mbox series

[for-4.18,0/2] golang: Binding fixes

Message ID 20231103194551.64448-1-jandryuk@gmail.com (mailing list archive)
Headers show
Series golang: Binding fixes | expand

Message

Jason Andryuk Nov. 3, 2023, 7:45 p.m. UTC
These two patches refresh the bindings which have gone a little stale.
I used two separate patches since the XSA-443 one may want backporting.

Jason Andryuk (2):
  golang: Fixup binding for Arm FF-A
  golang: Fix bindings after XSA-443

 tools/golang/xenlight/helpers.gen.go | 9 +++++++++
 tools/golang/xenlight/types.gen.go   | 3 +++
 2 files changed, 12 insertions(+)

Comments

Andrew Cooper Nov. 3, 2023, 7:50 p.m. UTC | #1
On 03/11/2023 7:45 pm, Jason Andryuk wrote:
> These two patches refresh the bindings which have gone a little stale.
> I used two separate patches since the XSA-443 one may want backporting.
>
> Jason Andryuk (2):
>   golang: Fixup binding for Arm FF-A
>   golang: Fix bindings after XSA-443

Bah - the security team did mess up there.

Also, we very clearly need the Golang bindings to be a blocking build
test in GitlabCI to avoid repeats of this in the future.

I'm tempted to the Gitlab change ought to be for-4.18 too.  Henry, thoughts?

~Andrew
Henry Wang Nov. 6, 2023, 5:38 a.m. UTC | #2
Hi Andrew,

> On Nov 4, 2023, at 03:50, Andrew Cooper <andcooper@tibco.com> wrote:
> 
> [andcooper@tibco.com appears similar to someone who previously sent you email, but may not be that person. Learn why this could be a risk at https://aka.ms/LearnAboutSenderIdentification ]

(+Your work email address since this reply somehow ended up with my filter,
I found this email a bit later, sorry for the late reply.)

> 
> On 03/11/2023 7:45 pm, Jason Andryuk wrote:
>> These two patches refresh the bindings which have gone a little stale.
>> I used two separate patches since the XSA-443 one may want backporting.
>> 
>> Jason Andryuk (2):
>>  golang: Fixup binding for Arm FF-A
>>  golang: Fix bindings after XSA-443
> 
> Bah - the security team did mess up there.
> 
> Also, we very clearly need the Golang bindings to be a blocking build
> test in GitlabCI to avoid repeats of this in the future.
> 
> I'm tempted to the Gitlab change ought to be for-4.18 too.  Henry, thoughts?

As long as the GitLab CI change can be submitted within the 4.18 release
timeline, I won’t object it. Thanks.

Kind regards,
Henry

> 
> ~Andrew
George Dunlap Nov. 6, 2023, 11:52 a.m. UTC | #3
On Fri, Nov 3, 2023 at 7:46 PM Jason Andryuk <jandryuk@gmail.com> wrote:
>
> These two patches refresh the bindings which have gone a little stale.
> I used two separate patches since the XSA-443 one may want backporting.

Thanks for these.  Out of curiosity, do you use the golang bindings,
and if so, are you happy with how you're consuming them now?  We had
talked at some point about exposing the bindings somewhere on xenbits
(or maybe gitlab) such that people should simply add the path to their
golang projects and have it download & build like "normal".  Is that
something you'd find useful?

Peace,
 -George
Roger Pau Monné Nov. 6, 2023, 12:45 p.m. UTC | #4
On Fri, Nov 03, 2023 at 03:45:49PM -0400, Jason Andryuk wrote:
> These two patches refresh the bindings which have gone a little stale.
> I used two separate patches since the XSA-443 one may want backporting.

The XSA-443 one is only relevant for unstable and 4.18.  Stable
branches at the time of the XSA didn't get those fields added as it
would break the API.

Regards, Roger.
Jason Andryuk Nov. 6, 2023, 1:04 p.m. UTC | #5
On Mon, Nov 6, 2023 at 6:52 AM George Dunlap <george.dunlap@cloud.com> wrote:
>
> On Fri, Nov 3, 2023 at 7:46 PM Jason Andryuk <jandryuk@gmail.com> wrote:
> >
> > These two patches refresh the bindings which have gone a little stale.
> > I used two separate patches since the XSA-443 one may want backporting.
>
> Thanks for these.  Out of curiosity, do you use the golang bindings,
> and if so, are you happy with how you're consuming them now?  We had
> talked at some point about exposing the bindings somewhere on xenbits
> (or maybe gitlab) such that people should simply add the path to their
> golang projects and have it download & build like "normal".  Is that
> something you'd find useful?

Hi George,

No, I do not use the bindings, so I cannot answer your questions.  I
just noticed the bindings were stale when doing a rebase.

I think Andrew's idea to check these via gitlab is a good one.  They
seem to often be missed.  I only noticed because I had installed go,
and I only installed go to check binding generation for a previous
patch series.

Since the bindings are auto generated, do they really need to be
checked into the tree?  If gitlab checks their generation, then maybe
they don't need to be checked in?

Regards,
Jason
Jason Andryuk Nov. 6, 2023, 1:06 p.m. UTC | #6
On Mon, Nov 6, 2023 at 7:45 AM Roger Pau Monné <roger.pau@citrix.com> wrote:
>
> On Fri, Nov 03, 2023 at 03:45:49PM -0400, Jason Andryuk wrote:
> > These two patches refresh the bindings which have gone a little stale.
> > I used two separate patches since the XSA-443 one may want backporting.
>
> The XSA-443 one is only relevant for unstable and 4.18.  Stable
> branches at the time of the XSA didn't get those fields added as it
> would break the API.

Oh, good.  Thanks for confirming, Roger.

Regards,
Jason
Andrew Cooper Nov. 6, 2023, 1:52 p.m. UTC | #7
On 06/11/2023 5:38 am, Henry Wang wrote:
> Hi Andrew,
>
>> On Nov 4, 2023, at 03:50, Andrew Cooper <andcooper@tibco.com> wrote:
>>
>> [andcooper@tibco.com appears similar to someone who previously sent you email, but may not be that person. Learn why this could be a risk at https://aka.ms/LearnAboutSenderIdentification ]
> (+Your work email address since this reply somehow ended up with my filter,
> I found this email a bit later, sorry for the late reply.)

sigh.

It seems a Thunderbird update has clobbered my configuration, and for
reasons of the corporate merger, @tibco is still our probably accounts.

I'll see about fixing up.

~Andrew