Re: [PATCH bpf-next] selftests/bpf: Replace %pK output with 0

[Sebastian Andrzej Siewior <bigeasy@linutronix.de>]

On 2026-09-11 15:03:37 [+0000], sashiko-bot@kernel.org wrote:
> > diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_udp4.c b/tools/testing/selftests/bpf/progs/bpf_iter_udp4.c
> > index 23b2aa2604de2..9b789f09f1462 100644
> > --- a/tools/testing/selftests/bpf/progs/bpf_iter_udp4.c
> > +++ b/tools/testing/selftests/bpf/progs/bpf_iter_udp4.c
> > @@ -57,13 +57,13 @@ int dump_udp4(struct bpf_iter__udp *ctx)
> >  	BPF_SEQ_PRINTF(seq, "%5d: %08X:%04X %08X:%04X ",
> >  		       ctx->bucket, src, srcp, dest, destp);
> >  
> > -	BPF_SEQ_PRINTF(seq, "%02X %08X:%08X %02X:%08lX %08X %5u %8d %lu %d %pK %u\n",
> > +	BPF_SEQ_PRINTF(seq, "%02X %08X:%08X %02X:%08lX %08X %5u %8d %lu %d %8d %u\n",
> 
> [Severity: Low]
> Does this %8d format incorrectly pad the 0 with spaces?
> 
> The networking subsystem replaced %pK with a literal 0, but replacing it with
> %8d here in dump_udp4() causes a mismatch with the native /proc/net/ output.
> Since BPF iterators are designed as exact drop-in replacements, could this
> mismatch break parsers that rely on strict spacing?

Indeed. I tried to match the pointer spacing for some reason. The
formatting in net does not align with the columns starting with the
inode which is %lu instead having 5 width same goes for ref. So I just
ignored and moved on with a plain 0. I could update if this is
considered important.

> [ ... ]
> 
> > diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_udp6.c b/tools/testing/selftests/bpf/progs/bpf_iter_udp6.c
> > index c48b05aa2a4b2..9766eff7069ff 100644
> > --- a/tools/testing/selftests/bpf/progs/bpf_iter_udp6.c
> > +++ b/tools/testing/selftests/bpf/progs/bpf_iter_udp6.c
> > @@ -65,13 +65,13 @@ int dump_udp6(struct bpf_iter__udp *ctx)
> >  		       dest->s6_addr32[0], dest->s6_addr32[1],
> >  		       dest->s6_addr32[2], dest->s6_addr32[3], destp);
> >  
> > -	BPF_SEQ_PRINTF(seq, "%02X %08X:%08X %02X:%08lX %08X %5u %8d %lu %d %pK %u\n",
> > +	BPF_SEQ_PRINTF(seq, "%02X %08X:%08X %02X:%08lX %08X %5u %8d %lu %d %8d %u\n",
> 
> [Severity: Low]
> Is the use of %8d here in dump_udp6() introducing the same formatting
> mismatch as in dump_udp4()?
> 
> It seems it should use a literal 0 instead of %8d to exactly match the
> networking subsystem's native output format.

Yes it does but the extra spacing intends to align with the columns…

Sebastian