On Thu, Feb 25, 2021 at 8:49 PM wrote:
>
> From: Luca Boccassi
>
> It consistently fails when ran on a build machine within the
> reproducible build environment, so move it out of there, so
> that we can run the fast suite at build time when packaging.
This test is quite simple.
Surprising that
Stephen,
No, I don't have a better proposal, but I think it is not correct to change
the behavior of KNI (making link down without a real response).
Even though we know that communicating with userspace under rtnl_lock is a
bad idea, it works as it is for many years already.
Elad,
I agree with y
Li Zhang (1):
ethdev: add meter PPS profile
lib/librte_ethdev/rte_mtr.h | 32 ++
--
2.27.0
Currently meter algorithms only supports rate is bytes per second(BPS).
Add this new meter srTCMp algorithm to support rate is packet per second.
So that it can meter traffic by packet per second.
The below structure will be extended:
rte_mtr_algorithm
rte_mtr_meter_profile
Signed-off-by: Li Zhang
On Fri, Feb 26, 2021 at 3:21 AM Ferruh Yigit wrote:
>
> On 2/26/2021 6:43 AM, Andrew Rybchenko wrote:
> > On 2/25/21 9:25 PM, Ferruh Yigit wrote:
> >> On 2/22/2021 7:18 PM, Ajit Khaparde wrote:
> >>> Add support for forced ethernet speed setting.
> >>> Currently testpmd tries to configure the Ethe
Hi, Jiawei
Thank you for the clarification. I missed the point that we have updated elts
array
with new allocated mbufs and are not able to retry packet building anymore.
Very good catch, thank you! Could you, please, add this extra explanation
to the commit message and send the v2 ?
With bes
Thanks Dmitry - will address and send v7.
On 28/02/2021 14:38, Dmitry Kozlyuk wrote:
2021-02-23 18:18, Nick Connolly:
Attaching to an NVMe disk on Windows using SPDK requires the
PCI class ID and device.bus fields. Decode the class ID from the PCI
device info strings if it is present and set de
> -Original Message-
> From: dev On Behalf Of Matan Azrad
> Sent: Monday, March 1, 2021 8:56 AM
> To: Anoob Joseph ; dev@dpdk.org
> Cc: Doherty, Declan ; Somalapuram Amaranath
> ; Ruifeng Wang ; Ajit
> Khaparde ; Zhang, Roy Fan
> ; Griffin, John ; De Lara
> Guarch, Pablo ; Michael Shami
Currently meter algorithms only supports rate is bytes per second(BPS).
Add this new meter srTCMp algorithm to support rate is packet per second.
So that it can meter traffic by packet per second.
The below structure will be extended:
rte_mtr_algorithm
rte_mtr_meter_profile
Signed-off-by: Li Zhang
v2:
* Rebase
Li Zhang (4):
ethdev: add meter PPS profile
common/mlx5: add meter mode definition in PRM file
net/mlx5: support meter PPS profile
app/testpmd: add meter pps mode cmd
app/test-pmd/cmdline.c| 4 +
app/test-pmd/cmdline_mtr.c| 105
Currently meter algorithms only supports bytes per second(BPS).
Such as Single Rate Three Color Marker (srTCM rfc2697)
Add this new meter srTCMp algorithm to support rate is packet per second.
So that it can meter traffic by packet per second (PPS).
Signed-off-by: Li Zhang
---
doc/guides/nics/ml
Add meter mode definition in PRM file
Signed-off-by: Li Zhang
---
drivers/common/mlx5/mlx5_prm.h | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/common/mlx5/mlx5_prm.h b/drivers/common/mlx5/mlx5_prm.h
index e3d4120849..609b1c3951 100644
--- a/drivers/common/mlx5/mlx5_prm.h
++
Support meter pps mode
Signed-off-by: Li Zhang
---
app/test-pmd/cmdline.c | 4 +
app/test-pmd/cmdline_mtr.c | 105
app/test-pmd/cmdline_mtr.h | 1 +
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 15 +++
4 files chan
On Sun, Feb 28, 2021 at 01:53:42PM +0100, Thomas Monjalon wrote:
> The page size is often retrieved from the macro PAGE_SIZE.
> If PAGE_SIZE is not defined, it is either using hard coded default,
> or getting the system value from the UNIX-only function sysconf().
>
> Such definitions are replaced
v2:
* Rebase
Li Zhang (4):
ethdev: add meter PPS profile
common/mlx5: add meter mode definition in PRM file
net/mlx5: support meter PPS profile
app/testpmd: add meter pps mode cmd
app/test-pmd/cmdline.c| 4 +
app/test-pmd/cmdline_mtr.c| 105
Currently meter algorithms only supports rate is bytes per second(BPS).
Add this new meter srTCMp algorithm to support rate is packet per second.
So that it can meter traffic by packet per second.
The below structure will be extended:
rte_mtr_algorithm
rte_mtr_meter_profile
Signed-off-by: Li Zhang
Add meter mode definition in PRM file
Signed-off-by: Li Zhang
---
drivers/common/mlx5/mlx5_prm.h | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/common/mlx5/mlx5_prm.h b/drivers/common/mlx5/mlx5_prm.h
index e3d4120849..609b1c3951 100644
--- a/drivers/common/mlx5/mlx5_prm.h
++
Currently meter algorithms only supports bytes per second(BPS).
Such as Single Rate Three Color Marker (srTCM rfc2697)
Add this new meter srTCMp algorithm to support rate is packet per second.
So that it can meter traffic by packet per second (PPS).
Signed-off-by: Li Zhang
---
doc/guides/nics/ml
Support meter pps mode
Signed-off-by: Li Zhang
---
app/test-pmd/cmdline.c | 4 +
app/test-pmd/cmdline_mtr.c | 105
app/test-pmd/cmdline_mtr.h | 1 +
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 15 +++
4 files chan
01/03/2021 10:41, Bruce Richardson:
> On Sun, Feb 28, 2021 at 01:53:42PM +0100, Thomas Monjalon wrote:
> > The page size is often retrieved from the macro PAGE_SIZE.
> > If PAGE_SIZE is not defined, it is either using hard coded default,
> > or getting the system value from the UNIX-only function s
Support meter pps mode
Signed-off-by: Li Zhang
---
app/test-pmd/cmdline.c | 4 +
app/test-pmd/cmdline_mtr.c | 105
app/test-pmd/cmdline_mtr.h | 1 +
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 15 +++
4 files chan
Currently meter algorithms only supports rate is bytes per second(BPS).
Add this new meter srTCMp algorithm to support rate is packet per second.
So that it can meter traffic by packet per second.
The below structure will be extended:
rte_mtr_algorithm
rte_mtr_meter_profile
Signed-off-by: Li Zhang
Currently meter algorithms only supports bytes per second(BPS).
Such as Single Rate Three Color Marker (srTCM rfc2697)
Add this new meter srTCMp algorithm to support rate is packet per second.
So that it can meter traffic by packet per second (PPS).
Signed-off-by: Li Zhang
---
doc/guides/nics/ml
v3:
* Redefine the struct name for PPS mode.
v2:
* Rebase
Li Zhang (4):
ethdev: add meter PPS profile
common/mlx5: add meter mode definition in PRM file
net/mlx5: support meter PPS profile
app/testpmd: add meter pps mode cmd
app/test-pmd/cmdline.c| 4 +
app/te
Add meter mode definition in PRM file
Signed-off-by: Li Zhang
---
drivers/common/mlx5/mlx5_prm.h | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/common/mlx5/mlx5_prm.h b/drivers/common/mlx5/mlx5_prm.h
index e3d4120849..609b1c3951 100644
--- a/drivers/common/mlx5/mlx5_prm.h
++
Attaching to an NVMe disk on Windows using SPDK requires the
PCI class ID and device.bus fields. Decode the class ID from the PCI
device info strings if it is present and set device.bus.
Signed-off-by: Nick Connolly
Acked-by: Tal Shnaiderman
Acked-by: Dmitry Kozlyuk
---
v7:
* Improve comments a
Fixes: c08bd191b13d ("eal/windows: initialize hugepage info")
Cc: sta...@dpdk.org
Reported-by: Stephen Hemminger
Signed-off-by: Dmitry Kozlyuk
Acked-by: Nick Connolly
On Mon, Mar 01, 2021 at 01:48:49AM +, bugzi...@dpdk.org wrote:
> https://bugs.dpdk.org/show_bug.cgi?id=644
>
> Bug ID: 644
>Summary: [dpdk-19.11.7]usertools/dpdk-devbind.py: bind NIC port
> to DPDK failed in vm.
>Product: DPDK
>
On Mon, Mar 01, 2021 at 10:53:46AM +0100, Thomas Monjalon wrote:
> 01/03/2021 10:41, Bruce Richardson:
> > On Sun, Feb 28, 2021 at 01:53:42PM +0100, Thomas Monjalon wrote:
> > > The page size is often retrieved from the macro PAGE_SIZE.
> > > If PAGE_SIZE is not defined, it is either using hard cod
Currently the flow Meter algorithms in rte_flow
only supports bytes per second(BPS).
Such as Single Rate Three Color Marker (srTCM rfc2697)
This RFC adds the packet per second definition in
Meter algorithms structure,
to support the rte_mtr APIs with type srTCM pps mode.
The below structure will be
Currently meter algorithms only supports rate is bytes per second(BPS).
Add this new meter srTCMp algorithm to support rate is packet per second.
So that it can meter traffic by packet per second.
The below structure will be extended:
rte_mtr_algorithm
rte_mtr_meter_profile
Signed-off-by: Li Zhang
Add meter mode definition in PRM file
Signed-off-by: Li Zhang
---
drivers/common/mlx5/mlx5_prm.h | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/common/mlx5/mlx5_prm.h b/drivers/common/mlx5/mlx5_prm.h
index e3d4120849..609b1c3951 100644
--- a/drivers/common/mlx5/mlx5_prm.h
++
Currently meter algorithms only supports bytes per second(BPS).
Such as Single Rate Three Color Marker (srTCM rfc2697)
Add this new meter srTCMp algorithm to support rate is packet per second.
So that it can meter traffic by packet per second (PPS).
Signed-off-by: Li Zhang
---
doc/guides/nics/ml
Support meter pps mode
Signed-off-by: Li Zhang
---
app/test-pmd/cmdline.c | 4 +
app/test-pmd/cmdline_mtr.c | 105
app/test-pmd/cmdline_mtr.h | 1 +
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 15 +++
4 files chan
The HW virtq object can be destroyed ether when the device is closed or
when the state of the virtq becomes disabled.
Some parameters of the virtq should continue to be managed when the
virtq state is changed but all of them must be initialized when the
device is closed.
Wrongly, the enable param
On 28-Feb-21 1:21 PM, Roy Shterman wrote:
On Mon, Feb 22, 2021 at 5:53 PM Burakov, Anatoly
mailto:anatoly.bura...@intel.com>> wrote:
On 22-Feb-21 10:41 AM, Roy Shterman wrote:
> When using huge_unlink we unlink the segment right
> after allocation. Although we unlink the file w
> -Original Message-
> From: Thomas Monjalon
> Sent: Wednesday, February 24, 2021 1:10 PM
> To: Juraj Linkeš
> Cc: oulijun ; ferruh.yi...@intel.com; dev@dpdk.org;
> linux...@openeuler.org
> Subject: Re: [dpdk-dev] [PATCH 1/2] config/arm: fix Hisilicon kunpeng920 SoC
> build
>
> 24/02/2
Prior to this commit, if rte_pktmbuf_alloc_bullk failed during rx queue
setup the error was not returned to the user and they may incorrectly
assume that the rx queue had been successfully set up. This commit ensures
that the error is returned to the user.
Bugzilla ID: 643
Fixes: d8a210774e1d ("ne
https://bugs.dpdk.org/show_bug.cgi?id=645
Bug ID: 645
Summary: mlx5 does not reflect lpbk_mode setting
Product: DPDK
Version: unspecified
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
The script dependencies list was incomplete,
this patch adds missing modules and removes an unnecessary entry.
The installation command was also added.
Fixes: f400e0b82bf1 ("app/crypto-perf: add script to graph perf results")
Signed-off-by: Ciara Power
---
doc/guides/tools/cryptoperf.rst | 8 ++
Please do not top post, message moved down.
On 2/26/2021 5:47 PM, 张 杨 wrote:
I think your idea is fine
What do you think just record file path in "pmd_pcap_probe()",
Perform an open operation only in "eth_dev_start()"?
When the secondary process add pcap vdev,
it send the request to primary
On Fri, Jan 15, 2021 at 8:33 AM Nithin Dabilpuram
wrote:
>
> Partial DMA unmap is not supported by VFIO type1 IOMMU
> in Linux. Though the return value is zero, the returned
> DMA unmap size is not same as expected size.
> So add test case and fix to both heap triggered DMA
> mapping and user trig
On 2/26/2021 4:18 PM, Andrew Boyer wrote:
On Feb 26, 2021, at 6:21 AM, Ferruh Yigit wrote:
On 2/26/2021 6:43 AM, Andrew Rybchenko wrote:
On 2/25/21 9:25 PM, Ferruh Yigit wrote:
On 2/22/2021 7:18 PM, Ajit Khaparde wrote:
Add support for forced ethernet speed setting.
Currently testpmd trie
This patch adds closing of the PMD after running the tests.
Signed-off-by: Adam Dybkowski
---
app/test/test_cryptodev.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index f91debc16..7dcd255de 100644
--- a/app/test/tes
This patch adds closing of the PMD after running the benchmark.
Signed-off-by: Adam Dybkowski
---
app/test-crypto-perf/main.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/app/test-crypto-perf/main.c b/app/test-crypto-perf/main.c
index 49af812d8..c1d338fa4 100644
--
Hello Asaf,
it is a while we were in touch regarding this topic. Finally, I am
again trying to get work this feature. I've seen that sampling is
already upstreamed which is great. However, I am not very successful
with that. There is nearly no documentation, just [1], I found no
examples, just com
> -Original Message-
> From: Ferruh Yigit
> Sent: Thursday, January 28, 2021 6:28 PM
> To: Li Zhang ; dek...@nvidia.com; or...@nvidia.com;
> viachesl...@nvidia.com; ma...@nvidia.com; Dumitrescu, Cristian
>
> Cc: dev@dpdk.org; tho...@monjalon.net; rasl...@nvidia.com
> Subject: Re: [dpdk-
> -Original Message-
> From: dev On Behalf Of Li Zhang
> Sent: Monday, March 1, 2021 10:35 AM
> To: dek...@nvidia.com; or...@nvidia.com; viachesl...@nvidia.com;
> ma...@nvidia.com
> Cc: dev@dpdk.org; tho...@monjalon.net; rasl...@nvidia.com;
> m...@smartsharesystems.com; ajit.khapa...@broad
On Wed, Sep 30, 2020 at 12:02:40PM +0100, Ferruh Yigit wrote:
> Also add missing .ini feature list documentation and document some
> existing features including new "Multiprocess aware" feature.
>
> Signed-off-by: Ferruh Yigit
> ---
> Cc: Kevin Laatz
> ---
> doc/guides/nics/features/ring.ini |
On 2/25/2021 11:46 AM, Heinrich Kuhn wrote:
From: "Chaoyong.He"
1. Fixup the suffix of the PCI ID to be consistent with the code.
2. Add specification of using MAC address to identify port.
Fixes: 979f2bae0 ("doc: improve multiport PF in nfp guide")
Cc: sta...@dpdk.org
Signed-off-by: Chaoyong
在 2021/2/26 23:25, Ferruh Yigit 写道:
On 2/24/2021 1:28 AM, Lijun Ou wrote:
From: Huisong Li
The normal operation of electrical interface devices depends on the
initialization and configuration of the PHY chip. The task of driving
the PHY chip is implemented in some firmware versions. If firm
> On Feb 28, 2021, at 7:53 AM, Thomas Monjalon wrote:
>
> The page size is often retrieved from the macro PAGE_SIZE.
> If PAGE_SIZE is not defined, it is either using hard coded default,
> or getting the system value from the UNIX-only function sysconf().
>
> Such definitions are replaced with
Hi, Jan
To use port action (I see it is in your sample action list) the flow should be
applied to the FDB domain,
ie "transfer" attribute should be specified:
flow validate 0 ingress transfer...
With best regards, Slava
> -Original Message-
> From: Jan Viktorin
> Sent: Monday, March 1
On Mon, 1 Mar 2021 14:34:07 +
Slava Ovsiienko wrote:
> Hi, Jan
>
> To use port action (I see it is in your sample action list) the flow should
> be applied to the FDB domain,
> ie "transfer" attribute should be specified:
>
> flow validate 0 ingress transfer...
As you can see (however, it
On 3/1/2021 2:17 PM, oulijun wrote:
在 2021/2/26 23:25, Ferruh Yigit 写道:
On 2/24/2021 1:28 AM, Lijun Ou wrote:
From: Huisong Li
The normal operation of electrical interface devices depends on the
initialization and configuration of the PHY chip. The task of driving
the PHY chip is implemente
On 2021/3/1 下午7:40, Ferruh Yigit wrote:
Please do not top post, message moved down.
On 2/26/2021 5:47 PM, 张 杨 wrote:
I think your idea is fine
What do you think just record file path in "pmd_pcap_probe()",
Perform an open operation only in "eth_dev_start()"?
When the secondary process add pc
pcap fd was opend when vdev probed,
but not closed when vdev removed.
This bug appears in dpdk-pdump
Signed-off-by: ZhangTengfei
---
drivers/net/pcap/rte_eth_pcap.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/pcap/rte_eth_pcap.c b/drivers/net/pcap/rte_eth_pcap.c
index 90f5d
On 2021/2/25 17:52, David Marchand wrote:
On Thu, Feb 25, 2021 at 5:00 AM 谢华伟(此时此刻） wrote:
Is the 'outb_p' to 'outb' conversion intentional? And if so why?
Same of the all 'outb_p', 'outw_p', 'outl_p'.
There is no need to delay for virtio device, as we can see in virtio
legacy driver.
IMO,
On 2021/2/25 1:52, David Marchand wrote:
On Wed, Feb 24, 2021 at 4:29 PM 谢华伟(此时此刻） wrote:
Did you check that virtio devices bound to uio_pci_generic still works
with legacy mode + PIO?
I had verified PIO, might under igb_uio driver.
Well, if you are unsure, please retest both cases, igb_uio
01/03/2021 14:20, Dumitrescu, Cristian:
> From: Li Zhang
> > As specified in the MAINTEINERS file of DPDK, I am the maintainer of this
> > API, so please make sure you add my email in the To: list of future
> > revisions of this patch set.
This kind of miss should be solved by using
--cc-cmd dev
From: "huawei.xhw"
Currently virtio PMD asssumes legacy device uses PIO bar.
There are three ways to get PIO(PortIO) address for virtio legacy device.
under igb_uio, get pio address from uio/uio# sysfs attribute
under uio_pci_generic:
for X86, get PIO address from /proc/ioport
From: "huawei.xhw"
With IO BAR, we get PIO(programmed IO) address.
With MMIO BAR, we get mapped virtual address.
We distinguish PIO(Programmed IO) and MMIO(memory mapped IO) by their address
like how kernel does.
ioread/write8/16/32 is provided to access PIO/MMIO.
By the way, for virtio on arch
From: "huawei.xhw"
virtio PMD assumes legacy device only supports PIO BAR resource. This is wrong.
As we need to create lots of devices, as PIO resource on x86 is very limited,
we expose MMIO(memory IO) BAR.
Kernel supports both PIO and MMIO BAR for legacy virtio-pci device, and for all
other p
On 2/24/2021 11:18 AM, Ciara Loftus wrote:
Prior to this the max size was 32 which was unnecessarily
small.
Can you please describe the impact? Why changed from 32 to 512?
I assume this is to improve the performance but can you please explicitly
document it in the commit log?
Also enforce
On 2/24/2021 11:18 AM, Ciara Loftus wrote:
This commit introduces support for preferred busy polling
to the AF_XDP PMD. This feature aims to improve single-core
performance for AF_XDP sockets under heavy load.
A new vdev arg is introduced called 'busy_budget' whose default
value is 64. busy_budg
On Mon, 1 Mar 2021 11:10:01 +0300
Igor Ryzhov wrote:
> Stephen,
>
> No, I don't have a better proposal, but I think it is not correct to change
> the behavior of KNI (making link down without a real response).
> Even though we know that communicating with userspace under rtnl_lock is a
> bad ide
On Wed, 24 Feb 2021 15:20:13 -0600
Honnappa Nagarahalli wrote:
> rte_eal_remote_launch and rte_eal_wait_lcore need to provide
> correct memory ordering to address the data communication from
> main core to worker core and vice versa.
>
> There are 2 use cases:
> 1) All the store operations (mean
Fixed issue could occur when Mbuf starvation happens
in a middle of reception of a segmented packet.
In such a situation, after release the segments of that
packet, it does not align consumer index to the next stride.
This would cause receive a wrong segmented packet.
Here is a possible error.
- w
This patch set removes DLB V1 from DPDK 21.05.
All DLB V1 customers obtain the source code for that device directly
from Intel, and will continue to do so in the future.
DLB V2 remains in DPDK, and will be supplemented with support
for DLB V2.5 in a future patch set tageted for DPDK 21.05.
Timot
Support for DLB v1 is being removed from dpdk in 21.05.
Signed-off-by: Timothy McDaniel
---
config/rte_config.h | 6 --
drivers/event/meson.build | 2 +-
2 files changed, 1 insertion(+), 7 deletions(-)
diff --git a/config/rte_config.h b/config/rte_config.h
index 55a2fc50e..aedb68c42 1
Hi, Slava
Thank you for your agreement. Here is the v2 patch:
https://patches.dpdk.org/project/dpdk/patch/1614617885-2650-1-git-send-email-17826875...@163.com/
With best regards,
Jiawei
On 2021/3/1 5:13 PM, Slava Ovsiienko wrote:
Hi, Jiawei
Thank you for the clarification. I missed the point
Support for DLB v1 is being removed from dpdk in 21.05.
Signed-off-by: Timothy McDaniel
---
doc/api/doxy-api-index.md | 1 -
doc/api/doxy-api.conf.in | 1 -
doc/guides/eventdevs/dlb.rst | 341 -
doc/guides/eventdevs/index.rst | 1 -
4 files chan
Support for DLB v1 is being removed from dpdk in 21.05.
Signed-off-by: Timothy McDaniel
---
doc/guides/rel_notes/release_21_05.rst | 2 ++
1 file changed, 2 insertions(+)
diff --git a/doc/guides/rel_notes/release_21_05.rst
b/doc/guides/rel_notes/release_21_05.rst
index 5aa9ed7db..a941d89ef 100
On 2/27/2021 12:32 PM, Dmitry Kozlyuk wrote:
Fixes: c08bd191b13d ("eal/windows: initialize hugepage info")
Cc: sta...@dpdk.org
Reported-by: Stephen Hemminger
Signed-off-by: Dmitry Kozlyuk
---
lib/librte_eal/windows/eal_hugepages.c | 4
1 file changed, 4 insertions(+)
diff --git a/lib/
Below is NVIDIA Mellanox's roadmap for DPDK21.05, on which we are currently
working:
rte_flow new APIs:
===
[1]Support a new action offload which perform connection tracking window
validation.
Motivation:
TCP connection tracking is needed for many applications that act as a mediator
Hi,
Sorry if I already asked this question.
Would it be possible to use rte_log in rte_telemetry
instead of returning telemetry_log_error at init?
Hi All,
Sorry to butt in on this, but I fixed this same issue about 3 years
ago in my application, but I was never able to get the changes
integrated and eventually just gave up trying.
The rule with KNI is:
1) The app should have a separate control thread per rte_kni which
just spins calling rte
This is from my git commit fixing this:
kni: separate releasing netdev from freeing KNI interface
Currently the rte_kni kernel driver suffers from a problem where
when the interface is released, it generates a callback to the DPDK
application to change the interface state to Down. Ho
There's a meson issue with `cc.has_function()`:
https://github.com/mesonbuild/meson/issues/5628
What if we just define RTE_INTERNAL for librte_eal/windows/include/rte_os.h
(and other public headers if need be) to distinguish the case when it's used
from within DPDK?
It's a pragmatic solution t
Complete removal of non-standard dependencies in headers is within a grasp.
Then we can remove shims and include whatever needed.
Thoughts?
Sounds good. A couple of 'gotchas' that I've come across (but may not be
an issue for DPDK):
* Memory allocation / free that spans a dll boundary (see
+%vmxnet3.Description%=netuio_Device, PCI\VEN_15AD&DEV_07B0
Acked-by: Nick Connolly
PCI hardware IDs verified against:
https://devicehunt.com/view/type/pci/vendor/15AD/device/07B0
Calculation of usec value is incorrect - 'ticks' is not adjusted for
epoch, whereas 'sec' has been, also subtraction has mismatched units -
ticks in 100ns and sec * USEC_PER_SEC in usecs.
On 14/02/2021 01:20, Dmitry Kozlyuk wrote:
+ GetSystemTimePreciseAsFileTime(&ft);
+ ticks = (
Use PRIu32 for uint32_t (found by -Wformat with Clang on Windows).
Acked-by: Nick Connolly
Calculation of usec value is incorrect - 'ticks' is not adjusted for
epoch, whereas 'sec' has been, also subtraction has mismatched units -
ticks in 100ns and sec * USEC_PER_SEC in usecs.
Ignore - fixed in later version which I hadn't spotted.
Many PMDs use POSIX gettimeofday(). Add rte_time_get_us() function
to obtain current time with microsecond precision on all platforms.
Acked-by: Nick Connolly
2021-03-01 21:43, Nick Connolly:
> > Complete removal of non-standard dependencies in headers is within a grasp.
> > Then we can remove shims and include whatever needed.
> > Thoughts?
> Sounds good. A couple of 'gotchas' that I've come across (but may not be
> an issue for DPDK):
>
> * Memory
2021-03-02 02:05, Dmitry Kozlyuk:
> 2021-03-01 21:43, Nick Connolly:
> > > Complete removal of non-standard dependencies in headers is within a
> > > grasp.
> > > Then we can remove shims and include whatever needed.
> > > Thoughts?
> > Sounds good. A couple of 'gotchas' that I've come across (b
2021-03-01 21:31, Nick Connolly:
> > There's a meson issue with `cc.has_function()`:
> > https://github.com/mesonbuild/meson/issues/5628
> >
> > What if we just define RTE_INTERNAL for librte_eal/windows/include/rte_os.h
> > (and other public headers if need be) to distinguish the case when it's us
Thanks Thomas.
I will use it next time.
Regards,
Li Zhang
> -Original Message-
> From: dev On Behalf Of Thomas Monjalon
> Sent: Monday, March 1, 2021 11:54 PM
> To: Li Zhang
> Cc: Dekel Peled ; Ori Kam ; Slava
> Ovsiienko ; Matan Azrad ;
> Dumitrescu, Cristian ; dev@dpdk.org; Raslan
> Da
Hi
> -Original Message-
> From: dev On Behalf Of Li Zhang
> Sent: Monday, March 1, 2021 18:36
> To: dek...@nvidia.com; or...@nvidia.com; viachesl...@nvidia.com;
> ma...@nvidia.com
> Cc: dev@dpdk.org; tho...@monjalon.net; rasl...@nvidia.com;
> m...@smartsharesystems.com; ajit.khapa...@broa
https://bugs.dpdk.org/show_bug.cgi?id=646
Bug ID: 646
Summary: mrvl: memset wrong struct when session clear
Product: DPDK
Version: 20.08
Hardware: ARM
OS: Linux
Status: UNCONFIRMED
Severity: minor
V3:
Clear the variable name of input_set_o and input_set_i.
Use input_set_o for tunnerl outer or non-tunnel fields, let
ice_pattern_match_item 3rd columns all none in ACL, RSS, switch.
Add new patch to use seg_tun[1] to record tunnel inner part.
Add new patch to clean GTPU inner flow type for PF.
Currently, the macro of input set use 2 bits, one bit for protocol and
inner/outer, another bit for src/dst field. But this could not distinguish
a rule with inner and outer fields for tunnel packet.
Redefine input set macro to make it clear. Only use these two bits for
protocol and field. Ignore t
This patch use input_set_o and input_set_i to distinguish inner/outer
input set. input_set_i is only used for inner field.
Signed-off-by: Zhirun Yan
---
drivers/net/ice/ice_ethdev.h | 4 ++--
drivers/net/ice/ice_fdir_filter.c | 6 +++---
2 files changed, 5 insertions(+), 5 deletions(-)
dif
Distinguish inner/outer input_set. And avoid too many nested
conditionals in each type's parser. input_set_o is used for
tunnel outer fields or non-tunnel fields , input_set_i is only
used for inner fields.
For GTPU, store the outer IP fields in inner part to align with
shared code behavior.
Sign
For tunnel or non-tunnel packet, the input set is in outer_input_set
and use seg_tun[0]. seg_tun[1] is only used for tunnel inner fields.
This patch make align with input_set inner/outer with seg_tun[] and
simplify it.
Signed-off-by: Zhirun Yan
---
drivers/net/ice/ice_fdir_filter.c | 12
Distinguish input_set_mask to inner and outer part. Use
input_set_mask_o for tunnel outer or non-tunnel input set.
input_set_mask_i is used for tunnel inner fields only. Adjust
indentation of ice_pattern_match_item list in switch, ACL, RSS and
FDIR for easy review. For switch, ACL and RSS, only use
Currently, FDIR only support GTPU outer fields in PF. Clean the
redundant GTPU inner info in flow type definition and align with
shared code.
Signed-off-by: Zhirun Yan
Signed-off-by: Junfeng Guo
---
drivers/net/ice/ice_fdir_filter.c | 19 ---
1 file changed, 8 insertions(+), 11
Hi Xiao yun,
Thanks for your comments.
We will follow other commands in our new command's helper_str.
Regards,
Li Zhang
> -Original Message-
> From: dev On Behalf Of Li, Xiaoyun
> Sent: Tuesday, March 2, 2021 9:48 AM
> To: Li Zhang ; Dekel Peled ; Ori Kam
> ; Slava Ovsiienko ; Matan
> Az
> >
> > > > Subject: [RFC 3/5] eal: lcore state FINISHED is not required
> > > >
> > > > FINISHED state seems to be used to indicate that the worker's
> > > > update of the 'state' is not visible to other threads. There seems
> > > > to be no requirement to have such a state.
> > >
> > > I am not
1 - 100 of 124 matches
Mail list logo
