Re: [dpdk-dev] [PATCH] net/mlx5: add support for 32bit systems

[Shahaf Shuler]

Hi Moty,

Few nits,

Also please fix the check patch warning :
### net/mlx5: add support for 32bit systems  
 
CHECK:OPEN_ENDED_LINE: Lines should not end with a '('   
#235: FILE: drivers/net/mlx5/mlx5_rxtx.c:1591:   
+   addr_64 = rte_cpu_to_be_64(  
 
total: 0 errors, 0 warnings, 1 checks, 311 lines checked 
 


Thursday, June 28, 2018 10:13 AM, Moti Haimovsky:
> Subject: [dpdk-dev] [PATCH] net/mlx5: add support for 32bit systems
> 
> This patch adds support for building and running mlx5 PMD on 32bit systems
> such as i686.
> 
> The main issue to tackle was handling the 32bit access to the UAR as quoted
> from the mlx5 PRM:
> QP and CQ DoorBells require 64-bit writes. For best performance, it is
> recommended to execute the QP/CQ DoorBell as a single 64-bit write
> operation. For platforms that do not support 64 bit writes, it is possible to
> issue the 64 bits DoorBells through two consecutive writes, each write 32
> bits, as described below:
> * The order of writing each of the Dwords is from lower to upper
>   addresses.
> * No other DoorBell can be rung (or even start ringing) in the midst of
>   an on-going write of a DoorBell over a given UAR page.
> The last rule implies that in a multi-threaded environment, the access to a
> UAR page (which can be accessible by all threads in the process) must be
> synchronized (for example, using a semaphore) unless an atomic write of 64
> bits in a single bus operation is guaranteed. Such a synchronization is not
> required for when ringing DoorBells on different UAR pages.
> 
> Signed-off-by: Moti Haimovsky 
> ---
>  doc/guides/nics/features/mlx5.ini |  1 +
>  doc/guides/nics/mlx5.rst  | 11 +++
>  drivers/net/mlx5/mlx5.c   |  8 -
>  drivers/net/mlx5/mlx5.h   |  5 +++
>  drivers/net/mlx5/mlx5_defs.h  | 18 --
>  drivers/net/mlx5/mlx5_rxq.c   |  6 +++-
>  drivers/net/mlx5/mlx5_rxtx.c  | 22 +++--
>  drivers/net/mlx5/mlx5_rxtx.h  | 69
> ++-
>  drivers/net/mlx5/mlx5_txq.c   | 13 +++-
>  9 files changed, 137 insertions(+), 16 deletions(-)
> 
> diff --git a/doc/guides/nics/features/mlx5.ini
> b/doc/guides/nics/features/mlx5.ini
> index e75b14b..b28b43e 100644
> --- a/doc/guides/nics/features/mlx5.ini
> +++ b/doc/guides/nics/features/mlx5.ini
> @@ -43,5 +43,6 @@ Multiprocess aware   = Y
>  Other kdrv   = Y
>  ARMv8= Y
>  Power8   = Y
> +x86-32   = Y
>  x86-64   = Y
>  Usage doc= Y
> diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst index
> 7dd9c1c..cb9d5d8 100644
> --- a/doc/guides/nics/mlx5.rst
> +++ b/doc/guides/nics/mlx5.rst
> @@ -50,6 +50,8 @@ Features
>  
> 
>  - Multi arch support: x86_64, POWER8, ARMv8.
> +- Support for i686 is available only when working with
> +  rdma-core version 18.0 or above, built with 32bit support.

I think we can just add i686 to the supported arch. The limitation on the 
rdma-core version is well documented below.

>  - Multiple TX and RX queues.
>  - Support for scattered TX and RX frames.
>  - IPv4, IPv6, TCPv4, TCPv6, UDPv4 and UDPv6 RSS on any number of queues.
> @@ -136,6 +138,11 @@ Limitations
>enabled (``rxq_cqe_comp_en``) at the same time, RSS hash result is not
> fully
>supported. Some Rx packets may not have PKT_RX_RSS_HASH.
> 
> +- Building for i686 is only supported with:
> +
> +  - rdma-core version 18.0 or above built with 32bit support.
> +  - Kernel version 4.14.41 or above.

Why the kernel is related? The rdma-core I understand. 

> +
>  Statistics
>  --
> 
> @@ -477,6 +484,10 @@ RMDA Core with Linux Kernel
>  - Minimal kernel version : v4.14 or the most recent 4.14-rc (see `Linux
> installation documentation`_)
>  - Minimal rdma-core version: v15+ commit 0c5f5765213a ("Merge pull
> request #227 from yishaih/tm")
>(see `RDMA Core installation documentation`_)
> +- When building for i686 use:
> +
> +  - rdma-core version 18.0 or above built with 32bit support.
> +  - Kernel version 4.14.41 or above.
> 
>  .. _`Linux installation documentation`:
> https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgit.
> kernel.org%2Fpub%2Fscm%2Flinux%2Fkernel%2Fgit%2Fstable%2Flinux-
> stable.git%2Fplain%2FDocumentation%2Fadmin-
> guide%2FREADME.rst&data=02%7C01%7Cshahafs%40mellanox.com%7C3793
> 359a175d46b47c2508d5dcc69ff1%7Ca652971c7d2e4d9ba6a4d149256f461b%7
> C0%7C0%7C636657668016130861&sdata=yFHd7tQET5SqIcPgj66BSuwJp3sydo
> ujC0ldCMkChVE%3D&reserved=0
>  .. _`RDMA Core installation documentation`:
> https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fra
> w.githubusercontent.com%2Flinux-rdma%2Frdma-
> core%2Fmaster%2FREADME.md&data=02%7C01%7Cshahafs%40mellanox.co
> m%7C379335

Re: [dpdk-dev] [PATCH v4 5/9] net/virtio: free in-order descriptors before device start

[Maxime Coquelin]

On 07/01/2018 01:50 AM, Marvin Liu wrote:

Add new function for freeing IN_ORDER descriptors. As descriptors will
be allocated and freed sequentially when IN_ORDER feature was
negotiated. There will be no need to utilize chain for freed descriptors
management, only index update is enough.

Signed-off-by: Marvin Liu 



Reviewed-by: Maxime Coquelin 

[dpdk-dev] [PATCH v7] net/fm10k: add support for check descriptor status APIs

[Zhao Wei]

rte_eth_rx_descritpr_status and rte_eth_tx_descriptor_status
are supported by fm10K.

Signed-off-by: Wei Zhao 

---

v2:
-fix DD check error in tx descriptor

v3:
-fix DD check index error

v4:
-fix error in RS bit list poll

v5:
-rebase code to branch and delete useless variable

v6:
-change release note

v7:
-update update fm10k.ini file
---
 doc/guides/nics/features/fm10k.ini |  2 +
 doc/guides/rel_notes/release_18_08.rst |  6 +++
 drivers/net/fm10k/fm10k.h  |  7 +++
 drivers/net/fm10k/fm10k_ethdev.c   |  2 +
 drivers/net/fm10k/fm10k_rxtx.c | 78 ++
 5 files changed, 95 insertions(+)

diff --git a/doc/guides/nics/features/fm10k.ini 
b/doc/guides/nics/features/fm10k.ini
index 58f58b9..0acdf0d 100644
--- a/doc/guides/nics/features/fm10k.ini
+++ b/doc/guides/nics/features/fm10k.ini
@@ -26,6 +26,8 @@ VLAN offload = Y
 L3 checksum offload  = Y
 L4 checksum offload  = Y
 Packet type parsing  = Y
+Rx descriptor status = Y
+Tx descriptor status = Y
 Basic stats  = Y
 Extended stats   = Y
 Stats per queue  = Y
diff --git a/doc/guides/rel_notes/release_18_08.rst 
b/doc/guides/rel_notes/release_18_08.rst
index 68aaee8..ac54121 100644
--- a/doc/guides/rel_notes/release_18_08.rst
+++ b/doc/guides/rel_notes/release_18_08.rst
@@ -54,6 +54,12 @@ New Features
   PMD does not provide any. The provision of such tuned values now includes
   the ixgbe PMD.
 
+* **Added fm10k ethernet driver to support check descriptor status APIs.**
+
+  rte_eth_rx_descritpr_status and rte_eth_tx_descriptor_status
+  are supported by fm10K.
+
+
 API Changes
 ---
 
diff --git a/drivers/net/fm10k/fm10k.h b/drivers/net/fm10k/fm10k.h
index ef30780..1bc2c18 100644
--- a/drivers/net/fm10k/fm10k.h
+++ b/drivers/net/fm10k/fm10k.h
@@ -329,6 +329,13 @@ uint16_t fm10k_recv_scattered_pkts(void *rx_queue,
 int
 fm10k_dev_rx_descriptor_done(void *rx_queue, uint16_t offset);
 
+int
+fm10k_dev_rx_descriptor_status(void *rx_queue, uint16_t offset);
+
+int
+fm10k_dev_tx_descriptor_status(void *rx_queue, uint16_t offset);
+
+
 uint16_t fm10k_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
uint16_t nb_pkts);
 
diff --git a/drivers/net/fm10k/fm10k_ethdev.c b/drivers/net/fm10k/fm10k_ethdev.c
index 3ff1b0e..ea2f2bf 100644
--- a/drivers/net/fm10k/fm10k_ethdev.c
+++ b/drivers/net/fm10k/fm10k_ethdev.c
@@ -2837,6 +2837,8 @@ static const struct eth_dev_ops fm10k_eth_dev_ops = {
.tx_queue_setup = fm10k_tx_queue_setup,
.tx_queue_release   = fm10k_tx_queue_release,
.rx_descriptor_done = fm10k_dev_rx_descriptor_done,
+   .rx_descriptor_status = fm10k_dev_rx_descriptor_status,
+   .tx_descriptor_status = fm10k_dev_tx_descriptor_status,
.rx_queue_intr_enable   = fm10k_dev_rx_queue_intr_enable,
.rx_queue_intr_disable  = fm10k_dev_rx_queue_intr_disable,
.reta_update= fm10k_reta_update,
diff --git a/drivers/net/fm10k/fm10k_rxtx.c b/drivers/net/fm10k/fm10k_rxtx.c
index 9320748..4a5b46e 100644
--- a/drivers/net/fm10k/fm10k_rxtx.c
+++ b/drivers/net/fm10k/fm10k_rxtx.c
@@ -389,6 +389,84 @@ fm10k_dev_rx_descriptor_done(void *rx_queue, uint16_t 
offset)
return ret;
 }
 
+int
+fm10k_dev_rx_descriptor_status(void *rx_queue, uint16_t offset)
+{
+   volatile union fm10k_rx_desc *rxdp;
+   struct fm10k_rx_queue *rxq = rx_queue;
+   uint16_t nb_hold, trigger_last;
+   uint16_t desc;
+   int ret;
+
+   if (unlikely(offset >= rxq->nb_desc)) {
+   PMD_DRV_LOG(ERR, "Invalid RX descriptor offset %u", offset);
+   return 0;
+   }
+
+   if (rxq->next_trigger < rxq->alloc_thresh)
+   trigger_last = rxq->next_trigger +
+   rxq->nb_desc - rxq->alloc_thresh;
+   else
+   trigger_last = rxq->next_trigger - rxq->alloc_thresh;
+
+   if (rxq->next_dd < trigger_last)
+   nb_hold = rxq->next_dd + rxq->nb_desc - trigger_last;
+   else
+   nb_hold = rxq->next_dd - trigger_last;
+
+   if (offset >= rxq->nb_desc - nb_hold)
+   return RTE_ETH_RX_DESC_UNAVAIL;
+
+   desc = rxq->next_dd + offset;
+   if (desc >= rxq->nb_desc)
+   desc -= rxq->nb_desc;
+
+   rxdp = &rxq->hw_ring[desc];
+
+   ret = !!(rxdp->w.status &
+   rte_cpu_to_le_16(FM10K_RXD_STATUS_DD));
+
+   return ret;
+}
+
+int
+fm10k_dev_tx_descriptor_status(void *tx_queue, uint16_t offset)
+{
+   volatile struct fm10k_tx_desc *txdp;
+   struct fm10k_tx_queue *txq = tx_queue;
+   uint16_t desc;
+   uint16_t next_rs = txq->nb_desc;
+   struct fifo rs_tracker = txq->rs_tracker;
+   struct fifo *r = &rs_tracker;
+
+   if (unlikely(offset >= txq->nb_desc))
+   return -EINVAL;
+
+   desc = txq->next_free + offset;
+   /* go to next desc that has the RS bit */
+   desc = (desc / txq->rs_thresh + 1) *

Re: [dpdk-dev] [PATCH v4] net/e1000: add support for check descriptor status APIs

[Zhao1, Wei]

Hi,  Ferruh

The doc of igb_vf.ini has enable "Rx  descriptor status" & "Tx 
descriptor status" features, so I do not need update this doc in this patch.

Thanks

> -Original Message-
> From: Zhao1, Wei
> Sent: Friday, June 29, 2018 9:53 AM
> To: dev@dpdk.org
> Cc: Zhang, Qi Z ; Zhao1, Wei 
> Subject: [PATCH v4] net/e1000: add support for check descriptor status APIs
> 
> rte_eth_rx_descritpr_status and rte_eth_tx_descriptor_status are
> supported by igb VF.
> 
> Signed-off-by: Wei Zhao 
> 
> ---
> 
> v2:
> -add release note document info
> 
> v3:
> -rebase code and change git log
> 
> v4:
> -rebase code
> ---
>  doc/guides/rel_notes/release_18_08.rst | 4 ++--
>  drivers/net/e1000/igb_ethdev.c | 3 +++
>  2 files changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/doc/guides/rel_notes/release_18_08.rst
> b/doc/guides/rel_notes/release_18_08.rst
> index ac54121..d7613a2 100644
> --- a/doc/guides/rel_notes/release_18_08.rst
> +++ b/doc/guides/rel_notes/release_18_08.rst
> @@ -54,10 +54,10 @@ New Features
>PMD does not provide any. The provision of such tuned values now
> includes
>the ixgbe PMD.
> 
> -* **Added fm10k ethernet driver to support check descriptor status
> APIs.**
> +* **Added fm10k and igb VF ethernet driver to support check descriptor
> +status APIs.**
> 
>rte_eth_rx_descritpr_status and rte_eth_tx_descriptor_status
> -  are supported by fm10K.
> +  are supported by fm10K and igb VF.
> 
> 
>  API Changes
> diff --git a/drivers/net/e1000/igb_ethdev.c
> b/drivers/net/e1000/igb_ethdev.c index edc7be3..f5b02b2 100644
> --- a/drivers/net/e1000/igb_ethdev.c
> +++ b/drivers/net/e1000/igb_ethdev.c
> @@ -435,6 +435,9 @@ static const struct eth_dev_ops igbvf_eth_dev_ops =
> {
>   .dev_supported_ptypes_get = eth_igb_supported_ptypes_get,
>   .rx_queue_setup   = eth_igb_rx_queue_setup,
>   .rx_queue_release = eth_igb_rx_queue_release,
> + .rx_descriptor_done   = eth_igb_rx_descriptor_done,
> + .rx_descriptor_status = eth_igb_rx_descriptor_status,
> + .tx_descriptor_status = eth_igb_tx_descriptor_status,
>   .tx_queue_setup   = eth_igb_tx_queue_setup,
>   .tx_queue_release = eth_igb_tx_queue_release,
>   .set_mc_addr_list = eth_igb_set_mc_addr_list,
> --
> 2.7.5

Re: [dpdk-dev] [PATCH v4 0/5] eventdev: add interrupt driven queues to Rx adapter

[Jerin Jacob]

-Original Message-
> Date: Mon, 2 Jul 2018 07:25:25 +0530
> From: Nikhil Rao 
> To: jerin.ja...@caviumnetworks.com
> CC: nikhil@intel.com, dev@dpdk.org
> Subject: [PATCH v4 0/5] eventdev: add interrupt driven queues to Rx adapter
> X-Mailer: git-send-email 1.8.3.1
> 
> 
> This patch series adds support for interrupt driven queues to the
> ethernet Rx adapter, the first 3 patches prepare the code to
> handle both poll and interrupt driven Rx queues, the 4th patch
> patch has code changes specific to interrupt driven queues and
> the final patch has test code.
> 
> Changelog:
> v3->v4:
> 
> * Fix FreeBSD build breakage.

# meson shared lib is failing with patch:

[226/1298] Compiling C object
'lib/lib@@rte_eventdev@sta/librte_eventdev_rte_event_eth_rx_adapter.c.o'.
FAILED:
lib/lib@@rte_eventdev@sta/librte_eventdev_rte_event_eth_rx_adapter.c.o
ccache gcc -Ilib/lib@@rte_eventdev@sta -Ilib -I../lib
-Ilib/librte_eventdev -I../lib/librte_eventdev -I. -I../ -Iconfig
-I../config -Ilib/librte_eal/common -I../lib/librte_eal/common
-Ilib/librte_eal/common/include -I../lib/librte_eal/common/include
-Ilib/librte_eal/common/include/arch/x86
-I../lib/librte_eal/common/include/arch/x86
-I../lib/librte_eal/linuxapp/eal/include
-Ilib/librte_eal/linuxapp/eal/../../../librte_compat
-I../lib/librte_eal/linuxapp/eal/../../../librte_compat -Ilib/librte_eal
-I../lib/librte_eal -Ilib/librte_compat -I../lib/librte_compat
-Ilib/librte_ring -I../lib/librte_ring -Ilib/librte_ethdev
-I../lib/librte_ethdev -Ilib/librte_net -I../lib/librte_net
-Ilib/librte_mbuf -I../lib/librte_mbuf -Ilib/librte_mempool
-I../lib/librte_mempool -Ilib/librte_kvargs -I../lib/librte_kvargs
-Ilib/librte_hash -I../lib/librte_hash -Ilib/librte_timer
-I../lib/librte_timer -Ilib/librte_cryptodev -I../lib/librte_cryptodev
-fdiagnostics-color=always -pipe -D_FILE_OFFSET_BITS=64 -Wall
-Winvalid-pch -Werror -O3 -include rte_config.h -Wsign-compare
-Wcast-qual -Wno-format-truncation -fPIC -march=native
-DALLOW_EXPERIMENTAL_API  -MD -MQ
'lib/lib@@rte_eventdev@sta/librte_eventdev_rte_event_eth_rx_adapter.c.o'
-MF
'lib/lib@@rte_eventdev@sta/librte_eventdev_rte_event_eth_rx_adapter.c.o.d'
-o
'lib/lib@@rte_eventdev@sta/librte_eventdev_rte_event_eth_rx_adapter.c.o'
-c ../lib/librte_eventdev/rte_event_eth_rx_adapter.c
../lib/librte_eventdev/rte_event_eth_rx_adapter.c: In function
‘rxa_epoll_create1’:
../lib/librte_eventdev/rte_event_eth_rx_adapter.c:1249:1: error: no
return statement in function returning non-void [-Werror=return-type]
 }
 ^
cc1: all warnings being treated as errors


# please rebase this and following patches to latest next-eventdev tree

http://patches.dpdk.org/patch/41795/



> 
> v2->v3:
> 
> * Fix shared build breakage.
> 
> * Fix FreeBSD build breakage.
> 
> * Reduce epoll maxevents parameter by 1, since thread wakeup
>   uses pthread_cancel as opposed to an exit message through a
>   file monitored by epoll_wait().
> 
> * Check intr_handle before access, it is NULL when zero Rx queue
>   interrupts are configured.
> 
> * Remove thread_stop flag, in the event of a pthread_cancel, it is
>   not possible to check this flag thread stack is unwound without
>   returning to rxa_intr_thread.
> 
> v1->v2:
> 
> * Move rte_service_component_runstate_set such that it
>   is called only when cap & RTE__EVENT_ETH_RX_ADAPTER_CAP_INTERNAL_PORT
>   is false. (Jerin Jacob)
> 
> * Fix meson build. (Jerin Jacob)
> 
> * Replace calls to pthread_* with rte_ctrl_thread_create().
>   (Jerin Jacob)
> 
> * Move adapter test code to separate patch. (Jerin Jacob)
> 
> Note: I haven't removed the note about devices created
> rte_event_eth_rx_adapter_create, will fix in a separate patch.
> 
> 
> Nikhil Rao (5):
>   eventdev: standardize Rx adapter internal function names
>   eventdev: improve err handling for Rx adapter queue add/del
>   eventdev: move Rx adapter eth Rx to separate function
>   eventdev: add interrupt driven queues to Rx adapter
>   eventdev: add Rx adapter tests for interrupt driven queues
> 
>  config/rte_config.h|1 +
>  lib/librte_eventdev/rte_event_eth_rx_adapter.h |5 +-
>  lib/librte_eventdev/rte_event_eth_rx_adapter.c | 1526 
> +---
>  test/test/test_event_eth_rx_adapter.c  |  261 +++-
>  .../prog_guide/event_ethernet_rx_adapter.rst   |   24 +
>  config/common_base |1 +
>  lib/librte_eventdev/Makefile   |9 +-
>  7 files changed, 1588 insertions(+), 239 deletions(-)
> 
> --
> 1.8.3.1
> 

[dpdk-dev] [PATCH v6 00/15] Vhost: add support to packed ring layout

[Maxime Coquelin]

This series is a handover from Jen's "[PATCH v4 00/20]
implement packed virtqueues", which only implements the
vhost side. Virtio PMD implementation will follow in a 
next series.

The series applies on top of previous reworks I posted
during this cycle that merges mergeable and non-mergeable
receive paths, and refactors transmit path to re-use
vector buffers.

I haven't run performance tests for now as the Virtio PMD
side isn't ready.

The series has been tested with Tiwei's series implementing
packed ring support to Kernel's virtio-net driver, and
with Wei series implementing the Qemu side.

To test it, I have used testpmd on host side with a vhost
vdev and a tap vdev forwarding to each other. Transferts
of big random files have been done in both ways with
integrity verified.

Tests have been run with Rx mrg ON/OFF and events suppression
ON/OFF.

Tests have also been run with legacy spli ring layout to
ensure no regression have been introduced.

Changes since v5:
=
- Remove duplicated VHOST_USER_F_PROTOCOL_FEATURES definition (Tiwei)
- Fix vq_is_ready (Maxime)
- Fix local used index overflow in flush_shadow_used_ring_packed (Tiwei)
- Make desc_is_avail() a bool (Tiwei)
- Improve desc_is_avail() logic (Tiwei)
- Remove split rings addr NULL assignment in the right patch (Tiwei)
- Make buffer id a uint16_t (Tiwei)

Jens Freimann (2):
  vhost: add virtio packed virtqueue defines
  vhost: add helpers for packed virtqueues

Maxime Coquelin (12):
  vhost: clear shadow used table index at flush time
  vhost: make indirect desc table copy desc type agnostic
  vhost: clear batch copy index at copy time
  vhost: extract split ring handling from Rx and Tx functions
  vhost: append shadow used ring function names with split
  vhost: add shadow used ring support for packed rings
  vhost: create descriptor mapping function
  vhost: add vector filling support for packed ring
  vhost: add Rx support for packed ring
  vhost: add Tx support for packed ring
  vhost: add notification for packed ring
  vhost: advertize packed ring layout support

Yuanhan Liu (1):
  vhost: vring address setup for packed queues

 lib/librte_vhost/vhost.c | 126 ++-
 lib/librte_vhost/vhost.h | 104 +-
 lib/librte_vhost/vhost_user.c| 128 ++-
 lib/librte_vhost/virtio-packed.h |  41 +++
 lib/librte_vhost/virtio_net.c| 772 ++-
 5 files changed, 963 insertions(+), 208 deletions(-)
 create mode 100644 lib/librte_vhost/virtio-packed.h

-- 
2.14.4

[dpdk-dev] [PATCH v6 01/15] vhost: add virtio packed virtqueue defines

[Maxime Coquelin]

From: Jens Freimann 

Signed-off-by: Jens Freimann 
---
 lib/librte_vhost/vhost.h |  3 +++
 lib/librte_vhost/virtio-packed.h | 22 ++
 2 files changed, 25 insertions(+)
 create mode 100644 lib/librte_vhost/virtio-packed.h

diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index 3f8fa3a78..cecd5c1ee 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -191,6 +191,9 @@ struct vhost_msg {
 #ifndef VIRTIO_F_VERSION_1
  #define VIRTIO_F_VERSION_1 32
 #endif
+#ifndef VIRTIO_F_RING_PACKED
+ #define VIRTIO_F_RING_PACKED 34
+#endif
 
 /* Features supported by this builtin vhost-user net driver. */
 #define VIRTIO_NET_SUPPORTED_FEATURES ((1ULL << VIRTIO_NET_F_MRG_RXBUF) | \
diff --git a/lib/librte_vhost/virtio-packed.h b/lib/librte_vhost/virtio-packed.h
new file mode 100644
index 0..744b3991b
--- /dev/null
+++ b/lib/librte_vhost/virtio-packed.h
@@ -0,0 +1,22 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) Red Hat Inc.
+ */
+
+#ifndef __VIRTIO_PACKED_H
+#define __VIRTIO_PACKED_H
+
+#define VRING_DESC_F_NEXT   1
+#define VRING_DESC_F_WRITE  2
+#define VRING_DESC_F_INDIRECT   4
+
+#define VRING_DESC_F_AVAIL  (1ULL << 7)
+#define VRING_DESC_F_USED  (1ULL << 15)
+
+struct vring_desc_packed {
+   uint64_t addr;
+   uint32_t len;
+   uint16_t index;
+   uint16_t flags;
+};
+
+#endif /* __VIRTIO_PACKED_H */
-- 
2.14.4

[dpdk-dev] [PATCH v6 04/15] vhost: clear shadow used table index at flush time

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/virtio_net.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 2c13a59a2..28ce2d073 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -115,6 +115,7 @@ flush_shadow_used_ring(struct virtio_net *dev, struct 
vhost_virtqueue *vq)
vhost_log_cache_sync(dev, vq);
 
*(volatile uint16_t *)&vq->used->idx += vq->shadow_used_idx;
+   vq->shadow_used_idx = 0;
vhost_log_used_vring(dev, vq, offsetof(struct vring_used, idx),
sizeof(vq->used->idx));
 }
@@ -561,7 +562,6 @@ virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
 
rte_prefetch0(&vq->avail->ring[vq->last_avail_idx & (vq->size - 1)]);
 
-   vq->shadow_used_idx = 0;
avail_head = *((volatile uint16_t *)&vq->avail->idx);
for (pkt_idx = 0; pkt_idx < count; pkt_idx++) {
uint32_t pkt_len = pkts[pkt_idx]->pkt_len + dev->vhost_hlen;
@@ -1048,7 +1048,6 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
goto out_access_unlock;
 
vq->batch_copy_nb_elems = 0;
-   vq->shadow_used_idx = 0;
 
if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
vhost_user_iotlb_rd_lock(vq);
@@ -1079,7 +1078,6 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
 
flush_shadow_used_ring(dev, vq);
vhost_vring_call(dev, vq);
-   vq->shadow_used_idx = 0;
}
 
rte_prefetch0(&vq->avail->ring[vq->last_avail_idx & (vq->size - 1)]);
-- 
2.14.4

[dpdk-dev] [PATCH v6 05/15] vhost: make indirect desc table copy desc type agnostic

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/virtio_net.c | 16 
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 28ce2d073..5cfc100a9 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -37,16 +37,15 @@ is_valid_virt_queue_idx(uint32_t idx, int is_tx, uint32_t 
nr_vring)
return (is_tx ^ (idx & 1)) == 0 && idx < nr_vring;
 }
 
-static __rte_always_inline struct vring_desc *
+static __rte_always_inline void *
 alloc_copy_ind_table(struct virtio_net *dev, struct vhost_virtqueue *vq,
-struct vring_desc *desc)
+uint64_t desc_addr, uint64_t desc_len)
 {
-   struct vring_desc *idesc;
+   void *idesc;
uint64_t src, dst;
-   uint64_t len, remain = desc->len;
-   uint64_t desc_addr = desc->addr;
+   uint64_t len, remain = desc_len;
 
-   idesc = rte_malloc(__func__, desc->len, 0);
+   idesc = rte_malloc(__func__, desc_len, 0);
if (unlikely(!idesc))
return 0;
 
@@ -72,7 +71,7 @@ alloc_copy_ind_table(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
 }
 
 static __rte_always_inline void
-free_ind_table(struct vring_desc *idesc)
+free_ind_table(void *idesc)
 {
rte_free(idesc);
 }
@@ -251,7 +250,8 @@ fill_vec_buf(struct virtio_net *dev, struct vhost_virtqueue 
*vq,
 * The indirect desc table is not contiguous
 * in process VA space, we have to copy it.
 */
-   idesc = alloc_copy_ind_table(dev, vq, &vq->desc[idx]);
+   idesc = alloc_copy_ind_table(dev, vq,
+   vq->desc[idx].addr, vq->desc[idx].len);
if (unlikely(!idesc))
return -1;
 
-- 
2.14.4

[dpdk-dev] [PATCH v6 03/15] vhost: vring address setup for packed queues

[Maxime Coquelin]

From: Yuanhan Liu 

Add code to set up packed queues when enabled.

Signed-off-by: Yuanhan Liu 
Signed-off-by: Jens Freimann 
Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/vhost.c  | 44 ++-
 lib/librte_vhost/vhost.h  |  7 ++-
 lib/librte_vhost/vhost_user.c | 40 +++
 3 files changed, 81 insertions(+), 10 deletions(-)

diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c
index afded4952..05d2bb182 100644
--- a/lib/librte_vhost/vhost.c
+++ b/lib/librte_vhost/vhost.c
@@ -23,6 +23,7 @@
 #include "iotlb.h"
 #include "vhost.h"
 #include "vhost_user.h"
+#include "virtio-packed.h"
 
 struct virtio_net *vhost_devices[MAX_VHOST_DEVICE];
 
@@ -115,14 +116,11 @@ free_device(struct virtio_net *dev)
rte_free(dev);
 }
 
-int
-vring_translate(struct virtio_net *dev, struct vhost_virtqueue *vq)
+static int
+vring_translate_split(struct virtio_net *dev, struct vhost_virtqueue *vq)
 {
uint64_t req_size, size;
 
-   if (!(dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM)))
-   goto out;
-
req_size = sizeof(struct vring_desc) * vq->size;
size = req_size;
vq->desc = (struct vring_desc *)(uintptr_t)vhost_iova_to_vva(dev, vq,
@@ -153,6 +151,40 @@ vring_translate(struct virtio_net *dev, struct 
vhost_virtqueue *vq)
if (!vq->used || size != req_size)
return -1;
 
+   return 0;
+}
+
+static int
+vring_translate_packed(struct virtio_net *dev, struct vhost_virtqueue *vq)
+{
+   uint64_t req_size, size;
+
+   req_size = sizeof(struct vring_desc_packed) * vq->size;
+   size = req_size;
+   vq->desc_packed =
+   (struct vring_desc_packed *)(uintptr_t)vhost_iova_to_vva(dev,
+   vq, vq->ring_addrs.desc_user_addr,
+   &size, VHOST_ACCESS_RW);
+   if (!vq->desc_packed || size != req_size)
+   return -1;
+
+   return 0;
+}
+
+int
+vring_translate(struct virtio_net *dev, struct vhost_virtqueue *vq)
+{
+
+   if (!(dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM)))
+   goto out;
+
+   if (vq_is_packed(dev)) {
+   if (vring_translate_packed(dev, vq) < 0)
+   return -1;
+   } else {
+   if (vring_translate_split(dev, vq) < 0)
+   return -1;
+   }
 out:
vq->access_ok = 1;
 
@@ -234,6 +266,8 @@ alloc_vring_queue(struct virtio_net *dev, uint32_t 
vring_idx)
dev->virtqueue[vring_idx] = vq;
init_vring_queue(dev, vring_idx);
rte_spinlock_init(&vq->access_lock);
+   vq->avail_wrap_counter = 1;
+   vq->used_wrap_counter = 1;
 
dev->nr_vring += 1;
 
diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index cabbc3ab7..35beff699 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -84,7 +84,10 @@ struct log_cache_entry {
  * Structure contains variables relevant to RX/TX virtqueues.
  */
 struct vhost_virtqueue {
-   struct vring_desc   *desc;
+   union {
+   struct vring_desc   *desc;
+   struct vring_desc_packed   *desc_packed;
+   };
struct vring_avail  *avail;
struct vring_used   *used;
uint32_tsize;
@@ -122,6 +125,8 @@ struct vhost_virtqueue {
 
struct batch_copy_elem  *batch_copy_elems;
uint16_tbatch_copy_nb_elems;
+   boolused_wrap_counter;
+   boolavail_wrap_counter;
 
struct log_cache_entry log_cache[VHOST_LOG_CACHE_NR];
uint16_t log_cache_nb_elem;
diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index 947290fc3..ae03ec29b 100644
--- a/lib/librte_vhost/vhost_user.c
+++ b/lib/librte_vhost/vhost_user.c
@@ -39,6 +39,7 @@
 #include "iotlb.h"
 #include "vhost.h"
 #include "vhost_user.h"
+#include "virtio-packed.h"
 
 #define VIRTIO_MIN_MTU 68
 #define VIRTIO_MAX_MTU 65535
@@ -477,6 +478,27 @@ translate_ring_addresses(struct virtio_net *dev, int 
vq_index)
struct vhost_vring_addr *addr = &vq->ring_addrs;
uint64_t len;
 
+   if (vq_is_packed(dev)) {
+   len = sizeof(struct vring_desc_packed) * vq->size;
+   vq->desc_packed = (struct vring_desc_packed *) ring_addr_to_vva
+   (dev, vq, addr->desc_user_addr, &len);
+   vq->log_guest_addr = 0;
+   if (vq->desc_packed == NULL ||
+   len != sizeof(struct vring_desc_packed) *
+   vq->size) {
+   RTE_LOG(DEBUG, VHOST_CONFIG,
+   "(%d) failed to map desc_packed ring.\n",
+   dev->vid);
+   return dev;
+   }
+
+   dev = numa_realloc(dev, vq_index);
+   vq = dev

[dpdk-dev] [PATCH v6 06/15] vhost: clear batch copy index at copy time

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/virtio_net.c | 8 
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 5cfc100a9..816d5fc1d 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -141,6 +141,8 @@ do_data_copy_enqueue(struct virtio_net *dev, struct 
vhost_virtqueue *vq)
vhost_log_cache_write(dev, vq, elem[i].log_addr, elem[i].len);
PRINT_PACKET(dev, (uintptr_t)elem[i].dst, elem[i].len, 0);
}
+
+   vq->batch_copy_nb_elems = 0;
 }
 
 static inline void
@@ -152,6 +154,8 @@ do_data_copy_dequeue(struct vhost_virtqueue *vq)
 
for (i = 0; i < count; i++)
rte_memcpy(elem[i].dst, elem[i].src, elem[i].len);
+
+   vq->batch_copy_nb_elems = 0;
 }
 
 /* avoid write operation when necessary, to lessen cache issues */
@@ -558,8 +562,6 @@ virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
if (count == 0)
goto out;
 
-   vq->batch_copy_nb_elems = 0;
-
rte_prefetch0(&vq->avail->ring[vq->last_avail_idx & (vq->size - 1)]);
 
avail_head = *((volatile uint16_t *)&vq->avail->idx);
@@ -1047,8 +1049,6 @@ rte_vhost_dequeue_burst(int vid, uint16_t queue_id,
if (unlikely(vq->enabled == 0))
goto out_access_unlock;
 
-   vq->batch_copy_nb_elems = 0;
-
if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
vhost_user_iotlb_rd_lock(vq);
 
-- 
2.14.4

[dpdk-dev] [PATCH v6 02/15] vhost: add helpers for packed virtqueues

[Maxime Coquelin]

From: Jens Freimann 

Add some helper functions to check descriptor flags
and check if a vring is of type packed.

Signed-off-by: Jens Freimann 
Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/vhost.h | 6 ++
 lib/librte_vhost/virtio-packed.h | 8 
 2 files changed, 14 insertions(+)

diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index cecd5c1ee..cabbc3ab7 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -318,6 +318,12 @@ struct virtio_net {
struct vhost_user_extern_ops extern_ops;
 } __rte_cache_aligned;
 
+static __rte_always_inline bool
+vq_is_packed(struct virtio_net *dev)
+{
+   return dev->features & (1ull << VIRTIO_F_RING_PACKED);
+}
+
 #define VHOST_LOG_PAGE 4096
 
 /*
diff --git a/lib/librte_vhost/virtio-packed.h b/lib/librte_vhost/virtio-packed.h
index 744b3991b..d386cb6df 100644
--- a/lib/librte_vhost/virtio-packed.h
+++ b/lib/librte_vhost/virtio-packed.h
@@ -19,4 +19,12 @@ struct vring_desc_packed {
uint16_t flags;
 };
 
+
+static inline bool
+desc_is_avail(struct vring_desc_packed *desc, bool wrap_counter)
+{
+   return wrap_counter == !!(desc->flags & VRING_DESC_F_AVAIL) &&
+   wrap_counter != !!(desc->flags & VRING_DESC_F_USED);
+}
+
 #endif /* __VIRTIO_PACKED_H */
-- 
2.14.4

[dpdk-dev] [PATCH v6 09/15] vhost: add shadow used ring support for packed rings

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/vhost.c  |  9 --
 lib/librte_vhost/vhost.h  | 13 ++--
 lib/librte_vhost/vhost_user.c | 64 --
 lib/librte_vhost/virtio_net.c | 72 +--
 4 files changed, 134 insertions(+), 24 deletions(-)

diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c
index 05d2bb182..8538302c9 100644
--- a/lib/librte_vhost/vhost.c
+++ b/lib/librte_vhost/vhost.c
@@ -94,9 +94,12 @@ cleanup_device(struct virtio_net *dev, int destroy)
 }
 
 void
-free_vq(struct vhost_virtqueue *vq)
+free_vq(struct virtio_net *dev, struct vhost_virtqueue *vq)
 {
-   rte_free(vq->shadow_used_ring);
+   if (vq_is_packed(dev))
+   rte_free(vq->shadow_used_packed);
+   else
+   rte_free(vq->shadow_used_split);
rte_free(vq->batch_copy_elems);
rte_mempool_free(vq->iotlb_pool);
rte_free(vq);
@@ -111,7 +114,7 @@ free_device(struct virtio_net *dev)
uint32_t i;
 
for (i = 0; i < dev->nr_vring; i++)
-   free_vq(dev->virtqueue[i]);
+   free_vq(dev, dev->virtqueue[i]);
 
rte_free(dev);
 }
diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index 35beff699..a7320469a 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -80,6 +80,12 @@ struct log_cache_entry {
unsigned long val;
 };
 
+struct vring_used_elem_packed {
+   uint16_t id;
+   uint32_t len;
+   uint32_t count;
+};
+
 /**
  * Structure contains variables relevant to RX/TX virtqueues.
  */
@@ -119,7 +125,10 @@ struct vhost_virtqueue {
struct zcopy_mbuf   *zmbufs;
struct zcopy_mbuf_list  zmbuf_list;
 
-   struct vring_used_elem  *shadow_used_ring;
+   union {
+   struct vring_used_elem  *shadow_used_split;
+   struct vring_used_elem_packed *shadow_used_packed;
+   };
uint16_tshadow_used_idx;
struct vhost_vring_addr ring_addrs;
 
@@ -552,7 +561,7 @@ void reset_device(struct virtio_net *dev);
 void vhost_destroy_device(int);
 
 void cleanup_vq(struct vhost_virtqueue *vq, int destroy);
-void free_vq(struct vhost_virtqueue *vq);
+void free_vq(struct virtio_net *dev, struct vhost_virtqueue *vq);
 
 int alloc_vring_queue(struct virtio_net *dev, uint32_t vring_idx);
 
diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index ae03ec29b..b2b57de57 100644
--- a/lib/librte_vhost/vhost_user.c
+++ b/lib/librte_vhost/vhost_user.c
@@ -244,7 +244,7 @@ vhost_user_set_features(struct virtio_net *dev, uint64_t 
features)
 
dev->virtqueue[dev->nr_vring] = NULL;
cleanup_vq(vq, 1);
-   free_vq(vq);
+   free_vq(dev, vq);
}
}
 
@@ -293,13 +293,26 @@ vhost_user_set_vring_num(struct virtio_net *dev,
TAILQ_INIT(&vq->zmbuf_list);
}
 
-   vq->shadow_used_ring = rte_malloc(NULL,
+   if (vq_is_packed(dev)) {
+   vq->shadow_used_packed = rte_malloc(NULL,
+   vq->size *
+   sizeof(struct vring_used_elem_packed),
+   RTE_CACHE_LINE_SIZE);
+   if (!vq->shadow_used_packed) {
+   RTE_LOG(ERR, VHOST_CONFIG,
+   "failed to allocate memory for shadow 
used ring.\n");
+   return -1;
+   }
+
+   } else {
+   vq->shadow_used_split = rte_malloc(NULL,
vq->size * sizeof(struct vring_used_elem),
RTE_CACHE_LINE_SIZE);
-   if (!vq->shadow_used_ring) {
-   RTE_LOG(ERR, VHOST_CONFIG,
-   "failed to allocate memory for shadow used ring.\n");
-   return -1;
+   if (!vq->shadow_used_split) {
+   RTE_LOG(ERR, VHOST_CONFIG,
+   "failed to allocate memory for shadow 
used ring.\n");
+   return -1;
+   }
}
 
vq->batch_copy_elems = rte_malloc(NULL,
@@ -326,7 +339,8 @@ numa_realloc(struct virtio_net *dev, int index)
struct virtio_net *old_dev;
struct vhost_virtqueue *old_vq, *vq;
struct zcopy_mbuf *new_zmbuf;
-   struct vring_used_elem *new_shadow_used_ring;
+   struct vring_used_elem *new_shadow_used_split;
+   struct vring_used_elem_packed *new_shadow_used_packed;
struct batch_copy_elem *new_batch_copy_elems;
int ret;
 
@@ -361,13 +375,26 @@ numa_realloc(struct virtio_net *dev, int index)
vq->zmbufs = new_zmbuf;
}
 
-   new_shadow_used_ring = rte_malloc_socket(NULL,
-   vq->size * sizeof(struct vring_used_elem),
-   RTE_CACHE_LINE_SIZ

[dpdk-dev] [PATCH v6 11/15] vhost: add vector filling support for packed ring

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/virtio_net.c | 111 ++
 1 file changed, 111 insertions(+)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 64664b7de..2d867e88e 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -448,6 +448,117 @@ reserve_avail_buf_split(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
return 0;
 }
 
+static __rte_always_inline int
+fill_vec_buf_packed_indirect(struct virtio_net *dev,
+   struct vhost_virtqueue *vq,
+   struct vring_desc_packed *desc, uint16_t *vec_idx,
+   struct buf_vector *buf_vec, uint16_t *len, uint8_t perm)
+{
+   uint16_t i;
+   uint32_t nr_decs;
+   uint16_t vec_id = *vec_idx;
+   uint64_t dlen;
+   struct vring_desc_packed *descs, *idescs = NULL;
+
+   dlen = desc->len;
+   descs = (struct vring_desc_packed *)(uintptr_t)
+   vhost_iova_to_vva(dev, vq, desc->addr, &dlen, VHOST_ACCESS_RO);
+   if (unlikely(!descs))
+   return -1;
+
+   if (unlikely(dlen < desc->len)) {
+   /*
+* The indirect desc table is not contiguous
+* in process VA space, we have to copy it.
+*/
+   idescs = alloc_copy_ind_table(dev, vq, desc->addr, desc->len);
+   if (unlikely(!idescs))
+   return -1;
+
+   descs = idescs;
+   }
+
+   nr_decs =  desc->len / sizeof(struct vring_desc_packed);
+   if (unlikely(nr_decs >= vq->size)) {
+   free_ind_table(idescs);
+   return -1;
+   }
+
+   for (i = 0; i < nr_decs; i++) {
+   if (unlikely(vec_id >= BUF_VECTOR_MAX)) {
+   free_ind_table(idescs);
+   return -1;
+   }
+
+   *len += descs[i].len;
+   if (unlikely(map_one_desc(dev, vq, buf_vec, &vec_id,
+   descs[i].addr, descs[i].len,
+   perm)))
+   return -1;
+   }
+   *vec_idx = vec_id;
+
+   if (unlikely(!!idescs))
+   free_ind_table(idescs);
+
+   return 0;
+}
+
+static inline int
+fill_vec_buf_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
+   uint16_t avail_idx, uint16_t *desc_count,
+   struct buf_vector *buf_vec, uint16_t *vec_idx,
+   uint16_t *buf_id, uint16_t *len, uint8_t perm)
+{
+   bool wrap_counter = vq->avail_wrap_counter;
+   struct vring_desc_packed *descs = vq->desc_packed;
+   uint16_t vec_id = *vec_idx;
+
+   if (avail_idx < vq->last_avail_idx)
+   wrap_counter ^= 1;
+
+   if (unlikely(!desc_is_avail(&descs[avail_idx], wrap_counter)))
+   return -1;
+
+   *desc_count = 0;
+
+   while (1) {
+   if (unlikely(vec_id >= BUF_VECTOR_MAX))
+   return -1;
+
+   *desc_count += 1;
+   *buf_id = descs[avail_idx].index;
+
+   if (descs[avail_idx].flags & VRING_DESC_F_INDIRECT) {
+   if (unlikely(fill_vec_buf_packed_indirect(dev, vq,
+   &descs[avail_idx],
+   &vec_id, buf_vec,
+   len, perm) < 0))
+   return -1;
+   } else {
+   *len += descs[avail_idx].len;
+
+   if (unlikely(map_one_desc(dev, vq, buf_vec, &vec_id,
+   descs[avail_idx].addr,
+   descs[avail_idx].len,
+   perm)))
+   return -1;
+   }
+
+   if ((descs[avail_idx].flags & VRING_DESC_F_NEXT) == 0)
+   break;
+
+   if (++avail_idx >= vq->size) {
+   avail_idx -= vq->size;
+   wrap_counter ^= 1;
+   }
+   }
+
+   *vec_idx = vec_id;
+
+   return 0;
+}
+
 static __rte_always_inline int
 copy_mbuf_to_desc(struct virtio_net *dev, struct vhost_virtqueue *vq,
struct rte_mbuf *m, struct buf_vector *buf_vec,
-- 
2.14.4

[dpdk-dev] [PATCH v6 10/15] vhost: create descriptor mapping function

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/virtio_net.c | 61 ++-
 1 file changed, 37 insertions(+), 24 deletions(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index f0e2e6a1f..64664b7de 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -292,6 +292,37 @@ virtio_enqueue_offload(struct rte_mbuf *m_buf, struct 
virtio_net_hdr *net_hdr)
}
 }
 
+static __rte_always_inline int
+map_one_desc(struct virtio_net *dev, struct vhost_virtqueue *vq,
+   struct buf_vector *buf_vec, uint16_t *vec_idx,
+   uint64_t desc_iova, uint64_t desc_len, uint8_t perm)
+{
+   uint16_t vec_id = *vec_idx;
+
+   while (desc_len) {
+   uint64_t desc_addr;
+   uint64_t desc_chunck_len = desc_len;
+
+   desc_addr = vhost_iova_to_vva(dev, vq,
+   desc_iova,
+   &desc_chunck_len,
+   perm);
+   if (unlikely(!desc_addr))
+   return -1;
+
+   buf_vec[vec_id].buf_iova = desc_iova;
+   buf_vec[vec_id].buf_addr = desc_addr;
+   buf_vec[vec_id].buf_len  = desc_chunck_len;
+
+   desc_len -= desc_chunck_len;
+   desc_iova += desc_chunck_len;
+   vec_id++;
+   }
+   *vec_idx = vec_id;
+
+   return 0;
+}
+
 static __rte_always_inline int
 fill_vec_buf_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
 uint32_t avail_idx, uint16_t *vec_idx,
@@ -301,7 +332,7 @@ fill_vec_buf_split(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
uint16_t idx = vq->avail->ring[avail_idx & (vq->size - 1)];
uint16_t vec_id = *vec_idx;
uint32_t len= 0;
-   uint64_t dlen, desc_avail, desc_iova;
+   uint64_t dlen;
struct vring_desc *descs = vq->desc;
struct vring_desc *idesc = NULL;
 
@@ -339,30 +370,12 @@ fill_vec_buf_split(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
}
 
len += descs[idx].len;
-   desc_avail = descs[idx].len;
-   desc_iova = descs[idx].addr;
-
-   while (desc_avail) {
-   uint64_t desc_addr;
-   uint64_t desc_chunck_len = desc_avail;
-
-   desc_addr = vhost_iova_to_vva(dev, vq,
-   desc_iova,
-   &desc_chunck_len,
-   perm);
-   if (unlikely(!desc_addr)) {
-   free_ind_table(idesc);
-   return -1;
-   }
-
-   buf_vec[vec_id].buf_iova = desc_iova;
-   buf_vec[vec_id].buf_addr = desc_addr;
-   buf_vec[vec_id].buf_len  = desc_chunck_len;
-   buf_vec[vec_id].desc_idx = idx;
 
-   desc_avail -= desc_chunck_len;
-   desc_iova += desc_chunck_len;
-   vec_id++;
+   if (unlikely(map_one_desc(dev, vq, buf_vec, &vec_id,
+   descs[idx].addr, descs[idx].len,
+   perm))) {
+   free_ind_table(idesc);
+   return -1;
}
 
if ((descs[idx].flags & VRING_DESC_F_NEXT) == 0)
-- 
2.14.4

[dpdk-dev] [PATCH v6 08/15] vhost: append shadow used ring function names with split

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/virtio_net.c | 28 +++-
 1 file changed, 15 insertions(+), 13 deletions(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 385876527..35f8cf90a 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -77,8 +77,9 @@ free_ind_table(void *idesc)
 }
 
 static __rte_always_inline void
-do_flush_shadow_used_ring(struct virtio_net *dev, struct vhost_virtqueue *vq,
- uint16_t to, uint16_t from, uint16_t size)
+do_flush_shadow_used_ring_split(struct virtio_net *dev,
+   struct vhost_virtqueue *vq,
+   uint16_t to, uint16_t from, uint16_t size)
 {
rte_memcpy(&vq->used->ring[to],
&vq->shadow_used_ring[from],
@@ -89,22 +90,22 @@ do_flush_shadow_used_ring(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
 }
 
 static __rte_always_inline void
-flush_shadow_used_ring(struct virtio_net *dev, struct vhost_virtqueue *vq)
+flush_shadow_used_ring_split(struct virtio_net *dev, struct vhost_virtqueue 
*vq)
 {
uint16_t used_idx = vq->last_used_idx & (vq->size - 1);
 
if (used_idx + vq->shadow_used_idx <= vq->size) {
-   do_flush_shadow_used_ring(dev, vq, used_idx, 0,
+   do_flush_shadow_used_ring_split(dev, vq, used_idx, 0,
  vq->shadow_used_idx);
} else {
uint16_t size;
 
/* update used ring interval [used_idx, vq->size] */
size = vq->size - used_idx;
-   do_flush_shadow_used_ring(dev, vq, used_idx, 0, size);
+   do_flush_shadow_used_ring_split(dev, vq, used_idx, 0, size);
 
/* update the left half used ring interval [0, left_size] */
-   do_flush_shadow_used_ring(dev, vq, 0, size,
+   do_flush_shadow_used_ring_split(dev, vq, 0, size,
  vq->shadow_used_idx - size);
}
vq->last_used_idx += vq->shadow_used_idx;
@@ -120,7 +121,7 @@ flush_shadow_used_ring(struct virtio_net *dev, struct 
vhost_virtqueue *vq)
 }
 
 static __rte_always_inline void
-update_shadow_used_ring(struct vhost_virtqueue *vq,
+update_shadow_used_ring_split(struct vhost_virtqueue *vq,
 uint16_t desc_idx, uint16_t len)
 {
uint16_t i = vq->shadow_used_idx++;
@@ -347,7 +348,7 @@ reserve_avail_buf_split(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
VHOST_ACCESS_RO) < 0))
return -1;
len = RTE_MIN(len, size);
-   update_shadow_used_ring(vq, head_idx, len);
+   update_shadow_used_ring_split(vq, head_idx, len);
size -= len;
 
cur_idx++;
@@ -573,7 +574,7 @@ virtio_dev_rx_split(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
do_data_copy_enqueue(dev, vq);
 
if (likely(vq->shadow_used_idx)) {
-   flush_shadow_used_ring(dev, vq);
+   flush_shadow_used_ring_split(dev, vq);
vhost_vring_call(dev, vq);
}
 
@@ -1041,7 +1042,8 @@ virtio_dev_tx_split(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
next = TAILQ_NEXT(zmbuf, next);
 
if (mbuf_is_consumed(zmbuf->mbuf)) {
-   update_shadow_used_ring(vq, zmbuf->desc_idx, 0);
+   update_shadow_used_ring_split(vq,
+   zmbuf->desc_idx, 0);
nr_updated += 1;
 
TAILQ_REMOVE(&vq->zmbuf_list, zmbuf, next);
@@ -1052,7 +1054,7 @@ virtio_dev_tx_split(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
}
}
 
-   flush_shadow_used_ring(dev, vq);
+   flush_shadow_used_ring_split(dev, vq);
vhost_vring_call(dev, vq);
}
 
@@ -1084,7 +1086,7 @@ virtio_dev_tx_split(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
break;
 
if (likely(dev->dequeue_zero_copy == 0))
-   update_shadow_used_ring(vq, head_idx, 0);
+   update_shadow_used_ring_split(vq, head_idx, 0);
 
rte_prefetch0((void *)(uintptr_t)buf_vec[0].buf_addr);
 
@@ -1131,7 +1133,7 @@ virtio_dev_tx_split(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
do_data_copy_dequeue(vq);
if (unlikely(i < count))
vq->shadow_used_idx = i;
-   flush_shadow_used_ring(dev, vq);
+   flush_shadow_used_ring_split(dev, vq);
vhost_vring_call(dev, vq);
}
 
-- 
2.14.4

[dpdk-dev] [PATCH v6 07/15] vhost: extract split ring handling from Rx and Tx functions

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/virtio_net.c | 240 +++---
 1 file changed, 131 insertions(+), 109 deletions(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 816d5fc1d..385876527 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -226,13 +226,13 @@ virtio_enqueue_offload(struct rte_mbuf *m_buf, struct 
virtio_net_hdr *net_hdr)
 }
 
 static __rte_always_inline int
-fill_vec_buf(struct virtio_net *dev, struct vhost_virtqueue *vq,
-uint32_t avail_idx, uint32_t *vec_idx,
+fill_vec_buf_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
+uint32_t avail_idx, uint16_t *vec_idx,
 struct buf_vector *buf_vec, uint16_t *desc_chain_head,
 uint16_t *desc_chain_len, uint8_t perm)
 {
uint16_t idx = vq->avail->ring[avail_idx & (vq->size - 1)];
-   uint32_t vec_id = *vec_idx;
+   uint16_t vec_id = *vec_idx;
uint32_t len= 0;
uint64_t dlen, desc_avail, desc_iova;
struct vring_desc *descs = vq->desc;
@@ -317,13 +317,13 @@ fill_vec_buf(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
  * Returns -1 on fail, 0 on success
  */
 static inline int
-reserve_avail_buf(struct virtio_net *dev, struct vhost_virtqueue *vq,
+reserve_avail_buf_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
uint32_t size, struct buf_vector *buf_vec,
uint16_t *num_buffers, uint16_t avail_head,
uint16_t *nr_vec)
 {
uint16_t cur_idx;
-   uint32_t vec_idx = 0;
+   uint16_t vec_idx = 0;
uint16_t max_tries, tries = 0;
 
uint16_t head_idx = 0;
@@ -341,7 +341,8 @@ reserve_avail_buf(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
if (unlikely(cur_idx == avail_head))
return -1;
 
-   if (unlikely(fill_vec_buf(dev, vq, cur_idx, &vec_idx, buf_vec,
+   if (unlikely(fill_vec_buf_split(dev, vq, cur_idx,
+   &vec_idx, buf_vec,
&head_idx, &len,
VHOST_ACCESS_RO) < 0))
return -1;
@@ -528,48 +529,22 @@ copy_mbuf_to_desc(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
 }
 
 static __rte_always_inline uint32_t
-virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
+virtio_dev_rx_split(struct virtio_net *dev, struct vhost_virtqueue *vq,
struct rte_mbuf **pkts, uint32_t count)
 {
-   struct vhost_virtqueue *vq;
uint32_t pkt_idx = 0;
uint16_t num_buffers;
struct buf_vector buf_vec[BUF_VECTOR_MAX];
uint16_t avail_head;
 
-   VHOST_LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
-   if (unlikely(!is_valid_virt_queue_idx(queue_id, 0, dev->nr_vring))) {
-   RTE_LOG(ERR, VHOST_DATA, "(%d) %s: invalid virtqueue idx %d.\n",
-   dev->vid, __func__, queue_id);
-   return 0;
-   }
-
-   vq = dev->virtqueue[queue_id];
-
-   rte_spinlock_lock(&vq->access_lock);
-
-   if (unlikely(vq->enabled == 0))
-   goto out_access_unlock;
-
-   if (dev->features & (1ULL << VIRTIO_F_IOMMU_PLATFORM))
-   vhost_user_iotlb_rd_lock(vq);
-
-   if (unlikely(vq->access_ok == 0))
-   if (unlikely(vring_translate(dev, vq) < 0))
-   goto out;
-
-   count = RTE_MIN((uint32_t)MAX_PKT_BURST, count);
-   if (count == 0)
-   goto out;
-
rte_prefetch0(&vq->avail->ring[vq->last_avail_idx & (vq->size - 1)]);
-
avail_head = *((volatile uint16_t *)&vq->avail->idx);
+
for (pkt_idx = 0; pkt_idx < count; pkt_idx++) {
uint32_t pkt_len = pkts[pkt_idx]->pkt_len + dev->vhost_hlen;
uint16_t nr_vec = 0;
 
-   if (unlikely(reserve_avail_buf(dev, vq,
+   if (unlikely(reserve_avail_buf_split(dev, vq,
pkt_len, buf_vec, &num_buffers,
avail_head, &nr_vec) < 0)) {
VHOST_LOG_DEBUG(VHOST_DATA,
@@ -602,6 +577,42 @@ virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
vhost_vring_call(dev, vq);
}
 
+   return pkt_idx;
+}
+
+static __rte_always_inline uint32_t
+virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
+   struct rte_mbuf **pkts, uint32_t count)
+{
+   struct vhost_virtqueue *vq;
+
+   VHOST_LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
+   if (unlikely(!is_valid_virt_queue_idx(queue_id, 0, dev->nr_vring))) {
+   RTE_LOG(ERR, VHOST_DATA, "(%d) %s: invalid virtqueue idx %d.\n",
+   dev->vid, __func__, queue

[dpdk-dev] [PATCH v6 12/15] vhost: add Rx support for packed ring

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/virtio_net.c | 117 +-
 1 file changed, 116 insertions(+), 1 deletion(-)

diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 2d867e88e..2e286e228 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -559,6 +559,65 @@ fill_vec_buf_packed(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
return 0;
 }
 
+/*
+ * Returns -1 on fail, 0 on success
+ */
+static inline int
+reserve_avail_buf_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
+   uint32_t size, struct buf_vector *buf_vec,
+   uint16_t *nr_vec, uint16_t *num_buffers,
+   uint16_t *nr_descs)
+{
+   uint16_t avail_idx;
+   uint16_t vec_idx = 0;
+   uint16_t max_tries, tries = 0;
+
+   uint16_t buf_id = 0;
+   uint16_t len = 0;
+   uint16_t desc_count;
+
+   *num_buffers = 0;
+   avail_idx = vq->last_avail_idx;
+
+   if (rxvq_is_mergeable(dev))
+   max_tries = vq->size;
+   else
+   max_tries = 1;
+
+   while (size > 0) {
+   if (unlikely(fill_vec_buf_packed(dev, vq,
+   avail_idx, &desc_count,
+   buf_vec, &vec_idx,
+   &buf_id, &len,
+   VHOST_ACCESS_RO) < 0))
+   return -1;
+
+   len = RTE_MIN(len, size);
+   update_shadow_used_ring_packed(vq, buf_id, len, desc_count);
+   size -= len;
+
+   avail_idx += desc_count;
+   if (avail_idx >= vq->size)
+   avail_idx -= vq->size;
+
+   *nr_descs += desc_count;
+   tries++;
+   *num_buffers += 1;
+
+   /*
+* if we tried all available ring items, and still
+* can't get enough buf, it means something abnormal
+* happened.
+*/
+   if (unlikely(tries > max_tries))
+   return -1;
+   }
+
+   *nr_vec = vec_idx;
+
+   return 0;
+}
+
 static __rte_always_inline int
 copy_mbuf_to_desc(struct virtio_net *dev, struct vhost_virtqueue *vq,
struct rte_mbuf *m, struct buf_vector *buf_vec,
@@ -771,6 +830,59 @@ virtio_dev_rx_split(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
return pkt_idx;
 }
 
+static __rte_always_inline uint32_t
+virtio_dev_rx_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
+   struct rte_mbuf **pkts, uint32_t count)
+{
+   uint32_t pkt_idx = 0;
+   uint16_t num_buffers;
+   struct buf_vector buf_vec[BUF_VECTOR_MAX];
+
+   for (pkt_idx = 0; pkt_idx < count; pkt_idx++) {
+   uint32_t pkt_len = pkts[pkt_idx]->pkt_len + dev->vhost_hlen;
+   uint16_t nr_vec = 0;
+   uint16_t nr_descs = 0;
+
+   if (unlikely(reserve_avail_buf_packed(dev, vq,
+   pkt_len, buf_vec, &nr_vec,
+   &num_buffers, &nr_descs) < 0)) {
+   VHOST_LOG_DEBUG(VHOST_DATA,
+   "(%d) failed to get enough desc from vring\n",
+   dev->vid);
+   vq->shadow_used_idx -= num_buffers;
+   break;
+   }
+
+   rte_prefetch0((void *)(uintptr_t)buf_vec[0].buf_addr);
+
+   VHOST_LOG_DEBUG(VHOST_DATA, "(%d) current index %d | end index 
%d\n",
+   dev->vid, vq->last_avail_idx,
+   vq->last_avail_idx + num_buffers);
+
+   if (copy_mbuf_to_desc(dev, vq, pkts[pkt_idx],
+   buf_vec, nr_vec,
+   num_buffers) < 0) {
+   vq->shadow_used_idx -= num_buffers;
+   break;
+   }
+
+   vq->last_avail_idx += nr_descs;
+   if (vq->last_avail_idx >= vq->size) {
+   vq->last_avail_idx -= vq->size;
+   vq->avail_wrap_counter ^= 1;
+   }
+   }
+
+   do_data_copy_enqueue(dev, vq);
+
+   if (likely(vq->shadow_used_idx)) {
+   flush_shadow_used_ring_packed(dev, vq);
+   vhost_vring_call(dev, vq);
+   }
+
+   return pkt_idx;
+}
+
 static __rte_always_inline uint32_t
 virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
struct rte_mbuf **pkts, uint32_t count)
@@ -802,7 +914,10 @@ virtio_dev_rx(struct virtio_net *dev, uint16_t queue_id,
if (count == 0)
goto out;
 
-   count = virtio_dev_rx_split(dev, vq, pkts, count);
+   if (

[dpdk-dev] [PATCH v6 13/15] vhost: add Tx support for packed ring

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/vhost.h  |   1 +
 lib/librte_vhost/virtio_net.c | 121 +-
 2 files changed, 121 insertions(+), 1 deletion(-)

diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index a7320469a..6ea8fb896 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -56,6 +56,7 @@ struct buf_vector {
 struct zcopy_mbuf {
struct rte_mbuf *mbuf;
uint32_t desc_idx;
+   uint16_t desc_count;
uint16_t in_use;
 
TAILQ_ENTRY(zcopy_mbuf) next;
diff --git a/lib/librte_vhost/virtio_net.c b/lib/librte_vhost/virtio_net.c
index 2e286e228..03dd38235 100644
--- a/lib/librte_vhost/virtio_net.c
+++ b/lib/librte_vhost/virtio_net.c
@@ -1445,6 +1445,122 @@ virtio_dev_tx_split(struct virtio_net *dev, struct 
vhost_virtqueue *vq,
return i;
 }
 
+static __rte_always_inline uint16_t
+virtio_dev_tx_packed(struct virtio_net *dev, struct vhost_virtqueue *vq,
+   struct rte_mempool *mbuf_pool, struct rte_mbuf **pkts, uint16_t count)
+{
+   uint16_t i;
+
+   rte_prefetch0(&vq->desc_packed[vq->last_avail_idx]);
+
+   if (unlikely(dev->dequeue_zero_copy)) {
+   struct zcopy_mbuf *zmbuf, *next;
+   int nr_updated = 0;
+
+   for (zmbuf = TAILQ_FIRST(&vq->zmbuf_list);
+zmbuf != NULL; zmbuf = next) {
+   next = TAILQ_NEXT(zmbuf, next);
+
+   if (mbuf_is_consumed(zmbuf->mbuf)) {
+   update_shadow_used_ring_packed(vq,
+   zmbuf->desc_idx,
+   0,
+   zmbuf->desc_count);
+   nr_updated += 1;
+
+   TAILQ_REMOVE(&vq->zmbuf_list, zmbuf, next);
+   restore_mbuf(zmbuf->mbuf);
+   rte_pktmbuf_free(zmbuf->mbuf);
+   put_zmbuf(zmbuf);
+   vq->nr_zmbuf -= 1;
+   }
+   }
+
+   flush_shadow_used_ring_packed(dev, vq);
+   vhost_vring_call(dev, vq);
+   }
+
+   VHOST_LOG_DEBUG(VHOST_DATA, "(%d) %s\n", dev->vid, __func__);
+
+   count = RTE_MIN(count, MAX_PKT_BURST);
+   VHOST_LOG_DEBUG(VHOST_DATA, "(%d) about to dequeue %u buffers\n",
+   dev->vid, count);
+
+   for (i = 0; i < count; i++) {
+   struct buf_vector buf_vec[BUF_VECTOR_MAX];
+   uint16_t buf_id, dummy_len;
+   uint16_t desc_count, nr_vec = 0;
+   int err;
+
+   if (unlikely(fill_vec_buf_packed(dev, vq,
+   vq->last_avail_idx, &desc_count,
+   buf_vec, &nr_vec,
+   &buf_id, &dummy_len,
+   VHOST_ACCESS_RW) < 0))
+   break;
+
+   if (likely(dev->dequeue_zero_copy == 0))
+   update_shadow_used_ring_packed(vq, buf_id, 0,
+   desc_count);
+
+   rte_prefetch0((void *)(uintptr_t)buf_vec[0].buf_addr);
+
+   pkts[i] = rte_pktmbuf_alloc(mbuf_pool);
+   if (unlikely(pkts[i] == NULL)) {
+   RTE_LOG(ERR, VHOST_DATA,
+   "Failed to allocate memory for mbuf.\n");
+   break;
+   }
+
+   err = copy_desc_to_mbuf(dev, vq, buf_vec, nr_vec, pkts[i],
+   mbuf_pool);
+   if (unlikely(err)) {
+   rte_pktmbuf_free(pkts[i]);
+   break;
+   }
+
+   if (unlikely(dev->dequeue_zero_copy)) {
+   struct zcopy_mbuf *zmbuf;
+
+   zmbuf = get_zmbuf(vq);
+   if (!zmbuf) {
+   rte_pktmbuf_free(pkts[i]);
+   break;
+   }
+   zmbuf->mbuf = pkts[i];
+   zmbuf->desc_idx = buf_id;
+   zmbuf->desc_count = desc_count;
+
+   /*
+* Pin lock the mbuf; we will check later to see
+* whether the mbuf is freed (when we are the last
+* user) or not. If that's the case, we then could
+* update the used ring safely.
+*/
+   rte_mbuf_refcnt_update(pkts[i], 1);
+
+   vq->nr_zmbuf += 1;
+   TAILQ_INSERT_TAIL(&vq->zmbuf_list, zmbuf, next);
+   }
+
+   vq->last_avail_idx += desc_count;
+   if (vq->last_avail_idx >

[dpdk-dev] [PATCH v6 14/15] vhost: add notification for packed ring

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/vhost.c | 73 
 lib/librte_vhost/vhost.h | 71 ++
 lib/librte_vhost/vhost_user.c| 24 +
 lib/librte_vhost/virtio-packed.h | 11 ++
 lib/librte_vhost/virtio_net.c| 12 +++
 5 files changed, 172 insertions(+), 19 deletions(-)

diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c
index 8538302c9..78f20c402 100644
--- a/lib/librte_vhost/vhost.c
+++ b/lib/librte_vhost/vhost.c
@@ -171,6 +171,24 @@ vring_translate_packed(struct virtio_net *dev, struct 
vhost_virtqueue *vq)
if (!vq->desc_packed || size != req_size)
return -1;
 
+   req_size = sizeof(struct vring_packed_desc_event);
+   size = req_size;
+   vq->driver_event = (struct vring_packed_desc_event *)(uintptr_t)
+   vhost_iova_to_vva(dev,
+   vq, vq->ring_addrs.avail_user_addr,
+   &size, VHOST_ACCESS_RW);
+   if (!vq->driver_event || size != req_size)
+   return -1;
+
+   req_size = sizeof(struct vring_packed_desc_event);
+   size = req_size;
+   vq->device_event = (struct vring_packed_desc_event *)(uintptr_t)
+   vhost_iova_to_vva(dev,
+   vq, vq->ring_addrs.used_user_addr,
+   &size, VHOST_ACCESS_RW);
+   if (!vq->device_event || size != req_size)
+   return -1;
+
return 0;
 }
 
@@ -595,7 +613,11 @@ rte_vhost_vring_call(int vid, uint16_t vring_idx)
if (!vq)
return -1;
 
-   vhost_vring_call(dev, vq);
+   if (vq_is_packed(dev))
+   vhost_vring_call_packed(dev, vq);
+   else
+   vhost_vring_call_split(dev, vq);
+
return 0;
 }
 
@@ -616,20 +638,59 @@ rte_vhost_avail_entries(int vid, uint16_t queue_id)
return *(volatile uint16_t *)&vq->avail->idx - vq->last_used_idx;
 }
 
+static inline int
+vhost_enable_notify_split(struct vhost_virtqueue *vq, int enable)
+{
+   if (enable)
+   vq->used->flags &= ~VRING_USED_F_NO_NOTIFY;
+   else
+   vq->used->flags |= VRING_USED_F_NO_NOTIFY;
+
+   return 0;
+}
+
+static inline int
+vhost_enable_notify_packed(struct virtio_net *dev,
+   struct vhost_virtqueue *vq, int enable)
+{
+   uint16_t flags;
+
+   if (!enable) {
+   vq->device_event->desc_event_flags = RING_EVENT_FLAGS_DISABLE;
+   return 0;
+   }
+
+   flags = RING_EVENT_FLAGS_ENABLE;
+   if (dev->features & (1ULL << VIRTIO_RING_F_EVENT_IDX)) {
+   flags = RING_EVENT_FLAGS_DESC;
+   vq->device_event->desc_event_off_wrap = vq->last_avail_idx |
+   vq->avail_wrap_counter << 15;
+   }
+
+   rte_smp_wmb();
+
+   vq->device_event->desc_event_flags = flags;
+
+   rte_smp_wmb();
+
+   return 0;
+}
+
 int
 rte_vhost_enable_guest_notification(int vid, uint16_t queue_id, int enable)
 {
struct virtio_net *dev = get_device(vid);
+   struct vhost_virtqueue *vq;
 
if (!dev)
return -1;
 
-   if (enable)
-   dev->virtqueue[queue_id]->used->flags &=
-   ~VRING_USED_F_NO_NOTIFY;
+   vq = dev->virtqueue[queue_id];
+
+   if (vq_is_packed(dev))
+   return vhost_enable_notify_packed(dev, vq, enable);
else
-   dev->virtqueue[queue_id]->used->flags |= VRING_USED_F_NO_NOTIFY;
-   return 0;
+   return vhost_enable_notify_split(vq, enable);
 }
 
 void
diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index 6ea8fb896..728fd2f6b 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -21,6 +21,7 @@
 
 #include "rte_vhost.h"
 #include "rte_vdpa.h"
+#include "virtio-packed.h"
 
 /* Used to indicate that the device is running on a data core */
 #define VIRTIO_DEV_RUNNING 1
@@ -95,8 +96,14 @@ struct vhost_virtqueue {
struct vring_desc   *desc;
struct vring_desc_packed   *desc_packed;
};
-   struct vring_avail  *avail;
-   struct vring_used   *used;
+   union {
+   struct vring_avail  *avail;
+   struct vring_packed_desc_event *driver_event;
+   };
+   union {
+   struct vring_used   *used;
+   struct vring_packed_desc_event *device_event;
+   };
uint32_tsize;
 
uint16_tlast_avail_idx;
@@ -613,7 +620,7 @@ vhost_need_event(uint16_t event_idx, uint16_t new_idx, 
uint16_t old)
 }
 
 static __rte_always_inline void
-vhost_vring_call(struct virtio_net *dev, struct vhost_virtqueue *vq)
+vhost_vring_call_split(struct virtio_net *dev, struct vhost_virtqueue *vq)
 {
/* Flush used->idx update before we read 

[dpdk-dev] [PATCH v6 15/15] vhost: advertize packed ring layout support

[Maxime Coquelin]

Signed-off-by: Maxime Coquelin 
---
 lib/librte_vhost/vhost.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index 728fd2f6b..ef2558795 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -241,7 +241,8 @@ struct vhost_msg {
(1ULL << VIRTIO_RING_F_INDIRECT_DESC) | \
(1ULL << VIRTIO_RING_F_EVENT_IDX) | \
(1ULL << VIRTIO_NET_F_MTU) | \
-   (1ULL << VIRTIO_F_IOMMU_PLATFORM))
+   (1ULL << VIRTIO_F_IOMMU_PLATFORM) | \
+   (1ULL << VIRTIO_F_RING_PACKED))
 
 
 struct guest_page {
-- 
2.14.4

[dpdk-dev] [PATCH 2/2] test/link_bonding_rssconf: disable HW CRC strip by default

[Phil Yang]

Since the tested null device has no capacity of HW CRC offload,
disable it for the link bonding rssconf test by default. Otherwise,
it will cause test unexpected failure.

Fixes: b219c8c("test: enable HW CRC strip by default")

Signed-off-by: Phil Yang 
---
 test/test/test_link_bonding_rssconf.c | 16 
 1 file changed, 4 insertions(+), 12 deletions(-)

diff --git a/test/test/test_link_bonding_rssconf.c 
b/test/test/test_link_bonding_rssconf.c
index 6a1a28d..b23e4e8 100644
--- a/test/test/test_link_bonding_rssconf.c
+++ b/test/test/test_link_bonding_rssconf.c
@@ -82,12 +82,8 @@ static struct rte_eth_conf default_pmd_conf = {
.rxmode = {
.mq_mode = ETH_MQ_RX_NONE,
.max_rx_pkt_len = ETHER_MAX_LEN,
-   .split_hdr_size = 0,
-   .header_split   = 0, /**< Header Split disabled */
-   .hw_ip_checksum = 0, /**< IP checksum offload enabled */
-   .hw_vlan_filter = 0, /**< VLAN filtering disabled */
-   .jumbo_frame= 0, /**< Jumbo Frame Support disabled */
-   .hw_strip_crc   = 1, /**< CRC stripped by hardware */
+   .offloads = 0,
+   .ignore_offload_bitfield = 1,
},
.txmode = {
.mq_mode = ETH_MQ_TX_NONE,
@@ -99,12 +95,8 @@ static struct rte_eth_conf rss_pmd_conf = {
.rxmode = {
.mq_mode = ETH_MQ_RX_RSS,
.max_rx_pkt_len = ETHER_MAX_LEN,
-   .split_hdr_size = 0,
-   .header_split   = 0, /**< Header Split disabled */
-   .hw_ip_checksum = 0, /**< IP checksum offload enabled */
-   .hw_vlan_filter = 0, /**< VLAN filtering disabled */
-   .jumbo_frame= 0, /**< Jumbo Frame Support disabled */
-   .hw_strip_crc   = 1, /**< CRC stripped by hardware */
+   .offloads = 0,
+   .ignore_offload_bitfield = 1,
},
.txmode = {
.mq_mode = ETH_MQ_TX_NONE,
-- 
2.7.4

[dpdk-dev] [PATCH 1/2] test/link_bonding: disable HW CRC strip by default.

[Phil Yang]

Since the tested virtual device has no capacity of HW CRC offload,
disable it for the link bonding test by default. Otherwise, it will
cause test unexpected failure.

Fixes: b219c8c("test: enable HW CRC strip by default")

Signed-off-by: Phil Yang 
---
 test/test/test_link_bonding.c | 18 --
 1 file changed, 4 insertions(+), 14 deletions(-)

diff --git a/test/test/test_link_bonding.c b/test/test/test_link_bonding.c
index 0ffd650..d996c02 100644
--- a/test/test/test_link_bonding.c
+++ b/test/test/test_link_bonding.c
@@ -137,14 +137,8 @@ static uint16_t vlan_id = 0x100;
 
 struct rte_eth_rxmode rx_mode = {
.max_rx_pkt_len = ETHER_MAX_LEN, /**< Default maximum frame length. */
-   .split_hdr_size = 0,
-   .header_split   = 0, /**< Header Split disabled. */
-   .hw_ip_checksum = 0, /**< IP checksum offload disabled. */
-   .hw_vlan_filter = 1, /**< VLAN filtering enabled. */
-   .hw_vlan_strip  = 1, /**< VLAN strip enabled. */
-   .hw_vlan_extend = 0, /**< Extended VLAN disabled. */
-   .jumbo_frame= 0, /**< Jumbo Frame Support disabled. */
-   .hw_strip_crc   = 1, /**< CRC stripping by hardware enabled. */
+   .offloads = 0,
+   .ignore_offload_bitfield = 1,
 };
 
 struct rte_fdir_conf fdir_conf = {
@@ -158,12 +152,8 @@ static struct rte_eth_conf default_pmd_conf = {
.rxmode = {
.mq_mode = ETH_MQ_RX_NONE,
.max_rx_pkt_len = ETHER_MAX_LEN,
-   .split_hdr_size = 0,
-   .header_split   = 0, /**< Header Split disabled */
-   .hw_ip_checksum = 0, /**< IP checksum offload enabled */
-   .hw_vlan_filter = 0, /**< VLAN filtering disabled */
-   .jumbo_frame= 0, /**< Jumbo Frame Support disabled */
-   .hw_strip_crc   = 1, /**< CRC stripped by hardware */
+   .offloads = 0,
+   .ignore_offload_bitfield = 1,
},
.txmode = {
.mq_mode = ETH_MQ_TX_NONE,
-- 
2.7.4

Re: [dpdk-dev] [PATCH v5 0/9] support in-order feature

[Maxime Coquelin]

On 07/02/2018 03:56 PM, Marvin Liu wrote:

In latest virtio-spec, new feature bit VIRTIO_F_IN_ORDER was introduced.
When this feature has been negotiated, virtio driver will use
descriptors in ring order: starting from offset 0 in the table, and
wrapping around at the end of the table. Vhost devices will always use
descriptors in the same order in which they have been made available.
This can reduce virtio accesses to used ring.

Based on updated virtio-spec, this series realized IN_ORDER prototype
in virtio driver. Due to new [RT]x path added into selection, also add
two new parameters mrg_rx and in_order into virtio-user vdev parameters
list. This will allow user to configure feature bits thus can impact
[RT]x path selection.

Performance of virtio user with IN_ORDER feature:

 Platform: Purely
 CPU: Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
 DPDK baseline: 18.05
 Setup: testpmd with vhost vdev + testpmd with virtio vdev

 +--+--+--+-+
 |Vhost->Virtio |1 Queue   |2 Queues  |4 Queues |
 +--+--+--+-+
 |Inorder   |12.0Mpps  |24.2Mpps  |26.0Mpps |
 |Normal|12.1Mpps  |18.5Mpps  |18.9Mpps |
 +--+--+--+-+
 
 +--+--++-+

 |Virtio->Vhost |1 Queue   |2 Queues|4 Queues |
 +--+--++-+
 |Inorder   |13.8Mpps  |10.7 ~ 15.2Mpps |11.5Mpps |
 |Normal|13.3Mpps  |9.8 ~ 14Mpps|10.5Mpps |
 +--+--++-+
 
 +-+--+++

 |Loopback |1 Queue   |2 Queues|4 Queues|
 +-+--+++
 |Inorder  |7.4Mpps   |9.1 ~ 11.6Mpps  |10.5 ~ 11.3Mpps |
 +-+--+++
 |Normal   |7.5Mpps   |7.7 ~ 9.0Mpps   |7.6 ~ 7.8Mpps   |
 +-+--+++

v5:
- disable simple Tx when in-order negotiated
- doc update

v4:
- disable simple [RT]x function for ARM
- squash doc update into relevant patches
- fix git-check-log and checkpatch errors

v3:
- refine [RT]x function selection logic
- fix in-order mergeable packets index error
- combine unsupport mask patch
- doc virtio in-order update
- fix checkpatch error

v2:
- merge to latest dpdk-net-virtio
- not use in_direct for normal xmit packets
- update available ring for each descriptor
- clean up IN_ORDER xmit function
- unmask feature bits when disabled in_order or mgr_rxbuf
- extract common part between IN_ORDER and normal functions
- update performance result

Marvin Liu (9):
   vhost: advertise support in-order feature
   net/virtio: add in-order feature bit definition
   net/virtio-user: add unsupported features mask
   net/virtio-user: add mrg-rxbuf and in-order vdev parameters
   net/virtio: free in-order descriptors before device start
   net/virtio: extract common part for in-order functions
   net/virtio: support in-order Rx and Tx
   net/virtio: add in-order Rx/Tx into selection
   net/virtio: advertise support in-order feature

  doc/guides/nics/virtio.rst|  23 +-
  drivers/net/virtio/virtio_ethdev.c|  32 +-
  drivers/net/virtio/virtio_ethdev.h|   7 +
  drivers/net/virtio/virtio_pci.h   |   8 +
  drivers/net/virtio/virtio_rxtx.c  | 639 --
  .../net/virtio/virtio_user/virtio_user_dev.c  |  30 +-
  .../net/virtio/virtio_user/virtio_user_dev.h  |   4 +-
  drivers/net/virtio/virtio_user_ethdev.c   |  47 +-
  drivers/net/virtio/virtqueue.c|   8 +
  drivers/net/virtio/virtqueue.h|   2 +
  lib/librte_vhost/socket.c |   6 +
  lib/librte_vhost/vhost.h  |  10 +-
  12 files changed, 736 insertions(+), 80 deletions(-)



Applied to dpdk-next-virtio/master.

Thanks,
Maxime

Re: [dpdk-dev] [RFC 0/3] ethdev: add IP address and TCP/UDP port rewrite actions to flow API

[Jack Min]

Hey,

I've one question: 

Do those APIs support modify inner IP headers? 

In our scenarios, we need to specific that  NAT action is on inner or outer IP 
header.

Thanks,
-Jack

> -Original Message-
> From: dev  On Behalf Of Rahul Lakkireddy
> Sent: June 22, 2018 17:56
> To: dev@dpdk.org
> Cc: shag...@chelsio.com; indra...@chelsio.com; nirran...@chelsio.com
> Subject: [dpdk-dev] [RFC 0/3] ethdev: add IP address and TCP/UDP port rewrite
> actions to flow API
> 
> This series of patches add support for actions:
> - OF_SET_NW_IPV4_SRC - set a new IPv4 source address.
> - OF_SET_NW_IPV4_DST - set a new IPv4 destination address.
> - OF_SET_NW_IPV6_SRC - set a new IPv6 source address.
> - OF_SET_NW_IPV6_DST - set a new IPv6 destination address.
> - OF_SET_TP_SRC - set a new TCP/UDP source port number.
> - OF_SET_TP_DST - set a new TCP/UDP destination port number.
> 
> These actions are useful in Network Address Translation use case to edit IP
> address and TCP/UDP port numbers before switching the packets out to the
> destination device port.
> 
> The IP address and TCP/UDP port rewrite actions are based on
> OFPAT_SET_NW_SRC, OFPAT_SET_NW_DST, OFPAT_SET_TP_SRC, and
> OFPAT_SET_TP_DST actions from OpenFlow Specification [1].
> 
> Patch 1 adds support for IP address rewrite to rte_flow and testpmd.
> 
> Patch 2 adds support for TCP/UDP port rewrite to rte_flow and testpmd.
> 
> Patch 3 shows CXGBE PMD example to offload these actions to hardware.
> 
> Feedback and suggestions will be much appreciated.
> 
> Thanks,
> Rahul
> 
> [1]
> https://emea01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.open
> networking.org%2Fimages%2Fopenflow-switch-
> v1.5.1.pdf&data=02%7C01%7Cxuemingl%40mellanox.com%7C74c8ce3b79954b2f
> 123208d5d8268617%7Ca652971c7d2e4d9ba6a4d149256f461b%7C0%7C1%7C636
> 652582334142929&sdata=uoFc1KjVTkGI9fdaPNXvXYRmh%2F6HLMKT79IwgvDSH
> 0k%3D&reserved=0
> 
> Shagun Agrawal (3):
>   ethdev: add flow api actions to modify IP addresses
>   ethdev: add flow api actions to modify TCP/UDP port numbers
>   net/cxgbe: add flow actions to modify IP and TCP/UDP port address
> 
>  app/test-pmd/cmdline_flow.c | 150
> 
>  app/test-pmd/config.c   |  12 +++
>  doc/guides/prog_guide/rte_flow.rst  | 102 +++
>  doc/guides/testpmd_app_ug/testpmd_funcs.rst |  24 +
>  drivers/net/cxgbe/base/common.h |   1 +
>  drivers/net/cxgbe/base/t4fw_interface.h |  50 ++
>  drivers/net/cxgbe/cxgbe_filter.c|  23 -
>  drivers/net/cxgbe/cxgbe_filter.h|  26 -
>  drivers/net/cxgbe/cxgbe_flow.c  | 127 +++
>  drivers/net/cxgbe/cxgbe_main.c  |  10 ++
>  lib/librte_ethdev/rte_flow.c|  12 +++
>  lib/librte_ethdev/rte_flow.h|  92 +
>  12 files changed, 625 insertions(+), 4 deletions(-)
> 
> --
> 2.14.1

[dpdk-dev] [PATCH v5 0/5] eventdev: add interrupt driven queues to Rx adapter

[Nikhil Rao]

This patch series adds support for interrupt driven queues to the
ethernet Rx adapter, the first 3 patches prepare the code to
handle both poll and interrupt driven Rx queues, the 4th patch
patch has code changes specific to interrupt driven queues and
the final patch has test code.

Changelog:

v4->v5:
* Fix meson build breakage

v3->v4:

* Fix FreeBSD build breakage.

v2->v3:

* Fix shared build breakage.

* Fix FreeBSD build breakage.

* Reduce epoll maxevents parameter by 1, since thread wakeup
  uses pthread_cancel as opposed to an exit message through a
  file monitored by epoll_wait().

* Check intr_handle before access, it is NULL when zero Rx queue
  interrupts are configured.

* Remove thread_stop flag, in the event of a pthread_cancel, it is
  not possible to check this flag thread stack is unwound without
  returning to rxa_intr_thread.

v1->v2:

* Move rte_service_component_runstate_set such that it
  is called only when cap & RTE__EVENT_ETH_RX_ADAPTER_CAP_INTERNAL_PORT
  is false. (Jerin Jacob)

* Fix meson build. (Jerin Jacob)

* Replace calls to pthread_* with rte_ctrl_thread_create().
  (Jerin Jacob)

* Move adapter test code to separate patch. (Jerin Jacob)

Note: I haven't removed the note about devices created
rte_event_eth_rx_adapter_create, will fix in a separate patch.

Nikhil Rao (5):
  eventdev: standardize Rx adapter internal function names
  eventdev: improve err handling for Rx adapter queue add/del
  eventdev: move Rx adapter eth Rx to separate function
  eventdev: add interrupt driven queues to Rx adapter
  eventdev: add Rx adapter tests for interrupt driven queues

 config/rte_config.h|1 +
 lib/librte_eventdev/rte_event_eth_rx_adapter.h |5 +-
 lib/librte_eventdev/rte_event_eth_rx_adapter.c | 1526 +---
 test/test/test_event_eth_rx_adapter.c  |  261 +++-
 .../prog_guide/event_ethernet_rx_adapter.rst   |   24 +
 config/common_base |1 +
 lib/librte_eventdev/Makefile   |9 +-
 lib/librte_eventdev/meson.build|9 +-
 8 files changed, 1596 insertions(+), 240 deletions(-)

-- 
1.8.3.1

[dpdk-dev] [PATCH v5 1/5] eventdev: standardize Rx adapter internal function names

[Nikhil Rao]

Add a common prefix to function names and rename
few to better match functionality

Signed-off-by: Nikhil Rao 
Acked-by: Jerin Jacob 
---
 lib/librte_eventdev/rte_event_eth_rx_adapter.c | 167 -
 1 file changed, 80 insertions(+), 87 deletions(-)

diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c 
b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
index ce1f62d..9361d48 100644
--- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
+++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
@@ -129,30 +129,30 @@ struct eth_rx_queue_info {
 static struct rte_event_eth_rx_adapter **event_eth_rx_adapter;
 
 static inline int
-valid_id(uint8_t id)
+rxa_validate_id(uint8_t id)
 {
return id < RTE_EVENT_ETH_RX_ADAPTER_MAX_INSTANCE;
 }
 
 #define RTE_EVENT_ETH_RX_ADAPTER_ID_VALID_OR_ERR_RET(id, retval) do { \
-   if (!valid_id(id)) { \
+   if (!rxa_validate_id(id)) { \
RTE_EDEV_LOG_ERR("Invalid eth Rx adapter id = %d\n", id); \
return retval; \
} \
 } while (0)
 
 static inline int
-sw_rx_adapter_queue_count(struct rte_event_eth_rx_adapter *rx_adapter)
+rxa_sw_adapter_queue_count(struct rte_event_eth_rx_adapter *rx_adapter)
 {
return rx_adapter->num_rx_polled;
 }
 
 /* Greatest common divisor */
-static uint16_t gcd_u16(uint16_t a, uint16_t b)
+static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
 {
uint16_t r = a % b;
 
-   return r ? gcd_u16(b, r) : b;
+   return r ? rxa_gcd_u16(b, r) : b;
 }
 
 /* Returns the next queue in the polling sequence
@@ -160,7 +160,7 @@ static uint16_t gcd_u16(uint16_t a, uint16_t b)
  * http://kb.linuxvirtualserver.org/wiki/Weighted_Round-Robin_Scheduling
  */
 static int
-wrr_next(struct rte_event_eth_rx_adapter *rx_adapter,
+rxa_wrr_next(struct rte_event_eth_rx_adapter *rx_adapter,
 unsigned int n, int *cw,
 struct eth_rx_poll_entry *eth_rx_poll, uint16_t max_wt,
 uint16_t gcd, int prev)
@@ -190,7 +190,7 @@ static uint16_t gcd_u16(uint16_t a, uint16_t b)
 
 /* Precalculate WRR polling sequence for all queues in rx_adapter */
 static int
-eth_poll_wrr_calc(struct rte_event_eth_rx_adapter *rx_adapter)
+rxa_calc_wrr_sequence(struct rte_event_eth_rx_adapter *rx_adapter)
 {
uint16_t d;
uint16_t q;
@@ -239,7 +239,7 @@ static uint16_t gcd_u16(uint16_t a, uint16_t b)
rx_poll[poll_q].eth_rx_qid = q;
max_wrr_pos += wt;
max_wt = RTE_MAX(max_wt, wt);
-   gcd = (gcd) ? gcd_u16(gcd, wt) : wt;
+   gcd = (gcd) ? rxa_gcd_u16(gcd, wt) : wt;
poll_q++;
}
}
@@ -259,7 +259,7 @@ static uint16_t gcd_u16(uint16_t a, uint16_t b)
int prev = -1;
int cw = -1;
for (i = 0; i < max_wrr_pos; i++) {
-   rx_wrr[i] = wrr_next(rx_adapter, poll_q, &cw,
+   rx_wrr[i] = rxa_wrr_next(rx_adapter, poll_q, &cw,
 rx_poll, max_wt, gcd, prev);
prev = rx_wrr[i];
}
@@ -276,7 +276,7 @@ static uint16_t gcd_u16(uint16_t a, uint16_t b)
 }
 
 static inline void
-mtoip(struct rte_mbuf *m, struct ipv4_hdr **ipv4_hdr,
+rxa_mtoip(struct rte_mbuf *m, struct ipv4_hdr **ipv4_hdr,
struct ipv6_hdr **ipv6_hdr)
 {
struct ether_hdr *eth_hdr = rte_pktmbuf_mtod(m, struct ether_hdr *);
@@ -315,7 +315,7 @@ static uint16_t gcd_u16(uint16_t a, uint16_t b)
 
 /* Calculate RSS hash for IPv4/6 */
 static inline uint32_t
-do_softrss(struct rte_mbuf *m, const uint8_t *rss_key_be)
+rxa_do_softrss(struct rte_mbuf *m, const uint8_t *rss_key_be)
 {
uint32_t input_len;
void *tuple;
@@ -324,7 +324,7 @@ static uint16_t gcd_u16(uint16_t a, uint16_t b)
struct ipv4_hdr *ipv4_hdr;
struct ipv6_hdr *ipv6_hdr;
 
-   mtoip(m, &ipv4_hdr, &ipv6_hdr);
+   rxa_mtoip(m, &ipv4_hdr, &ipv6_hdr);
 
if (ipv4_hdr) {
ipv4_tuple.src_addr = rte_be_to_cpu_32(ipv4_hdr->src_addr);
@@ -343,13 +343,13 @@ static uint16_t gcd_u16(uint16_t a, uint16_t b)
 }
 
 static inline int
-rx_enq_blocked(struct rte_event_eth_rx_adapter *rx_adapter)
+rxa_enq_blocked(struct rte_event_eth_rx_adapter *rx_adapter)
 {
return !!rx_adapter->enq_block_count;
 }
 
 static inline void
-rx_enq_block_start_ts(struct rte_event_eth_rx_adapter *rx_adapter)
+rxa_enq_block_start_ts(struct rte_event_eth_rx_adapter *rx_adapter)
 {
if (rx_adapter->rx_enq_block_start_ts)
return;
@@ -362,13 +362,13 @@ static uint16_t gcd_u16(uint16_t a, uint16_t b)
 }
 
 static inline void
-rx_enq_block_end_ts(struct rte_event_eth_rx_adapter *rx_adapter,
+rxa_enq_block_end_ts(struct rte_event_eth_rx_adapter *rx_adapter,
struct rte_event_eth_rx_adapter_stats *stats)
 {
if (unlikely(

[dpdk-dev] [PATCH v5 2/5] eventdev: improve err handling for Rx adapter queue add/del

[Nikhil Rao]

The new WRR sequence applicable after queue add/del is set
up after setting the new queue state, so a memory allocation
failure will leave behind an incorrect state.

This change separates the memory sizing + allocation for the
Rx poll and WRR array from calculation of the WRR sequence.
If there is a memory allocation failure, existing Rx queue
configuration remains unchanged.

Signed-off-by: Nikhil Rao 
---
 lib/librte_eventdev/rte_event_eth_rx_adapter.c | 418 ++---
 1 file changed, 302 insertions(+), 116 deletions(-)

diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c 
b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
index 9361d48..926f83a 100644
--- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
+++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
@@ -109,10 +109,16 @@ struct eth_device_info {
 * rx_adapter_stop callback needs to be invoked
 */
uint8_t dev_rx_started;
-   /* If nb_dev_queues > 0, the start callback will
+   /* Number of queues added for this device */
+   uint16_t nb_dev_queues;
+   /* If nb_rx_poll > 0, the start callback will
 * be invoked if not already invoked
 */
-   uint16_t nb_dev_queues;
+   uint16_t nb_rx_poll;
+   /* sum(wrr(q)) for all queues within the device
+* useful when deleting all device queues
+*/
+   uint32_t wrr_len;
 };
 
 /* Per Rx queue */
@@ -188,13 +194,170 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
}
 }
 
-/* Precalculate WRR polling sequence for all queues in rx_adapter */
+static inline int
+rxa_polled_queue(struct eth_device_info *dev_info,
+   int rx_queue_id)
+{
+   struct eth_rx_queue_info *queue_info;
+
+   queue_info = &dev_info->rx_queue[rx_queue_id];
+   return !dev_info->internal_event_port &&
+   dev_info->rx_queue &&
+   queue_info->queue_enabled && queue_info->wt != 0;
+}
+
+/* Calculate size of the eth_rx_poll and wrr_sched arrays
+ * after deleting poll mode rx queues
+ */
+static void
+rxa_calc_nb_post_poll_del(struct rte_event_eth_rx_adapter *rx_adapter,
+   struct eth_device_info *dev_info,
+   int rx_queue_id,
+   uint32_t *nb_rx_poll,
+   uint32_t *nb_wrr)
+{
+   uint32_t poll_diff;
+   uint32_t wrr_len_diff;
+
+   if (rx_queue_id == -1) {
+   poll_diff = dev_info->nb_rx_poll;
+   wrr_len_diff = dev_info->wrr_len;
+   } else {
+   poll_diff = rxa_polled_queue(dev_info, rx_queue_id);
+   wrr_len_diff = poll_diff ? dev_info->rx_queue[rx_queue_id].wt :
+   0;
+   }
+
+   *nb_rx_poll = rx_adapter->num_rx_polled - poll_diff;
+   *nb_wrr = rx_adapter->wrr_len - wrr_len_diff;
+}
+
+/* Calculate nb_rx_* after adding poll mode rx queues
+ */
+static void
+rxa_calc_nb_post_add_poll(struct rte_event_eth_rx_adapter *rx_adapter,
+   struct eth_device_info *dev_info,
+   int rx_queue_id,
+   uint16_t wt,
+   uint32_t *nb_rx_poll,
+   uint32_t *nb_wrr)
+{
+   uint32_t poll_diff;
+   uint32_t wrr_len_diff;
+
+   if (rx_queue_id == -1) {
+   poll_diff = dev_info->dev->data->nb_rx_queues -
+   dev_info->nb_rx_poll;
+   wrr_len_diff = wt*dev_info->dev->data->nb_rx_queues
+   - dev_info->wrr_len;
+   } else {
+   poll_diff = !rxa_polled_queue(dev_info, rx_queue_id);
+   wrr_len_diff = rxa_polled_queue(dev_info, rx_queue_id) ?
+   wt - dev_info->rx_queue[rx_queue_id].wt :
+   wt;
+   }
+
+   *nb_rx_poll = rx_adapter->num_rx_polled + poll_diff;
+   *nb_wrr = rx_adapter->wrr_len + wrr_len_diff;
+}
+
+/* Calculate nb_rx_* after adding rx_queue_id */
+static void
+rxa_calc_nb_post_add(struct rte_event_eth_rx_adapter *rx_adapter,
+   struct eth_device_info *dev_info,
+   int rx_queue_id,
+   uint16_t wt,
+   uint32_t *nb_rx_poll,
+   uint32_t *nb_wrr)
+{
+   rxa_calc_nb_post_add_poll(rx_adapter, dev_info, rx_queue_id,
+   wt, nb_rx_poll, nb_wrr);
+}
+
+/* Calculate nb_rx_* after deleting rx_queue_id */
+static void
+rxa_calc_nb_post_del(struct rte_event_eth_rx_adapter *rx_adapter,
+   struct eth_device_info *dev_info,
+   int rx_queue_id,
+   uint32_t *nb_rx_poll,
+   uint32_t *nb_wrr)
+{
+   rxa_calc_nb_post_poll_del(rx_adapter, dev_info, rx_queue_id, nb_rx_poll,
+   nb_wrr);
+}
+
+/*
+ * Allocate the rx_poll array
+ */
+static struct eth_rx_poll_entry *
+rxa_alloc_poll(struct rte_event_eth_rx_adapter *rx_adapter,
+   uint32_t nu

[dpdk-dev] [PATCH v5 4/5] eventdev: add interrupt driven queues to Rx adapter

[Nikhil Rao]

Add support for interrupt driven queues when eth device is
configured for rxq interrupts and servicing weight for the
queue is configured to be zero.

A interrupt driven packet received counter has been added to
rte_event_eth_rx_adapter_stats.

Signed-off-by: Nikhil Rao 
---
 config/rte_config.h|   1 +
 lib/librte_eventdev/rte_event_eth_rx_adapter.h |   5 +-
 lib/librte_eventdev/rte_event_eth_rx_adapter.c | 940 -
 .../prog_guide/event_ethernet_rx_adapter.rst   |  24 +
 config/common_base |   1 +
 lib/librte_eventdev/Makefile   |   9 +-
 lib/librte_eventdev/meson.build|   9 +-
 7 files changed, 958 insertions(+), 31 deletions(-)

diff --git a/config/rte_config.h b/config/rte_config.h
index 0ba0ead..b1fb8cd 100644
--- a/config/rte_config.h
+++ b/config/rte_config.h
@@ -64,6 +64,7 @@
 #define RTE_EVENT_MAX_DEVS 16
 #define RTE_EVENT_MAX_QUEUES_PER_DEV 64
 #define RTE_EVENT_TIMER_ADAPTER_NUM_MAX 32
+#define RTE_EVENT_ETH_INTR_RING_SIZE 1024
 #define RTE_EVENT_CRYPTO_ADAPTER_MAX_INSTANCE 32
 
 /* rawdev defines */
diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.h 
b/lib/librte_eventdev/rte_event_eth_rx_adapter.h
index 307b2b5..97f25e9 100644
--- a/lib/librte_eventdev/rte_event_eth_rx_adapter.h
+++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.h
@@ -64,8 +64,7 @@
  * the service function ID of the adapter in this case.
  *
  * Note:
- * 1) Interrupt driven receive queues are currently unimplemented.
- * 2) Devices created after an instance of rte_event_eth_rx_adapter_create
+ * 1) Devices created after an instance of rte_event_eth_rx_adapter_create
  *  should be added to a new instance of the rx adapter.
  */
 
@@ -199,6 +198,8 @@ struct rte_event_eth_rx_adapter_stats {
 * block cycles can be used to compute the percentage of
 * cycles the service is blocked by the event device.
 */
+   uint64_t rx_intr_packets;
+   /**< Received packet count for interrupt mode Rx queues */
 };
 
 /**
diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c 
b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
index 8fe037f..42dd7f8 100644
--- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
+++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
@@ -2,6 +2,11 @@
  * Copyright(c) 2017 Intel Corporation.
  * All rights reserved.
  */
+#if defined(LINUX)
+#include 
+#endif
+#include 
+
 #include 
 #include 
 #include 
@@ -11,6 +16,7 @@
 #include 
 #include 
 #include 
+#include 
 
 #include "rte_eventdev.h"
 #include "rte_eventdev_pmd.h"
@@ -24,6 +30,22 @@
 #define ETH_RX_ADAPTER_MEM_NAME_LEN32
 
 #define RSS_KEY_SIZE   40
+/* value written to intr thread pipe to signal thread exit */
+#define ETH_BRIDGE_INTR_THREAD_EXIT1
+/* Sentinel value to detect initialized file handle */
+#define INIT_FD-1
+
+/*
+ * Used to store port and queue ID of interrupting Rx queue
+ */
+union queue_data {
+   RTE_STD_C11
+   void *ptr;
+   struct {
+   uint16_t port;
+   uint16_t queue;
+   };
+};
 
 /*
  * There is an instance of this struct per polled Rx queue added to the
@@ -75,6 +97,30 @@ struct rte_event_eth_rx_adapter {
uint16_t enq_block_count;
/* Block start ts */
uint64_t rx_enq_block_start_ts;
+   /* epoll fd used to wait for Rx interrupts */
+   int epd;
+   /* Num of interrupt driven interrupt queues */
+   uint32_t num_rx_intr;
+   /* Used to send  of interrupting Rx queues from
+* the interrupt thread to the Rx thread
+*/
+   struct rte_ring *intr_ring;
+   /* Rx Queue data (dev id, queue id) for the last non-empty
+* queue polled
+*/
+   union queue_data qd;
+   /* queue_data is valid */
+   int qd_valid;
+   /* Interrupt ring lock, synchronizes Rx thread
+* and interrupt thread
+*/
+   rte_spinlock_t intr_ring_lock;
+   /* event array passed to rte_poll_wait */
+   struct rte_epoll_event *epoll_events;
+   /* Count of interrupt vectors in use */
+   uint32_t num_intr_vec;
+   /* Thread blocked on Rx interrupts */
+   pthread_t rx_intr_thread;
/* Configuration callback for rte_service configuration */
rte_event_eth_rx_adapter_conf_cb conf_cb;
/* Configuration callback argument */
@@ -93,6 +139,8 @@ struct rte_event_eth_rx_adapter {
uint32_t service_id;
/* Adapter started flag */
uint8_t rxa_started;
+   /* Adapter ID */
+   uint8_t id;
 } __rte_cache_aligned;
 
 /* Per eth device */
@@ -111,19 +159,40 @@ struct eth_device_info {
uint8_t dev_rx_started;
/* Number of queues added for this device */
uint16_t nb_dev_queues;
-   /* If nb_rx_poll > 0, the start callback will
+   /* Number of poll based queues
+* If nb_rx_poll > 0, the start callback will
 * be i

[dpdk-dev] [PATCH v5 5/5] eventdev: add Rx adapter tests for interrupt driven queues

[Nikhil Rao]

Add test for queue add and delete, the add/delete calls
also switch queues between poll and interrupt mode.

Signed-off-by: Nikhil Rao 
---
 test/test/test_event_eth_rx_adapter.c | 261 +++---
 1 file changed, 242 insertions(+), 19 deletions(-)

diff --git a/test/test/test_event_eth_rx_adapter.c 
b/test/test/test_event_eth_rx_adapter.c
index d432731..2337e54 100644
--- a/test/test/test_event_eth_rx_adapter.c
+++ b/test/test/test_event_eth_rx_adapter.c
@@ -25,28 +25,17 @@ struct event_eth_rx_adapter_test_params {
struct rte_mempool *mp;
uint16_t rx_rings, tx_rings;
uint32_t caps;
+   int rx_intr_port_inited;
+   uint16_t rx_intr_port;
 };
 
 static struct event_eth_rx_adapter_test_params default_params;
 
 static inline int
-port_init(uint8_t port, struct rte_mempool *mp)
+port_init_common(uint8_t port, const struct rte_eth_conf *port_conf,
+   struct rte_mempool *mp)
 {
-   static const struct rte_eth_conf port_conf_default = {
-   .rxmode = {
-   .mq_mode = ETH_MQ_RX_RSS,
-   .max_rx_pkt_len = ETHER_MAX_LEN
-   },
-   .rx_adv_conf = {
-   .rss_conf = {
-   .rss_hf = ETH_RSS_IP |
- ETH_RSS_TCP |
- ETH_RSS_UDP,
-   }
-   }
-   };
const uint16_t rx_ring_size = 512, tx_ring_size = 512;
-   struct rte_eth_conf port_conf = port_conf_default;
int retval;
uint16_t q;
struct rte_eth_dev_info dev_info;
@@ -54,7 +43,7 @@ struct event_eth_rx_adapter_test_params {
if (!rte_eth_dev_is_valid_port(port))
return -1;
 
-   retval = rte_eth_dev_configure(port, 0, 0, &port_conf);
+   retval = rte_eth_dev_configure(port, 0, 0, port_conf);
 
rte_eth_dev_info_get(port, &dev_info);
 
@@ -64,7 +53,7 @@ struct event_eth_rx_adapter_test_params {
 
/* Configure the Ethernet device. */
retval = rte_eth_dev_configure(port, default_params.rx_rings,
-   default_params.tx_rings, &port_conf);
+   default_params.tx_rings, port_conf);
if (retval != 0)
return retval;
 
@@ -104,6 +93,77 @@ struct event_eth_rx_adapter_test_params {
return 0;
 }
 
+static inline int
+port_init_rx_intr(uint8_t port, struct rte_mempool *mp)
+{
+   static const struct rte_eth_conf port_conf_default = {
+   .rxmode = {
+   .mq_mode = ETH_MQ_RX_RSS,
+   .max_rx_pkt_len = ETHER_MAX_LEN
+   },
+   .intr_conf = {
+   .rxq = 1,
+   },
+   };
+
+   return port_init_common(port, &port_conf_default, mp);
+}
+
+static inline int
+port_init(uint8_t port, struct rte_mempool *mp)
+{
+   static const struct rte_eth_conf port_conf_default = {
+   .rxmode = {
+   .mq_mode = ETH_MQ_RX_RSS,
+   .max_rx_pkt_len = ETHER_MAX_LEN
+   },
+   .rx_adv_conf = {
+   .rss_conf = {
+   .rss_hf = ETH_RSS_IP |
+   ETH_RSS_TCP |
+   ETH_RSS_UDP,
+   }
+   }
+   };
+
+   return port_init_common(port, &port_conf_default, mp);
+}
+
+static int
+init_port_rx_intr(int num_ports)
+{
+   int retval;
+   uint16_t portid;
+   int err;
+
+   default_params.mp = rte_pktmbuf_pool_create("packet_pool",
+  NB_MBUFS,
+  MBUF_CACHE_SIZE,
+  MBUF_PRIV_SIZE,
+  RTE_MBUF_DEFAULT_BUF_SIZE,
+  rte_socket_id());
+   if (!default_params.mp)
+   return -ENOMEM;
+
+   RTE_ETH_FOREACH_DEV(portid) {
+   retval = port_init_rx_intr(portid, default_params.mp);
+   if (retval)
+   continue;
+   err = rte_event_eth_rx_adapter_caps_get(TEST_DEV_ID, portid,
+   &default_params.caps);
+   if (err)
+   continue;
+   if (!(default_params.caps &
+   RTE_EVENT_ETH_RX_ADAPTER_CAP_INTERNAL_PORT)) {
+   default_params.rx_intr_port_inited = 1;
+   default_params.rx_intr_port = portid;
+   return 0;
+   }
+   rte_eth_dev_stop(portid);
+   }
+   return 0;
+}
+
 static int
 init_ports(int num_ports)
 {
@@ -181,6 +241,57 @@ struct event_eth_rx_adapter_test_params {

[dpdk-dev] [PATCH v5 3/5] eventdev: move Rx adapter eth Rx to separate function

[Nikhil Rao]

Create a separate function that handles eth receive and
enqueue to event buffer. This function will also be called for
interrupt driven receive queues.

Signed-off-by: Nikhil Rao 
Acked-by: Jerin Jacob 
---
 lib/librte_eventdev/rte_event_eth_rx_adapter.c | 67 ++
 1 file changed, 47 insertions(+), 20 deletions(-)

diff --git a/lib/librte_eventdev/rte_event_eth_rx_adapter.c 
b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
index 926f83a..8fe037f 100644
--- a/lib/librte_eventdev/rte_event_eth_rx_adapter.c
+++ b/lib/librte_eventdev/rte_event_eth_rx_adapter.c
@@ -616,6 +616,45 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
}
 }
 
+/* Enqueue packets fromto event buffer */
+static inline uint32_t
+rxa_eth_rx(struct rte_event_eth_rx_adapter *rx_adapter,
+   uint16_t port_id,
+   uint16_t queue_id,
+   uint32_t rx_count,
+   uint32_t max_rx)
+{
+   struct rte_mbuf *mbufs[BATCH_SIZE];
+   struct rte_eth_event_enqueue_buffer *buf =
+   &rx_adapter->event_enqueue_buffer;
+   struct rte_event_eth_rx_adapter_stats *stats =
+   &rx_adapter->stats;
+   uint16_t n;
+   uint32_t nb_rx = 0;
+
+   /* Don't do a batch dequeue from the rx queue if there isn't
+* enough space in the enqueue buffer.
+*/
+   while (BATCH_SIZE <= (RTE_DIM(buf->events) - buf->count)) {
+   if (buf->count >= BATCH_SIZE)
+   rxa_flush_event_buffer(rx_adapter);
+
+   stats->rx_poll_count++;
+   n = rte_eth_rx_burst(port_id, queue_id, mbufs, BATCH_SIZE);
+   if (unlikely(!n))
+   break;
+   rxa_buffer_mbufs(rx_adapter, port_id, queue_id, mbufs, n);
+   nb_rx += n;
+   if (rx_count + nb_rx > max_rx)
+   break;
+   }
+
+   if (buf->count >= BATCH_SIZE)
+   rxa_flush_event_buffer(rx_adapter);
+
+   return nb_rx;
+}
+
 /*
  * Polls receive queues added to the event adapter and enqueues received
  * packets to the event device.
@@ -633,17 +672,16 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
 rxa_poll(struct rte_event_eth_rx_adapter *rx_adapter)
 {
uint32_t num_queue;
-   uint16_t n;
uint32_t nb_rx = 0;
-   struct rte_mbuf *mbufs[BATCH_SIZE];
struct rte_eth_event_enqueue_buffer *buf;
uint32_t wrr_pos;
uint32_t max_nb_rx;
+   struct rte_event_eth_rx_adapter_stats *stats;
 
wrr_pos = rx_adapter->wrr_pos;
max_nb_rx = rx_adapter->max_nb_rx;
buf = &rx_adapter->event_enqueue_buffer;
-   struct rte_event_eth_rx_adapter_stats *stats = &rx_adapter->stats;
+   stats = &rx_adapter->stats;
 
/* Iterate through a WRR sequence */
for (num_queue = 0; num_queue < rx_adapter->wrr_len; num_queue++) {
@@ -658,32 +696,21 @@ static uint16_t rxa_gcd_u16(uint16_t a, uint16_t b)
rxa_flush_event_buffer(rx_adapter);
if (BATCH_SIZE > (ETH_EVENT_BUFFER_SIZE - buf->count)) {
rx_adapter->wrr_pos = wrr_pos;
-   return;
+   break;
}
 
-   stats->rx_poll_count++;
-   n = rte_eth_rx_burst(d, qid, mbufs, BATCH_SIZE);
-
-   if (n) {
-   stats->rx_packets += n;
-   /* The check before rte_eth_rx_burst() ensures that
-* all n mbufs can be buffered
-*/
-   rxa_buffer_mbufs(rx_adapter, d, qid, mbufs, n);
-   nb_rx += n;
-   if (nb_rx > max_nb_rx) {
-   rx_adapter->wrr_pos =
+   nb_rx += rxa_eth_rx(rx_adapter, d, qid, nb_rx, max_nb_rx);
+   if (nb_rx > max_nb_rx) {
+   rx_adapter->wrr_pos =
(wrr_pos + 1) % rx_adapter->wrr_len;
-   break;
-   }
+   break;
}
 
if (++wrr_pos == rx_adapter->wrr_len)
wrr_pos = 0;
}
 
-   if (buf->count >= BATCH_SIZE)
-   rxa_flush_event_buffer(rx_adapter);
+   stats->rx_packets += nb_rx;
 }
 
 static int
-- 
1.8.3.1

Re: [dpdk-dev] [PATCH v2] mempool/octeontx: fix pool to aura mapping

[Jerin Jacob]

-Original Message-
> Date: Mon,  2 Jul 2018 11:59:32 +0530
> From: Pavan Nikhilesh 
> To: jerin.ja...@caviumnetworks.com, santosh.shu...@caviumnetworks.com,
>  olivier.m...@6wind.com
> Cc: dev@dpdk.org, sta...@dpdk.org, Pavan Nikhilesh
>  
> Subject: [dpdk-dev] [PATCH v2] mempool/octeontx: fix pool to aura mapping
> X-Mailer: git-send-email 2.18.0
> 
> HW needs each pool to be mapped to an aura set of 16 auras.
> Previously, pool to aura mapping was considered to be 1:1.
> 
> Fixes: 02fd6c744350 ("mempool/octeontx: support allocation")
> Cc: sta...@dpdk.org
> 
> Signed-off-by: Pavan Nikhilesh 
> Acked-by: Santosh Shukla 


Acked-by: Jerin Jacob 

> ---
>  v2 Changes:
>  - use macro to avoid code duplication (Santosh).
>  - use uint16_t for gaura id.
> 
>  drivers/event/octeontx/timvf_evdev.c  |  2 +-
>  drivers/mempool/octeontx/octeontx_fpavf.c | 45 ++-
>  drivers/mempool/octeontx/octeontx_fpavf.h |  9 +
>  drivers/net/octeontx/octeontx_ethdev.c|  6 +--
>  drivers/net/octeontx/octeontx_rxtx.c  |  2 +-
>  5 files changed, 42 insertions(+), 22 deletions(-)
> 
> diff --git a/drivers/event/octeontx/timvf_evdev.c 
> b/drivers/event/octeontx/timvf_evdev.c
> index c4fbd2d86..8a045c250 100644
> --- a/drivers/event/octeontx/timvf_evdev.c
> +++ b/drivers/event/octeontx/timvf_evdev.c
> @@ -174,7 +174,7 @@ timvf_ring_start(const struct rte_event_timer_adapter 
> *adptr)
>   if (use_fpa) {
>   pool = (uintptr_t)((struct rte_mempool *)
>   timr->chunk_pool)->pool_id;
> - ret = octeontx_fpa_bufpool_gpool(pool);
> + ret = octeontx_fpa_bufpool_gaura(pool);
>   if (ret < 0) {
>   timvf_log_dbg("Unable to get gaura id");
>   ret = -ENOMEM;
> diff --git a/drivers/mempool/octeontx/octeontx_fpavf.c 
> b/drivers/mempool/octeontx/octeontx_fpavf.c
> index 7aecaa85d..e5918c866 100644
> --- a/drivers/mempool/octeontx/octeontx_fpavf.c
> +++ b/drivers/mempool/octeontx/octeontx_fpavf.c
> @@ -243,7 +243,7 @@ octeontx_fpapf_pool_setup(unsigned int gpool, unsigned 
> int buf_size,
>   POOL_LTYPE(0x2) | POOL_STYPE(0) | POOL_SET_NAT_ALIGN |
>   POOL_ENA;
> 
> - cfg.aid = 0;
> + cfg.aid = FPA_AURA_IDX(gpool);
>   cfg.pool_cfg = reg;
>   cfg.pool_stack_base = phys_addr;
>   cfg.pool_stack_end = phys_addr + memsz;
> @@ -327,7 +327,7 @@ octeontx_fpapf_aura_attach(unsigned int gpool_index)
>   hdr.vfid = gpool_index;
>   hdr.res_code = 0;
>   memset(&cfg, 0x0, sizeof(struct octeontx_mbox_fpa_cfg));
> - cfg.aid = gpool_index; /* gpool is guara */
> + cfg.aid = gpool_index << FPA_GAURA_SHIFT;
> 
>   ret = octeontx_mbox_send(&hdr, &cfg,
>   sizeof(struct octeontx_mbox_fpa_cfg),
> @@ -335,7 +335,8 @@ octeontx_fpapf_aura_attach(unsigned int gpool_index)
>   if (ret < 0) {
>   fpavf_log_err("Could not attach fpa ");
>   fpavf_log_err("aura %d to pool %d. Err=%d. FuncErr=%d\n",
> -   gpool_index, gpool_index, ret, hdr.res_code);
> +   gpool_index << FPA_GAURA_SHIFT, gpool_index, ret,
> +   hdr.res_code);
>   ret = -EACCES;
>   goto err;
>   }
> @@ -355,14 +356,15 @@ octeontx_fpapf_aura_detach(unsigned int gpool_index)
>   goto err;
>   }
> 
> - cfg.aid = gpool_index; /* gpool is gaura */
> + cfg.aid = gpool_index << FPA_GAURA_SHIFT;
>   hdr.coproc = FPA_COPROC;
>   hdr.msg = FPA_DETACHAURA;
>   hdr.vfid = gpool_index;
>   ret = octeontx_mbox_send(&hdr, &cfg, sizeof(cfg), NULL, 0);
>   if (ret < 0) {
>   fpavf_log_err("Couldn't detach FPA aura %d Err=%d FuncErr=%d\n",
> -   gpool_index, ret, hdr.res_code);
> +   gpool_index << FPA_GAURA_SHIFT, ret,
> +   hdr.res_code);
>   ret = -EINVAL;
>   }
> 
> @@ -469,6 +471,7 @@ octeontx_fpa_bufpool_free_count(uintptr_t handle)
>  {
>   uint64_t cnt, limit, avail;
>   uint8_t gpool;
> + uint16_t gaura;
>   uintptr_t pool_bar;
> 
>   if (unlikely(!octeontx_fpa_handle_valid(handle)))
> @@ -476,14 +479,16 @@ octeontx_fpa_bufpool_free_count(uintptr_t handle)
> 
>   /* get the gpool */
>   gpool = octeontx_fpa_bufpool_gpool(handle);
> + /* get the aura */
> + gaura = octeontx_fpa_bufpool_gaura(handle);
> 
>   /* Get pool bar address from handle */
>   pool_bar = handle & ~(uint64_t)FPA_GPOOL_MASK;
> 
>   cnt = fpavf_read64((void *)((uintptr_t)pool_bar +
> - FPA_VF_VHAURA_CNT(gpool)));
> + FPA_VF_VHAURA_CNT(gaura)));
>   limit = fpavf_read64((void *)((uintptr_t)pool_bar +
> - FPA_VF_VHAURA_CNT_LIMIT(gpool)));
> + FPA_VF_VHAURA_CN

Re: [dpdk-dev] 17.05 --> 17.11, minimum hash table key size

[Dumitrescu, Cristian]

> -Original Message-
> From: Bly, Mike [mailto:m...@ciena.com]
> Sent: Friday, June 29, 2018 9:01 PM
> To: Dumitrescu, Cristian ; Yeddula, Avinash
> ; dev@dpdk.org; dev ;
> us...@dpdk.org
> Subject: RE: 17.05 --> 17.11, minimum hash table key size
> 
> Cristian,
> 
> Thank you. I have done the recommended change. The next thing we have
> run into is the f_hash selection. We were historically using
> rte_hash_crc_8byte() from rte_has_crc.h. In looking at DPDK content, I do
> not see anything close in an rte_*.h file to use with the appropriate input
> parameter list. I did find use of hash_default_key8() in
> examples/ip_pipeline/pipeline/hash_func.h. Is there no formal rte_*.h
> replacement/update for this relative to the hash-table changes in 17.11+? Is
> the expectation that the implementer will create their own as done with the
> aforementioned ip_pipeline example?
> 
> -MikeB
> 

Hi Mike,

Yes, we are working to move the code from examples/ip_pipeline/hash_func.h into 
librte_table, Kevin is working on it right now.

Regards,
Cristian

[dpdk-dev] [PATCH v4 0/3] crypto/qat: move files to drivers/common directory

[Tomasz Jozwiak]

This patchset depends on QAT dynamic logging patchset and should be targetig on 
18.08.
Patchset refactors the PMD in order that files are split into several
places: common, crypto.
New drivers/common/qat are added and files split between locations.

Changes for v2:
  -  removed drivers/common/qat/qat
  -  updated meson.build files
  -  added description into qat.rst
  -  updated MAINTAINERS file 

Changes for v3:
  -  removed libcrypto detection from Makefile
  -  removed description about libcrypto detection from doc.
  -  renamed CONFIG_LIBCRYPTO_QAT define into BUILD_QAT_SYM

Changes for v4:
  -  fixed libcrypto dependency in 
 're-organise build file content' patch

Tomasz Jozwiak (3):
  crypto/qat: add weak functions
  crypto/qat: re-organise build file content
  crypto/qat: move common qat files to common dir

 MAINTAINERS|  1 +
 drivers/Makefile   |  2 +
 drivers/common/meson.build |  2 +-
 drivers/common/qat/Makefile| 49 ++
 drivers/common/qat/meson.build | 14 +++
 .../qat/qat_adf/adf_transport_access_macros.h  |  0
 .../{crypto => common}/qat/qat_adf/icp_qat_fw.h|  0
 .../{crypto => common}/qat/qat_adf/icp_qat_fw_la.h |  0
 .../{crypto => common}/qat/qat_adf/icp_qat_hw.h|  0
 drivers/{crypto => common}/qat/qat_common.c|  0
 drivers/{crypto => common}/qat/qat_common.h|  0
 drivers/{crypto => common}/qat/qat_device.c| 39 -
 drivers/{crypto => common}/qat/qat_device.h| 20 +
 drivers/{crypto => common}/qat/qat_logs.c  |  0
 drivers/{crypto => common}/qat/qat_logs.h  |  0
 drivers/{crypto => common}/qat/qat_qp.c|  0
 drivers/{crypto => common}/qat/qat_qp.h|  0
 drivers/crypto/Makefile|  1 -
 drivers/crypto/qat/Makefile| 40 --
 drivers/crypto/qat/README  |  8 
 drivers/crypto/qat/meson.build | 32 --
 drivers/crypto/qat/qat_asym_pmd.c  | 17 
 drivers/crypto/qat/qat_asym_pmd.h  | 15 ---
 drivers/crypto/qat/qat_comp_pmd.c  | 18 
 drivers/crypto/qat/qat_comp_pmd.h  | 29 -
 drivers/crypto/qat/qat_sym.h   |  8 
 drivers/crypto/qat/qat_sym_pmd.h   |  6 ++-
 27 files changed, 163 insertions(+), 138 deletions(-)
 create mode 100644 drivers/common/qat/Makefile
 create mode 100644 drivers/common/qat/meson.build
 rename drivers/{crypto => common}/qat/qat_adf/adf_transport_access_macros.h 
(100%)
 rename drivers/{crypto => common}/qat/qat_adf/icp_qat_fw.h (100%)
 rename drivers/{crypto => common}/qat/qat_adf/icp_qat_fw_la.h (100%)
 rename drivers/{crypto => common}/qat/qat_adf/icp_qat_hw.h (100%)
 rename drivers/{crypto => common}/qat/qat_common.c (100%)
 rename drivers/{crypto => common}/qat/qat_common.h (100%)
 rename drivers/{crypto => common}/qat/qat_device.c (88%)
 rename drivers/{crypto => common}/qat/qat_device.h (80%)
 rename drivers/{crypto => common}/qat/qat_logs.c (100%)
 rename drivers/{crypto => common}/qat/qat_logs.h (100%)
 rename drivers/{crypto => common}/qat/qat_qp.c (100%)
 rename drivers/{crypto => common}/qat/qat_qp.h (100%)
 delete mode 100644 drivers/crypto/qat/Makefile
 create mode 100644 drivers/crypto/qat/README
 delete mode 100644 drivers/crypto/qat/qat_asym_pmd.c
 delete mode 100644 drivers/crypto/qat/qat_asym_pmd.h
 delete mode 100644 drivers/crypto/qat/qat_comp_pmd.c
 delete mode 100644 drivers/crypto/qat/qat_comp_pmd.h

-- 
2.7.4

[dpdk-dev] [PATCH v4 2/3] crypto/qat: re-organise build file content

[Tomasz Jozwiak]

This patch groups sources and related dependencies into
common and sym sections in build files.

Signed-off-by: Tomasz Jozwiak 
Acked-by: Fiona Trahe 
---
 drivers/crypto/qat/Makefile  | 25 ++---
 drivers/crypto/qat/meson.build   | 16 ++--
 drivers/crypto/qat/qat_sym.h |  2 +-
 drivers/crypto/qat/qat_sym_pmd.h |  3 +--
 4 files changed, 26 insertions(+), 20 deletions(-)

diff --git a/drivers/crypto/qat/Makefile b/drivers/crypto/qat/Makefile
index 64f39fd..a939eca 100644
--- a/drivers/crypto/qat/Makefile
+++ b/drivers/crypto/qat/Makefile
@@ -15,19 +15,22 @@ CFLAGS += -O3
 
 # external library include paths
 CFLAGS += -I$(SRCDIR)/qat_adf
-LDLIBS += -lcrypto
-LDLIBS += -lrte_eal -lrte_mbuf -lrte_mempool -lrte_ring
+
+# library common source files
+SRCS-y += qat_device.c
+SRCS-y += qat_common.c
+SRCS-y += qat_logs.c
+SRCS-y += qat_qp.c
+
+# library symetric crypto source files
 LDLIBS += -lrte_cryptodev
-LDLIBS += -lrte_pci -lrte_bus_pci
+LDLIBS += -lcrypto
+SRCS-y += qat_sym.c
+SRCS-y += qat_sym_session.c
+SRCS-y += qat_sym_pmd.c
 
-# library source files
-SRCS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += qat_sym.c
-SRCS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += qat_device.c
-SRCS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += qat_qp.c
-SRCS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += qat_sym_session.c
-SRCS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += qat_common.c
-SRCS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += qat_logs.c
-SRCS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += qat_sym_pmd.c
+LDLIBS += -lrte_eal -lrte_mbuf -lrte_mempool
+LDLIBS += -lrte_pci -lrte_bus_pci
 
 # export include files
 SYMLINK-y-include +=
diff --git a/drivers/crypto/qat/meson.build b/drivers/crypto/qat/meson.build
index 6d01dac..0675426 100644
--- a/drivers/crypto/qat/meson.build
+++ b/drivers/crypto/qat/meson.build
@@ -2,15 +2,19 @@
 # Copyright(c) 2017-2018 Intel Corporation
 
 dep = dependency('libcrypto', required: false)
-if not dep.found()
-   build = false
-endif
+
+build = false
 sources = files('qat_common.c',
'qat_qp.c',
'qat_device.c',
-   'qat_logs.c',
-   'qat_sym_pmd.c', 'qat_sym.c', 'qat_sym_session.c')
+   'qat_logs.c')
+
+if dep.found()
+   sources += files('qat_sym_pmd.c', 'qat_sym.c', 'qat_sym_session.c')
+   pkgconfig_extra_libs += '-lcrypto'
+   build = true
+endif
+
 includes += include_directories('qat_adf')
 deps += ['bus_pci']
 ext_deps += dep
-pkgconfig_extra_libs += '-lcrypto'
diff --git a/drivers/crypto/qat/qat_sym.h b/drivers/crypto/qat/qat_sym.h
index f9e72a6..126c191 100644
--- a/drivers/crypto/qat/qat_sym.h
+++ b/drivers/crypto/qat/qat_sym.h
@@ -6,7 +6,6 @@
 #define _QAT_SYM_H_
 
 #include 
-
 #include 
 
 #include "qat_common.h"
@@ -153,4 +152,5 @@ qat_sym_process_response(void **op, uint8_t *resp)
}
*op = (void *)rx_op;
 }
+
 #endif /* _QAT_SYM_H_ */
diff --git a/drivers/crypto/qat/qat_sym_pmd.h b/drivers/crypto/qat/qat_sym_pmd.h
index 80a1987..1e2344c 100644
--- a/drivers/crypto/qat/qat_sym_pmd.h
+++ b/drivers/crypto/qat/qat_sym_pmd.h
@@ -10,7 +10,6 @@
 #include "qat_sym_capabilities.h"
 #include "qat_device.h"
 
-
 /**< Intel(R) QAT Symmetric Crypto PMD device name */
 #define CRYPTODEV_NAME_QAT_SYM_PMD crypto_qat
 #define QAT_SYM_PMD_MAX_NB_SESSIONS2048
@@ -31,10 +30,10 @@ struct qat_sym_dev_private {
/* QAT device symmetric crypto capabilities */
 };
 
-
 int
 qat_sym_dev_create(struct qat_pci_device *qat_pci_dev);
 
 int
 qat_sym_dev_destroy(struct qat_pci_device *qat_pci_dev);
+
 #endif /* _QAT_SYM_PMD_H_ */
-- 
2.7.4

[dpdk-dev] [PATCH v4 3/3] crypto/qat: move common qat files to common dir

[Tomasz Jozwiak]

  -  moved common qat files to common/qat dir.
  -  changed common/qat/Makefile, common/qat/meson.build,
 drivers/Makefile, crypto/Makefile
 to add possibility of using new files locations
  -  added README file into crypto/qat to clarify where
 the build is made from
  -  updated MAINTAINERS file

Signed-off-by: Tomasz Jozwiak 
Acked-by: Fiona Trahe 
---
 MAINTAINERS|  1 +
 drivers/Makefile   |  2 ++
 drivers/common/meson.build |  2 +-
 drivers/{crypto => common}/qat/Makefile| 20 +++-
 drivers/common/qat/meson.build | 14 +++
 .../qat/qat_adf/adf_transport_access_macros.h  |  0
 .../{crypto => common}/qat/qat_adf/icp_qat_fw.h|  0
 .../{crypto => common}/qat/qat_adf/icp_qat_fw_la.h |  0
 .../{crypto => common}/qat/qat_adf/icp_qat_hw.h|  0
 drivers/{crypto => common}/qat/qat_common.c|  0
 drivers/{crypto => common}/qat/qat_common.h|  0
 drivers/{crypto => common}/qat/qat_device.c|  0
 drivers/{crypto => common}/qat/qat_device.h|  0
 drivers/{crypto => common}/qat/qat_logs.c  |  0
 drivers/{crypto => common}/qat/qat_logs.h  |  0
 drivers/{crypto => common}/qat/qat_qp.c|  0
 drivers/{crypto => common}/qat/qat_qp.h|  0
 drivers/crypto/Makefile|  1 -
 drivers/crypto/qat/README  |  8 +++
 drivers/crypto/qat/meson.build | 28 --
 drivers/crypto/qat/qat_sym.h   |  8 +++
 drivers/crypto/qat/qat_sym_pmd.h   |  3 +++
 22 files changed, 67 insertions(+), 20 deletions(-)
 rename drivers/{crypto => common}/qat/Makefile (60%)
 create mode 100644 drivers/common/qat/meson.build
 rename drivers/{crypto => common}/qat/qat_adf/adf_transport_access_macros.h 
(100%)
 rename drivers/{crypto => common}/qat/qat_adf/icp_qat_fw.h (100%)
 rename drivers/{crypto => common}/qat/qat_adf/icp_qat_fw_la.h (100%)
 rename drivers/{crypto => common}/qat/qat_adf/icp_qat_hw.h (100%)
 rename drivers/{crypto => common}/qat/qat_common.c (100%)
 rename drivers/{crypto => common}/qat/qat_common.h (100%)
 rename drivers/{crypto => common}/qat/qat_device.c (100%)
 rename drivers/{crypto => common}/qat/qat_device.h (100%)
 rename drivers/{crypto => common}/qat/qat_logs.c (100%)
 rename drivers/{crypto => common}/qat/qat_logs.h (100%)
 rename drivers/{crypto => common}/qat/qat_qp.c (100%)
 rename drivers/{crypto => common}/qat/qat_qp.h (100%)
 create mode 100644 drivers/crypto/qat/README

diff --git a/MAINTAINERS b/MAINTAINERS
index 3bc928f..bc16078 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -775,6 +775,7 @@ M: John Griffin 
 M: Fiona Trahe 
 M: Deepak Kumar Jain 
 F: drivers/crypto/qat/
+F: drivers/common/qat/
 F: doc/guides/cryptodevs/qat.rst
 F: doc/guides/cryptodevs/features/qat.ini
 
diff --git a/drivers/Makefile b/drivers/Makefile
index c88638c..7566076 100644
--- a/drivers/Makefile
+++ b/drivers/Makefile
@@ -13,6 +13,8 @@ DIRS-$(CONFIG_RTE_LIBRTE_BBDEV) += baseband
 DEPDIRS-baseband := common bus mempool
 DIRS-$(CONFIG_RTE_LIBRTE_CRYPTODEV) += crypto
 DEPDIRS-crypto := common bus mempool
+DIRS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += common/qat
+DEPDIRS-common/qat := bus mempool
 DIRS-$(CONFIG_RTE_LIBRTE_COMPRESSDEV) += compress
 DEPDIRS-compress := bus mempool
 DIRS-$(CONFIG_RTE_LIBRTE_EVENTDEV) += event
diff --git a/drivers/common/meson.build b/drivers/common/meson.build
index 5f6341b..d7b7d8c 100644
--- a/drivers/common/meson.build
+++ b/drivers/common/meson.build
@@ -2,6 +2,6 @@
 # Copyright(c) 2018 Cavium, Inc
 
 std_deps = ['eal']
-drivers = ['octeontx']
+drivers = ['octeontx', 'qat']
 config_flag_fmt = 'RTE_LIBRTE_@0@_COMMON'
 driver_name_fmt = 'rte_common_@0@'
diff --git a/drivers/crypto/qat/Makefile b/drivers/common/qat/Makefile
similarity index 60%
rename from drivers/crypto/qat/Makefile
rename to drivers/common/qat/Makefile
index a939eca..069ac8c 100644
--- a/drivers/crypto/qat/Makefile
+++ b/drivers/common/qat/Makefile
@@ -13,8 +13,13 @@ LIBABIVER := 1
 CFLAGS += $(WERROR_FLAGS)
 CFLAGS += -O3
 
+# build directories
+QAT_CRYPTO_DIR := $(RTE_SDK)/drivers/crypto/qat
+
 # external library include paths
 CFLAGS += -I$(SRCDIR)/qat_adf
+CFLAGS += -I$(SRCDIR)
+CFLAGS += -I$(QAT_CRYPTO_DIR)
 
 # library common source files
 SRCS-y += qat_device.c
@@ -23,11 +28,14 @@ SRCS-y += qat_logs.c
 SRCS-y += qat_qp.c
 
 # library symetric crypto source files
-LDLIBS += -lrte_cryptodev
-LDLIBS += -lcrypto
-SRCS-y += qat_sym.c
-SRCS-y += qat_sym_session.c
-SRCS-y += qat_sym_pmd.c
+ifeq ($(CONFIG_RTE_LIBRTE_CRYPTODEV),y)
+   LDLIBS += -lrte_cryptodev
+   LDLIBS += -lcrypto
+   CFLAGS += -DBUILD_QAT_SYM
+   SRCS-y += $(QAT_CRYPTO_DIR)/qat_sym.c
+   SRCS-y += $(QAT_CRYPTO_DIR)/qat_sym_session.c
+   SRCS-y += $(QAT_CRYPTO_DIR)/qat_sym_pmd.c
+endif
 
 LDLIBS += -lrte_eal 

[dpdk-dev] [PATCH v4 1/3] crypto/qat: add weak functions

[Tomasz Jozwiak]

This patch adds following weak functions to facilitate conditional
compilation of code for those services:
  -  qat_sym_dev_create
  -  qat_asym_dev_create
  -  qat_comp_dev_create
  -  qat_sym_dev_destroy
  -  qat_asym_dev_destroy
  -  qat_comp_dev_destroy
and removes unused files with empty definitions of above functions.

Signed-off-by: Tomasz Jozwiak 
Acked-by: Fiona Trahe 
---
 drivers/crypto/qat/Makefile   |  2 --
 drivers/crypto/qat/meson.build|  4 +---
 drivers/crypto/qat/qat_asym_pmd.c | 17 -
 drivers/crypto/qat/qat_asym_pmd.h | 15 ---
 drivers/crypto/qat/qat_comp_pmd.c | 18 --
 drivers/crypto/qat/qat_comp_pmd.h | 29 -
 drivers/crypto/qat/qat_device.c   | 39 +--
 drivers/crypto/qat/qat_device.h   | 20 
 8 files changed, 58 insertions(+), 86 deletions(-)
 delete mode 100644 drivers/crypto/qat/qat_asym_pmd.c
 delete mode 100644 drivers/crypto/qat/qat_asym_pmd.h
 delete mode 100644 drivers/crypto/qat/qat_comp_pmd.c
 delete mode 100644 drivers/crypto/qat/qat_comp_pmd.h

diff --git a/drivers/crypto/qat/Makefile b/drivers/crypto/qat/Makefile
index ef4a567..64f39fd 100644
--- a/drivers/crypto/qat/Makefile
+++ b/drivers/crypto/qat/Makefile
@@ -28,8 +28,6 @@ SRCS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += qat_sym_session.c
 SRCS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += qat_common.c
 SRCS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += qat_logs.c
 SRCS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += qat_sym_pmd.c
-SRCS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += qat_asym_pmd.c
-SRCS-$(CONFIG_RTE_LIBRTE_PMD_QAT) += qat_comp_pmd.c
 
 # export include files
 SYMLINK-y-include +=
diff --git a/drivers/crypto/qat/meson.build b/drivers/crypto/qat/meson.build
index bcab16e..6d01dac 100644
--- a/drivers/crypto/qat/meson.build
+++ b/drivers/crypto/qat/meson.build
@@ -9,9 +9,7 @@ sources = files('qat_common.c',
'qat_qp.c',
'qat_device.c',
'qat_logs.c',
-   'qat_sym_pmd.c', 'qat_sym.c', 'qat_sym_session.c',
-   'qat_asym_pmd.c',
-   'qat_comp_pmd.c')
+   'qat_sym_pmd.c', 'qat_sym.c', 'qat_sym_session.c')
 includes += include_directories('qat_adf')
 deps += ['bus_pci']
 ext_deps += dep
diff --git a/drivers/crypto/qat/qat_asym_pmd.c 
b/drivers/crypto/qat/qat_asym_pmd.c
deleted file mode 100644
index 8d36300..000
--- a/drivers/crypto/qat/qat_asym_pmd.c
+++ /dev/null
@@ -1,17 +0,0 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright(c) 2018 Intel Corporation
- */
-
-#include "qat_asym_pmd.h"
-
-int
-qat_asym_dev_create(struct qat_pci_device *qat_pci_dev __rte_unused)
-{
-   return 0;
-}
-
-int
-qat_asym_dev_destroy(struct qat_pci_device *qat_pci_dev __rte_unused)
-{
-   return 0;
-}
diff --git a/drivers/crypto/qat/qat_asym_pmd.h 
b/drivers/crypto/qat/qat_asym_pmd.h
deleted file mode 100644
index 0465e03..000
--- a/drivers/crypto/qat/qat_asym_pmd.h
+++ /dev/null
@@ -1,15 +0,0 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright(c) 2018 Intel Corporation
- */
-
-#ifndef _QAT_ASYM_PMD_H_
-#define _QAT_ASYM_PMD_H_
-
-#include "qat_device.h"
-
-int
-qat_asym_dev_create(struct qat_pci_device *qat_pci_dev);
-
-int
-qat_asym_dev_destroy(struct qat_pci_device *qat_pci_dev);
-#endif /* _QAT_ASYM_PMD_H_ */
diff --git a/drivers/crypto/qat/qat_comp_pmd.c 
b/drivers/crypto/qat/qat_comp_pmd.c
deleted file mode 100644
index 547b3db..000
--- a/drivers/crypto/qat/qat_comp_pmd.c
+++ /dev/null
@@ -1,18 +0,0 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright(c) 2018 Intel Corporation
- */
-
-#include "qat_comp_pmd.h"
-
-
-int
-qat_comp_dev_create(struct qat_pci_device *qat_pci_dev __rte_unused)
-{
-   return 0;
-}
-
-int
-qat_comp_dev_destroy(struct qat_pci_device *qat_pci_dev __rte_unused)
-{
-   return 0;
-}
diff --git a/drivers/crypto/qat/qat_comp_pmd.h 
b/drivers/crypto/qat/qat_comp_pmd.h
deleted file mode 100644
index cc31246..000
--- a/drivers/crypto/qat/qat_comp_pmd.h
+++ /dev/null
@@ -1,29 +0,0 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright(c) 2018 Intel Corporation
- */
-
-#ifndef _QAT_COMP_PMD_H_
-#define _QAT_COMP_PMD_H_
-
-#include "qat_device.h"
-
-
-/**< Intel(R) QAT Compression PMD device name */
-#define COMPRESSDEV_NAME_QAT_PMD   comp_qat
-
-
-/** private data structure for a QAT compression device.
- * This QAT device is a device offering only a compression service,
- * there can be one of these on each qat_pci_device (VF).
- */
-struct qat_comp_dev_private {
-   struct qat_pci_device *qat_dev;
-   /**< The qat pci device hosting the service */
-};
-
-int
-qat_comp_dev_create(struct qat_pci_device *qat_pci_dev);
-
-int
-qat_comp_dev_destroy(struct qat_pci_device *qat_pci_dev);
-#endif /* _QAT_COMP_PMD_H_ */
diff --git a/drivers/crypto/qat/qat_device.c b/drivers/crypto/qat/qat_device.c
index 4b97c84..64f236e 100644
--- a/drivers/crypto/qat/qat_device.c
+++ b/drivers/crypto/qat/qat_device

Re: [dpdk-dev] [PATCH 1/2] test/link_bonding: disable HW CRC strip by default.

[Ferruh Yigit]

On 7/2/2018 9:26 AM, dev-boun...@dpdk.org wrote:
> Since the tested virtual device has no capacity of HW CRC offload,
> disable it for the link bonding test by default. Otherwise, it will
> cause test unexpected failure.
> 
> Fixes: b219c8c("test: enable HW CRC strip by default")
> 
> Signed-off-by: Phil Yang 

Hi Phil,

There is a patch to add DEV_RX_OFFLOAD_CRC_STRIP to virtual devices [1] and
there is another under development to remove old offload API [2], those both
looks like can replace your patch when merged.

[1]: [on next-net] d8b0cfa797d8 ("ethdev: add new offload flag to keep CRC")
[2]: https://patches.dpdk.org/patch/40956/

Re: [dpdk-dev] [PATCH] net/mlx5: add support for 32bit systems

[Mordechay Haimovsky]

Inline


> -Original Message-
> From: Shahaf Shuler
> Sent: Monday, July 2, 2018 10:05 AM
> To: Mordechay Haimovsky ; Yongseok Koh
> ; Adrien Mazarguil 
> Cc: dev@dpdk.org; Mordechay Haimovsky 
> Subject: RE: [dpdk-dev] [PATCH] net/mlx5: add support for 32bit systems
> 
> Hi Moty,
> 
> Few nits,
> 
> Also please fix the check patch warning :
> ### net/mlx5: add support for 32bit systems
> 
> CHECK:OPEN_ENDED_LINE: Lines should not end with a '('
> #235: FILE: drivers/net/mlx5/mlx5_rxtx.c:1591:
> +   addr_64 = rte_cpu_to_be_64(
> 
> total: 0 errors, 0 warnings, 1 checks, 311 lines checked
> 
> 
Will Do.
Is there a way to know against which kernel DPDK tools are testing ?
Checkpatch does not show this error when testing with  
4.14.0-0.rc4.git4.1.el7.x86_64 kernel for example.
> 
> Thursday, June 28, 2018 10:13 AM, Moti Haimovsky:
> > Subject: [dpdk-dev] [PATCH] net/mlx5: add support for 32bit systems
> >
> > This patch adds support for building and running mlx5 PMD on 32bit
> > systems such as i686.
> >
> > The main issue to tackle was handling the 32bit access to the UAR as
> > quoted from the mlx5 PRM:
> > QP and CQ DoorBells require 64-bit writes. For best performance, it is
> > recommended to execute the QP/CQ DoorBell as a single 64-bit write
> > operation. For platforms that do not support 64 bit writes, it is
> > possible to issue the 64 bits DoorBells through two consecutive
> > writes, each write 32 bits, as described below:
> > * The order of writing each of the Dwords is from lower to upper
> >   addresses.
> > * No other DoorBell can be rung (or even start ringing) in the midst of
> >   an on-going write of a DoorBell over a given UAR page.
> > The last rule implies that in a multi-threaded environment, the access
> > to a UAR page (which can be accessible by all threads in the process)
> > must be synchronized (for example, using a semaphore) unless an atomic
> > write of 64 bits in a single bus operation is guaranteed. Such a
> > synchronization is not required for when ringing DoorBells on different UAR
> pages.
> >
> > Signed-off-by: Moti Haimovsky 
> > ---
> >  doc/guides/nics/features/mlx5.ini |  1 +
> >  doc/guides/nics/mlx5.rst  | 11 +++
> >  drivers/net/mlx5/mlx5.c   |  8 -
> >  drivers/net/mlx5/mlx5.h   |  5 +++
> >  drivers/net/mlx5/mlx5_defs.h  | 18 --
> >  drivers/net/mlx5/mlx5_rxq.c   |  6 +++-
> >  drivers/net/mlx5/mlx5_rxtx.c  | 22 +++--
> >  drivers/net/mlx5/mlx5_rxtx.h  | 69
> > ++-
> >  drivers/net/mlx5/mlx5_txq.c   | 13 +++-
> >  9 files changed, 137 insertions(+), 16 deletions(-)
> >
> > diff --git a/doc/guides/nics/features/mlx5.ini
> > b/doc/guides/nics/features/mlx5.ini
> > index e75b14b..b28b43e 100644
> > --- a/doc/guides/nics/features/mlx5.ini
> > +++ b/doc/guides/nics/features/mlx5.ini
> > @@ -43,5 +43,6 @@ Multiprocess aware   = Y
> >  Other kdrv   = Y
> >  ARMv8= Y
> >  Power8   = Y
> > +x86-32   = Y
> >  x86-64   = Y
> >  Usage doc= Y
> > diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst index
> > 7dd9c1c..cb9d5d8 100644
> > --- a/doc/guides/nics/mlx5.rst
> > +++ b/doc/guides/nics/mlx5.rst
> > @@ -50,6 +50,8 @@ Features
> >  
> >
> >  - Multi arch support: x86_64, POWER8, ARMv8.
> > +- Support for i686 is available only when working with
> > +  rdma-core version 18.0 or above, built with 32bit support.
> 
> I think we can just add i686 to the supported arch. The limitation on the
> rdma-core version is well documented below.
> 
Will change this

> >  - Multiple TX and RX queues.
> >  - Support for scattered TX and RX frames.
> >  - IPv4, IPv6, TCPv4, TCPv6, UDPv4 and UDPv6 RSS on any number of
> queues.
> > @@ -136,6 +138,11 @@ Limitations
> >enabled (``rxq_cqe_comp_en``) at the same time, RSS hash result is
> > not fully
> >supported. Some Rx packets may not have PKT_RX_RSS_HASH.
> >
> > +- Building for i686 is only supported with:
> > +
> > +  - rdma-core version 18.0 or above built with 32bit support.
> > +  - Kernel version 4.14.41 or above.
> 
> Why the kernel is related? The rdma-core I understand.
> 
There was a patch  added to the kernel that fixed broken 32bit support
f2e9bfac13c9 RDMA/rxe: Fix uABI structure layouts for 32/64 compat
(SHA may have changed)
This patch was added to kernel 4.17 and backported to 4.14.41

> > +
> >  Statistics
> >  --
> >
> > @@ -477,6 +484,10 @@ RMDA Core with Linux Kernel
> >  - Minimal kernel version : v4.14 or the most recent 4.14-rc (see
> > `Linux installation documentation`_)
> >  - Minimal rdma-core version: v15+ commit 0c5f5765213a ("Merge pull
> > request #227 from yishaih/tm")
> >(see `RDMA Core installation documentation`_)
> > +- When building for i686 use:
> > +
> > +  - rdma-core version 18.0 or above built with 32bit support.
> > +  - Kernel version 4.14.41 or abov

Re: [dpdk-dev] [PATCH v7 0/2] app/testpmd implement VXLAN/NVGRE Encap/Decap

[Mohammad Abdul Awal]

On 27/06/2018 12:45, Nelio Laranjeiro wrote:

This series adds an easy and maintainable configuration version support for
those two actions for 18.08 by using global variables in testpmd to store the
necessary information for the tunnel encapsulation.  Those variables are used
in conjunction of RTE_FLOW_ACTION_{VXLAN,NVGRE}_ENCAP action to create easily
the action for flows.

A common way to use it:

  set vxlan ipv4 4 4 4 127.0.0.1 128.0.0.1 11:11:11:11:11:11 22:22:22:22:22:22
  flow create 0 ingress pattern end actions vxlan_encap / queue index 0 / end

  set vxlan ipv6 4 4 4 ::1 :: 11:11:11:11:11:11 22:22:22:22:22:22
  flow create 0 ingress pattern end actions vxlan_encap / queue index 0 / end

  set nvgre ipv4 4 127.0.0.1 128.0.0.1 11:11:11:11:11:11 22:22:22:22:22:22
  flow create 0 ingress pattern end actions nvgre_encap / queue index 0 / end

  set nvgre ipv6 4 ::1 :: 11:11:11:11:11:11 22:22:22:22:22:22
  flow create 0 ingress pattern end actions nvgre_encap / queue index 0 / end

This also replace the proposal done by Mohammad Abdul Awal [1] which handles
in a more complex way for the same work.

Note this API has already a modification planned for 18.11 [2] thus those
series should have a limited life for a single release.

[1] https://dpdk.org/ml/archives/dev/2018-May/101403.html
[2] https://dpdk.org/ml/archives/dev/2018-June/103485.html

Changes in v7:

- add missing documentation added in v5 and removed in v6 by mistake.

Changes in v6:

- fix compilation under redhat 7.5 with gcc 4.8.5 20150623

Changes in v5:

- fix documentation generation.
- add more explanation on how to generate several encapsulated flows.

Changes in v4:

- fix big endian issue on vni and tni.
- add samples to the documentation.
- set the VXLAN UDP source port to 0 by default to let the driver generate it
   from the inner hash as described in the RFC 7348.
- use default rte flow mask for each item.

Changes in v3:

- support VLAN in the outer encapsulation.
- fix the documentation with missing arguments.

Changes in v2:

- add default IPv6 values for NVGRE encapsulation.
- replace VXLAN to NVGRE in comments concerning NVGRE layer.

Nelio Laranjeiro (2):
   app/testpmd: add VXLAN encap/decap support
   app/testpmd: add NVGRE encap/decap support

  app/test-pmd/cmdline.c  | 252 ++
  app/test-pmd/cmdline_flow.c | 274 
  app/test-pmd/testpmd.c  |  32 +++
  app/test-pmd/testpmd.h  |  32 +++
  doc/guides/testpmd_app_ug/testpmd_funcs.rst |  82 ++
  5 files changed, 672 insertions(+)



Hi,

I have one concern in terms of usability though.
In testpmd, the rte_flow command line options have auto-completion with 
" " format which make using the command very 
much user friendly.


For the command "set vxlan ipv4 4 4 4 127.0.0.1 128.0.0.1 
11:11:11:11:11:11 22:22:22:22:22:22", it does not look much user 
friendly to me. A user may easily lose track of sequence of 9 param 
items. It would be much user friendly if the options would be like below 
and has auto-completion.


set vxlan ip_ver  vni  udp_src  
udp-dst  ip_src  ip_dst  
eth_src  eth_dst 


This way an user may never feel confused. Can maintainers comment on 
this point please?


Regards,
Awal.

Re: [dpdk-dev] [PATCH v3 09/16] cryptodev: remove max number of sessions parameter

[De Lara Guarch, Pablo]

Hi 

> -Original Message-
> From: Tomasz Duszynski [mailto:t...@semihalf.com]
> Sent: Friday, June 29, 2018 9:04 AM
> To: De Lara Guarch, Pablo 
> Cc: Doherty, Declan ; akhil.go...@nxp.com;
> shally.ve...@caviumnetworks.com; ravi1.ku...@amd.com;
> jerin.ja...@caviumnetworks.com; Zhang, Roy Fan ;
> Trahe, Fiona ; t...@semihalf.com;
> jianjay.z...@huawei.com; dev@dpdk.org
> Subject: Re: [PATCH v3 09/16] cryptodev: remove max number of sessions
> parameter
> 
> Hello Pablo,
> 
> On Thu, Jun 28, 2018 at 01:52:57AM +0100, Pablo de Lara wrote:
> > Most crypto PMDs do not have a limitation of the number of the
> > sessions that can be handled internally. The value that was set before
> > was not actually used at all, since the sessions are created at the
> > application level.
> > Therefore, this value is not parsed from the initial crypto parameters
> > anymore and it is set to 0, meaning that there is no actual limit.
> >
> 
> [...]
> 
> > diff --git a/drivers/crypto/mvsam/rte_mrvl_pmd_ops.c
> > b/drivers/crypto/mvsam/rte_mrvl_pmd_ops.c
> > index 3f8de37b7..07850098b 100644
> > --- a/drivers/crypto/mvsam/rte_mrvl_pmd_ops.c
> > +++ b/drivers/crypto/mvsam/rte_mrvl_pmd_ops.c
> > @@ -471,7 +471,8 @@ mrvl_crypto_pmd_info_get(struct rte_cryptodev *dev,
> > dev_info->feature_flags = dev->feature_flags;
> > dev_info->capabilities = mrvl_crypto_pmd_capabilities;
> > dev_info->max_nb_queue_pairs = internals->max_nb_qpairs;
> > -   dev_info->sym.max_nb_sessions = internals->max_nb_sessions;
> > +   /* No limit of number of sessions */
> > +   dev_info->sym.max_nb_sessions = 0;
> 
> In this case we want to set a limit, right?
> 

Hi Tomasz,

Yes, you are right, I made a mistake here. Thanks for spotting it :)

Will send another version with this fixed.

Pablo

[dpdk-dev] [PATCH v2] net/mlx5: add support for 32bit systems

[Moti Haimovsky]

This patch adds support for building and running mlx5 PMD on
32bit systems such as i686.

The main issue to tackle was handling the 32bit access to the UAR
as quoted from the mlx5 PRM:
QP and CQ DoorBells require 64-bit writes. For best performance, it
is recommended to execute the QP/CQ DoorBell as a single 64-bit write
operation. For platforms that do not support 64 bit writes, it is
possible to issue the 64 bits DoorBells through two consecutive writes,
each write 32 bits, as described below:
* The order of writing each of the Dwords is from lower to upper
  addresses.
* No other DoorBell can be rung (or even start ringing) in the midst of
  an on-going write of a DoorBell over a given UAR page.
The last rule implies that in a multi-threaded environment, the access
to a UAR page (which can be accessible by all threads in the process)
must be synchronized (for example, using a semaphore) unless an atomic
write of 64 bits in a single bus operation is guaranteed. Such a
synchronization is not required for when ringing DoorBells on different
UAR pages.

Signed-off-by: Moti Haimovsky 
---
v2:
* Fixed coding style issues.
* Modified documentation according to review inputs.
* Fixed merge conflicts.
---
 doc/guides/nics/features/mlx5.ini |  1 +
 doc/guides/nics/mlx5.rst  |  6 +++-
 drivers/net/mlx5/mlx5.c   |  8 -
 drivers/net/mlx5/mlx5.h   |  5 +++
 drivers/net/mlx5/mlx5_defs.h  | 18 --
 drivers/net/mlx5/mlx5_rxq.c   |  6 +++-
 drivers/net/mlx5/mlx5_rxtx.c  | 22 +++--
 drivers/net/mlx5/mlx5_rxtx.h  | 69 ++-
 drivers/net/mlx5/mlx5_txq.c   | 13 +++-
 9 files changed, 131 insertions(+), 17 deletions(-)

diff --git a/doc/guides/nics/features/mlx5.ini 
b/doc/guides/nics/features/mlx5.ini
index e75b14b..b28b43e 100644
--- a/doc/guides/nics/features/mlx5.ini
+++ b/doc/guides/nics/features/mlx5.ini
@@ -43,5 +43,6 @@ Multiprocess aware   = Y
 Other kdrv   = Y
 ARMv8= Y
 Power8   = Y
+x86-32   = Y
 x86-64   = Y
 Usage doc= Y
diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst
index 7dd9c1c..5fbad60 100644
--- a/doc/guides/nics/mlx5.rst
+++ b/doc/guides/nics/mlx5.rst
@@ -49,7 +49,7 @@ libibverbs.
 Features
 
 
-- Multi arch support: x86_64, POWER8, ARMv8.
+- Multi arch support: x86_64, POWER8, ARMv8, i686.
 - Multiple TX and RX queues.
 - Support for scattered TX and RX frames.
 - IPv4, IPv6, TCPv4, TCPv6, UDPv4 and UDPv6 RSS on any number of queues.
@@ -477,6 +477,10 @@ RMDA Core with Linux Kernel
 - Minimal kernel version : v4.14 or the most recent 4.14-rc (see `Linux 
installation documentation`_)
 - Minimal rdma-core version: v15+ commit 0c5f5765213a ("Merge pull request 
#227 from yishaih/tm")
   (see `RDMA Core installation documentation`_)
+- When building for i686 use:
+
+  - rdma-core version 18.0 or above built with 32bit support.
+  - Kernel version 4.14.41 or above.
 
 .. _`Linux installation documentation`: 
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git/plain/Documentation/admin-guide/README.rst
 .. _`RDMA Core installation documentation`: 
https://raw.githubusercontent.com/linux-rdma/rdma-core/master/README.md
diff --git a/drivers/net/mlx5/mlx5.c b/drivers/net/mlx5/mlx5.c
index f0e6ed7..5d0f706 100644
--- a/drivers/net/mlx5/mlx5.c
+++ b/drivers/net/mlx5/mlx5.c
@@ -567,7 +567,7 @@
rte_memseg_walk(find_lower_va_bound, &addr);
 
/* keep distance to hugepages to minimize potential conflicts. */
-   addr = RTE_PTR_SUB(addr, MLX5_UAR_OFFSET + MLX5_UAR_SIZE);
+   addr = RTE_PTR_SUB(addr, (uintptr_t)(MLX5_UAR_OFFSET + MLX5_UAR_SIZE));
/* anonymous mmap, no real memory consumption. */
addr = mmap(addr, MLX5_UAR_SIZE,
PROT_NONE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
@@ -953,6 +953,12 @@
priv->port = port;
priv->pd = pd;
priv->mtu = ETHER_MTU;
+#ifndef RTE_ARCH_64
+   /* Initialize UAR access locks for 32bit implementations. */
+   rte_spinlock_init(&priv->uar_lock_cq);
+   for (i = 0; i < MLX5_UAR_PAGE_NUM_MAX; i++)
+   rte_spinlock_init(&priv->uar_lock[i]);
+#endif
err = mlx5_args(&config, pci_dev->device.devargs);
if (err) {
err = rte_errno;
diff --git a/drivers/net/mlx5/mlx5.h b/drivers/net/mlx5/mlx5.h
index 997b04a..2da32cd 100644
--- a/drivers/net/mlx5/mlx5.h
+++ b/drivers/net/mlx5/mlx5.h
@@ -198,6 +198,11 @@ struct priv {
/* Context for Verbs allocator. */
int nl_socket; /* Netlink socket. */
uint32_t nl_sn; /* Netlink message sequence number. */
+#ifndef RTE_ARCH_64
+   rte_spinlock_t uar_lock_cq; /* CQs share a common distinct UAR */
+   rte_spinlock_t uar_lock[MLX5_UAR_PAGE_NUM_MAX];
+   /* UAR same-page access control required in 32bit i

Re: [dpdk-dev] [PATCH v5 8/9] net/virtio: add in-order Rx/Tx into selection

[Maxime Coquelin]

On 07/02/2018 03:56 PM, Marvin Liu wrote:

After IN_ORDER Rx/Tx paths added, need to update Rx/Tx path selection
logic.

Rx path select logic: If IN_ORDER and merge-able are enabled will select
IN_ORDER Rx path. If IN_ORDER is enabled, Rx offload and merge-able are
disabled will select simple Rx path. Otherwise will select normal Rx
path.

Tx path select logic: If IN_ORDER is enabled will select IN_ORDER Tx
path. Otherwise will select default Tx path.

Signed-off-by: Marvin Liu 
Reviewed-by: Maxime Coquelin 

diff --git a/doc/guides/nics/virtio.rst b/doc/guides/nics/virtio.rst
index 46e292c4d..7c099fb7c 100644
--- a/doc/guides/nics/virtio.rst
+++ b/doc/guides/nics/virtio.rst
@@ -201,7 +201,7 @@ The packet transmission flow is:
  Virtio PMD Rx/Tx Callbacks
  --
  
-Virtio driver has 3 Rx callbacks and 2 Tx callbacks.

+Virtio driver has 4 Rx callbacks and 3 Tx callbacks.
  
  Rx callbacks:
  
@@ -215,6 +215,9 @@ Rx callbacks:

 Vector version without mergeable Rx buffer support, also fixes the 
available
 ring indexes and uses vector instructions to optimize performance.
  
+#. ``virtio_recv_mergeable_pkts_inorder``:

+   In-order version with mergeable Rx buffer support.
+
  Tx callbacks:
  
  #. ``virtio_xmit_pkts``:

@@ -223,6 +226,8 @@ Tx callbacks:
  #. ``virtio_xmit_pkts_simple``:
 Vector version fixes the available ring indexes to optimize performance.
  
+#. ``virtio_xmit_pkts_inorder``:

+   In-order version.
  
  By default, the non-vector callbacks are used:
  
@@ -254,6 +259,12 @@ Example of using the vector version of the virtio poll mode driver in
  
 testpmd -l 0-2 -n 4 -- -i --tx-offloads=0x0 --rxq=1 --txq=1 --nb-cores=1
  
+In-order callbacks only work on simulated virtio user vdev.

+
+*   For Rx: If mergeable Rx buffers is enabled and in-order is enabled then
+``virtio_xmit_pkts_inorder`` is used.
+
+*   For Tx: If in-order is enabled then ``virtio_xmit_pkts_inorder`` is used.
  
  Interrupt mode

  --
diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index df50a571a..df7981ddb 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1320,6 +1320,11 @@ set_rxtx_funcs(struct rte_eth_dev *eth_dev)
PMD_INIT_LOG(INFO, "virtio: using simple Rx path on port %u",
eth_dev->data->port_id);
eth_dev->rx_pkt_burst = virtio_recv_pkts_vec;
+   } else if (hw->use_inorder_rx) {
+   PMD_INIT_LOG(INFO,
+   "virtio: using inorder mergeable buffer Rx path on port 
%u",
+   eth_dev->data->port_id);
+   eth_dev->rx_pkt_burst = &virtio_recv_mergeable_pkts_inorder;
} else if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) {
PMD_INIT_LOG(INFO,
"virtio: using mergeable buffer Rx path on port %u",
@@ -1335,6 +1340,10 @@ set_rxtx_funcs(struct rte_eth_dev *eth_dev)
PMD_INIT_LOG(INFO, "virtio: using simple Tx path on port %u",
eth_dev->data->port_id);
eth_dev->tx_pkt_burst = virtio_xmit_pkts_simple;
+   } else if (hw->use_inorder_tx) {
+   PMD_INIT_LOG(INFO, "virtio: using inorder Tx path on port %u",
+   eth_dev->data->port_id);
+   eth_dev->tx_pkt_burst = virtio_xmit_pkts_inorder;
} else {
PMD_INIT_LOG(INFO, "virtio: using standard Tx path on port %u",
eth_dev->data->port_id);
@@ -1874,20 +1883,27 @@ virtio_dev_configure(struct rte_eth_dev *dev)
hw->use_simple_rx = 1;
hw->use_simple_tx = 1;
  
+	if (vtpci_with_feature(hw, VIRTIO_F_IN_ORDER)) {

+   /* Simple Tx not compatible with in-order ring */
+   hw->use_inorder_tx = 1;
+   hw->use_simple_tx = 0;
+   if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) {
+   hw->use_inorder_rx = 1;
+   hw->use_simple_rx = 0;
+   } else {
+   hw->use_inorder_rx = 0;
+   if (rx_offloads & (DEV_RX_OFFLOAD_UDP_CKSUM |
+  DEV_RX_OFFLOAD_TCP_CKSUM))
+   hw->use_simple_rx = 0;

It is applied, but I think this is still not good.

Simple Rx is set to 1 by default, so if IN_ORDER isn't negotiated,
and UDP/TCP_CSUM is enabled, simple Rx keeps being selected.

I'll fix it in my series that I'm doing on top.

Regards,
Maxime


+   }
+   }
+
  #if defined RTE_ARCH_ARM64 || defined RTE_ARCH_ARM
if (!rte_cpu_get_flag_enabled(RTE_CPUFLAG_NEON)) {
hw->use_simple_rx = 0;
hw->use_simple_tx = 0;
}
  #endif
-   if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) {
-   hw->use_simple_rx = 0;
-   hw->use_simple_tx = 0;
-   }
-
-   if (r

Re: [dpdk-dev] [RFC 0/3] ethdev: add IP address and TCP/UDP port rewrite actions to flow API

[Rahul Lakkireddy]

Hi Jack,

On Monday, July 07/02/18, 2018 at 14:39:06 +0530, Jack Min wrote:
> Hey,
> 
> I've one question: 
> 
> Do those APIs support modify inner IP headers? 
> 

I'm afraid not. These API only modify outer IP addresses.

I can think of two ways to achieve this:

1. Enhance these API to accept an additional flag to indicate
   inner or outer. For example, to set inner IPv4 address, we can
   set inner to 1 in following structure.

   struct rte_flow_action_of_set_nw_ipv4 {
   uint32_t inner:1; /* 1 - Inner, 0 - Outer */
   uint32_t reserved:31;
   uint32_t ipv4_addr;
   };

2. Add new actions to indicate inner IP addresses; i.e.

   RTE_FLOW_ACTION_TYPE_OF_SET_INNER_NW_IPV4_SRC
   RTE_FLOW_ACTION_TYPE_OF_SET_INNER_NW_IPV4_DST
   RTE_FLOW_ACTION_TYPE_OF_SET_INNER_NW_IPV6_SRC
   RTE_FLOW_ACTION_TYPE_OF_SET_INNER_NW_IPV6_DST

Thanks,
Rahul

> In our scenarios, we need to specific that  NAT action is on inner or outer 
> IP header.
> 
> Thanks,
> -Jack
> 
> > -Original Message-
> > From: dev  On Behalf Of Rahul Lakkireddy
> > Sent: June 22, 2018 17:56
> > To: dev@dpdk.org
> > Cc: shag...@chelsio.com; indra...@chelsio.com; nirran...@chelsio.com
> > Subject: [dpdk-dev] [RFC 0/3] ethdev: add IP address and TCP/UDP port 
> > rewrite
> > actions to flow API
> > 
> > This series of patches add support for actions:
> > - OF_SET_NW_IPV4_SRC - set a new IPv4 source address.
> > - OF_SET_NW_IPV4_DST - set a new IPv4 destination address.
> > - OF_SET_NW_IPV6_SRC - set a new IPv6 source address.
> > - OF_SET_NW_IPV6_DST - set a new IPv6 destination address.
> > - OF_SET_TP_SRC - set a new TCP/UDP source port number.
> > - OF_SET_TP_DST - set a new TCP/UDP destination port number.
> > 
> > These actions are useful in Network Address Translation use case to edit IP
> > address and TCP/UDP port numbers before switching the packets out to the
> > destination device port.
> > 
> > The IP address and TCP/UDP port rewrite actions are based on
> > OFPAT_SET_NW_SRC, OFPAT_SET_NW_DST, OFPAT_SET_TP_SRC, and
> > OFPAT_SET_TP_DST actions from OpenFlow Specification [1].
> > 
> > Patch 1 adds support for IP address rewrite to rte_flow and testpmd.
> > 
> > Patch 2 adds support for TCP/UDP port rewrite to rte_flow and testpmd.
> > 
> > Patch 3 shows CXGBE PMD example to offload these actions to hardware.
> > 
> > Feedback and suggestions will be much appreciated.
> > 
> > Thanks,
> > Rahul
> > 
> > [1]
> > https://emea01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.open
> > networking.org%2Fimages%2Fopenflow-switch-
> > v1.5.1.pdf&data=02%7C01%7Cxuemingl%40mellanox.com%7C74c8ce3b79954b2f
> > 123208d5d8268617%7Ca652971c7d2e4d9ba6a4d149256f461b%7C0%7C1%7C636
> > 652582334142929&sdata=uoFc1KjVTkGI9fdaPNXvXYRmh%2F6HLMKT79IwgvDSH
> > 0k%3D&reserved=0
> > 
> > Shagun Agrawal (3):
> >   ethdev: add flow api actions to modify IP addresses
> >   ethdev: add flow api actions to modify TCP/UDP port numbers
> >   net/cxgbe: add flow actions to modify IP and TCP/UDP port address
> > 
> >  app/test-pmd/cmdline_flow.c | 150
> > 
> >  app/test-pmd/config.c   |  12 +++
> >  doc/guides/prog_guide/rte_flow.rst  | 102 +++
> >  doc/guides/testpmd_app_ug/testpmd_funcs.rst |  24 +
> >  drivers/net/cxgbe/base/common.h |   1 +
> >  drivers/net/cxgbe/base/t4fw_interface.h |  50 ++
> >  drivers/net/cxgbe/cxgbe_filter.c|  23 -
> >  drivers/net/cxgbe/cxgbe_filter.h|  26 -
> >  drivers/net/cxgbe/cxgbe_flow.c  | 127 +++
> >  drivers/net/cxgbe/cxgbe_main.c  |  10 ++
> >  lib/librte_ethdev/rte_flow.c|  12 +++
> >  lib/librte_ethdev/rte_flow.h|  92 +
> >  12 files changed, 625 insertions(+), 4 deletions(-)
> > 
> > --
> > 2.14.1
> 

Re: [dpdk-dev] Compilation of MLX5 driver

[Asaf Sinai]

Hi Shahaf,

I face similar problem with DPDK 18.05, where missing file is "mlx5dv.h":

== Build drivers/net/mlx5
  CC mlx5.o
In file included from 
/net/emcradware/CM_Users/asafsi/work/drivers/dpdk/dpdk-18.05/drivers/net/mlx5/mlx5.h:35:0,
 from 
/net/emcradware/CM_Users/asafsi/work/drivers/dpdk/dpdk-18.05/drivers/net/mlx5/mlx5.c:40:

/net/emcradware/CM_Users/asafsi/work/drivers/dpdk/dpdk-18.05/drivers/net/mlx5/mlx5_mr.h:19:31:
 fatal error: infiniband/mlx5dv.h: No such file ordirectory
 #include 
   ^
compilation terminated.

Can you please advise?

Thanks,
Asaf

-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Nitin Katiyar
Sent: Thursday, May 31, 2018 02:44 PM
To: Nélio Laranjeiro 
Cc: Shahaf Shuler ; dev@dpdk.org
Subject: Re: [dpdk-dev] Compilation of MLX5 driver

Thanks Shahaf, it worked after removing the options you specified.

Regards,
Nitin

-Original Message-
From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com] 
Sent: Thursday, May 31, 2018 3:23 PM
To: Nitin Katiyar 
Cc: Shahaf Shuler ; dev@dpdk.org
Subject: Re: [dpdk-dev] Compilation of MLX5 driver

On Thu, May 31, 2018 at 09:14:03AM +, Nitin Katiyar wrote:
> Yes,I installed it using --dpdk --upstream-libs. What is the way 
> forward now?

In v17.05 MLX5 PMD is still relying on libibverbs and libmlx5, the way Those 
options you used are necessary to select in their package the installation of 
libverbs,libmlx5 or rdma-core.
Doing this you have selected rdma-core which is not supported in v17.05 DPDK 
version.

You need to install Mellanox OFED without those two options to select 
libibverbs, libmlx5 to make it work.

Regards,
 
> Regards,
> Nitin
> 
> -Original Message-
> From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> Sent: Thursday, May 31, 2018 1:36 PM
> To: Nitin Katiyar 
> Cc: Shahaf Shuler ; dev@dpdk.org
> Subject: Re: [dpdk-dev] Compilation of MLX5 driver
> 
> On Thu, May 31, 2018 at 07:01:17AM +, Nitin Katiyar wrote:
> > Hi,
> > It has following files:
> > 
> > arch.h  ib.h  kern-abi.h  mlx4dv.h  mlx5dv.h  opcode.h  sa.h 
> > sa-kern-abi.h  verbs.h
> > 
> > I tried with both MLNX_OFED_LINUX-4.2-1.0.0.0 and
> > MLNX_OFED_LINUX-4.2-1.2.0.0-ubuntu14.04-x86_64
> 
> Did you installed Mellanox OFED with the --dpdk --upstream-libs arguments for 
> the installation script?
> 
> If it is the case, you should not add them for this version, those options 
> are for DPDK v17.11 and higher.
> 
> Regards,
> 
> > Regards,
> > Nitin
> > 
> > -Original Message-
> > From: Shahaf Shuler [mailto:shah...@mellanox.com]
> > Sent: Thursday, May 31, 2018 10:51 AM
> > To: Nitin Katiyar ; Nélio Laranjeiro 
> > 
> > Cc: dev@dpdk.org
> > Subject: RE: [dpdk-dev] Compilation of MLX5 driver
> > 
> > Wednesday, May 30, 2018 7:45 PM, Nitin Katiyar:
> > > 
> > > Hi,
> > > I was compiling 17.05.02.
> > > Regards,
> > > Nitin
> > > 
> > > -Original Message-
> > > From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> > > Sent: Wednesday, May 30, 2018 6:42 PM
> > > To: Nitin Katiyar 
> > > Cc: dev@dpdk.org
> > > Subject: Re: [dpdk-dev] Compilation of MLX5 driver
> > > 
> > > Hi,
> > > 
> > > On Wed, May 30, 2018 at 11:54:31AM +, Nitin Katiyar wrote:
> > > > Hi,
> > > > I am trying to compile MLX5 PMD driver by setting
> > > "CONFIG_RTE_LIBRTE_MLX5_PMD=y" and hitting following compilation 
> > > error.
> > > >
> > > > fatal error: infiniband/mlx5_hw.h: No such file or directory
> > 
> > Can you list the files you have under /usr/include/infiniband ? 
> > 
> > > >
> > > > I have installed MLNX_OFED _LINUX-4.2-1.2.0 on my Ubuntu 14.04 
> > > > machine
> > > but still hitting the same error. Am I missing some other package?
> > > 
> > > Which version of DPDK are you using (it is important to help)?
> > > 
> > > Regards,
> > > 
> > > --
> > > Nélio Laranjeiro
> > > 6WIND
> 
> --
> Nélio Laranjeiro
> 6WIND

--
Nélio Laranjeiro
6WIND

Re: [dpdk-dev] [PATCH v2 10/23] net/bnxt: move function check zero bytes to bnxt util.h

[Ferruh Yigit]

On 6/28/2018 9:15 PM, Ajit Khaparde wrote:
> From: Scott Branden 
> 
> Move check_zero_bytes into new bnxt_util.h file.
> 
> Signed-off-by: Scott Branden 
> Reviewed-by: Ajit Khaparde 
> ---
>  drivers/net/bnxt/Makefile  |  1 +
>  drivers/net/bnxt/bnxt_ethdev.c |  1 +
>  drivers/net/bnxt/bnxt_filter.c |  9 -
>  drivers/net/bnxt/bnxt_filter.h |  1 -
>  drivers/net/bnxt/bnxt_util.c   | 18 ++
>  drivers/net/bnxt/bnxt_util.h   | 11 +++
>  6 files changed, 31 insertions(+), 10 deletions(-)
>  create mode 100644 drivers/net/bnxt/bnxt_util.c
>  create mode 100644 drivers/net/bnxt/bnxt_util.h
> 
> diff --git a/drivers/net/bnxt/Makefile b/drivers/net/bnxt/Makefile
> index fd0cb5235..80db03ea8 100644
> --- a/drivers/net/bnxt/Makefile
> +++ b/drivers/net/bnxt/Makefile
> @@ -38,6 +38,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_txq.c
>  SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_txr.c
>  SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_vnic.c
>  SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_irq.c
> +SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_util.c

This breaks the meson build and similar change required for meson, same with
bnxt_flow.c in other patch, if there is no other issue I can fix this while 
merging.

[dpdk-dev] [PATCH v2] cryptodev: remove RTE_LIBRTE_CRYPTODEV_DEBUG

[Jananee Parthasarathy]

From: Jananee Parthasarathy  

For librte_cryptodev dynamic logging, conditional compilation of
debug logs would not be required anymore.

Signed-off-by: Jananee Parthasarathy  
Reviewed-by: Reshma Pattan 
Reviewed-by: Pablo de Lara Guarch 
---
v2: unused macro removed from config/common_base and rte_dev.h
---

 config/common_base  | 1 -
 lib/librte_cryptodev/rte_cryptodev.h| 8 
 lib/librte_eal/common/include/rte_dev.h | 1 -
 3 files changed, 10 deletions(-)

diff --git a/config/common_base b/config/common_base
index 6541ad5b2..84317a9bd 100644
--- a/config/common_base
+++ b/config/common_base
@@ -460,7 +460,6 @@ CONFIG_RTE_LIBRTE_PMD_BBDEV_TURBO_SW=n
 # Compile generic crypto device library
 #
 CONFIG_RTE_LIBRTE_CRYPTODEV=y
-CONFIG_RTE_LIBRTE_CRYPTODEV_DEBUG=n
 CONFIG_RTE_CRYPTO_MAX_DEVS=64
 
 #
diff --git a/lib/librte_cryptodev/rte_cryptodev.h 
b/lib/librte_cryptodev/rte_cryptodev.h
index b7a4cf0a2..ccc0f73fd 100644
--- a/lib/librte_cryptodev/rte_cryptodev.h
+++ b/lib/librte_cryptodev/rte_cryptodev.h
@@ -38,7 +38,6 @@ extern const char **rte_cyptodev_names;
RTE_FMT(RTE_FMT_HEAD(__VA_ARGS__,) "\n", \
RTE_FMT_TAIL(__VA_ARGS__,)))
 
-#ifdef RTE_LIBRTE_CRYPTODEV_DEBUG
 #define CDEV_LOG_DEBUG(...) \
RTE_LOG(DEBUG, CRYPTODEV, \
RTE_FMT("%s() line %u: " RTE_FMT_HEAD(__VA_ARGS__,) "\n", \
@@ -49,13 +48,6 @@ extern const char **rte_cyptodev_names;
RTE_FMT("[%s] %s: " RTE_FMT_HEAD(__VA_ARGS__,) "\n", \
dev, __func__, RTE_FMT_TAIL(__VA_ARGS__,)))
 
-#else
-#define CDEV_LOG_DEBUG(...) (void)0
-#define CDEV_PMD_TRACE(...) (void)0
-#endif
-
-
-
 /**
  * A macro that points to an offset from the start
  * of the crypto operation structure (rte_crypto_op)
diff --git a/lib/librte_eal/common/include/rte_dev.h 
b/lib/librte_eal/common/include/rte_dev.h
index 3879ff3ca..d4b43512c 100644
--- a/lib/librte_eal/common/include/rte_dev.h
+++ b/lib/librte_eal/common/include/rte_dev.h
@@ -70,7 +70,6 @@ rte_pmd_debug_trace(const char *func_name, const char *fmt, 
...)
  * RTE_*_RET() macros defined below is compiled in debug mode.
  */
 #if defined(RTE_LIBRTE_ETHDEV_DEBUG) || \
-   defined(RTE_LIBRTE_CRYPTODEV_DEBUG) || \
defined(RTE_LIBRTE_EVENTDEV_DEBUG)
 #define RTE_PMD_DEBUG_TRACE(...) \
rte_pmd_debug_trace(__func__, __VA_ARGS__)
-- 
2.16.3

Re: [dpdk-dev] [PATCH v5 8/9] net/virtio: add in-order Rx/Tx into selection

[Maxime Coquelin]

On 07/02/2018 01:24 PM, Maxime Coquelin wrote:



On 07/02/2018 03:56 PM, Marvin Liu wrote:

After IN_ORDER Rx/Tx paths added, need to update Rx/Tx path selection
logic.

Rx path select logic: If IN_ORDER and merge-able are enabled will select
IN_ORDER Rx path. If IN_ORDER is enabled, Rx offload and merge-able are
disabled will select simple Rx path. Otherwise will select normal Rx
path.

Tx path select logic: If IN_ORDER is enabled will select IN_ORDER Tx
path. Otherwise will select default Tx path.

Signed-off-by: Marvin Liu 
Reviewed-by: Maxime Coquelin 

diff --git a/doc/guides/nics/virtio.rst b/doc/guides/nics/virtio.rst
index 46e292c4d..7c099fb7c 100644
--- a/doc/guides/nics/virtio.rst
+++ b/doc/guides/nics/virtio.rst
@@ -201,7 +201,7 @@ The packet transmission flow is:
  Virtio PMD Rx/Tx Callbacks
  --
-Virtio driver has 3 Rx callbacks and 2 Tx callbacks.
+Virtio driver has 4 Rx callbacks and 3 Tx callbacks.
  Rx callbacks:
@@ -215,6 +215,9 @@ Rx callbacks:
 Vector version without mergeable Rx buffer support, also fixes 
the available

 ring indexes and uses vector instructions to optimize performance.
+#. ``virtio_recv_mergeable_pkts_inorder``:
+   In-order version with mergeable Rx buffer support.
+
  Tx callbacks:
  #. ``virtio_xmit_pkts``:
@@ -223,6 +226,8 @@ Tx callbacks:
  #. ``virtio_xmit_pkts_simple``:
 Vector version fixes the available ring indexes to optimize 
performance.

+#. ``virtio_xmit_pkts_inorder``:
+   In-order version.
  By default, the non-vector callbacks are used:
@@ -254,6 +259,12 @@ Example of using the vector version of the virtio 
poll mode driver in
 testpmd -l 0-2 -n 4 -- -i --tx-offloads=0x0 --rxq=1 --txq=1 
--nb-cores=1

+In-order callbacks only work on simulated virtio user vdev.
+
+*   For Rx: If mergeable Rx buffers is enabled and in-order is 
enabled then

+    ``virtio_xmit_pkts_inorder`` is used.
+
+*   For Tx: If in-order is enabled then ``virtio_xmit_pkts_inorder`` 
is used.

  Interrupt mode
  --
diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c

index df50a571a..df7981ddb 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1320,6 +1320,11 @@ set_rxtx_funcs(struct rte_eth_dev *eth_dev)
  PMD_INIT_LOG(INFO, "virtio: using simple Rx path on port %u",
  eth_dev->data->port_id);
  eth_dev->rx_pkt_burst = virtio_recv_pkts_vec;
+    } else if (hw->use_inorder_rx) {
+    PMD_INIT_LOG(INFO,
+    "virtio: using inorder mergeable buffer Rx path on port %u",
+    eth_dev->data->port_id);
+    eth_dev->rx_pkt_burst = &virtio_recv_mergeable_pkts_inorder;
  } else if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) {
  PMD_INIT_LOG(INFO,
  "virtio: using mergeable buffer Rx path on port %u",
@@ -1335,6 +1340,10 @@ set_rxtx_funcs(struct rte_eth_dev *eth_dev)
  PMD_INIT_LOG(INFO, "virtio: using simple Tx path on port %u",
  eth_dev->data->port_id);
  eth_dev->tx_pkt_burst = virtio_xmit_pkts_simple;
+    } else if (hw->use_inorder_tx) {
+    PMD_INIT_LOG(INFO, "virtio: using inorder Tx path on port %u",
+    eth_dev->data->port_id);
+    eth_dev->tx_pkt_burst = virtio_xmit_pkts_inorder;
  } else {
  PMD_INIT_LOG(INFO, "virtio: using standard Tx path on port %u",
  eth_dev->data->port_id);
@@ -1874,20 +1883,27 @@ virtio_dev_configure(struct rte_eth_dev *dev)
  hw->use_simple_rx = 1;
  hw->use_simple_tx = 1;
+    if (vtpci_with_feature(hw, VIRTIO_F_IN_ORDER)) {
+    /* Simple Tx not compatible with in-order ring */
+    hw->use_inorder_tx = 1;
+    hw->use_simple_tx = 0;
+    if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) {
+    hw->use_inorder_rx = 1;
+    hw->use_simple_rx = 0;
+    } else {
+    hw->use_inorder_rx = 0;
+    if (rx_offloads & (DEV_RX_OFFLOAD_UDP_CKSUM |
+   DEV_RX_OFFLOAD_TCP_CKSUM))
+    hw->use_simple_rx = 0;

It is applied, but I think this is still not good.

Simple Rx is set to 1 by default, so if IN_ORDER isn't negotiated,
and UDP/TCP_CSUM is enabled, simple Rx keeps being selected.

I'll fix it in my series that I'm doing on top.


Actually, after discussion with Ferruh, I fixed it directly in the patch.

Thanks,
Maxime


Regards,
Maxime


+    }
+    }
+
  #if defined RTE_ARCH_ARM64 || defined RTE_ARCH_ARM
  if (!rte_cpu_get_flag_enabled(RTE_CPUFLAG_NEON)) {
  hw->use_simple_rx = 0;
  hw->use_simple_tx = 0;
  }
  #endif
-    if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) {
-    hw->use_simple_rx = 0;
-    hw->use_simple_tx = 0;
-    }
-
-    if (rx_offloads & (DEV_RX_OFFLOAD_UDP_CKSUM |
-   DEV_RX_OFFLOAD_TCP_CKSUM))
-    hw->use_simple_rx = 0;
  return 0;
  }

Re: [dpdk-dev] Compilation of MLX5 driver

[Asaf Sinai]

Additional details:

Ubuntu: 12.04.3
Linux: 3.2.0-53-virtual
GCC: 4.7.2 (Ubuntu/Linaro 4.7.2-11precise2)

I tried to build DPDK 18.05 after installing the Mellanox OFED driver 
(MLNX_OFED_LINUX-4.0-2.0.0.1-ubuntu12.04-x86_64), but it failed with the 
mentioned error.
So I removed the OFED driver and installed the EN driver 
(mlnx-en-4.0-2.0.0.1-ubuntu12.04-x86_64), but it also failed with the same 
error.
Both drivers were compiled with "--dpdk" argument ("--upstream-libs" is not 
supported).

-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Asaf Sinai
Sent: Monday, July 2, 2018 03:20 PM
To: Shahaf Shuler 
Cc: dev@dpdk.org; Nitin Katiyar ; Nélio Laranjeiro 

Subject: Re: [dpdk-dev] Compilation of MLX5 driver

Hi Shahaf,

I face similar problem with DPDK 18.05, where missing file is "mlx5dv.h":

== Build drivers/net/mlx5
  CC mlx5.o
In file included from 
/net/emcradware/CM_Users/asafsi/work/drivers/dpdk/dpdk-18.05/drivers/net/mlx5/mlx5.h:35:0,
 from 
/net/emcradware/CM_Users/asafsi/work/drivers/dpdk/dpdk-18.05/drivers/net/mlx5/mlx5.c:40:

/net/emcradware/CM_Users/asafsi/work/drivers/dpdk/dpdk-18.05/drivers/net/mlx5/mlx5_mr.h:19:31:
 fatal error: infiniband/mlx5dv.h: No such file ordirectory
 #include 
   ^
compilation terminated.

Can you please advise?

Thanks,
Asaf

-Original Message-
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Nitin Katiyar
Sent: Thursday, May 31, 2018 02:44 PM
To: Nélio Laranjeiro 
Cc: Shahaf Shuler ; dev@dpdk.org
Subject: Re: [dpdk-dev] Compilation of MLX5 driver

Thanks Shahaf, it worked after removing the options you specified.

Regards,
Nitin

-Original Message-
From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
Sent: Thursday, May 31, 2018 3:23 PM
To: Nitin Katiyar 
Cc: Shahaf Shuler ; dev@dpdk.org
Subject: Re: [dpdk-dev] Compilation of MLX5 driver

On Thu, May 31, 2018 at 09:14:03AM +, Nitin Katiyar wrote:
> Yes,I installed it using --dpdk --upstream-libs. What is the way 
> forward now?

In v17.05 MLX5 PMD is still relying on libibverbs and libmlx5, the way Those 
options you used are necessary to select in their package the installation of 
libverbs,libmlx5 or rdma-core.
Doing this you have selected rdma-core which is not supported in v17.05 DPDK 
version.

You need to install Mellanox OFED without those two options to select 
libibverbs, libmlx5 to make it work.

Regards,
 
> Regards,
> Nitin
> 
> -Original Message-
> From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> Sent: Thursday, May 31, 2018 1:36 PM
> To: Nitin Katiyar 
> Cc: Shahaf Shuler ; dev@dpdk.org
> Subject: Re: [dpdk-dev] Compilation of MLX5 driver
> 
> On Thu, May 31, 2018 at 07:01:17AM +, Nitin Katiyar wrote:
> > Hi,
> > It has following files:
> > 
> > arch.h  ib.h  kern-abi.h  mlx4dv.h  mlx5dv.h  opcode.h  sa.h 
> > sa-kern-abi.h  verbs.h
> > 
> > I tried with both MLNX_OFED_LINUX-4.2-1.0.0.0 and
> > MLNX_OFED_LINUX-4.2-1.2.0.0-ubuntu14.04-x86_64
> 
> Did you installed Mellanox OFED with the --dpdk --upstream-libs arguments for 
> the installation script?
> 
> If it is the case, you should not add them for this version, those options 
> are for DPDK v17.11 and higher.
> 
> Regards,
> 
> > Regards,
> > Nitin
> > 
> > -Original Message-
> > From: Shahaf Shuler [mailto:shah...@mellanox.com]
> > Sent: Thursday, May 31, 2018 10:51 AM
> > To: Nitin Katiyar ; Nélio Laranjeiro 
> > 
> > Cc: dev@dpdk.org
> > Subject: RE: [dpdk-dev] Compilation of MLX5 driver
> > 
> > Wednesday, May 30, 2018 7:45 PM, Nitin Katiyar:
> > > 
> > > Hi,
> > > I was compiling 17.05.02.
> > > Regards,
> > > Nitin
> > > 
> > > -Original Message-
> > > From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> > > Sent: Wednesday, May 30, 2018 6:42 PM
> > > To: Nitin Katiyar 
> > > Cc: dev@dpdk.org
> > > Subject: Re: [dpdk-dev] Compilation of MLX5 driver
> > > 
> > > Hi,
> > > 
> > > On Wed, May 30, 2018 at 11:54:31AM +, Nitin Katiyar wrote:
> > > > Hi,
> > > > I am trying to compile MLX5 PMD driver by setting
> > > "CONFIG_RTE_LIBRTE_MLX5_PMD=y" and hitting following compilation 
> > > error.
> > > >
> > > > fatal error: infiniband/mlx5_hw.h: No such file or directory
> > 
> > Can you list the files you have under /usr/include/infiniband ? 
> > 
> > > >
> > > > I have installed MLNX_OFED _LINUX-4.2-1.2.0 on my Ubuntu 14.04 
> > > > machine
> > > but still hitting the same error. Am I missing some other package?
> > > 
> > > Which version of DPDK are you using (it is important to help)?
> > > 
> > > Regards,
> > > 
> > > --
> > > Nélio Laranjeiro
> > > 6WIND
> 
> --
> Nélio Laranjeiro
> 6WIND

--
Nélio Laranjeiro
6WIND

Re: [dpdk-dev] [PATCH v2 10/23] net/bnxt: move function check zero bytes to bnxt util.h

[Ferruh Yigit]

On 7/2/2018 1:20 PM, Ferruh Yigit wrote:
> On 6/28/2018 9:15 PM, Ajit Khaparde wrote:
>> From: Scott Branden 
>>
>> Move check_zero_bytes into new bnxt_util.h file.
>>
>> Signed-off-by: Scott Branden 
>> Reviewed-by: Ajit Khaparde 
>> ---
>>  drivers/net/bnxt/Makefile  |  1 +
>>  drivers/net/bnxt/bnxt_ethdev.c |  1 +
>>  drivers/net/bnxt/bnxt_filter.c |  9 -
>>  drivers/net/bnxt/bnxt_filter.h |  1 -
>>  drivers/net/bnxt/bnxt_util.c   | 18 ++
>>  drivers/net/bnxt/bnxt_util.h   | 11 +++
>>  6 files changed, 31 insertions(+), 10 deletions(-)
>>  create mode 100644 drivers/net/bnxt/bnxt_util.c
>>  create mode 100644 drivers/net/bnxt/bnxt_util.h
>>
>> diff --git a/drivers/net/bnxt/Makefile b/drivers/net/bnxt/Makefile
>> index fd0cb5235..80db03ea8 100644
>> --- a/drivers/net/bnxt/Makefile
>> +++ b/drivers/net/bnxt/Makefile
>> @@ -38,6 +38,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_txq.c
>>  SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_txr.c
>>  SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_vnic.c
>>  SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_irq.c
>> +SRCS-$(CONFIG_RTE_LIBRTE_BNXT_PMD) += bnxt_util.c
> 
> This breaks the meson build and similar change required for meson, same with
> bnxt_flow.c in other patch, if there is no other issue I can fix this while 
> merging.

This patch also breaks the Makefile build, 'bnxt_filter.c' requires:
 #include "bnxt_util.h"

Re: [dpdk-dev] [PATCH v4] net/e1000: add support for check descriptor status APIs

[Zhang, Qi Z]

> -Original Message-
> From: Zhao1, Wei
> Sent: Monday, July 2, 2018 3:42 PM
> To: Yigit, Ferruh 
> Cc: Zhang, Qi Z ; dev@dpdk.org
> Subject: RE: [PATCH v4] net/e1000: add support for check descriptor status
> APIs
> 
> Hi,  Ferruh
> 
>   The doc of igb_vf.ini has enable "Rx  descriptor status" & "Tx 
> descriptor
> status" features, so I do not need update this doc in this patch.

> 
> Thanks
> 
> > -Original Message-
> > From: Zhao1, Wei
> > Sent: Friday, June 29, 2018 9:53 AM
> > To: dev@dpdk.org
> > Cc: Zhang, Qi Z ; Zhao1, Wei
> > 
> > Subject: [PATCH v4] net/e1000: add support for check descriptor status
> > APIs
> >
> > rte_eth_rx_descritpr_status and rte_eth_tx_descriptor_status are
> > supported by igb VF.
> >
> > Signed-off-by: Wei Zhao 

So the patch looks like a fix, but not new feature enable, so I will remove the 
release notes update.

Acked-by: Qi Zhang 

Re: [dpdk-dev] [PATCH v4 0/3] crypto/qat: move files to drivers/common directory

[De Lara Guarch, Pablo]

> -Original Message-
> From: Jozwiak, TomaszX
> Sent: Monday, July 2, 2018 10:39 AM
> To: Trahe, Fiona ; De Lara Guarch, Pablo
> ; dev@dpdk.org
> Subject: [PATCH v4 0/3] crypto/qat: move files to drivers/common directory
> 
> This patchset depends on QAT dynamic logging patchset and should be targetig
> on 18.08.
> Patchset refactors the PMD in order that files are split into several
> places: common, crypto.
> New drivers/common/qat are added and files split between locations.

Series applied to dpdk-next-crypto.

Thanks,
Pablo

Re: [dpdk-dev] [PATCH v7] net/fm10k: add support for check descriptor status APIs

[Zhang, Qi Z]

> -Original Message-
> From: Zhao1, Wei
> Sent: Monday, July 2, 2018 3:16 PM
> To: dev@dpdk.org
> Cc: Zhang, Qi Z ; Yigit, Ferruh 
> ;
> Zhao1, Wei 
> Subject: [PATCH v7] net/fm10k: add support for check descriptor status APIs

<...>

> diff --git a/doc/guides/rel_notes/release_18_08.rst
> b/doc/guides/rel_notes/release_18_08.rst
> index 68aaee8..ac54121 100644
> --- a/doc/guides/rel_notes/release_18_08.rst
> +++ b/doc/guides/rel_notes/release_18_08.rst
> @@ -54,6 +54,12 @@ New Features
>PMD does not provide any. The provision of such tuned values now
> includes
>the ixgbe PMD.
> 
> +* **Added fm10k ethernet driver to support check descriptor status
> +APIs.**

Will change to "Added descriptor status check support for fm10k"

> +
> +  rte_eth_rx_descritpr_status and rte_eth_tx_descriptor_status  are
> + supported by fm10K.

API in release notes need be ``rte_eth.``, will fix during apply.

Acked-by: Qi Zhang 

<...>

Re: [dpdk-dev] [PATCH v7] net/fm10k: add support for check descriptor status APIs

[Zhang, Qi Z]

> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Zhang, Qi Z
> Sent: Monday, July 2, 2018 9:08 PM
> To: Zhao1, Wei ; dev@dpdk.org
> Cc: Yigit, Ferruh 
> Subject: Re: [dpdk-dev] [PATCH v7] net/fm10k: add support for check
> descriptor status APIs
> 
> > -Original Message-
> > From: Zhao1, Wei
> > Sent: Monday, July 2, 2018 3:16 PM
> > To: dev@dpdk.org
> > Cc: Zhang, Qi Z ; Yigit, Ferruh
> > ; Zhao1, Wei 
> > Subject: [PATCH v7] net/fm10k: add support for check descriptor status
> > APIs
> 
> <...>
> 
> > diff --git a/doc/guides/rel_notes/release_18_08.rst
> > b/doc/guides/rel_notes/release_18_08.rst
> > index 68aaee8..ac54121 100644
> > --- a/doc/guides/rel_notes/release_18_08.rst
> > +++ b/doc/guides/rel_notes/release_18_08.rst
> > @@ -54,6 +54,12 @@ New Features
> >PMD does not provide any. The provision of such tuned values now
> > includes
> >the ixgbe PMD.
> >
> > +* **Added fm10k ethernet driver to support check descriptor status
> > +APIs.**
> 
> Will change to "Added descriptor status check support for fm10k"
> 
> > +
> > +  rte_eth_rx_descritpr_status and rte_eth_tx_descriptor_status  are
> > + supported by fm10K.
> 
> API in release notes need be ``rte_eth.``, will fix during apply.
> 
> Acked-by: Qi Zhang 

Applied to dpdk-next-net-intel

Thanks!
Qi

Re: [dpdk-dev] [PATCH v4] net/e1000: add support for check descriptor status APIs

[Zhang, Qi Z]

> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Zhang, Qi Z
> Sent: Monday, July 2, 2018 9:04 PM
> To: Zhao1, Wei ; Yigit, Ferruh 
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v4] net/e1000: add support for check
> descriptor status APIs
> 
> 
> 
> > -Original Message-
> > From: Zhao1, Wei
> > Sent: Monday, July 2, 2018 3:42 PM
> > To: Yigit, Ferruh 
> > Cc: Zhang, Qi Z ; dev@dpdk.org
> > Subject: RE: [PATCH v4] net/e1000: add support for check descriptor
> > status APIs
> >
> > Hi,  Ferruh
> >
> > The doc of igb_vf.ini has enable "Rx  descriptor status" & "Tx
> > descriptor status" features, so I do not need update this doc in this patch.
> 
> >
> > Thanks
> >
> > > -Original Message-
> > > From: Zhao1, Wei
> > > Sent: Friday, June 29, 2018 9:53 AM
> > > To: dev@dpdk.org
> > > Cc: Zhang, Qi Z ; Zhao1, Wei
> > > 
> > > Subject: [PATCH v4] net/e1000: add support for check descriptor
> > > status APIs
> > >
> > > rte_eth_rx_descritpr_status and rte_eth_tx_descriptor_status are
> > > supported by igb VF.
> > >
> > > Signed-off-by: Wei Zhao 
> 
> So the patch looks like a fix, but not new feature enable, so I will remove 
> the
> release notes update.
> 
> Acked-by: Qi Zhang 

Applied to dpdk-next-net-intel

Thanks!
Qi

[dpdk-dev] [PATCH v1] lib/bitratestats: add NULL sanity check

[Remy Horton]

If rte_stats_bitrate_reg() is passed NULL, the result is a crash.
Fixed by adding a sanity check that makes sure the passed-in
pointer is not NULL.

Fixes: 2ad7ba9a6567 ("bitrate: add bitrate statistics library")

Signed-off-by: Remy Horton 
---
 lib/librte_bitratestats/rte_bitrate.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/lib/librte_bitratestats/rte_bitrate.c 
b/lib/librte_bitratestats/rte_bitrate.c
index 964e3c3..20e1755 100644
--- a/lib/librte_bitratestats/rte_bitrate.c
+++ b/lib/librte_bitratestats/rte_bitrate.c
@@ -47,6 +47,9 @@ rte_stats_bitrate_reg(struct rte_stats_bitrates *bitrate_data)
};
int return_value;
 
+   if (bitrate_data == NULL)
+   return -EINVAL;
+
return_value = rte_metrics_reg_names(&names[0], ARRAY_SIZE(names));
if (return_value >= 0)
bitrate_data->id_stats_set = return_value;
-- 
2.9.5

[dpdk-dev] [PATCH v1] lib/metrics: disallow NULL as metric name

[Remy Horton]

This patch adds a sanity check so that names passed into
rte_metrics_reg_names() and the wrapper rte_metrics_reg_name()
cannot be NULL.

Fixes: 349950ddb9c5 ("metrics: add information metrics library")

Signed-off-by: Remy Horton 
---
 lib/librte_metrics/rte_metrics.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/lib/librte_metrics/rte_metrics.c b/lib/librte_metrics/rte_metrics.c
index 258f058..2275244 100644
--- a/lib/librte_metrics/rte_metrics.c
+++ b/lib/librte_metrics/rte_metrics.c
@@ -96,6 +96,9 @@ rte_metrics_reg_names(const char * const *names, uint16_t 
cnt_names)
/* Some sanity checks */
if (cnt_names < 1 || names == NULL)
return -EINVAL;
+   for (idx_name = 0; idx_name < cnt_names; idx_name++)
+   if (names[idx_name] == NULL)
+   return -EINVAL;
 
memzone = rte_memzone_lookup(RTE_METRICS_MEMZONE_NAME);
if (memzone == NULL)
-- 
2.9.5

Re: [dpdk-dev] Compilation of MLX5 driver

[Shahaf Shuler]

Hi Asaf,

Monday, July 2, 2018 3:46 PM, Asaf Sinai:
> Subject: RE: [dpdk-dev] Compilation of MLX5 driver
> 
> Additional details:
> 
> Ubuntu: 12.04.3
> Linux: 3.2.0-53-virtual
> GCC: 4.7.2 (Ubuntu/Linaro 4.7.2-11precise2)
> 
> I tried to build DPDK 18.05 after installing the Mellanox OFED driver

DPDK version 18.05 supports only upstream-libs (rdma-core).  

> (MLNX_OFED_LINUX-4.0-2.0.0.1-ubuntu12.04-x86_64), but it failed with the
> mentioned error.
> So I removed the OFED driver and installed the EN driver (mlnx-en-4.0-
> 2.0.0.1-ubuntu12.04-x86_64), but it also failed with the same error.
> Both drivers were compiled with "--dpdk" argument ("--upstream-libs" is not
> supported).

Yes MLNX_OFED-4.0 is too old for upstream libs. You should use MLNX_OFED-4.2 
and above. 
Or alternatively use an older DPDK version (not recommended). 

> 
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Asaf Sinai
> Sent: Monday, July 2, 2018 03:20 PM
> To: Shahaf Shuler 
> Cc: dev@dpdk.org; Nitin Katiyar ; Nélio
> Laranjeiro 
> Subject: Re: [dpdk-dev] Compilation of MLX5 driver
> 
> Hi Shahaf,
> 
> I face similar problem with DPDK 18.05, where missing file is "mlx5dv.h":
> 
>   == Build drivers/net/mlx5
> CC mlx5.o
>   In file included from
> /net/emcradware/CM_Users/asafsi/work/drivers/dpdk/dpdk-
> 18.05/drivers/net/mlx5/mlx5.h:35:0,
>from
> /net/emcradware/CM_Users/asafsi/work/drivers/dpdk/dpdk-
> 18.05/drivers/net/mlx5/mlx5.c:40:
>   /net/emcradware/CM_Users/asafsi/work/drivers/dpdk/dpdk-
> 18.05/drivers/net/mlx5/mlx5_mr.h:19:31: fatal error: infiniband/mlx5dv.h:
> No such file or   directory
>#include 
>  ^
>   compilation terminated.
> 
> Can you please advise?
> 
> Thanks,
> Asaf
> 
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Nitin Katiyar
> Sent: Thursday, May 31, 2018 02:44 PM
> To: Nélio Laranjeiro 
> Cc: Shahaf Shuler ; dev@dpdk.org
> Subject: Re: [dpdk-dev] Compilation of MLX5 driver
> 
> Thanks Shahaf, it worked after removing the options you specified.
> 
> Regards,
> Nitin
> 
> -Original Message-
> From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> Sent: Thursday, May 31, 2018 3:23 PM
> To: Nitin Katiyar 
> Cc: Shahaf Shuler ; dev@dpdk.org
> Subject: Re: [dpdk-dev] Compilation of MLX5 driver
> 
> On Thu, May 31, 2018 at 09:14:03AM +, Nitin Katiyar wrote:
> > Yes,I installed it using --dpdk --upstream-libs. What is the way
> > forward now?
> 
> In v17.05 MLX5 PMD is still relying on libibverbs and libmlx5, the way Those
> options you used are necessary to select in their package the installation of
> libverbs,libmlx5 or rdma-core.
> Doing this you have selected rdma-core which is not supported in v17.05
> DPDK version.
> 
> You need to install Mellanox OFED without those two options to select
> libibverbs, libmlx5 to make it work.
> 
> Regards,
> 
> > Regards,
> > Nitin
> >
> > -Original Message-
> > From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> > Sent: Thursday, May 31, 2018 1:36 PM
> > To: Nitin Katiyar 
> > Cc: Shahaf Shuler ; dev@dpdk.org
> > Subject: Re: [dpdk-dev] Compilation of MLX5 driver
> >
> > On Thu, May 31, 2018 at 07:01:17AM +, Nitin Katiyar wrote:
> > > Hi,
> > > It has following files:
> > >
> > > arch.h  ib.h  kern-abi.h  mlx4dv.h  mlx5dv.h  opcode.h  sa.h
> > > sa-kern-abi.h  verbs.h
> > >
> > > I tried with both MLNX_OFED_LINUX-4.2-1.0.0.0 and
> > > MLNX_OFED_LINUX-4.2-1.2.0.0-ubuntu14.04-x86_64
> >
> > Did you installed Mellanox OFED with the --dpdk --upstream-libs
> arguments for the installation script?
> >
> > If it is the case, you should not add them for this version, those options 
> > are
> for DPDK v17.11 and higher.
> >
> > Regards,
> >
> > > Regards,
> > > Nitin
> > >
> > > -Original Message-
> > > From: Shahaf Shuler [mailto:shah...@mellanox.com]
> > > Sent: Thursday, May 31, 2018 10:51 AM
> > > To: Nitin Katiyar ; Nélio Laranjeiro
> > > 
> > > Cc: dev@dpdk.org
> > > Subject: RE: [dpdk-dev] Compilation of MLX5 driver
> > >
> > > Wednesday, May 30, 2018 7:45 PM, Nitin Katiyar:
> > > >
> > > > Hi,
> > > > I was compiling 17.05.02.
> > > > Regards,
> > > > Nitin
> > > >
> > > > -Original Message-
> > > > From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> > > > Sent: Wednesday, May 30, 2018 6:42 PM
> > > > To: Nitin Katiyar 
> > > > Cc: dev@dpdk.org
> > > > Subject: Re: [dpdk-dev] Compilation of MLX5 driver
> > > >
> > > > Hi,
> > > >
> > > > On Wed, May 30, 2018 at 11:54:31AM +, Nitin Katiyar wrote:
> > > > > Hi,
> > > > > I am trying to compile MLX5 PMD driver by setting
> > > > "CONFIG_RTE_LIBRTE_MLX5_PMD=y" and hitting following compilation
> > > > error.
> > > > >
> > > > > fatal error: infiniband/mlx5_hw.h: No such file or directory
> > >
> > > Can you list the files you have under /usr/include/infiniband ?
> > >
> > > > >
> > > > >

Re: [dpdk-dev] Compilation of MLX5 driver

[Asaf Sinai]

Hi Shahaf,

1. Can you please explain what is the meaning of upstream-libs?
2. We use Ubuntu 12.04. so if we would like to stay with it, and use 
MLNX_OFED-4.0, which latest DPDK version should be used?

Thanks,
Asaf
 
-Original Message-
From: Shahaf Shuler [mailto:shah...@mellanox.com] 
Sent: Monday, July 2, 2018 05:11 PM
To: Asaf Sinai 
Cc: dev@dpdk.org; Nitin Katiyar ; Nélio Laranjeiro 

Subject: RE: [dpdk-dev] Compilation of MLX5 driver

Hi Asaf,

Monday, July 2, 2018 3:46 PM, Asaf Sinai:
> Subject: RE: [dpdk-dev] Compilation of MLX5 driver
> 
> Additional details:
> 
> Ubuntu: 12.04.3
> Linux: 3.2.0-53-virtual
> GCC: 4.7.2 (Ubuntu/Linaro 4.7.2-11precise2)
> 
> I tried to build DPDK 18.05 after installing the Mellanox OFED driver

DPDK version 18.05 supports only upstream-libs (rdma-core).  

> (MLNX_OFED_LINUX-4.0-2.0.0.1-ubuntu12.04-x86_64), but it failed with 
> the mentioned error.
> So I removed the OFED driver and installed the EN driver (mlnx-en-4.0- 
> 2.0.0.1-ubuntu12.04-x86_64), but it also failed with the same error.
> Both drivers were compiled with "--dpdk" argument ("--upstream-libs" 
> is not supported).

Yes MLNX_OFED-4.0 is too old for upstream libs. You should use MLNX_OFED-4.2 
and above. 
Or alternatively use an older DPDK version (not recommended). 

> 
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Asaf Sinai
> Sent: Monday, July 2, 2018 03:20 PM
> To: Shahaf Shuler 
> Cc: dev@dpdk.org; Nitin Katiyar ; Nélio 
> Laranjeiro 
> Subject: Re: [dpdk-dev] Compilation of MLX5 driver
> 
> Hi Shahaf,
> 
> I face similar problem with DPDK 18.05, where missing file is "mlx5dv.h":
> 
>   == Build drivers/net/mlx5
> CC mlx5.o
>   In file included from
> /net/emcradware/CM_Users/asafsi/work/drivers/dpdk/dpdk-
> 18.05/drivers/net/mlx5/mlx5.h:35:0,
>from
> /net/emcradware/CM_Users/asafsi/work/drivers/dpdk/dpdk-
> 18.05/drivers/net/mlx5/mlx5.c:40:
>   /net/emcradware/CM_Users/asafsi/work/drivers/dpdk/dpdk-
> 18.05/drivers/net/mlx5/mlx5_mr.h:19:31: fatal error: infiniband/mlx5dv.h:
> No such file or   directory
>#include 
>  ^
>   compilation terminated.
> 
> Can you please advise?
> 
> Thanks,
> Asaf
> 
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Nitin Katiyar
> Sent: Thursday, May 31, 2018 02:44 PM
> To: Nélio Laranjeiro 
> Cc: Shahaf Shuler ; dev@dpdk.org
> Subject: Re: [dpdk-dev] Compilation of MLX5 driver
> 
> Thanks Shahaf, it worked after removing the options you specified.
> 
> Regards,
> Nitin
> 
> -Original Message-
> From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> Sent: Thursday, May 31, 2018 3:23 PM
> To: Nitin Katiyar 
> Cc: Shahaf Shuler ; dev@dpdk.org
> Subject: Re: [dpdk-dev] Compilation of MLX5 driver
> 
> On Thu, May 31, 2018 at 09:14:03AM +, Nitin Katiyar wrote:
> > Yes,I installed it using --dpdk --upstream-libs. What is the way 
> > forward now?
> 
> In v17.05 MLX5 PMD is still relying on libibverbs and libmlx5, the way 
> Those options you used are necessary to select in their package the 
> installation of
> libverbs,libmlx5 or rdma-core.
> Doing this you have selected rdma-core which is not supported in 
> v17.05 DPDK version.
> 
> You need to install Mellanox OFED without those two options to select 
> libibverbs, libmlx5 to make it work.
> 
> Regards,
> 
> > Regards,
> > Nitin
> >
> > -Original Message-
> > From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> > Sent: Thursday, May 31, 2018 1:36 PM
> > To: Nitin Katiyar 
> > Cc: Shahaf Shuler ; dev@dpdk.org
> > Subject: Re: [dpdk-dev] Compilation of MLX5 driver
> >
> > On Thu, May 31, 2018 at 07:01:17AM +, Nitin Katiyar wrote:
> > > Hi,
> > > It has following files:
> > >
> > > arch.h  ib.h  kern-abi.h  mlx4dv.h  mlx5dv.h  opcode.h  sa.h 
> > > sa-kern-abi.h  verbs.h
> > >
> > > I tried with both MLNX_OFED_LINUX-4.2-1.0.0.0 and
> > > MLNX_OFED_LINUX-4.2-1.2.0.0-ubuntu14.04-x86_64
> >
> > Did you installed Mellanox OFED with the --dpdk --upstream-libs
> arguments for the installation script?
> >
> > If it is the case, you should not add them for this version, those 
> > options are
> for DPDK v17.11 and higher.
> >
> > Regards,
> >
> > > Regards,
> > > Nitin
> > >
> > > -Original Message-
> > > From: Shahaf Shuler [mailto:shah...@mellanox.com]
> > > Sent: Thursday, May 31, 2018 10:51 AM
> > > To: Nitin Katiyar ; Nélio Laranjeiro 
> > > 
> > > Cc: dev@dpdk.org
> > > Subject: RE: [dpdk-dev] Compilation of MLX5 driver
> > >
> > > Wednesday, May 30, 2018 7:45 PM, Nitin Katiyar:
> > > >
> > > > Hi,
> > > > I was compiling 17.05.02.
> > > > Regards,
> > > > Nitin
> > > >
> > > > -Original Message-
> > > > From: Nélio Laranjeiro [mailto:nelio.laranje...@6wind.com]
> > > > Sent: Wednesday, May 30, 2018 6:42 PM
> > > > To: Nitin Katiyar 
> > > > Cc: dev@dpdk.org
> > > > Subject: Re: [dpdk-dev] Com

[dpdk-dev] [PATCH v1] lib/metrics: fix silent fail when uninitialised

[Remy Horton]

If rte_metrics_init() had not been called and hence the internal
metric storage is not allocated, rte_metrics_get_values() and
rte_metrics_get_name() would silently fail by returning zero
(i.e. no metrics registered). This patch changes the result of
this scenario to an explicit fail by returning -EIO.

Fixes: 349950ddb9c5 ("metrics: add information metrics library")

Signed-off-by: Remy Horton 
---
 lib/librte_metrics/rte_metrics.c | 7 +++
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/lib/librte_metrics/rte_metrics.c b/lib/librte_metrics/rte_metrics.c
index 258f058..fed0a33 100644
--- a/lib/librte_metrics/rte_metrics.c
+++ b/lib/librte_metrics/rte_metrics.c
@@ -200,9 +200,8 @@ rte_metrics_get_names(struct rte_metric_name *names,
int return_value;
 
memzone = rte_memzone_lookup(RTE_METRICS_MEMZONE_NAME);
-   /* If not allocated, fail silently */
if (memzone == NULL)
-   return 0;
+   return -EIO;
 
stats = memzone->addr;
rte_spinlock_lock(&stats->lock);
@@ -238,9 +237,9 @@ rte_metrics_get_values(int port_id,
return -EINVAL;
 
memzone = rte_memzone_lookup(RTE_METRICS_MEMZONE_NAME);
-   /* If not allocated, fail silently */
if (memzone == NULL)
-   return 0;
+   return -EIO;
+
stats = memzone->addr;
rte_spinlock_lock(&stats->lock);
 
-- 
2.9.5

Re: [dpdk-dev] [PATCH] ethdev: fix queue mapping documentation

[Ferruh Yigit]

On 6/29/2018 10:44 AM, Jerin Jacob wrote:
> The RTE_MAX_ETHPORT_QUEUE_STATS_MAPS does not exists, change
> to the correct definition(RTE_ETHDEV_QUEUE_STAT_CNTRS)
> 
> Fixes: 5de201df8927 ("ethdev: add stats per queue")
> Cc: sta...@dpdk.org
> 
> Signed-off-by: Jerin Jacob 
> ---
>  lib/librte_ethdev/rte_ethdev.h | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/lib/librte_ethdev/rte_ethdev.h b/lib/librte_ethdev/rte_ethdev.h
> index 36e3984ea..375ea24ce 100644
> --- a/lib/librte_ethdev/rte_ethdev.h
> +++ b/lib/librte_ethdev/rte_ethdev.h
> @@ -2144,7 +2144,7 @@ void rte_eth_xstats_reset(uint16_t port_id);
>   * @param stat_idx
>   *   The per-queue packet statistics functionality number that the transmit
>   *   queue is to be assigned.
> - *   The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 
> 1].
> + *   The value must be in the range [0, RTE_ETHDEV_QUEUE_STAT_CNTRS - 1].

Yes RTE_MAX_ETHPORT_QUEUE_STATS_MAPS doesn't exits and comment is wrong, but
RTE_ETHDEV_QUEUE_STAT_CNTRS also slightly not correct.

I think how testpmd uses it increase the confusion.

In ixgbe there is no stats registers per queue, 128 queues are represented by 16
register set. stat_idx here is the index of that 16 registers. You map queue to
stats requester to get queue stats.

Also there is RTE_ETHDEV_QUEUE_STAT_CNTRS config in the ethdev API, which is the
hardcoded size of the queue stats, its default value is 16. This limits number
of the queues we can get stats from but saves allocated space. (Why not 
dynamic?)

You can increase the RTE_ETHDEV_QUEUE_STAT_CNTRS to the max supported number of
queue and ethdev code will be all valid. But "stat_idx" can't go beyond 16 (for
ixgbe) because it is hardware limitation and it may change from hw to hw.

Also technically it should be possible to reduce RTE_ETHDEV_QUEUE_STAT_CNTRS to
a low number, like 2, but in ixgbe map two queues into stat registers 14 & 15
and display those two set as queue stat 0 and 1. It seems current implementation
prevents this and forces the queues mapped should be less than
RTE_ETHDEV_QUEUE_STAT_CNTRS. Overall it seems there is a mixed used of
RTE_ETHDEV_QUEUE_STAT_CNTRS and stats queue index values, I assume because both
are same values.

I suggest updating it as:
"
The value must be in the range:
 [0 - MIN(HW Stat Registers Size, RTE_ETHDEV_QUEUE_STAT_CNTRS) - 1]
"

>   * @return
>   *   Zero if successful. Non-zero otherwise.
>   */
> @@ -2164,7 +2164,7 @@ int rte_eth_dev_set_tx_queue_stats_mapping(uint16_t 
> port_id,
>   * @param stat_idx
>   *   The per-queue packet statistics functionality number that the receive
>   *   queue is to be assigned.
> - *   The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 
> 1].
> + *   The value must be in the range [0, RTE_ETHDEV_QUEUE_STAT_CNTRS - 1].
>   * @return
>   *   Zero if successful. Non-zero otherwise.
>   */
> 

Re: [dpdk-dev] [PATCH v5 8/9] net/virtio: add in-order Rx/Tx into selection

[Maxime Coquelin]

On 07/02/2018 02:41 PM, Maxime Coquelin wrote:



On 07/02/2018 01:24 PM, Maxime Coquelin wrote:



On 07/02/2018 03:56 PM, Marvin Liu wrote:

After IN_ORDER Rx/Tx paths added, need to update Rx/Tx path selection
logic.

Rx path select logic: If IN_ORDER and merge-able are enabled will select
IN_ORDER Rx path. If IN_ORDER is enabled, Rx offload and merge-able are
disabled will select simple Rx path. Otherwise will select normal Rx
path.

Tx path select logic: If IN_ORDER is enabled will select IN_ORDER Tx
path. Otherwise will select default Tx path.

Signed-off-by: Marvin Liu 
Reviewed-by: Maxime Coquelin 

diff --git a/doc/guides/nics/virtio.rst b/doc/guides/nics/virtio.rst
index 46e292c4d..7c099fb7c 100644
--- a/doc/guides/nics/virtio.rst
+++ b/doc/guides/nics/virtio.rst
@@ -201,7 +201,7 @@ The packet transmission flow is:
  Virtio PMD Rx/Tx Callbacks
  --
-Virtio driver has 3 Rx callbacks and 2 Tx callbacks.
+Virtio driver has 4 Rx callbacks and 3 Tx callbacks.
  Rx callbacks:
@@ -215,6 +215,9 @@ Rx callbacks:
 Vector version without mergeable Rx buffer support, also fixes 
the available

 ring indexes and uses vector instructions to optimize performance.
+#. ``virtio_recv_mergeable_pkts_inorder``:
+   In-order version with mergeable Rx buffer support.
+
  Tx callbacks:
  #. ``virtio_xmit_pkts``:
@@ -223,6 +226,8 @@ Tx callbacks:
  #. ``virtio_xmit_pkts_simple``:
 Vector version fixes the available ring indexes to optimize 
performance.

+#. ``virtio_xmit_pkts_inorder``:
+   In-order version.
  By default, the non-vector callbacks are used:
@@ -254,6 +259,12 @@ Example of using the vector version of the 
virtio poll mode driver in
 testpmd -l 0-2 -n 4 -- -i --tx-offloads=0x0 --rxq=1 --txq=1 
--nb-cores=1

+In-order callbacks only work on simulated virtio user vdev.
+
+*   For Rx: If mergeable Rx buffers is enabled and in-order is 
enabled then

+    ``virtio_xmit_pkts_inorder`` is used.
+
+*   For Tx: If in-order is enabled then ``virtio_xmit_pkts_inorder`` 
is used.

  Interrupt mode
  --
diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c

index df50a571a..df7981ddb 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1320,6 +1320,11 @@ set_rxtx_funcs(struct rte_eth_dev *eth_dev)
  PMD_INIT_LOG(INFO, "virtio: using simple Rx path on port %u",
  eth_dev->data->port_id);
  eth_dev->rx_pkt_burst = virtio_recv_pkts_vec;
+    } else if (hw->use_inorder_rx) {
+    PMD_INIT_LOG(INFO,
+    "virtio: using inorder mergeable buffer Rx path on port 
%u",

+    eth_dev->data->port_id);
+    eth_dev->rx_pkt_burst = &virtio_recv_mergeable_pkts_inorder;
  } else if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) {
  PMD_INIT_LOG(INFO,
  "virtio: using mergeable buffer Rx path on port %u",
@@ -1335,6 +1340,10 @@ set_rxtx_funcs(struct rte_eth_dev *eth_dev)
  PMD_INIT_LOG(INFO, "virtio: using simple Tx path on port %u",
  eth_dev->data->port_id);
  eth_dev->tx_pkt_burst = virtio_xmit_pkts_simple;
+    } else if (hw->use_inorder_tx) {
+    PMD_INIT_LOG(INFO, "virtio: using inorder Tx path on port %u",
+    eth_dev->data->port_id);
+    eth_dev->tx_pkt_burst = virtio_xmit_pkts_inorder;
  } else {
  PMD_INIT_LOG(INFO, "virtio: using standard Tx path on port 
%u",

  eth_dev->data->port_id);
@@ -1874,20 +1883,27 @@ virtio_dev_configure(struct rte_eth_dev *dev)
  hw->use_simple_rx = 1;
  hw->use_simple_tx = 1;
+    if (vtpci_with_feature(hw, VIRTIO_F_IN_ORDER)) {
+    /* Simple Tx not compatible with in-order ring */
+    hw->use_inorder_tx = 1;
+    hw->use_simple_tx = 0;
+    if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) {
+    hw->use_inorder_rx = 1;
+    hw->use_simple_rx = 0;
+    } else {
+    hw->use_inorder_rx = 0;
+    if (rx_offloads & (DEV_RX_OFFLOAD_UDP_CKSUM |
+   DEV_RX_OFFLOAD_TCP_CKSUM))
+    hw->use_simple_rx = 0;

It is applied, but I think this is still not good.

Simple Rx is set to 1 by default, so if IN_ORDER isn't negotiated,
and UDP/TCP_CSUM is enabled, simple Rx keeps being selected.

I'll fix it in my series that I'm doing on top.


Actually, after discussion with Ferruh, I fixed it directly in the patch.


Running some more tests, I noticed that it is still broken.
For example the case where host does not support IN_ORDER.
If mergeable buffer is negotiated, the simple path gets selected, which
is wrong.

I fixed that directly in the patch.

Regards,
Maxime

Thanks,
Maxime


Regards,
Maxime


+    }
+    }
+
  #if defined RTE_ARCH_ARM64 || defined RTE_ARCH_ARM
  if (!rte_cpu_get_flag_enabled(RTE_CPUFLAG_NEON)) {
  hw->use_simple_rx = 0;
  hw->use_simple_tx = 0;
  }
  #endif
-    if (vtpci_with_feature(hw, 

[dpdk-dev] [PATCH v4 1/4] net/virtio: remove simple Tx path

[Maxime Coquelin]

The simple Tx path does not comply with the Virtio specification.
Now that VIRTIO_F_IN_ORDER feature is supported by the Virtio PMD,
let's use this optimized path instead.

Signed-off-by: Maxime Coquelin 
---
 drivers/net/virtio/virtio_ethdev.c  | 11 +-
 drivers/net/virtio/virtio_pci.h |  1 -
 drivers/net/virtio/virtio_rxtx.c| 28 +-
 drivers/net/virtio/virtio_rxtx_simple.c | 67 -
 drivers/net/virtio/virtio_rxtx_simple.h | 49 
 drivers/net/virtio/virtio_user_ethdev.c |  1 -
 6 files changed, 2 insertions(+), 155 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index 02980e3d1..73c428734 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1336,11 +1336,7 @@ set_rxtx_funcs(struct rte_eth_dev *eth_dev)
eth_dev->rx_pkt_burst = &virtio_recv_pkts;
}
 
-   if (hw->use_simple_tx) {
-   PMD_INIT_LOG(INFO, "virtio: using simple Tx path on port %u",
-   eth_dev->data->port_id);
-   eth_dev->tx_pkt_burst = virtio_xmit_pkts_simple;
-   } else if (hw->use_inorder_tx) {
+   if (hw->use_inorder_tx) {
PMD_INIT_LOG(INFO, "virtio: using inorder Tx path on port %u",
eth_dev->data->port_id);
eth_dev->tx_pkt_burst = virtio_xmit_pkts_inorder;
@@ -1881,12 +1877,9 @@ virtio_dev_configure(struct rte_eth_dev *dev)
rte_spinlock_init(&hw->state_lock);
 
hw->use_simple_rx = 1;
-   hw->use_simple_tx = 1;
 
if (vtpci_with_feature(hw, VIRTIO_F_IN_ORDER)) {
-   /* Simple Tx not compatible with in-order ring */
hw->use_inorder_tx = 1;
-   hw->use_simple_tx = 0;
if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) {
hw->use_inorder_rx = 1;
hw->use_simple_rx = 0;
@@ -1898,12 +1891,10 @@ virtio_dev_configure(struct rte_eth_dev *dev)
 #if defined RTE_ARCH_ARM64 || defined RTE_ARCH_ARM
if (!rte_cpu_get_flag_enabled(RTE_CPUFLAG_NEON)) {
hw->use_simple_rx = 0;
-   hw->use_simple_tx = 0;
}
 #endif
if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF)) {
 hw->use_simple_rx = 0;
-hw->use_simple_tx = 0;
}
 
if (rx_offloads & (DEV_RX_OFFLOAD_UDP_CKSUM |
diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h
index 77f805df6..0ec216874 100644
--- a/drivers/net/virtio/virtio_pci.h
+++ b/drivers/net/virtio/virtio_pci.h
@@ -238,7 +238,6 @@ struct virtio_hw {
uint8_t use_msix;
uint8_t modern;
uint8_t use_simple_rx;
-   uint8_t use_simple_tx;
uint8_t use_inorder_rx;
uint8_t use_inorder_tx;
uint16_tport_id;
diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net/virtio/virtio_rxtx.c
index 6394071b8..522e5dd41 100644
--- a/drivers/net/virtio/virtio_rxtx.c
+++ b/drivers/net/virtio/virtio_rxtx.c
@@ -708,10 +708,6 @@ virtio_dev_tx_queue_setup(struct rte_eth_dev *dev,
 
PMD_INIT_FUNC_TRACE();
 
-   /* cannot use simple rxtx funcs with multisegs or offloads */
-   if (dev->data->dev_conf.txmode.offloads)
-   hw->use_simple_tx = 0;
-
if (nb_desc == 0 || nb_desc > vq->vq_nentries)
nb_desc = vq->vq_nentries;
vq->vq_free_cnt = RTE_MIN(vq->vq_free_cnt, nb_desc);
@@ -746,33 +742,11 @@ virtio_dev_tx_queue_setup_finish(struct rte_eth_dev *dev,
uint8_t vtpci_queue_idx = 2 * queue_idx + VTNET_SQ_TQ_QUEUE_IDX;
struct virtio_hw *hw = dev->data->dev_private;
struct virtqueue *vq = hw->vqs[vtpci_queue_idx];
-   uint16_t mid_idx = vq->vq_nentries >> 1;
-   struct virtnet_tx *txvq = &vq->txq;
-   uint16_t desc_idx;
 
PMD_INIT_FUNC_TRACE();
 
-   if (hw->use_simple_tx) {
-   for (desc_idx = 0; desc_idx < mid_idx; desc_idx++) {
-   vq->vq_ring.avail->ring[desc_idx] =
-   desc_idx + mid_idx;
-   vq->vq_ring.desc[desc_idx + mid_idx].next =
-   desc_idx;
-   vq->vq_ring.desc[desc_idx + mid_idx].addr =
-   txvq->virtio_net_hdr_mem +
-   offsetof(struct virtio_tx_region, tx_hdr);
-   vq->vq_ring.desc[desc_idx + mid_idx].len =
-   vq->hw->vtnet_hdr_size;
-   vq->vq_ring.desc[desc_idx + mid_idx].flags =
-   VRING_DESC_F_NEXT;
-   vq->vq_ring.desc[desc_idx].flags = 0;
-   }
-   for (desc_idx = mid_idx; desc_idx < vq->vq_nentries;
-desc_idx++)
-   vq->vq_ring.avail->ring[desc_idx] = desc

[dpdk-dev] [PATCH v4 0/4] net/virtio: Tx simple path removal and offload improvements

[Maxime Coquelin]

In this v3, the main change is the removal of the Tx simple path.
Indeed, this path is not compliant with the Virtio specification,
so could cause problems with some host implementations.

Since Marvin has introduced the in-order path, we have a good
replacement for simple Tx and so we think it is better to
remove it than disabling it by defaul

Maxime Coquelin (4):
  net/virtio: remove simple Tx path
  net/virtio: improve Tx offload features negotiation
  net/virtio: don't use simple Rx if TCP LRO or VLAN strip
  net/virtio: improve offload check performance

 drivers/net/virtio/virtio_ethdev.c  | 49 --
 drivers/net/virtio/virtio_ethdev.h  |  3 --
 drivers/net/virtio/virtio_pci.h |  4 +-
 drivers/net/virtio/virtio_rxtx.c| 74 ++---
 drivers/net/virtio/virtio_rxtx_simple.c | 67 -
 drivers/net/virtio/virtio_rxtx_simple.h | 49 --
 drivers/net/virtio/virtio_user_ethdev.c |  1 -
 7 files changed, 51 insertions(+), 196 deletions(-)

-- 
2.14.4

[dpdk-dev] [PATCH v4 4/4] net/virtio: improve offload check performance

[Maxime Coquelin]

Instead of checking the multiple Virtio features bits for
every packet, let's do the check once at configure time and
store it in virtio_hw struct.

Reviewed-by: Tiwei Bie 
Signed-off-by: Maxime Coquelin 
---
 drivers/net/virtio/virtio_ethdev.c | 19 
 drivers/net/virtio/virtio_pci.h|  3 +++
 drivers/net/virtio/virtio_rxtx.c   | 46 +-
 3 files changed, 33 insertions(+), 35 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index 15d5b4f79..8c636b5f9 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1798,6 +1798,22 @@ rte_virtio_pmd_init(void)
rte_pci_register(&rte_virtio_pmd);
 }
 
+static bool
+rx_offload_enabled(struct virtio_hw *hw)
+{
+   return vtpci_with_feature(hw, VIRTIO_NET_F_GUEST_CSUM) ||
+   vtpci_with_feature(hw, VIRTIO_NET_F_GUEST_TSO4) ||
+   vtpci_with_feature(hw, VIRTIO_NET_F_GUEST_TSO6);
+}
+
+static bool
+tx_offload_enabled(struct virtio_hw *hw)
+{
+   return vtpci_with_feature(hw, VIRTIO_NET_F_CSUM) ||
+   vtpci_with_feature(hw, VIRTIO_NET_F_HOST_TSO4) ||
+   vtpci_with_feature(hw, VIRTIO_NET_F_HOST_TSO6);
+}
+
 /*
  * Configure virtio device
  * It returns 0 on success.
@@ -1877,6 +1893,9 @@ virtio_dev_configure(struct rte_eth_dev *dev)
return -ENOTSUP;
}
 
+   hw->has_tx_offload = tx_offload_enabled(hw);
+   hw->has_rx_offload = rx_offload_enabled(hw);
+
if (dev->data->dev_flags & RTE_ETH_DEV_INTR_LSC)
/* Enable vector (0) for Link State Intrerrupt */
if (VTPCI_OPS(hw)->set_config_irq(hw, 0) ==
diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h
index 0ec216874..58fdd3d45 100644
--- a/drivers/net/virtio/virtio_pci.h
+++ b/drivers/net/virtio/virtio_pci.h
@@ -6,6 +6,7 @@
 #define _VIRTIO_PCI_H_
 
 #include 
+#include 
 
 #include 
 #include 
@@ -240,6 +241,8 @@ struct virtio_hw {
uint8_t use_simple_rx;
uint8_t use_inorder_rx;
uint8_t use_inorder_tx;
+   boolhas_tx_offload;
+   boolhas_rx_offload;
uint16_tport_id;
uint8_t mac_addr[ETHER_ADDR_LEN];
uint32_tnotify_off_multiplier;
diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net/virtio/virtio_rxtx.c
index 522e5dd41..7d59bb5c1 100644
--- a/drivers/net/virtio/virtio_rxtx.c
+++ b/drivers/net/virtio/virtio_rxtx.c
@@ -347,13 +347,6 @@ virtio_tso_fix_cksum(struct rte_mbuf *m)
}
 }
 
-static inline int
-tx_offload_enabled(struct virtio_hw *hw)
-{
-   return vtpci_with_feature(hw, VIRTIO_NET_F_CSUM) ||
-   vtpci_with_feature(hw, VIRTIO_NET_F_HOST_TSO4) ||
-   vtpci_with_feature(hw, VIRTIO_NET_F_HOST_TSO6);
-}
 
 /* avoid write operation when necessary, to lessen cache issues */
 #define ASSIGN_UNLESS_EQUAL(var, val) do { \
@@ -364,7 +357,7 @@ tx_offload_enabled(struct virtio_hw *hw)
 static inline void
 virtqueue_xmit_offload(struct virtio_net_hdr *hdr,
struct rte_mbuf *cookie,
-   int offload)
+   bool offload)
 {
if (offload) {
if (cookie->ol_flags & PKT_TX_TCP_SEG)
@@ -421,14 +414,11 @@ virtqueue_enqueue_xmit_inorder(struct virtnet_tx *txvq,
struct virtio_net_hdr *hdr;
uint16_t idx;
uint16_t head_size = vq->hw->vtnet_hdr_size;
-   int offload;
uint16_t i = 0;
 
idx = vq->vq_desc_head_idx;
start_dp = vq->vq_ring.desc;
 
-   offload = tx_offload_enabled(vq->hw);
-
while (i < num) {
idx = idx & (vq->vq_nentries - 1);
dxp = &vq->vq_descx[idx];
@@ -440,7 +430,7 @@ virtqueue_enqueue_xmit_inorder(struct virtnet_tx *txvq,
cookies[i]->pkt_len -= head_size;
 
/* if offload disabled, it is not zeroed below, do it now */
-   if (offload == 0) {
+   if (!vq->hw->has_tx_offload) {
ASSIGN_UNLESS_EQUAL(hdr->csum_start, 0);
ASSIGN_UNLESS_EQUAL(hdr->csum_offset, 0);
ASSIGN_UNLESS_EQUAL(hdr->flags, 0);
@@ -449,7 +439,8 @@ virtqueue_enqueue_xmit_inorder(struct virtnet_tx *txvq,
ASSIGN_UNLESS_EQUAL(hdr->hdr_len, 0);
}
 
-   virtqueue_xmit_offload(hdr, cookies[i], offload);
+   virtqueue_xmit_offload(hdr, cookies[i],
+   vq->hw->has_tx_offload);
 
start_dp[idx].addr  = VIRTIO_MBUF_DATA_DMA_ADDR(cookies[i], vq);
start_dp[idx].len   = cookies[i]->data_len;
@@ -478,9 +469,6 @@ virtqueue_enqueue_xmit(struct virtnet_tx *txvq, struct 
rte_mbuf *cookie,
uint16_t head_idx, idx;
uint16_t head_size = vq->hw->vtnet_hdr_size;
struct virtio_net_hdr *hdr;
-   int offload;
-
-  

[dpdk-dev] [PATCH v4 2/4] net/virtio: improve Tx offload features negotiation

[Maxime Coquelin]

This patch improves the Tx offload features selection depending
on whether the application request for offloads.

When the application doesn't request for Tx offload features,
the corresponding features bits aren't negotiated.

Signed-off-by: Maxime Coquelin 
---
 drivers/net/virtio/virtio_ethdev.c | 15 +--
 drivers/net/virtio/virtio_ethdev.h |  3 ---
 2 files changed, 13 insertions(+), 5 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index 73c428734..1d223d029 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1806,8 +1806,10 @@ static int
 virtio_dev_configure(struct rte_eth_dev *dev)
 {
const struct rte_eth_rxmode *rxmode = &dev->data->dev_conf.rxmode;
+   const struct rte_eth_txmode *txmode = &dev->data->dev_conf.txmode;
struct virtio_hw *hw = dev->data->dev_private;
uint64_t rx_offloads = rxmode->offloads;
+   uint64_t tx_offloads = txmode->offloads;
uint64_t req_features;
int ret;
 
@@ -1829,6 +1831,15 @@ virtio_dev_configure(struct rte_eth_dev *dev)
(1ULL << VIRTIO_NET_F_GUEST_TSO4) |
(1ULL << VIRTIO_NET_F_GUEST_TSO6);
 
+   if (tx_offloads & (DEV_TX_OFFLOAD_UDP_CKSUM |
+  DEV_TX_OFFLOAD_TCP_CKSUM))
+   req_features |= (1ULL << VIRTIO_NET_F_CSUM);
+
+   if (tx_offloads & DEV_TX_OFFLOAD_TCP_TSO)
+   req_features |=
+   (1ULL << VIRTIO_NET_F_HOST_TSO4) |
+   (1ULL << VIRTIO_NET_F_HOST_TSO6);
+
/* if request features changed, reinit the device */
if (req_features != hw->req_guest_features) {
ret = virtio_init_device(dev, req_features);
@@ -2155,14 +2166,14 @@ virtio_dev_info_get(struct rte_eth_dev *dev, struct 
rte_eth_dev_info *dev_info)
 
dev_info->tx_offload_capa = DEV_TX_OFFLOAD_MULTI_SEGS |
DEV_TX_OFFLOAD_VLAN_INSERT;
-   if (hw->guest_features & (1ULL << VIRTIO_NET_F_CSUM)) {
+   if (host_features & (1ULL << VIRTIO_NET_F_CSUM)) {
dev_info->tx_offload_capa |=
DEV_TX_OFFLOAD_UDP_CKSUM |
DEV_TX_OFFLOAD_TCP_CKSUM;
}
tso_mask = (1ULL << VIRTIO_NET_F_HOST_TSO4) |
(1ULL << VIRTIO_NET_F_HOST_TSO6);
-   if ((hw->guest_features & tso_mask) == tso_mask)
+   if ((host_features & tso_mask) == tso_mask)
dev_info->tx_offload_capa |= DEV_TX_OFFLOAD_TCP_TSO;
 }
 
diff --git a/drivers/net/virtio/virtio_ethdev.h 
b/drivers/net/virtio/virtio_ethdev.h
index 350e9ce73..f4d09df71 100644
--- a/drivers/net/virtio/virtio_ethdev.h
+++ b/drivers/net/virtio/virtio_ethdev.h
@@ -28,9 +28,6 @@
 1u << VIRTIO_NET_F_CTRL_VQ   | \
 1u << VIRTIO_NET_F_CTRL_RX   | \
 1u << VIRTIO_NET_F_CTRL_VLAN | \
-1u << VIRTIO_NET_F_CSUM  | \
-1u << VIRTIO_NET_F_HOST_TSO4 | \
-1u << VIRTIO_NET_F_HOST_TSO6 | \
 1u << VIRTIO_NET_F_MRG_RXBUF | \
 1u << VIRTIO_NET_F_MTU | \
 1ULL << VIRTIO_NET_F_GUEST_ANNOUNCE |  \
-- 
2.14.4

[dpdk-dev] [PATCH v4 3/4] net/virtio: don't use simple Rx if TCP LRO or VLAN strip

[Maxime Coquelin]

Reviewed-by: Tiwei Bie 
Signed-off-by: Maxime Coquelin 
---
 drivers/net/virtio/virtio_ethdev.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index 1d223d029..15d5b4f79 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1909,7 +1909,9 @@ virtio_dev_configure(struct rte_eth_dev *dev)
}
 
if (rx_offloads & (DEV_RX_OFFLOAD_UDP_CKSUM |
-  DEV_RX_OFFLOAD_TCP_CKSUM))
+  DEV_RX_OFFLOAD_TCP_CKSUM |
+  DEV_RX_OFFLOAD_TCP_LRO |
+  DEV_RX_OFFLOAD_VLAN_STRIP))
hw->use_simple_rx = 0;
 
return 0;
-- 
2.14.4

Re: [dpdk-dev] [PATCH] ethdev: fix queue mapping documentation

[Andrew Rybchenko]

On 07/02/2018 06:08 PM, Ferruh Yigit wrote:

On 6/29/2018 10:44 AM, Jerin Jacob wrote:

The RTE_MAX_ETHPORT_QUEUE_STATS_MAPS does not exists, change
to the correct definition(RTE_ETHDEV_QUEUE_STAT_CNTRS)

Fixes: 5de201df8927 ("ethdev: add stats per queue")
Cc: sta...@dpdk.org

Signed-off-by: Jerin Jacob 
---
  lib/librte_ethdev/rte_ethdev.h | 4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/lib/librte_ethdev/rte_ethdev.h b/lib/librte_ethdev/rte_ethdev.h
index 36e3984ea..375ea24ce 100644
--- a/lib/librte_ethdev/rte_ethdev.h
+++ b/lib/librte_ethdev/rte_ethdev.h
@@ -2144,7 +2144,7 @@ void rte_eth_xstats_reset(uint16_t port_id);
   * @param stat_idx
   *   The per-queue packet statistics functionality number that the transmit
   *   queue is to be assigned.
- *   The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
+ *   The value must be in the range [0, RTE_ETHDEV_QUEUE_STAT_CNTRS - 1].

Yes RTE_MAX_ETHPORT_QUEUE_STATS_MAPS doesn't exits and comment is wrong, but
RTE_ETHDEV_QUEUE_STAT_CNTRS also slightly not correct.

I think how testpmd uses it increase the confusion.

In ixgbe there is no stats registers per queue, 128 queues are represented by 16
register set. stat_idx here is the index of that 16 registers. You map queue to
stats requester to get queue stats.

Also there is RTE_ETHDEV_QUEUE_STAT_CNTRS config in the ethdev API, which is the
hardcoded size of the queue stats, its default value is 16. This limits number
of the queues we can get stats from but saves allocated space. (Why not 
dynamic?)

You can increase the RTE_ETHDEV_QUEUE_STAT_CNTRS to the max supported number of
queue and ethdev code will be all valid. But "stat_idx" can't go beyond 16 (for
ixgbe) because it is hardware limitation and it may change from hw to hw.

Also technically it should be possible to reduce RTE_ETHDEV_QUEUE_STAT_CNTRS to
a low number, like 2, but in ixgbe map two queues into stat registers 14 & 15
and display those two set as queue stat 0 and 1. It seems current implementation
prevents this and forces the queues mapped should be less than
RTE_ETHDEV_QUEUE_STAT_CNTRS. Overall it seems there is a mixed used of
RTE_ETHDEV_QUEUE_STAT_CNTRS and stats queue index values, I assume because both
are same values.

I suggest updating it as:
"
The value must be in the range:
  [0 - MIN(HW Stat Registers Size, RTE_ETHDEV_QUEUE_STAT_CNTRS) - 1]
"


Technically I think it is not a problem to specify more than HW supports.
The function should simply return error. RTE_ETHDEV_QUEUE_STAT_CNTRS is
a hard limit which should be checked by ethdev.
The reasonable next question is how to find out what is the maximum for 
PMD/HW.

I think it deserves entry in dev_info. May be not now.
"HW Stats Registers size" is too HW specific. It could be not HW, but 
the PMD
limitation and limits for Rx and Tx could be different. So, may be 
something like:

"Device max per Tx queue stats"


   * @return
   *   Zero if successful. Non-zero otherwise.
   */
@@ -2164,7 +2164,7 @@ int rte_eth_dev_set_tx_queue_stats_mapping(uint16_t 
port_id,
   * @param stat_idx
   *   The per-queue packet statistics functionality number that the receive
   *   queue is to be assigned.
- *   The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 1].
+ *   The value must be in the range [0, RTE_ETHDEV_QUEUE_STAT_CNTRS - 1].
   * @return
   *   Zero if successful. Non-zero otherwise.
   */

[dpdk-dev] [PATCH v1] examples/multi_process: fix build error

[Emma Kenny]

Fix bug with undeclared variable name and
calling a variable that is not member of struct.

CC main.o
l2fwd_fork/main.c:  In function ‘main’: l2fwd_fork/main.c:1043:33:
 error: ‘dev_info’ undeclared (first use in this function)
 rte_eth_dev_info_get(portid, &dev_info);

l2fwd_fork/main.c:1043:33: note: each undeclared identifier is
reported only once for each function it appears in
l2fwd_fork/main.c:1077:11: error: ‘struct rte_eth_txconf’
has no member named ‘tx_offloads’
   txq_conf.tx_offloads = local_port_conf.txmode.offloads;

make[1]: *** [main.o] Error 1
make: *** [all] Error 2

Fixes: f8c02ca878af ("examples/multi_process: convert to new ethdev offloads 
API")
Cc: shah...@mellanox.com

Signed-off-by: Emma Kenny 
---
 examples/multi_process/l2fwd_fork/main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/examples/multi_process/l2fwd_fork/main.c 
b/examples/multi_process/l2fwd_fork/main.c
index 94318ab..62e3937 100644
--- a/examples/multi_process/l2fwd_fork/main.c
+++ b/examples/multi_process/l2fwd_fork/main.c
@@ -1030,6 +1030,7 @@ main(int argc, char **argv)
struct rte_eth_rxconf rxq_conf;
struct rte_eth_txconf txq_conf;
struct rte_eth_conf local_port_conf = port_conf;
+   struct rte_eth_dev_info dev_info;
 
/* skip ports that are not enabled */
if ((l2fwd_enabled_port_mask & (1 << portid)) == 0) {
@@ -1074,7 +1075,7 @@ main(int argc, char **argv)
fflush(stdout);
txq_conf = dev_info.default_txconf;
txq_conf.txq_flags = ETH_TXQ_FLAGS_IGNORE;
-   txq_conf.tx_offloads = local_port_conf.txmode.offloads;
+   txq_conf.offloads = local_port_conf.txmode.offloads;
ret = rte_eth_tx_queue_setup(portid, 0, nb_txd,
rte_eth_dev_socket_id(portid),
&txq_conf);
-- 
2.9.5

--
Intel Research and Development Ireland Limited
Registered in Ireland
Registered Office: Collinstown Industrial Park, Leixlip, County Kildare
Registered Number: 308263


This e-mail and any attachments may contain confidential material for the sole
use of the intended recipient(s). Any review or distribution by others is
strictly prohibited. If you are not the intended recipient, please contact the
sender and delete all copies.

Re: [dpdk-dev] [PATCH] ethdev: fix queue mapping documentation

[Ferruh Yigit]

On 7/2/2018 4:32 PM, Andrew Rybchenko wrote:
> On 07/02/2018 06:08 PM, Ferruh Yigit wrote:
>> On 6/29/2018 10:44 AM, Jerin Jacob wrote:
>>> The RTE_MAX_ETHPORT_QUEUE_STATS_MAPS does not exists, change
>>> to the correct definition(RTE_ETHDEV_QUEUE_STAT_CNTRS)
>>>
>>> Fixes: 5de201df8927 ("ethdev: add stats per queue")
>>> Cc: sta...@dpdk.org
>>>
>>> Signed-off-by: Jerin Jacob 
>>> ---
>>>  lib/librte_ethdev/rte_ethdev.h | 4 ++--
>>>  1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/lib/librte_ethdev/rte_ethdev.h b/lib/librte_ethdev/rte_ethdev.h
>>> index 36e3984ea..375ea24ce 100644
>>> --- a/lib/librte_ethdev/rte_ethdev.h
>>> +++ b/lib/librte_ethdev/rte_ethdev.h
>>> @@ -2144,7 +2144,7 @@ void rte_eth_xstats_reset(uint16_t port_id);
>>>   * @param stat_idx
>>>   *   The per-queue packet statistics functionality number that the transmit
>>>   *   queue is to be assigned.
>>> - *   The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 
>>> 1].
>>> + *   The value must be in the range [0, RTE_ETHDEV_QUEUE_STAT_CNTRS - 1].
>> Yes RTE_MAX_ETHPORT_QUEUE_STATS_MAPS doesn't exits and comment is wrong, but
>> RTE_ETHDEV_QUEUE_STAT_CNTRS also slightly not correct.
>>
>> I think how testpmd uses it increase the confusion.
>>
>> In ixgbe there is no stats registers per queue, 128 queues are represented 
>> by 16
>> register set. stat_idx here is the index of that 16 registers. You map queue 
>> to
>> stats requester to get queue stats.
>>
>> Also there is RTE_ETHDEV_QUEUE_STAT_CNTRS config in the ethdev API, which is 
>> the
>> hardcoded size of the queue stats, its default value is 16. This limits 
>> number
>> of the queues we can get stats from but saves allocated space. (Why not 
>> dynamic?)
>>
>> You can increase the RTE_ETHDEV_QUEUE_STAT_CNTRS to the max supported number 
>> of
>> queue and ethdev code will be all valid. But "stat_idx" can't go beyond 16 
>> (for
>> ixgbe) because it is hardware limitation and it may change from hw to hw.
>>
>> Also technically it should be possible to reduce RTE_ETHDEV_QUEUE_STAT_CNTRS 
>> to
>> a low number, like 2, but in ixgbe map two queues into stat registers 14 & 15
>> and display those two set as queue stat 0 and 1. It seems current 
>> implementation
>> prevents this and forces the queues mapped should be less than
>> RTE_ETHDEV_QUEUE_STAT_CNTRS. Overall it seems there is a mixed used of
>> RTE_ETHDEV_QUEUE_STAT_CNTRS and stats queue index values, I assume because 
>> both
>> are same values.
>>
>> I suggest updating it as:
>> "
>> The value must be in the range:
>>  [0 - MIN(HW Stat Registers Size, RTE_ETHDEV_QUEUE_STAT_CNTRS) - 1]
>> "
> 
> Technically I think it is not a problem to specify more than HW supports.
> The function should simply return error.  RTE_ETHDEV_QUEUE_STAT_CNTRS is
> a hard limit which should be checked by ethdev.
> The reasonable next question is how to find out what is the maximum for 
> PMD/HW.
> I think it deserves entry in dev_info. May be not now.

Yes there is not a way to find out that limit by application, setting
RTE_ETHDEV_QUEUE_STAT_CNTRS to 16 and using it as limit solving the issue for 
now :)

> "HW Stats Registers size" is too HW specific. It could be not HW, but the PMD
> limitation and limits for Rx and Tx could be different. So, may be something 
> like:
> "Device max per Tx queue stats"
> 
>>>   * @return
>>>   *   Zero if successful. Non-zero otherwise.
>>>   */
>>> @@ -2164,7 +2164,7 @@ int rte_eth_dev_set_tx_queue_stats_mapping(uint16_t 
>>> port_id,
>>>   * @param stat_idx
>>>   *   The per-queue packet statistics functionality number that the receive
>>>   *   queue is to be assigned.
>>> - *   The value must be in the range [0, RTE_MAX_ETHPORT_QUEUE_STATS_MAPS - 
>>> 1].
>>> + *   The value must be in the range [0, RTE_ETHDEV_QUEUE_STAT_CNTRS - 1].
>>>   * @return
>>>   *   Zero if successful. Non-zero otherwise.
>>>   */
>>>
> 

Re: [dpdk-dev] [PATCH v2 00/23] bnxt patchset

[Ferruh Yigit]

On 6/28/2018 9:15 PM, Ajit Khaparde wrote:
> Patchset against dpdk-next-net. Please apply.
> 
> v1->v2:
> Takes care of the various comments made in the previous version.
> I am dropping the style changes for now. I will send them later
> after addressing the coding convention issues.
> 
> 
> Ajit Khaparde (16):
>   net/bnxt: fix clear port stats
>   net/bnxt: add Tx batching support
>   net/bnxt: optimize receive processing code
>   net/bnxt: set MIN/MAX descriptor count fox Tx and Rx Rings
>   net/bnxt: fix dev close operation
>   net/bnxt: set ring coalesce parameters for Stratus NIC
>   net/bnxt: fix HW Tx checksum offload check
>   net/bnxt: add support for VF id 0xd800
>   net/bnxt: fix Rx/Tx queue start/stop operations
>   net/bnxt: refactor filter/flow
>   net/bnxt: check filter type before clearing it
>   net/bnxt: fix set MTU
>   net/bnxt: fix incorrect IO address handling in Tx
>   net/bnxt: allocate RSS context only if RSS mode is enabled
>   net/bnxt: check VF resources if resource manager is enabled
>   net/bnxt: fix Rx ring count limitation
> 
> Jay Ding (1):
>   net/bnxt: check for invalid vnic id
> 
> Rob Miller (1):
>   net/bnxt: update HWRM API to v1.9.2.9
> 
> Scott Branden (1):
>   net/bnxt: move function check zero bytes to bnxt util.h
> 
> Somnath Kotur (3):
>   net/bnxt: revert reset of L2 filter id
>   net/bnxt: fix to move a flow to a different queue
>   net/bnxt: use correct flags during VLAN configuration
> 
> Xiaoxin Peng (1):
>   net/bnxt: fix Tx with multiple mbuf

Series applied to dpdk-next-net/master, thanks.


Build errors commented on patch fixed while applying, please double check.

Re: [dpdk-dev] [PATCH v2] cryptodev: remove RTE_LIBRTE_CRYPTODEV_DEBUG

[De Lara Guarch, Pablo]

> -Original Message-
> From: Parthasarathy, JananeeX M
> Sent: Monday, July 2, 2018 1:31 PM
> To: dev@dpdk.org
> Cc: De Lara Guarch, Pablo ; Pattan, Reshma
> ; Parthasarathy, JananeeX M
> 
> Subject: [PATCH v2] cryptodev: remove RTE_LIBRTE_CRYPTODEV_DEBUG
> 
> From: Jananee Parthasarathy 
> 
> For librte_cryptodev dynamic logging, conditional compilation of debug logs
> would not be required anymore.
> 
> Signed-off-by: Jananee Parthasarathy 
> Reviewed-by: Reshma Pattan 
> Reviewed-by: Pablo de Lara Guarch 
> ---
> v2: unused macro removed from config/common_base and rte_dev.h

Applied to dpdk-next-crypto.
Thanks,

Pablo

Re: [dpdk-dev] [PATCH v1] examples/multi_process: fix build error

[Ferruh Yigit]

On 7/2/2018 4:40 PM, Emma Kenny wrote:
> Fix bug with undeclared variable name and
> calling a variable that is not member of struct.
> 
> CC main.o
> l2fwd_fork/main.c:  In function ‘main’: l2fwd_fork/main.c:1043:33:
>  error: ‘dev_info’ undeclared (first use in this function)
>  rte_eth_dev_info_get(portid, &dev_info);
> 
> l2fwd_fork/main.c:1043:33: note: each undeclared identifier is
> reported only once for each function it appears in
> l2fwd_fork/main.c:1077:11: error: ‘struct rte_eth_txconf’
> has no member named ‘tx_offloads’
>txq_conf.tx_offloads = local_port_conf.txmode.offloads;
> 
> make[1]: *** [main.o] Error 1
> make: *** [all] Error 2
> 
> Fixes: f8c02ca878af ("examples/multi_process: convert to new ethdev offloads 
> API")
> Cc: shah...@mellanox.com
> 
> Signed-off-by: Emma Kenny 

Cc: sta...@dpdk.org

Acked-by: Ferruh Yigit 

[dpdk-dev] [PATCH] maintainers: update for mempool

[Olivier Matz]

Given its very good contributions to this library, add Andrew as
official maintainer for librte_mempool.

Signed-off-by: Olivier Matz 
---
 MAINTAINERS | 1 +
 1 file changed, 1 insertion(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index dabb12d65..7781e43c5 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -277,6 +277,7 @@ Core Libraries
 
 Memory pool
 M: Olivier Matz 
+M: Andrew Rybchenko 
 F: lib/librte_mempool/
 F: drivers/mempool/Makefile
 F: drivers/mempool/ring/
-- 
2.11.0

Re: [dpdk-dev] [PATCH] maintainers: update for mempool

[Andrew Rybchenko]

On 07/02/2018 06:55 PM, Olivier Matz wrote:

Given its very good contributions to this library, add Andrew as
official maintainer for librte_mempool.

Signed-off-by: Olivier Matz 
---
  MAINTAINERS | 1 +
  1 file changed, 1 insertion(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index dabb12d65..7781e43c5 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -277,6 +277,7 @@ Core Libraries
  
  Memory pool

  M: Olivier Matz 
+M: Andrew Rybchenko 
  F: lib/librte_mempool/
  F: drivers/mempool/Makefile
  F: drivers/mempool/ring/


Acked-by: Andrew Rybchenko 

Re: [dpdk-dev] [PATCH v5 7/9] net/virtio: support in-order Rx and Tx

[Ferruh Yigit]

On 7/2/2018 2:56 PM, Marvin Liu wrote:
> IN_ORDER Rx function depends on merge-able feature. Descriptors
> allocation and free will be done in bulk.
> 
> Virtio dequeue logic:
> dequeue_burst_rx(burst mbufs)
> for (each mbuf b) {
> if (b need merge) {
> merge remained mbufs
> add merged mbuf to return mbufs list
> } else {
> add mbuf to return mbufs list
> }
> }
> if (last mbuf c need merge) {
> dequeue_burst_rx(required mbufs)
> merge last mbuf c
> }
> refill_avail_ring_bulk()
> update_avail_ring()
> return mbufs list
> 
> IN_ORDER Tx function can support offloading features. Packets which
> matched "can_push" option will be handled by simple xmit function. Those
> packets can't match "can_push" will be handled by original xmit function
> with in-order flag.
> 
> Virtio enqueue logic:
> xmit_cleanup(used descs)
> for (each xmit mbuf b) {
> if (b can inorder xmit) {
> add mbuf b to inorder burst list
> continue
> } else {
> xmit inorder burst list
> xmit mbuf b by original function
> }
> }
> if (inorder burst list not empty) {
> xmit inorder burst list
> }
> update_avail_ring()
> 
> Signed-off-by: Marvin Liu 
> Reviewed-by: Maxime Coquelin 

<...>

> @@ -150,6 +188,83 @@ virtio_xmit_cleanup(struct virtqueue *vq, uint16_t num)
>   }
>  }
>  
> +/* Cleanup from completed inorder transmits. */
> +static void
> +virtio_xmit_cleanup_inorder(struct virtqueue *vq, uint16_t num)
> +{
> + uint16_t i, used_idx, desc_idx, last_idx;


Getting following build error [1], from code it looks like false positive, but
to get rid of the build error would it be OK to set initial value to "desc_idx"?


[1]
.../dpdk/drivers/net/virtio/virtio_rxtx.c:195:24: error: ‘desc_idx’ may be used
uninitialized in this function [-Werror=maybe-uninitialized]


  uint16_t i, used_idx, desc_idx, last_idx;



^~~~

[dpdk-dev] [PATCH v2] librte_lpm: Improve performance of the delete and add functions

[Alex Kiselev]

There are two major problems with the library:
first, there is no need to rebuild the whole LPM tree
when a rule is deleted and second, due to the current
rules algorithm with complexity O(n) it's almost
impossible to deal with large rule sets (50k or so rules).
This patch addresses those two issues.

Signed-off-by: Alex Kiselev 
---
 lib/librte_lpm/rte_lpm6.c | 1073 ++---
 1 file changed, 816 insertions(+), 257 deletions(-)

diff --git a/lib/librte_lpm/rte_lpm6.c b/lib/librte_lpm/rte_lpm6.c
index 149677eb1..438db0831 100644
--- a/lib/librte_lpm/rte_lpm6.c
+++ b/lib/librte_lpm/rte_lpm6.c
@@ -21,6 +21,10 @@
 #include 
 #include 
 #include 
+#include 
+#include 
+#include 
+#include 
 
 #include "rte_lpm6.h"
 
@@ -37,6 +41,9 @@
 #define BYTE_SIZE 8
 #define BYTES2_SIZE  16
 
+#define RULE_HASH_TABLE_EXTRA_SPACE 256
+#define TBL24_INDUINT32_MAX
+
 #define lpm6_tbl8_gindex next_hop
 
 /** Flags for setting an entry as valid/invalid. */
@@ -70,6 +77,23 @@ struct rte_lpm6_rule {
uint8_t depth; /**< Rule depth. */
 };
 
+/** Rules tbl entry key. */
+struct rte_lpm6_rule_key {
+   uint8_t ip[RTE_LPM6_IPV6_ADDR_SIZE]; /**< Rule IP address. */
+   uint8_t depth; /**< Rule depth. */
+};
+
+/* Header of tbl8 */
+struct rte_lpm_tbl8_hdr {
+   uint32_t owner_tbl_ind; /**< owner table: TBL24_IND if owner is tbl24,
+   * otherwise index of tbl8
+   */
+   uint32_t owner_entry_ind; /**< index of the owner table entry where
+   * pointer to the tbl8 is stored
+   */
+   uint32_t ref_cnt; /**< table reference counter */
+};
+
 /** LPM6 structure. */
 struct rte_lpm6 {
/* LPM metadata. */
@@ -77,12 +101,18 @@ struct rte_lpm6 {
uint32_t max_rules;  /**< Max number of rules. */
uint32_t used_rules; /**< Used rules so far. */
uint32_t number_tbl8s;   /**< Number of tbl8s to allocate. */
-   uint32_t next_tbl8;  /**< Next tbl8 to be used. */
 
/* LPM Tables. */
-   struct rte_lpm6_rule *rules_tbl; /**< LPM rules. */
+   struct rte_mempool *rules_pool; /**< LPM rules mempool. */
+   struct rte_hash *rules_tbl; /**< LPM rules. */
struct rte_lpm6_tbl_entry tbl24[RTE_LPM6_TBL24_NUM_ENTRIES]
__rte_cache_aligned; /**< LPM tbl24 table. */
+
+   uint32_t *tbl8_pool; /**< pool of indexes of free tbl8s */
+   uint32_t tbl8_pool_pos; /**< current position in the tbl8 pool */
+
+   struct rte_lpm_tbl8_hdr *tbl8_hdrs; /* array of tbl8 headers */
+
struct rte_lpm6_tbl_entry tbl8[0]
__rte_cache_aligned; /**< LPM tbl8 table. */
 };
@@ -93,22 +123,130 @@ struct rte_lpm6 {
  * and set the rest to 0.
  */
 static inline void
-mask_ip(uint8_t *ip, uint8_t depth)
+ip6_mask_addr(uint8_t *ip, uint8_t depth)
 {
-int16_t part_depth, mask;
-int i;
+   int16_t part_depth, mask;
+   int i;
 
-   part_depth = depth;
+   part_depth = depth;
 
-   for (i = 0; i < RTE_LPM6_IPV6_ADDR_SIZE; i++) {
-   if (part_depth < BYTE_SIZE && part_depth >= 0) {
-   mask = (uint16_t)(~(UINT8_MAX >> part_depth));
-   ip[i] = (uint8_t)(ip[i] & mask);
-   } else if (part_depth < 0) {
-   ip[i] = 0;
-   }
-   part_depth -= BYTE_SIZE;
-   }
+   for (i = 0; i < RTE_LPM6_IPV6_ADDR_SIZE; i++) {
+   if (part_depth < BYTE_SIZE && part_depth >= 0) {
+   mask = (uint16_t)(~(UINT8_MAX >> part_depth));
+   ip[i] = (uint8_t)(ip[i] & mask);
+   } else if (part_depth < 0)
+   ip[i] = 0;
+
+   part_depth -= BYTE_SIZE;
+   }
+}
+
+/* copy ipv6 address */
+static inline void
+ip6_copy_addr(uint8_t *dst, const uint8_t *src)
+{
+   rte_memcpy(dst, src, RTE_LPM6_IPV6_ADDR_SIZE);
+}
+
+/*
+ * LPM6 rule hash function
+ */
+static inline uint32_t
+rule_hash_crc(const void *data, __rte_unused uint32_t data_len,
+ uint32_t init_val)
+{
+   return rte_hash_crc(data, sizeof(struct rte_lpm6_rule_key), init_val);
+}
+
+/*
+ * Init pool of free tbl8 indexes
+ */
+static void
+tbl8_pool_init(struct rte_lpm6 *lpm)
+{
+   /* put entire range of indexes to the tbl8 pool */
+   uint32_t i;
+   for (i = 0; i < lpm->number_tbl8s; i++)
+   lpm->tbl8_pool[i] = i;
+
+   lpm->tbl8_pool_pos = 0;
+}
+
+/*
+ * Get an index of a free tbl8 from the pool
+ */
+static inline uint32_t
+tbl8_get(struct rte_lpm6 *lpm, uint32_t *tbl8_ind)
+{
+   if (lpm->tbl8_pool_pos == lpm

Re: [dpdk-dev] [PATCH v5 7/9] net/virtio: support in-order Rx and Tx

[Ferruh Yigit]

On 7/2/2018 5:41 PM, Ferruh Yigit wrote:
> On 7/2/2018 2:56 PM, Marvin Liu wrote:
>> IN_ORDER Rx function depends on merge-able feature. Descriptors
>> allocation and free will be done in bulk.
>>
>> Virtio dequeue logic:
>> dequeue_burst_rx(burst mbufs)
>> for (each mbuf b) {
>> if (b need merge) {
>> merge remained mbufs
>> add merged mbuf to return mbufs list
>> } else {
>> add mbuf to return mbufs list
>> }
>> }
>> if (last mbuf c need merge) {
>> dequeue_burst_rx(required mbufs)
>> merge last mbuf c
>> }
>> refill_avail_ring_bulk()
>> update_avail_ring()
>> return mbufs list
>>
>> IN_ORDER Tx function can support offloading features. Packets which
>> matched "can_push" option will be handled by simple xmit function. Those
>> packets can't match "can_push" will be handled by original xmit function
>> with in-order flag.
>>
>> Virtio enqueue logic:
>> xmit_cleanup(used descs)
>> for (each xmit mbuf b) {
>> if (b can inorder xmit) {
>> add mbuf b to inorder burst list
>> continue
>> } else {
>> xmit inorder burst list
>> xmit mbuf b by original function
>> }
>> }
>> if (inorder burst list not empty) {
>> xmit inorder burst list
>> }
>> update_avail_ring()
>>
>> Signed-off-by: Marvin Liu 
>> Reviewed-by: Maxime Coquelin 
> 
> <...>
> 
>> @@ -150,6 +188,83 @@ virtio_xmit_cleanup(struct virtqueue *vq, uint16_t num)
>>  }
>>  }
>>  
>> +/* Cleanup from completed inorder transmits. */
>> +static void
>> +virtio_xmit_cleanup_inorder(struct virtqueue *vq, uint16_t num)
>> +{
>> +uint16_t i, used_idx, desc_idx, last_idx;
> 
> 
> Getting following build error [1], from code it looks like false positive, but
> to get rid of the build error would it be OK to set initial value to 
> "desc_idx"?

I applied this while merging, if this is wrong please let me know, we can fix in
next-net, Thanks.

> 
> 
> [1]
> .../dpdk/drivers/net/virtio/virtio_rxtx.c:195:24: error: ‘desc_idx’ may be 
> used
> uninitialized in this function [-Werror=maybe-uninitialized]
> 
> 
>   uint16_t i, used_idx, desc_idx, last_idx;
> 
> 
> 
> ^~~~
> 

Re: [dpdk-dev] [PATCH v5 7/9] net/virtio: support in-order Rx and Tx

[Maxime Coquelin]

On 07/02/2018 06:52 PM, Ferruh Yigit wrote:

On 7/2/2018 5:41 PM, Ferruh Yigit wrote:

On 7/2/2018 2:56 PM, Marvin Liu wrote:

IN_ORDER Rx function depends on merge-able feature. Descriptors
allocation and free will be done in bulk.

Virtio dequeue logic:
 dequeue_burst_rx(burst mbufs)
 for (each mbuf b) {
 if (b need merge) {
 merge remained mbufs
 add merged mbuf to return mbufs list
 } else {
 add mbuf to return mbufs list
 }
 }
 if (last mbuf c need merge) {
 dequeue_burst_rx(required mbufs)
 merge last mbuf c
 }
 refill_avail_ring_bulk()
 update_avail_ring()
 return mbufs list

IN_ORDER Tx function can support offloading features. Packets which
matched "can_push" option will be handled by simple xmit function. Those
packets can't match "can_push" will be handled by original xmit function
with in-order flag.

Virtio enqueue logic:
 xmit_cleanup(used descs)
 for (each xmit mbuf b) {
 if (b can inorder xmit) {
 add mbuf b to inorder burst list
 continue
 } else {
 xmit inorder burst list
 xmit mbuf b by original function
 }
 }
 if (inorder burst list not empty) {
 xmit inorder burst list
 }
 update_avail_ring()

Signed-off-by: Marvin Liu 
Reviewed-by: Maxime Coquelin 


<...>


@@ -150,6 +188,83 @@ virtio_xmit_cleanup(struct virtqueue *vq, uint16_t num)
}
  }
  
+/* Cleanup from completed inorder transmits. */

+static void
+virtio_xmit_cleanup_inorder(struct virtqueue *vq, uint16_t num)
+{
+   uint16_t i, used_idx, desc_idx, last_idx;



Getting following build error [1], from code it looks like false positive, but
to get rid of the build error would it be OK to set initial value to "desc_idx"?


I applied this while merging, if this is wrong please let me know, we can fix in
next-net, Thanks.


Looks good to me. I didn't catch it with the GCC version I use.

Thanks,
Maxime




[1]
.../dpdk/drivers/net/virtio/virtio_rxtx.c:195:24: error: ‘desc_idx’ may be used
uninitialized in this function [-Werror=maybe-uninitialized]


   uint16_t i, used_idx, desc_idx, last_idx;



 ^~~~

[dpdk-dev] [PATCH v2 0/6] compress: add Octeontx ZIP compression PMD

[Shally Verma]

This patch series add compression PMD for cavium octeontx ZIP module
in DPDK compress drivers.
Currently PMD only tested for deflate, stateless compression
and decompression with direct memory buffers.

Changes in v2:
- enable OCTEONTX_ZIPVF bydefault, and remove static debug flag
- fix meson build in compress/driver/octeontx
- fix 32-bit compiler error
- remove global declared data structure
- header files in alphabetical order
- update doc with correct feature support

V1 support:
Octeontx ZIP PMD implementation
Confuguration and Makefile changes to build Octeontx ZIP PMD
Documentation ZIP PMD
Build support in driver/compress meson for Octeontx ZIP PMD

TBDs:
Scatter Gather support,
Stateful compression/decompression,
test for performance

This patchset is dependent upon compressdev API.

Ashish Gupta (6):
  compress/octeontx: add octeontx zip PMD support
  compress/octeontx: add device setup PMD ops
  compress/octeontx: add xform and stream create support
  compress/octeontx: add ops enq deq apis
  doc: add Octeonx zip guide
  usertools: update devbind for octeontx zip device

 MAINTAINERS|   5 +
 config/common_base |   5 +
 doc/guides/compressdevs/features/octeontx.ini  |  22 +
 doc/guides/compressdevs/index.rst  |   1 +
 doc/guides/compressdevs/octeontx.rst   | 107 +++
 drivers/compress/Makefile  |   1 +
 drivers/compress/meson.build   |   2 +-
 drivers/compress/octeontx/Makefile |  30 +
 drivers/compress/octeontx/include/zip_regs.h   | 721 +
 drivers/compress/octeontx/meson.build  |  11 +
 .../compress/octeontx/rte_pmd_octeontx_version.map |   3 +
 drivers/compress/octeontx/zip_pmd.c| 625 ++
 drivers/compress/octeontx/zipvf.c  | 185 ++
 drivers/compress/octeontx/zipvf.h  | 354 ++
 mk/rte.app.mk  |   1 +
 usertools/dpdk-devbind.py  |   9 +
 16 files changed, 2081 insertions(+), 1 deletion(-)
 create mode 100644 doc/guides/compressdevs/features/octeontx.ini
 create mode 100644 doc/guides/compressdevs/octeontx.rst
 create mode 100644 drivers/compress/octeontx/Makefile
 create mode 100644 drivers/compress/octeontx/include/zip_regs.h
 create mode 100644 drivers/compress/octeontx/meson.build
 create mode 100644 drivers/compress/octeontx/rte_pmd_octeontx_version.map
 create mode 100644 drivers/compress/octeontx/zip_pmd.c
 create mode 100644 drivers/compress/octeontx/zipvf.c
 create mode 100644 drivers/compress/octeontx/zipvf.h

-- 
2.14.3

[dpdk-dev] [PATCH v2 2/6] compress/octeontx: add device setup PMD ops

[Shally Verma]

From: Sunila Sahu 

implement device configure and PMD ops.
setup stream resource memory pool
setup and enable hardware queue

Signed-off-by: Ashish Gupta 
Signed-off-by: Shally Verma 
Signed-off-by: Sunila Sahu 
---
 drivers/compress/octeontx/zip_pmd.c | 251 
 drivers/compress/octeontx/zipvf.c   |  73 +++
 drivers/compress/octeontx/zipvf.h   |  56 +++-
 3 files changed, 378 insertions(+), 2 deletions(-)

diff --git a/drivers/compress/octeontx/zip_pmd.c 
b/drivers/compress/octeontx/zip_pmd.c
index 2011db37e..44c271e1a 100644
--- a/drivers/compress/octeontx/zip_pmd.c
+++ b/drivers/compress/octeontx/zip_pmd.c
@@ -9,8 +9,259 @@
 #include 
 #include 
 
+static const struct rte_compressdev_capabilities
+   octtx_zip_pmd_capabilities[] = {
+   {   .algo = RTE_COMP_ALGO_DEFLATE,
+   /* Deflate */
+   .comp_feature_flags = 0,
+   /* Non sharable Priv XFORM and Stateless */
+   .window_size = {
+   .min = 1,
+   .max = 14,
+   .increment = 1
+   /* size supported 2^1 to 2^14 */
+   },
+   },
+   RTE_COMP_END_OF_CAPABILITIES_LIST()
+};
+
+/** Configure device */
+static int
+zip_pmd_config(struct rte_compressdev *dev,
+   struct rte_compressdev_config *config)
+{
+   int nb_streams;
+   char res_pool[RTE_MEMZONE_NAMESIZE];
+   struct zip_vf *vf;
+   struct rte_mempool *zip_buf_mp;
+
+   if (!config || !dev)
+   return -EIO;
+
+   vf = (struct zip_vf *)(dev->data->dev_private);
+
+   /* create pool with maximum numbers of resources
+* required by streams
+*/
+
+   /* use common pool for non-shareable priv_xform and stream */
+   nb_streams = config->max_nb_priv_xforms + config->max_nb_streams;
+
+   snprintf(res_pool, RTE_MEMZONE_NAMESIZE, "octtx_zip_res_pool%u",
+dev->data->dev_id);
+
+   /** TBD Should we use the per core object cache for stream resources */
+   zip_buf_mp = rte_mempool_create(
+   res_pool,
+   nb_streams * MAX_BUFS_PER_STREAM,
+   ZIP_BUF_SIZE,
+   0,
+   0,
+   NULL,
+   NULL,
+   NULL,
+   NULL,
+   SOCKET_ID_ANY,
+   0);
+
+   if (zip_buf_mp == NULL) {
+   ZIP_PMD_ERR(
+   "Failed to create buf mempool octtx_zip_res_pool%u",
+   dev->data->dev_id);
+   return -1;
+   }
+
+   vf->zip_mp = zip_buf_mp;
+
+   return 0;
+}
+
+/** Start device */
+static int
+zip_pmd_start(__rte_unused struct rte_compressdev *dev)
+{
+   return 0;
+}
+
+/** Stop device */
+static void
+zip_pmd_stop(__rte_unused struct rte_compressdev *dev)
+{
+
+}
+
+/** Close device */
+static int
+zip_pmd_close(struct rte_compressdev *dev)
+{
+   if (dev == NULL)
+   return -1;
+
+   struct zip_vf *vf = (struct zip_vf *)dev->data->dev_private;
+   rte_mempool_free(vf->zip_mp);
+
+   return 0;
+}
+
+/** Get device statistics */
+static void
+zip_pmd_stats_get(struct rte_compressdev *dev,
+   struct rte_compressdev_stats *stats)
+{
+   int qp_id;
+
+   for (qp_id = 0; qp_id < dev->data->nb_queue_pairs; qp_id++) {
+   struct zipvf_qp *qp = dev->data->queue_pairs[qp_id];
+
+   stats->enqueued_count += qp->qp_stats.enqueued_count;
+   stats->dequeued_count += qp->qp_stats.dequeued_count;
+
+   stats->enqueue_err_count += qp->qp_stats.enqueue_err_count;
+   stats->dequeue_err_count += qp->qp_stats.dequeue_err_count;
+   }
+}
+
+/** Reset device statistics */
+static void
+zip_pmd_stats_reset(struct rte_compressdev *dev)
+{
+   int qp_id;
+
+   for (qp_id = 0; qp_id < dev->data->nb_queue_pairs; qp_id++) {
+   struct zipvf_qp *qp = dev->data->queue_pairs[qp_id];
+   memset(&qp->qp_stats, 0, sizeof(qp->qp_stats));
+   }
+}
+
+/** Get device info */
+static void
+zip_pmd_info_get(struct rte_compressdev *dev,
+   struct rte_compressdev_info *dev_info)
+{
+   struct zip_vf *vf = (struct zip_vf *)dev->data->dev_private;
+
+   if (dev_info != NULL) {
+   dev_info->driver_name = dev->device->driver->name;
+   dev_info->feature_flags = dev->feature_flags;
+   dev_info->capabilities = octtx_zip_pmd_capabilities;
+   dev_info->max_nb_queue_pairs = vf->max_nb_queue_pairs;
+   }
+}
+
+/** Release queue pair */
+static int
+zip_pmd_qp_release(struct rte_compressdev *dev, uint16_t qp_id)
+{
+   struct zipvf_qp *qp = dev->data->queue_pairs[qp_id];
+   struct rte_ring *r 

[dpdk-dev] [PATCH v2 1/6] compress/octeontx: add octeontx zip PMD support

[Shally Verma]

From: Sunila Sahu 

Add octeontx zip pmd support in compressdev driver.
Octeontx ZIP appears as PCI device.
Add device probe and remove support.
link zip pmd library in rtp.app.mk
Update meson.build and Makefile to build octeontx zip pmd

Signed-off-by: Ashish Gupta 
Signed-off-by: Shally Verma 
Signed-off-by: Sunila Sahu 
---
 MAINTAINERS|   5 +
 config/common_base |   5 +
 drivers/compress/Makefile  |   1 +
 drivers/compress/meson.build   |   2 +-
 drivers/compress/octeontx/Makefile |  30 +
 drivers/compress/octeontx/include/zip_regs.h   | 721 +
 drivers/compress/octeontx/meson.build  |  11 +
 .../compress/octeontx/rte_pmd_octeontx_version.map |   3 +
 drivers/compress/octeontx/zip_pmd.c| 120 
 drivers/compress/octeontx/zipvf.c  |  63 ++
 drivers/compress/octeontx/zipvf.h  | 111 
 mk/rte.app.mk  |   1 +
 12 files changed, 1072 insertions(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index bc1607844..78708d8af 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -850,6 +850,11 @@ F: drivers/compress/isal/
 F: doc/guides/compressdevs/isal.rst
 F: doc/guides/compressdevs/features/isal.ini
 
+Cavium OCTEONTX zipvf
+M: Ashish Gupta 
+F: drivers/compress/octeontx
+F: doc/guides/compressdevs/octeontx.rst
+F: doc/guides/compressdevs/features/octeontx.ini
 
 Eventdev Drivers
 
diff --git a/config/common_base b/config/common_base
index 6541ad5b2..396a09682 100644
--- a/config/common_base
+++ b/config/common_base
@@ -578,6 +578,11 @@ CONFIG_RTE_COMPRESS_MAX_DEVS=64
 #
 CONFIG_RTE_COMPRESSDEV_TEST=n
 
+#
+# Compile PMD for Octeontx ZIPVF compression device
+#
+CONFIG_RTE_LIBRTE_PMD_OCTEONTX_ZIPVF=y
+
 #
 # Compile PMD for ISA-L compression device
 #
diff --git a/drivers/compress/Makefile b/drivers/compress/Makefile
index 592497f51..25cced348 100644
--- a/drivers/compress/Makefile
+++ b/drivers/compress/Makefile
@@ -4,5 +4,6 @@
 include $(RTE_SDK)/mk/rte.vars.mk
 
 DIRS-$(CONFIG_RTE_LIBRTE_PMD_ISAL) += isal
+DIRS-$(CONFIG_RTE_LIBRTE_PMD_OCTEONTX_ZIPVF) += octeontx
 
 include $(RTE_SDK)/mk/rte.subdir.mk
diff --git a/drivers/compress/meson.build b/drivers/compress/meson.build
index fb136e1b2..a7d6f3a40 100644
--- a/drivers/compress/meson.build
+++ b/drivers/compress/meson.build
@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2018 Intel Corporation
 
-drivers = ['isal']
+drivers = ['isal', 'octeontx']
 
 std_deps = ['compressdev'] # compressdev pulls in all other needed deps
 config_flag_fmt = 'RTE_LIBRTE_@0@_PMD'
diff --git a/drivers/compress/octeontx/Makefile 
b/drivers/compress/octeontx/Makefile
new file mode 100644
index 0..a58f3df3a
--- /dev/null
+++ b/drivers/compress/octeontx/Makefile
@@ -0,0 +1,30 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2018 Cavium, Inc
+
+include $(RTE_SDK)/mk/rte.vars.mk
+
+# library name
+LIB = librte_pmd_octeontx_zip.a
+
+# library version
+LIBABIVER := 1
+
+# build flags
+CFLAGS += $(WERROR_FLAGS)
+CFLAGS += -O3
+CFLAGS += -DALLOW_EXPERIMENTAL_API
+CFLAGS += -I$(RTE_SDK)/drivers/compress/octeontx/include
+
+# external library include paths
+LDLIBS += -lrte_eal -lrte_mbuf -lrte_mempool -lrte_ring
+LDLIBS += -lrte_compressdev
+LDLIBS += -lrte_pci -lrte_bus_pci
+
+# library source files
+SRCS-$(CONFIG_RTE_LIBRTE_PMD_OCTEONTX_ZIPVF) += zip_pmd.c
+SRCS-$(CONFIG_RTE_LIBRTE_PMD_OCTEONTX_ZIPVF) += zipvf.c
+
+# versioning export map
+EXPORT_MAP := rte_pmd_octeontx_version.map
+
+include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/compress/octeontx/include/zip_regs.h 
b/drivers/compress/octeontx/include/zip_regs.h
new file mode 100644
index 0..0eec393c3
--- /dev/null
+++ b/drivers/compress/octeontx/include/zip_regs.h
@@ -0,0 +1,721 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2018 Cavium, Inc
+ */
+
+#ifndef _RTE_OCTEONTX_ZIP_REGS_H_
+#define _RTE_OCTEONTX_ZIP_REGS_H_
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include 
+#include 
+
+/**
+ * Enumeration zip_cc
+ *
+ * ZIP compression coding Enumeration
+ * Enumerates ZIP_INST_S[CC].
+ */
+enum {
+   ZIP_CC_DEFAULT = 0,
+   ZIP_CC_DYN_HUFF,
+   ZIP_CC_FIXED_HUFF,
+   ZIP_CC_LZS
+} zip_cc;
+
+/**
+ * Register (NCB) zip_vq#_ena
+ *
+ * ZIP VF Queue Enable Register
+ * If a queue is disabled, ZIP CTL stops fetching instructions from the queue.
+ */
+typedef union {
+   uint64_t u;
+   struct zip_vqx_ena_s {
+#if defined(__BIG_ENDIAN_BITFIELD) /* Word 0 - Big Endian */
+   uint64_t reserved_1_63 : 63;
+   uint64_t ena   : 1;
+#else /* Word 0 - Little Endian */
+   uint64_t ena   : 1;
+   uint64_t reserved_1_63 : 63;
+#endif /* Word 0 - End */
+   } s;
+   /* struct zip_vq

[dpdk-dev] [PATCH v2 3/6] compress/octeontx: add xform and stream create support

[Shally Verma]

From: Ashish Gupta 

implement private xform and stream create ops

Signed-off-by: Ashish Gupta 
Signed-off-by: Shally Verma 
Signed-off-by: Sunila Sahu 
---
 drivers/compress/octeontx/zip_pmd.c | 140 
 drivers/compress/octeontx/zipvf.h   |  24 ++-
 2 files changed, 163 insertions(+), 1 deletion(-)

diff --git a/drivers/compress/octeontx/zip_pmd.c 
b/drivers/compress/octeontx/zip_pmd.c
index 44c271e1a..c8feb96bc 100644
--- a/drivers/compress/octeontx/zip_pmd.c
+++ b/drivers/compress/octeontx/zip_pmd.c
@@ -25,6 +25,103 @@ static const struct rte_compressdev_capabilities
RTE_COMP_END_OF_CAPABILITIES_LIST()
 };
 
+/** Parse xform parameters and setup a stream */
+int
+zip_set_stream_parameters(struct rte_compressdev *dev,
+   const struct rte_comp_xform *xform,
+   struct zip_stream *z_stream)
+{
+   int ret;
+   union zip_inst_s *inst;
+   struct zip_vf *vf = (struct zip_vf *)dev->data->dev_private;
+   void *res;
+
+   ret = rte_mempool_get_bulk(vf->zip_mp,
+   z_stream->bufs, MAX_BUFS_PER_STREAM);
+   if (ret < 0)
+   return -1;
+
+   inst = (union zip_inst_s *)z_stream->bufs[CMD_BUF];
+   res = z_stream->bufs[RES_BUF];
+
+   /* get one command buffer from pool and set up */
+
+   memset(inst->u, 0, sizeof(inst->u));
+
+   /* set bf for only first ops of stream */
+   inst->s.bf = 1;
+
+   if (xform->type == RTE_COMP_COMPRESS) {
+   inst->s.op = ZIP_OP_E_COMP;
+
+   switch (xform->compress.deflate.huffman) {
+   case RTE_COMP_HUFFMAN_DEFAULT:
+   inst->s.cc = ZIP_CC_DEFAULT;
+   break;
+   case RTE_COMP_HUFFMAN_FIXED:
+   inst->s.cc = ZIP_CC_FIXED_HUFF;
+   break;
+   case RTE_COMP_HUFFMAN_DYNAMIC:
+   inst->s.cc = ZIP_CC_DYN_HUFF;
+   break;
+   default:
+   ret = -1;
+   goto err;
+   }
+
+   switch (xform->compress.level) {
+   case RTE_COMP_LEVEL_MIN:
+   inst->s.ss = ZIP_COMP_E_LEVEL_MIN;
+   break;
+   case RTE_COMP_LEVEL_MAX:
+   inst->s.ss = ZIP_COMP_E_LEVEL_MAX;
+   break;
+   case RTE_COMP_LEVEL_NONE:
+   ZIP_PMD_ERR("Compression level not supported");
+   ret = -1;
+   goto err;
+   default:
+   /* for any value between min and max , choose
+* PMD default.
+*/
+   inst->s.ss = ZIP_COMP_E_LEVEL_MED; /** PMD default **/
+   break;
+   }
+   } else if (xform->type == RTE_COMP_DECOMPRESS) {
+   inst->s.op = ZIP_OP_E_DECOMP;
+   /* from HRM,
+* For DEFLATE decompression, [CC] must be 0x0.
+* For decompression, [SS] must be 0x0
+*/
+   inst->s.cc = 0;
+   /* Speed bit should not be set for decompression */
+   inst->s.ss = 0;
+   /* decompression context is supported only for STATEFUL
+* operations. Currently we support STATELESS ONLY so
+* skip setting of ctx pointer
+*/
+
+   } else {
+   ZIP_PMD_ERR("\nxform type not supported");
+   ret = -1;
+   goto err;
+   }
+
+   inst->s.res_ptr_addr.s.addr = rte_mempool_virt2iova(res);
+   inst->s.res_ptr_ctl.s.length = 0;
+
+   z_stream->inst = inst;
+
+   return 0;
+
+err:
+   rte_mempool_put_bulk(vf->zip_mp,
+(void *)&(z_stream->bufs[0]),
+MAX_BUFS_PER_STREAM);
+
+   return ret;
+}
+
 /** Configure device */
 static int
 zip_pmd_config(struct rte_compressdev *dev,
@@ -248,6 +345,45 @@ zip_pmd_qp_setup(struct rte_compressdev *dev, uint16_t 
qp_id,
return -1;
 }
 
+static int
+zip_pmd_stream_create(struct rte_compressdev *dev,
+   const struct rte_comp_xform *xform, void **stream)
+{
+   int ret;
+   struct zip_stream *strm;
+
+   strm = rte_malloc(NULL,
+   sizeof(struct zip_stream), 0);
+   ret = zip_set_stream_parameters(dev, xform, strm);
+   if (ret < 0) {
+   ZIP_PMD_ERR("failed configure xform parameters");
+   rte_free(strm);
+   return ret;
+   }
+   *stream = strm;
+   return 0;
+}
+
+static int
+zip_pmd_stream_free(struct rte_compressdev *dev, void *stream)
+{
+   struct zip_vf *vf = (struct zip_vf *) (dev->data->dev_private);
+   struct zip_stream *z_stream;
+
+   if (stream == NULL)
+   return -1;
+
+  

[dpdk-dev] [PATCH v2 4/6] compress/octeontx: add ops enq deq apis

[Shally Verma]

From: Ashish Gupta 

implement enqueue and dequeue apis

Signed-off-by: Ashish Gupta 
Signed-off-by: Shally Verma 
Signed-off-by: Sunila Sahu 
---
 drivers/compress/octeontx/zip_pmd.c | 114 
 drivers/compress/octeontx/zipvf.c   |  49 +++
 drivers/compress/octeontx/zipvf.h   | 169 
 3 files changed, 332 insertions(+)

diff --git a/drivers/compress/octeontx/zip_pmd.c 
b/drivers/compress/octeontx/zip_pmd.c
index c8feb96bc..4a0eea41a 100644
--- a/drivers/compress/octeontx/zip_pmd.c
+++ b/drivers/compress/octeontx/zip_pmd.c
@@ -25,6 +25,67 @@ static const struct rte_compressdev_capabilities
RTE_COMP_END_OF_CAPABILITIES_LIST()
 };
 
+/*
+ * Reset session to default state for next set of stateless operation
+ */
+static inline void
+reset_stream(struct zip_stream *z_stream)
+{
+   union zip_inst_s *inst = (union zip_inst_s *)(z_stream->inst);
+
+   inst->s.bf = 1;
+   inst->s.ef = 0;
+}
+
+int
+zip_process_op(struct rte_comp_op *op,
+   struct zipvf_qp *qp,
+   struct zip_stream *zstrm)
+{
+   int ret;
+   union zip_inst_s *inst = zstrm->inst;
+   volatile union zip_zres_s *zresult = NULL;
+
+   zipvf_prepare_cmd_stateless(op, zstrm);
+
+   zresult = (union zip_zres_s *)zstrm->bufs[RES_BUF];
+   zresult->s.compcode = 0;
+
+#ifdef ZIP_DBG
+   zip_dump_instruction(inst);
+#endif
+
+   /* Submit zip command */
+   ret = zipvf_push_command(qp, (void *)inst);
+
+   /* Check and Process results in sync mode */
+   do {
+   } while (!zresult->s.compcode);
+
+   if (zresult->s.compcode == ZIP_COMP_E_SUCCESS) {
+   op->status = RTE_COMP_OP_STATUS_SUCCESS;
+   } else {
+   /* FATAL error cannot do anything */
+   ZIP_PMD_ERR("operation failed with error code:%d\n",
+   zresult->s.compcode);
+   if (zresult->s.compcode == ZIP_COMP_E_DSTOP)
+   op->status = RTE_COMP_OP_STATUS_OUT_OF_SPACE_TERMINATED;
+   else
+   op->status = RTE_COMP_OP_STATUS_ERROR;
+   }
+
+   ZIP_PMD_INFO("ret %d,written %d\n", ret, zresult->s.totalbyteswritten);
+
+   op->produced = zresult->s.totalbyteswritten;
+   op->consumed = zresult->s.totalbytesread;
+
+   if (zresult->s.ef == 1)
+   reset_stream(zstrm);
+
+   zresult->s.compcode = 0;
+   return ret;
+}
+
 /** Parse xform parameters and setup a stream */
 int
 zip_set_stream_parameters(struct rte_compressdev *dev,
@@ -111,6 +172,7 @@ zip_set_stream_parameters(struct rte_compressdev *dev,
inst->s.res_ptr_ctl.s.length = 0;
 
z_stream->inst = inst;
+   z_stream->func = zip_process_op;
 
return 0;
 
@@ -385,6 +447,56 @@ zip_pmd_stream_free(struct rte_compressdev *dev, void 
*stream)
return 0;
 }
 
+static uint16_t
+zip_pmd_enqueue_burst_sync(void *queue_pair,
+   struct rte_comp_op **ops, uint16_t nb_ops)
+{
+   struct zipvf_qp *qp = queue_pair;
+   struct rte_comp_op *op;
+   struct zip_stream *zstrm;
+   int ret, i;
+   uint16_t enqd = 0;
+
+   for (i = 0; i < nb_ops; i++) {
+   op = ops[i];
+   if (op->op_type == RTE_COMP_OP_STATEFUL)
+   op->status = RTE_COMP_OP_STATUS_INVALID_ARGS;
+   else {
+   /* process stateless ops */
+   zstrm = (struct zip_stream *)op->private_xform;
+   ret = zstrm->func(op, qp, zstrm);
+   }
+
+   /* Whatever is out of op, put it into completion queue with
+* its status
+*/
+   ret = rte_ring_enqueue(qp->processed_pkts, (void *)op);
+   if (unlikely(ret < 0)) {
+   /* increment count if failed to enqueue op*/
+   qp->qp_stats.enqueue_err_count++;
+   } else {
+   qp->qp_stats.enqueued_count++;
+   enqd++;
+   }
+   }
+   return enqd;
+}
+
+static uint16_t
+zip_pmd_dequeue_burst_sync(void *queue_pair,
+   struct rte_comp_op **ops, uint16_t nb_ops)
+{
+   struct zipvf_qp *qp = queue_pair;
+
+   unsigned int nb_dequeued = 0;
+
+   nb_dequeued = rte_ring_dequeue_burst(qp->processed_pkts,
+   (void **)ops, nb_ops, NULL);
+   qp->qp_stats.dequeued_count += nb_dequeued;
+
+   return nb_dequeued;
+}
+
 struct rte_compressdev_ops octtx_zip_pmd_ops = {
.dev_configure  = zip_pmd_config,
.dev_start  = zip_pmd_start,
@@ -446,6 +558,8 @@ zip_pci_probe(struct rte_pci_driver *pci_drv __rte_unused,
 
compressdev->dev_ops = &octtx_zip_pmd_ops;
/* register rx/tx burst functions for data path */
+   compressdev->dequeue_burst = zip_pmd_dequeue_burst_sync;
+   compressd

[dpdk-dev] [PATCH v2 5/6] doc: add Octeonx zip guide

[Shally Verma]

add zip pmd feature specification and overview documentation

Signed-off-by: Ashish Gupta 
Signed-off-by: Shally Verma 
Signed-off-by: Sunila Sahu 
---
 doc/guides/compressdevs/features/octeontx.ini |  22 ++
 doc/guides/compressdevs/index.rst |   1 +
 doc/guides/compressdevs/octeontx.rst  | 107 ++
 3 files changed, 130 insertions(+)

diff --git a/doc/guides/compressdevs/features/octeontx.ini 
b/doc/guides/compressdevs/features/octeontx.ini
new file mode 100644
index 0..224bcb8bb
--- /dev/null
+++ b/doc/guides/compressdevs/features/octeontx.ini
@@ -0,0 +1,22 @@
+;
+; Refer to default.ini for the full list of available PMD features.
+;
+; Supported features of 'OCTEONTX ZIP' compression driver.
+;
+[Features]
+HW Accelerated = Y
+CPU SSE=
+CPU AVX=
+CPU AVX2   =
+CPU AVX512 =
+CPU NEON   =
+Stateful   =
+By-Pass=
+Chained mbufs  =
+Deflate= Y
+LZS=
+Adler32=
+Crc32  =
+Adler32&Crc32  =
+Fixed  = Y
+Dynamic= Y
diff --git a/doc/guides/compressdevs/index.rst 
b/doc/guides/compressdevs/index.rst
index bc59ce810..cb467ab90 100644
--- a/doc/guides/compressdevs/index.rst
+++ b/doc/guides/compressdevs/index.rst
@@ -11,3 +11,4 @@ Compression Device Drivers
 
 overview
 isal
+octeontx
diff --git a/doc/guides/compressdevs/octeontx.rst 
b/doc/guides/compressdevs/octeontx.rst
new file mode 100644
index 0..d4010d31d
--- /dev/null
+++ b/doc/guides/compressdevs/octeontx.rst
@@ -0,0 +1,107 @@
+..  SPDX-License-Identifier: BSD-3-Clause
+Copyright(c) 2018 Cavium Networks.
+
+Octeontx ZIP Compression Poll Mode Driver
+=
+
+The Octeontx ZIP PMD (**librte_pmd_octeontx_zip**) provides poll mode
+compression & decompression driver for ZIP HW offload device, found in
+**Cavium OCTEONTX** SoC family.
+
+More information can be found at `Cavium, Inc Official Website
+`_.
+
+Features
+
+
+Octeontx ZIP PMD has support for:
+
+Compression/Decompression algorithm:
+
+* DEFLATE
+
+Huffman code type:
+
+* FIXED
+* DYNAMIC
+
+Window size support:
+
+* 2 to 16K (2^14)
+
+Limitations
+---
+
+* Chained mbufs are not supported.
+
+Supported OCTEONTX SoCs
+---
+
+- CN83xx
+
+Steps To Setup Platform
+---
+
+   Octeontx SDK includes kernel image which provides Octeontx ZIP PF
+   driver to manage configuration of ZIPVF device.
+   Required version of SDK is "OCTEONTX-SDK-6.2.0-build35" or above.
+
+   SDK can be install by using below command.
+   #rpm -ivh CTEONTX-SDK-6.2.0-build35.x86_64.rpm --force --nodeps
+   It will install OCTEONTX-SDK at following default location
+   /usr/local/Cavium_Networks/OCTEONTX-SDK/
+
+   For more information on building and booting linux kernel on OCTEONTX
+   please refer 
/usr/local/Cavium_Networks/OCTEONTX-SDK/docs/OcteonTX-SDK-UG_6.2.0.pdf.
+
+   SDK and related information can be obtained from: `Cavium support site 
`_.
+
+Installation
+
+
+Driver Compilation
+~~
+
+To compile the OCTEONTX ZIP PMD for Linux arm64 gcc target, run the
+following ``make`` command:
+
+   .. code-block:: console
+
+  cd 
+  make config T=arm64-thunderx-linuxapp-gcc install
+
+
+Initialization
+--
+
+The octeontx zip is exposed as pci device which consists of a set of
+PCIe VF devices. On EAL initialization, ZIP PCIe VF devices will be
+probed. To use the PMD in an application, user must:
+
+* run dev_bind script to bind eight ZIP PCIe VFs to the ``vfio-pci`` driver:
+
+   .. code-block:: console
+
+  ./usertools/dpdk-devbind.py -b vfio-pci 0001:04:00.1
+  ./usertools/dpdk-devbind.py -b vfio-pci 0001:04:00.2
+  ./usertools/dpdk-devbind.py -b vfio-pci 0001:04:00.3
+  ./usertools/dpdk-devbind.py -b vfio-pci 0001:04:00.4
+  ./usertools/dpdk-devbind.py -b vfio-pci 0001:04:00.5
+  ./usertools/dpdk-devbind.py -b vfio-pci 0001:04:00.6
+  ./usertools/dpdk-devbind.py -b vfio-pci 0001:04:00.7
+  ./usertools/dpdk-devbind.py -b vfio-pci 0001:04:01.0
+
+* The unit test cases can be tested as below:
+
+   .. code-block:: console
+
+  reserve enough huge pages
+  cd to the top-level DPDK directory
+  export RTE_TARGET=arm64-thunderx-linuxapp-gcc
+  export RTE_SDK=`pwd`
+  cd to test/test
+  type the command "make" to compile
+  run the tests with "./test"
+  type the command "compressdev_autotest" to test
+
+
-- 
2.14.3

[dpdk-dev] [PATCH v2 6/6] usertools: update devbind for octeontx zip device

[Shally Verma]

From: Ashish Gupta 

add the cavium octeontx zip pci device details.

Signed-off-by: Ashish Gupta 
Signed-off-by: Shally Verma 
Signed-off-by: Sunila Sahu 
---
 usertools/dpdk-devbind.py | 9 +
 1 file changed, 9 insertions(+)

diff --git a/usertools/dpdk-devbind.py b/usertools/dpdk-devbind.py
index d0c420906..1d48a6cba 100755
--- a/usertools/dpdk-devbind.py
+++ b/usertools/dpdk-devbind.py
@@ -24,6 +24,8 @@
   'SVendor': None, 'SDevice': None}
 cavium_tim = {'Class': '08', 'Vendor': '177d', 'Device': 'a051',
   'SVendor': None, 'SDevice': None}
+cavium_zip = {'Class': '12', 'Vendor': '177d', 'Device': 'a037',
+  'SVendor': None, 'SDevice': None}
 avp_vnic = {'Class': '05', 'Vendor': '1af4', 'Device': '1110',
   'SVendor': None, 'SDevice': None}
 
@@ -31,6 +33,7 @@
 crypto_devices = [encryption_class, intel_processor_class]
 eventdev_devices = [cavium_sso, cavium_tim]
 mempool_devices = [cavium_fpa]
+compress_devices = [cavium_zip]
 
 # global dict ethernet devices present. Dictionary indexed by PCI address.
 # Each device within this is itself a dictionary of device properties
@@ -569,6 +572,10 @@ def show_status():
 if status_dev == "mempool" or status_dev == "all":
 show_device_status(mempool_devices, "Mempool")
 
+if status_dev == "compress" or status_dev == "all":
+show_device_status(compress_devices , "Compress")
+
+
 def parse_args():
 '''Parses the command-line arguments given by the user and takes the
 appropriate action for each'''
@@ -642,6 +649,7 @@ def do_arg_actions():
 get_device_details(crypto_devices)
 get_device_details(eventdev_devices)
 get_device_details(mempool_devices)
+get_device_details(compress_devices)
 show_status()
 
 
@@ -654,6 +662,7 @@ def main():
 get_device_details(crypto_devices)
 get_device_details(eventdev_devices)
 get_device_details(mempool_devices)
+get_device_details(compress_devices)
 do_arg_actions()
 
 if __name__ == "__main__":
-- 
2.14.3

[dpdk-dev] [PATCH v2 0/5] compress: add ZLIB compression PMD

[Shally Verma]

This patch series add software zlib based compression PMD
in DPDK compress drivers.
Application must need to install zlib prior to compile and
run this PMD to avail compression/decompression services.
Currently driver only tested for deflate, stateless
compression and decompression with direct memory buffers.

Changes in v2:
- removed unused variables
- corrected capability to reflect current support
- add lookup for internally maintained mempool during device_configure
- optimized core compression/decompression logic in enq/deq APIs
- updated documentation with correct feature support

v1 includes:
- build changes to build zlib PMD
- zlib PMD implementation
- zlib PMD documentation
- meson build support

This patchset is dependent upon compressdev API.

Sunila Sahu (5):
  compress/zlib: add ZLIB PMD support
  compress/zlib: add device setup PMD ops
  compress/zlib: add xform and stream create support
  compress/zlib: add enq deq apis
  doc: add ZLIB PMD documentation

 MAINTAINERS|   5 +
 config/common_base |   5 +
 doc/guides/compressdevs/features/zlib.ini  |  22 ++
 doc/guides/compressdevs/zlib.rst   |  68 
 drivers/compress/Makefile  |   1 +
 drivers/compress/meson.build   |   2 +-
 drivers/compress/zlib/Makefile |  29 ++
 drivers/compress/zlib/meson.build  |  14 +
 drivers/compress/zlib/rte_pmd_zlib_version.map |   3 +
 drivers/compress/zlib/zlib_pmd.c   | 412 +
 drivers/compress/zlib/zlib_pmd_ops.c   | 311 +++
 drivers/compress/zlib/zlib_pmd_private.h   |  71 +
 mk/rte.app.mk  |   2 +
 13 files changed, 944 insertions(+), 1 deletion(-)
 create mode 100644 doc/guides/compressdevs/features/zlib.ini
 create mode 100644 doc/guides/compressdevs/zlib.rst
 create mode 100644 drivers/compress/zlib/Makefile
 create mode 100644 drivers/compress/zlib/meson.build
 create mode 100644 drivers/compress/zlib/rte_pmd_zlib_version.map
 create mode 100644 drivers/compress/zlib/zlib_pmd.c
 create mode 100644 drivers/compress/zlib/zlib_pmd_ops.c
 create mode 100644 drivers/compress/zlib/zlib_pmd_private.h

-- 
2.9.5

Re: [dpdk-dev] [PATCH v5 7/9] net/virtio: support in-order Rx and Tx

[Ferruh Yigit]

On 7/2/2018 5:53 PM, Maxime Coquelin wrote:
> 
> 
> On 07/02/2018 06:52 PM, Ferruh Yigit wrote:
>> On 7/2/2018 5:41 PM, Ferruh Yigit wrote:
>>> On 7/2/2018 2:56 PM, Marvin Liu wrote:
 IN_ORDER Rx function depends on merge-able feature. Descriptors
 allocation and free will be done in bulk.

 Virtio dequeue logic:
  dequeue_burst_rx(burst mbufs)
  for (each mbuf b) {
  if (b need merge) {
  merge remained mbufs
  add merged mbuf to return mbufs list
  } else {
  add mbuf to return mbufs list
  }
  }
  if (last mbuf c need merge) {
  dequeue_burst_rx(required mbufs)
  merge last mbuf c
  }
  refill_avail_ring_bulk()
  update_avail_ring()
  return mbufs list

 IN_ORDER Tx function can support offloading features. Packets which
 matched "can_push" option will be handled by simple xmit function. Those
 packets can't match "can_push" will be handled by original xmit function
 with in-order flag.

 Virtio enqueue logic:
  xmit_cleanup(used descs)
  for (each xmit mbuf b) {
  if (b can inorder xmit) {
  add mbuf b to inorder burst list
  continue
  } else {
  xmit inorder burst list
  xmit mbuf b by original function
  }
  }
  if (inorder burst list not empty) {
  xmit inorder burst list
  }
  update_avail_ring()

 Signed-off-by: Marvin Liu 
 Reviewed-by: Maxime Coquelin 
>>>
>>> <...>
>>>
 @@ -150,6 +188,83 @@ virtio_xmit_cleanup(struct virtqueue *vq, uint16_t 
 num)
}
   }
   
 +/* Cleanup from completed inorder transmits. */
 +static void
 +virtio_xmit_cleanup_inorder(struct virtqueue *vq, uint16_t num)
 +{
 +  uint16_t i, used_idx, desc_idx, last_idx;
>>>
>>>
>>> Getting following build error [1], from code it looks like false positive, 
>>> but
>>> to get rid of the build error would it be OK to set initial value to 
>>> "desc_idx"?
>>
>> I applied this while merging, if this is wrong please let me know, we can 
>> fix in
>> next-net, Thanks.
> 
> Looks good to me. I didn't catch it with the GCC version I use.

I didn't dig more but I also didn't get the error with regular build, the one
with all DEBUGs enabled and mach=default combination gave the error, not sure 
why.

> 
> Thanks,
> Maxime
> 
>>>
>>>
>>> [1]
>>> .../dpdk/drivers/net/virtio/virtio_rxtx.c:195:24: error: ‘desc_idx’ may be 
>>> used
>>> uninitialized in this function [-Werror=maybe-uninitialized]
>>>
>>>
>>>uint16_t i, used_idx, desc_idx, last_idx;
>>>
>>>
>>>
>>>  ^~~~
>>>
>>

[dpdk-dev] [PATCH v2 1/5] compress/zlib: add ZLIB PMD support

[Shally Verma]

From: Ashish Gupta 

Add sw zlib pmd support in compressdev driver.
Add device probe and remove support.
Add ZLIB build file support.

Signed-off-by: Sunila Sahu 
Signed-off-by: Shally Verma 
Signed-off-by: Ashish Gupta 
---
 MAINTAINERS|  3 +
 config/common_base |  5 ++
 drivers/compress/Makefile  |  1 +
 drivers/compress/meson.build   |  2 +-
 drivers/compress/zlib/Makefile | 28 +
 drivers/compress/zlib/meson.build  | 14 +
 drivers/compress/zlib/rte_pmd_zlib_version.map |  3 +
 drivers/compress/zlib/zlib_pmd.c   | 81 ++
 drivers/compress/zlib/zlib_pmd_private.h   | 33 +++
 mk/rte.app.mk  |  2 +
 10 files changed, 171 insertions(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index dabb12d..448bbe1 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -851,6 +851,9 @@ F: drivers/compress/isal/
 F: doc/guides/compressdevs/isal.rst
 F: doc/guides/compressdevs/features/isal.ini
 
+ZLIB
+M: Sunila Sahu 
+F: drivers/compress/zlib/
 
 Eventdev Drivers
 
diff --git a/config/common_base b/config/common_base
index 721e59b..dabc269 100644
--- a/config/common_base
+++ b/config/common_base
@@ -585,6 +585,11 @@ CONFIG_RTE_COMPRESSDEV_TEST=n
 CONFIG_RTE_LIBRTE_PMD_ISAL=n
 
 #
+# Compile PMD for ZLIB compression device
+#
+CONFIG_RTE_LIBRTE_PMD_ZLIB=n
+
+#
 # Compile generic event device library
 #
 CONFIG_RTE_LIBRTE_EVENTDEV=y
diff --git a/drivers/compress/Makefile b/drivers/compress/Makefile
index 592497f..1f159a5 100644
--- a/drivers/compress/Makefile
+++ b/drivers/compress/Makefile
@@ -4,5 +4,6 @@
 include $(RTE_SDK)/mk/rte.vars.mk
 
 DIRS-$(CONFIG_RTE_LIBRTE_PMD_ISAL) += isal
+DIRS-$(CONFIG_RTE_LIBRTE_PMD_ZLIB) += zlib
 
 include $(RTE_SDK)/mk/rte.subdir.mk
diff --git a/drivers/compress/meson.build b/drivers/compress/meson.build
index fb136e1..e4d5e5c 100644
--- a/drivers/compress/meson.build
+++ b/drivers/compress/meson.build
@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2018 Intel Corporation
 
-drivers = ['isal']
+drivers = ['isal','zlib']
 
 std_deps = ['compressdev'] # compressdev pulls in all other needed deps
 config_flag_fmt = 'RTE_LIBRTE_@0@_PMD'
diff --git a/drivers/compress/zlib/Makefile b/drivers/compress/zlib/Makefile
new file mode 100644
index 000..bd322c9
--- /dev/null
+++ b/drivers/compress/zlib/Makefile
@@ -0,0 +1,28 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2018 Cavium Networks
+
+include $(RTE_SDK)/mk/rte.vars.mk
+
+# library name
+LIB = librte_pmd_zlib.a
+
+# build flags
+CFLAGS += -O3
+CFLAGS += $(WERROR_FLAGS)
+CFLAGS += -DALLOW_EXPERIMENTAL_API
+
+# library version
+LIBABIVER := 1
+
+# versioning export map
+EXPORT_MAP := rte_pmd_zlib_version.map
+
+# external library dependencies
+LDLIBS += -lrte_eal -lrte_mbuf -lrte_mempool -lrte_ring -lz
+LDLIBS += -lrte_compressdev
+LDLIBS += -lrte_bus_vdev
+
+# library source files
+SRCS-$(CONFIG_RTE_LIBRTE_PMD_ZLIB) += zlib_pmd.c
+
+include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/compress/zlib/meson.build 
b/drivers/compress/zlib/meson.build
new file mode 100644
index 000..7748de2
--- /dev/null
+++ b/drivers/compress/zlib/meson.build
@@ -0,0 +1,14 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2018 Cavium Networks
+
+dep = dependency('zlib', required: false)
+if not dep.found()
+   build = false
+endif
+
+deps += 'bus_vdev'
+sources = files('zlib_pmd.c', 'zlib_pmd_ops.c')
+ext_deps += dep
+pkgconfig_extra_libs += '-lz'
+
+allow_experimental_apis = true
diff --git a/drivers/compress/zlib/rte_pmd_zlib_version.map 
b/drivers/compress/zlib/rte_pmd_zlib_version.map
new file mode 100644
index 000..1a99a33
--- /dev/null
+++ b/drivers/compress/zlib/rte_pmd_zlib_version.map
@@ -0,0 +1,3 @@
+18.08 {
+   local: *;
+};
diff --git a/drivers/compress/zlib/zlib_pmd.c b/drivers/compress/zlib/zlib_pmd.c
new file mode 100644
index 000..f667ccc
--- /dev/null
+++ b/drivers/compress/zlib/zlib_pmd.c
@@ -0,0 +1,81 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2018 Cavium Networks
+ */
+
+#include 
+#include 
+#include "zlib_pmd_private.h"
+
+static int
+zlib_create(const char *name,
+   struct rte_vdev_device *vdev,
+   struct rte_compressdev_pmd_init_params *init_params)
+{
+   struct rte_compressdev *dev;
+
+   dev = rte_compressdev_pmd_create(name, &vdev->device,
+   sizeof(struct zlib_private), init_params);
+   if (dev == NULL) {
+   ZLIB_PMD_ERR("driver %s: create failed", init_params->name);
+   return -ENODEV;
+   }
+
+   dev->feature_flags = RTE_COMP_FF_NONCOMPRESSED_BLOCKS;
+
+   return 0;
+}
+
+static int
+zlib_probe(struct rte_vdev_device *vdev)
+{
+   struct rte_compressdev_pmd_init_params init_params = {
+   "",
+   

[dpdk-dev] [PATCH v2 2/5] compress/zlib: add device setup PMD ops

[Shally Verma]

From: Ashish Gupta 

Implement device configure and PMD ops

Signed-off-by: Sunila Sahu 
Signed-off-by: Shally Verma 
Signed-off-by: Ashish Gupta 
---
 drivers/compress/zlib/Makefile   |   1 +
 drivers/compress/zlib/zlib_pmd.c |   2 +
 drivers/compress/zlib/zlib_pmd_ops.c | 236 +++
 drivers/compress/zlib/zlib_pmd_private.h |  34 +
 4 files changed, 273 insertions(+)

diff --git a/drivers/compress/zlib/Makefile b/drivers/compress/zlib/Makefile
index bd322c9..5cf8de6 100644
--- a/drivers/compress/zlib/Makefile
+++ b/drivers/compress/zlib/Makefile
@@ -24,5 +24,6 @@ LDLIBS += -lrte_bus_vdev
 
 # library source files
 SRCS-$(CONFIG_RTE_LIBRTE_PMD_ZLIB) += zlib_pmd.c
+SRCS-$(CONFIG_RTE_LIBRTE_PMD_ZLIB) += zlib_pmd_ops.c
 
 include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/compress/zlib/zlib_pmd.c b/drivers/compress/zlib/zlib_pmd.c
index f667ccc..c4f67bb 100644
--- a/drivers/compress/zlib/zlib_pmd.c
+++ b/drivers/compress/zlib/zlib_pmd.c
@@ -20,6 +20,8 @@ zlib_create(const char *name,
return -ENODEV;
}
 
+   dev->dev_ops = rte_zlib_pmd_ops;
+
dev->feature_flags = RTE_COMP_FF_NONCOMPRESSED_BLOCKS;
 
return 0;
diff --git a/drivers/compress/zlib/zlib_pmd_ops.c 
b/drivers/compress/zlib/zlib_pmd_ops.c
new file mode 100644
index 000..03b6da5
--- /dev/null
+++ b/drivers/compress/zlib/zlib_pmd_ops.c
@@ -0,0 +1,236 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2018 Cavium Networks
+ */
+
+#include 
+
+#include 
+#include 
+
+#include "zlib_pmd_private.h"
+
+static const struct rte_compressdev_capabilities zlib_pmd_capabilities[] = {
+   {   /* Deflate */
+   .algo = RTE_COMP_ALGO_DEFLATE,
+   .window_size = {
+   .min = 8,
+   .max = 15,
+   .increment = 1
+   },
+   },
+
+   RTE_COMP_END_OF_CAPABILITIES_LIST()
+
+};
+
+/** Configure device */
+static int
+zlib_pmd_config(struct rte_compressdev *dev,
+   struct rte_compressdev_config *config)
+{
+   struct rte_mempool *mp;
+
+   struct zlib_private *internals = dev->data->dev_private;
+   snprintf(internals->mp_name, RTE_MEMPOOL_NAMESIZE,
+   "stream_mp_%u", dev->data->dev_id);
+   mp = rte_mempool_lookup(internals->mp_name);
+   if (mp == NULL) {
+   mp = rte_mempool_create(internals->mp_name,
+   config->max_nb_priv_xforms +
+   config->max_nb_streams,
+   sizeof(struct zlib_priv_xform),
+   0, 0, NULL, NULL, NULL,
+   NULL, config->socket_id,
+   0);
+   if (mp == NULL) {
+   ZLIB_PMD_ERR("Cannot create private xform pool on "
+   "socket %d\n", config->socket_id);
+   return -ENOMEM;
+   }
+   }
+   return 0;
+}
+
+/** Start device */
+static int
+zlib_pmd_start(__rte_unused struct rte_compressdev *dev)
+{
+   return 0;
+}
+
+/** Stop device */
+static void
+zlib_pmd_stop(__rte_unused struct rte_compressdev *dev)
+{
+}
+
+/** Close device */
+static int
+zlib_pmd_close(struct rte_compressdev *dev)
+{
+   struct zlib_private *internals = dev->data->dev_private;
+   struct rte_mempool *mp = rte_mempool_lookup(internals->mp_name);
+   rte_mempool_free(mp);
+   return 0;
+}
+
+/** Get device statistics */
+static void
+zlib_pmd_stats_get(struct rte_compressdev *dev,
+   struct rte_compressdev_stats *stats)
+{
+   int qp_id;
+
+   for (qp_id = 0; qp_id < dev->data->nb_queue_pairs; qp_id++) {
+   struct zlib_qp *qp = dev->data->queue_pairs[qp_id];
+
+   stats->enqueued_count += qp->qp_stats.enqueued_count;
+   stats->dequeued_count += qp->qp_stats.dequeued_count;
+
+   stats->enqueue_err_count += qp->qp_stats.enqueue_err_count;
+   stats->dequeue_err_count += qp->qp_stats.dequeue_err_count;
+   }
+}
+
+/** Reset device statistics */
+static void
+zlib_pmd_stats_reset(struct rte_compressdev *dev)
+{
+   int qp_id;
+
+   for (qp_id = 0; qp_id < dev->data->nb_queue_pairs; qp_id++) {
+   struct zlib_qp *qp = dev->data->queue_pairs[qp_id];
+
+   memset(&qp->qp_stats, 0, sizeof(qp->qp_stats));
+   }
+}
+
+/** Get device info */
+static void
+zlib_pmd_info_get(struct rte_compressdev *dev,
+   struct rte_compressdev_info *dev_info)
+{
+   if (dev_info != NULL) {
+   dev_info->driver_name = dev->device->name;
+   dev_info->feature_flags = dev->feature_flags;
+   dev_info->capabilities = zlib_pmd_capabilities;
+   }
+}
+
+/** Release queue pair */
+static int
+zlib_pmd_qp_release(struct rte_compressdev *dev, uint16_t qp_id)
+{
+   struct zlib_qp 

[dpdk-dev] [PATCH v2 3/5] compress/zlib: add xform and stream create support

[Shally Verma]

From: Sunila Sahu 

Implement private xform and stream create ops

Signed-off-by: Sunila Sahu 
Signed-off-by: Shally Verma 
Signed-off-by: Ashish Gupta 
---
 drivers/compress/zlib/zlib_pmd.c | 93 
 drivers/compress/zlib/zlib_pmd_ops.c | 83 ++--
 drivers/compress/zlib/zlib_pmd_private.h |  4 ++
 3 files changed, 176 insertions(+), 4 deletions(-)

diff --git a/drivers/compress/zlib/zlib_pmd.c b/drivers/compress/zlib/zlib_pmd.c
index c4f67bb..7c2614e 100644
--- a/drivers/compress/zlib/zlib_pmd.c
+++ b/drivers/compress/zlib/zlib_pmd.c
@@ -6,6 +6,99 @@
 #include 
 #include "zlib_pmd_private.h"
 
+/** Parse comp xform and set private xform/stream parameters */
+int
+zlib_set_stream_parameters(const struct rte_comp_xform *xform,
+   struct zlib_stream *stream)
+{
+   int strategy, level, wbits;
+   z_stream *strm = &stream->strm;
+
+   /* allocate deflate state */
+   strm->zalloc = Z_NULL;
+   strm->zfree = Z_NULL;
+   strm->opaque = Z_NULL;
+
+   switch (xform->type) {
+   case RTE_COMP_COMPRESS:
+   /** Compression window bits */
+   switch (xform->compress.algo) {
+   case RTE_COMP_ALGO_DEFLATE:
+   wbits = -(xform->compress.window_size);
+   break;
+   default:
+   ZLIB_PMD_ERR("Compression algorithm not supported\n");
+   return -1;
+   }
+   /** Compression Level */
+   switch (xform->compress.level) {
+   case RTE_COMP_LEVEL_PMD_DEFAULT:
+   level = Z_DEFAULT_COMPRESSION;
+   break;
+   case RTE_COMP_LEVEL_NONE:
+   level = Z_NO_COMPRESSION;
+   break;
+   case RTE_COMP_LEVEL_MIN:
+   level = Z_BEST_SPEED;
+   break;
+   case RTE_COMP_LEVEL_MAX:
+   level = Z_BEST_COMPRESSION;
+   break;
+   default:
+   level = xform->compress.level;
+   if (level < RTE_COMP_LEVEL_MIN ||
+   level > RTE_COMP_LEVEL_MAX) {
+   ZLIB_PMD_ERR("Compression level %d "
+   "not supported\n",
+   level);
+   return -1;
+   }
+   }
+   /** Compression strategy */
+   switch (xform->compress.deflate.huffman) {
+   case RTE_COMP_HUFFMAN_DEFAULT:
+   strategy = Z_DEFAULT_STRATEGY;
+   break;
+   case RTE_COMP_HUFFMAN_FIXED:
+   strategy = Z_FIXED;
+   break;
+   case RTE_COMP_HUFFMAN_DYNAMIC:
+   strategy = Z_HUFFMAN_ONLY;
+   break;
+   default:
+   ZLIB_PMD_ERR("Compression strategy not supported\n");
+   return -1;
+   }
+   if (deflateInit2(strm, level,
+   Z_DEFLATED, wbits,
+   DEF_MEM_LEVEL, strategy) != Z_OK) {
+   ZLIB_PMD_ERR("Deflate init failed\n");
+   return -1;
+   }
+   break;
+
+   case RTE_COMP_DECOMPRESS:
+   /** window bits */
+   switch (xform->decompress.algo) {
+   case RTE_COMP_ALGO_DEFLATE:
+   wbits = -(xform->decompress.window_size);
+   break;
+   default:
+   ZLIB_PMD_ERR("Compression algorithm not supported\n");
+   return -1;
+   }
+
+   if (inflateInit2(strm, wbits) != Z_OK) {
+   ZLIB_PMD_ERR("Inflate init failed\n");
+   return -1;
+   }
+   break;
+   default:
+   return -1;
+   }
+   return 0;
+}
+
 static int
 zlib_create(const char *name,
struct rte_vdev_device *vdev,
diff --git a/drivers/compress/zlib/zlib_pmd_ops.c 
b/drivers/compress/zlib/zlib_pmd_ops.c
index 03b6da5..812406b 100644
--- a/drivers/compress/zlib/zlib_pmd_ops.c
+++ b/drivers/compress/zlib/zlib_pmd_ops.c
@@ -212,6 +212,81 @@ zlib_pmd_qp_setup(struct rte_compressdev *dev, uint16_t 
qp_id,
return -1;
 }
 
+/** Configure stream */
+static int
+zlib_pmd_stream_create(struct rte_compressdev *dev,
+   const struct rte_comp_xform *xform,
+   void **zstream)
+{
+   int ret = 0;
+   struct zlib_stream *stream;
+   struct zlib_private *internals = dev->data->dev_private;
+
+   if (xform == NULL) {
+   ZLIB_PMD_ERR("i

[dpdk-dev] [PATCH v2 4/5] compress/zlib: add enq deq apis

[Shally Verma]

From: Sunila Sahu 

implement enqueue and dequeue apis

Signed-off-by: Sunila Sahu 
Signed-off-by: Shally Verma 
Signed-off-by: Ashish Gupta 
---
 drivers/compress/zlib/zlib_pmd.c | 238 ++-
 1 file changed, 237 insertions(+), 1 deletion(-)

diff --git a/drivers/compress/zlib/zlib_pmd.c b/drivers/compress/zlib/zlib_pmd.c
index 7c2614e..aef23de 100644
--- a/drivers/compress/zlib/zlib_pmd.c
+++ b/drivers/compress/zlib/zlib_pmd.c
@@ -6,7 +6,198 @@
 #include 
 #include "zlib_pmd_private.h"
 
-/** Parse comp xform and set private xform/stream parameters */
+/** Compute next mbuf in the list, assign data buffer and len */
+#define COMPUTE_BUF(mbuf, data, len)   \
+   ((mbuf = mbuf->next) ?  \
+   (data = rte_pktmbuf_mtod(mbuf, uint8_t *)), \
+   (len = rte_pktmbuf_data_len(mbuf)) : 0)
+
+/** Set op->status to appropriate flag if we run out of mbuf */
+#define COMPUTE_DST_BUF(mbuf, dst, dlen)   \
+   ((COMPUTE_BUF(mbuf, dst, dlen)) ? 1 :   \
+   (!(op->status = \
+   RTE_COMP_OP_STATUS_OUT_OF_SPACE_TERMINATED)))
+
+static void
+process_zlib_deflate(struct rte_comp_op *op, z_stream *strm)
+{
+   int ret, flush, fin_flush;
+   struct rte_mbuf *mbuf_src = op->m_src;
+   struct rte_mbuf *mbuf_dst = op->m_dst;
+
+   switch (op->flush_flag) {
+   case RTE_COMP_FLUSH_FULL:
+   case RTE_COMP_FLUSH_FINAL:
+   fin_flush = Z_FINISH;
+   break;
+   default:
+   op->status = RTE_COMP_OP_STATUS_INVALID_ARGS;
+   ZLIB_PMD_ERR("\nInvalid flush value");
+
+   }
+
+   if (unlikely(!strm)) {
+   op->status = RTE_COMP_OP_STATUS_INVALID_ARGS;
+   ZLIB_PMD_ERR("\nInvalid z_stream");
+   return;
+   }
+   /* Update z_stream with the inputs provided by application */
+   strm->next_in = rte_pktmbuf_mtod_offset(mbuf_src, uint8_t *,
+   op->src.offset);
+
+   strm->avail_in = rte_pktmbuf_data_len(mbuf_src) - op->src.offset;
+
+   strm->next_out = rte_pktmbuf_mtod_offset(mbuf_dst, uint8_t *,
+   op->dst.offset);
+
+   strm->avail_out = rte_pktmbuf_data_len(mbuf_dst) - op->dst.offset;
+
+   /* Set flush value to NO_FLUSH unless it is last mbuf */
+   flush = Z_NO_FLUSH;
+   /* Initialize status to SUCCESS */
+   op->status = RTE_COMP_OP_STATUS_SUCCESS;
+
+   do {
+   /* Set flush value to Z_FINISH for last block */
+   if ((op->src.length - strm->total_in) <= strm->avail_in) {
+   strm->avail_in = (op->src.length - strm->total_in);
+   flush = fin_flush;
+   }
+   do {
+   ret = deflate(strm, flush);
+   if (unlikely(ret == Z_STREAM_ERROR)) {
+   /* error return, do not process further */
+   op->status =  RTE_COMP_OP_STATUS_ERROR;
+   break;
+   }
+   /* Break if Z_STREAM_END is encountered */
+   if (ret == Z_STREAM_END)
+   goto def_end;
+
+/* Break if current input buffer is consumed or
+ * there is no more space for compressed output
+ */
+   } while ((strm->avail_out == 0) &&
+   COMPUTE_DST_BUF(mbuf_dst,  strm->next_out,
+   strm->avail_out));
+
+   /* Update source buffer to next mbuf
+* Exit if op->status is not SUCCESS or input buffers are fully consumed
+*/
+   } while ((op->status == RTE_COMP_OP_STATUS_SUCCESS) &&
+   (COMPUTE_BUF(mbuf_src, strm->next_in, strm->avail_in)));
+
+def_end:
+   /* Update op stats */
+   switch (op->status) {
+   case RTE_COMP_OP_STATUS_SUCCESS:
+   op->consumed += strm->total_in;
+   case RTE_COMP_OP_STATUS_OUT_OF_SPACE_TERMINATED:
+   op->produced += strm->total_out;
+   break;
+   default:
+   ZLIB_PMD_ERR("stats not updated for status:%d\n",
+   op->status);
+   }
+
+   deflateReset(strm);
+}
+
+static void
+process_zlib_inflate(struct rte_comp_op *op, z_stream *strm)
+{
+   int ret, flush;
+   struct rte_mbuf *mbuf_src = op->m_src;
+   struct rte_mbuf *mbuf_dst = op->m_dst;
+
+   if (unlikely(!strm)) {
+   op->status = RTE_COMP_OP_STATUS_INVALID_ARGS;
+   ZLIB_PMD_ERR("\nInvalid z_stream");
+   return;
+   }
+   strm->next_in = rte_pktmbuf_mtod_offset(mbuf_src, uint8_t *,
+   op->src.offset);
+
+   strm->avail_in = rte_pktmbuf_data_len(mbuf_src) - op->src.offset;
+
+   strm->next_out = rte_pktmbuf_mtod_offset(mbu

[dpdk-dev] [PATCH v2 5/5] doc: add ZLIB PMD documentation

[Shally Verma]

add zlib pmd feature specification and overview documentation

Signed-off-by: Sunila Sahu 
Signed-off-by: Shally Verma 
Signed-off-by: Ashish Gupta 
---
 MAINTAINERS   |  2 +
 doc/guides/compressdevs/features/zlib.ini | 22 ++
 doc/guides/compressdevs/zlib.rst  | 68 +++
 3 files changed, 92 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 448bbe1..1c217b0 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -854,6 +854,8 @@ F: doc/guides/compressdevs/features/isal.ini
 ZLIB
 M: Sunila Sahu 
 F: drivers/compress/zlib/
+F: doc/guides/compressdevs/zlib.rst
+F: doc/guides/compressdevs/features/zlib.ini
 
 Eventdev Drivers
 
diff --git a/doc/guides/compressdevs/features/zlib.ini 
b/doc/guides/compressdevs/features/zlib.ini
new file mode 100644
index 000..bdc0fc4
--- /dev/null
+++ b/doc/guides/compressdevs/features/zlib.ini
@@ -0,0 +1,22 @@
+;
+; Refer to default.ini for the full list of available PMD features.
+;
+; Supported features of 'ZLIB' compression driver.
+;
+[Features]
+HW Accelerated =
+CPU SSE=
+CPU AVX=
+CPU AVX2   =
+CPU AVX512 =
+CPU NEON   =
+Stateful   =
+By-Pass=
+Chained mbufs  =
+Deflate= Y
+LZS=
+Adler32=
+Crc32  =
+Adler32&Crc32  =
+Fixed  = Y
+Dynamic= Y
diff --git a/doc/guides/compressdevs/zlib.rst b/doc/guides/compressdevs/zlib.rst
new file mode 100644
index 000..7dd5c74
--- /dev/null
+++ b/doc/guides/compressdevs/zlib.rst
@@ -0,0 +1,68 @@
+..  SPDX-License-Identifier: BSD-3-Clause
+Copyright(c) 2018 Cavium Networks.
+
+ZLIB Compression Poll Mode Driver
+==
+
+The ZLIB PMD (**librte_pmd_zlib**) provides poll mode compression &
+decompression driver based on SW zlib library,
+
+Features
+
+
+ZLIB PMD has support for:
+
+Compression/Decompression algorithm:
+
+* DEFLATE
+
+Huffman code type:
+
+* FIXED
+* DYNAMIC
+
+Window size support:
+
+* Min - 256 bytes
+* Max - 32K
+
+Limitations
+---
+
+* Chained mbufs are not supported.
+
+Installation
+
+
+* To build DPDK with ZLIB library, the user is required to download the 
``libz`` library.
+* Use following command for installation.
+
+*  For Fedora users ::
+  yum install zlib-devel
+*  For ubuntu users ::
+  apt-get install zlib1g-dev
+
+* Once downloaded, the user needs to build the library.
+
+* make can  be used to install the library on their system, before building 
DPDK::
+
+make
+sudo make install
+
+Initialization
+--
+
+In order to enable this virtual compression PMD, user must:
+
+* Set ``CONFIG_RTE_LIBRTE_PMD_ZLIB=y`` in config/common_base.
+
+To use the PMD in an application, user must:
+
+* Call ``rte_vdev_init("compress_zlib")`` within the application.
+
+* Use ``--vdev="compress_zlib"`` in the EAL options, which will call 
``rte_vdev_init()`` internally.
+
+The following parameter (optional) can be provided in the previous two calls:
+
+* ``socket_id:`` Specify the socket where the memory for the device is going 
to be allocated
+  (by default, socket_id will be the socket where the core that is creating 
the PMD is running on).
-- 
2.9.5

[dpdk-dev] [PATCH] common/qat: replace snprintf

[Fiona Trahe]

Replaced snprintf with strlcpy for safer string copy

Signed-off-by: Fiona Trahe 
Signed-off-by: Tomasz Jozwiak 
---
 drivers/common/qat/qat_device.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/common/qat/qat_device.c b/drivers/common/qat/qat_device.c
index 64f236e..f32d723 100644
--- a/drivers/common/qat/qat_device.c
+++ b/drivers/common/qat/qat_device.c
@@ -2,6 +2,8 @@
  * Copyright(c) 2018 Intel Corporation
  */
 
+#include 
+
 #include "qat_device.h"
 #include "adf_transport_access_macros.h"
 #include "qat_sym_pmd.h"
@@ -116,7 +118,7 @@ qat_pci_device_allocate(struct rte_pci_device *pci_dev)
 
qat_dev = qat_pci_get_dev(qat_dev_id);
memset(qat_dev, 0, sizeof(*qat_dev));
-   snprintf(qat_dev->name, QAT_DEV_NAME_MAX_LEN, "%s", name);
+   strlcpy(qat_dev->name, name, QAT_DEV_NAME_MAX_LEN);
qat_dev->qat_dev_id = qat_dev_id;
qat_dev->pci_dev = pci_dev;
switch (qat_dev->pci_dev->id.device_id) {
-- 
2.7.4

[dpdk-dev] [PATH 0/6] Use IOVAs check based on DMA mask

[Alejandro Lucero]

This patchset adds, mainly, a check for ensuring IOVAs are within a
restricted range due to addressing limitations with some devices. There
are two known cases: NFP and IOMMU VT-d emulation.

With this check IOVAs out of range are detected and PMDs can abort
initialization. For the VT-d case, IOVA VA mode is allowed as long as
IOVAs are within the supported range, avoiding to forbid IOVA VA by
default.

For the addressing limitations known cases, there are just 40(NFP) or
39(VT-d) bits for handling IOVAs. When using IOVA PA, those limitations
imply 1TB(NFP) or 512M(VT-d) as upper limits, which is likely enough for
most systems. With machines using more memory, the added check will
ensure IOVAs within the range.

With IOVA VA, and because the way the Linux kernel serves mmap calls
in 64 bits systems, 39 or 40 bits are not enough. It is possible to
give an address hint with a lower starting address than the default one
used by the kernel, and then ensuring the mmap uses that hint or hint plus
some offset. With 64 bits systems, the process virtual address space is
large enoguh for doing the hugepages mmaping within the supported range
when those addressing limitations exist. This patchset also adds a change
for using such a hint making the use of IOVA VA a more than likely
possibility when there are those addressing limitations.

The check is not done by default but just when it is required. This
patchset adds the check for NFP initialization and for setting the IOVA
mode is an emulated VT-d is detected.

This patchset applies on 17.11.3.

Similar changes will be submitted to main DPDK branch soon.

[dpdk-dev] [PATCH 5/6] net/nfp: check hugepages IOVAs based on DMA mask

[Alejandro Lucero]

NFP devices can not handle DMA addresses requiring more than
40 bits. This patch uses rte_dev_check_dma_mask with 40 bits
and avoids device initialization if memory out of NFP range.

Signed-off-by: Alejandro Lucero 
---
 drivers/net/nfp/nfp_net.c | 8 
 1 file changed, 8 insertions(+)

diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index d9cd047..5976f37 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -2649,6 +2649,14 @@ uint32_t nfp_net_txq_full(struct nfp_net_txq *txq)
 
pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev);
 
+   /* NFP can not handle DMA addresses requiring more than 40 bits */
+   if (rte_eth_dev_check_dma_mask(40) < 0) {
+   RTE_LOG(INFO, PMD, "device %s can not be used:",
+  pci_dev->device.name);
+   RTE_LOG(INFO, PMD, "\trestricted dma mask to 40 bits!\n");
+   return -ENODEV;
+   };
+
if ((pci_dev->id.device_id == PCI_DEVICE_ID_NFP4000_PF_NIC) ||
(pci_dev->id.device_id == PCI_DEVICE_ID_NFP6000_PF_NIC)) {
port = get_pf_port_number(eth_dev->data->name);
-- 
1.9.1

[dpdk-dev] [PATCH 1/6] mem: add function for checking memsegs IOVAs addresses

[Alejandro Lucero]

A device can suffer addressing limitations. This functions checks
memsegs have iovas within the supported range based on dma mask.

PMD should use this during initialization if supported devices
suffer addressing limitations, returning an error if this function
returns memsegs out of range.

Another potential usage is for emulated IOMMU hardware with addressing
limitations.

Signed-off-by: Alejandro Lucero 
---
 lib/librte_eal/common/eal_common_memory.c  | 37 ++
 lib/librte_eal/common/include/rte_memory.h |  3 +++
 2 files changed, 40 insertions(+)

diff --git a/lib/librte_eal/common/eal_common_memory.c 
b/lib/librte_eal/common/eal_common_memory.c
index fc6c44d..ca49e5c 100644
--- a/lib/librte_eal/common/eal_common_memory.c
+++ b/lib/librte_eal/common/eal_common_memory.c
@@ -109,6 +109,43 @@
}
 }
 
+/* check memseg iovas are within the required range based on dma mask */
+int
+rte_eal_check_dma_mask(uint8_t maskbits)
+{
+
+   const struct rte_mem_config *mcfg;
+   uint64_t mask;
+   int i;
+   int ret = 0;
+
+   /* create dma mask */
+   mask = 1ULL << maskbits;
+   mask -= 1;
+
+   /* get pointer to global configuration */
+   mcfg = rte_eal_get_configuration()->mem_config;
+
+   for (i = 0; i < RTE_MAX_MEMSEG; i++) {
+   if (mcfg->memseg[i].addr == NULL)
+   break;
+
+   if (mcfg->memseg[i].iova & ~mask) {
+   ret = -1;
+   break;
+   }
+   }
+
+   if (!ret)
+   return ret;
+
+   RTE_LOG(INFO, EAL, "memseg[%d] iova %"PRIx64" out of range:\n",
+  i, mcfg->memseg[i].iova);
+   RTE_LOG(INFO, EAL, "\tusing dma mask %"PRIx64"\n", mask);
+
+   return -1;
+}
+
 /* return the number of memory channels */
 unsigned rte_memory_get_nchannel(void)
 {
diff --git a/lib/librte_eal/common/include/rte_memory.h 
b/lib/librte_eal/common/include/rte_memory.h
index 80a8fc0..b2a0168 100644
--- a/lib/librte_eal/common/include/rte_memory.h
+++ b/lib/librte_eal/common/include/rte_memory.h
@@ -209,6 +209,9 @@ struct rte_memseg {
  */
 unsigned rte_memory_get_nrank(void);
 
+/* check memsegs iovas are within a range based on dma mask */
+int rte_eal_check_dma_mask(uint8_t maskbits);
+
 /**
  * Drivers based on uio will not load unless physical
  * addresses are obtainable. It is only possible to get
-- 
1.9.1

[dpdk-dev] [PATCH 2/6] ethdev: add function for checking IOVAs by a device

[Alejandro Lucero]

A PMD should invoke this function for checking memsegs iovas are within
the supported range by the device.

Signed-off-by: Alejandro Lucero 
---
 lib/librte_ether/rte_ethdev.h | 13 +
 1 file changed, 13 insertions(+)

diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
index eba11ca..e51a432 100644
--- a/lib/librte_ether/rte_ethdev.h
+++ b/lib/librte_ether/rte_ethdev.h
@@ -2799,6 +2799,19 @@ int rte_eth_dev_set_vlan_ether_type(uint16_t port_id,
 int rte_eth_dev_set_vlan_pvid(uint16_t port_id, uint16_t pvid, int on);
 
 /**
+ * check device dma mask within expected range based on dma mask.
+ *
+ * @param maskbits
+ *  mask length in bits
+ *
+ */
+static inline int
+rte_eth_dev_check_dma_mask(uint8_t maskbits)
+{
+   return rte_eal_check_dma_mask(maskbits);
+}
+
+/**
  *
  * Retrieve a burst of input packets from a receive queue of an Ethernet
  * device. The retrieved packets are stored in *rte_mbuf* structures whose
-- 
1.9.1

[dpdk-dev] [PATCH 3/6] bus/pci: use IOVAs check when setting IOVA mode

[Alejandro Lucero]

Although VT-d emulation currently only supports 39 bits, it could
be iovas being within that supported range. This patch allows
IOVA mode in such a case.

Indeed, memory initialization code can be modified for using lower
virtual addresses than those used by the kernel for 64 bits processes
by default, and therefore memsegs iovas can use 39 bits or less for
most system. And this is likely 100% true for VMs.

Signed-off-by: Alejandro Lucero 
---
 drivers/bus/pci/linux/pci.c | 15 +++
 1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/drivers/bus/pci/linux/pci.c b/drivers/bus/pci/linux/pci.c
index 74deef3..792c819 100644
--- a/drivers/bus/pci/linux/pci.c
+++ b/drivers/bus/pci/linux/pci.c
@@ -43,6 +43,7 @@
 #include 
 #include 
 #include 
+#include 
 
 #include "eal_private.h"
 #include "eal_filesystem.h"
@@ -613,10 +614,12 @@
fclose(fp);
 
mgaw = ((vtd_cap_reg & VTD_CAP_MGAW_MASK) >> VTD_CAP_MGAW_SHIFT) + 1;
-   if (mgaw < X86_VA_WIDTH)
+
+   if (!rte_eal_check_dma_mask(mgaw))
+   return true;
+   else
return false;
 
-   return true;
 }
 #elif defined(RTE_ARCH_PPC_64)
 static bool
@@ -640,13 +643,17 @@
 {
struct rte_pci_device *dev = NULL;
struct rte_pci_driver *drv = NULL;
+   int iommu_dma_mask_check_done = 0;
 
FOREACH_DRIVER_ON_PCIBUS(drv) {
FOREACH_DEVICE_ON_PCIBUS(dev) {
if (!rte_pci_match(drv, dev))
continue;
-   if (!pci_one_device_iommu_support_va(dev))
-   return false;
+   if (!iommu_dma_mask_check_done) {
+   if (pci_one_device_iommu_support_va(dev) < 0)
+   return false;
+   iommu_dma_mask_check_done  = 1;
+   }
}
}
return true;
-- 
1.9.1

[dpdk-dev] [PATCH 4/6] mem: use address hint for mapping hugepages

[Alejandro Lucero]

Linux kernel uses a really high address as starting address for
serving mmaps calls. If there exists addressing limitations and
IOVA mode is VA, this starting address is likely too high for
those devices. However, it is possible to use a lower address in
the process virtual address space as with 64 bits there is a lot
of available space.

This patch adds an address hint as starting address for 64 bits
systems.

Signed-off-by: Alejandro Lucero 
---
 lib/librte_eal/linuxapp/eal/eal_memory.c | 50 ++--
 1 file changed, 41 insertions(+), 9 deletions(-)

diff --git a/lib/librte_eal/linuxapp/eal/eal_memory.c 
b/lib/librte_eal/linuxapp/eal/eal_memory.c
index 17c20d4..0a134c4 100644
--- a/lib/librte_eal/linuxapp/eal/eal_memory.c
+++ b/lib/librte_eal/linuxapp/eal/eal_memory.c
@@ -88,6 +88,23 @@
 
 static uint64_t baseaddr_offset;
 
+#ifdef RTE_ARCH_64
+/*
+ * Linux kernel uses a really high address as starting address for serving
+ * mmaps calls. If there exists addressing limitations and IOVA mode is VA,
+ * this starting address is likely too high for those devices. However, it
+ * is possible to use a lower address in the process virtual address space
+ * as with 64 bits there is a lot of available space.
+ *
+ * Current known limitations are 39 or 40 bits. Setting the starting address
+ * at 4GB implies there are 508GB or 1020GB for mapping the available
+ * hugepages. This is likely enough for most systems, although a device with
+ * addressing limitations should call rte_dev_check_dma_mask for ensuring all
+ * memory is within supported range.
+ */
+static uint64_t baseaddr = 0x1;
+#endif
+
 static bool phys_addrs_available = true;
 
 #define RANDOMIZE_VA_SPACE_FILE "/proc/sys/kernel/randomize_va_space"
@@ -260,16 +277,10 @@
 static void *
 get_virtual_area(size_t *size, size_t hugepage_sz)
 {
-   void *addr;
+   void *addr, *addr_hint;
int fd;
long aligned_addr;
 
-   if (internal_config.base_virtaddr != 0) {
-   addr = (void*) (uintptr_t) (internal_config.base_virtaddr +
-   baseaddr_offset);
-   }
-   else addr = NULL;
-
RTE_LOG(DEBUG, EAL, "Ask a virtual area of 0x%zx bytes\n", *size);
 
fd = open("/dev/zero", O_RDONLY);
@@ -278,7 +289,22 @@
return NULL;
}
do {
-   addr = mmap(addr,
+   if (internal_config.base_virtaddr != 0) {
+   addr_hint = (void *) (uintptr_t)
+   (internal_config.base_virtaddr +
+baseaddr_offset);
+   }
+#ifdef RTE_ARCH_64
+   else {
+   addr_hint = (void *) (uintptr_t) (baseaddr +
+   baseaddr_offset);
+   }
+#else
+   else {
+   addr_hint = NULL;
+   }
+#endif
+   addr = mmap(addr_hint,
(*size) + hugepage_sz, PROT_READ,
 #ifdef RTE_ARCH_PPC_64
MAP_PRIVATE | MAP_ANONYMOUS | MAP_HUGETLB,
@@ -286,8 +312,14 @@
MAP_PRIVATE,
 #endif
fd, 0);
-   if (addr == MAP_FAILED)
+   if (addr == MAP_FAILED) {
+   /* map failed. Let's try with less memory */
*size -= hugepage_sz;
+   } else if (addr_hint && addr != addr_hint) {
+   /* map not using hint. Let's try with another offset */
+   addr = MAP_FAILED;
+   baseaddr_offset += 0x1;
+   }
} while (addr == MAP_FAILED && *size > 0);
 
if (addr == MAP_FAILED) {
-- 
1.9.1