+0x6d/0x80
This patch also attempts to remove this option.
References:
[1] https://www.kernel.org/doc/Documentation/timers/timers-howto.txt
Signed-off-by: Tudor Cornea
Acked-by: Padraig Connolly
Reviewed-by: Ferruh Yigit
---
v5:
* Rebased the patch on top of the dpdk-next-net branch
v4
Hi Stephen,
> NAK
> Doing this risks having a CPU lockup if userspace does not keep up
> or the DPDK application gets stuck.
>
> There are better ways to solve the TCP stack queue overrun issue:
> 1. Use a better queueing discipline on the kni device. The Linux default
>of pfifo_fast has buff
On Fri, 14 Jan 2022 at 18:44, Ferruh Yigit wrote:
> On 1/14/2022 4:24 PM, Stephen Hemminger wrote:
> > On Fri, 14 Jan 2022 17:18:19 +0200
> > Tudor Cornea wrote:
> >
> >> +module_param(min_scheduling_interval, long, 0644);
> >> +MODULE_PARM_DESC(min_sch
+0x6d/0x80
This patch also attempts to remove this option.
References:
[1] https://www.kernel.org/doc/Documentation/timers/timers-howto.txt
Signed-off-by: Tudor Cornea
Acked-by: Padraig Connolly
Reviewed-by: Ferruh Yigit
---
v6:
* Removed tabs and newline in the description of the
seem
to have an older version of the PF driver as default.
Signed-off-by: Tudor Cornea
---
drivers/net/ixgbe/ixgbe_ethdev.c | 5 +++--
drivers/net/ixgbe/ixgbe_rxtx.c | 1 -
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe
dk.org/api/rte__ethdev_8h.html
Thanks,
Tudor
On Fri, 15 Oct 2021 at 17:06, Tudor Cornea wrote:
> It seems that if the call to ixgbevf_rlpml_set_vf fails,
> we will not initialize dev_conf.rxmode.max_rx_pkt_len correctly anymore.
>
> This happens with a 82599EB NIC and a VMware ESXI 6.0 setup,
> an
> > -Original Message-
> > From: Yigit, Ferruh
> > Sent: Tuesday, October 19, 2021 8:58 PM
> > To: Tudor Cornea ; Zhang, Qi Z
> >
> > Cc: Zhang, AlvinX ; Wang, Haiyue
> > ; dev@dpdk.org
> > Subject: Re: [dpdk-dev] [PATCH] net/ixgbe: initi
consistent with the overall
kernel ixgbevf implementation.
[1]
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c#n2015
Signed-off-by: Tudor Cornea
---
v2:
* Change title
* Remove max_rx_pkt_len fix in ixgbe_ethdev.c
It
On Thu, 21 Oct 2021 at 18:33, Ferruh Yigit wrote:
> On 10/20/2021 7:13 PM, Tudor Cornea wrote:
> > On a VMware ESXi 6.0 setup with an Intel 82599 NIC the ports don't
> > seem to initialize anymore, while running testpmd.
> >
> > Configuring Port 0 (socket 0)
&g
[] watchdog_timer_fn+0x160/0x160
[] __run_hrtimer.isra.4+0x42/0xd0
[] hrtimer_interrupt+0xe7/0x1f0
[] smp_apic_timer_interrupt+0x67/0xa0
[] apic_timer_interrupt+0x6d/0x80
References:
[1] https://www.kernel.org/doc/Documentation/timers/timers-howto.txt
Signed-off-by: Tudor Cornea
---
doc/guides
On Tue, 26 Oct 2021 at 17:41, Ferruh Yigit wrote:
> Hi Tudor,
>
> I have used testpmd, 'txonly' forwarding. Tx recovers after interface up,
> but by adding some debug logs I can see 'poll()' returns with POLLOUT even
> there is no space in the buffer.
>
> According the logic in the PMD, when 'pol
consistent, and does not recover.
ifconfig eth1 down; sleep 7; ifconfig eth1 up
[1] http://code.dpdk.org/pktgen-dpdk/pktgen-20.11.2/source/INSTALL.md
Signed-off-by: Mihai Pogonaru
Signed-off-by: Tudor Cornea
---
v2:
* Added check for POLLERR
* Used tx_ring_status_available() fo
[] watchdog_timer_fn+0x160/0x160
[] __run_hrtimer.isra.4+0x42/0xd0
[] hrtimer_interrupt+0xe7/0x1f0
[] smp_apic_timer_interrupt+0x67/0xa0
[] apic_timer_interrupt+0x6d/0x80
References:
[1] https://www.kernel.org/doc/Documentation/timers/timers-howto.txt
Signed-off-by: Tudor Cornea
---
v2:
* Fixed some
consistent, and does not recover.
ifconfig eth1 down; sleep 7; ifconfig eth1 up
[1] http://code.dpdk.org/pktgen-dpdk/pktgen-20.11.2/source/INSTALL.md
Signed-off-by: Mihai Pogonaru
Signed-off-by: Tudor Cornea
Reviewed-by: Ferruh Yigit
---
v4:
* Updated comment regarding POLLERR
Removed
On Tue, 2 Nov 2021 at 17:53, Stephen Hemminger
wrote:
> On Tue, 2 Nov 2021 17:51:13 +0200
> Tudor Cornea wrote:
>
> > +#ifdef RTE_KNI_PREEMPT_DEFAULT
> > +module_param(min_scheduling_interval, long, 0644);
> > +MODULE_PARM_DESC(min_scheduling_interval,
> > +&q
+0x6d/0x80
References:
[1] https://www.kernel.org/doc/Documentation/timers/timers-howto.txt
Signed-off-by: Tudor Cornea
---
v3:
* Fixed unwrapped commit description warning
* Changed from hrtimers to Linux High Precision Timers in docs
* Added two tabs at the beginning of the new params
The poll call can return POLLERR which is ignored, or it can return
POLLOUT, even if there are no free frames in the mmap-ed area.
We can account for both of these cases by re-checking if the next
frame is empty before writing into it.
Signed-off-by: Mihai Pogonaru
Signed-off-by: Tudor Cornea
ed. This allows
compatibility with non-timestamp aware user-space code which checks
tp_status == TP_STATUS_AVAILABLE; not expecting additional timestamp
flags to be set in tp_status.
[1] https://www.spinics.net/lists/kernel/msg3959391.html
Signed-off-by: Mihai Pogonaru
Signed-off-by: Tudor Cornea
-
ed. This allows
compatibility with non-timestamp aware user-space code which checks
tp_status == TP_STATUS_AVAILABLE; not expecting additional timestamp
flags to be set in tp_status.
[1] https://www.spinics.net/lists/kernel/msg3959391.html
Signed-off-by: Mihai Pogonaru
Signed-off-by: Tudor Cornea
---
v2:
*
Thanks for the observation.
I have removed the compile-time kernel version check in v2 of the patch
On Mon, 13 Sept 2021 at 18:09, Stephen Hemminger
wrote:
> On Mon, 13 Sep 2021 17:09:11 +0300
> Tudor Cornea wrote:
>
> > +static inline bool tx_ring_status_unavailable(uin
Thanks, Ferruh
I will perform the suggested recommendations in version 3 of the patch.
On Mon, 20 Sept 2021 at 18:41, Ferruh Yigit wrote:
> On 9/8/2021 9:59 AM, Tudor Cornea wrote:
> > The af_packet pmd driver binds to a raw socket and allows
> > sending and receiving of pack
Thanks for the suggestion. I will send a new version of the patch with the
required changes.
Tudor
On Mon, 20 Sept 2021 at 20:49, Ferruh Yigit wrote:
> On 9/13/2021 6:23 PM, Tudor Cornea wrote:
> > We should eliminate the timestamp status from the packet
> > status. This should
ed. This allows
compatibility with non-timestamp aware user-space code which checks
tp_status == TP_STATUS_AVAILABLE; not expecting additional timestamp
flags to be set in tp_status.
[1] https://www.spinics.net/lists/kernel/msg3959391.html
Signed-off-by: Mihai Pogonaru
Signed-off-by: Tudor Cornea
---
v3:
*
terms of packet
rate
[1]
https://github.com/torvalds/linux/commit/bcc6d47903612c3861201cc3a866fb604f26b8b2
Signed-off-by: Tudor Cornea
---
v3:
* Updated release note and documentation
* Updated commit with performance measurements
v2:
* Added DEV_RX_OFFLOAD_VLAN_STRIP to rxmode->offloads
--
older kernel.
Setting the socket option only if it is specifically requested from
the EAL arguments, allows us to have a way to run the PMD compiled
against newer kernel headers, on platforms having older kernels.
Signed-off-by: Tudor Cornea
---
drivers/net/af_packet/rte_eth_af_packet.c | 16
s 5, the computation mentioned
above could potentially result in a value which might be bigger than MAX_USHORT.
The result will be that Jumbo Frames will not work properly
A similar fix was submitted for the ice driver
Signed-off-by: Tudor Cornea
---
drivers/net/iavf/iavf_ethdev.c | 7 ---
Hi Ferruh,
> As I tested both with KNI sample app and KNI PMD, change looks good,
> practically we can't change the scheduler delay with existing code but
> this patch enables it and lets configure performance/CPU usage trade of.
>
> Only possible change is to remove 'RTE_KNI_PREEMPT_DEFAULT' ma
+0x6d/0x80
This patch also attempts to remove this option.
References:
[1] https://www.kernel.org/doc/Documentation/timers/timers-howto.txt
Signed-off-by: Tudor Cornea
---
v4:
* Removed RTE_KNI_PREEMPT_DEFAULT configuration option
v3:
* Fixed unwrapped commit description warning
* Changed from
to returning NETDEV_TX_BUSY seems
to add some extra overhead, and degrade the control-plane performance
a bit.
Signed-off-by: Tudor Cornea
---
kernel/linux/kni/kni_net.c | 9 ++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/kernel/linux/kni/kni_net.c b/kernel/linux/kni
edge (probably excepting patches that are back-ported by Ubuntu
maintainers from newer releases).
On Mon, 20 Sept 2021 at 20:44, Ferruh Yigit wrote:
> On 9/13/2021 2:45 PM, Tudor Cornea wrote:
> > The poll call can return POLLERR which is ignored, or it can return
> > POLLOUT, even if th
terms
of packet rate.
[1]
https://github.com/torvalds/linux/commit/bcc6d47903612c3861201cc3a866fb604f26b8b2
Signed-off-by: Tudor Cornea
---
v4:
* Updated the af_packet documentation
v3:
* Updated release note and documentation
* Updated commit with performance measurements
v2:
* Added DEV_RX_OFF
Thanks Stephen, for the suggestion
I've sent v4 of the patch, which adds the succinct description in the
af_packet documentation..
On Fri, 24 Sept 2021 at 18:11, Stephen Hemminger
wrote:
> On Fri, 24 Sep 2021 14:44:45 +0300
> Tudor Cornea wrote:
>
> > +Features an
terms
of packet rate.
[1]
https://github.com/torvalds/linux/commit/bcc6d47903612c3861201cc3a866fb604f26b8b2
Signed-off-by: Tudor Cornea
---
v5:
* Updated the af_packet documentation
* Updated the af_packet release notes
v4:
* Updated the af_packet documentation
v3:
* Updated release note an
ruh Yigit wrote:
> On 9/29/2021 3:08 PM, Tudor Cornea wrote:
> > The af_packet pmd driver binds to a raw socket and allows
> > sending and receiving of packets through the kernel.
> >
> > Since commit [1], the kernel strips the vlan tags early in
> > __netif_receiv
-n 4 \
--proc-type=primary \
--no-pci \
--no-telemetry \
--no-huge \
-m 512 \
--vdev=net_af_packet0,iface=eth1,blocksz=16384,framesz=8192,framecnt=2048,qpairs=1,qdisc_bypass=0
\
-- \
-P \
-T \
-m "3.0" \
-f themes/black-yellow.theme
se
thing similar was done for the netvsc pmd.
[1]
https://github.com/torvalds/linux/commit/bcc6d47903612c3861201cc3a866fb604f26b8b2
Signed-off-by: Tudor Cornea
---
drivers/net/af_packet/rte_eth_af_packet.c | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/net/af_packet/rte_eth_af_packet
status.
Signed-off-by: Mihai Pogonaru
Signed-off-by: Tudor Cornea
---
drivers/net/af_packet/rte_eth_af_packet.c | 47 +--
1 file changed, 45 insertions(+), 2 deletions(-)
diff --git a/drivers/net/af_packet/rte_eth_af_packet.c
b/drivers/net/af_packet
effect of vlan stripping for the af_packet PMD, in an uniform way, similar
to other PMDs.
Would this be considered an acceptable solution ?
On Tue, 31 Aug 2021 at 18:31, Ferruh Yigit wrote:
> On 8/20/2021 1:46 PM, Tudor Cornea wrote:
> > The af_packet pmd driver binds to a raw sock
rate at first
glance, but maybe the experiment could be repeated on different setups as
I'm using a virtual environment.
Would it be reasonable if I send v2 of the patch for review, with the above
changes ?
On Thu, 2 Sept 2021 at 13:49, Ferruh Yigit wrote:
> On 9/1/2021 10:34 PM, Stephen He
s from an old config, but
only in the first 2048 packets (which is the queue size)
[1] https://www.spinics.net/lists/kernel/msg3959391.html
[2] https://www.spinics.net/lists/netdev/msg739372.html
On Wed, 1 Sept 2021 at 19:34, Ferruh Yigit wrote:
> On 8/20/2021 2:39 PM, Tudor Cornea wrote:
&
/commit/bcc6d47903612c3861201cc3a866fb604f26b8b2
Signed-off-by: Tudor Cornea
---
v2:
* Add DEV_RX_OFFLOAD_VLAN_STRIP to rxmode->offloads
---
drivers/net/af_packet/rte_eth_af_packet.c | 12
1 file changed, 12 insertions(+)
diff --git a/drivers/net/af_packet/rte_eth_af_packet.c
to have a large value (i.e 13312),
and ICE_SUPPORT_CHAIN_NUM is defined as 5, the computation mentioned
above could potentially result in a value which might be bigger than
MAX_USHORT.
The result will be that Jumbo Frames will not work properly
Signed-off-by: Tudor Cornea
---
drivers/net/
the computation mentioned
above could potentially result in a value which might be bigger than
MAX_USHORT.
The result will be that Jumbo Frames will not work properly
Signed-off-by: Tudor Cornea
---
drivers/net/ice/ice_dcf_ethdev.c | 7 ---
drivers/net/ice/ice_rxtx.c | 6 +++---
2 files chan
This allows us to control the algorithm used to spread traffic between
sockets, adding more fine grained control. If the user does not
specify a fanout mode, the PMD driver will default to
PACKET_FANOUT_HASH.
Signed-off-by: Tudor Cornea
---
v2:
* Renamed the patch
* Replaced packet_fanout
> This is a great idea. Would introducing a new devarg, (e.g
> 'fanout_mode') be the proper way to allow the application to customize
> fanout in more detail ?
>
> --vdev=net_af_packet0,iface=eth1,blocksz=4096,framesz=2048,framecnt=512,qpairs=1,fanout_mode=[fanout_hash|fanout_cpu|fanout_rnd|fanout_
pmd> port start 0
Port 0: CA:65:81:63:81:B2
Checking link statuses...
Done
Fixes: 364e08f2bbc0 ("af_packet: add PMD for AF_PACKET-based virtual devices")
Signed-off-by: Tudor Cornea
---
drivers/net/af_packet/rte_eth_af_packet.c | 30 +++
1 file changed, 15
> Applied to next-net
> Should this go to stable as well
I think it would probably make sense to have it in stable.
This allows us to control whether the PMD will attempt to use
the PACKET_FANOUT socket option, and allows the binary compiled
against newer kernel headers to run on an older kernel, which
lacks support for it.
Signed-off-by: Tudor Cornea
---
doc/guides/nics/af_packet.rst | 4
> Controlling fanout more is a good idea but not sure what this patch
> is trying to do with it.
I am maintaining a DPDK application which should run on a large number
of setups.
Unfortunately, I do not have a lot of control over the environment the
application runs on (e.g kernel).
The problem I
This allows us to control the algorithm used to spread traffic between
sockets, adding more fine grained control. If the user does not
specify a fanout mode, the PMD driver will default to
PACKET_FANOUT_HASH.
Signed-off-by: Tudor Cornea
---
v3:
* Removed PACKET_FANOUT ifdef. The feature has
50 matches
Mail list logo
