From: Beilei Xing
This patchset adds hairpin queue support.
It depends on the following two patch set:
https://patches.dpdk.org/project/dpdk/cover/20230406074245.82991-1-beilei.x...@intel.com/
https://patches.dpdk.org/project/dpdk/cover/20230413094502.1714755-1-wenjing.q...@intel.com/
Beilei Xin
From: Beilei Xing
This patch refines some structures to support hairpin queue,
cpfl_rx_queue/cpfl_tx_queue/cpfl_vport.
Signed-off-by: Mingxia Liu
Signed-off-by: Beilei Xing
---
drivers/net/cpfl/cpfl_ethdev.c | 85 +++-
drivers/net/cpfl/cpfl_ethdev.h | 6 +-
driver
From: Beilei Xing
This patch adds hairpin_cap_get ops support.
Signed-off-by: Xiao Wang
Signed-off-by: Mingxia Liu
Signed-off-by: Beilei Xing
---
drivers/net/cpfl/cpfl_ethdev.c | 13 +
drivers/net/cpfl/cpfl_rxtx.h | 4
2 files changed, 17 insertions(+)
diff --git a/drive
From: Beilei Xing
This patch adds queue group add/delete virtual channel support.
Signed-off-by: Mingxia Liu
Signed-off-by: Beilei Xing
---
drivers/common/idpf/idpf_common_virtchnl.c | 66 ++
drivers/common/idpf/idpf_common_virtchnl.h | 9 +++
drivers/common/idpf/version.
From: Beilei Xing
This patch adds haipin queue group during vpotr init.
Signed-off-by: Mingxia Liu
Signed-off-by: Beilei Xing
---
drivers/net/cpfl/cpfl_ethdev.c | 125 +
drivers/net/cpfl/cpfl_ethdev.h | 17 +
drivers/net/cpfl/cpfl_rxtx.h | 4 ++
3 file
From: Beilei Xing
Support hairpin Rx/Tx queue setup and release.
Signed-off-by: Xiao Wang
Signed-off-by: Mingxia Liu
Signed-off-by: Beilei Xing
---
drivers/net/cpfl/cpfl_ethdev.c | 6 +
drivers/net/cpfl/cpfl_ethdev.h | 10 +
drivers/net/cpfl/cpfl_rxtx.c| 373
From: Beilei Xing
This patch supports Rx/Tx hairpin queue configuration.
Signed-off-by: Xiao Wang
Signed-off-by: Mingxia Liu
Signed-off-by: Beilei Xing
---
drivers/common/idpf/idpf_common_virtchnl.c | 70 +++
drivers/common/idpf/idpf_common_virtchnl.h | 6 +
drivers/common/idpf/ve
From: Beilei Xing
This patch supports Rx/Tx hairpin queue start/stop.
Signed-off-by: Xiao Wang
Signed-off-by: Mingxia Liu
Signed-off-by: Beilei Xing
---
drivers/common/idpf/idpf_common_virtchnl.c | 2 +-
drivers/common/idpf/idpf_common_virtchnl.h | 3 +
drivers/common/idpf/version.map
From: Beilei Xing
This patch enabls write back based on ITR expire
(WR_ON_ITR) for hairpin queue.
Signed-off-by: Mingxia Liu
Signed-off-by: Beilei Xing
---
drivers/common/idpf/idpf_common_device.c | 75
drivers/common/idpf/idpf_common_device.h | 4 ++
drivers/common/
From: Beilei Xing
This patch supports get hairpin peer ports.
Signed-off-by: Xiao Wang
Signed-off-by: Beilei Xing
---
drivers/net/cpfl/cpfl_ethdev.c | 34 ++
1 file changed, 34 insertions(+)
diff --git a/drivers/net/cpfl/cpfl_ethdev.c b/drivers/net/cpfl/cpfl_e
From: Beilei Xing
This patch supports hairpin_bind/unbind ops.
Signed-off-by: Xiao Wang
Signed-off-by: Beilei Xing
---
drivers/net/cpfl/cpfl_ethdev.c | 137 +
drivers/net/cpfl/cpfl_rxtx.c | 28 +++
drivers/net/cpfl/cpfl_rxtx.h | 2 +
3 files changed,
Add timestamp offload feature support for ACC. Using alarm
to save main time to solve timestamp roll over issue.
Ajust timestamp mbuf registering at dev start.
Wenjing Qiao (7):
common/idpf: fix 64b timestamp roll over issue
net/idpf: save main time by alarm
net/cpfl: save main time by alarm
Using alarm to save main time from registers every 1 second.
Fixes: 8c6098afa075 ("common/idpf: add Rx/Tx data path")
Cc: sta...@dpdk.org
Signed-off-by: Wenjing Qiao
---
drivers/net/idpf/idpf_ethdev.c | 12
1 file changed, 12 insertions(+)
diff --git a/drivers/net/idpf/idpf_ethdev
Reading MTS register at first packet will cause timestamp
roll over issue. To support caculating 64b timestamp, need
an alarm to save master time from registers every 1 second.
Fixes: 8c6098afa075 ("common/idpf: add Rx/Tx data path")
Cc: sta...@dpdk.org
Signed-off-by: Wenjing Qiao
---
drivers/c
Using alarm to save main time from registers every 1 second.
Fixes: 8c6098afa075 ("common/idpf: add Rx/Tx data path")
Cc: sta...@dpdk.org
Signed-off-by: Wenjing Qiao
---
drivers/net/cpfl/cpfl_ethdev.c | 12
1 file changed, 12 insertions(+)
diff --git a/drivers/net/cpfl/cpfl_ethdev
For ACC, getting main time from MTS registers by shared memory.
Notice: it is a workaround, and it will be removed after generic
solution are provided.
Signed-off-by: Wenjing Qiao
---
config/meson.build | 3 ++
drivers/common/idpf/base/idpf_osdep.h | 48 +++
A rxq can be configured with timestamp offload.
So, add timestamp enable flag for rxq.
Fixes: 8c6098afa075 ("common/idpf: add Rx/Tx data path")
Cc: sta...@dpdk.org
Signed-off-by: Wenjing Qiao
Suggested-by: Jingjing Wu
---
drivers/common/idpf/idpf_common_rxtx.c | 3 ++-
drivers/common/idpf/idpf
Due to only support timestamp at port level, registering
timestamp mbuf should be at dev start stage.
Fixes: 8c6098afa075 ("common/idpf: add Rx/Tx data path")
Cc: sta...@dpdk.org
Signed-off-by: Wenjing Qiao
Suggested-by: Jingjing Wu
---
drivers/net/cpfl/cpfl_ethdev.c | 7 +++
drivers/net/c
Due to only support timestamp at port level, registering
timestamp mbuf should be at dev start stage.
Fixes: 8c6098afa075 ("common/idpf: add Rx/Tx data path")
Cc: sta...@dpdk.org
Signed-off-by: Wenjing Qiao
Suggested-by: Jingjing Wu
---
drivers/net/idpf/idpf_ethdev.c | 7 +++
drivers/net/i
21/04/2023 04:34, You, KaisenX:
> From: Thomas Monjalon
> > 13/04/2023 02:56, You, KaisenX:
> > > From: You, KaisenX
> > > > From: Thomas Monjalon
> > > > >
> > > > > I'm not comfortable with this patch.
> > > > >
> > > > > First, there is no comment in the code which helps to understand the
> >
This patch refines devargs parsing functions and use valid
variable max_vport_nb to replace IDPF_MAX_VPORT_NUM.
Signed-off-by: Mingxia Liu
---
drivers/net/idpf/idpf_ethdev.c | 61 +-
1 file changed, 30 insertions(+), 31 deletions(-)
diff --git a/drivers/net/idpf/
21/04/2023 03:47, Lu, Wenzhuo:
> Hi Stephen,
>
> From: Stephen Hemminger
> > On Tue, 18 Apr 2023 13:25:41 +0800
> > Wenzhuo Lu wrote:
> >
> > > The cores in a single CPU may be not all the same.
> > > The user tool is updated to show the
> > > difference of the cores.
> > >
> > > This patch add
20/04/2023 20:20, Tyler Retzlaff:
> On Thu, Apr 20, 2023 at 09:43:28AM +0200, Thomas Monjalon wrote:
> > 19/04/2023 16:51, Tyler Retzlaff:
> > > On Wed, Apr 19, 2023 at 11:36:34AM +0300, Ophir Munk wrote:
> > > > In current DPDK the RTE_MAX_MEMZONE definition is unconditionally hard
> > > > coded a
https://bugs.dpdk.org/show_bug.cgi?id=1218
Bug ID: 1218
Summary: memif zeor-copy mode mbuf->refcnt error
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
This patch set updates idpf shared code.
Wenjing Qiao (15):
common/idpf: remove virtchnl related shared code
common/idpf: fix ctlq message send and receive
common/idpf: fix ITR register definitions for AVF
common/idpf: remove qregion struct variables
common/idpf: move OEM capability to t
Since virtchnl related shared code of iavf
and idpf has the same sourse, so remove
virtual related shared code of idpf.
Signed-off-by: Wenjing Qiao
---
drivers/common/idpf/base/virtchnl.h | 2866 -
.../common/idpf/base/virtchnl_inline_ipsec.h | 567
drivers/commo
Fixes the ctlq send and receive functions to not cast the cookie field
to a u64 before programming. By doing a cast, it can cause endianness
issues as LE will swap the lower 32 and higher 32 bits whereas BE will
not. By treating this field as two 32 bit values, both BE and LE will
place the retval
Fix ITR register definitions for AVF1.0 and AVF2.0
Fixes: fb4ac04e9bfa ("common/idpf: introduce common library")
Cc: sta...@dpdk.org
Signed-off-by: Priyalee Kushwaha
Signed-off-by: Wenjing Qiao
---
drivers/common/idpf/base/idpf_lan_pf_regs.h | 9 +++--
drivers/common/idpf/base/idpf_lan_vf
Existing qregion variables are not well defined and cannot be
used for TC related stuff. Remove them from create vport
struct and add those freed bytes to a new reserved field.
Add appropriate comments on how to use the dynctl and itrn
register spacing variables.
Only VF reference was used in get
Move the existing OEM capability in VIRTCHNL2_OTHER_CAPS to
the last bit. This should not break any backward compatibility
as it is not used yet.
And VIRTCHNL2_MEV_DEVICE is no longer upstreamed.
Signed-off-by: Pavan Kumar Linga
Signed-off-by: Wenjing Qiao
---
drivers/common/idpf/base/virtchnl
Virtchnl2 error codes are required for meaningful failure
information sharing between CP and PF/VF. Introduce the
necessary error codes.
New error codes were introduced removing the old ones. So
the references to the old one should be modified to avoid
CI build failures.
Use appropriate error cod
- Driver assumes minimum packet length for sso as 17 bytes
but it should be a negotiated value from CP.
- Similarly, the number of header buffers for lso that are
supported by the device should also be a negotiated value.
Add min_sso_packet_len, max_hdr_buf_per_lso to address the
above.
Also, the
Virtchnl2 opcodes are no longer in the enum virtchnl_ops. So change
these parameters to allow int rather that compiler enum type checking.
Fixes: fb4ac04e9bfa ("common/idpf: introduce common library")
Cc: sta...@dpdk.org
Signed-off-by: Christopher Pau
Signed-off-by: Wenjing Qiao
---
drivers/co
To make the code more readable and make it clearer that the opcode goes
in cookie_high and retval goes in cookie_low.
Add macro definitions for filling opcode and retval.
Signed-off-by: Charles Stoll
Signed-off-by: Wenjing Qiao
---
drivers/common/idpf/base/idpf_controlq.c | 2 ++
drivers/c
idpf_init_hw needs to free it's q_info.
idpf_clean_arq_element needs to return buffers via post_rx_buffs
Fixes: fb4ac04e9bfa ("common/idpf: introduce common library")
Cc: sta...@dpdk.org
Signed-off-by: Christopher Pau
Signed-off-by: Wenjing Qiao
---
drivers/common/idpf/base/idpf_common.c | 12
Some OSs don't allow allocating DMA memory at runtime. So create an
initial static buffer at initialization to hold this data.
Signed-off-by: Christopher Pau
Signed-off-by: Wenjing Qiao
---
drivers/common/idpf/base/idpf_common.c | 26 +++---
1 file changed, 15 insertions(+),
Replace MAKEMASK to IDPF_M to avoid conflicts with MAKEMASK
redefinition from various subcomponents.
Signed-off-by: Priyalee Kushwaha
Signed-off-by: Wenjing Qiao
---
drivers/common/idpf/base/idpf_controlq.h | 3 --
drivers/common/idpf/base/idpf_lan_pf_regs.h | 26 +--
drivers/co
Add types for new two virtchnl commands: add & delete queue group
Signed-off-by: Nizan Zorea
Signed-off-by: Wenjing Qiao
---
drivers/common/idpf/base/virtchnl2.h | 189 +++
1 file changed, 189 insertions(+)
diff --git a/drivers/common/idpf/base/virtchnl2.h
b/drivers/co
Add 'idpf_ctlq_clean_sq_force' which will clean all descriptors on
given control queue. It is needed in case control plane is not
running and we need to do proper driver cleanup.
Signed-off-by: NorbertX Ciosek
Signed-off-by: Wenjing Qiao
---
drivers/common/idpf/base/idpf_controlq.c | 56 +++
Update license and README
Signed-off-by: Wenjing Qiao
---
.mailmap | 8
drivers/common/idpf/base/README| 4 ++--
drivers/common/idpf/base/idpf_alloc.h | 2 +-
drivers/common/idpf/base/idpf_common.c | 2 +-
drivers/co
From: Chengwen Feng
The media type information helps locate faults such as AN or rate
switching, so add dump media type in eth_dev_priv_dump ops.
Signed-off-by: Chengwen Feng
Signed-off-by: Dongdong Liu
---
drivers/net/hns3/hns3_dump.c | 2 ++
drivers/net/hns3/hns3_ethdev.c | 2 +-
drivers/
This patchset is to add some features for hns3 pmd.
Chengwen Feng (1):
net/hns3: support dump media type
Dongdong Liu (1):
net/hns3: simplify for hardware csum offloading
drivers/net/hns3/hns3_cmd.c| 3 ++
drivers/net/hns3/hns3_cmd.h| 1 +
drivers/net/hns3/hns3_dump.c | 3 ++
d
If the NIC support simple BD mode, the hardware will calculate
the checksum from the start position of checksum and fill the
checksum result to the offset position, which simple the
HW operations of calculating the type and header length of
L3/L4. Add this mode for hns3 pmd when the packet type
is
AESNI_MB PMD does not support Digest Encrypted. This patch adds partial
support for this feature.
Signed-off-by: Brian Dooley
---
Some out-of-place tests are still failing.
Only some in-place tests are passing.
Working on adding support for this feature in v2.
---
app/test/1.diff
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Friday, 21 April 2023 10.35
>
> 20/04/2023 20:20, Tyler Retzlaff:
> > On Thu, Apr 20, 2023 at 09:43:28AM +0200, Thomas Monjalon wrote:
> > > 19/04/2023 16:51, Tyler Retzlaff:
> > > > On Wed, Apr 19, 2023 at 11:36:34AM +0300, Ophir Munk w
This patchset adds some optimisations for AESNI_MB PMD, many based on
features that are available in intel-ipsec-mb v1.3 and future release v1.4.
Marcel Cornu (1):
crypto/ipsec_mb: use burst API in aesni_mb
Pablo de Lara (7):
crypto/ipsec_mb: use GMAC dedicated algorithms
crypto/ipsec_mb: u
From: Pablo de Lara
AES-GMAC can be done with auth-only enums
IMB_AES_GMAC_128/192/256, which allows another cipher
algorithm to be used, instead of being part of AES-GCM.
Signed-off-by: Pablo de Lara
Signed-off-by: Ciara Power
---
drivers/crypto/ipsec_mb/pmd_aesni_mb.c | 104 +++-
From: Marcel Cornu
Use new ipsec_mb burst API in dequeue burst function,
when ipsec_mb version is v1.3 or newer.
Signed-off-by: Marcel Cornu
Signed-off-by: Pablo de Lara
Signed-off-by: Ciara Power
---
drivers/crypto/ipsec_mb/pmd_aesni_mb.c | 133 -
1 file changed, 132
From: Pablo de Lara
Use new SGL API available from IPSec Multi-buffer v1.3,
where only one function call is required to submit
all segments to be processed in an SGL scenario.
Instead of having one call per segment, there is only
one call per buffer.
Signed-off-by: Pablo de Lara
Signed-off-by:
From: Pablo de Lara
Cipher direction, cipher mode and hash algorithm are
duplicated in crypto session.
Signed-off-by: Pablo de Lara
---
drivers/crypto/ipsec_mb/pmd_aesni_mb_priv.h | 3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/crypto/ipsec_mb/pmd_aesni_mb_priv.h
b/drivers/crypt
From: Pablo de Lara
Store template IMB_JOB in session that
will have filled all session-related fields.
These fields include cipher direction, chain order, cipher mode,
hash algorithm, key length, IV lengths, AAD length, digest length,
and key pointers.
Signed-off-by: Pablo de Lara
Signed-off-b
From: Pablo de Lara
Use a separate code path when dealing with AES-GCM.
Signed-off-by: Pablo de Lara
Signed-off-by: Ciara Power
---
drivers/crypto/ipsec_mb/pmd_aesni_mb.c | 88 +++---
1 file changed, 79 insertions(+), 9 deletions(-)
diff --git a/drivers/crypto/ipsec_mb/pm
From: Pablo de Lara
linear_sgl buffer only needs to be freed
if it was allocated previously.
Signed-off-by: Pablo de Lara
---
drivers/crypto/ipsec_mb/pmd_aesni_mb.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/crypto/ipsec_mb/pmd_aesni_mb.c
b/drivers/crypto
From: Pablo de Lara
When creating a session, get the session ID that
defines the fixed session parameters and store it in the private data.
When retrieving IMB_JOB's, if their internal session ID matches
the one in the private session data, these fixed session parameters
do not need to be filled
Fedora 35 has been declared EOL in 2022/12 (see [1]).
Fedora 36 will soon be EOL too.
Move to Fedora 37.
Fedora 37 libbpf does not support AF_XDP anymore, now provided by
libxdp.
1: https://docs.fedoraproject.org/en-US/releases/eol/
Signed-off-by: David Marchand
---
.github/workflows/build.yml
It uses sysfs interface and the vfio-cdx driver to discover
and initialize the CDX devices.
The CDX bus and VFIO support is available at Xilinx open source tree:
https://github.com/Xilinx/linux-xlnx (drivers/cdx/ and drivers/vfio/cdx)
Linux CDX bus patches has been added into linux next:
https://
CDX bus supports multiple type of devices, which can be
exposed to user-space via vfio-cdx.
vfio-cdx provides the MMIO IO_MEMORY regions as well as the
DMA interface for the device (IOMMU).
This support aims to enable the DPDK to support the cdx
devices in user-space using VFIO interface.
Signed
CDX bus can use VFIO interface for mapping and unmapping
of DMA addresses in the IOMMU. This change adds the callback
support for map and unmap APIs as well as fetching the IOMMU
class.
Signed-off-by: Nipun Gupta
---
drivers/bus/cdx/cdx.c | 40
1 file cha
MSI's are exposed to the devices using VFIO (vfio-cdx). This
patch uses the same to add support for MSI for the devices on
the cdx bus.
A couple of API's have been introduced in the EAL interrupt
framework:
- rte_intr_irq_count_set: This API is used to set the total
interrupts on the interrupt
This change adds support for plugging and unplugging
CDX devices on the CDX bus. Also, CDX dev iterator support
has been added for the CDX bus.
Signed-off-by: Nipun Gupta
---
drivers/bus/cdx/bus_cdx_driver.h | 1 +
drivers/bus/cdx/cdx.c| 122 +++
2 files
Adding support for AMD CDX devices
Signed-off-by: Nipun Gupta
---
config/arm/arm64_cdx_linux_gcc | 17 +
config/arm/meson.build | 14 ++
2 files changed, 31 insertions(+)
create mode 100644 config/arm/arm64_cdx_linux_gcc
diff --git a/config/arm/arm64_cdx_lin
21/04/2023 13:08, Morten Brørup:
> > From: Thomas Monjalon [mailto:tho...@monjalon.net]
> > Sent: Friday, 21 April 2023 10.35
> > 20/04/2023 20:20, Tyler Retzlaff:
> > > On Thu, Apr 20, 2023 at 09:43:28AM +0200, Thomas Monjalon wrote:
> > > > 19/04/2023 16:51, Tyler Retzlaff:
> > > > > On Wed, Apr
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Friday, 21 April 2023 16.57
>
> 21/04/2023 13:08, Morten Brørup:
> > > From: Thomas Monjalon [mailto:tho...@monjalon.net]
> > > Sent: Friday, 21 April 2023 10.35
> > > 20/04/2023 20:20, Tyler Retzlaff:
> > > > On Thu, Apr 20, 2023 at 09:
Version 2 fixes:
- misspellings
- whitespace issues
- signed off issues
Kishore Padmanabha (1):
net/bnxt: fix multi-root card support
Randy Schacher (9):
net/bnxt: remove deprecated features
net/bnxt: update bnxt hsi structure
net/bnxt: update copyright date and cleanup whitespace
net/b
- Deprecate shadow identifier
- Deprecate shadow tcam
Signed-off-by: Randy Schacher
Signed-off-by: Kishore Padmanabha
Reviewed-by: Peter Spreadborough
---
drivers/net/bnxt/bnxt_hwrm.c | 53 --
drivers/net/bnxt/bnxt_hwrm.h | 10 -
drivers/net/bnxt/tf_core/mes
Update the Copyright to 2023
Clean up extra blank lines
Clean up other whitespace issues
Signed-off-by: Randy Schacher
Reviewed-by: Kishore Padmanabha
---
drivers/net/bnxt/bnxt_cpr.c| 2 +-
drivers/net/bnxt/bnxt_cpr.h| 2 +-
drivers/net/bnxt/bnxt_filter.c
Add RTE meter support into the ULP layer
Signed-off-by: Randy Schacher
Reviewed-by: Jay Ding
---
drivers/net/bnxt/bnxt.h | 2 +
drivers/net/bnxt/bnxt_ethdev.c| 1 +
drivers/net/bnxt/tf_ulp/bnxt_ulp.c| 8 +
drivers/net/bnxt/tf_ulp/bnxt_ulp.
add locking and time stamp checks to ptp feature
Signed-off-by: Randy Schacher
Reviewed-by: Ajit Khaparde
---
drivers/net/bnxt/bnxt.h| 5 ++
drivers/net/bnxt/bnxt_ethdev.c | 11 +
drivers/net/bnxt/bnxt_hwrm.c | 11 -
drivers/net/bnxt/bnxt_ring.c | 3 ++
drivers/net/bnxt/bn
From: Kishore Padmanabha
Changed the logic to use device serial number to identify that
different ports belong to same physical card instead of the PCI
domain address.
Signed-off-by: Kishore Padmanabha
Reviewed-by: Shahaji Bhosle
---
drivers/net/bnxt/bnxt.h| 3 +++
drivers/net/bn
Add RTE ECPRI support into the ULP layer
Signed-off-by: Randy Schacher
Signed-off-by: Shahaji Bhosle
Reviewed-by: Manish Kurup
---
drivers/net/bnxt/bnxt.h | 4 +
drivers/net/bnxt/bnxt_ethdev.c| 35 +
drivers/net/bnxt/bnxt_hwrm.c |
From: Shuanglin Wang
On WH+, if rss mode isn't enabled, then there is no rss context.
Submitting HWRM_VNIC_RSS_CFG request to firmware would hit a failure.
The fix is to check the rss context. If no rss context, then don't
submit the hwrm request.
Signed-off-by: Shuanglin Wang
Signed-off-by: K
Tail load in __rte_ring_move_cons_head and __rte_ring_move_prod_head
can be changed to __ATOMIC_RELAXED from __ATOMIC_ACQUIRE.
Because to calculate the addresses of the dequeue
elements __rte_ring_dequeue_elems uses the old_head updated by the
__atomic_compare_exchange_n intrinsic used in
__rte_rin
Tail load in __rte_ring_move_cons_head and __rte_ring_move_prod_head
can be changed to __ATOMIC_RELAXED from __ATOMIC_ACQUIRE.
Because to calculate the addresses of the dequeue
elements __rte_ring_dequeue_elems uses the old_head updated by the
__atomic_compare_exchange_n intrinsic used in
__rte_rin
David Marchand writes:
> Fedora 35 has been declared EOL in 2022/12 (see [1]).
> Fedora 36 will soon be EOL too.
>
> Move to Fedora 37.
> Fedora 37 libbpf does not support AF_XDP anymore, now provided by
> libxdp.
>
> 1: https://docs.fedoraproject.org/en-US/releases/eol/
>
> Signed-off-by: David
Google cloud routes traffic using IP addresses without the support of MAC
addresses, so changing source IP address for txonly-multi-flow can have
negative performance implications for net/gve when using testpmd. This
patch updates txonly multiflow mode to modify source ports instead of
source IP ad
在 2023/4/21 9:12, zhoumin 写道:
On Fri, Apr 7, 2023 at 4:50PM, Min Zhou wrote:
Segmentation fault has been observed while running the
ixgbe_recv_pkts_lro() function to receive packets on the Loongson
3C5000 processor which has 64 cores and 4 NUMA nodes.
Reason is the read ordering of the statu
On 2023/04/20 16:20, Ruifeng Wang wrote:
-Original Message-
From: Akihiko Odaki
Sent: Friday, April 14, 2023 8:42 PM
To: Ruifeng Wang ; Bruce Richardson
Cc: dev@dpdk.org; Akihiko Odaki
Subject: [PATCH 2/2] config/arm: Enable NUMA for generic Arm build
We enable NUMA even if the prese
77 matches
Mail list logo
