FDIR ID parsing will not be handled correctly after queue reconfigured,
enable FDIR ID parsing per Q regardless of fdir_ref_cnt to fix it.
Fixes: f71dbf852d46 ("net/iavf: add flow director enabled switch value")
Signed-off-by: Leyi Rong
---
drivers/net/iavf/iavf_rxtx.h | 4 ++--
1 file changed,
On 05/18, Qi Zhang wrote:
>Use a common function when selecting the proper word and mask match for
>a tunnel type when programming switch rules.
>
>Store switch recipe field mask as little endian, which avoids needing to
>convert back to big endian after reading recipe from FW.
>
>Obtain word mask
Update the description and limitation about ice PMD according to the
product release strategy.
Signed-off-by: Xiaolong Ye
---
doc/guides/nics/ice.rst | 29 +
1 file changed, 5 insertions(+), 24 deletions(-)
diff --git a/doc/guides/nics/ice.rst b/doc/guides/nics/ice.r
> -Original Message-
> From: Xing, Beilei
> Sent: Wednesday, May 20, 2020 12:14 PM
> To: Zhao1, Wei ; dev@dpdk.org
> Cc: sta...@dpdk.org; Guo, Jia
> Subject: RE: [PATCH] net/i40e: fix flow FDIR enable issue
>
>
>
> > -Original Message-
> > From: Zhao1, Wei
> > Sent: Tuesday
> -Original Message-
> From: Zhao1, Wei
> Sent: Tuesday, May 19, 2020 3:17 PM
> To: dev@dpdk.org
> Cc: sta...@dpdk.org; Xing, Beilei ; Guo, Jia
> ; Zhao1, Wei
> Subject: [PATCH] net/i40e: fix flow FDIR enable issue
>
> When we flush flow FDIR, all queues are disabled for FDIR.
> If FD
Hi, guojia
> -Original Message-
> From: Guo, Jia
> Sent: Wednesday, May 20, 2020 11:19 AM
> To: Zhao1, Wei ; dev@dpdk.org
> Cc: sta...@dpdk.org; Xing, Beilei
> Subject: Re: [PATCH] net/i40e: fix flow FDIR enable issue
>
> hi, zhaowei
>
> On 5/19/2020 3:16 PM, Wei Zhao wrote:
> > When
Update documentation for 'show fwd' testpmd runtime function to show
CPU cycles/packet example.
Signed-off-by: Dharmik Thakkar
Reviewed-by: Honnappa Nagarahalli
Reviewed-by: Phil Yang
---
v3:
- Keep existing stats and add a note providing CPU cycles/packet
example (Ferruh)
- Remove platfor
Update testpmd documentation to include RECORD configuration options,
CONFIG_RTE_TEST_PMD_RECORD_CORE_CYCLES and
CONFIG_RTE_TEST_PMD_RECORD_BURST_STATS.
Signed-off-by: Dharmik Thakkar
Reviewed-by: Honnappa Nagarahalli
Reviewed-by: Phil Yang
---
v3:
- Replace config/common_base with build/.conf
hi, zhaowei
On 5/19/2020 3:16 PM, Wei Zhao wrote:
When we flush flow FDIR, all queues are disabled for FDIR.
If FDIR rule is created again, then the flow list is empty,
as it is the first time to create rule after flush fdir filter,
so we need to enable FDRI for all queues. And also, disable FDI
Hi, David Marchand && guys
I am using 20.05-rc2 version to test based on HNS3 NIC hardware, and
found that after starting l3fwd-power,
using ctrl+c cannot force quit. But I revert the patch(33666b4 service:
fix crash on exit) and it is ok.
the log as follows:
Initializing rx queues on l
Acked-by: Qiming Yang
> -Original Message-
> From: Zhang, Qi Z
> Sent: Monday, May 18, 2020 12:24
> To: Yang, Qiming
> Cc: dev@dpdk.org; Ye, Xiaolong ; Zhang, Qi Z
> ; sta...@dpdk.org; Nowlin, Dan
>
> Subject: [PATCH] net/ice: fix tunnel type match word handling
>
> Use a common funct
From: Tonghao Zhang
If user don't set the dv_xmeta_en to 1 or 2,
in the flow_dv_convert_action_set_meta function:
* flow_dv_get_metadata_reg may return the REG_NONE,
when MLX5_METADATA_FDB enabled for metadata set
action.
* reg_to_field(REG_NONE) return MLX5_MODI_OUT_NONE
which is invalid.
> On May 19, 2020, at 2:42 AM, Thomas Monjalon wrote:
>
>>> Update testpmd documentation to include RECORD configuration options,
>>> CONFIG_RTE_TEST_PMD_RECORD_CORE_CYCLES and
>>> CONFIG_RTE_TEST_PMD_RECORD_BURST_STATS.
>>>
>>> Signed-off-by: Dharmik Thakkar
>>> Reviewed-by: Honnappa Nagarah
On 5/19/2020 1:18 PM, Thomas Monjalon wrote:
19/05/2020 21:57, Dmitry Kozlyuk:
On Tue, 19 May 2020 20:49:50 +0200
Thomas Monjalon wrote:
+Cc more maintainers
19/05/2020 20:41, tal...@mellanox.com:
From: Tal Shnaiderman
Using uint32_t type bit-fields in Windows will pads the
'L2/L3/L4 and
19/05/2020 21:57, Dmitry Kozlyuk:
> On Tue, 19 May 2020 20:49:50 +0200
> Thomas Monjalon wrote:
>
> > +Cc more maintainers
> >
> > 19/05/2020 20:41, tal...@mellanox.com:
> > > From: Tal Shnaiderman
> > >
> > > Using uint32_t type bit-fields in Windows will pads the
> > > 'L2/L3/L4 and tunnel i
On Tue, 19 May 2020 20:49:50 +0200
Thomas Monjalon wrote:
> +Cc more maintainers
>
> 19/05/2020 20:41, tal...@mellanox.com:
> > From: Tal Shnaiderman
> >
> > Using uint32_t type bit-fields in Windows will pads the
> > 'L2/L3/L4 and tunnel information' union with additional bits.
> >
> > This
A new DPDK release candidate is ready for testing:
https://git.dpdk.org/dpdk/tag/?id=v20.05-rc3
There are 144 new patches in this snapshot.
Release notes:
http://doc.dpdk.org/guides/rel_notes/release_20_05.html
Some deprecation notices might be added if reviewed on time.
The -rc4
+Cc more maintainers
19/05/2020 20:41, tal...@mellanox.com:
> From: Tal Shnaiderman
>
> Using uint32_t type bit-fields in Windows will pads the
> 'L2/L3/L4 and tunnel information' union with additional bits.
>
> This padding causes rte_mbuf size misalignment and the total size
> increases to 3
From: Tal Shnaiderman
Using uint32_t type bit-fields in Windows will pads the
'L2/L3/L4 and tunnel information' union with additional bits.
This padding causes rte_mbuf size misalignment and the total size
increases to 3 cache-lines.
Changed packet_type bit-fields types from uint32_t to uint8_t
Hello,
I'm trying to build an accurate capture device based on Mellanox
Connect-X5 with following requirements:
- capture every incoming packets with hardware timestamps
- output: pcap with timestamps in nanoseconds
My problem is that the packets forwarded to `dpdk-pdump` carry raw
timestamps from
11/05/2020 11:22, Vladimir Medvedkin:
> Check returned value after strtok()
> CID 355674 (#1 of 1): Dereference null return value (NULL_RETURNS)
> 4. dereference: Dereferencing a pointer that might be NULL s when
> calling inet_pton
>
> Fixes: 103809d032cd ("app/test-fib: add test application for
14/05/2020 12:39, Nithin Dabilpuram:
> On Thu, May 14, 2020 at 01:52:17AM +0530, pbhagavat...@marvell.com wrote:
> > From: Pavan Nikhilesh
> >
> > Fix unchecked return values reported by coverity.
> >
> > Coverity issue: 350601
> > Fixes: ef853f1fd979 ("examples/l3fwd-graph: add ethdev configura
On Tue, May 19, 2020 at 5:35 PM David Marchand
wrote:
>
> On Thu, Apr 30, 2020 at 7:54 AM Haiyue Wang wrote:
> >
> > INTERNAL is new introduced version, update the shell script that checks
> > whether built libraries are versioned with expected ABI (current ABI,
> > current ABI + 1, EXPERIMENTAL,
When the primary device link state is queried, there is no
need to query the VF state as well. The application only sees
the state of the synthetic device.
Fixes: dc7680e8597c ("net/netvsc: support integrated VF")
Signed-off-by: Stephen Hemminger
---
drivers/net/netvsc/hn_ethdev.c | 4 +-
drive
The original code would deadlock itself if a link change event
happened with link state interrupt enabled. The problem is that
the link state changed message would be seen while reading
the host to guest ring (under lock) and then the driver would
send a query to the host to see the new link state.
Because of bugs in driver or host a reply to a request might
never occur. Better to give an error than spin forever.
Fixes: 4e9c73e96e83 ("net/netvsc: add Hyper-V network device")
Signed-off-by: Stephen Hemminger
---
drivers/net/netvsc/hn_rndis.c | 30 --
1 file chang
The code to unset owner of VF device was changing port to invalid
value before calling unset.
Fixes: 4a9efcddaddd ("net/netvsc: fix VF support with secondary process")
Signed-off-by: Stephen Hemminger
---
drivers/net/netvsc/hn_vf.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
dif
These functions are useful for applications and debugging.
The netvsc PMD also transparently handles the rx/tx descriptor
functions for underlying VF device.
Signed-off-by: Stephen Hemminger
---
drivers/net/netvsc/hn_ethdev.c | 3 ++
drivers/net/netvsc/hn_rxtx.c | 57 +
The PMD_TX_LOG and PMD_RX_LOG can hide errors since this
debug log is typically disabled. Change the code to use
PMD_DRV_LOG for errors.
Under load, the ring buffer to the host can fill.
Add some statistics to estimate the impact and see other errors.
Signed-off-by: Stephen Hemminger
---
driver
This set of patches includes bugfixes for races in VF handling
and implements the queue and descriptor status functions in netvsc PMD.
This patch set is likely to cause merge warnings for the automated
tools because it is based on top of my previous netvsc PMD patches
already in net-next branch.
There is not a lot of info here from this driver.
But worth supporting these additional info queries.
Signed-off-by: Stephen Hemminger
---
drivers/net/netvsc/hn_ethdev.c | 2 ++
drivers/net/netvsc/hn_rxtx.c | 20
drivers/net/netvsc/hn_var.h| 2 ++
3 files changed, 24
13/05/2020 22:20, pbhagavat...@marvell.com:
> From: Pavan Nikhilesh
>
> Fix unchecked return values reported by coverity.
>
> Coverity issue: 354235
> Fixes: 8bd537e9c6cf ("examples/l3fwd: add service core setup based on caps")
>
> Signed-off-by: Pavan Nikhilesh
Applied, thanks
Hi
Thank you for the feedback.
While I'm ok with the fix you suggested, the warning was placed in order to let
the user understand that his/her MTU settings may not impact DPDK. After all in
this case(no JUMBO) we are not doing anything besides changing the
dev->data->mtu.
I think there is som
15/05/2020 13:29, Ferruh Yigit:
> On 5/14/2020 5:09 AM, Rasesh Mody wrote:
> > This patch adds a fix to setup Tx queue when changing KNI interface MTU.
> > It ensures device can safely start txq post MTU change operation.
> >
> > Fixes: fc9ee41b7016 ("examples/kni: convert to new ethdev offloads A
11/05/2020 12:07, pbhagavat...@marvell.com:
> From: Pavan Nikhilesh
>
> Add device arguments to lock NPA aura and pool contexts in NDC cache.
> The device args take hexadecimal bitmask where each bit represent the
> corresponding aura/pool id.
> Example:
> -w 0002:02:00.0,npa_lock_mask=0xf
27/04/2020 17:04, Ferruh Yigit:
> On 4/27/2020 3:25 PM, Bruce Richardson wrote:
> > When anyone uses the make build system, they are to be informed
> > about upcoming plans to deprecate and subsequently remove that
> > system and to use meson and ninja instead.
> >
> > Signed-off-by: Bruce Richard
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Stephen Hemminger
> Sent: Tuesday, May 19, 2020 5:52 PM
>
> On Tue, 19 May 2020 15:27:25 +
> Morten Brørup wrote:
>
> > diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
> > index 9078e7c24..f67141482 100644
> > --- a/
11/05/2020 12:27, Bruce Richardson:
> On Mon, May 11, 2020 at 11:25:13AM +0100, Anatoly Burakov wrote:
> > Now that kernel modules aren't built by default, we can be more
> > strict with their build process, and fail the build if they were
> > requested to be built, but weren't.
> >
> > Signed-off
On Tue, 19 May 2020 15:27:25 +
Morten Brørup wrote:
> diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
> index 9078e7c24..f67141482 100644
> --- a/lib/librte_ring/rte_ring.h
> +++ b/lib/librte_ring/rte_ring.h
> @@ -733,7 +733,9 @@ rte_ring_full(const struct rte_ring *r)
>
On Thu, Apr 30, 2020 at 7:54 AM Haiyue Wang wrote:
>
> INTERNAL is new introduced version, update the shell script that checks
> whether built libraries are versioned with expected ABI (current ABI,
> current ABI + 1, EXPERIMENTAL, or INTERNAL).
>
> Signed-off-by: Haiyue Wang
> ---
> devtools/ch
Testing if the ring is empty is as simple as comparing the producer and
consumer pointers.
In theory, this optimization reduces the number of potential cache misses
from 3 to 2 by not having to read r->mask in rte_ring_count().
The modification of this function were also discussed in the RFC here
Fix coding style violations that checkpatch will complain about.
Add missing "int" after "unsigned".
Add missing spaces around "+=" and "+".
Remove superfluous type cast of numerical constant.
Suggested-by: Konstantin Ananyev
Signed-off-by: Morten Brørup
---
lib/librte_ring/rte_ring.c |
Testing if the ring is empty is as simple as comparing the producer and
consumer pointers.
Checkpatch complains about existing coding style violations, so the first
part of the patch fixes those, and contains no functional changes.
Morten Brørup (2):
ring: coding style cleanup
ring: empty opt
On Tue, 19 May 2020 16:22:58 +0530
Hemant Agrawal wrote:
> +/* To test free the resources from bitrate_reg test */
> +static int
> +test_stats_bitrate_free(void)
> +{
> + int ret = 0;
> +
> + ret = rte_metrics_deinit();
Coverity will complain about extra initialization like this
Meeting notes for the DPDK technical board meeting held on 2020-05-06
Attendees:
- Bruce Richardson
- Ferruh Yigit
- Hemant Agrawal
- Jerin Jacob (Chair)
- Kevin Traynor
- Konstantin Ananyev
- Olivier Matz
- Stephen Hemminger
On Tue, 19 May 2020 16:22:57 +0530
Hemant Agrawal wrote:
> diff --git a/lib/librte_metrics/rte_metrics.c
> b/lib/librte_metrics/rte_metrics.c
> index e07670219f..dba6409c27 100644
> --- a/lib/librte_metrics/rte_metrics.c
> +++ b/lib/librte_metrics/rte_metrics.c
> @@ -85,6 +85,7 @@ rte_metrics_de
On Thu, 14 May 2020 15:00:00 +0530
Hemant Agrawal wrote:
>
> +int
> +rte_stats_bitrate_free(struct rte_stats_bitrates *bitrate_data)
> +{
> + if (bitrate_data)
> + rte_free(bitrate_data);
> + return 0;
> +}
rte_free of NULL is allowed, the test is not needed.
Should be voi
On Thu, Apr 30, 2020 at 7:54 AM Haiyue Wang wrote:
>
> INTERNAL is new introduced version, update the script that automatically
> leaving internal section exactly as it is.
>
> Signed-off-by: Haiyue Wang
> ---
> devtools/update_version_map_abi.py | 37 +++---
> 1 file cha
On 5/18/2020 10:27 AM, Shy Shyman wrote:
> MTU is used in testpmd to set the maximum payload size for packets.
> According to testpmd, the setting influnce RX only.
> In rte_ethdev there's no relation between MTU setting and JUMBO offload or
> rx_max_pkt_len.
>
> The previous fix in patch referenc
07/05/2020 12:13, Darek Stojaczyk:
> Even when a PMD was disabled with meson's disable_drivers option
> its config file was still being parsed. Some of the PMD configs
> attempt to find a library they depend on and parse its header files
> with certain assumptions. If the library is found, but it's
19/05/2020 15:26, Dumitrescu, Cristian:
> From: Yigit, Ferruh
> >
> > On v20.02 some meter APIs have been matured and symbols moved from
> > EXPERIMENTAL to DPDK_20.0.1 block.
> >
> > This can break the applications that were using these mentioned APIs on
> > v19.11. Although there is no modific
On 19/05/2020 13:16, Ferruh Yigit wrote:
> On v20.02 some meter APIs have been matured and symbols moved from
> EXPERIMENTAL to DPDK_20.0.1 block.
>
> This can break the applications that were using these mentioned APIs on
> v19.11. Although there is no modification on the APIs and the action i
From: Tal Shnaiderman
Uses SetupAPI.h functions to scan PCI tree.
Uses DEVPKEY_Device_Numa_Node to get the PCI NUMA node.
Uses SPDRP_BUSNUMBER and SPDRP_BUSNUMBER to get the BDF.
scanning currently supports types RTE_KDRV_NONE.
Signed-off-by: Tal Shnaiderman
---
drivers/bus/pci/windows/pci.c
From: Tal Shnaiderman
import library (/IMPLIB) in meson.build should use
the 'drivers' and not 'libs' folder.
The error is: fatal error LNK1149: output filename matches input filename.
The fix uses the correct folder.
Fixes: 5ed3766981 ("drivers: process shared link dependencies as for libs")
From: Tal Shnaiderman
pmdinfogen generation is currently unsupported for Windows.
The relevant part in meson.build is skipped.
Signed-off-by: Tal Shnaiderman
---
drivers/meson.build | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/meson.build b/drivers/meson.build
From: Tal Shnaiderman
the struct rte_pci_addr defines domain as uint32_t variable however
the PCI_PRI_FMT macro used for logging the struct sets the format
of domain to uint16_t.
The mismatch causes the following warning messages
in Windows clang build:
format specifies type 'unsigned short' bu
From: Tal Shnaiderman
Changing all of PCIs Unix memory mapping to the
new memory allocation API wrapper.
Change all of PCI mapping function usage in
bus/pci to support the new API.
Signed-off-by: Tal Shnaiderman
---
drivers/bus/pci/bsd/pci.c| 2 +-
drivers/bus/pci/linux/pci_uio.c |
From: Tal Shnaiderman
Addition of stub eal and bus/pci functions to compile
bus/pci for Windows.
Signed-off-by: Tal Shnaiderman
---
drivers/baseband/meson.build | 4 +
drivers/bus/ifpga/meson.build | 6 ++
drivers/bus/pci/meson.build| 14 ++-
drivers/bus/pci/pci_common.
From: Tal Shnaiderman
Move common functions between Unix and Windows to eal_common_config.c.
Those simple functions are getter functions for IOVA,
configuration, Multi-process.
Move rte_config and runtime_dir to be defined in a common file.
Signed-off-by: Tal Shnaiderman
---
lib/librte_eal/c
From: Tal Shnaiderman
This patchset implements the EAL and PCI functions needed for probing PMDs
using RTE_KDRV_NONE on Windows.
Depends-on: series-9713 ("Windows basic memory management")
v4:
* Fixed various warnings and naming conventions(DmitryK).
* Fixed broken mingw-64 bui
From: Tal Shnaiderman
Added in Windows header file includes
to include off_t type since it is needed by PCI.
export per_lcore__rte_errno from eal since it is needed by PCI.
Signed-off-by: Tal Shnaiderman
---
lib/librte_eal/rte_eal_exports.def | 1 +
lib/librte_eal/rte_eal_version.map
Hi Shy
> -Original Message-
> From: Shy Shyman
> Sent: Monday, May 18, 2020 10:27 AM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo ; Xing, Beilei
> ; Iremonger, Bernard
> ; xavier.hu...@huawei.com
> Subject: [PATCH] app/testpmd: fix error detection in MTU command
>
> MTU is used in testpmd to set
On 18/05/2020 18:18, Thomas Monjalon wrote:
> 16/05/2020 13:53, Neil Horman:
>> On Fri, May 15, 2020 at 04:01:53PM +0100, Ray Kinsella wrote:
>>> From: Ferruh Yigit
>>>
>>> On v20.02 some APIs matured and symbols moved from EXPERIMENTAL to
>>> DPDK_20.0.1 block.
>>>
>>> This had the affect of b
On 18/05/2020 19:32, Ferruh Yigit wrote:
> On 5/18/2020 6:51 PM, Thomas Monjalon wrote:
>> 18/05/2020 19:34, Ferruh Yigit:
>>> On 5/18/2020 6:18 PM, Thomas Monjalon wrote:
16/05/2020 13:53, Neil Horman:
> On Fri, May 15, 2020 at 04:01:53PM +0100, Ray Kinsella wrote:
>> From: Ferruh
19/03/2020 09:33, David Marchand:
> On Fri, Mar 13, 2020 at 8:55 PM Kevin Traynor wrote:
> >
> > The versions currently listed as maintained have gone stale.
> >
> > Rather than having to keep updating this doc, point to the
> > dpdk.org stable roadmap.
> >
> > Signed-off-by: Kevin Traynor
>
> A
04/05/2020 15:45, jer...@marvell.com:
> From: Jerin Jacob
>
> Glob is the terminology used in fnmatch man page.
> Use glob terminology across DPDK for shell pattern.
>
> Signed-off-by: Jerin Jacob
> ---
>
> v3:
> - Address Thomas's comments
> http://patches.dpdk.org/patch/68225/
Acked-by: Tho
15/05/2020 18:50, Muhammad Bilal:
> Removed the typing error in doc/guides/eventdevs/index.rst,
> drivers/net/mlx5/mlx5.c and in lib/librte_vhost/rte_vhost.h
>
> Cc: dev@dpdk.org
> Cc: sta...@dpdk.org
> Cc: jer...@marvell.com
> Cc: ma...@mellanox.com
> Cc: shah...@mellanox.com
> Cc: maxime.coque..
On Fri, May 15, 2020 at 11:50 AM Hemant Agrawal wrote:
> This patchset moves the internal symbol to the INTERNAL sections
>
> This patch also change some of the shared variables to function
> calls. (Mostly control path used variables).
>
> Hemant Agrawal (13):
> common/dpaax: move internal symb
On 5/13/2020 6:07 AM, Lu, Wenzhuo wrote:
>
>> -Original Message-
>> From: Zhang, Qi Z
>> Sent: Thursday, April 30, 2020 5:28 PM
>> To: Ye, Xiaolong
>> Cc: Lu, Wenzhuo ; Yang, Qiming
>> ; Yigit, Ferruh ;
>> dev@dpdk.org; Zhang, Qi Z
>> Subject: [PATCH] maintainers: update for ice
>>
>> R
On 5/19/2020 8:13 AM, Ye Xiaolong wrote:
> Hi, Maxime & Ferruh
>
> This patch is used to fix the regression caused by commit
> 715070ea10e6 ("vhost: prevent zero-copy with incompatible client mode") which
> was introduced in RC2, I think we should get it in this release.
+1, v4 of the patch appl
On 5/19/2020 11:15 AM, Xuan Ding wrote:
> This patch fixes the situation where vhost-user cannot start as server with
> dequeue_zero_copy enabled.
>
> Using flag instead of vsocket->is_server to determine whether vhost-user is
> in client mode. Because vsocket->is_server is not ready at this time.
> -Original Message-
> From: Yigit, Ferruh
> Sent: Tuesday, May 19, 2020 1:16 PM
> To: Dumitrescu, Cristian ; Ray Kinsella
> ; Neil Horman ; Eelco
> Chaudron
> Cc: dev@dpdk.org; Yigit, Ferruh ; Thomas Monjalon
> ; David Marchand ;
> sta...@dpdk.org; Luca Boccassi ; Richardson, Bruce
>
On Tue, May 19, 2020 at 2:57 PM Medvedkin, Vladimir
wrote:
>
> Hi,
>
> On 19/05/2020 13:23, David Marchand wrote:
> > On Tue, May 19, 2020 at 2:15 PM Vladimir Medvedkin
> > wrote:
> >> This patch series implements vectorized lookup using AVX512 for
> >> ipv4 dir24_8 and ipv6 trie algorithms.
> >>
Hi,
On 19/05/2020 13:23, David Marchand wrote:
On Tue, May 19, 2020 at 2:15 PM Vladimir Medvedkin
wrote:
This patch series implements vectorized lookup using AVX512 for
ipv4 dir24_8 and ipv6 trie algorithms.
Also introduced rte_fib_set_lookup_fn() to change lookup function type.
Added option t
Hi Xiaoping:
According to i40e datasheet, for packet with multiple buffer, the L2TAG
should only be valid on last Rx descriptor.
So it is identical with what you observed and it looks like a bug in
the vector path.
Could you try if below change can fix the issue?
--- a/d
12/05/2020 17:28, Ciara Power:
>
> Ciara Power (5):
> telemetry: keep telemetry threads separate from data plane
> telemetry: fix error checking for strchr function
> telemetry: fix closing socket fd on error
> telemetry: fix checking error return for socket creation
> telemetry: fix buf
hi, xiaoping
Base on the HW behaviors, either i40e normal recv or i40e vector recv
should use the offload flags of the last rxd to resign to the first
segment, then restore the mbuf to the recv packets, i am not sure if
this could answer your question, Anyway, you could use below quick check
On Tue, May 19, 2020 at 2:15 PM Vladimir Medvedkin
wrote:
>
> This patch series implements vectorized lookup using AVX512 for
> ipv4 dir24_8 and ipv6 trie algorithms.
> Also introduced rte_fib_set_lookup_fn() to change lookup function type.
> Added option to select lookup function type in testfib
> -Original Message-
> From: Trahe, Fiona
> Sent: Tuesday, 19 May, 2020 12:29
> To: Dybkowski, AdamX ; Anoob Joseph
> ; dev@dpdk.org; akhil.go...@nxp.com
> Cc: sta...@dpdk.org; Trahe, Fiona
> Subject: RE: [dpdk-dev] [PATCH] cryptodev: fix SHA-1 digest enum comment
>
> Hi Adam, Anoob,
>
On v20.02 some meter APIs have been matured and symbols moved from
EXPERIMENTAL to DPDK_20.0.1 block.
This can break the applications that were using these mentioned APIs on
v19.11. Although there is no modification on the APIs and the action is
positive and matures the APIs, the affect can be neg
Added -v option to switch between different lookup implementations
to measure their performance and correctness.
Signed-off-by: Vladimir Medvedkin
---
app/test-fib/main.c | 58 ++---
1 file changed, 55 insertions(+), 3 deletions(-)
diff --git a/ap
Add new lookup implementation for FIB6 trie algorithm using
AVX512 instruction set
Signed-off-by: Vladimir Medvedkin
---
lib/librte_fib/Makefile | 10 ++
lib/librte_fib/meson.build | 9 ++
lib/librte_fib/rte_fib6.h| 3 +-
lib/librte_fib/trie.c| 21
lib/librte_fib/tr
Move trie table layout and lookup defenition into the
private header file. This is necessary for implementing a
vectorized lookup function in a separate .с file.
Signed-off-by: Vladimir Medvedkin
---
lib/librte_fib/trie.c | 121 --
lib/librte_fib/t
Add new lookup implementation for DIR24_8 algorithm using
AVX512 instruction set
Signed-off-by: Vladimir Medvedkin
---
lib/librte_fib/Makefile | 14
lib/librte_fib/dir24_8.c| 24 ++
lib/librte_fib/dir24_8_avx512.c | 165
lib/lib
Add type argument to trie_get_lookup_fn()
Now it only supports RTE_FIB6_TRIE_SCALAR
Add new rte_fib6_set_lookup_fn() - user can change lookup
function type runtime.
Signed-off-by: Vladimir Medvedkin
---
lib/librte_fib/rte_fib6.c | 19 ++-
lib/librte_fib/rte_fib6.h
Move dir24_8 table layout and lookup defenition into the
private header file. This is necessary for implementing a
vectorized lookup function in a separate .с file.
Signed-off-by: Vladimir Medvedkin
---
lib/librte_fib/dir24_8.c | 225 +--
lib/librte_fi
New data type to manipulate 512 bit AVX values.
Signed-off-by: Vladimir Medvedkin
---
lib/librte_eal/x86/include/rte_vect.h | 20
1 file changed, 20 insertions(+)
diff --git a/lib/librte_eal/x86/include/rte_vect.h
b/lib/librte_eal/x86/include/rte_vect.h
index df5a607..ffe4
This patch series implements vectorized lookup using AVX512 for
ipv4 dir24_8 and ipv6 trie algorithms.
Also introduced rte_fib_set_lookup_fn() to change lookup function type.
Added option to select lookup function type in testfib application.
v3:
- separate out the AVX-512 code into a separate fi
Add type argument to dir24_8_get_lookup_fn()
Now it supports 3 different lookup implementations:
RTE_FIB_DIR24_8_SCALAR_MACRO
RTE_FIB_DIR24_8_SCALAR_INLINE
RTE_FIB_DIR24_8_SCALAR_UNI
Add new rte_fib_set_lookup_fn() - user can change lookup
function type runtime.
Signed-off-by: Vladimir Medvedk
On Tue, May 19, 2020 at 12:55 PM Hemant Agrawal wrote:
>
> metrics_initialized shall be reset in deinit function
> This is currently causing issue in running
> metrics_autotest mulutiple times
>
> Fixes: 07c1b6925b65 ("telemetry: invert dependency on metrics library")
>
> Signed-off-by: Hemant Agr
This patch fixes improper SHA-1 digest size in the enum comment
and also adds the note about HMAC-SHA-1-96.
v2:
* add info about HMAC-SHA-1-96
Adam Dybkowski (1):
cryptodev: fix SHA-1 digest enum comment
lib/librte_cryptodev/rte_crypto_sym.h | 5 +++--
1 file changed, 3 insertions(+), 2 delet
This patch fixes improper SHA-1 digest size in the enum comment
and also adds the note about HMAC-SHA-1-96.
Fixes: 1bd407fac80b ("cryptodev: extract symmetric operations")
Cc: sta...@dpdk.org
Signed-off-by: Adam Dybkowski
---
lib/librte_cryptodev/rte_crypto_sym.h | 5 +++--
1 file changed, 3 in
Hi,
I tried to use gdb to print the rx descriptor, and it turned out that only the
rxd for the last segment has correct l2tag1 value.
Test step: ping xx.xx.xx.xx -s 2500, from a vlan interface with vlan id 1901,
MTU 9000. (this will sends 1 packet, and needs two mbuf segments to receive,
len
On 19/05/2020 12:39, Hemant Agrawal wrote:
> Hi Ray,
>
>> On 15/05/2020 10:47, Hemant Agrawal wrote:
>>> This patch moves the internal symbols to INTERNAL sections so that any
>>> change in them is not reported as ABI breakage.
>>>
>>> Signed-off-by: Hemant Agrawal
>>> ---
>>> devtools/libabi
Hi Ray,
> On 15/05/2020 10:47, Hemant Agrawal wrote:
> > This patch moves the internal symbols to INTERNAL sections so that any
> > change in them is not reported as ABI breakage.
> >
> > Signed-off-by: Hemant Agrawal
> > ---
> > devtools/libabigail.abignore | 2 ++
> > drivers/net/
On 19/05/2020 12:16, Hemant Agrawal wrote:
>
>> On 15/05/2020 10:47, Hemant Agrawal wrote:
>>> This patch moves the internal symbols to INTERNAL sections so that any
>>> change in them is not reported as ABI breakage.
>>>
>>> Signed-off-by: Hemant Agrawal
>>> ---
>>> devtools/libabigail.abig
On 15/05/2020 10:47, Hemant Agrawal wrote:
> rte_dpaa2_memsegs is not being used by any other library
> or even within bus.
>
> Signed-off-by: Hemant Agrawal
> ---
> drivers/bus/fslmc/portal/dpaa2_hw_dpbp.c| 6 --
> drivers/bus/fslmc/portal/dpaa2_hw_pvt.h | 3 ---
> drivers/bus/fs
On 15/05/2020 10:47, Hemant Agrawal wrote:
> Currently rte_mcp_ptr_list is being shared as a variable
> across libs. This is only used in control path.
> This patch change it to a exported function based access.
>
> Signed-off-by: Hemant Agrawal
> ---
> drivers/bus/fslmc/fslmc_vfio.c
On 15/05/2020 10:47, Hemant Agrawal wrote:
> This patch changes the export of fman port config
> as function call instead of direct variable access.
> This is in control path, so it will not impact perf.
>
> Signed-off-by: Hemant Agrawal
> ---
> drivers/bus/dpaa/dpaa_bus.c | 6 +
On 15/05/2020 10:47, Hemant Agrawal wrote:
> qman caam and pool portal ids are only used in control
> path. This patch changes their inter library access to
> function call instead of direct shared variable.
>
> Signed-off-by: Hemant Agrawal
> ---
> drivers/bus/dpaa/base/qbman/qman_driver.c |
1 - 100 of 156 matches
Mail list logo
