Re: [dpdk-dev] [PATCH 1/2] eventdev: implement ABI change

[Kinsella, Ray]

On 28/09/2020 06:53, Jerin Jacob wrote:
> On Thu, Sep 24, 2020 at 3:25 PM Sunil Kumar Kori  wrote:
>>
>>> -Original Message-
>>> From: dev  On Behalf Of Timothy McDaniel
>>> Sent: Friday, September 11, 2020 10:28 PM
>>> Cc: Jerin Jacob Kollanukkaran ;
>>> mattias.ronnb...@ericsson.com; liang.j...@intel.com;
>>> peter.mccar...@intel.com; nipun.gu...@nxp.com; Pavan Nikhilesh
>>> Bhagavatula ; dev@dpdk.org;
>>> erik.g.carri...@intel.com; gage.e...@intel.com; harry.van.haa...@intel.com;
>>> hemant.agra...@nxp.com; bruce.richard...@intel.com
>>> Subject: [dpdk-dev] [PATCH 1/2] eventdev: implement ABI change
> 
> Please change subject to "eventdev: express DLB PMD constraints" or
> something similar.
> 
>>>
>>> This commit implements the eventdev ABI changes required by the DLB PMD.
>>>
>>> The DLB hardware does not conform exactly to the eventdev interface.
>>> 1) It has a limit on the number of queues that may be linked to a port.
>>> 2) Some ports are further restricted to a maximum of 1 linked queue.
>>> 3) It does not (currently) have the ability to carry the flow_id as part of 
>>> the
>>> event (QE) payload.
>>>
>>> Due to the above, we would like to propose the following enhancements.
>>>
>>> 1) Add new fields to the rte_event_dev_info struct. These fields allow the
>>> device to advertise its capabilities so that applications can take the
>>> appropriate actions based on those capabilities.
>>>
>>> 2) Add a new field to the rte_event_dev_config struct. This field allows the
>>> application to specify how many of its ports are limited to a single link, 
>>> or will
>>> be used in single link mode.
>>>
>>> 3) Replace the dedicated implicit_release_disabled field with a bit field of
>>> explicit port capabilities. The implicit_release_disable functionality is 
>>> assigned
>>> to one bit, and a port-is-single-link-only attribute is assigned to 
>>> another, with
>>> the remaining bits available for future assignment.
>>>
>>> Signed-off-by: Timothy McDaniel 
>>> ---
>>> drivers/event/dpaa/dpaa_eventdev.c |  3 +-
>>> drivers/event/dpaa2/dpaa2_eventdev.c   |  5 +-
>>> drivers/event/dsw/dsw_evdev.c  |  3 +-
>>> drivers/event/octeontx/ssovf_evdev.c   |  5 +-
>>> drivers/event/octeontx2/otx2_evdev.c   |  3 +-
>>> drivers/event/opdl/opdl_evdev.c|  3 +-
>>> drivers/event/skeleton/skeleton_eventdev.c |  5 +-
>>> drivers/event/sw/sw_evdev.c|  8 ++--
>>> drivers/event/sw/sw_evdev_selftest.c   |  6 +--
>>> lib/librte_eventdev/rte_event_eth_tx_adapter.c |  2 +-
>>> lib/librte_eventdev/rte_eventdev.c | 66 
>>> +++---
>>> lib/librte_eventdev/rte_eventdev.h | 51 
>>> lib/librte_eventdev/rte_eventdev_pmd_pci.h |  1 -
>>> lib/librte_eventdev/rte_eventdev_trace.h   |  7 +--
>>> lib/librte_eventdev/rte_eventdev_version.map   |  4 +-
> 
> Missed to remove the announced depreciation notice.
> 
> 
>>> 15 files changed, 134 insertions(+), 38 deletions(-)
>>>
>>> /**
>>>  * @internal
>>>  * Wrapper for use by pci drivers as a .remove function to detach a event 
>>> diff
>>> --git a/lib/librte_eventdev/rte_eventdev_trace.h
>>> b/lib/librte_eventdev/rte_eventdev_trace.h
>>> index 4de6341..5ec43d8 100644
>>> --- a/lib/librte_eventdev/rte_eventdev_trace.h
>>> +++ b/lib/librte_eventdev/rte_eventdev_trace.h
>>> @@ -34,6 +34,7 @@ RTE_TRACE_POINT(
>>>   rte_trace_point_emit_u32(dev_conf-
 nb_event_port_dequeue_depth);
>>>   rte_trace_point_emit_u32(dev_conf-
 nb_event_port_enqueue_depth);
>>>   rte_trace_point_emit_u32(dev_conf->event_dev_cfg);
>>> +  rte_trace_point_emit_u8(dev_conf-
 nb_single_link_event_port_queues);
>>>   rte_trace_point_emit_int(rc);
>>> )
>>>
>> I tried running testpmd with trace support with this patch and got following 
>> error:
>>
>> EAL: __rte_trace_point_emit_field():442 CTF field is too long
>> EAL: __rte_trace_point_register():468 missing rte_trace_emit_header() in 
>> register fn
>>
>> It is because of memory (384 bytes) for CTF gets exhausted during 
>> eventdev_configure.
>> Did you also observe the same issue ? If yes, Then please send a separate 
>> patch to increase
>> Maximum buffer size TRACE_CTF_FIELD_SIZE. Recommended size is 448 bytes.
> 
> Please check this and send the next version for this and common code
> change[1]. I would like to merge-common code change first and then the
> DLB drivers.
> 
> [1]
> http://patches.dpdk.org/patch/77466/
> 
> 
>>
>>> @@ -59,7 +60,7 @@ RTE_TRACE_POINT(
>>>   rte_trace_point_emit_i32(port_conf->new_event_threshold);
>>>   rte_trace_point_emit_u16(port_conf->dequeue_depth);
>>>   rte_trace_point_emit_u16(port_conf->enqueue_depth);
>>> -  rte_trace_point_emit_u8(port_conf->disable_implicit_release);
>>> +  rte_trace_point_emit_u32(port_conf->event_port_cfg);
>>>   rte_trace_point_emit_int(rc);
>>> )
>>>
>>> @@ -165,7 +166,7 @@ RT

Re: [dpdk-dev] [RFC PATCH v3 1/3] ethdev: support device reset and recovery events

[Thomas Monjalon]

Hi,

Please use --cc-cmd devtools/get-maintainer.sh so all relevant
maintainers are Cc'ed. Adding Andrew.

> From: Kalesh AP 
> 
> Adding support for device reset and recovery events in the
> rte_eth_event framework. FW error and FW reset conditions would be
> managed internally by PMD without needing application intervention.
> In such cases, PMD would need reset/recovery events to notify application
> that PMD is undergoing a reset.

We already have this event:

RTE_ETH_EVENT_INTR_RESET,
/**< reset interrupt event, sent to VF on PF reset */

I don't know why "INTR" is in the name of this event,
and I think it does not need to be restricted to VF.
The application does not need to know whether the reset
is caused by the PF, the FW or the HW.
That's why I think you could share the same event.

> +   RTE_ETH_EVENT_RESET,/**< port resetting from an error */
> +   RTE_ETH_EVENT_RECOVERED, /**< port recovered from an error */

You ignored my previous comments:
"
What the application is supposed to do when receiving such event?
How the application knows that flow rules were resetted?
Is there any other configuration resetted?
These informations must be explicit in the doxygen comments.
"

Re: [dpdk-dev] [PATCH v5] bus/pci: netuio interface for windows

[Tal Shnaiderman]

> Subject: Re: [PATCH v5] bus/pci: netuio interface for windows
> 
> Hi Tal,
> 
> The issue I raised is slightly different: How do we distinguish between
> different GUID_DEVCLASS_NET class devices - those that can support DPDK
> and those that can't?
> 
> For instance, say we have 3 NICs in the system - MLX5 NIC, Intel i40e NIC and
> Intel ice NIC - all loaded with standard ethernet (NDIS) drivers.
> 
> Now if I load and bind the netuio driver to the Intel ice NIC, I have one
> GUID_DEVCLASS_NETUIO class device and two GUID_DEVCLASS_NET class
> devices.
> 
> If I run the pci scan now, it will detect all 3 devices and add them to the 
> list.
> However, the Intel i40e NIC which is correctly detected as a
> GUID_DEVCLASS_NET class and is bound to the standard NDIS driver cannot
> support DPDK and should be excluded from the list.
> 
> I think the suggestion at the community meeting was to define a DEV
> interface which can be queried for confirming DPDK support. In such a case,
> the MLX5 driver should support this interface, where as our standard i40e
> (NDIS) driver will not. This check can be used to exclude this device from the
> list.
> 
> ranjit m.

Thanks for the explanation Ranjit,

BTW, what will happen if a user does tries to use a NETUIO NIC recognized as 
NET?
It will have nulls for the needed mem_resource fields so will it fails or crash?

[dpdk-dev] [PATCH] crypto/mvsam: remove crypto end enumerators

[michaelsh]

From: Michael Shamis 

Remove enumerators RTE_CRYPTO_CIPHER_LIST_END,
RTE_CRYPTO_AUTH_LIST_END, RTE_CRYPTO_AEAD_LIST_END to prevent
some problems that may arise when adding new crypto algorithms.

Signed-off-by: Michael Shamis 
---
 drivers/crypto/mvsam/rte_mrvl_pmd.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/crypto/mvsam/rte_mrvl_pmd.c 
b/drivers/crypto/mvsam/rte_mrvl_pmd.c
index aaa40dced..bec51c9ff 100644
--- a/drivers/crypto/mvsam/rte_mrvl_pmd.c
+++ b/drivers/crypto/mvsam/rte_mrvl_pmd.c
@@ -67,7 +67,7 @@ __rte_aligned(32);
  * Map of supported cipher algorithms.
  */
 static const
-struct cipher_params_mapping cipher_map[RTE_CRYPTO_CIPHER_LIST_END] = {
+struct cipher_params_mapping cipher_map[] = {
[RTE_CRYPTO_CIPHER_NULL] = {
.supported = ALGO_SUPPORTED,
.cipher_alg = SAM_CIPHER_NONE },
@@ -107,7 +107,7 @@ struct cipher_params_mapping 
cipher_map[RTE_CRYPTO_CIPHER_LIST_END] = {
  * Map of supported auth algorithms.
  */
 static const
-struct auth_params_mapping auth_map[RTE_CRYPTO_AUTH_LIST_END] = {
+struct auth_params_mapping auth_map[] = {
[RTE_CRYPTO_AUTH_NULL] = {
.supported = ALGO_SUPPORTED,
.auth_alg = SAM_AUTH_NONE },
@@ -156,7 +156,7 @@ struct auth_params_mapping 
auth_map[RTE_CRYPTO_AUTH_LIST_END] = {
  * Map of supported aead algorithms.
  */
 static const
-struct cipher_params_mapping aead_map[RTE_CRYPTO_AEAD_LIST_END] = {
+struct cipher_params_mapping aead_map[] = {
[RTE_CRYPTO_AEAD_AES_GCM] = {
.supported = ALGO_SUPPORTED,
.cipher_alg = SAM_CIPHER_AES,
-- 
2.24.0

Re: [dpdk-dev] [PATCH v2 1/7] cmdline: make implementation opaque

[Kinsella, Ray]

On 05/08/2020 12:17, Dmitry Kozlyuk wrote:
> On Wed, 5 Aug 2020 10:31:31 +0100, Kinsella, Ray wrote:
>> On 30/07/2020 22:06, Dmitry Kozlyuk wrote:
>>> struct cmdline exposes platform-specific members it contains, most
>>> notably struct termios that is only available on Unix. Make the
>>> structure opaque.
>>>
>>> Remove tests checking struct cmdline content as meaningless.
>>>
>>> Add cmdline_get_rdline() to access history buffer.
>>> The new function is currently used only in tests.  
>>
>> Should it be INTERNAL then? Is it useful outside of the test cases?
> 
> There are already exposed rdline_*() functions that require struct rdline
> pointer, which is now only accessible via this function for struct cmdline
> instances. Thus, public API would be broken with INTERNAL for such use cases.
> 

Right but that runs a bit contrary to what you said in the commit log.

"Add cmdline_get_rdline() to access history buffer.
The new function is currently used only in tests."

In anycase, given the elapse in time since my feedback.
I will ACK the changes to MAP file. 

Ray K

Re: [dpdk-dev] [PATCH v3 1/7] cmdline: make implementation logically opaque

[Kinsella, Ray]

On 28/09/2020 22:50, Dmitry Kozlyuk wrote:
> struct cmdline exposes platform-specific members it contains, most
> notably struct termios that is only available on Unix. While ABI
> considerations prevent from hinding the definition on already supported
> platforms, struct cmdline is considered logically opaque from now on.
> Add a deprecation notice targeted at 20.11.
> 
> * Remove tests checking struct cmdline content as meaningless.
> 
> * Fix missing cmdline_free() in unit test.
> 
> * Add cmdline_get_rdline() to access history buffer indirectly.
>   The new function is currently used only in tests.
> 
> Signed-off-by: Dmitry Kozlyuk 
> Suggested-by: Olivier Matz 
> ---
>  app/test-cmdline/commands.c|  8 +++-
>  app/test/test_cmdline_lib.c| 44 +-
>  doc/guides/rel_notes/deprecation.rst   |  4 ++
>  lib/librte_cmdline/cmdline.c   |  9 +++--
>  lib/librte_cmdline/cmdline.h   | 18 -
>  lib/librte_cmdline/cmdline_parse.c |  3 --
>  lib/librte_cmdline/cmdline_socket.c|  5 +--
>  lib/librte_cmdline/rte_cmdline_version.map |  8 
>  8 files changed, 60 insertions(+), 39 deletions(-)

Acked-by: Ray Kinsella 

Re: [dpdk-dev] [PATCH V15 1/3] ethdev: introduce FEC API

[Thomas Monjalon]

29/09/2020 10:44, Min Hu (Connor):
> Hi, all,
>   Are there any suggestions for this set of patches?
>   I wish this patch can be merged into 20.11,and this
>   is urget. Please gives reply to me, thanks.

We all have urgent tasks to complete.
Please help by reviewing some other patches.

Re: [dpdk-dev] [PATCH v9 11/11] doc: support python3 only

[Kinsella, Ray]

On 29/09/2020 11:22, Kevin Laatz wrote:
> Remove references to python 2 from the documentation.
> 
> The deprecation notice for python 2 support has been removed. A section
> was added to the rel_notes for this patchset.
> 
> Cc: Ray Kinsella 
> Cc: Neil Horman 
> Cc: John McNamara 
> Cc: Marko Kovacevic 
> 
> Signed-off-by: Kevin Laatz 
> ---
>  doc/guides/contributing/coding_style.rst | 2 +-
>  doc/guides/linux_gsg/sys_reqs.rst| 2 +-
>  doc/guides/rel_notes/deprecation.rst | 6 --
>  doc/guides/rel_notes/release_20_11.rst   | 2 ++
>  4 files changed, 4 insertions(+), 8 deletions(-)
> 

Acked-by: Ray Kinsella 

[SNIP]

Re: [dpdk-dev] [PATCH] net/bnx2x: release port upon close

[Ferruh Yigit]

On 9/29/2020 7:20 PM, Jerin Jacob wrote:

On Mon, Sep 28, 2020 at 4:07 AM Rasesh Mody  wrote:


Set RTE_ETH_DEV_CLOSE_REMOVE upon probe so all the private resources
for the port can be freed by rte_eth_dev_close(). With this change the
private port resources are released in the .dev_close callback.

Signed-off-by: Rasesh Mody 



It looks like it has ICC compiler errors in this patch[1]. I don't
think, it is introduced by this patch. It is from CPT crypto driver.

[1]
http://mails.dpdk.org/archives/test-report/2020-September/155058.html



These patches seems pulled to the Thomas' "cleanup ethdev close operation" 
patchset:
[v3,17/29] net/qede: release port upon close 
(https://patches.dpdk.org/patch/79083/)
[v3,06/29] net/bnx2x: release port upon close 
(https://patches.dpdk.org/patch/79072/)


Can you please review the patches in that patchset? And they can be merged with 
the whole patchset.

I will mark this set as superseded.


Btw, agree that compile error looks unrelated.

Re: [dpdk-dev] [PATCH V5 1/2] dpdk: resolve compiling errors for per-queue stats

[Kinsella, Ray]

On 28/09/2020 09:59, Ferruh Yigit wrote:
> On 9/27/2020 4:16 AM, Min Hu (Connor) wrote:
>> From: Huisong Li 
>>
>> Currently, only statistics of rx/tx queues with queue_id less than
>> RTE_ETHDEV_QUEUE_STAT_CNTRS can be displayed. If there is a certain
>> application scenario that it needs to use 256 or more than 256 queues
>> and display all statistics of rx/tx queue. At this moment, we have to
>> change the macro to be equaled to the queue number.
>>
>> However, modifying the macro to be greater than 256 will trigger
>> many errors and warnings from test-pmd, PMD drivers and librte_ethdev
>> during compiling dpdk project. But it is possible and permitted that
>> rx/tx queue number is greater than 256 and all statistics of rx/tx
>> queue need to be displayed. In addition, the data type of rx/tx queue
>> number in rte_eth_dev_configure API is 'uint16_t'. So It is unreasonable
>> to use the 'uint8_t' type for variables that control which per-queue
>> statistics can be displayed.
>>
>> Fixes: ed30d9b691b2 ("app/testpmd: add stats per queue")
>> Fixes: 09c7e63a71f9 ("net/memif: introduce memory interface PMD")
>> Fixes: abf7275bbaa2 ("ixgbe: move to drivers/net/")
>> Fixes: e6defdfddc3b ("net/igc: enable statistics")
>> Fixes: 2265e4b4e84b ("net/octeontx2: add basic stats operation")
>> Fixes: 6c3169a3dc04 ("virtio: move to drivers/net/")
>>
>> Signed-off-by: Huisong Li 
>> Signed-off-by: Min Hu (Connor) 
>> Reviewed-by: Wei Hu (Xavier) 
>> Reviewed-by: Dongdong Liu 
>> ---
>> V4 -> V5:
>> add release notes updated.
>>
>> ---
>> v3->v4:
>> add a change in cmd_setqmap_mapvalue.
>>
>> ---
>> v2->v3:
>> change 'uint8_t i' to 'uint16_t i' in nic_stats_display function.
>>
>> ---
>>   app/proc-info/main.c   | 2 +-
>>   app/test-pmd/cmdline.c | 4 ++--
>>   app/test-pmd/config.c  | 4 ++--
>>   app/test-pmd/testpmd.c | 2 +-
>>   app/test-pmd/testpmd.h | 5 +++--
>>   doc/guides/rel_notes/release_20_11.rst | 5 +
>>   drivers/net/igc/igc_ethdev.c   | 4 ++--
>>   drivers/net/ixgbe/ixgbe_ethdev.c   | 4 ++--
>>   drivers/net/memif/rte_eth_memif.c  | 2 +-
>>   drivers/net/octeontx2/otx2_ethdev.h    | 2 +-
>>   drivers/net/octeontx2/otx2_stats.c | 2 +-
>>   drivers/net/virtio/virtio_ethdev.c | 4 ++--
>>   lib/librte_ethdev/rte_ethdev.c | 6 +++---
>>   lib/librte_ethdev/rte_ethdev.h | 4 ++--
>>   lib/librte_ethdev/rte_ethdev_driver.h  | 2 +-
>>   15 files changed, 29 insertions(+), 23 deletions(-)
>>
[SNIP]
> 
> cc'ed tech-board,
> 
> The patch breaks the ethdev ABI without a deprecation notice from previous 
> release(s).
> 
> It is mainly a fix to the port_id storage type, which we have updated from 
> uint8_t to uint16_t in past but some seems remained for 
> 'rte_eth_dev_set_tx_queue_stats_mapping()' & 
> 'rte_eth_dev_set_rx_queue_stats_mapping()' APIs.
> 
> Since the ethdev library already heavily breaks the ABI this release, I am 
> for getting this fix, instead of waiting the fix for one more year.
> 
> Can you please review the patch, is there any objectio to proceed with it?

After reading the rest of the thread, I understand that Thomas has suggested 
depreciating this entire API and using xstats instead.
My 2c is that if changing this value requires an ABI breakage and rebuild, its 
probably the wrong API.

Ray K

Re: [dpdk-dev] [PATCH v3 20/29] net/tap: release port upon close

[wangyunjian]

I have checked and tested it, no issue found.

Thanks,
Yunjian

> -Original Message-
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Tuesday, September 29, 2020 7:14 AM
> To: dev@dpdk.org
> Cc: ferruh.yi...@intel.com; arybche...@solarflare.com; wangyunjian
> ; Keith Wiles ; Anatoly
> Burakov 
> Subject: [PATCH v3 20/29] net/tap: release port upon close
> 
> From: Yunjian Wang 
> 
> The flag RTE_ETH_DEV_CLOSE_REMOVE is set so all port resources can be
> freed by rte_eth_dev_close().
> 
> Freeing of private port resources is moved from the ".remove(device)" to the
> ".dev_close(port)" operation.
> 
> Signed-off-by: Yunjian Wang 
> Signed-off-by: Thomas Monjalon 
> ---
>  drivers/net/tap/rte_eth_tap.c | 51 +++
>  1 file changed, 28 insertions(+), 23 deletions(-)
> 
> diff --git a/drivers/net/tap/rte_eth_tap.c b/drivers/net/tap/rte_eth_tap.c 
> index
> 2d1e4e365a..0d217f1486 100644
> --- a/drivers/net/tap/rte_eth_tap.c
> +++ b/drivers/net/tap/rte_eth_tap.c
> @@ -72,6 +72,10 @@
> 
>  static int tap_devices_count;
> 
> +static const char *tuntap_types[ETH_TUNTAP_TYPE_MAX] = {
> + "UNKNOWN", "TUN", "TAP"
> +};
> +
>  static const char *valid_arguments[] = {
>   ETH_TAP_IFACE_ARG,
>   ETH_TAP_REMOTE_ARG,
> @@ -1075,6 +1079,11 @@ tap_dev_close(struct rte_eth_dev *dev)
>   struct pmd_process_private *process_private = dev->process_private;
>   struct rx_queue *rxq;
> 
> + if (rte_eal_process_type() != RTE_PROC_PRIMARY) {
> + rte_free(dev->process_private);
> + return 0;
> + }
> +
>   tap_link_set_down(dev);
>   if (internals->nlsk_fd != -1) {
>   tap_flow_flush(dev, NULL);
> @@ -1112,6 +1121,23 @@ tap_dev_close(struct rte_eth_dev *dev)
>   close(internals->ka_fd);
>   internals->ka_fd = -1;
>   }
> +
> + /* mac_addrs must not be freed alone because part of dev_private */
> + dev->data->mac_addrs = NULL;
> +
> + internals = dev->data->dev_private;
> + TAP_LOG(DEBUG, "Closing %s Ethernet device on numa %u",
> + tuntap_types[internals->type], rte_socket_id());
> +
> + if (internals->ioctl_sock != -1) {
> + close(internals->ioctl_sock);
> + internals->ioctl_sock = -1;
> + }
> + rte_free(dev->process_private);
> + dev->process_private = NULL;
> + if (tap_devices_count == 1)
> + rte_mp_action_unregister(TAP_MP_KEY);
> + tap_devices_count--;
>   /*
>* Since TUN device has no more opened file descriptors
>* it will be removed from kernel
> @@ -1845,10 +1871,6 @@ static const struct eth_dev_ops ops = {
>   .filter_ctrl= tap_dev_filter_ctrl,
>  };
> 
> -static const char *tuntap_types[ETH_TUNTAP_TYPE_MAX] = {
> - "UNKNOWN", "TUN", "TAP"
> -};
> -
>  static int
>  eth_dev_tap_create(struct rte_vdev_device *vdev, const char *tap_name,
>  char *remote_iface, struct rte_ether_addr *mac_addr, @@
> -1900,7 +1922,7 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, const
> char *tap_name,
>   /* Setup some default values */
>   data = dev->data;
>   data->dev_private = pmd;
> - data->dev_flags = RTE_ETH_DEV_INTR_LSC;
> + data->dev_flags = RTE_ETH_DEV_INTR_LSC |
> RTE_ETH_DEV_CLOSE_REMOVE;
>   data->numa_node = numa_node;
> 
>   data->dev_link = pmd_link;
> @@ -2492,30 +2514,13 @@ static int
>  rte_pmd_tap_remove(struct rte_vdev_device *dev)  {
>   struct rte_eth_dev *eth_dev = NULL;
> - struct pmd_internals *internals;
> 
>   /* find the ethdev entry */
>   eth_dev = rte_eth_dev_allocated(rte_vdev_device_name(dev));
>   if (!eth_dev)
> - return -ENODEV;
> -
> - /* mac_addrs must not be freed alone because part of dev_private */
> - eth_dev->data->mac_addrs = NULL;
> -
> - if (rte_eal_process_type() != RTE_PROC_PRIMARY)
> - return rte_eth_dev_release_port(eth_dev);
> + return 0;
> 
>   tap_dev_close(eth_dev);
> -
> - internals = eth_dev->data->dev_private;
> - TAP_LOG(DEBUG, "Closing %s Ethernet device on numa %u",
> - tuntap_types[internals->type], rte_socket_id());
> -
> - close(internals->ioctl_sock);
> - rte_free(eth_dev->process_private);
> - if (tap_devices_count == 1)
> - rte_mp_action_unregister(TAP_MP_KEY);
> - tap_devices_count--;
>   rte_eth_dev_release_port(eth_dev);
> 
>   return 0;
> --
> 2.28.0

Re: [dpdk-dev] [PATCH] net/bonding: fix a possible unbalance packet receiving

[Wei Hu (Xavier)]

Hi, Li RongQing

Please add fixes statements in the commit log.

And for the patch [2/2] of this series:    Reviewed-by: Wei Hu (Xavier) 




On 2020/9/30 13:40, Li RongQing wrote:

ping

thanks

-Li

Re: [dpdk-dev] [PATCH 2/2] lpm: hide internal data

[Kevin Traynor]

On 16/09/2020 04:17, Ruifeng Wang wrote:
> 
>> -Original Message-
>> From: Medvedkin, Vladimir 
>> Sent: Wednesday, September 16, 2020 12:28 AM
>> To: Bruce Richardson ; Ruifeng Wang
>> 
>> Cc: dev@dpdk.org; Honnappa Nagarahalli
>> ; nd 
>> Subject: Re: [PATCH 2/2] lpm: hide internal data
>>
>> Hi Ruifeng,
>>
>> On 15/09/2020 17:02, Bruce Richardson wrote:
>>> On Mon, Sep 07, 2020 at 04:15:17PM +0800, Ruifeng Wang wrote:
 Fields except tbl24 and tbl8 in rte_lpm structure have no need to be
 exposed to the user.
 Hide the unneeded exposure of structure fields for better ABI
 maintainability.

 Suggested-by: David Marchand 
 Signed-off-by: Ruifeng Wang 
 Reviewed-by: Phil Yang 
 ---
   lib/librte_lpm/rte_lpm.c | 152 +++---
>> -
   lib/librte_lpm/rte_lpm.h |   7 --
   2 files changed, 91 insertions(+), 68 deletions(-)

>>> 
 diff --git a/lib/librte_lpm/rte_lpm.h b/lib/librte_lpm/rte_lpm.h
 index 03da2d37e..112d96f37 100644
 --- a/lib/librte_lpm/rte_lpm.h
 +++ b/lib/librte_lpm/rte_lpm.h
 @@ -132,17 +132,10 @@ struct rte_lpm_rule_info {

   /** @internal LPM structure. */
   struct rte_lpm {
 -  /* LPM metadata. */
 -  char name[RTE_LPM_NAMESIZE];/**< Name of the lpm. */
 -  uint32_t max_rules; /**< Max. balanced rules per lpm. */
 -  uint32_t number_tbl8s; /**< Number of tbl8s. */
 -  struct rte_lpm_rule_info rule_info[RTE_LPM_MAX_DEPTH]; /**<
>> Rule info table. */
 -
/* LPM Tables. */
struct rte_lpm_tbl_entry tbl24[RTE_LPM_TBL24_NUM_ENTRIES]
__rte_cache_aligned; /**< LPM tbl24 table. */
struct rte_lpm_tbl_entry *tbl8; /**< LPM tbl8 table. */
 -  struct rte_lpm_rule *rules_tbl; /**< LPM rules. */
   };

>>>
>>> Since this changes the ABI, does it not need advance notice?
>>>
>>> [Basically the return value point from rte_lpm_create() will be
>>> different, and that return value could be used by rte_lpm_lookup()
>>> which as a static inline function will be in the binary and using the
>>> old structure offsets.]
>>>
>>
>> Agree with Bruce, this patch breaks ABI, so it can't be accepted without 
>> prior
>> notice.
>>
> So if the change wants to happen in 20.11, a deprecation notice should have 
> been
> added in 20.08.
> I should have added a deprecation notice. This change will have to wait for 
> next ABI update window.
> 

Do you plan to extend? or is this just speculative?

A quick scan and there seems to be several projects using some of these
members that you are proposing to hide. e.g. BESS, NFF-Go, DPVS,
gatekeeper. I didn't look at the details to see if they are really needed.

Not sure how much notice they'd need or if they update DPDK much, but I
think it's worth having a closer look as to how they use lpm and what
the impact to them is.

> Thanks.
> Ruifeng
   /** LPM RCU QSBR configuration structure. */
 --
 2.17.1

>>
>> --
>> Regards,
>> Vladimir

Re: [dpdk-dev] [PATCH v2 5/6] ethdev: fix mis-named parameter

[Bruce Richardson]

On Tue, Sep 29, 2020 at 07:34:17PM +0200, Thomas Monjalon wrote:
> > The parameter to rte_eth_link_speed_to_str was called "speed_link" in
> > the function prototype in the header file, but "link_speed" everywhere
> > else.  This showed up as warnings when building the API docs, due to
> > missing and undocumented parameters from doxygen's viewpoint.
> > 
> > Rename the prototype value to the correct name to fix these issues.
> 
> It has been fixed by David and merged already by Ferruh.
> 
Yep, I guessed there was a patch outstanding for it from someone, though I
think the patch wasn't merged when I did this patch, so I included it to
guarantee that this patchset didn't break the build when werror was turned
on.

/Bruce

Re: [dpdk-dev] [dpdk-techboard] [PATCH v2] eal: simplify exit functions

[Kevin Traynor]

On 28/09/2020 01:00, Thomas Monjalon wrote:
> The option RTE_EAL_ALWAYS_PANIC_ON_ERROR was off by default,
> and not customizable with meson. It is completely removed.
> 
> The function rte_dump_registers is a trace of the bare metal support
> era, and was not supported in userland. It is completely removed.
> 
> Signed-off-by: Thomas Monjalon 
> Acked-by: Ray Kinsella 
> ---
> The deprecation notice for this removal has been missed.
> I assume it would not hurt anybody to remove this useless function
> from DPDK 20.11. Asking the Technical Board for confirmation.
> ---

Acked-by: Kevin Traynor 

>  app/test/test_debug.c|  3 ---
>  doc/guides/howto/debug_troubleshoot.rst  |  2 +-
>  doc/guides/rel_notes/release_20_11.rst   |  2 ++
>  lib/librte_eal/common/eal_common_debug.c | 17 +
>  lib/librte_eal/include/rte_debug.h   |  7 ---
>  lib/librte_eal/rte_eal_version.map   |  1 -
>  6 files changed, 4 insertions(+), 28 deletions(-)
> 
> diff --git a/app/test/test_debug.c b/app/test/test_debug.c
> index 25eab97e2a..834a7386f5 100644
> --- a/app/test/test_debug.c
> +++ b/app/test/test_debug.c
> @@ -66,13 +66,11 @@ test_exit_val(int exit_val)
>   }
>   wait(&status);
>   printf("Child process status: %d\n", status);
> -#ifndef RTE_EAL_ALWAYS_PANIC_ON_ERROR
>   if(!WIFEXITED(status) || WEXITSTATUS(status) != (uint8_t)exit_val){
>   printf("Child process terminated with incorrect status 
> (expected = %d)!\n",
>   exit_val);
>   return -1;
>   }
> -#endif
>   return 0;
>  }
>  
> @@ -113,7 +111,6 @@ static int
>  test_debug(void)
>  {
>   rte_dump_stack();
> - rte_dump_registers();
>   if (test_panic() < 0)
>   return -1;
>   if (test_exit() < 0)
> diff --git a/doc/guides/howto/debug_troubleshoot.rst 
> b/doc/guides/howto/debug_troubleshoot.rst
> index 5a46f5fba3..50bd32a8ef 100644
> --- a/doc/guides/howto/debug_troubleshoot.rst
> +++ b/doc/guides/howto/debug_troubleshoot.rst
> @@ -314,7 +314,7 @@ Custom worker function :numref:`dtg_distributor_worker`.
> * For high-performance execution logic ensure running it on correct NUMA
>   and non-master core.
>  
> -   * Analyze run logic with ``rte_dump_stack``, ``rte_dump_registers`` and
> +   * Analyze run logic with ``rte_dump_stack`` and
>   ``rte_memdump`` for more insights.
>  
> * Make use of objdump to ensure opcode is matching to the desired state.
> diff --git a/doc/guides/rel_notes/release_20_11.rst 
> b/doc/guides/rel_notes/release_20_11.rst
> index f377ab8e87..c0b83e9554 100644
> --- a/doc/guides/rel_notes/release_20_11.rst
> +++ b/doc/guides/rel_notes/release_20_11.rst
> @@ -184,6 +184,8 @@ ABI Changes
> Also, make sure to start the actual text at the margin.
> ===
>  
> +* eal: Removed the not implemented function ``rte_dump_registers()``.
> +
>  * ``ethdev`` changes
>  
>* Following device operation function pointers moved
> diff --git a/lib/librte_eal/common/eal_common_debug.c 
> b/lib/librte_eal/common/eal_common_debug.c
> index 722468754d..15418e957f 100644
> --- a/lib/librte_eal/common/eal_common_debug.c
> +++ b/lib/librte_eal/common/eal_common_debug.c
> @@ -7,14 +7,6 @@
>  #include 
>  #include 
>  
> -/* not implemented */
> -void
> -rte_dump_registers(void)
> -{
> - return;
> -}
> -
> -/* call abort(), it will generate a coredump if enabled */
>  void
>  __rte_panic(const char *funcname, const char *format, ...)
>  {
> @@ -25,8 +17,7 @@ __rte_panic(const char *funcname, const char *format, ...)
>   rte_vlog(RTE_LOG_CRIT, RTE_LOGTYPE_EAL, format, ap);
>   va_end(ap);
>   rte_dump_stack();
> - rte_dump_registers();
> - abort();
> + abort(); /* generate a coredump if enabled */
>  }
>  
>  /*
> @@ -46,14 +37,8 @@ rte_exit(int exit_code, const char *format, ...)
>   rte_vlog(RTE_LOG_CRIT, RTE_LOGTYPE_EAL, format, ap);
>   va_end(ap);
>  
> -#ifndef RTE_EAL_ALWAYS_PANIC_ON_ERROR
>   if (rte_eal_cleanup() != 0)
>   RTE_LOG(CRIT, EAL,
>   "EAL could not release all resources\n");
>   exit(exit_code);
> -#else
> - rte_dump_stack();
> - rte_dump_registers();
> - abort();
> -#endif
>  }
> diff --git a/lib/librte_eal/include/rte_debug.h 
> b/lib/librte_eal/include/rte_debug.h
> index 50052c5a90..c4bc71ce28 100644
> --- a/lib/librte_eal/include/rte_debug.h
> +++ b/lib/librte_eal/include/rte_debug.h
> @@ -26,13 +26,6 @@ extern "C" {
>   */
>  void rte_dump_stack(void);
>  
> -/**
> - * Dump the registers of the calling core to the console.
> - *
> - * Note: Not implemented in a userapp environment; use gdb instead.
> - */
> -void rte_dump_registers(void);
> -
>  /**
>   * Provide notification of a critical non-recoverable error and terminate
>   * execution abnormally.
> diff --git a/lib/librte_eal/rte_eal_version.map 
> b/lib/librte_eal/rte_eal_ver

[dpdk-dev] [PATCH] ethdev: support query of AGE action

[Dekel Peled]

Existing API supports AGE action to monitor the aging of a flow.
This patch implements RFC [1], introducing the response format for query
of an AGE action.
Application will be able to query the AGE action state.
The response will be returned in the format implemented here.

[1] https://mails.dpdk.org/archives/dev/2020-September/180061.html

Signed-off-by: Dekel Peled 
Acked-by: Matan Azrad 
---

Testpmd updates to follow shortly.

 doc/guides/prog_guide/rte_flow.rst | 17 +
 doc/guides/rel_notes/release_20_11.rst |  3 +++
 lib/librte_ethdev/rte_flow.h   | 14 ++
 3 files changed, 34 insertions(+)

diff --git a/doc/guides/prog_guide/rte_flow.rst 
b/doc/guides/prog_guide/rte_flow.rst
index 119b128..4b8d033 100644
--- a/doc/guides/prog_guide/rte_flow.rst
+++ b/doc/guides/prog_guide/rte_flow.rst
@@ -2666,6 +2666,23 @@ timeout passed without any matching on the flow.
| ``context``  | user input flow context |
+--+-+
 
+Query structure to retrieve ageing status information of a
+shared AGE action, or a flow rule using the AGE action:
+
+.. _table_rte_flow_query_age:
+
+.. table:: AGE query
+
+   +-+-+---+
+   | Field   | I/O | Value |
+   +=+=+===+
+   | ``aged``| out | Aging timeout expired |
+   +-+-+---+
+   | ``last_hit_time_valid`` | out | ``last_hit_time`` field is valid  |
+   +-+-+---+
+   | ``last_hit_time``   | out | Seconds since last traffic hit|
+   +-+-+---+
+
 Negative types
 ~~
 
diff --git a/doc/guides/rel_notes/release_20_11.rst 
b/doc/guides/rel_notes/release_20_11.rst
index 4bcf220..9a397d2 100644
--- a/doc/guides/rel_notes/release_20_11.rst
+++ b/doc/guides/rel_notes/release_20_11.rst
@@ -162,6 +162,9 @@ API Changes
 
 * ethdev: ``rte_eth_rx_descriptor_done()`` API has been deprecated.
 
+* ethdev: Added struct ``rte_flow_query_age`` for use in response to query
+  of AGE action.
+
 * Renamed internal ethdev APIs:
 
   * ``_rte_eth_dev_callback_process()`` -> ``rte_eth_dev_callback_process()``
diff --git a/lib/librte_ethdev/rte_flow.h b/lib/librte_ethdev/rte_flow.h
index da8bfa5..1a13802 100644
--- a/lib/librte_ethdev/rte_flow.h
+++ b/lib/librte_ethdev/rte_flow.h
@@ -2130,6 +2130,7 @@ enum rte_flow_action_type {
 * See struct rte_flow_action_age.
 * See function rte_flow_get_aged_flows
 * see enum RTE_ETH_EVENT_FLOW_AGED
+* See struct rte_flow_query_age
 */
RTE_FLOW_ACTION_TYPE_AGE,
 };
@@ -2194,6 +2195,19 @@ struct rte_flow_action_age {
 };
 
 /**
+ * RTE_FLOW_ACTION_TYPE_AGE (query)
+ *
+ * Query structure to retrieve the aging status information of a
+ * shared AGE action, or a flow rule using the AGE action.
+ */
+struct rte_flow_query_age {
+   uint32_t aged:1; /**< 1 if aging timeout expired, 0 otherwise. */
+   uint32_t last_hit_time_valid:1; /**< 1 if last_hit_time value valid. */
+   uint32_t reserved:6; /**< Reserved, must be zero. */
+   uint32_t last_hit_time:24; /**< Seconds since last traffic hit. */
+};
+
+/**
  * @warning
  * @b EXPERIMENTAL: this structure may change without prior notice
  *
-- 
1.8.3.1

Re: [dpdk-dev] [PATCH] net/bonding: fix a possible unbalance packet receiving

[Li,Rongqing]

Fixes: ae2a04864a9a "(net/bonding: reduce slave starvation on Rx poll)"

Thanks

-Li

From: Wei Hu (Xavier) [mailto:huwei...@chinasoftinc.com]
Sent: Wednesday, September 30, 2020 4:41 PM
To: Li,Rongqing 
Cc: dev@dpdk.org; Wei Hu (Xavier) 
Subject: Re: [dpdk-dev] [PATCH] net/bonding: fix a possible unbalance packet 
receiving


Hi, Li RongQing

Please add fixes statements in the commit log.

And for the patch [2/2] of this series:Reviewed-by: Wei Hu (Xavier) 

[dpdk-dev] [PATCH v3 0/4] Tunnel Offload API

[Gregory Etelson]

Tunnel Offload API provides hardware independent, unified model
to offload tunneled traffic. Key model elements are:
 - apply matches to both outer and inner packet headers
   during entire offload procedure;
 - restore outer header of partially offloaded packet;
 - model is implemented as a set of helper functions.

Eli Britstein (1):
  ethdev: tunnel offload model

Gregory Etelson (3):
  ethdev: allow negative values in flow rule types
  net/mlx5: implement tunnel offload API
  app/testpmd: add commands for tunnel offload API

 app/test-pmd/cmdline_flow.c | 170 -
 app/test-pmd/config.c   | 253 +++-
 app/test-pmd/testpmd.c  |   5 +-
 app/test-pmd/testpmd.h  |  34 +-
 app/test-pmd/util.c |  35 +-
 doc/guides/nics/mlx5.rst|   3 +
 doc/guides/prog_guide/rte_flow.rst  | 105 +++
 doc/guides/rel_notes/release_20_11.rst  |  10 +
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |  49 ++
 drivers/net/mlx5/linux/mlx5_os.c|  18 +
 drivers/net/mlx5/mlx5.c |   8 +-
 drivers/net/mlx5/mlx5.h |   3 +
 drivers/net/mlx5/mlx5_defs.h|   2 +
 drivers/net/mlx5/mlx5_flow.c| 678 +++-
 drivers/net/mlx5/mlx5_flow.h| 173 -
 drivers/net/mlx5/mlx5_flow_dv.c | 241 ++-
 lib/librte_ethdev/rte_ethdev_version.map|   6 +
 lib/librte_ethdev/rte_flow.c| 140 +++-
 lib/librte_ethdev/rte_flow.h| 195 ++
 lib/librte_ethdev/rte_flow_driver.h |  32 +
 20 files changed, 2095 insertions(+), 65 deletions(-)

-- 
2.25.1

[dpdk-dev] [PATCH v3 2/4] ethdev: tunnel offload model

[Gregory Etelson]

From: Eli Britstein 

Rte_flow API provides the building blocks for vendor agnostic flow
classification offloads.  The rte_flow match and action primitives are
fine grained, thus enabling DPDK applications the flexibility to
offload network stacks and complex pipelines.

Applications wishing to offload complex data structures (e.g. tunnel
virtual ports) are required to use the rte_flow primitives, such as
group, meta, mark, tag and others to model their high level objects.

The hardware model design for high level software objects is not
trivial.  Furthermore, an optimal design is often vendor specific.

The goal of this API is to provide applications with the hardware
offload model for common high level software objects which is optimal
in regards to the underlying hardware.

Tunnel ports are the first of such objects.

Tunnel ports

Ingress processing of tunneled traffic requires the classification of
the tunnel type followed by a decap action.

In software, once a packet is decapsulated the in_port field is
changed to a virtual port representing the tunnel type. The outer
header fields are stored as packet metadata members and may be matched
by proceeding flows.

Openvswitch, for example, uses two flows:
1. classification flow - setting the virtual port representing the
tunnel type For example: match on udp port 4789
actions=tnl_pop(vxlan_vport)
2. steering flow according to outer and inner header matches match on
in_port=vxlan_vport and outer/inner header matches actions=forward to
p ort X The benefits of multi-flow tables are described in [1].

Offloading tunnel ports
---
Tunnel ports introduce a new stateless field that can be matched on.
Currently the rte_flow library provides an API to encap, decap and
match on tunnel headers. However, there is no rte_flow primitive to
set and match tunnel virtual ports.

There are several possible hardware models for offloading virtual
tunnel port flows including, but not limited to, the following:
1. Setting the virtual port on a hw register using the
rte_flow_action_mark/ rte_flow_action_tag/rte_flow_set_meta objects.
2. Mapping a virtual port to an rte_flow group
3. Avoiding the need to match on transient objects by merging
multi-table flows to a single rte_flow rule.

Every approach has its pros and cons.  The preferred approach should
take into account the entire system architecture and is very often
vendor specific.

The proposed rte_flow_tunnel_decap_set helper function (drafted below)
is designed to provide a common, vendor agnostic, API for setting the
virtual port value.  The helper API enables PMD implementations to
return vendor specific combination of rte_flow actions realizing the
vendor's hardware model for setting a tunnel port.  Applications may
append the list of actions returned from the helper function when
creating an rte_flow rule in hardware.

Similarly, the rte_flow_tunnel_match helper (drafted below)
allows for multiple hardware implementations to return a list of
fte_flow items.

Miss handling
-
Packets going through multiple rte_flow groups are exposed to hw
misses due to partial packet processing. In such cases, the software
should continue the packet's processing from the point where the
hardware missed.

We propose a generic rte_flow_restore structure providing the state
that was stored in hardware when the packet missed.

Currently, the structure will provide the tunnel state of the packet
that missed, namely:
1. The group id that missed
2. The tunnel port that missed
3. Tunnel information that was stored in memory (due to decap action).
In the future, we may add additional fields as more state may be
stored in the device memory (e.g. ct_state).

Applications may query the state via a new
rte_flow_tunnel_get_restore_info(mbuf) API, thus allowing
a vendor specific implementation.

VXLAN Code example:
Assume application needs to do inner NAT on VXLAN packet.
The first  rule in group 0:

flow create  ingress group 0
  pattern eth / ipv4 / udp dst is 4789 / vxlan / end
  actions {pmd actions} / jump group 3 / end

First VXLAN packet that arrives matches the rule in group 0 and jumps
to group 3 In group 3 the packet will miss since there is no flow to
match and will be uploaded to application.  Application  will call
rte_flow_get_restore_info() to get the packet outer header.
Application will insert a new rule in group 3 to match outer and inner
headers:

flow create  ingress group 3
  pattern {pmd items} / eth / ipv4 dst is 172.10.10.1 /
  udp dst 4789 / vxlan vni is 10 /
  ipv4 dst is 184.1.2.3 / end
  actions  set_ipv4_dst  186.1.1.1 / queue index 3 / end

Resulting of rules will be that VXLAN packet with vni=10, outer IPv4
dst=172.10.10.1 and inner IPv4 dst=184.1.2.3 will be received decaped
on queue 3 with IPv4 dst=186.1.1.1

Note: Packet in group 3 is considered decaped. All actions in that
group will be done on header that was inner before decap. Application
may specify ou

[dpdk-dev] [PATCH v3 1/4] ethdev: allow negative values in flow rule types

[Gregory Etelson]

From: Gregory Etelson 

RTE flow items & actions use positive values in item & action type.
Negative values are reserved for PMD private types. PMD
items & actions usually are not exposed to application and are not
used to create RTE flows.

The patch allows applications with access to PMD flow
items & actions ability to integrate RTE and PMD items & actions
and use them to create flow rule.

RTE flow library functions cannot work with PMD private items and
actions (elements) because RTE flow has no API to query PMD flow
object size. In the patch, PMD flow elements use object pointer.
RTE flow library functions handle PMD element object size as
size of a pointer. PMD handles its objects internally.

Signed-off-by: Gregory Etelson 
Acked-by: Ori Kam 
Acked-by: Viacheslav Ovsiienko 
---
 lib/librte_ethdev/rte_flow.c | 28 ++--
 1 file changed, 22 insertions(+), 6 deletions(-)

diff --git a/lib/librte_ethdev/rte_flow.c b/lib/librte_ethdev/rte_flow.c
index f8fdd68fe9..c8c6d62a8b 100644
--- a/lib/librte_ethdev/rte_flow.c
+++ b/lib/librte_ethdev/rte_flow.c
@@ -564,7 +564,11 @@ rte_flow_conv_item_spec(void *buf, const size_t size,
}
break;
default:
-   off = rte_flow_desc_item[item->type].size;
+   /**
+* allow PMD private flow item
+*/
+   off = (int)item->type >= 0 ?
+ rte_flow_desc_item[item->type].size : sizeof(void *);
rte_memcpy(buf, data, (size > off ? off : size));
break;
}
@@ -667,7 +671,11 @@ rte_flow_conv_action_conf(void *buf, const size_t size,
}
break;
default:
-   off = rte_flow_desc_action[action->type].size;
+   /**
+* allow PMD private flow action
+*/
+   off = (int)action->type >= 0 ?
+ rte_flow_desc_action[action->type].size : sizeof(void *);
rte_memcpy(buf, action->conf, (size > off ? off : size));
break;
}
@@ -709,8 +717,12 @@ rte_flow_conv_pattern(struct rte_flow_item *dst,
unsigned int i;
 
for (i = 0, off = 0; !num || i != num; ++i, ++src, ++dst) {
-   if ((size_t)src->type >= RTE_DIM(rte_flow_desc_item) ||
-   !rte_flow_desc_item[src->type].name)
+   /**
+* allow PMD private flow item
+*/
+   if (((int)src->type >= 0) &&
+   ((size_t)src->type >= RTE_DIM(rte_flow_desc_item) ||
+   !rte_flow_desc_item[src->type].name))
return rte_flow_error_set
(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ITEM, src,
 "cannot convert unknown item type");
@@ -798,8 +810,12 @@ rte_flow_conv_actions(struct rte_flow_action *dst,
unsigned int i;
 
for (i = 0, off = 0; !num || i != num; ++i, ++src, ++dst) {
-   if ((size_t)src->type >= RTE_DIM(rte_flow_desc_action) ||
-   !rte_flow_desc_action[src->type].name)
+   /**
+* allow PMD private flow action
+*/
+   if (((int)src->type >= 0) &&
+   ((size_t)src->type >= RTE_DIM(rte_flow_desc_action) ||
+   !rte_flow_desc_action[src->type].name))
return rte_flow_error_set
(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ACTION,
 src, "cannot convert unknown action type");
-- 
2.25.1

[dpdk-dev] [PATCH v3 3/4] net/mlx5: implement tunnel offload API

[Gregory Etelson]

Tunnel Offload API provides hardware independent, unified model
to offload tunneled traffic. Key model elements are:
 - apply matches to both outer and inner packet headers
   during entire offload procedure;
 - restore outer header of partially offloaded packet;
 - model is implemented as a set of helper functions.

Implementation details:
* tunnel_offload PMD parameter must be set to 1 to enable the feature.
* application cannot use MARK and META flow actions whith tunnel.
* offload JUMP action is restricted to steering tunnel rule only.

Signed-off-by: Gregory Etelson 
Acked-by: Viacheslav Ovsiienko 
---
v2:
* introduce MLX5 PMD API implementation
v3:
* bug fixes
---
 doc/guides/nics/mlx5.rst |   3 +
 drivers/net/mlx5/linux/mlx5_os.c |  18 +
 drivers/net/mlx5/mlx5.c  |   8 +-
 drivers/net/mlx5/mlx5.h  |   3 +
 drivers/net/mlx5/mlx5_defs.h |   2 +
 drivers/net/mlx5/mlx5_flow.c | 678 ++-
 drivers/net/mlx5/mlx5_flow.h | 173 +++-
 drivers/net/mlx5/mlx5_flow_dv.c  | 241 +--
 8 files changed, 1080 insertions(+), 46 deletions(-)

diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst
index 211c0c5a6c..287fd23b43 100644
--- a/doc/guides/nics/mlx5.rst
+++ b/doc/guides/nics/mlx5.rst
@@ -815,6 +815,9 @@ Driver options
 24 bits. The actual supported width can be retrieved in runtime by
 series of rte_flow_validate() trials.
 
+  - 3, this engages tunnel offload mode. In E-Switch configuration, that
+mode implicitly activates ``dv_xmeta_en=1``.
+
   +--+---+---+-+-+
   | Mode | ``MARK``  | ``META``  | ``META`` Tx | FDB/Through |
   +==+===+===+=+=+
diff --git a/drivers/net/mlx5/linux/mlx5_os.c b/drivers/net/mlx5/linux/mlx5_os.c
index 81a2e99e71..ecf4d2f2a6 100644
--- a/drivers/net/mlx5/linux/mlx5_os.c
+++ b/drivers/net/mlx5/linux/mlx5_os.c
@@ -298,6 +298,12 @@ mlx5_alloc_shared_dr(struct mlx5_priv *priv)
sh->esw_drop_action = mlx5_glue->dr_create_flow_action_drop();
}
 #endif
+   if (!sh->tunnel_hub)
+   err = mlx5_alloc_tunnel_hub(sh);
+   if (err) {
+   DRV_LOG(ERR, "mlx5_alloc_tunnel_hub failed err=%d", err);
+   goto error;
+   }
if (priv->config.reclaim_mode == MLX5_RCM_AGGR) {
mlx5_glue->dr_reclaim_domain_memory(sh->rx_domain, 1);
mlx5_glue->dr_reclaim_domain_memory(sh->tx_domain, 1);
@@ -344,6 +350,10 @@ mlx5_alloc_shared_dr(struct mlx5_priv *priv)
mlx5_hlist_destroy(sh->tag_table, NULL, NULL);
sh->tag_table = NULL;
}
+   if (sh->tunnel_hub) {
+   mlx5_release_tunnel_hub(sh, priv->dev_port);
+   sh->tunnel_hub = NULL;
+   }
mlx5_free_table_hash_list(priv);
return err;
 }
@@ -405,6 +415,10 @@ mlx5_os_free_shared_dr(struct mlx5_priv *priv)
mlx5_hlist_destroy(sh->tag_table, NULL, NULL);
sh->tag_table = NULL;
}
+   if (sh->tunnel_hub) {
+   mlx5_release_tunnel_hub(sh, priv->dev_port);
+   sh->tunnel_hub = NULL;
+   }
mlx5_free_table_hash_list(priv);
 }
 
@@ -658,6 +672,10 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev,
strerror(rte_errno));
goto error;
}
+   if (config->dv_miss_info) {
+   if (switch_info->master || switch_info->representor)
+   config->dv_xmeta_en = MLX5_XMETA_MODE_META16;
+   }
mlx5_malloc_mem_select(config->sys_mem_en);
sh = mlx5_alloc_shared_dev_ctx(spawn, config);
if (!sh)
diff --git a/drivers/net/mlx5/mlx5.c b/drivers/net/mlx5/mlx5.c
index 4a807fb4fd..569a865da8 100644
--- a/drivers/net/mlx5/mlx5.c
+++ b/drivers/net/mlx5/mlx5.c
@@ -1590,13 +1590,17 @@ mlx5_args_check(const char *key, const char *val, void 
*opaque)
} else if (strcmp(MLX5_DV_XMETA_EN, key) == 0) {
if (tmp != MLX5_XMETA_MODE_LEGACY &&
tmp != MLX5_XMETA_MODE_META16 &&
-   tmp != MLX5_XMETA_MODE_META32) {
+   tmp != MLX5_XMETA_MODE_META32 &&
+   tmp != MLX5_XMETA_MODE_MISS_INFO) {
DRV_LOG(ERR, "invalid extensive "
 "metadata parameter");
rte_errno = EINVAL;
return -rte_errno;
}
-   config->dv_xmeta_en = tmp;
+   if (tmp != MLX5_XMETA_MODE_MISS_INFO)
+   config->dv_xmeta_en = tmp;
+   else
+   config->dv_miss_info = 1;
} else if (strcmp(MLX5_LACP_BY_USER, key) == 0) {
config->lacp_by_user = !!tmp;
} else if (strcmp(MLX5_MR_EXT_MEMSEG_EN, key) == 0) {
diff --git a/drivers/net/mlx5/mlx5.h b/drivers/net/mlx5/mlx5.h
index 0907506755..e12c4cee4b 1006

[dpdk-dev] [PATCH v3 4/4] app/testpmd: add commands for tunnel offload API

[Gregory Etelson]

Tunnel Offload API provides hardware independent, unified model
to offload tunneled traffic. Key model elements are:
 - apply matches to both outer and inner packet headers
   during entire offload procedure;
 - restore outer header of partially offloaded packet;
 - model is implemented as a set of helper functions.

Implementation details:

* Create application tunnel:
flow tunnel create  type 
On success, the command creates application tunnel object and returns
the tunnel descriptor. Tunnel descriptor is used in subsequent flow
creation commands to reference the tunnel.

* Create tunnel steering flow rule:
tunnel_set  parameter used with steering rule
template.

* Create tunnel matching flow rule:
tunnel_match  used with matching rule template.

* If tunnel steering rule was offloaded, outer header of a partially
offloaded packet is restored after miss.

Example:
test packet=
>>
>>> len(packet)
92

testpmd> flow flush 0
testpmd> port 0/queue 0: received 1 packets
src=50:6B:4B:CC:FC:E2 - dst=24:8A:07:8D:AE:D6 - type=0x0800 -
length=92

testpmd> flow tunnel 0 type vxlan
port 0: flow tunnel #1 type vxlan
testpmd> flow create 0 ingress group 0 tunnel_set 1
 pattern eth /ipv4 / udp dst is 4789 / vxlan / end
 actions  jump group 0 / end
Flow rule #0 created
testpmd> port 0/queue 0: received 1 packets
tunnel restore info: - vxlan tunnel - outer header present # <--
  src=50:6B:4B:CC:FC:E2 - dst=24:8A:07:8D:AE:D6 - type=0x0800 -
length=92

testpmd> flow create 0 ingress group 0 tunnel_match 1
 pattern eth / ipv4 / udp dst is 4789 / vxlan / eth / ipv4 /
 end
 actions set_mac_dst mac_addr 02:CA:FE:CA:FA:80 /
 queue index 0 / end
Flow rule #1 created
testpmd> port 0/queue 0: received 1 packets
  src=50:BB:BB:BB:BB:E2 - dst=02:CA:FE:CA:FA:80 - type=0x0800 -
length=42

* Destroy flow tunnel
flow tunnel destroy  id 

* Show existing flow tunnels
flow tunnel list 

Signed-off-by: Gregory Etelson 
---
v2:
* introduce testpmd support for tunnel offload API

v3:
* update flow tunnel commands
---
 app/test-pmd/cmdline_flow.c | 170 -
 app/test-pmd/config.c   | 253 +++-
 app/test-pmd/testpmd.c  |   5 +-
 app/test-pmd/testpmd.h  |  34 ++-
 app/test-pmd/util.c |  35 ++-
 doc/guides/testpmd_app_ug/testpmd_funcs.rst |  49 
 6 files changed, 533 insertions(+), 13 deletions(-)

diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c
index 6263d307ed..0fb61860cd 100644
--- a/app/test-pmd/cmdline_flow.c
+++ b/app/test-pmd/cmdline_flow.c
@@ -69,6 +69,14 @@ enum index {
LIST,
AGED,
ISOLATE,
+   TUNNEL,
+
+   /* Tunnel argumens. */
+   TUNNEL_CREATE,
+   TUNNEL_CREATE_TYPE,
+   TUNNEL_LIST,
+   TUNNEL_DESTROY,
+   TUNNEL_DESTROY_ID,
 
/* Destroy arguments. */
DESTROY_RULE,
@@ -88,6 +96,8 @@ enum index {
INGRESS,
EGRESS,
TRANSFER,
+   TUNNEL_SET,
+   TUNNEL_MATCH,
 
/* Validate/create pattern. */
PATTERN,
@@ -653,6 +663,7 @@ struct buffer {
union {
struct {
struct rte_flow_attr attr;
+   struct tunnel_ops tunnel_ops;
struct rte_flow_item *pattern;
struct rte_flow_action *actions;
uint32_t pattern_n;
@@ -713,10 +724,32 @@ static const enum index next_vc_attr[] = {
INGRESS,
EGRESS,
TRANSFER,
+   TUNNEL_SET,
+   TUNNEL_MATCH,
PATTERN,
ZERO,
 };
 
+static const enum index tunnel_create_attr[] = {
+   TUNNEL_CREATE,
+   TUNNEL_CREATE_TYPE,
+   END,
+   ZERO,
+};
+
+static const enum index tunnel_destroy_attr[] = {
+   TUNNEL_DESTROY,
+   TUNNEL_DESTROY_ID,
+   END,
+   ZERO,
+};
+
+static const enum index tunnel_list_attr[] = {
+   TUNNEL_LIST,
+   END,
+   ZERO,
+};
+
 static const enum index next_destroy_attr[] = {
DESTROY_RULE,
END,
@@ -1516,6 +1549,9 @@ static int parse_aged(struct context *, const struct 
token *,
 static int parse_isolate(struct context *, const struct token *,
 const char *, unsigned int,
 void *, unsigned int);
+static int parse_tunnel(struct context *, const struct token *,
+   const char *, unsigned int,
+   void *, unsigned int);
 static int parse_int(struct context *, const struct token *,
 const char *, unsigned int,
 void *, unsigned int);
@@ -1698,7 +1734,8 @@ static const struct token token_list[] = {
  LIST,
  AGED,
  QUERY,
- ISOLATE)),
+ ISOLATE,
+ TUNNEL)),

Re: [dpdk-dev] [RFC PATCH v3 1/3] ethdev: support device reset and recovery events

[Thomas Monjalon]

30/09/2020 10:35, Kalesh Anakkur Purayil:
> On Wed, Sep 30, 2020 at 1:21 PM Thomas Monjalon  wrote:
> > > From: Kalesh AP 
> > >
> > > Adding support for device reset and recovery events in the
> > > rte_eth_event framework. FW error and FW reset conditions would be
> > > managed internally by PMD without needing application intervention.
> > > In such cases, PMD would need reset/recovery events to notify application
> > > that PMD is undergoing a reset.
> >
> > We already have this event:
> >
> > RTE_ETH_EVENT_INTR_RESET,
> > /**< reset interrupt event, sent to VF on PF reset */
> >
> > I don't know why "INTR" is in the name of this event,
> > and I think it does not need to be restricted to VF.
> > The application does not need to know whether the reset
> > is caused by the PF, the FW or the HW.
> > That's why I think you could share the same event.
> >
> 
> [Kalesh]: Yes. As you mentioned, this event is used for some other purpose.
> I did not want to break the existing usage/purpose of this event.
> For example, upon receiving the RTE_ETH_EVENT_INTR_RESET event OVS
> application invokes rte_eth_dev_reset() to reset the port.
> The aim here is to recover from the device error condition without the
> intervention of Applications. PMD itself will recover from the error using
> the protocol with FW.
> 
> >
> > > +   RTE_ETH_EVENT_RESET,/**< port resetting from an error */
> > > +   RTE_ETH_EVENT_RECOVERED, /**< port recovered from an error */
> >
> > You ignored my previous comments:
> > "
> > What the application is supposed to do when receiving such event?
> > How the application knows that flow rules were resetted?
> > Is there any other configuration resetted?
> > These informations must be explicit in the doxygen comments.
> > "
> >
> [Kalesh]: Sorry, I missed it.
> I am not sure what you meant by "These information must be explicit in the
> doxygen comments ".
> Could you please elaborate a little how to/where to put these details?

/** is the start of a doxygen comment.
This is the place (in the .h file) to explain to application
developer what to do with the event.
The code + the comments is what we call "the API".

You should complete the description of RTE_ETH_EVENT_INTR_RESET
as well: the need for calling rte_eth_dev_reset() was not explicit.

Re: [dpdk-dev] [PATCH V15 1/3] ethdev: introduce FEC API

[Andrew Rybchenko]

On 9/29/20 4:03 AM, Min Hu (Connor) wrote:
> This patch adds Forward error correction(FEC) support for ethdev.
> Introduce APIs which support query and config FEC information in
> hardware.
> 
> Signed-off-by: Min Hu (Connor) 
> Reviewed-by: Wei Hu (Xavier) 
> Reviewed-by: Chengwen Feng 
> Reviewed-by: Chengchang Tang 
> Acked-by: Andrew Rybchenko 

I confirm that it LGTM. However, I'd be happy if soneone
else reread descriptions with fresh look. Thanks.

Re: [dpdk-dev] [PATCH] net/octeontx2: remove logically dead code

[wangyunjian]

> -Original Message-
> From: Jerin Jacob [mailto:jerinjac...@gmail.com]
> Sent: Wednesday, September 30, 2020 2:15 AM
> To: wangyunjian 
> Cc: dpdk-dev ; Jerin Jacob ; Nithin
> Dabilpuram ; Kiran Kumar K
> ; Lilijun (Jerry) ;
> xudingke ; dpdk stable 
> Subject: Re: [dpdk-dev] [PATCH] net/octeontx2: remove logically dead code
> 
> On Wed, Aug 26, 2020 at 4:48 PM wangyunjian 
> wrote:
> >
> > From: Yunjian Wang 
> >
> > Coverity issue: 357719
> > Fixes: da138cd47e06 ("net/octeontx2: handle port reconfigure")
> > Cc: sta...@dpdk.org
> >
> > Signed-off-by: Yunjian Wang 
> > ---
> >  drivers/net/octeontx2/otx2_ethdev.c | 2 --
> >  1 file changed, 2 deletions(-)
> >
> > diff --git a/drivers/net/octeontx2/otx2_ethdev.c
> > b/drivers/net/octeontx2/otx2_ethdev.c
> > index 33b72bd4d..3f9399cc8 100644
> > --- a/drivers/net/octeontx2/otx2_ethdev.c
> > +++ b/drivers/net/octeontx2/otx2_ethdev.c
> > @@ -1355,8 +1355,6 @@ nix_store_queue_cfg_and_then_release(struct
> > rte_eth_dev *eth_dev)
> >  fail:
> 
> See below
> 
> > if (tx_qconf)
> > free(tx_qconf);
> > -   if (rx_qconf)
> > -   free(rx_qconf);
> 
> I think, it is clean and maintainable code have free() if rx_qconf as if we 
> add
> some another exit error case in the future, we simply forget to add this check
> and it will fail. So I prefer to keep as-is for the sake of maintainability 
> as there is
> no harm.

Hi, Jerin

Thanks for your explanation.
According to Stephen's suggestion, is it need to remove unnecessary NULL check?

Thanks,
Yunjian

> 
> 
> >
> > return -ENOMEM;
> >  }
> > --
> > 2.23.0
> >
> >

[dpdk-dev] [PATCH v8 0/8] fib: implement AVX512 vector lookup

[Vladimir Medvedkin]

This patch series implements vectorized lookup using AVX512 for
ipv4 dir24_8 and ipv6 trie algorithms.
Also introduced rte_fib_set_lookup_fn() to change lookup function type.
Added option to select lookup function type in testfib application.

v8:
 - remove Makefile related changes
 - fix missing doxygen for lookup_type
 - add release notes

v7:
 - fix RTE_X86_ZMM_MASK macro

v6:
 - style fixes

v5:
 - prefix zmm macro in rte_vect.h with RTE_X86
 - remove unnecessary typedef for _x86_zmm_t
 - reword commit title
 - fix typos

v4:
 - use __rte_aligned() instead of using compiler attribute directly
 - rework and add comments to meson.build

v3:
 - separate out the AVX-512 code into a separate file

v2:
 - rename rte_zmm to __rte_x86_zmm to reflect its internal usage
 - make runtime decision to use avx512 lookup

Vladimir Medvedkin (8):
  eal/x86: introduce AVX 512-bit type
  fib: make lookup function type configurable
  fib: move lookup definition into the header file
  fib: introduce AVX512 lookup
  fib6: make lookup function type configurable
  fib6: move lookup definition into the header file
  fib6: introduce AVX512 lookup
  app/testfib: add support for different lookup functions

 app/test-fib/main.c|  58 ++-
 doc/guides/rel_notes/release_20_11.rst |   3 +
 lib/librte_eal/x86/include/rte_vect.h  |  19 +++
 lib/librte_fib/dir24_8.c   | 281 ++---
 lib/librte_fib/dir24_8.h   | 226 +-
 lib/librte_fib/dir24_8_avx512.c| 165 +++
 lib/librte_fib/dir24_8_avx512.h|  24 +++
 lib/librte_fib/meson.build |  31 
 lib/librte_fib/rte_fib.c   |  21 ++-
 lib/librte_fib/rte_fib.h   |  25 +++
 lib/librte_fib/rte_fib6.c  |  20 ++-
 lib/librte_fib/rte_fib6.h  |  24 +++
 lib/librte_fib/rte_fib_version.map |   2 +
 lib/librte_fib/trie.c  | 161 ---
 lib/librte_fib/trie.h  | 119 +-
 lib/librte_fib/trie_avx512.c   | 269 +++
 lib/librte_fib/trie_avx512.h   |  20 +++
 17 files changed, 1096 insertions(+), 372 deletions(-)
 create mode 100644 lib/librte_fib/dir24_8_avx512.c
 create mode 100644 lib/librte_fib/dir24_8_avx512.h
 create mode 100644 lib/librte_fib/trie_avx512.c
 create mode 100644 lib/librte_fib/trie_avx512.h

-- 
2.7.4

[dpdk-dev] [PATCH v8 2/8] fib: make lookup function type configurable

[Vladimir Medvedkin]

Add type argument to dir24_8_get_lookup_fn()
Now it supports 3 different lookup implementations:
 RTE_FIB_DIR24_8_SCALAR_MACRO
 RTE_FIB_DIR24_8_SCALAR_INLINE
 RTE_FIB_DIR24_8_SCALAR_UNI

Add new rte_fib_set_lookup_fn() - user can change lookup
function type runtime.

Signed-off-by: Vladimir Medvedkin 
---
 lib/librte_fib/dir24_8.c   | 32 
 lib/librte_fib/dir24_8.h   |  2 +-
 lib/librte_fib/rte_fib.c   | 21 -
 lib/librte_fib/rte_fib.h   | 24 
 lib/librte_fib/rte_fib_version.map |  1 +
 5 files changed, 66 insertions(+), 14 deletions(-)

diff --git a/lib/librte_fib/dir24_8.c b/lib/librte_fib/dir24_8.c
index c9dce3c..825d061 100644
--- a/lib/librte_fib/dir24_8.c
+++ b/lib/librte_fib/dir24_8.c
@@ -45,13 +45,6 @@ struct dir24_8_tbl {
 
 #define ROUNDUP(x, y)   RTE_ALIGN_CEIL(x, (1 << (32 - y)))
 
-enum lookup_type {
-   MACRO,
-   INLINE,
-   UNI
-};
-enum lookup_type test_lookup = MACRO;
-
 static inline void *
 get_tbl24_p(struct dir24_8_tbl *dp, uint32_t ip, uint8_t nh_sz)
 {
@@ -253,11 +246,18 @@ dir24_8_lookup_bulk_uni(void *p, const uint32_t *ips,
 }
 
 rte_fib_lookup_fn_t
-dir24_8_get_lookup_fn(struct rte_fib_conf *fib_conf)
+dir24_8_get_lookup_fn(void *p, enum rte_fib_dir24_8_lookup_type type)
 {
-   enum rte_fib_dir24_8_nh_sz nh_sz = fib_conf->dir24_8.nh_sz;
+   enum rte_fib_dir24_8_nh_sz nh_sz;
+   struct dir24_8_tbl *dp = p;
 
-   if (test_lookup == MACRO) {
+   if (dp == NULL)
+   return NULL;
+
+   nh_sz = dp->nh_sz;
+
+   switch (type) {
+   case RTE_FIB_DIR24_8_SCALAR_MACRO:
switch (nh_sz) {
case RTE_FIB_DIR24_8_1B:
return dir24_8_lookup_bulk_1b;
@@ -267,8 +267,10 @@ dir24_8_get_lookup_fn(struct rte_fib_conf *fib_conf)
return dir24_8_lookup_bulk_4b;
case RTE_FIB_DIR24_8_8B:
return dir24_8_lookup_bulk_8b;
+   default:
+   return NULL;
}
-   } else if (test_lookup == INLINE) {
+   case RTE_FIB_DIR24_8_SCALAR_INLINE:
switch (nh_sz) {
case RTE_FIB_DIR24_8_1B:
return dir24_8_lookup_bulk_0;
@@ -278,9 +280,15 @@ dir24_8_get_lookup_fn(struct rte_fib_conf *fib_conf)
return dir24_8_lookup_bulk_2;
case RTE_FIB_DIR24_8_8B:
return dir24_8_lookup_bulk_3;
+   default:
+   return NULL;
}
-   } else
+   case RTE_FIB_DIR24_8_SCALAR_UNI:
return dir24_8_lookup_bulk_uni;
+   default:
+   return NULL;
+   }
+
return NULL;
 }
 
diff --git a/lib/librte_fib/dir24_8.h b/lib/librte_fib/dir24_8.h
index 1ec437c..53c5dd2 100644
--- a/lib/librte_fib/dir24_8.h
+++ b/lib/librte_fib/dir24_8.h
@@ -22,7 +22,7 @@ void
 dir24_8_free(void *p);
 
 rte_fib_lookup_fn_t
-dir24_8_get_lookup_fn(struct rte_fib_conf *conf);
+dir24_8_get_lookup_fn(void *p, enum rte_fib_dir24_8_lookup_type type);
 
 int
 dir24_8_modify(struct rte_fib *fib, uint32_t ip, uint8_t depth,
diff --git a/lib/librte_fib/rte_fib.c b/lib/librte_fib/rte_fib.c
index e090808..b9f6efb 100644
--- a/lib/librte_fib/rte_fib.c
+++ b/lib/librte_fib/rte_fib.c
@@ -107,7 +107,8 @@ init_dataplane(struct rte_fib *fib, __rte_unused int 
socket_id,
fib->dp = dir24_8_create(dp_name, socket_id, conf);
if (fib->dp == NULL)
return -rte_errno;
-   fib->lookup = dir24_8_get_lookup_fn(conf);
+   fib->lookup = dir24_8_get_lookup_fn(fib->dp,
+   RTE_FIB_DIR24_8_SCALAR_MACRO);
fib->modify = dir24_8_modify;
return 0;
default:
@@ -317,3 +318,21 @@ rte_fib_get_rib(struct rte_fib *fib)
 {
return (fib == NULL) ? NULL : fib->rib;
 }
+
+int
+rte_fib_set_lookup_fn(struct rte_fib *fib,
+   enum rte_fib_dir24_8_lookup_type type)
+{
+   rte_fib_lookup_fn_t fn;
+
+   switch (fib->type) {
+   case RTE_FIB_DIR24_8:
+   fn = dir24_8_get_lookup_fn(fib->dp, type);
+   if (fn == NULL)
+   return -EINVAL;
+   fib->lookup = fn;
+   return 0;
+   default:
+   return -EINVAL;
+   }
+}
diff --git a/lib/librte_fib/rte_fib.h b/lib/librte_fib/rte_fib.h
index 84ee774..a9bd0da 100644
--- a/lib/librte_fib/rte_fib.h
+++ b/lib/librte_fib/rte_fib.h
@@ -58,6 +58,13 @@ enum rte_fib_dir24_8_nh_sz {
RTE_FIB_DIR24_8_8B
 };
 
+/** Type of lookup function implementation */
+enum rte_fib_dir24_8_lookup_type {
+   RTE_FIB_DIR24_8_SCALAR_MACRO,
+   RTE_FIB_DIR24_8_SCALAR_INLINE,
+   RTE_FIB_DIR24_8_SCALAR_UNI
+};
+
 /** FIB configuration structure */
 struct rte_fib_conf {
enum rte_fib_type type; /**< Type of FIB st

[dpdk-dev] [PATCH v8 1/8] eal/x86: introduce AVX 512-bit type

[Vladimir Medvedkin]

New data type to manipulate 512 bit AVX values.

Signed-off-by: Vladimir Medvedkin 
Acked-by: Konstantin Ananyev 
---
 lib/librte_eal/x86/include/rte_vect.h | 19 +++
 1 file changed, 19 insertions(+)

diff --git a/lib/librte_eal/x86/include/rte_vect.h 
b/lib/librte_eal/x86/include/rte_vect.h
index df5a607..64383c3 100644
--- a/lib/librte_eal/x86/include/rte_vect.h
+++ b/lib/librte_eal/x86/include/rte_vect.h
@@ -13,6 +13,7 @@
 
 #include 
 #include 
+#include 
 #include "generic/rte_vect.h"
 
 #if (defined(__ICC) || \
@@ -90,6 +91,24 @@ __extension__ ({ \
 })
 #endif /* (defined(__ICC) && __ICC < 1210) */
 
+#ifdef __AVX512F__
+
+#define RTE_X86_ZMM_SIZE   (sizeof(__m512i))
+#define RTE_X86_ZMM_MASK   (RTE_X86_ZMM_SIZE - 1)
+
+typedef union __rte_x86_zmm {
+   __m512i  z;
+   ymm_ty[RTE_X86_ZMM_SIZE / sizeof(ymm_t)];
+   xmm_tx[RTE_X86_ZMM_SIZE / sizeof(xmm_t)];
+   uint8_t  u8[RTE_X86_ZMM_SIZE / sizeof(uint8_t)];
+   uint16_t u16[RTE_X86_ZMM_SIZE / sizeof(uint16_t)];
+   uint32_t u32[RTE_X86_ZMM_SIZE / sizeof(uint32_t)];
+   uint64_t u64[RTE_X86_ZMM_SIZE / sizeof(uint64_t)];
+   double   pd[RTE_X86_ZMM_SIZE / sizeof(double)];
+} __rte_aligned(RTE_X86_ZMM_SIZE) __rte_x86_zmm_t;
+
+#endif /* __AVX512F__ */
+
 #ifdef __cplusplus
 }
 #endif
-- 
2.7.4

[dpdk-dev] [PATCH v8 3/8] fib: move lookup definition into the header file

[Vladimir Medvedkin]

Move dir24_8 table layout and lookup definition into the
private header file. This is necessary for implementing a
vectorized lookup function in a separate .с file.

Signed-off-by: Vladimir Medvedkin 
Acked-by: Konstantin Ananyev 
---
 lib/librte_fib/dir24_8.c | 225 +--
 lib/librte_fib/dir24_8.h | 224 ++
 2 files changed, 225 insertions(+), 224 deletions(-)

diff --git a/lib/librte_fib/dir24_8.c b/lib/librte_fib/dir24_8.c
index 825d061..9d74653 100644
--- a/lib/librte_fib/dir24_8.c
+++ b/lib/librte_fib/dir24_8.c
@@ -11,240 +11,17 @@
 
 #include 
 #include 
-#include 
 #include 
 #include 
-#include 
 
-#include 
 #include 
+#include 
 #include "dir24_8.h"
 
 #define DIR24_8_NAMESIZE   64
 
-#define DIR24_8_TBL24_NUM_ENT  (1 << 24)
-#define DIR24_8_TBL8_GRP_NUM_ENT   256U
-#define DIR24_8_EXT_ENT1
-#define DIR24_8_TBL24_MASK 0xff00
-
-#define BITMAP_SLAB_BIT_SIZE_LOG2  6
-#define BITMAP_SLAB_BIT_SIZE   (1 << BITMAP_SLAB_BIT_SIZE_LOG2)
-#define BITMAP_SLAB_BITMASK(BITMAP_SLAB_BIT_SIZE - 1)
-
-struct dir24_8_tbl {
-   uint32_tnumber_tbl8s;   /**< Total number of tbl8s */
-   uint32_trsvd_tbl8s; /**< Number of reserved tbl8s */
-   uint32_tcur_tbl8s;  /**< Current number of tbl8s */
-   enum rte_fib_dir24_8_nh_sz  nh_sz;  /**< Size of nexthop entry */
-   uint64_tdef_nh; /**< Default next hop */
-   uint64_t*tbl8;  /**< tbl8 table. */
-   uint64_t*tbl8_idxes;/**< bitmap containing free tbl8 idxes*/
-   /* tbl24 table. */
-   __extension__ uint64_t  tbl24[0] __rte_cache_aligned;
-};
-
 #define ROUNDUP(x, y)   RTE_ALIGN_CEIL(x, (1 << (32 - y)))
 
-static inline void *
-get_tbl24_p(struct dir24_8_tbl *dp, uint32_t ip, uint8_t nh_sz)
-{
-   return (void *)&((uint8_t *)dp->tbl24)[(ip &
-   DIR24_8_TBL24_MASK) >> (8 - nh_sz)];
-}
-
-static inline  uint8_t
-bits_in_nh(uint8_t nh_sz)
-{
-   return 8 * (1 << nh_sz);
-}
-
-static inline uint64_t
-get_max_nh(uint8_t nh_sz)
-{
-   return ((1ULL << (bits_in_nh(nh_sz) - 1)) - 1);
-}
-
-static  inline uint32_t
-get_tbl24_idx(uint32_t ip)
-{
-   return ip >> 8;
-}
-
-static  inline uint32_t
-get_tbl8_idx(uint32_t res, uint32_t ip)
-{
-   return (res >> 1) * DIR24_8_TBL8_GRP_NUM_ENT + (uint8_t)ip;
-}
-
-static inline uint64_t
-lookup_msk(uint8_t nh_sz)
-{
-   return ((1ULL << ((1 << (nh_sz + 3)) - 1)) << 1) - 1;
-}
-
-static inline uint8_t
-get_psd_idx(uint32_t val, uint8_t nh_sz)
-{
-   return val & ((1 << (3 - nh_sz)) - 1);
-}
-
-static inline uint32_t
-get_tbl_idx(uint32_t val, uint8_t nh_sz)
-{
-   return val >> (3 - nh_sz);
-}
-
-static inline uint64_t
-get_tbl24(struct dir24_8_tbl *dp, uint32_t ip, uint8_t nh_sz)
-{
-   return ((dp->tbl24[get_tbl_idx(get_tbl24_idx(ip), nh_sz)] >>
-   (get_psd_idx(get_tbl24_idx(ip), nh_sz) *
-   bits_in_nh(nh_sz))) & lookup_msk(nh_sz));
-}
-
-static inline uint64_t
-get_tbl8(struct dir24_8_tbl *dp, uint32_t res, uint32_t ip, uint8_t nh_sz)
-{
-   return ((dp->tbl8[get_tbl_idx(get_tbl8_idx(res, ip), nh_sz)] >>
-   (get_psd_idx(get_tbl8_idx(res, ip), nh_sz) *
-   bits_in_nh(nh_sz))) & lookup_msk(nh_sz));
-}
-
-static inline int
-is_entry_extended(uint64_t ent)
-{
-   return (ent & DIR24_8_EXT_ENT) == DIR24_8_EXT_ENT;
-}
-
-#define LOOKUP_FUNC(suffix, type, bulk_prefetch, nh_sz)
\
-static void dir24_8_lookup_bulk_##suffix(void *p, const uint32_t *ips, \
-   uint64_t *next_hops, const unsigned int n)  \
-{  \
-   struct dir24_8_tbl *dp = (struct dir24_8_tbl *)p;   \
-   uint64_t tmp;   \
-   uint32_t i; \
-   uint32_t prefetch_offset =  \
-   RTE_MIN((unsigned int)bulk_prefetch, n);\
-   \
-   for (i = 0; i < prefetch_offset; i++)   \
-   rte_prefetch0(get_tbl24_p(dp, ips[i], nh_sz));  \
-   for (i = 0; i < (n - prefetch_offset); i++) {   \
-   rte_prefetch0(get_tbl24_p(dp,   \
-   ips[i + prefetch_offset], nh_sz));  \
-   tmp = ((type *)dp->tbl24)[ips[i] >> 8]; \
-   if (unlikely(is_entry_extended(tmp)))   \
-   tmp = ((type *)dp->tbl8)[(uint8_t)ips[i] +  \
-   ((tmp >> 1) * DIR24_8_TBL8_GRP_NUM_ENT)]; \
-   next_hops[i] = tmp >> 1; 

[dpdk-dev] [PATCH v8 4/8] fib: introduce AVX512 lookup

[Vladimir Medvedkin]

Add new lookup implementation for DIR24_8 algorithm using
AVX512 instruction set

Signed-off-by: Vladimir Medvedkin 
Acked-by: Konstantin Ananyev 
---
 doc/guides/rel_notes/release_20_11.rst |   3 +
 lib/librte_fib/dir24_8.c   |  24 +
 lib/librte_fib/dir24_8_avx512.c| 165 +
 lib/librte_fib/dir24_8_avx512.h|  24 +
 lib/librte_fib/meson.build |  18 
 lib/librte_fib/rte_fib.h   |   3 +-
 6 files changed, 236 insertions(+), 1 deletion(-)
 create mode 100644 lib/librte_fib/dir24_8_avx512.c
 create mode 100644 lib/librte_fib/dir24_8_avx512.h

diff --git a/doc/guides/rel_notes/release_20_11.rst 
b/doc/guides/rel_notes/release_20_11.rst
index 4eb3224..26a7d8e 100644
--- a/doc/guides/rel_notes/release_20_11.rst
+++ b/doc/guides/rel_notes/release_20_11.rst
@@ -78,6 +78,9 @@ New Features
 ``--portmask=N``
 where N represents the hexadecimal bitmask of ports used.
 
+* **Added AVX512 lookup implementation for FIB.**
+
+  Added a AVX512 lookup functions implementation into FIB library.
 
 Removed Items
 -
diff --git a/lib/librte_fib/dir24_8.c b/lib/librte_fib/dir24_8.c
index 9d74653..0d7bf2c 100644
--- a/lib/librte_fib/dir24_8.c
+++ b/lib/librte_fib/dir24_8.c
@@ -18,6 +18,12 @@
 #include 
 #include "dir24_8.h"
 
+#ifdef CC_DIR24_8_AVX512_SUPPORT
+
+#include "dir24_8_avx512.h"
+
+#endif /* CC_DIR24_8_AVX512_SUPPORT */
+
 #define DIR24_8_NAMESIZE   64
 
 #define ROUNDUP(x, y)   RTE_ALIGN_CEIL(x, (1 << (32 - y)))
@@ -62,6 +68,24 @@ dir24_8_get_lookup_fn(void *p, enum 
rte_fib_dir24_8_lookup_type type)
}
case RTE_FIB_DIR24_8_SCALAR_UNI:
return dir24_8_lookup_bulk_uni;
+#ifdef CC_DIR24_8_AVX512_SUPPORT
+   case RTE_FIB_DIR24_8_VECTOR_AVX512:
+   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) <= 0)
+   return NULL;
+
+   switch (nh_sz) {
+   case RTE_FIB_DIR24_8_1B:
+   return rte_dir24_8_vec_lookup_bulk_1b;
+   case RTE_FIB_DIR24_8_2B:
+   return rte_dir24_8_vec_lookup_bulk_2b;
+   case RTE_FIB_DIR24_8_4B:
+   return rte_dir24_8_vec_lookup_bulk_4b;
+   case RTE_FIB_DIR24_8_8B:
+   return rte_dir24_8_vec_lookup_bulk_8b;
+   default:
+   return NULL;
+   }
+#endif
default:
return NULL;
}
diff --git a/lib/librte_fib/dir24_8_avx512.c b/lib/librte_fib/dir24_8_avx512.c
new file mode 100644
index 000..43dba28
--- /dev/null
+++ b/lib/librte_fib/dir24_8_avx512.c
@@ -0,0 +1,165 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include 
+#include 
+
+#include "dir24_8.h"
+#include "dir24_8_avx512.h"
+
+static __rte_always_inline void
+dir24_8_vec_lookup_x16(void *p, const uint32_t *ips,
+   uint64_t *next_hops, int size)
+{
+   struct dir24_8_tbl *dp = (struct dir24_8_tbl *)p;
+   __mmask16 msk_ext;
+   __mmask16 exp_msk = 0x;
+   __m512i ip_vec, idxes, res, bytes;
+   const __m512i zero = _mm512_set1_epi32(0);
+   const __m512i lsb = _mm512_set1_epi32(1);
+   const __m512i lsbyte_msk = _mm512_set1_epi32(0xff);
+   __m512i tmp1, tmp2, res_msk;
+   __m256i tmp256;
+   /* used to mask gather values if size is 1/2 (8/16 bit next hops) */
+   if (size == sizeof(uint8_t))
+   res_msk = _mm512_set1_epi32(UINT8_MAX);
+   else if (size == sizeof(uint16_t))
+   res_msk = _mm512_set1_epi32(UINT16_MAX);
+
+   ip_vec = _mm512_loadu_si512(ips);
+   /* mask 24 most significant bits */
+   idxes = _mm512_srli_epi32(ip_vec, 8);
+
+   /**
+* lookup in tbl24
+* Put it inside branch to make compiler happy with -O0
+*/
+   if (size == sizeof(uint8_t)) {
+   res = _mm512_i32gather_epi32(idxes, (const int *)dp->tbl24, 1);
+   res = _mm512_and_epi32(res, res_msk);
+   } else if (size == sizeof(uint16_t)) {
+   res = _mm512_i32gather_epi32(idxes, (const int *)dp->tbl24, 2);
+   res = _mm512_and_epi32(res, res_msk);
+   } else
+   res = _mm512_i32gather_epi32(idxes, (const int *)dp->tbl24, 4);
+
+   /* get extended entries indexes */
+   msk_ext = _mm512_test_epi32_mask(res, lsb);
+
+   if (msk_ext != 0) {
+   idxes = _mm512_srli_epi32(res, 1);
+   idxes = _mm512_slli_epi32(idxes, 8);
+   bytes = _mm512_and_epi32(ip_vec, lsbyte_msk);
+   idxes = _mm512_maskz_add_epi32(msk_ext, idxes, bytes);
+   if (size == sizeof(uint8_t)) {
+   idxes = _mm512_mask_i32gather_epi32(zero, msk_ext,
+   idxes, (const int *)dp->tbl8, 1);
+   idxes = _mm512_and_epi32(idxes, res_msk);
+   }

[dpdk-dev] [PATCH v8 5/8] fib6: make lookup function type configurable

[Vladimir Medvedkin]

Add type argument to trie_get_lookup_fn()
Now it only supports RTE_FIB6_TRIE_SCALAR

Add new rte_fib6_set_lookup_fn() - user can change lookup
function type runtime.

Signed-off-by: Vladimir Medvedkin 
---
 lib/librte_fib/rte_fib6.c  | 20 +++-
 lib/librte_fib/rte_fib6.h  | 23 +++
 lib/librte_fib/rte_fib_version.map |  1 +
 lib/librte_fib/trie.c  | 25 ++---
 lib/librte_fib/trie.h  |  2 +-
 5 files changed, 58 insertions(+), 13 deletions(-)

diff --git a/lib/librte_fib/rte_fib6.c b/lib/librte_fib/rte_fib6.c
index a1f0db8..566cd5f 100644
--- a/lib/librte_fib/rte_fib6.c
+++ b/lib/librte_fib/rte_fib6.c
@@ -107,7 +107,7 @@ init_dataplane(struct rte_fib6 *fib, __rte_unused int 
socket_id,
fib->dp = trie_create(dp_name, socket_id, conf);
if (fib->dp == NULL)
return -rte_errno;
-   fib->lookup = rte_trie_get_lookup_fn(conf);
+   fib->lookup = trie_get_lookup_fn(fib->dp, RTE_FIB6_TRIE_SCALAR);
fib->modify = trie_modify;
return 0;
default:
@@ -319,3 +319,21 @@ rte_fib6_get_rib(struct rte_fib6 *fib)
 {
return (fib == NULL) ? NULL : fib->rib;
 }
+
+int
+rte_fib6_set_lookup_fn(struct rte_fib6 *fib,
+   enum rte_fib_trie_lookup_type type)
+{
+   rte_fib6_lookup_fn_t fn;
+
+   switch (fib->type) {
+   case RTE_FIB6_TRIE:
+   fn = trie_get_lookup_fn(fib->dp, type);
+   if (fn == NULL)
+   return -EINVAL;
+   fib->lookup = fn;
+   return 0;
+   default:
+   return -EINVAL;
+   }
+}
diff --git a/lib/librte_fib/rte_fib6.h b/lib/librte_fib/rte_fib6.h
index bbfcf23..cc817ad 100644
--- a/lib/librte_fib/rte_fib6.h
+++ b/lib/librte_fib/rte_fib6.h
@@ -53,12 +53,18 @@ enum rte_fib6_op {
RTE_FIB6_DEL,
 };
 
+/** Size of nexthop (1 << nh_sz) bits for TRIE based FIB */
 enum rte_fib_trie_nh_sz {
RTE_FIB6_TRIE_2B = 1,
RTE_FIB6_TRIE_4B,
RTE_FIB6_TRIE_8B
 };
 
+/** Type of lookup function implementation */
+enum rte_fib_trie_lookup_type {
+   RTE_FIB6_TRIE_SCALAR
+};
+
 /** FIB configuration structure */
 struct rte_fib6_conf {
enum rte_fib6_type type; /**< Type of FIB struct */
@@ -201,6 +207,23 @@ __rte_experimental
 struct rte_rib6 *
 rte_fib6_get_rib(struct rte_fib6 *fib);
 
+/**
+ * Set lookup function based on type
+ *
+ * @param fib
+ *   FIB object handle
+ * @param type
+ *   type of lookup function
+ *
+ * @return
+ *-EINVAL on failure
+ *0 on success
+ */
+__rte_experimental
+int
+rte_fib6_set_lookup_fn(struct rte_fib6 *fib,
+   enum rte_fib_trie_lookup_type type);
+
 #ifdef __cplusplus
 }
 #endif
diff --git a/lib/librte_fib/rte_fib_version.map 
b/lib/librte_fib/rte_fib_version.map
index 216af66..9d1e181 100644
--- a/lib/librte_fib/rte_fib_version.map
+++ b/lib/librte_fib/rte_fib_version.map
@@ -19,6 +19,7 @@ EXPERIMENTAL {
rte_fib6_lookup_bulk;
rte_fib6_get_dp;
rte_fib6_get_rib;
+   rte_fib6_set_lookup_fn;
 
local: *;
 };
diff --git a/lib/librte_fib/trie.c b/lib/librte_fib/trie.c
index 2ae2add..63c519a 100644
--- a/lib/librte_fib/trie.c
+++ b/lib/librte_fib/trie.c
@@ -59,13 +59,6 @@ enum edge {
REDGE
 };
 
-enum lookup_type {
-   MACRO,
-   INLINE,
-   UNI
-};
-static enum lookup_type test_lookup = MACRO;
-
 static inline uint32_t
 get_tbl24_idx(const uint8_t *ip)
 {
@@ -154,11 +147,18 @@ LOOKUP_FUNC(4b, uint32_t, 2)
 LOOKUP_FUNC(8b, uint64_t, 3)
 
 rte_fib6_lookup_fn_t
-rte_trie_get_lookup_fn(struct rte_fib6_conf *conf)
+trie_get_lookup_fn(void *p, enum rte_fib_trie_lookup_type type)
 {
-   enum rte_fib_trie_nh_sz nh_sz = conf->trie.nh_sz;
+   enum rte_fib_trie_nh_sz nh_sz;
+   struct rte_trie_tbl *dp = p;
 
-   if (test_lookup == MACRO) {
+   if (dp == NULL)
+   return NULL;
+
+   nh_sz = dp->nh_sz;
+
+   switch (type) {
+   case RTE_FIB6_TRIE_SCALAR:
switch (nh_sz) {
case RTE_FIB6_TRIE_2B:
return rte_trie_lookup_bulk_2b;
@@ -166,9 +166,12 @@ rte_trie_get_lookup_fn(struct rte_fib6_conf *conf)
return rte_trie_lookup_bulk_4b;
case RTE_FIB6_TRIE_8B:
return rte_trie_lookup_bulk_8b;
+   default:
+   return NULL;
}
+   default:
+   return NULL;
}
-
return NULL;
 }
 
diff --git a/lib/librte_fib/trie.h b/lib/librte_fib/trie.h
index bb750c5..0d5ef9a 100644
--- a/lib/librte_fib/trie.h
+++ b/lib/librte_fib/trie.h
@@ -22,7 +22,7 @@ void
 trie_free(void *p);
 
 rte_fib6_lookup_fn_t
-rte_trie_get_lookup_fn(struct rte_fib6_conf *fib_conf);
+trie_get_lookup_fn(void *p, enum rte_fib_trie_lookup_type type);
 
 int
 trie_modify(struct rte_fib6 *fib, const uint8_t ip[RTE_FI

[dpdk-dev] [PATCH v8 7/8] fib6: introduce AVX512 lookup

[Vladimir Medvedkin]

Add new lookup implementation for FIB6 trie algorithm using
AVX512 instruction set

Signed-off-by: Vladimir Medvedkin 
Acked-by: Konstantin Ananyev 
---
 doc/guides/rel_notes/release_20_11.rst |   2 +-
 lib/librte_fib/meson.build |  13 ++
 lib/librte_fib/rte_fib6.h  |   3 +-
 lib/librte_fib/trie.c  |  21 +++
 lib/librte_fib/trie_avx512.c   | 269 +
 lib/librte_fib/trie_avx512.h   |  20 +++
 6 files changed, 326 insertions(+), 2 deletions(-)
 create mode 100644 lib/librte_fib/trie_avx512.c
 create mode 100644 lib/librte_fib/trie_avx512.h

diff --git a/doc/guides/rel_notes/release_20_11.rst 
b/doc/guides/rel_notes/release_20_11.rst
index 26a7d8e..cafd499 100644
--- a/doc/guides/rel_notes/release_20_11.rst
+++ b/doc/guides/rel_notes/release_20_11.rst
@@ -80,7 +80,7 @@ New Features
 
 * **Added AVX512 lookup implementation for FIB.**
 
-  Added a AVX512 lookup functions implementation into FIB library.
+  Added a AVX512 lookup functions implementation into FIB and FIB6 libraries.
 
 Removed Items
 -
diff --git a/lib/librte_fib/meson.build b/lib/librte_fib/meson.build
index d96ff02..98c8752 100644
--- a/lib/librte_fib/meson.build
+++ b/lib/librte_fib/meson.build
@@ -13,6 +13,8 @@ if arch_subdir == 'x86' and not 
machine_args.contains('-mno-avx512f')
if dpdk_conf.has('RTE_MACHINE_CPUFLAG_AVX512F')
cflags += ['-DCC_DIR24_8_AVX512_SUPPORT']
sources += files('dir24_8_avx512.c')
+   cflags += ['-DCC_TRIE_AVX512_SUPPORT']
+   sources += files('trie_avx512.c')
elif cc.has_multi_arguments('-mavx512f', '-mavx512dq')
dir24_8_avx512_tmp = static_library('dir24_8_avx512_tmp',
'dir24_8_avx512.c',
@@ -20,6 +22,17 @@ if arch_subdir == 'x86' and not 
machine_args.contains('-mno-avx512f')
c_args: cflags + ['-mavx512f', '-mavx512dq'])
objs += dir24_8_avx512_tmp.extract_objects('dir24_8_avx512.c')
cflags += '-DCC_DIR24_8_AVX512_SUPPORT'
+   # TRIE AVX512 implementation uses avx512bw intrinsics along with
+   # avx512f and avx512dq
+   if cc.has_argument('-mavx512bw')
+   trie_avx512_tmp = static_library('trie_avx512_tmp',
+   'trie_avx512.c',
+   dependencies: static_rte_eal,
+   c_args: cflags + ['-mavx512f', \
+   '-mavx512dq', '-mavx512bw'])
+   objs += trie_avx512_tmp.extract_objects('trie_avx512.c')
+   cflags += '-DCC_TRIE_AVX512_SUPPORT'
+   endif
endif
 endif
 
diff --git a/lib/librte_fib/rte_fib6.h b/lib/librte_fib/rte_fib6.h
index cc817ad..f53b076 100644
--- a/lib/librte_fib/rte_fib6.h
+++ b/lib/librte_fib/rte_fib6.h
@@ -62,7 +62,8 @@ enum rte_fib_trie_nh_sz {
 
 /** Type of lookup function implementation */
 enum rte_fib_trie_lookup_type {
-   RTE_FIB6_TRIE_SCALAR
+   RTE_FIB6_TRIE_SCALAR,
+   RTE_FIB6_TRIE_VECTOR_AVX512
 };
 
 /** FIB configuration structure */
diff --git a/lib/librte_fib/trie.c b/lib/librte_fib/trie.c
index 136e938..d0233ad 100644
--- a/lib/librte_fib/trie.c
+++ b/lib/librte_fib/trie.c
@@ -18,6 +18,12 @@
 #include 
 #include "trie.h"
 
+#ifdef CC_TRIE_AVX512_SUPPORT
+
+#include "trie_avx512.h"
+
+#endif /* CC_TRIE_AVX512_SUPPORT */
+
 #define TRIE_NAMESIZE  64
 
 enum edge {
@@ -48,6 +54,21 @@ trie_get_lookup_fn(void *p, enum rte_fib_trie_lookup_type 
type)
default:
return NULL;
}
+#ifdef CC_TRIE_AVX512_SUPPORT
+   case RTE_FIB6_TRIE_VECTOR_AVX512:
+   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) <= 0)
+   return NULL;
+   switch (nh_sz) {
+   case RTE_FIB6_TRIE_2B:
+   return rte_trie_vec_lookup_bulk_2b;
+   case RTE_FIB6_TRIE_4B:
+   return rte_trie_vec_lookup_bulk_4b;
+   case RTE_FIB6_TRIE_8B:
+   return rte_trie_vec_lookup_bulk_8b;
+   default:
+   return NULL;
+   }
+#endif
default:
return NULL;
}
diff --git a/lib/librte_fib/trie_avx512.c b/lib/librte_fib/trie_avx512.c
new file mode 100644
index 000..b1c9e4e
--- /dev/null
+++ b/lib/librte_fib/trie_avx512.c
@@ -0,0 +1,269 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include 
+#include 
+
+#include "trie.h"
+#include "trie_avx512.h"
+
+static __rte_always_inline void
+transpose_x16(uint8_t ips[16][RTE_FIB6_IPV6_ADDR_SIZE],
+   __m512i *first, __m512i *second, __m512i *third, __m512i *fourth)
+{
+   __m512i tmp1, tmp2, tmp3, tmp4;
+   __m512i tmp5, tmp6, tmp7, tmp8;
+   const

[dpdk-dev] [PATCH v8 8/8] app/testfib: add support for different lookup functions

[Vladimir Medvedkin]

Added -v option to switch between different lookup implementations
to measure their performance and correctness.

Signed-off-by: Vladimir Medvedkin 
---
 app/test-fib/main.c | 58 ++---
 1 file changed, 55 insertions(+), 3 deletions(-)

diff --git a/app/test-fib/main.c b/app/test-fib/main.c
index 9cf01b1..9c2d413 100644
--- a/app/test-fib/main.c
+++ b/app/test-fib/main.c
@@ -99,6 +99,7 @@ static struct {
uint8_t ent_sz;
uint8_t rnd_lookup_ips_ratio;
uint8_t print_fract;
+   uint8_t lookup_fn;
 } config = {
.routes_file = NULL,
.lookup_ips_file = NULL,
@@ -110,7 +111,8 @@ static struct {
.tbl8 = DEFAULT_LPM_TBL8,
.ent_sz = 4,
.rnd_lookup_ips_ratio = 0,
-   .print_fract = 10
+   .print_fract = 10,
+   .lookup_fn = 0
 };
 
 struct rt_rule_4 {
@@ -638,7 +640,11 @@ print_usage(void)
"1/2/4/8 (default 4)>]\n"
"[-g ]\n"
"[-w ]\n"
-   "[-u ]\n",
+   "[-u ]\n"
+   "[-v ]\n",
config.prgname);
 }
 
@@ -681,7 +687,7 @@ parse_opts(int argc, char **argv)
int opt;
char *endptr;
 
-   while ((opt = getopt(argc, argv, "f:t:n:d:l:r:c6ab:e:g:w:u:s")) !=
+   while ((opt = getopt(argc, argv, "f:t:n:d:l:r:c6ab:e:g:w:u:sv:")) !=
-1) {
switch (opt) {
case 'f':
@@ -769,6 +775,22 @@ parse_opts(int argc, char **argv)
rte_exit(-EINVAL, "Invalid option -g\n");
}
break;
+   case 'v':
+   if ((strcmp(optarg, "s1") == 0) ||
+   (strcmp(optarg, "s") == 0))
+   break;
+   else if (strcmp(optarg, "v") == 0) {
+   config.lookup_fn = 1;
+   break;
+   } else if (strcmp(optarg, "s2") == 0) {
+   config.lookup_fn = 2;
+   break;
+   } else if (strcmp(optarg, "s3") == 0) {
+   config.lookup_fn = 3;
+   break;
+   }
+   print_usage();
+   rte_exit(-EINVAL, "Invalid option -v %s\n", optarg);
default:
print_usage();
rte_exit(-EINVAL, "Invalid options\n");
@@ -846,6 +868,24 @@ run_v4(void)
return -rte_errno;
}
 
+   if (config.lookup_fn != 0) {
+   if (config.lookup_fn == 1)
+   ret = rte_fib_set_lookup_fn(fib,
+   RTE_FIB_DIR24_8_VECTOR_AVX512);
+   else if (config.lookup_fn == 2)
+   ret = rte_fib_set_lookup_fn(fib,
+   RTE_FIB_DIR24_8_SCALAR_INLINE);
+   else if (config.lookup_fn == 3)
+   ret = rte_fib_set_lookup_fn(fib,
+   RTE_FIB_DIR24_8_SCALAR_UNI);
+   else
+   ret = -EINVAL;
+   if (ret != 0) {
+   printf("Can not init lookup function\n");
+   return ret;
+   }
+   }
+
for (k = config.print_fract, i = 0; k > 0; k--) {
start = rte_rdtsc_precise();
for (j = 0; j < (config.nb_routes - i) / k; j++) {
@@ -1025,6 +1065,18 @@ run_v6(void)
return -rte_errno;
}
 
+   if (config.lookup_fn != 0) {
+   if (config.lookup_fn == 1)
+   ret = rte_fib6_set_lookup_fn(fib,
+   RTE_FIB6_TRIE_VECTOR_AVX512);
+   else
+   ret = -EINVAL;
+   if (ret != 0) {
+   printf("Can not init lookup function\n");
+   return ret;
+   }
+   }
+
for (k = config.print_fract, i = 0; k > 0; k--) {
start = rte_rdtsc_precise();
for (j = 0; j < (config.nb_routes - i) / k; j++) {
-- 
2.7.4

[dpdk-dev] [PATCH v8 6/8] fib6: move lookup definition into the header file

[Vladimir Medvedkin]

Move trie table layout and lookup definition into the
private header file. This is necessary for implementing a
vectorized lookup function in a separate .с file.

Signed-off-by: Vladimir Medvedkin 
Acked-by: Konstantin Ananyev 
---
 lib/librte_fib/trie.c | 121 --
 lib/librte_fib/trie.h | 117 
 2 files changed, 117 insertions(+), 121 deletions(-)

diff --git a/lib/librte_fib/trie.c b/lib/librte_fib/trie.c
index 63c519a..136e938 100644
--- a/lib/librte_fib/trie.c
+++ b/lib/librte_fib/trie.c
@@ -11,141 +11,20 @@
 
 #include 
 #include 
-#include 
 #include 
 #include 
-#include 
 
 #include 
 #include 
 #include "trie.h"
 
-/* @internal Total number of tbl24 entries. */
-#define TRIE_TBL24_NUM_ENT (1 << 24)
-
-/* Maximum depth value possible for IPv6 LPM. */
-#define TRIE_MAX_DEPTH 128
-
-/* @internal Number of entries in a tbl8 group. */
-#define TRIE_TBL8_GRP_NUM_ENT  256ULL
-
-/* @internal Total number of tbl8 groups in the tbl8. */
-#define TRIE_TBL8_NUM_GROUPS   65536
-
-/* @internal bitmask with valid and valid_group fields set */
-#define TRIE_EXT_ENT   1
-
 #define TRIE_NAMESIZE  64
 
-#define BITMAP_SLAB_BIT_SIZE_LOG2  6
-#define BITMAP_SLAB_BIT_SIZE   (1ULL << BITMAP_SLAB_BIT_SIZE_LOG2)
-#define BITMAP_SLAB_BITMASK(BITMAP_SLAB_BIT_SIZE - 1)
-
-struct rte_trie_tbl {
-   uint32_tnumber_tbl8s;   /**< Total number of tbl8s */
-   uint32_trsvd_tbl8s; /**< Number of reserved tbl8s */
-   uint32_tcur_tbl8s;  /**< Current cumber of tbl8s */
-   uint64_tdef_nh; /**< Default next hop */
-   enum rte_fib_trie_nh_sz nh_sz;  /**< Size of nexthop entry */
-   uint64_t*tbl8;  /**< tbl8 table. */
-   uint32_t*tbl8_pool; /**< bitmap containing free tbl8 idxes*/
-   uint32_ttbl8_pool_pos;
-   /* tbl24 table. */
-   __extension__ uint64_t  tbl24[0] __rte_cache_aligned;
-};
-
 enum edge {
LEDGE,
REDGE
 };
 
-static inline uint32_t
-get_tbl24_idx(const uint8_t *ip)
-{
-   return ip[0] << 16|ip[1] << 8|ip[2];
-}
-
-static inline void *
-get_tbl24_p(struct rte_trie_tbl *dp, const uint8_t *ip, uint8_t nh_sz)
-{
-   uint32_t tbl24_idx;
-
-   tbl24_idx = get_tbl24_idx(ip);
-   return (void *)&((uint8_t *)dp->tbl24)[tbl24_idx << nh_sz];
-}
-
-static inline uint8_t
-bits_in_nh(uint8_t nh_sz)
-{
-   return 8 * (1 << nh_sz);
-}
-
-static inline uint64_t
-get_max_nh(uint8_t nh_sz)
-{
-   return ((1ULL << (bits_in_nh(nh_sz) - 1)) - 1);
-}
-
-static inline uint64_t
-lookup_msk(uint8_t nh_sz)
-{
-   return ((1ULL << ((1 << (nh_sz + 3)) - 1)) << 1) - 1;
-}
-
-static inline uint8_t
-get_psd_idx(uint32_t val, uint8_t nh_sz)
-{
-   return val & ((1 << (3 - nh_sz)) - 1);
-}
-
-static inline uint32_t
-get_tbl_pos(uint32_t val, uint8_t nh_sz)
-{
-   return val >> (3 - nh_sz);
-}
-
-static inline uint64_t
-get_tbl_val_by_idx(uint64_t *tbl, uint32_t idx, uint8_t nh_sz)
-{
-   return ((tbl[get_tbl_pos(idx, nh_sz)] >> (get_psd_idx(idx, nh_sz) *
-   bits_in_nh(nh_sz))) & lookup_msk(nh_sz));
-}
-
-static inline void *
-get_tbl_p_by_idx(uint64_t *tbl, uint64_t idx, uint8_t nh_sz)
-{
-   return (uint8_t *)tbl + (idx << nh_sz);
-}
-
-static inline int
-is_entry_extended(uint64_t ent)
-{
-   return (ent & TRIE_EXT_ENT) == TRIE_EXT_ENT;
-}
-
-#define LOOKUP_FUNC(suffix, type, nh_sz)   \
-static void rte_trie_lookup_bulk_##suffix(void *p, \
-   uint8_t ips[][RTE_FIB6_IPV6_ADDR_SIZE], \
-   uint64_t *next_hops, const unsigned int n)  \
-{  \
-   struct rte_trie_tbl *dp = (struct rte_trie_tbl *)p; \
-   uint64_t tmp;   \
-   uint32_t i, j;  \
-   \
-   for (i = 0; i < n; i++) {   \
-   tmp = ((type *)dp->tbl24)[get_tbl24_idx(&ips[i][0])];   \
-   j = 3;  \
-   while (is_entry_extended(tmp)) {\
-   tmp = ((type *)dp->tbl8)[ips[i][j++] +  \
-   ((tmp >> 1) * TRIE_TBL8_GRP_NUM_ENT)];  \
-   }   \
-   next_hops[i] = tmp >> 1;\
-   }   \
-}
-LOOKUP_FUNC(2b, uint16_t, 1)
-LOOKUP_FUNC(4b, uint32_t, 2)
-LOOKUP_FUNC(8b, uint64_t, 3)
-
 rte_fib6_lookup_fn_t
 trie_get_lookup_fn(void *p, enum rte_fib_trie_lookup_t

Re: [dpdk-dev] [PATCH 2/2] ethdev: make rte flow API thread safe

[Ori Kam]

Hi Suanming,

Small comments,
PSB,

Best,
Ori

> -Original Message-
> From: Suanming Mou 
> Sent: Sunday, September 27, 2020 11:20 AM
> Cc: dev@dpdk.org
> Subject: [PATCH 2/2] ethdev: make rte flow API thread safe
> 
> Currently, the rte flow functions are not defined as thread safety.

For a given port. (between ports they are thread safe)

> DPDK applications either call the functions in single thread or add
> locks around the functions for the critical section.
> 
> For PMDs support the flow operations thread safe natively, the
> redundant protection in application hurts the performance of the
> rte flow operation functions.
> 
> And the restriction of thread safety not guaranteed for the rte
> flow functions also limits the applications' expectation.
> 
> This feature is going to change the rte flow functions to be thread
> safety. As different PMDs have different flow operations, some may

Safety => safe I think this is true also to other places in this patch.

> support thread safety already and others may not. For PMDs don't
> support flow thread safe operation, a new lock is defined in ethdev
> in order to protects thread unsafe PMDs from rte flow level.
> 
> A new RTE_ETH_DEV_FLOW_OPS_THREAD_SAFE device flag is added to
> determine whether the PMD supports thread safe flow operation or not.
> For PMDs support thread safe flow operations, set the
> RTE_ETH_DEV_FLOW_OPS_THREAD_SAFE flag, rte flow level functions will
> skip the thread safe helper lock for these PMDs. Again the rte flow
> level thread safe lock only works when PMD operation functions are
> not thread safety.
safe
> 
> For the PMDs which don't want the default mutex lock, just set the
> flag in the PMD, and add the perfer type of lock in the PMD. Then
> the default mutex lock is easily replaced by the PMD level lock.
> 
> The change has no effect on the current DPDK applications. No change
> is required for the current DPDK applications. If no lock contention
> with the added rte flow level mutex, the mutex only does the atomic
> increasing in pthread_mutex_lock() and decreasing in
> pthread_mutex_unlock(). No futex() syscall will be involved.
> 
> Signed-off-by: Suanming Mou 
> ---
>  doc/guides/prog_guide/rte_flow.rst  |  7 ++--
>  drivers/net/mlx5/linux/mlx5_os.c|  2 +
>  lib/librte_ethdev/rte_ethdev.c  |  2 +
>  lib/librte_ethdev/rte_ethdev.h  |  2 +
>  lib/librte_ethdev/rte_ethdev_core.h |  4 ++
>  lib/librte_ethdev/rte_flow.c| 84 
> -
>  6 files changed, 78 insertions(+), 23 deletions(-)
> 
> diff --git a/doc/guides/prog_guide/rte_flow.rst
> b/doc/guides/prog_guide/rte_flow.rst
> index 119b128..6f7997a 100644
> --- a/doc/guides/prog_guide/rte_flow.rst
> +++ b/doc/guides/prog_guide/rte_flow.rst
> @@ -3046,10 +3046,6 @@ Caveats
>  - API operations are synchronous and blocking (``EAGAIN`` cannot be
>returned).
> 
> -- There is no provision for re-entrancy/multi-thread safety, although nothing
> -  should prevent different devices from being configured at the same
> -  time. PMDs may protect their control path functions accordingly.
> -
>  - Stopping the data path (TX/RX) should not be necessary when managing flow
>rules. If this cannot be achieved naturally or with workarounds (such as
>temporarily replacing the burst function pointers), an appropriate error
> @@ -3101,6 +3097,9 @@ This interface additionally defines the following
> helper function:
>  - ``rte_flow_ops_get()``: get generic flow operations structure from a
>port.
> 
> +If PMD interfaces do not support re-entrancy/multi-thread safety, rte flow
> +level functions will do it by mutex.
> +

I think you should add which flag the app should test to see if the PMD
support thread safe.

>  More will be added over time.
> 
>  Device compatibility
> diff --git a/drivers/net/mlx5/linux/mlx5_os.c
> b/drivers/net/mlx5/linux/mlx5_os.c
> index 2a4beb0..daefb7d 100644
> --- a/drivers/net/mlx5/linux/mlx5_os.c
> +++ b/drivers/net/mlx5/linux/mlx5_os.c
> @@ -1351,6 +1351,8 @@
>   goto error;
>   }
>   }
> + if (priv->config.dv_flow_en)
> + eth_dev->data->dev_flags =
> RTE_ETH_DEV_FLOW_OPS_THREAD_SAFE;
>   return eth_dev;

I don't think this code is relevant for this patch.
I also think you are missing the | when assigning the value.

>  error:
>   if (priv) {
> diff --git a/lib/librte_ethdev/rte_ethdev.c b/lib/librte_ethdev/rte_ethdev.c
> index dfe5c1b..d6fd17e 100644
> --- a/lib/librte_ethdev/rte_ethdev.c
> +++ b/lib/librte_ethdev/rte_ethdev.c
> @@ -500,6 +500,7 @@ struct rte_eth_dev *
>   strlcpy(eth_dev->data->name, name, sizeof(eth_dev->data->name));
>   eth_dev->data->port_id = port_id;
>   eth_dev->data->mtu = RTE_ETHER_MTU;
> + pthread_mutex_init(ð_dev->data->fts_mutex, NULL);
> 
>  unlock:
>   rte_spinlock_unlock(&rte_eth_dev_shared_data->ownership_lock);
> @@ -562,6 +563,7 @@ struct rte_eth_dev *
>   rte_fre

[dpdk-dev] [PATCH] net/ring: support secondary process

[Ferruh Yigit]

Also add missing .ini feature list documentation and document some
existing features including new "Multiprocess aware" feature.

Signed-off-by: Ferruh Yigit 
---
Cc: Kevin Laatz 
---
 doc/guides/nics/features/ring.ini | 10 ++
 drivers/net/ring/rte_eth_ring.c   | 17 +
 2 files changed, 27 insertions(+)
 create mode 100644 doc/guides/nics/features/ring.ini

diff --git a/doc/guides/nics/features/ring.ini 
b/doc/guides/nics/features/ring.ini
new file mode 100644
index 00..fbfa6555d7
--- /dev/null
+++ b/doc/guides/nics/features/ring.ini
@@ -0,0 +1,10 @@
+;
+; Supported features of the 'ring' network poll mode driver.
+;
+; Refer to default.ini for the full list of available PMD features.
+;
+[Features]
+Basic stats  = Y
+Multiprocess aware   = Y
+ARMv8= Y
+x86-64   = Y
diff --git a/drivers/net/ring/rte_eth_ring.c b/drivers/net/ring/rte_eth_ring.c
index 12046f5a00..f33951ee2e 100644
--- a/drivers/net/ring/rte_eth_ring.c
+++ b/drivers/net/ring/rte_eth_ring.c
@@ -595,6 +595,23 @@ rte_pmd_ring_probe(struct rte_vdev_device *dev)
 
PMD_LOG(INFO, "Initializing pmd_ring for %s", name);
 
+   if (rte_eal_process_type() == RTE_PROC_SECONDARY) {
+   eth_dev = rte_eth_dev_attach_secondary(name);
+   if (!eth_dev) {
+   PMD_LOG(ERR, "Failed to probe %s", name);
+   return -1;
+   }
+   eth_dev->dev_ops = &ops;
+   eth_dev->device = &dev->device;
+
+   eth_dev->rx_pkt_burst = eth_ring_rx;
+   eth_dev->tx_pkt_burst = eth_ring_tx;
+
+   rte_eth_dev_probing_finish(eth_dev);
+
+   return 0;
+   }
+
if (params == NULL || params[0] == '\0') {
ret = eth_dev_ring_create(name, dev, rte_socket_id(), 
DEV_CREATE,
ð_dev);
-- 
2.26.2

Re: [dpdk-dev] [PATCH v7 2/8] fib: make lookup function type configurable

[Vladimir Medvedkin]

Hi Thomas,

чт, 16 июл. 2020 г. в 15:32, Thomas Monjalon :

> 13/07/2020 13:56, Vladimir Medvedkin:
> > Add type argument to dir24_8_get_lookup_fn()
> > Now it supports 3 different lookup implementations:
> >  RTE_FIB_DIR24_8_SCALAR_MACRO
> >  RTE_FIB_DIR24_8_SCALAR_INLINE
> >  RTE_FIB_DIR24_8_SCALAR_UNI
> >
> > Add new rte_fib_set_lookup_fn() - user can change lookup
> > function type runtime.
> >
> > Signed-off-by: Vladimir Medvedkin 
> [...]
> > --- a/lib/librte_fib/rte_fib.h
> > +++ b/lib/librte_fib/rte_fib.h
> > +enum rte_fib_dir24_8_lookup_type {
> > + RTE_FIB_DIR24_8_SCALAR_MACRO,
> > + RTE_FIB_DIR24_8_SCALAR_INLINE,
> > + RTE_FIB_DIR24_8_SCALAR_UNI
> > +};
>
> Doxygen missing.
>
> [...]
> > +/**
> > + * Set lookup function based on type
> > + *
> > + * @param fib
> > + *   FIB object handle
> > + * @param type
> > + *   type of lookup function
> > + *
> > + * @return
> > + *-EINVAL on failure
> > + *0 on success
> > + */
> > +__rte_experimental
> > +int
> > +rte_fib_set_lookup_fn(struct rte_fib *fib,
> > + enum rte_fib_dir24_8_lookup_type type);
>
> I think the types deserve to be documented,
> explaining why using one or the other.
>

I'm going to get rid of extra lookup types in next releases, so there will
be only well understandable SCALAR and VECTOR types

>
>


-- 
Regards,
Vladimir

[dpdk-dev] [PATCH v10 00/11] adding support for python 3 only

[Kevin Laatz]

This patch set converts all python scripts in the project to use
python3 only and removes all deprecation notices associated with these
changes. This is due to python2 being EOL in January 2020.

---
v6:
  - rebased, removing conflict with make removal patchset.
  - added changes to buildtools/map_to_win.py

v7:
  - typo in email Cc'ing David Marchand
  - added maintainers for buildtools patch

v8:
  - removed unrelated cleanup
  - replaced integer cast with integer division operator

v9:
  - include documentation updates

v10:
  - made doc/guides/conf.py support python3 only

Kevin Laatz (4):
  app/test-cmdline: support python3 only
  app/test: support python3 only
  buildtools: support python3 only
  doc: support python3 only

Louise Kilheeney (7):
  usertools/dpdk-telemetry: support python3 only
  usertools/dpdk-devbind: support python3 only
  usertools/dpdk-pmdinfo: support python3 only
  usertools/cpu_layout: support python3 only
  devtools: support python3 only
  config/arm: support python3 only
  app/test-bbdev: support python3 only

 app/test-bbdev/test-bbdev.py |  7 +--
 app/test-cmdline/cmdline_test.py |  9 ++---
 app/test-cmdline/cmdline_test_data.py|  1 +
 app/test/autotest.py |  7 +--
 app/test/autotest_data.py|  1 +
 app/test/autotest_runner.py  | 21 -
 app/test/autotest_test_funcs.py  |  1 +
 buildtools/map_to_win.py |  3 +--
 config/arm/armv8_machine.py  |  2 +-
 devtools/update_version_map_abi.py   |  7 +--
 doc/guides/conf.py   |  9 ++---
 doc/guides/contributing/coding_style.rst |  2 +-
 doc/guides/linux_gsg/sys_reqs.rst|  2 +-
 doc/guides/rel_notes/deprecation.rst |  6 --
 doc/guides/rel_notes/release_20_11.rst   |  2 ++
 usertools/cpu_layout.py  | 13 ++---
 usertools/dpdk-devbind.py| 22 --
 usertools/dpdk-pmdinfo.py|  7 +--
 usertools/dpdk-telemetry-client.py   | 18 +++---
 usertools/dpdk-telemetry.py  |  2 +-
 20 files changed, 35 insertions(+), 107 deletions(-)

-- 
2.25.1

[dpdk-dev] [PATCH v10 01/11] usertools/dpdk-telemetry: support python3 only

[Kevin Laatz]

From: Louise Kilheeney 

Changed script to explicitly use python3 only to avoid
maintaining python 2 and removed deprecation notice.

Cc: Kevin Laatz 

Signed-off-by: Louise Kilheeney 
Signed-off-by: Kevin Laatz 
Acked-by: Bruce Richardson 
Acked-by: Robin Jarry 
Acked-by: Stephen Hemminger 

---
v8: removed cleanup not related to python3 only support

v10: fix direct call to directly call to /usr/bin/python3 in
 dpdk-telemetry.py
---
 usertools/dpdk-telemetry-client.py | 18 +++---
 usertools/dpdk-telemetry.py|  2 +-
 2 files changed, 4 insertions(+), 16 deletions(-)

diff --git a/usertools/dpdk-telemetry-client.py 
b/usertools/dpdk-telemetry-client.py
index 98d28fa89b..d8e439027c 100755
--- a/usertools/dpdk-telemetry-client.py
+++ b/usertools/dpdk-telemetry-client.py
@@ -1,10 +1,7 @@
-#! /usr/bin/env python
+#! /usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2018 Intel Corporation
 
-from __future__ import print_function
-from __future__ import unicode_literals
-
 import socket
 import os
 import sys
@@ -18,15 +15,6 @@
 GLOBAL_METRICS_REQ = 
"{\"action\":0,\"command\":\"global_stat_values\",\"data\":null}"
 DEFAULT_FP = "/var/run/dpdk/default_client"
 
-try:
-raw_input  # Python 2
-except NameError:
-raw_input = input  # Python 3
-
-if sys.version_info.major < 3:
-print("WARNING: Python 2 is deprecated for use in DPDK, and will not work 
in future releases.", file=sys.stderr)
-print("Please use Python 3 instead", file=sys.stderr)
-
 class Socket:
 
 def __init__(self):
@@ -86,7 +74,7 @@ def requestMetrics(self): # Requests metrics for given client
 
 def repeatedlyRequestMetrics(self, sleep_time): # Recursively requests 
metrics for given client
 print("\nPlease enter the number of times you'd like to continuously 
request Metrics:")
-n_requests = int(raw_input("\n:"))
+n_requests = int(input("\n:"))
 print("\033[F") #Removes the user input from screen, cleans it up
 print("\033[K")
 for i in range(n_requests):
@@ -107,7 +95,7 @@ def interactiveMenu(self, sleep_time): # Creates Interactive 
menu within the scr
 print("[4] Unregister client")
 
 try:
-self.choice = int(raw_input("\n:"))
+self.choice = int(input("\n:"))
 print("\033[F") #Removes the user input for screen, cleans it 
up
 print("\033[K")
 if self.choice == 1:
diff --git a/usertools/dpdk-telemetry.py b/usertools/dpdk-telemetry.py
index 8e4039d574..181859658f 100755
--- a/usertools/dpdk-telemetry.py
+++ b/usertools/dpdk-telemetry.py
@@ -1,4 +1,4 @@
-#! /usr/bin/python3
+#! /usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2020 Intel Corporation
 
-- 
2.25.1

[dpdk-dev] [PATCH v10 02/11] usertools/dpdk-devbind: support python3 only

[Kevin Laatz]

From: Louise Kilheeney 

Changed script to explicitly use python3 only to avoid
maintaining python 2 and removed deprecation notice.

Signed-off-by: Louise Kilheeney 
Reviewed-by: Bruce Richardson 
Acked-by: Robin Jarry 
Acked-by: Stephen Hemminger 
---
 usertools/dpdk-devbind.py | 22 --
 1 file changed, 4 insertions(+), 18 deletions(-)

diff --git a/usertools/dpdk-devbind.py b/usertools/dpdk-devbind.py
index 094c2ffc8b..8278a748d4 100755
--- a/usertools/dpdk-devbind.py
+++ b/usertools/dpdk-devbind.py
@@ -1,9 +1,8 @@
-#! /usr/bin/env python
+#!/usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2010-2014 Intel Corporation
 #
 
-from __future__ import print_function
 import sys
 import os
 import getopt
@@ -12,10 +11,6 @@
 from os.path import exists, abspath, dirname, basename
 from os.path import join as path_join
 
-if sys.version_info.major < 3:
-print("WARNING: Python 2 is deprecated for use in DPDK, and will not work 
in future releases.", file=sys.stderr)
-print("Please use Python 3 instead", file=sys.stderr)
-
 # The PCI base class for all devices
 network_class = {'Class': '02', 'Vendor': None, 'Device': None,
 'SVendor': None, 'SDevice': None}
@@ -154,14 +149,6 @@ def usage():
 
 """ % locals())  # replace items from local variables
 
-
-# This is roughly compatible with check_output function in subprocess module
-# which is only available in python 2.7.
-def check_output(args, stderr=None):
-'''Run a command and capture its output'''
-return subprocess.Popen(args, stdout=subprocess.PIPE,
-stderr=stderr).communicate()[0]
-
 # check if a specific kernel module is loaded
 def module_is_loaded(module):
 global loaded_modules
@@ -218,8 +205,7 @@ def get_pci_device_details(dev_id, probe_lspci):
 device = {}
 
 if probe_lspci:
-extra_info = check_output(["lspci", "-vmmks", dev_id]).splitlines()
-
+extra_info = subprocess.check_output(["lspci", "-vmmks", 
dev_id]).splitlines()
 # parse lspci details
 for line in extra_info:
 if len(line) == 0:
@@ -255,7 +241,7 @@ def get_device_details(devices_type):
 # first loop through and read details for all devices
 # request machine readable format, with numeric IDs and String
 dev = {}
-dev_lines = check_output(["lspci", "-Dvmmnnk"]).splitlines()
+dev_lines = subprocess.check_output(["lspci", "-Dvmmnnk"]).splitlines()
 for dev_line in dev_lines:
 if len(dev_line) == 0:
 if device_type_match(dev, devices_type):
@@ -283,7 +269,7 @@ def get_device_details(devices_type):
 # check what is the interface if any for an ssh connection if
 # any to this host, so we can mark it later.
 ssh_if = []
-route = check_output(["ip", "-o", "route"])
+route = subprocess.check_output(["ip", "-o", "route"])
 # filter out all lines for 169.254 routes
 route = "\n".join(filter(lambda ln: not ln.startswith("169.254"),
  route.decode().splitlines()))
-- 
2.25.1

[dpdk-dev] [PATCH v10 04/11] usertools/cpu_layout: support python3 only

[Kevin Laatz]

From: Louise Kilheeney 

Changed script to explicitly use python3 only to avoid
maintaining python 2 and removed deprecation notice.

Signed-off-by: Louise Kilheeney 
Reviewed-by: Bruce Richardson 
Acked-by: Robin Jarry 
Acked-by: Stephen Hemminger 
---
 usertools/cpu_layout.py | 13 ++---
 1 file changed, 2 insertions(+), 11 deletions(-)

diff --git a/usertools/cpu_layout.py b/usertools/cpu_layout.py
index 5423c7965f..89a48cec46 100755
--- a/usertools/cpu_layout.py
+++ b/usertools/cpu_layout.py
@@ -1,18 +1,9 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2010-2014 Intel Corporation
 # Copyright(c) 2017 Cavium, Inc. All rights reserved.
 
-from __future__ import print_function
 import sys
-try:
-xrange # Python 2
-except NameError:
-xrange = range # Python 3
-
-if sys.version_info.major < 3:
-print("WARNING: Python 2 is deprecated for use in DPDK, and will not work 
in future releases.", file=sys.stderr)
-print("Please use Python 3 instead", file=sys.stderr)
 
 sockets = []
 cores = []
@@ -21,7 +12,7 @@
 fd = open("{}/kernel_max".format(base_path))
 max_cpus = int(fd.read())
 fd.close()
-for cpu in xrange(max_cpus + 1):
+for cpu in range(max_cpus + 1):
 try:
 fd = open("{}/cpu{}/topology/core_id".format(base_path, cpu))
 except IOError:
-- 
2.25.1

[dpdk-dev] [PATCH v10 06/11] app/test: support python3 only

[Kevin Laatz]

Changed script to explicitly use python3 only to avoid
maintaining python 2 and removed deprecation notice.

Signed-off-by: Louise Kilheeney 
Signed-off-by: Kevin Laatz 
Acked-by: Robin Jarry 
Acked-by: Stephen Hemminger 

---
v6:
  - Removed changes to mk/rte.sdktest.mk since the file no longer exists
---
 app/test/autotest.py|  7 +--
 app/test/autotest_data.py   |  1 +
 app/test/autotest_runner.py | 21 -
 app/test/autotest_test_funcs.py |  1 +
 4 files changed, 11 insertions(+), 19 deletions(-)

diff --git a/app/test/autotest.py b/app/test/autotest.py
index cf7584ccd7..9eef1efbe5 100644
--- a/app/test/autotest.py
+++ b/app/test/autotest.py
@@ -1,9 +1,8 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2010-2014 Intel Corporation
 
 # Script that uses either test app or qemu controlled by python-pexpect
-from __future__ import print_function
 import autotest_data
 import autotest_runner
 import sys
@@ -17,10 +16,6 @@ def usage():
 usage()
 sys.exit(1)
 
-if sys.version_info.major < 3:
-print("WARNING: Python 2 is deprecated for use in DPDK, and will not work 
in future releases.", file=sys.stderr)
-print("Please use Python 3 instead", file=sys.stderr)
-
 target = sys.argv[2]
 
 test_whitelist = None
diff --git a/app/test/autotest_data.py b/app/test/autotest_data.py
index 4b7da45e09..097638941f 100644
--- a/app/test/autotest_data.py
+++ b/app/test/autotest_data.py
@@ -1,3 +1,4 @@
+#!/usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2010-2014 Intel Corporation
 
diff --git a/app/test/autotest_runner.py b/app/test/autotest_runner.py
index 95e74c760d..998fe57a55 100644
--- a/app/test/autotest_runner.py
+++ b/app/test/autotest_runner.py
@@ -1,10 +1,10 @@
+#!/usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2010-2014 Intel Corporation
 
 # The main logic behind running autotests in parallel
 
-from __future__ import print_function
-import StringIO
+import io
 import csv
 from multiprocessing import Pool, Queue
 import pexpect
@@ -50,11 +50,7 @@ def first_cpu_on_node(node_nr):
 map(os.path.basename, cpu_path)
 )
 )
-# for compatibility between python 3 and 2 we need to make interable out
-# of filter return as it returns list in python 2 and a generator in 3
-m = next(iter(cpu_name))
-return int(m.group(1))
-
+return int(next(cpu_name).group(1))
 
 pool_child = None  # per-process child
 
@@ -78,7 +74,7 @@ def pool_init(queue, result_queue):
 cmdline = "%s %s" % (cmdline, prefix_cmdline)
 
 # prepare logging of init
-startuplog = StringIO.StringIO()
+startuplog = io.StringIO()
 
 # run test app
 try:
@@ -86,8 +82,7 @@ def pool_init(queue, result_queue):
 print("\n%s %s\n" % ("=" * 20, prefix), file=startuplog)
 print("\ncmdline=%s" % cmdline, file=startuplog)
 
-pool_child = pexpect.spawn(cmdline, logfile=startuplog)
-
+pool_child = pexpect.spawn(cmdline, logfile=startuplog, 
encoding='utf-8')
 # wait for target to boot
 if not wait_prompt(pool_child):
 pool_child.close()
@@ -138,7 +133,7 @@ def run_test(target, test):
 # create log buffer for each test
 # in multiprocessing environment, the logging would be
 # interleaved and will create a mess, hence the buffering
-logfile = StringIO.StringIO()
+logfile = io.StringIO()
 pool_child.logfile = logfile
 
 # make a note when the test started
@@ -210,9 +205,9 @@ def __init__(self, cmdline, target, blacklist, whitelist, 
n_processes):
 # parse the binary for available test commands
 binary = cmdline.split()[0]
 stripped = 'not stripped' not in \
-   subprocess.check_output(['file', binary])
+   subprocess.check_output(['file', binary]).decode()
 if not stripped:
-symbols = subprocess.check_output(['nm', binary]).decode('utf-8')
+symbols = subprocess.check_output(['nm', binary]).decode()
 self.avail_cmds = re.findall('test_register_(\w+)', symbols)
 else:
 self.avail_cmds = None
diff --git a/app/test/autotest_test_funcs.py b/app/test/autotest_test_funcs.py
index 26688b7132..775dfd1dc5 100644
--- a/app/test/autotest_test_funcs.py
+++ b/app/test/autotest_test_funcs.py
@@ -1,3 +1,4 @@
+#!/usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2010-2014 Intel Corporation
 
-- 
2.25.1

[dpdk-dev] [PATCH v10 03/11] usertools/dpdk-pmdinfo: support python3 only

[Kevin Laatz]

From: Louise Kilheeney 

Changed script to explicitly use python3 only to avoid
maintaining python 2 and removed deprecation notice.

Cc: Neil Horman 

Signed-off-by: Louise Kilheeney 
Reviewed-by: Bruce Richardson 
Acked-by: Neil Horman 
Acked-by: Robin Jarry 
Acked-by: Stephen Hemminger 
---
 usertools/dpdk-pmdinfo.py | 7 +--
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/usertools/dpdk-pmdinfo.py b/usertools/dpdk-pmdinfo.py
index f9ed755176..1661982791 100755
--- a/usertools/dpdk-pmdinfo.py
+++ b/usertools/dpdk-pmdinfo.py
@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2016  Neil Horman 
 
@@ -7,8 +7,6 @@
 # Utility to dump PMD_INFO_STRING support from an object file
 #
 # -
-from __future__ import print_function
-from __future__ import unicode_literals
 import json
 import io
 import os
@@ -28,9 +26,6 @@
 pcidb = None
 
 # ===
-if sys.version_info.major < 3:
-print("WARNING: Python 2 is deprecated for use in DPDK, and will not 
work in future releases.", file=sys.stderr)
-print("Please use Python 3 instead", file=sys.stderr)
 
 class Vendor:
 """
-- 
2.25.1

[dpdk-dev] [PATCH v10 05/11] app/test-cmdline: support python3 only

[Kevin Laatz]

Changed script to explicitly use python3 only to avoid
maintaining python 2 and removed deprecation notice.

Cc: Olivier Matz 

Signed-off-by: Louise Kilheeney 
Signed-off-by: Kevin Laatz 
Reviewed-by: Bruce Richardson 
Acked-by: Robin Jarry 
Acked-by: Stephen Hemminger 

---
v5:
  - fixed python3 issue causing script to fail. Divisions without the
typecast lead to a floating point result, which was messing up the
loop.

v6:
  - Removed changes to mk/rte.sdktest.mk since it no longer exists.

v8:
  - Replaced integer cast with integer division operator.
---
 app/test-cmdline/cmdline_test.py  | 9 ++---
 app/test-cmdline/cmdline_test_data.py | 1 +
 2 files changed, 3 insertions(+), 7 deletions(-)

diff --git a/app/test-cmdline/cmdline_test.py b/app/test-cmdline/cmdline_test.py
index 954428e2bf..f3377313e9 100755
--- a/app/test-cmdline/cmdline_test.py
+++ b/app/test-cmdline/cmdline_test.py
@@ -1,9 +1,8 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2010-2014 Intel Corporation
 
 # Script that runs cmdline_test app and feeds keystrokes into it.
-from __future__ import print_function
 import cmdline_test_data
 import os
 import pexpect
@@ -19,10 +18,6 @@ def runTest(child, test):
 return 0
 child.expect(test["Result"], 1)
 
-if sys.version_info.major < 3:
-print("WARNING: Python 2 is deprecated for use in DPDK, and will not work 
in future releases.", file=sys.stderr)
-print("Please use Python 3 instead", file=sys.stderr)
-
 #
 # history test is a special case
 #
@@ -43,7 +38,7 @@ def runHistoryTest(child):
 i = 0
 
 # fill the history with numbers
-while i < history_size / 10:
+while i < history_size // 10:
 # add 1 to prevent from parsing as octals
 child.send("1" + str(i).zfill(8) + cmdline_test_data.ENTER)
 # the app will simply print out the number
diff --git a/app/test-cmdline/cmdline_test_data.py 
b/app/test-cmdline/cmdline_test_data.py
index 114d2cb6a0..2d9b3262a6 100644
--- a/app/test-cmdline/cmdline_test_data.py
+++ b/app/test-cmdline/cmdline_test_data.py
@@ -1,3 +1,4 @@
+#!/usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2010-2014 Intel Corporation
 
-- 
2.25.1

[dpdk-dev] [PATCH v10 07/11] devtools: support python3 only

[Kevin Laatz]

From: Louise Kilheeney 

Changed script to explicitly use python3 only to avoid
maintaining python 2 and removed deprecation notice.

Cc: Neil Horman 
Cc: Ray Kinsella 

Signed-off-by: Louise Kilheeney 
Acked-by: Ray Kinsella 
Acked-by: Neil Horman 
Acked-by: Robin Jarry 
Acked-by: Stephen Hemminger 
---
 devtools/update_version_map_abi.py | 7 +--
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/devtools/update_version_map_abi.py 
b/devtools/update_version_map_abi.py
index 10c3bc8098..3536a54b44 100755
--- a/devtools/update_version_map_abi.py
+++ b/devtools/update_version_map_abi.py
@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2019 Intel Corporation
 
@@ -9,7 +9,6 @@
 from the devtools/update-abi.sh utility.
 """
 
-from __future__ import print_function
 import argparse
 import sys
 import re
@@ -160,10 +159,6 @@ def __generate_internal_abi(f_out, lines):
 print("};", file=f_out)
 
 def __main():
-if sys.version_info.major < 3:
-print("WARNING: Python 2 is deprecated for use in DPDK, and will not 
work in future releases.", file=sys.stderr)
-print("Please use Python 3 instead", file=sys.stderr)
-
 arg_parser = argparse.ArgumentParser(
 description='Merge versions in linker version script.')
 
-- 
2.25.1

[dpdk-dev] [PATCH v10 09/11] app/test-bbdev: support python3 only

[Kevin Laatz]

From: Louise Kilheeney 

Changed script to explicitly use python3 only to avoid
maintaining python 2 and removed deprecation notice.

Cc: Nicolas Chautru 

Signed-off-by: Louise Kilheeney 
Acked-by: Robin Jarry 
Acked-by: Stephen Hemminger 
---
 app/test-bbdev/test-bbdev.py | 7 +--
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/app/test-bbdev/test-bbdev.py b/app/test-bbdev/test-bbdev.py
index 5ae2dc6c49..7d67dbe30a 100755
--- a/app/test-bbdev/test-bbdev.py
+++ b/app/test-bbdev/test-bbdev.py
@@ -1,9 +1,8 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2017 Intel Corporation
 
-from __future__ import print_function
 import sys
 import os
 import argparse
@@ -16,10 +15,6 @@ def kill(process):
 print("ERROR: Test app timed out")
 process.kill()
 
-if sys.version_info.major < 3:
-print("WARNING: Python 2 is deprecated for use in DPDK, and will not work 
in future releases.", file=sys.stderr)
-print("Please use Python 3 instead", file=sys.stderr)
-
 if "RTE_SDK" in os.environ:
 dpdk_path = os.environ["RTE_SDK"]
 else:
-- 
2.25.1

[dpdk-dev] [PATCH v10 08/11] config/arm: support python3 only

[Kevin Laatz]

From: Louise Kilheeney 

Changed script to explicitly use python3 only to avoid
maintaining python 2.

Cc: Thomas Monjalon 

Signed-off-by: Louise Kilheeney 
Acked-by: Robin Jarry 
Acked-by: Stephen Hemminger 
---
 config/arm/armv8_machine.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/config/arm/armv8_machine.py b/config/arm/armv8_machine.py
index 404866d2f8..1f689d9a83 100755
--- a/config/arm/armv8_machine.py
+++ b/config/arm/armv8_machine.py
@@ -1,4 +1,4 @@
-#!/usr/bin/python
+#!/usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2017 Cavium, Inc
 
-- 
2.25.1

[dpdk-dev] [PATCH v10 10/11] buildtools: support python3 only

[Kevin Laatz]

Changed script to explicitly use python3 only to avoid
maintaining python 2 and removed deprecation notice.

Cc: Dmitry Kozlyuk 
Cc: Narcisa Ana Maria Vasile 
Cc: Dmitry Malloy 
Cc: Pallavi Kadam 

Signed-off-by: Kevin Laatz 
Acked-by: Robin Jarry 
Acked-by: Stephen Hemminger 

---
v7:
  - Cc Maintainers
---
 buildtools/map_to_win.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/buildtools/map_to_win.py b/buildtools/map_to_win.py
index 2990b58634..2a6cb88605 100644
--- a/buildtools/map_to_win.py
+++ b/buildtools/map_to_win.py
@@ -1,8 +1,7 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2019 Intel Corporation
 
-from __future__ import print_function
 import sys
 from os.path import dirname, basename, join, exists
 
-- 
2.25.1

[dpdk-dev] [PATCH v10 11/11] doc: support python3 only

[Kevin Laatz]

Changed script to explicitly use python3 only to avoid maintaining python 2
and removed references to python 2 from the documentation.

The deprecation notice for python 2 support has been removed. A section
was added to the rel_notes for this patchset.

Cc: Ray Kinsella 
Cc: Neil Horman 
Cc: John McNamara 
Cc: Marko Kovacevic 

Signed-off-by: Kevin Laatz 

---
v10:
  - changes to doc/guides/conf.py to only support python3
---
 doc/guides/conf.py   | 9 ++---
 doc/guides/contributing/coding_style.rst | 2 +-
 doc/guides/linux_gsg/sys_reqs.rst| 2 +-
 doc/guides/rel_notes/deprecation.rst | 6 --
 doc/guides/rel_notes/release_20_11.rst   | 2 ++
 5 files changed, 6 insertions(+), 15 deletions(-)

diff --git a/doc/guides/conf.py b/doc/guides/conf.py
index 9ebc26ed3f..ef550f68c0 100644
--- a/doc/guides/conf.py
+++ b/doc/guides/conf.py
@@ -1,7 +1,7 @@
+#!/usr/bin/env python3
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2010-2015 Intel Corporation
 
-from __future__ import print_function
 from docutils import nodes
 from distutils.version import LooseVersion
 from sphinx import __version__ as sphinx_version
@@ -13,12 +13,7 @@
 from os.path import dirname
 from os.path import join as path_join
 
-try:
-# Python 2.
-import ConfigParser as configparser
-except:
-# Python 3.
-import configparser
+import configparser
 
 try:
 import sphinx_rtd_theme
diff --git a/doc/guides/contributing/coding_style.rst 
b/doc/guides/contributing/coding_style.rst
index 0be9546a6a..4fc010ca49 100644
--- a/doc/guides/contributing/coding_style.rst
+++ b/doc/guides/contributing/coding_style.rst
@@ -765,7 +765,7 @@ specializations, run the ``app/test`` binary, and use the 
``dump_log_types``
 Python Code
 ---
 
-All Python code should work with Python 2.7+ and 3.2+ and be compliant with
+All Python code should work with 3.2+ and be compliant with
 `PEP8 (Style Guide for Python Code) 
`_.
 
 The ``pep8`` tool can be used for testing compliance with the guidelines.
diff --git a/doc/guides/linux_gsg/sys_reqs.rst 
b/doc/guides/linux_gsg/sys_reqs.rst
index 6b9f573fce..66d9723dd6 100644
--- a/doc/guides/linux_gsg/sys_reqs.rst
+++ b/doc/guides/linux_gsg/sys_reqs.rst
@@ -47,7 +47,7 @@ Compilation of the DPDK
 
 * Python v3.5+ is needed to build DPDK using meson and ninja
 
-* Python 2.7+ or 3.2+, to use various helper scripts included in the DPDK 
package.
+* Python 3.2+, to use various helper scripts included in the DPDK package.
 
 *   Meson (version 0.47.1+) and ninja
 
diff --git a/doc/guides/rel_notes/deprecation.rst 
b/doc/guides/rel_notes/deprecation.rst
index 1f888fa90e..0a7e874b2c 100644
--- a/doc/guides/rel_notes/deprecation.rst
+++ b/doc/guides/rel_notes/deprecation.rst
@@ -308,12 +308,6 @@ Deprecation Notices
   In this case the function will return -1 unless the environment is unset 
first
   (using ``rte_power_unset_env``). Other function usage scenarios will not 
change.
 
-* python: Since the beginning of 2020, Python 2 has officially reached
-  end-of-support: https://www.python.org/doc/sunset-python-2/.
-  Python 2 support will be completely removed in 20.11.
-  In 20.08, explicit deprecation warnings will be displayed when running
-  scripts with Python 2.
-
 * dpdk-setup.sh: This old script relies on deprecated stuff, and especially
   ``make``. Given environments are too much variables for such a simple script,
   it will be removed in DPDK 20.11.
diff --git a/doc/guides/rel_notes/release_20_11.rst 
b/doc/guides/rel_notes/release_20_11.rst
index cc726095a7..2eb83c8137 100644
--- a/doc/guides/rel_notes/release_20_11.rst
+++ b/doc/guides/rel_notes/release_20_11.rst
@@ -68,6 +68,8 @@ Removed Items
Also, make sure to start the actual text at the margin.
===
 
+* Removed python 2 support since it was EOL'd in January 2020.
+
 
 API Changes
 ---
-- 
2.25.1

Re: [dpdk-dev] [PATCH v3 1/3] eal: add API for bus close

[Ferruh Yigit]

On 8/26/2020 6:52 AM, rohit@nxp.com wrote:

From: Rohit Raj 

As per the current code we have API for bus probe, but the
bus close API is missing. This breaks the multi process
scenarios as objects are not cleaned while terminating the
secondary processes.

This patch adds a new API rte_bus_close() for cleanup of
bus objects which were acquired during probe.

Signed-off-by: Rohit Raj 
---

v3:
* nit: combined nested if statements

v2:
* Moved rte_bus_close call to rte_eal_cleanup path.

  lib/librte_eal/common/eal_common_bus.c | 32 +-
  lib/librte_eal/include/rte_bus.h   | 25 +++-
  lib/librte_eal/linux/eal.c |  1 +
  lib/librte_eal/rte_eal_version.map |  3 +++
  4 files changed, 59 insertions(+), 2 deletions(-)

diff --git a/lib/librte_eal/common/eal_common_bus.c 
b/lib/librte_eal/common/eal_common_bus.c
index baa5b532a..5fd7cf6c5 100644
--- a/lib/librte_eal/common/eal_common_bus.c
+++ b/lib/librte_eal/common/eal_common_bus.c
@@ -1,5 +1,5 @@
  /* SPDX-License-Identifier: BSD-3-Clause
- * Copyright 2016 NXP
+ * Copyright 2016,2020 NXP
   */
  
  #include 

@@ -56,6 +56,36 @@ rte_bus_scan(void)
return 0;
  }
  
+int

+rte_bus_close(void)
+{
+   int ret;
+   struct rte_bus *bus, *vbus = NULL;
+
+   TAILQ_FOREACH(bus, &rte_bus_list, next) {
+   if (!strcmp(bus->name, "vdev")) {
+   vbus = bus;
+   continue;
+   }


This special treatment for 'vdev' bus is done in probe to be sure physically 
device port ids start from '0',  I guess we don't need to do this for 'close'.



+
+   if (bus->close) {
+   ret = bus->close();
+   if (ret)
+   RTE_LOG(ERR, EAL, "Bus (%s) close failed.\n",
+   bus->name);
+   }
+   }
+
+   if (vbus && vbus->close) {
+   ret = vbus->close();
+   if (ret)
+   RTE_LOG(ERR, EAL, "Bus (%s) close failed.\n",
+   vbus->name);
+   }
+
+   return 0;
+}
+
  /* Probe all devices of all buses */
  int
  rte_bus_probe(void)
diff --git a/lib/librte_eal/include/rte_bus.h b/lib/librte_eal/include/rte_bus.h
index d3034d0ed..af4787b18 100644
--- a/lib/librte_eal/include/rte_bus.h
+++ b/lib/librte_eal/include/rte_bus.h
@@ -1,5 +1,5 @@
  /* SPDX-License-Identifier: BSD-3-Clause
- * Copyright 2016 NXP
+ * Copyright 2016,2020 NXP
   */
  
  #ifndef _RTE_BUS_H_

@@ -67,6 +67,18 @@ typedef int (*rte_bus_scan_t)(void);
   */
  typedef int (*rte_bus_probe_t)(void);
  
+/**

+ * Implementation specific close function which is responsible for closing
+ * devices on that bus.
+ *
+ * This is called while iterating over each registered bus.
+ *
+ * @return
+ * 0 for successful close
+ * !0 for any error while closing
+ */
+typedef int (*rte_bus_close_t)(void);
+


As I checked the 'rte_fslmc_bus->close()' ops, it iterates on all devices in the 
bus instead of doing a bus level close, in that case
instead of adding a new 'close' bus operations, will it work if existing 
'bus->unplug(dev)' used?
Whatever done in the 'rte_fslmc_bus->close()' per device, can it be done under 
the 'fslmc_bus_unplug()'?


And in that case a 'rte_bus_remove()' API can be added which can call 
'bus->unplug(dev)' for all buses and it will be beneficial for all buses, and it 
can fit well into the 'rte_eal_cleanup()'.


What do you think?

[dpdk-dev] [PATCH] app/procinfo: close devices before exit

[Ferruh Yigit]

As well as this is correct thing to close devices before exit, it is
also useful to test the closing devices from secondary process.

Signed-off-by: Ferruh Yigit 
---
Cc: Thomas Monjalon 
---
 app/proc-info/main.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/app/proc-info/main.c b/app/proc-info/main.c
index 64fb83b38d..b0de39489a 100644
--- a/app/proc-info/main.c
+++ b/app/proc-info/main.c
@@ -1349,6 +1349,9 @@ main(int argc, char **argv)
if (enable_iter_mempool)
iter_mempool(mempool_iter_name);
 
+   RTE_ETH_FOREACH_DEV(i)
+   rte_eth_dev_close(i);
+
ret = rte_eal_cleanup();
if (ret)
printf("Error from rte_eal_cleanup(), %d\n", ret);
-- 
2.26.2

Re: [dpdk-dev] [PATCH v3 02/29] ethdev: allow drivers to return error on close

[Ferruh Yigit]

On 9/29/2020 11:53 AM, Andrew Rybchenko wrote:

On 9/29/20 2:14 AM, Thomas Monjalon wrote:

The device operation .dev_close was returning void.
This driver interface is changed to return an int.

Note that the API rte_eth_dev_close() is still returning void,
although a deprecation notice is pending to change it as well.

Signed-off-by: Thomas Monjalon 
Reviewed-by: Rosen Xu 
Reviewed-by: Sachin Saxena 
Reviewed-by: Liron Himi 


Reviewed-by: Andrew Rybchenko 



Reviewed-by: Ferruh Yigit 

Re: [dpdk-dev] [PATCH v3 25/29] app/testpmd: reset port status on close notification

[Ferruh Yigit]

On 9/29/2020 12:14 AM, Thomas Monjalon wrote:

Since rte_eth_dev_release_port() is called on all port close operations,
the event RTE_ETH_EVENT_DESTROY can be reliably used for resetting
the port status on the application side.

The intermediate state RTE_PORT_HANDLING is removed in close_port()
because a port can also be closed by a PMD in a device remove operation.

In case multiple ports are closed, calling remove_invalid_ports()
only once is enough.

Signed-off-by: Thomas Monjalon 



Reviewed-by: Ferruh Yigit 

Re: [dpdk-dev] [PATCH v3 28/29] ethdev: reset all when releasing a port

[Ferruh Yigit]

On 9/29/2020 5:35 PM, Ferruh Yigit wrote:

On 9/29/2020 5:02 PM, Thomas Monjalon wrote:

29/09/2020 17:50, Ferruh Yigit:

On 9/29/2020 12:58 PM, Wang, Haiyue wrote:

From: Thomas Monjalon 

29/09/2020 12:26, Maxime Coquelin:

On 9/29/20 1:14 AM, Thomas Monjalon wrote:

The function rte_eth_dev_release_port() was resetting partially
the struct rte_eth_dev. The drivers were completing it
with more pointers set to NULL in the close or remove operations.

A full memset is done so most of those assignments become useless.

[...]

With this patch, I get following segfault at init time with Virtio PMD:

  Program received signal SIGSEGV, Segmentation fault.
  0x00854c9b in rte_eth_dev_callback_register (port_id=32,
  event=RTE_ETH_EVENT_UNKNOWN, cb_fn=0x4b24de
,
  cb_arg=0x0) at ../lib/librte_ethdev/rte_ethdev.c:4042
  4042
TAILQ_INSERT_TAIL(&(dev->link_intr_cbs),


Yes this is because after closing a port, everything is resetted,
including .link_intr_cbs which is set only once in a constructor:
http://git.dpdk.org/dpdk/commit/?id=9ec0b3869d8

I can change this patch to selectively set pointers to NULL.

Or if we prefer a big memset 0, we need to rework how RTE_ETH_ALL
is managed to register a callback for any event.
Instead of setting the callback for all ports, we could have
a special catch-call callback list which is called for all events.
This way we could revert initializing .link_intr_cbs in eth_dev_get().




Move 'struct rte_eth_dev_cb_list link_intr_cbs' to the end of 'struct 
rte_eth_dev',

and starting from link_intr_cbs, these members will be kept after closed ? :-)

memset(eth_dev, 0, offsetof(struct rte_eth_dev, link_intr_cbs));



This is similar version of a previous patch:
https://patches.dpdk.org/patch/65795/


I forgot this patch :)


That one is also waiting because I remember it was not safe for hotplug.


I don't think there is a safety issue.
It makes harder to play with dangling pointers, which is good.
Note: the .device pointer was already resetted by
rte_eth_dev_pci_generic_remove(), and generalized in patch 1.



As far as I remember, resetting '.device' in the 'close()' was causing some 
issues on hot remove. Yes patch one already does it, I will test the hot remove 
after close with it.




I didn't get any error with 1/29, will proceed with the patchset.


Instead of a big memset, I am for selectively set pointers to NULL.


I will prepare the patch with selective resets.

Re: [dpdk-dev] [PATCH v3 26/29] app/testpmd: align behaviour of multi-port detach

[Ferruh Yigit]

On 9/29/2020 12:14 AM, Thomas Monjalon wrote:

A port can be closed in multiple situations:
- close command calling close_port() -> rte_eth_dev_close()
- exit calling close_port() -> rte_eth_dev_close()
- hotplug calling close_port() -> rte_eth_dev_close()
- hotplug calling detach_device() -> rte_dev_remove()
- port detach command, detach_device() -> rte_dev_remove()
- device detach command, detach_devargs() -> rte_eal_hotplug_remove()

The flow rules are flushed before each close.
It was already done in close_port(), detach_devargs() and
detach_port_device() which calls detach_device(),
but not in detach_device(). As a consequence, it was missing for siblings
of port detach command and unplugged device.
The check before calling port_flow_flush() is moved inside the function.

The state of the port to close is checked to be stopped.
As above, this check was missing in detach_device(),
impacting the cases of a multi-port device unplugged or detached
with the port detach command.

Signed-off-by: Thomas Monjalon 



Reviewed-by: Ferruh Yigit 

[dpdk-dev] [RFC PATCH v4 0/3] librte_ethdev: error recovery support

[Kalesh A P]

From: Kalesh AP 

The error recovery solution is a protocol implemented between firmware
and bnxt PMD to recover from the fatal errors without a system reboot.
There is an alarm thread which constantly monitors the health of the
firmware and initiates a recovery when needed.

There are two scenarios here:

1. Hardware or firmware encountered an error which firmware detected.
   Firmware is in operational status here. In this case, firmware can
   reset the chip and notify the driver about the reset.
2. Hardware or firmware encountered an error but firmware is dead/hung.
   Firmware is not in operational status. In this case, the only possible
   way to recover the adapter is through host driver(bnxt PMD).

In both cases, bnxt PMD reinitializes with the FW again after the reset.
During that recovery process, data path will be halted and any control path
operation would fail. So, the PMD has to notify the application about this
reset/error event to prevent any activities from the application while
the PMD is recovering from the error.

This patch set adds support for the reset and recovery event in
the rte_eth_event framework. FW error and FW reset conditions would be
managed by the PMD. Driver uses RTE_ETH_EVENT_RESET event to notify
the applications about the FW reset or error. In such cases,
PMD use the RTE_ETH_EVENT_RECOVERED event to notify application about
PMD has recovered from FW reset or FW error.

v4: Added doxygen comments about new events.
V3: Fixed a typo in commit log.
V2: Added a new event RTE_ETH_EVENT_RESET instead of using the
RTE_ETH_EVENT_INTR_RESET to notify applications about device reset.

Kalesh AP (3):
  ethdev: support device reset and recovery events
  net/bnxt: notify applications about device reset/recovery
  app/testpmd: handle device recovery event

 app/test-pmd/testpmd.c |  6 +-
 drivers/net/bnxt/bnxt_cpr.c|  3 +++
 drivers/net/bnxt/bnxt_ethdev.c |  9 +
 lib/librte_ethdev/rte_ethdev.h | 17 +
 4 files changed, 34 insertions(+), 1 deletion(-)

-- 
2.10.1

[dpdk-dev] [RFC PATCH v4 1/3] ethdev: support device reset and recovery events

[Kalesh A P]

From: Kalesh AP 

Adding support for device reset and recovery events in the
rte_eth_event framework. FW error and FW reset conditions would be
managed internally by PMD without needing application intervention.
In such cases, PMD would need reset/recovery events to notify application
that PMD is undergoing a reset.

Signed-off-by: Somnath Kotur 
Signed-off-by: Kalesh AP 
Reviewed-by: Ajit Khaparde 
---
 lib/librte_ethdev/rte_ethdev.h | 17 +
 1 file changed, 17 insertions(+)

diff --git a/lib/librte_ethdev/rte_ethdev.h b/lib/librte_ethdev/rte_ethdev.h
index 9759f13..c0c90b7 100644
--- a/lib/librte_ethdev/rte_ethdev.h
+++ b/lib/librte_ethdev/rte_ethdev.h
@@ -3207,6 +3207,23 @@ enum rte_eth_event_type {
RTE_ETH_EVENT_DESTROY,  /**< port is released */
RTE_ETH_EVENT_IPSEC,/**< IPsec offload related event */
RTE_ETH_EVENT_FLOW_AGED,/**< New aged-out flows is detected */
+   RTE_ETH_EVENT_RESET,
+   /**< port resetting from an error
+*
+* PMD detected a FW reset or error condition.
+* PMD will try to recover from the error.
+* Data path will be halted and Control path operations
+* would fail at this time.
+*/
+   RTE_ETH_EVENT_RECOVERED,
+   /**< port recovered from an error
+*
+* PMD has recovered from the error condition.
+* Control path and Data path are up now.
+* Since the device undergone a reset, flow rules
+* offloaded prior to the reset will be lost and
+* the application has to recreate the rules again.
+*/
RTE_ETH_EVENT_MAX   /**< max value of this enum */
 };
 
-- 
2.10.1

[dpdk-dev] [RFC PATCH v4 2/3] net/bnxt: notify applications about device reset/recovery

[Kalesh A P]

From: Kalesh AP 

When the driver receives RESET_NOTIFY async event from FW or detected
an error condition, it should update the application that FW is going
to reset. Once the driver recoveres from the reset, update the reset
recovery status to application as well.

The recovery process is transparent to the application as the driver
itself tries to recover from FW reset or FW error conditions.

Signed-off-by: Kalesh AP 
Signed-off-by: Ajit Khaparde 
Reviewed-by: Somnath Kotur 
---
 drivers/net/bnxt/bnxt_cpr.c| 3 +++
 drivers/net/bnxt/bnxt_ethdev.c | 9 +
 2 files changed, 12 insertions(+)

diff --git a/drivers/net/bnxt/bnxt_cpr.c b/drivers/net/bnxt/bnxt_cpr.c
index 8311e26..3b19ef7 100644
--- a/drivers/net/bnxt/bnxt_cpr.c
+++ b/drivers/net/bnxt/bnxt_cpr.c
@@ -129,6 +129,9 @@ void bnxt_handle_async_event(struct bnxt *bp,
bp->flags |= BNXT_FLAG_FATAL_ERROR;
return;
}
+   rte_eth_dev_callback_process(bp->eth_dev,
+RTE_ETH_EVENT_RESET,
+NULL);
 
event_data = rte_le_to_cpu_32(async_cmp->event_data1);
/* timestamp_lo/hi values are in units of 100ms */
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index b99c712..9a47bfd 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -4566,6 +4566,9 @@ static void bnxt_dev_recover(void *arg)
goto err_start;
 
PMD_DRV_LOG(INFO, "Recovered from FW reset\n");
+   rte_eth_dev_callback_process(bp->eth_dev,
+RTE_ETH_EVENT_RECOVERED,
+NULL);
return;
 err_start:
bnxt_dev_stop_op(bp->eth_dev);
@@ -4573,6 +4576,9 @@ static void bnxt_dev_recover(void *arg)
bp->flags |= BNXT_FLAG_FATAL_ERROR;
bnxt_uninit_resources(bp, false);
PMD_DRV_LOG(ERR, "Failed to recover from FW reset\n");
+   rte_eth_dev_callback_process(bp->eth_dev,
+RTE_ETH_EVENT_INTR_RMV,
+NULL);
 }
 
 void bnxt_dev_reset_and_resume(void *arg)
@@ -4708,6 +4714,9 @@ static void bnxt_check_fw_health(void *arg)
bp->flags |= BNXT_FLAG_FW_RESET;
 
PMD_DRV_LOG(ERR, "Detected FW dead condition\n");
+   rte_eth_dev_callback_process(bp->eth_dev,
+RTE_ETH_EVENT_RESET,
+NULL);
 
if (bnxt_is_master_func(bp))
wait_msec = info->master_func_wait_period;
-- 
2.10.1

[dpdk-dev] [RFC PATCH v4 3/3] app/testpmd: handle device recovery event

[Kalesh A P]

From: Kalesh AP 

Added code to handle device reset and recovery event in testpmd.
This is an indication from the PMD that device has reset and
recovered error condition.

Signed-off-by: Kalesh AP 
Reviewed-by: Ajit Kumar Khaparde 
---
 app/test-pmd/testpmd.c | 6 +-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index fe6450c..1c8fb46 100644
--- a/app/test-pmd/testpmd.c
+++ b/app/test-pmd/testpmd.c
@@ -380,6 +380,8 @@ static const char * const eth_event_desc[] = {
[RTE_ETH_EVENT_NEW] = "device probed",
[RTE_ETH_EVENT_DESTROY] = "device released",
[RTE_ETH_EVENT_FLOW_AGED] = "flow aged",
+   [RTE_ETH_EVENT_RESET] = "device reset",
+   [RTE_ETH_EVENT_RECOVERED] = "device recovery",
[RTE_ETH_EVENT_MAX] = NULL,
 };
 
@@ -394,7 +396,9 @@ uint32_t event_print_mask = (UINT32_C(1) << 
RTE_ETH_EVENT_UNKNOWN) |
(UINT32_C(1) << RTE_ETH_EVENT_IPSEC) |
(UINT32_C(1) << RTE_ETH_EVENT_MACSEC) |
(UINT32_C(1) << RTE_ETH_EVENT_INTR_RMV) |
-   (UINT32_C(1) << RTE_ETH_EVENT_FLOW_AGED);
+   (UINT32_C(1) << RTE_ETH_EVENT_FLOW_AGED) |
+   (UINT32_C(1) << RTE_ETH_EVENT_RESET) |
+   (UINT32_C(1) << RTE_ETH_EVENT_RECOVERED);
 /*
  * Decide if all memory are locked for performance.
  */
-- 
2.10.1

[dpdk-dev] [PATCH] net/mlx5: fix DevX CQ attributes values

[Dekel Peled]

Previous patch wrongly used rdma-core defined values, when preparing
attributes for creating DevX CQ object.
This patch adds the correct value definition and uses them instead.

Fixes: 08d1838f645a ("net/mlx5: implement CQ for Rx using DevX API")
Cc: sta...@dpdk.org

Signed-off-by: Dekel Peled 
Acked-by: Matan Azrad 
---
 drivers/common/mlx5/mlx5_prm.h |  8 
 drivers/net/mlx5/mlx5_devx.c   | 14 --
 2 files changed, 12 insertions(+), 10 deletions(-)

diff --git a/drivers/common/mlx5/mlx5_prm.h b/drivers/common/mlx5/mlx5_prm.h
index 563e7c8..20f2fcc 100644
--- a/drivers/common/mlx5/mlx5_prm.h
+++ b/drivers/common/mlx5/mlx5_prm.h
@@ -2835,6 +2835,14 @@ struct mlx5_mini_cqe8 {
uint32_t byte_cnt;
 };
 
+/* Mini CQE responder format. */
+enum {
+   MLX5_CQE_RESP_FORMAT_HASH = 0x0,
+   MLX5_CQE_RESP_FORMAT_CSUM = 0x1,
+   MLX5_CQE_RESP_FORMAT_CSUM_FLOW_TAG = 0x2,
+   MLX5_CQE_RESP_FORMAT_CSUM_STRIDX = 0x3,
+};
+
 /* srTCM PRM flow meter parameters. */
 enum {
MLX5_FLOW_COLOR_RED = 0,
diff --git a/drivers/net/mlx5/mlx5_devx.c b/drivers/net/mlx5/mlx5_devx.c
index 3e81fcc..cb4a522 100644
--- a/drivers/net/mlx5/mlx5_devx.c
+++ b/drivers/net/mlx5/mlx5_devx.c
@@ -367,15 +367,11 @@
}
if (priv->config.cqe_comp && !rxq_data->hw_timestamp &&
!rxq_data->lro) {
-   cq_attr.cqe_comp_en = MLX5DV_CQ_INIT_ATTR_MASK_COMPRESSED_CQE;
-#ifdef HAVE_IBV_DEVICE_STRIDING_RQ_SUPPORT
+   cq_attr.cqe_comp_en = 1u;
cq_attr.mini_cqe_res_format =
mlx5_rxq_mprq_enabled(rxq_data) ?
-   MLX5DV_CQE_RES_FORMAT_CSUM_STRIDX :
-   MLX5DV_CQE_RES_FORMAT_HASH;
-#else
-   cq_attr.mini_cqe_res_format = MLX5DV_CQE_RES_FORMAT_HASH;
-#endif
+   MLX5_CQE_RESP_FORMAT_CSUM_STRIDX :
+   MLX5_CQE_RESP_FORMAT_HASH;
/*
 * For vectorized Rx, it must not be doubled in order to
 * make cq_ci and rq_ci aligned.
@@ -392,10 +388,8 @@
"Port %u Rx CQE compression is disabled for LRO.",
dev->data->port_id);
}
-#ifdef HAVE_IBV_MLX5_MOD_CQE_128B_PAD
if (priv->config.cqe_pad)
-   cq_attr.cqe_size = MLX5DV_CQ_INIT_ATTR_FLAGS_CQE_PAD;
-#endif
+   cq_attr.cqe_size = MLX5_CQE_SIZE_128B;
log_cqe_n = log2above(cqe_n);
cq_size = sizeof(struct mlx5_cqe) * (1 << log_cqe_n);
/* Query the EQN for this core. */
-- 
1.8.3.1

Re: [dpdk-dev] [PATCH v3 00/29] cleanup ethdev close operation

[Ferruh Yigit]

On 9/29/2020 12:14 AM, Thomas Monjalon wrote:

This is the end of a process started two years ago,
to have a close which reliably releases an ethdev port
without the need of removing the device (which can have more ports).

Unfortunately, some drivers might be broken because did not follow
the migration recommendations. We cannot wait more,
this should be merged before the 20.11-rc1 release.


v3:
   - integrate NXP patches for dpaa, dpaa2, enetc and pfe
   - fix crash in rte_eth_dev_release_port() because of early memset 0

v2 (big thanks to Ferruh for the reviews):
   - rebase on top of cxgbe updates
   - integrate updated patches for tap, bnx2x, qede
   - return 0 in atlantic PMD
   - fix failsafe mutex free
   - remove useless check in pcap
   - fix testpmd port state transition
   - add primary process check
   - reset more pointers in rte_eth_dev_release_port()
   - remove dev_started reset in rte_eth_dev_close()
   - return int from rte_eth_dev_close()

Note for future: fix old memory leak when releasing port,
i.e. free callbacks link_intr_cbs, post_rx_burst_cbs, pre_tx_burst_cbs.

The last 3 patches are optional changes to ethdev close/release
which were requested by Ferruh but may take longer to be accepted:
   - memset 0
   - keep start state
   - int return


Rasesh Mody (2):
   net/bnx2x: release port upon close
   net/qede: release port upon close

Sachin Saxena (4):
   net/dpaa: release port upon close
   net/dpaa2: release port upon close
   net/enetc: release port upon close
   net/pfe: release port upon close

Thomas Monjalon (22):
   ethdev: reset device and interrupt pointers on release
   ethdev: allow drivers to return error on close
   net/af_packet: release port upon close
   net/atlantic: release port upon close
   net/axgbe: release port upon close
   net/bonding: release port upon close
   net/failsafe: release port upon close
   net/mlx4: release port upon close
   net/null: release port upon close
   net/octeontx: release port upon close
   net/pcap: release port upon close
   net/ring: release port upon close
   net/softnic: release port upon close
   ethdev: remove old close behaviour
   drivers/net: accept removing device without any port
   drivers/net: check process type in close operation
   drivers/net: remove redundant MAC addresses freeing
   app/testpmd: reset port status on close notification
   app/testpmd: align behaviour of multi-port detach
   ethdev: remove forcing stopped state upon close
   ethdev: reset all when releasing a port
   ethdev: allow close function to return an error

Yunjian Wang (1):
   net/tap: release port upon close




Since some discussions going on them last three patches not merged, they can be 
sent separately.


Except from last three patches, 27/29, 28/29, 29/29,
Series applied to dpdk-next-net/main, thanks.

Re: [dpdk-dev] [RFC PATCH v3 1/3] ethdev: support device reset and recovery events

[Kalesh Anakkur Purayil]

Hi Thomas,

Please see my response inline.

Regards,
Kalesh

On Wed, Sep 30, 2020 at 1:21 PM Thomas Monjalon  wrote:

> Hi,
>
> Please use --cc-cmd devtools/get-maintainer.sh so all relevant
> maintainers are Cc'ed. Adding Andrew.
>
[Kalesh]: Thank you. Sure, I will follow next time.

>
> > From: Kalesh AP 
> >
> > Adding support for device reset and recovery events in the
> > rte_eth_event framework. FW error and FW reset conditions would be
> > managed internally by PMD without needing application intervention.
> > In such cases, PMD would need reset/recovery events to notify application
> > that PMD is undergoing a reset.
>
> We already have this event:
>
> RTE_ETH_EVENT_INTR_RESET,
> /**< reset interrupt event, sent to VF on PF reset */
>
> I don't know why "INTR" is in the name of this event,
> and I think it does not need to be restricted to VF.
> The application does not need to know whether the reset
> is caused by the PF, the FW or the HW.
> That's why I think you could share the same event.
>

[Kalesh]: Yes. As you mentioned, this event is used for some other purpose.
I did not want to break the existing usage/purpose of this event.
For example, upon receiving the RTE_ETH_EVENT_INTR_RESET event OVS
application invokes rte_eth_dev_reset() to reset the port.
The aim here is to recover from the device error condition without the
intervention of Applications. PMD itself will recover from the error using
the protocol with FW.

>
> > +   RTE_ETH_EVENT_RESET,/**< port resetting from an error */
> > +   RTE_ETH_EVENT_RECOVERED, /**< port recovered from an error */
>
> You ignored my previous comments:
> "
> What the application is supposed to do when receiving such event?
> How the application knows that flow rules were resetted?
> Is there any other configuration resetted?
> These informations must be explicit in the doxygen comments.
> "
>
[Kalesh]: Sorry, I missed it.
I am not sure what you meant by "These information must be explicit in the
doxygen comments ".
Could you please elaborate a little how to/where to put these details?


-- 
Regards,
Kalesh A P

Re: [dpdk-dev] [PATCH v4] net/mlx5: relaxed ordering for multi-packet RQ buffer refcnt

[Slava Ovsiienko]

Looks good to me, and we've checked the performance has no impact.
Thank you.

Acked-by: Viacheslav Ovsiienko 

> -Original Message-
> From: dev  On Behalf Of Phil Yang
> Sent: Tuesday, September 29, 2020 18:23
> To: Raslan Darawsheh ; Matan Azrad
> ; Shahaf Shuler 
> Cc: nd ; Alexander Kozyrev ;
> Honnappa Nagarahalli ; dev@dpdk.org;
> nd 
> Subject: Re: [dpdk-dev] [PATCH v4] net/mlx5: relaxed ordering for multi-
> packet RQ buffer refcnt
> 
> Hi Raslan,
> 
> It seems that there are no more comments for this patch.
> So shall we proceed further?
> 
> Thanks,
> Phil Yang
> 
> > -Original Message-
> > From: Alexander Kozyrev 
> > Sent: Thursday, September 10, 2020 9:37 AM
> > To: Honnappa Nagarahalli ; Phil Yang
> > ; akozy...@mellanox.com; rasl...@mellanox.com;
> > dev@dpdk.org
> > Cc: Phil Yang ; ma...@mellanox.com; Shahaf Shuler
> > ; viachesl...@mellanox.com; nd
> ; nd
> > 
> > Subject: RE: [PATCH v4] net/mlx5: relaxed ordering for multi-packet RQ
> > buffer refcnt
> >
> > > 
> > >
> > > >
> > > > Use c11 atomics with RELAXED ordering instead of the rte_atomic
> > > > ops which enforce unnecessary barriers on aarch64.
> > > >
> > > > Signed-off-by: Phil Yang 
> > > Looks good.
> > >
> > > Reviewed-by: Honnappa Nagarahalli 
> >
> > Acked-by: Alexander Kozyrev 
> >
> > >
> > > > ---
> > > > v4:
> > > > Remove the unnecessary ACQUIRE barrier in rx burst path.
> > > > (Honnappa)
> > > >
> > > > v3:
> > > > Split from the patchset:
> > > >
> >
> https://nam11.safelinks.protection.outlook.com/?url=http%3A%2F%2Fpatch
> > > >
> > >
> >
> work.dpdk.org%2Fcover%2F68159%2F&data=02%7C01%7Cakozyrev%4
> 0
> > nv
> > > idia.
> > > >
> > >
> >
> com%7Cf16ba4e8cfb145f5d82008d85529348e%7C43083d15727340c1b7db3
> 9e
> > f
> > > d9ccc
> > > >
> > >
> >
> 17a%7C0%7C0%7C637352982762038088&sdata=0HzTxbzh0Dqk0hZ5PI
> gE
> > V
> > > zieyV%
> > > > 2BnLTivsVIFFxXFAtI%3D&reserved=0
> > > >
> > > >  drivers/net/mlx5/mlx5_rxq.c  |  2 +-
> > > > drivers/net/mlx5/mlx5_rxtx.c | 16 +---
> > > > drivers/net/mlx5/mlx5_rxtx.h |  2 +-
> > > >  3 files changed, 11 insertions(+), 9 deletions(-)
> > > >
> > > > diff --git a/drivers/net/mlx5/mlx5_rxq.c
> > > > b/drivers/net/mlx5/mlx5_rxq.c index
> > > > 79eb8f8..40e0239 100644
> > > > --- a/drivers/net/mlx5/mlx5_rxq.c
> > > > +++ b/drivers/net/mlx5/mlx5_rxq.c
> > > > @@ -2012,7 +2012,7 @@ mlx5_mprq_buf_init(struct rte_mempool
> *mp,
> > void
> > > > *opaque_arg,
> > > >
> > > > memset(_m, 0, sizeof(*buf));
> > > > buf->mp = mp;
> > > > -   rte_atomic16_set(&buf->refcnt, 1);
> > > > +   __atomic_store_n(&buf->refcnt, 1, __ATOMIC_RELAXED);
> > > > for (j = 0; j != strd_n; ++j) {
> > > > shinfo = &buf->shinfos[j];
> > > > shinfo->free_cb = mlx5_mprq_buf_free_cb; diff --git
> > > > a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c
> > > > index 1b71e94..549477b 100644
> > > > --- a/drivers/net/mlx5/mlx5_rxtx.c
> > > > +++ b/drivers/net/mlx5/mlx5_rxtx.c
> > > > @@ -1626,10 +1626,11 @@ mlx5_mprq_buf_free_cb(void *addr
> > > __rte_unused,
> > > > void *opaque)  {
> > > > struct mlx5_mprq_buf *buf = opaque;
> > > >
> > > > -   if (rte_atomic16_read(&buf->refcnt) == 1) {
> > > > +   if (__atomic_load_n(&buf->refcnt, __ATOMIC_RELAXED) == 1) {
> > > > rte_mempool_put(buf->mp, buf);
> > > > -   } else if (rte_atomic16_add_return(&buf->refcnt, -1) == 0) {
> > > > -   rte_atomic16_set(&buf->refcnt, 1);
> > > > +   } else if (unlikely(__atomic_sub_fetch(&buf->refcnt, 1,
> > > > +  __ATOMIC_RELAXED) == 0)) 
> > > > {
> > > > +   __atomic_store_n(&buf->refcnt, 1, __ATOMIC_RELAXED);
> > > > rte_mempool_put(buf->mp, buf);
> > > > }
> > > >  }
> > > > @@ -1709,7 +1710,8 @@ mlx5_rx_burst_mprq(void *dpdk_rxq, struct
> > > > rte_mbuf **pkts, uint16_t pkts_n)
> > > >
> > > > if (consumed_strd == strd_n) {
> > > > /* Replace WQE only if the buffer is still in 
> > > > use. */
> > > > -   if (rte_atomic16_read(&buf->refcnt) > 1) {
> > > > +   if (__atomic_load_n(&buf->refcnt,
> > > > +   __ATOMIC_RELAXED) > 1) {
> > > > mprq_buf_replace(rxq, rq_ci & wq_mask,
> > > strd_n);
> > > > /* Release the old buffer. */
> > > > mlx5_mprq_buf_free(buf);
> > > > @@ -1821,9 +1823,9 @@ mlx5_rx_burst_mprq(void *dpdk_rxq, struct
> > > > rte_mbuf **pkts, uint16_t pkts_n)
> > > > void *buf_addr;
> > > >
> > > > /* Increment the refcnt of the whole chunk. */
> > > > -   rte_atomic16_add_return(&buf->refcnt, 1);
> > > > -   MLX5_ASSERT((uint16_t)rte_atomic16_read(&buf-
> > > > >refcnt) <=
> > > > -

Re: [dpdk-dev] [PATCH v4] net/mlx5: relaxed ordering for multi-packet RQ buffer refcnt

[Raslan Darawsheh]

Sure, I'll take the patch since I was waiting for performance impact testing, 
and since Slava just confirmed it, 
Then we can take it yes thank you for your contribution. 

Kindest regards,
Raslan Darawsheh

> -Original Message-
> From: Slava Ovsiienko 
> Sent: Wednesday, September 30, 2020 3:45 PM
> To: Phil Yang ; Raslan Darawsheh
> ; Matan Azrad ; Shahaf Shuler
> 
> Cc: nd ; Alexander Kozyrev ;
> Honnappa Nagarahalli ; dev@dpdk.org;
> nd 
> Subject: RE: [PATCH v4] net/mlx5: relaxed ordering for multi-packet RQ
> buffer refcnt
> 
> Looks good to me, and we've checked the performance has no impact.
> Thank you.
> 
> Acked-by: Viacheslav Ovsiienko 
> 
> > -Original Message-
> > From: dev  On Behalf Of Phil Yang
> > Sent: Tuesday, September 29, 2020 18:23
> > To: Raslan Darawsheh ; Matan Azrad
> > ; Shahaf Shuler 
> > Cc: nd ; Alexander Kozyrev ;
> > Honnappa Nagarahalli ; dev@dpdk.org;
> > nd 
> > Subject: Re: [dpdk-dev] [PATCH v4] net/mlx5: relaxed ordering for multi-
> > packet RQ buffer refcnt
> >
> > Hi Raslan,
> >
> > It seems that there are no more comments for this patch.
> > So shall we proceed further?
> >
> > Thanks,
> > Phil Yang
> >
> > > -Original Message-
> > > From: Alexander Kozyrev 
> > > Sent: Thursday, September 10, 2020 9:37 AM
> > > To: Honnappa Nagarahalli ; Phil Yang
> > > ; akozy...@mellanox.com;
> rasl...@mellanox.com;
> > > dev@dpdk.org
> > > Cc: Phil Yang ; ma...@mellanox.com; Shahaf
> Shuler
> > > ; viachesl...@mellanox.com; nd
> > ; nd
> > > 
> > > Subject: RE: [PATCH v4] net/mlx5: relaxed ordering for multi-packet RQ
> > > buffer refcnt
> > >
> > > > 
> > > >
> > > > >
> > > > > Use c11 atomics with RELAXED ordering instead of the rte_atomic
> > > > > ops which enforce unnecessary barriers on aarch64.
> > > > >
> > > > > Signed-off-by: Phil Yang 
> > > > Looks good.
> > > >
> > > > Reviewed-by: Honnappa Nagarahalli
> 
> > >
> > > Acked-by: Alexander Kozyrev 
> > >
> > > >
> > > > > ---
> > > > > v4:
> > > > > Remove the unnecessary ACQUIRE barrier in rx burst path.
> > > > > (Honnappa)
> > > > >
> > > > > v3:
> > > > > Split from the patchset:
> > > > >
> > >
> >
> https://nam11.safelinks.protection.outlook.com/?url=http%3A%2F%2Fpatch
> > > > >
> > > >
> > >
> >
> work.dpdk.org%2Fcover%2F68159%2F&data=02%7C01%7Cakozyrev%4
> > 0
> > > nv
> > > > idia.
> > > > >
> > > >
> > >
> > com%7Cf16ba4e8cfb145f5d82008d85529348e%7C43083d15727340c1b7db3
> > 9e
> > > f
> > > > d9ccc
> > > > >
> > > >
> > >
> > 17a%7C0%7C0%7C637352982762038088&sdata=0HzTxbzh0Dqk0hZ5PI
> > gE
> > > V
> > > > zieyV%
> > > > > 2BnLTivsVIFFxXFAtI%3D&reserved=0
> > > > >
> > > > >  drivers/net/mlx5/mlx5_rxq.c  |  2 +-
> > > > > drivers/net/mlx5/mlx5_rxtx.c | 16 +---
> > > > > drivers/net/mlx5/mlx5_rxtx.h |  2 +-
> > > > >  3 files changed, 11 insertions(+), 9 deletions(-)
> > > > >
> > > > > diff --git a/drivers/net/mlx5/mlx5_rxq.c
> > > > > b/drivers/net/mlx5/mlx5_rxq.c index
> > > > > 79eb8f8..40e0239 100644
> > > > > --- a/drivers/net/mlx5/mlx5_rxq.c
> > > > > +++ b/drivers/net/mlx5/mlx5_rxq.c
> > > > > @@ -2012,7 +2012,7 @@ mlx5_mprq_buf_init(struct rte_mempool
> > *mp,
> > > void
> > > > > *opaque_arg,
> > > > >
> > > > >   memset(_m, 0, sizeof(*buf));
> > > > >   buf->mp = mp;
> > > > > - rte_atomic16_set(&buf->refcnt, 1);
> > > > > + __atomic_store_n(&buf->refcnt, 1, __ATOMIC_RELAXED);
> > > > >   for (j = 0; j != strd_n; ++j) {
> > > > >   shinfo = &buf->shinfos[j];
> > > > >   shinfo->free_cb = mlx5_mprq_buf_free_cb; diff --git
> > > > > a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c
> > > > > index 1b71e94..549477b 100644
> > > > > --- a/drivers/net/mlx5/mlx5_rxtx.c
> > > > > +++ b/drivers/net/mlx5/mlx5_rxtx.c
> > > > > @@ -1626,10 +1626,11 @@ mlx5_mprq_buf_free_cb(void *addr
> > > > __rte_unused,
> > > > > void *opaque)  {
> > > > >   struct mlx5_mprq_buf *buf = opaque;
> > > > >
> > > > > - if (rte_atomic16_read(&buf->refcnt) == 1) {
> > > > > + if (__atomic_load_n(&buf->refcnt, __ATOMIC_RELAXED) ==
> 1) {
> > > > >   rte_mempool_put(buf->mp, buf);
> > > > > - } else if (rte_atomic16_add_return(&buf->refcnt, -1) == 0) {
> > > > > - rte_atomic16_set(&buf->refcnt, 1);
> > > > > + } else if (unlikely(__atomic_sub_fetch(&buf->refcnt, 1,
> > > > > +__ATOMIC_RELAXED) ==
> 0)) {
> > > > > + __atomic_store_n(&buf->refcnt, 1,
> __ATOMIC_RELAXED);
> > > > >   rte_mempool_put(buf->mp, buf);
> > > > >   }
> > > > >  }
> > > > > @@ -1709,7 +1710,8 @@ mlx5_rx_burst_mprq(void *dpdk_rxq,
> struct
> > > > > rte_mbuf **pkts, uint16_t pkts_n)
> > > > >
> > > > >   if (consumed_strd == strd_n) {
> > > > >   /* Replace WQE only if the buffer is still in
> use. */
> > > > > - if (rte_atomic16_read(&buf->refcnt) > 1) {
> > > > > + if (__atomic_lo

Re: [dpdk-dev] [PATCH v4] app/testpmd: fix the default RSS key configuration

[Ferruh Yigit]

On 9/24/2020 2:45 PM, Lijun Ou wrote:

It use the NIC valid default RSS key instead of the testpmd
dummy RSS key in the flow configuration when the RSS key is
not specified in the flow rule. If the NIC RSS key is
invalid, it will use testpmd dummy RSS key as the default
key.


Can you please describe the impact, what fails without this fix?



Fixes: ac8d22de2394 ("ethdev: flatten RSS configuration in flow API")
Cc: sta...@dpdk.org

Signed-off-by: Lijun Ou 
---
V3->V4:
-fix checkpatch warning and shorter commit content.

V2->V3:
-fix checkpatch warning.

V1->V2:
-fix the commit.
---
  app/test-pmd/cmdline_flow.c | 8 
  1 file changed, 8 insertions(+)

diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c
index 6263d30..e6648da 100644
--- a/app/test-pmd/cmdline_flow.c
+++ b/app/test-pmd/cmdline_flow.c
@@ -4312,6 +4312,7 @@ parse_vc_action_rss(struct context *ctx, const struct 
token *token,
action_rss_data->queue[i] = i;
if (!port_id_is_invalid(ctx->port, DISABLED_WARN) &&
ctx->port != (portid_t)RTE_PORT_ALL) {
+   struct rte_eth_rss_conf rss_conf = {0};
struct rte_eth_dev_info info;
int ret2;
  
@@ -4322,6 +4323,13 @@ parse_vc_action_rss(struct context *ctx, const struct token *token,

action_rss_data->conf.key_len =
RTE_MIN(sizeof(action_rss_data->key),
info.hash_key_size);
+
+   rss_conf.rss_key_len = sizeof(action_rss_data->key);
+   rss_conf.rss_key = action_rss_data->key;


'rss_conf.rss_key_len' is the input parameter, it looks like it has been used as 
the size of the 'rss_key', but as far as I can see it is not.


Because of this if 'info.hash_key_size' is bigger than 
'sizeof(action_rss_data->key)', won't PMD overwrite the stack for the remaining 
bytes?


Can you please check?


+   ret2 = rte_eth_dev_rss_hash_conf_get(ctx->port, &rss_conf);
+   if (ret2 != 0)
+   return ret2;


If PMD not implemented the 'rss_hash_conf_get' dev_ops, should it fail the RSS 
action?



+   action_rss_data->conf.key = rss_conf.rss_key;
}
action->conf = &action_rss_data->conf;
return ret;

[dpdk-dev] [PATCH v3 00/18] add max SIMD bitwidth to EAL

[Ciara Power]

A number of components in DPDK have optional AVX-512 or other vector
code paths which can be selected at runtime. Rather than having each
component provide its own mechanism to select a code path, this patchset
adds support for a single setting to control what code paths are used.
This can be used to enable some non-default code paths e.g. ones using
AVX-512, but also to limit the code paths to certain vector widths, or
to scalar code only, which is useful for testing.

The max SIMD bitwidth setting can be set by the app itself through use of
the available API, or can be overriden by a commandline argument passed by
the user.

v3:
 - Added patch to add check for LPM lib
 - Modified default max bitwidth for Arm to disable max SIMD bitwidth,
   which will allow for SVE.
 - Added "0" as an acceptable value for command-line flag, which internally
   is used as UINT16_MAX to essentially disable max SIMD bitwidth limits.
 - Made suggested changes to net lib patch.
 - Rebased onto main.
v2:
  - Added some documentation.
  - Modified default max bitwidth for Arm.
  - Moved mlx5 condition check into existing check vec support function.
  - Added max SIMD bitwidth checks to some libraries.

Ciara Power (18):
  eal: add max SIMD bitwidth
  eal: add default SIMD bitwidth values
  doc: add detail on using max SIMD bitwidth
  net/i40e: add checks for max SIMD bitwidth
  net/axgbe: add checks for max SIMD bitwidth
  net/bnxt: add checks for max SIMD bitwidth
  net/enic: add checks for max SIMD bitwidth
  net/fm10k: add checks for max SIMD bitwidth
  net/iavf: add checks for max SIMD bitwidth
  net/ice: add checks for max SIMD bitwidth
  net/ixgbe: add checks for max SIMD bitwidth
  net/mlx5: add checks for max SIMD bitwidth
  net/virtio: add checks for max SIMD bitwidth
  distributor: add checks for max SIMD bitwidth
  member: add checks for max SIMD bitwidth
  efd: add checks for max SIMD bitwidth
  net: add checks for max SIMD bitwidth
  lpm: choose vector path at runtime

 doc/guides/howto/avx512.rst   | 36 ++
 doc/guides/howto/index.rst|  1 +
 doc/guides/linux_gsg/eal_args.include.rst | 16 +
 .../prog_guide/env_abstraction_layer.rst  | 32 +
 drivers/net/axgbe/axgbe_rxtx.c|  3 +-
 drivers/net/bnxt/bnxt_ethdev.c|  6 +-
 drivers/net/enic/enic_rxtx_vec_avx2.c |  3 +-
 drivers/net/fm10k/fm10k_ethdev.c  | 11 ++-
 drivers/net/i40e/i40e_rxtx.c  | 19 --
 drivers/net/iavf/iavf_rxtx.c  | 16 +++--
 drivers/net/ice/ice_rxtx.c| 20 --
 drivers/net/ixgbe/ixgbe_rxtx.c|  7 +-
 drivers/net/mlx5/mlx5_rxtx_vec.c  |  2 +
 drivers/net/virtio/virtio_ethdev.c|  9 ++-
 lib/librte_distributor/rte_distributor.c  |  3 +-
 lib/librte_eal/arm/include/rte_vect.h |  2 +
 lib/librte_eal/common/eal_common_options.c| 67 +++
 lib/librte_eal/common/eal_internal_cfg.h  |  8 +++
 lib/librte_eal/common/eal_options.h   |  2 +
 lib/librte_eal/include/rte_eal.h  | 33 +
 lib/librte_eal/ppc/include/rte_vect.h |  2 +
 lib/librte_eal/rte_eal_version.map|  4 ++
 lib/librte_eal/x86/include/rte_vect.h |  2 +
 lib/librte_efd/rte_efd.c  |  7 +-
 lib/librte_lpm/rte_lpm.h  | 57 +---
 lib/librte_lpm/rte_lpm_altivec.h  |  2 +-
 lib/librte_lpm/rte_lpm_neon.h |  2 +-
 lib/librte_lpm/rte_lpm_sse.h  |  2 +-
 lib/librte_member/rte_member_ht.c |  3 +-
 lib/librte_net/rte_net_crc.c  | 26 ---
 lib/librte_net/rte_net_crc.h  |  3 +-
 31 files changed, 351 insertions(+), 55 deletions(-)
 create mode 100644 doc/guides/howto/avx512.rst

-- 
2.17.1

[dpdk-dev] [PATCH v3 01/18] eal: add max SIMD bitwidth

[Ciara Power]

This patch adds a max SIMD bitwidth EAL configuration. The API allows
for an app to set this value. It can also be set using EAL argument
--force-max-simd-bitwidth, which will lock the value and override any
modifications made by the app.

Signed-off-by: Ciara Power 

---
v3:
  - Added enum value to essentially disable using max SIMD to choose
paths, intended for use by ARM SVE.
  - Fixed parsing bitwidth argument to return an error for values
greater than uint16_t.
v2: Added to Doxygen comment for API.
---
 lib/librte_eal/common/eal_common_options.c | 64 ++
 lib/librte_eal/common/eal_internal_cfg.h   |  8 +++
 lib/librte_eal/common/eal_options.h|  2 +
 lib/librte_eal/include/rte_eal.h   | 33 +++
 lib/librte_eal/rte_eal_version.map |  4 ++
 5 files changed, 111 insertions(+)

diff --git a/lib/librte_eal/common/eal_common_options.c 
b/lib/librte_eal/common/eal_common_options.c
index a5426e1234..e9117a96af 100644
--- a/lib/librte_eal/common/eal_common_options.c
+++ b/lib/librte_eal/common/eal_common_options.c
@@ -102,6 +102,7 @@ eal_long_options[] = {
{OPT_MATCH_ALLOCATIONS, 0, NULL, OPT_MATCH_ALLOCATIONS_NUM},
{OPT_TELEMETRY, 0, NULL, OPT_TELEMETRY_NUM},
{OPT_NO_TELEMETRY,  0, NULL, OPT_NO_TELEMETRY_NUM },
+   {OPT_FORCE_MAX_SIMD_BITWIDTH, 1, NULL, OPT_FORCE_MAX_SIMD_BITWIDTH_NUM},
{0, 0, NULL, 0}
 };
 
@@ -1309,6 +1310,34 @@ eal_parse_iova_mode(const char *name)
return 0;
 }
 
+static int
+eal_parse_simd_bitwidth(const char *arg, bool locked)
+{
+   char *end;
+   unsigned long bitwidth;
+   int ret;
+   struct internal_config *internal_conf =
+   eal_get_internal_configuration();
+
+   if (arg == NULL || arg[0] == '\0')
+   return -1;
+
+   errno = 0;
+   bitwidth = strtoul(arg, &end, 0);
+
+   /* check for errors */
+   if (bitwidth > UINT16_MAX || errno != 0 || end == NULL || *end != '\0')
+   return -1;
+
+   if (bitwidth == 0)
+   bitwidth = UINT16_MAX;
+   ret = rte_set_max_simd_bitwidth(bitwidth);
+   if (ret < 0)
+   return -1;
+   internal_conf->max_simd_bitwidth.locked = locked;
+   return 0;
+}
+
 static int
 eal_parse_base_virtaddr(const char *arg)
 {
@@ -1707,6 +1736,13 @@ eal_parse_common_option(int opt, const char *optarg,
case OPT_NO_TELEMETRY_NUM:
conf->no_telemetry = 1;
break;
+   case OPT_FORCE_MAX_SIMD_BITWIDTH_NUM:
+   if (eal_parse_simd_bitwidth(optarg, 1) < 0) {
+   RTE_LOG(ERR, EAL, "invalid parameter for --"
+   OPT_FORCE_MAX_SIMD_BITWIDTH "\n");
+   return -1;
+   }
+   break;
 
/* don't know what to do, leave this to caller */
default:
@@ -1903,6 +1939,33 @@ eal_check_common_options(struct internal_config 
*internal_cfg)
return 0;
 }
 
+uint16_t
+rte_get_max_simd_bitwidth(void)
+{
+   const struct internal_config *internal_conf =
+   eal_get_internal_configuration();
+   return internal_conf->max_simd_bitwidth.bitwidth;
+}
+
+int
+rte_set_max_simd_bitwidth(uint16_t bitwidth)
+{
+   struct internal_config *internal_conf =
+   eal_get_internal_configuration();
+   if (internal_conf->max_simd_bitwidth.locked) {
+   RTE_LOG(NOTICE, EAL, "Cannot set max SIMD bitwidth - user 
runtime override enabled");
+   return -EPERM;
+   }
+
+   if (bitwidth != RTE_MAX_SIMD_DISABLE && (bitwidth < RTE_NO_SIMD ||
+   !rte_is_power_of_2(bitwidth))) {
+   RTE_LOG(ERR, EAL, "Invalid bitwidth value!\n");
+   return -EINVAL;
+   }
+   internal_conf->max_simd_bitwidth.bitwidth = bitwidth;
+   return 0;
+}
+
 void
 eal_common_usage(void)
 {
@@ -1981,6 +2044,7 @@ eal_common_usage(void)
   "  --"OPT_BASE_VIRTADDR" Base virtual address\n"
   "  --"OPT_TELEMETRY"   Enable telemetry support (on by 
default)\n"
   "  --"OPT_NO_TELEMETRY"   Disable telemetry support\n"
+  "  --"OPT_FORCE_MAX_SIMD_BITWIDTH" Force the max SIMD bitwidth\n"
   "\nEAL options for DEBUG use only:\n"
   "  --"OPT_HUGE_UNLINK"   Unlink hugepage files after init\n"
   "  --"OPT_NO_HUGE"   Use malloc instead of hugetlbfs\n"
diff --git a/lib/librte_eal/common/eal_internal_cfg.h 
b/lib/librte_eal/common/eal_internal_cfg.h
index 13f93388a7..367e0cc19e 100644
--- a/lib/librte_eal/common/eal_internal_cfg.h
+++ b/lib/librte_eal/common/eal_internal_cfg.h
@@ -33,6 +33,12 @@ struct hugepage_info {
int lock_descriptor;/**< file descriptor for hugepage dir */
 };
 
+struct simd_bitwidth {
+   /**< flag indicating if bitwidth is locked from further modi

[dpdk-dev] [PATCH v3 02/18] eal: add default SIMD bitwidth values

[Ciara Power]

Each arch has a define for the default SIMD bitwidth value, this is used
on EAL init to set the config max SIMD bitwidth.

Cc: Ruifeng Wang 
Cc: Jerin Jacob 
Cc: Honnappa Nagarahalli 
Cc: David Christensen 

Signed-off-by: Ciara Power 

---
v3:
  - Removed unnecessary define in generic rte_vect.h
  - Changed default bitwidth for ARM to UINT16_MAX, to allow for SVE.
v2: Changed default bitwidth for Arm to 128.
---
 lib/librte_eal/arm/include/rte_vect.h  | 2 ++
 lib/librte_eal/common/eal_common_options.c | 3 +++
 lib/librte_eal/ppc/include/rte_vect.h  | 2 ++
 lib/librte_eal/x86/include/rte_vect.h  | 2 ++
 4 files changed, 9 insertions(+)

diff --git a/lib/librte_eal/arm/include/rte_vect.h 
b/lib/librte_eal/arm/include/rte_vect.h
index 01c51712a1..a3508e69d5 100644
--- a/lib/librte_eal/arm/include/rte_vect.h
+++ b/lib/librte_eal/arm/include/rte_vect.h
@@ -14,6 +14,8 @@
 extern "C" {
 #endif
 
+#define RTE_DEFAULT_SIMD_BITWIDTH UINT16_MAX
+
 typedef int32x4_t xmm_t;
 
 #defineXMM_SIZE(sizeof(xmm_t))
diff --git a/lib/librte_eal/common/eal_common_options.c 
b/lib/librte_eal/common/eal_common_options.c
index e9117a96af..d412cae89b 100644
--- a/lib/librte_eal/common/eal_common_options.c
+++ b/lib/librte_eal/common/eal_common_options.c
@@ -35,6 +35,7 @@
 #ifndef RTE_EXEC_ENV_WINDOWS
 #include 
 #endif
+#include 
 
 #include "eal_internal_cfg.h"
 #include "eal_options.h"
@@ -344,6 +345,8 @@ eal_reset_internal_config(struct internal_config 
*internal_cfg)
internal_cfg->user_mbuf_pool_ops_name = NULL;
CPU_ZERO(&internal_cfg->ctrl_cpuset);
internal_cfg->init_complete = 0;
+   internal_cfg->max_simd_bitwidth.bitwidth = RTE_DEFAULT_SIMD_BITWIDTH;
+   internal_cfg->max_simd_bitwidth.locked = 0;
 }
 
 static int
diff --git a/lib/librte_eal/ppc/include/rte_vect.h 
b/lib/librte_eal/ppc/include/rte_vect.h
index b0545c878c..70fbd0c423 100644
--- a/lib/librte_eal/ppc/include/rte_vect.h
+++ b/lib/librte_eal/ppc/include/rte_vect.h
@@ -15,6 +15,8 @@
 extern "C" {
 #endif
 
+#define RTE_DEFAULT_SIMD_BITWIDTH 256
+
 typedef vector signed int xmm_t;
 
 #defineXMM_SIZE(sizeof(xmm_t))
diff --git a/lib/librte_eal/x86/include/rte_vect.h 
b/lib/librte_eal/x86/include/rte_vect.h
index df5a607623..b1df75aca7 100644
--- a/lib/librte_eal/x86/include/rte_vect.h
+++ b/lib/librte_eal/x86/include/rte_vect.h
@@ -35,6 +35,8 @@
 extern "C" {
 #endif
 
+#define RTE_DEFAULT_SIMD_BITWIDTH 256
+
 typedef __m128i xmm_t;
 
 #defineXMM_SIZE(sizeof(xmm_t))
-- 
2.17.1

[dpdk-dev] [PATCH v3 03/18] doc: add detail on using max SIMD bitwidth

[Ciara Power]

This patch adds documentation on the usage of the max SIMD bitwidth EAL
setting, and how to use it to enable AVX-512 at runtime.

Cc: Anatoly Burakov 
Cc: John McNamara 
Cc: Marko Kovacevic 

Signed-off-by: Ciara Power 

---
v3:
  - Added enum value for disabling use of max SIMD to doc.
  - Added entry to HowTo index.
---
 doc/guides/howto/avx512.rst   | 36 +++
 doc/guides/howto/index.rst|  1 +
 doc/guides/linux_gsg/eal_args.include.rst | 16 +
 .../prog_guide/env_abstraction_layer.rst  | 32 +
 4 files changed, 85 insertions(+)
 create mode 100644 doc/guides/howto/avx512.rst

diff --git a/doc/guides/howto/avx512.rst b/doc/guides/howto/avx512.rst
new file mode 100644
index 00..ebae0f2b4f
--- /dev/null
+++ b/doc/guides/howto/avx512.rst
@@ -0,0 +1,36 @@
+..  SPDX-License-Identifier: BSD-3-Clause
+Copyright(c) 2020 Intel Corporation.
+
+
+Using AVX-512 with DPDK
+===
+
+AVX-512 is not used by default in DPDK, but it can be selected at runtime by 
apps through the use of EAL API,
+and by the user with a commandline argument. DPDK has a setting for max SIMD 
bitwidth,
+which can be modified and will then limit the vector path taken by the code.
+
+
+Using the API in apps
+-
+
+Apps can request DPDK uses AVX-512 at runtime, if it provides improved 
application performance.
+This can be done by modifying the EAL setting for max SIMD bitwidth to 512, as 
by default it is 256,
+which does not allow for AVX-512.
+
+.. code-block:: c
+
+   rte_set_max_simd_bitwidth(RTE_MAX_512_SIMD);
+
+This API should only be called once at initialization, before EAL init.
+For more information on the possible enum values to use as a parameter, go to 
:ref:`max_simd_bitwidth`:
+
+
+Using the command-line argument
+-
+
+The user can select to use AVX-512 at runtime, using the following argument to 
set the max bitwidth::
+
+   ./app/dpdk-testpmd --force-max-simd-bitwidth=512
+
+This will override any further changes to the max SIMD bitwidth in DPDK,
+which is useful for testing purposes.
diff --git a/doc/guides/howto/index.rst b/doc/guides/howto/index.rst
index 5a97ea508c..c2a2c60ddb 100644
--- a/doc/guides/howto/index.rst
+++ b/doc/guides/howto/index.rst
@@ -20,3 +20,4 @@ HowTo Guides
 telemetry
 debug_troubleshoot
 openwrt
+avx512
diff --git a/doc/guides/linux_gsg/eal_args.include.rst 
b/doc/guides/linux_gsg/eal_args.include.rst
index 0fe4457968..a0bfbd1a98 100644
--- a/doc/guides/linux_gsg/eal_args.include.rst
+++ b/doc/guides/linux_gsg/eal_args.include.rst
@@ -210,3 +210,19 @@ Other options
 *``--no-telemetry``:
 
 Disable telemetry.
+
+*``--force-max-simd-bitwidth=``:
+
+Specify the maximum SIMD bitwidth size to handle. This limits which vector 
paths,
+if any, are taken, as any paths taken must use a bitwidth below the max 
bitwidth limit.
+For example, to allow all SIMD bitwidths up to and including AVX-512::
+
+--force-max-simd-bitwidth=512
+
+The following example shows limiting the bitwidth to 64-bits to disable 
all vector code::
+
+--force-max-simd-bitwidth=64
+
+To disable use of max SIMD bitwidth limit::
+
+--force-max-simd-bitwidth=0
diff --git a/doc/guides/prog_guide/env_abstraction_layer.rst 
b/doc/guides/prog_guide/env_abstraction_layer.rst
index f64ae953d1..58f591e921 100644
--- a/doc/guides/prog_guide/env_abstraction_layer.rst
+++ b/doc/guides/prog_guide/env_abstraction_layer.rst
@@ -486,6 +486,38 @@ the desired addressing mode when virtual devices that are 
not directly attached
 To facilitate forcing the IOVA mode to a specific value the EAL command line 
option ``--iova-mode`` can
 be used to select either physical addressing('pa') or virtual addressing('va').
 
+.. _max_simd_bitwidth:
+
+
+Max SIMD bitwidth
+~
+
+The EAL provides a single setting to limit the max SIMD bitwidth used by DPDK,
+which is used in determining the vector path, if any, chosen by a component.
+The value can be set at runtime by an application using the 
'rte_set_max_simd_bitwidth(uint16_t bitwidth)' function,
+which should only be called once at initialization, before EAL init.
+The value can be overridden by the user using the EAL command-line option 
'--force-max-simd-bitwidth'.
+
+When choosing a vector path, along with checking the CPU feature support,
+the value of the max SIMD bitwidth must also be checked, and can be retrieved 
using the 'rte_get_max_simd_bitwidth()' function.
+The value should be compared against the enum values for accepted max SIMD 
bitwidths:
+
+.. code-block:: c
+
+   enum rte_max_simd_t {
+   RTE_NO_SIMD = 64,
+   RTE_MAX_128_SIMD = 128,
+   RTE_MAX_256_SIMD = 256,
+   RTE_MAX_512_SIMD = 512,
+   RTE_MAX_SIMD_DISABLE = UINT16_MAX,
+   };
+
+if (rte_get_max_simd_bitwidth() >= RTE_MAX_512_SIMD)
+/* Take AVX-512 vect

[dpdk-dev] [PATCH v3 05/18] net/axgbe: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU enabled
path.

Cc: Somalapuram Amaranath 

Signed-off-by: Ciara Power 
---
 drivers/net/axgbe/axgbe_rxtx.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/axgbe/axgbe_rxtx.c b/drivers/net/axgbe/axgbe_rxtx.c
index bc93becaa5..6093ec7279 100644
--- a/drivers/net/axgbe/axgbe_rxtx.c
+++ b/drivers/net/axgbe/axgbe_rxtx.c
@@ -557,7 +557,8 @@ int axgbe_dev_tx_queue_setup(struct rte_eth_dev *dev, 
uint16_t queue_idx,
if (!pdata->tx_queues)
pdata->tx_queues = dev->data->tx_queues;
 
-   if (txq->vector_disable)
+   if (txq->vector_disable || rte_get_max_simd_bitwidth()
+   < RTE_MAX_128_SIMD)
dev->tx_pkt_burst = &axgbe_xmit_pkts;
else
 #ifdef RTE_ARCH_X86
-- 
2.17.1

[dpdk-dev] [PATCH v3 04/18] net/i40e: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU enabled
path.

Cc: Beilei Xing 
Cc: Jeff Guo 

Signed-off-by: Ciara Power 
---
 drivers/net/i40e/i40e_rxtx.c | 19 +--
 1 file changed, 13 insertions(+), 6 deletions(-)

diff --git a/drivers/net/i40e/i40e_rxtx.c b/drivers/net/i40e/i40e_rxtx.c
index 60b33d20a1..9b535b52fa 100644
--- a/drivers/net/i40e/i40e_rxtx.c
+++ b/drivers/net/i40e/i40e_rxtx.c
@@ -3098,7 +3098,8 @@ static eth_rx_burst_t
 i40e_get_latest_rx_vec(bool scatter)
 {
 #if defined(RTE_ARCH_X86) && defined(CC_AVX2_SUPPORT)
-   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2))
+   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) &&
+   rte_get_max_simd_bitwidth() >= RTE_MAX_256_SIMD)
return scatter ? i40e_recv_scattered_pkts_vec_avx2 :
 i40e_recv_pkts_vec_avx2;
 #endif
@@ -3115,7 +3116,8 @@ i40e_get_recommend_rx_vec(bool scatter)
 * use of AVX2 version to later plaforms, not all those that could
 * theoretically run it.
 */
-   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F))
+   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) &&
+   rte_get_max_simd_bitwidth() >= RTE_MAX_256_SIMD)
return scatter ? i40e_recv_scattered_pkts_vec_avx2 :
 i40e_recv_pkts_vec_avx2;
 #endif
@@ -3154,7 +3156,8 @@ i40e_set_rx_function(struct rte_eth_dev *dev)
}
}
 
-   if (ad->rx_vec_allowed) {
+   if (ad->rx_vec_allowed  && rte_get_max_simd_bitwidth()
+   >= RTE_MAX_128_SIMD) {
/* Vec Rx path */
PMD_INIT_LOG(DEBUG, "Vector Rx path will be used on port=%d.",
dev->data->port_id);
@@ -3268,7 +3271,8 @@ static eth_tx_burst_t
 i40e_get_latest_tx_vec(void)
 {
 #if defined(RTE_ARCH_X86) && defined(CC_AVX2_SUPPORT)
-   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2))
+   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) &&
+   rte_get_max_simd_bitwidth() >= RTE_MAX_256_SIMD)
return i40e_xmit_pkts_vec_avx2;
 #endif
return i40e_xmit_pkts_vec;
@@ -3283,7 +3287,8 @@ i40e_get_recommend_tx_vec(void)
 * use of AVX2 version to later plaforms, not all those that could
 * theoretically run it.
 */
-   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F))
+   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) &&
+   rte_get_max_simd_bitwidth() >= RTE_MAX_256_SIMD)
return i40e_xmit_pkts_vec_avx2;
 #endif
return i40e_xmit_pkts_vec;
@@ -3311,7 +3316,9 @@ i40e_set_tx_function(struct rte_eth_dev *dev)
}
 
if (ad->tx_simple_allowed) {
-   if (ad->tx_vec_allowed) {
+   if (ad->tx_vec_allowed &&
+   rte_get_max_simd_bitwidth()
+   >= RTE_MAX_128_SIMD) {
PMD_INIT_LOG(DEBUG, "Vector tx finally be used.");
if (ad->use_latest_vec)
dev->tx_pkt_burst =
-- 
2.17.1

[dpdk-dev] [PATCH v3 07/18] net/enic: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU enabled
path.

Cc: John Daley 
Cc: Hyong Youb Kim 

Acked-by: Hyong Youb Kim 
Signed-off-by: Ciara Power 
---
 drivers/net/enic/enic_rxtx_vec_avx2.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/enic/enic_rxtx_vec_avx2.c 
b/drivers/net/enic/enic_rxtx_vec_avx2.c
index 676b9f5fdb..5db43bdbb8 100644
--- a/drivers/net/enic/enic_rxtx_vec_avx2.c
+++ b/drivers/net/enic/enic_rxtx_vec_avx2.c
@@ -821,7 +821,8 @@ enic_use_vector_rx_handler(struct rte_eth_dev *eth_dev)
fconf = ð_dev->data->dev_conf.fdir_conf;
if (fconf->mode != RTE_FDIR_MODE_NONE)
return false;
-   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2)) {
+   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) &&
+   rte_get_max_simd_bitwidth() >= RTE_MAX_256_SIMD) {
ENICPMD_LOG(DEBUG, " use the non-scatter avx2 Rx handler");
eth_dev->rx_pkt_burst = &enic_noscatter_vec_recv_pkts;
enic->use_noscatter_vec_rx_handler = 1;
-- 
2.17.1

[dpdk-dev] [PATCH v3 06/18] net/bnxt: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU enabled
path.

Cc: Ajit Khaparde 
Cc: Somnath Kotur 

Signed-off-by: Ciara Power 
---
 drivers/net/bnxt/bnxt_ethdev.c | 6 --
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index 05e9a6abbf..5cd522f1fd 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -1122,7 +1122,8 @@ bnxt_receive_function(struct rte_eth_dev *eth_dev)
DEV_RX_OFFLOAD_OUTER_IPV4_CKSUM |
DEV_RX_OFFLOAD_RSS_HASH |
DEV_RX_OFFLOAD_VLAN_FILTER)) &&
-   !BNXT_TRUFLOW_EN(bp) && BNXT_NUM_ASYNC_CPR(bp)) {
+   !BNXT_TRUFLOW_EN(bp) && BNXT_NUM_ASYNC_CPR(bp) &&
+   rte_get_max_simd_bitwidth() >= RTE_MAX_128_SIMD) {
PMD_DRV_LOG(INFO, "Using vector mode receive for port %d\n",
eth_dev->data->port_id);
bp->flags |= BNXT_FLAG_RX_VECTOR_PKT_MODE;
@@ -1154,7 +1155,8 @@ bnxt_transmit_function(__rte_unused struct rte_eth_dev 
*eth_dev)
 */
if (!eth_dev->data->scattered_rx &&
!eth_dev->data->dev_conf.txmode.offloads &&
-   !BNXT_TRUFLOW_EN(bp)) {
+   !BNXT_TRUFLOW_EN(bp) &&
+   rte_get_max_simd_bitwidth() >= RTE_MAX_128_SIMD) {
PMD_DRV_LOG(INFO, "Using vector mode transmit for port %d\n",
eth_dev->data->port_id);
return bnxt_xmit_pkts_vec;
-- 
2.17.1

[dpdk-dev] [PATCH v3 09/18] net/iavf: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU enabled
path.

Cc: Jingjing Wu 
Cc: Beilei Xing 

Signed-off-by: Ciara Power 
---
 drivers/net/iavf/iavf_rxtx.c | 16 ++--
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/net/iavf/iavf_rxtx.c b/drivers/net/iavf/iavf_rxtx.c
index 05a7dd898a..b798d082a2 100644
--- a/drivers/net/iavf/iavf_rxtx.c
+++ b/drivers/net/iavf/iavf_rxtx.c
@@ -2105,14 +2105,16 @@ iavf_set_rx_function(struct rte_eth_dev *dev)
int i;
bool use_avx2 = false;
 
-   if (!iavf_rx_vec_dev_check(dev)) {
+   if (!iavf_rx_vec_dev_check(dev) &&
+   rte_get_max_simd_bitwidth() >= RTE_MAX_128_SIMD) {
for (i = 0; i < dev->data->nb_rx_queues; i++) {
rxq = dev->data->rx_queues[i];
(void)iavf_rxq_vec_setup(rxq);
}
 
-   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) == 1 ||
-   rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1)
+   if ((rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) == 1 ||
+   rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1) &&
+   rte_get_max_simd_bitwidth() >= RTE_MAX_256_SIMD)
use_avx2 = true;
 
if (dev->data->scattered_rx) {
@@ -2178,7 +2180,8 @@ iavf_set_tx_function(struct rte_eth_dev *dev)
int i;
bool use_avx2 = false;
 
-   if (!iavf_tx_vec_dev_check(dev)) {
+   if (!iavf_tx_vec_dev_check(dev) &&
+   rte_get_max_simd_bitwidth() >= RTE_MAX_128_SIMD) {
for (i = 0; i < dev->data->nb_tx_queues; i++) {
txq = dev->data->tx_queues[i];
if (!txq)
@@ -2186,8 +2189,9 @@ iavf_set_tx_function(struct rte_eth_dev *dev)
iavf_txq_vec_setup(txq);
}
 
-   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) == 1 ||
-   rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1)
+   if ((rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) == 1 ||
+   rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1) &&
+   rte_get_max_simd_bitwidth() >= RTE_MAX_256_SIMD)
use_avx2 = true;
 
PMD_DRV_LOG(DEBUG, "Using %sVector Tx (port %d).",
-- 
2.17.1

[dpdk-dev] [PATCH v3 08/18] net/fm10k: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU enabled
path.

Cc: Qi Zhang 
Cc: Xiao Wang 

Signed-off-by: Ciara Power 
---
 drivers/net/fm10k/fm10k_ethdev.c | 11 ---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/net/fm10k/fm10k_ethdev.c b/drivers/net/fm10k/fm10k_ethdev.c
index 5771d83b55..a8bc1036a3 100644
--- a/drivers/net/fm10k/fm10k_ethdev.c
+++ b/drivers/net/fm10k/fm10k_ethdev.c
@@ -2930,7 +2930,9 @@ fm10k_set_tx_function(struct rte_eth_dev *dev)
if (rte_eal_process_type() != RTE_PROC_PRIMARY) {
/* primary process has set the ftag flag and offloads */
txq = dev->data->tx_queues[0];
-   if (fm10k_tx_vec_condition_check(txq)) {
+   if (fm10k_tx_vec_condition_check(txq) ||
+   rte_get_max_simd_bitwidth()
+   < RTE_MAX_128_SIMD) {
dev->tx_pkt_burst = fm10k_xmit_pkts;
dev->tx_pkt_prepare = fm10k_prep_pkts;
PMD_INIT_LOG(DEBUG, "Use regular Tx func");
@@ -2949,7 +2951,8 @@ fm10k_set_tx_function(struct rte_eth_dev *dev)
txq = dev->data->tx_queues[i];
txq->tx_ftag_en = tx_ftag_en;
/* Check if Vector Tx is satisfied */
-   if (fm10k_tx_vec_condition_check(txq))
+   if (fm10k_tx_vec_condition_check(txq) ||
+   rte_get_max_simd_bitwidth() < RTE_MAX_128_SIMD)
use_sse = 0;
}
 
@@ -2983,7 +2986,9 @@ fm10k_set_rx_function(struct rte_eth_dev *dev)
 * conditions to be met.
 */
if (!fm10k_rx_vec_condition_check(dev) &&
-   dev_info->rx_vec_allowed && !rx_ftag_en) {
+   dev_info->rx_vec_allowed && !rx_ftag_en &&
+   rte_get_max_simd_bitwidth()
+   >= RTE_MAX_128_SIMD) {
if (dev->data->scattered_rx)
dev->rx_pkt_burst = fm10k_recv_scattered_pkts_vec;
else
-- 
2.17.1

[dpdk-dev] [PATCH v3 10/18] net/ice: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU enabled
path.

Cc: Qiming Yang 
Cc: Qi Zhang 

Signed-off-by: Ciara Power 
---
 drivers/net/ice/ice_rxtx.c | 20 ++--
 1 file changed, 14 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ice/ice_rxtx.c b/drivers/net/ice/ice_rxtx.c
index fef6ad4544..5a29af743c 100644
--- a/drivers/net/ice/ice_rxtx.c
+++ b/drivers/net/ice/ice_rxtx.c
@@ -2936,7 +2936,9 @@ ice_set_rx_function(struct rte_eth_dev *dev)
bool use_avx2 = false;
 
if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
-   if (!ice_rx_vec_dev_check(dev) && ad->rx_bulk_alloc_allowed) {
+   if (!ice_rx_vec_dev_check(dev) && ad->rx_bulk_alloc_allowed &&
+   rte_get_max_simd_bitwidth()
+   >= RTE_MAX_128_SIMD) {
ad->rx_vec_allowed = true;
for (i = 0; i < dev->data->nb_rx_queues; i++) {
rxq = dev->data->rx_queues[i];
@@ -2946,8 +2948,10 @@ ice_set_rx_function(struct rte_eth_dev *dev)
}
}
 
-   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) == 1 ||
-   rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1)
+   if ((rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) == 1 ||
+   rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1) &&
+   rte_get_max_simd_bitwidth()
+   >= RTE_MAX_256_SIMD)
use_avx2 = true;
 
} else {
@@ -3114,7 +3118,9 @@ ice_set_tx_function(struct rte_eth_dev *dev)
bool use_avx2 = false;
 
if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
-   if (!ice_tx_vec_dev_check(dev)) {
+   if (!ice_tx_vec_dev_check(dev) &&
+   rte_get_max_simd_bitwidth()
+   >= RTE_MAX_128_SIMD) {
ad->tx_vec_allowed = true;
for (i = 0; i < dev->data->nb_tx_queues; i++) {
txq = dev->data->tx_queues[i];
@@ -3124,8 +3130,10 @@ ice_set_tx_function(struct rte_eth_dev *dev)
}
}
 
-   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) == 1 ||
-   rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1)
+   if ((rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) == 1 ||
+   rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1) &&
+   rte_get_max_simd_bitwidth()
+   >= RTE_MAX_256_SIMD)
use_avx2 = true;
 
} else {
-- 
2.17.1

[dpdk-dev] [PATCH v3 11/18] net/ixgbe: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU enabled
path.

Cc: Wei Zhao 
Cc: Jeff Guo 

Signed-off-by: Ciara Power 
---
 drivers/net/ixgbe/ixgbe_rxtx.c | 7 +--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ixgbe/ixgbe_rxtx.c b/drivers/net/ixgbe/ixgbe_rxtx.c
index 977ecf5137..eadc7183f2 100644
--- a/drivers/net/ixgbe/ixgbe_rxtx.c
+++ b/drivers/net/ixgbe/ixgbe_rxtx.c
@@ -2503,7 +2503,9 @@ ixgbe_set_tx_function(struct rte_eth_dev *dev, struct 
ixgbe_tx_queue *txq)
dev->tx_pkt_prepare = NULL;
if (txq->tx_rs_thresh <= RTE_IXGBE_TX_MAX_FREE_BUF_SZ &&
(rte_eal_process_type() != RTE_PROC_PRIMARY ||
-   ixgbe_txq_vec_setup(txq) == 0)) {
+   ixgbe_txq_vec_setup(txq) == 0) &&
+   rte_get_max_simd_bitwidth()
+   >= RTE_MAX_128_SIMD) {
PMD_INIT_LOG(DEBUG, "Vector tx enabled.");
dev->tx_pkt_burst = ixgbe_xmit_pkts_vec;
} else
@@ -4743,7 +4745,8 @@ ixgbe_set_rx_function(struct rte_eth_dev *dev)
 * conditions to be met and Rx Bulk Allocation should be allowed.
 */
if (ixgbe_rx_vec_dev_conf_condition_check(dev) ||
-   !adapter->rx_bulk_alloc_allowed) {
+   !adapter->rx_bulk_alloc_allowed ||
+   rte_get_max_simd_bitwidth() < RTE_MAX_128_SIMD) {
PMD_INIT_LOG(DEBUG, "Port[%d] doesn't meet Vector Rx "
"preconditions",
 dev->data->port_id);
-- 
2.17.1

[dpdk-dev] [PATCH v3 12/18] net/mlx5: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU enabled
path.

Cc: Matan Azrad 
Cc: Shahaf Shuler 
Cc: Viacheslav Ovsiienko 

Signed-off-by: Ciara Power 

---
v2: Moved check for max bitwidth into existing check vec
support function.
---
 drivers/net/mlx5/mlx5_rxtx_vec.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/mlx5/mlx5_rxtx_vec.c b/drivers/net/mlx5/mlx5_rxtx_vec.c
index 711dcd35fa..c384c737dc 100644
--- a/drivers/net/mlx5/mlx5_rxtx_vec.c
+++ b/drivers/net/mlx5/mlx5_rxtx_vec.c
@@ -148,6 +148,8 @@ mlx5_check_vec_rx_support(struct rte_eth_dev *dev)
struct mlx5_priv *priv = dev->data->dev_private;
uint32_t i;
 
+   if (rte_get_max_simd_bitwidth() < RTE_MAX_128_SIMD)
+   return -ENOTSUP;
if (!priv->config.rx_vec_en)
return -ENOTSUP;
if (mlx5_mprq_enabled(dev))
-- 
2.17.1

[dpdk-dev] [PATCH v3 13/18] net/virtio: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU enabled
path.

Cc: Maxime Coquelin 
Cc: Chenbo Xia 
Cc: Zhihong Wang 

Signed-off-by: Ciara Power 

---
v3: Moved max SIMD bitwidth check to configure function with other vec
support checks.
---
 drivers/net/virtio/virtio_ethdev.c | 9 -
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index 013a2904e6..f749e81405 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -2306,7 +2306,8 @@ virtio_dev_configure(struct rte_eth_dev *dev)
if ((hw->use_vec_rx || hw->use_vec_tx) &&
(!rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) ||
 !vtpci_with_feature(hw, VIRTIO_F_IN_ORDER) ||
-!vtpci_with_feature(hw, VIRTIO_F_VERSION_1))) {
+!vtpci_with_feature(hw, VIRTIO_F_VERSION_1) ||
+rte_get_max_simd_bitwidth() < RTE_MAX_512_SIMD)) {
PMD_DRV_LOG(INFO,
"disabled packed ring vectorized path for 
requirements not met");
hw->use_vec_rx = 0;
@@ -2359,6 +2360,12 @@ virtio_dev_configure(struct rte_eth_dev *dev)
"disabled split ring vectorized rx for 
offloading enabled");
hw->use_vec_rx = 0;
}
+
+   if (rte_get_max_simd_bitwidth() < RTE_MAX_128_SIMD) {
+   PMD_DRV_LOG(INFO,
+   "disabled split ring vectorized rx, max 
SIMD bitwidth too low");
+   hw->use_vec_rx = 0;
+   }
}
}
 
-- 
2.17.1

[dpdk-dev] [PATCH v3 14/18] distributor: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU enabled
path.

Cc: David Hunt 

Signed-off-by: Ciara Power 
---
 lib/librte_distributor/rte_distributor.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/lib/librte_distributor/rte_distributor.c 
b/lib/librte_distributor/rte_distributor.c
index 1c047f065a..9f0a9b1d48 100644
--- a/lib/librte_distributor/rte_distributor.c
+++ b/lib/librte_distributor/rte_distributor.c
@@ -636,7 +636,8 @@ rte_distributor_create(const char *name,
 
d->dist_match_fn = RTE_DIST_MATCH_SCALAR;
 #if defined(RTE_ARCH_X86)
-   d->dist_match_fn = RTE_DIST_MATCH_VECTOR;
+   if (rte_get_max_simd_bitwidth() >= RTE_MAX_128_SIMD)
+   d->dist_match_fn = RTE_DIST_MATCH_VECTOR;
 #endif
 
/*
-- 
2.17.1

[dpdk-dev] [PATCH v3 15/18] member: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU
enabled path.

Cc: Yipeng Wang 
Cc: Sameh Gobriel 

Signed-off-by: Ciara Power 
---
 lib/librte_member/rte_member_ht.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/lib/librte_member/rte_member_ht.c 
b/lib/librte_member/rte_member_ht.c
index cbcd0d4407..71e3cf7b52 100644
--- a/lib/librte_member/rte_member_ht.c
+++ b/lib/librte_member/rte_member_ht.c
@@ -113,7 +113,8 @@ rte_member_create_ht(struct rte_member_setsum *ss,
}
 #if defined(RTE_ARCH_X86)
if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) &&
-   RTE_MEMBER_BUCKET_ENTRIES == 16)
+   RTE_MEMBER_BUCKET_ENTRIES == 16 &&
+   rte_get_max_simd_bitwidth() >= RTE_MAX_256_SIMD)
ss->sig_cmp_fn = RTE_MEMBER_COMPARE_AVX2;
else
 #endif
-- 
2.17.1

[dpdk-dev] [PATCH v3 16/18] efd: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU enabled
path.

Cc: Byron Marohn 
Cc: Yipeng Wang 

Signed-off-by: Ciara Power 
---
 lib/librte_efd/rte_efd.c | 7 +--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/lib/librte_efd/rte_efd.c b/lib/librte_efd/rte_efd.c
index 6a799556d4..509ecc8256 100644
--- a/lib/librte_efd/rte_efd.c
+++ b/lib/librte_efd/rte_efd.c
@@ -645,7 +645,9 @@ rte_efd_create(const char *name, uint32_t max_num_rules, 
uint32_t key_len,
 * For less than 4 bits, scalar function performs better
 * than vectorised version
 */
-   if (RTE_EFD_VALUE_NUM_BITS > 3 && 
rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2))
+   if (RTE_EFD_VALUE_NUM_BITS > 3
+   && rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2)
+   && rte_get_max_simd_bitwidth() >= RTE_MAX_256_SIMD)
table->lookup_fn = EFD_LOOKUP_AVX2;
else
 #endif
@@ -655,7 +657,8 @@ rte_efd_create(const char *name, uint32_t max_num_rules, 
uint32_t key_len,
 * than vectorised version
 */
if (RTE_EFD_VALUE_NUM_BITS > 16 &&
-   rte_cpu_get_flag_enabled(RTE_CPUFLAG_NEON))
+   rte_cpu_get_flag_enabled(RTE_CPUFLAG_NEON) &&
+   rte_get_max_simd_bitwidth() >= RTE_MAX_128_SIMD)
table->lookup_fn = EFD_LOOKUP_NEON;
else
 #endif
-- 
2.17.1

[dpdk-dev] [PATCH v3 18/18] lpm: choose vector path at runtime

[Ciara Power]

When choosing the vector path, max SIMD bitwidth is now checked to
ensure a vector path is allowable. To do this, rather than the vector
lookup functions being called directly from apps, a generic lookup
function is called which will call the vector functions if suitable.

Signed-off-by: Ciara Power 
---
 lib/librte_lpm/rte_lpm.h | 57 ++--
 lib/librte_lpm/rte_lpm_altivec.h |  2 +-
 lib/librte_lpm/rte_lpm_neon.h|  2 +-
 lib/librte_lpm/rte_lpm_sse.h |  2 +-
 4 files changed, 50 insertions(+), 13 deletions(-)

diff --git a/lib/librte_lpm/rte_lpm.h b/lib/librte_lpm/rte_lpm.h
index 03da2d37e0..edba7cafd5 100644
--- a/lib/librte_lpm/rte_lpm.h
+++ b/lib/librte_lpm/rte_lpm.h
@@ -397,8 +397,18 @@ rte_lpm_lookup_bulk_func(const struct rte_lpm *lpm, const 
uint32_t *ips,
 /* Mask four results. */
 #define RTE_LPM_MASKX4_RES UINT64_C(0x00ff00ff)
 
+#if defined(RTE_ARCH_ARM) || defined(RTE_ARCH_ARM64)
+#include "rte_lpm_neon.h"
+#elif defined(RTE_ARCH_PPC_64)
+#include "rte_lpm_altivec.h"
+#else
+#include "rte_lpm_sse.h"
+#endif
+
 /**
- * Lookup four IP addresses in an LPM table.
+ * Lookup four IP addresses in an LPM table individually by calling the
+ * lookup function for each ip. This is used when lookupx4 is called but
+ * the vector path is not suitable.
  *
  * @param lpm
  *   LPM object handle
@@ -417,16 +427,43 @@ rte_lpm_lookup_bulk_func(const struct rte_lpm *lpm, const 
uint32_t *ips,
  *   if lookup would fail.
  */
 static inline void
-rte_lpm_lookupx4(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
-   uint32_t defv);
+rte_lpm_lookupx4_scalar(struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
+   uint32_t defv)
+{
+   int i;
+   for (i = 0; i < 4; i++)
+   if (rte_lpm_lookup(lpm, ((uint32_t *) &ip)[i], &hop[i]) < 0)
+   hop[i] = defv; /* lookupx4 expected to set on failure */
+}
 
-#if defined(RTE_ARCH_ARM) || defined(RTE_ARCH_ARM64)
-#include "rte_lpm_neon.h"
-#elif defined(RTE_ARCH_PPC_64)
-#include "rte_lpm_altivec.h"
-#else
-#include "rte_lpm_sse.h"
-#endif
+/**
+ * Lookup four IP addresses in an LPM table.
+ *
+ * @param lpm
+ *   LPM object handle
+ * @param ip
+ *   Four IPs to be looked up in the LPM table
+ * @param hop
+ *   Next hop of the most specific rule found for IP (valid on lookup hit 
only).
+ *   This is an 4 elements array of two byte values.
+ *   If the lookup was successful for the given IP, then least significant byte
+ *   of the corresponding element is the  actual next hop and the most
+ *   significant byte is zero.
+ *   If the lookup for the given IP failed, then corresponding element would
+ *   contain default value, see description of then next parameter.
+ * @param defv
+ *   Default value to populate into corresponding element of hop[] array,
+ *   if lookup would fail.
+ */
+static inline void
+rte_lpm_lookupx4(struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
+   uint32_t defv)
+{
+   if (rte_get_max_simd_bitwidth() >= RTE_MAX_128_SIMD)
+   rte_lpm_lookupx4_vec(lpm, ip, hop, defv);
+   else
+   rte_lpm_lookupx4_scalar(lpm, ip, hop, defv);
+}
 
 #ifdef __cplusplus
 }
diff --git a/lib/librte_lpm/rte_lpm_altivec.h b/lib/librte_lpm/rte_lpm_altivec.h
index 228c41b38e..82142d3351 100644
--- a/lib/librte_lpm/rte_lpm_altivec.h
+++ b/lib/librte_lpm/rte_lpm_altivec.h
@@ -16,7 +16,7 @@ extern "C" {
 #endif
 
 static inline void
-rte_lpm_lookupx4(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
+rte_lpm_lookupx4_vec(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
uint32_t defv)
 {
vector signed int i24;
diff --git a/lib/librte_lpm/rte_lpm_neon.h b/lib/librte_lpm/rte_lpm_neon.h
index 6c131d3125..14b184515d 100644
--- a/lib/librte_lpm/rte_lpm_neon.h
+++ b/lib/librte_lpm/rte_lpm_neon.h
@@ -16,7 +16,7 @@ extern "C" {
 #endif
 
 static inline void
-rte_lpm_lookupx4(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
+rte_lpm_lookupx4_vec(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
uint32_t defv)
 {
uint32x4_t i24;
diff --git a/lib/librte_lpm/rte_lpm_sse.h b/lib/librte_lpm/rte_lpm_sse.h
index 44770b6ff8..cb5477c6cf 100644
--- a/lib/librte_lpm/rte_lpm_sse.h
+++ b/lib/librte_lpm/rte_lpm_sse.h
@@ -15,7 +15,7 @@ extern "C" {
 #endif
 
 static inline void
-rte_lpm_lookupx4(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
+rte_lpm_lookupx4_vec(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
uint32_t defv)
 {
__m128i i24;
-- 
2.17.1

[dpdk-dev] [PATCH v3 17/18] net: add checks for max SIMD bitwidth

[Ciara Power]

When choosing a vector path to take, an extra condition must be
satisfied to ensure the max SIMD bitwidth allows for the CPU enabled
path.

The vector path was initially chosen in RTE_INIT, however this is no
longer suitable as we cannot check the max SIMD bitwidth at that time.
The default chosen in RTE_INIT is now scalar. For best performance
and to use vector paths, apps must explicitly call the set algorithm
function before using other functions from this library, as this is
where vector handlers are now chosen.

Suggested-by: Jasvinder Singh 

Signed-off-by: Ciara Power 

---
v3:
  - Moved choosing vector paths out of RTE_INIT.
  - Moved checking max_simd_bitwidth into the set_alg function.
---
 lib/librte_net/rte_net_crc.c | 26 +-
 lib/librte_net/rte_net_crc.h |  3 ++-
 2 files changed, 19 insertions(+), 10 deletions(-)

diff --git a/lib/librte_net/rte_net_crc.c b/lib/librte_net/rte_net_crc.c
index 9fd4794a9d..241eb16399 100644
--- a/lib/librte_net/rte_net_crc.c
+++ b/lib/librte_net/rte_net_crc.c
@@ -9,6 +9,7 @@
 #include 
 #include 
 #include 
+#include 
 
 #if defined(RTE_ARCH_X86_64) && defined(RTE_MACHINE_CPUFLAG_PCLMULQDQ)
 #define X86_64_SSE42_PCLMULQDQ 1
@@ -60,6 +61,9 @@ static rte_net_crc_handler handlers_neon[] = {
 };
 #endif
 
+static uint16_t max_simd_bitwidth;
+#define RTE_LOGTYPE_NET RTE_LOGTYPE_USER1
+
 /**
  * Reflect the bits about the middle
  *
@@ -145,18 +149,26 @@ rte_crc32_eth_handler(const uint8_t *data, uint32_t 
data_len)
 void
 rte_net_crc_set_alg(enum rte_net_crc_alg alg)
 {
+   if (max_simd_bitwidth == 0)
+   max_simd_bitwidth = rte_get_max_simd_bitwidth();
+
switch (alg) {
 #ifdef X86_64_SSE42_PCLMULQDQ
case RTE_NET_CRC_SSE42:
-   handlers = handlers_sse42;
-   break;
+   if (max_simd_bitwidth >= RTE_MAX_128_SIMD) {
+   handlers = handlers_sse42;
+   return;
+   }
+   RTE_LOG(INFO, NET, "Max SIMD Bitwidth too low, using scalar\n");
 #elif defined ARM64_NEON_PMULL
/* fall-through */
case RTE_NET_CRC_NEON:
-   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_PMULL)) {
+   if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_PMULL) &&
+   max_simd_bitwidth >= RTE_MAX_128_SIMD) {
handlers = handlers_neon;
-   break;
+   return;
}
+   RTE_LOG(INFO, NET, "Max SIMD Bitwidth too low or CPU flag not 
enabled, using scalar\n");
 #endif
/* fall-through */
case RTE_NET_CRC_SCALAR:
@@ -184,19 +196,15 @@ rte_net_crc_calc(const void *data,
 /* Select highest available crc algorithm as default one */
 RTE_INIT(rte_net_crc_init)
 {
-   enum rte_net_crc_alg alg = RTE_NET_CRC_SCALAR;
-
rte_net_crc_scalar_init();
 
 #ifdef X86_64_SSE42_PCLMULQDQ
-   alg = RTE_NET_CRC_SSE42;
rte_net_crc_sse42_init();
 #elif defined ARM64_NEON_PMULL
if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_PMULL)) {
-   alg = RTE_NET_CRC_NEON;
rte_net_crc_neon_init();
}
 #endif
 
-   rte_net_crc_set_alg(alg);
+   rte_net_crc_set_alg(RTE_NET_CRC_SCALAR);
 }
diff --git a/lib/librte_net/rte_net_crc.h b/lib/librte_net/rte_net_crc.h
index 16e85ca970..7a45ebe193 100644
--- a/lib/librte_net/rte_net_crc.h
+++ b/lib/librte_net/rte_net_crc.h
@@ -28,7 +28,8 @@ enum rte_net_crc_alg {
 /**
  * This API set the CRC computation algorithm (i.e. scalar version,
  * x86 64-bit sse4.2 intrinsic version, etc.) and internal data
- * structure.
+ * structure. This should be called before any other functions, to
+ * choose the algorithm for best performance.
  *
  * @param alg
  *   This parameter is used to select the CRC implementation version.
-- 
2.17.1

Re: [dpdk-dev] [PATCH v4] RSS key use with testpmd

[Ferruh Yigit]

On 9/24/2020 2:45 PM, Lijun Ou wrote:

Consider the follow usage with testpmd:
1. first, startup testpmd:
testpmd> show port 0 rss-hash key
RSS functions:
  all ipv4-frag ipv4-other ipv6-frag ipv6-other ip
RSS key:
6D5A56DA255B0EC24167253D43A38FB0D0CA2BCBAE7B30B477CB2DA38030F
20C6A42B73BBEAC01FA
2. create a rss rule
testpmd> flow create 0 ingress pattern eth / ipv4 / udp / end actions rss \
types ipv4-udp end queues end / end

3. show rss-hash key
testpmd> show port 0 rss-hash key
RSS functions:
  all ipv4-udp udp
RSS key:
74657374706D6427732064656661756C74205253532068617368206B65792C206F
76657272696465

As a result, the step 3 with RSS key and the step 1 RSS key
is not the same. The patch[1] to solve the above problems.



This is interesting, can you please provide above information in the commit log 
too?

Also can you please provide the details on why this happens, callstack can help?

Thanks,
ferruh



Lijun Ou (1):
   app/testpmd: fix the default RSS key configuration

  app/test-pmd/cmdline_flow.c | 8 
  1 file changed, 8 insertions(+)

Re: [dpdk-dev] [PATCH v3 05/18] net/axgbe: add checks for max SIMD bitwidth

[Somalapuram, Amaranath]

-Original Message-
From: Ciara Power  
Sent: Wednesday, September 30, 2020 6:34 PM
To: dev@dpdk.org
Cc: Ciara Power ; Somalapuram, Amaranath 

Subject: [PATCH v3 05/18] net/axgbe: add checks for max SIMD bitwidth

[CAUTION: External Email]

When choosing a vector path to take, an extra condition must be satisfied to 
ensure the max SIMD bitwidth allows for the CPU enabled path.

Cc: Somalapuram Amaranath 

Signed-off-by: Ciara Power 
---
 drivers/net/axgbe/axgbe_rxtx.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/axgbe/axgbe_rxtx.c b/drivers/net/axgbe/axgbe_rxtx.c 
index bc93becaa5..6093ec7279 100644
--- a/drivers/net/axgbe/axgbe_rxtx.c
+++ b/drivers/net/axgbe/axgbe_rxtx.c
@@ -557,7 +557,8 @@ int axgbe_dev_tx_queue_setup(struct rte_eth_dev *dev, 
uint16_t queue_idx,
if (!pdata->tx_queues)
pdata->tx_queues = dev->data->tx_queues;

-   if (txq->vector_disable)
+   if (txq->vector_disable || rte_get_max_simd_bitwidth()
+   < RTE_MAX_128_SIMD)
dev->tx_pkt_burst = &axgbe_xmit_pkts;
else
 #ifdef RTE_ARCH_X86
--
2.17.1

Acked-by: Amaranath Somalapuram 

Re: [dpdk-dev] [PATCH v4] app/testpmd: fix the default RSS key configuration

[Ferruh Yigit]

On 9/24/2020 2:45 PM, Lijun Ou wrote:

It use the NIC valid default RSS key instead of the testpmd
dummy RSS key in the flow configuration when the RSS key is
not specified in the flow rule. If the NIC RSS key is
invalid, it will use testpmd dummy RSS key as the default
key.

Fixes: ac8d22de2394 ("ethdev: flatten RSS configuration in flow API")
Cc: sta...@dpdk.org

Signed-off-by: Lijun Ou 
---
V3->V4:
-fix checkpatch warning and shorter commit content.

V2->V3:
-fix checkpatch warning.

V1->V2:
-fix the commit.
---
  app/test-pmd/cmdline_flow.c | 8 
  1 file changed, 8 insertions(+)

diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c
index 6263d30..e6648da 100644
--- a/app/test-pmd/cmdline_flow.c
+++ b/app/test-pmd/cmdline_flow.c
@@ -4312,6 +4312,7 @@ parse_vc_action_rss(struct context *ctx, const struct 
token *token,
action_rss_data->queue[i] = i;
if (!port_id_is_invalid(ctx->port, DISABLED_WARN) &&
ctx->port != (portid_t)RTE_PORT_ALL) {
+   struct rte_eth_rss_conf rss_conf = {0};
struct rte_eth_dev_info info;
int ret2;
  
@@ -4322,6 +4323,13 @@ parse_vc_action_rss(struct context *ctx, const struct token *token,

action_rss_data->conf.key_len =
RTE_MIN(sizeof(action_rss_data->key),
info.hash_key_size);
+
+   rss_conf.rss_key_len = sizeof(action_rss_data->key);
+   rss_conf.rss_key = action_rss_data->key;
+   ret2 = rte_eth_dev_rss_hash_conf_get(ctx->port, &rss_conf);
+   if (ret2 != 0)
+   return ret2;
+   action_rss_data->conf.key = rss_conf.rss_key;


Do you need this last assignment at all?
'rss_conf.rss_key' point to 'action_rss_data->key'
'action_rss_data->conf.key' already point to 'action_rss_data->key'
so it just assigns same value back, no?

Re: [dpdk-dev] [PATCH v3 18/18] lpm: choose vector path at runtime

[Medvedkin, Vladimir]

Hi Ciara,


On 30/09/2020 14:04, Ciara Power wrote:

When choosing the vector path, max SIMD bitwidth is now checked to
ensure a vector path is allowable. To do this, rather than the vector
lookup functions being called directly from apps, a generic lookup
function is called which will call the vector functions if suitable.

Signed-off-by: Ciara Power 
---
  lib/librte_lpm/rte_lpm.h | 57 ++--
  lib/librte_lpm/rte_lpm_altivec.h |  2 +-
  lib/librte_lpm/rte_lpm_neon.h|  2 +-
  lib/librte_lpm/rte_lpm_sse.h |  2 +-
  4 files changed, 50 insertions(+), 13 deletions(-)

diff --git a/lib/librte_lpm/rte_lpm.h b/lib/librte_lpm/rte_lpm.h
index 03da2d37e0..edba7cafd5 100644
--- a/lib/librte_lpm/rte_lpm.h
+++ b/lib/librte_lpm/rte_lpm.h
@@ -397,8 +397,18 @@ rte_lpm_lookup_bulk_func(const struct rte_lpm *lpm, const 
uint32_t *ips,
  /* Mask four results. */
  #defineRTE_LPM_MASKX4_RES UINT64_C(0x00ff00ff)
  
+#if defined(RTE_ARCH_ARM) || defined(RTE_ARCH_ARM64)

+#include "rte_lpm_neon.h"
+#elif defined(RTE_ARCH_PPC_64)
+#include "rte_lpm_altivec.h"
+#else
+#include "rte_lpm_sse.h"
+#endif
+
  /**
- * Lookup four IP addresses in an LPM table.
+ * Lookup four IP addresses in an LPM table individually by calling the
+ * lookup function for each ip. This is used when lookupx4 is called but
+ * the vector path is not suitable.
   *
   * @param lpm
   *   LPM object handle
@@ -417,16 +427,43 @@ rte_lpm_lookup_bulk_func(const struct rte_lpm *lpm, const 
uint32_t *ips,
   *   if lookup would fail.
   */
  static inline void
-rte_lpm_lookupx4(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
-   uint32_t defv);
+rte_lpm_lookupx4_scalar(struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
+   uint32_t defv)
+{
+   int i;
+   for (i = 0; i < 4; i++)
+   if (rte_lpm_lookup(lpm, ((uint32_t *) &ip)[i], &hop[i]) < 0)
+   hop[i] = defv; /* lookupx4 expected to set on failure */
+}
  
-#if defined(RTE_ARCH_ARM) || defined(RTE_ARCH_ARM64)

-#include "rte_lpm_neon.h"
-#elif defined(RTE_ARCH_PPC_64)
-#include "rte_lpm_altivec.h"
-#else
-#include "rte_lpm_sse.h"
-#endif
+/**
+ * Lookup four IP addresses in an LPM table.
+ *
+ * @param lpm
+ *   LPM object handle
+ * @param ip
+ *   Four IPs to be looked up in the LPM table
+ * @param hop
+ *   Next hop of the most specific rule found for IP (valid on lookup hit 
only).
+ *   This is an 4 elements array of two byte values.
+ *   If the lookup was successful for the given IP, then least significant byte
+ *   of the corresponding element is the  actual next hop and the most
+ *   significant byte is zero.
+ *   If the lookup for the given IP failed, then corresponding element would
+ *   contain default value, see description of then next parameter.
+ * @param defv
+ *   Default value to populate into corresponding element of hop[] array,
+ *   if lookup would fail.
+ */
+static inline void
+rte_lpm_lookupx4(struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
+   uint32_t defv)
+{
+   if (rte_get_max_simd_bitwidth() >= RTE_MAX_128_SIMD)
+   rte_lpm_lookupx4_vec(lpm, ip, hop, defv);
+   else
+   rte_lpm_lookupx4_scalar(lpm, ip, hop, defv);
+}


I'm afraid this will lead to a drop in performance. rte_lpm_lookupx4 is 
used in the hot path, and a bulk size is too small to amortize the cost 
of adding this extra logic.


  
  #ifdef __cplusplus

  }
diff --git a/lib/librte_lpm/rte_lpm_altivec.h b/lib/librte_lpm/rte_lpm_altivec.h
index 228c41b38e..82142d3351 100644
--- a/lib/librte_lpm/rte_lpm_altivec.h
+++ b/lib/librte_lpm/rte_lpm_altivec.h
@@ -16,7 +16,7 @@ extern "C" {
  #endif
  
  static inline void

-rte_lpm_lookupx4(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
+rte_lpm_lookupx4_vec(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
uint32_t defv)
  {
vector signed int i24;
diff --git a/lib/librte_lpm/rte_lpm_neon.h b/lib/librte_lpm/rte_lpm_neon.h
index 6c131d3125..14b184515d 100644
--- a/lib/librte_lpm/rte_lpm_neon.h
+++ b/lib/librte_lpm/rte_lpm_neon.h
@@ -16,7 +16,7 @@ extern "C" {
  #endif
  
  static inline void

-rte_lpm_lookupx4(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
+rte_lpm_lookupx4_vec(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
uint32_t defv)
  {
uint32x4_t i24;
diff --git a/lib/librte_lpm/rte_lpm_sse.h b/lib/librte_lpm/rte_lpm_sse.h
index 44770b6ff8..cb5477c6cf 100644
--- a/lib/librte_lpm/rte_lpm_sse.h
+++ b/lib/librte_lpm/rte_lpm_sse.h
@@ -15,7 +15,7 @@ extern "C" {
  #endif
  
  static inline void

-rte_lpm_lookupx4(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
+rte_lpm_lookupx4_vec(const struct rte_lpm *lpm, xmm_t ip, uint32_t hop[4],
uint32_t defv)
  {
__m128i i24;



--
Regards,
Vladimir

Re: [dpdk-dev] [PATCH v4] net/mlx5: relaxed ordering for multi-packet RQ buffer refcnt

[Raslan Darawsheh]

Hi,

> -Original Message-
> From: Phil Yang 
> Sent: Thursday, September 3, 2020 5:53 AM
> To: akozy...@mellanox.com; rasl...@mellanox.com; dev@dpdk.org
> Cc: honnappa.nagaraha...@arm.com; phil.y...@arm.com;
> ma...@mellanox.com; Shahaf Shuler ;
> viachesl...@mellanox.com; n...@arm.com
> Subject: [PATCH v4] net/mlx5: relaxed ordering for multi-packet RQ buffer
> refcnt
> 
> Use c11 atomics with RELAXED ordering instead of the rte_atomic ops
> which enforce unnecessary barriers on aarch64.
> 
> Signed-off-by: Phil Yang 
> ---
> v4:
> Remove the unnecessary ACQUIRE barrier in rx burst path. (Honnappa)
> 
> v3:
> Split from the patchset:
> https://nam11.safelinks.protection.outlook.com/?url=http%3A%2F%2Fpatch
> work.dpdk.org%2Fcover%2F68159%2F&data=02%7C01%7Crasland%40n
> vidia.com%7Ca89037e982b84575f58808d84fb493d9%7C43083d15727340c1b7d
> b39efd9ccc17a%7C0%7C0%7C637346984280298342&sdata=hjWErvvV%2
> Fg%2Bpb%2FMMzepHuRK0Weftf%2BFC2NNWsyoWSyo%3D&reserved
> =0
> 
>  drivers/net/mlx5/mlx5_rxq.c  |  2 +-
>  drivers/net/mlx5/mlx5_rxtx.c | 16 +---
>  drivers/net/mlx5/mlx5_rxtx.h |  2 +-
>  3 files changed, 11 insertions(+), 9 deletions(-)
> 

Patch applied to next-net-mlx,

Kindest regards,
Raslan Darawsheh

[dpdk-dev] [PATCH 00/10] support match on L3 fragmented packets

[Dekel Peled]

This series implements support of matching on packets based on the
fragmentation attribute of the packet, i.e. if packet is a fragment
of a larger packet, or the opposite - packet is not a fragment.

In ethdev, add API to support IPv6 extension headers, and specifically
the IPv6 fragment extension header item.
In MLX5 PMD, support match on IPv4 fragmented packets, IPv6 fragmented
packets, and IPv6 fragment extension header item.
Testpmd CLI is updated accordingly.
Documentation is updated accordingly.

Dekel Peled (10):
  ethdev: add extensions attributes to IPv6 item
  ethdev: add IPv6 fragment extension header item
  app/testpmd: support IPv4 fragments
  app/testpmd: support IPv6 fragments
  app/testpmd: support IPv6 fragment extension item
  net/mlx5: remove handling of ICMP fragmented packets
  net/mlx5: support match on IPv4 fragment packets
  net/mlx5: support match on IPv6 fragment packets
  net/mlx5: support match on IPv6 fragment ext. item
  doc: update release notes for MLX5 L3 frag support

 app/test-pmd/cmdline_flow.c|  53 +
 doc/guides/prog_guide/rte_flow.rst |  28 ++-
 doc/guides/rel_notes/release_20_11.rst |  10 +
 drivers/net/mlx5/mlx5_flow.c   |  48 +++--
 drivers/net/mlx5/mlx5_flow.h   |  14 ++
 drivers/net/mlx5/mlx5_flow_dv.c| 382 +
 drivers/net/mlx5/mlx5_flow_verbs.c |   9 +-
 lib/librte_ethdev/rte_flow.c   |   1 +
 lib/librte_ethdev/rte_flow.h   |  46 +++-
 lib/librte_ip_frag/rte_ip_frag.h   |  26 +--
 lib/librte_net/rte_ip.h|  26 ++-
 11 files changed, 555 insertions(+), 88 deletions(-)

-- 
1.8.3.1

[dpdk-dev] [PATCH 01/10] ethdev: add extensions attributes to IPv6 item

[Dekel Peled]

Using the current implementation of DPDK, an application cannot match on
IPv6 packets, based on the existing extension headers, in a simple way.

Field 'Next Header' in IPv6 header indicates type of the first extension
header only. Following extension headers can't be identified by
inspecting the IPv6 header.
As a result, the existence or absence of specific extension headers
can't be used for packet matching.

For example, fragmented IPv6 packets contain a dedicated extension header
(which is implemented in a later patch of this series).
Non-fragmented packets don't contain the fragment extension header.
For an application to match on non-fragmented IPv6 packets, the current
implementation doesn't provide a suitable solution.
Matching on the Next Header field is not sufficient, since additional
extension headers might be present in the same packet.
To match on fragmented IPv6 packets, the same difficulty exists.

This patch implements the update as detailed in RFC [1].
A set of additional values will be added to IPv6 header struct.
These values will indicate the existence of every defined extension
header type, providing simple means for identification of existing
extensions in the packet header.
Continuing the above example, fragmented packets can be identified using
the specific value indicating existence of fragment extension header.

[1] https://mails.dpdk.org/archives/dev/2020-August/177257.html

Signed-off-by: Dekel Peled 
---
 doc/guides/prog_guide/rte_flow.rst | 16 +---
 lib/librte_ethdev/rte_flow.h   | 25 +++--
 2 files changed, 36 insertions(+), 5 deletions(-)

diff --git a/doc/guides/prog_guide/rte_flow.rst 
b/doc/guides/prog_guide/rte_flow.rst
index 119b128..0b476da 100644
--- a/doc/guides/prog_guide/rte_flow.rst
+++ b/doc/guides/prog_guide/rte_flow.rst
@@ -946,11 +946,21 @@ Item: ``IPV6``
 
 Matches an IPv6 header.
 
-Note: IPv6 options are handled by dedicated pattern items, see `Item:
-IPV6_EXT`_.
+Dedicated flags indicate existence of specific extension headers.
+Every type of extension header can use a dedicated pattern item, or
+the generic `Item: IPV6_EXT`_.
 
 - ``hdr``: IPv6 header definition (``rte_ip.h``).
-- Default ``mask`` matches source and destination addresses only.
+- ``hop_ext_exist``: Hop-by-Hop Options extension header exists.
+- ``rout_ext_exist``: Routing extension header exists.
+- ``frag_ext_exist``: Fragment extension header exists.
+- ``auth_ext_exist``: Authentication extension header exists.
+- ``esp_ext_exist``: Encapsulation Security Payload extension header exists.
+- ``dest_ext_exist``: Destination Options extension header exists.
+- ``mobil_ext_exist``: Mobility extension header exists.
+- ``hip_ext_exist``: Host Identity Protocol extension header exists.
+- ``shim6_ext_exist``: Shim6 Protocol extension header exists.
+- Default ``mask`` matches ``hdr`` source and destination addresses only.
 
 Item: ``ICMP``
 ^^
diff --git a/lib/librte_ethdev/rte_flow.h b/lib/librte_ethdev/rte_flow.h
index da8bfa5..5b5bed2 100644
--- a/lib/librte_ethdev/rte_flow.h
+++ b/lib/librte_ethdev/rte_flow.h
@@ -792,11 +792,32 @@ struct rte_flow_item_ipv4 {
  *
  * Matches an IPv6 header.
  *
- * Note: IPv6 options are handled by dedicated pattern items, see
- * RTE_FLOW_ITEM_TYPE_IPV6_EXT.
+ * Dedicated flags indicate existence of specific extension headers.
+ * Every type of extension header can use a dedicated pattern item, or
+ * the generic item RTE_FLOW_ITEM_TYPE_IPV6_EXT.
  */
 struct rte_flow_item_ipv6 {
struct rte_ipv6_hdr hdr; /**< IPv6 header definition. */
+   uint32_t hop_ext_exist:1;
+   /**< Hop-by-Hop Options extension header exists. */
+   uint32_t rout_ext_exist:1;
+   /**< Routing extension header exists. */
+   uint32_t frag_ext_exist:1;
+   /**< Fragment extension header exists. */
+   uint32_t auth_ext_exist:1;
+   /**< Authentication extension header exists. */
+   uint32_t esp_ext_exist:1;
+   /**< Encapsulation Security Payload extension header exists. */
+   uint32_t dest_ext_exist:1;
+   /**< Destination Options extension header exists. */
+   uint32_t mobil_ext_exist:1;
+   /**< Mobility extension header exists. */
+   uint32_t hip_ext_exist:1;
+   /**< Host Identity Protocol extension header exists. */
+   uint32_t shim6_ext_exist:1;
+   /**< Shim6 Protocol extension header exists. */
+   uint32_t reserved:23;
+   /**< Reserved for future extension headers, must be zero. */
 };
 
 /** Default mask for RTE_FLOW_ITEM_TYPE_IPV6. */
-- 
1.8.3.1

[dpdk-dev] [PATCH 02/10] ethdev: add IPv6 fragment extension header item

[Dekel Peled]

Applications handling fragmented IPv6 packets need to match on IPv6
fragment extension header, in order to identify the fragments order
and location in the packet.
This patch introduces the IPv6 fragment extension header item,
proposed in [1].

Relevant definitions are moved from lib/librte_ip_frag/rte_ip_frag.h
to lib/librte_net/rte_ip.h, as they are needed for IPv6 header handling.
struct ipv6_extension_fragment renamed to rte_ipv6_fragment_ext to
adapt it to the common naming convention.

Default mask is not defined, since all fields are optional.

[1] http://mails.dpdk.org/archives/dev/2020-March/160255.html

Signed-off-by: Dekel Peled 
---
 doc/guides/prog_guide/rte_flow.rst | 16 ++--
 lib/librte_ethdev/rte_flow.c   |  1 +
 lib/librte_ethdev/rte_flow.h   | 21 +
 lib/librte_ip_frag/rte_ip_frag.h   | 26 ++
 lib/librte_net/rte_ip.h| 26 --
 5 files changed, 62 insertions(+), 28 deletions(-)

diff --git a/doc/guides/prog_guide/rte_flow.rst 
b/doc/guides/prog_guide/rte_flow.rst
index 0b476da..826e45d 100644
--- a/doc/guides/prog_guide/rte_flow.rst
+++ b/doc/guides/prog_guide/rte_flow.rst
@@ -947,8 +947,8 @@ Item: ``IPV6``
 Matches an IPv6 header.
 
 Dedicated flags indicate existence of specific extension headers.
-Every type of extension header can use a dedicated pattern item, or
-the generic `Item: IPV6_EXT`_.
+Every type of extension header can use a dedicated pattern item,
+for example `Item: IPV6_FRAG_EXT`_, or the generic `Item: IPV6_EXT`_.
 
 - ``hdr``: IPv6 header definition (``rte_ip.h``).
 - ``hop_ext_exist``: Hop-by-Hop Options extension header exists.
@@ -1187,6 +1187,18 @@ Normally preceded by any of:
 - `Item: IPV6`_
 - `Item: IPV6_EXT`_
 
+Item: ``IPV6_FRAG_EXT``
+^^^
+
+Matches the presence of IPv6 fragment extension header.
+
+- ``hdr``: IPv6 fragment extension header definition (``rte_ip.h``).
+
+Normally preceded by any of:
+
+- `Item: IPV6`_
+- `Item: IPV6_EXT`_
+
 Item: ``ICMP6``
 ^^^
 
diff --git a/lib/librte_ethdev/rte_flow.c b/lib/librte_ethdev/rte_flow.c
index f8fdd68..c1f3132 100644
--- a/lib/librte_ethdev/rte_flow.c
+++ b/lib/librte_ethdev/rte_flow.c
@@ -72,6 +72,7 @@ struct rte_flow_desc_data {
MK_FLOW_ITEM(VXLAN_GPE, sizeof(struct rte_flow_item_vxlan_gpe)),
MK_FLOW_ITEM(ARP_ETH_IPV4, sizeof(struct rte_flow_item_arp_eth_ipv4)),
MK_FLOW_ITEM(IPV6_EXT, sizeof(struct rte_flow_item_ipv6_ext)),
+   MK_FLOW_ITEM(IPV6_FRAG_EXT, sizeof(struct rte_flow_item_ipv6_frag_ext)),
MK_FLOW_ITEM(ICMP6, sizeof(struct rte_flow_item_icmp6)),
MK_FLOW_ITEM(ICMP6_ND_NS, sizeof(struct rte_flow_item_icmp6_nd_ns)),
MK_FLOW_ITEM(ICMP6_ND_NA, sizeof(struct rte_flow_item_icmp6_nd_na)),
diff --git a/lib/librte_ethdev/rte_flow.h b/lib/librte_ethdev/rte_flow.h
index 5b5bed2..1443e6a 100644
--- a/lib/librte_ethdev/rte_flow.h
+++ b/lib/librte_ethdev/rte_flow.h
@@ -368,6 +368,13 @@ enum rte_flow_item_type {
RTE_FLOW_ITEM_TYPE_IPV6_EXT,
 
/**
+* Matches the presence of IPv6 fragment extension header.
+*
+* See struct rte_flow_item_ipv6_frag_ext.
+*/
+   RTE_FLOW_ITEM_TYPE_IPV6_FRAG_EXT,
+
+   /**
 * Matches any ICMPv6 header.
 *
 * See struct rte_flow_item_icmp6.
@@ -1188,6 +1195,20 @@ struct rte_flow_item_ipv6_ext 
rte_flow_item_ipv6_ext_mask = {
 #endif
 
 /**
+ * RTE_FLOW_ITEM_TYPE_IPV6_FRAG_EXT
+ *
+ * Matches the presence of IPv6 fragment extension header.
+ *
+ * Preceded by any of:
+ *
+ * - RTE_FLOW_ITEM_TYPE_IPV6
+ * - RTE_FLOW_ITEM_TYPE_IPV6_EXT
+ */
+struct rte_flow_item_ipv6_frag_ext {
+   struct rte_ipv6_fragment_ext hdr;
+};
+
+/**
  * RTE_FLOW_ITEM_TYPE_ICMP6
  *
  * Matches any ICMPv6 header.
diff --git a/lib/librte_ip_frag/rte_ip_frag.h b/lib/librte_ip_frag/rte_ip_frag.h
index 66edd7e..0bfe64b 100644
--- a/lib/librte_ip_frag/rte_ip_frag.h
+++ b/lib/librte_ip_frag/rte_ip_frag.h
@@ -110,30 +110,8 @@ struct rte_ip_frag_tbl {
__extension__ struct ip_frag_pkt pkt[0]; /**< hash table. */
 };
 
-/** IPv6 fragment extension header */
-#defineRTE_IPV6_EHDR_MF_SHIFT  0
-#defineRTE_IPV6_EHDR_MF_MASK   1
-#defineRTE_IPV6_EHDR_FO_SHIFT  3
-#defineRTE_IPV6_EHDR_FO_MASK   (~((1 << 
RTE_IPV6_EHDR_FO_SHIFT) - 1))
-#defineRTE_IPV6_EHDR_FO_ALIGN  (1 << 
RTE_IPV6_EHDR_FO_SHIFT)
-
-#define RTE_IPV6_FRAG_USED_MASK\
-   (RTE_IPV6_EHDR_MF_MASK | RTE_IPV6_EHDR_FO_MASK)
-
-#define RTE_IPV6_GET_MF(x) ((x) & 
RTE_IPV6_EHDR_MF_MASK)
-#define RTE_IPV6_GET_FO(x) ((x) >> 
RTE_IPV6_EHDR_FO_SHIFT)
-
-#define RTE_IPV6_SET_FRAG_DATA(fo, mf) \
-   (((fo) & RTE_IPV6_EHDR_FO_MASK) | ((mf) & RTE_IPV6_EHDR_MF_MASK))
-
-struct ipv6_extension_fragment {
-   uint

[dpdk-dev] [PATCH 03/10] app/testpmd: support IPv4 fragments

[Dekel Peled]

This patch updates testpmd CLI to support fragment_offset field of
IPv4 header item.

To match on non-fragmented IPv4 packets, need to use pattern:
... ipv4 fragment_offset spec 0 fragment_offset mask 0x3fff ...
To match on fragmented IPv4 packets, need to use pattern:
... ipv4 fragment_offset spec 1 fragment_offset last 0x3fff
fragment_offset mask 0x3fff ...
(Use the full available range 1 to 0x3fff to include all possible
values.)
To match on any IPv4 packets, fragmented and non-fragmented,
the fragment_offset field should not be specified for match.

Signed-off-by: Dekel Peled 
---
 app/test-pmd/cmdline_flow.c | 9 +
 1 file changed, 9 insertions(+)

diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c
index 6263d30..0c964c1 100644
--- a/app/test-pmd/cmdline_flow.c
+++ b/app/test-pmd/cmdline_flow.c
@@ -129,6 +129,7 @@ enum index {
ITEM_VLAN_INNER_TYPE,
ITEM_IPV4,
ITEM_IPV4_TOS,
+   ITEM_IPV4_FRAGMENT_OFFSET,
ITEM_IPV4_TTL,
ITEM_IPV4_PROTO,
ITEM_IPV4_SRC,
@@ -871,6 +872,7 @@ struct parse_action_priv {
 
 static const enum index item_ipv4[] = {
ITEM_IPV4_TOS,
+   ITEM_IPV4_FRAGMENT_OFFSET,
ITEM_IPV4_TTL,
ITEM_IPV4_PROTO,
ITEM_IPV4_SRC,
@@ -2093,6 +2095,13 @@ static int comp_set_raw_index(struct context *, const 
struct token *,
.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_ipv4,
 hdr.type_of_service)),
},
+   [ITEM_IPV4_FRAGMENT_OFFSET] = {
+   .name = "fragment_offset",
+   .help = "fragmentation flags and fragment offset",
+   .next = NEXT(item_ipv4, NEXT_ENTRY(UNSIGNED), item_param),
+   .args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_ipv4,
+hdr.fragment_offset)),
+   },
[ITEM_IPV4_TTL] = {
.name = "ttl",
.help = "time to live",
-- 
1.8.3.1

[dpdk-dev] [PATCH 04/10] app/testpmd: support IPv6 fragments

[Dekel Peled]

rte_flow update, following RFC [1], introduced frag_ext_exist field for
IPv6 header item, used to indicate match on fragmented/non-fragmented
packets.
This patch updates testpmd CLI to support the new field.

To match on non-fragmented IPv6 packets, need to use pattern:
... ipv6 frag_ext_exist spec 0 frag_ext_exist mask 1 ...
To match on fragmented IPv6 packets, need to use pattern:
... ipv6 frag_ext_exist spec 1 frag_ext_exist mask 1 ...
To match on any IPv6 packets, the frag_ext_exist field should
not be specified for match.

[1] https://mails.dpdk.org/archives/dev/2020-August/177257.html

Signed-off-by: Dekel Peled 
---
 app/test-pmd/cmdline_flow.c | 9 +
 1 file changed, 9 insertions(+)

diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c
index 0c964c1..55cc30f 100644
--- a/app/test-pmd/cmdline_flow.c
+++ b/app/test-pmd/cmdline_flow.c
@@ -141,6 +141,7 @@ enum index {
ITEM_IPV6_HOP,
ITEM_IPV6_SRC,
ITEM_IPV6_DST,
+   ITEM_IPV6_FRAG_EXT_EXIST,
ITEM_ICMP,
ITEM_ICMP_TYPE,
ITEM_ICMP_CODE,
@@ -888,6 +889,7 @@ struct parse_action_priv {
ITEM_IPV6_HOP,
ITEM_IPV6_SRC,
ITEM_IPV6_DST,
+   ITEM_IPV6_FRAG_EXT_EXIST,
ITEM_NEXT,
ZERO,
 };
@@ -2181,6 +2183,13 @@ static int comp_set_raw_index(struct context *, const 
struct token *,
.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_ipv6,
 hdr.dst_addr)),
},
+   [ITEM_IPV6_FRAG_EXT_EXIST] = {
+   .name = "frag_ext_exist",
+   .help = "fragment packet attribute",
+   .next = NEXT(item_ipv6, NEXT_ENTRY(UNSIGNED), item_param),
+   .args = ARGS(ARGS_ENTRY_BF(struct rte_flow_item_ipv6,
+  frag_ext_exist, 1)),
+   },
[ITEM_ICMP] = {
.name = "icmp",
.help = "match ICMP header",
-- 
1.8.3.1

[dpdk-dev] [PATCH 05/10] app/testpmd: support IPv6 fragment extension item

[Dekel Peled]

rte_flow update, following RFC [1], added to ethdev the rte_flow item
ipv6_frag_ext.
This patch updates testpmd CLI to support the new item and its fields.

To match on fragmented IPv6 packets, this item is added to pattern:
... ipv6 / ipv6_frag_ext ...

[1] http://mails.dpdk.org/archives/dev/2020-March/160255.html

Signed-off-by: Dekel Peled 
---
 app/test-pmd/cmdline_flow.c | 35 +++
 1 file changed, 35 insertions(+)

diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c
index 55cc30f..ef166e3 100644
--- a/app/test-pmd/cmdline_flow.c
+++ b/app/test-pmd/cmdline_flow.c
@@ -193,6 +193,9 @@ enum index {
ITEM_ARP_ETH_IPV4_TPA,
ITEM_IPV6_EXT,
ITEM_IPV6_EXT_NEXT_HDR,
+   ITEM_IPV6_FRAG_EXT,
+   ITEM_IPV6_FRAG_EXT_NEXT_HDR,
+   ITEM_IPV6_FRAG_EXT_FRAG_DATA,
ITEM_ICMP6,
ITEM_ICMP6_TYPE,
ITEM_ICMP6_CODE,
@@ -784,6 +787,7 @@ struct parse_action_priv {
ITEM_VXLAN_GPE,
ITEM_ARP_ETH_IPV4,
ITEM_IPV6_EXT,
+   ITEM_IPV6_FRAG_EXT,
ITEM_ICMP6,
ITEM_ICMP6_ND_NS,
ITEM_ICMP6_ND_NA,
@@ -1003,6 +1007,13 @@ struct parse_action_priv {
ZERO,
 };
 
+static const enum index item_ipv6_frag_ext[] = {
+   ITEM_IPV6_FRAG_EXT_NEXT_HDR,
+   ITEM_IPV6_FRAG_EXT_FRAG_DATA,
+   ITEM_NEXT,
+   ZERO,
+};
+
 static const enum index item_icmp6[] = {
ITEM_ICMP6_TYPE,
ITEM_ICMP6_CODE,
@@ -2560,6 +2571,30 @@ static int comp_set_raw_index(struct context *, const 
struct token *,
.args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_ipv6_ext,
 next_hdr)),
},
+   [ITEM_IPV6_FRAG_EXT] = {
+   .name = "ipv6_frag_ext",
+   .help = "match presence of IPv6 fragment extension header",
+   .priv = PRIV_ITEM(IPV6_FRAG_EXT,
+   sizeof(struct rte_flow_item_ipv6_frag_ext)),
+   .next = NEXT(item_ipv6_frag_ext),
+   .call = parse_vc,
+   },
+   [ITEM_IPV6_FRAG_EXT_NEXT_HDR] = {
+   .name = "next_hdr",
+   .help = "next header",
+   .next = NEXT(item_ipv6_frag_ext, NEXT_ENTRY(UNSIGNED),
+item_param),
+   .args = ARGS(ARGS_ENTRY(struct rte_flow_item_ipv6_frag_ext,
+   hdr.next_header)),
+   },
+   [ITEM_IPV6_FRAG_EXT_FRAG_DATA] = {
+   .name = "frag_data",
+   .help = "Fragment flags and offset",
+   .next = NEXT(item_ipv6_frag_ext, NEXT_ENTRY(UNSIGNED),
+item_param),
+   .args = ARGS(ARGS_ENTRY_HTON(struct rte_flow_item_ipv6_frag_ext,
+hdr.frag_data)),
+   },
[ITEM_ICMP6] = {
.name = "icmp6",
.help = "match any ICMPv6 header",
-- 
1.8.3.1

[dpdk-dev] [PATCH 06/10] net/mlx5: remove handling of ICMP fragmented packets

[Dekel Peled]

Commit [1] forced setting of match on 'frag' bit mask 1 and value 0.
Previous patch in this series added support of match on fragmented and
non-fragmented packets on L3 items, so this setting is now redundant.

This patch removes the changes done in [1].

[1] commit 85407db9f60d ("net/mlx5: fix matching for ICMP fragments")

Signed-off-by: Dekel Peled 
---
 drivers/net/mlx5/mlx5_flow_dv.c | 12 
 1 file changed, 12 deletions(-)

diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
index 79fdf34..0a0a5a4 100644
--- a/drivers/net/mlx5/mlx5_flow_dv.c
+++ b/drivers/net/mlx5/mlx5_flow_dv.c
@@ -7345,12 +7345,6 @@ struct field_modify_info modify_tcp[] = {
return;
if (!icmp6_m)
icmp6_m = &rte_flow_item_icmp6_mask;
-   /*
-* Force flow only to match the non-fragmented IPv6 ICMPv6 packets.
-* If only the protocol is specified, no need to match the frag.
-*/
-   MLX5_SET(fte_match_set_lyr_2_4, headers_m, frag, 1);
-   MLX5_SET(fte_match_set_lyr_2_4, headers_v, frag, 0);
MLX5_SET(fte_match_set_misc3, misc3_m, icmpv6_type, icmp6_m->type);
MLX5_SET(fte_match_set_misc3, misc3_v, icmpv6_type,
 icmp6_v->type & icmp6_m->type);
@@ -7398,12 +7392,6 @@ struct field_modify_info modify_tcp[] = {
return;
if (!icmp_m)
icmp_m = &rte_flow_item_icmp_mask;
-   /*
-* Force flow only to match the non-fragmented IPv4 ICMP packets.
-* If only the protocol is specified, no need to match the frag.
-*/
-   MLX5_SET(fte_match_set_lyr_2_4, headers_m, frag, 1);
-   MLX5_SET(fte_match_set_lyr_2_4, headers_v, frag, 0);
MLX5_SET(fte_match_set_misc3, misc3_m, icmp_type,
 icmp_m->hdr.icmp_type);
MLX5_SET(fte_match_set_misc3, misc3_v, icmp_type,
-- 
1.8.3.1

[dpdk-dev] [PATCH 07/10] net/mlx5: support match on IPv4 fragment packets

[Dekel Peled]

This patch adds to MLX5 PMD the support of matching on IPv4
fragmented and non-fragmented packets, using the IPv4 header
fragment_offset field.

Signed-off-by: Dekel Peled 
---
 drivers/net/mlx5/mlx5_flow.c   |  48 
 drivers/net/mlx5/mlx5_flow.h   |  10 +++
 drivers/net/mlx5/mlx5_flow_dv.c| 156 +++--
 drivers/net/mlx5/mlx5_flow_verbs.c |   9 ++-
 4 files changed, 178 insertions(+), 45 deletions(-)

diff --git a/drivers/net/mlx5/mlx5_flow.c b/drivers/net/mlx5/mlx5_flow.c
index ffa7646..906741f 100644
--- a/drivers/net/mlx5/mlx5_flow.c
+++ b/drivers/net/mlx5/mlx5_flow.c
@@ -474,6 +474,8 @@ struct mlx5_flow_tunnel_info {
  *   Bit-masks covering supported fields by the NIC to compare with user mask.
  * @param[in] size
  *   Bit-masks size in bytes.
+ * @param[in] range_accepted
+ *   True if range of values is accepted for specific fields, false otherwise.
  * @param[out] error
  *   Pointer to error structure.
  *
@@ -485,6 +487,7 @@ struct mlx5_flow_tunnel_info {
  const uint8_t *mask,
  const uint8_t *nic_mask,
  unsigned int size,
+ bool range_accepted,
  struct rte_flow_error *error)
 {
unsigned int i;
@@ -502,7 +505,7 @@ struct mlx5_flow_tunnel_info {
  RTE_FLOW_ERROR_TYPE_ITEM, item,
  "mask/last without a spec is not"
  " supported");
-   if (item->spec && item->last) {
+   if (item->spec && item->last && !range_accepted) {
uint8_t spec[size];
uint8_t last[size];
unsigned int i;
@@ -1277,7 +1280,8 @@ struct mlx5_flow_tunnel_info {
ret = mlx5_flow_item_acceptable
(item, (const uint8_t *)mask,
 (const uint8_t *)&rte_flow_item_icmp6_mask,
-sizeof(struct rte_flow_item_icmp6), error);
+sizeof(struct rte_flow_item_icmp6),
+MLX5_ITEM_RANGE_NOT_ACCEPTED, error);
if (ret < 0)
return ret;
return 0;
@@ -1329,7 +1333,8 @@ struct mlx5_flow_tunnel_info {
ret = mlx5_flow_item_acceptable
(item, (const uint8_t *)mask,
 (const uint8_t *)&rte_flow_item_icmp_mask,
-sizeof(struct rte_flow_item_icmp), error);
+sizeof(struct rte_flow_item_icmp),
+MLX5_ITEM_RANGE_NOT_ACCEPTED, error);
if (ret < 0)
return ret;
return 0;
@@ -1384,7 +1389,7 @@ struct mlx5_flow_tunnel_info {
ret = mlx5_flow_item_acceptable(item, (const uint8_t *)mask,
(const uint8_t *)&nic_mask,
sizeof(struct rte_flow_item_eth),
-   error);
+   MLX5_ITEM_RANGE_NOT_ACCEPTED, error);
return ret;
 }
 
@@ -1438,7 +1443,7 @@ struct mlx5_flow_tunnel_info {
ret = mlx5_flow_item_acceptable(item, (const uint8_t *)mask,
(const uint8_t *)&nic_mask,
sizeof(struct rte_flow_item_vlan),
-   error);
+   MLX5_ITEM_RANGE_NOT_ACCEPTED, error);
if (ret)
return ret;
if (!tunnel && mask->tci != RTE_BE16(0x0fff)) {
@@ -1502,6 +1507,7 @@ struct mlx5_flow_tunnel_info {
 uint64_t last_item,
 uint16_t ether_type,
 const struct rte_flow_item_ipv4 *acc_mask,
+bool range_accepted,
 struct rte_flow_error *error)
 {
const struct rte_flow_item_ipv4 *mask = item->mask;
@@ -1572,7 +1578,7 @@ struct mlx5_flow_tunnel_info {
acc_mask ? (const uint8_t *)acc_mask
 : (const uint8_t *)&nic_mask,
sizeof(struct rte_flow_item_ipv4),
-   error);
+   range_accepted, error);
if (ret < 0)
return ret;
return 0;
@@ -1592,6 +1598,8 @@ struct mlx5_flow_tunnel_info {
  * @param[in] acc_mask
  *   Acceptable mask, if NULL default internal default mask
  *   will be used to check whether item fields are supported.
+ * @param[in] range_accepted
+ *   True if range of values is accepted for specific fields, false otherwise.
  * @param[out] error
  *   Pointer to error structure.
  *
@@ -1671,7 +1679,7 @@ struct mlx5_flow_tunnel_info {
acc_mask ? (const uint8_t *)acc_mask
 : (const uint8_t *)&nic_mask,

[dpdk-dev] [PATCH 08/10] net/mlx5: support match on IPv6 fragment packets

[Dekel Peled]

This patch adds to MLX5 PMD the support of matching on IPv6
fragmented and non-fragmented packets, using the new field
frag_ext_exist, added to rte_flow following RFC [1].

[1] https://mails.dpdk.org/archives/dev/2020-August/177257.html

Signed-off-by: Dekel Peled 
---
 drivers/net/mlx5/mlx5_flow_dv.c | 5 +
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
index 3379caf..4403abc 100644
--- a/drivers/net/mlx5/mlx5_flow_dv.c
+++ b/drivers/net/mlx5/mlx5_flow_dv.c
@@ -5211,6 +5211,7 @@ struct field_modify_info modify_tcp[] = {
.proto = 0xff,
.hop_limits = 0xff,
},
+   .frag_ext_exist = 1,
};
const struct rte_flow_item_ecpri nic_ecpri_mask = {
.hdr = {
@@ -6519,6 +6520,10 @@ struct field_modify_info modify_tcp[] = {
 ipv6_m->hdr.hop_limits);
MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_ttl_hoplimit,
 ipv6_v->hdr.hop_limits & ipv6_m->hdr.hop_limits);
+   MLX5_SET(fte_match_set_lyr_2_4, headers_m, frag,
+!!(ipv6_m->frag_ext_exist));
+   MLX5_SET(fte_match_set_lyr_2_4, headers_v, frag,
+!!(ipv6_v->frag_ext_exist & ipv6_m->frag_ext_exist));
 }
 
 /**
-- 
1.8.3.1

[dpdk-dev] [PATCH 09/10] net/mlx5: support match on IPv6 fragment ext. item

[Dekel Peled]

rte_flow update, following RFC [1], added to ethdev the rte_flow item
ipv6_frag_ext.
This patch adds to MLX5 PMD the option to match on this item type.

[1] http://mails.dpdk.org/archives/dev/2020-March/160255.html

Signed-off-by: Dekel Peled 
---
 drivers/net/mlx5/mlx5_flow.h|   4 +
 drivers/net/mlx5/mlx5_flow_dv.c | 209 
 2 files changed, 213 insertions(+)

diff --git a/drivers/net/mlx5/mlx5_flow.h b/drivers/net/mlx5/mlx5_flow.h
index 1e30c93..376519f 100644
--- a/drivers/net/mlx5/mlx5_flow.h
+++ b/drivers/net/mlx5/mlx5_flow.h
@@ -122,6 +122,10 @@ enum mlx5_feature_name {
 /* Pattern eCPRI Layer bit. */
 #define MLX5_FLOW_LAYER_ECPRI (UINT64_C(1) << 29)
 
+/* IPv6 Fragment Extension Header bit. */
+#define MLX5_FLOW_LAYER_OUTER_L3_IPV6_FRAG_EXT (1u << 30)
+#define MLX5_FLOW_LAYER_INNER_L3_IPV6_FRAG_EXT (1u << 31)
+
 /* Outer Masks. */
 #define MLX5_FLOW_LAYER_OUTER_L3 \
(MLX5_FLOW_LAYER_OUTER_L3_IPV4 | MLX5_FLOW_LAYER_OUTER_L3_IPV6)
diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
index 4403abc..eb1db12 100644
--- a/drivers/net/mlx5/mlx5_flow_dv.c
+++ b/drivers/net/mlx5/mlx5_flow_dv.c
@@ -1901,6 +1901,120 @@ struct field_modify_info modify_tcp[] = {
 }
 
 /**
+ * Validate IPV6 fragment extension item.
+ *
+ * @param[in] item
+ *   Item specification.
+ * @param[in] item_flags
+ *   Bit-fields that holds the items detected until now.
+ * @param[out] error
+ *   Pointer to error structure.
+ *
+ * @return
+ *   0 on success, a negative errno value otherwise and rte_errno is set.
+ */
+static int
+flow_dv_validate_item_ipv6_frag_ext(const struct rte_flow_item *item,
+   uint64_t item_flags,
+   struct rte_flow_error *error)
+{
+   const struct rte_flow_item_ipv6_frag_ext *spec = item->spec;
+   const struct rte_flow_item_ipv6_frag_ext *last = item->last;
+   const struct rte_flow_item_ipv6_frag_ext *mask = item->mask;
+   rte_be16_t frag_data_spec = 0;
+   rte_be16_t frag_data_last = 0;
+   const int tunnel = !!(item_flags & MLX5_FLOW_LAYER_TUNNEL);
+   const uint64_t l4m = tunnel ? MLX5_FLOW_LAYER_INNER_L4 :
+ MLX5_FLOW_LAYER_OUTER_L4;
+   int ret = 0;
+   struct rte_flow_item_ipv6_frag_ext nic_mask = {
+   .hdr = {
+   .next_header = 0xff,
+   .frag_data = RTE_BE16(0x),
+   },
+   };
+
+   if (item_flags & l4m)
+   return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "ipv6 fragment extension item cannot "
+ "follow L4 item.");
+   if ((tunnel && !(item_flags & MLX5_FLOW_LAYER_INNER_L3_IPV6)) ||
+   (!tunnel && !(item_flags & MLX5_FLOW_LAYER_OUTER_L3_IPV6)))
+   return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "ipv6 fragment extension item must "
+ "follow ipv6 item");
+   if (spec && mask)
+   frag_data_spec = spec->hdr.frag_data & mask->hdr.frag_data;
+   if (!frag_data_spec)
+   return 0;
+   /*
+* spec and mask are valid, enforce using full mask to make sure the
+* complete value is used correctly.
+*/
+   if ((mask->hdr.frag_data & RTE_BE16(RTE_IPV6_FRAG_USED_MASK)) !=
+   RTE_BE16(RTE_IPV6_FRAG_USED_MASK))
+   return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM_MASK,
+ item, "must use full mask for"
+ " frag_data");
+   /*
+* Match on frag_data 0x1 means M is 1 and frag-offset is 0.
+* This is 1st fragment of fragmented packet.
+*/
+   if (frag_data_spec == RTE_BE16(RTE_IPV6_EHDR_MF_MASK))
+   return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "match on first fragment not "
+ "supported");
+   if (frag_data_spec && !last)
+   return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "specified value not supported");
+   ret = mlx5_flow_item_acceptable
+   (item, (const uint8_t *)mask,
+(const uint8_t *)&nic_mask,
+sizeof(struct rte_flow_item_ipv6_frag_ext),
+MLX5_ITEM_RANGE_ACCEPTED, error);
+   if (ret)
+   r

[dpdk-dev] [PATCH 10/10] doc: update release notes for MLX5 L3 frag support

[Dekel Peled]

This patch updates 20.11 release notes with the changes included in
patches of this series:
1) MLX5 support of matching on IPv4/IPv6 fragmented/non-fragmented
   packets.
2) ABI change in ethdev struct rte_flow_item_ipv6.

Signed-off-by: Dekel Peled 
---
 doc/guides/rel_notes/release_20_11.rst | 10 ++
 1 file changed, 10 insertions(+)

diff --git a/doc/guides/rel_notes/release_20_11.rst 
b/doc/guides/rel_notes/release_20_11.rst
index 4bcf220..60d3243 100644
--- a/doc/guides/rel_notes/release_20_11.rst
+++ b/doc/guides/rel_notes/release_20_11.rst
@@ -84,6 +84,11 @@ New Features
 
   * Added support for 200G PAM4 link speed.
 
+* **Updated Mellanox mlx5 driver.**
+
+  Updated Mellanox mlx5 driver with new features and improvements, including:
+
+  * Added support for matching on fragmented/non-fragmented IPv4/IPv6 packets.
 
 Removed Items
 -
@@ -209,6 +214,11 @@ ABI Changes
 
   * ``ethdev`` internal functions are marked with ``__rte_internal`` tag.
 
+  * Added extensions' attributes to struct ``rte_flow_item_ipv6``.
+A set of additional values added to struct, indicating the existence of
+every defined extension header type.
+Applications should use the new values for identification of existing
+extensions in the packet header.
 
 Known Issues
 
-- 
1.8.3.1