: Kobylinski, MichalX <michalx.kobylinski at intel.com> > Cc: dev at dpdk.org; viktorin at rehivetech.com > Subject: Re: [dpdk-dev] [PATCH v4 0/3] add lpm support for NEON > Importance: High > > On Tue, Feb 16, 2016 at 01:27:02PM +0000, Kobylinski, MichalX wrote: > > > > > > > -----Original Message----- > > > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Jerin Jacob > > > Sent: Friday, February 12, 2016 1:29 PM > > > To: dev at dpdk.org > > > Cc: viktorin at rehivetech.com > > > Subject: [dpdk-dev] [PATCH v4 0/3] add lpm support for NEON > > > > [snip] > > > > > > Jerin Jacob (3): > > > lpm: make rte_lpm_lookupx4 API definition architecture agnostic > > > lpm: add support for NEON > > > maintainers: claim responsibility for arm64 specific files of hash and > > > lpm > > > > > > MAINTAINERS | 3 + > > > app/test/test_lpm.c | 21 ++-- > > > app/test/test_xmmt_ops.h | 67 +++++++++++++ > > > config/defconfig_arm-armv7a-linuxapp-gcc | 3 - > > > config/defconfig_arm64-armv8a-linuxapp-gcc | 3 - > > > lib/librte_lpm/Makefile | 6 ++ > > > lib/librte_lpm/rte_lpm.h | 99 ++----------------- > > > lib/librte_lpm/rte_lpm_neon.h | 148 > +++++++++++++++++++++++++++++ > > > lib/librte_lpm/rte_lpm_sse.h | 143 > ++++++++++++++++++++++++++++ > > > 9 files changed, 386 insertions(+), 107 deletions(-) create mode > > > 100644 app/test/test_xmmt_ops.h create mode 100644 > > > lib/librte_lpm/rte_lpm_neon.h create mode 100644 > > > lib/librte_lpm/rte_lpm_sse.h > > > > > > -- > > > 2.1.0 > > > > Hi Jerin, > > Hi Michal, > > > Are you planning increase next_hop field for ARM? I extended next_hop field > from 8 bits to 24 bits and created structure to configure LPM for x86. > > Yes, I am planning to increase next_hop field for ARM as a separate patch. > Let > this base patchset get merges. > > I will make ARM specific changes for your new feature in 'rte_lpm_lookupx4' > as > a separate patch on top of your series. > So that in case if I want to go back to 8 bit then I can do it > > Jerin
Thank you for your answer. Do you prepare separate patch with changes for ARM architecture on the top my series? If you want I can support you with prepare new patch. Michal > > > Please look at my patchset with proposal increase next_hop field and > > structure > to configure. > > > > http://patchwork.dpdk.org/dev/patchwork/patch/10249/ > > http://patchwork.dpdk.org/dev/patchwork/patch/10250/ > > > > Best Regards, > > Michal > >
