[dpdk-dev] [PATCH v2] net/sfc: add device state check to reta update operation

[Andrew Rybchenko]

From: Ivan Malov 

The callback must not attempt to program RSS table to the HW
in non-started state; the new table must be remembered and
applied on the next start

Fixes: 32bcfb0a50b1 ("net/sfc: update RSS redirection table")
Cc: sta...@dpdk.org

Signed-off-by: Ivan Malov 
Signed-off-by: Andrew Rybchenko 
Reviewed-by: Andy Moreton 
---
v2:
 - fix clang build failure because of possibly uninitialized rc

 drivers/net/sfc/sfc_ethdev.c | 15 ++-
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/drivers/net/sfc/sfc_ethdev.c b/drivers/net/sfc/sfc_ethdev.c
index 9e65b6a..7a57472 100644
--- a/drivers/net/sfc/sfc_ethdev.c
+++ b/drivers/net/sfc/sfc_ethdev.c
@@ -1354,7 +1354,7 @@ sfc_dev_rss_reta_update(struct rte_eth_dev *dev,
struct sfc_port *port = &sa->port;
unsigned int *rss_tbl_new;
uint16_t entry;
-   int rc;
+   int rc = 0;
 
 
if (port->isolated)
@@ -1399,11 +1399,16 @@ sfc_dev_rss_reta_update(struct rte_eth_dev *dev,
}
}
 
-   rc = efx_rx_scale_tbl_set(sa->nic, EFX_RSS_CONTEXT_DEFAULT,
- rss_tbl_new, EFX_RSS_TBL_SIZE);
-   if (rc == 0)
-   rte_memcpy(sa->rss_tbl, rss_tbl_new, sizeof(sa->rss_tbl));
+   if (sa->state == SFC_ADAPTER_STARTED) {
+   rc = efx_rx_scale_tbl_set(sa->nic, EFX_RSS_CONTEXT_DEFAULT,
+ rss_tbl_new, EFX_RSS_TBL_SIZE);
+   if (rc != 0)
+   goto fail_scale_tbl_set;
+   }
+
+   rte_memcpy(sa->rss_tbl, rss_tbl_new, sizeof(sa->rss_tbl));
 
+fail_scale_tbl_set:
 bad_reta_entry:
sfc_adapter_unlock(sa);
 
-- 
2.7.4

Re: [dpdk-dev] [PATCH v3 7/7] app/crypto-perf: use single mempool

[De Lara Guarch, Pablo]

> -Original Message-
> From: Akhil Goyal [mailto:akhil.go...@nxp.com]
> Sent: Tuesday, September 26, 2017 10:21 AM
> To: De Lara Guarch, Pablo ; Doherty,
> Declan 
> Cc: dev@dpdk.org
> Subject: Re: [PATCH v3 7/7] app/crypto-perf: use single mempool
> 
> On 9/22/2017 1:25 PM, Pablo de Lara wrote:
> > In order to improve memory utilization, a single mempool is created,
> > containing the crypto operation and mbufs (one if operation is
> > in-place, two if out-of-place).
> > This way, a single object is allocated and freed per operation,
> > reducing the amount of memory in cache, which improves scalability.
> >
> > Signed-off-by: Pablo de Lara 
> > ---
> >   app/test-crypto-perf/cperf_ops.c |  96 --
> >   app/test-crypto-perf/cperf_ops.h |   2 +-
> >   app/test-crypto-perf/cperf_test_latency.c| 361 +++
> ---
> >   app/test-crypto-perf/cperf_test_pmd_cyclecount.c | 364 +++-
> --
> >   app/test-crypto-perf/cperf_test_throughput.c | 358 +++-
> --
> >   app/test-crypto-perf/cperf_test_verify.c | 367 
> > +++--
> --
> >   6 files changed, 793 insertions(+), 755 deletions(-)
> >
> 
> The patch set looks good to me. Except for one comment in the 6th patch of
> the series and one comment as below.
> 
> Is it possible to move the common code at a single place for all the latency,
> cycle_count, throughput, verify cases.
> 
> I can see a lot of duplicate code in these files.

Good point. I will send a v4 with an extra patch that moves this common code to 
another file
(at the start of the patchset, so it is easier to review the rest of the 
changes).

Thanks,
Pablo

> 
> -Akhil

Re: [dpdk-dev] [PATCH v3 3/4] doc: add mrvl net pmd documentation

[Tomasz Duszynski]

On Wed, Oct 04, 2017 at 01:29:01AM +0100, Ferruh Yigit wrote:

+cc Shlomi.
> On 10/3/2017 12:51 PM, Tomasz Duszynski wrote:
> > Add documentation for the MRVL NET PMD driver.
> >
> > Signed-off-by: Jacek Siuda 
> > Signed-off-by: Tomasz Duszynski 
>
> <...>
>
> > +;
> > +[Features]
> > +Speed capabilities   = Y
>
> This doesn't seems correct please check doc/guides/nics/features.rst
Right, speed capabilities are not returned via rte_eth_dev_info. Thanks
for pointing this out. Will fix that in v4.
>
> > +Link status  = Y
> > +MTU update   = Y
> > +Jumbo frame  = Y
> > +Promiscuous mode = Y
> > +Allmulticast mode= Y
> > +Unicast MAC filter   = Y
> > +Multicast MAC filter = Y
> > +RSS hash = Y
> > +VLAN filter  = Y
> > +CRC offload  = Y
> > +L3 checksum offload  = Y
> > +L4 checksum offload  = Y
> > +Packet type parsing  = Y
> > +Basic stats  = Y
> > +Stats per queue  = Y
> > +ARMv8= Y
>
> Is other architecture not supported or not tested?
Currently that NIC comes integrated into Armada 7k/8k SoCs and they are based
on ARMv8.
>
> > +Usage doc= Y
>
> <...>
>
> > +Prerequisites
> > +-
> > +
> > +- MUSDK (Marvell User-Space SDK) sources available
> > +  `here 
> > `_.
>
> Is this 17.08 by change related to DPDK version, I mean is there any
> relation between DPDK version and musdk library version?
No relation except similar versioning convention that is used. Musdk
library version needed by DPDK driver is mentioned explicitly in
documentation.
>
> > +
> > +MUSDK is a light-weight library that provides direct access to 
> > Marvell's
> > +PPv2 (Packet Processor v2). Alternatively prebuilt MUSDK library can be
> > +requested from `Marvell Extranet `_. Once
> > +approval has been granted, library can be found by typing ``musdk`` in
> > +search box.
>
> What is the condition of having an approval?
I think Marvell team would be more helpful here.
>
> > +
> > +- DPDK environment
> > +
> > +Follow the DPDK :ref:`Getting Started Guide for Linux ` to 
> > setup
> > +DPDK environment.
> > +
> > +
>
> <...>
>
> > +Building DPDK
> > +-
> > +
> > +Driver needs precompiled MUSDK library during compilation. Detailed build
> > +process is described in library's documentation under ``doc`` directory.
> > +
> > +Before the DPDK build process the environmental variable ``LIBMUSDK_PATH`` 
> > with
> > +the path to the MUSDK installation directory needs to be exported.
> > +
> > +Usage Example
> > +-
> > +
> > +MRVL PMD requires extra kernel modules to function properly:
> > +
> > +.. code-block:: console
> > +
> > +   insmod musdk_uio.ko
> > +   insmod mv_pp_uio.ko
> > +   insmod mvpp2x_sysfs.ko
>
> Why these kernel modules are required and how one can obtain them?
Kernel modules are used to map device memory regions to userspace.
musdk_uio and mv_pp_uio both come with MUSDK library. As for mvpp2x_sysfs
I think Marvell team could be more helpful here.
> These are out of tree kernel modules right?
Right. You have to either build them yourself or get precompiled from
Marvell-extranet.
>
> <...>
>

--
- Tomasz Duszyński

Re: [dpdk-dev] [PATCH v5 2/3] ethdev: introduce Tx queue offloads API

[Shahaf Shuler]

Tuesday, October 3, 2017 10:50 PM, Ferruh Yigit:
> <...>
> 
> > @@ -620,6 +628,15 @@ Supports packet type parsing and returns a list of
> supported types.
> >
> >  .. _nic_features_timesync:
> >
> > +Mbuf fast free
> > +--
> 
> I think this is not one of the current tracked features. Is this coming/planed
> with new patches?

This is not a new feature, rather re-wording and merging of the flags:
ETH_TXQ_FLAGS_NOREFCOUNT
ETH_TXQ_FLAGS_NOMULTMEMP

> 
> I suggest removing from this patch, and if required add with another patch
> that both updates default.ini and this documented.

I agree it makes more sense to have this "feature" on a different patch. 

> 
> > +
> > +Supports optimization for fast release of mbufs following successful Tx.
> > +Requires all mbufs to come from the same mempool and has refcnt = 1.
> > +
> > +* **[uses]   rte_eth_txconf,rte_eth_txmode**:
> ``offloads:DEV_TX_OFFLOAD_MBUF_FAST_FREE``.
> > +* **[provides]   rte_eth_dev_info**:
> ``tx_offload_capa,tx_queue_offload_capa:DEV_TX_OFFLOAD_MBUF_FAST_
> FREE``.
> > +
> >  Timesync
> >  
> >
> > diff --git a/lib/librte_ether/rte_ethdev.c
> > b/lib/librte_ether/rte_ethdev.c index 9b73d2377..59756dd82 100644
> > --- a/lib/librte_ether/rte_ethdev.c
> > +++ b/lib/librte_ether/rte_ethdev.c
> 
> <...>

Re: [dpdk-dev] [PATCH 1/2] net/tap: fix indentation in flow file

[Pascal Mazon]

Hi,

I'm surprised there's only one place in the file with indent problem,
but I'm ok with the patch otherwise.

Ferruh, I didn't get what you mean; Matan is modifying tap_flow.c in
patch 2/2, right?

Best regards,
Pascal

On 04/10/2017 00:07, Ferruh Yigit wrote:
> On 10/3/2017 3:55 PM, Matan Azrad wrote:
>> Missed 4 spaces were added after break if line.
>>
>> Fixes: 7a6811d78ed6 ("net/tap: fix flow and port commands")
> What do you think fixing this when that code has been touched next time?
>
>> Signed-off-by: Matan Azrad 
>> ---
>>  drivers/net/tap/tap_flow.c | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/net/tap/tap_flow.c b/drivers/net/tap/tap_flow.c
>> index eefa868..aa33960 100644
>> --- a/drivers/net/tap/tap_flow.c
>> +++ b/drivers/net/tap/tap_flow.c
>> @@ -1093,7 +1093,7 @@ struct tap_flow_items {
>>  goto exit_action_not_supported;
>>  action = 1;
>>  if (!queue ||
>> -(queue->index > pmd->dev->data->nb_rx_queues - 1))
>> +(queue->index > pmd->dev->data->nb_rx_queues - 1))
>>  goto exit_action_not_supported;
>>  if (flow)
>>  err = add_action_skbedit(flow, queue->index);
>>

[dpdk-dev] [PATCH v6 0/4] ethdev new offloads API

[Shahaf Shuler]

Tx offloads configuration is per queue. Tx offloads are enabled by default, 
and can be disabled using ETH_TXQ_FLAGS_NO* flags. 
This behaviour is not consistent with the Rx side where the Rx offloads
configuration is per port. Rx offloads are disabled by default and enabled 
according to bit field in rte_eth_rxmode structure.

Moreover, considering more Tx and Rx offloads will be added 
over time, the cost of managing them all inside the PMD will be tremendous,
as the PMD will need to check the matching for the entire offload set 
for each mbuf it handles.
In addition, on the current approach each Rx offload added breaks the
ABI compatibility as it requires to add entries to existing bit-fields.
 
The series address above issues by defining a new offloads API.
In the new API, offloads are divided into per-port and per-queue offloads,
with a corresponding capability for each.
The offloads are disabled by default. Each offload can be enabled or
disabled using the existing DEV_TX_OFFLOADS_* or DEV_RX_OFFLOADS_* flags.
Such API will enable to easily add or remove offloads, without breaking the
ABI compatibility.

In order to provide a smooth transition between the APIs the following actions
were taken:
*  The old offloads API is kept for the meanwhile.
*  Helper function which copy from old to new API were added to ethdev,
   enabling the PMD to support only one of the APIs.
*  Helper function which copy from new to old API were also added,
   to enable application to use the new API with PMD which still supports
   the old one.

Per discussion made on the RFC of this series [1], the integration plan which 
was
decided is to do the transition in two phases:
* ethdev API will move on 17.11.
* Apps and examples will move on 18.02.

This to enable PMD maintainers sufficient time to adopt the new API.

[1]
http://dpdk.org/ml/archives/dev/2017-August/072643.html

on v6:
 - Move mbuf fast free Tx offload to a seperate patch.

on v5:
 - Fix documentation.
 - Fix comments on port offloads configuration.

on v4:
 - Added another patch for documentation.
 - Fixed ETH_TXQ_FLAGS_IGNORE flag override.
 - Clarify the description of DEV_TX_OFFLOAD_MBUF_FAST_FREE offload.

on v3:
 - Introduce the DEV_TX_OFFLOAD_MBUF_FAST_FREE to act as an equivalent
   for the no refcnt and single mempool flags.
 - Fix features documentation.
 - Fix comment style.

on v2:
 - Taking new approach of dividing offloads into per-queue and per-port one.
 - Postpone the Tx/Rx public struct renaming to 18.02
 - Squash the helper functions into the Rx/Tx offloads intro patches.

Shahaf Shuler (4):
  ethdev: introduce Rx queue offloads API
  ethdev: introduce Tx queue offloads API
  ethdev: add mbuf fast free Tx offload
  doc: add details on ethdev offloads API

 doc/guides/nics/features.rst|  66 +---
 doc/guides/nics/features/default.ini|   1 +
 doc/guides/prog_guide/poll_mode_drv.rst |  20 +++
 lib/librte_ether/rte_ethdev.c   | 223 +--
 lib/librte_ether/rte_ethdev.h   |  89 ++-
 5 files changed, 359 insertions(+), 40 deletions(-)

Series-reviewed-by: Andrew Rybchenko 

-- 
2.12.0

[dpdk-dev] [PATCH v6 1/4] ethdev: introduce Rx queue offloads API

[Shahaf Shuler]

Introduce a new API to configure Rx offloads.

In the new API, offloads are divided into per-port and per-queue
offloads. The PMD reports capability for each of them.
Offloads are enabled using the existing DEV_RX_OFFLOAD_* flags.
To enable per-port offload, the offload should be set on both device
configuration and queue configuration. To enable per-queue offload, the
offloads can be set only on queue configuration.

Applications should set the ignore_offload_bitfield bit on rxmode
structure in order to move to the new API.

The old Rx offloads API is kept for the meanwhile, in order to enable a
smooth transition for PMDs and application to the new API.

Signed-off-by: Shahaf Shuler 
---
 doc/guides/nics/features.rst  |  33 
 lib/librte_ether/rte_ethdev.c | 156 +
 lib/librte_ether/rte_ethdev.h |  51 +++-
 3 files changed, 210 insertions(+), 30 deletions(-)

diff --git a/doc/guides/nics/features.rst b/doc/guides/nics/features.rst
index 37ffbc68c..4e68144ef 100644
--- a/doc/guides/nics/features.rst
+++ b/doc/guides/nics/features.rst
@@ -179,7 +179,7 @@ Jumbo frame
 
 Supports Rx jumbo frames.
 
-* **[uses]user config**: ``dev_conf.rxmode.jumbo_frame``,
+* **[uses]rte_eth_rxconf,rte_eth_rxmode**: 
``offloads:DEV_RX_OFFLOAD_JUMBO_FRAME``.
   ``dev_conf.rxmode.max_rx_pkt_len``.
 * **[related] rte_eth_dev_info**: ``max_rx_pktlen``.
 * **[related] API**: ``rte_eth_dev_set_mtu()``.
@@ -192,7 +192,7 @@ Scattered Rx
 
 Supports receiving segmented mbufs.
 
-* **[uses]   user config**: ``dev_conf.rxmode.enable_scatter``.
+* **[uses]   rte_eth_rxconf,rte_eth_rxmode**: 
``offloads:DEV_RX_OFFLOAD_SCATTER``.
 * **[implements] datapath**: ``Scattered Rx function``.
 * **[implements] rte_eth_dev_data**: ``scattered_rx``.
 * **[provides]   eth_dev_ops**: ``rxq_info_get:scattered_rx``.
@@ -206,11 +206,11 @@ LRO
 
 Supports Large Receive Offload.
 
-* **[uses]   user config**: ``dev_conf.rxmode.enable_lro``.
+* **[uses]   rte_eth_rxconf,rte_eth_rxmode**: 
``offloads:DEV_RX_OFFLOAD_TCP_LRO``.
 * **[implements] datapath**: ``LRO functionality``.
 * **[implements] rte_eth_dev_data**: ``lro``.
 * **[provides]   mbuf**: ``mbuf.ol_flags:PKT_RX_LRO``, ``mbuf.tso_segsz``.
-* **[provides]   rte_eth_dev_info**: 
``rx_offload_capa:DEV_RX_OFFLOAD_TCP_LRO``.
+* **[provides]   rte_eth_dev_info**: 
``rx_offload_capa,rx_queue_offload_capa:DEV_RX_OFFLOAD_TCP_LRO``.
 
 
 .. _nic_features_tso:
@@ -363,7 +363,7 @@ VLAN filter
 
 Supports filtering of a VLAN Tag identifier.
 
-* **[uses]   user config**: ``dev_conf.rxmode.hw_vlan_filter``.
+* **[uses]   rte_eth_rxconf,rte_eth_rxmode**: 
``offloads:DEV_RX_OFFLOAD_VLAN_FILTER``.
 * **[implements] eth_dev_ops**: ``vlan_filter_set``.
 * **[related]API**: ``rte_eth_dev_vlan_filter()``.
 
@@ -499,7 +499,7 @@ CRC offload
 
 Supports CRC stripping by hardware.
 
-* **[uses] user config**: ``dev_conf.rxmode.hw_strip_crc``.
+* **[uses] rte_eth_rxconf,rte_eth_rxmode**: 
``offloads:DEV_RX_OFFLOAD_CRC_STRIP``.
 
 
 .. _nic_features_vlan_offload:
@@ -509,11 +509,10 @@ VLAN offload
 
 Supports VLAN offload to hardware.
 
-* **[uses]   user config**: ``dev_conf.rxmode.hw_vlan_strip``,
-  ``dev_conf.rxmode.hw_vlan_filter``, ``dev_conf.rxmode.hw_vlan_extend``.
+* **[uses]   rte_eth_rxconf,rte_eth_rxmode**: 
``offloads:DEV_RX_OFFLOAD_VLAN_STRIP,DEV_RX_OFFLOAD_VLAN_FILTER,DEV_RX_OFFLOAD_VLAN_EXTEND``.
 * **[implements] eth_dev_ops**: ``vlan_offload_set``.
 * **[provides]   mbuf**: ``mbuf.ol_flags:PKT_RX_VLAN_STRIPPED``, 
``mbuf.vlan_tci``.
-* **[provides]   rte_eth_dev_info**: 
``rx_offload_capa:DEV_RX_OFFLOAD_VLAN_STRIP``,
+* **[provides]   rte_eth_dev_info**: 
``rx_offload_capa,rx_queue_offload_capa:DEV_RX_OFFLOAD_VLAN_STRIP``,
   ``tx_offload_capa:DEV_TX_OFFLOAD_VLAN_INSERT``.
 * **[related]API**: ``rte_eth_dev_set_vlan_offload()``,
   ``rte_eth_dev_get_vlan_offload()``.
@@ -526,10 +525,11 @@ QinQ offload
 
 Supports QinQ (queue in queue) offload.
 
+* **[uses] rte_eth_rxconf,rte_eth_rxmode**: 
``offloads:DEV_RX_OFFLOAD_QINQ_STRIP``.
 * **[uses] mbuf**: ``mbuf.ol_flags:PKT_TX_QINQ_PKT``.
 * **[provides] mbuf**: ``mbuf.ol_flags:PKT_RX_QINQ_STRIPPED``, 
``mbuf.vlan_tci``,
``mbuf.vlan_tci_outer``.
-* **[provides] rte_eth_dev_info**: 
``rx_offload_capa:DEV_RX_OFFLOAD_QINQ_STRIP``,
+* **[provides] rte_eth_dev_info**: 
``rx_offload_capa,rx_queue_offload_capa:DEV_RX_OFFLOAD_QINQ_STRIP``,
   ``tx_offload_capa:DEV_TX_OFFLOAD_QINQ_INSERT``.
 
 
@@ -540,13 +540,13 @@ L3 checksum offload
 
 Supports L3 checksum offload.
 
-* **[uses] user config**: ``dev_conf.rxmode.hw_ip_checksum``.
+* **[uses] rte_eth_rxconf,rte_eth_rxmode**: 
``offloads:DEV_RX_OFFLOAD_IPV4_CKSUM``.
 * **[uses] mbuf**: ``mbuf.ol_flags:PKT_TX_IP_CKSUM``,
   ``mbuf.ol_flags:PKT_TX_IPV4`` | ``PKT_TX_IPV6``.
 * **[provides] mbuf**: ``mbuf.ol_flags:PKT_RX_IP_CKSUM_UNKNOWN`` |
   ``PKT_RX_IP_CKSUM_BAD`` | ``PKT_RX_IP_CKSUM_GOOD`` |
   ``PK

[dpdk-dev] [PATCH v6 2/4] ethdev: introduce Tx queue offloads API

[Shahaf Shuler]

Introduce a new API to configure Tx offloads.

In the new API, offloads are divided into per-port and per-queue
offloads. The PMD reports capability for each of them.
Offloads are enabled using the existing DEV_TX_OFFLOAD_* flags.
To enable per-port offload, the offload should be set on both device
configuration and queue configuration. To enable per-queue offload, the
offloads can be set only on queue configuration.

In addition the Tx offloads will be disabled by default and be
enabled per application needs. This will much simplify PMD management of
the different offloads.

Applications should set the ETH_TXQ_FLAGS_IGNORE flag on txq_flags
field in order to move to the new API.

The old Tx offloads API is kept for the meanwhile, in order to enable a
smooth transition for PMDs and application to the new API.

Signed-off-by: Shahaf Shuler 
---
 doc/guides/nics/features.rst  | 24 ++-
 lib/librte_ether/rte_ethdev.c | 62 +-
 lib/librte_ether/rte_ethdev.h | 33 +++-
 3 files changed, 109 insertions(+), 10 deletions(-)

diff --git a/doc/guides/nics/features.rst b/doc/guides/nics/features.rst
index 4e68144ef..17745dace 100644
--- a/doc/guides/nics/features.rst
+++ b/doc/guides/nics/features.rst
@@ -131,7 +131,8 @@ Lock-free Tx queue
 If a PMD advertises DEV_TX_OFFLOAD_MT_LOCKFREE capable, multiple threads can
 invoke rte_eth_tx_burst() concurrently on the same Tx queue without SW lock.
 
-* **[provides] rte_eth_dev_info**: 
``tx_offload_capa:DEV_TX_OFFLOAD_MT_LOCKFREE``.
+* **[uses]rte_eth_txconf,rte_eth_txmode**: 
``offloads:DEV_TX_OFFLOAD_MT_LOCKFREE``.
+* **[provides] rte_eth_dev_info**: 
``tx_offload_capa,tx_queue_offload_capa:DEV_TX_OFFLOAD_MT_LOCKFREE``.
 * **[related]  API**: ``rte_eth_tx_burst()``.
 
 
@@ -220,11 +221,12 @@ TSO
 
 Supports TCP Segmentation Offloading.
 
+* **[uses]   rte_eth_txconf,rte_eth_txmode**: 
``offloads:DEV_TX_OFFLOAD_TCP_TSO``.
 * **[uses]   rte_eth_desc_lim**: ``nb_seg_max``, ``nb_mtu_seg_max``.
 * **[uses]   mbuf**: ``mbuf.ol_flags:PKT_TX_TCP_SEG``.
 * **[uses]   mbuf**: ``mbuf.tso_segsz``, ``mbuf.l2_len``, ``mbuf.l3_len``, 
``mbuf.l4_len``.
 * **[implements] datapath**: ``TSO functionality``.
-* **[provides]   rte_eth_dev_info**: 
``tx_offload_capa:DEV_TX_OFFLOAD_TCP_TSO,DEV_TX_OFFLOAD_UDP_TSO``.
+* **[provides]   rte_eth_dev_info**: 
``tx_offload_capa,tx_queue_offload_capa:DEV_TX_OFFLOAD_TCP_TSO,DEV_TX_OFFLOAD_UDP_TSO``.
 
 
 .. _nic_features_promiscuous_mode:
@@ -510,10 +512,11 @@ VLAN offload
 Supports VLAN offload to hardware.
 
 * **[uses]   rte_eth_rxconf,rte_eth_rxmode**: 
``offloads:DEV_RX_OFFLOAD_VLAN_STRIP,DEV_RX_OFFLOAD_VLAN_FILTER,DEV_RX_OFFLOAD_VLAN_EXTEND``.
+* **[uses]   rte_eth_txconf,rte_eth_txmode**: 
``offloads:DEV_TX_OFFLOAD_VLAN_INSERT``.
 * **[implements] eth_dev_ops**: ``vlan_offload_set``.
 * **[provides]   mbuf**: ``mbuf.ol_flags:PKT_RX_VLAN_STRIPPED``, 
``mbuf.vlan_tci``.
 * **[provides]   rte_eth_dev_info**: 
``rx_offload_capa,rx_queue_offload_capa:DEV_RX_OFFLOAD_VLAN_STRIP``,
-  ``tx_offload_capa:DEV_TX_OFFLOAD_VLAN_INSERT``.
+  ``tx_offload_capa,tx_queue_offload_capa:DEV_TX_OFFLOAD_VLAN_INSERT``.
 * **[related]API**: ``rte_eth_dev_set_vlan_offload()``,
   ``rte_eth_dev_get_vlan_offload()``.
 
@@ -526,11 +529,12 @@ QinQ offload
 Supports QinQ (queue in queue) offload.
 
 * **[uses] rte_eth_rxconf,rte_eth_rxmode**: 
``offloads:DEV_RX_OFFLOAD_QINQ_STRIP``.
+* **[uses] rte_eth_txconf,rte_eth_txmode**: 
``offloads:DEV_TX_OFFLOAD_QINQ_INSERT``.
 * **[uses] mbuf**: ``mbuf.ol_flags:PKT_TX_QINQ_PKT``.
 * **[provides] mbuf**: ``mbuf.ol_flags:PKT_RX_QINQ_STRIPPED``, 
``mbuf.vlan_tci``,
``mbuf.vlan_tci_outer``.
 * **[provides] rte_eth_dev_info**: 
``rx_offload_capa,rx_queue_offload_capa:DEV_RX_OFFLOAD_QINQ_STRIP``,
-  ``tx_offload_capa:DEV_TX_OFFLOAD_QINQ_INSERT``.
+  ``tx_offload_capa,tx_queue_offload_capa:DEV_TX_OFFLOAD_QINQ_INSERT``.
 
 
 .. _nic_features_l3_checksum_offload:
@@ -541,13 +545,14 @@ L3 checksum offload
 Supports L3 checksum offload.
 
 * **[uses] rte_eth_rxconf,rte_eth_rxmode**: 
``offloads:DEV_RX_OFFLOAD_IPV4_CKSUM``.
+* **[uses] rte_eth_txconf,rte_eth_txmode**: 
``offloads:DEV_TX_OFFLOAD_IPV4_CKSUM``.
 * **[uses] mbuf**: ``mbuf.ol_flags:PKT_TX_IP_CKSUM``,
   ``mbuf.ol_flags:PKT_TX_IPV4`` | ``PKT_TX_IPV6``.
 * **[provides] mbuf**: ``mbuf.ol_flags:PKT_RX_IP_CKSUM_UNKNOWN`` |
   ``PKT_RX_IP_CKSUM_BAD`` | ``PKT_RX_IP_CKSUM_GOOD`` |
   ``PKT_RX_IP_CKSUM_NONE``.
 * **[provides] rte_eth_dev_info**: 
``rx_offload_capa,rx_queue_offload_capa:DEV_RX_OFFLOAD_IPV4_CKSUM``,
-  ``tx_offload_capa:DEV_TX_OFFLOAD_IPV4_CKSUM``.
+  ``tx_offload_capa,tx_queue_offload_capa:DEV_TX_OFFLOAD_IPV4_CKSUM``.
 
 
 .. _nic_features_l4_checksum_offload:
@@ -558,6 +563,7 @@ L4 checksum offload
 Supports L4 checksum offload.
 
 * **[uses] rte_eth_rxconf,rte_eth_rxmode**: 
``offloads:DEV_RX_OFFLOAD_UDP_CKSUM,DEV_RX_OFFLOAD_TCP_CKSUM``.
+* **[uses]  

[dpdk-dev] [PATCH v6 4/4] doc: add details on ethdev offloads API

[Shahaf Shuler]

Add the programmers guide details on the new offloads API introduced
by commits:

commit 67a1a59b597f ("ethdev: introduce Rx queue offloads API")
commit f883eb32e2d4 ("ethdev: introduce Tx queue offloads API")

Signed-off-by: Shahaf Shuler 
Reviewed-by: John McNamara 
---
 doc/guides/prog_guide/poll_mode_drv.rst | 20 
 1 file changed, 20 insertions(+)

diff --git a/doc/guides/prog_guide/poll_mode_drv.rst 
b/doc/guides/prog_guide/poll_mode_drv.rst
index 8922e39f4..423170997 100644
--- a/doc/guides/prog_guide/poll_mode_drv.rst
+++ b/doc/guides/prog_guide/poll_mode_drv.rst
@@ -310,6 +310,26 @@ exported by each PMD. The list of flags and their precise 
meaning is
 described in the mbuf API documentation and in the in :ref:`Mbuf Library
 `, section "Meta Information".
 
+Per-Port and Per-Queue Offloads
+^^^
+
+In the DPDK offload API, offloads are divided into per-port and per-queue 
offloads.
+The different offloads capabilities can be queried using 
``rte_eth_dev_info_get()``.
+Supported offloads can be either per-port or per-queue.
+
+Offloads are enabled using the existing ``DEV_TX_OFFLOAD_*`` or 
``DEV_RX_OFFLOAD_*`` flags.
+Per-port offload configuration is set using ``rte_eth_dev_configure``.
+Per-queue offload configuration is set using ``rte_eth_rx_queue_setup`` and 
``rte_eth_tx_queue_setup``.
+To enable per-port offload, the offload should be set on both device 
configuration and queue setup.
+In case of a mixed configuration the queue setup shall return with an error.
+To enable per-queue offload, the offload can be set only on the queue setup.
+Offloads which are not enabled are disabled by default.
+
+For an application to use the Tx offloads API it should set the 
``ETH_TXQ_FLAGS_IGNORE`` flag in the ``txq_flags`` field located in 
``rte_eth_txconf`` struct.
+In such cases it is not required to set other flags in ``txq_flags``.
+For an application to use the Rx offloads API it should set the 
``ignore_offload_bitfield`` bit in the ``rte_eth_rxmode`` struct.
+In such cases it is not required to set other bitfield offloads in the 
``rxmode`` struct.
+
 Poll Mode Driver API
 
 
-- 
2.12.0

[dpdk-dev] [PATCH v6 3/4] ethdev: add mbuf fast free Tx offload

[Shahaf Shuler]

PMDs which expose this offload cap supports optimization for fast release
of mbufs following successful Tx.
Such optimization requires that per queue, all mbufs come from the same
mempool and has refcnt = 1.

Signed-off-by: Shahaf Shuler 
---
 doc/guides/nics/features.rst | 9 +
 doc/guides/nics/features/default.ini | 1 +
 lib/librte_ether/rte_ethdev.c| 5 +
 lib/librte_ether/rte_ethdev.h| 5 +
 4 files changed, 20 insertions(+)

diff --git a/doc/guides/nics/features.rst b/doc/guides/nics/features.rst
index 17745dace..6538470ac 100644
--- a/doc/guides/nics/features.rst
+++ b/doc/guides/nics/features.rst
@@ -628,6 +628,15 @@ Supports packet type parsing and returns a list of 
supported types.
 
 .. _nic_features_timesync:
 
+Mbuf fast free
+--
+
+Supports optimization for fast release of mbufs following successful Tx.
+Requires that per queue, all mbufs come from the same mempool and has refcnt = 
1.
+
+* **[uses]   rte_eth_txconf,rte_eth_txmode**: 
``offloads:DEV_TX_OFFLOAD_MBUF_FAST_FREE``.
+* **[provides]   rte_eth_dev_info**: 
``tx_offload_capa,tx_queue_offload_capa:DEV_TX_OFFLOAD_MBUF_FAST_FREE``.
+
 Timesync
 
 
diff --git a/doc/guides/nics/features/default.ini 
b/doc/guides/nics/features/default.ini
index 542430696..9a5990195 100644
--- a/doc/guides/nics/features/default.ini
+++ b/doc/guides/nics/features/default.ini
@@ -75,3 +75,4 @@ x86-64   =
 Usage doc=
 Design doc   =
 Perf doc =
+Mbuf fast free   =
diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
index 856a54a8e..59756dd82 100644
--- a/lib/librte_ether/rte_ethdev.c
+++ b/lib/librte_ether/rte_ethdev.c
@@ -1232,6 +1232,9 @@ rte_eth_convert_txq_flags(const uint32_t txq_flags, 
uint64_t *tx_offloads)
offloads |= DEV_TX_OFFLOAD_UDP_CKSUM;
if (!(txq_flags & ETH_TXQ_FLAGS_NOXSUMTCP))
offloads |= DEV_TX_OFFLOAD_TCP_CKSUM;
+   if ((txq_flags & ETH_TXQ_FLAGS_NOREFCOUNT) &&
+   (txq_flags & ETH_TXQ_FLAGS_NOMULTMEMP))
+   offloads |= DEV_TX_OFFLOAD_MBUF_FAST_FREE;
 
*tx_offloads = offloads;
 }
@@ -1254,6 +1257,8 @@ rte_eth_convert_txq_offloads(const uint64_t tx_offloads, 
uint32_t *txq_flags)
flags |= ETH_TXQ_FLAGS_NOXSUMUDP;
if (!(tx_offloads & DEV_TX_OFFLOAD_TCP_CKSUM))
flags |= ETH_TXQ_FLAGS_NOXSUMTCP;
+   if (tx_offloads & DEV_TX_OFFLOAD_MBUF_FAST_FREE)
+   flags |= (ETH_TXQ_FLAGS_NOREFCOUNT | ETH_TXQ_FLAGS_NOMULTMEMP);
 
*txq_flags = flags;
 }
diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
index da91f8740..78de045ed 100644
--- a/lib/librte_ether/rte_ethdev.h
+++ b/lib/librte_ether/rte_ethdev.h
@@ -991,6 +991,11 @@ struct rte_eth_conf {
  */
 #define DEV_TX_OFFLOAD_MULTI_SEGS  0x8000
 /**< Device supports multi segment send. */
+#define DEV_TX_OFFLOAD_MBUF_FAST_FREE  0x0001
+/**< Device supports optimization for fast release of mbufs.
+ *   When set application must guarantee that per-queue all mbufs comes from
+ *   the same mempool and has refcnt = 1.
+ */
 
 struct rte_pci_device;
 
-- 
2.12.0

Re: [dpdk-dev] [PATCH 2/2] net/tap: allow RSS flow action

[Pascal Mazon]

Hi,

What you say stands to reason, I'm basically ok with your patch, except
for one remark.
The rest of the file has no line breaks anywhere, please remove the two
you added.

Best regards,
Pascal

On 03/10/2017 16:55, Matan Azrad wrote:
> One of the main identified use cases for the tap PMD is to be used in
> combination with the fail-safe PMD as a fallback for a physical device.
>
> Fail-safe is very strict about making sure its current configuration is
> properly applied to all slave devices, they get rejected otherwise in
> order to maintain a consistent state.
>
> The problem is that tap's RSS support is currently limited to the
> default (non-Toeplitz) balancing performed by the kernel on all Rx
> queues. While proper RSS support emulation in the tap PMD is a work in
> progress, the lack of rte_flow counterpart prevents validation of the
> above use case in the meantime.
>
> Given that unlike most PMDs, tap is more about convenience than
> performance, support for the RSS action can be temporarily faked with
> a minimum amount of code and mostly correct behavior by treating it
> like a QUEUE action. Traffic is directed to the first queue of the set.
>
> Signed-off-by: Matan Azrad 
> ---
>  drivers/net/tap/tap_flow.c | 15 +++
>  1 file changed, 15 insertions(+)
>
> diff --git a/drivers/net/tap/tap_flow.c b/drivers/net/tap/tap_flow.c
> index aa33960..28d793f 100644
> --- a/drivers/net/tap/tap_flow.c
> +++ b/drivers/net/tap/tap_flow.c
> @@ -1089,6 +1089,7 @@ struct tap_flow_items {
>   const struct rte_flow_action_queue *queue =
>   (const struct rte_flow_action_queue *)
>   actions->conf;
> +
>   if (action)
>   goto exit_action_not_supported;
>   action = 1;
> @@ -1097,6 +1098,20 @@ struct tap_flow_items {
>   goto exit_action_not_supported;
>   if (flow)
>   err = add_action_skbedit(flow, queue->index);
> + } else if (actions->type == RTE_FLOW_ACTION_TYPE_RSS) {
> + /* Fake RSS support. */
> + const struct rte_flow_action_rss *rss =
> + (const struct rte_flow_action_rss *)
> + actions->conf;
> +
> + if (action)
> + goto exit_action_not_supported;
> + action = 1;
> + if (!rss || rss->num < 1 ||
> + (rss->queue[0] > pmd->dev->data->nb_rx_queues - 1))
> + goto exit_action_not_supported;
> + if (flow)
> + err = add_action_skbedit(flow, rss->queue[0]);
>   } else {
>   goto exit_action_not_supported;
>   }

[dpdk-dev] [PATCH] net/mlx5: fix checksum offload cap

[Shahaf Shuler]

The PMD was using the wrong flag to query device checksum capabilities

Fixes: bba710e6b99b ("net/mlx5: support upstream rdma-core")
Cc: shacha...@mellanox.com

Signed-off-by: Shahaf Shuler 
---
 drivers/net/mlx5/mlx5.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/mlx5/mlx5.c b/drivers/net/mlx5/mlx5.c
index 229b824bc..a4b718417 100644
--- a/drivers/net/mlx5/mlx5.c
+++ b/drivers/net/mlx5/mlx5.c
@@ -649,8 +649,8 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct 
rte_pci_device *pci_dev)
}
 
priv->hw_csum =
-   ((device_attr_ex.device_cap_flags_ex &
- IBV_DEVICE_UD_IP_CSUM));
+   !!(device_attr_ex.device_cap_flags_ex &
+  IBV_DEVICE_RAW_IP_CSUM);
DEBUG("checksum offloading is %ssupported",
  (priv->hw_csum ? "" : "not "));
 
-- 
2.12.0

Re: [dpdk-dev] [PATCH 2/2] net/tap: allow RSS flow action

[Pascal Mazon]

Hi,

I was a little too quick earlier. The line breaks are justified because
they follow the list of variable declarations.

So I ack the whole patch series.

Acked-by: Pascal Mazon 

On 04/10/2017 10:19, Pascal Mazon wrote:
> Hi,
>
> What you say stands to reason, I'm basically ok with your patch, except
> for one remark.
> The rest of the file has no line breaks anywhere, please remove the two
> you added.
>
> Best regards,
> Pascal
>
> On 03/10/2017 16:55, Matan Azrad wrote:
>> One of the main identified use cases for the tap PMD is to be used in
>> combination with the fail-safe PMD as a fallback for a physical device.
>>
>> Fail-safe is very strict about making sure its current configuration is
>> properly applied to all slave devices, they get rejected otherwise in
>> order to maintain a consistent state.
>>
>> The problem is that tap's RSS support is currently limited to the
>> default (non-Toeplitz) balancing performed by the kernel on all Rx
>> queues. While proper RSS support emulation in the tap PMD is a work in
>> progress, the lack of rte_flow counterpart prevents validation of the
>> above use case in the meantime.
>>
>> Given that unlike most PMDs, tap is more about convenience than
>> performance, support for the RSS action can be temporarily faked with
>> a minimum amount of code and mostly correct behavior by treating it
>> like a QUEUE action. Traffic is directed to the first queue of the set.
>>
>> Signed-off-by: Matan Azrad 
>> ---
>>  drivers/net/tap/tap_flow.c | 15 +++
>>  1 file changed, 15 insertions(+)
>>
>> diff --git a/drivers/net/tap/tap_flow.c b/drivers/net/tap/tap_flow.c
>> index aa33960..28d793f 100644
>> --- a/drivers/net/tap/tap_flow.c
>> +++ b/drivers/net/tap/tap_flow.c
>> @@ -1089,6 +1089,7 @@ struct tap_flow_items {
>>  const struct rte_flow_action_queue *queue =
>>  (const struct rte_flow_action_queue *)
>>  actions->conf;
>> +
>>  if (action)
>>  goto exit_action_not_supported;
>>  action = 1;
>> @@ -1097,6 +1098,20 @@ struct tap_flow_items {
>>  goto exit_action_not_supported;
>>  if (flow)
>>  err = add_action_skbedit(flow, queue->index);
>> +} else if (actions->type == RTE_FLOW_ACTION_TYPE_RSS) {
>> +/* Fake RSS support. */
>> +const struct rte_flow_action_rss *rss =
>> +(const struct rte_flow_action_rss *)
>> +actions->conf;
>> +
>> +if (action)
>> +goto exit_action_not_supported;
>> +action = 1;
>> +if (!rss || rss->num < 1 ||
>> +(rss->queue[0] > pmd->dev->data->nb_rx_queues - 1))
>> +goto exit_action_not_supported;
>> +if (flow)
>> +err = add_action_skbedit(flow, rss->queue[0]);
>>  } else {
>>  goto exit_action_not_supported;
>>  }

Re: [dpdk-dev] [PATCH v3 2/4] net/mrvl: add mrvl net pmd driver

[Tomasz Duszynski]

On Wed, Oct 04, 2017 at 01:24:27AM +0100, Ferruh Yigit wrote:
> On 10/3/2017 12:51 PM, Tomasz Duszynski wrote:
> > Add support for the Marvell PPv2 (Packet Processor v2) 1/10 Gbps adapter.
> > Driver is based on external, publicly available, light-weight Marvell
> > MUSDK library that provides access to network packet processor.
> >
> > Driver comes with support for the following features:
> >
> > * Speed capabilities
> > * Link status
> > * Queue start/stop
> > * MTU update
> > * Jumbo frame
> > * Promiscuous mode
> > * Allmulticast mode
> > * Unicast MAC filter
> > * Multicast MAC filter
> > * RSS hash
> > * VLAN filter
> > * CRC offload
> > * L3 checksum offload
> > * L4 checksum offload
> > * Packet type parsing
> > * Basic stats
> > * Stats per queue
>
> I have more detailed comments but in high level,
> what do you think splitting this patch into three patches:
> - Skeleton
> - Add Rx/Tx support
> - Add features, like MTU update or Promiscuous etc.. support
If it's how submission process works then I think you left me with no
other option than splitting driver into nice patchset :). On the other
hand driver is really a wrapper to MUSDK library and thus quite easy to
follow. What are the benefits of such 3-way split?
>
> >
> > Driver was engineered cooperatively by Semihalf and Marvell teams.
> >
> > Semihalf:
> > Jacek Siuda 
> > Tomasz Duszynski 
> >
> > Marvell:
> > Dmitri Epshtein 
> > Natalie Samsonov 
> >
> > Signed-off-by: Jacek Siuda 
> > Signed-off-by: Tomasz Duszynski 
>
> <...>
>
> > +static struct rte_vdev_driver pmd_mrvl_drv = {
> > +   .probe = rte_pmd_mrvl_probe,
> > +   .remove = rte_pmd_mrvl_remove,
> > +};
> > +
> > +RTE_PMD_REGISTER_VDEV(net_mrvl, pmd_mrvl_drv);
>
> Please help me understand.
>
> This driver implemented as virtual driver, because:
> With the help of custom kernel modules, musdk library already provides
> userspace datapath support. This PMD is an interface to musdk library.
> Is this correct?
That is right. Another reason this NIC is not PCI device.
>
> If so, just thinking loud:
> - Why not implement this PMD directly on top of kernel interface,
> removing musdk layer completely?
> - How big problem that this PMD depends on custom kernel code?
I think the main reason is that MUSDK is already used in different projects.
Keeping multiple codebases offering similar functionality would be quite
demanding in terms of extra work needed.
> - How library and custom kernel code delivered? For which platforms?
Kernel and library sources are hosted on publicly available repository.
Driver was tested on Armada 7k/8k SoCs.
>
> <>
>

--
- Tomasz Duszyński

[dpdk-dev] DPDK techboard minutes of September 15]

[Pavan Nikhilesh Bhagavatula]

On Wed, Sep 20, 2017 at 09:36:48PM +, Ananyev, Konstantin wrote:

Hi everyone,

As per the last techboard meeting's conclusion,


>
> 3) Request for feedback regarding the new license requirement for the 
> optimized version of 64bit division code from libdivide.
> It was pointed out that current 32-bit version is not BSD-licensed either:
> http://dpdk.org/browse/dpdk/tree/lib/librte_sched/rte_reciprocal.h
> TB recommendations:
> a) request LF legals to audit current lib/librte_sched/rte_reciprocal.h
> b) ask author of the patch for 64-bit version to:
> - try to find BSD licensed implementation of the same algorithm
> - contact the authors of the original code and ask their permission for 
> re-licensing in BSD

I have contacted the author and got the following response

---
"libdivide is made available under the zlib license, which is very permissive
and compatible with BSD licenses. libdivide may be freely used in BSD-licensed
software, or also GPL or commercial software.

You do not have to add any copyright notices if you use libdivide in your
binaries. If you simply add libdivide.h to your sources and use it, you are
complying with the license.

It is my hope with libdivide that anyone can use it for anything.
The license is only so that nobody else claims to be the original author."


"It’s fine to modify libdivide to remove pieces you don’t use, or just copy
and paste the parts you do use. However it’s not reasonable to ask me to
contribute libdivide "without the licensing part;” in fact I cannot do so
because libdivide has more than one contributor.

The good news is that using libdivide under its license is very easy!
You only have to include the 14 line license text (under "ZLib license”).
The simplest way is with a source code comment in the file itself - I see you
have several other acknowledgements in rte_reciprocal.h, so that seems like a
natural place. A separate acknowledgements file is also fine if you prefer.

I hope that explains things and thanks for understanding,
_fish"
---

The authors concern is contributing it to BSD licence will be a weak link
to the original author credits.

>

>
> Thanks,
> Konstantin

Thanks,
Pavan

[dpdk-dev] [PATCH v4] Policy Based Power Control for Guest

[David Hunt]

This patchset adds the facility for a guest VM to send a policy down to the
host that will allow the host to scale up/down cpu frequencies
depending on the policy criteria independently of the DPDK app running in
the guest.  This differs from the previous vm_power implementation where
individual scale up/down requests were send from the guest to the host via
virtio-serial.

V4 patchset changes:
  * None, replying into correct email thread. V3 was a reply to the
turbo patch set, should have been inband policy power patchset.

V3 patchset changes:
  * Changed to using is_same_ether_addr() instead of looping through
the mac address bytes to compare them.
  * Tweaked some comments and working in the i40e patch after review.
  * Added a patch to the set to add new i40e function to map file, so
as to allow shared library builds. The power library API needs a cleanup
in next release, so will add API/ABI warning for this cleanup in a
separate patch.

V2 patchset changes:
  * Removed API's in ethdev layer.
  * Now just a single new API in the i40e driver for mapping VF MAC to
VF index.
  * Moved new function from rte_rxtx.c to rte_pmd_i40e.c
  * Removed function for reading i40e register, moved to using the
standard stats API.
  * Renamed i40e function to rte_pmd_i40e_query_vfid_by_mac
  * Cleaned up policy generation code.

It's a modification of the vm_power_manager app that runs in the host, and
the guest_vm_power_app example app that runs in the guest. This allows the
guest to send down a policy to the host via virtio-serial, which then allows
the host to scale up/down based on the criteria in the policy, resulting in
quicker scale up/down than individual requests coming from the guest.
It also means that the DPDK application running in the guest does not need
to be modified in any way, it is unaware that it's cores are being scaled
up/down, reducing the effort in implementing a power-aware infrastructure.

The usage model is as follows:
1. Set up the VF's and assign to the guest in the usual way.
2. run vm_power_manager on the host, creating a channel to the guest.
3. Start the guest_vm_power_mgr app on the guest, which establishes
   a virtio-serial channel to the host.
4. Send down the profile for the guest using the "send_profile now" command.
   There is an example profile hard-coded into guest_vm_power_mgr.
5. Stop the guest_vm_power_mgr and run your normal power-unaware application.
6. Send traffic into the VFs at varying traffic rates.
   Observe the frequency change on the host (turbostat -i 1)

The sequence of code changes are as follows:

A new function has been aded to the i40e driver to allow mapping of
a VF MAC to VF index.

Next we make an addition to librte_power that adds an extra command to allow
the passing of a policy structure from the guest to the host. This struct
contains information like busy/quiet hour, packet throughput thresholds, etc.

The next addition adds functionality to convert the virtual CPU (vcpU0 IDs to
physical CPU (pcpu) IDs so that the host can scale up/down the cores used
in the guest.

The remaining patches are functionality to process the policy, and take action
when the relevant trigger occurs to cause a frequency change.

[1/9] net/i40e: add API to convert VF MAC to VF id
[2/9] lib/librte_power: add extra msg type for policies
[3/9] examples/vm_power_mgr: add vcpu to pcpu mapping
[4/9] examples/vm_power_mgr: add scale to medium freq fn
[5/9] examples/vm_power_mgr: add policy to channels
[6/9] examples/vm_power_mgr: add port initialisation
[7/9] power: add send channel msg function to map file
[8/9] examples/guest_cli: add send policy to host
[9/9] examples/vm_power_mgr: set MAC address of VF

[dpdk-dev] [PATCH v4 4/9] examples/vm_power_mgr: add scale to medium freq fn

[David Hunt]

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: Rory Sexton 
Signed-off-by: David Hunt 
---
 examples/vm_power_manager/power_manager.c | 15 +++
 examples/vm_power_manager/power_manager.h | 13 +
 2 files changed, 28 insertions(+)

diff --git a/examples/vm_power_manager/power_manager.c 
b/examples/vm_power_manager/power_manager.c
index 80705f9..c021c1d 100644
--- a/examples/vm_power_manager/power_manager.c
+++ b/examples/vm_power_manager/power_manager.c
@@ -286,3 +286,18 @@ power_manager_disable_turbo_core(unsigned int core_num)
POWER_SCALE_CORE(disable_turbo, core_num, ret);
return ret;
 }
+
+int
+power_manager_scale_core_med(unsigned int core_num)
+{
+   int ret = 0;
+
+   if (core_num >= POWER_MGR_MAX_CPUS)
+   return -1;
+   if (!(global_enabled_cpus & (1ULL << core_num)))
+   return -1;
+   rte_spinlock_lock(&global_core_freq_info[core_num].power_sl);
+   ret = rte_power_set_freq(core_num, 5);
+   rte_spinlock_unlock(&global_core_freq_info[core_num].power_sl);
+   return ret;
+}
diff --git a/examples/vm_power_manager/power_manager.h 
b/examples/vm_power_manager/power_manager.h
index b74d09b..b52fb4c 100644
--- a/examples/vm_power_manager/power_manager.h
+++ b/examples/vm_power_manager/power_manager.h
@@ -231,6 +231,19 @@ int power_manager_disable_turbo_core(unsigned int 
core_num);
  */
 uint32_t power_manager_get_current_frequency(unsigned core_num);
 
+/**
+ * Scale to medium frequency for the core specified by core_num.
+ * It is thread-safe.
+ *
+ * @param core_num
+ *  The core number to change frequency
+ *
+ * @return
+ *  - 1 on success.
+ *  - 0 if frequency not changed.
+ *  - Negative on error.
+ */
+int power_manager_scale_core_med(unsigned int core_num);
 
 #ifdef __cplusplus
 }
-- 
2.7.4

[dpdk-dev] [PATCH v4 2/9] lib/librte_power: add extra msg type for policies

[David Hunt]

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: Rory Sexton 
Signed-off-by: David Hunt 
---
 lib/librte_power/channel_commands.h | 52 +
 1 file changed, 52 insertions(+)

diff --git a/lib/librte_power/channel_commands.h 
b/lib/librte_power/channel_commands.h
index 484085b..1599706 100644
--- a/lib/librte_power/channel_commands.h
+++ b/lib/librte_power/channel_commands.h
@@ -46,6 +46,7 @@ extern "C" {
 /* Valid Commands */
 #define CPU_POWER   1
 #define CPU_POWER_CONNECT   2
+#define PKT_POLICY  3
 
 /* CPU Power Command Scaling */
 #define CPU_POWER_SCALE_UP  1
@@ -54,11 +55,62 @@ extern "C" {
 #define CPU_POWER_SCALE_MIN 4
 #define CPU_POWER_ENABLE_TURBO  5
 #define CPU_POWER_DISABLE_TURBO 6
+#define HOURS 24
+
+#ifdef RTE_LIBRTE_I40E_PMD
+#define MAX_VFS 10
+#endif
+
+#define MAX_VCPU_PER_VM 8
+
+typedef enum {false, true} bool;
+
+struct t_boost_status {
+   bool tbEnabled;
+};
+
+struct timer_profile {
+   int busy_hours[HOURS];
+   int quiet_hours[HOURS];
+#ifdef RTE_LIBRTE_I40E_PMD
+   int hours_to_use_traffic_profile[HOURS];
+#endif
+};
+
+enum workload {HIGH, MEDIUM, LOW};
+enum policy_to_use {
+#ifdef RTE_LIBRTE_I40E_PMD
+   TRAFFIC,
+#endif
+   TIME,
+   WORKLOAD
+};
+
+#ifdef RTE_LIBRTE_I40E_PMD
+struct traffic {
+   uint32_t min_packet_thresh;
+   uint32_t avg_max_packet_thresh;
+   uint32_t max_max_packet_thresh;
+};
+#endif
 
 struct channel_packet {
uint64_t resource_id; /**< core_num, device */
uint32_t unit;/**< scale down/up/min/max */
uint32_t command; /**< Power, IO, etc */
+   char vm_name[32];
+
+#ifdef RTE_LIBRTE_I40E_PMD
+   uint64_t vfid[MAX_VFS];
+   int nb_mac_to_monitor;
+   struct traffic traffic_policy;
+#endif
+   uint8_t vcpu_to_control[MAX_VCPU_PER_VM];
+   uint8_t num_vcpu;
+   struct timer_profile timer_policy;
+   enum workload workload;
+   enum policy_to_use policy_to_use;
+   struct t_boost_status t_boost_status;
 };
 
 
-- 
2.7.4

[dpdk-dev] [PATCH v4 3/9] examples/vm_power_mgr: add vcpu to pcpu mapping

[David Hunt]

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: Rory Sexton 
Signed-off-by: David Hunt 
---
 examples/vm_power_manager/channel_manager.c | 62 +
 examples/vm_power_manager/channel_manager.h | 25 
 2 files changed, 87 insertions(+)

diff --git a/examples/vm_power_manager/channel_manager.c 
b/examples/vm_power_manager/channel_manager.c
index e068ae2..03fa626 100644
--- a/examples/vm_power_manager/channel_manager.c
+++ b/examples/vm_power_manager/channel_manager.c
@@ -574,6 +574,68 @@ set_channel_status(const char *vm_name, unsigned 
*channel_list,
return num_channels_changed;
 }
 
+void
+get_all_vm(int *num_vm, int *num_cpu)
+{
+
+   virNodeInfo node_info;
+   virDomainPtr *domptr;
+   uint64_t mask;
+   int i, ii, numVcpus[MAX_VCPUS], cpu, n_vcpus;
+   unsigned int jj;
+   const char *vm_name;
+   unsigned int flags = VIR_CONNECT_LIST_DOMAINS_RUNNING |
+   VIR_CONNECT_LIST_DOMAINS_PERSISTENT;
+   unsigned int flag = VIR_DOMAIN_VCPU_CONFIG;
+
+
+   memset(global_cpumaps, 0, CHANNEL_CMDS_MAX_CPUS*global_maplen);
+   if (virNodeGetInfo(global_vir_conn_ptr, &node_info))
+   RTE_LOG(ERR, CHANNEL_MANAGER, "Unable to retrieve node Info\n");
+
+   /* Returns number of pcpus */
+   global_n_host_cpus = (unsigned int)node_info.cpus;
+
+   /* Returns number of active domains */
+   *num_vm = virConnectListAllDomains(global_vir_conn_ptr, &domptr, flags);
+   if (*num_vm <= 0)
+   RTE_LOG(ERR, CHANNEL_MANAGER, "No Active Domains Running\n");
+
+   for (i = 0; i < *num_vm; i++) {
+
+   /* Get Domain Names */
+   vm_name = virDomainGetName(domptr[i]);
+   lvm_info[i].vm_name = vm_name;
+
+   /* Get Number of Vcpus */
+   numVcpus[i] = virDomainGetVcpusFlags(domptr[i], flag);
+
+   /* Get Number of VCpus & VcpuPinInfo */
+   n_vcpus = virDomainGetVcpuPinInfo(domptr[i],
+   numVcpus[i], global_cpumaps,
+   global_maplen, flag);
+
+   if ((int)n_vcpus > 0) {
+   *num_cpu = n_vcpus;
+   lvm_info[i].num_cpus = n_vcpus;
+   }
+
+   /* Save pcpu in use by libvirt VMs */
+   for (ii = 0; ii < n_vcpus; ii++) {
+   mask = 0;
+   for (jj = 0; jj < global_n_host_cpus; jj++) {
+   if (VIR_CPU_USABLE(global_cpumaps,
+   global_maplen, ii, jj) > 0) {
+   mask |= 1ULL << jj;
+   }
+   }
+   ITERATIVE_BITMASK_CHECK_64(mask, cpu) {
+   lvm_info[i].pcpus[ii] = cpu;
+   }
+   }
+   }
+}
+
 int
 get_info_vm(const char *vm_name, struct vm_info *info)
 {
diff --git a/examples/vm_power_manager/channel_manager.h 
b/examples/vm_power_manager/channel_manager.h
index 47c3b9c..788c1e6 100644
--- a/examples/vm_power_manager/channel_manager.h
+++ b/examples/vm_power_manager/channel_manager.h
@@ -66,6 +66,17 @@ struct sockaddr_un _sockaddr_un;
 #define UNIX_PATH_MAX sizeof(_sockaddr_un.sun_path)
 #endif
 
+#define MAX_VMS 4
+#define MAX_VCPUS 20
+
+
+struct libvirt_vm_info {
+   const char *vm_name;
+   unsigned int pcpus[MAX_VCPUS];
+   uint8_t num_cpus;
+};
+
+struct libvirt_vm_info lvm_info[MAX_VMS];
 /* Communication Channel Status */
 enum channel_status { CHANNEL_MGR_CHANNEL_DISCONNECTED = 0,
CHANNEL_MGR_CHANNEL_CONNECTED,
@@ -319,6 +330,20 @@ int set_channel_status(const char *vm_name, unsigned 
*channel_list,
  */
 int get_info_vm(const char *vm_name, struct vm_info *info);
 
+/**
+ * Populates a table with all domains running and their physical cpu.
+ * All information is gathered through libvirt api.
+ *
+ * @param noVms
+ *  modified to store number of active VMs
+ *
+ * @param noVcpus
+modified to store number of vcpus active
+ *
+ * @return
+ *   void
+ */
+void get_all_vm(int *noVms, int *noVcpus);
 #ifdef __cplusplus
 }
 #endif
-- 
2.7.4

[dpdk-dev] [PATCH v4 1/9] net/i40e: add API to convert VF MAC to VF id

[David Hunt]

From: "Sexton, Rory" 

Need a way to convert a vf id to a pf id on the host so as to query the pf
for relevant statistics which are used for the frequency changes in the
vm_power_manager app. Used when profiles are passed down from the guest
to the host, allowing the host to map the vfs to pfs.

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: Rory Sexton 
Signed-off-by: David Hunt 
---
 drivers/net/i40e/rte_pmd_i40e.c   | 31 +++
 drivers/net/i40e/rte_pmd_i40e.h   | 13 +
 drivers/net/i40e/rte_pmd_i40e_version.map |  7 +++
 3 files changed, 51 insertions(+)

diff --git a/drivers/net/i40e/rte_pmd_i40e.c b/drivers/net/i40e/rte_pmd_i40e.c
index f12b7f4..21efb2f 100644
--- a/drivers/net/i40e/rte_pmd_i40e.c
+++ b/drivers/net/i40e/rte_pmd_i40e.c
@@ -2115,3 +2115,34 @@ int rte_pmd_i40e_ptype_mapping_replace(uint8_t port,
 
return 0;
 }
+
+uint64_t
+rte_pmd_i40e_query_vfid_by_mac(uint8_t port, uint64_t vf_mac)
+{
+   struct rte_eth_dev *dev;
+   struct ether_addr *vf_mac_addr = (struct ether_addr *)&vf_mac;
+   struct ether_addr *mac;
+   struct i40e_pf *pf;
+   int vf_id;
+   struct i40e_pf_vf *vf;
+   uint16_t vf_num;
+
+   RTE_ETH_VALID_PORTID_OR_ERR_RET(port, -ENODEV);
+   dev = &rte_eth_devices[port];
+
+   if (!is_i40e_supported(dev))
+   return -ENOTSUP;
+
+   pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+   vf_num = pf->vf_num;
+
+   for (vf_id = 0; vf_id < vf_num; vf_id++) {
+   vf = &pf->vfs[vf_id];
+   mac = &vf->mac_addr;
+
+   if (is_same_ether_addr(mac, vf_mac_addr))
+   return vf_id;
+   }
+
+   return -EINVAL;
+}
diff --git a/drivers/net/i40e/rte_pmd_i40e.h b/drivers/net/i40e/rte_pmd_i40e.h
index 356fa89..a7ae0f0 100644
--- a/drivers/net/i40e/rte_pmd_i40e.h
+++ b/drivers/net/i40e/rte_pmd_i40e.h
@@ -637,4 +637,17 @@ int rte_pmd_i40e_ptype_mapping_replace(uint8_t port,
   uint8_t mask,
   uint32_t pkt_type);
 
+/**
+ * On the PF, find VF index based on VF MAC address
+ *
+ * @param port
+ *pointer to port identifier of the device
+ * @param vf_mac
+ *the mac address of the vf to determine index of
+ * @return
+ *-(-22 EINVAL) the vf mac does not exist on this port
+ *-(!-22) the index of vfid in pf->vfs
+ */
+uint64_t rte_pmd_i40e_query_vfid_by_mac(uint8_t port, uint64_t vf_mac);
+
 #endif /* _PMD_I40E_H_ */
diff --git a/drivers/net/i40e/rte_pmd_i40e_version.map 
b/drivers/net/i40e/rte_pmd_i40e_version.map
index 20cc980..d8b74bd 100644
--- a/drivers/net/i40e/rte_pmd_i40e_version.map
+++ b/drivers/net/i40e/rte_pmd_i40e_version.map
@@ -45,3 +45,10 @@ DPDK_17.08 {
rte_pmd_i40e_get_ddp_info;
 
 } DPDK_17.05;
+
+DPDK_17.11 {
+   global:
+
+   rte_pmd_i40e_query_vfid_by_mac;
+
+} DPDK_17.08;
-- 
2.7.4

[dpdk-dev] [PATCH v4 5/9] examples/vm_power_mgr: add policy to channels

[David Hunt]

From: "Sexton, Rory" 

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: Rory Sexton 
Signed-off-by: David Hunt 
---
 examples/vm_power_manager/Makefile  |  16 ++
 examples/vm_power_manager/channel_monitor.c | 340 +++-
 examples/vm_power_manager/channel_monitor.h |  20 ++
 3 files changed, 370 insertions(+), 6 deletions(-)

diff --git a/examples/vm_power_manager/Makefile 
b/examples/vm_power_manager/Makefile
index 59a9641..9cf20a2 100644
--- a/examples/vm_power_manager/Makefile
+++ b/examples/vm_power_manager/Makefile
@@ -54,6 +54,22 @@ CFLAGS += $(WERROR_FLAGS)
 
 LDLIBS += -lvirt
 
+ifeq ($(CONFIG_RTE_BUILD_SHARED_LIB),y)
+
+ifeq ($(CONFIG_RTE_LIBRTE_IXGBE_PMD),y)
+LDLIBS += -lrte_pmd_ixgbe
+endif
+
+ifeq ($(CONFIG_RTE_LIBRTE_I40E_PMD),y)
+LDLIBS += -lrte_pmd_i40e
+endif
+
+ifeq ($(CONFIG_RTE_LIBRTE_BNXT_PMD),y)
+LDLIBS += -lrte_pmd_bnxt
+endif
+
+endif
+
 # workaround for a gcc bug with noreturn attribute
 # http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12603
 ifeq ($(CONFIG_RTE_TOOLCHAIN_GCC),y)
diff --git a/examples/vm_power_manager/channel_monitor.c 
b/examples/vm_power_manager/channel_monitor.c
index ac40dac..7db98ad 100644
--- a/examples/vm_power_manager/channel_monitor.c
+++ b/examples/vm_power_manager/channel_monitor.c
@@ -41,13 +41,20 @@
 #include 
 #include 
 #include 
+#include 
 
 #include 
 #include 
 #include 
 #include 
+#include 
+#include 
 
+#ifdef RTE_LIBRTE_I40E_PMD
+#include 
+#endif
 
+#include 
 #include "channel_monitor.h"
 #include "channel_commands.h"
 #include "channel_manager.h"
@@ -57,10 +64,17 @@
 
 #define MAX_EVENTS 256
 
+#ifdef RTE_LIBRTE_I40E_PMD
+uint64_t vsi_pkt_count_prev[384];
+uint64_t rdtsc_prev[384];
+#endif
 
+double time_period_s = 1;
 static volatile unsigned run_loop = 1;
 static int global_event_fd;
+static unsigned int policy_is_set;
 static struct epoll_event *global_events_list;
+static struct policy policies[MAX_VMS];
 
 void channel_monitor_exit(void)
 {
@@ -68,6 +82,302 @@ void channel_monitor_exit(void)
rte_free(global_events_list);
 }
 
+static void
+core_share(int pNo, int z, int x, int t)
+{
+   if (policies[pNo].core_share[z].pcpu == lvm_info[x].pcpus[t]) {
+   if (strcmp(policies[pNo].pkt.vm_name,
+   lvm_info[x].vm_name) != 0) {
+   policies[pNo].core_share[z].status = 1;
+   power_manager_scale_core_max(
+   policies[pNo].core_share[z].pcpu);
+   }
+   }
+}
+
+static void
+core_share_status(int pNo)
+{
+
+   int noVms, noVcpus, z, x, t;
+
+   get_all_vm(&noVms, &noVcpus);
+
+   /* Reset Core Share Status. */
+   for (z = 0; z < noVcpus; z++)
+   policies[pNo].core_share[z].status = 0;
+
+   /* Foreach vcpu in a policy. */
+   for (z = 0; z < policies[pNo].pkt.num_vcpu; z++) {
+   /* Foreach VM on the platform. */
+   for (x = 0; x < noVms; x++) {
+   /* Foreach vcpu of VMs on platform. */
+   for (t = 0; t < lvm_info[x].num_cpus; t++)
+   core_share(pNo, z, x, t);
+   }
+   }
+}
+
+static void
+get_pcpu_to_control(struct policy *pol)
+{
+
+   /* Convert vcpu to pcpu. */
+   struct vm_info info;
+   int pcpu, count;
+   uint64_t mask_u64b;
+
+   RTE_LOG(INFO, CHANNEL_MONITOR, "Looking for pcpu for %s\n",
+   pol->pkt.vm_name);
+   get_info_vm(pol->pkt.vm_name, &info);
+
+   for (count = 0; count < pol->pkt.num_vcpu; count++) {
+   mask_u64b = info.pcpu_mask[pol->pkt.vcpu_to_control[count]];
+   for (pcpu = 0; mask_u64b; mask_u64b &= ~(1ULL << pcpu++)) {
+   if ((mask_u64b >> pcpu) & 1)
+   pol->core_share[count].pcpu = pcpu;
+   }
+   }
+}
+
+#ifdef RTE_LIBRTE_I40E_PMD
+static int
+get_pfid(struct policy *pol)
+{
+
+   int i, x, ret = 0, nb_ports;
+
+   nb_ports = rte_eth_dev_count();
+   for (i = 0; i < pol->pkt.nb_mac_to_monitor; i++) {
+
+   for (x = 0; x < nb_ports; x++) {
+   ret = rte_pmd_i40e_query_vfid_by_mac(x,
+   pol->pkt.vfid[i]);
+   if (ret != -EINVAL) {
+   pol->port[i] = x;
+   break;
+   }
+   }
+   if (ret == -EINVAL) {
+   RTE_LOG(INFO, CHANNEL_MONITOR,
+   "Error with Policy. MAC not found on "
+   "attached ports ");
+   pol->enabled = 0;
+   return ret;
+   }
+   pol->pfid[i] = ret;
+   }
+   return 1;
+}
+#endif
+
+static int
+update_policy(struct channel_packet *pkt)
+{
+
+   unsigned int updated = 0;
+
+   for (int i = 0; i < 

[dpdk-dev] [PATCH v4 6/9] examples/vm_power_mgr: add port initialisation

[David Hunt]

We need to initialise the port's we're monitoring to be able to see
the throughput.

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: David Hunt 
---
 examples/vm_power_manager/main.c | 220 +++
 1 file changed, 220 insertions(+)

diff --git a/examples/vm_power_manager/main.c b/examples/vm_power_manager/main.c
index c33fcc9..698abca 100644
--- a/examples/vm_power_manager/main.c
+++ b/examples/vm_power_manager/main.c
@@ -49,6 +49,9 @@
 #include 
 #include 
 #include 
+#include 
+#include 
+#include 
 #include 
 
 #include "channel_manager.h"
@@ -56,6 +59,192 @@
 #include "power_manager.h"
 #include "vm_power_cli.h"
 
+#define RX_RING_SIZE 512
+#define TX_RING_SIZE 512
+
+#define NUM_MBUFS 8191
+#define MBUF_CACHE_SIZE 250
+#define BURST_SIZE 32
+
+static uint32_t enabled_port_mask;
+static volatile bool force_quit;
+
+//
+static const struct rte_eth_conf port_conf_default = {
+   .rxmode = { .max_rx_pkt_len = ETHER_MAX_LEN }
+};
+
+static inline int
+port_init(uint8_t port, struct rte_mempool *mbuf_pool)
+{
+   struct rte_eth_conf port_conf = port_conf_default;
+   const uint16_t rx_rings = 1, tx_rings = 1;
+   int retval;
+   uint16_t q;
+
+   if (port >= rte_eth_dev_count())
+   return -1;
+
+   /* Configure the Ethernet device. */
+   retval = rte_eth_dev_configure(port, rx_rings, tx_rings, &port_conf);
+   if (retval != 0)
+   return retval;
+
+   /* Allocate and set up 1 RX queue per Ethernet port. */
+   for (q = 0; q < rx_rings; q++) {
+   retval = rte_eth_rx_queue_setup(port, q, RX_RING_SIZE,
+   rte_eth_dev_socket_id(port), NULL, mbuf_pool);
+   if (retval < 0)
+   return retval;
+   }
+
+   /* Allocate and set up 1 TX queue per Ethernet port. */
+   for (q = 0; q < tx_rings; q++) {
+   retval = rte_eth_tx_queue_setup(port, q, TX_RING_SIZE,
+   rte_eth_dev_socket_id(port), NULL);
+   if (retval < 0)
+   return retval;
+   }
+
+   /* Start the Ethernet port. */
+   retval = rte_eth_dev_start(port);
+   if (retval < 0)
+   return retval;
+
+   /* Display the port MAC address. */
+   struct ether_addr addr;
+   rte_eth_macaddr_get(port, &addr);
+   printf("Port %u MAC: %02" PRIx8 " %02" PRIx8 " %02" PRIx8
+  " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 "\n",
+   (unsigned int)port,
+   addr.addr_bytes[0], addr.addr_bytes[1],
+   addr.addr_bytes[2], addr.addr_bytes[3],
+   addr.addr_bytes[4], addr.addr_bytes[5]);
+
+   /* Enable RX in promiscuous mode for the Ethernet device. */
+   rte_eth_promiscuous_enable(port);
+
+
+   return 0;
+}
+
+static int
+parse_portmask(const char *portmask)
+{
+   char *end = NULL;
+   unsigned long pm;
+
+   /* parse hexadecimal string */
+   pm = strtoul(portmask, &end, 16);
+   if ((portmask[0] == '\0') || (end == NULL) || (*end != '\0'))
+   return -1;
+
+   if (pm == 0)
+   return -1;
+
+   return pm;
+}
+/* Parse the argument given in the command line of the application */
+static int
+parse_args(int argc, char **argv)
+{
+   int opt, ret;
+   char **argvopt;
+   int option_index;
+   char *prgname = argv[0];
+   static struct option lgopts[] = {
+   { "mac-updating", no_argument, 0, 1},
+   { "no-mac-updating", no_argument, 0, 0},
+   {NULL, 0, 0, 0}
+   };
+   argvopt = argv;
+
+   while ((opt = getopt_long(argc, argvopt, "p:q:T:",
+ lgopts, &option_index)) != EOF) {
+
+   switch (opt) {
+   /* portmask */
+   case 'p':
+   enabled_port_mask = parse_portmask(optarg);
+   if (enabled_port_mask == 0) {
+   printf("invalid portmask\n");
+   return -1;
+   }
+   break;
+   /* long options */
+   case 0:
+   break;
+
+   default:
+   return -1;
+   }
+   }
+
+   if (optind >= 0)
+   argv[optind-1] = prgname;
+
+   ret = optind-1;
+   optind = 0; /* reset getopt lib */
+   return ret;
+}
+
+static void
+check_all_ports_link_status(uint8_t port_num, uint32_t port_mask)
+{
+#define CHECK_INTERVAL 100 /* 100ms */
+#define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */
+   uint8_t portid, count, all_ports_up, print_flag = 0;
+   struct rte_eth_link link;
+
+   printf("\nChecking link status");
+   fflush(stdout);
+   for (count = 0; count <= MAX_CHECK_TIME; count++) {
+   if (force_q

[dpdk-dev] [PATCH v4 7/9] power: add send channel msg function to map file

[David Hunt]

Adding new wrapper function to existing private (but unused 'till now)
function with an rte_power_ prefix.

The plan is to clean up all the header files in the next release so
that only the intended public functions are in the map file and only
the relevant headers have the rte_ prefix so that only they are
included in the documentation.

Signed-off-by: David Hunt 
---
 lib/librte_power/guest_channel.c   |  7 +++
 lib/librte_power/guest_channel.h   | 15 +++
 lib/librte_power/rte_power_version.map |  1 +
 3 files changed, 23 insertions(+)

diff --git a/lib/librte_power/guest_channel.c b/lib/librte_power/guest_channel.c
index 85c92fa..fa5de0f 100644
--- a/lib/librte_power/guest_channel.c
+++ b/lib/librte_power/guest_channel.c
@@ -148,6 +148,13 @@ guest_channel_send_msg(struct channel_packet *pkt, 
unsigned lcore_id)
return 0;
 }
 
+int rte_power_guest_channel_send_msg(struct channel_packet *pkt,
+   unsigned int lcore_id)
+{
+   return guest_channel_send_msg(pkt, lcore_id);
+}
+
+
 void
 guest_channel_host_disconnect(unsigned lcore_id)
 {
diff --git a/lib/librte_power/guest_channel.h b/lib/librte_power/guest_channel.h
index 9e18af5..741339c 100644
--- a/lib/librte_power/guest_channel.h
+++ b/lib/librte_power/guest_channel.h
@@ -81,6 +81,21 @@ void guest_channel_host_disconnect(unsigned lcore_id);
  */
 int guest_channel_send_msg(struct channel_packet *pkt, unsigned lcore_id);
 
+/**
+ * Send a message contained in pkt over the Virtio-Serial to the host endpoint.
+ *
+ * @param pkt
+ *  Pointer to a populated struct channel_packet
+ *
+ * @param lcore_id
+ *  lcore_id.
+ *
+ * @return
+ *  - 0 on success.
+ *  - Negative on error.
+ */
+int rte_power_guest_channel_send_msg(struct channel_packet *pkt,
+   unsigned int lcore_id);
 
 #ifdef __cplusplus
 }
diff --git a/lib/librte_power/rte_power_version.map 
b/lib/librte_power/rte_power_version.map
index 9ae0627..96dc42e 100644
--- a/lib/librte_power/rte_power_version.map
+++ b/lib/librte_power/rte_power_version.map
@@ -20,6 +20,7 @@ DPDK_2.0 {
 DPDK_17.11 {
global:
 
+   rte_power_guest_channel_send_msg;
rte_power_freq_disable_turbo;
rte_power_freq_enable_turbo;
rte_power_turbo_status;
-- 
2.7.4

[dpdk-dev] [PATCH v4 9/9] examples/vm_power_mgr: set MAC address of VF

[David Hunt]

We need to set vf mac from the host, so that they will be in sync on the
guest and the host. Otherwise, we'll have a random mac on the guest, and
a 00:00:00:00:00:00 mac on the host.

Signed-off-by: David Hunt 
---
 examples/vm_power_manager/main.c | 60 +++-
 1 file changed, 59 insertions(+), 1 deletion(-)

diff --git a/examples/vm_power_manager/main.c b/examples/vm_power_manager/main.c
index 698abca..18f5e7f 100644
--- a/examples/vm_power_manager/main.c
+++ b/examples/vm_power_manager/main.c
@@ -58,6 +58,15 @@
 #include "channel_monitor.h"
 #include "power_manager.h"
 #include "vm_power_cli.h"
+#ifdef RTE_LIBRTE_IXGBE_PMD
+#include 
+#endif
+#ifdef RTE_LIBRTE_I40E_PMD
+#include 
+#endif
+#ifdef RTE_LIBRTE_BNXT_PMD
+#include 
+#endif
 
 #define RX_RING_SIZE 512
 #define TX_RING_SIZE 512
@@ -222,7 +231,7 @@ check_all_ports_link_status(uint8_t port_num, uint32_t 
port_mask)
(uint8_t)portid);
continue;
}
-  /* clear all_ports_up flag if any link down */
+   /* clear all_ports_up flag if any link down */
if (link.link_status == ETH_LINK_DOWN) {
all_ports_up = 0;
break;
@@ -273,7 +282,9 @@ main(int argc, char **argv)
unsigned lcore_id;
unsigned int nb_ports;
struct rte_mempool *mbuf_pool;
+#ifdef RTE_LIBRTE_I40E_PMD
uint8_t portid;
+#endif
 
 
ret = rte_eal_init(argc, argv);
@@ -300,13 +311,60 @@ main(int argc, char **argv)
rte_exit(EXIT_FAILURE, "Cannot create mbuf pool\n");
 
/* Initialize ports. */
+#ifdef RTE_LIBRTE_I40E_PMD
for (portid = 0; portid < nb_ports; portid++) {
+   struct ether_addr eth;
+   int w, j;
+   int ret = -ENOTSUP;
+
if ((enabled_port_mask & (1 << portid)) == 0)
continue;
+
+   eth.addr_bytes[0] = 0xe0;
+   eth.addr_bytes[1] = 0xe0;
+   eth.addr_bytes[2] = 0xe0;
+   eth.addr_bytes[3] = 0xe0;
+   eth.addr_bytes[4] = portid + 0xf0;
+
if (port_init(portid, mbuf_pool) != 0)
rte_exit(EXIT_FAILURE, "Cannot init port %"PRIu8 "\n",
portid);
+
+   for (w = 0; w < MAX_VFS; w++) {
+   eth.addr_bytes[5] = w + 0xf0;
+
+#ifdef RTE_LIBRTE_IXGBE_PMD
+   if (ret == -ENOTSUP)
+   ret = rte_pmd_ixgbe_set_vf_mac_addr(portid,
+   w, ð);
+#endif
+#ifdef RTE_LIBRTE_I40E_PMD
+   if (ret == -ENOTSUP)
+   ret = rte_pmd_i40e_set_vf_mac_addr(portid,
+   w, ð);
+#endif
+#ifdef RTE_LIBRTE_BNXT_PMD
+   if (ret == -ENOTSUP)
+   ret = rte_pmd_bnxt_set_vf_mac_addr(portid,
+   w, ð);
+#endif
+
+
+   switch (ret) {
+   case 0:
+   printf("Port %d VF %d MAC: ",
+   portid, w);
+   for (j = 0; j < 6; j++) {
+   printf("%02x", eth.addr_bytes[j]);
+   if (j < 5)
+   printf(":");
+   }
+   printf("\n");
+   break;
+   }
+   }
}
+#endif
 
lcore_id = rte_get_next_lcore(-1, 1, 0);
if (lcore_id == RTE_MAX_LCORE) {
-- 
2.7.4

[dpdk-dev] [PATCH v4 8/9] examples/guest_cli: add send policy to host

[David Hunt]

From: "Sexton, Rory" 

Here we're adding an example of setting up a policy, and allowing the
vm_cli_guest app to send it to the host using the cli command
"send_policy now"

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: Rory Sexton 
Signed-off-by: David Hunt 
---
 .../guest_cli/vm_power_cli_guest.c | 105 +
 .../guest_cli/vm_power_cli_guest.h |   6 --
 2 files changed, 105 insertions(+), 6 deletions(-)

diff --git a/examples/vm_power_manager/guest_cli/vm_power_cli_guest.c 
b/examples/vm_power_manager/guest_cli/vm_power_cli_guest.c
index 4e982bd..fe0d77a 100644
--- a/examples/vm_power_manager/guest_cli/vm_power_cli_guest.c
+++ b/examples/vm_power_manager/guest_cli/vm_power_cli_guest.c
@@ -45,8 +45,10 @@
 #include 
 #include 
 #include 
+#include 
 
 #include 
+#include 
 
 #include "vm_power_cli_guest.h"
 
@@ -139,8 +141,111 @@ cmdline_parse_inst_t cmd_set_cpu_freq_set = {
},
 };
 
+struct cmd_send_policy_result {
+   cmdline_fixed_string_t send_policy;
+   cmdline_fixed_string_t cmd;
+};
+
+#ifdef RTE_LIBRTE_I40E_PMD
+union PFID {
+   struct ether_addr addr;
+   uint64_t pfid;
+};
+#endif
+
+static inline int
+send_policy(void)
+{
+   struct channel_packet pkt;
+   int ret;
+
+#ifdef RTE_LIBRTE_I40E_PMD
+   union PFID pfid;
+   /* Use port MAC address as the vfid */
+   rte_eth_macaddr_get(0, &pfid.addr);
+   printf("Port %u MAC: %02" PRIx8 ":%02" PRIx8 ":%02" PRIx8 ":"
+   "%02" PRIx8 ":%02" PRIx8 ":%02" PRIx8 "\n",
+   1,
+   pfid.addr.addr_bytes[0], pfid.addr.addr_bytes[1],
+   pfid.addr.addr_bytes[2], pfid.addr.addr_bytes[3],
+   pfid.addr.addr_bytes[4], pfid.addr.addr_bytes[5]);
+   pkt.vfid[0] = pfid.pfid;
+
+   pkt.nb_mac_to_monitor = 1;
+#endif
+   pkt.t_boost_status.tbEnabled = false;
+
+   pkt.vcpu_to_control[0] = 0;
+   pkt.vcpu_to_control[1] = 1;
+   pkt.num_vcpu = 2;
+   /* Dummy Population. */
+#ifdef RTE_LIBRTE_I40E_PMD
+   pkt.traffic_policy.min_packet_thresh = 96000;
+   pkt.traffic_policy.avg_max_packet_thresh = 180;
+   pkt.traffic_policy.max_max_packet_thresh = 200;
+#endif
+
+   pkt.timer_policy.busy_hours[0] = 3;
+   pkt.timer_policy.busy_hours[1] = 4;
+   pkt.timer_policy.busy_hours[2] = 5;
+   pkt.timer_policy.quiet_hours[0] = 11;
+   pkt.timer_policy.quiet_hours[1] = 12;
+   pkt.timer_policy.quiet_hours[2] = 13;
+
+#ifdef RTE_LIBRTE_I40E_PMD
+   pkt.timer_policy.hours_to_use_traffic_profile[0] = 8;
+   pkt.timer_policy.hours_to_use_traffic_profile[1] = 10;
+#endif
+
+   pkt.workload = LOW;
+   pkt.policy_to_use = TIME;
+   pkt.command = PKT_POLICY;
+   strcpy(pkt.vm_name, "ubintu2");
+   ret = rte_power_guest_channel_send_msg(&pkt, 1);
+   if (ret == 0)
+   return 1;
+   RTE_LOG(DEBUG, POWER, "Error sending message: %s\n",
+   ret > 0 ? strerror(ret) : "channel not connected");
+   return -1;
+}
+
+static void
+cmd_send_policy_parsed(void *parsed_result, struct cmdline *cl,
+  __attribute__((unused)) void *data)
+{
+   int ret = -1;
+   struct cmd_send_policy_result *res = parsed_result;
+
+   if (!strcmp(res->cmd, "now")) {
+   printf("Sending Policy down now!\n");
+   ret = send_policy();
+   }
+   if (ret != 1)
+   cmdline_printf(cl, "Error sending message: %s\n",
+   strerror(ret));
+}
+
+cmdline_parse_token_string_t cmd_send_policy =
+   TOKEN_STRING_INITIALIZER(struct cmd_send_policy_result,
+   send_policy, "send_policy");
+cmdline_parse_token_string_t cmd_send_policy_cmd_cmd =
+   TOKEN_STRING_INITIALIZER(struct cmd_send_policy_result,
+   cmd, "now");
+
+cmdline_parse_inst_t cmd_send_policy_set = {
+   .f = cmd_send_policy_parsed,
+   .data = NULL,
+   .help_str = "send_policy now",
+   .tokens = {
+   (void *)&cmd_send_policy,
+   (void *)&cmd_send_policy_cmd_cmd,
+   NULL,
+   },
+};
+
 cmdline_parse_ctx_t main_ctx[] = {
(cmdline_parse_inst_t *)&cmd_quit,
+   (cmdline_parse_inst_t *)&cmd_send_policy_set,
(cmdline_parse_inst_t *)&cmd_set_cpu_freq_set,
NULL,
 };
diff --git a/examples/vm_power_manager/guest_cli/vm_power_cli_guest.h 
b/examples/vm_power_manager/guest_cli/vm_power_cli_guest.h
index 0c4bdd5..277eab3 100644
--- a/examples/vm_power_manager/guest_cli/vm_power_cli_guest.h
+++ b/examples/vm_power_manager/guest_cli/vm_power_cli_guest.h
@@ -40,12 +40,6 @@ extern "C" {
 
 #include "channel_commands.h"
 
-int guest_channel_host_connect(unsigned lcore_id);
-
-int guest_channel_send_msg(struct channel_packet *pkt, unsigned lcore_id);
-
-void guest_channel_host_disconn

Re: [dpdk-dev] [PATCH v3] vhost: Expose virtio interrupt need on rte_vhost API

[Jan Scheurich]

Friendly reminder: 
Could somebody please have a look at this patch now that the DPDK
summit is over?

Thanks, Jan

> -Original Message-
> From: Jan Scheurich
> Sent: Saturday, 23 September, 2017 22:32
> To: 'dev@dpdk.org' 
> Subject: [PATCH v3] vhost: Expose virtio interrupt need on rte_vhost API
> 
> Performance tests with the OVS DPDK datapath have shown
> that the tx throughput over a vhostuser port into a VM with
> an interrupt-based virtio driver is limited by the overhead
> incurred by virtio interrupts. The OVS PMD spends up to 30%
> of its cycles in system calls kicking the eventfd. Also the core
> running the vCPU is heavily loaded with generating the virtio
> interrupts in KVM on the host and handling these interrupts
> in the virtio-net driver in the guest. This limits the throughput
> to about 500-700 Kpps with a single vCPU.
> 
> OVS is trying to address this issue by batching packets to a
> vhostuser port for some time to limit the virtio interrupt
> frequency. With a 50 us batching period we have measured an
> iperf3  throughput increase by 15% and a PMD utilization
> decrease from 45% to 30%.
> 
> On the other hand, guests using virtio PMDs do not profit from
> time-based tx batching. Instead they experience a 2-3%
> performance penalty and an average latency increase of
> 30-40 us. OVS therefore intends to apply time-based tx
> batching only for vhostuser tx queues that need to trigger
> virtio interrupts.
> 
> Today this information is hidden inside the rte_vhost library
> and not accessible to users of the API. This patch adds a
> function to the API to query it.
> 
> Signed-off-by: Jan Scheurich 
> 
> ---
> 
> v2 -> v3:
>   Fixed even more white-space errors and warnings
> v1 -> v2:
>   Fixed too long commit lines
>   Fixed white-space errors and warnings
> 
>  lib/librte_vhost/rte_vhost.h | 12 
>  lib/librte_vhost/vhost.c | 19 +++
>  2 files changed, 31 insertions(+)
> 
> diff --git a/lib/librte_vhost/rte_vhost.h b/lib/librte_vhost/rte_vhost.h
> index 8c974eb..d62338b 100644
> --- a/lib/librte_vhost/rte_vhost.h
> +++ b/lib/librte_vhost/rte_vhost.h
> @@ -444,6 +444,18 @@ int rte_vhost_get_vhost_vring(int vid, uint16_t 
> vring_idx,
>   */
>  uint32_t rte_vhost_rx_queue_count(int vid, uint16_t qid);
> 
> +/**
> + * Does the virtio driver request interrupts for a vhost tx queue?
> + *
> + * @param vid
> + *  vhost device ID
> + * @param qid
> + *  virtio queue index in mq case
> + * @return
> + *  1 if true, 0 if false
> + */
> +int rte_vhost_tx_interrupt_requested(int vid, uint16_t qid);
> +
>  #ifdef __cplusplus
>  }
>  #endif
> diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c
> index 0b6aa1c..c6e636e 100644
> --- a/lib/librte_vhost/vhost.c
> +++ b/lib/librte_vhost/vhost.c
> @@ -503,3 +503,22 @@ struct virtio_net *
> 
>   return *((volatile uint16_t *)&vq->avail->idx) - vq->last_avail_idx;
>  }
> +
> +int rte_vhost_tx_interrupt_requested(int vid, uint16_t qid)
> +{
> + struct virtio_net *dev;
> + struct vhost_virtqueue *vq;
> +
> + dev = get_device(vid);
> + if (dev == NULL)
> + return 0;
> +
> + vq = dev->virtqueue[qid];
> + if (vq == NULL)
> + return 0;
> +
> + if (unlikely(vq->enabled == 0 || vq->avail == NULL))
> + return 0;
> +
> + return !(vq->avail->flags & VRING_AVAIL_F_NO_INTERRUPT);
> +}

Re: [dpdk-dev] DPDK techboard minutes of September 15]

[Hemant Agrawal]

Hi Pavan,

On 10/4/2017 2:45 PM, Pavan Nikhilesh Bhagavatula wrote:

On Wed, Sep 20, 2017 at 09:36:48PM +, Ananyev, Konstantin wrote:

Hi everyone,

As per the last techboard meeting's conclusion,




3) Request for feedback regarding the new license requirement for the optimized 
version of 64bit division code from libdivide.
It was pointed out that current 32-bit version is not BSD-licensed either:
http://dpdk.org/browse/dpdk/tree/lib/librte_sched/rte_reciprocal.h
TB recommendations:
a) request LF legals to audit current lib/librte_sched/rte_reciprocal.h
b) ask author of the patch for 64-bit version to:
- try to find BSD licensed implementation of the same algorithm
- contact the authors of the original code and ask their permission for 
re-licensing in BSD


I have contacted the author and got the following response

---
"libdivide is made available under the zlib license, which is very permissive
and compatible with BSD licenses. libdivide may be freely used in BSD-licensed
software, or also GPL or commercial software.

You do not have to add any copyright notices if you use libdivide in your
binaries. If you simply add libdivide.h to your sources and use it, you are
complying with the license.

It is my hope with libdivide that anyone can use it for anything.
The license is only so that nobody else claims to be the original author."


"It’s fine to modify libdivide to remove pieces you don’t use, or just copy
and paste the parts you do use. However it’s not reasonable to ask me to
contribute libdivide "without the licensing part;” in fact I cannot do so
because libdivide has more than one contributor.

The good news is that using libdivide under its license is very easy!
You only have to include the 14 line license text (under "ZLib license”).
The simplest way is with a source code comment in the file itself - I see you
have several other acknowledgements in rte_reciprocal.h, so that seems like a
natural place. A separate acknowledgements file is also fine if you prefer.

I hope that explains things and thanks for understanding,
_fish"
---

The authors concern is contributing it to BSD licence will be a weak link
to the original author credits.


You will need explicit permission to use the code as BSD licensed, 
before it can be merged to DPDK.


It is less likely that Gov Board will allow any other license inclusion 
to DPDK main project at this moment. This will impact DPDK charter #6, 
which will mean long legal review cycles of charter.


In case of "lib/librte_sched/rte_reciprocal.h", Stephen has confirmed 
that the implementer gave permission to incorporate this into DPDK as 
BSD licensed.


Regards,
Hemant

> -Original Message-
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Friday, September 22, 2017 3:00 AM
> To: Hemant Agrawal 
> Cc: techbo...@dpdk.org
> Subject: Re: [dpdk-techboard] DPDK - legal check on code base
>
> On Sat, 16 Sep 2017 06:07:42 +
> Hemant Agrawal  wrote:
>
> >   1.  Implementer License - NOT BSD
> > ./lib/librte_sched/rte_reciprocal.h
> >AI - need to ask LF - legal.
>
> The implementer gave permission to incorporate this into DPDK as BSD
> licensed.
> My assumption is that not every file needs to have the BSD licensed
> boilerplate since the overall project is BSD licensed.










Thanks,
Konstantin


Thanks,
Pavan

Re: [dpdk-dev] [PATCH v3 6/7] app/crypto-perf: support multiple queue pairs

[De Lara Guarch, Pablo]

Hi Akhil,

> -Original Message-
> From: Akhil Goyal [mailto:akhil.go...@nxp.com]
> Sent: Tuesday, September 26, 2017 9:42 AM
> To: De Lara Guarch, Pablo ; Doherty,
> Declan 
> Cc: dev@dpdk.org
> Subject: Re: [PATCH v3 6/7] app/crypto-perf: support multiple queue pairs
> 
> Hi Pablo,
> On 9/22/2017 1:25 PM, Pablo de Lara wrote:
> > Add parameter "qps" in crypto performance app, to create multiple
> > queue pairs per device.
> >
> > This new parameter is useful to have multiple logical cores using a
> > single crypto device, without needing to initialize a crypto device
> > per core.
> >
> > Signed-off-by: Pablo de Lara 
> > ---
> >   app/test-crypto-perf/cperf_options.h |  2 +
> >   app/test-crypto-perf/cperf_options_parsing.c | 22 ++
> >   app/test-crypto-perf/cperf_test_latency.c| 14 +++---
> >   app/test-crypto-perf/cperf_test_pmd_cyclecount.c |  7 +--
> >   app/test-crypto-perf/cperf_test_throughput.c | 14 +++---
> >   app/test-crypto-perf/cperf_test_verify.c | 14 +++---
> >   app/test-crypto-perf/main.c  | 56 
> > ++--
> >   doc/guides/tools/cryptoperf.rst  |  4 ++
> >   8 files changed, 84 insertions(+), 49 deletions(-)
> >
> > diff --git a/app/test-crypto-perf/cperf_options.h
> > b/app/test-crypto-perf/cperf_options.h
> > index 6d339f4..468d5e2 100644
> > --- a/app/test-crypto-perf/cperf_options.h
> > +++ b/app/test-crypto-perf/cperf_options.h
> > @@ -15,6 +15,7 @@
> >   #define CPERF_DESC_NB ("desc-nb")
> >
> >   #define CPERF_DEVTYPE ("devtype")
> > +#define CPERF_QP_NB("qp-nb")
> >   #define CPERF_OPTYPE  ("optype")
> >   #define CPERF_SESSIONLESS ("sessionless")
> >   #define CPERF_OUT_OF_PLACE("out-of-place")
> > @@ -74,6 +75,7 @@ struct cperf_options {
> > uint32_t segment_sz;
> > uint32_t test_buffer_size;
> > uint32_t nb_descriptors;
> > +   uint32_t nb_qps;
> >
> > uint32_t sessionless:1;
> > uint32_t out_of_place:1;
> > diff --git a/app/test-crypto-perf/cperf_options_parsing.c
> > b/app/test-crypto-perf/cperf_options_parsing.c
> > index 89f86a2..441cd61 100644
> > --- a/app/test-crypto-perf/cperf_options_parsing.c
> > +++ b/app/test-crypto-perf/cperf_options_parsing.c
> > @@ -364,6 +364,24 @@ parse_desc_nb(struct cperf_options *opts,
> const char *arg)
> >   }
> >
> >   static int
> > +parse_qp_nb(struct cperf_options *opts, const char *arg) {
> > +   int ret = parse_uint32_t(&opts->nb_qps, arg);
> > +
> > +   if (ret) {
> > +   RTE_LOG(ERR, USER1, "failed to parse number of queue
> pairs\n");
> > +   return -1;
> > +   }
> > +
> > +   if ((opts->nb_qps == 0) || (opts->nb_qps > 256)) {
> Shouldn't this be a macro for max nb_qps.
> 
> Also a generic comment on this patch..  Why do we need an explicit
> parameter for nb-qps. Can't we do it similar to ipsec-secgw.
> It takes the devices and maps the queues with core as per the devices'
> capabilities.

I see... that looks like a good idea. I am implementing it, but will do it 
slightly different.
Instead of having the number of queue pairs per device equal to the number of 
logical cores,
I will divide the number of cores by the number of crypto devices.
So, if 4 cores are available and 2 crypto devices are used, 2 queue pairs will 
be set up. 

Thanks for your review,
Pablo

> 
> -Akhil

Re: [dpdk-dev] [PATCH v2 06/12] ethdev: extend ethdev to support security APIs

[Shahaf Shuler]

Tuesday, October 3, 2017 4:14 PM, Akhil Goyal:
> From: Declan Doherty 
> 
> rte_flow_action type and ethdev updated to support rte_security sessions
> for crypto offload to ethernet device.
> 
> Signed-off-by: Boris Pismenny 
> Signed-off-by: Aviad Yehezkel 
> Signed-off-by: Radu Nicolau 
> Signed-off-by: Declan Doherty 
> ---
>  lib/librte_ether/rte_ethdev.c   | 11 +++
>  lib/librte_ether/rte_ethdev.h   | 19 +--
>  lib/librte_ether/rte_ethdev_version.map |  7 +++
>  3 files changed, 35 insertions(+), 2 deletions(-)
> 
> diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
> index 0597641..f51c5a5 100644
> --- a/lib/librte_ether/rte_ethdev.c
> +++ b/lib/librte_ether/rte_ethdev.c
> @@ -302,6 +302,17 @@ rte_eth_dev_socket_id(uint8_t port_id)
>   return rte_eth_devices[port_id].data->numa_node;
>  }
> 
> +uint16_t
> +rte_eth_dev_get_sec_id(uint8_t port_id) {
> + RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -1);
> +
> + if (rte_eth_devices[port_id].data->dev_flags &
> RTE_ETH_DEV_SECURITY)
> + return rte_eth_devices[port_id].data->sec_id;
> +
> + return -1;
> +}
> +
>  uint8_t
>  rte_eth_dev_count(void)
>  {
> diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
> index 0adf327..193ad62 100644
> --- a/lib/librte_ether/rte_ethdev.h
> +++ b/lib/librte_ether/rte_ethdev.h
> @@ -180,6 +180,8 @@ extern "C" {
>  #include 
>  #include 
>  #include 
> +#include 
> +
>  #include "rte_ether.h"
>  #include "rte_eth_ctrl.h"
>  #include "rte_dev_info.h"
> @@ -357,7 +359,8 @@ struct rte_eth_rxmode {
>   jumbo_frame  : 1, /**< Jumbo Frame Receipt enable. */
>   hw_strip_crc : 1, /**< Enable CRC stripping by hardware. */
>   enable_scatter   : 1, /**< Enable scatter packets rx handler */
> - enable_lro   : 1; /**< Enable LRO */
> + enable_lro   : 1, /**< Enable LRO */
> + enable_sec   : 1; /**< Enable security offload */
>  };
> 
>  /**
> @@ -679,8 +682,10 @@ struct rte_eth_txmode {
>   /**< If set, reject sending out tagged pkts */
>   hw_vlan_reject_untagged : 1,
>   /**< If set, reject sending out untagged pkts */
> - hw_vlan_insert_pvid : 1;
> + hw_vlan_insert_pvid : 1,
>   /**< If set, enable port based VLAN insertion */
> + enable_sec   : 1;
> + /**< Enable security offload */

Already comment on it in the previous version [1]. 
I don't think there is a justification to introduce new approach to set Tx 
offloads given there is already patch set which provides such new API [2].
I think this patch should be on top of it.

>  };
> 
>  /**
> @@ -907,6 +912,7 @@ struct rte_eth_conf {  #define
> DEV_RX_OFFLOAD_QINQ_STRIP  0x0020  #define
> DEV_RX_OFFLOAD_OUTER_IPV4_CKSUM 0x0040
>  #define DEV_RX_OFFLOAD_MACSEC_STRIP 0x0080
> +#define DEV_RX_OFFLOAD_SECURITY 0x0100
> 
>  /**
>   * TX offload capabilities of a device.
> @@ -926,6 +932,8 @@ struct rte_eth_conf {
>  #define DEV_TX_OFFLOAD_GENEVE_TNL_TSO   0x1000/**< Used for
> tunneling packet. */
>  #define DEV_TX_OFFLOAD_MACSEC_INSERT0x2000
>  #define DEV_TX_OFFLOAD_MT_LOCKFREE  0x4000
> +#define DEV_TX_OFFLOAD_SECURITY 0x8000
> +#define DEV_TX_OFFLOAD_SEC_NEED_MDATA   0x0001

Isn't it better to have the DEV_TX_OFFLOAD_SEC_NEED_MDATA  as part of 
rte_security_capability struct?
The "need metadata" should be per security operation indication.
For example It is possible that PMD will be able to do IPSEC without the need 
in metadata and PDCP with the need in metadata. 

IMO the better model is for the PMD to advertise it support all kind of 
security offloads by setting the DEV_TX_OFFLOAD_SECURITY flag. Application will 
be able to query more fine-grained capabilities per security operation using 
rte_security APIs.

>  /**< Multiple threads can invoke rte_eth_tx_burst() concurrently on the
> same
>   * tx queue without SW lock.
>   */
> @@ -1651,6 +1659,9 @@ struct rte_eth_dev {
>   enum rte_eth_dev_state state; /**< Flag indicating the port state */
> } __rte_cache_aligned;
> 
> +uint16_t
> +rte_eth_dev_get_sec_id(uint8_t port_id);
> +
>  struct rte_eth_dev_sriov {
>   uint8_t active;   /**< SRIOV is active with 16, 32 or 64 
> pools */
>   uint8_t nb_q_per_pool;/**< rx queue number per pool */
> @@ -1711,6 +1722,8 @@ struct rte_eth_dev_data {
>   int numa_node;  /**< NUMA node connection */
>   struct rte_vlan_filter_conf vlan_filter_conf;
>   /**< VLAN filter configuration. */
> + uint16_t sec_id;
> + /**< security instance identifier */
>  };
> 
>  /** Device supports hotplug detach */
> @@ -1721,6 +1734,8 @@ struct rte_eth_dev_data {  #define
> RTE_ETH_DEV_BONDED_SLAVE 0x0004
>  /** Device supports device removal interrupt */
>  #define RTE_ETH_DEV_INTR_RMV  

[dpdk-dev] [PATCH 0/3] service: add attribute get and reset

[Harry van Haaren]

This patchset adds two new functions to the service cores API, allowing
the service library to expose generic attributes about the services,
and to reset them. These attributes can be the cycle counts that cores
have spent in the service-function. The 3rd patch in the series adds a
new attribute, enabling the application to retrieve the number of calls
to any service.

Harry van Haaren (3):
  service: add attribute get function
  service: add reset all attributes for service
  service: add attribute for number of invokations

 lib/librte_eal/bsdapp/eal/rte_eal_version.map   |  2 +
 lib/librte_eal/common/include/rte_service.h | 32 +
 lib/librte_eal/common/rte_service.c | 45 +++--
 lib/librte_eal/linuxapp/eal/rte_eal_version.map |  2 +
 test/test/test_service_cores.c  | 88 +
 5 files changed, 163 insertions(+), 6 deletions(-)

-- 
2.7.4

[dpdk-dev] [PATCH 1/3] service: add attribute get function

[Harry van Haaren]

This commit adds a new function to the service API to allow
the application to retrieve items about each individual service
in the system. A unit test checks the return values of a variety
of invalid and valid calls.

Signed-off-by: Harry van Haaren 
---
 lib/librte_eal/bsdapp/eal/rte_eal_version.map   |  1 +
 lib/librte_eal/common/include/rte_service.h | 17 +++
 lib/librte_eal/common/rte_service.c | 18 
 lib/librte_eal/linuxapp/eal/rte_eal_version.map |  1 +
 test/test/test_service_cores.c  | 61 +
 5 files changed, 98 insertions(+)

diff --git a/lib/librte_eal/bsdapp/eal/rte_eal_version.map 
b/lib/librte_eal/bsdapp/eal/rte_eal_version.map
index 47a09ea..8c9c5f6 100644
--- a/lib/librte_eal/bsdapp/eal/rte_eal_version.map
+++ b/lib/librte_eal/bsdapp/eal/rte_eal_version.map
@@ -212,6 +212,7 @@ EXPERIMENTAL {
rte_eal_devargs_remove;
rte_eal_hotplug_add;
rte_eal_hotplug_remove;
+   rte_service_attr_get;
rte_service_component_register;
rte_service_component_unregister;
rte_service_component_runstate_set;
diff --git a/lib/librte_eal/common/include/rte_service.h 
b/lib/librte_eal/common/include/rte_service.h
index 21da739..9267a0e 100644
--- a/lib/librte_eal/common/include/rte_service.h
+++ b/lib/librte_eal/common/include/rte_service.h
@@ -360,6 +360,23 @@ int32_t rte_service_lcore_count_services(uint32_t lcore);
  */
 int32_t rte_service_dump(FILE *f, uint32_t id);
 
+/**
+ * Returns the number of cycles that this service has consumed
+ */
+#define RTE_SERVICE_ATTR_CYCLES 0
+
+/**
+ * @warning
+ * @b EXPERIMENTAL: this API may change without prior notice
+ *
+ * Get an attribute from a service.
+ *
+ * @retval 0 Success, the attribute value has been written to *attr_value*.
+ * -EINVAL Invalid id, attr_id or attr_value was NULL.
+ */
+int32_t rte_service_attr_get(uint32_t id, uint32_t attr_id,
+uint32_t *attr_value);
+
 #ifdef __cplusplus
 }
 #endif
diff --git a/lib/librte_eal/common/rte_service.c 
b/lib/librte_eal/common/rte_service.c
index e598e16..4935804 100644
--- a/lib/librte_eal/common/rte_service.c
+++ b/lib/librte_eal/common/rte_service.c
@@ -644,6 +644,24 @@ rte_service_lcore_stop(uint32_t lcore)
return 0;
 }
 
+int32_t
+rte_service_attr_get(uint32_t id, uint32_t attr_id, uint32_t *attr_value)
+{
+   struct rte_service_spec_impl *s;
+   SERVICE_VALID_GET_OR_ERR_RET(id, s, -EINVAL);
+
+   if (!attr_value)
+   return -EINVAL;
+
+   switch (attr_id) {
+   case RTE_SERVICE_ATTR_CYCLES:
+   *attr_value = s->cycles_spent;
+   return 0;
+   default:
+   return -EINVAL;
+   }
+}
+
 static void
 rte_service_dump_one(FILE *f, struct rte_service_spec_impl *s,
 uint64_t all_cycles, uint32_t reset)
diff --git a/lib/librte_eal/linuxapp/eal/rte_eal_version.map 
b/lib/librte_eal/linuxapp/eal/rte_eal_version.map
index 8c08b8d..c2fad10 100644
--- a/lib/librte_eal/linuxapp/eal/rte_eal_version.map
+++ b/lib/librte_eal/linuxapp/eal/rte_eal_version.map
@@ -217,6 +217,7 @@ EXPERIMENTAL {
rte_eal_devargs_remove;
rte_eal_hotplug_add;
rte_eal_hotplug_remove;
+   rte_service_attr_get;
rte_service_component_register;
rte_service_component_unregister;
rte_service_component_runstate_set;
diff --git a/test/test/test_service_cores.c b/test/test/test_service_cores.c
index ee8313e..7d2e42d 100644
--- a/test/test/test_service_cores.c
+++ b/test/test/test_service_cores.c
@@ -266,6 +266,66 @@ service_name(void)
return unregister_all();
 }
 
+/* verify service attr get */
+static int
+service_attr_get(void)
+{
+   struct rte_service_spec service;
+   memset(&service, 0, sizeof(struct rte_service_spec));
+   service.callback = dummy_cb;
+   snprintf(service.name, sizeof(service.name), DUMMY_SERVICE_NAME);
+   service.capabilities |= RTE_SERVICE_CAP_MT_SAFE;
+   uint32_t id;
+   TEST_ASSERT_EQUAL(0, rte_service_component_register(&service, &id),
+   "Register of  service failed");
+   rte_service_component_runstate_set(id, 1);
+   TEST_ASSERT_EQUAL(0, rte_service_runstate_set(id, 1),
+   "Error: Service start returned non-zero");
+   rte_service_set_stats_enable(id, 1);
+
+   uint32_t attr_id = UINT32_MAX;
+   uint32_t attr_value = 0xdead;
+   /* check error return values */
+   TEST_ASSERT_EQUAL(-EINVAL, rte_service_attr_get(id, attr_id,
+   &attr_value),
+   "Invalid attr_id didn't return -EINVAL");
+
+   attr_id = RTE_SERVICE_ATTR_CYCLES;
+   TEST_ASSERT_EQUAL(-EINVAL, rte_service_attr_get(UINT32_MAX, attr_id,
+   &attr_value),
+   "Invalid service id didn't return -EINVAL");
+
+   

[dpdk-dev] [PATCH 3/3] service: add attribute for number of invokations

[Harry van Haaren]

This commit adds a new attribute to the service cores attributes
API, which allows the application to retrieve the number of times
that a service-core called the service to perform its action.

Signed-off-by: Harry van Haaren 
---
 lib/librte_eal/common/include/rte_service.h |  5 +
 lib/librte_eal/common/rte_service.c |  3 +++
 test/test/test_service_cores.c  | 16 
 3 files changed, 24 insertions(+)

diff --git a/lib/librte_eal/common/include/rte_service.h 
b/lib/librte_eal/common/include/rte_service.h
index bf7b50b..4684985 100644
--- a/lib/librte_eal/common/include/rte_service.h
+++ b/lib/librte_eal/common/include/rte_service.h
@@ -366,6 +366,11 @@ int32_t rte_service_dump(FILE *f, uint32_t id);
 #define RTE_SERVICE_ATTR_CYCLES 0
 
 /**
+ * Returns the count of invokations of this service function
+ */
+#define RTE_SERVICE_ATTR_CALL_COUNT 1
+
+/**
  * @warning
  * @b EXPERIMENTAL: this API may change without prior notice
  *
diff --git a/lib/librte_eal/common/rte_service.c 
b/lib/librte_eal/common/rte_service.c
index 3f274f4..fe0b8cf 100644
--- a/lib/librte_eal/common/rte_service.c
+++ b/lib/librte_eal/common/rte_service.c
@@ -657,6 +657,9 @@ rte_service_attr_get(uint32_t id, uint32_t attr_id, 
uint32_t *attr_value)
case RTE_SERVICE_ATTR_CYCLES:
*attr_value = s->cycles_spent;
return 0;
+   case RTE_SERVICE_ATTR_CALL_COUNT:
+   *attr_value = s->calls;
+   return 0;
default:
return -EINVAL;
}
diff --git a/test/test/test_service_cores.c b/test/test/test_service_cores.c
index dd3bf75..cabf5de 100644
--- a/test/test/test_service_cores.c
+++ b/test/test/test_service_cores.c
@@ -306,6 +306,12 @@ service_attr_get(void)
"Valid attr_get() call didn't return success");
TEST_ASSERT_EQUAL(0, attr_value,
"attr_get() call didn't set correct cycles (zero)");
+   /* check correct call count */
+   const int attr_calls = RTE_SERVICE_ATTR_CALL_COUNT;
+   TEST_ASSERT_EQUAL(0, rte_service_attr_get(id, attr_calls, &attr_value),
+   "Valid attr_get() call didn't return success");
+   TEST_ASSERT_EQUAL(0, attr_value,
+   "attr_get() call didn't get call count (zero)");
 
/* Call service to increment cycle count */
TEST_ASSERT_EQUAL(0, rte_service_lcore_add(slcore_id),
@@ -326,6 +332,11 @@ service_attr_get(void)
 
rte_service_lcore_stop(slcore_id);
 
+   TEST_ASSERT_EQUAL(0, rte_service_attr_get(id, attr_calls, &attr_value),
+   "Valid attr_get() call didn't return success");
+   TEST_ASSERT_EQUAL(1, (attr_value > 0),
+   "attr_get() call didn't get call count (zero)");
+
TEST_ASSERT_EQUAL(0, rte_service_attr_reset_all(id),
"Valid attr_reset_all() return success");
 
@@ -333,6 +344,11 @@ service_attr_get(void)
"Valid attr_get() call didn't return success");
TEST_ASSERT_EQUAL(0, attr_value,
"attr_get() call didn't set correct cycles (zero)");
+   /* ensure call count > zero */
+   TEST_ASSERT_EQUAL(0, rte_service_attr_get(id, attr_calls, &attr_value),
+   "Valid attr_get() call didn't return success");
+   TEST_ASSERT_EQUAL(0, (attr_value > 0),
+   "attr_get() call didn't get call count (zero)");
 
return unregister_all();
 }
-- 
2.7.4

[dpdk-dev] [PATCH 2/3] service: add reset all attributes for service

[Harry van Haaren]

This commit introduces a new API, allowing the application to
reset attributes of a service like the cycle count. Given this
functionality is now exposed to the user, remove the resetting
of stats during a dump() call.

Signed-off-by: Harry van Haaren 
---
 lib/librte_eal/bsdapp/eal/rte_eal_version.map   |  1 +
 lib/librte_eal/common/include/rte_service.h | 10 ++
 lib/librte_eal/common/rte_service.c | 24 ++--
 lib/librte_eal/linuxapp/eal/rte_eal_version.map |  1 +
 test/test/test_service_cores.c  | 11 +++
 5 files changed, 41 insertions(+), 6 deletions(-)

diff --git a/lib/librte_eal/bsdapp/eal/rte_eal_version.map 
b/lib/librte_eal/bsdapp/eal/rte_eal_version.map
index 8c9c5f6..e654283 100644
--- a/lib/librte_eal/bsdapp/eal/rte_eal_version.map
+++ b/lib/librte_eal/bsdapp/eal/rte_eal_version.map
@@ -213,6 +213,7 @@ EXPERIMENTAL {
rte_eal_hotplug_add;
rte_eal_hotplug_remove;
rte_service_attr_get;
+   rte_service_attr_reset_all;
rte_service_component_register;
rte_service_component_unregister;
rte_service_component_runstate_set;
diff --git a/lib/librte_eal/common/include/rte_service.h 
b/lib/librte_eal/common/include/rte_service.h
index 9267a0e..bf7b50b 100644
--- a/lib/librte_eal/common/include/rte_service.h
+++ b/lib/librte_eal/common/include/rte_service.h
@@ -377,6 +377,16 @@ int32_t rte_service_dump(FILE *f, uint32_t id);
 int32_t rte_service_attr_get(uint32_t id, uint32_t attr_id,
 uint32_t *attr_value);
 
+/**
+ * @warning
+ * @b EXPERIMENTAL: this API may change without prior notice
+ *
+ * Reset all attributes of a service.
+ * @retval 0 Successfully reset attributes
+ * -EINVAL Invalid service id provided
+ */
+int32_t rte_service_attr_reset_all(uint32_t id);
+
 #ifdef __cplusplus
 }
 #endif
diff --git a/lib/librte_eal/common/rte_service.c 
b/lib/librte_eal/common/rte_service.c
index 4935804..3f274f4 100644
--- a/lib/librte_eal/common/rte_service.c
+++ b/lib/librte_eal/common/rte_service.c
@@ -674,15 +674,27 @@ rte_service_dump_one(FILE *f, struct 
rte_service_spec_impl *s,
if (s->calls != 0)
calls = s->calls;
 
+   if (reset) {
+   s->cycles_spent = 0;
+   s->calls = 0;
+   return;
+   }
+
fprintf(f, "  %s: stats %d\tcalls %"PRIu64"\tcycles %"
PRIu64"\tavg: %"PRIu64"\n",
s->spec.name, service_stats_enabled(s), s->calls,
s->cycles_spent, s->cycles_spent / calls);
+}
 
-   if (reset) {
-   s->cycles_spent = 0;
-   s->calls = 0;
-   }
+int32_t
+rte_service_attr_reset_all(uint32_t id)
+{
+   struct rte_service_spec_impl *s;
+   SERVICE_VALID_GET_OR_ERR_RET(id, s, -EINVAL);
+
+   int reset = 1;
+   rte_service_dump_one(NULL, s, 0, reset);
+   return 0;
 }
 
 static void
@@ -730,7 +742,7 @@ int32_t rte_service_dump(FILE *f, uint32_t id)
for (i = 0; i < RTE_SERVICE_NUM_MAX; i++) {
if (!service_valid(i))
continue;
-   uint32_t reset = 1;
+   uint32_t reset = 0;
rte_service_dump_one(f, &rte_services[i], total_cycles, reset);
}
 
@@ -739,7 +751,7 @@ int32_t rte_service_dump(FILE *f, uint32_t id)
if (lcore_config[i].core_role != ROLE_SERVICE)
continue;
 
-   uint32_t reset = 1;
+   uint32_t reset = 0;
service_dump_calls_per_lcore(f, i, reset);
}
 
diff --git a/lib/librte_eal/linuxapp/eal/rte_eal_version.map 
b/lib/librte_eal/linuxapp/eal/rte_eal_version.map
index c2fad10..e5a0f04 100644
--- a/lib/librte_eal/linuxapp/eal/rte_eal_version.map
+++ b/lib/librte_eal/linuxapp/eal/rte_eal_version.map
@@ -218,6 +218,7 @@ EXPERIMENTAL {
rte_eal_hotplug_add;
rte_eal_hotplug_remove;
rte_service_attr_get;
+   rte_service_attr_reset_all;
rte_service_component_register;
rte_service_component_unregister;
rte_service_component_runstate_set;
diff --git a/test/test/test_service_cores.c b/test/test/test_service_cores.c
index 7d2e42d..dd3bf75 100644
--- a/test/test/test_service_cores.c
+++ b/test/test/test_service_cores.c
@@ -270,6 +270,9 @@ service_name(void)
 static int
 service_attr_get(void)
 {
+   /* ensure all services unregistered so cycle counts are zero */
+   unregister_all();
+
struct rte_service_spec service;
memset(&service, 0, sizeof(struct rte_service_spec));
service.callback = dummy_cb;
@@ -323,6 +326,14 @@ service_attr_get(void)
 
rte_service_lcore_stop(slcore_id);
 
+   TEST_ASSERT_EQUAL(0, rte_service_attr_reset_all(id),
+   "Valid attr_reset_all() return success");
+
+   TEST_ASSERT_EQUAL(0, rte_service_attr_get(id, attr_id, &attr_value),
+   "V

Re: [dpdk-dev] [PATCH v4 4/4] eventdev: Add tests for event eth Rx adapter APIs

[Rao, Nikhil]

On 10/2/2017 4:01 PM, Jerin Jacob wrote:

-Original Message-

Date: Sun, 24 Sep 2017 23:54:38 +0530
From: "Rao, Nikhil" 
To: Jerin Jacob 
CC: bruce.richard...@intel.com, gage.e...@intel.com, dev@dpdk.org,
  tho...@monjalon.net, harry.van.haa...@intel.com, hemant.agra...@nxp.com,
  nipun.gu...@nxp.com, narender.vang...@intel.com,
  erik.g.carri...@intel.com, abhinandan.guj...@intel.com,
  santosh.shu...@caviumnetworks.com
Subject: Re: [PATCH v4 4/4] eventdev: Add tests for event eth Rx adapter
  APIs
User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:52.0) Gecko/20100101
  Thunderbird/52.3.0

On 9/22/2017 5:42 PM, Jerin Jacob wrote:

-Original Message-

Date: Fri, 22 Sep 2017 02:47:14 +0530
From: Nikhil Rao 
To: jerin.ja...@caviumnetworks.com, bruce.richard...@intel.com
CC: gage.e...@intel.com, dev@dpdk.org, tho...@monjalon.net,
   harry.van.haa...@intel.com, hemant.agra...@nxp.com, nipun.gu...@nxp.com,
   narender.vang...@intel.com, erik.g.carri...@intel.com,
   abhinandan.guj...@intel.com, santosh.shu...@caviumnetworks.com
Subject: [PATCH v4 4/4] eventdev: Add tests for event eth Rx adapter APIs
X-Mailer: git-send-email 2.7.4

Add unit tests for rte_event_eth_rx_adapter_xxx() APIs






+#include 
+#include 
+#include 
+#include 
+#include 
+#include 
+
+#include 
+
+#include "test.h"
+
+/* i40e limits max to 64 */


This comment could be removed.


OK, I am documenting why the code doesn't just use dev_info.max_rx_queues,
won't the comment be useful to retain ?


OK. If dev_info.max_rx_queues for i40e is not 64 as expected then we
could the fix the i40e driver as well.


This change was added in c9eb97fb9212a38, will check with i40e PMD 
maintainer (this is a holiday week in PRC though), a quick check showed 
that it is possible to allocate dev_info.max_rx_queues - 
dev_info.vmdq_queue_num Rx queues


I will re post with the comment removed for now and post a patch later.

Nikhil

[dpdk-dev] [PATCH v4 1/8] app/crypto-perf: refactor common test code

[Pablo de Lara]

Currently, there is some duplication in all the test types,
in the crypto performance application.

In order to improve maintainability of this code,
and ease future work on it, common functions have been separated
in a different file that gets included in all the tests.

Signed-off-by: Pablo de Lara 
---
 app/test-crypto-perf/Makefile|   1 +
 app/test-crypto-perf/cperf_test_common.c | 234 +++
 app/test-crypto-perf/cperf_test_common.h |  61 ++
 app/test-crypto-perf/cperf_test_latency.c| 199 ++-
 app/test-crypto-perf/cperf_test_pmd_cyclecount.c | 188 ++
 app/test-crypto-perf/cperf_test_throughput.c | 194 ++-
 app/test-crypto-perf/cperf_test_verify.c | 191 ++
 7 files changed, 351 insertions(+), 717 deletions(-)
 create mode 100644 app/test-crypto-perf/cperf_test_common.c
 create mode 100644 app/test-crypto-perf/cperf_test_common.h

diff --git a/app/test-crypto-perf/Makefile b/app/test-crypto-perf/Makefile
index 821e8e5..25ae395 100644
--- a/app/test-crypto-perf/Makefile
+++ b/app/test-crypto-perf/Makefile
@@ -45,5 +45,6 @@ SRCS-y += cperf_test_latency.c
 SRCS-y += cperf_test_pmd_cyclecount.c
 SRCS-y += cperf_test_verify.c
 SRCS-y += cperf_test_vector_parsing.c
+SRCS-y += cperf_test_common.c
 
 include $(RTE_SDK)/mk/rte.app.mk
diff --git a/app/test-crypto-perf/cperf_test_common.c 
b/app/test-crypto-perf/cperf_test_common.c
new file mode 100644
index 000..a87d27e
--- /dev/null
+++ b/app/test-crypto-perf/cperf_test_common.c
@@ -0,0 +1,234 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) 2017 Intel Corporation. All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ *   notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ *   notice, this list of conditions and the following disclaimer in
+ *   the documentation and/or other materials provided with the
+ *   distribution.
+ * * Neither the name of Intel Corporation nor the names of its
+ *   contributors may be used to endorse or promote products derived
+ *   from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include 
+
+#include "cperf_test_common.h"
+
+static struct rte_mbuf *
+cperf_mbuf_create(struct rte_mempool *mempool,
+   uint32_t segments_nb,
+   const struct cperf_options *options,
+   const struct cperf_test_vector *test_vector)
+{
+   struct rte_mbuf *mbuf;
+   uint32_t segment_sz = options->max_buffer_size / segments_nb;
+   uint32_t last_sz = options->max_buffer_size % segments_nb;
+   uint8_t *mbuf_data;
+   uint8_t *test_data =
+   (options->cipher_op == RTE_CRYPTO_CIPHER_OP_ENCRYPT) ?
+   test_vector->plaintext.data :
+   test_vector->ciphertext.data;
+
+   mbuf = rte_pktmbuf_alloc(mempool);
+   if (mbuf == NULL)
+   goto error;
+
+   mbuf_data = (uint8_t *)rte_pktmbuf_append(mbuf, segment_sz);
+   if (mbuf_data == NULL)
+   goto error;
+
+   memcpy(mbuf_data, test_data, segment_sz);
+   test_data += segment_sz;
+   segments_nb--;
+
+   while (segments_nb) {
+   struct rte_mbuf *m;
+
+   m = rte_pktmbuf_alloc(mempool);
+   if (m == NULL)
+   goto error;
+
+   rte_pktmbuf_chain(mbuf, m);
+
+   mbuf_data = (uint8_t *)rte_pktmbuf_append(mbuf, segment_sz);
+   if (mbuf_data == NULL)
+   goto error;
+
+   memcpy(mbuf_data, test_data, segment_sz);
+   test_data += segment_sz;
+   segments_nb--;
+   }
+
+   if (last_sz) {
+   mbuf_data = (uint8_t *)rte_pktmbuf_append(mbuf, last_sz);
+  

[dpdk-dev] [PATCH v4 0/8] Crypto-perf app improvements

[Pablo de Lara]

This patchset includes some improvements in the Crypto
performance application, including app fixes and new parameter additions.

The last patch, in particular, introduces performance improvements.
Currently, crypto operations are allocated in a mempool and mbufs
in a different one. Then mbufs are extracted to an array,
which is looped through for all the crypto operations,
impacting greatly the performance, as much more memory is used.

Since crypto operations and mbufs are mapped 1:1, the can share
the same mempool object (similar to having the mbuf in the
private data of the crypto operation).
This improves performance, as it is only required to handle
a single mempool and the mbufs are obtained from the cache
of the mempoool, and not from an static array.

Changes in v4:
- Refactored test code, to minimize duplications
- Removed --qp-nb parameter. Now the number of queue pairs
  per device are calculated from the number of logical cores
  available and the number of crypto devices

Changes in v3:
- Renamed "number of queue pairs" option from "--qps" to "--qp-nb",
  for more consistency

Changes in v2:

- Added support for multiple queue pairs
- Mempool for crypto operations and mbufs is now created
  using rte_mempool_create_empty(), rte_mempool_set_ops_byname(),
  rte_mempool_populate_default() and rte_mempool_obj_iter(),
  so mempool handler is set, as per Akhil's request.

Pablo de Lara (8):
  app/crypto-perf: refactor common test code
  app/crypto-perf: set AAD after the crypto operation
  app/crypto-perf: parse AEAD data from vectors
  app/crypto-perf: parse segment size
  app/crypto-perf: overwrite mbuf when verifying
  app/crypto-perf: do not populate the mbufs at init
  app/crypto-perf: support multiple queue pairs
  app/crypto-perf: use single mempool

 app/test-crypto-perf/Makefile|   5 +
 app/test-crypto-perf/cperf_ops.c | 136 ---
 app/test-crypto-perf/cperf_ops.h |   2 +-
 app/test-crypto-perf/cperf_options.h |   5 +-
 app/test-crypto-perf/cperf_options_parsing.c |  47 ++--
 app/test-crypto-perf/cperf_test_common.c | 225 ++
 app/test-crypto-perf/cperf_test_common.h |  52 +
 app/test-crypto-perf/cperf_test_latency.c| 239 +++
 app/test-crypto-perf/cperf_test_pmd_cyclecount.c | 239 ---
 app/test-crypto-perf/cperf_test_throughput.c | 237 +++
 app/test-crypto-perf/cperf_test_vector_parsing.c |  55 +
 app/test-crypto-perf/cperf_test_verify.c | 278 ++-
 app/test-crypto-perf/main.c  | 100 +---
 doc/guides/tools/cryptoperf.rst  |   6 +-
 14 files changed, 715 insertions(+), 911 deletions(-)
 create mode 100644 app/test-crypto-perf/cperf_test_common.c
 create mode 100644 app/test-crypto-perf/cperf_test_common.h

-- 
2.9.4

[dpdk-dev] [PATCH v4 2/8] app/crypto-perf: set AAD after the crypto operation

[Pablo de Lara]

Instead of prepending the AAD (Additional Authenticated Data)
in the mbuf, it is easier to set after the crypto operation,
as it is a read-only value, like the IV, and then it is not
restricted to the size of the mbuf headroom.

Signed-off-by: Pablo de Lara 
---
 app/test-crypto-perf/cperf_ops.c | 16 
 app/test-crypto-perf/cperf_test_common.c | 15 +++
 app/test-crypto-perf/cperf_test_verify.c |  4 ++--
 3 files changed, 17 insertions(+), 18 deletions(-)

diff --git a/app/test-crypto-perf/cperf_ops.c b/app/test-crypto-perf/cperf_ops.c
index 88fb972..5be20d9 100644
--- a/app/test-crypto-perf/cperf_ops.c
+++ b/app/test-crypto-perf/cperf_ops.c
@@ -307,6 +307,8 @@ cperf_set_ops_aead(struct rte_crypto_op **ops,
uint16_t iv_offset)
 {
uint16_t i;
+   uint16_t aad_offset = iv_offset +
+   RTE_ALIGN_CEIL(test_vector->aead_iv.length, 16);
 
for (i = 0; i < nb_ops; i++) {
struct rte_crypto_sym_op *sym_op = ops[i]->sym;
@@ -318,11 +320,12 @@ cperf_set_ops_aead(struct rte_crypto_op **ops,
 
/* AEAD parameters */
sym_op->aead.data.length = options->test_buffer_size;
-   sym_op->aead.data.offset =
-   RTE_ALIGN_CEIL(options->aead_aad_sz, 16);
+   sym_op->aead.data.offset = 0;
 
-   sym_op->aead.aad.data = rte_pktmbuf_mtod(bufs_in[i], uint8_t *);
-   sym_op->aead.aad.phys_addr = rte_pktmbuf_mtophys(bufs_in[i]);
+   sym_op->aead.aad.data = rte_crypto_op_ctod_offset(ops[i],
+   uint8_t *, aad_offset);
+   sym_op->aead.aad.phys_addr = 
rte_crypto_op_ctophys_offset(ops[i],
+   aad_offset);
 
if (options->aead_op == RTE_CRYPTO_AEAD_OP_DECRYPT) {
sym_op->aead.digest.data = test_vector->digest.data;
@@ -360,6 +363,11 @@ cperf_set_ops_aead(struct rte_crypto_op **ops,
 
memcpy(iv_ptr, test_vector->aead_iv.data,
test_vector->aead_iv.length);
+
+   /* Copy AAD after the IV */
+   memcpy(ops[i]->sym->aead.aad.data,
+   test_vector->aad.data,
+   test_vector->aad.length);
}
}
 
diff --git a/app/test-crypto-perf/cperf_test_common.c 
b/app/test-crypto-perf/cperf_test_common.c
index a87d27e..ddf5641 100644
--- a/app/test-crypto-perf/cperf_test_common.c
+++ b/app/test-crypto-perf/cperf_test_common.c
@@ -94,16 +94,6 @@ cperf_mbuf_create(struct rte_mempool *mempool,
goto error;
}
 
-   if (options->op_type == CPERF_AEAD) {
-   uint8_t *aead = (uint8_t *)rte_pktmbuf_prepend(mbuf,
-   RTE_ALIGN_CEIL(options->aead_aad_sz, 16));
-
-   if (aead == NULL)
-   goto error;
-
-   memcpy(aead, test_vector->aad.data, test_vector->aad.length);
-   }
-
return mbuf;
 error:
if (mbuf != NULL)
@@ -183,9 +173,10 @@ cperf_alloc_common_memory(const struct cperf_options 
*options,
snprintf(pool_name, sizeof(pool_name), "cperf_op_pool_cdev_%d",
dev_id);
 
-   uint16_t priv_size = test_vector->cipher_iv.length +
+   uint16_t priv_size = RTE_ALIGN_CEIL(test_vector->cipher_iv.length +
test_vector->auth_iv.length + test_vector->aead_iv.length +
-   extra_op_priv_size;
+   extra_op_priv_size, 16) +
+   RTE_ALIGN_CEIL(options->aead_aad_sz, 16);
 
*crypto_op_pool = rte_crypto_op_pool_create(pool_name,
RTE_CRYPTO_OP_TYPE_SYMMETRIC, options->pool_sz,
diff --git a/app/test-crypto-perf/cperf_test_verify.c 
b/app/test-crypto-perf/cperf_test_verify.c
index d05124a..82e5e9f 100644
--- a/app/test-crypto-perf/cperf_test_verify.c
+++ b/app/test-crypto-perf/cperf_test_verify.c
@@ -197,9 +197,9 @@ cperf_verify_op(struct rte_crypto_op *op,
break;
case CPERF_AEAD:
cipher = 1;
-   cipher_offset = vector->aad.length;
+   cipher_offset = 0;
auth = 1;
-   auth_offset = vector->aad.length + options->test_buffer_size;
+   auth_offset = options->test_buffer_size;
break;
}
 
-- 
2.9.4

[dpdk-dev] [PATCH v4 3/8] app/crypto-perf: parse AEAD data from vectors

[Pablo de Lara]

Since DPDK 17.08, there is specific parameters
for AEAD algorithm, like AES-GCM. When verifying
crypto operations with test vectors, the parser
was not reading AEAD data (such as IV or key).

Fixes: 8a5b494a7f99 ("app/test-crypto-perf: add AEAD parameters")
Cc: sta...@dpdk.org

Signed-off-by: Pablo de Lara 
---
 app/test-crypto-perf/cperf_test_vector_parsing.c | 55 
 1 file changed, 55 insertions(+)

diff --git a/app/test-crypto-perf/cperf_test_vector_parsing.c 
b/app/test-crypto-perf/cperf_test_vector_parsing.c
index 148a604..3952632 100644
--- a/app/test-crypto-perf/cperf_test_vector_parsing.c
+++ b/app/test-crypto-perf/cperf_test_vector_parsing.c
@@ -116,6 +116,20 @@ show_test_vector(struct cperf_test_vector *test_vector)
printf("\n");
}
 
+   if (test_vector->aead_key.data) {
+   printf("\naead_key =\n");
+   for (i = 0; i < test_vector->aead_key.length; ++i) {
+   if ((i % wrap == 0) && (i != 0))
+   printf("\n");
+   if (i == (uint32_t)(test_vector->aead_key.length - 1))
+   printf("0x%02x", test_vector->aead_key.data[i]);
+   else
+   printf("0x%02x, ",
+   test_vector->aead_key.data[i]);
+   }
+   printf("\n");
+   }
+
if (test_vector->cipher_iv.data) {
printf("\ncipher_iv =\n");
for (i = 0; i < test_vector->cipher_iv.length; ++i) {
@@ -142,6 +156,19 @@ show_test_vector(struct cperf_test_vector *test_vector)
printf("\n");
}
 
+   if (test_vector->aead_iv.data) {
+   printf("\naead_iv =\n");
+   for (i = 0; i < test_vector->aead_iv.length; ++i) {
+   if ((i % wrap == 0) && (i != 0))
+   printf("\n");
+   if (i == (uint32_t)(test_vector->aead_iv.length - 1))
+   printf("0x%02x", test_vector->aead_iv.data[i]);
+   else
+   printf("0x%02x, ", 
test_vector->aead_iv.data[i]);
+   }
+   printf("\n");
+   }
+
if (test_vector->ciphertext.data) {
printf("\nciphertext =\n");
for (i = 0; i < test_vector->ciphertext.length; ++i) {
@@ -345,6 +372,20 @@ parse_entry(char *entry, struct cperf_test_vector *vector,
vector->auth_key.length = opts->auth_key_sz;
}
 
+   } else if (strstr(key_token, "aead_key")) {
+   rte_free(vector->aead_key.data);
+   vector->aead_key.data = data;
+   if (tc_found)
+   vector->aead_key.length = data_length;
+   else {
+   if (opts->aead_key_sz > data_length) {
+   printf("Global aead_key shorter than "
+   "aead_key_sz\n");
+   return -1;
+   }
+   vector->aead_key.length = opts->aead_key_sz;
+   }
+
} else if (strstr(key_token, "cipher_iv")) {
rte_free(vector->cipher_iv.data);
vector->cipher_iv.data = data;
@@ -373,6 +414,20 @@ parse_entry(char *entry, struct cperf_test_vector *vector,
vector->auth_iv.length = opts->auth_iv_sz;
}
 
+   } else if (strstr(key_token, "aead_iv")) {
+   rte_free(vector->aead_iv.data);
+   vector->aead_iv.data = data;
+   if (tc_found)
+   vector->aead_iv.length = data_length;
+   else {
+   if (opts->aead_iv_sz > data_length) {
+   printf("Global aead iv shorter than "
+   "aead_iv_sz\n");
+   return -1;
+   }
+   vector->aead_iv.length = opts->aead_iv_sz;
+   }
+
} else if (strstr(key_token, "ciphertext")) {
rte_free(vector->ciphertext.data);
vector->ciphertext.data = data;
-- 
2.9.4

[dpdk-dev] [PATCH v4 6/8] app/crypto-perf: do not populate the mbufs at init

[Pablo de Lara]

For throughput and latency tests, it is not required
to populate the mbufs with any test vector.
For verify test, there is already a function that rewrites
the mbufs every time they are going to be used with
crypto operations.

Signed-off-by: Pablo de Lara 
---
 app/test-crypto-perf/cperf_test_common.c | 31 +--
 1 file changed, 9 insertions(+), 22 deletions(-)

diff --git a/app/test-crypto-perf/cperf_test_common.c 
b/app/test-crypto-perf/cperf_test_common.c
index 0f62a2c..25eb970 100644
--- a/app/test-crypto-perf/cperf_test_common.c
+++ b/app/test-crypto-perf/cperf_test_common.c
@@ -38,15 +38,10 @@ static struct rte_mbuf *
 cperf_mbuf_create(struct rte_mempool *mempool,
uint32_t segment_sz,
uint32_t segments_nb,
-   const struct cperf_options *options,
-   const struct cperf_test_vector *test_vector)
+   const struct cperf_options *options)
 {
struct rte_mbuf *mbuf;
uint8_t *mbuf_data;
-   uint8_t *test_data =
-   (options->cipher_op == RTE_CRYPTO_CIPHER_OP_ENCRYPT) ?
-   test_vector->plaintext.data :
-   test_vector->ciphertext.data;
uint32_t remaining_bytes = options->max_buffer_size;
 
mbuf = rte_pktmbuf_alloc(mempool);
@@ -57,15 +52,11 @@ cperf_mbuf_create(struct rte_mempool *mempool,
if (mbuf_data == NULL)
goto error;
 
-   if (options->max_buffer_size <= segment_sz) {
-   memcpy(mbuf_data, test_data, options->max_buffer_size);
-   test_data += options->max_buffer_size;
+   if (options->max_buffer_size <= segment_sz)
remaining_bytes = 0;
-   } else {
-   memcpy(mbuf_data, test_data, segment_sz);
-   test_data += segment_sz;
+   else
remaining_bytes -= segment_sz;
-   }
+
segments_nb--;
 
while (remaining_bytes) {
@@ -81,15 +72,11 @@ cperf_mbuf_create(struct rte_mempool *mempool,
if (mbuf_data == NULL)
goto error;
 
-   if (remaining_bytes <= segment_sz) {
-   memcpy(mbuf_data, test_data, remaining_bytes);
+   if (remaining_bytes <= segment_sz)
remaining_bytes = 0;
-   test_data += remaining_bytes;
-   } else {
-   memcpy(mbuf_data, test_data, segment_sz);
+   else
remaining_bytes -= segment_sz;
-   test_data += segment_sz;
-   }
+
segments_nb--;
}
 
@@ -156,7 +143,7 @@ cperf_alloc_common_memory(const struct cperf_options 
*options,
*pkt_mbuf_pool_in,
options->segment_sz,
segments_nb,
-   options, test_vector);
+   options);
if ((*mbufs_in)[mbuf_idx] == NULL)
return -1;
}
@@ -180,7 +167,7 @@ cperf_alloc_common_memory(const struct cperf_options 
*options,
for (mbuf_idx = 0; mbuf_idx < options->pool_sz; mbuf_idx++) {
(*mbufs_out)[mbuf_idx] = cperf_mbuf_create(
*pkt_mbuf_pool_out, max_size,
-   1, options, test_vector);
+   1, options);
if ((*mbufs_out)[mbuf_idx] == NULL)
return -1;
}
-- 
2.9.4

[dpdk-dev] [PATCH v4 5/8] app/crypto-perf: overwrite mbuf when verifying

[Pablo de Lara]

When running the verify test, mbufs in the pool were
populated with the test vector loaded from a file.
To avoid limiting the number of operations to the pool size,
mbufs will be rewritten with the test vector, before
linking them to the crypto operations.

Signed-off-by: Pablo de Lara 
---
 app/test-crypto-perf/cperf_options_parsing.c |  7 --
 app/test-crypto-perf/cperf_test_verify.c | 35 
 2 files changed, 35 insertions(+), 7 deletions(-)

diff --git a/app/test-crypto-perf/cperf_options_parsing.c 
b/app/test-crypto-perf/cperf_options_parsing.c
index d372691..89f86a2 100644
--- a/app/test-crypto-perf/cperf_options_parsing.c
+++ b/app/test-crypto-perf/cperf_options_parsing.c
@@ -944,13 +944,6 @@ cperf_options_check(struct cperf_options *options)
}
 
if (options->test == CPERF_TEST_TYPE_VERIFY &&
-   options->total_ops > options->pool_sz) {
-   RTE_LOG(ERR, USER1, "Total number of ops must be less than or"
-   " equal to the pool size.\n");
-   return -EINVAL;
-   }
-
-   if (options->test == CPERF_TEST_TYPE_VERIFY &&
(options->inc_buffer_size != 0 ||
options->buffer_size_count > 1)) {
RTE_LOG(ERR, USER1, "Only one buffer size is allowed when "
diff --git a/app/test-crypto-perf/cperf_test_verify.c 
b/app/test-crypto-perf/cperf_test_verify.c
index 7a85aa7..9a806fa 100644
--- a/app/test-crypto-perf/cperf_test_verify.c
+++ b/app/test-crypto-perf/cperf_test_verify.c
@@ -224,6 +224,34 @@ cperf_verify_op(struct rte_crypto_op *op,
return !!res;
 }
 
+static void
+cperf_mbuf_set(struct rte_mbuf *mbuf,
+   const struct cperf_options *options,
+   const struct cperf_test_vector *test_vector)
+{
+   uint32_t segment_sz = options->segment_sz;
+   uint8_t *mbuf_data;
+   uint8_t *test_data =
+   (options->cipher_op == RTE_CRYPTO_CIPHER_OP_ENCRYPT) ?
+   test_vector->plaintext.data :
+   test_vector->ciphertext.data;
+   uint32_t remaining_bytes = options->max_buffer_size;
+
+   while (remaining_bytes) {
+   mbuf_data = rte_pktmbuf_mtod(mbuf, uint8_t *);
+
+   if (remaining_bytes <= segment_sz) {
+   memcpy(mbuf_data, test_data, remaining_bytes);
+   return;
+   }
+
+   memcpy(mbuf_data, test_data, segment_sz);
+   remaining_bytes -= segment_sz;
+   test_data += segment_sz;
+   mbuf = mbuf->next;
+   }
+}
+
 int
 cperf_verify_test_runner(void *test_ctx)
 {
@@ -294,6 +322,13 @@ cperf_verify_test_runner(void *test_ctx)
ops_needed, ctx->sess, ctx->options,
ctx->test_vector, iv_offset);
 
+
+   /* Populate the mbuf with the test vector, for verification */
+   for (i = 0; i < ops_needed; i++)
+   cperf_mbuf_set(ops[i]->sym->m_src,
+   ctx->options,
+   ctx->test_vector);
+
 #ifdef CPERF_LINEARIZATION_ENABLE
if (linearize) {
/* PMD doesn't support scatter-gather and source buffer
-- 
2.9.4

[dpdk-dev] [PATCH v4 4/8] app/crypto-perf: parse segment size

[Pablo de Lara]

Instead of parsing number of segments, from the command line,
parse segment size, as it is a more usual case to have
the segment size fixed and then different packet sizes
will require different number of segments.

Signed-off-by: Pablo de Lara 
---
 app/test-crypto-perf/cperf_ops.c | 24 
 app/test-crypto-perf/cperf_options.h |  4 +-
 app/test-crypto-perf/cperf_options_parsing.c | 38 
 app/test-crypto-perf/cperf_test_common.c | 74 ++--
 app/test-crypto-perf/cperf_test_latency.c|  2 +-
 app/test-crypto-perf/cperf_test_pmd_cyclecount.c |  2 +-
 app/test-crypto-perf/cperf_test_throughput.c |  2 +-
 app/test-crypto-perf/cperf_test_verify.c |  2 +-
 doc/guides/tools/cryptoperf.rst  |  6 +-
 9 files changed, 106 insertions(+), 48 deletions(-)

diff --git a/app/test-crypto-perf/cperf_ops.c b/app/test-crypto-perf/cperf_ops.c
index 5be20d9..ad32065 100644
--- a/app/test-crypto-perf/cperf_ops.c
+++ b/app/test-crypto-perf/cperf_ops.c
@@ -175,6 +175,14 @@ cperf_set_ops_auth(struct rte_crypto_op **ops,
offset -= tbuf->data_len;
tbuf = tbuf->next;
}
+   /*
+* If there is not enough room in segment,
+* place the digest in the next segment
+*/
+   if ((tbuf->data_len - offset) < 
options->digest_sz) {
+   tbuf = tbuf->next;
+   offset = 0;
+   }
buf = tbuf;
}
 
@@ -256,6 +264,14 @@ cperf_set_ops_cipher_auth(struct rte_crypto_op **ops,
offset -= tbuf->data_len;
tbuf = tbuf->next;
}
+   /*
+* If there is not enough room in segment,
+* place the digest in the next segment
+*/
+   if ((tbuf->data_len - offset) < 
options->digest_sz) {
+   tbuf = tbuf->next;
+   offset = 0;
+   }
buf = tbuf;
}
 
@@ -346,6 +362,14 @@ cperf_set_ops_aead(struct rte_crypto_op **ops,
offset -= tbuf->data_len;
tbuf = tbuf->next;
}
+   /*
+* If there is not enough room in segment,
+* place the digest in the next segment
+*/
+   if ((tbuf->data_len - offset) < 
options->digest_sz) {
+   tbuf = tbuf->next;
+   offset = 0;
+   }
buf = tbuf;
}
 
diff --git a/app/test-crypto-perf/cperf_options.h 
b/app/test-crypto-perf/cperf_options.h
index 2f42cb6..6d339f4 100644
--- a/app/test-crypto-perf/cperf_options.h
+++ b/app/test-crypto-perf/cperf_options.h
@@ -11,7 +11,7 @@
 #define CPERF_TOTAL_OPS("total-ops")
 #define CPERF_BURST_SIZE   ("burst-sz")
 #define CPERF_BUFFER_SIZE  ("buffer-sz")
-#define CPERF_SEGMENTS_NB  ("segments-nb")
+#define CPERF_SEGMENT_SIZE ("segment-sz")
 #define CPERF_DESC_NB  ("desc-nb")
 
 #define CPERF_DEVTYPE  ("devtype")
@@ -71,7 +71,7 @@ struct cperf_options {
 
uint32_t pool_sz;
uint32_t total_ops;
-   uint32_t segments_nb;
+   uint32_t segment_sz;
uint32_t test_buffer_size;
uint32_t nb_descriptors;
 
diff --git a/app/test-crypto-perf/cperf_options_parsing.c 
b/app/test-crypto-perf/cperf_options_parsing.c
index f3508a4..d372691 100644
--- a/app/test-crypto-perf/cperf_options_parsing.c
+++ b/app/test-crypto-perf/cperf_options_parsing.c
@@ -328,17 +328,17 @@ parse_buffer_sz(struct cperf_options *opts, const char 
*arg)
 }
 
 static int
-parse_segments_nb(struct cperf_options *opts, const char *arg)
+parse_segment_sz(struct cperf_options *opts, const char *arg)
 {
-   int ret = parse_uint32_t(&opts->segments_nb, arg);
+   int ret = parse_uint32_t(&opts->segment_sz, arg);
 
if (ret) {
-   RTE_LOG(ERR, USER1, "failed to parse segments number\n");
+   RTE_LOG(ERR, USER1, "failed to parse segment size\n");
return -1;
}
 
-   if ((opts->segments_nb == 0) || (opts->segments_nb > 255)) {
-   RTE_LOG(ERR, USER1, "invalid segment

[dpdk-dev] [PATCH v4 7/8] app/crypto-perf: support multiple queue pairs

[Pablo de Lara]

Add support for multiple queue pairs, when there are
more logical cores available than crypto devices enabled.
For instance, if there are 4 cores available and
2 crypto devices, each device will have two queue pairs.

This is useful to have multiple logical cores using
a single crypto device, without needing to initialize
a crypto device per core.

Signed-off-by: Pablo de Lara 
---
 app/test-crypto-perf/Makefile|   4 +
 app/test-crypto-perf/cperf_options.h |   1 +
 app/test-crypto-perf/cperf_options_parsing.c |   2 +
 app/test-crypto-perf/cperf_test_common.c |  15 ++--
 app/test-crypto-perf/cperf_test_common.h |   3 +-
 app/test-crypto-perf/cperf_test_latency.c|   4 +-
 app/test-crypto-perf/cperf_test_pmd_cyclecount.c |   2 +-
 app/test-crypto-perf/cperf_test_throughput.c |   4 +-
 app/test-crypto-perf/cperf_test_verify.c |   4 +-
 app/test-crypto-perf/main.c  | 100 +--
 10 files changed, 95 insertions(+), 44 deletions(-)

diff --git a/app/test-crypto-perf/Makefile b/app/test-crypto-perf/Makefile
index 25ae395..c75d7ed 100644
--- a/app/test-crypto-perf/Makefile
+++ b/app/test-crypto-perf/Makefile
@@ -47,4 +47,8 @@ SRCS-y += cperf_test_verify.c
 SRCS-y += cperf_test_vector_parsing.c
 SRCS-y += cperf_test_common.c
 
+ifeq ($(CONFIG_RTE_LIBRTE_PMD_CRYPTO_SCHEDULER),y)
+LDLIBS += -lrte_pmd_crypto_scheduler
+endif
+
 include $(RTE_SDK)/mk/rte.app.mk
diff --git a/app/test-crypto-perf/cperf_options.h 
b/app/test-crypto-perf/cperf_options.h
index 6d339f4..da4fb47 100644
--- a/app/test-crypto-perf/cperf_options.h
+++ b/app/test-crypto-perf/cperf_options.h
@@ -74,6 +74,7 @@ struct cperf_options {
uint32_t segment_sz;
uint32_t test_buffer_size;
uint32_t nb_descriptors;
+   uint16_t nb_qps;
 
uint32_t sessionless:1;
uint32_t out_of_place:1;
diff --git a/app/test-crypto-perf/cperf_options_parsing.c 
b/app/test-crypto-perf/cperf_options_parsing.c
index 89f86a2..e5fa4a8 100644
--- a/app/test-crypto-perf/cperf_options_parsing.c
+++ b/app/test-crypto-perf/cperf_options_parsing.c
@@ -747,6 +747,7 @@ cperf_options_default(struct cperf_options *opts)
 
strncpy(opts->device_type, "crypto_aesni_mb",
sizeof(opts->device_type));
+   opts->nb_qps = 1;
 
opts->op_type = CPERF_CIPHER_THEN_AUTH;
 
@@ -1032,6 +1033,7 @@ cperf_options_dump(struct cperf_options *opts)
printf("#\n");
printf("# cryptodev type: %s\n", opts->device_type);
printf("#\n");
+   printf("# number of queue pairs per device: %u\n", opts->nb_qps);
printf("# crypto operation: %s\n", cperf_op_type_strs[opts->op_type]);
printf("# sessionless: %s\n", opts->sessionless ? "yes" : "no");
printf("# out of place: %s\n", opts->out_of_place ? "yes" : "no");
diff --git a/app/test-crypto-perf/cperf_test_common.c 
b/app/test-crypto-perf/cperf_test_common.c
index 25eb970..65f0be7 100644
--- a/app/test-crypto-perf/cperf_test_common.c
+++ b/app/test-crypto-perf/cperf_test_common.c
@@ -108,7 +108,8 @@ cperf_mbuf_create(struct rte_mempool *mempool,
 int
 cperf_alloc_common_memory(const struct cperf_options *options,
const struct cperf_test_vector *test_vector,
-   uint8_t dev_id, size_t extra_op_priv_size,
+   uint8_t dev_id, uint16_t qp_id,
+   size_t extra_op_priv_size,
struct rte_mempool **pkt_mbuf_pool_in,
struct rte_mempool **pkt_mbuf_pool_out,
struct rte_mbuf ***mbufs_in,
@@ -118,8 +119,8 @@ cperf_alloc_common_memory(const struct cperf_options 
*options,
unsigned int mbuf_idx = 0;
char pool_name[32] = "";
 
-   snprintf(pool_name, sizeof(pool_name), "cperf_pool_in_cdev_%d",
-   dev_id);
+   snprintf(pool_name, sizeof(pool_name), "cperf_pool_in_cdev_%u_qp_%u",
+   dev_id, qp_id);
 
uint32_t max_size = options->max_buffer_size + options->digest_sz;
uint16_t segments_nb = (max_size % options->segment_sz) ?
@@ -153,8 +154,8 @@ cperf_alloc_common_memory(const struct cperf_options 
*options,
options->pool_sz), 0);
 
if (options->out_of_place == 1) {
-   snprintf(pool_name, sizeof(pool_name), "cperf_pool_out_cdev_%d",
-   dev_id);
+   snprintf(pool_name, sizeof(pool_name), 
"cperf_pool_out_cdev_%u_qp_%u",
+   dev_id, qp_id);
 
*pkt_mbuf_pool_out = rte_pktmbuf_pool_create(
pool_name, options->pool_sz, 0, 0,
@@ -173,8 +174,8 @@ cperf_alloc_common_memory(const struct cperf_options 
*options,
}
}
 
-   snprintf(pool_name, sizeof(pool_name), "cperf_op_pool_cdev_%d",
-   dev_id);
+   snprintf(pool_name, s

[dpdk-dev] [PATCH v4 8/8] app/crypto-perf: use single mempool

[Pablo de Lara]

In order to improve memory utilization, a single mempool
is created, containing the crypto operation and mbufs
(one if operation is in-place, two if out-of-place).
This way, a single object is allocated and freed
per operation, reducing the amount of memory in cache,
which improves scalability.

Signed-off-by: Pablo de Lara 
---
 app/test-crypto-perf/cperf_ops.c |  96 +--
 app/test-crypto-perf/cperf_ops.h |   2 +-
 app/test-crypto-perf/cperf_test_common.c | 322 +++
 app/test-crypto-perf/cperf_test_common.h |  16 +-
 app/test-crypto-perf/cperf_test_latency.c|  52 ++--
 app/test-crypto-perf/cperf_test_pmd_cyclecount.c |  67 ++---
 app/test-crypto-perf/cperf_test_throughput.c |  55 ++--
 app/test-crypto-perf/cperf_test_verify.c |  60 ++---
 8 files changed, 331 insertions(+), 339 deletions(-)

diff --git a/app/test-crypto-perf/cperf_ops.c b/app/test-crypto-perf/cperf_ops.c
index ad32065..f76dbdd 100644
--- a/app/test-crypto-perf/cperf_ops.c
+++ b/app/test-crypto-perf/cperf_ops.c
@@ -37,7 +37,7 @@
 
 static int
 cperf_set_ops_null_cipher(struct rte_crypto_op **ops,
-   struct rte_mbuf **bufs_in, struct rte_mbuf **bufs_out,
+   uint32_t src_buf_offset, uint32_t dst_buf_offset,
uint16_t nb_ops, struct rte_cryptodev_sym_session *sess,
const struct cperf_options *options,
const struct cperf_test_vector *test_vector __rte_unused,
@@ -48,10 +48,18 @@ cperf_set_ops_null_cipher(struct rte_crypto_op **ops,
for (i = 0; i < nb_ops; i++) {
struct rte_crypto_sym_op *sym_op = ops[i]->sym;
 
+   ops[i]->status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED;
rte_crypto_op_attach_sym_session(ops[i], sess);
 
-   sym_op->m_src = bufs_in[i];
-   sym_op->m_dst = bufs_out[i];
+   sym_op->m_src = (struct rte_mbuf *)((uint8_t *)ops[i] +
+   src_buf_offset);
+
+   /* Set dest mbuf to NULL if out-of-place (dst_buf_offset = 0) */
+   if (dst_buf_offset == 0)
+   sym_op->m_dst = NULL;
+   else
+   sym_op->m_dst = (struct rte_mbuf *)((uint8_t *)ops[i] +
+   dst_buf_offset);
 
/* cipher parameters */
sym_op->cipher.data.length = options->test_buffer_size;
@@ -63,7 +71,7 @@ cperf_set_ops_null_cipher(struct rte_crypto_op **ops,
 
 static int
 cperf_set_ops_null_auth(struct rte_crypto_op **ops,
-   struct rte_mbuf **bufs_in, struct rte_mbuf **bufs_out,
+   uint32_t src_buf_offset, uint32_t dst_buf_offset,
uint16_t nb_ops, struct rte_cryptodev_sym_session *sess,
const struct cperf_options *options,
const struct cperf_test_vector *test_vector __rte_unused,
@@ -74,10 +82,18 @@ cperf_set_ops_null_auth(struct rte_crypto_op **ops,
for (i = 0; i < nb_ops; i++) {
struct rte_crypto_sym_op *sym_op = ops[i]->sym;
 
+   ops[i]->status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED;
rte_crypto_op_attach_sym_session(ops[i], sess);
 
-   sym_op->m_src = bufs_in[i];
-   sym_op->m_dst = bufs_out[i];
+   sym_op->m_src = (struct rte_mbuf *)((uint8_t *)ops[i] +
+   src_buf_offset);
+
+   /* Set dest mbuf to NULL if out-of-place (dst_buf_offset = 0) */
+   if (dst_buf_offset == 0)
+   sym_op->m_dst = NULL;
+   else
+   sym_op->m_dst = (struct rte_mbuf *)((uint8_t *)ops[i] +
+   dst_buf_offset);
 
/* auth parameters */
sym_op->auth.data.length = options->test_buffer_size;
@@ -89,7 +105,7 @@ cperf_set_ops_null_auth(struct rte_crypto_op **ops,
 
 static int
 cperf_set_ops_cipher(struct rte_crypto_op **ops,
-   struct rte_mbuf **bufs_in, struct rte_mbuf **bufs_out,
+   uint32_t src_buf_offset, uint32_t dst_buf_offset,
uint16_t nb_ops, struct rte_cryptodev_sym_session *sess,
const struct cperf_options *options,
const struct cperf_test_vector *test_vector,
@@ -100,10 +116,18 @@ cperf_set_ops_cipher(struct rte_crypto_op **ops,
for (i = 0; i < nb_ops; i++) {
struct rte_crypto_sym_op *sym_op = ops[i]->sym;
 
+   ops[i]->status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED;
rte_crypto_op_attach_sym_session(ops[i], sess);
 
-   sym_op->m_src = bufs_in[i];
-   sym_op->m_dst = bufs_out[i];
+   sym_op->m_src = (struct rte_mbuf *)((uint8_t *)ops[i] +
+   src_buf_offset);
+
+   

Re: [dpdk-dev] [PATCH] net/mlx5: fix checksum offload cap

[Nélio Laranjeiro]

On Wed, Oct 04, 2017 at 11:28:43AM +0300, Shahaf Shuler wrote:
> The PMD was using the wrong flag to query device checksum capabilities
> 
> Fixes: bba710e6b99b ("net/mlx5: support upstream rdma-core")
> Cc: shacha...@mellanox.com
> 
> Signed-off-by: Shahaf Shuler 
> ---
>  drivers/net/mlx5/mlx5.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/mlx5/mlx5.c b/drivers/net/mlx5/mlx5.c
> index 229b824bc..a4b718417 100644
> --- a/drivers/net/mlx5/mlx5.c
> +++ b/drivers/net/mlx5/mlx5.c
> @@ -649,8 +649,8 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct 
> rte_pci_device *pci_dev)
>   }
>  
>   priv->hw_csum =
> - ((device_attr_ex.device_cap_flags_ex &
> -   IBV_DEVICE_UD_IP_CSUM));
> + !!(device_attr_ex.device_cap_flags_ex &
> +IBV_DEVICE_RAW_IP_CSUM);
>   DEBUG("checksum offloading is %ssupported",
> (priv->hw_csum ? "" : "not "));
>  
> -- 
> 2.12.0


Acked-by: Nelio Laranjeiro 

-- 
Nélio Laranjeiro
6WIND

[dpdk-dev] [PATCH v5 0/5] net/i40e: implement dynamic mapping of flow types to pctypes

[Kirill Rybalchenko]

Implement dynamic mapping of software flow types to hardware pctypes.
This allows to map new flow types to pctypes without changing
API of the driver.

v2:
Remove unnecessary check for new flow types.
Re-arrange patchset to avoid compillation errors.
Remove unnecessary usage of statically defined flow types and pctypes.

v3:
Remove unnecessary bit operations in I40E_PFQF_HENA and I40E_VFQF_HENA 
registers.
Add new definition in enum i40e_filter_pctype for for invalid pctype.
Fixed bugs in i40e_pctype_to_flowtype and i40e_flowtype_to_pctype functions.
Function rte_pmd_i40e_flow_type_mapping_get returns now full mapping table.
testpmd: changed command syntax from 'pctype mapping...' to
'port config pctype mapping...' and 'show port pctype mapping'
Various small modifications in code style after reviewing.

v4:
Change prototypes of some static functions.
Move declaration of automatic variables to beginning of function.
Move declaration of I40E_FILTER_PCTYPE_INVALID to i40e_ethdev.h
Fix some typos in source filea and documentation.

v5:
Fix code style warnings and documentation.
No functional changes.

Reviewed-by: Ferruh Yigit 

Kirill Rybalchenko (5):
  net/i40e: remove unnecessary bit operations
  net/i40e: implement dynamic mapping of sw flow types to hw pctypes
  net/i40e: add new functions to manipulate with pctype mapping table
  app/testpmd: add new commands to manipulate with  pctype mapping
  ethdev: remove unnecessary check for new flow type

 app/test-pmd/cmdline.c  | 338 -
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |  25 ++
 drivers/net/i40e/i40e_ethdev.c  | 366 +---
 drivers/net/i40e/i40e_ethdev.h  |  23 +-
 drivers/net/i40e/i40e_ethdev_vf.c   |  36 +--
 drivers/net/i40e/i40e_fdir.c|  54 ++--
 drivers/net/i40e/i40e_flow.c|   5 +-
 drivers/net/i40e/i40e_rxtx.c|  58 +
 drivers/net/i40e/i40e_rxtx.h|   1 +
 drivers/net/i40e/rte_pmd_i40e.c |  94 +++
 drivers/net/i40e/rte_pmd_i40e.h |  55 +
 drivers/net/i40e/rte_pmd_i40e_version.map   |   3 +
 lib/librte_ether/rte_ethdev.c   |   8 -
 13 files changed, 728 insertions(+), 338 deletions(-)

-- 
2.5.5

[dpdk-dev] [PATCH v5 1/5] net/i40e: remove unnecessary bit operations

[Kirill Rybalchenko]

Remove unnecessary bit operations in I40E_PFQF_HENA
and I40E_VFQF_HENA registers.

Signed-off-by: Kirill Rybalchenko 
---
 drivers/net/i40e/i40e_ethdev.c| 21 +++--
 drivers/net/i40e/i40e_ethdev_vf.c | 22 --
 2 files changed, 7 insertions(+), 36 deletions(-)

diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index acdf0de..41c4033 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -6717,16 +6717,9 @@ static void
 i40e_pf_disable_rss(struct i40e_pf *pf)
 {
struct i40e_hw *hw = I40E_PF_TO_HW(pf);
-   uint64_t hena;
 
-   hena = (uint64_t)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(0));
-   hena |= ((uint64_t)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(1))) << 32;
-   if (hw->mac.type == I40E_MAC_X722)
-   hena &= ~I40E_RSS_HENA_ALL_X722;
-   else
-   hena &= ~I40E_RSS_HENA_ALL;
-   i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), (uint32_t)hena);
-   i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), (uint32_t)(hena >> 32));
+   i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), 0);
+   i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), 0);
I40E_WRITE_FLUSH(hw);
 }
 
@@ -6798,7 +6791,6 @@ static int
 i40e_hw_rss_hash_set(struct i40e_pf *pf, struct rte_eth_rss_conf *rss_conf)
 {
struct i40e_hw *hw = I40E_PF_TO_HW(pf);
-   uint64_t rss_hf;
uint64_t hena;
int ret;
 
@@ -6807,14 +6799,7 @@ i40e_hw_rss_hash_set(struct i40e_pf *pf, struct 
rte_eth_rss_conf *rss_conf)
if (ret)
return ret;
 
-   rss_hf = rss_conf->rss_hf;
-   hena = (uint64_t)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(0));
-   hena |= ((uint64_t)i40e_read_rx_ctl(hw, I40E_PFQF_HENA(1))) << 32;
-   if (hw->mac.type == I40E_MAC_X722)
-   hena &= ~I40E_RSS_HENA_ALL_X722;
-   else
-   hena &= ~I40E_RSS_HENA_ALL;
-   hena |= i40e_config_hena(rss_hf, hw->mac.type);
+   hena = i40e_config_hena(rss_conf->rss_hf, hw->mac.type);
i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), (uint32_t)hena);
i40e_write_rx_ctl(hw, I40E_PFQF_HENA(1), (uint32_t)(hena >> 32));
I40E_WRITE_FLUSH(hw);
diff --git a/drivers/net/i40e/i40e_ethdev_vf.c 
b/drivers/net/i40e/i40e_ethdev_vf.c
index b35011a..a903deb 100644
--- a/drivers/net/i40e/i40e_ethdev_vf.c
+++ b/drivers/net/i40e/i40e_ethdev_vf.c
@@ -2498,7 +2498,7 @@ static int
 i40evf_hw_rss_hash_set(struct i40e_vf *vf, struct rte_eth_rss_conf *rss_conf)
 {
struct i40e_hw *hw = I40E_VF_TO_HW(vf);
-   uint64_t rss_hf, hena;
+   uint64_t hena;
int ret;
 
ret = i40evf_set_rss_key(&vf->vsi, rss_conf->rss_key,
@@ -2506,14 +2506,7 @@ i40evf_hw_rss_hash_set(struct i40e_vf *vf, struct 
rte_eth_rss_conf *rss_conf)
if (ret)
return ret;
 
-   rss_hf = rss_conf->rss_hf;
-   hena = (uint64_t)i40e_read_rx_ctl(hw, I40E_VFQF_HENA(0));
-   hena |= ((uint64_t)i40e_read_rx_ctl(hw, I40E_VFQF_HENA(1))) << 32;
-   if (hw->mac.type == I40E_MAC_X722)
-   hena &= ~I40E_RSS_HENA_ALL_X722;
-   else
-   hena &= ~I40E_RSS_HENA_ALL;
-   hena |= i40e_config_hena(rss_hf, hw->mac.type);
+   hena = i40e_config_hena(rss_conf->rss_hf, hw->mac.type);
i40e_write_rx_ctl(hw, I40E_VFQF_HENA(0), (uint32_t)hena);
i40e_write_rx_ctl(hw, I40E_VFQF_HENA(1), (uint32_t)(hena >> 32));
I40EVF_WRITE_FLUSH(hw);
@@ -2525,16 +2518,9 @@ static void
 i40evf_disable_rss(struct i40e_vf *vf)
 {
struct i40e_hw *hw = I40E_VF_TO_HW(vf);
-   uint64_t hena;
 
-   hena = (uint64_t)i40e_read_rx_ctl(hw, I40E_VFQF_HENA(0));
-   hena |= ((uint64_t)i40e_read_rx_ctl(hw, I40E_VFQF_HENA(1))) << 32;
-   if (hw->mac.type == I40E_MAC_X722)
-   hena &= ~I40E_RSS_HENA_ALL_X722;
-   else
-   hena &= ~I40E_RSS_HENA_ALL;
-   i40e_write_rx_ctl(hw, I40E_VFQF_HENA(0), (uint32_t)hena);
-   i40e_write_rx_ctl(hw, I40E_VFQF_HENA(1), (uint32_t)(hena >> 32));
+   i40e_write_rx_ctl(hw, I40E_VFQF_HENA(0), 0);
+   i40e_write_rx_ctl(hw, I40E_VFQF_HENA(1), 0);
I40EVF_WRITE_FLUSH(hw);
 }
 
-- 
2.5.5

[dpdk-dev] [PATCH v5 2/5] net/i40e: implement dynamic mapping of sw flow types to hw pctypes

[Kirill Rybalchenko]

Implement dynamic mapping of software flow types to hardware pctypes.
This allows to add new flow types and pctypes for DDP without changing
API of the driver. The mapping table is located in private
data area for particular network adapter and can be individually
modified with set of appropriate functions.

v2:
Re-arrange patchset to avoid compillation errors.
Remove usage of statically defined flow types and pctypes.

v3:
Change prototypes of some static functions.
Fixe bugs in i40e_pctype_to_flowtype and i40e_flowtype_to_pctype
functions.
Various small modifications after reviewing.

v4:
Change prototypes of some static functions.
Move declaration of automatic variables to beginning of function.
Move declaration of I40E_FILTER_PCTYPE_INVALID to i40e_ethdev.h

v5:
Fix code style warnings.

Signed-off-by: Kirill Rybalchenko 
---
 drivers/net/i40e/i40e_ethdev.c| 347 --
 drivers/net/i40e/i40e_ethdev.h|  23 ++-
 drivers/net/i40e/i40e_ethdev_vf.c |  16 +-
 drivers/net/i40e/i40e_fdir.c  |  54 +++---
 drivers/net/i40e/i40e_flow.c  |   5 +-
 drivers/net/i40e/i40e_rxtx.c  |  58 +++
 drivers/net/i40e/i40e_rxtx.h  |   1 +
 7 files changed, 218 insertions(+), 286 deletions(-)

diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 41c4033..0b151a0 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -1070,6 +1070,7 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
return 0;
}
i40e_set_default_ptype_table(dev);
+   i40e_set_default_pctype_table(dev);
pci_dev = RTE_ETH_DEV_TO_PCI(dev);
intr_handle = &pci_dev->intr_handle;
 
@@ -3020,7 +3021,7 @@ i40e_dev_info_get(struct rte_eth_dev *dev, struct 
rte_eth_dev_info *dev_info)
dev_info->hash_key_size = (I40E_PFQF_HKEY_MAX_INDEX + 1) *
sizeof(uint32_t);
dev_info->reta_size = pf->hash_lut_size;
-   dev_info->flow_type_rss_offloads = I40E_RSS_OFFLOAD_ALL;
+   dev_info->flow_type_rss_offloads = pf->adapter->flow_types_mask;
 
dev_info->default_rxconf = (struct rte_eth_rxconf) {
.rx_thresh = {
@@ -6611,104 +6612,36 @@ i40e_vsi_delete_mac(struct i40e_vsi *vsi, struct 
ether_addr *addr)
 
 /* Configure hash enable flags for RSS */
 uint64_t
-i40e_config_hena(uint64_t flags, enum i40e_mac_type type)
+i40e_config_hena(const struct i40e_adapter *adapter, uint64_t flags)
 {
uint64_t hena = 0;
+   int i;
 
if (!flags)
return hena;
 
-   if (flags & ETH_RSS_FRAG_IPV4)
-   hena |= 1ULL << I40E_FILTER_PCTYPE_FRAG_IPV4;
-   if (flags & ETH_RSS_NONFRAG_IPV4_TCP) {
-   if (type == I40E_MAC_X722) {
-   hena |= (1ULL << I40E_FILTER_PCTYPE_NONF_IPV4_TCP) |
-(1ULL << I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK);
-   } else
-   hena |= 1ULL << I40E_FILTER_PCTYPE_NONF_IPV4_TCP;
-   }
-   if (flags & ETH_RSS_NONFRAG_IPV4_UDP) {
-   if (type == I40E_MAC_X722) {
-   hena |= (1ULL << I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
-(1ULL << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) |
-(1ULL << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP);
-   } else
-   hena |= 1ULL << I40E_FILTER_PCTYPE_NONF_IPV4_UDP;
-   }
-   if (flags & ETH_RSS_NONFRAG_IPV4_SCTP)
-   hena |= 1ULL << I40E_FILTER_PCTYPE_NONF_IPV4_SCTP;
-   if (flags & ETH_RSS_NONFRAG_IPV4_OTHER)
-   hena |= 1ULL << I40E_FILTER_PCTYPE_NONF_IPV4_OTHER;
-   if (flags & ETH_RSS_FRAG_IPV6)
-   hena |= 1ULL << I40E_FILTER_PCTYPE_FRAG_IPV6;
-   if (flags & ETH_RSS_NONFRAG_IPV6_TCP) {
-   if (type == I40E_MAC_X722) {
-   hena |= (1ULL << I40E_FILTER_PCTYPE_NONF_IPV6_TCP) |
-(1ULL << I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN_NO_ACK);
-   } else
-   hena |= 1ULL << I40E_FILTER_PCTYPE_NONF_IPV6_TCP;
+   for (i = RTE_ETH_FLOW_UNKNOWN + 1; i < I40E_FLOW_TYPE_MAX; i++) {
+   if (flags & (1ULL << i))
+   hena |= adapter->pctypes_tbl[i];
}
-   if (flags & ETH_RSS_NONFRAG_IPV6_UDP) {
-   if (type == I40E_MAC_X722) {
-   hena |= (1ULL << I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
-(1ULL << I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) |
-(1ULL << I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP);
-   } else
-   hena |= 1ULL << I40E_FILTER_PCTYPE_NONF_IPV6_UDP;
-   }
-   if (flags & ETH_RSS_NONFRAG_IPV6_SCTP)
-   hena |= 1ULL << I40E_FILTER_PCTYPE_NONF_IPV6_SCTP;
-   if (flags & ETH_RSS_NONFRAG_IPV6_OTHER)
-   hena |= 1ULL << I40E_FILTER_PCTYPE_N

[dpdk-dev] [PATCH v5 3/5] net/i40e: add new functions to manipulate with pctype mapping table

[Kirill Rybalchenko]

Add new functions which allow modify, return or reset to default
the contents of flow type to pctype dynamic mapping table.

v3:
Function rte_pmd_i40e_flow_type_mapping_get returns now full
mapping table.

v4:
Fix typo in rte_pmd_i40e_version.map file.

v5:
Fix code style warnings.

Signed-off-by: Kirill Rybalchenko 
---
 drivers/net/i40e/rte_pmd_i40e.c   | 94 +++
 drivers/net/i40e/rte_pmd_i40e.h   | 55 ++
 drivers/net/i40e/rte_pmd_i40e_version.map |  3 +
 3 files changed, 152 insertions(+)

diff --git a/drivers/net/i40e/rte_pmd_i40e.c b/drivers/net/i40e/rte_pmd_i40e.c
index c08e07a..3b4582d 100644
--- a/drivers/net/i40e/rte_pmd_i40e.c
+++ b/drivers/net/i40e/rte_pmd_i40e.c
@@ -2161,3 +2161,97 @@ rte_pmd_i40e_add_vf_mac_addr(uint8_t port, uint16_t 
vf_id,
 
return 0;
 }
+
+int rte_pmd_i40e_flow_type_mapping_reset(uint8_t port)
+{
+   struct rte_eth_dev *dev;
+
+   RTE_ETH_VALID_PORTID_OR_ERR_RET(port, -ENODEV);
+
+   dev = &rte_eth_devices[port];
+
+   if (!is_i40e_supported(dev))
+   return -ENOTSUP;
+
+   i40e_set_default_pctype_table(dev);
+
+   return 0;
+}
+
+int rte_pmd_i40e_flow_type_mapping_get(
+   uint8_t port,
+   struct rte_pmd_i40e_flow_type_mapping *mapping_items)
+{
+   struct rte_eth_dev *dev;
+   struct i40e_adapter *ad;
+   uint16_t i;
+
+   RTE_ETH_VALID_PORTID_OR_ERR_RET(port, -ENODEV);
+
+   dev = &rte_eth_devices[port];
+
+   if (!is_i40e_supported(dev))
+   return -ENOTSUP;
+
+   ad = I40E_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private);
+
+   for (i = 0; i < I40E_FLOW_TYPE_MAX; i++) {
+   mapping_items[i].flow_type = i;
+   mapping_items[i].pctype = ad->pctypes_tbl[i];
+   }
+
+   return 0;
+}
+
+int
+rte_pmd_i40e_flow_type_mapping_update(
+   uint8_t port,
+   struct rte_pmd_i40e_flow_type_mapping *mapping_items,
+   uint16_t count,
+   uint8_t exclusive)
+{
+   struct rte_eth_dev *dev;
+   struct i40e_adapter *ad;
+   int i;
+
+   RTE_ETH_VALID_PORTID_OR_ERR_RET(port, -ENODEV);
+
+   dev = &rte_eth_devices[port];
+
+   if (!is_i40e_supported(dev))
+   return -ENOTSUP;
+
+   if (count > I40E_FLOW_TYPE_MAX)
+   return -EINVAL;
+
+   for (i = 0; i < count; i++)
+   if (mapping_items[i].flow_type >= I40E_FLOW_TYPE_MAX ||
+   mapping_items[i].flow_type == RTE_ETH_FLOW_UNKNOWN ||
+   (mapping_items[i].pctype &
+   (1ULL << I40E_FILTER_PCTYPE_INVALID)))
+   return -EINVAL;
+
+   ad = I40E_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private);
+
+   if (exclusive) {
+   for (i = 0; i < I40E_FLOW_TYPE_MAX; i++)
+   ad->pctypes_tbl[i] = 0ULL;
+   ad->flow_types_mask = 0ULL;
+   }
+
+   for (i = 0; i < count; i++) {
+   ad->pctypes_tbl[mapping_items[i].flow_type] =
+   mapping_items[i].pctype;
+   if (mapping_items[i].pctype)
+   ad->flow_types_mask |=
+   (1ULL << mapping_items[i].flow_type);
+   else
+   ad->flow_types_mask &=
+   ~(1ULL << mapping_items[i].flow_type);
+   }
+
+   for (i = 0, ad->pctypes_mask = 0ULL; i < I40E_FLOW_TYPE_MAX; i++)
+   ad->pctypes_mask |= ad->pctypes_tbl[i];
+
+   return 0;
+}
diff --git a/drivers/net/i40e/rte_pmd_i40e.h b/drivers/net/i40e/rte_pmd_i40e.h
index 155b7e8..004a8a5 100644
--- a/drivers/net/i40e/rte_pmd_i40e.h
+++ b/drivers/net/i40e/rte_pmd_i40e.h
@@ -657,4 +657,59 @@ int rte_pmd_i40e_ptype_mapping_replace(uint8_t port,
 int rte_pmd_i40e_add_vf_mac_addr(uint8_t port, uint16_t vf_id,
 struct ether_addr *mac_addr);
 
+#define RTE_PMD_I40E_PCTYPE_MAX64
+#define RTE_PMD_I40E_FLOW_TYPE_MAX 64
+
+struct rte_pmd_i40e_flow_type_mapping {
+   uint16_t flow_type; /**< software defined flow type*/
+   uint64_t pctype;/**< hardware defined pctype */
+};
+
+/**
+ * Update hardware defined pctype to software defined flow type
+ * mapping table.
+ *
+ * @param port
+ *pointer to port identifier of the device.
+ * @param mapping_items
+ *the base address of the mapping items array.
+ * @param count
+ *number of mapping items.
+ * @param exclusive
+ *the flag indicate different pctype mapping update method.
+ *-(0) only overwrite referred PCTYPE mapping,
+ * keep other PCTYPEs mapping unchanged.
+ *-(!0) overwrite referred PCTYPE mapping,
+ * set other PCTYPEs maps to PCTYPE_INVALID.
+ */
+int rte_pmd_i40e_flow_type_mapping_update(
+   uint8_t port,
+   

[dpdk-dev] [PATCH v5 4/5] app/testpmd: add new commands to manipulate with pctype mapping

[Kirill Rybalchenko]

Add new commands to manipulate with dynamic flow type to
pctype mapping table in i40e PMD.
Commands allow to print table, modify it and reset to default value.

v3:
changed command syntax from 'pctype mapping...' to
'port config pctype mapping...' and 'show port pctype mapping'

v4:
Fix typos in cmdline.c and documentation.
Move variable declaration to the beginning of function.

v5:
Fix code style warnings and documentation.

Signed-off-by: Kirill Rybalchenko 
---
 app/test-pmd/cmdline.c  | 338 +++-
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |  25 ++
 2 files changed, 353 insertions(+), 10 deletions(-)

diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 4f2d731..dc9833b 100644
--- a/app/test-pmd/cmdline.c
+++ b/app/test-pmd/cmdline.c
@@ -230,6 +230,10 @@ static void cmd_help_long_parsed(void *parsed_result,
 
"clear vf stats (port_id) (vf_id)\n"
"Reset a VF's statistics.\n\n"
+
+   "show port (port_id) pctype mapping\n"
+   "Get flow ptype to pctype mapping on a port\n\n"
+
);
}
 
@@ -681,7 +685,8 @@ static void cmd_help_long_parsed(void *parsed_result,
"Set 
crc-strip/scatter/rx-checksum/hardware-vlan/drop_en"
" for ports.\n\n"
 
-   "port config all rss 
(all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none)\n"
+   "port config all rss 
(all|ip|tcp|udp|sctp|ether|port|vxlan|"
+   "geneve|nvgre|none|)\n"
"Set the RSS mode.\n\n"
 
"port config port-id rss reta 
(hash,queue)[,(hash,queue)]\n"
@@ -716,6 +721,13 @@ static void cmd_help_long_parsed(void *parsed_result,
"port config (port_id|all) l2-tunnel E-tag"
" (enable|disable)\n"
"Enable/disable the E-tag support.\n\n"
+
+   "port config (port_id) pctype mapping reset\n"
+   "Reset flow type to pctype mapping on a port\n\n"
+
+   "port config (port_id) pctype mapping update"
+   " (pctype_id_0[,pctype_id_1]*) (flow_type_id)\n"
+   "Update a flow type to pctype mapping item on a 
port\n\n"
);
}
 
@@ -878,8 +890,8 @@ static void cmd_help_long_parsed(void *parsed_result,
"set_hash_input_set (port_id) (ipv4|ipv4-frag|"
"ipv4-tcp|ipv4-udp|ipv4-sctp|ipv4-other|ipv6|"
"ipv6-frag|ipv6-tcp|ipv6-udp|ipv6-sctp|ipv6-other|"
-   "l2_payload) (ovlan|ivlan|src-ipv4|dst-ipv4|src-ipv6|"
-   "dst-ipv6|ipv4-tos|ipv4-proto|ipv6-tc|"
+   "l2_payload|) 
(ovlan|ivlan|src-ipv4|dst-ipv4|"
+   "src-ipv6|dst-ipv6|ipv4-tos|ipv4-proto|ipv6-tc|"
"ipv6-next-header|udp-src-port|udp-dst-port|"
"tcp-src-port|tcp-dst-port|sctp-src-port|"
"sctp-dst-port|sctp-veri-tag|udp-key|gre-key|fld-1st|"
@@ -1720,6 +1732,9 @@ cmd_config_rss_parsed(void *parsed_result,
rss_conf.rss_hf = ETH_RSS_NVGRE;
else if (!strcmp(res->value, "none"))
rss_conf.rss_hf = 0;
+   else if (isdigit(res->value[0]) && atoi(res->value) > 0 &&
+   atoi(res->value) < 64)
+   rss_conf.rss_hf = 1ULL << atoi(res->value);
else {
printf("Unknown parameter\n");
return;
@@ -1743,14 +1758,13 @@ cmdline_parse_token_string_t cmd_config_rss_all =
 cmdline_parse_token_string_t cmd_config_rss_name =
TOKEN_STRING_INITIALIZER(struct cmd_config_rss, name, "rss");
 cmdline_parse_token_string_t cmd_config_rss_value =
-   TOKEN_STRING_INITIALIZER(struct cmd_config_rss, value,
-   "all#ip#tcp#udp#sctp#ether#port#vxlan#geneve#nvgre#none");
+   TOKEN_STRING_INITIALIZER(struct cmd_config_rss, value, NULL);
 
 cmdline_parse_inst_t cmd_config_rss = {
.f = cmd_config_rss_parsed,
.data = NULL,
.help_str = "port config all rss "
-   "all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none",
+   
"all|ip|tcp|udp|sctp|ether|port|vxlan|geneve|nvgre|none|",
.tokens = {
(void *)&cmd_config_rss_port,
(void *)&cmd_config_rss_keyword,
@@ -8991,6 +9005,10 @@ str2flowtype(char *string)
if (!strcmp(flowtype_str[i].str, string))
return flowtype_str[i].type;
}
+
+   if (isdigit(string[0]) && atoi(string) > 0 && atoi(string) < 64)
+   return (uint16_t)atoi(string);
+
return RTE_ETH_FLOW_UNKNOWN;
 }
 
@@ -10467,9 +10485,7 @@ cmdline_parse_token_num_t 
cmd_set_hash_input_set_port_id =

[dpdk-dev] [PATCH v5 5/5] ethdev: remove unnecessary check for new flow type

[Kirill Rybalchenko]

Remove unnecessary check for new flow type for rss hash
filter update.

Signed-off-by: Kirill Rybalchenko 
---
 lib/librte_ether/rte_ethdev.c | 8 
 1 file changed, 8 deletions(-)

diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
index 1849a3b..f3bf3e5 100644
--- a/lib/librte_ether/rte_ethdev.c
+++ b/lib/librte_ether/rte_ethdev.c
@@ -2291,16 +2291,8 @@ int
 rte_eth_dev_rss_hash_update(uint8_t port_id, struct rte_eth_rss_conf *rss_conf)
 {
struct rte_eth_dev *dev;
-   uint16_t rss_hash_protos;
 
RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV);
-   rss_hash_protos = rss_conf->rss_hf;
-   if ((rss_hash_protos != 0) &&
-   ((rss_hash_protos & ETH_RSS_PROTO_MASK) == 0)) {
-   RTE_PMD_DEBUG_TRACE("Invalid rss_hash_protos=0x%x\n",
-   rss_hash_protos);
-   return -EINVAL;
-   }
dev = &rte_eth_devices[port_id];
RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->rss_hash_update, -ENOTSUP);
return (*dev->dev_ops->rss_hash_update)(dev, rss_conf);
-- 
2.5.5

Re: [dpdk-dev] [PATCH] app: fix gcc -O3 compiler errors

[Burakov, Anatoly]

On 15-Aug-17 2:53 PM, Keith Wiles wrote:

When using EXTRA_CFLAGS="-g -O3" in the build the -O3 causes
compiler warnings for these two applications. Using Ubuntu 17.04
gcc compiler.

Signed-off-by: Keith Wiles 
---


Acked-by: Anatoly Burakov 

Re: [dpdk-dev] [PATCH] app/testpmd: wrong usage of fseek & ftell to determine filesize

[Mrzyglod, DanielX T]

>>
>> This issue was about passing unsigned argument where should be signed
>> number.
>> In reality this is about wrong usage of fseek and ftell to determine 
>> filesize.
>> This patch is compliant to suggestions from FIO19-C:
>> "Do not use fseek() and ftell() to compute the size of a regular file"
>>
>> Coverity issue: 143454
>> Fixes: a92a5a2cbbff ("app/testpmd: add command for loading DDP")
>>
>> Signed-off-by: Daniel Mrzyglod 
>
>Acked-by: Beilei Xing 


Hi Jingjing
I would like to ask for a feedback about status of proposed fix.
Will it go for integration or does it need any other additional work.

Best regards
Daniel

Re: [dpdk-dev] [PATCH v6 1/6] gso: add Generic Segmentation Offload API framework

[Ananyev, Konstantin]

> -Original Message-
> From: Kavanagh, Mark B
> Sent: Monday, October 2, 2017 5:46 PM
> To: dev@dpdk.org
> Cc: Hu, Jiayu ; Tan, Jianfeng ; 
> Ananyev, Konstantin ; Yigit,
> Ferruh ; tho...@monjalon.net; Kavanagh, Mark B 
> 
> Subject: [PATCH v6 1/6] gso: add Generic Segmentation Offload API framework
> 
> From: Jiayu Hu 
> 
> Generic Segmentation Offload (GSO) is a SW technique to split large
> packets into small ones. Akin to TSO, GSO enables applications to
> operate on large packets, thus reducing per-packet processing overhead.
> 
> To enable more flexibility to applications, DPDK GSO is implemented
> as a standalone library. Applications explicitly use the GSO library
> to segment packets. To segment a packet requires two steps. The first
> is to set proper flags to mbuf->ol_flags, where the flags are the same
> as that of TSO. The second is to call the segmentation API,
> rte_gso_segment(). This patch introduces the GSO API framework to DPDK.
> 
> rte_gso_segment() splits an input packet into small ones in each
> invocation. The GSO library refers to these small packets generated
> by rte_gso_segment() as GSO segments. Each of the newly-created GSO
> segments is organized as a two-segment MBUF, where the first segment is a
> standard MBUF, which stores a copy of packet header, and the second is an
> indirect MBUF which points to a section of data in the input packet.
> rte_gso_segment() reduces the refcnt of the input packet by 1. Therefore,
> when all GSO segments are freed, the input packet is freed automatically.
> Additionally, since each GSO segment has multiple MBUFs (i.e. 2 MBUFs),
> the driver of the interface which the GSO segments are sent to should
> support to transmit multi-segment packets.
> 
> The GSO framework clears the PKT_TX_TCP_SEG flag for both the input
> packet, and all produced GSO segments in the event of success, since
> segmentation in hardware is no longer required at that point.
> 
> Signed-off-by: Jiayu Hu 
> Signed-off-by: Mark Kavanagh 
> ---
>  config/common_base |   5 ++
>  doc/api/doxy-api-index.md  |   1 +
>  doc/api/doxy-api.conf  |   1 +
>  doc/guides/rel_notes/release_17_11.rst |   1 +
>  lib/Makefile   |   2 +
>  lib/librte_gso/Makefile|  49 +++
>  lib/librte_gso/rte_gso.c   |  52 
>  lib/librte_gso/rte_gso.h   | 145 
> +
>  lib/librte_gso/rte_gso_version.map |   7 ++
>  mk/rte.app.mk  |   1 +
>  10 files changed, 264 insertions(+)
>  create mode 100644 lib/librte_gso/Makefile
>  create mode 100644 lib/librte_gso/rte_gso.c
>  create mode 100644 lib/librte_gso/rte_gso.h
>  create mode 100644 lib/librte_gso/rte_gso_version.map
> 
> diff --git a/config/common_base b/config/common_base
> index 12f6be9..58ca5c0 100644
> --- a/config/common_base
> +++ b/config/common_base
> @@ -653,6 +653,11 @@ CONFIG_RTE_LIBRTE_IP_FRAG_TBL_STAT=n
>  CONFIG_RTE_LIBRTE_GRO=y
> 
>  #
> +# Compile GSO library
> +#
> +CONFIG_RTE_LIBRTE_GSO=y
> +
> +#
>  # Compile librte_meter
>  #
>  CONFIG_RTE_LIBRTE_METER=y
> diff --git a/doc/api/doxy-api-index.md b/doc/api/doxy-api-index.md
> index 19e0d4f..6512918 100644
> --- a/doc/api/doxy-api-index.md
> +++ b/doc/api/doxy-api-index.md
> @@ -101,6 +101,7 @@ The public API headers are grouped by topics:
>[TCP](@ref rte_tcp.h),
>[UDP](@ref rte_udp.h),
>[GRO](@ref rte_gro.h),
> +  [GSO](@ref rte_gso.h),
>[frag/reass] (@ref rte_ip_frag.h),
>[LPM IPv4 route] (@ref rte_lpm.h),
>[LPM IPv6 route] (@ref rte_lpm6.h),
> diff --git a/doc/api/doxy-api.conf b/doc/api/doxy-api.conf
> index 823554f..408f2e6 100644
> --- a/doc/api/doxy-api.conf
> +++ b/doc/api/doxy-api.conf
> @@ -47,6 +47,7 @@ INPUT   = doc/api/doxy-api-index.md \
>lib/librte_ether \
>lib/librte_eventdev \
>lib/librte_gro \
> +  lib/librte_gso \
>lib/librte_hash \
>lib/librte_ip_frag \
>lib/librte_jobstats \
> diff --git a/doc/guides/rel_notes/release_17_11.rst 
> b/doc/guides/rel_notes/release_17_11.rst
> index 8bf91bd..7508be7 100644
> --- a/doc/guides/rel_notes/release_17_11.rst
> +++ b/doc/guides/rel_notes/release_17_11.rst
> @@ -174,6 +174,7 @@ The libraries prepended with a plus sign were incremented 
> in this version.
>   librte_ethdev.so.7
>   librte_eventdev.so.2
>   librte_gro.so.1
> +   + librte_gso.so.1
>   librte_hash.so.2
>   librte_ip_frag.so.1
>   librte_jobstats.so.1
> diff --git a/lib/Makefile b/lib/Makefile
> index 86caba1..3d123f4 100644
> --- a/lib/Makefile
> +++ b/lib/Makefile
> @@ -108,6 +108,8 @@ DIRS-$(CONFIG_RTE_LIBRTE_REORDER) += librte_reorde

Re: [dpdk-dev] [PATCH v2] vfio: refactor PCI BAR mapping

[Burakov, Anatoly]

On 25-Sep-17 4:04 PM, Jonas Pfefferle wrote:

Split pci_vfio_map_resource for primary and secondary processes.
Save all relevant mapping data in primary process to allow
the secondary process to perform mappings.

Signed-off-by: Jonas Pfefferle 
---
v2:
* fix zero size and offset when trying to mmap non msix bar

  lib/librte_eal/common/include/rte_pci.h|   7 +
  lib/librte_eal/linuxapp/eal/eal_pci_vfio.c | 446 +
  2 files changed, 271 insertions(+), 182 deletions(-)

diff --git a/lib/librte_eal/common/include/rte_pci.h 
b/lib/librte_eal/common/include/rte_pci.h
index 8b12339..0821af9 100644
--- a/lib/librte_eal/common/include/rte_pci.h
+++ b/lib/librte_eal/common/include/rte_pci.h
@@ -214,6 +214,12 @@ struct pci_map {
uint64_t phaddr;
  };
  
+struct pci_msix_table {

+   int bar_index;
+   uint32_t offset;
+   uint32_t size;
+};
+
  /**
   * A structure describing a mapped PCI resource.
   * For multi-process we need to reproduce all PCI mappings in secondary
@@ -226,6 +232,7 @@ struct mapped_pci_resource {
char path[PATH_MAX];
int nb_maps;
struct pci_map maps[PCI_MAX_RESOURCE];
+   struct pci_msix_table msix_table;
  };
  
  /** mapped pci device list */

diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c 
b/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c
index aa9d96e..6443bd5 100644
--- a/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c
+++ b/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c
@@ -88,8 +88,7 @@ pci_vfio_write_config(const struct rte_intr_handle 
*intr_handle,
  
  /* get PCI BAR number where MSI-X interrupts are */

  static int
-pci_vfio_get_msix_bar(int fd, int *msix_bar, uint32_t *msix_table_offset,
- uint32_t *msix_table_size)
+pci_vfio_get_msix_bar(int fd, struct pci_msix_table *msix_table)
  {
int ret;
uint32_t reg;
@@ -161,9 +160,10 @@ pci_vfio_get_msix_bar(int fd, int *msix_bar, uint32_t 
*msix_table_offset,
return -1;
}
  
-			*msix_bar = reg & RTE_PCI_MSIX_TABLE_BIR;

-   *msix_table_offset = reg & RTE_PCI_MSIX_TABLE_OFFSET;
-   *msix_table_size = 16 * (1 + (flags & 
RTE_PCI_MSIX_FLAGS_QSIZE));
+   msix_table->bar_index = reg & RTE_PCI_MSIX_TABLE_BIR;
+   msix_table->offset = reg & RTE_PCI_MSIX_TABLE_OFFSET;
+   msix_table->size =
+   16 * (1 + (flags & RTE_PCI_MSIX_FLAGS_QSIZE));
  
  			return 0;

}
@@ -300,25 +300,150 @@ pci_vfio_setup_interrupts(struct rte_pci_device *dev, 
int vfio_dev_fd)
return -1;
  }
  
-/*

- * map the PCI resources of a PCI device in virtual memory (VFIO version).
- * primary and secondary processes follow almost exactly the same path
- */
-int
-pci_vfio_map_resource(struct rte_pci_device *dev)
+static int
+pci_vfio_is_ioport_bar(int vfio_dev_fd, int bar_index)
+{
+   uint32_t ioport_bar;
+   int ret;
+
+   ret = pread64(vfio_dev_fd, &ioport_bar, sizeof(ioport_bar),
+ VFIO_GET_REGION_ADDR(VFIO_PCI_CONFIG_REGION_INDEX)
+ + PCI_BASE_ADDRESS_0 + bar_index*4);
+   if (ret != sizeof(ioport_bar)) {
+   RTE_LOG(ERR, EAL, "Cannot read command (%x) from config 
space!\n",
+   PCI_BASE_ADDRESS_0 + bar_index*4);
+   return -1;
+   }
+
+   return ioport_bar & PCI_BASE_ADDRESS_SPACE_IO;


Not sure i like this. I think it's better to be explicit, e.g.
return ioport_bar & PCI_BASE_ADDRESS_SPACE_IO != 0;

Makes no difference (both because return value is non-zero and because 
PCI_BASE_ADDRESS_SPACE_IO is 0x01), but still, better to make intentions 
clear i think.


Otherwise, did a quick smoke-test and it works, so

Acked-by: Anatoly Burakov 

Keep the ack if you decide to submit a v3 :)

Re: [dpdk-dev] [PATCH v3 2/4] net/mrvl: add mrvl net pmd driver

[Tomasz Duszynski]

On Wed, Oct 04, 2017 at 01:28:47AM +0100, Ferruh Yigit wrote:
> On 10/3/2017 12:51 PM, Tomasz Duszynski wrote:
> > Add support for the Marvell PPv2 (Packet Processor v2) 1/10 Gbps adapter.
> > Driver is based on external, publicly available, light-weight Marvell
> > MUSDK library that provides access to network packet processor.
> >
> > Driver comes with support for the following features:
> >
> > * Speed capabilities
> > * Link status
> > * Queue start/stop
> > * MTU update
> > * Jumbo frame
> > * Promiscuous mode
> > * Allmulticast mode
> > * Unicast MAC filter
> > * Multicast MAC filter
> > * RSS hash
> > * VLAN filter
> > * CRC offload
> > * L3 checksum offload
> > * L4 checksum offload
> > * Packet type parsing
> > * Basic stats
> > * Stats per queue
> >
> > Driver was engineered cooperatively by Semihalf and Marvell teams.
> >
> > Semihalf:
> > Jacek Siuda 
> > Tomasz Duszynski 
> >
> > Marvell:
> > Dmitri Epshtein 
> > Natalie Samsonov 
> >
> > Signed-off-by: Jacek Siuda 
> > Signed-off-by: Tomasz Duszynski 
>
> <...>
>
> > +++ b/config/common_base
> > @@ -262,6 +262,13 @@ CONFIG_RTE_LIBRTE_NFP_PMD=n
> >  CONFIG_RTE_LIBRTE_NFP_DEBUG=n
> >
> >  #
> > +# Compile Marvell PMD driver
> > +#
> > +CONFIG_RTE_LIBRTE_MRVL_PMD=n
> > +CONFIG_RTE_LIBRTE_MRVL_DEBUG=n
> > +CONFIG_RTE_MRVL_MUSDK_DMA_MEMSIZE=41943040
>
> Is dma memsize needs to be a configuration option?

That config option is used both by NET and CRYPTO drivers. In case NET
and CRYPTO are used together i.e ipsec-secgw then DMA_MEMSIZE must be
the set to the same size. Putting this configuration option in .config
makes sure DMA_MEMSIZE stays synchronized.

>
> <...>
>
> > +include $(RTE_SDK)/mk/rte.vars.mk
> > +
> > +ifneq ($(MAKECMDGOALS),clean)
> > +ifneq ($(MAKECMDGOALS),config)
> > +ifeq ($(LIBMUSDK_PATH),)
> > +$(error "Please define LIBMUSDK_PATH environment variable")
>
> Not sure how to resolve this dependency.
> What do you think adding this as configuration option?

All other drivers with external dependencies follow the same approach.

>
> Or DPDK just adds the -lmusdk external dependency and while compiling
> for marvel EXTRA_LDFLAGS parameter should be pass with
> "-L$(LIBMUSDK_PATH)" and this can be documented in marvel doc. What do
> you think?

Both solutions are reasonable. The former was chosen because that's what the
other drivers do.

>
> > +endif
> > +ifeq ($(CONFIG_RTE_LIBRTE_CFGFILE),n)
> > +$(error "RTE_LIBRTE_CFGFILE must be enabled in configuration!")
>
> This can be also handled in drivers/net/Makefile, it can be possible to
> add check there for LIBRTE_CFGFILE dependency.
>

ACK

> > +endif
> > +endif
> > +endif
> > +
> > +# library name
> > +LIB = librte_pmd_mrvl.a
> > +
> > +# library version
> > +LIBABIVER := 1
> > +
> > +# versioning export map
> > +EXPORT_MAP := rte_pmd_mrvl_version.map
> > +
> > +# external library dependencies
> > +CFLAGS += -I$(LIBMUSDK_PATH)/include
> > +CFLAGS += -DMVCONF_ARCH_DMA_ADDR_T_64BIT
> > +CFLAGS += -DCONF_PP2_BPOOL_COOKIE_SIZE=32
> > +CFLAGS += $(WERROR_FLAGS)
> > +CFLAGS += -O3
> > +LDLIBS += -L$(LIBMUSDK_PATH)/lib
>
> This can be LDFLAGS instead of LDLIBS

Moving that to LDFLAGS will break compilation in case
CONFIG_RTE_BUILD_SHARED_LIB is set as -L... does not show up on command
line thus linker does not know where to look extra library up.
I may be wrong but it looks as if specifying LDFLAGS in driver's
Makefile is no-op.

On the other hand, if we are building static libraries both
-lmusdk and -L$(LIBMUSDK_PATH)/lib are added to specific _LDLIBS which in turn
ends up in LDLIBS.

>
> > +LDLIBS += -lmusdk
> > +
> > +# library source files
> > +SRCS-$(CONFIG_RTE_LIBRTE_MRVL_PMD) += mrvl_ethdev.c
> > +SRCS-$(CONFIG_RTE_LIBRTE_MRVL_PMD) += mrvl_qos.c
> > +
> > +# library dependencies
> > +DEPDIRS-$(CONFIG_RTE_LIBRTE_MRVL_PMD) += lib/librte_cfgfile
>
> These variables no more used, you can drop this. drivers/net/Makefile
> used for this, you are already updating that file, librte_cfgfile needs
> to be added there.
>

ACK

> > +
> > +include $(RTE_SDK)/mk/rte.lib.mk
>
> <...>
>
> > +/*
> > + * To use buffer harvesting based on loopback port shadow queue structure
> > + * was introduced for buffers information bookkeeping.
> > + *
> > + * Before sending the packet, related buffer information (pp2_buff_inf) is
> > + * stored in shadow queue. After packet is transmitted no longer used
> > + * packet buffer is released back to it's original hardware pool,
> > + * on condition it originated from interface.
> > + * In case it  was generated by application itself i.e: mbuf->port field is
> > + * 0xff then its released to software mempool.
>
> You already explained here but can you please give more details why
> shadow queue needed?

It's used for mbuf harvesting in tx-path. Instead of releasing pushed
out mbuf to mempool and allocating it once again later on, mbuf is
stored in the shadow queue and returned back to hardware buffer manager
after being sent.

>
> > + */
> > +struct mrvl_shadow_txq

Re: [dpdk-dev] [PATCH v6 1/6] gso: add Generic Segmentation Offload API framework

[Kavanagh, Mark B]

>From: Ananyev, Konstantin
>Sent: Wednesday, October 4, 2017 2:11 PM
>To: Kavanagh, Mark B ; dev@dpdk.org
>Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Yigit, Ferruh ; tho...@monjalon.net
>Subject: RE: [PATCH v6 1/6] gso: add Generic Segmentation Offload API
>framework
>
>
>
>> -Original Message-
>> From: Kavanagh, Mark B
>> Sent: Monday, October 2, 2017 5:46 PM
>> To: dev@dpdk.org
>> Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Ananyev, Konstantin ; Yigit,
>> Ferruh ; tho...@monjalon.net; Kavanagh, Mark B
>
>> Subject: [PATCH v6 1/6] gso: add Generic Segmentation Offload API framework
>>
>> From: Jiayu Hu 
>>
>> Generic Segmentation Offload (GSO) is a SW technique to split large
>> packets into small ones. Akin to TSO, GSO enables applications to
>> operate on large packets, thus reducing per-packet processing overhead.
>>
>> To enable more flexibility to applications, DPDK GSO is implemented
>> as a standalone library. Applications explicitly use the GSO library
>> to segment packets. To segment a packet requires two steps. The first
>> is to set proper flags to mbuf->ol_flags, where the flags are the same
>> as that of TSO. The second is to call the segmentation API,
>> rte_gso_segment(). This patch introduces the GSO API framework to DPDK.
>>
>> rte_gso_segment() splits an input packet into small ones in each
>> invocation. The GSO library refers to these small packets generated
>> by rte_gso_segment() as GSO segments. Each of the newly-created GSO
>> segments is organized as a two-segment MBUF, where the first segment is a
>> standard MBUF, which stores a copy of packet header, and the second is an
>> indirect MBUF which points to a section of data in the input packet.
>> rte_gso_segment() reduces the refcnt of the input packet by 1. Therefore,
>> when all GSO segments are freed, the input packet is freed automatically.
>> Additionally, since each GSO segment has multiple MBUFs (i.e. 2 MBUFs),
>> the driver of the interface which the GSO segments are sent to should
>> support to transmit multi-segment packets.
>>
>> The GSO framework clears the PKT_TX_TCP_SEG flag for both the input
>> packet, and all produced GSO segments in the event of success, since
>> segmentation in hardware is no longer required at that point.
>>
>> Signed-off-by: Jiayu Hu 
>> Signed-off-by: Mark Kavanagh 
>> ---
>>  config/common_base |   5 ++
>>  doc/api/doxy-api-index.md  |   1 +
>>  doc/api/doxy-api.conf  |   1 +
>>  doc/guides/rel_notes/release_17_11.rst |   1 +
>>  lib/Makefile   |   2 +
>>  lib/librte_gso/Makefile|  49 +++
>>  lib/librte_gso/rte_gso.c   |  52 
>>  lib/librte_gso/rte_gso.h   | 145
>+
>>  lib/librte_gso/rte_gso_version.map |   7 ++
>>  mk/rte.app.mk  |   1 +
>>  10 files changed, 264 insertions(+)
>>  create mode 100644 lib/librte_gso/Makefile
>>  create mode 100644 lib/librte_gso/rte_gso.c
>>  create mode 100644 lib/librte_gso/rte_gso.h
>>  create mode 100644 lib/librte_gso/rte_gso_version.map
>>
>> diff --git a/config/common_base b/config/common_base
>> index 12f6be9..58ca5c0 100644
>> --- a/config/common_base
>> +++ b/config/common_base
>> @@ -653,6 +653,11 @@ CONFIG_RTE_LIBRTE_IP_FRAG_TBL_STAT=n
>>  CONFIG_RTE_LIBRTE_GRO=y
>>
>>  #
>> +# Compile GSO library
>> +#
>> +CONFIG_RTE_LIBRTE_GSO=y
>> +
>> +#
>>  # Compile librte_meter
>>  #
>>  CONFIG_RTE_LIBRTE_METER=y
>> diff --git a/doc/api/doxy-api-index.md b/doc/api/doxy-api-index.md
>> index 19e0d4f..6512918 100644
>> --- a/doc/api/doxy-api-index.md
>> +++ b/doc/api/doxy-api-index.md
>> @@ -101,6 +101,7 @@ The public API headers are grouped by topics:
>>[TCP](@ref rte_tcp.h),
>>[UDP](@ref rte_udp.h),
>>[GRO](@ref rte_gro.h),
>> +  [GSO](@ref rte_gso.h),
>>[frag/reass] (@ref rte_ip_frag.h),
>>[LPM IPv4 route] (@ref rte_lpm.h),
>>[LPM IPv6 route] (@ref rte_lpm6.h),
>> diff --git a/doc/api/doxy-api.conf b/doc/api/doxy-api.conf
>> index 823554f..408f2e6 100644
>> --- a/doc/api/doxy-api.conf
>> +++ b/doc/api/doxy-api.conf
>> @@ -47,6 +47,7 @@ INPUT   = doc/api/doxy-api-index.md \
>>lib/librte_ether \
>>lib/librte_eventdev \
>>lib/librte_gro \
>> +  lib/librte_gso \
>>lib/librte_hash \
>>lib/librte_ip_frag \
>>lib/librte_jobstats \
>> diff --git a/doc/guides/rel_notes/release_17_11.rst
>b/doc/guides/rel_notes/release_17_11.rst
>> index 8bf91bd..7508be7 100644
>> --- a/doc/guides/rel_notes/release_17_11.rst
>> +++ b/doc/guides/rel_notes/release_17_11.rst
>> @@ -174,6 +174,7 @@ The libraries prepended with a plus sign were
>incremented in this version.
>>   librte_ethdev.so.7

Re: [dpdk-dev] [PATCH v6 2/6] gso: add TCP/IPv4 GSO support

[Ananyev, Konstantin]

Hi Mark,

> -Original Message-
> From: Kavanagh, Mark B
> Sent: Monday, October 2, 2017 5:46 PM
> To: dev@dpdk.org
> Cc: Hu, Jiayu ; Tan, Jianfeng ; 
> Ananyev, Konstantin ; Yigit,
> Ferruh ; tho...@monjalon.net; Kavanagh, Mark B 
> 
> Subject: [PATCH v6 2/6] gso: add TCP/IPv4 GSO support
> 
> From: Jiayu Hu 
> 
> This patch adds GSO support for TCP/IPv4 packets. Supported packets
> may include a single VLAN tag. TCP/IPv4 GSO doesn't check if input
> packets have correct checksums, and doesn't update checksums for
> output packets (the responsibility for this lies with the application).
> Additionally, TCP/IPv4 GSO doesn't process IP fragmented packets.
> 
> TCP/IPv4 GSO uses two chained MBUFs, one direct MBUF and one indrect
> MBUF, to organize an output packet. Note that we refer to these two
> chained MBUFs as a two-segment MBUF. The direct MBUF stores the packet
> header, while the indirect mbuf simply points to a location within the
> original packet's payload. Consequently, use of the GSO library requires
> multi-segment MBUF support in the TX functions of the NIC driver.
> 
> If a packet is GSO'd, TCP/IPv4 GSO reduces its MBUF refcnt by 1. As a
> result, when all of its GSOed segments are freed, the packet is freed
> automatically.
> 
> Signed-off-by: Jiayu Hu 
> Signed-off-by: Mark Kavanagh 
> Tested-by: Lei Yao 
> ---
>  doc/guides/rel_notes/release_17_11.rst  |  12 +++
>  lib/librte_eal/common/include/rte_log.h |   1 +
>  lib/librte_gso/Makefile |   2 +
>  lib/librte_gso/gso_common.c | 153 
> 
>  lib/librte_gso/gso_common.h | 141 +
>  lib/librte_gso/gso_tcp4.c   | 104 ++
>  lib/librte_gso/gso_tcp4.h   |  74 +++
>  lib/librte_gso/rte_gso.c|  52 ++-
>  8 files changed, 536 insertions(+), 3 deletions(-)
>  create mode 100644 lib/librte_gso/gso_common.c
>  create mode 100644 lib/librte_gso/gso_common.h
>  create mode 100644 lib/librte_gso/gso_tcp4.c
>  create mode 100644 lib/librte_gso/gso_tcp4.h
> 
> diff --git a/doc/guides/rel_notes/release_17_11.rst 
> b/doc/guides/rel_notes/release_17_11.rst
> index 7508be7..c414f73 100644
> --- a/doc/guides/rel_notes/release_17_11.rst
> +++ b/doc/guides/rel_notes/release_17_11.rst
> @@ -41,6 +41,18 @@ New Features
>   Also, make sure to start the actual text at the margin.
>   =
> 
> +* **Added the Generic Segmentation Offload Library.**
> +
> +  Added the Generic Segmentation Offload (GSO) library to enable
> +  applications to split large packets (e.g. MTU is 64KB) into small
> +  ones (e.g. MTU is 1500B). Supported packet types are:
> +
> +  * TCP/IPv4 packets, which may include a single VLAN tag.
> +
> +  The GSO library doesn't check if the input packets have correct
> +  checksums, and doesn't update checksums for output packets.
> +  Additionally, the GSO library doesn't process IP fragmented packets.
> +
> 
>  Resolved Issues
>  ---
> diff --git a/lib/librte_eal/common/include/rte_log.h 
> b/lib/librte_eal/common/include/rte_log.h
> index ec8dba7..2fa1199 100644
> --- a/lib/librte_eal/common/include/rte_log.h
> +++ b/lib/librte_eal/common/include/rte_log.h
> @@ -87,6 +87,7 @@ struct rte_logs {
>  #define RTE_LOGTYPE_CRYPTODEV 17 /**< Log related to cryptodev. */
>  #define RTE_LOGTYPE_EFD   18 /**< Log related to EFD. */
>  #define RTE_LOGTYPE_EVENTDEV  19 /**< Log related to eventdev. */
> +#define RTE_LOGTYPE_GSO   20 /**< Log related to GSO. */
> 
>  /* these log types can be used in an application */
>  #define RTE_LOGTYPE_USER1 24 /**< User-defined log type 1. */
> diff --git a/lib/librte_gso/Makefile b/lib/librte_gso/Makefile
> index aeaacbc..2be64d1 100644
> --- a/lib/librte_gso/Makefile
> +++ b/lib/librte_gso/Makefile
> @@ -42,6 +42,8 @@ LIBABIVER := 1
> 
>  #source files
>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += rte_gso.c
> +SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_common.c
> +SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_tcp4.c
> 
>  # install this header file
>  SYMLINK-$(CONFIG_RTE_LIBRTE_GSO)-include += rte_gso.h
> diff --git a/lib/librte_gso/gso_common.c b/lib/librte_gso/gso_common.c
> new file mode 100644
> index 000..ee75d4c
> --- /dev/null
> +++ b/lib/librte_gso/gso_common.c
> @@ -0,0 +1,153 @@
> +/*-
> + *   BSD LICENSE
> + *
> + *   Copyright(c) 2017 Intel Corporation. All rights reserved.
> + *   All rights reserved.
> + *
> + *   Redistribution and use in source and binary forms, with or without
> + *   modification, are permitted provided that the following conditions
> + *   are met:
> + *
> + * * Redistributions of source code must retain the above copyright
> + *   notice, this list of conditions and the following disclaimer.
> + * * Redistributions in binary form must reproduce the above copyright
> + *   notice, this list of conditions and the following di

Re: [dpdk-dev] [PATCH v6 2/6] gso: add TCP/IPv4 GSO support

[Ananyev, Konstantin]

> -Original Message-
> From: Ananyev, Konstantin
> Sent: Wednesday, October 4, 2017 2:32 PM
> To: Kavanagh, Mark B ; dev@dpdk.org
> Cc: Hu, Jiayu ; Tan, Jianfeng ; 
> Yigit, Ferruh ; tho...@monjalon.net
> Subject: RE: [PATCH v6 2/6] gso: add TCP/IPv4 GSO support
> 
> Hi Mark,
> 
> > -Original Message-
> > From: Kavanagh, Mark B
> > Sent: Monday, October 2, 2017 5:46 PM
> > To: dev@dpdk.org
> > Cc: Hu, Jiayu ; Tan, Jianfeng ; 
> > Ananyev, Konstantin ;
> Yigit,
> > Ferruh ; tho...@monjalon.net; Kavanagh, Mark B 
> > 
> > Subject: [PATCH v6 2/6] gso: add TCP/IPv4 GSO support
> >
> > From: Jiayu Hu 
> >
> > This patch adds GSO support for TCP/IPv4 packets. Supported packets
> > may include a single VLAN tag. TCP/IPv4 GSO doesn't check if input
> > packets have correct checksums, and doesn't update checksums for
> > output packets (the responsibility for this lies with the application).
> > Additionally, TCP/IPv4 GSO doesn't process IP fragmented packets.
> >
> > TCP/IPv4 GSO uses two chained MBUFs, one direct MBUF and one indrect
> > MBUF, to organize an output packet. Note that we refer to these two
> > chained MBUFs as a two-segment MBUF. The direct MBUF stores the packet
> > header, while the indirect mbuf simply points to a location within the
> > original packet's payload. Consequently, use of the GSO library requires
> > multi-segment MBUF support in the TX functions of the NIC driver.
> >
> > If a packet is GSO'd, TCP/IPv4 GSO reduces its MBUF refcnt by 1. As a
> > result, when all of its GSOed segments are freed, the packet is freed
> > automatically.
> >
> > Signed-off-by: Jiayu Hu 
> > Signed-off-by: Mark Kavanagh 
> > Tested-by: Lei Yao 
> > ---
> >  doc/guides/rel_notes/release_17_11.rst  |  12 +++
> >  lib/librte_eal/common/include/rte_log.h |   1 +
> >  lib/librte_gso/Makefile |   2 +
> >  lib/librte_gso/gso_common.c | 153 
> > 
> >  lib/librte_gso/gso_common.h | 141 +
> >  lib/librte_gso/gso_tcp4.c   | 104 ++
> >  lib/librte_gso/gso_tcp4.h   |  74 +++
> >  lib/librte_gso/rte_gso.c|  52 ++-
> >  8 files changed, 536 insertions(+), 3 deletions(-)
> >  create mode 100644 lib/librte_gso/gso_common.c
> >  create mode 100644 lib/librte_gso/gso_common.h
> >  create mode 100644 lib/librte_gso/gso_tcp4.c
> >  create mode 100644 lib/librte_gso/gso_tcp4.h
> >
> > diff --git a/doc/guides/rel_notes/release_17_11.rst 
> > b/doc/guides/rel_notes/release_17_11.rst
> > index 7508be7..c414f73 100644
> > --- a/doc/guides/rel_notes/release_17_11.rst
> > +++ b/doc/guides/rel_notes/release_17_11.rst
> > @@ -41,6 +41,18 @@ New Features
> >   Also, make sure to start the actual text at the margin.
> >   =
> >
> > +* **Added the Generic Segmentation Offload Library.**
> > +
> > +  Added the Generic Segmentation Offload (GSO) library to enable
> > +  applications to split large packets (e.g. MTU is 64KB) into small
> > +  ones (e.g. MTU is 1500B). Supported packet types are:
> > +
> > +  * TCP/IPv4 packets, which may include a single VLAN tag.

As a nit: I think it doesn't matter as you are relying on mbuf->l2_len.
Konstantin

Re: [dpdk-dev] [PATCH v6 7/7] doc: add membership documentation

[Mcnamara, John]

> -Original Message-
> From: Wang, Yipeng1
> Sent: Wednesday, October 4, 2017 4:12 AM
> To: dev@dpdk.org
> Cc: tho...@monjalon.net; Tai, Charlie ; Gobriel,
> Sameh ; De Lara Guarch, Pablo
> ; Mcnamara, John
> ; Wang, Yipeng1 
> Subject: [PATCH v6 7/7] doc: add membership documentation
> 
> This patch adds the documentation for membership library.
> 
> Signed-off-by: Yipeng Wang 
> Reviewed-by: John McNamara 
> Reviewed-by: Pablo de Lara 

Acked-by: John McNamara 

Re: [dpdk-dev] [PATCH v6 4/4] doc: add details on ethdev offloads API

[Mcnamara, John]

> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Shahaf Shuler
> Sent: Wednesday, October 4, 2017 9:18 AM
> To: Ananyev, Konstantin ;
> tho...@monjalon.net; arybche...@solarflare.com;
> jerin.ja...@caviumnetworks.com; Yigit, Ferruh 
> Cc: dev@dpdk.org
> Subject: [dpdk-dev] [PATCH v6 4/4] doc: add details on ethdev offloads API
> 
> Add the programmers guide details on the new offloads API introduced by
> commits:
> 
> commit 67a1a59b597f ("ethdev: introduce Rx queue offloads API") commit
> f883eb32e2d4 ("ethdev: introduce Tx queue offloads API")
> 
> Signed-off-by: Shahaf Shuler 
> Reviewed-by: John McNamara 

Acked-by: John McNamara 

Re: [dpdk-dev] [PATCH v6 6/6] doc: add GSO programmer's guide

[Mcnamara, John]

> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Mark Kavanagh
> Sent: Monday, October 2, 2017 5:46 PM
> To: dev@dpdk.org
> Cc: Hu, Jiayu ; Tan, Jianfeng
> ; Ananyev, Konstantin
> ; Yigit, Ferruh ;
> tho...@monjalon.net; Kavanagh, Mark B 
> Subject: [dpdk-dev] [PATCH v6 6/6] doc: add GSO programmer's guide
> 
> Add programmer's guide doc to explain the design and use of the
> GSO library.
> 
> Signed-off-by: Mark Kavanagh 
> Signed-off-by: Jiayu Hu 

Acked-by: John McNamara 

Re: [dpdk-dev] [PATCH] eal: add doc for constructor macros

[Mcnamara, John]

> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Thomas Monjalon
> Sent: Tuesday, October 3, 2017 10:15 AM
> To: dev@dpdk.org
> Cc: Yigit, Ferruh 
> Subject: Re: [dpdk-dev] [PATCH] eal: add doc for constructor macros
> 
> 02/10/2017 20:57, Ferruh Yigit:
> > On 10/2/2017 3:59 PM, Thomas Monjalon wrote:
> > > It is a reminder that the constructors without priority get the
> > > lowest priority.
> > >
> > > Signed-off-by: Thomas Monjalon 
> >
> > Reviewed-by: Ferruh Yigit 
> 

Acked-by: John McNamara 

[dpdk-dev] [PATCH v8 1/2] net/i40e: get information about protocols defined in ddp profile

[Kirill Rybalchenko]

This patch adds new package info types to get list of protocols,
pctypes and ptypes defined in a profile

---
v3
info_size parameter always represents size of the info buffer in bytes

v6
fix bug with wrong usage of info_size parameter

v7
change misleading variable names, change order of checking variable
for zero value

v8
Fix code style warnings.

Signed-off-by: Kirill Rybalchenko 
---
 drivers/net/i40e/rte_pmd_i40e.c | 171 
 drivers/net/i40e/rte_pmd_i40e.h |  25 ++
 2 files changed, 196 insertions(+)

diff --git a/drivers/net/i40e/rte_pmd_i40e.c b/drivers/net/i40e/rte_pmd_i40e.c
index c08e07a..56a2235 100644
--- a/drivers/net/i40e/rte_pmd_i40e.c
+++ b/drivers/net/i40e/rte_pmd_i40e.c
@@ -1706,6 +1706,27 @@ rte_pmd_i40e_process_ddp_package(uint8_t port, uint8_t 
*buff,
return status;
 }
 
+/* Get number of tvl records in the section */
+static unsigned int
+i40e_get_tlv_section_size(struct i40e_profile_section_header *sec)
+{
+   unsigned int i, nb_rec, nb_tlv = 0;
+   struct i40e_profile_tlv_section_record *tlv;
+
+   if (!sec)
+   return nb_tlv;
+
+   /* get number of records in the section */
+   nb_rec = sec->section.size /
+   sizeof(struct i40e_profile_tlv_section_record);
+   for (i = 0; i < nb_rec; ) {
+   tlv = (struct i40e_profile_tlv_section_record *)&sec[1 + i];
+   i += tlv->len;
+   nb_tlv++;
+   }
+   return nb_tlv;
+}
+
 int rte_pmd_i40e_get_ddp_info(uint8_t *pkg_buff, uint32_t pkg_size,
uint8_t *info_buff, uint32_t info_size,
enum rte_pmd_i40e_package_info type)
@@ -1860,6 +1881,156 @@ int rte_pmd_i40e_get_ddp_info(uint8_t *pkg_buff, 
uint32_t pkg_size,
return I40E_SUCCESS;
}
 
+   /* get number of protocols */
+   if (type == RTE_PMD_I40E_PKG_INFO_PROTOCOL_NUM) {
+   struct i40e_profile_section_header *proto;
+
+   if (info_size < sizeof(uint32_t)) {
+   PMD_DRV_LOG(ERR, "Invalid information buffer size");
+   return -EINVAL;
+   }
+   proto = i40e_find_section_in_profile(SECTION_TYPE_PROTO,
+   (struct i40e_profile_segment *)i40e_seg_hdr);
+   *(uint32_t *)info_buff = i40e_get_tlv_section_size(proto);
+   return I40E_SUCCESS;
+   }
+
+   /* get list of protocols */
+   if (type == RTE_PMD_I40E_PKG_INFO_PROTOCOL_LIST) {
+   uint32_t i, j, nb_tlv, nb_rec, nb_proto_info;
+   struct rte_pmd_i40e_proto_info *pinfo;
+   struct i40e_profile_section_header *proto;
+   struct i40e_profile_tlv_section_record *tlv;
+
+   pinfo = (struct rte_pmd_i40e_proto_info *)info_buff;
+   nb_proto_info = info_size /
+   sizeof(struct rte_pmd_i40e_proto_info);
+   for (i = 0; i < nb_proto_info; i++) {
+   pinfo[i].proto_id = RTE_PMD_I40E_PROTO_UNUSED;
+   memset(pinfo[i].name, 0, RTE_PMD_I40E_DDP_NAME_SIZE);
+   }
+   proto = i40e_find_section_in_profile(SECTION_TYPE_PROTO,
+   (struct i40e_profile_segment *)i40e_seg_hdr);
+   nb_tlv = i40e_get_tlv_section_size(proto);
+   if (nb_tlv == 0)
+   return I40E_SUCCESS;
+   if (nb_proto_info < nb_tlv) {
+   PMD_DRV_LOG(ERR, "Invalid information buffer size");
+   return -EINVAL;
+   }
+   /* get number of records in the section */
+   nb_rec = proto->section.size /
+   sizeof(struct i40e_profile_tlv_section_record);
+   tlv = (struct i40e_profile_tlv_section_record *)&proto[1];
+   for (i = j = 0; i < nb_rec; j++) {
+   pinfo[j].proto_id = tlv->data[0];
+   strncpy(pinfo[j].name, (const char *)&tlv->data[1],
+   I40E_DDP_NAME_SIZE);
+   i += tlv->len;
+   tlv = &tlv[tlv->len];
+   }
+   return I40E_SUCCESS;
+   }
+
+   /* get number of packet classification types */
+   if (type == RTE_PMD_I40E_PKG_INFO_PCTYPE_NUM) {
+   struct i40e_profile_section_header *pctype;
+
+   if (info_size < sizeof(uint32_t)) {
+   PMD_DRV_LOG(ERR, "Invalid information buffer size");
+   return -EINVAL;
+   }
+   pctype = i40e_find_section_in_profile(SECTION_TYPE_PCTYPE,
+   (struct i40e_profile_segment *)i40e_seg_hdr);
+   *(uint32_t *)info_buff = i40e_get_tlv_section_size(pctype);
+   return I40E_SUCCESS;
+   }
+
+   /* get list of packet classification type

[dpdk-dev] [PATCH v8 2/2] app/testpmd: get information about protocols defined in ddp profile

[Kirill Rybalchenko]

Update 'ddp get info' command to display protocols defined in  a profile

v5
buff_size parameter in rte_pmd_i40e_get_ddp_info function call
always represents buffer size in bytes

v6
fix bug with wrong usage of buff_size parameter

v8
Fix code style warnings.

Signed-off-by: Kirill Rybalchenko 
---
 app/test-pmd/cmdline.c | 123 ++---
 1 file changed, 116 insertions(+), 7 deletions(-)

diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 4f2d731..71f37f1 100644
--- a/app/test-pmd/cmdline.c
+++ b/app/test-pmd/cmdline.c
@@ -13427,12 +13427,20 @@ cmd_ddp_info_parsed(
uint32_t pkg_size;
int ret = -ENOTSUP;
 #ifdef RTE_LIBRTE_I40E_PMD
-   uint32_t i;
+   uint32_t i, j, n;
uint8_t *buff;
-   uint32_t buff_size;
+   uint32_t buff_size = 0;
struct rte_pmd_i40e_profile_info info;
-   uint32_t dev_num;
+   uint32_t dev_num = 0;
struct rte_pmd_i40e_ddp_device_id *devs;
+   uint32_t proto_num = 0;
+   struct rte_pmd_i40e_proto_info *proto;
+   uint32_t pctype_num = 0;
+   struct rte_pmd_i40e_ptype_info *pctype;
+   uint32_t ptype_num = 0;
+   struct rte_pmd_i40e_ptype_info *ptype;
+   uint8_t proto_id;
+
 #endif
 
pkg = open_ddp_package_file(res->filepath, &pkg_size);
@@ -13485,12 +13493,11 @@ cmd_ddp_info_parsed(
(uint8_t *)&dev_num, sizeof(dev_num),
RTE_PMD_I40E_PKG_INFO_DEVID_NUM);
if (!ret && dev_num) {
-   devs = (struct rte_pmd_i40e_ddp_device_id *)malloc(dev_num *
-   sizeof(struct rte_pmd_i40e_ddp_device_id));
+   buff_size = dev_num * sizeof(struct rte_pmd_i40e_ddp_device_id);
+   devs = (struct rte_pmd_i40e_ddp_device_id *)malloc(buff_size);
if (devs) {
ret = rte_pmd_i40e_get_ddp_info(pkg, pkg_size,
-   (uint8_t *)devs, dev_num *
-   sizeof(struct 
rte_pmd_i40e_ddp_device_id),
+   (uint8_t *)devs, buff_size,

RTE_PMD_I40E_PKG_INFO_DEVID_LIST);
if (!ret) {
printf("List of supported devices:\n");
@@ -13506,8 +13513,110 @@ cmd_ddp_info_parsed(
free(devs);
}
}
+
+   /* get information about protocols and packet types */
+   ret = rte_pmd_i40e_get_ddp_info(pkg, pkg_size,
+   (uint8_t *)&proto_num, sizeof(proto_num),
+   RTE_PMD_I40E_PKG_INFO_PROTOCOL_NUM);
+   if (ret || !proto_num)
+   goto no_print_return;
+
+   buff_size = proto_num * sizeof(struct rte_pmd_i40e_proto_info);
+   proto = (struct rte_pmd_i40e_proto_info *)malloc(buff_size);
+   if (!proto)
+   goto no_print_return;
+
+   ret = rte_pmd_i40e_get_ddp_info(pkg, pkg_size, (uint8_t *)proto,
+   buff_size,
+   RTE_PMD_I40E_PKG_INFO_PROTOCOL_LIST);
+   if (!ret) {
+   printf("List of used protocols:\n");
+   for (i = 0; i < proto_num; i++)
+   printf("  %2u: %s\n", proto[i].proto_id,
+  proto[i].name);
+   printf("\n");
+   }
+   ret = rte_pmd_i40e_get_ddp_info(pkg, pkg_size,
+   (uint8_t *)&pctype_num, sizeof(pctype_num),
+   RTE_PMD_I40E_PKG_INFO_PCTYPE_NUM);
+   if (ret || !pctype_num)
+   goto no_print_pctypes;
+
+   buff_size = pctype_num * sizeof(struct rte_pmd_i40e_ptype_info);
+   pctype = (struct rte_pmd_i40e_ptype_info *)malloc(buff_size);
+   if (!pctype)
+   goto no_print_pctypes;
+
+   ret = rte_pmd_i40e_get_ddp_info(pkg, pkg_size, (uint8_t *)pctype,
+   buff_size,
+   RTE_PMD_I40E_PKG_INFO_PCTYPE_LIST);
+   if (ret) {
+   free(pctype);
+   goto no_print_pctypes;
+   }
+
+   printf("List of defined packet classification types:\n");
+   for (i = 0; i < pctype_num; i++) {
+   printf("  %2u:", pctype[i].ptype_id);
+   for (j = 0; j < RTE_PMD_I40E_PROTO_NUM; j++) {
+   proto_id = pctype[i].protocols[j];
+   if (proto_id != RTE_PMD_I40E_PROTO_UNUSED) {
+   for (n = 0; n < proto_num; n++) {
+   if (proto[n].proto_id == proto_id) {
+   printf(" %s", proto[n].name);
+   break;
+   }
+   }
+   }
+   }
+   printf("\n");

[dpdk-dev] [PATCH v8 0/2] et/i40e: get information about protocols defined in ddp profile

[Kirill Rybalchenko]

This patch adds ability to request information about protocols defined in 
dynamic
device personalization profile

v2:
Some code style warnings were removed

v3:
info_size parameter always represents size of the info buffer in bytes;
fix code style;

v4:
another code style fixes

v5:
in testpmd buff_size parameter in rte_pmd_i40e_get_ddp_info function call
always represents buffer size in bytes

v6:
fix bug with wrong usage of buff_size parameter

v7:
change misleading variable names, change order of checking variable
for zero value

v8:
Fix code style warnings.

Kirill Rybalchenko (2):
  net/i40e: get information about protocols defined in ddp profile
  app/testpmd: get information about protocols defined in ddp profile

 app/test-pmd/cmdline.c  | 123 +++--
 drivers/net/i40e/rte_pmd_i40e.c | 171 
 drivers/net/i40e/rte_pmd_i40e.h |  25 ++
 3 files changed, 312 insertions(+), 7 deletions(-)

-- 
2.5.5

Re: [dpdk-dev] [PATCH v6 3/6] gso: add VxLAN GSO support

[Ananyev, Konstantin]

> -Original Message-
> From: Kavanagh, Mark B
> Sent: Monday, October 2, 2017 5:46 PM
> To: dev@dpdk.org
> Cc: Hu, Jiayu ; Tan, Jianfeng ; 
> Ananyev, Konstantin ; Yigit,
> Ferruh ; tho...@monjalon.net; Kavanagh, Mark B 
> 
> Subject: [PATCH v6 3/6] gso: add VxLAN GSO support
> 
> This patch adds a framework that allows GSO on tunneled packets.
> Furthermore, it leverages that framework to provide GSO support for
> VxLAN-encapsulated packets.
> 
> Supported VxLAN packets must have an outer IPv4 header (prepended by an
> optional VLAN tag), and contain an inner TCP/IPv4 packet (with an optional
> inner VLAN tag).
> 
> VxLAN GSO doesn't check if input packets have correct checksums and
> doesn't update checksums for output packets. Additionally, it doesn't
> process IP fragmented packets.
> 
> As with TCP/IPv4 GSO, VxLAN GSO uses a two-segment MBUF to organize each
> output packet, which mandates support for multi-segment mbufs in the TX
> functions of the NIC driver. Also, if a packet is GSOed, VxLAN GSO
> reduces its MBUF refcnt by 1. As a result, when all of its GSO'd segments
> are freed, the packet is freed automatically.
> 
> Signed-off-by: Mark Kavanagh 
> Signed-off-by: Jiayu Hu 
> ---
>  doc/guides/rel_notes/release_17_11.rst |   3 +
>  lib/librte_gso/Makefile|   1 +
>  lib/librte_gso/gso_common.h|  25 +++
>  lib/librte_gso/gso_tunnel_tcp4.c   | 123 
> +
>  lib/librte_gso/gso_tunnel_tcp4.h   |  75 
>  lib/librte_gso/rte_gso.c   |  13 +++-
>  6 files changed, 237 insertions(+), 3 deletions(-)
>  create mode 100644 lib/librte_gso/gso_tunnel_tcp4.c
>  create mode 100644 lib/librte_gso/gso_tunnel_tcp4.h
> 
> diff --git a/doc/guides/rel_notes/release_17_11.rst 
> b/doc/guides/rel_notes/release_17_11.rst
> index c414f73..25b8a78 100644
> --- a/doc/guides/rel_notes/release_17_11.rst
> +++ b/doc/guides/rel_notes/release_17_11.rst
> @@ -48,6 +48,9 @@ New Features
>ones (e.g. MTU is 1500B). Supported packet types are:
> 
>* TCP/IPv4 packets, which may include a single VLAN tag.
> +  * VxLAN packets, which must have an outer IPv4 header (prepended by
> +an optional VLAN tag), and contain an inner TCP/IPv4 packet (with
> +an optional VLAN tag).
> 
>The GSO library doesn't check if the input packets have correct
>checksums, and doesn't update checksums for output packets.
> diff --git a/lib/librte_gso/Makefile b/lib/librte_gso/Makefile
> index 2be64d1..e6d41df 100644
> --- a/lib/librte_gso/Makefile
> +++ b/lib/librte_gso/Makefile
> @@ -44,6 +44,7 @@ LIBABIVER := 1
>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += rte_gso.c
>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_common.c
>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_tcp4.c
> +SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_tunnel_tcp4.c
> 
>  # install this header file
>  SYMLINK-$(CONFIG_RTE_LIBRTE_GSO)-include += rte_gso.h
> diff --git a/lib/librte_gso/gso_common.h b/lib/librte_gso/gso_common.h
> index 8d9b94e..c051295 100644
> --- a/lib/librte_gso/gso_common.h
> +++ b/lib/librte_gso/gso_common.h
> @@ -39,6 +39,7 @@
>  #include 
>  #include 
>  #include 
> +#include 
> 
>  #define IS_FRAGMENTED(frag_off) (((frag_off) & IPV4_HDR_OFFSET_MASK) != 0 \
>   || ((frag_off) & IPV4_HDR_MF_FLAG) == IPV4_HDR_MF_FLAG)
> @@ -49,6 +50,30 @@
>  #define IS_IPV4_TCP(flag) (((flag) & (PKT_TX_TCP_SEG | PKT_TX_IPV4)) == \
>   (PKT_TX_TCP_SEG | PKT_TX_IPV4))
> 
> +#define IS_IPV4_VXLAN_TCP4(flag) (((flag) & (PKT_TX_TCP_SEG | PKT_TX_IPV4 | \
> + PKT_TX_OUTER_IPV4 | PKT_TX_TUNNEL_VXLAN)) == \
> + (PKT_TX_TCP_SEG | PKT_TX_IPV4 | PKT_TX_OUTER_IPV4 | \
> +  PKT_TX_TUNNEL_VXLAN))
> +
> +/**
> + * Internal function which updates the UDP header of a packet, following
> + * segmentation. This is required to update the header's datagram length 
> field.
> + *
> + * @param pkt
> + *  The packet containing the UDP header.
> + * @param udp_offset
> + *  The offset of the UDP header from the start of the packet.
> + */
> +static inline void
> +update_udp_header(struct rte_mbuf *pkt, uint16_t udp_offset)
> +{
> + struct udp_hdr *udp_hdr;
> +
> + udp_hdr = (struct udp_hdr *)(rte_pktmbuf_mtod(pkt, char *) +
> + udp_offset);
> + udp_hdr->dgram_len = rte_cpu_to_be_16(pkt->pkt_len - udp_offset);
> +}
> +
>  /**
>   * Internal function which updates the TCP header of a packet, following
>   * segmentation. This is required to update the header's 'sent' sequence
> diff --git a/lib/librte_gso/gso_tunnel_tcp4.c 
> b/lib/librte_gso/gso_tunnel_tcp4.c
> new file mode 100644
> index 000..34bbbd7
> --- /dev/null
> +++ b/lib/librte_gso/gso_tunnel_tcp4.c
> @@ -0,0 +1,123 @@
> +/*-
> + *   BSD LICENSE
> + *
> + *   Copyright(c) 2017 Intel Corporation. All rights reserved.
> + *   All rights reserved.
> + *
> + *   Redistribution and use in source and binary forms, with or without
> + *   modificat

Re: [dpdk-dev] [PATCH 0/3] Extend port and queue attribute get functions

[Van Haaren, Harry]

> From: Eads, Gage
> Sent: Wednesday, September 20, 2017 4:21 PM
> To: dev@dpdk.org
> Cc: jerin.ja...@caviumnetworks.com; Van Haaren, Harry
> 
> Subject: [PATCH 0/3] Extend port and queue attribute get functions
> 
> This patchset makes the full queue and port configuration structures
> accessible
> through the attribute get functions, adds tests for the new attributes, and
> fixes one bug in a port attribute get test.
> 
> This patchset is based on patches 29019 through 29023.
> 
> Gage Eads (3):
>   eventdev: extend queue attribute get function
>   eventdev: extend port attribute get function
>   test/eventdev: fix bug in port enqueue depth test
> 
>  lib/librte_eventdev/rte_eventdev.c | 125 --
>  lib/librte_eventdev/rte_eventdev.h |  26 --
>  test/test/test_eventdev.c  | 179
> +++--
>  3 files changed, 245 insertions(+), 85 deletions(-)

Thanks - good to see the attrs API get more value!

Series-Reviewed-by: Harry van Haaren 

Re: [dpdk-dev] [PATCH v6 4/6] gso: add GRE GSO support

[Ananyev, Konstantin]

> diff --git a/lib/librte_gso/rte_gso.c b/lib/librte_gso/rte_gso.c
> index 6095689..b748ab1 100644
> --- a/lib/librte_gso/rte_gso.c
> +++ b/lib/librte_gso/rte_gso.c
> @@ -60,8 +60,9 @@
> 
>   if ((gso_ctx->gso_size >= pkt->pkt_len) || (gso_ctx->gso_types &
>   (DEV_TX_OFFLOAD_TCP_TSO |
> -  DEV_TX_OFFLOAD_VXLAN_TNL_TSO)) !=
> - gso_ctx->gso_types) {
> +  DEV_TX_OFFLOAD_VXLAN_TNL_TSO |
> +  DEV_TX_OFFLOAD_GRE_TNL_TSO)) !=
> +  gso_ctx->gso_types) {
>   pkt->ol_flags &= (~PKT_TX_TCP_SEG);
>   pkts_out[0] = pkt;
>   return 1;
> @@ -73,7 +74,8 @@
>   ipid_delta = (gso_ctx->ipid_flag != RTE_GSO_IPID_FIXED);
>   ol_flags = pkt->ol_flags;
> 
> - if (IS_IPV4_VXLAN_TCP4(pkt->ol_flags)) {
> + if (IS_IPV4_VXLAN_TCP4(pkt->ol_flags) ||
> + IS_IPV4_GRE_TCP4(pkt->ol_flags)) {

Same comment as for previous patch: user might want that ctx to 
Segment vxlan packets and not segment gro packets.
Konstantin

>   pkt->ol_flags &= (~PKT_TX_TCP_SEG);
>   ret = gso_tunnel_tcp4_segment(pkt, gso_size, ipid_delta,
>   direct_pool, indirect_pool,
> --
> 1.9.3

Re: [dpdk-dev] [PATCH v6 2/6] gso: add TCP/IPv4 GSO support

[Kavanagh, Mark B]

>-Original Message-
>From: Ananyev, Konstantin
>Sent: Wednesday, October 4, 2017 2:36 PM
>To: Kavanagh, Mark B ; dev@dpdk.org
>Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Yigit, Ferruh ; tho...@monjalon.net
>Subject: RE: [PATCH v6 2/6] gso: add TCP/IPv4 GSO support
>
>
>
>> -Original Message-
>> From: Ananyev, Konstantin
>> Sent: Wednesday, October 4, 2017 2:32 PM
>> To: Kavanagh, Mark B ; dev@dpdk.org
>> Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Yigit, Ferruh ; tho...@monjalon.net
>> Subject: RE: [PATCH v6 2/6] gso: add TCP/IPv4 GSO support
>>
>> Hi Mark,
>>
>> > -Original Message-
>> > From: Kavanagh, Mark B
>> > Sent: Monday, October 2, 2017 5:46 PM
>> > To: dev@dpdk.org
>> > Cc: Hu, Jiayu ; Tan, Jianfeng
>; Ananyev, Konstantin ;
>> Yigit,
>> > Ferruh ; tho...@monjalon.net; Kavanagh, Mark B
>
>> > Subject: [PATCH v6 2/6] gso: add TCP/IPv4 GSO support
>> >
>> > From: Jiayu Hu 
>> >
>> > This patch adds GSO support for TCP/IPv4 packets. Supported packets
>> > may include a single VLAN tag. TCP/IPv4 GSO doesn't check if input
>> > packets have correct checksums, and doesn't update checksums for
>> > output packets (the responsibility for this lies with the application).
>> > Additionally, TCP/IPv4 GSO doesn't process IP fragmented packets.
>> >
>> > TCP/IPv4 GSO uses two chained MBUFs, one direct MBUF and one indrect
>> > MBUF, to organize an output packet. Note that we refer to these two
>> > chained MBUFs as a two-segment MBUF. The direct MBUF stores the packet
>> > header, while the indirect mbuf simply points to a location within the
>> > original packet's payload. Consequently, use of the GSO library requires
>> > multi-segment MBUF support in the TX functions of the NIC driver.
>> >
>> > If a packet is GSO'd, TCP/IPv4 GSO reduces its MBUF refcnt by 1. As a
>> > result, when all of its GSOed segments are freed, the packet is freed
>> > automatically.
>> >
>> > Signed-off-by: Jiayu Hu 
>> > Signed-off-by: Mark Kavanagh 
>> > Tested-by: Lei Yao 
>> > ---
>> >  doc/guides/rel_notes/release_17_11.rst  |  12 +++
>> >  lib/librte_eal/common/include/rte_log.h |   1 +
>> >  lib/librte_gso/Makefile |   2 +
>> >  lib/librte_gso/gso_common.c | 153
>
>> >  lib/librte_gso/gso_common.h | 141
>+
>> >  lib/librte_gso/gso_tcp4.c   | 104 ++
>> >  lib/librte_gso/gso_tcp4.h   |  74 +++
>> >  lib/librte_gso/rte_gso.c|  52 ++-
>> >  8 files changed, 536 insertions(+), 3 deletions(-)
>> >  create mode 100644 lib/librte_gso/gso_common.c
>> >  create mode 100644 lib/librte_gso/gso_common.h
>> >  create mode 100644 lib/librte_gso/gso_tcp4.c
>> >  create mode 100644 lib/librte_gso/gso_tcp4.h
>> >
>> > diff --git a/doc/guides/rel_notes/release_17_11.rst
>b/doc/guides/rel_notes/release_17_11.rst
>> > index 7508be7..c414f73 100644
>> > --- a/doc/guides/rel_notes/release_17_11.rst
>> > +++ b/doc/guides/rel_notes/release_17_11.rst
>> > @@ -41,6 +41,18 @@ New Features
>> >   Also, make sure to start the actual text at the margin.
>> >   =
>> >
>> > +* **Added the Generic Segmentation Offload Library.**
>> > +
>> > +  Added the Generic Segmentation Offload (GSO) library to enable
>> > +  applications to split large packets (e.g. MTU is 64KB) into small
>> > +  ones (e.g. MTU is 1500B). Supported packet types are:
>> > +
>> > +  * TCP/IPv4 packets, which may include a single VLAN tag.
>
>As a nit: I think it doesn't matter as you are relying on mbuf->l2_len.
>Konstantin
>

Okay, I'll remove any mention of VLAN tags in the description - thanks!

Re: [dpdk-dev] [PATCH v6 2/6] gso: add TCP/IPv4 GSO support

[Kavanagh, Mark B]

>-Original Message-
>From: Ananyev, Konstantin
>Sent: Wednesday, October 4, 2017 2:32 PM
>To: Kavanagh, Mark B ; dev@dpdk.org
>Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Yigit, Ferruh ; tho...@monjalon.net
>Subject: RE: [PATCH v6 2/6] gso: add TCP/IPv4 GSO support
>
>Hi Mark,
>
>> -Original Message-
>> From: Kavanagh, Mark B
>> Sent: Monday, October 2, 2017 5:46 PM
>> To: dev@dpdk.org
>> Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Ananyev, Konstantin ; Yigit,
>> Ferruh ; tho...@monjalon.net; Kavanagh, Mark B
>
>> Subject: [PATCH v6 2/6] gso: add TCP/IPv4 GSO support
>>
>> From: Jiayu Hu 
>>
>> This patch adds GSO support for TCP/IPv4 packets. Supported packets
>> may include a single VLAN tag. TCP/IPv4 GSO doesn't check if input
>> packets have correct checksums, and doesn't update checksums for
>> output packets (the responsibility for this lies with the application).
>> Additionally, TCP/IPv4 GSO doesn't process IP fragmented packets.
>>
>> TCP/IPv4 GSO uses two chained MBUFs, one direct MBUF and one indrect
>> MBUF, to organize an output packet. Note that we refer to these two
>> chained MBUFs as a two-segment MBUF. The direct MBUF stores the packet
>> header, while the indirect mbuf simply points to a location within the
>> original packet's payload. Consequently, use of the GSO library requires
>> multi-segment MBUF support in the TX functions of the NIC driver.
>>
>> If a packet is GSO'd, TCP/IPv4 GSO reduces its MBUF refcnt by 1. As a
>> result, when all of its GSOed segments are freed, the packet is freed
>> automatically.
>>
>> Signed-off-by: Jiayu Hu 
>> Signed-off-by: Mark Kavanagh 
>> Tested-by: Lei Yao 
>> ---
>>  doc/guides/rel_notes/release_17_11.rst  |  12 +++
>>  lib/librte_eal/common/include/rte_log.h |   1 +
>>  lib/librte_gso/Makefile |   2 +
>>  lib/librte_gso/gso_common.c | 153
>
>>  lib/librte_gso/gso_common.h | 141 +
>>  lib/librte_gso/gso_tcp4.c   | 104 ++
>>  lib/librte_gso/gso_tcp4.h   |  74 +++
>>  lib/librte_gso/rte_gso.c|  52 ++-
>>  8 files changed, 536 insertions(+), 3 deletions(-)
>>  create mode 100644 lib/librte_gso/gso_common.c
>>  create mode 100644 lib/librte_gso/gso_common.h
>>  create mode 100644 lib/librte_gso/gso_tcp4.c
>>  create mode 100644 lib/librte_gso/gso_tcp4.h
>>
>> diff --git a/doc/guides/rel_notes/release_17_11.rst
>b/doc/guides/rel_notes/release_17_11.rst
>> index 7508be7..c414f73 100644
>> --- a/doc/guides/rel_notes/release_17_11.rst
>> +++ b/doc/guides/rel_notes/release_17_11.rst
>> @@ -41,6 +41,18 @@ New Features
>>   Also, make sure to start the actual text at the margin.
>>   =
>>
>> +* **Added the Generic Segmentation Offload Library.**
>> +
>> +  Added the Generic Segmentation Offload (GSO) library to enable
>> +  applications to split large packets (e.g. MTU is 64KB) into small
>> +  ones (e.g. MTU is 1500B). Supported packet types are:
>> +
>> +  * TCP/IPv4 packets, which may include a single VLAN tag.
>> +
>> +  The GSO library doesn't check if the input packets have correct
>> +  checksums, and doesn't update checksums for output packets.
>> +  Additionally, the GSO library doesn't process IP fragmented packets.
>> +
>>
>>  Resolved Issues
>>  ---
>> diff --git a/lib/librte_eal/common/include/rte_log.h
>b/lib/librte_eal/common/include/rte_log.h
>> index ec8dba7..2fa1199 100644
>> --- a/lib/librte_eal/common/include/rte_log.h
>> +++ b/lib/librte_eal/common/include/rte_log.h
>> @@ -87,6 +87,7 @@ struct rte_logs {
>>  #define RTE_LOGTYPE_CRYPTODEV 17 /**< Log related to cryptodev. */
>>  #define RTE_LOGTYPE_EFD   18 /**< Log related to EFD. */
>>  #define RTE_LOGTYPE_EVENTDEV  19 /**< Log related to eventdev. */
>> +#define RTE_LOGTYPE_GSO   20 /**< Log related to GSO. */
>>
>>  /* these log types can be used in an application */
>>  #define RTE_LOGTYPE_USER1 24 /**< User-defined log type 1. */
>> diff --git a/lib/librte_gso/Makefile b/lib/librte_gso/Makefile
>> index aeaacbc..2be64d1 100644
>> --- a/lib/librte_gso/Makefile
>> +++ b/lib/librte_gso/Makefile
>> @@ -42,6 +42,8 @@ LIBABIVER := 1
>>
>>  #source files
>>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += rte_gso.c
>> +SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_common.c
>> +SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_tcp4.c
>>
>>  # install this header file
>>  SYMLINK-$(CONFIG_RTE_LIBRTE_GSO)-include += rte_gso.h
>> diff --git a/lib/librte_gso/gso_common.c b/lib/librte_gso/gso_common.c
>> new file mode 100644
>> index 000..ee75d4c
>> --- /dev/null
>> +++ b/lib/librte_gso/gso_common.c
>> @@ -0,0 +1,153 @@
>> +/*-
>> + *   BSD LICENSE
>> + *
>> + *   Copyright(c) 2017 Intel Corporation. All rights reserved.
>> + *   All rights reserved.
>> + *
>> + *   Redistribution and use in source and binary forms, with or without
>> + *   modification, are permitte

Re: [dpdk-dev] [PATCH v6 3/6] gso: add VxLAN GSO support

[Kavanagh, Mark B]

>-Original Message-
>From: Ananyev, Konstantin
>Sent: Wednesday, October 4, 2017 3:12 PM
>To: Kavanagh, Mark B ; dev@dpdk.org
>Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Yigit, Ferruh ; tho...@monjalon.net
>Subject: RE: [PATCH v6 3/6] gso: add VxLAN GSO support
>
>
>
>> -Original Message-
>> From: Kavanagh, Mark B
>> Sent: Monday, October 2, 2017 5:46 PM
>> To: dev@dpdk.org
>> Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Ananyev, Konstantin ; Yigit,
>> Ferruh ; tho...@monjalon.net; Kavanagh, Mark B
>
>> Subject: [PATCH v6 3/6] gso: add VxLAN GSO support
>>
>> This patch adds a framework that allows GSO on tunneled packets.
>> Furthermore, it leverages that framework to provide GSO support for
>> VxLAN-encapsulated packets.
>>
>> Supported VxLAN packets must have an outer IPv4 header (prepended by an
>> optional VLAN tag), and contain an inner TCP/IPv4 packet (with an optional
>> inner VLAN tag).
>>
>> VxLAN GSO doesn't check if input packets have correct checksums and
>> doesn't update checksums for output packets. Additionally, it doesn't
>> process IP fragmented packets.
>>
>> As with TCP/IPv4 GSO, VxLAN GSO uses a two-segment MBUF to organize each
>> output packet, which mandates support for multi-segment mbufs in the TX
>> functions of the NIC driver. Also, if a packet is GSOed, VxLAN GSO
>> reduces its MBUF refcnt by 1. As a result, when all of its GSO'd segments
>> are freed, the packet is freed automatically.
>>
>> Signed-off-by: Mark Kavanagh 
>> Signed-off-by: Jiayu Hu 
>> ---
>>  doc/guides/rel_notes/release_17_11.rst |   3 +
>>  lib/librte_gso/Makefile|   1 +
>>  lib/librte_gso/gso_common.h|  25 +++
>>  lib/librte_gso/gso_tunnel_tcp4.c   | 123
>+
>>  lib/librte_gso/gso_tunnel_tcp4.h   |  75 
>>  lib/librte_gso/rte_gso.c   |  13 +++-
>>  6 files changed, 237 insertions(+), 3 deletions(-)
>>  create mode 100644 lib/librte_gso/gso_tunnel_tcp4.c
>>  create mode 100644 lib/librte_gso/gso_tunnel_tcp4.h
>>
>> diff --git a/doc/guides/rel_notes/release_17_11.rst
>b/doc/guides/rel_notes/release_17_11.rst
>> index c414f73..25b8a78 100644
>> --- a/doc/guides/rel_notes/release_17_11.rst
>> +++ b/doc/guides/rel_notes/release_17_11.rst
>> @@ -48,6 +48,9 @@ New Features
>>ones (e.g. MTU is 1500B). Supported packet types are:
>>
>>* TCP/IPv4 packets, which may include a single VLAN tag.
>> +  * VxLAN packets, which must have an outer IPv4 header (prepended by
>> +an optional VLAN tag), and contain an inner TCP/IPv4 packet (with
>> +an optional VLAN tag).
>>
>>The GSO library doesn't check if the input packets have correct
>>checksums, and doesn't update checksums for output packets.
>> diff --git a/lib/librte_gso/Makefile b/lib/librte_gso/Makefile
>> index 2be64d1..e6d41df 100644
>> --- a/lib/librte_gso/Makefile
>> +++ b/lib/librte_gso/Makefile
>> @@ -44,6 +44,7 @@ LIBABIVER := 1
>>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += rte_gso.c
>>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_common.c
>>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_tcp4.c
>> +SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_tunnel_tcp4.c
>>
>>  # install this header file
>>  SYMLINK-$(CONFIG_RTE_LIBRTE_GSO)-include += rte_gso.h
>> diff --git a/lib/librte_gso/gso_common.h b/lib/librte_gso/gso_common.h
>> index 8d9b94e..c051295 100644
>> --- a/lib/librte_gso/gso_common.h
>> +++ b/lib/librte_gso/gso_common.h
>> @@ -39,6 +39,7 @@
>>  #include 
>>  #include 
>>  #include 
>> +#include 
>>
>>  #define IS_FRAGMENTED(frag_off) (((frag_off) & IPV4_HDR_OFFSET_MASK) != 0 \
>>  || ((frag_off) & IPV4_HDR_MF_FLAG) == IPV4_HDR_MF_FLAG)
>> @@ -49,6 +50,30 @@
>>  #define IS_IPV4_TCP(flag) (((flag) & (PKT_TX_TCP_SEG | PKT_TX_IPV4)) == \
>>  (PKT_TX_TCP_SEG | PKT_TX_IPV4))
>>
>> +#define IS_IPV4_VXLAN_TCP4(flag) (((flag) & (PKT_TX_TCP_SEG | PKT_TX_IPV4 |
>\
>> +PKT_TX_OUTER_IPV4 | PKT_TX_TUNNEL_VXLAN)) == \
>> +(PKT_TX_TCP_SEG | PKT_TX_IPV4 | PKT_TX_OUTER_IPV4 | \
>> + PKT_TX_TUNNEL_VXLAN))
>> +
>> +/**
>> + * Internal function which updates the UDP header of a packet, following
>> + * segmentation. This is required to update the header's datagram length
>field.
>> + *
>> + * @param pkt
>> + *  The packet containing the UDP header.
>> + * @param udp_offset
>> + *  The offset of the UDP header from the start of the packet.
>> + */
>> +static inline void
>> +update_udp_header(struct rte_mbuf *pkt, uint16_t udp_offset)
>> +{
>> +struct udp_hdr *udp_hdr;
>> +
>> +udp_hdr = (struct udp_hdr *)(rte_pktmbuf_mtod(pkt, char *) +
>> +udp_offset);
>> +udp_hdr->dgram_len = rte_cpu_to_be_16(pkt->pkt_len - udp_offset);
>> +}
>> +
>>  /**
>>   * Internal function which updates the TCP header of a packet, following
>>   * segmentation. This is required to update the header's 'sent' sequence
>> diff --git a/lib/librte_gso/gso_tunnel_tcp4.c
>b/lib/librte_gso/gso_tunnel_tcp4.c
>> new 

Re: [dpdk-dev] [PATCH v6 4/6] gso: add GRE GSO support

[Kavanagh, Mark B]

>-Original Message-
>From: Ananyev, Konstantin
>Sent: Wednesday, October 4, 2017 3:16 PM
>To: Kavanagh, Mark B ; dev@dpdk.org
>Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Yigit, Ferruh ; tho...@monjalon.net
>Subject: RE: [PATCH v6 4/6] gso: add GRE GSO support
>
>
>> diff --git a/lib/librte_gso/rte_gso.c b/lib/librte_gso/rte_gso.c
>> index 6095689..b748ab1 100644
>> --- a/lib/librte_gso/rte_gso.c
>> +++ b/lib/librte_gso/rte_gso.c
>> @@ -60,8 +60,9 @@
>>
>>  if ((gso_ctx->gso_size >= pkt->pkt_len) || (gso_ctx->gso_types &
>>  (DEV_TX_OFFLOAD_TCP_TSO |
>> - DEV_TX_OFFLOAD_VXLAN_TNL_TSO)) !=
>> -gso_ctx->gso_types) {
>> + DEV_TX_OFFLOAD_VXLAN_TNL_TSO |
>> + DEV_TX_OFFLOAD_GRE_TNL_TSO)) !=
>> + gso_ctx->gso_types) {
>>  pkt->ol_flags &= (~PKT_TX_TCP_SEG);
>>  pkts_out[0] = pkt;
>>  return 1;
>> @@ -73,7 +74,8 @@
>>  ipid_delta = (gso_ctx->ipid_flag != RTE_GSO_IPID_FIXED);
>>  ol_flags = pkt->ol_flags;
>>
>> -if (IS_IPV4_VXLAN_TCP4(pkt->ol_flags)) {
>> +if (IS_IPV4_VXLAN_TCP4(pkt->ol_flags) ||
>> +IS_IPV4_GRE_TCP4(pkt->ol_flags)) {
>
>Same comment as for previous patch: user might want that ctx to
>Segment vxlan packets and not segment gro packets.
>Konstantin

Thanks Konstantin - I'll update appropriately.

>
>>  pkt->ol_flags &= (~PKT_TX_TCP_SEG);
>>  ret = gso_tunnel_tcp4_segment(pkt, gso_size, ipid_delta,
>>  direct_pool, indirect_pool,
>> --
>> 1.9.3

Re: [dpdk-dev] [PATCH v6 2/6] gso: add TCP/IPv4 GSO support

[Ananyev, Konstantin]

> >>  int
> >>  rte_gso_segment(struct rte_mbuf *pkt,
> >> @@ -41,12 +46,53 @@
> >>struct rte_mbuf **pkts_out,
> >>uint16_t nb_pkts_out)
> >>  {
> >> +  struct rte_mempool *direct_pool, *indirect_pool;
> >> +  struct rte_mbuf *pkt_seg;
> >> +  uint64_t ol_flags;
> >> +  uint16_t gso_size;
> >> +  uint8_t ipid_delta;
> >> +  int ret = 1;
> >> +
> >>if (pkt == NULL || pkts_out == NULL || gso_ctx == NULL ||
> >>nb_pkts_out < 1)
> >>return -EINVAL;
> >>
> >> -  pkt->ol_flags &= (~PKT_TX_TCP_SEG);
> >> -  pkts_out[0] = pkt;
> >> +  if ((gso_ctx->gso_size >= pkt->pkt_len) || (gso_ctx->gso_types &
> >> +  DEV_TX_OFFLOAD_TCP_TSO) !=
> >> +  gso_ctx->gso_types) {
> >> +  pkt->ol_flags &= (~PKT_TX_TCP_SEG);
> >> +  pkts_out[0] = pkt;
> >> +  return 1;
> >> +  }
> >> +
> >> +  direct_pool = gso_ctx->direct_pool;
> >> +  indirect_pool = gso_ctx->indirect_pool;
> >> +  gso_size = gso_ctx->gso_size;
> >> +  ipid_delta = (gso_ctx->ipid_flag != RTE_GSO_IPID_FIXED);
> >> +  ol_flags = pkt->ol_flags;
> >> +
> >> +  if (IS_IPV4_TCP(pkt->ol_flags)) {
> >> +  pkt->ol_flags &= (~PKT_TX_TCP_SEG);
> >> +  ret = gso_tcp4_segment(pkt, gso_size, ipid_delta,
> >> +  direct_pool, indirect_pool,
> >> +  pkts_out, nb_pkts_out);
> >> +  } else {
> >> +  pkt->ol_flags &= (~PKT_TX_TCP_SEG);
> >
> >Not sure why do you clean this flag if you don't support that packet type
> >and no action was perfomed?
> >Suppose you have a mix ipv4 and ipv6 packets - gso lib would do ipv4 and
> >someone else
> >(HW?) can do ipv4 segmentation.
> 
> I can't say for definite, since I didn't implement this change. However, I 
> can only presume that the assumption here is that since
> segmentation is being done in S/W that the underlying H/W does not support 
> TSO.
> Since the underlying HW can't segment the packet in HW, we should clear the 
> flag; otherwise, if an mbuf marked for TCP segmentation is
> passed to the driver of a NIC that does not support/understand that feature, 
> the behavior is undefined.
> Is this a fair assumption in your opinion, or is it the case that the packet 
> would simply be transmitted un-segmented in that case, and so we
> shouldn't clear the flag?

Yes, I think if we shouldn't clear the flag if we didn't do any segmentation 
(we just encounter a packet type that we don't support).
Konstantin

> 
> Thanks again,
> Mark
> 
> >BTW, did you notice that building of shared target fails?
> >Konstantin
> 
> I didn't, but I'll take a look right now - thanks for the catch!
> 
> >
> >
> >> +  pkts_out[0] = pkt;
> >> +  RTE_LOG(WARNING, GSO, "Unsupported packet type\n");
> >> +  return 1;
> >> +  }
> >> +
> >> +  if (ret > 1) {
> >> +  pkt_seg = pkt;
> >> +  while (pkt_seg) {
> >> +  rte_mbuf_refcnt_update(pkt_seg, -1);
> >> +  pkt_seg = pkt_seg->next;
> >> +  }
> >> +  } else if (ret < 0) {
> >> +  /* Revert the ol_flags in the event of failure. */
> >> +  pkt->ol_flags = ol_flags;
> >> +  }
> >>
> >> -  return 1;
> >> +  return ret;
> >>  }
> >> --
> >> 1.9.3

Re: [dpdk-dev] [PATCH v6 2/6] gso: add TCP/IPv4 GSO support

[Kavanagh, Mark B]

>From: Ananyev, Konstantin
>Sent: Wednesday, October 4, 2017 3:49 PM
>To: Kavanagh, Mark B ; dev@dpdk.org
>Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Yigit, Ferruh ; tho...@monjalon.net
>Subject: RE: [PATCH v6 2/6] gso: add TCP/IPv4 GSO support
>
>> >>  int
>> >>  rte_gso_segment(struct rte_mbuf *pkt,
>> >> @@ -41,12 +46,53 @@
>> >>   struct rte_mbuf **pkts_out,
>> >>   uint16_t nb_pkts_out)
>> >>  {
>> >> + struct rte_mempool *direct_pool, *indirect_pool;
>> >> + struct rte_mbuf *pkt_seg;
>> >> + uint64_t ol_flags;
>> >> + uint16_t gso_size;
>> >> + uint8_t ipid_delta;
>> >> + int ret = 1;
>> >> +
>> >>   if (pkt == NULL || pkts_out == NULL || gso_ctx == NULL ||
>> >>   nb_pkts_out < 1)
>> >>   return -EINVAL;
>> >>
>> >> - pkt->ol_flags &= (~PKT_TX_TCP_SEG);
>> >> - pkts_out[0] = pkt;
>> >> + if ((gso_ctx->gso_size >= pkt->pkt_len) || (gso_ctx->gso_types &
>> >> + DEV_TX_OFFLOAD_TCP_TSO) !=
>> >> + gso_ctx->gso_types) {
>> >> + pkt->ol_flags &= (~PKT_TX_TCP_SEG);
>> >> + pkts_out[0] = pkt;
>> >> + return 1;
>> >> + }
>> >> +
>> >> + direct_pool = gso_ctx->direct_pool;
>> >> + indirect_pool = gso_ctx->indirect_pool;
>> >> + gso_size = gso_ctx->gso_size;
>> >> + ipid_delta = (gso_ctx->ipid_flag != RTE_GSO_IPID_FIXED);
>> >> + ol_flags = pkt->ol_flags;
>> >> +
>> >> + if (IS_IPV4_TCP(pkt->ol_flags)) {
>> >> + pkt->ol_flags &= (~PKT_TX_TCP_SEG);
>> >> + ret = gso_tcp4_segment(pkt, gso_size, ipid_delta,
>> >> + direct_pool, indirect_pool,
>> >> + pkts_out, nb_pkts_out);
>> >> + } else {
>> >> + pkt->ol_flags &= (~PKT_TX_TCP_SEG);
>> >
>> >Not sure why do you clean this flag if you don't support that packet type
>> >and no action was perfomed?
>> >Suppose you have a mix ipv4 and ipv6 packets - gso lib would do ipv4 and
>> >someone else
>> >(HW?) can do ipv4 segmentation.
>>
>> I can't say for definite, since I didn't implement this change. However, I
>can only presume that the assumption here is that since
>> segmentation is being done in S/W that the underlying H/W does not support
>TSO.
>> Since the underlying HW can't segment the packet in HW, we should clear the
>flag; otherwise, if an mbuf marked for TCP segmentation is
>> passed to the driver of a NIC that does not support/understand that feature,
>the behavior is undefined.
>> Is this a fair assumption in your opinion, or is it the case that the packet
>would simply be transmitted un-segmented in that case, and so we
>> shouldn't clear the flag?
>
>Yes, I think if we shouldn't clear the flag if we didn't do any segmentation
>(we just encounter a packet type that we don't support).
>Konstantin

Okay, thanks for clarifying - I'll update the code accordingly.
-Mark

>
>>
>> Thanks again,
>> Mark
>>
>> >BTW, did you notice that building of shared target fails?
>> >Konstantin
>>
>> I didn't, but I'll take a look right now - thanks for the catch!
>>
>> >
>> >
>> >> + pkts_out[0] = pkt;
>> >> + RTE_LOG(WARNING, GSO, "Unsupported packet type\n");
>> >> + return 1;
>> >> + }
>> >> +
>> >> + if (ret > 1) {
>> >> + pkt_seg = pkt;
>> >> + while (pkt_seg) {
>> >> + rte_mbuf_refcnt_update(pkt_seg, -1);
>> >> + pkt_seg = pkt_seg->next;
>> >> + }
>> >> + } else if (ret < 0) {
>> >> + /* Revert the ol_flags in the event of failure. */
>> >> + pkt->ol_flags = ol_flags;
>> >> + }
>> >>
>> >> - return 1;
>> >> + return ret;
>> >>  }
>> >> --
>> >> 1.9.3

Re: [dpdk-dev] [PATCH v6 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO

[Ananyev, Konstantin]

> -Original Message-
> From: Kavanagh, Mark B
> Sent: Monday, October 2, 2017 5:46 PM
> To: dev@dpdk.org
> Cc: Hu, Jiayu ; Tan, Jianfeng ; 
> Ananyev, Konstantin ; Yigit,
> Ferruh ; tho...@monjalon.net; Kavanagh, Mark B 
> 
> Subject: [PATCH v6 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO
> 
> From: Jiayu Hu 
> 
> This patch adds GSO support to the csum forwarding engine. Oversized
> packets transmitted over a GSO-enabled port will undergo segmentation
> (with the exception of packet-types unsupported by the GSO library).
> GSO support is disabled by default.
> 
> GSO support may be toggled on a per-port basis, using the command:
> 
> "set port  gso on|off"
> 
> The maximum packet length (including the packet header and payload) for
> GSO segments may be set with the command:
> 
> "set gso segsz "
> 
> Show GSO configuration for a given port with the command:
> 
>   "show port  gso"
> 
> Signed-off-by: Jiayu Hu 
> Signed-off-by: Mark Kavanagh 
> ---
>  app/test-pmd/cmdline.c  | 178 
> 
>  app/test-pmd/config.c   |  24 
>  app/test-pmd/csumonly.c |  69 ++-
>  app/test-pmd/testpmd.c  |  13 ++
>  app/test-pmd/testpmd.h  |  10 ++
>  doc/guides/testpmd_app_ug/testpmd_funcs.rst |  46 +++
>  6 files changed, 335 insertions(+), 5 deletions(-)
> 
> diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
> index ccdf239..05b0ce8 100644
> --- a/app/test-pmd/cmdline.c
> +++ b/app/test-pmd/cmdline.c
> @@ -431,6 +431,17 @@ static void cmd_help_long_parsed(void *parsed_result,
>   "Set max flow number and max packet number per-flow"
>   " for GRO.\n\n"
> 
> + "set port (port_id) gso (on|off)"
> + "Enable or disable Generic Segmentation Offload in"
> + " csum forwarding engine.\n\n"
> +
> + "set gso segsz (length)\n"
> + "Set max packet length for output GSO segments,"
> + " including packet header and payload.\n\n"

Probably a  good future improvement would be to allow user to specify gso_type 
too.

> +
> + "show port (port_id) gso\n"
> + "Show GSO configuration.\n\n"
> +
>   "set fwd (%s)\n"
>   "Set packet forwarding mode.\n\n"
> 
> @@ -3967,6 +3978,170 @@ struct cmd_gro_set_result {
>   },
>  };
> 
> +/* *** ENABLE/DISABLE GSO *** */
> +struct cmd_gso_enable_result {
> + cmdline_fixed_string_t cmd_set;
> + cmdline_fixed_string_t cmd_port;
> + cmdline_fixed_string_t cmd_keyword;
> + cmdline_fixed_string_t cmd_mode;
> + uint8_t cmd_pid;
> +};
> +
> +static void
> +cmd_gso_enable_parsed(void *parsed_result,
> + __attribute__((unused)) struct cmdline *cl,
> + __attribute__((unused)) void *data)
> +{
> + struct cmd_gso_enable_result *res;
> +
> + res = parsed_result;
> + if (!strcmp(res->cmd_keyword, "gso"))
> + setup_gso(res->cmd_mode, res->cmd_pid);
> +}
> +
> +cmdline_parse_token_string_t cmd_gso_enable_set =
> + TOKEN_STRING_INITIALIZER(struct cmd_gso_enable_result,
> + cmd_set, "set");
> +cmdline_parse_token_string_t cmd_gso_enable_port =
> + TOKEN_STRING_INITIALIZER(struct cmd_gso_enable_result,
> + cmd_port, "port");
> +cmdline_parse_token_string_t cmd_gso_enable_keyword =
> + TOKEN_STRING_INITIALIZER(struct cmd_gso_enable_result,
> + cmd_keyword, "gso");
> +cmdline_parse_token_string_t cmd_gso_enable_mode =
> + TOKEN_STRING_INITIALIZER(struct cmd_gso_enable_result,
> + cmd_mode, "on#off");
> +cmdline_parse_token_num_t cmd_gso_enable_pid =
> + TOKEN_NUM_INITIALIZER(struct cmd_gso_enable_result,
> + cmd_pid, UINT8);
> +
> +cmdline_parse_inst_t cmd_gso_enable = {
> + .f = cmd_gso_enable_parsed,
> + .data = NULL,
> + .help_str = "set port  gso on|off",
> + .tokens = {
> + (void *)&cmd_gso_enable_set,
> + (void *)&cmd_gso_enable_port,
> + (void *)&cmd_gso_enable_pid,
> + (void *)&cmd_gso_enable_keyword,
> + (void *)&cmd_gso_enable_mode,
> + NULL,
> + },
> +};
> +
> +/* *** SET MAX PACKET LENGTH FOR GSO SEGMENTS *** */
> +struct cmd_gso_size_result {
> + cmdline_fixed_string_t cmd_set;
> + cmdline_fixed_string_t cmd_keyword;
> + cmdline_fixed_string_t cmd_segsz;
> + uint16_t cmd_size;
> +};
> +
> +static void
> +cmd_gso_size_parsed(void *parsed_result,
> +__attribute__((unused)) struct cmdline *cl,
> +__attribute__((unused)) void *data)
> +{
> + struct cmd_gso_size_result *res = parsed_result;
> +
> + if (test_done == 0) {
> + p

[dpdk-dev] [PATCH v5 0/9] Policy Based Power Control for Guest

[David Hunt]

Policy Based Power Control for Guest

This patchset adds the facility for a guest VM to send a policy down to the
host that will allow the host to scale up/down cpu frequencies
depending on the policy criteria independently of the DPDK app running in
the guest.  This differs from the previous vm_power implementation where
individual scale up/down requests were send from the guest to the host via
virtio-serial.

V5 patchset changes:
  * Removed most of the #ifdef I40_PMD from the example code as it will
be applicable to other PMDs in the future.
  * Changed the parameter of rte_pmd_i40e_query_vfid_by_mac from a uint64
to a const struct ether_addr *, rather than casting it later in the
function.

V4 patchset changes:
  * None, re-post to mailing list under the correct email thread.

V3 patchset changes:
  * Changed to using is_same_ether_addr() instead of looping through
the mac address bytes to compare them.
  * Tweaked some comments and working in the i40e patch after review.
  * Added a patch to the set to add new i40e function to map file, so
as to allow shared library builds. The power library API needs a cleanup
in next release, so will add API/ABI warning for this cleanup in a
separate patch.

V2 patchset changes:
  * Removed API's in ethdev layer.
  * Now just a single new API in the i40e driver for mapping VF MAC to
VF index.
  * Moved new function from rte_rxtx.c to rte_pmd_i40e.c
  * Removed function for reading i40e register, moved to using the
standard stats API.
  * Renamed i40e function to rte_pmd_i40e_query_vfid_by_mac
  * Cleaned up policy generation code.

It's a modification of the vm_power_manager app that runs in the host, and
the guest_vm_power_app example app that runs in the guest. This allows the
guest to send down a policy to the host via virtio-serial, which then allows
the host to scale up/down based on the criteria in the policy, resulting in
quicker scale up/down than individual requests coming from the guest.
It also means that the DPDK application running in the guest does not need
to be modified in any way, it is unaware that it's cores are being scaled
up/down, reducing the effort in implementing a power-aware infrastructure.

The usage model is as follows:
1. Set up the VF's and assign to the guest in the usual way.
2. run vm_power_manager on the host, creating a channel to the guest.
3. Start the guest_vm_power_mgr app on the guest, which establishes
   a virtio-serial channel to the host.
4. Send down the profile for the guest using the "send_profile now" command.
   There is an example profile hard-coded into guest_vm_power_mgr.
5. Stop the guest_vm_power_mgr and run your normal power-unaware application.
6. Send traffic into the VFs at varying traffic rates.
   Observe the frequency change on the host (turbostat -i 1)

The sequence of code changes are as follows:

A new function has been aded to the i40e driver to allow mapping of
a VF MAC to VF index.

Next we make an addition to librte_power that adds an extra command to allow
the passing of a policy structure from the guest to the host. This struct
contains information like busy/quiet hour, packet throughput thresholds, etc.

The next addition adds functionality to convert the virtual CPU (vcpU0 IDs to
physical CPU (pcpu) IDs so that the host can scale up/down the cores used
in the guest.

The remaining patches are functionality to process the policy, and take action
when the relevant trigger occurs to cause a frequency change.

[1/9] net/i40e: add API to convert VF MAC to VF id
[2/9] lib/librte_power: add extra msg type for policies
[3/9] examples/vm_power_mgr: add vcpu to pcpu mapping
[4/9] examples/vm_power_mgr: add scale to medium freq fn
[5/9] examples/vm_power_mgr: add policy to channels
[6/9] examples/vm_power_mgr: add port initialisation
[7/9] power: add send channel msg function to map file
[8/9] examples/guest_cli: add send policy to host
[9/9] examples/vm_power_mgr: set MAC address of VF

[dpdk-dev] [PATCH v5 1/9] net/i40e: add API to convert VF MAC to VF id

[David Hunt]

From: "Sexton, Rory" 

Need a way to convert a vf id to a pf id on the host so as to query the pf
for relevant statistics which are used for the frequency changes in the
vm_power_manager app. Used when profiles are passed down from the guest
to the host, allowing the host to map the vfs to pfs.

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: Rory Sexton 
Signed-off-by: David Hunt 
---
 drivers/net/i40e/rte_pmd_i40e.c   | 30 ++
 drivers/net/i40e/rte_pmd_i40e.h   | 14 ++
 drivers/net/i40e/rte_pmd_i40e_version.map |  7 +++
 3 files changed, 51 insertions(+)

diff --git a/drivers/net/i40e/rte_pmd_i40e.c b/drivers/net/i40e/rte_pmd_i40e.c
index f12b7f4..08e6b16 100644
--- a/drivers/net/i40e/rte_pmd_i40e.c
+++ b/drivers/net/i40e/rte_pmd_i40e.c
@@ -2115,3 +2115,33 @@ int rte_pmd_i40e_ptype_mapping_replace(uint8_t port,
 
return 0;
 }
+
+uint64_t
+rte_pmd_i40e_query_vfid_by_mac(uint8_t port, const struct ether_addr *vf_mac)
+{
+   struct rte_eth_dev *dev;
+   struct ether_addr *mac;
+   struct i40e_pf *pf;
+   int vf_id;
+   struct i40e_pf_vf *vf;
+   uint16_t vf_num;
+
+   RTE_ETH_VALID_PORTID_OR_ERR_RET(port, -ENODEV);
+   dev = &rte_eth_devices[port];
+
+   if (!is_i40e_supported(dev))
+   return -ENOTSUP;
+
+   pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+   vf_num = pf->vf_num;
+
+   for (vf_id = 0; vf_id < vf_num; vf_id++) {
+   vf = &pf->vfs[vf_id];
+   mac = &vf->mac_addr;
+
+   if (is_same_ether_addr(mac, vf_mac))
+   return vf_id;
+   }
+
+   return -EINVAL;
+}
diff --git a/drivers/net/i40e/rte_pmd_i40e.h b/drivers/net/i40e/rte_pmd_i40e.h
index 356fa89..9798103 100644
--- a/drivers/net/i40e/rte_pmd_i40e.h
+++ b/drivers/net/i40e/rte_pmd_i40e.h
@@ -637,4 +637,18 @@ int rte_pmd_i40e_ptype_mapping_replace(uint8_t port,
   uint8_t mask,
   uint32_t pkt_type);
 
+/**
+ * On the PF, find VF index based on VF MAC address
+ *
+ * @param port
+ *pointer to port identifier of the device
+ * @param vf_mac
+ *the mac address of the vf to determine index of
+ * @return
+ *-(-22 EINVAL) the vf mac does not exist on this port
+ *-(!-22) the index of vfid in pf->vfs
+ */
+uint64_t rte_pmd_i40e_query_vfid_by_mac(uint8_t port,
+   const struct ether_addr *vf_mac);
+
 #endif /* _PMD_I40E_H_ */
diff --git a/drivers/net/i40e/rte_pmd_i40e_version.map 
b/drivers/net/i40e/rte_pmd_i40e_version.map
index 20cc980..d8b74bd 100644
--- a/drivers/net/i40e/rte_pmd_i40e_version.map
+++ b/drivers/net/i40e/rte_pmd_i40e_version.map
@@ -45,3 +45,10 @@ DPDK_17.08 {
rte_pmd_i40e_get_ddp_info;
 
 } DPDK_17.05;
+
+DPDK_17.11 {
+   global:
+
+   rte_pmd_i40e_query_vfid_by_mac;
+
+} DPDK_17.08;
-- 
2.7.4

[dpdk-dev] [PATCH v5 3/9] examples/vm_power_mgr: add vcpu to pcpu mapping

[David Hunt]

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: Rory Sexton 
Signed-off-by: David Hunt 
---
 examples/vm_power_manager/channel_manager.c | 62 +
 examples/vm_power_manager/channel_manager.h | 25 
 2 files changed, 87 insertions(+)

diff --git a/examples/vm_power_manager/channel_manager.c 
b/examples/vm_power_manager/channel_manager.c
index e068ae2..03fa626 100644
--- a/examples/vm_power_manager/channel_manager.c
+++ b/examples/vm_power_manager/channel_manager.c
@@ -574,6 +574,68 @@ set_channel_status(const char *vm_name, unsigned 
*channel_list,
return num_channels_changed;
 }
 
+void
+get_all_vm(int *num_vm, int *num_cpu)
+{
+
+   virNodeInfo node_info;
+   virDomainPtr *domptr;
+   uint64_t mask;
+   int i, ii, numVcpus[MAX_VCPUS], cpu, n_vcpus;
+   unsigned int jj;
+   const char *vm_name;
+   unsigned int flags = VIR_CONNECT_LIST_DOMAINS_RUNNING |
+   VIR_CONNECT_LIST_DOMAINS_PERSISTENT;
+   unsigned int flag = VIR_DOMAIN_VCPU_CONFIG;
+
+
+   memset(global_cpumaps, 0, CHANNEL_CMDS_MAX_CPUS*global_maplen);
+   if (virNodeGetInfo(global_vir_conn_ptr, &node_info))
+   RTE_LOG(ERR, CHANNEL_MANAGER, "Unable to retrieve node Info\n");
+
+   /* Returns number of pcpus */
+   global_n_host_cpus = (unsigned int)node_info.cpus;
+
+   /* Returns number of active domains */
+   *num_vm = virConnectListAllDomains(global_vir_conn_ptr, &domptr, flags);
+   if (*num_vm <= 0)
+   RTE_LOG(ERR, CHANNEL_MANAGER, "No Active Domains Running\n");
+
+   for (i = 0; i < *num_vm; i++) {
+
+   /* Get Domain Names */
+   vm_name = virDomainGetName(domptr[i]);
+   lvm_info[i].vm_name = vm_name;
+
+   /* Get Number of Vcpus */
+   numVcpus[i] = virDomainGetVcpusFlags(domptr[i], flag);
+
+   /* Get Number of VCpus & VcpuPinInfo */
+   n_vcpus = virDomainGetVcpuPinInfo(domptr[i],
+   numVcpus[i], global_cpumaps,
+   global_maplen, flag);
+
+   if ((int)n_vcpus > 0) {
+   *num_cpu = n_vcpus;
+   lvm_info[i].num_cpus = n_vcpus;
+   }
+
+   /* Save pcpu in use by libvirt VMs */
+   for (ii = 0; ii < n_vcpus; ii++) {
+   mask = 0;
+   for (jj = 0; jj < global_n_host_cpus; jj++) {
+   if (VIR_CPU_USABLE(global_cpumaps,
+   global_maplen, ii, jj) > 0) {
+   mask |= 1ULL << jj;
+   }
+   }
+   ITERATIVE_BITMASK_CHECK_64(mask, cpu) {
+   lvm_info[i].pcpus[ii] = cpu;
+   }
+   }
+   }
+}
+
 int
 get_info_vm(const char *vm_name, struct vm_info *info)
 {
diff --git a/examples/vm_power_manager/channel_manager.h 
b/examples/vm_power_manager/channel_manager.h
index 47c3b9c..788c1e6 100644
--- a/examples/vm_power_manager/channel_manager.h
+++ b/examples/vm_power_manager/channel_manager.h
@@ -66,6 +66,17 @@ struct sockaddr_un _sockaddr_un;
 #define UNIX_PATH_MAX sizeof(_sockaddr_un.sun_path)
 #endif
 
+#define MAX_VMS 4
+#define MAX_VCPUS 20
+
+
+struct libvirt_vm_info {
+   const char *vm_name;
+   unsigned int pcpus[MAX_VCPUS];
+   uint8_t num_cpus;
+};
+
+struct libvirt_vm_info lvm_info[MAX_VMS];
 /* Communication Channel Status */
 enum channel_status { CHANNEL_MGR_CHANNEL_DISCONNECTED = 0,
CHANNEL_MGR_CHANNEL_CONNECTED,
@@ -319,6 +330,20 @@ int set_channel_status(const char *vm_name, unsigned 
*channel_list,
  */
 int get_info_vm(const char *vm_name, struct vm_info *info);
 
+/**
+ * Populates a table with all domains running and their physical cpu.
+ * All information is gathered through libvirt api.
+ *
+ * @param noVms
+ *  modified to store number of active VMs
+ *
+ * @param noVcpus
+modified to store number of vcpus active
+ *
+ * @return
+ *   void
+ */
+void get_all_vm(int *noVms, int *noVcpus);
 #ifdef __cplusplus
 }
 #endif
-- 
2.7.4

[dpdk-dev] [PATCH v5 2/9] lib/librte_power: add extra msg type for policies

[David Hunt]

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: Rory Sexton 
Signed-off-by: David Hunt 
---
 lib/librte_power/channel_commands.h | 42 +
 1 file changed, 42 insertions(+)

diff --git a/lib/librte_power/channel_commands.h 
b/lib/librte_power/channel_commands.h
index 484085b..020d9fe 100644
--- a/lib/librte_power/channel_commands.h
+++ b/lib/librte_power/channel_commands.h
@@ -46,6 +46,7 @@ extern "C" {
 /* Valid Commands */
 #define CPU_POWER   1
 #define CPU_POWER_CONNECT   2
+#define PKT_POLICY  3
 
 /* CPU Power Command Scaling */
 #define CPU_POWER_SCALE_UP  1
@@ -54,11 +55,52 @@ extern "C" {
 #define CPU_POWER_SCALE_MIN 4
 #define CPU_POWER_ENABLE_TURBO  5
 #define CPU_POWER_DISABLE_TURBO 6
+#define HOURS 24
+
+#define MAX_VFS 10
+
+#define MAX_VCPU_PER_VM 8
+
+typedef enum {false, true} bool;
+
+struct t_boost_status {
+   bool tbEnabled;
+};
+
+struct timer_profile {
+   int busy_hours[HOURS];
+   int quiet_hours[HOURS];
+   int hours_to_use_traffic_profile[HOURS];
+};
+
+enum workload {HIGH, MEDIUM, LOW};
+enum policy_to_use {
+   TRAFFIC,
+   TIME,
+   WORKLOAD
+};
+
+struct traffic {
+   uint32_t min_packet_thresh;
+   uint32_t avg_max_packet_thresh;
+   uint32_t max_max_packet_thresh;
+};
 
 struct channel_packet {
uint64_t resource_id; /**< core_num, device */
uint32_t unit;/**< scale down/up/min/max */
uint32_t command; /**< Power, IO, etc */
+   char vm_name[32];
+
+   uint64_t vfid[MAX_VFS];
+   int nb_mac_to_monitor;
+   struct traffic traffic_policy;
+   uint8_t vcpu_to_control[MAX_VCPU_PER_VM];
+   uint8_t num_vcpu;
+   struct timer_profile timer_policy;
+   enum workload workload;
+   enum policy_to_use policy_to_use;
+   struct t_boost_status t_boost_status;
 };
 
 
-- 
2.7.4

[dpdk-dev] [PATCH v5 4/9] examples/vm_power_mgr: add scale to medium freq fn

[David Hunt]

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: Rory Sexton 
Signed-off-by: David Hunt 
---
 examples/vm_power_manager/power_manager.c | 15 +++
 examples/vm_power_manager/power_manager.h | 13 +
 2 files changed, 28 insertions(+)

diff --git a/examples/vm_power_manager/power_manager.c 
b/examples/vm_power_manager/power_manager.c
index 80705f9..c021c1d 100644
--- a/examples/vm_power_manager/power_manager.c
+++ b/examples/vm_power_manager/power_manager.c
@@ -286,3 +286,18 @@ power_manager_disable_turbo_core(unsigned int core_num)
POWER_SCALE_CORE(disable_turbo, core_num, ret);
return ret;
 }
+
+int
+power_manager_scale_core_med(unsigned int core_num)
+{
+   int ret = 0;
+
+   if (core_num >= POWER_MGR_MAX_CPUS)
+   return -1;
+   if (!(global_enabled_cpus & (1ULL << core_num)))
+   return -1;
+   rte_spinlock_lock(&global_core_freq_info[core_num].power_sl);
+   ret = rte_power_set_freq(core_num, 5);
+   rte_spinlock_unlock(&global_core_freq_info[core_num].power_sl);
+   return ret;
+}
diff --git a/examples/vm_power_manager/power_manager.h 
b/examples/vm_power_manager/power_manager.h
index b74d09b..b52fb4c 100644
--- a/examples/vm_power_manager/power_manager.h
+++ b/examples/vm_power_manager/power_manager.h
@@ -231,6 +231,19 @@ int power_manager_disable_turbo_core(unsigned int 
core_num);
  */
 uint32_t power_manager_get_current_frequency(unsigned core_num);
 
+/**
+ * Scale to medium frequency for the core specified by core_num.
+ * It is thread-safe.
+ *
+ * @param core_num
+ *  The core number to change frequency
+ *
+ * @return
+ *  - 1 on success.
+ *  - 0 if frequency not changed.
+ *  - Negative on error.
+ */
+int power_manager_scale_core_med(unsigned int core_num);
 
 #ifdef __cplusplus
 }
-- 
2.7.4

[dpdk-dev] [PATCH v5 5/9] examples/vm_power_mgr: add policy to channels

[David Hunt]

From: "Sexton, Rory" 

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: Rory Sexton 
Signed-off-by: David Hunt 
---
 examples/vm_power_manager/Makefile  |  16 ++
 examples/vm_power_manager/channel_monitor.c | 321 +++-
 examples/vm_power_manager/channel_monitor.h |  18 ++
 3 files changed, 348 insertions(+), 7 deletions(-)

diff --git a/examples/vm_power_manager/Makefile 
b/examples/vm_power_manager/Makefile
index 59a9641..9cf20a2 100644
--- a/examples/vm_power_manager/Makefile
+++ b/examples/vm_power_manager/Makefile
@@ -54,6 +54,22 @@ CFLAGS += $(WERROR_FLAGS)
 
 LDLIBS += -lvirt
 
+ifeq ($(CONFIG_RTE_BUILD_SHARED_LIB),y)
+
+ifeq ($(CONFIG_RTE_LIBRTE_IXGBE_PMD),y)
+LDLIBS += -lrte_pmd_ixgbe
+endif
+
+ifeq ($(CONFIG_RTE_LIBRTE_I40E_PMD),y)
+LDLIBS += -lrte_pmd_i40e
+endif
+
+ifeq ($(CONFIG_RTE_LIBRTE_BNXT_PMD),y)
+LDLIBS += -lrte_pmd_bnxt
+endif
+
+endif
+
 # workaround for a gcc bug with noreturn attribute
 # http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12603
 ifeq ($(CONFIG_RTE_TOOLCHAIN_GCC),y)
diff --git a/examples/vm_power_manager/channel_monitor.c 
b/examples/vm_power_manager/channel_monitor.c
index ac40dac..f16358d 100644
--- a/examples/vm_power_manager/channel_monitor.c
+++ b/examples/vm_power_manager/channel_monitor.c
@@ -41,13 +41,17 @@
 #include 
 #include 
 #include 
+#include 
 
 #include 
 #include 
 #include 
 #include 
+#include 
+#include 
+#include 
 
-
+#include 
 #include "channel_monitor.h"
 #include "channel_commands.h"
 #include "channel_manager.h"
@@ -57,10 +61,15 @@
 
 #define MAX_EVENTS 256
 
+uint64_t vsi_pkt_count_prev[384];
+uint64_t rdtsc_prev[384];
 
+double time_period_s = 1;
 static volatile unsigned run_loop = 1;
 static int global_event_fd;
+static unsigned int policy_is_set;
 static struct epoll_event *global_events_list;
+static struct policy policies[MAX_VMS];
 
 void channel_monitor_exit(void)
 {
@@ -68,6 +77,286 @@ void channel_monitor_exit(void)
rte_free(global_events_list);
 }
 
+static void
+core_share(int pNo, int z, int x, int t)
+{
+   if (policies[pNo].core_share[z].pcpu == lvm_info[x].pcpus[t]) {
+   if (strcmp(policies[pNo].pkt.vm_name,
+   lvm_info[x].vm_name) != 0) {
+   policies[pNo].core_share[z].status = 1;
+   power_manager_scale_core_max(
+   policies[pNo].core_share[z].pcpu);
+   }
+   }
+}
+
+static void
+core_share_status(int pNo)
+{
+
+   int noVms, noVcpus, z, x, t;
+
+   get_all_vm(&noVms, &noVcpus);
+
+   /* Reset Core Share Status. */
+   for (z = 0; z < noVcpus; z++)
+   policies[pNo].core_share[z].status = 0;
+
+   /* Foreach vcpu in a policy. */
+   for (z = 0; z < policies[pNo].pkt.num_vcpu; z++) {
+   /* Foreach VM on the platform. */
+   for (x = 0; x < noVms; x++) {
+   /* Foreach vcpu of VMs on platform. */
+   for (t = 0; t < lvm_info[x].num_cpus; t++)
+   core_share(pNo, z, x, t);
+   }
+   }
+}
+
+static void
+get_pcpu_to_control(struct policy *pol)
+{
+
+   /* Convert vcpu to pcpu. */
+   struct vm_info info;
+   int pcpu, count;
+   uint64_t mask_u64b;
+
+   RTE_LOG(INFO, CHANNEL_MONITOR, "Looking for pcpu for %s\n",
+   pol->pkt.vm_name);
+   get_info_vm(pol->pkt.vm_name, &info);
+
+   for (count = 0; count < pol->pkt.num_vcpu; count++) {
+   mask_u64b = info.pcpu_mask[pol->pkt.vcpu_to_control[count]];
+   for (pcpu = 0; mask_u64b; mask_u64b &= ~(1ULL << pcpu++)) {
+   if ((mask_u64b >> pcpu) & 1)
+   pol->core_share[count].pcpu = pcpu;
+   }
+   }
+}
+
+static int
+get_pfid(struct policy *pol)
+{
+
+   int i, x, ret = 0, nb_ports;
+
+   nb_ports = rte_eth_dev_count();
+   for (i = 0; i < pol->pkt.nb_mac_to_monitor; i++) {
+
+   for (x = 0; x < nb_ports; x++) {
+   ret = rte_pmd_i40e_query_vfid_by_mac(x,
+   (struct ether_addr *)&(pol->pkt.vfid[i]));
+   if (ret != -EINVAL) {
+   pol->port[i] = x;
+   break;
+   }
+   }
+   if (ret == -EINVAL || ret == -ENOTSUP || ret == ENODEV) {
+   RTE_LOG(INFO, CHANNEL_MONITOR,
+   "Error with Policy. MAC not found on "
+   "attached ports ");
+   pol->enabled = 0;
+   return ret;
+   }
+   pol->pfid[i] = ret;
+   }
+   return 1;
+}
+
+static int
+update_policy(struct channel_packet *pkt)
+{
+
+   unsigned int updated = 0;
+
+   for (int i = 0; i < MAX_VMS; i++) {
+   if (strcmp(policies[i].p

[dpdk-dev] [PATCH v5 6/9] examples/vm_power_mgr: add port initialisation

[David Hunt]

We need to initialise the port's we're monitoring to be able to see
the throughput.

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: David Hunt 
---
 examples/vm_power_manager/main.c | 220 +++
 1 file changed, 220 insertions(+)

diff --git a/examples/vm_power_manager/main.c b/examples/vm_power_manager/main.c
index c33fcc9..698abca 100644
--- a/examples/vm_power_manager/main.c
+++ b/examples/vm_power_manager/main.c
@@ -49,6 +49,9 @@
 #include 
 #include 
 #include 
+#include 
+#include 
+#include 
 #include 
 
 #include "channel_manager.h"
@@ -56,6 +59,192 @@
 #include "power_manager.h"
 #include "vm_power_cli.h"
 
+#define RX_RING_SIZE 512
+#define TX_RING_SIZE 512
+
+#define NUM_MBUFS 8191
+#define MBUF_CACHE_SIZE 250
+#define BURST_SIZE 32
+
+static uint32_t enabled_port_mask;
+static volatile bool force_quit;
+
+//
+static const struct rte_eth_conf port_conf_default = {
+   .rxmode = { .max_rx_pkt_len = ETHER_MAX_LEN }
+};
+
+static inline int
+port_init(uint8_t port, struct rte_mempool *mbuf_pool)
+{
+   struct rte_eth_conf port_conf = port_conf_default;
+   const uint16_t rx_rings = 1, tx_rings = 1;
+   int retval;
+   uint16_t q;
+
+   if (port >= rte_eth_dev_count())
+   return -1;
+
+   /* Configure the Ethernet device. */
+   retval = rte_eth_dev_configure(port, rx_rings, tx_rings, &port_conf);
+   if (retval != 0)
+   return retval;
+
+   /* Allocate and set up 1 RX queue per Ethernet port. */
+   for (q = 0; q < rx_rings; q++) {
+   retval = rte_eth_rx_queue_setup(port, q, RX_RING_SIZE,
+   rte_eth_dev_socket_id(port), NULL, mbuf_pool);
+   if (retval < 0)
+   return retval;
+   }
+
+   /* Allocate and set up 1 TX queue per Ethernet port. */
+   for (q = 0; q < tx_rings; q++) {
+   retval = rte_eth_tx_queue_setup(port, q, TX_RING_SIZE,
+   rte_eth_dev_socket_id(port), NULL);
+   if (retval < 0)
+   return retval;
+   }
+
+   /* Start the Ethernet port. */
+   retval = rte_eth_dev_start(port);
+   if (retval < 0)
+   return retval;
+
+   /* Display the port MAC address. */
+   struct ether_addr addr;
+   rte_eth_macaddr_get(port, &addr);
+   printf("Port %u MAC: %02" PRIx8 " %02" PRIx8 " %02" PRIx8
+  " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 "\n",
+   (unsigned int)port,
+   addr.addr_bytes[0], addr.addr_bytes[1],
+   addr.addr_bytes[2], addr.addr_bytes[3],
+   addr.addr_bytes[4], addr.addr_bytes[5]);
+
+   /* Enable RX in promiscuous mode for the Ethernet device. */
+   rte_eth_promiscuous_enable(port);
+
+
+   return 0;
+}
+
+static int
+parse_portmask(const char *portmask)
+{
+   char *end = NULL;
+   unsigned long pm;
+
+   /* parse hexadecimal string */
+   pm = strtoul(portmask, &end, 16);
+   if ((portmask[0] == '\0') || (end == NULL) || (*end != '\0'))
+   return -1;
+
+   if (pm == 0)
+   return -1;
+
+   return pm;
+}
+/* Parse the argument given in the command line of the application */
+static int
+parse_args(int argc, char **argv)
+{
+   int opt, ret;
+   char **argvopt;
+   int option_index;
+   char *prgname = argv[0];
+   static struct option lgopts[] = {
+   { "mac-updating", no_argument, 0, 1},
+   { "no-mac-updating", no_argument, 0, 0},
+   {NULL, 0, 0, 0}
+   };
+   argvopt = argv;
+
+   while ((opt = getopt_long(argc, argvopt, "p:q:T:",
+ lgopts, &option_index)) != EOF) {
+
+   switch (opt) {
+   /* portmask */
+   case 'p':
+   enabled_port_mask = parse_portmask(optarg);
+   if (enabled_port_mask == 0) {
+   printf("invalid portmask\n");
+   return -1;
+   }
+   break;
+   /* long options */
+   case 0:
+   break;
+
+   default:
+   return -1;
+   }
+   }
+
+   if (optind >= 0)
+   argv[optind-1] = prgname;
+
+   ret = optind-1;
+   optind = 0; /* reset getopt lib */
+   return ret;
+}
+
+static void
+check_all_ports_link_status(uint8_t port_num, uint32_t port_mask)
+{
+#define CHECK_INTERVAL 100 /* 100ms */
+#define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */
+   uint8_t portid, count, all_ports_up, print_flag = 0;
+   struct rte_eth_link link;
+
+   printf("\nChecking link status");
+   fflush(stdout);
+   for (count = 0; count <= MAX_CHECK_TIME; count++) {
+   if (force_q

[dpdk-dev] [PATCH v5 8/9] examples/guest_cli: add send policy to host

[David Hunt]

From: "Sexton, Rory" 

Here we're adding an example of setting up a policy, and allowing the
vm_cli_guest app to send it to the host using the cli command
"send_policy now"

Signed-off-by: Nemanja Marjanovic 
Signed-off-by: Rory Sexton 
Signed-off-by: David Hunt 
---
 .../guest_cli/vm_power_cli_guest.c | 97 ++
 .../guest_cli/vm_power_cli_guest.h |  6 --
 2 files changed, 97 insertions(+), 6 deletions(-)

diff --git a/examples/vm_power_manager/guest_cli/vm_power_cli_guest.c 
b/examples/vm_power_manager/guest_cli/vm_power_cli_guest.c
index 4e982bd..dc9efc2 100644
--- a/examples/vm_power_manager/guest_cli/vm_power_cli_guest.c
+++ b/examples/vm_power_manager/guest_cli/vm_power_cli_guest.c
@@ -45,8 +45,10 @@
 #include 
 #include 
 #include 
+#include 
 
 #include 
+#include 
 
 #include "vm_power_cli_guest.h"
 
@@ -139,8 +141,103 @@ cmdline_parse_inst_t cmd_set_cpu_freq_set = {
},
 };
 
+struct cmd_send_policy_result {
+   cmdline_fixed_string_t send_policy;
+   cmdline_fixed_string_t cmd;
+};
+
+union PFID {
+   struct ether_addr addr;
+   uint64_t pfid;
+};
+
+static inline int
+send_policy(void)
+{
+   struct channel_packet pkt;
+   int ret;
+
+   union PFID pfid;
+   /* Use port MAC address as the vfid */
+   rte_eth_macaddr_get(0, &pfid.addr);
+   printf("Port %u MAC: %02" PRIx8 ":%02" PRIx8 ":%02" PRIx8 ":"
+   "%02" PRIx8 ":%02" PRIx8 ":%02" PRIx8 "\n",
+   1,
+   pfid.addr.addr_bytes[0], pfid.addr.addr_bytes[1],
+   pfid.addr.addr_bytes[2], pfid.addr.addr_bytes[3],
+   pfid.addr.addr_bytes[4], pfid.addr.addr_bytes[5]);
+   pkt.vfid[0] = pfid.pfid;
+
+   pkt.nb_mac_to_monitor = 1;
+   pkt.t_boost_status.tbEnabled = false;
+
+   pkt.vcpu_to_control[0] = 0;
+   pkt.vcpu_to_control[1] = 1;
+   pkt.num_vcpu = 2;
+   /* Dummy Population. */
+   pkt.traffic_policy.min_packet_thresh = 96000;
+   pkt.traffic_policy.avg_max_packet_thresh = 180;
+   pkt.traffic_policy.max_max_packet_thresh = 200;
+
+   pkt.timer_policy.busy_hours[0] = 3;
+   pkt.timer_policy.busy_hours[1] = 4;
+   pkt.timer_policy.busy_hours[2] = 5;
+   pkt.timer_policy.quiet_hours[0] = 11;
+   pkt.timer_policy.quiet_hours[1] = 12;
+   pkt.timer_policy.quiet_hours[2] = 13;
+
+   pkt.timer_policy.hours_to_use_traffic_profile[0] = 8;
+   pkt.timer_policy.hours_to_use_traffic_profile[1] = 10;
+
+   pkt.workload = LOW;
+   pkt.policy_to_use = TIME;
+   pkt.command = PKT_POLICY;
+   strcpy(pkt.vm_name, "ubuntu2");
+   ret = rte_power_guest_channel_send_msg(&pkt, 1);
+   if (ret == 0)
+   return 1;
+   RTE_LOG(DEBUG, POWER, "Error sending message: %s\n",
+   ret > 0 ? strerror(ret) : "channel not connected");
+   return -1;
+}
+
+static void
+cmd_send_policy_parsed(void *parsed_result, struct cmdline *cl,
+  __attribute__((unused)) void *data)
+{
+   int ret = -1;
+   struct cmd_send_policy_result *res = parsed_result;
+
+   if (!strcmp(res->cmd, "now")) {
+   printf("Sending Policy down now!\n");
+   ret = send_policy();
+   }
+   if (ret != 1)
+   cmdline_printf(cl, "Error sending message: %s\n",
+   strerror(ret));
+}
+
+cmdline_parse_token_string_t cmd_send_policy =
+   TOKEN_STRING_INITIALIZER(struct cmd_send_policy_result,
+   send_policy, "send_policy");
+cmdline_parse_token_string_t cmd_send_policy_cmd_cmd =
+   TOKEN_STRING_INITIALIZER(struct cmd_send_policy_result,
+   cmd, "now");
+
+cmdline_parse_inst_t cmd_send_policy_set = {
+   .f = cmd_send_policy_parsed,
+   .data = NULL,
+   .help_str = "send_policy now",
+   .tokens = {
+   (void *)&cmd_send_policy,
+   (void *)&cmd_send_policy_cmd_cmd,
+   NULL,
+   },
+};
+
 cmdline_parse_ctx_t main_ctx[] = {
(cmdline_parse_inst_t *)&cmd_quit,
+   (cmdline_parse_inst_t *)&cmd_send_policy_set,
(cmdline_parse_inst_t *)&cmd_set_cpu_freq_set,
NULL,
 };
diff --git a/examples/vm_power_manager/guest_cli/vm_power_cli_guest.h 
b/examples/vm_power_manager/guest_cli/vm_power_cli_guest.h
index 0c4bdd5..277eab3 100644
--- a/examples/vm_power_manager/guest_cli/vm_power_cli_guest.h
+++ b/examples/vm_power_manager/guest_cli/vm_power_cli_guest.h
@@ -40,12 +40,6 @@ extern "C" {
 
 #include "channel_commands.h"
 
-int guest_channel_host_connect(unsigned lcore_id);
-
-int guest_channel_send_msg(struct channel_packet *pkt, unsigned lcore_id);
-
-void guest_channel_host_disconnect(unsigned lcore_id);
-
 void run_cli(__attribute__((unused)) void *arg);
 
 #ifdef __cplusplus
-- 
2.7.4

[dpdk-dev] [PATCH v5 7/9] power: add send channel msg function to map file

[David Hunt]

Adding new wrapper function to existing private (but unused 'till now)
function with an rte_power_ prefix.

The plan is to clean up all the header files in the next release so
that only the intended public functions are in the map file and only
the relevant headers have the rte_ prefix so that only they are
included in the documentation.

Signed-off-by: David Hunt 
---
 lib/librte_power/guest_channel.c   |  7 +++
 lib/librte_power/guest_channel.h   | 15 +++
 lib/librte_power/rte_power_version.map |  1 +
 3 files changed, 23 insertions(+)

diff --git a/lib/librte_power/guest_channel.c b/lib/librte_power/guest_channel.c
index 85c92fa..fa5de0f 100644
--- a/lib/librte_power/guest_channel.c
+++ b/lib/librte_power/guest_channel.c
@@ -148,6 +148,13 @@ guest_channel_send_msg(struct channel_packet *pkt, 
unsigned lcore_id)
return 0;
 }
 
+int rte_power_guest_channel_send_msg(struct channel_packet *pkt,
+   unsigned int lcore_id)
+{
+   return guest_channel_send_msg(pkt, lcore_id);
+}
+
+
 void
 guest_channel_host_disconnect(unsigned lcore_id)
 {
diff --git a/lib/librte_power/guest_channel.h b/lib/librte_power/guest_channel.h
index 9e18af5..741339c 100644
--- a/lib/librte_power/guest_channel.h
+++ b/lib/librte_power/guest_channel.h
@@ -81,6 +81,21 @@ void guest_channel_host_disconnect(unsigned lcore_id);
  */
 int guest_channel_send_msg(struct channel_packet *pkt, unsigned lcore_id);
 
+/**
+ * Send a message contained in pkt over the Virtio-Serial to the host endpoint.
+ *
+ * @param pkt
+ *  Pointer to a populated struct channel_packet
+ *
+ * @param lcore_id
+ *  lcore_id.
+ *
+ * @return
+ *  - 0 on success.
+ *  - Negative on error.
+ */
+int rte_power_guest_channel_send_msg(struct channel_packet *pkt,
+   unsigned int lcore_id);
 
 #ifdef __cplusplus
 }
diff --git a/lib/librte_power/rte_power_version.map 
b/lib/librte_power/rte_power_version.map
index 9ae0627..96dc42e 100644
--- a/lib/librte_power/rte_power_version.map
+++ b/lib/librte_power/rte_power_version.map
@@ -20,6 +20,7 @@ DPDK_2.0 {
 DPDK_17.11 {
global:
 
+   rte_power_guest_channel_send_msg;
rte_power_freq_disable_turbo;
rte_power_freq_enable_turbo;
rte_power_turbo_status;
-- 
2.7.4

[dpdk-dev] [PATCH v5 9/9] examples/vm_power_mgr: set MAC address of VF

[David Hunt]

We need to set vf mac from the host, so that they will be in sync on the
guest and the host. Otherwise, we'll have a random mac on the guest, and
a 00:00:00:00:00:00 mac on the host.

Signed-off-by: David Hunt 
---
 examples/vm_power_manager/main.c | 43 +++-
 1 file changed, 42 insertions(+), 1 deletion(-)

diff --git a/examples/vm_power_manager/main.c b/examples/vm_power_manager/main.c
index 698abca..5147789 100644
--- a/examples/vm_power_manager/main.c
+++ b/examples/vm_power_manager/main.c
@@ -58,6 +58,9 @@
 #include "channel_monitor.h"
 #include "power_manager.h"
 #include "vm_power_cli.h"
+#include 
+#include 
+#include 
 
 #define RX_RING_SIZE 512
 #define TX_RING_SIZE 512
@@ -222,7 +225,7 @@ check_all_ports_link_status(uint8_t port_num, uint32_t 
port_mask)
(uint8_t)portid);
continue;
}
-  /* clear all_ports_up flag if any link down */
+   /* clear all_ports_up flag if any link down */
if (link.link_status == ETH_LINK_DOWN) {
all_ports_up = 0;
break;
@@ -301,11 +304,49 @@ main(int argc, char **argv)
 
/* Initialize ports. */
for (portid = 0; portid < nb_ports; portid++) {
+   struct ether_addr eth;
+   int w, j;
+   int ret = -ENOTSUP;
+
if ((enabled_port_mask & (1 << portid)) == 0)
continue;
+
+   eth.addr_bytes[0] = 0xe0;
+   eth.addr_bytes[1] = 0xe0;
+   eth.addr_bytes[2] = 0xe0;
+   eth.addr_bytes[3] = 0xe0;
+   eth.addr_bytes[4] = portid + 0xf0;
+
if (port_init(portid, mbuf_pool) != 0)
rte_exit(EXIT_FAILURE, "Cannot init port %"PRIu8 "\n",
portid);
+
+   for (w = 0; w < MAX_VFS; w++) {
+   eth.addr_bytes[5] = w + 0xf0;
+
+   if (ret == -ENOTSUP)
+   ret = rte_pmd_ixgbe_set_vf_mac_addr(portid,
+   w, ð);
+   if (ret == -ENOTSUP)
+   ret = rte_pmd_i40e_set_vf_mac_addr(portid,
+   w, ð);
+   if (ret == -ENOTSUP)
+   ret = rte_pmd_bnxt_set_vf_mac_addr(portid,
+   w, ð);
+
+   switch (ret) {
+   case 0:
+   printf("Port %d VF %d MAC: ",
+   portid, w);
+   for (j = 0; j < 6; j++) {
+   printf("%02x", eth.addr_bytes[j]);
+   if (j < 5)
+   printf(":");
+   }
+   printf("\n");
+   break;
+   }
+   }
}
 
lcore_id = rte_get_next_lcore(-1, 1, 0);
-- 
2.7.4

Re: [dpdk-dev] [PATCH v4 1/9] net/i40e: add API to convert VF MAC to VF id

[santosh]

On Wednesday 04 October 2017 02:45 PM, David Hunt wrote:
> From: "Sexton, Rory" 
>
> Need a way to convert a vf id to a pf id on the host so as to query the pf
> for relevant statistics which are used for the frequency changes in the
> vm_power_manager app. Used when profiles are passed down from the guest
> to the host, allowing the host to map the vfs to pfs.
>
> Signed-off-by: Nemanja Marjanovic 
> Signed-off-by: Rory Sexton 
> Signed-off-by: David Hunt 
> ---
>  drivers/net/i40e/rte_pmd_i40e.c   | 31 
> +++
>  drivers/net/i40e/rte_pmd_i40e.h   | 13 +
>  drivers/net/i40e/rte_pmd_i40e_version.map |  7 +++
>  3 files changed, 51 insertions(+)
>
> diff --git a/drivers/net/i40e/rte_pmd_i40e.c b/drivers/net/i40e/rte_pmd_i40e.c
> index f12b7f4..21efb2f 100644
> --- a/drivers/net/i40e/rte_pmd_i40e.c
> +++ b/drivers/net/i40e/rte_pmd_i40e.c
> @@ -2115,3 +2115,34 @@ int rte_pmd_i40e_ptype_mapping_replace(uint8_t port,
>  
>   return 0;
>  }
> +
> +uint64_t
> +rte_pmd_i40e_query_vfid_by_mac(uint8_t port, uint64_t vf_mac)
> +{
> + struct rte_eth_dev *dev;
> + struct ether_addr *vf_mac_addr = (struct ether_addr *)&vf_mac;
> + struct ether_addr *mac;
> + struct i40e_pf *pf;
> + int vf_id;
> + struct i40e_pf_vf *vf;
> + uint16_t vf_num;
> +
> + RTE_ETH_VALID_PORTID_OR_ERR_RET(port, -ENODEV);
> + dev = &rte_eth_devices[port];
> +
> + if (!is_i40e_supported(dev))
> + return -ENOTSUP;
> +
> + pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
> + vf_num = pf->vf_num;
> +
> + for (vf_id = 0; vf_id < vf_num; vf_id++) {
> + vf = &pf->vfs[vf_id];
> + mac = &vf->mac_addr;
> +
> + if (is_same_ether_addr(mac, vf_mac_addr))
> + return vf_id;
> + }
> +
> + return -EINVAL;
> +}
> diff --git a/drivers/net/i40e/rte_pmd_i40e.h b/drivers/net/i40e/rte_pmd_i40e.h
> index 356fa89..a7ae0f0 100644
> --- a/drivers/net/i40e/rte_pmd_i40e.h
> +++ b/drivers/net/i40e/rte_pmd_i40e.h
> @@ -637,4 +637,17 @@ int rte_pmd_i40e_ptype_mapping_replace(uint8_t port,
>  uint8_t mask,
>  uint32_t pkt_type);
>  
> +/**
> + * On the PF, find VF index based on VF MAC address
> + *
> + * @param port
> + *pointer to port identifier of the device
> + * @param vf_mac
> + *the mac address of the vf to determine index of
> + * @return
> + *-(-22 EINVAL) the vf mac does not exist on this port
> + *-(!-22) the index of vfid in pf->vfs
> + */
> +uint64_t rte_pmd_i40e_query_vfid_by_mac(uint8_t port, uint64_t vf_mac);
> +

On @return:
- The index of vfid If successful.
- -EINVAL: vf mac address does not exits for this port
- -ENOTSUP: i40e not supported for this port.

and return type should s/uint64_t/int since error case -ve value.
Thanks.

Re: [dpdk-dev] [PATCH] eal/bsdapp: add FreeBSD module compilation to meson build

[Van Haaren, Harry]

> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Bruce Richardson
> Sent: Thursday, September 21, 2017 3:55 PM
> To: dev@dpdk.org
> Cc: Richardson, Bruce 
> Subject: [dpdk-dev] [PATCH] eal/bsdapp: add FreeBSD module compilation to
> meson build
> 
> Support compiling the FreeBSD kernel modules using meson and ninja.
> 
> Signed-off-by: Bruce Richardson 

Tested on BSD box, both existing GMake system, and Meson / Ninja working as 
expected here.
With inline comment below fixed;

Tested-by: Harry van Haaren 


> ---
>  config/rte_config.h |  5 
>  lib/librte_eal/bsdapp/BSDmakefile.meson | 42
> +
>  lib/librte_eal/bsdapp/contigmem/meson.build | 32 ++
>  lib/librte_eal/bsdapp/meson.build   | 23 
>  lib/librte_eal/bsdapp/nic_uio/meson.build   | 32 ++
>  5 files changed, 134 insertions(+)
>  create mode 100644 lib/librte_eal/bsdapp/BSDmakefile.meson
>  create mode 100644 lib/librte_eal/bsdapp/contigmem/meson.build
>  create mode 100644 lib/librte_eal/bsdapp/nic_uio/meson.build
> 
> diff --git a/config/rte_config.h b/config/rte_config.h
> index 3e26d5118..26656137c 100644
> --- a/config/rte_config.h
> +++ b/config/rte_config.h
> @@ -58,6 +58,11 @@
>  #define RTE_BACKTRACE 1
>  #define RTE_EAL_VFIO 1
> 
> +/* bsd module defines */
> +#define RTE_CONTIGMEM_MAX_NUM_BUFS 64
> +#define RTE_CONTIGMEM_DEFAULT_NUM_BUFS 1
> +#define RTE_CONTIGMEM_DEFAULT_BUF_SIZE 512*1024*1024

ERROR:COMPLEX_MACRO: Macros with complex values should be enclosed in 
parentheses
#39: FILE: config/rte_config.h:64:
+#define RTE_CONTIGMEM_DEFAULT_BUF_SIZE 512*1024*1024

Re: [dpdk-dev] [PATCH v4 2/9] lib/librte_power: add extra msg type for policies

[santosh]

Hi David,


On Wednesday 04 October 2017 02:45 PM, David Hunt wrote:
> Signed-off-by: Nemanja Marjanovic 
> Signed-off-by: Rory Sexton 
> Signed-off-by: David Hunt 
> ---

my 2cent:
General comment on implementation approach:
IMO, we should avoid PMD details in common lib area.
example: file channel_commons.h has ifdef clutter referencing
i40e pmds all over.

Perhaps we should introduce opaque handle example void * or introduce pmd
specific callback/handle which points to PMD specific metadata in power library.

Example:
struct channel_packet {
  void *pmd_specific_metadata;
}

Or someway via callback (I'm not sure at the moment)
so that we could hide PMD details in common area.

Thanks.

>  lib/librte_power/channel_commands.h | 52 
> +
>  1 file changed, 52 insertions(+)
>
> diff --git a/lib/librte_power/channel_commands.h 
> b/lib/librte_power/channel_commands.h
> index 484085b..1599706 100644
> --- a/lib/librte_power/channel_commands.h
> +++ b/lib/librte_power/channel_commands.h
> @@ -46,6 +46,7 @@ extern "C" {
>  /* Valid Commands */
>  #define CPU_POWER   1
>  #define CPU_POWER_CONNECT   2
> +#define PKT_POLICY  3
>  
>  /* CPU Power Command Scaling */
>  #define CPU_POWER_SCALE_UP  1
> @@ -54,11 +55,62 @@ extern "C" {
>  #define CPU_POWER_SCALE_MIN 4
>  #define CPU_POWER_ENABLE_TURBO  5
>  #define CPU_POWER_DISABLE_TURBO 6
> +#define HOURS 24
> +
> +#ifdef RTE_LIBRTE_I40E_PMD
> +#define MAX_VFS 10
> +#endif
> +
> +#define MAX_VCPU_PER_VM 8
> +
> +typedef enum {false, true} bool;
> +
> +struct t_boost_status {
> + bool tbEnabled;
> +};
> +
> +struct timer_profile {
> + int busy_hours[HOURS];
> + int quiet_hours[HOURS];
> +#ifdef RTE_LIBRTE_I40E_PMD
> + int hours_to_use_traffic_profile[HOURS];
> +#endif
> +};
> +
> +enum workload {HIGH, MEDIUM, LOW};
> +enum policy_to_use {
> +#ifdef RTE_LIBRTE_I40E_PMD
> + TRAFFIC,
> +#endif
> + TIME,
> + WORKLOAD
> +};
> +
> +#ifdef RTE_LIBRTE_I40E_PMD
> +struct traffic {
> + uint32_t min_packet_thresh;
> + uint32_t avg_max_packet_thresh;
> + uint32_t max_max_packet_thresh;
> +};
> +#endif
>  
>  struct channel_packet {
>   uint64_t resource_id; /**< core_num, device */
>   uint32_t unit;/**< scale down/up/min/max */
>   uint32_t command; /**< Power, IO, etc */
> + char vm_name[32];
> +
> +#ifdef RTE_LIBRTE_I40E_PMD
> + uint64_t vfid[MAX_VFS];
> + int nb_mac_to_monitor;
> + struct traffic traffic_policy;
> +#endif
> + uint8_t vcpu_to_control[MAX_VCPU_PER_VM];
> + uint8_t num_vcpu;
> + struct timer_profile timer_policy;
> + enum workload workload;
> + enum policy_to_use policy_to_use;
> + struct t_boost_status t_boost_status;
>  };
>  
>  

Re: [dpdk-dev] [PATCH v5 1/9] net/i40e: add API to convert VF MAC to VF id

[santosh]

Hi David,


On Wednesday 04 October 2017 08:55 PM, David Hunt wrote:
> From: "Sexton, Rory" 
>
> Need a way to convert a vf id to a pf id on the host so as to query the pf
> for relevant statistics which are used for the frequency changes in the
> vm_power_manager app. Used when profiles are passed down from the guest
> to the host, allowing the host to map the vfs to pfs.
>
> Signed-off-by: Nemanja Marjanovic 
> Signed-off-by: Rory Sexton 
> Signed-off-by: David Hunt 
> ---

I see that you just now sent out v5;)
But I guess v4 comment on this patch [1]
is still applicable (imo).
Thanks.

[1] http://dpdk.org/dev/patchwork/patch/29577/

Re: [dpdk-dev] [PATCH v5 2/9] lib/librte_power: add extra msg type for policies

[santosh]

Hi David,


On Wednesday 04 October 2017 08:55 PM, David Hunt wrote:
> Signed-off-by: Nemanja Marjanovic 
> Signed-off-by: Rory Sexton 
> Signed-off-by: David Hunt 
> ---

Glad that ifdef clutter removed.
Few nits though..

>  lib/librte_power/channel_commands.h | 42 
> +
>  1 file changed, 42 insertions(+)
>
> diff --git a/lib/librte_power/channel_commands.h 
> b/lib/librte_power/channel_commands.h
> index 484085b..020d9fe 100644
> --- a/lib/librte_power/channel_commands.h
> +++ b/lib/librte_power/channel_commands.h
> @@ -46,6 +46,7 @@ extern "C" {
>  /* Valid Commands */
>  #define CPU_POWER   1
>  #define CPU_POWER_CONNECT   2
> +#define PKT_POLICY  3
>  
>  /* CPU Power Command Scaling */
>  #define CPU_POWER_SCALE_UP  1
> @@ -54,11 +55,52 @@ extern "C" {
>  #define CPU_POWER_SCALE_MIN 4
>  #define CPU_POWER_ENABLE_TURBO  5
>  #define CPU_POWER_DISABLE_TURBO 6
> +#define HOURS 24
> +
> +#define MAX_VFS 10
> +
> +#define MAX_VCPU_PER_VM 8
> +
> +typedef enum {false, true} bool;
> +

do we really need typedef for bool; can't we simply
use bool data-type?

> +struct t_boost_status {
> + bool tbEnabled;
> +};
> +
> +struct timer_profile {
> + int busy_hours[HOURS];
> + int quiet_hours[HOURS];
> + int hours_to_use_traffic_profile[HOURS];
> +};
> +
> +enum workload {HIGH, MEDIUM, LOW};
> +enum policy_to_use {
> + TRAFFIC,
> + TIME,
> + WORKLOAD
> +};
> +
> +struct traffic {
> + uint32_t min_packet_thresh;
> + uint32_t avg_max_packet_thresh;
> + uint32_t max_max_packet_thresh;
> +};
>  
>  struct channel_packet {
>   uint64_t resource_id; /**< core_num, device */
>   uint32_t unit;/**< scale down/up/min/max */
>   uint32_t command; /**< Power, IO, etc */
> + char vm_name[32];
> +

How about const char * Or in case not possible then #define RTE_xx 32 Or
use existing RTE_ for same purpose or some micro local to power lib.

> + uint64_t vfid[MAX_VFS];
> + int nb_mac_to_monitor;
> + struct traffic traffic_policy;
> + uint8_t vcpu_to_control[MAX_VCPU_PER_VM];
> + uint8_t num_vcpu;
> + struct timer_profile timer_policy;
> + enum workload workload;
> + enum policy_to_use policy_to_use;
> + struct t_boost_status t_boost_status;
>  };
>  
>  

Re: [dpdk-dev] [PATCH v5 3/9] examples/vm_power_mgr: add vcpu to pcpu mapping

[santosh]

Hi David,


On Wednesday 04 October 2017 08:55 PM, David Hunt wrote:
> Signed-off-by: Nemanja Marjanovic 
> Signed-off-by: Rory Sexton 
> Signed-off-by: David Hunt 
> ---
>  examples/vm_power_manager/channel_manager.c | 62 
> +
>  examples/vm_power_manager/channel_manager.h | 25 
>  2 files changed, 87 insertions(+)
>
> diff --git a/examples/vm_power_manager/channel_manager.c 
> b/examples/vm_power_manager/channel_manager.c
> index e068ae2..03fa626 100644
> --- a/examples/vm_power_manager/channel_manager.c
> +++ b/examples/vm_power_manager/channel_manager.c
> @@ -574,6 +574,68 @@ set_channel_status(const char *vm_name, unsigned 
> *channel_list,
>   return num_channels_changed;
>  }
>  
> +void
> +get_all_vm(int *num_vm, int *num_cpu)
> +{

nits:
s/*num_cpu/*num_vcpu

> +
> + virNodeInfo node_info;
> + virDomainPtr *domptr;
> + uint64_t mask;
> + int i, ii, numVcpus[MAX_VCPUS], cpu, n_vcpus;
> + unsigned int jj;
> + const char *vm_name;
> + unsigned int flags = VIR_CONNECT_LIST_DOMAINS_RUNNING |
> + VIR_CONNECT_LIST_DOMAINS_PERSISTENT;
> + unsigned int flag = VIR_DOMAIN_VCPU_CONFIG;
> +

nits:
Perhaps add more clear name example:
s/flags/conn_flags
s/flag/domain_flags

> +
> + memset(global_cpumaps, 0, CHANNEL_CMDS_MAX_CPUS*global_maplen);
> + if (virNodeGetInfo(global_vir_conn_ptr, &node_info))
> + RTE_LOG(ERR, CHANNEL_MANAGER, "Unable to retrieve node Info\n");
> +

Should return from here.. since node info not retrieve ops errored out.

> + /* Returns number of pcpus */
> + global_n_host_cpus = (unsigned int)node_info.cpus;
> +
> + /* Returns number of active domains */
> + *num_vm = virConnectListAllDomains(global_vir_conn_ptr, &domptr, flags);
> + if (*num_vm <= 0)
> + RTE_LOG(ERR, CHANNEL_MANAGER, "No Active Domains Running\n");
> +

ditto..

> + for (i = 0; i < *num_vm; i++) {
> +
> + /* Get Domain Names */
> + vm_name = virDomainGetName(domptr[i]);
> + lvm_info[i].vm_name = vm_name;
> +
> + /* Get Number of Vcpus */
> + numVcpus[i] = virDomainGetVcpusFlags(domptr[i], flag);
> +
> + /* Get Number of VCpus & VcpuPinInfo */
> + n_vcpus = virDomainGetVcpuPinInfo(domptr[i],
> + numVcpus[i], global_cpumaps,
> + global_maplen, flag);
> +
> + if ((int)n_vcpus > 0) {
> + *num_cpu = n_vcpus;
> + lvm_info[i].num_cpus = n_vcpus;
> + }
> +
> + /* Save pcpu in use by libvirt VMs */
> + for (ii = 0; ii < n_vcpus; ii++) {
> + mask = 0;
> + for (jj = 0; jj < global_n_host_cpus; jj++) {
> + if (VIR_CPU_USABLE(global_cpumaps,
> + global_maplen, ii, jj) > 0) {
> + mask |= 1ULL << jj;
> + }
> + }
> + ITERATIVE_BITMASK_CHECK_64(mask, cpu) {
> + lvm_info[i].pcpus[ii] = cpu;
> + }
> + }
> + }
> +}
> +
>  int
>  get_info_vm(const char *vm_name, struct vm_info *info)
>  {
> diff --git a/examples/vm_power_manager/channel_manager.h 
> b/examples/vm_power_manager/channel_manager.h
> index 47c3b9c..788c1e6 100644
> --- a/examples/vm_power_manager/channel_manager.h
> +++ b/examples/vm_power_manager/channel_manager.h
> @@ -66,6 +66,17 @@ struct sockaddr_un _sockaddr_un;
>  #define UNIX_PATH_MAX sizeof(_sockaddr_un.sun_path)
>  #endif
>  
> +#define MAX_VMS 4
> +#define MAX_VCPUS 20
> +
> +
> +struct libvirt_vm_info {
> + const char *vm_name;
> + unsigned int pcpus[MAX_VCPUS];
> + uint8_t num_cpus;
> +};
> +
> +struct libvirt_vm_info lvm_info[MAX_VMS];
>  /* Communication Channel Status */
>  enum channel_status { CHANNEL_MGR_CHANNEL_DISCONNECTED = 0,
>   CHANNEL_MGR_CHANNEL_CONNECTED,
> @@ -319,6 +330,20 @@ int set_channel_status(const char *vm_name, unsigned 
> *channel_list,
>   */
>  int get_info_vm(const char *vm_name, struct vm_info *info);
>  
> +/**
> + * Populates a table with all domains running and their physical cpu.
> + * All information is gathered through libvirt api.
> + *
> + * @param noVms
> + *  modified to store number of active VMs
> + *
> + * @param noVcpus
> +modified to store number of vcpus active
> + *
> + * @return
> + *   void
> + */
> +void get_all_vm(int *noVms, int *noVcpus);

nits: perhaps,
void
get_all_vm(int *num_vm, int *num_vcpu)
Thanks. 

>  #ifdef __cplusplus
>  }
>  #endif

Re: [dpdk-dev] [PATCH v5 4/9] examples/vm_power_mgr: add scale to medium freq fn

[santosh]

Hi David,


On Wednesday 04 October 2017 08:55 PM, David Hunt wrote:
> Signed-off-by: Nemanja Marjanovic 
> Signed-off-by: Rory Sexton 
> Signed-off-by: David Hunt 
> ---
>  examples/vm_power_manager/power_manager.c | 15 +++
>  examples/vm_power_manager/power_manager.h | 13 +
>  2 files changed, 28 insertions(+)
>
> diff --git a/examples/vm_power_manager/power_manager.c 
> b/examples/vm_power_manager/power_manager.c
> index 80705f9..c021c1d 100644
> --- a/examples/vm_power_manager/power_manager.c
> +++ b/examples/vm_power_manager/power_manager.c
> @@ -286,3 +286,18 @@ power_manager_disable_turbo_core(unsigned int core_num)
>   POWER_SCALE_CORE(disable_turbo, core_num, ret);
>   return ret;
>  }
> +
> +int
> +power_manager_scale_core_med(unsigned int core_num)
> +{
> + int ret = 0;
> +
> + if (core_num >= POWER_MGR_MAX_CPUS)
> + return -1;
> + if (!(global_enabled_cpus & (1ULL << core_num)))
> + return -1;
> + rte_spinlock_lock(&global_core_freq_info[core_num].power_sl);
> + ret = rte_power_set_freq(core_num, 5);

nits:
what is 5? also should be enum or macro.

Thanks.

Re: [dpdk-dev] [PATCH v6 0/4] ethdev new offloads API

[Ananyev, Konstantin]

> -Original Message-
> From: Shahaf Shuler [mailto:shah...@mellanox.com]
> Sent: Wednesday, October 4, 2017 9:18 AM
> To: Ananyev, Konstantin ; tho...@monjalon.net; 
> arybche...@solarflare.com;
> jerin.ja...@caviumnetworks.com; Yigit, Ferruh 
> Cc: dev@dpdk.org
> Subject: [PATCH v6 0/4] ethdev new offloads API
> 
> Tx offloads configuration is per queue. Tx offloads are enabled by default,
> and can be disabled using ETH_TXQ_FLAGS_NO* flags.
> This behaviour is not consistent with the Rx side where the Rx offloads
> configuration is per port. Rx offloads are disabled by default and enabled
> according to bit field in rte_eth_rxmode structure.
> 
> Moreover, considering more Tx and Rx offloads will be added
> over time, the cost of managing them all inside the PMD will be tremendous,
> as the PMD will need to check the matching for the entire offload set
> for each mbuf it handles.
> In addition, on the current approach each Rx offload added breaks the
> ABI compatibility as it requires to add entries to existing bit-fields.
> 
> The series address above issues by defining a new offloads API.
> In the new API, offloads are divided into per-port and per-queue offloads,
> with a corresponding capability for each.
> The offloads are disabled by default. Each offload can be enabled or
> disabled using the existing DEV_TX_OFFLOADS_* or DEV_RX_OFFLOADS_* flags.
> Such API will enable to easily add or remove offloads, without breaking the
> ABI compatibility.
> 
> In order to provide a smooth transition between the APIs the following actions
> were taken:
> *  The old offloads API is kept for the meanwhile.
> *  Helper function which copy from old to new API were added to ethdev,
>enabling the PMD to support only one of the APIs.
> *  Helper function which copy from new to old API were also added,
>to enable application to use the new API with PMD which still supports
>the old one.
> 
> Per discussion made on the RFC of this series [1], the integration plan which 
> was
> decided is to do the transition in two phases:
> * ethdev API will move on 17.11.
> * Apps and examples will move on 18.02.
> 
> This to enable PMD maintainers sufficient time to adopt the new API.
> 
> [1]
> http://dpdk.org/ml/archives/dev/2017-August/072643.html
> 
> on v6:
>  - Move mbuf fast free Tx offload to a seperate patch.
> 
> on v5:
>  - Fix documentation.
>  - Fix comments on port offloads configuration.
> 
> on v4:
>  - Added another patch for documentation.
>  - Fixed ETH_TXQ_FLAGS_IGNORE flag override.
>  - Clarify the description of DEV_TX_OFFLOAD_MBUF_FAST_FREE offload.
> 
> on v3:
>  - Introduce the DEV_TX_OFFLOAD_MBUF_FAST_FREE to act as an equivalent
>for the no refcnt and single mempool flags.
>  - Fix features documentation.
>  - Fix comment style.
> 
> on v2:
>  - Taking new approach of dividing offloads into per-queue and per-port one.
>  - Postpone the Tx/Rx public struct renaming to 18.02
>  - Squash the helper functions into the Rx/Tx offloads intro patches.
> 
> Shahaf Shuler (4):
>   ethdev: introduce Rx queue offloads API
>   ethdev: introduce Tx queue offloads API
>   ethdev: add mbuf fast free Tx offload
>   doc: add details on ethdev offloads API
> 
>  doc/guides/nics/features.rst|  66 +---
>  doc/guides/nics/features/default.ini|   1 +
>  doc/guides/prog_guide/poll_mode_drv.rst |  20 +++
>  lib/librte_ether/rte_ethdev.c   | 223 +--
>  lib/librte_ether/rte_ethdev.h   |  89 ++-
>  5 files changed, 359 insertions(+), 40 deletions(-)
> 
> Series-reviewed-by: Andrew Rybchenko 
> 
> --
Acked-by: Konstantin Ananyev 

> 2.12.0

Re: [dpdk-dev] [PATCH v6 3/6] gso: add VxLAN GSO support

[Kavanagh, Mark B]

>-Original Message-
>From: Ananyev, Konstantin
>Sent: Wednesday, October 4, 2017 3:12 PM
>To: Kavanagh, Mark B ; dev@dpdk.org
>Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Yigit, Ferruh ; tho...@monjalon.net
>Subject: RE: [PATCH v6 3/6] gso: add VxLAN GSO support
>
>
>
>> -Original Message-
>> From: Kavanagh, Mark B
>> Sent: Monday, October 2, 2017 5:46 PM
>> To: dev@dpdk.org
>> Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Ananyev, Konstantin ; Yigit,
>> Ferruh ; tho...@monjalon.net; Kavanagh, Mark B
>
>> Subject: [PATCH v6 3/6] gso: add VxLAN GSO support
>>
>> This patch adds a framework that allows GSO on tunneled packets.
>> Furthermore, it leverages that framework to provide GSO support for
>> VxLAN-encapsulated packets.
>>
>> Supported VxLAN packets must have an outer IPv4 header (prepended by an
>> optional VLAN tag), and contain an inner TCP/IPv4 packet (with an optional
>> inner VLAN tag).
>>
>> VxLAN GSO doesn't check if input packets have correct checksums and
>> doesn't update checksums for output packets. Additionally, it doesn't
>> process IP fragmented packets.
>>
>> As with TCP/IPv4 GSO, VxLAN GSO uses a two-segment MBUF to organize each
>> output packet, which mandates support for multi-segment mbufs in the TX
>> functions of the NIC driver. Also, if a packet is GSOed, VxLAN GSO
>> reduces its MBUF refcnt by 1. As a result, when all of its GSO'd segments
>> are freed, the packet is freed automatically.
>>
>> Signed-off-by: Mark Kavanagh 
>> Signed-off-by: Jiayu Hu 
>> ---
>>  doc/guides/rel_notes/release_17_11.rst |   3 +
>>  lib/librte_gso/Makefile|   1 +
>>  lib/librte_gso/gso_common.h|  25 +++
>>  lib/librte_gso/gso_tunnel_tcp4.c   | 123
>+
>>  lib/librte_gso/gso_tunnel_tcp4.h   |  75 
>>  lib/librte_gso/rte_gso.c   |  13 +++-
>>  6 files changed, 237 insertions(+), 3 deletions(-)
>>  create mode 100644 lib/librte_gso/gso_tunnel_tcp4.c
>>  create mode 100644 lib/librte_gso/gso_tunnel_tcp4.h
>>
>> diff --git a/doc/guides/rel_notes/release_17_11.rst
>b/doc/guides/rel_notes/release_17_11.rst
>> index c414f73..25b8a78 100644
>> --- a/doc/guides/rel_notes/release_17_11.rst
>> +++ b/doc/guides/rel_notes/release_17_11.rst
>> @@ -48,6 +48,9 @@ New Features
>>ones (e.g. MTU is 1500B). Supported packet types are:
>>
>>* TCP/IPv4 packets, which may include a single VLAN tag.
>> +  * VxLAN packets, which must have an outer IPv4 header (prepended by
>> +an optional VLAN tag), and contain an inner TCP/IPv4 packet (with
>> +an optional VLAN tag).
>>
>>The GSO library doesn't check if the input packets have correct
>>checksums, and doesn't update checksums for output packets.
>> diff --git a/lib/librte_gso/Makefile b/lib/librte_gso/Makefile
>> index 2be64d1..e6d41df 100644
>> --- a/lib/librte_gso/Makefile
>> +++ b/lib/librte_gso/Makefile
>> @@ -44,6 +44,7 @@ LIBABIVER := 1
>>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += rte_gso.c
>>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_common.c
>>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_tcp4.c
>> +SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_tunnel_tcp4.c
>>
>>  # install this header file
>>  SYMLINK-$(CONFIG_RTE_LIBRTE_GSO)-include += rte_gso.h
>> diff --git a/lib/librte_gso/gso_common.h b/lib/librte_gso/gso_common.h
>> index 8d9b94e..c051295 100644
>> --- a/lib/librte_gso/gso_common.h
>> +++ b/lib/librte_gso/gso_common.h
>> @@ -39,6 +39,7 @@
>>  #include 
>>  #include 
>>  #include 
>> +#include 
>>
>>  #define IS_FRAGMENTED(frag_off) (((frag_off) & IPV4_HDR_OFFSET_MASK) != 0 \
>>  || ((frag_off) & IPV4_HDR_MF_FLAG) == IPV4_HDR_MF_FLAG)
>> @@ -49,6 +50,30 @@
>>  #define IS_IPV4_TCP(flag) (((flag) & (PKT_TX_TCP_SEG | PKT_TX_IPV4)) == \
>>  (PKT_TX_TCP_SEG | PKT_TX_IPV4))
>>
>> +#define IS_IPV4_VXLAN_TCP4(flag) (((flag) & (PKT_TX_TCP_SEG | PKT_TX_IPV4 |
>\
>> +PKT_TX_OUTER_IPV4 | PKT_TX_TUNNEL_VXLAN)) == \
>> +(PKT_TX_TCP_SEG | PKT_TX_IPV4 | PKT_TX_OUTER_IPV4 | \
>> + PKT_TX_TUNNEL_VXLAN))
>> +
>> +/**
>> + * Internal function which updates the UDP header of a packet, following
>> + * segmentation. This is required to update the header's datagram length
>field.
>> + *
>> + * @param pkt
>> + *  The packet containing the UDP header.
>> + * @param udp_offset
>> + *  The offset of the UDP header from the start of the packet.
>> + */
>> +static inline void
>> +update_udp_header(struct rte_mbuf *pkt, uint16_t udp_offset)
>> +{
>> +struct udp_hdr *udp_hdr;
>> +
>> +udp_hdr = (struct udp_hdr *)(rte_pktmbuf_mtod(pkt, char *) +
>> +udp_offset);
>> +udp_hdr->dgram_len = rte_cpu_to_be_16(pkt->pkt_len - udp_offset);
>> +}
>> +
>>  /**
>>   * Internal function which updates the TCP header of a packet, following
>>   * segmentation. This is required to update the header's 'sent' sequence
>> diff --git a/lib/librte_gso/gso_tunnel_tcp4.c
>b/lib/librte_gso/gso_tunnel_tcp4.c
>> new fi

Re: [dpdk-dev] [PATCH v1 6/6] bbdev: auxiliary changes to DPDK framework

[Burakov, Anatoly]

On 30-Sep-17 2:37 AM, Amr Mokhtar wrote:

diff --git a/mk/rte.app.mk b/mk/rte.app.mk
index c25fdd9..7286ad9 100644
--- a/mk/rte.app.mk
+++ b/mk/rte.app.mk
@@ -93,6 +93,7 @@ _LDLIBS-$(CONFIG_RTE_LIBRTE_NET)+= -lrte_net
  _LDLIBS-$(CONFIG_RTE_LIBRTE_ETHER)  += -lrte_ethdev
  _LDLIBS-$(CONFIG_RTE_LIBRTE_CRYPTODEV)  += -lrte_cryptodev
  _LDLIBS-$(CONFIG_RTE_LIBRTE_EVENTDEV)   += -lrte_eventdev
+_LDLIBS-$(CONFIG_RTE_LIBRTE_BBDEV)  += -lrte_bbdev
  _LDLIBS-$(CONFIG_RTE_LIBRTE_MEMPOOL)+= -lrte_mempool
  _LDLIBS-$(CONFIG_RTE_DRIVER_MEMPOOL_RING)   += -lrte_mempool_ring
  _LDLIBS-$(CONFIG_RTE_LIBRTE_RING)   += -lrte_ring
@@ -146,6 +147,18 @@ endif # $(CONFIG_RTE_LIBRTE_VHOST)
  _LDLIBS-$(CONFIG_RTE_LIBRTE_VMXNET3_PMD)+= -lrte_pmd_vmxnet3_uio
  _LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_XENVIRT)+= -lrte_pmd_xenvirt -lxenstore
  
+ifeq ($(CONFIG_RTE_LIBRTE_BBDEV),y)

+_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_BBDEV_NULL) += -lrte_pmd_bbdev_null
+
+# TURBO SOFTWARE PMD is dependent on the BBLIB library
+_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_BBDEV_TURBO_SW) += -lrte_pmd_bbdev_turbo_sw
+_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_BBDEV_TURBO_SW) += -L$(BBLIB_PATH)/lib_common 
-lcommon
+_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_BBDEV_TURBO_SW) += -L$(BBLIB_PATH)/lib_crc 
-lcrc
+_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_BBDEV_TURBO_SW) += -L$(BBLIB_PATH)/lib_turbo 
-lturbo
+_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_BBDEV_TURBO_SW) += 
-L$(BBLIB_PATH)/lib_rate_matching -lrate_matching
+_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_BBDEV_TURBO_SW) += -lirc -limf -lstdc++ -lipps
+endif # CONFIG_RTE_LIBRTE_BBDEV
+
  ifeq ($(CONFIG_RTE_LIBRTE_CRYPTODEV),y)
  _LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_AESNI_MB)+= -lrte_pmd_aesni_mb
  _LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_AESNI_MB)+= 
-L$(AESNI_MULTI_BUFFER_LIB_PATH) -lIPSec_MB



Changes to makefiles should probably go in with the patches that 
introduce the libraries. As it stands, if something breaks the build 
inside one of the PMD's or BBDEV itself (which is the case right now), 
we won't know about it until this patch.


--
Thanks,
Anatoly

Re: [dpdk-dev] [PATCH v6 3/6] gso: add VxLAN GSO support

[Ananyev, Konstantin]

> -Original Message-
> From: Kavanagh, Mark B
> Sent: Wednesday, October 4, 2017 5:14 PM
> To: Ananyev, Konstantin ; dev@dpdk.org
> Cc: Hu, Jiayu ; Tan, Jianfeng ; 
> Yigit, Ferruh ; tho...@monjalon.net
> Subject: RE: [PATCH v6 3/6] gso: add VxLAN GSO support
> 
> >-Original Message-
> >From: Ananyev, Konstantin
> >Sent: Wednesday, October 4, 2017 3:12 PM
> >To: Kavanagh, Mark B ; dev@dpdk.org
> >Cc: Hu, Jiayu ; Tan, Jianfeng ;
> >Yigit, Ferruh ; tho...@monjalon.net
> >Subject: RE: [PATCH v6 3/6] gso: add VxLAN GSO support
> >
> >
> >
> >> -Original Message-
> >> From: Kavanagh, Mark B
> >> Sent: Monday, October 2, 2017 5:46 PM
> >> To: dev@dpdk.org
> >> Cc: Hu, Jiayu ; Tan, Jianfeng ;
> >Ananyev, Konstantin ; Yigit,
> >> Ferruh ; tho...@monjalon.net; Kavanagh, Mark B
> >
> >> Subject: [PATCH v6 3/6] gso: add VxLAN GSO support
> >>
> >> This patch adds a framework that allows GSO on tunneled packets.
> >> Furthermore, it leverages that framework to provide GSO support for
> >> VxLAN-encapsulated packets.
> >>
> >> Supported VxLAN packets must have an outer IPv4 header (prepended by an
> >> optional VLAN tag), and contain an inner TCP/IPv4 packet (with an optional
> >> inner VLAN tag).
> >>
> >> VxLAN GSO doesn't check if input packets have correct checksums and
> >> doesn't update checksums for output packets. Additionally, it doesn't
> >> process IP fragmented packets.
> >>
> >> As with TCP/IPv4 GSO, VxLAN GSO uses a two-segment MBUF to organize each
> >> output packet, which mandates support for multi-segment mbufs in the TX
> >> functions of the NIC driver. Also, if a packet is GSOed, VxLAN GSO
> >> reduces its MBUF refcnt by 1. As a result, when all of its GSO'd segments
> >> are freed, the packet is freed automatically.
> >>
> >> Signed-off-by: Mark Kavanagh 
> >> Signed-off-by: Jiayu Hu 
> >> ---
> >>  doc/guides/rel_notes/release_17_11.rst |   3 +
> >>  lib/librte_gso/Makefile|   1 +
> >>  lib/librte_gso/gso_common.h|  25 +++
> >>  lib/librte_gso/gso_tunnel_tcp4.c   | 123
> >+
> >>  lib/librte_gso/gso_tunnel_tcp4.h   |  75 
> >>  lib/librte_gso/rte_gso.c   |  13 +++-
> >>  6 files changed, 237 insertions(+), 3 deletions(-)
> >>  create mode 100644 lib/librte_gso/gso_tunnel_tcp4.c
> >>  create mode 100644 lib/librte_gso/gso_tunnel_tcp4.h
> >>
> >> diff --git a/doc/guides/rel_notes/release_17_11.rst
> >b/doc/guides/rel_notes/release_17_11.rst
> >> index c414f73..25b8a78 100644
> >> --- a/doc/guides/rel_notes/release_17_11.rst
> >> +++ b/doc/guides/rel_notes/release_17_11.rst
> >> @@ -48,6 +48,9 @@ New Features
> >>ones (e.g. MTU is 1500B). Supported packet types are:
> >>
> >>* TCP/IPv4 packets, which may include a single VLAN tag.
> >> +  * VxLAN packets, which must have an outer IPv4 header (prepended by
> >> +an optional VLAN tag), and contain an inner TCP/IPv4 packet (with
> >> +an optional VLAN tag).
> >>
> >>The GSO library doesn't check if the input packets have correct
> >>checksums, and doesn't update checksums for output packets.
> >> diff --git a/lib/librte_gso/Makefile b/lib/librte_gso/Makefile
> >> index 2be64d1..e6d41df 100644
> >> --- a/lib/librte_gso/Makefile
> >> +++ b/lib/librte_gso/Makefile
> >> @@ -44,6 +44,7 @@ LIBABIVER := 1
> >>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += rte_gso.c
> >>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_common.c
> >>  SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_tcp4.c
> >> +SRCS-$(CONFIG_RTE_LIBRTE_GSO) += gso_tunnel_tcp4.c
> >>
> >>  # install this header file
> >>  SYMLINK-$(CONFIG_RTE_LIBRTE_GSO)-include += rte_gso.h
> >> diff --git a/lib/librte_gso/gso_common.h b/lib/librte_gso/gso_common.h
> >> index 8d9b94e..c051295 100644
> >> --- a/lib/librte_gso/gso_common.h
> >> +++ b/lib/librte_gso/gso_common.h
> >> @@ -39,6 +39,7 @@
> >>  #include 
> >>  #include 
> >>  #include 
> >> +#include 
> >>
> >>  #define IS_FRAGMENTED(frag_off) (((frag_off) & IPV4_HDR_OFFSET_MASK) != 0 
> >> \
> >>|| ((frag_off) & IPV4_HDR_MF_FLAG) == IPV4_HDR_MF_FLAG)
> >> @@ -49,6 +50,30 @@
> >>  #define IS_IPV4_TCP(flag) (((flag) & (PKT_TX_TCP_SEG | PKT_TX_IPV4)) == \
> >>(PKT_TX_TCP_SEG | PKT_TX_IPV4))
> >>
> >> +#define IS_IPV4_VXLAN_TCP4(flag) (((flag) & (PKT_TX_TCP_SEG | PKT_TX_IPV4 
> >> |
> >\
> >> +  PKT_TX_OUTER_IPV4 | PKT_TX_TUNNEL_VXLAN)) == \
> >> +  (PKT_TX_TCP_SEG | PKT_TX_IPV4 | PKT_TX_OUTER_IPV4 | \
> >> +   PKT_TX_TUNNEL_VXLAN))
> >> +
> >> +/**
> >> + * Internal function which updates the UDP header of a packet, following
> >> + * segmentation. This is required to update the header's datagram length
> >field.
> >> + *
> >> + * @param pkt
> >> + *  The packet containing the UDP header.
> >> + * @param udp_offset
> >> + *  The offset of the UDP header from the start of the packet.
> >> + */
> >> +static inline void
> >> +update_udp_header(struct rte_mbuf *pkt, uint16_t ud

Re: [dpdk-dev] [PATCH v5 7/9] power: add send channel msg function to map file

[santosh]

Hi David,


On Wednesday 04 October 2017 08:55 PM, David Hunt wrote:
> Adding new wrapper function to existing private (but unused 'till now)
> function with an rte_power_ prefix.
>
> The plan is to clean up all the header files in the next release so
> that only the intended public functions are in the map file and only
> the relevant headers have the rte_ prefix so that only they are
> included in the documentation.
>
> Signed-off-by: David Hunt 
> ---

lgtm:
Reviewed-by: Santosh Shukla 
Thanks.

Re: [dpdk-dev] [PATCH v6 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO

[Ananyev, Konstantin]

> -Original Message-
> From: Kavanagh, Mark B
> Sent: Wednesday, October 4, 2017 5:23 PM
> To: Ananyev, Konstantin ; dev@dpdk.org
> Cc: Hu, Jiayu ; Tan, Jianfeng ; 
> Yigit, Ferruh ; tho...@monjalon.net
> Subject: RE: [PATCH v6 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO
> 
> 
> 
> >-Original Message-
> >From: Ananyev, Konstantin
> >Sent: Wednesday, October 4, 2017 4:09 PM
> >To: Kavanagh, Mark B ; dev@dpdk.org
> >Cc: Hu, Jiayu ; Tan, Jianfeng ;
> >Yigit, Ferruh ; tho...@monjalon.net
> >Subject: RE: [PATCH v6 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO
> >
> >
> >
> >> -Original Message-
> >> From: Kavanagh, Mark B
> >> Sent: Monday, October 2, 2017 5:46 PM
> >> To: dev@dpdk.org
> >> Cc: Hu, Jiayu ; Tan, Jianfeng ;
> >Ananyev, Konstantin ; Yigit,
> >> Ferruh ; tho...@monjalon.net; Kavanagh, Mark B
> >
> >> Subject: [PATCH v6 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO
> >>
> >> From: Jiayu Hu 
> >>
> >> This patch adds GSO support to the csum forwarding engine. Oversized
> >> packets transmitted over a GSO-enabled port will undergo segmentation
> >> (with the exception of packet-types unsupported by the GSO library).
> >> GSO support is disabled by default.
> >>
> >> GSO support may be toggled on a per-port basis, using the command:
> >>
> >> "set port  gso on|off"
> >>
> >> The maximum packet length (including the packet header and payload) for
> >> GSO segments may be set with the command:
> >>
> >> "set gso segsz "
> >>
> >> Show GSO configuration for a given port with the command:
> >>
> >>"show port  gso"
> >>
> >> Signed-off-by: Jiayu Hu 
> >> Signed-off-by: Mark Kavanagh 
> >> ---
> >>  app/test-pmd/cmdline.c  | 178
> >
> >>  app/test-pmd/config.c   |  24 
> >>  app/test-pmd/csumonly.c |  69 ++-
> >>  app/test-pmd/testpmd.c  |  13 ++
> >>  app/test-pmd/testpmd.h  |  10 ++
> >>  doc/guides/testpmd_app_ug/testpmd_funcs.rst |  46 +++
> >>  6 files changed, 335 insertions(+), 5 deletions(-)
> >>
> >> diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
> >> index ccdf239..05b0ce8 100644
> >> --- a/app/test-pmd/cmdline.c
> >> +++ b/app/test-pmd/cmdline.c
> >> @@ -431,6 +431,17 @@ static void cmd_help_long_parsed(void *parsed_result,
> >>"Set max flow number and max packet number per-flow"
> >>" for GRO.\n\n"
> >>
> >> +  "set port (port_id) gso (on|off)"
> >> +  "Enable or disable Generic Segmentation Offload in"
> >> +  " csum forwarding engine.\n\n"
> >> +
> >> +  "set gso segsz (length)\n"
> >> +  "Set max packet length for output GSO segments,"
> >> +  " including packet header and payload.\n\n"
> >
> >Probably a  good future improvement would be to allow user to specify 
> >gso_type
> >too.
> 
> Would you like to see that change implemented in time for the 17.11 release?

I think that's too late for such change in 17.11.
My thought was about 18.02 here.
Konstantin

Re: [dpdk-dev] [PATCH v6 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO

[Kavanagh, Mark B]

>-Original Message-
>From: Ananyev, Konstantin
>Sent: Wednesday, October 4, 2017 4:09 PM
>To: Kavanagh, Mark B ; dev@dpdk.org
>Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Yigit, Ferruh ; tho...@monjalon.net
>Subject: RE: [PATCH v6 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO
>
>
>
>> -Original Message-
>> From: Kavanagh, Mark B
>> Sent: Monday, October 2, 2017 5:46 PM
>> To: dev@dpdk.org
>> Cc: Hu, Jiayu ; Tan, Jianfeng ;
>Ananyev, Konstantin ; Yigit,
>> Ferruh ; tho...@monjalon.net; Kavanagh, Mark B
>
>> Subject: [PATCH v6 5/6] app/testpmd: enable TCP/IPv4, VxLAN and GRE GSO
>>
>> From: Jiayu Hu 
>>
>> This patch adds GSO support to the csum forwarding engine. Oversized
>> packets transmitted over a GSO-enabled port will undergo segmentation
>> (with the exception of packet-types unsupported by the GSO library).
>> GSO support is disabled by default.
>>
>> GSO support may be toggled on a per-port basis, using the command:
>>
>> "set port  gso on|off"
>>
>> The maximum packet length (including the packet header and payload) for
>> GSO segments may be set with the command:
>>
>> "set gso segsz "
>>
>> Show GSO configuration for a given port with the command:
>>
>>  "show port  gso"
>>
>> Signed-off-by: Jiayu Hu 
>> Signed-off-by: Mark Kavanagh 
>> ---
>>  app/test-pmd/cmdline.c  | 178
>
>>  app/test-pmd/config.c   |  24 
>>  app/test-pmd/csumonly.c |  69 ++-
>>  app/test-pmd/testpmd.c  |  13 ++
>>  app/test-pmd/testpmd.h  |  10 ++
>>  doc/guides/testpmd_app_ug/testpmd_funcs.rst |  46 +++
>>  6 files changed, 335 insertions(+), 5 deletions(-)
>>
>> diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
>> index ccdf239..05b0ce8 100644
>> --- a/app/test-pmd/cmdline.c
>> +++ b/app/test-pmd/cmdline.c
>> @@ -431,6 +431,17 @@ static void cmd_help_long_parsed(void *parsed_result,
>>  "Set max flow number and max packet number per-flow"
>>  " for GRO.\n\n"
>>
>> +"set port (port_id) gso (on|off)"
>> +"Enable or disable Generic Segmentation Offload in"
>> +" csum forwarding engine.\n\n"
>> +
>> +"set gso segsz (length)\n"
>> +"Set max packet length for output GSO segments,"
>> +" including packet header and payload.\n\n"
>
>Probably a  good future improvement would be to allow user to specify gso_type
>too.

Would you like to see that change implemented in time for the 17.11 release?

>
>> +
>> +"show port (port_id) gso\n"
>> +"Show GSO configuration.\n\n"
>> +
>>  "set fwd (%s)\n"
>>  "Set packet forwarding mode.\n\n"
>>
>> @@ -3967,6 +3978,170 @@ struct cmd_gro_set_result {
>>  },
>>  };
>>
>> +/* *** ENABLE/DISABLE GSO *** */
>> +struct cmd_gso_enable_result {
>> +cmdline_fixed_string_t cmd_set;
>> +cmdline_fixed_string_t cmd_port;
>> +cmdline_fixed_string_t cmd_keyword;
>> +cmdline_fixed_string_t cmd_mode;
>> +uint8_t cmd_pid;
>> +};
>> +
>> +static void
>> +cmd_gso_enable_parsed(void *parsed_result,
>> +__attribute__((unused)) struct cmdline *cl,
>> +__attribute__((unused)) void *data)
>> +{
>> +struct cmd_gso_enable_result *res;
>> +
>> +res = parsed_result;
>> +if (!strcmp(res->cmd_keyword, "gso"))
>> +setup_gso(res->cmd_mode, res->cmd_pid);
>> +}
>> +
>> +cmdline_parse_token_string_t cmd_gso_enable_set =
>> +TOKEN_STRING_INITIALIZER(struct cmd_gso_enable_result,
>> +cmd_set, "set");
>> +cmdline_parse_token_string_t cmd_gso_enable_port =
>> +TOKEN_STRING_INITIALIZER(struct cmd_gso_enable_result,
>> +cmd_port, "port");
>> +cmdline_parse_token_string_t cmd_gso_enable_keyword =
>> +TOKEN_STRING_INITIALIZER(struct cmd_gso_enable_result,
>> +cmd_keyword, "gso");
>> +cmdline_parse_token_string_t cmd_gso_enable_mode =
>> +TOKEN_STRING_INITIALIZER(struct cmd_gso_enable_result,
>> +cmd_mode, "on#off");
>> +cmdline_parse_token_num_t cmd_gso_enable_pid =
>> +TOKEN_NUM_INITIALIZER(struct cmd_gso_enable_result,
>> +cmd_pid, UINT8);
>> +
>> +cmdline_parse_inst_t cmd_gso_enable = {
>> +.f = cmd_gso_enable_parsed,
>> +.data = NULL,
>> +.help_str = "set port  gso on|off",
>> +.tokens = {
>> +(void *)&cmd_gso_enable_set,
>> +(void *)&cmd_gso_enable_port,
>> +(void *)&cmd_gso_enable_pid,
>> +(void *)&cmd_gso_enable_keyword,
>> +(void *)&cmd_gso_enable_mode,
>> +NULL,
>> +},
>> +};
>> +
>> +/* *** SET MAX PACKET LENGTH FOR GSO SEGMENTS *** */
>> +struct cmd_gso_size_result {
>> +cmdline_fixed_string_t cmd_set;

Re: [dpdk-dev] [PATCH] eal/bsdapp: add FreeBSD module compilation to meson build

[Bruce Richardson]

On Wed, Oct 04, 2017 at 04:34:11PM +0100, Van Haaren, Harry wrote:
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Bruce Richardson
> > Sent: Thursday, September 21, 2017 3:55 PM
> > To: dev@dpdk.org
> > Cc: Richardson, Bruce 
> > Subject: [dpdk-dev] [PATCH] eal/bsdapp: add FreeBSD module compilation to
> > meson build
> > 
> > Support compiling the FreeBSD kernel modules using meson and ninja.
> > 
> > Signed-off-by: Bruce Richardson 
> 
> Tested on BSD box, both existing GMake system, and Meson / Ninja working as 
> expected here.
> With inline comment below fixed;
> 
> Tested-by: Harry van Haaren 
> 
Applied to dpdk-next-build