In the rte_eth_rx_queue_setup API function, the local variable named
mbp_buf_size, which is the data room size of the input parameter mp,
is checked to guarantee that each memory chunck used for net device
in the mbuf is bigger than the min_rx_bufsize. But if mbp_buf_size is
less than RTE_PKTMBUF_H
Currently, there is a potential problem that calling the API function
rte_eth_dev_set_vlan_offload to start VLAN hardware offloads which the
driver does not support. If the PMD driver does not support certain VLAN
hardware offloads and does not check for it, the hardware setting will
not change, bu
This series are minor bugfixes for rte_ethdev.c.
Wei Hu (Xavier) (2):
ethdev: fix data room size verification in Rx queue setup
ethdev: fix VLAN offloads set if no relative capabilities
drivers/net/dpaa2/dpaa2_ethdev.c | 5 -
drivers/net/enic/enic_ethdev.c | 12 ---
Hi, Andrew Rybchenko
On 2020/6/19 16:37, Andrew Rybchenko wrote:
On 6/19/20 6:42 AM, Wei Hu (Xavier) wrote:
From: Chengchang Tang
Currently, there is a potential problem that calling the API function
rte_eth_dev_set_vlan_offload to start a vlan hardware offloads which the
driver does not sup
Reviewed-by: Wei Zhao
> -Original Message-
> From: Ivan Dyukov
> Sent: Monday, June 15, 2020 5:02 PM
> To: dev@dpdk.org; i.dyu...@samsung.com; v.kurams...@samsung.com;
> tho...@monjalon.net; david.march...@redhat.com; Yigit, Ferruh
> ; arybche...@solarflare.com; Zhao1, Wei
> ; Guo, Jia
Hi, Ferruh
> -Original Message-
> From: Yigit, Ferruh
> Sent: Thursday, June 18, 2020 7:12 PM
> To: Zhao1, Wei ; i.dyu...@samsung.com;
> dev@dpdk.org; v.kurams...@samsung.com; tho...@monjalon.net;
> david.march...@redhat.com; arybche...@solarflare.com; Guo, Jia
> ; Xing, Beilei ; Yang, Qi
Reviewed-by: Wei Zhao
> -Original Message-
> From: dev On Behalf Of Renata Saiakhova
> Sent: Wednesday, May 13, 2020 9:14 PM
> To: dev@dpdk.org
> Cc: Renata Saiakhova
> Subject: [dpdk-dev] [PATCH v3 0/4] Memory corruption due to HW rings
> allocation
>
> igb and ixgbe and some other d
> -Original Message-
> From: Yigit, Ferruh
> Sent: Saturday, June 20, 2020 12:56 AM
> To: Zhao1, Wei ; Renata Saiakhova
> ; dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v3 3/4] drivers/net: Fix in i40e HW rings
> memory overlap
>
> On 6/1/2020 8:58 AM, Zhao1, Wei wrote:
> > Hi, Renata
On Fri, 19 Jun 2020 18:22:43 +0200
David Marchand wrote:
> + rte_rwlock_read_lock(&lcore_lock);
I see you converted a spin lock to a reader lock.
Are you sure this is a good idea, although conceptually faster,
the implementation on most cpu's is slower than a simple spin lock.
https://www.k
On 5/15/2020 9:04 AM, Jeff Guo wrote:
> hi, renata
>
> On 5/13/2020 9:14 PM, Renata Saiakhova wrote:
>> Free previously allocated memzone for HW rings
>>
>> Signed-off-by: Renata Saiakhova
>> ---
>> lib/librte_ethdev/rte_ethdev.c | 28 ++--
>> lib/librte_ethdev/rt
On 6/1/2020 8:58 AM, Zhao1, Wei wrote:
> Hi, Renata Saiakhova
>
>I think this patch is very important, It seems all kind of NIC has memory
> leak problem that used for store
> Tx or rx descriptor. If that is true, memory point by rxq-> rx_ring/ txq->
> tx_ring will never be freed even if dev
On 6/3/2020 2:36 AM, Zhao1, Wei wrote:
> I am sure i40e FDIR filter also allocate ring for filter programing, it also
> has the same issue. Ice NIC the same.
> Although we can avoid this issue by change ring name to "igb_tx_ring" and so
> on, but the issue of memory for ring not freed after close
On 5/18/2020 10:48 AM, Renata Saiakhova wrote:
> Hi Ferruh,
>
> thanks for comments,
>
> are the rte_eth_dma_zone_reserve() calls always used to allocate HW rings? It
> is
> not totally clear to me. That was partly the reason I don't do the fix for
> every
> driver which uses this API. I made f
DPDK components and applications can have their say when a new lcore is
initialized. For this, they can register a callback for initializing and
releasing their private data.
Signed-off-by: David Marchand
---
Changes since v1:
- added unit test (since missing some coverage, for v3),
- preferred c
Convert to new lcore API to support non-EAL lcores.
Signed-off-by: David Marchand
---
drivers/mempool/bucket/rte_mempool_bucket.c | 131
1 file changed, 82 insertions(+), 49 deletions(-)
diff --git a/drivers/mempool/bucket/rte_mempool_bucket.c
b/drivers/mempool/bucket/rte_
DPDK allows calling some part of its API from a non-EAL thread but this
has some limitations.
OVS (and other applications) has its own thread management but still
want to avoid such limitations by hacking RTE_PER_LCORE(_lcore_id) and
faking EAL threads potentially unknown of some DPDK component.
I
Add a helper to iterate all lcores.
The iterator callback is read-only wrt the lcores list.
Implement a dump function on top of this for debugging.
Signed-off-by: David Marchand
---
Changes since v1:
- introduced lcore iterators and implemented rte_lcore_dump,
this iterator mechanism can then
This is a preparation step for dynamically unregistering threads.
Since we explicitly allocate a per thread trace buffer in
rte_thread_init, add an internal helper to free this buffer.
Signed-off-by: David Marchand
---
Note: I preferred renaming the current internal function to free all
threads
Introduce a helper responsible for initialising the per thread context.
We can then have a unified context for EAL and non-EAL threads and
remove copy/paste'd OS-specific helpers.
Per EAL thread CPU affinity setting is separated from the thread init.
It is to accommodate with Windows EAL where CPU
Because of the inline accessor + static declaration in rte_gettid(),
we end up with multiple symbols for RTE_PER_LCORE(_thread_id).
Each compilation unit will pay a cost when accessing this information
for the first time.
$ nm build/app/dpdk-testpmd | grep per_lcore__thread_id
0054 d p
For consistency sake, move all lcore role code in the dedicated
compilation unit / header.
Signed-off-by: David Marchand
---
lib/librte_eal/common/eal_common_lcore.c | 11 +++
lib/librte_eal/common/eal_common_thread.c | 11 ---
lib/librte_eal/include/rte_eal.h | 9 --
lib/
We have per lcore thread symbols scattered in OS implementations but
common code relies on them.
Move all of them in common.
RTE_PER_LCORE(_socket_id) and RTE_PER_LCORE(_cpuset) have public
accessors and are not exported through the library map, they can be
made static.
Signed-off-by: David March
OVS and some other applications have been hacking into DPDK internals to
fake EAL threads and avoid performance penalty of only having non-EAL
threads.
This series proposes to add a new type of lcores and maps those threads
to such lcores.
non-EAL threads won't run the DPDK eal mainloop.
As a cons
> On Jun 19, 2020, at 10:43 AM, Thomas Monjalon wrote:
>
> 19/06/2020 17:38, Dharmik Thakkar:
>> + Bruce
>>
>>> On Jun 17, 2020, at 2:48 PM, Thomas Monjalon wrote:
>>>
>>> 17/06/2020 20:21, Ferruh Yigit:
On 5/20/2020 4:20 AM, Dharmik Thakkar wrote:
> Update testpmd documentation to
19/06/2020 17:38, Dharmik Thakkar:
> + Bruce
>
> > On Jun 17, 2020, at 2:48 PM, Thomas Monjalon wrote:
> >
> > 17/06/2020 20:21, Ferruh Yigit:
> >> On 5/20/2020 4:20 AM, Dharmik Thakkar wrote:
> >>> Update testpmd documentation to include RECORD configuration options,
> >>> CONFIG_RTE_TEST_PMD_R
+ Bruce
> On Jun 17, 2020, at 2:48 PM, Thomas Monjalon wrote:
>
> 17/06/2020 20:21, Ferruh Yigit:
>> On 5/20/2020 4:20 AM, Dharmik Thakkar wrote:
>>> Update testpmd documentation to include RECORD configuration options,
>>> CONFIG_RTE_TEST_PMD_RECORD_CORE_CYCLES and
>>> CONFIG_RTE_TEST_PMD_RECOR
On 6/19/20 1:14 PM, Adrian Moreno wrote:
>
>
> On 6/11/20 11:37 PM, Maxime Coquelin wrote:
>> This patch makes the vDPA framework to no more
>> support only PCI devices, but any devices by relying
>> on the generic device name as identifier.
>>
>> Signed-off-by: Maxime Coquelin
>> ---
>> dri
On 6/19/20 5:02 PM, Ferruh Yigit wrote:
> On 6/18/2020 7:59 PM, Matan Azrad wrote:
>> A new vDPA driver feature was added to query the virtq statistics from
>> the HW.
>>
>> Use this feature to show the HW queues statistics for the virtqs.
>>
>> Command description: stats X Y.
>> X is the device
On 6/18/2020 7:59 PM, Matan Azrad wrote:
> A new vDPA driver feature was added to query the virtq statistics from
> the HW.
>
> Use this feature to show the HW queues statistics for the virtqs.
>
> Command description: stats X Y.
> X is the device ID.
> Y is the queue ID, Y=0x to show all the
On 6/11/2020 9:43 AM, Qi Zhang wrote:
> This patch add support to get tunnel type of recipe
> after get recipe from fw. This will fix the issue in
> function ice_find_recp() for tunnel type comparing.
>
> Signed-off-by: Wei Zhao
> Signed-off-by: Paul M Stillwell Jr
> Signed-off-by: Qi Zhang
<.
On 6/19/20 3:28 PM, Matan Azrad wrote:
>
>
> From: Maxime Coquelin:
>> On 6/18/20 6:28 PM, Matan Azrad wrote:
>>> As an arrangement to per queue operations in the vDPA device it is
>>> needed to change the next experimental API:
>>>
>>> The API ``rte_vhost_host_notifier_ctrl`` was changed to b
An issue has been observed where epoll file descriptor
list rebuilds every time an interrupt/alarm event is
received.
eal_intr_process_interrupts() should notify pipe fd only
if any source is removed from the source list i.e (rv > 0)
Fixes: 0c7ce182a760 ("eal: add pending interrupt callback unreg
Hi Matan,
On 6/19/20 3:11 PM, Matan Azrad wrote:
> Hi Maxime
>
> Thanks for the fast review.
> This is first version, let's review it carefully to be sure it is correct.
> @Xiao Wang, it will be good to hear your idea too.
> We also need to understand the effect on IFC driver/device...
> Just to
From: Maxime Coquelin:
> On 6/18/20 6:28 PM, Matan Azrad wrote:
> > As an arrangement to per queue operations in the vDPA device it is
> > needed to change the next experimental API:
> >
> > The API ``rte_vhost_host_notifier_ctrl`` was changed to be per queue
> > instead of per device.
> >
> > A
Hi Maxime
Thanks for the fast review.
This is first version, let's review it carefully to be sure it is correct.
@Xiao Wang, it will be good to hear your idea too.
We also need to understand the effect on IFC driver/device...
Just to update that I checked this code with the mlx5 adjustments and I
Hi Luca,
All testcases pass for DPDK LTS 19.11.3. Failed testcases below were double
checked and No issues found.
Test Report:
DPDK
https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgit.dpdk.org%2Fdpdk-stable%2Fsnapshot%2Fdpdk-stable-19.11.3-rc1.tar.gz&data=02%7C01%7CAbhishek.M
> -Original Message-
> From: David Marchand
> Sent: Friday, June 19, 2020 5:58 PM
> To: dev@dpdk.org
> Cc: Gavin Hu ; Phil Yang ; Joyce
> Kong
> Subject: [PATCH] test: fix bitops test name
>
> Caught by code review, bitops test name is incorrect.
>
> Fixes: 7660614c11e2 ("test/bitops: a
Add rte_write32_wc and rte_write32_wc_relaxed functions
that implement 32bit stores using write combining memory protocol.
Provided generic stubs and x86 implementation.
Signed-off-by: Radu Nicolau
---
v2 rework new eal io functions
lib/librte_eal/include/generic/rte_io.h | 47 +
Performance improvement: use a write combining store
instead of a regular mmio write to update queue tail
registers.
Signed-off-by: Radu Nicolau
---
drivers/net/i40e/base/i40e_osdep.h| 6 ++
drivers/net/i40e/i40e_rxtx.c | 8
drivers/net/i40e/i40e_rxtx_vec_avx2.c | 4 ++-
Hi Matan,
On 6/19/20 9:41 AM, Maxime Coquelin wrote:
>
>
> On 6/18/20 6:28 PM, Matan Azrad wrote:
>> Some guest drivers may not configure disabled virtio queues.
>>
>> In this case, the vhost management never triggers the vDPA device
>> configuration because it waits to the device to be ready.
>
On 6/11/20 11:37 PM, Maxime Coquelin wrote:
> This patch makes the vDPA framework to no more
> support only PCI devices, but any devices by relying
> on the generic device name as identifier.
>
> Signed-off-by: Maxime Coquelin
> ---
> drivers/vdpa/ifc/ifcvf_vdpa.c | 6 +-
> drivers/
Make the coding style more consistent, and the init logic control flow
more explicit.
Signed-off-by: Anatoly Burakov
---
examples/l3fwd-power/main.c | 18 +++---
1 file changed, 11 insertions(+), 7 deletions(-)
diff --git a/examples/l3fwd-power/main.c b/examples/l3fwd-power/main.c
i
Currently, legacy mode is the implicit default, but it is not possible
to directly request using legacy mode. Add the argument to enable
requesting legacy mode, and also make it the default.
Signed-off-by: Anatoly Burakov
---
examples/l3fwd-power/main.c | 28 +++-
1 file
Currently, there is no way to know if the power management env is
supported without trying to initialize it. The init API also does
not distinguish between failure due to some error and failure due to
power management not being available on the platform in the first
place.
Thus, add an API that pr
Since 20.05, l3fwd-power has become much more stringent about
whether it allows initialization without initializing the
librte_power library with it. This means that while previously
the app could have been used to test RX interrupts functionality
even if the app itself was in a half-working state,
In addition to existing modes, add a mode which is very similar to
legacy mode, but does not do frequency scaling, and thus does not
depend on the power library.
Signed-off-by: Anatoly Burakov
Tested-by: Yinan Wang
---
examples/l3fwd-power/main.c | 188 +++-
1 fi
Currently, the application does support running without the power
library being initialized, but it has to be specifically requested. On
platforms without support for frequency scaling using the power library,
we can just enable interrupt-only mode by default.
Signed-off-by: Anatoly Burakov
Sugge
Currently, l3fwd-power will attempt to run even if the power env
is set to KVM, which is not supported. Fix this by preventing the
app from initializing unless the env is set to one of the supported
modes.
Signed-off-by: Anatoly Burakov
Tested-by: Yinan Wang
---
examples/l3fwd-power/main.c | 9
Currently, interrupts are enabled in telemetry and empty poll modes, but
they are not used. Switch to disabling interrupts by default, and only
enable interrupts for modes that require them.
Signed-off-by: Anatoly Burakov
Tested-by: Yinan Wang
---
examples/l3fwd-power/main.c | 12 +++-
Waiting for reviews please.
On 19/05/2020 13:12, 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
Improve cache hit and increase packet rate on benchmarks.
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/sfc_tweak.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/sfc/sfc_tweak.h b/drivers/net/sfc/sfc_tweak.h
index bd3de5313a..426e48f593 100644
--- a/driver
Caught by code review, bitops test name is incorrect.
Fixes: 7660614c11e2 ("test/bitops: add bit operations test case")
Signed-off-by: David Marchand
---
app/test/autotest_data.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/app/test/autotest_data.py b/app/test/autotest_d
On 19-Jun-20 8:37 AM, Harman Kalra wrote:
On Thu, Jun 18, 2020 at 06:18:29PM +0100, Anatoly Burakov wrote:
External Email
--
Currently, the application does support running without the power
library being initialized, but it ha
From: Ivan Malov
Add named constants for deprecated QinQ TPIDs.
Update drivers which have already been using existing
TPID named constants from librte_net to use the
new named constants rather than magic numbers.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
---
drivers/net/bnxt/b
From: Qi Zhang
Add support to add and delete MAC address filter in DCF.
Signed-off-by: Qi Zhang
Signed-off-by: Ting Xu
---
drivers/net/ice/ice_dcf.c| 42
drivers/net/ice/ice_dcf.h| 1 +
drivers/net/ice/ice_dcf_ethdev.c | 7 ++
3 files ch
Add doc for DCF datapath configuration in DPDK 20.08 release note.
Signed-off-by: Ting Xu
---
doc/guides/rel_notes/release_20_08.rst | 6 ++
1 file changed, 6 insertions(+)
diff --git a/doc/guides/rel_notes/release_20_08.rst
b/doc/guides/rel_notes/release_20_08.rst
index dee4ccbb5..1a3a4cd
From: Qi Zhang
Enable Rx queues initialization during device start in DCF.
Signed-off-by: Qi Zhang
Signed-off-by: Ting Xu
---
drivers/net/ice/ice_dcf.h| 1 +
drivers/net/ice/ice_dcf_ethdev.c | 83
2 files changed, 84 insertions(+)
diff --git a/driver
From: Qi Zhang
Add support to get and reset Rx/Tx stats in DCF. Query stats
from PF.
Signed-off-by: Qi Zhang
Signed-off-by: Ting Xu
---
drivers/net/ice/ice_dcf.c| 27
drivers/net/ice/ice_dcf.h| 4 ++
drivers/net/ice/ice_dcf_ethdev.c | 102 ++
From: Qi Zhang
Add queue start and stop in DCF. Support queue enable and disable
through virtual channel. Add support for Rx queue mbufs allocation
and queue reset.
Signed-off-by: Qi Zhang
Signed-off-by: Ting Xu
---
drivers/net/ice/ice_dcf.c| 57 ++
drivers/net/ice/ice_dcf.h
From: Qi Zhang
Add stop flag for DCF device start and stop.
Signed-off-by: Qi Zhang
Signed-off-by: Ting Xu
---
drivers/net/ice/ice_dcf_ethdev.c | 14 ++
drivers/net/ice/ice_dcf_parent.c | 1 +
2 files changed, 15 insertions(+)
diff --git a/drivers/net/ice/ice_dcf_ethdev.c b/driv
From: Qi Zhang
Enable RSS initialization during DCF start. Add RSS LUT and
RSS key configuration functions.
Signed-off-by: Qi Zhang
Signed-off-by: Ting Xu
---
drivers/net/ice/ice_dcf.c| 117 +++
drivers/net/ice/ice_dcf.h| 1 +
drivers/net/ice/ice_
From: Qi Zhang
Add queues and Rx queue irqs configuration during device start
in DCF. The setup is sent to PF via virtchnl.
Signed-off-by: Qi Zhang
Signed-off-by: Ting Xu
---
drivers/net/ice/ice_dcf.c| 111 +++
drivers/net/ice/ice_dcf.h| 6 ++
drivers
From: Qi Zhang
Delete original DCF queue setup functions and use ice
queue setup and release functions instead.
Signed-off-by: Qi Zhang
Signed-off-by: Ting Xu
---
drivers/net/ice/ice_dcf_ethdev.c | 42 +++-
drivers/net/ice/ice_dcf_ethdev.h | 3 ---
drivers/net/ice
From: Qi Zhang
Add support to get complete device information for DCF, including
Rx/Tx offload capabilities and default configuration.
Signed-off-by: Qi Zhang
Signed-off-by: Ting Xu
---
drivers/net/ice/ice_dcf_ethdev.c | 70 ++--
1 file changed, 67 insertions(+), 3
From: Qi Zhang
Enable device configuration function in DCF.
Signed-off-by: Qi Zhang
Signed-off-by: Ting Xu
---
drivers/net/ice/ice_dcf_ethdev.c | 11 ++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ice/ice_dcf_ethdev.c b/drivers/net/ice/ice_dcf_ethdev.c
i
From: Qi Zhang
Enable RSS parameters initialization and get the supported
flexible descriptor RXDIDs bitmap from PF during DCF init.
Signed-off-by: Qi Zhang
Signed-off-by: Ting Xu
---
drivers/net/ice/ice_dcf.c | 54 ++-
drivers/net/ice/ice_dcf.h | 3 +++
2
From: Xu Ting
This patchset adds support to configure DCF datapath, including
Rx/Tx queues setup, start and stop, device configuration, RSS
and flexible descriptor RXDID initialization and MAC filter setup.
Qi Zhang (11):
net/ice: init RSS and supported RXDID in DCF
net/ice: complete device
On 6/19/20 6:42 AM, Wei Hu (Xavier) wrote:
> From: Chengchang Tang
>
> Currently, there is a potential problem that calling the API function
> rte_eth_dev_set_vlan_offload to start a vlan hardware offloads which the
> driver does not support. if the PMD driver does not support the relative
> hard
Hi all,
Here is a new stable release:
https://fast.dpdk.org/rel/dpdk-19.11.3.tar.xz
The git tree is at:
https://dpdk.org/browse/dpdk-stable/?h=19.11
Luca Boccassi
---
.travis.yml| 2 +-
VERSION
On 6/19/20 6:42 AM, Wei Hu (Xavier) wrote:
> From: Chengchang Tang
>
> In the rte_eth_rx_queue_setup API function, the local variable named
> mbp_buf_size, which is the data room size of the input parameter mp,
> is checked to guarantee that each memory chunck used for net device
> in the mbuf is
This patch moved the RSS initialization from dev start to
dev configure, to fix the issue that RSS redirection table
can not be kept after restarting port.
Fixes: 69dd4c3d0898 ("net/avf: enable queue and device")
Cc: sta...@dpdk.org
Signed-off-by: Junyu Jiang
---
drivers/net/iavf/iavf_ethdev.c
On 6/18/20 6:28 PM, Matan Azrad wrote:
> Some guest drivers may not configure disabled virtio queues.
>
> In this case, the vhost management never triggers the vDPA device
> configuration because it waits to the device to be ready.
This is not vDPA-only, even with SW datapath the application's
On Thu, Jun 18, 2020 at 06:18:29PM +0100, Anatoly Burakov wrote:
> External Email
>
> --
> Currently, the application does support running without the power
> library being initialized, but it has to be specifically requested. On
The creation of DBR can be used by a number of different
Mellanox PMDs. for example RegEx / Net / VDPA.
This commits moves the DBR creation and relase functions to common
folder.
Signed-off-by: Ori Kam
Acked-by: Viacheslav Ovsiienko
---
v2:
* No changes.
---
drivers/common/mlx5/linux/mlx5_comm
From: Ophir Munk
Getter functions such as: 'mlx5_os_get_ctx_device_name',
'mlx5_os_get_ctx_device_path', 'mlx5_os_get_dev_device_name',
'mlx5_os_get_umem_id' are implemented under net directory. To enable
additional devices (e.g. regex, vdpa) to access these getter functions
they are moved under
75 matches
Mail list logo
