From: Olivier Matz <olivier.matz@6wind.com>
To: dev@dpdk.org, rkerur@gmail.com
Subject: [dpdk-dev] [PATCH v5] mbuf: introduce common functions to get dma address
Date: Fri, 4 Mar 2016 10:09:40 +0100 [thread overview]
Message-ID: <1457082580-19289-1-git-send-email-olivier.matz@6wind.com> (raw)
In-Reply-To: <1443700690-16524-1-git-send-email-rkerur@gmail.com>
From: Ravi Kerur <rkerur@gmail.com>
Macros RTE_MBUF_DATA_DMA_ADDR and RTE_MBUF_DATA_DMA_ADDR_DEFAULT
are defined in each PMD driver file. Convert macros to inline
functions and move them to common lib/librte_mbuf/rte_mbuf.h file.
PMD drivers include rte_mbuf.h file directly/indirectly hence no
additioanl header file inclusion is necessary.
Signed-off-by: Ravi Kerur <rkerur@gmail.com>
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
---
Sorry, I missed this patch since a long time. So, instead of doing
very late comments on the v4, I prefered to submit an updated v5 with
some minor fixes.
v5:
> Rebase on master branch (new occurences to replace in nfp
driver and in virtio)
> add API comments above the new functions
> use phys_addr_t instead of uint64_t
v4:
> Remove un-necessary cast and braces in inline functions.
> Add 'const' for function parameters
v3:
> Changed converted macro->inline names to lower-case
> Fix checkpatch.pl errors and warnings
(camelcase warning is not fixed)
> Compiled following targets
> x86_64-native-linuxapp-clang
> x86_64-native-linuxapp-gcc
> i686-native-linuxapp-gcc
> x86_64-native-bsdapp-gcc
> x86_64-native-bsdapp-clang
> Tested on:
> x86_64 ubuntu 14.04, make test
v2:
> Changed both macros to inline functions in all PMD
> Changed macro to rte_pktmbuf_mtod in xenvirt module
> Compiled following targets:
> x86_64-native-linuxapp-clang
> x86_64-native-linuxapp-gcc
> i686-native-linuxapp-gcc
> Tested on:
> x86_64 ubuntu 14.04, testpmd and make test
v1:
> Move macros into common rte_mbuf header file.
> Compiled following targets:
> x86_64-native-linuxapp-clang
> x86_64-native-linuxapp-gcc
> i686-native-linuxapp-gcc
> x86_64-native-bsdapp-gcc
> x86_64-native-bsdapp-clang
Tested on:
> x86_64 Ubuntu 14.04, testpmd and make test
> FreeBSD 10.1, testpmd
drivers/net/bnx2x/bnx2x.c | 2 +-
drivers/net/bnx2x/bnx2x.h | 3 ---
drivers/net/cxgbe/sge.c | 3 ---
drivers/net/e1000/em_rxtx.c | 15 +++++----------
drivers/net/e1000/igb_rxtx.c | 14 ++++----------
drivers/net/i40e/i40e_rxtx.c | 20 +++++++-------------
drivers/net/ixgbe/ixgbe_rxtx.c | 14 +++++++-------
drivers/net/ixgbe/ixgbe_rxtx.h | 6 ------
drivers/net/nfp/nfp_net.c | 2 +-
drivers/net/nfp/nfp_net_pmd.h | 3 ---
drivers/net/virtio/virtio_rxtx.c | 2 +-
drivers/net/virtio/virtio_rxtx_simple.c | 4 ++--
drivers/net/virtio/virtqueue.h | 3 ---
drivers/net/vmxnet3/vmxnet3_rxtx.c | 11 +++--------
drivers/net/xenvirt/virtqueue.h | 9 +--------
lib/librte_mbuf/rte_mbuf.h | 32 ++++++++++++++++++++++++++++++++
16 files changed, 64 insertions(+), 79 deletions(-)
diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
index 5a38c32..9d640da 100644
--- a/drivers/net/bnx2x/bnx2x.c
+++ b/drivers/net/bnx2x/bnx2x.c
@@ -2147,7 +2147,7 @@ int bnx2x_tx_encap(struct bnx2x_tx_queue *txq, struct rte_mbuf **m_head, int m_p
tx_start_bd = &txq->tx_ring[TX_BD(bd_prod, txq)].start_bd;
tx_start_bd->addr =
- rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR(m0));
+ rte_cpu_to_le_64(rte_mbuf_data_dma_addr(m0));
tx_start_bd->nbytes = rte_cpu_to_le_16(m0->data_len);
tx_start_bd->bd_flags.as_bitfield = ETH_TX_BD_FLAGS_START_BD;
tx_start_bd->general_data =
diff --git a/drivers/net/bnx2x/bnx2x.h b/drivers/net/bnx2x/bnx2x.h
index 1b5e26e..5f52450 100644
--- a/drivers/net/bnx2x/bnx2x.h
+++ b/drivers/net/bnx2x/bnx2x.h
@@ -128,9 +128,6 @@ struct bnx2x_device_type {
char *bnx2x_name;
};
-#define RTE_MBUF_DATA_DMA_ADDR(mb) \
- ((uint64_t)((mb)->buf_physaddr + (mb)->data_off))
-
#define BNX2X_PAGE_SHIFT 12
#define BNX2X_PAGE_SIZE (1 << BNX2X_PAGE_SHIFT)
#define BNX2X_PAGE_MASK (~(BNX2X_PAGE_SIZE - 1))
diff --git a/drivers/net/cxgbe/sge.c b/drivers/net/cxgbe/sge.c
index 3c62d03..ab5a842 100644
--- a/drivers/net/cxgbe/sge.c
+++ b/drivers/net/cxgbe/sge.c
@@ -1332,9 +1332,6 @@ static struct rte_mbuf *t4_pktgl_to_mbuf(const struct pkt_gl *gl)
return t4_pktgl_to_mbuf_usembufs(gl);
}
-#define RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mb) \
- ((dma_addr_t) ((mb)->buf_physaddr + (mb)->data_off))
-
/**
* t4_ethrx_handler - process an ingress ethernet packet
* @q: the response queue that received the packet
diff --git a/drivers/net/e1000/em_rxtx.c b/drivers/net/e1000/em_rxtx.c
index 7ef965c..441ccad 100644
--- a/drivers/net/e1000/em_rxtx.c
+++ b/drivers/net/e1000/em_rxtx.c
@@ -88,12 +88,6 @@ rte_rxmbuf_alloc(struct rte_mempool *mp)
return m;
}
-#define RTE_MBUF_DATA_DMA_ADDR(mb) \
- (uint64_t) ((mb)->buf_physaddr + (mb)->data_off)
-
-#define RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mb) \
- (uint64_t) ((mb)->buf_physaddr + RTE_PKTMBUF_HEADROOM)
-
/**
* Structure associated with each descriptor of the RX ring of a RX queue.
*/
@@ -585,7 +579,7 @@ eth_em_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
* Set up Transmit Data Descriptor.
*/
slen = m_seg->data_len;
- buf_dma_addr = RTE_MBUF_DATA_DMA_ADDR(m_seg);
+ buf_dma_addr = rte_mbuf_data_dma_addr(m_seg);
txd->buffer_addr = rte_cpu_to_le_64(buf_dma_addr);
txd->lower.data = rte_cpu_to_le_32(cmd_type_len | slen);
@@ -769,7 +763,7 @@ eth_em_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
rxm = rxe->mbuf;
rxe->mbuf = nmb;
dma_addr =
- rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(nmb));
+ rte_cpu_to_le_64(rte_mbuf_data_dma_addr_default(nmb));
rxdp->buffer_addr = dma_addr;
rxdp->status = 0;
@@ -949,7 +943,7 @@ eth_em_recv_scattered_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
*/
rxm = rxe->mbuf;
rxe->mbuf = nmb;
- dma = rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(nmb));
+ dma = rte_cpu_to_le_64(rte_mbuf_data_dma_addr_default(nmb));
rxdp->buffer_addr = dma;
rxdp->status = 0;
@@ -1575,7 +1569,8 @@ em_alloc_rx_queue_mbufs(struct em_rx_queue *rxq)
return -ENOMEM;
}
- dma_addr = rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mbuf));
+ dma_addr =
+ rte_cpu_to_le_64(rte_mbuf_data_dma_addr_default(mbuf));
/* Clear HW ring memory */
rxq->rx_ring[i] = rxd_init;
diff --git a/drivers/net/e1000/igb_rxtx.c b/drivers/net/e1000/igb_rxtx.c
index e7c30b7..74587b5 100644
--- a/drivers/net/e1000/igb_rxtx.c
+++ b/drivers/net/e1000/igb_rxtx.c
@@ -89,12 +89,6 @@ rte_rxmbuf_alloc(struct rte_mempool *mp)
return m;
}
-#define RTE_MBUF_DATA_DMA_ADDR(mb) \
- (uint64_t) ((mb)->buf_physaddr + (mb)->data_off)
-
-#define RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mb) \
- (uint64_t) ((mb)->buf_physaddr + RTE_PKTMBUF_HEADROOM)
-
/**
* Structure associated with each descriptor of the RX ring of a RX queue.
*/
@@ -596,7 +590,7 @@ eth_igb_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
* Set up transmit descriptor.
*/
slen = (uint16_t) m_seg->data_len;
- buf_dma_addr = RTE_MBUF_DATA_DMA_ADDR(m_seg);
+ buf_dma_addr = rte_mbuf_data_dma_addr(m_seg);
txd->read.buffer_addr =
rte_cpu_to_le_64(buf_dma_addr);
txd->read.cmd_type_len =
@@ -875,7 +869,7 @@ eth_igb_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
rxm = rxe->mbuf;
rxe->mbuf = nmb;
dma_addr =
- rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(nmb));
+ rte_cpu_to_le_64(rte_mbuf_data_dma_addr_default(nmb));
rxdp->read.hdr_addr = 0;
rxdp->read.pkt_addr = dma_addr;
@@ -1061,7 +1055,7 @@ eth_igb_recv_scattered_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
*/
rxm = rxe->mbuf;
rxe->mbuf = nmb;
- dma = rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(nmb));
+ dma = rte_cpu_to_le_64(rte_mbuf_data_dma_addr_default(nmb));
rxdp->read.pkt_addr = dma;
rxdp->read.hdr_addr = 0;
@@ -1970,7 +1964,7 @@ igb_alloc_rx_queue_mbufs(struct igb_rx_queue *rxq)
return -ENOMEM;
}
dma_addr =
- rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mbuf));
+ rte_cpu_to_le_64(rte_mbuf_data_dma_addr_default(mbuf));
rxd = &rxq->rx_ring[i];
rxd->read.hdr_addr = 0;
rxd->read.pkt_addr = dma_addr;
diff --git a/drivers/net/i40e/i40e_rxtx.c b/drivers/net/i40e/i40e_rxtx.c
index 40cffc1..8931b8e 100644
--- a/drivers/net/i40e/i40e_rxtx.c
+++ b/drivers/net/i40e/i40e_rxtx.c
@@ -78,12 +78,6 @@
PKT_TX_L4_MASK | \
PKT_TX_OUTER_IP_CKSUM)
-#define RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mb) \
- (uint64_t) ((mb)->buf_physaddr + RTE_PKTMBUF_HEADROOM)
-
-#define RTE_MBUF_DATA_DMA_ADDR(mb) \
- ((uint64_t)((mb)->buf_physaddr + (mb)->data_off))
-
static uint16_t i40e_xmit_pkts_simple(void *tx_queue,
struct rte_mbuf **tx_pkts,
uint16_t nb_pkts);
@@ -1098,7 +1092,7 @@ i40e_rx_alloc_bufs(struct i40e_rx_queue *rxq)
mb->nb_segs = 1;
mb->port = rxq->port_id;
dma_addr = rte_cpu_to_le_64(\
- RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mb));
+ rte_mbuf_data_dma_addr_default(mb));
rxdp[i].read.hdr_addr = 0;
rxdp[i].read.pkt_addr = dma_addr;
}
@@ -1245,7 +1239,7 @@ i40e_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
rxm = rxe->mbuf;
rxe->mbuf = nmb;
dma_addr =
- rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(nmb));
+ rte_cpu_to_le_64(rte_mbuf_data_dma_addr_default(nmb));
rxdp->read.hdr_addr = 0;
rxdp->read.pkt_addr = dma_addr;
@@ -1356,7 +1350,7 @@ i40e_recv_scattered_pkts(void *rx_queue,
rxm = rxe->mbuf;
rxe->mbuf = nmb;
dma_addr =
- rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(nmb));
+ rte_cpu_to_le_64(rte_mbuf_data_dma_addr_default(nmb));
/* Set data buffer address and data length of the mbuf */
rxdp->read.hdr_addr = 0;
@@ -1691,7 +1685,7 @@ i40e_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
/* Setup TX Descriptor */
slen = m_seg->data_len;
- buf_dma_addr = RTE_MBUF_DATA_DMA_ADDR(m_seg);
+ buf_dma_addr = rte_mbuf_data_dma_addr(m_seg);
PMD_TX_LOG(DEBUG, "mbuf: %p, TDD[%u]:\n"
"buf_dma_addr: %#"PRIx64";\n"
@@ -1790,7 +1784,7 @@ tx4(volatile struct i40e_tx_desc *txdp, struct rte_mbuf **pkts)
uint32_t i;
for (i = 0; i < 4; i++, txdp++, pkts++) {
- dma_addr = RTE_MBUF_DATA_DMA_ADDR(*pkts);
+ dma_addr = rte_mbuf_data_dma_addr(*pkts);
txdp->buffer_addr = rte_cpu_to_le_64(dma_addr);
txdp->cmd_type_offset_bsz =
i40e_build_ctob((uint32_t)I40E_TD_CMD, 0,
@@ -1804,7 +1798,7 @@ tx1(volatile struct i40e_tx_desc *txdp, struct rte_mbuf **pkts)
{
uint64_t dma_addr;
- dma_addr = RTE_MBUF_DATA_DMA_ADDR(*pkts);
+ dma_addr = rte_mbuf_data_dma_addr(*pkts);
txdp->buffer_addr = rte_cpu_to_le_64(dma_addr);
txdp->cmd_type_offset_bsz =
i40e_build_ctob((uint32_t)I40E_TD_CMD, 0,
@@ -2741,7 +2735,7 @@ i40e_alloc_rx_queue_mbufs(struct i40e_rx_queue *rxq)
mbuf->port = rxq->port_id;
dma_addr =
- rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mbuf));
+ rte_cpu_to_le_64(rte_mbuf_data_dma_addr_default(mbuf));
rxd = &rxq->rx_ring[i];
rxd->read.pkt_addr = dma_addr;
diff --git a/drivers/net/ixgbe/ixgbe_rxtx.c b/drivers/net/ixgbe/ixgbe_rxtx.c
index e95e6b7..54278ce 100644
--- a/drivers/net/ixgbe/ixgbe_rxtx.c
+++ b/drivers/net/ixgbe/ixgbe_rxtx.c
@@ -171,7 +171,7 @@ tx4(volatile union ixgbe_adv_tx_desc *txdp, struct rte_mbuf **pkts)
int i;
for (i = 0; i < 4; ++i, ++txdp, ++pkts) {
- buf_dma_addr = RTE_MBUF_DATA_DMA_ADDR(*pkts);
+ buf_dma_addr = rte_mbuf_data_dma_addr(*pkts);
pkt_len = (*pkts)->data_len;
/* write data to descriptor */
@@ -194,7 +194,7 @@ tx1(volatile union ixgbe_adv_tx_desc *txdp, struct rte_mbuf **pkts)
uint64_t buf_dma_addr;
uint32_t pkt_len;
- buf_dma_addr = RTE_MBUF_DATA_DMA_ADDR(*pkts);
+ buf_dma_addr = rte_mbuf_data_dma_addr(*pkts);
pkt_len = (*pkts)->data_len;
/* write data to descriptor */
@@ -816,7 +816,7 @@ ixgbe_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
* Set up Transmit Data Descriptor.
*/
slen = m_seg->data_len;
- buf_dma_addr = RTE_MBUF_DATA_DMA_ADDR(m_seg);
+ buf_dma_addr = rte_mbuf_data_dma_addr(m_seg);
txd->read.buffer_addr =
rte_cpu_to_le_64(buf_dma_addr);
txd->read.cmd_type_len =
@@ -1152,7 +1152,7 @@ ixgbe_rx_alloc_bufs(struct ixgbe_rx_queue *rxq, bool reset_mbuf)
mb->data_off = RTE_PKTMBUF_HEADROOM;
/* populate the descriptors */
- dma_addr = rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mb));
+ dma_addr = rte_cpu_to_le_64(rte_mbuf_data_dma_addr_default(mb));
rxdp[i].read.hdr_addr = 0;
rxdp[i].read.pkt_addr = dma_addr;
}
@@ -1379,7 +1379,7 @@ ixgbe_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
rxm = rxe->mbuf;
rxe->mbuf = nmb;
dma_addr =
- rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(nmb));
+ rte_cpu_to_le_64(rte_mbuf_data_dma_addr_default(nmb));
rxdp->read.hdr_addr = 0;
rxdp->read.pkt_addr = dma_addr;
@@ -1672,7 +1672,7 @@ next_desc:
if (!bulk_alloc) {
__le64 dma =
- rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(nmb));
+ rte_cpu_to_le_64(rte_mbuf_data_dma_addr_default(nmb));
/*
* Update RX descriptor with the physical address of the
* new data buffer of the new allocated mbuf.
@@ -3594,7 +3594,7 @@ ixgbe_alloc_rx_queue_mbufs(struct ixgbe_rx_queue *rxq)
mbuf->port = rxq->port_id;
dma_addr =
- rte_cpu_to_le_64(RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mbuf));
+ rte_cpu_to_le_64(rte_mbuf_data_dma_addr_default(mbuf));
rxd = &rxq->rx_ring[i];
rxd->read.hdr_addr = 0;
rxd->read.pkt_addr = dma_addr;
diff --git a/drivers/net/ixgbe/ixgbe_rxtx.h b/drivers/net/ixgbe/ixgbe_rxtx.h
index 475a800..42eb449 100644
--- a/drivers/net/ixgbe/ixgbe_rxtx.h
+++ b/drivers/net/ixgbe/ixgbe_rxtx.h
@@ -61,12 +61,6 @@
#define RTE_IXGBE_DESCS_PER_LOOP 4
-#define RTE_MBUF_DATA_DMA_ADDR(mb) \
- (uint64_t) ((mb)->buf_physaddr + (mb)->data_off)
-
-#define RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mb) \
- (uint64_t) ((mb)->buf_physaddr + RTE_PKTMBUF_HEADROOM)
-
#ifdef RTE_IXGBE_INC_VECTOR
#define RTE_IXGBE_RXQ_REARM_THRESH 32
#define RTE_IXGBE_MAX_RX_BURST RTE_IXGBE_RXQ_REARM_THRESH
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index fd4dd39..9c4f218 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -1985,7 +1985,7 @@ nfp_net_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
rte_pktmbuf_free_seg(*lmbuf);
dma_size = pkt->data_len;
- dma_addr = RTE_MBUF_DATA_DMA_ADDR(pkt);
+ dma_addr = rte_mbuf_data_dma_addr(pkt);
PMD_TX_LOG(DEBUG, "Working with mbuf at dma address:"
"%" PRIx64 "\n", dma_addr);
diff --git a/drivers/net/nfp/nfp_net_pmd.h b/drivers/net/nfp/nfp_net_pmd.h
index a7f9386..232ce5c 100644
--- a/drivers/net/nfp/nfp_net_pmd.h
+++ b/drivers/net/nfp/nfp_net_pmd.h
@@ -75,9 +75,6 @@ struct nfp_net_adapter;
/* Interrupt definitions */
#define NFP_NET_IRQ_LSC_IDX 0
-#define RTE_MBUF_DATA_DMA_ADDR(mb) \
- ((uint64_t)((mb)->buf_physaddr + (mb)->data_off))
-
/* Default values for RX/TX configuration */
#define DEFAULT_RX_FREE_THRESH 32
#define DEFAULT_RX_PTHRESH 8
diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net/virtio/virtio_rxtx.c
index ec0b8de..e96352c 100644
--- a/drivers/net/virtio/virtio_rxtx.c
+++ b/drivers/net/virtio/virtio_rxtx.c
@@ -239,7 +239,7 @@ virtqueue_enqueue_xmit(struct virtqueue *txvq, struct rte_mbuf *cookie)
for (; ((seg_num > 0) && (cookie != NULL)); seg_num--) {
idx = start_dp[idx].next;
- start_dp[idx].addr = RTE_MBUF_DATA_DMA_ADDR(cookie);
+ start_dp[idx].addr = rte_mbuf_data_dma_addr(cookie);
start_dp[idx].len = cookie->data_len;
start_dp[idx].flags = VRING_DESC_F_NEXT;
cookie = cookie->next;
diff --git a/drivers/net/virtio/virtio_rxtx_simple.c b/drivers/net/virtio/virtio_rxtx_simple.c
index 3a1de9d..8f5293d 100644
--- a/drivers/net/virtio/virtio_rxtx_simple.c
+++ b/drivers/net/virtio/virtio_rxtx_simple.c
@@ -366,7 +366,7 @@ virtio_xmit_pkts_simple(void *tx_queue, struct rte_mbuf **tx_pkts,
txvq->vq_descx[desc_idx + i].cookie = tx_pkts[i];
for (i = 0; i < nb_tail; i++) {
start_dp[desc_idx].addr =
- RTE_MBUF_DATA_DMA_ADDR(*tx_pkts);
+ rte_mbuf_data_dma_addr(*tx_pkts);
start_dp[desc_idx].len = (*tx_pkts)->pkt_len;
tx_pkts++;
desc_idx++;
@@ -377,7 +377,7 @@ virtio_xmit_pkts_simple(void *tx_queue, struct rte_mbuf **tx_pkts,
for (i = 0; i < nb_commit; i++)
txvq->vq_descx[desc_idx + i].cookie = tx_pkts[i];
for (i = 0; i < nb_commit; i++) {
- start_dp[desc_idx].addr = RTE_MBUF_DATA_DMA_ADDR(*tx_pkts);
+ start_dp[desc_idx].addr = rte_mbuf_data_dma_addr(*tx_pkts);
start_dp[desc_idx].len = (*tx_pkts)->pkt_len;
tx_pkts++;
desc_idx++;
diff --git a/drivers/net/virtio/virtqueue.h b/drivers/net/virtio/virtqueue.h
index 99d4fa9..68e0b4b 100644
--- a/drivers/net/virtio/virtqueue.h
+++ b/drivers/net/virtio/virtqueue.h
@@ -66,9 +66,6 @@ struct rte_mbuf;
#define VIRTQUEUE_MAX_NAME_SZ 32
-#define RTE_MBUF_DATA_DMA_ADDR(mb) \
- (uint64_t) ((mb)->buf_physaddr + (mb)->data_off)
-
#define VTNET_SQ_RQ_QUEUE_IDX 0
#define VTNET_SQ_TQ_QUEUE_IDX 1
#define VTNET_SQ_CQ_QUEUE_IDX 2
diff --git a/drivers/net/vmxnet3/vmxnet3_rxtx.c b/drivers/net/vmxnet3/vmxnet3_rxtx.c
index 4de5d89..8385478 100644
--- a/drivers/net/vmxnet3/vmxnet3_rxtx.c
+++ b/drivers/net/vmxnet3/vmxnet3_rxtx.c
@@ -77,12 +77,6 @@
#include "vmxnet3_logs.h"
#include "vmxnet3_ethdev.h"
-#define RTE_MBUF_DATA_DMA_ADDR(mb) \
- (uint64_t) ((mb)->buf_physaddr + (mb)->data_off)
-
-#define RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mb) \
- (uint64_t) ((mb)->buf_physaddr + RTE_PKTMBUF_HEADROOM)
-
static const uint32_t rxprod_reg[2] = {VMXNET3_REG_RXPROD, VMXNET3_REG_RXPROD2};
static int vmxnet3_post_rx_bufs(vmxnet3_rx_queue_t*, uint8_t);
@@ -377,7 +371,7 @@ vmxnet3_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
transmit buffer size (16K) is greater than
maximum sizeof mbuf segment size. */
gdesc = txq->cmd_ring.base + txq->cmd_ring.next2fill;
- gdesc->txd.addr = RTE_MBUF_DATA_DMA_ADDR(m_seg);
+ gdesc->txd.addr = rte_mbuf_data_dma_addr(m_seg);
gdesc->dword[2] = dw2 | m_seg->data_len;
gdesc->dword[3] = 0;
@@ -475,7 +469,8 @@ vmxnet3_post_rx_bufs(vmxnet3_rx_queue_t *rxq, uint8_t ring_id)
buf_info->m = mbuf;
buf_info->len = (uint16_t)(mbuf->buf_len -
RTE_PKTMBUF_HEADROOM);
- buf_info->bufPA = RTE_MBUF_DATA_DMA_ADDR_DEFAULT(mbuf);
+ buf_info->bufPA =
+ rte_mbuf_data_dma_addr_default(mbuf);
/* Load Rx Descriptor with the buffer's GPA */
rxd->addr = buf_info->bufPA;
diff --git a/drivers/net/xenvirt/virtqueue.h b/drivers/net/xenvirt/virtqueue.h
index 1d1bbcc..e425a04 100644
--- a/drivers/net/xenvirt/virtqueue.h
+++ b/drivers/net/xenvirt/virtqueue.h
@@ -50,13 +50,6 @@ struct rte_mbuf;
/* The alignment to use between consumer and producer parts of vring. */
#define VIRTIO_PCI_VRING_ALIGN 4096
-/*
- * Address translatio is between gva<->hva,
- * rather than gpa<->hva in virito spec.
- */
-#define RTE_MBUF_DATA_DMA_ADDR(mb) \
- ((uint64_t)(uintptr_t)rte_pktmbuf_mtod(mb, void *))
-
enum { VTNET_RQ = 0, VTNET_TQ = 1, VTNET_CQ = 2 };
/**
@@ -238,7 +231,7 @@ virtqueue_enqueue_xmit(struct virtqueue *txvq, struct rte_mbuf *cookie)
start_dp[idx].flags = VRING_DESC_F_NEXT;
start_dp[idx].addr = (uintptr_t)NULL;
idx = start_dp[idx].next;
- start_dp[idx].addr = RTE_MBUF_DATA_DMA_ADDR(cookie);
+ start_dp[idx].addr = rte_pktmbuf_mtod(cookie, uint64_t);
start_dp[idx].len = cookie->data_len;
start_dp[idx].flags = 0;
idx = start_dp[idx].next;
diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
index c1f6bc4..e39ad28 100644
--- a/lib/librte_mbuf/rte_mbuf.h
+++ b/lib/librte_mbuf/rte_mbuf.h
@@ -855,6 +855,38 @@ struct rte_mbuf {
static inline uint16_t rte_pktmbuf_priv_size(struct rte_mempool *mp);
/**
+ * Return the DMA address of the beginning of the mbuf data
+ *
+ * @param mb
+ * The pointer to the mbuf.
+ * @return
+ * The physical address of the beginning of the mbuf data
+ */
+static inline phys_addr_t
+rte_mbuf_data_dma_addr(const struct rte_mbuf *mb)
+{
+ return mb->buf_physaddr + mb->data_off;
+}
+
+/**
+ * Return the default DMA address of the beginning of the mbuf data
+ *
+ * This function is used by drivers in their receive function, as it
+ * returns the location where data should be written by the NIC, taking
+ * the default headroom in account.
+ *
+ * @param mb
+ * The pointer to the mbuf.
+ * @return
+ * The physical address of the beginning of the mbuf data
+ */
+static inline phys_addr_t
+rte_mbuf_data_dma_addr_default(const struct rte_mbuf *mb)
+{
+ return mb->buf_physaddr + RTE_PKTMBUF_HEADROOM;
+}
+
+/**
* Return the mbuf owning the data buffer address of an indirect mbuf.
*
* @param mi
--
2.1.4
next prev parent reply other threads:[~2016-03-04 9:10 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-01 11:58 [dpdk-dev] [PATCH v4] Move rte_mbuf macros to common header file Ravi Kerur
2016-03-04 9:09 ` Olivier Matz [this message]
2016-03-04 14:52 ` [dpdk-dev] [PATCH v5] mbuf: introduce common functions to get dma address Thomas Monjalon
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1457082580-19289-1-git-send-email-olivier.matz@6wind.com \
--to=olivier.matz@6wind.com \
--cc=dev@dpdk.org \
--cc=rkerur@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).