* [dpdk-stable] [PATCH v3 02/13] net/enetc: fix SMMU unhandled context fault [not found] ` <20190412105105.24351-1-g.singh@nxp.com> @ 2019-04-12 11:01 ` Gagandeep Singh 2019-04-12 11:01 ` [dpdk-stable] [PATCH v3 03/13] net/enetc: fix big endian build and correct buffer allocation Gagandeep Singh ` (2 subsequent siblings) 3 siblings, 0 replies; 6+ messages in thread From: Gagandeep Singh @ 2019-04-12 11:01 UTC (permalink / raw) To: dev, ferruh.yigit; +Cc: Gagandeep Singh, stable First configure ring with BDs properly then enable the ring. Fixes: 469c6111a799 ("net/enetc: enable Rx and Tx") Cc: stable@dpdk.org Signed-off-by: Gagandeep Singh <g.singh@nxp.com> --- drivers/net/enetc/enetc_ethdev.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/enetc/enetc_ethdev.c b/drivers/net/enetc/enetc_ethdev.c index 7a9a97d..f1807b9 100644 --- a/drivers/net/enetc/enetc_ethdev.c +++ b/drivers/net/enetc/enetc_ethdev.c @@ -492,15 +492,15 @@ static int enetc_tx_queue_setup(struct rte_eth_dev *dev, uint16_t tx_queue_id, ENETC_RTBLENR_LEN(rx_ring->bd_count)); rx_ring->mb_pool = mb_pool; - /* enable ring */ - enetc_rxbdr_wr(hw, idx, ENETC_RBMR, ENETC_RBMR_EN); - enetc_rxbdr_wr(hw, idx, ENETC_RBPIR, 0); rx_ring->rcir = (void *)((size_t)hw->reg + ENETC_BDR(RX, idx, ENETC_RBCIR)); enetc_refill_rx_ring(rx_ring, (enetc_bd_unused(rx_ring))); buf_size = (uint16_t)(rte_pktmbuf_data_room_size(rx_ring->mb_pool) - RTE_PKTMBUF_HEADROOM); enetc_rxbdr_wr(hw, idx, ENETC_RBBSR, buf_size); + /* enable ring */ + enetc_rxbdr_wr(hw, idx, ENETC_RBMR, ENETC_RBMR_EN); + enetc_rxbdr_wr(hw, idx, ENETC_RBPIR, 0); } static int -- 1.9.1 ^ permalink raw reply [flat|nested] 6+ messages in thread
* [dpdk-stable] [PATCH v3 03/13] net/enetc: fix big endian build and correct buffer allocation [not found] ` <20190412105105.24351-1-g.singh@nxp.com> 2019-04-12 11:01 ` [dpdk-stable] [PATCH v3 02/13] net/enetc: fix SMMU unhandled context fault Gagandeep Singh @ 2019-04-12 11:01 ` Gagandeep Singh 2019-04-12 11:01 ` [dpdk-stable] [PATCH v3 13/13] net/enetc: fix crash at high speed traffic Gagandeep Singh [not found] ` <20190412122840.1908-1-g.singh@nxp.com> 3 siblings, 0 replies; 6+ messages in thread From: Gagandeep Singh @ 2019-04-12 11:01 UTC (permalink / raw) To: dev, ferruh.yigit; +Cc: Gagandeep Singh, thomas, stable There was an error at rte_constant_bswap64 while compiling with big endian toolchain. so fixing it by adding type cast. Also, rte_pktmbuf_alloc API should be used to allocate mbuf instead of rte_pktmbuf_raw_alloc to avoid use of stale mbuf information. Fixes: 469c6111a799 ("net/enetc: enable Rx and Tx") Cc: thomas@monjalon.net Cc: stable@dpdk.org Signed-off-by: Gagandeep Singh <g.singh@nxp.com> --- drivers/net/enetc/enetc_rxtx.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/enetc/enetc_rxtx.c b/drivers/net/enetc/enetc_rxtx.c index a31a387..42f16ca 100644 --- a/drivers/net/enetc/enetc_rxtx.c +++ b/drivers/net/enetc/enetc_rxtx.c @@ -88,8 +88,9 @@ rx_swbd = &rx_ring->q_swbd[i]; rxbd = ENETC_RXBD(*rx_ring, i); for (j = 0; j < buff_cnt; j++) { - rx_swbd->buffer_addr = - rte_cpu_to_le_64(rte_mbuf_raw_alloc(rx_ring->mb_pool)); + rx_swbd->buffer_addr = (void *)(uintptr_t) + rte_cpu_to_le_64((uint64_t)(uintptr_t) + rte_pktmbuf_alloc(rx_ring->mb_pool)); rxbd->w.addr = (uint64_t)(uintptr_t) rx_swbd->buffer_addr->buf_iova + rx_swbd->buffer_addr->data_off; -- 1.9.1 ^ permalink raw reply [flat|nested] 6+ messages in thread
* [dpdk-stable] [PATCH v3 13/13] net/enetc: fix crash at high speed traffic [not found] ` <20190412105105.24351-1-g.singh@nxp.com> 2019-04-12 11:01 ` [dpdk-stable] [PATCH v3 02/13] net/enetc: fix SMMU unhandled context fault Gagandeep Singh 2019-04-12 11:01 ` [dpdk-stable] [PATCH v3 03/13] net/enetc: fix big endian build and correct buffer allocation Gagandeep Singh @ 2019-04-12 11:01 ` Gagandeep Singh [not found] ` <20190412122840.1908-1-g.singh@nxp.com> 3 siblings, 0 replies; 6+ messages in thread From: Gagandeep Singh @ 2019-04-12 11:01 UTC (permalink / raw) To: dev, ferruh.yigit; +Cc: Gagandeep Singh, stable On xmit side, there should be a check whether BD ring has free BDs available before transmit a packet to avoid data corruption and buffer leak issue. Fixes: 469c6111a799 ("net/enetc: enable Rx and Tx") Cc: stable@dpdk.org Signed-off-by: Gagandeep Singh <g.singh@nxp.com> --- drivers/net/enetc/enetc_rxtx.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/enetc/enetc_rxtx.c b/drivers/net/enetc/enetc_rxtx.c index 0ce7dbe..81b0ef3 100644 --- a/drivers/net/enetc/enetc_rxtx.c +++ b/drivers/net/enetc/enetc_rxtx.c @@ -49,11 +49,16 @@ uint16_t nb_pkts) { struct enetc_swbd *tx_swbd; - int i, start; + int i, start, bds_to_use; struct enetc_tx_bd *txbd; struct enetc_bdr *tx_ring = (struct enetc_bdr *)tx_queue; i = tx_ring->next_to_use; + + bds_to_use = enetc_bd_unused(tx_ring); + if (bds_to_use < nb_pkts) + nb_pkts = bds_to_use; + start = 0; while (nb_pkts--) { enetc_clean_tx_ring(tx_ring); -- 1.9.1 ^ permalink raw reply [flat|nested] 6+ messages in thread
[parent not found: <20190412122840.1908-1-g.singh@nxp.com>]
* [dpdk-stable] [PATCH v4 02/13] net/enetc: fix SMMU unhandled context fault [not found] ` <20190412122840.1908-1-g.singh@nxp.com> @ 2019-04-12 12:28 ` Gagandeep Singh 2019-04-12 12:28 ` [dpdk-stable] [PATCH v4 03/13] net/enetc: fix big endian build and correct buffer allocation Gagandeep Singh 2019-04-12 12:29 ` [dpdk-stable] [PATCH v4 13/13] net/enetc: fix crash at high speed traffic Gagandeep Singh 2 siblings, 0 replies; 6+ messages in thread From: Gagandeep Singh @ 2019-04-12 12:28 UTC (permalink / raw) To: dev, ferruh.yigit; +Cc: Gagandeep Singh, stable First configure ring with BDs properly then enable the ring. Fixes: 469c6111a799 ("net/enetc: enable Rx and Tx") Cc: stable@dpdk.org Signed-off-by: Gagandeep Singh <g.singh@nxp.com> --- drivers/net/enetc/enetc_ethdev.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/enetc/enetc_ethdev.c b/drivers/net/enetc/enetc_ethdev.c index 7a9a97deb..f1807b901 100644 --- a/drivers/net/enetc/enetc_ethdev.c +++ b/drivers/net/enetc/enetc_ethdev.c @@ -492,15 +492,15 @@ enetc_setup_rxbdr(struct enetc_hw *hw, struct enetc_bdr *rx_ring, ENETC_RTBLENR_LEN(rx_ring->bd_count)); rx_ring->mb_pool = mb_pool; - /* enable ring */ - enetc_rxbdr_wr(hw, idx, ENETC_RBMR, ENETC_RBMR_EN); - enetc_rxbdr_wr(hw, idx, ENETC_RBPIR, 0); rx_ring->rcir = (void *)((size_t)hw->reg + ENETC_BDR(RX, idx, ENETC_RBCIR)); enetc_refill_rx_ring(rx_ring, (enetc_bd_unused(rx_ring))); buf_size = (uint16_t)(rte_pktmbuf_data_room_size(rx_ring->mb_pool) - RTE_PKTMBUF_HEADROOM); enetc_rxbdr_wr(hw, idx, ENETC_RBBSR, buf_size); + /* enable ring */ + enetc_rxbdr_wr(hw, idx, ENETC_RBMR, ENETC_RBMR_EN); + enetc_rxbdr_wr(hw, idx, ENETC_RBPIR, 0); } static int -- 2.19.1 ^ permalink raw reply [flat|nested] 6+ messages in thread
* [dpdk-stable] [PATCH v4 03/13] net/enetc: fix big endian build and correct buffer allocation [not found] ` <20190412122840.1908-1-g.singh@nxp.com> 2019-04-12 12:28 ` [dpdk-stable] [PATCH v4 02/13] net/enetc: fix SMMU unhandled context fault Gagandeep Singh @ 2019-04-12 12:28 ` Gagandeep Singh 2019-04-12 12:29 ` [dpdk-stable] [PATCH v4 13/13] net/enetc: fix crash at high speed traffic Gagandeep Singh 2 siblings, 0 replies; 6+ messages in thread From: Gagandeep Singh @ 2019-04-12 12:28 UTC (permalink / raw) To: dev, ferruh.yigit; +Cc: Gagandeep Singh, stable There was an error at rte_constant_bswap64 while compiling with big endian toolchain. so fixing it by adding type cast. Also, rte_pktmbuf_alloc API should be used to allocate mbuf instead of rte_pktmbuf_raw_alloc to avoid use of stale mbuf information. Fixes: 469c6111a799 ("net/enetc: enable Rx and Tx") Cc: stable@dpdk.org Suggested-by: Thomas Monjalon <thomas@monjalon.net> Signed-off-by: Gagandeep Singh <g.singh@nxp.com> --- drivers/net/enetc/enetc_rxtx.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/enetc/enetc_rxtx.c b/drivers/net/enetc/enetc_rxtx.c index a31a38725..42f16cab1 100644 --- a/drivers/net/enetc/enetc_rxtx.c +++ b/drivers/net/enetc/enetc_rxtx.c @@ -88,8 +88,9 @@ enetc_refill_rx_ring(struct enetc_bdr *rx_ring, const int buff_cnt) rx_swbd = &rx_ring->q_swbd[i]; rxbd = ENETC_RXBD(*rx_ring, i); for (j = 0; j < buff_cnt; j++) { - rx_swbd->buffer_addr = - rte_cpu_to_le_64(rte_mbuf_raw_alloc(rx_ring->mb_pool)); + rx_swbd->buffer_addr = (void *)(uintptr_t) + rte_cpu_to_le_64((uint64_t)(uintptr_t) + rte_pktmbuf_alloc(rx_ring->mb_pool)); rxbd->w.addr = (uint64_t)(uintptr_t) rx_swbd->buffer_addr->buf_iova + rx_swbd->buffer_addr->data_off; -- 2.19.1 ^ permalink raw reply [flat|nested] 6+ messages in thread
* [dpdk-stable] [PATCH v4 13/13] net/enetc: fix crash at high speed traffic [not found] ` <20190412122840.1908-1-g.singh@nxp.com> 2019-04-12 12:28 ` [dpdk-stable] [PATCH v4 02/13] net/enetc: fix SMMU unhandled context fault Gagandeep Singh 2019-04-12 12:28 ` [dpdk-stable] [PATCH v4 03/13] net/enetc: fix big endian build and correct buffer allocation Gagandeep Singh @ 2019-04-12 12:29 ` Gagandeep Singh 2 siblings, 0 replies; 6+ messages in thread From: Gagandeep Singh @ 2019-04-12 12:29 UTC (permalink / raw) To: dev, ferruh.yigit; +Cc: Gagandeep Singh, stable On xmit side, there should be a check whether BD ring has free BDs available before transmit a packet to avoid data corruption and buffer leak issue. Fixes: 469c6111a799 ("net/enetc: enable Rx and Tx") Cc: stable@dpdk.org Signed-off-by: Gagandeep Singh <g.singh@nxp.com> --- drivers/net/enetc/enetc_rxtx.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/enetc/enetc_rxtx.c b/drivers/net/enetc/enetc_rxtx.c index 0ce7dbee7..81b0ef3b1 100644 --- a/drivers/net/enetc/enetc_rxtx.c +++ b/drivers/net/enetc/enetc_rxtx.c @@ -49,11 +49,16 @@ enetc_xmit_pkts(void *tx_queue, uint16_t nb_pkts) { struct enetc_swbd *tx_swbd; - int i, start; + int i, start, bds_to_use; struct enetc_tx_bd *txbd; struct enetc_bdr *tx_ring = (struct enetc_bdr *)tx_queue; i = tx_ring->next_to_use; + + bds_to_use = enetc_bd_unused(tx_ring); + if (bds_to_use < nb_pkts) + nb_pkts = bds_to_use; + start = 0; while (nb_pkts--) { enetc_clean_tx_ring(tx_ring); -- 2.19.1 ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2019-04-12 12:30 UTC | newest] Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- [not found] <1555075598-4988-1-git-send-email-g.singh@nxp.com> [not found] ` <20190412105105.24351-1-g.singh@nxp.com> 2019-04-12 11:01 ` [dpdk-stable] [PATCH v3 02/13] net/enetc: fix SMMU unhandled context fault Gagandeep Singh 2019-04-12 11:01 ` [dpdk-stable] [PATCH v3 03/13] net/enetc: fix big endian build and correct buffer allocation Gagandeep Singh 2019-04-12 11:01 ` [dpdk-stable] [PATCH v3 13/13] net/enetc: fix crash at high speed traffic Gagandeep Singh [not found] ` <20190412122840.1908-1-g.singh@nxp.com> 2019-04-12 12:28 ` [dpdk-stable] [PATCH v4 02/13] net/enetc: fix SMMU unhandled context fault Gagandeep Singh 2019-04-12 12:28 ` [dpdk-stable] [PATCH v4 03/13] net/enetc: fix big endian build and correct buffer allocation Gagandeep Singh 2019-04-12 12:29 ` [dpdk-stable] [PATCH v4 13/13] net/enetc: fix crash at high speed traffic Gagandeep Singh
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).