* [PATCH] net/pcap: use pcap_next_ex to track errors
@ 2024-09-05 16:10 Stephen Hemminger
2024-10-10 2:14 ` Ferruh Yigit
2024-10-13 15:47 ` Ferruh Yigit
0 siblings, 2 replies; 6+ messages in thread
From: Stephen Hemminger @ 2024-09-05 16:10 UTC (permalink / raw)
To: dev; +Cc: Acked-by : Ferruh Yigit, Stephen Hemminger, Ofer Dagan
Use pcap_next_ex rather than just pcap_next because pcap_next
always blocks if there is no packets to receive.
Bugzilla ID: 1526
Reported-by: Ofer Dagan <ofer.d@claroty.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
---
drivers/net/pcap/pcap_ethdev.c | 33 +++++++++++++++++----------------
1 file changed, 17 insertions(+), 16 deletions(-)
diff --git a/drivers/net/pcap/pcap_ethdev.c b/drivers/net/pcap/pcap_ethdev.c
index bfec085045..261997be5c 100644
--- a/drivers/net/pcap/pcap_ethdev.c
+++ b/drivers/net/pcap/pcap_ethdev.c
@@ -274,7 +274,7 @@ static uint16_t
eth_pcap_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
{
unsigned int i;
- struct pcap_pkthdr header;
+ struct pcap_pkthdr *header;
struct pmd_process_private *pp;
const u_char *packet;
struct rte_mbuf *mbuf;
@@ -294,9 +294,13 @@ eth_pcap_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
*/
for (i = 0; i < nb_pkts; i++) {
/* Get the next PCAP packet */
- packet = pcap_next(pcap, &header);
- if (unlikely(packet == NULL))
+ int ret = pcap_next_ex(pcap, &header, &packet);
+ if (ret != 1) {
+ if (ret == PCAP_ERROR)
+ pcap_q->rx_stat.err_pkts++;
+
break;
+ }
mbuf = rte_pktmbuf_alloc(pcap_q->mb_pool);
if (unlikely(mbuf == NULL)) {
@@ -304,33 +308,30 @@ eth_pcap_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
break;
}
- if (header.caplen <= rte_pktmbuf_tailroom(mbuf)) {
+ uint32_t len = header->caplen;
+ if (len <= rte_pktmbuf_tailroom(mbuf)) {
/* pcap packet will fit in the mbuf, can copy it */
- rte_memcpy(rte_pktmbuf_mtod(mbuf, void *), packet,
- header.caplen);
- mbuf->data_len = (uint16_t)header.caplen;
+ rte_memcpy(rte_pktmbuf_mtod(mbuf, void *), packet, len);
+ mbuf->data_len = len;
} else {
/* Try read jumbo frame into multi mbufs. */
if (unlikely(eth_pcap_rx_jumbo(pcap_q->mb_pool,
- mbuf,
- packet,
- header.caplen) == -1)) {
+ mbuf, packet, len) == -1)) {
pcap_q->rx_stat.err_pkts++;
rte_pktmbuf_free(mbuf);
break;
}
}
- mbuf->pkt_len = (uint16_t)header.caplen;
- *RTE_MBUF_DYNFIELD(mbuf, timestamp_dynfield_offset,
- rte_mbuf_timestamp_t *) =
- (uint64_t)header.ts.tv_sec * 1000000 +
- header.ts.tv_usec;
+ mbuf->pkt_len = len;
+ uint64_t us = (uint64_t)header->ts.tv_sec * US_PER_S + header->ts.tv_usec;
+
+ *RTE_MBUF_DYNFIELD(mbuf, timestamp_dynfield_offset, rte_mbuf_timestamp_t *) = us;
mbuf->ol_flags |= timestamp_rx_dynflag;
mbuf->port = pcap_q->port_id;
bufs[num_rx] = mbuf;
num_rx++;
- rx_bytes += header.caplen;
+ rx_bytes += len;
}
pcap_q->rx_stat.pkts += num_rx;
pcap_q->rx_stat.bytes += rx_bytes;
--
2.45.2
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] net/pcap: use pcap_next_ex to track errors
2024-09-05 16:10 [PATCH] net/pcap: use pcap_next_ex to track errors Stephen Hemminger
@ 2024-10-10 2:14 ` Ferruh Yigit
2024-10-10 2:21 ` Stephen Hemminger
2024-10-13 15:47 ` Ferruh Yigit
1 sibling, 1 reply; 6+ messages in thread
From: Ferruh Yigit @ 2024-10-10 2:14 UTC (permalink / raw)
To: Stephen Hemminger, dev; +Cc: Ofer Dagan, Tyler Retzlaff
On 9/5/2024 5:10 PM, Stephen Hemminger wrote:
> Use pcap_next_ex rather than just pcap_next because pcap_next
> always blocks if there is no packets to receive.
>
Hi Stephen,
Do you know if using 'pcap_next_ex()' (instead of 'pcap_next()') has any
dependency impact?
Like can we rely that all libraries that support 'pcap_next()', also
supports 'pcap_next_ex()'?
> Bugzilla ID: 1526
> Reported-by: Ofer Dagan <ofer.d@claroty.com>
> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
>
Hi Ofer, Can you please verify this fix?
> ---
> drivers/net/pcap/pcap_ethdev.c | 33 +++++++++++++++++----------------
> 1 file changed, 17 insertions(+), 16 deletions(-)
>
> diff --git a/drivers/net/pcap/pcap_ethdev.c b/drivers/net/pcap/pcap_ethdev.c
> index bfec085045..261997be5c 100644
> --- a/drivers/net/pcap/pcap_ethdev.c
> +++ b/drivers/net/pcap/pcap_ethdev.c
> @@ -274,7 +274,7 @@ static uint16_t
> eth_pcap_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
> {
> unsigned int i;
> - struct pcap_pkthdr header;
> + struct pcap_pkthdr *header;
> struct pmd_process_private *pp;
> const u_char *packet;
> struct rte_mbuf *mbuf;
> @@ -294,9 +294,13 @@ eth_pcap_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
> */
> for (i = 0; i < nb_pkts; i++) {
> /* Get the next PCAP packet */
> - packet = pcap_next(pcap, &header);
> - if (unlikely(packet == NULL))
> + int ret = pcap_next_ex(pcap, &header, &packet);
> + if (ret != 1) {
> + if (ret == PCAP_ERROR)
> + pcap_q->rx_stat.err_pkts++;
> +
> break;
> + }
>
> mbuf = rte_pktmbuf_alloc(pcap_q->mb_pool);
> if (unlikely(mbuf == NULL)) {
> @@ -304,33 +308,30 @@ eth_pcap_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
> break;
> }
>
> - if (header.caplen <= rte_pktmbuf_tailroom(mbuf)) {
> + uint32_t len = header->caplen;
> + if (len <= rte_pktmbuf_tailroom(mbuf)) {
> /* pcap packet will fit in the mbuf, can copy it */
> - rte_memcpy(rte_pktmbuf_mtod(mbuf, void *), packet,
> - header.caplen);
> - mbuf->data_len = (uint16_t)header.caplen;
> + rte_memcpy(rte_pktmbuf_mtod(mbuf, void *), packet, len);
> + mbuf->data_len = len;
> } else {
> /* Try read jumbo frame into multi mbufs. */
> if (unlikely(eth_pcap_rx_jumbo(pcap_q->mb_pool,
> - mbuf,
> - packet,
> - header.caplen) == -1)) {
> + mbuf, packet, len) == -1)) {
> pcap_q->rx_stat.err_pkts++;
> rte_pktmbuf_free(mbuf);
> break;
> }
> }
>
> - mbuf->pkt_len = (uint16_t)header.caplen;
> - *RTE_MBUF_DYNFIELD(mbuf, timestamp_dynfield_offset,
> - rte_mbuf_timestamp_t *) =
> - (uint64_t)header.ts.tv_sec * 1000000 +
> - header.ts.tv_usec;
> + mbuf->pkt_len = len;
> + uint64_t us = (uint64_t)header->ts.tv_sec * US_PER_S + header->ts.tv_usec;
> +
> + *RTE_MBUF_DYNFIELD(mbuf, timestamp_dynfield_offset, rte_mbuf_timestamp_t *) = us;
> mbuf->ol_flags |= timestamp_rx_dynflag;
> mbuf->port = pcap_q->port_id;
> bufs[num_rx] = mbuf;
> num_rx++;
> - rx_bytes += header.caplen;
> + rx_bytes += len;
> }
> pcap_q->rx_stat.pkts += num_rx;
> pcap_q->rx_stat.bytes += rx_bytes;
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] net/pcap: use pcap_next_ex to track errors
2024-10-10 2:14 ` Ferruh Yigit
@ 2024-10-10 2:21 ` Stephen Hemminger
2024-10-10 2:24 ` Ferruh Yigit
0 siblings, 1 reply; 6+ messages in thread
From: Stephen Hemminger @ 2024-10-10 2:21 UTC (permalink / raw)
To: Ferruh Yigit; +Cc: dev, Ofer Dagan, Tyler Retzlaff
On Thu, 10 Oct 2024 03:14:48 +0100
Ferruh Yigit <ferruh.yigit@amd.com> wrote:
> On 9/5/2024 5:10 PM, Stephen Hemminger wrote:
> > Use pcap_next_ex rather than just pcap_next because pcap_next
> > always blocks if there is no packets to receive.
> >
>
> Hi Stephen,
>
> Do you know if using 'pcap_next_ex()' (instead of 'pcap_next()') has any
> dependency impact?
> Like can we rely that all libraries that support 'pcap_next()', also
> supports 'pcap_next_ex()'?
The code depends on libpcap, and that API has been in pcap since the
early days. See it in a version from 2003!
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] net/pcap: use pcap_next_ex to track errors
2024-10-10 2:21 ` Stephen Hemminger
@ 2024-10-10 2:24 ` Ferruh Yigit
2024-10-10 2:34 ` Stephen Hemminger
0 siblings, 1 reply; 6+ messages in thread
From: Ferruh Yigit @ 2024-10-10 2:24 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: dev, Ofer Dagan, Tyler Retzlaff
On 10/10/2024 3:21 AM, Stephen Hemminger wrote:
> On Thu, 10 Oct 2024 03:14:48 +0100
> Ferruh Yigit <ferruh.yigit@amd.com> wrote:
>
>> On 9/5/2024 5:10 PM, Stephen Hemminger wrote:
>>> Use pcap_next_ex rather than just pcap_next because pcap_next
>>> always blocks if there is no packets to receive.
>>>
>>
>> Hi Stephen,
>>
>> Do you know if using 'pcap_next_ex()' (instead of 'pcap_next()') has any
>> dependency impact?
>> Like can we rely that all libraries that support 'pcap_next()', also
>> supports 'pcap_next_ex()'?
>
> The code depends on libpcap, and that API has been in pcap since the
> early days. See it in a version from 2003!
>
Thanks for confirming. I assume it is same with the 'iphlpapi' that
windows uses.
Let wait a little more, it would be good if we get a test result from
Ofer first.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] net/pcap: use pcap_next_ex to track errors
2024-10-10 2:24 ` Ferruh Yigit
@ 2024-10-10 2:34 ` Stephen Hemminger
0 siblings, 0 replies; 6+ messages in thread
From: Stephen Hemminger @ 2024-10-10 2:34 UTC (permalink / raw)
To: Ferruh Yigit; +Cc: dev, Ofer Dagan, Tyler Retzlaff
On Thu, 10 Oct 2024 03:24:42 +0100
Ferruh Yigit <ferruh.yigit@amd.com> wrote:
> On 10/10/2024 3:21 AM, Stephen Hemminger wrote:
> > On Thu, 10 Oct 2024 03:14:48 +0100
> > Ferruh Yigit <ferruh.yigit@amd.com> wrote:
> >
> >> On 9/5/2024 5:10 PM, Stephen Hemminger wrote:
> >>> Use pcap_next_ex rather than just pcap_next because pcap_next
> >>> always blocks if there is no packets to receive.
> >>>
> >>
> >> Hi Stephen,
> >>
> >> Do you know if using 'pcap_next_ex()' (instead of 'pcap_next()') has any
> >> dependency impact?
> >> Like can we rely that all libraries that support 'pcap_next()', also
> >> supports 'pcap_next_ex()'?
> >
> > The code depends on libpcap, and that API has been in pcap since the
> > early days. See it in a version from 2003!
> >
>
> Thanks for confirming. I assume it is same with the 'iphlpapi' that
> windows uses.
>
> Let wait a little more, it would be good if we get a test result from
> Ofer first.
>
I think the Windows build would have caught any issue.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] net/pcap: use pcap_next_ex to track errors
2024-09-05 16:10 [PATCH] net/pcap: use pcap_next_ex to track errors Stephen Hemminger
2024-10-10 2:14 ` Ferruh Yigit
@ 2024-10-13 15:47 ` Ferruh Yigit
1 sibling, 0 replies; 6+ messages in thread
From: Ferruh Yigit @ 2024-10-13 15:47 UTC (permalink / raw)
To: Stephen Hemminger, dev; +Cc: Ofer Dagan
On 9/5/2024 5:10 PM, Stephen Hemminger wrote:
> Use pcap_next_ex rather than just pcap_next because pcap_next
> always blocks if there is no packets to receive.
>
> Bugzilla ID: 1526
> Fixes: 4c173302c307 ("pcap: add new driver")
> Cc: stable@dpdk.org
>
> Reported-by: Ofer Dagan <ofer.d@claroty.com>
> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
>
According Bugzilla comment:
Tested-by: Ofer Dagan <ofer.d@claroty.com>
Applied to dpdk-next-net/main, thanks.
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2024-10-13 15:48 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2024-09-05 16:10 [PATCH] net/pcap: use pcap_next_ex to track errors Stephen Hemminger
2024-10-10 2:14 ` Ferruh Yigit
2024-10-10 2:21 ` Stephen Hemminger
2024-10-10 2:24 ` Ferruh Yigit
2024-10-10 2:34 ` Stephen Hemminger
2024-10-13 15:47 ` Ferruh Yigit
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).