On Mon, May 28, 2018 at 05:40:18AM +, Fu, Qiaobin wrote:
> The new action inheritdsfield copies the field DS of
> IPv4 and IPv6 packets into skb->priority. This enables
> later classification of packets based on the DS field.
>
> Original idea by Jamal Hadi Salim
>
>
While compile-testing on arm64 with gcc-8.1, I ran into a build diagnostic:
drivers/net/ethernet/freescale/fec_main.c: In function 'fec_probe':
drivers/net/ethernet/freescale/fec_main.c:3517:25: error: '%d' directive
writing between 1 and 10 bytes into a region of size 5
On Sat, May 26, 2018 at 05:39:12AM -0400, Eric S. Raymond wrote:
> I'm trying to untangle some nasty code in the Mills implementation of
> NTP. I could simplify it a lot if there were a way to query a packet
> to find out the name of the network interface it arrived on. (At the
> moment the code
At this point skb->len is including tcp header length, so it is meaningless
to user. data length could be more helpful, with which we can easily filter
out the packet without payload.
Cc: Eric Dumazet
Acked-by: Song Liu
Signed-off-by: Yafang Shao
On 26-May-18 7:53 AM, Jakub Kicinski wrote:
Offload simple RED configurations. For now support only DCTCP
like scenarios where min and max are the same.
Signed-off-by: Jakub Kicinski
---
drivers/net/ethernet/netronome/nfp/abm/main.c | 82 +++
On 28/05/18 18:44, Petr Machata wrote:
> Callers of br_fdb_find() need to hold the hash lock, which
> br_fdb_find_port() doesn't do. Add the missing lock/unlock
> pair.
>
> Signed-off-by: Petr Machata
> ---
> net/bridge/br_fdb.c | 2 ++
> 1 file changed, 2 insertions(+)
>
>
On Sun, May 27, 2018 at 01:42:34PM +0300, Leon Romanovsky wrote:
> From: Yonatan Cohen
>
> The new mini-CQE format includes the stride index, byte count and
> packet checksum.
> Stride index is needed for striding WQ feature.
> This patch exposes this capability and
Callers of br_fdb_find() need to hold the hash lock, which
br_fdb_find_port() doesn't do. Add the missing lock/unlock
pair.
Signed-off-by: Petr Machata
---
net/bridge/br_fdb.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c
On 28/05/18 11:42 AM, Marcelo Ricardo Leitner wrote:
On Mon, May 28, 2018 at 05:40:18AM +, Fu, Qiaobin wrote:
The new action inheritdsfield copies the field DS of
IPv4 and IPv6 packets into skb->priority. This enables
later classification of packets based on the DS field.
Original idea by
The previous code was optimistic, accepting the offload of whole action
chain when there was a single known action (drop/redirect). This results
in offloading a rule which should not be offloaded, because its behavior
cannot be reproduced in the hardware.
For example:
$ tc filter add dev eno1
On Fri, May 25, 2018 at 05:32:52PM -0700, Greg Thelen wrote:
> On Fri, May 25, 2018 at 2:32 PM Arnd Bergmann wrote:
>
> > Several subsystems depend on INFINIBAND_ADDR_TRANS, which in turn depends
> > on INFINIBAND. However, when with CONFIG_INIFIBAND=m, this leads to a
> > link
On 5/28/2018 7:11 PM, Jason Gunthorpe wrote:
On Sun, May 27, 2018 at 01:42:34PM +0300, Leon Romanovsky wrote:
From: Yonatan Cohen
The new mini-CQE format includes the stride index, byte count and
packet checksum.
Stride index is needed for striding WQ feature.
This patch exposes this
On Sun, May 27, 2018 at 01:23:44PM +0300, Leon Romanovsky wrote:
> + if (!mcounters->hw_cntrs_hndl) {
> + mcounters->hw_cntrs_hndl =
> + (void
> *)mlx5_fc_create(to_mdev(ibcounters->device)->mdev,
> +false);
No
syzbot has found a reproducer for the following crash on:
HEAD commit:7a1a98c171ea Merge branch 'bpf-sendmsg-hook'
git tree: bpf-next
console output: https://syzkaller.appspot.com/x/log.txt?x=149ae2b780
kernel config: https://syzkaller.appspot.com/x/.config?x=e4078980b886800c
On Mon, May 28, 2018 at 07:52:03PM +0300, Yishai Hadas wrote:
> On 5/28/2018 7:11 PM, Jason Gunthorpe wrote:
> >On Sun, May 27, 2018 at 01:42:34PM +0300, Leon Romanovsky wrote:
> >>From: Yonatan Cohen
> >>
> >>The new mini-CQE format includes the stride index, byte count and
> >>packet checksum.
On Mon, 28 May 2018 17:44:16 +0200
Petr Machata wrote:
> Callers of br_fdb_find() need to hold the hash lock, which
> br_fdb_find_port() doesn't do. Add the missing lock/unlock
> pair.
>
> Signed-off-by: Petr Machata
> ---
> net/bridge/br_fdb.c | 2 ++
> 1 file changed, 2 insertions(+)
>
>
> On May 26, 2018, at 12:43 PM, John Fastabend wrote:
>
> On 05/25/2018 12:46 PM, Song Liu wrote:
>> On Fri, May 25, 2018 at 11:11 AM, Song Liu wrote:
>>> Summary:
>>>
>>> At the end of sch_direct_xmit(), we are in the else path of
>>> !dev_xmit_complete(ret), which means ret ==
On Sun, May 20, 2018 at 04:26:03PM +0800, Xin Long wrote:
> On Sat, May 19, 2018 at 11:57 PM, Eric Dumazet wrote:
> > SCTP experts, please take a look.
> >
> > On 05/19/2018 08:55 AM, syzbot wrote:
> >> Hello,
> >>
> >> syzbot found the following crash on:
> >>
> >> HEAD commit:73fcb1a370c7
Thanks for clarifying that first ip will be used as primary ip.
I have 2 further queries on this.
1. How can this survive across reboots without having a custom script
on boot up? Like some ifcfg file,etc.
2. is there a way to tell to make a given ip as primary, irrespective of order?
On Mon, May
On Mon, May 28, 2018 at 07:47:48PM +0200, Linus Walleij wrote:
> This is a second RFC version of the DSA driver for Realtek
> RTL8366x especially RTL8366RB.
>
> I've been beating my head against this one and I'm not really
> clear on why my ethernet frames are not coming through to the
> CPU port
> +Examples:
> +
> +switch {
> + compatible = "realtek,rtl8366rb";
> + /* 22 = MDIO (has input reads), 21 = MDC (clock, output only) */
> + mdc-gpios = < 21 GPIO_ACTIVE_HIGH>;
> + mdio-gpios = < 22 GPIO_ACTIVE_HIGH>;
> + reset-gpios = < 14 GPIO_ACTIVE_LOW>;
> +
> +
On Sun, May 27, 2018 at 10:58:35AM +0200, Michael Tuexen wrote:
...
> >>> Patch looks fine, you probably want to note this hard minimum in man(7)
> >>> sctp as
> >>> well
> >>>
> >> I'm aware of some signalling networks which use RTO.min of smaller values
> >> than 200ms.
> >> So could this be
This adds the Ethernet and Realtek switch device to the
D-Link DIR-685 Gemini-based device.
Signed-off-by: Linus Walleij
---
ChangeLog RFCv1->RFCv2
- Rebased, use the new DT bindings
---
arch/arm/boot/dts/gemini-dlink-dir-685.dts | 153 -
1 file changed, 152 insertions(+), 1
The RTL8366RB is an ASIC with five internal PHYs for
LAN0..LAN3 and WAN. The PHYs are spawn off the main
device so they can be handled in a distributed manner
by the Realtek PHY driver. All that is really needed
is the power save feature enablement and letting the
PHY driver core pick up the IRQ
This is a second RFC version of the DSA driver for Realtek
RTL8366x especially RTL8366RB.
I've been beating my head against this one and I'm not really
clear on why my ethernet frames are not coming through to the
CPU port on the chip.
It appears when using ethtool -S on the ports that packets
On Sun, May 20, 2018 at 04:26:03PM +0800, Xin Long wrote:
> On Sat, May 19, 2018 at 11:57 PM, Eric Dumazet wrote:
> > SCTP experts, please take a look.
> >
> > On 05/19/2018 08:55 AM, syzbot wrote:
> >> Hello,
> >>
> >> syzbot found the following crash on:
> >>
> >> HEAD commit:73fcb1a370c7
Jason Gunthorpe wrote:
On Fri, May 25, 2018 at 05:32:52PM -0700, Greg Thelen wrote:
On Fri, May 25, 2018 at 2:32 PM Arnd Bergmann wrote:
> Several subsystems depend on INFINIBAND_ADDR_TRANS, which in turn
depends
> on INFINIBAND. However, when with CONFIG_INIFIBAND=m, this leads to a
>
The Realtek SMI family is a set of DSA chips that provide
switching in routers. This binding just follows the pattern
set by other switches but with the introduction of an embedded
irqchip to demux and handle the interrupts fired by the single
line from the chip.
This interrupt construction is
This adds a driver core for the Realtek SMI chips and a subdriver
for the RTL8366RB. I just added this chip simply because it is
all I can test.
The code is a massaged variant of the code that has been sitting
out-of-tree in OpenWRT for years in the absence of a proper switch
subsystem. I have
> +#ifdef CONFIG_RTL8366_SMI_DEBUG_FS
> + struct dentry *debugfs_root;
> + u8 dbg_vlan_4k_page;
> +#endif
Hi Linus
I think you forgot to remove this.
Andrew
On Mon, 28 May 2018 13:48:28 +0300, Or Gerlitz wrote:
> On Fri, May 25, 2018 at 5:25 AM, Jakub Kicinski wrote:
> > This series from John adds the ability to replay filter offload requests
> > when new offload callback is being registered on a TC block. This is most
> > likely to take place for
On Mon, May 28, 2018 at 04:27:46PM +0200, Toke Høiland-Jørgensen wrote:
[...]
> diff --git a/net/netfilter/core.c b/net/netfilter/core.c
> index 0f6b8172fb9a..520565198f0e 100644
> --- a/net/netfilter/core.c
> +++ b/net/netfilter/core.c
> @@ -572,6 +572,27 @@ void nf_conntrack_destroy(struct
> +struct rtl8366_mib_counter {
> + unsignedbase;
> + unsignedoffset;
> + unsignedlength;
> + const char *name;
> +};
> +void rtl8366_get_strings(struct dsa_switch *ds, int port, uint8_t *data)
> +{
> + struct realtek_smi *smi = ds->priv;
> +
--
Hello
I have been trying to contact you. Did you get my business proposal?
Best Regards,
Miss.Zeliha ömer faruk
Esentepe Mahallesi Büyükdere
Caddesi Kristal Kule Binasi
No:215
Sisli - Istanbul, Turke
On Mon, May 28, 2018 at 04:27:46PM +0200, Toke Høiland-Jørgensen wrote:
> When CAKE is deployed on a gateway that also performs NAT (which is a
> common deployment mode), the host fairness mechanism cannot distinguish
> internal hosts from each other, and so fails to work correctly.
>
> To fix
Michal Kubecek :
> > 1. Is there a cmsg_level/cmsg_type combination that will return the
> > name of the device the packet arrived through?
>
> Not name directly, AFAIK, but you can set SOL_IP / IP_PKTINFO (or
> SOL_IPV6 / IPV6_RECVPKTINFO) socket option and get IP_PKTINFO
> (IPV6_PKTINFO)
Pablo Neira Ayuso writes:
> On Mon, May 28, 2018 at 04:27:46PM +0200, Toke Høiland-Jørgensen wrote:
>> When CAKE is deployed on a gateway that also performs NAT (which is a
>> common deployment mode), the host fairness mechanism cannot distinguish
>> internal hosts from each other, and so fails
At lower bandwidths, the transmission time of a single GSO segment can add
an unacceptable amount of latency due to HOL blocking. Furthermore, with a
software shaper, any tuning mechanism employed by the kernel to control the
maximum size of GSO segments is thrown off by the artificial limit on
This commit adds configurable overhead compensation support to the rate
shaper. With this feature, userspace can configure the actual bottleneck
link overhead and encapsulation mode used, which will be used by the shaper
to calculate the precise duration of each packet on the wire.
This feature
When CAKE is deployed on a gateway that also performs NAT (which is a
common deployment mode), the host fairness mechanism cannot distinguish
internal hosts from each other, and so fails to work correctly.
To fix this, we add an optional NAT awareness mode, which will query the
kernel conntrack
The ACK filter is an optional feature of CAKE which is designed to improve
performance on links with very asymmetrical rate limits. On such links
(which are unfortunately quite prevalent, especially for DSL and cable
subscribers), the downstream throughput can be limited by the number of
ACKs
This patch series adds the CAKE qdisc, and has been split up to ease
review.
I have attempted to split out each configurable feature into its own patch.
The first commit adds the base shaper and packet scheduler, while
subsequent commits add the optional features. The full userspace API and
most
This adds support for DiffServ-based priority queueing to CAKE. If the
shaper is in use, each priority tier gets its own virtual clock, which
limits that tier's rate to a fraction of the overall shaped rate, to
discourage trying to game the priority mechanism.
CAKE defaults to a simple,
The ingress mode is meant to be enabled when CAKE runs downlink of the
actual bottleneck (such as on an IFB device). The mode changes the shaper
to also account dropped packets to the shaped rate, as these have already
traversed the bottleneck.
Enabling ingress mode will also tune the AQM to
sch_cake targets the home router use case and is intended to squeeze the
most bandwidth and latency out of even the slowest ISP links and routers,
while presenting an API simple enough that even an ISP can configure it.
Example of use on a cable ISP uplink:
tc qdisc add dev eth0 cake bandwidth
This adds a global netfilter function to extract a conntrack tuple from an
skb. The function uses a new function added to nf_ct_hook, which will try
to get the tuple from skb->_nfct, and do a full lookup if that fails. This
makes it possible to use the lookup function before the skb has passed
On Sat, May 26, 2018 at 09:48:53AM +, Wei Yongjun wrote:
> Fixes the following sparse warning:
>
> net/netfilter/nf_nat_core.c:1039:20: warning:
> symbol 'nat_hook' was not declared. Should it be static?
Applied, thanks.
On Sun, May 27, 2018 at 06:39:01PM +0530, Arushi Singhal wrote:
> There is a difference between error messages in iptables and
> iptables-compat:
>
> #sudo iptables-compat -D INPUT 4
> iptables: No chain/target/match by that name.
>
> #sudo iptables -D INPUT 4
> iptables: Index of deletion too
On Mon, May 28, 2018 at 12:17:29AM +0300, Vlad Buslov wrote:
...
> -int tcf_action_destroy(struct list_head *actions, int bind)
> +int tcf_action_destroy(struct tc_action *actions[], int bind)
> {
> const struct tc_action_ops *ops;
> - struct tc_action *a, *tmp;
> - int ret = 0;
> +
Pablo Neira Ayuso writes:
> On Mon, May 28, 2018 at 04:27:46PM +0200, Toke Høiland-Jørgensen wrote:
> [...]
>> diff --git a/net/netfilter/core.c b/net/netfilter/core.c
>> index 0f6b8172fb9a..520565198f0e 100644
>> --- a/net/netfilter/core.c
>> +++ b/net/netfilter/core.c
>> @@ -572,6 +572,27 @@
On Sat, May 26, 2018 at 09:01:00PM -0400, Neil Horman wrote:
> On Sat, May 26, 2018 at 05:50:39PM +0200, Dmitry Vyukov wrote:
> > On Sat, May 26, 2018 at 5:42 PM, Michael Tuexen
> > wrote:
> > >> On 25. May 2018, at 21:13, Neil Horman wrote:
> > >>
> > >> On Sat, May 26, 2018 at 01:41:02AM
Hi Nogah!
On Mon, 28 May 2018 18:49:51 +0300, Nogah Frankel wrote:
> > +static int
> > +nfp_abm_red_replace(struct net_device *netdev, struct nfp_abm_link *alink,
> > + struct tc_red_qopt_offload *opt)
> > +{
> > + struct nfp_port *port = nfp_port_from_netdev(netdev);
> > + int
On Mon, May 28, 2018 at 12:17:27AM +0300, Vlad Buslov wrote:
> Change action API to assume that action init function always takes
> reference to action, even when overwriting existing action. This is
> necessary because action API continues to use action pointer after init
> function is done. At
On Mon, May 28, 2018 at 12:17:28AM +0300, Vlad Buslov wrote:
> Implement function that atomically checks if action exists and either takes
> reference to it, or allocates idr slot for action index to prevent
> concurrent allocations of actions with same index. Use EBUSY error pointer
> to indicate
On Mon, May 28, 2018 at 12:17:23AM +0300, Vlad Buslov wrote:
> Implement new action API function that atomically finds and deletes action
> from idr by index. Intended to be used by lockless actions that do not rely
> on rtnl lock.
>
> Signed-off-by: Vlad Buslov
Reviewed-by: Marcelo Ricardo
On Mon, May 28, 2018 at 12:17:22AM +0300, Vlad Buslov wrote:
> Without rtnl lock protection it is no longer safe to use pointer to tc
> action without holding reference to it. (it can be destroyed concurrently)
>
> Remove unsafe action idr lookup function. Instead of it, implement safe tcf
> idr
On Mon, May 28, 2018 at 12:17:25AM +0300, Vlad Buslov wrote:
> Implement helper delete function that uses new action ops 'delete', instead
> of destroying action directly. This is required so act API could delete
> actions by index, without holding any references to action that is being
> deleted.
On Mon, May 28, 2018 at 12:17:21AM +0300, Vlad Buslov wrote:
> Add additional 'rtnl_held' argument to act API init functions. It is
> required to implement actions that need to release rtnl lock before loading
> kernel module and reacquire if afterwards.
>
> Signed-off-by: Vlad Buslov
On Mon, May 28, 2018 at 12:17:26AM +0300, Vlad Buslov wrote:
> Return from action init function with reference to action taken,
> even when overwriting existing action.
>
> Action init API initializes its fourth argument (pointer to pointer to tc
> action) to either existing action with same
On Mon, May 28, 2018 at 12:17:19AM +0300, Vlad Buslov wrote:
> Implement functions to atomically update and free action cookie
> using rcu mechanism.
>
> Signed-off-by: Vlad Buslov
> Signed-off-by: Jiri Pirko
Reviewed-by: Marcelo Ricardo Leitner
> ---
> include/net/act_api.h | 2 +-
>
On Mon, May 28, 2018 at 12:17:24AM +0300, Vlad Buslov wrote:
> Extend action ops with 'delete' function. Each action type to implements
> its own delete function that doesn't depend on rtnl lock.
>
> Implement delete function that is required to delete actions without
> holding rtnl lock. Use
On Mon, May 28, 2018 at 12:17:20AM +0300, Vlad Buslov wrote:
> Change type of action reference counter to refcount_t.
>
> Change type of action bind counter to atomic_t.
> This type is used to allow decrementing bind counter without testing
> for 0 result.
>
> Signed-off-by: Vlad Buslov
>
On Sat, May 26, 2018 at 6:43 AM, David Miller wrote:
> From: Jon Maxwell
> Date: Fri, 25 May 2018 07:38:29 +1000
>
>> Fixup the checksum for CHECKSUM_COMPLETE when pulling skbs on RX path.
>> Otherwise we get splats when tc mirred is used to redirect packets to ifb.
>>
>> Before fix:
>>
>> nic:
Hello,
syzbot found the following crash on:
HEAD commit:7a1a98c171ea Merge branch 'bpf-sendmsg-hook'
git tree: bpf-next
console output: https://syzkaller.appspot.com/x/log.txt?x=131f406780
kernel config: https://syzkaller.appspot.com/x/.config?x=e4078980b886800c
dashboard link:
Check sch_direct_xmit() at the end of sch_direct_xmit() will be bypassed.
This is because "ret" from sch_direct_xmit() will be either NETDEV_TX_OK
or NETDEV_TX_BUSY, and only ret == NETDEV_TX_OK == 0 will reach the
condition:
if (ret && netif_xmit_frozen_or_stopped(txq))
return false;
Hello,
syzbot found the following crash on:
HEAD commit:861d9dd37526 Merge tag 'kbuild-fixes-v4.17-2' of git://git..
git tree: upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=10bffd0f80
kernel config: https://syzkaller.appspot.com/x/.config?x=982e2df1b9e60b02
Hi David,
The following patchset contains Netfilter/IPVS fixes for your net tree:
1) Null pointer dereference when dumping conntrack helper configuration,
from Taehee Yoo.
2) Missing sanitization in ebtables extension name through compat,
from Paolo Abeni.
3) Broken fetch of tracing
From: Paolo Abeni
strlcpy() can't be safely used on a user-space provided string,
as it can try to read beyond the buffer's end, if the latter is
not NULL terminated.
Leveraging the above, syzbot has been able to trigger the following
splat:
BUG: KASAN: stack-out-of-bounds in strlcpy
This patch fixes the following splat.
[118709.054937] BUG: using smp_processor_id() in preemptible [] code:
test/1571
[118709.054970] caller is nft_update_chain_stats.isra.4+0x53/0x97 [nf_tables]
[118709.054980] CPU: 2 PID: 1571 Comm: test Not tainted 4.17.0-rc6+ #335
[...]
From: Taehee Yoo
The table field in nft_obj_filter is not an array. In order to check
tablename, we should check if the pointer is set.
Test commands:
%nft add table ip filter
%nft add counter ip filter ct1
%nft reset counters
Splat looks like:
[ 306.510504] kasan:
From: Julian Anastasov
syzkaller reports for buffer overflow for interface name
when starting sync daemons [1]
What we do is that we copy user structure into larger stack
buffer but later we search NUL past the stack buffer.
The same happens for sched_name when adding/editing virtual server.
Credit calculations for the packet ratelimiting are not correct, as per
the applied ratelimit of 25/second and burst 8, a total of 33 packets
should have been accepted. This is true in iptables(33) but not in
nftables (~65). For packet ratelimiting, use:
div_u64(limit->nsecs,
From: Taehee Yoo
In the nft_ct_helper_obj_dump(), always priv->helper4 is dereferenced.
But if family is ipv6, priv->helper6 should be dereferenced.
Steps to reproduces:
#test.nft
table ip6 filter {
ct helper ftp {
type "ftp" protocol tcp
}
From: Taehee Yoo
In the nft_meta_set_eval, nftrace value is dereferenced as u32 from sreg.
But correct type is u8. so that sometimes incorrect value is dereferenced.
Steps to reproduce:
%nft add table ip filter
%nft add chain ip filter input { type filter hook input priority 4\; }
From: Eric Dumazet
Recent patch forgot to remove nla_data(), upsetting syzkaller a bit.
BUG: KASAN: slab-out-of-bounds in nla_strlcpy+0x13d/0x150 lib/nlattr.c:314
Read of size 1 at addr 8801ad1f4fdd by task syz-executor189/4509
CPU: 1 PID: 4509 Comm: syz-executor189 Not tainted 4.17.0-rc6+
From: Taehee Yoo
When a chain is updated, a counter can be attached. if so,
the nft_counters_enabled should be increased.
test commands:
%nft add table ip filter
%nft add chain ip filter input { type filter hook input priority 4\; }
%iptables-compat -Z input
%nft delete chain ip
On Tue, May 29, 2018 at 12:36 AM, Eric Dumazet wrote:
> On Mon, May 28, 2018 at 8:36 AM Yafang Shao wrote:
>
>> This is additional to the commit ea1627c20c34 ("tcp: minor optimizations
> around tcp_hdr() usage").
>> At this point, skb->data is same with tcp_hdr() as tcp header has not
>> been
From: Arnd Bergmann Sent: 2018年5月28日 23:50
> While compile-testing on arm64 with gcc-8.1, I ran into a build diagnostic:
>
> drivers/net/ethernet/freescale/fec_main.c: In function 'fec_probe':
> drivers/net/ethernet/freescale/fec_main.c:3517:25: error: '%d' directive
> writing between 1 and 10
We assumes used ring update is the only user for vhost_put_user() in
the past. This may not be the case for the incoming packed ring which
may update the descriptor ring for used. So introduce a new type
parameter.
Signed-off-by: Jason Wang
---
drivers/vhost/vhost.c | 14 +++---
1 file
Signed-off-by: Jason Wang
---
drivers/vhost/net.c | 13 +-
drivers/vhost/vhost.c | 585 ++
drivers/vhost/vhost.h | 13 +-
3 files changed, 566 insertions(+), 45 deletions(-)
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index
This patch introduces basic support for event suppression aka driver
and device area.
Signed-off-by: Jason Wang
---
drivers/vhost/vhost.c| 191 ---
drivers/vhost/vhost.h| 10 +-
include/uapi/linux/virtio_ring.h | 19
3 files
Signed-off-by: Jason Wang
---
include/uapi/linux/virtio_config.h | 9 +
include/uapi/linux/virtio_ring.h | 13 +
2 files changed, 22 insertions(+)
diff --git a/include/uapi/linux/virtio_config.h
b/include/uapi/linux/virtio_config.h
index 308e209..5903d51 100644
---
Move get_rx_bufs() to vhost.c and rename it to
vhost_get_bufs(). This helps to hide vring internal layout from
specific device implementation. Packed ring implementation will
benefit from this.
Signed-off-by: Jason Wang
---
drivers/vhost/net.c | 83
We used to return descriptor head by vhost_get_vq_desc() to device and
pass it back to vhost_add_used() and its friends. This exposes the
internal used ring layout to device which makes it hard to be extended for
e.g packed ring layout.
So this patch tries to hide the used ring layout by
-
Two helpers of setting/getting used len were introduced to avoid
explicitly manipulating vhost_used_elem in zerocopy code. This will be
used to hide used_elem internals and simplify packed ring
implementation.
Signed-off-by: Jason Wang
---
drivers/vhost/net.c | 11 +--
Hi all:
This RFC implement packed ring layout. The code were tested with
Tiwei's RFC V5 at https://lkml.org/lkml/2018/5/22/138. Some fixups and
tweaks were needed on top of Tiwei's code to make it run for event
index.
Pktgen reports about 20% improvement on TX PPS when doing pktgen from
guest to
Instead of depending on the exported vring_used_elem, this patch
switches to use a new internal structure vhost_used_elem which embed
vring_used_elem in itself. This could be used to let vhost to record
extra metadata for the incoming packed ring layout.
Signed-off-by: Jason Wang
---
On 2018年05月28日 18:37, Toshiaki Makita wrote:
Calling XDP redirection requires bh disabled. Softirq can call another
XDP function and redirection functions, then the percpu static variable
ri->map can be overwritten to NULL.
This is a generic XDP case called from tun.
[ 3535.736058] BUG:
From: Pablo Neira Ayuso
Date: Tue, 29 May 2018 01:42:12 +0200
> The following patchset contains Netfilter/IPVS fixes for your net tree:
>
> 1) Null pointer dereference when dumping conntrack helper configuration,
>from Taehee Yoo.
>
> 2) Missing sanitization in ebtables extension name
The local variable is only used while CONFIG_IPV6 enabled
net/core/filter.c: In function ‘sk_msg_convert_ctx_access’:
net/core/filter.c:6489:6: warning: unused variable ‘off’ [-Wunused-variable]
int off;
^
This puts it into #ifdef.
Fixes: 303def35f64e ("bpf: allow sk_msg programs to read
On 2018年05月22日 16:16, Tiwei Bie wrote:
This commit introduces the support for creating packed ring.
All split ring specific functions are added _split suffix.
Some necessary stubs for packed ring are also added.
Signed-off-by: Tiwei Bie
---
drivers/virtio/virtio_ring.c | 801
Remove and coalesce formats when there is an unnecessary
character after a logging newline. These extra characters
cause logging defects.
Miscellanea:
o Coalesce formats
Signed-off-by: Joe Perches
---
drivers/net/ethernet/cavium/liquidio/lio_main.c | 2 +-
From: Sudarsana Reddy Kalluru
Date: Thu, 24 May 2018 04:57:51 -0700
> Tx-timeout mostly happens due to some issue in the device. In such cases,
> debug dump would be helpful for identifying the cause of the issue.
> This patch adds support to spill debug data during the Tx timeout. Here
>
From: Thierry Reding
Date: Thu, 24 May 2018 16:09:07 +0200
> From: Thierry Reding
>
> Some drivers, such as DWC EQOS on Tegra, need to perform operations that
> can sleep under this lock (clk_set_rate() in tegra_eqos_fix_speed()) for
> proper operation. Since there is no need for this lock to
From: Davide Caratti
Date: Thu, 24 May 2018 17:49:35 +0200
> SCTP sockets originated in a VRF can improve their performance if CRC32c
> computation is delegated to underlying devices: update device features,
> setting NETIF_F_SCTP_CRC. Iterating the following command in the topology
> proposed
Hi all,
Today's linux-next merge of the net-next tree got a conflict in:
arch/s390/net/Makefile
between commit:
866f4c8e0e26 ("s390/net: add pnetid support")
from the s390 tree and commit:
e1cf4befa297 ("bpf, s390x: remove ld_abs/ld_ind")
from the net-next tree.
I fixed it up (see
From: Sridhar Samudrala
Date: Thu, 24 May 2018 09:55:12 -0700
> The main motivation for this patch is to enable cloud service providers
> to provide an accelerated datapath to virtio-net enabled VMs in a
> transparent manner with no/minimal guest userspace changes. This also
> enables
From: Heiner Kallweit
Date: Thu, 24 May 2018 22:40:12 +0200
> Add RTL8211B suspend / resume callbacks.
>
> Signed-off-by: Heiner Kallweit
Applied.
From: Jon Maxwell
Date: Fri, 25 May 2018 07:38:29 +1000
> Fixup the checksum for CHECKSUM_COMPLETE when pulling skbs on RX path.
> Otherwise we get splats when tc mirred is used to redirect packets to ifb.
>
> Before fix:
>
> nic: hw csum failure
>
> Signed-off-by: Jon Maxwell
Applied.
From: Yangbo Lu
Date: Fri, 25 May 2018 12:40:34 +0800
> gianfar_ptp was the PTP clock driver for 1588 timer
> module of Freescale QorIQ eTSEC (Enhanced Three-Speed
> Ethernet Controllers) platforms. Actually QorIQ DPAA
> (Data Path Acceleration Architecture) platforms is
> also using the same
1 - 100 of 205 matches
Mail list logo