The Key Wrap approach is used by applications in order to protect keys
located in untrusted storage or transmitted over untrusted
communications networks. The constructions are typically built from
standard primitives such as block ciphers and cryptographic hash
functions.
The Key Wrap method and
> -Original Message-
> From: Kalesh A P
> Sent: Tuesday, March 16, 2021 14:52
> To: dev@dpdk.org
> Cc: Li, Xiaoyun ; Yigit, Ferruh
> Subject: [dpdk-dev] [PATCH] app/testpmd: fix unchecked return value
>
> From: Kalesh AP
>
> This patch checks return value for rte_eth_dev_info_get() in
From: Kalesh AP
This patch checks return value for rte_eth_dev_info_get() in show_macs().
Coverity issue: 353629
Fixes: e1d44d0ad623 ("app/testpmd: show MAC addresses added to a port")
Cc: sta...@dpdk.org
Signed-off-by: Kalesh AP
---
app/test-pmd/config.c | 3 ++-
1 file changed, 2 insertions
From: Vijay Kumar Srivastava
Move function to get efx family from net driver into common driver.
Signed-off-by: Vijay Kumar Srivastava
Signed-off-by: Andrew Rybchenko
---
drivers/common/meson.build | 2 +-
drivers/common/sfc_efx/meson.build | 6 +++
drivers/common/sfc_efx/sfc_efx.c
From: Vijay Kumar Srivastava
Driver would be probed only for the net device class.
Signed-off-by: Vijay Kumar Srivastava
Signed-off-by: Andrew Rybchenko
---
doc/guides/nics/sfc_efx.rst | 8
drivers/net/sfc/sfc.h| 1 +
drivers/net/sfc/sfc_ethdev.c | 7 +++
drivers/net/sfc
From: Vijay Kumar Srivastava
Device class argument would be used to select compatible driver.
Driver probe would be skipped for incompatible device class.
Signed-off-by: Vijay Kumar Srivastava
Signed-off-by: Andrew Rybchenko
---
drivers/common/sfc_efx/sfc_efx.c | 49
From: Vijay Kumar Srivastava
Add an API to verify virtio features supported by device.
Signed-off-by: Vijay Kumar Srivastava
Signed-off-by: Andrew Rybchenko
---
drivers/common/sfc_efx/base/efx.h | 7
drivers/common/sfc_efx/base/efx_impl.h | 2 +
drivers/common/sfc_efx/base
From: Vijay Kumar Srivastava
Add an API to get virtio features supported by device.
Signed-off-by: Vijay Kumar Srivastava
Signed-off-by: Andrew Rybchenko
---
drivers/common/sfc_efx/base/efx.h | 14 ++
drivers/common/sfc_efx/base/efx_impl.h | 2 +
drivers/common/sfc_efx/base/
From: Vijay Kumar Srivastava
Add EFSYS_HAS_UINT64 build dependency on EFSYS_OPT_VIRTIO.
virtio features are represented as bitmask in 64-bit unsigned
integer.
Signed-off-by: Vijay Kumar Srivastava
Signed-off-by: Andrew Rybchenko
---
drivers/common/sfc_efx/base/efx_check.h | 3 +++
1 file chan
From: Vijay Srivastava
In the vDPA mode, only data path is offloaded in the hardware and
control path still goes through the hypervisor and it configures
virtqueues via vDPA driver so new virtqueue APIs are required.
Implement virtio init/fini and virtqueue create/destroy APIs.
Signed-off-by: V
From: Vijay Srivastava
Add an API to query the virtqueue doorbell offset in the BAR for a VI.
For vDPA, the virtio net driver notifies the device directly by writing
doorbell. This API would be invoked from vDPA client driver.
Signed-off-by: Vijay Srivastava
Signed-off-by: Andrew Rybchenko
---
Update base driver to provide functionality required by vDPA driver.
Factor out helper functions to be shared by net and vDPA drivers.
v3:
- one more attempt to fix windows build breakage
v2:
- fix windows build breakage - do not build common/sfc_efx in the case
of windows
- remove undefin
>On Mon, Mar 8, 2021 at 7:44 PM Jerin Jacob
>wrote:
>> Summary:
>> 1) Ideal way of adding this feature is to add elements in the
>> existing structure as mentioned
>> in ("eventdev: introduce event vector Rx capability") in this series.
>> 2) Since this breaking ABI, Introducing a new structure
Hi
> -Original Message-
> From: Kalesh A P
> Sent: Tuesday, March 16, 2021 12:55
> To: dev@dpdk.org
> Cc: Li, Xiaoyun ; Yigit, Ferruh
> Subject: [dpdk-dev] [PATCH] app/testpmd: fix unchecked return value
>
> From: Kalesh AP
>
> CID 353629 (#1 of 1): Unchecked return value (CHECKED_RET
Series Acked-by: Anoob Joseph
>
> This series adds support for block cipher DIGEST_ENCRYPTED mode in
> OCTEON TX, OCTEON TX2 PMDs and sample unit test application.
>
> Tejasree Kondoj (2):
> common/cpt: support DIGEST_ENCRYPTED mode
> test/crypto: support block cipher DIGEST_ENCRYPTED mode
From: Kalesh AP
CID 363716 (#1 of 1): Unchecked return value (CHECKED_RETURN)
check_return: Calling rte_pci_write_config without checking
return value (as is done elsewhere 46 out of 49 times).
Coverity issue: 363716
Fixes: be14720def9c ("net/bnxt: support FW reset")
Cc: sta...@dpdk.org
Signed
From: Kalesh AP
CID 353629 (#1 of 1): Unchecked return value (CHECKED_RETURN)
check_return: Calling rte_eth_dev_info_get without checking
return value (as is done elsewhere 110 out of 117 times).
Coverity issue: 353629
Fixes: e1d44d0ad623 ("app/testpmd: show MAC addresses added to a port")
Cc:
For NVGRE protocol, the default value of c_k_s_rsvd0_ver
must be 0x2000, and protocol type must be 0x6558 in the NVGRE
header.
This patch updates these two configuraitons while parser the nvgre
encap.
Fixes: dcd962fc6b ("app/testpmd: add NVGRE encap/decap")
Cc: sta...@dpdk.org
Signed-off-by: Jia
Hi Jan,
> -Original Message-
> From: Jan Viktorin
> Sent: Monday, March 15, 2021 9:22 PM
> To: Jiawei(Jonny) Wang
> Cc: Slava Ovsiienko ; Asaf Penso
> ; dev@dpdk.org; Ori Kam
> Subject: Re: [dpdk-dev] Duplicating traffic with RTE Flow
>
> Hello Jiawei,
>
> On Fri, 12 Mar 2021 09:32:4
Use UINT64_MAX and UINT32_MAX instead of -1 or ~0 literal variations
of different explicit widths when creating masks and sentinel values.
some compilers generate a warning when applying a '-' to an unsigned
literal so avoid this by initializing with unsigned preprocessor
definitions where appropr
Check for failure, while here just increment len once after checking for
failure instead of duplicating len + 1 math in two different argument
lists.
Signed-off-by: Tyler Retzlaff
---
lib/librte_eal/common/eal_common_devargs.c | 9 ++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff
10/03/2021 13:48, Tal Shnaiderman:
> rename the key opaque pointer from rte_tls_key to
> rte_thread_key to avoid confusion with transport layer security.
>
> Also rename and remove the "_tls" term from the following
> functions to avoid redundancy:
>
> rte_thread_tls_key_create
> rte_thread_tls_k
10/03/2021 13:48, Tal Shnaiderman:
> --- a/lib/librte_eal/include/rte_thread.h
> +++ b/lib/librte_eal/include/rte_thread.h
> @@ -59,7 +59,9 @@ void rte_thread_get_affinity(rte_cpuset_t *cpusetp);
> *
> * @return
> * On success, zero.
> - * On failure, a negative number.
> + * On failure
09/02/2021 16:06, Ferruh Yigit:
> In Linux by default DPDK log goes to stdout, as well as syslog.
>
> It is possible for an application to change the library output stream
> via 'rte_openlog_stream()' API, to set it to stderr, it can be used as:
> rte_openlog_stream(stderr);
>
> But still updatin
Meter arrays are stateful objects that are updated by the data plane
and configured & monitored by the control plane. The meters implement
the RFC 2698 Two Rate Three Color Marker (trTCM) algorithm.
Signed-off-by: Cristian Dumitrescu
---
examples/pipeline/cli.c | 471 ++
Register arrays are stateful objects that can be read & modified by
both the data plane and the control plane, as opposed to tables, which
are read-only for data plane. One key use-case is the implementation
of stats counters.
Signed-off-by: Cristian Dumitrescu
---
examples/pipeline/cli.c
11/02/2021 13:59, Juraj Linkeš:
> The kni linux module is using a custom target for building, which
> doesn't take into account any cross compilation arguments. The arguments
> in question are ARCH, CROSS_COMPILE (for gcc, clang) and CC, LD (for
> clang). Get those from the cross file and pass them
16/02/2021 16:13, Bruce Richardson:
> For using a DPDK application, such as OVS, which is dynamically linked, the
> DPDK version in use should always report the actual version, not the
> version used at build time. This incorrect behaviour can be seen by
> building OVS against one version of DPDK a
Meter arrays are stateful objects that are updated by the data plane
and configured & monitored by the control plane. The meters implement
the RFC 2698 Two Rate Three Color Marker (trTCM) algorithm.
Signed-off-by: Cristian Dumitrescu
---
examples/pipeline/cli.c | 471 ++
Register arrays are stateful objects that can be read & modified by
both the data plane and the control plane, as opposed to tables, which
are read-only for data plane. One key use-case is the implementation
of stats counters.
Signed-off-by: Cristian Dumitrescu
---
examples/pipeline/cli.c
When link goes down, disable the port's Rx path to drop the incoming
traffic closer to the wire, instead of accepting them in for further
Rx processing, only to eventually drop them at the port's Rxqs. This
prevents unnecessary congestion in the Rx path. The port's Rx path
will be re-enabled once t
From: Asaf Penso
The current define for MLX5_DRIVER_NAME refers specially for the PCI
driver.
The define itself does not mention PCI and this is confusing.
Rename from MLX5_DRIVER_NAME to MLX5_PCI_DRIVER_NAME.
Signed-off-by: Asaf Penso
---
drivers/common/mlx5/mlx5_common.h | 2 +-
driver
From: Asaf Penso
Some mlx5 pmds define the log prefix as "mlx5_pmd" while others as
"pmd_mlx5".
The patch aligns all pmds to use the "mlx5_pmd" format.
Signed-off-by: Asaf Penso
---
drivers/common/mlx5/mlx5_common.h | 4
drivers/common/mlx5/mlx5_common_utils.h | 2 +-
drivers/net/ml
On Mon, 15 Mar 2021 19:52:46 +
Slava Ovsiienko wrote:
> Hi, Jan
>
> > -Original Message-
> > From: Jan Viktorin
> > Sent: Monday, March 15, 2021 21:49
> > To: Slava Ovsiienko
> > Cc: dev@dpdk.org; Asaf Penso ; Shahaf Shuler
> > ; Matan Azrad
> > Subject: Re: [PATCH 2/2] net/mlx5:
Hi, Jan
> -Original Message-
> From: Jan Viktorin
> Sent: Monday, March 15, 2021 21:49
> To: Slava Ovsiienko
> Cc: dev@dpdk.org; Asaf Penso ; Shahaf Shuler
> ; Matan Azrad
> Subject: Re: [PATCH 2/2] net/mlx5: avoid unbind step to enable switchdev
> mode
>
> On Mon, 15 Mar 2021 19:31:25
On Mon, 15 Mar 2021 19:31:25 +
Slava Ovsiienko wrote:
> Hi, Jan
Hello Salva,
I am sorry, but I didn't get the point of your response. For me, there
was 1 PF :05:00.1, SRIOV disabled. So, I followed the guide [1].
1. Enable SRIOV via mlxconfig. Done.
2. Configure max number of VFs. Don
On 3/11/21 8:11 AM, Piotr Kubaj wrote:
There are couple of issues when building with Clang:
1. vector is a keyword and should not be used in code. I undefined it,
but it would probably be better to just change the variable name.
2. vector long is deprecated by Clang and should not be used. I s
On Mon, 15 Mar 2021 12:27:20 -0700
Stephen Hemminger wrote:
> This fixes two places where pthread_mutex was being unsafely
> used between primary secondary process.
>
> These patches are necessary but not sufficient to address Bug 662
>
> Stephen Hemminger (2):
> ethdev: make flow API primary
The proposal has resulted from request to review [1] the following
functions where there appeared to be inconsistency in return type
or parameter type selections for the following inline functions.
rte_bsf32()
rte_bsf32_safe()
rte_bsf64()
rte_bsf64_safe()
rte_fls_u32()
rte_fls_u64()
rte_log2_u32()
Hi, Jan
1. Let's suppose we have the one port NIC, PCI BDF 08:00.0, SR-IOV enabled, 0
VFs is configured,
netdev name is ens1f0 (depends on distro and kernel version).
2. Then, we configure 3 VFs with SR-IOV, we get :
- 08:00.0 - PF, ens1f0
- 08:00.1 - VF0, ens1f1
- 08:00.2 - VF1, ens1f2
- 08:00
Set mutex used in failsafe driver to protect when used by
both primary and secondary process. Without this fix, the failsafe
lock is not really locking when there are multiple secondary processes.
Bugzilla ID: 662
Signed-off-by: Stephen Hemminger
Fixes: 655fcd68c7d2 ("net/failsafe: fix hotplug ra
This fixes two places where pthread_mutex was being unsafely
used between primary secondary process.
These patches are necessary but not sufficient to address Bug 662
Stephen Hemminger (2):
ethdev: make flow API primary/secondary process safe
net/failsafe: fix primary/secondary mutex
driver
Posix mutex are not by default safe for protecting for usage
from multiple processes. The flow ops mutex could be used by
both primary and secondary processes.
Bugzilla ID: 662
Signed-off-by: Stephen Hemminger
Fixes: 80d1a9aff7f6 ("ethdev: make flow API thread safe")
Cc: suanmi...@nvidia.com
---
https://bugs.dpdk.org/show_bug.cgi?id=662
Bug ID: 662
Summary: Pthread_mutex usage is not safe between
primar/secondary process
Product: DPDK
Version: 18.11
Hardware: All
OS: All
Status: UNCONFIR
> -Original Message-
> From: David Marchand
> Sent: Monday, March 15, 2021 4:17 PM
> To: Wang, Xiao W
> Cc: Xia, Chenbo ; Maxime Coquelin
> ; Liu, Yong ; dev
> ; Ananyev, Konstantin ; dpdk
> stable
> Subject: Re: [dpdk-stable] [PATCH v2] vhost: add header check in dequeue
> offload
OK, after firmware update, it seems that it works now.
Thank you for help!
Jan
On Mon, 15 Mar 2021 14:22:16 +0100
Jan Viktorin wrote:
> Hello Jiawei,
>
> On Fri, 12 Mar 2021 09:32:44 +
> "Jiawei(Jonny) Wang" wrote:
>
> > Hi Jan,
> >
> > > -Original Message-
> > > From: Jan Vi
Hi Wenzhuo
> -Original Message-
> From: dev On Behalf Of Wenzhuo Lu
> Sent: Friday, March 12, 2021 1:27 AM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo ; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH 3/3] net/i40e: fix segment fault in AVX512
>
> Fix segment fault when failing to get the memory f
Hi Wenzhuo
> -Original Message-
> From: dev On Behalf Of Wenzhuo Lu
> Sent: Friday, March 12, 2021 1:27 AM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo ; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH 2/3] net/ice: fix segment fault in AVX512
>
> Fix segment fault when failing to get the memory fr
On 2/25/2021 2:32 PM, Elad Nachman wrote:
This part of the series includes my fixes for the issues reported
by Ferruh and Igor (and Igor comments for v3 of the patch)
on top of part 1 of the patch series:
A. KNI sync lock is being locked while rtnl is held.
If two threads are calling kni_net_pro
On 2/25/2021 2:32 PM, Elad Nachman wrote:
This first part of v4 of the patch re-introduces Stephen Hemminger's
patch 64106 . This part changes the parameter kni_net_process_request()
gets and introduces the initial rtnl unlocking mechanism.
Signed-off-by: Elad Nachman
---
v4:
* for if down case
Hi Wenzhuo
> -Original Message-
> From: dev On Behalf Of Wenzhuo Lu
> Sent: Friday, March 12, 2021 1:27 AM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo ; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH 1/3] net/iavf: fix segment fault in AVX512
>
> Fix segment fault when failing to get the memory f
12/03/2021 15:56, Bruce Richardson:
> The meson.build files in each example directory is simply to support
> building the example as part of the main SDK build, and these should not
> be installed with the example's source code and makefile. The exclude of
> "meson.build" only filters out the top-l
On 3/15/21 5:57 PM, David Marchand wrote:
> On Mon, Mar 15, 2021 at 5:46 PM Maxime Coquelin
> wrote:
>> @@ -612,6 +622,8 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t
>> queue_idx)
>>
>> clean_vq:
>> hw->cvq = NULL;
>> + rte_free(fake_mbuf);
>> +free_sw_ring:
>>
On 15/03/2021 15:59, Stephen Hemminger wrote:
> On Mon, 15 Mar 2021 11:52:13 +0100
> Thomas Monjalon wrote:
>
>> 15/03/2021 11:42, Kinsella, Ray:
>>>
>>> On 15/03/2021 10:31, Bruce Richardson wrote:
On Mon, Mar 15, 2021 at 10:19:47AM +, Kinsella, Ray wrote:
>
>
> On 12
On Mon, Mar 15, 2021 at 5:46 PM Maxime Coquelin
wrote:
>
> This series optimizes the cache usage of virtqueue struct,
> by making a "fake" mbuf being dynamically allocated in Rx
> virtnet struct, by removing a useless virtuque pointer
> into the virtnet structs and by moving a few fields
> to pack
On 3/1/2021 4:38 PM, Stephen Hemminger wrote:
On Mon, 1 Mar 2021 11:10:01 +0300
Igor Ryzhov wrote:
Stephen,
No, I don't have a better proposal, but I think it is not correct to change
the behavior of KNI (making link down without a real response).
Even though we know that communicating with u
On Mon, Mar 15, 2021 at 5:46 PM Maxime Coquelin
wrote:
> @@ -612,6 +622,8 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t
> queue_idx)
>
> clean_vq:
> hw->cvq = NULL;
> + rte_free(fake_mbuf);
> +free_sw_ring:
> rte_free(sw_ring);
> free_hdr_mz:
> rte_memzone
This patch optimizes packing of the virtqueue
struct by moving fields around to fill holes.
Offset field is not used and so can be removed.
Signed-off-by: Maxime Coquelin
Reviewed-by: Chenbo Xia
---
drivers/net/virtio/virtqueue.h | 18 +-
1 file changed, 9 insertions(+), 9 dele
While it is worth clarifying whether the fake mbuf
in virtnet_rx struct is really necessary, it is sure
that it heavily impacts cache usage by being part of
the struct. Indeed, it uses two cachelines, and
requires alignment on a cacheline.
Before this series, it means it took 120 bytes in
virtnet_
This patch improves the error path of virtio_init_queue(),
by cleaning in reversing order all resources that have
been allocated.
Suggested-by: Chenbo Xia
Signed-off-by: Maxime Coquelin
---
drivers/net/virtio/virtio_ethdev.c | 15 ++-
1 file changed, 10 insertions(+), 5 deletions(-)
This series optimizes the cache usage of virtqueue struct,
by making a "fake" mbuf being dynamically allocated in Rx
virtnet struct, by removing a useless virtuque pointer
into the virtnet structs and by moving a few fields
to pack holes.
With these 3 patches, the virtqueue struct size goes from
5
Vrings are part of the virtqueues, so we don't need
to have a pointer to it in Vrings descriptions.
Instead, let's just subtract from its offset to
calculate virtqueue address.
Signed-off-by: Maxime Coquelin
Reviewed-by: Chenbo Xia
---
drivers/net/virtio/virtio_ethdev.c| 36 +++
> -Original Message-
> From: dev On Behalf Of Savinay Dharmappa
> Sent: Tuesday, March 9, 2021 4:10 PM
> To: Singh, Jasvinder ; Dumitrescu, Cristian
> ; dev@dpdk.org
> Cc: Dharmappa, Savinay
> Subject: [dpdk-dev] [PATCH v2] sched : Initialize tc ov watermark.
>
> tc ov watermark is ini
On 3/15/21 4:50 PM, David Marchand wrote:
> On Mon, Mar 15, 2021 at 4:20 PM Maxime Coquelin
> wrote:
>> @@ -550,10 +551,18 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t
>> queue_idx)
>> goto free_hdr_mz;
>> }
>>
>> + fake_mbuf = ma
From: Jan Viktorin
The step 4 is a contradiction. It advices to unbind the device from the
mlx5_core which removes the associated system network interface (e.g.
eth0). In the step 5, the same system network interface (e.g. eth0) is
required to exist.
Signed-off-by: Jan Viktorin
---
doc/guides/
From: Jan Viktorin
Signed-off-by: Jan Viktorin
---
doc/guides/nics/mlx5.rst | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst
index 7c50497fb..0a2dc3dee 100644
--- a/doc/guides/nics/mlx5.rst
+++ b/doc/guides/nics/mlx5.rst
Updates the documentation with the added support for sample actions VXLAN
and NVGRE encap in E-Switch steering flow.
Signed-off-by: Salem Sol
---
doc/guides/nics/mlx5.rst | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst
i
Update documentation for sample action usage in testpmd utilizing
rte_flow_action_vxlan_encap and rte_flow_action_nvgre_encap and
show the command line example.
Signed-off-by: Salem Sol
---
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 21 +
1 file changed, 21 insertions(+)
Add support for rte_flow_action_nvge_encap as a sample action.
The example of test-pmd command:
1. set nvgre ip-version ... tni ... ip-src ... ip-dst ...
set raw_encap 1 eth src... / ipv4... /...
set sample_actions 2 nvgre / port_id id 0 / end
flow create 0 ... pattern eth / end acti
Add support for NVGRE encap as a sample action
and validate it.
Signed-off-by: Salem Sol
---
drivers/net/mlx5/mlx5_flow_dv.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
index 4b2db47e39..590abdc822 100644
--- a/drivers/n
Add support for rte_flow_action_vxlan_encap as a sample action.
The example of test-pmd command:
1. set vxlan ip-version ... vni ... udp-src ...
set raw_encap 1 eth src.../ ipv4.../...
set sample_actions 2 vxlan_encap / port_id id 0 / end
flow create 0 ... pattern eth / end actions
Add support for VXLAN encap as a sample action
and validate it.
Signed-off-by: Salem Sol
---
drivers/net/mlx5/mlx5_flow_dv.c | 11 +++
1 file changed, 11 insertions(+)
diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
index 1a74d5ac2b..4b2db47e39 100644
---
From: Jiawei Wang
With the current code the VXLAN/NVGRE parsing routine
stored the configuration of the header on stack, this
might lead to overwriting the data on the stack.
This patch stores the external data of vxlan and nvgre encap
into global data as a pre-step to supporting vxlan and nvgre
This series adds support for VXLAN and NVGRE encap as a sample actions with the
proper
documentation, this series depends on [1] for the documentation part.
[1]
http://patches.dpdk.org/project/dpdk/patch/1615774238-51875-1-git-send-email-jiaw...@nvidia.com/
Jiawei Wang (1):
app/testpmd: sto
On Mon, Mar 15, 2021 at 4:52 PM Xiao Wang wrote:
>
> When parsing the virtio net header and packet header for dequeue offload,
> we need to perform sanity check on the packet header to ensure:
> - No out-of-boundary memory access.
> - The packet header and virtio_net header are valid and align
Hello Salva,
On Mon, 15 Mar 2021 15:53:51 +
Slava Ovsiienko wrote:
> Hi, Jan
>
> Yes, bullet [4] explicitly requires to unbind VFs, and detach the netdevs
> from the mlx5_core driver.
> Otherwise, kernel driver refuses to be configured with switchdev mode in [5].
> So, [4] can't be skippe
On Mon, 15 Mar 2021 11:52:13 +0100
Thomas Monjalon wrote:
> 15/03/2021 11:42, Kinsella, Ray:
> >
> > On 15/03/2021 10:31, Bruce Richardson wrote:
> > > On Mon, Mar 15, 2021 at 10:19:47AM +, Kinsella, Ray wrote:
> > >>
> > >>
> > >> On 12/03/2021 18:17, Thomas Monjalon wrote:
> > >>> Th
On 3/15/2021 1:58 PM, Andrew Rybchenko wrote:
Update base driver to provide functionality required by vDPA driver.
Factor out helper functions to be shared by net and vDPA drivers.
v2:
- fix windows build breakage - do not build common/sfc_efx in the case
of windows
- remove undefined e
Hi, Jan
Yes, bullet [4] explicitly requires to unbind VFs, and detach the netdevs from
the mlx5_core driver.
Otherwise, kernel driver refuses to be configured with switchdev mode in [5].
So, [4] can't be skipped.
After setting swithdev mode, VFs can be bound back (if it is needed, and these
one
When parsing the virtio net header and packet header for dequeue offload,
we need to perform sanity check on the packet header to ensure:
- No out-of-boundary memory access.
- The packet header and virtio_net header are valid and aligned.
Fixes: d0cf91303d73 ("vhost: add Tx offload capabilitie
On Mon, Mar 15, 2021 at 4:20 PM Maxime Coquelin
wrote:
>
> This patch optimizes packing of the virtuqueue
virtqueue ? and same typo in the title.
> struct by moving fields around to fill holes.
>
> Offset field is not used and so can be removed.
>
> Signed-off-by: Maxime Coquelin
> Reviewed-by:
On Mon, Mar 15, 2021 at 4:20 PM Maxime Coquelin
wrote:
> @@ -550,10 +551,18 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t
> queue_idx)
> goto free_hdr_mz;
> }
>
> + fake_mbuf = malloc(sizeof(*fake_mbuf));
> + if (!fake_m
On 3/15/21 4:38 PM, David Marchand wrote:
> On Mon, Mar 15, 2021 at 4:20 PM Maxime Coquelin
> wrote:
>> @@ -604,15 +604,22 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t
>> queue_idx)
>>
>> if (VIRTIO_OPS(hw)->setup_queue(hw, vq) < 0) {
>> PMD_INIT_LOG(ERR, "set
On Mon, Mar 15, 2021 at 4:20 PM Maxime Coquelin
wrote:
> @@ -604,15 +604,22 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t
> queue_idx)
>
> if (VIRTIO_OPS(hw)->setup_queue(hw, vq) < 0) {
> PMD_INIT_LOG(ERR, "setup_queue failed");
> - return -EINVAL;
>
> Subject: Re: [PATCH v4 1/2] eal: error number enhancement for thread TLS
> API
>
> On Wed, Mar 10, 2021 at 02:48:55PM +0200, Tal Shnaiderman wrote:
> > add error number reporting to rte_errno in all functions in the
> > rte_thread_tls_* API.
> >
> > Suggested-by: Anatoly Burakov
> > Signed-off-
use UINT64_MAX instead of -1ULL when manipulating uint64_t masks and
initializing sentinel values.
some compilers generate a warning when applying a '-' to an unsigned
literal so avoid this by initializing with unsigned preprocessor
definitions where appropriate.
Signed-off-by: Tyler Retzlaff
--
Support the PMD power management API in MLX5 driver.
The monitor policy of this API puts a CPU core to sleep until
a data in some monitored memory address is changed by the NIC.
Implement the get_monitor_addr function to return an address
of a CQE owner bit to monitor the arrival of a new packet.
On 2021/3/15 18:19, David Marchand wrote:
#else
#define IO_COND(addr, is_pio, is_mmio) do { \
is_mmio; \
} while (0)
#endif
We should not just copy/paste kernel code.
Plus here, this seems a bit overkill.
And there are other parts in thi
This patch optimizes packing of the virtuqueue
struct by moving fields around to fill holes.
Offset field is not used and so can be removed.
Signed-off-by: Maxime Coquelin
Reviewed-by: Chenbo Xia
---
drivers/net/virtio/virtqueue.h | 18 +-
1 file changed, 9 insertions(+), 9 del
While it is worth clarifying whether the fake mbuf
in virtnet_rx struct is really necessary, it is sure
that it heavily impacts cache usage by being part of
the struct. Indeed, it uses two cachelines, and
requires alignement on a cacheline.
Before this series, it means it took 120 bytes in
virtnet
Vrings are part of the virtqueues, so we don't need
to have a pointer to it in Vrings descriptions.
Instead, let's just subtract from its offset to
calculate virtqueue address.
Signed-off-by: Maxime Coquelin
Reviewed-by: Chenbo Xia
---
drivers/net/virtio/virtio_ethdev.c| 36 +++
This patch improves the error path of virtio_init_queue(),
by cleaning in reversing order all resources that have
been allocated.
Suggested-by: Chenbo Xia
Signed-off-by: Maxime Coquelin
---
drivers/net/virtio/virtio_ethdev.c | 19 +--
1 file changed, 13 insertions(+), 6 deletion
This series optimizes the cache usage of virtqueue struct,
by making a "fake" mbuf being dynamically allocated in Rx
virtnet struct, by removing a useless virtuque pointer
into the virtnet structs and by moving a few fields
to pack holes.
With these 3 patches, the virtqueue struct size goes from
5
> -Original Message-
> From: dev On Behalf Of Mattias Rönnblom
> Sent: Monday, March 15, 2021 2:45 PM
> To: Jerin Jacob
> Cc: Jerin Jacob ; dpdk-dev ; Richardson,
> Bruce
> Subject: Re: [dpdk-dev] [RFC] eventdev: introduce event dispatcher
>
> On 2021-03-07 14:04, Jerin Jacob wrote:
> >
On 2021-03-07 14:04, Jerin Jacob wrote:
> On Fri, Feb 26, 2021 at 1:31 PM Mattias Rönnblom
> wrote:
>> On 2021-02-25 13:32, Jerin Jacob wrote:
>>> On Fri, Feb 19, 2021 at 12:00 AM Mattias Rönnblom
>>> wrote:
The purpose of the event dispatcher is primarily to decouple different
parts of
10/03/2021 23:20, Ferruh Yigit:
> Signed-off-by: Ferruh Yigit
> ---
> v2:
> * there won't be a new driver for dlb2.5
> * reword thash library support
Applied with minor updates for sorting things, thanks.
On Wed, Mar 10, 2021 at 6:37 PM 谢华伟(此时此刻） wrote:
>
> virtio PMD assumes legacy device only supports PIO(port-mapped) BAR
> resource. This is wrong. As we need to create lots of devices, adn PIO
> resource on x86 is very limited, we expose MMIO(memory-mapped I/O) BAR.
>
> Kernel supports both PIO a
From: Vijay Kumar Srivastava
Move function to get efx family from net driver into common driver.
Signed-off-by: Vijay Kumar Srivastava
Signed-off-by: Andrew Rybchenko
---
drivers/common/meson.build | 2 +-
drivers/common/sfc_efx/meson.build | 5 +++
drivers/common/sfc_efx/sfc_efx.c
On 3/14/21 3:36 AM, Ferruh Yigit wrote:
> On 3/11/2021 11:03 AM, Andrew Rybchenko wrote:
>> From: Vijay Kumar Srivastava
>>
>> Move function to get efx family from net driver into common driver.
>>
>> Signed-off-by: Vijay Kumar Srivastava
>> Signed-off-by: Andrew Rybchenko
>
> <...>
>
>> diff --
From: Vijay Kumar Srivastava
Device class argument would be used to select compatible driver.
Driver probe would be skipped for incompatible device class.
Signed-off-by: Vijay Kumar Srivastava
Signed-off-by: Andrew Rybchenko
---
drivers/common/sfc_efx/sfc_efx.c | 49
1 - 100 of 153 matches
Mail list logo
