From: Lothar Waßmann Sent: Monday, November 30, 2015
2:56 PM
> To: Duan Fugang-B38611
> Cc: Andrew Lunn; David S. Miller; Estevam Fabio-R49496; Kevin Hao; Lucas
> Stach; Philippe Reynes; Russell King; Uwe Kleine-K?nig; linux-
> ker...@vger.kernel.org; netdev@vger.kernel.org; Stefan Agner
> Subjec
From: Amit Khatri
txrate variable might be NULL and passing inside function
without NULL check.
Signed-off-by: Amit Khatri
Signed-off-by: Rahul Jain
---
drivers/net/wireless/mac80211_hwsim.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/wireless/mac80211_hwsim.c
b/driver
Hi,
> From: Lothar Waßmann Sent: Friday, November 27,
> 2015 9:39 PM
> > To: Andrew Lunn; David S. Miller; Estevam Fabio-R49496; Kevin Hao; Lothar
> > Waßmann; Lucas Stach; Duan Fugang-B38611; Philippe Reynes; Russell King;
> > Uwe Kleine-König; linux-ker...@vger.kernel.org; netdev@vger.kernel.o
On 11/26/2015 11:56 AM, Alexander Duyck wrote:
> I am not saying you cannot modify the drivers, however what you are
doing is far too invasive. Do you seriously plan on modifying all of
the PCI device drivers out there in order to allow any device that
might be direct assigned to a port to suppo
On Mon, Nov 23, 2015 at 12:02:44AM +0100, Francois Romieu wrote:
> Jonathan Woithe :
> [...]
> > I'm a little confused though as to which patch you want me to apply. There
> > was an inline patch against r8169.c in your message, and then there was
> > another patch to r8169.c in the form of an at
>>> there was a work by David Decotigny that should have solved the out
>>> of bits problem here [1]. Maybe it should be revived.
>>>
>>> [1] https://lkml.org/lkml/2015/1/26/882
>>
>> Yes, it should.
>
> A repost would strongly facilitate that.
>
> Just if anyone ever thinks something is being ig
Hi,
> On Fri, Nov 27, 2015 at 02:39:10PM +0100, Lothar Waßmann wrote:
> > When ENET_OUT is being used as reference clock for an external PHY,
> > the clock must not be disabled while the PHY is active. Otherwise the
> > PHY may lose its internal state and require a reset to become
> > functional a
From: Luuk Paulussen
Date: Mon, 30 Nov 2015 04:10:43 +
> On 11/30/2015 02:58 PM, David Miller wrote:
>> If you guys, really anyone, can find a way to remove some other 32-bit
>> item from sk_buff, you can expand skb->mark to 64-bits. But otherwise,
>> I'm going to be strongly against it. sk
On 11/30/2015 02:58 PM, David Miller wrote:
> If you guys, really anyone, can find a way to remove some other 32-bit
> item from sk_buff, you can expand skb->mark to 64-bits. But otherwise,
> I'm going to be strongly against it. sk_buff is already enormous and
> larger than it should be. So I'm
Dmitry provided a syzkaller (http://github.com/google/syzkaller)
triggering a fault in sock_wake_async() when async IO is requested.
Said program stressed af_unix sockets, but the issue is generic
and should be addressed in core networking stack.
The problem is that by the time sock_wake_async()
This patch is a cleanup to make following patch easier to
review.
Goal is to move SOCK_ASYNC_NOSPACE and SOCK_ASYNC_WAITDATA
from (struct socket)->flags to a (struct socket_wq)->flags
to benefit from RCU protection in sock_wake_async()
To ease backports, we rename both constants.
Two new helpers
From: Ameen
Date: Wed, 25 Nov 2015 23:55:26 +0200
> if (csum)
> - tx_cmd_b |= TX_CMD_B_CSUM_ENABLE;
> - cpu_to_le32s(&tx_cmd_b);
> - memcpy(skb->data, &tx_cmd_b, 4);
> + tx_cmds.cmd_b |= TX_CMD_B_CSUM_ENABLE;
You've corrupted the indentation here.
--
To unsubscrib
From: Paolo Abeni
Date: Wed, 25 Nov 2015 18:05:27 +0100
> After 614732eaa12d, no refcount is mainteined for the vport-vxlan module.
> This allows the userspace to remove such module while vport-vxlan
> devices still exist, which lead to later oops, i.e. when
> the openvswitch module is removed.
>
From: Eric Dumazet
Date: Wed, 25 Nov 2015 09:02:10 -0800
> From: Eric Dumazet
>
> Once TX has been enabled on a NIC, it is illegal to access skb,
> as this skb might have been freed by another cpu, from TX completion
> handler.
>
> Signed-off-by: Eric Dumazet
> Reported-by: Mark Rutland
> Te
From: Bert Kenward
Date: Wed, 25 Nov 2015 16:06:14 +
> if (!(nic_data->datapath_caps &
> - (1 << MC_CMD_GET_CAPABILITIES_OUT_TX_TSO_LBN))) {
> - netif_err(efx, drv, efx->net_dev,
Can you please keep code properly indented...
> +
> + if (nic_data->datapath_cap
From: LABBE Corentin
Date: Wed, 25 Nov 2015 14:44:41 +0100
> The simple_strtol function is obsolete.
> This patch replace it by kstrtoint.
> This will simplify code, since some error case not handled by
> simple_strtol are handled by kstrtoint.
>
> Signed-off-by: LABBE Corentin
kstrtoint() act
From: Anup Limbu
Date: Wed, 25 Nov 2015 15:37:21 +0530
> replace kmalloc + memset with kmemdup
>
> Signed-off-by: Anup Limbu
I agree with Bjorn that this microturfing cleanup misses
the real higher level opportunity to use the usbnet_read_cmd()
and usbnet_write_cmd() helpers that already exist
From: Kalle Valo
Date: Wed, 25 Nov 2015 10:59:43 +0200
> here's first wireless-drivers pull request for 4.5. I'll start the cycle
> with a big one as I'm adding vendor directories to wireless drivers.
> This patch has been on linux-next since Monday and I haven't seen any
> reports of breakage so
From: Eric Dumazet
This patch addresses multiple problems :
UDP/RAW sendmsg() need to get a stable struct ipv6_txoptions
while socket is not locked : Other threads can change np->opt
concurrently. Dmitry posted a syzkaller
(http://github.com/google/syzkaller) program desmonstrating
use-after-fre
From: Jason Wang
Date: Wed, 25 Nov 2015 15:11:26 +0800
> This series tries to add basic busy polling for vhost net. The idea is
> simple: at the end of tx/rx processing, busy polling for new tx added
> descriptor and rx receive socket for a while. The maximum number of
> time (in us) could be spe
On Sat, 2015-11-28 at 09:23 -0800, Eric Dumazet wrote:
> On Sat, 2015-11-28 at 09:11 -0800, Eric Dumazet wrote:
> > On Sat, 2015-11-28 at 12:00 +0100, Dmitry Vyukov wrote:
> > > Hello,
> > >
> > > The following program triggers use-after-free in ip6_setup_cork:
> > >
> > > // autogenerated by syz
From: Tom Herbert
Date: Tue, 24 Nov 2015 09:32:11 -0800
>>
>> FWIW, I've brought the issue to the attention of the architects here,
>> and we will likely be able to make changes in this space. Intel
>> hardware (as demonstrated by your patches) already is able to deal with
>> this de-ossificatio
From: Tom Herbert
Date: Mon, 23 Nov 2015 13:53:44 -0800
> The bad effect of this model is that it is encourages HW vendors to
> continue implement HW protocol specific support for encapsulations, we
> get so much more benefit if they implement protocol generic
> mechanisms.
+1
--
To unsubscribe
From: Martin Blumenstingl
Date: Sun, 22 Nov 2015 17:46:09 +0100
> Commit 9c7077622dd91 ("packet: make packet_snd fail on len smaller
> than l2 header") added validation for the packet size in packet_snd.
> This change enforces that every packet needs a header (with at least
> hard_header_len byte
From: Ben Hutchings
Date: Mon, 30 Nov 2015 00:40:54 +
> On Sun, 2015-11-29 at 23:39 +0200, Or Gerlitz wrote:
>> On Sun, Nov 29, 2015 at 3:53 PM, Yuval Mintz wrote:
>> > [I suspect there was probably some discussion over this,
>> > but I couldn't find it; References would be welcome]
>>
>> t
Lorenzo Colitti wrote:
> On Mon, Nov 30, 2015 at 11:08 AM, Florian Westphal wrote:
> > We have ctnetlink which allows direct setting of ctmark/ctlabels.
>
> Can those labels be set early enough in the stack to affect the source
> address chosen for outgoing connections?
No, same restrictions as
From: Andrew Lunn Sent: Sunday, November 29, 2015 12:44 AM
> To: Duan Fugang-B38611
> Cc: Lothar Wa?mann; David S. Miller; Estevam Fabio-R49496; Kevin Hao;
> Lucas Stach; Philippe Reynes; Russell King; Uwe Kleine-K?nig; linux-
> ker...@vger.kernel.org; netdev@vger.kernel.org; Stefan Agner
> Subjec
On Mon, Nov 30, 2015 at 11:08 AM, Florian Westphal wrote:
>
> > While connlabels provide 128bits of space skb->mark is still only 32
> > bits. Since we are using connection tracking to simply restore skb->mark
> > the use of connlabels by itself doesn't solve the problem I outlined
> > above. skb-
Matt Bennett wrote:
> On Tue, 2015-11-24 at 21:36 +0100, Florian Westphal wrote:
> > Matt Bennett wrote:
> > > I'm emailing this list for feedback on the feasibility of increasing
> > > skb->mark or adding a new field for marking. Perhaps this extension
> > > could be done under a new CONFIG opti
On Fri, Nov 27, 2015 at 02:39:10PM +0100, Lothar Waßmann wrote:
> When ENET_OUT is being used as reference clock for an external PHY,
> the clock must not be disabled while the PHY is active. Otherwise the
> PHY may lose its internal state and require a reset to become
> functional again.
>
> A sy
From: Marcin Wojtas
Date: Sun, 29 Nov 2015 14:21:35 +0100
>> Looking at your patches, it was not entirely clear to me how the buffer
>> manager on these Marvell SoCs work, but other networking products have
>> something similar, like Broadcom's Cable Modem SoCs (BCM33xx) FPM, and
>> maybe Freesca
> Do you mean PHY switch also use the enet_out clock ?
Yes, the Marvell switches i've have connected to a Vybrid use enet_out
clock. There was a recent change to IMX pinctrl which broke the muxing
for ENET_OUT on Vydrid, which broke the probing of these switches. It
was no longer possible to mux t
From: Lorenzo Colitti
Date: Sun, 29 Nov 2015 17:37:11 +0900
> On Wed, Nov 25, 2015 at 5:32 AM, Matt Bennett
> wrote:
>> I'm emailing this list for feedback on the feasibility of increasing
>> skb->mark or adding a new field for marking. Perhaps this extension
>> could be done under a new CONFIG
Hi Sebastian,
On 11/27/15 8:00 AM, Sebastian Andrzej Siewior wrote:
It looks like an attempt to use CPU notifier here which was never
completed. Nobody tried to wire it up completely since 2k9. So I unwind
this code and get rid of everything not required. Oh look! 19 lines were
removed while cod
From: Alexei Starovoitov
For large map->value_size the user space can trigger memory allocation warnings
like:
WARNING: CPU: 2 PID: 11122 at mm/page_alloc.c:2989
__alloc_pages_nodemask+0x695/0x14e0()
Call Trace:
[< inline >] __dump_stack lib/dump_stack.c:15
[] dump_stack+0x68/0x92 lib/
On Mon, Nov 30, 2015 at 12:14:21AM +0300, Sergei Shtylyov wrote:
> Hello.
>
> On 11/29/2015 10:26 PM, Geert Uytterhoeven wrote:
>
> >>Add support for yet another ARM member of the R-Car family, R-Car M2, also
>
> >R-Car M2-W?
>
>Right, forgot about the postfixes.
>
> >>known as R8A7791.
>
On Sun, 2015-11-29 at 23:39 +0200, Or Gerlitz wrote:
> On Sun, Nov 29, 2015 at 3:53 PM, Yuval Mintz wrote:
> > [I suspect there was probably some discussion over this,
> > but I couldn't find it; References would be welcome]
>
> there was a work by David Decotigny that should have solved the out
On zo, 2015-11-29 at 21:26 +0100, Paul Bolle wrote:
> If the above is correct it would be nice to know the .config of the
> kernel used by syzkaller.
>
> Anyhow, without further details of the chain of events that triggered
> this warning, I'm afraid it will be hard to determine which struct
> tim
Hi!
By spawning new network and user namesapces an unprivileged user
is able to execute /sbin/bridge-stp within the initial mount namespace
with global root rights.
While this cannot directly be used to break out of a container or gain
global root rights it could be used by exploit writers as valu
Hello.
On 11/28/2015 5:51 PM, Pavel Machek wrote:
atl1c driver is doing order-4 allocation with GFP_ATOMIC
priority. That often breaks networking after resume. Switch to
GFP_KERNEL. Still not ideal, but should be significantly better.
Signed-off-by: Pavel Machek
diff --git a/drivers/net/eth
On Sun, 2015-11-29 at 17:40 +0900, Lorenzo Colitti wrote:
> On Thu, Nov 26, 2015 at 1:58 AM, Radu Rendec wrote:
> > I tracked down the problem to ip_route_input_slow() in net/ipv4/route.c
> > [...]
> > The next problem is that packets are seen as "martians" and dropped,
> > most probably because o
On Sun, Nov 29, 2015 at 3:53 PM, Yuval Mintz wrote:
> [I suspect there was probably some discussion over this,
> but I couldn't find it; References would be welcome]
there was a work by David Decotigny that should have solved the out of
bits problem here [1]. Maybe it should be revived.
[1] http
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On Sun, Nov 29, 2015 at 12:07:52PM -0800, Stephen Hemminger wrote:
> On Thu, 26 Nov 2015 14:26:05 +0100
> Phil Sutter wrote:
>
> > This macro aims to simplify most netlink users' pattern to prepare a
> > request, which is to create an unnamed struct
Hello.
On 11/29/2015 10:26 PM, Geert Uytterhoeven wrote:
Add support for yet another ARM member of the R-Car family, R-Car M2, also
R-Car M2-W?
Right, forgot about the postfixes.
known as R8A7791.
There's also R-Car M2-N, aka R8A7793, but you probably know that ;-)
Will fix.
This will probably earn me a reputation as the most single-minded
monomaniac on this planet (insofar there's still anything to earn in
this respect) but this issue has been irking me "ever since".
NB: This is somewhat loser formatted than a proper patch submission in
order to explain the backgroun
On zo, 2015-11-29 at 19:47 +0100, Tilman Schmidt wrote:
> Btw I don't see a timer_list object in struct platform_device either.
> Nor in the embedded struct device.
I found two instances of struct timer_list, rather deep down struct
ser_cardstate:
struct ser_cardstate {
struct platform_de
On Thu, 26 Nov 2015 14:26:05 +0100
Phil Sutter wrote:
> This macro aims to simplify most netlink users' pattern to prepare a
> request, which is to create an unnamed struct and initialize it:
>
> | struct {
> | struct nlmsghdr n;
> | struct whatever foo;
> | char buf[arbitrary number
On Thu, 26 Nov 2015 15:38:46 +0100
Daniel Borkmann wrote:
> I've added three examples to examples/bpf/ that demonstrate how one can
> implement eBPF tail calls in tc with f.e. multiple levels of nesting.
> That should act as a good starting point, but also as test cases for the
> ELF loader and k
On Fri, 27 Nov 2015 10:23:43 -0800
Tom Herbert wrote:
> This patch adds support to remote checksum checksum offload
> to VXLAN. This patch adds remcsumtx and remcsumrx to ip vxlan
> configuration to enable remote checksum offload for transmit
> and receive on the VXLAN tunnel.
>
> https://tools.
On Sat, 28 Nov 2015 01:00:00 +0100
Phil Sutter wrote:
> The primary goal was to get rid of the -Wunused-result warnings emitted
> during compiling. While adding the necessary checks, I found a few
> functions which could benefit from a bigger review (patches 1, 2 and 3).
> Patch 4 then adds the r
On Tue, 24 Nov 2015 15:30:59 +0100
Phil Sutter wrote:
> This patch series aims to silence spurious error messages when flushing
> an interface with many addresses assigned and provides a few cleanups
> found along the way.
>
> The original issue was a test-case adding 40k IPv4 addresses to an
>
Hi Sergei,
On Fri, Nov 27, 2015 at 11:06 PM, Sergei Shtylyov
wrote:
> Add support for yet another ARM member of the R-Car family, R-Car M2, also
R-Car M2-W?
> known as R8A7791.
There's also R-Car M2-N, aka R8A7793, but you probably know that ;-)
Gr{oetje,eeting}s,
Ge
Am 29.11.2015 um 19:22 schrieb Peter Hurley:
>>> [ 413.536749] WARNING: CPU: 6 PID: 25400 at lib/debugobjects.c:263
>>> debug_print_object+0x1c4/0x1e0()
>>> [ 413.538111] ODEBUG: free active (active state 0) object type: timer_list
>>> hint: delayed_work_timer_fn+0x0/0x90
>>
>> This message se
Am 29.11.2015 um 19:22 schrieb Peter Hurley:
> On 11/29/2015 10:30 AM, Tilman Schmidt wrote:
>>
>> Judging from the backtrace below this must be the call
>>
>> kfree(cs->hw.ser);
>>
>> in drivers/isdn/gigaset/ser-gigaset.c line 375.
>> cs->hw.ser is of type struct ser_cardstate *.
>> struct
On Sun, Nov 29, 2015 at 02:18:29PM +0100, Dmitry Vyukov wrote:
> ca.key_size = 1;
> ca.value_size = 0xfff9;
> ca.max_entries = 10;
> int fd = syscall(SYS_bpf, BPF_MAP_CREATE, &ca, sizeof(ca));
...
> [ cut here ]
> WARNING: CPU: 2 PID: 1112
Hi Tilman,
On 11/29/2015 10:30 AM, Tilman Schmidt wrote:
> Hi Sasha,
>
> thanks for the report. As the original author of the code in question, I
> am somewhat at a loss what to make of it.
>
> Am 27.11.2015 um 16:19 schrieb Sasha Levin:
>> Fuzzing with syzkaller on the latest -next kernel produ
> If I didn't know that though, my expectations would be that the
> Fair Queue packet scheduler would behave just like its name says, fair. :P
>
Only if the flows have same characteristics. This cannot be ensured for
hours long flows. This should be obvious.
If one flow is not trying to push en
On 11/29/2015 6:27 PM, Julia Lawall wrote:
Please check line 442. It seems clear that match_c was intended.
Sure, will fix that and run more static checkers before submitting V2, seems
like checkpatch, smatch and sparse are not enough... thanks for pointing
this out.
Or.
--
To unsubscribe f
drivers/net/ethernet/mellanox/mlx5/core/en_main.c:1970:2-3: Unneeded semicolon
drivers/net/ethernet/mellanox/mlx5/core/en_main.c:1959:2-3: Unneeded semicolon
Remove unneeded semicolon.
Generated by: scripts/coccinelle/misc/semicolon.cocci
CC: Saeed Mahameed
Signed-off-by: Fengguang Wu
---
Hi Saeed,
[auto build test WARNING on: net-next/master]
[also build test WARNING on: v4.4-rc2 next-20151127]
url:
https://github.com/0day-ci/linux/commits/Or-Gerlitz/Introducing-ConnectX-4-Ethernet-SRIOV/20151129-234342
coccinelle warnings: (new ones prefixed by >>)
>>
https://github.com/0day-ci/linux/commits/Or-Gerlitz/Introducing-ConnectX-4-Ethernet-SRIOV/20151129-234342
> :: branch date: 40 minutes ago
> :: commit date: 40 minutes ago
>
> >> drivers/net/ethernet/mellanox/mlx5/core/eswitch.c:442:5-13: duplicated
> >> argumen
From: Saeed Mahameed
Implement set VF mac/link state and query VF config
to be used later in nedev VF ndos or any other management API.
Signed-off-by: Saeed Mahameed
Signed-off-by: Or Gerlitz
---
drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 61 +++
drivers/net/ether
From: Saeed Mahameed
Implement and enable SR-IOV ndos to manage SR-IOV configuration via
netdev netlink API.
Signed-off-by: Saeed Mahameed
Signed-off-by: Or Gerlitz
---
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 84 ++-
1 file changed, 83 insertions(+), 1 deletion
From: Saeed Mahameed
Each Vport/vNIC must notify underlying e-Switch layer
for vlan table changes in-order to update SR-IOV FDB tables.
We do that at vlan_rx_add_vid and vlan_rx_kill_vid ndos.
Signed-off-by: Saeed Mahameed
Signed-off-by: Or Gerlitz
---
drivers/net/ethernet/mellanox/mlx5/core
From: Saeed Mahameed
E-Switch is the software entity that represents and manages ConnectX4
inter-HCA ethernet l2 switching.
E-Switch has its own Virtual Ports, each Vport/vNIC/VF can be
connected to the device through a vport of an e-switch.
Each e-switch is managed by one vNIC identified by
HC
From: Saeed Mahameed
Those functions are needed to notify the upcoming SR-IOV
E-Switch(FDB) manager(PF), of the NIC vport (vf) promisc mode changes.
Preperation for ethernet sriov and l2 table management.
Signed-off-by: Saeed Mahameed
Signed-off-by: Or Gerlitz
---
drivers/net/ethernet/mellan
From: Saeed Mahameed
Each Vport/vNIC must notify underlying e-Switch layer
for UC/MC list and promisc mode updates, in-order to update
l2 tables and SR-IOV FDB tables.
We do that at set_rx_mode ndo.
preperation for ethernet-SRIOV and l2 table management.
Signed-off-by: Saeed Mahameed
Signed-o
From: Saeed Mahameed
Enabling E-Switch SRIOV for nvfs+1 vports.
Create E-Switch FDB for L2 UC/MC mac steering between VFs/PF and
external vport (Uplink).
FDB contains forwarding rules such as:
UC MAC0 -> vport0(PF).
UC MAC1 -> vport1.
UC MAC2 -> vport2.
MC MACX -
From: Saeed Mahameed
E-Switch vport context is unlike NIC vport context, managed by the
E-Switch manager or vport_group_manager and not by the NIC(VF) driver.
The E-Switch manager can access (read/modify) any of its vports
E-Switch context.
Currently E-Switch vport context includes only clietnt
From: Saeed Mahameed
Add query and modify functions to control client vlan and qos
striping or insertion, in E-Switch vports contexts.
Signed-off-by: Saeed Mahameed
Signed-off-by: Or Gerlitz
---
drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 134 +-
drivers/net/ethern
From: Saeed Mahameed
Define needed hardware structures and capabilities needed
for E-Switch FDB flow tables and read them on driver load.
Signed-off-by: Saeed Mahameed
Signed-off-by: Or Gerlitz
---
drivers/net/ethernet/mellanox/mlx5/core/fw.c | 13 +
include/linux/mlx5/device.h
From: Saeed Mahameed
In preparation for SR-IOV we add here an API to enable each e-switch
client (PF/VF) to configure its L2 MAC addresses and for the e-switch
manager (usually the PF) to access them in order to be able to
configure them into the e-switch.
Therefore we now pass vport num paramete
From: Saeed Mahameed
Those functions are needed to notify the upcoming L2 table and SR-IOV
E-Switch(FDB) manager(PF), of the NIC vport (vf) vlan table changes.
preperation for ethernet sriov and l2 table management.
Signed-off-by: Saeed Mahameed
Signed-off-by: Or Gerlitz
---
drivers/net/ethe
From: Saeed Mahameed
Those functions are needed to notify the upcoming L2 table and SR-IOV
E-Switch(FDB) manager(PF), of the NIC vport (vf) UC/MC mac lists
changes.
preperation for ethernet sriov and l2 table management.
Signed-off-by: Saeed Mahameed
Signed-off-by: Or Gerlitz
---
drivers/net
From: Saeed Mahameed
Add support to get VF statistics using query vport
counter command.
Signed-off-by: Saeed Mahameed
Signed-off-by: Or Gerlitz
---
drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 67 +++
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 3 +
2 file
From: Saeed Mahameed
In preparation for SR-IOV we add here an API to enable each e-switch
manager (PF) to configure its VFs link states in e-switch
preparation for ethernet sriov.
Signed-off-by: Saeed Mahameed
Signed-off-by: Or Gerlitz
---
drivers/net/ethernet/mellanox/mlx5/core/en_main.c |
From: Eli Cohen
This patch adds SRIOV base support for mlx5 supported devices. The same
driver is used for both PFs and VFs; VFs are identified by the driver
through the flag MLX5_PCI_DEV_IS_VF added to the pci table entries.
Virtual functions are created as usual through writing a value to the
s
Hi Dave,
This patchset introduces the support of Ethernet SRIOV in ConnectX-4
family of 100G Ethernet NICs.
Some features are still missing, but all the basic SRIOV functionalities
are there already.
Basic Introduction:
ConnectX-4 HW architecture provides two kinds of underlying HW switches.
M
From: Eli Cohen
Modify these functions to have func_id argument to state which device we
are referring to. This is done as a preparation for SRIOV support where
a PF driver needs to control its virtual functions.
Signed-off-by: Eli Cohen
Signed-off-by: Or Gerlitz
---
drivers/net/ethernet/mell
From: Saeed Mahameed
Update HCA capabilities and HW struct to include needed
capabilities for upcoming Ethernet Switch (SR-IOV E-Switch).
Signed-off-by: Saeed Mahameed
Signed-off-by: Or Gerlitz
---
include/linux/mlx5/mlx5_ifc.h | 26 +++---
1 file changed, 15 insertions(+)
Hi Sasha,
thanks for the report. As the original author of the code in question, I
am somewhat at a loss what to make of it.
Am 27.11.2015 um 16:19 schrieb Sasha Levin:
> Fuzzing with syzkaller on the latest -next kernel produced this error:
Is there a way to know the actual sequence of events t
On Sat, Nov 28, 2015 at 2:27 AM, Sasha Levin wrote:
> On 11/27/2015 08:20 PM, Peter Hurley wrote:
>> It would really help if you included the syzkaller-generated applet with
>> the bug reports; state previously established by the applet can be
>> crucial in understanding why the call stack looks t
[I suspect there was probably some discussion over this,
but I couldn't find it; References would be welcome]
So it appears we're almost out of bits for new advertised speeds;
I'm currently looking at adding 100g.
Assuming we want [and we do want] to have those in the `supported',
`advertising' a
On Thu, Nov 19, 2015 at 5:40 PM, Or Gerlitz wrote:
> On Sat, Nov 14, 2015 at 8:45 AM, Joe Stringer wrote:
>> On 13 November 2015 at 06:46, Or Gerlitz wrote:
>>> On Fri, Nov 13, 2015 at 10:14 AM, Joe Stringer
>>> wrote:
>
I agree that this UNSPEC issue on v2.3 could do with a bit of a clos
Hi Florian,
>
> Looking at your patches, it was not entirely clear to me how the buffer
> manager on these Marvell SoCs work, but other networking products have
> something similar, like Broadcom's Cable Modem SoCs (BCM33xx) FPM, and
> maybe Freescale's FMAN/DPAA seems to do something similar.
>
>
On 14/11/2015 08:45, Joe Stringer wrote:
> On 13 November 2015 at 06:46, Or Gerlitz wrote:
>> > On Fri, Nov 13, 2015 at 10:14 AM, Joe Stringer
>> > wrote:
>> >
>>> >> I don't follow the logic. You observed one flow which matched on
>>> >> TTL=64, therefore all vxlan packets terminated at OVS hav
Hello,
The following program triggers a WARNING in kmalloc:
// autogenerated by syzkaller (http://github.com/google/syzkaller)
#include
#include
#include
#define SYS_bpf 321
#define BPF_MAP_CREATE0
#define BPF_MAP_UPDATE_ELEM2
union bpf_attr {
This patch introduces register and unregister functionality for lowpan
interfaces. While register a lowpan interface there are several things
which need to be initialize by the 6lowpan subsystem. Upcoming
functionality need to register/unregister per interface components e.g.
debugfs entry.
Signed
This patch adds a static inline function ipv6_addr_prefix_copy which
copies a ipv6 address prefix(argument pfx) into the ipv6 address prefix.
The prefix len is given by plen as bits. This function mainly based on
ipv6_addr_prefix which copies one address prefix from address into a new
ipv6 address
This patch introduce support for IPHC stateful address compression. It
will offer three debugfs per interface entries, which are:
- dci_table: destination context indentifier table
- sci_table: source context indentifier table
- mcast_sci_table: multicast context identifier table
Example to se
This patch will introduce a 6lowpan entry into the debugfs if enabled.
Inside this 6lowpan directory we create a subdirectories of all 6lowpan
interfaces to offer a per interface debugfs support.
Signed-off-by: Alexander Aring
---
include/net/6lowpan.h | 3 +++
net/6lowpan/6lowpan_i.h | 28 ++
Hi,
This patch based on the work of Lukasz Duda. It doesn't contain any interface
lookup methods since we have a generic private dataroom per 6lowpan interface.
Also I dropped the list implementation and do a simple array one. It also
contains support for multicast stateful compression.
To acces
Sun, Nov 29, 2015 at 10:04:07AM CET, gerlitz...@gmail.com wrote:
>[..]
>> +enum mlxsw_hwmon_attr_type {
>> + MLXSW_HWMON_ATTR_TYPE_TEMP,
>> + MLXSW_HWMON_ATTR_TYPE_TEMP_MAX,
>> +};
>> +
>> +static void mlxsw_hwmon_attr_add(struct mlxsw_hwmon *mlxsw_hwmon,
>> +
[..]
> +enum mlxsw_hwmon_attr_type {
> + MLXSW_HWMON_ATTR_TYPE_TEMP,
> + MLXSW_HWMON_ATTR_TYPE_TEMP_MAX,
> +};
> +
> +static void mlxsw_hwmon_attr_add(struct mlxsw_hwmon *mlxsw_hwmon,
> +enum mlxsw_hwmon_attr_type attr_type,
> +
On Thu, Nov 26, 2015 at 1:58 AM, Radu Rendec wrote:
> I tracked down the problem to ip_route_input_slow() in net/ipv4/route.c
> [...]
> The next problem is that packets are seen as "martians" and dropped,
> most probably because of __fib_validate_source() failing due to the
> fact that the input i
On Wed, Nov 25, 2015 at 5:32 AM, Matt Bennett
wrote:
> I'm emailing this list for feedback on the feasibility of increasing
> skb->mark or adding a new field for marking. Perhaps this extension
> could be done under a new CONFIG option.
64-bit marks (both skb->mark and sk->sk_mark) would be usefu
96 matches
Mail list logo
