Hi Bruce, they should not be merged, they are solving different issues.
On Thu, Aug 22, 2024 at 09:56:08AM +, Soumyadeep Hore wrote:
> Correct the logic for determining the maximum PFA offset to include
> the extra last word. Additionally, make the driver robust against
> overflows by using
In roc batch alloc wait code, __ATOMIC_RELAXED is changed to
__ATOMIC_ACQUIRE in order to avoid potential out of order loads.
Signed-off-by: Nawal Kishor
---
.mailmap | 1 +
drivers/common/cnxk/roc_npa.h | 2 +-
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/.
From: Pavan Nikhilesh
Move DMA ops to use mempool to prevent using the
same ops before completion.
This also allows us to measure forward latency.
Signed-off-by: Pavan Nikhilesh
---
app/test-eventdev/test_perf_atq.c| 14 ++--
app/test-eventdev/test_perf_common.c | 106 ++--
On Thu, 22 Aug 2024 23:27:53 +0300
Alexander Kozyrev wrote:
> +/**
> + * @warning
> + * @b EXPERIMENTAL: this API may change without prior notice.
> + *
> + * Enqueue rule creation operation.
> + *
> + * @param port_id
> + * Port identifier of Ethernet device.
> + * @param queue_id
> + * Flow
On 8/22/2024 6:18 PM, Bruce Richardson wrote:
> On Thu, Aug 22, 2024 at 06:14:55PM +0100, Bruce Richardson wrote:
>> On Thu, Aug 22, 2024 at 05:53:27PM +0100, Ferruh Yigit wrote:
>>> On 8/22/2024 11:41 AM, Bruce Richardson wrote:
Testpmd supports the "--cmdline-file" parameter to read a set of
Introduce new JUMP_TO_TABLE and JUMP_TO_TABLE_INDEX actions.
They allow bypassing a hierarchy of groups and going directly
to a specified flow table. That gives a user the flexibility
to jump between different priorities in a group and eliminates
the need to do a table lookup in the group hierarchy
#
August 22, 2024
Attendees
1. Patrick Robb
2. Tomas Durovec
3. Dean Marx
4. Luca Vizzarro
5. Jeremy Spewock
6. Ali Alnubani
7. Juraj Linkeš
8. Paul Szczepanek
#
Extends the functionality of the function responsible for
checking if the firmware supports FEC (Forward Error Correction)
disable in Auto FEC mode. It now includes an additional check to
determine if the adapter is a E830 model. With this
change, the function will enable FEC auto-detect support fo
Update E830 50G branding strings from "E830-XXV" to "E830-L".
Signed-off-by: Soumyadeep Hore
---
drivers/net/ice/base/ice_common.c | 6 +++---
drivers/net/ice/base/ice_devids.h | 12 ++--
drivers/net/ice/ice_ethdev.c | 6 +++---
3 files changed, 12 insertions(+), 12 deletions(-)
An optional flag field has been added to the signature segment
header. The flags field contains two flags, a valid flag, and
a last segment flag that indicates whether the segment is the
last segment that will be downloaded to firmware.
If the flag field's valid bit is NOT set, then as was done
be
In E825C, regarding the Get Tx Topology AQ command, there is
a change in the way that the RD bit must be set. For E825C,
the RD bit must be cleared for the Get Tx Topology operation,
whereas for E810 devices, the RD bit must be set.
Signed-off-by: Soumyadeep Hore
---
drivers/net/ice/base/ice_ddp
Correct the logic for determining the maximum PFA offset to include the
extra last word. Additionally, make the driver robust against overflows
by using check_add_overflow. This ensures that even if the NVM
provides bogus data, the driver will not overflow, and will instead log
a useful warning mes
When setting PHY configuration during driver initialization,
200G link speed is not being advertised even when the PHY
is capable. This is because the get PHY capabilities link
speed response is being masked by ICE_AQ_LINK_SPEED_M,
which does not include 200G link speed bit.
Fixes: d13ad9cf1721 ("
Visual Studio C++ compiler does not pass 32->16 or
16->8 bits conversions because of possible loss of data.
Signed-off-by: Soumyadeep Hore
---
drivers/net/ice/base/ice_ptp_hw.c | 31 ---
1 file changed, 4 insertions(+), 27 deletions(-)
diff --git a/drivers/net/ice/ba
Add Bit macro to init PHY 1 for E825C devices.
Signed-off-by: Soumyadeep Hore
---
drivers/net/ice/base/ice_ptp_hw.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ice/base/ice_ptp_hw.c
b/drivers/net/ice/base/ice_ptp_hw.c
index e574ae6d4f..e61810cbdc 100644
The ice_get_pfa_module_tlv() function iterates over the Preserved Fields
Area to read data from the Shadow RAM, including the Part Board Assembly
data, among others.
If the specific TLV being requested is not found in the current NVM, the
code will read past the end of the PFA, misinterpreting the
Add E830_GLTCLAN_TSYN_REG_RANGE_ENFORCE* defines to
unified_manual.inc to make them available externally.
Signed-off-by: Soumyadeep Hore
---
drivers/net/ice/base/ice_hw_autogen.h | 14 ++
1 file changed, 14 insertions(+)
diff --git a/drivers/net/ice/base/ice_hw_autogen.h
b/drivers/
The implementation was done incorrectly assuming
the TS PLL parameters would be similar to E822/E823
devices. Fix it by using proper values.
Define access to SB (sideband) for second PHY and
CGU devices in case of E825C devices.
In E825C soft straps of CGU cannot be read from HW,
and therefore it
Firmware was giving a number for the MSIX vectors that was
way too big and obviously not right. Because of the wrong
format specifier, this big number ended up looking like a
tiny negative number in the logs. This was fixed by using
the right format specifier everywhere it's needed.
Signed-off-by:
Updating the latest shared code patches to ICE base driver.
---
v2:
- Addressed comments given by reviewer
- Corrected errors in Camel Case
---
Soumyadeep Hore (12):
net/ice: use correct format specifiers for unsigned ints
net/ice: updates for ptp init in E825C
net/ice: add new tag definiti
> From: Robin Jarry [mailto:rja...@redhat.com]
>
> Morten Brørup, Aug 22, 2024 at 17:13:
> > Maybe the 'a' field in the DPDK IPv6 address structure should be named
> 'b' for "byte" instead, to prepare it for adding a union with a 'w' (for
> "word") field if we want to introduce 2-byte alignment.
Anatoly Burakov, Aug 22, 2024 at 12:38:
In traditional NUMA case, NUMA nodes and physical sockets were used
interchangeably, but there are cases where there can be multiple NUMA
nodes per socket, as well as all CPU's being assigned NUMA node 0 even in
cases of multiple sockets. Use sysfs to print
On Thu, Aug 22, 2024 at 06:14:55PM +0100, Bruce Richardson wrote:
> On Thu, Aug 22, 2024 at 05:53:27PM +0100, Ferruh Yigit wrote:
> > On 8/22/2024 11:41 AM, Bruce Richardson wrote:
> > > Testpmd supports the "--cmdline-file" parameter to read a set of initial
> > > commands from a file. However, th
On Thu, Aug 22, 2024 at 06:06:39PM +0100, Bruce Richardson wrote:
> On Tue, Aug 06, 2024 at 08:35:27AM +0800, Kaiwen Deng wrote:
> > Physical link instability may cause a core dump.
> > Unstable physical links can result in a large number of link
> > change events.Link change events captured by vf
On Thu, Aug 22, 2024 at 05:53:27PM +0100, Ferruh Yigit wrote:
> On 8/22/2024 11:41 AM, Bruce Richardson wrote:
> > Testpmd supports the "--cmdline-file" parameter to read a set of initial
> > commands from a file. However, the only indication that this has been
> > done successfully on startup is a
On Tue, Aug 06, 2024 at 08:35:27AM +0800, Kaiwen Deng wrote:
> Physical link instability may cause a core dump.
> Unstable physical links can result in a large number of link
> change events.Link change events captured by vf before vf
> resources are allocated will result in a core dump.
>
> This
On 8/22/2024 11:41 AM, Bruce Richardson wrote:
> Testpmd supports the "--cmdline-file" parameter to read a set of initial
> commands from a file. However, the only indication that this has been
> done successfully on startup is a single-line message, no output from
> the commands is seen.
>
For u
On Wed, Jul 31, 2024 at 07:23:03AM +, Praveen Shetty wrote:
> CPFL parser was incorrectly parsing the mask value of the
> next_proto_id field from recipe.json file as a string
> instead of unsigned integer.
>
> Fixes: 41f20298ee8c ("net/cpfl: parse flow offloading hint from JSON")
> Cc: sta...
Adds a new script which automatically re-generates the JSON schema file
based on the Pydantic configuration models.
Moreover, update the JSON schema with this script for the first time.
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
doc/guides/tools/dts.rst |
The introduction of TestSuiteSpec adds auto-discovery of test suites,
which are also automatically imported. This causes double imports as the
runner loads the test suites. This changes the behaviour of the runner
to load the imported classes from TestSuiteSpec instead of importing
them again.
Sig
This change brings in Pydantic in place of Warlock. Pydantic offers
a built-in model validation system in the classes, which allows for
a more resilient and simpler code. As a consequence of this change:
- most validation is now built-in
- further validation is added to verify:
- cross referenci
Add Pydantic to the project dependencies while dropping Warlock.
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
dts/poetry.lock| 346 +
dts/pyproject.toml | 3 +-
2 files changed, 135 insertions(+), 214 deletions(-)
diff --git a/
Currently there is a lack of a definition which identifies all the test
suites available to test. This change intends to simplify the process to
discover all the test suites and idenfity them.
Signed-off-by: Luca Vizzarro
Reviewed-by: Paul Szczepanek
---
dts/framework/test_suite.py | 182 ++
Hello,
sending the first version for the Pydantic configuration update work.
Best,
Luca
Luca Vizzarro (5):
dts: add TestSuiteSpec class and discovery
dts: add Pydantic and remove Warlock
dts: use Pydantic in the configuration
dts: use TestSuiteSpec class imports
dts: add JSON schema ge
On Tue, Jul 30, 2024 at 11:45:40AM +, Praveen Shetty wrote:
> In case of CPFL PMD, port_representor action is used for
> the local vport and represented_port action is used for
> the remote port(remote port in this case is either the idpf
> pf or the vf port that is being represeted by the cpfl
On Tue, Jul 30, 2024 at 11:45:40AM +, Praveen Shetty wrote:
> In case of CPFL PMD, port_representor action is used for
> the local vport and represented_port action is used for
> the remote port(remote port in this case is either the idpf
> pf or the vf port that is being represeted by the cpfl
On Thu, Aug 22, 2024 at 04:58:59PM +0100, Bruce Richardson wrote:
> On Fri, Jul 12, 2024 at 07:30:47PM +0800, Jun Wang wrote:
> > The code to update link status is not safe in secondary process.
> > If called from secondary it will crash, example from dumpcap:
> > eth_em_link_update
> >
> > Si
On 8/22/2024 8:51 AM, rajesh goel wrote:
> Hi All,
> Need info if DPDK hash library supports bihash table where for multi-
> thread and multi-process we can update/del/lookup entries per bucket level.
>
>
+ hash library maintainers.
On Fri, Jul 12, 2024 at 07:30:47PM +0800, Jun Wang wrote:
> The code to update link status is not safe in secondary process.
> If called from secondary it will crash, example from dumpcap:
> eth_em_link_update
>
> Signed-off-by: Jun Wang
> ---
> drivers/net/e1000/em_ethdev.c | 3 +++
> 1 fil
Support the IP Encapsulating Security Payload (ESP) in transport mode.
Currently, we have RTE_PTYPE_TUNNEL_ESP for the ESP tunnel mode.
Transport mode can be detected by parsing the "Next Header" field.
The Next Header is TCP for the transport mode and IP for the tunnel mode.
Add RTE_PTYPE_L4_ESP f
Morten Brørup, Aug 22, 2024 at 17:13:
The types seem to be different in the Windows API (than in Linux/BSD):
https://github.com/tpn/winsdk-10/blob/9b69fd26ac0c7d0b83d378dba01080e93349c2ed/Include/10.0.14393.0/shared/in6addr.h#L25
https://github.com/tpn/winsdk-10/blob/9b69fd26ac0c7d0b83d378dba0108
Stephen Hemminger, Aug 22, 2024 at 16:13:
Why does DPDK need to have its own definitions of IPv4 and IPv6 addresses?
The structures in_addr and in6_addr exist on Linux, BSD, and Windows.
It creates lots of code reimplementation (see inet_ntop etc).
By the way, you can call inet_ntop(AF_INET6) w
Stephen Hemminger, Aug 22, 2024 at 16:13:
Let me ask a couple of questions about this.
Why does DPDK need to have its own definitions of IPv4 and IPv6 addresses?
The structures in_addr and in6_addr exist on Linux, BSD, and Windows.
It creates lots of code reimplementation (see inet_ntop etc).
W
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
>
> On Wed, 21 Aug 2024 18:25:17 +0200
> Robin Jarry wrote:
>
> > Hi everyone,
> >
> > As discussed recently [1], here is a first draft of the IPv6 APIs rework.
> The
> > API change was announced before the 24.07 release [2]. This ser
On Thu, Aug 22, 2024 at 09:56:08AM +, Soumyadeep Hore wrote:
> Correct the logic for determining the maximum PFA offset to include the
> extra last word. Additionally, make the driver robust against overflows
> by using check_add_overflow. This ensures that even if the NVM
> provides bogus data
On Thu, Aug 22, 2024 at 09:56:04AM +, Soumyadeep Hore wrote:
> The ice_get_pfa_module_tlv() function iterates over the Preserved Fields
> Area to read data from the Shadow RAM, including the Part Board Assembly
> data, among others.
>
> If the specific TLV being requested is not found in the c
https://bugs.dpdk.org/show_bug.cgi?id=1526
Bug ID: 1526
Summary: rte_eth_rx_burst is blocking in libpcap pmd on no
packets
Product: DPDK
Version: 24.03
Hardware: x86
OS: Linux
Status: UNCONFIRMED
Hi all,
Here is a list of patches targeted for stable release 23.11.2.
The planned date for the final release is 31th August.
Please help with testing and validation of your use cases and report
any issues/results with reply-all to this mail. For the final release
the fixes and reported validati
Extends the functionality of the function responsible for
checking if the firmware supports FEC (Forward Error Correction)
disable in Auto FEC mode. It now includes an additional check to
determine if the adapter is a Connorsville model. With this
change, the function will enable FEC auto-detect su
Update E830 50G branding strings from "E830-XXV" to "E830-L".
Signed-off-by: Soumyadeep Hore
---
drivers/net/ice/base/ice_common.c | 6 +++---
drivers/net/ice/base/ice_devids.h | 12 ++--
drivers/net/ice/ice_ethdev.c | 6 +++---
3 files changed, 12 insertions(+), 12 deletions(-)
An optional flag field has been added to the signature segment
header. The flags field contains two flags, a valid flag, and
a last segment flag that indicates whether the segment is the
last segment that will be downloaded to firmware.
If the flag field's valid bit is NOT set, then as was done
be
In E825C, regarding the Get Tx Topology AQ command, there is
a change in the way that the RD bit must be set. For E825C,
the RD bit must be cleared for the Get Tx Topology operation,
whereas for E810 devices, the RD bit must be set.
Signed-off-by: Soumyadeep Hore
---
drivers/net/ice/base/ice_ddp
Correct the logic for determining the maximum PFA offset to include the
extra last word. Additionally, make the driver robust against overflows
by using check_add_overflow. This ensures that even if the NVM
provides bogus data, the driver will not overflow, and will instead log
a useful warning mes
When setting PHY configuration during driver initialization,
200G link speed is not being advertised even when the PHY
is capable. This is because the get PHY capabilities link
speed response is being masked by ICE_AQ_LINK_SPEED_M,
which does not include 200G link speed bit.
Fixes: d13ad9cf1721 ("
Visual Studio C++ compiler does not pass 32->16 or
16->8 bits conversions because of possible loss of data.
Signed-off-by: Soumyadeep Hore
---
drivers/net/ice/base/ice_ptp_hw.c | 31 ---
1 file changed, 4 insertions(+), 27 deletions(-)
diff --git a/drivers/net/ice/ba
Add BIt macro to init PHY 1 for GNRD.
Signed-off-by: Soumyadeep Hore
---
drivers/net/ice/base/ice_ptp_hw.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ice/base/ice_ptp_hw.c
b/drivers/net/ice/base/ice_ptp_hw.c
index e574ae6d4f..e61810cbdc 100644
--- a/dr
The ice_get_pfa_module_tlv() function iterates over the Preserved Fields
Area to read data from the Shadow RAM, including the Part Board Assembly
data, among others.
If the specific TLV being requested is not found in the current NVM, the
code will read past the end of the PFA, misinterpreting the
Add E830_GLTCLAN_TSYN_REG_RANGE_ENFORCE* defines to
unified_manual.inc to make them available externally.
Signed-off-by: Soumyadeep Hore
---
drivers/net/ice/base/ice_hw_autogen.h | 14 ++
1 file changed, 14 insertions(+)
diff --git a/drivers/net/ice/base/ice_hw_autogen.h
b/drivers/
The implementation was done incorrectly assuming
the TS PLL parameters would be similar to E822/E823
devices. Fix it by using proper values.
Define access to SB (sideband) for second PHY and
CGU devices in case of E825C devices.
In E825C soft straps of CGU cannot be read from HW,
and therefore it
Firmware was giving a number for the MSIX vectors that was
way too big and obviously not right. Because of the wrong
format specifier, this big number ended up looking like a
tiny negative number in the logs. This was fixed by using
the right format specifier everywhere it's needed.
Signed-off-by:
Updating the latest shared code patches to ICE base driver.
Soumyadeep Hore (12):
net/ice: use correct format specifiers for unsigned ints
net/ice: updates for ptp init GNRD
net/ice: add new tag definitions
net/ice: avoid reading past end of PFA
net/ice: update PTP init
net/ice: addres
Testpmd supports the "--cmdline-file" parameter to read a set of initial
commands from a file. However, the only indication that this has been
done successfully on startup is a single-line message, no output from
the commands is seen.
To improve usability here, we can use cmdline_new rather than
c
Currently, devbind does not print out any NUMA information, which makes
figuring out which NUMA node device belongs to not trivial. Add printouts
for NUMA information if NUMA support is enabled on the system.
Signed-off-by: Anatoly Burakov
Acked-by: Robin Jarry
---
Notes:
v1 -> v2:
- Ad
Update coding style:
- make the code PEP-484 compliant
- add more comments, improve readability, use f-strings everywhere
- address all Python static analysis (e.g. mypy, pylint) warnings
- format code with Ruff
- improve error handling
- refactor printing and sysfs/procfs access functions
- sort
Update coding style:
- make it PEP-484 compliant
- format code with Ruff
- address all mypy etc. warnings
- use f-strings in place of old-style string interpolation
- refactor printing to make the code more readable
- read valid CPU ID's from "online" sysfs node
Signed-off-by: Anatoly Burakov
Ac
In traditional NUMA case, NUMA nodes and physical sockets were used
interchangeably, but there are cases where there can be multiple NUMA
nodes per socket, as well as all CPU's being assigned NUMA node 0 even in
cases of multiple sockets. Use sysfs to print out NUMA information.
Signed-off-by: Ana
Testpmd supports the "--cmdline-file" parameter to read a set of initial
commands from a file. However, the only indication that this has been
done successfully on startup is a single-line message, no output from
the commands is seen.
To improve usability here, we can use cmdline_new rather than
c
Add a callback for error recovery and register it with cryptodev.
Add a unit test to verify the error recovery of cryptodev.
The unit test generates an error by passing an mbuf to cryptodev
allocated from heap memory. The registered callback will be called as
part of error recovery.
The unit test v
Add support for reset of a specific queue pair on cnxk platform.
Signed-off-by: Vidya Sagar Velumuri
diff --git a/drivers/common/cnxk/roc_cpt.c b/drivers/common/cnxk/roc_cpt.c
index aba2a49d19..3855d5e7a5 100644
--- a/drivers/common/cnxk/roc_cpt.c
+++ b/drivers/common/cnxk/roc_cpt.c
@@ -953,6 +9
The API will reset the specific queue pair of a cryptodev.
The current API, cryptodev_queue_pair_setup(), requires the cryptodev
to be stopped before reconfiguring any queue pair. Stopping the
cryptodev in one thread can result in a segmentation fault when
multiple queues are used for enqueue and d
Hi All,
Need info if DPDK hash library supports bihash table where for multi-thread
and multi-process we can update/del/lookup entries per bucket level.
Thanks
Rajesh
Release status meeting minutes 2024-08-22
=
Agenda:
* Release Dates
* Subtrees
* Roadmaps
* LTS
* Defects
* Opens
Participants:
* AMD
* ARM
* Debian/Microsoft
* Intel
* Marvell
* Red Hat
Release Dates
-
The following are the current/updated w
+def configure_port_mtu_all(self, mtu_length: int) -> None:
+"""Set the MTU length on all designated ports.
+
+Args:
+mtu_length: The MTU length to be set on all ports.
+"""
+for port in self.show_port_info_all():
+self.send_command(f"por
On 20/08/2024 16:54, Stephen Hemminger wrote:
> On Tue, 20 Aug 2024 14:42:56 +0100
> Alex Chapman wrote:
>
>> Hi Stephen,
>>
>> I have gone through your patch series and the hexdump option would be
>> quite valuable for use in DTS.
>>
>> However I am currently facing the issue of distinguishing
> -Original Message-
> From: pbhagavat...@marvell.com
> Sent: Thursday, August 22, 2024 2:08 AM
> To: Jerin Jacob ; Amit Prakash Shukla
>
> Cc: dev@dpdk.org; Pavan Nikhilesh Bhagavatula
>
> Subject: [PATCH] test/event: fix incorrect target event queue
>
> From: Pavan Nikhilesh
>
> In
Add asserts to validate the max lengths set.
Signed-off-by: Anoob Joseph
---
app/test/test_cryptodev.c | 2 +-
app/test/test_cryptodev.h | 2 ++
app/test/test_cryptodev_aead_test_vectors.h | 14 +-
3 files changed, 16 insertions(+), 2 deletions(-
Remove redefinition of TRUE & FALSE. These are defined by stdbool. Can
use the same.
Signed-off-by: Anoob Joseph
---
app/test/test_cryptodev.c | 5 +++--
app/test/test_cryptodev.h | 3 ---
2 files changed, 3 insertions(+), 5 deletions(-)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cr
Enable larger packet sizes with TLS. Add wrapper for existing
create_segmented_mbuf() function to get allocations from both pools.
Signed-off-by: Anoob Joseph
---
app/test/test_cryptodev.c | 13 +--
app/test/test_cryptodev.h | 38 +++
a
The macros are not used. Remove the same.
Signed-off-by: Anoob Joseph
---
app/test/test_cryptodev.c | 2 --
app/test/test_cryptodev.h | 5 -
2 files changed, 7 deletions(-)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index 169e2f3b48..1491ff03ba 100644
--- a/app/test/
Start opening brace of the function body in new line as suggested in
coding guideline.
Signed-off-by: Anoob Joseph
---
app/test/test_cryptodev.h | 10 ++
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/app/test/test_cryptodev.h b/app/test/test_cryptodev.h
index c56ef2fe3f..
Free the pools allocated in testsuite_setup.
Signed-off-by: Anoob Joseph
---
app/test/test_cryptodev.c | 13 +
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index 1491ff03ba..6079d10078 100644
--- a/app/test/test_c
Enable larger packet sizes in TLS tests and add minor improvements in
crypto unit tests.
Anoob Joseph (6):
test/crypto: add asserts to validate test lengths
test/crypto: enable larger packet sizes with TLS
test/crypto: remove redefinition
test/crypto: remove unused macros
test/crypto: st
Please see ENA updated roadmap:
API Change:
- Restructuring of the llq policy user setting.
Bug fix:
- Addition of a missing memory barrier in aarch64 for instances based on Nitro
v4 and later.
Features:
- Support for dynamic HRSS size.
- Rework of the Rx refill threshold setting.
- Support of
Hello all, we would like to share the upcoming changes and updates in the AWS
ENA driver roadmap for version 24.11:
API Change:
- Restructuring of the llq policy user setting.
Bug fix:
- Addition of a missing memory barrier in aarch64 for instances based on Nitro
v4 and later.
Features:
- Supp
84 matches
Mail list logo
