Re: [PATCH bpf-next v1 5/8] selftests/bpf: bad_struct_ops test

[David Vernet <void@manifault.com>]

[-- Attachment #1: Type: text/plain, Size: 2084 bytes --]

On Wed, Feb 28, 2024 at 10:06:21PM +0200, Eduard Zingerman wrote:
> On Wed, 2024-02-28 at 12:15 -0600, David Vernet wrote:
> [...]
> 
> > > +static libbpf_print_fn_t old_print_cb;
> > > +static bool msg_found;
> > > +
> > > +static int print_cb(enum libbpf_print_level level, const char *fmt, va_list args)
> > > +{
> > > +	old_print_cb(level, fmt, args);
> > > +	if (level == LIBBPF_WARN && strncmp(fmt, EXPECTED_MSG, strlen(EXPECTED_MSG)) == 0)
> > > +		msg_found = true;
> > > +
> > > +	return 0;
> > > +}
> > 
> > Not necessary at all for this patch set / just an observation, but it would be
> > nice to have this be something offered by the core prog_tests framework
> > (meaning, the ability to assert libbpf output for a testcase).
> 
> This might be useful, I will add a utility function for it (probably two).
> 
> [...]
> 
> > > diff --git a/tools/testing/selftests/bpf/progs/bad_struct_ops.c b/tools/testing/selftests/bpf/progs/bad_struct_ops.c
> > > new file mode 100644
> > > index 000000000000..9c103afbfdb1
> > > --- /dev/null
> > > +++ b/tools/testing/selftests/bpf/progs/bad_struct_ops.c
> > > @@ -0,0 +1,17 @@
> > > +// SPDX-License-Identifier: GPL-2.0
> > > +
> > > +#include <vmlinux.h>
> > > +#include <bpf/bpf_helpers.h>
> > > +#include <bpf/bpf_tracing.h>
> > > +#include "../bpf_testmod/bpf_testmod.h"
> > > +
> > > +char _license[] SEC("license") = "GPL";
> > > +
> > > +SEC("struct_ops/test_1")
> > > +int BPF_PROG(test_1) { return 0; }
> > > +
> > > +SEC(".struct_ops.link")
> > > +struct bpf_testmod_ops testmod_1 = { .test_1 = (void *)test_1 };
> > 
> > Just to make be 100% sure that we're isolating the issue under test, should we
> > also add a .test_2 prog and add it to the struct bpf_testmod_ops map?
> 
> You are concerned that error might be confused with libbpf insisting
> that '.test_2' should be present, right?
> libbpf allows NULL members but I can add '.test_2' here, no problem.

Correct, and yes that's true. Feel free to ignore if you think it's cleaner
without, totally up to you.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]