> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ferruh Yigit
> Sent: Thursday, January 28, 2021 7:28 PM
>
> On 1/25/2021 1:20 AM, Li Zhang wrote:
> > Currently the flow Meter algorithms in rte_flow
> > only supports bytes per second(BPS).
> > Such as Single Rate Three Color Marker (srTCM rf
This is a new type of reader-writer lock that provides better fairness
guarantees. According to the research it will be better for real
time applications such as DPDK.
A similar implementation is Concurrency Kit available in FreeBSD.
For information see:
"Reader-Writer Synchronization for Shar
>
> 14/01/2021 17:55, Stephen Hemminger:
> > The implementation of reader/writer locks in DPDK (from first release)
> > is simple and fast. But it can lead to writer starvation issues.
> >
> > It is not easy to fix this without changing ABI and potentially
> > breaking customer applications that
Add support for querying receive queue count in order to allow
the rte_eth_dev rx_queue_count() API to be exercised and tested.
Signed-off-by: Lance Richardson
---
app/test-pmd/cmdline.c | 65 +
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 6 ++
2 files
11/02/2021 13:08, Mcnamara, John:
> Thomas,
>
> I didn't remove the "Known Issues" section in case someone else
> adds something to it. If not you can remove it.
OK, I will remove.
> Also, there probably should be a note about the change to pmdinfogen
> in this release. I didn't know enough abo
On Tue, Feb 9, 2021 at 6:40 AM Dmitry Kozlyuk wrote:
>
> On Mon, 8 Feb 2021 14:58:29 +0800, Feng Li wrote:
> > > > +/**
> > > > + * Usage function typedef used by the application usage function.
> > >
> > > It's unrelated to the following typedef purpose, is it?
> > It's borrowed from the front ty
A new DPDK release candidate is ready for testing:
https://git.dpdk.org/dpdk/tag/?id=v21.02-rc4
There are only 20 new patches in this snapshot,
meaning this release cycle is almost closed.
Release notes:
https://doc.dpdk.org/guides/rel_notes/release_21_02.html
Please test and rep
On Thu, 11 Feb 2021 23:05:44 +0100
Thomas Monjalon wrote:
> The default page size can be None, and the page size from user request
> can be 0 kB if lower than 1024. In these cases, a division will fail.
> In order to avoid a Python exception, the page size is checked
> and an error message "Inval
On Thu, 11 Feb 2021 22:44:03 +0100
Thomas Monjalon wrote:
> 11/02/2021 09:20, Sarosh Arif:
> > If user requests a hugepage size which is not supported by the system,
> > currently user gets an error message saying that the requested size
> > is not a valid system huge page size. In addition to th
14/01/2021 17:55, Stephen Hemminger:
> The implementation of reader/writer locks in DPDK (from first release)
> is simple and fast. But it can lead to writer starvation issues.
>
> It is not easy to fix this without changing ABI and potentially
> breaking customer applications that are expect the
11/02/2021 20:07, Stephen Hemminger:
> On Thu, 11 Feb 2021 19:16:59 +0100
> Thomas Monjalon wrote:
>
> > The tool dpdk-hugepages.py, added in DPDK 20.11,
> > is referenced in the guides instead of more complicate commands.
> >
> > The original Linux commands are kept in linux_gsg/sys_reqs.rst
>
> >> A driver can be loaded as a dynamic module or a built-in module.
> >> In commit 681a67288655 ("usertools: check if module is loaded before
> >> binding"), script only checks modules in /sys/module/.
> >>
> >> However, for built-in kernel driver, it only shows up in /sys/module/, if
> >> it has
The default page size can be None, and the page size from user request
can be 0 kB if lower than 1024. In these cases, a division will fail.
In order to avoid a Python exception, the page size is checked
and an error message "Invalid page size" is printed.
A similar error message is printed in set
11/02/2021 09:20, Sarosh Arif:
> If user requests a hugepage size which is not supported by the system,
> currently user gets an error message saying that the requested size
> is not a valid system huge page size. In addition to this if we display
> the valid hugepage sizes it will be convenient fo
> Subject: [PATCH v2] bus/pci: fix probing for non-netuio bound devices
>
> External email: Use caution opening links or attachments
>
>
> Implement rte_pci_map_device() to distinguish between the devices bound
> to netuio and NDIS devices.
> Only return success for the netuio devices.
>
> v2 c
On Thu, 11 Feb 2021 19:16:59 +0100
Thomas Monjalon wrote:
> The tool dpdk-hugepages.py, added in DPDK 20.11,
> is referenced in the guides instead of more complicate commands.
>
> The original Linux commands are kept in linux_gsg/sys_reqs.rst
> and nics/build_and_test.rst.
>
> Suggested-by: Ste
06/12/2020 18:43, Stephen Hemminger:
> On Mon, 7 Dec 2020 01:31:14 +0800
> walimis...@gmail.com wrote:
>
> > From: Wang Liming
> >
> > Remove redundant "nr_hugepages" words from mlx4 and mlx5 doc.
> >
> > Signed-off-by: Wang Liming
> > ---
> > doc/guides/nics/mlx4.rst | 2 +-
> > doc/guides/
The tool dpdk-hugepages.py, added in DPDK 20.11,
is referenced in the guides instead of more complicate commands.
The original Linux commands are kept in linux_gsg/sys_reqs.rst
and nics/build_and_test.rst.
Suggested-by: Stephen Hemminger
Signed-off-by: Thomas Monjalon
---
doc/guides/cryptodevs
Adding wrapper script for cppcheck code analysis tool.
usage: cppcheck.sh [-h] [path]
Without any argument current folder scanned with all sub folders, if
'path' argument provided, that 'path' is scanned with all its
subfolders, like:
"./devtools/cppcheck.sh lib/librte_ethdev/"
The output log is
On Thu, 11 Feb 2021 13:20:04 +0500
Sarosh Arif wrote:
> If user requests a hugepage size which is not supported by the system,
> currently user gets an error message saying that the requested size
> is not a valid system huge page size. In addition to this if we display
> the valid hugepage sizes
Hi Aaron,
I have disabled the pipeline. Here is the bug ticket so that we can
keep track of the status: https://bugs.dpdk.org/show_bug.cgi?id=637
Attached is the dockerfile and the script that it uses to run SPDK compilation.
Thanks,
Brandon
On Thu, Feb 11, 2021 at 9:02 AM Aaron Conole wrote:
>
On Mon, 14 Sep 2020 14:04:05 +0100
Anatoly Burakov wrote:
> Currently, we don't detach the shared memory on EAL cleanup, which
> leaves the page table descriptors still holding on to the file
> descriptors as well as memory space occupied by them. Fix it by adding
> another detach stage that clos
11/02/2021 10:31, Bruce Richardson:
> On Thu, Feb 11, 2021 at 12:19:37PM +0500, Sarosh Arif wrote:
> > replace testpmd with dpdk-testpmd in all commands
> > because on compilation through meson, dpdk-testpmd is the default
> > application name.
> >
> > Signed-off-by: Sarosh Arif
> > ---
> Thanks
1. It is not a error if no network device available. One can
only use crypto device
2. Improve logging for failure in detecting the bus
Signed-off-by: Hemant Agrawal
---
drivers/bus/dpaa/dpaa_bus.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/drivers/bus/dpaa/dpaa_
From: Nipun Gupta
The current driver only support single buffer pool on a given
pmd instance. return error, if trying to configure more.
Signed-off-by: Nipun Gupta
---
drivers/mempool/dpaa/dpaa_mempool.c | 1 +
drivers/net/dpaa/dpaa_ethdev.c | 6 ++
2 files changed, 7 insertions(+)
d
From: Nipun Gupta
DPAA can share an interface on classification criteria with kernel.
This patch enables default kernel driver to be used as a shared MAC
interface with DPDK interface. (provided that VSP is enabled on that
interface.)
Signed-off-by: Nipun Gupta
---
drivers/bus/dpaa/base/fman/
Secondary process also need the access the qman and bman
ccsr map.
Signed-off-by: Hemant Agrawal
---
drivers/bus/dpaa/dpaa_bus.c | 26 --
1 file changed, 12 insertions(+), 14 deletions(-)
diff --git a/drivers/bus/dpaa/dpaa_bus.c b/drivers/bus/dpaa/dpaa_bus.c
index 662cbf
From: Nipun Gupta
CGR are automatically freed up in the kernel. As we do not cleanup
the queues, if we release the CGR here, kernel reports them in use.
So have them freed up in the kernel
Signed-off-by: Nipun Gupta
---
drivers/net/dpaa/dpaa_ethdev.c | 6 --
1 file changed, 6 deletions(-)
DPAA2 DPBP object access need availability of MCP object
pointer. In case of secondary process, we need to use local
MCP pointer instead of primary process.
Signed-off-by: Nipun Gupta
Signed-off-by: Hemant Agrawal
---
drivers/mempool/dpaa2/dpaa2_hw_mempool.c | 8 +++-
1 file changed, 7 inse
From: Nipun Gupta
In case error packets are received by the ethernet interface,
this patch enables the receival of packets on the error queue,
printing the error and the error packet.
to enable, use the dev_arg as : fslmc:dpni.1,drv_error_queue=1
Signed-off-by: Nipun Gupta
Acked-by: Hemant Agr
Add support for dev arg ``fslmc:dpni.1,drv_tx_conf=1``
It is optional for dpaa2 to use TX confirmation. DPAA2
can free the transmitted packets. However some use-case
requires the TX confirmation to be explicit.
Signed-off-by: Hemant Agrawal
---
doc/guides/nics/dpaa2.rst| 4
driver
DPDMUX should not drop parse error packets. They shall
be left to the decision of the connected DPNI interfaces
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa2/dpaa2_mux.c | 18
drivers/net/dpaa2/mc/dpdmux.c | 37 +
drivers/net/dpaa2/mc/fsl_dpdmux.h | 113 +
From: Apeksha Gupta
This is require as the interface is shared with linux and we do not want
the dpdmux to be reset. rather the default interface to kernel shall
continue.
command dpdmux_set_resetable is used to skip DPDMUX reset.
By default the DPDMUX_RESET command will reset the DPDMUX complete
From: Akhil Goyal
Added support for flow raw pattern and check that the call
for dpdmux_set_custom_key() which should be called
only once for a particular DPDMUX as all previous rules
will be erased with this call.
Hence calling it for the first time only.
Signed-off-by: Akhil Goyal
---
driver
From: Gagandeep Singh
Add basic support for scheduling and shaping on dpaa2
platform.
HW supports 2 level of scheduling and shaping.
However the current patch only support single level.
Signed-off-by: Gagandeep Singh
Acked-by: Hemant Agrawal
---
doc/guides/nics/dpaa2.rst | 120
This patch introduce a new pmd api, which can help the applications
to configure the max framelen for a given dpdmux interface
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa2/dpaa2_mux.c | 28 +++-
drivers/net/dpaa2/rte_pmd_dpaa2.h | 18 +-
drivers
This patch support tx of external buffers
Signed-off-by: Hemant Agrawal
---
drivers/net/dpaa/dpaa_rxtx.c | 39 +++-
drivers/net/dpaa/dpaa_rxtx.h | 8 +---
2 files changed, 31 insertions(+), 16 deletions(-)
diff --git a/drivers/net/dpaa/dpaa_rxtx.c b/drivers/
This patch enables support to use the first segment
headroom to build the HW requied Scatter Gather
Table. (if space is available).
This will avoid 1 less buffer for SG buffer creation.
Signed-off-by: Sachin Saxena
Signed-off-by: Hemant Agrawal
---
drivers/bus/fslmc/portal/dpaa2_hw_pvt.h | 5 +
From: Nipun Gupta
This patch support tx of external allocated buffers.
Signed-off-by: Nipun Gupta
Signed-off-by: Sachin Saxena
Acked-by: Hemant Agrawal
---
drivers/net/dpaa2/dpaa2_rxtx.c | 42 ++
1 file changed, 42 insertions(+)
diff --git a/drivers/net/dpaa2
From: Rohit Raj
According to DPDK Documentation, rte_eth_link_get API can wait upto 9
seconds for auto-negotiation to finish and then returns link status.
In current implementation of rte_eth_link_get API in DPAA drivers, it
wasn't waiting for auto negotiation to finish and was returning link
st
From: Rohit Raj
According to DPDK Documentation, rte_eth_link_get API can wait upto 9
seconds for auto-negotiation to finish and then returns link status.
In current implementation of rte_eth_link_get API in DPAA2 drivers, it
wasn't waiting for auto negotiation to finish and was returning link
s
From: Nipun Gupta
Reading of word un-aligned values after reading word aligned
values lead to corrution of memory, such that older value.
This patch make changes such that word alignedaccess is made,
before making an un-aligned access
Fixes: 6d6b4f49a155 ("bus/dpaa: add FMAN hardware operations"
From: Youri Querry
Random portal hangs observed on device with QBMAN 5.0
This fixes few random packet hang issues in event mode.
Few things fixed it.
1. generally, pi == ci, no need for extra checks.
2. the proper initilizations in init with ci helped
Fixes: 1b49352f41be ("bus/fslmc: rename por
This patch series adds changes to dpaaX ethernet pmd
1. Basic Traffic Management support for shaper and scheduler
2. Support external buffer in Transmit side.
3. Optimize the SG packet handling
4. Add support APIs in DPDMUX (SRIOV equivalent for DPAA2)
5. Fix isssues
v2: add more patches and add
Brandon Lo writes:
> Hi again everyone,
>
> I have checked the pipelines with SPDK branch v21.01.x on the main DPDK
> branch.
> It still seems to have an issue with compilation, and I have attached
> a log of a Fedora SPDK compilation.
> There are some undefined references to "rte_ether_unformat
11/02/2021 14:25, Hemant Agrawal:
> Add tested platforms with NXP SoCs to the 21.02 release notes.
>
> Signed-off-by: Hemant Agrawal
> ---
> + * ARMv8 SoC combinations from NXP (with integrated NICs)
For consistency, I reword this line to:
* NXP ARMv8 SoCs (with integrated NICs)
> +
>
Add tested platforms with NXP SoCs to the 21.02 release notes.
Signed-off-by: Hemant Agrawal
---
doc/guides/rel_notes/release_21_02.rst | 12
1 file changed, 12 insertions(+)
diff --git a/doc/guides/rel_notes/release_21_02.rst
b/doc/guides/rel_notes/release_21_02.rst
index 84b04a0
Add tested platforms with NXP SoCs to the 21.02 release notes.
Signed-off-by: Hemant Agrawal
---
doc/guides/rel_notes/release_21_02.rst | 12
1 file changed, 12 insertions(+)
diff --git a/doc/guides/rel_notes/release_21_02.rst
b/doc/guides/rel_notes/release_21_02.rst
index 84b04a0
On 14-Sep-20 12:26 PM, Anatoly Burakov wrote:
Currently, we don't detach the shared memory on EAL cleanup, which
leaves the page table descriptors still holding on to the file
descriptors as well as memory space occupied by them. Fix it by adding
another detach stage that closes the internal memo
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 to the custom
target.
The user supplie
>* Release pushed to *Sunday, 14 February 2021*
Probably pure coincidence that the DPDK February releases from the last few
years usually takes place on St. Valentine's Day :-)
I propose May 5th as the release date for 21.05 to mark Napoleon's 200 years
death anniversary.
> -Original Message-
> From: Mcnamara, John
> Sent: Thursday, February 11, 2021 12:04 PM
> To: dev@dpdk.org
> Cc: tho...@monjalon.net; Mcnamara, John
> Subject: [PATCH v1] doc: update release notes for 21.02
>
> Fix grammar, spelling and formatting of DPDK 21.02 release notes.
Thomas,
Fix grammar, spelling and formatting of DPDK 21.02 release notes.
Signed-off-by: John McNamara
---
doc/guides/rel_notes/release_21_02.rst | 24
1 file changed, 12 insertions(+), 12 deletions(-)
diff --git a/doc/guides/rel_notes/release_21_02.rst
b/doc/guides/rel_notes/
Meeting minutes of 11 February 2021
---
Agenda:
* Release Dates
* -rc3 status
* Subtrees
* LTS
* Opens
Participants:
* Canonical
* Debian/Microsoft
* Intel
* NXP
* Nvidia
Release Dates
-
* v21.02 dates
* -rc3 is released on Friday, 5 February 2021
On 19-Nov-20 3:52 AM, Ferruh Yigit wrote:
Removing unused functions, reported by cppcheck.
Easy way to remove clutter, since the code is already in the git repo,
they can be added back when needed.
Signed-off-by: Ferruh Yigit
---
diff --git a/lib/librte_eal/linux/eal_memory.c
b/lib/librt
On 15-Jan-21 7:32 AM, Nithin Dabilpuram wrote:
Currently external memory test uses 4K page size.
VFIO DMA mapping works only with system page granularity.
Earlier it was working because all the contiguous mappings
were coalesced and mapped in one-go which ended up becoming
a lot bigger page. Now
11/02/2021 07:22, Ajit Khaparde:
> Add tested platforms for Broadcom NICs.
>
> Signed-off-by: Ajit Khaparde
Applied, thanks
10/02/2021 13:30, Raslan Darawsheh:
> Add tested platforms with Mellanox NICs to the 21.02 release notes.
>
> Signed-off-by: Raslan Darawsheh
Applied, thanks
On 24-Nov-20 2:27 PM, Ajit Khaparde wrote:
On Tue, Nov 24, 2020 at 5:26 AM Anatoly Burakov
wrote:
Currently, compilation instructions for igb_uio are missing. Add them,
as well as a top-level INSTALL file referring users to per-OS
subdirectories.
Signed-off-by: Anatoly Burakov
Reviewed-by: F
On 27-Nov-20 1:21 PM, Thomas Monjalon wrote:
27/11/2020 13:56, Burakov, Anatoly:
On 22-Nov-20 6:16 PM, Thomas Monjalon wrote:
That's a pity we didn't get this patch in DPDK 20.11.
Anatoly, Stephen, what happened? It is not interesting anymore?
Anyone else to review?
It is a good patch and
On 17-Aug-20 3:08 PM, Renata Saiakhova wrote:
Avoid race with unregister interrupt hanlder if interrupt
source has some active callbacks at the moment, use wrapper
around rte_intr_callback_unregister() to check for -EAGAIN
return value and to loop until rte_intr_callback_unregister()
succeeds.
S
On 03-Dec-20 8:25 AM, Liu, Yongxin wrote:
Hi Anatoly,
Do you have any further comments on this v5?
Or you can submit your own patch directly.
I am really expecting this issue to be fixed.
Thank you very much.
Yongxin
-Original Message-
From: dev On Behalf Of Yongxin Liu
Sent: Mond
Add tested platforms for Broadcom NICs.
Signed-off-by: Ajit Khaparde
---
doc/guides/rel_notes/release_21_02.rst | 63 ++
1 file changed, 63 insertions(+)
diff --git a/doc/guides/rel_notes/release_21_02.rst
b/doc/guides/rel_notes/release_21_02.rst
index baf32151f3..9ac44
09/02/2021 12:17, Xueqin Lin:
> From: ChenBo
>
> Add tested Intel platforms with Intel NICs to v21.02 release note.
>
> Signed-off-by: ChenBo
Applied, thanks
On Thu, Feb 11, 2021 at 12:19:37PM +0500, Sarosh Arif wrote:
> replace testpmd with dpdk-testpmd in all commands
> because on compilation through meson, dpdk-testpmd is the default
> application name.
>
> Signed-off-by: Sarosh Arif
> ---
Thanks.
Acked-by: Bruce Richardson
If user requests a hugepage size which is not supported by the system,
currently user gets an error message saying that the requested size
is not a valid system huge page size. In addition to this if we display
the valid hugepage sizes it will be convenient for the user to request
the right size ne
66 matches
Mail list logo
