On Mon, Sep 12, 2016 at 07:07:36PM +0300, Ram Amrani wrote:
> Adds a skeletal implementation of the qed* RoCE driver -
> basically the ability to communicate with the qede driver and
> receive notifications from it regarding various init/exit events.
>
> Signed-off-by: Rajesh Borundia
> Signed-off
On Mon, Sep 12, 2016 at 07:17:35PM +, Yuval Mintz wrote:
> >> +uint debug;
> >> +module_param(debug, uint, 0);
> >> +MODULE_PARM_DESC(debug, "Default debug msglevel");
>
> >Why are you adding this as a module parameter?
>
> I believe this is mostly to follow same line as qede which also defines
On Mon, Sep 12, 2016 at 07:26:23PM +0200, Hannes Frederic Sowa wrote:
> > Actually, I'm convinced I must be doing something wrong here. The setup
> > for the issue is quite trivial, someone would have tripped over it
> > already. The only condition is that one host has multiple interfaces
> > with
On 12/09/2016 19:07, Ram Amrani wrote:
> Allocate and setup RoCE resources, interrupts and completion queues.
> Adds device attributes.
>
> Signed-off-by: Rajesh Borundia
> Signed-off-by: Ram Amrani
> ---
> drivers/infiniband/hw/qedr/main.c | 408 +++-
> drivers/infiniban
On Mon, Sep 12, 2016 at 05:39:35PM +, Yuval Mintz wrote:
> >>> include/linux/qed/common_hsi.h | 1 +
> >>> include/linux/qed/qed_if.h | 9 +-
> >>> include/linux/qed/qed_ll2_if.h | 140 +
> >>> include/linux/qed/qed_roce_if.h
Add SWITCHDEV_OBJ_ID_IPV4_FIB support to the DSA layer.
Signed-off-by: John Crispin
---
Changes in V2
* rebase on latest net-next to fix compile errors
Changes in V3
* fix subject prefix. this needs to go into the next tree
Documentation/networking/dsa/dsa.txt | 18 +++
include/n
Add SWITCHDEV_OBJ_ID_IPV4_FIB support to the DSA layer.
Signed-off-by: John Crispin
---
Changes in V2
* rebase on latest net-next to fix compile errors
Documentation/networking/dsa/dsa.txt | 18 +++
include/net/dsa.h| 13 +++
net/dsa/slave.c
On Mon, Sep 12, 2016 at 8:00 PM, Alexander Duyck
wrote:
> On Mon, Sep 12, 2016 at 3:13 PM, John Fastabend
> wrote:
>> The BQL API does not reference the sk_buff nor does the driver need to
>> reference the sk_buff to calculate the length of a transmitted frame.
>> This patch removes an sk_buff re
On Mon, Sep 12, 2016 at 3:13 PM, John Fastabend
wrote:
> From: Alexei Starovoitov
>
> This patch adds initial support for XDP on e1000 driver. Note e1000
> driver does not support page recycling in general which could be
> added as a further improvement. However XDP_DROP case will recycle.
> XDP_
From: Javier Martinez Canillas
Date: Mon, 12 Sep 2016 10:03:31 -0400
> This trivial series is similar to [0] for net/ that you already merged, but
> for drivers/net. The patches replaces the open coding to check for a Kconfig
> symbol being built-in or module, with IS_ENABLED() macro that does th
On (09/13/16 04:42), Hannes Frederic Sowa wrote:
> > But a couple of unexpected things I noticed in linux: the link-local
> > prefix should have a prefixlen of /10 according to
> > http://www.iana.org/assignments/ipv6-address-space/ipv6-address-space.xhtml
> > but "ip -6 route show" lists this as
Sowmini Varadhan wrote:
> On Mon, Sep 12, 2016 at 1:26 PM, Hannes Frederic Sowa
> wrote:
>> Hello,
>>
>> On 12.09.2016 16:27, Andreas Hübner wrote:
>
>>>
>>> I have the following setup:
>>> - 2 directly connected hosts (A+B), both have only link local addresses
>>> configured (interface on
On Mon, Sep 12, 2016 at 3:13 PM, John Fastabend
wrote:
> The BQL API does not reference the sk_buff nor does the driver need to
> reference the sk_buff to calculate the length of a transmitted frame.
> This patch removes an sk_buff reference from the xmit irq path and
> also allows packets sent fr
Hi Ram,
Just a few thoughts
On 12/09/2016 19:07, Ram Amrani wrote:
> Adds a skeletal implementation of the qed* RoCE driver -
> basically the ability to communicate with the qede driver and
> receive notifications from it regarding various init/exit events.
>
> Signed-off-by: Rajesh Borundia
On 13.09.2016 04:03, Sowmini Varadhan wrote:
> On Mon, Sep 12, 2016 at 1:26 PM, Hannes Frederic Sowa
> wrote:
>> Hello,
>>
>> On 12.09.2016 16:27, Andreas Hübner wrote:
>
>>>
>>> I have the following setup:
>>> - 2 directly connected hosts (A+B), both have only link local addresses
>>> conf
On Mon, Sep 12, 2016 at 1:26 PM, Hannes Frederic Sowa
wrote:
> Hello,
>
> On 12.09.2016 16:27, Andreas Hübner wrote:
>>
>> I have the following setup:
>> - 2 directly connected hosts (A+B), both have only link local addresses
>> configured (interface on both hosts is eth0)
:
>>
On Mon, Sep 12, 2016 at 05:03:25PM -0700, Tom Herbert wrote:
> On Mon, Sep 12, 2016 at 4:46 PM, Eric Dumazet wrote:
> > On Mon, 2016-09-12 at 16:07 -0700, Alexei Starovoitov wrote:
> >
> >> yep. there are various ways to shoot yourself in the foot with xdp.
> >> The simplest program that drops all
> +static int
> +qca8k_setup(struct dsa_switch *ds)
> +{
> + struct qca8k_priv *priv = qca8k_to_priv(ds);
> + int ret, i, phy_mode = -1;
> +
> + /* Keep track of which port is the connected to the cpu. This can be
> + * phy 11 / port 0 or phy 5 / port 6.
> + */
> + switch
On Mon, Sep 12, 2016 at 04:46:08PM -0700, Eric Dumazet wrote:
> On Mon, 2016-09-12 at 16:07 -0700, Alexei Starovoitov wrote:
>
> > yep. there are various ways to shoot yourself in the foot with xdp.
> > The simplest program that drops all the packets will make the box
> > unpingable.
>
> Well, m
Hi John
> +static u16
> +qca8k_phy_mmd_read(struct qca8k_priv *priv, int phy_addr, u16 addr, u16 reg)
> +{
> + u16 data;
> +
> + mutex_lock(&priv->bus->mdio_lock);
> +
> + priv->bus->write(priv->bus, phy_addr, MII_ATH_MMD_ADDR, addr);
> + priv->bus->write(priv->bus, phy_addr, MII_A
From: Jamal Hadi Salim
This action is intended to be an upgrade from a usability perspective
from pedit (as well as operational debugability).
Compare this:
sudo tc filter add dev $ETH parent 1: protocol ip prio 10 \
u32 match ip protocol 1 0xff flowid 1:2 \
action pedit munge offset -14 u8 set
Hi all,
Today's linux-next merge of the netfilter-next tree got a conflict in:
net/netfilter/nf_tables_netdev.c
between commit:
c73c24849011 ("netfilter: nf_tables_netdev: remove redundant ip_hdr
assignment")
from the net tree and commit:
ddc8b6027ad0 ("netfilter: introduce nft_set_pkt
Ok, screwup. I used version 5 again ;-< I will resend and incorporate
Florian's comment.
cheers,
jamal
From: Jamal Hadi Salim
This action is intended to be an upgrade from a usability perspective
from pedit (as well as operational debugability).
Compare this:
sudo tc filter add dev $ETH parent 1: protocol ip prio 10 \
u32 match ip protocol 1 0xff flowid 1:2 \
action pedit munge offset -14 u8 set
On Mon, Sep 12, 2016 at 4:46 PM, Eric Dumazet wrote:
> On Mon, 2016-09-12 at 16:07 -0700, Alexei Starovoitov wrote:
>
>> yep. there are various ways to shoot yourself in the foot with xdp.
>> The simplest program that drops all the packets will make the box unpingable.
>
> Well, my comment was abo
Jamal Hadi Salim wrote:
> +#define MAX_EDIT_LEN ETH_HLEN
This define is only referenced in a comment and seems unused otherwise.
On Mon, 2016-09-12 at 19:56 -0400, Jamal Hadi Salim wrote:
>
> Yes, I did ;->
> Another version coming. Is there a good way to catch
> something like this (ala userspace valgrind)?
You might try CONFIG_DEBUG_KMEMLEAK=y,
or simply run your script deleting/inserting the action one million
time ;)
On 16-09-12 07:40 PM, Eric Dumazet wrote:
On Mon, 2016-09-12 at 19:30 -0400, Jamal Hadi Salim wrote:
From: Jamal Hadi Salim
+
+static struct tc_action_ops act_skbmod_ops = {
+ .kind = "skbmod",
+ .type = TCA_ACT_SKBMOD,
+ .owner =
On Mon, 2016-09-12 at 16:07 -0700, Alexei Starovoitov wrote:
> yep. there are various ways to shoot yourself in the foot with xdp.
> The simplest program that drops all the packets will make the box unpingable.
Well, my comment was about XDP_TX only, not about XDP_DROP or driving a
scooter on 101
On Mon, Sep 12, 2016 at 3:14 PM, John Fastabend
wrote:
> e1000 supports a single TX queue so it is being shared with the stack
> when XDP runs XDP_TX action. This requires taking the xmit lock to
> ensure we don't corrupt the tx ring. To avoid taking and dropping the
> lock per packet this patch a
On Mon, 2016-09-12 at 19:30 -0400, Jamal Hadi Salim wrote:
> From: Jamal Hadi Salim
> +
> +static struct tc_action_ops act_skbmod_ops = {
> + .kind = "skbmod",
> + .type = TCA_ACT_SKBMOD,
> + .owner = THIS_MODULE,
> + .act
On 16-09-12 07:20 PM, Eric Dumazet wrote:
On Mon, 2016-09-12 at 16:10 -0700, Eric Dumazet wrote:
diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c
index
508e051304fb62627e61b5065b2325edd1b84f2e..dc9dd8ae7d5405f76c775278dac7689655b21041
100644
--- a/net/core/gen_stats.c
+++ b/net/core/
From: Jamal Hadi Salim
This action is intended to be an upgrade from a usability perspective
from pedit (as well as operational debugability).
Compare this:
sudo tc filter add dev $ETH parent 1: protocol ip prio 10 \
u32 match ip protocol 1 0xff flowid 1:2 \
action pedit munge offset -14 u8 set
On 16-09-12 07:02 PM, Jamal Hadi Salim wrote:
Looks like typical starvation caused by aggressive softirq.
Well, then it is strange that in one case a tc dump of the rule
was immediate and in the other case it was consistent for 5-15
seconds.
I may be going nuts because i cant reproduce tha
On Mon, 2016-09-12 at 06:39 +, Y.B. Lu wrote:
> Hi Scott,
>
> Thanks for your review :)
> See my comment inline.
>
> >
> > -Original Message-
> > From: Scott Wood [mailto:o...@buserror.net]
> > Sent: Friday, September 09, 2016 11:47 AM
> > To: Y.B. Lu; linux-...@vger.kernel.org; ulf.
On Mon, 2016-09-12 at 16:10 -0700, Eric Dumazet wrote:
>
> diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c
> index
> 508e051304fb62627e61b5065b2325edd1b84f2e..dc9dd8ae7d5405f76c775278dac7689655b21041
> 100644
> --- a/net/core/gen_stats.c
> +++ b/net/core/gen_stats.c
> @@ -142,10 +142,1
On Mon, Sep 12, 2016 at 10:43:00PM +, Adit Ranadive wrote:
> On Mon, Sep 12, 2016 at 11:03:39 -0700, Jason Gunthorpe wrote:
> > On Sun, Sep 11, 2016 at 09:49:10PM -0700, Adit Ranadive wrote:
> > > [2] Libpvrdma User-level library -
> > > http://git.openfabrics.org/?p=~aditr/libpvrdma.git;a=sum
On Mon, 2016-09-12 at 19:02 -0400, Jamal Hadi Salim wrote:
> On 16-09-12 06:26 PM, Eric Dumazet wrote:
> > On Mon, 2016-09-12 at 18:14 -0400, Jamal Hadi Salim wrote:
> >
> >> I noticed some very weird issues when I took that out.
> >> Running sufficiently large amount of traffic (ping -f is suffici
On Mon, Sep 12, 2016 at 03:46:39PM -0700, Eric Dumazet wrote:
> On Mon, 2016-09-12 at 15:13 -0700, John Fastabend wrote:
> > From: Alexei Starovoitov
>
> > +static void e1000_xmit_raw_frame(struct e1000_rx_buffer *rx_buffer_info,
> > +u32 len,
> > +
From: Jamal Hadi Salim
With the batch changes that translated transient actions into
a temporary list lost in the translation was the fact that
tcf_action_destroy() will eventually delete the action from
the permanent location if the refcount is zero.
Example of what broke:
...add a gact action
On 16-09-12 06:26 PM, Eric Dumazet wrote:
On Mon, 2016-09-12 at 18:14 -0400, Jamal Hadi Salim wrote:
I noticed some very weird issues when I took that out.
Running sufficiently large amount of traffic (ping -f is sufficient)
I saw that when i did a dump it took anywhere between 6-15 seconds.
Wi
Hi John
> +#include
> +#include
> +#include
> +#include
> +#include
> +#include
One linux/phy.h is enough.
> + /* Setup connection between CPU ports & PHYs */
> + for (i = 0; i < DSA_MAX_PORTS; i++) {
> + /* CPU port gets connected to all PHYs in the switch */
> +
On Mon, 2016-09-12 at 15:13 -0700, John Fastabend wrote:
> From: Alexei Starovoitov
> +static void e1000_xmit_raw_frame(struct e1000_rx_buffer *rx_buffer_info,
> + u32 len,
> + struct net_device *netdev,
> + st
On Mon, Sep 12, 2016 at 11:03:39 -0700, Jason Gunthorpe wrote:
> On Sun, Sep 11, 2016 at 09:49:10PM -0700, Adit Ranadive wrote:
> > [2] Libpvrdma User-level library -
> > http://git.openfabrics.org/?p=~aditr/libpvrdma.git;a=summary
>
> You will probably find that rdma-plumbing will be the best wa
On Mon, 2016-09-12 at 18:14 -0400, Jamal Hadi Salim wrote:
> I noticed some very weird issues when I took that out.
> Running sufficiently large amount of traffic (ping -f is sufficient)
> I saw that when i did a dump it took anywhere between 6-15 seconds.
> With the read_lock in place response wa
On 16-09-12 06:01 PM, Eric Dumazet wrote:
On Mon, 2016-09-12 at 16:46 -0400, Jamal Hadi Salim wrote:
+
+static int tcf_skbmod_dump(struct sk_buff *skb, struct tc_action *a,
+ int bind, int ref)
+{
+ struct tcf_skbmod *d = to_skbmod(a);
+ unsigned char *b = s
e1000 supports a single TX queue so it is being shared with the stack
when XDP runs XDP_TX action. This requires taking the xmit lock to
ensure we don't corrupt the tx ring. To avoid taking and dropping the
lock per packet this patch adds a bundling implementation to submit
a bundle of packets to t
From: Alexei Starovoitov
This patch adds initial support for XDP on e1000 driver. Note e1000
driver does not support page recycling in general which could be
added as a further improvement. However XDP_DROP case will recycle.
XDP_TX and XDP_PASS do not support recycling.
e1000 only supports a si
On Mon, Sep 12, 2016 at 10:52:26 -0700, Jason Gunthorpe wrote:
> On Sun, Sep 11, 2016 at 09:49:26PM -0700, Adit Ranadive wrote:
> > Add maintainer info for the PVRDMA driver.
>
> You can probably squash the last three patches.
Thanks for taking a look.
Since Doug mentioned that he would squash m
The BQL API does not reference the sk_buff nor does the driver need to
reference the sk_buff to calculate the length of a transmitted frame.
This patch removes an sk_buff reference from the xmit irq path and
also allows packets sent from XDP to use BQL.
Signed-off-by: John Fastabend
---
drivers/
This patch implements XDP on e1000 a few comments below:
The XDP_TX path does not violate BQL in this series so we have to increment
and decrement the bql counters correctly. When a TX queue can have both stack
traffic and XDP traffic I believe user configured BQL should be correct. If
users do no
On 16-09-12 05:58 PM, Eric Dumazet wrote:
On Mon, 2016-09-12 at 16:46 -0400, Jamal Hadi Salim wrote:
From: Jamal Hadi Salim
+
+ /* XXX: if you are going to edit more fields beyond ethernet header
+* (example when you add IP header replacement or vlan swap)
+* then MAX_E
On Mon, 2016-09-12 at 16:46 -0400, Jamal Hadi Salim wrote:
> +
> +static int tcf_skbmod_dump(struct sk_buff *skb, struct tc_action *a,
> +int bind, int ref)
> +{
> + struct tcf_skbmod *d = to_skbmod(a);
> + unsigned char *b = skb_tail_pointer(skb);
> + struct tc
On Mon, 2016-09-12 at 16:46 -0400, Jamal Hadi Salim wrote:
> From: Jamal Hadi Salim
> +
> + /* XXX: if you are going to edit more fields beyond ethernet header
> + * (example when you add IP header replacement or vlan swap)
> + * then MAX_EDIT_LEN needs to change appropriately
> +
Hi everyone
08.09.2016, 18:39, "Alban Crequy" :
> The act of a process creating or joining a namespace via clone(),
> unshare() or setns() is a useful signal for monitoring applications.
> + if (old_ns->mnt_ns != new_ns->mnt_ns)
> + proc_ns_connector(tsk, CLONE_NEWNS, PROC_NM_REASON_CLONE, old_mn
On Mon, Sep 12, 2016 at 3:15 AM, Jesper Dangaard Brouer
wrote:
> On Fri, 9 Sep 2016 18:03:09 +0300
> Saeed Mahameed wrote:
>
>> On Fri, Sep 9, 2016 at 6:22 AM, Alexei Starovoitov via iovisor-dev
>> wrote:
>> > On Thu, Sep 08, 2016 at 10:11:47AM +0200, Jesper Dangaard Brouer wrote:
>> >>
>> >> I'
We have a small skb_at_tc_ingress() helper for testing for ingress, so
make use of it. cls_bpf already uses it and so should act_bpf.
Signed-off-by: Daniel Borkmann
Acked-by: Alexei Starovoitov
---
net/sched/act_bpf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/sched
The skb_mac_header_was_set() test in cls_bpf's and act_bpf's fast-path is
actually unnecessary and can be removed altogether. This was added by
commit a166151cbe33 ("bpf: fix bpf helpers to use skb->mac_header relative
offsets"), which was later on improved by 3431205e0397 ("bpf: make programs
see
Two minor improvements to {cls,act}_bpf. For details please see
individual patches.
Thanks!
Daniel Borkmann (2):
bpf: drop unnecessary test in cls_bpf_classify and tcf_bpf
bpf: use skb_at_tc_ingress helper in tcf_bpf
net/sched/act_bpf.c | 5 +
net/sched/cls_bpf.c | 3 ---
2 files change
Hi Matt,
On Sun, Sep 11, 2016 at 10:57 PM, Matt Corallo wrote:
> The general advice is to force it into 100M mode then it seems to work fine.
> There is some issue with the driver seemingly not fully supporting 1G (at
> least on Odroid C2) which needs to be worked out.
thanks for the hint - by "f
Hi Alexandre,
On Mon, Sep 12, 2016 at 6:37 PM, Alexandre Torgue
wrote:
> Which Synopsys IP version do you use ?
found this in a dmesg log:
[1.504784] stmmac - user ID: 0x11, Synopsys ID: 0x37
[1.509785] Ring mode enabled
[1.512796] DMA HW capability register supported
[1.517286]
On Mon, 12 Sep 2016, Jarkko Sakkinen wrote:
> On Mon, Sep 12, 2016 at 04:43:58PM +0300, Felipe Balbi wrote:
> >
> > Hi,
> >
> > Jarkko Sakkinen writes:
> > > On Mon, Sep 12, 2016 at 10:54:07AM +0200, Julia Lawall wrote:
> > >>
> > >>
> > >> On Sun, 11 Sep 2016, Jarkko Sakkinen wrote:
> > >>
> >
On 16-09-12 01:18 PM, Cong Wang wrote:
On Mon, Sep 12, 2016 at 3:21 AM, Jamal Hadi Salim wrote:
+ if (ovr)
+ a->tcfa_refcnt-=1;
How about tcfa_bindcnt?
We dont touch it when mucking around only with actions (as is in
this case).
+ }
+}
+
int
On Mon, 12 Sep 2016 12:56:28 -0700
Alexei Starovoitov wrote:
> On Mon, Sep 12, 2016 at 01:30:25PM +0200, Jesper Dangaard Brouer wrote:
> > On Thu, 8 Sep 2016 23:30:50 -0700
> > Alexei Starovoitov wrote:
> >
> > > On Fri, Sep 09, 2016 at 07:36:52AM +0200, Jesper Dangaard Brouer wrote:
> > [.
From: Jamal Hadi Salim
This action is intended to be an upgrade from a usability perspective
from pedit (as well as operational debugability).
Compare this:
sudo tc filter add dev $ETH parent 1: protocol ip prio 10 \
u32 match ip protocol 1 0xff flowid 1:2 \
action pedit munge offset -14 u8 set
Hi,
On Fri, Sep 09, 2016 at 02:45:17PM +0200, Corentin Labbe wrote:
> This patch add pm_runtime support to sun8i-emac.
> For the moment, only basic support is added, (the device is marked as
> used when net/open)
>
> Signed-off-by: Corentin Labbe
> ---
> drivers/net/ethernet/allwinner/sun8i-ema
On Mon, Sep 12, 2016 at 7:55 AM, wrote:
> From: Kan Liang
>
> When the device tries to transmit a packet, netdev_pick_tx is called to
> find the available tx queues. If the net policy is applied, it picks up
> the assigned tx queue from net policy subsystem, and redirect the
> traffic to the ass
On Mon, Sep 12, 2016 at 04:43:58PM +0300, Felipe Balbi wrote:
>
> Hi,
>
> Jarkko Sakkinen writes:
> > On Mon, Sep 12, 2016 at 10:54:07AM +0200, Julia Lawall wrote:
> >>
> >>
> >> On Sun, 11 Sep 2016, Jarkko Sakkinen wrote:
> >>
> >> > On Sun, Sep 11, 2016 at 03:05:42PM +0200, Julia Lawall wro
With centralized MTU checking, there's nothing productive done by
eth_change_mtu that isn't already done in dev_set_mtu, so mark it as
deprecated and remove all usage of it in the kernel. All callers have been
audited for calls to alloc_etherdev* or ether_setup directly, which means
they all have a
While looking into an MTU issue with sfc, I started noticing that almost
every NIC driver with an ndo_change_mtu function implemented almost
exactly the same range checks, and in many cases, that was the only
practical thing their ndo_change_mtu function was doing. Quite a few
drivers have either 6
While looking into an MTU issue with sfc, I started noticing that almost
every NIC driver with an ndo_change_mtu function implemented almost
exactly the same range checks, and in many cases, that was the only
practical thing their ndo_change_mtu function was doing. Quite a few
drivers have either 6
On Mon, Sep 12, 2016 at 3:26 PM, kbuild test robot wrote:
> Hi Willem,
>
> [auto build test ERROR on net-next/master]
>
> url:
> https://github.com/0day-ci/linux/commits/Willem-de-Bruijn/uapi-include-time-h-from-errqueue-h/20160913-020431
> config: i386-defconfig (attached as .config)
> compil
On Mon, Sep 12, 2016 at 01:30:25PM +0200, Jesper Dangaard Brouer wrote:
> On Thu, 8 Sep 2016 23:30:50 -0700
> Alexei Starovoitov wrote:
>
> > On Fri, Sep 09, 2016 at 07:36:52AM +0200, Jesper Dangaard Brouer wrote:
> [...]
> > > Imagine you have packets intermixed towards the stack and XDP_TX.
>
On Sat, Sep 10, 2016 at 12:03:53AM +0800, Xin Long wrote:
> > That said, have you considered the retransmit case? That is to say, if you
> > queue and flush the outq, and some packets fail delivery, and in the time
> > between the intial send and the expiration of the RTX timer (during which
> >
On Thu, Sep 08, 2016 at 05:31:47PM +0800, Xin Long wrote:
> Last patch "sctp: do not return the transmit err back to sctp_sendmsg"
> made sctp_primitive_SEND return err only when asoc state is unavailable.
> In this case, chunks are not enqueued, they have no chance to be freed if
> we don't take c
On Fri, Sep 9, 2016 at 1:41 PM, Thadeu Lima de Souza Cascardo
wrote:
> Instead of using flow stats per NUMA node, use it per CPU. When using
> megaflows, the stats lock can be a bottleneck in scalability.
>
> On a E5-2690 12-core system, usual throughput went from ~4Mpps to
> ~15Mpps when forwardi
Hi Willem,
[auto build test ERROR on net-next/master]
url:
https://github.com/0day-ci/linux/commits/Willem-de-Bruijn/uapi-include-time-h-from-errqueue-h/20160913-020431
config: i386-defconfig (attached as .config)
compiler: gcc-6 (Debian 6.1.1-9) 6.1.1 20160705
reproduce:
# save the a
On Mon, Sep 12, 2016 at 09:21:10AM -0700, Tom Herbert wrote:
> In order to use XDP with non multi-queue drivers (such as e1000) we need
> a method to handle XDP_TX return code from xdp program. Since there is
> only one queue the XDP transmit path needs to share that queue with the
> normal stack p
On 9/12/16 11:26 AM, Hannes Frederic Sowa wrote:
> Hello,
>
> On 12.09.2016 16:27, Andreas Hübner wrote:
>> Hi,
>>
>> I'm currently debugging a potential issue with the icmpv6 stack and
>> hopefully this is the correct place to ask. (Was actually looking for a
>> more specific list, but didn't fin
>> +uint debug;
>> +module_param(debug, uint, 0);
>> +MODULE_PARM_DESC(debug, "Default debug msglevel");
>Why are you adding this as a module parameter?
I believe this is mostly to follow same line as qede which also defines
'debug' module parameter for allowing easy user control of debug
prints
On Fri, Sep 09, 2016 at 02:33:58PM +0800, Jia He wrote:
> This is to use the generic interface snmp_get_cpu_field{,64}_batch to
> aggregate the data by going through all the items of each cpu sequentially.
>
> Signed-off-by: Jia He
> ---
> net/ipv6/proc.c | 32 +++-
>
On Fri, Sep 09, 2016 at 02:33:57PM +0800, Jia He wrote:
> This is to use the generic interface snmp_get_cpu_field{,64}_batch to
> aggregate the data by going through all the items of each cpu sequentially.
> Then snmp_seq_show and netstat_seq_show are split into 2 parts to avoid build
> warning "t
On Mon, Sep 12, 2016 at 03:52:08PM +0200, Julia Lawall wrote:
>
>
> On Mon, 12 Sep 2016, Felipe Balbi wrote:
>
> >
> > Hi,
> >
> > Jarkko Sakkinen writes:
> > > On Mon, Sep 12, 2016 at 10:54:07AM +0200, Julia Lawall wrote:
> > >>
> > >>
> > >> On Sun, 11 Sep 2016, Jarkko Sakkinen wrote:
> > >>
On Sat, Sep 10, 2016 at 11:11:23PM +0800, Xin Long wrote:
> Since commit 4f0087812648 ("sctp: apply rhashtable api to send/recv
> path"), sctp uses transport rhashtable with .obj_cmpfn sctp_hash_cmp,
> in which it compares the members of the transport with the rhashtable
> args to check if it's the
On 12/09/2016 16:00, Andrew Lunn wrote:
> Hi John
>
>> +if (!ds->drv->ipv4_fib_prepare || !ds->drv->ipv4_fib_add)
>
> drv recently got renamed to ops, which is what 0-day is complaining
> about.
>
> Andrew
>
i based this on a net-next from 5-6 days ago, will rebase/resend tomorrow.
Le 12/09/2016 à 16:35, David Ahern a écrit :
On 9/12/16 12:02 AM, Christophe JAILLET wrote:
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index abfbe492ebfe..795af25cf84c 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -1134,7 +1134,6 @@ int inet_diag_handler_get_info(
On 16-09-12 05:17 AM, Jesper Dangaard Brouer wrote:
> On Fri, 09 Sep 2016 14:29:38 -0700
> John Fastabend wrote:
>
>> e1000 supports a single TX queue so it is being shared with the stack
>> when XDP runs XDP_TX action. This requires taking the xmit lock to
>> ensure we don't corrupt the tx ring.
From: Mahesh Bandewar
In a typical IPvlan L3 setup where master is in default-ns and
each slave is into different (slave) ns. In this setup egress
packet processing for traffic originating from slave-ns will
hit all NF_HOOKs in slave-ns as well as default-ns. However same
is not true for ingress
On Mon, Sep 12, 2016 at 07:07:34PM +0300, Ram Amrani wrote:
> drivers/infiniband/hw/qedr/main.c | 907 ++
> drivers/infiniband/hw/qedr/qedr.h | 494
> drivers/infiniband/hw/qedr/qedr_cm.c | 626 +
> drivers/infiniband/hw/qedr/qedr_cm.h
From: Mahesh Bandewar
Add _nf_register_hooks() and _nf_unregister_hooks() calls which allow
caller to hold RTNL mutex.
Signed-off-by: Mahesh Bandewar
CC: Pablo Neira Ayuso
---
include/linux/netfilter.h | 2 ++
net/netfilter/core.c | 51 ++-
2
On Sun, Sep 11, 2016 at 09:49:10PM -0700, Adit Ranadive wrote:
> [2] Libpvrdma User-level library -
> http://git.openfabrics.org/?p=~aditr/libpvrdma.git;a=summary
You will probably find that rdma-plumbing will be the best way to get
your userspace component into the distributors.
http://www.sp
From: Mahesh Bandewar
Make ip6_route_input_lookup available outside of ipv6 the module
similar to ip_route_input_noref in the IPv4 world.
Signed-off-by: Mahesh Bandewar
---
include/net/ip6_route.h | 3 +++
net/ipv6/route.c| 7 ---
2 files changed, 7 insertions(+), 3 deletions(-)
d
From: Mahesh Bandewar
Same old problem with new approach especially from suggestions from
earlier patch-series.
First thing is that this is introduced as a new mode rather than
modifying the old (L3) mode. So the behavior of the existing modes is
preserved as it is and the new L3s mode obeys ipt
From: Willem de Bruijn
Add libc-compat workaround for definitions in linux/time.h that
duplicate those in libc time.h, sys/time.h and bits/time.h.
With this change, userspace builds succeeds when linux/time.h is
included after libc time.h and when it is included after sys/time.h.
The inverse re
On Mon, Sep 12, 2016 at 10:56:55AM +0200, Jesper Dangaard Brouer wrote:
> On Thu, 8 Sep 2016 23:30:50 -0700
> Alexei Starovoitov wrote:
>
> > On Fri, Sep 09, 2016 at 07:36:52AM +0200, Jesper Dangaard Brouer wrote:
> > > > > Lets do bundling/bulking from the start!
> > > >
> > > > mlx4 alrea
From: Willem de Bruijn
struct scm_timestamping has fields of type struct timespec. Now that
it is safe to include linux/time.h and time.h at the same time,
include linux/time.h directly in linux/errqueue.h
Without this patch, when compiling the following program after
make headers_install:
gcc
On Sun, Sep 11, 2016 at 09:49:26PM -0700, Adit Ranadive wrote:
> Add maintainer info for the PVRDMA driver.
You can probably squash the last three patches.
.. and fix the __u32 stuff throughout the entire driver please.
Jason
From: Willem de Bruijn
It was reported that linux/errqueue.h requires linux/time.h, but that
adding the include directly may cause userspace conflicts between
linux/time.h and glibc time.h:
https://lkml.org/lkml/2016/7/10/10
Address the conflicts using the standard libc-compat approach, then
On Sun, Sep 11, 2016 at 09:49:13PM -0700, Adit Ranadive wrote:
> + __u8raw[16];
> + struct {
> + __be64 subnet_prefix;
> + __be64 interface_id;
> + } global;
If this is not a userspace header do not use the __ varients..
Jason
On Sun, Sep 11, 2016 at 09:49:12PM -0700, Adit Ranadive wrote:
> We share some common structures with the user-level driver. This patch
> adds those structures and shared functions to traverse the QP/CQ rings.
> create mode 100644 drivers/infiniband/hw/pvrdma/pvrdma_uapi.h
> create mode 100644 d
1 - 100 of 285 matches
Mail list logo
