> From: Morten Brørup [mailto:m...@smartsharesystems.com]
>
> > From: Gregory Etelson [mailto:getel...@nvidia.com]
> > Sent: Tuesday, 18 June 2024 07.18
> >
> > DPDK IPv6 header definition combined the `version`, `traffic class`
> > and `flow label` header fields into a single 32 bits structure me
On Tuesday, June 18, 2024 2:49 PM, dmarc...@redhat.com wrote:
> On Mon, Jun 17, 2024 at 11:54 AM Jiawen Wu wrote:
> >
> > Support WOL and NCSI capability for devices. And there is one OEM
> > NCSI NIC which can not be identified from sub-system ID, it needs
> > to check NCSI pin status in firmware
The problem of Tx hang also occurs on Wangxun 10Gb NICs, when stop device
under heavy traffic.
refer to commit ac6c5e9af56a ("net/ngbe: fix Tx hang on queue disable")
Disable PCIe bus master to clear BME when stop hardware, and verify there
are no pending requests. Move disabling Tx queue after d
The outer-ipv6 tunnel packet was parsed to the wrong packet type, remove
the default RTE_PTYPE_L2_ETHER and RTE_PTYPE_L3_IPV4 flags for tunnel
packets. And correct the calculation of tunnel length for GRE and GENEVE
packets.
Fixes: ca46fcd753b1 ("net/txgbe: support Tx with hardware offload")
Fixes
In virtualization mode, target pool should be determined for the
filters. For ether type filter, virtualization mode must be enabled
to filter broadcast/multicast packets due to hardware limitations.
Fixes: f8e2cfc7702b ("net/txgbe: support ethertype filter add and delete")
Fixes: 77a72b4d9dc0 ("n
There is a hardware limitation that Rx ring config register is not writable
when Rx ring is enabled, i.e. the TXGBE_RXCFG_ENA bit is set. But disabling
the ring when there is traffic will cause ring get stuck. So restrict the
configuration of VLAN strip offload only if device is started.
Fixes: 22
YT8531SH-CA PHY will switch to SDS space automatically when UTP and
SDS media are not present, causing failure to link up. Add the special
configuration to fix it.
Fixes: 3d0af7066759 ("net/ngbe: setup PHY link")
Fixes: 1c44384fce76 ("net/ngbe: support custom PHY interfaces")
Cc: sta...@dpdk.org
The configuration of allmulti and promiscuous modes conflicts
together. For instance, if we enable promiscuous mode, then enable and
disable allmulti, then the promiscuous mode is wrongly disabled.
Fix this behavior by:
- doing nothing when we set/unset allmulti if promiscuous mode is on
- restort
The internal PHY will be set to default power down after LAN reset,
but the external PHY will not. To keep the PHY behavior consistent,
set PHY power down uniformly here.
Fixes: 708ebe7d0399 ("net/ngbe: fix external PHY power down")
Cc: sta...@dpdk.org
Signed-off-by: Jiawen Wu
---
drivers/net/n
This bug occurs in OpenvSwitch. After adding a port bound to vfio-pci
to ovs, detach it from ovs and then unbind it from vfio-pci(that is
hotplug) will cause operating system to get stuck.
Fixes: 7dc117068a7c ("net/txgbe: support probe and remove")
Cc: sta...@dpdk.org
Signed-off-by: Jiawen Wu
--
This bug occurs in OpenvSwitch. After adding a port bound to vfio-pci
to ovs, detach it from ovs and then unbind it from vfio-pci(that is
hotplug) will cause operating system to get stuck.
Fixes: 6ee7e574cd48 ("net/ngbe: support probe and remove")
Cc: sta...@dpdk.org
Signed-off-by: Jiawen Wu
---
Support WOL and NCSI capability for devices. And there is one OEM
NCSI NIC which can not be identified from sub-system ID, it needs
to check NCSI pin status in firmware.
Signed-off-by: Jiawen Wu
---
drivers/net/ngbe/base/ngbe_hw.c | 30 ++--
drivers/net/ngbe/base/ngb
The valid range of MTU is 68 to 9414. Set min_mtu and max_mtu in dev_info.
Fixes: 3926214fd80d ("net/txgbe: support MTU set")
Cc: sta...@dpdk.org
Signed-off-by: Jiawen Wu
---
drivers/net/txgbe/txgbe_ethdev.c | 12 +---
drivers/net/txgbe/txgbe_ethdev.h | 2 +-
2 files changed, 6 inserti
The valid range of MTU is 68 to 9414. Set min_mtu and max_mtu in dev_info.
Fixes: 07baabb6a51a ("net/ngbe: support MTU set")
Cc: sta...@dpdk.org
Signed-off-by: Jiawen Wu
---
drivers/net/ngbe/ngbe_ethdev.c | 4 +++-
drivers/net/ngbe/ngbe_ethdev.h | 1 +
2 files changed, 4 insertions(+), 1 deleti
Fix some memory leaks caused by not release resource in time.
Fixes: e1698e383c2a ("net/txgbe: add device init and uninit")
Fixes: 635c21354f9a ("net/txgbe: add flow director filter init and uninit")
Fixes: c13f84a71b2d ("net/txgbe: add L2 tunnel filter init and uninit")
Fixes: 3a123ba60a71 ("net/
Fix Rx interrupt enable failure.
Fixes: a5682d28f134 ("net/txgbe: support Rx interrupt")
Cc: sta...@dpdk.org
Signed-off-by: Jiawen Wu
---
drivers/net/txgbe/txgbe_ethdev.c| 18 +-
drivers/net/txgbe/txgbe_ethdev_vf.c | 2 +-
2 files changed, 10 insertions(+), 10 deletions(-)
In the new firmware versions, LLDP is enabled by default to implement
new features in other drivers. But it is useless in DPDK. So disable
it in device initialization to prevent it from affecting hardware
default behavior.
Signed-off-by: Jiawen Wu
---
drivers/net/txgbe/base/txgbe_eeprom.h | 3 +
Implement Rx queue interrupt enable/disable functions.
Signed-off-by: Jiawen Wu
---
doc/guides/nics/features/ngbe.ini | 1 +
doc/guides/nics/ngbe.rst | 1 +
drivers/net/ngbe/ngbe_ethdev.c| 33 ++-
3 files changed, 34 insertions(+), 1 deletion(-)
diff -
Fix some memory leaks caused by not release resource in time.
Fixes: 43b7e5ea60ac ("net/ngbe: support Rx queue setup/release")
Fixes: a58e7c312c6b ("net/ngbe: support Tx queue setup/release")
Cc: sta...@dpdk.org
Signed-off-by: Jiawen Wu
---
drivers/net/ngbe/ngbe_rxtx.c | 8
drivers/net
In the new firmware versions, LLDP is enabled by default to implement
new features in other drivers. But it is useless in DPDK. So disable
it in device initialization to prevent it from affecting hardware
default behavior.
Signed-off-by: Jiawen Wu
---
drivers/net/ngbe/base/ngbe_eeprom.h | 4 ++
When link status changes, there is a probability that no more packets can
be received on the port, due to hardware defects. These MAC Rx registers
should be reconfigured to fix this problem.
Fixes: 950a6954df13 ("net/txgbe: reconfigure MAC Rx when link update")
Cc: sta...@dpdk.org
Signed-off-by:
On 6/17/24 08:26, Chaoyong He wrote:
From: Xinying Yu
Add new contributor.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zhang
---
.mailmap | 1 +
1 file changed, 1 insertion(+)
diff --git a/.mailmap b/.mailmap
index 6b396107d0..01c31b8c70
Hi,
The series looks good to me, will apply it shortly.
For next time, when sending a new revision, could you please apply
all the R-By tags provided by reviewers on previous revision if the
patch did not change?
Thanks,
Maxime
On 6/17/24 08:26, Chaoyong He wrote:
This patch series aims to ad
> -Original Message-
> From: Maxime Coquelin
> Sent: Tuesday, June 18, 2024 3:46 PM
> To: Chaoyong He ; dev@dpdk.org
> Cc: oss-drivers
> Subject: Re: [PATCH v3 00/11] support software live migration
>
> Hi,
>
> The series looks good to me, will apply it shortly.
>
> For next time, wh
On 6/17/24 08:26, Chaoyong He wrote:
From: Xinying Yu
Reconfigure the NIC will fail because lack of the
initialization logic of queue configuration pointer.
Fix this by adding the correct initialization logic.
Fixes: d89f4990c14e ("vdpa/nfp: add hardware init")
Cc: chaoyong...@corigine.com
On 6/17/24 08:27, Chaoyong He wrote:
From: Xinying Yu
The ctrl words of vDPA is located on the extended word, so it
should use the 'nfp_ext_reconfig()' rather than 'nfp_reconfig()'.
Also replace the misuse of 'NFP_NET_CFG_CTRL_SCATTER' macro
with 'NFP_NET_CFG_CTRL_VIRTIO'.
Fixes: b47a03739
DPDK IPv6 header definition combined the `version`, `traffic class`
and `flow label` header fields into a single 32 bits structure member
`vtc_flow`.
The patch expands IPv6 header definition with dedicated structure
members for the `version`, `traffic class` and `flow label` fields.
The `traffic c
Hello,
I've posted v2 patch with fixes.
Regards,
Gregory
> From: Gregory Etelson [mailto:getel...@nvidia.com]
> Sent: Tuesday, 18 June 2024 09.58
>
> DPDK IPv6 header definition combined the `version`, `traffic class`
> and `flow label` header fields into a single 32 bits structure member
> `vtc_flow`.
>
> The patch expands IPv6 header definition with
On 6/17/24 08:27, Chaoyong He wrote:
From: Xinying Yu
Add the basic logic of software live migration.
Unset the ring notify area to stop the direct IO datapath if the
device support, then we can setup the vring relay to help the
live migration.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoy
On 6/17/24 08:27, Chaoyong He wrote:
From: Xinying Yu
Add the interrupt setup logic of vring relay.
The epoll fd is provided here so host can get the interrupt from device
on Rx direction, all other operations on vring relay are based on this.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoyo
On 6/17/24 08:27, Chaoyong He wrote:
From: Xinying Yu
In order to add the new configuration logic of software live
migration, split the datapath update logic into two parts,
queue configuration and VF configuration.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
On 6/17/24 08:27, Chaoyong He wrote:
From: Xinying Yu
Create the relay vring on host and then set the address of Rx
used ring to the VF config bar. So the device can DMA the
used ring information to host rather than directly to VM.
Use 'NFP_NET_CFG_CTRL_LM_RELAY' notify the device side. And
On 6/17/24 08:27, Chaoyong He wrote:
From: Xinying Yu
After migrating to new host, the vring information is
recovered by the value in offset 'NFP_NET_CFG_TX_USED_INDEX'
and 'NFP_NET_CFG_RX_USED_INDEX'.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: P
On 6/17/24 08:27, Chaoyong He wrote:
From: Xinying Yu
Setup the vring relay thread to monitor the interruption from
device. And do the dirty page logging or notify device according
to event data.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zh
On 6/17/24 08:27, Chaoyong He wrote:
From: Xinying Yu
Add the 'VHOST_F_LOG_ALL' feature bits inorder to enable the
live migration function.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zhang
---
drivers/vdpa/nfp/nfp_vdpa_core.c | 1 +
1 fi
On 6/17/24 08:27, Chaoyong He wrote:
From: Xinying Yu
Add the software assisted vDPA live migration feature
into NFP document.
Signed-off-by: Xinying Yu
Reviewed-by: Chaoyong He
Reviewed-by: Long Wu
Reviewed-by: Peng Zhang
---
doc/guides/vdpadevs/nfp.rst | 9 +
1 file changed
On 17. 6. 2024 13:44, Luca Vizzarro wrote:
On 06/06/2024 10:19, Juraj Linkeš wrote:
The static method in the other patch is called compose and does
essentially the same thing, right? Can we use the same name (or a
similar one)?
Also, what is the difference in approaches between the two pat
On 17. 6. 2024 14:13, Luca Vizzarro wrote:
On 06/06/2024 19:03, Juraj Linkeš wrote:
+class DPDKShell(InteractiveShell, ABC):
+ """The base class for managing DPDK-based interactive shells.
+
+ This class shouldn't be instantiated directly, but instead be
extended.
+ It automatically
The graph id is determined based on a global variable that is
incremented every time a graph is created, and decremented every time
a graph is destroyed. This only works if graphs are destroyed in the
reverse order in which they have been created.
The following code produces duplicate graph IDs wh
From: Robin Jarry
Sent: Tuesday, June 18, 2024 2:53 PM
To: dev@dpdk.org; Jerin Jacob ; Kiran Kumar Kokkilagadda
; Nithin Kumar Dabilpuram ;
Zhirun Yan
Subject: [EXTERNAL] [PATCH v3] graph: avoid id collisions
The graph id is determined based on a global variable that is incremented every
ti
DPDK IPv6 header definition combined the `version`, `traffic class`
and `flow label` header fields into a single 32 bits structure member
`vtc_flow`.
The patch expands IPv6 header definition with dedicated structure
members for the `version`, `traffic class` and `flow label` fields.
The `traffic c
https://bugs.dpdk.org/show_bug.cgi?id=1463
Bug ID: 1463
Summary: disable vhost guest notification not work when split
queue and VIRTIO_RING_F_EVENT_IDX negotiated
Product: DPDK
Version: unspecified
Hardware: x86
Release status meeting minutes 2024-06-13
=
Agenda:
* Release Dates
* Subtrees
* Roadmaps
* LTS
* Defects
* Opens
Participants:
* AMD [No]
* ARM
* Debian/Microsoft [No]
* Intel
* Marvell
* Nvidia
* Red Hat
Release Dates
-
The following are t
> struct rte_ipv6_hdr {
> - rte_be32_t vtc_flow;/**< IP version, traffic class & flow label.
> */
> + union {
> + rte_be32_t vtc_flow;/**< IP version, traffic class &
> flow
> label. */
> + __extension__
> + struct {
> +#if RTE_BYTE_ORDER ==
> -Original Message-
> From: Dariusz Sosnowski
> Sent: Wednesday, May 29, 2024 2:21 PM
> To: Slava Ovsiienko ; Ori Kam ;
> Suanming Mou ; Matan Azrad
> ; Bing Zhao ; NBU-Contact-Thomas
> Monjalon (EXTERNAL)
> Cc: dev@dpdk.org
> Subject: [PATCH] maintainers: update for networking mlx5 driv
Update IDPF VF device id to 145C.
Also added device ID for S-IOV device.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/idpf_devids.h | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/common/idpf/base/idpf_devids.h
b/drivers/common/idpf/base/idpf_dev
These patches integrate the latest changes in MEV TS IDPF Base driver.
---
v4:
- Removed 1st patch as we are not using NVME_CPF flag
- Addressed comments
---
v3:
- Removed additional whitespace changes
- Fixed warnings of CI
- Updated documentation relating to MEV TS FW release
---
v2:
- Changed
Removed unused VIRTCHNL2_CAP_ADQ capability and use that bit for
VIRTCHNL2_CAP_INLINE_FLOW_STEER capability.
Added VIRTCHNL2_VPORT_INLINE_FLOW_STEER_ENA port flag to allow
enable/disable per vport.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/virtchnl2.h | 7 ---
1 file chang
There is an issue of recursive header file includes in accessing the
idpf_hw structure. The controlq.h has the structure definition and osdep
header file needs that. The problem is the controlq.h also needs
the osdep header file contents, basically both dependent on each other.
Moving the definiti
Based on the upstream feedback, driver should not use any
defensive programming strategy by checking for NULL pointers
and other conditional checks unnecessarily in the code flow
to fall back, instead fail and fix the bug in a proper way.
As the control queue is freed and deleted from the list aft
For bitmaps greater than 32 bits, use BIT_ULL instead of BIT
macro as reported by compiler.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/virtchnl2.h | 70 ++--
1 file changed, 35 insertions(+), 35 deletions(-)
diff --git a/drivers/common/idpf/base/virtchnl
'u32' data type is used for the struct members in
'virtchnl2_version_info' which should be '__le32'.
Make the change accordingly.
It is a Little Endian specific type defination.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/virtchnl2.h | 4 ++--
1 file changed, 2 insertions(+), 2
Instead of using the long RXDID definitions, introduce a
macro which uses common part of the RXDID definitions i.e.
VIRTCHNL2_RXDID_ and the bit passed to generate a mask.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/virtchnl2_lan_desc.h | 31 ++-
1 file changed, 1
Instead of using 'divide by 0' to check the struct length,
use the static_assert macro
Removed redundant CHECK_UNION_LEN macro.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/virtchnl2.h | 13 +
1 file changed, 5 insertions(+), 8 deletions(-)
diff --git a/drivers/commo
Mask for VIRTCHNL2_RX_FLEX_DESC_ADV_FF1_M was defined wrongly
and this patch fixes it.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/virtchnl2_lan_desc.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/common/idpf/base/virtchnl2_lan_desc.h
b/drivers/com
'pad' naming is used if the field is actually a padding byte
and is also used for bytes meant for future addition of new
fields, whereas 'reserved' is only used if the field is reserved
and cannot be used for any other purpose.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/virtchnl
Changes all groups of related defines to enums. The names of
the enums are chosen to follow the common part of the naming
pattern as much as possible.
Replaced the common labels from the comments with the enum names.
While at it, modify header description based on upstream feedback.
Some variabl
Don't initialize the variables if not needed.
Also use 'err' instead of 'status', 'ret_code', 'ret' etc.
for consistency and change the return label 'sq_send_command_out'
to 'err_unlock'.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/idpf_controlq.c | 60 +--
When the message for getting timestamp latches is sent by the driver,
number of latches is equal to 0. Current implementation of message
validation function incorrectly notifies this kind of message length as
invalid.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/virtchnl2.h | 4 ++
This capability bit indicates both inline as well as side band flow
steering capability.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/virtchnl2.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/common/idpf/base/virtchnl2.h
b/drivers/common/idpf/base/vi
Introduced through customer's feedback in their attempt to address some
bugs this introduces a memory barrier before posting ctlq tail. This
makes sure memory writes have a chance to take place before HW starts
messing with the descriptors.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/
With the internal Linux upstream feedback that is received on
IDPF driver and also some references available online, it
is discouraged to use 1-sized array fields in the structures,
especially in the new Linux drivers that are going to be
upstreamed. Instead, it is recommended to use flex array fie
Added virtchnl2_flow_types to be used for flow steering.
Added flow steer cap flags for vport create.
Add flow steer flow types and action types for vport create.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/virtchnl2.h | 60 ++--
1 file changed, 57 inser
Adding a new structure for the context descriptor that contains
the support for timesync packets, where the index for timestamping is set.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/idpf_lan_txrx.h | 20 +++-
1 file changed, 19 insertions(+), 1 deletion(-)
diff
Adding an argument named type to define queue type
in idpf_vc_queue_switch(). This solves the issue of
improper queue type in virtchnl2 message.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/idpf_common_virtchnl.c | 8 ++--
drivers/common/idpf/idpf_common_virtchnl.h | 2 +-
driver
The file is redundant in our implementation and is not required
further.
Signed-off-by: Soumyadeep Hore
---
drivers/common/idpf/base/idpf_common.c | 382 -
drivers/common/idpf/base/meson.build | 1 -
2 files changed, 383 deletions(-)
delete mode 100644 drivers/common
Updated the latest support for cpfl pmd in MEV TS
firmware version which is 1.4.
Signed-off-by: Soumyadeep Hore
---
doc/guides/nics/cpfl.rst | 2 ++
1 file changed, 2 insertions(+)
diff --git a/doc/guides/nics/cpfl.rst b/doc/guides/nics/cpfl.rst
index 9b7a99c894..528c809819 100644
--- a/doc/gui
> -Original Message-
> From: Slava Ovsiienko
> Sent: Tuesday, June 18, 2024 7:17 PM
> To: Dariusz Sosnowski ; Ori Kam
> ; Suanming Mou ; Matan
> Azrad ; Bing Zhao ;
> NBU-Contact-Thomas Monjalon (EXTERNAL)
> Cc: dev@dpdk.org
> Subject: RE: [PATCH] maintainers: update for networking mlx
On Mon, May 27, 2024 at 07:19:21PM +0300, Igor Gutorov wrote:
> According to the Intel X710/XXV710/XL710 Datasheet, the maximum receive
> queue descriptor length is 0x1FE0 (8160 in base 10). This is specified
> as QLEN in table 8-12, page 1083.
>
> I've tested this change with an XXV710 NIC and it
Hi Morten,
Thanks for your review.
在 2024/6/14 16:04, Morten Brørup 写道:
+#define PM_QOS_SYSFILE_RESUME_LATENCY_US \
+ "/sys/devices/system/cpu/cpu%u/power/pm_qos_resume_latency_us"
Is it OK to access this path using the lcore_id as CPU parameter to
open_core_sysfs_file(), or must
DPDK IPv6 header definition combined the `version`, `traffic class`
and `flow label` header fields into a single 32 bits structure member
`vtc_flow`.
The patch expands IPv6 header definition with dedicated structure
members for the `version`, `traffic class` and `flow label` fields.
The `traffic c
Re Robin,
On Wed, Mar 27, 2024 at 10:17 AM Robin Jarry wrote:
>
> In some cases, the node context data is used to store two pointers
> because the data is larger than the reserved 16 bytes. Having to define
> intermediate structures just to be able to cast is tedious. And without
> intermediate s
> From: Gregory Etelson [mailto:getel...@nvidia.com]
>
> DPDK IPv6 header definition combined the `version`, `traffic class`
> and `flow label` header fields into a single 32 bits structure member
> `vtc_flow`.
>
> The patch expands IPv6 header definition with dedicated structure
> members for th
Similarly to libraries, when setting a list of desired applications,
report if some application cannot be built because of a missing
dependency.
Signed-off-by: David Marchand
---
app/meson.build | 10 ++
1 file changed, 10 insertions(+)
diff --git a/app/meson.build b/app/meson.build
ind
> From: lihuisong (C) [mailto:lihuis...@huawei.com]
>
> Hi Morten,
>
> Thanks for your review.
>
>
> 在 2024/6/14 16:04, Morten Brørup 写道:
> >> +#define PM_QOS_SYSFILE_RESUME_LATENCY_US \
> >> + "/sys/devices/system/cpu/cpu%u/power/pm_qos_resume_latency_us"
> > Is it OK to access this path usi
On Tue, Jun 18, 2024 at 02:51:12PM +0200, David Marchand wrote:
> Similarly to libraries, when setting a list of desired applications,
> report if some application cannot be built because of a missing
> dependency.
>
> Signed-off-by: David Marchand
> ---
> app/meson.build | 10 ++
> 1 fi
On Mon, Jun 17, 2024 at 4:49 PM Stephen Hemminger
wrote:
>
> On Fri, 31 May 2024 10:41:18 -0700
> Stephen Hemminger wrote:
>
> > This reverts commit 628362c94a0b567a39a0177539c12c97d999.
> >
> > That commit broke use of dumpcap with vdev's and probably
> > other uses of secondary processes wi
The rte_vhost_driver_unregister() vhost_user_read_cb()
vhost_user_client_reconnect() can be called at the same time by 3 threads.
when memory of vsocket is freed in rte_vhost_driver_unregister(),
then vhost_user_read_cb() maybe add vsocket to reconn_list,
the invalid memory of vsocket is accessed
On Tue, Jun 18, 2024 at 11:24 AM Robin Jarry wrote:
>
> The graph id is determined based on a global variable that is
> incremented every time a graph is created, and decremented every time
> a graph is destroyed. This only works if graphs are destroyed in the
> reverse order in which they have be
On Mon, Apr 1, 2024 at 10:37 PM Robin Jarry wrote:
>
> In rte_graph_cluster_stats_get, the walk model of the first graph is
> checked to determine if multi-core dispatch specific counters should be
> updated or not. This global list is accessed without any locks.
>
> If the global list is modified
On Wed, Mar 20, 2024 at 6:11 PM Robin Jarry wrote:
>
> * Quote graph name to avoid parsing errors when it contains a dash.
> * Use fixed margin and smaller font for a more compact layout.
> * Use sans-serif font, the default is times new roman which is not the
> best choice for a packet processi
On Mon, Jun 10, 2024 at 10:56:00AM +0200, Morten Brørup wrote:
> PING (again) for review.
>
> Many applications use bursts of more than 32 packets,
> and some applications buffer more than 512 packets.
>
> This patch updates the mempool perf test accordingly.
>
> -Morten
>
> > From: Morten Brør
On Thu, Apr 25, 2024 at 5:28 PM Stephen Hemminger
wrote:
> > As per Intel, this is not supported, and the librte-bpf test fails on 32bit
> > x86
> > kernels, so disable the library and the pmd.
Cc: sta...@dpdk.org
> >
> > Signed-off-by: Luca Boccassi
> Acked-by: Stephen Hemminger
Applied, th
On Fri, May 10, 2024 at 10:40 AM Kiran Kumar Kokkilagadda
wrote:
> > When the order of graph destroy is not the reverse order of create, that is,
> > when it is destroyed at will, the newly created graph id will be the same as
> > the existing graph id, which is not the expected unique graph id. T
> From: Bruce Richardson [mailto:bruce.richard...@intel.com]
>
> On Mon, Jun 10, 2024 at 10:56:00AM +0200, Morten Brørup wrote:
> > PING (again) for review.
> >
> > Many applications use bursts of more than 32 packets,
> > and some applications buffer more than 512 packets.
> >
> > This patch upda
On Mon, May 20, 2024 at 8:51 PM Honnappa Nagarahalli
wrote:
> > On May 15, 2024, at 5:54 AM, Abdullah Ömer Yamaç
> > wrote:
> >
> > This patch adds a new feature to the hash library to allow the user to
> > reclaim the defer queue. This is useful when the user wants to force
> > reclaim resource
On Tue, May 14, 2024 at 10:52 PM Honnappa Nagarahalli
wrote:
> > On May 13, 2024, at 11:35 AM, Abdullah Ömer Yamaç
> > wrote:
> >
> > Set the maximum reclamation size to user provided value
> >
> > Fixes: 769b2de7fb52 ("hash: implement RCU resources reclamation")
> > Cc: sta...@dpdk.org
> >
> >
When using ring based ethdev, mbuf->port is not set on received packets.
For applications that use the mbuf->port to identify the incoming port,
especially when eventdev RX adapter pulls the packet on a different core
and the application running on a worker core has no clue on the incoming
port. T
The type parameter for malloc is only used for tracing.
Fix documentation and don't pass through heap routines.
Stephen Hemminger (2):
rte_malloc: document that type is for tracing
eal: remove type argument from internal routines
v3 - drop event/sw patch (already fixed)
lib/eal/common/eal_c
The string type is only used for tracing and not used as
documented by dump_stats.
Signed-off-by: Stephen Hemminger
---
lib/eal/include/rte_malloc.h | 27 +--
1 file changed, 13 insertions(+), 14 deletions(-)
diff --git a/lib/eal/include/rte_malloc.h b/lib/eal/include/rt
The type argument is carried through malloc heap routines but
never used there. It is only used a rte_malloc for tracing.
Signed-off-by: Stephen Hemminger
---
lib/eal/common/eal_common_memzone.c | 6 ++---
lib/eal/common/malloc_heap.c| 39 -
lib/eal/common/ma
On Wed, Jun 12, 2024 at 04:00:39PM +0100, Anatoly Burakov wrote:
> From: Ian Stokes
>
> If FW supports the corresponding functionality, the driver allows PFs to
> subscribe the same switch recipes. Then when the PF is done with a switch
> recipe, the PF can ask the FW to free that switch recipe.
On Wed, Jun 12, 2024 at 04:00:49PM +0100, Anatoly Burakov wrote:
> From: Ian Stokes
>
> Code using ice_ptp_one_port_cmd() (or the device specific variants for E822
> and
> ETH56G) has a subtle bug where-in the executing of ice_ptp_exec_tmr_cmd() will
> cause all other ports to execute their last
On Wed, Jun 12, 2024 at 04:00:51PM +0100, Anatoly Burakov wrote:
> From: Ian Stokes
>
> Replace enum ice_status with int. Also replcae ICE_SUCCESS instances with 0.
>
> Signed-off-by: Przemek Kitszel
> Signed-off-by: Ian Stokes
> ---
> drivers/net/ice/base/ice_common.c | 12 ++--
> drivers/ne
Tweaked logic on sending and verifying packets for
more concise code, added verbose and promisc
function calls from pmd shell module.
Signed-off-by: Dean Marx
---
dts/tests/TestSuite_vlan.py | 59 +
1 file changed, 27 insertions(+), 32 deletions(-)
diff --g
Ported over the promisc and verbose mode functions from
v2 of the queue start/stop suite to use for the VLAN suite.
Tweaked some of the verification methods to be
more concise, changed some docstrings to be more specific.
Signed-off-by: Dean Marx
---
dts/framework/remote_session/testpmd_shell.
Configuration to run vlan test suite
Signed-off-by: Dean Marx
---
dts/framework/config/conf_yaml_schema.json | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/dts/framework/config/conf_yaml_schema.json
b/dts/framework/config/conf_yaml_schema.json
index 4731f4511d..eca8244f27
On 13. 6. 2024 20:15, jspew...@iol.unh.edu wrote:
From: Jeremy Spewock
Interactive shells are managed in a way currently where they are closed
and cleaned up at the time of garbage collection. Due to there being no
guarantee of when this garbage collection happens in Python, there is no
way
Add the option to create one session for the PMD, and share it across
all of the queue pairs. This may help to discover/debug concurrency
issues (both correctness and performance) that can occur when using this
configuration.
Signed-off-by: Jack Bond-Preston
Acked-by: Brian Dooley
Reviewed-by: W
1 - 100 of 167 matches
Mail list logo
