DPDK patches and discussions
 help / color / mirror / Atom feed
* [PATCH] net/nfp: fix length comparison for max DMA length
@ 2023-02-03  2:23 Chaoyong He
  2023-02-11  0:53 ` Ferruh Yigit
  0 siblings, 1 reply; 2+ messages in thread
From: Chaoyong He @ 2023-02-03  2:23 UTC (permalink / raw)
  To: dev; +Cc: oss-drivers, Richard Donkin, stable, Chaoyong He

From: Richard Donkin <richard.donkin@corigine.com>

The DMA transfer length is allowed to be as large as
NFDK_TX_MAX_DATA_PER_HEAD. The existing check would not allow a simple-
type descriptor to be created for a packet of size
NFDK_TX_MAX_DATA_PER_HEAD, but it would also not enable gather-type
descriptors to be created correctly later.

Change the limit of simple-type descriptors to include packets of size
NFDK_TX_MAX_DATA_PER_HEAD.

In practice this fixes traffic with packets of size 4096. Previously
such packets would not be transmitted correctly.

Fixes: c73dced48c8c ("net/nfp: add NFDk Tx")
Cc: stable@dpdk.org

Signed-off-by: Richard Donkin <richard.donkin@corigine.com>
Reviewed-by: Chaoyong He <chaoyong.he@corigine.com>
---
 drivers/net/nfp/nfp_rxtx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/nfp/nfp_rxtx.c b/drivers/net/nfp/nfp_rxtx.c
index 01cffdfde0..fa4cc94a43 100644
--- a/drivers/net/nfp/nfp_rxtx.c
+++ b/drivers/net/nfp/nfp_rxtx.c
@@ -1214,7 +1214,7 @@ nfp_net_nfdk_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t nb_pk
 		if ((hw->cap & NFP_NET_CFG_CTRL_LSO_ANY) &&
 				(pkt->ol_flags & RTE_MBUF_F_TX_TCP_SEG)) {
 			type = NFDK_DESC_TX_TYPE_TSO;
-		} else if (!pkt->next && dma_len < NFDK_TX_MAX_DATA_PER_HEAD) {
+		} else if (!pkt->next && dma_len <= NFDK_TX_MAX_DATA_PER_HEAD) {
 			type = NFDK_DESC_TX_TYPE_SIMPLE;
 		} else {
 			type = NFDK_DESC_TX_TYPE_GATHER;
-- 
2.29.3


^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [PATCH] net/nfp: fix length comparison for max DMA length
  2023-02-03  2:23 [PATCH] net/nfp: fix length comparison for max DMA length Chaoyong He
@ 2023-02-11  0:53 ` Ferruh Yigit
  0 siblings, 0 replies; 2+ messages in thread
From: Ferruh Yigit @ 2023-02-11  0:53 UTC (permalink / raw)
  To: Chaoyong He, dev; +Cc: oss-drivers, Richard Donkin, stable

On 2/3/2023 2:23 AM, Chaoyong He wrote:
> From: Richard Donkin <richard.donkin@corigine.com>
> 
> The DMA transfer length is allowed to be as large as
> NFDK_TX_MAX_DATA_PER_HEAD. The existing check would not allow a simple-
> type descriptor to be created for a packet of size
> NFDK_TX_MAX_DATA_PER_HEAD, but it would also not enable gather-type
> descriptors to be created correctly later.
> 
> Change the limit of simple-type descriptors to include packets of size
> NFDK_TX_MAX_DATA_PER_HEAD.
> 
> In practice this fixes traffic with packets of size 4096. Previously
> such packets would not be transmitted correctly.
> 
> Fixes: c73dced48c8c ("net/nfp: add NFDk Tx")
> Cc: stable@dpdk.org
> 
> Signed-off-by: Richard Donkin <richard.donkin@corigine.com>
> Reviewed-by: Chaoyong He <chaoyong.he@corigine.com>

Applied to dpdk-next-net/main, thanks.


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2023-02-11  0:54 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-02-03  2:23 [PATCH] net/nfp: fix length comparison for max DMA length Chaoyong He
2023-02-11  0:53 ` 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).