* [PATCH 1/4] batman-adv: fix DAT candidate selection (must use vid)
From: Antonio Quartulli @ 2016-04-30 17:06 UTC (permalink / raw)
To: davem; +Cc: netdev, b.a.t.m.a.n, Antonio Quartulli, Sven Eckelmann,
Marek Lindner
In-Reply-To: <1462035963-28970-1-git-send-email-a@unstable.cc>
Now that DAT is VLAN aware, it must use the VID when
computing the DHT address of the candidate nodes where
an entry is going to be stored/retrieved.
Fixes: be1db4f6615b ("batman-adv: make the Distributed ARP Table vlan aware")
Signed-off-by: Antonio Quartulli <a@unstable.cc>
[sven@narfation.org: fix conflicts with current version]
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
---
net/batman-adv/distributed-arp-table.c | 17 ++++++++++-------
1 file changed, 10 insertions(+), 7 deletions(-)
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index e96d7c745b4a..3e6b2624f980 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -568,6 +568,7 @@ static void batadv_choose_next_candidate(struct batadv_priv *bat_priv,
* be sent to
* @bat_priv: the bat priv with all the soft interface information
* @ip_dst: ipv4 to look up in the DHT
+ * @vid: VLAN identifier
*
* An originator O is selected if and only if its DHT_ID value is one of three
* closest values (from the LEFT, with wrap around if needed) then the hash
@@ -576,7 +577,8 @@ static void batadv_choose_next_candidate(struct batadv_priv *bat_priv,
* Return: the candidate array of size BATADV_DAT_CANDIDATE_NUM.
*/
static struct batadv_dat_candidate *
-batadv_dat_select_candidates(struct batadv_priv *bat_priv, __be32 ip_dst)
+batadv_dat_select_candidates(struct batadv_priv *bat_priv, __be32 ip_dst,
+ unsigned short vid)
{
int select;
batadv_dat_addr_t last_max = BATADV_DAT_ADDR_MAX, ip_key;
@@ -592,7 +594,7 @@ batadv_dat_select_candidates(struct batadv_priv *bat_priv, __be32 ip_dst)
return NULL;
dat.ip = ip_dst;
- dat.vid = 0;
+ dat.vid = vid;
ip_key = (batadv_dat_addr_t)batadv_hash_dat(&dat,
BATADV_DAT_ADDR_MAX);
@@ -612,6 +614,7 @@ batadv_dat_select_candidates(struct batadv_priv *bat_priv, __be32 ip_dst)
* @bat_priv: the bat priv with all the soft interface information
* @skb: payload to send
* @ip: the DHT key
+ * @vid: VLAN identifier
* @packet_subtype: unicast4addr packet subtype to use
*
* This function copies the skb with pskb_copy() and is sent as unicast packet
@@ -622,7 +625,7 @@ batadv_dat_select_candidates(struct batadv_priv *bat_priv, __be32 ip_dst)
*/
static bool batadv_dat_send_data(struct batadv_priv *bat_priv,
struct sk_buff *skb, __be32 ip,
- int packet_subtype)
+ unsigned short vid, int packet_subtype)
{
int i;
bool ret = false;
@@ -631,7 +634,7 @@ static bool batadv_dat_send_data(struct batadv_priv *bat_priv,
struct sk_buff *tmp_skb;
struct batadv_dat_candidate *cand;
- cand = batadv_dat_select_candidates(bat_priv, ip);
+ cand = batadv_dat_select_candidates(bat_priv, ip, vid);
if (!cand)
goto out;
@@ -1022,7 +1025,7 @@ bool batadv_dat_snoop_outgoing_arp_request(struct batadv_priv *bat_priv,
ret = true;
} else {
/* Send the request to the DHT */
- ret = batadv_dat_send_data(bat_priv, skb, ip_dst,
+ ret = batadv_dat_send_data(bat_priv, skb, ip_dst, vid,
BATADV_P_DAT_DHT_GET);
}
out:
@@ -1150,8 +1153,8 @@ void batadv_dat_snoop_outgoing_arp_reply(struct batadv_priv *bat_priv,
/* Send the ARP reply to the candidates for both the IP addresses that
* the node obtained from the ARP reply
*/
- batadv_dat_send_data(bat_priv, skb, ip_src, BATADV_P_DAT_DHT_PUT);
- batadv_dat_send_data(bat_priv, skb, ip_dst, BATADV_P_DAT_DHT_PUT);
+ batadv_dat_send_data(bat_priv, skb, ip_src, vid, BATADV_P_DAT_DHT_PUT);
+ batadv_dat_send_data(bat_priv, skb, ip_dst, vid, BATADV_P_DAT_DHT_PUT);
}
/**
--
2.8.1
^ permalink raw reply related
* [PATCH 4/4] batman-adv: Fix reference counting of hardif_neigh_node object for neigh_node
From: Antonio Quartulli @ 2016-04-30 17:06 UTC (permalink / raw)
To: davem; +Cc: netdev, b.a.t.m.a.n, Sven Eckelmann, Marek Lindner,
Antonio Quartulli
In-Reply-To: <1462035963-28970-1-git-send-email-a@unstable.cc>
From: Sven Eckelmann <sven@narfation.org>
The batadv_neigh_node was specific to a batadv_hardif_neigh_node and held
an implicit reference to it. But this reference was never stored in form of
a pointer in the batadv_neigh_node itself. Instead
batadv_neigh_node_release depends on a consistent state of
hard_iface->neigh_list and that batadv_hardif_neigh_get always returns the
batadv_hardif_neigh_node object which it has a reference for. But
batadv_hardif_neigh_get cannot guarantee that because it is working only
with rcu_read_lock on this list. It can therefore happen that a neigh_addr
is in this list twice or that batadv_hardif_neigh_get cannot find the
batadv_hardif_neigh_node for an neigh_addr due to some other list
operations taking place at the same time.
Instead add a batadv_hardif_neigh_node pointer directly in
batadv_neigh_node which will be used for the reference counter decremented
on release of batadv_neigh_node.
Fixes: cef63419f7db ("batman-adv: add list of unique single hop neighbors per hard-interface")
Signed-off-by: Sven Eckelmann <sven@narfation.org>
Signed-off-by: Marek Lindner <mareklindner@neomailbox.ch>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
---
net/batman-adv/originator.c | 16 +++++-----------
net/batman-adv/types.h | 2 ++
2 files changed, 7 insertions(+), 11 deletions(-)
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index d52f67a0c057..c355a824713c 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -250,7 +250,6 @@ static void batadv_neigh_node_release(struct kref *ref)
{
struct hlist_node *node_tmp;
struct batadv_neigh_node *neigh_node;
- struct batadv_hardif_neigh_node *hardif_neigh;
struct batadv_neigh_ifinfo *neigh_ifinfo;
struct batadv_algo_ops *bao;
@@ -262,13 +261,7 @@ static void batadv_neigh_node_release(struct kref *ref)
batadv_neigh_ifinfo_put(neigh_ifinfo);
}
- hardif_neigh = batadv_hardif_neigh_get(neigh_node->if_incoming,
- neigh_node->addr);
- if (hardif_neigh) {
- /* batadv_hardif_neigh_get() increases refcount too */
- batadv_hardif_neigh_put(hardif_neigh);
- batadv_hardif_neigh_put(hardif_neigh);
- }
+ batadv_hardif_neigh_put(neigh_node->hardif_neigh);
if (bao->bat_neigh_free)
bao->bat_neigh_free(neigh_node);
@@ -665,6 +658,10 @@ batadv_neigh_node_new(struct batadv_orig_node *orig_node,
neigh_node->orig_node = orig_node;
neigh_node->last_seen = jiffies;
+ /* increment unique neighbor refcount */
+ kref_get(&hardif_neigh->refcount);
+ neigh_node->hardif_neigh = hardif_neigh;
+
/* extra reference for return */
kref_init(&neigh_node->refcount);
kref_get(&neigh_node->refcount);
@@ -673,9 +670,6 @@ batadv_neigh_node_new(struct batadv_orig_node *orig_node,
hlist_add_head_rcu(&neigh_node->list, &orig_node->neigh_list);
spin_unlock_bh(&orig_node->neigh_list_lock);
- /* increment unique neighbor refcount */
- kref_get(&hardif_neigh->refcount);
-
batadv_dbg(BATADV_DBG_BATMAN, orig_node->bat_priv,
"Creating new neighbor %pM for orig_node %pM on interface %s\n",
neigh_addr, orig_node->orig, hard_iface->net_dev->name);
diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h
index 65afd090ab3e..1e47fbe8bb7b 100644
--- a/net/batman-adv/types.h
+++ b/net/batman-adv/types.h
@@ -433,6 +433,7 @@ struct batadv_hardif_neigh_node {
* @ifinfo_lock: lock protecting private ifinfo members and list
* @if_incoming: pointer to incoming hard-interface
* @last_seen: when last packet via this neighbor was received
+ * @hardif_neigh: hardif_neigh of this neighbor
* @refcount: number of contexts the object is used
* @rcu: struct used for freeing in an RCU-safe manner
*/
@@ -444,6 +445,7 @@ struct batadv_neigh_node {
spinlock_t ifinfo_lock; /* protects ifinfo_list and its members */
struct batadv_hard_iface *if_incoming;
unsigned long last_seen;
+ struct batadv_hardif_neigh_node *hardif_neigh;
struct kref refcount;
struct rcu_head rcu;
};
--
2.8.1
^ permalink raw reply related
* [PATCH net-next] net: relax expensive skb_unclone() in iptunnel_handle_offloads()
From: Eric Dumazet @ 2016-04-30 17:19 UTC (permalink / raw)
To: David Miller; +Cc: netdev
From: Eric Dumazet <edumazet@google.com>
Locally generated TCP GSO packets having to go through a GRE/SIT/IPIP
tunnel have to go through an expensive skb_unclone()
Reallocating skb->head is a lot of work.
Test should really check if a 'real clone' of the packet was done.
TCP does not care if the original gso_type is changed while the packet
travels in the stack.
This adds skb_header_unclone() which is a variant of skb_clone()
using skb_header_cloned() check instead of skb_cloned().
This variant can probably be used from other points.
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
include/linux/skbuff.h | 10 ++++++++++
net/ipv4/ip_tunnel_core.c | 2 +-
2 files changed, 11 insertions(+), 1 deletion(-)
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index c84a5a1078c5528bf6fc84573f63f3c6f470ce8f..c413c588a24f854be9e4df78d8a6872b6b1ff9f3 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1325,6 +1325,16 @@ static inline int skb_header_cloned(const struct sk_buff *skb)
return dataref != 1;
}
+static inline int skb_header_unclone(struct sk_buff *skb, gfp_t pri)
+{
+ might_sleep_if(gfpflags_allow_blocking(pri));
+
+ if (skb_header_cloned(skb))
+ return pskb_expand_head(skb, 0, 0, pri);
+
+ return 0;
+}
+
/**
* skb_header_release - release reference to header
* @skb: buffer to operate on
diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c
index 786fa7ca28e0db00865cdf9fab472836866b7d6d..9118b0e640ba3a96850cf28aa8caab4e615254d9 100644
--- a/net/ipv4/ip_tunnel_core.c
+++ b/net/ipv4/ip_tunnel_core.c
@@ -157,7 +157,7 @@ int iptunnel_handle_offloads(struct sk_buff *skb,
}
if (skb_is_gso(skb)) {
- err = skb_unclone(skb, GFP_ATOMIC);
+ err = skb_header_unclone(skb, GFP_ATOMIC);
if (unlikely(err))
return err;
skb_shinfo(skb)->gso_type |= gso_type_mask;
^ permalink raw reply related
* Re: [PATCH 3.2 085/115] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good.
From: Ben Hutchings @ 2016-04-30 18:33 UTC (permalink / raw)
To: Sabrina Dubroca, Ben Greear
Cc: Hannes Frederic Sowa, linux-kernel, stable, akpm, David S. Miller,
Vijay Pandurangan, Cong Wang, netdev, Evan Jones, Nicolas Dichtel,
Phil Sutter, Toshiaki Makita, xiyou.wangcong
In-Reply-To: <20160428102953.GA7656@bistromath.localdomain>
[-- Attachment #1: Type: text/plain, Size: 1967 bytes --]
On Thu, 2016-04-28 at 12:29 +0200, Sabrina Dubroca wrote:
> Hello,
>
> 2016-04-27, 17:14:44 -0700, Ben Greear wrote:
> >
> > On 04/27/2016 05:00 PM, Hannes Frederic Sowa wrote:
> > >
> > > Hi Ben,
> > >
> > > On Wed, Apr 27, 2016, at 20:07, Ben Hutchings wrote:
> > > >
> > > > On Wed, 2016-04-27 at 08:59 -0700, Ben Greear wrote:
> > > > >
> > > > > On 04/26/2016 04:02 PM, Ben Hutchings wrote:
> > > > > >
> > > > > >
> > > > > > 3.2.80-rc1 review patch. If anyone has any objections, please let me know.
> > > > > I would be careful about this. It causes regressions when sending
> > > > > PACKET_SOCKET buffers from user-space to veth devices.
> > > > >
> > > > > There was a proposed upstream fix for the regression, but it has not gone
> > > > > into the tree as far as I know.
> > > > >
> > > > > http://www.spinics.net/lists/netdev/msg370436.html
> > > > [...]
> > > >
> > > > OK, I'll drop this for now.
> > > The fall out from not having this patch is in my opinion a bigger
> > > fallout than not having this patch. This patch fixes silent data
> > > corruption vs. the problem Ben Greear is talking about, which might not
> > > be that a common usage.
> > >
> > > What do others think?
> > >
> > > Bye,
> > > Hannes
> > >
> > This patch from Cong Wang seems to fix the regression for me, I think it should be added and
> > tested in the main tree, and then apply them to stable as a pair.
> >
> > http://dmz2.candelatech.com/?p=linux-4.4.dev.y/.git;a=commitdiff;h=8153e983c0e5eba1aafe1fc296248ed2a553f1ac;hp=454b07405d694dad52e7f41af5816eed0190da8a
> Actually, no, this is not really a regression.
[...]
It really is. Even though the old behaviour was a bug (raw packets
should not be changed), if there are real applications that depend on
that then we have to keep those applications working somehow.
Ben.
--
Ben Hutchings
Tomorrow will be cancelled due to lack of interest.
[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply
* Re: [PATCH 3.2 085/115] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good.
From: Ben Hutchings @ 2016-04-30 19:18 UTC (permalink / raw)
To: Ben Greear, Sabrina Dubroca
Cc: Hannes Frederic Sowa, linux-kernel, stable, akpm, David S. Miller,
Vijay Pandurangan, Cong Wang, netdev, Evan Jones, Nicolas Dichtel,
Phil Sutter, Toshiaki Makita, xiyou.wangcong
In-Reply-To: <5722140E.80702@candelatech.com>
[-- Attachment #1: Type: text/plain, Size: 1079 bytes --]
On Thu, 2016-04-28 at 06:45 -0700, Ben Greear wrote:
>
> On 04/28/2016 03:29 AM, Sabrina Dubroca wrote:
[...]
> > Your use case is invalid, it just happened to work because of a
> > bug. If you want the stack to fill checksums so that you want capture
> > and reinject packets, you have to disable checksum offloading (or
> > compute the checksum yourself in userspace).
> Disabling checksum offloading or computing in user-space (and then
> recomputing in veth to verify the checksum?) is a huge performance loss.
>
> Maybe we could add a socket option to enable Cong's patch on a per-socket
> basis? That way my use-case can still work and you can have this new
> behaviour by default?
It does sound like a useful option to have. If there are other
applications that depend on veth's checksum-fixing behaviour and are
being distributed in binary form, then a per-device option might be
necessary so users can keep those applications working before they're
updated.
Ben.
--
Ben Hutchings
Tomorrow will be cancelled due to lack of interest.
[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply
* __napi_alloc_skb failures locking up the box
From: Aaro Koskinen @ 2016-04-30 19:24 UTC (permalink / raw)
To: David S. Miller, netdev, linux-mm; +Cc: linux-kernel
Hi,
I have old NAS box (Thecus N2100) with 512 MB RAM, where rsync from NFS ->
disk reliably results in temporary out-of-memory conditions.
When this happens the dmesg gets flooded with below logs. If the serial
console logging is enabled, this will lock up the box completely and
the backup is not making any progress.
Shouldn't these allocation failures be ratelimited somehow (or even made
silent)? It doesn't sound right if I can lock up the system simply by
copying files...
...
[ 1706.105842] kworker/0:1H: page allocation failure: order:0, mode:0x2200020(GFP_NOWAIT|__GFP_HIGH|__GFP_NOTRACK)
[ 1706.105904] CPU: 0 PID: 519 Comm: kworker/0:1H Not tainted 4.6.0-rc5-iop32x-los_a50bb #1
[ 1706.105917] Hardware name: Thecus N2100
[ 1706.105973] Workqueue: rpciod rpc_async_schedule
[ 1706.105993] Backtrace:
[ 1706.106037] [<c000d460>] (dump_backtrace) from [<c000d658>] (show_stack+0x18/0x1c)
[ 1706.106068] r7:00000000 r6:00000060 r5:00000000 r4:00000000
[ 1706.106142] [<c000d640>] (show_stack) from [<c01c1e84>] (dump_stack+0x20/0x28)
[ 1706.106185] [<c01c1e64>] (dump_stack) from [<c007b9b0>] (warn_alloc_failed+0xf0/0x134)
[ 1706.106216] [<c007b8c4>] (warn_alloc_failed) from [<c007df78>] (__alloc_pages_nodemask+0x284/0x8fc)
[ 1706.106233] r3:02200020 r2:00000000
[ 1706.106269] r5:00000000 r4:c0524c40
[ 1706.106336] [<c007dcf4>] (__alloc_pages_nodemask) from [<c00b1948>] (new_slab+0x3c4/0x430)
[ 1706.106352] r10:00000000 r9:c0321fdc r8:00000000 r7:df401d00 r6:00000015 r5:02000020
[ 1706.106422] r4:df401d00
[ 1706.106460] [<c00b1584>] (new_slab) from [<c00b2bd8>] (___slab_alloc.constprop.8+0x238/0x298)
[ 1706.106477] r10:00000000 r9:c0321fdc r8:02080020 r7:df401d00 r6:dfbef060 r5:00000000
[ 1706.106547] r4:00000000
[ 1706.106581] [<c00b29a0>] (___slab_alloc.constprop.8) from [<c00b2f68>] (kmem_cache_alloc+0xbc/0xf8)
[ 1706.106597] r10:00167e93 r9:340285ee r8:e11e1920 r7:60000013 r6:02080020 r5:df401d00
[ 1706.106668] r4:00000000
[ 1706.106707] [<c00b2eac>] (kmem_cache_alloc) from [<c0321fdc>] (__build_skb+0x2c/0x98)
[ 1706.106723] r7:cc592240 r6:000006e0 r5:de3523f0 r4:000006e0
[ 1706.106784] [<c0321fb0>] (__build_skb) from [<c032226c>] (__napi_alloc_skb+0xb0/0xfc)
[ 1706.106801] r9:340285ee r8:e11e1920 r7:cc592240 r6:c0520ab8 r5:de3523f0 r4:000006e0
[ 1706.106903] [<c03221bc>] (__napi_alloc_skb) from [<c028a2fc>] (rtl8169_poll+0x3a0/0x588)
[ 1706.106920] r7:de364000 r6:c000fd08 r5:000005ea r4:de3523f0
[ 1706.106986] [<c0289f5c>] (rtl8169_poll) from [<c032d098>] (net_rx_action+0x1cc/0x2ec)
[ 1706.107002] r10:00022548 r9:df471ba8 r8:c05257e0 r7:0000012c r6:00000040 r5:00000001
[ 1706.107073] r4:de3523f0
[ 1706.107131] [<c032cecc>] (net_rx_action) from [<c001bd68>] (__do_softirq+0xf4/0x254)
[ 1706.107147] r10:00000101 r9:c052618c r8:40000001 r7:c0526188 r6:df470000 r5:00000003
[ 1706.107218] r4:00000000
[ 1706.107255] [<c001bc74>] (__do_softirq) from [<c001bf58>] (do_softirq.part.2+0x34/0x40)
[ 1706.107271] r10:00000001 r9:c0525a26 r8:c05555c0 r7:00000000 r6:deae8a80 r5:00000000
[ 1706.107341] r4:20000013
[ 1706.107380] [<c001bf24>] (do_softirq.part.2) from [<c001c010>] (__local_bh_enable_ip+0xac/0xcc)
[ 1706.107396] r5:00000000 r4:00000200
[ 1706.107459] [<c001bf64>] (__local_bh_enable_ip) from [<c0319538>] (release_sock+0x12c/0x158)
[ 1706.107477] r5:00000000 r4:00000000
[ 1706.107538] [<c031940c>] (release_sock) from [<c03671b4>] (tcp_sendmsg+0xf8/0xa90)
[ 1706.107557] r10:00004040 r9:deae8a80 r8:df471d74 r7:88cd7146 r6:0000059c r5:de8ccb00
[ 1706.107629] r4:0000007c r3:00000001
[ 1706.107678] [<c03670bc>] (tcp_sendmsg) from [<c038d7f0>] (inet_sendmsg+0x3c/0x74)
[ 1706.107695] r10:df471e2c r9:00000000 r8:df7eb604 r7:00000000 r6:00000000 r5:df02c780
[ 1706.107766] r4:deae8a80
[ 1706.107800] [<c038d7b4>] (inet_sendmsg) from [<c03157f0>] (sock_sendmsg+0x1c/0x30)
[ 1706.107816] r4:df471d74
[ 1706.107849] [<c03157d4>] (sock_sendmsg) from [<c03158f4>] (kernel_sendmsg+0x38/0x40)
[ 1706.107873] [<c03158bc>] (kernel_sendmsg) from [<c03afa2c>] (xs_send_kvec+0x94/0x9c)
[ 1706.107891] r5:00000000 r4:df02c780
[ 1706.107934] [<c03af998>] (xs_send_kvec) from [<c03afaa0>] (xs_sendpages+0x6c/0x244)
[ 1706.107950] r9:00000000 r8:df02c780 r7:0000007c r6:df7eb604 r5:00000001 r4:00000000
[ 1706.108026] [<c03afa34>] (xs_sendpages) from [<c03afd7c>] (xs_tcp_send_request+0x80/0x134)
[ 1706.108043] r10:00000000 r9:00000000 r8:de8e8000 r7:de0d7258 r6:df7eb604 r5:00000001
[ 1706.108114] r4:df7eb600
[ 1706.108172] [<c03afcfc>] (xs_tcp_send_request) from [<c03ad670>] (xprt_transmit+0x58/0x214)
[ 1706.108192] r10:de92cc60 r9:00000000 r8:d3697fdd r7:df7eb674 r6:de0d7258 r5:df7eb600
[ 1706.108265] r4:de8e8000
[ 1706.108303] [<c03ad618>] (xprt_transmit) from [<c03aac74>] (call_transmit+0x18c/0x230)
[ 1706.108321] r7:df7eb600 r6:00000001 r5:df7eb600 r4:de0d7258
[ 1706.108388] [<c03aaae8>] (call_transmit) from [<c03b1ed8>] (__rpc_execute+0x54/0x2c4)
[ 1706.108404] r8:c0508940 r7:00000000 r6:c03b1f08 r5:00000001 r4:de0d7258
[ 1706.108475] [<c03b1e84>] (__rpc_execute) from [<c03b215c>] (rpc_async_schedule+0x14/0x18)
[ 1706.108493] r7:00000000 r6:dfbf0800 r5:de92cc60 r4:de0d727c
[ 1706.108564] [<c03b2148>] (rpc_async_schedule) from [<c002c79c>] (process_one_work+0x130/0x3ec)
[ 1706.108587] [<c002c66c>] (process_one_work) from [<c002caa8>] (worker_thread+0x50/0x5ac)
[ 1706.108604] r10:de92cc60 r9:c0508940 r8:00000008 r7:c050b220 r6:c0508954 r5:de92cc78
[ 1706.108675] r4:c0508940
[ 1706.108718] [<c002ca58>] (worker_thread) from [<c00319cc>] (kthread+0xc8/0xe4)
[ 1706.108734] r10:00000000 r9:00000000 r8:00000000 r7:c002ca58 r6:de92cc60 r5:00000000
[ 1706.108804] r4:de385880
[ 1706.108844] [<c0031904>] (kthread) from [<c000a390>] (ret_from_fork+0x14/0x24)
[ 1706.108861] r7:00000000 r6:00000000 r5:c0031904 r4:de385880
[ 1706.108909] Mem-Info:
[ 1706.108950] active_anon:2694 inactive_anon:2799 isolated_anon:0
[ 1706.108950] active_file:43728 inactive_file:71243 isolated_file:0
[ 1706.108950] unevictable:0 dirty:1486 writeback:0 unstable:0
[ 1706.108950] slab_reclaimable:3099 slab_unreclaimable:710
[ 1706.108950] mapped:1763 shmem:19 pagetables:186 bounce:0
[ 1706.108950] free:290 free_pcp:56 free_cma:0
[ 1706.109041] Normal free:1160kB min:2868kB low:3584kB high:4300kB active_anon:10776kB inactive_anon:11196kB active_file:174912kB inactive_file:284972kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:524288kB managed:514412kB mlocked:0kB dirty:5944kB writeback:0kB mapped:7052kB shmem:76kB slab_reclaimable:12396kB slab_unreclaimable:2840kB kernel_stack:584kB pagetables:744kB unstable:0kB bounce:0kB free_pcp:224kB local_pcp:224kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
[ 1706.109059] lowmem_reserve[]: 0 0
[ 1706.109075] Normal: 80*4kB (UME) 105*8kB (UE) 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1160kB
[ 1706.109137] 114997 total pagecache pages
[ 1706.109154] 0 pages in swap cache
[ 1706.109168] Swap cache stats: add 0, delete 0, find 0/0
[ 1706.109179] Free swap = 1048572kB
[ 1706.109189] Total swap = 1048572kB
[ 1706.109199] 131072 pages RAM
[ 1706.109208] 0 pages HighMem/MovableOnly
[ 1706.109218] 2469 pages reserved
[ 1706.109238] SLUB: Unable to allocate memory on node -1, gfp=0x2080020(GFP_ATOMIC)
[ 1706.109255] cache: kmalloc-192, object size: 192, buffer size: 192, default order: 0, min order: 0
[ 1706.109272] node 0: slabs: 12, objs: 252, free: 0
[ 1706.109320] kworker/0:1H: page allocation failure: order:0, mode:0x2200020(GFP_NOWAIT|__GFP_HIGH|__GFP_NOTRACK)
[ 1706.109341] CPU: 0 PID: 519 Comm: kworker/0:1H Not tainted 4.6.0-rc5-iop32x-los_a50bb #1
[ 1706.109354] Hardware name: Thecus N2100
[ 1706.109371] Workqueue: rpciod rpc_async_schedule
[ 1706.109381] Backtrace:
[ 1706.109412] [<c000d460>] (dump_backtrace) from [<c000d658>] (show_stack+0x18/0x1c)
[ 1706.109429] r7:00000000 r6:00000060 r5:00000000 r4:00000000
[ 1706.109500] [<c000d640>] (show_stack) from [<c01c1e84>] (dump_stack+0x20/0x28)
[ 1706.109547] [<c01c1e64>] (dump_stack) from [<c007b9b0>] (warn_alloc_failed+0xf0/0x134)
[ 1706.109578] [<c007b8c4>] (warn_alloc_failed) from [<c007df78>] (__alloc_pages_nodemask+0x284/0x8fc)
[ 1706.109595] r3:02200020 r2:00000000
[ 1706.109631] r5:00000000 r4:c0524c40
[ 1706.109696] [<c007dcf4>] (__alloc_pages_nodemask) from [<c00b1948>] (new_slab+0x3c4/0x430)
[ 1706.109715] r10:00000000 r9:c0321fdc r8:00000000 r7:df401d00 r6:00000015 r5:02000020
[ 1706.109786] r4:df401d00
[ 1706.109823] [<c00b1584>] (new_slab) from [<c00b2bd8>] (___slab_alloc.constprop.8+0x238/0x298)
[ 1706.109839] r10:00000000 r9:c0321fdc r8:02080020 r7:df401d00 r6:dfbef060 r5:00000000
[ 1706.109909] r4:00000000
[ 1706.109943] [<c00b29a0>] (___slab_alloc.constprop.8) from [<c00b2f68>] (kmem_cache_alloc+0xbc/0xf8)
[ 1706.109959] r10:00167e94 r9:340285ee r8:e11e1930 r7:60000013 r6:02080020 r5:df401d00
[ 1706.110030] r4:00000000
[ 1706.110069] [<c00b2eac>] (kmem_cache_alloc) from [<c0321fdc>] (__build_skb+0x2c/0x98)
[ 1706.110086] r7:cc593920 r6:000006e0 r5:de3523f0 r4:000006e0
[ 1706.110147] [<c0321fb0>] (__build_skb) from [<c032226c>] (__napi_alloc_skb+0xb0/0xfc)
[ 1706.110163] r9:340285ee r8:e11e1930 r7:cc593920 r6:c0520ab8 r5:de3523f0 r4:000006e0
[ 1706.110267] [<c03221bc>] (__napi_alloc_skb) from [<c028a2fc>] (rtl8169_poll+0x3a0/0x588)
[ 1706.110286] r7:de368000 r6:c000fd08 r5:000005ea r4:de3523f0
[ 1706.110352] [<c0289f5c>] (rtl8169_poll) from [<c032d098>] (net_rx_action+0x1cc/0x2ec)
[ 1706.110369] r10:00022548 r9:df471ba8 r8:c05257e0 r7:0000012c r6:00000040 r5:00000001
[ 1706.110438] r4:de3523f0
[ 1706.110499] [<c032cecc>] (net_rx_action) from [<c001bd68>] (__do_softirq+0xf4/0x254)
[ 1706.110518] r10:00000101 r9:c052618c r8:40000001 r7:c0526188 r6:df470000 r5:00000003
[ 1706.110589] r4:00000000
[ 1706.110628] [<c001bc74>] (__do_softirq) from [<c001bf58>] (do_softirq.part.2+0x34/0x40)
[ 1706.110644] r10:00000001 r9:c0525a26 r8:c05555c0 r7:00000000 r6:deae8a80 r5:00000000
[ 1706.110714] r4:20000013
[ 1706.110754] [<c001bf24>] (do_softirq.part.2) from [<c001c010>] (__local_bh_enable_ip+0xac/0xcc)
[ 1706.110770] r5:00000000 r4:00000200
[ 1706.110833] [<c001bf64>] (__local_bh_enable_ip) from [<c0319538>] (release_sock+0x12c/0x158)
[ 1706.110849] r5:00000000 r4:00000000
[ 1706.110906] [<c031940c>] (release_sock) from [<c03671b4>] (tcp_sendmsg+0xf8/0xa90)
[ 1706.110922] r10:00004040 r9:deae8a80 r8:df471d74 r7:88cd7146 r6:0000059c r5:de8ccb00
[ 1706.110993] r4:0000007c r3:00000001
[ 1706.111040] [<c03670bc>] (tcp_sendmsg) from [<c038d7f0>] (inet_sendmsg+0x3c/0x74)
[ 1706.111058] r10:df471e2c r9:00000000 r8:df7eb604 r7:00000000 r6:00000000 r5:df02c780
[ 1706.111127] r4:deae8a80
[ 1706.111161] [<c038d7b4>] (inet_sendmsg) from [<c03157f0>] (sock_sendmsg+0x1c/0x30)
[ 1706.111177] r4:df471d74
[ 1706.111210] [<c03157d4>] (sock_sendmsg) from [<c03158f4>] (kernel_sendmsg+0x38/0x40)
[ 1706.111236] [<c03158bc>] (kernel_sendmsg) from [<c03afa2c>] (xs_send_kvec+0x94/0x9c)
[ 1706.111253] r5:00000000 r4:df02c780
[ 1706.111295] [<c03af998>] (xs_send_kvec) from [<c03afaa0>] (xs_sendpages+0x6c/0x244)
[ 1706.111311] r9:00000000 r8:df02c780 r7:0000007c r6:df7eb604 r5:00000001 r4:00000000
[ 1706.111387] [<c03afa34>] (xs_sendpages) from [<c03afd7c>] (xs_tcp_send_request+0x80/0x134)
[ 1706.111403] r10:00000000 r9:00000000 r8:de8e8000 r7:de0d7258 r6:df7eb604 r5:00000001
[ 1706.111473] r4:df7eb600
[ 1706.111527] [<c03afcfc>] (xs_tcp_send_request) from [<c03ad670>] (xprt_transmit+0x58/0x214)
[ 1706.111549] r10:de92cc60 r9:00000000 r8:d3697fdd r7:df7eb674 r6:de0d7258 r5:df7eb600
[ 1706.111619] r4:de8e8000
[ 1706.111656] [<c03ad618>] (xprt_transmit) from [<c03aac74>] (call_transmit+0x18c/0x230)
[ 1706.111672] r7:df7eb600 r6:00000001 r5:df7eb600 r4:de0d7258
[ 1706.111740] [<c03aaae8>] (call_transmit) from [<c03b1ed8>] (__rpc_execute+0x54/0x2c4)
[ 1706.111757] r8:c0508940 r7:00000000 r6:c03b1f08 r5:00000001 r4:de0d7258
[ 1706.111828] [<c03b1e84>] (__rpc_execute) from [<c03b215c>] (rpc_async_schedule+0x14/0x18)
[ 1706.111844] r7:00000000 r6:dfbf0800 r5:de92cc60 r4:de0d727c
[ 1706.111912] [<c03b2148>] (rpc_async_schedule) from [<c002c79c>] (process_one_work+0x130/0x3ec)
[ 1706.111936] [<c002c66c>] (process_one_work) from [<c002caa8>] (worker_thread+0x50/0x5ac)
[ 1706.111953] r10:de92cc60 r9:c0508940 r8:00000008 r7:c050b220 r6:c0508954 r5:de92cc78
[ 1706.112023] r4:c0508940
[ 1706.112069] [<c002ca58>] (worker_thread) from [<c00319cc>] (kthread+0xc8/0xe4)
[ 1706.112087] r10:00000000 r9:00000000 r8:00000000 r7:c002ca58 r6:de92cc60 r5:00000000
[ 1706.112157] r4:de385880
[ 1706.112196] [<c0031904>] (kthread) from [<c000a390>] (ret_from_fork+0x14/0x24)
[ 1706.112212] r7:00000000 r6:00000000 r5:c0031904 r4:de385880
[ 1706.112261] Mem-Info:
[ 1706.112299] active_anon:2694 inactive_anon:2799 isolated_anon:0
[ 1706.112299] active_file:43728 inactive_file:71243 isolated_file:0
[ 1706.112299] unevictable:0 dirty:1486 writeback:0 unstable:0
[ 1706.112299] slab_reclaimable:3099 slab_unreclaimable:710
[ 1706.112299] mapped:1763 shmem:19 pagetables:186 bounce:0
[ 1706.112299] free:290 free_pcp:55 free_cma:0
[ 1706.112387] Normal free:1160kB min:2868kB low:3584kB high:4300kB active_anon:10776kB inactive_anon:11196kB active_file:174912kB inactive_file:284972kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:524288kB managed:514412kB mlocked:0kB dirty:5944kB writeback:0kB mapped:7052kB shmem:76kB slab_reclaimable:12396kB slab_unreclaimable:2840kB kernel_stack:584kB pagetables:744kB unstable:0kB bounce:0kB free_pcp:220kB local_pcp:220kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
[ 1706.112406] lowmem_reserve[]: 0 0
[ 1706.112420] Normal: 80*4kB (UME) 105*8kB (UE) 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1160kB
[ 1706.112481] 114997 total pagecache pages
[ 1706.112499] 0 pages in swap cache
[ 1706.112512] Swap cache stats: add 0, delete 0, find 0/0
[ 1706.112522] Free swap = 1048572kB
[ 1706.112532] Total swap = 1048572kB
[ 1706.112543] 131072 pages RAM
[ 1706.112552] 0 pages HighMem/MovableOnly
[ 1706.112561] 2469 pages reserved
[ 1706.112582] SLUB: Unable to allocate memory on node -1, gfp=0x2080020(GFP_ATOMIC)
[ 1706.112601] cache: kmalloc-192, object size: 192, buffer size: 192, default order: 0, min order: 0
[ 1706.112617] node 0: slabs: 12, objs: 252, free: 0
[ 1706.112656] kworker/0:1H: page allocation failure: order:0, mode:0x2200020(GFP_NOWAIT|__GFP_HIGH|__GFP_NOTRACK)
[ 1706.112677] CPU: 0 PID: 519 Comm: kworker/0:1H Not tainted 4.6.0-rc5-iop32x-los_a50bb #1
[ 1706.112689] Hardware name: Thecus N2100
[ 1706.112707] Workqueue: rpciod rpc_async_schedule
[ 1706.112717] Backtrace:
[ 1706.112750] [<c000d460>] (dump_backtrace) from [<c000d658>] (show_stack+0x18/0x1c)
[ 1706.112769] r7:00000000 r6:00000060 r5:00000000 r4:00000000
[ 1706.112843] [<c000d640>] (show_stack) from [<c01c1e84>] (dump_stack+0x20/0x28)
[ 1706.112884] [<c01c1e64>] (dump_stack) from [<c007b9b0>] (warn_alloc_failed+0xf0/0x134)
[ 1706.112914] [<c007b8c4>] (warn_alloc_failed) from [<c007df78>] (__alloc_pages_nodemask+0x284/0x8fc)
[ 1706.112932] r3:02200020 r2:00000000
[ 1706.112967] r5:00000000 r4:c0524c40
[ 1706.113034] [<c007dcf4>] (__alloc_pages_nodemask) from [<c00b1948>] (new_slab+0x3c4/0x430)
[ 1706.113052] r10:00000000 r9:c0321fdc r8:00000000 r7:df401d00 r6:00000015 r5:02000020
[ 1706.113122] r4:df401d00
[ 1706.113160] [<c00b1584>] (new_slab) from [<c00b2bd8>] (___slab_alloc.constprop.8+0x238/0x298)
[ 1706.113176] r10:00000000 r9:c0321fdc r8:02080020 r7:df401d00 r6:dfbef060 r5:00000000
[ 1706.113246] r4:00000000
[ 1706.113279] [<c00b29a0>] (___slab_alloc.constprop.8) from [<c00b2f68>] (kmem_cache_alloc+0xbc/0xf8)
[ 1706.113297] r10:00167e95 r9:340285ee r8:e11e1940 r7:60000013 r6:02080020 r5:df401d00
[ 1706.113367] r4:00000000
[ 1706.113407] [<c00b2eac>] (kmem_cache_alloc) from [<c0321fdc>] (__build_skb+0x2c/0x98)
[ 1706.113425] r7:cc593240 r6:000006e0 r5:de3523f0 r4:000006e0
[ 1706.113485] [<c0321fb0>] (__build_skb) from [<c032226c>] (__napi_alloc_skb+0xb0/0xfc)
[ 1706.113502] r9:340285ee r8:e11e1940 r7:cc593240 r6:c0520ab8 r5:de3523f0 r4:000006e0
[ 1706.113603] [<c03221bc>] (__napi_alloc_skb) from [<c028a2fc>] (rtl8169_poll+0x3a0/0x588)
[ 1706.113623] r7:de36c000 r6:c000fd08 r5:000005ea r4:de3523f0
[ 1706.113689] [<c0289f5c>] (rtl8169_poll) from [<c032d098>] (net_rx_action+0x1cc/0x2ec)
[ 1706.113706] r10:00022548 r9:df471ba8 r8:c05257e0 r7:0000012c r6:00000040 r5:00000001
[ 1706.113776] r4:de3523f0
[ 1706.113832] [<c032cecc>] (net_rx_action) from [<c001bd68>] (__do_softirq+0xf4/0x254)
[ 1706.113849] r10:00000101 r9:c052618c r8:40000001 r7:c0526188 r6:df470000 r5:00000003
[ 1706.113920] r4:00000000
[ 1706.113957] [<c001bc74>] (__do_softirq) from [<c001bf58>] (do_softirq.part.2+0x34/0x40)
[ 1706.113974] r10:00000001 r9:c0525a26 r8:c05555c0 r7:00000000 r6:deae8a80 r5:00000000
[ 1706.114044] r4:20000013
[ 1706.114083] [<c001bf24>] (do_softirq.part.2) from [<c001c010>] (__local_bh_enable_ip+0xac/0xcc)
[ 1706.114100] r5:00000000 r4:00000200
[ 1706.114160] [<c001bf64>] (__local_bh_enable_ip) from [<c0319538>] (release_sock+0x12c/0x158)
[ 1706.114176] r5:00000000 r4:00000000
[ 1706.114236] [<c031940c>] (release_sock) from [<c03671b4>] (tcp_sendmsg+0xf8/0xa90)
[ 1706.114255] r10:00004040 r9:deae8a80 r8:df471d74 r7:88cd7146 r6:0000059c r5:de8ccb00
[ 1706.114325] r4:0000007c r3:00000001
[ 1706.114371] [<c03670bc>] (tcp_sendmsg) from [<c038d7f0>] (inet_sendmsg+0x3c/0x74)
[ 1706.114389] r10:df471e2c r9:00000000 r8:df7eb604 r7:00000000 r6:00000000 r5:df02c780
[ 1706.114460] r4:deae8a80
[ 1706.114493] [<c038d7b4>] (inet_sendmsg) from [<c03157f0>] (sock_sendmsg+0x1c/0x30)
[ 1706.114510] r4:df471d74
[ 1706.114544] [<c03157d4>] (sock_sendmsg) from [<c03158f4>] (kernel_sendmsg+0x38/0x40)
[ 1706.114569] [<c03158bc>] (kernel_sendmsg) from [<c03afa2c>] (xs_send_kvec+0x94/0x9c)
[ 1706.114585] r5:00000000 r4:df02c780
[ 1706.114626] [<c03af998>] (xs_send_kvec) from [<c03afaa0>] (xs_sendpages+0x6c/0x244)
[ 1706.114642] r9:00000000 r8:df02c780 r7:0000007c r6:df7eb604 r5:00000001 r4:00000000
[ 1706.114718] [<c03afa34>] (xs_sendpages) from [<c03afd7c>] (xs_tcp_send_request+0x80/0x134)
[ 1706.114734] r10:00000000 r9:00000000 r8:de8e8000 r7:de0d7258 r6:df7eb604 r5:00000001
[ 1706.114805] r4:df7eb600
[ 1706.114860] [<c03afcfc>] (xs_tcp_send_request) from [<c03ad670>] (xprt_transmit+0x58/0x214)
[ 1706.114880] r10:de92cc60 r9:00000000 r8:d3697fdd r7:df7eb674 r6:de0d7258 r5:df7eb600
[ 1706.114951] r4:de8e8000
[ 1706.114990] [<c03ad618>] (xprt_transmit) from [<c03aac74>] (call_transmit+0x18c/0x230)
[ 1706.115006] r7:df7eb600 r6:00000001 r5:df7eb600 r4:de0d7258
[ 1706.115072] [<c03aaae8>] (call_transmit) from [<c03b1ed8>] (__rpc_execute+0x54/0x2c4)
[ 1706.115089] r8:c0508940 r7:00000000 r6:c03b1f08 r5:00000001 r4:de0d7258
[ 1706.115160] [<c03b1e84>] (__rpc_execute) from [<c03b215c>] (rpc_async_schedule+0x14/0x18)
[ 1706.115176] r7:00000000 r6:dfbf0800 r5:de92cc60 r4:de0d727c
[ 1706.115246] [<c03b2148>] (rpc_async_schedule) from [<c002c79c>] (process_one_work+0x130/0x3ec)
[ 1706.115271] [<c002c66c>] (process_one_work) from [<c002caa8>] (worker_thread+0x50/0x5ac)
[ 1706.115288] r10:de92cc60 r9:c0508940 r8:00000008 r7:c050b220 r6:c0508954 r5:de92cc78
[ 1706.115359] r4:c0508940
[ 1706.115404] [<c002ca58>] (worker_thread) from [<c00319cc>] (kthread+0xc8/0xe4)
[ 1706.115422] r10:00000000 r9:00000000 r8:00000000 r7:c002ca58 r6:de92cc60 r5:00000000
[ 1706.115492] r4:de385880
[ 1706.115532] [<c0031904>] (kthread) from [<c000a390>] (ret_from_fork+0x14/0x24)
[ 1706.115548] r7:00000000 r6:00000000 r5:c0031904 r4:de385880
[ 1706.115596] Mem-Info:
[ 1706.115635] active_anon:2694 inactive_anon:2799 isolated_anon:0
[ 1706.115635] active_file:43728 inactive_file:71243 isolated_file:0
[ 1706.115635] unevictable:0 dirty:1486 writeback:0 unstable:0
[ 1706.115635] slab_reclaimable:3099 slab_unreclaimable:710
[ 1706.115635] mapped:1763 shmem:19 pagetables:186 bounce:0
[ 1706.115635] free:290 free_pcp:55 free_cma:0
[ 1706.115725] Normal free:1160kB min:2868kB low:3584kB high:4300kB active_anon:10776kB inactive_anon:11196kB active_file:174912kB inactive_file:284972kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:524288kB managed:514412kB mlocked:0kB dirty:5944kB writeback:0kB mapped:7052kB shmem:76kB slab_reclaimable:12396kB slab_unreclaimable:2840kB kernel_stack:584kB pagetables:744kB unstable:0kB bounce:0kB free_pcp:220kB local_pcp:220kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
[ 1706.115744] lowmem_reserve[]: 0 0
[ 1706.115758] Normal: 80*4kB (UME) 105*8kB (UE) 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1160kB
[ 1706.115819] 114997 total pagecache pages
[ 1706.115836] 0 pages in swap cache
[ 1706.115849] Swap cache stats: add 0, delete 0, find 0/0
[ 1706.115859] Free swap = 1048572kB
[ 1706.115869] Total swap = 1048572kB
[ 1706.115879] 131072 pages RAM
[ 1706.115888] 0 pages HighMem/MovableOnly
[ 1706.115898] 2469 pages reserved
[ 1706.115917] SLUB: Unable to allocate memory on node -1, gfp=0x2080020(GFP_ATOMIC)
[ 1706.115935] cache: kmalloc-192, object size: 192, buffer size: 192, default order: 0, min order: 0
[ 1706.115950] node 0: slabs: 12, objs: 252, free: 0
[ 1706.116178] kworker/0:1H: page allocation failure: order:0, mode:0x2200020(GFP_NOWAIT|__GFP_HIGH|__GFP_NOTRACK)
[ 1706.116220] CPU: 0 PID: 519 Comm: kworker/0:1H Not tainted 4.6.0-rc5-iop32x-los_a50bb #1
[ 1706.116234] Hardware name: Thecus N2100
[ 1706.116269] Workqueue: rpciod rpc_async_schedule
[ 1706.116285] Backtrace:
[ 1706.116320] [<c000d460>] (dump_backtrace) from [<c000d658>] (show_stack+0x18/0x1c)
[ 1706.116348] r7:00000000 r6:00000060 r5:00000000 r4:00000000
[ 1706.116422] [<c000d640>] (show_stack) from [<c01c1e84>] (dump_stack+0x20/0x28)
[ 1706.116462] [<c01c1e64>] (dump_stack) from [<c007b9b0>] (warn_alloc_failed+0xf0/0x134)
[ 1706.116494] [<c007b8c4>] (warn_alloc_failed) from [<c007df78>] (__alloc_pages_nodemask+0x284/0x8fc)
[ 1706.116512] r3:02200020 r2:00000000
[ 1706.116547] r5:00000000 r4:c0524c40
[ 1706.116611] [<c007dcf4>] (__alloc_pages_nodemask) from [<c00b1948>] (new_slab+0x3c4/0x430)
[ 1706.116630] r10:00000000 r9:c0321fdc r8:00000000 r7:df401d00 r6:00000015 r5:02000020
[ 1706.116701] r4:df401d00
[ 1706.116739] [<c00b1584>] (new_slab) from [<c00b2bd8>] (___slab_alloc.constprop.8+0x238/0x298)
[ 1706.116755] r10:00000000 r9:c0321fdc r8:02080020 r7:df401d00 r6:dfbef060 r5:00000000
[ 1706.116825] r4:00000000
[ 1706.116859] [<c00b29a0>] (___slab_alloc.constprop.8) from [<c00b2f68>] (kmem_cache_alloc+0xbc/0xf8)
[ 1706.116876] r10:00167e96 r9:340285ee r8:e11e1950 r7:60000013 r6:02080020 r5:df401d00
[ 1706.116946] r4:00000000
[ 1706.116986] [<c00b2eac>] (kmem_cache_alloc) from [<c0321fdc>] (__build_skb+0x2c/0x98)
[ 1706.117003] r7:cc593920 r6:000006e0 r5:de3523f0 r4:000006e0
[ 1706.117063] [<c0321fb0>] (__build_skb) from [<c032226c>] (__napi_alloc_skb+0xb0/0xfc)
[ 1706.117079] r9:340285ee r8:e11e1950 r7:cc593920 r6:c0520ab8 r5:de3523f0 r4:000006e0
[ 1706.117182] [<c03221bc>] (__napi_alloc_skb) from [<c028a2fc>] (rtl8169_poll+0x3a0/0x588)
[ 1706.117203] r7:de370000 r6:c000fd08 r5:000005ea r4:de3523f0
[ 1706.117270] [<c0289f5c>] (rtl8169_poll) from [<c032d098>] (net_rx_action+0x1cc/0x2ec)
[ 1706.117287] r10:00022548 r9:df471ba8 r8:c05257e0 r7:0000012c r6:00000040 r5:00000001
[ 1706.117356] r4:de3523f0
[ 1706.117413] [<c032cecc>] (net_rx_action) from [<c001bd68>] (__do_softirq+0xf4/0x254)
[ 1706.117430] r10:00000101 r9:c052618c r8:40000001 r7:c0526188 r6:df470000 r5:00000003
[ 1706.117500] r4:00000000
[ 1706.117537] [<c001bc74>] (__do_softirq) from [<c001bf58>] (do_softirq.part.2+0x34/0x40)
[ 1706.117554] r10:00000001 r9:c0525a26 r8:c05555c0 r7:00000000 r6:deae8a80 r5:00000000
[ 1706.117623] r4:20000013
[ 1706.117662] [<c001bf24>] (do_softirq.part.2) from [<c001c010>] (__local_bh_enable_ip+0xac/0xcc)
[ 1706.117678] r5:00000000 r4:00000200
[ 1706.117740] [<c001bf64>] (__local_bh_enable_ip) from [<c0319538>] (release_sock+0x12c/0x158)
[ 1706.117757] r5:00000000 r4:00000000
[ 1706.117813] [<c031940c>] (release_sock) from [<c03671b4>] (tcp_sendmsg+0xf8/0xa90)
[ 1706.117831] r10:00004040 r9:deae8a80 r8:df471d74 r7:88cd7146 r6:0000059c r5:de8ccb00
[ 1706.117901] r4:0000007c r3:00000001
[ 1706.117947] [<c03670bc>] (tcp_sendmsg) from [<c038d7f0>] (inet_sendmsg+0x3c/0x74)
[ 1706.117965] r10:df471e2c r9:00000000 r8:df7eb604 r7:00000000 r6:00000000 r5:df02c780
[ 1706.118035] r4:deae8a80
[ 1706.118070] [<c038d7b4>] (inet_sendmsg) from [<c03157f0>] (sock_sendmsg+0x1c/0x30)
[ 1706.118086] r4:df471d74
[ 1706.118119] [<c03157d4>] (sock_sendmsg) from [<c03158f4>] (kernel_sendmsg+0x38/0x40)
[ 1706.118144] [<c03158bc>] (kernel_sendmsg) from [<c03afa2c>] (xs_send_kvec+0x94/0x9c)
[ 1706.118161] r5:00000000 r4:df02c780
[ 1706.118202] [<c03af998>] (xs_send_kvec) from [<c03afaa0>] (xs_sendpages+0x6c/0x244)
[ 1706.118218] r9:00000000 r8:df02c780 r7:0000007c r6:df7eb604 r5:00000001 r4:00000000
[ 1706.118293] [<c03afa34>] (xs_sendpages) from [<c03afd7c>] (xs_tcp_send_request+0x80/0x134)
[ 1706.118309] r10:00000000 r9:00000000 r8:de8e8000 r7:de0d7258 r6:df7eb604 r5:00000001
[ 1706.118380] r4:df7eb600
[ 1706.118436] [<c03afcfc>] (xs_tcp_send_request) from [<c03ad670>] (xprt_transmit+0x58/0x214)
[ 1706.118455] r10:de92cc60 r9:00000000 r8:d3697fdd r7:df7eb674 r6:de0d7258 r5:df7eb600
[ 1706.118528] r4:de8e8000
[ 1706.118567] [<c03ad618>] (xprt_transmit) from [<c03aac74>] (call_transmit+0x18c/0x230)
[ 1706.118583] r7:df7eb600 r6:00000001 r5:df7eb600 r4:de0d7258
[ 1706.118649] [<c03aaae8>] (call_transmit) from [<c03b1ed8>] (__rpc_execute+0x54/0x2c4)
[ 1706.118667] r8:c0508940 r7:00000000 r6:c03b1f08 r5:00000001 r4:de0d7258
[ 1706.118736] [<c03b1e84>] (__rpc_execute) from [<c03b215c>] (rpc_async_schedule+0x14/0x18)
[ 1706.118751] r7:00000000 r6:dfbf0800 r5:de92cc60 r4:de0d727c
[ 1706.118821] [<c03b2148>] (rpc_async_schedule) from [<c002c79c>] (process_one_work+0x130/0x3ec)
[ 1706.118847] [<c002c66c>] (process_one_work) from [<c002caa8>] (worker_thread+0x50/0x5ac)
[ 1706.118864] r10:de92cc60 r9:c0508940 r8:00000008 r7:c050b220 r6:c0508954 r5:de92cc78
[ 1706.118935] r4:c0508940
[ 1706.118980] [<c002ca58>] (worker_thread) from [<c00319cc>] (kthread+0xc8/0xe4)
[ 1706.118997] r10:00000000 r9:00000000 r8:00000000 r7:c002ca58 r6:de92cc60 r5:00000000
[ 1706.119067] r4:de385880
[ 1706.119106] [<c0031904>] (kthread) from [<c000a390>] (ret_from_fork+0x14/0x24)
[ 1706.119123] r7:00000000 r6:00000000 r5:c0031904 r4:de385880
[ 1706.119171] Mem-Info:
[ 1706.119213] active_anon:2694 inactive_anon:2799 isolated_anon:0
[ 1706.119213] active_file:43728 inactive_file:71243 isolated_file:0
[ 1706.119213] unevictable:0 dirty:1486 writeback:0 unstable:0
[ 1706.119213] slab_reclaimable:3099 slab_unreclaimable:710
[ 1706.119213] mapped:1763 shmem:19 pagetables:186 bounce:0
[ 1706.119213] free:290 free_pcp:55 free_cma:0
[ 1706.119304] Normal free:1160kB min:2868kB low:3584kB high:4300kB active_anon:10776kB inactive_anon:11196kB active_file:174912kB inactive_file:284972kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:524288kB managed:514412kB mlocked:0kB dirty:5944kB writeback:0kB mapped:7052kB shmem:76kB slab_reclaimable:12396kB slab_unreclaimable:2840kB kernel_stack:584kB pagetables:744kB unstable:0kB bounce:0kB free_pcp:220kB local_pcp:220kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
[ 1706.119325] lowmem_reserve[]: 0 0
[ 1706.119341] Normal: 80*4kB (UME) 105*8kB (UE) 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1160kB
[ 1706.119402] 114997 total pagecache pages
[ 1706.119420] 0 pages in swap cache
[ 1706.119434] Swap cache stats: add 0, delete 0, find 0/0
[ 1706.119444] Free swap = 1048572kB
[ 1706.119454] Total swap = 1048572kB
[ 1706.119463] 131072 pages RAM
[ 1706.119473] 0 pages HighMem/MovableOnly
[ 1706.119483] 2469 pages reserved
[ 1706.119507] SLUB: Unable to allocate memory on node -1, gfp=0x2080020(GFP_ATOMIC)
[ 1706.119525] cache: kmalloc-192, object size: 192, buffer size: 192, default order: 0, min order: 0
[ 1706.119541] node 0: slabs: 12, objs: 252, free: 0
[ 1706.119581] kworker/0:1H: page allocation failure: order:0, mode:0x2200020(GFP_NOWAIT|__GFP_HIGH|__GFP_NOTRACK)
[ 1706.119603] CPU: 0 PID: 519 Comm: kworker/0:1H Not tainted 4.6.0-rc5-iop32x-los_a50bb #1
[ 1706.119616] Hardware name: Thecus N2100
[ 1706.119636] Workqueue: rpciod rpc_async_schedule
[ 1706.119646] Backtrace:
[ 1706.119678] [<c000d460>] (dump_backtrace) from [<c000d658>] (show_stack+0x18/0x1c)
[ 1706.119697] r7:00000000 r6:00000060 r5:00000000 r4:00000000
[ 1706.119768] [<c000d640>] (show_stack) from [<c01c1e84>] (dump_stack+0x20/0x28)
[ 1706.119808] [<c01c1e64>] (dump_stack) from [<c007b9b0>] (warn_alloc_failed+0xf0/0x134)
[ 1706.119839] [<c007b8c4>] (warn_alloc_failed) from [<c007df78>] (__alloc_pages_nodemask+0x284/0x8fc)
[ 1706.119857] r3:02200020 r2:00000000
[ 1706.119893] r5:00000000 r4:c0524c40
[ 1706.119959] [<c007dcf4>] (__alloc_pages_nodemask) from [<c00b1948>] (new_slab+0x3c4/0x430)
[ 1706.119976] r10:00000000 r9:c0321fdc r8:00000000 r7:df401d00 r6:00000015 r5:02000020
[ 1706.120047] r4:df401d00
[ 1706.120084] [<c00b1584>] (new_slab) from [<c00b2bd8>] (___slab_alloc.constprop.8+0x238/0x298)
[ 1706.120100] r10:00000000 r9:c0321fdc r8:02080020 r7:df401d00 r6:dfbef060 r5:00000000
[ 1706.120170] r4:00000000
[ 1706.120203] [<c00b29a0>] (___slab_alloc.constprop.8) from [<c00b2f68>] (kmem_cache_alloc+0xbc/0xf8)
[ 1706.120219] r10:00167e97 r9:340285ee r8:e11e1960 r7:60000013 r6:02080020 r5:df401d00
[ 1706.120290] r4:00000000
[ 1706.120328] [<c00b2eac>] (kmem_cache_alloc) from [<c0321fdc>] (__build_skb+0x2c/0x98)
[ 1706.120345] r7:cc593240 r6:000006e0 r5:de3523f0 r4:000006e0
[ 1706.120405] [<c0321fb0>] (__build_skb) from [<c032226c>] (__napi_alloc_skb+0xb0/0xfc)
[ 1706.120421] r9:340285ee r8:e11e1960 r7:cc593240 r6:c0520ab8 r5:de3523f0 r4:000006e0
[ 1706.120523] [<c03221bc>] (__napi_alloc_skb) from [<c028a2fc>] (rtl8169_poll+0x3a0/0x588)
[ 1706.120541] r7:de374000 r6:c000fd08 r5:000005ea r4:de3523f0
[ 1706.120607] [<c0289f5c>] (rtl8169_poll) from [<c032d098>] (net_rx_action+0x1cc/0x2ec)
[ 1706.120624] r10:00022548 r9:df471ba8 r8:c05257e0 r7:0000012c r6:00000040 r5:00000001
[ 1706.120694] r4:de3523f0
[ 1706.120751] [<c032cecc>] (net_rx_action) from [<c001bd68>] (__do_softirq+0xf4/0x254)
[ 1706.120771] r10:00000101 r9:c052618c r8:40000001 r7:c0526188 r6:df470000 r5:00000003
[ 1706.120842] r4:00000000
[ 1706.120879] [<c001bc74>] (__do_softirq) from [<c001bf58>] (do_softirq.part.2+0x34/0x40)
[ 1706.120895] r10:00000001 r9:c0525a26 r8:c05555c0 r7:00000000 r6:deae8a80 r5:00000000
[ 1706.120966] r4:20000013
[ 1706.121006] [<c001bf24>] (do_softirq.part.2) from [<c001c010>] (__local_bh_enable_ip+0xac/0xcc)
[ 1706.121023] r5:00000000 r4:00000200
[ 1706.121086] [<c001bf64>] (__local_bh_enable_ip) from [<c0319538>] (release_sock+0x12c/0x158)
[ 1706.121104] r5:00000000 r4:00000000
[ 1706.121162] [<c031940c>] (release_sock) from [<c03671b4>] (tcp_sendmsg+0xf8/0xa90)
[ 1706.121178] r10:00004040 r9:deae8a80 r8:df471d74 r7:88cd7146 r6:0000059c r5:de8ccb00
[ 1706.121249] r4:0000007c r3:00000001
[ 1706.121295] [<c03670bc>] (tcp_sendmsg) from [<c038d7f0>] (inet_sendmsg+0x3c/0x74)
[ 1706.121313] r10:df471e2c r9:00000000 r8:df7eb604 r7:00000000 r6:00000000 r5:df02c780
[ 1706.121382] r4:deae8a80
[ 1706.121416] [<c038d7b4>] (inet_sendmsg) from [<c03157f0>] (sock_sendmsg+0x1c/0x30)
[ 1706.121432] r4:df471d74
[ 1706.121466] [<c03157d4>] (sock_sendmsg) from [<c03158f4>] (kernel_sendmsg+0x38/0x40)
[ 1706.121491] [<c03158bc>] (kernel_sendmsg) from [<c03afa2c>] (xs_send_kvec+0x94/0x9c)
[ 1706.121508] r5:00000000 r4:df02c780
[ 1706.121549] [<c03af998>] (xs_send_kvec) from [<c03afaa0>] (xs_sendpages+0x6c/0x244)
[ 1706.121566] r9:00000000 r8:df02c780 r7:0000007c r6:df7eb604 r5:00000001 r4:00000000
[ 1706.121641] [<c03afa34>] (xs_sendpages) from [<c03afd7c>] (xs_tcp_send_request+0x80/0x134)
[ 1706.121657] r10:00000000 r9:00000000 r8:de8e8000 r7:de0d7258 r6:df7eb604 r5:00000001
[ 1706.121727] r4:df7eb600
[ 1706.121782] [<c03afcfc>] (xs_tcp_send_request) from [<c03ad670>] (xprt_transmit+0x58/0x214)
[ 1706.121802] r10:de92cc60 r9:00000000 r8:d3697fdd r7:df7eb674 r6:de0d7258 r5:df7eb600
[ 1706.121873] r4:de8e8000
[ 1706.121912] [<c03ad618>] (xprt_transmit) from [<c03aac74>] (call_transmit+0x18c/0x230)
[ 1706.121928] r7:df7eb600 r6:00000001 r5:df7eb600 r4:de0d7258
[ 1706.121994] [<c03aaae8>] (call_transmit) from [<c03b1ed8>] (__rpc_execute+0x54/0x2c4)
[ 1706.122011] r8:c0508940 r7:00000000 r6:c03b1f08 r5:00000001 r4:de0d7258
[ 1706.122081] [<c03b1e84>] (__rpc_execute) from [<c03b215c>] (rpc_async_schedule+0x14/0x18)
[ 1706.122098] r7:00000000 r6:dfbf0800 r5:de92cc60 r4:de0d727c
[ 1706.122167] [<c03b2148>] (rpc_async_schedule) from [<c002c79c>] (process_one_work+0x130/0x3ec)
[ 1706.122192] [<c002c66c>] (process_one_work) from [<c002caa8>] (worker_thread+0x50/0x5ac)
[ 1706.122209] r10:de92cc60 r9:c0508940 r8:00000008 r7:c050b220 r6:c0508954 r5:de92cc78
[ 1706.122279] r4:c0508940
[ 1706.122324] [<c002ca58>] (worker_thread) from [<c00319cc>] (kthread+0xc8/0xe4)
[ 1706.122341] r10:00000000 r9:00000000 r8:00000000 r7:c002ca58 r6:de92cc60 r5:00000000
[ 1706.122412] r4:de385880
[ 1706.122449] [<c0031904>] (kthread) from [<c000a390>] (ret_from_fork+0x14/0x24)
[ 1706.122465] r7:00000000 r6:00000000 r5:c0031904 r4:de385880
[ 1706.122513] Mem-Info:
[ 1706.122553] active_anon:2694 inactive_anon:2799 isolated_anon:0
[ 1706.122553] active_file:43728 inactive_file:71243 isolated_file:0
[ 1706.122553] unevictable:0 dirty:1486 writeback:0 unstable:0
[ 1706.122553] slab_reclaimable:3099 slab_unreclaimable:710
[ 1706.122553] mapped:1763 shmem:19 pagetables:186 bounce:0
[ 1706.122553] free:290 free_pcp:55 free_cma:0
[ 1706.122638] Normal free:1160kB min:2868kB low:3584kB high:4300kB active_anon:10776kB inactive_anon:11196kB active_file:174912kB inactive_file:284972kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:524288kB managed:514412kB mlocked:0kB dirty:5944kB writeback:0kB mapped:7052kB shmem:76kB slab_reclaimable:12396kB slab_unreclaimable:2840kB kernel_stack:584kB pagetables:744kB unstable:0kB bounce:0kB free_pcp:220kB local_pcp:220kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
[ 1706.122658] lowmem_reserve[]: 0 0
[ 1706.122674] Normal: 80*4kB (UME) 105*8kB (UE) 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1160kB
[ 1706.122734] 114997 total pagecache pages
[ 1706.122752] 0 pages in swap cache
[ 1706.122767] Swap cache stats: add 0, delete 0, find 0/0
[ 1706.122777] Free swap = 1048572kB
[ 1706.122787] Total swap = 1048572kB
[ 1706.122796] 131072 pages RAM
[ 1706.122805] 0 pages HighMem/MovableOnly
[ 1706.122815] 2469 pages reserved
[ 1706.122835] SLUB: Unable to allocate memory on node -1, gfp=0x2080020(GFP_ATOMIC)
[ 1706.122851] cache: kmalloc-192, object size: 192, buffer size: 192, default order: 0, min order: 0
[ 1706.122868] node 0: slabs: 12, objs: 252, free: 0
[ 1706.122903] kworker/0:1H: page allocation failure: order:0, mode:0x2200020(GFP_NOWAIT|__GFP_HIGH|__GFP_NOTRACK)
[ 1706.122927] CPU: 0 PID: 519 Comm: kworker/0:1H Not tainted 4.6.0-rc5-iop32x-los_a50bb #1
[ 1706.122939] Hardware name: Thecus N2100
[ 1706.122959] Workqueue: rpciod rpc_async_schedule
[ 1706.122968] Backtrace:
[ 1706.123001] [<c000d460>] (dump_backtrace) from [<c000d658>] (show_stack+0x18/0x1c)
[ 1706.123019] r7:00000000 r6:00000060 r5:00000000 r4:00000000
[ 1706.123091] [<c000d640>] (show_stack) from [<c01c1e84>] (dump_stack+0x20/0x28)
[ 1706.123131] [<c01c1e64>] (dump_stack) from [<c007b9b0>] (warn_alloc_failed+0xf0/0x134)
[ 1706.123161] [<c007b8c4>] (warn_alloc_failed) from [<c007df78>] (__alloc_pages_nodemask+0x284/0x8fc)
[ 1706.123177] r3:02200020 r2:00000000
[ 1706.123213] r5:00000000 r4:c0524c40
[ 1706.123279] [<c007dcf4>] (__alloc_pages_nodemask) from [<c00b1948>] (new_slab+0x3c4/0x430)
[ 1706.123296] r10:00000000 r9:c0321fdc r8:00000000 r7:df401d00 r6:00000015 r5:02000020
[ 1706.123367] r4:df401d00
[ 1706.123405] [<c00b1584>] (new_slab) from [<c00b2bd8>] (___slab_alloc.constprop.8+0x238/0x298)
[ 1706.123421] r10:00000000 r9:c0321fdc r8:02080020 r7:df401d00 r6:dfbef060 r5:00000000
[ 1706.123491] r4:00000000
[ 1706.123524] [<c00b29a0>] (___slab_alloc.constprop.8) from [<c00b2f68>] (kmem_cache_alloc+0xbc/0xf8)
[ 1706.123541] r10:00167e98 r9:340285ee r8:e11e1970 r7:60000013 r6:02080020 r5:df401d00
[ 1706.123611] r4:00000000
[ 1706.123650] [<c00b2eac>] (kmem_cache_alloc) from [<c0321fdc>] (__build_skb+0x2c/0x98)
[ 1706.123668] r7:cc593920 r6:000006e0 r5:de3523f0 r4:000006e0
[ 1706.123729] [<c0321fb0>] (__build_skb) from [<c032226c>] (__napi_alloc_skb+0xb0/0xfc)
[ 1706.123745] r9:340285ee r8:e11e1970 r7:cc593920 r6:c0520ab8 r5:de3523f0 r4:000006e0
[ 1706.123848] [<c03221bc>] (__napi_alloc_skb) from [<c028a2fc>] (rtl8169_poll+0x3a0/0x588)
[ 1706.123868] r7:de378000 r6:c000fd08 r5:000005ea r4:de3523f0
[ 1706.123936] [<c0289f5c>] (rtl8169_poll) from [<c032d098>] (net_rx_action+0x1cc/0x2ec)
[ 1706.123953] r10:00022548 r9:df471ba8 r8:c05257e0 r7:0000012c r6:00000040 r5:00000001
[ 1706.124023] r4:de3523f0
[ 1706.124077] [<c032cecc>] (net_rx_action) from [<c001bd68>] (__do_softirq+0xf4/0x254)
[ 1706.124095] r10:00000101 r9:c052618c r8:40000001 r7:c0526188 r6:df470000 r5:00000003
[ 1706.124165] r4:00000000
[ 1706.124203] [<c001bc74>] (__do_softirq) from [<c001bf58>] (do_softirq.part.2+0x34/0x40)
[ 1706.124218] r10:00000001 r9:c0525a26 r8:c05555c0 r7:00000000 r6:deae8a80 r5:00000000
[ 1706.124288] r4:20000013
[ 1706.124327] [<c001bf24>] (do_softirq.part.2) from [<c001c010>] (__local_bh_enable_ip+0xac/0xcc)
[ 1706.124343] r5:00000000 r4:00000200
[ 1706.124406] [<c001bf64>] (__local_bh_enable_ip) from [<c0319538>] (release_sock+0x12c/0x158)
[ 1706.124421] r5:00000000 r4:00000000
[ 1706.124480] [<c031940c>] (release_sock) from [<c03671b4>] (tcp_sendmsg+0xf8/0xa90)
[ 1706.124497] r10:00004040 r9:deae8a80 r8:df471d74 r7:88cd7146 r6:0000059c r5:de8ccb00
[ 1706.124568] r4:0000007c r3:00000001
[ 1706.124615] [<c03670bc>] (tcp_sendmsg) from [<c038d7f0>] (inet_sendmsg+0x3c/0x74)
[ 1706.124633] r10:df471e2c r9:00000000 r8:df7eb604 r7:00000000 r6:00000000 r5:df02c780
[ 1706.124704] r4:deae8a80
[ 1706.124738] [<c038d7b4>] (inet_sendmsg) from [<c03157f0>] (sock_sendmsg+0x1c/0x30)
[ 1706.124754] r4:df471d74
[ 1706.124787] [<c03157d4>] (sock_sendmsg) from [<c03158f4>] (kernel_sendmsg+0x38/0x40)
[ 1706.124813] [<c03158bc>] (kernel_sendmsg) from [<c03afa2c>] (xs_send_kvec+0x94/0x9c)
[ 1706.124828] r5:00000000 r4:df02c780
[ 1706.124870] [<c03af998>] (xs_send_kvec) from [<c03afaa0>] (xs_sendpages+0x6c/0x244)
[ 1706.124887] r9:00000000 r8:df02c780 r7:0000007c r6:df7eb604 r5:00000001 r4:00000000
[ 1706.124963] [<c03afa34>] (xs_sendpages) from [<c03afd7c>] (xs_tcp_send_request+0x80/0x134)
[ 1706.124980] r10:00000000 r9:00000000 r8:de8e8000 r7:de0d7258 r6:df7eb604 r5:00000001
[ 1706.125050] r4:df7eb600
[ 1706.125106] [<c03afcfc>] (xs_tcp_send_request) from [<c03ad670>] (xprt_transmit+0x58/0x214)
[ 1706.125126] r10:de92cc60 r9:00000000 r8:d3697fdd r7:df7eb674 r6:de0d7258 r5:df7eb600
[ 1706.125197] r4:de8e8000
[ 1706.125234] [<c03ad618>] (xprt_transmit) from [<c03aac74>] (call_transmit+0x18c/0x230)
[ 1706.125251] r7:df7eb600 r6:00000001 r5:df7eb600 r4:de0d7258
[ 1706.125316] [<c03aaae8>] (call_transmit) from [<c03b1ed8>] (__rpc_execute+0x54/0x2c4)
[ 1706.125333] r8:c0508940 r7:00000000 r6:c03b1f08 r5:00000001 r4:de0d7258
[ 1706.125402] [<c03b1e84>] (__rpc_execute) from [<c03b215c>] (rpc_async_schedule+0x14/0x18)
[ 1706.125419] r7:00000000 r6:dfbf0800 r5:de92cc60 r4:de0d727c
[ 1706.125487] [<c03b2148>] (rpc_async_schedule) from [<c002c79c>] (process_one_work+0x130/0x3ec)
[ 1706.125511] [<c002c66c>] (process_one_work) from [<c002caa8>] (worker_thread+0x50/0x5ac)
[ 1706.125528] r10:de92cc60 r9:c0508940 r8:00000008 r7:c050b220 r6:c0508954 r5:de92cc78
[ 1706.125597] r4:c0508940
[ 1706.125643] [<c002ca58>] (worker_thread) from [<c00319cc>] (kthread+0xc8/0xe4)
[ 1706.125660] r10:00000000 r9:00000000 r8:00000000 r7:c002ca58 r6:de92cc60 r5:00000000
[ 1706.125730] r4:de385880
[ 1706.125767] [<c0031904>] (kthread) from [<c000a390>] (ret_from_fork+0x14/0x24)
[ 1706.125784] r7:00000000 r6:00000000 r5:c0031904 r4:de385880
[ 1706.125832] Mem-Info:
[ 1706.125871] active_anon:2694 inactive_anon:2799 isolated_anon:0
[ 1706.125871] active_file:43728 inactive_file:71243 isolated_file:0
[ 1706.125871] unevictable:0 dirty:1486 writeback:0 unstable:0
[ 1706.125871] slab_reclaimable:3099 slab_unreclaimable:710
[ 1706.125871] mapped:1763 shmem:19 pagetables:186 bounce:0
[ 1706.125871] free:290 free_pcp:55 free_cma:0
[ 1706.125956] Normal free:1160kB min:2868kB low:3584kB high:4300kB active_anon:10776kB inactive_anon:11196kB active_file:174912kB inactive_file:284972kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:524288kB managed:514412kB mlocked:0kB dirty:5944kB writeback:0kB mapped:7052kB shmem:76kB slab_reclaimable:12396kB slab_unreclaimable:2840kB kernel_stack:584kB pagetables:744kB unstable:0kB bounce:0kB free_pcp:220kB local_pcp:220kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
[ 1706.125976] lowmem_reserve[]: 0 0
[ 1706.125990] Normal: 80*4kB (UME) 105*8kB (UE) 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1160kB
[ 1706.126051] 114997 total pagecache pages
[ 1706.126068] 0 pages in swap cache
[ 1706.126081] Swap cache stats: add 0, delete 0, find 0/0
[ 1706.126091] Free swap = 1048572kB
[ 1706.126100] Total swap = 1048572kB
[ 1706.126110] 131072 pages RAM
[ 1706.126119] 0 pages HighMem/MovableOnly
[ 1706.126128] 2469 pages reserved
[ 1706.126149] SLUB: Unable to allocate memory on node -1, gfp=0x2080020(GFP_ATOMIC)
[ 1706.126168] cache: kmalloc-192, object size: 192, buffer size: 192, default order: 0, min order: 0
[ 1706.126183] node 0: slabs: 12, objs: 252, free: 0
[ 1706.126360] kworker/0:1H: page allocation failure: order:0, mode:0x2200020(GFP_NOWAIT|__GFP_HIGH|__GFP_NOTRACK)
[ 1706.126399] CPU: 0 PID: 519 Comm: kworker/0:1H Not tainted 4.6.0-rc5-iop32x-los_a50bb #1
[ 1706.126411] Hardware name: Thecus N2100
[ 1706.126441] Workqueue: rpciod rpc_async_schedule
[ 1706.126456] Backtrace:
[ 1706.126493] [<c000d460>] (dump_backtrace) from [<c000d658>] (show_stack+0x18/0x1c)
[ 1706.126519] r7:00000000 r6:00000060 r5:00000000 r4:00000000
[ 1706.126594] [<c000d640>] (show_stack) from [<c01c1e84>] (dump_stack+0x20/0x28)
[ 1706.126638] [<c01c1e64>] (dump_stack) from [<c007b9b0>] (warn_alloc_failed+0xf0/0x134)
[ 1706.126669] [<c007b8c4>] (warn_alloc_failed) from [<c007df78>] (__alloc_pages_nodemask+0x284/0x8fc)
[ 1706.126686] r3:02200020 r2:00000000
[ 1706.126721] r5:00000000 r4:c0524c40
[ 1706.126789] [<c007dcf4>] (__alloc_pages_nodemask) from [<c00b1948>] (new_slab+0x3c4/0x430)
[ 1706.126807] r10:00000000 r9:c0321fdc r8:00000000 r7:df401d00 r6:00000015 r5:02000020
[ 1706.126878] r4:df401d00
[ 1706.126916] [<c00b1584>] (new_slab) from [<c00b2bd8>] (___slab_alloc.constprop.8+0x238/0x298)
[ 1706.126934] r10:00000000 r9:c0321fdc r8:02080020 r7:df401d00 r6:dfbef060 r5:00000000
[ 1706.127003] r4:00000000
[ 1706.127037] [<c00b29a0>] (___slab_alloc.constprop.8) from [<c00b2f68>] (kmem_cache_alloc+0xbc/0xf8)
[ 1706.127054] r10:00167e99 r9:340285ee r8:e11e1980 r7:60000013 r6:02080020 r5:df401d00
[ 1706.127124] r4:00000000
[ 1706.127162] [<c00b2eac>] (kmem_cache_alloc) from [<c0321fdc>] (__build_skb+0x2c/0x98)
[ 1706.127178] r7:cc593240 r6:000006e0 r5:de3523f0 r4:000006e0
[ 1706.127238] [<c0321fb0>] (__build_skb) from [<c032226c>] (__napi_alloc_skb+0xb0/0xfc)
[ 1706.127254] r9:340285ee r8:e11e1980 r7:cc593240 r6:c0520ab8 r5:de3523f0 r4:000006e0
[ 1706.127354] [<c03221bc>] (__napi_alloc_skb) from [<c028a2fc>] (rtl8169_poll+0x3a0/0x588)
[ 1706.127373] r7:de37c000 r6:c000fd08 r5:000005ea r4:de3523f0
[ 1706.127438] [<c0289f5c>] (rtl8169_poll) from [<c032d098>] (net_rx_action+0x1cc/0x2ec)
[ 1706.127454] r10:00022548 r9:df471ba8 r8:c05257e0 r7:0000012c r6:00000040 r5:00000001
[ 1706.127524] r4:de3523f0
[ 1706.127581] [<c032cecc>] (net_rx_action) from [<c001bd68>] (__do_softirq+0xf4/0x254)
[ 1706.127598] r10:00000101 r9:c052618c r8:40000001 r7:c0526188 r6:df470000 r5:00000003
[ 1706.127668] r4:00000000
[ 1706.127705] [<c001bc74>] (__do_softirq) from [<c001bf58>] (do_softirq.part.2+0x34/0x40)
[ 1706.127722] r10:00000001 r9:c0525a26 r8:c05555c0 r7:00000000 r6:deae8a80 r5:00000000
[ 1706.127793] r4:20000013
[ 1706.127831] [<c001bf24>] (do_softirq.part.2) from [<c001c010>] (__local_bh_enable_ip+0xac/0xcc)
[ 1706.127848] r5:00000000 r4:00000200
[ 1706.127910] [<c001bf64>] (__local_bh_enable_ip) from [<c0319538>] (release_sock+0x12c/0x158)
[ 1706.127926] r5:00000000 r4:00000000
[ 1706.127981] [<c031940c>] (release_sock) from [<c03671b4>] (tcp_sendmsg+0xf8/0xa90)
[ 1706.127998] r10:00004040 r9:deae8a80 r8:df471d74 r7:88cd7146 r6:0000059c r5:de8ccb00
[ 1706.128068] r4:0000007c r3:00000001
[ 1706.128114] [<c03670bc>] (tcp_sendmsg) from [<c038d7f0>] (inet_sendmsg+0x3c/0x74)
[ 1706.128134] r10:df471e2c r9:00000000 r8:df7eb604 r7:00000000 r6:00000000 r5:df02c780
[ 1706.128204] r4:deae8a80
[ 1706.128238] [<c038d7b4>] (inet_sendmsg) from [<c03157f0>] (sock_sendmsg+0x1c/0x30)
[ 1706.128254] r4:df471d74
[ 1706.128287] [<c03157d4>] (sock_sendmsg) from [<c03158f4>] (kernel_sendmsg+0x38/0x40)
[ 1706.128312] [<c03158bc>] (kernel_sendmsg) from [<c03afa2c>] (xs_send_kvec+0x94/0x9c)
[ 1706.128329] r5:00000000 r4:df02c780
[ 1706.128371] [<c03af998>] (xs_send_kvec) from [<c03afaa0>] (xs_sendpages+0x6c/0x244)
[ 1706.128388] r9:00000000 r8:df02c780 r7:0000007c r6:df7eb604 r5:00000001 r4:00000000
[ 1706.128463] [<c03afa34>] (xs_sendpages) from [<c03afd7c>] (xs_tcp_send_request+0x80/0x134)
[ 1706.128479] r10:00000000 r9:00000000 r8:de8e8000 r7:de0d7258 r6:df7eb604 r5:00000001
[ 1706.128548] r4:df7eb600
[ 1706.128607] [<c03afcfc>] (xs_tcp_send_request) from [<c03ad670>] (xprt_transmit+0x58/0x214)
[ 1706.128626] r10:de92cc60 r9:00000000 r8:d3697fdd r7:df7eb674 r6:de0d7258 r5:df7eb600
[ 1706.128697] r4:de8e8000
[ 1706.128737] [<c03ad618>] (xprt_transmit) from [<c03aac74>] (call_transmit+0x18c/0x230)
[ 1706.128753] r7:df7eb600 r6:00000001 r5:df7eb600 r4:de0d7258
[ 1706.128821] [<c03aaae8>] (call_transmit) from [<c03b1ed8>] (__rpc_execute+0x54/0x2c4)
[ 1706.128838] r8:c0508940 r7:00000000 r6:c03b1f08 r5:00000001 r4:de0d7258
[ 1706.128909] [<c03b1e84>] (__rpc_execute) from [<c03b215c>] (rpc_async_schedule+0x14/0x18)
[ 1706.128926] r7:00000000 r6:dfbf0800 r5:de92cc60 r4:de0d727c
[ 1706.128996] [<c03b2148>] (rpc_async_schedule) from [<c002c79c>] (process_one_work+0x130/0x3ec)
[ 1706.129024] [<c002c66c>] (process_one_work) from [<c002caa8>] (worker_thread+0x50/0x5ac)
[ 1706.129041] r10:de92cc60 r9:c0508940 r8:00000008 r7:c050b220 r6:c0508954 r5:de92cc78
[ 1706.129112] r4:c0508940
[ 1706.129158] [<c002ca58>] (worker_thread) from [<c00319cc>] (kthread+0xc8/0xe4)
[ 1706.129174] r10:00000000 r9:00000000 r8:00000000 r7:c002ca58 r6:de92cc60 r5:00000000
[ 1706.129243] r4:de385880
[ 1706.129281] [<c0031904>] (kthread) from [<c000a390>] (ret_from_fork+0x14/0x24)
[ 1706.129296] r7:00000000 r6:00000000 r5:c0031904 r4:de385880
[ 1706.129344] Mem-Info:
[ 1706.129386] active_anon:2694 inactive_anon:2799 isolated_anon:0
[ 1706.129386] active_file:43728 inactive_file:71243 isolated_file:0
[ 1706.129386] unevictable:0 dirty:1486 writeback:0 unstable:0
[ 1706.129386] slab_reclaimable:3099 slab_unreclaimable:710
[ 1706.129386] mapped:1763 shmem:19 pagetables:186 bounce:0
[ 1706.129386] free:290 free_pcp:55 free_cma:0
[ 1706.129475] Normal free:1160kB min:2868kB low:3584kB high:4300kB active_anon:10776kB inactive_anon:11196kB active_file:174912kB inactive_file:284972kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:524288kB managed:514412kB mlocked:0kB dirty:5944kB writeback:0kB mapped:7052kB shmem:76kB slab_reclaimable:12396kB slab_unreclaimable:2840kB kernel_stack:584kB pagetables:744kB unstable:0kB bounce:0kB free_pcp:220kB local_pcp:220kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
[ 1706.129499] lowmem_reserve[]: 0 0
[ 1706.129513] Normal: 80*4kB (UME) 105*8kB (UE) 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1160kB
[ 1706.129577] 114997 total pagecache pages
[ 1706.129595] 0 pages in swap cache
[ 1706.129607] Swap cache stats: add 0, delete 0, find 0/0
[ 1706.129617] Free swap = 1048572kB
[ 1706.129627] Total swap = 1048572kB
[ 1706.129636] 131072 pages RAM
[ 1706.129646] 0 pages HighMem/MovableOnly
[ 1706.129655] 2469 pages reserved
[ 1706.129676] SLUB: Unable to allocate memory on node -1, gfp=0x2080020(GFP_ATOMIC)
[ 1706.129694] cache: kmalloc-192, object size: 192, buffer size: 192, default order: 0, min order: 0
[ 1706.129709] node 0: slabs: 12, objs: 252, free: 0
[ 1706.129748] kworker/0:1H: page allocation failure: order:0, mode:0x2200020(GFP_NOWAIT|__GFP_HIGH|__GFP_NOTRACK)
[ 1706.129769] CPU: 0 PID: 519 Comm: kworker/0:1H Not tainted 4.6.0-rc5-iop32x-los_a50bb #1
[ 1706.129780] Hardware name: Thecus N2100
[ 1706.129799] Workqueue: rpciod rpc_async_schedule
[ 1706.129808] Backtrace:
[ 1706.129840] [<c000d460>] (dump_backtrace) from [<c000d658>] (show_stack+0x18/0x1c)
[ 1706.129857] r7:00000000 r6:00000060 r5:00000000 r4:00000000
[ 1706.129927] [<c000d640>] (show_stack) from [<c01c1e84>] (dump_stack+0x20/0x28)
[ 1706.129968] [<c01c1e64>] (dump_stack) from [<c007b9b0>] (warn_alloc_failed+0xf0/0x134)
[ 1706.129997] [<c007b8c4>] (warn_alloc_failed) from [<c007df78>] (__alloc_pages_nodemask+0x284/0x8fc)
[ 1706.130016] r3:02200020 r2:00000000
[ 1706.130052] r5:00000000 r4:c0524c40
[ 1706.130119] [<c007dcf4>] (__alloc_pages_nodemask) from [<c00b1948>] (new_slab+0x3c4/0x430)
[ 1706.130137] r10:00000000 r9:c0321fdc r8:00000000 r7:df401d00 r6:00000015 r5:02000020
[ 1706.130209] r4:df401d00
[ 1706.130246] [<c00b1584>] (new_slab) from [<c00b2bd8>] (___slab_alloc.constprop.8+0x238/0x298)
[ 1706.130262] r10:00000000 r9:c0321fdc r8:02080020 r7:df401d00 r6:dfbef060 r5:00000000
[ 1706.130333] r4:00000000
[ 1706.130367] [<c00b29a0>] (___slab_alloc.constprop.8) from [<c00b2f68>] (kmem_cache_alloc+0xbc/0xf8)
[ 1706.130384] r10:00167e9a r9:340285ee r8:e11e1990 r7:60000013 r6:02080020 r5:df401d00
[ 1706.130454] r4:00000000
[ 1706.130495] [<c00b2eac>] (kmem_cache_alloc) from [<c0321fdc>] (__build_skb+0x2c/0x98)
[ 1706.130512] r7:cc593920 r6:000006e0 r5:de3523f0 r4:000006e0
[ 1706.130572] [<c0321fb0>] (__build_skb) from [<c032226c>] (__napi_alloc_skb+0xb0/0xfc)
[ 1706.130589] r9:340285ee r8:e11e1990 r7:cc593920 r6:c0520ab8 r5:de3523f0 r4:000006e0
[ 1706.130687] [<c03221bc>] (__napi_alloc_skb) from [<c028a2fc>] (rtl8169_poll+0x3a0/0x588)
[ 1706.130707] r7:de860000 r6:c000fd08 r5:000005ea r4:de3523f0
[ 1706.130777] [<c0289f5c>] (rtl8169_poll) from [<c032d098>] (net_rx_action+0x1cc/0x2ec)
[ 1706.130794] r10:00022548 r9:df471ba8 r8:c05257e0 r7:0000012c r6:00000040 r5:00000001
[ 1706.130865] r4:de3523f0
[ 1706.130922] [<c032cecc>] (net_rx_action) from [<c001bd68>] (__do_softirq+0xf4/0x254)
[ 1706.130939] r10:00000101 r9:c052618c r8:40000001 r7:c0526188 r6:df470000 r5:00000003
[ 1706.131010] r4:00000000
[ 1706.131048] [<c001bc74>] (__do_softirq) from [<c001bf58>] (do_softirq.part.2+0x34/0x40)
[ 1706.131064] r10:00000001 r9:c0525a26 r8:c05555c0 r7:00000000 r6:deae8a80 r5:00000000
[ 1706.131134] r4:20000013
[ 1706.131173] [<c001bf24>] (do_softirq.part.2) from [<c001c010>] (__local_bh_enable_ip+0xac/0xcc)
[ 1706.131189] r5:00000000 r4:00000200
[ 1706.131251] [<c001bf64>] (__local_bh_enable_ip) from [<c0319538>] (release_sock+0x12c/0x158)
[ 1706.131268] r5:00000000 r4:00000000
[ 1706.131324] [<c031940c>] (release_sock) from [<c03671b4>] (tcp_sendmsg+0xf8/0xa90)
[ 1706.131342] r10:00004040 r9:deae8a80 r8:df471d74 r7:88cd7146 r6:0000059c r5:de8ccb00
[ 1706.131413] r4:0000007c r3:00000001
[ 1706.131459] [<c03670bc>] (tcp_sendmsg) from [<c038d7f0>] (inet_sendmsg+0x3c/0x74)
[ 1706.131477] r10:df471e2c r9:00000000 r8:df7eb604 r7:00000000 r6:00000000 r5:df02c780
[ 1706.131546] r4:deae8a80
[ 1706.131581] [<c038d7b4>] (inet_sendmsg) from [<c03157f0>] (sock_sendmsg+0x1c/0x30)
[ 1706.131597] r4:df471d74
[ 1706.131630] [<c03157d4>] (sock_sendmsg) from [<c03158f4>] (kernel_sendmsg+0x38/0x40)
[ 1706.131655] [<c03158bc>] (kernel_sendmsg) from [<c03afa2c>] (xs_send_kvec+0x94/0x9c)
[ 1706.131670] r5:00000000 r4:df02c780
[ 1706.131713] [<c03af998>] (xs_send_kvec) from [<c03afaa0>] (xs_sendpages+0x6c/0x244)
[ 1706.131728] r9:00000000 r8:df02c780 r7:0000007c r6:df7eb604 r5:00000001 r4:00000000
[ 1706.131805] [<c03afa34>] (xs_sendpages) from [<c03afd7c>] (xs_tcp_send_request+0x80/0x134)
[ 1706.131821] r10:00000000 r9:00000000 r8:de8e8000 r7:de0d7258 r6:df7eb604 r5:00000001
[ 1706.131891] r4:df7eb600
[ 1706.131945] [<c03afcfc>] (xs_tcp_send_request) from [<c03ad670>] (xprt_transmit+0x58/0x214)
[ 1706.131963] r10:de92cc60 r9:00000000 r8:d3697fdd r7:df7eb674 r6:de0d7258 r5:df7eb600
[ 1706.132035] r4:de8e8000
[ 1706.132072] [<c03ad618>] (xprt_transmit) from [<c03aac74>] (call_transmit+0x18c/0x230)
[ 1706.132088] r7:df7eb600 r6:00000001 r5:df7eb600 r4:de0d7258
[ 1706.132153] [<c03aaae8>] (call_transmit) from [<c03b1ed8>] (__rpc_execute+0x54/0x2c4)
[ 1706.132169] r8:c0508940 r7:00000000 r6:c03b1f08 r5:00000001 r4:de0d7258
[ 1706.132239] [<c03b1e84>] (__rpc_execute) from [<c03b215c>] (rpc_async_schedule+0x14/0x18)
[ 1706.132256] r7:00000000 r6:dfbf0800 r5:de92cc60 r4:de0d727c
[ 1706.132327] [<c03b2148>] (rpc_async_schedule) from [<c002c79c>] (process_one_work+0x130/0x3ec)
[ 1706.132351] [<c002c66c>] (process_one_work) from [<c002caa8>] (worker_thread+0x50/0x5ac)
[ 1706.132368] r10:de92cc60 r9:c0508940 r8:00000008 r7:c050b220 r6:c0508954 r5:de92cc78
[ 1706.132439] r4:c0508940
[ 1706.132483] [<c002ca58>] (worker_thread) from [<c00319cc>] (kthread+0xc8/0xe4)
[ 1706.132500] r10:00000000 r9:00000000 r8:00000000 r7:c002ca58 r6:de92cc60 r5:00000000
[ 1706.132570] r4:de385880
[ 1706.132608] [<c0031904>] (kthread) from [<c000a390>] (ret_from_fork+0x14/0x24)
[ 1706.132625] r7:00000000 r6:00000000 r5:c0031904 r4:de385880
[ 1706.132673] Mem-Info:
[ 1706.132714] active_anon:2694 inactive_anon:2799 isolated_anon:0
[ 1706.132714] active_file:43728 inactive_file:71243 isolated_file:0
[ 1706.132714] unevictable:0 dirty:1486 writeback:0 unstable:0
[ 1706.132714] slab_reclaimable:3099 slab_unreclaimable:710
[ 1706.132714] mapped:1763 shmem:19 pagetables:186 bounce:0
[ 1706.132714] free:290 free_pcp:55 free_cma:0
[ 1706.132799] Normal free:1160kB min:2868kB low:3584kB high:4300kB active_anon:10776kB inactive_anon:11196kB active_file:174912kB inactive_file:284972kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:524288kB managed:514412kB mlocked:0kB dirty:5944kB writeback:0kB mapped:7052kB shmem:76kB slab_reclaimable:12396kB slab_unreclaimable:2840kB kernel_stack:584kB pagetables:744kB unstable:0kB bounce:0kB free_pcp:220kB local_pcp:220kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
[ 1706.132820] lowmem_reserve[]: 0 0
[ 1706.132835] Normal: 80*4kB (UME) 105*8kB (UE) 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1160kB
[ 1706.132896] 114997 total pagecache pages
[ 1706.132914] 0 pages in swap cache
[ 1706.132927] Swap cache stats: add 0, delete 0, find 0/0
[ 1706.132937] Free swap = 1048572kB
[ 1706.132947] Total swap = 1048572kB
[ 1706.132957] 131072 pages RAM
[ 1706.132966] 0 pages HighMem/MovableOnly
[ 1706.132976] 2469 pages reserved
[ 1706.132996] SLUB: Unable to allocate memory on node -1, gfp=0x2080020(GFP_ATOMIC)
[ 1706.133014] cache: kmalloc-192, object size: 192, buffer size: 192, default order: 0, min order: 0
[ 1706.133029] node 0: slabs: 12, objs: 252, free: 0
[ 1706.133064] kworker/0:1H: page allocation failure: order:0, mode:0x2200020(GFP_NOWAIT|__GFP_HIGH|__GFP_NOTRACK)
[ 1706.133087] CPU: 0 PID: 519 Comm: kworker/0:1H Not tainted 4.6.0-rc5-iop32x-los_a50bb #1
[ 1706.133099] Hardware name: Thecus N2100
[ 1706.133118] Workqueue: rpciod rpc_async_schedule
[ 1706.133130] Backtrace:
[ 1706.133162] [<c000d460>] (dump_backtrace) from [<c000d658>] (show_stack+0x18/0x1c)
[ 1706.133179] r7:00000000 r6:00000060 r5:00000000 r4:00000000
[ 1706.133251] [<c000d640>] (show_stack) from [<c01c1e84>] (dump_stack+0x20/0x28)
[ 1706.133292] [<c01c1e64>] (dump_stack) from [<c007b9b0>] (warn_alloc_failed+0xf0/0x134)
[ 1706.133324] [<c007b8c4>] (warn_alloc_failed) from [<c007df78>] (__alloc_pages_nodemask+0x284/0x8fc)
[ 1706.133341] r3:02200020 r2:00000000
[ 1706.133377] r5:00000000 r4:c0524c40
[ 1706.133444] [<c007dcf4>] (__alloc_pages_nodemask) from [<c00b1948>] (new_slab+0x3c4/0x430)
[ 1706.133460] r10:00000000 r9:c0321fdc r8:00000000 r7:df401d00 r6:00000015 r5:02000020
[ 1706.133531] r4:df401d00
[ 1706.133569] [<c00b1584>] (new_slab) from [<c00b2bd8>] (___slab_alloc.constprop.8+0x238/0x298)
[ 1706.133585] r10:00000000 r9:c0321fdc r8:02080020 r7:df401d00 r6:dfbef060 r5:00000000
[ 1706.133655] r4:00000000
[ 1706.133688] [<c00b29a0>] (___slab_alloc.constprop.8) from [<c00b2f68>] (kmem_cache_alloc+0xbc/0xf8)
[ 1706.133705] r10:00167e9b r9:340285ee r8:e11e19a0 r7:60000013 r6:02080020 r5:df401d00
[ 1706.133775] r4:00000000
[ 1706.133816] [<c00b2eac>] (kmem_cache_alloc) from [<c0321fdc>] (__build_skb+0x2c/0x98)
[ 1706.133833] r7:cc593240 r6:000006e0 r5:de3523f0 r4:000006e0
[ 1706.133893] [<c0321fb0>] (__build_skb) from [<c032226c>] (__napi_alloc_skb+0xb0/0xfc)
[ 1706.133910] r9:340285ee r8:e11e19a0 r7:cc593240 r6:c0520ab8 r5:de3523f0 r4:000006e0
[ 1706.134009] [<c03221bc>] (__napi_alloc_skb) from [<c028a2fc>] (rtl8169_poll+0x3a0/0x588)
[ 1706.134028] r7:de864000 r6:c000fd08 r5:000005ea r4:de3523f0
[ 1706.134094] [<c0289f5c>] (rtl8169_poll) from [<c032d098>] (net_rx_action+0x1cc/0x2ec)
[ 1706.134111] r10:00022548 r9:df471ba8 r8:c05257e0 r7:0000012c r6:00000040 r5:00000001
[ 1706.134181] r4:de3523f0
[ 1706.134239] [<c032cecc>] (net_rx_action) from [<c001bd68>] (__do_softirq+0xf4/0x254)
[ 1706.134257] r10:00000101 r9:c052618c r8:40000001 r7:c0526188 r6:df470000 r5:00000003
[ 1706.134327] r4:00000000
[ 1706.134364] [<c001bc74>] (__do_softirq) from [<c001bf58>] (do_softirq.part.2+0x34/0x40)
[ 1706.134380] r10:00000001 r9:c0525a26 r8:c05555c0 r7:00000000 r6:deae8a80 r5:00000000
[ 1706.134451] r4:20000013
[ 1706.134489] [<c001bf24>] (do_softirq.part.2) from [<c001c010>] (__local_bh_enable_ip+0xac/0xcc)
[ 1706.134506] r5:00000000 r4:00000200
[ 1706.134566] [<c001bf64>] (__local_bh_enable_ip) from [<c0319538>] (release_sock+0x12c/0x158)
[ 1706.134582] r5:00000000 r4:00000000
[ 1706.134639] [<c031940c>] (release_sock) from [<c03671b4>] (tcp_sendmsg+0xf8/0xa90)
[ 1706.134657] r10:00004040 r9:deae8a80 r8:df471d74 r7:88cd7146 r6:0000059c r5:de8ccb00
[ 1706.134727] r4:0000007c r3:00000001
[ 1706.134774] [<c03670bc>] (tcp_sendmsg) from [<c038d7f0>] (inet_sendmsg+0x3c/0x74)
[ 1706.134790] r10:df471e2c r9:00000000 r8:df7eb604 r7:00000000 r6:00000000 r5:df02c780
[ 1706.134860] r4:deae8a80
[ 1706.134893] [<c038d7b4>] (inet_sendmsg) from [<c03157f0>] (sock_sendmsg+0x1c/0x30)
[ 1706.134909] r4:df471d74
[ 1706.134943] [<c03157d4>] (sock_sendmsg) from [<c03158f4>] (kernel_sendmsg+0x38/0x40)
[ 1706.134968] [<c03158bc>] (kernel_sendmsg) from [<c03afa2c>] (xs_send_kvec+0x94/0x9c)
[ 1706.134985] r5:00000000 r4:df02c780
[ 1706.135027] [<c03af998>] (xs_send_kvec) from [<c03afaa0>] (xs_sendpages+0x6c/0x244)
[ 1706.135043] r9:00000000 r8:df02c780 r7:0000007c r6:df7eb604 r5:00000001 r4:00000000
[ 1706.135118] [<c03afa34>] (xs_sendpages) from [<c03afd7c>] (xs_tcp_send_request+0x80/0x134)
[ 1706.135134] r10:00000000 r9:00000000 r8:de8e8000 r7:de0d7258 r6:df7eb604 r5:00000001
[ 1706.135203] r4:df7eb600
[ 1706.135260] [<c03afcfc>] (xs_tcp_send_request) from [<c03ad670>] (xprt_transmit+0x58/0x214)
[ 1706.135280] r10:de92cc60 r9:00000000 r8:d3697fdd r7:df7eb674 r6:de0d7258 r5:df7eb600
[ 1706.135350] r4:de8e8000
[ 1706.135388] [<c03ad618>] (xprt_transmit) from [<c03aac74>] (call_transmit+0x18c/0x230)
[ 1706.135407] r7:df7eb600 r6:00000001 r5:df7eb600 r4:de0d7258
[ 1706.135472] [<c03aaae8>] (call_transmit) from [<c03b1ed8>] (__rpc_execute+0x54/0x2c4)
[ 1706.135489] r8:c0508940 r7:00000000 r6:c03b1f08 r5:00000001 r4:de0d7258
[ 1706.135560] [<c03b1e84>] (__rpc_execute) from [<c03b215c>] (rpc_async_schedule+0x14/0x18)
[ 1706.135577] r7:00000000 r6:dfbf0800 r5:de92cc60 r4:de0d727c
[ 1706.135648] [<c03b2148>] (rpc_async_schedule) from [<c002c79c>] (process_one_work+0x130/0x3ec)
[ 1706.135673] [<c002c66c>] (process_one_work) from [<c002caa8>] (worker_thread+0x50/0x5ac)
[ 1706.135690] r10:de92cc60 r9:c0508940 r8:00000008 r7:c050b220 r6:c0508954 r5:de92cc78
[ 1706.135760] r4:c0508940
[ 1706.135805] [<c002ca58>] (worker_thread) from [<c00319cc>] (kthread+0xc8/0xe4)
[ 1706.135820] r10:00000000 r9:00000000 r8:00000000 r7:c002ca58 r6:de92cc60 r5:00000000
[ 1706.135890] r4:de385880
[ 1706.135928] [<c0031904>] (kthread) from [<c000a390>] (ret_from_fork+0x14/0x24)
[ 1706.135943] r7:00000000 r6:00000000 r5:c0031904 r4:de385880
[ 1706.135991] Mem-Info:
[ 1706.136030] active_anon:2694 inactive_anon:2799 isolated_anon:0
[ 1706.136030] active_file:43728 inactive_file:71243 isolated_file:0
[ 1706.136030] unevictable:0 dirty:1486 writeback:0 unstable:0
[ 1706.136030] slab_reclaimable:3099 slab_unreclaimable:710
[ 1706.136030] mapped:1763 shmem:19 pagetables:186 bounce:0
[ 1706.136030] free:290 free_pcp:55 free_cma:0
[ 1706.136116] Normal free:1160kB min:2868kB low:3584kB high:4300kB active_anon:10776kB inactive_anon:11196kB active_file:174912kB inactive_file:284972kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:524288kB managed:514412kB mlocked:0kB dirty:5944kB writeback:0kB mapped:7052kB shmem:76kB slab_reclaimable:12396kB slab_unreclaimable:2840kB kernel_stack:584kB pagetables:744kB unstable:0kB bounce:0kB free_pcp:220kB local_pcp:220kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
[ 1706.136136] lowmem_reserve[]: 0 0
[ 1706.136151] Normal: 80*4kB (UME) 105*8kB (UE) 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1160kB
[ 1706.136215] 114997 total pagecache pages
[ 1706.136232] 0 pages in swap cache
[ 1706.136245] Swap cache stats: add 0, delete 0, find 0/0
[ 1706.136256] Free swap = 1048572kB
[ 1706.136265] Total swap = 1048572kB
[ 1706.136274] 131072 pages RAM
[ 1706.136284] 0 pages HighMem/MovableOnly
[ 1706.136293] 2469 pages reserved
[ 1706.136312] SLUB: Unable to allocate memory on node -1, gfp=0x2080020(GFP_ATOMIC)
[ 1706.136331] cache: kmalloc-192, object size: 192, buffer size: 192, default order: 0, min order: 0
[ 1706.136347] node 0: slabs: 12, objs: 252, free: 0
[ 1706.136517] kworker/0:1H: page allocation failure: order:0, mode:0x2200020(GFP_NOWAIT|__GFP_HIGH|__GFP_NOTRACK)
[ 1706.136556] CPU: 0 PID: 519 Comm: kworker/0:1H Not tainted 4.6.0-rc5-iop32x-los_a50bb #1
[ 1706.136569] Hardware name: Thecus N2100
[ 1706.136599] Workqueue: rpciod rpc_async_schedule
[ 1706.136613] Backtrace:
[ 1706.136648] [<c000d460>] (dump_backtrace) from [<c000d658>] (show_stack+0x18/0x1c)
[ 1706.136673] r7:00000000 r6:00000060 r5:00000000 r4:00000000
[ 1706.136748] [<c000d640>] (show_stack) from [<c01c1e84>] (dump_stack+0x20/0x28)
[ 1706.136790] [<c01c1e64>] (dump_stack) from [<c007b9b0>] (warn_alloc_failed+0xf0/0x134)
[ 1706.136820] [<c007b8c4>] (warn_alloc_failed) from [<c007df78>] (__alloc_pages_nodemask+0x284/0x8fc)
[ 1706.136837] r3:02200020 r2:00000000
[ 1706.136872] r5:00000000 r4:c0524c40
[ 1706.136937] [<c007dcf4>] (__alloc_pages_nodemask) from [<c00b1948>] (new_slab+0x3c4/0x430)
[ 1706.136955] r10:00000000 r9:c0321fdc r8:00000000 r7:df401d00 r6:00000015 r5:02000020
[ 1706.137025] r4:df401d00
[ 1706.137064] [<c00b1584>] (new_slab) from [<c00b2bd8>] (___slab_alloc.constprop.8+0x238/0x298)
[ 1706.137081] r10:00000000 r9:c0321fdc r8:02080020 r7:df401d00 r6:dfbef060 r5:00000000
[ 1706.137151] r4:00000000
[ 1706.137184] [<c00b29a0>] (___slab_alloc.constprop.8) from [<c00b2f68>] (kmem_cache_alloc+0xbc/0xf8)
[ 1706.137200] r10:00167e9c r9:340285ee r8:e11e19b0 r7:60000013 r6:02080020 r5:df401d00
[ 1706.137271] r4:00000000
[ 1706.137312] [<c00b2eac>] (kmem_cache_alloc) from [<c0321fdc>] (__build_skb+0x2c/0x98)
[ 1706.137329] r7:cc593920 r6:000006e0 r5:de3523f0 r4:000006e0
[ 1706.137389] [<c0321fb0>] (__build_skb) from [<c032226c>] (__napi_alloc_skb+0xb0/0xfc)
[ 1706.137405] r9:340285ee r8:e11e19b0 r7:cc593920 r6:c0520ab8 r5:de3523f0 r4:000006e0
[ 1706.137509] [<c03221bc>] (__napi_alloc_skb) from [<c028a2fc>] (rtl8169_poll+0x3a0/0x588)
[ 1706.137531] r7:de868000 r6:c000fd08 r5:000005ea r4:de3523f0
[ 1706.137598] [<c0289f5c>] (rtl8169_poll) from [<c032d098>] (net_rx_action+0x1cc/0x2ec)
[ 1706.137616] r10:00022548 r9:df471ba8 r8:c05257e0 r7:0000012c r6:00000040 r5:00000001
[ 1706.137686] r4:de3523f0
[ 1706.137744] [<c032cecc>] (net_rx_action) from [<c001bd68>] (__do_softirq+0xf4/0x254)
[ 1706.137762] r10:00000101 r9:c052618c r8:40000001 r7:c0526188 r6:df470000 r5:00000003
[ 1706.137832] r4:00000000
[ 1706.137869] [<c001bc74>] (__do_softirq) from [<c001bf58>] (do_softirq.part.2+0x34/0x40)
[ 1706.137886] r10:00000001 r9:c0525a26 r8:c05555c0 r7:00000000 r6:deae8a80 r5:00000000
[ 1706.137955] r4:20000013
[ 1706.137994] [<c001bf24>] (do_softirq.part.2) from [<c001c010>] (__local_bh_enable_ip+0xac/0xcc)
[ 1706.138009] r5:00000000 r4:00000200
[ 1706.138070] [<c001bf64>] (__local_bh_enable_ip) from [<c0319538>] (release_sock+0x12c/0x158)
[ 1706.138087] r5:00000000 r4:00000000
[ 1706.138143] [<c031940c>] (release_sock) from [<c03671b4>] (tcp_sendmsg+0xf8/0xa90)
[ 1706.138161] r10:00004040 r9:deae8a80 r8:df471d74 r7:88cd7146 r6:0000059c r5:de8ccb00
[ 1706.138231] r4:0000007c r3:00000001
[ 1706.138278] [<c03670bc>] (tcp_sendmsg) from [<c038d7f0>] (inet_sendmsg+0x3c/0x74)
[ 1706.138296] r10:df471e2c r9:00000000 r8:df7eb604 r7:00000000 r6:00000000 r5:df02c780
[ 1706.138366] r4:deae8a80
[ 1706.138400] [<c038d7b4>] (inet_sendmsg) from [<c03157f0>] (sock_sendmsg+0x1c/0x30)
[ 1706.138417] r4:df471d74
[ 1706.138451] [<c03157d4>] (sock_sendmsg) from [<c03158f4>] (kernel_sendmsg+0x38/0x40)
[ 1706.138476] [<c03158bc>] (kernel_sendmsg) from [<c03afa2c>] (xs_send_kvec+0x94/0x9c)
[ 1706.138492] r5:00000000 r4:df02c780
[ 1706.138534] [<c03af998>] (xs_send_kvec) from [<c03afaa0>] (xs_sendpages+0x6c/0x244)
[ 1706.138550] r9:00000000 r8:df02c780 r7:0000007c r6:df7eb604 r5:00000001 r4:00000000
[ 1706.138625] [<c03afa34>] (xs_sendpages) from [<c03afd7c>] (xs_tcp_send_request+0x80/0x134)
[ 1706.138641] r10:00000000 r9:00000000 r8:de8e8000 r7:de0d7258 r6:df7eb604 r5:00000001
[ 1706.138711] r4:df7eb600
[ 1706.138767] [<c03afcfc>] (xs_tcp_send_request) from [<c03ad670>] (xprt_transmit+0x58/0x214)
[ 1706.138785] r10:de92cc60 r9:00000000 r8:d3697fdd r7:df7eb674 r6:de0d7258 r5:df7eb600
[ 1706.138855] r4:de8e8000
[ 1706.138895] [<c03ad618>] (xprt_transmit) from [<c03aac74>] (call_transmit+0x18c/0x230)
[ 1706.138911] r7:df7eb600 r6:00000001 r5:df7eb600 r4:de0d7258
[ 1706.138978] [<c03aaae8>] (call_transmit) from [<c03b1ed8>] (__rpc_execute+0x54/0x2c4)
[ 1706.138994] r8:c0508940 r7:00000000 r6:c03b1f08 r5:00000001 r4:de0d7258
[ 1706.139064] [<c03b1e84>] (__rpc_execute) from [<c03b215c>] (rpc_async_schedule+0x14/0x18)
[ 1706.139080] r7:00000000 r6:dfbf0800 r5:de92cc60 r4:de0d727c
[ 1706.139152] [<c03b2148>] (rpc_async_schedule) from [<c002c79c>] (process_one_work+0x130/0x3ec)
[ 1706.139176] [<c002c66c>] (process_one_work) from [<c002caa8>] (worker_thread+0x50/0x5ac)
[ 1706.139193] r10:de92cc60 r9:c0508940 r8:00000008 r7:c050b220 r6:c0508954 r5:de92cc78
[ 1706.139265] r4:c0508940
[ 1706.139310] [<c002ca58>] (worker_thread) from [<c00319cc>] (kthread+0xc8/0xe4)
[ 1706.139326] r10:00000000 r9:00000000 r8:00000000 r7:c002ca58 r6:de92cc60 r5:00000000
[ 1706.139396] r4:de385880
[ 1706.139434] [<c0031904>] (kthread) from [<c000a390>] (ret_from_fork+0x14/0x24)
[ 1706.139450] r7:00000000 r6:00000000 r5:c0031904 r4:de385880
[ 1706.139498] Mem-Info:
[ 1706.139539] active_anon:2694 inactive_anon:2799 isolated_anon:0
[ 1706.139539] active_file:43728 inactive_file:71243 isolated_file:0
[ 1706.139539] unevictable:0 dirty:1486 writeback:0 unstable:0
[ 1706.139539] slab_reclaimable:3099 slab_unreclaimable:710
[ 1706.139539] mapped:1763 shmem:19 pagetables:186 bounce:0
[ 1706.139539] free:290 free_pcp:55 free_cma:0
[ 1706.139630] Normal free:1160kB min:2868kB low:3584kB high:4300kB active_anon:10776kB inactive_anon:11196kB active_file:174912kB inactive_file:284972kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:524288kB managed:514412kB mlocked:0kB dirty:5944kB writeback:0kB mapped:7052kB shmem:76kB slab_reclaimable:12396kB slab_unreclaimable:2840kB kernel_stack:584kB pagetables:744kB unstable:0kB bounce:0kB free_pcp:220kB local_pcp:220kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
[ 1706.139649] lowmem_reserve[]: 0 0
[ 1706.139664] Normal: 80*4kB (UME) 105*8kB (UE) 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 1160kB
[ 1706.139727] 114997 total pagecache pages
[ 1706.139745] 0 pages in swap cache
[ 1706.139758] Swap cache stats: add 0, delete 0, find 0/0
[ 1706.139768] Free swap = 1048572kB
[ 1706.139778] Total swap = 1048572kB
[ 1706.139787] 131072 pages RAM
[ 1706.139797] 0 pages HighMem/MovableOnly
[ 1706.139806] 2469 pages reserved
[ 1706.139829] SLUB: Unable to allocate memory on node -1, gfp=0x2080020(GFP_ATOMIC)
[ 1706.139847] cache: kmalloc-192, object size: 192, buffer size: 192, default order: 0, min order: 0
[ 1706.139862] node 0: slabs: 12, objs: 252, free: 0
A.
--
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 3.2 085/115] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good.
From: Ben Greear @ 2016-04-30 19:40 UTC (permalink / raw)
To: Ben Hutchings, Sabrina Dubroca
Cc: Hannes Frederic Sowa, linux-kernel, stable, akpm, David S. Miller,
Vijay Pandurangan, Cong Wang, netdev, Evan Jones, Nicolas Dichtel,
Phil Sutter, Toshiaki Makita, xiyou.wangcong
In-Reply-To: <1462041181.17662.3.camel@decadent.org.uk>
On 04/30/2016 11:33 AM, Ben Hutchings wrote:
> On Thu, 2016-04-28 at 12:29 +0200, Sabrina Dubroca wrote:
>> Hello,
>>> http://dmz2.candelatech.com/?p=linux-4.4.dev.y/.git;a=commitdiff;h=8153e983c0e5eba1aafe1fc296248ed2a553f1ac;hp=454b07405d694dad52e7f41af5816eed0190da8a
>> Actually, no, this is not really a regression.
> [...]
>
> It really is. Even though the old behaviour was a bug (raw packets
> should not be changed), if there are real applications that depend on
> that then we have to keep those applications working somehow.
To be honest, I fail to see why the old behaviour is a bug when sending
raw packets from user-space. If raw packets should not be changed, then
we need some way to specify what the checksum setting is to begin with,
otherwise, user-space has not enough control.
A socket option for new programs, and sysctl configurable defaults for raw sockets
for old binary programs would be sufficient I think.
Thanks,
Ben
--
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: __napi_alloc_skb failures locking up the box
From: Eric Dumazet @ 2016-04-30 19:54 UTC (permalink / raw)
To: Aaro Koskinen; +Cc: David S. Miller, netdev, linux-mm, linux-kernel
In-Reply-To: <20160430192402.GA8366@raspberrypi.musicnaut.iki.fi>
On Sat, 2016-04-30 at 22:24 +0300, Aaro Koskinen wrote:
> Hi,
>
> I have old NAS box (Thecus N2100) with 512 MB RAM, where rsync from NFS ->
> disk reliably results in temporary out-of-memory conditions.
>
> When this happens the dmesg gets flooded with below logs. If the serial
> console logging is enabled, this will lock up the box completely and
> the backup is not making any progress.
>
> Shouldn't these allocation failures be ratelimited somehow (or even made
> silent)? It doesn't sound right if I can lock up the system simply by
> copying files...
Agreed.
All napi_alloc_skb() callers handle failure just fine.
If they did not, a NULL deref would produce a proper stack dump.
When memory gets this tight, other traces will be dumped anyway.
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 15d0df943466..0652709fe81a 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -2423,7 +2423,7 @@ struct sk_buff *__napi_alloc_skb(struct napi_struct *napi,
static inline struct sk_buff *napi_alloc_skb(struct napi_struct *napi,
unsigned int length)
{
- return __napi_alloc_skb(napi, length, GFP_ATOMIC);
+ return __napi_alloc_skb(napi, length, GFP_ATOMIC | __GFP_NOWARN);
}
void napi_consume_skb(struct sk_buff *skb, int budget);
--
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 related
* Re: [PATCH 3.2 085/115] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good.
From: Tom Herbert @ 2016-04-30 19:54 UTC (permalink / raw)
To: Ben Greear
Cc: Ben Hutchings, Sabrina Dubroca, Hannes Frederic Sowa, LKML,
stable, akpm, David S. Miller, Vijay Pandurangan, Cong Wang,
Linux Kernel Network Developers, Evan Jones, Nicolas Dichtel,
Phil Sutter, Toshiaki Makita, Cong Wang
In-Reply-To: <57250A17.5090804@candelatech.com>
We've put considerable effort into cleaning up the checksum interface
to make it as unambiguous as possible, please be very careful to
follow it. Broken checksum processing is really hard to detect and
debug.
CHECKSUM_UNNECESSARY means that some number of _specific_ checksums
(indicated by csum_level) have been verified to be correct in a
packet. Blindly promoting CHECKSUM_NONE to CHECKSUM_UNNECESSARY is
never right. If CHECKSUM_UNNECESSARY is set in such a manner but the
checksum it would refer to has not been verified and is incorrect this
is a major bug.
Tom
On Sat, Apr 30, 2016 at 12:40 PM, Ben Greear <greearb@candelatech.com> wrote:
>
>
> On 04/30/2016 11:33 AM, Ben Hutchings wrote:
>>
>> On Thu, 2016-04-28 at 12:29 +0200, Sabrina Dubroca wrote:
>>>
>>> Hello,
>
>
>>>>
>>>> http://dmz2.candelatech.com/?p=linux-4.4.dev.y/.git;a=commitdiff;h=8153e983c0e5eba1aafe1fc296248ed2a553f1ac;hp=454b07405d694dad52e7f41af5816eed0190da8a
>>>
>>> Actually, no, this is not really a regression.
>>
>> [...]
>>
>> It really is. Even though the old behaviour was a bug (raw packets
>> should not be changed), if there are real applications that depend on
>> that then we have to keep those applications working somehow.
>
>
> To be honest, I fail to see why the old behaviour is a bug when sending
> raw packets from user-space. If raw packets should not be changed, then
> we need some way to specify what the checksum setting is to begin with,
> otherwise, user-space has not enough control.
>
> A socket option for new programs, and sysctl configurable defaults for raw
> sockets
> for old binary programs would be sufficient I think.
>
>
> Thanks,
> Ben
>
> --
> Ben Greear <greearb@candelatech.com>
> Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: [PATCH 3.2 085/115] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good.
From: Vijay Pandurangan @ 2016-04-30 20:15 UTC (permalink / raw)
To: Sabrina Dubroca
Cc: Ben Greear, Hannes Frederic Sowa, Ben Hutchings, LKML, stable,
akpm, David S. Miller, Cong Wang, Linux Kernel Network Developers,
Evan Jones, Nicolas Dichtel, Phil Sutter, Toshiaki Makita,
Cong Wang
In-Reply-To: <20160428102953.GA7656@bistromath.localdomain>
[oops – resending this because I was using gmail in HTML mode before
by accident]
There was a discussion on a separate thread about this. I agree with
Sabrina fully. I believe veth should provide an abstraction layer that
correctly emulates a physical network in all ways.
Consider an environment where we have multiple physical computers.
Each computer runs one or more containers, each of which has a
publicly routable ip address. When adding a new app to the cluster, a
scheduler might decide to run this container on any physical machine
of its choice, assuming that apps have a way of routing traffic to
their backends (we did something similar Google >10 years ago). This
is something we might imagine happening with docker and ipv6 for
instance.
If you have an app, A, which sends raw ethernet frames (the simplest
case I could imagine) with TCP data that has invalid checksums to app
B, which is receiving it, the behaviour of the system _will be
different_ depending upon whether app B is scheduled to run on the
same machine as app A or not. This seems like a clear bug and a broken
abstraction (especially as the default case), and something we should
endeavour to avoid.
I do see Ben's point about enabling sw checksum verification as
potentially incurring a huge performance penalty (I haven't had a
chance to measure it) that is completely wasteful in the vast majority
of cases.
Unfortunately I just don't see how we can solve this problem in a way
that preserves a correct abstraction layer while also avoiding excess
work. I guess the first piece of data that would be helpful is to
determine just how big of a performance penalty this is. If it's
small, then maybe it doesn't matter.
On Thu, Apr 28, 2016 at 6:29 AM, Sabrina Dubroca <sd@queasysnail.net> wrote:
> Hello,
>
> 2016-04-27, 17:14:44 -0700, Ben Greear wrote:
>> On 04/27/2016 05:00 PM, Hannes Frederic Sowa wrote:
>> > Hi Ben,
>> >
>> > On Wed, Apr 27, 2016, at 20:07, Ben Hutchings wrote:
>> > > On Wed, 2016-04-27 at 08:59 -0700, Ben Greear wrote:
>> > > > On 04/26/2016 04:02 PM, Ben Hutchings wrote:
>> > > > >
>> > > > > 3.2.80-rc1 review patch. If anyone has any objections, please let me know.
>> > > > I would be careful about this. It causes regressions when sending
>> > > > PACKET_SOCKET buffers from user-space to veth devices.
>> > > >
>> > > > There was a proposed upstream fix for the regression, but it has not gone
>> > > > into the tree as far as I know.
>> > > >
>> > > > http://www.spinics.net/lists/netdev/msg370436.html
>> > > [...]
>> > >
>> > > OK, I'll drop this for now.
>> >
>> > The fall out from not having this patch is in my opinion a bigger
>> > fallout than not having this patch. This patch fixes silent data
>> > corruption vs. the problem Ben Greear is talking about, which might not
>> > be that a common usage.
>> >
>> > What do others think?
>> >
>> > Bye,
>> > Hannes
>> >
>>
>> This patch from Cong Wang seems to fix the regression for me, I think it should be added and
>> tested in the main tree, and then apply them to stable as a pair.
>>
>> http://dmz2.candelatech.com/?p=linux-4.4.dev.y/.git;a=commitdiff;h=8153e983c0e5eba1aafe1fc296248ed2a553f1ac;hp=454b07405d694dad52e7f41af5816eed0190da8a
>
> Actually, no, this is not really a regression.
>
> If you capture packets on a device with checksum offloading enabled,
> the TCP/UDP checksum isn't filled. veth also behaves that way. What
> the "veth: don't modify ip_summed" patch does is enable proper
> checksum validation on veth. This really was a bug in veth.
>
> Cong's patch would also break cases where we choose to inject packets
> with invalid checksums, and they would now be accepted as correct.
>
> Your use case is invalid, it just happened to work because of a
> bug. If you want the stack to fill checksums so that you want capture
> and reinject packets, you have to disable checksum offloading (or
> compute the checksum yourself in userspace).
>
> Thanks.
>
> --
> Sabrina
^ permalink raw reply
* [PATCH] pxa168_eth: fix mdiobus_scan() error check
From: Sergei Shtylyov @ 2016-04-30 20:35 UTC (permalink / raw)
To: netdev; +Cc: arnd
In-Reply-To: <19240252.825Vlx6L0H@wasted.cogentembedded.com>
Since mdiobus_scan() returns either an error code or NULL on error, the
driver should check for both, not only for NULL, otherwise a crash is
imminent...
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
---
The patch is against DaveM's 'net.git' repo.
drivers/net/ethernet/marvell/pxa168_eth.c | 2 ++
1 file changed, 2 insertions(+)
Index: net/drivers/net/ethernet/marvell/pxa168_eth.c
===================================================================
--- net.orig/drivers/net/ethernet/marvell/pxa168_eth.c
+++ net/drivers/net/ethernet/marvell/pxa168_eth.c
@@ -979,6 +979,8 @@ static int pxa168_init_phy(struct net_de
return 0;
pep->phy = mdiobus_scan(pep->smi_bus, pep->phy_addr);
+ if (IS_ERR(pep->phy))
+ return PTR_ERR(pep->phy);
if (!pep->phy)
return -ENODEV;
^ permalink raw reply
* Re: [PATCH 3.2 085/115] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good.
From: Ben Greear @ 2016-04-30 20:59 UTC (permalink / raw)
To: Tom Herbert
Cc: Ben Hutchings, Sabrina Dubroca, Hannes Frederic Sowa, LKML,
stable, akpm, David S. Miller, Vijay Pandurangan, Cong Wang,
Linux Kernel Network Developers, Evan Jones, Nicolas Dichtel,
Phil Sutter, Toshiaki Makita, Cong Wang
In-Reply-To: <CALx6S36cqecPH+Zd8pGVdFHRi7bmWgAwm2UgFVprt5JOuO47UA@mail.gmail.com>
On 04/30/2016 12:54 PM, Tom Herbert wrote:
> We've put considerable effort into cleaning up the checksum interface
> to make it as unambiguous as possible, please be very careful to
> follow it. Broken checksum processing is really hard to detect and
> debug.
>
> CHECKSUM_UNNECESSARY means that some number of _specific_ checksums
> (indicated by csum_level) have been verified to be correct in a
> packet. Blindly promoting CHECKSUM_NONE to CHECKSUM_UNNECESSARY is
> never right. If CHECKSUM_UNNECESSARY is set in such a manner but the
> checksum it would refer to has not been verified and is incorrect this
> is a major bug.
Suppose I know that the packet received on a packet-socket has
already been verified by a NIC that supports hardware checksumming.
Then, I want to transmit it on a veth interface using a second
packet socket. I do not want veth to recalculate the checksum on
transmit, nor to validate it on the peer veth on receive, because I do
not want to waste the CPU cycles. I am assuming that my app is not
accidentally corrupting frames, so the checksum can never be bad.
How should the checksumming be configured for the packets going into
the packet-socket from user-space?
Also, I might want to send raw frames that do have
broken checksums (lets assume a real NIC, not veth), and I want them
to hit the wire with those bad checksums.
How do I configure the checksumming in this case?
Thanks,
Ben
>
> Tom
>
> On Sat, Apr 30, 2016 at 12:40 PM, Ben Greear <greearb@candelatech.com> wrote:
>>
>>
>> On 04/30/2016 11:33 AM, Ben Hutchings wrote:
>>>
>>> On Thu, 2016-04-28 at 12:29 +0200, Sabrina Dubroca wrote:
>>>>
>>>> Hello,
>>
>>
>>>>>
>>>>> http://dmz2.candelatech.com/?p=linux-4.4.dev.y/.git;a=commitdiff;h=8153e983c0e5eba1aafe1fc296248ed2a553f1ac;hp=454b07405d694dad52e7f41af5816eed0190da8a
>>>>
>>>> Actually, no, this is not really a regression.
>>>
>>> [...]
>>>
>>> It really is. Even though the old behaviour was a bug (raw packets
>>> should not be changed), if there are real applications that depend on
>>> that then we have to keep those applications working somehow.
>>
>>
>> To be honest, I fail to see why the old behaviour is a bug when sending
>> raw packets from user-space. If raw packets should not be changed, then
>> we need some way to specify what the checksum setting is to begin with,
>> otherwise, user-space has not enough control.
>>
>> A socket option for new programs, and sysctl configurable defaults for raw
>> sockets
>> for old binary programs would be sufficient I think.
>>
>>
>> Thanks,
>> Ben
>>
>> --
>> Ben Greear <greearb@candelatech.com>
>> Candela Technologies Inc http://www.candelatech.com
>
--
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: [PATCH 3.2 085/115] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good.
From: Vijay Pandurangan @ 2016-04-30 21:13 UTC (permalink / raw)
To: Ben Greear
Cc: Tom Herbert, Ben Hutchings, Sabrina Dubroca, Hannes Frederic Sowa,
LKML, stable, akpm, David S. Miller, Cong Wang,
Linux Kernel Network Developers, Evan Jones, Nicolas Dichtel,
Phil Sutter, Toshiaki Makita, Cong Wang
In-Reply-To: <57251CB3.1040504@candelatech.com>
On Sat, Apr 30, 2016 at 4:59 PM, Ben Greear <greearb@candelatech.com> wrote:
>
>
> On 04/30/2016 12:54 PM, Tom Herbert wrote:
>>
>> We've put considerable effort into cleaning up the checksum interface
>> to make it as unambiguous as possible, please be very careful to
>> follow it. Broken checksum processing is really hard to detect and
>> debug.
>>
>> CHECKSUM_UNNECESSARY means that some number of _specific_ checksums
>> (indicated by csum_level) have been verified to be correct in a
>> packet. Blindly promoting CHECKSUM_NONE to CHECKSUM_UNNECESSARY is
>> never right. If CHECKSUM_UNNECESSARY is set in such a manner but the
>> checksum it would refer to has not been verified and is incorrect this
>> is a major bug.
>
>
> Suppose I know that the packet received on a packet-socket has
> already been verified by a NIC that supports hardware checksumming.
>
> Then, I want to transmit it on a veth interface using a second
> packet socket. I do not want veth to recalculate the checksum on
> transmit, nor to validate it on the peer veth on receive, because I do
> not want to waste the CPU cycles. I am assuming that my app is not
> accidentally corrupting frames, so the checksum can never be bad.
>
> How should the checksumming be configured for the packets going into
> the packet-socket from user-space?
It seems like that only the receiver should decide whether or not to
checksum packets on the veth, not the sender.
How about:
We could add a receiving socket option for "don't checksum packets
received from a veth when the other side has marked them as
elide-checksum-suggested" (similar to UDP_NOCHECKSUM), and a sending
socket option for "mark all data sent via this socket to a veth as
elide-checksum-suggested".
So the process would be:
Writer:
1. open read socket
2. open write socket, with option elide-checksum-for-veth-suggested
3. write data
Reader:
1. open read socket with "follow-elide-checksum-suggestions-on-veth"
2. read data
The kernel / module would then need to persist the flag on all packets
that traverse a veth, and drop these data when they leave the veth
module.
>
>
> Also, I might want to send raw frames that do have
> broken checksums (lets assume a real NIC, not veth), and I want them
> to hit the wire with those bad checksums.
>
>
> How do I configure the checksumming in this case?
Correct me if I'm wrong but I think this is already possible now. You
can have packets with incorrect checksum hitting the wire as is. What
you cannot do is instruct the receiving end to ignore the checksum
from the sending end when using a physical device (and something I
think we should mimic on the sending device).
>
>
>
> Thanks,
> Ben
>
>
>
>>
>> Tom
>>
>> On Sat, Apr 30, 2016 at 12:40 PM, Ben Greear <greearb@candelatech.com> wrote:
>>>
>>>
>>>
>>> On 04/30/2016 11:33 AM, Ben Hutchings wrote:
>>>>
>>>>
>>>> On Thu, 2016-04-28 at 12:29 +0200, Sabrina Dubroca wrote:
>>>>>
>>>>>
>>>>> Hello,
>>>
>>>
>>>
>>>>>>
>>>>>> http://dmz2.candelatech.com/?p=linux-4.4.dev.y/.git;a=commitdiff;h=8153e983c0e5eba1aafe1fc296248ed2a553f1ac;hp=454b07405d694dad52e7f41af5816eed0190da8a
>>>>>
>>>>>
>>>>> Actually, no, this is not really a regression.
>>>>
>>>>
>>>> [...]
>>>>
>>>> It really is. Even though the old behaviour was a bug (raw packets
>>>> should not be changed), if there are real applications that depend on
>>>> that then we have to keep those applications working somehow.
>>>
>>>
>>>
>>> To be honest, I fail to see why the old behaviour is a bug when sending
>>> raw packets from user-space. If raw packets should not be changed, then
>>> we need some way to specify what the checksum setting is to begin with,
>>> otherwise, user-space has not enough control.
>>>
>>> A socket option for new programs, and sysctl configurable defaults for raw
>>> sockets
>>> for old binary programs would be sufficient I think.
>>>
>>>
>>> Thanks,
>>> Ben
>>>
>>> --
>>> Ben Greear <greearb@candelatech.com>
>>> Candela Technologies Inc http://www.candelatech.com
>>
>>
>
> --
> Ben Greear <greearb@candelatech.com>
> Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: [PATCH 3.2 085/115] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good.
From: Ben Greear @ 2016-04-30 21:29 UTC (permalink / raw)
To: Vijay Pandurangan
Cc: Tom Herbert, Ben Hutchings, Sabrina Dubroca, Hannes Frederic Sowa,
LKML, stable, akpm, David S. Miller, Cong Wang,
Linux Kernel Network Developers, Evan Jones, Nicolas Dichtel,
Phil Sutter, Toshiaki Makita, Cong Wang
In-Reply-To: <CAKUBDd-_5RouHduHvjOqgOpyAArEBd80BEy4KBDPLRQoTtWb2Q@mail.gmail.com>
On 04/30/2016 02:13 PM, Vijay Pandurangan wrote:
> On Sat, Apr 30, 2016 at 4:59 PM, Ben Greear <greearb@candelatech.com> wrote:
>>
>>
>> On 04/30/2016 12:54 PM, Tom Herbert wrote:
>>>
>>> We've put considerable effort into cleaning up the checksum interface
>>> to make it as unambiguous as possible, please be very careful to
>>> follow it. Broken checksum processing is really hard to detect and
>>> debug.
>>>
>>> CHECKSUM_UNNECESSARY means that some number of _specific_ checksums
>>> (indicated by csum_level) have been verified to be correct in a
>>> packet. Blindly promoting CHECKSUM_NONE to CHECKSUM_UNNECESSARY is
>>> never right. If CHECKSUM_UNNECESSARY is set in such a manner but the
>>> checksum it would refer to has not been verified and is incorrect this
>>> is a major bug.
>>
>>
>> Suppose I know that the packet received on a packet-socket has
>> already been verified by a NIC that supports hardware checksumming.
>>
>> Then, I want to transmit it on a veth interface using a second
>> packet socket. I do not want veth to recalculate the checksum on
>> transmit, nor to validate it on the peer veth on receive, because I do
>> not want to waste the CPU cycles. I am assuming that my app is not
>> accidentally corrupting frames, so the checksum can never be bad.
>>
>> How should the checksumming be configured for the packets going into
>> the packet-socket from user-space?
>
>
> It seems like that only the receiver should decide whether or not to
> checksum packets on the veth, not the sender.
>
> How about:
>
> We could add a receiving socket option for "don't checksum packets
> received from a veth when the other side has marked them as
> elide-checksum-suggested" (similar to UDP_NOCHECKSUM), and a sending
> socket option for "mark all data sent via this socket to a veth as
> elide-checksum-suggested".
>
> So the process would be:
>
> Writer:
> 1. open read socket
> 2. open write socket, with option elide-checksum-for-veth-suggested
> 3. write data
>
> Reader:
> 1. open read socket with "follow-elide-checksum-suggestions-on-veth"
> 2. read data
>
> The kernel / module would then need to persist the flag on all packets
> that traverse a veth, and drop these data when they leave the veth
> module.
I'm not sure this works completely. In my app, the packet flow might be:
eth0 <-> raw-socket <-> user-space-bridge <-> raw-socket <-> vethA <-> vethB <-> [kernel router/bridge logic ...] <-> eth1
There may be no sockets on the vethB port. And reader/writer is not
a good way to look at it since I am implementing a bi-directional bridge in
user-space and each packet-socket is for both rx and tx.
>> Also, I might want to send raw frames that do have
>> broken checksums (lets assume a real NIC, not veth), and I want them
>> to hit the wire with those bad checksums.
>>
>>
>> How do I configure the checksumming in this case?
>
>
> Correct me if I'm wrong but I think this is already possible now. You
> can have packets with incorrect checksum hitting the wire as is. What
> you cannot do is instruct the receiving end to ignore the checksum
> from the sending end when using a physical device (and something I
> think we should mimic on the sending device).
Yes, it does work currently (or, last I checked)...I just want to make sure it keeps working.
Thanks,
Ben
--
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: [PATCH 3.2 085/115] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good.
From: Vijay Pandurangan @ 2016-04-30 21:36 UTC (permalink / raw)
To: Ben Greear
Cc: Tom Herbert, Ben Hutchings, Sabrina Dubroca, Hannes Frederic Sowa,
LKML, stable, akpm, David S. Miller, Cong Wang,
Linux Kernel Network Developers, Evan Jones, Nicolas Dichtel,
Phil Sutter, Toshiaki Makita, Cong Wang
In-Reply-To: <572523C4.4080307@candelatech.com>
On Sat, Apr 30, 2016 at 5:29 PM, Ben Greear <greearb@candelatech.com> wrote:
>
>
> On 04/30/2016 02:13 PM, Vijay Pandurangan wrote:
>>
>> On Sat, Apr 30, 2016 at 4:59 PM, Ben Greear <greearb@candelatech.com>
>> wrote:
>>>
>>>
>>>
>>> On 04/30/2016 12:54 PM, Tom Herbert wrote:
>>>>
>>>>
>>>> We've put considerable effort into cleaning up the checksum interface
>>>> to make it as unambiguous as possible, please be very careful to
>>>> follow it. Broken checksum processing is really hard to detect and
>>>> debug.
>>>>
>>>> CHECKSUM_UNNECESSARY means that some number of _specific_ checksums
>>>> (indicated by csum_level) have been verified to be correct in a
>>>> packet. Blindly promoting CHECKSUM_NONE to CHECKSUM_UNNECESSARY is
>>>> never right. If CHECKSUM_UNNECESSARY is set in such a manner but the
>>>> checksum it would refer to has not been verified and is incorrect this
>>>> is a major bug.
>>>
>>>
>>>
>>> Suppose I know that the packet received on a packet-socket has
>>> already been verified by a NIC that supports hardware checksumming.
>>>
>>> Then, I want to transmit it on a veth interface using a second
>>> packet socket. I do not want veth to recalculate the checksum on
>>> transmit, nor to validate it on the peer veth on receive, because I do
>>> not want to waste the CPU cycles. I am assuming that my app is not
>>> accidentally corrupting frames, so the checksum can never be bad.
>>>
>>> How should the checksumming be configured for the packets going into
>>> the packet-socket from user-space?
>>
>>
>>
>> It seems like that only the receiver should decide whether or not to
>> checksum packets on the veth, not the sender.
>>
>> How about:
>>
>> We could add a receiving socket option for "don't checksum packets
>> received from a veth when the other side has marked them as
>> elide-checksum-suggested" (similar to UDP_NOCHECKSUM), and a sending
>> socket option for "mark all data sent via this socket to a veth as
>> elide-checksum-suggested".
>>
>> So the process would be:
>>
>> Writer:
>> 1. open read socket
>> 2. open write socket, with option elide-checksum-for-veth-suggested
>> 3. write data
>>
>> Reader:
>> 1. open read socket with "follow-elide-checksum-suggestions-on-veth"
>> 2. read data
>>
>> The kernel / module would then need to persist the flag on all packets
>> that traverse a veth, and drop these data when they leave the veth
>> module.
>
>
> I'm not sure this works completely. In my app, the packet flow might be:
>
> eth0 <-> raw-socket <-> user-space-bridge <-> raw-socket <-> vethA <-> vethB
> <-> [kernel router/bridge logic ...] <-> eth1
Good point, so if you had:
eth0 <-> raw <-> user space-bridge <-> raw <-> vethA <-> veth B <->
userspace-stub <->eth1
and user-space hub enabled this elide flag, things would work, right?
Then, it seems like what we need is a way to tell the kernel
router/bridge logic to follow elide signals in packets coming from
veth. I'm not sure what the best way to do this is because I'm less
familiar with conventions in that part of the kernel, but assuming
there's a way to do this, would it be acceptable?
>
> There may be no sockets on the vethB port. And reader/writer is not
> a good way to look at it since I am implementing a bi-directional bridge in
> user-space and each packet-socket is for both rx and tx.
Sure, but we could model a bidrectional connection as two
unidirectional sockets for our discussions here, right?
>
>>> Also, I might want to send raw frames that do have
>>> broken checksums (lets assume a real NIC, not veth), and I want them
>>> to hit the wire with those bad checksums.
>>>
>>>
>>> How do I configure the checksumming in this case?
>>
>>
>>
>> Correct me if I'm wrong but I think this is already possible now. You
>> can have packets with incorrect checksum hitting the wire as is. What
>> you cannot do is instruct the receiving end to ignore the checksum
>> from the sending end when using a physical device (and something I
>> think we should mimic on the sending device).
>
>
> Yes, it does work currently (or, last I checked)...I just want to make sure
> it keeps working.
Yeah, good point. It would be great if we could write a test, or at
the very least, a list of invariants about what kinds of things should
work somewhere.
>
>
> Thanks,
> Ben
>
> --
> Ben Greear <greearb@candelatech.com>
> Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: New DSA design for cross-chip operations
From: Andrew Lunn @ 2016-04-30 21:38 UTC (permalink / raw)
To: Vivien Didelot
Cc: netdev, Florian Fainelli, David S. Miller, Jiri Pirko, kernel
In-Reply-To: <1570407369.390644.1461965624673.JavaMail.zimbra@savoirfairelinux.com>
On Fri, Apr 29, 2016 at 05:33:44PM -0400, Vivien Didelot wrote:
> Hi,
>
> Here's a proposal for a new design of DSA. It is meant to discuss the
> actual state and future implementation of the D (for distributed) in
> DSA, which consists of supporting several interconnected switch chips,
> exposed to the user as one big switch unit.
>
> There's still a lot of work to finish in DSA before running into
> cross-chip operations, but it'll help to start thinking about it now.
Hi Vivien
One thing we should keep in mind. Very few switch chips actually
support D in DSA. In fact, at the moment, i don't know of any other
than Marvell. The Broadcom SF2 does not. The Microchip LAN935X
don't. The Micrel KS8995 does not. Do any of the switches supported by
OpenWRT via swconfig?
We should try to limit adding complexity to the DSA core and the API
to the drivers which the majority of drivers will never need. I would
make use of the layering we already have. For the Marvell devices, add
a cross chip layer between the Marvell drivers and the DSA core. And
even then, we might want to keep this layer optional for Marvell
devices. There are only a few niche boards with multiple switch
chips. There are many more WiFi access points and cable modems, etc,
with just a single switch.
Andrew
^ permalink raw reply
* Re: [PATCH 3.2 085/115] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good.
From: Ben Greear @ 2016-04-30 21:52 UTC (permalink / raw)
To: Vijay Pandurangan
Cc: Tom Herbert, Ben Hutchings, Sabrina Dubroca, Hannes Frederic Sowa,
LKML, stable, akpm, David S. Miller, Cong Wang,
Linux Kernel Network Developers, Evan Jones, Nicolas Dichtel,
Phil Sutter, Toshiaki Makita, Cong Wang
In-Reply-To: <CAKUBDd8fksttZOW3T5zdKD4rzbcZzZer-DCMMsA2zKF4A_hXQw@mail.gmail.com>
On 04/30/2016 02:36 PM, Vijay Pandurangan wrote:
> On Sat, Apr 30, 2016 at 5:29 PM, Ben Greear <greearb@candelatech.com> wrote:
>>
>>
>> On 04/30/2016 02:13 PM, Vijay Pandurangan wrote:
>>>
>>> On Sat, Apr 30, 2016 at 4:59 PM, Ben Greear <greearb@candelatech.com>
>>> wrote:
>>>>
>>>>
>>>>
>>>> On 04/30/2016 12:54 PM, Tom Herbert wrote:
>>>>>
>>>>>
>>>>> We've put considerable effort into cleaning up the checksum interface
>>>>> to make it as unambiguous as possible, please be very careful to
>>>>> follow it. Broken checksum processing is really hard to detect and
>>>>> debug.
>>>>>
>>>>> CHECKSUM_UNNECESSARY means that some number of _specific_ checksums
>>>>> (indicated by csum_level) have been verified to be correct in a
>>>>> packet. Blindly promoting CHECKSUM_NONE to CHECKSUM_UNNECESSARY is
>>>>> never right. If CHECKSUM_UNNECESSARY is set in such a manner but the
>>>>> checksum it would refer to has not been verified and is incorrect this
>>>>> is a major bug.
>>>>
>>>>
>>>>
>>>> Suppose I know that the packet received on a packet-socket has
>>>> already been verified by a NIC that supports hardware checksumming.
>>>>
>>>> Then, I want to transmit it on a veth interface using a second
>>>> packet socket. I do not want veth to recalculate the checksum on
>>>> transmit, nor to validate it on the peer veth on receive, because I do
>>>> not want to waste the CPU cycles. I am assuming that my app is not
>>>> accidentally corrupting frames, so the checksum can never be bad.
>>>>
>>>> How should the checksumming be configured for the packets going into
>>>> the packet-socket from user-space?
>>>
>>>
>>>
>>> It seems like that only the receiver should decide whether or not to
>>> checksum packets on the veth, not the sender.
>>>
>>> How about:
>>>
>>> We could add a receiving socket option for "don't checksum packets
>>> received from a veth when the other side has marked them as
>>> elide-checksum-suggested" (similar to UDP_NOCHECKSUM), and a sending
>>> socket option for "mark all data sent via this socket to a veth as
>>> elide-checksum-suggested".
>>>
>>> So the process would be:
>>>
>>> Writer:
>>> 1. open read socket
>>> 2. open write socket, with option elide-checksum-for-veth-suggested
>>> 3. write data
>>>
>>> Reader:
>>> 1. open read socket with "follow-elide-checksum-suggestions-on-veth"
>>> 2. read data
>>>
>>> The kernel / module would then need to persist the flag on all packets
>>> that traverse a veth, and drop these data when they leave the veth
>>> module.
>>
>>
>> I'm not sure this works completely. In my app, the packet flow might be:
>>
>> eth0 <-> raw-socket <-> user-space-bridge <-> raw-socket <-> vethA <-> vethB
>> <-> [kernel router/bridge logic ...] <-> eth1
>
> Good point, so if you had:
>
> eth0 <-> raw <-> user space-bridge <-> raw <-> vethA <-> veth B <->
> userspace-stub <->eth1
>
> and user-space hub enabled this elide flag, things would work, right?
> Then, it seems like what we need is a way to tell the kernel
> router/bridge logic to follow elide signals in packets coming from
> veth. I'm not sure what the best way to do this is because I'm less
> familiar with conventions in that part of the kernel, but assuming
> there's a way to do this, would it be acceptable?
You cannot receive on one veth without transmitting on the other, so
I think the elide csum logic can go on the raw-socket, and apply to packets
in the transmit-from-user-space direction. Just allowing the socket to make
the veth behave like it used to before this patch in question should be good
enough, since that worked for us for years. So, just an option to modify the
ip_summed for pkts sent on a socket is probably sufficient.
>> There may be no sockets on the vethB port. And reader/writer is not
>> a good way to look at it since I am implementing a bi-directional bridge in
>> user-space and each packet-socket is for both rx and tx.
>
> Sure, but we could model a bidrectional connection as two
> unidirectional sockets for our discussions here, right?
Best not to I think, you want to make sure that one socket can
correctly handle tx and rx. As long as that works, then using
uni-directional sockets should work too.
Thanks,
Ben
--
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: [PATCH 3.2 085/115] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good.
From: Vijay Pandurangan @ 2016-04-30 22:01 UTC (permalink / raw)
To: Ben Greear
Cc: Tom Herbert, Ben Hutchings, Sabrina Dubroca, Hannes Frederic Sowa,
LKML, stable, akpm, David S. Miller, Cong Wang,
Linux Kernel Network Developers, Evan Jones, Nicolas Dichtel,
Phil Sutter, Toshiaki Makita, Cong Wang
In-Reply-To: <57252918.7070302@candelatech.com>
On Sat, Apr 30, 2016 at 5:52 PM, Ben Greear <greearb@candelatech.com> wrote:
>>
>> Good point, so if you had:
>>
>> eth0 <-> raw <-> user space-bridge <-> raw <-> vethA <-> veth B <->
>> userspace-stub <->eth1
>>
>> and user-space hub enabled this elide flag, things would work, right?
>> Then, it seems like what we need is a way to tell the kernel
>> router/bridge logic to follow elide signals in packets coming from
>> veth. I'm not sure what the best way to do this is because I'm less
>> familiar with conventions in that part of the kernel, but assuming
>> there's a way to do this, would it be acceptable?
>
>
> You cannot receive on one veth without transmitting on the other, so
> I think the elide csum logic can go on the raw-socket, and apply to packets
> in the transmit-from-user-space direction. Just allowing the socket to make
> the veth behave like it used to before this patch in question should be good
> enough, since that worked for us for years. So, just an option to modify
> the
> ip_summed for pkts sent on a socket is probably sufficient.
I don't think this is right. Consider:
- App A sends out corrupt packets 50% of the time and discards inbound data.
- App B doesn't care about corrupt packets and is happy to receive
them and has some way of dealing with them (special case)
- App C is a regular app, say nc or something.
In your world, where A decides what happens to data it transmits,
then
A<--veth-->B and A<---wire-->B will have the same behaviour
but
A<-- veth --> C and A<-- wire --> C will have _different_ behaviour: C
will behave incorrectly if it's connected over veth but correctly if
connected with a wire. That is a bug.
Since A cannot know what the app it's talking to will desire, I argue
that both sides of a message must be opted in to this optimization.
>
>>> There may be no sockets on the vethB port. And reader/writer is not
>>> a good way to look at it since I am implementing a bi-directional bridge
>>> in
>>> user-space and each packet-socket is for both rx and tx.
>>
>>
>> Sure, but we could model a bidrectional connection as two
>> unidirectional sockets for our discussions here, right?
>
>
> Best not to I think, you want to make sure that one socket can
> correctly handle tx and rx. As long as that works, then using
> uni-directional sockets should work too.
>
>
> Thanks,
> Ben
>
> --
> Ben Greear <greearb@candelatech.com>
> Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: [PATCH 3.2 085/115] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good.
From: Tom Herbert @ 2016-04-30 22:42 UTC (permalink / raw)
To: Ben Greear
Cc: Ben Hutchings, Sabrina Dubroca, Hannes Frederic Sowa, LKML,
stable, akpm, David S. Miller, Vijay Pandurangan, Cong Wang,
Linux Kernel Network Developers, Evan Jones, Nicolas Dichtel,
Phil Sutter, Toshiaki Makita, Cong Wang
In-Reply-To: <57251CB3.1040504@candelatech.com>
On Sat, Apr 30, 2016 at 1:59 PM, Ben Greear <greearb@candelatech.com> wrote:
>
> On 04/30/2016 12:54 PM, Tom Herbert wrote:
>>
>> We've put considerable effort into cleaning up the checksum interface
>> to make it as unambiguous as possible, please be very careful to
>> follow it. Broken checksum processing is really hard to detect and
>> debug.
>>
>> CHECKSUM_UNNECESSARY means that some number of _specific_ checksums
>> (indicated by csum_level) have been verified to be correct in a
>> packet. Blindly promoting CHECKSUM_NONE to CHECKSUM_UNNECESSARY is
>> never right. If CHECKSUM_UNNECESSARY is set in such a manner but the
>> checksum it would refer to has not been verified and is incorrect this
>> is a major bug.
>
>
> Suppose I know that the packet received on a packet-socket has
> already been verified by a NIC that supports hardware checksumming.
>
> Then, I want to transmit it on a veth interface using a second
> packet socket. I do not want veth to recalculate the checksum on
> transmit, nor to validate it on the peer veth on receive, because I do
> not want to waste the CPU cycles. I am assuming that my app is not
> accidentally corrupting frames, so the checksum can never be bad.
>
> How should the checksumming be configured for the packets going into
> the packet-socket from user-space?
>
Checksum unnecessary conversion to checksum complete should be done
and then the computed value can be sent to user space. If you want to
take advantage of the value on transmit then we would to change the
interface. But I'm not sure why recalculating the the checksum in the
host is even a problem. With all the advances in checksum offload,
LCO, RCO it seems like that shouldn't be a common case anyway.
> Also, I might want to send raw frames that do have
> broken checksums (lets assume a real NIC, not veth), and I want them
> to hit the wire with those bad checksums.
>
> How do I configure the checksumming in this case?
Just send raw packets with bad checksums (no checksum offload).
Tom
>
>
> Thanks,
> Ben
>
>
>
>>
>> Tom
>>
>> On Sat, Apr 30, 2016 at 12:40 PM, Ben Greear <greearb@candelatech.com>
>> wrote:
>>>
>>>
>>>
>>> On 04/30/2016 11:33 AM, Ben Hutchings wrote:
>>>>
>>>>
>>>> On Thu, 2016-04-28 at 12:29 +0200, Sabrina Dubroca wrote:
>>>>>
>>>>>
>>>>> Hello,
>>>
>>>
>>>
>>>>>>
>>>>>>
>>>>>> http://dmz2.candelatech.com/?p=linux-4.4.dev.y/.git;a=commitdiff;h=8153e983c0e5eba1aafe1fc296248ed2a553f1ac;hp=454b07405d694dad52e7f41af5816eed0190da8a
>>>>>
>>>>>
>>>>> Actually, no, this is not really a regression.
>>>>
>>>>
>>>> [...]
>>>>
>>>> It really is. Even though the old behaviour was a bug (raw packets
>>>> should not be changed), if there are real applications that depend on
>>>> that then we have to keep those applications working somehow.
>>>
>>>
>>>
>>> To be honest, I fail to see why the old behaviour is a bug when sending
>>> raw packets from user-space. If raw packets should not be changed, then
>>> we need some way to specify what the checksum setting is to begin with,
>>> otherwise, user-space has not enough control.
>>>
>>> A socket option for new programs, and sysctl configurable defaults for
>>> raw
>>> sockets
>>> for old binary programs would be sufficient I think.
>>>
>>>
>>> Thanks,
>>> Ben
>>>
>>> --
>>> Ben Greear <greearb@candelatech.com>
>>> Candela Technologies Inc http://www.candelatech.com
>>
>>
>
> --
> Ben Greear <greearb@candelatech.com>
> Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: [PATCH 3.2 085/115] veth: don’t modify ip_summed; doing so treats packets with bad checksums as good.
From: Ben Greear @ 2016-04-30 22:43 UTC (permalink / raw)
To: Vijay Pandurangan
Cc: Tom Herbert, Ben Hutchings, Sabrina Dubroca, Hannes Frederic Sowa,
LKML, stable, akpm, David S. Miller, Cong Wang,
Linux Kernel Network Developers, Evan Jones, Nicolas Dichtel,
Phil Sutter, Toshiaki Makita, Cong Wang
In-Reply-To: <CAKUBDd8S2=YcJ5wtM24D6-vw+bMud=Om9TBykMvSYcJ4X5tryw@mail.gmail.com>
On 04/30/2016 03:01 PM, Vijay Pandurangan wrote:
> On Sat, Apr 30, 2016 at 5:52 PM, Ben Greear <greearb@candelatech.com> wrote:
>>>
>>> Good point, so if you had:
>>>
>>> eth0 <-> raw <-> user space-bridge <-> raw <-> vethA <-> veth B <->
>>> userspace-stub <->eth1
>>>
>>> and user-space hub enabled this elide flag, things would work, right?
>>> Then, it seems like what we need is a way to tell the kernel
>>> router/bridge logic to follow elide signals in packets coming from
>>> veth. I'm not sure what the best way to do this is because I'm less
>>> familiar with conventions in that part of the kernel, but assuming
>>> there's a way to do this, would it be acceptable?
>>
>>
>> You cannot receive on one veth without transmitting on the other, so
>> I think the elide csum logic can go on the raw-socket, and apply to packets
>> in the transmit-from-user-space direction. Just allowing the socket to make
>> the veth behave like it used to before this patch in question should be good
>> enough, since that worked for us for years. So, just an option to modify
>> the
>> ip_summed for pkts sent on a socket is probably sufficient.
>
> I don't think this is right. Consider:
>
> - App A sends out corrupt packets 50% of the time and discards inbound data.
> - App B doesn't care about corrupt packets and is happy to receive
> them and has some way of dealing with them (special case)
> - App C is a regular app, say nc or something.
>
> In your world, where A decides what happens to data it transmits,
> then
> A<--veth-->B and A<---wire-->B will have the same behaviour
>
> but
>
> A<-- veth --> C and A<-- wire --> C will have _different_ behaviour: C
> will behave incorrectly if it's connected over veth but correctly if
> connected with a wire. That is a bug.
>
> Since A cannot know what the app it's talking to will desire, I argue
> that both sides of a message must be opted in to this optimization.
How can you make a generic app C know how to do this? The path could be,
for instance:
eth0 <-> user-space-A <-> vethA <-> vethB <-> { kernel routing logic } <-> vethC <-> vethD <-> appC
There are no sockets on vethB, but it does need to have special behaviour to elide
csums. Even if appC is hacked to know how to twiddle some thing on it's veth port,
mucking with vethD will have no effect on vethB.
With regard to your example above, why would A corrupt packets? My guess:
1) It has bugs (so, fix the bugs, it could equally create incorrect data with proper checksums,
so just enabling checksumming adds no useful protection.)
2) It means to corrupt frames. In that case, someone must expect that C should receive incorrect
frames, otherwise why bother making App-A corrupt them in the first place?
3) You are explicitly trying to test the kernel checksum logic, so you want the kernel to
detect the bad checksum and throw away the packet. In this case, just don't set the socket
option in appA to elide checksums and the packet will be thrown away.
Any other cases you can think of?
Thanks,
Ben
--
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* [PATCH] macb: fix mdiobus_scan() error check
From: Sergei Shtylyov @ 2016-04-30 22:47 UTC (permalink / raw)
To: netdev, nicolas.ferre
Now mdiobus_scan() returns ERR_PTR(-ENODEV) instead of NULL if the PHY
device ID was read as all ones. As this was not an error before, this
value should be filtered out now in this driver.
Fixes: b74766a0a0fe ("phylib: don't return NULL from get_phy_device()")
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
---
The patch is against DaveM's 'net-next.git' repo.
drivers/net/ethernet/cadence/macb.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
Index: net-next/drivers/net/ethernet/cadence/macb.c
===================================================================
--- net-next.orig/drivers/net/ethernet/cadence/macb.c
+++ net-next/drivers/net/ethernet/cadence/macb.c
@@ -458,7 +458,8 @@ static int macb_mii_init(struct macb *bp
struct phy_device *phydev;
phydev = mdiobus_scan(bp->mii_bus, i);
- if (IS_ERR(phydev)) {
+ if (IS_ERR(phydev) &&
+ PTR_ERR(phydev) != -ENODEV) {
err = PTR_ERR(phydev);
break;
}
^ permalink raw reply
* Re: [PATCH net-next 10/12] net/mlx5e: Create aRFS flow tables
From: Alexei Starovoitov @ 2016-05-01 2:18 UTC (permalink / raw)
To: Saeed Mahameed
Cc: David S. Miller, netdev, Or Gerlitz, Tal Alon, Eran Ben Elisha,
Maor Gottlieb
In-Reply-To: <1461883002-8912-11-git-send-email-saeedm@mellanox.com>
On Fri, Apr 29, 2016 at 01:36:40AM +0300, Saeed Mahameed wrote:
> From: Maor Gottlieb <maorg@mellanox.com>
>
> Create the following four flow tables for aRFS usage:
> 1. IPv4 TCP - filtering 4-tuple of IPv4 TCP packets.
> 2. IPv6 TCP - filtering 4-tuple of IPv6 TCP packets.
> 3. IPv4 UDP - filtering 4-tuple of IPv4 UDP packets.
> 4. IPv6 UDP - filtering 4-tuple of IPv6 UDP packets.
>
> Each flow table has two flow groups: one for the 4-tuple
> filtering (full match) and the other contains * rule for miss rule.
>
> Full match rule means a hit for aRFS and packet will be forwarded
> to the dedicated RQ/Core, miss rule packets will be forwarded to
> default RSS hashing.
>
> Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
> ---
> drivers/net/ethernet/mellanox/mlx5/core/Makefile | 1 +
> drivers/net/ethernet/mellanox/mlx5/core/en.h | 41 ++++
> drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c | 251 +++++++++++++++++++++
> drivers/net/ethernet/mellanox/mlx5/core/en_fs.c | 23 +-
> drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 8 +-
> drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 3 +-
> 6 files changed, 313 insertions(+), 14 deletions(-)
> create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c
>
> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
> index 4fc45ee..679e18f 100644
> --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
> +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
> @@ -9,3 +9,4 @@ mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o \
> en_txrx.o en_clock.o vxlan.o en_tc.o
>
> mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) += en_dcbnl.o
> +mlx5_core-$(CONFIG_RFS_ACCEL) += en_arfs.o
this line breaks the build when RFS is on
and CONFIG_MLX5_CORE_EN is off.
^ permalink raw reply
* Re: sfc: Please add rss_cpus=hyperthreads support
From: Andy Lutomirski @ 2016-05-01 2:27 UTC (permalink / raw)
To: Robert Stonehouse
Cc: Shradha Shah, Edward Cree, Network Development,
Solarflare linux maintainers
In-Reply-To: <5548CA3D.7010408@solarflare.com>
On Tue, May 5, 2015 at 6:48 AM, Robert Stonehouse
<rstonehouse@solarflare.com> wrote:
> On 02/05/15 01:03, Andy Lutomirski wrote:
>>
>> Can you please port the rss_cpus=hyperthreads feature from the
>> OpenOnload driver to the upstream driver? The fact that sticking:
>>
>> options sfc rss_cpus=hyperthreads
>>
>> into modprobe.conf causes sfc to fail to load when OpenOnload isn't
>> installed is incredibly annoying, and the option is genuinely useful.
>
>
> Sure; I will make sure that happens soon.
> I can see how that would be annoying.
Ping?
Thanks,
Andy
^ permalink raw reply
* Re: [PATCH 3.2 085/115] veth: don???t modify ip_summed; doing so treats packets with bad checksums as good.
From: Willy Tarreau @ 2016-05-01 5:30 UTC (permalink / raw)
To: Ben Greear
Cc: Vijay Pandurangan, Tom Herbert, Ben Hutchings, Sabrina Dubroca,
Hannes Frederic Sowa, LKML, stable, akpm, David S. Miller,
Cong Wang, Linux Kernel Network Developers, Evan Jones,
Nicolas Dichtel, Phil Sutter, Toshiaki Makita, Cong Wang
In-Reply-To: <57253527.7010009@candelatech.com>
On Sat, Apr 30, 2016 at 03:43:51PM -0700, Ben Greear wrote:
> On 04/30/2016 03:01 PM, Vijay Pandurangan wrote:
> > Consider:
> >
> > - App A sends out corrupt packets 50% of the time and discards inbound data.
(...)
> How can you make a generic app C know how to do this? The path could be,
> for instance:
>
> eth0 <-> user-space-A <-> vethA <-> vethB <-> { kernel routing logic } <-> vethC <-> vethD <-> appC
>
> There are no sockets on vethB, but it does need to have special behaviour to elide
> csums. Even if appC is hacked to know how to twiddle some thing on it's veth port,
> mucking with vethD will have no effect on vethB.
>
> With regard to your example above, why would A corrupt packets? My guess:
>
> 1) It has bugs (so, fix the bugs, it could equally create incorrect data with proper checksums,
> so just enabling checksumming adds no useful protection.)
I agree with Ben here, what he needs is the ability for userspace to be
trusted when *forwarding* a packet. Ideally you'd only want to receive
the csum status per packet on the packet socket and pass the same value
on the vethA interface, with this status being kept when the packet
reaches vethB.
If A purposely corrupts packet, it's A's problem. It's similar to designing
a NIC which intentionally corrupts packets and reports "checksum good".
The real issue is that in order to do things right, the userspace bridge
(here, "A") would really need to pass this status. In Ben's case as he
says, bad checksum packets are dropped before reaching A, so that
simplifies the process quite a bit and that might be what causes some
confusion, but ideally we'd rather have recvmsg() and sendmsg() with
these flags.
I faced the exact same issue 3 years ago when playing with netmap, it was
slow as hell because it would lose all checksum information when packets
were passing through userland, resulting in GRO/GSO etc being disabled,
and had to modify it to let userland preserve it. That's especially
important when you have to deal with possibly corrupted packets not yet
detected in the chain because the NIC did not validate their checksums.
Willy
^ permalink raw reply
* Re: [PATCH net-next 10/12] net/mlx5e: Create aRFS flow tables
From: Saeed Mahameed @ 2016-05-01 8:12 UTC (permalink / raw)
To: Alexei Starovoitov
Cc: Saeed Mahameed, David S. Miller, Linux Netdev List, Or Gerlitz,
Tal Alon, Eran Ben Elisha, Maor Gottlieb
In-Reply-To: <20160501021845.GA10348@ast-mbp.thefacebook.com>
On Sun, May 1, 2016 at 5:18 AM, Alexei Starovoitov
<alexei.starovoitov@gmail.com> wrote:
> On Fri, Apr 29, 2016 at 01:36:40AM +0300, Saeed Mahameed wrote:
>> From: Maor Gottlieb <maorg@mellanox.com>
>> mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) += en_dcbnl.o
>> +mlx5_core-$(CONFIG_RFS_ACCEL) += en_arfs.o
>
> this line breaks the build when RFS is on
> and CONFIG_MLX5_CORE_EN is off.
>
Will Fix.
Thank you Alexei.
^ permalink raw reply
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