On 1/16/24 21:41, Stephen Hemminger wrote:
RTE_BUILD_BUG_ON() was being used with a non-constant value.
The inline function rte_is_power_of_2() is not constant since
inline expansion happens later in the compile process.
Replace it with macro which will be constant.
Fixes: 4236ce9bf5bf ("event/o
On 1/16/24 21:41, Stephen Hemminger wrote:
The macro RTE_MIN has some hidden assignments to provide type
safety which means the statement can not be fully evaluated in
first pass of compiler. Replace RTE_MIN() with equivalent macro.
This will cause errors from checkpatch about multiple evaluatio
On 2024-01-16 19:41, Stephen Hemminger wrote:
Both Gcc, clang and MSVC have better way to do compile time
assertions rather than using out of bounds array access.
The old method would fail if -Wvla is enabled because compiler
can't determine size in that code. Also, the use of new
_Static_assert
On 1/16/24 21:41, Stephen Hemminger wrote:
Both Gcc, clang and MSVC have better way to do compile time
assertions rather than using out of bounds array access.
The old method would fail if -Wvla is enabled because compiler
can't determine size in that code. Also, the use of new
_Static_assert wi
To reduce flow migration overhead, replace the array-based
representation of which set of ports are bound to a particular queue
by a bitmask-based one.
The maximum number of DSW event ports remains 64, but after this
change can no longer easily be increased by modifying DSW_MAX_PORTS
and recompili
>
>
>> <...>
>>
>> > gve_dev_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *stats)
>> > {
>> > uint16_t i;
>> > + if (gve_is_gqi(dev->data->dev_private))
>> > + gve_get_imissed_from_nic(dev);
>> >
>>
>> This updates imissed in RxQ struct for all queues for basic st
On Tue, 16 Jan 2024 05:43:00 +
Madhuker Mythri wrote:
> Hi Stephen,
>
> Yes, I found that the Kernel we use has more Verification added for safe
> execution:
> https://blogs.oracle.com/linux/post/bpf-in-depth-the-bpf-bytecode-and-the-bpf-verifier
>
> Here they mentioned about SKB pointer d
From: Qiming Yang
Remove the maintainer that no longer work.
Signed-off-by: Qiming Yang
---
MAINTAINERS | 2 --
1 file changed, 2 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 0d1c8126e3..230549ca94 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -749,7 +749,6 @@ F: doc/guides/nics/
On Tue, Jan 16, 2024 at 11:45 AM Honnappa Nagarahalli <
honnappa.nagaraha...@arm.com> wrote:
>
> 3. How well does DPDK run on Hyperscaler environments?
> - Can UNH take the task of running DPDK on Hyperscaler environments? UNH
> seems to have lot of things already.
>
Sorry, I could not make techb
On Tue, 16 Jan 2024 23:14:36 +0100
Morten Brørup wrote:
> > +1 for #2 just make it a block.
>
> I prefer that you implement the workaround in the RTE_BUILD_BUG_ON() macro,
> by surrounding it by "do { } while (0)", like this:
>
> #define RTE_BUILD_BUG_ON(condition) do { static_assert(!(condi
> From: Morten Brørup [mailto:m...@smartsharesystems.com]
> Sent: Tuesday, 16 January 2024 23.15
>
> > From: Tyler Retzlaff [mailto:roret...@linux.microsoft.com]
> > Sent: Tuesday, 16 January 2024 22.51
> >
> > On Tue, Jan 16, 2024 at 09:03:01AM -0800, Stephen Hemminger wrote:
> > > Ran into a cor
> From: Tyler Retzlaff [mailto:roret...@linux.microsoft.com]
> Sent: Tuesday, 16 January 2024 22.51
>
> On Tue, Jan 16, 2024 at 09:03:01AM -0800, Stephen Hemminger wrote:
> > Ran into a corner case issue, so sending to mailing list for wider
> discussion.
> >
> > One improvement to DPDK code base
On Tue, Jan 16, 2024 at 09:03:01AM -0800, Stephen Hemminger wrote:
> Ran into a corner case issue, so sending to mailing list for wider discussion.
>
> One improvement to DPDK code base planned is getting rid of variable length
> arrays.
> VLA's can cause bugs and are not supported by the Windows
From: Juraj Linkeš
The Dockerfile defines development and CI runner images.
Signed-off-by: Juraj Linkeš
Signed-off-by: Jeremy Spewock
---
v4:
Remove an example from and updated a comment in the devcontainer.json
and added the --no-root flag to the README to comply with the warning
message and
Both Gcc, clang and MSVC have better way to do compile time
assertions rather than using out of bounds array access.
The old method would fail if -Wvla is enabled because compiler
can't determine size in that code. Also, the use of new
_Static_assert will catch broken code that is passing non-cons
Clang does not handle casts in static_assert() expressions.
It doesn't like use of floating point to calculate threshold.
Use a different expression with same effect; yes this will cause
checkpatch nag.
Signed-off-by: Stephen Hemminger
---
lib/mempool/rte_mempool.c | 4 +---
1 file changed, 1 in
Clang does not allow const variable in a static_assert
expression.
Signed-off-by: Stephen Hemminger
---
drivers/net/i40e/i40e_ethdev.h | 1 +
drivers/net/i40e/i40e_rxtx_vec_sse.c | 10 --
2 files changed, 5 insertions(+), 6 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev.h
The macro RTE_MIN has some hidden assignments to provide type
safety which means the statement can not be fully evaluated in
first pass of compiler. Replace RTE_MIN() with equivalent macro.
This will cause errors from checkpatch about multiple evaluations
of same expression in macro but it is ok i
RTE_BUILD_BUG_ON() was being used with a non-constant value.
The inline function rte_is_power_of_2() is not constant since
inline expansion happens later in the compile process.
Replace it with macro which will be constant.
Fixes: 4236ce9bf5bf ("event/opdl: add OPDL ring infrastructure library")
C
This series fixes a couple places where expressions that could not
be evaluated as constant early in compiler passes were used. And then
converts RTE_BUILD_BUG_ON() with static_assert.
static_assert() is more picky about the expression has to
be a constant, which also catches some existing undefin
With modern CPUs, it is possible to have higher
CPU count thus we can have higher RTE_MAX_LCORES.
In DPDK sample applications, the current config
lcore options are hard limited to 255.
The patchset fixes these constraints by allowing
all lcore IDs up to RTE_MAX_LCORES.
v4:
- fixed build errors w
Currently the config option allows lcore IDs up to 255,
irrespective of RTE_MAX_LCORES and needs to be fixed.
The patch allows config options based on DPDK config.
Fixes: 0e8f47491f09 ("examples/vm_power: add command to query CPU frequency")
Cc: marcinx.hajkow...@intel.com
Cc: sta...@dpdk.org
Si
Currently the config option allows lcore IDs up to 255,
irrespective of RTE_MAX_LCORES and needs to be fixed.
The patch allows config options based on DPDK config.
Fixes: de3cfa2c9823 ("sched: initial import")
Cc: sta...@dpdk.org
Signed-off-by: Sivaprasad Tummala
---
examples/qos_sched/args.c
Currently the config option allows lcore IDs up to 255,
irrespective of RTE_MAX_LCORES and needs to be fixed.
The patch allows config options based on DPDK config.
Fixes: d299106e8e31 ("examples/ipsec-secgw: add IPsec sample application")
Cc: sergio.gonzalez.mon...@intel.com
Cc: sta...@dpdk.org
Currently the config option allows lcore IDs up to 255,
irrespective of RTE_MAX_LCORES and needs to be fixed.
The patch allows config options based on DPDK config.
Fixes: f88e7c175a68 ("examples/l3fwd-power: add high/regular perf cores
options")
Cc: radu.nico...@intel.com
Cc: sta...@dpdk.org
Si
Currently the config option allows lcore IDs up to 255,
irrespective of RTE_MAX_LCORES and needs to be fixed.
The patch allows config options based on DPDK config.
Fixes: 08bd1a174461 ("examples/l3fwd-graph: add graph-based l3fwd skeleton")
Cc: ndabilpu...@marvell.com
Cc: sta...@dpdk.org
Signed-
Currently the config option allows lcore IDs up to 255,
irrespective of RTE_MAX_LCORES and needs to be fixed.
The patch allows config options based on DPDK config.
Fixes: af75078fece3 ("first public release")
Cc: sta...@dpdk.org
Signed-off-by: Sivaprasad Tummala
Acked-by: Konstantin Ananyev
--
With modern CPUs, it is possible to have higher
CPU count thus we can have higher RTE_MAX_LCORES.
In DPDK sample applications, the current config
lcore options are hard limited to 255.
The patchset fixes these constraints by allowing
all lcore IDs up to RTE_MAX_LCORES.
v4:
- fixed build errors w
Ran into a corner case issue, so sending to mailing list for wider discussion.
One improvement to DPDK code base planned is getting rid of variable length
arrays.
VLA's can cause bugs and are not supported by the Windows compiler.
Gcc and Clang have a flag to warn on use of VLA's (-Wvla).
In DPD
Reviewed-by: Juraj Linkeš
On Tue, Jan 16, 2024 at 12:46 PM Luca Vizzarro wrote:
>
> Document the configuration schema in the docs, describing all of the
> relevant definitions and properties.
>
> Reviewed-by: Paul Szczepanek
> Signed-off-by: Luca Vizzarro
Reviewed-by: Juraj Linkeš
On Tue, Jan 16, 2024 at 12:46 PM Luca Vizzarro wrote:
>
> Improve instructions for installing dependencies, configuring and
> launching the project. Add mypy in the list of developer tools.
>
> Finally, fix a bug where Poetry expects to install DTS as a package
> as wel
TB Attendees
-
Bruce
Hemant
Honnappa
Jerin
Kevin
Konstantin
Maxime
Morten
Thomas
NOTE: The technical board meetings are on every second Wednesday at 3 pm UTC.
Meetings are public, and DPDK community members are welcome to attend.
Link to join:
https://zoom-lfx.platform.linuxfound
On Sat, 13 Jan 2024 17:50:19 +
Madhuker Mythri wrote:
> With the original code BPF verifier throws error, due to undefined array
> access and thus could failed to load the BPF instructions onto the Kernel.
> For safe execution on the Kernel, the array access with such undefined value
> thro
On Tue, 16 Jan 2024 05:43:00 +
Madhuker Mythri wrote:
> Hi Stephen,
>
> Yes, I found that the Kernel we use has more Verification added for safe
> execution:
> https://blogs.oracle.com/linux/post/bpf-in-depth-the-bpf-bytecode-and-the-bpf-verifier
>
> Here they mentioned about SKB pointer d
On Tue, 16 Jan 2024 12:33:48 +
"Tummala, Sivaprasad" wrote:
> > > Not sure why cast is even needed, assigning uint32_t to uint16_t is
> > > not going to generate a warning with current compiler settings.
> > >
> >
> > I was assuming compiler will complain when assigning uint32_t to uint16_t
There was no comment on this doc.
It is RFC, is it ready to merge?
Instead of the Linux guide, should we add it to the platform guides?
doc/guides/platform/
We may want to create an entry in MAINTAINERS as well.
More details below.
10/10/2023 17:34, Vipin Varghese:
> Add AMD EPYC SoC t
On Wed, Oct 25, 2023 at 09:40:54AM +0200, Mattias Rönnblom wrote:
> Another option would be to have the current RTE_EVENT_DEV_CAP_BURST_MODE
> capable PMDs start using the "impl_opaque" field for the purpose of matching
> in and out events. It would require applications to actually start adhering
在 2024/1/9 上午12:04, Kumara Parameshwaran 写道:
In the current implementation when a packet is received with
special TCP flag(s) set, only that packet is delivered out of order.
There could be already coalesced packets in the GRO table
belonging to the same flow but not delivered.
This fix makes s
On 1/5/24 22:15, Hernan Vargas wrote:
Cosmetic changes for comments.
No functional impact.
Signed-off-by: Hernan Vargas
---
drivers/baseband/fpga_5gnr_fec/agx100_pmd.h | 4 +-
.../baseband/fpga_5gnr_fec/fpga_5gnr_fec.h| 49 ++--
.../fpga_5gnr_fec/rte_fpga_5gnr_fec.c | 24
Recent version of firmware extended capabilities of setting link state
by adding two extra parameters i.e. timeout and flag disabling auto
enable of rx/tx during linkup.
This change adds supports for both.
Signed-off-by: Tomasz Duszynski
---
drivers/common/cnxk/roc_bphy_cgx.c | 11 +
[AMD Official Use Only - General]
Hi Stephen,
> -Original Message-
> From: Yigit, Ferruh
> Sent: Tuesday, January 9, 2024 8:47 PM
> To: Stephen Hemminger ; Tummala, Sivaprasad
>
> Cc: david.h...@intel.com; anatoly.bura...@intel.com; jer...@marvell.com;
> radu.nico...@intel.com; gak...@m
To reduce flow migration overhead, replace the array-based
representation of which set of ports are bound to a particular queue
by a bitmask-based one.
The maximum number of DSW event ports remains 64, but after this
change can no longer easily be increased by modifying DSW_MAX_PORTS
and recompili
Document the configuration schema in the docs, describing all of the
relevant definitions and properties.
Reviewed-by: Paul Szczepanek
Signed-off-by: Luca Vizzarro
---
v2:
- changed arrays, objects terminology
doc/guides/tools/dts.rst | 208 +--
1 file chang
Improve instructions for installing dependencies, configuring and
launching the project. Add mypy in the list of developer tools.
Finally, fix a bug where Poetry expects to install DTS as a package
as well while resolving the dependencies. Because DTS is not setup
as a package, this results in an
Adding support for parsing multiple representor devargs strings
passed to a PCI BDF. There may be scenario where port representors
for various PFs or VFs under PFs are required and all these are
representor ports shall be backed by single pci device. In such
case port representors can be created us
Following series adds support to enable creation of multiple representors
under one base device. There may be scenarios where port representors for
multiple PFs or VFs under PF are required and all these representor ports
created under a single pci device. Marvell CNXK port representor solution
is
> Subject: [PATCH 00/14] Add TLS record test suite
>
> Add TLS record test suite in cryptodev autotest framework. The test
> suite would run based on capabilities exposed by the cryptodev.
>
> The test suite framework is similar to the framework used in case of
> IPsec tests. To avoid duplication
> Subject: [PATCH v2 00/24] Fixes and improvements in crypto cnxk
>
> Add following features
> - TLS record processing offload (TLS 1.2-1.3, DTLS 1.2)
> - Rx inject to allow lookaside packets to be injected to ethdev Rx
> - Use PDCP_CHAIN opcode instead of PDCP opcode for cipher-only and auth
>
Remove myself from maintainers.
Signed-off-by: Simei Su
---
MAINTAINERS | 2 --
1 file changed, 2 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 0d1c812..11043da 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -739,7 +739,6 @@ F: doc/guides/nics/hinic.rst
F: doc/guides/nics/features/h
> Subject: [PATCH v2 07/24] crypto/cnxk: enable Rx inject in security lookaside
>
> From: Vidya Sagar Velumuri
>
> Add Rx inject fastpath API.
> Add devargs to specify an LF to be used for Rx inject.
It is better to specify the name of the devarg in description.
> When the RX inject feature fla
50 matches
Mail list logo
