[Pkg-samba-maint] Got one: "smbc_off_t_should_be_at_least_64bits_use_LFS_CFLAGS"

[Michael Tokarev]

30.05.2022 22:50, Andreas Hasenack wrote:
> Hi,

Hello!

> I got a build failure[0] on armhf with the new[1] approach to fix #221618[2]:

This is not a fix, this is a bug-on, so to say. We can't fix it in libsmbclient,
and the "fix" which closed that bug did not actually fix anything at all.

> # github.com/ubuntu/adsys/vendor/github.com/mvo5/libsmbclient-go
> In file included from
> src/github.com/ubuntu/adsys/vendor/github.com/mvo5/libsmbclient-go/libsmbclient.go:17:
> /usr/include/samba-4.0/libsmbclient.h:84:13: error: size of array
> 'smbc_off_t_should_be_at_least_64bits_use_LFS_CFLAGS' is too large
>     84 | typedef int
> smbc_off_t_should_be_at_least_64bits_use_LFS_CFLAGS[sizeof(off_t)-7];
>        |             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 
> I'm currently building all other reverse dependencies of libsmbclient
> that I can find (checking via reverse-deps -b libsmbclient-dev). So
> far it looks like adsys[3] is the only affected one, on armhf.

I did check all rdepends of libsmbclient on debian i386, - haven't found
any problematic cases. That's why I enabled this error with ease: it does
not, to my knowlege, actually break anything.

> Have you seen this failure in the wild yet, with other rdeps of
> libsmbclient? I understand it's the package that failed to build that
> has to be fixed. I'm just trying to get an idea of how many could be
> out there.

Nope, I know no failures like this in the wild, - you're reporting the
fist actual failure, it seems.

Please note this error is not 100% accurate: a package can be using
libsmbclient functions which does not operate with off_t (just open/read/
close), - in this case this error is a false positive, there's no actual
requirement to use LFS in this case. But unfortunately we can't (easily)
detect this in libsmbclient.

Thank you for understanding and patience!

/mjt