If skb_recv_datagram returns an skb, we should ignore the err
value returned. Otherwise, datagram receives will return EAGAIN
when they have to wait for a datagram.
Acked-by: Adit Ranadive
Signed-off-by: Jorgen Hansen
---
net/vmw_vsock/vmci_transport.c |7 ++-
1 files changed, 2 inserti
On Mon, Apr 18, 2016 at 04:55:35PM +0200, Julia Lawall wrote:
> Add __init attribute on a function that is only called from other __init
> functions and that is not inlined, at least with gcc version 4.8.4 on an
> x86 machine with allyesconfig. Currently, the function is put in the
> .text.unlikel
On Mon, Apr 18, 2016 at 6:46 PM, Martin KaFai Lau wrote:
> After receiving sacks, tcp_shifted_skb() will collapse
> skbs if possible. tx_flags/tskey/txstamp_ack also has
> to be merged in this case.
>
> This patch resues the tcp_skb_collapse_tstamp() to handle
> them.
>
> BPF Output Before:
> ~~~
On Mon, Apr 18, 2016 at 6:46 PM, Martin KaFai Lau wrote:
> If two skbs are merged/collapsed during retransmission, the current
> logic does not merge the tx_flags, tskey and txstamp_ack. The end
> result is the SCM_TSTAMP_ACK timestamp could be missing for a
> packet that the end-user has specifi
On Mon, Apr 18, 2016 at 6:46 PM, Martin KaFai Lau wrote:
> When a tcp skb is sliced into two smaller skbs (e.g. in
> tcp_fragment() and tso_fragment()), it does not carry
> the txstamp_ack bit to the newly created skb if it is needed.
> The end result is a timestamping event (SCM_TSTAMP_ACK) will
The bulk of the following patches have been sitting in Eugene's Github tree for
quite some time. They fix various issues existing in the mainline drivers, so
they should be merged there too.
Also included are two new fixes, of my own; the important one being the
reordering of deletion of the bss,
From: Pontus Fuchs
Needed for coming improvements. No functional changes.
Signed-off-by: Pontus Fuchs
[bjorn: restored BEACON_TEMPLATE_SIZE define to 0x180]
Signed-off-by: Bjorn Andersson
---
Changes since v2:
- Subtract sizeof(beacon_length) instead of modifying BEACON_TEMPLATE_SIZE,
which
From: Arnd Bergmann
Date: Mon, 18 Apr 2016 23:58:30 +0200
> The w5100-spi driver front-end only makes sense when the w5100
> core driver is enabled, not for a configuration that only has w5300:
...
> This adds an appropriate Kconfig dependency.
>
> Signed-off-by: Arnd Bergmann
> Fixes: 630cf09
From: Pontus Fuchs
While poking at this I also change two related things. I rename one
variable to make the names consistent. I also move one assignment of
priv_sta to the declaration to save a few lines.
Signed-off-by: Pontus Fuchs
Signed-off-by: Bjorn Andersson
---
drivers/net/wireless/ath/
From: Pontus Fuchs
The wcn36xx FW expects a fixed size TIM PVM in the beacon template. If
supplied with a shorter than expected PVM it will overwrite the IE
following the TIM.
Squashed with fix from Jason Mobarak :
Patch "wcn36xx: Pad TIM PVM if needed" has caused a regression in mesh
beaconing.
From: Pontus Fuchs
Signed-off-by: Pontus Fuchs
Signed-off-by: Bjorn Andersson
---
drivers/net/wireless/ath/wcn36xx/hal.h | 2 +-
drivers/net/wireless/ath/wcn36xx/main.c | 4 ++--
drivers/net/wireless/ath/wcn36xx/smd.c | 2 +-
3 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/dri
From: Pontus Fuchs
Makes the code a little easier to read.
Signed-off-by: Pontus Fuchs
Signed-off-by: Bjorn Andersson
---
drivers/net/wireless/ath/wcn36xx/debug.c | 12 +++-
drivers/net/wireless/ath/wcn36xx/main.c| 16 +++-
drivers/net/wireless/ath/wcn36xx/pmc.c
From: Pontus Fuchs
This does not work with multiple sta's in a vif.
Signed-off-by: Pontus Fuchs
Signed-off-by: Bjorn Andersson
---
drivers/net/wireless/ath/wcn36xx/main.c| 3 ---
drivers/net/wireless/ath/wcn36xx/smd.c | 28 +++-
drivers/net/wireless/ath/wcn36x
From: Pontus Fuchs
For consistency with other code.
Signed-off-by: Pontus Fuchs
Signed-off-by: Bjorn Andersson
---
Changes since v1:
- Reordered after the now previous patch, to make wcn36xx_sta_to_priv()
available before we use it
drivers/net/wireless/ath/wcn36xx/main.c | 2 +-
1 file ch
From: Pontus Fuchs
Pass the multicast list to FW.
This patch also adds a way to build the smd command in place. This is
needed because the MC list command is too big for the stack.
Signed-off-by: Pontus Fuchs
[bjorn: dropped FIF_PROMISC_IN_BSS usage]
Signed-off-by: Bjorn Andersson
---
driver
From: Pontus Fuchs
Signed-off-by: Pontus Fuchs
Signed-off-by: Bjorn Andersson
---
drivers/net/wireless/ath/wcn36xx/smd.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/wireless/ath/wcn36xx/smd.c
b/drivers/net/wireless/ath/wcn36xx/smd.c
index ebb446272d21..e0d5
From: Pontus Fuchs
Knowing the association state is needed for mc filtering.
Signed-off-by: Pontus Fuchs
Signed-off-by: Bjorn Andersson
---
drivers/net/wireless/ath/wcn36xx/main.c| 2 ++
drivers/net/wireless/ath/wcn36xx/wcn36xx.h | 1 +
2 files changed, 3 insertions(+)
diff --git a/drive
From: Pontus Fuchs
This message does not follow the canonical format and needs it's own
parser.
Signed-off-by: Pontus Fuchs
Signed-off-by: Bjorn Andersson
---
drivers/net/wireless/ath/wcn36xx/smd.c | 14 --
1 file changed, 12 insertions(+), 2 deletions(-)
diff --git a/drivers/net
From: Eric Dumazet
Date: Mon, 18 Apr 2016 21:32:04 -0700
> On Mon, 2016-04-18 at 23:41 -0400, David Miller wrote:
>>
>> +/* Add a zero length NOP attribute so that the nla_data()
>> + * of the IFLA_STATS64 will be 64-bit aligned.
>> + */
>> +#ifndef HAVE_EFFICIENT_UNALIGNED_ACCESS
>
From: Pontus Fuchs
EXIT_BMPS_REQ was using the command struct for ENTER_BMPS_REQ. I
spotted this when looking at command dumps.
Signed-off-by: Pontus Fuchs
Signed-off-by: Bjorn Andersson
---
drivers/net/wireless/ath/wcn36xx/smd.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --g
From: Pontus Fuchs
Some code used priv_vif and some used vif_priv. Convert all to vif_priv
for consistency.
Signed-off-by: Pontus Fuchs
Signed-off-by: Bjorn Andersson
---
drivers/net/wireless/ath/wcn36xx/smd.c | 28 ++--
1 file changed, 14 insertions(+), 14 deletions(-
From: Pontus Fuchs
When converting to version 1 of the config_sta struct not all
members where copied. This fixes the problem of multicast frames
not being delivered on an encrypted network.
Signed-off-by: Pontus Fuchs
Signed-off-by: Bjorn Andersson
---
drivers/net/wireless/ath/wcn36xx/smd.c
The WCN36XX_HAL_RMV_BSSKEY_RSP carries a single u32 with "status", so we
can use the standard status check function for decoding the result.
This is the last user of the v2 status checker, so remove the struct and
helper function.
Signed-off-by: Bjorn Andersson
---
Changes since v2:
- Added thi
When disabling the beacon we must delete the bss before idling the link.
Signed-off-by: Bjorn Andersson
---
Changes since v2:
- Added this patch to the series
drivers/net/wireless/ath/wcn36xx/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/wireless/ath/wc
Fill in the capability list with more values from the downstream driver.
Signed-off-by: Bjorn Andersson
---
Changes since v2:
- Added this patch to the series
drivers/net/wireless/ath/wcn36xx/hal.h | 39
drivers/net/wireless/ath/wcn36xx/main.c | 40 +++
From: Pontus Fuchs
This fixes a problem connecting to an open network after being
connected to an encrypted network.
Signed-off-by: Pontus Fuchs
Signed-off-by: Bjorn Andersson
---
drivers/net/wireless/ath/wcn36xx/main.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/wireles
On 4/18/16, 8:41 PM, David Miller wrote:
> From: Roopa Prabhu
> Date: Mon, 18 Apr 2016 14:10:19 -0700
>
>> This patch adds a new RTM_GETSTATS message to query link stats via
>> netlink from the kernel. RTM_NEWLINK also dumps stats today, but
>> RTM_NEWLINK returns a lot more than just stats and is
On Mon, 2016-04-18 at 23:41 -0400, David Miller wrote:
>
> + /* Add a zero length NOP attribute so that the nla_data()
> + * of the IFLA_STATS64 will be 64-bit aligned.
> + */
> +#ifndef HAVE_EFFICIENT_UNALIGNED_ACCESS
> + attr = nla_reserve(skb, IFLA_PAD, 0);
> + if (!attr)
On Mon, 2016-04-18 at 23:41 -0400, David Miller wrote:
> From: Roopa Prabhu
> Date: Mon, 18 Apr 2016 14:10:19 -0700
>
> > This patch adds a new RTM_GETSTATS message to query link stats via
> > netlink from the kernel. RTM_NEWLINK also dumps stats today, but
> > RTM_NEWLINK returns a lot more than
From: David Miller
Date: Mon, 18 Apr 2016 21:48:51 -0400 (EDT)
> I think the time has probably come to have a new netlink attribute
> format that doesn't have this multi-decade old problem.
...
Scratch this whole idea, I think the padding attribute works a lot
better and is backwards compatible
From: Roopa Prabhu
Date: Mon, 18 Apr 2016 19:40:08 -0700
> On 4/18/16, 7:22 PM, Eric Dumazet wrote:
>> On Mon, 2016-04-18 at 21:48 -0400, David Miller wrote:
>>
>>> And anyways, I get unaligned accesses without Roopa's changes :-/
>>>
>>> davem@patience:~$ ip l l
>>> [3391066.656729] Kernel unali
From: Roopa Prabhu
Date: Mon, 18 Apr 2016 14:10:19 -0700
> This patch adds a new RTM_GETSTATS message to query link stats via
> netlink from the kernel. RTM_NEWLINK also dumps stats today, but
> RTM_NEWLINK returns a lot more than just stats and is expensive in
> some cases when frequent polling
On Mon, 2016-04-18 at 20:18 -0700, Martin KaFai Lau wrote:
> On Mon, Apr 18, 2016 at 07:50:41PM -0700, Eric Dumazet wrote:
> > I believe it is slightly wrong (to do the goto new_segment if there is
> > no data to send)
> Aha. Thanks for pointing it out.
>
> >
> > I would instead use this fast path
On Mon, Apr 18, 2016 at 07:50:41PM -0700, Eric Dumazet wrote:
> I believe it is slightly wrong (to do the goto new_segment if there is
> no data to send)
Aha. Thanks for pointing it out.
>
> I would instead use this fast path, doing the test _when_ we already
> have an skb to test for.
The v1 was
move trace_call_bpf() into helper function to minimize the size
of perf_trace_*() tracepoint handlers.
text data bss dechex filename
105416795526646 2945024 190133491221ee5 vmlinux_before
105094225526646 2945024 18981092121a0e4 vmli
On Mon, 18 Apr 2016 18:15:04 -0700
Alexei Starovoitov wrote:
> On 4/18/16 3:16 PM, Steven Rostedt wrote:
> Yes. That what I referred to in below 'a struct to pass args'...
> But, fine, will try to optimize the size further.
> Frankly much bigger .text savings will come from combining
> trace_eve
On Mon, 2016-04-18 at 19:27 -0700, Martin KaFai Lau wrote:
> On Mon, Apr 18, 2016 at 05:06:57PM -0700, Eric Dumazet wrote:
> > It should only be a request from user space to ask TCP to not aggregate
> > stuff on future sendpage()/sendmsg() on the skb carrying this new flag.
> >
> How about somethin
On 4/18/16, 7:22 PM, Eric Dumazet wrote:
> On Mon, 2016-04-18 at 21:48 -0400, David Miller wrote:
>
>> And anyways, I get unaligned accesses without Roopa's changes :-/
>>
>> davem@patience:~$ ip l l
>> [3391066.656729] Kernel unaligned access at TPC[7d6c14]
>> loopback_get_stats64+0x74/0xa0
>> [3
Hi Alexei,
On Sat, Apr 16, 2016 at 05:47:42PM -0700, Alexei Starovoitov wrote:
> On Sat, Apr 16, 2016 at 10:29:33PM +0200, Arnd Bergmann wrote:
> > Two new functions in bpf contain a cast from a 'u64' to a
> > pointer. This works on 64-bit architectures but causes a warning
> > on all 32-bit archi
On 4/18/16, 5:57 PM, David Miller wrote:
> From: Eric Dumazet
> Date: Mon, 18 Apr 2016 14:35:56 -0700
>
>> On Mon, 2016-04-18 at 14:10 -0700, Roopa Prabhu wrote:
>>
>>> + if (filter_mask & IFLA_STATS_FILTER_BIT(IFLA_STATS_LINK_64)) {
>>> + struct rtnl_link_stats64 *sp;
>>> +
>>> +
On Mon, Apr 18, 2016 at 05:06:57PM -0700, Eric Dumazet wrote:
> It should only be a request from user space to ask TCP to not aggregate
> stuff on future sendpage()/sendmsg() on the skb carrying this new flag.
>
How about something like this. Please advise if tcp_sendmsg_noappend can
be simpler.
On Mon, 2016-04-18 at 21:48 -0400, David Miller wrote:
>
> And anyways, I get unaligned accesses without Roopa's changes :-/
>
> davem@patience:~$ ip l l
> [3391066.656729] Kernel unaligned access at TPC[7d6c14]
> loopback_get_stats64+0x74/0xa0
> [3391066.672020] Kernel unaligned access at TPC[
From: David Miller
Date: Mon, 18 Apr 2016 20:57:55 -0400 (EDT)
> From: Eric Dumazet
> Date: Mon, 18 Apr 2016 14:35:56 -0700
>
>> On Mon, 2016-04-18 at 14:10 -0700, Roopa Prabhu wrote:
>>
>>> + if (filter_mask & IFLA_STATS_FILTER_BIT(IFLA_STATS_LINK_64)) {
>>> + struct rtnl_link_sta
On 4/18/16 3:16 PM, Steven Rostedt wrote:
On Mon, 18 Apr 2016 14:43:07 -0700
Alexei Starovoitov wrote:
I was worried about this too, but single 'if' and two calls
(as in commit 98b5c2c65c295) is a better way, since it's faster, cleaner
and doesn't need to refactor the whole perf_trace_buf_sub
From: Eric Dumazet
Date: Mon, 18 Apr 2016 14:35:56 -0700
> On Mon, 2016-04-18 at 14:10 -0700, Roopa Prabhu wrote:
>
>> +if (filter_mask & IFLA_STATS_FILTER_BIT(IFLA_STATS_LINK_64)) {
>> +struct rtnl_link_stats64 *sp;
>> +
>> +attr = nla_reserve(skb, IFLA_STATS_LINK_64
On Mon, 2016-04-18 at 16:43 -0700, ka...@fb.com wrote:
> >
> > netperf could then get an option to set this MSG_EOR ;)
> Not sure how it is related. Can you share how netperf can
> benefit from MSG_EOR in TCP tests without any of the
> SOF_TIMESTAMPING_TX_RECORD_MASK.
Simply setting MSG_EOR woul
On Mon, Apr 18, 2016 at 04:18:13PM -0700, Eric Dumazet wrote:
> On Mon, 2016-04-18 at 15:46 -0700, Martin KaFai Lau wrote:
> > This patch allows the user process to use MSG_EOR during
> > tcp_sendmsg to tell the kernel that it is the last byte
> > of an application response message.
> >
> > It is c
On Mon, 2016-04-18 at 15:46 -0700, Martin KaFai Lau wrote:
> This patch allows the user process to use MSG_EOR during
> tcp_sendmsg to tell the kernel that it is the last byte
> of an application response message.
>
> It is currently useful when the end-user has turned on any bit of the
> SOF_TIME
If two skbs are merged/collapsed during retransmission, the current
logic does not merge the tx_flags, tskey and txstamp_ack. The end
result is the SCM_TSTAMP_ACK timestamp could be missing for a
packet that the end-user has specifically turned on
SOF_TIMESTAMPING_TX_ACK (e.g. by cmsg).
The patch
When a tcp skb is sliced into two smaller skbs (e.g. in
tcp_fragment() and tso_fragment()), it does not carry
the txstamp_ack bit to the newly created skb if it is needed.
The end result is a timestamping event (SCM_TSTAMP_ACK) will
be missing from the sk->sk_error_queue.
This patch carries this
Like txstamp_ack bit, if needed, the eor_info bit should also be carried
to the new skb2 when splitting a skb
or
to the prev skb from the next_skb when collapsing skbs.
Signed-off-by: Martin KaFai Lau
Cc: Eric Dumazet
Cc: Neal Cardwell
Cc: Soheil Hassas Yeganeh
Cc: Willem de Bruijn
Cc: Yuchun
It reuses the tcp_sendmsg_noappend() to decide if a new_segment
is needed before entering the loop. More checks could be added
later for the tcp_sendpage case to decide if a new_segment is
needed immediately.
Signed-off-by: Martin KaFai Lau
Cc: Eric Dumazet
Cc: Neal Cardwell
Cc: Soheil Hassas
After receiving sacks, tcp_shifted_skb() will collapse
skbs if possible. tx_flags/tskey/txstamp_ack also has
to be merged in this case.
This patch resues the tcp_skb_collapse_tstamp() to handle
them.
BPF Output Before:
~
BPF Output After:
~
<...>-2024 [007] d.s.88.644374: : ee_dat
This patch allows the user process to use MSG_EOR during
tcp_sendmsg to tell the kernel that it is the last byte
of an application response message.
It is currently useful when the end-user has turned on any bit of the
SOF_TIMESTAMPING_TX_RECORD_MASK (either by setsockopt or cmsg).
The kernel will
When collapsing skbs during tcp_collapse_retrans and tcp_shift_skb_data,
this patch is to avoid collapsing to a prev skb that has eor_info mark
and the next_skb also has the tskey set (i.e. the prev skb will lose
a eor marked tskey info).
Signed-off-by: Martin KaFai Lau
Cc: Eric Dumazet
Cc: Neal
v2:
~ Rework based on the recent work
"add TX timestamping via cmsg" by
Soheil Hassas Yeganeh
~ This version takes the MSG_EOR bit as a signal of
end-of-response-message and leave the selective
timestamping job to the cmsg
~ Changes based on the v1 feedback (like avoid
unlikely check in
On 18/04/2016 at 15:17:58 -0700, Florian Fainelli wrote :
> Yes, seems like it, how about adding this:
>
> diff --git a/drivers/net/ethernet/cadence/macb.c
> b/drivers/net/ethernet/cadence/macb.c
> index 98b99149ce0b..21096dfb0e83 100644
> --- a/drivers/net/ethernet/cadence/macb.c
> +++ b/drivers/
Assuming SOF_TIMESTAMPING_TX_ACK is on. When dup acks are received,
it could incorrectly think that a skb has already
been acked and queue a SCM_TSTAMP_ACK cmsg to the
sk->sk_error_queue.
In tcp_ack_tstamp(), it checks
'between(shinfo->tskey, prior_snd_una, tcp_sk(sk)->snd_una - 1)'.
If prior_snd_
Having the tag protocol in dsa_switch_driver for setup time and in
dsa_switch_tree for runtime is enough. Remove dsa_switch's one.
Signed-off-by: Vivien Didelot
---
include/net/dsa.h | 5 -
net/dsa/dsa.c | 5 ++---
2 files changed, 2 insertions(+), 8 deletions(-)
diff --git a/include/ne
On 18/04/16 15:14, Alexandre Belloni wrote:
> On 15/04/2016 at 15:23:39 -0700, Florian Fainelli wrote :
>> On 15/04/16 15:17, Alexandre Belloni wrote:
>>> On 16/04/2016 at 00:05:08 +0200, Andrew Lunn wrote :
> Trace without my patch:
> libphy: MACB_mii_bus: probed
> macb f802.ethern
On Mon, 18 Apr 2016 14:43:07 -0700
Alexei Starovoitov wrote:
> I was worried about this too, but single 'if' and two calls
> (as in commit 98b5c2c65c295) is a better way, since it's faster, cleaner
> and doesn't need to refactor the whole perf_trace_buf_submit() to pass
> extra event_call argume
On 15/04/2016 at 15:23:39 -0700, Florian Fainelli wrote :
> On 15/04/16 15:17, Alexandre Belloni wrote:
> > On 16/04/2016 at 00:05:08 +0200, Andrew Lunn wrote :
> >>> Trace without my patch:
> >>> libphy: MACB_mii_bus: probed
> >>> macb f802.ethernet eth0: Cadence GEM rev 0x00020120 at 0xf80200
The references to some arrays in the rtl8xxxu driver were moved inside
of an #ifdef, but the symbols remain outside, resulting in build warnings:
rtl8xxxu/rtl8xxxu.c:1506:33: error: 'rtl8188ru_radioa_1t_highpa_table' defined
but not used
rtl8xxxu/rtl8xxxu.c:1431:33: error: 'rtl8192cu_radioa_1t_in
The w5100-spi driver front-end only makes sense when the w5100
core driver is enabled, not for a configuration that only has w5300:
drivers/net/built-in.o: In function `w5100_spi_remove':
drivers/net/ethernet/wiznet/w5100-spi.c:277: undefined reference to
`w5100_remove'
drivers/net/built-in.o: In
This is the IPv6 counterpart to commit 8282f27449bf ("inet: frag: Always
orphan skbs inside ip_defrag()").
Prior to commit 029f7f3b8701 ("netfilter: ipv6: nf_defrag: avoid/free
clone operations"), ipv6 fragments sent to nf_ct_frag6_gather() would be
cloned (implicitly orphaning) prior to queueing
On 18 April 2016 at 11:35, Pablo Neira Ayuso wrote:
> On Thu, Apr 14, 2016 at 05:35:39PM -0700, Joe Stringer wrote:
>> On 14 April 2016 at 03:35, Pablo Neira Ayuso wrote:
>> > On Thu, Apr 14, 2016 at 10:40:15AM +0200, Florian Westphal wrote:
>> >> David Laight wrote:
>> >> > From: Joe Stringer
>
Hi Dave,
On Mon, 18 Apr 2016 12:43:16 -0400 (EDT) David Miller
wrote:
>
> From: Herbert Xu
> Date: Mon, 18 Apr 2016 18:43:36 +0800
>
> > Right, the problem is that nothing within crypto ever selects
> > CRYPTO since it's also used as a way of hiding the crypto menu
> > options.
>
> As far a
On 4/18/16 1:29 PM, Steven Rostedt wrote:
On Mon, 4 Apr 2016 21:52:48 -0700
Alexei Starovoitov wrote:
introduce BPF_PROG_TYPE_TRACEPOINT program type and allow it to be
attached to tracepoints.
The tracepoint will copy the arguments in the per-cpu buffer and pass
it to the bpf program as its f
Francesco Ruggeri writes:
> On Mon, Apr 18, 2016 at 11:48 AM, Eric W. Biederman
> wrote:
>>
>> These interactions all seem a little bit funny. At a quick skim it
>> would make more sense to increment the port count in macvlan_init,
>> and completely remove the need to mess with port counts anyw
On Mon, 2016-04-18 at 14:10 -0700, Roopa Prabhu wrote:
> + if (filter_mask & IFLA_STATS_FILTER_BIT(IFLA_STATS_LINK_64)) {
> + struct rtnl_link_stats64 *sp;
> +
> + attr = nla_reserve(skb, IFLA_STATS_LINK_64,
> +sizeof(struct rtnl_link_sta
On 4/18/16 1:47 PM, Steven Rostedt wrote:
On Mon, 18 Apr 2016 12:51:43 -0700
Alexei Starovoitov wrote:
yeah, it could be added to ftrace as well, but it won't be as effective
as perf_trace, since the cost of trace_event_buffer_reserve() in
trace_event_raw_event_() handler is significantly hig
On Mon, Apr 18, 2016 at 1:56 PM, Eric Dumazet wrote:
> Linux TCP stack painfully segments all TSO/GSO packets before retransmits.
>
> This was fine back in the days when TSO/GSO were emerging, with their
> bugs, but we believe the dark age is over.
>
> Keeping big packets in write queues, but also
On 4/18/16, 9:28 AM, David Miller wrote:
> From: Roopa Prabhu
> Date: Sun, 17 Apr 2016 22:35:35 -0700
>
>> This patch adds a new RTM_GETSTATS message to query link stats via netlink
>> from the kernel. RTM_NEWLINK also dumps stats today, but RTM_NEWLINK
>> returns a lot more than just stats and is
From: Roopa Prabhu
This patch adds a new RTM_GETSTATS message to query link stats via netlink
from the kernel. RTM_NEWLINK also dumps stats today, but RTM_NEWLINK
returns a lot more than just stats and is expensive in some cases when
frequent polling for stats from userspace is a common operation
Linux TCP stack painfully segments all TSO/GSO packets before retransmits.
This was fine back in the days when TSO/GSO were emerging, with their
bugs, but we believe the dark age is over.
Keeping big packets in write queues, but also in stack traversal
has a lot of benefits.
- Less memory overhe
On Mon, 18 Apr 2016 12:51:43 -0700
Alexei Starovoitov wrote:
> yeah, it could be added to ftrace as well, but it won't be as effective
> as perf_trace, since the cost of trace_event_buffer_reserve() in
> trace_event_raw_event_() handler is significantly higher than
> perf_trace_buf_alloc() in p
On Mon, 4 Apr 2016 21:52:48 -0700
Alexei Starovoitov wrote:
> introduce BPF_PROG_TYPE_TRACEPOINT program type and allow it to be
> attached to tracepoints.
> The tracepoint will copy the arguments in the per-cpu buffer and pass
> it to the bpf program as its first argument.
> The layout of the fi
Hi, how are you? My name is J Eric Denials, External Financial Auditor at
Lloyds Banking Group plc., London. It is a pleasure to contact you at this time
through this medium. I have a cool and legitimate deal to do with you as you're
a foreigner, it will be mutually beneficial to both. If you’re
On Mon, Apr 18, 2016 at 10:04:43PM +0200, Arnd Bergmann wrote:
> On Monday 18 April 2016 20:43:36 Pablo Neira Ayuso wrote:
> > On Mon, Apr 18, 2016 at 08:33:15PM +0200, Arnd Bergmann wrote:
> > > On Monday 18 April 2016 20:16:59 Pablo Neira Ayuso wrote:
> > > > On Sat, Apr 16, 2016 at 10:17:43PM +0
On Mon, Apr 18, 2016 at 11:48 AM, Eric W. Biederman
wrote:
>
> These interactions all seem a little bit funny. At a quick skim it
> would make more sense to increment the port count in macvlan_init,
> and completely remove the need to mess with port counts anywhere except
> macvlan_init and macvl
The dsa_slave_priv structure does not need a pointer to its net_device.
Kill it.
Signed-off-by: Vivien Didelot
---
net/dsa/dsa_priv.h | 5 -
net/dsa/slave.c| 9 -
2 files changed, 4 insertions(+), 10 deletions(-)
diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h
index 1d1a546
On Monday 18 April 2016 20:43:36 Pablo Neira Ayuso wrote:
> On Mon, Apr 18, 2016 at 08:33:15PM +0200, Arnd Bergmann wrote:
> > On Monday 18 April 2016 20:16:59 Pablo Neira Ayuso wrote:
> > > On Sat, Apr 16, 2016 at 10:17:43PM +0200, Arnd Bergmann wrote:
> > > > A recent patch removed many 'inline'
On 4/18/16 9:13 AM, Steven Rostedt wrote:
On Mon, 4 Apr 2016 21:52:46 -0700
Alexei Starovoitov wrote:
Hi Steven, Peter,
last time we discussed bpf+tracepoints it was a year ago [1] and the reason
we didn't proceed with that approach was that bpf would make arguments
arg1, arg2 to trace_xx(arg
> -Original Message-
> From: Joe Perches [mailto:j...@perches.com]
> Sent: Monday, April 18, 2016 10:00 AM
> To: KY Srinivasan ; Alexander Duyck
>
> Cc: David Miller ; Netdev
> ; linux-ker...@vger.kernel.org;
> de...@linuxdriverproject.org; o...@aepfle.de; Robo Bot
> ; Jason Wang ;
> e..
Currently all drivers depend and autoload the vxlan module because how
vxlan_get_rx_port is linked into them. Remove this dependency:
By using a new event type in the netdevice notifier call chain we proxy
the request from the drivers to flush and resetup the vxlan ports not
directly via function
Equivalent to "vxlan: break dependency with netdev drivers", don't
autoload geneve module in case the driver is loaded. Instead make the
coupling weaker by using netdevice notifiers as proxy.
Cc: Jesse Gross
Signed-off-by: Hannes Frederic Sowa
---
drivers/net/geneve.c | 31
vxlan_get_rx_port requires rtnl_lock to be held.
Cc: Jeff Kirsher
Cc: Jesse Brandeburg
Cc: Shannon Nelson
Cc: Carolyn Wyborny
Cc: Don Skidmore
Cc: Bruce Allan
Cc: John Ronciak
Cc: Mitch Williams
Signed-off-by: Hannes Frederic Sowa
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 2 ++
qlcnic_attach_func requires rtnl_lock to be held.
Cc: dept-gelinuxnic...@qlogic.com
Signed-off-by: Hannes Frederic Sowa
---
drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c | 10 --
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ma
mlx4_en_start_port requires rtnl_lock to be held.
Cc: Eugenia Emantayev
Cc: Yishai Hadas
Signed-off-by: Hannes Frederic Sowa
---
drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
b/drivers/net/et
be_open calls down to functions which expects rtnl lock to be held.
Cc: Sathya Perla
Cc: Ajit Khaparde
Cc: Padmanabh Ratnakar
Cc: Sriharsha Basavapatna
Cc: Somnath Kotur
Signed-off-by: Hannes Frederic Sowa
---
drivers/net/ethernet/emulex/benet/be_main.c | 10 ++
1 file changed, 6 in
fm10k_open requires rtnl_lock to be held.
Cc: Jeff Kirsher
Cc: Jesse Brandeburg
Cc: Shannon Nelson
Cc: Carolyn Wyborny
Cc: Don Skidmore
Cc: Bruce Allan
Cc: John Ronciak
Cc: Mitch Williams
Signed-off-by: Hannes Frederic Sowa
---
drivers/net/ethernet/intel/fm10k/fm10k_pci.c | 2 ++
1 file
This patchset removes the dependency of network drivers on vxlan or
geneve, so those don't get autoloaded when the nic driver is loaded.
Also audited the code such that vxlan_get_rx_port and geneve_get_rx_port
are not called without rtnl lock.
Hannes Frederic Sowa (7):
benet: be_resume needs to
Add a BPF_F_CURRENT_CPU flag to optimize the use-case where user space has
per-CPU ring buffers and the eBPF program pushes the data into the current
CPU's ring buffer which saves us an extra helper function call in eBPF.
Also, make sure to properly reserve the remaining flags which are not used.
This minor set adds a new helper bpf_event_output() for eBPF cls/act
program types which allows to pass events to user space applications.
For details, please see individual patches.
Thanks!
v1 -> v2:
- Address kbuild bot found compile issue in patch 2
- Rest as is
Daniel Borkmann (2):
bpf
This patch adds a new helper for cls/act programs that can push events
to user space applications. For networking, this can be f.e. for sampling,
debugging, logging purposes or pushing of arbitrary wake-up events. The
idea is similar to a43eec304259 ("bpf: introduce bpf_perf_event_output()
helper")
Francesco Ruggeri writes:
> Resending, did not include netdev the first time ...
>
> If a macvlan/macvtap creation fails in register_netdevice in
> call_netdevice_notifiers(NETDEV_REGISTER) then while cleaning things up in
> rollback_registered_many it invokes macvlan_uninit. This results in
> p
From: Govindarajulu Varadarajan
Date: Sat, 16 Apr 2016 00:40:43 +0530
> From: Govindarajulu Varadarajan <_gov...@gmx.com>
>
> Driver sets vlan_feature to netdev->features as hardware supports all of
> them on vlan interface.
>
> Signed-off-by: Govindarajulu Varadarajan <_gov...@gmx.com>
Applie
From: Philippe Reynes
Date: Fri, 15 Apr 2016 00:34:58 +0200
> Ethtool has a new api {get|set}_link_ksettings that deprecate
> the old api {get|set}_settings. We update the fec driver to use
> this new ethtool api.
>
> For this first version, I've converted old u32 value in phy structure
> to lin
On Mon, Apr 18, 2016 at 08:33:15PM +0200, Arnd Bergmann wrote:
> On Monday 18 April 2016 20:16:59 Pablo Neira Ayuso wrote:
> > On Sat, Apr 16, 2016 at 10:17:43PM +0200, Arnd Bergmann wrote:
> > > A recent patch removed many 'inline' annotations for static
> > > functions in this file, which has cau
From: Paolo Abeni
Date: Thu, 14 Apr 2016 18:39:39 +0200
> The current tun_net_xmit() implementation don't need any external
> lock since it relies on rcu protection for the tun data structure
> and on socket queue lock for skb queuing.
>
> This patch set the NETIF_F_LLTX feature bit in the tun d
1 - 100 of 211 matches
Mail list logo