On Tue, 2023-09-05 at 15:00 +0200, Phil Sutter wrote:
> On Mon, Sep 04, 2023 at 11:06:30AM +0200, Florian Westphal wrote:
> > Running selftests on older kernels makes some of them fail very
> > early
> > because some tests use features that are not available on older
> > kernels, e.g. -stable releases.
> >
> > Known examples:
> > - inner header matching
> > - anonymous chains
> > - elem delete from packet path
> >
> > Also, some test cases might fail because a feature isn't
> > compiled in, such as netdev chains for example.
> >
> > This adds a feature-probing to the shell tests.
> >
> > Simply drop a 'nft -f' compatible file with a .nft suffix into
> > tests/shell/features.
> >
> > run-tests.sh will load it via --check and will add
> >
> > NFT_TESTS_HAVE_${filename}=$?
>
> Maybe make this:
>
> > truefalse=(true false)
> > NFT_TESTS_HAVE_${filename}=${truefalse[$?]}
>
> [...]
>
> > [ $NFT_HAVE_chain_binding -eq 1 ] && test_chain_binding
>
> So this becomes:
>
> > $NFT_HAVE_chain_binding && test_chain_binding
>
> Use of true/false appears to work in dash, so might be POSIX sh
> compatible?
>
> Cheers, Phil
>
That's possible. But I would not do such non-obvious magic.
The existing variables like VERBOSE,VALGRIND,DUMPGEN use "y" for true
and everything else is false. I'd stick with that form.
The form would be (exactly):
if [ "$NFT_TESTS_HAVE_chain_binding" = y ] ; then
Btw, as a matter of principle, I think that all variables in the patch
set should be quoted.
Thomas
