On 07/04/2018 04:34 PM, Jonathan Maxwell wrote:
> Let's wait for Eric to review. Then I'll put together the next version.
Sorry for the delay (I was travelling last week) , please respin a v3, thanks !
Jakub Kicinski [Mon, 2018-07-09 11:01 -0700]:
> libbpf can guess program type based on ELF section names. As libbpf
> becomes more popular its association between section name strings and
> types becomes more of a standard. Allow libbpf users to use the same
> logic for matching strings to
Jakub Kicinski [Mon, 2018-07-09 11:01 -0700]:
> Similarly to bpf_prog_load() users of bpf_object__open() may need
> to specify the expected program type. Program type is needed at
> open to avoid the kernel version check for program types which don't
> require it.
>
> Signed-off-by: Jakub
Jakub Kicinski [Mon, 2018-07-09 19:49 -0700]:
> On Mon, 9 Jul 2018 13:22:54 -0700, Andrey Ignatov wrote:
> > Jakub Kicinski [Mon, 2018-07-09 11:01 -0700]:
> > > More advanced applications may want to only replace programs without
> > > destroying associated maps. Allow libbpf users to achieve
On Mon, Jul 09, 2018 at 10:59:41AM -0700, Jakub Kicinski wrote:
> bpf_prog_load() is a very useful helper but it doesn't give us full
> flexibility of modifying the BPF objects before loading. Open code
> bpf_prog_load() in bpftool so we can add extra logic in following
> commits.
>
>
From: Stephen Hemminger
If option or qdisc is unknown, still output in JSON.
Signed-off-by: Stephen Hemminger
---
tc/tc.c | 19 +++
1 file changed, 11 insertions(+), 8 deletions(-)
diff --git a/tc/tc.c b/tc/tc.c
index 0d223281ba25..b6d2a6151793 100644
--- a/tc/tc.c
+++
From: Stephen Hemminger
Only sprint_size is used, so fold it in.
Signed-off-by: Stephen Hemminger
---
tc/tc_util.c | 7 ++-
tc/tc_util.h | 1 -
2 files changed, 2 insertions(+), 6 deletions(-)
diff --git a/tc/tc_util.c b/tc/tc_util.c
index f5ffe3443892..01e131b5c5d7 100644
---
From: Stephen Hemminger
Only sprint_time is used.
Signed-off-by: Stephen Hemminger
---
tc/tc_util.c | 8 ++--
tc/tc_util.h | 1 -
2 files changed, 2 insertions(+), 7 deletions(-)
diff --git a/tc/tc_util.c b/tc/tc_util.c
index 01e131b5c5d7..95cb49b98612 100644
--- a/tc/tc_util.c
+++
From: Stephen Hemminger
This function is not used, only sprint_rate is used.
Signed-off-by: Stephen Hemminger
---
tc/tc_util.c | 10 +++---
tc/tc_util.h | 1 -
2 files changed, 3 insertions(+), 8 deletions(-)
diff --git a/tc/tc_util.c b/tc/tc_util.c
index 05b6c97563b3..f5ffe3443892
From: Stephen Hemminger
Convert compatiablity statistics to print in JSON.
Signed-off-by: Stephen Hemminger
---
tc/tc_util.c | 32 ++--
1 file changed, 22 insertions(+), 10 deletions(-)
diff --git a/tc/tc_util.c b/tc/tc_util.c
index e8e1241d083d..05b6c97563b3
Thanks Richard,
I'll add it in my next revision.
> -Original Message-
> From: Richard Cochran [mailto:richardcoch...@gmail.com]
> Sent: Friday, July 6, 2018 5:25 PM
> To: Bryan Whitehead - C21958
> Cc: da...@davemloft.net; netdev@vger.kernel.org; UNGLinuxDriver
>
> Subject: Re: [PATCH
On Mon, Jul 09, 2018 at 02:18:33PM -0400, Michel Machado wrote:
> On 07/09/2018 11:44 AM, Marcelo Ricardo Leitner wrote:
> > On Sat, Jul 07, 2018 at 03:43:55PM +0530, Nishanth Devarajan wrote:
> > > net/sched: add skbprio scheduer
> > >
> > > Skbprio (SKB Priority Queue) is a queueing discipline
On Mon, Jul 09, 2018 at 08:26:47PM +0300, Vlad Buslov wrote:
> Fix action attribute size calculation function to take rcu read lock and
> access act_cookie pointer with rcu dereference.
>
> Fixes: eec94fdb0480 ("net: sched: use rcu for action cookie update")
> Reported-by: Marcelo Ricardo Leitner
On 07/09/2018 03:53 PM, Marcelo Ricardo Leitner wrote:
On Mon, Jul 09, 2018 at 02:18:33PM -0400, Michel Machado wrote:
On 07/09/2018 11:44 AM, Marcelo Ricardo Leitner wrote:
On Sat, Jul 07, 2018 at 03:43:55PM +0530, Nishanth Devarajan wrote:
net/sched: add skbprio scheduer
Skbprio (SKB
Jakub Kicinski [Mon, 2018-07-09 11:01 -0700]:
> More advanced applications may want to only replace programs without
> destroying associated maps. Allow libbpf users to achieve that.
> Instead of always creating all of the maps at load time, expose to
> users an API to reconstruct the map object
On 07/09/2018 01:34 PM, Marcelo Ricardo Leitner wrote:
> I am not sure if this is enough to fix the entire issue. Now it will
> fetch the length correctly but, what guarantees that when it tries to
> actually copy the key (tcf_action_dump_1), the same act_cookie pointer
> will be used? As in,
On Mon 09 Jul 2018 at 20:34, Marcelo Ricardo Leitner
wrote:
> On Mon, Jul 09, 2018 at 08:26:47PM +0300, Vlad Buslov wrote:
>> Fix action attribute size calculation function to take rcu read lock and
>> access act_cookie pointer with rcu dereference.
>>
>> Fixes: eec94fdb0480 ("net: sched: use
On 03.07.2018 18:48, Florian Fainelli wrote:
>
>
> On 07/02/2018 02:31 PM, Heiner Kallweit wrote:
>> On 02.07.2018 23:20, Andrew Lunn wrote:
>>> On Mon, Jul 02, 2018 at 09:37:08PM +0200, Heiner Kallweit wrote:
Change rtl_speed_down() to use phylib.
Signed-off-by: Heiner Kallweit
On Mon, Jul 09, 2018 at 05:03:31PM -0400, Michel Machado wrote:
> On 07/09/2018 03:53 PM, Marcelo Ricardo Leitner wrote:
> > On Mon, Jul 09, 2018 at 02:18:33PM -0400, Michel Machado wrote:
> > > On 07/09/2018 11:44 AM, Marcelo Ricardo Leitner wrote:
> > > > On Sat, Jul 07, 2018 at 03:43:55PM
Mark reported that syzkaller triggered a KASAN detected slab-out-of-bounds
bug in ___bpf_prog_run() with a BPF_LD | BPF_ABS word load at offset 0x8001.
After further investigation it became clear that the issue was the
BPF_LDX_MEM() which takes offset as an argument whereas it cannot encode
larger
We are not supposed to use u32 in uapi, so change the flags member of
struct sock_txtime from u32 to __u32 instead.
Fixes: 80b14dee2bea ("net: Add a new socket option for a future transmit time")
Signed-off-by: Jesus Sanchez-Palencia
---
include/uapi/linux/net_tstamp.h | 2 +-
1 file changed, 1
On 07/09/2018 04:08 PM, Jesus Sanchez-Palencia wrote:
> We are not supposed to use u32 in uapi, so change the flags member of
> struct sock_txtime from u32 to __u32 instead.
>
> Fixes: 80b14dee2bea ("net: Add a new socket option for a future transmit
> time")
> Signed-off-by: Jesus
From: Michael Chan
Date: Mon, 9 Jul 2018 02:24:46 -0400
> These are bug fixes in error code paths, TC Flower VLAN TCI flow
> checking bug fix, proper filtering of Broadcast packets if IFF_BROADCAST
> is not set, and a bug fix in bnxt_get_max_rings() to return 0 ring
> parameters when the return
On Tue, Jul 3, 2018 at 10:45 PM, Konstantin Khlebnikov
wrote:
> I'm seeing problems with tunnelled traffic with Mellanox Technologies
> MT27710 Family [ConnectX-4 Lx] using vanilla driver from linux 4.4.y
>
> Packets with payload bigger than 116 bytes are not exmited.
> Smaller packets and normal
From: Julian Wiedmann
Date: Mon, 9 Jul 2018 09:45:14 +0200
> Since commit 74d4a8f8d378 ("tcp: remove sk_can_gso() use"), the code
> doesn't care whether the interface supports SG.
>
> Signed-off-by: Julian Wiedmann
> ---
> v2: remove the whole comment (Eric)
Applied.
On 07/09/2018 10:32 AM, David Ahern wrote:
> On 7/9/18 9:48 AM, Jesus Sanchez-Palencia wrote:
>> Hi David,
>>
>>
>> On 07/06/2018 08:58 AM, David Ahern wrote:
>>> On 7/5/18 4:42 PM, Jesus Sanchez-Palencia wrote:
>>>
+static int get_clockid(__s32 *val, const char *arg)
+{
+ const
On 07/09/2018 04:20 PM, Jesus Sanchez-Palencia wrote:
> We are not supposed to use u32 in uapi, so change the flags member of
> struct sock_txtime from u32 to __u32 instead.
>
> Fixes: 80b14dee2bea ("net: Add a new socket option for a future transmit
> time")
> Reported-by: Eric Dumazet
>
From: Taehee Yoo
Date: Sun, 8 Jul 2018 11:55:51 +0900
> rhashtable_free_and_destroy() cancels re-hash deferred work
> then walks and destroys elements. at this moment, some elements can be
> still in future_tbl. that elements are not destroyed.
>
> test case:
> nft_rhash_destroy() calls
From: Edward Cree
Date: Mon, 9 Jul 2018 18:08:38 +0100
> A couple of bugs in skb list handling were spotted by Dan Carpenter, with
> the help of Smatch; following up on them I found a couple more similar
> cases. This series fixes them by changing the relevant loops to use the
>
From: Alexander Duyck
This patch makes it so that instead of passing a void pointer as the
accel_priv we instead pass a net_device pointer as sb_dev. Making this
change allows us to pass the subordinate device through to the fallback
function eventually so that we can keep the actual code in the
From: Alexander Duyck
This patch makes it so that we use the tc_to_txq mapping in the macvlan
device in order to select the Tx queue for outgoing packets.
The idea here is to try and move away from using ixgbe_select_queue and to
come up with a generic way to make this work for devices going
This patch series is meant to allow support for the L2 forward offload, aka
MACVLAN offload without the need for using ndo_select_queue.
The existing solution currently requires that we use ndo_select_queue in
the transmit path if we want to associate specific Tx queues with a given
MACVLAN
From: Alexander Duyck
This patch adds a generic version of the ndo_select_queue functions for
either returning 0 or selecting a queue based on the processor ID. This is
generally meant to just reduce the number of functions we have to change
in the future when we have to deal with
From: Ido Schimmel
Date: Sun, 8 Jul 2018 23:51:15 +0300
> This is the second and last set of preparations towards initial
> Spectrum-2 support in mlxsw. It mainly re-arranges parts of the code
> that need to work with both ASICs, but somewhat differ.
>
> The first three patches allow different
We are not supposed to use u32 in uapi, so change the flags member of
struct sock_txtime from u32 to __u32 instead.
Fixes: 80b14dee2bea ("net: Add a new socket option for a future transmit time")
Reported-by: Eric Dumazet
Signed-off-by: Jesus Sanchez-Palencia
---
On 07/09/2018 04:18 PM, Eric Dumazet wrote:
>
>
> On 07/09/2018 04:08 PM, Jesus Sanchez-Palencia wrote:
>> We are not supposed to use u32 in uapi, so change the flags member of
>> struct sock_txtime from u32 to __u32 instead.
>>
>> Fixes: 80b14dee2bea ("net: Add a new socket option for a
From: Alexander Duyck
This change makes it so that we can support the concept of subordinate
device traffic classes to the core networking code. In doing this we can
start pulling out the driver specific bits needed to support selecting a
queue based on an upper device.
The solution at is
From: Alexander Duyck
For most of these calls we can just pass NULL through to the fallback
function as the sb_dev. The only cases where we cannot are the cases where
we might be dealing with either an upper device or a driver that would
have configured things to support an sb_dev itself.
The
From: Alexander Duyck
This patch is meant to provide the basic tools needed to allow us to create
subordinate device traffic classes. The general idea here is to allow
subdividing the queues of a device into queue groups accessible through an
upper device such as a macvlan.
The idea here is to
From: Alexander Duyck
This patch makes it so that we do not report the traffic class or allow XPS
configuration on single queue devices. This is mostly to avoid unnecessary
complexity with changes I have planned that will allow us to reuse
the unused tc_to_txq and XPS configuration on a single
On 07/07/2018 05:44 PM, Eric Dumazet wrote:
>
>
> On 07/03/2018 03:42 PM, Jesus Sanchez-Palencia wrote:
>> From: Richard Cochran
>>
>> This patch introduces SO_TXTIME. User space enables this option in
>> order to pass a desired future transmit time in a CMSG when calling
>> sendmsg(2). The
From: Boris Pismenny
Date: Sun, 8 Jul 2018 12:36:14 +0300
> The following series provides TLS TX inline crypto offload.
I think this patch series provides RX inline offload not TX inline
offload.
From: Jesus Sanchez-Palencia
Date: Mon, 9 Jul 2018 16:20:56 -0700
> We are not supposed to use u32 in uapi, so change the flags member of
> struct sock_txtime from u32 to __u32 instead.
>
> Fixes: 80b14dee2bea ("net: Add a new socket option for a future transmit
> time")
> Reported-by: Eric
From: Vinicius Costa Gomes
The "Earliest TxTime First" (ETF) queueing discipline allows precise
control of the transmission time of packets by providing a sorted
time-based scheduling of packets.
The syntax is:
tc qdisc add dev DEV parent NODE etf delta
clockid [offload]
fixes since v3:
- Add support for clock names with the "CLOCK_" prefix;
- Print clock name on print_opt();
- Use strcasecmp() instead of strncasecmp().
The ETF (earliest txtime first) qdisc was recently merged into net-next
[1], so this patchset adds support for it through the tc command line
Add an initial manpage for tc-etf covering all config options, basic
concepts and operation modes.
Signed-off-by: Jesus Sanchez-Palencia
---
man/man8/tc-etf.8 | 141 ++
1 file changed, 141 insertions(+)
create mode 100644 man/man8/tc-etf.8
diff
This should come from the next uapi headers update.
Sending it now just as a convenience so anyone can build tc with etf
and taprio support.
Signed-off-by: Jesus Sanchez-Palencia
---
include/uapi/linux/pkt_sched.h | 21 +
1 file changed, 21 insertions(+)
diff --git
On 7/9/18, 12:32 PM, "Yuchung Cheng" wrote:
On Sat, Jul 7, 2018 at 7:07 AM, Neal Cardwell wrote:
> On Sat, Jul 7, 2018 at 7:15 AM David Miller wrote:
>>
>> From: Lawrence Brakmo
>> Date: Tue, 3 Jul 2018 09:26:13 -0700
>>
>> > When have observed high tail latencies
From: Stephen Hemminger
Update core of TC command and library to do more JSON.
Most of this patch set is about getting tc utility functions
to be more friendly to the json_print infrastructure.
Stephen Hemminger (9):
tc: use JSON in error handling
tc: use const char in util
tc: convert
From: Stephen Hemminger
Convert to use JSON
Signed-off-by: Stephen Hemminger
---
tc/q_sfq.c | 65 --
1 file changed, 39 insertions(+), 26 deletions(-)
diff --git a/tc/q_sfq.c b/tc/q_sfq.c
index 6a1d853b7c93..cc8ce0dddf7e 100644
---
From: Stephen Hemminger
For prefix and path, can use const char.
Signed-off-by: Stephen Hemminger
---
tc/tc_util.c | 8 +---
tc/tc_util.h | 6 +++---
2 files changed, 8 insertions(+), 6 deletions(-)
diff --git a/tc/tc_util.c b/tc/tc_util.c
index e0c96291ade0..e8e1241d083d 100644
---
On Mon, Jul 09, 2018 at 10:59:44AM -0700, Jakub Kicinski wrote:
> Add map parameter to prog load which will allow reuse of existing
> maps instead of creating new ones.
>
> We need feature detection and compat code for reallocarray, since
> it's not available in many libc versions.
>
>
From: Stephen Hemminger
All other places in tc use sprint_rate.
Signed-off-by: Stephen Hemminger
---
tc/q_cbq.c | 15 ++-
1 file changed, 6 insertions(+), 9 deletions(-)
diff --git a/tc/q_cbq.c b/tc/q_cbq.c
index e7f1a3bfaf5d..ad0170c41858 100644
--- a/tc/q_cbq.c
+++ b/tc/q_cbq.c
From: Stephen Hemminger
Add a helper to print rate and size in numeric or pretty format
based on JSON flag.
Signed-off-by: Stephen Hemminger
---
tc/tc_util.c | 59 ++--
tc/tc_util.h | 2 ++
2 files changed, 31 insertions(+), 30 deletions(-)
Today macvlan ignores the notification when a lower device goes
administratively down, preventing the lack of connectivity from
bubbling up.
Processing NETDEV_DOWN results in a macvlan state of LOWERLAYERDOWN
with NO-CARRIER which should be easy to interpret in userspace.
2: lower: mtu 1500
On Mon, 9 Jul 2018 12:48:20 -0700, Alexei Starovoitov wrote:
> On Mon, Jul 09, 2018 at 10:59:44AM -0700, Jakub Kicinski wrote:
> > Add map parameter to prog load which will allow reuse of existing
> > maps instead of creating new ones.
> >
> > We need feature detection and compat code for
On Mon, 9 Jul 2018 13:22:54 -0700, Andrey Ignatov wrote:
> Jakub Kicinski [Mon, 2018-07-09 11:01 -0700]:
> > More advanced applications may want to only replace programs without
> > destroying associated maps. Allow libbpf users to achieve that.
> > Instead of always creating all of the maps at
The current code returns -ENOMEM and does not bother to set the output
parameters to 0 when no rings are available. Some callers, such as
bnxt_get_channels() will display garbage ring numbers when that happens.
Fix it by always setting the output parameters.
Fixes: 6e6c5a57fbe1 ("bnxt_en: Modify
These are bug fixes in error code paths, TC Flower VLAN TCI flow
checking bug fix, proper filtering of Broadcast packets if IFF_BROADCAST
is not set, and a bug fix in bnxt_get_max_rings() to return 0 ring
parameters when the return value is -ENOMEM.
Michael Chan (4):
bnxt_en: Fix inconsistent
Calling bnxt_set_max_func_irqs() to modify the max IRQ count requested or
freed by the RDMA driver is flawed. The max IRQ count is checked when
re-initializing the IRQ vectors and this can happen multiple times
during ifup or ethtool -L. If the max IRQ is reduced and the RDMA
driver is
If there aren't enough RX rings available, the driver will attempt to
use a single RX ring without the aggregation ring. If that also
fails, the BNXT_FLAG_AGG_RINGS flag is cleared but the other ring
parameters are not set consistently to reflect that. If more RX
rings become available at the
From: Venkat Duvvuru
It is possible that OVS may set don’t care for DEI/CFI bit in
vlan_tci mask. Hence, checking for vlan_tci exact match will endup
in a vlan flow rejection.
This patch fixes the problem by checking for vlan_pcp and vid
separately, instead of checking for the entire vlan_tci.
From: Vikas Gupta
Fix bug in the error code path when bnxt_request_irq() returns failure.
bnxt_disable_napi() should not be called in this error path because
NAPI has not been enabled yet.
Fixes: c0c050c58d84 ("bnxt_en: New Broadcom ethernet driver.")
Signed-off-by: Vikas Gupta
Signed-off-by:
Currently, the driver assumes IFF_BROADCAST is always set and always sets
the broadcast filter. Modify the code to set or clear the broadcast
filter according to the IFF_BROADCAST flag.
Signed-off-by: Michael Chan
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 9 +++--
1 file changed, 7
On 07/09/2018 02:43 AM, Paolo Abeni wrote:
> On Fri, 2018-07-06 at 07:20 -0700, Eric Dumazet wrote:
>> I will test/polish it later, I am coming back from vacations and have a
>> backlog.
>>
>> Here are my results : (Note that I have _not_ changed
>> /proc/sys/net/ipv4/ipfrag_time )
>>
>>
Free params if tcf_idr_check_alloc() returned error.
Fixes: 0190c1d452a9 ("net: sched: atomically check-allocate action")
Reported-by: Dan Carpenter
Signed-off-by: Vlad Buslov
---
net/sched/act_ife.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/net/sched/act_ife.c
On Fri, 2018-07-06 at 07:20 -0700, Eric Dumazet wrote:
> I will test/polish it later, I am coming back from vacations and have a
> backlog.
>
> Here are my results : (Note that I have _not_ changed
> /proc/sys/net/ipv4/ipfrag_time )
>
> lpaa6:~# grep . /proc/sys/net/ipv4/ipfrag_* ; grep FRAG
Extend struct tcf_walker with additional 'cookie' field. It is intended to
be used by classifier walk implementations to continue iteration directly
from particular filter, instead of iterating 'skip' number of times.
Change flower walk implementation to save filter handle in 'cookie'. Each
time
From: Arjun Vynipadath
congestion argument passed to t4_sge_alloc_rxq() is used
to differentiate between nic/ofld queues.
Signed-off-by: Arjun Vynipadath
Signed-off-by: Ganesh Goudar
---
drivers/net/ethernet/chelsio/cxgb4/sge.c | 4 +++-
drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h |
inet6_ifla6_size() is called to check how much space is needed by
inet6_fill_link_af() and inet6_fill_ifinfo(), both of which include
the IFLA_INET6_ADDR_GEN_MODE attribute. Reserve some room for it.
Fixes: bc91b0f07ada ("ipv6: addrconf: implement address generation modes")
Signed-off-by: Sabrina
The value has already been copied from this netns's devconf_dflt, it
shouldn't be reset to the global kernel default.
Fixes: d35a00b8e33d ("net/ipv6: allow sysctl to change link-local address
generation mode")
Signed-off-by: Sabrina Dubroca
Reviewed-by: David Ahern
---
net/ipv6/addrconf.c | 2
addrconf_sysctl_addr_gen_mode() has multiple problems. First, it ignores
the errors returned by proc_dointvec().
addrconf_sysctl_addr_gen_mode() calls proc_dointvec() directly, which
writes the value to memory, and then checks if it's valid and may return
EINVAL. If a bad value is given, the
This aligns the addr_gen_mode sysctl with the expected behavior of the
"all" variant.
Fixes: d35a00b8e33d ("net/ipv6: allow sysctl to change link-local address
generation mode")
Suggested-by: David Ahern
Signed-off-by: Sabrina Dubroca
---
net/ipv6/addrconf.c | 12
1 file changed,
This series fixes bugs in handling of the addr_gen_mode option, mainly
related to the sysctl. A minor netlink issue was also present in the
initial commit introducing the option on a per-netdevice basis.
v2: add patch 4, requested by David Ahern during review of v1
add patch 5, missing
addr_gen_mode was introduced in without documentation, add it now.
Fixes: d35a00b8e33d ("net/ipv6: allow sysctl to change link-local address
generation mode")
Signed-off-by: Sabrina Dubroca
---
Documentation/networking/ip-sysctl.txt | 9 +
1 file changed, 9 insertions(+)
diff --git
On 07/09/2018 04:34 AM, Eric Dumazet wrote:
> and number of tx queues?
>
> You seem to self inflict losses on the sender, and that is terrible for the
> (convoluted) stress test you want to run.
>
> I use mq + fq : no losses on the sender.
>
> Do not send patches to solve a problem that
Since commit 74d4a8f8d378 ("tcp: remove sk_can_gso() use"), the code
doesn't care whether the interface supports SG.
Signed-off-by: Julian Wiedmann
---
v2: remove the whole comment (Eric)
net/ipv4/tcp.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index
Expose stats obtained from firmware via debugfs. These stats can't
be part of ethtool -S because the slow firmware mailbox can cause
packet drops under heavy load.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Ganesh Goudar
---
drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c | 164
Some stats are fetched via slow firmware mailbox, which can cause
packet drops under heavy load. So, this series removes these stats
from ethtool -S and expose them via debugfs.
Patch 1 removes stats fetched via firmware from ethtool -S.
Patch 2 exposes stats removed in Patch 1 via debugfs.
On 7/9/18 9:17 AM, Eric Dumazet wrote:
>
>
> On 07/09/2018 07:59 AM, David Ahern wrote:
>> On 7/8/18 10:53 PM, Lorenzo Colitti wrote:
>>> On Sat, Jul 7, 2018 at 10:29 PM Eric Dumazet wrote:
>> Tested: passes Android sock_diag_test.py, which exercises this codepath
>
> If this
When running ethtool -S, some stats are requested from firmware.
Since getting these stats via firmware mailbox is slow, some packets
get dropped under heavy load while running ethtool -S.
So, remove these stats from ethtool -S.
Signed-off-by: Rahul Lakkireddy
Signed-off-by: Ganesh Goudar
---
On 07/09/2018 09:14 AM, David Ahern wrote:
> Perhaps it is something with my config, settings, ss version or test
> program, but I do not see it on 4.11:
>
Maybe some vrf issue, I dunno.
On Mon, Jul 9, 2018 at 9:05 AM, Deepti Raghavan wrote:
> Congestion control algorithms, which access the rate sample
> through the tcp_cong_control function, only have access to the maximum
> of the send and receive interval, for cases where the acknowledgment
> rate may be inaccurate due to ACK
My name is Smadar Barber-Tsadik, I'm the Chief Executive Officer (C.P.A) of the
First International Bank of Israel (FIBI). I'm getting in touch with you in
regards to a very important and urgent matter. Kindly respond back at your
earliest convinience so I can provide you the details.
This patch makes it so that we do not report the traffic class or allow XPS
configuration on single queue devices. This is mostly to avoid unnecessary
complexity with changes I have planned that will allow us to reuse
the unused tc_to_txq and XPS configuration on a single queue device to
allow it
This patch is meant to provide the basic tools needed to allow us to create
subordinate device traffic classes. The general idea here is to allow
subdividing the queues of a device into queue groups accessible through an
upper device such as a macvlan.
The idea here is to enforce the idea that an
This patch makes it so that instead of passing a void pointer as the
accel_priv we instead pass a net_device pointer as sb_dev. Making this
change allows us to pass the subordinate device through to the fallback
function eventually so that we can keep the actual code in the
ndo_select_queue call
This patch makes it so that we use the tc_to_txq mapping in the macvlan
device in order to select the Tx queue for outgoing packets.
The idea here is to try and move away from using ixgbe_select_queue and to
come up with a generic way to make this work for devices going forward. By
encoding this
For most of these calls we can just pass NULL through to the fallback
function as the sb_dev. The only cases where we cannot are the cases where
we might be dealing with either an upper device or a driver that would
have configured things to support an sb_dev itself.
The only driver that has any
This patch adds a generic version of the ndo_select_queue functions for
either returning 0 or selecting a queue based on the processor ID. This is
generally meant to just reduce the number of functions we have to change
in the future when we have to deal with ndo_select_queue changes.
+10
Yan Markman
Tel. 05-44732819
-Original Message-
From: Antoine Tenart [mailto:antoine.ten...@bootlin.com]
Sent: Monday, July 09, 2018 6:01 PM
To: da...@davemloft.net
Cc: Antoine Tenart ; netdev@vger.kernel.org;
linux-ker...@vger.kernel.org; thomas.petazz...@bootlin.com;
This change makes it so that we can support the concept of subordinate
device traffic classes to the core networking code. In doing this we can
start pulling out the driver specific bits needed to support selecting a
queue based on an upper device.
The solution at is currently stands is only
This patch series is meant to allow support for the L2 forward offload, aka
MACVLAN offload without the need for using ndo_select_queue.
The existing solution currently requires that we use ndo_select_queue in
the transmit path if we want to associate specific Tx queues with a given
MACVLAN
On Sat, Jul 7, 2018 at 7:07 AM, Neal Cardwell wrote:
> On Sat, Jul 7, 2018 at 7:15 AM David Miller wrote:
>>
>> From: Lawrence Brakmo
>> Date: Tue, 3 Jul 2018 09:26:13 -0700
>>
>> > When have observed high tail latencies when using DCTCP for RPCs as
>> > compared to using Cubic. For example, in
nf_hook() can free the skb, so we need to remove it from the list before
calling, and add passed skbs to a sublist afterwards.
Fixes: 17266ee93984 ("net: ipv4: listified version of ip_rcv")
Reported-by: Dan Carpenter
Signed-off-by: Edward Cree
---
include/linux/netfilter.h | 10 +++---
1
In netif_receive_skb_list_internal(), all of skb_defer_rx_timestamp(),
do_xdp_generic() and enqueue_to_backlog() can lead to kfree(skb). Thus,
we cannot wait until after they return to remove the skb from the list;
instead, we remove it first and, in the pass case, add it to a sublist
On 07/06/2018 11:14 PM, Sowmini Varadhan wrote:
- In net/rds/rds.h;
/* The following ports, 16385, 18634, 18635, are registered with IANA as
* the ports to be used for RDS over TCP and UDP. 18634 is the historical
What is "RDS over TCP and UDP"? There is no such thing as
On 07/07/2018 01:26 AM, Santosh Shilimkar wrote:
Hi Ka-Cheong,
On 7/6/2018 8:25 AM, Sowmini Varadhan wrote:
On (07/06/18 23:08), Ka-Cheong Poon wrote:
As mentioned in a previous mail, it is unclear why the
port number is transport specific. Most Internet services
use the same port number
On 07/09/2018 04:39 AM, Eric Dumazet wrote:
> Alternatively, you could try to patch fq_codel to drop all frags of one UDP
> datagram
> instead of few of them.
A first step would be to make sure fq_codel_hash() (using skb_get_hash(skb))
selects
the same bucket for all frags of a datagram :/
A couple of bugs in skb list handling were spotted by Dan Carpenter, with
the help of Smatch; following up on them I found a couple more similar
cases. This series fixes them by changing the relevant loops to use the
dequeue-enqueue model (rather than in-place list modification).
v2: dropped
1 - 100 of 140 matches
Mail list logo
