Extend fl_flow_mask structure with reference counter to allow parallel
modification without relying on rtnl lock. Use rcu read lock to safely
lookup mask and increment reference counter in order to accommodate
concurrent deletes.
Signed-off-by: Vlad Buslov
Acked-by: Jiri Pirko
---
net/sched/cls
Extend flower filters with reference counting in order to remove dependency
on rtnl lock in flower ops and allow to modify filters concurrently.
Reference to flower filter can be taken/released concurrently as soon as it
is marked as 'unlocked' by last patch in this series. Use atomic reference
cou
Flower classifier only changes root pointer during init and destroy. Cls
API implements reference counting for tcf_proto, so there is no danger of
concurrent access to tp when it is being destroyed, even without protection
provided by rtnl lock.
Implement new function fl_head_dereference() to dere
Currently, all netlink protocol handlers for updating rules, actions and
qdiscs are protected with single global rtnl lock which removes any
possibility for parallelism. This patch set is a third step to remove
rtnl lock dependency from TC rules update path.
Recently, new rtnl registration flag RT
In order to prevent double deletion of filter by concurrent tasks when rtnl
lock is not used for synchronization, add 'deleted' filter field. Check
value of this field when modifying filters and return error if concurrent
deletion is detected.
Refactor __fl_delete() to accept pointer to 'last' boo
Use 'rtnl_held' flag to track if caller holds rtnl lock. Propagate the flag
to internal functions that need to know rtnl lock state. Take rtnl lock
before calling tcf APIs that require it (hw offload, bind filter, etc.).
Signed-off-by: Vlad Buslov
Acked-by: Jiri Pirko
---
net/sched/cls_flower.c
Check if user specified a handle and another filter with the same handle
was inserted concurrently. Return EAGAIN to retry filter processing (in
case it is an overwrite request).
Signed-off-by: Vlad Buslov
Acked-by: Jiri Pirko
---
net/sched/cls_flower.c | 9 +
1 file changed, 9 insertio
Protect modifications of flower masks list with spinlock to remove
dependency on rtnl lock and allow concurrent access.
Signed-off-by: Vlad Buslov
Acked-by: Jiri Pirko
---
net/sched/cls_flower.c | 8
1 file changed, 8 insertions(+)
diff --git a/net/sched/cls_flower.c b/net/sched/cls_f
As a preparation for using classifier spinlock instead of relying on
external rtnl lock, rearrange code in fl_change. The goal is to group the
code which changes classifier state in single block in order to allow
following commits in this set to protect it from parallel modification with
tp->lock.
Without rtnl lock protection tcf proto can be deleted concurrently. Check
tcf proto 'deleting' flag after taking tcf spinlock to verify that no
concurrent deletion is in progress. Return EAGAIN error if concurrent
deletion detected, which will cause caller to retry and possibly create new
instance
Set TCF_PROTO_OPS_DOIT_UNLOCKED for flower classifier to indicate that its
ops callbacks don't require caller to hold rtnl lock.
Signed-off-by: Vlad Buslov
Acked-by: Jiri Pirko
---
net/sched/cls_flower.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/net/sched/cls_flower.c b/net/sched/cls_
Without rtnl lock protection masks with same key can be inserted
concurrently. Insert temporary mask with reference count zero to masks
hashtable. This will cause any concurrent modifications to retry.
Wait for rcu grace period to complete after removing temporary mask from
masks hashtable to acco
struct tcf_proto was extended with spinlock to be used by classifiers
instead of global rtnl lock. Use it to protect shared flower classifier
data structures (handle_idr, mask hashtable and list) and fields of
individual filters that can be accessed concurrently. This patch set uses
tcf_proto->lock
Dear all,
https://www.mail-archive.com/netdev@vger.kernel.org/msg256527.html
as we concerned before at above mail thread,
we faced a problem cased by not removed socket.
(from now, 'the socket' means the socket alloced at 0xFFC0051E5E00)
#1. the socket is state in TIME_WAIT1. maybe it's pr
在 2019/2/14 12:08, Florian Fainelli 写道:
>
>
> On 2/13/2019 8:31 PM, Jian Shen wrote:
>> The default led configuration of marvell 88E1510 is not fit
>> for hns3 driver, this patch fixes it.
>>
>> Signed-off-by: Jian Shen
>> ---
>> .../net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c | 19
>>
在 2019/2/14 12:06, Florian Fainelli 写道:
>
>
> On 2/13/2019 8:31 PM, Jian Shen wrote:
>> The default m88e1510 LED configuration is 0x1177, used LED[0]
>> for 1000M link, LED[1] for 100M link, and LED[2] for active.
>> But for our boards, we want to use 0x1040, which use LED[0] for
>> link, and
Hi,
i used iperf2 on my odroid c2 , since this can do duplex tests ! iperf2 -c
10.10.11.1
-i1 -d
Do you really need the exact numbers ?
NON-duplex:
4.14.29 reached 930 MBits without the -d ( duplex option ), 5.0rc5 only
reaches 600MBits there
DUPLEX:
With the -d option things get worse on 4.1
Each tls context maintains two cipher contexts (one each for tx and rx
directions). For each tls session, the constants such as protocol
version, ciphersuite, iv size, associated data size etc are same for
both the directions and need to be stored only once per tls context.
Hence these are moved fr
> -Original Message-
> From: Guenter Roeck On Behalf Of Guenter Roeck
> Sent: Wednesday, February 13, 2019 5:03 PM
> To: Andrew Lunn ; Ido Schimmel
> Cc: netdev@vger.kernel.org; da...@davemloft.net; Jiri Pirko
> ; mlxsw ; Vadim Pasternak
>
> Subject: Re: [PATCH net-next 09/12] mlxsw: c
On Tue, 2019-01-29 at 11:21 +0800, YueHaibing wrote:
> Use struct_size() in kzalloc instead of the 'regd_to_copy'
>
> Signed-off-by: YueHaibing
> ---
Applied to our internal tree and it will reach the mainline following
our normal upstreaming process.
Thanks!
--
Luca.
It's confusing to call PTR_ERR(v). The PTR_ERR() function is basically
a fancy cast to long so it makes you wonder, was IS_ERR() intended? But
that doesn't make sense because vcc_walk() doesn't return error
pointers.
This patch doesn't affect runtime, it's just a cleanup.
Signed-off-by: Dan Car
On Tue, 2019-01-29 at 11:21 +0800, YueHaibing wrote:
> Use kmemdup rather than duplicating its implementation
>
> Signed-off-by: YueHaibing
> ---
> drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c | 8 ++--
> 1 file changed, 2 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/net/wir
For common stmmac:
- Add support to set CSR Clock range selection in DT
For stm32mpu:
- Glue codes to support magic packet
- Glue codes to support all PHY config :
PHY_MODE (MII,GMII, RMII, RGMII) and in normal,
PHY wo crystal (25Mhz),
In Documentation stmmac.txt there is possibility to
fixed CSR Clock range selection with property clk_csr.
This patch add the management of this property
For example to use it, add in your ethernet node DT:
clk_csr = <3>;
Signed-off-by: Christophe Roullier
---
drivers/net/ethernet/stmicr
Update glue codes to support all PHY config on stm32mp157c
PHY_MODE (MII,GMII, RMII, RGMII) and in normal, PHY wo crystal (25Mhz),
PHY wo crystal (50Mhz), No 125Mhz from PHY config.
Signed-off-by: Christophe Roullier
---
drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c | 102
Update glue codes to be compatible with MCU family.
Signed-off-by: Christophe Roullier
---
drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c | 51 ++-
1 file changed, 41 insertions(+), 10 deletions(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
b/drivers/
Synopsys GMAC 4.10 is used. And Phy mode for eval and disco is RMII
with PHY SMSC LAN8742
Signed-off-by: Christophe Roullier
---
arch/arm/boot/dts/stm32h743-pinctrl.dtsi | 15 +++
arch/arm/boot/dts/stm32h743.dtsi | 19 +++
arch/arm/boot/dts/stm32h743i-disco.dt
This patch add syscfg clock support for Ethernet of the STM32F429 SoC.
Needed if bootloader do not manage it.
Signed-off-by: Christophe Roullier
---
arch/arm/boot/dts/stm32f429.dtsi | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/arch/arm/boot/dts/stm32f429.dtsi b/arch/
Need syscfg clock for MCU family in case bootloader does not
activate it.
Signed-off-by: Christophe Roullier
---
Documentation/devicetree/bindings/net/stm32-dwmac.txt | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/Documentation/devicetree/bindings/net/stm32-dwmac.txt
b/
Add glue codes to support magic packet on stm32mp157c
Signed-off-by: Christophe Roullier
---
drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c | 30 ++-
1 file changed, 29 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c
b/drivers/
Add properties to support all Phy config
PHY_MODE (MII,GMII, RMII, RGMII) and in normal, PHY wo crystal (25Mhz),
PHY wo crystal (50Mhz), No 125Mhz from PHY config.
Signed-off-by: Christophe Roullier
---
Documentation/devicetree/bindings/net/stm32-dwmac.txt | 6 +++---
1 file changed, 3 i
The skb should be freed by dev_consume_skb_any() efx_tx_tso_fallback()
when skb is still used. The skb is be replaced by segments, so the
original skb should be consumed(not drop).
Signed-off-by: Huang Zijiang
---
drivers/net/ethernet/sfc/tx.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-
The of_find_device_by_node() takes a reference to the underlying device
structure, we should release that reference.
Signed-off-by: Huang Zijiang
---
drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ds
dev_consume_skb_irq() should be called when skb xmit
done.It makes drop profiles more friendly.
Signed-off-by: Huang Zijiang
---
drivers/net/ethernet/dlink/dl2k.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/dlink/dl2k.c
b/drivers/net/ethernet/dlink/d
The skb should be freed by dev_consume_skb_any() in macb_pad_and_fcs()
when *skb is still used. The *skb is be replaced by nskb, so the
original *skb should be consumed(not drop).
Signed-off-by: Huang Zijiang
---
drivers/net/ethernet/cadence/macb_main.c | 2 +-
1 file changed, 1 insertion(+), 1
modify the code style in order to removing the following warning
when excute the script checkpatch.pl
WARNING: space prohibited between function name and open parenthesis '('
Signed-off-by: Huang Zijiang
---
drivers/net/ethernet/dlink/dl2k.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions
The skb should be freed by dev_consume_skb_any() in hfcpci_fill_fifo()
when bcs->tx_skb is still used. The bcs->tx_skb is be replaced by
skb_dequeue(&bcs->squeue), so the original bcs->tx_skb should
be consumed(not drop).
Signed-off-by: Huang Zijiang
---
drivers/isdn/hisax/hfc_pci.c | 2 +-
1 fi
On Wed, Feb 13, 2019 at 09:57:25PM -0800, Stanislav Fomichev wrote:
>
> > That 'stuck with __sk_buff' is what bothers me.
> I might have use the wrong word here. I don't think there is another
> option to be honest. Using __sk_buff makes flow dissector programs work
> with fragmented packets;
goo
From: Ido Schimmel
Date: Wed, 13 Feb 2019 11:28:43 +
> Vadim says:
>
> This patchset contains various improvements to hwmon and thermal code in
> mlxsw. The most significant improvement is the ability to read modules'
> temperature attributes (input, fault, critical and emergency thresholds)
tree: https://git.kernel.org/pub/scm/linux/kernel/git/davem/net.git master
head: af98c5a78517c04adb5fd68bb64b1ad6fe3d473f
commit: fc228abc2347e106a44c0e9b29ab70b712c4ca51 [27/28] sctp: call
gso_reset_checksum when computing checksum in sctp_gso_segment
reproduce:
# apt-get install spar
From: Simon Wunderlich
Date: Wed, 13 Feb 2019 10:55:03 +0100
> here is another pull request for batman-adv to go into net-next. The main
> patchset is Svens netlink restructure series, which was also discussed on
> netdev
> for quite some time now and should be ready to be integrated.
>
> Pleas
Hi David,
On 14.02.2019 03:45, David Chang wrote:
> Hi Heiner,
>
> On Feb 05, 2019 at 19:50:30 +0100, Heiner Kallweit wrote:
>> Hi David,
>>
>> meanwhile there's the following bug report matching what reported.
>> It's even the same chip version (RTL8168h).
>> https://bugzilla.redhat.com/show_bug
From: Dan Carpenter
Date: Wed, 13 Feb 2019 11:59:31 +0300
> We need to set the error message on this path otherwise some of the
> callers, such as test_hints_case(), print from an uninitialized pointer.
>
> We had a similar bug earlier and set "errmsg" to NULL in the caller,
> test_delta_action_
From: Dan Carpenter
Date: Wed, 13 Feb 2019 11:56:50 +0300
> We need to set the error code on this path otherwise we return
> ERR_PTR(0) which would result in a NULL dereference in the caller.
>
> Fixes: 9069a3817d82 ("lib: objagg: implement optimization hints assembly and
> use hints for object
From: Dan Carpenter
Date: Wed, 13 Feb 2019 11:23:04 +0300
> The value of ->num_ports comes from bcm_sf2_sw_probe() and it is less
> than or equal to DSA_MAX_PORTS. The ds->ports[] array is used inside
> the dsa_is_user_port() and dsa_is_cpu_port() functions. The ds->ports[]
> array is allocated
From: Dan Carpenter
Date: Wed, 13 Feb 2019 11:58:20 +0300
> There is a typo here. We intended to check "objagg2" but we instead
> test "objagg" which is not an error pointer.
>
> Fixes: 9069a3817d82 ("lib: objagg: implement optimization hints assembly and
> use hints for object creation")
> Si
From: Vishal Kulkarni
Date: Wed, 13 Feb 2019 10:44:03 +0530
> T6 introduced a Timer Mechanism in SGE called the
> SGE Doorbell Queue Timer. With this we can now configure
> TX Queues to get CIDX Updates when:
>
> Time(CIDX == PIDX) >= Timer
>
> Previously we rely on TX Queue Status Page upd
On Wed, Feb 13, 2019 at 8:21 PM Alexei Starovoitov
wrote:
>
> On Wed, Feb 13, 2019 at 08:44:51PM -0700, David Ahern wrote:
> > On 2/13/19 7:39 PM, Alexei Starovoitov wrote:
> > > On Wed, Feb 13, 2019 at 05:46:26PM -0700, David Ahern wrote:
> > >> On 2/13/19 12:53 PM, Peter Oskolkov wrote:
> > >>>
From: Vishal Kulkarni
Date: Wed, 13 Feb 2019 10:48:52 +0530
> CR4_QSFP 10G Speed technology should be 1baseKR_Full
> And also report available FEC modes.
>
> Signed-off-by: Vishal Kulkarni
Applied, thanks.
On error the skb should be freed. Tested with diff/steps
provided by David Ahern.
Reported-by: David Ahern
Fixes: 3bd0b15281af ("bpf: add handling of BPF_LWT_REROUTE to lwt_bpf.c")
Signed-off-by: Peter Oskolkov
---
net/core/lwt_bpf.c | 24
1 file changed, 16 insertions(
From: YueHaibing
Date: Wed, 13 Feb 2019 01:42:00 +
> Remove duplicated include.
>
> Signed-off-by: YueHaibing
Applied, thanks.
From: Jesper Dangaard Brouer
Date: Wed, 13 Feb 2019 02:55:34 +0100
> As pointed out by David Miller in [1] the current page_pool implementation
> stores dma_addr_t in page->private. This won't work on 32-bit platforms with
> 64-bit DMA addresses since the page->private is an unsigned long and the
On 02/13, Peter Oskolkov wrote:
> This patch adds a bpf self-test to cover BPF_LWT_ENCAP_IP mode
> in bpf_lwt_push_encap.
>
> Covered:
> - encapping in LWT_IN and LWT_XMIT
> - IPv4 and IPv6
>
> A follow-up patch will add GSO and VRF-enabled tests.
>
> Signed-off-by: Peter Oskolkov
> ---
> tool
On 02/13, Alexei Starovoitov wrote:
> On Tue, Feb 12, 2019 at 09:02:32AM -0800, Stanislav Fomichev wrote:
> > On 02/05, Stanislav Fomichev wrote:
> > > On 02/05, Alexei Starovoitov wrote:
> > > > On Tue, Feb 05, 2019 at 07:56:19PM -0800, Stanislav Fomichev wrote:
> > > > > On 02/05, Alexei Starovoi
Hi Herbert,
url:
https://github.com/0day-ci/linux/commits/Herbert-Xu/mac80211-Fix-incorrect-usage-of-rhashtable-walk-API/20190213-181325
base: https://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211-next.git
master
smatch warnings:
net/mac80211/mesh_pathtbl.c:439 mesh_path_add
On Wed, Feb 13, 2019 at 8:21 PM Alexei Starovoitov
wrote:
>
> On Wed, Feb 13, 2019 at 08:44:51PM -0700, David Ahern wrote:
> > On 2/13/19 7:39 PM, Alexei Starovoitov wrote:
> > > On Wed, Feb 13, 2019 at 05:46:26PM -0700, David Ahern wrote:
> > >> On 2/13/19 12:53 PM, Peter Oskolkov wrote:
> > >>>
From: John Hurley
Date: Wed, 13 Feb 2019 00:23:52 +
> With the introduction of flow_stats_update(), drivers now update the stats
> fields of the passed tc_cls_flower_offload struct, rather than call
> tcf_exts_stats_update() directly to update the stats of offloaded TC
> flower rules. However
From: Vlad Buslov
Date: Tue, 12 Feb 2019 23:39:06 +0200
> Recently introduced tc_setup_flow_action() can fail when parsing tcf_exts
> on some unsupported action commands. However, this should not affect the
> case when user did not explicitly request hw offload by setting skip_sw
> flag. Modify t
From: Eric Dumazet
Date: Tue, 12 Feb 2019 12:26:27 -0800
> With many active TCP sockets, fat TCP sockets could fool
> __sk_mem_raise_allocated() thanks to an overflow.
>
> They would increase their share of the memory, instead
> of decreasing it.
>
> Signed-off-by: Eric Dumazet
Applied, thank
From: Colin King
Date: Tue, 12 Feb 2019 16:08:07 +
> From: Colin Ian King
>
> There are some statements that are indented incorrectly. Fix these.
>
> Signed-off-by: Colin Ian King
Applied to net-next.
From: Colin King
Date: Tue, 12 Feb 2019 16:01:53 +
> From: Colin Ian King
>
> There are some statements in an if-block that are not correctly
> indented. Fix these.
>
> Signed-off-by: Colin Ian King
Applied to net-next.
From: Yang Wei
Date: Wed, 13 Feb 2019 00:00:02 +0800
> From: Yang Wei
>
> dev_consume_skb_irq() should be called in at91ether_interrupt() when
> skb xmit done. It makes drop profiles(dropwatch, perf) more friendly.
>
> Signed-off-by: Yang Wei
Applied, thanks.
From: Yang Wei
Date: Tue, 12 Feb 2019 23:59:04 +0800
> From: Yang Wei
>
> dev_consume_skb_irq() should be called when skb xmit done. It makes
> drop profiles(dropwatch, perf) more friendly.
>
> Signed-off-by: Yang Wei
Applied to net-next.
From: Yang Wei
Date: Wed, 13 Feb 2019 00:01:11 +0800
> From: Yang Wei
>
> dev_consume_skb_irq() should be called in eth_txdone_irq() when skb
> xmit done. It makes drop profiles(dropwatch, perf) more friendly.
>
> Signed-off-by: Yang Wei
Applied.
From: Yang Wei
Date: Tue, 12 Feb 2019 23:49:57 +0800
> From: Yang Wei
>
> dev_consume_skb_irq() should be called in emac_mac_tx_process() when
> skb xmit done. It makes drop profiles(dropwatch, perf) more friendly.
>
> Signed-off-by: Yang Wei
Applied to net-next.
From: Yang Wei
Date: Tue, 12 Feb 2019 23:56:00 +0800
> From: Yang Wei
>
> dev_consume_skb_irq() should be called in moxart_tx_finished() when
> skb xmit done. It makes drop profiles(dropwatch, perf) more friendly.
>
> Signed-off-by: Yang Wei
Applied, thanks.
From: Yang Wei
Date: Tue, 12 Feb 2019 23:52:53 +0800
> From: Yang Wei
>
> dev_consume_skb_irq() should be called in mace_interrupt() when skb
> xmit done. It makes drop profiles(dropwatch, perf) more friendly.
>
> Signed-off-by: Yang Wei
Applied to net-next.
From: Yang Wei
Date: Tue, 12 Feb 2019 23:47:31 +0800
> From: Yang Wei
>
> dev_consume_skb_irq() should be called when skb xmit done. It makes
> drop profiles(dropwatch, perf) more friendly.
>
> Signed-off-by: Yang Wei
Applied to net-next.
From: Yang Wei
Date: Tue, 12 Feb 2019 23:51:45 +0800
> From: Yang Wei
>
> dev_consume_skb_irq() should be called when skb xmit done. It makes
> drop profiles(dropwatch, perf) more friendly.
>
> Signed-off-by: Yang Wei
Applied.
> Ok, all done.
Thanks
> Should I queue just this one for -stable? I didn't queue up Heiner's change
> for
> -stable because it fixes a 5.0-rcX regression.
Yes please.
Andrew
On February 13, 2019 8:17:16 AM PST, Ivan Khoronzhuk
wrote:
>On Tue, Jan 22, 2019 at 03:12:41PM +0200, Ivan Khoronzhuk wrote:
>>On Mon, Jan 21, 2019 at 03:37:41PM -0800, Florian Fainelli wrote:
>>>On 12/4/18 3:42 PM, Ivan Khoronzhuk wrote:
On Tue, Dec 04, 2018 at 11:49:27AM -0800, Florian
From: Andrew Lunn
Date: Thu, 14 Feb 2019 03:07:23 +0100
> On Mon, Feb 11, 2019 at 01:40:21PM -0500, John David Anglin wrote:
>> The GPIO interrupt controller on the espressobin board only supports edge
>> interrupts.
>> If one enables the use of hardware interrupts in the device tree for the
>>
From: Heiner Kallweit
Date: Tue, 12 Feb 2019 19:56:15 +0100
> phylib enables interrupts before phy_start() has been called, and if
> we receive an interrupt in a non-started state, the interrupt handler
> returns IRQ_NONE. This causes problems with at least one Marvell chip
> as reported by Andre
On 2/13/19 8:32 PM, Andrew Lunn wrote:
>>> +For information describing the SFP cage in DT, please see the binding
>>> +documentation in the kernel source tree
>>> +``Documentation/devicetree/bindings/net/sff,sfp.txt``
>> oh, so SFP means "Small Form-factor Pluggable".
>>
>> I see that this source f
On Tue, Feb 12, 2019 at 09:02:32AM -0800, Stanislav Fomichev wrote:
> On 02/05, Stanislav Fomichev wrote:
> > On 02/05, Alexei Starovoitov wrote:
> > > On Tue, Feb 05, 2019 at 07:56:19PM -0800, Stanislav Fomichev wrote:
> > > > On 02/05, Alexei Starovoitov wrote:
> > > > > On Tue, Feb 05, 2019 at 0
> > +For information describing the SFP cage in DT, please see the binding
> > +documentation in the kernel source tree
> > +``Documentation/devicetree/bindings/net/sff,sfp.txt``
> oh, so SFP means "Small Form-factor Pluggable".
>
> I see that this source file:
> ./drivers/net/ethernet/intel/i40e/
On Wed, Feb 13, 2019 at 08:44:51PM -0700, David Ahern wrote:
> On 2/13/19 7:39 PM, Alexei Starovoitov wrote:
> > On Wed, Feb 13, 2019 at 05:46:26PM -0700, David Ahern wrote:
> >> On 2/13/19 12:53 PM, Peter Oskolkov wrote:
> >>> This patchset implements BPF_LWT_ENCAP_IP mode in bpf_lwt_push_encap
>
On 2/13/2019 11:12 AM, Heiner Kallweit wrote:
> Russell reported the following race in the phylib state machine
> (quoting from his mail):
>
> if (phy_polling_mode(phydev) && phy_is_started(phydev))
> phy_queue_state_machine(phydev, PHY_STATE_TIME);
>
> state = PHY_UP
> thread 0
On 2/13/2019 11:11 AM, Heiner Kallweit wrote:
> Russell suggested to remove the locking from phy_is_started() because
> the read is atomic anyway and actually the locking may be more
> misleading.
>
> Fixes: 2b3e88ea6528 ("net: phy: improve phy state checking")
> Suggested-by: Russell King - AR
On 2/13/2019 8:31 PM, Jian Shen wrote:
> The default led configuration of marvell 88E1510 is not fit
> for hns3 driver, this patch fixes it.
>
> Signed-off-by: Jian Shen
> ---
> .../net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c | 19
> +++
> 1 file changed, 19 insertions(
On 2/12/2019 10:56 AM, Heiner Kallweit wrote:
> phylib enables interrupts before phy_start() has been called, and if
> we receive an interrupt in a non-started state, the interrupt handler
> returns IRQ_NONE. This causes problems with at least one Marvell chip
> as reported by Andrew.
> Fix this
On 2/13/2019 8:31 PM, Jian Shen wrote:
> The default m88e1510 LED configuration is 0x1177, used LED[0]
> for 1000M link, LED[1] for 100M link, and LED[2] for active.
> But for our boards, we want to use 0x1040, which use LED[0] for
> link, and LED[1] for active.
>
> This patch adds a new m88e15
On 2/5/19 7:58 AM, Russell King wrote:
> Add some phylink documentation to the networking book detailing how
> to convert network drivers from phylib to phylink.
>
> Signed-off-by: Russell King
> ---
> Version 2 adds the "Modes of operation" section, as it appears mvpp2 is
> non-conformant (which
The default led configuration of marvell 88E1510 is not fit
for hns3 driver, this patch fixes it.
Signed-off-by: Jian Shen
---
.../net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c | 19 +++
1 file changed, 19 insertions(+)
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf
Currently, the m88e1510 phy driver used LED[0] and LED[1] for link,
LED[2] for active. It's incompatible with some boards which uses
LED[1] for active. This patchset adds new led configuration for
HNS3 driver.
Jian Shen (2):
net: phy: marvell: add new m88e1510 LED configuration
net: hns3: add
The default m88e1510 LED configuration is 0x1177, used LED[0]
for 1000M link, LED[1] for 100M link, and LED[2] for active.
But for our boards, we want to use 0x1040, which use LED[0] for
link, and LED[1] for active.
This patch adds a new m88e1510 LED configuration for it.
Signed-off-by: Jian Shen
On 2/13/19 7:39 PM, Alexei Starovoitov wrote:
> On Wed, Feb 13, 2019 at 05:46:26PM -0700, David Ahern wrote:
>> On 2/13/19 12:53 PM, Peter Oskolkov wrote:
>>> This patchset implements BPF_LWT_ENCAP_IP mode in bpf_lwt_push_encap
>>> BPF helper. It enables BPF programs (specifically, BPF_PROG_TYPE_LW
Hi Heiner,
On Feb 05, 2019 at 19:50:30 +0100, Heiner Kallweit wrote:
> Hi David,
>
> meanwhile there's the following bug report matching what reported.
> It's even the same chip version (RTL8168h).
> https://bugzilla.redhat.com/show_bug.cgi?id=1671958
>
> Symptom there is also a significant numb
Created a way to clear the multicast forwarding cache on a socket
without having to either remove the entries manually using the delete
entry socket option or destroy and recreate the multicast socket.
Calling the socket option MRT_FLUSH will allow any combination of the
four flag options to be cl
Currently the only way to clear the forwarding cache was to delete the
entries one by one using the MRT_DEL_MFC socket option or to destroy and
recreate the socket.
Create a new socket option which with the use of optional flags can
clear any combination of multicast entries (static or not static)
On Wed, Feb 13, 2019 at 05:46:26PM -0700, David Ahern wrote:
> On 2/13/19 12:53 PM, Peter Oskolkov wrote:
> > This patchset implements BPF_LWT_ENCAP_IP mode in bpf_lwt_push_encap
> > BPF helper. It enables BPF programs (specifically, BPF_PROG_TYPE_LWT_IN
> > and BPF_PROG_TYPE_LWT_XMIT prog types) t
From: Tonghao Zhang
This patch is a little improvement. If user use the
command shown as below, we should print the info [1]
instead of [2]. The eth0 exists actually, and it may
confuse user.
$ echo "eth0" > /sys/class/net/bond4/bonding/slaves
[1] "bond4: no command found in slaves file - use +
On Mon, Feb 11, 2019 at 01:40:21PM -0500, John David Anglin wrote:
> The GPIO interrupt controller on the espressobin board only supports edge
> interrupts.
> If one enables the use of hardware interrupts in the device tree for the
> 88E6341, it is
> possible to miss an edge. When this happens,
On Tue, Feb 12, 2019 at 07:56:15PM +0100, Heiner Kallweit wrote:
> phylib enables interrupts before phy_start() has been called, and if
> we receive an interrupt in a non-started state, the interrupt handler
> returns IRQ_NONE. This causes problems with at least one Marvell chip
> as reported by An
[ Upstream commit 356d1833b638bd465672aefeb71def3ab93fc17d ]
Note that when a new netns is created, it inherits its
sysctl_tcp_rmem and sysctl_tcp_wmem from initial netns.
This change is needed so that we can refine TCP rcvbuf autotuning,
to take RTT into consideration.
Signed-off-by: Eric Dumaz
Eric reported that, with 10 million sockets, ss -emoi (about 1000 bytes
output per socket) can easily lead to OOM (buffer would grow to 10GB of
memory).
Limit the maximum size of the buffer to five chunks, 1M each. Render and
flush buffers whenever we reach that.
This might make the resulting blo
On 2/13/19 12:53 PM, Peter Oskolkov wrote:
> This patchset implements BPF_LWT_ENCAP_IP mode in bpf_lwt_push_encap
> BPF helper. It enables BPF programs (specifically, BPF_PROG_TYPE_LWT_IN
> and BPF_PROG_TYPE_LWT_XMIT prog types) to add IP encapsulation headers
> to packets (e.g. IP/GRE, GUE, IPIP).
On Tue, Feb 12, 2019 at 12:43:21PM -0800, Joe Stringer wrote:
>
> Do you see any value in having incremental support in libbpf that
> could be used as a fallback for older kernels like in patch #2 of this
> series? I could imagine libbpf probing kernel support for
> global/static variables and att
From: Xin Long
Date: Tue, 12 Feb 2019 18:51:01 +0800
> In sctp_stream_init(), after sctp_stream_outq_migrate() freed the
> surplus streams' ext, but sctp_stream_alloc_out() returns -ENOMEM,
> stream->outcnt will not be set to 'outcnt'.
>
> With the bigger value on stream->outcnt, when closing th
From: Xin Long
Date: Tue, 12 Feb 2019 18:47:30 +0800
> Jianlin reported a panic when running sctp gso over gre over vlan device:
...
> It was caused by SKB_GSO_CB(skb)->csum_start not set in sctp_gso_segment.
> sctp_gso_segment() calls skb_segment() with 'feature | NETIF_F_HW_CSUM',
> which caus
1 - 100 of 300 matches
Mail list logo
