Re: [dpdk-dev] [PATCH] mbuf: display more fields in dump

Re: [dpdk-dev] [PATCH] mbuf: display more fields in dump

[Morten Brørup]

I think the name also changed from in_port.Sent from a smartphone. Please pardon brevity and spelling.
-------- Oprindelig besked --------Fra: Stephen Hemminger <stephen@networkplumber.org> Dato: 21/11/2019  17.43  (GMT+01:00) Til: Morten Brørup <mb@smartsharesystems.com> Cc: dev@dpdk.org Emne: Re: [dpdk-dev] [PATCH] mbuf: display more fields in dump On Thu, 21 Nov 2019 17:16:42 +0100Morten Brørup <mb@smartsharesystems.com> wrote:> Are all these explicit type casts really needed? buf_iova is already uint64_t, buflen is uint16_t, data_off is uint16_t, rte_mbuf_refcnt_read() returns uint16_t, pkt_len is uint32_t, nb_segs is uint16_t, port is uint16_t, vlan_tci is uint16_t.Normally, they are not needed.If you turn up some of the warning levels in Gcc, then yes.> > m->port is not only the input port anymore (ref. the mbuf documentation), so please change in_port=%u to port=%u.Makes sense, but since this in internal would rather keep existing to print the name of the field.> > For consistency with ol_flags and vlan_tci format, consider changing iova=% to iova=%#.good idea, personal preference is to always use %#x over %x

[dpdk-dev] [PATCH] mbuf: display more fields in dump

[Stephen Hemminger]

The rte_pktmbuf_dump should display offset, refcount, and vlan
info since these are often useful during debugging.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
---
 lib/librte_mbuf/rte_mbuf.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/lib/librte_mbuf/rte_mbuf.c b/lib/librte_mbuf/rte_mbuf.c
index 35df1c4c38a5..a964464bd598 100644
--- a/lib/librte_mbuf/rte_mbuf.c
+++ b/lib/librte_mbuf/rte_mbuf.c
@@ -473,11 +473,14 @@ rte_pktmbuf_dump(FILE *f, const struct rte_mbuf *m, unsigned dump_len)
 
 	__rte_mbuf_sanity_check(m, 1);
 
-	fprintf(f, "dump mbuf at %p, iova=%"PRIx64", buf_len=%u\n",
-	       m, (uint64_t)m->buf_iova, (unsigned)m->buf_len);
-	fprintf(f, "  pkt_len=%"PRIu32", ol_flags=%"PRIx64", nb_segs=%u, "
-	       "in_port=%u\n", m->pkt_len, m->ol_flags,
-	       (unsigned)m->nb_segs, (unsigned)m->port);
+	fprintf(f, "dump mbuf at %p, iova=%"PRIx64", buf_len=%u,"
+		" offs=%u, refcnt=%u\n",
+		m, (uint64_t)m->buf_iova, (unsigned)m->buf_len,
+		(unsigned)m->data_off, (unsigned)rte_mbuf_refcnt_read(m));
+	fprintf(f, "  pkt_len=%u, ol_flags=%#"PRIx64","
+		" nb_segs=%u, in_port=%u, vlan_tci=%#x\n",
+		(unsigned)m->pkt_len, m->ol_flags,
+		(unsigned)m->nb_segs, (unsigned)m->port, (unsigned)m->vlan_tci);
 	nb_segs = m->nb_segs;
 
 	while (m && nb_segs != 0) {
-- 
2.20.1

Re: [dpdk-dev] [PATCH] mbuf: display more fields in dump

[Morten Brørup]

> -----Original Message-----
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Stephen Hemminger
> Sent: Wednesday, November 20, 2019 6:41 PM
> 
> The rte_pktmbuf_dump should display offset, refcount, and vlan
> info since these are often useful during debugging.
> 
> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
> ---
>  lib/librte_mbuf/rte_mbuf.c | 13 ++++++++-----
>  1 file changed, 8 insertions(+), 5 deletions(-)
> 
> diff --git a/lib/librte_mbuf/rte_mbuf.c b/lib/librte_mbuf/rte_mbuf.c
> index 35df1c4c38a5..a964464bd598 100644
> --- a/lib/librte_mbuf/rte_mbuf.c
> +++ b/lib/librte_mbuf/rte_mbuf.c
> @@ -473,11 +473,14 @@ rte_pktmbuf_dump(FILE *f, const struct rte_mbuf
> *m, unsigned dump_len)
> 
>  	__rte_mbuf_sanity_check(m, 1);
> 
> -	fprintf(f, "dump mbuf at %p, iova=%"PRIx64", buf_len=%u\n",
> -	       m, (uint64_t)m->buf_iova, (unsigned)m->buf_len);
> -	fprintf(f, "  pkt_len=%"PRIu32", ol_flags=%"PRIx64", nb_segs=%u,
> "
> -	       "in_port=%u\n", m->pkt_len, m->ol_flags,
> -	       (unsigned)m->nb_segs, (unsigned)m->port);
> +	fprintf(f, "dump mbuf at %p, iova=%"PRIx64", buf_len=%u,"
> +		" offs=%u, refcnt=%u\n",
> +		m, (uint64_t)m->buf_iova, (unsigned)m->buf_len,
> +		(unsigned)m->data_off, (unsigned)rte_mbuf_refcnt_read(m));
> +	fprintf(f, "  pkt_len=%u, ol_flags=%#"PRIx64","
> +		" nb_segs=%u, in_port=%u, vlan_tci=%#x\n",
> +		(unsigned)m->pkt_len, m->ol_flags,
> +		(unsigned)m->nb_segs, (unsigned)m->port, (unsigned)m-
> >vlan_tci);

Are all these explicit type casts really needed? buf_iova is already uint64_t, buflen is uint16_t, data_off is uint16_t, rte_mbuf_refcnt_read() returns uint16_t, pkt_len is uint32_t, nb_segs is uint16_t, port is uint16_t, vlan_tci is uint16_t.

m->port is not only the input port anymore (ref. the mbuf documentation), so please change in_port=%u to port=%u.

For consistency with ol_flags and vlan_tci format, consider changing iova=% to iova=%#.

>  	nb_segs = m->nb_segs;
> 
>  	while (m && nb_segs != 0) {
> --
> 2.20.1
> 

Re: [dpdk-dev] [PATCH] mbuf: display more fields in dump

[Stephen Hemminger]

On Thu, 21 Nov 2019 17:16:42 +0100
Morten Brørup <mb@smartsharesystems.com> wrote:

> Are all these explicit type casts really needed? buf_iova is already uint64_t, buflen is uint16_t, data_off is uint16_t, rte_mbuf_refcnt_read() returns uint16_t, pkt_len is uint32_t, nb_segs is uint16_t, port is uint16_t, vlan_tci is uint16_t.

Normally, they are not needed.
If you turn up some of the warning levels in Gcc, then yes.

> 
> m->port is not only the input port anymore (ref. the mbuf documentation), so please change in_port=%u to port=%u.

Makes sense, but since this in internal would rather keep existing
 to print the name of the field.

> 
> For consistency with ol_flags and vlan_tci format, consider changing iova=% to iova=%#.

good idea, personal preference is to always use %#x over %x