[dpdk-stable] [PATCH] test/compress: replace test vector

Trahe, Fiona fiona.trahe at intel.com
Mon Jan 20 16:00:28 CET 2020


Hi Thomas

> -----Original Message-----
> From: Thomas Monjalon <thomas at monjalon.net>
> Sent: Monday, January 20, 2020 2:33 PM
> To: Trahe, Fiona <fiona.trahe at intel.com>; Trybula, ArturX <arturx.trybula at intel.com>; Mcnamara,
> John <john.mcnamara at intel.com>
> Cc: dev at dpdk.org; stable at dpdk.org; shallyv at marvell.com; Dybkowski, AdamX
> <adamx.dybkowski at intel.com>; Danilewicz, MarcinX <marcinx.danilewicz at intel.com>; Akhil Goyal
> <akhil.goyal at nxp.com>
> Subject: Re: [dpdk-stable] [PATCH] test/compress: replace test vector
> 
> 20/01/2020 14:54, Mcnamara, John:
> > From: stable <stable-bounces at dpdk.org> On Behalf Of Thomas Monjalon
> > > 15/01/2020 16:54, Akhil Goyal:
> > > > > >
> > > > > > This patch replaces existing test vector with a new one containing
> > > > > > C code to fix license issue.
> > > > > >
> > > > > > Signed-off-by: Artur Trybula <arturx.trybula at intel.com>
> > > > > Acked-by: Fiona Trahe <fiona.trahe at intel.com>
> > > > Applied to dpdk-next-crypto
> > >
> > > This is a terrible idea.
> > > Adding C code in a C project as text, and worst, its own code, will make
> > > grep matching on fake text.
> >
> > Hi,
> >
> > I don't think this is a valid criticism. The patch replace code in Pascal with some code in C as a
> compression test. The grep part isn't really relevant, it should be clear to anyone who finds a match.
> >
> > If you have a strong objection we will replace it with some more public domain text.
> 
> Yes please, it will be less confusing.
> 
> > > Why do you need so much text to compress?
> 
> Please why do we need some code at all?
[Fiona] compression engines are usually tested with different types of data to represent various use-cases.
So include for example ASCII text, binary chunks, excel files, etc. Compression of code is a typical compression workload - and
as code has lots of patterns may be quite compressible, at different ratios to standard English language text.
That said, it's not so important in DPDK to cover all data types as our focus is not on verifying the compression
engines, but just the path through the API and PMD to them. So we can replace this text if it's less confusing.

> 
> > > Why not just opening our own code files as text?
> >
> > That probably isn't worth the effort.
> >
> > > Why stable at dpdk.org is Cc'ed in this email?
> >
> > Ok. We'll do that.
> 
> I say Why? You say Ok :-)
[Fiona] Simple mistake - sorry 😊
> 
> > > What is the license issue?
> >
> > The issue is that the Pascal code from the Calgary Corpus (a standard body of compression test
> input) doesn't have a license or explicitly say that it is in the public domain so it isn't clear if we can
> relicense  it as BSD-3. We are being overly cautious here but we didn't want to run into any potential
> license issues.
> 
> OK
> Please explain in the commit log.
> 
[Fiona] ok


More information about the stable mailing list