On Mon, Jul 6, 2015 at 8:41 AM, Eric Dumazet wrote:
> As mentioned by Alexei last week in Budapest, it is a bit weird
> to take a spinlock in order to drop a packet in a tc filter...
>
Arg, please ignore v2, I messed my git send-email command
--
To unsubscribe from this list: send the line "unsub
Second step for gact RCU operation :
We want to get rid of the spinlock protecting gact operations.
Stats (packets/bytes) will soon be per cpu.
gact_determ() would not work without a central packet counter,
so lets add it for this mode.
Signed-off-by: Eric Dumazet
Cc: Alexei Starovoitov
Acked-
First step for gact RCU operation :
Instead of testing if tcfg_pval is zero or not, just make it 1.
No change in behavior, but slightly faster code.
The smp_rmb()/smp_wmb() barriers, while not strictly needed at this
stage are added for upcoming spinlock removal.
Signed-off-by: Eric Dumazet
Ac
Reuse existing percpu infrastructure John Fastabend added for qdisc.
This patch adds a new cpustats parameter to tcf_hash_create() and all
actions pass false, meaning this patch should have no effect yet.
Signed-off-by: Eric Dumazet
Cc: Alexei Starovoitov
Acked-by: Jamal Hadi Salim
Acked-by: J
First step for gact RCU operation :
Instead of testing if tcfg_pval is zero or not, just make it 1.
No change in behavior, but slightly faster code.
The smp_rmb()/smp_wmb() barriers, while not strictly needed at this
stage are added for upcoming spinlock removal.
Signed-off-by: Eric Dumazet
Ac
qdisc_bstats_update_cpu() and other helpers were added to support
percpu stats for qdisc.
We want to add percpu stats for tc action, so this patch add common
helpers.
qdisc_bstats_update_cpu() is renamed to qdisc_bstats_cpu_update()
qdisc_qstats_drop_cpu() is renamed to qdisc_qstats_cpu_drop()
S
As mentioned by Alexei last week in Budapest, it is a bit weird
to take a spinlock in order to drop a packet in a tc filter...
Lets add percpu infra for tc actions and use it for gact & mirred.
Before changes, my host with 8 RX queues was handling 5 Mpps with gact,
and more than 11 Mpps after.
M
Enable SG support for Zynq SOC family devices.
Signed-off-by: Punnaiah Choudary Kalluri
---
drivers/net/ethernet/cadence/macb.c |6 ++
1 files changed, 2 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/cadence/macb.c
b/drivers/net/ethernet/cadence/macb.c
index caeb395..
On Sun, Jul 5, 2015 at 8:50 PM, Simon Horman wrote:
> Is it possible to get a 'Fixes:' tag?
I suppose it'd be appropriate to say
Fixes: 41063e9dd119 ("ipv4: Early TCP socket demux.")
As that is what introduces tcp early_demux, but that's just a guess as
I haven't bisected it (not even sure my t
On Sun, Jul 05, 2015 at 03:19:27PM -0700, Alex Gartrell wrote:
> On Sun, Jul 5, 2015 at 3:13 PM, Julian Anastasov wrote:
> > May be the patch fixes crashes? If yes, Simon
> > should apply it for ipvs/net tree, otherwise after
> > the merge window...
>
> Yeah this is definitely a crash-fix
Reworked with "use_oif_addr".
Thanks,
-Erik
On 3 July 2015 at 16:03, YOSHIFUJI Hideaki
wrote:
> Hi,
>
> Erik Kline wrote:
>> Per RFC 6724, section 4, "Candidate Source Addresses":
>>
>> It is RECOMMENDED that the candidate source addresses be the set
>> of unicast addresses assigned to t
Per RFC 6724, section 4, "Candidate Source Addresses":
It is RECOMMENDED that the candidate source addresses be the set
of unicast addresses assigned to the interface that will be used
to send to the destination (the "outgoing" interface).
Add a sysctl to enable this behaviour.
Signe
This commit implements the switchdev operations to add, delete and dump
VLANs for the Marvell 88E6352 and compatible switch chips.
This allows to access the switch VLAN Table Unit from standard userspace
commands such as "bridge vlan".
A configuration like "1t 2t 3t 4u" for VLAN 10 is achieved li
This patch adds the glue between DSA and switchdev operations to add,
delete and dump SWITCHDEV_OBJ_PORT_VLAN objects.
This is a first step to link the "bridge vlan" command with hardware
entries for DSA compatible switch chips.
Signed-off-by: Vivien Didelot
---
include/net/dsa.h | 9
n
Implement the Get Next and Load Purge operations for the VLAN Table
Unit, and a "vtu" debugfs file to read and write the hardware VLANs.
A populated VTU look like this:
# cat /sys/kernel/debug/dsa0/vtu
VID FID SID 0 1 2 3 4 5 6
550 5620 x x x u x t x
1000 10
Hi all,
This patchset brings full support for hardware VLANs in DSA, and the Marvell
88E6xxx compatible switch chips.
The first patch adds the VTU operations to the mv88e6xxx code, as well as a
"vtu" debugfs file to read and modify the hardware VLAN table.
The second patch adds the glue between
There are two duplicated xenvif_zerocopy_callback() definitions.
Remove one of them.
Signed-off-by: Liang Li
---
drivers/net/xen-netback/common.h | 3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index 8a495b3..c6cb85a 10064
On Sun, Jul 5, 2015 at 3:13 PM, Julian Anastasov wrote:
> May be the patch fixes crashes? If yes, Simon
> should apply it for ipvs/net tree, otherwise after
> the merge window...
Yeah this is definitely a crash-fix and it's existed since at least 3.10.
--
Alex Gartrell
--
To unsubscrib
Hello,
On Sun, 5 Jul 2015, Alex Gartrell wrote:
> It is possible that we bind against a local socket in early_demux when we
> are actually going to want to forward it. In this case, the socket serves
> no purpose and only serves to confuse things (particularly functions which
> implicit
On 7/6/2015 12:33 AM, Eric Dumazet wrote:
On Mon, 2015-07-06 at 00:16 +0300, Ido Shamay wrote:
We can have a relaxation of the condition by looking only at TCP/UDP
CQE indication (without check-sum indications)
This can cover us also when device rx-checksuming feature is off.
Do we want it or
On Mon, 2015-07-06 at 00:16 +0300, Ido Shamay wrote:
> We can have a relaxation of the condition by looking only at TCP/UDP
> CQE indication (without check-sum indications)
> This can cover us also when device rx-checksuming feature is off.
> Do we want it or a correlation between check-sum and
It is possible that we bind against a local socket in early_demux when we
are actually going to want to forward it. In this case, the socket serves
no purpose and only serves to confuse things (particularly functions which
implicitly expect sk_fullsock to be true, like ip_local_out).
Additionally,
On 7/2/2015 2:24 PM, Eric Dumazet wrote:
From: Eric Dumazet
Mellanox driver has the knowledge if rxhash is a L4 hash,
if it receives a non fragmented TCP or UDP frame and
NETIF_F_RXCSUM is enabled on netdev.
ip_summed value is CHECKSUM_UNNECESSARY in this case.
Signed-off-by: Eric Dumazet
Cc
Hello,
On Sun, 5 Jul 2015, Alex Gartrell wrote:
> + /* Remove the early_demux association unless it's bound for the
> + * exact same port and address on this host after translation.
> + */
> + if (!local || cp->vport != cp->dport ||
> + !ip_vs_addr_equal(cp->af,
It is possible that we bind against a local socket in early_demux when we
are actually going to want to forward it. In this case, the socket serves
no purpose and only serves to confuse things (particularly functions which
implicitly expect sk_fullsock to be true, like ip_local_out).
Additionally,
On Sat, Jul 04, 2015 at 03:33:00AM +0800, Liang Li wrote:
> There are two duplicated xenvif_zerocopy_callback() definitions.
> Remove one of them.
>
> Signed-off-by: Liang Li
Acked-by: Wei Liu
Please fix the time of your computer and resend.
Wei.
> ---
> drivers/net/xen-netback/common.h | 3
On UML builds, mdio-mux-mmioreg.c fails to compile:
drivers/net/phy/mdio-mux-mmioreg.c:50:3: error: implicit declaration of
function ‘ioremap’ [-Werror=implicit-function-declaration]
drivers/net/phy/mdio-mux-mmioreg.c:63:3: error: implicit declaration of
function ‘iounmap’ [-Werror=implicit-func
On Sat, Jul 04, 2015 at 12:44:36AM -0700, Vinson Lee wrote:
> Hi.
>
> With the latest Linux 4.2-rc1, I am hitting this build error with GCC
> 4.4.7 on CentOS 6.
>
> CC net/netfilter/ipset/ip_set_hash_netnet.o
> net/netfilter/ipset/ip_set_hash_netnet.c: In function ‘hash_netnet4_uadt’:
> ne
Hello,
On Fri, 3 Jul 2015, Alex Gartrell wrote:
> > - if packets go to local server IPVS should not touch
> > skb->dst, skb->sk, etc (NF_ACCEPT case)
>
> Yeah, the thing is that early demux could totally match for a socket
> that existed before we created the service, an
Hi Thomas,
On Fri, Jul 03, 2015 at 04:46:24PM +0200, Thomas Petazzoni wrote:
> Maxime,
>
> On Fri, 3 Jul 2015 16:25:51 +0200, Maxime Ripard wrote:
>
> > +static void mvneta_percpu_enable(void *arg)
> > +{
> > + struct mvneta_port *pp = arg;
> > +
> > + enable_percpu_irq(pp->dev->irq, IRQ_TY
Hi Maxime,
On Fri, Jul 03, 2015 at 04:25:49PM +0200, Maxime Ripard wrote:
> Now that our interrupt controller is allowing us to use per-CPU interrupts,
> actually use it in the mvneta driver.
>
> This involves obviously reworking the driver to have a CPU-local NAPI
> structure, and report for inc
Hi.
With the latest Linux 4.2-rc1, I am hitting this build error with GCC
4.4.7 on CentOS 6.
CC net/netfilter/ipset/ip_set_hash_netnet.o
net/netfilter/ipset/ip_set_hash_netnet.c: In function ‘hash_netnet4_uadt’:
net/netfilter/ipset/ip_set_hash_netnet.c:163: error: unknown field
‘cidr’ spec
Thanks guys for the review.
For completeness, I'll add smp_wmb() here :
gact->tcfg_pval= max_t(u16, 1, p_parm->pval);
smp_wmb();
gact->tcfg_ptype = p_parm->ptype;
And corresponding smp_rmb()
On Fri, Jul 3, 2015 at 12:49 PM, Jamal Hadi Salim wrote:
> On 07/02/15 09:07, Eric Du
When sending lots of small packets, this patch will generate an "Unable to
handle kernel paging request" in the memset call:
ndp16 = memset(ctx->delayed_ndp16, 0, ctx->max_ndp_size);
And I don't know why.
Any comment or suggestion would be greatly apreciated.
This has been reproduced in a QEMU X8
> Looks like routing by definition can not divert skbs with
> early-demux socket because input routing is not called.
Only if found socket has a valid sk->sk_rx_dst
Early demux :
1) if TCP lookup found a matching socket, we do the attachment
skb->sk = sk;
skb->destructor = sock_ede
On 7/3/15, 3:00 AM, Thomas Graf wrote:
On 06/18/15 at 09:49pm, Roopa Prabhu wrote:
+#ifdef CONFIG_LWTUNNEL
+ if (fi->fib_nh->nh_lwtstate) {
+ struct lwtunnel_state *lwtstate;
+
+ lwtstate = fi->fib_nh->nh_lwtstate;
+
> Cc: linux-ker...@vger.kernel.org; ian.campb...@citrix.com;
> wei.l...@citrix.com; xen-de...@lists.xenproject.org;
> netdev@vger.kernel.org
> Subject: Re: [PATCH] xen-netback: remove duplicated function definition
>
> From: Liang Li
> Date: Sat, 4 Jul 2015 03:33:00 +0800
>
> > There are two du
Since 1d5da757da860a6916adbf68b09e868062b4b3b8 (ax25: Stop using magic
neighbour cache operations.) any attempt to transmit IP packets over
a bpqether device will result in a message like "Dead loop on virtual
device bpq0, fix it urgently!"
Fix suggested by Eric W. Biederman .
Signed-off-by: Ralf
> The new FW will allow us to utilize some new features in our driver,
> mainly adding vlan filtering offload and vxlan offload support.
>
> In addition, this fixes several issues:
> 1. Packets from a VF with pvid configured which were sent with a
>different vlan were transmitted instead of be
39 matches
Mail list logo
