From: Sunil Goutham
Adjusted nicvf structure such that all elements used in hot
path like napi, xmit e.t.c fall into same cache line. This reduced
no of cache misses and resulted in ~2% increase in no of packets
handled on a core.
Also modified elements with :1 notation to boolean, to be
consist
From: Sunil Goutham
Instead of calling get_page() for every receive buffer carved out
of page, set page's usage count at the end, to reduce no of atomic
calls.
Signed-off-by: Sunil Goutham
---
drivers/net/ethernet/cavium/thunder/nic.h |1 +
drivers/net/ethernet/cavium/thunder/nicv
From: Sunil Goutham
Below patches attempts to improve performance by reducing
no of atomic operations while allocating new receive buffers
and reducing cache misses by adjusting nicvf structure elements.
Sunil Goutham (2):
net: thunderx: Set recevie buffer page usage count in bulk
net: thund
On Fri, Mar 4, 2016 at 4:35 PM, Doug Ledford wrote:
[...]
> So, at *best*, the solution you are suggesting for existing setups is a
> partial solution that leaves things only half done.
[...]
Doug,
Point/s taken, justifying the code re-route of libvirt attempting to
issue set_vf_mac on the PF t
I know that in Germany a good amount of land-line telephone line are
still using ISDN. Some telco company try to move people to IP only, but
this is currently still a process.
Especially company line are using ISDN still, and there are some Linux
programs that act on then, e.g. Asterisk and derive
This patch fixes the checkpatch.pl error to netlabel_domainhash.c:
ERROR: do not initialise statics to NULL
Signed-off-by: Wei Tang
---
net/netlabel/netlabel_domainhash.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/netlabel/netlabel_domainhash.c
b/net/netlabel/n
This patch fixes the checkpatch.pl error to netlabel_unlabeled.c:
ERROR: do not initialise statics to 0 or NULL
Signed-off-by: Wei Tang
---
net/netlabel/netlabel_unlabeled.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/net/netlabel/netlabel_unlabeled.c
b/net/netlab
This fix is for dsmark similar to commit 3557619f0f6f7496ed453d4825e249
("net_sched: prio: use qdisc_dequeue_peeked")
and makes use of qdisc_dequeue_peeked() instead of direct dequeue() call.
First time, wrr peeks dsmark, which will then peek into sfq.
sfq dequeues an skb and it's stored in sch->g
I will upload new patch soon.
Kyeong
On Sun, 2016-03-06 at 22:49 -0500, David Miller wrote:
> From: Kyeong Yoo
> Date: Thu, 3 Mar 2016 10:25:19 +
...
> Your patch is corrupted by your email client, it has MS-DOS style
> newlines, etc.
From: Jon Maloy
Date: Thu, 3 Mar 2016 14:23:21 -0500
> Until now, we have kept a pre-allocated protocol message header
> aggregated into struct tipc_link. Apart from adding unnecessary
> footprint to the link instances, this requires extra code both to
> initialize and re-initialize it.
>
> We
From: Parthasarathy Bhuvaragan
Date: Thu, 3 Mar 2016 17:54:54 +0100
> commit 4d5cfcba2f6e ('tipc: fix connection abort during subscription
> cancel'), removes the check for a valid subscription before calling
> tipc_nametbl_subscribe().
>
> This will lead to a nullptr exception when we process a
From: Colin King
Date: Thu, 3 Mar 2016 13:43:34 +
> From: Colin Ian King
>
> priv->mdio->irq used to be allocated and required freeing, but it
> is now a fixed sized array and should no longer be free'd.
>
> Issue detected using static analysis with CoverityScan
>
> Fixes: e7f4dc3536a400
From: Colin King
Date: Thu, 3 Mar 2016 13:47:18 +
> From: Colin Ian King
>
> priv->mdio->irq used to be allocated and required freeing, but it
> is now a fixed sized array and should no longer be free'd.
>
> Issue detected using static analysis with CoverityScan
>
> Fixes: e7f4dc3536a400
From: Richard Alpe
Date: Thu, 3 Mar 2016 14:20:43 +0100
> Make sure the user has provided a scope for multicast and link local
> addresses used locally by a UDP bearer.
>
> Signed-off-by: Richard Alpe
> Acked-by: Jon Maloy
> Reviewed-by: Erik Hugne
Applied.
From: Phil Sutter
Date: Thu, 3 Mar 2016 14:34:14 +0100
> From: Phil Sutter
>
> Before calling the destroy() or target() callbacks, the family parameter
> field has to be initialized. Otherwise at least the LOG target will
> refuse to work and upon removal oops the kernel.
>
> Cc: Jamal Hadi S
From: Richard Alpe
Date: Thu, 3 Mar 2016 14:20:42 +0100
> The netlink policy for TIPC_NLA_UDP_LOCAL and TIPC_NLA_UDP_REMOTE
> is of type binary with a defined length. This causes the policy
> framework to threat the defined length as maximum length.
>
> There is however no protection against a u
From: Colin King
Date: Thu, 3 Mar 2016 13:27:56 +
> From: Colin Ian King
>
> Used to be allocated and required freeing, but now
> priv->mdio->irq is now a fixed sized array and should no longer be
> free'd.
>
> Issue detected using static analysis with CoverityScan
>
> Fixes: e7f4dc3536a
From: Richard Alpe
Date: Thu, 3 Mar 2016 14:20:41 +0100
> Make sure we have a link before checking if it has been reset or not.
>
> Prior to this patch tipc_link_is_reset() could be called with a non
> existing link, resulting in a null pointer dereference.
>
> Signed-off-by: Richard Alpe
> Ac
From: Richard Alpe
Date: Thu, 3 Mar 2016 14:20:40 +0100
> Prior to this patch enabling a IPv4 UDP bearer caused a null pointer
> dereference in iptunnel_xmit_stats(), when it tried to dereference the
> net device from the skb. To resolve this we now point the skb device
> to the net device resolv
From: Kyeong Yoo
Date: Thu, 3 Mar 2016 10:25:19 +
> This fix is for dsmark similar to commit 3557619f0f6f7496ed453d4825e249
> ("net_sched: prio: use qdisc_dequeue_peeked")
> and makes use of qdisc_dequeue_peeked() instead of direct dequeue() call.
>
> First time, wrr peeks dsmark, which will
From: Bernhard Walle
Date: Thu, 3 Mar 2016 10:15:55 +0100
> From the perspective of RESET, the meaning of the new property
> is actually "active high". Thanks for Troy Kisky for pointing
> that out.
>
> Since the patch is in linux-next, this patch is incremental and doesn't
> replace the origin
From: kbuild test robot
Date: Thu, 3 Mar 2016 09:55:19 +0800
> drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c:115:15-21: ERROR:
> application of sizeof to pointer
>
> sizeof when applied to a pointer typed expression gives the size of
> the pointer
>
> Generated by: scripts/coccinelle
From: Atsushi Nemoto
Date: Thu, 3 Mar 2016 09:07:51 +0900
> Enable workaround for MPC8548E erratum eTSEC 106,
> "Excess delays when transmitting TOE=1 large frames".
> (see commit 53fad77375ce "gianfar: Enable eTSEC-20 erratum w/a
> for P2020 Rev1")
>
> This erratum was fixed in Rev 3.1.x.
>
>
From: Cong Wang
Date: Wed, 2 Mar 2016 11:20:36 -0800
> Cc: Jamal Hadi Salim
> Signed-off-by: Cong Wang
Applied, thanks.
From: Andy Gospodarek
Date: Wed, 2 Mar 2016 11:43:06 -0500
> Any change to sysctl net.ipv4.conf.all.ignore_routes_with_linkdown does
> not result in a change to all interfaces on the system. This means that
> any devices initialized before sysctl settings are applied on boot do
> not see a chan
Hi Pablo,
please consider these IPVS fixes for v4.5 or
if it is too late please consider them for v4.6.
* Arnd Bergman has corrected an error whereby the SIP persistence engine
may incorrectly access protocol fields
* Julian Anastasov has corrected a problem reported by Jiri Bohac with the
co
From: Julian Anastasov
Jiri Bohac is reporting for a problem where the attempt
to reschedule existing connection to another real server
needs proper redirect for the conntrack used by the IPVS
connection. For example, when IPVS connection is created
to NAT-ed real server we alter the reply direct
From: Arnd Bergmann
ip_vs_fill_iph_skb_off() may not find an IP header, and gcc has
determined that ip_vs_sip_fill_param() then incorrectly accesses
the protocol fields:
net/netfilter/ipvs/ip_vs_pe_sip.c: In function 'ip_vs_sip_fill_param':
net/netfilter/ipvs/ip_vs_pe_sip.c:76:5: error: 'iph.pro
From: Julian Anastasov
"RFC 5961, 4.2. Mitigation" describes a mechanism to request
client to confirm with RST the restart of TCP connection
before resending its SYN. As result, IPVS can see SYNs for
existing connection in CLOSE state. Add check to allow
rescheduling in this state.
Signed-off-by
From: Marco Angaroni
The IPVS SIP persistence engine is not able to parse the SIP header
"Call-ID" when such header is inserted in the first positions of
the SIP message.
When IPVS is configured with "--pe sip" option, like for example:
ipvsadm -A -u 1.2.3.4:5060 -s rr --pe sip -p 120 -o
some pa
Hi Herbert,
Today's linux-next merge of the crypto tree got a conflict in:
net/rxrpc/rxkad.c
between commit:
0d12f8a4027d ("rxrpc: Keep the skb private record of the Rx header in host
byte order")
from the net-next tree and commit:
1afe593b4239 ("rxrpc: Use skcipher")
from the crypto
this test calls bpf programs from different contexts:
from inside of slub, from rcu, from pretty much everywhere,
since it kprobes all spin_lock functions.
It stresses the bpf hash and percpu map pre-allocation,
deallocation logic and call_rcu mechanisms.
User space part adding more stress by walki
move ksym search from offwaketime into library to be reused
in other tests
Signed-off-by: Alexei Starovoitov
---
samples/bpf/bpf_load.c | 62 ++
samples/bpf/bpf_load.h | 6
samples/bpf/offwaketime_user.c | 67 +
map creation is typically the first one to fail when rlimits are
too low, not enough memory, etc
Make this failure scenario more verbose
Signed-off-by: Alexei Starovoitov
---
samples/bpf/bpf_load.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/samples/bpf/bpf_load.c b/s
if kprobe is placed within update or delete hash map helpers
that hold bucket spin lock and triggered bpf program is trying to
grab the spinlock for the same bucket on the same cpu, it will
deadlock.
Fix it by extending existing recursion prevention mechanism.
Note, map_lookup and other tracing he
performance tests for hash map and per-cpu hash map
with and without pre-allocation
Signed-off-by: Alexei Starovoitov
---
samples/bpf/Makefile | 4 +
samples/bpf/map_perf_test_kern.c | 100 +
samples/bpf/map_perf_test_user.c | 155 +++
note old loader is compatible with new kernel.
map_flags are optional
Signed-off-by: Alexei Starovoitov
---
samples/bpf/bpf_helpers.h | 1 +
samples/bpf/bpf_load.c | 3 ++-
samples/bpf/fds_example.c | 2 +-
samples/bpf/libbpf.c| 5 +++--
samples/bpf/libbpf.h| 2 +-
If kprobe is placed on spin_unlock then calling kmalloc/kfree from
bpf programs is not safe, since the following dead lock is possible:
kfree->spin_lock(kmem_cache_node->lock)...spin_unlock->kprobe->
bpf_prog->map_update->kmalloc->spin_lock(of the same kmem_cache_node->lock)
and deadlocks.
The fol
extend test coveraged to include pre-allocated and run-time alloc maps
Signed-off-by: Alexei Starovoitov
---
samples/bpf/test_maps.c | 10 +-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/samples/bpf/test_maps.c b/samples/bpf/test_maps.c
index af02f7518c0a..d1e63f48e39c 10
Hi,
this path set switches bpf hash map to use pre-allocation by default
and introduces BPF_F_NO_PREALLOC flag to keep old behavior for cases
where full map pre-allocation is too memory expensive.
Some time back Daniel Wagner reported crashes when bpf hash map is
used to compute time intervals be
Introduce simple percpu_freelist to keep single list of elements
spread across per-cpu singly linked lists.
/* push element into the list */
void pcpu_freelist_push(struct pcpu_freelist *, struct pcpu_freelist_node *);
/* pop element from the list */
struct pcpu_freelist_node *pcpu_freelist_pop(s
On Wed, Mar 2, 2016 at 2:04 AM, Guo-Fu Tseng wrote:
> Dear Diego and who concerns about this driver:
>
> Thanks for reporting it. I can still provide some help, but it's just I'm
> quite busy.
> So that you have not seem me sending patch for a long time.
>
> I wrote this driver when I wan an inte
Le 05/03/2016 14:09, Sergei Shtylyov a écrit :
> Hello.
>
>I have a need to de-assert the active-low PHY hardware reset signal
> (mapped to a GPIO) before the MDIO bus scansince it's left asserted by
> the bootloader (U-Boot). I have a device tree probed MAX driver (ravb)
> and I'm somewhat at
On dim., 2016-03-06 at 13:52 +0800, Xin Long wrote:
> On Sun, Mar 6, 2016 at 12:42 AM, Eric Dumazet wrote:
> > On sam., 2016-03-05 at 23:59 +0800, Xin Long wrote:
> >
> > Are you sure idr_alloc(... GFP_KERNEL) makes sense inside spin_lock_bh()
> > section ?
> >
> > idr_alloc() has :
> >
> > might_
On 03/06/2016 09:56 PM, Stephen Hemminger wrote:
> On Thu, 3 Mar 2016 16:19:34 +0100
> Nikolay Aleksandrov wrote:
>
>> +static int is_temp_mcast_rtr(__u8 type)
>> {
>> +if (type == MDB_RTR_TYPE_TEMP_QUERY || type == MDB_RTR_TYPE_TEMP)
>> +return 1;
>> +else
>> +r
On Thu, 3 Mar 2016 16:19:34 +0100
Nikolay Aleksandrov wrote:
> +static int is_temp_mcast_rtr(__u8 type)
> {
> + if (type == MDB_RTR_TYPE_TEMP_QUERY || type == MDB_RTR_TYPE_TEMP)
> + return 1;
> + else
> + return 0;
> +}
> +
Minor style issues, why not much simpl
-CLEMENT/API-set-for-HW-Buffer-management/20160306-064411
config: arm-allmodconfig (attached as .config)
reproduce:
wget
https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross
-O ~/bin/make.cross
chmod +x ~/bin/make.cross
# save the attached
On Wed, 2 Mar 2016 16:56:27 +0100
Phil Sutter wrote:
> Since the relevant code (and it's bugs) is identical in both files, fix
> them in one go. This patch fixes multiple issues:
>
> * Using 'int' for the 'tdiff' variable does not suffice on 64bit
> systems, the assigned initial time differen
On Mon, 22 Feb 2016 15:26:13 +0100
Jiri Pirko wrote:
> From: Elad Raz
>
> Mark MDB entries which are offloaded to HW with "offload" flag
>
> Signed-off-by: Elad Raz
> Signed-off-by: Jiri Pirko
> ---
> bridge/mdb.c | 5 +++--
> 1 file changed, 3 insertions(+), 2 deletions(-)
Applied, took s
On Sat, Mar 05, 2016 at 11:38:20PM +0100, Gregory CLEMENT wrote:
> From: Marcin Wojtas
>
> Since mvneta driver supports using hardware buffer management (BM), in
> order to use it, board files have to be adjusted accordingly. This commit
> enables BM on:
> * A385-DB-AP - each port has its own poo
Signed-off-by: Moritz Fischer
---
drivers/net/ethernet/cadence/macb.c | 63 -
drivers/net/ethernet/cadence/macb.h | 1 +
2 files changed, 43 insertions(+), 21 deletions(-)
diff --git a/drivers/net/ethernet/cadence/macb.c
b/drivers/net/ethernet/cadence/macb.c
Hi all,
I'm trying to make the dt 'fixed-link' work with the Zynq's GEM on my board.
The endgoal will be to hook it up to an ethernet switch in the PL (FPGA) logic.
Does the attached patch look halfway reasonable? Note the code is pretty hacky,
and it's not 100% working, so I'm more looking for p
Hi Gregory,
> +int hwbm_pool_add(struct hwbm_pool *bm_pool, unsigned int buf_num, gfp_t gfp)
> +{
> + int err, i;
> + unsigned long flags;
> +
> + spin_lock_irqsave(&bm_pool->lock, flags);
> + if (bm_pool->buf_num == bm_pool->size) {
'size' field is used as a 'frag_size'
From: Eric Dumazet
If final packet (ACK) of 3WHS is lost, it appears we do not properly
account the following incoming segment into tcpi_segs_in
While we are at it, starts segs_in with one, to count the SYN packet.
We do not yet count number of SYN we received for a request sock, we
might add t
On Sun, Mar 06, 2016 at 08:23:12AM -0800, Eric Dumazet wrote:
> On dim., 2016-03-06 at 13:09 +0300, Cyrill Gorcunov wrote:
>
> > Anyway, then I run this script for 255 as parameter
> > in one pass which gen. requests to create 65025 addresses
> > and kernel start complaining:
> >
> > Perf output
From: nick Sent: Sunday, March 06, 2016 11:50 AM
> To: Fugang Duan ; da...@davemloft.net
> Cc: b38...@freescale.com; and...@lunn.ch; fabio.este...@freescale.com;
> l.st...@pengutronix.de; rmk+ker...@arm.linux.org.uk; trem...@gmail.com;
> johan...@sipsolutions.net; u.kleine-koe...@pengutronix.de;
>
On dim., 2016-03-06 at 13:09 +0300, Cyrill Gorcunov wrote:
> Anyway, then I run this script for 255 as parameter
> in one pass which gen. requests to create 65025 addresses
> and kernel start complaining:
>
> Perf output
> ---
> 24.95% [kernel] [k] __local_bh_enabl
Please rewrite your title to be without (...).
On Fri, Mar 04, 2016 at 04:41:14PM +0800, Wei Hu(Xavier) wrote:
> It added hns_dsaf_roce_reset routine for roce driver.
> RoCE is a feature of hns.
> In hip06 SOC, in roce reset process, it's needed to configure
> dsaf channel reset,port and sl map in
On Fri, Mar 04, 2016 at 04:41:13PM +0800, Wei Hu(Xavier) wrote:
> The Hisilicon Network Substem(hns) is a long term evolution IP which is
> supposed to be used in Hisilicon ICT SoC. RoCE is a feature of hns.
> The driver for Hisilicon RoCE engine is a platform driver.
> The driver will support muli
On Fri, Mar 04, 2016 at 04:41:17PM +0800, Wei Hu(Xavier) wrote:
> This submit add Kconfig.
It needs be part of previous patch.
>
> Signed-off-by: Wei Hu(Xavier)
> Signed-off-by: oulijun
^^^.. -- Is this first name or second name?
> ---
> drivers/infiniband/Kconfig |
Hello.
On 3/6/2016 10:01 AM, Michael Chan wrote:
The supported bits and advertising bits in ethtool have the same
definitions. The same is true for the firmware bits. So use the
common function to handle the conversion for both supported and
advertising bits.
Signed-off-by: Michael Chan
---
On 03/03/2016 04:38 PM, Ramesh Shanmugasundaram wrote:
> Changes since v1:
> * Removed testmodes & debugfs code (suggested by Oliver H)
> * Fixed tx path race issue by introducing lock (suggested by Marc K)
> * Removed __maybe_unused attribute of rcar_canfd_of_table
>
> obj-$
Hello Nicholas,
On Fri, Mar 04, 2016 at 03:00:14PM -0500, Nicholas Krause wrote:
> This fixes error checking in the function fec_enet_init to properly
> check if the internal call to the function fec_enet_alloc_queue fails
> and if so immediately return the error code to the caller for it to
> han
François,
Thank you for providing information about rose_rebuild_header history.
I was not able to find Jonathan Naylor G4KLX email.
However as rose_xmit() has been recently changed by Eric Biederman, he
may have more precise information about rose_route_frame()
NULL argument.
Meanwhile, I browse
On Sat, Mar 05, 2016 at 09:44:59PM +0300, Cyrill Gorcunov wrote:
> On Sat, Mar 05, 2016 at 11:33:12AM -0500, David Miller wrote:
> ...
> >
> > Probably the same optimization can be applied there, see patch below.
> > And if that doesn't do it, there is a really easy way to batch the
> > delete by
On Fri, Mar 04, 2016 at 09:01:39AM -0800, John Fastabend wrote:
> On 16-03-03 06:55 AM, Amir Vadai wrote:
> > This patch is based on a patch made by John Fastabend.
> > It adds support for offloading cls_flower.
> > when NETIF_F_HW_TC is on:
> > flags = 0 => Rule will be processed twice - b
On Fri, Mar 04, 2016 at 10:20:18AM -0800, Cong Wang wrote:
> On Thu, Mar 3, 2016 at 11:51 AM, Amir Vadai wrote:
> > On Thu, Mar 03, 2016 at 09:45:28AM -0800, Cong Wang wrote:
> >>
> >> So why?
> > The struct will not be used, and without exposing it, the compiler will
> > complain on code like I h
67 matches
Mail list logo
