On 6/29/2016 11:25 PM, Jiri Pirko wrote:
Thu, Jun 30, 2016 at 06:04:39AM CEST, john.fastab...@gmail.com wrote:
On 16-06-29 08:35 PM, John Fastabend wrote:
On 16-06-29 03:09 PM, John Fastabend wrote:
On 16-06-29 02:33 PM, Or Gerlitz wrote:
On Wed, Jun 29, 2016 at 7:35 PM, John Fastabend
did you get my mail..?
On 16-06-29 03:47 PM, John Fastabend wrote:
Add another xmit_mode to pktgen to allow testing xmit functionality
of qdiscs. The new mode "queue_xmit" injects packets at
__dev_queue_xmit() so that qdisc is called.
Signed-off-by: John Fastabend
---
net/core/pktgen.c
Ideally, usbnet_resume is sufficient for device resume operation.
since usbcore function usb_resume_device() sets udev->reset_resume
flag as a quirk solution, reset_resume() quirk we can keep in
class driver also. We checked on dongle DA-Queen UFE20C,
without reset function resume operation
On Wed, Jun 22, 2016 at 06:11:00PM +0200, ggar...@abra.uab.cat wrote:
> +int __vsock_remove_tap(struct vsock_tap *vt) {
This function should be static, it's not used outside this source file.
> +static int __vsock_deliver_tap_skb(struct sk_buff *skb,
> + struct
2016-06-29 18:59 GMT+03:00 Jay Vosburgh :
> Veli-Matti Lintu wrote:
> [...]
>>Thanks for the patch. I have been now testing it and the reselection
>>seems to be working now in most cases, but I hit one case that seems
>>to consistently fail
> -Original Message-
> From: Dan Carpenter [mailto:dan.carpen...@oracle.com]
>
> "num_vec" needs to be signed for the error handling to work.
>
> Fixes: e261768e9e39 ('be2net: support asymmetric rx/tx queue counts')
> Signed-off-by: Dan Carpenter
>
> diff --git
Hi,
On Monday 06 June 2016 06:11 PM, Pramod Kumar wrote:
> Add PCI Phy support for Broadcom Northstar2 SoCs. This driver uses the
> interface from the iproc mdio mux driver to enable the devices
> respective phys.
>
> Reviewed-by: Andrew Lunn
> Signed-off-by: Jon Mason
From: Andrey Vagin
Date: Mon, 27 Jun 2016 15:33:56 -0700
> We found that sometimes a restored tcp socket doesn't work.
>
> A reason of this bug is incorrect window parameters and in this case
> tcp_acceptable_seq() returns tcp_wnd_end(tp) instead of tp->snd_nxt. The
> other
From:
Date: Wed, 29 Jun 2016 02:10:13 -0700
> +static void auto_nego_timer_callback(unsigned long data)
> +{
> + u16 val = 0;
> + u16 speed = 0;
> + u16 duplex = 0;
> +
> + struct tse_pcs *pcs = (struct tse_pcs *)data;
> + void __iomem *tse_pcs_base =
From: Masanari Iida
Date: Wed, 29 Jun 2016 23:36:20 +0900
> This patch fix a spelling typo in keystone-netcp.txt
>
> Signed-off-by: Masanari Iida
Applied.
From: Eric Dumazet
ether_addr_equal_64bits() requires some care about its arguments,
namely that 8 bytes might be read, even if last 2 byte values are not
used.
KASan detected a violation with null_mac_addr and lacpdu_mcast_addr
in bond_3ad.c
Fixes: 815117adaf5b ("bonding:
From: Dan Carpenter
Date: Wed, 29 Jun 2016 17:39:43 +0300
> "num_vec" needs to be signed for the error handling to work.
>
> Fixes: e261768e9e39 ('be2net: support asymmetric rx/tx queue counts')
> Signed-off-by: Dan Carpenter
Applied.
On Tue, Jun 28, 2016 at 09:01:12AM -0400, David Miller wrote:
> From: Joe Perches
> Date: Fri, 24 Jun 2016 11:32:26 -0700
>
> > There are code duplications of a masked ethernet address comparison here
> > so make it a separate function instead.
> >
> > Miscellanea:
> >
> > o
From: Mateusz Bajorski
Date: Wed, 29 Jun 2016 09:22:10 +0200
> When adding rule with NLM_F_EXCL flag then check if the same rule exist.
> If yes then exit with -EEXIST.
>
> This is already implemented in iproute2:
> if (cmd == RTM_NEWRULE) {
>
From: Dexuan Cui
Date: Wed, 29 Jun 2016 11:30:40 +
> @@ -1509,4 +1509,18 @@ static inline void commit_rd_index(struct
> vmbus_channel *channel)
> }
>
>
> +struct vmpipe_proto_header {
> + u32 pkt_type;
It is wasteful to have two empty lines before this
From: Jeff Kirsher
Date: Wed, 29 Jun 2016 09:30:02 -0700
> This series contains fixes to e1000e and ixgbevf.
Pulled, thanks Jeff.
Thu, Jun 30, 2016 at 09:57:21AM CEST, john.fastab...@gmail.com wrote:
>On 16-06-30 12:41 AM, Jiri Pirko wrote:
>> Thu, Jun 30, 2016 at 09:13:55AM CEST, sridhar.samudr...@intel.com wrote:
>>>
>>>
>>> On 6/29/2016 11:25 PM, Jiri Pirko wrote:
Thu, Jun 30, 2016 at 06:04:39AM CEST,
On Wed, Jun 22, 2016 at 06:11:02PM +0200, ggar...@abra.uab.cat wrote:
> + hdr = (struct af_vsockmon_hdr *) skb_push(skb, sizeof(*hdr));
> +
> + hdr->src_cid = pkt->hdr.src_cid;
> + hdr->src_port = pkt->hdr.src_port;
> + hdr->dst_cid = pkt->hdr.dst_cid;
> + hdr->dst_port =
From: "Seymour, Shane M"
Date: Tue, 28 Jun 2016 23:06:48 +
> In previous commit 01f83d69844d307be2aa6fea88b0e8fe5cbdb2f4
> the following comments were added:
>
> "When peer uses tiny windows, there is no use in packetizing to sub-MSS
> pieces for the sake of SWS or
From: John Crispin
Date: Wed, 29 Jun 2016 13:38:07 +0200
> This series contains 2 small code cleanups that are leftovers from the
> MIPS support. There is also a small fix that adds proper locking to the
> code accessing the IRQ registers. Without this fix we saw deadlocks
[ resending as a new email, as I'm assuming people do not sort their
INBOX via last email on thread, thus my last email is sitting in the
bottom of everyone's INBOX ]
I've hit this again. Not sure when it started, but I applied my old
debug trace_printk() patch (attached) and rebooted
On Thu, 2016-06-30 at 15:25 +0800, Yisen Zhuang wrote:
> From: Daode Huang
>
> This patch adds maintainers for hisilicon network subsystem driver
[]
> diff --git a/MAINTAINERS b/MAINTAINERS
[]
> @@ -5421,6 +5421,15 @@ F: include/uapi/linux/if_hippi.h
> F:
Hi,
We're struggling to implement the PCI SR-IOV management features for
administering Virtual Functions which represent networking devices using
the current Linux APIs. The problem is that these APIs incorporate all
sorts of assumptions which don't match chelsio networking cards.
For instance,
Chelsio T4/T5 cards have SR-IOV Capabilities on Physical Functions
0..3 and the administrative Driver(cxgb4) attaches to Physical Function 4.
Each of the Physical Functions 0..3 can support up to 16 Virtual
Functions. With the current Linux APIs, a 2-Port card would only be
able to use the Virtual
Add ndo_set_vf_mac support which allows to set the MAC address
for cxgb4vf interfaces from the host.
Signed-off-by: Hariprasad Shenai
---
drivers/net/ethernet/chelsio/cxgb4/cxgb4.h |3 +
drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c| 33 ++-
Add options to specify PF number associated with the VF for MAC ACL.
The current API assumes that there is one to one mapping between
Network Ports, Physical Functions and the SR-IOV Virtual Function.
But that's not true in the case of Chelsio, our cards have SR-IOV
Capabilities on Physical
Sergei Shtylyov writes:
> On 6/21/2016 2:45 PM, Chaehyun Lim wrote:
>
>> is not used anymore, so just remove it.
>
>s/it/#include/?
Fixed that in the pending branch.
--
Kalle Valo
From: Hariprasad Shenai
Date: Wed, 29 Jun 2016 09:01:19 +0530
> If the ctrl queue is full, just follows current path by allocating an
> skb. If that fails then caller will just have to handle that case as
> before.
>
> Signed-off-by: Hariprasad Shenai
On Thu, Jun 30, 2016 at 2:56 PM, Eric Dumazet wrote:
> From: Eric Dumazet
>
> ether_addr_equal_64bits() requires some care about its arguments,
> namely that 8 bytes might be read, even if last 2 byte values are not
> used.
>
> KASan detected a
From: Jeff Kirsher
Date: Wed, 29 Jun 2016 14:32:47 -0700
> This series contains updates and fixes to e1000e, igb, ixgbe and
> fm10k. A true smorgasbord of changes.
Pulled, thanks Jeff.
From: Mohamad Haj Yahia
Change the netdev watchdog to poll only the real active tx queues
instead of polling all tx queues.
The netdev driver doesn't necessarily have to start/stop all the
tx queues including the inactive tx queues.
Fixes: fd2ea0a79faa ('net: Use queue
On 30/06/16 16:13, Eric Dumazet wrote:
> From: Eric Dumazet
>
> ether_addr_equal_64bits() requires some care about its arguments,
> namely that 8 bytes might be read, even if last 2 byte values are not
> used.
>
> KASan detected a violation with null_mac_addr and
From: soohoon@f5.com
Date: Wed, 29 Jun 2016 15:07:21 -0400
> When MTU is changed unlink_urbs() flushes RX Q but mean while usbnet_bh()
> can fill up the Q at the same time.
> Depends on which HCD is down there unlink takes long time then the flush
> never ends.
>
> Signed-off-by: Soohoon Lee
On 30/06/16 14:56, Eric Dumazet wrote:
> From: Eric Dumazet
>
> ether_addr_equal_64bits() requires some care about its arguments,
> namely that 8 bytes might be read, even if last 2 byte values are not
> used.
>
> KASan detected a violation with null_mac_addr and
When looking up a client connection to which to route a packet, we need to
check that the packet came from the correct source so that a peer can't try
to muck around with another peer's connection.
Signed-off-by: David Howells
---
net/rxrpc/conn_object.c |4 +++-
1
Provide refcount helper functions for connections so that the code doesn't
touch conn->usage directly.
Also provide queueing helper functions so that the queueing of local and
connection objects can be fixed later.
Signed-off-by: David Howells
---
net/rxrpc/ar-internal.h
lso:
http://git.kernel.org/cgit/linux/kernel/git/dhowells/linux-fs.git/log/?h=rxrpc-rewrite
Tagged thusly:
git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs.git
rxrpc-rewrite-20160630
David
---
David Howells (18):
rxrpc: Check the source of a packet to a cli
On Thu, Jun 30, 2016 at 4:13 PM, Eric Dumazet wrote:
> From: Eric Dumazet
>
> ether_addr_equal_64bits() requires some care about its arguments,
> namely that 8 bytes might be read, even if last 2 byte values are not
> used.
>
> KASan detected a
When a call is disconnected, clear the call's pointer to the connection and
release the associated ref on that connection. This means that the call no
longer pins the connection and the connection can be discarded even before
the call is.
As the code currently stands, the call struct is
The main connection list is used for two independent purposes: primarily it
is used to find connections to reap and secondarily it is used to list
connections in procfs.
Split the procfs list out from the reap list. This allows the reap list to
be phased out in stages as the client conns and
Rather than calling rxrpc_get_connection() manually before calling
rxrpc_queue_conn(), do it inside the queue wrapper.
This allows us to do some important fixes:
(1) If the usage count is 0, do nothing. This prevents connections from
being reanimated once they're dead.
(2) If
Check that the client conns cache is empty before module removal and bug if
not, listing any offending connections that are still present. Unfortunately,
if there are connections still around, then the transport socket is still
unexpectedly open and active, so we can't just unallocate the
Turn the connection event and state #define lists into enums and move
outside of the struct definition.
Whilst we're at it, change _SERVER to _SERVICE in those identifiers and add
EV_ into the event name to distinguish them from flags and states.
Also add a symbol indicating the number of states
Sigh.
It seems I misunderstood the comment on read_seqbegin_or_lock() and failed to
initialise seq in rxrpc_find_service_conn_rcu(). For some reason my compiler
didn't notice, though Fengguang's magic compile farm did.
So, please ignore this version of the patchset and I'll post a new set.
> + /* verify MAC addr is valid */
> + if (!is_zero_ether_addr(mac) && !is_valid_ether_addr(mac) &&
> + is_multicast_ether_addr(mac)) {
This is really odd as verification goes; Currently this is a very elaborate
way of checking for multicast, but I guess it's probably a mistake.
If rxrpc_connect_call() fails during the creation of a client connection,
there are two bugs that we can hit that need fixing:
(1) The call state should be moved to RXRPC_CALL_DEAD before the call
cleanup phase is invoked. If not, this can cause an assertion failure
later.
(2)
From: Daniel Jurgens
Add a timeout to avoid an infinite loop waiting for RQ's to flush. This
occurs during AER/EEH and will also happen if the device stops posting
completions due to internal error or reset, or if moving the RQ to the
error state fails. Also cleanup posted
From: Matthew Finlay
ConnectX4-Lx uses an inline wqe mode that currently defaults to
requiring the entire L2 header be included in the wqe.
This patch fixes mlx5e_get_inline_hdr_size() to account for
all L2 headers (VLAN, QinQ, etc) using skb_network_offset(skb).
Fixes:
From: Rana Shahout
Valid weight assigned to ETS TClass values are 1-100
Fixes: 08fb1dacdd76 ('net/mlx5e: Support DCBNL IEEE ETS')
Signed-off-by: Rana Shahout
Signed-off-by: Saeed Mahameed
---
From: Daniel Jurgens
The device ID for VFs is in a different location than PFs. This results
in the poll always timing out for VFs. There's no good way to read the
VF device ID without using the PF's configuration space. Switch to waiting
for the health poll to start
From: Daniel Jurgens
Add callback to handle TX timeouts.
Fixes: f62b8bb8f2d3 ('net/mlx5: Extend mlx5_core to support ConnectX-4 Ethernet
functionality')
Signed-off-by: Daniel Jurgens
Signed-off-by: Saeed Mahameed
---
Hi Dave,
This series provides two set of fixes to the mlx5 driver:
- Resiliency fixes for reset flow and internal pci errors
- xmit path fixes
Please consider queuing those patches for -stable (4.6).
Reset flow fixes for core driver:
- Add more commands to the list of
From: Rana Shahout
The default fallback function used by mlx5e select queue can return
any TX queues in range [0..dev->num_real_tx_queues).
The current implementation assumes that the fallback function returns
a number in the range [0.. number of channels). Actually
From: Mohamad Haj Yahia
The current implementation does not handle timeout in case of command
with callback request, and this can lead to deadlock if the command
doesn't get fw response.
Add delayed callback timeout work before posting the command to fw.
In case of real fw
From: Daniel Jurgens
Change page cleanup flow when in internal error to properly decrement
the page counts when reclaiming pages. The prevents timing out waiting
for extra pages that were actually cleaned up previously.
fixes: 89d44f0a6c73 ('net/mlx5_core: Add pci error
From: Mohamad Haj Yahia
In internal error state the health poll thread will eventually call
synchronize_irq() (to safely trigger command completions) which might
sleep, so we are calling sleeping function from atomic context which is
invalid.
Here we move
From: Mohamad Haj Yahia
Call command completion handler in case of timeout when working in
interrupts mode.
Avoid flushing the commands workqueue after acquiring the semaphores to
prevent a potential deadlock.
Fixes: e126ba97dba9 ('mlx5: Add driver for Mellanox Connect-IB
On Thu, 2016-06-30 at 15:35 +0200, Dmitry Vyukov wrote:
>
> What does nm -S say?
> I've tried to compile a sample program with gcc6 and
> __attribute__((aligned(8))) does in fact force alignment of start of
> the symbol, but size still stays 6. (and asan still barks on my test
> program).
Add RCU destruction for connections and calls as the RCU lookup from the
transport socket data_ready handler is going to come along shortly.
Whilst we're at it, move the cleanup workqueue flushing and RCU barrierage
into the destruction code for the objects that need it (locals and
connections)
From: Eric Dumazet
ether_addr_equal_64bits() requires some care about its arguments,
namely that 8 bytes might be read, even if last 2 byte values are not
used.
KASan detected a violation with null_mac_addr and lacpdu_mcast_addr
in bond_3ad.c
Same problem with mac_bcast[]
Split the client-specific connection code out into its own file. It will
behave somewhat differently from the service-specific connection code, so
it makes sense to separate them.
Signed-off-by: David Howells
---
net/rxrpc/ar-internal.h |7 +
net/rxrpc/conn_client.c |
From: Herbert Xu
rxkad uses stack memory in SG lists which would not work if stacks were
allocated from vmalloc memory. In fact, in most cases this isn't even
necessary as the stack memory ends up getting copied over to kmalloc
memory.
This patch eliminates all the
Move to using RCU access to a peer's service connection tree when routing
an incoming packet. This is done using a seqlock to trigger retrying of
the tree walk if a change happened.
Further, we no longer get a ref on the connection looked up in the
data_ready handler unless we queue the
Prune the contents of the rxrpc_conn_proto struct. Most of the fields aren't
used anymore.
Signed-off-by: David Howells
---
net/rxrpc/ar-internal.h | 20 +++-
net/rxrpc/call_object.c |2 +-
net/rxrpc/conn_client.c | 11 ---
Split the service-specific connection code out into into its own file. The
client-specific code has already been split out. This will leave just the
common code in the original file.
Signed-off-by: David Howells
---
net/rxrpc/Makefile |1
The socket's accept queue (socket->acceptq) should be accessed under
socket->call_lock, not under the connection lock.
Signed-off-by: David Howells
---
net/rxrpc/call_event.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/rxrpc/call_event.c
On Thu, Jun 30, 2016 at 1:52 PM, Jiri Pirko wrote:
> Why not to have 2 modes:
> 1) lagacy - the current solution, blackbox eswitch, undefined behaviour
> 2) switchdev - with representors, all features possible as on physical
> switches, whitebox eswitch configured using
Since the function won't ever change the data 'kind' is pointing at, it
can sanely be made const.
Fixes: e0513807f6dbb ("ip-address: Support filtering by slave type, too")
Suggested-by: Stephen Hemminger
Signed-off-by: Phil Sutter
---
ip/ipaddress.c | 2 +-
1
On Thu, Jun 30, 2016 at 5:28 PM, Eric Dumazet wrote:
> On Thu, 2016-06-30 at 16:58 +0300, Saeed Mahameed wrote:
>> - for (i = 0; i < dev->num_tx_queues; i++) {
>> + for (i = 0; i < dev->real_num_tx_queues; i++) {
>>
From: Shmulik Ladkani
Date: Wed, 29 Jun 2016 21:47:03 +0300
> ip_skb_dst_mtu uses skb->sk, assuming it is an AF_INET socket (e.g. it
> calls ip_sk_use_pmtu which casts sk as an inet_sk).
>
> However, in the case of UDP tunneling, the skb->sk is not necessarily an
>
> On Jun 30, 2016, at 08:59, Steven Rostedt wrote:
>
> [ resending as a new email, as I'm assuming people do not sort their
> INBOX via last email on thread, thus my last email is sitting in the
> bottom of everyone's INBOX ]
>
> I've hit this again. Not sure when it
On Thu, 2016-06-30 at 15:04 +0200, Dmitry Vyukov wrote:
> On Thu, Jun 30, 2016 at 2:56 PM, Eric Dumazet wrote:
> > From: Eric Dumazet
> >
> > ether_addr_equal_64bits() requires some care about its arguments,
> > namely that 8 bytes might be read, even
On Thu, Jun 30, 2016 at 3:25 PM, Eric Dumazet wrote:
> On Thu, 2016-06-30 at 15:04 +0200, Dmitry Vyukov wrote:
>> On Thu, Jun 30, 2016 at 2:56 PM, Eric Dumazet wrote:
>> > From: Eric Dumazet
>> >
>> > ether_addr_equal_64bits()
On Thu, 2016-06-30 at 16:58 +0300, Saeed Mahameed wrote:
> From: Mohamad Haj Yahia
>
> Change the netdev watchdog to poll only the real active tx queues
> instead of polling all tx queues.
> The netdev driver doesn't necessarily have to start/stop all the
> tx queues
From: Daniel Jurgens
Avoid an infinite loop by timing out waiting for the SQ to flush. Also
clean up the TX descriptors if that happens.
Fixes: f62b8bb8f2d3 ('net/mlx5: Extend mlx5_core to support ConnectX-4 Ethernet
functionality')
Signed-off-by: Daniel Jurgens
From: Shaker Daibes
Add Link UP/Down prints to kernel log when link state changes
Signed-off-by: Shaker Daibes
Signed-off-by: Saeed Mahameed
---
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 7 +--
1 file changed, 5
From: Mohamad Haj Yahia
In case of internal error state we will simulate the commands status
through the return value translation function, but we need to simulate
all the teardown fw commands as successful so we will not have fw
command failure prints.
This also fix memory
From: Jakub Kicinski
Date: Wed, 29 Jun 2016 21:55:52 +0100
> Three small patches for net-next. First and second patches
> improve the code quality by spelling things correctly and
> removing unused parameters. Third patch hooks-in standard
> kernel implementation
On Thu, Jun 30, 2016 at 4:08 PM, Eric Dumazet wrote:
> On Thu, 2016-06-30 at 15:35 +0200, Dmitry Vyukov wrote:
>
>>
>> What does nm -S say?
>> I've tried to compile a sample program with gcc6 and
>> __attribute__((aligned(8))) does in fact force alignment of start of
>>
Move the peer lookup done in input.c by data_ready into
rxrpc_find_connection().
Signed-off-by: David Howells
---
net/rxrpc/ar-internal.h |3 --
net/rxrpc/conn_object.c | 73 ---
net/rxrpc/input.c | 30
> From: David Miller [mailto:da...@davemloft.net]
> Sent: Thursday, June 30, 2016 20:45
> To: Dexuan Cui
> Cc: gre...@linuxfoundation.org; netdev@vger.kernel.org; linux-
> ker...@vger.kernel.org; de...@linuxdriverproject.org; o...@aepfle.de;
> a...@canonical.com;
Overhaul the usage count accounting for the rxrpc_connection struct to make
it easier to implement RCU access from the data_ready handler.
The problem is that currently we're using a lock to prevent the garbage
collector from trying to clean up a connection that we're contemplating
unidling. We
Move the lookup of a peer from a call that's being accepted into the
function that creates a new incoming connection. This will allow us to
avoid incrementing the peer's usage count in some cases in future.
Note that I haven't bother to integrate rxrpc_get_addr_from_skb() with
Each channel on a connection has a separate, independent number space from
which to allocate callNumber values. It is entirely possible, for example,
to have a connection with four active calls, each with call number 1.
Note that the callNumber values for any particular channel don't have to
On Wed, 29 Jun 2016 13:03:06 -0700
John Fastabend wrote:
> Add another xmit_mode to pktgen to allow testing xmit functionality
> of qdiscs. The new mode "queue_xmit" injects packets at
> __dev_queue_xmit() so that qdisc is called.
>
> Signed-off-by: John Fastabend
From: Daode Huang
This patch fixes the code style in hns driver. Change it from
"buff = buff + xxx" to "buff += xxx". The reveiw comments is
from andy.
Reviewed-by: Andriy Shevchenko
Signed-off-by: Daode Huang
From: Daode Huang
In comment line, some time miss a space before */, so this
patch adds a space before */.
Signed-off-by: Daode Huang
---
drivers/net/ethernet/hisilicon/hns/hns_enet.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
From: Kejian Yan
Since the registers of subctrl may be different, it is better to
mv the registers from hns mdio driver routine to device tree node.
Signed-off-by: Kejian Yan
---
change log:
v2: fix the wrong indentations
v1: first submit
link:
From: Kejian Yan
It is PORT_TP type if the service port is GE mode. It is wrong to
judge the port type by using if it is service port. Adding the media
type to know port type.
Reported-by: Jinchuan Tian
Signed-off-by: Kejian Yan
From: Daode Huang
This patch fixes code sytle of hns driver to make it
simple.
Signed-off-by: Daode Huang
---
drivers/net/ethernet/hisilicon/hns/hns_dsaf_misc.c | 16 +++-
1 file changed, 7 insertions(+), 9 deletions(-)
diff
Thu, Jun 30, 2016 at 09:13:55AM CEST, sridhar.samudr...@intel.com wrote:
>
>
>On 6/29/2016 11:25 PM, Jiri Pirko wrote:
>>Thu, Jun 30, 2016 at 06:04:39AM CEST, john.fastab...@gmail.com wrote:
>>>On 16-06-29 08:35 PM, John Fastabend wrote:
On 16-06-29 03:09 PM, John Fastabend wrote:
>On
Signed-off-by: Michael S. Tsirkin
Signed-off-by: Jason Wang
---
include/linux/skb_array.h | 9 +
1 file changed, 9 insertions(+)
diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h
index 2dd0d1e..f4dfade 100644
---
This patch introduces a new event - NETDEV_CHANGE_TX_QUEUE_LEN, this
will be triggered when tx_queue_len. It could be used by net device
who want to do some processing at that time. An example is tun who may
want to resize tx array when tx_queue_len is changed.
Cc: John Fastabend
We used to queue tx packets in sk_receive_queue, this is less
efficient since it requires spinlocks to synchronize between producer
and consumer.
This patch tries to address this by:
- switch from sk_receive_queue to a skb_array, and resize it when
tx_queue_len was changed.
- introduce a new
This patch uses xmit_more optimization to reduce
number of TX doorbells write per packet.
Signed-off-by: Manish
Signed-off-by: Yuval Mintz
---
drivers/net/ethernet/qlogic/qede/qede_main.c | 37 +---
1 file changed, 23
When handling IP fragmented packets with csum in their
transport header, the csum isn't changed as part of the
fragmentation. As a result, the packet containing the
transport headers would have the correct csum of the original
packet, but one that mismatches the actual packet that
passes on the
This patch cleanups qede_poll() routine a bit
and allows qede_poll() to do single iteration to handle
TX completion [As under heavy TX load qede_poll() might
run for indefinite time in the while(1) loop for TX
completion processing and cause CPU stuck].
Signed-off-by: Manish
Signed-off-by: Manish
Signed-off-by: Yuval Mintz
---
drivers/net/ethernet/qlogic/qede/qede.h | 3 ++
drivers/net/ethernet/qlogic/qede/qede_ethtool.c | 46 +
drivers/net/ethernet/qlogic/qede/qede_main.c| 3 +-
On 30/06/2016 08:03, kbuild test robot wrote:
> Hi,
>
> [auto build test ERROR on net/master]
> [also build test ERROR on v4.7-rc5 next-20160629]
> [cannot apply to net-next/master]
> [if your patch is applied to the wrong git tree, please drop us a note to
> help improve the system]
>
> url:
1 - 100 of 255 matches
Mail list logo