* memif jumbo fixes
@ 2025-01-02 21:42 Mike Bly
2025-01-06 0:13 ` Stephen Hemminger
0 siblings, 1 reply; 2+ messages in thread
From: Mike Bly @ 2025-01-02 21:42 UTC (permalink / raw)
To: dev
[-- Attachment #1: Type: text/plain, Size: 2720 bytes --]
Please review/accept the following fixes for drivers/net/memif. This is in
regards to https://bugs.dpdk.org/show_bug.cgi?id=1609.
RX and TX fixes to correct linking/handling of multiple memif buffers for
jumbo frames (>2KB) in mixed traffic scenarios.
From fa1f5ab578bb06aa22b356b8ce5cfbdb52dd4a48 Mon Sep 17 00:00:00 2001
From: Mike Bly <bly454@gmail.com>
Date: Thu, 2 Jan 2025 13:00:57 -0800
Subject: [PATCH 1/2] Moving where dst_off is set in memif_rx function Fixes:
aa17df86089 ("net/memif: add a Rx fast path")
---
drivers/net/memif/rte_eth_memif.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/net/memif/rte_eth_memif.c
b/drivers/net/memif/rte_eth_memif.c
index 1eb41bb471..29fd890ed1 100644
--- a/drivers/net/memif/rte_eth_memif.c
+++ b/drivers/net/memif/rte_eth_memif.c
@@ -414,13 +414,12 @@ eth_memif_rx(void *queue, struct rte_mbuf **bufs,
uint16_t nb_pkts)
goto no_free_bufs;
mbuf = mbuf_head;
mbuf->port = mq->in_port;
-
+ dst_off = 0;
next_slot2:
s0 = cur_slot & mask;
d0 = &ring->desc[s0];
src_len = d0->length;
- dst_off = 0;
src_off = 0;
do {
--
2.43.5
From eb1770036ce829e193953af9994d8ae16e75272b Mon Sep 17 00:00:00 2001
From: Mike Bly <bly454@gmail.com>
Date: Thu, 2 Jan 2025 13:20:23 -0800
Subject: [PATCH 2/2] Added missing initialization of memif flags field per
memif buffer Fixes: 09c7e63a71f ("net/memif: introduce memory interface
PMD")
---
drivers/net/memif/rte_eth_memif.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/memif/rte_eth_memif.c
b/drivers/net/memif/rte_eth_memif.c
index 29fd890ed1..57216d7096 100644
--- a/drivers/net/memif/rte_eth_memif.c
+++ b/drivers/net/memif/rte_eth_memif.c
@@ -709,6 +709,7 @@ eth_memif_tx(void *queue, struct rte_mbuf **bufs,
uint16_t nb_pkts)
next_in_chain1:
d0 = &ring->desc[slot & mask];
cp_len = rte_pktmbuf_data_len(mbuf);
+ d0->flags = 0;
rte_memcpy((uint8_t
*)memif_get_buffer(proc_private, d0),
rte_pktmbuf_mtod(mbuf, void *), cp_len);
@@ -744,6 +745,7 @@ eth_memif_tx(void *queue, struct rte_mbuf **bufs,
uint16_t nb_pkts)
dst_off = 0;
dst_len = (type == MEMIF_RING_C2S) ?
pmd->run.pkt_buffer_size : d0->length;
+ d0->flags = 0;
next_in_chain2:
src_off = 0;
--
2.43.5
-Mike
[-- Attachment #2: Type: text/html, Size: 3496 bytes --]
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: memif jumbo fixes
2025-01-02 21:42 memif jumbo fixes Mike Bly
@ 2025-01-06 0:13 ` Stephen Hemminger
0 siblings, 0 replies; 2+ messages in thread
From: Stephen Hemminger @ 2025-01-06 0:13 UTC (permalink / raw)
To: Mike Bly; +Cc: dev
On Thu, 2 Jan 2025 13:42:10 -0800
Mike Bly <bly454@gmail.com> wrote:
> Please review/accept the following fixes for drivers/net/memif. This is in
> regards to https://bugs.dpdk.org/show_bug.cgi?id=1609.
>
> RX and TX fixes to correct linking/handling of multiple memif buffers for
> jumbo frames (>2KB) in mixed traffic scenarios.
>
> From fa1f5ab578bb06aa22b356b8ce5cfbdb52dd4a48 Mon Sep 17 00:00:00 2001
> From: Mike Bly <bly454@gmail.com>
> Date: Thu, 2 Jan 2025 13:00:57 -0800
> Subject: [PATCH 1/2] Moving where dst_off is set in memif_rx function Fixes:
> aa17df86089 ("net/memif: add a Rx fast path")
>
Patch is missing some things:
- Developer Certificate of Origin (DCO) i.e. Signed-off-by is required
- Please use Bugzilla ID: tag to match with bug report
- Please add Fixes: tag
- Please use Cc: stable@dpdk.org for bugs fixes that should be backported
- Please Cc: the maintainer of memif Jakub Grajciar <jgrajcia@cisco.com>
See:
https://doc.dpdk.org/guides/contributing/patches.html
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2025-01-06 0:13 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2025-01-02 21:42 memif jumbo fixes Mike Bly
2025-01-06 0:13 ` Stephen Hemminger
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).