* [PATCH net v2 0/5] fsl/fman: fixes for ARM
From: Madalin Bucur @ 2016-12-19 9:22 UTC (permalink / raw)
To: netdev; +Cc: scott.wood, linuxppc-dev, linux-kernel, davem
The patch set fixes advertised speeds for QSGMII interfaces, disables
A007273 erratum workaround on non-PowerPC platforms where it does not
apply, enables compilation on ARM64 and addresses a probing issue on
non PPC platforms.
Changes from v1: unifying probing on all supported platforms
Igal Liberman (1):
fsl/fman: call of_platform_populate()
Madalin Bucur (4):
fsl/fman: fix 1G support for QSGMII interfaces
powerpc: remove fsl,fman from of_device_ids[]
fsl/fman: A007273 only applies to PPC SoCs
fsl/fman: enable compilation on ARM64
arch/powerpc/platforms/85xx/corenet_generic.c | 3 ---
drivers/net/ethernet/freescale/fman/Kconfig | 2 +-
drivers/net/ethernet/freescale/fman/fman.c | 16 ++++++++++++++++
drivers/net/ethernet/freescale/fman/mac.c | 1 +
4 files changed, 18 insertions(+), 4 deletions(-)
--
2.1.0
^ permalink raw reply
* Re: [PATCH 5/8] linux: drop __bitwise__ everywhere
From: Luca Coelho @ 2016-12-19 9:09 UTC (permalink / raw)
To: Michael S. Tsirkin, linux-kernel
Cc: Kukjin Kim, Krzysztof Kozlowski, Javier Martinez Canillas,
Russell King, Alasdair Kergon, Mike Snitzer, dm-devel, Shaohua Li,
Johannes Berg, Emmanuel Grumbach, Intel Linux Wireless,
Kalle Valo, Greg Kroah-Hartman, Jiri Slaby, Lee Duncan,
Chris Leech, James E.J. Bottomley, Martin K. Petersen,
Nicholas A. Bellinger, Jason Wang, Alexander Aring,
Stefan Schmidt, David S. Miller
In-Reply-To: <1481778865-27667-6-git-send-email-mst@redhat.com>
On Thu, 2016-12-15 at 07:15 +0200, Michael S. Tsirkin wrote:
> __bitwise__ used to mean "yes, please enable sparse checks
> unconditionally", but now that we dropped __CHECK_ENDIAN__
> __bitwise is exactly the same.
> There aren't many users, replace it by __bitwise everywhere.
>
> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
> ---
> arch/arm/plat-samsung/include/plat/gpio-cfg.h | 2 +-
> drivers/md/dm-cache-block-types.h | 6 +++---
> drivers/net/ethernet/sun/sunhme.h | 2 +-
> drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h | 4 ++--
For drivers/net/wireless/intel/iwlwifi/iwl-fw-file.h:
Acked-by: Luca Coelho <luciano.coelho@intel.com>
--
Luca.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply
* Re: [PATCH 8/8] Makefile: drop -D__CHECK_ENDIAN__ from cflags
From: Luca Coelho @ 2016-12-19 9:08 UTC (permalink / raw)
To: Michael S. Tsirkin, linux-kernel-u79uwXL29TY76Z2rM5mHXA
Cc: Emmanuel Grumbach, Stanislaw Gruszka, Gustavo Padovan,
Arend van Spriel, devel-gWbeCf7V1WCQmaza687I9mD2FQJk+8+b,
Jakub Kicinski, Stefan Schmidt,
linux-mediatek-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
wil6210-A+ZNKFmMK5xy9aJCnZT0Uw,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r, Chris Snook,
Wolfgang Grandegger, Jay Cliburn,
linux-wpan-u79uwXL29TY76Z2rM5mHXA, Johan Hedberg, Johannes Berg,
Intel Linux Wireless, Alexander Aring, Marcel Holtmann,
Hante Meuleman, linux-can-u79uwXL29TY76Z2rM5mHXA,
Marc Kleine-Budde, Matthias Brugger, Kall
In-Reply-To: <1481778865-27667-9-git-send-email-mst-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
On Thu, 2016-12-15 at 07:15 +0200, Michael S. Tsirkin wrote:
> That's the default now, no need for makefiles to set it.
>
> Signed-off-by: Michael S. Tsirkin <mst-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
> ---
> drivers/bluetooth/Makefile | 2 --
> drivers/net/can/Makefile | 1 -
> drivers/net/ethernet/altera/Makefile | 1 -
> drivers/net/ethernet/atheros/alx/Makefile | 1 -
> drivers/net/ethernet/freescale/Makefile | 2 --
> drivers/net/wireless/ath/Makefile | 2 --
> drivers/net/wireless/ath/wil6210/Makefile | 2 --
> drivers/net/wireless/broadcom/brcm80211/brcmfmac/Makefile | 2 --
> drivers/net/wireless/broadcom/brcm80211/brcmsmac/Makefile | 1 -
> drivers/net/wireless/intel/iwlegacy/Makefile | 2 --
> drivers/net/wireless/intel/iwlwifi/Makefile | 2 +-
> drivers/net/wireless/intel/iwlwifi/dvm/Makefile | 2 +-
> drivers/net/wireless/intel/iwlwifi/mvm/Makefile | 2 +-
For the drivers/net/wireless/intel/iwlwifi/ part:
Acked-by: Luca Coelho <luciano.coelho-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
--
Luca.
^ permalink raw reply
* ovs internal_dev mtu
From: nickcooper-zhangtonghao @ 2016-12-19 8:59 UTC (permalink / raw)
To: netdev
The upstream net-next has removed the “internal_dev_change_mtu”.
It will be ok for ovs(e.g. mtu_request) ? I think that code should
be keep. Can you provide more information ？
^ permalink raw reply
* Re: "rfkill: Add rfkill-any LED trigger" causes deadlock
From: Johannes Berg @ 2016-12-19 8:56 UTC (permalink / raw)
To: Михаил Кринкин,
kernel, linux-wireless, davem, netdev
In-Reply-To: <CACpa5=_chGuU3zVcVR3nkNySwz1cGAQTs39v+BjwcDNSvOqhTw@mail.gmail.com>
Hi Mike,
Thanks for the report. I'm sorry I missed this in review - obviously we
can't call something that acquires the mutex from rfkill_set_sw_state()
which clearly states, in the documentation:
* This function can be called in any context, even from within rfkill
* callbacks.
I've reverted the change (and the follow-up fix) now.
Michał, if you want to resubmit with this fixed, please also make sure
you don't reintroduce the unused label warning and have the appropriate
#ifdef that Arnd had later added for your change.
Thanks,
johannes
^ permalink raw reply
* Re: [PATCH v2 1/4] net: hix5hd2_gmac: add generic compatible string
From: Dongpo Li @ 2016-12-19 8:14 UTC (permalink / raw)
To: Rob Herring, David Miller
Cc: Mark Rutland, Michael Turquette, Stephen Boyd, Russell King,
Zhangfei Gao, Yisen Zhuang, salil.mehta, Arnd Bergmann,
Andrew Lunn, Jiancheng Xue, benjamin.chenhao, caizhiyong, netdev,
devicetree@vger.kernel.org, linux-kernel@vger.kernel.org
In-Reply-To: <CAL_Jsq+d5xHsxMKxbWSYy3cDXnC+HQj3=wun1zxf5PZ_BYrD-Q@mail.gmail.com>
Hi Rob and David,
On 2016/12/12 22:21, Rob Herring wrote:
> On Mon, Dec 12, 2016 at 5:16 AM, Dongpo Li <lidongpo@hisilicon.com> wrote:
>> Hi Rob,
>>
>> On 2016/12/10 6:35, Rob Herring wrote:
>>> On Mon, Dec 05, 2016 at 09:27:58PM +0800, Dongpo Li wrote:
>>>> The "hix5hd2" is SoC name, add the generic ethernet driver name.
>>>> The "hisi-gemac-v1" is the basic version and "hisi-gemac-v2" adds
>>>> the SG/TXCSUM/TSO/UFO features.
>>>>
>>>> Signed-off-by: Dongpo Li <lidongpo@hisilicon.com>
>>>> ---
>>>> .../devicetree/bindings/net/hisilicon-hix5hd2-gmac.txt | 9 +++++++--
>>>> drivers/net/ethernet/hisilicon/hix5hd2_gmac.c | 15 +++++++++++----
>>>> 2 files changed, 18 insertions(+), 6 deletions(-)
>>>>
>>>> diff --git a/Documentation/devicetree/bindings/net/hisilicon-hix5hd2-gmac.txt b/Documentation/devicetree/bindings/net/hisilicon-hix5hd2-gmac.txt
>>>> index 75d398b..75920f0 100644
>>>> --- a/Documentation/devicetree/bindings/net/hisilicon-hix5hd2-gmac.txt
>>>> +++ b/Documentation/devicetree/bindings/net/hisilicon-hix5hd2-gmac.txt
>>>> @@ -1,7 +1,12 @@
>>>> Hisilicon hix5hd2 gmac controller
>>>>
>>>> Required properties:
>>>> -- compatible: should be "hisilicon,hix5hd2-gmac".
>>>> +- compatible: should contain one of the following SoC strings:
>>>> + * "hisilicon,hix5hd2-gemac"
>>>> + * "hisilicon,hi3798cv200-gemac"
>>>> + and one of the following version string:
>>>> + * "hisilicon,hisi-gemac-v1"
>>>> + * "hisilicon,hisi-gemac-v2"
>>>
>>> What combinations are valid? I assume both chips don't have both v1 and
>>> v2. 2 SoCs and 2 versions so far, I don't think there is much point to
>>> have the v1 and v2 compatible strings.
>>>
>> The v1 and v2 are generic MAC compatible strings, many HiSilicon SoCs may
>> use the same MAC version. For example,
>> hix5hd2, hi3716cv200 SoCs use the v1 MAC version,
>> hi3798cv200, hi3516a SoCs use the v2 MAC version,
>> and there may be more SoCs added in future.
>> So I think the generic compatible strings are okay here.
>> Should I add the hi3716cv200, hi3516a SoCs compatible here?
>
> Yes.
>
>> Do you have any good advice?
>>
>>>> - reg: specifies base physical address(s) and size of the device registers.
>>>> The first region is the MAC register base and size.
>>>> The second region is external interface control register.
>>>> @@ -20,7 +25,7 @@ Required properties:
>>>>
>>>> Example:
>>>> gmac0: ethernet@f9840000 {
>>>> - compatible = "hisilicon,hix5hd2-gmac";
>>>> + compatible = "hisilicon,hix5hd2-gemac", "hisilicon,hisi-gemac-v1";
>>>
>>> You can't just change compatible strings.
>>>
>> Okay, maybe I should name all the compatible string with the suffix "-gmac" instead of
>> "-gemac". This can keep the compatible strings with the same suffix. Is this okay?
>> Can I just add the generic compatible string without changing the SoCs compatible string?
>> Like following:
>> gmac0: ethernet@f9840000 {
>> - compatible = "hisilicon,hix5hd2-gmac";
>> + compatible = "hisilicon,hix5hd2-gmac", "hisilicon,hisi-gmac-v1";
>
> Yes, this is fine.
As the patch series have been applied to net-next branch,
in which way should I commit this compatible fix?
Should I send a new patch fixing this compatible string error with "Fixes: xxx"?
Looking forward to your reply!
Regards,
Dongpo
.
^ permalink raw reply
* Re: [RFC v2 03/10] IB/hfi-vnic: Virtual Network Interface Controller (VNIC) netdev
From: Vishwanathapura, Niranjana @ 2016-12-19 6:43 UTC (permalink / raw)
To: Jason Gunthorpe
Cc: dledford-H+wXaHxf7aLQT0dZR+AlfA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA, netdev-u79uwXL29TY76Z2rM5mHXA,
dennis.dalessandro-ral2JQCrhuEAvxtiuMwx3w,
ira.weiny-ral2JQCrhuEAvxtiuMwx3w, Sadanand Warrier, Sudeep Dutt,
Tanya K Jajodia, Andrzej Kacprowski
In-Reply-To: <20161216042420.GB3797-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
On Thu, Dec 15, 2016 at 09:24:20PM -0700, Jason Gunthorpe wrote:
>> >>+struct __hfi_vesw_info {
>> >>+ u16 fabric_id;
>> >>+ u16 vesw_id;
>> >>+
>> >>+ u8 rsvd0[6];
>> >>+ u16 def_port_mask;
>> >>+
>> >>+ u8 rsvd1[2];
>> >>+ u16 pkey;
>> >>+
>> >>+ u8 rsvd2[4];
>> >>+ u32 u_mcast_dlid;
>> >>+ u32 u_ucast_dlid[HFI_VESW_MAX_NUM_DEF_PORT];
>> >>+
>> >>+ u8 rsvd3[44];
>> >>+ u16 eth_mtu[HFI_VNIC_MAX_NUM_PCP];
>> >>+ u16 eth_mtu_non_vlan;
>> >>+ u8 rsvd4[2];
>> >>+} __packed;
>> >
>> >This goes on the network too? Also looks like it has endian problems.
>> >
>> >Ditto for all the __packed structures.
>> >
>>
>> This is in CPU format. There is a separate big endian version of
>> this
>
>Why are CPU handled structures packed and full of reserved fields?
>Don't pack them if they are not pushed out to the network..
>
>There were lots of __packed structures, any that go on the network
>need be/le annoations.
>
Well, driver treats the reserved fields to be sticky. ie., information
block returned (upon GET) to EM is not changed (from SET) except few fields
which driver is expected to modify.
Structures that go on wire are big endian __packed structures in
hfi_vnic_encap.h. Ok, I will remove the __packed attribute from CPU handled
structures here.
Niranjana
>Jason
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* RE: [PATCH 3/3] net: fec: Fix typo in error msg and comment
From: Andy Duan @ 2016-12-19 2:14 UTC (permalink / raw)
To: Peter Meerwald-Stadler, netdev@vger.kernel.org; +Cc: trivial@kernel.org
In-Reply-To: <1481894616-4517-1-git-send-email-pmeerw@pmeerw.net>
From: Peter Meerwald-Stadler <pmeerw@pmeerw.net> Sent: Friday, December 16, 2016 9:24 PM
>To: netdev@vger.kernel.org
>Cc: Peter Meerwald-Stadler <pmeerw@pmeerw.net>; Andy Duan
><fugang.duan@nxp.com>; trivial@kernel.org
>Subject: [PATCH 3/3] net: fec: Fix typo in error msg and comment
>
>Signed-off-by: Peter Meerwald-Stadler <pmeerw@pmeerw.net>
>Cc: Fugang Duan <fugang.duan@nxp.com>
>Cc: trivial@kernel.org
>---
> drivers/net/ethernet/freescale/fec_main.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
Acked-by: Fugang Duan <fugang.duan@nxp.com>
>diff --git a/drivers/net/ethernet/freescale/fec_main.c
>b/drivers/net/ethernet/freescale/fec_main.c
>index 38160c2..250b790 100644
>--- a/drivers/net/ethernet/freescale/fec_main.c
>+++ b/drivers/net/ethernet/freescale/fec_main.c
>@@ -2001,7 +2001,7 @@ static int fec_enet_mii_init(struct platform_device
>*pdev)
> mii_speed--;
> if (mii_speed > 63) {
> dev_err(&pdev->dev,
>- "fec clock (%lu) to fast to get right mii speed\n",
>+ "fec clock (%lu) too fast to get right mii speed\n",
> clk_get_rate(fep->clk_ipg));
> err = -EINVAL;
> goto err_out;
>@@ -2950,7 +2950,7 @@ static void set_multicast_list(struct net_device
>*ndev)
> }
>
> /* only upper 6 bits (FEC_HASH_BITS) are used
>- * which point to specific bit in he hash registers
>+ * which point to specific bit in the hash registers
> */
> hash = (crc >> (32 - FEC_HASH_BITS)) & 0x3f;
>
>--
>1.7.10.4
^ permalink raw reply
* Re: [PATCH iproute2 v2 1/3] ifstat: Add extended statistics to ifstat
From: Or Gerlitz @ 2016-12-19 4:10 UTC (permalink / raw)
To: Nogah Frankel
Cc: Linux Netdev List, Stephen Hemminger, Roopa Prabhu, Jiri Pirko,
Elad Raz, Yotam Gigi, Ido Schimmel, Or Gerlitz
In-Reply-To: <1481806845-63384-2-git-send-email-nogahf@mellanox.com>
On Thu, Dec 15, 2016 at 3:00 PM, Nogah Frankel <nogahf@mellanox.com> wrote:
> +/* Note: if one xstat name in subset of another, it should be before it in this
> + * list.
> + * Name length must be under 64 chars.
> + */
nit "in subset" --> "is subset" ?
^ permalink raw reply
* [PATCH net-next 1/1] driver: ipvlan: Define common functions to decrease duplicated codes used to add or del IP address
From: fgao @ 2016-12-19 1:24 UTC (permalink / raw)
To: davem, maheshb, edumazet, netdev, gfree.wind; +Cc: Gao Feng
From: Gao Feng <fgao@ikuai8.com>
There are some duplicated codes in ipvlan_add_addr6/4 and
ipvlan_del_addr6/4. Now define two common functions ipvlan_add_addr
and ipvlan_del_addr to decrease the duplicated codes.
It could be helful to maintain the codes.
Signed-off-by: Gao Feng <fgao@ikuai8.com>
---
It is sent again because the first email is sent during net-next closing.
drivers/net/ipvlan/ipvlan_main.c | 68 +++++++++++++++++-----------------------
1 file changed, 29 insertions(+), 39 deletions(-)
diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c
index 693ec5b..5874d30 100644
--- a/drivers/net/ipvlan/ipvlan_main.c
+++ b/drivers/net/ipvlan/ipvlan_main.c
@@ -669,23 +669,22 @@ static int ipvlan_device_event(struct notifier_block *unused,
return NOTIFY_DONE;
}
-static int ipvlan_add_addr6(struct ipvl_dev *ipvlan, struct in6_addr *ip6_addr)
+static int ipvlan_add_addr(struct ipvl_dev *ipvlan, void *iaddr, bool is_v6)
{
struct ipvl_addr *addr;
- if (ipvlan_addr_busy(ipvlan->port, ip6_addr, true)) {
- netif_err(ipvlan, ifup, ipvlan->dev,
- "Failed to add IPv6=%pI6c addr for %s intf\n",
- ip6_addr, ipvlan->dev->name);
- return -EINVAL;
- }
addr = kzalloc(sizeof(struct ipvl_addr), GFP_ATOMIC);
if (!addr)
return -ENOMEM;
addr->master = ipvlan;
- memcpy(&addr->ip6addr, ip6_addr, sizeof(struct in6_addr));
- addr->atype = IPVL_IPV6;
+ if (is_v6) {
+ memcpy(&addr->ip6addr, iaddr, sizeof(struct in6_addr));
+ addr->atype = IPVL_IPV6;
+ } else {
+ memcpy(&addr->ip4addr, iaddr, sizeof(struct in_addr));
+ addr->atype = IPVL_IPV4;
+ }
list_add_tail(&addr->anode, &ipvlan->addrs);
/* If the interface is not up, the address will be added to the hash
@@ -697,11 +696,11 @@ static int ipvlan_add_addr6(struct ipvl_dev *ipvlan, struct in6_addr *ip6_addr)
return 0;
}
-static void ipvlan_del_addr6(struct ipvl_dev *ipvlan, struct in6_addr *ip6_addr)
+static void ipvlan_del_addr(struct ipvl_dev *ipvlan, void *iaddr, bool is_v6)
{
struct ipvl_addr *addr;
- addr = ipvlan_find_addr(ipvlan, ip6_addr, true);
+ addr = ipvlan_find_addr(ipvlan, iaddr, is_v6);
if (!addr)
return;
@@ -712,6 +711,23 @@ static void ipvlan_del_addr6(struct ipvl_dev *ipvlan, struct in6_addr *ip6_addr)
return;
}
+static int ipvlan_add_addr6(struct ipvl_dev *ipvlan, struct in6_addr *ip6_addr)
+{
+ if (ipvlan_addr_busy(ipvlan->port, ip6_addr, true)) {
+ netif_err(ipvlan, ifup, ipvlan->dev,
+ "Failed to add IPv6=%pI6c addr for %s intf\n",
+ ip6_addr, ipvlan->dev->name);
+ return -EINVAL;
+ }
+
+ return ipvlan_add_addr(ipvlan, ip6_addr, true);
+}
+
+static void ipvlan_del_addr6(struct ipvl_dev *ipvlan, struct in6_addr *ip6_addr)
+{
+ return ipvlan_del_addr(ipvlan, ip6_addr, true);
+}
+
static int ipvlan_addr6_event(struct notifier_block *unused,
unsigned long event, void *ptr)
{
@@ -745,45 +761,19 @@ static int ipvlan_addr6_event(struct notifier_block *unused,
static int ipvlan_add_addr4(struct ipvl_dev *ipvlan, struct in_addr *ip4_addr)
{
- struct ipvl_addr *addr;
-
if (ipvlan_addr_busy(ipvlan->port, ip4_addr, false)) {
netif_err(ipvlan, ifup, ipvlan->dev,
"Failed to add IPv4=%pI4 on %s intf.\n",
ip4_addr, ipvlan->dev->name);
return -EINVAL;
}
- addr = kzalloc(sizeof(struct ipvl_addr), GFP_KERNEL);
- if (!addr)
- return -ENOMEM;
-
- addr->master = ipvlan;
- memcpy(&addr->ip4addr, ip4_addr, sizeof(struct in_addr));
- addr->atype = IPVL_IPV4;
- list_add_tail(&addr->anode, &ipvlan->addrs);
-
- /* If the interface is not up, the address will be added to the hash
- * list by ipvlan_open.
- */
- if (netif_running(ipvlan->dev))
- ipvlan_ht_addr_add(ipvlan, addr);
- return 0;
+ return ipvlan_add_addr(ipvlan, ip4_addr, false);
}
static void ipvlan_del_addr4(struct ipvl_dev *ipvlan, struct in_addr *ip4_addr)
{
- struct ipvl_addr *addr;
-
- addr = ipvlan_find_addr(ipvlan, ip4_addr, false);
- if (!addr)
- return;
-
- ipvlan_ht_addr_del(addr);
- list_del(&addr->anode);
- kfree_rcu(addr, rcu);
-
- return;
+ return ipvlan_del_addr(ipvlan, ip4_addr, false);
}
static int ipvlan_addr4_event(struct notifier_block *unused,
--
1.9.1
^ permalink raw reply related
* [PATCH net-next 1/1] driver: ipvlan: Remove unnecessary ipvlan NULL check in ipvlan_count_rx
From: fgao @ 2016-12-19 1:24 UTC (permalink / raw)
To: davem, maheshb, edumazet, netdev, gfree.wind; +Cc: Gao Feng
From: Gao Feng <fgao@ikuai8.com>
There are three functions which would invoke the ipvlan_count_rx. They
are ipvlan_process_multicast, ipvlan_rcv_frame, and ipvlan_nf_input.
The former two functions already use the ipvlan directly before
ipvlan_count_rx, and ipvlan_nf_input gets the ipvlan from
ipvl_addr->master, it is not possible to be NULL too.
So the ipvlan pointer check is unnecessary in ipvlan_count_rx.
Signed-off-by: Gao Feng <fgao@ikuai8.com>
---
drivers/net/ipvlan/ipvlan_core.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c
index b4e9907..082f9f1 100644
--- a/drivers/net/ipvlan/ipvlan_core.c
+++ b/drivers/net/ipvlan/ipvlan_core.c
@@ -19,9 +19,6 @@ void ipvlan_init_secret(void)
static void ipvlan_count_rx(const struct ipvl_dev *ipvlan,
unsigned int len, bool success, bool mcast)
{
- if (!ipvlan)
- return;
-
if (likely(success)) {
struct ipvl_pcpu_stats *pcptr;
--
1.9.1
^ permalink raw reply related
* Re: [PATCH net] netfilter: check duplicate config when initializing in ipt_CLUSTERIP
From: Marcelo Ricardo Leitner @ 2016-12-19 1:02 UTC (permalink / raw)
To: Xin Long; +Cc: network dev, netfilter-devel, davem
In-Reply-To: <9b3e1f76b5670d33727e63b6e166ae416b0d65af.1481776300.git.lucien.xin@gmail.com>
On Thu, Dec 15, 2016 at 12:31:40PM +0800, Xin Long wrote:
> Now when adding an ipt_CLUSTERIP rule, it only checks duplicate config in
> clusterip_config_find_get(). But after that, there may be still another
> thread to insert a config with the same ip, then it leaves proc_create_data
> to do duplicate check.
>
> It's more reasonable to check duplicate config by ipt_CLUSTERIP itself,
> instead of checking it by proc fs duplicate file check. Before, when proc
> fs allowed duplicate name files in a directory, It could even crash kernel
> because of use-after-free.
>
> This patch is to check duplicate config under the protection of clusterip
> net lock when initializing a new config.
>
> Note that it also moves proc file node creation after adding new config, as
> proc_create_data may sleep, it couldn't be called under the clusterip_net
> lock. clusterip_config_find_get returns NULL if c->pde is null to make sure
> it can't be used until the proc file node creation is done.
>
> Suggested-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
> Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
> ---
> net/ipv4/netfilter/ipt_CLUSTERIP.c | 26 +++++++++++++++++++-------
> 1 file changed, 19 insertions(+), 7 deletions(-)
>
> diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c
> index 21db00d..0e71cac 100644
> --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c
> +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c
> @@ -144,7 +144,7 @@ clusterip_config_find_get(struct net *net, __be32 clusterip, int entry)
> rcu_read_lock_bh();
> c = __clusterip_config_find(net, clusterip);
> if (c) {
> - if (unlikely(!atomic_inc_not_zero(&c->refcount)))
> + if (!c->pde || unlikely(!atomic_inc_not_zero(&c->refcount)))
> c = NULL;
> else if (entry)
> atomic_inc(&c->entries);
> @@ -166,10 +166,11 @@ clusterip_config_init_nodelist(struct clusterip_config *c,
>
> static struct clusterip_config *
> clusterip_config_init(const struct ipt_clusterip_tgt_info *i, __be32 ip,
> - struct net_device *dev)
> + struct net_device *dev)
> {
> + struct net *net = dev_net(dev);
> struct clusterip_config *c;
> - struct clusterip_net *cn = net_generic(dev_net(dev), clusterip_net_id);
> + struct clusterip_net *cn = net_generic(net, clusterip_net_id);
>
> c = kzalloc(sizeof(*c), GFP_ATOMIC);
> if (!c)
> @@ -185,6 +186,17 @@ clusterip_config_init(const struct ipt_clusterip_tgt_info *i, __be32 ip,
> atomic_set(&c->refcount, 1);
> atomic_set(&c->entries, 1);
>
> + spin_lock_bh(&cn->lock);
> + if (__clusterip_config_find(net, ip)) {
> + spin_unlock_bh(&cn->lock);
> + kfree(c);
> +
> + return NULL;
> + }
> +
> + list_add_rcu(&c->list, &cn->configs);
> + spin_unlock_bh(&cn->lock);
> +
> #ifdef CONFIG_PROC_FS
> {
> char buffer[16];
> @@ -195,16 +207,16 @@ clusterip_config_init(const struct ipt_clusterip_tgt_info *i, __be32 ip,
> cn->procdir,
> &clusterip_proc_fops, c);
> if (!c->pde) {
> + spin_lock_bh(&cn->lock);
> + list_del_rcu(&c->list);
> + spin_unlock_bh(&cn->lock);
> kfree(c);
> +
> return NULL;
> }
> }
> #endif
>
> - spin_lock_bh(&cn->lock);
> - list_add_rcu(&c->list, &cn->configs);
> - spin_unlock_bh(&cn->lock);
> -
> return c;
> }
>
> --
> 2.1.0
>
> --
> To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
^ permalink raw reply
* RE: [PATCH iproute2 v2 1/3] ifstat: Add extended statistics to ifstat
From: Nogah Frankel @ 2016-12-18 14:22 UTC (permalink / raw)
To: Rami Rosen
Cc: Stephen Hemminger, netdev@vger.kernel.org,
roopa@cumulusnetworks.com, Jiri Pirko, Elad Raz, Yotam Gigi,
Ido Schimmel, Or Gerlitz
In-Reply-To: <CAKoUArk5oSseSGDdZ_TEGLbG-jabdArUmz3pZ7kdoZKiNpqPrQ@mail.gmail.com>
> -----Original Message-----
> From: Rami Rosen [mailto:roszenrami@gmail.com]
> Sent: Friday, December 16, 2016 5:21 PM
> To: Nogah Frankel <nogahf@mellanox.com>
> Cc: Stephen Hemminger <stephen@networkplumber.org>; netdev@vger.kernel.org;
> roopa@cumulusnetworks.com; Jiri Pirko <jiri@mellanox.com>; Elad Raz
> <eladr@mellanox.com>; Yotam Gigi <yotamg@mellanox.com>; Ido Schimmel
> <idosch@mellanox.com>; Or Gerlitz <ogerlitz@mellanox.com>
> Subject: Re: [PATCH iproute2 v2 1/3] ifstat: Add extended statistics to ifstat
>
> Hi,
>
> >Thanks, I'll fix it.
> Another minor nit, on this occasion:
>
> bool is_extanded should be: bool is_extended
>
> Regards,
> Rami Rosen
Thank you, I will fix it.
Nogah
^ permalink raw reply
* [PATCH] at86rf230: Allow slow GPIO pins for "rstn"
From: Andrey Smirnov @ 2016-12-18 23:25 UTC (permalink / raw)
To: linux-wpan
Cc: Andrey Smirnov, Alexander Aring, netdev, linux-kernel,
Chris Healy
Driver code never touches "rstn" signal in atomic context, so there's
no need to implicitly put such restriction on it by using gpio_set_value
to manipulate it. Replace gpio_set_value to gpio_set_value_cansleep to
fix that.
As a an example of where such restriction might be inconvenient,
consider a hardware design where "rstn" is connected to a pin of I2C/SPI
GPIO expander chip.
Cc: Chris Healy <cphealy@gmail.com>
Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
---
drivers/net/ieee802154/at86rf230.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ieee802154/at86rf230.c b/drivers/net/ieee802154/at86rf230.c
index 9f10da6..7700690 100644
--- a/drivers/net/ieee802154/at86rf230.c
+++ b/drivers/net/ieee802154/at86rf230.c
@@ -1710,9 +1710,9 @@ static int at86rf230_probe(struct spi_device *spi)
/* Reset */
if (gpio_is_valid(rstn)) {
udelay(1);
- gpio_set_value(rstn, 0);
+ gpio_set_value_cansleep(rstn, 0);
udelay(1);
- gpio_set_value(rstn, 1);
+ gpio_set_value_cansleep(rstn, 1);
usleep_range(120, 240);
}
--
2.5.5
^ permalink raw reply related
* Re: regression: ath_tx_edma_tasklet() Illegal idle entry in RCU read-side critical section
From: Gabriel C @ 2016-12-18 23:11 UTC (permalink / raw)
To: paulmck, Valo, Kalle
Cc: Tobias Klausmann, lkml, ath9k-devel,
linux-wireless@vger.kernel.org, ath9k-devel@lists.ath9k.org,
netdev@vger.kernel.org, nbd@nbd.name
In-Reply-To: <20161218201410.GS3924@linux.vnet.ibm.com>
On 18.12.2016 21:14, Paul E. McKenney wrote:
> On Sun, Dec 18, 2016 at 07:57:42PM +0000, Valo, Kalle wrote:
>> Tobias Klausmann <tobias.johannes.klausmann@mni.thm.de> writes:
>>
>>> A patch for this is already floating on the ML for a while now latest:
>>> (ath9k: do not return early to fix rcu unlocking)
>>
>> It's here:
>>
>> https://patchwork.kernel.org/patch/9472709/
>
> Feel free to add:
>
> Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
>
This patch fixes the bug for me.
You can add my Tested-by if you wish.
Tested-by: Gabriel Craciunescu <nix.or.die@gmail.com>
Regards,
Gabriel
^ permalink raw reply
* Re: [PATCH v3] net: macb: Added PCI wrapper for Platform Driver.
From: Andy Shevchenko @ 2016-12-18 22:43 UTC (permalink / raw)
To: Bartosz Folta
Cc: Nicolas Ferre, David S. Miller, Niklas Cassel, Alexandre Torgue,
Satanand Burla, Raghu Vatsavayi, Simon Horman,
linux-kernel@vger.kernel.org, netdev@vger.kernel.org,
Rafal Ozieblo
In-Reply-To: <SN1PR0701MB19516D29452ABA2695B2F061CC9A0@SN1PR0701MB1951.namprd07.prod.outlook.com>
On Wed, Dec 14, 2016 at 8:39 AM, Bartosz Folta <bfolta@cadence.com> wrote:
> There are hardware PCI implementations of Cadence GEM network
> controller. This patch will allow to use such hardware with reuse of
> existing Platform Driver.
Since it's already applied, perhaps you would consider addressing my
below comments as follow up clean up.
> +++ b/drivers/net/ethernet/cadence/macb_pci.c
> @@ -0,0 +1,153 @@
> +/**
> + * macb_pci.c - Cadence GEM PCI wrapper.
If at some point file is renamed it will be a bit more work to change
this line. Perhaps drop file name?
> +static int macb_probe(struct pci_dev *pdev, const struct pci_device_id *id)
> +{
> + int err;
> + struct platform_device *plat_dev;
> + struct platform_device_info plat_info;
> + struct macb_platform_data plat_data;
> + struct resource res[2];
> +
> + /* sanity check */
> + if (!id)
> + return -EINVAL;
Is it even possible in modern kernels?
> +
> + /* enable pci device */
> + err = pci_enable_device(pdev);
pcim_enable_device();
> + if (err < 0) {
> + dev_err(&pdev->dev, "Enabling PCI device has failed: 0x%04X",
%x for error code?! Confusing.
> + err);
> + return -EACCES;
return err;
> + }
> +
> + pci_set_master(pdev);
> +
> + /* set up resources */
> + memset(res, 0x00, sizeof(struct resource) * ARRAY_SIZE(res));
> + res[0].start = pdev->resource[0].start;
> + res[0].end = pdev->resource[0].end;
pci_resource_start()
pci_resource_end()
> + res[0].name = PCI_DRIVER_NAME;
> + res[0].flags = IORESOURCE_MEM;
> + res[1].start = pdev->irq;
Btw, does it support MSI? If so and are planning to use it,
pci_irq_...() API would help here.
> + res[1].name = PCI_DRIVER_NAME;
> + res[1].flags = IORESOURCE_IRQ;
> +
> + dev_info(&pdev->dev, "EMAC physical base addr = 0x%p\n",
> + (void *)(uintptr_t)pci_resource_start(pdev, 0));
Oh, %pa
> +
> + /* set up macb platform data */
> + memset(&plat_data, 0, sizeof(plat_data));
> +
> + /* initialize clocks */
> + plat_data.pclk = clk_register_fixed_rate(&pdev->dev, "pclk", NULL, 0,
> + GEM_PCLK_RATE);
> + if (IS_ERR(plat_data.pclk)) {
> + err = PTR_ERR(plat_data.pclk);
> + goto err_pclk_register;
> + }
> +
> + plat_data.hclk = clk_register_fixed_rate(&pdev->dev, "hclk", NULL, 0,
> + GEM_HCLK_RATE);
> + if (IS_ERR(plat_data.hclk)) {
> + err = PTR_ERR(plat_data.hclk);
> + goto err_hclk_register;
> + }
> +
> + /* set up platform device info */
> + memset(&plat_info, 0, sizeof(plat_info));
> + plat_info.parent = &pdev->dev;
> + plat_info.fwnode = pdev->dev.fwnode;
> + plat_info.name = PLAT_DRIVER_NAME;
> + plat_info.id = pdev->devfn;
> + plat_info.res = res;
> + plat_info.num_res = ARRAY_SIZE(res);
> + plat_info.data = &plat_data;
> + plat_info.size_data = sizeof(plat_data);
> + plat_info.dma_mask = DMA_BIT_MASK(32);
> +
> + /* register platform device */
> + plat_dev = platform_device_register_full(&plat_info);
> + if (IS_ERR(plat_dev)) {
> + err = PTR_ERR(plat_dev);
> + goto err_plat_dev_register;
> + }
> +
> + pci_set_drvdata(pdev, plat_dev);
> +
> + return 0;
> +
> +err_plat_dev_register:
> + clk_unregister(plat_data.hclk);
> +
> +err_hclk_register:
> + clk_unregister(plat_data.pclk);
> +
> +err_pclk_register:
> + pci_disable_device(pdev);
Redundant when use pcim_enable_device().
> + return err;
> +}
> +
> +static void macb_remove(struct pci_dev *pdev)
> +{
> + struct platform_device *plat_dev = pci_get_drvdata(pdev);
> + struct macb_platform_data *plat_data = dev_get_platdata(&plat_dev->dev);
> +
> + platform_device_unregister(plat_dev);
> + pci_disable_device(pdev);
Ditto.
> + clk_unregister(plat_data->pclk);
> + clk_unregister(plat_data->hclk);
> +}
--
With Best Regards,
Andy Shevchenko
^ permalink raw reply
* [PATCH RFC net-next 7/7] net: pending_confirm is not used anymore
From: Julian Anastasov @ 2016-12-18 20:56 UTC (permalink / raw)
To: netdev; +Cc: linux-sctp, YueHaibing
In-Reply-To: <1482094596-26046-1-git-send-email-ja@ssi.bg>
When same struct dst_entry can be used for many different
neighbours we can not use it for pending confirmations.
As last step, we can remove the pending_confirm flag.
Reported-by: YueHaibing <yuehaibing@huawei.com>
Fixes: 5110effee8fd ("net: Do delayed neigh confirmation.")
Fixes: f2bb4bedf35d ("ipv4: Cache output routes in fib_info nexthops.")
Signed-off-by: Julian Anastasov <ja@ssi.bg>
---
include/net/dst.h | 14 ++------------
net/core/dst.c | 1 -
2 files changed, 2 insertions(+), 13 deletions(-)
diff --git a/include/net/dst.h b/include/net/dst.h
index 3a3b34b..84a1043 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -59,8 +59,6 @@ struct dst_entry {
#define DST_XFRM_QUEUE 0x0100
#define DST_METADATA 0x0200
- unsigned short pending_confirm;
-
short error;
/* A non-zero value of dst->obsolete forces by-hand validation
@@ -78,6 +76,8 @@ struct dst_entry {
#define DST_OBSOLETE_KILL -2
unsigned short header_len; /* more space at head required */
unsigned short trailer_len; /* space to reserve at tail */
+ unsigned short __pad3;
+
#ifdef CONFIG_IP_ROUTE_CLASSID
__u32 tclassid;
#else
@@ -440,7 +440,6 @@ static inline void dst_rcu_free(struct rcu_head *head)
static inline void dst_confirm(struct dst_entry *dst)
{
- dst->pending_confirm = 1;
}
static inline int dst_neigh_output(struct dst_entry *dst, struct neighbour *n,
@@ -448,15 +447,6 @@ static inline int dst_neigh_output(struct dst_entry *dst, struct neighbour *n,
{
const struct hh_cache *hh;
- if (dst->pending_confirm) {
- unsigned long now = jiffies;
-
- dst->pending_confirm = 0;
- /* avoid dirtying neighbour */
- if (n->confirmed != now)
- n->confirmed = now;
- }
-
hh = &n->hh;
if ((n->nud_state & NUD_CONNECTED) && hh->hh_len)
return neigh_hh_output(hh, skb);
diff --git a/net/core/dst.c b/net/core/dst.c
index b5cbbe0..960e503 100644
--- a/net/core/dst.c
+++ b/net/core/dst.c
@@ -190,7 +190,6 @@ void dst_init(struct dst_entry *dst, struct dst_ops *ops,
dst->__use = 0;
dst->lastuse = jiffies;
dst->flags = flags;
- dst->pending_confirm = 0;
dst->next = NULL;
if (!(flags & DST_NOCOUNT))
dst_entries_add(ops, 1);
--
1.9.3
^ permalink raw reply related
* [PATCH RFC net-next 1/7] sock: add sk_dst_pending_confirm flag
From: Julian Anastasov @ 2016-12-18 20:56 UTC (permalink / raw)
To: netdev; +Cc: linux-sctp, YueHaibing
In-Reply-To: <1482094596-26046-1-git-send-email-ja@ssi.bg>
Add new sock flag to allow sockets to confirm neighbour.
When same struct dst_entry can be used for many different
neighbours we can not use it for pending confirmations.
As not all call paths lock the socket use full word for
the flag.
Add sk_dst_confirm as replacement for dst_confirm when
called for received packets.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
---
include/net/sock.h | 14 +++++++++++++-
net/core/sock.c | 2 ++
2 files changed, 15 insertions(+), 1 deletion(-)
diff --git a/include/net/sock.h b/include/net/sock.h
index 282d065..e83bb01 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -239,6 +239,7 @@ struct sock_common {
* @sk_wq: sock wait queue and async head
* @sk_rx_dst: receive input route used by early demux
* @sk_dst_cache: destination cache
+ * @sk_dst_pending_confirm: need to confirm neighbour
* @sk_policy: flow policy
* @sk_receive_queue: incoming packets
* @sk_wmem_alloc: transmit queue bytes committed
@@ -381,7 +382,7 @@ struct sock {
#endif
struct dst_entry *sk_rx_dst;
struct dst_entry __rcu *sk_dst_cache;
- atomic_t sk_omem_alloc;
+ int sk_dst_pending_confirm;
int sk_sndbuf;
/* ===== cache line for TX ===== */
@@ -405,6 +406,8 @@ struct sock {
unsigned int sk_gso_max_size;
gfp_t sk_allocation;
__u32 sk_txhash;
+ atomic_t sk_omem_alloc;
+ /* Note: 32bit hole on 64bit arches */
/*
* Because of non atomicity rules, all
@@ -1761,6 +1764,7 @@ static inline void dst_negative_advice(struct sock *sk)
if (ndst != dst) {
rcu_assign_pointer(sk->sk_dst_cache, ndst);
sk_tx_queue_clear(sk);
+ sk->sk_dst_pending_confirm = 0;
}
}
}
@@ -1771,6 +1775,7 @@ static inline void dst_negative_advice(struct sock *sk)
struct dst_entry *old_dst;
sk_tx_queue_clear(sk);
+ sk->sk_dst_pending_confirm = 0;
/*
* This can be called while sk is owned by the caller only,
* with no state that can be checked in a rcu_dereference_check() cond
@@ -1786,6 +1791,7 @@ static inline void dst_negative_advice(struct sock *sk)
struct dst_entry *old_dst;
sk_tx_queue_clear(sk);
+ sk->sk_dst_pending_confirm = 0;
old_dst = xchg((__force struct dst_entry **)&sk->sk_dst_cache, dst);
dst_release(old_dst);
}
@@ -1806,6 +1812,12 @@ static inline void dst_negative_advice(struct sock *sk)
struct dst_entry *sk_dst_check(struct sock *sk, u32 cookie);
+static inline void sk_dst_confirm(struct sock *sk)
+{
+ if (!sk->sk_dst_pending_confirm)
+ sk->sk_dst_pending_confirm = 1;
+}
+
bool sk_mc_loop(struct sock *sk);
static inline bool sk_can_gso(const struct sock *sk)
diff --git a/net/core/sock.c b/net/core/sock.c
index 9fa46b9..8af5296 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -502,6 +502,7 @@ struct dst_entry *__sk_dst_check(struct sock *sk, u32 cookie)
if (dst && dst->obsolete && dst->ops->check(dst, cookie) == NULL) {
sk_tx_queue_clear(sk);
+ sk->sk_dst_pending_confirm = 0;
RCU_INIT_POINTER(sk->sk_dst_cache, NULL);
dst_release(dst);
return NULL;
@@ -1522,6 +1523,7 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority)
af_family_clock_key_strings[newsk->sk_family]);
newsk->sk_dst_cache = NULL;
+ newsk->sk_dst_pending_confirm = 0;
newsk->sk_wmem_queued = 0;
newsk->sk_forward_alloc = 0;
atomic_set(&newsk->sk_drops, 0);
--
1.9.3
^ permalink raw reply related
* [PATCH RFC net-next 6/7] net: use dst_confirm_neigh for UDP, RAW, ICMP, L2TP
From: Julian Anastasov @ 2016-12-18 20:56 UTC (permalink / raw)
To: netdev; +Cc: linux-sctp, YueHaibing
In-Reply-To: <1482094596-26046-1-git-send-email-ja@ssi.bg>
When same struct dst_entry can be used for many different
neighbours we can not use it for pending confirmations.
The datagram protocols can use MSG_CONFIRM to confirm the
neighbour. When used with MSG_PROBE we do not reach the
code where neighbour is confirmed, so we have to do the
same slow lookup by using the dst_confirm_neigh() helper.
When MSG_PROBE is not used, ip_append_data/ip6_append_data
will set the skb flag dst_pending_confirm.
Reported-by: YueHaibing <yuehaibing@huawei.com>
Fixes: 5110effee8fd ("net: Do delayed neigh confirmation.")
Fixes: f2bb4bedf35d ("ipv4: Cache output routes in fib_info nexthops.")
Signed-off-by: Julian Anastasov <ja@ssi.bg>
---
net/ipv4/ip_output.c | 6 ++++++
net/ipv4/ping.c | 3 ++-
net/ipv4/raw.c | 6 +++++-
net/ipv4/udp.c | 3 ++-
net/ipv6/ip6_output.c | 6 ++++++
net/ipv6/raw.c | 6 +++++-
net/ipv6/route.c | 27 ++++++++++++++-------------
net/ipv6/udp.c | 3 ++-
net/l2tp/l2tp_ip6.c | 3 ++-
9 files changed, 44 insertions(+), 19 deletions(-)
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index fbe63cc..3afab90 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -889,6 +889,9 @@ static inline int ip_ufo_append_data(struct sock *sk,
skb->csum = 0;
+ if (flags & MSG_CONFIRM)
+ skb->dst_pending_confirm = 1;
+
__skb_queue_tail(queue, skb);
} else if (skb_is_gso(skb)) {
goto append;
@@ -1089,6 +1092,9 @@ static int __ip_append_data(struct sock *sk,
exthdrlen = 0;
csummode = CHECKSUM_NONE;
+ if ((flags & MSG_CONFIRM) && !skb_prev)
+ skb->dst_pending_confirm = 1;
+
/*
* Put the packet on the pending queue.
*/
diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index 86cca61..f3f6f50 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -848,7 +848,8 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
return err;
do_confirm:
- dst_confirm(&rt->dst);
+ if (msg->msg_flags & MSG_PROBE)
+ dst_confirm_neigh(&rt->dst, &fl4.daddr);
if (!(msg->msg_flags & MSG_PROBE) || len)
goto back_from_confirm;
err = 0;
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 2300fae..47c5451 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -383,6 +383,9 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4,
sock_tx_timestamp(sk, sockc->tsflags, &skb_shinfo(skb)->tx_flags);
+ if (flags & MSG_CONFIRM)
+ skb->dst_pending_confirm = 1;
+
skb->transport_header = skb->network_header;
err = -EFAULT;
if (memcpy_from_msg(iph, msg, length))
@@ -666,7 +669,8 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
return len;
do_confirm:
- dst_confirm(&rt->dst);
+ if (msg->msg_flags & MSG_PROBE)
+ dst_confirm_neigh(&rt->dst, &fl4.daddr);
if (!(msg->msg_flags & MSG_PROBE) || len)
goto back_from_confirm;
err = 0;
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 9ca279b..22a1eb1 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1113,7 +1113,8 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
return err;
do_confirm:
- dst_confirm(&rt->dst);
+ if (msg->msg_flags & MSG_PROBE)
+ dst_confirm_neigh(&rt->dst, &fl4->daddr);
if (!(msg->msg_flags&MSG_PROBE) || len)
goto back_from_confirm;
err = 0;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 285aa9f..efe8b3f 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1145,6 +1145,9 @@ static inline int ip6_ufo_append_data(struct sock *sk,
skb->protocol = htons(ETH_P_IPV6);
skb->csum = 0;
+ if (flags & MSG_CONFIRM)
+ skb->dst_pending_confirm = 1;
+
__skb_queue_tail(queue, skb);
} else if (skb_is_gso(skb)) {
goto append;
@@ -1517,6 +1520,9 @@ static int __ip6_append_data(struct sock *sk,
exthdrlen = 0;
dst_exthdrlen = 0;
+ if ((flags & MSG_CONFIRM) && !skb_prev)
+ skb->dst_pending_confirm = 1;
+
/*
* Put the packet on the pending queue
*/
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 291ebc2..0597548 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -650,6 +650,9 @@ static int rawv6_send_hdrinc(struct sock *sk, struct msghdr *msg, int length,
skb->ip_summed = CHECKSUM_NONE;
+ if (flags & MSG_CONFIRM)
+ skb->dst_pending_confirm = 1;
+
skb->transport_header = skb->network_header;
err = memcpy_from_msg(iph, msg, length);
if (err)
@@ -930,7 +933,8 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
txopt_put(opt_to_free);
return err < 0 ? err : len;
do_confirm:
- dst_confirm(dst);
+ if (msg->msg_flags & MSG_PROBE)
+ dst_confirm_neigh(dst, &fl6.daddr);
if (!(msg->msg_flags & MSG_PROBE) || len)
goto back_from_confirm;
err = 0;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 4056d0e..c8e3730 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1375,6 +1375,7 @@ static bool rt6_cache_allowed_for_pmtu(const struct rt6_info *rt)
static void __ip6_rt_update_pmtu(struct dst_entry *dst, const struct sock *sk,
const struct ipv6hdr *iph, u32 mtu)
{
+ const struct in6_addr *daddr, *saddr;
struct rt6_info *rt6 = (struct rt6_info *)dst;
if (rt6->rt6i_flags & RTF_LOCAL)
@@ -1383,26 +1384,26 @@ static void __ip6_rt_update_pmtu(struct dst_entry *dst, const struct sock *sk,
if (dst_metric_locked(dst, RTAX_MTU))
return;
- dst_confirm(dst);
+ if (iph) {
+ daddr = &iph->daddr;
+ saddr = &iph->saddr;
+ } else if (sk) {
+ daddr = &sk->sk_v6_daddr;
+ saddr = &inet6_sk(sk)->saddr;
+ } else {
+ daddr = NULL;
+ saddr = NULL;
+ }
+ dst_confirm_neigh(dst, daddr);
mtu = max_t(u32, mtu, IPV6_MIN_MTU);
if (mtu >= dst_mtu(dst))
return;
if (!rt6_cache_allowed_for_pmtu(rt6)) {
rt6_do_update_pmtu(rt6, mtu);
- } else {
- const struct in6_addr *daddr, *saddr;
+ } else if (daddr) {
struct rt6_info *nrt6;
- if (iph) {
- daddr = &iph->daddr;
- saddr = &iph->saddr;
- } else if (sk) {
- daddr = &sk->sk_v6_daddr;
- saddr = &inet6_sk(sk)->saddr;
- } else {
- return;
- }
nrt6 = ip6_rt_cache_alloc(rt6, daddr, saddr);
if (nrt6) {
rt6_do_update_pmtu(nrt6, mtu);
@@ -2274,7 +2275,7 @@ static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_bu
* Look, redirects are sent only in response to data packets,
* so that this nexthop apparently is reachable. --ANK
*/
- dst_confirm(&rt->dst);
+ dst_confirm_neigh(&rt->dst, &ipv6_hdr(skb)->saddr);
neigh = __neigh_lookup(&nd_tbl, &msg->target, skb->dev, 1);
if (!neigh)
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 649efc2..9956b7e 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1295,7 +1295,8 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
return err;
do_confirm:
- dst_confirm(dst);
+ if (msg->msg_flags & MSG_PROBE)
+ dst_confirm_neigh(dst, &fl6.daddr);
if (!(msg->msg_flags&MSG_PROBE) || len)
goto back_from_confirm;
err = 0;
diff --git a/net/l2tp/l2tp_ip6.c b/net/l2tp/l2tp_ip6.c
index f092ac4..060dbae 100644
--- a/net/l2tp/l2tp_ip6.c
+++ b/net/l2tp/l2tp_ip6.c
@@ -658,7 +658,8 @@ static int l2tp_ip6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
return err < 0 ? err : len;
do_confirm:
- dst_confirm(dst);
+ if (msg->msg_flags & MSG_PROBE)
+ dst_confirm_neigh(dst, &fl6.daddr);
if (!(msg->msg_flags & MSG_PROBE) || len)
goto back_from_confirm;
err = 0;
--
1.9.3
^ permalink raw reply related
* [PATCH RFC net-next 0/7] net: dst_confirm replacement
From: Julian Anastasov @ 2016-12-18 20:56 UTC (permalink / raw)
To: netdev; +Cc: linux-sctp, YueHaibing
This patchset addresses the problem of neighbour
confirmation where received replies from one nexthop
can cause confirmation of different nexthop when using
the same dst. Thanks to YueHaibing <yuehaibing@huawei.com>
for tracking the dst->pending_confirm problem.
Sockets can obtain cached output route. Such
routes can be to known nexthop (rt_gateway=IP) or to be
used simultaneously for different nexthop IPs by different
subnet prefixes (nh->nh_scope = RT_SCOPE_HOST, rt_gateway=0).
At first look, there are more problems:
- dst_confirm() sets flag on dst and not on dst->path,
as result, indication is lost when XFRM is used
- DNAT can change the nexthop, so the really used nexthop is
not confirmed
So, the following solution is to avoid using
dst->pending_confirm.
The current dst_confirm() usage is as follows:
Protocols confirming dst on received packets:
- TCP (1 dst per socket)
- SCTP (1 dst per transport)
- CXGB*
Protocols supporting sendmsg with MSG_CONFIRM [ | MSG_PROBE ] to
confirm neighbour:
- UDP IPv4/IPv6
- ICMPv4 PING
- RAW IPv4/IPv6
- L2TP/IPv6
MSG_CONFIRM for other purposes (fix not needed):
- CAN
Sending without locking the socket:
- UDP (when no cork)
- RAW (when hdrincl=1)
Redirects from old to new GW:
- rt6_do_redirect
The patchset includes the following changes:
1. sock: add sk_dst_pending_confirm flag
- used only by TCP with patch 4 to remember the received
indication in sk->sk_dst_pending_confirm
2. net: add dst_pending_confirm flag to skbuff
- skb->dst_pending_confirm will be used by all protocols
in following patches
3. sctp: add dst_pending_confirm flag
- SCTP uses per-transport dsts and can not use
sk->sk_dst_pending_confirm like TCP
4. tcp: replace dst_confirm with sk_dst_confirm
5. net: add confirm_neigh method to dst_ops
- IPv4 and IPv6 provision for slow neigh lookups for MSG_PROBE users.
I decided to use neigh lookup only for this case because on
MSG_PROBE the skb may pass MTU checks but it does not reach
the neigh confirmation code. This patch will be used from patch 6.
- xfrm_confirm_neigh: support is incomplete here, only routes with
known nexthops (gateway) are supported because the tunnel address
is slow to obtain. Or there is solution to this problem?
6. net: use dst_confirm_neigh for UDP, RAW, ICMP, L2TP
- dst_confirm conversion for UDP, RAW, ICMP and L2TP/IPv6
- these protocols use MSG_CONFIRM propagated by ip*_append_data
to skb->dst_pending_confirm. sk->sk_dst_pending_confirm is not
used because some sending paths do not lock the socket. For
MSG_PROBE we use the slow lookup (dst_confirm_neigh).
- there are also 2 cases that need the slow lookup:
__ip6_rt_update_pmtu and rt6_do_redirect. I hope
&ipv6_hdr(skb)->saddr is the correct nexthop address to use here.
7. net: pending_confirm is not used anymore
- I failed to understand the CXGB* code, I see dst_confirm()
calls but I'm not sure dst_neigh_output() was called. For now
I just removed the dst->pending_confirm flag and left all
dst_confirm() calls there. Any better idea?
- Now may be old function neigh_output() should be restored
instead of dst_neigh_output?
Julian Anastasov (7):
sock: add sk_dst_pending_confirm flag
net: add dst_pending_confirm flag to skbuff
sctp: add dst_pending_confirm flag
tcp: replace dst_confirm with sk_dst_confirm
net: add confirm_neigh method to dst_ops
net: use dst_confirm_neigh for UDP, RAW, ICMP, L2TP
net: pending_confirm is not used anymore
drivers/net/vrf.c | 5 ++++-
include/linux/skbuff.h | 4 +++-
include/net/arp.h | 16 ++++++++++++++++
include/net/dst.h | 21 +++++++++------------
include/net/dst_ops.h | 2 ++
include/net/ndisc.h | 17 +++++++++++++++++
include/net/sctp/sctp.h | 6 ++----
include/net/sctp/structs.h | 4 ++++
include/net/sock.h | 28 +++++++++++++++++++++++++++-
net/core/dst.c | 1 -
net/core/sock.c | 2 ++
net/ipv4/ip_output.c | 11 ++++++++++-
net/ipv4/ping.c | 3 ++-
net/ipv4/raw.c | 6 +++++-
net/ipv4/route.c | 19 +++++++++++++++++++
net/ipv4/tcp_input.c | 12 +++---------
net/ipv4/tcp_metrics.c | 7 ++-----
net/ipv4/tcp_output.c | 2 ++
net/ipv4/udp.c | 3 ++-
net/ipv6/ip6_output.c | 7 +++++++
net/ipv6/raw.c | 6 +++++-
net/ipv6/route.c | 43 ++++++++++++++++++++++++++++++-------------
net/ipv6/udp.c | 3 ++-
net/l2tp/l2tp_ip6.c | 3 ++-
net/sctp/associola.c | 3 +--
net/sctp/output.c | 10 +++++++++-
net/sctp/outqueue.c | 2 +-
net/sctp/sm_make_chunk.c | 6 ++----
net/sctp/sm_sideeffect.c | 2 +-
net/sctp/socket.c | 4 ++--
net/sctp/transport.c | 18 +++++++++++++++++-
net/xfrm/xfrm_policy.c | 16 ++++++++++++++++
32 files changed, 226 insertions(+), 66 deletions(-)
--
1.9.3
^ permalink raw reply
* [PATCH RFC net-next 3/7] sctp: add dst_pending_confirm flag
From: Julian Anastasov @ 2016-12-18 20:56 UTC (permalink / raw)
To: netdev; +Cc: linux-sctp, YueHaibing
In-Reply-To: <1482094596-26046-1-git-send-email-ja@ssi.bg>
Add new transport flag to allow sockets to confirm neighbour.
When same struct dst_entry can be used for many different
neighbours we can not use it for pending confirmations.
The flag is propagated from transport to every packet.
It is reset when cached dst is reset.
Reported-by: YueHaibing <yuehaibing@huawei.com>
Fixes: 5110effee8fd ("net: Do delayed neigh confirmation.")
Fixes: f2bb4bedf35d ("ipv4: Cache output routes in fib_info nexthops.")
Signed-off-by: Julian Anastasov <ja@ssi.bg>
---
include/net/sctp/sctp.h | 6 ++----
include/net/sctp/structs.h | 4 ++++
net/sctp/associola.c | 3 +--
net/sctp/output.c | 10 +++++++++-
net/sctp/outqueue.c | 2 +-
net/sctp/sm_make_chunk.c | 6 ++----
net/sctp/sm_sideeffect.c | 2 +-
net/sctp/socket.c | 4 ++--
net/sctp/transport.c | 18 +++++++++++++++++-
9 files changed, 39 insertions(+), 16 deletions(-)
diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index f0dcaeb..85d9083 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -586,10 +586,8 @@ static inline void sctp_v4_map_v6(union sctp_addr *addr)
*/
static inline struct dst_entry *sctp_transport_dst_check(struct sctp_transport *t)
{
- if (t->dst && !dst_check(t->dst, t->dst_cookie)) {
- dst_release(t->dst);
- t->dst = NULL;
- }
+ if (t->dst && !dst_check(t->dst, t->dst_cookie))
+ sctp_transport_dst_release(t);
return t->dst;
}
diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 92daabd..e842e84 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -838,6 +838,8 @@ struct sctp_transport {
__u32 burst_limited; /* Holds old cwnd when max.burst is applied */
+ __u32 dst_pending_confirm; /* need to confirm neighbour */
+
/* Destination */
struct dst_entry *dst;
/* Source address. */
@@ -980,6 +982,8 @@ void sctp_transport_route(struct sctp_transport *, union sctp_addr *,
void sctp_transport_reset(struct sctp_transport *);
void sctp_transport_update_pmtu(struct sock *, struct sctp_transport *, u32);
void sctp_transport_immediate_rtx(struct sctp_transport *);
+void sctp_transport_dst_release(struct sctp_transport *t);
+void sctp_transport_dst_confirm(struct sctp_transport *t);
/* This is the structure we use to queue packets as they come into
diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index 68428e1..7bd26e0 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -820,8 +820,7 @@ void sctp_assoc_control_transport(struct sctp_association *asoc,
if (transport->state != SCTP_UNCONFIRMED)
transport->state = SCTP_INACTIVE;
else {
- dst_release(transport->dst);
- transport->dst = NULL;
+ sctp_transport_dst_release(transport);
ulp_notify = false;
}
diff --git a/net/sctp/output.c b/net/sctp/output.c
index f5320a8..4684a00 100644
--- a/net/sctp/output.c
+++ b/net/sctp/output.c
@@ -550,6 +550,7 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp)
struct sctp_association *asoc = tp->asoc;
struct sctp_chunk *chunk, *tmp;
int pkt_count, gso = 0;
+ int confirm;
struct dst_entry *dst;
struct sk_buff *head;
struct sctphdr *sh;
@@ -628,7 +629,14 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp)
asoc->peer.last_sent_to = tp;
}
head->ignore_df = packet->ipfragok;
- tp->af_specific->sctp_xmit(head, tp);
+ confirm = tp->dst_pending_confirm;
+ if (confirm)
+ head->dst_pending_confirm = 1;
+ /* neighbour should be confirmed on successful transmission or
+ * positive error
+ */
+ if (tp->af_specific->sctp_xmit(head, tp) >= 0 && confirm)
+ tp->dst_pending_confirm = 0;
out:
list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) {
diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index e540826..8234799 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -1641,7 +1641,7 @@ static void sctp_check_transmitted(struct sctp_outq *q,
if (forward_progress) {
if (transport->dst)
- dst_confirm(transport->dst);
+ sctp_transport_dst_confirm(transport);
}
}
diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index 9e9690b..6fb15bf 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -3317,8 +3317,7 @@ static void sctp_asconf_param_success(struct sctp_association *asoc,
local_bh_enable();
list_for_each_entry(transport, &asoc->peer.transport_addr_list,
transports) {
- dst_release(transport->dst);
- transport->dst = NULL;
+ sctp_transport_dst_release(transport);
}
break;
case SCTP_PARAM_DEL_IP:
@@ -3332,8 +3331,7 @@ static void sctp_asconf_param_success(struct sctp_association *asoc,
local_bh_enable();
list_for_each_entry(transport, &asoc->peer.transport_addr_list,
transports) {
- dst_release(transport->dst);
- transport->dst = NULL;
+ sctp_transport_dst_release(transport);
}
break;
default:
diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c
index c345bf1..9255b29 100644
--- a/net/sctp/sm_sideeffect.c
+++ b/net/sctp/sm_sideeffect.c
@@ -723,7 +723,7 @@ static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds,
* forward progress.
*/
if (t->dst)
- dst_confirm(t->dst);
+ sctp_transport_dst_confirm(t);
/* The receiver of the HEARTBEAT ACK should also perform an
* RTT measurement for that destination transport address
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 318c678..9ee676a 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -588,7 +588,7 @@ static int sctp_send_asconf_add_ip(struct sock *sk,
list_for_each_entry(trans,
&asoc->peer.transport_addr_list, transports) {
/* Clear the source and route cache */
- dst_release(trans->dst);
+ sctp_transport_dst_release(trans);
trans->cwnd = min(4*asoc->pathmtu, max_t(__u32,
2*asoc->pathmtu, 4380));
trans->ssthresh = asoc->peer.i.a_rwnd;
@@ -839,7 +839,7 @@ static int sctp_send_asconf_del_ip(struct sock *sk,
*/
list_for_each_entry(transport, &asoc->peer.transport_addr_list,
transports) {
- dst_release(transport->dst);
+ sctp_transport_dst_release(transport);
sctp_transport_route(transport, NULL,
sctp_sk(asoc->base.sk));
}
diff --git a/net/sctp/transport.c b/net/sctp/transport.c
index ce54dce..db66514 100644
--- a/net/sctp/transport.c
+++ b/net/sctp/transport.c
@@ -227,7 +227,7 @@ void sctp_transport_pmtu(struct sctp_transport *transport, struct sock *sk)
{
/* If we don't have a fresh route, look one up */
if (!transport->dst || transport->dst->obsolete) {
- dst_release(transport->dst);
+ sctp_transport_dst_release(transport);
transport->af_specific->get_dst(transport, &transport->saddr,
&transport->fl, sk);
}
@@ -659,3 +659,19 @@ void sctp_transport_immediate_rtx(struct sctp_transport *t)
sctp_transport_hold(t);
}
}
+
+/* Drop dst */
+void sctp_transport_dst_release(struct sctp_transport *t)
+{
+ dst_release(t->dst);
+ t->dst = NULL;
+ t->dst_pending_confirm = 0;
+}
+
+/* Schedule neighbour confirm */
+void sctp_transport_dst_confirm(struct sctp_transport *t)
+{
+ if (!t->dst_pending_confirm)
+ t->dst_pending_confirm = 1;
+}
+
--
1.9.3
^ permalink raw reply related
* [PATCH RFC net-next 4/7] tcp: replace dst_confirm with sk_dst_confirm
From: Julian Anastasov @ 2016-12-18 20:56 UTC (permalink / raw)
To: netdev; +Cc: linux-sctp, YueHaibing
In-Reply-To: <1482094596-26046-1-git-send-email-ja@ssi.bg>
When same struct dst_entry can be used for many different
neighbours we can not use it for pending confirmations.
Use the new sk_dst_confirm() helper to propagate the
indication from received packets to sock_confirm_neigh().
Reported-by: YueHaibing <yuehaibing@huawei.com>
Fixes: 5110effee8fd ("net: Do delayed neigh confirmation.")
Fixes: f2bb4bedf35d ("ipv4: Cache output routes in fib_info nexthops.")
Signed-off-by: Julian Anastasov <ja@ssi.bg>
---
net/ipv4/tcp_input.c | 12 +++---------
net/ipv4/tcp_metrics.c | 7 ++-----
net/ipv4/tcp_output.c | 2 ++
3 files changed, 7 insertions(+), 14 deletions(-)
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 6c79075..005c428 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3698,11 +3698,8 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
if (tp->tlp_high_seq)
tcp_process_tlp_ack(sk, ack, flag);
- if ((flag & FLAG_FORWARD_PROGRESS) || !(flag & FLAG_NOT_DUP)) {
- struct dst_entry *dst = __sk_dst_get(sk);
- if (dst)
- dst_confirm(dst);
- }
+ if ((flag & FLAG_FORWARD_PROGRESS) || !(flag & FLAG_NOT_DUP))
+ sk_dst_confirm(sk);
if (icsk->icsk_pending == ICSK_TIME_RETRANS)
tcp_schedule_loss_probe(sk);
@@ -6022,7 +6019,6 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb)
break;
case TCP_FIN_WAIT1: {
- struct dst_entry *dst;
int tmo;
/* If we enter the TCP_FIN_WAIT1 state and we are a
@@ -6049,9 +6045,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb)
tcp_set_state(sk, TCP_FIN_WAIT2);
sk->sk_shutdown |= SEND_SHUTDOWN;
- dst = __sk_dst_get(sk);
- if (dst)
- dst_confirm(dst);
+ sk_dst_confirm(sk);
if (!sock_flag(sk, SOCK_DEAD)) {
/* Wake up lingering close() */
diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c
index d46f4d5..1b335d6 100644
--- a/net/ipv4/tcp_metrics.c
+++ b/net/ipv4/tcp_metrics.c
@@ -375,12 +375,10 @@ void tcp_update_metrics(struct sock *sk)
u32 val;
int m;
+ sk_dst_confirm(sk);
if (sysctl_tcp_nometrics_save || !dst)
return;
- if (dst->flags & DST_HOST)
- dst_confirm(dst);
-
rcu_read_lock();
if (icsk->icsk_backoff || !tp->srtt_us) {
/* This session failed to estimate rtt. Why?
@@ -493,11 +491,10 @@ void tcp_init_metrics(struct sock *sk)
struct tcp_metrics_block *tm;
u32 val, crtt = 0; /* cached RTT scaled by 8 */
+ sk_dst_confirm(sk);
if (!dst)
goto reset;
- dst_confirm(dst);
-
rcu_read_lock();
tm = tcp_get_metrics(sk, dst, true);
if (!tm) {
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index b45101f..dc01f35 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -974,6 +974,8 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it,
skb_set_hash_from_sk(skb, sk);
atomic_add(skb->truesize, &sk->sk_wmem_alloc);
+ skb->dst_pending_confirm = sk->sk_dst_pending_confirm;
+
/* Build TCP header and checksum it. */
th = (struct tcphdr *)skb->data;
th->source = inet->inet_sport;
--
1.9.3
^ permalink raw reply related
* [PATCH RFC net-next 2/7] net: add dst_pending_confirm flag to skbuff
From: Julian Anastasov @ 2016-12-18 20:56 UTC (permalink / raw)
To: netdev; +Cc: linux-sctp, YueHaibing
In-Reply-To: <1482094596-26046-1-git-send-email-ja@ssi.bg>
Add new skbuff flag to allow protocols to confirm neighbour.
When same struct dst_entry can be used for many different
neighbours we can not use it for pending confirmations.
Add sock_confirm_neigh() helper to confirm the neighbour and
use it for IPv4, IPv6 and VRF before dst_neigh_output.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
---
drivers/net/vrf.c | 5 ++++-
include/linux/skbuff.h | 4 +++-
include/net/sock.h | 14 ++++++++++++++
net/ipv4/ip_output.c | 5 ++++-
net/ipv6/ip6_output.c | 1 +
5 files changed, 26 insertions(+), 3 deletions(-)
diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c
index 7532646..b118d2b 100644
--- a/drivers/net/vrf.c
+++ b/drivers/net/vrf.c
@@ -377,6 +377,7 @@ static int vrf_finish_output6(struct net *net, struct sock *sk,
if (unlikely(!neigh))
neigh = __neigh_create(&nd_tbl, nexthop, dst->dev, false);
if (!IS_ERR(neigh)) {
+ sock_confirm_neigh(skb, neigh);
ret = dst_neigh_output(dst, neigh, skb);
rcu_read_unlock_bh();
return ret;
@@ -573,8 +574,10 @@ static int vrf_finish_output(struct net *net, struct sock *sk, struct sk_buff *s
neigh = __ipv4_neigh_lookup_noref(dev, nexthop);
if (unlikely(!neigh))
neigh = __neigh_create(&arp_tbl, &nexthop, dev, false);
- if (!IS_ERR(neigh))
+ if (!IS_ERR(neigh)) {
+ sock_confirm_neigh(skb, neigh);
ret = dst_neigh_output(dst, neigh, skb);
+ }
rcu_read_unlock_bh();
err:
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index ac7fa34..94d7c36 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -610,6 +610,7 @@ static inline bool skb_mstamp_after(const struct skb_mstamp *t1,
* @wifi_acked_valid: wifi_acked was set
* @wifi_acked: whether frame was acked on wifi or not
* @no_fcs: Request NIC to treat last 4 bytes as Ethernet FCS
+ * @dst_pending_confirm: need to confirm neighbour
* @napi_id: id of the NAPI struct this skb came from
* @secmark: security marking
* @mark: Generic packet mark
@@ -740,6 +741,7 @@ struct sk_buff {
__u8 csum_level:2;
__u8 csum_bad:1;
+ __u8 dst_pending_confirm:1;
#ifdef CONFIG_IPV6_NDISC_NODETYPE
__u8 ndisc_nodetype:2;
#endif
@@ -749,7 +751,7 @@ struct sk_buff {
#ifdef CONFIG_NET_SWITCHDEV
__u8 offload_fwd_mark:1;
#endif
- /* 2, 4 or 5 bit hole */
+ /* 1, 3 or 4 bit hole */
#ifdef CONFIG_NET_SCHED
__u16 tc_index; /* traffic control index */
diff --git a/include/net/sock.h b/include/net/sock.h
index e83bb01..bd63d4d 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1818,6 +1818,20 @@ static inline void sk_dst_confirm(struct sock *sk)
sk->sk_dst_pending_confirm = 1;
}
+static inline void sock_confirm_neigh(struct sk_buff *skb, struct neighbour *n)
+{
+ if (unlikely(skb->dst_pending_confirm)) {
+ struct sock *sk = skb->sk;
+ unsigned long now = jiffies;
+
+ /* avoid dirtying neighbour */
+ if (n->confirmed != now)
+ n->confirmed = now;
+ if (sk && sk->sk_dst_pending_confirm)
+ sk->sk_dst_pending_confirm = 0;
+ }
+}
+
bool sk_mc_loop(struct sock *sk);
static inline bool sk_can_gso(const struct sock *sk)
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 6c9615c..fbe63cc 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -222,7 +222,10 @@ static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *s
if (unlikely(!neigh))
neigh = __neigh_create(&arp_tbl, &nexthop, dev, false);
if (!IS_ERR(neigh)) {
- int res = dst_neigh_output(dst, neigh, skb);
+ int res;
+
+ sock_confirm_neigh(skb, neigh);
+ res = dst_neigh_output(dst, neigh, skb);
rcu_read_unlock_bh();
return res;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 70d0de40..285aa9f 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -119,6 +119,7 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff *
if (unlikely(!neigh))
neigh = __neigh_create(&nd_tbl, nexthop, dst->dev, false);
if (!IS_ERR(neigh)) {
+ sock_confirm_neigh(skb, neigh);
ret = dst_neigh_output(dst, neigh, skb);
rcu_read_unlock_bh();
return ret;
--
1.9.3
^ permalink raw reply related
* [PATCH RFC net-next 5/7] net: add confirm_neigh method to dst_ops
From: Julian Anastasov @ 2016-12-18 20:56 UTC (permalink / raw)
To: netdev; +Cc: linux-sctp, YueHaibing
In-Reply-To: <1482094596-26046-1-git-send-email-ja@ssi.bg>
Add confirm_neigh method to dst_ops and use it from IPv4 and IPv6
to lookup and confirm the neighbour. Its usage via the new helper
dst_confirm_neigh() should be restricted to MSG_PROBE users for
performance reasons.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
---
include/net/arp.h | 16 ++++++++++++++++
include/net/dst.h | 7 +++++++
include/net/dst_ops.h | 2 ++
include/net/ndisc.h | 17 +++++++++++++++++
net/ipv4/route.c | 19 +++++++++++++++++++
net/ipv6/route.c | 16 ++++++++++++++++
net/xfrm/xfrm_policy.c | 16 ++++++++++++++++
7 files changed, 93 insertions(+)
diff --git a/include/net/arp.h b/include/net/arp.h
index 5e0f891..65619a2 100644
--- a/include/net/arp.h
+++ b/include/net/arp.h
@@ -35,6 +35,22 @@ static inline struct neighbour *__ipv4_neigh_lookup(struct net_device *dev, u32
return n;
}
+static inline void __ipv4_confirm_neigh(struct net_device *dev, u32 key)
+{
+ struct neighbour *n;
+
+ rcu_read_lock_bh();
+ n = __ipv4_neigh_lookup_noref(dev, key);
+ if (n) {
+ unsigned long now = jiffies;
+
+ /* avoid dirtying neighbour */
+ if (n->confirmed != now)
+ n->confirmed = now;
+ }
+ rcu_read_unlock_bh();
+}
+
void arp_init(void);
int arp_ioctl(struct net *net, unsigned int cmd, void __user *arg);
void arp_send(int type, int ptype, __be32 dest_ip,
diff --git a/include/net/dst.h b/include/net/dst.h
index 6835d22..3a3b34b 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -477,6 +477,13 @@ static inline struct neighbour *dst_neigh_lookup_skb(const struct dst_entry *dst
return IS_ERR(n) ? NULL : n;
}
+static inline void dst_confirm_neigh(const struct dst_entry *dst,
+ const void *daddr)
+{
+ if (dst->ops->confirm_neigh)
+ dst->ops->confirm_neigh(dst, daddr);
+}
+
static inline void dst_link_failure(struct sk_buff *skb)
{
struct dst_entry *dst = skb_dst(skb);
diff --git a/include/net/dst_ops.h b/include/net/dst_ops.h
index a0d443c..fc1bdb4 100644
--- a/include/net/dst_ops.h
+++ b/include/net/dst_ops.h
@@ -33,6 +33,8 @@ struct dst_ops {
struct neighbour * (*neigh_lookup)(const struct dst_entry *dst,
struct sk_buff *skb,
const void *daddr);
+ void (*confirm_neigh)(const struct dst_entry *dst,
+ const void *daddr);
struct kmem_cache *kmem_cachep;
diff --git a/include/net/ndisc.h b/include/net/ndisc.h
index d562a2f..8a02146 100644
--- a/include/net/ndisc.h
+++ b/include/net/ndisc.h
@@ -391,6 +391,23 @@ static inline struct neighbour *__ipv6_neigh_lookup(struct net_device *dev, cons
return n;
}
+static inline void __ipv6_confirm_neigh(struct net_device *dev,
+ const void *pkey)
+{
+ struct neighbour *n;
+
+ rcu_read_lock_bh();
+ n = __ipv6_neigh_lookup_noref(dev, pkey);
+ if (n) {
+ unsigned long now = jiffies;
+
+ /* avoid dirtying neighbour */
+ if (n->confirmed != now)
+ n->confirmed = now;
+ }
+ rcu_read_unlock_bh();
+}
+
int ndisc_init(void);
int ndisc_late_init(void);
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index fa5c037..682ff24 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -154,6 +154,7 @@ static u32 *ipv4_cow_metrics(struct dst_entry *dst, unsigned long old)
static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst,
struct sk_buff *skb,
const void *daddr);
+static void ipv4_confirm_neigh(const struct dst_entry *dst, const void *daddr);
static struct dst_ops ipv4_dst_ops = {
.family = AF_INET,
@@ -168,6 +169,7 @@ static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst,
.redirect = ip_do_redirect,
.local_out = __ip_local_out,
.neigh_lookup = ipv4_neigh_lookup,
+ .confirm_neigh = ipv4_confirm_neigh,
};
#define ECN_OR_COST(class) TC_PRIO_##class
@@ -461,6 +463,23 @@ static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst,
return neigh_create(&arp_tbl, pkey, dev);
}
+static void ipv4_confirm_neigh(const struct dst_entry *dst, const void *daddr)
+{
+ struct net_device *dev = dst->dev;
+ const __be32 *pkey = daddr;
+ const struct rtable *rt;
+
+ rt = (const struct rtable *)dst;
+ if (rt->rt_gateway)
+ pkey = (const __be32 *)&rt->rt_gateway;
+ else if (!daddr ||
+ (rt->rt_flags &
+ (RTCF_MULTICAST | RTCF_BROADCAST | RTCF_LOCAL)))
+ return;
+
+ __ipv4_confirm_neigh(dev, *(__force u32 *)pkey);
+}
+
#define IP_IDENTS_SZ 2048u
static atomic_t *ip_idents __read_mostly;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 890acac..4056d0e 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -217,6 +217,21 @@ static struct neighbour *ip6_neigh_lookup(const struct dst_entry *dst,
return neigh_create(&nd_tbl, daddr, dst->dev);
}
+static void ip6_confirm_neigh(const struct dst_entry *dst, const void *daddr)
+{
+ struct net_device *dev = dst->dev;
+ struct rt6_info *rt = (struct rt6_info *)dst;
+
+ daddr = choose_neigh_daddr(rt, NULL, daddr);
+ if (!daddr)
+ return;
+ if (dev->flags & (IFF_NOARP | IFF_LOOPBACK))
+ return;
+ if (ipv6_addr_is_multicast((const struct in6_addr *)daddr))
+ return;
+ __ipv6_confirm_neigh(dev, daddr);
+}
+
static struct dst_ops ip6_dst_ops_template = {
.family = AF_INET6,
.gc = ip6_dst_gc,
@@ -233,6 +248,7 @@ static struct neighbour *ip6_neigh_lookup(const struct dst_entry *dst,
.redirect = rt6_do_redirect,
.local_out = __ip6_local_out,
.neigh_lookup = ip6_neigh_lookup,
+ .confirm_neigh = ip6_confirm_neigh,
};
static unsigned int ip6_blackhole_mtu(const struct dst_entry *dst)
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 177e208..c010ee0 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -2856,6 +2856,20 @@ static struct neighbour *xfrm_neigh_lookup(const struct dst_entry *dst,
return dst->path->ops->neigh_lookup(dst, skb, daddr);
}
+static void xfrm_confirm_neigh(const struct dst_entry *dst, const void *daddr)
+{
+ const struct dst_entry *path = dst->path;
+
+ if (path == dst) {
+ dst->ops->confirm_neigh(dst, daddr);
+ } else {
+ /* daddr can be from different family and we need the
+ * tunnel address. How to get it?
+ */
+ path->ops->confirm_neigh(path, NULL);
+ }
+}
+
int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo)
{
int err = 0;
@@ -2882,6 +2896,8 @@ int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo)
dst_ops->link_failure = xfrm_link_failure;
if (likely(dst_ops->neigh_lookup == NULL))
dst_ops->neigh_lookup = xfrm_neigh_lookup;
+ if (likely(!dst_ops->confirm_neigh))
+ dst_ops->confirm_neigh = xfrm_confirm_neigh;
if (likely(afinfo->garbage_collect == NULL))
afinfo->garbage_collect = xfrm_garbage_collect_deferred;
rcu_assign_pointer(xfrm_policy_afinfo[afinfo->family], afinfo);
--
1.9.3
^ permalink raw reply related
* [PATCH] stmmac: fix memory barriers
From: Pavel Machek @ 2016-12-18 20:38 UTC (permalink / raw)
To: LinoSanfilippo, peppe.cavallaro, alexandre.torgue, davem,
linux-kernel, netdev, niklas.cassel, Joao.Pinto
[-- Attachment #1: Type: text/plain, Size: 3034 bytes --]
Fix up memory barriers in stmmac driver. They are meant to protect
against DMA engine, so smp_ variants are certainly wrong, and dma_
variants are preferable.
Signed-off-by: Pavel Machek <pavel@denx.de>
diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c
index a340fc8..8816515 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_descs.c
@@ -334,7 +334,7 @@ static void dwmac4_rd_prepare_tx_desc(struct dma_desc *p, int is_fs, int len,
* descriptors for the same frame has to be set before, to
* avoid race condition.
*/
- wmb();
+ dma_wmb();
p->des3 = cpu_to_le32(tdes3);
}
@@ -377,7 +377,7 @@ static void dwmac4_rd_prepare_tso_tx_desc(struct dma_desc *p, int is_fs,
* descriptors for the same frame has to be set before, to
* avoid race condition.
*/
- wmb();
+ dma_wmb();
p->des3 = cpu_to_le32(tdes3);
}
diff --git a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
index ce97e52..f0d8632 100644
--- a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
@@ -350,7 +350,7 @@ static void enh_desc_prepare_tx_desc(struct dma_desc *p, int is_fs, int len,
* descriptors for the same frame has to be set before, to
* avoid race condition.
*/
- wmb();
+ dma_wmb();
p->des0 = cpu_to_le32(tdes0);
}
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 3e40578..bb40382 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -2125,7 +2125,7 @@ static netdev_tx_t stmmac_tso_xmit(struct sk_buff *skb, struct net_device *dev)
* descriptor and then barrier is needed to make sure that
* all is coherent before granting the DMA engine.
*/
- smp_wmb();
+ dma_wmb();
if (netif_msg_pktdata(priv)) {
pr_info("%s: curr=%d dirty=%d f=%d, e=%d, f_p=%p, nfrags %d\n",
@@ -2338,7 +2338,7 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
* descriptor and then barrier is needed to make sure that
* all is coherent before granting the DMA engine.
*/
- smp_wmb();
+ dma_wmb();
}
netdev_sent_queue(dev, skb->len);
@@ -2443,14 +2443,14 @@ static inline void stmmac_rx_refill(struct stmmac_priv *priv)
netif_dbg(priv, rx_status, priv->dev,
"refill entry #%d\n", entry);
}
- wmb();
+ dma_wmb();
if (unlikely(priv->synopsys_id >= DWMAC_CORE_4_00))
priv->hw->desc->init_rx_desc(p, priv->use_riwt, 0, 0);
else
priv->hw->desc->set_rx_owner(p);
- wmb();
+ dma_wmb();
entry = STMMAC_GET_ENTRY(entry, DMA_RX_SIZE);
}
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox