Hello,
syzbot found the following crash on:
HEAD commit:04c4b677 Add linux-next specific files for 20190513
git tree: linux-next
console output: https://syzkaller.appspot.com/x/log.txt?x=10a413c8a0
kernel config: https://syzkaller.appspot.com/x/.config?x=b8e08a763b62ad3a
Hello Heiner and net-folks,
On Sat, May 11, 2019 at 04:56:56PM +0200, Heiner Kallweit wrote:
> On 11.05.2019 16:46, Vicente Bergas wrote:
> > On Friday, May 10, 2019 10:28:06 PM CEST, Heiner Kallweit wrote:
> >> On 10.05.2019 17:05, Vicente Bergas wrote:
> >>> Hello,
> >>> there is a regression on
Now that the mvpp2 driver supports classification offloading, we must
add the NETIF_F_NTUPLE to the features list.
Since the current code doesn't allow disabling the feature, we don't set
the flag in dev->hw_features.
Fixes: 90b509b39ac9 ("net: mvpp2: cls: Add Classification offload support")
Rep
On Mon, May 13, 2019 at 09:33:13AM +0800, Weilong Chen wrote:
> The remote host answers to an ICMP timestamp request.
> This allows an attacker to know the time and date on your host.
Why is that a problem? If it is, does it also mean that it is a security
problem to have your time in sync (becaus
Andy Duan [2019-05-13 03:31:59]:
> From: Andrew Lunn Sent: Saturday, May 11, 2019 2:18 AM
> > On Fri, May 10, 2019 at 08:24:03AM +, Andy Duan wrote:
> > > If MAC address read from nvmem cell and it is valid mac address,
> > > .of_get_mac_addr_nvmem() add new property "nvmem-mac-address" in
>
On 11/05/2019 15:44, Petr Štetiar wrote:
};
+Another example where we've MAC address for eth1 stored in the NOR EEPROM as
+following sequence of bytes (output of hexdump -C /dev/mtdX):
+
+ 0180 66 61 63 5f 6d 61 63 20 3d 20 44 34 3a 45 45 3a |fac_mac = D4:EE:|
+ 0190 3
From: Petr Štetiar Sent: Monday, May 13, 2019 4:00 PM
> Andy Duan [2019-05-13 03:31:59]:
>
> > From: Andrew Lunn Sent: Saturday, May 11, 2019 2:18
> > AM
> > > On Fri, May 10, 2019 at 08:24:03AM +, Andy Duan wrote:
> > > > If MAC address read from nvmem cell and it is valid mac address,
> >
From: Simon Huelck
Date: Sat, May 11, 2019 at 15:53:34
> ethtool -S gave me some counts for mmc_rx_fifo_overflow, which i didnt
> recognize before.
Flow Control can prevent this to happen. Please check if your DT FIFO
bindings are >= 4096.
> Do we have new ideas / new direction to dig for ?
G
Srinivas Kandagatla [2019-05-13 09:25:55]:
Hi,
> My initial idea was to add compatible strings to the cell so that most of
> the encoding information can be derived from it. For example if the encoding
> representing in your example is pretty standard or vendor specific we could
> just do with a
Hi Jagdish,
On Thu, May 09, 2019 at 12:01:14AM +0530, Jagdish Motwani wrote:
> From: Jagdish Motwani
>
> In case of more than 1 nf_queues, hooks between them are being executed
> more than once.
This refers to NF_REPEAT, correct?
I think this broke with 960632ece6949. If so, it would be good t
Maxime Ripard [2019-05-12 14:19:10]:
> > @@ -29,6 +31,19 @@ Optional properties:
> > bits: Is pair of bit location and number of bits, which specifies offset
> > in bit and number of bits within the address range specified by reg
> > property.
> > Offset takes values from 0-7.
On Thu, 9 May 2019 at 20:36, Alexei Starovoitov wrote:
>
> On 5/9/19 9:12 AM, Jonathan Lemon wrote:
> > On 9 May 2019, at 4:48, Björn Töpel wrote:
> >
> >> On Thu, 9 May 2019 at 01:07, Jonathan Lemon
> >> wrote:
> >>>
> >>> Currently, the AF_XDP code uses a separate map in order to
> >>> determin
Andy Duan [2019-05-13 03:38:32]:
> From: Maxime Ripard Sent: Friday, May 10, 2019
> 7:32 PM
> >
> > It looks to me that it should be abstracted away by the nvmem interface and
> > done at the provider level, not the customer.
> >
> If to implement add above features like Petr Štetiar describe
of_find_device_by_node takes a reference to the embedded struct device
which needs to be dropped after use.
Fixes: d01f449c008a ("of_net: add NVMEM support to of_get_mac_address")
Reported-by: kbuild test robot
Reported-by: Julia Lawall
Signed-off-by: Petr Štetiar
---
drivers/of/of_net.c | 5 +
On Thu, Apr 25, 2019 at 09:43:53PM +0800, linmiaohe wrote:
> From: Miaohe Lin
>
> When firewalld is enabled with ipv4/ipv6 rpfilter, vrf
> ipv4/ipv6 packets will be dropped because in device is
> vrf but out device is an enslaved device. So failed with
> the check of the rpfilter.
>
> Signed-off
From: Florian Westphal
If userspace provides a rule blob with trailing data after last target,
we trigger a splat, then convert ruleset to 64bit format (with trailing
data), then pass that to do_replace_finish() which then returns -EINVAL.
Erroring out right away avoids the splat plus unneeded t
From: Florian Westphal
Following splat gets triggered when nfnetlink monitor is running while
xtables-nft selftests are running:
net/netfilter/nf_tables_api.c:1272 suspicious rcu_dereference_check() usage!
other info that might help us debug this:
1 lock held by xtables-nft-mul/27006:
#0:
From: Subash Abhinov Kasiviswanathan
CONFIG_NF_CONNTRACK_IPV6 has been deprecated so replace it with a check
for IPV6 instead.
Use nf_ip6_route6() instead of v6ops->route() and keep the IS_MODULE()
in nf_ipv6_ops as mentioned by Florian so that direct calls are used
when IPV6 is builtin and indi
This is fixing flow offload for UDP traffic where packets only follow
one single direction.
The flow_offload_fixup_tcp() mechanism works fine in case that the
offloaded entry remains in SYN_RECV state, given sequence tracking is
reset and that conntrack handles syn+ack packets as a retransmission,
From: Taehee Yoo
Conntrack entries can be deleted by the masquerade module. In that case,
flow offload should be deleted too, but GC and data-path of flow offload
do not check for conntrack status bits, hence flow offload entries will
be removed only by the timeout.
Update garbage collector and
From: Florian Westphal
When we process a long ruleset of the form
chain input {
type filter hook input priority filter; policy drop;
...
}
Then the base chain gets registered early on, we then continue to
process/validate the next messages coming in the same transaction.
Problem is that
From: Florian Westphal
should be same as NFT_LOGLEVEL_AUDIT, so use -, not +.
Fixes: 7eced5ab5a73 ("netfilter: nf_tables: add NFT_LOGLEVEL_* enumeration and
use it")
Signed-off-by: Florian Westphal
Signed-off-by: Pablo Neira Ayuso
---
include/uapi/linux/netfilter/nf_tables.h | 2 +-
1 file c
From: Jakub Jankowski
Since commit bc7d811ace4a ("netfilter: nf_ct_h323: Convert
CHECK_BOUND macro to function"), NAT traversal for H.323
doesn't work, failing to parse H323-UserInformation.
nf_h323_error_boundary() compares contents of the bitstring,
not the addresses, preventing valid H.323 pac
From: Taehee Yoo
nf_flow_offload_ip_hook() and nf_flow_offload_ipv6_hook() do not check
ttl value. So, ttl value overflow may occur.
Fixes: 97add9f0d66d ("netfilter: flow table support for IPv4")
Fixes: 0995210753a2 ("netfilter: flow table support for IPv6")
Signed-off-by: Taehee Yoo
Signed-off
From: Taehee Yoo
flow_offload_alloc() calls nf_route() to get a dst_entry. Internally,
nf_route() calls ip_route_output_key() that allocates a dst_entry and
holds it. So, a dst_entry should be released by dst_release() if
nf_route() is successful.
Otherwise, netns exit routine cannot be finished
Hi David,
The following patchset contains Netfilter fixes for net:
1) Postpone chain policy update to drop after transaction is complete,
from Florian Westphal.
2) Add entry to flowtable after confirmation to fix UDP flows with
packets going in one single direction.
3) Reference count lea
From: Kristian Evensen
Commit 59c08c69c278 ("netfilter: ctnetlink: Support L3 protocol-filter
on flush") introduced a user-space regression when flushing connection
track entries. Before this commit, the nfgen_family field was not used
by the kernel and all entries were removed. Since this commit
Never used anywhere in the code.
Fixes: 7e0b2b57f01d ("netfilter: nft_ct: add ct timeout support")
Reported-by: Stéphane Veyret
Signed-off-by: Pablo Neira Ayuso
---
include/uapi/linux/netfilter/nf_tables.h | 2 --
1 file changed, 2 deletions(-)
diff --git a/include/uapi/linux/netfilter/nf_tabl
From: Taehee Yoo
rhashtable_insert_fast() may return an error value when memory
allocation fails, but flow_offload_add() does not check for errors.
This patch just adds missing error checking.
Fixes: ac2a5e23 ("netfilter: add generic flow table infrastructure")
Signed-off-by: Taehee Yoo
Sig
On 13/05/2019 10:07, Petr Štetiar wrote:
Srinivas Kandagatla [2019-05-13 09:25:55]:
Hi,
My initial idea was to add compatible strings to the cell so that most of
the encoding information can be derived from it. For example if the encoding
representing in your example is pretty standard or
Hello Vincente,
On Sat, May 11, 2019 at 05:06:04PM +0200, Vicente Bergas wrote:
> On Saturday, May 11, 2019 4:56:56 PM CEST, Heiner Kallweit wrote:
> > On 11.05.2019 16:46, Vicente Bergas wrote:
> > > On Friday, May 10, 2019 10:28:06 PM CEST, Heiner Kallweit wrote:
> > > > On 10.05.2019 17:05, Vic
Hello Guenter,
On Sun, May 12, 2019 at 10:41:32PM -0700, Guenter Roeck wrote:
> Hi,
>
> On Sat, Apr 27, 2019 at 12:21:11AM +0300, Serge Semin wrote:
> > There are two chip pins named TXDLY and RXDLY which actually adds the 2ns
> > delays to TXC and RXC for TXD/RXD latching. Alas this is the only
On Mon, May 13, 2019 at 10:36:51AM +, Jagdish Motwani wrote:
> Hi Pablo,
>
> The case I am referring to is : If there are more than 1 hooks
> returning NF_QUEUE verdict. When the first queue reinjects the
> packet, 'nf_reinject' starts traversing hooks with hook_index (i).
> However if it ag
Hi Pablo,
The case I am referring to is : If there are more than 1 hooks returning
NF_QUEUE verdict.
When the first queue reinjects the packet, 'nf_reinject' starts traversing
hooks with hook_index (i).
However if it again receives a NF_QUEUE verdict (by some other netfilter hook),
it queue wi
On Mon, May 13, 2019 at 01:29:42PM +0300, Serge Semin wrote:
> Hello Vincente,
>
> On Sat, May 11, 2019 at 05:06:04PM +0200, Vicente Bergas wrote:
> > On Saturday, May 11, 2019 4:56:56 PM CEST, Heiner Kallweit wrote:
> > > On 11.05.2019 16:46, Vicente Bergas wrote:
> > > > On Friday, May 10, 2019
The old MIPS implementation of dma_cache_sync() didn't use the dev argument,
but commit c9eb6172c328 ("dma-mapping: turn dma_cache_sync into a
dma_map_ops method") changed that, so we now need to set dev.parent.
Signed-off-by: Thomas Bogendoerfer
---
drivers/net/ethernet/seeq/sgiseeq.c | 1 +
1
Srinivas Kandagatla [2019-05-13 11:06:48]:
> On 13/05/2019 10:07, Petr Štetiar wrote:
> > Srinivas Kandagatla [2019-05-13 09:25:55]:
> >
> > > My initial idea was to add compatible strings to the cell so that most of
> > > the encoding information can be derived from it. For example if the
> >
On 2019/5/13 15:49, Michal Kubecek wrote:
On Mon, May 13, 2019 at 09:33:13AM +0800, Weilong Chen wrote:
The remote host answers to an ICMP timestamp request.
This allows an attacker to know the time and date on your host.
Why is that a problem? If it is, does it also mean that it is a securi
On Mon, May 13, 2019 at 07:38:37PM +0800, Weilong Chen wrote:
>
> On 2019/5/13 15:49, Michal Kubecek wrote:
> > On Mon, May 13, 2019 at 09:33:13AM +0800, Weilong Chen wrote:
> > > The remote host answers to an ICMP timestamp request.
> > > This allows an attacker to know the time and date on your
On 2019/5/13 19:49, Michal Kubecek wrote:
On Mon, May 13, 2019 at 07:38:37PM +0800, Weilong Chen wrote:
On 2019/5/13 15:49, Michal Kubecek wrote:
On Mon, May 13, 2019 at 09:33:13AM +0800, Weilong Chen wrote:
The remote host answers to an ICMP timestamp request.
This allows an attacker to know
459d69c407f9 disabled runtime PM for CNP+ chips. This broke my I219-LM
on a SPT chip. So disable runtime PM also for SPT.
Signed-off-by: Julian Wollrath
---
drivers/net/ethernet/intel/e1000e/netdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/e
On Mon, May 13, 2019 at 08:06:37PM +0800, Weilong Chen wrote:
> On 2019/5/13 19:49, Michal Kubecek wrote:
> > One idea is that there may be applications using current time as a seed
> > for random number generator - but then such application is the real
> > problem, not having correct time.
> >
>
On Thu, May 09, 2019 at 05:32:21PM +0200, Ulrich Hecht wrote:
>
> > On May 9, 2019 at 12:10 PM Simon Horman wrote:
> >
> >
> > On Thu, May 09, 2019 at 08:57:44AM +0200, Ulrich Hecht wrote:
> > >
> > > > On May 8, 2019 at 6:52 PM Niklas Söderlund
> > > > wrote:
> > > >
> > > >
> > > > Hi Se
Weilong Chen wrote:
> On 2019/5/13 15:49, Michal Kubecek wrote:
> > On Mon, May 13, 2019 at 09:33:13AM +0800, Weilong Chen wrote:
> > > The remote host answers to an ICMP timestamp request.
> > > This allows an attacker to know the time and date on your host.
> >
> > Why is that a problem? If it
On Monday, May 13, 2019 12:51:05 PM CEST, Serge Semin wrote:
On Mon, May 13, 2019 at 01:29:42PM +0300, Serge Semin wrote:
Hello Vincente,
On Sat, May 11, 2019 at 05:06:04PM +0200, Vicente Bergas wrote: ...
Hmm, just figured out, that in the datasheet RXDLY/TXDLY pins
are actually grounded, s
On 2019/5/13 20:11, Michal Kubecek wrote:
On Mon, May 13, 2019 at 08:06:37PM +0800, Weilong Chen wrote:
On 2019/5/13 19:49, Michal Kubecek wrote:
One idea is that there may be applications using current time as a seed
for random number generator - but then such application is the real
problem,
On Mon, May 13, 2019 at 08:26:18PM +0800, Weilong Chen wrote:
> On 2019/5/13 20:11, Michal Kubecek wrote:
> > On Mon, May 13, 2019 at 08:06:37PM +0800, Weilong Chen wrote:
> > > On 2019/5/13 19:49, Michal Kubecek wrote:
> > > > One idea is that there may be applications using current time as a seed
Hello,
On Mon, May 13, 2019 at 02:19:17PM +0200, Vicente Bergas wrote:
> On Monday, May 13, 2019 12:51:05 PM CEST, Serge Semin wrote:
> > On Mon, May 13, 2019 at 01:29:42PM +0300, Serge Semin wrote:
> > > Hello Vincente,
> > >
> > > On Sat, May 11, 2019 at 05:06:04PM +0200, Vicente Bergas wrote:
On Mon, May 13, 2019 at 03:41:45PM +0900, Kunihiko Hayashi wrote:
> After calling phy_select_page() and until calling phy_restore_page(),
> the mutex 'mdio_lock' is already locked, so the driver should use
> non-locked version of phy functions. Or there will be a deadlock with
> 'mdio_lock'.
>
> T
> Ahh, I see. Then using lock-less version of the access methods must fix the
> problem. You could try something like this:
Kunihiko Hayash is way ahead of you.
Andrew
The 05/03/2019 16:08, Edward Cree wrote:
> Plumb it through from the flow_dissector.
>
> Signed-off-by: Edward Cree
> ---
> include/net/flow_offload.h | 2 ++
> net/core/flow_offload.c| 7 +++
> 2 files changed, 9 insertions(+)
>
> diff --git a/include/net/flow_offload.h b/include/net/f
On Mon, May 13, 2019 at 11:41:39AM +0200, Petr Štetiar wrote:
> of_find_device_by_node takes a reference to the embedded struct device
> which needs to be dropped after use.
>
> Fixes: d01f449c008a ("of_net: add NVMEM support to of_get_mac_address")
> Reported-by: kbuild test robot
> Reported-by:
On Mon, May 13, 2019 at 02:51:03PM +0200, Andrew Lunn wrote:
> > Ahh, I see. Then using lock-less version of the access methods must fix the
> > problem. You could try something like this:
>
> Kunihiko Hayash is way ahead of you.
>
>Andrew
I wouldn't say that five hours is "way ahead". B
Currently, nla_put_iflink() doesn't put the IFLA_LINK attribute when
iflink == ifindex.
In some cases, a device can be created in a different netns with the
same ifindex as its parent. That device will not dump its IFLA_LINK
attribute, which can confuse some userspace software that expects it.
For
When adding more MAC address to a dwmac-sun8i interface, the device goes
directly in promiscuous mode.
This is due to IFF_UNICAST_FLT missing flag.
So since the hardware support unicast filtering, let's add IFF_UNICAST_FLT.
Fixes: 9f93ac8d4085 ("net-next: stmmac: Add dwmac-sun8i")
Signed-off-by:
The remote host answers to an ICMP timestamp request.
This allows an attacker to know the time and date on your host.
This path is an another way contrast to iptables rules:
iptables -A input -p icmp --icmp-type timestamp-request -j DROP
iptables -A output -p icmp --icmp-type timestamp-reply -j DR
Fix reported build fail:
ERROR: "cpsw_ale_flush_multicast"
[drivers/net/ethernet/ti/keystone_netcp_ethss.ko] undefined!
ERROR: "cpsw_ale_create" [drivers/net/ethernet/ti/keystone_netcp_ethss.ko]
undefined!
ERROR: "cpsw_ale_add_vlan" [drivers/net/ethernet/ti/keystone_netcp_ethss.ko]
undefined!
F
Le 13/05/2019 à 15:01, Sabrina Dubroca a écrit :
> Currently, nla_put_iflink() doesn't put the IFLA_LINK attribute when
> iflink == ifindex.
>
> In some cases, a device can be created in a different netns with the
> same ifindex as its parent. That device will not dump its IFLA_LINK
> attribute, w
Do you see kernel WARNING while running bpf: test_sock test case ?
This kernel warning is popping up continuously for ~725 times on all
devices ( arm64, arm, x86_64 and i386).
selftests: bpf: test_sock
Test case: bind4 load with invalid access: src_ip6 .. [PASS]
Test case: bind4 load with invalid
On 2019/5/13 17:42, Pablo Neira Ayuso wrote:
> On Thu, Apr 25, 2019 at 09:43:53PM +0800, linmiaohe wrote:
>> From: Miaohe Lin
>>
>> When firewalld is enabled with ipv4/ipv6 rpfilter, vrf
>> ipv4/ipv6 packets will be dropped because in device is
>> vrf but out device is an enslaved device. So fa
On 13/05/2019 13:54, Jianbo Liu wrote:
> Could you please push to 5.1 and 5.0-stable? The original patch brought a bug
> in mlx5_core driver. Need your patch to fix.
>
> --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
> +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
> @@ -1615,7 +1615,
On Monday, May 13, 2019 3:01:33 PM CEST, Serge Semin wrote:
On Mon, May 13, 2019 at 02:51:03PM +0200, Andrew Lunn wrote:
Ahh, I see. Then using lock-less version of the access
methods must fix the
problem. You could try something like this:
Kunihiko Hayash is way ahead of you.
Andre
Killer series loadded IWL_22000_HR_B_FW_PRE prefixed firmware instead
IWL_CC_A_FW_PRE prefixed firmware.
Add killer series to the check logic as iwl_ax200_cfg_cc.
Signed-off-by: Cyrus Lien
---
drivers/net/wireless/intel/iwlwifi/pcie/trans.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(
2019-05-13, 15:17:33 +0200, Nicolas Dichtel wrote:
> Le 13/05/2019 à 15:01, Sabrina Dubroca a écrit :
> > Currently, nla_put_iflink() doesn't put the IFLA_LINK attribute when
> > iflink == ifindex.
> >
> > In some cases, a device can be created in a different netns with the
> > same ifindex as its
Currently, nla_put_iflink() doesn't put the IFLA_LINK attribute when
iflink == ifindex.
In some cases, a device can be created in a different netns with the
same ifindex as its parent. That device will not dump its IFLA_LINK
attribute, which can confuse some userspace software that expects it.
For
I tested yesterdays 5.2 devel git and it failed to boot on my Sun Fire V445
(4x UltraSparc III). Init is started and it hangs there:
[ 38.414436] Run /sbin/init as init process
[ 38.530711] random: fast init done
[ 39.580678] systemd[1]: Inserted module 'autofs4'
[ 39.721577] systemd[1]:
Weilong Chen wrote:
> The remote host answers to an ICMP timestamp request.
> This allows an attacker to know the time and date on your host.
No, it does not, I already told you so in V1 :-/
If you really think that its a problem that one can discover
milliseconds-since-midnight please just chan
A couple fixes for sockmap code. Previously this was bundled with a tls
fix for unhash() path however, that is becoming a larger fix so push
these on their own.
---
John Fastabend (3):
bpf: sockmap, only stop/flush strp if it was enabled at some point
bpf: sockmap remove duplicate que
If we try to call strp_done on a parser that has never been
initialized, because the sockmap user is only using TX side for
example we get the following error.
[ 883.422081] WARNING: CPU: 1 PID: 208 at kernel/workqueue.c:3030
__flush_work+0x1ca/0x1e0
...
[ 883.422095] Workqueue: events s
In tcp bpf remove we free the cork list and purge the ingress msg
list. However we do this before the ref count reaches zero so it
could be possible some other access is in progress. In this case
(tcp close and/or tcp_unhash) we happen to also hold the sock
lock so no path exists but lets fix it ot
When converting a skb to msg->sg we forget to set the size after the
latest ktls/tls code conversion. This patch can be reached by doing
a redir into ingress path from BPF skb sock recv hook. Then trying to
read the size fails.
Fix this by setting the size.
Fixes: 604326b41a6fb ("bpf, sockmap: co
Le 13/05/2019 à 15:47, Sabrina Dubroca a écrit :
> Currently, nla_put_iflink() doesn't put the IFLA_LINK attribute when
> iflink == ifindex.
>
> In some cases, a device can be created in a different netns with the
> same ifindex as its parent. That device will not dump its IFLA_LINK
> attribute, w
On 2019-05-08 16:06, Magnus Karlsson wrote:
> On Tue, May 7, 2019 at 4:19 PM Maxim Mikityanskiy
> wrote:
>>
>> On 2019-05-06 17:23, Magnus Karlsson wrote:
>>> On Mon, May 6, 2019 at 3:46 PM Maxim Mikityanskiy
>>> wrote:
On 2019-05-04 20:26, Björn Töpel wrote:
> On Tue, 30 Apr 2019
A recent bug in systemd [1] triggered the following kernel warning:
A link change request failed with some changes committed already.
Interface eth1 may have been left with an inconsistent configuration,
please check.
do_setlink() performs multiple configuration updates, and if any of them
inet6_set_link_af requires that at least one of IFLA_INET6_TOKEN or
IFLA_INET6_ADDR_GET_MODE is passed. If none of them is passed, it
returns -EINVAL, which may cause do_setlink() to fail in the middle of
processing other commands and give the following warning message:
A link change request fai
inet6_set_link_af requires that at least one of IFLA_INET6_TOKEN or
IFLA_INET6_ADDR_GET_MODE is passed. If none of them is passed, it
returns -EINVAL, which may cause do_setlink() to fail in the middle of
processing other commands and give the following warning message:
A link change request fai
On 5/13/19 7:47 AM, Sabrina Dubroca wrote:
> Currently, nla_put_iflink() doesn't put the IFLA_LINK attribute when
> iflink == ifindex.
>
> In some cases, a device can be created in a different netns with the
> same ifindex as its parent. That device will not dump its IFLA_LINK
> attribute, which c
2019-05-13, 16:50:51 +0200, Nicolas Dichtel wrote:
> Le 13/05/2019 à 15:47, Sabrina Dubroca a écrit :
> > Currently, nla_put_iflink() doesn't put the IFLA_LINK attribute when
> > iflink == ifindex.
> >
> > In some cases, a device can be created in a different netns with the
> > same ifindex as its
Le 13/05/2019 à 17:08, Sabrina Dubroca a écrit :
> 2019-05-13, 16:50:51 +0200, Nicolas Dichtel wrote:
>> Le 13/05/2019 à 15:47, Sabrina Dubroca a écrit :
>>> Currently, nla_put_iflink() doesn't put the IFLA_LINK attribute when
>>> iflink == ifindex.
>>>
>>> In some cases, a device can be created in
Le 13/05/2019 à 17:07, David Ahern a écrit :
> On 5/13/19 7:47 AM, Sabrina Dubroca wrote:
>> Currently, nla_put_iflink() doesn't put the IFLA_LINK attribute when
>> iflink == ifindex.
>>
>> In some cases, a device can be created in a different netns with the
>> same ifindex as its parent. That devi
On Sun, 12 May 2019 11:19:45 +0800, Rocky Liao wrote:
> This patch adds an optional device property "firmware-name" to allow the
> driver to load customized nvm firmware file based on this property.
>
> Signed-off-by: Rocky Liao
> ---
> Changes in v4:
> * rebased the code base and merge with la
On Thu, May 9, 2019 at 11:57 AM David Ahern wrote:
>
> From: David Ahern
>
> Add support for timestamping netlink messages. If a socket wants a
> timestamp, it is added when the skb clone is queued to the socket.
>
> Allow userspace to know the actual time an event happened. In a
> busy system th
Hello!
On 05/13/2019 04:06 PM, Corentin Labbe wrote:
> When adding more MAC address to a dwmac-sun8i interface, the device goes
Addresses?
> directly in promiscuous mode.
> This is due to IFF_UNICAST_FLT missing flag.
>
> So since the hardware support unicast filtering, let's add IFF_UNICAS
From: Maxime Chevallier
Date: Mon, 13 May 2019 09:30:33 +0200
> Now that the mvpp2 driver supports classification offloading, we must
> add the NETIF_F_NTUPLE to the features list.
>
> Since the current code doesn't allow disabling the feature, we don't set
> the flag in dev->hw_features.
>
> F
From: Petr Štetiar
Date: Mon, 13 May 2019 11:41:39 +0200
> of_find_device_by_node takes a reference to the embedded struct device
> which needs to be dropped after use.
>
> Fixes: d01f449c008a ("of_net: add NVMEM support to of_get_mac_address")
> Reported-by: kbuild test robot
> Reported-by: Ju
From: Pablo Neira Ayuso
Date: Mon, 13 May 2019 11:56:17 +0200
> The following patchset contains Netfilter fixes for net:
...
> This batch comes with a conflict that can be fixed with this patch:
Thanks for this.
> You can pull these changes from:
>
> git://git.kernel.org/pub/scm/linux/kerne
From: Leonard Zgrablic
Currently there is only a dump version of RTM_GETNEIGH for PF_UNSPEC in
RTNETLINK that dumps neighbor entries, no non-dump version that can be used to
retrieve a single neighbor entry.
Add support for the non-dump (doit) version of RTM_GETNEIGH for PF_UNSPEC so
that a sing
From: Thomas Bogendoerfer
Date: Mon, 13 May 2019 13:15:17 +0200
> The old MIPS implementation of dma_cache_sync() didn't use the dev argument,
> but commit c9eb6172c328 ("dma-mapping: turn dma_cache_sync into a
> dma_map_ops method") changed that, so we now need to set dev.parent.
>
> Signed-off
From: Michal Kubecek
Date: Mon, 13 May 2019 14:11:45 +0200
> I'm sorry but I cannot agree with that. Seeding PRNG with current time
> is known to be a bad practice and if some application does it, the
> solution is to fix the application, not obfuscating system time.
+1 +1 +1
From: Weilong Chen
Date: Mon, 13 May 2019 20:06:37 +0800
> So, the 'time' may become sensitive information. The OS should not
> leak it out.
The current time of day is a globally synchronized value everyone on
the planet has access to.
I don't buy this line of reasoning at all, time is not sens
From: Andrew Lunn
Date: Mon, 13 May 2019 14:50:02 +0200
> On Mon, May 13, 2019 at 03:41:45PM +0900, Kunihiko Hayashi wrote:
>> After calling phy_select_page() and until calling phy_restore_page(),
>> the mutex 'mdio_lock' is already locked, so the driver should use
>> non-locked version of phy fu
From: Jerome Brunet
Date: Sun, 12 May 2019 23:12:37 +0200
> The phy id chosen by Amlogic is incorrectly set in the mdio mux and
> does not match the phy driver.
>
> It was not detected before because DT forces the use the correct driver
> for the internal PHY.
>
> Fixes: 7090425104db ("net: phy
From: Weilong Chen
Date: Mon, 13 May 2019 21:28:57 +0800
> The remote host answers to an ICMP timestamp request.
> This allows an attacker to know the time and date on your host.
>
> This path is an another way contrast to iptables rules:
> iptables -A input -p icmp --icmp-type timestamp-request
From: Nicolas Dichtel
Date: Mon, 13 May 2019 16:50:51 +0200
> Le 13/05/2019 à 15:47, Sabrina Dubroca a écrit :
>> Currently, nla_put_iflink() doesn't put the IFLA_LINK attribute when
>> iflink == ifindex.
>>
>> In some cases, a device can be created in a different netns with the
>> same ifindex
Functionally this patch looks fine, but it has several style things that
need to be fixed.
The Subject line of the mail should be:
[PATCH net-next] getneigh: add nondump to retrieve single entry
Also, your timing is wrong. net-next is still closed.
Since there are multiple style errors, learn t
From: Maxim Mikityanskiy
Date: Mon, 13 May 2019 15:05:28 +
> A recent bug in systemd [1] triggered the following kernel warning:
>
> A link change request failed with some changes committed already.
> Interface eth1 may have been left with an inconsistent configuration,
> please check.
On Sun, May 12, 2019 at 4:52 PM Daniel Borkmann wrote:
>
> On 05/12/2019 03:09 AM, Stanislav Fomichev wrote:
> > On 05/11, Andrii Nakryiko wrote:
> >> On Fri, May 10, 2019 at 3:00 PM Stanislav Fomichev
> >> wrote:
> >>> On 05/10, Andrii Nakryiko wrote:
> On Fri, May 10, 2019 at 2:36 PM Stan
Various things in eBPF really require us to disable preemption
before running an eBPF program.
syzbot reported :
BUG: assuming atomic context at net/core/flow_dissector.c:737
in_atomic(): 0, irqs_disabled(): 0, pid: 24710, name: syz-executor.3
2 locks held by syz-executor.3/24710:
#0: e8
Jarod Wilson wrote:
>On 5/10/19 6:53 PM, Jay Vosburgh wrote:
>> Jarod Wilson wrote:
>>
>>> There's currently a problem with toggling arp_validate on and off with an
>>> active-backup bond. At the moment, you can start up a bond, like so:
>>>
>>> modprobe bonding mode=1 arp_interval=100 arp_valid
From: Jay Vosburgh
Date: Mon, 13 May 2019 09:43:30 -0700
> That would be my preference, as the 29c4948293bf commit looks to
> be the change actually being fixed.
Sorry I pushed the original commit message out :-(
But isn't the Fixes: tag he choose the one where the logic actually
causes p
1 - 100 of 194 matches
Mail list logo
