Re: [dpdk-dev] [PATCH v2 02/15] common/octeontx2: add routine to check if sec capable otx2

[Anoob Joseph]

Hi Akhil,

Please see inline.

Thanks,
Anoob

> -Original Message-
> From: Akhil Goyal 
> Sent: Monday, January 27, 2020 12:31 PM
> To: Anoob Joseph ; Declan Doherty
> ; Thomas Monjalon 
> Cc: Vamsi Krishna Attunuru ; Jerin Jacob
> Kollanukkaran ; Narayana Prasad Raju Athreya
> ; Kiran Kumar Kokkilagadda
> ; Nithin Kumar Dabilpuram
> ; Pavan Nikhilesh Bhagavatula
> ; Ankur Dwivedi ;
> Archana Muniganti ; Tejasree Kondoj
> ; Lukas Bartosik ;
> dev@dpdk.org
> Subject: [EXT] RE: [PATCH v2 02/15] common/octeontx2: add routine to check if
> sec capable otx2
> 
> External Email
> 
> --
> Hi Anoob,
> 
> >
> > From: Vamsi Attunuru 
> >
> > This routine returns true if given rte_eth_dev is security offload
> > capable and belongs to octeontx2.
> >
> > Signed-off-by: Anoob Joseph 
> > Signed-off-by: Tejasree Kondoj 
> > Signed-off-by: Vamsi Attunuru 
> > ---
> >  drivers/common/octeontx2/otx2_common.c   | 20
> > 
> >  drivers/common/octeontx2/otx2_common.h   |  2 ++
> >  .../octeontx2/rte_common_octeontx2_version.map   |  1 +
> >  3 files changed, 23 insertions(+)
> >
> > diff --git a/drivers/common/octeontx2/otx2_common.c
> > b/drivers/common/octeontx2/otx2_common.c
> > index 7e45366..2f9b167 100644
> > --- a/drivers/common/octeontx2/otx2_common.c
> > +++ b/drivers/common/octeontx2/otx2_common.c
> > @@ -3,6 +3,7 @@
> >   */
> >
> >  #include 
> > +#include 
> >  #include 
> >  #include 
> >
> > @@ -23,6 +24,25 @@ otx2_npa_set_defaults(struct otx2_idev_cfg *idev)
> >
> >  /**
> >   * @internal
> > + * Check if rte_eth_dev is security offload capable otx2_eth_dev  */
> > +uint8_t otx2_ethdev_is_sec_capable(struct rte_eth_dev *eth_dev) {
> > +   struct rte_pci_device *pci_dev;
> > +
> > +   pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev);
> > +
> > +   if (pci_dev->id.device_id == PCI_DEVID_OCTEONTX2_RVU_PF ||
> > +   pci_dev->id.device_id == PCI_DEVID_OCTEONTX2_RVU_VF ||
> > +   pci_dev->id.device_id == PCI_DEVID_OCTEONTX2_RVU_AF_VF)
> > +   return 1;
> > +
> > +   return 0;
> > +}
> 
> Why is this API defined in common?
> This is a capability of ethernet device and is only specific to OCTEONTX2 
> driver.

[Anoob] Crypto PMD is querying eth dev device for 'security' capability. If we 
put the call in ethdev PMD, then there will be a build dependency. 

This is required because we need to map crypto queues to eth devices. This is 
required for outbound processing.
 
> 
> - Akhil

Re: [dpdk-dev] [PATCH v2 02/15] common/octeontx2: add routine to check if sec capable otx2

[Anoob Joseph]

Hi Akhil,

Minor correction. Please see inline.

Thanks,
Anoob

> -Original Message-
> From: dev  On Behalf Of Anoob Joseph
> Sent: Monday, January 27, 2020 1:33 PM
> To: Akhil Goyal ; Declan Doherty
> ; Thomas Monjalon 
> Cc: Vamsi Krishna Attunuru ; Jerin Jacob
> Kollanukkaran ; Narayana Prasad Raju Athreya
> ; Kiran Kumar Kokkilagadda
> ; Nithin Kumar Dabilpuram
> ; Pavan Nikhilesh Bhagavatula
> ; Ankur Dwivedi ;
> Archana Muniganti ; Tejasree Kondoj
> ; Lukas Bartosik ;
> dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v2 02/15] common/octeontx2: add routine to
> check if sec capable otx2
> 
> Hi Akhil,
> 
> Please see inline.
> 
> Thanks,
> Anoob
> 
> > -Original Message-
> > From: Akhil Goyal 
> > Sent: Monday, January 27, 2020 12:31 PM
> > To: Anoob Joseph ; Declan Doherty
> > ; Thomas Monjalon 
> > Cc: Vamsi Krishna Attunuru ; Jerin Jacob
> > Kollanukkaran ; Narayana Prasad Raju Athreya
> > ; Kiran Kumar Kokkilagadda
> > ; Nithin Kumar Dabilpuram
> > ; Pavan Nikhilesh Bhagavatula
> > ; Ankur Dwivedi ;
> > Archana Muniganti ; Tejasree Kondoj
> > ; Lukas Bartosik ;
> > dev@dpdk.org
> > Subject: [EXT] RE: [PATCH v2 02/15] common/octeontx2: add routine to
> > check if sec capable otx2
> >
> > External Email
> >
> > --
> > Hi Anoob,
> >
> > >
> > > From: Vamsi Attunuru 
> > >
> > > This routine returns true if given rte_eth_dev is security offload
> > > capable and belongs to octeontx2.
> > >
> > > Signed-off-by: Anoob Joseph 
> > > Signed-off-by: Tejasree Kondoj 
> > > Signed-off-by: Vamsi Attunuru 
> > > ---
> > >  drivers/common/octeontx2/otx2_common.c   | 20
> > > 
> > >  drivers/common/octeontx2/otx2_common.h   |  2 ++
> > >  .../octeontx2/rte_common_octeontx2_version.map   |  1 +
> > >  3 files changed, 23 insertions(+)
> > >
> > > diff --git a/drivers/common/octeontx2/otx2_common.c
> > > b/drivers/common/octeontx2/otx2_common.c
> > > index 7e45366..2f9b167 100644
> > > --- a/drivers/common/octeontx2/otx2_common.c
> > > +++ b/drivers/common/octeontx2/otx2_common.c
> > > @@ -3,6 +3,7 @@
> > >   */
> > >
> > >  #include 
> > > +#include 
> > >  #include 
> > >  #include 
> > >
> > > @@ -23,6 +24,25 @@ otx2_npa_set_defaults(struct otx2_idev_cfg *idev)
> > >
> > >  /**
> > >   * @internal
> > > + * Check if rte_eth_dev is security offload capable otx2_eth_dev
> > > +*/ uint8_t otx2_ethdev_is_sec_capable(struct rte_eth_dev *eth_dev) {
> > > + struct rte_pci_device *pci_dev;
> > > +
> > > + pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev);
> > > +
> > > + if (pci_dev->id.device_id == PCI_DEVID_OCTEONTX2_RVU_PF ||
> > > + pci_dev->id.device_id == PCI_DEVID_OCTEONTX2_RVU_VF ||
> > > + pci_dev->id.device_id == PCI_DEVID_OCTEONTX2_RVU_AF_VF)
> > > + return 1;
> > > +
> > > + return 0;
> > > +}
> >
> > Why is this API defined in common?
> > This is a capability of ethernet device and is only specific to OCTEONTX2 
> > driver.
> 
> [Anoob] Crypto PMD is querying eth dev device for 'security' capability. If 
> we put
> the "definition" in ethdev PMD, then there will be a build dependency.

[Anoob] Minor correction in the above line. 

> 
> This is required because we need to map crypto queues to eth devices. This is
> required for outbound processing.
> 
> >
> > - Akhil

[dpdk-dev] [PATCH 1/2] crypto/dpaa_sec: fix performance issue

[Gagandeep Singh]

DPAA sec driver is using virtual to physical address
translation in its data path and driver is using
dpaax_iova_table_update() API in every address translation
which is very costly.
This patch moves dpaax_iova_table_update() calling to rte_dpaa_mem_ptov(),
only if it fails to found translation from DPAAX table.

Fixes: 033968bdc28 ("drivers/crypto/dpaa_sec: update DPAA iova table in 
dpaa_mem_vtop")
Signed-off-by: Gagandeep Singh 
---
 drivers/bus/dpaa/rte_dpaa_bus.h |  18 -
 drivers/crypto/dpaa_sec/dpaa_sec.c  | 117 +++-
 drivers/crypto/dpaa_sec/meson.build |   2 +-
 3 files changed, 64 insertions(+), 73 deletions(-)

diff --git a/drivers/bus/dpaa/rte_dpaa_bus.h b/drivers/bus/dpaa/rte_dpaa_bus.h
index 9bf2cd9d6..373aca978 100644
--- a/drivers/bus/dpaa/rte_dpaa_bus.h
+++ b/drivers/bus/dpaa/rte_dpaa_bus.h
@@ -132,7 +132,23 @@ static inline void *rte_dpaa_mem_ptov(phys_addr_t paddr)
}
 
/* If not, Fallback to full memseg list searching */
-   return rte_mem_iova2virt(paddr);
+   va = rte_mem_iova2virt(paddr);
+
+   dpaax_iova_table_update(paddr, va, RTE_CACHE_LINE_SIZE);
+
+   return va;
+}
+
+static inline rte_iova_t
+rte_dpaa_mem_vtop(void *vaddr)
+{
+   const struct rte_memseg *ms;
+
+   ms = rte_mem_virt2memseg(vaddr, NULL);
+   if (ms)
+   return ms->iova + RTE_PTR_DIFF(vaddr, ms->addr);
+
+   return (size_t)NULL;
 }
 
 /**
diff --git a/drivers/crypto/dpaa_sec/dpaa_sec.c 
b/drivers/crypto/dpaa_sec/dpaa_sec.c
index e0b307cec..df684d265 100644
--- a/drivers/crypto/dpaa_sec/dpaa_sec.c
+++ b/drivers/crypto/dpaa_sec/dpaa_sec.c
@@ -94,31 +94,6 @@ dpaa_sec_alloc_ctx(dpaa_sec_session *ses, int sg_count)
return ctx;
 }
 
-static inline rte_iova_t
-dpaa_mem_vtop(void *vaddr)
-{
-   const struct rte_memseg *ms;
-
-   ms = rte_mem_virt2memseg(vaddr, NULL);
-   if (ms) {
-   dpaax_iova_table_update(ms->iova, ms->addr, ms->len);
-   return ms->iova + RTE_PTR_DIFF(vaddr, ms->addr);
-   }
-   return (size_t)NULL;
-}
-
-static inline void *
-dpaa_mem_ptov(rte_iova_t paddr)
-{
-   void *va;
-
-   va = (void *)dpaax_iova_table_get_va(paddr);
-   if (likely(va))
-   return va;
-
-   return rte_mem_iova2virt(paddr);
-}
-
 static void
 ern_sec_fq_handler(struct qman_portal *qm __rte_unused,
   struct qman_fq *fq,
@@ -183,7 +158,7 @@ dqrr_out_fq_cb_rx(struct qman_portal *qm __always_unused,
 * sg[0] is for output
 * sg[1] for input
 */
-   job = dpaa_mem_ptov(qm_fd_addr_get64(fd));
+   job = rte_dpaa_mem_ptov(qm_fd_addr_get64(fd));
 
ctx = container_of(job, struct dpaa_sec_op_ctx, job);
ctx->fd_status = fd->status;
@@ -304,12 +279,12 @@ dpaa_sec_prep_pdcp_cdb(dpaa_sec_session *ses)
 
if (!(cdb->sh_desc[2] & 1) && cipherdata.keylen) {
cipherdata.key =
-   (size_t)dpaa_mem_vtop((void *)(size_t)cipherdata.key);
+   (size_t)rte_dpaa_mem_vtop((void 
*)(size_t)cipherdata.key);
cipherdata.key_type = RTA_DATA_PTR;
}
if (!(cdb->sh_desc[2] & (1 << 1)) &&  authdata.keylen) {
authdata.key =
-   (size_t)dpaa_mem_vtop((void *)(size_t)authdata.key);
+   (size_t)rte_dpaa_mem_vtop((void *)(size_t)authdata.key);
authdata.key_type = RTA_DATA_PTR;
}
 
@@ -405,14 +380,14 @@ dpaa_sec_prep_ipsec_cdb(dpaa_sec_session *ses)
if (cdb->sh_desc[2] & 1)
cipherdata.key_type = RTA_DATA_IMM;
else {
-   cipherdata.key = (size_t)dpaa_mem_vtop(
+   cipherdata.key = (size_t)rte_dpaa_mem_vtop(
(void *)(size_t)cipherdata.key);
cipherdata.key_type = RTA_DATA_PTR;
}
if (cdb->sh_desc[2] & (1<<1))
authdata.key_type = RTA_DATA_IMM;
else {
-   authdata.key = (size_t)dpaa_mem_vtop(
+   authdata.key = (size_t)rte_dpaa_mem_vtop(
(void *)(size_t)authdata.key);
authdata.key_type = RTA_DATA_PTR;
}
@@ -591,14 +566,14 @@ dpaa_sec_prep_cdb(dpaa_sec_session *ses)
if (cdb->sh_desc[2] & 1)
alginfo_c.key_type = RTA_DATA_IMM;
else {
-   alginfo_c.key = (size_t)dpaa_mem_vtop(
+   alginfo_c.key = (size_t)rte_dpaa_mem_vtop(
(void *)(size_t)alginfo_c.key);
alginfo_c.key_type = RTA_DATA_PTR;
}
if (cdb->sh_desc[2] & (1<<1))
alginfo_a.key_type = RTA_DATA_IMM;
else {
-   alginfo_a.key = (size_t)dpaa_mem_vtop(
+   alginfo_a.key = (size_t)rte_dpaa_mem_vtop(
 

[dpdk-dev] [PATCH 2/2] crypto/dpaa_sec: reorganize structure members

[Gagandeep Singh]

This patch reorganize the members of a structure
used by driver in its data-path to improve
performance.

Signed-off-by: Gagandeep Singh 
---
 drivers/crypto/dpaa_sec/dpaa_sec.h | 8 
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/crypto/dpaa_sec/dpaa_sec.h 
b/drivers/crypto/dpaa_sec/dpaa_sec.h
index 3ecc7eae5..2cd50cc8d 100644
--- a/drivers/crypto/dpaa_sec/dpaa_sec.h
+++ b/drivers/crypto/dpaa_sec/dpaa_sec.h
@@ -1,6 +1,6 @@
 /* SPDX-License-Identifier: BSD-3-Clause
  *
- *   Copyright 2016 NXP
+ *   Copyright 2016-2020 NXP
  *
  */
 
@@ -117,6 +117,9 @@ struct sec_pdcp_ctxt {
 };
 #endif
 typedef struct dpaa_sec_session_entry {
+   struct sec_cdb cdb; /**< cmd block associated with qp */
+   struct dpaa_sec_qp *qp[MAX_DPAA_CORES];
+   struct qman_fq *inq[MAX_DPAA_CORES];
uint8_t dir; /*!< Operation Direction */
uint8_t ctxt;   /*!< Session Context Type */
enum rte_crypto_cipher_algorithm cipher_alg; /*!< Cipher Algorithm*/
@@ -169,9 +172,6 @@ typedef struct dpaa_sec_session_entry {
struct sec_pdcp_ctxt pdcp;
 #endif
};
-   struct dpaa_sec_qp *qp[MAX_DPAA_CORES];
-   struct qman_fq *inq[MAX_DPAA_CORES];
-   struct sec_cdb cdb; /**< cmd block associated with qp */
 } dpaa_sec_session;
 
 struct dpaa_sec_qp {
-- 
2.17.1

[dpdk-dev] [PATCH 0/2] app/testpmd: add portlist option to the testpmd

[Hariprasad Govindharajan]

This patchset contains 2 patches.
The first one will be changes to testpmd where a new option
portlist is added. 
The second one adds a function eal_parse_optionlist which will
parse user input

The following new command line parameter is added:
--portlist: user can specify the list of forwarding ports
In current version, we are setting the ports using portmask.
With portmask, we can use only upto 64 ports. 
This portlist option enables the user to use more than 64 ports.
--portlist [-p2][,p3[-p4],...]

The following new API is added:
eal_parse_optionlist()
In current version, there is a function which parses
the corelist based on user value. A new generic
function eal_parse_optionlist is added which will
parse corelist as well as similar user input from
the command line

Hariprasad Govindharajan (2):
  app/testpmd: add portlist option to the testpmd
  eal: add eal_parse_optionlist to parse user input

 app/test-pmd/config.c  | 25 +
 app/test-pmd/parameters.c  |  5 
 app/test-pmd/testpmd.h |  3 ++
 lib/librte_eal/common/eal_common_options.c | 45 ++
 lib/librte_eal/common/include/rte_eal.h| 34 ++
 lib/librte_eal/rte_eal_version.map |  1 +
 6 files changed, 95 insertions(+), 18 deletions(-)

-- 
2.7.4

[dpdk-dev] [PATCH 1/2] app/testpmd: add portlist option to the testpmd

[Hariprasad Govindharajan]

In current version, we are setting the ports
using portmask. With portmask, we can use only
upto 64 ports. This portlist option enables the user
to use more than 64 ports.
Now we can specify the ports in 2 different ways
- Using portmask (-p [0x]nnn): mask must be in hex format
- Using portlist in the following format
  --portlist [-p2][,p3[-p4],...]

--portmask 0x2 is same as --portlist 1
--portmask 0x3 is same as --portlist 0-1

Signed-off-by: Hariprasad Govindharajan 
---
 app/test-pmd/config.c | 25 +
 app/test-pmd/parameters.c |  5 +
 app/test-pmd/testpmd.h|  3 +++
 3 files changed, 33 insertions(+)

diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index 9669cbd..49662cf 100644
--- a/app/test-pmd/config.c
+++ b/app/test-pmd/config.c
@@ -2588,6 +2588,31 @@ set_fwd_ports_list(unsigned int *portlist, unsigned int 
nb_pt)
 }
 
 void
+parse_fwd_portlist(const char *portlist)
+{
+   unsigned int portcount = 0;
+   int portindex[RTE_MAX_ETHPORTS];
+   unsigned int ports[RTE_MAX_ETHPORTS];
+   unsigned int idx;
+   /*
+* eal_parse_portlist() will mark the portindex array
+* with -1 if the port is not listed and with a positive value
+* for the listed ports. however, the function set_fwd_ports_list()
+* requires a list of portids' so we use 2 arrays to do the
+* conversion between 2 formats.
+*/
+   if (eal_parse_optionlist(portlist, portindex, RTE_MAX_ETHPORTS) < 0)
+   rte_exit(EXIT_FAILURE, "Invalid fwd port list\n");
+
+   RTE_ETH_FOREACH_DEV(idx) {
+   if (portindex[idx] != -1)
+   ports[portcount++] = idx;
+   }
+
+   set_fwd_ports_list(ports, portcount);
+}
+
+void
 set_fwd_ports_mask(uint64_t portmask)
 {
unsigned int portlist[64];
diff --git a/app/test-pmd/parameters.c b/app/test-pmd/parameters.c
index 6340104..404dba2 100644
--- a/app/test-pmd/parameters.c
+++ b/app/test-pmd/parameters.c
@@ -57,6 +57,7 @@ usage(char* progname)
   "[--help|-h] | [--auto-start|-a] | ["
   "--tx-first | --stats-period=PERIOD | "
   "--coremask=COREMASK --portmask=PORTMASK --numa "
+  "--portlist=PORTLIST "
   "--mbuf-size= | --total-num-mbufs= | "
   "--nb-cores= | --nb-ports= | "
 #ifdef RTE_LIBRTE_CMDLINE
@@ -92,6 +93,7 @@ usage(char* progname)
   "packet forwarding.\n");
printf("  --portmask=PORTMASK: hexadecimal bitmask of ports used "
   "by the packet forwarding test.\n");
+   printf("  --portlist=PORTLIST: list of forwarding ports\n");
printf("  --numa: enable NUMA-aware allocation of RX/TX rings and of "
   "RX memory buffers (mbufs).\n");
printf("  --port-numa-config=(port,socket)[,(port,socket)]: "
@@ -587,6 +589,7 @@ launch_args_parse(int argc, char** argv)
{ "nb-ports",   1, 0, 0 },
{ "coremask",   1, 0, 0 },
{ "portmask",   1, 0, 0 },
+   { "portlist",   1, 0, 0 },
{ "numa",   0, 0, 0 },
{ "no-numa",0, 0, 0 },
{ "mp-anon",0, 0, 0 },
@@ -825,6 +828,8 @@ launch_args_parse(int argc, char** argv)
parse_fwd_coremask(optarg);
if (!strcmp(lgopts[opt_idx].name, "portmask"))
parse_fwd_portmask(optarg);
+   if (!strcmp(lgopts[opt_idx].name, "portlist"))
+   parse_fwd_portlist(optarg);
if (!strcmp(lgopts[opt_idx].name, "no-numa"))
numa_support = 0;
if (!strcmp(lgopts[opt_idx].name, "numa"))
diff --git a/app/test-pmd/testpmd.h b/app/test-pmd/testpmd.h
index 3dd5fc7..33ef3e2 100644
--- a/app/test-pmd/testpmd.h
+++ b/app/test-pmd/testpmd.h
@@ -614,6 +614,9 @@ lcore_num(void)
rte_panic("lcore_id of current thread not found in 
fwd_lcores_cpuids\n");
 }
 
+void
+parse_fwd_portlist(const char *port);
+
 static inline struct fwd_lcore *
 current_fwd_lcore(void)
 {
-- 
2.7.4

[dpdk-dev] [PATCH 2/2] eal: add eal_parse_optionlist to parse user input

[Hariprasad Govindharajan]

In current version, there is a function which parses
the corelist based on user value. A new generic
function eal_parse_optionlist is added which will
parse corelist as well as similar user input so
that we can use it as a public API too.

Signed-off-by: Hariprasad Govindharajan 
---
 lib/librte_eal/common/eal_common_options.c | 45 ++
 lib/librte_eal/common/include/rte_eal.h| 34 ++
 lib/librte_eal/rte_eal_version.map |  1 +
 3 files changed, 62 insertions(+), 18 deletions(-)

diff --git a/lib/librte_eal/common/eal_common_options.c 
b/lib/librte_eal/common/eal_common_options.c
index 5920233..aa59f8b 100644
--- a/lib/librte_eal/common/eal_common_options.c
+++ b/lib/librte_eal/common/eal_common_options.c
@@ -571,33 +571,36 @@ eal_parse_service_corelist(const char *corelist)
return 0;
 }
 
-static int
-eal_parse_corelist(const char *corelist, int *cores)
+int
+eal_parse_optionlist(const char *list, int *values, int maxsize)
 {
unsigned count = 0;
char *end = NULL;
int min, max;
int idx;
 
-   for (idx = 0; idx < RTE_MAX_LCORE; idx++)
-   cores[idx] = -1;
+   if (list == NULL || values == NULL || maxsize < 0)
+   return -1;
+
+   for (idx = 0; idx < maxsize; idx++)
+   values[idx] = -1;
 
/* Remove all blank characters ahead */
-   while (isblank(*corelist))
-   corelist++;
+   while (isblank(*list))
+   list++;
+
+   min = maxsize;
 
-   /* Get list of cores */
-   min = RTE_MAX_LCORE;
do {
-   while (isblank(*corelist))
-   corelist++;
-   if (*corelist == '\0')
+   while (isblank(*list))
+   list++;
+   if (*list == '\0')
return -1;
errno = 0;
-   idx = strtol(corelist, &end, 10);
+   idx = strtol(list, &end, 10);
if (errno || end == NULL)
return -1;
-   if (idx < 0 || idx >= RTE_MAX_LCORE)
+   if (idx < 0 || idx >= maxsize)
return -1;
while (isblank(*end))
end++;
@@ -605,18 +608,18 @@ eal_parse_corelist(const char *corelist, int *cores)
min = idx;
} else if ((*end == ',') || (*end == '\0')) {
max = idx;
-   if (min == RTE_MAX_LCORE)
+   if (min == maxsize)
min = idx;
for (idx = min; idx <= max; idx++) {
-   if (cores[idx] == -1) {
-   cores[idx] = count;
+   if (values[idx] == -1) {
+   values[idx] = count;
count++;
}
}
-   min = RTE_MAX_LCORE;
+   min = maxsize;
} else
return -1;
-   corelist = end + 1;
+   list = end + 1;
} while (*end != '\0');
 
if (count == 0)
@@ -624,6 +627,12 @@ eal_parse_corelist(const char *corelist, int *cores)
return 0;
 }
 
+static int
+eal_parse_corelist(const char *corelist, int *cores)
+{
+   return eal_parse_optionlist(corelist, cores, RTE_MAX_LCORE);
+}
+
 /* Changes the lcore id of the master thread */
 static int
 eal_parse_master_lcore(const char *arg)
diff --git a/lib/librte_eal/common/include/rte_eal.h 
b/lib/librte_eal/common/include/rte_eal.h
index 2f9ed29..567b754 100644
--- a/lib/librte_eal/common/include/rte_eal.h
+++ b/lib/librte_eal/common/include/rte_eal.h
@@ -71,6 +71,40 @@ enum rte_proc_type_t rte_eal_process_type(void);
 int rte_eal_iopl_init(void);
 
 /**
+ * Parse the user input
+ *
+ * This function can be used to read and parse the user input
+ * from the command line. For example, when the user specifies
+ * corelist or port list this function will read the input
+ * and set the forwarding cores or ports
+ *
+ * @param[in] list
+ *   String containing the user input. User can specify
+ *   in these formats 1,3,5 or 1-3 or 1-2,5 or 3,5-6.
+ *   For example, if the user wants to use all the available
+ *   4 ports in his system, then the input can be 0-3 or 0,1,2,3.
+ *   If the user wants to use only the ports 1,2 then the input
+ *   is 1,2.
+ *   valid characters are '-' and ','
+ *   invalid chars like '.' or '#' will result in
+ *   EAL: Error - exiting with code: 1
+ * Cause: Invalid fwd port list
+ * @param[in] values
+ *   An array pointer, used by this function to set the
+ *   array contents to a positive value if they are listed
+ *   in the input
+ *   else sets it to -1
+ * @param[in] maxsize
+ *   This is the maximum value the list string can contain
+ 

Re: [dpdk-dev] [PATCH v2 00/15] add OCTEONTX2 inline IPsec support

[Akhil Goyal]

> 
> On Wed, Jan 22, 2020 at 6:26 PM Akhil Goyal  wrote:
> >
> > > > >
> > > > > Hi Jerin,
> > > > >
> > > > > Will  do the suggested change (RX/rx-> Rx & TX/tx->Tx). Do you want me
> > > > > to trim the headline as well?
> > > > >
> > > >
> > > > Hi Anoob,
> > > >
> > > > > @Akhil, did you get a chance to review the series? Do you have any
> > > > > comments on the patches?
> > > > >
> > > >
> > > > You are adding inline ipsec support to ethernet device and not a crypto
> > > > device.
> > > > These patches should not be part of crypto PMD. There will be cyclic
> > > > dependency Between ethernet device and crypto device which can be
> easily
> > > > avoided.
> > >
> > > [Anoob] We have plans to use lookaside protocol to handle the "fallback"
> > > session. And that involves session sharing between inline and lookaside
> protocol
> > > offloads. Also, though the feature is exposed as a feature of ethdev, on 
> > > our
> > > platform, it's the crypto block which primarily implements the feature. 
> > > And
> so, if
> > > the code is moved to ethdev dir, there would be lot of code duplication. 
> > > The
> > > idea is to have all security related code in one place.
> > >
> > > Also, the PMDs don't have any calls to each other. The communication
> between
> > > the two happens via common. The crypto dev PMD will register the required
> > > security ops to a common structure and ethdev would get it from there. So
> there
> > > won't be an issue of build dependency.
> > >
> >
> > - The code that need to be duplicated can be moved to drivers/common/
> 
> I would like to keep the common code that is common to all the coprocessors.
> 
> Moreover, there are logistic issues in that case where
> a) drivers/common/octeontx2/ going through master repo. So we will be
> creating unnecessary dependency with that 'master' tree.
> b) crypto and ethdev work is done by different teams so we would like
> to make responsibly clear wrt the review and ownership.

I can agree upon that the code shall not be placed in the common/octeontx2, 
But you can have a header file in crypto/octeontx2/ which has some inline 
functions
Which can be called from the ethernet as well as crypto device for inline proto
And lookaside cases respectively. I think with that the maintainability would 
not
be an issue.

IMO, following approach can be looked upon and may be cleaner
- define security ctx and its ops in net/octeontx2.c or 
net/octeontx2_security.c 
These are all control path and should not be any issue.

- define crypto specific code (inline functions) in crypto/octeontx2_security.h
Which can be called for both inline protocol as well as lookaside proto case.

- for data path you can have the processing as is. I hope all dependencies can
Be dealt with as the code will be there in crypto driver for all the mapping of 
Event and crypto queues.

- all APIs which are common for ethernet device and crypto device in
octeontx2_security.h should not prefix eth as is the case in current APIs.

I hope this will clear the ugliness of the code.
> 
> I would like to keep the security/crypto-related code to driver/crypto and
> hook to driver/net/octeontx2 with required functionalities over the
> driver/common
> using a few function pointers to remove the cyclic build dependency.
> 
> Considering there is no cyclic build and shared library dependency
> now, Can we make forward progress
> with the existing scheme?
> 
> > - You may not need to include rte_ethdev.h inside drivers/crypto
> > - otx2_sec_eth_ctx_create should be part of ethdev and you would need
> similar API
> > for crypto device as well when you would support the fallback session 
> > support.
> > So that would go in crypto. Session creation code may be common and can go
> > in drivers/common.
> > - You would have separate security_ctx for both eth device and crypto device
> and that should
> > In net device and crypto device separately. Similarly security->ops should 
> > be
> different in both of them.
> > However if they may have same session creation code and that can go in
> common.
> >

Re: [dpdk-dev] 17.11.10 (LTS) patches review and test

[Luca Boccassi]

On Sun, 2020-01-26 at 14:58 +, Ali Alnubani wrote:
> Hi Luca,
> Sorry for the late response 😊
> 
> > -Original Message-
> > From: Luca Boccassi <
> > bl...@debian.org
> > >
> > Sent: Thursday, January 23, 2020 1:06 PM
> > To: Ali Alnubani <
> > alia...@mellanox.com
> > >; dpdk stable <
> > sta...@dpdk.org
> > >
> > Cc: 
> > dev@dpdk.org
> > ; Raslan Darawsheh <
> > rasl...@mellanox.com
> > >; Thomas
> > Monjalon <
> > tho...@monjalon.net
> > >; Slava Ovsiienko
> > <
> > viachesl...@mellanox.com
> > >; Matan Azrad <
> > ma...@mellanox.com
> > >; Phil
> > Yang <
> > phil.y...@arm.com
> > >; David Marchand
> > <
> > david.march...@redhat.com
> > >
> > Subject: Re: 17.11.10 (LTS) patches review and test
> > 
> > On Wed, 2020-01-22 at 16:34 +, Luca Boccassi wrote:
> > > On Wed, 2020-01-22 at 16:06 +, Ali Alnubani wrote:
> > > > Hi,
> > > > 
> > > > > -Original Message-
> > > > > From: Luca Boccassi <
> > > > > bl...@debian.org
> > > > > 
> > > > > 
> > > > > 
> > > > > Sent: Monday, January 13, 2020 2:47 PM
> > > > > To: dpdk stable <
> > > > > sta...@dpdk.org
> > > > > 
> > > > > 
> > > > > 
> > > > > Cc:
> > > > > dev@dpdk.org
> > > > > 
> > > > > 
> > > > > ; Akhil Goyal <
> > > > > akhil.go...@nxp.com
> > > > > 
> > > > > 
> > > > > > ; Ali Alnubani
> > > > > 
> > > > > <
> > > > > alia...@mellanox.com
> > > > > 
> > > > > 
> > > > > > ;
> > > > > 
> > > > > benjamin.wal...@intel.com
> > > > > 
> > > > > 
> > > > > ; David Christensen
> > > > > <
> > > > > d...@linux.vnet.ibm.com
> > > > > 
> > > > > 
> > > > > > ; Hemant Agrawal <
> > > > > 
> > > > > hemant.agra...@nxp.com
> > > > > 
> > > > > 
> > > > > > ;
> > > > > 
> > > > > Ian Stokes <
> > > > > ian.sto...@intel.com
> > > > > 
> > > > > 
> > > > > > ; Jerin Jacob <
> > > > > 
> > > > > jer...@marvell.com
> > > > > 
> > > > > 
> > > > > > ; John
> > > > > 
> > > > > McNamara <
> > > > > john.mcnam...@intel.com
> > > > > 
> > > > > 
> > > > > > ; Kevin Traynor
> > > > > 
> > > > > <
> > > > > ktray...@redhat.com
> > > > > 
> > > > > 
> > > > > > ; Pei Zhang <
> > > > > 
> > > > > pezh...@redhat.com
> > > > > 
> > > > > 
> > > > > > ;
> > > > > 
> > > > > pingx...@intel.com
> > > > > 
> > > > > 
> > > > > ;
> > > > > qian.q...@intel.com
> > > > > 
> > > > > 
> > > > > ; Raslan Darawsheh
> > > > > <
> > > > > rasl...@mellanox.com
> > > > > 
> > > > > 
> > > > > > ; Thomas Monjalon <
> > > > > 
> > > > > tho...@monjalon.net
> > > > > 
> > > > > 
> > > > > > ;
> > > > > 
> > > > > yuan.p...@intel.com
> > > > > 
> > > > > 
> > > > > ; Chen, Zhaoyan <
> > > > > zhaoyan.c...@intel.com
> > > > > 
> > > > > 
> > > > > 
> > > > > Subject: 17.11.10 (LTS) patches review and test
> > > > > 
> > > > > Hi all,
> > > > > 
> > > > > Here is a list of patches targeted for LTS release 17.11.10.
> > > > > This will be the last 17.11 release.
> > > > > 
> > > > > The planned date for the final release is the 30th of
> > > > > January.
> > > > > 
> > > > > Please help with testing and validation of your use cases and
> > > > > report any issues/results with reply-all to this mail. For
> > > > > the
> > > > > final release the fixes and reported validations will be
> > > > > added to
> > > > > the release notes.
> > > > > 
> > > > 
> > > > The following cases were tested on Mellanox devices:
> > > > - Verify sending and receiving multiple types of traffic.
> > > > - testpmd xstats counter tests.
> > > > - testpmd timestamp tests.
> > > > - Changing/checking link status through testpmd.
> > > > - RTE flow and flow_director tests.
> > > > - Some RSS tests.
> > > > - VLAN stripping and insertion tests.
> > > > - Checksum and TSO tests.
> > > > - ptype tests.
> > > > - Multi-process tests.
> > > > 
> > > > Testing matrix:
> > > > - NIC: ConnectX-4 Lx / OS: RHEL7.4 / Driver: MLNX_OFED_LINUX-
> > > > 4.7-
> > > > 3.2.9.0 / Firmware: 14.26.4012
> > > > - NIC: ConnectX-5 / OS: RHEL7.4 / Driver: MLNX_OFED_LINUX-4.7-
> > > > 3.2.9.0
> > > > / Firmware: 16.26.4012
> > > > 
> > > > Current issues:
> > > > - The message “PMD: Failed to send netlink message: Bad file
> > > > descriptor (9)” is shown at testpmd startup. Still didn’t debug
> > > > its
> > > > effect on functionality.
> > > > - ibv_icmd_create_cq () from /lib64/libibverbs.so.1 segfaults
> > > > after
> > > > setting the number of queues and restarting the ports.
> > > > - Enabling MLX4_PMD causes a build failure with OFED-4.7-
> > > > 3.2.9.0
> > > > (“error: unknown type name ‘off_t’”).
> > > > - Count action isn’t supported in rte_flow.
> > > > - Error creating rte_flow rules with vlan items.
> > > > 
> > > > 
> > > > Regards,
> > > > Ali
> > > 
> > > Thanks - are these new regressions since 17.11.9?
> 
> Only the segfault is caused by new changes. It is caused by commit:
> https://git.dpdk.org/dpdk-stable/commit/?h=17.11&id=a10a7083d164fb48cc3978067820c3d256899f3e
>  (Added Xueming Li to CC list.)
> I should've reported the rest of the issues earlier, my bad.

No problem! Let's wait and see if

Re: [dpdk-dev] [PATCH 2/5] devtools: add license tag to cocci.sh

[Mcnamara, John]

> -Original Message-
> From: dev  On Behalf Of Stephen Hemminger
> Sent: Wednesday, December 4, 2019 3:55 PM
> To: dev@dpdk.org
> Cc: Stephen Hemminger 
> Subject: [dpdk-dev] [PATCH 2/5] devtools: add license tag to cocci.sh
> 
> This file still had original BSD boilerplate text.
> Replace it with SPDX tag.
> 
> Signed-off-by: Stephen Hemminger 
> ---
>  devtools/cocci.sh | 30 +-
>  1 file changed, 1 insertion(+), 29 deletions(-)
> 
> diff --git a/devtools/cocci.sh b/devtools/cocci.sh index
> 8b17a8cebac3..77b7447cdfcd 100755
> --- a/devtools/cocci.sh
> +++ b/devtools/cocci.sh
> @@ -1,34 +1,6 @@
>  #! /bin/sh
> -
> -# BSD LICENSE
> -#
> +# SPDX-License-Identifier: BSD-3-Clause
>  # Copyright 2015 EZchip Semiconductor Ltd.


Acked-by: John McNamara  

Can we please merge this (and the other SPDX patches but this one in 
particular) since the header gets flagged in an IP compliance tool that we use 
in Intel.

Since the original copyright was with EZchip I presume that Mellanox can now 
sign off on this or change the copyright if they wish. CC olga and Thomas.

John
 

Re: [dpdk-dev] [dpdk-stable] [PATCH] test/compress: replace test vector

[Mcnamara, John]

> -Original Message-
> From: dev  On Behalf Of Mcnamara, John
> Sent: Monday, January 20, 2020 6:00 PM
> To: Thomas Monjalon ; Trahe, Fiona
> ; Trybula, ArturX 
> Cc: dev@dpdk.org; sta...@dpdk.org; shal...@marvell.com; Dybkowski, AdamX
> ; Danilewicz, MarcinX
> ; Akhil Goyal 
> Subject: Re: [dpdk-dev] [dpdk-stable] [PATCH] test/compress: replace test
> vector


This patch also needs a BSD SPDX header. Please add that.

John

Re: [dpdk-dev] SPDX license nag

[Mcnamara, John]

> -Original Message-
> From: dev  On Behalf Of Stephen Hemminger
> Sent: Wednesday, January 22, 2020 4:19 PM
> To: dev@dpdk.org
> Subject: [dpdk-dev] SPDX license nag
> 
> Files without SPDX License
> --
> app/test-pmd/flowgen.c   - Tilera/Mellanox
> app/test-pmd/macswap.c   - Tilera/Mellanox
> app/test/test_compressdev_test_buffer.h  - Intel
> app/test/test_timer_racecond.c   - Akamai
> devtools/cocci.sh- EZchip/Mellanox
> devtools/load-devel-config   - Canonical
> examples/ipsec-secgw/test/trs_aesgcm_inline_crypto_fallback_defs.sh - Intel
> examples/ipsec-secgw/test/tun_aesgcm_inline_crypto_fallback_defs.sh - Intel
> examples/performance-thread/l3fwd-thread/test.sh - Intel
> lib/librte_ethdev/rte_ethdev_pci.h   - Brocade/Jan Blunck/Broadcom
> lib/librte_ethdev/rte_ethdev_vdev.h  - Brocade/Jan Blunck/Broadcom
> 
> Some patches for these are outstanding but haven't been merged


I've looked at the headers/logs for these files and added the owners (or 
potential owners after mergers) above.

Can everyone on the CC list please take care of these for your company/files. 
It is a simple patch to fix and Stephen has already provided some.

Thanks Stephen for keeping the focus on this.

John
 

Re: [dpdk-dev] [PATCH v2 00/15] add OCTEONTX2 inline IPsec support

[Jerin Jacob]

On Mon, Jan 27, 2020 at 4:10 PM Akhil Goyal  wrote:
>
>
> >
> > On Wed, Jan 22, 2020 at 6:26 PM Akhil Goyal  wrote:
> > >
> > > > > >
> > > > > > Hi Jerin,
> > > > > >
> > > > > > Will  do the suggested change (RX/rx-> Rx & TX/tx->Tx). Do you want 
> > > > > > me
> > > > > > to trim the headline as well?
> > > > > >
> > > > >
> > > > > Hi Anoob,
> > > > >
> > > > > > @Akhil, did you get a chance to review the series? Do you have any
> > > > > > comments on the patches?
> > > > > >
> > > > >
> > > > > You are adding inline ipsec support to ethernet device and not a 
> > > > > crypto
> > > > > device.
> > > > > These patches should not be part of crypto PMD. There will be cyclic
> > > > > dependency Between ethernet device and crypto device which can be
> > easily
> > > > > avoided.
> > > >
> > > > [Anoob] We have plans to use lookaside protocol to handle the "fallback"
> > > > session. And that involves session sharing between inline and lookaside
> > protocol
> > > > offloads. Also, though the feature is exposed as a feature of ethdev, 
> > > > on our
> > > > platform, it's the crypto block which primarily implements the feature. 
> > > > And
> > so, if
> > > > the code is moved to ethdev dir, there would be lot of code 
> > > > duplication. The
> > > > idea is to have all security related code in one place.
> > > >
> > > > Also, the PMDs don't have any calls to each other. The communication
> > between
> > > > the two happens via common. The crypto dev PMD will register the 
> > > > required
> > > > security ops to a common structure and ethdev would get it from there. 
> > > > So
> > there
> > > > won't be an issue of build dependency.
> > > >
> > >
> > > - The code that need to be duplicated can be moved to drivers/common/
> >
> > I would like to keep the common code that is common to all the coprocessors.
> >
> > Moreover, there are logistic issues in that case where
> > a) drivers/common/octeontx2/ going through master repo. So we will be
> > creating unnecessary dependency with that 'master' tree.
> > b) crypto and ethdev work is done by different teams so we would like
> > to make responsibly clear wrt the review and ownership.
>
> I can agree upon that the code shall not be placed in the common/octeontx2,
> But you can have a header file in crypto/octeontx2/ which has some inline 
> functions
> Which can be called from the ethernet as well as crypto device for inline 
> proto
> And lookaside cases respectively. I think with that the maintainability would 
> not
> be an issue.
>
> IMO, following approach can be looked upon and may be cleaner
> - define security ctx and its ops in net/octeontx2.c or 
> net/octeontx2_security.c
> These are all control path and should not be any issue.
>
> - define crypto specific code (inline functions) in 
> crypto/octeontx2_security.h
> Which can be called for both inline protocol as well as lookaside proto case.

One problem with such an approach is we need to have fat inline functions.
In some case, those inline functions to needs  accessing the
array/driver specific symbols
in another driver namespace then those array needs to be exported in
map file and hence the
build dependency comes.


>
> - for data path you can have the processing as is. I hope all dependencies can
> Be dealt with as the code will be there in crypto driver for all the mapping 
> of
> Event and crypto queues.
>
> - all APIs which are common for ethernet device and crypto device in
> octeontx2_security.h should not prefix eth as is the case in current APIs.

If I understand it correctly, You have a concern in having the
*rte_eth* symbols in driver/crypto/octeontx2?
If so, we can check what can be done. Let us know the exact your
concern in managing the code in this
model?

>
> I hope this will clear the ugliness of the code.

It is relative, I think, having fat inline functions and accessing
both drivers is ugly.

> >
> > I would like to keep the security/crypto-related code to driver/crypto and
> > hook to driver/net/octeontx2 with required functionalities over the
> > driver/common
> > using a few function pointers to remove the cyclic build dependency.
> >
> > Considering there is no cyclic build and shared library dependency
> > now, Can we make forward progress
> > with the existing scheme?
> >
> > > - You may not need to include rte_ethdev.h inside drivers/crypto
> > > - otx2_sec_eth_ctx_create should be part of ethdev and you would need
> > similar API
> > > for crypto device as well when you would support the fallback session 
> > > support.
> > > So that would go in crypto. Session creation code may be common and can go
> > > in drivers/common.
> > > - You would have separate security_ctx for both eth device and crypto 
> > > device
> > and that should
> > > In net device and crypto device separately. Similarly security->ops 
> > > should be
> > different in both of them.
> > > However if they may have same session creation code and that can go in
> > common.
> > >
>

Re: [dpdk-dev] [RFC v2 0/7] RFC: Support MACSEC offload in the RTE_SECURITY infrastructure.

[Akhil Goyal]

Hi Pavel,

> From: Pavel Belous 
> 
> This RFC suggest possible API to implement generic MACSEC HW
> offload in DPDK infrastructure.
> 
> Right now two PMDs implementing MACSEC hw offload via private
> API: ixgbe (Intel) and atlantic (Aquantia).
> 
> During that private API discussion it was decided to go further
> with well defined public API, based most probably on rte_security
> infrastructure.
> 
> Here is that previous discussion:
> 
> http://inbox.dpdk.org/dev/20190416101145.nVecHKp3w14Ptd_hne-DqHhKyzbre88PwNI-OAowXJM@z/
> 
> Declaring macsec API via rte_security gives a good data-centric view on
> parameters
> and operations macsec supports. Old, pure functional API (basically ixbe only 
> API)
> presented function calls with big argument lists which is hard to extend and
> analyse.
> 
> However, I'd like to note rte_security has to be used via explicitly created
> mempools - this hardens abit the usage.
> It also may be hard to extend the structures in the ABI compatible way.
> 
> One of the problems with MACSEC is that internally implementation and
> hardware
> support could be either very simple, doing only endpoint encryption with a 
> single
> TX SC (Secure Connection), or quite complex, capable to do flexible filtering
> and SC matching based on mac, vlan, ethertype and other.
> 
> Different macsec hardware supports some custom features and from our
> experience
> users would like to configure these as well. Therefore there will probably be
> needed a number of PMD specific macsec operators support.
> 
> Examples include: custom in-the-clear tag (matched by vlan id or mask),
> configurable internal logic to allow both secure and unsecure traffic,
> bypass filters on specific ethertypes.
> To support such extensions, suggest use rte_security_macsec_op enum with
> vendor specific operation codes.
> 
> In context of rte_security, MACSEC operations should normally be based on
> security session create and update calls.
> 
> Session create is used to setup overall session. Thats equivalent of old
> `macsec enable` operation.
> 
> Session update is used to update security connections and associations.
> Here xform->op contains the required operation: rx/tx session/association
> add/update/removal.
> 

The patches look good from rte_security perspective. You can send the formal
Patches for 20.05 window.

Re: [dpdk-dev] [PATCH v4] build: allow using wildcards to disable drivers

[Andrzej Ostruszka]

On 1/24/20 4:10 PM, Bruce Richardson wrote:
> Rather than having to explicitly list each and every driver to disable in a
> build, we can use a small python script and the python glob library to
> expand out the wildcards. This means that we can configure meson using e.g.
> 
> meson -Ddisable_drivers=crypto/*,event/* build

Maybe documenting that in 'build-sdk-meson.txt' would be beneficial?

With regards
Andrzej Ostruszka

Re: [dpdk-dev] [PATCH v2 00/15] add OCTEONTX2 inline IPsec support

[Akhil Goyal]

Hi Jerin,

> 
> On Mon, Jan 27, 2020 at 4:10 PM Akhil Goyal  wrote:
> >
> >
> > >
> > > On Wed, Jan 22, 2020 at 6:26 PM Akhil Goyal  wrote:
> > > >
> > > > > > >
> > > > > > > Hi Jerin,
> > > > > > >
> > > > > > > Will  do the suggested change (RX/rx-> Rx & TX/tx->Tx). Do you 
> > > > > > > want
> me
> > > > > > > to trim the headline as well?
> > > > > > >
> > > > > >
> > > > > > Hi Anoob,
> > > > > >
> > > > > > > @Akhil, did you get a chance to review the series? Do you have any
> > > > > > > comments on the patches?
> > > > > > >
> > > > > >
> > > > > > You are adding inline ipsec support to ethernet device and not a 
> > > > > > crypto
> > > > > > device.
> > > > > > These patches should not be part of crypto PMD. There will be cyclic
> > > > > > dependency Between ethernet device and crypto device which can be
> > > easily
> > > > > > avoided.
> > > > >
> > > > > [Anoob] We have plans to use lookaside protocol to handle the 
> > > > > "fallback"
> > > > > session. And that involves session sharing between inline and 
> > > > > lookaside
> > > protocol
> > > > > offloads. Also, though the feature is exposed as a feature of ethdev, 
> > > > > on
> our
> > > > > platform, it's the crypto block which primarily implements the 
> > > > > feature.
> And
> > > so, if
> > > > > the code is moved to ethdev dir, there would be lot of code 
> > > > > duplication.
> The
> > > > > idea is to have all security related code in one place.
> > > > >
> > > > > Also, the PMDs don't have any calls to each other. The communication
> > > between
> > > > > the two happens via common. The crypto dev PMD will register the
> required
> > > > > security ops to a common structure and ethdev would get it from there.
> So
> > > there
> > > > > won't be an issue of build dependency.
> > > > >
> > > >
> > > > - The code that need to be duplicated can be moved to drivers/common/
> > >
> > > I would like to keep the common code that is common to all the
> coprocessors.
> > >
> > > Moreover, there are logistic issues in that case where
> > > a) drivers/common/octeontx2/ going through master repo. So we will be
> > > creating unnecessary dependency with that 'master' tree.
> > > b) crypto and ethdev work is done by different teams so we would like
> > > to make responsibly clear wrt the review and ownership.
> >
> > I can agree upon that the code shall not be placed in the common/octeontx2,
> > But you can have a header file in crypto/octeontx2/ which has some inline
> functions
> > Which can be called from the ethernet as well as crypto device for inline 
> > proto
> > And lookaside cases respectively. I think with that the maintainability 
> > would
> not
> > be an issue.
> >
> > IMO, following approach can be looked upon and may be cleaner
> > - define security ctx and its ops in net/octeontx2.c or 
> > net/octeontx2_security.c
> > These are all control path and should not be any issue.
> >
> > - define crypto specific code (inline functions) in 
> > crypto/octeontx2_security.h
> > Which can be called for both inline protocol as well as lookaside proto 
> > case.
> 
> One problem with such an approach is we need to have fat inline functions.
> In some case, those inline functions to needs  accessing the
> array/driver specific symbols
> in another driver namespace then those array needs to be exported in
> map file and hence the
> build dependency comes.

How many such symbols are there. I don’t they will be many. Can they be
Passed as argument in the APIs to avoid build dependencies.

> 
> 
> >
> > - for data path you can have the processing as is. I hope all dependencies 
> > can
> > Be dealt with as the code will be there in crypto driver for all the 
> > mapping of
> > Event and crypto queues.
> >
> > - all APIs which are common for ethernet device and crypto device in
> > octeontx2_security.h should not prefix eth as is the case in current APIs.
> 
> If I understand it correctly, You have a concern in having the
> *rte_eth* symbols in driver/crypto/octeontx2?
> If so, we can check what can be done. Let us know the exact your
> concern in managing the code in this
> model?

Yes, rte_eth* symbols should not be there in crypto driver.
Because crypto driver is not leveraging any ethernet functionality,
It is the other way, ethernet device is using the crypto functionality/
Structs etc for supporting inline IPSEC.

Also, the security ctx should be part of ethdev  and its ops should be
Defined in ethernet device which may call some inline APIs placed in
Drivers/crypto/octeontx2/

> 
> >
> > I hope this will clear the ugliness of the code.
> 
> It is relative, I think, having fat inline functions and accessing
> both drivers is ugly.

Breaking the way an API need to be defined and used is even more uglier.
IMO, having fat inline functions will act as external library functions which
Are independent of the device which is calling it.
Something similar to drivers/common/dpaax/caamflib/.

My original suggestion was to put it in common, bu

Re: [dpdk-dev] [PATCH] cryptodev: fix set_sym_session_private_data error in sessionless mode

[Akhil Goyal]

> 
> Bugzilla ID: 377
> 
> Signed-off-by: Junxiao Shi 
> ---
Please add appropriate description about the bug and also add
Fixes line to show the original commit.

Change the patch title also as
"drivers/crypto: fix ..."

Re: [dpdk-dev] [PATCH v4] build: allow using wildcards to disable drivers

[Luca Boccassi]

On Fri, 2020-01-24 at 15:10 +, Bruce Richardson wrote:
> Rather than having to explicitly list each and every driver to
> disable in a
> build, we can use a small python script and the python glob library
> to
> expand out the wildcards. This means that we can configure meson
> using e.g.
> 
> meson -Ddisable_drivers=crypto/*,event/* build
> 
> to do a build omitting all the crypto and event drivers. Explicitly
> specified drivers e.g. net/i40e, work as before, and can be mixed
> with
> wildcarded drivers as required.
> 
> Signed-off-by: Bruce Richardson <
> bruce.richard...@intel.com
> >
> ---
> 
> V4:
> - get pep8/pycodestyle compliance
> - simplify getting the root directory
> 
> V3:
> - added check for correct number of params
> - replaced list comprehension with loops for simplicity
> - allow running without meson environmental vars set (for easier
> testing)
> 
> V2:
> - fixed file suffix
> - since it's being called from meson, make this python3 only
> - remove use of chdir()
> - use '\n' rather than ',' between entries

Acked-by: Luca Boccassi 

-- 
Kind regards,
Luca Boccassi

Re: [dpdk-dev] [PATCH v4] build: allow using wildcards to disable drivers

[Bruce Richardson]

On Mon, Jan 27, 2020 at 12:26:53PM +0100, Andrzej Ostruszka wrote:
> On 1/24/20 4:10 PM, Bruce Richardson wrote:
> > Rather than having to explicitly list each and every driver to disable in a
> > build, we can use a small python script and the python glob library to
> > expand out the wildcards. This means that we can configure meson using e.g.
> > 
> > meson -Ddisable_drivers=crypto/*,event/* build
> 
> Maybe documenting that in 'build-sdk-meson.txt' would be beneficial?
> 
Yes, yes it would. Thanks for the suggestion.

/Bruce

[dpdk-dev] [dpdk-dev PATCH] raw/octeontx2_ep: fix coverity scan reported defects

[Mahipal Challa]

Defects reported by coverity scan are resolved.
Coverity issue: CID 353611 - Error handling.
Coverity issue: CID 353622 - Error handling.
Coverity issue: CID 353632 - NULL pointer dereference.

Signed-off-by: Mahipal Challa 
---
 drivers/raw/octeontx2_ep/otx2_ep_enqdeq.c | 18 ++
 1 file changed, 10 insertions(+), 8 deletions(-)

diff --git a/drivers/raw/octeontx2_ep/otx2_ep_enqdeq.c 
b/drivers/raw/octeontx2_ep/otx2_ep_enqdeq.c
index 1ba27c9..9f1e5ed 100644
--- a/drivers/raw/octeontx2_ep/otx2_ep_enqdeq.c
+++ b/drivers/raw/octeontx2_ep/otx2_ep_enqdeq.c
@@ -257,8 +257,8 @@
void *buf;
 
for (idx = 0; idx < droq->nb_desc; idx++) {
-   rte_mempool_get(sdpvf->enqdeq_mpool, &buf);
-   if (buf == NULL) {
+   if (rte_mempool_get(sdpvf->enqdeq_mpool, &buf) ||
+   (buf == NULL)) {
otx2_err("OQ buffer alloc failed");
droq->stats.rx_alloc_failure++;
/* sdp_droq_destroy_ring_buffers(droq);*/
@@ -662,11 +662,11 @@
if (droq->recv_buf_list[droq->refill_idx].buffer != NULL)
break;
 
-   rte_mempool_get(sdpvf->enqdeq_mpool, &buf);
-   /* If a buffer could not be allocated, no point in
-* continuing
-*/
-   if (buf == NULL) {
+   if (rte_mempool_get(sdpvf->enqdeq_mpool, &buf) ||
+   (buf == NULL)) {
+   /* If a buffer could not be allocated, no point in
+* continuing
+*/
droq->stats.rx_alloc_failure++;
break;
}
@@ -780,7 +780,7 @@
droq = sdpvf->droq[q_no];
if (!droq) {
otx2_err("Invalid droq[%d]", q_no);
-   goto deq_fail;
+   goto droq_err;
}
 
/* Grab the lock */
@@ -840,5 +840,7 @@
 
 deq_fail:
rte_spinlock_unlock(&droq->lock);
+
+droq_err:
return SDP_OQ_RECV_FAILED;
 }
-- 
1.8.3.1

[dpdk-dev] Guidance about how to carry out flow control at L2

[nisarg joshi]

Hello community,
I am working on a project in ns-3(network simulator) where we have a
NetDevice implemented called "DpdkNetDevice" which emulates a net device
using dpdk. The traffic control module/layer having the qdisc is a part of
ns-3 and the transmit ring is a part of the dpdk environment. Both the
queues are not in sync and we want to implement flow control between these
2 queues. We have a mechanism where we can signal the tc qdisc to stop and
start, but for that, we require to know the empty space in the transmit
ring. Can you please help us out with the correct API we should be using?
Or is there some mechanism that dpdk has implemented for signaling the
upper layers from the transmit ring? We have come across a function called
rte_eth_tx_descriptor_status  using which we are planning to check the
status of the tail to judge if there is empty space. Can you guys help us
in any way regarding this?

Thank you
Regards

[dpdk-dev] app/testpmd: configure mapping of port queues to cores

[Asaf Sinai]

Hi,

Could not find a way to map specific port queues to specific cores.
For example, mapping port 0 to cores 1-7, port 1 to cores 8-14 and port 2 to 
cores 15-21 is very easy in 'l3fwd' test application:

--config '(0,0,1),(0,1,2),(0,2,3),(0,3,4),(0,4,5),(0,5,6),(0,6,7),
(1,0,8),(1,1,9),(1,2,10),(1,3,11),(1,4,12),(1,5,13),(1,6,14),
(2,0,15),(2,1,16),(2,2,17),(2,3,18),(2,4,19),(2,5,20),(2,6,21)'

Unfortunately, I did not find such option or similar one in 'testpmd'.
Anyone knows if it is possible, and how?

Thanks,
Asaf

[Radware]
Asaf Sinai
ND SW Engineer
Email: asa...@radware.com
T:+972-72-3917050
M:+972-50-6518541
F:+972-3-6488662
[Check out the latest and greatest from 
Radware]

www.radware.com

[Blog]  
[https://www.radware.com/images/signature/linkedin.jpg] 
 
[https://www.radware.com/images/signature/twitter.jpg] 
   [youtube] 

Confidentiality note: This message, and any attachments to it, contains 
privileged/confidential information of RADWARE Ltd./RADWARE Inc. and may not be 
disclosed, used, copied, or transmitted in any form or by any means without 
prior written permission from RADWARE. If you are not the intended recipient, 
delete the message and any attachments from your system without reading or 
copying it, and kindly notify the sender by e-mail. Thank you.
P Please consider your environmental responsibility before printing this e-mail

[dpdk-dev] [PATCH v5] build: allow using wildcards to disable drivers

[Bruce Richardson]

Rather than having to explicitly list each and every driver to disable in a
build, we can use a small python script and the python glob library to
expand out the wildcards. This means that we can configure meson using e.g.

meson -Ddisable_drivers=crypto/*,event/* build

to do a build omitting all the crypto and event drivers. Explicitly
specified drivers e.g. net/i40e, work as before, and can be mixed with
wildcarded drivers as required.

Signed-off-by: Bruce Richardson 
Reviewed-by: Robin Jarry 
Acked-by: Luca Boccassi 
---
V5:
- add a couple of lines to the doc as an example of using this

V4:
- get pep8/pycodestyle compliance
- simplify getting the root directory

V3:
- added check for correct number of params
- replaced list comprehension with loops for simplicity
- allow running without meson environmental vars set (for easier testing)

V2:
- fixed file suffix
- since it's being called from meson, make this python3 only
- remove use of chdir()
- use '\n' rather than ',' between entries
---
 buildtools/list-dir-globs.py | 19 +++
 buildtools/meson.build   |  2 +-
 doc/build-sdk-meson.txt  |  5 -
 drivers/meson.build  |  3 ++-
 4 files changed, 26 insertions(+), 3 deletions(-)
 create mode 100755 buildtools/list-dir-globs.py

diff --git a/buildtools/list-dir-globs.py b/buildtools/list-dir-globs.py
new file mode 100755
index 0..80b5e801f
--- /dev/null
+++ b/buildtools/list-dir-globs.py
@@ -0,0 +1,19 @@
+#! /usr/bin/env python3
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2020 Intel Corporation
+
+import sys
+import os
+from glob import iglob
+
+if len(sys.argv) != 2:
+print("Usage: {0} [,[,...]]".format(sys.argv[0]))
+sys.exit(1)
+
+root = os.path.join(os.getenv('MESON_SOURCE_ROOT', '.'),
+os.getenv('MESON_SUBDIR', '.'))
+
+for path in sys.argv[1].split(','):
+for p in iglob(os.path.join(root, path)):
+if os.path.isdir(p):
+print(os.path.relpath(p))
diff --git a/buildtools/meson.build b/buildtools/meson.build
index cd1d05403..0f563d89a 100644
--- a/buildtools/meson.build
+++ b/buildtools/meson.build
@@ -4,7 +4,7 @@
 subdir('pmdinfogen')
 
 pmdinfo = find_program('gen-pmdinfo-cfile.sh')
-
+list_dir_globs = find_program('list-dir-globs.py')
 check_experimental_syms = find_program('check-experimental-syms.sh')
 
 # set up map-to-def script using python, either built-in or external
diff --git a/doc/build-sdk-meson.txt b/doc/build-sdk-meson.txt
index fc7fe37b5..319a19ef6 100644
--- a/doc/build-sdk-meson.txt
+++ b/doc/build-sdk-meson.txt
@@ -84,7 +84,10 @@ Project-specific options are passed used -Doption=value::
 
meson -Dmachine=default  # use builder-independent baseline -march
 
-Examples of setting the same options using meson configure::
+   meson -Ddisable_drivers=event/*,net/tap  # disable tap driver and all
+   # eventdev PMDs for a smaller build
+
+Examples of setting some of the same options using meson configure::
 
meson configure -Dwerror=true
 
diff --git a/drivers/meson.build b/drivers/meson.build
index 29708cc2b..3ee998d80 100644
--- a/drivers/meson.build
+++ b/drivers/meson.build
@@ -17,7 +17,8 @@ dpdk_driver_classes = ['common',
   'event',   # depends on common, bus, mempool and net.
   'baseband'] # depends on common and bus.
 
-disabled_drivers = get_option('disable_drivers').split(',')
+disabled_drivers = run_command(list_dir_globs, get_option('disable_drivers'),
+   ).stdout().split()
 
 default_cflags = machine_args
 if cc.has_argument('-Wno-format-truncation')
-- 
2.20.1

[dpdk-dev] [PATCH v3 2/5] net/mlx4: use mlx4 debug flag instead of NDEBUG

[Alexander Kozyrev]

Use the RTE_LIBRTE_MLX4_DEBUG compilation flag to get rid of dependency
on the NDEBUG definition. This is a preparation step to switch
from standard assert clauses to DPDK RTE_ASSERT ones in MLX4 driver.

Signed-off-by: Alexander Kozyrev 
Acked-by: Viacheslav Ovsiienko 
---
 drivers/net/mlx4/Makefile |  4 ++--
 drivers/net/mlx4/meson.build  |  4 ++--
 drivers/net/mlx4/mlx4.c   |  4 ++--
 drivers/net/mlx4/mlx4_mr.c|  8 
 drivers/net/mlx4/mlx4_rxtx.c  | 10 +-
 drivers/net/mlx4/mlx4_utils.h |  8 
 6 files changed, 19 insertions(+), 19 deletions(-)

diff --git a/drivers/net/mlx4/Makefile b/drivers/net/mlx4/Makefile
index 329569d..c5543ed 100644
--- a/drivers/net/mlx4/Makefile
+++ b/drivers/net/mlx4/Makefile
@@ -65,13 +65,13 @@ endif
 
 # User-defined CFLAGS.
 ifeq ($(CONFIG_RTE_LIBRTE_MLX4_DEBUG),y)
-CFLAGS += -pedantic -UNDEBUG
+CFLAGS += -pedantic
 ifneq ($(CONFIG_RTE_TOOLCHAIN_ICC),y)
 CFLAGS += -DPEDANTIC
 endif
 AUTO_CONFIG_CFLAGS += -Wno-pedantic
 else
-CFLAGS += -DNDEBUG -UPEDANTIC
+CFLAGS += -UPEDANTIC
 endif
 
 include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/net/mlx4/meson.build b/drivers/net/mlx4/meson.build
index 9eb4988..0260c5d 100644
--- a/drivers/net/mlx4/meson.build
+++ b/drivers/net/mlx4/meson.build
@@ -67,9 +67,9 @@ if build
endif
endforeach
if get_option('buildtype').contains('debug')
-   cflags += [ '-pedantic', '-UNDEBUG', '-DPEDANTIC' ]
+   cflags += [ '-pedantic', '-DPEDANTIC' ]
else
-   cflags += [ '-DNDEBUG', '-UPEDANTIC' ]
+   cflags += [ '-UPEDANTIC' ]
endif
# To maintain the compatibility with the make build system
# mlx4_autoconf.h file is still generated.
diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/mlx4.c
index ab5e6c6..122d37c 100644
--- a/drivers/net/mlx4/mlx4.c
+++ b/drivers/net/mlx4/mlx4.c
@@ -305,7 +305,7 @@ struct mlx4_conf {
  (void *)dev, strerror(-ret));
goto err;
}
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX4_DEBUG
mlx4_mr_dump_dev(dev);
 #endif
ret = mlx4_rxq_intr_enable(priv);
@@ -1305,7 +1305,7 @@ struct mlx4_conf {
return;
assert(mlx4_glue);
 #endif
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX4_DEBUG
/* Glue structure must not contain any NULL pointers. */
{
unsigned int i;
diff --git a/drivers/net/mlx4/mlx4_mr.c b/drivers/net/mlx4/mlx4_mr.c
index 470cee0..4b2e8fe 100644
--- a/drivers/net/mlx4/mlx4_mr.c
+++ b/drivers/net/mlx4/mlx4_mr.c
@@ -242,7 +242,7 @@ struct mr_update_mp_data {
memset(bt, 0, sizeof(*bt));
 }
 
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX4_DEBUG
 /**
  * Dump all the entries in a B-tree
  *
@@ -962,7 +962,7 @@ struct mr_update_mp_data {
rte_smp_wmb();
}
rte_rwlock_write_unlock(&priv->mr.rwlock);
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX4_DEBUG
if (rebuild)
mlx4_mr_dump_dev(dev);
 #endif
@@ -1380,7 +1380,7 @@ struct mr_update_mp_data {
return data.ret;
 }
 
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX4_DEBUG
 /**
  * Dump all the created MRs and the global cache entries.
  *
@@ -1440,7 +1440,7 @@ struct mr_update_mp_data {
rte_rwlock_write_lock(&mlx4_shared_data->mem_event_rwlock);
LIST_REMOVE(priv, mem_event_cb);
rte_rwlock_write_unlock(&mlx4_shared_data->mem_event_rwlock);
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX4_DEBUG
mlx4_mr_dump_dev(dev);
 #endif
rte_rwlock_write_lock(&priv->mr.rwlock);
diff --git a/drivers/net/mlx4/mlx4_rxtx.c b/drivers/net/mlx4/mlx4_rxtx.c
index 4dc0adb..b56436e 100644
--- a/drivers/net/mlx4/mlx4_rxtx.c
+++ b/drivers/net/mlx4/mlx4_rxtx.c
@@ -321,7 +321,7 @@ struct tso_info {
if (unlikely(!!(cqe->owner_sr_opcode & MLX4_CQE_OWNER_MASK) ^
!!(cons_index & cq->cqe_cnt)))
break;
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX4_DEBUG
/*
 * Make sure we read the CQE after we read the ownership bit.
 */
@@ -336,7 +336,7 @@ struct tso_info {
  cqe_err->syndrome);
break;
}
-#endif /* NDEBUG */
+#endif /* RTE_LIBRTE_MLX4_DEBUG */
cons_index++;
} while (1);
completed = (cons_index - cq->cons_index) * txq->elts_comp_cd_init;
@@ -488,14 +488,14 @@ struct tso_info {
((uintptr_t)dseg & (MLX4_TXBB_SIZE - 1))) >>
   MLX4_SEG_SHIFT;
switch (nb_segs_txbb) {
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX4_DEBUG
default:
/* Should never happen. */
rte_panic("%p: Invalid number of SGEs(%d) for a TXBB",
(void *)txq, nb_segs_txbb);
/* rte_panic never returns. */
break;
-#endif /* NDEBUG */
+#end

[dpdk-dev] [PATCH v3 4/5] net/mlx5: use mlx5 debug flag instead of NDEBUG

[Alexander Kozyrev]

Use the RTE_LIBRTE_MLX5_DEBUG configuration flag to get rid of dependency
on the NDEBUG definition. This is a preparation step to switch
from standard assert clauses to DPDK RTE_ASSERT ones in MLX5 driver.

Signed-off-by: Alexander Kozyrev 
Acked-by: Viacheslav Ovsiienko 
---
 drivers/net/mlx5/Makefile   |  4 ++--
 drivers/net/mlx5/meson.build|  4 ++--
 drivers/net/mlx5/mlx5.c |  6 +++---
 drivers/net/mlx5/mlx5_flow_dv.c |  2 +-
 drivers/net/mlx5/mlx5_mr.c  |  4 ++--
 drivers/net/mlx5/mlx5_nl.c  |  2 +-
 drivers/net/mlx5/mlx5_rxtx.c|  8 
 drivers/net/mlx5/mlx5_rxtx.h|  6 +++---
 drivers/net/mlx5/mlx5_txq.c |  2 +-
 drivers/net/mlx5/mlx5_utils.h   | 14 +++---
 10 files changed, 26 insertions(+), 26 deletions(-)

diff --git a/drivers/net/mlx5/Makefile b/drivers/net/mlx5/Makefile
index 0466d9d..1626860 100644
--- a/drivers/net/mlx5/Makefile
+++ b/drivers/net/mlx5/Makefile
@@ -85,13 +85,13 @@ endif
 
 # User-defined CFLAGS.
 ifeq ($(CONFIG_RTE_LIBRTE_MLX5_DEBUG),y)
-CFLAGS += -pedantic -UNDEBUG
+CFLAGS += -pedantic
 ifneq ($(CONFIG_RTE_TOOLCHAIN_ICC),y)
 CFLAGS += -DPEDANTIC
 endif
 AUTO_CONFIG_CFLAGS += -Wno-pedantic
 else
-CFLAGS += -DNDEBUG -UPEDANTIC
+CFLAGS += -UPEDANTIC
 endif
 
 include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/net/mlx5/meson.build b/drivers/net/mlx5/meson.build
index 3ad4f02..c7b835b 100644
--- a/drivers/net/mlx5/meson.build
+++ b/drivers/net/mlx5/meson.build
@@ -84,9 +84,9 @@ if build
endif
endforeach
if get_option('buildtype').contains('debug')
-   cflags += [ '-pedantic', '-UNDEBUG', '-DPEDANTIC' ]
+   cflags += [ '-pedantic', '-DPEDANTIC' ]
else
-   cflags += [ '-DNDEBUG', '-UPEDANTIC' ]
+   cflags += [ '-UPEDANTIC' ]
endif
# To maintain the compatibility with the make build system
# mlx5_autoconf.h file is still generated.
diff --git a/drivers/net/mlx5/mlx5.c b/drivers/net/mlx5/mlx5.c
index ffee39c..6c6a66d 100644
--- a/drivers/net/mlx5/mlx5.c
+++ b/drivers/net/mlx5/mlx5.c
@@ -654,7 +654,7 @@ struct mlx5_flow_id_pool *
 mlx5_free_shared_ibctx(struct mlx5_ibv_shared *sh)
 {
pthread_mutex_lock(&mlx5_ibv_list_mutex);
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX5_DEBUG
/* Check the object presence in the list. */
struct mlx5_ibv_shared *lctx;
 
@@ -2612,7 +2612,7 @@ struct mlx5_flow_id_pool *
mac.addr_bytes[0], mac.addr_bytes[1],
mac.addr_bytes[2], mac.addr_bytes[3],
mac.addr_bytes[4], mac.addr_bytes[5]);
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX5_DEBUG
{
char ifname[IF_NAMESIZE];
 
@@ -3658,7 +3658,7 @@ struct mlx5_flow_id_pool *
return;
assert(mlx5_glue);
 #endif
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX5_DEBUG
/* Glue structure must not contain any NULL pointers. */
{
unsigned int i;
diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
index 5a1b426..2c0cb43 100644
--- a/drivers/net/mlx5/mlx5_flow_dv.c
+++ b/drivers/net/mlx5/mlx5_flow_dv.c
@@ -5145,7 +5145,7 @@ struct field_modify_info modify_tcp[] = {
return dev_flow;
 }
 
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX5_DEBUG
 /**
  * Sanity check for match mask and value. Similar to check_valid_spec() in
  * kernel driver. If unmasked bit is present in value, it returns failure.
diff --git a/drivers/net/mlx5/mlx5_mr.c b/drivers/net/mlx5/mlx5_mr.c
index 0d549b6..07ce8df 100644
--- a/drivers/net/mlx5/mlx5_mr.c
+++ b/drivers/net/mlx5/mlx5_mr.c
@@ -240,7 +240,7 @@ struct mr_update_mp_data {
 void
 mlx5_mr_btree_dump(struct mlx5_mr_btree *bt __rte_unused)
 {
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX5_DEBUG
int idx;
struct mlx5_mr_cache *lkp_tbl;
 
@@ -1551,7 +1551,7 @@ struct mr_update_mp_data {
 void
 mlx5_mr_dump_dev(struct mlx5_ibv_shared *sh __rte_unused)
 {
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX5_DEBUG
struct mlx5_mr *mr;
int mr_n = 0;
int chunk_n = 0;
diff --git a/drivers/net/mlx5/mlx5_nl.c b/drivers/net/mlx5/mlx5_nl.c
index e7ba034..ce37460 100644
--- a/drivers/net/mlx5/mlx5_nl.c
+++ b/drivers/net/mlx5/mlx5_nl.c
@@ -356,7 +356,7 @@ struct mlx5_nl_ifindex_data {
rte_errno = ENOMEM;
return -rte_errno;
}
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX5_DEBUG
char m[18];
 
rte_ether_format_addr(m, 18, RTA_DATA(attribute));
diff --git a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c
index 5e31f01..4b69409 100644
--- a/drivers/net/mlx5/mlx5_rxtx.c
+++ b/drivers/net/mlx5/mlx5_rxtx.c
@@ -2179,11 +2179,11 @@ enum mlx5_txcmp_code {
last->cseg.flags = RTE_BE32(MLX5_COMP_ALWAYS <<
MLX5_COMP_MODE_OFFSET);
/* Save elts_head in dedicated free on completion queue. */

[dpdk-dev] [PATCH v3 0/5] net/mlx: assert cleanup in mlx drivers

[Alexander Kozyrev]

The Mellanox PMD uses the NDEBUG definition to control built-in debug
features including the asserting one. The DPDK uses a bit different
approach and provides RTE_ASSERT macro and appropriate global
configuration option: CONFIG_RTE_ENABLE_ASSERT.
The patch set introduces the MLX_ASSERT macros that allows to follow
the DPDK approach in a unified fashion and, at the same time,
provides the opportunity to turn on Mellanox PMD assert feature with
dedicated local configuration options: CONFIG_RTE_LIBRTE_MLX_DEBUG.
Note that the direct configuration MLX_DEBUG is clearer than
double negation "ifndef NDEBUG" used before.
This patch set triggers another false positive warning with ICC.
To spare future development efforts we disable the treatments of
compilation warnings as errors in ICC config. GCC stays as a guard.

Signed-off-by: Alexander Kozyrev 
---
v1: http://patches.dpdk.org/cover/65082/
v2: http://patches.dpdk.org/cover/65089/ 
 - Corrects typos and make commit messages more accurate.
 - Fixes broken compilation due to an undefined function in debug mode.
v3:
 - Uses RTE_LIBRTE_MLX_DEBUG directly instead of MLX_DEBUG

Alexander Kozyrev (5):
  mk/icc: disable treatment of warnings as errors
  net/mlx4: use mlx4 debug flag instead of NDEBUG
  net/mlx4: introduce the mlx4 version of the assert
  net/mlx5: use mlx5 debug flag instead of NDEBUG
  net/mlx5: introduce the mlx5 version of the assert

 drivers/net/mlx4/Makefile|   4 +-
 drivers/net/mlx4/meson.build |   4 +-
 drivers/net/mlx4/mlx4.c  |  29 ++--
 drivers/net/mlx4/mlx4_ethdev.c   |   5 +-
 drivers/net/mlx4/mlx4_flow.c |  34 ++--
 drivers/net/mlx4/mlx4_intr.c |   3 +-
 drivers/net/mlx4/mlx4_mp.c   |  25 ++-
 drivers/net/mlx4/mlx4_mr.c   |  74 
 drivers/net/mlx4/mlx4_rxq.c  |  53 +++---
 drivers/net/mlx4/mlx4_rxtx.c |  29 ++--
 drivers/net/mlx4/mlx4_txq.c  |  17 +-
 drivers/net/mlx4/mlx4_utils.c|   3 +-
 drivers/net/mlx4/mlx4_utils.h|  15 +-
 drivers/net/mlx5/Makefile|   4 +-
 drivers/net/mlx5/meson.build |   4 +-
 drivers/net/mlx5/mlx5.c  |  83 +
 drivers/net/mlx5/mlx5_devx_cmds.c|   4 +-
 drivers/net/mlx5/mlx5_ethdev.c   |  69 
 drivers/net/mlx5/mlx5_flow.c |  69 
 drivers/net/mlx5/mlx5_flow_dv.c  |  97 ++-
 drivers/net/mlx5/mlx5_flow_meter.c   |  12 +-
 drivers/net/mlx5/mlx5_flow_verbs.c   |   4 +-
 drivers/net/mlx5/mlx5_mac.c  |   5 +-
 drivers/net/mlx5/mlx5_mp.c   |  29 ++--
 drivers/net/mlx5/mlx5_mr.c   |  71 
 drivers/net/mlx5/mlx5_nl.c   |  26 +--
 drivers/net/mlx5/mlx5_prm.h  |   3 +-
 drivers/net/mlx5/mlx5_rss.c  |   3 +-
 drivers/net/mlx5/mlx5_rxq.c  |  41 +++--
 drivers/net/mlx5/mlx5_rxtx.c | 289 ---
 drivers/net/mlx5/mlx5_rxtx.h |   6 +-
 drivers/net/mlx5/mlx5_rxtx_vec.c |   1 -
 drivers/net/mlx5/mlx5_rxtx_vec.h |   7 +-
 drivers/net/mlx5/mlx5_rxtx_vec_altivec.h |  11 +-
 drivers/net/mlx5/mlx5_rxtx_vec_neon.h|  11 +-
 drivers/net/mlx5/mlx5_rxtx_vec_sse.h |  11 +-
 drivers/net/mlx5/mlx5_socket.c   |   4 +-
 drivers/net/mlx5/mlx5_stats.c|   2 +-
 drivers/net/mlx5/mlx5_txq.c  |  59 ---
 drivers/net/mlx5/mlx5_utils.c|   8 +-
 drivers/net/mlx5/mlx5_utils.h|  36 ++--
 drivers/net/mlx5/mlx5_vlan.c |   7 +-
 mk/toolchain/icc/rte.vars.mk |   4 -
 43 files changed, 632 insertions(+), 643 deletions(-)

-- 
1.8.3.1

[dpdk-dev] [PATCH v3 1/5] mk/icc: disable treatment of warnings as errors

[Alexander Kozyrev]

Remove -Werror-all flag in ICC configuration file to stop treating ICC
warnings as errors in DPDK due to many false positives. We are using
GCC and Clang as a benchmark for warnings anyway for simplification.

Suggested-by: Thomas Monjalon 
Signed-off-by: Alexander Kozyrev 
Acked-by: Viacheslav Ovsiienko 
Acked-by: Thomas Monjalon 
---
 mk/toolchain/icc/rte.vars.mk | 4 
 1 file changed, 4 deletions(-)

diff --git a/mk/toolchain/icc/rte.vars.mk b/mk/toolchain/icc/rte.vars.mk
index 8aa87aa..1729f3d 100644
--- a/mk/toolchain/icc/rte.vars.mk
+++ b/mk/toolchain/icc/rte.vars.mk
@@ -47,10 +47,6 @@ WERROR_FLAGS += -diag-disable 13368 -diag-disable 15527
 WERROR_FLAGS += -diag-disable 188
 WERROR_FLAGS += -diag-disable 11074 -diag-disable 11076 -Wdeprecated
 
-ifeq ($(RTE_DEVEL_BUILD),y)
-WERROR_FLAGS += -Werror-all
-endif
-
 # process cpu flags
 include $(RTE_SDK)/mk/toolchain/$(RTE_TOOLCHAIN)/rte.toolchain-compat.mk
 
-- 
1.8.3.1

[dpdk-dev] [PATCH v3 5/5] net/mlx5: introduce the mlx5 version of the assert

[Alexander Kozyrev]

Use the MLX5_ASSERT macros instead of the standard assert clause.
Depends on the RTE_LIBRTE_MLX5_DEBUG configuration option to define it.
If RTE_LIBRTE_MLX5_DEBUG is enabled MLX5_ASSERT is equal to RTE_VERIFY
to bypass the global CONFIG_RTE_ENABLE_ASSERT option.
If RTE_LIBRTE_MLX5_DEBUG is disabled, the global CONFIG_RTE_ENABLE_ASSERT
can still make this assert active by calling RTE_VERIFY inside RTE_ASSERT.

Signed-off-by: Alexander Kozyrev 
Acked-by: Viacheslav Ovsiienko 
---
 drivers/net/mlx5/mlx5.c  |  77 +
 drivers/net/mlx5/mlx5_devx_cmds.c|   4 +-
 drivers/net/mlx5/mlx5_ethdev.c   |  69 
 drivers/net/mlx5/mlx5_flow.c |  69 
 drivers/net/mlx5/mlx5_flow_dv.c  |  95 ++-
 drivers/net/mlx5/mlx5_flow_meter.c   |  12 +-
 drivers/net/mlx5/mlx5_flow_verbs.c   |   4 +-
 drivers/net/mlx5/mlx5_mac.c  |   5 +-
 drivers/net/mlx5/mlx5_mp.c   |  29 ++--
 drivers/net/mlx5/mlx5_mr.c   |  67 
 drivers/net/mlx5/mlx5_nl.c   |  24 +--
 drivers/net/mlx5/mlx5_prm.h  |   3 +-
 drivers/net/mlx5/mlx5_rss.c  |   3 +-
 drivers/net/mlx5/mlx5_rxq.c  |  41 +++--
 drivers/net/mlx5/mlx5_rxtx.c | 281 ---
 drivers/net/mlx5/mlx5_rxtx_vec.c |   1 -
 drivers/net/mlx5/mlx5_rxtx_vec.h |   7 +-
 drivers/net/mlx5/mlx5_rxtx_vec_altivec.h |  11 +-
 drivers/net/mlx5/mlx5_rxtx_vec_neon.h|  11 +-
 drivers/net/mlx5/mlx5_rxtx_vec_sse.h |  11 +-
 drivers/net/mlx5/mlx5_socket.c   |   4 +-
 drivers/net/mlx5/mlx5_stats.c|   2 +-
 drivers/net/mlx5/mlx5_txq.c  |  57 +++
 drivers/net/mlx5/mlx5_utils.c|   8 +-
 drivers/net/mlx5/mlx5_utils.h|  22 ++-
 drivers/net/mlx5/mlx5_vlan.c |   7 +-
 26 files changed, 462 insertions(+), 462 deletions(-)

diff --git a/drivers/net/mlx5/mlx5.c b/drivers/net/mlx5/mlx5.c
index 6c6a66d..99d4c94 100644
--- a/drivers/net/mlx5/mlx5.c
+++ b/drivers/net/mlx5/mlx5.c
@@ -6,7 +6,6 @@
 #include 
 #include 
 #include 
-#include 
 #include 
 #include 
 #include 
@@ -290,7 +289,7 @@ struct mlx5_flow_id_pool *
if (pool->curr == pool->last) {
size = pool->curr - pool->free_arr;
size2 = size * MLX5_ID_GENERATION_ARRAY_FACTOR;
-   assert(size2 > size);
+   MLX5_ASSERT(size2 > size);
mem = rte_malloc("", size2 * sizeof(uint32_t), 0);
if (!mem) {
DRV_LOG(ERR, "can't allocate mem for id pool");
@@ -436,7 +435,7 @@ struct mlx5_flow_id_pool *
char *env;
int value;
 
-   assert(rte_eal_process_type() == RTE_PROC_PRIMARY);
+   MLX5_ASSERT(rte_eal_process_type() == RTE_PROC_PRIMARY);
/* Get environment variable to store. */
env = getenv(MLX5_SHUT_UP_BF);
value = env ? !!strcmp(env, "0") : MLX5_ARG_UNSET;
@@ -451,7 +450,7 @@ struct mlx5_flow_id_pool *
 static void
 mlx5_restore_doorbell_mapping_env(int value)
 {
-   assert(rte_eal_process_type() == RTE_PROC_PRIMARY);
+   MLX5_ASSERT(rte_eal_process_type() == RTE_PROC_PRIMARY);
/* Restore the original environment variable state. */
if (value == MLX5_ARG_UNSET)
unsetenv(MLX5_SHUT_UP_BF);
@@ -491,9 +490,9 @@ struct mlx5_flow_id_pool *
struct mlx5_devx_tis_attr tis_attr = { 0 };
 #endif
 
-   assert(spawn);
+   MLX5_ASSERT(spawn);
/* Secondary process should not create the shared context. */
-   assert(rte_eal_process_type() == RTE_PROC_PRIMARY);
+   MLX5_ASSERT(rte_eal_process_type() == RTE_PROC_PRIMARY);
pthread_mutex_lock(&mlx5_ibv_list_mutex);
/* Search for IB context by device name. */
LIST_FOREACH(sh, &mlx5_ibv_list, next) {
@@ -503,7 +502,7 @@ struct mlx5_flow_id_pool *
}
}
/* No device found, we have to create new shared context. */
-   assert(spawn->max_port);
+   MLX5_ASSERT(spawn->max_port);
sh = rte_zmalloc("ethdev shared ib context",
 sizeof(struct mlx5_ibv_shared) +
 spawn->max_port *
@@ -626,7 +625,7 @@ struct mlx5_flow_id_pool *
return sh;
 error:
pthread_mutex_unlock(&mlx5_ibv_list_mutex);
-   assert(sh);
+   MLX5_ASSERT(sh);
if (sh->tis)
claim_zero(mlx5_devx_cmd_destroy(sh->tis));
if (sh->td)
@@ -638,7 +637,7 @@ struct mlx5_flow_id_pool *
if (sh->flow_id_pool)
mlx5_flow_id_pool_release(sh->flow_id_pool);
rte_free(sh);
-   assert(err > 0);
+   MLX5_ASSERT(err > 0);
rte_errno = err;
return NULL;
 }
@@ -661,16 +660,16 @@ struct mlx5_flow_id_pool *
LIST_FOREACH(lctx, &mlx5_ibv_list, next)
if (lctx == sh)
break;
-   assert(lctx);
+   MLX5_ASSERT(lctx);
  

[dpdk-dev] [PATCH v3 3/5] net/mlx4: introduce the mlx4 version of the assert

[Alexander Kozyrev]

Use the MLX4_ASSERT macros instead of the standard assert clause.
Depends on the RTE_LIBRTE_MLX4_DEBUG configuration option to define it.
If RTE_LIBRTE_MLX4_DEBUG is enabled MLX4_ASSERT is equal to RTE_VERIFY
to bypass the global CONFIG_RTE_ENABLE_ASSERT option.
If RTE_LIBRTE_MLX4_DEBUG is disabled, the global CONFIG_RTE_ENABLE_ASSERT
can still make this assert active by calling RTE_VERIFY inside RTE_ASSERT.

Signed-off-by: Alexander Kozyrev 
Acked-by: Viacheslav Ovsiienko 
---
 drivers/net/mlx4/mlx4.c| 25 
 drivers/net/mlx4/mlx4_ethdev.c |  5 ++--
 drivers/net/mlx4/mlx4_flow.c   | 34 +++---
 drivers/net/mlx4/mlx4_intr.c   |  3 +-
 drivers/net/mlx4/mlx4_mp.c | 25 
 drivers/net/mlx4/mlx4_mr.c | 66 +-
 drivers/net/mlx4/mlx4_rxq.c| 53 +
 drivers/net/mlx4/mlx4_rxtx.c   | 19 ++--
 drivers/net/mlx4/mlx4_txq.c| 17 +--
 drivers/net/mlx4/mlx4_utils.c  |  3 +-
 drivers/net/mlx4/mlx4_utils.h  |  7 +++--
 11 files changed, 125 insertions(+), 132 deletions(-)

diff --git a/drivers/net/mlx4/mlx4.c b/drivers/net/mlx4/mlx4.c
index 122d37c..5522eb5 100644
--- a/drivers/net/mlx4/mlx4.c
+++ b/drivers/net/mlx4/mlx4.c
@@ -8,7 +8,6 @@
  * mlx4 driver initialization.
  */
 
-#include 
 #include 
 #include 
 #include 
@@ -162,7 +161,7 @@ struct mlx4_conf {
 
socket = rxq->socket;
}
-   assert(data != NULL);
+   MLX4_ASSERT(data != NULL);
ret = rte_malloc_socket(__func__, size, alignment, socket);
if (!ret && size)
rte_errno = ENOMEM;
@@ -180,7 +179,7 @@ struct mlx4_conf {
 static void
 mlx4_free_verbs_buf(void *ptr, void *data __rte_unused)
 {
-   assert(data != NULL);
+   MLX4_ASSERT(data != NULL);
rte_free(ptr);
 }
 #endif
@@ -392,11 +391,11 @@ struct mlx4_conf {
mlx4_proc_priv_uninit(dev);
mlx4_mr_release(dev);
if (priv->pd != NULL) {
-   assert(priv->ctx != NULL);
+   MLX4_ASSERT(priv->ctx != NULL);
claim_zero(mlx4_glue->dealloc_pd(priv->pd));
claim_zero(mlx4_glue->close_device(priv->ctx));
} else
-   assert(priv->ctx == NULL);
+   MLX4_ASSERT(priv->ctx == NULL);
mlx4_intr_uninstall(priv);
memset(priv, 0, sizeof(*priv));
 }
@@ -705,7 +704,7 @@ struct mlx4_conf {
if (mlx4_init_shared_data())
return -rte_errno;
sd = mlx4_shared_data;
-   assert(sd);
+   MLX4_ASSERT(sd);
rte_spinlock_lock(&sd->lock);
switch (rte_eal_process_type()) {
case RTE_PROC_PRIMARY:
@@ -775,16 +774,16 @@ struct mlx4_conf {
  strerror(rte_errno));
return -rte_errno;
}
-   assert(pci_drv == &mlx4_driver);
+   MLX4_ASSERT(pci_drv == &mlx4_driver);
list = mlx4_glue->get_device_list(&i);
if (list == NULL) {
rte_errno = errno;
-   assert(rte_errno);
+   MLX4_ASSERT(rte_errno);
if (rte_errno == ENOSYS)
ERROR("cannot list devices, is ib_uverbs loaded?");
return -rte_errno;
}
-   assert(i >= 0);
+   MLX4_ASSERT(i >= 0);
/*
 * For each listed device, check related sysfs entry against
 * the provided PCI ID.
@@ -821,7 +820,7 @@ struct mlx4_conf {
ERROR("cannot use device, are drivers up to date?");
return -rte_errno;
}
-   assert(err > 0);
+   MLX4_ASSERT(err > 0);
rte_errno = err;
return -rte_errno;
}
@@ -846,7 +845,7 @@ struct mlx4_conf {
err = ENODEV;
goto error;
}
-   assert(device_attr.max_sge >= MLX4_MAX_SGE);
+   MLX4_ASSERT(device_attr.max_sge >= MLX4_MAX_SGE);
for (i = 0; i < device_attr.phys_port_cnt; i++) {
uint32_t port = i + 1; /* ports are indexed from one */
struct ibv_context *ctx = NULL;
@@ -1303,7 +1302,7 @@ struct mlx4_conf {
 #ifdef RTE_IBVERBS_LINK_DLOPEN
if (mlx4_glue_init())
return;
-   assert(mlx4_glue);
+   MLX4_ASSERT(mlx4_glue);
 #endif
 #ifdef RTE_LIBRTE_MLX4_DEBUG
/* Glue structure must not contain any NULL pointers. */
@@ -1311,7 +1310,7 @@ struct mlx4_conf {
unsigned int i;
 
for (i = 0; i != sizeof(*mlx4_glue) / sizeof(void *); ++i)
-   assert(((const void *const *)mlx4_glue)[i]);
+   MLX4_ASSERT(((const void *const *)mlx4_glue)[i]);
}
 #endif
if (strcmp(mlx4_glue->version, MLX4_GLUE_VERSION)) {
diff --git a/drivers/net/mlx4/mlx4_ethdev.c b/drivers/net/mlx4/mlx4_ethdev.c
index dfb24c2..e500ec4 100644
--- a/drivers/net/mlx4/mlx4_ethdev.c
+++ b/drivers/net/mlx4/mlx4_ethdev.c

Re: [dpdk-dev] [PATCH v2 00/15] add OCTEONTX2 inline IPsec support

[Anoob Joseph]

Hi Jerin, Akhil,

Let me summarize the design changes from the discussions below.

Currently, drivers/crypto/octeontx2/otx2_security.c defines all security ctx 
ops for the ethdev (idea was to add all crypto security ctx for lookaside also 
there). That will be moved to drivers/net/octeontx2 as is. The routines which 
are doing qp_add & qp_remove would be moved to common (discussed below). 
Otherwise, the rest should remain as is. If Jerin/Akhil wants further 
isolation, please do share specifics. Almost all functions in otx2_security.c 
is dereferencing 'rte_eth_dev'. So having (void *) will not help.

The functions in otx2_security.c is calling inline functions in otx2_ipsec_fp.h 
(which has lower level implementations of session create etc). This will remain 
as is in drivers/crypto/octeontx2 but would be called from 
drivers/net/octeontx2/otx2_security.c.

We will need to include otx2_cryptodev_qp.h (internal header in 
drivers/crypto/octeontx2) since the crypto queue pair is required for outbound 
processing. Since otx2_cryptodev_qp.h has dependency on rte_cryptodev.h, the 
ethdev file will have dependency on rte_cryptodev.h.

I want all the maintainers (Akhil, Jerin & Ferruh) to ack the above behavior so 
that I can proceed with the restructuring. (Currently issue is rte_ethdev.h 
getting included in a cryptodev PMD file. The case we are proposing is the 
exact mirror of that)

Currently, the cryptodev has to do qp-eth port mapping and save it somewhere 
for eth dev to use during security session create. This will have to be saved 
in drivers/common/octeontx2.

@Ferruh, do you agree with the suggestions here? With the proposed changes, 
parts of the patches would go into ethdev space (for reviews and merge), and 
the rest would be in crypto space.

Thanks,
Anoob

> -Original Message-
> From: Akhil Goyal 
> Sent: Monday, January 27, 2020 5:18 PM
> To: Jerin Jacob 
> Cc: Anoob Joseph ; Declan Doherty
> ; Thomas Monjalon ; Jerin
> Jacob Kollanukkaran ; Narayana Prasad Raju Athreya
> ; Kiran Kumar Kokkilagadda
> ; Nithin Kumar Dabilpuram
> ; Pavan Nikhilesh Bhagavatula
> ; Ankur Dwivedi ;
> Archana Muniganti ; Tejasree Kondoj
> ; Vamsi Krishna Attunuru ;
> Lukas Bartosik ; dpdk-dev 
> Subject: [EXT] RE: [dpdk-dev] [PATCH v2 00/15] add OCTEONTX2 inline IPsec
> support
> 
> External Email
> 
> --
> Hi Jerin,
> 
> >
> > On Mon, Jan 27, 2020 at 4:10 PM Akhil Goyal  wrote:
> > >
> > >
> > > >
> > > > On Wed, Jan 22, 2020 at 6:26 PM Akhil Goyal 
> wrote:
> > > > >
> > > > > > > >
> > > > > > > > Hi Jerin,
> > > > > > > >
> > > > > > > > Will  do the suggested change (RX/rx-> Rx & TX/tx->Tx). Do
> > > > > > > > you want
> > me
> > > > > > > > to trim the headline as well?
> > > > > > > >
> > > > > > >
> > > > > > > Hi Anoob,
> > > > > > >
> > > > > > > > @Akhil, did you get a chance to review the series? Do you
> > > > > > > > have any comments on the patches?
> > > > > > > >
> > > > > > >
> > > > > > > You are adding inline ipsec support to ethernet device and
> > > > > > > not a crypto device.
> > > > > > > These patches should not be part of crypto PMD. There will
> > > > > > > be cyclic dependency Between ethernet device and crypto
> > > > > > > device which can be
> > > > easily
> > > > > > > avoided.
> > > > > >
> > > > > > [Anoob] We have plans to use lookaside protocol to handle the
> "fallback"
> > > > > > session. And that involves session sharing between inline and
> > > > > > lookaside
> > > > protocol
> > > > > > offloads. Also, though the feature is exposed as a feature of
> > > > > > ethdev, on
> > our
> > > > > > platform, it's the crypto block which primarily implements the 
> > > > > > feature.
> > And
> > > > so, if
> > > > > > the code is moved to ethdev dir, there would be lot of code 
> > > > > > duplication.
> > The
> > > > > > idea is to have all security related code in one place.
> > > > > >
> > > > > > Also, the PMDs don't have any calls to each other. The
> > > > > > communication
> > > > between
> > > > > > the two happens via common. The crypto dev PMD will register
> > > > > > the
> > required
> > > > > > security ops to a common structure and ethdev would get it from 
> > > > > > there.
> > So
> > > > there
> > > > > > won't be an issue of build dependency.
> > > > > >
> > > > >
> > > > > - The code that need to be duplicated can be moved to
> > > > > drivers/common/
> > > >
> > > > I would like to keep the common code that is common to all the
> > coprocessors.
> > > >
> > > > Moreover, there are logistic issues in that case where
> > > > a) drivers/common/octeontx2/ going through master repo. So we will
> > > > be creating unnecessary dependency with that 'master' tree.
> > > > b) crypto and ethdev work is done by different teams so we would
> > > > like to make responsibly clear wrt the review and ownership.
> > >
> > > I can agree upon that the code shall not be placed in the
> > > common/octeontx2, B

Re: [dpdk-dev] [dpdk-web] DPDK Programmer's Guide :ERRATA

[Honnappa Nagarahalli]

Thanks Prateek for the comments. Agree with both the corrections. Would you 
like to submit a patch?

Thanks,
Honnappa

> -Original Message-
> From: Thomas Monjalon 
> Sent: Sunday, January 26, 2020 3:36 PM
> To: prateekag 
> Cc: dev@dpdk.org; Honnappa Nagarahalli 
> Subject: Re: [dpdk-web] DPDK Programmer's Guide :ERRATA
> 
> Thanks for the comments.
> I did not look at this doc but I'm sure Honnappa will have answers.
> 
> Note: you may want to update the doc yourself by making a patch for the
> DPDK git.
> 
> Rerouting this email to dpdk-dev and adding Honnappa.
> 
> 26/01/2020 19:46, prateekag:
> > Dear,
> >
> > I am currently going through the DPDK programmer's guide 20.02-rc1
> > updated on January 21, 2020, as a part of my Master's Project.
> > I found the following errors in the documentation.
> >   a) In section 5.1 (What is quiescent state): In paragraph 5, The
> > line "So reader thread 1 will not have a reference to the deleted
> > entry" should be replaced with "So reader thread 3 will not have a
> > reference to the deleted entry." We are talking about reader thread 3
> > and its behavior in this paragraph and not reader thread 1."
> > b) If what I am thinking is right, this one is a much serious error.
> > In section 6.5.3 Multiple Producers enqueue Figure 6.10, the global
> > prod_head(ring-> prod_head) should point one place to the right of the
> > obj5 (i.e. shift it from its current position to one place right) As
> > in third step, core 2 CAS is successful as mention in the text.
> > If I am not mailing to the correct mailing list, kindly forward the
> > message to the correct mailing list.
> >
> > Regards
> > Prateek Agarwal
> 
> 

Re: [dpdk-dev] [PATCH v2 1/5] mk/icc: disable treatment of warnings as errors

[Ferruh Yigit]

On 1/24/2020 7:37 PM, Thomas Monjalon wrote:
> 24/01/2020 17:36, Ferruh Yigit:
>> On 1/23/2020 6:20 PM, Alexander Kozyrev wrote:
>>> Remove -Werror-all flag in ICC configuration file to stop treating ICC
>>> warnings as errors in DPDK due to many false positives. We are using
>>> GCC and Clang as a benchmark for warnings anyway for simplification.
>>>
>>> Suggested-by: Thomas Monjalon 
>>> Signed-off-by: Alexander Kozyrev 
>>> Acked-by: Viacheslav Ovsiienko 
>>> Acked-by: Thomas Monjalon 
>>> ---
>>>  mk/toolchain/icc/rte.vars.mk | 4 
>>>  1 file changed, 4 deletions(-)
>>>
>>> diff --git a/mk/toolchain/icc/rte.vars.mk b/mk/toolchain/icc/rte.vars.mk
>>> index 8aa87aa..1729f3d 100644
>>> --- a/mk/toolchain/icc/rte.vars.mk
>>> +++ b/mk/toolchain/icc/rte.vars.mk
>>> @@ -47,10 +47,6 @@ WERROR_FLAGS += -diag-disable 13368 -diag-disable 15527
>>>  WERROR_FLAGS += -diag-disable 188
>>>  WERROR_FLAGS += -diag-disable 11074 -diag-disable 11076 -Wdeprecated
>>>  
>>> -ifeq ($(RTE_DEVEL_BUILD),y)
>>> -WERROR_FLAGS += -Werror-all
>>> -endif
>>
>> Not sure about removing this globally, as of now the ICC builds fine. If 
>> this is
>> for the coming changes in mlx, why not disable warnings in mlx driver only?
> 
> Adding special handling for ICC in the driver means it is kind of supported.
> ICC support is on best-effort basis as a favor to Intel and its CI.
> 
> I don't see any good reason to spend time disabling ICC warnings each time
> we hit a false positive.
> And I would like we stop doing strange things in the code to make ICC happy.
> We do not need to support ICC, or at least we do not need to make ICC build
> warning-free.
> 
> This patch will solve all future annoying issues with ICC in the future.
> 
> Now let me ask the reverse question: why the flag -Werror-all is set for ICC?
> We set this flag for gcc and clang because we use gcc and clang as static
> code analyzers (like coverity) and we want to be sure to catch any issue.
> ICC is not used as code analyzer, this is just an optimization for some
> Intel projects. I won't elaborate on the packaging and licensing of ICC...
> 
> I hope you will be convinced to acknowledge this change.
> 

To support the ICC or not is a valid question, but for me it is better to
support more compilers in case different ones catch an additional issues that is
a benefit for us, although I agree false positives are annoying, which is not
limited to icc.

"-Werror-all" is forced in DEVEL_BUILD, in this cause I was assuming to force
developer to fix the warning to increase the code quality, independent from
compiler.

As of now, all DPDK compiles with icc without warning. But we are allowing the
icc warnings just because the warnings with the change in the mlx driver. And
when we let is once, it is for sure warnings will increase by time.

If we support ICC, I am for keeping this flag and support it properly.

[dpdk-dev] [PATCH v2 3/4] build: allow hiding dependencies from pkg-config

[Thomas Monjalon]

If a dependency is required for a driver build,
but should not be exposed to the application (via pkg-config),
it can be declared in the array hidden_deps.

The hidden_deps are used as internal dependencies,
when building the shared library and the first stage of static library.
The final static library does not include the hidden_deps
because this library object is used to generate the .pc file.

Signed-off-by: Thomas Monjalon 
---
 doc/guides/contributing/coding_style.rst | 6 ++
 drivers/meson.build  | 9 ++---
 2 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/doc/guides/contributing/coding_style.rst 
b/doc/guides/contributing/coding_style.rst
index 841ef6d5c8..c96457273f 100644
--- a/doc/guides/contributing/coding_style.rst
+++ b/doc/guides/contributing/coding_style.rst
@@ -975,6 +975,12 @@ deps
 ext_deps
As above.
 
+hidden_deps
+   **Default Value = []**.
+   Used to specify external dependencies, same as ``ext_deps``, except
+   the libraries won't be exposed as requirements of the static driver,
+   in the section ``Requires.private`` of the generated ``.pc`` file.
+
 includes
**Default Value = ** Some drivers include a base
directory for additional source files and headers, so we have this
diff --git a/drivers/meson.build b/drivers/meson.build
index 29708cc2bb..759263c309 100644
--- a/drivers/meson.build
+++ b/drivers/meson.build
@@ -56,6 +56,7 @@ foreach class:dpdk_driver_classes
# too, so that it can be reflected in the pkgconfig output for
# static builds.
ext_deps = []
+   hidden_deps = []
pkgconfig_extra_libs = []
 
# pull in driver directory which should assign to each of the 
above
@@ -71,8 +72,9 @@ foreach class:dpdk_driver_classes
endforeach
if build
# get dependency objs from strings
-   shared_deps = ext_deps
-   static_deps = ext_deps
+   shared_deps = ext_deps + hidden_deps
+   static_deps = ext_deps + hidden_deps
+   static_pub_deps = ext_deps
foreach d:deps
if not is_variable('shared_rte_' + d)
build = false
@@ -82,6 +84,7 @@ foreach class:dpdk_driver_classes
else
shared_deps += 
[get_variable('shared_rte_' + d)]
static_deps += 
[get_variable('static_rte_' + d)]
+   static_pub_deps += 
[get_variable('static_rte_' + d)]
endif
endforeach
endif
@@ -144,7 +147,7 @@ foreach class:dpdk_driver_classes
sources,
objects: objs,
include_directories: includes,
-   dependencies: static_deps,
+   dependencies: static_pub_deps, # skip 
hidden_deps
c_args: cflags,
install: true)
 
-- 
2.24.1

[dpdk-dev] [PATCH v2 2/4] buildtools: get static mlx dependencies for meson

[Thomas Monjalon]

The shell script options-ibverbs-static.sh was used with make
in forcing static linkage of ibverbs libraries.

When choosing to link with a static dependency in meson,
the generated .pc file will not force such static linkage.
The solution will rely on using this script in meson.

If linking with libraries installed in a non-standard path,
an option -L is provided via EXTRA_LDFLAGS in case of using make.
With meson, tuning PKG_CONFIG_PATH for pkg-config should be enough.
When statically linking an application, the .pc file must save the
-L path so the application link will work without any extra option.
That's why --libs-only-l is replaced with --libs which includes -L.
Options which are neither -l or -L are filtered out because not needed
and can cause compilation issues with the legacy system using make.

The other change in this script is to move the main library file
(libiverbs.a) at the end of the list of dependencies.
It fixes some undefined references when linking a static application
using libdpdk.pc.

Signed-off-by: Thomas Monjalon 
---
 buildtools/options-ibverbs-static.sh | 10 --
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/buildtools/options-ibverbs-static.sh 
b/buildtools/options-ibverbs-static.sh
index 0f285a343b..43578a37f3 100755
--- a/buildtools/options-ibverbs-static.sh
+++ b/buildtools/options-ibverbs-static.sh
@@ -9,6 +9,12 @@
 #
 # PKG_CONFIG_PATH may be required to be set if libibverbs.pc is not installed.
 
-pkg-config --libs-only-l --static libibverbs |
+lib='libibverbs'
+deps='pthread|nl'
+
+pkg-config --libs --static $lib |
tr '[:space:]' '\n' |
-   sed -r '/^-l(pthread|nl)/! s,(^-l)(.*),\1:lib\2.a,'
+   sed -r "/^-l($deps)/! s,(^-l)(.*),\1:lib\2.a," |   # explicit .a
+   sed -n '/^-[Ll]/p' |   # extra link options may break with make
+   sed "/$lib/d"   # move main lib at the end
+echo -l:$lib.a
-- 
2.24.1

[dpdk-dev] [PATCH v2 0/4] add static ibverbs in meson

[Thomas Monjalon]

This is the follow-up of the feature I added one year ago:
static linkage of libibverbs in mlx PMDs.
The first implementation was focused on "make".
This second round does the same with "meson".


changes in v2:
- split mlx patch for normal addition and workarounds
- fix ldflags for ibverbs installed in a standard directory
- fix libs order leading to undefined references
- add doc for hidden_deps
- improve explanations in commit logs


Thomas Monjalon (4):
  net/mlx: add static ibverbs linkage with meson
  buildtools: get static mlx dependencies for meson
  build: allow hiding dependencies from pkg-config
  net/mlx: workaround static linkage with meson


 buildtools/meson.build   |  2 ++
 buildtools/options-ibverbs-static.sh | 10 --
 doc/guides/contributing/coding_style.rst |  6 ++
 doc/guides/nics/mlx4.rst |  4 
 doc/guides/nics/mlx5.rst |  4 
 drivers/meson.build  |  9 ++---
 drivers/net/mlx4/meson.build | 19 +++
 drivers/net/mlx5/meson.build | 19 +++
 meson_options.txt|  4 ++--
 9 files changed, 62 insertions(+), 15 deletions(-)

-- 
2.24.1

[dpdk-dev] [PATCH v2 4/4] net/mlx: workaround static linkage with meson

[Thomas Monjalon]

If ibverbs_link is static and the application choose to link DPDK
as static libraries, both PMD and ibverbs libraries must be linked
as static libraries. And the dependencies of ibverbs (netlink) must
still be linked as shared libraries.

Unfortunately, meson forget about the static requirement for ibverbs
when generating the .pc file.
As a result, libibverbs, libmlx4, libmlx5 are listed in Requires.private
section (allowing to be linked as shared libraries) and libnl is missing.

A fix is in progress for meson, but anyway we will have to live without
such a fix until a better version of meson is widely available:
https://github.com/mesonbuild/meson/pull/6393

In order to not allow meson suggesting shared libraries in the section
Requires.private of the .pc file, the cflags are passed as hidden_deps.
The list of required dependencies is replaced with ldflags, forcing
static flavor of ibverbs libraries thanks to this syntax:
-l:libfoo.a

Fixes: 83fff714bd27 ("net/mlx: add static ibverbs linkage with meson")

Signed-off-by: Thomas Monjalon 
---
 buildtools/meson.build   |  2 ++
 drivers/net/mlx4/meson.build | 14 --
 drivers/net/mlx5/meson.build | 14 --
 3 files changed, 26 insertions(+), 4 deletions(-)

diff --git a/buildtools/meson.build b/buildtools/meson.build
index cd1d054036..2a25527385 100644
--- a/buildtools/meson.build
+++ b/buildtools/meson.build
@@ -7,6 +7,8 @@ pmdinfo = find_program('gen-pmdinfo-cfile.sh')
 
 check_experimental_syms = find_program('check-experimental-syms.sh')
 
+ldflags_ibverbs_static = find_program('options-ibverbs-static.sh')
+
 # set up map-to-def script using python, either built-in or external
 python3 = import('python').find_installation(required: false)
 if python3.found()
diff --git a/drivers/net/mlx4/meson.build b/drivers/net/mlx4/meson.build
index 51d9784ee2..724256bfb4 100644
--- a/drivers/net/mlx4/meson.build
+++ b/drivers/net/mlx4/meson.build
@@ -30,16 +30,26 @@ foreach libname:libnames
lib = cc.find_library(libname, required:false)
endif
if lib.found()
-   libs += [ lib ]
+   libs += lib
+   if static_ibverbs
+   # Build without adding shared libs to Requires.private
+   hidden_deps += lib.partial_dependency(compile_args:true)
+   else
+   ext_deps += lib
+   endif
else
build = false
reason = 'missing dependency, "' + libname + '"'
endif
 endforeach
+if build and static_ibverbs
+   # Add static deps ldflags to internal apps and Libs.private
+   ldflags = run_command(ldflags_ibverbs_static, check:true).stdout()
+   ext_deps += declare_dependency(link_args:ldflags.split())
+endif
 
 if build
allow_experimental_apis = true
-   ext_deps += libs
sources = files(
'mlx4.c',
'mlx4_ethdev.c',
diff --git a/drivers/net/mlx5/meson.build b/drivers/net/mlx5/meson.build
index a4f2f79f22..e04c8947e0 100644
--- a/drivers/net/mlx5/meson.build
+++ b/drivers/net/mlx5/meson.build
@@ -30,17 +30,27 @@ foreach libname:libnames
lib = cc.find_library(libname, required:false)
endif
if lib.found()
-   libs += [ lib ]
+   libs += lib
+   if static_ibverbs
+   # Build without adding shared libs to Requires.private
+   hidden_deps += lib.partial_dependency(compile_args:true)
+   else
+   ext_deps += lib
+   endif
else
build = false
reason = 'missing dependency, "' + libname + '"'
endif
 endforeach
+if build and static_ibverbs
+   # Add static deps ldflags to internal apps and Libs.private
+   ldflags = run_command(ldflags_ibverbs_static, check:true).stdout()
+   ext_deps += declare_dependency(link_args:ldflags.split())
+endif
 
 if build
allow_experimental_apis = true
deps += ['hash']
-   ext_deps += libs
sources = files(
'mlx5.c',
'mlx5_ethdev.c',
-- 
2.24.1

[dpdk-dev] [PATCH v2 1/4] net/mlx: add static ibverbs linkage with meson

[Thomas Monjalon]

The libibverbs (and libmlx4/5) can be statically embedded
in the shared PMD library, or in the application with the static PMD.
It was supported with make build system in
commit 2c0dd7b69fb0 ("config: add static linkage of mlx dependency").

The same feature is enabled with meson when using pkg-config
(i.e. only if the call to dependency() is successful).
The fallback method for searching library with cc.find_library()
is not supported because the dependencies of the found library
would not be linked (no such info in .a file unlike .so).

The main difference, in meson build system, is the generated .pc file
giving arguments to link DPDK with the application.
Unfortunately the .pc file will not keep memory of the static linkage
option for libibverbs.

Signed-off-by: Thomas Monjalon 
---
 doc/guides/nics/mlx4.rst | 4 
 doc/guides/nics/mlx5.rst | 4 
 drivers/net/mlx4/meson.build | 5 +++--
 drivers/net/mlx5/meson.build | 5 +++--
 meson_options.txt| 4 ++--
 5 files changed, 16 insertions(+), 6 deletions(-)

diff --git a/doc/guides/nics/mlx4.rst b/doc/guides/nics/mlx4.rst
index d0e8a8b2ff..4b1d1aceb2 100644
--- a/doc/guides/nics/mlx4.rst
+++ b/doc/guides/nics/mlx4.rst
@@ -92,6 +92,10 @@ These options can be modified in the ``.config`` file.
   adds additional run-time checks and debugging messages at the cost of
   lower performance.
 
+This option is available in meson:
+
+- ``ibverbs_link`` can be ``static``, ``shared``, or ``dlopen``.
+
 Environment variables
 ~
 
diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst
index 66997f1ae7..b3bd3bba34 100644
--- a/doc/guides/nics/mlx5.rst
+++ b/doc/guides/nics/mlx5.rst
@@ -278,6 +278,10 @@ These options can be modified in the ``.config`` file.
64. Default armv8a configuration of make build and meson build set it to 128
then brings performance degradation.
 
+This option is available in meson:
+
+- ``ibverbs_link`` can be ``static``, ``shared``, or ``dlopen``.
+
 Environment variables
 ~
 
diff --git a/drivers/net/mlx4/meson.build b/drivers/net/mlx4/meson.build
index 9eb4988420..51d9784ee2 100644
--- a/drivers/net/mlx4/meson.build
+++ b/drivers/net/mlx4/meson.build
@@ -9,6 +9,7 @@ if not is_linux
 endif
 build = true
 
+static_ibverbs = (get_option('ibverbs_link') == 'static')
 pmd_dlopen = (get_option('ibverbs_link') == 'dlopen')
 LIB_GLUE_BASE = 'librte_pmd_mlx4_glue.so'
 LIB_GLUE_VERSION = '18.02.0'
@@ -24,8 +25,8 @@ endif
 libnames = [ 'mlx4', 'ibverbs' ]
 libs = []
 foreach libname:libnames
-   lib = dependency('lib' + libname, required:false)
-   if not lib.found()
+   lib = dependency('lib' + libname, static:static_ibverbs, required:false)
+   if not lib.found() and not static_ibverbs
lib = cc.find_library(libname, required:false)
endif
if lib.found()
diff --git a/drivers/net/mlx5/meson.build b/drivers/net/mlx5/meson.build
index 3ad4f02a53..a4f2f79f22 100644
--- a/drivers/net/mlx5/meson.build
+++ b/drivers/net/mlx5/meson.build
@@ -9,6 +9,7 @@ if not is_linux
 endif
 build = true
 
+static_ibverbs = (get_option('ibverbs_link') == 'static')
 pmd_dlopen = (get_option('ibverbs_link') == 'dlopen')
 LIB_GLUE_BASE = 'librte_pmd_mlx5_glue.so'
 LIB_GLUE_VERSION = '20.02.0'
@@ -24,8 +25,8 @@ endif
 libnames = [ 'mlx5', 'ibverbs' ]
 libs = []
 foreach libname:libnames
-   lib = dependency('lib' + libname, required:false)
-   if not lib.found()
+   lib = dependency('lib' + libname, static:static_ibverbs, required:false)
+   if not lib.found() and not static_ibverbs
lib = cc.find_library(libname, required:false)
endif
if lib.found()
diff --git a/meson_options.txt b/meson_options.txt
index 53dfe13c30..0a02f2faf9 100644
--- a/meson_options.txt
+++ b/meson_options.txt
@@ -12,8 +12,8 @@ option('examples', type: 'string', value: '',
description: 'Comma-separated list of examples to build by default')
 option('flexran_sdk', type: 'string', value: '',
description: 'Path to FlexRAN SDK optional Libraries for BBDEV device')
-option('ibverbs_link', type: 'combo', choices : ['shared', 'dlopen'], value: 
'shared',
-   description: 'Linkage method (shared/dlopen) for Mellanox PMDs with 
ibverbs dependencies.')
+option('ibverbs_link', type: 'combo', choices : ['static', 'shared', 
'dlopen'], value: 'shared',
+   description: 'Linkage method (static/shared/dlopen) for Mellanox PMDs 
with ibverbs dependencies.')
 option('include_subdir_arch', type: 'string', value: '',
description: 'subdirectory where to install arch-dependent headers')
 option('kernel_dir', type: 'string', value: '',
-- 
2.24.1

Re: [dpdk-dev] [RFC] DPDK Trace support

[Aaron Conole]

Jerin Jacob Kollanukkaran  writes:

> Hi All,
>
> I would like to add tracing support for DPDK.
> I am planning to add this support in v20.05 release.
>
> This RFC attempts to get feedback from the community on
>
> a) Tracing Use cases.
> b) Tracing Requirements.
> b) Implementation choices.
> c) Trace format.
>
> Use-cases
> -
> - Most of the cases, The DPDK provider will not have access to the DPDK 
> customer applications.
> To debug/analyze the slow path and fast path DPDK API usage from the field,
> we need to have integrated trace support in DPDK.
>
> - Need a low overhead Fast path multi-core PMD driver debugging/analysis
> infrastructure in DPDK to fix the functional and performance issue(s) of PMD.
>
> - Post trace analysis tools can provide various status across the system such
> as cpu_idle() using the timestamp added in the trace.
>
>
> Requirements:
> -
> - Support for Linux, FreeBSD and Windows OS
> - Open trace format
> - Multi-platform Open source trace viewer
> - Absolute low overhead trace API for DPDK fast path tracing/debugging.
> - Dynamic enable/disable of trace events
>
>
> To enable trace support in DPDK, following items need to work out: 
>
> a) Add the DPDK trace points in the DPDK source code.
>
> - This includes updating DPDK functions such as,
> rte_eth_dev_configure(), rte_eth_dev_start(), rte_eth_dev_rx_burst() to emit 
> the trace.

I wonder for these if it makes sense to use librte_bpf and a helper
function to actually emit events.  That way rather than static trace
point data, a user can implement some C-code and pull the exact data
that they want.

There could be some downside with the approach (because we might lose
some inlining or variable eliding), but I think it makes the trace point
concept quite a bit more powerful.  Have you given it any thought?

> b) Choosing suitable serialization-format
>
> - Common Trace Format, CTF, is an open format and language to describe trace 
> formats.
> This enables tool reuse, of which line-textual (babeltrace) and 
> graphical (TraceCompass) variants already exist.
>
> CTF should look familiar to C programmers but adds stronger typing. 
> See CTF - A Flexible, High-performance Binary Trace Format.
>
> https://diamon.org/ctf/
>
> c) Writing the on-target serialization code,
>
> See the section below.(Lttng CTF trace emitter vs DPDK specific CTF trace 
> emitter)
>  
> d) Deciding on and writing the I/O transport mechanics,
>
> For performance reasons, it should be backed by a huge-page and write to file 
> IO.
>
> e) Writing the PC-side deserializer/parser,
>
> Both the babletrace(CLI tool) and Trace Compass(GUI tool) support CTF.
> See: 
> https://lttng.org/viewers/
>
> f) Writing tools for filtering and presentation.
>
> See item (e)
>
>
> Lttng CTF trace emitter vs DPDK specific CTF trace emitter
> --
>
> I have written a performance evaluation application to measure the overhead
> of Lttng CTF emitter(The fastpath infrastructure used by
> https://lttng.org/ library to emit the trace)
>
> https://github.com/jerinjacobk/lttng-overhead
> https://github.com/jerinjacobk/lttng-overhead/blob/master/README
>
> I could improve the performance by 30% by adding the "DPDK"
> based plugin for get_clock() and get_cpu(),
> Here are the performance numbers after adding the plugin on 
> x86 and various arm64 board that I have access to,
>
> On high-end x86, it comes around 236 cycles/~100ns @ 2.4GHz (See the
> last line in the log(ZERO_ARG))
> On arm64, it varies from 312 cycles to 1100 cycles(based on the class of CPU).
> In short, Based on the "IPC capabilities", The cost would be around 100ns to 
> 400ns
> for single void trace(a trace without any argument)
>
>
> [lttng-overhead-x86] $ sudo ./calibrate/build/app/calibrate -c 0xc0
> make[1]: Entering directory '/export/lttng-overhead-x86/calibrate'
> make[1]: Leaving directory '/export/lttng-overhead-x86/calibrate'
> EAL: Detected 56 lcore(s)
> EAL: Detected 2 NUMA nodes
> EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
> EAL: Selected IOVA mode 'PA'
> EAL: Probing VFIO support...
> EAL: PCI device :01:00.0 on NUMA socket 0
> EAL:   probe driver: 8086:1521 net_e1000_igb
> EAL: PCI device :01:00.1 on NUMA socket 0
> EAL:   probe driver: 8086:1521 net_e1000_igb
> CPU Timer freq is 2600.00MHz
> NOP: cycles=0.194834 ns=0.074936
> GET_CLOCK: cyclesG.854658 ns.405638
> GET_CPU: cycles0.995892 ns.921497
> ZERO_ARG: cycles#6.945113 ns‘.132736
>
>
> We will have only 16.75ns to process 59.2 mpps(40Gbps), So IMO, Lttng CTF 
> emitter
> may not fit the DPDK fast path purpose due to the cost associated with 
> generic Lttng features.
>
> One option could be to have, native CTF emitter in EAL/DPDK to emit the
> trace in a hugepage. I think it would be a handful of cycles if we limit the 
> features
> to the requirements above:
>
> The upside of using Lttng CTF emitter:
> a) No need to write a new CTF trace emitter

Re: [dpdk-dev] [PATCH v5 00/11] example/l3fwd: introduce event device support

[Jerin Jacob]

On Fri, Jan 24, 2020 at 9:35 AM  wrote:
>
> From: Pavan Nikhilesh 
>
> This patchset adds support for event mode in l3fwd.
> The poll mode is still the default mode of operation.
>
> The following new command line parameters are added:
>  --mode: Dictates the mode of operation either poll or event. If unset then
>  by default poll mode is used.
>  --eventq_sched: Dictates event synchronization mode i.e. either ordered,
>  atomic or parallel.
>  --event-eth-rxqs: Number of ethernet RX queues per device.
>
> Based on event device capability the configuration is done as follows:
> - A single event device is enabled.
> - The number of event ports is equal to the number of worker
>   cores enabled in the core mask. Additional event ports might
>   be configured based on Rx/Tx adapter capability.
> - The number of event queues is equal to the number of ethernet
>   ports. If Tx adapter doesn't have internal port capability then
>   an additional single link event queue is used to enqueue events
>   to Tx adapter.
> - Each event port is linked to all existing event queues.
> - Dedicated Rx/Tx adapters for each Ethernet port.

This series looks good to me.
I have verified that there is no poll mode per core performance impact
with this change.

I will merge the next version on Tuesday for rc2 if there are no more
comments from the mailing list.

With below changes, Series-Acked-by: Jerin Jacob 

# Please rebase the release note to next-eventdev
# Fix the following typos

### examples/l3fwd: add event device configuration

WARNING:TYPO_SPELLING: 'configurtion' may be misspelled - perhaps
'configuration'?
#119: FILE: examples/l3fwd/l3fwd_event_generic.c:34:
+   /* Event device configurtion */

WARNING:TYPO_SPELLING: 'configurtion' may be misspelled - perhaps
'configuration'?
#207: FILE: examples/l3fwd/l3fwd_event_internal_port.c:34:
+   /* Event device configurtion */

total: 0 errors, 2 warnings, 230 lines checked

### examples/l3fwd: add event eth Rx/Tx adapter setup

WARNING:TYPO_SPELLING: 'memery' may be misspelled - perhaps 'memory'?
#67: FILE: examples/l3fwd/l3fwd_event_generic.c:208:
+   rte_panic("Failed to allocate memery for Rx adapter\n");

WARNING:TYPO_SPELLING: 'memery' may be misspelled - perhaps 'memory'?
#111: FILE: examples/l3fwd/l3fwd_event_generic.c:252:
+   rte_panic("Failed to allocate memery for Rx adapter\n");

WARNING:TYPO_SPELLING: 'memery' may be misspelled - perhaps 'memory'?
#198: FILE: examples/l3fwd/l3fwd_event_internal_port.c:207:
+   rte_panic("Failed to allocate memery for Rx adapter\n");

WARNING:TYPO_SPELLING: 'memery' may be misspelled - perhaps 'memory'?
#236: FILE: examples/l3fwd/l3fwd_event_internal_port.c:245:
+   rte_panic("Failed to allocate memery for Rx adapter\n");

total: 0 errors, 4 warnings, 239 lines checked

>
> v5 Changes:
> --
> - Update release notes and MAINTAINERS file.
> - Fix typo in doc.
>
> v4 Changes:
> --
> - Update documentation about --event-eth-rxqs. (Jerin)
> - Add validation for command line options that are specific to event/poll 
> mode.
> - Fix event device service initialization.
>
> v3 Changes:
> --
> - Unify argument parsing.
> - Segregate poll mode and event mode initialization.
> - Simplify event resource creation and accesses.
> - Integrate http://patches.dpdk.org/project/dpdk/list/?series=8002 series.
> - Reduce code duplication in lpm.
>
> v2 Changes:
> --
> - Shorten the structure name `s/event_rsrc/evt_rsrc` `s/l2fwd_rsrc/rsrc`.
> - Use rte_panic instead of rte_exit.
> - Rebase on top of Tx adapter change http://patches.dpdk.org/patch/60971.
> - Add parallel mode support.
> - Fix missing eventdev args parsing.
>
> Pavan Nikhilesh (7):
>   examples/l3fwd: add event device configuration
>   examples/l3fwd: add event eth Rx/Tx adapter setup
>   examples/l3fwd: add service core setup based on caps
>   examples/l3fwd: add event lpm main loop
>   examples/l3fwd: add event em main loop
>   examples/l3fwd: add graceful teardown for eventdevice
>   doc: update l3fwd user guide to support eventdev
>
> Sunil Kumar Kori (4):
>   examples/l3fwd: add framework for event device
>   examples/l3fwd: split pipelines based on capability
>   examples/l3fwd: add ethdev setup based on eventdev
>   examples/l3fwd: add event port and queue setup
>
>  MAINTAINERS|   1 +
>  doc/guides/rel_notes/release_20_02.rst |   5 +
>  doc/guides/sample_app_ug/l3_forward.rst|  79 -
>  examples/l3fwd/Makefile|   3 +-
>  examples/l3fwd/l3fwd.h |  30 ++
>  examples/l3fwd/l3fwd_em.c  | 177 +++
>  examples/l3fwd/l3fwd_em.h  | 159 +++---
>  examples/l3fwd/l3fwd_em_hlm.h  | 131 
>  examples/l3fwd/l3fwd_em_sequential.h   |  26 ++
>  examples/l3fwd/l3fwd_event.c   | 263 +

Re: [dpdk-dev] [dpdk-stable] [PATCH] app/testpmd: use correct RFC addresses for tx-only

[Ferruh Yigit]

On 1/27/2020 3:34 AM, Stephen Hemminger wrote:
> The previous patch to change default IP addresses for tx only
> mode got the wrong values (typo).
> 
> Fixes: bf5b2126bf44 ("app/testpmd: add ability to set Tx IP and UDP 
> parameters")
> Cc: sta...@dpdk.org
> Signed-off-by: Stephen Hemminger 
> ---
>  app/test-pmd/txonly.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/app/test-pmd/txonly.c b/app/test-pmd/txonly.c
> index 871cf6c1547f..4b5bec443b44 100644
> --- a/app/test-pmd/txonly.c
> +++ b/app/test-pmd/txonly.c
> @@ -45,8 +45,8 @@ uint16_t tx_udp_src_port = 9;
>  uint16_t tx_udp_dst_port = 9;
>  
>  /* use RFC5735 / RFC2544 reserved network test addresses */
> -uint32_t tx_ip_src_addr = (192U << 24) | (18 << 16) | (0 << 8) | 1;
> -uint32_t tx_ip_dst_addr = (192U << 24) | (18 << 16) | (0 << 8) | 2;
> +uint32_t tx_ip_src_addr = (198U << 24) | (18 << 16) | (0 << 8) | 1;
> +uint32_t tx_ip_dst_addr = (198U << 24) | (18 << 16) | (0 << 8) | 2;
>  
>  #define IP_DEFTTL  64   /* from RFC 1340. */
>  


I confirm from rfc5735:

198.18.0.0/15 - This block has been allocated for use in benchmark
   tests of network interconnect devices.  [RFC2544] explains that this
   range was assigned to minimize the chance of conflict in case a
   testing device were to be accidentally connected to part of the
   Internet.  Packets with source addresses from this range are not
   meant to be forwarded across the Internet.

Reviewed-by: Ferruh Yigit 

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

Re: [dpdk-dev] [RFC] DPDK Trace support

[Jerin Jacob]

On Mon, Jan 27, 2020 at 9:43 PM Aaron Conole  wrote:
>
> Jerin Jacob Kollanukkaran  writes:
>
> > Hi All,
> >
> > I would like to add tracing support for DPDK.
> > I am planning to add this support in v20.05 release.
> >
> > This RFC attempts to get feedback from the community on
> >
> > a) Tracing Use cases.
> > b) Tracing Requirements.
> > b) Implementation choices.
> > c) Trace format.
> >
> > Use-cases
> > -
> > - Most of the cases, The DPDK provider will not have access to the DPDK 
> > customer applications.
> > To debug/analyze the slow path and fast path DPDK API usage from the field,
> > we need to have integrated trace support in DPDK.
> >
> > - Need a low overhead Fast path multi-core PMD driver debugging/analysis
> > infrastructure in DPDK to fix the functional and performance issue(s) of 
> > PMD.
> >
> > - Post trace analysis tools can provide various status across the system 
> > such
> > as cpu_idle() using the timestamp added in the trace.
> >
> >
> > Requirements:
> > -
> > - Support for Linux, FreeBSD and Windows OS
> > - Open trace format
> > - Multi-platform Open source trace viewer
> > - Absolute low overhead trace API for DPDK fast path tracing/debugging.
> > - Dynamic enable/disable of trace events
> >
> >
> > To enable trace support in DPDK, following items need to work out:
> >
> > a) Add the DPDK trace points in the DPDK source code.
> >
> > - This includes updating DPDK functions such as,
> > rte_eth_dev_configure(), rte_eth_dev_start(), rte_eth_dev_rx_burst() to 
> > emit the trace.
>
> I wonder for these if it makes sense to use librte_bpf and a helper
> function to actually emit events.  That way rather than static trace
> point data, a user can implement some C-code and pull the exact data
> that they want.
>
> There could be some downside with the approach (because we might lose
> some inlining or variable eliding), but I think it makes the trace point
> concept quite a bit more powerful.  Have you given it any thought?

I think the reasoning for the same to have control over whether to
emit the trace or not. Right?
i.e only when specific conditions are met with runtime data then only
emit the trace to buffer.

I think, a couple of challenges would be
1) Performance in fast-path tracing
2) Need to write eBPF class for all the events that we need to trace to have
control the arguments for tracing

I think, once we have the base framework in C which support
enable/disable the event at runtime and
then we can give provision to hook eBPF program to control more runtime nature.
example emit rte_eth_dev_configure() trace only when port_id == 2 and
nb_rx_q == 4.

Re: [dpdk-dev] [PATCH v2 1/5] mk/icc: disable treatment of warnings as errors

[Thomas Monjalon]

27/01/2020 16:37, Ferruh Yigit:
> On 1/24/2020 7:37 PM, Thomas Monjalon wrote:
> > 24/01/2020 17:36, Ferruh Yigit:
> >> On 1/23/2020 6:20 PM, Alexander Kozyrev wrote:
> >>> Remove -Werror-all flag in ICC configuration file to stop treating ICC
> >>> warnings as errors in DPDK due to many false positives. We are using
> >>> GCC and Clang as a benchmark for warnings anyway for simplification.
> >>>
> >>> Suggested-by: Thomas Monjalon 
> >>> Signed-off-by: Alexander Kozyrev 
> >>> Acked-by: Viacheslav Ovsiienko 
> >>> Acked-by: Thomas Monjalon 
> >>> ---
> >>>  mk/toolchain/icc/rte.vars.mk | 4 
> >>>  1 file changed, 4 deletions(-)
> >>>
> >>> diff --git a/mk/toolchain/icc/rte.vars.mk b/mk/toolchain/icc/rte.vars.mk
> >>> index 8aa87aa..1729f3d 100644
> >>> --- a/mk/toolchain/icc/rte.vars.mk
> >>> +++ b/mk/toolchain/icc/rte.vars.mk
> >>> @@ -47,10 +47,6 @@ WERROR_FLAGS += -diag-disable 13368 -diag-disable 15527
> >>>  WERROR_FLAGS += -diag-disable 188
> >>>  WERROR_FLAGS += -diag-disable 11074 -diag-disable 11076 -Wdeprecated
> >>>  
> >>> -ifeq ($(RTE_DEVEL_BUILD),y)
> >>> -WERROR_FLAGS += -Werror-all
> >>> -endif
> >>
> >> Not sure about removing this globally, as of now the ICC builds fine. If 
> >> this is
> >> for the coming changes in mlx, why not disable warnings in mlx driver only?
> > 
> > Adding special handling for ICC in the driver means it is kind of supported.
> > ICC support is on best-effort basis as a favor to Intel and its CI.
> > 
> > I don't see any good reason to spend time disabling ICC warnings each time
> > we hit a false positive.
> > And I would like we stop doing strange things in the code to make ICC happy.
> > We do not need to support ICC, or at least we do not need to make ICC build
> > warning-free.
> > 
> > This patch will solve all future annoying issues with ICC in the future.
> > 
> > Now let me ask the reverse question: why the flag -Werror-all is set for 
> > ICC?
> > We set this flag for gcc and clang because we use gcc and clang as static
> > code analyzers (like coverity) and we want to be sure to catch any issue.
> > ICC is not used as code analyzer, this is just an optimization for some
> > Intel projects. I won't elaborate on the packaging and licensing of ICC...
> > 
> > I hope you will be convinced to acknowledge this change.
> > 
> 
> To support the ICC or not is a valid question, but for me it is better to
> support more compilers in case different ones catch an additional issues that 
> is
> a benefit for us, although I agree false positives are annoying, which is not
> limited to icc.
> 
> "-Werror-all" is forced in DEVEL_BUILD, in this cause I was assuming to force
> developer to fix the warning to increase the code quality, independent from
> compiler.

I doubt that ICC can help to increase code quality.
And it is really annoying to fix any ICC warning,
because it is not a truly open compiler.

> As of now, all DPDK compiles with icc without warning.

There is no warning currently because we avoid some known patterns
that ICC does not like. I don't see this fact as an advantage.

> But we are allowing the
> icc warnings just because the warnings with the change in the mlx driver. And
> when we let is once, it is for sure warnings will increase by time.

Yes, this is the intent of this patch: let the ICC warnings grow.

> If we support ICC, I am for keeping this flag and support it properly.

We do not really support ICC.
See the first item of the requirements:

http://doc.dpdk.org/guides/linux_gsg/sys_reqs.html#compilation-of-the-dpdk
"supported C compiler such as gcc (version 4.9+) or clang (version 
3.4+)"
and the consensus in the techboard:
http://mails.dpdk.org/archives/dev/2019-June/135847.html
"Agreement that all DPDK PMDs themselves must be possible to
 compile with a DPDK supported compiler - GCC and/or clang"

I add the Technical Board as Cc, in case a confirmation is required.

[dpdk-dev] [PATCH] drivers/crypto: fix set_sym_session_private_data error in sessionless mode

[Junxiao Shi]

When OpenSSL or AESNI-MB cryptodev is being used in sessionless mode
for symmetric crypto operation (e.g. SHA256 hash), the driver prints
error message:

CRYPTODEV: set_sym_session_private_data() line 489:
   Set private data for driver 0 not allowed

Then, AESNI-MB driver segfaults in post_process_mb_job().

Bugzilla ID: 377
Fixes: b3bbd9e ("cryptodev: support device independent sessions")
Fixes: c68d7aa ("crypto/aesni_mb: use architecture independent macros")

Signed-off-by: Junxiao Shi 
---
 drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c | 4 ++--
 drivers/crypto/openssl/rte_openssl_pmd.c   | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c 
b/drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c
index 97d9f81..33f4167 100644
--- a/drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c
+++ b/drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c
@@ -729,10 +729,10 @@ get_session(struct aesni_mb_qp *qp, struct rte_crypto_op 
*op)
op->sym->session,
cryptodev_driver_id);
} else {
-   void *_sess = NULL;
+   void *_sess = rte_cryptodev_sym_session_create(qp->sess_mp);
void *_sess_private_data = NULL;

-   if (rte_mempool_get(qp->sess_mp, (void **)&_sess))
+   if (_sess == NULL)
return NULL;

if (rte_mempool_get(qp->sess_mp_priv,
diff --git a/drivers/crypto/openssl/rte_openssl_pmd.c 
b/drivers/crypto/openssl/rte_openssl_pmd.c
index 91f0283..199097b 100644
--- a/drivers/crypto/openssl/rte_openssl_pmd.c
+++ b/drivers/crypto/openssl/rte_openssl_pmd.c
@@ -762,10 +762,10 @@ get_session(struct openssl_qp *qp, struct rte_crypto_op 
*op)
return NULL;

/* provide internal session */
-   void *_sess = NULL;
+   void *_sess = rte_cryptodev_sym_session_create(qp->sess_mp);
void *_sess_private_data = NULL;

-   if (rte_mempool_get(qp->sess_mp, (void **)&_sess))
+   if (_sess == NULL)
return NULL;

if (rte_mempool_get(qp->sess_mp_priv,
--
2.7.4

[dpdk-dev] [PATCH v2] net/regexdev: introduce regexdev subsystem

[Ori Kam]

Even though there are some vendors which offer Regex HW offload, due to
lack of standard API, It is diffcult for DPDK consumer to use them
in a portable way.

This _RFC_ attempts to standardize the RegEx/DPI offload APIs for DPDK.

This RFC crafted based on SW Regex API frameworks such as libpcre and
hyperscan and a few of the RegEx HW IPs which I am aware of.

RegEx pattern matching applications:
* Next Generation Firewalls (NGFW)
* Deep Packet and Flow Inspection (DPI)
* Intrusion Prevention Systems (IPS)
* DDoS Mitigation
* Network Monitoring
* Data Loss Prevention (DLP)
* Smart NICs
* Grammar based content processing
* URL, spam and adware filtering
* Advanced auditing and policing of user/application security policies
* Financial data mining - parsing of streamed financial feeds
* Application recognition.
* Dmemory introspection.
* Natural Language Processing (NLP)
* Sentiment Analysis.
* Big data databse acceleration.
* Computational storage.

Request to review from HW and SW RegEx vendors and RegEx application
users to have portable DPDK API for RegEx.

The API schematics are based cryptodev, eventdev and ethdev existing
device API.

Signed-off-by: Jerin Jacob 
Signed-off-by: Pavan Nikhilesh 
Signed-off-by: Ori Kam 
---
V2:
 - Update RFC based on ML comments.

RTE RegEx Device API


Defines RTE RegEx Device APIs for RegEx operations and its provisioning.

The RegEx Device API is composed of two parts:

- The application-oriented RegEx API that includes functions to setup
  a RegEx device (configure it, setup its queue pairs and start it),
  update the rule database and so on.

- The driver-oriented RegEx API that exports a function allowing
  a RegEx poll Mode Driver (PMD) to simultaneously register itself as
  a RegEx device driver.

RegEx device components and definitions:

+-+
| |
| o-+rte_regex_[en|de]queue_burst()
|   PCRE basedo--+  |   |
|  RegEx pattern  |  |  |  ++   |
| matching engine o--+--+--o|   |+--+
| |  |  |  | queue  |<==o===>|Core 0|
| o+ |  |  | pair 0 ||  |
| || |  |  +++--+
+-+| |  |
   ^   | |  |  ++
   |   | |  |  ||+--+
   |   | +--+--o queue  |<==>|Core 1|
   Rule|Database   ||  | pair 1 ||  |
+--+--+||  +++--+
| Group 0 |||
| +-+ |||  +++--+
| | Rules 0..n  | |||  |||Core 2|
| +-+ ||+--o queue  |<==>|  |
| Group 1 ||   | pair 2 |+--+
| +-+ ||   ++
| | Rules 0..n  | ||
| +-+ ||   ++
| Group 2 ||   ||+--+
| +-+ ||   | queue  |<==>|Core n|
| | Rules 0..n  | |+---o pair n ||  |
| +-+ |+++--+
| Group n |
| +-+ |<---rte_regex_rule_db_update()
| | Rules 0..n  | |<---rte_regex_rule_db_import()
| +-+ |--->rte_regex_rule_db_export()
+-+

RegEx: A regular expression is a concise and flexible means for matching
strings of text, such as particular characters, words, or patterns of
characters. A common abbreviation for this is ?RegEx?.

RegEx device: A hardware or software-based implementation of RegEx
device API for PCRE based pattern matching syntax and semantics.

PCRE RegEx syntax and semantics specification:
http://regexkit.sourceforge.net/Documentation/pcre/pcrepattern.html

RegEx queue pair: Each RegEx device should have one or more queue pair
to
transmit a burst of pattern matching request and receive a burst of
receive the pattern matching response. The pattern matching
request/response
embedded in *rte_regex_ops* structure.

Rule: A pattern matching rule expressed in PCRE RegEx syntax along with
Match ID and Group ID to identify the rule upon the match.

Rule database: The RegEx device accepts regular expressions and converts
them
into a compiled rule database that can then be used to scan data.
Compilation allows the device to analyze the given pattern(s) and
pre-determine how to scan for these patterns in an optimized fashion
that
would be far too expensive to compute at run-time. A rule database
contains
a set of rules that compiled in device specific binary form.

Match ID or Rule ID: A unique identifier provided at the time of rule
creation for the application to identify the rule upon match.

Group ID: Group of rules can be grouped under one group ID to enable
rule isolation and effective pattern

[dpdk-dev] [PATCH] bpf: properly install headers in meson build

[Junxiao Shi]

Previously, when librte_bpf is built with meson+ninja, its
headers such as bpf_def is not installed to the system.
This commit fixes this problem.

Fixes: 94972f3 ("bpf: add BPF loading and execution framework")

Signed-off-by: Junxiao Shi 
---
 lib/librte_bpf/meson.build | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/librte_bpf/meson.build b/lib/librte_bpf/meson.build
index 13fc02d..52cfaf9 100644
--- a/lib/librte_bpf/meson.build
+++ b/lib/librte_bpf/meson.build
@@ -14,7 +14,7 @@ elif dpdk_conf.has('RTE_ARCH_ARM64')
sources += files('bpf_jit_arm64.c')
 endif
 
-install_headers = files('bpf_def.h',
+install_headers('bpf_def.h',
'rte_bpf.h',
'rte_bpf_ethdev.h')
 
-- 
2.7.4

[dpdk-dev] ACL priority field

[Bly, Mike]

Hello,

Can someone clarify what I am interpreting  as a documentation conflict 
regarding the "priority" field for rte_table_acl_rule_add_params? Below 
documentation says "highest priority wins", but the header file comment says 0 
is highest priority. Based on my testing with conflicting entries, I would like 
ask if we can/should update the documentation/descriptions to state "the lowest 
non-negative integer priority value will be selected". Highest priority implies 
select X, when X > Y >= 0. However, based on my testing, that is not the case. 
Instead, Y is selected.

From: https://doc.dpdk.org/guides/prog_guide/packet_classif_access_ctrl.html

When creating a set of rules, for each rule, additional information must be 
supplied also:
* priority: A weight to measure the priority of the rules (higher is 
better)... If the input tuple matches more than one rule, then the rule with 
the higher priority is returned. Note that if the input tuple matches more than 
one rule and these rules have equal priority, it is undefined which rule is 
returned as a match. It is recommended to assign a unique priority for each 
rule.
From: http://doc.dpdk.org/api/structrte__table__acl__rule__add__params.html

int32_t priority
ACL rule priority, with 0 as the highest priority
Regards,
Mike

Re: [dpdk-dev] [PATCH v5 3/9] eal: add windows compatible header files

[Thomas Monjalon]

13/01/2020 22:55, Pallavi Kadam:
> Modified \common\include\arch\x86\rte_vect.h
> to include SSE4 header for Windows.
> 
> Adding dlfcn.h on Windows to support common code.
> 
> Adding eal_filesystem.h to support functions and
> path defines for files and directories on Windows.

I don't see any relationship between these 3 items,
so I think they should be 3 separate patches.

> --- a/lib/librte_eal/common/include/arch/x86/rte_vect.h
> +++ b/lib/librte_eal/common/include/arch/x86/rte_vect.h
> @@ -15,7 +15,9 @@
>  #include 
>  #include "generic/rte_vect.h"
>  
> -#if (defined(__ICC) || (__GNUC__ == 4 &&  __GNUC_MINOR__ < 4))
> +#if (defined(__ICC) || \
> + (defined(_WIN64)) || \
> + (__GNUC__ == 4 &&  __GNUC_MINOR__ < 4))
>  
>  #include  /* SSE4 */

I trust you on this change :)

> --- /dev/null
> +++ b/lib/librte_eal/windows/eal/include/dlfcn.h
> @@ -0,0 +1,21 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + * Copyright(c) 2019 Intel Corporation
> + */
> +
> +#ifndef _DLFCN_H_
> +#define _DLFCN_H_
> +
> +/**
> + * This file is added to support common code in eal_common_options.c
> + * as Microsoft libc does not contain dlfcn.h. This may be removed
> + * in future releases.
> + */
> +
> +/* The windows port does not currently support dynamic loading of libraries,
> + * so fail these calls
> + */
> +#define dlopen(lib, flag)   (0)
> +#define RTLD_NOW 0
> +#define dlerror()   ("Not supported!")

This is only for the function eal_plugins_init().
The plugin logic and directory search might be different on Windows.
I believe it would be better handled directly in eal_plugins_init().
For now, my advice is to not compile this function (and not call it
in Windows init of course):

int
eal_plugins_init(void)
{
#ifndef RTE_EXEC_ENV_WINDOWS

The right fix would be to move this "common" function in an UNIX-only file:
What about creating eal_unix_options.c file?
Could be in lib/librte_eal/unix/ directory?


> --- /dev/null
> +++ b/lib/librte_eal/windows/eal/include/eal_filesystem.h
> @@ -0,0 +1,99 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + * Copyright(c) 2019 Intel Corporation
> + */
> +
> +/**
> + * @file
> + * Stores functions and path defines for files and directories
> + * on the filesystem for Windows, that are used by the Windows EAL.
> + */
[...]
> +/** Function to read a single numeric value from a file on the filesystem.
> + * Used to read information from files on /sys
> + */
> +int eal_parse_sysfs_value(const char *filename, unsigned long *val);

Given that sysfs is a Linux system, I guess you don't need this function at all.

Re: [dpdk-dev] [PATCH v5 7/9] eal: remove syslog support for windows

[Thomas Monjalon]

13/01/2020 22:55, Pallavi Kadam:
> Added #ifndef WIN64 to exclude syslog definitions and parameters
> from Windows builds.
> 
> Signed-off-by: Pallavi Kadam 
> Reviewed-by: Ranjit Menon 
> ---
> --- a/lib/librte_eal/common/eal_common_options.c
> +++ b/lib/librte_eal/common/eal_common_options.c
> @@ -204,9 +206,9 @@ eal_reset_internal_config(struct internal_config 
> *internal_cfg)
>   internal_cfg->hugepage_info[i].lock_descriptor = -1;
>   }
>   internal_cfg->base_virtaddr = 0;
> -

Please keep blank lines

> +#ifndef _WIN64

Could it be #ifdef LOG_DAEMON?

>   internal_cfg->syslog_facility = LOG_DAEMON;
> -
> +#endif
[..]
> @@ -1391,7 +1395,7 @@ eal_parse_common_option(int opt, const char *optarg,
> +#ifndef _WIN64
>   case OPT_SYSLOG_NUM:
>   if (eal_parse_syslog(optarg, conf) < 0) {

Instead of adding #ifdef, I think we could introduce eal_parse_unix_option()
in a separate file, and it would call eal_parse_common_option().
So in Windows, you just skip it by calling directly eal_parse_common_option()
which would be truly common.

Re: [dpdk-dev] [PATCH v5 8/9] build: add additional common files support

[Thomas Monjalon]

13/01/2020 22:55, Pallavi Kadam:
> Added support for additional common files in meson build
> to expand Windows EAL and to support the lcore parsing
> feature on Windows.
> 
> Signed-off-by: Pallavi Kadam 
> Signed-off-by: Antara Ganesh Kolar 
> Reviewed-by: Ranjit Menon 
> Reviewed-by: Keith Wiles 
> ---
> --- a/lib/librte_eal/windows/eal/meson.build
> +++ b/lib/librte_eal/windows/eal/meson.build
>  common_sources = files(
> + '../../common/eal_common_bus.c',
> + '../../common/eal_common_class.c',
> + '../../common/eal_common_devargs.c',
>   '../../common/eal_common_errno.c',
>   '../../common/eal_common_launch.c',
>   '../../common/eal_common_lcore.c',
> - '../../common/eal_common_log.c'
> + '../../common/eal_common_log.c',
> + '../../common/eal_common_options.c',
> + '../../common/eal_common_thread.c',
> + '../../common/rte_option.c'

Please add a comma after the last item of the list, so the next patch
won't need to change this line.

[dpdk-dev] [PATCH v4 0/8] Introduce CPU crypto mode

[Marcin Smoczynski]

Originally both SW and HW crypto PMDs use rte_crypot_op based API to
process the crypto workload asynchronously. This way provides uniformity
to both PMD types, but also introduce unnecessary performance penalty to
SW PMDs that have to "simulate" HW async behavior (crypto-ops
enqueue/dequeue, HW addresses computations, storing/dereferencing user
provided data (mbuf) for each crypto-op, etc).

The aim is to introduce a new optional API for SW crypto-devices
to perform crypto processing in a synchronous manner.

v3 to v4 changes:
 - add feature discovery in the ipsec example application when
   using cpu-crypto
 - add gmac in aesni-gcm
 - add tests for aesni-gcm/cpu crypto mode
 - add documentation: pg and rel notes
 - remove xform flags as no longer needed
 - add some extra API comments
 - remove compilation error from v3

Marcin Smoczynski (8):
  cryptodev: introduce cpu crypto support API
  crypto/aesni_gcm: cpu crypto support
  test/crypto: add CPU crypto tests
  security: add cpu crypto action type
  ipsec: introduce support for cpu crypto mode
  examples/ipsec-secgw: cpu crypto support
  examples/ipsec-secgw: cpu crypto testing
  doc: add cpu crypto related documentation

 app/test/Makefile |   1 +
 app/test/cpu_crypto_all_gcm_perf_test_cases.h |  11 +
 app/test/cpu_crypto_all_gcm_unit_test_cases.h |  49 +
 .../cpu_crypto_all_gmac_unit_test_cases.h |   7 +
 app/test/meson.build  |   1 +
 app/test/test_cryptodev_cpu_crypto.c  | 930 ++
 doc/guides/cryptodevs/aesni_gcm.rst   |   5 +
 doc/guides/prog_guide/cryptodev_lib.rst   |  31 +
 doc/guides/prog_guide/ipsec_lib.rst   |   8 +
 doc/guides/prog_guide/rte_security.rst|  15 +-
 doc/guides/rel_notes/release_20_02.rst|   8 +
 drivers/crypto/aesni_gcm/aesni_gcm_ops.h  |   9 +
 drivers/crypto/aesni_gcm/aesni_gcm_pmd.c  | 220 -
 drivers/crypto/aesni_gcm/aesni_gcm_pmd_ops.c  |   3 +
 .../crypto/aesni_gcm/aesni_gcm_pmd_private.h  |  11 +-
 examples/ipsec-secgw/ipsec.c  |  23 +-
 examples/ipsec-secgw/ipsec_process.c  | 134 ++-
 examples/ipsec-secgw/sa.c |  28 +-
 examples/ipsec-secgw/test/common_defs.sh  |  21 +
 examples/ipsec-secgw/test/linux_test4.sh  |  11 +-
 examples/ipsec-secgw/test/linux_test6.sh  |  11 +-
 .../test/trs_3descbc_sha1_common_defs.sh  |   8 +-
 .../test/trs_aescbc_sha1_common_defs.sh   |   8 +-
 .../test/trs_aesctr_sha1_common_defs.sh   |   8 +-
 .../test/tun_3descbc_sha1_common_defs.sh  |   8 +-
 .../test/tun_aescbc_sha1_common_defs.sh   |   8 +-
 .../test/tun_aesctr_sha1_common_defs.sh   |   8 +-
 lib/librte_cryptodev/rte_crypto_sym.h |  61 ++
 lib/librte_cryptodev/rte_cryptodev.c  |  33 +
 lib/librte_cryptodev/rte_cryptodev.h  |  20 +
 lib/librte_cryptodev/rte_cryptodev_pmd.h  |  19 +
 .../rte_cryptodev_version.map |   1 +
 lib/librte_ipsec/esp_inb.c| 154 ++-
 lib/librte_ipsec/esp_outb.c   | 134 ++-
 lib/librte_ipsec/misc.h   | 118 +++
 lib/librte_ipsec/rte_ipsec.h  |  18 +-
 lib/librte_ipsec/sa.c | 112 ++-
 lib/librte_ipsec/sa.h |  17 +
 lib/librte_ipsec/ses.c|   3 +-
 lib/librte_security/rte_security.h|   6 +-
 40 files changed, 2119 insertions(+), 162 deletions(-)
 create mode 100644 app/test/cpu_crypto_all_gcm_perf_test_cases.h
 create mode 100644 app/test/cpu_crypto_all_gcm_unit_test_cases.h
 create mode 100644 app/test/cpu_crypto_all_gmac_unit_test_cases.h
 create mode 100644 app/test/test_cryptodev_cpu_crypto.c

-- 
2.17.1

[dpdk-dev] [PATCH v4 1/8] cryptodev: introduce cpu crypto support API

[Marcin Smoczynski]

Add new API allowing to process crypto operations in a synchronous
manner. Operations are performed on a set of SG arrays.

Sync mode is selected by setting appropriate flag in an xform
type number. Cryptodevs which allows CPU crypto operation mode have to
use RTE_CRYPTODEV_FF_SYM_CPU_CRYPTO capability.

Signed-off-by: Konstantin Ananyev 
Signed-off-by: Marcin Smoczynski 
---
 lib/librte_cryptodev/rte_crypto_sym.h | 61 +++
 lib/librte_cryptodev/rte_cryptodev.c  | 33 ++
 lib/librte_cryptodev/rte_cryptodev.h  | 20 ++
 lib/librte_cryptodev/rte_cryptodev_pmd.h  | 19 ++
 .../rte_cryptodev_version.map |  1 +
 5 files changed, 134 insertions(+)

diff --git a/lib/librte_cryptodev/rte_crypto_sym.h 
b/lib/librte_cryptodev/rte_crypto_sym.h
index bc356f6ff..da1530093 100644
--- a/lib/librte_cryptodev/rte_crypto_sym.h
+++ b/lib/librte_cryptodev/rte_crypto_sym.h
@@ -25,6 +25,67 @@ extern "C" {
 #include 
 #include 
 
+/**
+ * Crypto IO Vector (in analogy with struct iovec)
+ * Supposed be used to pass input/output data buffers for crypto data-path
+ * functions.
+ */
+struct rte_crypto_vec {
+   /** virtual address of the data buffer */
+   void *base;
+   /** IOVA of the data buffer */
+   rte_iova_t *iova;
+   /** length of the data buffer */
+   uint32_t len;
+};
+
+/**
+ * Crypto scatter-gather list descriptor. Consists of a pointer to an array
+ * of Crypto IO vectors with its size.
+ */
+struct rte_crypto_sgl {
+   /** start of an array of vectors */
+   struct rte_crypto_vec *vec;
+   /** size of an array of vectors */
+   uint32_t num;
+};
+
+/**
+ * Synchronous operation descriptor.
+ * Supposed to be used with CPU crypto API call.
+ */
+struct rte_crypto_sym_vec {
+   /** array of SGL vectors */
+   struct rte_crypto_sgl *sgl;
+   /** array of pointers to IV */
+   void **iv;
+   /** array of pointers to AAD */
+   void **aad;
+   /** array of pointers to digest */
+   void **digest;
+   /**
+* array of statuses for each operation:
+*  - 0 on success
+*  - errno on error
+*/
+   int32_t *status;
+   /** number of operations to perform */
+   uint32_t num;
+};
+
+/**
+ * used for cpu_crypto_process_bulk() to specify head/tail offsets
+ * for auth/cipher processing.
+ */
+union rte_crypto_sym_ofs {
+   uint64_t raw;
+   struct {
+   struct {
+   uint16_t head;
+   uint16_t tail;
+   } auth, cipher;
+   } ofs;
+};
 
 /** Symmetric Cipher Algorithms */
 enum rte_crypto_cipher_algorithm {
diff --git a/lib/librte_cryptodev/rte_cryptodev.c 
b/lib/librte_cryptodev/rte_cryptodev.c
index 5c6359b5c..410b22867 100644
--- a/lib/librte_cryptodev/rte_cryptodev.c
+++ b/lib/librte_cryptodev/rte_cryptodev.c
@@ -494,6 +494,8 @@ rte_cryptodev_get_feature_name(uint64_t flag)
return "RSA_PRIV_OP_KEY_QT";
case RTE_CRYPTODEV_FF_DIGEST_ENCRYPTED:
return "DIGEST_ENCRYPTED";
+   case RTE_CRYPTODEV_FF_SYM_CPU_CRYPTO:
+   return "SYM_CPU_CRYPTO";
default:
return NULL;
}
@@ -1619,6 +1621,37 @@ rte_cryptodev_sym_session_get_user_data(
return (void *)(sess->sess_data + sess->nb_drivers);
 }
 
+static inline void
+sym_crypto_fill_status(struct rte_crypto_sym_vec *vec, int32_t errnum)
+{
+   uint32_t i;
+   for (i = 0; i < vec->num; i++)
+   vec->status[i] = errnum;
+}
+
+uint32_t
+rte_cryptodev_sym_cpu_crypto_process(uint8_t dev_id,
+   struct rte_cryptodev_sym_session *sess, union rte_crypto_sym_ofs ofs,
+   struct rte_crypto_sym_vec *vec)
+{
+   struct rte_cryptodev *dev;
+
+   if (!rte_cryptodev_pmd_is_valid_dev(dev_id)) {
+   sym_crypto_fill_status(vec, EINVAL);
+   return 0;
+   }
+
+   dev = rte_cryptodev_pmd_get_dev(dev_id);
+
+   if (*dev->dev_ops->sym_cpu_process == NULL ||
+   !(dev->feature_flags & RTE_CRYPTODEV_FF_SYM_CPU_CRYPTO)) {
+   sym_crypto_fill_status(vec, ENOTSUP);
+   return 0;
+   }
+
+   return dev->dev_ops->sym_cpu_process(dev, sess, ofs, vec);
+}
+
 /** Initialise rte_crypto_op mempool element */
 static void
 rte_crypto_op_init(struct rte_mempool *mempool,
diff --git a/lib/librte_cryptodev/rte_cryptodev.h 
b/lib/librte_cryptodev/rte_cryptodev.h
index c6ffa3b35..8786dfb90 100644
--- a/lib/librte_cryptodev/rte_cryptodev.h
+++ b/lib/librte_cryptodev/rte_cryptodev.h
@@ -450,6 +450,8 @@ rte_cryptodev_asym_get_xform_enum(enum 
rte_crypto_asym_xform_type *xform_enum,
 /**< Support encrypted-digest operations where digest is appended to data */
 #define RTE_CRYPTODEV_FF_ASYM_SESSIONLESS  (1ULL << 20)
 /**< Support asymmetric session-less operations */
+#defineRTE_CRYPTODEV_FF_SYM_CPU_CRYPTO (1ULL << 21)
+/**< Supp

[dpdk-dev] [PATCH v4 2/8] crypto/aesni_gcm: cpu crypto support

[Marcin Smoczynski]

Add support for CPU crypto mode by introducing required handler.
Crypto mode (sync/async) is chosen during sym session create if an
appropriate flag is set in an xform type number.

Authenticated encryption and decryption are supported with tag
generation/verification.

Signed-off-by: Marcin Smoczynski 
---
 drivers/crypto/aesni_gcm/aesni_gcm_ops.h  |   9 +
 drivers/crypto/aesni_gcm/aesni_gcm_pmd.c  | 220 +-
 drivers/crypto/aesni_gcm/aesni_gcm_pmd_ops.c  |   3 +
 .../crypto/aesni_gcm/aesni_gcm_pmd_private.h  |  11 +-
 4 files changed, 237 insertions(+), 6 deletions(-)

diff --git a/drivers/crypto/aesni_gcm/aesni_gcm_ops.h 
b/drivers/crypto/aesni_gcm/aesni_gcm_ops.h
index e272f1067..404c0adff 100644
--- a/drivers/crypto/aesni_gcm/aesni_gcm_ops.h
+++ b/drivers/crypto/aesni_gcm/aesni_gcm_ops.h
@@ -65,4 +65,13 @@ struct aesni_gcm_ops {
aesni_gcm_finalize_t finalize_dec;
 };
 
+/** GCM per-session operation handlers */
+struct aesni_gcm_session_ops {
+   aesni_gcm_t cipher;
+   aesni_gcm_pre_t pre;
+   aesni_gcm_init_t init;
+   aesni_gcm_update_t update;
+   aesni_gcm_finalize_t finalize;
+};
+
 #endif /* _AESNI_GCM_OPS_H_ */
diff --git a/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c 
b/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
index 1a03be31d..9901c811b 100644
--- a/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
+++ b/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
@@ -15,6 +15,31 @@
 
 static uint8_t cryptodev_driver_id;
 
+/* setup session handlers */
+static void
+set_func_ops(struct aesni_gcm_session *s, const struct aesni_gcm_ops *gcm_ops)
+{
+   s->ops.pre = gcm_ops->pre;
+   s->ops.init = gcm_ops->init;
+
+   switch (s->op) {
+   case AESNI_GCM_OP_AUTHENTICATED_ENCRYPTION:
+   s->ops.cipher = gcm_ops->enc;
+   s->ops.update = gcm_ops->update_enc;
+   s->ops.finalize = gcm_ops->finalize_enc;
+   break;
+   case AESNI_GCM_OP_AUTHENTICATED_DECRYPTION:
+   s->ops.cipher = gcm_ops->dec;
+   s->ops.update = gcm_ops->update_dec;
+   s->ops.finalize = gcm_ops->finalize_dec;
+   break;
+   case AESNI_GMAC_OP_GENERATE:
+   case AESNI_GMAC_OP_VERIFY:
+   s->ops.finalize = gcm_ops->finalize_enc;
+   break;
+   }
+}
+
 /** Parse crypto xform chain and set private session parameters */
 int
 aesni_gcm_set_session_parameters(const struct aesni_gcm_ops *gcm_ops,
@@ -65,6 +90,7 @@ aesni_gcm_set_session_parameters(const struct aesni_gcm_ops 
*gcm_ops,
/* Select Crypto operation */
if (aead_xform->aead.op == RTE_CRYPTO_AEAD_OP_ENCRYPT)
sess->op = AESNI_GCM_OP_AUTHENTICATED_ENCRYPTION;
+   /* op == RTE_CRYPTO_AEAD_OP_DECRYPT */
else
sess->op = AESNI_GCM_OP_AUTHENTICATED_DECRYPTION;
 
@@ -78,7 +104,6 @@ aesni_gcm_set_session_parameters(const struct aesni_gcm_ops 
*gcm_ops,
return -ENOTSUP;
}
 
-
/* IV check */
if (sess->iv.length != 16 && sess->iv.length != 12 &&
sess->iv.length != 0) {
@@ -102,6 +127,10 @@ aesni_gcm_set_session_parameters(const struct 
aesni_gcm_ops *gcm_ops,
return -EINVAL;
}
 
+   /* setup session handlers */
+   set_func_ops(sess, &gcm_ops[sess->key]);
+
+   /* pre-generate key */
gcm_ops[sess->key].pre(key, &sess->gdata_key);
 
/* Digest check */
@@ -356,6 +385,191 @@ process_gcm_crypto_op(struct aesni_gcm_qp *qp, struct 
rte_crypto_op *op,
return 0;
 }
 
+static inline void
+aesni_gcm_fill_error_code(struct rte_crypto_sym_vec *vec, int32_t errnum)
+{
+   uint32_t i;
+
+   for (i = 0; i < vec->num; i++)
+   vec->status[i] = errnum;
+}
+
+
+static inline int32_t
+aesni_gcm_sgl_op_finalize_encryption(const struct aesni_gcm_session *s,
+   struct gcm_context_data *gdata_ctx, uint8_t *digest)
+{
+   if (s->req_digest_length != s->gen_digest_length) {
+   uint8_t tmpdigest[s->gen_digest_length];
+
+   s->ops.finalize(&s->gdata_key, gdata_ctx, tmpdigest,
+   s->gen_digest_length);
+   memcpy(digest, tmpdigest, s->req_digest_length);
+   } else {
+   s->ops.finalize(&s->gdata_key, gdata_ctx, digest,
+   s->gen_digest_length);
+   }
+
+   return 0;
+}
+
+static inline int32_t
+aesni_gcm_sgl_op_finalize_decryption(const struct aesni_gcm_session *s,
+   struct gcm_context_data *gdata_ctx, uint8_t *digest)
+{
+   uint8_t tmpdigest[s->gen_digest_length];
+
+   s->ops.finalize(&s->gdata_key, gdata_ctx, tmpdigest,
+   s->gen_digest_length);
+
+   return memcmp(digest, tmpdigest, s->req_digest_length) == 0 ? 0 :
+   EBADMSG;
+}
+
+static inline void
+aesni_gcm_process_gcm_sgl_op(const struct aesni_gcm_session *s,
+   struct gcm_context_da

[dpdk-dev] [PATCH v4 4/8] security: add cpu crypto action type

[Marcin Smoczynski]

Introduce CPU crypto action type allowing to differentiate between
regular async 'none security' and synchronous, CPU crypto accelerated
sessions.

Signed-off-by: Marcin Smoczynski 
---
 lib/librte_security/rte_security.h | 6 +-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/lib/librte_security/rte_security.h 
b/lib/librte_security/rte_security.h
index 546779df2..309f7311c 100644
--- a/lib/librte_security/rte_security.h
+++ b/lib/librte_security/rte_security.h
@@ -307,10 +307,14 @@ enum rte_security_session_action_type {
/**< All security protocol processing is performed inline during
 * transmission
 */
-   RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL
+   RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL,
/**< All security protocol processing including crypto is performed
 * on a lookaside accelerator
 */
+   RTE_SECURITY_ACTION_TYPE_CPU_CRYPTO
+   /**< Crypto processing for security protocol is processed by CPU
+* synchronously
+*/
 };
 
 /** Security session protocol definition */
-- 
2.17.1

[dpdk-dev] [PATCH v4 3/8] test/crypto: add CPU crypto tests

[Marcin Smoczynski]

Add unit and performance tests for CPU crypto mode currently implemented
by AESNI-GCM cryptodev. Unit tests cover AES-GCM and GMAC test vectors.

Signed-off-by: Marcin Smoczynski 
---
 app/test/Makefile |   1 +
 app/test/cpu_crypto_all_gcm_perf_test_cases.h |  11 +
 app/test/cpu_crypto_all_gcm_unit_test_cases.h |  49 +
 .../cpu_crypto_all_gmac_unit_test_cases.h |   7 +
 app/test/meson.build  |   1 +
 app/test/test_cryptodev_cpu_crypto.c  | 930 ++
 6 files changed, 999 insertions(+)
 create mode 100644 app/test/cpu_crypto_all_gcm_perf_test_cases.h
 create mode 100644 app/test/cpu_crypto_all_gcm_unit_test_cases.h
 create mode 100644 app/test/cpu_crypto_all_gmac_unit_test_cases.h
 create mode 100644 app/test/test_cryptodev_cpu_crypto.c

diff --git a/app/test/Makefile b/app/test/Makefile
index 57930c00b..b8f0169ef 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -203,6 +203,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_PMD_RING) += test_pmd_ring_perf.c
 SRCS-$(CONFIG_RTE_LIBRTE_CRYPTODEV) += test_cryptodev_blockcipher.c
 SRCS-$(CONFIG_RTE_LIBRTE_CRYPTODEV) += test_cryptodev.c
 SRCS-$(CONFIG_RTE_LIBRTE_CRYPTODEV) += test_cryptodev_asym.c
+SRCS-$(CONFIG_RTE_LIBRTE_CRYPTODEV) += test_cryptodev_cpu_crypto.c
 SRCS-$(CONFIG_RTE_LIBRTE_SECURITY) += test_cryptodev_security_pdcp.c
 
 SRCS-$(CONFIG_RTE_LIBRTE_METRICS) += test_metrics.c
diff --git a/app/test/cpu_crypto_all_gcm_perf_test_cases.h 
b/app/test/cpu_crypto_all_gcm_perf_test_cases.h
new file mode 100644
index 0..ee9545abc
--- /dev/null
+++ b/app/test/cpu_crypto_all_gcm_perf_test_cases.h
@@ -0,0 +1,11 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2019 Intel Corporation
+ */
+
+TEST_EXPAND(_128, 16, SGL_ONE_SEG)
+TEST_EXPAND(_192, 24, SGL_ONE_SEG)
+TEST_EXPAND(_256, 32, SGL_ONE_SEG)
+
+TEST_EXPAND(_128, 16, SGL_MAX_SEG)
+TEST_EXPAND(_192, 24, SGL_MAX_SEG)
+TEST_EXPAND(_256, 32, SGL_MAX_SEG)
diff --git a/app/test/cpu_crypto_all_gcm_unit_test_cases.h 
b/app/test/cpu_crypto_all_gcm_unit_test_cases.h
new file mode 100644
index 0..ed40c1632
--- /dev/null
+++ b/app/test/cpu_crypto_all_gcm_unit_test_cases.h
@@ -0,0 +1,49 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2019 Intel Corporation
+ */
+
+TEST_EXPAND(gcm_test_case_1, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_2, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_3, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_4, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_5, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_6, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_7, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_8, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_192_1, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_192_2, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_192_3, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_192_4, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_192_5, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_192_6, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_192_7, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_256_1, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_256_2, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_256_3, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_256_4, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_256_5, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_256_6, SGL_ONE_SEG)
+TEST_EXPAND(gcm_test_case_256_7, SGL_ONE_SEG)
+
+TEST_EXPAND(gcm_test_case_1, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_2, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_3, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_4, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_5, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_6, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_7, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_8, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_192_1, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_192_2, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_192_3, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_192_4, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_192_5, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_192_6, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_192_7, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_256_1, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_256_2, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_256_3, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_256_4, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_256_5, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_256_6, SGL_MAX_SEG)
+TEST_EXPAND(gcm_test_case_256_7, SGL_MAX_SEG)
diff --git a/app/test/cpu_crypto_all_gmac_unit_test_cases.h 
b/app/test/cpu_crypto_all_gmac_unit_test_cases.h
new file mode 100644
index 0..b6ebce936
--- /dev/null
+++ b/app/test/cpu_crypto_all_gmac_unit_test_cases.h
@@ -0,0 +1,7 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2019 Intel Corporation
+ */
+
+TEST_EXPAND(gmac_test_case_1, SGL_ONE_SEG)
+TEST_EXPAND(gmac_test_case_2, SGL_ONE_SEG)
+TEST_EXPAND(gmac_test_case_3, SGL_ONE_SEG)
diff --git a/app/test/meson.build b/app/test/meson.build
index 22b0cefaa..5a218affe 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -30,6 +30,7 @@ test_sources = files('commands.c',
'test_cryptodev.c',
'test_

[dpdk-dev] [PATCH v4 5/8] ipsec: introduce support for cpu crypto mode

[Marcin Smoczynski]

Update library to handle CPU cypto security mode which utilizes
cryptodev's synchronous, CPU accelerated crypto operations.

Signed-off-by: Konstantin Ananyev 
Signed-off-by: Marcin Smoczynski 
---
 lib/librte_ipsec/esp_inb.c   | 154 ++-
 lib/librte_ipsec/esp_outb.c  | 134 +++---
 lib/librte_ipsec/misc.h  | 118 +++
 lib/librte_ipsec/rte_ipsec.h |  18 +++-
 lib/librte_ipsec/sa.c| 112 +
 lib/librte_ipsec/sa.h|  17 
 lib/librte_ipsec/ses.c   |   3 +-
 7 files changed, 506 insertions(+), 50 deletions(-)

diff --git a/lib/librte_ipsec/esp_inb.c b/lib/librte_ipsec/esp_inb.c
index 5c653dd39..58b3dec1b 100644
--- a/lib/librte_ipsec/esp_inb.c
+++ b/lib/librte_ipsec/esp_inb.c
@@ -105,6 +105,39 @@ inb_cop_prepare(struct rte_crypto_op *cop,
}
 }
 
+static inline uint32_t
+inb_cpu_crypto_prepare(const struct rte_ipsec_sa *sa, struct rte_mbuf *mb,
+   uint32_t *pofs, uint32_t plen, void *iv)
+{
+   struct aead_gcm_iv *gcm;
+   struct aesctr_cnt_blk *ctr;
+   uint64_t *ivp;
+   uint32_t clen;
+
+   ivp = rte_pktmbuf_mtod_offset(mb, uint64_t *,
+   *pofs + sizeof(struct rte_esp_hdr));
+   clen = 0;
+
+   switch (sa->algo_type) {
+   case ALGO_TYPE_AES_GCM:
+   gcm = (struct aead_gcm_iv *)iv;
+   aead_gcm_iv_fill(gcm, ivp[0], sa->salt);
+   break;
+   case ALGO_TYPE_AES_CBC:
+   case ALGO_TYPE_3DES_CBC:
+   copy_iv(iv, ivp, sa->iv_len);
+   break;
+   case ALGO_TYPE_AES_CTR:
+   ctr = (struct aesctr_cnt_blk *)iv;
+   aes_ctr_cnt_blk_fill(ctr, ivp[0], sa->salt);
+   break;
+   }
+
+   *pofs += sa->ctp.auth.offset;
+   clen = plen - sa->ctp.auth.length;
+   return clen;
+}
+
 /*
  * Helper function for prepare() to deal with situation when
  * ICV is spread by two segments. Tries to move ICV completely into the
@@ -157,17 +190,12 @@ inb_pkt_xprepare(const struct rte_ipsec_sa *sa, 
rte_be64_t sqc,
}
 }
 
-/*
- * setup/update packet data and metadata for ESP inbound tunnel case.
- */
-static inline int32_t
-inb_pkt_prepare(const struct rte_ipsec_sa *sa, const struct replay_sqn *rsn,
-   struct rte_mbuf *mb, uint32_t hlen, union sym_op_data *icv)
+static inline int
+inb_get_sqn(const struct rte_ipsec_sa *sa, const struct replay_sqn *rsn,
+   struct rte_mbuf *mb, uint32_t hlen, rte_be64_t *sqc)
 {
int32_t rc;
uint64_t sqn;
-   uint32_t clen, icv_len, icv_ofs, plen;
-   struct rte_mbuf *ml;
struct rte_esp_hdr *esph;
 
esph = rte_pktmbuf_mtod_offset(mb, struct rte_esp_hdr *, hlen);
@@ -179,12 +207,21 @@ inb_pkt_prepare(const struct rte_ipsec_sa *sa, const 
struct replay_sqn *rsn,
sqn = rte_be_to_cpu_32(esph->seq);
if (IS_ESN(sa))
sqn = reconstruct_esn(rsn->sqn, sqn, sa->replay.win_sz);
+   *sqc = rte_cpu_to_be_64(sqn);
 
+   /* check IPsec window */
rc = esn_inb_check_sqn(rsn, sa, sqn);
-   if (rc != 0)
-   return rc;
 
-   sqn = rte_cpu_to_be_64(sqn);
+   return rc;
+}
+
+/* prepare packet for upcoming processing */
+static inline int32_t
+inb_prepare(const struct rte_ipsec_sa *sa, struct rte_mbuf *mb,
+   uint32_t hlen, union sym_op_data *icv)
+{
+   uint32_t clen, icv_len, icv_ofs, plen;
+   struct rte_mbuf *ml;
 
/* start packet manipulation */
plen = mb->pkt_len;
@@ -217,7 +254,8 @@ inb_pkt_prepare(const struct rte_ipsec_sa *sa, const struct 
replay_sqn *rsn,
 
icv_ofs += sa->sqh_len;
 
-   /* we have to allocate space for AAD somewhere,
+   /*
+* we have to allocate space for AAD somewhere,
 * right now - just use free trailing space at the last segment.
 * Would probably be more convenient to reserve space for AAD
 * inside rte_crypto_op itself
@@ -238,10 +276,28 @@ inb_pkt_prepare(const struct rte_ipsec_sa *sa, const 
struct replay_sqn *rsn,
mb->pkt_len += sa->sqh_len;
ml->data_len += sa->sqh_len;
 
-   inb_pkt_xprepare(sa, sqn, icv);
return plen;
 }
 
+static inline int32_t
+inb_pkt_prepare(const struct rte_ipsec_sa *sa, const struct replay_sqn *rsn,
+   struct rte_mbuf *mb, uint32_t hlen, union sym_op_data *icv)
+{
+   int rc;
+   rte_be64_t sqn;
+
+   rc = inb_get_sqn(sa, rsn, mb, hlen, &sqn);
+   if (rc != 0)
+   return rc;
+
+   rc = inb_prepare(sa, mb, hlen, icv);
+   if (rc < 0)
+   return rc;
+
+   inb_pkt_xprepare(sa, sqn, icv);
+   return rc;
+}
+
 /*
  * setup/update packets and crypto ops for ESP inbound case.
  */
@@ -270,17 +326,17 @@ esp_inb_pkt_prepare(const struct rte_ipsec_session *ss, 
struct rte_mbuf *mb[],
lksd_none_cop_prepare(cop[k], cs, mb[i]);
inb_cop_prepare

[dpdk-dev] [PATCH v4 8/8] doc: add cpu crypto related documentation

[Marcin Smoczynski]

Update documentation with a description of cpu crypto in cryptodev,
ipsec and security libraries.

Add release notes for 20.02.

Signed-off-by: Marcin Smoczynski 
---
 doc/guides/cryptodevs/aesni_gcm.rst |  5 
 doc/guides/prog_guide/cryptodev_lib.rst | 31 +
 doc/guides/prog_guide/ipsec_lib.rst |  8 +++
 doc/guides/prog_guide/rte_security.rst  | 15 
 doc/guides/rel_notes/release_20_02.rst  |  8 +++
 5 files changed, 63 insertions(+), 4 deletions(-)

diff --git a/doc/guides/cryptodevs/aesni_gcm.rst 
b/doc/guides/cryptodevs/aesni_gcm.rst
index 151aa3060..6b1a3d2a0 100644
--- a/doc/guides/cryptodevs/aesni_gcm.rst
+++ b/doc/guides/cryptodevs/aesni_gcm.rst
@@ -9,6 +9,11 @@ The AES-NI GCM PMD (**librte_pmd_aesni_gcm**) provides poll 
mode crypto driver
 support for utilizing Intel multi buffer library (see AES-NI Multi-buffer PMD 
documentation
 to learn more about it, including installation).
 
+The AES-NI GCM PMD supports synchronous mode of operation with
+``rte_cryptodev_sym_cpu_crypto_process`` function call for both AES-GCM and
+GMAC, however GMAC support is limited to one segment per operation. Please
+refer to ``rte_crypto`` programmer's guide for more detail.
+
 Features
 
 
diff --git a/doc/guides/prog_guide/cryptodev_lib.rst 
b/doc/guides/prog_guide/cryptodev_lib.rst
index ac1643774..1a01e1bda 100644
--- a/doc/guides/prog_guide/cryptodev_lib.rst
+++ b/doc/guides/prog_guide/cryptodev_lib.rst
@@ -600,6 +600,37 @@ chain.
 };
 };
 
+Synchronous mode
+
+
+Some cryptodevs support synchronous mode alongside with a standard asynchronous
+mode. In that case operations are performed directly when calling
+``rte_cryptodev_sym_cpu_crypto_process`` method instead of enqueuing and
+dequeuing an operation before. This mode of operation allows cryptodevs which
+utilize CPU cryptographic acceleration to have significant performance boost
+comparing to standard asynchronous approach. Cryptodevs supporting synchronous
+mode have ``RTE_CRYPTODEV_FF_SYM_CPU_CRYPTO`` feature flag set.
+
+To perform a synchronous operation a call to
+``rte_cryptodev_sym_cpu_crypto_process`` has to be made with vectorized
+operation descriptor (``struct rte_crypto_sym_vec``) containing:
+
+- ``num`` - number of operations to perform,
+- pointer to an array of size ``num`` containing a scatter-gather list
+  descriptors of performed operations (``struct rte_crypto_sgl``). Each 
instance
+  of ``struct rte_crypto_sgl`` consists of a number of segments and a pointer 
to
+  an array of segment descriptors ``struct rte_crypto_vec``;
+- pointers to arrays of size ``num`` containing IV, AAD and digest information,
+- pointer to an array of size ``num`` where status information will be stored
+  for each operation.
+
+Function returns a number of successfully completed operations and sets
+appropriate status number for each operation in the status array provided as
+a call argument. Status different than zero must be treated as error.
+
+For more details, e.g. how to convert an mbuf to an SGL, please refer to an
+example usage in the IPsec library implementation.
+
 Sample code
 ---
 
diff --git a/doc/guides/prog_guide/ipsec_lib.rst 
b/doc/guides/prog_guide/ipsec_lib.rst
index 1ce0db453..e6a21fae6 100644
--- a/doc/guides/prog_guide/ipsec_lib.rst
+++ b/doc/guides/prog_guide/ipsec_lib.rst
@@ -81,6 +81,14 @@ In that mode the library functions perform
   - verify that crypto device operations (encryption, ICV generation)
 were completed successfully
 
+RTE_SECURITY_ACTION_TYPE_CPU_CRYPTO
+~~~
+
+In that mode the library functions perform same operations as in
+``RTE_SECURITY_ACTION_TYPE_NONE``. The only differnce is that crypto operations
+are performed with CPU crypto synchronous API.
+
+
 RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO
 ~~
 
diff --git a/doc/guides/prog_guide/rte_security.rst 
b/doc/guides/prog_guide/rte_security.rst
index f77fb89dc..a911c676b 100644
--- a/doc/guides/prog_guide/rte_security.rst
+++ b/doc/guides/prog_guide/rte_security.rst
@@ -511,13 +511,20 @@ Offload.
 /**< No security actions */
 RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO,
 /**< Crypto processing for security protocol is processed inline
- * during transmission */
+ * during transmission
+ */
 RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL,
 /**< All security protocol processing is performed inline during
- * transmission */
-RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL
+ * transmission
+ */
+RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL,
 /**< All security protocol processing including crypto is performed
- * on a lookaside accelerator */
+ * on a lookaside accelerator
+ */
+RTE_SECURITY_ACTION_TYPE_CPU_CRYPTO
+/**< Crypto processing for security protocol is processed b

[dpdk-dev] [PATCH v4 6/8] examples/ipsec-secgw: cpu crypto support

[Marcin Smoczynski]

Add support for CPU accelerated crypto. 'cpu-crypto' SA type has
been introduced in configuration allowing to use abovementioned
acceleration.

Legacy mode is not currently supported.

Signed-off-by: Konstantin Ananyev 
Signed-off-by: Marcin Smoczynski 
---
 examples/ipsec-secgw/ipsec.c |  23 -
 examples/ipsec-secgw/ipsec_process.c | 134 +--
 examples/ipsec-secgw/sa.c|  28 --
 3 files changed, 128 insertions(+), 57 deletions(-)

diff --git a/examples/ipsec-secgw/ipsec.c b/examples/ipsec-secgw/ipsec.c
index d4b57121a..49a947990 100644
--- a/examples/ipsec-secgw/ipsec.c
+++ b/examples/ipsec-secgw/ipsec.c
@@ -10,6 +10,7 @@
 #include 
 #include 
 #include 
+#include 
 #include 
 #include 
 #include 
@@ -86,7 +87,8 @@ create_lookaside_session(struct ipsec_ctx *ipsec_ctx, struct 
ipsec_sa *sa,
ipsec_ctx->tbl[cdev_id_qp].id,
ipsec_ctx->tbl[cdev_id_qp].qp);
 
-   if (ips->type != RTE_SECURITY_ACTION_TYPE_NONE) {
+   if (ips->type != RTE_SECURITY_ACTION_TYPE_NONE &&
+   ips->type != RTE_SECURITY_ACTION_TYPE_CPU_CRYPTO) {
struct rte_security_session_conf sess_conf = {
.action_type = ips->type,
.protocol = RTE_SECURITY_PROTOCOL_IPSEC,
@@ -126,6 +128,18 @@ create_lookaside_session(struct ipsec_ctx *ipsec_ctx, 
struct ipsec_sa *sa,
return -1;
}
} else {
+   if (ips->type == RTE_SECURITY_ACTION_TYPE_CPU_CRYPTO) {
+   struct rte_cryptodev_info info;
+   uint16_t cdev_id;
+
+   cdev_id = ipsec_ctx->tbl[cdev_id_qp].id;
+   rte_cryptodev_info_get(cdev_id, &info);
+   if (!(info.feature_flags &
+   RTE_CRYPTODEV_FF_SYM_CPU_CRYPTO))
+   return -ENOTSUP;
+
+   ips->crypto.dev_id = cdev_id;
+   }
ips->crypto.ses = rte_cryptodev_sym_session_create(
ipsec_ctx->session_pool);
rte_cryptodev_sym_session_init(ipsec_ctx->tbl[cdev_id_qp].id,
@@ -476,6 +490,13 @@ ipsec_enqueue(ipsec_xform_fn xform_func, struct ipsec_ctx 
*ipsec_ctx,
rte_security_attach_session(&priv->cop,
ips->security.ses);
break;
+
+   case RTE_SECURITY_ACTION_TYPE_CPU_CRYPTO:
+   RTE_LOG(ERR, IPSEC, "CPU crypto is not supported by the"
+   " legacy mode.");
+   rte_pktmbuf_free(pkts[i]);
+   continue;
+
case RTE_SECURITY_ACTION_TYPE_NONE:
 
priv->cop.type = RTE_CRYPTO_OP_TYPE_SYMMETRIC;
diff --git a/examples/ipsec-secgw/ipsec_process.c 
b/examples/ipsec-secgw/ipsec_process.c
index 2eb5c8b34..576a9fa8a 100644
--- a/examples/ipsec-secgw/ipsec_process.c
+++ b/examples/ipsec-secgw/ipsec_process.c
@@ -92,7 +92,8 @@ fill_ipsec_session(struct rte_ipsec_session *ss, struct 
ipsec_ctx *ctx,
int32_t rc;
 
/* setup crypto section */
-   if (ss->type == RTE_SECURITY_ACTION_TYPE_NONE) {
+   if (ss->type == RTE_SECURITY_ACTION_TYPE_NONE ||
+   ss->type == RTE_SECURITY_ACTION_TYPE_CPU_CRYPTO) {
RTE_ASSERT(ss->crypto.ses == NULL);
rc = create_lookaside_session(ctx, sa, ss);
if (rc != 0)
@@ -215,6 +216,62 @@ ipsec_prepare_crypto_group(struct ipsec_ctx *ctx, struct 
ipsec_sa *sa,
return k;
 }
 
+/*
+ * helper routine for inline and cpu(synchronous) processing
+ * this is just to satisfy inbound_sa_check() and get_hop_for_offload_pkt().
+ * Should be removed in future.
+ */
+static inline void
+prep_process_group(void *sa, struct rte_mbuf *mb[], uint32_t cnt)
+{
+   uint32_t j;
+   struct ipsec_mbuf_metadata *priv;
+
+   for (j = 0; j != cnt; j++) {
+   priv = get_priv(mb[j]);
+   priv->sa = sa;
+   }
+}
+
+/*
+ * finish processing of packets successfully decrypted by an inline processor
+ */
+static uint32_t
+ipsec_process_inline_group(struct rte_ipsec_session *ips, void *sa,
+   struct ipsec_traffic *trf, struct rte_mbuf *mb[], uint32_t cnt)
+{
+   uint64_t satp;
+   uint32_t k;
+
+   /* get SA type */
+   satp = rte_ipsec_sa_type(ips->sa);
+   prep_process_group(sa, mb, cnt);
+
+   k = rte_ipsec_pkt_process(ips, mb, cnt);
+   copy_to_trf(trf, satp, mb, k);
+   return k;
+}
+
+/*
+ * process packets synchronously
+ */
+static uint32_t
+ipsec_process_cpu_group(struct rte_ipsec_session *ips, void *sa,
+   struct ipsec_traffic *trf, struct rte_mbuf *mb[], uint32_t cnt)
+{
+   uint64_t satp;
+   uint32_t k;
+
+   /* get SA type */
+   satp = rte_ipsec_sa_type(ips->sa);
+   prep_pro

[dpdk-dev] [PATCH v4 7/8] examples/ipsec-secgw: cpu crypto testing

[Marcin Smoczynski]

Enable cpu-crypto mode testing by adding dedicated environmental
variable CRYPTO_PRIM_TYPE. Setting it to 'type cpu-crypto' allows
to run test scenario with cpu crypto acceleration.

Signed-off-by: Konstantin Ananyev 
Signed-off-by: Marcin Smoczynski 
---
 examples/ipsec-secgw/test/common_defs.sh  | 21 +++
 examples/ipsec-secgw/test/linux_test4.sh  | 11 +-
 examples/ipsec-secgw/test/linux_test6.sh  | 11 +-
 .../test/trs_3descbc_sha1_common_defs.sh  |  8 +++
 .../test/trs_aescbc_sha1_common_defs.sh   |  8 +++
 .../test/trs_aesctr_sha1_common_defs.sh   |  8 +++
 .../test/tun_3descbc_sha1_common_defs.sh  |  8 +++
 .../test/tun_aescbc_sha1_common_defs.sh   |  8 +++
 .../test/tun_aesctr_sha1_common_defs.sh   |  8 +++
 9 files changed, 47 insertions(+), 44 deletions(-)

diff --git a/examples/ipsec-secgw/test/common_defs.sh 
b/examples/ipsec-secgw/test/common_defs.sh
index 4aac4981a..6b6ae06f3 100644
--- a/examples/ipsec-secgw/test/common_defs.sh
+++ b/examples/ipsec-secgw/test/common_defs.sh
@@ -42,6 +42,27 @@ DPDK_BUILD=${RTE_TARGET:-x86_64-native-linux-gcc}
 DEF_MTU_LEN=1400
 DEF_PING_LEN=1200
 
+#upsate operation mode based on env vars values
+select_mode()
+{
+   # select sync/async mode
+   if [[ -n "${CRYPTO_PRIM_TYPE}" && -n "${SGW_CMD_XPRM}" ]]; then
+   echo "${CRYPTO_PRIM_TYPE} is enabled"
+   SGW_CFG_XPRM="${SGW_CFG_XPRM} ${CRYPTO_PRIM_TYPE}"
+   fi
+
+   #make linux to generate fragmented packets
+   if [[ -n "${MULTI_SEG_TEST}" && -n "${SGW_CMD_XPRM}" ]]; then
+   echo "multi-segment test is enabled"
+   SGW_CMD_XPRM="${SGW_CMD_XPRM} ${MULTI_SEG_TEST}"
+   PING_LEN=5000
+   MTU_LEN=1500
+   else
+   PING_LEN=${DEF_PING_LEN}
+   MTU_LEN=${DEF_MTU_LEN}
+   fi
+}
+
 #setup mtu on local iface
 set_local_mtu()
 {
diff --git a/examples/ipsec-secgw/test/linux_test4.sh 
b/examples/ipsec-secgw/test/linux_test4.sh
index 760451000..fb8ae1023 100644
--- a/examples/ipsec-secgw/test/linux_test4.sh
+++ b/examples/ipsec-secgw/test/linux_test4.sh
@@ -45,16 +45,7 @@ MODE=$1
  . ${DIR}/common_defs.sh
  . ${DIR}/${MODE}_defs.sh
 
-#make linux to generate fragmented packets
-if [[ -n "${MULTI_SEG_TEST}" && -n "${SGW_CMD_XPRM}" ]]; then
-   echo "multi-segment test is enabled"
-   SGW_CMD_XPRM="${SGW_CMD_XPRM} ${MULTI_SEG_TEST}"
-   PING_LEN=5000
-   MTU_LEN=1500
-else
-   PING_LEN=${DEF_PING_LEN}
-   MTU_LEN=${DEF_MTU_LEN}
-fi
+select_mode
 
 config_secgw
 
diff --git a/examples/ipsec-secgw/test/linux_test6.sh 
b/examples/ipsec-secgw/test/linux_test6.sh
index 479f29be3..dbcca7936 100644
--- a/examples/ipsec-secgw/test/linux_test6.sh
+++ b/examples/ipsec-secgw/test/linux_test6.sh
@@ -46,16 +46,7 @@ MODE=$1
  . ${DIR}/common_defs.sh
  . ${DIR}/${MODE}_defs.sh
 
-#make linux to generate fragmented packets
-if [[ -n "${MULTI_SEG_TEST}" && -n "${SGW_CMD_XPRM}" ]]; then
-   echo "multi-segment test is enabled"
-   SGW_CMD_XPRM="${SGW_CMD_XPRM} ${MULTI_SEG_TEST}"
-   PING_LEN=5000
-   MTU_LEN=1500
-else
-   PING_LEN=${DEF_PING_LEN}
-   MTU_LEN=${DEF_MTU_LEN}
-fi
+select_mode
 
 config_secgw
 
diff --git a/examples/ipsec-secgw/test/trs_3descbc_sha1_common_defs.sh 
b/examples/ipsec-secgw/test/trs_3descbc_sha1_common_defs.sh
index 3c5c18afd..62118bb3f 100644
--- a/examples/ipsec-secgw/test/trs_3descbc_sha1_common_defs.sh
+++ b/examples/ipsec-secgw/test/trs_3descbc_sha1_common_defs.sh
@@ -33,14 +33,14 @@ cipher_key \
 de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef \
 auth_algo sha1-hmac \
 auth_key de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef \
-mode transport
+mode transport ${SGW_CFG_XPRM}
 
 sa in 9 cipher_algo 3des-cbc \
 cipher_key \
 de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef \
 auth_algo sha1-hmac \
 auth_key de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef \
-mode transport
+mode transport ${SGW_CFG_XPRM}
 
 #SA out rules
 sa out 7 cipher_algo 3des-cbc \
@@ -48,7 +48,7 @@ cipher_key \
 de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef \
 auth_algo sha1-hmac \
 auth_key de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef \
-mode transport
+mode transport ${SGW_CFG_XPRM}
 
 #SA out rules
 sa out 9 cipher_algo 3des-cbc \
@@ -56,7 +56,7 @@ cipher_key \
 de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef \
 auth_algo sha1-hmac \
 auth_key de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef:de:ad:be:ef \
-mode transport
+mode transport ${SGW_CFG_XPRM}
 
 #Routing rules
 rt ipv4 dst ${REMOTE_IPV4}/32 port 0
diff --git a/examples/ipsec-secgw/test/trs_aescbc_sha1_common_defs.sh 
b/examples/ipsec-secgw/test/trs_aescbc_sha1_common_defs.sh
index 9dbdd1765..7ddeb2b5a 100644
--- a/examples/ipsec-secgw/test/trs_aescbc_sha1_common_defs.sh
+++ b/examples/i

Re: [dpdk-dev] [PATCH v2 00/12] add eventmode to ipsec-secgw

[Anoob Joseph]

Hi Akhil, Konstantin,

Do you have any further comments?

Thanks,
Anoob

> -Original Message-
> From: dev  On Behalf Of Anoob Joseph
> Sent: Monday, January 20, 2020 7:15 PM
> To: Akhil Goyal ; Radu Nicolau
> ; Thomas Monjalon 
> Cc: Anoob Joseph ; Jerin Jacob Kollanukkaran
> ; Narayana Prasad Raju Athreya
> ; Ankur Dwivedi ;
> Archana Muniganti ; Tejasree Kondoj
> ; Vamsi Krishna Attunuru
> ; Lukas Bartosik ;
> Konstantin Ananyev ; dev@dpdk.org
> Subject: [dpdk-dev] [PATCH v2 00/12] add eventmode to ipsec-secgw
> 
> This series introduces event-mode additions to ipsec-secgw. This effort is
> parallel to the similar changes in l2fwd (l2fwd-event app) & l3fwd.
> 
> With this series, ipsec-secgw would be able to run in eventmode. The worker
> thread (executing loop) would be receiving events and would be submitting
> it back to the eventdev after the processing. This way, multicore scaling and
> h/w assisted scheduling is achieved by making use of the eventdev
> capabilities.
> 
> Since the underlying event device would be having varying capabilities, the
> worker thread could be drafted differently to maximize performance.
> This series introduces usage of multiple worker threads, among which the
> one to be used will be determined by the operating conditions and the
> underlying device capabilities.
> 
> For example, if an event device - eth device pair has Tx internal port, then
> application can do tx_adapter_enqueue() instead of regular
> event_enqueue(). So a thread making an assumption that the device pair has
> internal port will not be the right solution for another pair. The 
> infrastructure
> added with these patches aims to help application to have multiple worker
> threads, there by extracting maximum performance from every device
> without affecting existing paths/use cases.
> 
> The eventmode configuration is predefined. All packets reaching one eth
> port will hit one event queue. All event queues will be mapped to all event
> ports. So all cores will be able to receive traffic from all ports.
> When schedule_type is set as RTE_SCHED_TYPE_ORDERED/ATOMIC, event
> device will ensure the ordering. Ordering would be lost when tried in
> PARALLEL.
> 
> Following command line options are introduced,
> 
> --transfer-mode: to choose between poll mode & event mode
> --schedule-type: to specify the scheduling type
>  (RTE_SCHED_TYPE_ORDERED/
>   RTE_SCHED_TYPE_ATOMIC/
>   RTE_SCHED_TYPE_PARALLEL)
> 
> Additionally the event mode introduces two modes of processing packets:
> 
> Driver-mode: This mode will have bare minimum changes in the application
>  to support ipsec. There woudn't be any lookup etc done in
>  the application. And for inline-protocol use case, the
>  thread would resemble l2fwd as the ipsec processing would be
>  done entirely in the h/w. This mode can be used to benchmark
>  the raw performance of the h/w. All the application side
>  steps (like lookup) can be redone based on the requirement
>  of the end user. Hence the need for a mode which would
>  report the raw performance.
> 
> App-mode: This mode will have all the features currently implemented with
>   ipsec-secgw (non librte_ipsec mode). All the lookups etc
>   would follow the existing methods and would report numbers
>   that can be compared against regular ipsec-secgw benchmark
>   numbers.
> 
> The driver mode is selected with existing --single-sa option (used also by 
> poll
> mode). When --single-sa option is used in conjution with event mode then
> index passed to --single-sa is ignored.
> 
> Example commands to execute ipsec-secgw in various modes on OCTEON
> TX2 platform,
> 
> #Inbound and outbound app mode
> ipsec-secgw -w 0002:02:00.0,ipsec_in_max_spi=128 -w
> 0002:03:00.0,ipsec_in_max_spi=128 -w 0002:0e:00.0 -w 0002:10:00.1 --log-
> level=8 -c 0x1 -- -P -p 0x3 -u 0x1 --config "(1,0,0),(0,0,0)" -f aes-gcm.cfg 
> --
> transfer-mode event --schedule-type parallel
> 
> #Inbound and outbound driver mode
> ipsec-secgw -w 0002:02:00.0,ipsec_in_max_spi=128 -w
> 0002:03:00.0,ipsec_in_max_spi=128 -w 0002:0e:00.0 -w 0002:10:00.1 --log-
> level=8 -c 0x1 -- -P -p 0x3 -u 0x1 --config "(1,0,0),(0,0,0)" -f aes-gcm.cfg 
> --
> transfer-mode event --schedule-type parallel --single-sa 0
> 
> This series adds non burst tx internal port workers only. It provides
> infrastructure for non internal port workers, however does not define any.
> Also, only inline ipsec protocol mode is supported by the worker threads
> added.
> 
> Following are planned features,
> 1. Add burst mode workers.
> 2. Add non internal port workers.
> 3. Verify support for Rx core (the support is added but lack of h/w to 
> verify).
> 4. Add lookaside protocol support.
> 
> Following are features that Marvell won't be attempting.
> 1. Inline crypto support.
> 2. Lookaside crypto su

[dpdk-dev] [PATCH v6 00/11] example/l3fwd: introduce event device support

[pbhagavatula]

From: Pavan Nikhilesh 

This patchset adds support for event mode in l3fwd.
The poll mode is still the default mode of operation.

The following new command line parameters are added:
 --mode: Dictates the mode of operation either poll or event. If unset then
 by default poll mode is used.
 --eventq_sched: Dictates event synchronization mode i.e. either ordered,
 atomic or parallel.
 --event-eth-rxqs: Number of ethernet RX queues per device.

Based on event device capability the configuration is done as follows:
- A single event device is enabled.
- The number of event ports is equal to the number of worker
  cores enabled in the core mask. Additional event ports might
  be configured based on Rx/Tx adapter capability.
- The number of event queues is equal to the number of ethernet
  ports. If Tx adapter doesn't have internal port capability then
  an additional single link event queue is used to enqueue events
  to Tx adapter.
- Each event port is linked to all existing event queues.
- Dedicated Rx/Tx adapters for each Ethernet port.

v6 Changes:
--
- Fix typo in comments.

v5 Changes:
--
- Update release notes and MAINTAINERS file.
- Fix typo in doc.

v4 Changes:
--
- Update documentation about --event-eth-rxqs. (Jerin)
- Add validation for command line options that are specific to event/poll mode.
- Fix event device service initialization.

v3 Changes:
--
- Unify argument parsing.
- Segregate poll mode and event mode initialization.
- Simplify event resource creation and accesses.
- Integrate http://patches.dpdk.org/project/dpdk/list/?series=8002 series.
- Reduce code duplication in lpm.

v2 Changes:
--
- Shorten the structure name `s/event_rsrc/evt_rsrc` `s/l2fwd_rsrc/rsrc`.
- Use rte_panic instead of rte_exit.
- Rebase on top of Tx adapter change http://patches.dpdk.org/patch/60971.
- Add parallel mode support.
- Fix missing eventdev args parsing.


Pavan Nikhilesh (7):
  examples/l3fwd: add event device configuration
  examples/l3fwd: add event eth Rx/Tx adapter setup
  examples/l3fwd: add service core setup based on caps
  examples/l3fwd: add event lpm main loop
  examples/l3fwd: add event em main loop
  examples/l3fwd: add graceful teardown for eventdevice
  doc: update l3fwd user guide to support eventdev

Sunil Kumar Kori (4):
  examples/l3fwd: add framework for event device
  examples/l3fwd: split pipelines based on capability
  examples/l3fwd: add ethdev setup based on eventdev
  examples/l3fwd: add event port and queue setup

 MAINTAINERS|   1 +
 doc/guides/rel_notes/release_20_02.rst |   5 +
 doc/guides/sample_app_ug/l3_forward.rst|  79 -
 examples/l3fwd/Makefile|   3 +-
 examples/l3fwd/l3fwd.h |  30 ++
 examples/l3fwd/l3fwd_em.c  | 177 +++
 examples/l3fwd/l3fwd_em.h  | 159 +++---
 examples/l3fwd/l3fwd_em_hlm.h  | 131 
 examples/l3fwd/l3fwd_em_sequential.h   |  26 ++
 examples/l3fwd/l3fwd_event.c   | 263 
 examples/l3fwd/l3fwd_event.h   |  86 ++
 examples/l3fwd/l3fwd_event_generic.c   | 303 ++
 examples/l3fwd/l3fwd_event_internal_port.c | 279 +
 examples/l3fwd/l3fwd_lpm.c | 203 
 examples/l3fwd/main.c  | 341 +
 examples/l3fwd/meson.build |   5 +-
 16 files changed, 1976 insertions(+), 115 deletions(-)
 create mode 100644 examples/l3fwd/l3fwd_event.c
 create mode 100644 examples/l3fwd/l3fwd_event.h
 create mode 100644 examples/l3fwd/l3fwd_event_generic.c
 create mode 100644 examples/l3fwd/l3fwd_event_internal_port.c

--
2.17.1

[dpdk-dev] [PATCH v6 02/11] examples/l3fwd: split pipelines based on capability

[pbhagavatula]

From: Sunil Kumar Kori 

Add infra to split eventdev framework based on event Tx adapter
capability.
If event Tx adapter has internal port capability then we use
`rte_event_eth_tx_adapter_enqueue` to transmitting packets else
we use a SINGLE_LINK event queue to enqueue packets to a service
core which is responsible for transmitting packets.

Signed-off-by: Sunil Kumar Kori 
Acked-by: Jerin Jacob 
---
 examples/l3fwd/Makefile|  1 +
 examples/l3fwd/l3fwd_event.c   | 31 ++
 examples/l3fwd/l3fwd_event.h   | 20 ++
 examples/l3fwd/l3fwd_event_generic.c   | 14 ++
 examples/l3fwd/l3fwd_event_internal_port.c | 14 ++
 examples/l3fwd/meson.build |  3 ++-
 6 files changed, 82 insertions(+), 1 deletion(-)
 create mode 100644 examples/l3fwd/l3fwd_event_generic.c
 create mode 100644 examples/l3fwd/l3fwd_event_internal_port.c

diff --git a/examples/l3fwd/Makefile b/examples/l3fwd/Makefile
index c892b867b..59a110d12 100644
--- a/examples/l3fwd/Makefile
+++ b/examples/l3fwd/Makefile
@@ -6,6 +6,7 @@ APP = l3fwd
 
 # all source are stored in SRCS-y
 SRCS-y := main.c l3fwd_lpm.c l3fwd_em.c l3fwd_event.c
+SRCS-y += l3fwd_event_generic.c l3fwd_event_internal_port.c
 
 # Build using pkg-config variables if possible
 ifeq ($(shell pkg-config --exists libdpdk && echo 0),0)
diff --git a/examples/l3fwd/l3fwd_event.c b/examples/l3fwd/l3fwd_event.c
index 1040da4ea..62218f3ca 100644
--- a/examples/l3fwd/l3fwd_event.c
+++ b/examples/l3fwd/l3fwd_event.c
@@ -30,6 +30,31 @@ l3fwd_get_eventdev_rsrc(void)
return NULL;
 }
 
+static void
+l3fwd_event_capability_setup(void)
+{
+   struct l3fwd_event_resources *evt_rsrc = l3fwd_get_eventdev_rsrc();
+   uint32_t caps = 0;
+   uint16_t i;
+   int ret;
+
+   RTE_ETH_FOREACH_DEV(i) {
+   ret = rte_event_eth_tx_adapter_caps_get(0, i, &caps);
+   if (ret)
+   rte_exit(EXIT_FAILURE,
+"Invalid capability for Tx adptr port %d\n",
+i);
+
+   evt_rsrc->tx_mode_q |= !(caps &
+  RTE_EVENT_ETH_TX_ADAPTER_CAP_INTERNAL_PORT);
+   }
+
+   if (evt_rsrc->tx_mode_q)
+   l3fwd_event_set_generic_ops(&evt_rsrc->ops);
+   else
+   l3fwd_event_set_internal_port_ops(&evt_rsrc->ops);
+}
+
 void
 l3fwd_event_resource_setup(void)
 {
@@ -37,4 +62,10 @@ l3fwd_event_resource_setup(void)
 
if (!evt_rsrc->enabled)
return;
+
+   if (!rte_event_dev_count())
+   rte_exit(EXIT_FAILURE, "No Eventdev found");
+
+   /* Setup eventdev capability callbacks */
+   l3fwd_event_capability_setup();
 }
diff --git a/examples/l3fwd/l3fwd_event.h b/examples/l3fwd/l3fwd_event.h
index 4c23c4e1a..d25c8d222 100644
--- a/examples/l3fwd/l3fwd_event.h
+++ b/examples/l3fwd/l3fwd_event.h
@@ -7,17 +7,37 @@
 
 #include 
 #include 
+#include 
 #include 
 
 #include "l3fwd.h"
 
+typedef uint32_t (*event_device_setup_cb)(void);
+typedef void (*event_queue_setup_cb)(uint32_t event_queue_cfg);
+typedef void (*event_port_setup_cb)(void);
+typedef void (*adapter_setup_cb)(void);
+typedef int (*event_loop_cb)(void *);
+
+struct l3fwd_event_setup_ops {
+   event_device_setup_cb event_device_setup;
+   event_queue_setup_cb event_queue_setup;
+   event_port_setup_cb event_port_setup;
+   adapter_setup_cb adapter_setup;
+   event_loop_cb lpm_event_loop;
+   event_loop_cb em_event_loop;
+};
+
 struct l3fwd_event_resources {
+   struct l3fwd_event_setup_ops ops;
uint8_t sched_type;
+   uint8_t tx_mode_q;
uint8_t enabled;
uint8_t eth_rx_queues;
 };
 
 struct l3fwd_event_resources *l3fwd_get_eventdev_rsrc(void);
 void l3fwd_event_resource_setup(void);
+void l3fwd_event_set_generic_ops(struct l3fwd_event_setup_ops *ops);
+void l3fwd_event_set_internal_port_ops(struct l3fwd_event_setup_ops *ops);
 
 #endif /* __L3FWD_EVENTDEV_H__ */
diff --git a/examples/l3fwd/l3fwd_event_generic.c 
b/examples/l3fwd/l3fwd_event_generic.c
new file mode 100644
index 0..7fff850e5
--- /dev/null
+++ b/examples/l3fwd/l3fwd_event_generic.c
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(C) 2019 Marvell International Ltd.
+ */
+
+#include 
+
+#include "l3fwd.h"
+#include "l3fwd_event.h"
+
+void
+l3fwd_event_set_generic_ops(struct l3fwd_event_setup_ops *ops)
+{
+   RTE_SET_USED(ops);
+}
diff --git a/examples/l3fwd/l3fwd_event_internal_port.c 
b/examples/l3fwd/l3fwd_event_internal_port.c
new file mode 100644
index 0..085e9c825
--- /dev/null
+++ b/examples/l3fwd/l3fwd_event_internal_port.c
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(C) 2019 Marvell International Ltd.
+ */
+
+#include 
+
+#include "l3fwd.h"
+#include "l3fwd_event.h"
+
+void
+l3fwd_event_set_internal_port_ops(struct l3fwd_event_setup_ops *op

[dpdk-dev] [PATCH v6 01/11] examples/l3fwd: add framework for event device

[pbhagavatula]

From: Sunil Kumar Kori 

Add framework to enable event device as a producer of packets.
To switch between event mode and poll mode the following options
have been added:
`--mode="eventdev"` or `--mode="poll"`
Allow the user to select the schedule type to be either
RTE_SCHED_TYPE_ORDERED, RTE_SCHED_TYPE_ATOMIC or RTE_SCHED_TYPE_PARALLEL
through:
`--eventq-sched="ordered"` or `--eventq-sched="atomic"` or
`--eventq-sched="parallel"`
Allow the user to specify the number of Rx queues to be connected to
event queue using:
`--event-eth-rxqs`

Poll mode is still the default operation mode.

Signed-off-by: Sunil Kumar Kori 
Signed-off-by: Nipun Gupta 
Series-Acked-by: Jerin Jacob 
---
 examples/l3fwd/Makefile  |   2 +-
 examples/l3fwd/l3fwd.h   |   1 +
 examples/l3fwd/l3fwd_event.c |  40 +
 examples/l3fwd/l3fwd_event.h |  23 
 examples/l3fwd/main.c| 110 ++-
 examples/l3fwd/meson.build   |   4 +-
 6 files changed, 174 insertions(+), 6 deletions(-)
 create mode 100644 examples/l3fwd/l3fwd_event.c
 create mode 100644 examples/l3fwd/l3fwd_event.h

diff --git a/examples/l3fwd/Makefile b/examples/l3fwd/Makefile
index b2dbf2607..c892b867b 100644
--- a/examples/l3fwd/Makefile
+++ b/examples/l3fwd/Makefile
@@ -5,7 +5,7 @@
 APP = l3fwd
 
 # all source are stored in SRCS-y
-SRCS-y := main.c l3fwd_lpm.c l3fwd_em.c
+SRCS-y := main.c l3fwd_lpm.c l3fwd_em.c l3fwd_event.c
 
 # Build using pkg-config variables if possible
 ifeq ($(shell pkg-config --exists libdpdk && echo 0),0)
diff --git a/examples/l3fwd/l3fwd.h b/examples/l3fwd/l3fwd.h
index 293fb1fa2..e2399d76b 100644
--- a/examples/l3fwd/l3fwd.h
+++ b/examples/l3fwd/l3fwd.h
@@ -5,6 +5,7 @@
 #ifndef __L3_FWD_H__
 #define __L3_FWD_H__
 
+#include 
 #include 
 
 #define DO_RFC_1812_CHECKS
diff --git a/examples/l3fwd/l3fwd_event.c b/examples/l3fwd/l3fwd_event.c
new file mode 100644
index 0..1040da4ea
--- /dev/null
+++ b/examples/l3fwd/l3fwd_event.c
@@ -0,0 +1,40 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(C) 2019 Marvell International Ltd.
+ */
+
+#include 
+#include 
+
+#include 
+
+#include "l3fwd.h"
+#include "l3fwd_event.h"
+
+struct l3fwd_event_resources *
+l3fwd_get_eventdev_rsrc(void)
+{
+   static struct l3fwd_event_resources *rsrc;
+
+   if (rsrc != NULL)
+   return rsrc;
+
+   rsrc = rte_zmalloc("l3fwd", sizeof(struct l3fwd_event_resources), 0);
+   if (rsrc != NULL) {
+   rsrc->sched_type = RTE_SCHED_TYPE_ATOMIC;
+   rsrc->eth_rx_queues = 1;
+   return rsrc;
+   }
+
+   rte_exit(EXIT_FAILURE, "Unable to allocate memory for eventdev cfg\n");
+
+   return NULL;
+}
+
+void
+l3fwd_event_resource_setup(void)
+{
+   struct l3fwd_event_resources *evt_rsrc = l3fwd_get_eventdev_rsrc();
+
+   if (!evt_rsrc->enabled)
+   return;
+}
diff --git a/examples/l3fwd/l3fwd_event.h b/examples/l3fwd/l3fwd_event.h
new file mode 100644
index 0..4c23c4e1a
--- /dev/null
+++ b/examples/l3fwd/l3fwd_event.h
@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(C) 2019 Marvell International Ltd.
+ */
+
+#ifndef __L3FWD_EVENTDEV_H__
+#define __L3FWD_EVENTDEV_H__
+
+#include 
+#include 
+#include 
+
+#include "l3fwd.h"
+
+struct l3fwd_event_resources {
+   uint8_t sched_type;
+   uint8_t enabled;
+   uint8_t eth_rx_queues;
+};
+
+struct l3fwd_event_resources *l3fwd_get_eventdev_rsrc(void);
+void l3fwd_event_resource_setup(void);
+
+#endif /* __L3FWD_EVENTDEV_H__ */
diff --git a/examples/l3fwd/main.c b/examples/l3fwd/main.c
index 4dea12a65..8a5cdb52e 100644
--- a/examples/l3fwd/main.c
+++ b/examples/l3fwd/main.c
@@ -19,6 +19,7 @@
 #include 
 #include 
 #include 
+#include 
 #include 
 #include 
 #include 
@@ -33,7 +34,6 @@
 #include 
 #include 
 #include 
-#include 
 #include 
 #include 
 #include 
@@ -46,6 +46,7 @@
 #include 
 
 #include "l3fwd.h"
+#include "l3fwd_event.h"
 
 /*
  * Configurable number of RX/TX ring descriptors
@@ -289,7 +290,9 @@ print_usage(const char *prgname)
" [--hash-entry-num]"
" [--ipv6]"
" [--parse-ptype]"
-   " [--per-port-pool]\n\n"
+   " [--per-port-pool]"
+   " [--mode]"
+   " [--eventq-sched]\n\n"
 
"  -p PORTMASK: Hexadecimal bitmask of ports to configure\n"
"  -P : Enable promiscuous mode\n"
@@ -304,7 +307,16 @@ print_usage(const char *prgname)
"  --hash-entry-num: Specify the hash entry number in 
hexadecimal to be setup\n"
"  --ipv6: Set if running ipv6 packets\n"
"  --parse-ptype: Set to use software to analyze packet type\n"
-   "  --per-port-pool: Use separate buffer pool per port\n\n",
+   "  --per-port-pool: Use separate buffer pool per port\n"
+   "  --mode: Packet transfer mode for I/O, poll

[dpdk-dev] [PATCH v6 04/11] examples/l3fwd: add ethdev setup based on eventdev

[pbhagavatula]

From: Sunil Kumar Kori 

Add ethernet port Rx/Tx queue setup for event device which are later
used for setting up event eth Rx/Tx adapters.

Signed-off-by: Sunil Kumar Kori 
Signed-off-by: Nipun Gupta 
---
 examples/l3fwd/l3fwd.h   |  10 +++
 examples/l3fwd/l3fwd_event.c | 138 ++-
 examples/l3fwd/l3fwd_event.h |   2 +-
 examples/l3fwd/main.c|  15 ++--
 4 files changed, 153 insertions(+), 12 deletions(-)

diff --git a/examples/l3fwd/l3fwd.h b/examples/l3fwd/l3fwd.h
index e2399d76b..ed5ba5f3b 100644
--- a/examples/l3fwd/l3fwd.h
+++ b/examples/l3fwd/l3fwd.h
@@ -16,9 +16,16 @@
 #define NO_HASH_MULTI_LOOKUP 1
 #endif
 
+/*
+ * Configurable number of RX/TX ring descriptors
+ */
+#define RTE_TEST_RX_DESC_DEFAULT 1024
+#define RTE_TEST_TX_DESC_DEFAULT 1024
+
 #define MAX_PKT_BURST 32
 #define BURST_TX_DRAIN_US 100 /* TX drain every ~100us */
 
+#define MEMPOOL_CACHE_SIZE 256
 #define MAX_RX_QUEUE_PER_LCORE 16
 
 /*
@@ -170,6 +177,9 @@ is_valid_ipv4_pkt(struct rte_ipv4_hdr *pkt, uint32_t 
link_len)
 }
 #endif /* DO_RFC_1812_CHECKS */
 
+int
+init_mem(uint16_t portid, unsigned int nb_mbuf);
+
 /* Function pointers for LPM or EM functionality. */
 void
 setup_lpm(const int socketid);
diff --git a/examples/l3fwd/l3fwd_event.c b/examples/l3fwd/l3fwd_event.c
index 9cb46c711..f9491ecc6 100644
--- a/examples/l3fwd/l3fwd_event.c
+++ b/examples/l3fwd/l3fwd_event.c
@@ -10,6 +10,14 @@
 #include "l3fwd.h"
 #include "l3fwd_event.h"
 
+static void
+print_ethaddr(const char *name, const struct rte_ether_addr *eth_addr)
+{
+   char buf[RTE_ETHER_ADDR_FMT_SIZE];
+   rte_ether_format_addr(buf, RTE_ETHER_ADDR_FMT_SIZE, eth_addr);
+   printf("%s%s", name, buf);
+}
+
 struct l3fwd_event_resources *
 l3fwd_get_eventdev_rsrc(void)
 {
@@ -30,6 +38,131 @@ l3fwd_get_eventdev_rsrc(void)
return NULL;
 }
 
+static void
+l3fwd_eth_dev_port_setup(struct rte_eth_conf *port_conf)
+{
+   struct l3fwd_event_resources *evt_rsrc = l3fwd_get_eventdev_rsrc();
+   uint16_t nb_ports = rte_eth_dev_count_avail();
+   uint16_t nb_rxd = RTE_TEST_RX_DESC_DEFAULT;
+   uint16_t nb_txd = RTE_TEST_TX_DESC_DEFAULT;
+   unsigned int nb_lcores = rte_lcore_count();
+   struct rte_eth_conf local_port_conf;
+   struct rte_eth_dev_info dev_info;
+   struct rte_eth_txconf txconf;
+   struct rte_eth_rxconf rxconf;
+   unsigned int nb_mbuf;
+   uint16_t port_id;
+   uint8_t eth_qid;
+   int32_t ret;
+
+   /* initialize all ports */
+   RTE_ETH_FOREACH_DEV(port_id) {
+   local_port_conf = *port_conf;
+   /* skip ports that are not enabled */
+   if ((evt_rsrc->port_mask & (1 << port_id)) == 0) {
+   printf("\nSkipping disabled port %d\n", port_id);
+   continue;
+   }
+
+   /* init port */
+   printf("Initializing port %d ... ", port_id);
+   fflush(stdout);
+   printf("Creating queues: nb_rxq=%d nb_txq=1...\n",
+  evt_rsrc->eth_rx_queues);
+
+   rte_eth_dev_info_get(port_id, &dev_info);
+   if (dev_info.tx_offload_capa & DEV_TX_OFFLOAD_MBUF_FAST_FREE)
+   local_port_conf.txmode.offloads |=
+   DEV_TX_OFFLOAD_MBUF_FAST_FREE;
+
+   local_port_conf.rx_adv_conf.rss_conf.rss_hf &=
+   dev_info.flow_type_rss_offloads;
+   if (local_port_conf.rx_adv_conf.rss_conf.rss_hf !=
+   port_conf->rx_adv_conf.rss_conf.rss_hf) {
+   printf("Port %u modified RSS hash function "
+  "based on hardware support,"
+  "requested:%#"PRIx64" configured:%#"PRIx64"\n",
+  port_id,
+  port_conf->rx_adv_conf.rss_conf.rss_hf,
+  local_port_conf.rx_adv_conf.rss_conf.rss_hf);
+   }
+
+   ret = rte_eth_dev_configure(port_id, evt_rsrc->eth_rx_queues,
+   1, &local_port_conf);
+   if (ret < 0)
+   rte_exit(EXIT_FAILURE,
+"Cannot configure device: err=%d, port=%d\n",
+ret, port_id);
+
+   ret = rte_eth_dev_adjust_nb_rx_tx_desc(port_id, &nb_rxd,
+  &nb_txd);
+   if (ret < 0)
+   rte_exit(EXIT_FAILURE,
+"Cannot adjust number of descriptors: err=%d, "
+"port=%d\n", ret, port_id);
+
+   rte_eth_macaddr_get(port_id, &ports_eth_addr[port_id]);
+   print_ethaddr(" Address:", &ports_eth_addr[port_id]);
+   printf(", ");
+   print_ethaddr("Des

[dpdk-dev] [PATCH v6 03/11] examples/l3fwd: add event device configuration

[pbhagavatula]

From: Pavan Nikhilesh 

Add event device configuration based on the capabilities of the
probed event device.

Signed-off-by: Pavan Nikhilesh 
---
 examples/l3fwd/l3fwd_event.c   |  3 +
 examples/l3fwd/l3fwd_event.h   | 36 ++
 examples/l3fwd/l3fwd_event_generic.c   | 76 +-
 examples/l3fwd/l3fwd_event_internal_port.c | 76 +-
 4 files changed, 189 insertions(+), 2 deletions(-)

diff --git a/examples/l3fwd/l3fwd_event.c b/examples/l3fwd/l3fwd_event.c
index 62218f3ca..9cb46c711 100644
--- a/examples/l3fwd/l3fwd_event.c
+++ b/examples/l3fwd/l3fwd_event.c
@@ -68,4 +68,7 @@ l3fwd_event_resource_setup(void)
 
/* Setup eventdev capability callbacks */
l3fwd_event_capability_setup();
+
+   /* Event device configuration */
+   evt_rsrc->ops.event_device_setup();
 }
diff --git a/examples/l3fwd/l3fwd_event.h b/examples/l3fwd/l3fwd_event.h
index d25c8d222..fc3862ec1 100644
--- a/examples/l3fwd/l3fwd_event.h
+++ b/examples/l3fwd/l3fwd_event.h
@@ -8,6 +8,7 @@
 #include 
 #include 
 #include 
+#include 
 #include 
 
 #include "l3fwd.h"
@@ -18,6 +19,29 @@ typedef void (*event_port_setup_cb)(void);
 typedef void (*adapter_setup_cb)(void);
 typedef int (*event_loop_cb)(void *);
 
+struct l3fwd_event_queues {
+   uint8_t *event_q_id;
+   uint8_t nb_queues;
+};
+
+struct l3fwd_event_ports {
+   uint8_t *event_p_id;
+   uint8_t nb_ports;
+   rte_spinlock_t lock;
+};
+
+struct l3fwd_event_rx_adptr {
+   uint32_t service_id;
+   uint8_t nb_rx_adptr;
+   uint8_t *rx_adptr;
+};
+
+struct l3fwd_event_tx_adptr {
+   uint32_t service_id;
+   uint8_t nb_tx_adptr;
+   uint8_t *tx_adptr;
+};
+
 struct l3fwd_event_setup_ops {
event_device_setup_cb event_device_setup;
event_queue_setup_cb event_queue_setup;
@@ -28,9 +52,21 @@ struct l3fwd_event_setup_ops {
 };
 
 struct l3fwd_event_resources {
+   struct rte_event_port_conf def_p_conf;
+   struct l3fwd_event_rx_adptr rx_adptr;
+   struct l3fwd_event_tx_adptr tx_adptr;
+   uint8_t disable_implicit_release;
struct l3fwd_event_setup_ops ops;
+   struct rte_mempool * (*pkt_pool)[NB_SOCKETS];
+   struct l3fwd_event_queues evq;
+   struct l3fwd_event_ports evp;
+   uint32_t port_mask;
+   uint8_t per_port_pool;
+   uint8_t event_d_id;
uint8_t sched_type;
uint8_t tx_mode_q;
+   uint8_t deq_depth;
+   uint8_t has_burst;
uint8_t enabled;
uint8_t eth_rx_queues;
 };
diff --git a/examples/l3fwd/l3fwd_event_generic.c 
b/examples/l3fwd/l3fwd_event_generic.c
index 7fff850e5..ce06fa0e4 100644
--- a/examples/l3fwd/l3fwd_event_generic.c
+++ b/examples/l3fwd/l3fwd_event_generic.c
@@ -7,8 +7,82 @@
 #include "l3fwd.h"
 #include "l3fwd_event.h"
 
+static uint32_t
+l3fwd_event_device_setup_generic(void)
+{
+   struct l3fwd_event_resources *evt_rsrc = l3fwd_get_eventdev_rsrc();
+   struct rte_event_dev_config event_d_conf = {
+   .nb_events_limit  = 4096,
+   .nb_event_queue_flows = 1024,
+   .nb_event_port_dequeue_depth = 128,
+   .nb_event_port_enqueue_depth = 128
+   };
+   struct rte_event_dev_info dev_info;
+   const uint8_t event_d_id = 0; /* Always use first event device only */
+   uint32_t event_queue_cfg = 0;
+   uint16_t ethdev_count = 0;
+   uint16_t num_workers = 0;
+   uint16_t port_id;
+   int ret;
+
+   RTE_ETH_FOREACH_DEV(port_id) {
+   if ((evt_rsrc->port_mask & (1 << port_id)) == 0)
+   continue;
+   ethdev_count++;
+   }
+
+   /* Event device configuration */
+   rte_event_dev_info_get(event_d_id, &dev_info);
+   /* Enable implicit release */
+   if (dev_info.event_dev_cap & RTE_EVENT_DEV_CAP_IMPLICIT_RELEASE_DISABLE)
+   evt_rsrc->disable_implicit_release = 0;
+
+   if (dev_info.event_dev_cap & RTE_EVENT_DEV_CAP_QUEUE_ALL_TYPES)
+   event_queue_cfg |= RTE_EVENT_QUEUE_CFG_ALL_TYPES;
+
+   /* One queue for each ethdev port + one Tx adapter Single link queue. */
+   event_d_conf.nb_event_queues = ethdev_count + 1;
+   if (dev_info.max_event_queues < event_d_conf.nb_event_queues)
+   event_d_conf.nb_event_queues = dev_info.max_event_queues;
+
+   if (dev_info.max_num_events < event_d_conf.nb_events_limit)
+   event_d_conf.nb_events_limit = dev_info.max_num_events;
+
+   if (dev_info.max_event_queue_flows < event_d_conf.nb_event_queue_flows)
+   event_d_conf.nb_event_queue_flows =
+   dev_info.max_event_queue_flows;
+
+   if (dev_info.max_event_port_dequeue_depth <
+   event_d_conf.nb_event_port_dequeue_depth)
+   event_d_conf.nb_event_port_dequeue_depth =
+   dev_info.max_event_port_dequeue_depth;
+
+ 

[dpdk-dev] [PATCH v6 08/11] examples/l3fwd: add event lpm main loop

[pbhagavatula]

From: Pavan Nikhilesh 

Add lpm main loop for handling events based on capabilities of the
event device.

Signed-off-by: Pavan Nikhilesh 
---
 examples/l3fwd/l3fwd.h   |   9 ++
 examples/l3fwd/l3fwd_event.c |   9 ++
 examples/l3fwd/l3fwd_event.h |   5 +
 examples/l3fwd/l3fwd_lpm.c   | 203 +++
 examples/l3fwd/main.c|   1 +
 5 files changed, 227 insertions(+)

diff --git a/examples/l3fwd/l3fwd.h b/examples/l3fwd/l3fwd.h
index ed5ba5f3b..b3cdcd496 100644
--- a/examples/l3fwd/l3fwd.h
+++ b/examples/l3fwd/l3fwd.h
@@ -207,6 +207,15 @@ em_main_loop(__attribute__((unused)) void *dummy);
 int
 lpm_main_loop(__attribute__((unused)) void *dummy);
 
+int
+lpm_event_main_loop_tx_d(__attribute__((unused)) void *dummy);
+int
+lpm_event_main_loop_tx_d_burst(__attribute__((unused)) void *dummy);
+int
+lpm_event_main_loop_tx_q(__attribute__((unused)) void *dummy);
+int
+lpm_event_main_loop_tx_q_burst(__attribute__((unused)) void *dummy);
+
 /* Return ipv4/ipv6 fwd lookup struct for LPM or EM. */
 void *
 em_get_ipv4_l3fwd_lookup_struct(const int socketid);
diff --git a/examples/l3fwd/l3fwd_event.c b/examples/l3fwd/l3fwd_event.c
index 1ed42c3ab..ebddd8f97 100644
--- a/examples/l3fwd/l3fwd_event.c
+++ b/examples/l3fwd/l3fwd_event.c
@@ -211,6 +211,12 @@ void
 l3fwd_event_resource_setup(struct rte_eth_conf *port_conf)
 {
struct l3fwd_event_resources *evt_rsrc = l3fwd_get_eventdev_rsrc();
+   const event_loop_cb lpm_event_loop[2][2] = {
+   [0][0] = lpm_event_main_loop_tx_d,
+   [0][1] = lpm_event_main_loop_tx_d_burst,
+   [1][0] = lpm_event_main_loop_tx_q,
+   [1][1] = lpm_event_main_loop_tx_q_burst,
+   };
uint32_t event_queue_cfg;
int ret;
 
@@ -242,4 +248,7 @@ l3fwd_event_resource_setup(struct rte_eth_conf *port_conf)
ret = rte_event_dev_start(evt_rsrc->event_d_id);
if (ret < 0)
rte_exit(EXIT_FAILURE, "Error in starting eventdev");
+
+   evt_rsrc->ops.lpm_event_loop = lpm_event_loop[evt_rsrc->tx_mode_q]
+  [evt_rsrc->has_burst];
 }
diff --git a/examples/l3fwd/l3fwd_event.h b/examples/l3fwd/l3fwd_event.h
index b1ae970d4..0e4616417 100644
--- a/examples/l3fwd/l3fwd_event.h
+++ b/examples/l3fwd/l3fwd_event.h
@@ -14,6 +14,11 @@
 
 #include "l3fwd.h"
 
+#define L3FWD_EVENT_SINGLE 0x1
+#define L3FWD_EVENT_BURST  0x2
+#define L3FWD_EVENT_TX_DIRECT  0x4
+#define L3FWD_EVENT_TX_ENQ 0x8
+
 typedef uint32_t (*event_device_setup_cb)(void);
 typedef void (*event_queue_setup_cb)(uint32_t event_queue_cfg);
 typedef void (*event_port_setup_cb)(void);
diff --git a/examples/l3fwd/l3fwd_lpm.c b/examples/l3fwd/l3fwd_lpm.c
index 349de2703..058a60b1d 100644
--- a/examples/l3fwd/l3fwd_lpm.c
+++ b/examples/l3fwd/l3fwd_lpm.c
@@ -28,6 +28,7 @@
 #include 
 
 #include "l3fwd.h"
+#include "l3fwd_event.h"
 
 struct ipv4_l3fwd_lpm_route {
uint32_t ip;
@@ -254,6 +255,208 @@ lpm_main_loop(__attribute__((unused)) void *dummy)
return 0;
 }
 
+static __rte_always_inline uint16_t
+lpm_process_event_pkt(const struct lcore_conf *lconf, struct rte_mbuf *mbuf)
+{
+   mbuf->port = lpm_get_dst_port(lconf, mbuf, mbuf->port);
+
+#if defined RTE_ARCH_X86 || defined RTE_MACHINE_CPUFLAG_NEON \
+   || defined RTE_ARCH_PPC_64
+   process_packet(mbuf, &mbuf->port);
+#else
+
+   struct rte_ether_hdr *eth_hdr = rte_pktmbuf_mtod(mbuf,
+   struct rte_ether_hdr *);
+#ifdef DO_RFC_1812_CHECKS
+   struct rte_ipv4_hdr *ipv4_hdr;
+   if (RTE_ETH_IS_IPV4_HDR(mbuf->packet_type)) {
+   /* Handle IPv4 headers.*/
+   ipv4_hdr = rte_pktmbuf_mtod_offset(mbuf,
+   struct rte_ipv4_hdr *,
+   sizeof(struct rte_ether_hdr));
+
+   if (is_valid_ipv4_pkt(ipv4_hdr, mbuf->pkt_len)
+   < 0) {
+   mbuf->port = BAD_PORT;
+   continue;
+   }
+   /* Update time to live and header checksum */
+   --(ipv4_hdr->time_to_live);
+   ++(ipv4_hdr->hdr_checksum);
+   }
+#endif
+   /* dst addr */
+   *(uint64_t *)ð_hdr->d_addr = dest_eth_addr[mbuf->port];
+
+   /* src addr */
+   rte_ether_addr_copy(&ports_eth_addr[mbuf->port],
+   ð_hdr->s_addr);
+#endif
+   return mbuf->port;
+}
+
+static __rte_always_inline void
+lpm_event_loop_single(struct l3fwd_event_resources *evt_rsrc,
+   const uint8_t flags)
+{
+   const int event_p_id = l3fwd_get_free_event_port(evt_rsrc);
+   const uint8_t tx_q_id = evt_rsrc->evq.event_q_id[
+   evt_rsrc->evq.nb_queues - 1];
+   const uint8_t event_d_id = evt_rsrc->event_d_id;
+   struct lcore_conf *lconf;
+   unsigned int lcore_id;
+   struct rte_event ev;
+
+   if (event_p_id < 0)
+   return;
+

[dpdk-dev] [PATCH v6 05/11] examples/l3fwd: add event port and queue setup

[pbhagavatula]

From: Sunil Kumar Kori 

Add event device queue and port setup based on event eth Tx adapter
capabilities.

Signed-off-by: Sunil Kumar Kori 
---
 examples/l3fwd/l3fwd_event.c   |  28 +-
 examples/l3fwd/l3fwd_event.h   |   1 +
 examples/l3fwd/l3fwd_event_generic.c   | 103 +
 examples/l3fwd/l3fwd_event_internal_port.c |  98 
 4 files changed, 229 insertions(+), 1 deletion(-)

diff --git a/examples/l3fwd/l3fwd_event.c b/examples/l3fwd/l3fwd_event.c
index f9491ecc6..b58f9b79a 100644
--- a/examples/l3fwd/l3fwd_event.c
+++ b/examples/l3fwd/l3fwd_event.c
@@ -188,10 +188,30 @@ l3fwd_event_capability_setup(void)
l3fwd_event_set_internal_port_ops(&evt_rsrc->ops);
 }
 
+int
+l3fwd_get_free_event_port(struct l3fwd_event_resources *evt_rsrc)
+{
+   static int index;
+   int port_id;
+
+   rte_spinlock_lock(&evt_rsrc->evp.lock);
+   if (index >= evt_rsrc->evp.nb_ports) {
+   printf("No free event port is available\n");
+   return -1;
+   }
+
+   port_id = evt_rsrc->evp.event_p_id[index];
+   index++;
+   rte_spinlock_unlock(&evt_rsrc->evp.lock);
+
+   return port_id;
+}
+
 void
 l3fwd_event_resource_setup(struct rte_eth_conf *port_conf)
 {
struct l3fwd_event_resources *evt_rsrc = l3fwd_get_eventdev_rsrc();
+   uint32_t event_queue_cfg;
 
if (!evt_rsrc->enabled)
return;
@@ -206,5 +226,11 @@ l3fwd_event_resource_setup(struct rte_eth_conf *port_conf)
l3fwd_eth_dev_port_setup(port_conf);
 
/* Event device configuration */
-   evt_rsrc->ops.event_device_setup();
+   event_queue_cfg = evt_rsrc->ops.event_device_setup();
+
+   /* Event queue configuration */
+   evt_rsrc->ops.event_queue_setup(event_queue_cfg);
+
+   /* Event port configuration */
+   evt_rsrc->ops.event_port_setup();
 }
diff --git a/examples/l3fwd/l3fwd_event.h b/examples/l3fwd/l3fwd_event.h
index 53feea069..4bceca920 100644
--- a/examples/l3fwd/l3fwd_event.h
+++ b/examples/l3fwd/l3fwd_event.h
@@ -73,6 +73,7 @@ struct l3fwd_event_resources {
 
 struct l3fwd_event_resources *l3fwd_get_eventdev_rsrc(void);
 void l3fwd_event_resource_setup(struct rte_eth_conf *port_conf);
+int l3fwd_get_free_event_port(struct l3fwd_event_resources *eventdev_rsrc);
 void l3fwd_event_set_generic_ops(struct l3fwd_event_setup_ops *ops);
 void l3fwd_event_set_internal_port_ops(struct l3fwd_event_setup_ops *ops);
 
diff --git a/examples/l3fwd/l3fwd_event_generic.c 
b/examples/l3fwd/l3fwd_event_generic.c
index ce06fa0e4..4d0cbec05 100644
--- a/examples/l3fwd/l3fwd_event_generic.c
+++ b/examples/l3fwd/l3fwd_event_generic.c
@@ -81,8 +81,111 @@ l3fwd_event_device_setup_generic(void)
return event_queue_cfg;
 }
 
+static void
+l3fwd_event_port_setup_generic(void)
+{
+   struct l3fwd_event_resources *evt_rsrc = l3fwd_get_eventdev_rsrc();
+   uint8_t event_d_id = evt_rsrc->event_d_id;
+   struct rte_event_port_conf event_p_conf = {
+   .dequeue_depth = 32,
+   .enqueue_depth = 32,
+   .new_event_threshold = 4096
+   };
+   struct rte_event_port_conf def_p_conf;
+   uint8_t event_p_id;
+   int32_t ret;
+
+   evt_rsrc->evp.event_p_id = (uint8_t *)malloc(sizeof(uint8_t) *
+   evt_rsrc->evp.nb_ports);
+   if (!evt_rsrc->evp.event_p_id)
+   rte_panic("No space is available\n");
+
+   memset(&def_p_conf, 0, sizeof(struct rte_event_port_conf));
+   rte_event_port_default_conf_get(event_d_id, 0, &def_p_conf);
+
+   if (def_p_conf.new_event_threshold < event_p_conf.new_event_threshold)
+   event_p_conf.new_event_threshold =
+   def_p_conf.new_event_threshold;
+
+   if (def_p_conf.dequeue_depth < event_p_conf.dequeue_depth)
+   event_p_conf.dequeue_depth = def_p_conf.dequeue_depth;
+
+   if (def_p_conf.enqueue_depth < event_p_conf.enqueue_depth)
+   event_p_conf.enqueue_depth = def_p_conf.enqueue_depth;
+
+   event_p_conf.disable_implicit_release =
+   evt_rsrc->disable_implicit_release;
+   evt_rsrc->deq_depth = def_p_conf.dequeue_depth;
+
+   for (event_p_id = 0; event_p_id < evt_rsrc->evp.nb_ports;
+   event_p_id++) {
+   ret = rte_event_port_setup(event_d_id, event_p_id,
+  &event_p_conf);
+   if (ret < 0)
+   rte_panic("Error in configuring event port %d\n",
+ event_p_id);
+
+   ret = rte_event_port_link(event_d_id, event_p_id,
+ evt_rsrc->evq.event_q_id,
+ NULL,
+ evt_rsrc->evq.nb_queues - 1);
+   if (ret != (evt_rsrc->evq.nb_queues - 1))

[dpdk-dev] [PATCH v6 07/11] examples/l3fwd: add service core setup based on caps

[pbhagavatula]

From: Pavan Nikhilesh 

Add service core setup when eventdev and Rx/Tx adapter don't have
internal port capability.

Signed-off-by: Pavan Nikhilesh 
---
 examples/l3fwd/l3fwd_event.c |   6 ++
 examples/l3fwd/main.c| 188 +++
 2 files changed, 150 insertions(+), 44 deletions(-)

diff --git a/examples/l3fwd/l3fwd_event.c b/examples/l3fwd/l3fwd_event.c
index 0a75e39ee..1ed42c3ab 100644
--- a/examples/l3fwd/l3fwd_event.c
+++ b/examples/l3fwd/l3fwd_event.c
@@ -212,6 +212,7 @@ l3fwd_event_resource_setup(struct rte_eth_conf *port_conf)
 {
struct l3fwd_event_resources *evt_rsrc = l3fwd_get_eventdev_rsrc();
uint32_t event_queue_cfg;
+   int ret;
 
if (!evt_rsrc->enabled)
return;
@@ -236,4 +237,9 @@ l3fwd_event_resource_setup(struct rte_eth_conf *port_conf)
 
/* Rx/Tx adapters configuration */
evt_rsrc->ops.adapter_setup();
+
+   /* Start event device */
+   ret = rte_event_dev_start(evt_rsrc->event_d_id);
+   if (ret < 0)
+   rte_exit(EXIT_FAILURE, "Error in starting eventdev");
 }
diff --git a/examples/l3fwd/main.c b/examples/l3fwd/main.c
index 04526343f..4bb00a48b 100644
--- a/examples/l3fwd/main.c
+++ b/examples/l3fwd/main.c
@@ -899,49 +899,18 @@ prepare_ptype_parser(uint16_t portid, uint16_t queueid)
return 0;
 }
 
-int
-main(int argc, char **argv)
+static void
+l3fwd_poll_resource_setup(void)
 {
-   struct l3fwd_event_resources *evt_rsrc;
-   struct lcore_conf *qconf;
+   uint8_t nb_rx_queue, queue, socketid;
struct rte_eth_dev_info dev_info;
+   uint32_t n_tx_queue, nb_lcores;
struct rte_eth_txconf *txconf;
-   int ret;
-   unsigned nb_ports;
+   struct lcore_conf *qconf;
uint16_t queueid, portid;
-   unsigned lcore_id;
-   uint32_t n_tx_queue, nb_lcores;
-   uint8_t nb_rx_queue, queue, socketid;
-
-   /* init EAL */
-   ret = rte_eal_init(argc, argv);
-   if (ret < 0)
-   rte_exit(EXIT_FAILURE, "Invalid EAL parameters\n");
-   argc -= ret;
-   argv += ret;
-
-   force_quit = false;
-   signal(SIGINT, signal_handler);
-   signal(SIGTERM, signal_handler);
-
-   /* pre-init dst MACs for all ports to 02:00:00:00:00:xx */
-   for (portid = 0; portid < RTE_MAX_ETHPORTS; portid++) {
-   dest_eth_addr[portid] =
-   RTE_ETHER_LOCAL_ADMIN_ADDR + ((uint64_t)portid << 40);
-   *(uint64_t *)(val_eth + portid) = dest_eth_addr[portid];
-   }
-
-   evt_rsrc = l3fwd_get_eventdev_rsrc();
-   /* parse application arguments (after the EAL ones) */
-   ret = parse_args(argc, argv);
-   if (ret < 0)
-   rte_exit(EXIT_FAILURE, "Invalid L3FWD parameters\n");
-
-   evt_rsrc->per_port_pool = per_port_pool;
-   evt_rsrc->pkt_pool = pktmbuf_pool;
-   evt_rsrc->port_mask = enabled_port_mask;
-   /* Configure eventdev parameters if user has requested */
-   l3fwd_event_resource_setup(&port_conf);
+   unsigned int nb_ports;
+   unsigned int lcore_id;
+   int ret;
 
if (check_lcore_params() < 0)
rte_exit(EXIT_FAILURE, "check_lcore_params failed\n");
@@ -957,9 +926,6 @@ main(int argc, char **argv)
 
nb_lcores = rte_lcore_count();
 
-   /* Setup function pointers for lookup method. */
-   setup_l3fwd_lookup_tables();
-
/* initialize all ports */
RTE_ETH_FOREACH_DEV(portid) {
struct rte_eth_conf local_port_conf = port_conf;
@@ -1127,7 +1093,142 @@ main(int argc, char **argv)
}
}
 
-   printf("\n");
+
+}
+
+static inline int
+l3fwd_service_enable(uint32_t service_id)
+{
+   uint8_t min_service_count = UINT8_MAX;
+   uint32_t slcore_array[RTE_MAX_LCORE];
+   unsigned int slcore = 0;
+   uint8_t service_count;
+   int32_t slcore_count;
+
+   if (!rte_service_lcore_count())
+   return -ENOENT;
+
+   slcore_count = rte_service_lcore_list(slcore_array, RTE_MAX_LCORE);
+   if (slcore_count < 0)
+   return -ENOENT;
+   /* Get the core which has least number of services running. */
+   while (slcore_count--) {
+   /* Reset default mapping */
+   rte_service_map_lcore_set(service_id,
+   slcore_array[slcore_count], 0);
+   service_count = rte_service_lcore_count_services(
+   slcore_array[slcore_count]);
+   if (service_count < min_service_count) {
+   slcore = slcore_array[slcore_count];
+   min_service_count = service_count;
+   }
+   }
+   if (rte_service_map_lcore_set(service_id, slcore, 1))
+   return -ENOENT;
+   rte_service_lcore_start(slcore);
+
+   return 0;
+}
+
+static void
+l3fwd_event_service_setup(void)
+{
+   struct l3fwd_event_resources *evt_r

[dpdk-dev] [PATCH v6 06/11] examples/l3fwd: add event eth Rx/Tx adapter setup

[pbhagavatula]

From: Pavan Nikhilesh 

Add event eth Rx/Tx adapter setup for both generic and internal port
event device pipelines.

Signed-off-by: Sunil Kumar Kori 
Signed-off-by: Pavan Nikhilesh 
---
 examples/l3fwd/l3fwd_event.c   |   3 +
 examples/l3fwd/l3fwd_event.h   |   1 +
 examples/l3fwd/l3fwd_event_generic.c   | 112 +
 examples/l3fwd/l3fwd_event_internal_port.c |  93 +
 4 files changed, 209 insertions(+)

diff --git a/examples/l3fwd/l3fwd_event.c b/examples/l3fwd/l3fwd_event.c
index b58f9b79a..0a75e39ee 100644
--- a/examples/l3fwd/l3fwd_event.c
+++ b/examples/l3fwd/l3fwd_event.c
@@ -233,4 +233,7 @@ l3fwd_event_resource_setup(struct rte_eth_conf *port_conf)
 
/* Event port configuration */
evt_rsrc->ops.event_port_setup();
+
+   /* Rx/Tx adapters configuration */
+   evt_rsrc->ops.adapter_setup();
 }
diff --git a/examples/l3fwd/l3fwd_event.h b/examples/l3fwd/l3fwd_event.h
index 4bceca920..b1ae970d4 100644
--- a/examples/l3fwd/l3fwd_event.h
+++ b/examples/l3fwd/l3fwd_event.h
@@ -7,6 +7,7 @@
 
 #include 
 #include 
+#include 
 #include 
 #include 
 #include 
diff --git a/examples/l3fwd/l3fwd_event_generic.c 
b/examples/l3fwd/l3fwd_event_generic.c
index 4d0cbec05..c69c611dd 100644
--- a/examples/l3fwd/l3fwd_event_generic.c
+++ b/examples/l3fwd/l3fwd_event_generic.c
@@ -182,10 +182,122 @@ l3fwd_event_queue_setup_generic(uint32_t event_queue_cfg)
evt_rsrc->evq.event_q_id[event_q_id] = event_q_id;
 }
 
+static void
+l3fwd_rx_tx_adapter_setup_generic(void)
+{
+   struct l3fwd_event_resources *evt_rsrc = l3fwd_get_eventdev_rsrc();
+   struct rte_event_eth_rx_adapter_queue_conf eth_q_conf;
+   uint8_t event_d_id = evt_rsrc->event_d_id;
+   uint8_t rx_adptr_id = 0;
+   uint8_t tx_adptr_id = 0;
+   uint8_t tx_port_id = 0;
+   uint16_t port_id;
+   uint32_t service_id;
+   int32_t ret, i = 0;
+
+   memset(ð_q_conf, 0, sizeof(eth_q_conf));
+   eth_q_conf.ev.priority = RTE_EVENT_DEV_PRIORITY_NORMAL;
+
+   /* Rx adapter setup */
+   evt_rsrc->rx_adptr.nb_rx_adptr = 1;
+   evt_rsrc->rx_adptr.rx_adptr = (uint8_t *)malloc(sizeof(uint8_t) *
+   evt_rsrc->rx_adptr.nb_rx_adptr);
+   if (!evt_rsrc->rx_adptr.rx_adptr) {
+   free(evt_rsrc->evp.event_p_id);
+   free(evt_rsrc->evq.event_q_id);
+   rte_panic("Failed to allocate memory for Rx adapter\n");
+   }
+
+   ret = rte_event_eth_rx_adapter_create(rx_adptr_id, event_d_id,
+ &evt_rsrc->def_p_conf);
+   if (ret)
+   rte_panic("Failed to create rx adapter\n");
+
+   /* Configure user requested sched type */
+   eth_q_conf.ev.sched_type = evt_rsrc->sched_type;
+   RTE_ETH_FOREACH_DEV(port_id) {
+   if ((evt_rsrc->port_mask & (1 << port_id)) == 0)
+   continue;
+   eth_q_conf.ev.queue_id = evt_rsrc->evq.event_q_id[i];
+   ret = rte_event_eth_rx_adapter_queue_add(rx_adptr_id, port_id,
+-1, ð_q_conf);
+   if (ret)
+   rte_panic("Failed to add queues to Rx adapter\n");
+   if (i < evt_rsrc->evq.nb_queues)
+   i++;
+   }
+
+   ret = rte_event_eth_rx_adapter_service_id_get(rx_adptr_id, &service_id);
+   if (ret != -ESRCH && ret != 0)
+   rte_panic("Error getting the service ID for rx adptr\n");
+
+   rte_service_runstate_set(service_id, 1);
+   rte_service_set_runstate_mapped_check(service_id, 0);
+   evt_rsrc->rx_adptr.service_id = service_id;
+
+   ret = rte_event_eth_rx_adapter_start(rx_adptr_id);
+   if (ret)
+   rte_panic("Rx adapter[%d] start Failed\n", rx_adptr_id);
+
+   evt_rsrc->rx_adptr.rx_adptr[0] = rx_adptr_id;
+
+   /* Tx adapter setup */
+   evt_rsrc->tx_adptr.nb_tx_adptr = 1;
+   evt_rsrc->tx_adptr.tx_adptr = (uint8_t *)malloc(sizeof(uint8_t) *
+   evt_rsrc->tx_adptr.nb_tx_adptr);
+   if (!evt_rsrc->tx_adptr.tx_adptr) {
+   free(evt_rsrc->rx_adptr.rx_adptr);
+   free(evt_rsrc->evp.event_p_id);
+   free(evt_rsrc->evq.event_q_id);
+   rte_panic("Failed to allocate memory for Rx adapter\n");
+   }
+
+   ret = rte_event_eth_tx_adapter_create(tx_adptr_id, event_d_id,
+ &evt_rsrc->def_p_conf);
+   if (ret)
+   rte_panic("Failed to create tx adapter\n");
+
+   RTE_ETH_FOREACH_DEV(port_id) {
+   if ((evt_rsrc->port_mask & (1 << port_id)) == 0)
+   continue;
+   ret = rte_event_eth_tx_adapter_queue_add(tx_adptr_id, port_id,
+-1);
+   if (ret)
+   rte_pan

[dpdk-dev] [PATCH v6 10/11] examples/l3fwd: add graceful teardown for eventdevice

[pbhagavatula]

From: Pavan Nikhilesh 

Add graceful teardown that addresses both event mode and poll mode.

Signed-off-by: Pavan Nikhilesh 
---
 examples/l3fwd/main.c | 50 +--
 1 file changed, 34 insertions(+), 16 deletions(-)

diff --git a/examples/l3fwd/main.c b/examples/l3fwd/main.c
index c95b1655e..dda430d68 100644
--- a/examples/l3fwd/main.c
+++ b/examples/l3fwd/main.c
@@ -1092,8 +1092,6 @@ l3fwd_poll_resource_setup(void)
ret, portid);
}
}
-
-
 }
 
 static inline int
@@ -1191,7 +1189,7 @@ main(int argc, char **argv)
uint16_t queueid, portid;
unsigned int lcore_id;
uint8_t queue;
-   int ret;
+   int i, ret;
 
/* init EAL */
ret = rte_eal_init(argc, argv);
@@ -1280,21 +1278,41 @@ main(int argc, char **argv)
ret = 0;
/* launch per-lcore init on every lcore */
rte_eal_mp_remote_launch(l3fwd_lkp.main_loop, NULL, CALL_MASTER);
-   RTE_LCORE_FOREACH_SLAVE(lcore_id) {
-   if (rte_eal_wait_lcore(lcore_id) < 0) {
-   ret = -1;
-   break;
+   if (evt_rsrc->enabled) {
+   for (i = 0; i < evt_rsrc->rx_adptr.nb_rx_adptr; i++)
+   rte_event_eth_rx_adapter_stop(
+   evt_rsrc->rx_adptr.rx_adptr[i]);
+   for (i = 0; i < evt_rsrc->tx_adptr.nb_tx_adptr; i++)
+   rte_event_eth_tx_adapter_stop(
+   evt_rsrc->tx_adptr.tx_adptr[i]);
+
+   RTE_ETH_FOREACH_DEV(portid) {
+   if ((enabled_port_mask & (1 << portid)) == 0)
+   continue;
+   rte_eth_dev_stop(portid);
}
-   }
 
-   /* stop ports */
-   RTE_ETH_FOREACH_DEV(portid) {
-   if ((enabled_port_mask & (1 << portid)) == 0)
-   continue;
-   printf("Closing port %d...", portid);
-   rte_eth_dev_stop(portid);
-   rte_eth_dev_close(portid);
-   printf(" Done\n");
+   rte_eal_mp_wait_lcore();
+   RTE_ETH_FOREACH_DEV(portid) {
+   if ((enabled_port_mask & (1 << portid)) == 0)
+   continue;
+   rte_eth_dev_close(portid);
+   }
+
+   rte_event_dev_stop(evt_rsrc->event_d_id);
+   rte_event_dev_close(evt_rsrc->event_d_id);
+
+   } else {
+   rte_eal_mp_wait_lcore();
+
+   RTE_ETH_FOREACH_DEV(portid) {
+   if ((enabled_port_mask & (1 << portid)) == 0)
+   continue;
+   printf("Closing port %d...", portid);
+   rte_eth_dev_stop(portid);
+   rte_eth_dev_close(portid);
+   printf(" Done\n");
+   }
}
printf("Bye...\n");
 
-- 
2.17.1

[dpdk-dev] [PATCH v6 09/11] examples/l3fwd: add event em main loop

[pbhagavatula]

From: Pavan Nikhilesh 

Add em main loop for handling events based on capabilities of the
event device.

Signed-off-by: Pavan Nikhilesh 
---
 examples/l3fwd/l3fwd.h   |  10 ++
 examples/l3fwd/l3fwd_em.c| 177 +++
 examples/l3fwd/l3fwd_em.h| 159 +---
 examples/l3fwd/l3fwd_em_hlm.h| 131 
 examples/l3fwd/l3fwd_em_sequential.h |  26 
 examples/l3fwd/l3fwd_event.c |   9 ++
 examples/l3fwd/main.c|   5 +-
 7 files changed, 470 insertions(+), 47 deletions(-)

diff --git a/examples/l3fwd/l3fwd.h b/examples/l3fwd/l3fwd.h
index b3cdcd496..c786f9062 100644
--- a/examples/l3fwd/l3fwd.h
+++ b/examples/l3fwd/l3fwd.h
@@ -216,6 +216,16 @@ lpm_event_main_loop_tx_q(__attribute__((unused)) void 
*dummy);
 int
 lpm_event_main_loop_tx_q_burst(__attribute__((unused)) void *dummy);
 
+int
+em_event_main_loop_tx_d(__attribute__((unused)) void *dummy);
+int
+em_event_main_loop_tx_d_burst(__attribute__((unused)) void *dummy);
+int
+em_event_main_loop_tx_q(__attribute__((unused)) void *dummy);
+int
+em_event_main_loop_tx_q_burst(__attribute__((unused)) void *dummy);
+
+
 /* Return ipv4/ipv6 fwd lookup struct for LPM or EM. */
 void *
 em_get_ipv4_l3fwd_lookup_struct(const int socketid);
diff --git a/examples/l3fwd/l3fwd_em.c b/examples/l3fwd/l3fwd_em.c
index 74a7c8fa4..1a8bc9168 100644
--- a/examples/l3fwd/l3fwd_em.c
+++ b/examples/l3fwd/l3fwd_em.c
@@ -26,6 +26,7 @@
 #include 
 
 #include "l3fwd.h"
+#include "l3fwd_event.h"
 
 #if defined(RTE_ARCH_X86) || defined(RTE_MACHINE_CPUFLAG_CRC32)
 #define EM_HASH_CRC 1
@@ -699,6 +700,182 @@ em_main_loop(__attribute__((unused)) void *dummy)
return 0;
 }
 
+static __rte_always_inline void
+em_event_loop_single(struct l3fwd_event_resources *evt_rsrc,
+   const uint8_t flags)
+{
+   const int event_p_id = l3fwd_get_free_event_port(evt_rsrc);
+   const uint8_t tx_q_id = evt_rsrc->evq.event_q_id[
+   evt_rsrc->evq.nb_queues - 1];
+   const uint8_t event_d_id = evt_rsrc->event_d_id;
+   struct lcore_conf *lconf;
+   unsigned int lcore_id;
+   struct rte_event ev;
+
+   if (event_p_id < 0)
+   return;
+
+   lcore_id = rte_lcore_id();
+   lconf = &lcore_conf[lcore_id];
+
+   RTE_LOG(INFO, L3FWD, "entering %s on lcore %u\n", __func__, lcore_id);
+   while (!force_quit) {
+   if (!rte_event_dequeue_burst(event_d_id, event_p_id, &ev, 1, 0))
+   continue;
+
+   struct rte_mbuf *mbuf = ev.mbuf;
+
+#if defined RTE_ARCH_X86 || defined RTE_MACHINE_CPUFLAG_NEON
+   mbuf->port = em_get_dst_port(lconf, mbuf, mbuf->port);
+   process_packet(mbuf, &mbuf->port);
+#else
+   l3fwd_em_simple_process(mbuf, lconf);
+#endif
+   if (mbuf->port == BAD_PORT) {
+   rte_pktmbuf_free(mbuf);
+   continue;
+   }
+
+   if (flags & L3FWD_EVENT_TX_ENQ) {
+   ev.queue_id = tx_q_id;
+   ev.op = RTE_EVENT_OP_FORWARD;
+   while (rte_event_enqueue_burst(event_d_id, event_p_id,
+   &ev, 1) && !force_quit)
+   ;
+   }
+
+   if (flags & L3FWD_EVENT_TX_DIRECT) {
+   rte_event_eth_tx_adapter_txq_set(mbuf, 0);
+   while (!rte_event_eth_tx_adapter_enqueue(event_d_id,
+   event_p_id, &ev, 1, 0) &&
+   !force_quit)
+   ;
+   }
+   }
+}
+
+static __rte_always_inline void
+em_event_loop_burst(struct l3fwd_event_resources *evt_rsrc,
+   const uint8_t flags)
+{
+   const int event_p_id = l3fwd_get_free_event_port(evt_rsrc);
+   const uint8_t tx_q_id = evt_rsrc->evq.event_q_id[
+   evt_rsrc->evq.nb_queues - 1];
+   const uint8_t event_d_id = evt_rsrc->event_d_id;
+   const uint16_t deq_len = evt_rsrc->deq_depth;
+   struct rte_event events[MAX_PKT_BURST];
+   struct lcore_conf *lconf;
+   unsigned int lcore_id;
+   int i, nb_enq, nb_deq;
+
+   if (event_p_id < 0)
+   return;
+
+   lcore_id = rte_lcore_id();
+
+   lconf = &lcore_conf[lcore_id];
+
+   RTE_LOG(INFO, L3FWD, "entering %s on lcore %u\n", __func__, lcore_id);
+
+   while (!force_quit) {
+   /* Read events from RX queues */
+   nb_deq = rte_event_dequeue_burst(event_d_id, event_p_id,
+   events, deq_len, 0);
+   if (nb_deq == 0) {
+   rte_pause();
+   continue;
+   }
+
+#if defined RTE_ARCH_X86 || defined RTE_MACHINE_CPUFLAG_NEON
+   l3fwd_em_process_events(nb_deq, (struct rte_event **)&event

[dpdk-dev] [PATCH v6 11/11] doc: update l3fwd user guide to support eventdev

[pbhagavatula]

From: Pavan Nikhilesh 

Update l3fwd user guide to include event device related information.
Update release notes and MAINTAINERS file.

Signed-off-by: Pavan Nikhilesh 
---
 MAINTAINERS |  1 +
 doc/guides/rel_notes/release_20_02.rst  |  5 ++
 doc/guides/sample_app_ug/l3_forward.rst | 79 +++--
 3 files changed, 79 insertions(+), 6 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 94bccae6d..8047aaf2a 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -430,6 +430,7 @@ T: git://dpdk.org/next/dpdk-next-eventdev
 F: lib/librte_eventdev/
 F: drivers/event/skeleton/
 F: app/test/test_eventdev.c
+F: examples/l3fwd/l3fwd_event*
 
 Eventdev Ethdev Rx Adapter API
 M: Nikhil Rao 
diff --git a/doc/guides/rel_notes/release_20_02.rst 
b/doc/guides/rel_notes/release_20_02.rst
index 50e2c1484..9cb69e53b 100644
--- a/doc/guides/rel_notes/release_20_02.rst
+++ b/doc/guides/rel_notes/release_20_02.rst
@@ -143,6 +143,11 @@ New Features
   Added a new OCTEON TX2 rawdev PMD for End Point mode of operation.
   See the :doc:`../rawdevs/octeontx2_ep` for more details on this new PMD.
 
+* **Add event device support for l3fwd sample application.**
+
+  Add event device support for ``l3fwd`` sample application. It demonstrates
+  usage of poll and event mode IO mechanism under a single application.
+
 
 Removed Items
 -
diff --git a/doc/guides/sample_app_ug/l3_forward.rst 
b/doc/guides/sample_app_ug/l3_forward.rst
index 4cb4b18da..153e46893 100644
--- a/doc/guides/sample_app_ug/l3_forward.rst
+++ b/doc/guides/sample_app_ug/l3_forward.rst
@@ -4,16 +4,23 @@
 L3 Forwarding Sample Application
 
 
-The L3 Forwarding application is a simple example of packet processing using 
the DPDK.
+The L3 Forwarding application is a simple example of packet processing using
+DPDK to demonstrate usage of poll and event mode packet I/O mechanism.
 The application performs L3 forwarding.
 
 Overview
 
 
-The application demonstrates the use of the hash and LPM libraries in the DPDK 
to implement packet forwarding.
-The initialization and run-time paths are very similar to those of the 
:doc:`l2_forward_real_virtual`.
-The main difference from the L2 Forwarding sample application is that the 
forwarding decision
-is made based on information read from the input packet.
+The application demonstrates the use of the hash and LPM libraries in the DPDK
+to implement packet forwarding using poll or event mode PMDs for packet I/O.
+The initialization and run-time paths are very similar to those of the
+:doc:`l2_forward_real_virtual` and :doc:`l2_forward_event`.
+The main difference from the L2 Forwarding sample application is that 
optionally
+packet can be Rx/Tx from/to eventdev instead of port directly and forwarding
+decision is made based on information read from the input packet.
+
+Eventdev can optionally use S/W or H/W (if supported by platform) scheduler
+implementation for packet I/O based on run time parameters.
 
 The lookup method is either hash-based or LPM-based and is selected at run 
time. When the selected lookup method is hash-based,
 a hash object is used to emulate the flow classification stage.
@@ -56,6 +63,9 @@ The application has a number of command line options::
  [--ipv6]
  [--parse-ptype]
  [--per-port-pool]
+ [--mode]
+ [--eventq-sched]
+ [--event-eth-rxqs]
 
 Where,
 
@@ -86,6 +96,13 @@ Where,
 
 * ``--per-port-pool:`` Optional, set to use independent buffer pools per port. 
Without this option, single buffer pool is used for all ports.
 
+* ``--mode:`` Optional, Packet transfer mode for I/O, poll or eventdev.
+
+* ``--eventq-sched:`` Optional, Event queue synchronization method, Ordered, 
Atomic or Parallel. Only valid if --mode=eventdev.
+
+* ``--event-eth-rxqs:`` Optional, Number of ethernet RX queues per device. 
Only valid if --mode=eventdev.
+
+
 For example, consider a dual processor socket platform with 8 physical cores, 
where cores 0-7 and 16-23 appear on socket 0,
 while cores 8-15 and 24-31 appear on socket 1.
 
@@ -116,6 +133,51 @@ In this command:
 |  |   |   | |
 +--+---+---+-+
 
+To use eventdev mode with sync method **ordered** on above mentioned 
environment,
+Following is the sample command:
+
+.. code-block:: console
+
+./build/l3fwd -l 0-3 -n 4 -w  -- -p 0x3 
--eventq-sched=ordered
+
+or
+
+.. code-block:: console
+
+./build/l3fwd -l 0-3 -n 4 -w  -- -p 0x03 --mode=eventdev 
--eventq-sched=ordered
+
+In this command:
+
+*   -w option whitelist the event device supported by platform. Way to pass 
this device may vary based on platform.
+
+*   The --mode option defines PMD to be used for packet I/O.
+
+*   The --eventq-sche

Re: [dpdk-dev] [PATCH] event/dsw: use custom element size ring for control

[Jerin Jacob]

On Wed, Jan 22, 2020 at 10:36 AM Honnappa Nagarahalli
 wrote:
>
> 
>
> > Subject: [PATCH] event/dsw: use custom element size ring for control
> >
> > Replace DSW's use of regular DPDK rings (and code for packing/unpacking
> > control messages into void pointers) with custom size rings.
> >
> > In addition to cleaner code, this change allows DSW to support up to the
> > eventdev API's maximum of 255 ports by tweaking DSW_MAX_PORTS.
> >
> > Signed-off-by: Mattias Rönnblom 

> > +
> > +# Depends on rte_ring_elem_*()
> > +allow_experimental_apis = true
> The rte_ring_elem_xxx API changes look good.
>
> Reviewed-by: Honnappa Nagarahalli 

Applied to dpdk-next-eventdev/master. Thanks.


>
> > --
> > 2.17.1
>

Re: [dpdk-dev] [PATCH] test/compress: add cycle-count mode to the perf tool

[Akhil Goyal]

> >
> > This commit adds cycle-count mode to the compression perf tool.
> > The new mode enhances the compression performance tool to allow
> > cycle-count measurement of both hardware and softwate PMDs.
> >
> > Signed-off-by: Artur Trybula 
> Acked-by: Fiona Trahe 

Applied to dpdk-next-crypto

Thanks.

Re: [dpdk-dev] [PATCH v1] crypto/ccp: add support for new CCP device ID

[Akhil Goyal]

> 
> 
> Looks good to me.
> 
> Acked-by: Ravi Kumar 
> 
Applied to dpdk-next-crypto

Thanks

Re: [dpdk-dev] [PATCH] examples/l2fwd-crypto:skip dev configure for masked devices

[Akhil Goyal]

> 
> The devices which are masked by cryptodev mask should not be initialized
> and skipped while traversing the device list.
> 
> Fixes: 61a7018ba0 ("examples/l2fwd-crypto:skip dev configure for masked
> devices")
> Cc: sta...@dpdk.org
> 
> Signed-off-by: Apeksha Gupta 
> ---
Acked-by: Akhil Goyal 

Applied to dpdk-next-crypto

Thanks.

Re: [dpdk-dev] [PATCH 1/2] crypto/dpaa_sec: fix performance issue

[Akhil Goyal]

> 
> DPAA sec driver is using virtual to physical address
> translation in its data path and driver is using
> dpaax_iova_table_update() API in every address translation
> which is very costly.
> This patch moves dpaax_iova_table_update() calling to rte_dpaa_mem_ptov(),
> only if it fails to found translation from DPAAX table.
> 
> Fixes: 033968bdc28 ("drivers/crypto/dpaa_sec: update DPAA iova table in
> dpaa_mem_vtop")
> Signed-off-by: Gagandeep Singh 
> ---
Series
Acked-by: Akhil Goyal 

>From next time, please update the version of the updated patch in title
And specify change log after '---'

Series applied to dpdk-next-crypto

Thanks.

Re: [dpdk-dev] [PATCH v2 1/1] test/crypto: remove repeated test and enable on QAT

[Akhil Goyal]

> >
> > Remove one AES GCM scatter-gather unit test repetition
> > and enable this test on QAT (after a fix included in the commit
> > b26ef1a11f21).
> >
> > Fixes: 157d0b11d204 ("test/crypto: add capability checks")
> > Fixes: b26ef1a11f21 ("test/crypto: fix missing operation status check")
> >
> > Signed-off-by: Adam Dybkowski 
> Acked-by: Fiona Trahe 

Squashed to original commit as they were not applied to Master

Thanks.

Re: [dpdk-dev] [PATCH v1] crypto/ccp: aes-gcm driver bug fix

[Akhil Goyal]

> 
> Looks good to me.
> 
> Acked-by: Ravi Kumar 
> 
> Regards,
> Ravi
> >
> >
> >
> >fixes ccp crypto driver to make aes-gcm output match with openssl/NIST
> output
> >
> >Signed-off-by: Selwin Sebastian 
> >---
Please mention commit id for the original patch for which this fix is submitted.
"Fixes: <12 digits of commit id> ("patch title")"
Title should be in format "crypto/ccp: fix ."
Also explain the issue briefly in description and cc sta...@dpdk.org

[dpdk-dev] [PATCH 5/8] net/bnxt: remove redundant macro

[Somnath Kotur]

From: Kalesh AP 

Use "dev->data->dev_started" state, instead of local BNXT_FLAG_INIT_DONE
to check whether device has been initialised or not.

Fixes: ed2ced6fe927 ("net/bnxt: check initialization before accessing stats")

Reviewed-by: Santoshkumar Karanappa Rastapur 
Reviewed-by: Ajit Kumar Khaparde 
Signed-off-by: Kalesh AP 
---
 drivers/net/bnxt/bnxt.h| 1 -
 drivers/net/bnxt/bnxt_ethdev.c | 2 --
 drivers/net/bnxt/bnxt_stats.c  | 4 ++--
 3 files changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index bca9ad4..95f1f1a 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -520,7 +520,6 @@ struct bnxt {
 #define BNXT_FLAG_FW_HEALTH_CHECK_SCHEDULEDBIT(18)
 #define BNXT_FLAG_EXT_STATS_SUPPORTED  BIT(19)
 #define BNXT_FLAG_NEW_RM   BIT(20)
-#define BNXT_FLAG_INIT_DONEBIT(21)
 #define BNXT_FLAG_FW_CAP_ONE_STEP_TX_TSBIT(22)
 #define BNXT_FLAG_ADV_FLOW_MGMTBIT(23)
 #define BNXT_FLAG_RX_VECTOR_PKT_MODE   BIT(24)
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index 267a361..b04685c 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -887,7 +887,6 @@ static int bnxt_dev_start_op(struct rte_eth_dev *eth_dev)
eth_dev->rx_pkt_burst = bnxt_receive_function(eth_dev);
eth_dev->tx_pkt_burst = bnxt_transmit_function(eth_dev);
 
-   bp->flags |= BNXT_FLAG_INIT_DONE;
eth_dev->data->dev_started = 1;
pthread_mutex_lock(&bp->def_cp_lock);
bnxt_schedule_fw_health_check(bp);
@@ -947,7 +946,6 @@ static void bnxt_dev_stop_op(struct rte_eth_dev *eth_dev)
 
bnxt_cancel_fw_health_check(bp);
 
-   bp->flags &= ~BNXT_FLAG_INIT_DONE;
bnxt_dev_set_link_down_op(eth_dev);
 
/* Wait for link to be reset and the async notification to process.
diff --git a/drivers/net/bnxt/bnxt_stats.c b/drivers/net/bnxt/bnxt_stats.c
index 32f112b..4668629 100644
--- a/drivers/net/bnxt/bnxt_stats.c
+++ b/drivers/net/bnxt/bnxt_stats.c
@@ -389,7 +389,7 @@ int bnxt_stats_get_op(struct rte_eth_dev *eth_dev,
if (rc)
return rc;
 
-   if (!(bp->flags & BNXT_FLAG_INIT_DONE))
+   if (!eth_dev->data->dev_started)
return -EIO;
 
num_q_stats = RTE_MIN(bp->rx_cp_nr_rings,
@@ -434,7 +434,7 @@ int bnxt_stats_reset_op(struct rte_eth_dev *eth_dev)
if (ret)
return ret;
 
-   if (!(bp->flags & BNXT_FLAG_INIT_DONE)) {
+   if (!eth_dev->data->dev_started) {
PMD_DRV_LOG(ERR, "Device Initialization not complete!\n");
return -EINVAL;
}
-- 
1.8.3.1

[dpdk-dev] [PATCH 4/8] net/bnxt: remove redundant if statement

[Somnath Kotur]

From: Kalesh AP 

Since "eth_dev->data->dev_started" has been assigned to 0 at the
beginning of bnxt_dev_stop_op() routine, the code inside the if()
condition is redundant. Remove it.

Anyways "eth_dev->data->dev_link.link_status" will be set to 0 in
bnxt_dev_set_link_down_op() later in the routine.

Fixes: 316e412299fd ("net/bnxt: fix crash when closing")

Signed-off-by: Kalesh AP 
Reviewed-by: Ajit Kumar Khaparde 
Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt_ethdev.c | 4 
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index fc3f1a8..267a361 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -948,10 +948,6 @@ static void bnxt_dev_stop_op(struct rte_eth_dev *eth_dev)
bnxt_cancel_fw_health_check(bp);
 
bp->flags &= ~BNXT_FLAG_INIT_DONE;
-   if (bp->eth_dev->data->dev_started) {
-   /* TBD: STOP HW queues DMA */
-   eth_dev->data->dev_link.link_status = 0;
-   }
bnxt_dev_set_link_down_op(eth_dev);
 
/* Wait for link to be reset and the async notification to process.
-- 
1.8.3.1

[dpdk-dev] [PATCH 1/8] net/bnxt: fix bnxt_alloc_filter() to use a common routine

[Somnath Kotur]

Invoke bnxt_get_unused_filter() inside bnxt_alloc_filter() so that
all filters are allocated from one common routine.

Fixes: f92735db1e4c ("net/bnxt: add L2 filter alloc/init/free")

Reviewed-by: Kalesh Anakkur Purayil 
Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt_filter.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/bnxt/bnxt_filter.c b/drivers/net/bnxt/bnxt_filter.c
index b31f104..e218433 100644
--- a/drivers/net/bnxt/bnxt_filter.c
+++ b/drivers/net/bnxt/bnxt_filter.c
@@ -26,13 +26,11 @@ struct bnxt_filter_info *bnxt_alloc_filter(struct bnxt *bp)
 {
struct bnxt_filter_info *filter;
 
-   /* Find the 1st unused filter from the free_filter_list pool*/
-   filter = STAILQ_FIRST(&bp->free_filter_list);
+   filter = bnxt_get_unused_filter(bp);
if (!filter) {
PMD_DRV_LOG(ERR, "No more free filter resources\n");
return NULL;
}
-   STAILQ_REMOVE_HEAD(&bp->free_filter_list, next);
 
filter->mac_index = INVALID_MAC_INDEX;
/* Default to L2 MAC Addr filter */
-- 
1.8.3.1

[dpdk-dev] [PATCH 0/8] bnxt patch set

[Somnath Kotur]

Please apply.

Kalesh AP (4):
  net/bnxt: remove redundant if statement
  net/bnxt: remove redundant macro
  net/bnxt: remove unnecessary structure variable
  net/bnxt: remove a redundant variable

Somnath Kotur (4):
  net/bnxt: fix bnxt_alloc_filter() to use a common routine
  net/bnxt: fix bumping of L2 filter reference count
  net/bnxt: fix to allow group ID 0 for RSS action
  net/bnxt: fix to support zero mark id along with RSS action

 drivers/net/bnxt/bnxt.h| 29 +--
 drivers/net/bnxt/bnxt_ethdev.c | 31 +++--
 drivers/net/bnxt/bnxt_filter.c |  7 ++
 drivers/net/bnxt/bnxt_flow.c   | 52 +++---
 drivers/net/bnxt/bnxt_hwrm.c   |  6 +++--
 drivers/net/bnxt/bnxt_rxr.c| 25 
 drivers/net/bnxt/bnxt_rxr.h|  6 +++--
 drivers/net/bnxt/bnxt_stats.c  |  4 ++--
 8 files changed, 80 insertions(+), 80 deletions(-)

-- 
1.8.3.1

[dpdk-dev] [PATCH 2/8] net/bnxt: fix bumping of L2 filter reference count

[Somnath Kotur]

Now that the L2 filter reference count is bumped up in all cases
including bnxt_alloc_filter() which is issued in init, just move this
ref count bump inside the routine issuing the HWRM cmd so that it is
bumped up only if the cmd is successful.

Fixes: f0f6b5e6cf94 ("fix reusing L2 filter")

Reviewed-by: Kalesh Anakkur Purayil 
Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt_filter.c | 3 +--
 drivers/net/bnxt/bnxt_flow.c   | 1 -
 drivers/net/bnxt/bnxt_hwrm.c   | 2 ++
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/bnxt/bnxt_filter.c b/drivers/net/bnxt/bnxt_filter.c
index e218433..a1463a0 100644
--- a/drivers/net/bnxt/bnxt_filter.c
+++ b/drivers/net/bnxt/bnxt_filter.c
@@ -39,8 +39,7 @@ struct bnxt_filter_info *bnxt_alloc_filter(struct bnxt *bp)
HWRM_CFA_L2_FILTER_ALLOC_INPUT_ENABLES_L2_ADDR_MASK;
memcpy(filter->l2_addr, bp->mac_addr, RTE_ETHER_ADDR_LEN);
memset(filter->l2_addr_mask, 0xff, RTE_ETHER_ADDR_LEN);
-   /* bump up the reference count of filter */
-   filter->l2_ref_cnt++;
+
return filter;
 }
 
diff --git a/drivers/net/bnxt/bnxt_flow.c b/drivers/net/bnxt/bnxt_flow.c
index 5564c53..4b3b597 100644
--- a/drivers/net/bnxt/bnxt_flow.c
+++ b/drivers/net/bnxt/bnxt_flow.c
@@ -867,7 +867,6 @@
bnxt_free_filter(bp, filter1);
return NULL;
}
-   filter1->l2_ref_cnt++;
return filter1;
 }
 
diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index 3b01339..460cc48 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -488,6 +488,8 @@ int bnxt_hwrm_set_l2_filter(struct bnxt *bp,
filter->flow_id = rte_le_to_cpu_32(resp->flow_id);
HWRM_UNLOCK();
 
+   filter->l2_ref_cnt++;
+
return rc;
 }
 
-- 
1.8.3.1

[dpdk-dev] [PATCH 3/8] net/bnxt: fix to allow group ID 0 for RSS action

[Somnath Kotur]

Allow RSS action with group ID 0. The RSS match check will ensure if
requested RSS action configuration parameters should be allowed as per
the VNIC's RSS configuration or not and if it does match as it is
by design in OVS-DPDK use case, the default vnic can be used to create
the filter. As part of this ensure that rx_queue_cnt for the default
vnic is setup during init as this field was being set only for non-zero
vnics while handling RSS action for flow create.
Check for outofboundds erorr while accessing the vnic array.

Fixes: adc0f81c6552d ("net/bnxt: support RSS action")

Reviewed-by: Ajit Kumar Khaparde 
Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt.h| 14 +-
 drivers/net/bnxt/bnxt_ethdev.c |  4 
 drivers/net/bnxt/bnxt_flow.c   | 33 -
 3 files changed, 21 insertions(+), 30 deletions(-)

diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index ddb2681..bca9ad4 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -696,11 +696,23 @@ int bnxt_link_update(struct rte_eth_dev *eth_dev, int 
wait_to_complete,
 #define bnxt_release_flow_lock(bp) \
pthread_mutex_unlock(&(bp)->flow_lock)
 
+#define BNXT_VALID_VNIC_OR_RET(bp, vnic_id) do { \
+   if ((vnic_id) >= (bp)->max_vnics) { \
+   rte_flow_error_set(error, \
+   EINVAL, \
+   RTE_FLOW_ERROR_TYPE_ATTR_GROUP, \
+   NULL, \
+   "Group id is invalid!"); \
+   rc = -rte_errno; \
+   goto ret; \
+   } \
+} while (0)
+
 extern int bnxt_logtype_driver;
 #define PMD_DRV_LOG_RAW(level, fmt, args...) \
rte_log(RTE_LOG_ ## level, bnxt_logtype_driver, "%s(): " fmt, \
__func__, ## args)
 
 #define PMD_DRV_LOG(level, fmt, args...) \
-   PMD_DRV_LOG_RAW(level, fmt, ## args)
+ PMD_DRV_LOG_RAW(level, fmt, ## args)
 #endif
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index 2ef1169..fc3f1a8 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -295,8 +295,12 @@ static int bnxt_setup_one_vnic(struct bnxt *bp, uint16_t 
vnic_id)
 
if (BNXT_HAS_RING_GRPS(bp) && rxq->rx_deferred_start)
rxq->vnic->fw_grp_ids[j] = INVALID_HW_RING_ID;
+   else
+   vnic->rx_queue_cnt++;
}
 
+   PMD_DRV_LOG(DEBUG, "vnic->rx_queue_cnt = %d\n", vnic->rx_queue_cnt);
+
rc = bnxt_vnic_rss_configure(bp, vnic);
if (rc)
goto err_out;
diff --git a/drivers/net/bnxt/bnxt_flow.c b/drivers/net/bnxt/bnxt_flow.c
index 4b3b597..bd6c726 100644
--- a/drivers/net/bnxt/bnxt_flow.c
+++ b/drivers/net/bnxt/bnxt_flow.c
@@ -1059,16 +1059,9 @@ static int match_vnic_rss_cfg(struct bnxt *bp,
vnic_id = act_q->index;
}
 
+   BNXT_VALID_VNIC_OR_RET(bp, vnic_id);
+
vnic = &bp->vnic_info[vnic_id];
-   if (vnic == NULL) {
-   rte_flow_error_set(error,
-  EINVAL,
-  RTE_FLOW_ERROR_TYPE_ACTION,
-  act,
-  "No matching VNIC found.");
-   rc = -rte_errno;
-   goto ret;
-   }
if (vnic->rx_queue_cnt) {
if (vnic->start_grp_id != act_q->index) {
PMD_DRV_LOG(ERR,
@@ -1268,28 +1261,10 @@ static int match_vnic_rss_cfg(struct bnxt *bp,
rss = (const struct rte_flow_action_rss *)act->conf;
 
vnic_id = attr->group;
-   if (!vnic_id) {
-   PMD_DRV_LOG(ERR, "Group id cannot be 0\n");
-   rte_flow_error_set(error,
-  EINVAL,
-  RTE_FLOW_ERROR_TYPE_ATTR,
-  NULL,
-  "Group id cannot be 0");
-   rc = -rte_errno;
-   goto ret;
-   }
+
+   BNXT_VALID_VNIC_OR_RET(bp, vnic_id);
 
vnic = &bp->vnic_info[vnic_id];
-   if (vnic == NULL) {
-   rte_flow_error_set(error,
-  EINVAL,
-  RTE_FLOW_ERROR_TYPE_ACTION,
-  act,
-  "No matching VNIC for RSS group.");
-   rc = -rte_errno;
-   goto ret;
-   }
-   PMD_DRV_LOG(DEBUG, "VNIC found\n");
 
/* Check if requested RSS config matches RSS config of VNIC
 * only if it is not a fre

[dpdk-dev] [PATCH 8/8] net/bnxt: fix to support zero mark id along with RSS action

[Somnath Kotur]

Certain applications(Ex: OVS-DPDK) can issue rte_flow_create with mark id
set to 0. The mark table entry creation in the driver was assuming
non-zero mark ids. Fix it to have a valid flag for each entry.
Also, it is possible that both RSS flags and cfa_code/ mark id are
set as part of the Rx packet completion descriptor if the 'action'
for a flow is MARK + RSS.
Fix Rx completion processing to look for both fields as opposed to
only either.

Fixes: 94eb699bc82e ("net/bnxt: support flow mark action")

Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt.h  |  9 +++--
 drivers/net/bnxt/bnxt_flow.c | 20 +++-
 drivers/net/bnxt/bnxt_rxr.c  | 25 +++--
 drivers/net/bnxt/bnxt_rxr.h  |  6 --
 4 files changed, 41 insertions(+), 19 deletions(-)

diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index 1a5d542..68786a8 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -470,6 +470,11 @@ struct bnxt_error_recovery_info {
uint32_tlast_reset_counter;
 };
 
+struct bnxt_mark_info {
+   uint32_tmark_id;
+   boolvalid;
+};
+
 /* address space location of register */
 #define BNXT_FW_STATUS_REG_TYPE_MASK   3
 /* register is located in PCIe config space */
@@ -668,10 +673,10 @@ struct bnxt {
 
/* Struct to hold adapter error recovery related info */
struct bnxt_error_recovery_info *recovery_info;
-#define BNXT_MARK_TABLE_SZ (sizeof(uint32_t)  * 64 * 1024)
+#define BNXT_MARK_TABLE_SZ (sizeof(struct bnxt_mark_info)  * 64 * 1024)
 /* TCAM and EM should be 16-bit only. Other modes not supported. */
 #define BNXT_FLOW_ID_MASK  0x
-   uint32_t*mark_table;
+   struct bnxt_mark_info   *mark_table;
 };
 
 int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu);
diff --git a/drivers/net/bnxt/bnxt_flow.c b/drivers/net/bnxt/bnxt_flow.c
index bd6c726..9fb6dbd 100644
--- a/drivers/net/bnxt/bnxt_flow.c
+++ b/drivers/net/bnxt/bnxt_flow.c
@@ -1263,7 +1263,6 @@ static int match_vnic_rss_cfg(struct bnxt *bp,
vnic_id = attr->group;
 
BNXT_VALID_VNIC_OR_RET(bp, vnic_id);
-
vnic = &bp->vnic_info[vnic_id];
 
/* Check if requested RSS config matches RSS config of VNIC
@@ -1641,7 +1640,7 @@ struct bnxt_vnic_info *find_matching_vnic(struct bnxt *bp,
bool update_flow = false;
struct rte_flow *flow;
int ret = 0;
-   uint32_t tun_type;
+   uint32_t tun_type, flow_id;
 
if (BNXT_VF(bp) && !BNXT_VF_IS_TRUSTED(bp)) {
rte_flow_error_set(error, EINVAL,
@@ -1773,8 +1772,16 @@ struct bnxt_vnic_info *find_matching_vnic(struct bnxt 
*bp,
/* TCAM and EM should be 16-bit only.
 * Other modes not supported.
 */
-   bp->mark_table[filter->flow_id & BNXT_FLOW_ID_MASK] =
-   filter->mark;
+   flow_id = filter->flow_id & BNXT_FLOW_ID_MASK;
+   if (bp->mark_table[flow_id].valid) {
+   PMD_DRV_LOG(ERR,
+   "Entry for Mark id 0x%x occupied"
+   " flow id 0x%x\n",
+   filter->mark, filter->flow_id);
+   goto free_filter;
+   }
+   bp->mark_table[flow_id].valid = true;
+   bp->mark_table[flow_id].mark_id = filter->mark;
}
bnxt_release_flow_lock(bp);
return flow;
@@ -1850,6 +1857,7 @@ static int bnxt_handle_tunnel_redirect_destroy(struct 
bnxt *bp,
struct bnxt_filter_info *filter;
struct bnxt_vnic_info *vnic;
int ret = 0;
+   uint32_t flow_id;
 
filter = flow->filter;
vnic = flow->vnic;
@@ -1868,7 +1876,9 @@ static int bnxt_handle_tunnel_redirect_destroy(struct 
bnxt *bp,
PMD_DRV_LOG(ERR, "Could not find matching flow\n");
 
if (filter->valid_flags & BNXT_FLOW_MARK_FLAG) {
-   bp->mark_table[filter->flow_id & BNXT_FLOW_ID_MASK] = 0;
+   flow_id = filter->flow_id & BNXT_FLOW_ID_MASK;
+   memset(&bp->mark_table[flow_id], 0,
+  sizeof(bp->mark_table[flow_id]));
filter->flow_id = 0;
}
 
diff --git a/drivers/net/bnxt/bnxt_rxr.c b/drivers/net/bnxt/bnxt_rxr.c
index 1ae0c3c..1960b05 100644
--- a/drivers/net/bnxt/bnxt_rxr.c
+++ b/drivers/net/bnxt/bnxt_rxr.c
@@ -488,11 +488,10 @@ static int bnxt_rx_pkt(struct rte_mbuf **rx_pkt,
if (flags_type & RX_PKT_CMPL_FLAGS_RSS_VALID) {
mbuf->hash.rss = rxcmp->rss_hash;
mbuf->ol_flags |= PKT_RX_RSS_HASH;
-   } else {
-   mbuf->hash.fdir.id = bnxt_get_cfa_code_or_mark_id(rxq->bp,
- 

[dpdk-dev] [PATCH 7/8] net/bnxt: remove a redundant variable

[Somnath Kotur]

From: Kalesh AP 

Use "dev->data->dev_started" state, instead of local "dev_stopped"
to check whether port has been started or not.

Fixes: 316e412299fd ("net/bnxt: fix crash when closing")

Reviewed-by: Santoshkumar Karanappa Rastapur 
Reviewed-by: Somnath Kotur 
Signed-off-by: Kalesh AP 
---
 drivers/net/bnxt/bnxt.h|  1 -
 drivers/net/bnxt/bnxt_ethdev.c | 21 +
 2 files changed, 9 insertions(+), 13 deletions(-)

diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index 434ce28..1a5d542 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -648,7 +648,6 @@ struct bnxt {
 #define BNXT_OUTER_TPID_BD_SHFT16
uint32_touter_tpid_bd;
struct bnxt_pf_info pf;
-   uint8_t dev_stopped;
uint8_t vxlan_port_cnt;
uint8_t geneve_port_cnt;
uint16_tvxlan_port;
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index b04685c..72e5441 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -872,9 +872,9 @@ static int bnxt_dev_start_op(struct rte_eth_dev *eth_dev)
goto error;
 
eth_dev->data->scattered_rx = bnxt_scattered_rx(eth_dev);
+   eth_dev->data->dev_started = 1;
 
bnxt_link_update(eth_dev, 1, ETH_LINK_UP);
-   bp->dev_stopped = 0;
 
if (rx_offloads & DEV_RX_OFFLOAD_VLAN_FILTER)
vlan_mask |= ETH_VLAN_FILTER_MASK;
@@ -887,7 +887,6 @@ static int bnxt_dev_start_op(struct rte_eth_dev *eth_dev)
eth_dev->rx_pkt_burst = bnxt_receive_function(eth_dev);
eth_dev->tx_pkt_burst = bnxt_transmit_function(eth_dev);
 
-   eth_dev->data->dev_started = 1;
pthread_mutex_lock(&bp->def_cp_lock);
bnxt_schedule_fw_health_check(bp);
pthread_mutex_unlock(&bp->def_cp_lock);
@@ -898,7 +897,7 @@ static int bnxt_dev_start_op(struct rte_eth_dev *eth_dev)
bnxt_shutdown_nic(bp);
bnxt_free_tx_mbufs(bp);
bnxt_free_rx_mbufs(bp);
-   bp->dev_stopped = 1;
+   eth_dev->data->dev_started = 0;
return rc;
 }
 
@@ -973,7 +972,6 @@ static void bnxt_dev_stop_op(struct rte_eth_dev *eth_dev)
bp->mark_table = NULL;
 
bp->flags &= ~BNXT_FLAG_RX_VECTOR_PKT_MODE;
-   bp->dev_stopped = 1;
bp->rx_cosq_cnt = 0;
 }
 
@@ -981,7 +979,7 @@ static void bnxt_dev_close_op(struct rte_eth_dev *eth_dev)
 {
struct bnxt *bp = eth_dev->data->dev_private;
 
-   if (bp->dev_stopped == 0)
+   if (eth_dev->data->dev_started)
bnxt_dev_stop_op(eth_dev);
 
bnxt_uninit_resources(bp, false);
@@ -1174,7 +1172,7 @@ static int bnxt_promiscuous_enable_op(struct rte_eth_dev 
*eth_dev)
return rc;
 
/* Filter settings will get applied when port is started */
-   if (bp->dev_stopped == 1)
+   if (!eth_dev->data->dev_started)
return 0;
 
if (bp->vnic_info == NULL)
@@ -1203,7 +1201,7 @@ static int bnxt_promiscuous_disable_op(struct rte_eth_dev 
*eth_dev)
return rc;
 
/* Filter settings will get applied when port is started */
-   if (bp->dev_stopped == 1)
+   if (!eth_dev->data->dev_started)
return 0;
 
if (bp->vnic_info == NULL)
@@ -1232,7 +1230,7 @@ static int bnxt_allmulticast_enable_op(struct rte_eth_dev 
*eth_dev)
return rc;
 
/* Filter settings will get applied when port is started */
-   if (bp->dev_stopped == 1)
+   if (!eth_dev->data->dev_started)
return 0;
 
if (bp->vnic_info == NULL)
@@ -1261,7 +1259,7 @@ static int bnxt_allmulticast_disable_op(struct 
rte_eth_dev *eth_dev)
return rc;
 
/* Filter settings will get applied when port is started */
-   if (bp->dev_stopped == 1)
+   if (!eth_dev->data->dev_started)
return 0;
 
if (bp->vnic_info == NULL)
@@ -1985,7 +1983,7 @@ static int bnxt_free_one_vnic(struct bnxt *bp, uint16_t 
vnic_id)
return rc;
 
/* Filter settings will get applied when port is started */
-   if (bp->dev_stopped == 1)
+   if (!dev->data->dev_started)
return 0;
 
if (mask & ETH_VLAN_FILTER_MASK) {
@@ -3887,7 +3885,7 @@ static void bnxt_dev_cleanup(struct bnxt *bp)
 {
bnxt_set_hwrm_link_config(bp, false);
bp->link_info.link_up = 0;
-   if (bp->dev_stopped == 0)
+   if (bp->eth_dev->data->dev_started)
bnxt_dev_stop_op(bp->eth_dev);
 
bnxt_uninit_resources(bp, true);
@@ -4824,7 +4822,6 @@ static int bnxt_init_resources(struct bnxt *bp, bool 
reconfig_dev)
 
bp = eth_dev->data->dev_private;
 
-   bp->dev_stopped = 1;
bp->flags &= ~BNXT_FLAG_RX_VECTOR_PKT_MODE;
 
if (bnxt_vf_pciid(pci_dev->id.device_id))
-- 
1.8.3.1

[dpdk-dev] [PATCH 6/8] net/bnxt: remove unnecessary structure variable

[Somnath Kotur]

From: Kalesh AP 

This change could help in reducing the size of bnxt PMD private
data structure by converting a uint8_t variable to use bit map flag.

Fixes: 5cd0e2889c432 ("net/bnxt: support NIC Partitioning")

Reviewed-by: Santoshkumar Karanappa Rastapur 
Reviewed-by: Somnath Kotur 
Signed-off-by: Kalesh AP 
---
 drivers/net/bnxt/bnxt.h  | 4 ++--
 drivers/net/bnxt/bnxt_hwrm.c | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index 95f1f1a..434ce28 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -520,12 +520,13 @@ struct bnxt {
 #define BNXT_FLAG_FW_HEALTH_CHECK_SCHEDULEDBIT(18)
 #define BNXT_FLAG_EXT_STATS_SUPPORTED  BIT(19)
 #define BNXT_FLAG_NEW_RM   BIT(20)
+#define BNXT_FLAG_NPAR_PF  BIT(21)
 #define BNXT_FLAG_FW_CAP_ONE_STEP_TX_TSBIT(22)
 #define BNXT_FLAG_ADV_FLOW_MGMTBIT(23)
 #define BNXT_FLAG_RX_VECTOR_PKT_MODE   BIT(24)
 #define BNXT_PF(bp)(!((bp)->flags & BNXT_FLAG_VF))
 #define BNXT_VF(bp)((bp)->flags & BNXT_FLAG_VF)
-#define BNXT_NPAR(bp)  ((bp)->port_partition_type)
+#define BNXT_NPAR(bp)  ((bp)->flags & BNXT_FLAG_NPAR_PF)
 #define BNXT_MH(bp) ((bp)->flags & BNXT_FLAG_MULTI_HOST)
 #define BNXT_SINGLE_PF(bp)  (BNXT_PF(bp) && !BNXT_NPAR(bp) && !BNXT_MH(bp))
 #define BNXT_USE_CHIMP_MB  0 //For non-CFA commands, everything uses Chimp.
@@ -647,7 +648,6 @@ struct bnxt {
 #define BNXT_OUTER_TPID_BD_SHFT16
uint32_touter_tpid_bd;
struct bnxt_pf_info pf;
-   uint8_t port_partition_type;
uint8_t dev_stopped;
uint8_t vxlan_port_cnt;
uint8_t geneve_port_cnt;
diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index 460cc48..f325aff 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -2952,10 +2952,10 @@ int bnxt_hwrm_func_qcfg(struct bnxt *bp, uint16_t *mtu)
case HWRM_FUNC_QCFG_OUTPUT_PORT_PARTITION_TYPE_NPAR1_5:
case HWRM_FUNC_QCFG_OUTPUT_PORT_PARTITION_TYPE_NPAR2_0:
/* FALLTHROUGH */
-   bp->port_partition_type = resp->port_partition_type;
+   bp->flags |= BNXT_FLAG_NPAR_PF;
break;
default:
-   bp->port_partition_type = 0;
+   bp->flags &= ~BNXT_FLAG_NPAR_PF;
break;
}
 
-- 
1.8.3.1

Re: [dpdk-dev] [PATCH v2 2/2] examples/crypto: crypto test application

[Akhil Goyal]

Hi,
> 
> From: Sardar Shamsher Singh 
> 
> test application to test Cipher/Authentication/AEAD in AMD CCP
> a crypto controller. This test application can be used as
> reference to develop/test crypto algo.
> 
> Signed-off-by: Sardar Shamsher Singh 
> ---
Why do you need a separate example application to test an algo. 
Is it not possible to test it with app/test/test_cryptodev.c and 
Examples/l2fwd-crypto?

-Akhil

[dpdk-dev] [PATCH 2/9] net/bnxt: fix bumping of L2 filter reference count

[Somnath Kotur]

Now that the L2 filter reference count is bumped up in all cases
including bnxt_alloc_filter() which is issued in init, just move this
ref count bump inside the routine issuing the HWRM cmd so that it is
bumped up only if the cmd is successful.

Fixes: f0f6b5e6cf94 ("fix reusing L2 filter")

Reviewed-by: Kalesh Anakkur Purayil 
Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt_filter.c | 3 +--
 drivers/net/bnxt/bnxt_flow.c   | 1 -
 drivers/net/bnxt/bnxt_hwrm.c   | 2 ++
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/bnxt/bnxt_filter.c b/drivers/net/bnxt/bnxt_filter.c
index e218433..a1463a0 100644
--- a/drivers/net/bnxt/bnxt_filter.c
+++ b/drivers/net/bnxt/bnxt_filter.c
@@ -39,8 +39,7 @@ struct bnxt_filter_info *bnxt_alloc_filter(struct bnxt *bp)
HWRM_CFA_L2_FILTER_ALLOC_INPUT_ENABLES_L2_ADDR_MASK;
memcpy(filter->l2_addr, bp->mac_addr, RTE_ETHER_ADDR_LEN);
memset(filter->l2_addr_mask, 0xff, RTE_ETHER_ADDR_LEN);
-   /* bump up the reference count of filter */
-   filter->l2_ref_cnt++;
+
return filter;
 }
 
diff --git a/drivers/net/bnxt/bnxt_flow.c b/drivers/net/bnxt/bnxt_flow.c
index 5564c53..4b3b597 100644
--- a/drivers/net/bnxt/bnxt_flow.c
+++ b/drivers/net/bnxt/bnxt_flow.c
@@ -867,7 +867,6 @@
bnxt_free_filter(bp, filter1);
return NULL;
}
-   filter1->l2_ref_cnt++;
return filter1;
 }
 
diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index 3b01339..460cc48 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -488,6 +488,8 @@ int bnxt_hwrm_set_l2_filter(struct bnxt *bp,
filter->flow_id = rte_le_to_cpu_32(resp->flow_id);
HWRM_UNLOCK();
 
+   filter->l2_ref_cnt++;
+
return rc;
 }
 
-- 
1.8.3.1

[dpdk-dev] [PATCH 4/9] net/bnxt: remove redundant if statement

[Somnath Kotur]

From: Kalesh AP 

Since "eth_dev->data->dev_started" has been assigned to 0 at the
beginning of bnxt_dev_stop_op() routine, the code inside the if()
condition is redundant. Remove it.

Anyways "eth_dev->data->dev_link.link_status" will be set to 0 in
bnxt_dev_set_link_down_op() later in the routine.

Fixes: 316e412299fd ("net/bnxt: fix crash when closing")

Signed-off-by: Kalesh AP 
Reviewed-by: Ajit Kumar Khaparde 
Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt_ethdev.c | 4 
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index fc3f1a8..267a361 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -948,10 +948,6 @@ static void bnxt_dev_stop_op(struct rte_eth_dev *eth_dev)
bnxt_cancel_fw_health_check(bp);
 
bp->flags &= ~BNXT_FLAG_INIT_DONE;
-   if (bp->eth_dev->data->dev_started) {
-   /* TBD: STOP HW queues DMA */
-   eth_dev->data->dev_link.link_status = 0;
-   }
bnxt_dev_set_link_down_op(eth_dev);
 
/* Wait for link to be reset and the async notification to process.
-- 
1.8.3.1

[dpdk-dev] [PATCH 5/9] net/bnxt: remove redundant macro

[Somnath Kotur]

From: Kalesh AP 

Use "dev->data->dev_started" state, instead of local BNXT_FLAG_INIT_DONE
to check whether device has been initialised or not.

Fixes: ed2ced6fe927 ("net/bnxt: check initialization before accessing stats")

Reviewed-by: Santoshkumar Karanappa Rastapur 
Reviewed-by: Ajit Kumar Khaparde 
Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt.h| 1 -
 drivers/net/bnxt/bnxt_ethdev.c | 2 --
 drivers/net/bnxt/bnxt_stats.c  | 4 ++--
 3 files changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index bca9ad4..95f1f1a 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -520,7 +520,6 @@ struct bnxt {
 #define BNXT_FLAG_FW_HEALTH_CHECK_SCHEDULEDBIT(18)
 #define BNXT_FLAG_EXT_STATS_SUPPORTED  BIT(19)
 #define BNXT_FLAG_NEW_RM   BIT(20)
-#define BNXT_FLAG_INIT_DONEBIT(21)
 #define BNXT_FLAG_FW_CAP_ONE_STEP_TX_TSBIT(22)
 #define BNXT_FLAG_ADV_FLOW_MGMTBIT(23)
 #define BNXT_FLAG_RX_VECTOR_PKT_MODE   BIT(24)
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index 267a361..b04685c 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -887,7 +887,6 @@ static int bnxt_dev_start_op(struct rte_eth_dev *eth_dev)
eth_dev->rx_pkt_burst = bnxt_receive_function(eth_dev);
eth_dev->tx_pkt_burst = bnxt_transmit_function(eth_dev);
 
-   bp->flags |= BNXT_FLAG_INIT_DONE;
eth_dev->data->dev_started = 1;
pthread_mutex_lock(&bp->def_cp_lock);
bnxt_schedule_fw_health_check(bp);
@@ -947,7 +946,6 @@ static void bnxt_dev_stop_op(struct rte_eth_dev *eth_dev)
 
bnxt_cancel_fw_health_check(bp);
 
-   bp->flags &= ~BNXT_FLAG_INIT_DONE;
bnxt_dev_set_link_down_op(eth_dev);
 
/* Wait for link to be reset and the async notification to process.
diff --git a/drivers/net/bnxt/bnxt_stats.c b/drivers/net/bnxt/bnxt_stats.c
index 32f112b..4668629 100644
--- a/drivers/net/bnxt/bnxt_stats.c
+++ b/drivers/net/bnxt/bnxt_stats.c
@@ -389,7 +389,7 @@ int bnxt_stats_get_op(struct rte_eth_dev *eth_dev,
if (rc)
return rc;
 
-   if (!(bp->flags & BNXT_FLAG_INIT_DONE))
+   if (!eth_dev->data->dev_started)
return -EIO;
 
num_q_stats = RTE_MIN(bp->rx_cp_nr_rings,
@@ -434,7 +434,7 @@ int bnxt_stats_reset_op(struct rte_eth_dev *eth_dev)
if (ret)
return ret;
 
-   if (!(bp->flags & BNXT_FLAG_INIT_DONE)) {
+   if (!eth_dev->data->dev_started) {
PMD_DRV_LOG(ERR, "Device Initialization not complete!\n");
return -EINVAL;
}
-- 
1.8.3.1

[dpdk-dev] [PATCH 1/9] net/bnxt: fix bnxt_alloc_filter() to use a common routine

[Somnath Kotur]

Invoke bnxt_get_unused_filter() inside bnxt_alloc_filter() so that
all filters are allocated from one common routine.

Fixes: f92735db1e4c ("net/bnxt: add L2 filter alloc/init/free")

Reviewed-by: Kalesh Anakkur Purayil 
Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt_filter.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/bnxt/bnxt_filter.c b/drivers/net/bnxt/bnxt_filter.c
index b31f104..e218433 100644
--- a/drivers/net/bnxt/bnxt_filter.c
+++ b/drivers/net/bnxt/bnxt_filter.c
@@ -26,13 +26,11 @@ struct bnxt_filter_info *bnxt_alloc_filter(struct bnxt *bp)
 {
struct bnxt_filter_info *filter;
 
-   /* Find the 1st unused filter from the free_filter_list pool*/
-   filter = STAILQ_FIRST(&bp->free_filter_list);
+   filter = bnxt_get_unused_filter(bp);
if (!filter) {
PMD_DRV_LOG(ERR, "No more free filter resources\n");
return NULL;
}
-   STAILQ_REMOVE_HEAD(&bp->free_filter_list, next);
 
filter->mac_index = INVALID_MAC_INDEX;
/* Default to L2 MAC Addr filter */
-- 
1.8.3.1

[dpdk-dev] [PATCH v2 0/9] bnxt patchset

[Somnath Kotur]

Please apply.

Kalesh AP (5):
  net/bnxt: remove redundant if statement
  net/bnxt: remove redundant macro
  net/bnxt: remove unnecessary structure variable
  net/bnxt: remove a redundant variable
  net/bnxt: fix coverity warnings

Somnath Kotur (4):
  net/bnxt: fix bnxt_alloc_filter() to use a common routine
  net/bnxt: fix bumping of L2 filter reference count
  net/bnxt: fix to allow group ID 0 for RSS action
  net/bnxt: fix to support zero mark id along with RSS action

 drivers/net/bnxt/bnxt.h| 29 +--
 drivers/net/bnxt/bnxt_ethdev.c | 49 ++-
 drivers/net/bnxt/bnxt_filter.c |  7 ++
 drivers/net/bnxt/bnxt_flow.c   | 52 +++---
 drivers/net/bnxt/bnxt_hwrm.c   |  6 +++--
 drivers/net/bnxt/bnxt_rxr.c| 25 
 drivers/net/bnxt/bnxt_rxr.h|  6 +++--
 drivers/net/bnxt/bnxt_stats.c  |  4 ++--
 8 files changed, 95 insertions(+), 83 deletions(-)

-- 
v1->v2: Added one more patch to the set ("net/bnxt: fix coverity warnings")
1.8.3.1

[dpdk-dev] [PATCH 6/9] net/bnxt: remove unnecessary structure variable

[Somnath Kotur]

From: Kalesh AP 

This change could help in reducing the size of bnxt PMD private
data structure by converting a uint8_t variable to use bit map flag.

Fixes: 5cd0e2889c432 ("net/bnxt: support NIC Partitioning")

Reviewed-by: Santoshkumar Karanappa Rastapur 
Reviewed-by: Somnath Kotur 
Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt.h  | 4 ++--
 drivers/net/bnxt/bnxt_hwrm.c | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index 95f1f1a..434ce28 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -520,12 +520,13 @@ struct bnxt {
 #define BNXT_FLAG_FW_HEALTH_CHECK_SCHEDULEDBIT(18)
 #define BNXT_FLAG_EXT_STATS_SUPPORTED  BIT(19)
 #define BNXT_FLAG_NEW_RM   BIT(20)
+#define BNXT_FLAG_NPAR_PF  BIT(21)
 #define BNXT_FLAG_FW_CAP_ONE_STEP_TX_TSBIT(22)
 #define BNXT_FLAG_ADV_FLOW_MGMTBIT(23)
 #define BNXT_FLAG_RX_VECTOR_PKT_MODE   BIT(24)
 #define BNXT_PF(bp)(!((bp)->flags & BNXT_FLAG_VF))
 #define BNXT_VF(bp)((bp)->flags & BNXT_FLAG_VF)
-#define BNXT_NPAR(bp)  ((bp)->port_partition_type)
+#define BNXT_NPAR(bp)  ((bp)->flags & BNXT_FLAG_NPAR_PF)
 #define BNXT_MH(bp) ((bp)->flags & BNXT_FLAG_MULTI_HOST)
 #define BNXT_SINGLE_PF(bp)  (BNXT_PF(bp) && !BNXT_NPAR(bp) && !BNXT_MH(bp))
 #define BNXT_USE_CHIMP_MB  0 //For non-CFA commands, everything uses Chimp.
@@ -647,7 +648,6 @@ struct bnxt {
 #define BNXT_OUTER_TPID_BD_SHFT16
uint32_touter_tpid_bd;
struct bnxt_pf_info pf;
-   uint8_t port_partition_type;
uint8_t dev_stopped;
uint8_t vxlan_port_cnt;
uint8_t geneve_port_cnt;
diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index 460cc48..f325aff 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -2952,10 +2952,10 @@ int bnxt_hwrm_func_qcfg(struct bnxt *bp, uint16_t *mtu)
case HWRM_FUNC_QCFG_OUTPUT_PORT_PARTITION_TYPE_NPAR1_5:
case HWRM_FUNC_QCFG_OUTPUT_PORT_PARTITION_TYPE_NPAR2_0:
/* FALLTHROUGH */
-   bp->port_partition_type = resp->port_partition_type;
+   bp->flags |= BNXT_FLAG_NPAR_PF;
break;
default:
-   bp->port_partition_type = 0;
+   bp->flags &= ~BNXT_FLAG_NPAR_PF;
break;
}
 
-- 
1.8.3.1

[dpdk-dev] [PATCH 3/9] net/bnxt: fix to allow group ID 0 for RSS action

[Somnath Kotur]

Allow RSS action with group ID 0. The RSS match check will ensure if
requested RSS action configuration parameters should be allowed as per
the VNIC's RSS configuration or not and if it does match as it is
by design in OVS-DPDK use case, the default vnic can be used to create
the filter. As part of this ensure that rx_queue_cnt for the default
vnic is setup during init as this field was being set only for non-zero
vnics while handling RSS action for flow create.
Check for outofboundds erorr while accessing the vnic array.

Fixes: adc0f81c6552d ("net/bnxt: support RSS action")

Reviewed-by: Ajit Kumar Khaparde 
Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt.h| 14 +-
 drivers/net/bnxt/bnxt_ethdev.c |  4 
 drivers/net/bnxt/bnxt_flow.c   | 33 -
 3 files changed, 21 insertions(+), 30 deletions(-)

diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index ddb2681..bca9ad4 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -696,11 +696,23 @@ int bnxt_link_update(struct rte_eth_dev *eth_dev, int 
wait_to_complete,
 #define bnxt_release_flow_lock(bp) \
pthread_mutex_unlock(&(bp)->flow_lock)
 
+#define BNXT_VALID_VNIC_OR_RET(bp, vnic_id) do { \
+   if ((vnic_id) >= (bp)->max_vnics) { \
+   rte_flow_error_set(error, \
+   EINVAL, \
+   RTE_FLOW_ERROR_TYPE_ATTR_GROUP, \
+   NULL, \
+   "Group id is invalid!"); \
+   rc = -rte_errno; \
+   goto ret; \
+   } \
+} while (0)
+
 extern int bnxt_logtype_driver;
 #define PMD_DRV_LOG_RAW(level, fmt, args...) \
rte_log(RTE_LOG_ ## level, bnxt_logtype_driver, "%s(): " fmt, \
__func__, ## args)
 
 #define PMD_DRV_LOG(level, fmt, args...) \
-   PMD_DRV_LOG_RAW(level, fmt, ## args)
+ PMD_DRV_LOG_RAW(level, fmt, ## args)
 #endif
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index 2ef1169..fc3f1a8 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -295,8 +295,12 @@ static int bnxt_setup_one_vnic(struct bnxt *bp, uint16_t 
vnic_id)
 
if (BNXT_HAS_RING_GRPS(bp) && rxq->rx_deferred_start)
rxq->vnic->fw_grp_ids[j] = INVALID_HW_RING_ID;
+   else
+   vnic->rx_queue_cnt++;
}
 
+   PMD_DRV_LOG(DEBUG, "vnic->rx_queue_cnt = %d\n", vnic->rx_queue_cnt);
+
rc = bnxt_vnic_rss_configure(bp, vnic);
if (rc)
goto err_out;
diff --git a/drivers/net/bnxt/bnxt_flow.c b/drivers/net/bnxt/bnxt_flow.c
index 4b3b597..bd6c726 100644
--- a/drivers/net/bnxt/bnxt_flow.c
+++ b/drivers/net/bnxt/bnxt_flow.c
@@ -1059,16 +1059,9 @@ static int match_vnic_rss_cfg(struct bnxt *bp,
vnic_id = act_q->index;
}
 
+   BNXT_VALID_VNIC_OR_RET(bp, vnic_id);
+
vnic = &bp->vnic_info[vnic_id];
-   if (vnic == NULL) {
-   rte_flow_error_set(error,
-  EINVAL,
-  RTE_FLOW_ERROR_TYPE_ACTION,
-  act,
-  "No matching VNIC found.");
-   rc = -rte_errno;
-   goto ret;
-   }
if (vnic->rx_queue_cnt) {
if (vnic->start_grp_id != act_q->index) {
PMD_DRV_LOG(ERR,
@@ -1268,28 +1261,10 @@ static int match_vnic_rss_cfg(struct bnxt *bp,
rss = (const struct rte_flow_action_rss *)act->conf;
 
vnic_id = attr->group;
-   if (!vnic_id) {
-   PMD_DRV_LOG(ERR, "Group id cannot be 0\n");
-   rte_flow_error_set(error,
-  EINVAL,
-  RTE_FLOW_ERROR_TYPE_ATTR,
-  NULL,
-  "Group id cannot be 0");
-   rc = -rte_errno;
-   goto ret;
-   }
+
+   BNXT_VALID_VNIC_OR_RET(bp, vnic_id);
 
vnic = &bp->vnic_info[vnic_id];
-   if (vnic == NULL) {
-   rte_flow_error_set(error,
-  EINVAL,
-  RTE_FLOW_ERROR_TYPE_ACTION,
-  act,
-  "No matching VNIC for RSS group.");
-   rc = -rte_errno;
-   goto ret;
-   }
-   PMD_DRV_LOG(DEBUG, "VNIC found\n");
 
/* Check if requested RSS config matches RSS config of VNIC
 * only if it is not a fre

[dpdk-dev] [PATCH 9/9] net/bnxt: fix coverity warnings

[Somnath Kotur]

From: Kalesh AP 

return value stored in "ret" but it has been overwritten before use.

CID 353621:  Code maintainability issues  (UNUSED_VALUE)
Fixes: 57813868 ("net/bnxt: fix VLAN strip")
Fixes: df6cd7c1 ("net/bnxt: handle reset notify async event from FW")

Signed-off-by: Kalesh AP 
Reviewed-by: Ajit Khaparde 
Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt_ethdev.c | 18 +++---
 1 file changed, 15 insertions(+), 3 deletions(-)

diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index 72e5441..c1cb401 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -1956,9 +1956,15 @@ static int bnxt_free_one_vnic(struct bnxt *bp, uint16_t 
vnic_id)
if (bp->eth_dev->data->dev_conf.rxmode.offloads &
DEV_RX_OFFLOAD_VLAN_FILTER) {
rc = bnxt_add_vlan_filter(bp, 0);
-   bnxt_restore_vlan_filters(bp);
+   if (rc)
+   return rc;
+   rc = bnxt_restore_vlan_filters(bp);
+   if (rc)
+   return rc;
} else {
rc = bnxt_add_mac_filter(bp, vnic, NULL, 0, 0);
+   if (rc)
+   return rc;
}
 
rc = bnxt_hwrm_cfa_l2_set_rx_mask(bp, vnic, 0, NULL);
@@ -3961,10 +3967,16 @@ static int bnxt_restore_filters(struct bnxt *bp)
struct rte_eth_dev *dev = bp->eth_dev;
int ret = 0;
 
-   if (dev->data->all_multicast)
+   if (dev->data->all_multicast) {
ret = bnxt_allmulticast_enable_op(dev);
-   if (dev->data->promiscuous)
+   if (ret)
+   return ret;
+   }
+   if (dev->data->promiscuous) {
ret = bnxt_promiscuous_enable_op(dev);
+   if (ret)
+   return ret;
+   }
 
ret = bnxt_restore_mac_filters(bp);
if (ret)
-- 
1.8.3.1

[dpdk-dev] [PATCH 7/9] net/bnxt: remove a redundant variable

[Somnath Kotur]

From: Kalesh AP 

Use "dev->data->dev_started" state, instead of local "dev_stopped"
to check whether port has been started or not.

Fixes: 316e412299fd ("net/bnxt: fix crash when closing")

Reviewed-by: Santoshkumar Karanappa Rastapur 
Reviewed-by: Somnath Kotur 
Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt.h|  1 -
 drivers/net/bnxt/bnxt_ethdev.c | 21 +
 2 files changed, 9 insertions(+), 13 deletions(-)

diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index 434ce28..1a5d542 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -648,7 +648,6 @@ struct bnxt {
 #define BNXT_OUTER_TPID_BD_SHFT16
uint32_touter_tpid_bd;
struct bnxt_pf_info pf;
-   uint8_t dev_stopped;
uint8_t vxlan_port_cnt;
uint8_t geneve_port_cnt;
uint16_tvxlan_port;
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index b04685c..72e5441 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -872,9 +872,9 @@ static int bnxt_dev_start_op(struct rte_eth_dev *eth_dev)
goto error;
 
eth_dev->data->scattered_rx = bnxt_scattered_rx(eth_dev);
+   eth_dev->data->dev_started = 1;
 
bnxt_link_update(eth_dev, 1, ETH_LINK_UP);
-   bp->dev_stopped = 0;
 
if (rx_offloads & DEV_RX_OFFLOAD_VLAN_FILTER)
vlan_mask |= ETH_VLAN_FILTER_MASK;
@@ -887,7 +887,6 @@ static int bnxt_dev_start_op(struct rte_eth_dev *eth_dev)
eth_dev->rx_pkt_burst = bnxt_receive_function(eth_dev);
eth_dev->tx_pkt_burst = bnxt_transmit_function(eth_dev);
 
-   eth_dev->data->dev_started = 1;
pthread_mutex_lock(&bp->def_cp_lock);
bnxt_schedule_fw_health_check(bp);
pthread_mutex_unlock(&bp->def_cp_lock);
@@ -898,7 +897,7 @@ static int bnxt_dev_start_op(struct rte_eth_dev *eth_dev)
bnxt_shutdown_nic(bp);
bnxt_free_tx_mbufs(bp);
bnxt_free_rx_mbufs(bp);
-   bp->dev_stopped = 1;
+   eth_dev->data->dev_started = 0;
return rc;
 }
 
@@ -973,7 +972,6 @@ static void bnxt_dev_stop_op(struct rte_eth_dev *eth_dev)
bp->mark_table = NULL;
 
bp->flags &= ~BNXT_FLAG_RX_VECTOR_PKT_MODE;
-   bp->dev_stopped = 1;
bp->rx_cosq_cnt = 0;
 }
 
@@ -981,7 +979,7 @@ static void bnxt_dev_close_op(struct rte_eth_dev *eth_dev)
 {
struct bnxt *bp = eth_dev->data->dev_private;
 
-   if (bp->dev_stopped == 0)
+   if (eth_dev->data->dev_started)
bnxt_dev_stop_op(eth_dev);
 
bnxt_uninit_resources(bp, false);
@@ -1174,7 +1172,7 @@ static int bnxt_promiscuous_enable_op(struct rte_eth_dev 
*eth_dev)
return rc;
 
/* Filter settings will get applied when port is started */
-   if (bp->dev_stopped == 1)
+   if (!eth_dev->data->dev_started)
return 0;
 
if (bp->vnic_info == NULL)
@@ -1203,7 +1201,7 @@ static int bnxt_promiscuous_disable_op(struct rte_eth_dev 
*eth_dev)
return rc;
 
/* Filter settings will get applied when port is started */
-   if (bp->dev_stopped == 1)
+   if (!eth_dev->data->dev_started)
return 0;
 
if (bp->vnic_info == NULL)
@@ -1232,7 +1230,7 @@ static int bnxt_allmulticast_enable_op(struct rte_eth_dev 
*eth_dev)
return rc;
 
/* Filter settings will get applied when port is started */
-   if (bp->dev_stopped == 1)
+   if (!eth_dev->data->dev_started)
return 0;
 
if (bp->vnic_info == NULL)
@@ -1261,7 +1259,7 @@ static int bnxt_allmulticast_disable_op(struct 
rte_eth_dev *eth_dev)
return rc;
 
/* Filter settings will get applied when port is started */
-   if (bp->dev_stopped == 1)
+   if (!eth_dev->data->dev_started)
return 0;
 
if (bp->vnic_info == NULL)
@@ -1985,7 +1983,7 @@ static int bnxt_free_one_vnic(struct bnxt *bp, uint16_t 
vnic_id)
return rc;
 
/* Filter settings will get applied when port is started */
-   if (bp->dev_stopped == 1)
+   if (!dev->data->dev_started)
return 0;
 
if (mask & ETH_VLAN_FILTER_MASK) {
@@ -3887,7 +3885,7 @@ static void bnxt_dev_cleanup(struct bnxt *bp)
 {
bnxt_set_hwrm_link_config(bp, false);
bp->link_info.link_up = 0;
-   if (bp->dev_stopped == 0)
+   if (bp->eth_dev->data->dev_started)
bnxt_dev_stop_op(bp->eth_dev);
 
bnxt_uninit_resources(bp, true);
@@ -4824,7 +4822,6 @@ static int bnxt_init_resources(struct bnxt *bp, bool 
reconfig_dev)
 
bp = eth_dev->data->dev_private;
 
-   bp->dev_stopped = 1;
bp->flags &= ~BNXT_FLAG_RX_VECTOR_PKT_MODE;
 
if (bnxt_vf_pciid(pci_dev->id.device_id))
-- 
1.8.3.1

[dpdk-dev] [PATCH 8/9] net/bnxt: fix to support zero mark id along with RSS action

[Somnath Kotur]

Certain applications(Ex: OVS-DPDK) can issue rte_flow_create with mark id
set to 0. The mark table entry creation in the driver was assuming
non-zero mark ids. Fix it to have a valid flag for each entry.
Also, it is possible that both RSS flags and cfa_code/ mark id are
set as part of the Rx packet completion descriptor if the 'action'
for a flow is MARK + RSS.
Fix Rx completion processing to look for both fields as opposed to
only either.

Fixes: 94eb699bc82e ("net/bnxt: support flow mark action")

Signed-off-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt.h  |  9 +++--
 drivers/net/bnxt/bnxt_flow.c | 20 +++-
 drivers/net/bnxt/bnxt_rxr.c  | 25 +++--
 drivers/net/bnxt/bnxt_rxr.h  |  6 --
 4 files changed, 41 insertions(+), 19 deletions(-)

diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index 1a5d542..68786a8 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -470,6 +470,11 @@ struct bnxt_error_recovery_info {
uint32_tlast_reset_counter;
 };
 
+struct bnxt_mark_info {
+   uint32_tmark_id;
+   boolvalid;
+};
+
 /* address space location of register */
 #define BNXT_FW_STATUS_REG_TYPE_MASK   3
 /* register is located in PCIe config space */
@@ -668,10 +673,10 @@ struct bnxt {
 
/* Struct to hold adapter error recovery related info */
struct bnxt_error_recovery_info *recovery_info;
-#define BNXT_MARK_TABLE_SZ (sizeof(uint32_t)  * 64 * 1024)
+#define BNXT_MARK_TABLE_SZ (sizeof(struct bnxt_mark_info)  * 64 * 1024)
 /* TCAM and EM should be 16-bit only. Other modes not supported. */
 #define BNXT_FLOW_ID_MASK  0x
-   uint32_t*mark_table;
+   struct bnxt_mark_info   *mark_table;
 };
 
 int bnxt_mtu_set_op(struct rte_eth_dev *eth_dev, uint16_t new_mtu);
diff --git a/drivers/net/bnxt/bnxt_flow.c b/drivers/net/bnxt/bnxt_flow.c
index bd6c726..9fb6dbd 100644
--- a/drivers/net/bnxt/bnxt_flow.c
+++ b/drivers/net/bnxt/bnxt_flow.c
@@ -1263,7 +1263,6 @@ static int match_vnic_rss_cfg(struct bnxt *bp,
vnic_id = attr->group;
 
BNXT_VALID_VNIC_OR_RET(bp, vnic_id);
-
vnic = &bp->vnic_info[vnic_id];
 
/* Check if requested RSS config matches RSS config of VNIC
@@ -1641,7 +1640,7 @@ struct bnxt_vnic_info *find_matching_vnic(struct bnxt *bp,
bool update_flow = false;
struct rte_flow *flow;
int ret = 0;
-   uint32_t tun_type;
+   uint32_t tun_type, flow_id;
 
if (BNXT_VF(bp) && !BNXT_VF_IS_TRUSTED(bp)) {
rte_flow_error_set(error, EINVAL,
@@ -1773,8 +1772,16 @@ struct bnxt_vnic_info *find_matching_vnic(struct bnxt 
*bp,
/* TCAM and EM should be 16-bit only.
 * Other modes not supported.
 */
-   bp->mark_table[filter->flow_id & BNXT_FLOW_ID_MASK] =
-   filter->mark;
+   flow_id = filter->flow_id & BNXT_FLOW_ID_MASK;
+   if (bp->mark_table[flow_id].valid) {
+   PMD_DRV_LOG(ERR,
+   "Entry for Mark id 0x%x occupied"
+   " flow id 0x%x\n",
+   filter->mark, filter->flow_id);
+   goto free_filter;
+   }
+   bp->mark_table[flow_id].valid = true;
+   bp->mark_table[flow_id].mark_id = filter->mark;
}
bnxt_release_flow_lock(bp);
return flow;
@@ -1850,6 +1857,7 @@ static int bnxt_handle_tunnel_redirect_destroy(struct 
bnxt *bp,
struct bnxt_filter_info *filter;
struct bnxt_vnic_info *vnic;
int ret = 0;
+   uint32_t flow_id;
 
filter = flow->filter;
vnic = flow->vnic;
@@ -1868,7 +1876,9 @@ static int bnxt_handle_tunnel_redirect_destroy(struct 
bnxt *bp,
PMD_DRV_LOG(ERR, "Could not find matching flow\n");
 
if (filter->valid_flags & BNXT_FLOW_MARK_FLAG) {
-   bp->mark_table[filter->flow_id & BNXT_FLOW_ID_MASK] = 0;
+   flow_id = filter->flow_id & BNXT_FLOW_ID_MASK;
+   memset(&bp->mark_table[flow_id], 0,
+  sizeof(bp->mark_table[flow_id]));
filter->flow_id = 0;
}
 
diff --git a/drivers/net/bnxt/bnxt_rxr.c b/drivers/net/bnxt/bnxt_rxr.c
index 1ae0c3c..1960b05 100644
--- a/drivers/net/bnxt/bnxt_rxr.c
+++ b/drivers/net/bnxt/bnxt_rxr.c
@@ -488,11 +488,10 @@ static int bnxt_rx_pkt(struct rte_mbuf **rx_pkt,
if (flags_type & RX_PKT_CMPL_FLAGS_RSS_VALID) {
mbuf->hash.rss = rxcmp->rss_hash;
mbuf->ol_flags |= PKT_RX_RSS_HASH;
-   } else {
-   mbuf->hash.fdir.id = bnxt_get_cfa_code_or_mark_id(rxq->bp,
- 

[dpdk-dev] [PATCH v1] net/axgbe: swupport flow control API

[asomalap]

From: Amaranath Somalapuram 

adding api for flow_ctrl_get and flow_ctrl_set
by default axgbe driver flow control is dissabled
adding dpdk flow control to set water high and low

Signed-off-by: Amaranath Somalapuram 
---
 drivers/net/axgbe/axgbe_ethdev.c | 85 +++-
 drivers/net/axgbe/axgbe_ethdev.h | 10 
 2 files changed, 94 insertions(+), 1 deletion(-)

diff --git a/drivers/net/axgbe/axgbe_ethdev.c b/drivers/net/axgbe/axgbe_ethdev.c
index d0b6f091f..b88ad55ac 100644
--- a/drivers/net/axgbe/axgbe_ethdev.c
+++ b/drivers/net/axgbe/axgbe_ethdev.c
@@ -43,7 +43,11 @@ axgbe_dev_xstats_get_names_by_id(struct rte_eth_dev *dev,
 unsigned int size);
 static int axgbe_dev_xstats_reset(struct rte_eth_dev *dev);
 static int  axgbe_dev_info_get(struct rte_eth_dev *dev,
-  struct rte_eth_dev_info *dev_info);
+   struct rte_eth_dev_info *dev_info);
+static int axgbe_flow_ctrl_get(struct rte_eth_dev *dev,
+   struct rte_eth_fc_conf *fc_conf);
+static int axgbe_flow_ctrl_set(struct rte_eth_dev *dev,
+   struct rte_eth_fc_conf *fc_conf);
 
 struct axgbe_xstats {
char name[RTE_ETH_XSTATS_NAME_SIZE];
@@ -170,6 +174,8 @@ static const struct eth_dev_ops axgbe_eth_dev_ops = {
.rx_queue_release = axgbe_dev_rx_queue_release,
.tx_queue_setup   = axgbe_dev_tx_queue_setup,
.tx_queue_release = axgbe_dev_tx_queue_release,
+   .flow_ctrl_get= axgbe_flow_ctrl_get,
+   .flow_ctrl_set= axgbe_flow_ctrl_set,
 };
 
 static int axgbe_phy_reset(struct axgbe_port *pdata)
@@ -815,6 +821,83 @@ axgbe_dev_info_get(struct rte_eth_dev *dev, struct 
rte_eth_dev_info *dev_info)
 
return 0;
 }
+static int
+axgbe_flow_ctrl_get(struct rte_eth_dev *dev, struct rte_eth_fc_conf *fc_conf)
+{
+   struct axgbe_port *pdata = dev->data->dev_private;
+   struct xgbe_fc_info fc = pdata->fc;
+   unsigned int reg, reg_val = 0;
+
+   reg = MAC_Q0TFCR;
+   reg_val = AXGMAC_IOREAD(pdata, reg);
+   fc.low_water[0] =  AXGMAC_MTL_IOREAD_BITS(pdata, 0, MTL_Q_RQFCR, RFA);
+   fc.high_water[0] =  AXGMAC_MTL_IOREAD_BITS(pdata, 0, MTL_Q_RQFCR, RFD);
+   fc.pause_time[0] = AXGMAC_GET_BITS(reg_val, MAC_Q0TFCR, PT);
+   fc.autoneg = pdata->pause_autoneg;
+
+   if (pdata->rx_pause && pdata->tx_pause)
+   fc.mode = RTE_FC_FULL;
+   else if (pdata->rx_pause)
+   fc.mode = RTE_FC_RX_PAUSE;
+   else if (pdata->tx_pause)
+   fc.mode = RTE_FC_TX_PAUSE;
+   else
+   fc.mode = RTE_FC_NONE;
+
+   fc_conf->high_water =  (1024 + (fc.low_water[0] << 9)) / 1024;
+   fc_conf->low_water =  (1024 + (fc.high_water[0] << 9)) / 1024;
+   fc_conf->pause_time = fc.pause_time[0];
+   fc_conf->send_xon = fc.send_xon;
+   fc_conf->mode = fc.mode;
+
+   return 0;
+}
+
+static int
+axgbe_flow_ctrl_set(struct rte_eth_dev *dev, struct rte_eth_fc_conf *fc_conf)
+{
+   struct axgbe_port *pdata = dev->data->dev_private;
+   struct xgbe_fc_info fc = pdata->fc;
+   unsigned int reg, reg_val = 0;
+   reg = MAC_Q0TFCR;
+
+   pdata->pause_autoneg = fc_conf->autoneg;
+   pdata->phy.pause_autoneg = pdata->pause_autoneg;
+   fc.send_xon = fc_conf->send_xon;
+   AXGMAC_MTL_IOWRITE_BITS(pdata, 0, MTL_Q_RQFCR, RFA,
+   AXGMAC_FLOW_CONTROL_VALUE(1024 * fc_conf->high_water));
+   AXGMAC_MTL_IOWRITE_BITS(pdata, 0, MTL_Q_RQFCR, RFD,
+   AXGMAC_FLOW_CONTROL_VALUE(1024 * fc_conf->low_water));
+   AXGMAC_SET_BITS(reg_val, MAC_Q0TFCR, PT, fc_conf->pause_time);
+   AXGMAC_IOWRITE(pdata, reg, reg_val);
+   fc.mode = fc_conf->mode;
+
+   if (fc.mode == RTE_FC_FULL) {
+   pdata->tx_pause = 1;
+   pdata->rx_pause = 1;
+   } else if (fc.mode == RTE_FC_RX_PAUSE) {
+   pdata->tx_pause = 0;
+   pdata->rx_pause = 1;
+   } else if (fc.mode == RTE_FC_TX_PAUSE) {
+   pdata->tx_pause = 1;
+   pdata->rx_pause = 0;
+   } else {
+   pdata->tx_pause = 0;
+   pdata->rx_pause = 0;
+   }
+
+   if (pdata->tx_pause != (unsigned int)pdata->phy.tx_pause)
+   pdata->hw_if.config_tx_flow_control(pdata);
+
+   if (pdata->rx_pause != (unsigned int)pdata->phy.rx_pause)
+   pdata->hw_if.config_rx_flow_control(pdata);
+
+   pdata->hw_if.config_flow_control(pdata);
+   pdata->phy.tx_pause = pdata->tx_pause;
+   pdata->phy.rx_pause = pdata->rx_pause;
+
+   return 0;
+}
 
 static void axgbe_get_all_hw_features(struct axgbe_port *pdata)
 {
diff --git a/drivers/net/axgbe/axgbe_ethdev.h b/drivers/net/axgbe/axgbe_ethdev.h
index a1083b17b..746fb2f15 100644
--- a/drivers/net/axgbe/axgbe_ethdev.h
+++ b/drivers/net/axgbe/axgbe_ethdev.h
@@ -485,6 +485,15 @@ str