unit test build failure in phosphor-hwmon

unit test build failure in phosphor-hwmon

[Brad Bishop]

I was trying to get meson going in phosphor-hwmon today and I stumbled on a  
unit test that doesn’t build under gcc 9.1.0.  The failing test is the  
average test.

I just thought I’d quick throw this out there in case anyone feels like  
having a look.  Also, I think this means our repository CI environment has  
gotten out of sync with upstream Yocto?

I realize “patches welcome” applies here for both issues :-) but I just  
wanted to broadcast them.

thx -brad

Re: unit test build failure in phosphor-hwmon

[William Kennington]

Well, the unit test environment is using gcc 9.x.x for what that's worth.

On Mon, Oct 7, 2019 at 1:02 PM Brad Bishop <bradleyb@fuzziesquirrel.com> wrote:
>
> I was trying to get meson going in phosphor-hwmon today and I stumbled on a
> unit test that doesn’t build under gcc 9.1.0.  The failing test is the
> average test.
>
> I just thought I’d quick throw this out there in case anyone feels like
> having a look.  Also, I think this means our repository CI environment has
> gotten out of sync with upstream Yocto?
>
> I realize “patches welcome” applies here for both issues :-) but I just
> wanted to broadcast them.
>
> thx -brad

Re: unit test build failure in phosphor-hwmon

[Kun Yi]

What's the test build error message with meson?

On Mon, Oct 7, 2019 at 1:05 PM William Kennington <wak@google.com> wrote:
>
> Well, the unit test environment is using gcc 9.x.x for what that's worth.
>
> On Mon, Oct 7, 2019 at 1:02 PM Brad Bishop <bradleyb@fuzziesquirrel.com> wrote:
> >
> > I was trying to get meson going in phosphor-hwmon today and I stumbled on a
> > unit test that doesn’t build under gcc 9.1.0.  The failing test is the
> > average test.
> >
> > I just thought I’d quick throw this out there in case anyone feels like
> > having a look.  Also, I think this means our repository CI environment has
> > gotten out of sync with upstream Yocto?
> >
> > I realize “patches welcome” applies here for both issues :-) but I just
> > wanted to broadcast them.
> >
> > thx -brad



-- 
Regards,
Kun

Re: unit test build failure in phosphor-hwmon

[Patrick Venture]

On Mon, Oct 7, 2019 at 1:22 PM Kun Yi <kunyi@google.com> wrote:
>
> What's the test build error message with meson?
>
> On Mon, Oct 7, 2019 at 1:05 PM William Kennington <wak@google.com> wrote:
> >
> > Well, the unit test environment is using gcc 9.x.x for what that's worth.
> >
> > On Mon, Oct 7, 2019 at 1:02 PM Brad Bishop <bradleyb@fuzziesquirrel.com> wrote:
> > >
> > > I was trying to get meson going in phosphor-hwmon today and I stumbled on a
> > > unit test that doesn’t build under gcc 9.1.0.  The failing test is the
> > > average test.
> > >
> > > I just thought I’d quick throw this out there in case anyone feels like
> > > having a look.  Also, I think this means our repository CI environment has
> > > gotten out of sync with upstream Yocto?
> > >
> > > I realize “patches welcome” applies here for both issues :-) but I just
> > > wanted to broadcast them.
> > >
> > > thx -brad
>

I don't see your patches on gerrit.  I don't mind building against
experimental patches :)

>
>
> --
> Regards,
> Kun

Re: unit test build failure in phosphor-hwmon

[Brad Bishop]

at 4:22 PM, Kun Yi <kunyi@google.com> wrote:

> What's the test build error message with meson?

It fails the same way with autotools...

I created a gist with the failure here:

https://gist.github.com/bradbishop/0f18c71f734a8459dcab0ea886528b4f

thanks Kun!

Re: unit test build failure in phosphor-hwmon

[Patrick Venture]

On Mon, Oct 7, 2019 at 1:31 PM Brad Bishop <bradleyb@fuzziesquirrel.com> wrote:
>
> at 4:22 PM, Kun Yi <kunyi@google.com> wrote:
>
> > What's the test build error message with meson?
>
> It fails the same way with autotools...
>
> I created a gist with the failure here:
>
> https://gist.github.com/bradbishop/0f18c71f734a8459dcab0ea886528b4f

Thanks, my CI checkout must be stale.

>
> thanks Kun!

Re: unit test build failure in phosphor-hwmon

[William Kennington]

Looks like someone is comparing longs to uint64_t's or something,
which works on 64 bit arches but not 32-bit ones where unint64_t is
long long int.

On Mon, Oct 7, 2019 at 1:42 PM Patrick Venture <venture@google.com> wrote:
>
> On Mon, Oct 7, 2019 at 1:31 PM Brad Bishop <bradleyb@fuzziesquirrel.com> wrote:
> >
> > at 4:22 PM, Kun Yi <kunyi@google.com> wrote:
> >
> > > What's the test build error message with meson?
> >
> > It fails the same way with autotools...
> >
> > I created a gist with the failure here:
> >
> > https://gist.github.com/bradbishop/0f18c71f734a8459dcab0ea886528b4f
>
> Thanks, my CI checkout must be stale.
>
> >
> > thanks Kun!

Re: unit test build failure in phosphor-hwmon

[Patrick Venture]

On Mon, Oct 7, 2019 at 1:42 PM Patrick Venture <venture@google.com> wrote:
>
> On Mon, Oct 7, 2019 at 1:31 PM Brad Bishop <bradleyb@fuzziesquirrel.com> wrote:
> >
> > at 4:22 PM, Kun Yi <kunyi@google.com> wrote:
> >
> > > What's the test build error message with meson?
> >
> > It fails the same way with autotools...
> >
> > I created a gist with the failure here:
> >
> > https://gist.github.com/bradbishop/0f18c71f734a8459dcab0ea886528b4f
>
> Thanks, my CI checkout must be stale.

Updated my CI checkout and it builds "fine."  It builds without the
comparison failures you're seeing at least.

>
> >
> > thanks Kun!

Re: unit test build failure in phosphor-hwmon

[Brad Bishop]

at 4:29 PM, Patrick Venture <venture@google.com> wrote:

>>> On Mon, Oct 7, 2019 at 1:02 PM Brad Bishop  
>>> <bradleyb@fuzziesquirrel.com> wrote:
>>>> I was trying to get meson going in phosphor-hwmon today and I stumbled  
>>>> on a
>>>> unit test that doesn’t build under gcc 9.1.0.  The failing test is the
>>>> average test.
>>>>
>>>> I just thought I’d quick throw this out there in case anyone feels like
>>>> having a look.  Also, I think this means our repository CI environment  
>>>> has
>>>> gotten out of sync with upstream Yocto?
>>>>
>>>> I realize “patches welcome” applies here for both issues :-) but I just
>>>> wanted to broadcast them.
>>>>
>>>> thx -brad
>
> I don't see your patches on gerrit.  I don't mind building against
> experimental patches :)

Right…I don’t have any patches for either of these problems.  I did just  
put my patches that add meson support out there; however:

https://gerrit.openbmc-project.xyz/25877

thanks!

Re: unit test build failure in phosphor-hwmon

[Brad Bishop]

at 4:43 PM, William Kennington <wak@google.com> wrote:

> Looks like someone is comparing longs to uint64_t's or something,
> which works on 64 bit arches but not 32-bit ones where unint64_t is
> long long int.

This was most definitely the problem:

https://gerrit.openbmc-project.xyz/c/openbmc/phosphor-hwmon/+/25879/1

Thanks William!

-brad