Re: [PATCH v4 2/2] config/arm: use common cpu arch for cross files
On Tue, 5 Dec 2023 03:52:59 + Joyce Kong wrote: > The cpu info in some cross files is inconsistent with > that in SoC flags. The mismatch doesn't cause any issue > because the cpu field in the cross file takes no effect > and machine_args in config/arm/meson.build actually works. > Use a common one in cross files to remove any confusion. > > Reported-by: Honnappa Nagarahalli > Signed-off-by: Joyce Kong > Reviewed-by: Ruifeng Wang Good to see consolidation instead of copy/paste. Acked-by: Stephen Hemminger
Re: [PATCH v3 0/4] net/hns3: refactor mailbox
On 2023/12/7 20:31, Ferruh Yigit wrote: On 12/7/2023 1:37 AM, Jie Hai wrote: This patchset refactors mailbox codes. -- v3: 1. fix the checkpatch warning on __atomic_xxx. -- Dengdui Huang (4): net/hns3: refactor VF mailbox message struct net/hns3: refactor PF mailbox message struct net/hns3: refactor send mailbox function net/hns3: refactor handle mailbox function Hi Jie, Dengdui, Set fails to build with clang and stdatomic [1], it can be reproduced with command [2]. Can you please check? [1] https://mails.dpdk.org/archives/test-report/2023-December/525771.html [2] CC=clang meson setup --buildtype=debugoptimized --werror -Denable_stdatomic=true build && ninja -C build . Hi, Ferruh, Thanks, I'will check and fix it. Best regards, Jie Hai
[PATCH 0/1] docs:programmers guide index reflow
Enhanced Organization of the Programmer's Guide for Better Clarity
- Restructured main sections for coherent topic grouping; added 'Foundation
Principles',
'Memory Management', 'CPU Management', and 'Device Libraries'.
- 'Foundation Principles' now includes 'Introduction', 'Overview', 'Source
Organization',
and a new 'Glossary' section.
- Consolidated memory-related topics under 'Memory Management', enhancing topic
discoverability.
- Introduced 'CPU Management' to centralize CPU-related libraries and features.
- Reorganized 'Device Libraries' for clearer categorization of device-specific
documentation.
- Grouped packet processing topics under 'CPU Packet Processing' for focused
guidance.
- 'Utility Libraries' section newly created to encompass supporting libraries
like Timer,
RCU, and Ring libraries.
- Refined 'Protocol Processing Libraries' section, emphasizing
protocol-specific information.
- New 'High-Level Libraries' section highlights advanced libraries like Packet
Framework and
Graph Library.
- Added 'Howto Guides' and 'Tips & Tricks' for practical, hands-on advice and
optimization
techniques.
- Updated filenames and page titles for consistency and clarity, including the
addition
of 'library' in titles where it was previously missing.
- This reorganization simplifies guide navigation, facilitating quicker access
to information.
David Young (1):
Modified Programmer's Guide index for better organization and
readability
doc/guides/prog_guide/bpf_lib.rst | 4 +-
doc/guides/prog_guide/dmadev.rst | 6 +-
doc/guides/prog_guide/efd_lib.rst | 4 +-
.../prog_guide/env_abstraction_layer.rst | 4 +-
doc/guides/prog_guide/ethdev/index.rst| 12 ++
.../prog_guide/{ => ethdev}/poll_mode_drv.rst | 0
.../prog_guide/{ => ethdev}/qos_framework.rst | 0
.../prog_guide/{ => ethdev}/rte_flow.rst | 0
.../{ => ethdev}/switch_representation.rst| 0
.../{ => ethdev}/traffic_management.rst | 0
.../traffic_metering_and_policing.rst | 0
.../{ => eventdev}/event_crypto_adapter.rst | 0
.../event_ethernet_rx_adapter.rst | 0
.../event_ethernet_tx_adapter.rst | 0
.../{ => eventdev}/event_timer_adapter.rst| 0
.../prog_guide/{ => eventdev}/eventdev.rst| 2 +-
doc/guides/prog_guide/eventdev/index.rst | 11 ++
doc/guides/prog_guide/fib_lib.rst | 4 +-
.../generic_receive_offload_lib.rst | 4 +-
.../generic_segmentation_offload_lib.rst | 4 +-
doc/guides/prog_guide/gpudev.rst | 4 +-
doc/guides/prog_guide/index.rst | 145 --
doc/guides/prog_guide/lpm6_lib.rst| 4 +-
doc/guides/prog_guide/lpm_lib.rst | 4 +-
doc/guides/prog_guide/mbuf_lib.rst| 6 +-
doc/guides/prog_guide/mempool_lib.rst | 4 +-
doc/guides/prog_guide/mldev.rst | 6 +-
.../prog_guide/packet_classif_access_ctrl.rst | 4 +-
doc/guides/prog_guide/packet_framework.rst| 4 +-
doc/guides/prog_guide/rawdev.rst | 4 +-
doc/guides/prog_guide/regexdev.rst| 4 +-
doc/guides/prog_guide/rib_lib.rst | 4 +-
...y_dpdk_functions.rst => thread_safety.rst} | 4 +-
33 files changed, 166 insertions(+), 86 deletions(-)
create mode 100644 doc/guides/prog_guide/ethdev/index.rst
rename doc/guides/prog_guide/{ => ethdev}/poll_mode_drv.rst (100%)
rename doc/guides/prog_guide/{ => ethdev}/qos_framework.rst (100%)
rename doc/guides/prog_guide/{ => ethdev}/rte_flow.rst (100%)
rename doc/guides/prog_guide/{ => ethdev}/switch_representation.rst (100%)
rename doc/guides/prog_guide/{ => ethdev}/traffic_management.rst (100%)
rename doc/guides/prog_guide/{ => ethdev}/traffic_metering_and_policing.rst
(100%)
rename doc/guides/prog_guide/{ => eventdev}/event_crypto_adapter.rst (100%)
rename doc/guides/prog_guide/{ => eventdev}/event_ethernet_rx_adapter.rst
(100%)
rename doc/guides/prog_guide/{ => eventdev}/event_ethernet_tx_adapter.rst
(100%)
rename doc/guides/prog_guide/{ => eventdev}/event_timer_adapter.rst (100%)
rename doc/guides/prog_guide/{ => eventdev}/eventdev.rst (99%)
create mode 100644 doc/guides/prog_guide/eventdev/index.rst
rename doc/guides/prog_guide/{thread_safety_dpdk_functions.rst =>
thread_safety.rst} (98%)
--
2.41.0.windows.1
[PATCH 1/1] Modified Programmer's Guide index for better organization and readability
---
doc/guides/prog_guide/bpf_lib.rst | 4 +-
doc/guides/prog_guide/dmadev.rst | 6 +-
doc/guides/prog_guide/efd_lib.rst | 4 +-
.../prog_guide/env_abstraction_layer.rst | 4 +-
doc/guides/prog_guide/ethdev/index.rst| 12 ++
.../prog_guide/{ => ethdev}/poll_mode_drv.rst | 0
.../prog_guide/{ => ethdev}/qos_framework.rst | 0
.../prog_guide/{ => ethdev}/rte_flow.rst | 0
.../{ => ethdev}/switch_representation.rst| 0
.../{ => ethdev}/traffic_management.rst | 0
.../traffic_metering_and_policing.rst | 0
.../{ => eventdev}/event_crypto_adapter.rst | 0
.../event_ethernet_rx_adapter.rst | 0
.../event_ethernet_tx_adapter.rst | 0
.../{ => eventdev}/event_timer_adapter.rst| 0
.../prog_guide/{ => eventdev}/eventdev.rst| 2 +-
doc/guides/prog_guide/eventdev/index.rst | 11 ++
doc/guides/prog_guide/fib_lib.rst | 4 +-
.../generic_receive_offload_lib.rst | 4 +-
.../generic_segmentation_offload_lib.rst | 4 +-
doc/guides/prog_guide/gpudev.rst | 4 +-
doc/guides/prog_guide/index.rst | 145 --
doc/guides/prog_guide/lpm6_lib.rst| 4 +-
doc/guides/prog_guide/lpm_lib.rst | 4 +-
doc/guides/prog_guide/mbuf_lib.rst| 6 +-
doc/guides/prog_guide/mempool_lib.rst | 4 +-
doc/guides/prog_guide/mldev.rst | 6 +-
.../prog_guide/packet_classif_access_ctrl.rst | 4 +-
doc/guides/prog_guide/packet_framework.rst| 4 +-
doc/guides/prog_guide/rawdev.rst | 4 +-
doc/guides/prog_guide/regexdev.rst| 4 +-
doc/guides/prog_guide/rib_lib.rst | 4 +-
...y_dpdk_functions.rst => thread_safety.rst} | 4 +-
33 files changed, 166 insertions(+), 86 deletions(-)
create mode 100644 doc/guides/prog_guide/ethdev/index.rst
rename doc/guides/prog_guide/{ => ethdev}/poll_mode_drv.rst (100%)
rename doc/guides/prog_guide/{ => ethdev}/qos_framework.rst (100%)
rename doc/guides/prog_guide/{ => ethdev}/rte_flow.rst (100%)
rename doc/guides/prog_guide/{ => ethdev}/switch_representation.rst (100%)
rename doc/guides/prog_guide/{ => ethdev}/traffic_management.rst (100%)
rename doc/guides/prog_guide/{ => ethdev}/traffic_metering_and_policing.rst
(100%)
rename doc/guides/prog_guide/{ => eventdev}/event_crypto_adapter.rst (100%)
rename doc/guides/prog_guide/{ => eventdev}/event_ethernet_rx_adapter.rst
(100%)
rename doc/guides/prog_guide/{ => eventdev}/event_ethernet_tx_adapter.rst
(100%)
rename doc/guides/prog_guide/{ => eventdev}/event_timer_adapter.rst (100%)
rename doc/guides/prog_guide/{ => eventdev}/eventdev.rst (99%)
create mode 100644 doc/guides/prog_guide/eventdev/index.rst
rename doc/guides/prog_guide/{thread_safety_dpdk_functions.rst =>
thread_safety.rst} (98%)
diff --git a/doc/guides/prog_guide/bpf_lib.rst
b/doc/guides/prog_guide/bpf_lib.rst
index 1cf2d59429..8c820328b9 100644
--- a/doc/guides/prog_guide/bpf_lib.rst
+++ b/doc/guides/prog_guide/bpf_lib.rst
@@ -1,8 +1,8 @@
.. SPDX-License-Identifier: BSD-3-Clause
Copyright(c) 2018 Intel Corporation.
-Berkeley Packet Filter Library
-==
+Berkeley Packet Filter (BPF) Library
+
The DPDK provides an BPF library that gives the ability
to load and execute Enhanced Berkeley Packet Filter (eBPF) bytecode within
diff --git a/doc/guides/prog_guide/dmadev.rst b/doc/guides/prog_guide/dmadev.rst
index 2aa26d33b8..2b31dc904f 100644
--- a/doc/guides/prog_guide/dmadev.rst
+++ b/doc/guides/prog_guide/dmadev.rst
@@ -1,8 +1,10 @@
.. SPDX-License-Identifier: BSD-3-Clause
Copyright 2021 HiSilicon Limited
-DMA Device Library
-==
+.. _direct_memory_access_dma:
+
+Direct Memory Access (DMA) Device Library
+=
The DMA library provides a DMA device framework for management and provisioning
of hardware and software DMA poll mode drivers, defining generic API which
diff --git a/doc/guides/prog_guide/efd_lib.rst
b/doc/guides/prog_guide/efd_lib.rst
index 2b355ff2a6..4bee7237be 100644
--- a/doc/guides/prog_guide/efd_lib.rst
+++ b/doc/guides/prog_guide/efd_lib.rst
@@ -3,8 +3,8 @@
.. _Efd_Library:
-Elastic Flow Distributor Library
-
+Elastic Flow Distributor (EFD) Library
+==
Introduction
diff --git a/doc/guides/prog_guide/env_abstraction_layer.rst
b/doc/guides/prog_guide/env_abstraction_layer.rst
index 6debf54efb..5ce80a4496 100644
--- a/doc/guides/prog_guide/env_abstraction_layer.rst
+++ b/doc/guides/prog_guide/env_abstraction_layer.rst
@@ -3,8 +3,8 @@
.. _Environment_Abstraction_Layer:
-Environment Abstraction Layer
-=
+Environment Abstraction Layer (EAL) Library
+===
The E
Re: [PATCH RESEND v6 0/5] app/testpmd: support multiple process attach and detach port
Hi Ferruh and Thomas, This series have been discussing for over a year and a half. Looking back on previous discussions, we have also made some progress and consensus. I am sticking to track it. Because they resolve a real exist issue. Can you take a look at it again? BR, /Huisong 在 2023/10/30 20:17, lihuisong (C) 写道: Hi Ferruh and Thomas, This series have been discussing more than one year. Kindly ping for reivew. 在 2023/10/9 18:34, lihuisong (C) 写道: Hi Ferruh and Thomas, Can you take a look at this series? They've been over a year on disscussion. 在 2023/8/2 11:15, Huisong Li 写道: This patchset fix some bugs and support attaching and detaching port in primary and secondary. --- -v6: adjust rte_eth_dev_is_used position based on alphabetical order in version.map -v5: move 'ALLOCATED' state to the back of 'REMOVED' to avoid abi break. -v4: fix a misspelling. -v3: #1 merge patch 1/6 and patch 2/6 into patch 1/5, and add modification for other bus type. #2 add a RTE_ETH_DEV_ALLOCATED state in rte_eth_dev_state to resolve the probelm in patch 2/5. -v2: resend due to CI unexplained failure. Huisong Li (5): drivers/bus: restore driver assignment at front of probing ethdev: fix skip valid port in probing callback app/testpmd: check the validity of the port app/testpmd: add attach and detach port for multiple process app/testpmd: stop forwarding in new or destroy event app/test-pmd/testpmd.c | 47 +++- app/test-pmd/testpmd.h | 1 - drivers/bus/auxiliary/auxiliary_common.c | 9 - drivers/bus/dpaa/dpaa_bus.c | 9 - drivers/bus/fslmc/fslmc_bus.c | 8 +++- drivers/bus/ifpga/ifpga_bus.c | 12 -- drivers/bus/pci/pci_common.c | 9 - drivers/bus/vdev/vdev.c | 10 - drivers/bus/vmbus/vmbus_common.c | 9 - drivers/net/bnxt/bnxt_ethdev.c | 3 +- drivers/net/bonding/bonding_testpmd.c | 1 - drivers/net/mlx5/mlx5.c | 2 +- lib/ethdev/ethdev_driver.c | 13 +-- lib/ethdev/ethdev_driver.h | 12 ++ lib/ethdev/ethdev_pci.h | 2 +- lib/ethdev/rte_class_eth.c | 2 +- lib/ethdev/rte_ethdev.c | 4 +- lib/ethdev/rte_ethdev.h | 4 +- lib/ethdev/version.map | 1 + 19 files changed, 114 insertions(+), 44 deletions(-) . .
[PATCH v1] net/memif: remove extra mbuf refcnt update in zero copy Tx
The refcnt update of stored mbufs in memif driver is redundant since
those mbufs are only freed in eth_memif_tx_zc(). No other place
can free those stored mbufs quietly. So remove the redundant mbuf
refcnt update in dpdk memif driver to avoid extra heavy cost.
Performance of dpdk memif zero copy tx is improved with this change.
Signed-off-by: Liangxing Wang
Reviewed-by: Ruifeng Wang
---
drivers/net/memif/rte_eth_memif.c | 6 --
1 file changed, 6 deletions(-)
diff --git a/drivers/net/memif/rte_eth_memif.c
b/drivers/net/memif/rte_eth_memif.c
index 7cc8c0da91..962d390b90 100644
--- a/drivers/net/memif/rte_eth_memif.c
+++ b/drivers/net/memif/rte_eth_memif.c
@@ -265,8 +265,6 @@ memif_free_stored_mbufs(struct pmd_process_private
*proc_private, struct memif_q
cur_tail = __atomic_load_n(&ring->tail, __ATOMIC_ACQUIRE);
while (mq->last_tail != cur_tail) {
RTE_MBUF_PREFETCH_TO_FREE(mq->buffers[(mq->last_tail + 1) &
mask]);
- /* Decrement refcnt and free mbuf. (current segment) */
- rte_mbuf_refcnt_update(mq->buffers[mq->last_tail & mask], -1);
rte_pktmbuf_free_seg(mq->buffers[mq->last_tail & mask]);
mq->last_tail++;
}
@@ -825,10 +823,6 @@ memif_tx_one_zc(struct pmd_process_private *proc_private,
struct memif_queue *mq
next_in_chain:
/* store pointer to mbuf to free it later */
mq->buffers[slot & mask] = mbuf;
- /* Increment refcnt to make sure the buffer is not freed before server
-* receives it. (current segment)
-*/
- rte_mbuf_refcnt_update(mbuf, 1);
/* populate descriptor */
d0 = &ring->desc[slot & mask];
d0->length = rte_pktmbuf_data_len(mbuf);
--
2.25.1
RE: [PATCH v4 1/2] config: correct cpu instruction set for cross build
> -Original Message-
> From: Stephen Hemminger
> Sent: Friday, December 8, 2023 6:41 AM
> To: Joyce Kong
> Cc: tho...@monjalon.net; Ruifeng Wang ;
> bruce.richard...@intel.com; dev@dpdk.org; nd ;
> sta...@dpdk.org
> Subject: Re: [PATCH v4 1/2] config: correct cpu instruction set for cross
> build
>
> On Tue, 5 Dec 2023 03:52:58 +
> Joyce Kong wrote:
>
> > The platform value would be 'native' only when not cross build.
> > Move the operation about modifying cpu_instruction_set while platform
> > equals 'native' to the not cross build branch.
> >
> > Fixes: bf66003b51ec ("build: use platform for generic and native
> > builds")
> > Cc: sta...@dpdk.org
> >
> > Signed-off-by: Joyce Kong
> > Reviewed-by: Ruifeng Wang
>
> Looks right, what is the impact on existing users doing current cross builds?
The original operation changes cpu_instruction_set to 'native' when both
'platform == native' and 'cpu_instruction_set == auto'.
There was no impact on the existing users for current cross build, as the cross
files set cpu field to a specific arch, such as 'armv8-a', and this didn't
trigger the modification for cpu_instruction_set.
However, if we want to use a common arch in cross files like 'auto', the
modification operation would override 'native' value to cpu_instruction_set and
b7676fcccab4 ("config: verify machine arch flag") would break the build as it
tries to test -march=native.
>
> Acked-by: Stephen Hemminger
[PATCH 1/2] net/virtio-user: improve kick performance with notification area mapping
This patch introduces new virtio-user callback to map the vq
notification area and implements it for the vhost-vDPA backend.
This is simply done by using mmap()/munmap() for
the vhost-vDPA fd.
This patch also adds a parameter for configuring feature bit
VIRTIO_NET_F_NOTIFICATION_DATA. If feature is disabled, also
update corresponding unsupported feature bit. And also adds
code to write to queue notify address in notify callback.
This will help in increasing the kick performance.
Signed-off-by: Srujana Challa
---
doc/guides/nics/virtio.rst| 5 ++
drivers/net/virtio/virtio_user/vhost.h| 1 +
drivers/net/virtio/virtio_user/vhost_vdpa.c | 56 ++
.../net/virtio/virtio_user/virtio_user_dev.c | 52 +++--
.../net/virtio/virtio_user/virtio_user_dev.h | 5 +-
drivers/net/virtio/virtio_user_ethdev.c | 57 ---
6 files changed, 162 insertions(+), 14 deletions(-)
diff --git a/doc/guides/nics/virtio.rst b/doc/guides/nics/virtio.rst
index c22ce56a02..11dd6359e5 100644
--- a/doc/guides/nics/virtio.rst
+++ b/doc/guides/nics/virtio.rst
@@ -349,6 +349,11 @@ Below devargs are supported by the virtio-user vdev:
election.
(Default: 0 (disabled))
+#. ``notification_data``:
+
+It is used to enable virtio device notification data feature.
+(Default: 1 (enabled))
+
Virtio paths Selection and Usage
diff --git a/drivers/net/virtio/virtio_user/vhost.h
b/drivers/net/virtio/virtio_user/vhost.h
index f817cab77a..1bce00c7ac 100644
--- a/drivers/net/virtio/virtio_user/vhost.h
+++ b/drivers/net/virtio/virtio_user/vhost.h
@@ -90,6 +90,7 @@ struct virtio_user_backend_ops {
int (*server_disconnect)(struct virtio_user_dev *dev);
int (*server_reconnect)(struct virtio_user_dev *dev);
int (*get_intr_fd)(struct virtio_user_dev *dev);
+ int (*map_notification_area)(struct virtio_user_dev *dev, bool map);
};
extern struct virtio_user_backend_ops virtio_ops_user;
diff --git a/drivers/net/virtio/virtio_user/vhost_vdpa.c
b/drivers/net/virtio/virtio_user/vhost_vdpa.c
index 2c36b26224..1eb0f9ec48 100644
--- a/drivers/net/virtio/virtio_user/vhost_vdpa.c
+++ b/drivers/net/virtio/virtio_user/vhost_vdpa.c
@@ -5,6 +5,7 @@
#include
#include
#include
+#include
#include
#include
#include
@@ -622,6 +623,60 @@ vhost_vdpa_get_intr_fd(struct virtio_user_dev *dev
__rte_unused)
return -1;
}
+static int
+unmap_notification_area(struct virtio_user_dev *dev, int nr_vrings)
+{
+ int i;
+
+ for (i = 0; i < nr_vrings; i++) {
+ if (dev->notify_area[i])
+ munmap(dev->notify_area[i], getpagesize());
+ }
+ free(dev->notify_area);
+
+ return 0;
+}
+
+static int
+vhost_vdpa_map_notification_area(struct virtio_user_dev *dev, bool map)
+{
+ struct vhost_vdpa_data *data = dev->backend_data;
+ int nr_vrings, i, page_size = getpagesize();
+ uint16_t **notify_area;
+
+ nr_vrings = dev->max_queue_pairs * 2;
+ if (dev->device_features & (1ull << VIRTIO_NET_F_CTRL_VQ))
+ nr_vrings++;
+
+ if (!map)
+ return unmap_notification_area(dev, nr_vrings);
+
+ notify_area = malloc(nr_vrings * sizeof(*notify_area));
+ if (!notify_area) {
+ PMD_DRV_LOG(ERR, "(%s) Failed to allocate notify area array",
dev->path);
+ return -1;
+ }
+ for (i = 0; i < nr_vrings; i++) {
+ notify_area[i] = mmap(NULL, page_size, PROT_WRITE, MAP_SHARED |
MAP_FILE,
+ data->vhostfd, i * page_size);
+ if (notify_area[i] == MAP_FAILED) {
+ PMD_DRV_LOG(ERR, "(%s) Map failed for notify address of
queue %d\n",
+ dev->path, i);
+ goto map_err;
+ }
+ }
+ dev->notify_area = notify_area;
+
+ return 0;
+
+map_err:
+ i--;
+ for (; i >= 0; i--)
+ munmap(notify_area[i], page_size);
+ free(notify_area);
+ return -1;
+}
+
struct virtio_user_backend_ops virtio_ops_vdpa = {
.setup = vhost_vdpa_setup,
.destroy = vhost_vdpa_destroy,
@@ -646,4 +701,5 @@ struct virtio_user_backend_ops virtio_ops_vdpa = {
.dma_unmap = vhost_vdpa_dma_unmap_batch,
.update_link_state = vhost_vdpa_update_link_state,
.get_intr_fd = vhost_vdpa_get_intr_fd,
+ .map_notification_area = vhost_vdpa_map_notification_area,
};
diff --git a/drivers/net/virtio/virtio_user/virtio_user_dev.c
b/drivers/net/virtio/virtio_user/virtio_user_dev.c
index af1f8c8237..578877d089 100644
--- a/drivers/net/virtio/virtio_user/virtio_user_dev.c
+++ b/drivers/net/virtio/virtio_user/virtio_user_dev.c
@@ -18,6 +18,7 @@
#include
#include
#include
+#include
#include "vhost.h"
#include "virtio_user_dev.h"
@@ -413,6 +414,12 @@ virtio_user_dev_init_noti
[PATCH 2/2] net/virtio-user: add VIRTIO_NET_F_RSS to supported features
This patch introduces new function to get rss device config
and adds code to forward the RSS control command to backend
through hw control queue if RSS feature is negotiated.
This patch will help to negotiate VIRTIO_NET_F_RSS feature
if vhost-vdpa backend supports RSS in HW.
Signed-off-by: Srujana Challa
---
.../net/virtio/virtio_user/virtio_user_dev.c | 31 ++-
.../net/virtio/virtio_user/virtio_user_dev.h | 2 ++
drivers/net/virtio/virtio_user_ethdev.c | 3 ++
3 files changed, 35 insertions(+), 1 deletion(-)
diff --git a/drivers/net/virtio/virtio_user/virtio_user_dev.c
b/drivers/net/virtio/virtio_user/virtio_user_dev.c
index 578877d089..b0937c9df9 100644
--- a/drivers/net/virtio/virtio_user/virtio_user_dev.c
+++ b/drivers/net/virtio/virtio_user/virtio_user_dev.c
@@ -304,6 +304,24 @@ virtio_user_dev_init_max_queue_pairs(struct
virtio_user_dev *dev, uint32_t user_
return 0;
}
+int
+virtio_user_dev_get_rss_config(struct virtio_user_dev *dev, void *dst, size_t
offset, int length)
+{
+ int ret = 0;
+
+ if (!(dev->device_features & (1ULL << VIRTIO_NET_F_RSS)))
+ return -ENOTSUP;
+
+ if (!dev->ops->get_config)
+ return -ENOTSUP;
+
+ ret = dev->ops->get_config(dev, dst, offset, length);
+ if (ret)
+ PMD_DRV_LOG(ERR, "(%s) Failed to get rss config in device",
dev->path);
+
+ return ret;
+}
+
int
virtio_user_dev_set_mac(struct virtio_user_dev *dev)
{
@@ -687,7 +705,8 @@ virtio_user_free_vrings(struct virtio_user_dev *dev)
1ULL << VIRTIO_F_IN_ORDER | \
1ULL << VIRTIO_F_VERSION_1 | \
1ULL << VIRTIO_F_RING_PACKED | \
-1ULL << VIRTIO_F_NOTIFICATION_DATA)
+1ULL << VIRTIO_F_NOTIFICATION_DATA | \
+1ULL << VIRTIO_NET_F_RSS)
int
virtio_user_dev_init(struct virtio_user_dev *dev, char *path, uint16_t queues,
@@ -903,6 +922,11 @@ virtio_user_handle_ctrl_msg_split(struct virtio_user_dev
*dev, struct vring *vri
queues = *(uint16_t *)(uintptr_t)vring->desc[idx_data].addr;
status = virtio_user_handle_mq(dev, queues);
+ } else if (hdr->class == VIRTIO_NET_CTRL_MQ && hdr->cmd ==
VIRTIO_NET_CTRL_MQ_RSS_CONFIG) {
+ struct virtio_net_ctrl_rss *rss;
+
+ rss = (struct virtio_net_ctrl_rss
*)(uintptr_t)vring->desc[idx_data].addr;
+ status = virtio_user_handle_mq(dev, rss->max_tx_vq);
} else if (hdr->class == VIRTIO_NET_CTRL_RX ||
hdr->class == VIRTIO_NET_CTRL_MAC ||
hdr->class == VIRTIO_NET_CTRL_VLAN) {
@@ -964,6 +988,11 @@ virtio_user_handle_ctrl_msg_packed(struct virtio_user_dev
*dev,
queues = *(uint16_t *)(uintptr_t)
vring->desc[idx_data].addr;
status = virtio_user_handle_mq(dev, queues);
+ } else if (hdr->class == VIRTIO_NET_CTRL_MQ && hdr->cmd ==
VIRTIO_NET_CTRL_MQ_RSS_CONFIG) {
+ struct virtio_net_ctrl_rss *rss;
+
+ rss = (struct virtio_net_ctrl_rss
*)(uintptr_t)vring->desc[idx_data].addr;
+ status = virtio_user_handle_mq(dev, rss->max_tx_vq);
} else if (hdr->class == VIRTIO_NET_CTRL_RX ||
hdr->class == VIRTIO_NET_CTRL_MAC ||
hdr->class == VIRTIO_NET_CTRL_VLAN) {
diff --git a/drivers/net/virtio/virtio_user/virtio_user_dev.h
b/drivers/net/virtio/virtio_user/virtio_user_dev.h
index 29ec386da5..39b3eec0f2 100644
--- a/drivers/net/virtio/virtio_user/virtio_user_dev.h
+++ b/drivers/net/virtio/virtio_user/virtio_user_dev.h
@@ -86,6 +86,8 @@ int virtio_user_dev_update_status(struct virtio_user_dev
*dev);
int virtio_user_dev_update_link_state(struct virtio_user_dev *dev);
int virtio_user_dev_set_mac(struct virtio_user_dev *dev);
int virtio_user_dev_get_mac(struct virtio_user_dev *dev);
+int virtio_user_dev_get_rss_config(struct virtio_user_dev *dev, void *dst,
size_t offset,
+ int length);
void virtio_user_dev_delayed_disconnect_handler(void *param);
int virtio_user_dev_server_reconnect(struct virtio_user_dev *dev);
extern const char * const virtio_user_backend_strings[];
diff --git a/drivers/net/virtio/virtio_user_ethdev.c
b/drivers/net/virtio/virtio_user_ethdev.c
index 241465ecdd..6c10e8f6c0 100644
--- a/drivers/net/virtio/virtio_user_ethdev.c
+++ b/drivers/net/virtio/virtio_user_ethdev.c
@@ -52,6 +52,9 @@ virtio_user_read_dev_config(struct virtio_hw *hw, size_t
offset,
if (offset == offsetof(struct virtio_net_config, max_virtqueue_pairs))
*(uint16_t *)dst = dev->max_queue_pairs;
+
+ if (offset >= offsetof(struct virtio_net_config, rss_max_key_size))
+ virtio_user_dev_get_rss_config(dev, dst, offset, length);
}
static void
--
2.25.1
[PATCH] bus/uacce: introduce UACCE bus
UACCE (Unified/User-space-access-intended Accelerator Framework) was
upstream to Linux kernel version 5.7, and it targets to provide Shared
Virtual Addressing (SVA) between accelerators and processes. So
accelerator can access any data structure of the main cpu. [1] for more
information.
This commit introduces UACCE bus, so that the accelerator devices could
seen at DPDK and could be further registered as a compress, crypto, dma
and ethdev device.
[1] https://docs.kernel.org/misc-devices/uacce.html
Signed-off-by: Chengwen Feng
---
MAINTAINERS | 4 +
drivers/bus/meson.build | 1 +
drivers/bus/uacce/bus_uacce_driver.h | 254 ++
drivers/bus/uacce/meson.build| 12 +
drivers/bus/uacce/uacce.c| 702 +++
drivers/bus/uacce/version.map| 15 +
6 files changed, 988 insertions(+)
create mode 100644 drivers/bus/uacce/bus_uacce_driver.h
create mode 100644 drivers/bus/uacce/meson.build
create mode 100644 drivers/bus/uacce/uacce.c
create mode 100644 drivers/bus/uacce/version.map
diff --git a/MAINTAINERS b/MAINTAINERS
index 0d1c8126e3..89711029d5 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -604,6 +604,10 @@ Platform bus driver
M: Tomasz Duszynski
F: drivers/bus/platform/
+UACCE bus driver
+M: Chengwen Feng
+F: drivers/bus/uacce/
+
VDEV bus driver
F: drivers/bus/vdev/
F: app/test/test_vdev.c
diff --git a/drivers/bus/meson.build b/drivers/bus/meson.build
index a78b4283bf..d67db8576b 100644
--- a/drivers/bus/meson.build
+++ b/drivers/bus/meson.build
@@ -9,6 +9,7 @@ drivers = [
'ifpga',
'pci',
'platform',
+'uacce',
'vdev',
'vmbus',
]
diff --git a/drivers/bus/uacce/bus_uacce_driver.h
b/drivers/bus/uacce/bus_uacce_driver.h
new file mode 100644
index 00..0276154658
--- /dev/null
+++ b/drivers/bus/uacce/bus_uacce_driver.h
@@ -0,0 +1,254 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2023 HiSilicon Limited
+ */
+
+#ifndef BUS_UACCE_DRIVER_H
+#define BUS_UACCE_DRIVER_H
+
+/**
+ * @file
+ *
+ * HiSilicon UACCE bus interface.
+ */
+
+#include
+#include
+#include
+
+#include
+#include
+#include
+
+#ifdef __cplusplus
+extern "C" {
+#endif /* __cplusplus */
+
+#define RTE_UACCE_DEV_PATH_SIZE256
+#define RTE_UACCE_API_NAME_SIZE64
+#define RTE_UACCE_ALGS_NAME_SIZE 384
+#define RTE_UACCE_ATTR_MAX_SIZE384
+
+/*
+ * Definition for queue file region type.
+ */
+enum rte_uacce_qfrt {
+ RTE_UACCE_QFRT_MMIO = 0, /**< Device mmio region. */
+ RTE_UACCE_QFRT_DUS, /**< Device user share region. */
+ RTE_UACCE_QFRT_BUTT
+};
+
+struct rte_uacce_driver;
+
+/**
+ * A structure describing a UACCE device.
+ */
+struct rte_uacce_device {
+ RTE_TAILQ_ENTRY(rte_uacce_device) next; /**< Next in device list. */
+ struct rte_device device;/**< Inherit core device. */
+ struct rte_uacce_driver *driver; /**< Driver used in probing. */
+ char name[RTE_DEV_NAME_MAX_LEN]; /**< Device name. */
+ char dev_root[RTE_UACCE_DEV_PATH_SIZE]; /**< Sysfs path with device
name. */
+ char cdev_path[RTE_UACCE_DEV_PATH_SIZE]; /**< Device path in devfs. */
+ char api[RTE_UACCE_API_NAME_SIZE]; /**< Device context type. */
+ char algs[RTE_UACCE_ALGS_NAME_SIZE]; /**< Device supported
algorithms. */
+ uint32_t flags; /**< Device flags. */
+ int numa_node; /**< NUMA node connection, -1
if unknown. */
+ uint32_t qfrt_sz[RTE_UACCE_QFRT_BUTT]; /**< Queue file region type's
size. */
+};
+
+/**
+ * @internal
+ * Helper macro for drivers that need to convert to struct rte_uacce_device.
+ */
+#define RTE_DEV_TO_UACCE_DEV(ptr) \
+ container_of(ptr, struct rte_uacce_device, device)
+
+#define RTE_DEV_TO_UACCE_DEV_CONST(ptr) \
+ container_of(ptr, const struct rte_uacce_device, device)
+
+/**
+ * A structure describing an ID for a UACCE driver. Each driver provides a
+ * table of these IDs for each device that it supports.
+ */
+struct rte_uacce_id {
+ const char *dev_api; /**< Device context type. */
+ /** Device algorithm.
+* If this field is NULL, only dev_api is matched. Otherwise, in
+* addition to match dev_api, dev_alg must be a subset of device's
+* algs.
+*/
+ const char *dev_alg;
+};
+
+/**
+ * Initialization function for the driver called during probing.
+ */
+typedef int (rte_uacce_probe_t)(struct rte_uacce_driver *, struct
rte_uacce_device *);
+
+/**
+ * Uninitialization function for the driver called during hotplugging.
+ */
+typedef int (rte_uacce_remove_t)(struct rte_uacce_device *);
+
+/**
+ * A structure describing a UACCE driver.
+ */
+struct rte_uacce_driver {
+ RTE_TAILQ_ENTRY(rte_uacce_driver) next; /**< Next in list. */
+ struct rte_driver driver;
[Bug 1334] Failures of testpmd on XXV710 card in VMs under Openshift Container Platform
https://bugs.dpdk.org/show_bug.cgi?id=1334 Bug ID: 1334 Summary: Failures of testpmd on XXV710 card in VMs under Openshift Container Platform Product: DPDK Version: 22.11 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: major Priority: Normal Component: testpmd Assignee: dev@dpdk.org Reporter: niny.gu.2...@gmail.com Target Milestone: --- I met the following issue when running dpdk-testpmd under the OpenShift Virtualization environment recently. We did the test on 2 VFs, each passed through to a VM, and they come from the same PF(XXV710) on the host. However, it's found there are the following errors when I quit the test on the receiver side. # dpdk-testpmd --socket-mem 256 --lcores 0,1 -n 1 -a 00:05:00.0 --proc-type auto --file-prefix rx -- --forward-mode=rxonly --port-topology=chained --disable-rss -i --rxq=1 --txq=1 --rxd=256 --txd=256 --nb-cores=1 --auto-start EAL: Detected CPU lcores: 2 EAL: Detected NUMA nodes: 1 EAL: Auto-detected process type: PRIMARY EAL: Detected shared linkage of DPDK EAL: Multi-process socket /var/run/dpdk/rx/mp_socket EAL: Selected IOVA mode 'PA' EAL: VFIO support initialized EAL: Using IOMMU type 8 (No-IOMMU) EAL: Probe PCI driver: net_iavf (8086:154c) device: :05:00.0 (socket -1) .. iavf_dev_init_vlan(): Failed to update vlan offload iavf_dev_configure(): configure VLAN failed: -95 iavf_set_rx_function(): request RXDID[1] in Queue[0] is legacy, set rx_pkt_burst as legacy for all queues .. testpmd> quit Telling cores to stop... Waiting for lcores to finish... -- Forward statistics for port 0 -- RX-packets: 158837911 RX-dropped: 0 RX-total: 158837911 TX-packets: 0 TX-dropped: 0 TX-total: 0 +++ Accumulated forward statistics for all ports+++ RX-packets: 158837911 RX-dropped: 0 RX-total: 158837911 TX-packets: 0 TX-dropped: 0 TX-total: 0 Done. Stopping port 0... Stopping ports... iavf_execute_vf_cmd(): Return failure -37 for cmd 9 iavf_disable_queues(): Failed to execute command of OP_DISABLE_QUEUES iavf_stop_queues(): Fail to stop queues iavf_stop_queues(): Fail to stop queues iavf_execute_vf_cmd(): No response for cmd 11 iavf_add_del_all_mac_addr(): fail to execute command OP_DEL_ETHER_ADDRESS Done Shutting down port 0... Closing ports... Port 0 is closed Done Bye... -- You are receiving this mail because: You are the assignee for the bug.
Re: [PATCH v2 4/4] net/hns3: use stdatomic API
On 2023/12/7 21:57, Ferruh Yigit wrote: On 12/7/2023 1:42 AM, Jie Hai wrote: Replace the use of gcc builtin __atomic_xxx intrinsics with corresponding rte_atomic_xxx optional stdatomic API. Signed-off-by: Jie Hai <...> static void @@ -643,8 +643,8 @@ hns3vf_get_push_lsc_cap(struct hns3_hw *hw) uint16_t exp = HNS3_PF_PUSH_LSC_CAP_UNKNOWN; struct hns3_vf *vf = HNS3_DEV_HW_TO_VF(hw); - __atomic_store_n(&vf->pf_push_lsc_cap, HNS3_PF_PUSH_LSC_CAP_UNKNOWN, -__ATOMIC_RELEASE); + rte_atomic_load_explicit(&vf->pf_push_lsc_cap, + HNS3_PF_PUSH_LSC_CAP_UNKNOWN, rte_memory_order_release); I guess intension here is to use 'rte_atomic_store_explicit()', current code cause a build error [1]. [1] https://mails.dpdk.org/archives/test-report/2023-December/525759.html Hi, Ferruh, Thank, It's my mistake to use 'rte_atomic_load_explicit', will fix it. Best regards, Jie Hai .
[PATCH v4 0/4] net/hns3: refactor mailbox
This patchset refactors mailbox codes. We will send a patch fix for all __atomic_xxx, so this patchset still use __atomic_xxx. -- v4: 1. use __atomic_xxx instead of rte_atomic_XXX. 2. use '__rte_packed' instead of '#pragma pack()'. v3: 1. fix the checkpatch warning on __atomic_xxx. -- Dengdui Huang (4): net/hns3: refactor VF mailbox message struct net/hns3: refactor PF mailbox message struct net/hns3: refactor send mailbox function net/hns3: refactor handle mailbox function drivers/net/hns3/hns3_ethdev.c| 2 +- drivers/net/hns3/hns3_ethdev_vf.c | 181 +- drivers/net/hns3/hns3_mbx.c | 165 +++ drivers/net/hns3/hns3_mbx.h | 92 +++ drivers/net/hns3/hns3_rxtx.c | 18 ++- 5 files changed, 276 insertions(+), 182 deletions(-) -- 2.30.0
[PATCH v4 1/4] net/hns3: refactor VF mailbox message struct
From: Dengdui Huang
The data region in VF to PF mbx memssage command is
used to communicate with PF driver. And this data
region exists as an array. As a result, some complicated
feature commands, like setting promisc mode, map/unmap
ring vector and setting VLAN id, have to use magic number
to set them. This isn't good for maintenance of driver.
So this patch refactors these messages by extracting an
hns3_vf_to_pf_msg structure.
In addition, the PF link change event message is reported
by the firmware and is reported in hns3_mbx_vf_to_pf_cmd
format, it also needs to be modified.
Fixes: 463e748964f5 ("net/hns3: support mailbox")
Cc: sta...@dpdk.org
Signed-off-by: Dengdui Huang
Signed-off-by: Jie Hai
---
drivers/net/hns3/hns3_ethdev_vf.c | 54 ++---
drivers/net/hns3/hns3_mbx.c | 24 ++---
drivers/net/hns3/hns3_mbx.h | 56 ++-
3 files changed, 76 insertions(+), 58 deletions(-)
diff --git a/drivers/net/hns3/hns3_ethdev_vf.c
b/drivers/net/hns3/hns3_ethdev_vf.c
index 916cc0fb1b62..19e734ca8d8e 100644
--- a/drivers/net/hns3/hns3_ethdev_vf.c
+++ b/drivers/net/hns3/hns3_ethdev_vf.c
@@ -254,11 +254,12 @@ hns3vf_set_promisc_mode(struct hns3_hw *hw, bool
en_bc_pmc,
*the packets with vlan tag in promiscuous mode.
*/
hns3_cmd_setup_basic_desc(&desc, HNS3_OPC_MBX_VF_TO_PF, false);
- req->msg[0] = HNS3_MBX_SET_PROMISC_MODE;
- req->msg[1] = en_bc_pmc ? 1 : 0;
- req->msg[2] = en_uc_pmc ? 1 : 0;
- req->msg[3] = en_mc_pmc ? 1 : 0;
- req->msg[4] = hw->promisc_mode == HNS3_LIMIT_PROMISC_MODE ? 1 : 0;
+ req->msg.code = HNS3_MBX_SET_PROMISC_MODE;
+ req->msg.en_bc = en_bc_pmc ? 1 : 0;
+ req->msg.en_uc = en_uc_pmc ? 1 : 0;
+ req->msg.en_mc = en_mc_pmc ? 1 : 0;
+ req->msg.en_limit_promisc =
+ hw->promisc_mode == HNS3_LIMIT_PROMISC_MODE ? 1 : 0;
ret = hns3_cmd_send(hw, &desc, 1);
if (ret)
@@ -347,30 +348,28 @@ hns3vf_bind_ring_with_vector(struct hns3_hw *hw, uint16_t
vector_id,
bool mmap, enum hns3_ring_type queue_type,
uint16_t queue_id)
{
- struct hns3_vf_bind_vector_msg bind_msg;
+#define HNS3_RING_VERCTOR_DATA_SIZE14
+ struct hns3_vf_to_pf_msg req = {0};
const char *op_str;
- uint16_t code;
int ret;
- memset(&bind_msg, 0, sizeof(bind_msg));
- code = mmap ? HNS3_MBX_MAP_RING_TO_VECTOR :
+ req.code = mmap ? HNS3_MBX_MAP_RING_TO_VECTOR :
HNS3_MBX_UNMAP_RING_TO_VECTOR;
- bind_msg.vector_id = (uint8_t)vector_id;
+ req.vector_id = (uint8_t)vector_id;
+ req.ring_num = 1;
if (queue_type == HNS3_RING_TYPE_RX)
- bind_msg.param[0].int_gl_index = HNS3_RING_GL_RX;
+ req.ring_param[0].int_gl_index = HNS3_RING_GL_RX;
else
- bind_msg.param[0].int_gl_index = HNS3_RING_GL_TX;
-
- bind_msg.param[0].ring_type = queue_type;
- bind_msg.ring_num = 1;
- bind_msg.param[0].tqp_index = queue_id;
+ req.ring_param[0].int_gl_index = HNS3_RING_GL_TX;
+ req.ring_param[0].ring_type = queue_type;
+ req.ring_param[0].tqp_index = queue_id;
op_str = mmap ? "Map" : "Unmap";
- ret = hns3_send_mbx_msg(hw, code, 0, (uint8_t *)&bind_msg,
- sizeof(bind_msg), false, NULL, 0);
+ ret = hns3_send_mbx_msg(hw, req.code, 0, (uint8_t *)&req.vector_id,
+ HNS3_RING_VERCTOR_DATA_SIZE, false, NULL, 0);
if (ret)
- hns3_err(hw, "%s TQP %u fail, vector_id is %u, ret is %d.",
-op_str, queue_id, bind_msg.vector_id, ret);
+ hns3_err(hw, "%s TQP %u fail, vector_id is %u, ret = %d.",
+op_str, queue_id, req.vector_id, ret);
return ret;
}
@@ -965,19 +964,16 @@ hns3vf_update_link_status(struct hns3_hw *hw, uint8_t
link_status,
static int
hns3vf_vlan_filter_configure(struct hns3_adapter *hns, uint16_t vlan_id, int
on)
{
-#define HNS3VF_VLAN_MBX_MSG_LEN 5
+ struct hns3_mbx_vlan_filter vlan_filter = {0};
struct hns3_hw *hw = &hns->hw;
- uint8_t msg_data[HNS3VF_VLAN_MBX_MSG_LEN];
- uint16_t proto = htons(RTE_ETHER_TYPE_VLAN);
- uint8_t is_kill = on ? 0 : 1;
- msg_data[0] = is_kill;
- memcpy(&msg_data[1], &vlan_id, sizeof(vlan_id));
- memcpy(&msg_data[3], &proto, sizeof(proto));
+ vlan_filter.is_kill = on ? 0 : 1;
+ vlan_filter.proto = rte_cpu_to_le_16(RTE_ETHER_TYPE_VLAN);
+ vlan_filter.vlan_id = rte_cpu_to_le_16(vlan_id);
return hns3_send_mbx_msg(hw, HNS3_MBX_SET_VLAN, HNS3_MBX_VLAN_FILTER,
-msg_data, HNS3VF_VLAN_MBX_MSG_LEN, true, NULL,
-0);
+ (uint8_t *)&vlan_filter, sizeof(vlan_filter),
+
[PATCH v4 2/4] net/hns3: refactor PF mailbox message struct
From: Dengdui Huang
The data region in PF to VF mbx memssage command is used
to communicate with VF driver. And this data region exists
as an array. As a result, some complicated feature commands,
like mailbox response, link change event, close promisc mode,
reset request and update pvid state, have to use magic number
to set them. This isn't good for maintenance of driver. So
this patch refactors these messages by extracting an
hns3_pf_to_vf_msg structure.
Fixes: 463e748964f5 ("net/hns3: support mailbox")
Cc: sta...@dpdk.org
Signed-off-by: Dengdui Huang
Signed-off-by: Jie Hai
---
drivers/net/hns3/hns3_mbx.c | 38 ++---
drivers/net/hns3/hns3_mbx.h | 25 +++-
2 files changed, 43 insertions(+), 20 deletions(-)
diff --git a/drivers/net/hns3/hns3_mbx.c b/drivers/net/hns3/hns3_mbx.c
index ad5ec555b39e..c90f5d59ba21 100644
--- a/drivers/net/hns3/hns3_mbx.c
+++ b/drivers/net/hns3/hns3_mbx.c
@@ -192,17 +192,17 @@ static void
hns3vf_handle_link_change_event(struct hns3_hw *hw,
struct hns3_mbx_pf_to_vf_cmd *req)
{
+ struct hns3_mbx_link_status *link_info =
+ (struct hns3_mbx_link_status *)req->msg.msg_data;
uint8_t link_status, link_duplex;
- uint16_t *msg_q = req->msg;
uint8_t support_push_lsc;
uint32_t link_speed;
- memcpy(&link_speed, &msg_q[2], sizeof(link_speed));
- link_status = rte_le_to_cpu_16(msg_q[1]);
- link_duplex = (uint8_t)rte_le_to_cpu_16(msg_q[4]);
- hns3vf_update_link_status(hw, link_status, link_speed,
- link_duplex);
- support_push_lsc = (*(uint8_t *)&msg_q[5]) & 1u;
+ link_status = (uint8_t)rte_le_to_cpu_16(link_info->link_status);
+ link_speed = rte_le_to_cpu_32(link_info->speed);
+ link_duplex = (uint8_t)rte_le_to_cpu_16(link_info->duplex);
+ hns3vf_update_link_status(hw, link_status, link_speed, link_duplex);
+ support_push_lsc = (link_info->flag) & 1u;
hns3vf_update_push_lsc_cap(hw, support_push_lsc);
}
@@ -211,7 +211,6 @@ hns3_handle_asserting_reset(struct hns3_hw *hw,
struct hns3_mbx_pf_to_vf_cmd *req)
{
enum hns3_reset_level reset_level;
- uint16_t *msg_q = req->msg;
/*
* PF has asserted reset hence VF should go in pending
@@ -219,7 +218,7 @@ hns3_handle_asserting_reset(struct hns3_hw *hw,
* has been completely reset. After this stack should
* eventually be re-initialized.
*/
- reset_level = rte_le_to_cpu_16(msg_q[1]);
+ reset_level = rte_le_to_cpu_16(req->msg.reset_level);
hns3_atomic_set_bit(reset_level, &hw->reset.pending);
hns3_warn(hw, "PF inform reset level %d", reset_level);
@@ -241,8 +240,9 @@ hns3_handle_mbx_response(struct hns3_hw *hw, struct
hns3_mbx_pf_to_vf_cmd *req)
* to match the request.
*/
if (req->match_id == resp->match_id) {
- resp->resp_status = hns3_resp_to_errno(req->msg[3]);
- memcpy(resp->additional_info, &req->msg[4],
+ resp->resp_status =
+ hns3_resp_to_errno(req->msg.resp_status);
+ memcpy(resp->additional_info, &req->msg.resp_data,
HNS3_MBX_MAX_RESP_DATA_SIZE);
rte_io_wmb();
resp->received_match_resp = true;
@@ -255,7 +255,8 @@ hns3_handle_mbx_response(struct hns3_hw *hw, struct
hns3_mbx_pf_to_vf_cmd *req)
* support copy request's match_id to its response. So VF follows the
* original scheme to process.
*/
- msg_data = (uint32_t)req->msg[1] << HNS3_MBX_RESP_CODE_OFFSET |
req->msg[2];
+ msg_data = (uint32_t)req->msg.vf_mbx_msg_code <<
+ HNS3_MBX_RESP_CODE_OFFSET | req->msg.vf_mbx_msg_subcode;
if (resp->req_msg_data != msg_data) {
hns3_warn(hw,
"received response tag (%u) is mismatched with
requested tag (%u)",
@@ -263,8 +264,8 @@ hns3_handle_mbx_response(struct hns3_hw *hw, struct
hns3_mbx_pf_to_vf_cmd *req)
return;
}
- resp->resp_status = hns3_resp_to_errno(req->msg[3]);
- memcpy(resp->additional_info, &req->msg[4],
+ resp->resp_status = hns3_resp_to_errno(req->msg.resp_status);
+ memcpy(resp->additional_info, &req->msg.resp_data,
HNS3_MBX_MAX_RESP_DATA_SIZE);
rte_io_wmb();
resp->received_match_resp = true;
@@ -305,8 +306,7 @@ static void
hns3_update_port_base_vlan_info(struct hns3_hw *hw,
struct hns3_mbx_pf_to_vf_cmd *req)
{
-#define PVID_STATE_OFFSET 1
- uint16_t new_pvid_state = req->msg[PVID_STATE_OFFSET] ?
+ uint16_t new_pvid_state = req->msg.pvid_state ?
HNS3_PORT_BASE_VLAN_ENABLE : HNS3_PORT_BA
[PATCH v4 3/4] net/hns3: refactor send mailbox function
From: Dengdui Huang
The 'hns3_send_mbx_msg' function has following problem:
1. the name is vague, missing caller indication
2. too many input parameters because the filling messages
are placed in commands the send command.
Therefore, a common interface is encapsulated to fill in
the mailbox message before sending it.
Fixes: 463e748964f5 ("net/hns3: support mailbox")
Cc: sta...@dpdk.org
Signed-off-by: Dengdui Huang
Signed-off-by: Jie Hai
---
drivers/net/hns3/hns3_ethdev_vf.c | 141 ++
drivers/net/hns3/hns3_mbx.c | 50 ---
drivers/net/hns3/hns3_mbx.h | 8 +-
drivers/net/hns3/hns3_rxtx.c | 18 ++--
4 files changed, 116 insertions(+), 101 deletions(-)
diff --git a/drivers/net/hns3/hns3_ethdev_vf.c
b/drivers/net/hns3/hns3_ethdev_vf.c
index 19e734ca8d8e..b0d0c29df191 100644
--- a/drivers/net/hns3/hns3_ethdev_vf.c
+++ b/drivers/net/hns3/hns3_ethdev_vf.c
@@ -91,11 +91,13 @@ hns3vf_add_uc_mac_addr(struct hns3_hw *hw, struct
rte_ether_addr *mac_addr)
{
/* mac address was checked by upper level interface */
char mac_str[RTE_ETHER_ADDR_FMT_SIZE];
+ struct hns3_vf_to_pf_msg req;
int ret;
- ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_UNICAST,
- HNS3_MBX_MAC_VLAN_UC_ADD, mac_addr->addr_bytes,
- RTE_ETHER_ADDR_LEN, false, NULL, 0);
+ hns3vf_mbx_setup(&req, HNS3_MBX_SET_UNICAST,
+HNS3_MBX_MAC_VLAN_UC_ADD);
+ memcpy(req.data, mac_addr->addr_bytes, RTE_ETHER_ADDR_LEN);
+ ret = hns3vf_mbx_send(hw, &req, false, NULL, 0);
if (ret) {
hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
mac_addr);
@@ -110,12 +112,13 @@ hns3vf_remove_uc_mac_addr(struct hns3_hw *hw, struct
rte_ether_addr *mac_addr)
{
/* mac address was checked by upper level interface */
char mac_str[RTE_ETHER_ADDR_FMT_SIZE];
+ struct hns3_vf_to_pf_msg req;
int ret;
- ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_UNICAST,
- HNS3_MBX_MAC_VLAN_UC_REMOVE,
- mac_addr->addr_bytes, RTE_ETHER_ADDR_LEN,
- false, NULL, 0);
+ hns3vf_mbx_setup(&req, HNS3_MBX_SET_UNICAST,
+HNS3_MBX_MAC_VLAN_UC_REMOVE);
+ memcpy(req.data, mac_addr->addr_bytes, RTE_ETHER_ADDR_LEN);
+ ret = hns3vf_mbx_send(hw, &req, false, NULL, 0);
if (ret) {
hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
mac_addr);
@@ -134,6 +137,7 @@ hns3vf_set_default_mac_addr(struct rte_eth_dev *dev,
struct rte_ether_addr *old_addr;
uint8_t addr_bytes[HNS3_TWO_ETHER_ADDR_LEN]; /* for 2 MAC addresses */
char mac_str[RTE_ETHER_ADDR_FMT_SIZE];
+ struct hns3_vf_to_pf_msg req;
int ret;
/*
@@ -146,9 +150,10 @@ hns3vf_set_default_mac_addr(struct rte_eth_dev *dev,
memcpy(&addr_bytes[RTE_ETHER_ADDR_LEN], old_addr->addr_bytes,
RTE_ETHER_ADDR_LEN);
- ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_UNICAST,
- HNS3_MBX_MAC_VLAN_UC_MODIFY, addr_bytes,
- HNS3_TWO_ETHER_ADDR_LEN, true, NULL, 0);
+ hns3vf_mbx_setup(&req, HNS3_MBX_SET_UNICAST,
+HNS3_MBX_MAC_VLAN_UC_MODIFY);
+ memcpy(req.data, addr_bytes, HNS3_TWO_ETHER_ADDR_LEN);
+ ret = hns3vf_mbx_send(hw, &req, true, NULL, 0);
if (ret) {
/*
* The hns3 VF PMD depends on the hns3 PF kernel ethdev
@@ -185,12 +190,13 @@ hns3vf_add_mc_mac_addr(struct hns3_hw *hw,
struct rte_ether_addr *mac_addr)
{
char mac_str[RTE_ETHER_ADDR_FMT_SIZE];
+ struct hns3_vf_to_pf_msg req;
int ret;
- ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_MULTICAST,
- HNS3_MBX_MAC_VLAN_MC_ADD,
- mac_addr->addr_bytes, RTE_ETHER_ADDR_LEN, false,
- NULL, 0);
+ hns3vf_mbx_setup(&req, HNS3_MBX_SET_MULTICAST,
+HNS3_MBX_MAC_VLAN_MC_ADD);
+ memcpy(req.data, mac_addr->addr_bytes, RTE_ETHER_ADDR_LEN);
+ ret = hns3vf_mbx_send(hw, &req, false, NULL, 0);
if (ret) {
hns3_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
mac_addr);
@@ -206,12 +212,13 @@ hns3vf_remove_mc_mac_addr(struct hns3_hw *hw,
struct rte_ether_addr *mac_addr)
{
char mac_str[RTE_ETHER_ADDR_FMT_SIZE];
+ struct hns3_vf_to_pf_msg req;
int ret;
- ret = hns3_send_mbx_msg(hw, HNS3_MBX_SET_MULTICAST,
- HNS3_MBX_MAC_VLAN_MC_REMOVE,
- mac_addr->addr_bytes, RTE_ETHER_ADDR_LEN, false,
-
[PATCH v4 4/4] net/hns3: refactor handle mailbox function
From: Dengdui Huang
The mailbox messages of the PF and VF are processed in
the same function. The PF and VF call the same function
to process the messages. This code is excessive coupling
and isn't good for maintenance. Therefore, this patch
separates the interfaces that handle PF mailbox message
and handle VF mailbox message.
Fixes: 463e748964f5 ("net/hns3: support mailbox")
Fixes: 109e4dd1bd7a ("net/hns3: get link state change through mailbox")
Cc: sta...@dpdk.org
Signed-off-by: Dengdui Huang
Signed-off-by: Jie Hai
---
drivers/net/hns3/hns3_ethdev.c| 2 +-
drivers/net/hns3/hns3_ethdev_vf.c | 4 +-
drivers/net/hns3/hns3_mbx.c | 69 ---
drivers/net/hns3/hns3_mbx.h | 3 +-
4 files changed, 58 insertions(+), 20 deletions(-)
diff --git a/drivers/net/hns3/hns3_ethdev.c b/drivers/net/hns3/hns3_ethdev.c
index ae81368f68ae..bccd9db0dd4d 100644
--- a/drivers/net/hns3/hns3_ethdev.c
+++ b/drivers/net/hns3/hns3_ethdev.c
@@ -380,7 +380,7 @@ hns3_interrupt_handler(void *param)
hns3_warn(hw, "received reset interrupt");
hns3_schedule_reset(hns);
} else if (event_cause == HNS3_VECTOR0_EVENT_MBX) {
- hns3_dev_handle_mbx_msg(hw);
+ hns3pf_handle_mbx_msg(hw);
} else if (event_cause != HNS3_VECTOR0_EVENT_PTP) {
hns3_warn(hw, "received unknown event: vector0_int_stat:0x%x "
"ras_int_stat:0x%x cmdq_int_stat:0x%x",
diff --git a/drivers/net/hns3/hns3_ethdev_vf.c
b/drivers/net/hns3/hns3_ethdev_vf.c
index b0d0c29df191..f5a7a2b1f46c 100644
--- a/drivers/net/hns3/hns3_ethdev_vf.c
+++ b/drivers/net/hns3/hns3_ethdev_vf.c
@@ -618,7 +618,7 @@ hns3vf_interrupt_handler(void *param)
hns3_schedule_reset(hns);
break;
case HNS3VF_VECTOR0_EVENT_MBX:
- hns3_dev_handle_mbx_msg(hw);
+ hns3vf_handle_mbx_msg(hw);
break;
default:
break;
@@ -670,7 +670,7 @@ hns3vf_get_push_lsc_cap(struct hns3_hw *hw)
* driver has to actively handle the HNS3_MBX_LINK_STAT_CHANGE
* mailbox from PF driver to get this capability.
*/
- hns3_dev_handle_mbx_msg(hw);
+ hns3vf_handle_mbx_msg(hw);
if (__atomic_load_n(&vf->pf_push_lsc_cap, __ATOMIC_ACQUIRE) !=
HNS3_PF_PUSH_LSC_CAP_UNKNOWN)
break;
diff --git a/drivers/net/hns3/hns3_mbx.c b/drivers/net/hns3/hns3_mbx.c
index 43195ff184b1..9cdbc1668a17 100644
--- a/drivers/net/hns3/hns3_mbx.c
+++ b/drivers/net/hns3/hns3_mbx.c
@@ -78,7 +78,7 @@ hns3_get_mbx_resp(struct hns3_hw *hw, uint16_t code, uint16_t
subcode,
return -EIO;
}
- hns3_dev_handle_mbx_msg(hw);
+ hns3vf_handle_mbx_msg(hw);
rte_delay_us(HNS3_WAIT_RESP_US);
if (hw->mbx_resp.received_match_resp)
@@ -372,9 +372,57 @@ hns3_handle_mbx_msg_out_intr(struct hns3_hw *hw)
}
void
-hns3_dev_handle_mbx_msg(struct hns3_hw *hw)
+hns3pf_handle_mbx_msg(struct hns3_hw *hw)
+{
+ struct hns3_cmq_ring *crq = &hw->cmq.crq;
+ struct hns3_mbx_vf_to_pf_cmd *req;
+ struct hns3_cmd_desc *desc;
+ uint16_t flag;
+
+ rte_spinlock_lock(&hw->cmq.crq.lock);
+
+ while (!hns3_cmd_crq_empty(hw)) {
+ if (__atomic_load_n(&hw->reset.disable_cmd, __ATOMIC_RELAXED)) {
+ rte_spinlock_unlock(&hw->cmq.crq.lock);
+ return;
+ }
+ desc = &crq->desc[crq->next_to_use];
+ req = (struct hns3_mbx_vf_to_pf_cmd *)desc->data;
+
+ flag = rte_le_to_cpu_16(crq->desc[crq->next_to_use].flag);
+ if (unlikely(!hns3_get_bit(flag, HNS3_CMDQ_RX_OUTVLD_B))) {
+ hns3_warn(hw,
+ "dropped invalid mailbox message, code = %u",
+ req->msg.code);
+
+ /* dropping/not processing this invalid message */
+ crq->desc[crq->next_to_use].flag = 0;
+ hns3_mbx_ring_ptr_move_crq(crq);
+ continue;
+ }
+
+ switch (req->msg.code) {
+ case HNS3_MBX_PUSH_LINK_STATUS:
+ hns3pf_handle_link_change_event(hw, req);
+ break;
+ default:
+ hns3_err(hw, "received unsupported(%u) mbx msg",
+req->msg.code);
+ break;
+ }
+ crq->desc[crq->next_to_use].flag = 0;
+ hns3_mbx_ring_ptr_move_crq(crq);
+ }
+
+ /* Write back CMDQ_RQ header pointer, IMP need this pointer */
+ hns3_write_dev(hw, HNS3_CMDQ_RX_HEAD_REG, crq->next_to_use);
+
+ rte_spinlock_unlock(&hw->cmq.crq.lock);
+}
+
+
Re: [PATCH v2 4/4] net/hns3: use stdatomic API
On 2023/12/8 1:42, Ferruh Yigit wrote: On 12/7/2023 1:57 PM, Ferruh Yigit wrote: On 12/7/2023 1:42 AM, Jie Hai wrote: Replace the use of gcc builtin __atomic_xxx intrinsics with corresponding rte_atomic_xxx optional stdatomic API. Signed-off-by: Jie Hai <...> static void @@ -643,8 +643,8 @@ hns3vf_get_push_lsc_cap(struct hns3_hw *hw) uint16_t exp = HNS3_PF_PUSH_LSC_CAP_UNKNOWN; struct hns3_vf *vf = HNS3_DEV_HW_TO_VF(hw); - __atomic_store_n(&vf->pf_push_lsc_cap, HNS3_PF_PUSH_LSC_CAP_UNKNOWN, -__ATOMIC_RELEASE); + rte_atomic_load_explicit(&vf->pf_push_lsc_cap, + HNS3_PF_PUSH_LSC_CAP_UNKNOWN, rte_memory_order_release); I guess intension here is to use 'rte_atomic_store_explicit()', current code cause a build error [1]. [1] https://mails.dpdk.org/archives/test-report/2023-December/525759.html Another build error is when stdatomic enabled [2]. [2] meson setup --werror -Denable_stdatomic=true build && ninja -C build Hi,Ferruh, Thanks, and I will fix it. Since both the following two patches[1][2](in different patchsets) uses __atomic_XXX, I will keep using __atomic_XX in the patches first. After the two patches are accepted, resend the fixed patch[3] and change all __atomic_XX to rte_atomic_XXX. [1][PATCH] net/hns3: fix VF multiple count on one reset [2][PATCH] net/hns3: refactor handle mailbox function [3][PATCH] net/hns3: use stdatomic API Best regards, Jie Hai
[PATCH v3 0/3] net/hns3: bugfix on reset
This patchset fixes some bugs on reset, and the replaces of gcc builtin __atomic_xxx with rte_atomic_xxx is removed and will send later separately. Dengdui Huang (3): net/hns3: fix VF multiple count on one reset net/hns3: fix disable command with firmware net/hns3: fix incorrect reset level comparison drivers/net/hns3/hns3_ethdev.c| 28 +++- drivers/net/hns3/hns3_ethdev_vf.c | 44 --- 2 files changed, 38 insertions(+), 34 deletions(-) -- 2.30.0
[PATCH v3 1/3] net/hns3: fix VF multiple count on one reset
From: Dengdui Huang
There are two ways for the hns3 VF driver to know reset event, namely,
interrupt task and periodic detection task. For the latter, the real
reset process will delay several microseconds to execute. Both tasks
cause the count to increase by 1.
However, the periodic detection task also detects a reset event A
after interrupt task receive a reset event A. As a result, the reset
count will be double.
So this patch adds the comparison of reset level for VF in case of the
multiple reset count.
Fixes: a5475d61fa34 ("net/hns3: support VF")
Cc: sta...@dpdk.org
Signed-off-by: Dengdui Huang
Signed-off-by: Jie Hai
---
drivers/net/hns3/hns3_ethdev_vf.c | 44 ---
1 file changed, 29 insertions(+), 15 deletions(-)
diff --git a/drivers/net/hns3/hns3_ethdev_vf.c
b/drivers/net/hns3/hns3_ethdev_vf.c
index 916cc0fb1b62..089df146f76e 100644
--- a/drivers/net/hns3/hns3_ethdev_vf.c
+++ b/drivers/net/hns3/hns3_ethdev_vf.c
@@ -563,13 +563,8 @@ hns3vf_check_event_cause(struct hns3_adapter *hns,
uint32_t *clearval)
val = hns3_read_dev(hw, HNS3_VF_RST_ING);
hns3_write_dev(hw, HNS3_VF_RST_ING, val | HNS3_VF_RST_ING_BIT);
val = cmdq_stat_reg & ~BIT(HNS3_VECTOR0_RST_INT_B);
- if (clearval) {
- hw->reset.stats.global_cnt++;
- hns3_warn(hw, "Global reset detected, clear reset
status");
- } else {
- hns3_schedule_delayed_reset(hns);
- hns3_warn(hw, "Global reset detected, don't clear reset
status");
- }
+ hw->reset.stats.global_cnt++;
+ hns3_warn(hw, "Global reset detected, clear reset status");
ret = HNS3VF_VECTOR0_EVENT_RST;
goto out;
@@ -584,9 +579,9 @@ hns3vf_check_event_cause(struct hns3_adapter *hns, uint32_t
*clearval)
val = 0;
ret = HNS3VF_VECTOR0_EVENT_OTHER;
+
out:
- if (clearval)
- *clearval = val;
+ *clearval = val;
return ret;
}
@@ -1709,11 +1704,25 @@ is_vf_reset_done(struct hns3_hw *hw)
return true;
}
+static enum hns3_reset_level
+hns3vf_detect_reset_event(struct hns3_hw *hw)
+{
+ enum hns3_reset_level reset = HNS3_NONE_RESET;
+ uint32_t cmdq_stat_reg;
+
+ cmdq_stat_reg = hns3_read_dev(hw, HNS3_VECTOR0_CMDQ_STAT_REG);
+ if (BIT(HNS3_VECTOR0_RST_INT_B) & cmdq_stat_reg)
+ reset = HNS3_VF_RESET;
+
+ return reset;
+}
+
bool
hns3vf_is_reset_pending(struct hns3_adapter *hns)
{
+ enum hns3_reset_level last_req;
struct hns3_hw *hw = &hns->hw;
- enum hns3_reset_level reset;
+ enum hns3_reset_level new_req;
/*
* According to the protocol of PCIe, FLR to a PF device resets the PF
@@ -1736,13 +1745,18 @@ hns3vf_is_reset_pending(struct hns3_adapter *hns)
if (rte_eal_process_type() != RTE_PROC_PRIMARY)
return false;
- hns3vf_check_event_cause(hns, NULL);
- reset = hns3vf_get_reset_level(hw, &hw->reset.pending);
- if (hw->reset.level != HNS3_NONE_RESET && reset != HNS3_NONE_RESET &&
- hw->reset.level < reset) {
- hns3_warn(hw, "High level reset %d is pending", reset);
+ new_req = hns3vf_detect_reset_event(hw);
+ if (new_req == HNS3_NONE_RESET)
+ return false;
+
+ last_req = hns3vf_get_reset_level(hw, &hw->reset.pending);
+ if (last_req == HNS3_NONE_RESET || last_req < new_req) {
+ __atomic_store_n(&hw->reset.disable_cmd, 1, __ATOMIC_RELAXED);
+ hns3_schedule_delayed_reset(hns);
+ hns3_warn(hw, "High level reset detected, delay do reset");
return true;
}
+
return false;
}
--
2.30.0
[PATCH v3 2/3] net/hns3: fix disable command with firmware
From: Dengdui Huang
Disable command only when need to delay handle reset.
This patch fixes it.
Fixes: 5be38fc6c0fc ("net/hns3: fix multiple reset detected log")
Cc: sta...@dpdk.org
Signed-off-by: Dengdui Huang
Signed-off-by: Jie Hai
---
drivers/net/hns3/hns3_ethdev.c | 8 +++-
1 file changed, 3 insertions(+), 5 deletions(-)
diff --git a/drivers/net/hns3/hns3_ethdev.c b/drivers/net/hns3/hns3_ethdev.c
index ae81368f68ae..76fc401bd62c 100644
--- a/drivers/net/hns3/hns3_ethdev.c
+++ b/drivers/net/hns3/hns3_ethdev.c
@@ -5552,18 +5552,16 @@ hns3_detect_reset_event(struct hns3_hw *hw)
last_req = hns3_get_reset_level(hns, &hw->reset.pending);
vector0_intr_state = hns3_read_dev(hw, HNS3_VECTOR0_OTHER_INT_STS_REG);
- if (BIT(HNS3_VECTOR0_IMPRESET_INT_B) & vector0_intr_state) {
- __atomic_store_n(&hw->reset.disable_cmd, 1, __ATOMIC_RELAXED);
+ if (BIT(HNS3_VECTOR0_IMPRESET_INT_B) & vector0_intr_state)
new_req = HNS3_IMP_RESET;
- } else if (BIT(HNS3_VECTOR0_GLOBALRESET_INT_B) & vector0_intr_state) {
- __atomic_store_n(&hw->reset.disable_cmd, 1, __ATOMIC_RELAXED);
+ else if (BIT(HNS3_VECTOR0_GLOBALRESET_INT_B) & vector0_intr_state)
new_req = HNS3_GLOBAL_RESET;
- }
if (new_req == HNS3_NONE_RESET)
return HNS3_NONE_RESET;
if (last_req == HNS3_NONE_RESET || last_req < new_req) {
+ __atomic_store_n(&hw->reset.disable_cmd, 1, __ATOMIC_RELAXED);
hns3_schedule_delayed_reset(hns);
hns3_warn(hw, "High level reset detected, delay do reset");
}
--
2.30.0
[PATCH v3 3/3] net/hns3: fix incorrect reset level comparison
From: Dengdui Huang
Currently, there are two problems in hns3vf_is_reset_pending():
1. The new detect reset level is not HNS3_NONE_RESET, but the
last reset level is HNS3_NONE_RESET, this function returns false.
2. Comparison between last_req and new_req is opposite.
In addition, the reset level comparison in hns3_detect_reset_event()
is similar to the hns3vf_is_reset_pending(). So this patch fixes
above the problems and merges the logic of reset level comparison.
Fixes: 5be38fc6c0fc ("net/hns3: fix multiple reset detected log")
Cc: sta...@dpdk.org
Signed-off-by: Dengdui Huang
Signed-off-by: Jie Hai
---
drivers/net/hns3/hns3_ethdev.c | 22 +++---
1 file changed, 7 insertions(+), 15 deletions(-)
diff --git a/drivers/net/hns3/hns3_ethdev.c b/drivers/net/hns3/hns3_ethdev.c
index 76fc401bd62c..b8f7e408d1e0 100644
--- a/drivers/net/hns3/hns3_ethdev.c
+++ b/drivers/net/hns3/hns3_ethdev.c
@@ -5545,27 +5545,15 @@ is_pf_reset_done(struct hns3_hw *hw)
static enum hns3_reset_level
hns3_detect_reset_event(struct hns3_hw *hw)
{
- struct hns3_adapter *hns = HNS3_DEV_HW_TO_ADAPTER(hw);
enum hns3_reset_level new_req = HNS3_NONE_RESET;
- enum hns3_reset_level last_req;
uint32_t vector0_intr_state;
- last_req = hns3_get_reset_level(hns, &hw->reset.pending);
vector0_intr_state = hns3_read_dev(hw, HNS3_VECTOR0_OTHER_INT_STS_REG);
if (BIT(HNS3_VECTOR0_IMPRESET_INT_B) & vector0_intr_state)
new_req = HNS3_IMP_RESET;
else if (BIT(HNS3_VECTOR0_GLOBALRESET_INT_B) & vector0_intr_state)
new_req = HNS3_GLOBAL_RESET;
- if (new_req == HNS3_NONE_RESET)
- return HNS3_NONE_RESET;
-
- if (last_req == HNS3_NONE_RESET || last_req < new_req) {
- __atomic_store_n(&hw->reset.disable_cmd, 1, __ATOMIC_RELAXED);
- hns3_schedule_delayed_reset(hns);
- hns3_warn(hw, "High level reset detected, delay do reset");
- }
-
return new_req;
}
@@ -5584,10 +5572,14 @@ hns3_is_reset_pending(struct hns3_adapter *hns)
return false;
new_req = hns3_detect_reset_event(hw);
+ if (new_req == HNS3_NONE_RESET)
+ return false;
+
last_req = hns3_get_reset_level(hns, &hw->reset.pending);
- if (last_req != HNS3_NONE_RESET && new_req != HNS3_NONE_RESET &&
- new_req < last_req) {
- hns3_warn(hw, "High level reset %d is pending", last_req);
+ if (last_req == HNS3_NONE_RESET || last_req < new_req) {
+ __atomic_store_n(&hw->reset.disable_cmd, 1, __ATOMIC_RELAXED);
+ hns3_schedule_delayed_reset(hns);
+ hns3_warn(hw, "High level reset detected, delay do reset");
return true;
}
last_req = hns3_get_reset_level(hns, &hw->reset.request);
--
2.30.0
[PATCH] lib/dmadev: get DMA device using device ID
DMA library has a function to get DMA device based on device name but
there is no function to get DMA device using device id.
Added a function that lookup for the dma device using device id and
returns the pointer to the same.
Signed-off-by: Amit Prakash Shukla
---
lib/dmadev/rte_dmadev.c | 9 +
lib/dmadev/rte_dmadev_pmd.h | 14 ++
lib/dmadev/version.map | 1 +
3 files changed, 24 insertions(+)
diff --git a/lib/dmadev/rte_dmadev.c b/lib/dmadev/rte_dmadev.c
index 4e5e420c82..83f49e77f2 100644
--- a/lib/dmadev/rte_dmadev.c
+++ b/lib/dmadev/rte_dmadev.c
@@ -397,6 +397,15 @@ rte_dma_is_valid(int16_t dev_id)
rte_dma_devices[dev_id].state != RTE_DMA_DEV_UNUSED;
}
+struct rte_dma_dev *
+rte_dma_pmd_get_dev_by_id(const int dev_id)
+{
+ if (!rte_dma_is_valid(dev_id))
+ return NULL;
+
+ return &rte_dma_devices[dev_id];
+}
+
uint16_t
rte_dma_count_avail(void)
{
diff --git a/lib/dmadev/rte_dmadev_pmd.h b/lib/dmadev/rte_dmadev_pmd.h
index c61cedfb23..f68c3ac6aa 100644
--- a/lib/dmadev/rte_dmadev_pmd.h
+++ b/lib/dmadev/rte_dmadev_pmd.h
@@ -167,6 +167,20 @@ struct rte_dma_dev *rte_dma_pmd_allocate(const char *name,
int numa_node,
__rte_internal
int rte_dma_pmd_release(const char *name);
+/**
+ * @internal
+ * Get the rte_dma_dev structure device pointer for the device id.
+ *
+ * @param dev_id
+ * Device ID value to select the device structure.
+ *
+ * @return
+ * - rte_dma_dev structure pointer for the given device ID on success, NULL
+ * otherwise.
+ */
+__rte_internal
+struct rte_dma_dev *rte_dma_pmd_get_dev_by_id(const int dev_id);
+
#ifdef __cplusplus
}
#endif
diff --git a/lib/dmadev/version.map b/lib/dmadev/version.map
index 2a3736514c..2cd57c43d3 100644
--- a/lib/dmadev/version.map
+++ b/lib/dmadev/version.map
@@ -26,6 +26,7 @@ INTERNAL {
rte_dma_fp_objs;
rte_dma_pmd_allocate;
rte_dma_pmd_release;
+ rte_dma_pmd_get_dev_by_id;
local: *;
};
--
2.25.1
RE: [EXT] [PATCH v4 1/3] node: support to add next node to ethdev Rx node
> -Original Message-
> From: Rakesh Kudurumalla
> Sent: Tuesday, December 5, 2023 2:57 PM
> To: Nithin Kumar Dabilpuram ; Pavan
> Nikhilesh Bhagavatula
> Cc: dev@dpdk.org; Jerin Jacob Kollanukkaran ;
> Rakesh Kudurumalla
> Subject: [EXT] [PATCH v4 1/3] node: support to add next node to ethdev Rx
> node
>
> External Email
>
> --
> By default all packets received on ethdev_rx node is forwarded to pkt_cls
> node.This patch provides library support to add a new node as next node to
> ethdev_rx node and forward packet to new node from rx node.
>
> Signed-off-by: Rakesh Kudurumalla
> ---
> v4: Resolve compilation issues
>
> lib/node/ethdev_ctrl.c | 45
> +
> lib/node/rte_node_eth_api.h | 19
> lib/node/version.map| 1 +
> 3 files changed, 65 insertions(+)
>
> diff --git a/lib/node/ethdev_ctrl.c b/lib/node/ethdev_ctrl.c index
> d564b80e37..0faf4c19f2 100644
> --- a/lib/node/ethdev_ctrl.c
> +++ b/lib/node/ethdev_ctrl.c
> @@ -3,6 +3,7 @@
> */
>
> #include
> +#include
>
> #include
> #include
> @@ -129,3 +130,47 @@ rte_node_eth_config(struct
> rte_node_ethdev_config *conf, uint16_t nb_confs,
> ctrl.nb_graphs = nb_graphs;
> return 0;
> }
> +
> +int
> +rte_node_ethdev_rx_next_update(rte_node_t id, const char *edge_name) {
> + struct ethdev_rx_node_main *data;
> + ethdev_rx_node_elem_t *elem;
> + char **next_nodes;
> + int rc = -EINVAL;
> + uint32_t count;
> + uint16_t i = 0;
> +
> + if (edge_name == NULL)
> + return rc;
> +
Better to use goto scheme.
> + count = rte_node_edge_get(id, NULL);
> +
> + if (count == RTE_NODE_ID_INVALID)
> + return rc;
> +
> + next_nodes = malloc(count);
> + if (next_nodes == NULL)
> + return -ENOMEM;
> +
> + count = rte_node_edge_get(id, next_nodes);
> +
Here it looks like that loop can misbehave for invalid edge_name. I might be
wrong.
Consider the case.
1. edge_name = "test_name"
2. number of elements in next_nodes is 2 and no element is matching with
edge_name.
In this case, loop might misbehave as there will not be any NULL entry.
Did you try a negative test case on this API ?
> + while (next_nodes[i] != NULL) {
> + if (strcmp(edge_name, next_nodes[i]) == 0) {
> + data = ethdev_rx_get_node_data_get();
> + elem = data->head;
> + while (elem->next != data->head) {
> + if (elem->nid == id) {
> + elem->ctx.cls_next = i;
> + rc = 0;
> + goto exit;
> + }
> + elem = elem->next;
> + }
> + }
> + i++;
> + }
> +exit:
> + free(next_nodes);
> + return rc;
> +}
> diff --git a/lib/node/rte_node_eth_api.h b/lib/node/rte_node_eth_api.h index
> eaae50772d..b082a5bec1 100644
> --- a/lib/node/rte_node_eth_api.h
> +++ b/lib/node/rte_node_eth_api.h
> @@ -23,6 +23,7 @@ extern "C" {
> #include
> #include
> #include
> +#include
>
> /**
> * Port config for ethdev_rx and ethdev_tx node.
> @@ -57,6 +58,24 @@ struct rte_node_ethdev_config {
> */
> int rte_node_eth_config(struct rte_node_ethdev_config *cfg,
> uint16_t cnt, uint16_t nb_graphs);
> +
> +/**
> + * Update ethdev rx next node.
> + *
> + * @param id
> + * Node id whose edge is to be updated.
> + * @param edge_name
> + * Name of the next node.
> + *
> + * @return
> + * RTE_EDGE_ID_INVALID if id is invalid
I think, This is not needed now. Below error code handles for invalid ID too.
> + * ENINVAL: Either of input parameters are invalid
> + * ENOMEM: If memory allocation failed
> + * 0 on successful initialization.
> + */
> +__rte_experimental
> +int rte_node_ethdev_rx_next_update(rte_node_t id, const char
> +*edge_name);
> +
> #ifdef __cplusplus
> }
> #endif
> diff --git a/lib/node/version.map b/lib/node/version.map index
> 99ffcdd414..07abc3a79f 100644
> --- a/lib/node/version.map
> +++ b/lib/node/version.map
> @@ -16,6 +16,7 @@ EXPERIMENTAL {
> rte_node_ip6_route_add;
>
> # added in 23.11
> + rte_node_ethdev_rx_next_update;
> rte_node_ip4_reassembly_configure;
> rte_node_udp4_dst_port_add;
> rte_node_udp4_usr_node_add;
> --
> 2.25.1
RE: [PATCH v5 09/19] net/mlx[45]: remove word sanity
> -Original Message- > From: Stephen Hemminger > Sent: Wednesday, November 29, 2023 18:26 > To: dev@dpdk.org > Cc: Stephen Hemminger ; Dariusz Sosnowski > ; Slava Ovsiienko ; Ori > Kam ; Suanming Mou ; Matan > Azrad > Subject: [PATCH v5 09/19] net/mlx[45]: remove word sanity > > The term "sanity" is on non-inclusive naming list. > Remove it from the Nvidia drivers. > > Signed-off-by: Stephen Hemminger > --- > drivers/common/mlx5/linux/mlx5_common_os.c | 2 +- > drivers/net/mlx4/mlx4.c| 2 +- > drivers/net/mlx4/mlx4_flow.c | 6 +++--- > drivers/net/mlx5/mlx5_flow_dv.c| 2 +- > drivers/net/mlx5/mlx5_flow_hw.c| 6 ++ > drivers/net/mlx5/mlx5_rxq.c| 2 +- > drivers/net/mlx5/mlx5_rxtx_vec.h | 2 +- > 7 files changed, 10 insertions(+), 12 deletions(-) Acked-by: Dariusz Sosnowski Thank you.
TODAY - webinar about 5G with DPDK
Exploring DPDK’s Role in 5G Architecture Experts from Intel, Ericsson and NVIDIA will talk about using DPDK for 5G. Join us at 2pm UTC for 1 hour (Q&A included), this Thursday 7 December. Register with this link: https://zoom.us/webinar/register/WN_LSyHZj9DRf2Q0XFoev60Sg#/registration
NVIDIA roadmap for 24.03
Please find below NVIDIA roadmap for 24.03 release: A. rte_flow new APIs = 1. Add NAT64 action. Adding a new action for header re-write of NAT64. NAT64 (Network Address Translation 64) is a networking technology that facilitates communication between IPv6 and IPv4 networks. 1. Support more modify fields of IP and TCP headers. Supporting IPv4 IHL & total length, IPv6 payload length and TCP data offset in the rte_flow modify field actions. For example, by using the ADD operation, the length of the IP and TCP headers can be calculated. When it is equal to the total length of an IP packet, it means that there is no encrypted header and the IPSec matching can be skipped. 1. Template API performance tuning. Improve code to enable faster rule insertion/deletion. Improve indexed pool management for faster allocations. 1. Align the source port in the tunnel header calculation between SW and HW When doing ECMP (Equal-Cost Multipath Routing) in the underlay network with a tunnel encapsulation, for example, GENEVE, the outer UDP source port can be used as an identifier. The source port calculation in the slow path SW and data path HW offloading should be aligned to solve the path selection mismatch. 1. Support compares two fields with rte_flow_item, including EQ/NE/LT/LE/GT/GE. Adding new item RTE_FLOW_ITEM_TYPE_COMPARE to compare 2 fields with reusing the MODIFY_FIELD Provide a more flexible comparison. 1. Match random inside a range in template API. Enable matching random inside a range. An enhancement of the random item matching. Used for sampling the packets. 1. Support Geneve encap/decap and modify actions. Added support for Geneve raw_encap/decap as well as Geneve options modify with template API in addition to previously supported match on Geneve options. 1. Support table resize in template API. The table size (maximal number of rules supported) is fixed when creating a template table and the ASO actions number is also fixed when configuring the port. Reduce the number of tables to be created. Using rehash mechanism to handle the table resize for this stage. B. Net/mlx5 PMD updates = 1. Return detailed error if the match pattern cannot be created due to matching fields limit. Different NIC generations has different matching size limits (number of matching fields). Once application receive such error, it should perform relevant enhancements to build an alternate matching pipe that fits to the NIC’s limits. 1. Support more than 32 ports for CT No ports number limitation when creating an indirect connection tracking action in the template API. 1. Support VXLAN-GPE matching Support VXLAN-GPE matching in template API, including the flag and reserved bits. C. Test apps updates = 1. Support the changes in rte_flow listed above in testpmd. 2. Support hairpin with 4 physical ports in testpmd. For Connectx -7 Nic's that support 4 physical ports, detect port number, and update hairpin logic to enable use of all ports. Current testpmd enable sequential pair for hairpin, this feature is meant to enable any pair. Regards, Maayan Kashani
Re: [PATCH v4 1/3] node: support to add next node to ethdev Rx node
On Tue, Dec 5, 2023 at 10:27 AM Rakesh Kudurumalla
wrote:
> diff --git a/lib/node/version.map b/lib/node/version.map
> index 99ffcdd414..07abc3a79f 100644
> --- a/lib/node/version.map
> +++ b/lib/node/version.map
> @@ -16,6 +16,7 @@ EXPERIMENTAL {
> rte_node_ip6_route_add;
>
> # added in 23.11
> + rte_node_ethdev_rx_next_update;
> rte_node_ip4_reassembly_configure;
> rte_node_udp4_dst_port_add;
> rte_node_udp4_usr_node_add;
This is added in v24.03.
--
David Marchand
[PATCH v2] app/test-pmd: fix tcp/udp cksum with padding data
IEEE 802 packets may have a minimum size limit. The data fields
should be padded when necessary. In some cases, the padding data
is not zero. Testpmd does not trim these IP packets to the true
length of the frame, so errors will occur when calculating TCP
or UDP checksum.
This commit fixes this issue by triming IP packets to the true
length of the frame in testpmd.
Fixes: 03d17e4d0179 ("app/testpmd: do not change IP addrs in checksum engine")
Cc: sta...@dpdk.org
Signed-off-by: Kaiwen Deng
---
app/test-pmd/csumonly.c | 32
1 file changed, 32 insertions(+)
diff --git a/app/test-pmd/csumonly.c b/app/test-pmd/csumonly.c
index 21210aff43..f7c8af96d7 100644
--- a/app/test-pmd/csumonly.c
+++ b/app/test-pmd/csumonly.c
@@ -860,12 +860,14 @@ pkt_burst_checksum_forward(struct fwd_stream *fs)
uint16_t nb_rx;
uint16_t nb_prep;
uint16_t i;
+ uint16_t pad_len;
uint64_t rx_ol_flags, tx_ol_flags;
uint64_t tx_offloads;
uint32_t rx_bad_ip_csum;
uint32_t rx_bad_l4_csum;
uint32_t rx_bad_outer_l4_csum;
uint32_t rx_bad_outer_ip_csum;
+ uint32_t l3_off;
struct testpmd_offload_info info;
/* receive a burst of packet */
@@ -987,6 +989,36 @@ pkt_burst_checksum_forward(struct fwd_stream *fs)
l3_hdr = (char *)l3_hdr + info.outer_l3_len +
info.l2_len;
}
+ if (info.is_tunnel) {
+ l3_off = info.outer_l2_len +
+ info.outer_l3_len +
+ info.l2_len;
+ } else {
+ l3_off = info.l2_len;
+ }
+ switch (info.ethertype) {
+ case _htons(RTE_ETHER_TYPE_IPV4):
+ pad_len = rte_pktmbuf_data_len(m) -
+ (l3_off +
+ rte_be_to_cpu_16(
+ ((struct rte_ipv4_hdr
*)l3_hdr)->total_length));
+ break;
+ case _htons(RTE_ETHER_TYPE_IPV6):
+ pad_len = rte_pktmbuf_data_len(m) -
+ (l3_off + info.l3_len +
+ rte_be_to_cpu_16(
+ ((struct rte_ipv6_hdr
*)l3_hdr)->payload_len));
+ break;
+ default:
+ pad_len = 0;
+ break;
+ }
+
+ if (pad_len) {
+ rte_pktmbuf_data_len(m) = rte_pktmbuf_data_len(m) -
pad_len;
+ rte_pktmbuf_pkt_len(m) = rte_pktmbuf_data_len(m);
+ }
+
/* step 2: depending on user command line configuration,
* recompute checksum either in software or flag the
* mbuf to offload the calculation to the NIC. If TSO
--
2.34.1
Re: Next year of LTSes
On Thu, 9 Nov 2023 at 10:30, Luca Boccassi wrote: > > Hi, > > We should start thinking about the next year of LTSes and how to organize. > > 20.11 will be EOL after the next upcoming release being prepared now, > as it's now 3 years old. So long, and thanks for all the fixes. > > Currently Kevin is taking care of 21.11 and Xueming of 22.11. > > Xueming, what do you think about, from next year (ie, after the > current set is done) me taking over 22.11, and you switching to 23.11? > The reason is that Debian uses "even" releases due to release cadence, > so just like I had vested interest for 20.11 for Debian 11, Debian 12 > uses 22.11. Bumping the thread now that 23.11 is out
Re: [PATCH] version: 24.03-rc0
Hello maintainers, On Thu, Nov 30, 2023 at 10:23 AM David Marchand wrote: > For subtree maintainers. > > It should change nothing to you guys. > Please rebase your trees on v24.03-rc0 and push it to dpdk.org. > And then double check that the mirror happened (looking at git push > output, and checking https://github.com/DPDK/dpdk/branches). > If you hit some issue, ping me. We are still missing the next-net-brcm mirror, but thanks to all for rebasing your branches. A special note: with this mirroring in place, you get automatic testing in GHA when you push to your branches. You won't get a notification when the tests are done though. You will need to check manually by looking at https://github.com/DPDK/dpdk/actions. -- David Marchand
RE: [PATCH v8 0/3] test/dma: add vchan reconfig and SG tests
Hi, Could this patch series be reviewed ? As of now, patches are rebased without any conflicts. Thanks, Gowrishankar > -Original Message- > From: Gowrishankar Muthukrishnan > Sent: Thursday, November 16, 2023 11:15 PM > To: dev@dpdk.org > Cc: Anoob Joseph ; Chengwen Feng > ; Vamsi Krishna Attunuru > ; Amit Prakash Shukla > ; Vidya Sagar Velumuri > ; Kevin Laatz ; Bruce > Richardson ; Gowrishankar Muthukrishnan > > Subject: [PATCH v8 0/3] test/dma: add vchan reconfig and SG tests > > This patch series reworks DMA tests to follow unit test framework followed > by new vchan reconfig and SG tests. > > v8: > - Addressed compilation issue in Windows platform. > > Gowrishankar Muthukrishnan (3): > test/dma: use unit test framework > test/dma: test multiple vchan > test/dma: add SG copy tests > > app/test/test_dmadev.c | 374 +--- > - > app/test/test_dmadev_api.c | 316 +++ > 2 files changed, 534 insertions(+), 156 deletions(-) > > -- > 2.25.1
RE: [PATCH v8 0/4] PCI Dev and SG copy support
Hi, Could this patch series be reviewed ? As of now, patches are rebased without any conflicts. Thanks, Gowrishankar > -Original Message- > From: Gowrishankar Muthukrishnan > Sent: Wednesday, November 22, 2023 4:36 PM > To: dev@dpdk.org > Cc: Anoob Joseph ; Cheng Jiang > ; Kevin Laatz ; Bruce > Richardson ; Pavan Nikhilesh Bhagavatula > ; Amit Prakash Shukla > ; Chengwen Feng > ; Gowrishankar Muthukrishnan > > Subject: [PATCH v8 0/4] PCI Dev and SG copy support > > Improve dma-perf application to support PCI dev and SG copy, along with > additional supports below: > - validate copied memory > - skip tests if not opted. > > v8: > - updated skip test patch. > > Amit Prakash Shukla (2): > app/dma-perf: add skip support > app/dma-perf: add PCI device support > > Gowrishankar Muthukrishnan (2): > app/dma-perf: validate copied memory > app/dma-perf: add SG copy support > > app/test-dma-perf/benchmark.c | 383 > +++--- > app/test-dma-perf/config.ini | 56 + > app/test-dma-perf/main.c | 161 +++--- > app/test-dma-perf/main.h | 12 +- > 4 files changed, 551 insertions(+), 61 deletions(-) > > -- > 2.25.1
Re: [PATCH] app/graph: generate cmdline boilerplate
On Thu, Dec 07, 2023 at 12:42:05PM +0530, sk...@marvell.com wrote:
> From: Sunil Kumar Kori
>
> Use the dpdk-cmdline-gen script to autogenerate all the boilerplate
> structs and defines for the commandline part of the application.
>
> Depends-on: series-30459 ("enhancements for dpdk-cmdline-gen script")
>
> Signed-off-by: Sunil Kumar Kori
> ---
> app/graph/cli.c | 27 ---
> app/graph/commands.list | 32
> app/graph/ethdev.c| 288 +-
> app/graph/ethdev.h| 9 -
> app/graph/ethdev_priv.h | 63 ---
> app/graph/ethdev_rx.c | 67 +--
> app/graph/ethdev_rx.h | 2 -
> app/graph/ethdev_rx_priv.h| 16 --
> app/graph/examples/l3fwd.cli | 12 +-
> app/graph/examples/l3fwd_pcap.cli | 12 +-
> app/graph/graph.c | 131 ++
> app/graph/graph.h | 5 -
> app/graph/graph_priv.h| 35
> app/graph/ip4_route.c | 86 ++---
> app/graph/ip6_route.c | 88 ++---
> app/graph/mempool.c | 68 +--
> app/graph/mempool.h | 3 -
> app/graph/mempool_priv.h | 18 --
> app/graph/meson.build | 8 +
> app/graph/module_api.h| 1 +
> app/graph/neigh.c | 103 ++-
> app/graph/neigh.h | 4 -
> app/graph/neigh_priv.h| 21 ---
> app/graph/route.h | 5 -
> app/graph/route_priv.h| 34
> 25 files changed, 164 insertions(+), 974 deletions(-)
> create mode 100644 app/graph/commands.list
>
I really like the diff-stats of this patch! More than 800 lines fewer in our
codebase.
Acked-by: Bruce Richardson
Re: [RFT] net/nfb: use dynamic logtype
Thanks for patch! There are some issues.
On Wed, 2023-12-06 at 09:51 -0800, Stephen Hemminger wrote:
>
> diff --git a/drivers/net/nfb/nfb_log.h b/drivers/net/nfb/nfb_log.h
> new file mode 100644
> index ..fac66a38d4b3
> --- /dev/null
> +++ b/drivers/net/nfb/nfb_log.h
> @@ -0,0 +1,13 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + */
> +
> +#ifndef _NFB_STATS_H_
> +#define _NFB_STATS_H_
use the _NFB_LOG_H_ guards
> +
> +extern int nfb_logtype;
> +
> +#define NFB_LOG(level, fmt, args...) \
> + rte_log(RTE_LOG_ ## level, nfb_logtype, "%s(): " fmt "\n", \
> + __func__, ## args)
> +
> +#endif /* _NFB_STATS_H_ */
use the _NFB_LOG_H_ guard
> diff --git a/drivers/net/nfb/nfb_rx.c b/drivers/net/nfb/nfb_rx.c
> index 8a9b232305f2..e39592d04737 100644
> --- a/drivers/net/nfb/nfb_rx.c
> +++ b/drivers/net/nfb/nfb_rx.c
>
> @@ -19,7 +20,7 @@ nfb_eth_rx_queue_start(struct rte_eth_dev *dev, uint16_t
> rxq_id)
> int ret;
>
> if (rxq->queue == NULL) {
> - RTE_LOG(ERR, PMD, "RX NDP queue is NULL!\n");
> + NFP_LOG(ERR, "RX NDP queue is NULL");
typo, should be NFB_LOG instead of NFP_LOG
> return -EINVAL;
> }
>
>
Also, the nfb_rx.h and nfb_tx.h files use the macro NFB_LOG inside,
please add '#include "nfb_log.h"' into them (then the include in
nfb_rx.c will be duplicate). Otherwise, all .c sources, which include
main nfb.h, don't compile.
With these changes, the driver works.
Thank you!
Martin
RE: [PATCH v4 2/3] app/graph: add ethdev forward command
> -Original Message-
> From: Rakesh Kudurumalla
> Sent: Tuesday, December 5, 2023 2:57 PM
> To: Sunil Kumar Kori ; Rakesh Kudurumalla
>
> Cc: dev@dpdk.org; Jerin Jacob Kollanukkaran ; Nithin
> Kumar Dabilpuram
> Subject: [PATCH v4 2/3] app/graph: add ethdev forward command
>
> Adds a txport to forward packet for every rxport
>
> Mapping will be used to forward packets to txport received on rxport
>
> Following commands are exposed:
> - ethdev forward "
>
> Signed-off-by: Rakesh Kudurumalla
> ---
> app/graph/cli.c | 1 +
> app/graph/ethdev.c | 62
> +
> app/graph/ethdev.h | 1 +
> app/graph/ethdev_priv.h | 8 ++
> 4 files changed, 72 insertions(+)
>
> diff --git a/app/graph/cli.c b/app/graph/cli.c index
> 30b12312d6..76f5b8e670 100644
> --- a/app/graph/cli.c
> +++ b/app/graph/cli.c
> @@ -32,6 +32,7 @@ cmdline_parse_ctx_t modules_ctx[] = {
> (cmdline_parse_inst_t *)ðdev_prom_mode_cmd_ctx,
> (cmdline_parse_inst_t *)ðdev_ip4_cmd_ctx,
> (cmdline_parse_inst_t *)ðdev_ip6_cmd_ctx,
> + (cmdline_parse_inst_t *)ðdev_forward_cmd_ctx,
> (cmdline_parse_inst_t *)ðdev_cmd_ctx,
> (cmdline_parse_inst_t *)ðdev_help_cmd_ctx,
> (cmdline_parse_inst_t *)ðdev_rx_cmd_ctx, diff --git
> a/app/graph/ethdev.c b/app/graph/ethdev.c index
> c9b09168c1..bceee659a2 100644
> --- a/app/graph/ethdev.c
> +++ b/app/graph/ethdev.c
> @@ -38,6 +38,10 @@ cmd_ethdev_ip4_addr_help[] = "ethdev
> ip4 addr add netmask cmd_ethdev_ip6_addr_help[] = "ethdev ip6 addr add
> netmask ";
>
> +static const char
> +cmd_ethdev_forward_help[] = "ethdev forward
> +";
> +
> +
Remove extra line.
> static struct rte_eth_conf port_conf_default = {
> .link_speeds = 0,
> .rxmode = {
> @@ -888,3 +892,61 @@ cmdline_parse_inst_t ethdev_help_cmd_ctx = {
> NULL,
> },
> };
> +
> +static int
> +ethdev_forward_config(char *tx_name, char *rx_name) {
Replace tx_name -> tx_dev and rx_name -> rx_dev
> + struct ethdev *port;
> + uint16_t portid_rx = 0;
> + uint16_t portid_tx = 0;
> + int rc = -EINVAL;
> +
Use reverse x-max tree method to declare the variables.
> + rc = rte_eth_dev_get_port_by_name(tx_name, &portid_tx);
> + if (rc < 0)
> + return rc;
> +
> + rc = rte_eth_dev_get_port_by_name(rx_name, &portid_rx);
> + if (rc < 0)
> + return rc;
> +
> + port = ethdev_port_by_id(portid_rx);
> + if (port) {
> + port->config.tx_port_id = portid_tx;
> + rc = 0;
> + }
> +
If port is NULL then rc will be returned with value >= 0. Means in case of
failure, success is returned.
> + return rc;
> +}
> +
> +static void
> +cli_ethdev_forward(void *parsed_result, __rte_unused struct cmdline
> +*cl, void *data __rte_unused) {
> + struct ethdev_fwd_cmd_tokens *res = parsed_result;
> + int rc = -EINVAL;
> +
> + rc = ethdev_forward_config(res->tx_dev, res->rx_dev);
> + if (rc < 0)
> + printf(MSG_CMD_FAIL, res->cmd);
> +}
> +
> +cmdline_parse_token_string_t ethdev_l2_cmd =
Better to rename as ethdev_forward_xyz to align with other's naming convention.
> + TOKEN_STRING_INITIALIZER(struct ethdev_fwd_cmd_tokens, cmd,
> "ethdev");
> +cmdline_parse_token_string_t ethdev_fwd_cmd =
> + TOKEN_STRING_INITIALIZER(struct ethdev_fwd_cmd_tokens, fwd,
> +"forward"); cmdline_parse_token_string_t ethdev_tx_device =
> + TOKEN_STRING_INITIALIZER(struct ethdev_fwd_cmd_tokens,
> tx_dev, NULL);
> +cmdline_parse_token_string_t ethdev_rx_device =
> + TOKEN_STRING_INITIALIZER(struct ethdev_fwd_cmd_tokens,
> rx_dev, NULL);
> +
> +cmdline_parse_inst_t ethdev_forward_cmd_ctx = {
> + .f = cli_ethdev_forward,
> + .data = NULL,
> + .help_str = cmd_ethdev_forward_help,
> + .tokens = {
> +(void *)ðdev_l2_cmd,
> +(void *)ðdev_fwd_cmd,
> +(void *)ðdev_tx_device,
> +(void *)ðdev_rx_device,
> +NULL,
> + },
> +};
> diff --git a/app/graph/ethdev.h b/app/graph/ethdev.h index
> 94d3247a2c..836052046b 100644
> --- a/app/graph/ethdev.h
> +++ b/app/graph/ethdev.h
> @@ -15,6 +15,7 @@ extern cmdline_parse_inst_t ethdev_mtu_cmd_ctx;
> extern cmdline_parse_inst_t ethdev_prom_mode_cmd_ctx; extern
> cmdline_parse_inst_t ethdev_ip4_cmd_ctx; extern cmdline_parse_inst_t
> ethdev_ip6_cmd_ctx;
> +extern cmdline_parse_inst_t ethdev_forward_cmd_ctx;
> extern cmdline_parse_inst_t ethdev_cmd_ctx; extern
> cmdline_parse_inst_t ethdev_help_cmd_ctx;
>
> diff --git a/app/graph/ethdev_priv.h b/app/graph/ethdev_priv.h index
> f231f3f3e1..e5e5fbc9ae 100644
> --- a/app/graph/ethdev_priv.h
> +++ b/app/graph/ethdev_priv.h
> @@ -61,6 +61,13 @@ struct ethdev_ip6_cmd_tokens {
> cmdline_fixed_string_t mask;
> };
>
> +struct ethdev_fwd_cmd_tokens {
> + cmdline_fixed_string_t cmd;
> + cmdline_fixed_string_t fwd;
> + cmdline_fixed_string_t tx_dev;
> + cmdline
RE: [PATCH v4 3/3] app/graph: implement port forward usecase
> -Original Message-
> From: Rakesh Kudurumalla
> Sent: Tuesday, December 5, 2023 2:57 PM
> To: Sunil Kumar Kori ; Rakesh Kudurumalla
>
> Cc: dev@dpdk.org; Jerin Jacob Kollanukkaran ; Nithin
> Kumar Dabilpuram
> Subject: [PATCH v4 3/3] app/graph: implement port forward usecase
>
> Added portforward usecase.In this usecase packets received Rx port is
> forwarded to respective Tx port.
>
> Signed-off-by: Rakesh Kudurumalla
> ---
> app/graph/ethdev.c | 12 ++
> app/graph/ethdev.h | 1 +
> app/graph/examples/l2fwd.cli | 41 +
> app/graph/examples/l2fwd_pcap.cli| 37 +
> app/graph/graph.c| 8 +-
> app/graph/l2fwd.c| 148 +++
> app/graph/l2fwd.h| 11 ++
> app/graph/meson.build| 1 +
> app/graph/module_api.h | 1 +
> doc/guides/tools/graph.rst | 27
> doc/guides/tools/img/graph-usecase-l2fwd.svg | 84 +++
> 11 files changed, 370 insertions(+), 1 deletion(-) create mode 100644
> app/graph/examples/l2fwd.cli create mode 100644
> app/graph/examples/l2fwd_pcap.cli create mode 100644
> app/graph/l2fwd.c create mode 100644 app/graph/l2fwd.h create mode
> 100644 doc/guides/tools/img/graph-usecase-l2fwd.svg
>
> diff --git a/app/graph/ethdev.c b/app/graph/ethdev.c index
> bceee659a2..d048a6 100644
> --- a/app/graph/ethdev.c
> +++ b/app/graph/ethdev.c
> @@ -77,6 +77,18 @@ ethdev_port_by_id(uint16_t port_id)
> return NULL;
> }
>
> +int16_t
> +find_txport_by_rxport(uint16_t portid_rx) {
> + int portid = -EINVAL;
> + struct ethdev *port;
Need one line space here.
> + port = ethdev_port_by_id(portid_rx);
> + if (port)
> + portid = port->config.tx_port_id;
> +
> + return portid;
> +}
> +
> void *
> ethdev_mempool_list_by_portid(uint16_t portid) { diff --git
> a/app/graph/ethdev.h b/app/graph/ethdev.h index
> 836052046b..946e14d801 100644
> --- a/app/graph/ethdev.h
> +++ b/app/graph/ethdev.h
> @@ -33,6 +33,7 @@ extern uint32_t enabled_port_mask;
>
> void ethdev_start(void);
> void ethdev_stop(void);
> +int16_t find_txport_by_rxport(uint16_t portid_rx);
> void *ethdev_mempool_list_by_portid(uint16_t portid); int16_t
> ethdev_portid_by_ip4(uint32_t ip, uint32_t mask); int16_t
> ethdev_portid_by_ip6(uint8_t *ip, uint8_t *mask); diff --git
> a/app/graph/examples/l2fwd.cli b/app/graph/examples/l2fwd.cli new file
> mode 100644 index 00..af24a5836a
> --- /dev/null
> +++ b/app/graph/examples/l2fwd.cli
> @@ -0,0 +1,41 @@
> +; SPDX-License-Identifier: BSD-3-Clause ; Copyright(c) 2023 Marvell.
> +
> +;
> +; Graph configuration for given usecase ; graph l2fwd coremask 0xff bsz
> +32 tmo 10 model default pcap_enable 1 num_pcap_pkts 10 pcap_file
> +/tmp/output.pcap
> +
> +;
> +; Mempools to be attached with ethdev
> +;
> +mempool mempool0 size 8192 buffers 4000 cache 256 numa 0
> +
> +;
> +; DPDK devices and configuration.
> +;
> +; Note: Customize the parameters below to match your setup.
> +;
> +ethdev 0002:01:00.1 rxq 1 txq 8 mempool0 ethdev 0002:01:00.4 rxq 1 txq
> +8 mempool0 ethdev 0002:01:00.6 rxq 1 txq 8 mempool0 ethdev
> 0002:02:00.0
> +rxq 1 txq 8 mempool0
> +
> +;
> +; L2 mac forwarding rules
> +;
Replace "L2 mac forwarding rules" --> "Rx/Tx port mapping"
> +ethdev forward 0002:01:00.4 0002:02:00.0 ethdev forward 0002:01:00.1
> +0002:01:00.6
> +
> +;
> +; Port-Queue-Core mapping for ethdev_rx node ; ethdev_rx map port
> +0002:02:00.0 queue 0 core 1 ethdev_rx map port 0002:01:00.6 queue 0
> +core 2
> +
> +;
> +; Graph start command to create graph.
> +;
> +; Note: No more command should come after this.
> +;
> +graph start
> diff --git a/app/graph/examples/l2fwd_pcap.cli
> b/app/graph/examples/l2fwd_pcap.cli
> new file mode 100644
> index 00..718347f568
> --- /dev/null
> +++ b/app/graph/examples/l2fwd_pcap.cli
> @@ -0,0 +1,37 @@
> +; SPDX-License-Identifier: BSD-3-Clause ; Copyright(c) 2023 Marvell.
> +
> +;
> +; Graph configuration for given usecase ; graph l2fwd coremask 0xff bsz
> +32 tmo 10 model default pcap_enable 1 num_pcap_pkts 10 pcap_file
> +/tmp/output.pcap
> +
> +;
> +; Mempools to be attached with ethdev
> +;
> +mempool mempool0 size 8192 buffers 4000 cache 256 numa 0
> +
> +;
> +; DPDK devices and configuration.
> +;
> +; Note: Customize the parameters below to match your setup.
> +;
> +ethdev net_pcap0 rxq 1 txq 8 mempool0
> +ethdev net_pcap1 rxq 1 txq 8 mempool0
> +
> +;
> +; L2 mac forwarding rules
> +;
> +ethdev forward net_pcap1 net_pcap0
> +
> +;
> +; Port-Queue-Core mapping for ethdev_rx node ; ethdev_rx map port
> +net_pcap0 queue 0 core 1
> +
> +;
> +; Graph start command to create graph.
> +;
> +; Note: No more command should come after this.
> +;
> +graph start
> diff --git a/app/graph/graph.c b/app/graph/graph.c index
> a65723a196..4e
RE: NVIDIA roadmap for 24.03
Small correction, moved Geneve to mlx5 PMD updates. Regards, Maayan Kashani From: Maayan Kashani Sent: Thursday, 7 December 2023 11:29 To: dpdk-dev Subject: NVIDIA roadmap for 24.03 Please find below NVIDIA roadmap for 24.03 release: A. rte_flow new APIs = 1. Add NAT64 action. Adding a new action for header re-write of NAT64. NAT64 (Network Address Translation 64) is a networking technology that facilitates communication between IPv6 and IPv4 networks. 1. Support more modify fields of IP and TCP headers. Supporting IPv4 IHL & total length, IPv6 payload length and TCP data offset in the rte_flow modify field actions. For example, by using the ADD operation, the length of the IP and TCP headers can be calculated. When it is equal to the total length of an IP packet, it means that there is no encrypted header and the IPSec matching can be skipped. 1. Template API performance tuning. Improve code to enable faster rule insertion/deletion. Improve indexed pool management for faster allocations. 1. Align the source port in the tunnel header calculation between SW and HW When doing ECMP (Equal-Cost Multipath Routing) in the underlay network with a tunnel encapsulation, for example, GENEVE, the outer UDP source port can be used as an identifier. The source port calculation in the slow path SW and data path HW offloading should be aligned to solve the path selection mismatch. 1. Support compares two fields with rte_flow_item, including EQ/NE/LT/LE/GT/GE. Adding new item RTE_FLOW_ITEM_TYPE_COMPARE to compare 2 fields with reusing the MODIFY_FIELD Provide a more flexible comparison. 1. Match random inside a range in template API. Enable matching random inside a range. An enhancement of the random item matching. Used for sampling the packets. 1. Support table resize in template API. The table size (maximal number of rules supported) is fixed when creating a template table and the ASO actions number is also fixed when configuring the port. Reduce the number of tables to be created. Using rehash mechanism to handle the table resize for this stage. B. Net/mlx5 PMD updates = 1. Return detailed error if the match pattern cannot be created due to matching fields limit. Different NIC generations has different matching size limits (number of matching fields). Once application receive such error, it should perform relevant enhancements to build an alternate matching pipe that fits to the NIC’s limits. 1. Support more than 32 ports for CT No ports number limitation when creating an indirect connection tracking action in the template API. 1. Support VXLAN-GPE matching Support VXLAN-GPE matching in template API, including the flag and reserved bits. 1. Support Geneve encap/decap and modify actions. Added support for Geneve raw_encap/decap as well as Geneve options modify with template API in addition to previously supported match on Geneve options. C. Test apps updates = 1. Support the changes in rte_flow listed above in testpmd. 2. Support hairpin with 4 physical ports in testpmd. For Connectx -7 Nic's that support 4 physical ports, detect port number, and update hairpin logic to enable use of all ports. Current testpmd enable sequential pair for hairpin, this feature is meant to enable any pair. Regards, Maayan Kashani
[PATCH 1/2] net/virtio-user: improve kick performance with notification area mapping
This patch introduces new virtio-user callback to map the vq
notification area and implements it for the vhost-vDPA backend.
This is simply done by using mmap()/munmap() for
the vhost-vDPA fd.
This patch also adds a parameter for configuring feature bit
VIRTIO_NET_F_NOTIFICATION_DATA. If feature is disabled, also
update corresponding unsupported feature bit. And also adds
code to write to queue notify address in notify callback.
This will help in increasing the kick performance.
Signed-off-by: Srujana Challa
---
doc/guides/nics/virtio.rst| 5 ++
drivers/net/virtio/virtio_user/vhost.h| 1 +
drivers/net/virtio/virtio_user/vhost_vdpa.c | 56 ++
.../net/virtio/virtio_user/virtio_user_dev.c | 52 +++--
.../net/virtio/virtio_user/virtio_user_dev.h | 5 +-
drivers/net/virtio/virtio_user_ethdev.c | 57 ---
6 files changed, 162 insertions(+), 14 deletions(-)
diff --git a/doc/guides/nics/virtio.rst b/doc/guides/nics/virtio.rst
index c22ce56a02..11dd6359e5 100644
--- a/doc/guides/nics/virtio.rst
+++ b/doc/guides/nics/virtio.rst
@@ -349,6 +349,11 @@ Below devargs are supported by the virtio-user vdev:
election.
(Default: 0 (disabled))
+#. ``notification_data``:
+
+It is used to enable virtio device notification data feature.
+(Default: 1 (enabled))
+
Virtio paths Selection and Usage
diff --git a/drivers/net/virtio/virtio_user/vhost.h
b/drivers/net/virtio/virtio_user/vhost.h
index f817cab77a..1bce00c7ac 100644
--- a/drivers/net/virtio/virtio_user/vhost.h
+++ b/drivers/net/virtio/virtio_user/vhost.h
@@ -90,6 +90,7 @@ struct virtio_user_backend_ops {
int (*server_disconnect)(struct virtio_user_dev *dev);
int (*server_reconnect)(struct virtio_user_dev *dev);
int (*get_intr_fd)(struct virtio_user_dev *dev);
+ int (*map_notification_area)(struct virtio_user_dev *dev, bool map);
};
extern struct virtio_user_backend_ops virtio_ops_user;
diff --git a/drivers/net/virtio/virtio_user/vhost_vdpa.c
b/drivers/net/virtio/virtio_user/vhost_vdpa.c
index 2c36b26224..1eb0f9ec48 100644
--- a/drivers/net/virtio/virtio_user/vhost_vdpa.c
+++ b/drivers/net/virtio/virtio_user/vhost_vdpa.c
@@ -5,6 +5,7 @@
#include
#include
#include
+#include
#include
#include
#include
@@ -622,6 +623,60 @@ vhost_vdpa_get_intr_fd(struct virtio_user_dev *dev
__rte_unused)
return -1;
}
+static int
+unmap_notification_area(struct virtio_user_dev *dev, int nr_vrings)
+{
+ int i;
+
+ for (i = 0; i < nr_vrings; i++) {
+ if (dev->notify_area[i])
+ munmap(dev->notify_area[i], getpagesize());
+ }
+ free(dev->notify_area);
+
+ return 0;
+}
+
+static int
+vhost_vdpa_map_notification_area(struct virtio_user_dev *dev, bool map)
+{
+ struct vhost_vdpa_data *data = dev->backend_data;
+ int nr_vrings, i, page_size = getpagesize();
+ uint16_t **notify_area;
+
+ nr_vrings = dev->max_queue_pairs * 2;
+ if (dev->device_features & (1ull << VIRTIO_NET_F_CTRL_VQ))
+ nr_vrings++;
+
+ if (!map)
+ return unmap_notification_area(dev, nr_vrings);
+
+ notify_area = malloc(nr_vrings * sizeof(*notify_area));
+ if (!notify_area) {
+ PMD_DRV_LOG(ERR, "(%s) Failed to allocate notify area array",
dev->path);
+ return -1;
+ }
+ for (i = 0; i < nr_vrings; i++) {
+ notify_area[i] = mmap(NULL, page_size, PROT_WRITE, MAP_SHARED |
MAP_FILE,
+ data->vhostfd, i * page_size);
+ if (notify_area[i] == MAP_FAILED) {
+ PMD_DRV_LOG(ERR, "(%s) Map failed for notify address of
queue %d\n",
+ dev->path, i);
+ goto map_err;
+ }
+ }
+ dev->notify_area = notify_area;
+
+ return 0;
+
+map_err:
+ i--;
+ for (; i >= 0; i--)
+ munmap(notify_area[i], page_size);
+ free(notify_area);
+ return -1;
+}
+
struct virtio_user_backend_ops virtio_ops_vdpa = {
.setup = vhost_vdpa_setup,
.destroy = vhost_vdpa_destroy,
@@ -646,4 +701,5 @@ struct virtio_user_backend_ops virtio_ops_vdpa = {
.dma_unmap = vhost_vdpa_dma_unmap_batch,
.update_link_state = vhost_vdpa_update_link_state,
.get_intr_fd = vhost_vdpa_get_intr_fd,
+ .map_notification_area = vhost_vdpa_map_notification_area,
};
diff --git a/drivers/net/virtio/virtio_user/virtio_user_dev.c
b/drivers/net/virtio/virtio_user/virtio_user_dev.c
index af1f8c8237..578877d089 100644
--- a/drivers/net/virtio/virtio_user/virtio_user_dev.c
+++ b/drivers/net/virtio/virtio_user/virtio_user_dev.c
@@ -18,6 +18,7 @@
#include
#include
#include
+#include
#include "vhost.h"
#include "virtio_user_dev.h"
@@ -413,6 +414,12 @@ virtio_user_dev_init_noti
[PATCH 2/2] net/virtio-user: add VIRTIO_NET_F_RSS to supported features
This patch introduces new function to get rss device config
and adds code to forward the RSS control command to backend
through hw control queue if RSS feature is negotiated.
This patch will help to negotiate VIRTIO_NET_F_RSS feature
if vhost-vdpa backend supports RSS in HW.
Signed-off-by: Srujana Challa
---
.../net/virtio/virtio_user/virtio_user_dev.c | 31 ++-
.../net/virtio/virtio_user/virtio_user_dev.h | 2 ++
drivers/net/virtio/virtio_user_ethdev.c | 3 ++
3 files changed, 35 insertions(+), 1 deletion(-)
diff --git a/drivers/net/virtio/virtio_user/virtio_user_dev.c
b/drivers/net/virtio/virtio_user/virtio_user_dev.c
index 578877d089..f876cb6ed0 100644
--- a/drivers/net/virtio/virtio_user/virtio_user_dev.c
+++ b/drivers/net/virtio/virtio_user/virtio_user_dev.c
@@ -304,6 +304,24 @@ virtio_user_dev_init_max_queue_pairs(struct
virtio_user_dev *dev, uint32_t user_
return 0;
}
+int
+virtio_user_dev_get_rss_config(struct virtio_user_dev *dev, void *dst, size_t
offset, int length)
+{
+ int ret = 0;
+
+ if (!(dev->device_features & (1ULL << VIRTIO_NET_F_RSS)))
+ return -ENOTSUP;
+
+ if (!dev->ops->get_config)
+ return -ENOTSUP;
+
+ ret = dev->ops->get_config(dev, dst, offset, length);
+ if (ret)
+ PMD_DRV_LOG(ERR, "(%s) Failed to get rss config in device",
dev->path);
+
+ return ret;
+}
+
int
virtio_user_dev_set_mac(struct virtio_user_dev *dev)
{
@@ -687,7 +705,8 @@ virtio_user_free_vrings(struct virtio_user_dev *dev)
1ULL << VIRTIO_F_IN_ORDER | \
1ULL << VIRTIO_F_VERSION_1 | \
1ULL << VIRTIO_F_RING_PACKED | \
-1ULL << VIRTIO_F_NOTIFICATION_DATA)
+1ULL << VIRTIO_F_NOTIFICATION_DATA | \
+1ULL << VIRTIO_NET_F_RSS)
int
virtio_user_dev_init(struct virtio_user_dev *dev, char *path, uint16_t queues,
@@ -903,6 +922,11 @@ virtio_user_handle_ctrl_msg_split(struct virtio_user_dev
*dev, struct vring *vri
queues = *(uint16_t *)(uintptr_t)vring->desc[idx_data].addr;
status = virtio_user_handle_mq(dev, queues);
+ } else if (hdr->class == VIRTIO_NET_CTRL_MQ && hdr->cmd ==
VIRTIO_NET_CTRL_MQ_RSS_CONFIG) {
+ struct virtio_net_ctrl_rss *rss;
+
+ rss = (struct virtio_net_ctrl_rss *)vring->desc[idx_data].addr;
+ status = virtio_user_handle_mq(dev, rss->max_tx_vq);
} else if (hdr->class == VIRTIO_NET_CTRL_RX ||
hdr->class == VIRTIO_NET_CTRL_MAC ||
hdr->class == VIRTIO_NET_CTRL_VLAN) {
@@ -964,6 +988,11 @@ virtio_user_handle_ctrl_msg_packed(struct virtio_user_dev
*dev,
queues = *(uint16_t *)(uintptr_t)
vring->desc[idx_data].addr;
status = virtio_user_handle_mq(dev, queues);
+ } else if (hdr->class == VIRTIO_NET_CTRL_MQ && hdr->cmd ==
VIRTIO_NET_CTRL_MQ_RSS_CONFIG) {
+ struct virtio_net_ctrl_rss *rss;
+
+ rss = (struct virtio_net_ctrl_rss *)vring->desc[idx_data].addr;
+ status = virtio_user_handle_mq(dev, rss->max_tx_vq);
} else if (hdr->class == VIRTIO_NET_CTRL_RX ||
hdr->class == VIRTIO_NET_CTRL_MAC ||
hdr->class == VIRTIO_NET_CTRL_VLAN) {
diff --git a/drivers/net/virtio/virtio_user/virtio_user_dev.h
b/drivers/net/virtio/virtio_user/virtio_user_dev.h
index 29ec386da5..39b3eec0f2 100644
--- a/drivers/net/virtio/virtio_user/virtio_user_dev.h
+++ b/drivers/net/virtio/virtio_user/virtio_user_dev.h
@@ -86,6 +86,8 @@ int virtio_user_dev_update_status(struct virtio_user_dev
*dev);
int virtio_user_dev_update_link_state(struct virtio_user_dev *dev);
int virtio_user_dev_set_mac(struct virtio_user_dev *dev);
int virtio_user_dev_get_mac(struct virtio_user_dev *dev);
+int virtio_user_dev_get_rss_config(struct virtio_user_dev *dev, void *dst,
size_t offset,
+ int length);
void virtio_user_dev_delayed_disconnect_handler(void *param);
int virtio_user_dev_server_reconnect(struct virtio_user_dev *dev);
extern const char * const virtio_user_backend_strings[];
diff --git a/drivers/net/virtio/virtio_user_ethdev.c
b/drivers/net/virtio/virtio_user_ethdev.c
index 241465ecdd..6c10e8f6c0 100644
--- a/drivers/net/virtio/virtio_user_ethdev.c
+++ b/drivers/net/virtio/virtio_user_ethdev.c
@@ -52,6 +52,9 @@ virtio_user_read_dev_config(struct virtio_hw *hw, size_t
offset,
if (offset == offsetof(struct virtio_net_config, max_virtqueue_pairs))
*(uint16_t *)dst = dev->max_queue_pairs;
+
+ if (offset >= offsetof(struct virtio_net_config, rss_max_key_size))
+ virtio_user_dev_get_rss_config(dev, dst, offset, length);
}
static void
--
2.25.1
Re: [PATCH v3 0/4] net/hns3: refactor mailbox
On 12/7/2023 1:37 AM, Jie Hai wrote: > This patchset refactors mailbox codes. > > -- > v3: > 1. fix the checkpatch warning on __atomic_xxx. > -- > > Dengdui Huang (4): > net/hns3: refactor VF mailbox message struct > net/hns3: refactor PF mailbox message struct > net/hns3: refactor send mailbox function > net/hns3: refactor handle mailbox function > > Hi Jie, Dengdui, Set fails to build with clang and stdatomic [1], it can be reproduced with command [2]. Can you please check? [1] https://mails.dpdk.org/archives/test-report/2023-December/525771.html [2] CC=clang meson setup --buildtype=debugoptimized --werror -Denable_stdatomic=true build && ninja -C build
Re: [PATCH v3 1/4] net/hns3: refactor VF mailbox message struct
On 12/7/2023 1:37 AM, Jie Hai wrote:
> From: Dengdui Huang
>
> The data region in VF to PF mbx message command is
> used to communicate with PF driver. And this data
> region exists as an array. As a result, some complicated
> feature commands, like setting promisc mode, map/unmap
> ring vector and setting VLAN id, have to use magic number
> to set them. This isn't good for maintenance of driver.
> So this patch refactors these messages by extracting an
> hns3_vf_to_pf_msg structure.
>
> In addition, the PF link change event message is reported
> by the firmware and is reported in hns3_mbx_vf_to_pf_cmd
> format, it also needs to be modified.
>
> Fixes: 463e748964f5 ("net/hns3: support mailbox")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Dengdui Huang
> Signed-off-by: Jie Hai
<...>
> @@ -107,6 +106,48 @@ struct hns3_mbx_resp_status {
> uint8_t additional_info[HNS3_MBX_MAX_RESP_DATA_SIZE];
> };
>
> +struct hns3_ring_chain_param {
> + uint8_t ring_type;
> + uint8_t tqp_index;
> + uint8_t int_gl_index;
> +};
> +
> +#pragma pack(1)
> +struct hns3_mbx_vlan_filter {
> + uint8_t is_kill;
> + uint16_t vlan_id;
> + uint16_t proto;
> +};
> +#pragma pack()
> +
>
Please prefer '__rte_packed' instead of "#pragma pack()", as it is more
consisted way for same purpose.
But I see multiple instances of "#pragma pack()" already exists.
@Tyler,
as for as I understand '__attribute__((__packed__))' (__rte_packed) is
GCC way and "#pragma pack()" is Windows way.
Is __rte_packed causing problem with latest windows compilers?
And should we have an abstract __rte_packed that works for both windows
compiler and gcc?
[PATCH 00/14] Add TLS record test suite
Add TLS record test suite in cryptodev autotest framework. The test suite would run based on capabilities exposed by the cryptodev. The test suite framework is similar to the framework used in case of IPsec tests. To avoid duplication of code, protocol independent code is moved to common files and the functions are renamed accordingly. TLS record test suite has known vector tests as well as combined mode tests. Known vector tests leverages vectors generated with kTLS and gnuTLS utilities. The test suite supports testing both operations (read or decrypt, write or encrypt) with a single vector. Write or encrypt test would get skipped if cryptodev doesn't support disabling of IV generation. Combined mode tests are targetted at testing protocol features with all combinations of cipher-authentication algorithms. Combined mode performs record write operation first and feeds that back to record read operation. Individual test cases may update the input to record write operation based on the test case and the test framework validates the output obtained (not complete text, but protocol specific fields such as TLS header). Once it is validated, the output will be submitted for record read operation which would give back the original data. Currently this framework supports testing of multi-segmented mbuf as input with TLS record. The same would be enhanced to support more cases such as ICV corruption, incorrect padding etc. Enhancements planned for future, - Add TLS 1.3 suite - Add negative tests such as ICV corruption and incorrect padding - Add session expiry tests - Add anti-replay tests with DTLS Sample output with crypto_cn10k: + --- + [67/18944] + --- + + Test Suite : TLS 1.2 Record Protocol Unit Test Suite + --- + + TestCase [ 0] : Write record known vector AES-GCM-128 (vector 1) succeeded + TestCase [ 1] : Write record known vector AES-GCM-128 (vector 2) succeeded + TestCase [ 2] : Write record known vector AES-GCM-256 succeeded + TestCase [ 3] : Write record known vector AES-CBC-128-SHA1 succeeded + TestCase [ 4] : Write record known vector AES-128-CBC-SHA256 succeeded + TestCase [ 5] : Write record known vector AES-256-CBC-SHA1 succeeded + TestCase [ 6] : Write record known vector AES-256-CBC-SHA256 succeeded + TestCase [ 7] : Write record known vector 3DES-CBC-SHA1-HMAC succeeded USER1: Cipher crypto capabilities not supported + TestCase [ 8] : Write record known vector NULL-SHA1-HMAC skipped USER1: Crypto capabilities not supported + TestCase [ 9] : Write record known vector CHACHA20-POLY1305 skipped + TestCase [10] : Read record known vector AES-GCM-128 (vector 1) succeeded + TestCase [11] : Read record known vector AES-GCM-128 (vector 2) succeeded + TestCase [12] : Read record known vector AES-GCM-256 succeeded + TestCase [13] : Read record known vector AES-128-CBC-SHA1 succeeded + TestCase [14] : Read record known vector AES-128-CBC-SHA256 succeeded + TestCase [15] : Read record known vector AES-256-CBC-SHA1 succeeded + TestCase [16] : Read record known vector AES-256-CBC-SHA256 succeeded + TestCase [17] : Read record known vector 3DES-CBC-SHA1-HMAC succeeded USER1: Cipher crypto capabilities not supported + TestCase [18] : Read record known vector NULL-SHA1-HMAC skipped USER1: Crypto capabilities not supported + TestCase [19] : Read record known vector CHACHA20-POLY1305 skipped 3des-cbc [192] sha1-hmac [20B ICV] aes-cbc [128] sha1-hmac [20B ICV] aes-cbc [128] sha2-256-hmac [32B ICV] aes-cbc [256] sha1-hmac [20B ICV] aes-cbc [256] sha2-256-hmac [32B ICV] + TestCase [20] : Combined test alg list succeeded + TestCase [21] : Multi-segmented mode succeeded + --- + + Test Suite Summary : TLS 1.2 Record Protocol Unit Test Suite + --- + + Tests Total : 22 + Tests Skipped : 4 + Tests Executed :22 + Tests Unsupported: 0 + Tests Passed : 18 + Tests Failed : 0 + --- + + --- + + Test Suite : DTLS 1.2 Record Protocol Unit Test Suite + --- + + TestCase [ 0] : Write record known vector AES-GCM-128 succeeded + TestCase [ 1] : Write record known vector AES-GCM-256 succeeded + TestCase [ 2] : Write record known vector AES-128-CBC-SHA1 succeeded + TestCase [ 3] : Write record known vector AES-128-CBC-SHA256 succeeded + TestCase [ 4] : Write record known vector AES-256-CBC-SHA1 succeeded + TestCase [ 5] : Write record known vector AES-256-CBC-SHA256 succeeded + TestCase [ 6] : Write record known vector 3DES-CBC-SHA
[PATCH 01/14] test/crypto: move security caps checks to separate file
Move routines performing security caps verifications to a separate file
to allow the usage of same for other protocol tests such as TLS record.
Signed-off-by: Anoob Joseph
Signed-off-by: Vidya Sagar Velumuri
---
app/test-security-perf/meson.build | 1 +
app/test-security-perf/test_security_perf.c | 17 ++---
app/test/meson.build| 1 +
app/test/test_cryptodev.c | 17 ++---
app/test/test_cryptodev_security_ipsec.c| 78 ---
app/test/test_cryptodev_security_ipsec.h| 12 ---
app/test/test_security_inline_proto.c | 12 +--
app/test/test_security_proto.c | 83 +
app/test/test_security_proto.h | 20 +
9 files changed, 121 insertions(+), 120 deletions(-)
create mode 100644 app/test/test_security_proto.c
create mode 100644 app/test/test_security_proto.h
diff --git a/app/test-security-perf/meson.build
b/app/test-security-perf/meson.build
index 076999022e..547de9c908 100644
--- a/app/test-security-perf/meson.build
+++ b/app/test-security-perf/meson.build
@@ -10,5 +10,6 @@ endif
sources = files(
'test_security_perf.c',
'../test/test_cryptodev_security_ipsec.c',
+'../test/test_security_proto.c',
)
deps += ['security', 'cmdline']
diff --git a/app/test-security-perf/test_security_perf.c
b/app/test-security-perf/test_security_perf.c
index 4dfaca4800..c64f20e76c 100644
--- a/app/test-security-perf/test_security_perf.c
+++ b/app/test-security-perf/test_security_perf.c
@@ -16,6 +16,7 @@
#include
#include
#include
+#include
#define NB_DESC 4096
#define DEF_NB_SESSIONS (16 * 10 * 1024) /* 16 * 10K tunnels */
@@ -258,9 +259,7 @@ sec_conf_init(struct lcore_conf *conf,
conf->aead_xform.aead.iv.offset = IV_OFFSET;
/* Verify crypto capabilities */
- if (test_ipsec_crypto_caps_aead_verify(
- sec_cap,
- &conf->aead_xform) != 0) {
+ if (test_sec_crypto_caps_aead_verify(sec_cap,
&conf->aead_xform) != 0) {
RTE_LOG(ERR, USER1,
"Crypto capabilities not supported\n");
return -1;
@@ -270,9 +269,7 @@ sec_conf_init(struct lcore_conf *conf,
sizeof(conf->auth_xform));
conf->auth_xform.auth.key.data = td[0].auth_key.data;
- if (test_ipsec_crypto_caps_auth_verify(
- sec_cap,
- &conf->auth_xform) != 0) {
+ if (test_sec_crypto_caps_auth_verify(sec_cap,
&conf->auth_xform) != 0) {
RTE_LOG(INFO, USER1,
"Auth crypto capabilities not supported\n");
return -1;
@@ -288,17 +285,13 @@ sec_conf_init(struct lcore_conf *conf,
/* Verify crypto capabilities */
- if (test_ipsec_crypto_caps_cipher_verify(
- sec_cap,
- &conf->cipher_xform) != 0) {
+ if (test_sec_crypto_caps_cipher_verify(sec_cap,
&conf->cipher_xform) != 0) {
RTE_LOG(ERR, USER1,
"Cipher crypto capabilities not supported\n");
return -1;
}
- if (test_ipsec_crypto_caps_auth_verify(
- sec_cap,
- &conf->auth_xform) != 0) {
+ if (test_sec_crypto_caps_auth_verify(sec_cap,
&conf->auth_xform) != 0) {
RTE_LOG(ERR, USER1,
"Auth crypto capabilities not supported\n");
return -1;
diff --git a/app/test/meson.build b/app/test/meson.build
index dcc93f4a43..58e120a6ab 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -169,6 +169,7 @@ source_file_deps = {
'test_security.c': ['net', 'security'],
'test_security_inline_macsec.c': ['ethdev', 'security'],
'test_security_inline_proto.c': ['ethdev', 'security', 'eventdev'] +
test_cryptodev_deps,
+'test_security_proto.c' : ['cryptodev', 'security'],
'test_seqlock.c': [],
'test_service_cores.c': [],
'test_spinlock.c': [],
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index 58561ededf..9644566acc 100644
--- a/app/test/test_cryptodev.c
+++ b/app/test/test_cryptodev.c
@@ -53,6 +53,7 @@
#include "test_cryptodev_security_pdcp_sdap_test_vectors.h"
#include "test_cryptodev_security_pdcp_test_func.h"
#include "test_cryptodev_security_docsis_test_vectors.h"
+#include "test_security_proto.h"
#define SDAP_DISABLED 0
#define SDAP_ENABLED 1
@@ -10229,9 +10230,7 @@ test_ipsec_proto_process(const struct ipsec_test_data
td[],
ut_params->aead_xform.aead.iv.offset = IV_OFFSET;
/* Verify crypto capabilities */
- if (t
[PATCH 02/14] test/crypto: move algorithm list to common
From: Vidya Sagar Velumuri
The algorithm list and associated structs are used for IPsec combined
mode tests. The same list can be employed for other security protocol
testing as well. Move the same to test_security_proto.h for the same.
Signed-off-by: Anoob Joseph
Signed-off-by: Vidya Sagar Velumuri
---
app/test/test_cryptodev_security_ipsec.h | 126 +--
app/test/test_security_proto.h | 124 ++
2 files changed, 126 insertions(+), 124 deletions(-)
diff --git a/app/test/test_cryptodev_security_ipsec.h
b/app/test/test_cryptodev_security_ipsec.h
index dc1b4c4a80..77cb5fa7f9 100644
--- a/app/test/test_cryptodev_security_ipsec.h
+++ b/app/test/test_cryptodev_security_ipsec.h
@@ -8,6 +8,8 @@
#include
#include
+#include "test_security_proto.h"
+
#define IPSEC_TEST_PACKETS_MAX 32
#define IPSEC_TEXT_MAX_LEN 16384u
@@ -115,130 +117,6 @@ struct ipsec_test_flags {
bool rx_inject;
};
-struct crypto_param {
- enum rte_crypto_sym_xform_type type;
- union {
- enum rte_crypto_cipher_algorithm cipher;
- enum rte_crypto_auth_algorithm auth;
- enum rte_crypto_aead_algorithm aead;
- } alg;
- uint16_t key_length;
- uint16_t iv_length;
- uint16_t digest_length;
-};
-
-static const struct crypto_param aead_list[] = {
- {
- .type = RTE_CRYPTO_SYM_XFORM_AEAD,
- .alg.aead = RTE_CRYPTO_AEAD_AES_GCM,
- .key_length = 16,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_AEAD,
- .alg.aead = RTE_CRYPTO_AEAD_AES_GCM,
- .key_length = 24,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_AEAD,
- .alg.aead = RTE_CRYPTO_AEAD_AES_GCM,
- .key_length = 32,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_AEAD,
- .alg.aead = RTE_CRYPTO_AEAD_AES_CCM,
- .key_length = 32
- },
-};
-
-static const struct crypto_param cipher_list[] = {
- {
- .type = RTE_CRYPTO_SYM_XFORM_CIPHER,
- .alg.cipher = RTE_CRYPTO_CIPHER_NULL,
- .key_length = 0,
- .iv_length = 0,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_CIPHER,
- .alg.cipher = RTE_CRYPTO_CIPHER_DES_CBC,
- .key_length = 8,
- .iv_length = 8,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_CIPHER,
- .alg.cipher = RTE_CRYPTO_CIPHER_3DES_CBC,
- .key_length = 24,
- .iv_length = 8,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_CIPHER,
- .alg.cipher = RTE_CRYPTO_CIPHER_AES_CBC,
- .key_length = 16,
- .iv_length = 16,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_CIPHER,
- .alg.cipher = RTE_CRYPTO_CIPHER_AES_CTR,
- .key_length = 16,
- .iv_length = 16,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_CIPHER,
- .alg.cipher = RTE_CRYPTO_CIPHER_AES_CTR,
- .key_length = 24,
- .iv_length = 16,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_CIPHER,
- .alg.cipher = RTE_CRYPTO_CIPHER_AES_CTR,
- .key_length = 32,
- .iv_length = 16,
- },
-};
-
-static const struct crypto_param auth_list[] = {
- {
- .type = RTE_CRYPTO_SYM_XFORM_AUTH,
- .alg.auth = RTE_CRYPTO_AUTH_NULL,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_AUTH,
- .alg.auth = RTE_CRYPTO_AUTH_MD5_HMAC,
- .key_length = 16,
- .digest_length = 12,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_AUTH,
- .alg.auth = RTE_CRYPTO_AUTH_SHA256_HMAC,
- .key_length = 32,
- .digest_length = 16,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_AUTH,
- .alg.auth = RTE_CRYPTO_AUTH_SHA384_HMAC,
- .key_length = 48,
- .digest_length = 24,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_AUTH,
- .alg.auth = RTE_CRYPTO_AUTH_SHA512_HMAC,
- .key_length = 64,
- .digest_length = 32,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_AUTH,
- .alg.auth = RTE_CRYPTO_AUTH_AES_XCBC_MAC,
- .key_length = 16,
- .digest_length = 12,
- },
- {
- .type = RTE_CRYPTO_SYM_XFORM_AUTH,
- .alg.auth = RTE_CRYPTO_AUTH_AES_GMAC,
- .key_length = 16,
- .digest_length = 16,
- .iv_length = 12,
- },
-};
-
struct crypto_param_comb {
const struct crypto_param *param1;
const struct crypto_param *param2;
diff --git a/app/
[PATCH 03/14] test/crypto: move algorithm display routines to common
The functions that are used for printing algorithms in case of IPsec can
be moved to the common header, test_security_proto.h. The same could be
used for other security protocol testing.
Signed-off-by: Anoob Joseph
Signed-off-by: Vidya Sagar Velumuri
---
app/test-security-perf/test_security_perf.c | 3 +--
app/test/test_cryptodev.c | 6 ++---
app/test/test_cryptodev_security_ipsec.c| 26 -
app/test/test_cryptodev_security_ipsec.h| 3 ---
app/test/test_security_inline_proto.c | 9 +++
app/test/test_security_proto.c | 25
app/test/test_security_proto.h | 2 ++
7 files changed, 33 insertions(+), 41 deletions(-)
diff --git a/app/test-security-perf/test_security_perf.c
b/app/test-security-perf/test_security_perf.c
index c64f20e76c..84726de7b0 100644
--- a/app/test-security-perf/test_security_perf.c
+++ b/app/test-security-perf/test_security_perf.c
@@ -532,8 +532,7 @@ main(int argc, char **argv)
for (ctx.td_idx = 0; ctx.td_idx < RTE_DIM(alg_list); ctx.td_idx++) {
printf("\n\nAlgorithm combination:");
- test_ipsec_display_alg(alg_list[ctx.td_idx].param1,
- alg_list[ctx.td_idx].param2);
+ test_sec_alg_display(alg_list[ctx.td_idx].param1,
alg_list[ctx.td_idx].param2);
printf("--");
printf("\n%20s%20s%20s%20s%20s%20s\n\n",
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index 9644566acc..218761f0a9 100644
--- a/app/test/test_cryptodev.c
+++ b/app/test/test_cryptodev.c
@@ -10540,8 +10540,7 @@ test_ipsec_proto_all(const struct ipsec_test_flags
*flags)
return TEST_FAILED;
if (flags->display_alg)
- test_ipsec_display_alg(alg_list[i].param1,
- alg_list[i].param2);
+ test_sec_alg_display(alg_list[i].param1,
alg_list[i].param2);
pass_cnt++;
}
@@ -10586,8 +10585,7 @@ test_ipsec_ah_proto_all(const struct ipsec_test_flags
*flags)
return TEST_FAILED;
if (flags->display_alg)
- test_ipsec_display_alg(ah_alg_list[i].param1,
- ah_alg_list[i].param2);
+ test_sec_alg_display(ah_alg_list[i].param1,
ah_alg_list[i].param2);
pass_cnt++;
}
diff --git a/app/test/test_cryptodev_security_ipsec.c
b/app/test/test_cryptodev_security_ipsec.c
index 01e0a45ffd..dafe3244e8 100644
--- a/app/test/test_cryptodev_security_ipsec.c
+++ b/app/test/test_cryptodev_security_ipsec.c
@@ -549,32 +549,6 @@ test_ipsec_td_update(struct ipsec_test_data td_inb[],
}
}
-void
-test_ipsec_display_alg(const struct crypto_param *param1,
- const struct crypto_param *param2)
-{
- if (param1->type == RTE_CRYPTO_SYM_XFORM_AEAD) {
- printf("\t%s [%d]",
- rte_cryptodev_get_aead_algo_string(param1->alg.aead),
- param1->key_length * 8);
- } else if (param1->type == RTE_CRYPTO_SYM_XFORM_AUTH) {
- printf("\t%s",
- rte_cryptodev_get_auth_algo_string(param1->alg.auth));
- if (param1->alg.auth != RTE_CRYPTO_AUTH_NULL)
- printf(" [%dB ICV]", param1->digest_length);
- } else {
- printf("\t%s",
-
rte_cryptodev_get_cipher_algo_string(param1->alg.cipher));
- if (param1->alg.cipher != RTE_CRYPTO_CIPHER_NULL)
- printf(" [%d]", param1->key_length * 8);
- printf(" %s",
- rte_cryptodev_get_auth_algo_string(param2->alg.auth));
- if (param2->alg.auth != RTE_CRYPTO_AUTH_NULL)
- printf(" [%dB ICV]", param2->digest_length);
- }
- printf("\n");
-}
-
static int
test_ipsec_tunnel_hdr_len_get(const struct ipsec_test_data *td)
{
diff --git a/app/test/test_cryptodev_security_ipsec.h
b/app/test/test_cryptodev_security_ipsec.h
index 77cb5fa7f9..41aa865a0f 100644
--- a/app/test/test_cryptodev_security_ipsec.h
+++ b/app/test/test_cryptodev_security_ipsec.h
@@ -155,9 +155,6 @@ void test_ipsec_td_update(struct ipsec_test_data td_inb[],
int nb_td,
const struct ipsec_test_flags *flags);
-void test_ipsec_display_alg(const struct crypto_param *param1,
- const struct crypto_param *param2);
-
int test_ipsec_post_process(const struct rte_mbuf *m,
const struct ipsec_test_data *td,
struct ipsec_test_data *res_d, bool silent,
diff --git a/app/test/test_security_inline_proto.c
b/app/test/test_security_inline_proto.c
index 9644a3c39c..6cda9fe7aa 100644
--- a/app/
[PATCH 04/14] test/security: add sha1-hmac to auth list
Add SHA1-HMAC with 12B digest to the auth list. This algorithm would be
used in IPsec tests and combined tests would run on devices that support
the capabilities.
Signed-off-by: Anoob Joseph
Signed-off-by: Vidya Sagar Velumuri
---
app/test/test_security_proto.h | 6 ++
1 file changed, 6 insertions(+)
diff --git a/app/test/test_security_proto.h b/app/test/test_security_proto.h
index 596402fff9..d6a00018e3 100644
--- a/app/test/test_security_proto.h
+++ b/app/test/test_security_proto.h
@@ -99,6 +99,12 @@ static const struct crypto_param auth_list[] = {
.key_length = 16,
.digest_length = 12,
},
+ {
+ .type = RTE_CRYPTO_SYM_XFORM_AUTH,
+ .alg.auth = RTE_CRYPTO_AUTH_SHA1_HMAC,
+ .key_length = 20,
+ .digest_length = 12,
+ },
{
.type = RTE_CRYPTO_SYM_XFORM_AUTH,
.alg.auth = RTE_CRYPTO_AUTH_SHA256_HMAC,
--
2.25.1
[PATCH 05/14] test/crypto: move algorithm framework to common
From: Vidya Sagar Velumuri
Move algorithm list and the associated framework to common files
(test_security_proto.c & .h) to be able to use same framework for other
security offload tests such as TLS record.
Signed-off-by: Anoob Joseph
Signed-off-by: Vidya Sagar Velumuri
---
app/test-security-perf/test_security_perf.c | 17 +++---
app/test/test_cryptodev.c | 35 ++--
app/test/test_cryptodev_security_ipsec.c| 60 +++--
app/test/test_cryptodev_security_ipsec.h| 16 --
app/test/test_security_inline_proto.c | 27 +-
app/test/test_security_proto.c | 46
app/test/test_security_proto.h | 16 ++
7 files changed, 109 insertions(+), 108 deletions(-)
diff --git a/app/test-security-perf/test_security_perf.c
b/app/test-security-perf/test_security_perf.c
index 84726de7b0..b0d55cbab0 100644
--- a/app/test-security-perf/test_security_perf.c
+++ b/app/test-security-perf/test_security_perf.c
@@ -474,8 +474,8 @@ args_parse(int argc, char **argv)
int
main(int argc, char **argv)
{
- struct ipsec_test_data td_outb[RTE_DIM(alg_list)];
- struct ipsec_test_data td_inb[RTE_DIM(alg_list)];
+ struct ipsec_test_data td_outb[RTE_DIM(sec_alg_list)];
+ struct ipsec_test_data td_inb[RTE_DIM(sec_alg_list)];
struct ipsec_test_flags flags;
uint32_t lcore_id;
uint8_t nb_lcores;
@@ -513,11 +513,11 @@ main(int argc, char **argv)
if (ret)
goto cryptodev_fini;
- test_ipsec_alg_list_populate();
+ test_sec_alg_list_populate();
- for (i = 0; i < RTE_DIM(alg_list); i++) {
- test_ipsec_td_prepare(alg_list[i].param1,
- alg_list[i].param2,
+ for (i = 0; i < RTE_DIM(sec_alg_list); i++) {
+ test_ipsec_td_prepare(sec_alg_list[i].param1,
+ sec_alg_list[i].param2,
&flags,
&td_outb[i],
1);
@@ -529,10 +529,11 @@ main(int argc, char **argv)
if (ctx.is_inbound)
ctx.td = td_inb;
- for (ctx.td_idx = 0; ctx.td_idx < RTE_DIM(alg_list); ctx.td_idx++) {
+ for (ctx.td_idx = 0; ctx.td_idx < RTE_DIM(sec_alg_list); ctx.td_idx++) {
printf("\n\nAlgorithm combination:");
- test_sec_alg_display(alg_list[ctx.td_idx].param1,
alg_list[ctx.td_idx].param2);
+ test_sec_alg_display(sec_alg_list[ctx.td_idx].param1,
+sec_alg_list[ctx.td_idx].param2);
printf("--");
printf("\n%20s%20s%20s%20s%20s%20s\n\n",
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index 218761f0a9..fbb97d5456 100644
--- a/app/test/test_cryptodev.c
+++ b/app/test/test_cryptodev.c
@@ -838,8 +838,8 @@ ipsec_proto_testsuite_setup(void)
ret = TEST_SKIPPED;
}
- test_ipsec_alg_list_populate();
- test_ipsec_ah_alg_list_populate();
+ test_sec_alg_list_populate();
+ test_sec_auth_only_alg_list_populate();
/*
* Stop the device. Device would be started again by individual test
@@ -10483,19 +10483,19 @@ test_ipsec_proto_known_vec_inb_rx_inject(const void
*test_data)
static int
test_ipsec_proto_all(const struct ipsec_test_flags *flags)
{
- struct ipsec_test_data td_outb[IPSEC_TEST_PACKETS_MAX];
- struct ipsec_test_data td_inb[IPSEC_TEST_PACKETS_MAX];
+ struct ipsec_test_data td_outb[TEST_SEC_PKTS_MAX];
+ struct ipsec_test_data td_inb[TEST_SEC_PKTS_MAX];
unsigned int i, nb_pkts = 1, pass_cnt = 0;
int ret;
if (flags->iv_gen ||
flags->sa_expiry_pkts_soft ||
flags->sa_expiry_pkts_hard)
- nb_pkts = IPSEC_TEST_PACKETS_MAX;
+ nb_pkts = TEST_SEC_PKTS_MAX;
- for (i = 0; i < RTE_DIM(alg_list); i++) {
- test_ipsec_td_prepare(alg_list[i].param1,
- alg_list[i].param2,
+ for (i = 0; i < RTE_DIM(sec_alg_list); i++) {
+ test_ipsec_td_prepare(sec_alg_list[i].param1,
+ sec_alg_list[i].param2,
flags,
td_outb,
nb_pkts);
@@ -10540,7 +10540,7 @@ test_ipsec_proto_all(const struct ipsec_test_flags
*flags)
return TEST_FAILED;
if (flags->display_alg)
- test_sec_alg_display(alg_list[i].param1,
alg_list[i].param2);
+ test_sec_alg_display(sec_alg_list[i].param1,
sec_alg_list[i].param2);
pass_cnt++;
}
@@ -10554,14 +10554,14 @@ test_ipsec_proto_all(const struct ipsec_test_flags
*flags)
static int
test_ipsec_ah_proto_al
[PATCH 06/14] test/crypto: add TLS record tests
Add framework for testing TLS record cases. The framework supports
testing with known vector tests in both session types (read & write).
Signed-off-by: Anoob Joseph
Signed-off-by: Vidya Sagar Velumuri
---
app/test/meson.build | 1 +
app/test/test_cryptodev.c | 282 +-
app/test/test_cryptodev.h | 2 +
app/test/test_cryptodev_security_tls_record.c | 151 ++
app/test/test_cryptodev_security_tls_record.h | 71 +
...yptodev_security_tls_record_test_vectors.h | 16 +
6 files changed, 515 insertions(+), 8 deletions(-)
create mode 100644 app/test/test_cryptodev_security_tls_record.c
create mode 100644 app/test/test_cryptodev_security_tls_record.h
create mode 100644 app/test/test_cryptodev_security_tls_record_test_vectors.h
diff --git a/app/test/meson.build b/app/test/meson.build
index 58e120a6ab..f9e81eda2e 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -56,6 +56,7 @@ source_file_deps = {
'test_cryptodev_crosscheck.c': test_cryptodev_deps,
'test_cryptodev_security_ipsec.c': test_cryptodev_deps,
'test_cryptodev_security_pdcp.c': test_cryptodev_deps,
+'test_cryptodev_security_tls_record.c': ['cryptodev', 'security'],
'test_cycles.c': [],
'test_debug.c': [],
'test_devargs.c': ['kvargs'],
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index fbb97d5456..cd83370790 100644
--- a/app/test/test_cryptodev.c
+++ b/app/test/test_cryptodev.c
@@ -53,6 +53,7 @@
#include "test_cryptodev_security_pdcp_sdap_test_vectors.h"
#include "test_cryptodev_security_pdcp_test_func.h"
#include "test_cryptodev_security_docsis_test_vectors.h"
+#include "test_cryptodev_security_tls_record.h"
#include "test_security_proto.h"
#define SDAP_DISABLED 0
@@ -807,7 +808,7 @@ crypto_gen_testsuite_setup(void)
#ifdef RTE_LIB_SECURITY
static int
-ipsec_proto_testsuite_setup(void)
+sec_proto_testsuite_setup(enum rte_security_session_protocol protocol)
{
struct crypto_testsuite_params *ts_params = &testsuite_params;
struct crypto_unittest_params *ut_params = &unittest_params;
@@ -817,8 +818,8 @@ ipsec_proto_testsuite_setup(void)
rte_cryptodev_info_get(ts_params->valid_devs[0], &dev_info);
if (!(dev_info.feature_flags & RTE_CRYPTODEV_FF_SECURITY)) {
- RTE_LOG(INFO, USER1, "Feature flag requirements for IPsec Proto
"
- "testsuite not met\n");
+ RTE_LOG(INFO, USER1,
+ "Feature flag requirements for security protocol
testsuite not met\n");
return TEST_SKIPPED;
}
@@ -830,11 +831,9 @@ ipsec_proto_testsuite_setup(void)
/* Set action type */
ut_params->type = RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL;
- if (security_proto_supported(
- RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL,
- RTE_SECURITY_PROTOCOL_IPSEC) < 0) {
- RTE_LOG(INFO, USER1, "Capability requirements for IPsec Proto "
- "test not met\n");
+ if
(security_proto_supported(RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL,
protocol) < 0) {
+ RTE_LOG(INFO, USER1,
+ "Capability requirements for security protocol test not
met\n");
ret = TEST_SKIPPED;
}
@@ -850,6 +849,18 @@ ipsec_proto_testsuite_setup(void)
return ret;
}
+static int
+ipsec_proto_testsuite_setup(void)
+{
+ return sec_proto_testsuite_setup(RTE_SECURITY_PROTOCOL_IPSEC);
+}
+
+static int
+tls_record_proto_testsuite_setup(void)
+{
+ return sec_proto_testsuite_setup(RTE_SECURITY_PROTOCOL_TLS_RECORD);
+}
+
static int
pdcp_proto_testsuite_setup(void)
{
@@ -11654,6 +11665,244 @@ test_docsis_proto_downlink(const void *data)
return ret;
}
+
+static void
+test_tls_record_imp_nonce_update(const struct tls_record_test_data *td,
+struct rte_security_tls_record_xform
*tls_record_xform)
+{
+ unsigned int imp_nonce_len;
+ uint8_t *imp_nonce;
+
+ switch (tls_record_xform->ver) {
+ case RTE_SECURITY_VERSION_TLS_1_2:
+ imp_nonce_len = RTE_SECURITY_TLS_1_2_IMP_NONCE_LEN;
+ imp_nonce = tls_record_xform->tls_1_2.imp_nonce;
+ break;
+ case RTE_SECURITY_VERSION_DTLS_1_2:
+ imp_nonce_len = RTE_SECURITY_DTLS_1_2_IMP_NONCE_LEN;
+ imp_nonce = tls_record_xform->dtls_1_2.imp_nonce;
+ break;
+ case RTE_SECURITY_VERSION_TLS_1_3:
+ imp_nonce_len = RTE_SECURITY_TLS_1_3_IMP_NONCE_LEN;
+ imp_nonce = tls_record_xform->tls_1_3.imp_nonce;
+ break;
+ default:
+ return;
+ }
+
+ imp_nonce_len = RTE_MIN(imp_nonce_len, td[0].imp_nonce.len);
+ memcpy(imp_nonce, td[0].imp_nonce.data, imp_nonce_len);
+}
+
+static int
+test_tls
[PATCH 07/14] test/crypto: add AES-GCM 128 TLS 1.2 vector
From: Tejasree Kondoj
Add AES-GCM-128 test vector generated with kTLS.
Signed-off-by: Tejasree Kondoj
Signed-off-by: Vidya Sagar Velumuri
---
app/test/test_cryptodev.c | 8 +--
app/test/test_cryptodev_security_tls_record.h | 2 +-
...yptodev_security_tls_record_test_vectors.h | 67 ++-
3 files changed, 71 insertions(+), 6 deletions(-)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index cd83370790..e7e4b93c20 100644
--- a/app/test/test_cryptodev.c
+++ b/app/test/test_cryptodev.c
@@ -16814,13 +16814,13 @@ static struct unit_test_suite
tls12_record_proto_testsuite = {
.setup = tls_record_proto_testsuite_setup,
.unit_test_cases = {
TEST_CASE_NAMED_WITH_DATA(
- "Known vector TBD",
+ "Write record known vector AES-GCM-128 (vector 1)",
ut_setup_security, ut_teardown,
- test_tls_record_proto_known_vec, &tls_test_data1),
+ test_tls_record_proto_known_vec,
&tls_test_data_aes_128_gcm_v1),
TEST_CASE_NAMED_WITH_DATA(
- "Known vector TBD",
+ "Read record known vector AES-GCM-128 (vector 1)",
ut_setup_security, ut_teardown,
- test_tls_record_proto_known_vec_read, &tls_test_data1),
+ test_tls_record_proto_known_vec_read,
&tls_test_data_aes_128_gcm_v1),
TEST_CASES_END() /**< NULL terminate unit test array */
}
};
diff --git a/app/test/test_cryptodev_security_tls_record.h
b/app/test/test_cryptodev_security_tls_record.h
index 9a0cf70218..b9cadd5cd4 100644
--- a/app/test/test_cryptodev_security_tls_record.h
+++ b/app/test/test_cryptodev_security_tls_record.h
@@ -55,7 +55,7 @@ struct tls_record_test_flags {
bool display_alg;
};
-extern struct tls_record_test_data tls_test_data1;
+extern struct tls_record_test_data tls_test_data_aes_128_gcm_v1;
int test_tls_record_status_check(struct rte_crypto_op *op);
diff --git a/app/test/test_cryptodev_security_tls_record_test_vectors.h
b/app/test/test_cryptodev_security_tls_record_test_vectors.h
index 5aa0d27fda..52e97e8557 100644
--- a/app/test/test_cryptodev_security_tls_record_test_vectors.h
+++ b/app/test/test_cryptodev_security_tls_record_test_vectors.h
@@ -11,6 +11,71 @@
#include "test_cryptodev.h"
#include "test_cryptodev_security_tls_record.h"
-struct tls_record_test_data tls_test_data1;
+/* TLS 1.2 AES-GCM 128 vector generated with kTLS. */
+struct tls_record_test_data tls_test_data_aes_128_gcm_v1 = {
+ .key = {
+ .data = {
+ 0x77, 0xc4, 0x9b, 0x0c, 0x2b, 0xe2, 0xd2, 0x4e,
+ 0xf5, 0x36, 0xb4, 0xea, 0x16, 0xb0, 0xed, 0x1f
+ },
+ },
+ .input_text = {
+ .data = {
+ 0x61, 0x62, 0x63, 0x64, 0x31, 0x32, 0x33, 0x34,
+ 0x61, 0x62, 0x63, 0x64, 0x31, 0x32, 0x33, 0x34,
+ 0xa
+ },
+ .len = 17,
+ },
+ .output_text = {
+ .data = {
+ 0x17, 0x03, 0x03, 0x00, 0x29,
+ 0x3a, 0xcd, 0x41, 0x5d, 0x42, 0xac, 0xce, 0x36,
+ 0xad, 0xd1, 0x83, 0xa0, 0x16, 0x84, 0xce, 0x84,
+ 0xfc, 0xc2, 0xa7, 0x6b, 0xa1, 0x89, 0x5d, 0xf2,
+ 0xeb,
+ 0x9b, 0x90, 0x01, 0x56, 0xf8, 0x9d, 0xac, 0x0c,
+ 0xa8, 0x95, 0xba, 0xcc, 0x1f, 0xf3, 0x36, 0x2b
+ },
+ .len = 46,
+ },
+ .imp_nonce = {
+ .data = {
+ 0xda, 0xf8, 0x8f, 0x12
+ },
+ .len = 4,
+ },
+
+ .iv = {
+ .data = {
+ 0x3a, 0xcd, 0x41, 0x5d, 0x42, 0xac, 0xce, 0x36
+ },
+ },
+
+ .xform = {
+ .aead = {
+ .next = NULL,
+ .type = RTE_CRYPTO_SYM_XFORM_AEAD,
+ .aead = {
+ .op = RTE_CRYPTO_AEAD_OP_ENCRYPT,
+ .algo = RTE_CRYPTO_AEAD_AES_GCM,
+ .key.length = 16,
+ .iv.length = 12,
+ .iv.offset = IV_OFFSET,
+ .digest_length = 16,
+ .aad_length = 13,
+ },
+ },
+ },
+
+ .tls_record_xform = {
+ .ver = RTE_SECURITY_VERSION_TLS_1_2,
+ .type = RTE_SECURITY_TLS_SESS_TYPE_WRITE,
+ .tls_1_2.seq_no = 0x1,
+ },
+
+ .aead = true,
+ .app_type = 0x17,
+};
#endif
--
2.25.1
[PATCH 08/14] test/crypto: add TLS1.2 vectors
From: Akhil Goyal
Added TLS 1.2 test vectors for AES-128-CBC-SHA1
Signed-off-by: Akhil Goyal
---
app/test/test_cryptodev.c | 8 ++
app/test/test_cryptodev_security_tls_record.h | 1 +
...yptodev_security_tls_record_test_vectors.h | 91 +++
3 files changed, 100 insertions(+)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index e7e4b93c20..a82f971135 100644
--- a/app/test/test_cryptodev.c
+++ b/app/test/test_cryptodev.c
@@ -16817,10 +16817,18 @@ static struct unit_test_suite
tls12_record_proto_testsuite = {
"Write record known vector AES-GCM-128 (vector 1)",
ut_setup_security, ut_teardown,
test_tls_record_proto_known_vec,
&tls_test_data_aes_128_gcm_v1),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Write record known vector AES-128-CBC-SHA1",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec,
&tls_test_data_aes_128_cbc_sha1_hmac),
TEST_CASE_NAMED_WITH_DATA(
"Read record known vector AES-GCM-128 (vector 1)",
ut_setup_security, ut_teardown,
test_tls_record_proto_known_vec_read,
&tls_test_data_aes_128_gcm_v1),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Read record known vector AES-128-CBC-SHA1",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec_read,
&tls_test_data_aes_128_cbc_sha1_hmac),
TEST_CASES_END() /**< NULL terminate unit test array */
}
};
diff --git a/app/test/test_cryptodev_security_tls_record.h
b/app/test/test_cryptodev_security_tls_record.h
index b9cadd5cd4..3e42db2c03 100644
--- a/app/test/test_cryptodev_security_tls_record.h
+++ b/app/test/test_cryptodev_security_tls_record.h
@@ -56,6 +56,7 @@ struct tls_record_test_flags {
};
extern struct tls_record_test_data tls_test_data_aes_128_gcm_v1;
+extern struct tls_record_test_data tls_test_data_aes_128_cbc_sha1_hmac;
int test_tls_record_status_check(struct rte_crypto_op *op);
diff --git a/app/test/test_cryptodev_security_tls_record_test_vectors.h
b/app/test/test_cryptodev_security_tls_record_test_vectors.h
index 52e97e8557..85ad2df500 100644
--- a/app/test/test_cryptodev_security_tls_record_test_vectors.h
+++ b/app/test/test_cryptodev_security_tls_record_test_vectors.h
@@ -78,4 +78,95 @@ struct tls_record_test_data tls_test_data_aes_128_gcm_v1 = {
.app_type = 0x17,
};
+struct tls_record_test_data tls_test_data_aes_128_cbc_sha1_hmac = {
+ .key = {
+ .data = {
+ 0xee, 0x2b, 0x2a, 0x59, 0xbf, 0x60, 0x36, 0xba,
+ 0xe4, 0xdc, 0x7d, 0x8b, 0xc7, 0x0d, 0xbb, 0x85,
+ },
+ },
+ .auth_key = {
+ .data = {
+ 0xcf, 0x4f, 0xc1, 0x7d, 0x6d, 0x4c, 0x0d, 0x13,
+ 0x3d, 0x8f, 0x95, 0xb5, 0xdd, 0xb0, 0x85, 0x08,
+ 0x38, 0x2c, 0xa1, 0xa0,
+ },
+ },
+ .input_text = {
+ .data = {
+ /* plain text data */
+ 0x54, 0x68, 0x69, 0x73, 0x20, 0x69, 0x73, 0x20,
+ 0x54, 0x4c, 0x53, 0x20, 0x31, 0x2e, 0x32, 0x20,
+ 0x41, 0x45, 0x53, 0x2d, 0x31, 0x32, 0x38, 0x2d,
+ 0x43, 0x42, 0x43, 0x20, 0x53, 0x48, 0x41, 0x31,
+ 0x20, 0x65, 0x78, 0x61, 0x6d, 0x70, 0x6c, 0x65,
+ 0x20, 0x76, 0x65, 0x63, 0x74, 0x6f, 0x72, 0x0a,
+ },
+ .len = 48,
+ },
+ .output_text = {
+ .data = {
+ /* TLS header */
+ 0x17, 0x03, 0x03, 0x00, 0x60,
+ /* Encrypted data */
+ 0x30, 0x76, 0x10, 0x2e, 0xb2, 0xe3, 0xb6, 0x93,
+ 0x70, 0xd4, 0xdf, 0x2a, 0x4f, 0x92, 0x1d, 0x3c,
+ 0x9a, 0x87, 0x38, 0xb0, 0x8e, 0x82, 0xf1, 0x33,
+ 0xa2, 0x5e, 0xce, 0xaf, 0x51, 0xf1, 0x51, 0x17,
+ 0x4f, 0x63, 0x55, 0x60, 0xa5, 0x3f, 0xc5, 0x42,
+ 0x9a, 0x80, 0x7a, 0x19, 0xb9, 0x77, 0x5b, 0x6f,
+ 0x1b, 0x2e, 0x2c, 0x0a, 0xa1, 0xa9, 0xf4, 0xb1,
+ 0xc7, 0x12, 0x7d, 0xfc, 0x92, 0x2f, 0xf4, 0xfc,
+ 0x20, 0x16, 0x1a, 0xa3, 0x94, 0xc6, 0xfe, 0x64,
+ 0x8d, 0x16, 0xd2, 0x52, 0x56, 0x83, 0x75, 0x3f,
+ 0x5b, 0x89, 0xf4, 0x29, 0x72, 0x46, 0x19, 0xd2,
+ 0xae, 0xea, 0xe3, 0xfc, 0x23, 0xd9, 0x00, 0x6c,
+ },
+ .len = 101,
+ },
+ .iv = {
+ .data = {
+ 0x30, 0x76, 0x10, 0x2e, 0xb2, 0xe3, 0xb6, 0x93,
+
[PATCH 09/14] test/crypto: add TLS1.2/DTLS1.2 AES-128/256-GCM vectors
From: Akhil Goyal
Added test vectors for TLS1.2 and DTLS 1.2 for AES-128-GCM
and AES-256-GCM.
These vectors are generated using GnuTLS server and client applications.
Signed-off-by: Akhil Goyal
Signed-off-by: Vidya Sagar Velumuri
---
app/test/test_cryptodev.c | 43 ++-
app/test/test_cryptodev_security_tls_record.h | 4 +
...yptodev_security_tls_record_test_vectors.h | 286 ++
3 files changed, 332 insertions(+), 1 deletion(-)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index a82f971135..0360f30bd1 100644
--- a/app/test/test_cryptodev.c
+++ b/app/test/test_cryptodev.c
@@ -16818,13 +16818,29 @@ static struct unit_test_suite
tls12_record_proto_testsuite = {
ut_setup_security, ut_teardown,
test_tls_record_proto_known_vec,
&tls_test_data_aes_128_gcm_v1),
TEST_CASE_NAMED_WITH_DATA(
- "Write record known vector AES-128-CBC-SHA1",
+ "Write record known vector AES-GCM-128 (vector 2)",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec,
&tls_test_data_aes_128_gcm_v2),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Write record known vector AES-GCM-256",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec,
&tls_test_data_aes_256_gcm),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Write record known vector AES-CBC-128-SHA1",
ut_setup_security, ut_teardown,
test_tls_record_proto_known_vec,
&tls_test_data_aes_128_cbc_sha1_hmac),
TEST_CASE_NAMED_WITH_DATA(
"Read record known vector AES-GCM-128 (vector 1)",
ut_setup_security, ut_teardown,
test_tls_record_proto_known_vec_read,
&tls_test_data_aes_128_gcm_v1),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Read record known vector AES-GCM-128 (vector 2)",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec_read,
&tls_test_data_aes_128_gcm_v2),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Read record known vector AES-GCM-256",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec_read,
&tls_test_data_aes_256_gcm),
TEST_CASE_NAMED_WITH_DATA(
"Read record known vector AES-128-CBC-SHA1",
ut_setup_security, ut_teardown,
@@ -16833,6 +16849,30 @@ static struct unit_test_suite
tls12_record_proto_testsuite = {
}
};
+static struct unit_test_suite dtls12_record_proto_testsuite = {
+ .suite_name = "DTLS 1.2 Record Protocol Unit Test Suite",
+ .setup = tls_record_proto_testsuite_setup,
+ .unit_test_cases = {
+ TEST_CASE_NAMED_WITH_DATA(
+ "Write record known vector AES-GCM-128",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec,
&dtls_test_data_aes_128_gcm),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Write record known vector AES-GCM-256",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec,
&dtls_test_data_aes_256_gcm),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Read record known vector AES-GCM-128",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec_read,
&dtls_test_data_aes_128_gcm),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Read record known vector AES-GCM-256",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec_read,
&dtls_test_data_aes_256_gcm),
+ TEST_CASES_END() /**< NULL terminate unit test array */
+ }
+};
+
#define ADD_UPLINK_TESTCASE(data)
\
TEST_CASE_NAMED_WITH_DATA(data.test_descr_uplink, ut_setup_security,
\
ut_teardown, test_docsis_proto_uplink, (const void *) &data),
\
@@ -17841,6 +17881,7 @@ run_cryptodev_testsuite(const char *pmd_name)
&pdcp_proto_testsuite,
&docsis_proto_testsuite,
&tls12_record_proto_testsuite,
+ &dtls12_record_proto_testsuite,
#endif
&end_testsuite
};
diff --git a/app/test/test_cryptodev_security_tls_record.h
b/app/test/test_cryptodev_security_tls_record.h
index 3e42db2c03..18d689253b 100644
--- a/app/test/test_cryptodev_security_tls_record.h
+++ b/app/test/test_cryptodev_security_tls_record.h
@@ -56,6 +56,10 @@ struct tls_record_test_flags {
};
extern struct tls_record_test_data
[PATCH 10/14] test/crypto: add combined mode cases
From: Vidya Sagar Velumuri
Add cases to try TLS record write(encrypt) + read(decrypt) operations.
This is used for testing TLS record features with all algorithms
supported by the security device.
Signed-off-by: Anoob Joseph
Signed-off-by: Vidya Sagar Velumuri
---
app/test/test_cryptodev.c | 67 ++-
app/test/test_cryptodev_security_tls_record.c | 60 +
app/test/test_cryptodev_security_tls_record.h | 11 +++
3 files changed, 136 insertions(+), 2 deletions(-)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index 0360f30bd1..6070e7f336 100644
--- a/app/test/test_cryptodev.c
+++ b/app/test/test_cryptodev.c
@@ -11704,6 +11704,7 @@ test_tls_record_proto_process(const struct
tls_record_test_data td[],
struct rte_security_tls_record_xform tls_record_xform;
struct rte_security_capability_idx sec_cap_idx;
const struct rte_security_capability *sec_cap;
+ struct tls_record_test_data *res_d_tmp = NULL;
enum rte_security_tls_sess_type sess_type;
uint8_t dev_id = ts_params->valid_devs[0];
struct rte_security_ctx *ctx;
@@ -11841,7 +11842,10 @@ test_tls_record_proto_process(const struct
tls_record_test_data td[],
if (ret != TEST_SUCCESS)
goto crypto_op_free;
- ret = test_tls_record_post_process(ut_params->ibuf, &td[i],
NULL, silent);
+ if (res_d != NULL)
+ res_d_tmp = &res_d[i];
+
+ ret = test_tls_record_post_process(ut_params->ibuf, &td[i],
res_d_tmp, silent);
if (ret != TEST_SUCCESS)
goto crypto_op_free;
@@ -11864,7 +11868,6 @@ test_tls_record_proto_process(const struct
tls_record_test_data td[],
rte_security_session_destroy(ctx, ut_params->sec_session);
ut_params->sec_session = NULL;
- RTE_SET_USED(res_d);
RTE_SET_USED(flags);
return ret;
@@ -11903,6 +11906,58 @@ test_tls_record_proto_known_vec_read(const void
*test_data)
return test_tls_record_proto_process(&td_inb, NULL, 1, false, &flags);
}
+static int
+test_tls_record_proto_all(const struct tls_record_test_flags *flags)
+{
+ struct tls_record_test_data td_outb[TEST_SEC_PKTS_MAX];
+ struct tls_record_test_data td_inb[TEST_SEC_PKTS_MAX];
+ unsigned int i, nb_pkts = 1, pass_cnt = 0;
+ int ret;
+
+ for (i = 0; i < RTE_DIM(sec_alg_list); i++) {
+ test_tls_record_td_prepare(sec_alg_list[i].param1,
sec_alg_list[i].param2, flags,
+ td_outb, nb_pkts);
+
+ ret = test_tls_record_proto_process(td_outb, td_inb, nb_pkts,
true, flags);
+ if (ret == TEST_SKIPPED)
+ continue;
+
+ if (ret == TEST_FAILED)
+ return TEST_FAILED;
+
+ test_tls_record_td_update(td_inb, td_outb, nb_pkts, flags);
+
+ ret = test_tls_record_proto_process(td_inb, NULL, nb_pkts,
true, flags);
+ if (ret == TEST_SKIPPED)
+ continue;
+
+ if (ret == TEST_FAILED)
+ return TEST_FAILED;
+
+ if (flags->display_alg)
+ test_sec_alg_display(sec_alg_list[i].param1,
sec_alg_list[i].param2);
+
+ pass_cnt++;
+ }
+
+ if (pass_cnt > 0)
+ return TEST_SUCCESS;
+ else
+ return TEST_SKIPPED;
+}
+
+static int
+test_tls_record_proto_display_list(void)
+{
+ struct tls_record_test_flags flags;
+
+ memset(&flags, 0, sizeof(flags));
+
+ flags.display_alg = true;
+
+ return test_tls_record_proto_all(&flags);
+}
+
#endif
static int
@@ -16845,6 +16900,10 @@ static struct unit_test_suite
tls12_record_proto_testsuite = {
"Read record known vector AES-128-CBC-SHA1",
ut_setup_security, ut_teardown,
test_tls_record_proto_known_vec_read,
&tls_test_data_aes_128_cbc_sha1_hmac),
+ TEST_CASE_NAMED_ST(
+ "Combined test alg list",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_display_list),
TEST_CASES_END() /**< NULL terminate unit test array */
}
};
@@ -16869,6 +16928,10 @@ static struct unit_test_suite
dtls12_record_proto_testsuite = {
"Read record known vector AES-GCM-256",
ut_setup_security, ut_teardown,
test_tls_record_proto_known_vec_read,
&dtls_test_data_aes_256_gcm),
+ TEST_CASE_NAMED_ST(
+ "Combined test alg list",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_display_list),
TEST_CASES_END() /**< NULL terminate unit test array */
}
};
d
[PATCH 11/14] test/crypto: add verification of TLS headers
Add verification of TLS headers in protocol offload tests.
Signed-off-by: Anoob Joseph
Signed-off-by: Vidya Sagar Velumuri
---
app/test/test_cryptodev_security_tls_record.c | 118 +-
1 file changed, 117 insertions(+), 1 deletion(-)
diff --git a/app/test/test_cryptodev_security_tls_record.c
b/app/test/test_cryptodev_security_tls_record.c
index 6f106050c2..bcb2eba4ff 100644
--- a/app/test/test_cryptodev_security_tls_record.c
+++ b/app/test/test_cryptodev_security_tls_record.c
@@ -3,6 +3,8 @@
*/
#include
+#include
+#include
#include "test.h"
#include "test_cryptodev_security_tls_record.h"
@@ -62,8 +64,8 @@ test_tls_record_td_prepare(const struct crypto_param *param1,
const struct crypt
const struct tls_record_test_flags *flags,
struct tls_record_test_data *td_array, int nb_td)
{
+ int i, min_padding, hdr_len, tls_pkt_size, mac_len = 0, exp_nonce_len =
0, roundup_len = 0;
struct tls_record_test_data *td = NULL;
- int i;
memset(td_array, 0, nb_td * sizeof(*td));
@@ -94,6 +96,59 @@ test_tls_record_td_prepare(const struct crypto_param
*param1, const struct crypt
}
}
+ tls_pkt_size = td->input_text.len;
+
+ if (!td->aead) {
+ mac_len = td->xform.chain.auth.auth.digest_length;
+ switch (td->xform.chain.cipher.cipher.algo) {
+ case RTE_CRYPTO_CIPHER_3DES_CBC:
+ roundup_len = 8;
+ exp_nonce_len = 8;
+ break;
+ case RTE_CRYPTO_CIPHER_AES_CBC:
+ roundup_len = 16;
+ exp_nonce_len = 16;
+ break;
+ default:
+ roundup_len = 0;
+ exp_nonce_len = 0;
+ break;
+ }
+ } else {
+ mac_len = td->xform.aead.aead.digest_length;
+ exp_nonce_len = 8;
+ }
+
+ switch (td->tls_record_xform.ver) {
+ case RTE_SECURITY_VERSION_TLS_1_2:
+ case RTE_SECURITY_VERSION_TLS_1_3:
+ hdr_len = sizeof(struct rte_tls_hdr);
+ min_padding = 1;
+ break;
+ case RTE_SECURITY_VERSION_DTLS_1_2:
+ hdr_len = sizeof(struct rte_dtls_hdr);
+ min_padding = 0;
+ break;
+ default:
+ hdr_len = 0;
+ min_padding = 0;
+ break;
+ }
+
+ tls_pkt_size += mac_len;
+
+ /* Padding */
+ tls_pkt_size += min_padding;
+ tls_pkt_size = RTE_ALIGN_MUL_CEIL(tls_pkt_size, roundup_len);
+
+ /* Explicit nonce */
+ tls_pkt_size += exp_nonce_len;
+
+ /* Add TLS header */
+ tls_pkt_size += hdr_len;
+
+ td->output_text.len = tls_pkt_size;
+
RTE_SET_USED(flags);
}
@@ -160,6 +215,60 @@ test_tls_record_res_d_prepare(const uint8_t *output_text,
uint32_t len,
return TEST_SUCCESS;
}
+static int
+tls_record_hdr_verify(const struct tls_record_test_data *td, const uint8_t
*output_text)
+{
+ uint16_t length, hdr_len;
+ uint8_t content_type;
+
+ if (td->tls_record_xform.ver == RTE_SECURITY_VERSION_TLS_1_2) {
+ const struct rte_tls_hdr *hdr = (const struct rte_tls_hdr
*)output_text;
+ if (rte_be_to_cpu_16(hdr->version) != RTE_TLS_VERSION_1_2) {
+ printf("Incorrect header version [expected - %4x,
received - %4x]\n",
+ RTE_TLS_VERSION_1_2,
rte_be_to_cpu_16(hdr->version));
+ return TEST_FAILED;
+ }
+ content_type = hdr->type;
+ length = rte_be_to_cpu_16(hdr->length);
+ hdr_len = sizeof(struct rte_tls_hdr);
+ } else if (td->tls_record_xform.ver == RTE_SECURITY_VERSION_TLS_1_3) {
+ const struct rte_tls_hdr *hdr = (const struct rte_tls_hdr
*)output_text;
+ if (rte_be_to_cpu_16(hdr->version) != RTE_TLS_VERSION_1_3) {
+ printf("Incorrect header version [expected - %4x,
received - %4x]\n",
+ RTE_TLS_VERSION_1_3,
rte_be_to_cpu_16(hdr->version));
+ return TEST_FAILED;
+ }
+ content_type = hdr->type;
+ length = rte_be_to_cpu_16(hdr->length);
+ hdr_len = sizeof(struct rte_tls_hdr);
+ } else if (td->tls_record_xform.ver == RTE_SECURITY_VERSION_DTLS_1_2) {
+ const struct rte_dtls_hdr *hdr = (const struct rte_dtls_hdr
*)output_text;
+ if (rte_be_to_cpu_16(hdr->version) != RTE_DTLS_VERSION_1_2) {
+ printf("Incorrect header version [expected - %4x,
received - %4x]\n",
+ RTE_DTLS_VERSION_1_2,
rte_be_to_cpu_16(hdr->version));
+ return TEST_FAILED;
+ }
+
[PATCH 12/14] test/security: add more algos to combined tests
From: Vidya Sagar Velumuri
Add following algos to combined mode cases. These are intended primarily
for TLS record combined mode tests.
- AES256-CBC
- SHA256-HMAC
- SHA1-HMAC [20B]
Signed-off-by: Anoob Joseph
Signed-off-by: Vidya Sagar Velumuri
---
app/test/test_security_proto.h | 18 ++
1 file changed, 18 insertions(+)
diff --git a/app/test/test_security_proto.h b/app/test/test_security_proto.h
index 6797e8b8c2..efa023b99d 100644
--- a/app/test/test_security_proto.h
+++ b/app/test/test_security_proto.h
@@ -70,6 +70,12 @@ static const struct crypto_param cipher_list[] = {
.key_length = 16,
.iv_length = 16,
},
+ {
+ .type = RTE_CRYPTO_SYM_XFORM_CIPHER,
+ .alg.cipher = RTE_CRYPTO_CIPHER_AES_CBC,
+ .key_length = 32,
+ .iv_length = 16,
+ },
{
.type = RTE_CRYPTO_SYM_XFORM_CIPHER,
.alg.cipher = RTE_CRYPTO_CIPHER_AES_CTR,
@@ -107,12 +113,24 @@ static const struct crypto_param auth_list[] = {
.key_length = 20,
.digest_length = 12,
},
+ {
+ .type = RTE_CRYPTO_SYM_XFORM_AUTH,
+ .alg.auth = RTE_CRYPTO_AUTH_SHA1_HMAC,
+ .key_length = 20,
+ .digest_length = 20,
+ },
{
.type = RTE_CRYPTO_SYM_XFORM_AUTH,
.alg.auth = RTE_CRYPTO_AUTH_SHA256_HMAC,
.key_length = 32,
.digest_length = 16,
},
+ {
+ .type = RTE_CRYPTO_SYM_XFORM_AUTH,
+ .alg.auth = RTE_CRYPTO_AUTH_SHA256_HMAC,
+ .key_length = 32,
+ .digest_length = 32,
+ },
{
.type = RTE_CRYPTO_SYM_XFORM_AUTH,
.alg.auth = RTE_CRYPTO_AUTH_SHA384_HMAC,
--
2.25.1
[PATCH 13/14] test/security: add TLS 1.2 and DTLS 1.2 vectors
From: Akhil Goyal
Add TLS v1.2 vectors generated using gnuTLS for
AES-256, SHA256, 3DES, CHACHA-POLY
Add DTLS v1.2 test vectors generated using gnuTLS for
AES-128-CBC, AES-256-CBC, SHA1, SHA256, 3DES-CBC,
CHACHA20-POLY1305 and NULL cipher
Signed-off-by: Akhil Goyal
Signed-off-by: Vidya Sagar Velumuri
---
app/test/test_cryptodev.c | 121 ++
app/test/test_cryptodev_security_tls_record.h | 13 +
...yptodev_security_tls_record_test_vectors.h | 1126 +
3 files changed, 1260 insertions(+)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index 6070e7f336..86e2ccd483 100644
--- a/app/test/test_cryptodev.c
+++ b/app/test/test_cryptodev.c
@@ -16884,6 +16884,31 @@ static struct unit_test_suite
tls12_record_proto_testsuite = {
"Write record known vector AES-CBC-128-SHA1",
ut_setup_security, ut_teardown,
test_tls_record_proto_known_vec,
&tls_test_data_aes_128_cbc_sha1_hmac),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Write record known vector AES-128-CBC-SHA256",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec,
&tls_test_data_aes_128_cbc_sha256_hmac),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Write record known vector AES-256-CBC-SHA1",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec,
&tls_test_data_aes_256_cbc_sha1_hmac),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Write record known vector AES-256-CBC-SHA256",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec,
&tls_test_data_aes_256_cbc_sha256_hmac),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Write record known vector 3DES-CBC-SHA1-HMAC",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec,
&tls_test_data_3des_cbc_sha1_hmac),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Write record known vector NULL-SHA1-HMAC",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec,
&tls_test_data_null_cipher_sha1_hmac),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Write record known vector CHACHA20-POLY1305",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec,
&tls_test_data_chacha20_poly1305),
+
TEST_CASE_NAMED_WITH_DATA(
"Read record known vector AES-GCM-128 (vector 1)",
ut_setup_security, ut_teardown,
@@ -16900,6 +16925,33 @@ static struct unit_test_suite
tls12_record_proto_testsuite = {
"Read record known vector AES-128-CBC-SHA1",
ut_setup_security, ut_teardown,
test_tls_record_proto_known_vec_read,
&tls_test_data_aes_128_cbc_sha1_hmac),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Read record known vector AES-128-CBC-SHA256",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec_read,
+ &tls_test_data_aes_128_cbc_sha256_hmac),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Read record known vector AES-256-CBC-SHA1",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec_read,
&tls_test_data_aes_256_cbc_sha1_hmac),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Read record known vector AES-256-CBC-SHA256",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec_read,
+ &tls_test_data_aes_256_cbc_sha256_hmac),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Read record known vector 3DES-CBC-SHA1-HMAC",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec_read,
&tls_test_data_3des_cbc_sha1_hmac),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Read record known vector NULL-SHA1-HMAC",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec_read,
&tls_test_data_null_cipher_sha1_hmac),
+ TEST_CASE_NAMED_WITH_DATA(
+ "Read record known vector CHACHA20-POLY1305",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_known_vec_read,
&tls_test_data_chacha20_poly1305),
+
TEST_CASE_NAMED_ST(
"Combined test alg list",
ut_setup_security, ut_teardown,
@@ -16920,6 +16972,40 @@ static struct unit_test_suite
dtls12_record_proto_tests
[PATCH 14/14] test/crypto: add multi segmented cases
From: Tejasree Kondoj
Add multi segmented buffer test cases in TLS suites.
Signed-off-by: Tejasree Kondoj
Signed-off-by: Vidya Sagar Velumuri
---
app/test/test_cryptodev.c | 29 ++-
app/test/test_cryptodev_security_tls_record.h | 1 +
doc/guides/rel_notes/release_24_03.rst| 4 +++
3 files changed, 33 insertions(+), 1 deletion(-)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index 86e2ccd483..b208cca4e2 100644
--- a/app/test/test_cryptodev.c
+++ b/app/test/test_cryptodev.c
@@ -11699,6 +11699,7 @@ test_tls_record_proto_process(const struct
tls_record_test_data td[],
struct tls_record_test_data res_d[], int nb_td,
bool silent,
const struct tls_record_test_flags *flags)
{
+ int nb_segs = flags->nb_segs_in_mbuf ? flags->nb_segs_in_mbuf : 1;
struct crypto_testsuite_params *ts_params = &testsuite_params;
struct crypto_unittest_params *ut_params = &unittest_params;
struct rte_security_tls_record_xform tls_record_xform;
@@ -11801,7 +11802,7 @@ test_tls_record_proto_process(const struct
tls_record_test_data td[],
for (i = 0; i < nb_td; i++) {
/* Setup source mbuf payload */
ut_params->ibuf = create_segmented_mbuf(ts_params->mbuf_pool,
td[i].input_text.len,
- 1, 0);
+ nb_segs, 0);
pktmbuf_write(ut_params->ibuf, 0, td[i].input_text.len,
td[i].input_text.data);
/* Generate crypto op data structure */
@@ -11958,6 +11959,24 @@ test_tls_record_proto_display_list(void)
return test_tls_record_proto_all(&flags);
}
+static int
+test_tls_record_proto_sgl(void)
+{
+ struct tls_record_test_flags flags = {
+ .nb_segs_in_mbuf = 5
+ };
+ struct crypto_testsuite_params *ts_params = &testsuite_params;
+ struct rte_cryptodev_info dev_info;
+
+ rte_cryptodev_info_get(ts_params->valid_devs[0], &dev_info);
+ if (!(dev_info.feature_flags & RTE_CRYPTODEV_FF_IN_PLACE_SGL)) {
+ printf("Device doesn't support in-place scatter-gather. Test
Skipped.\n");
+ return TEST_SKIPPED;
+ }
+
+ return test_tls_record_proto_all(&flags);
+}
+
#endif
static int
@@ -16956,6 +16975,10 @@ static struct unit_test_suite
tls12_record_proto_testsuite = {
"Combined test alg list",
ut_setup_security, ut_teardown,
test_tls_record_proto_display_list),
+ TEST_CASE_NAMED_ST(
+ "Multi-segmented mode",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_sgl),
TEST_CASES_END() /**< NULL terminate unit test array */
}
};
@@ -17053,6 +17076,10 @@ static struct unit_test_suite
dtls12_record_proto_testsuite = {
"Combined test alg list",
ut_setup_security, ut_teardown,
test_tls_record_proto_display_list),
+ TEST_CASE_NAMED_ST(
+ "Multi-segmented mode",
+ ut_setup_security, ut_teardown,
+ test_tls_record_proto_sgl),
TEST_CASES_END() /**< NULL terminate unit test array */
}
};
diff --git a/app/test/test_cryptodev_security_tls_record.h
b/app/test/test_cryptodev_security_tls_record.h
index 82a1776265..0bd83f88f0 100644
--- a/app/test/test_cryptodev_security_tls_record.h
+++ b/app/test/test_cryptodev_security_tls_record.h
@@ -55,6 +55,7 @@ struct tls_record_test_data {
struct tls_record_test_flags {
bool display_alg;
+ int nb_segs_in_mbuf;
};
extern struct tls_record_test_data tls_test_data_aes_128_gcm_v1;
diff --git a/doc/guides/rel_notes/release_24_03.rst
b/doc/guides/rel_notes/release_24_03.rst
index e9c9717706..0d226e9644 100644
--- a/doc/guides/rel_notes/release_24_03.rst
+++ b/doc/guides/rel_notes/release_24_03.rst
@@ -55,6 +55,10 @@ New Features
Also, make sure to start the actual text at the margin.
===
+* **Added TLS record test suite.**
+
+ Added a new test suite in cryptodev autotest framework to test TLS record
+ security offload by supported cryptodevs.
Removed Items
-
--
2.25.1
[PATCH v2 1/2] common/cnxk: support to dump debug info to file
This patch dumps contents of receviced packet descriptor from CQ
for debug to file
Signed-off-by: Rakesh Kudurumalla
---
V2: Resolve Warnings
drivers/common/cnxk/roc_cpt.h | 2 +-
drivers/common/cnxk/roc_cpt_debug.c | 56 +
drivers/common/cnxk/roc_nix.h | 2 +-
drivers/common/cnxk/roc_nix_debug.c | 3 +-
4 files changed, 36 insertions(+), 27 deletions(-)
diff --git a/drivers/common/cnxk/roc_cpt.h b/drivers/common/cnxk/roc_cpt.h
index 787bccb27d..523a5964a3 100644
--- a/drivers/common/cnxk/roc_cpt.h
+++ b/drivers/common/cnxk/roc_cpt.h
@@ -191,7 +191,7 @@ void __roc_api roc_cpt_iq_enable(struct roc_cpt_lf *lf);
int __roc_api roc_cpt_lmtline_init(struct roc_cpt *roc_cpt,
struct roc_cpt_lmtline *lmtline, int lf_id);
-void __roc_api roc_cpt_parse_hdr_dump(const struct cpt_parse_hdr_s *cpth);
+void __roc_api roc_cpt_parse_hdr_dump(FILE *file, const struct cpt_parse_hdr_s
*cpth);
int __roc_api roc_cpt_ctx_write(struct roc_cpt_lf *lf, void *sa_dptr,
void *sa_cptr, uint16_t sa_len);
diff --git a/drivers/common/cnxk/roc_cpt_debug.c
b/drivers/common/cnxk/roc_cpt_debug.c
index dce3638507..8e69b0a0e5 100644
--- a/drivers/common/cnxk/roc_cpt_debug.c
+++ b/drivers/common/cnxk/roc_cpt_debug.c
@@ -5,38 +5,48 @@
#include "roc_api.h"
#include "roc_priv.h"
+#define cpt_dump(file, fmt, ...) do {
\
+ if ((file) == NULL)
\
+ plt_dump(fmt, ##__VA_ARGS__);
\
+ else
\
+ fprintf(file, fmt "\n", ##__VA_ARGS__);
\
+} while (0)
+
void
-roc_cpt_parse_hdr_dump(const struct cpt_parse_hdr_s *cpth)
+roc_cpt_parse_hdr_dump(FILE *file, const struct cpt_parse_hdr_s *cpth)
{
struct cpt_frag_info_s *frag_info;
uint32_t offset;
uint64_t *slot;
- plt_print("CPT_PARSE \t0x%p:", cpth);
+ cpt_dump(file, "CPT_PARSE \t0x%p:", cpth);
/* W0 */
- plt_print("W0: cookie \t0x%x\t\tmatch_id \t0x%04x\t\terr_sum \t%u \t",
- cpth->w0.cookie, cpth->w0.match_id, cpth->w0.err_sum);
- plt_print("W0: reas_sts \t0x%x\t\tet_owr \t%u\t\tpkt_fmt \t%u \t",
+ cpt_dump(file, "W0: cookie \t0x%x\t\tmatch_id \t0x%04x \t",
+ cpth->w0.cookie, cpth->w0.match_id);
+ cpt_dump(file, "W0: err_sum \t%u \t", cpth->w0.err_sum);
+ cpt_dump(file, "W0: reas_sts \t0x%x\t\tet_owr \t%u\t\tpkt_fmt \t%u \t",
cpth->w0.reas_sts, cpth->w0.et_owr, cpth->w0.pkt_fmt);
- plt_print("W0: pad_len \t%u\t\tnum_frags \t%u\t\tpkt_out \t%u \t",
+ cpt_dump(file, "W0: pad_len \t%u\t\tnum_frags \t%u\t\tpkt_out \t%u \t",
cpth->w0.pad_len, cpth->w0.num_frags, cpth->w0.pkt_out);
/* W1 */
- plt_print("W1: wqe_ptr \t0x%016lx\t", plt_be_to_cpu_64(cpth->wqe_ptr));
+ cpt_dump(file, "W1: wqe_ptr \t0x%016lx\t",
+ plt_be_to_cpu_64(cpth->wqe_ptr));
/* W2 */
- plt_print("W2: frag_age \t0x%x\t\torig_pf_func \t0x%04x",
+ cpt_dump(file, "W2: frag_age \t0x%x\t\torig_pf_func \t0x%04x",
cpth->w2.frag_age, cpth->w2.orig_pf_func);
- plt_print("W2: il3_off \t0x%x\t\tfi_pad \t0x%x\t\tfi_offset \t0x%x \t",
- cpth->w2.il3_off, cpth->w2.fi_pad, cpth->w2.fi_offset);
+ cpt_dump(file, "W2: il3_off \t0x%x\t\tfi_pad \t0x%x \t",
+ cpth->w2.il3_off, cpth->w2.fi_pad);
+ cpt_dump(file, "W2: fi_offset \t0x%x \t", cpth->w2.fi_offset);
/* W3 */
- plt_print("W3: hw_ccode \t0x%x\t\tuc_ccode \t0x%x\t\tspi \t0x%08x",
+ cpt_dump(file, "W3: hw_ccode \t0x%x\t\tuc_ccode \t0x%x\t\tspi \t0x%08x",
cpth->w3.hw_ccode, cpth->w3.uc_ccode, cpth->w3.spi);
/* W4 */
- plt_print("W4: esn \t%" PRIx64 " \t OR frag1_wqe_ptr \t0x%" PRIx64,
+ cpt_dump(file, "W4: esn \t%" PRIx64 " \t OR frag1_wqe_ptr \t0x%" PRIx64,
cpth->esn, plt_be_to_cpu_64(cpth->frag1_wqe_ptr));
/* offset of 0 implies 256B, otherwise it implies offset*8B */
@@ -44,24 +54,24 @@ roc_cpt_parse_hdr_dump(const struct cpt_parse_hdr_s *cpth)
offset = (((offset - 1) & 0x1f) + 1) * 8;
frag_info = PLT_PTR_ADD(cpth, offset);
- plt_print("CPT Fraginfo \t0x%p:", frag_info);
+ cpt_dump(file, "CPT Fraginfo \t0x%p:", frag_info);
/* W0 */
- plt_print("W0: f0.info \t0x%x", frag_info->w0.f0.info);
- plt_print("W0: f1.info \t0x%x", frag_info->w0.f1.info);
- plt_print("W0: f2.info \t0x%x", frag_info->w0.f2.info);
- plt_print("W0: f3.info \t0x%x", frag_info->w0.f3.info);
+ cpt_dump(file, "W0: f0.info \t0x%x", frag_info->w0.f0.info);
+ cpt_dump(file, "W0: f1.info \t0x%x", frag_info->w0.f1.info
[PATCH v2 2/2] net/cnxk: dump Rx descriptor info to file
Add support for eth_rx_descriptor_dump for cn9k and cn10k.
This patch dumps contents of receviced packet descriptor from CQ
for debug to file
Signed-off-by: Rakesh Kudurumalla
---
drivers/net/cnxk/cn10k_ethdev.c | 67 +
drivers/net/cnxk/cn9k_ethdev.c | 53 ++
2 files changed, 120 insertions(+)
diff --git a/drivers/net/cnxk/cn10k_ethdev.c b/drivers/net/cnxk/cn10k_ethdev.c
index 4a4e97287c..a2e943a3d0 100644
--- a/drivers/net/cnxk/cn10k_ethdev.c
+++ b/drivers/net/cnxk/cn10k_ethdev.c
@@ -656,6 +656,72 @@ cn10k_nix_reassembly_conf_set(struct rte_eth_dev *eth_dev,
return rc;
}
+static int
+cn10k_nix_rx_avail_get(struct cn10k_eth_rxq *rxq)
+{
+ uint32_t qmask = rxq->qmask;
+ uint64_t reg, head, tail;
+ int available;
+
+ /* Use LDADDA version to avoid reorder */
+ reg = roc_atomic64_add_sync(rxq->wdata, rxq->cq_status);
+ /* CQ_OP_STATUS operation error */
+ if (reg & BIT_ULL(NIX_CQ_OP_STAT_OP_ERR) ||
+ reg & BIT_ULL(NIX_CQ_OP_STAT_CQ_ERR))
+ return 0;
+ tail = reg & 0xF;
+ head = (reg >> 20) & 0xF;
+ if (tail < head)
+ available = tail - head + qmask + 1;
+ else
+ available = tail - head;
+
+ return available;
+}
+
+static int
+cn10k_rx_descriptor_dump(const struct rte_eth_dev *eth_dev, uint16_t qid,
+uint16_t offset, uint16_t num, FILE *file)
+{
+ struct cn10k_eth_rxq *rxq = eth_dev->data->rx_queues[qid];
+ const uint64_t data_off = rxq->data_off;
+ const uint32_t qmask = rxq->qmask;
+ const uintptr_t desc = rxq->desc;
+ struct cpt_parse_hdr_s *cpth;
+ uint32_t head = rxq->head;
+ struct nix_cqe_hdr_s *cq;
+ uint16_t count = 0;
+ int available_pkts;
+ uint64_t cq_w1;
+
+ available_pkts = cn10k_nix_rx_avail_get(rxq);
+
+ if ((offset + num - 1) >= available_pkts) {
+ plt_err("Invalid BD num=%u\n", num);
+ return -EINVAL;
+ }
+
+ while (count < num) {
+ cq = (struct nix_cqe_hdr_s *)(desc + CQE_SZ(head) +
+ count + offset);
+ cq_w1 = *((const uint64_t *)cq + 1);
+ if (cq_w1 & BIT(11)) {
+ rte_iova_t buff = *((rte_iova_t *)((uint64_t *)cq + 9));
+ struct rte_mbuf *mbuf =
+ (struct rte_mbuf *)(buff - data_off);
+ cpth = (struct cpt_parse_hdr_s *)
+ ((uintptr_t)mbuf + (uint16_t)data_off);
+ roc_cpt_parse_hdr_dump(file, cpth);
+ } else {
+ roc_nix_cqe_dump(file, cq);
+ }
+
+ count++;
+ head &= qmask;
+ }
+ return 0;
+}
+
static int
cn10k_nix_tm_mark_vlan_dei(struct rte_eth_dev *eth_dev, int mark_green,
int mark_yellow, int mark_red,
@@ -794,6 +860,7 @@ nix_eth_dev_ops_override(void)
cn10k_nix_reassembly_capability_get;
cnxk_eth_dev_ops.ip_reassembly_conf_get = cn10k_nix_reassembly_conf_get;
cnxk_eth_dev_ops.ip_reassembly_conf_set = cn10k_nix_reassembly_conf_set;
+ cnxk_eth_dev_ops.eth_rx_descriptor_dump = cn10k_rx_descriptor_dump;
}
/* Update platform specific tm ops */
diff --git a/drivers/net/cnxk/cn9k_ethdev.c b/drivers/net/cnxk/cn9k_ethdev.c
index bae4dda5e2..67f21a9c7f 100644
--- a/drivers/net/cnxk/cn9k_ethdev.c
+++ b/drivers/net/cnxk/cn9k_ethdev.c
@@ -664,6 +664,58 @@ cn9k_nix_tm_mark_ip_dscp(struct rte_eth_dev *eth_dev, int
mark_green,
return rc;
}
+static int
+cn9k_nix_rx_avail_get(struct cn9k_eth_rxq *rxq)
+{
+ uint32_t qmask = rxq->qmask;
+ uint64_t reg, head, tail;
+ int available;
+
+ /* Use LDADDA version to avoid reorder */
+ reg = roc_atomic64_add_sync(rxq->wdata, rxq->cq_status);
+ /* CQ_OP_STATUS operation error */
+ if (reg & BIT_ULL(NIX_CQ_OP_STAT_OP_ERR) ||
+ reg & BIT_ULL(NIX_CQ_OP_STAT_CQ_ERR))
+ return 0;
+ tail = reg & 0xF;
+ head = (reg >> 20) & 0xF;
+ if (tail < head)
+ available = tail - head + qmask + 1;
+ else
+ available = tail - head;
+
+ return available;
+}
+
+static int
+cn9k_rx_descriptor_dump(const struct rte_eth_dev *eth_dev, uint16_t qid,
+uint16_t offset, uint16_t num, FILE *file)
+{
+ struct cn9k_eth_rxq *rxq = eth_dev->data->rx_queues[qid];
+ const uint32_t qmask = rxq->qmask;
+ const uintptr_t desc = rxq->desc;
+ uint32_t head = rxq->head;
+ struct nix_cqe_hdr_s *cq;
+ uint16_t count = 0;
+ int available_pkts;
+
+ available_pkts = cn9k_nix_rx_avail_get(rxq);
+
+ if ((offset + num - 1) >= available_pkts) {
+ plt_err("
Re: [PATCH v2 4/4] net/hns3: use stdatomic API
On 12/7/2023 1:42 AM, Jie Hai wrote: > Replace the use of gcc builtin __atomic_xxx intrinsics with > corresponding rte_atomic_xxx optional stdatomic API. > > Signed-off-by: Jie Hai > <...> > static void > @@ -643,8 +643,8 @@ hns3vf_get_push_lsc_cap(struct hns3_hw *hw) > uint16_t exp = HNS3_PF_PUSH_LSC_CAP_UNKNOWN; > struct hns3_vf *vf = HNS3_DEV_HW_TO_VF(hw); > > - __atomic_store_n(&vf->pf_push_lsc_cap, HNS3_PF_PUSH_LSC_CAP_UNKNOWN, > - __ATOMIC_RELEASE); > + rte_atomic_load_explicit(&vf->pf_push_lsc_cap, > + HNS3_PF_PUSH_LSC_CAP_UNKNOWN, rte_memory_order_release); > I guess intension here is to use 'rte_atomic_store_explicit()', current code cause a build error [1]. [1] https://mails.dpdk.org/archives/test-report/2023-December/525759.html
Re: [PATCH v2 4/4] net/hns3: use stdatomic API
On 12/7/2023 1:42 AM, Jie Hai wrote:
> Replace the use of gcc builtin __atomic_xxx intrinsics with
> corresponding rte_atomic_xxx optional stdatomic API.
>
> Signed-off-by: Jie Hai
>
<...>
> diff --git a/drivers/net/hns3/hns3_ethdev.c b/drivers/net/hns3/hns3_ethdev.c
> index b8f7e408d1e0..457247609811 100644
> --- a/drivers/net/hns3/hns3_ethdev.c
> +++ b/drivers/net/hns3/hns3_ethdev.c
> @@ -99,7 +99,7 @@ static const struct rte_eth_fec_capa speed_fec_capa_tbl[] =
> {
> };
>
> static enum hns3_reset_level hns3_get_reset_level(struct hns3_adapter *hns,
> - uint64_t *levels);
> + RTE_ATOMIC(uint64_t *)levels);
>
I don't know if is there a defined syntax, but I think better to use
macro for pointers as "RTE_ATOMIC(uint64_t) *levels", because of same
reason to use "int *a", following usage can be confusing:
"RTE_ATOMIC(uint64_t *) level1, level2;"
Re: [PATCH v2] app/test-pmd: fix tcp/udp cksum with padding data
On 12/7/2023 8:53 AM, Kaiwen Deng wrote:
> IEEE 802 packets may have a minimum size limit. The data fields
> should be padded when necessary. In some cases, the padding data
> is not zero. Testpmd does not trim these IP packets to the true
> length of the frame, so errors will occur when calculating TCP
> or UDP checksum.
>
> This commit fixes this issue by triming IP packets to the true
> length of the frame in testpmd.
>
> Fixes: 03d17e4d0179 ("app/testpmd: do not change IP addrs in checksum engine")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Kaiwen Deng
>
I think decision was to fix 'rte_ipv4_udptcp_cksum_mbuf()' &
'rte_ipv6_udptcp_cksum_mbuf()'.
[PATCH v3 0/8] add new command line argument parsing library
A recent thread on the mailing list[1] discussed corelist and coremask parsing and the idea of a new library dedicated to command line parsing was mentioned[2]. This patchset adds the library, along with the new APIs, and edits the existing EAL, DLB2 driver and some example application functions to use these APIs, rather than each implementing their own copies. The new APIs work similar to the existing functions in EAL, however instead of filling a core array like this: [1, -1, -1, 2, 3] (a non -1 refers to an 'active core' at that index) It fills it like this: [0, 3, 4] (with the value at each index being an 'active core'). The new APIs will also return the number of cores contained in the passed corelist/coremask, so in the above example, 3 would be returned. New in the v3, is a heuristic parser which searches for key markers in the core string, returning a enum value based off this search to indicate if a parameter is likely a coremask or a corelist. This heuristic function is also wrapped in a parser function allowing apps to handle both coremasks and corelists simultaneously. [1] https://mails.dpdk.org/archives/dev/2023-November/280957.html [2] https://mails.dpdk.org/archives/dev/2023-November/280966.html v3: * new 'combined core string parser' and 'heuristic parser' * changes to eventdev_pipeline and l3fwd-power example applications * various struct optimisations in arg_parser.c * fix for windows build relating to RTE_SWAP() * minor changes for issues flagged during review v2: * changes to EAL service core related parsers to call API. * various optimisations in core_bit related functions in arg_parser.c. * add lib to list for windows build. * minor changes for issues flagged during review. Euan Bourke (8): arg_parser: new library for command line parsing arg_parser: add new coremask parsing API eal: add support for new arg parsing library eal: update to service core related parsers event/dlb2: add new arg parsing library API support arg_parser: added common core string and heuristic parsers examples/eventdev_pipeline: update to call arg parser API examples/l3fwd-power: update to call arg parser API .mailmap | 1 + MAINTAINERS | 4 + doc/api/doxy-api-index.md| 3 +- doc/api/doxy-api.conf.in | 1 + drivers/event/dlb2/dlb2_priv.h | 4 +- drivers/event/dlb2/pf/base/dlb2_resource.c | 51 ++-- examples/eventdev_pipeline/main.c| 65 + examples/eventdev_pipeline/pipeline_common.h | 1 + examples/l3fwd-power/perf_core.c | 51 +--- lib/arg_parser/arg_parser.c | 229 +++ lib/arg_parser/meson.build | 7 + lib/arg_parser/rte_arg_parser.h | 160 +++ lib/arg_parser/version.map | 13 + lib/eal/common/eal_common_options.c | 285 --- lib/eal/meson.build | 2 +- lib/meson.build | 2 + 16 files changed, 513 insertions(+), 366 deletions(-) create mode 100644 lib/arg_parser/arg_parser.c create mode 100644 lib/arg_parser/meson.build create mode 100644 lib/arg_parser/rte_arg_parser.h create mode 100644 lib/arg_parser/version.map -- 2.34.1
[PATCH v3 1/8] arg_parser: new library for command line parsing
Add a new library to make it easier for eal and other libraries to parse
command line arguments.
The first function in this library is one to parse a corelist string
into an array of individual core ids. The function will then return the
total number of cores described in the corelist.
Signed-off-by: Euan Bourke
---
.mailmap| 1 +
MAINTAINERS | 4 ++
doc/api/doxy-api-index.md | 3 +-
doc/api/doxy-api.conf.in| 1 +
lib/arg_parser/arg_parser.c | 108
lib/arg_parser/meson.build | 7 +++
lib/arg_parser/rte_arg_parser.h | 66 +++
lib/arg_parser/version.map | 10 +++
lib/meson.build | 2 +
9 files changed, 201 insertions(+), 1 deletion(-)
create mode 100644 lib/arg_parser/arg_parser.c
create mode 100644 lib/arg_parser/meson.build
create mode 100644 lib/arg_parser/rte_arg_parser.h
create mode 100644 lib/arg_parser/version.map
diff --git a/.mailmap b/.mailmap
index ab0742a382..528bc68a30 100644
--- a/.mailmap
+++ b/.mailmap
@@ -379,6 +379,7 @@ Eric Zhang
Erik Gabriel Carrillo
Erik Ziegenbalg
Erlu Chen
+Euan Bourke
Eugenio Pérez
Eugeny Parshutin
Evan Swanson
diff --git a/MAINTAINERS b/MAINTAINERS
index 0d1c8126e3..68ef5ba14b 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1756,6 +1756,10 @@ M: Nithin Dabilpuram
M: Pavan Nikhilesh
F: lib/node/
+Argument parsing
+M: Bruce Richardson
+F: lib/arg_parser/
+
Test Applications
-
diff --git a/doc/api/doxy-api-index.md b/doc/api/doxy-api-index.md
index a6a768bd7c..f711010140 100644
--- a/doc/api/doxy-api-index.md
+++ b/doc/api/doxy-api-index.md
@@ -221,7 +221,8 @@ The public API headers are grouped by topics:
[config file](@ref rte_cfgfile.h),
[key/value args](@ref rte_kvargs.h),
[string](@ref rte_string_fns.h),
- [thread](@ref rte_thread.h)
+ [thread](@ref rte_thread.h),
+ [argument parsing](@ref rte_arg_parser.h)
- **debug**:
[jobstats](@ref rte_jobstats.h),
diff --git a/doc/api/doxy-api.conf.in b/doc/api/doxy-api.conf.in
index e94c9e4e46..05718ba6ed 100644
--- a/doc/api/doxy-api.conf.in
+++ b/doc/api/doxy-api.conf.in
@@ -28,6 +28,7 @@ INPUT = @TOPDIR@/doc/api/doxy-api-index.md \
@TOPDIR@/lib/eal/include \
@TOPDIR@/lib/eal/include/generic \
@TOPDIR@/lib/acl \
+ @TOPDIR@/lib/arg_parser \
@TOPDIR@/lib/bbdev \
@TOPDIR@/lib/bitratestats \
@TOPDIR@/lib/bpf \
diff --git a/lib/arg_parser/arg_parser.c b/lib/arg_parser/arg_parser.c
new file mode 100644
index 00..240f63d8e1
--- /dev/null
+++ b/lib/arg_parser/arg_parser.c
@@ -0,0 +1,108 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2023 Intel Corporation
+ */
+
+#include "errno.h"
+#include "stdlib.h"
+#include "ctype.h"
+#include "string.h"
+#include "stdbool.h"
+
+#include
+#include
+
+
+struct core_bits {
+ uint8_t bits[(UINT16_MAX + 1) / CHAR_BIT];
+ uint16_t max_bit_set;
+ uint16_t min_bit_set;
+ uint32_t total_bits_set;
+};
+
+static inline bool
+get_core_bit(struct core_bits *mask, uint16_t idx)
+{
+ return !!(mask->bits[idx / CHAR_BIT] & (1 << (idx % CHAR_BIT)));
+}
+
+static inline void
+set_core_bit(struct core_bits *mask, uint16_t idx)
+{
+ if (get_core_bit(mask, idx))
+ return;
+
+ mask->bits[idx / CHAR_BIT] |= 1 << (idx % CHAR_BIT);
+ /* Update min and max bit if its first time setting a bit */
+ if (++(mask->total_bits_set) == 1) {
+ mask->min_bit_set = idx;
+ mask->max_bit_set = idx;
+ return;
+ }
+
+ if (idx > mask->max_bit_set)
+ mask->max_bit_set = idx;
+
+ if (idx < mask->min_bit_set)
+ mask->min_bit_set = idx;
+}
+
+static inline uint32_t
+corebits_to_array(struct core_bits *mask, uint16_t *cores, size_t max_cores)
+{
+ uint32_t count = 0;
+ for (uint32_t i = mask->min_bit_set; i <= mask->max_bit_set && count <
max_cores; i++) {
+ if (get_core_bit(mask, i))
+ cores[count++] = i;
+ }
+ return mask->total_bits_set;
+}
+
+
+int
+rte_arg_parse_corelist(const char *corelist, uint16_t *cores, uint32_t
cores_len)
+{
+ struct core_bits mask = {0};
+ int32_t min = -1;
+ char *end = NULL;
+
+ min = -1;
+ do {
+ int64_t idx;
+ int32_t max;
+
+ while (isblank(*corelist))
+ corelist++;
+ if (!isdigit(*corelist))
+ return -1;
+
+ errno = 0;
+ idx = strtol(corelist, &end, 10);
+ if (errno || end == NULL || idx > UINT16_MAX)
+ return -1;
+ while (isblank(*end))
+
[PATCH v3 2/8] arg_parser: add new coremask parsing API
Add new coremask parsing API. This API behaves similarly
to the corelist parsing API, taking a coremask string, a cores
array and a cores_len int. Parsing the coremask string, filling
its values into the cores array up to cores_len.
The API also returns a 'count' which corresponds to the total number
of cores in the coremask string.
Signed-off-by: Euan Bourke
---
lib/arg_parser/arg_parser.c | 53 +
lib/arg_parser/rte_arg_parser.h | 34 +
lib/arg_parser/version.map | 1 +
3 files changed, 88 insertions(+)
diff --git a/lib/arg_parser/arg_parser.c b/lib/arg_parser/arg_parser.c
index 240f63d8e1..cebab9e2f8 100644
--- a/lib/arg_parser/arg_parser.c
+++ b/lib/arg_parser/arg_parser.c
@@ -11,6 +11,9 @@
#include
#include
+#define BITS_PER_HEX 4
+#define MAX_COREMASK_SIZE ((UINT16_MAX + 1) / BITS_PER_HEX)
+
struct core_bits {
uint8_t bits[(UINT16_MAX + 1) / CHAR_BIT];
@@ -57,6 +60,15 @@ corebits_to_array(struct core_bits *mask, uint16_t *cores,
size_t max_cores)
return mask->total_bits_set;
}
+static int xdigit2val(unsigned char c)
+{
+ if (isdigit(c))
+ return c - '0';
+ else if (isupper(c))
+ return c - 'A' + 10;
+ else
+ return c - 'a' + 10;
+}
int
rte_arg_parse_corelist(const char *corelist, uint16_t *cores, uint32_t
cores_len)
@@ -106,3 +118,44 @@ rte_arg_parse_corelist(const char *corelist, uint16_t
*cores, uint32_t cores_len
return total_count;
}
+
+int
+rte_arg_parse_coremask(const char *coremask, uint16_t *cores, uint32_t
cores_len)
+{
+ struct core_bits mask = {0};
+
+ /* Remove all blank characters ahead and after .
+* Remove 0x/0X if exists.
+*/
+ while (isblank(*coremask))
+ coremask++;
+ if (coremask[0] == '0' && ((coremask[1] == 'x') || (coremask[1] ==
'X')))
+ coremask += 2;
+
+ int32_t i = strlen(coremask);
+ while ((i > 0) && isblank(coremask[i - 1]))
+ i--;
+ if (i == 0 || i > MAX_COREMASK_SIZE)
+ return -1;
+
+ uint32_t idx = 0;
+
+ for (i = i - 1; i >= 0; i--) {
+ int val;
+ char c = coremask[i];
+
+ if (isxdigit(c) == 0)
+ return -1;
+
+ val = xdigit2val(c);
+
+ for (uint8_t j = 0; j < BITS_PER_HEX; j++, idx++) {
+ if ((1 << j) & val)
+ set_core_bit(&mask, idx);
+ }
+ }
+
+ uint32_t total_count = corebits_to_array(&mask, cores, cores_len);
+
+ return total_count;
+}
diff --git a/lib/arg_parser/rte_arg_parser.h b/lib/arg_parser/rte_arg_parser.h
index 80579f8cf3..359d40e305 100644
--- a/lib/arg_parser/rte_arg_parser.h
+++ b/lib/arg_parser/rte_arg_parser.h
@@ -58,6 +58,40 @@ __rte_experimental
int
rte_arg_parse_corelist(const char *corelist, uint16_t *cores, uint32_t
cores_len);
+/**
+ * Convert a string describing a bitmask of core ids into an array of core ids.
+ *
+ * On success, the passed array is filled with the core ids present in the
+ * bitmask up to the "cores_len", and the number of unique cores present in the
+ * "coremask" is returned.
+ * For example, passing a 0xA "coremask" results in an array of [1, 3]
+ * and would return 2.
+ *
+ * NOTE: if the length of the input array is insufficient to hold the number
of core ids
+ * in "coremask" the input array is filled to capacity but the return value is
the
+ * number of elements which would have been written to the array, had enough
space been
+ * available. [This is similar to the behaviour of the snprintf function].
Because of
+ * this, the number of core values in the "coremask" may be determined by
calling the
+ * function with a NULL array pointer and array length given as 0.
+ *
+ * @param coremask
+ * A string containing a bitmask of core ids.
+ * @param cores
+ * An array where to store the core ids.
+ * Array can be NULL if "cores_len" is 0.
+ * @param cores_len
+ * The length of the "cores" array.
+ * If the size is smaller than that needed to hold all cores from "coremask",
+ * only "cores_len" elements will be written to the array.
+ * @return
+ * n: the number of unique cores present in "coremask".
+ * -1 if the string was invalid.
+ * NOTE: if n > "cores_len", then only "cores_len" elements in the "cores"
array are valid.
+ */
+__rte_experimental
+int
+rte_arg_parse_coremask(const char *coremask, uint16_t *cores, uint32_t
cores_len);
+
#ifdef __cplusplus
}
diff --git a/lib/arg_parser/version.map b/lib/arg_parser/version.map
index b0caaac569..b44d4b02b7 100644
--- a/lib/arg_parser/version.map
+++ b/lib/arg_parser/version.map
@@ -7,4 +7,5 @@ EXPERIMENTAL {
# added in 24.03
rte_arg_parse_corelist;
+ rte_arg_parse_coremask;
};
--
2.34.1
[PATCH v3 3/8] eal: add support for new arg parsing library
Update to eal functions relating to corelist and coremask
parsing to support the new arg parsing library. Functions
now call the API instead of implementing their own version.
Signed-off-by: Euan Bourke
---
lib/eal/common/eal_common_options.c | 114
lib/eal/meson.build | 2 +-
2 files changed, 14 insertions(+), 102 deletions(-)
diff --git a/lib/eal/common/eal_common_options.c
b/lib/eal/common/eal_common_options.c
index a6d21f1cba..60ba12a368 100644
--- a/lib/eal/common/eal_common_options.c
+++ b/lib/eal/common/eal_common_options.c
@@ -30,6 +30,7 @@
#include
#include
#include
+#include
#include
#ifndef RTE_EXEC_ENV_WINDOWS
#include
@@ -706,7 +707,7 @@ update_lcore_config(int *cores)
}
static int
-check_core_list(int *lcores, unsigned int count)
+check_core_list(uint16_t *lcores, unsigned int count)
{
char lcorestr[RTE_MAX_LCORE * 10];
bool overflow = false;
@@ -746,60 +747,18 @@ check_core_list(int *lcores, unsigned int count)
int
rte_eal_parse_coremask(const char *coremask, int *cores)
{
- const char *coremask_orig = coremask;
- int lcores[RTE_MAX_LCORE];
- unsigned int count = 0;
- int i, j, idx;
- int val;
- char c;
+ int count;
+ uint16_t lcores[RTE_MAX_LCORE];
+ int idx;
for (idx = 0; idx < RTE_MAX_LCORE; idx++)
cores[idx] = -1;
- idx = 0;
- /* Remove all blank characters ahead and after .
-* Remove 0x/0X if exists.
-*/
- while (isblank(*coremask))
- coremask++;
- if (coremask[0] == '0' && ((coremask[1] == 'x')
- || (coremask[1] == 'X')))
- coremask += 2;
- i = strlen(coremask);
- while ((i > 0) && isblank(coremask[i - 1]))
- i--;
- if (i == 0) {
- RTE_LOG(ERR, EAL, "No lcores in coremask: [%s]\n",
- coremask_orig);
- return -1;
- }
+ /* Call public coremask parsing API */
+ count = rte_arg_parse_coremask(coremask, lcores, RTE_MAX_LCORE);
- for (i = i - 1; i >= 0; i--) {
- c = coremask[i];
- if (isxdigit(c) == 0) {
- /* invalid characters */
- RTE_LOG(ERR, EAL, "invalid characters in coremask:
[%s]\n",
- coremask_orig);
- return -1;
- }
- val = xdigit2val(c);
- for (j = 0; j < BITS_PER_HEX; j++, idx++)
- {
- if ((1 << j) & val) {
- if (count >= RTE_MAX_LCORE) {
- RTE_LOG(ERR, EAL, "Too many lcores
provided. Cannot exceed RTE_MAX_LCORE (%d)\n",
- RTE_MAX_LCORE);
- return -1;
- }
- lcores[count++] = idx;
- }
- }
- }
- if (count == 0) {
- RTE_LOG(ERR, EAL, "No lcores in coremask: [%s]\n",
- coremask_orig);
+ if (count <= 0 || count > RTE_MAX_LCORE)
return -1;
- }
if (check_core_list(lcores, count))
return -1;
@@ -898,64 +857,17 @@ eal_parse_service_corelist(const char *corelist)
static int
eal_parse_corelist(const char *corelist, int *cores)
{
- unsigned int count = 0, i;
- int lcores[RTE_MAX_LCORE];
- char *end = NULL;
- int min, max;
+ int count;
+ uint16_t lcores[RTE_MAX_LCORE];
int idx;
for (idx = 0; idx < RTE_MAX_LCORE; idx++)
cores[idx] = -1;
- /* Remove all blank characters ahead */
- while (isblank(*corelist))
- corelist++;
+ /* Call public corelist parsing API */
+ count = rte_arg_parse_corelist(corelist, lcores, RTE_MAX_LCORE);
- /* Get list of cores */
- min = -1;
- do {
- while (isblank(*corelist))
- corelist++;
- if (*corelist == '\0')
- return -1;
- errno = 0;
- idx = strtol(corelist, &end, 10);
- if (errno || end == NULL)
- return -1;
- if (idx < 0)
- return -1;
- while (isblank(*end))
- end++;
- if (*end == '-') {
- min = idx;
- } else if ((*end == ',') || (*end == '\0')) {
- max = idx;
- if (min == -1)
- min = idx;
- for (idx = min; idx <= max; idx++) {
- bool dup = false;
-
- /* Check if this idx is already present */
- for (
[PATCH v3 4/8] eal: update to service core related parsers
Updates to the parse service cores functions in EAL to call the arg
parser API instead of implementing its own versions.
Signed-off-by: Euan Bourke
---
lib/eal/common/eal_common_options.c | 171 +++-
1 file changed, 41 insertions(+), 130 deletions(-)
diff --git a/lib/eal/common/eal_common_options.c
b/lib/eal/common/eal_common_options.c
index 60ba12a368..d44654f621 100644
--- a/lib/eal/common/eal_common_options.c
+++ b/lib/eal/common/eal_common_options.c
@@ -573,97 +573,49 @@ eal_plugins_init(void)
* the global configuration (core role and core count) with the parsed
* value.
*/
-static int xdigit2val(unsigned char c)
-{
- int val;
-
- if (isdigit(c))
- val = c - '0';
- else if (isupper(c))
- val = c - 'A' + 10;
- else
- val = c - 'a' + 10;
- return val;
-}
-
static int
eal_parse_service_coremask(const char *coremask)
{
struct rte_config *cfg = rte_eal_get_configuration();
- int i, j, idx = 0;
- unsigned int count = 0;
- char c;
- int val;
+ uint16_t cores[RTE_MAX_LCORE];
+ int64_t count;
uint32_t taken_lcore_count = 0;
- if (coremask == NULL)
- return -1;
- /* Remove all blank characters ahead and after .
-* Remove 0x/0X if exists.
-*/
- while (isblank(*coremask))
- coremask++;
- if (coremask[0] == '0' && ((coremask[1] == 'x')
- || (coremask[1] == 'X')))
- coremask += 2;
- i = strlen(coremask);
- while ((i > 0) && isblank(coremask[i - 1]))
- i--;
-
- if (i == 0)
+ count = rte_arg_parse_coremask(coremask, cores, RTE_DIM(cores));
+
+ if (count == 0 || count == -1)
return -1;
- for (i = i - 1; i >= 0 && idx < RTE_MAX_LCORE; i--) {
- c = coremask[i];
- if (isxdigit(c) == 0) {
- /* invalid characters */
+ for (int i = 0; i < count; i++) {
+ uint32_t lcore = cores[i];
+ if (main_lcore_parsed &&
+ cfg->main_lcore == lcore) {
+ RTE_LOG(ERR, EAL,
+ "lcore %u is main lcore, cannot use as service
core\n",
+ cores[i]);
return -1;
}
- val = xdigit2val(c);
- for (j = 0; j < BITS_PER_HEX && idx < RTE_MAX_LCORE;
- j++, idx++) {
- if ((1 << j) & val) {
- /* handle main lcore already parsed */
- uint32_t lcore = idx;
- if (main_lcore_parsed &&
- cfg->main_lcore == lcore) {
- RTE_LOG(ERR, EAL,
- "lcore %u is main lcore, cannot
use as service core\n",
- idx);
- return -1;
- }
- if (eal_cpu_detected(idx) == 0) {
- RTE_LOG(ERR, EAL,
- "lcore %u unavailable\n", idx);
- return -1;
- }
-
- if (cfg->lcore_role[idx] == ROLE_RTE)
- taken_lcore_count++;
-
- lcore_config[idx].core_role = ROLE_SERVICE;
- count++;
- }
- }
- }
-
- for (; i >= 0; i--)
- if (coremask[i] != '0')
+ if (eal_cpu_detected(cores[i]) == 0) {
+ RTE_LOG(ERR, EAL,
+ "lcore %u unavailable\n", cores[i]);
return -1;
+ }
- for (; idx < RTE_MAX_LCORE; idx++)
- lcore_config[idx].core_index = -1;
-
- if (count == 0)
- return -1;
+ if (cfg->lcore_role[cores[i]] == ROLE_RTE)
+ taken_lcore_count++;
+ lcore_config[cores[i]].core_role = ROLE_SERVICE;
+ }
if (core_parsed && taken_lcore_count != count) {
RTE_LOG(WARNING, EAL,
- "Not all service cores are in the coremask. "
+ "Not all service cores were in the coremask. "
"Please ensure -c or -l includes service cores\n");
}
+ for (uint16_t j = count*4; j < RTE_MAX_LCORE; j++)
+ lcore_config[j].core_index = -1;
+
cfg->service_lcore_count = count;
return 0;
}
@@ -780,70 +732,29 @@ static int
eal_parse_service_corelist(const char *corelist)
{
struct rte_config
[PATCH v3 5/8] event/dlb2: add new arg parsing library API support
Switched the dlb2 driver to call the new arg parsing library instead of
eal for coremask parsing, and updated the resource probe function to
support the changed formatting of the API.
Signed-off-by: Euan Bourke
---
drivers/event/dlb2/dlb2_priv.h | 4 +-
drivers/event/dlb2/pf/base/dlb2_resource.c | 51 +-
2 files changed, 21 insertions(+), 34 deletions(-)
diff --git a/drivers/event/dlb2/dlb2_priv.h b/drivers/event/dlb2/dlb2_priv.h
index 31a3beeb6c..c14d83da5b 100644
--- a/drivers/event/dlb2/dlb2_priv.h
+++ b/drivers/event/dlb2/dlb2_priv.h
@@ -10,6 +10,7 @@
#include
#include
+#include
#include "dlb2_user.h"
#include "dlb2_log.h"
#include "rte_pmd_dlb2.h"
@@ -729,9 +730,6 @@ void dlb2_event_build_hcws(struct dlb2_port *qm_port,
uint8_t *sched_type,
uint8_t *queue_id);
-/* Extern functions */
-extern int rte_eal_parse_coremask(const char *coremask, int *cores);
-
/* Extern globals */
extern struct process_local_port_data dlb2_port[][DLB2_NUM_PORT_TYPES];
diff --git a/drivers/event/dlb2/pf/base/dlb2_resource.c
b/drivers/event/dlb2/pf/base/dlb2_resource.c
index 7ce3e3531c..b65de9350a 100644
--- a/drivers/event/dlb2/pf/base/dlb2_resource.c
+++ b/drivers/event/dlb2/pf/base/dlb2_resource.c
@@ -922,49 +922,38 @@ dlb2_resource_probe(struct dlb2_hw *hw, const void
*probe_args)
{
const struct dlb2_devargs *args = (const struct dlb2_devargs
*)probe_args;
const char *mask = args ? args->producer_coremask : NULL;
- int cpu = 0, cnt = 0, cores[RTE_MAX_LCORE], i;
+ int cpu = 0, i;
+ uint16_t cores[RTE_MAX_LCORE];
if (args) {
mask = (const char *)args->producer_coremask;
}
- if (mask && rte_eal_parse_coremask(mask, cores)) {
+ int ret = rte_arg_parse_coremask(mask, cores, RTE_DIM(cores));
+
+ if (mask && ret == -1) {
DLB2_LOG_ERR(": Invalid producer coremask=%s", mask);
return -1;
}
- hw->num_prod_cores = 0;
- for (i = 0; i < RTE_MAX_LCORE; i++) {
- bool is_pcore = (mask && cores[i] != -1);
-
- if (rte_lcore_is_enabled(i)) {
- if (is_pcore) {
- /*
-* Populate the producer cores from parsed
-* coremask
-*/
- hw->prod_core_list[cores[i]] = i;
- hw->num_prod_cores++;
-
- } else if ((++cnt == DLB2_EAL_PROBE_CORE ||
- rte_lcore_count() < DLB2_EAL_PROBE_CORE)) {
- /*
-* If no producer coremask is provided, use the
-* second EAL core to probe
-*/
- cpu = i;
- break;
- }
- } else if (is_pcore) {
+ hw->num_prod_cores = ret;
+ /* Check for no producer cores and then get the second EAL core */
+ if (hw->num_prod_cores > 0)
+ cpu = cores[0];
+ else if (rte_lcore_count() < DLB2_EAL_PROBE_CORE)
+ cpu = rte_get_main_lcore();
+ else
+ cpu = rte_get_next_lcore(-1, 1, 0);
+
+ /* check our producer list is valid and error out if not */
+ for (i = 0; i < hw->num_prod_cores; i++) {
+ if (!rte_lcore_is_enabled(cores[i])) {
DLB2_LOG_ERR("Producer coremask(%s) must be a subset of
EAL coremask",
-mask);
+ mask);
return -1;
- }
-
}
- /* Use the first core in producer coremask to probe */
- if (hw->num_prod_cores)
- cpu = hw->prod_core_list[0];
+ hw->prod_core_list[i] = cores[i];
+}
dlb2_get_pp_allocation(hw, cpu, DLB2_LDB_PORT);
dlb2_get_pp_allocation(hw, cpu, DLB2_DIR_PORT);
--
2.34.1
[PATCH v3 6/8] arg_parser: added common core string and heuristic parsers
Two new functions, the first is a 'heuristic parser' which examines a
string describing a set of cores and determines based off heuristics
whether its a coremask or a corelist.
Second is a 'combined parser' which calls the first function and then
based off the returned value will call the relevant core string parser.
This function also takes a 'default_type' int which corresponds to
which parser should be used in the case of an ambiguous string.
Signed-off-by: Euan Bourke
---
lib/arg_parser/arg_parser.c | 68 +
lib/arg_parser/rte_arg_parser.h | 60 +
lib/arg_parser/version.map | 2 +
3 files changed, 130 insertions(+)
diff --git a/lib/arg_parser/arg_parser.c b/lib/arg_parser/arg_parser.c
index cebab9e2f8..95cbc50c13 100644
--- a/lib/arg_parser/arg_parser.c
+++ b/lib/arg_parser/arg_parser.c
@@ -7,10 +7,15 @@
#include "ctype.h"
#include "string.h"
#include "stdbool.h"
+#include "stdio.h"
#include
#include
+#define RTE_ARG_PARSE_TYPE_COREMASK 0
+#define RTE_ARG_PARSE_TYPE_CORELIST 1
+#define RTE_ARG_PARSE_TYPE_UNKNOWN 2
+
#define BITS_PER_HEX 4
#define MAX_COREMASK_SIZE ((UINT16_MAX + 1) / BITS_PER_HEX)
@@ -22,6 +27,7 @@ struct core_bits {
uint32_t total_bits_set;
};
+
static inline bool
get_core_bit(struct core_bits *mask, uint16_t idx)
{
@@ -159,3 +165,65 @@ rte_arg_parse_coremask(const char *coremask, uint16_t
*cores, uint32_t cores_len
return total_count;
}
+
+int
+rte_arg_parse_arg_type(const char *core_string)
+{
+ /* Remove leading whitespace */
+ while (isblank(*core_string))
+ core_string++;
+
+ /* Check for 0x prefix */
+ if (core_string[0] == '0' && tolower(core_string[1]) == 'x') {
+ if (core_string[2] != '\0')
+ return RTE_ARG_PARSE_TYPE_COREMASK;
+ return -1;
+ }
+
+ int i = 0, idx = 0;
+ /* Check for ',' and '-' and check for A-F */
+ do {
+ while (isblank(core_string[idx]))
+ idx++;
+
+ if (core_string[idx] == ',' || core_string[idx] == '-')
+ return RTE_ARG_PARSE_TYPE_CORELIST;
+
+ if (isalpha(core_string[idx])) {
+ if (isxdigit(core_string[idx]))
+ return RTE_ARG_PARSE_TYPE_COREMASK;
+ return -1;
+ }
+ idx++;
+ i++;
+ } while (core_string[idx] != '\0');
+
+ /* Check length of core_string if ambiguous as max length of a uint16_t
is 5 digits
+* implying its a coremask.
+*/
+ if (i > 5)
+ return RTE_ARG_PARSE_TYPE_COREMASK;
+
+ return -1;
+}
+
+int
+rte_arg_parse_core_string(const char *core_string, uint16_t *cores, uint32_t
cores_len,
+ int default_type)
+{
+ if (default_type != RTE_ARG_PARSE_TYPE_COREMASK &&
+ default_type != RTE_ARG_PARSE_TYPE_CORELIST) {
+ return -1;
+ }
+ switch (rte_arg_parse_arg_type(core_string)) {
+ case RTE_ARG_PARSE_TYPE_COREMASK:
+ return rte_arg_parse_coremask(core_string, cores, cores_len);
+ case RTE_ARG_PARSE_TYPE_CORELIST:
+ return rte_arg_parse_corelist(core_string, cores, cores_len);
+ default:
+ return default_type == RTE_ARG_PARSE_TYPE_COREMASK ?
+ rte_arg_parse_coremask(core_string, cores, cores_len) :
+ rte_arg_parse_corelist(core_string, cores, cores_len);
+ return -1;
+ }
+}
diff --git a/lib/arg_parser/rte_arg_parser.h b/lib/arg_parser/rte_arg_parser.h
index 359d40e305..125ca9524c 100644
--- a/lib/arg_parser/rte_arg_parser.h
+++ b/lib/arg_parser/rte_arg_parser.h
@@ -92,6 +92,66 @@ __rte_experimental
int
rte_arg_parse_coremask(const char *coremask, uint16_t *cores, uint32_t
cores_len);
+/**
+ * Use heuristics to determine if a string contains a coremask or a corelist.
+ *
+ * This function will check a series of conditions and return an int
representing which
+ * core type (mask or list) the string represents or UNKNOWN if the string is
ambiguous.
+ *
+ * @param core_string
+ * A string describing the intended cores to be parsed
+ * @return
+ * int representing the core type
+ * -1: error.
+ * 0: coremask.
+ * 1: corelist.
+ * 2: unknown (ambiguous).
+ */
+__rte_experimental
+int
+rte_arg_parse_arg_type(const char *core_string);
+
+/**
+ * Convert a string describing either a corelist or coremask into an array of
core ids.
+ *
+ * This function will fill the "cores" array up to "cores_len" with the core
ids described
+ * in the "core_string". The string can either describe a corelist or a
coremask, and
+ * will be parsed accordingly. The number of unique core ids in the string is
then returned.
+ * For example:
+ * "1-4" is treated as a corelist and results in an arr
[PATCH v3 7/8] examples/eventdev_pipeline: update to call arg parser API
Update to the eventdev_pipeline example application to call the arg parser
library for its 'combined core string parser' instead of implementing its
own coremask parser. The default_type passed into the function call is
a coremask.
Signed-off-by: Euan Bourke
---
examples/eventdev_pipeline/main.c| 65 +++-
examples/eventdev_pipeline/pipeline_common.h | 1 +
2 files changed, 10 insertions(+), 56 deletions(-)
diff --git a/examples/eventdev_pipeline/main.c
b/examples/eventdev_pipeline/main.c
index 0c995d1a70..c59d01e7a5 100644
--- a/examples/eventdev_pipeline/main.c
+++ b/examples/eventdev_pipeline/main.c
@@ -56,69 +56,22 @@ core_in_use(unsigned int lcore_id) {
fdata->tx_core[lcore_id] || fdata->worker_core[lcore_id]);
}
-/*
- * Parse the coremask given as argument (hexadecimal string) and fill
- * the global configuration (core role and core count) with the parsed
- * value.
- */
-static int xdigit2val(unsigned char c)
-{
- int val;
-
- if (isdigit(c))
- val = c - '0';
- else if (isupper(c))
- val = c - 'A' + 10;
- else
- val = c - 'a' + 10;
- return val;
-}
-
static uint64_t
parse_coremask(const char *coremask)
{
- int i, j, idx = 0;
- unsigned int count = 0;
- char c;
- int val;
+ int count;
+ uint16_t i;
uint64_t mask = 0;
- const int32_t BITS_HEX = 4;
+ uint16_t cores[RTE_MAX_LCORE];
- if (coremask == NULL)
- return -1;
- /* Remove all blank characters ahead and after .
-* Remove 0x/0X if exists.
-*/
- while (isblank(*coremask))
- coremask++;
- if (coremask[0] == '0' && ((coremask[1] == 'x')
- || (coremask[1] == 'X')))
- coremask += 2;
- i = strlen(coremask);
- while ((i > 0) && isblank(coremask[i - 1]))
- i--;
- if (i == 0)
- return -1;
+ count = rte_arg_parse_core_string(coremask, cores, RTE_DIM(cores), 0);
- for (i = i - 1; i >= 0 && idx < MAX_NUM_CORE; i--) {
- c = coremask[i];
- if (isxdigit(c) == 0) {
- /* invalid characters */
- return -1;
- }
- val = xdigit2val(c);
- for (j = 0; j < BITS_HEX && idx < MAX_NUM_CORE; j++, idx++) {
- if ((1 << j) & val) {
- mask |= (1ULL << idx);
- count++;
- }
- }
- }
- for (; i >= 0; i--)
- if (coremask[i] != '0')
- return -1;
- if (count == 0)
+ if (count == 0 || count == -1)
return -1;
+
+ for (i = 0; i < count; i++)
+ mask |= 1ULL << cores[i];
+
return mask;
}
diff --git a/examples/eventdev_pipeline/pipeline_common.h
b/examples/eventdev_pipeline/pipeline_common.h
index 28b6ab85ff..2b97a29bfc 100644
--- a/examples/eventdev_pipeline/pipeline_common.h
+++ b/examples/eventdev_pipeline/pipeline_common.h
@@ -6,6 +6,7 @@
#include
+#include
#include
#include
#include
--
2.34.1
[PATCH v3 8/8] examples/l3fwd-power: update to call arg parser API
Update to the l3fwd-power example application to call the arg parser
library for its 'combined core string parser' instead of implementing its
own corelist parser. The default_type passed into the function call is
a corelist.
Signed-off-by: Euan Bourke
---
examples/l3fwd-power/perf_core.c | 51 +---
1 file changed, 8 insertions(+), 43 deletions(-)
diff --git a/examples/l3fwd-power/perf_core.c b/examples/l3fwd-power/perf_core.c
index 41ef6d0c9a..f8511e30b3 100644
--- a/examples/l3fwd-power/perf_core.c
+++ b/examples/l3fwd-power/perf_core.c
@@ -12,6 +12,7 @@
#include
#include
#include
+#include
#include "perf_core.h"
#include "main.h"
@@ -177,56 +178,20 @@ parse_perf_config(const char *q_arg)
int
parse_perf_core_list(const char *corelist)
{
- int i, idx = 0;
- unsigned int count = 0;
- char *end = NULL;
- int min, max;
+ int count;
+ uint16_t cores[RTE_MAX_LCORE];
if (corelist == NULL) {
printf("invalid core list\n");
return -1;
}
+ count = rte_arg_parse_core_string(corelist, cores, RTE_DIM(cores), 1);
- /* Remove all blank characters ahead and after */
- while (isblank(*corelist))
- corelist++;
- i = strlen(corelist);
- while ((i > 0) && isblank(corelist[i - 1]))
- i--;
+ for (int i = 0; i < count; i++)
+ hp_lcores[i] = cores[i];
- /* Get list of cores */
- min = RTE_MAX_LCORE;
- do {
- while (isblank(*corelist))
- corelist++;
- if (*corelist == '\0')
- return -1;
- errno = 0;
- idx = strtoul(corelist, &end, 10);
- if (errno || end == NULL)
- return -1;
- while (isblank(*end))
- end++;
- if (*end == '-') {
- min = idx;
- } else if ((*end == ',') || (*end == '\0')) {
- max = idx;
- if (min == RTE_MAX_LCORE)
- min = idx;
- for (idx = min; idx <= max; idx++) {
- hp_lcores[count] = idx;
- count++;
- }
- min = RTE_MAX_LCORE;
- } else {
- printf("invalid core list\n");
- return -1;
- }
- corelist = end + 1;
- } while (*end != '\0');
-
- if (count == 0) {
+ if (count == 0 || count == -1) {
printf("invalid core list\n");
return -1;
}
@@ -234,7 +199,7 @@ parse_perf_core_list(const char *corelist)
nb_hp_lcores = count;
printf("Configured %d high performance cores\n", nb_hp_lcores);
- for (i = 0; i < nb_hp_lcores; i++)
+ for (uint16_t i = 0; i < nb_hp_lcores; i++)
printf("\tHigh performance core %d %d\n",
i, hp_lcores[i]);
--
2.34.1
Re: [PATCH v3 1/8] arg_parser: new library for command line parsing
On Thu, Dec 07, 2023 at 04:18:11PM +, Euan Bourke wrote:
> Add a new library to make it easier for eal and other libraries to parse
> command line arguments.
>
> The first function in this library is one to parse a corelist string
> into an array of individual core ids. The function will then return the
> total number of cores described in the corelist.
>
> Signed-off-by: Euan Bourke
There are other patches outstanding on the list for argument parsing too
[1], and we need to reconcile what goes into what libraries with what names!
Probably they can all be merged into one, but we need to check.
[1] http://patches.dpdk.org/project/dpdk/list/?series=30439
Some minor comments inline below for any v4.
> ---
> .mailmap| 1 +
> MAINTAINERS | 4 ++
> doc/api/doxy-api-index.md | 3 +-
> doc/api/doxy-api.conf.in| 1 +
> lib/arg_parser/arg_parser.c | 108
> lib/arg_parser/meson.build | 7 +++
> lib/arg_parser/rte_arg_parser.h | 66 +++
> lib/arg_parser/version.map | 10 +++
> lib/meson.build | 2 +
> 9 files changed, 201 insertions(+), 1 deletion(-)
> create mode 100644 lib/arg_parser/arg_parser.c
> create mode 100644 lib/arg_parser/meson.build
> create mode 100644 lib/arg_parser/rte_arg_parser.h
> create mode 100644 lib/arg_parser/version.map
>
> diff --git a/.mailmap b/.mailmap
> index ab0742a382..528bc68a30 100644
> --- a/.mailmap
> +++ b/.mailmap
> @@ -379,6 +379,7 @@ Eric Zhang
> Erik Gabriel Carrillo
> Erik Ziegenbalg
> Erlu Chen
> +Euan Bourke
> Eugenio Pérez
> Eugeny Parshutin
> Evan Swanson
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 0d1c8126e3..68ef5ba14b 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -1756,6 +1756,10 @@ M: Nithin Dabilpuram
> M: Pavan Nikhilesh
> F: lib/node/
>
> +Argument parsing
> +M: Bruce Richardson
> +F: lib/arg_parser/
> +
>
> Test Applications
> -
> diff --git a/doc/api/doxy-api-index.md b/doc/api/doxy-api-index.md
> index a6a768bd7c..f711010140 100644
> --- a/doc/api/doxy-api-index.md
> +++ b/doc/api/doxy-api-index.md
> @@ -221,7 +221,8 @@ The public API headers are grouped by topics:
>[config file](@ref rte_cfgfile.h),
>[key/value args](@ref rte_kvargs.h),
>[string](@ref rte_string_fns.h),
> - [thread](@ref rte_thread.h)
> + [thread](@ref rte_thread.h),
> + [argument parsing](@ref rte_arg_parser.h)
Not sure what order, if any, these elements are in, but I think argument
parsing could well be further up the list. It also reduces the diff by one
line, since you don't need to append the comma on the end of rte_thread
line.
>
> - **debug**:
>[jobstats](@ref rte_jobstats.h),
> diff --git a/doc/api/doxy-api.conf.in b/doc/api/doxy-api.conf.in
> index e94c9e4e46..05718ba6ed 100644
> --- a/doc/api/doxy-api.conf.in
> +++ b/doc/api/doxy-api.conf.in
> @@ -28,6 +28,7 @@ INPUT =
> @TOPDIR@/doc/api/doxy-api-index.md \
>@TOPDIR@/lib/eal/include \
>@TOPDIR@/lib/eal/include/generic \
>@TOPDIR@/lib/acl \
> + @TOPDIR@/lib/arg_parser \
>@TOPDIR@/lib/bbdev \
>@TOPDIR@/lib/bitratestats \
>@TOPDIR@/lib/bpf \
> diff --git a/lib/arg_parser/arg_parser.c b/lib/arg_parser/arg_parser.c
> new file mode 100644
> index 00..240f63d8e1
> --- /dev/null
> +++ b/lib/arg_parser/arg_parser.c
> @@ -0,0 +1,108 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + * Copyright(c) 2023 Intel Corporation
> + */
> +
> +#include "errno.h"
> +#include "stdlib.h"
> +#include "ctype.h"
> +#include "string.h"
> +#include "stdbool.h"
> +
> +#include
> +#include
> +
> +
> +struct core_bits {
> + uint8_t bits[(UINT16_MAX + 1) / CHAR_BIT];
> + uint16_t max_bit_set;
> + uint16_t min_bit_set;
> + uint32_t total_bits_set;
> +};
> +
> +static inline bool
> +get_core_bit(struct core_bits *mask, uint16_t idx)
> +{
> + return !!(mask->bits[idx / CHAR_BIT] & (1 << (idx % CHAR_BIT)));
> +}
> +
> +static inline void
> +set_core_bit(struct core_bits *mask, uint16_t idx)
> +{
> + if (get_core_bit(mask, idx))
> + return;
> +
> + mask->bits[idx / CHAR_BIT] |= 1 << (idx % CHAR_BIT);
> + /* Update min and max bit if its first time setting a bit */
> + if (++(mask->total_bits_set) == 1) {
> + mask->min_bit_set = idx;
> + mask->max_bit_set = idx;
> + return;
> + }
> +
> + if (idx > mask->max_bit_set)
> + mask->max_bit_set = idx;
> +
> + if (idx < mask->min_bit_set)
> + mask->min_bit_set = idx;
> +}
> +
> +static inline uint32_t
> +corebits_to_array(struct core_bits *mask, uint16_t *cores, size_t max_cores)
> +{
> + uint32_t count = 0;
> + for (uint32_t i = mask->min_bit_set
Re: [PATCH v3 6/8] arg_parser: added common core string and heuristic parsers
On Thu, Dec 07, 2023 at 04:18:16PM +, Euan Bourke wrote:
> Two new functions, the first is a 'heuristic parser' which examines a
> string describing a set of cores and determines based off heuristics
> whether its a coremask or a corelist.
>
> Second is a 'combined parser' which calls the first function and then
> based off the returned value will call the relevant core string parser.
> This function also takes a 'default_type' int which corresponds to
> which parser should be used in the case of an ambiguous string.
>
> Signed-off-by: Euan Bourke
> ---
> lib/arg_parser/arg_parser.c | 68 +
> lib/arg_parser/rte_arg_parser.h | 60 +
> lib/arg_parser/version.map | 2 +
> 3 files changed, 130 insertions(+)
>
> diff --git a/lib/arg_parser/arg_parser.c b/lib/arg_parser/arg_parser.c
> index cebab9e2f8..95cbc50c13 100644
> --- a/lib/arg_parser/arg_parser.c
> +++ b/lib/arg_parser/arg_parser.c
> @@ -7,10 +7,15 @@
> #include "ctype.h"
> #include "string.h"
> #include "stdbool.h"
> +#include "stdio.h"
>
> #include
> #include
>
> +#define RTE_ARG_PARSE_TYPE_COREMASK 0
> +#define RTE_ARG_PARSE_TYPE_CORELIST 1
> +#define RTE_ARG_PARSE_TYPE_UNKNOWN 2
> +
As these are used as return values, they need to be defined in the header
file so that applications can use them.
> #define BITS_PER_HEX 4
> #define MAX_COREMASK_SIZE ((UINT16_MAX + 1) / BITS_PER_HEX)
>
> @@ -22,6 +27,7 @@ struct core_bits {
> uint32_t total_bits_set;
> };
>
> +
Stray newline added to patch.
> static inline bool
> get_core_bit(struct core_bits *mask, uint16_t idx)
> {
> @@ -159,3 +165,65 @@ rte_arg_parse_coremask(const char *coremask, uint16_t
> *cores, uint32_t cores_len
>
> return total_count;
> }
> +
> +int
> +rte_arg_parse_arg_type(const char *core_string)
> +{
> + /* Remove leading whitespace */
> + while (isblank(*core_string))
> + core_string++;
> +
> + /* Check for 0x prefix */
> + if (core_string[0] == '0' && tolower(core_string[1]) == 'x') {
> + if (core_string[2] != '\0')
> + return RTE_ARG_PARSE_TYPE_COREMASK;
> + return -1;
> + }
> +
> + int i = 0, idx = 0;
> + /* Check for ',' and '-' and check for A-F */
> + do {
> + while (isblank(core_string[idx]))
> + idx++;
> +
> + if (core_string[idx] == ',' || core_string[idx] == '-')
> + return RTE_ARG_PARSE_TYPE_CORELIST;
> +
> + if (isalpha(core_string[idx])) {
> + if (isxdigit(core_string[idx]))
> + return RTE_ARG_PARSE_TYPE_COREMASK;
> + return -1;
> + }
> + idx++;
> + i++;
> + } while (core_string[idx] != '\0');
> +
> + /* Check length of core_string if ambiguous as max length of a uint16_t
> is 5 digits
> + * implying its a coremask.
> + */
> + if (i > 5)
> + return RTE_ARG_PARSE_TYPE_COREMASK;
> +
> + return -1;
Rather than returning -1, I think in most/all cases above, the function
should return -EINVAL as error code, since it's invalid input passed.
> +}
> +
> +int
> +rte_arg_parse_core_string(const char *core_string, uint16_t *cores, uint32_t
> cores_len,
> + int default_type)
> +{
> + if (default_type != RTE_ARG_PARSE_TYPE_COREMASK &&
> + default_type != RTE_ARG_PARSE_TYPE_CORELIST) {
> + return -1;
> + }
> + switch (rte_arg_parse_arg_type(core_string)) {
> + case RTE_ARG_PARSE_TYPE_COREMASK:
> + return rte_arg_parse_coremask(core_string, cores, cores_len);
> + case RTE_ARG_PARSE_TYPE_CORELIST:
> + return rte_arg_parse_corelist(core_string, cores, cores_len);
> + default:
> + return default_type == RTE_ARG_PARSE_TYPE_COREMASK ?
> + rte_arg_parse_coremask(core_string, cores, cores_len) :
> + rte_arg_parse_corelist(core_string, cores, cores_len);
> + return -1;
> + }
> +}
> diff --git a/lib/arg_parser/rte_arg_parser.h b/lib/arg_parser/rte_arg_parser.h
> index 359d40e305..125ca9524c 100644
> --- a/lib/arg_parser/rte_arg_parser.h
> +++ b/lib/arg_parser/rte_arg_parser.h
> @@ -92,6 +92,66 @@ __rte_experimental
> int
> rte_arg_parse_coremask(const char *coremask, uint16_t *cores, uint32_t
> cores_len);
>
> +/**
> + * Use heuristics to determine if a string contains a coremask or a corelist.
> + *
> + * This function will check a series of conditions and return an int
> representing which
> + * core type (mask or list) the string represents or UNKNOWN if the string
> is ambiguous.
"or report the type as unknown if it is ambiguous"
> + *
> + * @param core_string
> + * A string describing the intended cores to be parsed
> + * @return
> + * int representing the core type
> + * -1: error.
Suggest "negative
Re: [RFT] net/nfb: use dynamic logtype
On Thu, 07 Dec 2023 11:37:52 +0100 Martin Spinler wrote: > Also, the nfb_rx.h and nfb_tx.h files use the macro NFB_LOG inside, > please add '#include "nfb_log.h"' into them (then the include in > nfb_rx.c will be duplicate). Otherwise, all .c sources, which include > main nfb.h, don't compile. > > With these changes, the driver works. > > Thank you! > Martin I found a few more leftover bits in nfb driver. Will make a new patchset, and it looks like can build test in a Redhat VM.
Re: [PATCH v3 0/8] add new command line argument parsing library
On Thu, 7 Dec 2023 16:18:10 + Euan Bourke wrote: > A recent thread on the mailing list[1] discussed corelist and coremask > parsing and the idea of a new library dedicated to command line parsing > was mentioned[2]. This patchset adds the library, along with the new > APIs, and edits the existing EAL, DLB2 driver and some example > application functions to use these APIs, rather than each implementing > their own copies. > > The new APIs work similar to the existing functions in EAL, however > instead of filling a core array like this: > [1, -1, -1, 2, 3] (a non -1 refers to an 'active core' at that index) > It fills it like this: > [0, 3, 4] (with the value at each index being an 'active core'). > > The new APIs will also return the number of cores contained in the > passed corelist/coremask, so in the above example, 3 would be returned. > > New in the v3, is a heuristic parser which searches for key markers in > the core string, returning a enum value based off this search to > indicate if a parameter is likely a coremask or a corelist. This > heuristic function is also wrapped in a parser function allowing > apps to handle both coremasks and corelists simultaneously. > > [1] https://mails.dpdk.org/archives/dev/2023-November/280957.html > [2] https://mails.dpdk.org/archives/dev/2023-November/280966.html > To avoid confusion with the other work to handle better args in DPDK, maybe this should be call coreparse library?
Re: [PATCH v2 4/4] net/hns3: use stdatomic API
On 12/7/2023 1:57 PM, Ferruh Yigit wrote: > On 12/7/2023 1:42 AM, Jie Hai wrote: >> Replace the use of gcc builtin __atomic_xxx intrinsics with >> corresponding rte_atomic_xxx optional stdatomic API. >> >> Signed-off-by: Jie Hai >> > > <...> > >> static void >> @@ -643,8 +643,8 @@ hns3vf_get_push_lsc_cap(struct hns3_hw *hw) >> uint16_t exp = HNS3_PF_PUSH_LSC_CAP_UNKNOWN; >> struct hns3_vf *vf = HNS3_DEV_HW_TO_VF(hw); >> >> -__atomic_store_n(&vf->pf_push_lsc_cap, HNS3_PF_PUSH_LSC_CAP_UNKNOWN, >> - __ATOMIC_RELEASE); >> +rte_atomic_load_explicit(&vf->pf_push_lsc_cap, >> +HNS3_PF_PUSH_LSC_CAP_UNKNOWN, rte_memory_order_release); >> > > I guess intension here is to use 'rte_atomic_store_explicit()', current > code cause a build error [1]. > > [1] > https://mails.dpdk.org/archives/test-report/2023-December/525759.html > > Another build error is when stdatomic enabled [2]. [2] meson setup --werror -Denable_stdatomic=true build && ninja -C build
[PATCH 1/3] net/nfb: remove unused device args
The driver has no entries in VALID_KEYS array so there are
no device args. And after parsing it just frees the result.
Looks like it was copy/pasted from some other driver.
Signed-off-by: Stephen Hemminger
---
drivers/net/nfb/nfb.h| 2 --
drivers/net/nfb/nfb_ethdev.c | 16
2 files changed, 18 deletions(-)
diff --git a/drivers/net/nfb/nfb.h b/drivers/net/nfb/nfb.h
index 7dc5bd29e44c..21cd8f78f641 100644
--- a/drivers/net/nfb/nfb.h
+++ b/drivers/net/nfb/nfb.h
@@ -36,8 +36,6 @@
#define RTE_NFB_DRIVER_NAME net_nfb
-/* Device arguments */
-static const char * const VALID_KEYS[] = {NULL};
struct pmd_internals {
uint16_t max_rxmac;
diff --git a/drivers/net/nfb/nfb_ethdev.c b/drivers/net/nfb/nfb_ethdev.c
index defd118bd0ee..4c4e2e3273e6 100644
--- a/drivers/net/nfb/nfb_ethdev.c
+++ b/drivers/net/nfb/nfb_ethdev.c
@@ -518,7 +518,6 @@ nfb_eth_dev_init(struct rte_eth_dev *dev)
struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(dev);
struct rte_pci_addr *pci_addr = &pci_dev->addr;
struct rte_ether_addr eth_addr_init;
- struct rte_kvargs *kvlist;
RTE_LOG(INFO, PMD, "Initializing NFB device (" PCI_PRI_FMT ")\n",
pci_addr->domain, pci_addr->bus, pci_addr->devid,
@@ -529,21 +528,6 @@ nfb_eth_dev_init(struct rte_eth_dev *dev)
pci_addr->domain, pci_addr->bus, pci_addr->devid,
pci_addr->function);
- /* Check validity of device args */
- if (dev->device->devargs != NULL &&
- dev->device->devargs->args != NULL &&
- strlen(dev->device->devargs->args) > 0) {
- kvlist = rte_kvargs_parse(dev->device->devargs->args,
- VALID_KEYS);
- if (kvlist == NULL) {
- RTE_LOG(ERR, PMD, "Failed to parse device arguments %s",
- dev->device->devargs->args);
- rte_kvargs_free(kvlist);
- return -EINVAL;
- }
- rte_kvargs_free(kvlist);
- }
-
/*
* Get number of available DMA RX and TX queues, which is maximum
* number of queues that can be created and store it in private device
--
2.42.0
[PATCH 0/3] net/nfb: driver cleanups
Replace static logtype with dynamic logtype and remove dead code. Compile tested on Fedora. Stephen Hemminger (3): net/nfb: remove unused device args net/nfb: make device path local to init function net/nfb: use dynamic logtype drivers/net/nfb/nfb.h| 10 drivers/net/nfb/nfb_ethdev.c | 44 ++-- drivers/net/nfb/nfb_rx.c | 9 drivers/net/nfb/nfb_rx.h | 2 +- drivers/net/nfb/nfb_tx.c | 9 drivers/net/nfb/nfb_tx.h | 2 +- 6 files changed, 27 insertions(+), 49 deletions(-) -- 2.42.0
[PATCH 2/3] net/nfb: make device path local to init function
The device path is only used to call nfb_open() it does
not have to be stored in internal structure.
Signed-off-by: Stephen Hemminger
---
drivers/net/nfb/nfb.h| 2 --
drivers/net/nfb/nfb_ethdev.c | 8
2 files changed, 4 insertions(+), 6 deletions(-)
diff --git a/drivers/net/nfb/nfb.h b/drivers/net/nfb/nfb.h
index 21cd8f78f641..2707f3db6240 100644
--- a/drivers/net/nfb/nfb.h
+++ b/drivers/net/nfb/nfb.h
@@ -42,8 +42,6 @@ struct pmd_internals {
uint16_t max_txmac;
struct nc_rxmac *rxmac[RTE_MAX_NC_RXMAC];
struct nc_txmac *txmac[RTE_MAX_NC_TXMAC];
-
- char nfb_dev[PATH_MAX];
struct nfb_device *nfb;
};
diff --git a/drivers/net/nfb/nfb_ethdev.c b/drivers/net/nfb/nfb_ethdev.c
index 4c4e2e3273e6..892abe81f3d6 100644
--- a/drivers/net/nfb/nfb_ethdev.c
+++ b/drivers/net/nfb/nfb_ethdev.c
@@ -518,12 +518,13 @@ nfb_eth_dev_init(struct rte_eth_dev *dev)
struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(dev);
struct rte_pci_addr *pci_addr = &pci_dev->addr;
struct rte_ether_addr eth_addr_init;
+ char nfb_dev[PATH_MAX];
RTE_LOG(INFO, PMD, "Initializing NFB device (" PCI_PRI_FMT ")\n",
pci_addr->domain, pci_addr->bus, pci_addr->devid,
pci_addr->function);
- snprintf(internals->nfb_dev, PATH_MAX,
+ snprintf(nfb_dev, sizeof(nfb_dev),
"/dev/nfb/by-pci-slot/" PCI_PRI_FMT,
pci_addr->domain, pci_addr->bus, pci_addr->devid,
pci_addr->function);
@@ -533,10 +534,9 @@ nfb_eth_dev_init(struct rte_eth_dev *dev)
* number of queues that can be created and store it in private device
* data structure.
*/
- internals->nfb = nfb_open(internals->nfb_dev);
+ internals->nfb = nfb_open(nfb_dev);
if (internals->nfb == NULL) {
- RTE_LOG(ERR, PMD, "nfb_open(): failed to open %s",
- internals->nfb_dev);
+ RTE_LOG(ERR, PMD, "nfb_open(): failed to open %s", nfb_dev);
return -EINVAL;
}
data->nb_rx_queues = ndp_get_rx_queue_available_count(internals->nfb);
--
2.42.0
[PATCH 3/3] net/nfb: use dynamic logtype
All drivers should be using dynamic logtype.
Fixes: 6435f9a0ac22 ("net/nfb: add new netcope driver")
Signed-off-by: Stephen Hemminger
---
drivers/net/nfb/nfb.h| 6 +-
drivers/net/nfb/nfb_ethdev.c | 22 +-
drivers/net/nfb/nfb_rx.c | 9 -
drivers/net/nfb/nfb_rx.h | 2 +-
drivers/net/nfb/nfb_tx.c | 9 -
drivers/net/nfb/nfb_tx.h | 2 +-
6 files changed, 24 insertions(+), 26 deletions(-)
diff --git a/drivers/net/nfb/nfb.h b/drivers/net/nfb/nfb.h
index 2707f3db6240..3a27457d6e33 100644
--- a/drivers/net/nfb/nfb.h
+++ b/drivers/net/nfb/nfb.h
@@ -12,6 +12,11 @@
#include
#include
+extern int nfb_logtype;
+#define NFB_LOG(level, fmt, args...) \
+ rte_log(RTE_LOG_ ## level, nfb_logtype, "%s(): " fmt "\n", \
+ __func__, ## args)
+
#include "nfb_rx.h"
#include "nfb_tx.h"
@@ -36,7 +41,6 @@
#define RTE_NFB_DRIVER_NAME net_nfb
-
struct pmd_internals {
uint16_t max_rxmac;
uint16_t max_txmac;
diff --git a/drivers/net/nfb/nfb_ethdev.c b/drivers/net/nfb/nfb_ethdev.c
index 892abe81f3d6..fc510db7390c 100644
--- a/drivers/net/nfb/nfb_ethdev.c
+++ b/drivers/net/nfb/nfb_ethdev.c
@@ -12,11 +12,9 @@
#include
#include
+#include "nfb.h"
#include "nfb_stats.h"
-#include "nfb_rx.h"
-#include "nfb_tx.h"
#include "nfb_rxmode.h"
-#include "nfb.h"
/**
* Default MAC addr
@@ -192,8 +190,7 @@ nfb_eth_dev_configure(struct rte_eth_dev *dev __rte_unused)
(&nfb_timestamp_dynfield_offset,
&nfb_timestamp_rx_dynflag);
if (ret != 0) {
- RTE_LOG(ERR, PMD, "Cannot register Rx timestamp"
- " field/flag %d\n", ret);
+ NFB_LOG(ERR, "Cannot register Rx timestamp field/flag
%d", ret);
nfb_close(internals->nfb);
return -rte_errno;
}
@@ -520,7 +517,7 @@ nfb_eth_dev_init(struct rte_eth_dev *dev)
struct rte_ether_addr eth_addr_init;
char nfb_dev[PATH_MAX];
- RTE_LOG(INFO, PMD, "Initializing NFB device (" PCI_PRI_FMT ")\n",
+ NFB_LOG(INFO, "Initializing NFB device (" PCI_PRI_FMT ")",
pci_addr->domain, pci_addr->bus, pci_addr->devid,
pci_addr->function);
@@ -536,13 +533,13 @@ nfb_eth_dev_init(struct rte_eth_dev *dev)
*/
internals->nfb = nfb_open(nfb_dev);
if (internals->nfb == NULL) {
- RTE_LOG(ERR, PMD, "nfb_open(): failed to open %s", nfb_dev);
+ NFB_LOG(ERR, "nfb_open(): failed to open %s", nfb_dev);
return -EINVAL;
}
data->nb_rx_queues = ndp_get_rx_queue_available_count(internals->nfb);
data->nb_tx_queues = ndp_get_tx_queue_available_count(internals->nfb);
- RTE_LOG(INFO, PMD, "Available NDP queues RX: %u TX: %u\n",
+ NFB_LOG(INFO, "Available NDP queues RX: %u TX: %u\n",
data->nb_rx_queues, data->nb_tx_queues);
nfb_nc_rxmac_init(internals->nfb,
@@ -567,7 +564,7 @@ nfb_eth_dev_init(struct rte_eth_dev *dev)
data->mac_addrs = rte_zmalloc(data->name,
sizeof(struct rte_ether_addr) * mac_count, RTE_CACHE_LINE_SIZE);
if (data->mac_addrs == NULL) {
- RTE_LOG(ERR, PMD, "Could not alloc space for MAC address!\n");
+ NFB_LOG(ERR, "Could not alloc space for MAC address");
nfb_close(internals->nfb);
return -EINVAL;
}
@@ -585,8 +582,7 @@ nfb_eth_dev_init(struct rte_eth_dev *dev)
dev->data->dev_flags |= RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS;
- RTE_LOG(INFO, PMD, "NFB device ("
- PCI_PRI_FMT ") successfully initialized\n",
+ NFB_LOG(INFO, "NFB device (" PCI_PRI_FMT ") successfully initialized",
pci_addr->domain, pci_addr->bus, pci_addr->devid,
pci_addr->function);
@@ -610,8 +606,7 @@ nfb_eth_dev_uninit(struct rte_eth_dev *dev)
nfb_eth_dev_close(dev);
- RTE_LOG(INFO, PMD, "NFB device ("
- PCI_PRI_FMT ") successfully uninitialized\n",
+ NFB_LOG(INFO, "NFB device (" PCI_PRI_FMT ") successfully uninitialized",
pci_addr->domain, pci_addr->bus, pci_addr->devid,
pci_addr->function);
@@ -674,3 +669,4 @@ static struct rte_pci_driver nfb_eth_driver = {
RTE_PMD_REGISTER_PCI(RTE_NFB_DRIVER_NAME, nfb_eth_driver);
RTE_PMD_REGISTER_PCI_TABLE(RTE_NFB_DRIVER_NAME, nfb_pci_id_table);
RTE_PMD_REGISTER_KMOD_DEP(RTE_NFB_DRIVER_NAME, "* nfb");
+RTE_LOG_REGISTER_DEFAULT(nfb_logtype, NOTICE);
diff --git a/drivers/net/nfb/nfb_rx.c b/drivers/net/nfb/nfb_rx.c
index 8a9b232305f2..a20f7b9b4b41 100644
--- a/drivers/net/nfb/nfb_rx.c
+++ b/drivers/net/nfb/nfb_rx.c
@@ -6,7 +6,6 @@
#include
-#include "nfb_rx.h"
#include "nfb.h"
uint64_t nfb_timestamp_rx_dynflag;
@@ -19,7 +18,7 @@ nfb_eth_rx_que
[PATCH] eal: initialize shared plugins on Windows
When EAL is built with MSVC it is possible to dynamically load plugins on Windows. Hook eal_plugins_init into rte_eal_init if built with MSVC and provide code to load plugins on Windows. Tyler Retzlaff (1): eal: initialize shared plugins on Windows lib/eal/common/eal_common_options.c | 92 ++--- lib/eal/windows/eal.c | 8 2 files changed, 84 insertions(+), 16 deletions(-) -- 1.8.3.1
[PATCH] eal: initialize shared plugins on Windows
When EAL is built with MSVC it is possible to dynamically load plugins
on Windows. Hook eal_plugins_init into rte_eal_init if built with MSVC
and provide code to load plugins on Windows.
Signed-off-by: Tyler Retzlaff
---
lib/eal/common/eal_common_options.c | 92 ++---
lib/eal/windows/eal.c | 8
2 files changed, 84 insertions(+), 16 deletions(-)
diff --git a/lib/eal/common/eal_common_options.c
b/lib/eal/common/eal_common_options.c
index a6d21f1..3f9a030 100644
--- a/lib/eal/common/eal_common_options.c
+++ b/lib/eal/common/eal_common_options.c
@@ -18,9 +18,7 @@
#include
#endif
#include
-#ifndef RTE_EXEC_ENV_WINDOWS
#include
-#endif
#include
#include
@@ -74,7 +72,7 @@
{OPT_HELP, 0, NULL, OPT_HELP_NUM },
{OPT_HUGE_DIR, 1, NULL, OPT_HUGE_DIR_NUM },
{OPT_HUGE_UNLINK, 2, NULL, OPT_HUGE_UNLINK_NUM },
- {OPT_IOVA_MODE, 1, NULL, OPT_IOVA_MODE_NUM},
+ {OPT_IOVA_MODE, 1, NULL, OPT_IOVA_MODE_NUM},
{OPT_LCORES,1, NULL, OPT_LCORES_NUM },
{OPT_LOG_LEVEL, 1, NULL, OPT_LOG_LEVEL_NUM},
{OPT_TRACE, 1, NULL, OPT_TRACE_NUM},
@@ -123,10 +121,8 @@ struct shared_driver {
static struct shared_driver_list solib_list =
TAILQ_HEAD_INITIALIZER(solib_list);
-#ifndef RTE_EXEC_ENV_WINDOWS
/* Default path of external loadable drivers */
static const char *default_solib_dir = RTE_EAL_PMD_PATH;
-#endif
/*
* Stringified version of solib path used by dpdk-pmdinfo.py
@@ -371,12 +367,12 @@ struct device_option {
}
#ifdef RTE_EXEC_ENV_WINDOWS
-int
-eal_plugins_init(void)
-{
- return 0;
-}
+#define SOEXT".dll"
#else
+#define SOEXT".so"
+#endif
+
+#define SOABIEXT SOEXT"."ABI_VERSION
static int
eal_plugindir_init(const char *path)
@@ -397,12 +393,14 @@ struct device_option {
while ((dent = readdir(d)) != NULL) {
struct stat sb;
- int nlen = strnlen(dent->d_name, sizeof(dent->d_name));
+ size_t nlen = strnlen(dent->d_name, sizeof(dent->d_name));
- /* check if name ends in .so or .so.ABI_VERSION */
- if (strcmp(&dent->d_name[nlen - 3], ".so") != 0 &&
- strcmp(&dent->d_name[nlen - 4 - strlen(ABI_VERSION)],
- ".so."ABI_VERSION) != 0)
+ if (nlen < strlen(SOABIEXT))
+ continue;
+
+ /* check if name ends in SOEXT or SOABIEXT */
+ if (strcmp(&dent->d_name[nlen - strlen(SOEXT)], SOEXT) != 0 &&
+ strcmp(&dent->d_name[nlen - strlen(SOABIEXT)], SOABIEXT) !=
0)
continue;
snprintf(sopath, sizeof(sopath), "%s/%s", path, dent->d_name);
@@ -420,6 +418,68 @@ struct device_option {
return (dent == NULL) ? 0 : -1;
}
+#ifdef RTE_EXEC_ENV_WINDOWS
+static void*
+eal_dlopen(const char *pathname)
+{
+ void *retval = NULL;
+ struct stat pathstat;
+ char *fullpath = _fullpath(NULL, pathname, 0);
+
+ const char *loadpath = fullpath;
+ DWORD loadflags = 0;
+
+ if (fullpath == NULL) {
+ RTE_LOG(ERR, EAL, "Error expanding full path for %s, %s\n",
+ pathname, strerror(errno));
+ } else {
+ /* Verify that the path exists */
+ if ((stat(fullpath, &pathstat) != 0) && (errno == ENOENT)) {
+ /* not a full or relative path, try a load from default
dirs */
+ loadpath = pathname;
+ loadflags = LOAD_LIBRARY_SEARCH_DEFAULT_DIRS;
+ }
+
+ retval = LoadLibraryExA(loadpath, NULL, loadflags);
+ if (retval == NULL)
+ RTE_LOG(ERR, EAL, "Error loading %s, error code: %"
PRIu32 "\n",
+ loadpath, GetLastError());
+ }
+
+ free(fullpath);
+ return retval;
+}
+
+static int
+is_shared_build(void)
+{
+ int shared = 0;
+ HMODULE apphandle = NULL;
+ HMODULE libhandle = NULL;
+
+ /* if fail to get handle, assume statically linked */
+ apphandle = GetModuleHandleA(NULL);
+ if (apphandle == NULL) {
+ RTE_LOG(ERR, EAL, "Cannot get handle to the app\n");
+ goto out;
+ }
+
+ /* if fail to get handle, assume statically linked */
+ if (GetModuleHandleExA(GET_MODULE_HANDLE_EX_FLAG_FROM_ADDRESS |
+ GET_MODULE_HANDLE_EX_FLAG_UNCHANGED_REFCOUNT,
+ (LPCSTR)&eal_plugins_init,
+ &libhandle)) {
+ if (apphandle != libhandle) {
+ /* lib and app handles are different. */
+ /* Therefore lib is dynamically linked */
+ shared = 1;
+ }
+ }
+
+out:
+ return shared;
+}
+#else
static
Re: [PATCH v3 1/8] arg_parser: new library for command line parsing
On Thu, Dec 07, 2023 at 04:18:11PM +, Euan Bourke wrote:
> Add a new library to make it easier for eal and other libraries to parse
> command line arguments.
>
> The first function in this library is one to parse a corelist string
> into an array of individual core ids. The function will then return the
> total number of cores described in the corelist.
>
> Signed-off-by: Euan Bourke
> ---
> .mailmap| 1 +
> MAINTAINERS | 4 ++
> doc/api/doxy-api-index.md | 3 +-
> doc/api/doxy-api.conf.in| 1 +
> lib/arg_parser/arg_parser.c | 108
> lib/arg_parser/meson.build | 7 +++
> lib/arg_parser/rte_arg_parser.h | 66 +++
> lib/arg_parser/version.map | 10 +++
> lib/meson.build | 2 +
> 9 files changed, 201 insertions(+), 1 deletion(-)
> create mode 100644 lib/arg_parser/arg_parser.c
> create mode 100644 lib/arg_parser/meson.build
> create mode 100644 lib/arg_parser/rte_arg_parser.h
> create mode 100644 lib/arg_parser/version.map
>
> diff --git a/.mailmap b/.mailmap
> index ab0742a382..528bc68a30 100644
> --- a/.mailmap
> +++ b/.mailmap
> @@ -379,6 +379,7 @@ Eric Zhang
> Erik Gabriel Carrillo
> Erik Ziegenbalg
> Erlu Chen
> +Euan Bourke
> Eugenio Pérez
> Eugeny Parshutin
> Evan Swanson
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 0d1c8126e3..68ef5ba14b 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -1756,6 +1756,10 @@ M: Nithin Dabilpuram
> M: Pavan Nikhilesh
> F: lib/node/
>
> +Argument parsing
> +M: Bruce Richardson
> +F: lib/arg_parser/
> +
>
> Test Applications
> -
> diff --git a/doc/api/doxy-api-index.md b/doc/api/doxy-api-index.md
> index a6a768bd7c..f711010140 100644
> --- a/doc/api/doxy-api-index.md
> +++ b/doc/api/doxy-api-index.md
> @@ -221,7 +221,8 @@ The public API headers are grouped by topics:
>[config file](@ref rte_cfgfile.h),
>[key/value args](@ref rte_kvargs.h),
>[string](@ref rte_string_fns.h),
> - [thread](@ref rte_thread.h)
> + [thread](@ref rte_thread.h),
> + [argument parsing](@ref rte_arg_parser.h)
>
> - **debug**:
>[jobstats](@ref rte_jobstats.h),
> diff --git a/doc/api/doxy-api.conf.in b/doc/api/doxy-api.conf.in
> index e94c9e4e46..05718ba6ed 100644
> --- a/doc/api/doxy-api.conf.in
> +++ b/doc/api/doxy-api.conf.in
> @@ -28,6 +28,7 @@ INPUT =
> @TOPDIR@/doc/api/doxy-api-index.md \
>@TOPDIR@/lib/eal/include \
>@TOPDIR@/lib/eal/include/generic \
>@TOPDIR@/lib/acl \
> + @TOPDIR@/lib/arg_parser \
>@TOPDIR@/lib/bbdev \
>@TOPDIR@/lib/bitratestats \
>@TOPDIR@/lib/bpf \
> diff --git a/lib/arg_parser/arg_parser.c b/lib/arg_parser/arg_parser.c
> new file mode 100644
> index 00..240f63d8e1
> --- /dev/null
> +++ b/lib/arg_parser/arg_parser.c
> @@ -0,0 +1,108 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + * Copyright(c) 2023 Intel Corporation
> + */
> +
> +#include "errno.h"
> +#include "stdlib.h"
> +#include "ctype.h"
> +#include "string.h"
> +#include "stdbool.h"
should be <> not "" quotes for system or standard library headers includes.
> +
> +#include
> +#include
> +
> +
> +struct core_bits {
> + uint8_t bits[(UINT16_MAX + 1) / CHAR_BIT];
> + uint16_t max_bit_set;
> + uint16_t min_bit_set;
> + uint32_t total_bits_set;
> +};
> +
> +static inline bool
> +get_core_bit(struct core_bits *mask, uint16_t idx)
nit: use size_t typed parameters for things that are used as an offset /
subscript value. applies to all instances in the series.
Re: [PATCH] eal: initialize shared plugins on Windows
On Thu, Dec 07, 2023 at 11:20:51AM -0800, Tyler Retzlaff wrote:
> When EAL is built with MSVC it is possible to dynamically load plugins
> on Windows. Hook eal_plugins_init into rte_eal_init if built with MSVC
> and provide code to load plugins on Windows.
>
> Signed-off-by: Tyler Retzlaff
> ---
> lib/eal/common/eal_common_options.c | 92
> ++---
> lib/eal/windows/eal.c | 8
> 2 files changed, 84 insertions(+), 16 deletions(-)
>
> diff --git a/lib/eal/common/eal_common_options.c
> b/lib/eal/common/eal_common_options.c
> index a6d21f1..3f9a030 100644
> --- a/lib/eal/common/eal_common_options.c
> +++ b/lib/eal/common/eal_common_options.c
> @@ -18,9 +18,7 @@
> #include
> #endif
> #include
> -#ifndef RTE_EXEC_ENV_WINDOWS
> #include
> -#endif
>
> #include
> #include
> @@ -74,7 +72,7 @@
> {OPT_HELP, 0, NULL, OPT_HELP_NUM },
> {OPT_HUGE_DIR, 1, NULL, OPT_HUGE_DIR_NUM },
> {OPT_HUGE_UNLINK, 2, NULL, OPT_HUGE_UNLINK_NUM },
> - {OPT_IOVA_MODE, 1, NULL, OPT_IOVA_MODE_NUM},
> + {OPT_IOVA_MODE, 1, NULL, OPT_IOVA_MODE_NUM},
accidentally converted 8 spaces to a tab here, will submit a new version
to remove the change from this series (or maintainer can adjust).
thanks
[PATCH] maintainers: update Windows maintainers
Remove Windows maintainers who are no longer working on DPDK and add myself. Signed-off-by: Tyler Retzlaff --- MAINTAINERS | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 0d1c812..4ac25b1 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -360,8 +360,7 @@ F: kernel/freebsd/nic_uio/ Windows support M: Dmitry Kozlyuk -M: Narcisa Ana Maria Vasile -M: Dmitry Malloy +M: Tyler Retzlaff M: Pallavi Kadam F: lib/eal/windows/ F: buildtools/map_to_win.py -- 1.8.3.1
[PATCH v2] eal: initialize shared plugins on Windows
When EAL is built with MSVC it is possible to dynamically load plugins on Windows. Hook eal_plugins_init into rte_eal_init if built with MSVC and provide code to load plugins on Windows. v2: * revert unintended / unrelated whitespace change * include inttypes.h for use of PRIu32 in log format string Tyler Retzlaff (1): eal: initialize shared plugins on Windows lib/eal/common/eal_common_options.c | 91 +++-- lib/eal/windows/eal.c | 8 2 files changed, 84 insertions(+), 15 deletions(-) -- 1.8.3.1
[PATCH v2] eal: initialize shared plugins on Windows
When EAL is built with MSVC it is possible to dynamically load plugins
on Windows. Hook eal_plugins_init into rte_eal_init if built with MSVC
and provide code to load plugins on Windows.
Signed-off-by: Tyler Retzlaff
---
lib/eal/common/eal_common_options.c | 91 +++--
lib/eal/windows/eal.c | 8
2 files changed, 84 insertions(+), 15 deletions(-)
diff --git a/lib/eal/common/eal_common_options.c
b/lib/eal/common/eal_common_options.c
index a6d21f1..55be61d 100644
--- a/lib/eal/common/eal_common_options.c
+++ b/lib/eal/common/eal_common_options.c
@@ -5,6 +5,7 @@
#include
#include
+#include
#include
#ifndef RTE_EXEC_ENV_WINDOWS
#include
@@ -18,9 +19,7 @@
#include
#endif
#include
-#ifndef RTE_EXEC_ENV_WINDOWS
#include
-#endif
#include
#include
@@ -123,10 +122,8 @@ struct shared_driver {
static struct shared_driver_list solib_list =
TAILQ_HEAD_INITIALIZER(solib_list);
-#ifndef RTE_EXEC_ENV_WINDOWS
/* Default path of external loadable drivers */
static const char *default_solib_dir = RTE_EAL_PMD_PATH;
-#endif
/*
* Stringified version of solib path used by dpdk-pmdinfo.py
@@ -371,12 +368,12 @@ struct device_option {
}
#ifdef RTE_EXEC_ENV_WINDOWS
-int
-eal_plugins_init(void)
-{
- return 0;
-}
+#define SOEXT".dll"
#else
+#define SOEXT".so"
+#endif
+
+#define SOABIEXT SOEXT"."ABI_VERSION
static int
eal_plugindir_init(const char *path)
@@ -397,12 +394,14 @@ struct device_option {
while ((dent = readdir(d)) != NULL) {
struct stat sb;
- int nlen = strnlen(dent->d_name, sizeof(dent->d_name));
+ size_t nlen = strnlen(dent->d_name, sizeof(dent->d_name));
- /* check if name ends in .so or .so.ABI_VERSION */
- if (strcmp(&dent->d_name[nlen - 3], ".so") != 0 &&
- strcmp(&dent->d_name[nlen - 4 - strlen(ABI_VERSION)],
- ".so."ABI_VERSION) != 0)
+ if (nlen < strlen(SOABIEXT))
+ continue;
+
+ /* check if name ends in SOEXT or SOABIEXT */
+ if (strcmp(&dent->d_name[nlen - strlen(SOEXT)], SOEXT) != 0 &&
+ strcmp(&dent->d_name[nlen - strlen(SOABIEXT)], SOABIEXT) !=
0)
continue;
snprintf(sopath, sizeof(sopath), "%s/%s", path, dent->d_name);
@@ -420,6 +419,68 @@ struct device_option {
return (dent == NULL) ? 0 : -1;
}
+#ifdef RTE_EXEC_ENV_WINDOWS
+static void*
+eal_dlopen(const char *pathname)
+{
+ void *retval = NULL;
+ struct stat pathstat;
+ char *fullpath = _fullpath(NULL, pathname, 0);
+
+ const char *loadpath = fullpath;
+ DWORD loadflags = 0;
+
+ if (fullpath == NULL) {
+ RTE_LOG(ERR, EAL, "Error expanding full path for %s, %s\n",
+ pathname, strerror(errno));
+ } else {
+ /* Verify that the path exists */
+ if ((stat(fullpath, &pathstat) != 0) && (errno == ENOENT)) {
+ /* not a full or relative path, try a load from default
dirs */
+ loadpath = pathname;
+ loadflags = LOAD_LIBRARY_SEARCH_DEFAULT_DIRS;
+ }
+
+ retval = LoadLibraryExA(loadpath, NULL, loadflags);
+ if (retval == NULL)
+ RTE_LOG(ERR, EAL, "Error loading %s, error code: %"
PRIu32 "\n",
+ loadpath, GetLastError());
+ }
+
+ free(fullpath);
+ return retval;
+}
+
+static int
+is_shared_build(void)
+{
+ int shared = 0;
+ HMODULE apphandle = NULL;
+ HMODULE libhandle = NULL;
+
+ /* if fail to get handle, assume statically linked */
+ apphandle = GetModuleHandleA(NULL);
+ if (apphandle == NULL) {
+ RTE_LOG(ERR, EAL, "Cannot get handle to the app\n");
+ goto out;
+ }
+
+ /* if fail to get handle, assume statically linked */
+ if (GetModuleHandleExA(GET_MODULE_HANDLE_EX_FLAG_FROM_ADDRESS |
+ GET_MODULE_HANDLE_EX_FLAG_UNCHANGED_REFCOUNT,
+ (LPCSTR)&eal_plugins_init,
+ &libhandle)) {
+ if (apphandle != libhandle) {
+ /* lib and app handles are different. */
+ /* Therefore lib is dynamically linked */
+ shared = 1;
+ }
+ }
+
+out:
+ return shared;
+}
+#else
static int
verify_perms(const char *dirpath)
{
@@ -527,6 +588,7 @@ struct device_option {
RTE_LOG(INFO, EAL, "Detected static linkage of DPDK\n");
return 0;
}
+#endif
int
eal_plugins_init(void)
@@ -565,7 +627,6 @@ struct device_option {
}
return 0;
}
-#endif
/*
* Parse the coremask given as argument (hexadecimal string) and fill
diff --git a/lib/eal/windows/eal.c b/lib/eal/windows/eal.c
index 7ec2152..af51bf3 100644
--- a/lib
Re: [PATCH v4 1/2] config: correct cpu instruction set for cross build
On Tue, 5 Dec 2023 03:52:58 +
Joyce Kong wrote:
> The platform value would be 'native' only when not cross build.
> Move the operation about modifying cpu_instruction_set while
> platform equals 'native' to the not cross build branch.
>
> Fixes: bf66003b51ec ("build: use platform for generic and native builds")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Joyce Kong
> Reviewed-by: Ruifeng Wang
Looks right, what is the impact on existing users doing current cross builds?
Acked-by: Stephen Hemminger
