05/05/2021 05:43, Chengwen Feng:
> Currently, the mp uses gettimeofday() API to get the time, and used as
> timeout parameter.
>
> But the time which gets from gettimeofday() API isn't monotonically
> increasing. The process may fail if the system time is changed.
>
> This fixes it by using clock
02/05/2021 04:33, Dmitry Kozlyuk:
> Interrupt manager in Windows EAL allocates on IOCP and starts
> a control thread that runs indefinitely. At DPDK cleanup
> this thread was not stopped and IOCP handle was not closed.
>
> Gracefully stop interrupt-handling in rte_eal_cleanup().
> The thread alrea
> -Original Message-
> From: dev On Behalf Of Xueming Li
> Sent: Monday, May 10, 2021 16:14
> Cc: dev@dpdk.org; Xueming(Steven) Li ;
> sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH v2] net/mlx5: fix LAG representor probe on PF1
> PCI
>
> In case of bonding, orchestrator wants to use same d
https://bugs.dpdk.org/show_bug.cgi?id=701
Bug ID: 701
Summary: mlx5: reading xstats smashes stack
Product: DPDK
Version: 20.11
Hardware: x86
OS: Linux
Status: UNCONFIRMED
Severity: normal
Priorit
10/05/2021 18:04, Lance Richardson:
> On Mon, May 10, 2021 at 11:55 AM Thomas Monjalon
> wrote:
>
> > 10/05/2021 17:31, Burakov, Anatoly:
> > > On 10-May-21 1:42 PM, Thomas Monjalon wrote:
> > > > +Cc Anatoly
> > > >
> > > > 08/05/2021 16:27, Lance Richardson:
> > > >> For 32-bit targets, size_t
2021-05-11 09:41 (UTC+0200), Thomas Monjalon:
> 02/05/2021 04:33, Dmitry Kozlyuk:
> > Interrupt manager in Windows EAL allocates on IOCP and starts
> > a control thread that runs indefinitely. At DPDK cleanup
> > this thread was not stopped and IOCP handle was not closed.
> >
> > Gracefully stop i
Tunnel offload model requires application to obtain PMD related flow
items or actions to construct a flow rule. These elements acquire
internal PMD flow resources that must be explicitly released.
The patch destroys tunnel offload PMD resources after flow creation
failure.
Fixes: 1b9f274623b8 ("a
> -Original Message-
> From: Zhang, AlvinX
> Sent: Tuesday, May 11, 2021 10:39 AM
> To: Xing, Beilei ; Rong, Leyi
> Cc: dev@dpdk.org; Zhang, AlvinX ; sta...@dpdk.org
> Subject: [PATCH v2] net/i40e: fix offload flag checking in simple Tx datapath
>
> Tx offload flags 'PKT_TX_IPV6, PKT_T
08/05/2021 16:27, Lance Richardson:
> For 32-bit targets, size_t is normally a 32-bit type and
> does not have sufficient range to represent 64-bit offsets
> that can are needed when mapping PCI addresses. Use uint64_t
> instead.
>
> Found when attempting to run 32-bit Linux dpdk-testpmd
> using V
Hi, Feifei
Please, see below.
> -Original Message-
> From: Feifei Wang
> Sent: Saturday, May 8, 2021 6:13
> To: Slava Ovsiienko ; Matan Azrad
> ; Shahaf Shuler
> Cc: dev@dpdk.org; nd ; sta...@dpdk.org; Ruifeng Wang
> ; nd ; nd
> Subject: 回复: [PATCH v1 3/4] net/mlx5: fix rebuild bug for
> -Original Message-
> From: dev On Behalf Of Alvin Zhang
> Sent: Monday, May 10, 2021 8:21 PM
> To: Xing, Beilei ; Xu, Ting
> Cc: dev@dpdk.org; Zhang, AlvinX
> Subject: [dpdk-dev] [PATCH] net/iavf: fix error logs
>
> Remove some useless logs which are duplicated of the lower-level f
On Fri, May 07, 2021 at 11:06:38AM +0200, David Marchand wrote:
> On Fri, May 7, 2021 at 12:24 AM Dmitry Kozlyuk
> wrote:
> >
> > 2021-05-06 17:14 (UTC+0200), David Marchand:
> > > Spotted by sparse in OVS build:
> > > ../../lib/netdev-dpdk.c: note: in included file (through
> > > /home/runner/wo
On Fri, May 07, 2021 at 12:12:09PM -0700, Ben Pfaff wrote:
> On Fri, May 07, 2021 at 11:06:38AM +0200, David Marchand wrote:
> > On Fri, May 7, 2021 at 12:24 AM Dmitry Kozlyuk
> > wrote:
> > >
> > > 2021-05-06 17:14 (UTC+0200), David Marchand:
> > > > Spotted by sparse in OVS build:
> > > > ../..
> -Original Message-
> From: Yigit, Ferruh
> Sent: Monday, May 10, 2021 10:32 AM
> To: Wang, Haiyue ; Yang, Qiming
> ; Zhang, Qi Z ; Stillwell Jr,
> Paul
> M ; Lu, Wenzhuo ;
> Rong, Leyi ; Shukla, Shivanshu
>
> Cc: dev@dpdk.org; sta...@dpdk.org; Kevin Traynor ;
> Ajit Khaparde
> Subject
>> -Original Message-
>> From: pbhagavat...@marvell.com
>> Sent: Tuesday, May 11, 2021 03:40
>> To: david.march...@redhat.com; Wang, Haiyue
>; Matan Azrad ;
>> Shahaf Shuler ; Viacheslav Ovsiienko
>; Jiawen Wu
>> ; Jian Wang ;
>Chautru, Nicolas
>> ; Thomas Monjalon
>; Yigit, Ferruh
>> ; An
Tx offload flags 'PKT_TX_IPV6, PKT_TX_IPV4, PKT_TX_OUTER_IPV6,
PKT_TX_OUTER_IPV4' are supported in simple datapath.
This patch removes these offload flags from packet checking in simple
Tx datapath and defines 2 macro I40E_TX_OFFLOAD_SIMPLE_SUP_MASK
and I40E_TX_OFFLOAD_SIMPLE_NOTSUP_MASK.
Fixes:
>
>
>>
>> >
>> > The patch still holds true for CRC though as it is listed separately
>> > below
>> > https://urldefense.proofpoint.com/v2/url?u=https-
>3A__developer.arm.com_architectures_cpu-2Darchitecture_a-
>2D&d=DwIFAg&c=nKjWec2b6R0mOyPaz7xtfQ&r=E3SgYMjtKCMVsB-
>fmvgGV3o-
>g_fjLhk5Pupi9ijohpc&
On 10/05/2021 18:28, Ferruh Yigit wrote:
> On 5/10/2021 6:04 PM, Wang, Haiyue wrote:
>>> -Original Message-
>>> From: dev On Behalf Of Ferruh Yigit
>>> Sent: Monday, May 10, 2021 23:03
>>> To: Yang, Qiming ; Zhang, Qi Z
>>> ; Stillwell Jr, Paul M
>>> ; Lu, Wenzhuo ; Rong,
>>> Leyi ;
>>>
> -Original Message-
> From: Zhang, AlvinX
> Sent: Tuesday, May 11, 2021 4:46 PM
> To: Xing, Beilei ; Rong, Leyi
> Cc: dev@dpdk.org; Zhang, AlvinX ; sta...@dpdk.org
> Subject: [PATCH v3] net/i40e: fix offload flag checking in simple Tx datapath
>
> Tx offload flags 'PKT_TX_IPV6, PKT_TX
> -Original Message-
> From: dev On Behalf Of Chengwen Feng
> Sent: Monday, May 10, 2021 15:06
> To: NBU-Contact-Thomas Monjalon ;
> ferruh.yi...@intel.com
> Cc: dev@dpdk.org
> Subject: [dpdk-dev] [PATCH 1/2] net/mlx4: fix memory barrier incorrectly
> placed
>
> The memory barrier is used
> -Original Message-
> From: dev On Behalf Of Chengwen Feng
> Sent: Monday, May 10, 2021 15:06
> To: NBU-Contact-Thomas Monjalon ;
> ferruh.yi...@intel.com
> Cc: dev@dpdk.org
> Subject: [dpdk-dev] [PATCH 2/2] net/mlx5: fix memory barrier incorrectly
> placed
>
> The memory barrier is used
Currently action RTE_FLOW_ACTION_TYPE_METER_COLOR is defined.
Add the CLI for this action in meter policy: color type (types)
There are three types: green, yellow and red.
Example for the new policy meter CLIs:
add port meter policy 0 1 g_actions color type green / end y_actions
color t
On 5/11/2021 2:59 AM, Wang, Haiyue wrote:
>> -Original Message-
>> From: Yigit, Ferruh
>> Sent: Tuesday, May 11, 2021 01:28
>> To: Wang, Haiyue ; Yang, Qiming
>> ; Zhang, Qi Z
>> ; Stillwell Jr, Paul M
>> ; Lu, Wenzhuo
>> ; Rong, Leyi ; Shukla, Shivanshu
>>
>> Cc: dev@dpdk.org; sta...@d
On Fri, Mar 19, 2021 at 9:38 PM David Marchand
wrote:
>
> Enforce that added logtypes follow the convention:
> - for lib/librte_XXX: lib.XXX and lib.XXX.*
> - for drivers/class/XXX: pmd.class.XXX and pmd.class.XXX.*
As discussed offlist, I dropped this idea.
Instead we can use helpers that provi
On 10/05/2021 14:47, Xueming Li wrote:
> Auxiliary [1] provides a way to split function into child-devices
> representing sub-domains of functionality. Each auxiliary_device
> represents a part of its parent functionality.
>
> Auxiliary device is identified by unique device name, sysfs path:
>
From: David Marchand
Let's try to enforce the convention where most drivers use a pmd. logtype
with their class reflected in it, and libraries use a lib. logtype.
Introduce two new macros:
- RTE_LOG_REGISTER_DEFAULT can be used when a single logtype is
used in a component. It is associated to
Currently, the mp uses gettimeofday() API to get the time, and used as
timeout parameter.
But the time which gets from gettimeofday() API isn't monotonically
increasing. The process may fail if the system time is changed.
This fixes it by using clock_gettime() API with monotonic attribution.
Fix
On 2021/5/11 15:36, Thomas Monjalon wrote:
> 05/05/2021 05:43, Chengwen Feng:
>> Currently, the mp uses gettimeofday() API to get the time, and used as
>> timeout parameter.
>>
>> But the time which gets from gettimeofday() API isn't monotonically
>> increasing. The process may fail if the syste
On 2021/5/11 17:23, Slava Ovsiienko wrote:
>> -Original Message-
>> From: dev On Behalf Of Chengwen Feng
>> Sent: Monday, May 10, 2021 15:06
>> To: NBU-Contact-Thomas Monjalon ;
>> ferruh.yi...@intel.com
>> Cc: dev@dpdk.org
>> Subject: [dpdk-dev] [PATCH 1/2] net/mlx4: fix memory barrier
On Tue, May 11, 2021 at 12:09 PM Thomas Monjalon wrote:
>
> From: David Marchand
>
> Let's try to enforce the convention where most drivers use a pmd. logtype
> with their class reflected in it, and libraries use a lib. logtype.
>
> Introduce two new macros:
> - RTE_LOG_REGISTER_DEFAULT can be us
All,
Update the test status for Intel part. Till now dpdk21.05-rc2 test is almost
finished, find 1 critical bad commit id: "b7aaf417f9 raw/ioat: add bus driver
for device scanning automatically",
which blocked most of VF/kni test. Bruce has worked on this and has provided
patch(http://patches.d
On 2021/5/9 2:46, Honnappa Nagarahalli wrote:
>
>
>>
>> Thanks for your suggestions, we found that the -fno-tree-vectorize option
>> works.
>> PS: This option is not successfully added in the earliest test.
>>
>> Solution:
>> 1. use the -fno-tree-vectorize option to prevent compiler generate a
A terminated pthread should be joined or detached so that its associated
resources are released.
The "ice-reset-" threads are used to service some reset task in the
background, but they are never joined by the thread that created them.
The easiest solution is to detach new threads.
The Windows EA
After looking at the control thread API use in recent changes, I noticed
that some thread resources are not properly released.
This series fixes two net drivers.
Additional notes:
- I noticed that event/dlb2 never kills/terminates its service thread.
It means unplug of this device (if supported)
A terminated pthread should be joined or detached so that its associated
resources are released.
The "ark-delay-pg" thread is just used to delay some task but it is never
joined by the thread that created it.
The easiest solution is to detach the new thread.
Fixes: 727b3fe292bc ("net/ark: integra
On 5/11/2021 10:28 AM, Haifei Luo wrote:
> Currently action RTE_FLOW_ACTION_TYPE_METER_COLOR is defined.
> Add the CLI for this action in meter policy: color type (types)
> There are three types: green, yellow and red.
>
> Example for the new policy meter CLIs:
>add port meter policy 0 1 g_act
Reproduced with '--buildtype=debugoptimized' config,
compiler version: gcc (GCC) 12.0.0 20210509 (experimental)
Build error:
In file included from ../drivers/net/bnx2x/bnx2x_rxtx.c:8:
../drivers/net/bnx2x/bnx2x_rxtx.c: In function ‘bnx2x_upd_rx_prod_fast’:
../drivers/net/bnx2x/bnx2x.h:1528:35:
Reproduced with '--buildtype=debugoptimized' config,
compiler version: gcc (GCC) 12.0.0 20210509 (experimental)
Build error:
In file included from ../drivers/net/bnx2x/bnx2x.c:16:
../drivers/net/bnx2x/bnx2x.c: In function ‘bnx2x_hc_ack_sb’:
../drivers/net/bnx2x/bnx2x.h:1528:35:
warning: ‘
Reproduced with '--buildtype=debugoptimized' config,
compiler version: gcc (GCC) 12.0.0 20210509 (experimental)
There are multiple build errors, like:
../drivers/net/ice/base/ice_switch.c: In function ‘ice_add_marker_act’:
../drivers/net/ice/base/ice_switch.c:3727:15:
warning: array subscr
Reproduced with '--buildtype=debugoptimized' config,
compiler version: gcc (GCC) 12.0.0 20210509 (experimental)
There are multiple build errors, like:
In file included from ../drivers/net/tap/tap_flow.c:13:
In function ‘rte_jhash_2hashes’,
inlined from ‘rte_jhash’ at ../lib/hash/rte_jhash.h:28
On 4/29/2021 9:10 AM, Raslan Darawsheh wrote:
> Define new rte header for gtp PDU session container
> based on RFC 38415-g30
>
> Signed-off-by: Raslan Darawsheh
This patch has not urgent for this release, postponing it to next release.
On Tue, May 11, 2021 at 12:09:27PM +0200, Thomas Monjalon wrote:
> From: David Marchand
>
> Let's try to enforce the convention where most drivers use a pmd. logtype
> with their class reflected in it, and libraries use a lib. logtype.
>
> Introduce two new macros:
> - RTE_LOG_REGISTER_DEFAULT c
On Tue, May 11, 2021 at 12:09:27PM +0200, Thomas Monjalon wrote:
> From: David Marchand
>
> Let's try to enforce the convention where most drivers use a pmd. logtype
> with their class reflected in it, and libraries use a lib. logtype.
>
> Introduce two new macros:
> - RTE_LOG_REGISTER_DEFAULT c
11/05/2021 13:49, Bruce Richardson:
> On Tue, May 11, 2021 at 12:09:27PM +0200, Thomas Monjalon wrote:
> > +log_prefix
> > + **Default Value = pmd.**
> > + Some driver classes have a customized log prefix.
> > + This prefix is combined with the driver name
> > + when registering a default l
> -Original Message-
> From: fengchengwen
> Sent: Tuesday, May 11, 2021 14:09
> To: Slava Ovsiienko ; NBU-Contact-Thomas
> Monjalon ; ferruh.yi...@intel.com
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 1/2] net/mlx4: fix memory barrier incorrectly
> placed
>
>
>
> On 2021/5/11 17
11/05/2021 13:59, Bruce Richardson:
> On Tue, May 11, 2021 at 12:09:27PM +0200, Thomas Monjalon wrote:
> > --- a/drivers/meson.build
> > +++ b/drivers/meson.build
> > @@ -59,6 +59,11 @@ foreach subpath:subdirs
> > drivers += driver_path[1]
> > else
> > class = subpath
> > +
On 5/11/2021 7:45 AM, Michal Krawczyk wrote:
> To make it possible to the app to determine if the hash was calculated
> for the packet or not, the PKT_RX_RSS_HASH should be set in the mbuf's
> ol_flags.
>
> As the PMD wasn't setting that, the application couldn't check if there
> is a hash in a pr
On 5/11/2021 7:45 AM, Michal Krawczyk wrote:
> From: Stanislaw Kardach
>
> rte_pci_device and rte_eth_dev are process-local structures. Therefore
> ena_adapter::pdev and ena_adapter::rte_dev cannot be used universally.
> Switch this to extracting those structures via rte_eth_devices indexing
> an
On Tue, May 11, 2021 at 02:10:47PM +0200, Thomas Monjalon wrote:
> 11/05/2021 13:59, Bruce Richardson:
> > On Tue, May 11, 2021 at 12:09:27PM +0200, Thomas Monjalon wrote:
> > > --- a/drivers/meson.build
> > > +++ b/drivers/meson.build
> > > @@ -59,6 +59,11 @@ foreach subpath:subdirs
> > >
On 11/05/2021 12:44, Ferruh Yigit wrote:
> Reproduced with '--buildtype=debugoptimized' config,
> compiler version: gcc (GCC) 12.0.0 20210509 (experimental)
>
> Build error:
> In file included from ../drivers/net/bnx2x/bnx2x_rxtx.c:8:
> ../drivers/net/bnx2x/bnx2x_rxtx.c: In function ‘bnx2x_upd_rx_
On 11/05/2021 12:44, Ferruh Yigit wrote:
> Reproduced with '--buildtype=debugoptimized' config,
> compiler version: gcc (GCC) 12.0.0 20210509 (experimental)
>
> Build error:
> In file included from ../drivers/net/bnx2x/bnx2x.c:16:
> ../drivers/net/bnx2x/bnx2x.c: In function ‘bnx2x_hc_ack_sb’:
> ..
On 10/05/2021 16:03, Ferruh Yigit wrote:
> Reproduced with '--buildtype=debugoptimized' config,
> compiler version: gcc (GCC) 12.0.0 20210509 (experimental)
>
> There are multiple build errors, like:
> In file included from ../drivers/net/tap/tap_flow.c:13:
> In function ‘rte_jhash_2hashes’,
>
On 5/11/2021 7:45 AM, Michal Krawczyk wrote:
> From: Stanislaw Kardach
>
> For dev_ops not supported in SMP, either return -EPERM or return without
> doing anything. In both cases log a warning.
>
'SMP' can be confusing in this context, this is not exactly related to SMP. In
DPDK we tend to cal
> -Original Message-
> From: Yigit, Ferruh
> Sent: Tuesday, May 11, 2021 12:35 PM
> To: Haifei Luo ; ma...@nvidia.com; or...@nvidia.com;
> viachesl...@nvidia.com; Li, Xiaoyun
> Cc: dev@dpdk.org; tho...@monjalon.net; rasl...@nvidia.com; Singh,
> Jasvinder ; Dumitrescu, Cristian
>
> Subj
From: David Marchand
Let's try to enforce the convention where most drivers use a pmd. logtype
with their class reflected in it, and libraries use a lib. logtype.
Introduce two new macros:
- RTE_LOG_REGISTER_DEFAULT can be used when a single logtype is
used in a component. It is associated to
10/05/2021 22:48, Thomas Monjalon:
> Sphinx 4.0 became stricter with permalink configuration:
> "
> html_add_permalinks has been deprecated since v3.5.0.
> Please use html_permalinks and html_permalinks_icon instead.
> "
>
> The new variable is used while keeping compatibility
> with older Sphinx
On Mon, May 10, 2021 at 10:48:07PM +0200, Thomas Monjalon wrote:
> Sphinx 4.0 became stricter with permalink configuration:
> "
> html_add_permalinks has been deprecated since v3.5.0.
> Please use html_permalinks and html_permalinks_icon instead.
> "
>
> The new variable is used while keeping comp
On Tue, May 11, 2021 at 02:38:10PM +0200, Thomas Monjalon wrote:
> From: David Marchand
>
> Let's try to enforce the convention where most drivers use a pmd. logtype
> with their class reflected in it, and libraries use a lib. logtype.
>
> Introduce two new macros:
> - RTE_LOG_REGISTER_DEFAULT c
On 5/11/2021 7:45 AM, Michal Krawczyk wrote:
> Hi,
>
> this version updates the driver to version 2.3.0, which fixes multiple
> bugs, contains part of the work on making the ENA PMD fully MP aware, and
> also updates HAL to the latest version.
>
> More detailed list of changes:
> * memcpy mappi
On Thu, May 06, 2021 at 05:14:25PM +0200, David Marchand wrote:
> Spotted by sparse in OVS build:
>
> /home/runner/work/ovs/ovs/dpdk-dir/build/include/rte_flow.h:789:27:
> error: incorrect type in initializer (different base types)
> /home/runner/work/ovs/ovs/dpdk-dir/build/include/rte_flow.h:789:
On Thu, May 06, 2021 at 05:14:26PM +0200, David Marchand wrote:
> Spotted by sparse in OVS build:
> ../../lib/netdev-dpdk.c: note: in included file (through
> /home/runner/work/ovs/ovs/dpdk-dir/build/include/rte_ip.h,
> /home/runner/work/ovs/ovs/dpdk-dir/build/include/rte_flow.h, ...):
> ../../incl
Reproduced with '--buildtype=debugoptimized' config,
compiler version: gcc (GCC) 12.0.0 20210509 (experimental)
Build error:
In file included from ../drivers/net/bnx2x/bnx2x_rxtx.c:8:
../drivers/net/bnx2x/bnx2x_rxtx.c: In function ‘bnx2x_upd_rx_prod_fast’:
../drivers/net/bnx2x/bnx2x.h:1528:35:
Reproduced with '--buildtype=debugoptimized' config,
compiler version: gcc (GCC) 12.0.0 20210509 (experimental)
Build error:
In file included from ../drivers/net/bnx2x/bnx2x.c:16:
../drivers/net/bnx2x/bnx2x.c: In function ‘bnx2x_hc_ack_sb’:
../drivers/net/bnx2x/bnx2x.h:1528:35:
warning: ‘
Reproduced with '--buildtype=debugoptimized' config,
compiler version: gcc (GCC) 12.0.0 20210509 (experimental)
There are multiple build errors, like:
../drivers/net/ice/base/ice_switch.c: In function ‘ice_add_marker_act’:
../drivers/net/ice/base/ice_switch.c:3727:15:
warning: array subscr
Reproduced with '--buildtype=debugoptimized' config,
compiler version: gcc (GCC) 12.0.0 20210509 (experimental)
There are multiple build errors, like:
In file included from ../drivers/net/tap/tap_flow.c:13:
In function ‘rte_jhash_2hashes’,
inlined from ‘rte_jhash’ at ../lib/hash/rte_jhash.h:28
On 5/11/2021 1:34 PM, Dumitrescu, Cristian wrote:
>
>
>> -Original Message-
>> From: Yigit, Ferruh
>> Sent: Tuesday, May 11, 2021 12:35 PM
>> To: Haifei Luo ; ma...@nvidia.com; or...@nvidia.com;
>> viachesl...@nvidia.com; Li, Xiaoyun
>> Cc: dev@dpdk.org; tho...@monjalon.net; rasl...@nvi
11/05/2021 14:52, Bruce Richardson:
> On Tue, May 11, 2021 at 02:38:10PM +0200, Thomas Monjalon wrote:
> > From: David Marchand
> >
> > Let's try to enforce the convention where most drivers use a pmd. logtype
> > with their class reflected in it, and libraries use a lib. logtype.
> >
> > Introd
Any comment please?
29/04/2021 02:43, Min Hu (Connor):
>
> 在 2021/4/28 23:13, Pattan, Reshma 写道:
> >
> >
> >> -Original Message-
> >> From: dev On Behalf Of Min Hu (Connor)
> >>
> >> + if (lcore_id >= RTE_MAX_LCORE) {
> >> + RTE_LOG(ERR, GUEST_CHANNEL, "Channel(%u) is out of
On Tue, May 11, 2021 at 2:41 PM Thomas Monjalon wrote:
>
> 10/05/2021 22:48, Thomas Monjalon:
> > Sphinx 4.0 became stricter with permalink configuration:
> > "
> > html_add_permalinks has been deprecated since v3.5.0.
> > Please use html_permalinks and html_permalinks_icon instead.
> > "
> >
> >
This patch series is stucked.
That's too often the case for the power lib/example/test patches.
29/04/2021 12:34, Pattan, Reshma:
> From: dev On Behalf Of David Hunt
> > +TEST_POWER_SYSFILE_SCALING_FREQ, lcore_id) < 0) {
>
> > return 0;
>
> [Reshma]: Do we need to return -1 here and in other
10/05/2021 11:44, Anoob Joseph:
> Add checks to catch overflow of any offsets. Offset control word
> specifies,
>
> 1. 16 bits encryption offset
> 2. 8 bits IV offset
> 3. 8 bits auth offset
>
> Signed-off-by: Anoob Joseph
Should it be merged in DPDK 21.05-rc3, or wait for 21.08?
On 5/10/2021 5:45 PM, Heinrich Kuhn wrote:
> Before this change the dev_infos callback always reported RSS
> capabilities regardless of whether the capability is supported by the
> device or not. First check the capabilities field in the BAR of the
> device and advertise RSS functionality according
wt., 11 maj 2021 o 15:07 Ferruh Yigit napisał(a):
>
> On 5/11/2021 7:45 AM, Michal Krawczyk wrote:
> > Hi,
> >
> > this version updates the driver to version 2.3.0, which fixes multiple
> > bugs, contains part of the work on making the ENA PMD fully MP aware, and
> > also updates HAL to the latest
> -Original Message-
> From: Min Hu (Connor)
> > Another suggestion if you would like to improve the code, is below,
> > You can move "global_fds[lcore_id]" check to immediate after the line
> "fds.fd = global_fds[lcore_id]; "
> Hi, thanks Pattan.
> But I think "global_fds[lcore_id]"
Hi Thomas,
> Should it be merged in DPDK 21.05-rc3, or wait for 21.08?
Either is fine. This patch is just adding some extra parameter checks.
Thanks,
Anoob
> -Original Message-
> From: Thomas Monjalon
> Sent: Tuesday, May 11, 2021 6:59 PM
> To: Jerin Jacob Kollanukkaran ; Anoob Joseph
On Thu, May 6, 2021 at 5:15 PM David Marchand wrote:
>
> Testing OVS build against 21.05-rc2 revealed some issues.
> Bits were also missing on OVS side, so for people interested, a
> rebased dpdk-latest OVS branch is available at:
> https://github.com/david-marchand/ovs/commits/dpdk-latest
Series
On Thu, May 6, 2021 at 6:15 PM Brandon Lo wrote:
>
> Hi all,
>
> dpdk-next-qos and dpdk-next-pipeline remotes have been removed.
> dpdk-next-virtio and dpdk-next-crypto remotes have been switched to
> track main and for-main branches respectively.
Thank you Brandon.
Just an additional question.
On Sat, May 8, 2021 at 2:25 AM Piotr Kubaj wrote:
>
> Including below.
> From 87ad59cc894b227444936fb6a7968932abdc460f Mon Sep 17 00:00:00 2001
> From: Piotr Kubaj
> Date: Tue, 23 Mar 2021 00:40:47 +0100
> Subject: [PATCH] ppc64le: fix build without glibc
>
> __ppc_get_timebase() is only present
> >
> >>
> >> Thanks for your suggestions, we found that the -fno-tree-vectorize
> >> option works.
> >> PS: This option is not successfully added in the earliest test.
> >>
> >> Solution:
> >> 1. use the -fno-tree-vectorize option to prevent compiler generate
> >> auto vetorization
> >>code,
On 5/10/2021 9:47 AM, Ciara Loftus wrote:
> Prior to this change, two implementations of rx_syscall_handler
> existed although only one was needed (for the zero copy path which
> is only available from kernel 5.4 and onwards). Remove the second
> definition from compat.h and move the first definiti
On 5/5/2021 7:39 PM, John Daley wrote:
> The admin-configured vNIC settings (i.e. via CIMC or UCSM) now include
> Geneve offload. Use that setting to decide whether to enable or
> disable Geneve offload and remove the devarg 'geneve-opt'.
>
> Also, the firmware now allows the driver to change the
On 5/6/2021 2:30 PM, Min Hu (Connor) wrote:
> This patch adds more description for hns3 PMD document, that is:
> 'Sample Application Notes', 'Statistics','Performance tuning'.
>
> Signed-off-by: Min Hu (Connor)
Applied to dpdk-next-net/main, thanks.
On Fri, Mar 26, 2021 at 5:37 PM Lance Richardson
wrote:
>
> Fix two issues found when writing PMD unit tests for HW ptype and
> L4 checksum offload:
Would those unit tests be interesting to other pmd driver writers?
>
>- The version field in the IPv6 header was being set to zero,
> whi
On 29/4/2021 11:34 AM, Pattan, Reshma wrote:
-Original Message-
From: dev On Behalf Of David Hunt
+TEST_POWER_SYSFILE_SCALING_FREQ, lcore_id) < 0) {
return 0;
[Reshma]: Do we need to return -1 here and in other failure scenarios below.
Hi Reshma,
We might do, but that's not s
> -Original Message-
> From: Hunt, David
> > [Reshma]: Do we need to return -1 here and in other failure scenarios
> > below.
>
>
> Hi Reshma,
>
> We might do, but that's not something I had intended to address in this patch
> set.
>
> We should look at this in a future patch, OK?
11/05/2021 15:25, David Marchand:
> On Tue, May 11, 2021 at 2:41 PM Thomas Monjalon wrote:
> >
> > 10/05/2021 22:48, Thomas Monjalon:
> > > Sphinx 4.0 became stricter with permalink configuration:
> > > "
> > > html_add_permalinks has been deprecated since v3.5.0.
> > > Please use html_permalinks
Previously, the semantics of power monitor were such that we were
checking current value against the expected value, and if they matched,
then the sleep was aborted. This is somewhat inflexible, because it only
allowed us to check for a specific value.
We can reverse the check, and instead have mo
Implement support for .get_monitor_addr in AF_XDP driver.
Signed-off-by: Anatoly Burakov
---
drivers/net/af_xdp/rte_eth_af_xdp.c | 52 -
1 file changed, 37 insertions(+), 15 deletions(-)
diff --git a/drivers/net/af_xdp/rte_eth_af_xdp.c
b/drivers/net/af_xdp/rte_eth_a
On 11-May-21 4:31 PM, Anatoly Burakov wrote:
Implement support for .get_monitor_addr in AF_XDP driver.
Signed-off-by: Anatoly Burakov
---
static const struct eth_dev_ops ops = {
- .dev_start = eth_dev_start,
- .dev_stop = eth_dev_stop,
- .dev_close = eth_dev_close,
-
On 11-May-21 4:31 PM, Anatoly Burakov wrote:
Previously, the semantics of power monitor were such that we were
checking current value against the expected value, and if they matched,
then the sleep was aborted. This is somewhat inflexible, because it only
allowed us to check for a specific value.
2021-05-11 11:56 (UTC-0400), Lance Richardson:
> > v4: Identical to v1, with now included in rte_eal_paging.h
> > to (hopefully) make off_t available for Windows builds.
>
> With this version, using off_t is no longer a problem, however based on the
> new compilation error it appears that
On 5/7/2021 10:08 AM, Min Hu (Connor) wrote:
> From: Chengwen Feng
>
> This patch adds link status event pre-conditions.
>
> Signed-off-by: Chengwen Feng
> Signed-off-by: Min Hu (Connor)
> ---
> doc/guides/nics/hns3.rst | 10 ++
> 1 file changed, 10 insertions(+)
>
> diff --git a/doc
> -Original Message-
> From: Burakov, Anatoly
> Sent: Tuesday, May 11, 2021 10:32 AM
> To: dev@dpdk.org; McDaniel, Timothy ; Xing,
> Beilei ; Wu, Jingjing ; Yang,
> Qiming ; Zhang, Qi Z ; Wang,
> Haiyue ; Matan Azrad ; Shahaf
> Shuler ; Viacheslav Ovsiienko ;
> Richardson, Bruce ; Ananyev,
On 5/7/2021 10:08 AM, Min Hu (Connor) wrote:
> This patch set contains 6 bugfixes for hns3 PMD.
>
> Chengwen Feng (5):
> net/hns3: return code when PCI config space write fail
> net/hns3: fix logging info when clear all FDIR rules
> net/hns3: clear hash map when clear all FDIR rules
> net/
Hi All,
I just need few clarifications in the GRO code that is present
In the below code where the merge fails because the IP length is greater
than 65536 we do a insert_new_item but we send prev_idx rather than the
cur_idx. This would work for index 0 and 1 but when cur_idx is 1 and merge
fails,
On 5/8/2021 7:35 AM, Jiawen Wu wrote:
> Support to enable and disable QINQ hardware strip, when configure vlan
> offload with QINQ strip mask. If there are packets have QINQ tag to RSS,
> users should enable QINQ strip before configuring the RSS.
>
> Fixes: 220b0e49bc47 ("net/txgbe: support VLAN")
> >>
> >> >
> >> > The patch still holds true for CRC though as it is listed
> >> > separately below
> >> > https://urldefense.proofpoint.com/v2/url?u=https-
> >3A__developer.arm.com_architectures_cpu-2Darchitecture_a-
> >2D&d=DwIFAg&c=nKjWec2b6R0mOyPaz7xtfQ&r=E3SgYMjtKCMVsB-
> >fmvgGV3o-
> >g_fj
On 5/11/2021 12:59 AM, Dmitry Kozlyuk wrote:
2021-05-11 09:41 (UTC+0200), Thomas Monjalon:
02/05/2021 04:33, Dmitry Kozlyuk:
Interrupt manager in Windows EAL allocates on IOCP and starts
a control thread that runs indefinitely. At DPDK cleanup
this thread was not stopped and IOCP handle was n
On 5/1/2021 7:33 PM, Dmitry Kozlyuk wrote:
Interrupt manager in Windows EAL allocates on IOCP and starts
a control thread that runs indefinitely. At DPDK cleanup
this thread was not stopped and IOCP handle was not closed.
Gracefully stop interrupt-handling in rte_eal_cleanup().
The thread alre
> >
> > > > > +
> > > > > +fmt_name = 'mlx5_crypto'
> > > > > +deps += ['common_mlx5', 'eal', 'cryptodev']
> > > >
> > > > I don't see a reason for eal dependency. Do you really need it?
> > >
> > > See RTE_LOG_REGISTER.
> >
> > Compilation works fine even after removing this dependency.
> > I beli
1 - 100 of 163 matches
Mail list logo
