[dpdk-dev] [PATCH v7 4/5] hash: add lock-free read-write concurrency

[Jerin Jacob]

-----Original Message-----
> Date: Tue, 06 Nov 2018 10:13:53 +0100
> From: Thomas Monjalon <thomas at monjalon.net>
> To: Jerin Jacob <jerin.jacob at caviumnetworks.com>
> Cc: Honnappa Nagarahalli <Honnappa.Nagarahalli at arm.com>,
>  "bruce.richardson at intel.com" <bruce.richardson at intel.com>,
>  "pablo.de.lara.guarch at intel.com" <pablo.de.lara.guarch at intel.com>,
>  "dev at dpdk.org" <dev at dpdk.org>, "yipeng1.wang at intel.com"
>  <yipeng1.wang at intel.com>, Dharmik Thakkar <Dharmik.Thakkar at arm.com>,
>  "Gavin Hu (Arm Technology China)" <Gavin.Hu at arm.com>, nd <nd at arm.com>,
>  "ferruh.yigit at intel.com" <ferruh.yigit at intel.com>,
>  "hemant.agrawal at nxp.com" <hemant.agrawal at nxp.com>,
>  "chaozhu at linux.vnet.ibm.com" <chaozhu at linux.vnet.ibm.com>, "Kapoor,
>  Prasun" <Prasun.Kapoor at cavium.com>
> Subject: Re: [dpdk-dev] [PATCH v7 4/5] hash: add lock-free read-write
>  concurrency
> 
> 06/11/2018 10:10, Jerin Jacob:
> > From: Honnappa Nagarahalli <Honnappa.Nagarahalli at arm.com>
> > > Jerin, thank you for running this test and all the analysis. I have not run this test. I was focused on simultaneous reads and writes. You can look at file test_hash_readwrite_lf.c to look for the kind of the use cases.
> > >
> > > I am trying to reproduce this, I will get back with more details soon.
> >
> > OK. Since RC2 approaching, bit worried about next steps if we cannot
> > avoid the regression with this new feature for this release.
> 
> 18.11-rc2 is already out.
> 
> > 24% drop of performance regression not seen in past for a specific
> > usecase.So not sure what we did past for similar situations.
> >
> > Thomas,
> > Any thought of this?
> 
> Everything is possible, you just need to agree on a patch.
> Please let's fix (or disable or revert) it in -rc3 this week.

I let Honnappa to decide fix, disable or revert action for rc3 release.
But it should either one of them for rc3 not this patch as is.

> 
>