On Thu, 13 Aug 2015 19:01:07 +0200
Phil Sutter wrote:
> Handle IFF_NO_QUEUE as alternative to tx_queue_len being zero.
>
> Signed-off-by: Phil Sutter
Acked-by: Jesper Dangaard Brouer
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majord...@vge
On Thu, 13 Aug 2015 19:01:06 +0200
Phil Sutter wrote:
> This private net_device flag can be set by drivers to inform that a
> device runs fine without a qdisc attached. This was formerly done by
> setting tx_queue_len to zero.
>
> Signed-off-by: Phil Sutter
Acked-by: Jesper Dangaard Brouer
-
On Fri, 14 Aug 2015 10:41:53 +0200 Phil Sutter wrote:
> On Thu, Aug 13, 2015 at 12:11:57PM -0700, Stephen Hemminger wrote:
[...]
> >
> > But adding a flag risks breaking external scripts.
>
> Could you please elaborate on this? As far as I can tell, introducing a
> separate flag is the only so
Mon, Aug 17, 2015 at 05:32:06AM CEST, sfel...@gmail.com wrote:
>From: Scott Feldman
>
>> tree /sys/kernel/debug/rocker
>/sys/kernel/debug/rocker
>└── 525400123501
>├── fdb_tbl
>├── internal_vlan_tbl
>├── neigh_tbl
>├── of_dpa_flow_tbl
>└── of_dpa_group_tbl
>
>1 directory, 5
From: Noam Camus
We set controller to drop control frames and not trying
to pass them on. This is only needed for debug reasons.
Signed-off-by: Noam Camus
---
drivers/net/ethernet/ezchip/nps_enet.c |8
1 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethe
From: Noam Camus
TX handler is minimalistic and there is no need to schedule
a NAPI job.
Tx done will be processed during hardware interrupt context.
Signed-off-by: Noam Camus
---
drivers/net/ethernet/ezchip/nps_enet.c | 17 ++---
1 files changed, 10 insertions(+), 7 deletions(-)
From: Noam Camus
We need to set tx_skb pointer before send frame.
If we receive interrupt before we set pointer we will try
to free SKB with wrong pointer.
Now we are sure that SKB pointer will never be NULL during
handling TX done and check is removed.
Signed-off-by: Noam Camus
---
drivers/ne
From: Noam Camus
We define buf_int_enable in the minimal namespace it is used.
Signed-off-by: Noam Camus
---
drivers/net/ethernet/ezchip/nps_enet.c |7 ---
1 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/ezchip/nps_enet.c
b/drivers/net/ethernet/ezchi
From: Noam Camus
This is needed for when TX done interrupt is in
"level mode".
For example it is true for some simulators of this device.
Signed-off-by: Noam Camus
---
drivers/net/ethernet/ezchip/nps_enet.c |3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/drivers/net/
From: Noam Camus
When interrupt is received we read directly from control
register for RX/TX instead of reading cause register
since this register fails to indicate TX done when
TX interrupt is "edge mode".
Signed-off-by: Noam Camus
---
drivers/net/ethernet/ezchip/nps_enet.c |9 +
From: Noam Camus
This patch set is a bunch of fixes to make nps_enet work correctly with
all platforms, i.e. real device, emulation system, and simulation system.
The main trigger for this patch set was that in our emulation system
the TX end interrupt is "edge-sensitive" and therefore we cannot
Management firmware tells driver in case bandwidth configuration for
a specific function exists, but [regretably] the same field has different
meanings depending on the multi-function mode - it can either be
a percentile value or an actual speed.
For newer multi-function modes current logic is inc
From: Eric Dumazet
Date: Sat, 15 Aug 2015 10:54:07 -0700
> From: Eric Dumazet
>
> fib_lookup() forces FIB_LOOKUP_NOREF flag, while fib_table_lookup()
> does not.
>
> This patch solves the typical message at reboot time or device
> dismantle :
>
> unregister_netdevice: waiting for eth0 to beco
Hi Michal,
Thank you for your comments. I will wait for the review comments on kernel
changes, and repost this patch after that.
-Prem
-Original Message-
From: Michal Kubecek [mailto:mkube...@suse.cz]
Sent: Friday, August 14, 2015 4:42 PM
To: Premkumar Jonnala
Cc: netdev@vger.kernel.o
From: Scott Feldman
> tree /sys/kernel/debug/rocker
/sys/kernel/debug/rocker
└── 525400123501
├── fdb_tbl
├── internal_vlan_tbl
├── neigh_tbl
├── of_dpa_flow_tbl
└── of_dpa_group_tbl
1 directory, 5 files
> cat /sys/kernel/debug/rocker/525400123501/internal_vlan_tbl
i
Function i40e_clean_rx_irq() tries to reuse memory pages allocated
from the nearest node. To better support memoryless node, use
numa_mem_id() instead of numa_node_id() to get the nearest node with
memory.
This change should only affect performance.
Signed-off-by: Jiang Liu
---
drivers/net/ethe
Function ovs_flow_stats_update() allocates memory with __GFP_THISNODE
flag set, which may cause permanent memory allocation failure on
memoryless node. So replace cpu_to_node() with cpu_to_mem() to better
support memoryless node. For node with memory, cpu_to_mem() is the same
as cpu_to_node().
Thi
Function i40e_clean_rx_irq() tries to reuse memory pages allocated
from the nearest node. To better support memoryless node, use
numa_mem_id() instead of numa_node_id() to get the nearest node with
memory.
This change should only affect performance.
Signed-off-by: Jiang Liu
---
drivers/net/ethe
On 08/16/2015 09:42 PM, Victor Kaplansky wrote:
> Sometimes it is essential for libvirt to be able to configure MTU
> on guest's NICs to a value different from 1500.
>
> The change adds a new field to configuration area of network
> devices. It will be used to pass initial MTU from the device to
On 08/16/2015 09:42 PM, Victor Kaplansky wrote:
> Clarify general description of the mac, status and
> max_virtqueue_pairs fields. Specifically, the old description is
> vague about configuration layout and fields offsets when some of
> the fields are non valid.
>
> Also clarify that validity of
On 2015/7/30 13:55, Emmanuel Grumbach wrote:
> On Thu, Jul 30, 2015 at 8:18 AM, Fu, Zhonghui
> wrote:
>> Enable wiphy device to suspend/resume asynchronously. This can improve
>> system suspend/resume speed.
>>
> How will that impact the timing with respect to the suspend call
> coming from the
Table lookup compiles out when VRF is not enabled.
Signed-off-by: David Ahern
---
include/net/vrf.h | 24
net/ipv4/af_inet.c | 10 +-
2 files changed, 25 insertions(+), 9 deletions(-)
diff --git a/include/net/vrf.h b/include/net/vrf.h
index 0484d29d4589..40e379
On Sun, Aug 16, 2015 at 08:12:35PM +0200, Florian Westphal wrote:
> Phil Sutter wrote:
> > After having tested insertion, lookup, table walk and removal, spawn a
> > number of threads running operations on the same rhashtable. Each of
> > them will:
>
> [..]
>
> > + if (down_interruptible(&sta
Richard and Angela Maxwell has made a donation in your favour, for more details
contact : richardandmaxwe...@gmail.com
God bless you and happy Celebrations in advance.
Regard,
Richard and Angela Maxwell
---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.c
Phil Sutter wrote:
> After having tested insertion, lookup, table walk and removal, spawn a
> number of threads running operations on the same rhashtable. Each of
> them will:
[..]
> + if (down_interruptible(&startup_sem))
> + pr_err(" thread[%d]: down_interruptible failed\n", t
kbuild test robot reported:
tree: git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git master
head: d52736e24fe2e927c26817256f8d1a3c8b5d51a0
commit: 4e3c89920cd3a6cfce22c6f537690747c26128dd [751/762] net: Introduce VRF
related flags and helpers
reproduce: make htmldocs
>> Warning(
On 08/15/15 at 12:37am, Phil Sutter wrote:
> After having tested insertion, lookup, table walk and removal, spawn a
> number of threads running operations on the same rhashtable. Each of
> them will:
>
> 1) insert it's own set of objects,
> 2) lookup every successfully inserted object and finally
As Eric noted netif_index_is_vrf is not called with rcu_read_lock held,
so wrap the dev_get_by_index_rcu in rcu_read_lock and unlock.
If VRF is not enabled or oif is 0 skip the device lookup. In both cases
index cannot be the VRF master.
Signed-off-by: David Ahern
---
v2:
- per Florian's suggest
Sometimes it is essential for libvirt to be able to configure MTU
on guest's NICs to a value different from 1500.
The change adds a new field to configuration area of network
devices. It will be used to pass initial MTU from the device to
the driver, and to pass modified MTU from driver to the dev
This set of two patches adds a new field called default_mtu to
the configuration area of network devices. The motivation is to
allow libvirt to set initial MTU different from 1500 on guests
virtual NICs. We also propose to use this new field to report
MTU changes by the guest OS to the device to f
Clarify general description of the mac, status and
max_virtqueue_pairs fields. Specifically, the old description is
vague about configuration layout and fields offsets when some of
the fields are non valid.
Also clarify that validity of two status bits depends on two
different feature flags.
Sign
On 8/15/15 6:39 PM, Florian Westphal wrote:
David Ahern wrote:
As Eric noted netif_index_is_vrf is not called with rcu_read_lock held,
so use dev_get_by_index instead of dev_get_by_index_rcu.
If VRF is not enabled or oif is 0 skip the device lookup.
Signed-off-by: David Ahern
Why not
s
Hi Dave,
This patchset contains bug fixes, new RSS and pause parameters ethtool
options, and support for RX CHECKSUM_COMPLETE.
Patchset was applied and tested over commit adc6310 ("Merge branch
'mv88e6xxx-switchdev-fdb'").
Thanks,
Achiad
Achiad Shochat (8):
net/mlx5e: Have a single RSS Toepli
The indirection table size was defined by a variable that
was actually assigned a constant value.
Since we do not have any forseen intension to make it configurable
we simply made it a constant.
We also limit the number of channels such that the RSS indirection
table could always populate all RX r
- Change the maximum LRO session size from 16KB to 64KB
- Reduce the LRO session timeout from 512us to 32us in
order to reduce the TCP latency of non-LRO'ed flows.
- Fix skb_shinfo(skb)->gso_size and set skb_shinfo(skb)->gso_type.
- Fix a bug accessing un-initialized mdev pointer.
Signed-off-by:
- get_rxfh_key_size
- get_rxfh_indir_size
- get/set_rxfh indirection table and RSS Toeplitz hash key
- get_rxnfc
Signed-off-by: Achiad Shochat
---
drivers/net/ethernet/mellanox/mlx5/core/en.h | 7 ++-
.../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 71 +++---
drivers/
- Port speed settings are applied by the device only upon
port admin status transition from DOWN to UP.
So we enforce this transition regardless of the port's
current operation state (which may be occasionally DOWN if
for example the network cable is disconnected).
- Fix the PORT_UP/DOWN de
We un-intentionally limited the minimum rings size too much.
TX minimum ring size reduced from 128 to 64.
RX minimum ring size reduced from 128 to 2.
Signed-off-by: Achiad Shochat
---
drivers/net/ethernet/mellanox/mlx5/core/en.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --
No need to generate a unique key per TIR.
Generating a single key per netdev and copying it to all
its TIRs.
Signed-off-by: Achiad Shochat
---
drivers/net/ethernet/mellanox/mlx5/core/en.h | 1 +
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 5 -
2 files changed, 5 insertions(+), 1
Only rx/tx pause settings.
Autoneg setting is currently not supported.
Signed-off-by: Achiad Shochat
---
.../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 38
drivers/net/ethernet/mellanox/mlx5/core/port.c | 42 ++
include/linux/mlx5/driver.h
Only for packets with first ethertype set to IPv4/6 for now.
Signed-off-by: Achiad Shochat
---
drivers/net/ethernet/mellanox/mlx5/core/en.h | 8 +++--
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 5 ++-
drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 42 ++-
On 08/15/2015 04:31 AM, Willem de Bruijn wrote:
From: Willem de Bruijn
Add fanout mode PACKET_FANOUT_EBPF that accepts an en extended BPF
program to select a socket.
Update the internal eBPF program by passing to socket option
SOL_PACKET/PACKET_FANOUT_DATA a file descriptor returned by bpf().
On 08/15/2015 04:31 AM, Willem de Bruijn wrote:
From: Willem de Bruijn
Add fanout mode PACKET_FANOUT_CBPF that accepts a classic BPF program
to select a socket.
This avoids having to keep adding special case fanout modes. One
example use case is application layer load balancing. The QUIC
proto
43 matches
Mail list logo
