Re: [dpdk-dev] [PATCH 10/12] vhost: support to kick in secondary process

[Tan, Jianfeng]

> -Original Message-
> From: Yuanhan Liu [mailto:y...@fridaylinux.org]
> Sent: Wednesday, September 27, 2017 5:36 PM
> To: Tan, Jianfeng
> Cc: dev@dpdk.org; maxime.coque...@redhat.com; mtetsu...@gmail.com
> Subject: Re: [PATCH 10/12] vhost: support to kick in secondary process
> 
> On Fri, Sep 22, 2017 at 02:30:21AM +, Tan, Jianfeng wrote:
> >
> >
> > > -Original Message-
> > > From: Yuanhan Liu [mailto:y...@fridaylinux.org]
> > > Sent: Thursday, September 21, 2017 5:18 PM
> > > To: Tan, Jianfeng
> > > Cc: dev@dpdk.org; maxime.coque...@redhat.com;
> mtetsu...@gmail.com
> > > Subject: Re: [PATCH 10/12] vhost: support to kick in secondary process
> > >
> > > On Thu, Sep 21, 2017 at 03:04:39PM +0800, Tan, Jianfeng wrote:
> > > > >On Fri, Aug 25, 2017 at 09:40:50AM +, Jianfeng Tan wrote:
> > > > >>To support kick in secondary process, we propose callfd_pri and
> > > > >>kickfd_pri to store the value in primary process; and by a new
> > > > >>API, rte_vhost_set_vring_effective_fd(), we can set effective
> > > > >>callfd and kickfd which can be used by secondary process.
> > > > >>
> > > > >>Note in this case, either primary process or the secondary process
> > > > >>can kick the frontend; that is, they cannot kick a vring at the
> > > > >>same time.
> > > > >Since only one can work, why not just overwriting the fd? Say, you
> > > > >could introudce some APIs like "rte_vhost_set_vring_callfd", then
> > > > >you don't need to introduce few more fields like "callfd_pri".
> > > >
> > > > That cannot address the below case:
> > > > 1. Primary starts;
> > > > 2. Secondary one starts; (if we overwrite it without storing it in some
> > > > other fields)
> > > > 3. Secondary one exits;
> > > > 4. Secondary two starts. (primary cannot share the fd with this
> secondary
> > > > process now, as this fd does not mean anything to the primary process)
> > >
> > > I was thinking that those fds will be retrieved by the primary process
> > > once? So thsoe it got at beginning are still valid?
> >
> > Yes, the FDs are valid to primary process at step 1. But After overwriting 
> > in
> step 2, those FDs are not valid to primary.
> 
> Yes, but the primary process has already got its correct fds saved, right?
> With the saved fds, it then could share it with another secondary process.
> 
> Actually, the key (and typical) issue of multi-process here is the fds are
> process specific, while they are stored in the shared memory. That means
> only one will take effect eventually. Worse, the old ones are lost.
> 
> So, I think to make it right in this case, you should move the fds from
> the shared memory and store them in the memory of the corresponding
> process.
> If that's done, all processes could have its own valid fds, then every
> process could do the kick (if that's really necessary).
> 
> You could check following commit for more info.
> 553f45932fb7 ("net/virtio: store PCI operators pointer locally")

Have referred to the above solution, but seems not feasible for this case since 
there are too many queues. For example, if we define an array like this:
  int vhost_callfds[index_by_vid][index_by_queue_id];
The size would be MAX_VHOST_DEVICE * VHOST_MAX_VRING * 8Byte = 2Mbyte.

Instead, can we propose something like process_id to index array located at 
shared memory?

Thanks,
Jianfeng

[dpdk-dev] [PATCH v5 2/8] net/i40e: update ptype and pctype info

[Beilei Xing]

Update new packet type and new pctype info when downloading
profile.

Signed-off-by: Beilei Xing 
---
 drivers/net/i40e/i40e_ethdev.c  | 312 
 drivers/net/i40e/i40e_ethdev.h  |  24 
 drivers/net/i40e/rte_pmd_i40e.c |   6 +-
 3 files changed, 341 insertions(+), 1 deletion(-)

diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 720f067..d6b0d50 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -65,6 +65,7 @@
 #include "i40e_rxtx.h"
 #include "i40e_pf.h"
 #include "i40e_regs.h"
+#include "rte_pmd_i40e.h"
 
 #define ETH_I40E_FLOATING_VEB_ARG  "enable_floating_veb"
 #define ETH_I40E_FLOATING_VEB_LIST_ARG "floating_veb_list"
@@ -1036,6 +1037,21 @@ i40e_init_fdir_filter_list(struct rte_eth_dev *dev)
return ret;
 }
 
+static void
+i40e_init_customized_info(struct i40e_pf *pf)
+{
+   int i;
+
+   /* Initialize customized pctype */
+   for (i = I40E_CUSTOMIZED_GTPC; i < I40E_CUSTOMIZED_MAX; i++) {
+   pf->customized_pctype[i].index = i;
+   pf->customized_pctype[i].pctype = I40E_FILTER_PCTYPE_INVALID;
+   pf->customized_pctype[i].valid = false;
+   }
+
+   pf->gtp_support = false;
+}
+
 static int
 eth_i40e_dev_init(struct rte_eth_dev *dev)
 {
@@ -1301,6 +1317,9 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
/* initialize Traffic Manager configuration */
i40e_tm_conf_init(dev);
 
+   /* Initialize customized information */
+   i40e_init_customized_info(pf);
+
ret = i40e_init_ethtype_filter_list(dev);
if (ret < 0)
goto err_init_ethtype_filter_list;
@@ -10893,6 +10912,299 @@ is_i40e_supported(struct rte_eth_dev *dev)
return is_device_supported(dev, &rte_i40e_pmd);
 }
 
+struct i40e_customized_pctype*
+i40e_find_customized_pctype(struct i40e_pf *pf, uint8_t index)
+{
+   int i;
+
+   for (i = 0; i < I40E_CUSTOMIZED_MAX; i++) {
+   if (pf->customized_pctype[i].index == index)
+   return &pf->customized_pctype[i];
+   }
+   return NULL;
+}
+
+static int
+i40e_update_customized_pctype(struct rte_eth_dev *dev, uint8_t *pkg,
+ uint32_t pkg_size, uint32_t proto_num,
+ struct rte_pmd_i40e_proto_info *proto)
+{
+   struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+   uint32_t pctype_num;
+   struct rte_pmd_i40e_ptype_info *pctype;
+   uint32_t buff_size;
+   struct i40e_customized_pctype *new_pctype = NULL;
+   uint8_t proto_id;
+   uint8_t pctype_value;
+   char name[64];
+   uint32_t i, j, n;
+   int ret;
+
+   ret = rte_pmd_i40e_get_ddp_info(pkg, pkg_size,
+   (uint8_t *)&pctype_num, sizeof(pctype_num),
+   RTE_PMD_I40E_PKG_INFO_PCTYPE_NUM);
+   if (ret) {
+   PMD_DRV_LOG(ERR, "Failed to get pctype number");
+   return -1;
+   }
+   if (!pctype_num) {
+   PMD_DRV_LOG(INFO, "No new pctype added");
+   return -1;
+   }
+
+   buff_size = pctype_num * sizeof(struct rte_pmd_i40e_proto_info);
+   pctype = rte_zmalloc("new_pctype", buff_size, 0);
+   if (!pctype) {
+   PMD_DRV_LOG(ERR, "Failed to allocate memory");
+   return -1;
+   }
+   /* get information about new pctype list */
+   ret = rte_pmd_i40e_get_ddp_info(pkg, pkg_size,
+   (uint8_t *)pctype, buff_size,
+   RTE_PMD_I40E_PKG_INFO_PCTYPE_LIST);
+   if (ret) {
+   PMD_DRV_LOG(ERR, "Failed to get pctype list");
+   rte_free(pctype);
+   return -1;
+   }
+
+   /* Update customized pctype. */
+   for (i = 0; i < pctype_num; i++) {
+   pctype_value = pctype[i].ptype_id;
+   memset(name, 0, sizeof(name));
+   for (j = 0; j < RTE_PMD_I40E_PROTO_NUM; j++) {
+   proto_id = pctype[i].protocols[j];
+   if (proto_id == RTE_PMD_I40E_PROTO_UNUSED)
+   continue;
+   for (n = 0; n < proto_num; n++) {
+   if (proto[n].proto_id != proto_id)
+   continue;
+   strcat(name, proto[n].name);
+   strcat(name, "_");
+   break;
+   }
+   }
+   name[strlen(name) - 1] = '\0';
+   if (!strcmp(name, "GTPC"))
+   new_pctype =
+   i40e_find_customized_pctype(pf,
+ I40E_CUSTOMIZED_GTPC);
+   else if (!strcmp(name, "GTPU_IPV4"))
+   new_pctype =
+   

[dpdk-dev] [PATCH v5 3/8] net/i40e: support RSS for new pctype

[Beilei Xing]

Enable RSS for new pctypes after downloading
new profile.

Signed-off-by: Beilei Xing 
---
 drivers/net/i40e/i40e_ethdev.c | 29 +
 1 file changed, 29 insertions(+)

diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index d6b0d50..aba35a5 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -1928,6 +1928,31 @@ i40e_apply_link_speed(struct rte_eth_dev *dev)
return i40e_phy_conf_link(hw, abilities, speed, true);
 }
 
+static void
+i40e_customized_pctype_hash_set(struct i40e_pf *pf, bool enable)
+{
+   struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+   uint64_t hena;
+   int i;
+
+   hena = (uint64_t)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(0));
+   hena |= ((uint64_t)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(1))) << 32;
+
+   for (i = 0; i < I40E_CUSTOMIZED_MAX; i++) {
+   if (pf->customized_pctype[i].valid) {
+   if (enable)
+   hena |= 1ULL << pf->customized_pctype[i].pctype;
+   else
+   hena &= ~(1ULL <<
+ pf->customized_pctype[i].pctype);
+   }
+   }
+
+   i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), (uint32_t)hena);
+   i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), (uint32_t)(hena >> 32));
+   I40E_WRITE_FLUSH(hw);
+}
+
 static int
 i40e_dev_start(struct rte_eth_dev *dev)
 {
@@ -2075,6 +2100,8 @@ i40e_dev_start(struct rte_eth_dev *dev)
"please call hierarchy_commit() "
"before starting the port");
 
+   i40e_customized_pctype_hash_set(pf, true);
+
return I40E_SUCCESS;
 
 err_up:
@@ -2155,6 +2182,8 @@ i40e_dev_close(struct rte_eth_dev *dev)
uint32_t reg;
int i;
 
+   i40e_customized_pctype_hash_set(pf, false);
+
PMD_INIT_FUNC_TRACE();
 
i40e_dev_stop(dev);
-- 
2.5.5

[dpdk-dev] [PATCH v5 1/8] mbuf: support GTP in software packet type parser

[Beilei Xing]

Add support of GTP-C and GTP-U tunnels in rte_net_get_ptype().

Signed-off-by: Beilei Xing 
Acked-by: Olivier Matz 
---
 lib/librte_mbuf/rte_mbuf_ptype.c |  2 ++
 lib/librte_mbuf/rte_mbuf_ptype.h | 24 
 2 files changed, 26 insertions(+)

diff --git a/lib/librte_mbuf/rte_mbuf_ptype.c b/lib/librte_mbuf/rte_mbuf_ptype.c
index e5c4fae..a450814 100644
--- a/lib/librte_mbuf/rte_mbuf_ptype.c
+++ b/lib/librte_mbuf/rte_mbuf_ptype.c
@@ -89,6 +89,8 @@ const char *rte_get_ptype_tunnel_name(uint32_t ptype)
case RTE_PTYPE_TUNNEL_NVGRE: return "TUNNEL_NVGRE";
case RTE_PTYPE_TUNNEL_GENEVE: return "TUNNEL_GENEVE";
case RTE_PTYPE_TUNNEL_GRENAT: return "TUNNEL_GRENAT";
+   case RTE_PTYPE_TUNNEL_GTPC: return "TUNNEL_GTPC";
+   case RTE_PTYPE_TUNNEL_GTPU: return "TUNNEL_GTPU";
default: return "TUNNEL_UNKNOWN";
}
 }
diff --git a/lib/librte_mbuf/rte_mbuf_ptype.h b/lib/librte_mbuf/rte_mbuf_ptype.h
index acd70bb..eb7cd2c 100644
--- a/lib/librte_mbuf/rte_mbuf_ptype.h
+++ b/lib/librte_mbuf/rte_mbuf_ptype.h
@@ -383,6 +383,30 @@ extern "C" {
  */
 #define RTE_PTYPE_TUNNEL_GRENAT 0x6000
 /**
+ * GTP-C (GPRS Tunnelling Protocol) control tunneling packet type.
+ * Packet format:
+ * <'ether type'=0x0800
+ * | 'version'=4, 'protocol'=17
+ * | 'destination port'=2123>
+ * or,
+ * <'ether type'=0x86DD
+ * | 'version'=6, 'next header'=17
+ * | 'destination port'=2123>
+ */
+#define RTE_PTYPE_TUNNEL_GTPC   0x7000
+/**
+ * GTP-U (GPRS Tunnelling Protocol) user data tunneling packet type.
+ * Packet format:
+ * <'ether type'=0x0800
+ * | 'version'=4, 'protocol'=17
+ * | 'destination port'=2152>
+ * or,
+ * <'ether type'=0x86DD
+ * | 'version'=6, 'next header'=17
+ * | 'destination port'=2152>
+ */
+#define RTE_PTYPE_TUNNEL_GTPU   0x8000
+/**
  * Mask of tunneling packet types.
  */
 #define RTE_PTYPE_TUNNEL_MASK   0xf000
-- 
2.5.5

[dpdk-dev] [PATCH v5 4/8] ethdev: add GTP items to support flow API

[Beilei Xing]

This patch adds GTP, GTPC and GTPU items for
generic flow API, and also exposes item fields
through the flow command.

Signed-off-by: Beilei Xing 
Acked-by: Adrien Mazarguil 
---
 app/test-pmd/cmdline_flow.c | 40 ++
 app/test-pmd/config.c   |  3 ++
 doc/guides/prog_guide/rte_flow.rst  | 18 ++
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |  4 +++
 lib/librte_ether/rte_flow.h | 52 +
 5 files changed, 117 insertions(+)

diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c
index a17a004..26c3e4f 100644
--- a/app/test-pmd/cmdline_flow.c
+++ b/app/test-pmd/cmdline_flow.c
@@ -171,6 +171,10 @@ enum index {
ITEM_GRE_PROTO,
ITEM_FUZZY,
ITEM_FUZZY_THRESH,
+   ITEM_GTP,
+   ITEM_GTP_TEID,
+   ITEM_GTPC,
+   ITEM_GTPU,
 
/* Validate/create actions. */
ACTIONS,
@@ -451,6 +455,9 @@ static const enum index next_item[] = {
ITEM_MPLS,
ITEM_GRE,
ITEM_FUZZY,
+   ITEM_GTP,
+   ITEM_GTPC,
+   ITEM_GTPU,
ZERO,
 };
 
@@ -588,6 +595,12 @@ static const enum index item_gre[] = {
ZERO,
 };
 
+static const enum index item_gtp[] = {
+   ITEM_GTP_TEID,
+   ITEM_NEXT,
+   ZERO,
+};
+
 static const enum index next_action[] = {
ACTION_END,
ACTION_VOID,
@@ -1421,6 +1434,33 @@ static const struct token token_list[] = {
.args = ARGS(ARGS_ENTRY(struct rte_flow_item_fuzzy,
thresh)),
},
+   [ITEM_GTP] = {
+   .name = "gtp",
+   .help = "match GTP header",
+   .priv = PRIV_ITEM(GTP, sizeof(struct rte_flow_item_gtp)),
+   .next = NEXT(item_gtp),
+   .call = parse_vc,
+   },
+   [ITEM_GTP_TEID] = {
+   .name = "teid",
+   .help = "tunnel endpoint identifier",
+   .next = NEXT(item_gtp, NEXT_ENTRY(UNSIGNED), item_param),
+   .args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_gtp, teid)),
+   },
+   [ITEM_GTPC] = {
+   .name = "gtpc",
+   .help = "match GTP header",
+   .priv = PRIV_ITEM(GTPC, sizeof(struct rte_flow_item_gtp)),
+   .next = NEXT(item_gtp),
+   .call = parse_vc,
+   },
+   [ITEM_GTPU] = {
+   .name = "gtpu",
+   .help = "match GTP header",
+   .priv = PRIV_ITEM(GTPU, sizeof(struct rte_flow_item_gtp)),
+   .next = NEXT(item_gtp),
+   .call = parse_vc,
+   },
 
/* Validate/create actions. */
[ACTIONS] = {
diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index e8e311c..9b09bbd 100644
--- a/app/test-pmd/config.c
+++ b/app/test-pmd/config.c
@@ -949,6 +949,9 @@ static const struct {
MK_FLOW_ITEM(MPLS, sizeof(struct rte_flow_item_mpls)),
MK_FLOW_ITEM(GRE, sizeof(struct rte_flow_item_gre)),
MK_FLOW_ITEM(FUZZY, sizeof(struct rte_flow_item_fuzzy)),
+   MK_FLOW_ITEM(GTP, sizeof(struct rte_flow_item_gtp)),
+   MK_FLOW_ITEM(GTPC, sizeof(struct rte_flow_item_gtp)),
+   MK_FLOW_ITEM(GTPU, sizeof(struct rte_flow_item_gtp)),
 };
 
 /** Compute storage space needed by item specification. */
diff --git a/doc/guides/prog_guide/rte_flow.rst 
b/doc/guides/prog_guide/rte_flow.rst
index 662a912..1bc8f19 100644
--- a/doc/guides/prog_guide/rte_flow.rst
+++ b/doc/guides/prog_guide/rte_flow.rst
@@ -955,6 +955,24 @@ Usage example, fuzzy match a TCPv4 packets:
| 4 | END  |
+---+--+
 
+Item: ``GTP``, ``GTPC``, ``GTPU``
+^
+
+Matches a GTP header.
+
+Note: GTP, GTPC and GTPU use the same structure. Since only UDP destination 
port
+is used to distinguish GTP_C (port is 2123) and GTP_U packets (port is 2152),
+GTPC and GTPU item are defined for a user-friendly API when creating GTP-C and
+GTP-U flow.
+
+- ``v_pt_rsv_flags``: version (3b), protocol type (1b), reserved (1b),
+  extension header flag (1b), sequence number flag (1b), N-PDU number
+  flag (1b).
+- ``msg_type``: message type.
+- ``msg_len``: message length.
+- ``teid``: tunnel endpoint identifier.
+- Default ``mask`` matches teid only.
+
 Actions
 ~~~
 
diff --git a/doc/guides/testpmd_app_ug/testpmd_funcs.rst 
b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
index 2ed62f5..8cc2399 100644
--- a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
+++ b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
@@ -2696,6 +2696,10 @@ This section lists supported pattern items and their 
attributes, if any.
 
   - ``thresh {unsigned}``: accuracy threshold.
 
+- ``gtp``, ``gtpc``, ``gtpu``: match GTP header.
+
+  - ``teid {unsigned}``: tunnel endpoint identifier.
+
 Actions list
 
 
diff --git a/lib/librte_ether/rte_flow.h b/lib/librte_ether/rte_flow.h
index bba6169..5da3aff 100644
--- a/lib/librte_ether/rte_flow.h
+++ b

[dpdk-dev] [PATCH v5 0/8] GPT-C and GTP-U enabling

[Beilei Xing]

This patch set enables RSS/FDIR/cloud filter for GPT-C and GTP-U.
It depends on Kirill's patch:
http://dpdk.org/ml/archives/dev/2017-September/076035.html.
However, Kirill's patchset needs to be updated.

v5 changes:
 - Fix code style.
 - Reword commit log.

v4 changes:
 - Refine fdir related code.
 - Rework profile metadata parsing function.
 - Fix code style.

v3 changes:
 - Rework implementation to support the new profile.
 - Add GTPC and GTPU tunnel type in software packet type parser.
 - Update ptype info when loading profile.
 - Fix bug of updating pctype info.


v2 changes:
 - Enable RSS/FDIR/cloud filter dinamicly by checking profile
 - Add GTPC and GTPU items to distinguish rule for GTP-C or GTP-U
 - Rework FDIR/cloud filter enabling function

Beilei Xing (8):
  mbuf: support GTP in software packet type parser
  net/i40e: update ptype and pctype info
  net/i40e: support RSS for new pctype
  ethdev: add GTP items to support flow API
  net/i40e: finish integration FDIR with generic flow API
  net/i40e: add FDIR support for GTP-C and GTP-U
  net/i40e: add cloud filter parsing function for GTP
  net/i40e: enable cloud filter for GTP-C and GTP-U

 app/test-pmd/cmdline_flow.c |  40 ++
 app/test-pmd/config.c   |   3 +
 doc/guides/prog_guide/rte_flow.rst  |  18 +
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |   4 +
 drivers/net/i40e/i40e_ethdev.c  | 534 +-
 drivers/net/i40e/i40e_ethdev.h  | 156 +++-
 drivers/net/i40e/i40e_fdir.c| 570 +++-
 drivers/net/i40e/i40e_flow.c| 496 
 drivers/net/i40e/rte_pmd_i40e.c |   6 +-
 lib/librte_ether/rte_flow.h |  52 +++
 lib/librte_mbuf/rte_mbuf_ptype.c|   2 +
 lib/librte_mbuf/rte_mbuf_ptype.h|  24 ++
 12 files changed, 1775 insertions(+), 130 deletions(-)

-- 
2.5.5

[dpdk-dev] [PATCH v5 5/8] net/i40e: finish integration FDIR with generic flow API

[Beilei Xing]

rte_eth_fdir_* structures are still used in FDIR functions.
This patch adds i40e private FDIR related structures and
functions to finish integration FDIR with generic flow API.

Signed-off-by: Beilei Xing 
---
 drivers/net/i40e/i40e_ethdev.h |  83 ++-
 drivers/net/i40e/i40e_fdir.c   | 488 +++--
 drivers/net/i40e/i40e_flow.c   |  76 +++
 3 files changed, 584 insertions(+), 63 deletions(-)

diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index 73fb5c3..4d690a1 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -461,6 +461,80 @@ struct i40e_vmdq_info {
 #define I40E_FDIR_IPv6_TC_OFFSET   20
 
 /*
+ * A union contains the inputs for all types of flow
+ * items in flows need to be in big endian
+ */
+union i40e_fdir_flow {
+   struct rte_eth_l2_flow l2_flow;
+   struct rte_eth_udpv4_flow  udp4_flow;
+   struct rte_eth_tcpv4_flow  tcp4_flow;
+   struct rte_eth_sctpv4_flow sctp4_flow;
+   struct rte_eth_ipv4_flow   ip4_flow;
+   struct rte_eth_udpv6_flow  udp6_flow;
+   struct rte_eth_tcpv6_flow  tcp6_flow;
+   struct rte_eth_sctpv6_flow sctp6_flow;
+   struct rte_eth_ipv6_flow   ipv6_flow;
+};
+
+/* A structure used to contain extend input of flow */
+struct i40e_fdir_flow_ext {
+   uint16_t vlan_tci;
+   uint8_t flexbytes[RTE_ETH_FDIR_MAX_FLEXLEN];
+   /* It is filled by the flexible payload to match. */
+   uint8_t is_vf;   /* 1 for VF, 0 for port dev */
+   uint16_t dst_id; /* VF ID, available when is_vf is 1*/
+};
+
+/* A structure used to define the input for a flow director filter entry */
+struct i40e_fdir_input {
+   enum i40e_filter_pctype pctype;
+   union i40e_fdir_flow flow;
+   /* Flow fields to match, dependent on flow_type */
+   struct i40e_fdir_flow_ext flow_ext;
+   /* Additional fields to match */
+};
+
+/* Behavior will be taken if FDIR match */
+enum i40e_fdir_behavior {
+   I40E_FDIR_ACCEPT = 0,
+   I40E_FDIR_REJECT,
+   I40E_FDIR_PASSTHRU,
+};
+
+/* Flow director report status
+ * It defines what will be reported if FDIR entry is matched.
+ */
+enum i40e_fdir_status {
+   I40E_FDIR_NO_REPORT_STATUS = 0, /* Report nothing. */
+   I40E_FDIR_REPORT_ID,/* Only report FD ID. */
+   I40E_FDIR_REPORT_ID_FLEX_4, /* Report FD ID and 4 flex bytes. */
+   I40E_FDIR_REPORT_FLEX_8,/* Report 8 flex bytes. */
+};
+
+/* A structure used to define an action when match FDIR packet filter. */
+struct i40e_fdir_action {
+   uint16_t rx_queue;/* Queue assigned to if FDIR match. */
+   enum i40e_fdir_behavior behavior; /* Behavior will be taken */
+   enum i40e_fdir_status report_status;  /* Status report option */
+   /* If report_status is I40E_FDIR_REPORT_ID_FLEX_4 or
+* I40E_FDIR_REPORT_FLEX_8, flex_off specifies where the reported
+* flex bytes start from in flexible payload.
+*/
+   uint8_t flex_off;
+};
+
+/* A structure used to define the flow director filter entry by filter_ctrl API
+ * It supports RTE_ETH_FILTER_FDIR with RTE_ETH_FILTER_ADD and
+ * RTE_ETH_FILTER_DELETE operations.
+ */
+struct i40e_fdir_filter_conf {
+   uint32_t soft_id;
+   /* ID, an unique value is required when deal with FDIR entry */
+   struct i40e_fdir_input input;/* Input set */
+   struct i40e_fdir_action action;  /* Action taken when match */
+};
+
+/*
  * Structure to store flex pit for flow diretor.
  */
 struct i40e_fdir_flex_pit {
@@ -483,7 +557,7 @@ struct i40e_fdir_flex_mask {
 
 struct i40e_fdir_filter {
TAILQ_ENTRY(i40e_fdir_filter) rules;
-   struct rte_eth_fdir_filter fdir;
+   struct i40e_fdir_filter_conf fdir;
 };
 
 TAILQ_HEAD(i40e_fdir_filter_list, i40e_fdir_filter);
@@ -907,7 +981,7 @@ extern const struct rte_flow_ops i40e_flow_ops;
 
 union i40e_filter_t {
struct rte_eth_ethertype_filter ethertype_filter;
-   struct rte_eth_fdir_filter fdir_filter;
+   struct i40e_fdir_filter_conf fdir_filter;
struct rte_eth_tunnel_filter_conf tunnel_filter;
struct i40e_tunnel_filter_conf consistent_tunnel_filter;
 };
@@ -981,7 +1055,7 @@ i40e_sw_ethertype_filter_lookup(struct i40e_ethertype_rule 
*ethertype_rule,
 int i40e_sw_ethertype_filter_del(struct i40e_pf *pf,
 struct i40e_ethertype_filter_input *input);
 int i40e_sw_fdir_filter_del(struct i40e_pf *pf,
-   struct rte_eth_fdir_input *input);
+   struct i40e_fdir_input *input);
 struct i40e_tunnel_filter *
 i40e_sw_tunnel_filter_lookup(struct i40e_tunnel_rule *tunnel_rule,
 const struct i40e_tunnel_filter_input *input);
@@ -994,6 +1068,9 @@ int i40e_ethertype_filter_set(struct i40e_pf *pf,
 int i40e_add_del_fdir_filter(struct rte_eth_dev *dev,
 const struct rte_eth_fdir_filter 

[dpdk-dev] [PATCH v5 6/8] net/i40e: add FDIR support for GTP-C and GTP-U

[Beilei Xing]

This patch adds FDIR support for GTP-C and GTP-U. The
input set of GTP-C and GTP-U is TEID.

Signed-off-by: Beilei Xing 
---
 drivers/net/i40e/i40e_ethdev.h |  30 +
 drivers/net/i40e/i40e_fdir.c   | 200 ++-
 drivers/net/i40e/i40e_flow.c   | 263 +++--
 3 files changed, 396 insertions(+), 97 deletions(-)

diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index 4d690a1..502f6c6 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -460,6 +460,25 @@ struct i40e_vmdq_info {
 #define I40E_FLEX_WORD_MASK(off) (0x80 >> (off))
 #define I40E_FDIR_IPv6_TC_OFFSET   20
 
+/* A structure used to define the input for GTP flow */
+struct i40e_gtp_flow {
+   struct rte_eth_udpv4_flow udp; /* IPv4 UDP fields to match. */
+   uint8_t msg_type;  /* Message type. */
+   uint32_t teid; /* TEID in big endian. */
+};
+
+/* A structure used to define the input for GTP IPV4 flow */
+struct i40e_gtp_ipv4_flow {
+   struct i40e_gtp_flow gtp;
+   struct rte_eth_ipv4_flow ip4;
+};
+
+/* A structure used to define the input for GTP IPV6 flow */
+struct i40e_gtp_ipv6_flow {
+   struct i40e_gtp_flow gtp;
+   struct rte_eth_ipv6_flow ip6;
+};
+
 /*
  * A union contains the inputs for all types of flow
  * items in flows need to be in big endian
@@ -474,6 +493,14 @@ union i40e_fdir_flow {
struct rte_eth_tcpv6_flow  tcp6_flow;
struct rte_eth_sctpv6_flow sctp6_flow;
struct rte_eth_ipv6_flow   ipv6_flow;
+   struct i40e_gtp_flow   gtp_flow;
+   struct i40e_gtp_ipv4_flow  gtp_ipv4_flow;
+   struct i40e_gtp_ipv6_flow  gtp_ipv6_flow;
+};
+
+enum i40e_fdir_ip_type {
+   I40E_FDIR_IPTYPE_IPV4,
+   I40E_FDIR_IPTYPE_IPV6,
 };
 
 /* A structure used to contain extend input of flow */
@@ -483,6 +510,9 @@ struct i40e_fdir_flow_ext {
/* It is filled by the flexible payload to match. */
uint8_t is_vf;   /* 1 for VF, 0 for port dev */
uint16_t dst_id; /* VF ID, available when is_vf is 1*/
+   bool inner_ip;   /* If there is inner ip */
+   enum i40e_fdir_ip_type iip_type; /* ip type for inner ip */
+   bool customized_pctype; /* If customized pctype is used */
 };
 
 /* A structure used to define the input for a flow director filter entry */
diff --git a/drivers/net/i40e/i40e_fdir.c b/drivers/net/i40e/i40e_fdir.c
index 1072a24..55c86ee 100644
--- a/drivers/net/i40e/i40e_fdir.c
+++ b/drivers/net/i40e/i40e_fdir.c
@@ -71,6 +71,9 @@
 #define I40E_FDIR_IPv6_DEFAULT_HOP_LIMITS   0xFF
 #define I40E_FDIR_IPv6_PAYLOAD_LEN  380
 #define I40E_FDIR_UDP_DEFAULT_LEN   400
+#define I40E_FDIR_GTP_DEFAULT_LEN   384
+#define I40E_FDIR_INNER_IP_DEFAULT_LEN  384
+#define I40E_FDIR_INNER_IPv6_DEFAULT_LEN344
 
 /* Wait time for fdir filter programming */
 #define I40E_FDIR_MAX_WAIT_US 1
@@ -939,16 +942,34 @@ i40e_fdir_construct_pkt(struct i40e_pf *pf,
return 0;
 }
 
+static struct i40e_customized_pctype *
+i40e_flow_fdir_find_customized_pctype(struct i40e_pf *pf, uint8_t pctype)
+{
+   struct i40e_customized_pctype *cus_pctype;
+   enum i40e_new_pctype i = I40E_CUSTOMIZED_GTPC;
+
+   for (; i < I40E_CUSTOMIZED_MAX; i++) {
+   cus_pctype = &pf->customized_pctype[i];
+   if (pctype == cus_pctype->pctype)
+   return cus_pctype;
+   }
+   return NULL;
+}
+
 static inline int
-i40e_flow_fdir_fill_eth_ip_head(const struct i40e_fdir_input *fdir_input,
+i40e_flow_fdir_fill_eth_ip_head(struct i40e_pf *pf,
+   const struct i40e_fdir_input *fdir_input,
unsigned char *raw_pkt,
bool vlan)
 {
+   struct i40e_customized_pctype *cus_pctype = NULL;
static uint8_t vlan_frame[] = {0x81, 0, 0, 0};
uint16_t *ether_type;
uint8_t len = 2 * sizeof(struct ether_addr);
struct ipv4_hdr *ip;
struct ipv6_hdr *ip6;
+   uint8_t pctype = fdir_input->pctype;
+   bool is_customized_pctype = fdir_input->flow_ext.customized_pctype;
static const uint8_t next_proto[] = {
[I40E_FILTER_PCTYPE_FRAG_IPV4] = IPPROTO_IP,
[I40E_FILTER_PCTYPE_NONF_IPV4_TCP] = IPPROTO_TCP,
@@ -975,27 +996,30 @@ i40e_flow_fdir_fill_eth_ip_head(const struct 
i40e_fdir_input *fdir_input,
raw_pkt += sizeof(uint16_t);
len += sizeof(uint16_t);
 
-   switch (fdir_input->pctype) {
-   case I40E_FILTER_PCTYPE_L2_PAYLOAD:
+   if (is_customized_pctype) {
+   cus_pctype = i40e_flow_fdir_find_customized_pctype(pf, pctype);
+   if (!cus_pctype)
+   PMD_DRV_LOG(ERR, "unknown pctype %u.",
+   fdir_input->pctype);
+   }
+
+   if (pctype == I40E_FILTER_PCTYPE_L2_PAYLOAD)
*ether_type

[dpdk-dev] [PATCH v5 7/8] net/i40e: add cloud filter parsing function for GTP

[Beilei Xing]

This patch adds i40e_flow_parse_gtp_filter parsing
function for GTP-C and GTP-U to support cloud filter.

Signed-off-by: Beilei Xing 
---
 drivers/net/i40e/i40e_ethdev.h |   2 +
 drivers/net/i40e/i40e_flow.c   | 151 +
 2 files changed, 153 insertions(+)

diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index 502f6c6..436ca2c 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -703,6 +703,8 @@ enum i40e_tunnel_type {
I40E_TUNNEL_TYPE_MPLSoUDP,
I40E_TUNNEL_TYPE_MPLSoGRE,
I40E_TUNNEL_TYPE_QINQ,
+   I40E_TUNNEL_TYPE_GTPC,
+   I40E_TUNNEL_TYPE_GTPU,
I40E_TUNNEL_TYPE_MAX,
 };
 
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index ea81ecb..2bf7098 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -125,6 +125,12 @@ static int i40e_flow_parse_mpls_filter(struct rte_eth_dev 
*dev,
   const struct rte_flow_action actions[],
   struct rte_flow_error *error,
   union i40e_filter_t *filter);
+static int i40e_flow_parse_gtp_filter(struct rte_eth_dev *dev,
+ const struct rte_flow_attr *attr,
+ const struct rte_flow_item pattern[],
+ const struct rte_flow_action actions[],
+ struct rte_flow_error *error,
+ union i40e_filter_t *filter);
 static int i40e_flow_destroy_ethertype_filter(struct i40e_pf *pf,
  struct i40e_ethertype_filter *filter);
 static int i40e_flow_destroy_tunnel_filter(struct i40e_pf *pf,
@@ -1808,6 +1814,11 @@ static struct i40e_valid_pattern 
i40e_supported_patterns[] = {
{ pattern_mpls_2, i40e_flow_parse_mpls_filter },
{ pattern_mpls_3, i40e_flow_parse_mpls_filter },
{ pattern_mpls_4, i40e_flow_parse_mpls_filter },
+   /* GTP-C & GTP-U */
+   { pattern_fdir_ipv4_gtpc, i40e_flow_parse_gtp_filter },
+   { pattern_fdir_ipv4_gtpu, i40e_flow_parse_gtp_filter },
+   { pattern_fdir_ipv6_gtpc, i40e_flow_parse_gtp_filter },
+   { pattern_fdir_ipv6_gtpu, i40e_flow_parse_gtp_filter },
/* QINQ */
{ pattern_qinq_1, i40e_flow_parse_qinq_filter },
 };
@@ -3823,6 +3834,146 @@ i40e_flow_parse_mpls_filter(struct rte_eth_dev *dev,
 }
 
 /* 1. Last in item should be NULL as range is not supported.
+ * 2. Supported filter types: GTP TEID.
+ * 3. Mask of fields which need to be matched should be
+ *filled with 1.
+ * 4. Mask of fields which needn't to be matched should be
+ *filled with 0.
+ */
+static int
+i40e_flow_parse_gtp_pattern(struct rte_eth_dev *dev,
+   const struct rte_flow_item *pattern,
+   struct rte_flow_error *error,
+   struct i40e_tunnel_filter_conf *filter)
+{
+   struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+   const struct rte_flow_item *item = pattern;
+   const struct rte_flow_item_gtp *gtp_spec;
+   const struct rte_flow_item_gtp *gtp_mask;
+   enum rte_flow_item_type item_type;
+
+   if (!pf->gtp_support) {
+   rte_flow_error_set(error, EINVAL,
+  RTE_FLOW_ERROR_TYPE_ITEM,
+  item,
+  "GTP is not supported by default.");
+   return -rte_errno;
+   }
+
+   for (; item->type != RTE_FLOW_ITEM_TYPE_END; item++) {
+   if (item->last) {
+   rte_flow_error_set(error, EINVAL,
+  RTE_FLOW_ERROR_TYPE_ITEM,
+  item,
+  "Not support range");
+   return -rte_errno;
+   }
+   item_type = item->type;
+   switch (item_type) {
+   case RTE_FLOW_ITEM_TYPE_ETH:
+   if (item->spec || item->mask) {
+   rte_flow_error_set(error, EINVAL,
+  RTE_FLOW_ERROR_TYPE_ITEM,
+  item,
+  "Invalid ETH item");
+   return -rte_errno;
+   }
+   break;
+   case RTE_FLOW_ITEM_TYPE_IPV4:
+   filter->ip_type = I40E_TUNNEL_IPTYPE_IPV4;
+   /* IPv4 is used to describe protocol,
+* spec and mask should be NULL.
+*/
+   if (item->spec || item->mask) {
+   rte_flow_error_set(error, EINVAL,
+

[dpdk-dev] [PATCH v5 8/8] net/i40e: enable cloud filter for GTP-C and GTP-U

[Beilei Xing]

This patch sets TEID of GTP-C and GTP-U as filter type
by replacing existed filter types inner_mac and TUNNEL_KEY.
This configuration will be set when adding GTP-C or
GTP-U filter rules, and it will be invalid only by
NIC core reset.

Signed-off-by: Beilei Xing 
---
 drivers/net/i40e/i40e_ethdev.c | 193 +
 drivers/net/i40e/i40e_ethdev.h |  17 ++--
 drivers/net/i40e/i40e_flow.c   |  12 +--
 3 files changed, 191 insertions(+), 31 deletions(-)

diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index aba35a5..18aa376 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -7161,7 +7161,7 @@ i40e_status_code i40e_replace_mpls_l1_filter(struct 
i40e_pf *pf)
/* create L1 filter */
filter_replace.old_filter_type =
I40E_AQC_REPLACE_CLOUD_CMD_INPUT_FV_IMAC;
-   filter_replace.new_filter_type = I40E_AQC_ADD_L1_FILTER_TEID_MPLS;
+   filter_replace.new_filter_type = I40E_AQC_ADD_L1_FILTER_0X11;
filter_replace.tr_bit = 0;
 
/* Prepare the buffer, 3 entries */
@@ -7209,12 +7209,12 @@ i40e_status_code i40e_replace_mpls_cloud_filter(struct 
i40e_pf *pf)
I40E_AQC_MIRROR_CLOUD_FILTER;
filter_replace.old_filter_type = I40E_AQC_ADD_CLOUD_FILTER_IIP;
filter_replace.new_filter_type =
-   I40E_AQC_ADD_CLOUD_FILTER_TEID_MPLSoUDP;
+   I40E_AQC_ADD_CLOUD_FILTER_0X11;
/* Prepare the buffer, 2 entries */
filter_replace_buf.data[0] = I40E_AQC_REPLACE_CLOUD_CMD_INPUT_FV_STAG;
filter_replace_buf.data[0] |=
I40E_AQC_REPLACE_CLOUD_CMD_INPUT_VALIDATED;
-   filter_replace_buf.data[4] = I40E_AQC_ADD_L1_FILTER_TEID_MPLS;
+   filter_replace_buf.data[4] = I40E_AQC_ADD_L1_FILTER_0X11;
filter_replace_buf.data[4] |=
I40E_AQC_REPLACE_CLOUD_CMD_INPUT_VALIDATED;
status = i40e_aq_replace_cloud_filters(hw, &filter_replace,
@@ -7232,12 +7232,131 @@ i40e_status_code i40e_replace_mpls_cloud_filter(struct 
i40e_pf *pf)
I40E_AQC_MIRROR_CLOUD_FILTER;
filter_replace.old_filter_type = I40E_AQC_ADD_CLOUD_FILTER_IMAC;
filter_replace.new_filter_type =
-   I40E_AQC_ADD_CLOUD_FILTER_TEID_MPLSoGRE;
+   I40E_AQC_ADD_CLOUD_FILTER_0X12;
/* Prepare the buffer, 2 entries */
filter_replace_buf.data[0] = I40E_AQC_REPLACE_CLOUD_CMD_INPUT_FV_STAG;
filter_replace_buf.data[0] |=
I40E_AQC_REPLACE_CLOUD_CMD_INPUT_VALIDATED;
-   filter_replace_buf.data[4] = I40E_AQC_ADD_L1_FILTER_TEID_MPLS;
+   filter_replace_buf.data[4] = I40E_AQC_ADD_L1_FILTER_0X11;
+   filter_replace_buf.data[4] |=
+   I40E_AQC_REPLACE_CLOUD_CMD_INPUT_VALIDATED;
+
+   status = i40e_aq_replace_cloud_filters(hw, &filter_replace,
+  &filter_replace_buf);
+   return status;
+}
+
+static enum i40e_status_code
+i40e_replace_gtp_l1_filter(struct i40e_pf *pf)
+{
+   struct i40e_aqc_replace_cloud_filters_cmd  filter_replace;
+   struct i40e_aqc_replace_cloud_filters_cmd_buf  filter_replace_buf;
+   struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+   enum i40e_status_code status = I40E_SUCCESS;
+
+   /* For GTP-C */
+   memset(&filter_replace, 0,
+  sizeof(struct i40e_aqc_replace_cloud_filters_cmd));
+   memset(&filter_replace_buf, 0,
+  sizeof(struct i40e_aqc_replace_cloud_filters_cmd_buf));
+   /* create L1 filter */
+   filter_replace.old_filter_type =
+   I40E_AQC_REPLACE_CLOUD_CMD_INPUT_FV_IMAC;
+   filter_replace.new_filter_type = I40E_AQC_ADD_L1_FILTER_0X12;
+   filter_replace.tr_bit = I40E_AQC_NEW_TR_22 |
+   I40E_AQC_REPLACE_CLOUD_CMD_INPUT_VALIDATED;
+   /* Prepare the buffer, 2 entries */
+   filter_replace_buf.data[0] =
+   I40E_AQC_REPLACE_CLOUD_CMD_INPUT_FV_TEID_WORD0;
+   filter_replace_buf.data[0] |=
+   I40E_AQC_REPLACE_CLOUD_CMD_INPUT_VALIDATED;
+   filter_replace_buf.data[2] = 0xFF;
+   filter_replace_buf.data[3] = 0xFF;
+   filter_replace_buf.data[4] =
+   I40E_AQC_REPLACE_CLOUD_CMD_INPUT_FV_TEID_WORD1;
+   filter_replace_buf.data[4] |=
+   I40E_AQC_REPLACE_CLOUD_CMD_INPUT_VALIDATED;
+   filter_replace_buf.data[6] = 0xFF;
+   filter_replace_buf.data[7] = 0xFF;
+   status = i40e_aq_replace_cloud_filters(hw, &filter_replace,
+  &filter_replace_buf);
+   if (status < 0)
+   return status;
+
+   /* for GTP-U */
+   memset(&filter_replace, 0,
+  sizeof(struct i40e_aqc_replace_cloud_filters_cmd));
+   memset(&filter_replace_buf, 0,
+  sizeof(struct i40e_aqc_replace_cloud_filters_cmd_buf));
+   /* create L1 filter */
+   filter_replace.old_filter_type =
+   I40E_AQC_REPLACE_C

Re: [dpdk-dev] [PATCH v4 2/4] net/softnic: add traffic management support

[Singh, Jasvinder]

> 
> Hi Jasvinder,
> 
> > -Original Message-
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Jasvinder Singh
> > Sent: Monday, September 18, 2017 5:10 PM
> > To: dev@dpdk.org
> > Cc: Dumitrescu, Cristian ; Yigit,
> > Ferruh ; tho...@monjalon.net
> > Subject: [dpdk-dev] [PATCH v4 2/4] net/softnic: add traffic management
> > support
> >
> > Add ethdev Traffic Management API support to SoftNIC PMD.
> >
> > Signed-off-by: Cristian Dumitrescu 
> > Signed-off-by: Jasvinder Singh 
> > ---
> > v3 changes:
> > - add more confguration parameters (tm rate, tm queue sizes)
> >
> >  drivers/net/softnic/Makefile|   1 +
> >  drivers/net/softnic/rte_eth_softnic.c   | 252
> > +++-
> >  drivers/net/softnic/rte_eth_softnic.h   |  16 ++
> >  drivers/net/softnic/rte_eth_softnic_internals.h | 106 +-
> >  drivers/net/softnic/rte_eth_softnic_tm.c| 181 +
> >  5 files changed, 553 insertions(+), 3 deletions(-)  create mode
> > 100644 drivers/net/softnic/rte_eth_softnic_tm.c
> 
> 
> 
> >
> >  static void
> > @@ -293,6 +325,77 @@ rte_pmd_softnic_run_default(struct rte_eth_dev
> > *dev)
> > return 0;
> >  }
> >
> > +static __rte_always_inline int
> > +rte_pmd_softnic_run_tm(struct rte_eth_dev *dev) {
> This function name seems a little misleading. If it's a inline function not an
> API,  better just name it 'softnic_run_tm".
> And a common comments for the names, like, pmd_feature,
> tm_params_check, tm_init ... if they're only for soft nic, better add the 
> prefix
> 'softnc_' for them.

Ok, will change the above function name for clarity. Thanks.

Re: [dpdk-dev] [PATCH v4 3/4] net/softnic: add TM capabilities ops

[Singh, Jasvinder]

> Hi Jasvinder,
> 
> > -Original Message-
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Jasvinder Singh
> > Sent: Monday, September 18, 2017 5:10 PM
> > To: dev@dpdk.org
> > Cc: Dumitrescu, Cristian ; Yigit,
> > Ferruh ; tho...@monjalon.net
> > Subject: [dpdk-dev] [PATCH v4 3/4] net/softnic: add TM capabilities
> > ops
> >
> > Implement ethdev TM capability APIs in SoftNIC PMD.
> >
> > Signed-off-by: Cristian Dumitrescu 
> > Signed-off-by: Jasvinder Singh 
> > ---
> >  drivers/net/softnic/rte_eth_softnic.c   |  12 +-
> >  drivers/net/softnic/rte_eth_softnic_internals.h |  32 ++
> >  drivers/net/softnic/rte_eth_softnic_tm.c| 500
> > 
> >  3 files changed, 543 insertions(+), 1 deletion(-)
> The same concern of the naming as patch 2.The function and structure
> names are too common. Better add the prefix 'softnic_' too.
> Except that, the patch looks good to me.

Ok, Thanks. Will clarity and add comments as well.

Re: [dpdk-dev] [PATCH v4 4/4] net/softnic: add TM hierarchy related ops

[Singh, Jasvinder]

> Hi Jasvinder,
> 
> 
> > -Original Message-
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Jasvinder Singh
> > Sent: Monday, September 18, 2017 5:10 PM
> > To: dev@dpdk.org
> > Cc: Dumitrescu, Cristian ; Yigit,
> > Ferruh ; tho...@monjalon.net
> > Subject: [dpdk-dev] [PATCH v4 4/4] net/softnic: add TM hierarchy
> > related ops
> >
> > Implement ethdev TM hierarchy related APIs in SoftNIC PMD.
> >
> > Signed-off-by: Cristian Dumitrescu 
> > Signed-off-by: Jasvinder Singh 
> > ---
> >  drivers/net/softnic/rte_eth_softnic_internals.h |   41 +
> >  drivers/net/softnic/rte_eth_softnic_tm.c| 2776
> > ++-
> >  2 files changed, 2813 insertions(+), 4 deletions(-)
> 
> 
> > +
> > +static uint32_t
> > +tm_node_subport_id(struct rte_eth_dev *dev, struct tm_node
> > *subport_node)
> > +{
> > +   struct pmd_internals *p = dev->data->dev_private;
> > +   struct tm_node_list *nl = &p->soft.tm.h.nodes;
> > +   struct tm_node *ns;
> > +   uint32_t subport_id;
> > +
> > +   subport_id = 0;
> > +   TAILQ_FOREACH(ns, nl, node) {
> > +   if (ns->level != TM_NODE_LEVEL_SUBPORT)
> > +   continue;
> > +
> > +   if (ns->node_id == subport_node->node_id)
> > +   return subport_id;
> > +
> > +   subport_id++;
> > +   }
> > +
> > +   return UINT32_MAX;
> UINT32_MAX means invalid number, right? Better define a specific MACRO
> for the invalid number in case you may not want to use 0xff.. or uint32.
> The same suggestion for the below functions.

Ok.

> > +static int
> > +shaper_profile_check(struct rte_eth_dev *dev,
> > +   uint32_t shaper_profile_id,
> > +   struct rte_tm_shaper_params *profile,
> > +   struct rte_tm_error *error)
> > +{
> > +   struct tm_shaper_profile *sp;
> > +
> > +   /* Shaper profile ID must not be NONE. */
> > +   if (shaper_profile_id == RTE_TM_SHAPER_PROFILE_ID_NONE)
> > +   return -rte_tm_error_set(error,
> > +   EINVAL,
> > +   RTE_TM_ERROR_TYPE_SHAPER_PROFILE_ID,
> > +   NULL,
> > +   rte_strerror(EINVAL));
> > +
> > +   /* Shaper profile must not exist. */
> > +   sp = tm_shaper_profile_search(dev, shaper_profile_id);
> > +   if (sp)
> > +   return -rte_tm_error_set(error,
> > +   EEXIST,
> > +   RTE_TM_ERROR_TYPE_SHAPER_PROFILE_ID,
> > +   NULL,
> > +   rte_strerror(EEXIST));
> > +
> > +   /* Profile must not be NULL. */
> > +   if (profile == NULL)
> > +   return -rte_tm_error_set(error,
> > +   EINVAL,
> > +   RTE_TM_ERROR_TYPE_SHAPER_PROFILE,
> > +   NULL,
> > +   rte_strerror(EINVAL));
> A slight suggestion. We can do the easiest check at first.

We preferred to perform checks as per the arguments order specified in the 
function definition so that all the parameter could 
be scanned in a systematic manner instead of picking them randomly.

> > +
> > +/* Traffic manager shaper profile add */ static int
> > +pmd_tm_shaper_profile_add(struct rte_eth_dev *dev,
> > +   uint32_t shaper_profile_id,
> > +   struct rte_tm_shaper_params *profile,
> > +   struct rte_tm_error *error)
> > +{
> > +   struct pmd_internals *p = dev->data->dev_private;
> > +   struct tm_shaper_profile_list *spl = &p->soft.tm.h.shaper_profiles;
> > +   struct tm_shaper_profile *sp;
> > +   int status;
> > +
> > +   /* Check input params */
> > +   status = shaper_profile_check(dev, shaper_profile_id, profile, error);
> > +   if (status)
> > +   return status;
> > +
> > +   /* Memory allocation */
> > +   sp = calloc(1, sizeof(struct tm_shaper_profile));
> Just curious, why not use rte_zmalloc?

This relates to high level hierarchy specification objects which doesn't need 
to be allocated on specific
numa node as it is not used once the hierarchy is committed. All these objects 
gets eventually translated into
TM implementation  (librte_sched) specific objects. These objects are allocated 
using rte_zmalloc and  needs
lots of memory (approx. 2M mbufs for single instance of TM hierarchy ports) on 
specific numa node.

> > +   if (sp == NULL)
> > +   return -rte_tm_error_set(error,
> > +   ENOMEM,
> > +   RTE_TM_ERROR_TYPE_UNSPECIFIED,
> > +   NULL,
> > +   rte_strerror(ENOMEM));
> > +
> > +   /* Fill in */
> > +   sp->shaper_profile_id = shaper_profile_id;
> > +   memcpy(&sp->params, profile, sizeof(sp->params));
> > +
> > +   /* Add to list */
> > +   TAILQ_INSERT_TAIL(spl, sp, node);
> > +   p->soft.tm.h.n_shaper_profiles++;
> > +
> > +   return 0;
> > +}
> > +
> 
> > +
> > +static struct tm_node *
> > +tm_shared_shaper_get_tc(struct rte_eth_dev *dev,
> > +   struct tm_shared_shaper *ss)
> > +{
> > +   struct pmd_internals *p = dev->data->dev_private;
> > +   struct tm_node_list *nl = &p->soft.tm.h.nodes;
> > +   struct tm_node *n;
> > +
> > 

[dpdk-dev] [PATCH v4 0/3] net/i40e: API to configure queue regions for RSS

[Wei Zhao]

The patches mainly finish following functions:
1) Command line for input queue region configure parameters.
2) Set or flush queue region configutation.

v2:
-fix bug for packet into queue region code.
-change code style for DPDK C programing.
-change that "flowtype" name to  " hw_flowtype " in struct 
rte_i40e_rss_region_conf

v3:
-fix patch check warning
-fix a bug in cmdline.c

v4:
-fix bug for flowtype config.
-add a new command for get all queue region configuration.
-change function name to "rss" related.
-add document info about queue region CLI.

 app/test-pmd/cmdline.c  | 379 
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |  40 +++
 drivers/net/i40e/i40e_ethdev.c  |  27 +-
 drivers/net/i40e/i40e_ethdev.h  |  39 ++
 drivers/net/i40e/rte_pmd_i40e.c | 536 
 drivers/net/i40e/rte_pmd_i40e.h |  60 
 drivers/net/i40e/rte_pmd_i40e_version.map   |   1 +
 7 files changed, 1076 insertions(+), 6 deletions(-)

-- 
2.7.4

[dpdk-dev] [PATCH v4 1/3] net/i40e: queue region set and flush

[Wei Zhao]

This feature enable queue regions configuration for RSS in PF,
so that different traffic classes or different packet
classification types can be separated to different queues in
different queue regions.This patch can set queue region range,
it include queue number in a region and the index of first queue.
This patch enable mapping between different priorities (UP) and
different traffic classes.It also enable mapping between a region
index and a sepcific flowtype(PCTYPE).It also provide the solution
of flush all configuration about queue region the above described.

Signed-off-by: Wei Zhao 
---
 drivers/net/i40e/i40e_ethdev.c|  27 +-
 drivers/net/i40e/i40e_ethdev.h|  39 +++
 drivers/net/i40e/rte_pmd_i40e.c   | 536 ++
 drivers/net/i40e/rte_pmd_i40e.h   |  60 
 drivers/net/i40e/rte_pmd_i40e_version.map |   1 +
 5 files changed, 657 insertions(+), 6 deletions(-)

diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 720f067..1a5b73e 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -137,10 +137,6 @@
 #define I40E_PRTTSYN_TSYNTYPE0x0e00
 #define I40E_CYCLECOUNTER_MASK   0xULL
 
-#define I40E_MAX_PERCENT100
-#define I40E_DEFAULT_DCB_APP_NUM1
-#define I40E_DEFAULT_DCB_APP_PRIO   3
-
 /**
  * Below are values for writing un-exposed registers suggested
  * by silicon experts
@@ -309,7 +305,6 @@ static int i40e_pf_parameter_init(struct rte_eth_dev *dev);
 static int i40e_pf_setup(struct i40e_pf *pf);
 static int i40e_dev_rxtx_init(struct i40e_pf *pf);
 static int i40e_vmdq_setup(struct rte_eth_dev *dev);
-static int i40e_dcb_init_configure(struct rte_eth_dev *dev, bool sw_dcb);
 static int i40e_dcb_setup(struct rte_eth_dev *dev);
 static void i40e_stat_update_32(struct i40e_hw *hw, uint32_t reg,
bool offset_loaded, uint64_t *offset, uint64_t *stat);
@@ -1036,6 +1031,20 @@ i40e_init_fdir_filter_list(struct rte_eth_dev *dev)
return ret;
 }
 
+void
+i40e_init_queue_region_conf(struct rte_eth_dev *dev)
+{
+   struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+   struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+   struct i40e_queue_region_info *info = &pf->queue_region;
+   uint16_t i;
+
+   for (i = 0; i < I40E_PFQF_HREGION_MAX_INDEX; i++)
+   i40e_write_rx_ctl(hw, I40E_PFQF_HREGION(i), 0);
+
+   memset(info, 0, sizeof(struct i40e_queue_region_info));
+}
+
 static int
 eth_i40e_dev_init(struct rte_eth_dev *dev)
 {
@@ -1311,6 +1320,9 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
if (ret < 0)
goto err_init_fdir_filter_list;
 
+   /* initialize queue region configuration */
+   i40e_init_queue_region_conf(dev);
+
return 0;
 
 err_init_fdir_filter_list:
@@ -1466,6 +1478,9 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
/* Remove all Traffic Manager configuration */
i40e_tm_conf_uninit(dev);
 
+   /* Remove all the queue region configuration */
+   i40e_flush_queue_region_all_conf(dev, hw, pf, 0);
+
return 0;
 }
 
@@ -10419,7 +10434,7 @@ i40e_dcb_hw_configure(struct i40e_pf *pf,
  *
  * Returns 0 on success, negative value on failure
  */
-static int
+int
 i40e_dcb_init_configure(struct rte_eth_dev *dev, bool sw_dcb)
 {
struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index ad80f0f..3d237cd 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -260,6 +260,12 @@ enum i40e_flxpld_layer_idx {
 #define I40E_QOS_BW_WEIGHT_MIN 1
 /* The max bandwidth weight is 127. */
 #define I40E_QOS_BW_WEIGHT_MAX 127
+/* The max queue region index is 7. */
+#define I40E_REGION_MAX_INDEX 7
+
+#define I40E_MAX_PERCENT100
+#define I40E_DEFAULT_DCB_APP_NUM1
+#define I40E_DEFAULT_DCB_APP_PRIO   3
 
 /**
  * The overhead from MTU to max frame size.
@@ -541,6 +547,34 @@ struct i40e_ethertype_rule {
struct rte_hash *hash_table;
 };
 
+/* queue region info */
+struct i40e_region_info {
+   /* the region id for this configuration */
+   uint8_t region_id;
+   /* the start queue index for this region */
+   uint8_t queue_start_index;
+   /* the total queue number of this queue region */
+   uint8_t queue_num;
+   /* the total number of user priority for this region */
+   uint8_t user_priority_num;
+   /* the packet's user priority for this region */
+   uint8_t user_priority[I40E_MAX_USER_PRIORITY];
+   /* the total number of flowtype for this region */
+   uint8_t flowtype_num;
+   /**
+* the pctype or hardware flowtype of packet,
+* the specific index for each type has been defined
+* in file i40e_type.h as enum i40e_filter_pctype.
+*/
+   uint8_t hw_flo

[dpdk-dev] [PATCH v4 2/3] app/testpmd: add API for configuration of queue region

[Wei Zhao]

This patch add a API configuration of queue region in rss.
It can parse the parameters of region index, queue number,
queue start index, user priority, traffic classes and so on.
According to commands from command line, it will call i40e
private API and start the process of set or flush queue region
configure. As this feature is specific for i40e, so private API
will be used.

Signed-off-by: Wei Zhao 
---
 app/test-pmd/cmdline.c | 379 +
 1 file changed, 379 insertions(+)

diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 4f2d731..fc2444f 100644
--- a/app/test-pmd/cmdline.c
+++ b/app/test-pmd/cmdline.c
@@ -637,6 +637,24 @@ static void cmd_help_long_parsed(void *parsed_result,
"ptype mapping update (port_id) (hw_ptype) (sw_ptype)\n"
"Update a ptype mapping item on a port\n\n"
 
+   "queue-region set port (port_id) region_id (value) "
+   "queue_start_index (value) queue_num (value)\n"
+   "Set a queue region on a port\n\n"
+
+   "queue-region set port (port_id) region_id (value) "
+   "flowtype (value)\n"
+   "Set a flowtype region index on a port\n\n"
+
+   "queue-region set port (port_id) UP (value) region_id 
(value)\n"
+   "Set the mapping of User Priority to "
+   "queue region on a port\n\n"
+
+   "queue-region flush (on|off) port (port_id)\n"
+   "flush all queue region related configuration\n\n"
+
+   "queue-region get port (port_id)\n"
+   "get all queue region related configuration 
info\n\n"
+
, list_pkt_forwarding_modes()
);
}
@@ -8228,6 +8246,362 @@ cmdline_parse_inst_t cmd_syn_filter = {
NULL,
},
 };
+/* *** queue region set *** */
+struct cmd_queue_region_result {
+   cmdline_fixed_string_t cmd;
+   cmdline_fixed_string_t set;
+   cmdline_fixed_string_t port;
+   uint8_t  port_id;
+   cmdline_fixed_string_t region;
+   uint8_t  region_id;
+   cmdline_fixed_string_t queue_start_index;
+   uint8_t  queue_id;
+   cmdline_fixed_string_t queue_num;
+   uint8_t  queue_num_value;
+};
+
+static void
+cmd_queue_region_parsed(void *parsed_result,
+   __attribute__((unused)) struct cmdline *cl,
+   __attribute__((unused)) void *data)
+{
+   struct cmd_queue_region_result *res = parsed_result;
+   struct rte_i40e_rss_region_conf region_conf;
+   int ret = 0;
+
+   memset(®ion_conf, 0, sizeof(region_conf));
+   region_conf.op = RTE_PMD_I40E_QUEUE_REGION_SET;
+   region_conf.region_id = res->region_id;
+   region_conf.queue_num = res->queue_num_value;
+   region_conf.queue_start_index = res->queue_id;
+
+   ret = rte_pmd_i40e_rss_queue_region_conf(res->port_id, ®ion_conf);
+
+   if (ret < 0)
+   printf("queue region config programming error: (%s)\n",
+   strerror(-ret));
+}
+
+cmdline_parse_token_string_t cmd_queue_region_cmd =
+   TOKEN_STRING_INITIALIZER(struct cmd_queue_region_result,
+cmd, "queue-region");
+cmdline_parse_token_string_t cmd_queue_region_set =
+TOKEN_STRING_INITIALIZER(struct cmd_queue_region_result,
+   set, "set");
+cmdline_parse_token_string_t cmd_queue_region_port =
+   TOKEN_STRING_INITIALIZER(struct cmd_queue_region_result, port, "port");
+cmdline_parse_token_num_t cmd_queue_region_port_id =
+   TOKEN_NUM_INITIALIZER(struct cmd_queue_region_result,
+   port_id, UINT8);
+cmdline_parse_token_string_t cmd_queue_region_id =
+   TOKEN_STRING_INITIALIZER(struct cmd_queue_region_result,
+   region, "region_id");
+cmdline_parse_token_num_t cmd_queue_region_index =
+   TOKEN_NUM_INITIALIZER(struct cmd_queue_region_result,
+   region_id, UINT8);
+cmdline_parse_token_string_t cmd_queue_region_queue_start_index =
+   TOKEN_STRING_INITIALIZER(struct cmd_queue_region_result,
+   queue_start_index, "queue_start_index");
+cmdline_parse_token_num_t cmd_queue_region_queue_id =
+   TOKEN_NUM_INITIALIZER(struct cmd_queue_region_result,
+   queue_id, UINT8);
+cmdline_parse_token_string_t cmd_queue_region_queue_num =
+   TOKEN_STRING_INITIALIZER(struct cmd_queue_region_result,
+   queue_num, "queue_num");
+cmdline_parse_token_num_t cmd_queue_region_queue_num_value =
+   TOKEN_NUM_INITIALIZER(struct cmd_queue_region_result,
+   queue_num_value, UINT8);
+
+cmdline_parse_inst_t cmd_queue_region = {
+   .f = cmd_

[dpdk-dev] [PATCH v4 3/3] doc/testpmd_app_ug: add doc info for queue region

[Wei Zhao]

Queue region only support PF by now, so this document is
only for configuration of queue region on PF port.

Signed-off-by: Wei Zhao 
---
 doc/guides/testpmd_app_ug/testpmd_funcs.rst | 40 +
 1 file changed, 40 insertions(+)

diff --git a/doc/guides/testpmd_app_ug/testpmd_funcs.rst 
b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
index 2ed62f5..f31ec5f 100644
--- a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
+++ b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
@@ -842,6 +842,46 @@ Where:
 
Check the NIC Datasheet for hardware limits.
 
+RSS queue region
+
+
+Set RSS queue region span on a port::
+
+   testpmd> queue-region set port (port_id) region_id (value) 
queue_start_index (value) queue_num (value)
+
+Set flowtype mapping on a RSS queue region on a port::
+
+   testpmd> queue-region set port (port_id) region_id (value) flowtype (value)
+
+where:
+
+* For the flowtype(pctype) of packet,the specific index for each type has
+  been defined in file i40e_type.h as enum i40e_filter_pctype.
+
+Set user priority mapping on a RSS queue region on a port::
+
+   testpmd> queue-region set port (port_id) UP (value) region_id (value)
+
+Flush all queue region related configuration on a port::
+
+   testpmd> queue-region flush (on|off) port (port_id)
+
+where:
+* "on"is just an enable function which server for other configuration,
+  it is for all configuration about queue region from up layer,
+  at first will only keep in DPDK softwarestored in driver,
+  only after "flush on", it commit all configuration to HW.
+  "off" is just clean all configuration about queue region just now,
+  and restore all to DPDK i40e driver default config when start up.
+
+Get all queue region related configuration info on a port::
+
+   testpmd> queue-region get port (port_id)
+
+.. note::
+Queue region only support on PF by now, so thses command is
+only for configuration of queue region on PF port.
+
 csum parse-tunnel
 ~
 
-- 
2.7.4

[dpdk-dev] [PATCH v4 0/3] net/i40e: API to configure queue regions for RSS

[Wei Zhao]

The patches mainly finish following functions:
1) Command line for input queue region configure parameters.
2) Set or flush queue region configutation.

v2:
-fix bug for packet into queue region code.
-change code style for DPDK C programing.
-change that "flowtype" name to  " hw_flowtype " in struct 
rte_i40e_rss_region_conf

v3:
-fix patch check warning
-fix a bug in cmdline.c

v4:
-fix bug for flowtype config.
-add a new command for get all queue region configuration.
-change function name to "rss" related.
-add document info about queue region CLI.

 app/test-pmd/cmdline.c  | 379 
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |  40 +++
 drivers/net/i40e/i40e_ethdev.c  |  27 +-
 drivers/net/i40e/i40e_ethdev.h  |  39 ++
 drivers/net/i40e/rte_pmd_i40e.c | 536 
 drivers/net/i40e/rte_pmd_i40e.h |  60 
 drivers/net/i40e/rte_pmd_i40e_version.map   |   1 +
 7 files changed, 1076 insertions(+), 6 deletions(-)

-- 
2.7.4

[dpdk-dev] [PATCH v4 2/3] app/testpmd: add API for configuration of queue region

[Wei Zhao]

This patch add a API configuration of queue region in rss.
It can parse the parameters of region index, queue number,
queue start index, user priority, traffic classes and so on.
According to commands from command line, it will call i40e
private API and start the process of set or flush queue region
configure. As this feature is specific for i40e, so private API
will be used.

Signed-off-by: Wei Zhao 
---
 app/test-pmd/cmdline.c | 379 +
 1 file changed, 379 insertions(+)

diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 4f2d731..fc2444f 100644
--- a/app/test-pmd/cmdline.c
+++ b/app/test-pmd/cmdline.c
@@ -637,6 +637,24 @@ static void cmd_help_long_parsed(void *parsed_result,
"ptype mapping update (port_id) (hw_ptype) (sw_ptype)\n"
"Update a ptype mapping item on a port\n\n"
 
+   "queue-region set port (port_id) region_id (value) "
+   "queue_start_index (value) queue_num (value)\n"
+   "Set a queue region on a port\n\n"
+
+   "queue-region set port (port_id) region_id (value) "
+   "flowtype (value)\n"
+   "Set a flowtype region index on a port\n\n"
+
+   "queue-region set port (port_id) UP (value) region_id 
(value)\n"
+   "Set the mapping of User Priority to "
+   "queue region on a port\n\n"
+
+   "queue-region flush (on|off) port (port_id)\n"
+   "flush all queue region related configuration\n\n"
+
+   "queue-region get port (port_id)\n"
+   "get all queue region related configuration 
info\n\n"
+
, list_pkt_forwarding_modes()
);
}
@@ -8228,6 +8246,362 @@ cmdline_parse_inst_t cmd_syn_filter = {
NULL,
},
 };
+/* *** queue region set *** */
+struct cmd_queue_region_result {
+   cmdline_fixed_string_t cmd;
+   cmdline_fixed_string_t set;
+   cmdline_fixed_string_t port;
+   uint8_t  port_id;
+   cmdline_fixed_string_t region;
+   uint8_t  region_id;
+   cmdline_fixed_string_t queue_start_index;
+   uint8_t  queue_id;
+   cmdline_fixed_string_t queue_num;
+   uint8_t  queue_num_value;
+};
+
+static void
+cmd_queue_region_parsed(void *parsed_result,
+   __attribute__((unused)) struct cmdline *cl,
+   __attribute__((unused)) void *data)
+{
+   struct cmd_queue_region_result *res = parsed_result;
+   struct rte_i40e_rss_region_conf region_conf;
+   int ret = 0;
+
+   memset(®ion_conf, 0, sizeof(region_conf));
+   region_conf.op = RTE_PMD_I40E_QUEUE_REGION_SET;
+   region_conf.region_id = res->region_id;
+   region_conf.queue_num = res->queue_num_value;
+   region_conf.queue_start_index = res->queue_id;
+
+   ret = rte_pmd_i40e_rss_queue_region_conf(res->port_id, ®ion_conf);
+
+   if (ret < 0)
+   printf("queue region config programming error: (%s)\n",
+   strerror(-ret));
+}
+
+cmdline_parse_token_string_t cmd_queue_region_cmd =
+   TOKEN_STRING_INITIALIZER(struct cmd_queue_region_result,
+cmd, "queue-region");
+cmdline_parse_token_string_t cmd_queue_region_set =
+TOKEN_STRING_INITIALIZER(struct cmd_queue_region_result,
+   set, "set");
+cmdline_parse_token_string_t cmd_queue_region_port =
+   TOKEN_STRING_INITIALIZER(struct cmd_queue_region_result, port, "port");
+cmdline_parse_token_num_t cmd_queue_region_port_id =
+   TOKEN_NUM_INITIALIZER(struct cmd_queue_region_result,
+   port_id, UINT8);
+cmdline_parse_token_string_t cmd_queue_region_id =
+   TOKEN_STRING_INITIALIZER(struct cmd_queue_region_result,
+   region, "region_id");
+cmdline_parse_token_num_t cmd_queue_region_index =
+   TOKEN_NUM_INITIALIZER(struct cmd_queue_region_result,
+   region_id, UINT8);
+cmdline_parse_token_string_t cmd_queue_region_queue_start_index =
+   TOKEN_STRING_INITIALIZER(struct cmd_queue_region_result,
+   queue_start_index, "queue_start_index");
+cmdline_parse_token_num_t cmd_queue_region_queue_id =
+   TOKEN_NUM_INITIALIZER(struct cmd_queue_region_result,
+   queue_id, UINT8);
+cmdline_parse_token_string_t cmd_queue_region_queue_num =
+   TOKEN_STRING_INITIALIZER(struct cmd_queue_region_result,
+   queue_num, "queue_num");
+cmdline_parse_token_num_t cmd_queue_region_queue_num_value =
+   TOKEN_NUM_INITIALIZER(struct cmd_queue_region_result,
+   queue_num_value, UINT8);
+
+cmdline_parse_inst_t cmd_queue_region = {
+   .f = cmd_

[dpdk-dev] [PATCH v4 3/3] doc/testpmd_app_ug: add doc info for queue region

[Wei Zhao]

Queue region only support PF by now, so this document is
only for configuration of queue region on PF port.

Signed-off-by: Wei Zhao 
---
 doc/guides/testpmd_app_ug/testpmd_funcs.rst | 40 +
 1 file changed, 40 insertions(+)

diff --git a/doc/guides/testpmd_app_ug/testpmd_funcs.rst 
b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
index 2ed62f5..f31ec5f 100644
--- a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
+++ b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
@@ -842,6 +842,46 @@ Where:
 
Check the NIC Datasheet for hardware limits.
 
+RSS queue region
+
+
+Set RSS queue region span on a port::
+
+   testpmd> queue-region set port (port_id) region_id (value) 
queue_start_index (value) queue_num (value)
+
+Set flowtype mapping on a RSS queue region on a port::
+
+   testpmd> queue-region set port (port_id) region_id (value) flowtype (value)
+
+where:
+
+* For the flowtype(pctype) of packet,the specific index for each type has
+  been defined in file i40e_type.h as enum i40e_filter_pctype.
+
+Set user priority mapping on a RSS queue region on a port::
+
+   testpmd> queue-region set port (port_id) UP (value) region_id (value)
+
+Flush all queue region related configuration on a port::
+
+   testpmd> queue-region flush (on|off) port (port_id)
+
+where:
+* "on"is just an enable function which server for other configuration,
+  it is for all configuration about queue region from up layer,
+  at first will only keep in DPDK softwarestored in driver,
+  only after "flush on", it commit all configuration to HW.
+  "off" is just clean all configuration about queue region just now,
+  and restore all to DPDK i40e driver default config when start up.
+
+Get all queue region related configuration info on a port::
+
+   testpmd> queue-region get port (port_id)
+
+.. note::
+Queue region only support on PF by now, so these command is
+only for configuration of queue region on PF port.
+
 csum parse-tunnel
 ~
 
-- 
2.7.4

[dpdk-dev] [PATCH v4 1/3] net/i40e: queue region set and flush

[Wei Zhao]

This feature enable queue regions configuration for RSS in PF,
so that different traffic classes or different packet
classification types can be separated to different queues in
different queue regions.This patch can set queue region range,
it include queue number in a region and the index of first queue.
This patch enable mapping between different priorities (UP) and
different traffic classes.It also enable mapping between a region
index and a sepcific flowtype(PCTYPE).It also provide the solution
of flush all configuration about queue region the above described.

Signed-off-by: Wei Zhao 
---
 drivers/net/i40e/i40e_ethdev.c|  27 +-
 drivers/net/i40e/i40e_ethdev.h|  39 +++
 drivers/net/i40e/rte_pmd_i40e.c   | 536 ++
 drivers/net/i40e/rte_pmd_i40e.h   |  60 
 drivers/net/i40e/rte_pmd_i40e_version.map |   1 +
 5 files changed, 657 insertions(+), 6 deletions(-)

diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 720f067..1a5b73e 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -137,10 +137,6 @@
 #define I40E_PRTTSYN_TSYNTYPE0x0e00
 #define I40E_CYCLECOUNTER_MASK   0xULL
 
-#define I40E_MAX_PERCENT100
-#define I40E_DEFAULT_DCB_APP_NUM1
-#define I40E_DEFAULT_DCB_APP_PRIO   3
-
 /**
  * Below are values for writing un-exposed registers suggested
  * by silicon experts
@@ -309,7 +305,6 @@ static int i40e_pf_parameter_init(struct rte_eth_dev *dev);
 static int i40e_pf_setup(struct i40e_pf *pf);
 static int i40e_dev_rxtx_init(struct i40e_pf *pf);
 static int i40e_vmdq_setup(struct rte_eth_dev *dev);
-static int i40e_dcb_init_configure(struct rte_eth_dev *dev, bool sw_dcb);
 static int i40e_dcb_setup(struct rte_eth_dev *dev);
 static void i40e_stat_update_32(struct i40e_hw *hw, uint32_t reg,
bool offset_loaded, uint64_t *offset, uint64_t *stat);
@@ -1036,6 +1031,20 @@ i40e_init_fdir_filter_list(struct rte_eth_dev *dev)
return ret;
 }
 
+void
+i40e_init_queue_region_conf(struct rte_eth_dev *dev)
+{
+   struct i40e_hw *hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+   struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+   struct i40e_queue_region_info *info = &pf->queue_region;
+   uint16_t i;
+
+   for (i = 0; i < I40E_PFQF_HREGION_MAX_INDEX; i++)
+   i40e_write_rx_ctl(hw, I40E_PFQF_HREGION(i), 0);
+
+   memset(info, 0, sizeof(struct i40e_queue_region_info));
+}
+
 static int
 eth_i40e_dev_init(struct rte_eth_dev *dev)
 {
@@ -1311,6 +1320,9 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
if (ret < 0)
goto err_init_fdir_filter_list;
 
+   /* initialize queue region configuration */
+   i40e_init_queue_region_conf(dev);
+
return 0;
 
 err_init_fdir_filter_list:
@@ -1466,6 +1478,9 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
/* Remove all Traffic Manager configuration */
i40e_tm_conf_uninit(dev);
 
+   /* Remove all the queue region configuration */
+   i40e_flush_queue_region_all_conf(dev, hw, pf, 0);
+
return 0;
 }
 
@@ -10419,7 +10434,7 @@ i40e_dcb_hw_configure(struct i40e_pf *pf,
  *
  * Returns 0 on success, negative value on failure
  */
-static int
+int
 i40e_dcb_init_configure(struct rte_eth_dev *dev, bool sw_dcb)
 {
struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index ad80f0f..3d237cd 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -260,6 +260,12 @@ enum i40e_flxpld_layer_idx {
 #define I40E_QOS_BW_WEIGHT_MIN 1
 /* The max bandwidth weight is 127. */
 #define I40E_QOS_BW_WEIGHT_MAX 127
+/* The max queue region index is 7. */
+#define I40E_REGION_MAX_INDEX 7
+
+#define I40E_MAX_PERCENT100
+#define I40E_DEFAULT_DCB_APP_NUM1
+#define I40E_DEFAULT_DCB_APP_PRIO   3
 
 /**
  * The overhead from MTU to max frame size.
@@ -541,6 +547,34 @@ struct i40e_ethertype_rule {
struct rte_hash *hash_table;
 };
 
+/* queue region info */
+struct i40e_region_info {
+   /* the region id for this configuration */
+   uint8_t region_id;
+   /* the start queue index for this region */
+   uint8_t queue_start_index;
+   /* the total queue number of this queue region */
+   uint8_t queue_num;
+   /* the total number of user priority for this region */
+   uint8_t user_priority_num;
+   /* the packet's user priority for this region */
+   uint8_t user_priority[I40E_MAX_USER_PRIORITY];
+   /* the total number of flowtype for this region */
+   uint8_t flowtype_num;
+   /**
+* the pctype or hardware flowtype of packet,
+* the specific index for each type has been defined
+* in file i40e_type.h as enum i40e_filter_pctype.
+*/
+   uint8_t hw_flo

Re: [dpdk-dev] [PATCH v4 41/41] net/dpaa: support for extended statistics

[Ferruh Yigit]

On 9/28/2017 3:52 AM, Shreyansh Jain wrote:
>> -Original Message-
>> From: Shreyansh Jain
>> Sent: Thursday, September 28, 2017 7:59 AM
>> To: 'Ferruh Yigit' 
>> Cc: dev@dpdk.org; Hemant Agrawal 
>> Subject: RE: [PATCH v4 41/41] net/dpaa: support for extended statistics
>>
>> Hi Ferruh,
>>
>>> -Original Message-
>>> From: Ferruh Yigit [mailto:ferruh.yi...@intel.com]
>>> Sent: Thursday, September 28, 2017 5:07 AM
>>> To: Shreyansh Jain 
>>> Cc: dev@dpdk.org; Hemant Agrawal 
>>> Subject: Re: [PATCH v4 41/41] net/dpaa: support for extended statistics
>>>
>>> On 9/27/2017 9:26 AM, Shreyansh Jain wrote:
 On Thursday 21 September 2017 06:56 PM, Shreyansh Jain wrote:
> On Monday 18 September 2017 08:27 PM, Ferruh Yigit wrote:
>> On 9/9/2017 12:21 PM, Shreyansh Jain wrote:
>>> From: Hemant Agrawal 
>>>
>>> Signed-off-by: Hemant Agrawal 
>>
>> <...>
>>
>>> +static int
>>> +dpaa_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat
>>> *xstats,
>>> +    unsigned int n)
>>> +{
>>> +    struct dpaa_if *dpaa_intf = dev->data->dev_private;
>>> +    unsigned int i = 0, num = RTE_DIM(dpaa_xstats_strings);
>>> +    uint64_t values[sizeof(struct dpaa_if_stats) / 8];
>>> +
>>> +    if (xstats == NULL)
>>> +    return 0;
>>
>> This is a little not clear from API definition, but I guess when xstats
>> is NULL, it should return num of available stats, "num" for this case.
>> I
>> guess there are PMDs implements both, can you please double check?
>
> Ok. I will check again.

 I checked a number of other ethev implementations. Some like i40e/e1000
 also return 0 when xstats is NULL. Others, like bnx2x and qede don't
 handle this situation.
 All return "num" when passed argument is larger than number of elements
 in the table.

 Though, I think the logic that get_xstats should return its size (num)
 when passed with NULL, looks good to me.
 How does one standardize such semantics for existing APIs?
>>>
>>> Thanks for checking, I guess first we should clarify the API and the
>>> expected behavior [1] and later update required PMDs.
>>>
>>> So for now I think PMD is OK as it is.
>>>
>>>
>>> [1]
>>> I double checked the rte_eth_xstats_get(). It does:
>>>
>>> If xstats == NULL
>>> xcount = dev_ops->xstats_get(dev, NULL, 0);
>>> return count + xcount;
>>>
>>> Intention looks like to returning number of available stats, otherwise
>>> returning "count + 0" will be useless.
>>
>> Makes sense. I missed this and kept looking for implementations.
>> I will at least fix dpaa code.
>  
> On a second though: there might be another issue.
> Application calls rte_eth_xstats_get_names and finds that 'N' xstats exist.
> Thereafter, in a call to rte_eth_xstats_get, xstats==NULL but n=N, so the API 
> would return:
> 
> if (n < count + xcount || xstats == NULL)   
> return count + xcount;
> 
> 'count' is size of generic stats. If drivers->xstats_get were to return 
> xcount='N', the application would think that it has got a positive response.
> See the doxygen page [3] - it states:
> 
> --
> Returns:
> * A positive value lower or equal to size: success.
>   The return value is the number of entries filled in the
>   stats table
> --
> 
> There might be a case where the generic stats are exactly equal to xstats 
> size and application would attempt to access the array.
> 
> I am not even sure why the xstats_get is returning (count + xcount) when the 
> API definition doesn't say that generic+xstat is returned.
> Am I missing something?

Even for rte_eth_xstats_get_names(), returned N is generic + xstat.

dev_ops->xstats_get() manages xstat only, rte_eth_xstats_xxx() on top of
them manages generic + xstat, this seems how it is designed.

for rte_eth_xstats_get(), I guess there is an assumption that when app
provides xstats == NULL, n also should be 0. Perhaps this should be
implemented into API.

> 
> [3] 
> http://dpdk.org/doc/api/rte__ethdev_8h.html#adad5c65f659487db1fefba7d7d902973
> 
>>
>>>
>>> So it looks like expectation from eth_xstats_get_t for that case is
>>> returning xstats size, but this not clear and not documented in API
>> comment.
>>>

 (I can add this info to the API document that you created - but only
 once we know if others will agree to change)
>>
>> Probably this info should be in Doxygen APIs [2].
>>
>> [2]
>> http://dpdk.org/doc/api/rte__ethdev_8h.html#adad5c65f659487db1fefba7d7d902973
>>
> 

Re: [dpdk-dev] dpdk was compiled with CONFIG_RTE_LIBRTE_VHOST_NUMA=n , but on ovs-2.8 compilation, "configure: error: unable to find libnuma, install the dependency package "

[Burakov, Anatoly]

On 28-Sep-17 6:33 AM, Joo Kim wrote:

Hello,


My dpdk was compiled with   CONFIG_RTE_LIBRTE_VHOST_NUMA=n  (which is a
default in config/common_base).

But, I see following error when I  compile ovs-2.8 with dpdk enabled. Is
this expected?   Also ignorable error?

"
. . .
checking for library containing get_mempolicy... no
configure: error: unable to find libnuma, install the dependency package
"


Hi Joo,

Libnuma is a required dependency for building for NUMA support, but if 
you don't want it, please also set CONFIG_RTE_EAL_NUMA_AWARE_HUGEPAGES 
to "n".


--
Thanks,
Anatoly

[dpdk-dev] [PATCH v2 0/4] add net mrvl pmd driver

[Tomasz Duszynski]

Hello,

This patch series introduces the net driver for Marvell Armada 7k/8k
SoCs along with documentation.

Below you can find the list of features which net pmd supports:
* Speed capabilities
* Link status
* Queue start/stop
* MTU update
* Jumbo frame
* Promiscuous mode
* Allmulticast mode
* Unicast MAC filter
* Multicast MAC filter
* RSS hash
* VLAN filter
* CRC offload
* L3 checksum offload
* L4 checksum offload
* Packet type parsing
* Basic stats
* Stats per queue

Tomasz Duszynski (4):
  app: link the whole rte_cfgfile library
  net/mrvl: add mrvl net pmd driver
  doc: add mrvl net pmd documentation
  maintainers: add maintainers for the mrvl net pmd

 MAINTAINERS   |   10 +
 config/common_base|7 +
 doc/guides/nics/features/mrvl.ini |   24 +
 doc/guides/nics/index.rst |1 +
 doc/guides/nics/mrvl.rst  |  151 ++
 drivers/net/Makefile  |2 +
 drivers/net/mrvl/Makefile |   69 +
 drivers/net/mrvl/mrvl_ethdev.c| 2277 +
 drivers/net/mrvl/mrvl_ethdev.h|  115 ++
 drivers/net/mrvl/mrvl_qos.c   |  627 
 drivers/net/mrvl/mrvl_qos.h   |  112 ++
 drivers/net/mrvl/rte_pmd_mrvl_version.map |3 +
 mk/rte.app.mk |3 +-
 13 files changed, 3400 insertions(+), 1 deletion(-)
 create mode 100644 doc/guides/nics/features/mrvl.ini
 create mode 100644 doc/guides/nics/mrvl.rst
 create mode 100644 drivers/net/mrvl/Makefile
 create mode 100644 drivers/net/mrvl/mrvl_ethdev.c
 create mode 100644 drivers/net/mrvl/mrvl_ethdev.h
 create mode 100644 drivers/net/mrvl/mrvl_qos.c
 create mode 100644 drivers/net/mrvl/mrvl_qos.h
 create mode 100644 drivers/net/mrvl/rte_pmd_mrvl_version.map

--
2.7.4

[dpdk-dev] [PATCH v2 1/4] app: link the whole rte_cfgfile library

[Tomasz Duszynski]

MRVL net pmd needs rte_cfgfile to parse QoS configuration file thus
librte_pmd_mrvl.a contains undefined symbols from librte_cfgfile.a.

As a result linking applications under app/ directory will fail
because librte_cfgfile.a comes before librte_pmd_mrvl.a during
the linking stage.

Linking the whole librte_cfgfile.a solves the issue.

Signed-off-by: Jacek Siuda 
Signed-off-by: Tomasz Duszynski 
---
 v2:
 * Changed commit message to explain problem better.

 mk/rte.app.mk | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mk/rte.app.mk b/mk/rte.app.mk
index c25fdd9..94568a8 100644
--- a/mk/rte.app.mk
+++ b/mk/rte.app.mk
@@ -81,10 +81,10 @@ _LDLIBS-$(CONFIG_RTE_LIBRTE_POWER)  += -lrte_power

 _LDLIBS-$(CONFIG_RTE_LIBRTE_TIMER)  += -lrte_timer
 _LDLIBS-$(CONFIG_RTE_LIBRTE_EFD)+= -lrte_efd
-_LDLIBS-$(CONFIG_RTE_LIBRTE_CFGFILE)+= -lrte_cfgfile

 _LDLIBS-y += --whole-archive

+_LDLIBS-$(CONFIG_RTE_LIBRTE_CFGFILE)+= -lrte_cfgfile
 _LDLIBS-$(CONFIG_RTE_LIBRTE_HASH)   += -lrte_hash
 _LDLIBS-$(CONFIG_RTE_LIBRTE_VHOST)  += -lrte_vhost
 _LDLIBS-$(CONFIG_RTE_LIBRTE_KVARGS) += -lrte_kvargs
--
2.7.4

[dpdk-dev] [PATCH v2 3/4] doc: add mrvl net pmd documentation

[Tomasz Duszynski]

Add documentation for the MRVL NET PMD driver.

Signed-off-by: Jacek Siuda 
Signed-off-by: Tomasz Duszynski 
---
 doc/guides/nics/features/mrvl.ini |  24 ++
 doc/guides/nics/index.rst |   1 +
 doc/guides/nics/mrvl.rst  | 151 ++
 3 files changed, 176 insertions(+)
 create mode 100644 doc/guides/nics/features/mrvl.ini
 create mode 100644 doc/guides/nics/mrvl.rst

diff --git a/doc/guides/nics/features/mrvl.ini 
b/doc/guides/nics/features/mrvl.ini
new file mode 100644
index 000..781d6dc
--- /dev/null
+++ b/doc/guides/nics/features/mrvl.ini
@@ -0,0 +1,24 @@
+;
+; Supported features of the 'mrvl' network poll mode driver.
+;
+; Refer to default.ini for the full list of available PMD features.
+;
+[Features]
+Speed capabilities   = Y
+Link status  = Y
+MTU update   = Y
+Jumbo frame  = Y
+Promiscuous mode = Y
+Allmulticast mode= Y
+Unicast MAC filter   = Y
+Multicast MAC filter = Y
+RSS hash = Y
+VLAN filter  = Y
+CRC offload  = Y
+L3 checksum offload  = Y
+L4 checksum offload  = Y
+Packet type parsing  = Y
+Basic stats  = Y
+Stats per queue  = Y
+ARMv8= Y
+Usage doc= Y
diff --git a/doc/guides/nics/index.rst b/doc/guides/nics/index.rst
index 36f4f3f..7192f63 100644
--- a/doc/guides/nics/index.rst
+++ b/doc/guides/nics/index.rst
@@ -55,6 +55,7 @@ Network Interface Controller Drivers
 liquidio
 mlx4
 mlx5
+mrvl
 nfp
 qede
 sfc_efx
diff --git a/doc/guides/nics/mrvl.rst b/doc/guides/nics/mrvl.rst
new file mode 100644
index 000..241d89b
--- /dev/null
+++ b/doc/guides/nics/mrvl.rst
@@ -0,0 +1,151 @@
+..  BSD LICENSE
+Copyright(c) 2017 Semihalf. All rights reserved.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions
+are met:
+
+  * Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+  * Redistributions in binary form must reproduce the above copyright
+notice, this list of conditions and the following disclaimer in
+the documentation and/or other materials provided with the
+distribution.
+  * Neither the name of Semihalf nor the names of its
+contributors may be used to endorse or promote products derived
+from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+MRVL Poll Mode Driver
+==
+
+The MRVL PMD (librte_pmd_mrvl) provides poll mode driver support
+for the Marvell PPv2 (Packet Processor v2) 1/10 Gbps adapter.
+
+.. Note::
+
+   Due to external dependencies, this driver is disabled by default. It must
+   be enabled manually by setting relevant configuration option manually.
+   Please refer to `Config File Options`_ section for further details.
+
+
+Features
+
+
+Features of the MRVL PMD are:
+
+- Speed capabilities
+- Link status
+- Queue start/stop
+- MTU update
+- Jumbo frame
+- Promiscuous mode
+- Allmulticast mode
+- Unicast MAC filter
+- Multicast MAC filter
+- RSS hash
+- VLAN filter
+- CRC offload
+- L3 checksum offload
+- L4 checksum offload
+- Packet type parsing
+- Basic stats
+- Stats per queue
+
+
+Limitations
+---
+
+- Number of lcores is limited to 9 by MUSDK internal design. If more lcores
+  need to be allocated, locking will have to be considered. Number of available
+  lcores can be changed via ``MRVL_MUSDK_HIFS_RESERVED`` define in
+  ``mrvl_ethdev.c`` source file.
+
+- Flushing vlans added for filtering is not possible due to MUSDK missing
+  functionality. Current workaround is to reset board so that PPv2 has a
+  chance to start in a sane state.
+
+
+Prerequisites
+-
+
+- MUSDK (Marvell User-Space SDK) sources available
+  `here 
`_.
+
+MUSDK is a light-weight library that provides direct access to Marvell's
+PPv2 (Packet Processor v2). Alternatively prebuilt MUSDK library can be
+requested from `

[dpdk-dev] [PATCH v2 4/4] maintainers: add maintainers for the mrvl net pmd

[Tomasz Duszynski]

Signed-off-by: Jacek Siuda 
Signed-off-by: Tomasz Duszynski 
---
 MAINTAINERS | 10 ++
 1 file changed, 10 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index a0cd75e..d4810cf 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -393,6 +393,16 @@ F: drivers/net/mlx5/
 F: doc/guides/nics/mlx5.rst
 F: doc/guides/nics/features/mlx5.ini
 
+Marvell mrvl
+M: Jacek Siuda 
+M: Tomasz Duszynski 
+M: Dmitri Epshtein 
+M: Natalie Samsonov 
+M: Jianbo Liu 
+F: drivers/net/mrvl/
+F: doc/guides/nics/mrvl.rst
+F: doc/guides/nics/features/mrvl.ini
+
 Netcope szedata2
 M: Matej Vido 
 F: drivers/net/szedata2/
-- 
2.7.4

[dpdk-dev] [PATCH v2 2/4] net/mrvl: add mrvl net pmd driver

[Tomasz Duszynski]

Add support for the Marvell PPv2 (Packet Processor v2) 1/10 Gbps adapter.
Driver is based on external, publicly available, light-weight Marvell
MUSDK library that provides access to network packet processor.

Driver comes with support for the following features:

* Speed capabilities
* Link status
* Queue start/stop
* MTU update
* Jumbo frame
* Promiscuous mode
* Allmulticast mode
* Unicast MAC filter
* Multicast MAC filter
* RSS hash
* VLAN filter
* CRC offload
* L3 checksum offload
* L4 checksum offload
* Packet type parsing
* Basic stats
* Stats per queue

Driver was engineered cooperatively by Semihalf and Marvell teams.

Semihalf:
Jacek Siuda 
Tomasz Duszynski 

Marvell:
Dmitri Epshtein 
Natalie Samsonov 

Signed-off-by: Jacek Siuda 
Signed-off-by: Tomasz Duszynski 
---
 v2:
 * Removed bunch of checkpatch warnings about unnecessary parentheses.

 config/common_base|7 +
 drivers/net/Makefile  |2 +
 drivers/net/mrvl/Makefile |   69 +
 drivers/net/mrvl/mrvl_ethdev.c| 2277 +
 drivers/net/mrvl/mrvl_ethdev.h|  115 ++
 drivers/net/mrvl/mrvl_qos.c   |  627 
 drivers/net/mrvl/mrvl_qos.h   |  112 ++
 drivers/net/mrvl/rte_pmd_mrvl_version.map |3 +
 mk/rte.app.mk |1 +
 9 files changed, 3213 insertions(+)
 create mode 100644 drivers/net/mrvl/Makefile
 create mode 100644 drivers/net/mrvl/mrvl_ethdev.c
 create mode 100644 drivers/net/mrvl/mrvl_ethdev.h
 create mode 100644 drivers/net/mrvl/mrvl_qos.c
 create mode 100644 drivers/net/mrvl/mrvl_qos.h
 create mode 100644 drivers/net/mrvl/rte_pmd_mrvl_version.map

diff --git a/config/common_base b/config/common_base
index 5e97a08..d05a60c 100644
--- a/config/common_base
+++ b/config/common_base
@@ -262,6 +262,13 @@ CONFIG_RTE_LIBRTE_NFP_PMD=n
 CONFIG_RTE_LIBRTE_NFP_DEBUG=n

 #
+# Compile Marvell PMD driver
+#
+CONFIG_RTE_LIBRTE_MRVL_PMD=n
+CONFIG_RTE_LIBRTE_MRVL_DEBUG=n
+CONFIG_RTE_MRVL_MUSDK_DMA_MEMSIZE=41943040
+
+#
 # Compile burst-oriented Broadcom BNXT PMD driver
 #
 CONFIG_RTE_LIBRTE_BNXT_PMD=y
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index d33c959..4a3a205 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -73,6 +73,8 @@ DIRS-$(CONFIG_RTE_LIBRTE_MLX4_PMD) += mlx4
 DEPDIRS-mlx4 = $(core-libs)
 DIRS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += mlx5
 DEPDIRS-mlx5 = $(core-libs)
+DIRS-$(CONFIG_RTE_LIBRTE_MRVL_PMD) += mrvl
+DEPDIRS-mrvl = $(core-libs)
 DIRS-$(CONFIG_RTE_LIBRTE_NFP_PMD) += nfp
 DEPDIRS-nfp = $(core-libs)
 DIRS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt
diff --git a/drivers/net/mrvl/Makefile b/drivers/net/mrvl/Makefile
new file mode 100644
index 000..ab53f49
--- /dev/null
+++ b/drivers/net/mrvl/Makefile
@@ -0,0 +1,69 @@
+#   BSD LICENSE
+#
+#   Copyright(c) 2017 Semihalf. All rights reserved.
+#
+#   Redistribution and use in source and binary forms, with or without
+#   modification, are permitted provided that the following conditions
+#   are met:
+#
+# * Redistributions of source code must retain the above copyright
+#   notice, this list of conditions and the following disclaimer.
+# * Redistributions in binary form must reproduce the above copyright
+#   notice, this list of conditions and the following disclaimer in
+#   the documentation and/or other materials provided with the
+#   distribution.
+# * Neither the name of Semihalf nor the names of its
+#   contributors may be used to endorse or promote products derived
+#   from this software without specific prior written permission.
+#
+#   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+#   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+#   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+#   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+#   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+#   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+#   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+#   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+#   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+#   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+#   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+include $(RTE_SDK)/mk/rte.vars.mk
+
+ifneq ($(MAKECMDGOALS),clean)
+ifneq ($(MAKECMDGOALS),config)
+ifeq ($(LIBMUSDK_PATH),)
+$(error "Please define LIBMUSDK_PATH environment variable")
+endif
+ifeq ($(CONFIG_RTE_LIBRTE_CFGFILE),n)
+$(error "RTE_LIBRTE_CFGFILE must be enabled in configuration!")
+endif
+endif
+endif
+
+# library name
+LIB = librte_pmd_mrvl.a
+
+# library version
+LIBABIVER := 1
+
+# versioning export map
+EXPORT_MAP := rte_pmd_mrvl_version.map
+
+# external library dependencies
+CFLAGS += -I$(LIBMUSDK_PATH)/include

[dpdk-dev] [PATCH v2 0/4] add crypto mrvl pmd driver

[Tomasz Duszynski]

Hello,

This patch series introduces crypto driver for Marvell Armada 7k/8k
SoCs along with documentation and crypto pmd driver tests.

Below you can find the list of features which crypto pmd supports:
* Symmetric crypto
* Sym operation chaining
* AES CBC (128)
* AES CBC (192)
* AES CBC (256)
* AES CTR (128)
* AES CTR (192)
* AES CTR (256)
* 3DES CBC
* 3DES CTR
* MD5
* MD5 HMAC
* SHA1
* SHA1 HMAC
* SHA256
* SHA256 HMAC
* SHA384
* SHA384 HMAC
* SHA512
* SHA512 HMAC
* AES GCM (128)

Tomasz Duszynski (4):
  crypto/mrvl: add mrvl crypto pmd driver
  doc: add mrvl crypto pmd documentation
  maintainers: add maintainers for the mrvl crypto pmd
  test: add mrvl crypto pmd unit tests

 MAINTAINERS  |  10 +
 config/common_base   |   6 +
 doc/guides/cryptodevs/features/mrvl.ini  |  42 ++
 doc/guides/cryptodevs/index.rst  |   1 +
 doc/guides/cryptodevs/mrvl.rst   | 198 ++
 drivers/crypto/Makefile  |   2 +
 drivers/crypto/mrvl/Makefile |  63 ++
 drivers/crypto/mrvl/rte_mrvl_compat.h|  48 ++
 drivers/crypto/mrvl/rte_mrvl_pmd.c   | 870 +++
 drivers/crypto/mrvl/rte_mrvl_pmd_ops.c   | 787 
 drivers/crypto/mrvl/rte_mrvl_pmd_private.h   | 121 
 drivers/crypto/mrvl/rte_mrvl_pmd_version.map |   3 +
 mk/rte.app.mk|   1 +
 test/test/test_cryptodev.c   | 168 ++
 test/test/test_cryptodev.h   |   1 +
 test/test/test_cryptodev_aes_test_vectors.h  |  72 ++-
 test/test/test_cryptodev_blockcipher.c   |   9 +-
 test/test/test_cryptodev_blockcipher.h   |   1 +
 test/test/test_cryptodev_des_test_vectors.h  |  24 +-
 19 files changed, 2394 insertions(+), 33 deletions(-)
 create mode 100644 doc/guides/cryptodevs/features/mrvl.ini
 create mode 100644 doc/guides/cryptodevs/mrvl.rst
 create mode 100644 drivers/crypto/mrvl/Makefile
 create mode 100644 drivers/crypto/mrvl/rte_mrvl_compat.h
 create mode 100644 drivers/crypto/mrvl/rte_mrvl_pmd.c
 create mode 100644 drivers/crypto/mrvl/rte_mrvl_pmd_ops.c
 create mode 100644 drivers/crypto/mrvl/rte_mrvl_pmd_private.h
 create mode 100644 drivers/crypto/mrvl/rte_mrvl_pmd_version.map

--
2.7.4

[dpdk-dev] [PATCH v2 2/4] doc: add mrvl crypto pmd documentation

[Tomasz Duszynski]

Add documentation for the MRVL CRYPTO PMD driver.

Signed-off-by: Jacek Siuda 
Signed-off-by: Tomasz Duszynski 
---
 doc/guides/cryptodevs/features/mrvl.ini |  42 +++
 doc/guides/cryptodevs/index.rst |   1 +
 doc/guides/cryptodevs/mrvl.rst  | 198 
 3 files changed, 241 insertions(+)
 create mode 100644 doc/guides/cryptodevs/features/mrvl.ini
 create mode 100644 doc/guides/cryptodevs/mrvl.rst

diff --git a/doc/guides/cryptodevs/features/mrvl.ini 
b/doc/guides/cryptodevs/features/mrvl.ini
new file mode 100644
index 000..6d2fe6a
--- /dev/null
+++ b/doc/guides/cryptodevs/features/mrvl.ini
@@ -0,0 +1,42 @@
+; Supported features of the 'mrvl' crypto driver.
+;
+; Refer to default.ini for the full list of available PMD features.
+;
+[Features]
+Symmetric crypto   = Y
+Sym operation chaining = Y
+
+;
+; Supported crypto algorithms of a default crypto driver.
+;
+[Cipher]
+AES CBC (128)  = Y
+AES CBC (192)  = Y
+AES CBC (256)  = Y
+AES CTR (128)  = Y
+AES CTR (192)  = Y
+AES CTR (256)  = Y
+3DES CBC   = Y
+3DES CTR   = Y
+
+;
+; Supported authentication algorithms of a default crypto driver.
+;
+[Auth]
+MD5  = Y
+MD5 HMAC = Y
+SHA1 = Y
+SHA1 HMAC= Y
+SHA256   = Y
+SHA256 HMAC  = Y
+SHA384   = Y
+SHA384 HMAC  = Y
+SHA512   = Y
+SHA512 HMAC  = Y
+AES GMAC = Y
+
+;
+; Supported AEAD algorithms of a default crypto driver.
+;
+[AEAD]
+AES GCM (128) = Y
diff --git a/doc/guides/cryptodevs/index.rst b/doc/guides/cryptodevs/index.rst
index 361b82d..a8ee0eb 100644
--- a/doc/guides/cryptodevs/index.rst
+++ b/doc/guides/cryptodevs/index.rst
@@ -42,6 +42,7 @@ Crypto Device Drivers
 dpaa2_sec
 kasumi
 openssl
+mrvl
 null
 scheduler
 snow3g
diff --git a/doc/guides/cryptodevs/mrvl.rst b/doc/guides/cryptodevs/mrvl.rst
new file mode 100644
index 000..f5a83dc
--- /dev/null
+++ b/doc/guides/cryptodevs/mrvl.rst
@@ -0,0 +1,198 @@
+..  BSD LICENSE
+Copyright(c) 2017 Semihalf. All rights reserved.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions
+are met:
+
+  * Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+  * Redistributions in binary form must reproduce the above copyright
+notice, this list of conditions and the following disclaimer in
+the documentation and/or other materials provided with the
+distribution.
+  * Neither the name of Semihalf nor the names of its
+contributors may be used to endorse or promote products derived
+from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+MRVL Crypto Poll Mode Driver
+
+
+The MRVL CRYPTO PMD (**librte_crypto_mrvl_pmd**) provides poll mode crypto 
driver
+support by utilizing MUSDK library, which provides cryptographic operations
+acceleration by using Security Acceleration Engine (EIP197) directly from
+user-space with minimum overhead and high performance.
+
+Features
+
+
+MRVL CRYPTO PMD has support for:
+
+* Symmetric crypto
+* Sym operation chaining
+* AES CBC (128)
+* AES CBC (192)
+* AES CBC (256)
+* AES CTR (128)
+* AES CTR (192)
+* AES CTR (256)
+* 3DES CBC
+* 3DES CTR
+* MD5
+* MD5 HMAC
+* SHA1
+* SHA1 HMAC
+* SHA256
+* SHA256 HMAC
+* SHA384
+* SHA384 HMAC
+* SHA512
+* SHA512 HMAC
+* AES GCM (128)
+
+Limitations
+---
+
+* Hardware only supports scenarios where ICV (digest buffer) is placed just
+  after the authenticated data. Other placement will result in error.
+
+Installation
+
+
+MRVL CRYPTO PMD driver compilation is disabled by default due to external 
dependencies.
+Currently there are two driver specific compilation options in
+``config/common_base`` available:
+
+- ``CONFIG_RTE_LIBRTE_MRVL_CRYPTO`` (default ``n``)
+
+Toggle compilation of the librte_pmd_mrvl driver.
+
+- ``CONFIG_RTE_LIBRTE_MRVL_CRYPTO_DEBUG`` (default ``n``)
+
+Toggle display of debugging messages.
+
+During compilation external MUSD

[dpdk-dev] [PATCH v2 1/4] crypto/mrvl: add mrvl crypto pmd driver

[Tomasz Duszynski]

Add support for the Marvell Security Crypto Accelerator EIP197.
Driver is based on external, publicly available, Marvell MUSDK
library that provides access to the hardware with minimum overhead
and high performance.

Driver comes with support for the following features:

* Symmetric crypto
* Sym operation chaining
* AES CBC (128)
* AES CBC (192)
* AES CBC (256)
* AES CTR (128)
* AES CTR (192)
* AES CTR (256)
* 3DES CBC
* 3DES CTR
* MD5
* MD5 HMAC
* SHA1
* SHA1 HMAC
* SHA256
* SHA256 HMAC
* SHA384
* SHA384 HMAC
* SHA512
* SHA512 HMAC
* AES GCM (128)

Driver was engineered cooperatively by Semihalf and Marvell teams.

Semihalf:
Jacek Siuda 
Tomasz Duszynski 

Marvell:
Dmitri Epshtein 
Natalie Samsonov 

Signed-off-by: Jacek Siuda 
Signed-off-by: Tomasz Duszynski 
---
 config/common_base   |   6 +
 drivers/crypto/Makefile  |   2 +
 drivers/crypto/mrvl/Makefile |  63 ++
 drivers/crypto/mrvl/rte_mrvl_compat.h|  48 ++
 drivers/crypto/mrvl/rte_mrvl_pmd.c   | 870 +++
 drivers/crypto/mrvl/rte_mrvl_pmd_ops.c   | 787 
 drivers/crypto/mrvl/rte_mrvl_pmd_private.h   | 121 
 drivers/crypto/mrvl/rte_mrvl_pmd_version.map |   3 +
 mk/rte.app.mk|   1 +
 9 files changed, 1901 insertions(+)
 create mode 100644 drivers/crypto/mrvl/Makefile
 create mode 100644 drivers/crypto/mrvl/rte_mrvl_compat.h
 create mode 100644 drivers/crypto/mrvl/rte_mrvl_pmd.c
 create mode 100644 drivers/crypto/mrvl/rte_mrvl_pmd_ops.c
 create mode 100644 drivers/crypto/mrvl/rte_mrvl_pmd_private.h
 create mode 100644 drivers/crypto/mrvl/rte_mrvl_pmd_version.map

diff --git a/config/common_base b/config/common_base
index d05a60c..e9b181e 100644
--- a/config/common_base
+++ b/config/common_base
@@ -521,6 +521,12 @@ CONFIG_RTE_LIBRTE_PMD_CRYPTO_SCHEDULER_DEBUG=n
 CONFIG_RTE_LIBRTE_PMD_NULL_CRYPTO=y
 
 #
+# Compile PMD for Marvell Crypto device
+#
+CONFIG_RTE_LIBRTE_PMD_MRVL_CRYPTO=n
+CONFIG_RTE_LIBRTE_PMD_MRVL_CRYPTO_DEBUG=n
+
+#
 # Compile generic event device library
 #
 CONFIG_RTE_LIBRTE_EVENTDEV=y
diff --git a/drivers/crypto/Makefile b/drivers/crypto/Makefile
index 7a719b9..1d7db5b 100644
--- a/drivers/crypto/Makefile
+++ b/drivers/crypto/Makefile
@@ -51,6 +51,8 @@ DIRS-$(CONFIG_RTE_LIBRTE_PMD_KASUMI) += kasumi
 DEPDIRS-kasumi = $(core-libs)
 DIRS-$(CONFIG_RTE_LIBRTE_PMD_ZUC) += zuc
 DEPDIRS-zuc = $(core-libs)
+DIRS-$(CONFIG_RTE_LIBRTE_PMD_MRVL_CRYPTO) += mrvl
+DEPDIRS-mrvl = $(core-libs)
 DIRS-$(CONFIG_RTE_LIBRTE_PMD_NULL_CRYPTO) += null
 DEPDIRS-null = $(core-libs)
 DIRS-$(CONFIG_RTE_LIBRTE_PMD_DPAA2_SEC) += dpaa2_sec
diff --git a/drivers/crypto/mrvl/Makefile b/drivers/crypto/mrvl/Makefile
new file mode 100644
index 000..a00a19e
--- /dev/null
+++ b/drivers/crypto/mrvl/Makefile
@@ -0,0 +1,63 @@
+#   BSD LICENSE
+#
+#   Copyright (C) Semihalf 2017. All rights reserved.
+#
+#   Redistribution and use in source and binary forms, with or without
+#   modification, are permitted provided that the following conditions
+#   are met:
+#
+# * Redistributions of source code must retain the above copyright
+#   notice, this list of conditions and the following disclaimer.
+# * Redistributions in binary form must reproduce the above copyright
+#   notice, this list of conditions and the following disclaimer in
+#   the documentation and/or other materials provided with the
+#   distribution.
+# * Neither the name of Semihalf nor the names of its
+#   contributors may be used to endorse or promote products derived
+#   from this software without specific prior written permission.
+#
+#   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+#   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+#   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+#   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+#   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+#   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+#   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+#   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+#   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+#   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+#   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+include $(RTE_SDK)/mk/rte.vars.mk
+
+ifneq ($(MAKECMDGOALS),clean)
+ifneq ($(MAKECMDGOALS),config)
+ifeq ($(LIBMUSDK_PATH),)
+$(error "Please define LIBMUSDK_PATH environment variable")
+endif
+endif
+endif
+
+# library name
+LIB = librte_pmd_mrvl_crypto.a
+
+# build flags
+CFLAGS += -O3
+CFLAGS += $(WERROR_FLAGS)
+CFLAGS += -I$(LIBMUSDK_PATH)/include
+CFLAGS += -DMVCONF_ARCH_DMA_ADDR_T_64BIT
+
+# library version
+LIBABIVER := 1
+
+# versioning export map
+EXPORT_MAP := rte_mrvl_pmd_version

[dpdk-dev] [PATCH v2 3/4] maintainers: add maintainers for the mrvl crypto pmd

[Tomasz Duszynski]

Signed-off-by: Jacek Siuda 
Signed-off-by: Tomasz Duszynski 
---
 MAINTAINERS | 10 ++
 1 file changed, 10 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index d4810cf..93800d2 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -561,6 +561,16 @@ F: drivers/crypto/dpaa2_sec/
 F: doc/guides/cryptodevs/dpaa2_sec.rst
 F: doc/guides/cryptodevs/features/dpaa2_sec.ini
 
+MARVELL MRVL PMD
+M: Jacek Siuda 
+M: Tomasz Duszynski 
+M: Dmitri Epshtein 
+M: Natalie Samsonov 
+M: Jianbo Liu 
+F: drivers/crypto/mrvl/
+F: doc/guides/cryptodevs/mrvl.rst
+F: doc/guides/cryptodevs/mrvl.ini
+
 SNOW 3G PMD
 M: Pablo de Lara 
 F: drivers/crypto/snow3g/
-- 
2.7.4

[dpdk-dev] [PATCH v2 4/4] test: add mrvl crypto pmd unit tests

[Tomasz Duszynski]

Add unit tests for MRVL CRYPTO PMD driver.

Signed-off-by: Jacek Siuda 
Signed-off-by: Tomasz Duszynski 
---
 test/test/test_cryptodev.c  | 168 
 test/test/test_cryptodev.h  |   1 +
 test/test/test_cryptodev_aes_test_vectors.h |  72 
 test/test/test_cryptodev_blockcipher.c  |   9 +-
 test/test/test_cryptodev_blockcipher.h  |   1 +
 test/test/test_cryptodev_des_test_vectors.h |  24 ++--
 6 files changed, 242 insertions(+), 33 deletions(-)

diff --git a/test/test/test_cryptodev.c b/test/test/test_cryptodev.c
index a4116c6..9fb6a2c 100644
--- a/test/test/test_cryptodev.c
+++ b/test/test/test_cryptodev.c
@@ -341,6 +341,28 @@ testsuite_setup(void)
}
}
 
+   /* Create a MRVL device if required */
+   if (gbl_driver_id == rte_cryptodev_driver_id_get(
+   RTE_STR(CRYPTODEV_MRVL_PMD))) {
+#ifndef RTE_LIBRTE_PMD_MRVL_CRYPTO
+   RTE_LOG(ERR, USER1, "CONFIG_RTE_LIBRTE_PMD_MRVL_CRYPTO must be"
+   " enabled in config file to run this testsuite.\n");
+   return TEST_FAILED;
+#endif
+   nb_devs = rte_cryptodev_device_count_by_driver(
+   rte_cryptodev_driver_id_get(
+   RTE_STR(CRYPTODEV_NAME_MRVL_PMD)));
+   if (nb_devs < 1) {
+   ret = rte_vdev_init(
+   RTE_STR(CRYPTODEV_NAME_MRVL_PMD),
+   NULL);
+
+   TEST_ASSERT(ret == 0, "Failed to create "
+   "instance of pmd : %s",
+   RTE_STR(CRYPTODEV_NAME_MRVL_PMD));
+   }
+   }
+
 #ifdef RTE_LIBRTE_PMD_CRYPTO_SCHEDULER
if (gbl_driver_id == rte_cryptodev_driver_id_get(
RTE_STR(CRYPTODEV_NAME_SCHEDULER_PMD))) {
@@ -1862,6 +1884,101 @@ test_AES_chain_armv8_all(void)
return TEST_SUCCESS;
 }
 
+static int
+test_AES_chain_mrvl_all(void)
+{
+   struct crypto_testsuite_params *ts_params = &testsuite_params;
+   int status;
+
+   status = test_blockcipher_all_tests(ts_params->mbuf_pool,
+   ts_params->op_mpool,
+   ts_params->session_mpool,
+   ts_params->valid_devs[0],
+   rte_cryptodev_driver_id_get(
+   RTE_STR(CRYPTODEV_NAME_MRVL_PMD)),
+   BLKCIPHER_AES_CHAIN_TYPE);
+
+   TEST_ASSERT_EQUAL(status, 0, "Test failed");
+
+   return TEST_SUCCESS;
+}
+
+static int
+test_AES_cipheronly_mrvl_all(void)
+{
+   struct crypto_testsuite_params *ts_params = &testsuite_params;
+   int status;
+
+   status = test_blockcipher_all_tests(ts_params->mbuf_pool,
+   ts_params->op_mpool,
+   ts_params->session_mpool,
+   ts_params->valid_devs[0],
+   rte_cryptodev_driver_id_get(
+   RTE_STR(CRYPTODEV_NAME_MRVL_PMD)),
+   BLKCIPHER_AES_CIPHERONLY_TYPE);
+
+   TEST_ASSERT_EQUAL(status, 0, "Test failed");
+
+   return TEST_SUCCESS;
+}
+
+static int
+test_authonly_mrvl_all(void)
+{
+   struct crypto_testsuite_params *ts_params = &testsuite_params;
+   int status;
+
+   status = test_blockcipher_all_tests(ts_params->mbuf_pool,
+   ts_params->op_mpool,
+   ts_params->session_mpool,
+   ts_params->valid_devs[0],
+   rte_cryptodev_driver_id_get(
+   RTE_STR(CRYPTODEV_NAME_MRVL_PMD)),
+   BLKCIPHER_AUTHONLY_TYPE);
+
+   TEST_ASSERT_EQUAL(status, 0, "Test failed");
+
+   return TEST_SUCCESS;
+}
+
+static int
+test_3DES_chain_mrvl_all(void)
+{
+   struct crypto_testsuite_params *ts_params = &testsuite_params;
+   int status;
+
+   status = test_blockcipher_all_tests(ts_params->mbuf_pool,
+   ts_params->op_mpool,
+   ts_params->session_mpool,
+   ts_params->valid_devs[0],
+   rte_cryptodev_driver_id_get(
+   RTE_STR(CRYPTODEV_NAME_MRVL_PMD)),
+   BLKCIPHER_3DES_CHAIN_TYPE);
+
+   TEST_ASSERT_EQUAL(status, 0, "Test failed");
+
+   return TEST_SUCCESS;
+}
+
+static int
+test_3DES_cipheronly_mrvl_all(void)
+{
+   struct crypto_testsuite_params *ts_params = &testsuite_params;
+   int status;
+
+   status = test_blockcipher_all_tests(ts_params->mbuf_pool,
+   ts_params->op_mpool,
+   ts_params->session_mpool,
+   ts_params->valid_devs[0],
+   rte_cryptodev_driver_id_get(
+   RTE_STR(CRYPTODEV_NAME_MRVL_PMD)),
+   BLKCIPHER_3DES_CIPHERONLY_TYPE);
+
+   TEST_ASSERT_EQUAL(status, 0, "Test failed");
+
+   return TEST_SUCCESS;
+}
+
 /* * SNOW 3G Tests * */
 static int
 create_wireless_algo_hash_session(uint8_t dev_id,
@@ -8939,6 +9056,40 @@ static struct unit_test_suite cryptodev_armv8_testsuite  
= {
 

[dpdk-dev] How to use ixgbe module log?

[Sam]

Hi all,

I'm working on ixgbe, I want to enable "PMD_RX_LOG" and "PMD_TX_LOG", how
to do  this?

As for normal application, I could use --log-level param to start, but
ixgbe is kernel module, how to do this?

Thank you~

Re: [dpdk-dev] [PATCH v2 0/4] add net mrvl pmd driver

[Ferruh Yigit]

On 9/28/2017 11:22 AM, Tomasz Duszynski wrote:
> Hello,
> 
> This patch series introduces the net driver for Marvell Armada 7k/8k
> SoCs along with documentation.
> 
> Below you can find the list of features which net pmd supports:
> * Speed capabilities
> * Link status
> * Queue start/stop
> * MTU update
> * Jumbo frame
> * Promiscuous mode
> * Allmulticast mode
> * Unicast MAC filter
> * Multicast MAC filter
> * RSS hash
> * VLAN filter
> * CRC offload
> * L3 checksum offload
> * L4 checksum offload
> * Packet type parsing
> * Basic stats
> * Stats per queue
> 
> Tomasz Duszynski (4):
>   app: link the whole rte_cfgfile library
>   net/mrvl: add mrvl net pmd driver
>   doc: add mrvl net pmd documentation
>   maintainers: add maintainers for the mrvl net pmd

Hi Tomasz,

Good to see more vendors are contributing, welcome!

This patchset arrived a little late for this release cycle, I still
would like to get it but priority will be the existing waiting drivers,
so no promises :)

Thanks,
ferruh

<...>

[dpdk-dev] [PATCH v5 02/40] bus/dpaa: introduce NXP DPAA Bus driver skeleton

[Shreyansh Jain]

Signed-off-by: Shreyansh Jain 
Signed-off-by: Hemant Agrawal 
---
 MAINTAINERS   |   5 +
 config/common_base|   3 +
 config/defconfig_arm64-dpaa-linuxapp-gcc  |   4 +
 drivers/bus/Makefile  |   3 +
 drivers/bus/dpaa/Makefile |  58 +
 drivers/bus/dpaa/dpaa_bus.c   | 207 ++
 drivers/bus/dpaa/rte_bus_dpaa_version.map |   8 ++
 drivers/bus/dpaa/rte_dpaa_bus.h   | 148 +
 drivers/bus/dpaa/rte_dpaa_logs.h  |  65 ++
 9 files changed, 501 insertions(+)
 create mode 100644 drivers/bus/dpaa/Makefile
 create mode 100644 drivers/bus/dpaa/dpaa_bus.c
 create mode 100644 drivers/bus/dpaa/rte_bus_dpaa_version.map
 create mode 100644 drivers/bus/dpaa/rte_dpaa_bus.h
 create mode 100644 drivers/bus/dpaa/rte_dpaa_logs.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 8df2a7f..c566962 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -408,6 +408,11 @@ F: drivers/net/nfp/
 F: doc/guides/nics/nfp.rst
 F: doc/guides/nics/features/nfp.ini
 
+NXP dpaa
+M: Hemant Agrawal 
+M: Shreyansh Jain 
+F: drivers/bus/dpaa/
+
 NXP dpaa2
 M: Hemant Agrawal 
 M: Shreyansh Jain 
diff --git a/config/common_base b/config/common_base
index 439f3cc..fc1cdca 100644
--- a/config/common_base
+++ b/config/common_base
@@ -301,6 +301,9 @@ CONFIG_RTE_LIBRTE_LIO_DEBUG_TX=n
 CONFIG_RTE_LIBRTE_LIO_DEBUG_MBOX=n
 CONFIG_RTE_LIBRTE_LIO_DEBUG_REGS=n
 
+# NXP DPAA Bus
+CONFIG_RTE_LIBRTE_DPAA_BUS=n
+
 #
 # Compile NXP DPAA2 FSL-MC Bus
 #
diff --git a/config/defconfig_arm64-dpaa-linuxapp-gcc 
b/config/defconfig_arm64-dpaa-linuxapp-gcc
index 5bca012..8316fc9 100644
--- a/config/defconfig_arm64-dpaa-linuxapp-gcc
+++ b/config/defconfig_arm64-dpaa-linuxapp-gcc
@@ -45,3 +45,7 @@ CONFIG_RTE_MAX_LCORE=4
 CONFIG_RTE_MAX_NUMA_NODES=1
 CONFIG_RTE_CACHE_LINE_SIZE=64
 CONFIG_RTE_PKTMBUF_HEADROOM=128
+
+# NXP DPAA Bus
+CONFIG_RTE_LIBRTE_DPAA_BUS=y
+CONFIG_RTE_LIBRTE_DPAA_DEBUG_DRIVER=n
diff --git a/drivers/bus/Makefile b/drivers/bus/Makefile
index 0224214..6cb6466 100644
--- a/drivers/bus/Makefile
+++ b/drivers/bus/Makefile
@@ -32,6 +32,9 @@ include $(RTE_SDK)/mk/rte.vars.mk
 
 core-libs := librte_eal librte_mbuf librte_mempool librte_ring librte_ether
 
+DIRS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += dpaa
+DEPDIRS-dpaa = $(core-libs)
+
 DIRS-$(CONFIG_RTE_LIBRTE_FSLMC_BUS) += fslmc
 DEPDIRS-fslmc = $(core-libs)
 
diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
new file mode 100644
index 000..28694c0
--- /dev/null
+++ b/drivers/bus/dpaa/Makefile
@@ -0,0 +1,58 @@
+#   BSD LICENSE
+#
+#   Copyright 2016 NXP.
+#
+#   Redistribution and use in source and binary forms, with or without
+#   modification, are permitted provided that the following conditions
+#   are met:
+#
+# * Redistributions of source code must retain the above copyright
+#   notice, this list of conditions and the following disclaimer.
+# * Redistributions in binary form must reproduce the above copyright
+#   notice, this list of conditions and the following disclaimer in
+#   the documentation and/or other materials provided with the
+#   distribution.
+# * Neither the name of NXP nor the names of its
+#   contributors may be used to endorse or promote products derived
+#   from this software without specific prior written permission.
+#
+#   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+#   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+#   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+#   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+#   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+#   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+#   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+#   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+#   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+#   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+#   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+include $(RTE_SDK)/mk/rte.vars.mk
+RTE_BUS_DPAA=$(RTE_SDK)/drivers/bus/dpaa
+
+#
+# library name
+#
+LIB = librte_bus_dpaa.a
+
+CFLAGS := -I$(SRCDIR) $(CFLAGS)
+CFLAGS += -O3 $(WERROR_FLAGS)
+CFLAGS += -I$(RTE_BUS_DPAA)/
+CFLAGS += -I$(RTE_SDK)/lib/librte_eal/linuxapp/eal
+CFLAGS += -I$(RTE_SDK)/lib/librte_eal/common/include
+
+# versioning export map
+EXPORT_MAP := rte_bus_dpaa_version.map
+
+LIBABIVER := 1
+
+# all source are stored in SRCS-y
+#
+SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
+   dpaa_bus.c
+
+# Link Pthread
+LDLIBS += -lpthread
+
+include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/bus/dpaa/dpaa_bus.c b/drivers/bus/dpaa/dpaa_bus.c
new file mode 100644
index 000..cc343b3
--- /dev/null
+++ b/drivers/bus/dpaa/dpaa_bus.c
@@ -0,0 +1,207 @@
+/*-
+ *

[dpdk-dev] [PATCH v5 00/40] Introduce NXP DPAA Bus, Mempool and PMD

[Shreyansh Jain]

Change Log:


v5:
 - rebased over net-next/master (9d660ac)   
 - restructuring debugging macros. Removed a few and combined
   others. DPAA now reflects the dynamic logging with segragated
   DP logging
 - updated documentation for missing configuration option
 - fixed map file; shared build was broken earlier
 - other minor fixes from review comments

v4:
 - Some checkpatch fixes which were reported by checkpatch@dpdk
 - adding extra stats feature patch (patch 41)

v3:
 - Rebasing over 17.11-rc0 (85238f50)
 - Checkpatch fixes
   (There are still 2 errors which I think are false positives)
 - Implement rte_bus.find_device() interface
 - Various other minor updates/cleanups

v2:
 - Fixing various comments from Ferruh, but broadly:
  -) Logging is been changed to reflect rte_log_register
  -) Logs across Bus, Mempool and PMD updated
  -) fixed incorrect feature claimed in dpaa.ini
 - Removed 24/40/48 bit swapping macro from EAL.
   These are defined in dpaa/bus now (compat.h)
 - Added missing memory cleanup operation
 - Updated documentation with some missing information

Introduction


RFC was posted here -> [R3]
V4 was posted here  -> [R7]

This patch series adds NXP's QorIQ-Layerscape DPAA Architecture based
bus driver, mempool driver and PMD. This version of driver supports NXP
LS1043A/LS1023A, LS1046A/LS1026A family of network SoCs. [R1]

DPAA, or Datapath Acceleration Architecture [R2], is a set of hardware
components designed for high-speed network packet processing. This
architecture provides the infrastructure to support simplified sharing of
networking interfaces and accelerators by multiple CPU cores, and the
accelerators themselves.

This patchset introduces the following:
1. DPAA Bus (drivers/bus/dpaa)
 The core of DPAA bus is implemented using 3 main hardware blocks: QMan,
 or Queue Manager; BMan, or Buffer Manager and FMan, or Frame Manager.
 The patches introduce necessary layers to expose the DPAA hardware
 blocks for interfacing with RTE framework.

2. DPAA Mempool (drivers/mempool/dpaa)
 BMan, or Buffer Manager, block of DPAA features a hardware offloaded
 mempool. These patches add support for a driver to manage the BMan
 block. This driver allows for mempool creation, deletion, buffer
 acquire and release, as per the RTE APIs.

3. DPAA PMD (drivers/net/dpaa)
 The Poll Mode Driver for DPAA NIC Interfaces.

Patch Layout


01: Add DPAA SoC build configuration
02~16: Add DPAA Bus support and features, incrementally
17: Add Documentation
18~21: Add DPAA Mempool support
22~40: Add PMD and its various features, incrementally

References
==

[R1] 
http://www.nxp.com/products/microcontrollers-and-processors/arm-processors/qoriq-layerscape-arm-processors:QORIQ-ARM
[R2] http://www.nxp.com/assets/documents/data/en/white-papers/QORIQDPAAWP.pdf
[R3] RFC: http://dpdk.org/ml/archives/dev/2017-May/066675.html
[R4] v1: http://dpdk.org/ml/archives/dev/2017-June/068020.html
[R5] v2: http://dpdk.org/ml/archives/dev/2017-July/070113.html
[R6] v3: http://dpdk.org/ml/archives/dev/2017-August/073269.html
[R7] v4: http://dpdk.org/ml/archives/dev/2017-September/074936.html

Hemant Agrawal (3):
  bus/dpaa: add compatibility and helper macros
  net/dpaa: support firmware version get API
  net/dpaa: support extended statistics

Shreyansh Jain (37):
  config: add NXP DPAA SoC build configuration
  bus/dpaa: introduce NXP DPAA Bus driver skeleton
  bus/dpaa: add OF parser for device scanning
  bus/dpaa: introducing FMan configurations
  bus/dpaa: add FMan hardware operations
  bus/dpaa: enable DPAA IOCTL portal driver
  bus/dpaa: add layer for interrupt emulation using pthread
  bus/dpaa: add routines for managing a RB tree
  bus/dpaa: add QMAN interface driver
  bus/dpaa: add QMan driver core routines
  bus/dpaa: add BMAN driver core
  bus/dpaa: support FMAN frame queue lookup
  bus/dpaa: add BMan hardware interfaces
  bus/dpaa: add fman flow control threshold setting
  bus/dpaa: integrate DPAA Bus with hardware blocks
  doc: add NXP DPAA PMD documentation
  bus/dpaa: add DPAA mempool logging macros
  mempool/dpaa: support NXP DPAA Mempool
  config: enable compilation of DPAA Mempool driver
  bus/dpaa: add DPAA PMD logging macros
  net/dpaa: add NXP DPAA PMD driver skeleton
  config: enable NXP DPAA PMD compilation
  net/dpaa: support Tx and Rx queue setup
  net/dpaa: support MTU update
  net/dpaa: support jumbo frames
  net/dpaa: support link status update
  net/dpaa: support device info and speed capability
  net/dpaa: support promiscuous toggle
  net/dpaa: support multicast toggle
  net/dpaa: support MAC address update
  net/dpaa: support basic stats
  net/dpaa: support flow control
  net/dpaa: support hashed RSS
  net/dpaa: support packet type parsing
  net/dpaa: support checksum offload
  net/dpaa: support Scattered Rx
  net/dpaa: add packet dump for debugging

 MAINTAINERS   |9 +
 config/common_base  

[dpdk-dev] [PATCH v5 01/40] config: add NXP DPAA SoC build configuration

[Shreyansh Jain]

This patch adds skeleton build configuration for DPAA platform.

Signed-off-by: Shreyansh Jain 
---
 config/defconfig_arm64-dpaa-linuxapp-gcc | 47 
 mk/machine/dpaa/rte.vars.mk  | 61 
 2 files changed, 108 insertions(+)
 create mode 100644 config/defconfig_arm64-dpaa-linuxapp-gcc
 create mode 100644 mk/machine/dpaa/rte.vars.mk

diff --git a/config/defconfig_arm64-dpaa-linuxapp-gcc 
b/config/defconfig_arm64-dpaa-linuxapp-gcc
new file mode 100644
index 000..5bca012
--- /dev/null
+++ b/config/defconfig_arm64-dpaa-linuxapp-gcc
@@ -0,0 +1,47 @@
+#   BSD LICENSE
+#
+#   Copyright 2016 Freescale Semiconductor, Inc.
+#   Copyright 2017 NXP.
+#
+#   Redistribution and use in source and binary forms, with or without
+#   modification, are permitted provided that the following conditions
+#   are met:
+#
+# * Redistributions of source code must retain the above copyright
+#   notice, this list of conditions and the following disclaimer.
+# * Redistributions in binary form must reproduce the above copyright
+#   notice, this list of conditions and the following disclaimer in
+#   the documentation and/or other materials provided with the
+#   distribution.
+# * Neither the name of NXP nor the names of its
+#   contributors may be used to endorse or promote products derived
+#   from this software without specific prior written permission.
+#
+#   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+#   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+#   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+#   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+#   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+#   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+#   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+#   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+#   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+#   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+#   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+#
+
+#include "defconfig_arm64-armv8a-linuxapp-gcc"
+
+# NXP (Freescale) - Soc Architecture with FMAN, QMAN & BMAN support
+CONFIG_RTE_MACHINE="dpaa"
+CONFIG_RTE_ARCH_ARM_TUNE="cortex-a72"
+CONFIG_RTE_LIBRTE_VHOST_NUMA=n
+CONFIG_RTE_EAL_NUMA_AWARE_HUGEPAGES=n
+
+#
+# Compile Environment Abstraction Layer
+#
+CONFIG_RTE_MAX_LCORE=4
+CONFIG_RTE_MAX_NUMA_NODES=1
+CONFIG_RTE_CACHE_LINE_SIZE=64
+CONFIG_RTE_PKTMBUF_HEADROOM=128
diff --git a/mk/machine/dpaa/rte.vars.mk b/mk/machine/dpaa/rte.vars.mk
new file mode 100644
index 000..356a6af
--- /dev/null
+++ b/mk/machine/dpaa/rte.vars.mk
@@ -0,0 +1,61 @@
+#   BSD LICENSE
+#
+#   Copyright (c) 2016 Freescale Semiconductor, Inc. All rights reserved.
+#   Copyright 2017 NXP.
+#
+#   Redistribution and use in source and binary forms, with or without
+#   modification, are permitted provided that the following conditions
+#   are met:
+#
+# * Redistributions of source code must retain the above copyright
+#   notice, this list of conditions and the following disclaimer.
+# * Redistributions in binary form must reproduce the above copyright
+#   notice, this list of conditions and the following disclaimer in
+#   the documentation and/or other materials provided with the
+#   distribution.
+# * Neither the name of NXP nor the names of its
+#   contributors may be used to endorse or promote products derived
+#   from this software without specific prior written permission.
+#
+#   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+#   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+#   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+#   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+#   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+#   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+#   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+#   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+#   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+#   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+#   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+#
+# machine:
+#
+#   - can define ARCH variable (overridden by cmdline value)
+#   - can define CROSS variable (overridden by cmdline value)
+#   - define MACHINE_CFLAGS variable (overridden by cmdline value)
+#   - define MACHINE_LDFLAGS variable (overridden by cmdline value)
+#   - define MACHINE_ASFLAGS variable (overridden by cmdline value)
+#   - can define CPU_CFLAGS variable (overridden by cmdline value) that
+# overrides the 

[dpdk-dev] [PATCH v5 04/40] bus/dpaa: add OF parser for device scanning

[Shreyansh Jain]

This layer is used by Bus driver's scan function. Devices are parsed
using OF parser and added to DPAA device list.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile   |   7 +
 drivers/bus/dpaa/base/fman/of.c | 576 
 drivers/bus/dpaa/include/of.h   | 190 +
 3 files changed, 773 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/fman/of.c
 create mode 100644 drivers/bus/dpaa/include/of.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index 28694c0..30a3a5d 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -38,7 +38,11 @@ LIB = librte_bus_dpaa.a
 
 CFLAGS := -I$(SRCDIR) $(CFLAGS)
 CFLAGS += -O3 $(WERROR_FLAGS)
+CFLAGS += -Wno-pointer-arith
+CFLAGS += -Wno-cast-qual
+CFLAGS += -D _GNU_SOURCE
 CFLAGS += -I$(RTE_BUS_DPAA)/
+CFLAGS += -I$(RTE_BUS_DPAA)/include
 CFLAGS += -I$(RTE_SDK)/lib/librte_eal/linuxapp/eal
 CFLAGS += -I$(RTE_SDK)/lib/librte_eal/common/include
 
@@ -52,6 +56,9 @@ LIBABIVER := 1
 SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
dpaa_bus.c
 
+SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
+   base/fman/of.c \
+
 # Link Pthread
 LDLIBS += -lpthread
 
diff --git a/drivers/bus/dpaa/base/fman/of.c b/drivers/bus/dpaa/base/fman/of.c
new file mode 100644
index 000..b2d7c02
--- /dev/null
+++ b/drivers/bus/dpaa/base/fman/of.c
@@ -0,0 +1,576 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2010-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+#include 
+
+static int alive;
+static struct dt_dir root_dir;
+static const char *base_dir;
+static COMPAT_LIST_HEAD(linear);
+
+static int
+of_open_dir(const char *relative_path, struct dirent ***d)
+{
+   int ret;
+   char full_path[PATH_MAX];
+
+   snprintf(full_path, PATH_MAX, "%s/%s", base_dir, relative_path);
+   ret = scandir(full_path, d, 0, versionsort);
+   if (ret < 0)
+   DPAA_BUS_LOG(ERR, "Failed to open directory %s",
+full_path);
+   return ret;
+}
+
+static void
+of_close_dir(struct dirent **d, int num)
+{
+   while (num--)
+   free(d[num]);
+   free(d);
+}
+
+static int
+of_open_file(const char *relative_path)
+{
+   int ret;
+   char full_path[PATH_MAX];
+
+   snprintf(full_path, PATH_MAX, "%s/%s", base_dir, relative_path);
+   ret = open(full_path, O_RDONLY);
+   if (ret < 0)
+   DPAA_BUS_LOG(ERR, "Failed to open directory %s",
+full_path);
+   return ret;
+}
+
+static void
+process_file(struct dirent *dent, struct dt_dir *parent)
+{
+   int fd;
+   struct dt_file *f = malloc(sizeof(*f));
+
+   if (!f) {
+   DPAA_BUS_LOG(DEBUG, "Unable to allocate memory for file node");
+   return;
+   }
+   f->node.is_file = 1;
+   snprintf(f->node.node.name, NAME_MAX, "%s", dent->d_name);
+   snprintf(f->node.node.full_name, PATH_MAX, "%s/%s",
+parent->node.node.full_name, 

[dpdk-dev] [PATCH v5 03/40] bus/dpaa: add compatibility and helper macros

[Shreyansh Jain]

From: Hemant Agrawal 

Linked list, bit operations and compatibility macros.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Hemant Agrawal 
---
 drivers/bus/dpaa/include/compat.h| 385 +++
 drivers/bus/dpaa/include/dpaa_bits.h |  65 ++
 drivers/bus/dpaa/include/dpaa_list.h | 101 +
 3 files changed, 551 insertions(+)
 create mode 100644 drivers/bus/dpaa/include/compat.h
 create mode 100644 drivers/bus/dpaa/include/dpaa_bits.h
 create mode 100644 drivers/bus/dpaa/include/dpaa_list.h

diff --git a/drivers/bus/dpaa/include/compat.h 
b/drivers/bus/dpaa/include/compat.h
new file mode 100644
index 000..42733ae
--- /dev/null
+++ b/drivers/bus/dpaa/include/compat.h
@@ -0,0 +1,385 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2011 Freescale Semiconductor, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef __COMPAT_H
+#define __COMPAT_H
+
+#include 
+
+#ifndef _GNU_SOURCE
+#define _GNU_SOURCE
+#endif
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+
+/* The following definitions are primarily to allow the single-source driver
+ * interfaces to be included by arbitrary program code. Ie. for interfaces that
+ * are also available in kernel-space, these definitions provide compatibility
+ * with certain attributes and types used in those interfaces.
+ */
+
+/* Required compiler attributes */
+#define __maybe_unused __rte_unused
+#define __always_unused__rte_unused
+#define __packed   __rte_packed
+#define noinline   __attribute__((noinline))
+
+#define L1_CACHE_BYTES 64
+#define cacheline_aligned __attribute__((aligned(L1_CACHE_BYTES)))
+#define __stringify_1(x) #x
+#define __stringify(x) __stringify_1(x)
+
+#ifdef ARRAY_SIZE
+#undef ARRAY_SIZE
+#endif
+#define ARRAY_SIZE(a) (sizeof(a) / sizeof((a)[0]))
+
+/* Debugging */
+#define prflush(fmt, args...) \
+   do { \
+   printf(fmt, ##args); \
+   fflush(stdout); \
+   } while (0)
+
+#define pr_crit(fmt, args...)   prflush("CRIT:" fmt, ##args)
+#define pr_err(fmt, args...)prflush("ERR:" fmt, ##args)
+#define pr_warn(fmt, args...)   prflush("WARN:" fmt, ##args)
+#define pr_info(fmt, args...)   prflush(fmt, ##args)
+
+#ifdef RTE_LIBRTE_DPAA_DEBUG_BUS
+#ifdef pr_debug
+#undef pr_debug
+#endif
+#define pr_debug(fmt, args...) printf(fmt, ##args)
+#else
+#define pr_debug(fmt, args...) {}
+#endif
+
+#define DPAA_BUG_ON(x) RTE_ASSERT(x)
+
+/* Required types */
+typedef uint8_tu8;
+typedef uint16_t   u16;
+typedef uint32_t   u32;
+typedef uint64_t   u64;
+typedef uint64_t   dma_addr_t;
+typedef cpu_set_t  cpumask_t;
+typedef uint32_t   phandle;
+typedef uint32_t   gfp_t;
+typedef uint32_t   irqreturn_t;
+
+#define IRQ_HANDLED0
+#define request_irqqbman_request_irq
+#define free_irq   qb

[dpdk-dev] [PATCH v5 06/40] bus/dpaa: add FMan hardware operations

[Shreyansh Jain]

Signed-off-by: Geoff Thorpe 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile |   1 +
 drivers/bus/dpaa/base/fman/fman_hw.c  | 562 ++
 drivers/bus/dpaa/include/fsl_fman.h   | 174 +
 drivers/bus/dpaa/include/fsl_fman_crc64.h | 263 ++
 4 files changed, 1000 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/fman/fman_hw.c
 create mode 100644 drivers/bus/dpaa/include/fsl_fman.h
 create mode 100644 drivers/bus/dpaa/include/fsl_fman_crc64.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index f6e504d..fe65276 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -58,6 +58,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
 
 SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/fman.c \
+   base/fman/fman_hw.c \
base/fman/of.c \
base/fman/netcfg_layer.c
 
diff --git a/drivers/bus/dpaa/base/fman/fman_hw.c 
b/drivers/bus/dpaa/base/fman/fman_hw.c
new file mode 100644
index 000..a7ca661
--- /dev/null
+++ b/drivers/bus/dpaa/base/fman/fman_hw.c
@@ -0,0 +1,562 @@
+/*-
+ *   BSD LICENSE
+ *
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+#include 
+#include 
+#include 
+/* This header declares things about Fman hardware itself (the format of status
+ * words and an inline implementation of CRC64). We include it only in order to
+ * instantiate the one global variable it depends on.
+ */
+#include 
+#include 
+
+/* Instantiate the global variable that the inline CRC64 implementation (in
+ * ) depends on.
+ */
+DECLARE_FMAN_CRC64_TABLE();
+
+#define ETH_ADDR_TO_UINT64(eth_addr)  \
+   (uint64_t)(((uint64_t)(eth_addr)[0] << 40) |   \
+   ((uint64_t)(eth_addr)[1] << 32) |   \
+   ((uint64_t)(eth_addr)[2] << 24) |   \
+   ((uint64_t)(eth_addr)[3] << 16) |   \
+   ((uint64_t)(eth_addr)[4] << 8) |\
+   ((uint64_t)(eth_addr)[5]))
+
+void
+fman_if_set_mcast_filter_table(struct fman_if *p)
+{
+   struct __fman_if *__if = container_of(p, struct __fman_if, __if);
+   void *hashtable_ctrl;
+   uint32_t i;
+
+   hashtable_ctrl = &((struct memac_regs *)__if->ccsr_map)->hashtable_ctrl;
+   for (i = 0; i < 64; i++)
+   out_be32(hashtable_ctrl, i|HASH_CTRL_MCAST_EN);
+}
+
+void
+fman_if_reset_mcast_filter_table(struct fman_if *p)
+{
+   struct __fman_if *__if = container_of(p, struct __fman_if, __if);
+   void *hashtable_ctrl;
+   uint32_t i;
+
+   hashtable_ctrl = &((struct memac_regs *)__if->ccsr_map)->hashtable_ctrl;
+   for (i = 0; i < 64; i++)
+   out_be32(hashtable_ctrl, i & ~HASH_CTRL_MCAST_EN);
+}
+
+static
+uint32_t get_mac_hash_code(uint64_t eth_addr)
+{
+   uint64_tmask1, mask2;
+   uint32_txorVal = 0;
+   uint8_t i, j;
+
+   for (i = 0; i < 6; i++) {
+   mask1 = eth_addr & (uint64_t)0x01;
+   eth_addr >>= 1;
+
+   for (j = 0; j < 7; j++) {
+   mask2 = eth_addr & (uint64_t)0x01;
+   mask1 ^= mask2;
+   eth_addr >>= 1;
+   }
+
+   xorVal |= (mask1 << (5 - i));
+   }
+
+   return xorVal;
+}
+
+int
+fman_if_add_hash_mac_addr(struct fman_if *p, uint8_t *eth)
+{
+   uint64_t eth_addr;
+   void *hashtable_ctrl;
+   uint32_t hash;
+
+   struct __fman_if *__if = container_o

[dpdk-dev] [PATCH v5 08/40] bus/dpaa: add layer for interrupt emulation using pthread

[Shreyansh Jain]

An interrupt manager is implemented by emulating over pthreads.
Handlers are registered by QBMAN layer for being notified about
any interrupt request from DPAA blocks in userspace.

Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile  |   3 +-
 drivers/bus/dpaa/base/qbman/dpaa_sys.c | 136 +
 drivers/bus/dpaa/base/qbman/dpaa_sys.h |  61 +++
 3 files changed, 199 insertions(+), 1 deletion(-)
 create mode 100644 drivers/bus/dpaa/base/qbman/dpaa_sys.c
 create mode 100644 drivers/bus/dpaa/base/qbman/dpaa_sys.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index f06521c..5b76a4b 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -62,7 +62,8 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/fman_hw.c \
base/fman/of.c \
base/fman/netcfg_layer.c \
-   base/qbman/process.c
+   base/qbman/process.c \
+   base/qbman/dpaa_sys.c
 
 # Link Pthread
 LDLIBS += -lpthread
diff --git a/drivers/bus/dpaa/base/qbman/dpaa_sys.c 
b/drivers/bus/dpaa/base/qbman/dpaa_sys.c
new file mode 100644
index 000..0017da5
--- /dev/null
+++ b/drivers/bus/dpaa/base/qbman/dpaa_sys.c
@@ -0,0 +1,136 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+#include "dpaa_sys.h"
+
+struct process_interrupt {
+   int irq;
+   irqreturn_t (*isr)(int irq, void *arg);
+   unsigned long flags;
+   const char *name;
+   void *arg;
+   struct list_head node;
+};
+
+static COMPAT_LIST_HEAD(process_irq_list);
+static pthread_mutex_t process_irq_lock = PTHREAD_MUTEX_INITIALIZER;
+
+static void process_interrupt_install(struct process_interrupt *irq)
+{
+   int ret;
+   /* Add the irq to the end of the list */
+   ret = pthread_mutex_lock(&process_irq_lock);
+   assert(!ret);
+   list_add_tail(&irq->node, &process_irq_list);
+   ret = pthread_mutex_unlock(&process_irq_lock);
+   assert(!ret);
+}
+
+static void process_interrupt_remove(struct process_interrupt *irq)
+{
+   int ret;
+
+   ret = pthread_mutex_lock(&process_irq_lock);
+   assert(!ret);
+   list_del(&irq->node);
+   ret = pthread_mutex_unlock(&process_irq_lock);
+   assert(!ret);
+}
+
+static struct process_interrupt *process_interrupt_find(int irq_num)
+{
+   int ret;
+   struct process_interrupt *i = NULL;
+
+   ret = pthread_mutex_lock(&process_irq_lock);
+   assert(!ret);
+   list_for_each_entry(i, &process_irq_list, node) {
+   if (i->irq == irq_num)
+   goto done;
+   }
+done:
+   ret = pthread_mutex_unlock(&process_irq_lock);
+   assert(!ret);
+   return i;
+}
+
+/* This is the interface from the platform-agnostic driver code to (de)register
+ * interrupt handlers. We simply create/destroy corresponding structs.
+ */
+int qbman_request_irq(int irq, irqreturn_t (*isr)(in

[dpdk-dev] [PATCH v5 05/40] bus/dpaa: introducing FMan configurations

[Shreyansh Jain]

FMan or Frame Manager, inspects traffic, splits it into queueson ingress.
It is also responsible for directing traffic on queues on egress.

This patch introduces FMan configurational interfaces. This layer is
used by Bus driver for configuring the hardware block.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile |   2 +
 drivers/bus/dpaa/base/fman/fman.c | 611 ++
 drivers/bus/dpaa/base/fman/netcfg_layer.c | 214 +++
 drivers/bus/dpaa/include/fman.h   | 458 ++
 drivers/bus/dpaa/include/netcfg.h |  96 +
 5 files changed, 1381 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/fman/fman.c
 create mode 100644 drivers/bus/dpaa/base/fman/netcfg_layer.c
 create mode 100644 drivers/bus/dpaa/include/fman.h
 create mode 100644 drivers/bus/dpaa/include/netcfg.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index 30a3a5d..f6e504d 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -57,7 +57,9 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
dpaa_bus.c
 
 SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
+   base/fman/fman.c \
base/fman/of.c \
+   base/fman/netcfg_layer.c
 
 # Link Pthread
 LDLIBS += -lpthread
diff --git a/drivers/bus/dpaa/base/fman/fman.c 
b/drivers/bus/dpaa/base/fman/fman.c
new file mode 100644
index 000..2c6029e
--- /dev/null
+++ b/drivers/bus/dpaa/base/fman/fman.c
@@ -0,0 +1,611 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2010-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+#include 
+#include 
+
+#include 
+
+/* This header declares the driver interface we implement */
+#include 
+#include 
+#include 
+
+#define QMI_PORT_REGS_OFFSET   0x400
+
+/* CCSR map address to access ccsr based register */
+void *fman_ccsr_map;
+/* fman version info */
+u16 fman_ip_rev;
+static int get_once;
+u32 fman_dealloc_bufs_mask_hi;
+u32 fman_dealloc_bufs_mask_lo;
+
+int fman_ccsr_map_fd = -1;
+static COMPAT_LIST_HEAD(__ifs);
+
+/* This is the (const) global variable that callers have read-only access to.
+ * Internally, we have read-write access directly to __ifs.
+ */
+const struct list_head *fman_if_list = &__ifs;
+
+static void
+if_destructor(struct __fman_if *__if)
+{
+   struct fman_if_bpool *bp, *tmpbp;
+
+   if (__if->__if.mac_type == fman_offline)
+   goto cleanup;
+
+   list_for_each_entry_safe(bp, tmpbp, &__if->__if.bpool_list, node) {
+   list_del(&bp->node);
+   rte_free(bp);
+   }
+cleanup:
+   rte_free(__if);
+}
+
+static int
+fman_get_ip_rev(const struct device_node *fman_node)
+{
+   const uint32_t *fman_addr;
+   uint64_t phys_addr;
+   uint64_t regs_size;
+   uint32_t ip_rev_1;
+   int _errno;
+
+   fman_addr = of_get_address(fman_node, 0, ®s_size, NULL);
+   if (!fman_addr) {
+   pr_err("of_get_address cannot return fman address\n")

[dpdk-dev] [PATCH v5 07/40] bus/dpaa: enable DPAA IOCTL portal driver

[Shreyansh Jain]

Userspace applications interact with DPAA blocks using this IOCTL driver.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile |   4 +-
 drivers/bus/dpaa/base/qbman/process.c | 331 ++
 drivers/bus/dpaa/include/fsl_usd.h|  88 +
 drivers/bus/dpaa/include/process.h| 107 +++
 4 files changed, 529 insertions(+), 1 deletion(-)
 create mode 100644 drivers/bus/dpaa/base/qbman/process.c
 create mode 100644 drivers/bus/dpaa/include/fsl_usd.h
 create mode 100644 drivers/bus/dpaa/include/process.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index fe65276..f06521c 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -43,6 +43,7 @@ CFLAGS += -Wno-cast-qual
 CFLAGS += -D _GNU_SOURCE
 CFLAGS += -I$(RTE_BUS_DPAA)/
 CFLAGS += -I$(RTE_BUS_DPAA)/include
+CFLAGS += -I$(RTE_BUS_DPAA)/base/qbman
 CFLAGS += -I$(RTE_SDK)/lib/librte_eal/linuxapp/eal
 CFLAGS += -I$(RTE_SDK)/lib/librte_eal/common/include
 
@@ -60,7 +61,8 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/fman.c \
base/fman/fman_hw.c \
base/fman/of.c \
-   base/fman/netcfg_layer.c
+   base/fman/netcfg_layer.c \
+   base/qbman/process.c
 
 # Link Pthread
 LDLIBS += -lpthread
diff --git a/drivers/bus/dpaa/base/qbman/process.c 
b/drivers/bus/dpaa/base/qbman/process.c
new file mode 100644
index 000..b8ec539
--- /dev/null
+++ b/drivers/bus/dpaa/base/qbman/process.c
@@ -0,0 +1,331 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2011-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+#include 
+#include 
+#include 
+#include 
+
+#include "process.h"
+
+#include 
+
+/* As higher-level drivers will be built on top of this (dma_mem, qbman, ...),
+ * it's preferable that the process driver itself not provide any exported API.
+ * As such, combined with the fact that none of these operations are
+ * performance critical, it is justified to use lazy initialisation, so that's
+ * what the lock is for.
+ */
+static int fd = -1;
+static pthread_mutex_t fd_init_lock = PTHREAD_MUTEX_INITIALIZER;
+
+static int check_fd(void)
+{
+   int ret;
+
+   if (fd >= 0)
+   return 0;
+   ret = pthread_mutex_lock(&fd_init_lock);
+   assert(!ret);
+   /* check again with the lock held */
+   if (fd < 0)
+   fd = open(PROCESS_PATH, O_RDWR);
+   ret = pthread_mutex_unlock(&fd_init_lock);
+   assert(!ret);
+   return (fd >= 0) ? 0 : -ENODEV;
+}
+
+#define DPAA_IOCTL_MAGIC 'u'
+struct dpaa_ioctl_id_alloc {
+   uint32_t base; /* Return value, the start of the allocated range */
+   enum dpaa_id_type id_type; /* what kind of resource(s) to allocate */
+   uint32_t num; /* how many IDs to allocate (and return value) */
+   uint32_t align; /* must be a power of 2, 0 is treated like 1 */
+   int partial; /* whether to allow less than 'num' */
+};
+
+struct dpaa_ioctl_id_release {
+   /* Inp

[dpdk-dev] [PATCH v5 09/40] bus/dpaa: add routines for managing a RB tree

[Shreyansh Jain]

QMAN frames are managed over a RB tree data structure.
This patch introduces necessary routines for implementing a RB tree.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/include/dpaa_rbtree.h | 143 +
 1 file changed, 143 insertions(+)
 create mode 100644 drivers/bus/dpaa/include/dpaa_rbtree.h

diff --git a/drivers/bus/dpaa/include/dpaa_rbtree.h 
b/drivers/bus/dpaa/include/dpaa_rbtree.h
new file mode 100644
index 000..f8c9b59
--- /dev/null
+++ b/drivers/bus/dpaa/include/dpaa_rbtree.h
@@ -0,0 +1,143 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright 2017 NXP.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ *   notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ *   notice, this list of conditions and the following disclaimer in
+ *   the documentation and/or other materials provided with the
+ *   distribution.
+ * * Neither the name of NXP nor the names of its
+ *   contributors may be used to endorse or promote products derived
+ *   from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef __DPAA_RBTREE_H
+#define __DPAA_RBTREE_H
+
+#include 
+//
+/* RB-trees */
+//
+
+/* Linux has a good RB-tree implementation, that we can't use (GPL). It also 
has
+ * a flat/hooked-in interface that virtually requires license-contamination in
+ * order to write a caller-compatible implementation. Instead, I've created an
+ * RB-tree encapsulation on top of linux's primitives (it does some of the work
+ * the client logic would normally do), and this gives us something we can
+ * reimplement on LWE. Unfortunately there's no good+free RB-tree
+ * implementations out there that are license-compatible and "flat" (ie. no
+ * dynamic allocation). I did find a malloc-based one that I could convert, but
+ * that will be a task for later on. For now, LWE's RB-tree is implemented 
using
+ * an ordered linked-list.
+ *
+ * Note, the only linux-esque type is "struct rb_node", because it's used
+ * statically in the exported header, so it can't be opaque. Our version 
doesn't
+ * include a "rb_parent_color" field because we're doing linked-list instead of
+ * a true rb-tree.
+ */
+
+struct rb_node {
+   struct rb_node *prev, *next;
+};
+
+struct dpa_rbtree {
+   struct rb_node *head, *tail;
+};
+
+#define DPAA_RBTREE { NULL, NULL }
+static inline void dpa_rbtree_init(struct dpa_rbtree *tree)
+{
+   tree->head = tree->tail = NULL;
+}
+
+#define QMAN_NODE2OBJ(ptr, type, node_field) \
+   (type *)((char *)ptr - offsetof(type, node_field))
+
+#define IMPLEMENT_DPAA_RBTREE(name, type, node_field, val_field) \
+static inline int name##_push(struct dpa_rbtree *tree, type *obj) \
+{ \
+   struct rb_node *node = tree->head; \
+   if (!node) { \
+   tree->head = tree->tail = &obj->node_field; \
+   obj->node_field.prev = obj->node_field.next = NULL; \
+   return 0; \
+   } \
+   while (node) { \
+   type *item = QMAN_NODE2OBJ(node, type, node_field); \
+   if (obj->val_field == item->val_field) \
+   return -EBUSY; \
+   if (obj->val_field < item->val_field) { \
+   if (tree->head == node) \
+   tree->head = &obj->node_field; \
+   else \
+   node->prev->next = &obj->node_field; \
+   obj->node_field.prev = node->prev; \
+   obj->node_field.next = node; \
+   node->prev = &obj->node_field; \
+   return 0; \
+   } \
+   node = node->next; \
+   } \
+   obj->node_field.prev = tree->tail; \
+   obj->node_field.next = NULL; \
+   tree->tail->next = &obj->nod

[dpdk-dev] [PATCH v5 10/40] bus/dpaa: add QMAN interface driver

[Shreyansh Jain]

The Queue Manager (QMan) is a hardware queue management block that
allows software and accelerators on the datapath to enqueue and dequeue
frames in order to communicate.

This part of QBMAN DPAA Block.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile |1 +
 drivers/bus/dpaa/base/qbman/qman_driver.c |  271 +++
 drivers/bus/dpaa/base/qbman/qman_priv.h   |  303 +++
 drivers/bus/dpaa/include/fsl_qman.h   | 1254 +
 drivers/bus/dpaa/include/fsl_usd.h|   13 +
 5 files changed, 1842 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/qbman/qman_driver.c
 create mode 100644 drivers/bus/dpaa/base/qbman/qman_priv.h
 create mode 100644 drivers/bus/dpaa/include/fsl_qman.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index 5b76a4b..c9c15f8 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -63,6 +63,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/of.c \
base/fman/netcfg_layer.c \
base/qbman/process.c \
+   base/qbman/qman_driver.c \
base/qbman/dpaa_sys.c
 
 # Link Pthread
diff --git a/drivers/bus/dpaa/base/qbman/qman_driver.c 
b/drivers/bus/dpaa/base/qbman/qman_driver.c
new file mode 100644
index 000..80dde20
--- /dev/null
+++ b/drivers/bus/dpaa/base/qbman/qman_driver.c
@@ -0,0 +1,271 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2008-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+#include 
+#include "qman_priv.h"
+#include 
+#include 
+
+/* Global variable containing revision id (even on non-control plane systems
+ * where CCSR isn't available).
+ */
+u16 qman_ip_rev;
+u16 qm_channel_pool1 = QMAN_CHANNEL_POOL1;
+u16 qm_channel_caam = QMAN_CHANNEL_CAAM;
+u16 qm_channel_pme = QMAN_CHANNEL_PME;
+
+/* Ccsr map address to access ccsrbased register */
+void *qman_ccsr_map;
+/* The qman clock frequency */
+u32 qman_clk;
+
+static __thread int fd = -1;
+static __thread struct qm_portal_config pcfg;
+static __thread struct dpaa_ioctl_portal_map map = {
+   .type = dpaa_portal_qman
+};
+
+static int fsl_qman_portal_init(uint32_t index, int is_shared)
+{
+   cpu_set_t cpuset;
+   int loop, ret;
+   struct dpaa_ioctl_irq_map irq_map;
+
+   /* Verify the thread's cpu-affinity */
+   ret = pthread_getaffinity_np(pthread_self(), sizeof(cpu_set_t),
+&cpuset);
+   if (ret) {
+   error(0, ret, "pthread_getaffinity_np()");
+   return ret;
+   }
+   pcfg.cpu = -1;
+   for (loop = 0; loop < CPU_SETSIZE; loop++)
+   if (CPU_ISSET(loop, &cpuset)) {
+   if (pcfg.cpu != -1) {
+   pr_err("Thread is not affine to 1 cpu\n");
+   return -EINVAL;
+   }
+   pcfg.cpu = loop;
+   }
+   if (pcfg.cpu == -1) {
+ 

[dpdk-dev] [PATCH v5 11/40] bus/dpaa: add QMan driver core routines

[Shreyansh Jain]

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 config/defconfig_arm64-dpaa-linuxapp-gcc  |1 +
 drivers/bus/dpaa/Makefile |2 +
 drivers/bus/dpaa/base/qbman/dpaa_alloc.c  |   88 ++
 drivers/bus/dpaa/base/qbman/qman.c| 2402 +
 drivers/bus/dpaa/base/qbman/qman.h|  888 +++
 drivers/bus/dpaa/base/qbman/qman_driver.c |   12 +
 drivers/bus/dpaa/include/fsl_qman.h   |  755 +
 drivers/bus/dpaa/include/fsl_usd.h|1 +
 8 files changed, 4149 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/qbman/dpaa_alloc.c
 create mode 100644 drivers/bus/dpaa/base/qbman/qman.c
 create mode 100644 drivers/bus/dpaa/base/qbman/qman.h

diff --git a/config/defconfig_arm64-dpaa-linuxapp-gcc 
b/config/defconfig_arm64-dpaa-linuxapp-gcc
index 8316fc9..4d6b046 100644
--- a/config/defconfig_arm64-dpaa-linuxapp-gcc
+++ b/config/defconfig_arm64-dpaa-linuxapp-gcc
@@ -49,3 +49,4 @@ CONFIG_RTE_PKTMBUF_HEADROOM=128
 # NXP DPAA Bus
 CONFIG_RTE_LIBRTE_DPAA_BUS=y
 CONFIG_RTE_LIBRTE_DPAA_DEBUG_DRIVER=n
+CONFIG_RTE_LIBRTE_DPAA_HWDEBUG=n
diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index c9c15f8..5957c15 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -63,7 +63,9 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/of.c \
base/fman/netcfg_layer.c \
base/qbman/process.c \
+   base/qbman/qman.c \
base/qbman/qman_driver.c \
+   base/qbman/dpaa_alloc.c \
base/qbman/dpaa_sys.c
 
 # Link Pthread
diff --git a/drivers/bus/dpaa/base/qbman/dpaa_alloc.c 
b/drivers/bus/dpaa/base/qbman/dpaa_alloc.c
new file mode 100644
index 000..690576a
--- /dev/null
+++ b/drivers/bus/dpaa/base/qbman/dpaa_alloc.c
@@ -0,0 +1,88 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2009-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include "dpaa_sys.h"
+#include 
+#include 
+
+int qman_alloc_fqid_range(u32 *result, u32 count, u32 align, int partial)
+{
+   return process_alloc(dpaa_id_fqid, result, count, align, partial);
+}
+
+void qman_release_fqid_range(u32 fqid, u32 count)
+{
+   process_release(dpaa_id_fqid, fqid, count);
+}
+
+int qman_reserve_fqid_range(u32 fqid, unsigned int count)
+{
+   return process_reserve(dpaa_id_fqid, fqid, count);
+}
+
+int qman_alloc_pool_range(u32 *result, u32 count, u32 align, int partial)
+{
+   return process_alloc(dpaa_id_qpool, result, count, align, partial);
+}
+
+void qman_release_pool_range(u32 pool, u32 count)
+{
+   process_release(dpaa_id_qpool, pool, count);
+}
+
+int qman_reserve_pool_range(u32 pool, u32 count)
+{
+   return process_reserve(dpaa_id_qpool, pool, count);
+}
+
+int qman_alloc_cgrid_range(u32 *result, u32 count, u32 align, int partial)
+{
+   return process_alloc(dpaa_id_cgrid, result, count, align, partial);
+}
+
+void qman_release_cgrid_range(u32 cgrid, u32 count)
+{
+   process_release(dpaa_id_cgrid

[dpdk-dev] [PATCH v5 13/40] bus/dpaa: support FMAN frame queue lookup

[Shreyansh Jain]

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/base/qbman/qman.c| 99 ++-
 drivers/bus/dpaa/base/qbman/qman_driver.c |  7 ++-
 drivers/bus/dpaa/base/qbman/qman_priv.h   |  7 +++
 drivers/bus/dpaa/include/fsl_qman.h   | 12 
 4 files changed, 122 insertions(+), 3 deletions(-)

diff --git a/drivers/bus/dpaa/base/qbman/qman.c 
b/drivers/bus/dpaa/base/qbman/qman.c
index 9b1630b..8c8d270 100644
--- a/drivers/bus/dpaa/base/qbman/qman.c
+++ b/drivers/bus/dpaa/base/qbman/qman.c
@@ -176,6 +176,65 @@ static inline struct qman_fq *table_find_fq(struct 
qman_portal *p, u32 fqid)
return fqtree_find(&p->retire_table, fqid);
 }
 
+#ifdef CONFIG_FSL_QMAN_FQ_LOOKUP
+static void **qman_fq_lookup_table;
+static size_t qman_fq_lookup_table_size;
+
+int qman_setup_fq_lookup_table(size_t num_entries)
+{
+   num_entries++;
+   /* Allocate 1 more entry since the first entry is not used */
+   qman_fq_lookup_table = vmalloc((num_entries * sizeof(void *)));
+   if (!qman_fq_lookup_table) {
+   pr_err("QMan: Could not allocate fq lookup table\n");
+   return -ENOMEM;
+   }
+   memset(qman_fq_lookup_table, 0, num_entries * sizeof(void *));
+   qman_fq_lookup_table_size = num_entries;
+   pr_debug("QMan: Allocated lookup table at %p, entry count %lu\n",
+   qman_fq_lookup_table,
+   (unsigned long)qman_fq_lookup_table_size);
+   return 0;
+}
+
+/* global structure that maintains fq object mapping */
+static DEFINE_SPINLOCK(fq_hash_table_lock);
+
+static int find_empty_fq_table_entry(u32 *entry, struct qman_fq *fq)
+{
+   u32 i;
+
+   spin_lock(&fq_hash_table_lock);
+   /* Can't use index zero because this has special meaning
+* in context_b field.
+*/
+   for (i = 1; i < qman_fq_lookup_table_size; i++) {
+   if (qman_fq_lookup_table[i] == NULL) {
+   *entry = i;
+   qman_fq_lookup_table[i] = fq;
+   spin_unlock(&fq_hash_table_lock);
+   return 0;
+   }
+   }
+   spin_unlock(&fq_hash_table_lock);
+   return -ENOMEM;
+}
+
+static void clear_fq_table_entry(u32 entry)
+{
+   spin_lock(&fq_hash_table_lock);
+   DPAA_BUG_ON(entry >= qman_fq_lookup_table_size);
+   qman_fq_lookup_table[entry] = NULL;
+   spin_unlock(&fq_hash_table_lock);
+}
+
+static inline struct qman_fq *get_fq_table_entry(u32 entry)
+{
+   DPAA_BUG_ON(entry >= qman_fq_lookup_table_size);
+   return qman_fq_lookup_table[entry];
+}
+#endif
+
 static inline void cpu_to_hw_fqd(struct qm_fqd *fqd)
 {
/* Byteswap the FQD to HW format */
@@ -766,8 +825,13 @@ static u32 __poll_portal_slow(struct qman_portal *p, u32 
is)
break;
case QM_MR_VERB_FQPN:
/* Parked */
+#ifdef CONFIG_FSL_QMAN_FQ_LOOKUP
+   fq = get_fq_table_entry(
+   be32_to_cpu(msg->fq.contextB));
+#else
fq = (void *)(uintptr_t)
be32_to_cpu(msg->fq.contextB);
+#endif
fq_state_change(p, fq, msg, verb);
if (fq->cb.fqs)
fq->cb.fqs(p, fq, &swapped_msg);
@@ -792,7 +856,11 @@ static u32 __poll_portal_slow(struct qman_portal *p, u32 
is)
}
} else {
/* Its a software ERN */
+#ifdef CONFIG_FSL_QMAN_FQ_LOOKUP
+   fq = get_fq_table_entry(be32_to_cpu(msg->ern.tag));
+#else
fq = (void *)(uintptr_t)be32_to_cpu(msg->ern.tag);
+#endif
fq->cb.ern(p, fq, &swapped_msg);
}
num++;
@@ -907,7 +975,11 @@ static inline unsigned int __poll_portal_fast(struct 
qman_portal *p,
clear_vdqcr(p, fq);
} else {
/* SDQCR: context_b points to the FQ */
+#ifdef CONFIG_FSL_QMAN_FQ_LOOKUP
+   fq = get_fq_table_entry(dq->contextB);
+#else
fq = (void *)(uintptr_t)dq->contextB;
+#endif
/* Now let the callback do its stuff */
res = fq->cb.dqrr(p, fq, dq);
/*
@@ -1119,7 +1191,12 @@ int qman_create_fq(u32 fqid, u32 flags, struct qman_fq 
*fq)
fq->flags = flags;
fq->state = qman_fq_state_oos;
fq->cgr_groupid = 0;
-
+#ifdef CONFIG_FSL_QMAN_FQ_LOOKUP
+   if (unlikely(find_empty_fq_table_entry(&fq->key, fq))) {
+   pr_info("Find empty table entry failed\n");
+   return -ENOMEM;
+   }
+#endif
if (!(flags & QMAN_FQ_FLAG_AS_IS) || (f

[dpdk-dev] [PATCH v5 12/40] bus/dpaa: add BMAN driver core

[Shreyansh Jain]

The Buffer Manager (BMan) is a hardware buffer pool management block that
allows software and accelerators on the datapath to acquire and release
buffers in order to build frames.

This patch adds the core routines.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile |   1 +
 drivers/bus/dpaa/base/qbman/bman_driver.c | 311 +
 drivers/bus/dpaa/base/qbman/bman_priv.h   | 125 ++
 drivers/bus/dpaa/include/fsl_bman.h   | 375 ++
 drivers/bus/dpaa/include/fsl_usd.h|   5 +
 5 files changed, 817 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/qbman/bman_driver.c
 create mode 100644 drivers/bus/dpaa/base/qbman/bman_priv.h
 create mode 100644 drivers/bus/dpaa/include/fsl_bman.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index 5957c15..e1415e4 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -63,6 +63,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/of.c \
base/fman/netcfg_layer.c \
base/qbman/process.c \
+   base/qbman/bman_driver.c \
base/qbman/qman.c \
base/qbman/qman_driver.c \
base/qbman/dpaa_alloc.c \
diff --git a/drivers/bus/dpaa/base/qbman/bman_driver.c 
b/drivers/bus/dpaa/base/qbman/bman_driver.c
new file mode 100644
index 000..fb3c50e
--- /dev/null
+++ b/drivers/bus/dpaa/base/qbman/bman_driver.c
@@ -0,0 +1,311 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2008-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+
+#include 
+#include 
+#include "bman_priv.h"
+#include 
+
+/*
+ * Global variables of the max portal/pool number this bman version supported
+ */
+u16 bman_ip_rev;
+u16 bman_pool_max;
+void *bman_ccsr_map;
+
+/*/
+/* Portal driver */
+/*/
+
+static __thread int fd = -1;
+static __thread struct bm_portal_config pcfg;
+static __thread struct dpaa_ioctl_portal_map map = {
+   .type = dpaa_portal_bman
+};
+
+static int fsl_bman_portal_init(uint32_t idx, int is_shared)
+{
+   cpu_set_t cpuset;
+   int loop, ret;
+   struct dpaa_ioctl_irq_map irq_map;
+
+   /* Verify the thread's cpu-affinity */
+   ret = pthread_getaffinity_np(pthread_self(), sizeof(cpu_set_t),
+&cpuset);
+   if (ret) {
+   error(0, ret, "pthread_getaffinity_np()");
+   return ret;
+   }
+   pcfg.cpu = -1;
+   for (loop = 0; loop < CPU_SETSIZE; loop++)
+   if (CPU_ISSET(loop, &cpuset)) {
+   if (pcfg.cpu != -1) {
+   pr_err("Thread is not affine to 1 cpu");
+   return -EINVAL;
+   }
+   pcfg.cpu = loop;
+   }
+   if (pcfg.cpu == -1) {
+   pr_err("Bug in getaffinity handling!");
+   return -EINVAL;
+   }
+  

[dpdk-dev] [PATCH v5 14/40] bus/dpaa: add BMan hardware interfaces

[Shreyansh Jain]

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile |   1 +
 drivers/bus/dpaa/base/qbman/bman.c| 394 +
 drivers/bus/dpaa/base/qbman/bman.h| 550 ++
 drivers/bus/dpaa/base/qbman/bman_driver.c |  12 +
 drivers/bus/dpaa/base/qbman/dpaa_alloc.c  |  16 +
 5 files changed, 973 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/qbman/bman.c
 create mode 100644 drivers/bus/dpaa/base/qbman/bman.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index e1415e4..61b6432 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -63,6 +63,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/of.c \
base/fman/netcfg_layer.c \
base/qbman/process.c \
+   base/qbman/bman.c \
base/qbman/bman_driver.c \
base/qbman/qman.c \
base/qbman/qman_driver.c \
diff --git a/drivers/bus/dpaa/base/qbman/bman.c 
b/drivers/bus/dpaa/base/qbman/bman.c
new file mode 100644
index 000..0480caa
--- /dev/null
+++ b/drivers/bus/dpaa/base/qbman/bman.c
@@ -0,0 +1,394 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2008-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include "bman.h"
+#include 
+
+/* Compilation constants */
+#define RCR_THRESH 2   /* reread h/w CI when running out of space */
+#define IRQNAME"BMan portal %d"
+#define MAX_IRQNAME16  /* big enough for "BMan portal %d" */
+
+struct bman_portal {
+   struct bm_portal p;
+   /* 2-element array. pools[0] is mask, pools[1] is snapshot. */
+   struct bman_depletion *pools;
+   int thresh_set;
+   unsigned long irq_sources;
+   u32 slowpoll;   /* only used when interrupts are off */
+   /* When the cpu-affine portal is activated, this is non-NULL */
+   const struct bm_portal_config *config;
+   char irqname[MAX_IRQNAME];
+};
+
+static cpumask_t affine_mask;
+static DEFINE_SPINLOCK(affine_mask_lock);
+static RTE_DEFINE_PER_LCORE(struct bman_portal, bman_affine_portal);
+
+static inline struct bman_portal *get_affine_portal(void)
+{
+   return &RTE_PER_LCORE(bman_affine_portal);
+}
+
+/*
+ * This object type refers to a pool, it isn't *the* pool. There may be
+ * more than one such object per BMan buffer pool, eg. if different users of
+ * the pool are operating via different portals.
+ */
+struct bman_pool {
+   struct bman_pool_params params;
+   /* Used for hash-table admin when using depletion notifications. */
+   struct bman_portal *portal;
+   struct bman_pool *next;
+#ifdef RTE_LIBRTE_DPAA_HWDEBUG
+   atomic_t in_use;
+#endif
+};
+
+static inline
+struct bman_portal *bman_create_portal(struct bman_portal *portal,
+  const struct bm_portal_config *c)
+{
+   struct bm_portal *p;
+   const struct bman_depletion *pools = &c->mask;
+   int ret;
+   u8 bpid = 0;

[dpdk-dev] [PATCH v5 16/40] bus/dpaa: integrate DPAA Bus with hardware blocks

[Shreyansh Jain]

Now that QBMAN (QMAN, BMAN) and FMAN drivers are available, this patch
integrates the DPAA Bus driver for using the drivers for scanning
devices and calling the PMD registered probe callbacks.

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/dpaa_bus.c   | 248 ++
 drivers/bus/dpaa/rte_bus_dpaa_version.map |  47 ++
 drivers/bus/dpaa/rte_dpaa_bus.h   |  25 +++
 3 files changed, 320 insertions(+)

diff --git a/drivers/bus/dpaa/dpaa_bus.c b/drivers/bus/dpaa/dpaa_bus.c
index cc343b3..8017df3 100644
--- a/drivers/bus/dpaa/dpaa_bus.c
+++ b/drivers/bus/dpaa/dpaa_bus.c
@@ -63,9 +63,21 @@
 #include 
 #include 
 
+#include 
+#include 
+#include 
+#include 
+#include 
+
 int dpaa_logtype_bus;
 
 struct rte_dpaa_bus rte_dpaa_bus;
+struct netcfg_info *dpaa_netcfg;
+
+/* define a variable to hold the portal_key, once created.*/
+pthread_key_t dpaa_portal_key;
+
+RTE_DEFINE_PER_LCORE(bool, _dpaa_io);
 
 static inline void
 dpaa_add_to_device_list(struct rte_dpaa_device *dev)
@@ -79,11 +91,247 @@ dpaa_remove_from_device_list(struct rte_dpaa_device *dev)
TAILQ_INSERT_TAIL(&rte_dpaa_bus.device_list, dev, next);
 }
 
+static void dpaa_clean_device_list(void);
+
+static int
+dpaa_create_device_list(void)
+{
+   int i;
+   int ret;
+   struct rte_dpaa_device *dev;
+   struct fm_eth_port_cfg *cfg;
+   struct fman_if *fman_intf;
+
+   /* Creating Ethernet Devices */
+   for (i = 0; i < dpaa_netcfg->num_ethports; i++) {
+   dev = calloc(1, sizeof(struct rte_dpaa_device));
+   if (!dev) {
+   DPAA_BUS_LOG(ERR, "Failed to allocate ETH devices");
+   ret = -ENOMEM;
+   goto cleanup;
+   }
+
+   cfg = &dpaa_netcfg->port_cfg[i];
+   fman_intf = cfg->fman_if;
+
+   /* Device identifiers */
+   dev->id.fman_id = fman_intf->fman_idx + 1;
+   dev->id.mac_id = fman_intf->mac_idx;
+   dev->device_type = FSL_DPAA_ETH;
+   dev->id.dev_id = i;
+
+   /* Create device name */
+   memset(dev->name, 0, RTE_ETH_NAME_MAX_LEN);
+   sprintf(dev->name, "fm%d-mac%d", (fman_intf->fman_idx + 1),
+   fman_intf->mac_idx);
+   DPAA_BUS_LOG(DEBUG, "Device added: %s", dev->name);
+   dev->device.name = dev->name;
+
+   dpaa_add_to_device_list(dev);
+   }
+
+   rte_dpaa_bus.device_count = i;
+
+   return 0;
+
+cleanup:
+   dpaa_clean_device_list();
+   return ret;
+}
+
+static void
+dpaa_clean_device_list(void)
+{
+   struct rte_dpaa_device *dev = NULL;
+   struct rte_dpaa_device *tdev = NULL;
+
+   TAILQ_FOREACH_SAFE(dev, &rte_dpaa_bus.device_list, next, tdev) {
+   TAILQ_REMOVE(&rte_dpaa_bus.device_list, dev, next);
+   free(dev);
+   dev = NULL;
+   }
+}
+
+/** XXX move this function into a separate file */
+static int
+_dpaa_portal_init(void *arg)
+{
+   cpu_set_t cpuset;
+   pthread_t id;
+   uint32_t cpu = rte_lcore_id();
+   int ret;
+   struct dpaa_portal *dpaa_io_portal;
+
+   BUS_INIT_FUNC_TRACE();
+
+   if ((uint64_t)arg == 1 || cpu == LCORE_ID_ANY)
+   cpu = rte_get_master_lcore();
+   /* if the core id is not supported */
+   else
+   if (cpu >= RTE_MAX_LCORE)
+   return -1;
+
+   /* Set CPU affinity for this thread */
+   CPU_ZERO(&cpuset);
+   CPU_SET(cpu, &cpuset);
+   id = pthread_self();
+   ret = pthread_setaffinity_np(id, sizeof(cpu_set_t), &cpuset);
+   if (ret) {
+   DPAA_BUS_LOG(ERR, "pthread_setaffinity_np failed on "
+   "core :%d with ret: %d", cpu, ret);
+   return ret;
+   }
+
+   /* Initialise bman thread portals */
+   ret = bman_thread_init();
+   if (ret) {
+   DPAA_BUS_LOG(ERR, "bman_thread_init failed on "
+   "core %d with ret: %d", cpu, ret);
+   return ret;
+   }
+
+   DPAA_BUS_LOG(DEBUG, "BMAN thread initialized");
+
+   /* Initialise qman thread portals */
+   ret = qman_thread_init();
+   if (ret) {
+   DPAA_BUS_LOG(ERR, "bman_thread_init failed on "
+   "core %d with ret: %d", cpu, ret);
+   bman_thread_finish();
+   return ret;
+   }
+
+   DPAA_BUS_LOG(DEBUG, "QMAN thread initialized");
+
+   dpaa_io_portal = rte_malloc(NULL, sizeof(struct dpaa_portal),
+   RTE_CACHE_LINE_SIZE);
+   if (!dpaa_io_portal) {
+   DPAA_BUS_LOG(ERR, "Unable to allocate memory");
+   bman_thread_finish();
+   qman_thread_finish();
+   return -ENOMEM;
+   }
+
+   dpaa_io_portal->qman_idx = qman_get_portal_

[dpdk-dev] [PATCH v5 15/40] bus/dpaa: add fman flow control threshold setting

[Shreyansh Jain]

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/base/fman/fman_hw.c | 28 
 drivers/bus/dpaa/include/fsl_fman.h  |  7 +++
 2 files changed, 35 insertions(+)

diff --git a/drivers/bus/dpaa/base/fman/fman_hw.c 
b/drivers/bus/dpaa/base/fman/fman_hw.c
index a7ca661..077c17c 100644
--- a/drivers/bus/dpaa/base/fman/fman_hw.c
+++ b/drivers/bus/dpaa/base/fman/fman_hw.c
@@ -37,6 +37,7 @@
  */
 #include 
 #include 
+#include 
 
 /* Instantiate the global variable that the inline CRC64 implementation (in
  * ) depends on.
@@ -393,6 +394,33 @@ fman_if_set_bp(struct fman_if *fm_if, unsigned num 
__always_unused,
 }
 
 int
+fman_if_get_fc_threshold(struct fman_if *fm_if)
+{
+   struct __fman_if *__if = container_of(fm_if, struct __fman_if, __if);
+   unsigned int *fmbm_mpd;
+
+   assert(fman_ccsr_map_fd != -1);
+
+   fmbm_mpd = &((struct rx_bmi_regs *)__if->bmi_map)->fmbm_mpd;
+   return in_be32(fmbm_mpd);
+}
+
+int
+fman_if_set_fc_threshold(struct fman_if *fm_if, u32 high_water,
+u32 low_water, u32 bpid)
+{
+   struct __fman_if *__if = container_of(fm_if, struct __fman_if, __if);
+   unsigned int *fmbm_mpd;
+
+   assert(fman_ccsr_map_fd != -1);
+
+   fmbm_mpd = &((struct rx_bmi_regs *)__if->bmi_map)->fmbm_mpd;
+   out_be32(fmbm_mpd, FMAN_ENABLE_BPOOL_DEPLETION);
+   return bm_pool_set_hw_threshold(bpid, low_water, high_water);
+
+}
+
+int
 fman_if_get_fc_quanta(struct fman_if *fm_if)
 {
struct __fman_if *__if = container_of(fm_if, struct __fman_if, __if);
diff --git a/drivers/bus/dpaa/include/fsl_fman.h 
b/drivers/bus/dpaa/include/fsl_fman.h
index ac38082..95aee67 100644
--- a/drivers/bus/dpaa/include/fsl_fman.h
+++ b/drivers/bus/dpaa/include/fsl_fman.h
@@ -112,6 +112,13 @@ void fman_if_loopback_disable(struct fman_if *p);
 void fman_if_set_bp(struct fman_if *fm_if, unsigned int num, int bpid,
size_t bufsize);
 
+/* Get Flow Control threshold parameters on specific interface */
+int fman_if_get_fc_threshold(struct fman_if *fm_if);
+
+/* Enable and Set Flow Control threshold parameters on specific interface */
+int fman_if_set_fc_threshold(struct fman_if *fm_if,
+   u32 high_water, u32 low_water, u32 bpid);
+
 /* Get Flow Control pause quanta on specific interface */
 int fman_if_get_fc_quanta(struct fman_if *fm_if);
 
-- 
2.9.3

[dpdk-dev] [PATCH v5 17/40] doc: add NXP DPAA PMD documentation

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 MAINTAINERS   |   2 +
 doc/guides/nics/dpaa.rst  | 377 ++
 doc/guides/nics/features/dpaa.ini |   8 +
 doc/guides/nics/index.rst |   1 +
 4 files changed, 388 insertions(+)
 create mode 100644 doc/guides/nics/dpaa.rst
 create mode 100644 doc/guides/nics/features/dpaa.ini

diff --git a/MAINTAINERS b/MAINTAINERS
index c566962..dad876f 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -412,6 +412,8 @@ NXP dpaa
 M: Hemant Agrawal 
 M: Shreyansh Jain 
 F: drivers/bus/dpaa/
+F: doc/guides/nics/dpaa.rst
+F: doc/guides/nics/features/dpaa.ini
 
 NXP dpaa2
 M: Hemant Agrawal 
diff --git a/doc/guides/nics/dpaa.rst b/doc/guides/nics/dpaa.rst
new file mode 100644
index 000..7d054d7
--- /dev/null
+++ b/doc/guides/nics/dpaa.rst
@@ -0,0 +1,377 @@
+..  BSD LICENSE
+Copyright 2017 NXP.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions
+are met:
+
+* Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+* Redistributions in binary form must reproduce the above copyright
+notice, this list of conditions and the following disclaimer in
+the documentation and/or other materials provided with the
+distribution.
+* Neither the name of NXP nor the names of its
+contributors may be used to endorse or promote products derived
+from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+DPAA Poll Mode Driver
+=
+
+The DPAA NIC PMD (**librte_pmd_dpaa**) provides poll mode driver
+support for the inbuilt NIC found in the **NXP DPAA** SoC family.
+
+More information can be found at `NXP Official Website
+`_.
+
+NXP DPAA (Data Path Acceleration Architecture - Gen 1)
+--
+
+This section provides an overview of the NXP DPAA architecture
+and how it is integrated into the DPDK.
+
+Contents summary
+
+- DPAA overview
+- DPAA driver architecture overview
+
+.. _dpaa_overview:
+
+DPAA Overview
+~
+
+Reference: `FSL DPAA Architecture 
`_.
+
+The QorIQ Data Path Acceleration Architecture (DPAA) is a set of hardware
+components on specific QorIQ series multicore processors. This architecture
+provides the infrastructure to support simplified sharing of networking
+interfaces and accelerators by multiple CPU cores, and the accelerators
+themselves.
+
+DPAA includes:
+
+- Cores
+- Network and packet I/O
+- Hardware offload accelerators
+- Infrastructure required to facilitate flow of packets between the components 
above
+
+Infrastructure components are:
+
+- The Queue Manager (QMan) is a hardware accelerator that manages frame queues.
+  It allows  CPUs and other accelerators connected to the SoC datapath to
+  enqueue and dequeue ethernet frames, thus providing the infrastructure for
+  data exchange among CPUs and datapath accelerators.
+- The Buffer Manager (BMan) is a hardware buffer pool management block that
+  allows software and accelerators on the datapath to acquire and release
+  buffers in order to build frames.
+
+Hardware accelerators are:
+
+- SEC - Cryptographic accelerator
+- PME - Pattern matching engine
+
+The Network and packet I/O component:
+
+- The Frame Manager (FMan) is a key component in the DPAA and makes use of the
+  DPAA infrastructure (QMan and BMan). FMan  is responsible for packet
+  distribution and policing. Each frame can be parsed, classified and results
+  may be attached to the frame. This meta data can be used to select
+  particular QMan queue, which the packet is forwarded to.
+
+
+DPAA DPDK - Poll Mode Driver Overview
+-
+
+This section provides an overview of the drivers for DPAA:
+
+* Bus driver and associated "DPAA infrastructure" drivers
+* Functional object dr

[dpdk-dev] [PATCH v5 19/40] mempool/dpaa: support NXP DPAA Mempool

[Shreyansh Jain]

This Mempool driver works with DPAA BMan hardware block. This block
manages data buffers in memory, and provides efficient interface with
other hardware and software components for buffer requests.

This patch adds support for BMan. Compilation would be enabled in
subsequent patches.

Signed-off-by: Shreyansh Jain 
---
 MAINTAINERS   |   1 +
 drivers/mempool/Makefile  |   2 +
 drivers/mempool/dpaa/Makefile |  58 +
 drivers/mempool/dpaa/dpaa_mempool.c   | 286 ++
 drivers/mempool/dpaa/dpaa_mempool.h   |  77 ++
 drivers/mempool/dpaa/rte_mempool_dpaa_version.map |   8 +
 6 files changed, 432 insertions(+)
 create mode 100644 drivers/mempool/dpaa/Makefile
 create mode 100644 drivers/mempool/dpaa/dpaa_mempool.c
 create mode 100644 drivers/mempool/dpaa/dpaa_mempool.h
 create mode 100644 drivers/mempool/dpaa/rte_mempool_dpaa_version.map

diff --git a/MAINTAINERS b/MAINTAINERS
index dad876f..022715f 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -412,6 +412,7 @@ NXP dpaa
 M: Hemant Agrawal 
 M: Shreyansh Jain 
 F: drivers/bus/dpaa/
+F: drivers/mempool/dpaa/
 F: doc/guides/nics/dpaa.rst
 F: doc/guides/nics/features/dpaa.ini
 
diff --git a/drivers/mempool/Makefile b/drivers/mempool/Makefile
index efd55f2..bfc5f00 100644
--- a/drivers/mempool/Makefile
+++ b/drivers/mempool/Makefile
@@ -32,6 +32,8 @@ include $(RTE_SDK)/mk/rte.vars.mk
 
 core-libs := librte_eal librte_mempool librte_ring
 
+DIRS-$(CONFIG_RTE_LIBRTE_DPAA_MEMPOOL) += dpaa
+DEPDIRS-dpaa = $(core-libs)
 DIRS-$(CONFIG_RTE_LIBRTE_DPAA2_MEMPOOL) += dpaa2
 DEPDIRS-dpaa2 = $(core-libs)
 DIRS-$(CONFIG_RTE_DRIVER_MEMPOOL_RING) += ring
diff --git a/drivers/mempool/dpaa/Makefile b/drivers/mempool/dpaa/Makefile
new file mode 100644
index 000..25312a0
--- /dev/null
+++ b/drivers/mempool/dpaa/Makefile
@@ -0,0 +1,58 @@
+#   BSD LICENSE
+#
+#   Copyright 2016 NXP.
+#
+#   Redistribution and use in source and binary forms, with or without
+#   modification, are permitted provided that the following conditions
+#   are met:
+#
+# * Redistributions of source code must retain the above copyright
+#   notice, this list of conditions and the following disclaimer.
+# * Redistributions in binary form must reproduce the above copyright
+#   notice, this list of conditions and the following disclaimer in
+#   the documentation and/or other materials provided with the
+#   distribution.
+# * Neither the name of NXP nor the names of its
+#   contributors may be used to endorse or promote products derived
+#   from this software without specific prior written permission.
+#
+#   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+#   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+#   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+#   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+#   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+#   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+#   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+#   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+#   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+#   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+#   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+include $(RTE_SDK)/mk/rte.vars.mk
+
+#
+# library name
+#
+LIB = librte_mempool_dpaa.a
+
+CFLAGS := -I$(SRCDIR) $(CFLAGS)
+CFLAGS += -O3 $(WERROR_FLAGS)
+CFLAGS += -D _GNU_SOURCE
+CFLAGS += -I$(RTE_SDK)/drivers/bus/dpaa
+CFLAGS += -I$(RTE_SDK)/drivers/bus/dpaa/include/
+CFLAGS += -I$(RTE_SDK)/drivers/mempool/dpaa
+CFLAGS += -I$(RTE_SDK)/lib/librte_mempool
+
+# versioning export map
+EXPORT_MAP := rte_mempool_dpaa_version.map
+
+# Lbrary version
+LIBABIVER := 1
+
+# all source are stored in SRCS-y
+#
+SRCS-$(CONFIG_RTE_LIBRTE_DPAA_MEMPOOL) += dpaa_mempool.c
+
+LDLIBS += -lrte_bus_dpaa
+
+include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/mempool/dpaa/dpaa_mempool.c 
b/drivers/mempool/dpaa/dpaa_mempool.c
new file mode 100644
index 000..921c36b
--- /dev/null
+++ b/drivers/mempool/dpaa/dpaa_mempool.c
@@ -0,0 +1,286 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright 2017 NXP.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ *   notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ *   notice, this list of conditions and the following disclaimer in
+ *   the documentation and/or other materials provided with the
+ *   distribution.
+ * * Neither the name of 

[dpdk-dev] [PATCH v5 21/40] bus/dpaa: add DPAA PMD logging macros

[Shreyansh Jain]

Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/dpaa_bus.c   |  5 +
 drivers/bus/dpaa/rte_bus_dpaa_version.map |  1 +
 drivers/bus/dpaa/rte_dpaa_logs.h  | 22 ++
 3 files changed, 28 insertions(+)

diff --git a/drivers/bus/dpaa/dpaa_bus.c b/drivers/bus/dpaa/dpaa_bus.c
index dc2b3ad..7ae5bfa 100644
--- a/drivers/bus/dpaa/dpaa_bus.c
+++ b/drivers/bus/dpaa/dpaa_bus.c
@@ -71,6 +71,7 @@
 
 int dpaa_logtype_bus;
 int dpaa_logtype_mempool;
+int dpaa_logtype_pmd;
 
 struct rte_dpaa_bus rte_dpaa_bus;
 struct netcfg_info *dpaa_netcfg;
@@ -457,4 +458,8 @@ dpaa_init_log(void)
dpaa_logtype_mempool = rte_log_register("mempool.dpaa");
if (dpaa_logtype_mempool >= 0)
rte_log_set_level(dpaa_logtype_mempool, RTE_LOG_NOTICE);
+
+   dpaa_logtype_pmd = rte_log_register("pmd.dpaa");
+   if (dpaa_logtype_pmd >= 0)
+   rte_log_set_level(dpaa_logtype_pmd, RTE_LOG_NOTICE);
 }
diff --git a/drivers/bus/dpaa/rte_bus_dpaa_version.map 
b/drivers/bus/dpaa/rte_bus_dpaa_version.map
index a2394b8..64a05a9 100644
--- a/drivers/bus/dpaa/rte_bus_dpaa_version.map
+++ b/drivers/bus/dpaa/rte_bus_dpaa_version.map
@@ -9,6 +9,7 @@ DPDK_17.11 {
bman_query_free_buffers;
bman_release;
dpaa_logtype_mempool;
+   dpaa_logtype_pmd;
dpaa_netcfg;
fman_ccsr_map_fd;
fman_dealloc_bufs_mask_hi;
diff --git a/drivers/bus/dpaa/rte_dpaa_logs.h b/drivers/bus/dpaa/rte_dpaa_logs.h
index 5335fd8..037c96b 100644
--- a/drivers/bus/dpaa/rte_dpaa_logs.h
+++ b/drivers/bus/dpaa/rte_dpaa_logs.h
@@ -37,6 +37,7 @@
 
 extern int dpaa_logtype_bus;
 extern int dpaa_logtype_mempool;
+extern int dpaa_logtype_pmd;
 
 #define DPAA_BUS_LOG(level, fmt, args...) \
rte_log(RTE_LOG_ ## level, dpaa_logtype_bus, "%s(): " fmt "\n", \
@@ -82,4 +83,25 @@ extern int dpaa_logtype_mempool;
 #define DPAA_MEMPOOL_WARN(fmt, args...) \
DPAA_MEMPOOL_LOG(WARNING, fmt, ## args)
 
+/* PMD related logs */
+
+#define DPAA_PMD_LOG(level, fmt, args...) \
+   rte_log(RTE_LOG_ ## level, dpaa_logtype_pmd, "%s(): " fmt "\n", \
+   __func__, ##args)
+
+#define PMD_INIT_FUNC_TRACE() DPAA_PMD_LOG(DEBUG, " >>")
+
+#define DPAA_PMD_DEBUG(fmt, args...) \
+   DPAA_PMD_LOG(DEBUG, fmt, ## args)
+#define DPAA_PMD_ERR(fmt, args...) \
+   DPAA_PMD_LOG(ERR, fmt, ## args)
+#define DPAA_PMD_INFO(fmt, args...) \
+   DPAA_PMD_LOG(INFO, fmt, ## args)
+#define DPAA_PMD_WARN(fmt, args...) \
+   DPAA_PMD_LOG(WARNING, fmt, ## args)
+
+/* DP Logs, toggled out at compile time if level lower than current level */
+#define DPAA_DP_LOG(level, fmt, args...) \
+   RTE_LOG_DP(level, PMD, fmt, ## args)
+
 #endif /* _DPAA_LOGS_H_ */
-- 
2.9.3

[dpdk-dev] [PATCH v5 18/40] bus/dpaa: add DPAA mempool logging macros

[Shreyansh Jain]

Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/dpaa_bus.c   |  5 +
 drivers/bus/dpaa/rte_bus_dpaa_version.map |  1 +
 drivers/bus/dpaa/rte_dpaa_logs.h  | 20 
 3 files changed, 26 insertions(+)

diff --git a/drivers/bus/dpaa/dpaa_bus.c b/drivers/bus/dpaa/dpaa_bus.c
index 8017df3..dc2b3ad 100644
--- a/drivers/bus/dpaa/dpaa_bus.c
+++ b/drivers/bus/dpaa/dpaa_bus.c
@@ -70,6 +70,7 @@
 #include 
 
 int dpaa_logtype_bus;
+int dpaa_logtype_mempool;
 
 struct rte_dpaa_bus rte_dpaa_bus;
 struct netcfg_info *dpaa_netcfg;
@@ -452,4 +453,8 @@ dpaa_init_log(void)
dpaa_logtype_bus = rte_log_register("bus.dpaa");
if (dpaa_logtype_bus >= 0)
rte_log_set_level(dpaa_logtype_bus, RTE_LOG_NOTICE);
+
+   dpaa_logtype_mempool = rte_log_register("mempool.dpaa");
+   if (dpaa_logtype_mempool >= 0)
+   rte_log_set_level(dpaa_logtype_mempool, RTE_LOG_NOTICE);
 }
diff --git a/drivers/bus/dpaa/rte_bus_dpaa_version.map 
b/drivers/bus/dpaa/rte_bus_dpaa_version.map
index 853bc47..a2394b8 100644
--- a/drivers/bus/dpaa/rte_bus_dpaa_version.map
+++ b/drivers/bus/dpaa/rte_bus_dpaa_version.map
@@ -8,6 +8,7 @@ DPDK_17.11 {
bman_new_pool;
bman_query_free_buffers;
bman_release;
+   dpaa_logtype_mempool;
dpaa_netcfg;
fman_ccsr_map_fd;
fman_dealloc_bufs_mask_hi;
diff --git a/drivers/bus/dpaa/rte_dpaa_logs.h b/drivers/bus/dpaa/rte_dpaa_logs.h
index cc10937..5335fd8 100644
--- a/drivers/bus/dpaa/rte_dpaa_logs.h
+++ b/drivers/bus/dpaa/rte_dpaa_logs.h
@@ -36,6 +36,7 @@
 #include 
 
 extern int dpaa_logtype_bus;
+extern int dpaa_logtype_mempool;
 
 #define DPAA_BUS_LOG(level, fmt, args...) \
rte_log(RTE_LOG_ ## level, dpaa_logtype_bus, "%s(): " fmt "\n", \
@@ -62,4 +63,23 @@ extern int dpaa_logtype_bus;
 #define DPAA_BUS_WARN(fmt, args...) \
DPAA_BUS_LOG(WARNING, fmt, ## args)
 
+/* Mempool related logs */
+
+#define DPAA_MEMPOOL_LOG(level, fmt, args...) \
+   rte_log(RTE_LOG_ ## level, dpaa_logtype_mempool, "%s(): " fmt "\n", \
+   __func__, ##args)
+
+#define MEMPOOL_INIT_FUNC_TRACE() DPAA_MEMPOOL_LOG(DEBUG, " >>")
+
+#define DPAA_MEMPOOL_DPDEBUG(fmt, args...) \
+   RTE_LOG_DP(DEBUG, PMD, fmt, ## args)
+#define DPAA_MEMPOOL_DEBUG(fmt, args...) \
+   DPAA_MEMPOOL_LOG(DEBUG, fmt, ## args)
+#define DPAA_MEMPOOL_ERR(fmt, args...) \
+   DPAA_MEMPOOL_LOG(ERR, fmt, ## args)
+#define DPAA_MEMPOOL_INFO(fmt, args...) \
+   DPAA_MEMPOOL_LOG(INFO, fmt, ## args)
+#define DPAA_MEMPOOL_WARN(fmt, args...) \
+   DPAA_MEMPOOL_LOG(WARNING, fmt, ## args)
+
 #endif /* _DPAA_LOGS_H_ */
-- 
2.9.3

[dpdk-dev] [PATCH v5 20/40] config: enable compilation of DPAA Mempool driver

[Shreyansh Jain]

This patch also adds configuration necessary for compilation of DPAA
Mempool driver into the DPAA specific config file.
CONFIG_RTE_MBUF_DEFAULT_MEMPOOL_OPS=dpaa is also configured to allow
applications to use DPAA mempool as default.

Signed-off-by: Shreyansh Jain 
---
 config/common_base   | 1 +
 config/defconfig_arm64-dpaa-linuxapp-gcc | 4 
 2 files changed, 5 insertions(+)

diff --git a/config/common_base b/config/common_base
index fc1cdca..fe287b0 100644
--- a/config/common_base
+++ b/config/common_base
@@ -303,6 +303,7 @@ CONFIG_RTE_LIBRTE_LIO_DEBUG_REGS=n
 
 # NXP DPAA Bus
 CONFIG_RTE_LIBRTE_DPAA_BUS=n
+CONFIG_RTE_LIBRTE_DPAA_MEMPOOL=n
 
 #
 # Compile NXP DPAA2 FSL-MC Bus
diff --git a/config/defconfig_arm64-dpaa-linuxapp-gcc 
b/config/defconfig_arm64-dpaa-linuxapp-gcc
index 4d6b046..3e11718 100644
--- a/config/defconfig_arm64-dpaa-linuxapp-gcc
+++ b/config/defconfig_arm64-dpaa-linuxapp-gcc
@@ -50,3 +50,7 @@ CONFIG_RTE_PKTMBUF_HEADROOM=128
 CONFIG_RTE_LIBRTE_DPAA_BUS=y
 CONFIG_RTE_LIBRTE_DPAA_DEBUG_DRIVER=n
 CONFIG_RTE_LIBRTE_DPAA_HWDEBUG=n
+
+# NXP DPAA Mempool
+CONFIG_RTE_LIBRTE_DPAA_MEMPOOL=y
+CONFIG_RTE_MBUF_DEFAULT_MEMPOOL_OPS="dpaa"
-- 
2.9.3

[dpdk-dev] [PATCH v5 22/40] net/dpaa: add NXP DPAA PMD driver skeleton

[Shreyansh Jain]

A skeleton which would be called after bus device scan. It currently
fails to identify the device.

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 MAINTAINERS   |   1 +
 drivers/net/dpaa/Makefile |  57 +++
 drivers/net/dpaa/dpaa_ethdev.c| 256 ++
 drivers/net/dpaa/dpaa_ethdev.h| 127 +++
 drivers/net/dpaa/rte_pmd_dpaa_version.map |   4 +
 5 files changed, 445 insertions(+)
 create mode 100644 drivers/net/dpaa/Makefile
 create mode 100644 drivers/net/dpaa/dpaa_ethdev.c
 create mode 100644 drivers/net/dpaa/dpaa_ethdev.h
 create mode 100644 drivers/net/dpaa/rte_pmd_dpaa_version.map

diff --git a/MAINTAINERS b/MAINTAINERS
index 022715f..9eec984 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -413,6 +413,7 @@ M: Hemant Agrawal 
 M: Shreyansh Jain 
 F: drivers/bus/dpaa/
 F: drivers/mempool/dpaa/
+F: drivers/net/dpaa/
 F: doc/guides/nics/dpaa.rst
 F: doc/guides/nics/features/dpaa.ini
 
diff --git a/drivers/net/dpaa/Makefile b/drivers/net/dpaa/Makefile
new file mode 100644
index 000..bb305ca
--- /dev/null
+++ b/drivers/net/dpaa/Makefile
@@ -0,0 +1,57 @@
+#   BSD LICENSE
+#
+#   Copyright 2017 NXP.
+#
+#   Redistribution and use in source and binary forms, with or without
+#   modification, are permitted provided that the following conditions
+#   are met:
+#
+# * Redistributions of source code must retain the above copyright
+#   notice, this list of conditions and the following disclaimer.
+# * Redistributions in binary form must reproduce the above copyright
+#   notice, this list of conditions and the following disclaimer in
+#   the documentation and/or other materials provided with the
+#   distribution.
+# * Neither the name of NXP nor the names of its
+#   contributors may be used to endorse or promote products derived
+#   from this software without specific prior written permission.
+#
+#   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+#   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+#   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+#   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+#   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+#   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+#   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+#   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+#   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+#   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+#   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+include $(RTE_SDK)/mk/rte.vars.mk
+RTE_SDK_DPAA=$(RTE_SDK)/drivers/net/dpaa
+
+#
+# library name
+#
+LIB = librte_pmd_dpaa.a
+
+CFLAGS := -I$(SRCDIR) $(CFLAGS)
+CFLAGS += -O3 $(WERROR_FLAGS)
+CFLAGS += -I$(RTE_SDK_DPAA)/
+CFLAGS += -I$(RTE_SDK_DPAA)/include
+CFLAGS += -I$(RTE_SDK)/drivers/bus/dpaa
+CFLAGS += -I$(RTE_SDK)/drivers/bus/dpaa/include/
+CFLAGS += -I$(RTE_SDK)/lib/librte_eal/common/include
+CFLAGS += -I$(RTE_SDK)/lib/librte_eal/linuxapp/eal/include
+
+EXPORT_MAP := rte_pmd_dpaa_version.map
+
+LIBABIVER := 1
+
+# Interfaces with DPDK
+SRCS-$(CONFIG_RTE_LIBRTE_DPAA_PMD) += dpaa_ethdev.c
+
+LDLIBS += -lrte_bus_dpaa
+
+include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
new file mode 100644
index 000..4543dfc
--- /dev/null
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -0,0 +1,256 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright 2016 Freescale Semiconductor, Inc. All rights reserved.
+ *   Copyright 2017 NXP.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ *   notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ *   notice, this list of conditions and the following disclaimer in
+ *   the documentation and/or other materials provided with the
+ *   distribution.
+ * * Neither the name of  Freescale Semiconductor, Inc nor the names of its
+ *   contributors may be used to endorse or promote products derived
+ *   from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUD

[dpdk-dev] [PATCH v5 23/40] config: enable NXP DPAA PMD compilation

[Shreyansh Jain]

Signed-off-by: Shreyansh Jain 
---
 config/common_base   | 1 +
 config/defconfig_arm64-dpaa-linuxapp-gcc | 3 +++
 drivers/net/Makefile | 2 ++
 mk/rte.app.mk| 5 +
 4 files changed, 11 insertions(+)

diff --git a/config/common_base b/config/common_base
index fe287b0..ca47615 100644
--- a/config/common_base
+++ b/config/common_base
@@ -304,6 +304,7 @@ CONFIG_RTE_LIBRTE_LIO_DEBUG_REGS=n
 # NXP DPAA Bus
 CONFIG_RTE_LIBRTE_DPAA_BUS=n
 CONFIG_RTE_LIBRTE_DPAA_MEMPOOL=n
+CONFIG_RTE_LIBRTE_DPAA_PMD=n
 
 #
 # Compile NXP DPAA2 FSL-MC Bus
diff --git a/config/defconfig_arm64-dpaa-linuxapp-gcc 
b/config/defconfig_arm64-dpaa-linuxapp-gcc
index 3e11718..f59834c 100644
--- a/config/defconfig_arm64-dpaa-linuxapp-gcc
+++ b/config/defconfig_arm64-dpaa-linuxapp-gcc
@@ -54,3 +54,6 @@ CONFIG_RTE_LIBRTE_DPAA_HWDEBUG=n
 # NXP DPAA Mempool
 CONFIG_RTE_LIBRTE_DPAA_MEMPOOL=y
 CONFIG_RTE_MBUF_DEFAULT_MEMPOOL_OPS="dpaa"
+
+# Compile software NXP DPAA PMD
+CONFIG_RTE_LIBRTE_DPAA_PMD=y
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index d33c959..2bd42f8 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -51,6 +51,8 @@ DIRS-$(CONFIG_RTE_LIBRTE_PMD_BOND) += bonding
 DEPDIRS-bonding = $(core-libs) librte_cmdline
 DIRS-$(CONFIG_RTE_LIBRTE_CXGBE_PMD) += cxgbe
 DEPDIRS-cxgbe = $(core-libs)
+DIRS-$(CONFIG_RTE_LIBRTE_DPAA_PMD) += dpaa
+DEPDIRS-dpaa = $(core-libs)
 DIRS-$(CONFIG_RTE_LIBRTE_DPAA2_PMD) += dpaa2
 DEPDIRS-dpaa2 = $(core-libs)
 DIRS-$(CONFIG_RTE_LIBRTE_E1000_PMD) += e1000
diff --git a/mk/rte.app.mk b/mk/rte.app.mk
index c25fdd9..9c5a171 100644
--- a/mk/rte.app.mk
+++ b/mk/rte.app.mk
@@ -116,6 +116,7 @@ _LDLIBS-$(CONFIG_RTE_LIBRTE_BNX2X_PMD)  += 
-lrte_pmd_bnx2x -lz
 _LDLIBS-$(CONFIG_RTE_LIBRTE_BNXT_PMD)   += -lrte_pmd_bnxt
 _LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_BOND)   += -lrte_pmd_bond
 _LDLIBS-$(CONFIG_RTE_LIBRTE_CXGBE_PMD)  += -lrte_pmd_cxgbe
+_LDLIBS-$(CONFIG_RTE_LIBRTE_DPAA_PMD)   += -lrte_pmd_dpaa
 _LDLIBS-$(CONFIG_RTE_LIBRTE_DPAA2_PMD)  += -lrte_pmd_dpaa2
 _LDLIBS-$(CONFIG_RTE_LIBRTE_E1000_PMD)  += -lrte_pmd_e1000
 _LDLIBS-$(CONFIG_RTE_LIBRTE_ENA_PMD)+= -lrte_pmd_ena
@@ -182,6 +183,10 @@ _LDLIBS-$(CONFIG_RTE_LIBRTE_DPAA2_PMD)  += 
-lrte_bus_fslmc
 _LDLIBS-$(CONFIG_RTE_LIBRTE_DPAA2_PMD)  += -lrte_mempool_dpaa2
 endif # CONFIG_RTE_LIBRTE_DPAA2_PMD
 
+ifeq ($(CONFIG_RTE_LIBRTE_DPAA_PMD),y)
+_LDLIBS-$(CONFIG_RTE_LIBRTE_DPAA_PMD)   += -lrte_bus_dpaa
+endif
+
 endif # !CONFIG_RTE_BUILD_SHARED_LIBS
 
 _LDLIBS-y += --no-whole-archive
-- 
2.9.3

[dpdk-dev] [PATCH v5 24/40] net/dpaa: support Tx and Rx queue setup

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/net/dpaa/Makefile  |   4 +
 drivers/net/dpaa/dpaa_ethdev.c | 296 -
 drivers/net/dpaa/dpaa_rxtx.c   | 370 +
 drivers/net/dpaa/dpaa_rxtx.h   |  61 +++
 mk/rte.app.mk  |   1 +
 5 files changed, 729 insertions(+), 3 deletions(-)
 create mode 100644 drivers/net/dpaa/dpaa_rxtx.c
 create mode 100644 drivers/net/dpaa/dpaa_rxtx.h

diff --git a/drivers/net/dpaa/Makefile b/drivers/net/dpaa/Makefile
index bb305ca..c77384c 100644
--- a/drivers/net/dpaa/Makefile
+++ b/drivers/net/dpaa/Makefile
@@ -38,10 +38,12 @@ LIB = librte_pmd_dpaa.a
 
 CFLAGS := -I$(SRCDIR) $(CFLAGS)
 CFLAGS += -O3 $(WERROR_FLAGS)
+CFLAGS += -Wno-pointer-arith
 CFLAGS += -I$(RTE_SDK_DPAA)/
 CFLAGS += -I$(RTE_SDK_DPAA)/include
 CFLAGS += -I$(RTE_SDK)/drivers/bus/dpaa
 CFLAGS += -I$(RTE_SDK)/drivers/bus/dpaa/include/
+CFLAGS += -I$(RTE_SDK)/drivers/mempool/dpaa
 CFLAGS += -I$(RTE_SDK)/lib/librte_eal/common/include
 CFLAGS += -I$(RTE_SDK)/lib/librte_eal/linuxapp/eal/include
 
@@ -51,7 +53,9 @@ LIBABIVER := 1
 
 # Interfaces with DPDK
 SRCS-$(CONFIG_RTE_LIBRTE_DPAA_PMD) += dpaa_ethdev.c
+SRCS-$(CONFIG_RTE_LIBRTE_DPAA_PMD) += dpaa_rxtx.c
 
 LDLIBS += -lrte_bus_dpaa
+LDLIBS += -lrte_mempool_dpaa
 
 include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index 4543dfc..2db7d99 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -62,8 +62,15 @@
 
 #include 
 #include 
+#include 
 
 #include 
+#include 
+
+#include 
+#include 
+#include 
+#include 
 
 /* Keep track of whether QMAN and BMAN have been globally initialized */
 static int is_global_init;
@@ -78,20 +85,104 @@ dpaa_eth_dev_configure(struct rte_eth_dev *dev 
__rte_unused)
 
 static int dpaa_eth_dev_start(struct rte_eth_dev *dev)
 {
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
PMD_INIT_FUNC_TRACE();
 
/* Change tx callback to the real one */
-   dev->tx_pkt_burst = NULL;
+   dev->tx_pkt_burst = dpaa_eth_queue_tx;
+   fman_if_enable_rx(dpaa_intf->fif);
 
return 0;
 }
 
 static void dpaa_eth_dev_stop(struct rte_eth_dev *dev)
 {
-   dev->tx_pkt_burst = NULL;
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   PMD_INIT_FUNC_TRACE();
+
+   fman_if_disable_rx(dpaa_intf->fif);
+   dev->tx_pkt_burst = dpaa_eth_tx_drop_all;
 }
 
-static void dpaa_eth_dev_close(struct rte_eth_dev *dev __rte_unused)
+static void dpaa_eth_dev_close(struct rte_eth_dev *dev)
+{
+   PMD_INIT_FUNC_TRACE();
+
+   dpaa_eth_dev_stop(dev);
+}
+
+static
+int dpaa_eth_rx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx,
+   uint16_t nb_desc __rte_unused,
+   unsigned int socket_id __rte_unused,
+   const struct rte_eth_rxconf *rx_conf __rte_unused,
+   struct rte_mempool *mp)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   PMD_INIT_FUNC_TRACE();
+
+   DPAA_PMD_INFO("Rx queue setup for queue index: %d", queue_idx);
+
+   if (!dpaa_intf->bp_info || dpaa_intf->bp_info->mp != mp) {
+   struct fman_if_ic_params icp;
+   uint32_t fd_offset;
+   uint32_t bp_size;
+
+   if (!mp->pool_data) {
+   DPAA_PMD_ERR("Not an offloaded buffer pool!");
+   return -1;
+   }
+   dpaa_intf->bp_info = DPAA_MEMPOOL_TO_POOL_INFO(mp);
+
+   memset(&icp, 0, sizeof(icp));
+   /* set ICEOF for to the default value , which is 0*/
+   icp.iciof = DEFAULT_ICIOF;
+   icp.iceof = DEFAULT_RX_ICEOF;
+   icp.icsz = DEFAULT_ICSZ;
+   fman_if_set_ic_params(dpaa_intf->fif, &icp);
+
+   fd_offset = RTE_PKTMBUF_HEADROOM + DPAA_HW_BUF_RESERVE;
+   fman_if_set_fdoff(dpaa_intf->fif, fd_offset);
+
+   /* Buffer pool size should be equal to Dataroom Size*/
+   bp_size = rte_pktmbuf_data_room_size(mp);
+   fman_if_set_bp(dpaa_intf->fif, mp->size,
+  dpaa_intf->bp_info->bpid, bp_size);
+   dpaa_intf->valid = 1;
+   DPAA_PMD_INFO("if =%s - fd_offset = %d offset = %d",
+   dpaa_intf->name, fd_offset,
+   fman_if_get_fdoff(dpaa_intf->fif));
+   }
+   dev->data->rx_queues[queue_idx] = &dpaa_intf->rx_queues[queue_idx];
+
+   return 0;
+}
+
+static
+void dpaa_eth_rx_queue_release(void *rxq __rte_unused)
+{
+   PMD_INIT_FUNC_TRACE();
+}
+
+static
+int dpaa_eth_tx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx,
+   uint16_t nb_desc __rte_unused,
+   unsigned int socket_id __rte_unused,
+   const struct rte_eth_txc

[dpdk-dev] [PATCH v5 25/40] net/dpaa: support MTU update

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 doc/guides/nics/features/dpaa.ini |  1 +
 drivers/net/dpaa/dpaa_ethdev.c| 21 +
 2 files changed, 22 insertions(+)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index 9e8befc..59ef23d 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -4,5 +4,6 @@
 ; Refer to default.ini for the full list of available PMD features.
 ;
 [Features]
+MTU update   = Y
 ARMv8= Y
 Usage doc= Y
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index 2db7d99..ad0a092 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -76,6 +76,26 @@
 static int is_global_init;
 
 static int
+dpaa_mtu_set(struct rte_eth_dev *dev, uint16_t mtu)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   PMD_INIT_FUNC_TRACE();
+
+   if (mtu < ETHER_MIN_MTU)
+   return -EINVAL;
+   if (mtu > ETHER_MAX_LEN)
+   return -1;
+
+   dev->data->dev_conf.rxmode.jumbo_frame = 0;
+   dev->data->dev_conf.rxmode.max_rx_pkt_len = mtu;
+
+   fman_if_set_maxfrm(dpaa_intf->fif, mtu);
+
+   return 0;
+}
+
+static int
 dpaa_eth_dev_configure(struct rte_eth_dev *dev __rte_unused)
 {
PMD_INIT_FUNC_TRACE();
@@ -197,6 +217,7 @@ static struct eth_dev_ops dpaa_devops = {
.tx_queue_setup   = dpaa_eth_tx_queue_setup,
.rx_queue_release = dpaa_eth_rx_queue_release,
.tx_queue_release = dpaa_eth_tx_queue_release,
+   .mtu_set  = dpaa_mtu_set,
 };
 
 /* Initialise an Rx FQ */
-- 
2.9.3

[dpdk-dev] [PATCH v5 26/40] net/dpaa: support jumbo frames

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 doc/guides/nics/features/dpaa.ini |  1 +
 drivers/net/dpaa/dpaa_ethdev.c| 13 +++--
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index 59ef23d..e62812c 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -4,6 +4,7 @@
 ; Refer to default.ini for the full list of available PMD features.
 ;
 [Features]
+Jumbo frame  = Y
 MTU update   = Y
 ARMv8= Y
 Usage doc= Y
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index ad0a092..c013a84 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -85,9 +85,10 @@ dpaa_mtu_set(struct rte_eth_dev *dev, uint16_t mtu)
if (mtu < ETHER_MIN_MTU)
return -EINVAL;
if (mtu > ETHER_MAX_LEN)
-   return -1;
+   dev->data->dev_conf.rxmode.jumbo_frame = 1;
+   else
+   dev->data->dev_conf.rxmode.jumbo_frame = 0;
 
-   dev->data->dev_conf.rxmode.jumbo_frame = 0;
dev->data->dev_conf.rxmode.max_rx_pkt_len = mtu;
 
fman_if_set_maxfrm(dpaa_intf->fif, mtu);
@@ -100,6 +101,14 @@ dpaa_eth_dev_configure(struct rte_eth_dev *dev 
__rte_unused)
 {
PMD_INIT_FUNC_TRACE();
 
+   if (dev->data->dev_conf.rxmode.jumbo_frame == 1) {
+   if (dev->data->dev_conf.rxmode.max_rx_pkt_len <=
+   DPAA_MAX_RX_PKT_LEN)
+   return dpaa_mtu_set(dev,
+   dev->data->dev_conf.rxmode.max_rx_pkt_len);
+   else
+   return -1;
+   }
return 0;
 }
 
-- 
2.9.3

[dpdk-dev] [PATCH v5 27/40] net/dpaa: support link status update

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 doc/guides/nics/features/dpaa.ini |  1 +
 drivers/net/dpaa/dpaa_ethdev.c| 42 +++
 2 files changed, 43 insertions(+)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index e62812c..132f94b 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -4,6 +4,7 @@
 ; Refer to default.ini for the full list of available PMD features.
 ;
 [Features]
+Link status  = Y
 Jumbo frame  = Y
 MTU update   = Y
 ARMv8= Y
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index c013a84..804c89f 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -142,6 +142,28 @@ static void dpaa_eth_dev_close(struct rte_eth_dev *dev)
dpaa_eth_dev_stop(dev);
 }
 
+static int dpaa_eth_link_update(struct rte_eth_dev *dev,
+   int wait_to_complete __rte_unused)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+   struct rte_eth_link *link = &dev->data->dev_link;
+
+   PMD_INIT_FUNC_TRACE();
+
+   if (dpaa_intf->fif->mac_type == fman_mac_1g)
+   link->link_speed = 1000;
+   else if (dpaa_intf->fif->mac_type == fman_mac_10g)
+   link->link_speed = 1;
+   else
+   DPAA_PMD_ERR("invalid link_speed: %s, %d",
+dpaa_intf->name, dpaa_intf->fif->mac_type);
+
+   link->link_status = dpaa_intf->valid;
+   link->link_duplex = ETH_LINK_FULL_DUPLEX;
+   link->link_autoneg = ETH_LINK_AUTONEG;
+   return 0;
+}
+
 static
 int dpaa_eth_rx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx,
uint16_t nb_desc __rte_unused,
@@ -216,6 +238,22 @@ static void dpaa_eth_tx_queue_release(void *txq 
__rte_unused)
PMD_INIT_FUNC_TRACE();
 }
 
+static int dpaa_link_down(struct rte_eth_dev *dev)
+{
+   PMD_INIT_FUNC_TRACE();
+
+   dpaa_eth_dev_stop(dev);
+   return 0;
+}
+
+static int dpaa_link_up(struct rte_eth_dev *dev)
+{
+   PMD_INIT_FUNC_TRACE();
+
+   dpaa_eth_dev_start(dev);
+   return 0;
+}
+
 static struct eth_dev_ops dpaa_devops = {
.dev_configure= dpaa_eth_dev_configure,
.dev_start= dpaa_eth_dev_start,
@@ -226,7 +264,11 @@ static struct eth_dev_ops dpaa_devops = {
.tx_queue_setup   = dpaa_eth_tx_queue_setup,
.rx_queue_release = dpaa_eth_rx_queue_release,
.tx_queue_release = dpaa_eth_tx_queue_release,
+
+   .link_update  = dpaa_eth_link_update,
.mtu_set  = dpaa_mtu_set,
+   .dev_set_link_down= dpaa_link_down,
+   .dev_set_link_up  = dpaa_link_up,
 };
 
 /* Initialise an Rx FQ */
-- 
2.9.3

[dpdk-dev] [PATCH v5 29/40] net/dpaa: support promiscuous toggle

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 doc/guides/nics/features/dpaa.ini |  1 +
 drivers/net/dpaa/dpaa_ethdev.c| 21 +
 2 files changed, 22 insertions(+)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index 19beada..b2dfd81 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -8,5 +8,6 @@ Speed capabilities   = P
 Link status  = Y
 Jumbo frame  = Y
 MTU update   = Y
+Promiscuous mode = Y
 ARMv8= Y
 Usage doc= Y
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index 384be8e..56a6039 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -183,6 +183,25 @@ static int dpaa_eth_link_update(struct rte_eth_dev *dev,
return 0;
 }
 
+
+static void dpaa_eth_promiscuous_enable(struct rte_eth_dev *dev)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   PMD_INIT_FUNC_TRACE();
+
+   fman_if_promiscuous_enable(dpaa_intf->fif);
+}
+
+static void dpaa_eth_promiscuous_disable(struct rte_eth_dev *dev)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   PMD_INIT_FUNC_TRACE();
+
+   fman_if_promiscuous_disable(dpaa_intf->fif);
+}
+
 static
 int dpaa_eth_rx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx,
uint16_t nb_desc __rte_unused,
@@ -286,6 +305,8 @@ static struct eth_dev_ops dpaa_devops = {
.tx_queue_release = dpaa_eth_tx_queue_release,
 
.link_update  = dpaa_eth_link_update,
+   .promiscuous_enable   = dpaa_eth_promiscuous_enable,
+   .promiscuous_disable  = dpaa_eth_promiscuous_disable,
.mtu_set  = dpaa_mtu_set,
.dev_set_link_down= dpaa_link_down,
.dev_set_link_up  = dpaa_link_up,
-- 
2.9.3

[dpdk-dev] [PATCH v5 28/40] net/dpaa: support device info and speed capability

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 doc/guides/nics/features/dpaa.ini |  1 +
 drivers/net/dpaa/dpaa_ethdev.c| 20 
 2 files changed, 21 insertions(+)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index 132f94b..19beada 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -4,6 +4,7 @@
 ; Refer to default.ini for the full list of available PMD features.
 ;
 [Features]
+Speed capabilities   = P
 Link status  = Y
 Jumbo frame  = Y
 MTU update   = Y
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index 804c89f..384be8e 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -142,6 +142,25 @@ static void dpaa_eth_dev_close(struct rte_eth_dev *dev)
dpaa_eth_dev_stop(dev);
 }
 
+static void dpaa_eth_dev_info(struct rte_eth_dev *dev,
+ struct rte_eth_dev_info *dev_info)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   PMD_INIT_FUNC_TRACE();
+
+   dev_info->max_rx_queues = dpaa_intf->nb_rx_queues;
+   dev_info->max_tx_queues = dpaa_intf->nb_tx_queues;
+   dev_info->min_rx_bufsize = DPAA_MIN_RX_BUF_SIZE;
+   dev_info->max_rx_pktlen = DPAA_MAX_RX_PKT_LEN;
+   dev_info->max_mac_addrs = DPAA_MAX_MAC_FILTER;
+   dev_info->max_hash_mac_addrs = 0;
+   dev_info->max_vfs = 0;
+   dev_info->max_vmdq_pools = ETH_16_POOLS;
+   dev_info->speed_capa = (ETH_LINK_SPEED_1G |
+   ETH_LINK_SPEED_10G);
+}
+
 static int dpaa_eth_link_update(struct rte_eth_dev *dev,
int wait_to_complete __rte_unused)
 {
@@ -259,6 +278,7 @@ static struct eth_dev_ops dpaa_devops = {
.dev_start= dpaa_eth_dev_start,
.dev_stop = dpaa_eth_dev_stop,
.dev_close= dpaa_eth_dev_close,
+   .dev_infos_get= dpaa_eth_dev_info,
 
.rx_queue_setup   = dpaa_eth_rx_queue_setup,
.tx_queue_setup   = dpaa_eth_tx_queue_setup,
-- 
2.9.3

[dpdk-dev] [PATCH v5 30/40] net/dpaa: support multicast toggle

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 doc/guides/nics/features/dpaa.ini |  1 +
 drivers/net/dpaa/dpaa_ethdev.c| 20 
 2 files changed, 21 insertions(+)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index b2dfd81..f21a85f 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -9,5 +9,6 @@ Link status  = Y
 Jumbo frame  = Y
 MTU update   = Y
 Promiscuous mode = Y
+Allmulticast mode= Y
 ARMv8= Y
 Usage doc= Y
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index 56a6039..f6a2e22 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -202,6 +202,24 @@ static void dpaa_eth_promiscuous_disable(struct 
rte_eth_dev *dev)
fman_if_promiscuous_disable(dpaa_intf->fif);
 }
 
+static void dpaa_eth_multicast_enable(struct rte_eth_dev *dev)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   PMD_INIT_FUNC_TRACE();
+
+   fman_if_set_mcast_filter_table(dpaa_intf->fif);
+}
+
+static void dpaa_eth_multicast_disable(struct rte_eth_dev *dev)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   PMD_INIT_FUNC_TRACE();
+
+   fman_if_reset_mcast_filter_table(dpaa_intf->fif);
+}
+
 static
 int dpaa_eth_rx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx,
uint16_t nb_desc __rte_unused,
@@ -307,6 +325,8 @@ static struct eth_dev_ops dpaa_devops = {
.link_update  = dpaa_eth_link_update,
.promiscuous_enable   = dpaa_eth_promiscuous_enable,
.promiscuous_disable  = dpaa_eth_promiscuous_disable,
+   .allmulticast_enable  = dpaa_eth_multicast_enable,
+   .allmulticast_disable = dpaa_eth_multicast_disable,
.mtu_set  = dpaa_mtu_set,
.dev_set_link_down= dpaa_link_down,
.dev_set_link_up  = dpaa_link_up,
-- 
2.9.3

[dpdk-dev] [PATCH v5 31/40] net/dpaa: support MAC address update

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 doc/guides/nics/features/dpaa.ini |  1 +
 drivers/net/dpaa/dpaa_ethdev.c| 48 +++
 2 files changed, 49 insertions(+)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index f21a85f..cdf5e46 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -10,5 +10,6 @@ Jumbo frame  = Y
 MTU update   = Y
 Promiscuous mode = Y
 Allmulticast mode= Y
+Unicast MAC filter   = Y
 ARMv8= Y
 Usage doc= Y
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index f6a2e22..f1d0f75 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -310,6 +310,50 @@ static int dpaa_link_up(struct rte_eth_dev *dev)
return 0;
 }
 
+static int
+dpaa_dev_add_mac_addr(struct rte_eth_dev *dev,
+struct ether_addr *addr,
+uint32_t index,
+__rte_unused uint32_t pool)
+{
+   int ret;
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   PMD_INIT_FUNC_TRACE();
+
+   ret = fman_if_add_mac_addr(dpaa_intf->fif, addr->addr_bytes, index);
+
+   if (ret)
+   RTE_LOG(ERR, PMD, "error: Adding the MAC ADDR failed:"
+   " err = %d", ret);
+   return 0;
+}
+
+static void
+dpaa_dev_remove_mac_addr(struct rte_eth_dev *dev,
+ uint32_t index)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   PMD_INIT_FUNC_TRACE();
+
+   fman_if_clear_mac_addr(dpaa_intf->fif, index);
+}
+
+static void
+dpaa_dev_set_mac_addr(struct rte_eth_dev *dev,
+  struct ether_addr *addr)
+{
+   int ret;
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   PMD_INIT_FUNC_TRACE();
+
+   ret = fman_if_add_mac_addr(dpaa_intf->fif, addr->addr_bytes, 0);
+   if (ret)
+   RTE_LOG(ERR, PMD, "error: Setting the MAC ADDR failed %d", ret);
+}
+
 static struct eth_dev_ops dpaa_devops = {
.dev_configure= dpaa_eth_dev_configure,
.dev_start= dpaa_eth_dev_start,
@@ -330,6 +374,10 @@ static struct eth_dev_ops dpaa_devops = {
.mtu_set  = dpaa_mtu_set,
.dev_set_link_down= dpaa_link_down,
.dev_set_link_up  = dpaa_link_up,
+   .mac_addr_add = dpaa_dev_add_mac_addr,
+   .mac_addr_remove  = dpaa_dev_remove_mac_addr,
+   .mac_addr_set = dpaa_dev_set_mac_addr,
+
 };
 
 /* Initialise an Rx FQ */
-- 
2.9.3

[dpdk-dev] [PATCH v5 32/40] net/dpaa: support basic stats

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 doc/guides/nics/features/dpaa.ini |  1 +
 drivers/net/dpaa/dpaa_ethdev.c| 20 
 2 files changed, 21 insertions(+)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index cdf5e46..c09efd8 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -11,5 +11,6 @@ MTU update   = Y
 Promiscuous mode = Y
 Allmulticast mode= Y
 Unicast MAC filter   = Y
+Basic stats  = Y
 ARMv8= Y
 Usage doc= Y
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index f1d0f75..23a9efd 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -183,6 +183,24 @@ static int dpaa_eth_link_update(struct rte_eth_dev *dev,
return 0;
 }
 
+static void dpaa_eth_stats_get(struct rte_eth_dev *dev,
+  struct rte_eth_stats *stats)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   PMD_INIT_FUNC_TRACE();
+
+   fman_if_stats_get(dpaa_intf->fif, stats);
+}
+
+static void dpaa_eth_stats_reset(struct rte_eth_dev *dev)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   PMD_INIT_FUNC_TRACE();
+
+   fman_if_stats_reset(dpaa_intf->fif);
+}
 
 static void dpaa_eth_promiscuous_enable(struct rte_eth_dev *dev)
 {
@@ -367,6 +385,8 @@ static struct eth_dev_ops dpaa_devops = {
.tx_queue_release = dpaa_eth_tx_queue_release,
 
.link_update  = dpaa_eth_link_update,
+   .stats_get= dpaa_eth_stats_get,
+   .stats_reset  = dpaa_eth_stats_reset,
.promiscuous_enable   = dpaa_eth_promiscuous_enable,
.promiscuous_disable  = dpaa_eth_promiscuous_disable,
.allmulticast_enable  = dpaa_eth_multicast_enable,
-- 
2.9.3

[dpdk-dev] [PATCH v5 36/40] net/dpaa: support checksum offload

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 doc/guides/nics/features/dpaa.ini |  2 +
 drivers/net/dpaa/dpaa_ethdev.c|  4 ++
 drivers/net/dpaa/dpaa_rxtx.c  | 89 +++
 drivers/net/dpaa/dpaa_rxtx.h  | 23 +-
 4 files changed, 117 insertions(+), 1 deletion(-)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index 2ef1b56..23626c0 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -13,6 +13,8 @@ Allmulticast mode= Y
 Unicast MAC filter   = Y
 RSS hash = Y
 Flow control = Y
+L3 checksum offload  = Y
+L4 checksum offload  = Y
 Packet type parsing  = Y
 Basic stats  = Y
 ARMv8= Y
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index cf2c0c7..4eb9b62 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -186,6 +186,10 @@ static void dpaa_eth_dev_info(struct rte_eth_dev *dev,
(DEV_RX_OFFLOAD_IPV4_CKSUM |
DEV_RX_OFFLOAD_UDP_CKSUM   |
DEV_RX_OFFLOAD_TCP_CKSUM);
+   dev_info->tx_offload_capa =
+   (DEV_TX_OFFLOAD_IPV4_CKSUM  |
+   DEV_TX_OFFLOAD_UDP_CKSUM   |
+   DEV_TX_OFFLOAD_TCP_CKSUM);
 }
 
 static int dpaa_eth_link_update(struct rte_eth_dev *dev,
diff --git a/drivers/net/dpaa/dpaa_rxtx.c b/drivers/net/dpaa/dpaa_rxtx.c
index f8ac711..976268b 100644
--- a/drivers/net/dpaa/dpaa_rxtx.c
+++ b/drivers/net/dpaa/dpaa_rxtx.c
@@ -200,6 +200,82 @@ static inline void dpaa_eth_packet_info(struct rte_mbuf *m,
/* Packet received without stripping the vlan */
 }
 
+static inline void dpaa_checksum(struct rte_mbuf *mbuf)
+{
+   struct ether_hdr *eth_hdr = rte_pktmbuf_mtod(mbuf, struct ether_hdr *);
+   char *l3_hdr = (char *)eth_hdr + mbuf->l2_len;
+   struct ipv4_hdr *ipv4_hdr = (struct ipv4_hdr *)l3_hdr;
+   struct ipv6_hdr *ipv6_hdr = (struct ipv6_hdr *)l3_hdr;
+
+   DPAA_DP_LOG(DEBUG, "Calculating checksum for mbuf: %p", mbuf);
+
+   if (((mbuf->packet_type & RTE_PTYPE_L3_MASK) == RTE_PTYPE_L3_IPV4) ||
+   ((mbuf->packet_type & RTE_PTYPE_L3_MASK) ==
+   RTE_PTYPE_L3_IPV4_EXT)) {
+   ipv4_hdr = (struct ipv4_hdr *)l3_hdr;
+   ipv4_hdr->hdr_checksum = 0;
+   ipv4_hdr->hdr_checksum = rte_ipv4_cksum(ipv4_hdr);
+   } else if (((mbuf->packet_type & RTE_PTYPE_L3_MASK) ==
+  RTE_PTYPE_L3_IPV6) ||
+  ((mbuf->packet_type & RTE_PTYPE_L3_MASK) ==
+  RTE_PTYPE_L3_IPV6_EXT))
+   ipv6_hdr = (struct ipv6_hdr *)l3_hdr;
+
+   if ((mbuf->packet_type & RTE_PTYPE_L4_MASK) == RTE_PTYPE_L4_TCP) {
+   struct tcp_hdr *tcp_hdr = (struct tcp_hdr *)(l3_hdr +
+ mbuf->l3_len);
+   tcp_hdr->cksum = 0;
+   if (eth_hdr->ether_type == htons(ETHER_TYPE_IPv4))
+   tcp_hdr->cksum = rte_ipv4_udptcp_cksum(ipv4_hdr,
+  tcp_hdr);
+   else /* assume ethertype == ETHER_TYPE_IPv6 */
+   tcp_hdr->cksum = rte_ipv6_udptcp_cksum(ipv6_hdr,
+  tcp_hdr);
+   } else if ((mbuf->packet_type & RTE_PTYPE_L4_MASK) ==
+  RTE_PTYPE_L4_UDP) {
+   struct udp_hdr *udp_hdr = (struct udp_hdr *)(l3_hdr +
+mbuf->l3_len);
+   udp_hdr->dgram_cksum = 0;
+   if (eth_hdr->ether_type == htons(ETHER_TYPE_IPv4))
+   udp_hdr->dgram_cksum = rte_ipv4_udptcp_cksum(ipv4_hdr,
+udp_hdr);
+   else /* assume ethertype == ETHER_TYPE_IPv6 */
+   udp_hdr->dgram_cksum = rte_ipv6_udptcp_cksum(ipv6_hdr,
+udp_hdr);
+   }
+}
+
+static inline void dpaa_checksum_offload(struct rte_mbuf *mbuf,
+struct qm_fd *fd, char *prs_buf)
+{
+   struct dpaa_eth_parse_results_t *prs;
+
+   DPAA_DP_LOG(DEBUG, " Offloading checksum for mbuf: %p", mbuf);
+
+   prs = GET_TX_PRS(prs_buf);
+   prs->l3r = 0;
+   prs->l4r = 0;
+   if (((mbuf->packet_type & RTE_PTYPE_L3_MASK) == RTE_PTYPE_L3_IPV4) ||
+  ((mbuf->packet_type & RTE_PTYPE_L3_MASK) ==
+  RTE_PTYPE_L3_IPV4_EXT))
+   prs->l3r = DPAA_L3_PARSE_RESULT_IPV4;
+   else if (((mbuf->packet_type & RTE_PTYPE_L3_MASK) ==
+  RTE_PTYPE_L3_IPV6) ||
+((mbuf->packet_type & RTE_PTYPE_L3_MASK) ==
+   RTE_PTYPE_L3_IPV6_EXT))
+   prs->l3r = DPAA_L3_PARSE_RESULT_IPV6;
+
+   if ((mbuf->packet_type & RTE_PTYPE_L4_MAS

[dpdk-dev] [PATCH v5 35/40] net/dpaa: support packet type parsing

[Shreyansh Jain]

Add support for parsing the packet type and L2/L3 checksum offload
capability information.

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 doc/guides/nics/features/dpaa.ini |   2 +
 drivers/net/dpaa/dpaa_ethdev.c|  27 +
 drivers/net/dpaa/dpaa_rxtx.c  | 116 +
 drivers/net/dpaa/dpaa_rxtx.h  | 206 ++
 4 files changed, 351 insertions(+)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index 1ba6b11..2ef1b56 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -11,7 +11,9 @@ MTU update   = Y
 Promiscuous mode = Y
 Allmulticast mode= Y
 Unicast MAC filter   = Y
+RSS hash = Y
 Flow control = Y
+Packet type parsing  = Y
 Basic stats  = Y
 ARMv8= Y
 Usage doc= Y
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index 5860dfa..cf2c0c7 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -112,6 +112,28 @@ dpaa_eth_dev_configure(struct rte_eth_dev *dev 
__rte_unused)
return 0;
 }
 
+static const uint32_t *
+dpaa_supported_ptypes_get(struct rte_eth_dev *dev)
+{
+   static const uint32_t ptypes[] = {
+   /*todo -= add more types */
+   RTE_PTYPE_L2_ETHER,
+   RTE_PTYPE_L3_IPV4,
+   RTE_PTYPE_L3_IPV4_EXT,
+   RTE_PTYPE_L3_IPV6,
+   RTE_PTYPE_L3_IPV6_EXT,
+   RTE_PTYPE_L4_TCP,
+   RTE_PTYPE_L4_UDP,
+   RTE_PTYPE_L4_SCTP
+   };
+
+   PMD_INIT_FUNC_TRACE();
+
+   if (dev->rx_pkt_burst == dpaa_eth_queue_rx)
+   return ptypes;
+   return NULL;
+}
+
 static int dpaa_eth_dev_start(struct rte_eth_dev *dev)
 {
struct dpaa_if *dpaa_intf = dev->data->dev_private;
@@ -160,6 +182,10 @@ static void dpaa_eth_dev_info(struct rte_eth_dev *dev,
dev_info->flow_type_rss_offloads = DPAA_RSS_OFFLOAD_ALL;
dev_info->speed_capa = (ETH_LINK_SPEED_1G |
ETH_LINK_SPEED_10G);
+   dev_info->rx_offload_capa =
+   (DEV_RX_OFFLOAD_IPV4_CKSUM |
+   DEV_RX_OFFLOAD_UDP_CKSUM   |
+   DEV_RX_OFFLOAD_TCP_CKSUM);
 }
 
 static int dpaa_eth_link_update(struct rte_eth_dev *dev,
@@ -458,6 +484,7 @@ static struct eth_dev_ops dpaa_devops = {
.dev_stop = dpaa_eth_dev_stop,
.dev_close= dpaa_eth_dev_close,
.dev_infos_get= dpaa_eth_dev_info,
+   .dev_supported_ptypes_get = dpaa_supported_ptypes_get,
 
.rx_queue_setup   = dpaa_eth_rx_queue_setup,
.tx_queue_setup   = dpaa_eth_tx_queue_setup,
diff --git a/drivers/net/dpaa/dpaa_rxtx.c b/drivers/net/dpaa/dpaa_rxtx.c
index c4e67f5..f8ac711 100644
--- a/drivers/net/dpaa/dpaa_rxtx.c
+++ b/drivers/net/dpaa/dpaa_rxtx.c
@@ -85,6 +85,121 @@
(_fd)->bpid = _bpid; \
} while (0)
 
+static inline void dpaa_slow_parsing(struct rte_mbuf *m __rte_unused,
+uint64_t prs __rte_unused)
+{
+   DPAA_DP_LOG(DEBUG, "Slow parsing");
+   /*TBD:XXX: to be implemented*/
+}
+
+static inline void dpaa_eth_packet_info(struct rte_mbuf *m,
+   uint64_t fd_virt_addr)
+{
+   struct annotations_t *annot = GET_ANNOTATIONS(fd_virt_addr);
+   uint64_t prs = *((uint64_t *)(&annot->parse)) & DPAA_PARSE_MASK;
+
+   DPAA_DP_LOG(DEBUG, " Parsing mbuf: %p with annotations: %p", m, annot);
+
+   switch (prs) {
+   case DPAA_PKT_TYPE_NONE:
+   m->packet_type = 0;
+   break;
+   case DPAA_PKT_TYPE_ETHER:
+   m->packet_type = RTE_PTYPE_L2_ETHER;
+   break;
+   case DPAA_PKT_TYPE_IPV4:
+   m->packet_type = RTE_PTYPE_L2_ETHER |
+   RTE_PTYPE_L3_IPV4;
+   break;
+   case DPAA_PKT_TYPE_IPV6:
+   m->packet_type = RTE_PTYPE_L2_ETHER |
+   RTE_PTYPE_L3_IPV6;
+   break;
+   case DPAA_PKT_TYPE_IPV4_FRAG:
+   case DPAA_PKT_TYPE_IPV4_FRAG_UDP:
+   case DPAA_PKT_TYPE_IPV4_FRAG_TCP:
+   case DPAA_PKT_TYPE_IPV4_FRAG_SCTP:
+   m->packet_type = RTE_PTYPE_L2_ETHER |
+   RTE_PTYPE_L3_IPV4 | RTE_PTYPE_L4_FRAG;
+   break;
+   case DPAA_PKT_TYPE_IPV6_FRAG:
+   case DPAA_PKT_TYPE_IPV6_FRAG_UDP:
+   case DPAA_PKT_TYPE_IPV6_FRAG_TCP:
+   case DPAA_PKT_TYPE_IPV6_FRAG_SCTP:
+   m->packet_type = RTE_PTYPE_L2_ETHER |
+   RTE_PTYPE_L3_IPV6 | RTE_PTYPE_L4_FRAG;
+   break;
+   case DPAA_PKT_TYPE_IPV4_EXT:
+   m->packet_type = RTE_PTYPE_L2_ETHER |
+   RTE_PTYPE_L3_IPV4_EXT;
+   break;
+   case DPAA_PKT_TYPE_IPV6_EXT:
+   m->pa

[dpdk-dev] [PATCH v5 33/40] net/dpaa: support flow control

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 doc/guides/nics/features/dpaa.ini |   1 +
 drivers/net/dpaa/dpaa_ethdev.c| 112 ++
 2 files changed, 113 insertions(+)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index c09efd8..1ba6b11 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -11,6 +11,7 @@ MTU update   = Y
 Promiscuous mode = Y
 Allmulticast mode= Y
 Unicast MAC filter   = Y
+Flow control = Y
 Basic stats  = Y
 ARMv8= Y
 Usage doc= Y
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index 23a9efd..ebceb8d 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -329,6 +329,85 @@ static int dpaa_link_up(struct rte_eth_dev *dev)
 }
 
 static int
+dpaa_flow_ctrl_set(struct rte_eth_dev *dev,
+  struct rte_eth_fc_conf *fc_conf)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+   struct rte_eth_fc_conf *net_fc;
+
+   PMD_INIT_FUNC_TRACE();
+
+   if (!(dpaa_intf->fc_conf)) {
+   dpaa_intf->fc_conf = rte_zmalloc(NULL,
+   sizeof(struct rte_eth_fc_conf), MAX_CACHELINE);
+   if (!dpaa_intf->fc_conf) {
+   DPAA_PMD_ERR("unable to save flow control info");
+   return -ENOMEM;
+   }
+   }
+   net_fc = dpaa_intf->fc_conf;
+
+   if (fc_conf->high_water < fc_conf->low_water) {
+   DPAA_PMD_ERR("Incorrect Flow Control Configuration");
+   return -EINVAL;
+   }
+
+   if (fc_conf->mode == RTE_FC_NONE) {
+   return 0;
+   } else if (fc_conf->mode == RTE_FC_TX_PAUSE ||
+fc_conf->mode == RTE_FC_FULL) {
+   fman_if_set_fc_threshold(dpaa_intf->fif, fc_conf->high_water,
+fc_conf->low_water,
+   dpaa_intf->bp_info->bpid);
+   if (fc_conf->pause_time)
+   fman_if_set_fc_quanta(dpaa_intf->fif,
+ fc_conf->pause_time);
+   }
+
+   /* Save the information in dpaa device */
+   net_fc->pause_time = fc_conf->pause_time;
+   net_fc->high_water = fc_conf->high_water;
+   net_fc->low_water = fc_conf->low_water;
+   net_fc->send_xon = fc_conf->send_xon;
+   net_fc->mac_ctrl_frame_fwd = fc_conf->mac_ctrl_frame_fwd;
+   net_fc->mode = fc_conf->mode;
+   net_fc->autoneg = fc_conf->autoneg;
+
+   return 0;
+}
+
+static int
+dpaa_flow_ctrl_get(struct rte_eth_dev *dev,
+  struct rte_eth_fc_conf *fc_conf)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+   struct rte_eth_fc_conf *net_fc = dpaa_intf->fc_conf;
+   int ret;
+
+   PMD_INIT_FUNC_TRACE();
+
+   if (net_fc) {
+   fc_conf->pause_time = net_fc->pause_time;
+   fc_conf->high_water = net_fc->high_water;
+   fc_conf->low_water = net_fc->low_water;
+   fc_conf->send_xon = net_fc->send_xon;
+   fc_conf->mac_ctrl_frame_fwd = net_fc->mac_ctrl_frame_fwd;
+   fc_conf->mode = net_fc->mode;
+   fc_conf->autoneg = net_fc->autoneg;
+   return 0;
+   }
+   ret = fman_if_get_fc_threshold(dpaa_intf->fif);
+   if (ret) {
+   fc_conf->mode = RTE_FC_TX_PAUSE;
+   fc_conf->pause_time = fman_if_get_fc_quanta(dpaa_intf->fif);
+   } else {
+   fc_conf->mode = RTE_FC_NONE;
+   }
+
+   return 0;
+}
+
+static int
 dpaa_dev_add_mac_addr(struct rte_eth_dev *dev,
 struct ether_addr *addr,
 uint32_t index,
@@ -384,6 +463,9 @@ static struct eth_dev_ops dpaa_devops = {
.rx_queue_release = dpaa_eth_rx_queue_release,
.tx_queue_release = dpaa_eth_tx_queue_release,
 
+   .flow_ctrl_get= dpaa_flow_ctrl_get,
+   .flow_ctrl_set= dpaa_flow_ctrl_set,
+
.link_update  = dpaa_eth_link_update,
.stats_get= dpaa_eth_stats_get,
.stats_reset  = dpaa_eth_stats_reset,
@@ -400,6 +482,33 @@ static struct eth_dev_ops dpaa_devops = {
 
 };
 
+static int dpaa_fc_set_default(struct dpaa_if *dpaa_intf)
+{
+   struct rte_eth_fc_conf *fc_conf;
+   int ret;
+
+   PMD_INIT_FUNC_TRACE();
+
+   if (!(dpaa_intf->fc_conf)) {
+   dpaa_intf->fc_conf = rte_zmalloc(NULL,
+   sizeof(struct rte_eth_fc_conf), MAX_CACHELINE);
+   if (!dpaa_intf->fc_conf) {
+   DPAA_PMD_ERR("unable to save flow control info");
+   return -ENOMEM;
+   }
+   }
+   fc_conf = dpaa_intf->fc_conf;
+   ret = fman_if_get_fc_thres

[dpdk-dev] [PATCH v5 37/40] net/dpaa: support Scattered Rx

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 doc/guides/nics/features/dpaa.ini |   1 +
 drivers/net/dpaa/dpaa_rxtx.c  | 159 ++
 drivers/net/dpaa/dpaa_rxtx.h  |   9 +++
 3 files changed, 169 insertions(+)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index 23626c0..0e7956c 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -8,6 +8,7 @@ Speed capabilities   = P
 Link status  = Y
 Jumbo frame  = Y
 MTU update   = Y
+Scattered Rx = Y
 Promiscuous mode = Y
 Allmulticast mode= Y
 Unicast MAC filter   = Y
diff --git a/drivers/net/dpaa/dpaa_rxtx.c b/drivers/net/dpaa/dpaa_rxtx.c
index 976268b..9c25d8c 100644
--- a/drivers/net/dpaa/dpaa_rxtx.c
+++ b/drivers/net/dpaa/dpaa_rxtx.c
@@ -276,18 +276,82 @@ static inline void dpaa_checksum_offload(struct rte_mbuf 
*mbuf,
fd->cmd = DPAA_FD_CMD_RPD | DPAA_FD_CMD_DTC;
 }
 
+struct rte_mbuf *
+dpaa_eth_sg_to_mbuf(struct qm_fd *fd, uint32_t ifid)
+{
+   struct dpaa_bp_info *bp_info = DPAA_BPID_TO_POOL_INFO(fd->bpid);
+   struct rte_mbuf *first_seg, *prev_seg, *cur_seg, *temp;
+   struct qm_sg_entry *sgt, *sg_temp;
+   void *vaddr, *sg_vaddr;
+   int i = 0;
+   uint8_t fd_offset = fd->offset;
+
+   DPAA_DP_LOG(DEBUG, "Received an SG frame");
+
+   vaddr = rte_dpaa_mem_ptov(qm_fd_addr(fd));
+   if (!vaddr) {
+   DPAA_PMD_ERR("unable to convert physical address");
+   return NULL;
+   }
+   sgt = vaddr + fd_offset;
+   sg_temp = &sgt[i++];
+   hw_sg_to_cpu(sg_temp);
+   temp = (struct rte_mbuf *)((char *)vaddr - bp_info->meta_data_size);
+   sg_vaddr = rte_dpaa_mem_ptov(qm_sg_entry_get64(sg_temp));
+
+   first_seg = (struct rte_mbuf *)((char *)sg_vaddr -
+   bp_info->meta_data_size);
+   first_seg->data_off = sg_temp->offset;
+   first_seg->data_len = sg_temp->length;
+   first_seg->pkt_len = sg_temp->length;
+   rte_mbuf_refcnt_set(first_seg, 1);
+
+   first_seg->port = ifid;
+   first_seg->nb_segs = 1;
+   first_seg->ol_flags = 0;
+   prev_seg = first_seg;
+   while (i < DPAA_SGT_MAX_ENTRIES) {
+   sg_temp = &sgt[i++];
+   hw_sg_to_cpu(sg_temp);
+   sg_vaddr = rte_dpaa_mem_ptov(qm_sg_entry_get64(sg_temp));
+   cur_seg = (struct rte_mbuf *)((char *)sg_vaddr -
+ bp_info->meta_data_size);
+   cur_seg->data_off = sg_temp->offset;
+   cur_seg->data_len = sg_temp->length;
+   first_seg->pkt_len += sg_temp->length;
+   first_seg->nb_segs += 1;
+   rte_mbuf_refcnt_set(cur_seg, 1);
+   prev_seg->next = cur_seg;
+   if (sg_temp->final) {
+   cur_seg->next = NULL;
+   break;
+   }
+   prev_seg = cur_seg;
+   }
+
+   dpaa_eth_packet_info(first_seg, (uint64_t)vaddr);
+   rte_pktmbuf_free_seg(temp);
+
+   return first_seg;
+}
+
 static inline struct rte_mbuf *dpaa_eth_fd_to_mbuf(struct qm_fd *fd,
uint32_t ifid)
 {
struct dpaa_bp_info *bp_info = DPAA_BPID_TO_POOL_INFO(fd->bpid);
struct rte_mbuf *mbuf;
void *ptr;
+   uint8_t format =
+   (fd->opaque & DPAA_FD_FORMAT_MASK) >> DPAA_FD_FORMAT_SHIFT;
uint16_t offset =
(fd->opaque & DPAA_FD_OFFSET_MASK) >> DPAA_FD_OFFSET_SHIFT;
uint32_t length = fd->opaque & DPAA_FD_LENGTH_MASK;
 
DPAA_DP_LOG(DEBUG, " FD--->MBUF");
 
+   if (unlikely(format == qm_fd_sg))
+   return dpaa_eth_sg_to_mbuf(fd, ifid);
+
/* Ignoring case when format != qm_fd_contig */
ptr = rte_dpaa_mem_ptov(fd->addr);
/* Ignoring case when ptr would be NULL. That is only possible incase
@@ -390,6 +454,95 @@ static struct rte_mbuf *dpaa_get_dmable_mbuf(struct 
rte_mbuf *mbuf,
return dpaa_mbuf;
 }
 
+int
+dpaa_eth_mbuf_to_sg_fd(struct rte_mbuf *mbuf,
+   struct qm_fd *fd,
+   uint32_t bpid)
+{
+   struct rte_mbuf *cur_seg = mbuf, *prev_seg = NULL;
+   struct dpaa_bp_info *bp_info = DPAA_BPID_TO_POOL_INFO(bpid);
+   struct rte_mbuf *temp, *mi;
+   struct qm_sg_entry *sg_temp, *sgt;
+   int i = 0;
+
+   DPAA_DP_LOG(DEBUG, "Creating SG FD to transmit");
+
+   temp = rte_pktmbuf_alloc(bp_info->mp);
+   if (!temp) {
+   DPAA_PMD_ERR("Failure in allocation of mbuf");
+   return -1;
+   }
+   if (temp->buf_len < ((mbuf->nb_segs * sizeof(struct qm_sg_entry))
+   + temp->data_off)) {
+   DPAA_PMD_ERR("Insufficient space in mbuf for SG entries");
+   return -1;
+   }

[dpdk-dev] [PATCH v5 34/40] net/dpaa: support hashed RSS

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/net/dpaa/dpaa_ethdev.c |  1 +
 drivers/net/dpaa/dpaa_ethdev.h | 10 ++
 2 files changed, 11 insertions(+)

diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index ebceb8d..5860dfa 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -157,6 +157,7 @@ static void dpaa_eth_dev_info(struct rte_eth_dev *dev,
dev_info->max_hash_mac_addrs = 0;
dev_info->max_vfs = 0;
dev_info->max_vmdq_pools = ETH_16_POOLS;
+   dev_info->flow_type_rss_offloads = DPAA_RSS_OFFLOAD_ALL;
dev_info->speed_capa = (ETH_LINK_SPEED_1G |
ETH_LINK_SPEED_10G);
 }
diff --git a/drivers/net/dpaa/dpaa_ethdev.h b/drivers/net/dpaa/dpaa_ethdev.h
index 2f25acb..e1e062e 100644
--- a/drivers/net/dpaa/dpaa_ethdev.h
+++ b/drivers/net/dpaa/dpaa_ethdev.h
@@ -88,6 +88,16 @@
 #define DPAA_DEBUG_FQ_RX_ERROR   0
 #define DPAA_DEBUG_FQ_TX_ERROR   1
 
+#define DPAA_RSS_OFFLOAD_ALL ( \
+   ETH_RSS_FRAG_IPV4 | \
+   ETH_RSS_NONFRAG_IPV4_TCP | \
+   ETH_RSS_NONFRAG_IPV4_UDP | \
+   ETH_RSS_NONFRAG_IPV4_SCTP | \
+   ETH_RSS_FRAG_IPV6 | \
+   ETH_RSS_NONFRAG_IPV6_TCP | \
+   ETH_RSS_NONFRAG_IPV6_UDP | \
+   ETH_RSS_NONFRAG_IPV6_SCTP)
+
 #define DPAA_TX_CKSUM_OFFLOAD_MASK ( \
PKT_TX_IP_CKSUM |\
PKT_TX_TCP_CKSUM |   \
-- 
2.9.3

[dpdk-dev] [PATCH v5 40/40] net/dpaa: support extended statistics

[Shreyansh Jain]

From: Hemant Agrawal 

Signed-off-by: Hemant Agrawal 
---
 doc/guides/nics/features/dpaa.ini |   1 +
 drivers/net/dpaa/dpaa_ethdev.c| 143 ++
 drivers/net/dpaa/dpaa_ethdev.h|  40 +++
 3 files changed, 184 insertions(+)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index 09b9bd9..24cfd85 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -18,6 +18,7 @@ L3 checksum offload  = Y
 L4 checksum offload  = Y
 Packet type parsing  = Y
 Basic stats  = Y
+Extended stats   = Y
 FW version   = Y
 ARMv8= Y
 Usage doc= Y
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index 0002324..db4921f 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -75,6 +75,40 @@
 /* Keep track of whether QMAN and BMAN have been globally initialized */
 static int is_global_init;
 
+struct rte_dpaa_xstats_name_off {
+   char name[RTE_ETH_XSTATS_NAME_SIZE];
+   uint32_t offset;
+};
+
+static const struct rte_dpaa_xstats_name_off dpaa_xstats_strings[] = {
+   {"rx_align_err",
+   offsetof(struct dpaa_if_stats, raln)},
+   {"rx_valid_pause",
+   offsetof(struct dpaa_if_stats, rxpf)},
+   {"rx_fcs_err",
+   offsetof(struct dpaa_if_stats, rfcs)},
+   {"rx_vlan_frame",
+   offsetof(struct dpaa_if_stats, rvlan)},
+   {"rx_frame_err",
+   offsetof(struct dpaa_if_stats, rerr)},
+   {"rx_drop_err",
+   offsetof(struct dpaa_if_stats, rdrp)},
+   {"rx_undersized",
+   offsetof(struct dpaa_if_stats, rund)},
+   {"rx_oversize_err",
+   offsetof(struct dpaa_if_stats, rovr)},
+   {"rx_fragment_pkt",
+   offsetof(struct dpaa_if_stats, rfrg)},
+   {"tx_valid_pause",
+   offsetof(struct dpaa_if_stats, txpf)},
+   {"tx_fcs_err",
+   offsetof(struct dpaa_if_stats, terr)},
+   {"tx_vlan_frame",
+   offsetof(struct dpaa_if_stats, tvlan)},
+   {"rx_undersized",
+   offsetof(struct dpaa_if_stats, tund)},
+};
+
 static int
 dpaa_mtu_set(struct rte_eth_dev *dev, uint16_t mtu)
 {
@@ -268,6 +302,110 @@ static void dpaa_eth_stats_reset(struct rte_eth_dev *dev)
fman_if_stats_reset(dpaa_intf->fif);
 }
 
+static int
+dpaa_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats,
+   unsigned int n)
+{
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+   unsigned int i = 0, num = RTE_DIM(dpaa_xstats_strings);
+   uint64_t values[sizeof(struct dpaa_if_stats) / 8];
+
+   if (xstats == NULL)
+   return 0;
+
+   if (n < num)
+   return num;
+
+   fman_if_stats_get_all(dpaa_intf->fif, values,
+ sizeof(struct dpaa_if_stats) / 8);
+
+   for (i = 0; i < num; i++) {
+   xstats[i].id = i;
+   xstats[i].value = values[dpaa_xstats_strings[i].offset / 8];
+   }
+   return i;
+}
+
+static int
+dpaa_xstats_get_names(__rte_unused struct rte_eth_dev *dev,
+ struct rte_eth_xstat_name *xstats_names,
+ __rte_unused unsigned int limit)
+{
+   unsigned int i, stat_cnt = RTE_DIM(dpaa_xstats_strings);
+
+   if (xstats_names != NULL)
+   for (i = 0; i < stat_cnt; i++)
+   snprintf(xstats_names[i].name,
+sizeof(xstats_names[i].name),
+"%s",
+dpaa_xstats_strings[i].name);
+
+   return stat_cnt;
+}
+
+static int
+dpaa_xstats_get_by_id(struct rte_eth_dev *dev, const uint64_t *ids,
+ uint64_t *values, unsigned int n)
+{
+   unsigned int i, stat_cnt = RTE_DIM(dpaa_xstats_strings);
+   uint64_t values_copy[sizeof(struct dpaa_if_stats) / 8];
+
+   if (!ids) {
+   struct dpaa_if *dpaa_intf = dev->data->dev_private;
+
+   if (n < stat_cnt)
+   return stat_cnt;
+
+   if (!values)
+   return 0;
+
+   fman_if_stats_get_all(dpaa_intf->fif, values_copy,
+ sizeof(struct dpaa_if_stats));
+
+   for (i = 0; i < stat_cnt; i++)
+   values[i] =
+   values_copy[dpaa_xstats_strings[i].offset / 8];
+
+   return stat_cnt;
+   }
+
+   dpaa_xstats_get_by_id(dev, NULL, values_copy, stat_cnt);
+
+   for (i = 0; i < n; i++) {
+   if (ids[i] >= stat_cnt) {
+   DPAA_PMD_ERR("id value isn't valid");
+   return -1;
+   }
+   values[i] = values_copy[ids[i]];
+   }
+   return n;
+}
+
+static int
+dpaa_xstats_get_names_by_id(
+   struct rte_e

[dpdk-dev] [PATCH v5 38/40] net/dpaa: add packet dump for debugging

[Shreyansh Jain]

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/net/dpaa/dpaa_ethdev.c | 42 ++
 drivers/net/dpaa/dpaa_rxtx.c   | 26 ++
 2 files changed, 68 insertions(+)

diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index 4eb9b62..ad9fea8 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -618,6 +618,39 @@ static int dpaa_tx_queue_init(struct qman_fq *fq,
return ret;
 }
 
+#ifdef RTE_LIBRTE_DPAA_DEBUG_DRIVER
+/* Initialise a DEBUG FQ ([rt]x_error, rx_default). */
+static int dpaa_debug_queue_init(struct qman_fq *fq, uint32_t fqid)
+{
+   struct qm_mcc_initfq opts;
+   int ret;
+
+   PMD_INIT_FUNC_TRACE();
+
+   ret = qman_reserve_fqid(fqid);
+   if (ret) {
+   DPAA_PMD_ERR("Reserve debug fqid %d failed with ret: %d",
+   fqid, ret);
+   return -EINVAL;
+   }
+   /* "map" this Rx FQ to one of the interfaces Tx FQID */
+   DPAA_PMD_DEBUG("Creating debug fq %p, fqid %d", fq, fqid);
+   ret = qman_create_fq(fqid, QMAN_FQ_FLAG_NO_ENQUEUE, fq);
+   if (ret) {
+   DPAA_PMD_ERR("create debug fqid %d failed with ret: %d",
+   fqid, ret);
+   return ret;
+   }
+   opts.we_mask = QM_INITFQ_WE_DESTWQ | QM_INITFQ_WE_FQCTRL;
+   opts.fqd.dest.wq = DPAA_IF_DEBUG_PRIORITY;
+   ret = qman_init_fq(fq, 0, &opts);
+   if (ret)
+   DPAA_PMD_ERR("init debug fqid %d failed with ret: %d",
+   fqid, ret);
+   return ret;
+}
+#endif
+
 /* Initialise a network interface */
 static int
 dpaa_dev_init(struct rte_eth_dev *eth_dev)
@@ -692,6 +725,15 @@ dpaa_dev_init(struct rte_eth_dev *eth_dev)
}
dpaa_intf->nb_tx_queues = num_cores;
 
+#ifdef RTE_LIBRTE_DPAA_DEBUG_DRIVER
+   dpaa_debug_queue_init(&dpaa_intf->debug_queues[
+   DPAA_DEBUG_FQ_RX_ERROR], fman_intf->fqid_rx_err);
+   dpaa_intf->debug_queues[DPAA_DEBUG_FQ_RX_ERROR].dpaa_intf = dpaa_intf;
+   dpaa_debug_queue_init(&dpaa_intf->debug_queues[
+   DPAA_DEBUG_FQ_TX_ERROR], fman_intf->fqid_tx_err);
+   dpaa_intf->debug_queues[DPAA_DEBUG_FQ_TX_ERROR].dpaa_intf = dpaa_intf;
+#endif
+
DPAA_PMD_DEBUG("All frame queues created");
 
/* Get the initial configuration for flow control */
diff --git a/drivers/net/dpaa/dpaa_rxtx.c b/drivers/net/dpaa/dpaa_rxtx.c
index 9c25d8c..d73f9cb 100644
--- a/drivers/net/dpaa/dpaa_rxtx.c
+++ b/drivers/net/dpaa/dpaa_rxtx.c
@@ -85,6 +85,31 @@
(_fd)->bpid = _bpid; \
} while (0)
 
+#if (defined RTE_LIBRTE_DPAA_DEBUG_DRIVER)
+void dpaa_display_frame(const struct qm_fd *fd)
+{
+   int ii;
+   char *ptr;
+
+   printf("%s::bpid %x addr %08x%08x, format %d off %d, len %d stat %x\n",
+  __func__, fd->bpid, fd->addr_hi, fd->addr_lo, fd->format,
+   fd->offset, fd->length20, fd->status);
+
+   ptr = (char *)rte_dpaa_mem_ptov(fd->addr);
+   ptr += fd->offset;
+   printf("%02x ", *ptr);
+   for (ii = 1; ii < fd->length20; ii++) {
+   printf("%02x ", *ptr);
+   if ((ii % 16) == 0)
+   printf("\n");
+   ptr++;
+   }
+   printf("\n");
+}
+#else
+#define dpaa_display_frame(a)
+#endif
+
 static inline void dpaa_slow_parsing(struct rte_mbuf *m __rte_unused,
 uint64_t prs __rte_unused)
 {
@@ -353,6 +378,7 @@ static inline struct rte_mbuf *dpaa_eth_fd_to_mbuf(struct 
qm_fd *fd,
return dpaa_eth_sg_to_mbuf(fd, ifid);
 
/* Ignoring case when format != qm_fd_contig */
+   dpaa_display_frame(fd);
ptr = rte_dpaa_mem_ptov(fd->addr);
/* Ignoring case when ptr would be NULL. That is only possible incase
 * of a corrupted packet
-- 
2.9.3

[dpdk-dev] [PATCH v5 39/40] net/dpaa: support firmware version get API

[Shreyansh Jain]

From: Hemant Agrawal 

Signed-off-by: Hemant Agrawal 
---
 doc/guides/nics/features/dpaa.ini |  1 +
 drivers/net/dpaa/dpaa_ethdev.c| 36 
 drivers/net/dpaa/dpaa_ethdev.h|  5 +
 3 files changed, 42 insertions(+)

diff --git a/doc/guides/nics/features/dpaa.ini 
b/doc/guides/nics/features/dpaa.ini
index 0e7956c..09b9bd9 100644
--- a/doc/guides/nics/features/dpaa.ini
+++ b/doc/guides/nics/features/dpaa.ini
@@ -18,5 +18,6 @@ L3 checksum offload  = Y
 L4 checksum offload  = Y
 Packet type parsing  = Y
 Basic stats  = Y
+FW version   = Y
 ARMv8= Y
 Usage doc= Y
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index ad9fea8..0002324 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -164,6 +164,41 @@ static void dpaa_eth_dev_close(struct rte_eth_dev *dev)
dpaa_eth_dev_stop(dev);
 }
 
+static int
+dpaa_fw_version_get(struct rte_eth_dev *dev __rte_unused,
+char *fw_version,
+size_t fw_size)
+{
+   int ret;
+   FILE *svr_file = NULL;
+   unsigned int svr_ver = 0;
+
+   PMD_INIT_FUNC_TRACE();
+
+   svr_file = fopen(DPAA_SOC_ID_FILE, "r");
+   if (!svr_file) {
+   DPAA_PMD_ERR("Unable to open SoC device");
+   return -ENOTSUP; /* Not supported on this infra */
+   }
+
+   ret = fscanf(svr_file, "svr:%x", &svr_ver);
+   if (ret <= 0) {
+   DPAA_PMD_ERR("Unable to read SoC device");
+   return -ENOTSUP; /* Not supported on this infra */
+   }
+
+   ret = snprintf(fw_version, fw_size,
+  "svr:%x-fman-v%x",
+  svr_ver,
+  fman_ip_rev);
+
+   ret += 1; /* add the size of '\0' */
+   if (fw_size < (uint32_t)ret)
+   return ret;
+   else
+   return 0;
+}
+
 static void dpaa_eth_dev_info(struct rte_eth_dev *dev,
  struct rte_eth_dev_info *dev_info)
 {
@@ -512,6 +547,7 @@ static struct eth_dev_ops dpaa_devops = {
.mac_addr_remove  = dpaa_dev_remove_mac_addr,
.mac_addr_set = dpaa_dev_set_mac_addr,
 
+   .fw_version_get   = dpaa_fw_version_get,
 };
 
 static int dpaa_fc_set_default(struct dpaa_if *dpaa_intf)
diff --git a/drivers/net/dpaa/dpaa_ethdev.h b/drivers/net/dpaa/dpaa_ethdev.h
index e1e062e..a980262 100644
--- a/drivers/net/dpaa/dpaa_ethdev.h
+++ b/drivers/net/dpaa/dpaa_ethdev.h
@@ -43,6 +43,11 @@
 #include 
 #include 
 
+/* DPAA SoC identifier; If this is not available, it can be concluded
+ * that board is non-DPAA. Single slot is currently supported.
+ */
+#define DPAA_SOC_ID_FILE   "sys/devices/soc0/soc_id"
+
 #define DPAA_MBUF_HW_ANNOTATION64
 #define DPAA_FD_PTA_SIZE   64
 
-- 
2.9.3

Re: [dpdk-dev] [PATCH v4 11/41] bus/dpaa: add QMan driver core routines

[Shreyansh Jain]

On Tuesday 19 September 2017 07:48 PM, Shreyansh Jain wrote:

On Monday 18 September 2017 08:23 PM, Ferruh Yigit wrote:

On 9/9/2017 12:21 PM, Shreyansh Jain wrote:

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 


<...>

+#ifdef RTE_LIBRTE_DPAA_CHECKING


This is not defined anywhere, it looks this will come from config file
in further patches, config file update can be moved to this patch.


Its more of a debugging macro and it was introduced in later patches.
Not that I see any reason why it can't be introduced here. I will fix this.




+    eqcr->busy = 0;
+    eqcr->pmode = pmode;
+#endif




In the v5, taking cue from yours and Thomas' comments, I have removed a 
few debugging macros and combined a couple. I have also changed this to 
"HWDEBUG" and documented this. This macro is necessary for enabling some 
internally debugging configurations (like error queues).

Re: [dpdk-dev] [PATCH v4 13/41] bus/dpaa: add support for FMAN frame queue lookup

[Shreyansh Jain]

On Monday 18 September 2017 08:21 PM, Ferruh Yigit wrote:

On 9/9/2017 12:21 PM, Shreyansh Jain wrote:

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 


<...>

  
+#if !defined(CONFIG_FSL_QMAN_FQ_LOOKUP) && defined(RTE_ARCH_ARM64)

+#error "_ARM64 requires _FSL_QMAN_FQ_LOOKUP"
+#endif


This PMD enabled with new added config
"defconfig_arm64-armv8a-linuxapp-gcc", which is 64bits. So this means
CONFIG_FSL_QMAN_FQ_LOOKUP always defined for the bus.

Does is make sense to keep above check, but for rest of the code assume
CONFIG_FSL_QMAN_FQ_LOOKUP always set and remove the #ifdefs, to simplify
the code?

<...>



I have removed these lines in v5. They were indeed unnecessary.
Thanks for highlighting.

Re: [dpdk-dev] [PATCH v4 17/41] doc: add NXP DPAA PMD documentation

[Shreyansh Jain]

On Tuesday 19 September 2017 07:55 PM, Shreyansh Jain wrote:

On Monday 18 September 2017 08:23 PM, Ferruh Yigit wrote:

On 9/9/2017 12:21 PM, Shreyansh Jain wrote:

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 




[...]

+  This is not a DPAA specific configuration - it is a generic RTE 
config.
+  For optimal performance and hardware utilization, it is expected 
that DPAA
+  Mempool driver is used for mempools. For that, this configuration 
needs to

+  enabled.
+
+Environment Variables
+~
+
+DPAA drivers uses the following environment variables to configure its
+state during application initialization:
+
+- ``DPAA_NUM_RX_QUEUES`` (default 1)


Why not getting this value as device arg?


We had this discussion during DPAA2 as well. This time, I was not sure 
of how the device argument patches are turning out to be after the 
re-shuffle being done by Gaetan. So, I kept this as it is.


In the v5, I have continued with the old way.
I am still reviewing the devargs patches from Gaetan - once that is 
done, probably after 1711, I will push another patch for removing this 
environment variable.


In fact, in an internal code change, we have removed need to rely on 
this. but, that is slightly ahead in future.

Re: [dpdk-dev] [PATCH v2 0/4] add net mrvl pmd driver

[Amit Tomer]

>> This patch series introduces the net driver for Marvell Armada 7k/8k
>> SoCs along with documentation.

Wondering, if it could be tested on Marvell Armada 3700 based
ESPRESSObin board ?

Thanks
Amit

Re: [dpdk-dev] [PATCH v4 25/41] net/dpaa: add support for Tx and Rx queue setup

[Shreyansh Jain]

On Thursday 21 September 2017 06:29 PM, Shreyansh Jain wrote:

Hello Ferruh,

Apologies for delay in response for these, I am already working to get 
the next version based on your comments. Meanwhile, some comments inline...


On Monday 18 September 2017 08:25 PM, Ferruh Yigit wrote:

On 9/9/2017 12:21 PM, Shreyansh Jain wrote:

Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 


<...>


[...]




+    }
+
+    /* Populate ethdev structure */
  eth_dev->dev_ops = &dpaa_devops;
+    eth_dev->rx_pkt_burst = dpaa_eth_queue_rx;
+    eth_dev->tx_pkt_burst = dpaa_eth_tx_drop_all;
+
+    /* Allocate memory for storing MAC addresses */
+    eth_dev->data->mac_addrs = rte_zmalloc("mac_addr",
+    ETHER_ADDR_LEN * DPAA_MAX_MAC_FILTER, 0);
+    if (eth_dev->data->mac_addrs == NULL) {
+    DPAA_PMD_ERR("Failed to allocate %d bytes needed to "
+    "store MAC addresses",
+    ETHER_ADDR_LEN * DPAA_MAX_MAC_FILTER);


free dpaa_intf->rx_queues, tx_queues ?


yes, certainly an issue. I will fix it.


I have fixed this in v5.






+    return -ENOMEM;
+    }
+
+    /* copy the primary mac address */
+    memcpy(eth_dev->data->mac_addrs[0].addr_bytes,
+    fman_intf->mac_addr.addr_bytes,
+    ETHER_ADDR_LEN);


Instead can use ether_addr_copy() instead.


:) Yes, I can.


Unfortunately, I forgot to fix this in v5.
If you want, I can send a small patch against this. Sending a v6 because 
of this would be overkill. But, this is definitely a valid comment.

Sorry.





<...>

Re: [dpdk-dev] How to use ixgbe module log?

[Burakov, Anatoly]

On 28-Sep-17 11:26 AM, Sam wrote:

Hi all,

I'm working on ixgbe, I want to enable "PMD_RX_LOG" and "PMD_TX_LOG", how
to do  this?

As for normal application, I could use --log-level param to start, but
ixgbe is kernel module, how to do this?

Thank you~



Hi Sam,

In DPDK, ixgbe is not a kernel module - these are customized, userspace 
drivers. In order to enable RX and TX logging, corresponding config 
options should be set at compile time. For example, if you want to 
enable RX/TX logging for IXGBE driver, here's a rough list of steps you 
need to do:


1) create a custom config file in config/ directory. Copy whatever it is 
that you're using for your application (e.g. 
defconfig_x86_64-native-linuxapp-gcc), and name it something starting 
with defconfig_ (e.g. defconfig_myconfig)
2) Set CONFIG_RTE_LIBRTE_IXGBE_DEBUG_RX and 
CONFIG_RTE_LIBRTE_IXGBE_DEBUG_TX (and whatever other options you may 
need, see [1] and [2]) to "y" in your custom config
3) do "make config T=myconfig && make" (or whatever other command you 
use to build/install DPDK - make sure to specify your custom config as 
target)


After that, you should be able to see RX/TX logs in DPDK applications.

[1] http://dpdk.org/browse/dpdk/tree/config/common_base#n171
[2] http://dpdk.org/browse/dpdk/tree/drivers/net/ixgbe/ixgbe_logs.h

--
Thanks,
Anatoly

Re: [dpdk-dev] [PATCH v2 0/4] add net mrvl pmd driver

[Marcin Wojtas]

Hi Amit,

2017-09-28 13:40 GMT+02:00 Amit Tomer :
>
> >> This patch series introduces the net driver for Marvell Armada 7k/8k
> >> SoCs along with documentation.
>
> Wondering, if it could be tested on Marvell Armada 3700 based
> ESPRESSObin board ?
>

Different SoC. This one is intended to work with Armada 8040 MacchiatoBin board.

Best regards,
Marcin

[dpdk-dev] [PATCH v6 00/40] Introduce NXP DPAA Bus, Mempool and PMD

[Shreyansh Jain]

Change Log:


v6:
 - rebased over net-next/master (9d660ac) 
 - fixed mk/rte.app.mk (Thomas's comment). It had incorrect
   style of adding library linking
 - changed from manual memcpy of etheraddr to ether_addr_copy
   as suggested by Ferruh
 (these were minor changes missed in v5)

 v5:
 - rebased over net-next/master (9d660ac)   
 - restructuring debugging macros. Removed a few and combined
   others. DPAA now reflects the dynamic logging with segragated
   DP logging
 - updated documentation for missing configuration option
 - fixed map file; shared build was broken earlier
 - other minor fixes from review comments

v4:
 - Some checkpatch fixes which were reported by checkpatch@dpdk
 - adding extra stats feature patch (patch 41)

v3:
 - Rebasing over 17.11-rc0 (85238f50)
 - Checkpatch fixes
   (There are still 2 errors which I think are false positives)
 - Implement rte_bus.find_device() interface
 - Various other minor updates/cleanups

v2:
 - Fixing various comments from Ferruh, but broadly:
  -) Logging is been changed to reflect rte_log_register
  -) Logs across Bus, Mempool and PMD updated
  -) fixed incorrect feature claimed in dpaa.ini
 - Removed 24/40/48 bit swapping macro from EAL.
   These are defined in dpaa/bus now (compat.h)
 - Added missing memory cleanup operation
 - Updated documentation with some missing information

Introduction


RFC was posted here -> [R3]
V5 was posted here  -> [R8]

This patch series adds NXP's QorIQ-Layerscape DPAA Architecture based
bus driver, mempool driver and PMD. This version of driver supports NXP
LS1043A/LS1023A, LS1046A/LS1026A family of network SoCs. [R1]

DPAA, or Datapath Acceleration Architecture [R2], is a set of hardware
components designed for high-speed network packet processing. This
architecture provides the infrastructure to support simplified sharing of
networking interfaces and accelerators by multiple CPU cores, and the
accelerators themselves.

This patchset introduces the following:
1. DPAA Bus (drivers/bus/dpaa)
 The core of DPAA bus is implemented using 3 main hardware blocks: QMan,
 or Queue Manager; BMan, or Buffer Manager and FMan, or Frame Manager.
 The patches introduce necessary layers to expose the DPAA hardware
 blocks for interfacing with RTE framework.

2. DPAA Mempool (drivers/mempool/dpaa)
 BMan, or Buffer Manager, block of DPAA features a hardware offloaded
 mempool. These patches add support for a driver to manage the BMan
 block. This driver allows for mempool creation, deletion, buffer
 acquire and release, as per the RTE APIs.

3. DPAA PMD (drivers/net/dpaa)
 The Poll Mode Driver for DPAA NIC Interfaces.

Patch Layout


01: Add DPAA SoC build configuration
02~16: Add DPAA Bus support and features, incrementally
17: Add Documentation
18~21: Add DPAA Mempool support
22~40: Add PMD and its various features, incrementally

References
==

[R1] 
http://www.nxp.com/products/microcontrollers-and-processors/arm-processors/qoriq-layerscape-arm-processors:QORIQ-ARM
[R2] http://www.nxp.com/assets/documents/data/en/white-papers/QORIQDPAAWP.pdf
[R3] RFC: http://dpdk.org/ml/archives/dev/2017-May/066675.html
[R4] v1: http://dpdk.org/ml/archives/dev/2017-June/068020.html
[R5] v2: http://dpdk.org/ml/archives/dev/2017-July/070113.html
[R6] v3: http://dpdk.org/ml/archives/dev/2017-August/073269.html
[R7] v4: http://dpdk.org/ml/archives/dev/2017-September/074936.html
[R8] v5: http://dpdk.org/dev/patchwork/patch/29245/

Hemant Agrawal (3):
  bus/dpaa: add compatibility and helper macros
  net/dpaa: support firmware version get API
  net/dpaa: support extended statistics

Shreyansh Jain (37):
  config: add NXP DPAA SoC build configuration
  bus/dpaa: introduce NXP DPAA Bus driver skeleton
  bus/dpaa: add OF parser for device scanning
  bus/dpaa: introducing FMan configurations
  bus/dpaa: add FMan hardware operations
  bus/dpaa: enable DPAA IOCTL portal driver
  bus/dpaa: add layer for interrupt emulation using pthread
  bus/dpaa: add routines for managing a RB tree
  bus/dpaa: add QMAN interface driver
  bus/dpaa: add QMan driver core routines
  bus/dpaa: add BMAN driver core
  bus/dpaa: support FMAN frame queue lookup
  bus/dpaa: add BMan hardware interfaces
  bus/dpaa: add fman flow control threshold setting
  bus/dpaa: integrate DPAA Bus with hardware blocks
  doc: add NXP DPAA PMD documentation
  bus/dpaa: add DPAA mempool logging macros
  mempool/dpaa: support NXP DPAA Mempool
  config: enable compilation of DPAA Mempool driver
  bus/dpaa: add DPAA PMD logging macros
  net/dpaa: add NXP DPAA PMD driver skeleton
  config: enable NXP DPAA PMD compilation
  net/dpaa: support Tx and Rx queue setup
  net/dpaa: support MTU update
  net/dpaa: support jumbo frames
  net/dpaa: support link status update
  net/dpaa: support device info and speed capability
  net/dpaa: support promiscuous toggle
  net/dpaa: support multicast toggle
  net/dpaa: support MAC address update
  net/dpaa: support basic sta

[dpdk-dev] [PATCH v6 01/40] config: add NXP DPAA SoC build configuration

[Shreyansh Jain]

This patch adds skeleton build configuration for DPAA platform.

Signed-off-by: Shreyansh Jain 
---
 config/defconfig_arm64-dpaa-linuxapp-gcc | 47 
 mk/machine/dpaa/rte.vars.mk  | 61 
 2 files changed, 108 insertions(+)
 create mode 100644 config/defconfig_arm64-dpaa-linuxapp-gcc
 create mode 100644 mk/machine/dpaa/rte.vars.mk

diff --git a/config/defconfig_arm64-dpaa-linuxapp-gcc 
b/config/defconfig_arm64-dpaa-linuxapp-gcc
new file mode 100644
index 000..5bca012
--- /dev/null
+++ b/config/defconfig_arm64-dpaa-linuxapp-gcc
@@ -0,0 +1,47 @@
+#   BSD LICENSE
+#
+#   Copyright 2016 Freescale Semiconductor, Inc.
+#   Copyright 2017 NXP.
+#
+#   Redistribution and use in source and binary forms, with or without
+#   modification, are permitted provided that the following conditions
+#   are met:
+#
+# * Redistributions of source code must retain the above copyright
+#   notice, this list of conditions and the following disclaimer.
+# * Redistributions in binary form must reproduce the above copyright
+#   notice, this list of conditions and the following disclaimer in
+#   the documentation and/or other materials provided with the
+#   distribution.
+# * Neither the name of NXP nor the names of its
+#   contributors may be used to endorse or promote products derived
+#   from this software without specific prior written permission.
+#
+#   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+#   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+#   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+#   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+#   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+#   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+#   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+#   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+#   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+#   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+#   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+#
+
+#include "defconfig_arm64-armv8a-linuxapp-gcc"
+
+# NXP (Freescale) - Soc Architecture with FMAN, QMAN & BMAN support
+CONFIG_RTE_MACHINE="dpaa"
+CONFIG_RTE_ARCH_ARM_TUNE="cortex-a72"
+CONFIG_RTE_LIBRTE_VHOST_NUMA=n
+CONFIG_RTE_EAL_NUMA_AWARE_HUGEPAGES=n
+
+#
+# Compile Environment Abstraction Layer
+#
+CONFIG_RTE_MAX_LCORE=4
+CONFIG_RTE_MAX_NUMA_NODES=1
+CONFIG_RTE_CACHE_LINE_SIZE=64
+CONFIG_RTE_PKTMBUF_HEADROOM=128
diff --git a/mk/machine/dpaa/rte.vars.mk b/mk/machine/dpaa/rte.vars.mk
new file mode 100644
index 000..356a6af
--- /dev/null
+++ b/mk/machine/dpaa/rte.vars.mk
@@ -0,0 +1,61 @@
+#   BSD LICENSE
+#
+#   Copyright (c) 2016 Freescale Semiconductor, Inc. All rights reserved.
+#   Copyright 2017 NXP.
+#
+#   Redistribution and use in source and binary forms, with or without
+#   modification, are permitted provided that the following conditions
+#   are met:
+#
+# * Redistributions of source code must retain the above copyright
+#   notice, this list of conditions and the following disclaimer.
+# * Redistributions in binary form must reproduce the above copyright
+#   notice, this list of conditions and the following disclaimer in
+#   the documentation and/or other materials provided with the
+#   distribution.
+# * Neither the name of NXP nor the names of its
+#   contributors may be used to endorse or promote products derived
+#   from this software without specific prior written permission.
+#
+#   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+#   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+#   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+#   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+#   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+#   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+#   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+#   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+#   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+#   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+#   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+#
+# machine:
+#
+#   - can define ARCH variable (overridden by cmdline value)
+#   - can define CROSS variable (overridden by cmdline value)
+#   - define MACHINE_CFLAGS variable (overridden by cmdline value)
+#   - define MACHINE_LDFLAGS variable (overridden by cmdline value)
+#   - define MACHINE_ASFLAGS variable (overridden by cmdline value)
+#   - can define CPU_CFLAGS variable (overridden by cmdline value) that
+# overrides the 

[dpdk-dev] [PATCH v6 02/40] bus/dpaa: introduce NXP DPAA Bus driver skeleton

[Shreyansh Jain]

Signed-off-by: Shreyansh Jain 
Signed-off-by: Hemant Agrawal 
---
 MAINTAINERS   |   5 +
 config/common_base|   3 +
 config/defconfig_arm64-dpaa-linuxapp-gcc  |   4 +
 drivers/bus/Makefile  |   3 +
 drivers/bus/dpaa/Makefile |  58 +
 drivers/bus/dpaa/dpaa_bus.c   | 207 ++
 drivers/bus/dpaa/rte_bus_dpaa_version.map |   8 ++
 drivers/bus/dpaa/rte_dpaa_bus.h   | 148 +
 drivers/bus/dpaa/rte_dpaa_logs.h  |  65 ++
 9 files changed, 501 insertions(+)
 create mode 100644 drivers/bus/dpaa/Makefile
 create mode 100644 drivers/bus/dpaa/dpaa_bus.c
 create mode 100644 drivers/bus/dpaa/rte_bus_dpaa_version.map
 create mode 100644 drivers/bus/dpaa/rte_dpaa_bus.h
 create mode 100644 drivers/bus/dpaa/rte_dpaa_logs.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 8df2a7f..c566962 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -408,6 +408,11 @@ F: drivers/net/nfp/
 F: doc/guides/nics/nfp.rst
 F: doc/guides/nics/features/nfp.ini
 
+NXP dpaa
+M: Hemant Agrawal 
+M: Shreyansh Jain 
+F: drivers/bus/dpaa/
+
 NXP dpaa2
 M: Hemant Agrawal 
 M: Shreyansh Jain 
diff --git a/config/common_base b/config/common_base
index 439f3cc..fc1cdca 100644
--- a/config/common_base
+++ b/config/common_base
@@ -301,6 +301,9 @@ CONFIG_RTE_LIBRTE_LIO_DEBUG_TX=n
 CONFIG_RTE_LIBRTE_LIO_DEBUG_MBOX=n
 CONFIG_RTE_LIBRTE_LIO_DEBUG_REGS=n
 
+# NXP DPAA Bus
+CONFIG_RTE_LIBRTE_DPAA_BUS=n
+
 #
 # Compile NXP DPAA2 FSL-MC Bus
 #
diff --git a/config/defconfig_arm64-dpaa-linuxapp-gcc 
b/config/defconfig_arm64-dpaa-linuxapp-gcc
index 5bca012..8316fc9 100644
--- a/config/defconfig_arm64-dpaa-linuxapp-gcc
+++ b/config/defconfig_arm64-dpaa-linuxapp-gcc
@@ -45,3 +45,7 @@ CONFIG_RTE_MAX_LCORE=4
 CONFIG_RTE_MAX_NUMA_NODES=1
 CONFIG_RTE_CACHE_LINE_SIZE=64
 CONFIG_RTE_PKTMBUF_HEADROOM=128
+
+# NXP DPAA Bus
+CONFIG_RTE_LIBRTE_DPAA_BUS=y
+CONFIG_RTE_LIBRTE_DPAA_DEBUG_DRIVER=n
diff --git a/drivers/bus/Makefile b/drivers/bus/Makefile
index 0224214..6cb6466 100644
--- a/drivers/bus/Makefile
+++ b/drivers/bus/Makefile
@@ -32,6 +32,9 @@ include $(RTE_SDK)/mk/rte.vars.mk
 
 core-libs := librte_eal librte_mbuf librte_mempool librte_ring librte_ether
 
+DIRS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += dpaa
+DEPDIRS-dpaa = $(core-libs)
+
 DIRS-$(CONFIG_RTE_LIBRTE_FSLMC_BUS) += fslmc
 DEPDIRS-fslmc = $(core-libs)
 
diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
new file mode 100644
index 000..28694c0
--- /dev/null
+++ b/drivers/bus/dpaa/Makefile
@@ -0,0 +1,58 @@
+#   BSD LICENSE
+#
+#   Copyright 2016 NXP.
+#
+#   Redistribution and use in source and binary forms, with or without
+#   modification, are permitted provided that the following conditions
+#   are met:
+#
+# * Redistributions of source code must retain the above copyright
+#   notice, this list of conditions and the following disclaimer.
+# * Redistributions in binary form must reproduce the above copyright
+#   notice, this list of conditions and the following disclaimer in
+#   the documentation and/or other materials provided with the
+#   distribution.
+# * Neither the name of NXP nor the names of its
+#   contributors may be used to endorse or promote products derived
+#   from this software without specific prior written permission.
+#
+#   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+#   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+#   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+#   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+#   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+#   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+#   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+#   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+#   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+#   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+#   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+include $(RTE_SDK)/mk/rte.vars.mk
+RTE_BUS_DPAA=$(RTE_SDK)/drivers/bus/dpaa
+
+#
+# library name
+#
+LIB = librte_bus_dpaa.a
+
+CFLAGS := -I$(SRCDIR) $(CFLAGS)
+CFLAGS += -O3 $(WERROR_FLAGS)
+CFLAGS += -I$(RTE_BUS_DPAA)/
+CFLAGS += -I$(RTE_SDK)/lib/librte_eal/linuxapp/eal
+CFLAGS += -I$(RTE_SDK)/lib/librte_eal/common/include
+
+# versioning export map
+EXPORT_MAP := rte_bus_dpaa_version.map
+
+LIBABIVER := 1
+
+# all source are stored in SRCS-y
+#
+SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
+   dpaa_bus.c
+
+# Link Pthread
+LDLIBS += -lpthread
+
+include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/bus/dpaa/dpaa_bus.c b/drivers/bus/dpaa/dpaa_bus.c
new file mode 100644
index 000..cc343b3
--- /dev/null
+++ b/drivers/bus/dpaa/dpaa_bus.c
@@ -0,0 +1,207 @@
+/*-
+ *

[dpdk-dev] [PATCH v6 03/40] bus/dpaa: add compatibility and helper macros

[Shreyansh Jain]

From: Hemant Agrawal 

Linked list, bit operations and compatibility macros.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Hemant Agrawal 
---
 drivers/bus/dpaa/include/compat.h| 385 +++
 drivers/bus/dpaa/include/dpaa_bits.h |  65 ++
 drivers/bus/dpaa/include/dpaa_list.h | 101 +
 3 files changed, 551 insertions(+)
 create mode 100644 drivers/bus/dpaa/include/compat.h
 create mode 100644 drivers/bus/dpaa/include/dpaa_bits.h
 create mode 100644 drivers/bus/dpaa/include/dpaa_list.h

diff --git a/drivers/bus/dpaa/include/compat.h 
b/drivers/bus/dpaa/include/compat.h
new file mode 100644
index 000..42733ae
--- /dev/null
+++ b/drivers/bus/dpaa/include/compat.h
@@ -0,0 +1,385 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2011 Freescale Semiconductor, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef __COMPAT_H
+#define __COMPAT_H
+
+#include 
+
+#ifndef _GNU_SOURCE
+#define _GNU_SOURCE
+#endif
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+
+/* The following definitions are primarily to allow the single-source driver
+ * interfaces to be included by arbitrary program code. Ie. for interfaces that
+ * are also available in kernel-space, these definitions provide compatibility
+ * with certain attributes and types used in those interfaces.
+ */
+
+/* Required compiler attributes */
+#define __maybe_unused __rte_unused
+#define __always_unused__rte_unused
+#define __packed   __rte_packed
+#define noinline   __attribute__((noinline))
+
+#define L1_CACHE_BYTES 64
+#define cacheline_aligned __attribute__((aligned(L1_CACHE_BYTES)))
+#define __stringify_1(x) #x
+#define __stringify(x) __stringify_1(x)
+
+#ifdef ARRAY_SIZE
+#undef ARRAY_SIZE
+#endif
+#define ARRAY_SIZE(a) (sizeof(a) / sizeof((a)[0]))
+
+/* Debugging */
+#define prflush(fmt, args...) \
+   do { \
+   printf(fmt, ##args); \
+   fflush(stdout); \
+   } while (0)
+
+#define pr_crit(fmt, args...)   prflush("CRIT:" fmt, ##args)
+#define pr_err(fmt, args...)prflush("ERR:" fmt, ##args)
+#define pr_warn(fmt, args...)   prflush("WARN:" fmt, ##args)
+#define pr_info(fmt, args...)   prflush(fmt, ##args)
+
+#ifdef RTE_LIBRTE_DPAA_DEBUG_BUS
+#ifdef pr_debug
+#undef pr_debug
+#endif
+#define pr_debug(fmt, args...) printf(fmt, ##args)
+#else
+#define pr_debug(fmt, args...) {}
+#endif
+
+#define DPAA_BUG_ON(x) RTE_ASSERT(x)
+
+/* Required types */
+typedef uint8_tu8;
+typedef uint16_t   u16;
+typedef uint32_t   u32;
+typedef uint64_t   u64;
+typedef uint64_t   dma_addr_t;
+typedef cpu_set_t  cpumask_t;
+typedef uint32_t   phandle;
+typedef uint32_t   gfp_t;
+typedef uint32_t   irqreturn_t;
+
+#define IRQ_HANDLED0
+#define request_irqqbman_request_irq
+#define free_irq   qb

[dpdk-dev] [PATCH v6 04/40] bus/dpaa: add OF parser for device scanning

[Shreyansh Jain]

This layer is used by Bus driver's scan function. Devices are parsed
using OF parser and added to DPAA device list.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile   |   7 +
 drivers/bus/dpaa/base/fman/of.c | 576 
 drivers/bus/dpaa/include/of.h   | 190 +
 3 files changed, 773 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/fman/of.c
 create mode 100644 drivers/bus/dpaa/include/of.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index 28694c0..30a3a5d 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -38,7 +38,11 @@ LIB = librte_bus_dpaa.a
 
 CFLAGS := -I$(SRCDIR) $(CFLAGS)
 CFLAGS += -O3 $(WERROR_FLAGS)
+CFLAGS += -Wno-pointer-arith
+CFLAGS += -Wno-cast-qual
+CFLAGS += -D _GNU_SOURCE
 CFLAGS += -I$(RTE_BUS_DPAA)/
+CFLAGS += -I$(RTE_BUS_DPAA)/include
 CFLAGS += -I$(RTE_SDK)/lib/librte_eal/linuxapp/eal
 CFLAGS += -I$(RTE_SDK)/lib/librte_eal/common/include
 
@@ -52,6 +56,9 @@ LIBABIVER := 1
 SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
dpaa_bus.c
 
+SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
+   base/fman/of.c \
+
 # Link Pthread
 LDLIBS += -lpthread
 
diff --git a/drivers/bus/dpaa/base/fman/of.c b/drivers/bus/dpaa/base/fman/of.c
new file mode 100644
index 000..b2d7c02
--- /dev/null
+++ b/drivers/bus/dpaa/base/fman/of.c
@@ -0,0 +1,576 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2010-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+#include 
+
+static int alive;
+static struct dt_dir root_dir;
+static const char *base_dir;
+static COMPAT_LIST_HEAD(linear);
+
+static int
+of_open_dir(const char *relative_path, struct dirent ***d)
+{
+   int ret;
+   char full_path[PATH_MAX];
+
+   snprintf(full_path, PATH_MAX, "%s/%s", base_dir, relative_path);
+   ret = scandir(full_path, d, 0, versionsort);
+   if (ret < 0)
+   DPAA_BUS_LOG(ERR, "Failed to open directory %s",
+full_path);
+   return ret;
+}
+
+static void
+of_close_dir(struct dirent **d, int num)
+{
+   while (num--)
+   free(d[num]);
+   free(d);
+}
+
+static int
+of_open_file(const char *relative_path)
+{
+   int ret;
+   char full_path[PATH_MAX];
+
+   snprintf(full_path, PATH_MAX, "%s/%s", base_dir, relative_path);
+   ret = open(full_path, O_RDONLY);
+   if (ret < 0)
+   DPAA_BUS_LOG(ERR, "Failed to open directory %s",
+full_path);
+   return ret;
+}
+
+static void
+process_file(struct dirent *dent, struct dt_dir *parent)
+{
+   int fd;
+   struct dt_file *f = malloc(sizeof(*f));
+
+   if (!f) {
+   DPAA_BUS_LOG(DEBUG, "Unable to allocate memory for file node");
+   return;
+   }
+   f->node.is_file = 1;
+   snprintf(f->node.node.name, NAME_MAX, "%s", dent->d_name);
+   snprintf(f->node.node.full_name, PATH_MAX, "%s/%s",
+parent->node.node.full_name, 

[dpdk-dev] [PATCH v6 07/40] bus/dpaa: enable DPAA IOCTL portal driver

[Shreyansh Jain]

Userspace applications interact with DPAA blocks using this IOCTL driver.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile |   4 +-
 drivers/bus/dpaa/base/qbman/process.c | 331 ++
 drivers/bus/dpaa/include/fsl_usd.h|  88 +
 drivers/bus/dpaa/include/process.h| 107 +++
 4 files changed, 529 insertions(+), 1 deletion(-)
 create mode 100644 drivers/bus/dpaa/base/qbman/process.c
 create mode 100644 drivers/bus/dpaa/include/fsl_usd.h
 create mode 100644 drivers/bus/dpaa/include/process.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index fe65276..f06521c 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -43,6 +43,7 @@ CFLAGS += -Wno-cast-qual
 CFLAGS += -D _GNU_SOURCE
 CFLAGS += -I$(RTE_BUS_DPAA)/
 CFLAGS += -I$(RTE_BUS_DPAA)/include
+CFLAGS += -I$(RTE_BUS_DPAA)/base/qbman
 CFLAGS += -I$(RTE_SDK)/lib/librte_eal/linuxapp/eal
 CFLAGS += -I$(RTE_SDK)/lib/librte_eal/common/include
 
@@ -60,7 +61,8 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/fman.c \
base/fman/fman_hw.c \
base/fman/of.c \
-   base/fman/netcfg_layer.c
+   base/fman/netcfg_layer.c \
+   base/qbman/process.c
 
 # Link Pthread
 LDLIBS += -lpthread
diff --git a/drivers/bus/dpaa/base/qbman/process.c 
b/drivers/bus/dpaa/base/qbman/process.c
new file mode 100644
index 000..b8ec539
--- /dev/null
+++ b/drivers/bus/dpaa/base/qbman/process.c
@@ -0,0 +1,331 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2011-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+#include 
+#include 
+#include 
+#include 
+
+#include "process.h"
+
+#include 
+
+/* As higher-level drivers will be built on top of this (dma_mem, qbman, ...),
+ * it's preferable that the process driver itself not provide any exported API.
+ * As such, combined with the fact that none of these operations are
+ * performance critical, it is justified to use lazy initialisation, so that's
+ * what the lock is for.
+ */
+static int fd = -1;
+static pthread_mutex_t fd_init_lock = PTHREAD_MUTEX_INITIALIZER;
+
+static int check_fd(void)
+{
+   int ret;
+
+   if (fd >= 0)
+   return 0;
+   ret = pthread_mutex_lock(&fd_init_lock);
+   assert(!ret);
+   /* check again with the lock held */
+   if (fd < 0)
+   fd = open(PROCESS_PATH, O_RDWR);
+   ret = pthread_mutex_unlock(&fd_init_lock);
+   assert(!ret);
+   return (fd >= 0) ? 0 : -ENODEV;
+}
+
+#define DPAA_IOCTL_MAGIC 'u'
+struct dpaa_ioctl_id_alloc {
+   uint32_t base; /* Return value, the start of the allocated range */
+   enum dpaa_id_type id_type; /* what kind of resource(s) to allocate */
+   uint32_t num; /* how many IDs to allocate (and return value) */
+   uint32_t align; /* must be a power of 2, 0 is treated like 1 */
+   int partial; /* whether to allow less than 'num' */
+};
+
+struct dpaa_ioctl_id_release {
+   /* Inp

[dpdk-dev] [PATCH v6 05/40] bus/dpaa: introducing FMan configurations

[Shreyansh Jain]

FMan or Frame Manager, inspects traffic, splits it into queueson ingress.
It is also responsible for directing traffic on queues on egress.

This patch introduces FMan configurational interfaces. This layer is
used by Bus driver for configuring the hardware block.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile |   2 +
 drivers/bus/dpaa/base/fman/fman.c | 611 ++
 drivers/bus/dpaa/base/fman/netcfg_layer.c | 214 +++
 drivers/bus/dpaa/include/fman.h   | 458 ++
 drivers/bus/dpaa/include/netcfg.h |  96 +
 5 files changed, 1381 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/fman/fman.c
 create mode 100644 drivers/bus/dpaa/base/fman/netcfg_layer.c
 create mode 100644 drivers/bus/dpaa/include/fman.h
 create mode 100644 drivers/bus/dpaa/include/netcfg.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index 30a3a5d..f6e504d 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -57,7 +57,9 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
dpaa_bus.c
 
 SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
+   base/fman/fman.c \
base/fman/of.c \
+   base/fman/netcfg_layer.c
 
 # Link Pthread
 LDLIBS += -lpthread
diff --git a/drivers/bus/dpaa/base/fman/fman.c 
b/drivers/bus/dpaa/base/fman/fman.c
new file mode 100644
index 000..2c6029e
--- /dev/null
+++ b/drivers/bus/dpaa/base/fman/fman.c
@@ -0,0 +1,611 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2010-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+#include 
+#include 
+
+#include 
+
+/* This header declares the driver interface we implement */
+#include 
+#include 
+#include 
+
+#define QMI_PORT_REGS_OFFSET   0x400
+
+/* CCSR map address to access ccsr based register */
+void *fman_ccsr_map;
+/* fman version info */
+u16 fman_ip_rev;
+static int get_once;
+u32 fman_dealloc_bufs_mask_hi;
+u32 fman_dealloc_bufs_mask_lo;
+
+int fman_ccsr_map_fd = -1;
+static COMPAT_LIST_HEAD(__ifs);
+
+/* This is the (const) global variable that callers have read-only access to.
+ * Internally, we have read-write access directly to __ifs.
+ */
+const struct list_head *fman_if_list = &__ifs;
+
+static void
+if_destructor(struct __fman_if *__if)
+{
+   struct fman_if_bpool *bp, *tmpbp;
+
+   if (__if->__if.mac_type == fman_offline)
+   goto cleanup;
+
+   list_for_each_entry_safe(bp, tmpbp, &__if->__if.bpool_list, node) {
+   list_del(&bp->node);
+   rte_free(bp);
+   }
+cleanup:
+   rte_free(__if);
+}
+
+static int
+fman_get_ip_rev(const struct device_node *fman_node)
+{
+   const uint32_t *fman_addr;
+   uint64_t phys_addr;
+   uint64_t regs_size;
+   uint32_t ip_rev_1;
+   int _errno;
+
+   fman_addr = of_get_address(fman_node, 0, ®s_size, NULL);
+   if (!fman_addr) {
+   pr_err("of_get_address cannot return fman address\n")

[dpdk-dev] [PATCH v6 06/40] bus/dpaa: add FMan hardware operations

[Shreyansh Jain]

Signed-off-by: Geoff Thorpe 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile |   1 +
 drivers/bus/dpaa/base/fman/fman_hw.c  | 562 ++
 drivers/bus/dpaa/include/fsl_fman.h   | 174 +
 drivers/bus/dpaa/include/fsl_fman_crc64.h | 263 ++
 4 files changed, 1000 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/fman/fman_hw.c
 create mode 100644 drivers/bus/dpaa/include/fsl_fman.h
 create mode 100644 drivers/bus/dpaa/include/fsl_fman_crc64.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index f6e504d..fe65276 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -58,6 +58,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
 
 SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/fman.c \
+   base/fman/fman_hw.c \
base/fman/of.c \
base/fman/netcfg_layer.c
 
diff --git a/drivers/bus/dpaa/base/fman/fman_hw.c 
b/drivers/bus/dpaa/base/fman/fman_hw.c
new file mode 100644
index 000..a7ca661
--- /dev/null
+++ b/drivers/bus/dpaa/base/fman/fman_hw.c
@@ -0,0 +1,562 @@
+/*-
+ *   BSD LICENSE
+ *
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+#include 
+#include 
+#include 
+/* This header declares things about Fman hardware itself (the format of status
+ * words and an inline implementation of CRC64). We include it only in order to
+ * instantiate the one global variable it depends on.
+ */
+#include 
+#include 
+
+/* Instantiate the global variable that the inline CRC64 implementation (in
+ * ) depends on.
+ */
+DECLARE_FMAN_CRC64_TABLE();
+
+#define ETH_ADDR_TO_UINT64(eth_addr)  \
+   (uint64_t)(((uint64_t)(eth_addr)[0] << 40) |   \
+   ((uint64_t)(eth_addr)[1] << 32) |   \
+   ((uint64_t)(eth_addr)[2] << 24) |   \
+   ((uint64_t)(eth_addr)[3] << 16) |   \
+   ((uint64_t)(eth_addr)[4] << 8) |\
+   ((uint64_t)(eth_addr)[5]))
+
+void
+fman_if_set_mcast_filter_table(struct fman_if *p)
+{
+   struct __fman_if *__if = container_of(p, struct __fman_if, __if);
+   void *hashtable_ctrl;
+   uint32_t i;
+
+   hashtable_ctrl = &((struct memac_regs *)__if->ccsr_map)->hashtable_ctrl;
+   for (i = 0; i < 64; i++)
+   out_be32(hashtable_ctrl, i|HASH_CTRL_MCAST_EN);
+}
+
+void
+fman_if_reset_mcast_filter_table(struct fman_if *p)
+{
+   struct __fman_if *__if = container_of(p, struct __fman_if, __if);
+   void *hashtable_ctrl;
+   uint32_t i;
+
+   hashtable_ctrl = &((struct memac_regs *)__if->ccsr_map)->hashtable_ctrl;
+   for (i = 0; i < 64; i++)
+   out_be32(hashtable_ctrl, i & ~HASH_CTRL_MCAST_EN);
+}
+
+static
+uint32_t get_mac_hash_code(uint64_t eth_addr)
+{
+   uint64_tmask1, mask2;
+   uint32_txorVal = 0;
+   uint8_t i, j;
+
+   for (i = 0; i < 6; i++) {
+   mask1 = eth_addr & (uint64_t)0x01;
+   eth_addr >>= 1;
+
+   for (j = 0; j < 7; j++) {
+   mask2 = eth_addr & (uint64_t)0x01;
+   mask1 ^= mask2;
+   eth_addr >>= 1;
+   }
+
+   xorVal |= (mask1 << (5 - i));
+   }
+
+   return xorVal;
+}
+
+int
+fman_if_add_hash_mac_addr(struct fman_if *p, uint8_t *eth)
+{
+   uint64_t eth_addr;
+   void *hashtable_ctrl;
+   uint32_t hash;
+
+   struct __fman_if *__if = container_o

[dpdk-dev] [PATCH v6 09/40] bus/dpaa: add routines for managing a RB tree

[Shreyansh Jain]

QMAN frames are managed over a RB tree data structure.
This patch introduces necessary routines for implementing a RB tree.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/include/dpaa_rbtree.h | 143 +
 1 file changed, 143 insertions(+)
 create mode 100644 drivers/bus/dpaa/include/dpaa_rbtree.h

diff --git a/drivers/bus/dpaa/include/dpaa_rbtree.h 
b/drivers/bus/dpaa/include/dpaa_rbtree.h
new file mode 100644
index 000..f8c9b59
--- /dev/null
+++ b/drivers/bus/dpaa/include/dpaa_rbtree.h
@@ -0,0 +1,143 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright 2017 NXP.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ *   notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ *   notice, this list of conditions and the following disclaimer in
+ *   the documentation and/or other materials provided with the
+ *   distribution.
+ * * Neither the name of NXP nor the names of its
+ *   contributors may be used to endorse or promote products derived
+ *   from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef __DPAA_RBTREE_H
+#define __DPAA_RBTREE_H
+
+#include 
+//
+/* RB-trees */
+//
+
+/* Linux has a good RB-tree implementation, that we can't use (GPL). It also 
has
+ * a flat/hooked-in interface that virtually requires license-contamination in
+ * order to write a caller-compatible implementation. Instead, I've created an
+ * RB-tree encapsulation on top of linux's primitives (it does some of the work
+ * the client logic would normally do), and this gives us something we can
+ * reimplement on LWE. Unfortunately there's no good+free RB-tree
+ * implementations out there that are license-compatible and "flat" (ie. no
+ * dynamic allocation). I did find a malloc-based one that I could convert, but
+ * that will be a task for later on. For now, LWE's RB-tree is implemented 
using
+ * an ordered linked-list.
+ *
+ * Note, the only linux-esque type is "struct rb_node", because it's used
+ * statically in the exported header, so it can't be opaque. Our version 
doesn't
+ * include a "rb_parent_color" field because we're doing linked-list instead of
+ * a true rb-tree.
+ */
+
+struct rb_node {
+   struct rb_node *prev, *next;
+};
+
+struct dpa_rbtree {
+   struct rb_node *head, *tail;
+};
+
+#define DPAA_RBTREE { NULL, NULL }
+static inline void dpa_rbtree_init(struct dpa_rbtree *tree)
+{
+   tree->head = tree->tail = NULL;
+}
+
+#define QMAN_NODE2OBJ(ptr, type, node_field) \
+   (type *)((char *)ptr - offsetof(type, node_field))
+
+#define IMPLEMENT_DPAA_RBTREE(name, type, node_field, val_field) \
+static inline int name##_push(struct dpa_rbtree *tree, type *obj) \
+{ \
+   struct rb_node *node = tree->head; \
+   if (!node) { \
+   tree->head = tree->tail = &obj->node_field; \
+   obj->node_field.prev = obj->node_field.next = NULL; \
+   return 0; \
+   } \
+   while (node) { \
+   type *item = QMAN_NODE2OBJ(node, type, node_field); \
+   if (obj->val_field == item->val_field) \
+   return -EBUSY; \
+   if (obj->val_field < item->val_field) { \
+   if (tree->head == node) \
+   tree->head = &obj->node_field; \
+   else \
+   node->prev->next = &obj->node_field; \
+   obj->node_field.prev = node->prev; \
+   obj->node_field.next = node; \
+   node->prev = &obj->node_field; \
+   return 0; \
+   } \
+   node = node->next; \
+   } \
+   obj->node_field.prev = tree->tail; \
+   obj->node_field.next = NULL; \
+   tree->tail->next = &obj->nod

[dpdk-dev] [PATCH v6 08/40] bus/dpaa: add layer for interrupt emulation using pthread

[Shreyansh Jain]

An interrupt manager is implemented by emulating over pthreads.
Handlers are registered by QBMAN layer for being notified about
any interrupt request from DPAA blocks in userspace.

Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile  |   3 +-
 drivers/bus/dpaa/base/qbman/dpaa_sys.c | 136 +
 drivers/bus/dpaa/base/qbman/dpaa_sys.h |  61 +++
 3 files changed, 199 insertions(+), 1 deletion(-)
 create mode 100644 drivers/bus/dpaa/base/qbman/dpaa_sys.c
 create mode 100644 drivers/bus/dpaa/base/qbman/dpaa_sys.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index f06521c..5b76a4b 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -62,7 +62,8 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/fman_hw.c \
base/fman/of.c \
base/fman/netcfg_layer.c \
-   base/qbman/process.c
+   base/qbman/process.c \
+   base/qbman/dpaa_sys.c
 
 # Link Pthread
 LDLIBS += -lpthread
diff --git a/drivers/bus/dpaa/base/qbman/dpaa_sys.c 
b/drivers/bus/dpaa/base/qbman/dpaa_sys.c
new file mode 100644
index 000..0017da5
--- /dev/null
+++ b/drivers/bus/dpaa/base/qbman/dpaa_sys.c
@@ -0,0 +1,136 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2013-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+#include "dpaa_sys.h"
+
+struct process_interrupt {
+   int irq;
+   irqreturn_t (*isr)(int irq, void *arg);
+   unsigned long flags;
+   const char *name;
+   void *arg;
+   struct list_head node;
+};
+
+static COMPAT_LIST_HEAD(process_irq_list);
+static pthread_mutex_t process_irq_lock = PTHREAD_MUTEX_INITIALIZER;
+
+static void process_interrupt_install(struct process_interrupt *irq)
+{
+   int ret;
+   /* Add the irq to the end of the list */
+   ret = pthread_mutex_lock(&process_irq_lock);
+   assert(!ret);
+   list_add_tail(&irq->node, &process_irq_list);
+   ret = pthread_mutex_unlock(&process_irq_lock);
+   assert(!ret);
+}
+
+static void process_interrupt_remove(struct process_interrupt *irq)
+{
+   int ret;
+
+   ret = pthread_mutex_lock(&process_irq_lock);
+   assert(!ret);
+   list_del(&irq->node);
+   ret = pthread_mutex_unlock(&process_irq_lock);
+   assert(!ret);
+}
+
+static struct process_interrupt *process_interrupt_find(int irq_num)
+{
+   int ret;
+   struct process_interrupt *i = NULL;
+
+   ret = pthread_mutex_lock(&process_irq_lock);
+   assert(!ret);
+   list_for_each_entry(i, &process_irq_list, node) {
+   if (i->irq == irq_num)
+   goto done;
+   }
+done:
+   ret = pthread_mutex_unlock(&process_irq_lock);
+   assert(!ret);
+   return i;
+}
+
+/* This is the interface from the platform-agnostic driver code to (de)register
+ * interrupt handlers. We simply create/destroy corresponding structs.
+ */
+int qbman_request_irq(int irq, irqreturn_t (*isr)(in

[dpdk-dev] [PATCH v6 11/40] bus/dpaa: add QMan driver core routines

[Shreyansh Jain]

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 config/defconfig_arm64-dpaa-linuxapp-gcc  |1 +
 drivers/bus/dpaa/Makefile |2 +
 drivers/bus/dpaa/base/qbman/dpaa_alloc.c  |   88 ++
 drivers/bus/dpaa/base/qbman/qman.c| 2402 +
 drivers/bus/dpaa/base/qbman/qman.h|  888 +++
 drivers/bus/dpaa/base/qbman/qman_driver.c |   12 +
 drivers/bus/dpaa/include/fsl_qman.h   |  755 +
 drivers/bus/dpaa/include/fsl_usd.h|1 +
 8 files changed, 4149 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/qbman/dpaa_alloc.c
 create mode 100644 drivers/bus/dpaa/base/qbman/qman.c
 create mode 100644 drivers/bus/dpaa/base/qbman/qman.h

diff --git a/config/defconfig_arm64-dpaa-linuxapp-gcc 
b/config/defconfig_arm64-dpaa-linuxapp-gcc
index 8316fc9..4d6b046 100644
--- a/config/defconfig_arm64-dpaa-linuxapp-gcc
+++ b/config/defconfig_arm64-dpaa-linuxapp-gcc
@@ -49,3 +49,4 @@ CONFIG_RTE_PKTMBUF_HEADROOM=128
 # NXP DPAA Bus
 CONFIG_RTE_LIBRTE_DPAA_BUS=y
 CONFIG_RTE_LIBRTE_DPAA_DEBUG_DRIVER=n
+CONFIG_RTE_LIBRTE_DPAA_HWDEBUG=n
diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index c9c15f8..5957c15 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -63,7 +63,9 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/of.c \
base/fman/netcfg_layer.c \
base/qbman/process.c \
+   base/qbman/qman.c \
base/qbman/qman_driver.c \
+   base/qbman/dpaa_alloc.c \
base/qbman/dpaa_sys.c
 
 # Link Pthread
diff --git a/drivers/bus/dpaa/base/qbman/dpaa_alloc.c 
b/drivers/bus/dpaa/base/qbman/dpaa_alloc.c
new file mode 100644
index 000..690576a
--- /dev/null
+++ b/drivers/bus/dpaa/base/qbman/dpaa_alloc.c
@@ -0,0 +1,88 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2009-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include "dpaa_sys.h"
+#include 
+#include 
+
+int qman_alloc_fqid_range(u32 *result, u32 count, u32 align, int partial)
+{
+   return process_alloc(dpaa_id_fqid, result, count, align, partial);
+}
+
+void qman_release_fqid_range(u32 fqid, u32 count)
+{
+   process_release(dpaa_id_fqid, fqid, count);
+}
+
+int qman_reserve_fqid_range(u32 fqid, unsigned int count)
+{
+   return process_reserve(dpaa_id_fqid, fqid, count);
+}
+
+int qman_alloc_pool_range(u32 *result, u32 count, u32 align, int partial)
+{
+   return process_alloc(dpaa_id_qpool, result, count, align, partial);
+}
+
+void qman_release_pool_range(u32 pool, u32 count)
+{
+   process_release(dpaa_id_qpool, pool, count);
+}
+
+int qman_reserve_pool_range(u32 pool, u32 count)
+{
+   return process_reserve(dpaa_id_qpool, pool, count);
+}
+
+int qman_alloc_cgrid_range(u32 *result, u32 count, u32 align, int partial)
+{
+   return process_alloc(dpaa_id_cgrid, result, count, align, partial);
+}
+
+void qman_release_cgrid_range(u32 cgrid, u32 count)
+{
+   process_release(dpaa_id_cgrid

[dpdk-dev] [PATCH v6 10/40] bus/dpaa: add QMAN interface driver

[Shreyansh Jain]

The Queue Manager (QMan) is a hardware queue management block that
allows software and accelerators on the datapath to enqueue and dequeue
frames in order to communicate.

This part of QBMAN DPAA Block.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile |1 +
 drivers/bus/dpaa/base/qbman/qman_driver.c |  271 +++
 drivers/bus/dpaa/base/qbman/qman_priv.h   |  303 +++
 drivers/bus/dpaa/include/fsl_qman.h   | 1254 +
 drivers/bus/dpaa/include/fsl_usd.h|   13 +
 5 files changed, 1842 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/qbman/qman_driver.c
 create mode 100644 drivers/bus/dpaa/base/qbman/qman_priv.h
 create mode 100644 drivers/bus/dpaa/include/fsl_qman.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index 5b76a4b..c9c15f8 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -63,6 +63,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/of.c \
base/fman/netcfg_layer.c \
base/qbman/process.c \
+   base/qbman/qman_driver.c \
base/qbman/dpaa_sys.c
 
 # Link Pthread
diff --git a/drivers/bus/dpaa/base/qbman/qman_driver.c 
b/drivers/bus/dpaa/base/qbman/qman_driver.c
new file mode 100644
index 000..80dde20
--- /dev/null
+++ b/drivers/bus/dpaa/base/qbman/qman_driver.c
@@ -0,0 +1,271 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2008-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+#include 
+#include "qman_priv.h"
+#include 
+#include 
+
+/* Global variable containing revision id (even on non-control plane systems
+ * where CCSR isn't available).
+ */
+u16 qman_ip_rev;
+u16 qm_channel_pool1 = QMAN_CHANNEL_POOL1;
+u16 qm_channel_caam = QMAN_CHANNEL_CAAM;
+u16 qm_channel_pme = QMAN_CHANNEL_PME;
+
+/* Ccsr map address to access ccsrbased register */
+void *qman_ccsr_map;
+/* The qman clock frequency */
+u32 qman_clk;
+
+static __thread int fd = -1;
+static __thread struct qm_portal_config pcfg;
+static __thread struct dpaa_ioctl_portal_map map = {
+   .type = dpaa_portal_qman
+};
+
+static int fsl_qman_portal_init(uint32_t index, int is_shared)
+{
+   cpu_set_t cpuset;
+   int loop, ret;
+   struct dpaa_ioctl_irq_map irq_map;
+
+   /* Verify the thread's cpu-affinity */
+   ret = pthread_getaffinity_np(pthread_self(), sizeof(cpu_set_t),
+&cpuset);
+   if (ret) {
+   error(0, ret, "pthread_getaffinity_np()");
+   return ret;
+   }
+   pcfg.cpu = -1;
+   for (loop = 0; loop < CPU_SETSIZE; loop++)
+   if (CPU_ISSET(loop, &cpuset)) {
+   if (pcfg.cpu != -1) {
+   pr_err("Thread is not affine to 1 cpu\n");
+   return -EINVAL;
+   }
+   pcfg.cpu = loop;
+   }
+   if (pcfg.cpu == -1) {
+ 

[dpdk-dev] [PATCH v6 13/40] bus/dpaa: support FMAN frame queue lookup

[Shreyansh Jain]

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/base/qbman/qman.c| 99 ++-
 drivers/bus/dpaa/base/qbman/qman_driver.c |  7 ++-
 drivers/bus/dpaa/base/qbman/qman_priv.h   |  7 +++
 drivers/bus/dpaa/include/fsl_qman.h   | 12 
 4 files changed, 122 insertions(+), 3 deletions(-)

diff --git a/drivers/bus/dpaa/base/qbman/qman.c 
b/drivers/bus/dpaa/base/qbman/qman.c
index 9b1630b..8c8d270 100644
--- a/drivers/bus/dpaa/base/qbman/qman.c
+++ b/drivers/bus/dpaa/base/qbman/qman.c
@@ -176,6 +176,65 @@ static inline struct qman_fq *table_find_fq(struct 
qman_portal *p, u32 fqid)
return fqtree_find(&p->retire_table, fqid);
 }
 
+#ifdef CONFIG_FSL_QMAN_FQ_LOOKUP
+static void **qman_fq_lookup_table;
+static size_t qman_fq_lookup_table_size;
+
+int qman_setup_fq_lookup_table(size_t num_entries)
+{
+   num_entries++;
+   /* Allocate 1 more entry since the first entry is not used */
+   qman_fq_lookup_table = vmalloc((num_entries * sizeof(void *)));
+   if (!qman_fq_lookup_table) {
+   pr_err("QMan: Could not allocate fq lookup table\n");
+   return -ENOMEM;
+   }
+   memset(qman_fq_lookup_table, 0, num_entries * sizeof(void *));
+   qman_fq_lookup_table_size = num_entries;
+   pr_debug("QMan: Allocated lookup table at %p, entry count %lu\n",
+   qman_fq_lookup_table,
+   (unsigned long)qman_fq_lookup_table_size);
+   return 0;
+}
+
+/* global structure that maintains fq object mapping */
+static DEFINE_SPINLOCK(fq_hash_table_lock);
+
+static int find_empty_fq_table_entry(u32 *entry, struct qman_fq *fq)
+{
+   u32 i;
+
+   spin_lock(&fq_hash_table_lock);
+   /* Can't use index zero because this has special meaning
+* in context_b field.
+*/
+   for (i = 1; i < qman_fq_lookup_table_size; i++) {
+   if (qman_fq_lookup_table[i] == NULL) {
+   *entry = i;
+   qman_fq_lookup_table[i] = fq;
+   spin_unlock(&fq_hash_table_lock);
+   return 0;
+   }
+   }
+   spin_unlock(&fq_hash_table_lock);
+   return -ENOMEM;
+}
+
+static void clear_fq_table_entry(u32 entry)
+{
+   spin_lock(&fq_hash_table_lock);
+   DPAA_BUG_ON(entry >= qman_fq_lookup_table_size);
+   qman_fq_lookup_table[entry] = NULL;
+   spin_unlock(&fq_hash_table_lock);
+}
+
+static inline struct qman_fq *get_fq_table_entry(u32 entry)
+{
+   DPAA_BUG_ON(entry >= qman_fq_lookup_table_size);
+   return qman_fq_lookup_table[entry];
+}
+#endif
+
 static inline void cpu_to_hw_fqd(struct qm_fqd *fqd)
 {
/* Byteswap the FQD to HW format */
@@ -766,8 +825,13 @@ static u32 __poll_portal_slow(struct qman_portal *p, u32 
is)
break;
case QM_MR_VERB_FQPN:
/* Parked */
+#ifdef CONFIG_FSL_QMAN_FQ_LOOKUP
+   fq = get_fq_table_entry(
+   be32_to_cpu(msg->fq.contextB));
+#else
fq = (void *)(uintptr_t)
be32_to_cpu(msg->fq.contextB);
+#endif
fq_state_change(p, fq, msg, verb);
if (fq->cb.fqs)
fq->cb.fqs(p, fq, &swapped_msg);
@@ -792,7 +856,11 @@ static u32 __poll_portal_slow(struct qman_portal *p, u32 
is)
}
} else {
/* Its a software ERN */
+#ifdef CONFIG_FSL_QMAN_FQ_LOOKUP
+   fq = get_fq_table_entry(be32_to_cpu(msg->ern.tag));
+#else
fq = (void *)(uintptr_t)be32_to_cpu(msg->ern.tag);
+#endif
fq->cb.ern(p, fq, &swapped_msg);
}
num++;
@@ -907,7 +975,11 @@ static inline unsigned int __poll_portal_fast(struct 
qman_portal *p,
clear_vdqcr(p, fq);
} else {
/* SDQCR: context_b points to the FQ */
+#ifdef CONFIG_FSL_QMAN_FQ_LOOKUP
+   fq = get_fq_table_entry(dq->contextB);
+#else
fq = (void *)(uintptr_t)dq->contextB;
+#endif
/* Now let the callback do its stuff */
res = fq->cb.dqrr(p, fq, dq);
/*
@@ -1119,7 +1191,12 @@ int qman_create_fq(u32 fqid, u32 flags, struct qman_fq 
*fq)
fq->flags = flags;
fq->state = qman_fq_state_oos;
fq->cgr_groupid = 0;
-
+#ifdef CONFIG_FSL_QMAN_FQ_LOOKUP
+   if (unlikely(find_empty_fq_table_entry(&fq->key, fq))) {
+   pr_info("Find empty table entry failed\n");
+   return -ENOMEM;
+   }
+#endif
if (!(flags & QMAN_FQ_FLAG_AS_IS) || (f

[dpdk-dev] [PATCH v6 12/40] bus/dpaa: add BMAN driver core

[Shreyansh Jain]

The Buffer Manager (BMan) is a hardware buffer pool management block that
allows software and accelerators on the datapath to acquire and release
buffers in order to build frames.

This patch adds the core routines.

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile |   1 +
 drivers/bus/dpaa/base/qbman/bman_driver.c | 311 +
 drivers/bus/dpaa/base/qbman/bman_priv.h   | 125 ++
 drivers/bus/dpaa/include/fsl_bman.h   | 375 ++
 drivers/bus/dpaa/include/fsl_usd.h|   5 +
 5 files changed, 817 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/qbman/bman_driver.c
 create mode 100644 drivers/bus/dpaa/base/qbman/bman_priv.h
 create mode 100644 drivers/bus/dpaa/include/fsl_bman.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index 5957c15..e1415e4 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -63,6 +63,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/of.c \
base/fman/netcfg_layer.c \
base/qbman/process.c \
+   base/qbman/bman_driver.c \
base/qbman/qman.c \
base/qbman/qman_driver.c \
base/qbman/dpaa_alloc.c \
diff --git a/drivers/bus/dpaa/base/qbman/bman_driver.c 
b/drivers/bus/dpaa/base/qbman/bman_driver.c
new file mode 100644
index 000..fb3c50e
--- /dev/null
+++ b/drivers/bus/dpaa/base/qbman/bman_driver.c
@@ -0,0 +1,311 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2008-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+
+#include 
+#include 
+#include "bman_priv.h"
+#include 
+
+/*
+ * Global variables of the max portal/pool number this bman version supported
+ */
+u16 bman_ip_rev;
+u16 bman_pool_max;
+void *bman_ccsr_map;
+
+/*/
+/* Portal driver */
+/*/
+
+static __thread int fd = -1;
+static __thread struct bm_portal_config pcfg;
+static __thread struct dpaa_ioctl_portal_map map = {
+   .type = dpaa_portal_bman
+};
+
+static int fsl_bman_portal_init(uint32_t idx, int is_shared)
+{
+   cpu_set_t cpuset;
+   int loop, ret;
+   struct dpaa_ioctl_irq_map irq_map;
+
+   /* Verify the thread's cpu-affinity */
+   ret = pthread_getaffinity_np(pthread_self(), sizeof(cpu_set_t),
+&cpuset);
+   if (ret) {
+   error(0, ret, "pthread_getaffinity_np()");
+   return ret;
+   }
+   pcfg.cpu = -1;
+   for (loop = 0; loop < CPU_SETSIZE; loop++)
+   if (CPU_ISSET(loop, &cpuset)) {
+   if (pcfg.cpu != -1) {
+   pr_err("Thread is not affine to 1 cpu");
+   return -EINVAL;
+   }
+   pcfg.cpu = loop;
+   }
+   if (pcfg.cpu == -1) {
+   pr_err("Bug in getaffinity handling!");
+   return -EINVAL;
+   }
+  

[dpdk-dev] [PATCH v6 15/40] bus/dpaa: add fman flow control threshold setting

[Shreyansh Jain]

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/base/fman/fman_hw.c | 28 
 drivers/bus/dpaa/include/fsl_fman.h  |  7 +++
 2 files changed, 35 insertions(+)

diff --git a/drivers/bus/dpaa/base/fman/fman_hw.c 
b/drivers/bus/dpaa/base/fman/fman_hw.c
index a7ca661..077c17c 100644
--- a/drivers/bus/dpaa/base/fman/fman_hw.c
+++ b/drivers/bus/dpaa/base/fman/fman_hw.c
@@ -37,6 +37,7 @@
  */
 #include 
 #include 
+#include 
 
 /* Instantiate the global variable that the inline CRC64 implementation (in
  * ) depends on.
@@ -393,6 +394,33 @@ fman_if_set_bp(struct fman_if *fm_if, unsigned num 
__always_unused,
 }
 
 int
+fman_if_get_fc_threshold(struct fman_if *fm_if)
+{
+   struct __fman_if *__if = container_of(fm_if, struct __fman_if, __if);
+   unsigned int *fmbm_mpd;
+
+   assert(fman_ccsr_map_fd != -1);
+
+   fmbm_mpd = &((struct rx_bmi_regs *)__if->bmi_map)->fmbm_mpd;
+   return in_be32(fmbm_mpd);
+}
+
+int
+fman_if_set_fc_threshold(struct fman_if *fm_if, u32 high_water,
+u32 low_water, u32 bpid)
+{
+   struct __fman_if *__if = container_of(fm_if, struct __fman_if, __if);
+   unsigned int *fmbm_mpd;
+
+   assert(fman_ccsr_map_fd != -1);
+
+   fmbm_mpd = &((struct rx_bmi_regs *)__if->bmi_map)->fmbm_mpd;
+   out_be32(fmbm_mpd, FMAN_ENABLE_BPOOL_DEPLETION);
+   return bm_pool_set_hw_threshold(bpid, low_water, high_water);
+
+}
+
+int
 fman_if_get_fc_quanta(struct fman_if *fm_if)
 {
struct __fman_if *__if = container_of(fm_if, struct __fman_if, __if);
diff --git a/drivers/bus/dpaa/include/fsl_fman.h 
b/drivers/bus/dpaa/include/fsl_fman.h
index ac38082..95aee67 100644
--- a/drivers/bus/dpaa/include/fsl_fman.h
+++ b/drivers/bus/dpaa/include/fsl_fman.h
@@ -112,6 +112,13 @@ void fman_if_loopback_disable(struct fman_if *p);
 void fman_if_set_bp(struct fman_if *fm_if, unsigned int num, int bpid,
size_t bufsize);
 
+/* Get Flow Control threshold parameters on specific interface */
+int fman_if_get_fc_threshold(struct fman_if *fm_if);
+
+/* Enable and Set Flow Control threshold parameters on specific interface */
+int fman_if_set_fc_threshold(struct fman_if *fm_if,
+   u32 high_water, u32 low_water, u32 bpid);
+
 /* Get Flow Control pause quanta on specific interface */
 int fman_if_get_fc_quanta(struct fman_if *fm_if);
 
-- 
2.9.3

[dpdk-dev] [PATCH v6 14/40] bus/dpaa: add BMan hardware interfaces

[Shreyansh Jain]

Signed-off-by: Geoff Thorpe 
Signed-off-by: Roy Pledge 
Signed-off-by: Hemant Agrawal 
Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/Makefile |   1 +
 drivers/bus/dpaa/base/qbman/bman.c| 394 +
 drivers/bus/dpaa/base/qbman/bman.h| 550 ++
 drivers/bus/dpaa/base/qbman/bman_driver.c |  12 +
 drivers/bus/dpaa/base/qbman/dpaa_alloc.c  |  16 +
 5 files changed, 973 insertions(+)
 create mode 100644 drivers/bus/dpaa/base/qbman/bman.c
 create mode 100644 drivers/bus/dpaa/base/qbman/bman.h

diff --git a/drivers/bus/dpaa/Makefile b/drivers/bus/dpaa/Makefile
index e1415e4..61b6432 100644
--- a/drivers/bus/dpaa/Makefile
+++ b/drivers/bus/dpaa/Makefile
@@ -63,6 +63,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_DPAA_BUS) += \
base/fman/of.c \
base/fman/netcfg_layer.c \
base/qbman/process.c \
+   base/qbman/bman.c \
base/qbman/bman_driver.c \
base/qbman/qman.c \
base/qbman/qman_driver.c \
diff --git a/drivers/bus/dpaa/base/qbman/bman.c 
b/drivers/bus/dpaa/base/qbman/bman.c
new file mode 100644
index 000..0480caa
--- /dev/null
+++ b/drivers/bus/dpaa/base/qbman/bman.c
@@ -0,0 +1,394 @@
+/*-
+ * This file is provided under a dual BSD/GPLv2 license. When using or
+ * redistributing this file, you may do so under either license.
+ *
+ *   BSD LICENSE
+ *
+ * Copyright 2008-2016 Freescale Semiconductor Inc.
+ * Copyright 2017 NXP.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of the above-listed copyright holders nor the
+ * names of any contributors may be used to endorse or promote products
+ * derived from this software without specific prior written permission.
+ *
+ *   GPL LICENSE SUMMARY
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDERS OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include "bman.h"
+#include 
+
+/* Compilation constants */
+#define RCR_THRESH 2   /* reread h/w CI when running out of space */
+#define IRQNAME"BMan portal %d"
+#define MAX_IRQNAME16  /* big enough for "BMan portal %d" */
+
+struct bman_portal {
+   struct bm_portal p;
+   /* 2-element array. pools[0] is mask, pools[1] is snapshot. */
+   struct bman_depletion *pools;
+   int thresh_set;
+   unsigned long irq_sources;
+   u32 slowpoll;   /* only used when interrupts are off */
+   /* When the cpu-affine portal is activated, this is non-NULL */
+   const struct bm_portal_config *config;
+   char irqname[MAX_IRQNAME];
+};
+
+static cpumask_t affine_mask;
+static DEFINE_SPINLOCK(affine_mask_lock);
+static RTE_DEFINE_PER_LCORE(struct bman_portal, bman_affine_portal);
+
+static inline struct bman_portal *get_affine_portal(void)
+{
+   return &RTE_PER_LCORE(bman_affine_portal);
+}
+
+/*
+ * This object type refers to a pool, it isn't *the* pool. There may be
+ * more than one such object per BMan buffer pool, eg. if different users of
+ * the pool are operating via different portals.
+ */
+struct bman_pool {
+   struct bman_pool_params params;
+   /* Used for hash-table admin when using depletion notifications. */
+   struct bman_portal *portal;
+   struct bman_pool *next;
+#ifdef RTE_LIBRTE_DPAA_HWDEBUG
+   atomic_t in_use;
+#endif
+};
+
+static inline
+struct bman_portal *bman_create_portal(struct bman_portal *portal,
+  const struct bm_portal_config *c)
+{
+   struct bm_portal *p;
+   const struct bman_depletion *pools = &c->mask;
+   int ret;
+   u8 bpid = 0;

[dpdk-dev] [PATCH v6 18/40] bus/dpaa: add DPAA mempool logging macros

[Shreyansh Jain]

Signed-off-by: Shreyansh Jain 
---
 drivers/bus/dpaa/dpaa_bus.c   |  5 +
 drivers/bus/dpaa/rte_bus_dpaa_version.map |  1 +
 drivers/bus/dpaa/rte_dpaa_logs.h  | 20 
 3 files changed, 26 insertions(+)

diff --git a/drivers/bus/dpaa/dpaa_bus.c b/drivers/bus/dpaa/dpaa_bus.c
index 8017df3..dc2b3ad 100644
--- a/drivers/bus/dpaa/dpaa_bus.c
+++ b/drivers/bus/dpaa/dpaa_bus.c
@@ -70,6 +70,7 @@
 #include 
 
 int dpaa_logtype_bus;
+int dpaa_logtype_mempool;
 
 struct rte_dpaa_bus rte_dpaa_bus;
 struct netcfg_info *dpaa_netcfg;
@@ -452,4 +453,8 @@ dpaa_init_log(void)
dpaa_logtype_bus = rte_log_register("bus.dpaa");
if (dpaa_logtype_bus >= 0)
rte_log_set_level(dpaa_logtype_bus, RTE_LOG_NOTICE);
+
+   dpaa_logtype_mempool = rte_log_register("mempool.dpaa");
+   if (dpaa_logtype_mempool >= 0)
+   rte_log_set_level(dpaa_logtype_mempool, RTE_LOG_NOTICE);
 }
diff --git a/drivers/bus/dpaa/rte_bus_dpaa_version.map 
b/drivers/bus/dpaa/rte_bus_dpaa_version.map
index 853bc47..a2394b8 100644
--- a/drivers/bus/dpaa/rte_bus_dpaa_version.map
+++ b/drivers/bus/dpaa/rte_bus_dpaa_version.map
@@ -8,6 +8,7 @@ DPDK_17.11 {
bman_new_pool;
bman_query_free_buffers;
bman_release;
+   dpaa_logtype_mempool;
dpaa_netcfg;
fman_ccsr_map_fd;
fman_dealloc_bufs_mask_hi;
diff --git a/drivers/bus/dpaa/rte_dpaa_logs.h b/drivers/bus/dpaa/rte_dpaa_logs.h
index cc10937..5335fd8 100644
--- a/drivers/bus/dpaa/rte_dpaa_logs.h
+++ b/drivers/bus/dpaa/rte_dpaa_logs.h
@@ -36,6 +36,7 @@
 #include 
 
 extern int dpaa_logtype_bus;
+extern int dpaa_logtype_mempool;
 
 #define DPAA_BUS_LOG(level, fmt, args...) \
rte_log(RTE_LOG_ ## level, dpaa_logtype_bus, "%s(): " fmt "\n", \
@@ -62,4 +63,23 @@ extern int dpaa_logtype_bus;
 #define DPAA_BUS_WARN(fmt, args...) \
DPAA_BUS_LOG(WARNING, fmt, ## args)
 
+/* Mempool related logs */
+
+#define DPAA_MEMPOOL_LOG(level, fmt, args...) \
+   rte_log(RTE_LOG_ ## level, dpaa_logtype_mempool, "%s(): " fmt "\n", \
+   __func__, ##args)
+
+#define MEMPOOL_INIT_FUNC_TRACE() DPAA_MEMPOOL_LOG(DEBUG, " >>")
+
+#define DPAA_MEMPOOL_DPDEBUG(fmt, args...) \
+   RTE_LOG_DP(DEBUG, PMD, fmt, ## args)
+#define DPAA_MEMPOOL_DEBUG(fmt, args...) \
+   DPAA_MEMPOOL_LOG(DEBUG, fmt, ## args)
+#define DPAA_MEMPOOL_ERR(fmt, args...) \
+   DPAA_MEMPOOL_LOG(ERR, fmt, ## args)
+#define DPAA_MEMPOOL_INFO(fmt, args...) \
+   DPAA_MEMPOOL_LOG(INFO, fmt, ## args)
+#define DPAA_MEMPOOL_WARN(fmt, args...) \
+   DPAA_MEMPOOL_LOG(WARNING, fmt, ## args)
+
 #endif /* _DPAA_LOGS_H_ */
-- 
2.9.3