* [dpdk-dev] Process for removing __rte_experimental
@ 2018-05-25 9:35 Shreyansh Jain
2018-05-25 9:49 ` Thomas Monjalon
2018-05-25 11:23 ` Neil Horman
0 siblings, 2 replies; 5+ messages in thread
From: Shreyansh Jain @ 2018-05-25 9:35 UTC (permalink / raw)
To: dev; +Cc: Ferruh Yigit, Thomas Monjalon, bluca, nhorman
While reading through [1], I couldn't understand if __rte_experimental
and EXPERIMENTAL tag removed from code and map file, respectively, are
categorized under ABI breakage or not. Thus, whether deprecation for
them should be sent or not in a release cycle for removal in subsequent
release.
Logically, EXPERIMENTAL APIs are not part of stable APIs/ABIs and hence
they don't really break an ABI. Thus, deprecation for them doesn't make
sense. (or, as Luca noted on IRC, removing experimental is like
de-deprecation :) )
On IRC, Luca pointed out that in one of the Tech Board meeting minutes
[2], this was recorded but that too has slight ambiguity to it.
Any thoughts on this? Or, was this documented somewhere other than [1]?
[1] http://dpdk.org/doc/guides/contributing/versioning.html
[2] https://dpdk.org/ml/archives/dev/2017-October/079961.html (Section 2-b)
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] Process for removing __rte_experimental
2018-05-25 9:35 [dpdk-dev] Process for removing __rte_experimental Shreyansh Jain
@ 2018-05-25 9:49 ` Thomas Monjalon
2018-05-25 10:17 ` Luca Boccassi
2018-05-25 11:23 ` Neil Horman
1 sibling, 1 reply; 5+ messages in thread
From: Thomas Monjalon @ 2018-05-25 9:49 UTC (permalink / raw)
To: Shreyansh Jain; +Cc: dev, Ferruh Yigit, bluca, nhorman
25/05/2018 11:35, Shreyansh Jain:
> While reading through [1], I couldn't understand if __rte_experimental
> and EXPERIMENTAL tag removed from code and map file, respectively, are
> categorized under ABI breakage or not. Thus, whether deprecation for
> them should be sent or not in a release cycle for removal in subsequent
> release.
>
> Logically, EXPERIMENTAL APIs are not part of stable APIs/ABIs and hence
> they don't really break an ABI. Thus, deprecation for them doesn't make
> sense. (or, as Luca noted on IRC, removing experimental is like
> de-deprecation :) )
I agree, no need for prior notice, in my opinion.
> On IRC, Luca pointed out that in one of the Tech Board meeting minutes
> [2], this was recorded but that too has slight ambiguity to it.
>
> Any thoughts on this? Or, was this documented somewhere other than [1]?
>
> [1] http://dpdk.org/doc/guides/contributing/versioning.html
> [2] https://dpdk.org/ml/archives/dev/2017-October/079961.html (Section 2-b)
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] Process for removing __rte_experimental
2018-05-25 9:49 ` Thomas Monjalon
@ 2018-05-25 10:17 ` Luca Boccassi
2018-05-25 11:16 ` Shreyansh Jain
0 siblings, 1 reply; 5+ messages in thread
From: Luca Boccassi @ 2018-05-25 10:17 UTC (permalink / raw)
To: Thomas Monjalon, Shreyansh Jain; +Cc: dev, Ferruh Yigit, nhorman
On Fri, 2018-05-25 at 11:49 +0200, Thomas Monjalon wrote:
> 25/05/2018 11:35, Shreyansh Jain:
> > While reading through [1], I couldn't understand if
> > __rte_experimental
> > and EXPERIMENTAL tag removed from code and map file, respectively,
> > are
> > categorized under ABI breakage or not. Thus, whether deprecation
> > for
> > them should be sent or not in a release cycle for removal in
> > subsequent
> > release.
> >
> > Logically, EXPERIMENTAL APIs are not part of stable APIs/ABIs and
> > hence
> > they don't really break an ABI. Thus, deprecation for them doesn't
> > make
> > sense. (or, as Luca noted on IRC, removing experimental is like
> > de-deprecation :) )
>
> I agree, no need for prior notice, in my opinion.
Yeah that makes sense for me as well.
> > On IRC, Luca pointed out that in one of the Tech Board meeting
> > minutes
> > [2], this was recorded but that too has slight ambiguity to it.
> >
> > Any thoughts on this? Or, was this documented somewhere other than
> > [1]?
> >
> > [1] http://dpdk.org/doc/guides/contributing/versioning.html
> > [2] https://dpdk.org/ml/archives/dev/2017-October/079961.html
> > (Section 2-b)
--
Kind regards,
Luca Boccassi
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] Process for removing __rte_experimental
2018-05-25 10:17 ` Luca Boccassi
@ 2018-05-25 11:16 ` Shreyansh Jain
0 siblings, 0 replies; 5+ messages in thread
From: Shreyansh Jain @ 2018-05-25 11:16 UTC (permalink / raw)
To: Luca Boccassi, Thomas Monjalon; +Cc: dev, Ferruh Yigit, nhorman
> -----Original Message-----
> From: Luca Boccassi [mailto:bluca@debian.org]
> Sent: Friday, May 25, 2018 3:47 PM
> To: Thomas Monjalon <thomas@monjalon.net>; Shreyansh Jain
> <shreyansh.jain@nxp.com>
> Cc: dev@dpdk.org; Ferruh Yigit <ferruh.yigit@intel.com>;
> nhorman@tuxdriver.com
> Subject: Re: Process for removing __rte_experimental
>
> On Fri, 2018-05-25 at 11:49 +0200, Thomas Monjalon wrote:
> > 25/05/2018 11:35, Shreyansh Jain:
> > > While reading through [1], I couldn't understand if
> > > __rte_experimental
> > > and EXPERIMENTAL tag removed from code and map file, respectively,
> > > are
> > > categorized under ABI breakage or not. Thus, whether deprecation
> > > for
> > > them should be sent or not in a release cycle for removal in
> > > subsequent
> > > release.
> > >
> > > Logically, EXPERIMENTAL APIs are not part of stable APIs/ABIs and
> > > hence
> > > they don't really break an ABI. Thus, deprecation for them doesn't
> > > make
> > > sense. (or, as Luca noted on IRC, removing experimental is like
> > > de-deprecation :) )
> >
> > I agree, no need for prior notice, in my opinion.
>
> Yeah that makes sense for me as well.
[...]
Ok. I will send across a small documentation edit for this clarification.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] Process for removing __rte_experimental
2018-05-25 9:35 [dpdk-dev] Process for removing __rte_experimental Shreyansh Jain
2018-05-25 9:49 ` Thomas Monjalon
@ 2018-05-25 11:23 ` Neil Horman
1 sibling, 0 replies; 5+ messages in thread
From: Neil Horman @ 2018-05-25 11:23 UTC (permalink / raw)
To: Shreyansh Jain; +Cc: dev, Ferruh Yigit, Thomas Monjalon, bluca
On Fri, May 25, 2018 at 03:05:34PM +0530, Shreyansh Jain wrote:
> While reading through [1], I couldn't understand if __rte_experimental and
> EXPERIMENTAL tag removed from code and map file, respectively, are
> categorized under ABI breakage or not. Thus, whether deprecation for them
> should be sent or not in a release cycle for removal in subsequent release.
>
> Logically, EXPERIMENTAL APIs are not part of stable APIs/ABIs and hence they
> don't really break an ABI. Thus, deprecation for them doesn't make sense.
> (or, as Luca noted on IRC, removing experimental is like de-deprecation :) )
>
> On IRC, Luca pointed out that in one of the Tech Board meeting minutes [2],
> this was recorded but that too has slight ambiguity to it.
>
> Any thoughts on this? Or, was this documented somewhere other than [1]?
>
> [1] http://dpdk.org/doc/guides/contributing/versioning.html
> [2] https://dpdk.org/ml/archives/dev/2017-October/079961.html (Section 2-b)
>
Concur with Thomas and Luca. Use of __rte_experimental denotes APIs that are
not bound by ABI requirements. Removing the experimental tag is,as you say
de-deprication.
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2018-05-25 11:24 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-05-25 9:35 [dpdk-dev] Process for removing __rte_experimental Shreyansh Jain
2018-05-25 9:49 ` Thomas Monjalon
2018-05-25 10:17 ` Luca Boccassi
2018-05-25 11:16 ` Shreyansh Jain
2018-05-25 11:23 ` Neil Horman
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).