On Fri, Sep 21, 2018 at 07:48:25AM +, Jorgen S. Hansen wrote:
> Hi Hangbin,
>
> I finaly got to the bottom of this - the issue was indeed in the VMCI driver.
> The patch is posted here:
>
> https://lkml.org/lkml/2018/9/21/326
>
> I used your reproduce.log to test the fix. Thanks for discove
On Mon, 2018-09-17 at 13:30 +0300, Leon Romanovsky wrote:
> From: Leon Romanovsky
>
> Hi,
>
> This is short series from Mark which extends handling of loopback
> traffic. Originally mlx5 IB dynamically enabled/disabled both unicast
> and multicast based on number of users. However RAW ethernet Q
On 09/20/2018 09:52 AM, Prashant Bhole wrote:
> following commit:
> commit d58e468b1112 ("flow_dissector: implements flow dissector BPF hook")
> added struct bpf_flow_keys which conflicts with the struct with
> same name in sockex2_kern.c and sockex3_kern.c
>
> similar to commit:
> commit 534e0e52
On 09/18/2018 06:01 PM, John Fastabend wrote:
> Eric noted that using the close callback is not sufficient
> to catch all transitions from ESTABLISHED state to a LISTEN
> state. So this series does two things. First, only allow
> adding socks in ESTABLISH state and second use unhash callback
> to c
On Sat, 2018-09-22 at 00:40 +0300, Leon Romanovsky wrote:
> On Fri, Sep 21, 2018 at 04:05:53PM -0400, Doug Ledford wrote:
> > On Fri, 2018-09-21 at 22:33 +0300, Leon Romanovsky wrote:
> > > Hope it makes it clear now.
> >
> > Clear enough. Between yours and Jason's explanation I think it's well
>
In both 802.3z and SGMII modes we need to configure the MAC accordingly
to flip between Fiber and SGMII modes, and we need to read the MAC
status from the SGMII in-band control word.
Fixes: 0e01491de646 ("net: dsa: b53: Add SerDes support")
Signed-off-by: Florian Fainelli
---
drivers/net/dsa/b53
Maths went wrong, to get 0x20, we need to do 0x1e + (x) * 2, not 0x18,
fix that offset so we access the correct registers. This would make us
not access the correct SerDes Digital control words, status would be
fine and so we would not be correctly flipping between Fiber and SGMII
modes resulting i
Hi David,
Here are two additional fixes that are required in order for SGMII to
work correctly. This was discovered with using a copper SFP which would
make us use SGMII mode, we would actually leave the HW configured in its
default mode: Fiber.
Florian Fainelli (2):
net: dsa: b53: Fix B53_SERD
PHYLINK takes care of filing the right information into
state->an_enabled, get rid of the read from the SerDes's BMCR register.
Fixes: 0e01491de646 ("net: dsa: b53: Add SerDes support")
Signed-off-by: Florian Fainelli
---
drivers/net/dsa/b53/b53_serdes.c | 5 +
1 file changed, 1 insertion(+)
On Fri, 21 Sep 2018 15:27:51 -0700, Eric Dumazet wrote:
> As diagnosed by Song Liu, ndo_poll_controller() can
> be very dangerous on loaded hosts, since the cpu
> calling ndo_poll_controller() might steal all NAPI
> contexts (for all RX/TX queues of the NIC). This capture
> can last for unlimited a
Actually there's nothing wrong with the two changes, they just
revealed a problem which has been existing before.
Core of the problem is that phy_suspend() suspends the PHY when it
should not because of WoL. phy_suspend() checks for WoL already, but
this works only if the PHY driver handles WoL (w
On Wed, Sep 19, 2018 at 3:51 PM Alexei Starovoitov wrote:
>
> Recognize JITed bpf prog load/unload events.
> Add/remove kernel symbols accordingly.
>
> Signed-off-by: Alexei Starovoitov
Acked-by: Song Liu
> ---
> tools/perf/util/machine.c | 27 +++
> tools/perf/util/sy
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
can last for unlimited amount of time, since one
cpu is generally not able to drain all th
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
can last for unlimited amount of time, since one
cpu is generally not able to drain all th
On 19.09.2018 18:34, David Arendt wrote:
> Hi,
>
> the networking problem did not occur for 12 hours now, so I think this
> patch resolved the problem.
@Heiner:
It seems that the regression was introduced by your
commit 4fd48c4ac0a0 ("r8169: move common initializations to tp->hw_start").
Will yo
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
can last for unlimited amount of time, since one
cpu is generally not able to drain all th
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC).
This capture, showing one ksoftirqd eating all cycles
can last for unlimited amount of time, since one
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
can last for unlimited amount of time, since one
cpu is generally not able to drain all th
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
can last for unlimited amount of time, since one
cpu is generally not able to drain all th
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
can last for unlimited amount of time, since one
cpu is generally not able to drain all th
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
can last for unlimited amount of time, since one
cpu is generally not able to drain all th
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
can last for unlimited amount of time, since one
cpu is generally not able to drain all th
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
can last for unlimited amount of time, since one
cpu is generally not able to drain all th
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
can last for unlimited amount of time, since one
cpu is generally not able to drain all th
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
can last for unlimited amount of time, since one
cpu is generally not able to drain all th
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
can last for unlimited amount of time, since one
cpu is generally not able to drain all th
We want to allow NAPI drivers to no longer provide
ndo_poll_controller() method, as it has been proven problematic.
team driver must not look at its presence, but instead call
netpoll_poll_dev() which factorize the needed actions.
Signed-off-by: Eric Dumazet
Cc: Jay Vosburgh
Cc: Veaceslav Falic
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
lasts for unlimited amount of time, since one
cpu is generally not able to drain all the q
As diagnosed by Song Liu, ndo_poll_controller() can
be very dangerous on loaded hosts, since the cpu
calling ndo_poll_controller() might steal all NAPI
contexts (for all RX/TX queues of the NIC). This capture
can last for unlimited amount of time, since one
cpu is generally not able to drain all th
On Fri, Sep 21, 2018 at 09:25:00AM -0300, Arnaldo Carvalho de Melo wrote:
>
> > I have considered adding MUNMAP to match existing MMAP, but went
> > without it because I didn't want to introduce new bit in perf_event_attr
> > and emit these new events in a misbalanced conditional way for prog
>
On Fri, Sep 21, 2018 at 04:05:53PM -0400, Doug Ledford wrote:
> On Fri, 2018-09-21 at 22:33 +0300, Leon Romanovsky wrote:
> > Hope it makes it clear now.
>
> Clear enough. Between yours and Jason's explanation I think it's well
> covered.
>
> > Are you ok with me/Saeed taking first patch to our br
Summary:
This appears to be necessary and sufficient change to enable `MPLS` on
`ip6gre` tunnels (RFC4023).
This diff allows IP6GRE devices to be recognized by MPLS kernel module
and hence user can configure interface to accept packets with mpls
headers as well setup mpls routes on them.
Test Pl
Summary:
This appears to be necessary and sufficient change to enable `MPLS` on
`ip6gre` tunnels (RFC4023).
This diff allows IP6GRE devices to be recognized by MPLS kernel module
and hence user can configure interface to accept packets with mpls
headers as well setup mpls routes on them.
Test Pl
Signed-off-by: Petr Vorel
---
Makefile | 4
1 file changed, 4 insertions(+)
diff --git a/Makefile b/Makefile
index 25de3893..b7488add 100644
--- a/Makefile
+++ b/Makefile
@@ -77,6 +77,7 @@ help:
@echo " clean - remove products of build"
@echo " distclean
On 9/21/18 9:16 AM, Stephen Hemminger wrote:
> Add help target to Makefile.
>
> Signed-off-by: Stephen Hemminger
> ---
> Makefile | 12
> 1 file changed, 12 insertions(+)
>
Acked-by: David Ahern
On Fri, 2018-09-21 at 22:33 +0300, Leon Romanovsky wrote:
> Hope it makes it clear now.
Clear enough. Between yours and Jason's explanation I think it's well
covered.
> Are you ok with me/Saeed taking first patch to our branch so you will be
> able to take the rest?
Yep. Let me know a tag when
On Fri, Sep 21, 2018 at 03:14:36PM -0400, Doug Ledford wrote:
> On Mon, 2018-09-17 at 13:30 +0300, Leon Romanovsky wrote:
> > From: Leon Romanovsky
> >
> > Hi,
> >
> > This is short series from Mark which extends handling of loopback
> > traffic. Originally mlx5 IB dynamically enabled/disabled b
On Thu, Sep 20, 2018 at 12:36 AM Vlad Buslov wrote:
>
>
> On Wed 19 Sep 2018 at 22:09, Cong Wang wrote:
> > On Mon, Sep 17, 2018 at 12:19 AM Vlad Buslov wrote:
> >> @@ -482,16 +483,25 @@ static int tcf_block_insert(struct tcf_block *block,
> >> struct net *net,
> >>
On Fri, Sep 21, 2018 at 03:14:36PM -0400, Doug Ledford wrote:
> On Mon, 2018-09-17 at 13:30 +0300, Leon Romanovsky wrote:
> > From: Leon Romanovsky
> >
> > Hi,
> >
> > This is short series from Mark which extends handling of loopback
> > traffic. Originally mlx5 IB dynamically enabled/disabled bot
On Thu, Sep 20, 2018 at 12:21 AM Vlad Buslov wrote:
>
>
> On Wed 19 Sep 2018 at 22:04, Cong Wang wrote:
> > On Mon, Sep 17, 2018 at 12:19 AM Vlad Buslov wrote:
> >> +static void tcf_qdisc_put(struct Qdisc *q, bool rtnl_held)
> >> +{
> >> + if (!q)
> >> + return;
> >> +
> >> +
On Mon, 2018-09-17 at 13:30 +0300, Leon Romanovsky wrote:
> From: Leon Romanovsky
>
> Hi,
>
> This is short series from Mark which extends handling of loopback
> traffic. Originally mlx5 IB dynamically enabled/disabled both unicast
> and multicast based on number of users. However RAW ethernet Q
On Thu, Sep 20, 2018 at 1:58 AM Li RongQing wrote:
>
> if skb->head is vmalloc address, when this skb is delivered, full
> allocation for this skb is required, if there are many devices,
> the full allocation will be called for every devices
So why do you in practice need many netlink tap devices
On Wed, Sep 19, 2018 at 6:04 AM Paolo Abeni wrote:
> diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
> index 419960b0ba16..a0b6932c3afd 100644
> --- a/net/ipv6/ip6_tunnel.c
> +++ b/net/ipv6/ip6_tunnel.c
> @@ -1234,7 +1234,7 @@ static inline int
> ip4ip6_tnl_xmit(struct sk_buff *skb, st
On 2018-09-21 11:40, Eric Dumazet wrote:
On 09/21/2018 10:09 AM, stran...@codeaurora.org wrote:
I also tried reverting 7f6b9dbd5afb ("af_key: locking change") and
running the
test there and I still see the crash, so it doesn't seem to be an RCU
specific
issue.
Is there anything else that cou
Currently, ip[6]frag_high_thresh sysctl values in new namespaces are
hard-limited to those of the root/init ns.
There are at least two use cases when it would be desirable to
set the high_thresh values higher in a child namespace vs the global hard
limit:
- a security/ddos protection policy may l
This is similar to how ipv4 now behaves:
commit 0ff89efb5246 ("ip: fail fast on IP defrag errors").
Signed-off-by: Peter Oskolkov
---
net/ipv6/reassembly.c | 11 ++-
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index f1b1ff30
On Fri, 2018-09-21 at 09:35 +0200, Björn Töpel wrote:
> > --- a/drivers/net/ethernet/intel/i40e/i40e_xsk.c
> > +++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.c
> > @@ -830,3 +830,33 @@ int i40e_xsk_async_xmit(struct net_device
> > *dev, u32 queue_id)
> >
> > return 0;
> > }
> > +
> > +/
This patch adds ipv6 defragmentation tests to ip_defrag selftest,
to complement existing ipv4 tests.
Signed-off-by: Peter Oskolkov
---
tools/testing/selftests/net/ip_defrag.c | 249 +++
tools/testing/selftests/net/ip_defrag.sh | 39 ++--
2 files changed, 190 insertions(+),
On 9/21/18 10:58 AM, Eric Dumazet wrote:
> net/ipv4/fib_frontend.c: In function 'fib_info_nh_uses_dev':
> net/ipv4/fib_frontend.c:322:6: error: unused variable 'ret'
> [-Werror=unused-variable]
> cc1: all warnings being treated as errors
>
> Fixes: 78f2756c5fc0 ("net/ipv4: Move device validation
On 09/21/2018 10:58 AM, Eric Dumazet wrote:
> net/ipv4/fib_frontend.c: In function 'fib_info_nh_uses_dev':
> net/ipv4/fib_frontend.c:322:6: error: unused variable 'ret'
> [-Werror=unused-variable]
> cc1: all warnings being treated as errors
>
> Fixes: 78f2756c5fc0 ("net/ipv4: Move device valid
net/ipv4/fib_frontend.c: In function 'fib_info_nh_uses_dev':
net/ipv4/fib_frontend.c:322:6: error: unused variable 'ret'
[-Werror=unused-variable]
cc1: all warnings being treated as errors
Fixes: 78f2756c5fc0 ("net/ipv4: Move device validation to helper")
Signed-off-by: Eric Dumazet
Cc: David Ah
On 09/21/2018 10:09 AM, stran...@codeaurora.org wrote:
> I also tried reverting 7f6b9dbd5afb ("af_key: locking change") and running the
> test there and I still see the crash, so it doesn't seem to be an RCU specific
> issue.
>
> Is there anything else that could be causing this?
What about y
Hello!
You forgot to CC me. :-/
On 09/21/2018 04:52 PM, Andrew Lunn wrote:
> The previous commit to ravb had the side effect of making the PHY
> advertise Pause and Asym Pause, which previously did not happen. By
> default, phydev->supported has both forms of pause enabled, but
> phydev->adv
Document the new pointer types in the verifier and how the pointer ID
tracking works to ensure that references which are taken are later
released.
Signed-off-by: Joe Stringer
Acked-by: Alexei Starovoitov
---
Documentation/networking/filter.txt | 64 +
1 file changed,
Add some tests that demonstrate and test the balanced lookup/free
nature of socket lookup. Section names that start with "fail" represent
programs that are expected to fail verification; all others should
succeed.
Signed-off-by: Joe Stringer
Acked-by: Alexei Starovoitov
---
tools/testing/selfte
reference tracking: leak potential reference
reference tracking: leak potential reference on stack
reference tracking: leak potential reference on stack 2
reference tracking: zero potential reference
reference tracking: copy and zero potential references
reference tracking: release reference withou
Allow the individual program load to be invoked. This will help with
testing, where a single ELF may contain several sections, some of which
denote subprograms that are expected to fail verification, along with
some which are expected to pass verification. By allowing programs to be
iterated and in
Teach the verifier a little bit about a new type of pointer, a
PTR_TO_SOCKET. This pointer type is accessed from BPF through the
'struct bpf_sock' structure.
Signed-off-by: Joe Stringer
---
v2: Reuse reg_type_mismatch() in more places
Reduce the number of passes at convert_ctx_access()
---
An upcoming commit will need very similar copy/realloc boilerplate, so
refactor the existing stack copy/realloc functions into macros to
simplify it.
Signed-off-by: Joe Stringer
Acked-by: Alexei Starovoitov
---
kernel/bpf/verifier.c | 106 --
1 file chang
This patch adds new BPF helper functions, bpf_sk_lookup_tcp() and
bpf_sk_lookup_udp() which allows BPF programs to find out if there is a
socket listening on this host, and returns a socket pointer which the
BPF program can then access to determine, for instance, whether to
forward or drop traffic.
An upcoming commit will add another two pointer types that need very
similar behaviour, so generalise this function now.
Signed-off-by: Joe Stringer
Acked-by: Alexei Starovoitov
---
kernel/bpf/verifier.c | 22 ++---
tools/testing/selftests/bpf/test_verifier
Allow helper functions to acquire a reference and return it into a
register. Specific pointer types such as the PTR_TO_SOCKET will
implicitly represent such a reference. The verifier must ensure that
these references are released exactly once in each path through the
program.
To achieve this, this
This check will be reused by an upcoming commit for conditional jump
checks for sockets. Refactor it a bit to simplify the later commit.
Signed-off-by: Joe Stringer
Acked-by: Alexei Starovoitov
---
kernel/bpf/verifier.c | 43 +--
1 file changed, 25 insert
Add this iterator for spilled registers, it concentrates the details of
how to get the current frame's spilled registers into a single macro
while clarifying the intention of the code which is calling the macro.
Signed-off-by: Joe Stringer
Acked-by: Alexei Starovoitov
---
include/linux/bpf_veri
This series proposes a new helper for the BPF API which allows BPF programs to
perform lookups for sockets in a network namespace. This would allow programs
to determine early on in processing whether the stack is expecting to receive
the packet, and perform some action (eg drop, forward somewhere)
As long as one skb has sock_rfree has its destructor, the socket
attached to
this skb can not be released. There is no race here.
Note that skb_clone() does not propagate the destructor.
The issue here is that in the rcu lookup, we can find a socket that
has been
dismantled, with a 0 refcou
> On Fri, 21 Sep 2018 15:34:25 +0200
> Lorenzo Bianconi wrote:
>
> > Take into account the configured preferred_family if neither saddr or
> > daddr are provided since otherwise vxlan kernel module will use IPv4 as
> > default remote inet family neglecting the one provided by userspace.
> > This
On Thu, 20 Sep 2018 01:36:21 +0200
Petr Vorel wrote:
> Hi,
>
> here are simply fixes to restore 'make alltests'.
> Currently it does not run.
>
> Kind regards,
> Petr
>
> Petr Vorel (3):
> testsuite: Fix missing generate_nlmsg
> testsuite: Generate generate_nlmsg when needed
> testsuite:
Add help target to Makefile.
Signed-off-by: Stephen Hemminger
---
Makefile | 12
1 file changed, 12 insertions(+)
diff --git a/Makefile b/Makefile
index ea2f797c933f..25de3893cae4 100644
--- a/Makefile
+++ b/Makefile
@@ -71,6 +71,18 @@ all: config.mk
for i in $(SUBDIRS); \
David Ahern wrote:
> On 9/21/18 8:55 AM, Florian Westphal wrote:
> > David Ahern wrote:
> >>> David, i hope this will silence the warning, would be nice
> >>> if you could test it.
> >>
> >> I still the warning.
> >
> > Wait. Do you see this warning everywhere or just in xfrm?
> >
>
> just
On Fri, 21 Sep 2018 15:34:25 +0200
Lorenzo Bianconi wrote:
> Take into account the configured preferred_family if neither saddr or
> daddr are provided since otherwise vxlan kernel module will use IPv4 as
> default remote inet family neglecting the one provided by userspace.
> This behaviour was
On 9/21/18 8:55 AM, Florian Westphal wrote:
> David Ahern wrote:
>>> David, i hope this will silence the warning, would be nice
>>> if you could test it.
>>
>> I still the warning.
>
> Wait. Do you see this warning everywhere or just in xfrm?
>
just the one file.
David Ahern wrote:
> > David, i hope this will silence the warning, would be nice
> > if you could test it.
>
> I still the warning.
Wait. Do you see this warning everywhere or just in xfrm?
On 9/21/18 3:35 AM, Florian Westphal wrote:
> Unfortunately some versions of gcc emit following warning:
> linux/compiler.h:252:20: warning: array subscript is above array bounds
> [-Warray-bounds]
> hook_head = rcu_dereference(net->nf.hooks_arp[hook]);
> ^~~~
Next patch will use tcp_wstamp_ns to feed internal
TCP pacing timer, so switch to CLOCK_TAI to share same base.
Signed-off-by: Eric Dumazet
---
net/ipv4/tcp_output.c | 2 +-
net/ipv4/tcp_timer.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/ipv4/tcp_output.c b/net/i
Switch internal TCP skb->skb_mstamp to skb->skb_mstamp_ns,
from usec units to nsec units.
Do not clear skb->tstamp before entering IP stacks in TX,
so that qdisc or devices can implement pacing based on the
earliest departure time instead of socket sk->sk_pacing_rate
Packets are fed with tcp_wsta
With the earliest departure time model, we no longer plan
special casing TCP retransmits. We therefore remove dead
code (since most compilers understood skb_is_retransmit()
was false)
Signed-off-by: Eric Dumazet
---
net/sched/sch_fq.c | 58 --
1 file c
TCP keeps track of tcp_wstamp_ns by itself, meaning sch_fq
no longer has to do it.
Thanks to this model, TCP can get more accurate RTT samples,
since pacing no longer inflates them.
This has the nice effect of removing some delays caused by FQ
quantum mechanism, causing inflated max/P99 latencies
Now TCP keeps track of tcp_wstamp_ns, recording the earliest
departure time of next packet, we can remove duplicate code
from tcp_internal_pacing()
This removes one ktime_get_tai_ns() call, and a divide.
Signed-off-by: Eric Dumazet
---
net/ipv4/tcp_output.c | 17 -
1 file change
In the early days, pacing has been implemented in sch_fq (FQ)
in a generic way :
- SO_MAX_PACING_RATE could be used by any sockets.
- TCP would vary effective pacing rate based on CWND*MSS/SRTT
- FQ would ensure delays between packets based on current
sk->sk_pacing_rate, but with some quantum
TCP will soon provide per skb->tstamp with earliest departure time,
so that sch_fq does not have to determine departure time by looking
at socket sk_pacing_rate.
We chose in linux-4.19 CLOCK_TAI as the clock base for transports,
qdiscs, and NIC offloads.
Signed-off-by: Eric Dumazet
---
net/sche
TCP will soon provide earliest departure time on TX skbs.
It needs to track this in a new variable.
tcp_mstamp_refresh() needs to update this variable, and
became too big to stay an inline.
Signed-off-by: Eric Dumazet
---
include/linux/tcp.h | 2 ++
include/net/tcp.h | 12 +---
n
There are few places where TCP reads skb->skb_mstamp expecting
a value in usec unit.
skb->tstamp (aka skb->skb_mstamp) will soon store CLOCK_TAI nsec value.
Add tcp_skb_timestamp_us() to provide proper conversion when needed.
Signed-off-by: Eric Dumazet
---
include/net/tcp.h | 8 +++
TCP pacing is either implemented in sch_fq or internally.
We have the goal of being able to offload pacing on the NICS.
TCP will soon provide per skb skb->tstamp as early departure time.
Like ETF in commit 25db26a91364 ("net/sched: Introduce the ETF Qdisc")
we chose CLOCK_T as the clock base, so
On Fri, 21 Sep 2018 16:16:31 +0530, Arjun Vynipadath wrote:
> On Tuesday, September 09/18/18, 2018 at 11:39:14 -0700, Jakub Kicinski wrote:
> > On Tue, 18 Sep 2018 18:37:23 +0530, Arjun Vynipadath wrote:
> > > Forcing link up of virtual interfaces even when physical link is down
> > > causes pack
On 09/21/2018 08:14 AM, Alexei Starovoitov wrote:
> On 9/21/18 7:59 AM, Eric Dumazet wrote:
>>
>>
>> On 09/21/2018 07:55 AM, Alexei Starovoitov wrote:
>>
>>>
>>> should we remove ndo_poll_controller then?
>>> My understanding that the patch helps by not letting
>>> drivers do napi_schedule() for
On 9/21/18 7:59 AM, Eric Dumazet wrote:
>
>
> On 09/21/2018 07:55 AM, Alexei Starovoitov wrote:
>
>>
>> should we remove ndo_poll_controller then?
>> My understanding that the patch helps by not letting
>> drivers do napi_schedule() for all queues into this_cpu, right?
>> But most of the drivers do
On 09/21/2018 07:55 AM, Alexei Starovoitov wrote:
>
> should we remove ndo_poll_controller then?
> My understanding that the patch helps by not letting
> drivers do napi_schedule() for all queues into this_cpu, right?
> But most of the drivers do exactly that in their ndo_poll_controller
> imp
On 9/21/18 6:33 AM, Eric Dumazet wrote:
On 09/21/2018 12:17 AM, Song Liu wrote:
On Sep 20, 2018, at 4:49 PM, Eric Dumazet wrote:
On 09/20/2018 04:43 PM, Song Liu wrote:
I tried to totally skip ndo_poll_controller() here. It did avoid hitting
the issue. However, netpoll will drop (f
On 09/20/2018 08:27 PM, Li,Rongqing wrote:
>
> The below change seems simple, but it increase skb allocation and
> free one time,
Seem fine to me. An extra skb_clone() for vmalloc-skb-users is absolute noise,
compared to vmalloc()vfree() cost.
Thanks.
From: Lorenzo Colitti
Date: Fri, 21 Sep 2018 18:46:25 +0800
> Would you take a patch to add a one-line comment saying that this is
> the way it is for backwards compatibility? If that comment were there
> anyone else who finds this will not spend time debugging it and
> immediately know what's go
From: Xin Long
Date: Fri, 21 Sep 2018 15:55:34 +0800
> It's under the protection of the sock lock, I think any other places
> that want to access the address also need to acquire this sock lock
> first.
Hash table lookups don't even have a socket context yet, so can't hold
the sock lock, but loo
From: David Ahern
Date: Thu, 20 Sep 2018 21:28:43 -0700
> I need some time to review and more importantly test this patch set
> before it is committed. I am traveling tomorrow afternoon through Sunday
> evening, so I need a few days into next week to get to this.
Sure, no problem.
On Fri, Sep 21, 2018 at 09:25:00AM -0300, Arnaldo Carvalho de Melo wrote:
> There is another longstanding TODO list entry: PERF_RECORD_MMAP records
> should include a build-id
I throught the problem was that the kernel doesn't have the build-id in
the first place. So it cannot hand them out.
On Fri, Sep 21, 2018 at 09:25:00AM -0300, Arnaldo Carvalho de Melo wrote:
> > I consider synthetic perf events to be non-ABI. Meaning they're
> > emitted by perf user space into perf.data and there is a convention
> > on names, but it's not a kernel abi. Like RECORD_MMAP with
> > event.filename ==
On 09/21/2018 02:19 AM, Jose Abreu wrote:
> Hello,
>
> I'm getting a race in stmmac coalesce timer and the
> napi_schedule() interrupt and I'm asking for advice. Currently,
> we are scheduling NAPI in coalesce timer but this leads to
> stmmac_tx_clean() deadlock because this function tries to a
The previous commit to ravb had the side effect of making the PHY
advertise Pause and Asym Pause, which previously did not happen. By
default, phydev->supported has both forms of pause enabled, but
phydev->advertising does not. The new phy_remove_link_mode() copies
phydev->supported to phydev->adv
On Fri, Sep 14, 2018 at 10:17 PM, Jesse Brandeburg
wrote:
> On Fri, 14 Sep 2018 12:10:45 +0300 Or wrote:
>> On Fri, Sep 14, 2018 at 1:31 AM, Jesse Brandeburg
>> wrote:
>> on what HW ring format do you standardize? do i40e/Fortville and
>> ice/what's-the-intel-code-name? HWs can/use the same post
Take into account the configured preferred_family if neither saddr or
daddr are provided since otherwise vxlan kernel module will use IPv4 as
default remote inet family neglecting the one provided by userspace.
This behaviour was originally in commit 97d564b90ccb ("vxlan: use
preferred address fami
On 09/21/2018 12:17 AM, Song Liu wrote:
>
>
>> On Sep 20, 2018, at 4:49 PM, Eric Dumazet wrote:
>>
>>
>>
>> On 09/20/2018 04:43 PM, Song Liu wrote:
>>>
>>
>>> I tried to totally skip ndo_poll_controller() here. It did avoid hitting
>>> the issue. However, netpoll will drop (fail to send) more
1 - 100 of 116 matches
Mail list logo
