> Subject: Re: [PATCH] net: phylink: set the autoneg state in
> phylink_phy_change
>
> On Thu, Jun 13, 2019 at 02:32:06PM +, Ioana Ciornei wrote:
> >
> > > Subject: Re: [PATCH] net: phylink: set the autoneg state in
> > > phylink_phy_change
> > >
> > > On Thu, Jun 13, 2019 at 08:55:16AM +,
Current flower mask creating code assumes that temporary mask that is used
when inserting new filter is stack allocated. To prevent race condition
with data patch synchronize_rcu() is called every time fl_create_new_mask()
replaces temporary stack allocated mask. As reported by Jiri, this
increases
On Thu, 13 Jun 2019 at 16:11, Björn Töpel wrote:
>
>
> On 2019-06-13 16:01, Maxim Mikityanskiy wrote:
> > On 2019-06-13 15:58, Björn Töpel wrote:
> >> On Wed, 12 Jun 2019 at 22:49, Jakub Kicinski
> >> wrote:
> >>>
> >>> On Wed, 12 Jun 2019 15:56:33 +, Maxim Mikityanskiy wrote:
> UAPI is
On Thu, Jun 13, 2019 at 03:52:27AM +, Sunil Muthuswamy wrote:
> The current vsock code for removal of socket from the list is both
> subject to race and inefficient. It takes the lock, checks whether
> the socket is in the list, drops the lock and if the socket was on the
> list, deletes it fro
On Thu, 13 Jun 2019 14:01:39 +
Maxim Mikityanskiy wrote:
> On 2019-06-12 23:23, Jakub Kicinski wrote:
> > On Wed, 12 Jun 2019 15:56:48 +, Maxim Mikityanskiy wrote:
> >> Currently, libbpf uses the number of combined channels as the maximum
> >> queue number. However, the kernel has a dif
On Thu, Jun 13, 2019 at 02:32:06PM +, Ioana Ciornei wrote:
>
> > Subject: Re: [PATCH] net: phylink: set the autoneg state in
> > phylink_phy_change
> >
> > On Thu, Jun 13, 2019 at 08:55:16AM +, Ioana Ciornei wrote:
> > > > Subject: Re: [PATCH] net: phylink: set the autoneg state in
> > >
> This whole discussion has zero to do with what text format 'ss' outputs.
Would you like to support the clarification of additional data formats
for the usage of specific information by tools for socket statistics
(from the directory “/proc/net” and related ones)?
Regards,
Markus
> Subject: Re: [PATCH] net: phylink: set the autoneg state in
> phylink_phy_change
>
> On Thu, Jun 13, 2019 at 08:55:16AM +, Ioana Ciornei wrote:
> > > Subject: Re: [PATCH] net: phylink: set the autoneg state in
> > > phylink_phy_change
> > >
> > > On Thu, Jun 13, 2019 at 09:37:51AM +0300, I
in the case of IPoIB with SRIOV enabled hardware
ip link show command incorrecly prints
0 instead of a VF hardware address. To correcly print the address
add a new field to specify an address length.
Before:
11: ib1: mtu 2044 qdisc pfifo_fast
state UP mode DEFAULT group default qlen 256
link/
in IPoIB case we can't see a VF broadcast address for but
can see for PF
Before:
11: ib1: mtu 2044 qdisc pfifo_fast
state UP mode DEFAULT group default qlen 256
link/infiniband
80:00:00:66:fe:80:00:00:00:00:00:00:24:8a:07:03:00:a4:3e:7c brd
00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:f
in the case of IPoIB with SRIOV enabled hardware
ip link show command incorrecly prints
0 instead of a VF hardware address. To correcly print the address
add a new field to specify an address length.
Before:
11: ib1: mtu 2044 qdisc pfifo_fast
state UP mode DEFAULT group default qlen 256
link/
in IPoIB case we can't see a VF broadcast address for but
can see for PF
Before:
11: ib1: mtu 2044 qdisc pfifo_fast
state UP mode DEFAULT group default qlen 256
link/infiniband
80:00:00:66:fe:80:00:00:00:00:00:00:24:8a:07:03:00:a4:3e:7c brd
00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:f
Thu, Jun 13, 2019 at 01:26:17PM CEST, vla...@mellanox.com wrote:
>
>On Thu 13 Jun 2019 at 14:11, Jiri Pirko wrote:
>> Thu, Jun 13, 2019 at 12:09:32PM CEST, vla...@mellanox.com wrote:
>>>On Thu 13 Jun 2019 at 11:11, Jiri Pirko wrote:
I made a mistake during measurements, sorry about that.
>>>
On 2019-06-13 16:01, Maxim Mikityanskiy wrote:
On 2019-06-13 15:58, Björn Töpel wrote:
On Wed, 12 Jun 2019 at 22:49, Jakub Kicinski
wrote:
On Wed, 12 Jun 2019 15:56:33 +, Maxim Mikityanskiy wrote:
UAPI is not changed, XSK RX queues are exposed to the kernel. The lower
half of the avail
Paul Blakey writes:
> On 6/11/2019 11:23 PM, Toke Høiland-Jørgensen wrote:
>> Marcelo Ricardo Leitner writes:
>>
>>> On Tue, Jun 11, 2019 at 05:34:50PM +0200, Toke Høiland-Jørgensen wrote:
Paul Blakey writes:
> On 6/11/2019 4:59 PM, Toke Høiland-Jørgensen wrote:
>> Paul Blakey
On 2019-06-13 15:58, Björn Töpel wrote:
> On Wed, 12 Jun 2019 at 22:49, Jakub Kicinski
> wrote:
>>
>> On Wed, 12 Jun 2019 15:56:33 +, Maxim Mikityanskiy wrote:
>>> UAPI is not changed, XSK RX queues are exposed to the kernel. The lower
>>> half of the available amount of RX queues are regular
On 2019-06-12 23:10, Jakub Kicinski wrote:
> On Wed, 12 Jun 2019 15:56:43 +, Maxim Mikityanskiy wrote:
>> The typical XDP memory scheme is one packet per page. Change the AF_XDP
>> frame size in libbpf to 4096, which is the page size on x86, to allow
>> libbpf to be used with the drivers with t
On 2019-06-12 23:23, Jakub Kicinski wrote:
> On Wed, 12 Jun 2019 15:56:48 +, Maxim Mikityanskiy wrote:
>> Currently, libbpf uses the number of combined channels as the maximum
>> queue number. However, the kernel has a different limitation:
>>
>> - xdp_reg_umem_at_qid() allows up to max(RX queu
On 2019-06-12 23:48, Jakub Kicinski wrote:
> On Wed, 12 Jun 2019 15:56:33 +, Maxim Mikityanskiy wrote:
>> UAPI is not changed, XSK RX queues are exposed to the kernel. The lower
>> half of the available amount of RX queues are regular queues, and the
>> upper half are XSK RX queues.
>
> If I h
On Thu, Jun 13, 2019 at 2:20 PM Davide Caratti wrote:
>
> hello John!
>
> On Wed, 2019-06-12 at 13:51 +0100, John Hurley wrote:
> > Currently, TC offers the ability to match on the MPLS fields of a packet
> > through the use of the flow_dissector_key_mpls struct. However, as yet, TC
> > actions do
On 6/12/2019 10:55 PM, Florian Westphal wrote:
> Paul Blakey wrote:
>> +/* The conntrack module expects to be working at L3. */
> It also expects that IP stack has validated ip(v6)
> headers and has pulled the ip header into linear area.
>
> What are your plans wrt. IP fragments? AFAICS right
On 6/12/2019 9:04 PM, Marcelo Ricardo Leitner wrote:
> On Tue, Jun 11, 2019 at 04:28:31PM +0300, Paul Blakey wrote:
>> Allow sending a packet to conntrack and set conntrack zone, mark,
>> labels and nat parameters.
>>
>> Signed-off-by: Paul Blakey
>> Signed-off-by: Marcelo Ricardo Leitner
>> Sig
On 6/13/2019 1:49 PM, Davide Caratti wrote:
> On Thu, 2019-06-13 at 12:16 +0200, Davide Caratti wrote:
>> hello Paul!
>>
>> On Tue, 2019-06-11 at 16:28 +0300, Paul Blakey wrote:
>>
>>> +#endif /* __NET_TC_CT_H */
>>> diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h
>>> inde
On Wed, Jun 12, 2019 at 7:03 PM Marcelo Ricardo Leitner
wrote:
>
> On Wed, Jun 12, 2019 at 01:51:44PM +0100, John Hurley wrote:
> > Currently, TC offers the ability to match on the MPLS fields of a packet
> > through the use of the flow_dissector_key_mpls struct. However, as yet, TC
> > actions do
On 6/11/2019 11:23 PM, Toke Høiland-Jørgensen wrote:
> Marcelo Ricardo Leitner writes:
>
>> On Tue, Jun 11, 2019 at 05:34:50PM +0200, Toke Høiland-Jørgensen wrote:
>>> Paul Blakey writes:
>>>
On 6/11/2019 4:59 PM, Toke Høiland-Jørgensen wrote:
> Paul Blakey writes:
>
>> Allow s
Hi Reinhard,
Il giorno mer 12 giu 2019 alle ore 19:02 Reinhard Speyerer
ha scritto:
>
> This series addresses the following issues observed when using the
> QMAP support of the qmi_wwan driver:
>
> 1. The QMAP code in the qmi_wwan driver is based on the CodeAurora
>GobiNet driver ([1], [2]) w
Hi Rob,
On Mon, Jun 10, 2019 at 12:59:29PM -0600, Rob Herring wrote:
> On Mon, Jun 10, 2019 at 8:31 AM Andrew Lunn wrote:
> >
> > > +required:
> > > + - compatible
> > > + - reg
> > > + - interrupts
> > > + - clocks
> > > + - phy
> > > + - allwinner,sram
> >
> > Quoting ethernet.txt:
> >
>
In preparation for using R12 indexing instructions in BPF JIT code, add
support for generating the x86 SIB byte.
Signed-off-by: Josh Poimboeuf
---
arch/x86/net/bpf_jit_comp.c | 69 +
1 file changed, 54 insertions(+), 15 deletions(-)
diff --git a/arch/x86/net/
hello John!
On Wed, 2019-06-12 at 13:51 +0100, John Hurley wrote:
> Currently, TC offers the ability to match on the MPLS fields of a packet
> through the use of the flow_dissector_key_mpls struct. However, as yet, TC
> actions do not allow the modification or manipulation of such fields.
>
> Add
On Thu, Jun 13, 2019 at 03:28:04PM +0530, Raju Lakkaraju wrote:
> Hi Andrew,
>
> Look like these patches are not create from "net-next" branch.
Correct. As the cover note says, they are dependent on the work by
Michal Kubecek implementing ethtool-nl. He is currently using an older
tree, or at lea
On Wed, 12 Jun 2019 at 22:49, Jakub Kicinski
wrote:
>
> On Wed, 12 Jun 2019 15:56:33 +, Maxim Mikityanskiy wrote:
> > UAPI is not changed, XSK RX queues are exposed to the kernel. The lower
> > half of the available amount of RX queues are regular queues, and the
> > upper half are XSK RX queu
On Thu, Jun 13, 2019 at 01:18:52PM +0200, Kevin Darbyshire-Bryant wrote:
> Use extack error reporting mechanism in addition to returning -EINVAL
>
> Signed-off-by: Kevin Darbyshire-Bryant
Nice. LGTM!
> ---
> net/sched/act_ctinfo.c | 23 ++-
> 1 file changed, 18 insertions(+
On Wed, 12 Jun 2019 at 20:05, Maxim Mikityanskiy wrote:
>
> Query XDP_OPTIONS in libbpf to determine if the zero-copy mode is active
> or not.
>
Acked-by: Björn Töpel
> Signed-off-by: Maxim Mikityanskiy
> Reviewed-by: Tariq Toukan
> Acked-by: Saeed Mahameed
> ---
> tools/lib/bpf/xsk.c | 12
On Wed, 12 Jun 2019 at 20:05, Maxim Mikityanskiy wrote:
>
> Add a function that checks whether the Fill Ring has the specified
> amount of descriptors available. It will be useful for mlx5e that wants
> to check in advance, whether it can allocate a bulk of RX descriptors,
> to get the best perfor
On Wed, 12 Jun 2019 at 20:05, Maxim Mikityanskiy wrote:
>
> Make it possible for the application to determine whether the AF_XDP
> socket is running in zero-copy mode. To achieve this, add a new
> getsockopt option XDP_OPTIONS that returns flags. The only flag
> supported for now is the zero-copy
On Wed, 12 Jun 2019 at 20:05, Maxim Mikityanskiy wrote:
>
> Some drivers want to access the data transmitted in order to implement
> acceleration features of the NICs. It is also useful in AF_XDP TX flow.
>
> Change the xsk_umem_consume_tx API to return the whole xdp_desc, that
> contains the data
On Wed, 12 Jun 2019 at 22:24, Jakub Kicinski
wrote:
>
> On Wed, 12 Jun 2019 15:56:48 +, Maxim Mikityanskiy wrote:
> > Currently, libbpf uses the number of combined channels as the maximum
> > queue number. However, the kernel has a different limitation:
> >
> > - xdp_reg_umem_at_qid() allows u
On Thu, 2019-06-13 at 16:20 +0200, Denis Kirjanov wrote:
> in the case of IPoIB with SRIOV enabled hardware
> ip link show command incorrecly prints
> 0 instead of a VF hardware address. To correcly print the address
> add a new field to specify an address length.
>
> Before:
> 11: ib1: mtu 2044
On Thu, 13 Jun 2019 18:39:59 +0900
Toshiaki Makita wrote:
> XDP_TX is similar to XDP_REDIRECT as it essentially redirects packets to
> the device itself. XDP_REDIRECT has bulk transmit mechanism to avoid the
> heavy cost of indirect call but it also reduces lock acquisition on the
> destination d
On Thu, 13 Jun 2019 18:39:58 +0900
Toshiaki Makita wrote:
> This is introduced for admins to check what is happening on XDP_TX when
> bulk XDP_TX is in use, which will be first introduced in veth in next
> commit.
>
> v3:
> - Add act field to be in line with other XDP tracepoints.
Thanks
Acked
Pointer members of an object with static storage duration, if not
explicitly initialized, will be initialized to a NULL pointer. The
net namespace API checks if this pointer is not NULL before using it,
it are safe to remove the function.
Signed-off-by: Li RongQing
---
net/xfrm/xfrm_interface.c
On Thu 13 Jun 2019 at 14:11, Jiri Pirko wrote:
> Thu, Jun 13, 2019 at 12:09:32PM CEST, vla...@mellanox.com wrote:
>>On Thu 13 Jun 2019 at 11:11, Jiri Pirko wrote:
>>> I made a mistake during measurements, sorry about that.
>>>
>>> This is the correct script:
>>> -
Hi Andrew,
Finally, I'm able to figureout the branch details (kernel:
ff24e4980a68)and apply your patches.
Give me sometime to review your code and sent my comments.
The 06/12/2019 18:05, Andrew Lunn wrote:
> External E-Mail
>
>
> This patchset adds support for executing Ethernet PHY cable test
From: Toke Høiland-Jørgensen
The socket map uses a linked list instead of a bitmap to keep track of
which entries to flush. Do the same for devmap and cpumap, as this means we
don't have to care about the map index when enqueueing things into the
map (and so we can cache the map lookup).
Signed-
When using the bpf_redirect_map() helper to redirect packets from XDP, the eBPF
program cannot currently know whether the redirect will succeed, which makes it
impossible to gracefully handle errors. To properly fix this will probably
require deeper changes to the way TX resources are allocated, bu
Use extack error reporting mechanism in addition to returning -EINVAL
Signed-off-by: Kevin Darbyshire-Bryant
---
net/sched/act_ctinfo.c | 23 ++-
1 file changed, 18 insertions(+), 5 deletions(-)
diff --git a/net/sched/act_ctinfo.c b/net/sched/act_ctinfo.c
index e78b60e47c0f.
From: Toke Høiland-Jørgensen
We don't currently allow lookups into a devmap from eBPF, because the map
lookup returns a pointer directly to the dev->ifindex, which shouldn't be
modifiable from eBPF.
However, being able to do lookups in devmaps is useful to know (e.g.)
whether forwarding to a spe
From: Toke Høiland-Jørgensen
The bpf_redirect_map() helper used by XDP programs doesn't return any
indication of whether it can successfully redirect to the map index it was
given. Instead, BPF programs have to track this themselves, leading to
programs using duplicate maps to track which entries
Thu, Jun 13, 2019 at 12:09:32PM CEST, vla...@mellanox.com wrote:
>On Thu 13 Jun 2019 at 11:11, Jiri Pirko wrote:
>> I made a mistake during measurements, sorry about that.
>>
>> This is the correct script:
>> ---
>> #!/bin/bash
>>
> On 11 Jun 2019, at 16:15, Paul Blakey wrote:
>
>
> On 6/11/2019 4:59 PM, Toke Høiland-Jørgensen wrote:
>> Paul Blakey writes:
>>
>>> Allow sending a packet to conntrack and set conntrack zone, mark,
>>> labels and nat parameters.
>> How is this different from the newly merged ctinfo action
Andrii Nakryiko writes:
> On Tue, Jun 11, 2019 at 9:42 AM Toke Høiland-Jørgensen
> wrote:
>>
>> From: Toke Høiland-Jørgensen
>>
>> The socket map uses a linked list instead of a bitmap to keep track of
>> which entries to flush. Do the same for devmap and cpumap, as this means we
>> don't have
Kevin 'ldir' Darbyshire-Bryant writes:
>> On 13 Jun 2019, at 10:33, Simon Horman wrote:
>>
>> On Wed, Jun 12, 2019 at 11:46:27AM -0700, Jakub Kicinski wrote:
>>> On Wed, 12 Jun 2019 15:02:39 -0300, Marcelo Ricardo Leitner wrote:
On Tue, May 28, 2019 at 05:03:50PM +, Kevin 'ldir' Darbys
On Thu, 2019-06-13 at 12:16 +0200, Davide Caratti wrote:
> hello Paul!
>
> On Tue, 2019-06-11 at 16:28 +0300, Paul Blakey wrote:
>
> > +#endif /* __NET_TC_CT_H */
> > diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h
> > index a93680f..c5264d7 100644
> > --- a/include/uapi/
On Tue, Jun 11, 2019 at 10:29:56AM +0300, Nikolay Aleksandrov wrote:
>
> Have you considered using the flow dissector and doing something similar to
> the bonding ?
> It does a full flow dissect via skb_flow_dissect_flow_keys() and uses
> whatever headers
> it needs, but that will support any tu
> On 13 Jun 2019, at 11:09, Kevin 'ldir' Darbyshire-Bryant
> wrote:
>
>
> Warning: Not even compile tested! Am I heading in the right direction?
>
I think this is even better. Does it help? Clues before I send as proper
patch :-)
Subject: [PATCH] sched: act_ctinfo: use extack error r
Based on comments from Xin, even after fixes for our recent syzbot
report of cookie memory leaks, its possible to get a resend of an INIT
chunk which would lead to us leaking cookie memory.
To ensure that we don't leak cookie memory, free any previously
allocated cookie first.
Change notes
v1->v2
hello Paul!
On Tue, 2019-06-11 at 16:28 +0300, Paul Blakey wrote:
> +#endif /* __NET_TC_CT_H */
> diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h
> index a93680f..c5264d7 100644
> --- a/include/uapi/linux/pkt_cls.h
> +++ b/include/uapi/linux/pkt_cls.h
> @@ -83,6 +83,7 @@
On Thu 13 Jun 2019 at 11:11, Jiri Pirko wrote:
> I made a mistake during measurements, sorry about that.
>
> This is the correct script:
> ---
> #!/bin/bash
>
> dev=testdummy
> ip link add name $dev type dummy
> ip link set dev $d
Hi Andrew,
Look like these patches are not create from "net-next" branch.
Can you please share branch detail where i can apply patches and check the code
flow?
The 06/12/2019 18:05, Andrew Lunn wrote:
> External E-Mail
>
>
> This patchset adds support for executing Ethernet PHY cable tests and
Em Wed, 12 Jun 2019 17:25:39 -0700
"Srivatsa S. Bhat" escreveu:
> On 6/12/19 10:52 AM, Mauro Carvalho Chehab wrote:
> > Convert the PM documents to ReST, in order to allow them to
> > build with Sphinx.
> >
> > The conversion is actually:
> > - add blank lines and identation in order to identi
XDP_TX is similar to XDP_REDIRECT as it essentially redirects packets to
the device itself. XDP_REDIRECT has bulk transmit mechanism to avoid the
heavy cost of indirect call but it also reduces lock acquisition on the
destination device that needs locks like veth and tun.
XDP_TX does not use indir
This introduces bulk XDP_TX in veth.
Improves XDP_TX performance by approximately 9%. The detailed
explanation and performance numbers are shown in patch 2.
v2:
- Use stack for bulk queue instead of a global variable.
v3:
- Add act field to xdp_bulk_tx tracepoint to be in line with other XDP
tr
This is introduced for admins to check what is happening on XDP_TX when
bulk XDP_TX is in use, which will be first introduced in veth in next
commit.
v3:
- Add act field to be in line with other XDP tracepoints.
Signed-off-by: Toshiaki Makita
---
include/trace/events/xdp.h | 29
On Wed, Jun 12, 2019 at 01:56:27PM -0700, Jakub Kicinski wrote:
> On Wed, 12 Jun 2019 16:25:16 -0400, Willem de Bruijn wrote:
> > On Wed, Jun 12, 2019 at 3:59 PM Jakub Kicinski
> > wrote:
> > >
> > > On Wed, 12 Jun 2019 15:44:09 -0400, Willem de Bruijn wrote:
> > > > From: Willem de Bruijn
> >
On Thu, Jun 13, 2019 at 08:55:16AM +, Ioana Ciornei wrote:
> > Subject: Re: [PATCH] net: phylink: set the autoneg state in
> > phylink_phy_change
> >
> > On Thu, Jun 13, 2019 at 09:37:51AM +0300, Ioana Ciornei wrote:
> > > The phy_state field of phylink should carry only valid information
> >
Thu, Jun 13, 2019 at 04:54:04PM CEST, vla...@mellanox.com wrote:
>Current flower mask creating code assumes that temporary mask that is used
>when inserting new filter is stack allocated. To prevent race condition
>with data patch synchronize_rcu() is called every time fl_create_new_mask()
>replace
From: Weihang Li
This patch does following things:
1. add the keyword const before some variables which won't be modified
in functions.
2. changes some variables from signed to unsigned to avoid bitwise
operation on signed variables.
3. adds or removes initialization of some variables.
4. d
From: Yonglong Liu
In hns3_nic_init_irq(), if request irq fail at index i,
the function return directly without releasing irq resources
that already requested, and nowhere else will release them.
Signed-off-by: Yonglong Liu
Signed-off-by: Peng Li
Signed-off-by: Huazhong Tan
---
drivers/net/e
From: Shiju Jose
Presently the HNS driver enables the HNS H/W error interrupts after
the dev initialization is completed. However some exceptions such as
NCSI errors can occur when the network port driver is not loaded
and those errors required reporting to the BMC.
Therefore the firmware enabled
From: Weihang Li
This patch modifies print message of rx_q_search_miss from error to dfx to
prevent misleading users, because this interrupt may occur if we receive
packets during initialization of HNS3 driver.
Otherwise, this patch masks 28th bit of PPU_MPF_ABNORMAL_SRC2 which is now
meaningless
From: Yunsheng Lin
If hns3_nic_net_xmit does not return NETDEV_TX_BUSY when doing
a loopback selftest, the skb is not freed in hns3_clean_tx_ring
or hns3_nic_net_xmit, which causes skb not freed problem.
This patch fixes it by freeing skb when hns3_nic_net_xmit does
not return NETDEV_TX_OK.
Fix
> On 13 Jun 2019, at 10:33, Simon Horman wrote:
>
> On Wed, Jun 12, 2019 at 11:46:27AM -0700, Jakub Kicinski wrote:
>> On Wed, 12 Jun 2019 15:02:39 -0300, Marcelo Ricardo Leitner wrote:
>>> On Tue, May 28, 2019 at 05:03:50PM +, Kevin 'ldir' Darbyshire-Bryant
>>> wrote:
>>> ...
+static
On Wed, 12 Jun 2019 at 23:15, Jakub Kicinski
wrote:
>
> On Wed, 12 Jun 2019 16:14:18 +, Maxim Mikityanskiy wrote:
> > dev_change_xdp_fd doesn't perform any checks in case it uninstalls an
> > XDP program. It means that the driver's ndo_bpf can be called with
> > XDP_SETUP_PROG asking to set it
> Subject: Re: [PATCH] net: phylink: set the autoneg state in
> phylink_phy_change
>
> On Thu, Jun 13, 2019 at 09:37:51AM +0300, Ioana Ciornei wrote:
> > The phy_state field of phylink should carry only valid information
> > especially when this can be passed to the .mac_config callback.
> > Updat
The longstanding prohibition against using 0.0.0.0/8 dates back
to two issues with the early internet.
There was an interoperability problem with BSD 4.2 in 1984, fixed in
BSD 4.3 in 1986. BSD 4.2 has long since been retired.
Secondly, addresses of the form 0.x.y.z were initially defined only as
My talk's slides and video at netdev 0x13 about
"Potential IPv4 Unicast expansions" is up, here:
https://netdevconf.org/0x13/session.html?talk-ipv4-unicast-expansions
There are roughly 419 million IPv4 addresses that are unallocated and
unused in the 0, localhost, reserved future multicast, and
To remove rtnl lock dependency in tc filter update API when using clsact
Qdisc, set QDISC_CLASS_OPS_DOIT_UNLOCKED flag in clsact Qdisc_class_ops.
Clsact Qdisc ops don't require any modifications to be used without rtnl
lock on tc filter update path. Implementation never changes its q->block
and on
Use the doorbell recovery mechanism to register rdma related doorbells
that will be restored in case there is a doorbell overflow attention.
Signed-off-by: Ariel Elior
Signed-off-by: Michal Kalderon
---
drivers/infiniband/hw/qedr/qedr.h | 13 +-
drivers/infiniband/hw/qedr/verbs.c | 351 ++
This patch series used the doorbell overflow recovery mechanism
introduced in
commit 36907cd5cd72 ("qed: Add doorbell overflow recovery mechanism")
for rdma ( RoCE and iWARP )
rdma-core pull request #493
Changes from V2:
- Don't use long-lived kmap. Instead use user-trigger mmap for the
doorbel
This patch adds the iWARP specific doorbells to the doorbell
recovery mechanism
Signed-off-by: Ariel Elior
Signed-off-by: Michal Kalderon
---
drivers/infiniband/hw/qedr/qedr.h | 12 +++-
drivers/infiniband/hw/qedr/verbs.c | 38 +-
2 files changed, 44
Several casts were required around dpi_addr parameter in qed_rdma_if.h
This is an address on the doorbell bar and should therefore be marked
with __iomem.
Reported-by: Jason Gunthorpe
Signed-off-by: Michal Kalderon
---
drivers/infiniband/hw/qedr/main.c | 2 +-
drivers/infiniband/hw/qed
From: Willem de Bruijn
NAPI tx mode improves TCP behavior by enabling TCP small queues (TSQ).
TSQ reduces queuing ("bufferbloat") and burstiness.
Previous measurements have shown significant improvement for
TCP_STREAM style workloads. Such as those in commit 86a5df1495cc
("Merge branch 'virtio-n
This series contains a few small fixes related to iWARP.
Michal Kalderon (4):
qed: iWARP - Use READ_ONCE and smp_store_release to access ep->state
qed: iWARP - fix uninitialized callback
qed: iWARP - Fix tc for MPA ll2 connection
qed: iWARP - Fix default window size to be based on chip
The default window size is calculated for best performance based
on internal hw buffer sizes. The size differs between the
different chips and modes.
Fixes: 67b40dccc45f ("qed: Implement iWARP initialization, teardown and qp
operations")
Signed-off-by: Ariel Elior
Signed-off-by: Michal Kalderon
Fix uninitialized variable warning by static checker.
Fixes: ae3488ff37dc ("qed: Add ll2 connection for processing unaligned MPA
packets")
Signed-off-by: Ariel Elior
Signed-off-by: Michal Kalderon
---
drivers/net/ethernet/qlogic/qed/qed_iwarp.c | 1 +
1 file changed, 1 insertion(+)
diff --git
On Wed, Jun 12, 2019 at 11:46:27AM -0700, Jakub Kicinski wrote:
> On Wed, 12 Jun 2019 15:02:39 -0300, Marcelo Ricardo Leitner wrote:
> > On Tue, May 28, 2019 at 05:03:50PM +, Kevin 'ldir' Darbyshire-Bryant
> > wrote:
> > ...
> > > +static int tcf_ctinfo_init(struct net *net, struct nlattr *nla
The driver needs to assign a lossless traffic class for the MPA ll2
connection to ensure no packets are dropped when returning from the
driver as they will never be re-transmitted by the peer.
Fixes: ae3488ff37dc ("qed: Add ll2 connection for processing unaligned MPA
packets")
Signed-off-by: Arie
On 06/13/2019 09:06 AM, Vlad Buslov wrote:
> On Wed 12 Jun 2019 at 19:33, Daniel Borkmann wrote:
>> On 06/12/2019 09:14 AM, Vlad Buslov wrote:
>>> To remove rtnl lock dependency in tc filter update API when using ingress
>>> Qdisc, set QDISC_CLASS_OPS_DOIT_UNLOCKED flag in ingress Qdisc_class_ops.
Destroy QP waits for it's ep object state to be set to CLOSED
before proceeding. ep->state can be updated from a different
context. Add smp_store_release/READ_ONCE to synchronize.
Fixes: fc4c6065e661 ("qed: iWARP implement disconnect flows")
Signed-off-by: Ariel Elior
Signed-off-by: Michal Kalder
On Thu, Jun 13, 2019 at 09:37:51AM +0300, Ioana Ciornei wrote:
> The phy_state field of phylink should carry only valid information
> especially when this can be passed to the .mac_config callback.
> Update the an_enabled field with the autoneg state in the
> phylink_phy_change function.
an_enable
I made a mistake during measurements, sorry about that.
This is the correct script:
---
#!/bin/bash
dev=testdummy
ip link add name $dev type dummy
ip link set dev $dev up
tc qdisc add dev $dev ingress
tmp_file_name=$(date +"/tmp
Support for indicating interface name has an ip address in parallel
with specifying ip address when activating UDP bearer.
This liberates the user from keeping track of the current ip address
for each device.
Old command syntax:
$tipc bearer enable media udp name NAME localip IP
New command synta
> Subject: [PATCH] net: phylink: set the autoneg state in phylink_phy_change
>
> The phy_state field of phylink should carry only valid information especially
> when this can be passed to the .mac_config callback.
> Update the an_enabled field with the autoneg state in the phylink_phy_change
> fun
This commit adds using the need_sleep flag to the xdpsock sample
application. It is turned on by default as we think it is a feature
that seems to always produce a performance benefit, if the application
has been written taking advantage of it. It can be turned off in the
sample app by using the '-
This patch adds support for the need_wakeup feature of AF_XDP. If the
application has told the kernel that it might sleep using the new bind
flag XDP_USE_NEED_WAKEUP, the driver will then set this flag if it has
no more buffers on the NIC Rx ring and yield to the application. For
Tx, it will set th
This patch adds support for the need_wakeup feature of AF_XDP. If the
application has told the kernel that it might sleep using the new bind
flag XDP_USE_NEED_WAKEUP, the driver will then set this flag if it has
no more buffers on the NIC Rx ring and yield to the application. For
Tx, it will set th
This commit adds support for a new flag called need_wakeup in the
AF_XDP Tx and fill rings. When this flag is set, it means that the
application has to explicitly wake up the kernel Rx (for the bit in
the fill ring) or kernel Tx (for bit in the Tx ring) processing by
issuing a syscall. Poll() can w
This commit replaces ndo_xsk_async_xmit with ndo_xsk_wakeup. This new
ndo provides the same functionality as before but with the addition of
a new flags field that is used to specifiy if Rx, Tx or both should be
woken up. The previous ndo only woke up Tx, as implied by the
name. The i40e and ixgbe
This patch set adds support for a new flag called need_wakeup in the
AF_XDP Tx and fill rings. When this flag is set by the driver, it
means that the application has to explicitly wake up the kernel Rx
(for the bit in the fill ring) or kernel Tx (for bit in the Tx ring)
processing by issuing a sy
This commit adds support for the new need_wakeup flag in AF_XDP. The
xsk_socket__create function is updated to handle this and a new
function is introduced called xsk_ring_prod__needs_wakeup(). This
function can be used by the application to check if Rx and/or Tx
processing needs to be explicitly w
1 - 100 of 331 matches
Mail list logo
