On 19/05/2015 21:35, Jason Gunthorpe wrote:
...
> The share count should be 'listen_sharecount' because it *only* works
> for listen.
>
> The above test in cm_destroy_id should only be in the listen branch of
> the if.
Okay.
>
>> + * Create a new listening ib_cm_id and listen on the given serv
On Wed, May 20, 2015 at 05:30:40PM -0700, Eric Dumazet wrote:
> On Wed, 2015-05-20 at 16:46 -0700, Cong Wang wrote:
>
> > There is very little to do on ingress side since there is no queue at all,
> > not to mention priority, you could try ifb to see if it fits your need.
>
> Note that if the nee
On Wed, 20 May 2015 16:13:54 +0300
Deniz Eren wrote:
> Hi,
>
> I'm having problem with packet capturing performance on my linux server.
>
> I am using Intel ixgbe 10g NIC with v3.19.1 version driver over Linux
> 3.15.9 based system. Naturally I can route 3.8Mpps packet from spoof
> (random sou
On Wed, May 20, 2015 at 10:05:07PM -0700, sfel...@gmail.com wrote:
> From: Scott Feldman
>
> In review of Simon's patchset "rocker: transaction fixes". it was noted
> that rocker->neigh_tbl_next_index was unprotected in the call path below
> and could race with other contexts calling rocker_port_
On 05/19/2015 03:25 PM, ext Eric Dumazet wrote:
On Tue, 2015-05-19 at 10:41 +0200, Jetchko Jekov wrote:
Hello,
I hope this is the right way to report a bug regarding iproute.
While playing with gre[tap] tunnels I run in the following:
# modprobe ip-gre
# ip l add gre1 type gre remote 1.1.1.
Thu, May 21, 2015 at 07:05:07AM CEST, sfel...@gmail.com wrote:
>From: Scott Feldman
>
>In review of Simon's patchset "rocker: transaction fixes". it was noted
>that rocker->neigh_tbl_next_index was unprotected in the call path below
>and could race with other contexts calling rocker_port_ipv4_neig
On 20/05/2015 01:35, Jason Gunthorpe wrote:
> On Tue, May 19, 2015 at 12:35:45PM -0600, Jason Gunthorpe wrote:
>> On Sun, May 17, 2015 at 08:51:01AM +0300, Haggai Eran wrote:
>>> @@ -212,6 +212,8 @@ struct cm_id_private {
>>> spinlock_t lock;/* Do not acquire inside cm.lock */
>>> s
On 2015/05/21 14:05, sfel...@gmail.com wrote:
> From: Scott Feldman
>
> In review of Simon's patchset "rocker: transaction fixes". it was noted
> that rocker->neigh_tbl_next_index was unprotected in the call path below
> and could race with other contexts calling rocker_port_ipv4_neigh():
>
>
On 05/20/2015 08:45 PM, Andy Grover wrote:
> On 05/13/2015 03:12 PM, Chris Leech wrote:
>> This is only about the structures and functionality involved in
>> maintaining the
>> iSCSI session, the SCSI host along with it's discovered targets
>> and devices has
>> no association with network namespac
From: Stock David (NHS GREAT YARMOUTH AND WAVENEY CCG)
Sent: 21 May 2015 10:05
To: Stock David (NHS GREAT YARMOUTH AND WAVENEY CCG)
Subject:
You have won contact Allen On allemwilliam10...@gmail.com for info.
Allen Williams
+27612909541
You have won contact Allen On allemwilliam10...@gmail.com for info.
Allen Williams
+27612909541
This message may contain confidential information. If you are not the intended
re
On Wed, 20 May 2015 14:23:17 -0700
Cong Wang wrote:
> On Tue, May 19, 2015 at 2:36 PM, Jesper Dangaard Brouer
> wrote:
> > +
> > +# More generic replacement for pgset(), that does not depend on global
> > +# variable for proc file.
> > +function proc_cmd() {
> > +local result
> > +local
unix_stream_recvmsg is refactored to unix_stream_read_generic in this
patch and enhanced to deal with pipe splicing. The refactoring is
inneglible, we mostly have to deal with a non-existing struct msghdr
argument.
Signed-off-by: Hannes Frederic Sowa
---
v2:
* checked that CONFIG_UNIX=m still wor
Signed-off-by: Hannes Frederic Sowa
---
v2:
* remove now unused variable i in ip_append_page (thanks, Cong!)
* switched to EXPORT_SYMBOL_GPL for skb_append_pagefrags
include/linux/skbuff.h | 3 +++
net/core/skbuff.c | 18 ++
net/ipv4/ip_output.c | 15 ---
3 fi
This patch implements sendpage support for AF_UNIX SOCK_STREAM
sockets. This is also required for a complete splice implementation.
The implementation is a bit tricky because we append to already existing
skbs and so have to hold unix_sk->readlock to protect the reading side
from either advancing
This series implements zerocopy support for AF_UNIX SOCK_STREAM sockets.
Changelog in the specific patches. Thanks to all the reviewers!
Hannes Frederic Sowa (4):
net: skbuff: add skb_append_pagefrags and use it
net: af_unix: implement stream sendpage support
net: make skb_splice_bits more
Prepare skb_splice_bits to be able to deal with AF_UNIX sockets.
AF_UNIX sockets don't use lock_sock/release_sock and thus we have to
use a callback to make the locking and unlocking configureable.
Signed-off-by: Hannes Frederic Sowa
Acked-by: Eric Dumazet
---
v2:
* no changes
include/linux/s
On Wed, 20 May 2015 14:33:42 -0700
Cong Wang wrote:
> On Tue, May 19, 2015 at 2:36 PM, Jesper Dangaard Brouer
> wrote:
> > +# Threads are specified with parameter -t value in $THREADS
> > +for ((thread = 0; thread < $THREADS; thread++)); do
> > +dev=${DEV}@${thread}
> > +
> > +# Add remo
On 20.05.2015 02:33, Mahesh Bandewar wrote:
On Thu, May 14, 2015 at 6:56 AM, Konstantin Khlebnikov
wrote:
ipvlan_start_xmit() is called with rcu_read_lock_bh() while its internal
structures requre normal rcu_read_lock().
Signed-off-by: Konstantin Khlebnikov
---
[ 802.945151] =
v2: address nitpicks from Cong Wang
- Remove useless cat's, but keep them for old pgset()
- Comment on: Due to pgctrl, cannot use exit code $? from grep
- Use arithmetic compare in pktgen_sample03_burst_single_flow.sh
This patchset is focused on making pktgen easier to use and better
documented
And cleanup some whitespaces in pktgen.txt.
Signed-off-by: Jesper Dangaard Brouer
---
0 files changed, 0 insertions(+), 0 deletions(-)
diff --git a/Documentation/networking/pktgen.txt
b/Documentation/networking/pktgen.txt
index 747facc..62f5ebb 100644
--- a/Documentation/networking/pktgen.txt
Too many spaces were introduced in commit 63adc6fb8ac0 ("pktgen: cleanup
checkpatch warnings"), thus misaligning "src_min:" to other columns.
Fixes: 63adc6fb8ac0 ("pktgen: cleanup checkpatch warnings")
Signed-off-by: Jesper Dangaard Brouer
---
0 files changed, 0 insertions(+), 0 deletions(-)
d
And cleanup some whitespaces in pktgen.txt.
Signed-off-by: Jesper Dangaard Brouer
---
Documentation/networking/pktgen.txt | 18 --
1 files changed, 8 insertions(+), 10 deletions(-)
diff --git a/Documentation/networking/pktgen.txt
b/Documentation/networking/pktgen.txt
index 7
v3:
- Aborted v2 send due it was not generating diff stat
(this is a bug in stg-mail, if not in the root directory)
v2: address nitpicks from Cong Wang
- Remove useless cat's, but keep them for old pgset()
- Comment on: Due to pgctrl, cannot use exit code $? from grep
- Use arithmetic compa
Too many spaces were introduced in commit 63adc6fb8ac0 ("pktgen: cleanup
checkpatch warnings"), thus misaligning "src_min:" to other columns.
Fixes: 63adc6fb8ac0 ("pktgen: cleanup checkpatch warnings")
Signed-off-by: Jesper Dangaard Brouer
---
net/core/pktgen.c |2 +-
1 files changed, 1 ins
Giving /proc/net/pktgen/pgctrl an invalid command just returns shell
success and prints a warning in dmesg. This is not very useful for
shell scripting, as it can only detect the error by parsing dmesg.
Instead return -EINVAL when the command is unknown, as this provides
userspace shell scripting
The pktgen.txt documentation over available config options were not complete.
Making the list complete by adding the following.
Pgcontrol commands:
reset
Device commands:
burst
queue_map_min
queue_map_max
skb_priority
tos
traffic_class
node
spi
dst6_max
dst6_min
vlan_cfi
vlan_id
vla
The pktgen.txt documentation still claimed that adding same device to
multiple threads were not supported, but it have been since 2008 via
commit e6fce5b916cd7 ("pktgen: multiqueue etc.").
Document this and describe the naming scheme dev@X, as the procfile name
still need to be unique.
Fixes: e6f
On Thu, May 21, 2015 at 12:38:12AM +0800, Herbert Xu wrote:
> As we're now always including the high bits of the sequence number
> in the IV generation process we need to ensure that they don't
> contain crap.
>
> This patch ensures that the high sequence bits are always zeroed
> so that we don't
Add the pktgen samples script pktgen_sample02_multiqueue.sh that
demonstrates generating packets on multiqueue NICs.
Specifically notice the options "-t" that specifies how many
kernel threads to activate. Also notice the flag QUEUE_MAP_CPU,
which cause the SKB TX queue to be mapped to the CPU ru
Add the first basic pktgen samples script pktgen_sample01_simple.sh,
which demonstrates the a simple use of the helper functions.
Removing pktgen.conf-1-1 as that example should be covered now.
The naming scheme pktgen_sampleNN, where NN is a number, should encourage
reading the samples in a speci
Add the pktgen samples script pktgen_sample03_burst_single_flow.sh
that demonstrates how to acheive maximum performance.
If correctly tuned[1] single CPU 10Gbit/s wirespeed small pkts is
possible[2] which is 14.88Mpps. The trick is to take advantage of the
"burst" feature introduced in commit 38b
Preparing for removing existing samples/pktgen/ scripts, and
replacing these with easier to use samples.
This commit provides two helper shell files, that can
be "included" by shell source'ing. Namely "functions.sh"
and "parameters.sh".
The parameters.sh file support easy and consistant parameter
This script pktgen_bench_xmit_mode_netif_receive.sh is a benchmark
script, which can be used for benchmarking part of the network stack.
This can be used for performance improving or catching regression in
that area.
The script is developed for benchmarking ingress qdisc path, original
idea by Ale
On Thu, 21 May 2015 12:11:42 +0200 Jesper Dangaard Brouer
wrote:
> ---
>
> 0 files changed, 0 insertions(+), 0 deletions(-)
Aborting V2 stg-mail sending, because it's not generating the correct
diff stat. This is a bug in stg if not in the root of the git dir.
--
Best regards,
Jesper Da
On Thu, May 21, 2015 at 9:33 AM, Haggai Eran wrote:
>> you can use non default pkeys as well here.
>> $ ip link add link ib0 name ib0.1 type ipoib pkey 0x8001
> Right. I think when we started development of the namespaces patches
> these child interfaces (rtnetlink with pkey) didn't work for us,
Hi:
This series of patches convert all in-tree AEAD users that I
could find to the new single SG list interface. For IPsec it
also adopts the new explicit IV generator scheme.
To recap, the old AEAD interface takes an associated data (AD)
SG list in addition to the plain/cipher text SG list(s).
From: Wilson Kok
Check in fdb_add_entry() if the source port should learn, similar
check is used in br_fdb_update.
Note that new fdb entries which are added manually or
as local ones are still permitted.
This patch has been tested by running traffic via a bridge port and
switching the port's stat
This patch adds IV generator information for each AEAD and block
cipher to xfrm_algo_desc. This will be used to access the new
AEAD interface.
Signed-off-by: Herbert Xu
---
include/net/xfrm.h |2 ++
net/xfrm/xfrm_algo.c | 16
2 files changed, 18 insertions(+)
diff --g
This patch makes use of the new AEAD interface which uses a single
SG list instead of separate lists for the AD and plain text.
Signed-off-by: Herbert Xu
---
net/mac802154/llsec.c | 41 ++---
1 file changed, 14 insertions(+), 27 deletions(-)
diff --git a/n
This patch makes use of the new AEAD interface which uses a single
SG list instead of separate lists for the AD and plain text. The
IV generation is also now carried out through normal AEAD methods.
Signed-off-by: Herbert Xu
---
net/ipv4/esp4.c | 197 ++
This patch makes use of the new AEAD interface which uses a single
SG list instead of separate lists for the AD and plain text.
Signed-off-by: Herbert Xu
---
net/mac80211/aes_ccm.c | 30 ++
net/mac80211/aes_gcm.c | 30 ++
net/mac8021
This patch adds IV generator information to xfrm_state. This
is currently obtained from our own list of algorithm descriptions.
Signed-off-by: Herbert Xu
---
include/net/xfrm.h |1 +
net/key/af_key.c |1 +
net/xfrm/xfrm_user.c | 40 +++-
3 fi
This patch makes use of the new AEAD interface which uses a single
SG list instead of separate lists for the AD and plain text. The
IV generation is also now carried out through normal AEAD methods.
Signed-off-by: Herbert Xu
---
net/ipv6/esp6.c | 197 ++
This patch makes use of the new AEAD interface which uses a single
SG list instead of separate lists for the AD and plain text.
Signed-off-by: Herbert Xu
---
crypto/testmgr.c | 84 +++
1 file changed, 48 insertions(+), 36 deletions(-)
diff
You have won contact Allen On allemwilliam10...@gmail.com for info.
Allen Williams
+27612909541
This message may contain confidential information. If you are not the intended
re
On Thu, 2015-05-21 at 18:44 +0800, Herbert Xu wrote:
> This patch makes use of the new AEAD interface which uses a single
> SG list instead of separate lists for the AD and plain text.
Looks fine - want me to run any tests on it?
johannes
--
To unsubscribe from this list: send the line "unsubscr
On Tue, May 19, 2015 at 08:55:17PM -0400, Dave Jones wrote:
> After improving setsockopt() coverage in trinity, I started triggering
> vmalloc failures pretty reliably from this code path:
>
> warn_alloc_failed+0xe9/0x140
> __vmalloc_node_range+0x1be/0x270
> vzalloc+0x4b/0x50
> __do_replace+0x52/0
On 20.05.2015 02:59, Mahesh Bandewar wrote:
On Thu, May 14, 2015 at 6:56 AM, Konstantin Khlebnikov
wrote:
All ipvlan ports use one MAC address, this way ipv6 RA tries to assign
one ipv6 address to all of them. This patch assigns unique dev_id to each
ipvlan port. This field is used instead of c
On Thu, May 21, 2015, at 11:51, Konstantin Khlebnikov wrote:
> On 20.05.2015 02:33, Mahesh Bandewar wrote:
> > On Thu, May 14, 2015 at 6:56 AM, Konstantin Khlebnikov
> > wrote:
> >>
> >> ipvlan_start_xmit() is called with rcu_read_lock_bh() while its internal
> >> structures requre normal rcu_read
On Thu, May 21, 2015 at 01:20:49PM +0200, Johannes Berg wrote:
> On Thu, 2015-05-21 at 18:44 +0800, Herbert Xu wrote:
> > This patch makes use of the new AEAD interface which uses a single
> > SG list instead of separate lists for the AD and plain text.
>
> Looks fine - want me to run any tests on
Hi Takashi,
>> The data is cached in RAM. More specifically, the former loaded
>> firmware files are reloaded and saved at suspend for each device
>> object. See fw_pm_notify() in firmware_class.c.
>
> OK, this may be a stupid idea, but do we know the firmware
> was succ
On Thu, May 21, 2015, at 13:38, Konstantin Khlebnikov wrote:
> On 20.05.2015 02:59, Mahesh Bandewar wrote:
> > On Thu, May 14, 2015 at 6:56 AM, Konstantin Khlebnikov
> > wrote:
> >> All ipvlan ports use one MAC address, this way ipv6 RA tries to assign
> >> one ipv6 address to all of them. This pa
Signed-off-by: Geert Uytterhoeven
---
drivers/net/ethernet/cisco/enic/enic_clsf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/cisco/enic/enic_clsf.c
b/drivers/net/ethernet/cisco/enic/enic_clsf.c
index 6739ebc08c473f7a..a31b57adcb372917 100644
--- a/dr
Hi Dave,
This series further enhances the support for mlx4 single ported VFs
introduced in 3.15 to work over IB ports too.
Just as quick reminder, the ConnectX3 device family exposes one PCI device
which serves both ports.
This can be non-optimal under virtualization schemes where the admin
Single port VFs always provide port = 1 (even if the actual physical
port used is port 2). As such, we need to convert the port provided
by the VF to the physical port before calling into the firmware.
It turns out that the Linux mlx4 VF RoCE driver maintains a copy of
the GID table and hence this
When multiplexling a MAD sent from VF, we should convert the port used
by the guest to send the packet to the actual physical port which will be
used to transmit the packet, before building the relevant address-handle (AH).
This is needed under VPI for single ported VFs, since the code that builds
Some VF drivers flow set the schedule queue in the QP context but
without setting none of OPTPAR_SCHED_QUEUE or OPTPAR_PRIMARY_ADDR_PATH.
To allow for such non-modified drivers to function as single ported
IB VFs, we must adjust the schedule queue port whenever being set,
e.g as currently done for
It's legal for drivers to provide the QP port through the
QPC schedule-queue field on the reset-to-init QP state change.
Add adjusting of the schedule queue port in the SRIOV wrapper
for that operation too.
Signed-off-by: Or Gerlitz
Signed-off-by: Jack Morgenstein
---
.../net/ethernet/mellanox
Remove the limitation that disallows configuring single ported VFs
in the presence of IB ports, after addressing the issues that
prevented that to work.
SMI (QP0) requests/responses are still not supported for single
ported IB VFs.
Signed-off-by: Or Gerlitz
Signed-off-by: Jack Morgenstein
---
As part of enabling single ported VFs over IB ports we need to handle
some of the flows for generting EQ events for VFs which don't come
into play under Eth ports.
This mainly includes port management events derived from changes of the
phyiscal port (lid change, client re-register, down/up, etc),
On Thu, 2015-05-21 at 19:50 +0800, Herbert Xu wrote:
> On Thu, May 21, 2015 at 01:20:49PM +0200, Johannes Berg wrote:
> > On Thu, 2015-05-21 at 18:44 +0800, Herbert Xu wrote:
> > > This patch makes use of the new AEAD interface which uses a single
> > > SG list instead of separate lists for the AD
Am Donnerstag, 21. Mai 2015, 18:39:39 schrieb Herbert Xu:
Hi Herbert,
>Hi:
>
>This series of patches convert all in-tree AEAD users that I
>could find to the new single SG list interface. For IPsec it
>also adopts the new explicit IV generator scheme.
>
>To recap, the old AEAD interface takes an
At Thu, 21 May 2015 14:07:11 +0200,
Marcel Holtmann wrote:
>
> Hi Takashi,
>
> >> The data is cached in RAM. More specifically, the former loaded
> >> firmware files are reloaded and saved at suspend for each device
> >> object. See fw_pm_notify() in firmware_class.c.
> >
> >>>
On Wed, May 20, 2015 at 05:52:34PM -0700, Eric Dumazet wrote:
> On Wed, 2015-05-20 at 17:48 -0700, Rick Jones wrote:
> > On 05/20/2015 05:37 PM, Eric Dumazet wrote:
> >
> > > Anyway, if we can send tcp data at 100Gbits on one flow, I guess we are
> > > doing a terrific job and do not need to tweak
> It remains to clean up ib_cm's ib_cm_listen interface now that
> compare_data isn't used, but I'm not sure this belongs in this series.
This patch series is changing the behavior that the compare data solves.
Currently, the ib_cm handles all of the multiplexing for the rdma_cm -- that's
the r
On Wed, May 20, 2015 at 02:04:02PM +0200, Arnd Bergmann wrote:
> On Wednesday 20 May 2015 06:52:03 Suravee Suthikulanit wrote:
> > On 5/20/2015 5:01 AM, Catalin Marinas wrote:
> > > On Fri, May 15, 2015 at 04:23:09PM -0500, Suravee Suthikulpanit wrote:
> > >> +static inline bool acpi_dma_is_support
From: Hiroshi Shimamoto
Introduce module parameter "disable_hw_vlan_filter" to disable HW VLAN
filter on ixgbe module load.
>From the hardware limitation, there are only 64 VLAN entries for HW VLAN
filter, and it leads to limit the number of VLANs up to 64 among PF and
VFs. For SDN/NFV case, we
Yuval Mintz writes:
> Regardless, I'll give it a more thorough review tomorrow.
> [If those are all the "problems" we'll find with it, I don't think we'll
> need to re-spin this once more; that is, unless Dave insists]
As a follow up, here is a new version that fixes the style issues that Yuval
On Thursday 21 May 2015 14:01:16 Catalin Marinas wrote:
> On Wed, May 20, 2015 at 02:04:02PM +0200, Arnd Bergmann wrote:
> > On Wednesday 20 May 2015 06:52:03 Suravee Suthikulanit wrote:
> > >
> > > Ok. That seems to be what Arnd would prefer as well. Let's just leave
> > > the support for _CCA=
On 05/20/15 20:26, Florian Westphal wrote:
Jamal points out that this header also contains kernel internal magic that
cannot be used from userspace for anything meaningful.
Lets remove what the kernel doesn't use anymore and wrap remainder with
__KERNEL__.
Suggested-by: Jamal Hadi Salim
Sugges
On 05/21/15 03:07, Uwe Kleine-König wrote:
On Wed, May 20, 2015 at 05:30:40PM -0700, Eric Dumazet wrote:
On Wed, 2015-05-20 at 16:46 -0700, Cong Wang wrote:
There is very little to do on ingress side since there is no queue at all,
not to mention priority, you could try ifb to see if it fits y
Nicholas Mc Guire writes:
> On Thu, 14 May 2015, Kalle Valo wrote:
>
>> Nicholas Mc Guire writes:
>>
>> > Patch is against 4.1-rc3 (localversion-next is -next-20150514)
>>
>> BTW, this info should be under "---" line so that it doesn't get stored
>> to the actual commit log.
>>
> that is where
On Thu, 2015-05-21 at 09:07 +0200, Uwe Kleine-König wrote:
> On Wed, May 20, 2015 at 05:30:40PM -0700, Eric Dumazet wrote:
> > On Wed, 2015-05-20 at 16:46 -0700, Cong Wang wrote:
> >
> > > There is very little to do on ingress side since there is no queue at all,
> > > not to mention priority, you
On Thu, 2015-05-21 at 11:39 +0200, Hannes Frederic Sowa wrote:
> Signed-off-by: Hannes Frederic Sowa
> ---
> v2:
> * remove now unused variable i in ip_append_page (thanks, Cong!)
> * switched to EXPORT_SYMBOL_GPL for skb_append_pagefrags
>
> include/linux/skbuff.h | 3 +++
> net/core/skbuff.c
On Thu, 2015-05-21 at 11:39 +0200, Hannes Frederic Sowa wrote:
> This patch implements sendpage support for AF_UNIX SOCK_STREAM
> sockets. This is also required for a complete splice implementation.
>
> The implementation is a bit tricky because we append to already existing
> skbs and so have to
On Thu, 21 May 2015, Takashi Iwai wrote:
> Then avoiding the failed firmware is no solution, indeed.
> If it's a new probe, it should be never executed during resume.
Can you expand this comment? What's wrong with probing during resume?
The USB stack does carry out probes during resume under ce
On Thu, 2015-05-21 at 09:47 +0200, Jetchko Jekov wrote:
> I am confused by this reply, sorry.
> What does it mean?
> Is my fix correct one or its just workaround which doesn't fix root
> cause of the problem? And if is not, what would be the correct way?
> Because the current state ip command f
On Thu, 2015-05-21 at 11:39 +0200, Hannes Frederic Sowa wrote:
> unix_stream_recvmsg is refactored to unix_stream_read_generic in this
> patch and enhanced to deal with pipe splicing. The refactoring is
> inneglible, we mostly have to deal with a non-existing struct msghdr
> argument.
>
> +static
Hi Alan,
>> Then avoiding the failed firmware is no solution, indeed.
>> If it's a new probe, it should be never executed during resume.
>
> Can you expand this comment? What's wrong with probing during resume?
>
> The USB stack does carry out probes during resume under certain
> circumstances.
On 21/05/15 00:33, Cong Wang wrote:
> On Wed, May 20, 2015 at 2:10 PM, Alex Gartrell wrote:
>> Hey everyone,
>>
>> tl;dr; pure python generic netlink library with simple clients for ipvs and
>> taskstats here: https://github.com/facebook/gnlpy
> libnl should have python support for generic netlink
On 05/20/2015 09:50 PM, Todd Bezenek wrote:
I'm pushing 10,000 frames per second of UDP traffic through a Linux
system with a bridge configured between two 1GbE ports.
Iptables is installed and running, but the default rule is ACCEPT with
no other rules.
When I make the packets 1179 bytes in si
On Thu, May 21, 2015, at 16:23, Eric Dumazet wrote:
> On Thu, 2015-05-21 at 11:39 +0200, Hannes Frederic Sowa wrote:
> > unix_stream_recvmsg is refactored to unix_stream_read_generic in this
> > patch and enhanced to deal with pipe splicing. The refactoring is
> > inneglible, we mostly have to deal
This series implements zerocopy support for AF_UNIX SOCK_STREAM sockets.
Changelog in the specific patches. Thanks to all the reviewers!
Hannes Frederic Sowa (4):
net: skbuff: add skb_append_pagefrags and use it
net: af_unix: implement stream sendpage support
net: make skb_splice_bits more
unix_stream_recvmsg is refactored to unix_stream_read_generic in this
patch and enhanced to deal with pipe splicing. The refactoring is
inneglible, we mostly have to deal with a non-existing struct msghdr
argument.
Signed-off-by: Hannes Frederic Sowa
---
v2:
* checked that CONFIG_UNIX=m still wor
Signed-off-by: Hannes Frederic Sowa
Acked-by: Eric Dumazet
---
v2:
* remove now unused variable i in ip_append_page (thanks, Cong!)
* switched to EXPORT_SYMBOL_GPL for skb_append_pagefrags
v3:
* unchanged
include/linux/skbuff.h | 3 +++
net/core/skbuff.c | 18 ++
net/ipv4
This patch implements sendpage support for AF_UNIX SOCK_STREAM
sockets. This is also required for a complete splice implementation.
The implementation is a bit tricky because we append to already existing
skbs and so have to hold unix_sk->readlock to protect the reading side
from either advancing
Prepare skb_splice_bits to be able to deal with AF_UNIX sockets.
AF_UNIX sockets don't use lock_sock/release_sock and thus we have to
use a callback to make the locking and unlocking configureable.
Signed-off-by: Hannes Frederic Sowa
Acked-by: Eric Dumazet
---
v2:
* no changes
v3:
* no changes
>From 68f0dda45a4f1bca070849f3f04a409de897ee0f Mon Sep 17 00:00:00 2001
From: Jetchko Jekov
Date: Thu, 21 May 2015 16:32:24 +0200
Subject: [PATCH] Raising the size of the buffer holding nl messages.
Now it matches the size for the answer defined in rtnl_talk()
and prevents stack corruption wi
At Thu, 21 May 2015 10:18:08 -0400 (EDT),
Alan Stern wrote:
>
> On Thu, 21 May 2015, Takashi Iwai wrote:
>
> > Then avoiding the failed firmware is no solution, indeed.
> > If it's a new probe, it should be never executed during resume.
>
> Can you expand this comment? What's wrong with probing
On Thu, 21 May 2015, Marcel Holtmann wrote:
> Hi Alan,
>
> >> Then avoiding the failed firmware is no solution, indeed.
> >> If it's a new probe, it should be never executed during resume.
> >
> > Can you expand this comment? What's wrong with probing during resume?
> >
> > The USB stack does
At Thu, 21 May 2015 11:26:17 -0400 (EDT),
Alan Stern wrote:
>
> On Thu, 21 May 2015, Takashi Iwai wrote:
>
> > At Thu, 21 May 2015 10:18:08 -0400 (EDT),
> > Alan Stern wrote:
> > >
> > > On Thu, 21 May 2015, Takashi Iwai wrote:
> > >
> > > > Then avoiding the failed firmware is no solution, ind
On 5/20/15, 10:46 PM, Scott Feldman wrote:
On Tue, May 19, 2015 at 1:28 PM, David Miller wrote:
From: Andy Gospodarek
Date: Tue, 19 May 2015 15:47:32 -0400
Are you actually saying that if users complain loudly enough about
the current behavior (not the change Roopa has proposed) that you
wou
Hi Bjorn,
I'm going to check this and will get back to you with
input/questions/resolution.
Thanks,
Yanir
> -Original Message-
> From: Intel-wired-lan [mailto:intel-wired-lan-boun...@lists.osuosl.org] On
> Behalf Of Bjorn Helgaas
> Sent: Wednesday, May 20, 2015 22:48
> To: Yinghai Lu;
From: Herbert Xu
Date: Thu, 21 May 2015 18:39:39 +0800
> This series of patches convert all in-tree AEAD users that I
> could find to the new single SG list interface. For IPsec it
> also adopts the new explicit IV generator scheme.
No objections on my end.
I assume since the dependencies exis
Fix in send of emac regs dump to ethtool which
causing in wrong data interpretation on ethtool
layer for MII and EMAC.
Signed-off-by: Ivan Mikhaylov
---
drivers/net/ethernet/ibm/emac/core.c | 16 ++--
drivers/net/ethernet/ibm/emac/core.h |7 ++-
2 files changed, 8 insertion
From: Hiroshi Shimamoto
Date: Thu, 21 May 2015 13:10:49 +
> diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
> b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
> index 263cb40..b45570f 100644
> --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
> +++ b/drivers/net/ethernet/intel/ix
In ibm_emac.c in ethtool size of emac structure which passing through to driver
is nailed down and not correlating with current emac_regs structure.
Signed-off-by: Ivan Mikhaylov
---
ibm_emac.c | 143 +---
1 file changed, 107 insertions(+)
On 05/20/2015 01:59 AM, Alexei Starovoitov wrote:
introduce bpf_tail_call(ctx, &jmp_table, index) helper function
which can be used from BPF programs like:
int bpf_prog(struct pt_regs *ctx)
{
...
bpf_tail_call(ctx, &jmp_table, index);
...
}
that is roughly equivalent to:
int bpf_prog(str
On Tue, May 19, 2015 at 5:18 PM, Alexei Starovoitov wrote:
> On 5/19/15 5:13 PM, Andy Lutomirski wrote:
>>
>>
>> IMO this is starting to get a bit ugly. Would it be possible to have
>> the program dereference the subprogram reference itself from the jump
>> table? There would have to be a verifi
1 - 100 of 210 matches
Mail list logo
