* Re: Oops with latest (netfilter) nf-next tree, when unloading iptable_nat
From: Pablo Neira Ayuso @ 2012-09-14 12:07 UTC (permalink / raw)
To: Florian Westphal
Cc: Jesper Dangaard Brouer, netfilter-devel, netdev, yongjun_wei,
kaber
In-Reply-To: <20120912213627.GJ14750@breakpoint.cc>
On Wed, Sep 12, 2012 at 11:36:27PM +0200, Florian Westphal wrote:
> Jesper Dangaard Brouer <brouer@redhat.com> wrote:
>
> [ CC'd Patrick ]
>
> > I'm hitting this general protection fault, when unloading iptables_nat.
> > [ 524.591067] Pid: 5842, comm: modprobe Not tainted 3.6.0-rc3-pablo-nf-next+ #1 Red Hat KVM
> > [ 524.591067] RIP: 0010:[<ffffffffa002c2fd>] [<ffffffffa002c2fd>] nf_nat_proto_clean+0x6d/0xc0 [nf_nat]
> > [ 524.591067] RSP: 0018:ffff880073203e18 EFLAGS: 00010246
> > [ 524.591067] RAX: 0000000000000000 RBX: ffff880077dff2c8 RCX: ffff8800797fab70
> > [ 524.591067] RDX: dead000000200200 RSI: ffff880073203e88 RDI: ffffffffa002f208
> > [ 524.591067] RBP: ffff880073203e28 R08: ffff880073202000 R09: 0000000000000000
> > [ 524.591067] R10: dead000000200200 R11: dead000000100100 R12: ffffffff81c6dc00
> > list corruption? ^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^
>
> Yep, looks like it.
>
> > [ 524.591067] [<ffffffffa002c290>] ? nf_nat_net_exit+0x50/0x50 [nf_nat]
> > [ 524.591067] [<ffffffff815614e3>] nf_ct_iterate_cleanup+0xc3/0x170
> > [ 524.591067] [<ffffffffa002c54a>] nf_nat_l3proto_unregister+0x8a/0x100 [nf_nat]
> > [ 524.591067] [<ffffffff812a0303>] ? compat_prepare_timeout+0x13/0xb0
> > [ 524.591067] [<ffffffffa0035848>] nf_nat_l3proto_ipv4_exit+0x10/0x23 [nf_nat_ipv4]
>
> On module removal nf_nat_ipv4 calls nf_iterate_cleanup which invokes
> nf_nat_proto_clean() for each conntrack. That will then call
> hlist_del_rcu(&nat->bysource) using eachs conntracks nat ext area.
>
> Problem is that nf_nat_proto_clean() is called multiple times for the same
> conntrack:
> a) nf_ct_iterate_cleanup() returns each ct twice (origin, reply)
> b) we call it both for l3 and for l4 protocol ids
>
> We barf in hlist_del_rcu the 2nd time because ->pprev is poisoned.
>
> This was introduced with the ipv6 nat patches.
>
> --- a/net/netfilter/nf_nat_core.c
> +++ b/net/netfilter/nf_nat_core.c
> @@ -487,7 +487,7 @@ static int nf_nat_proto_clean(struct nf_conn *i, void *data)
>
> if (clean->hash) {
> spin_lock_bh(&nf_nat_lock);
> - hlist_del_rcu(&nat->bysource);
> + hlist_del_init_rcu(&nat->bysource);
> spin_unlock_bh(&nf_nat_lock);
> } else {
>
> Would probably avoid it. I guess it would be nicer to only call this
> once for each ct.
>
> Patrick, any other idea?
I already discussed this with Florian (I've been having problems with
two out of three of my email accounts this week... so I couldn't reply
to this email in the mailing list).
We can add nf_nat_iterate_cleanup that can iterate over the NAT
hashtable to replace current usage of nf_ct_iterate_cleanup.
^ permalink raw reply
* Re: [PATCH] netfilter: Allow xt_nat.c and x_tables.c to compiled in
From: Patrick McHardy @ 2012-09-14 11:54 UTC (permalink / raw)
To: Eric W. Biederman; +Cc: David Miller, netdev, netfilter-devel
In-Reply-To: <87627hfi69.fsf@xmission.com>
On Thu, 13 Sep 2012, Eric W. Biederman wrote:
> xt_init in x_tables.c must be called before xt_nat_init in xt_nat.c
> Reorder the makefile so that x_tables.o comes before xt_nat.o in
> netfilter.o.
>
> This allows me to built a kernel with both of these modules compiled in.
Thanks, we've already fixed that, the patch is queued in Pablo's tree.
^ permalink raw reply
* [PATCH net] batman-adv: make batadv_test_bit() return 0 or 1 only
From: Antonio Quartulli @ 2012-09-14 10:40 UTC (permalink / raw)
To: davem-fT/PcQaiUtIeIZ0/mPfg9Q
Cc: netdev-u79uwXL29TY76Z2rM5mHXA,
b.a.t.m.a.n-ZwoEplunGu2X36UT3dwllkB+6BGkLq7r
From: Linus Lüssing <linus.luessing-S0/GAf8tV78@public.gmane.org>
On some architectures test_bit() can return other values than 0 or 1:
With a generic x86 OpenWrt image in a kvm setup (batadv_)test_bit()
frequently returns -1 for me, leading to batadv_iv_ogm_update_seqnos()
wrongly signaling a protected seqno window.
This patch tries to fix this issue by making batadv_test_bit() return 0
or 1 only.
Signed-off-by: Linus Lüssing <linus.luessing-S0/GAf8tV78@public.gmane.org>
Acked-by: Sven Eckelmann <sven-KaDOiPu9UxWEi8DpZVb4nw@public.gmane.org>
Signed-off-by: Antonio Quartulli <ordex-GaUfNO9RBHfsrOwW+9ziJQ@public.gmane.org>
---
net/batman-adv/bitarray.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/net/batman-adv/bitarray.h b/net/batman-adv/bitarray.h
index a081ce1..cebaae7 100644
--- a/net/batman-adv/bitarray.h
+++ b/net/batman-adv/bitarray.h
@@ -20,8 +20,8 @@
#ifndef _NET_BATMAN_ADV_BITARRAY_H_
#define _NET_BATMAN_ADV_BITARRAY_H_
-/* returns true if the corresponding bit in the given seq_bits indicates true
- * and curr_seqno is within range of last_seqno
+/* Returns 1 if the corresponding bit in the given seq_bits indicates true
+ * and curr_seqno is within range of last_seqno. Otherwise returns 0.
*/
static inline int batadv_test_bit(const unsigned long *seq_bits,
uint32_t last_seqno, uint32_t curr_seqno)
@@ -32,7 +32,7 @@ static inline int batadv_test_bit(const unsigned long *seq_bits,
if (diff < 0 || diff >= BATADV_TQ_LOCAL_WINDOW_SIZE)
return 0;
else
- return test_bit(diff, seq_bits);
+ return test_bit(diff, seq_bits) != 0;
}
/* turn corresponding bit on, so we can remember that we got the packet */
--
1.7.12
^ permalink raw reply related
* [PATCH] i825xx: znet: fix compiler warnings when building a 64-bit kernel
From: Mika Westerberg @ 2012-09-14 10:33 UTC (permalink / raw)
To: netdev; +Cc: davem, Mika Westerberg
When building 64-bit kernel with this driver we get following warnings from
the compiler:
drivers/net/ethernet/i825xx/znet.c: In function ‘hardware_init’:
drivers/net/ethernet/i825xx/znet.c:863:29: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
drivers/net/ethernet/i825xx/znet.c:870:29: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
Fix these by calling isa_virt_to_bus() before passing the pointers to
set_dma_addr().
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
---
drivers/net/ethernet/i825xx/znet.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/i825xx/znet.c b/drivers/net/ethernet/i825xx/znet.c
index bd1f1ef..e843624 100644
--- a/drivers/net/ethernet/i825xx/znet.c
+++ b/drivers/net/ethernet/i825xx/znet.c
@@ -860,14 +860,14 @@ static void hardware_init(struct net_device *dev)
disable_dma(znet->rx_dma); /* reset by an interrupting task. */
clear_dma_ff(znet->rx_dma);
set_dma_mode(znet->rx_dma, DMA_RX_MODE);
- set_dma_addr(znet->rx_dma, (unsigned int) znet->rx_start);
+ set_dma_addr(znet->rx_dma, isa_virt_to_bus(znet->rx_start));
set_dma_count(znet->rx_dma, RX_BUF_SIZE);
enable_dma(znet->rx_dma);
/* Now set up the Tx channel. */
disable_dma(znet->tx_dma);
clear_dma_ff(znet->tx_dma);
set_dma_mode(znet->tx_dma, DMA_TX_MODE);
- set_dma_addr(znet->tx_dma, (unsigned int) znet->tx_start);
+ set_dma_addr(znet->tx_dma, isa_virt_to_bus(znet->tx_start));
set_dma_count(znet->tx_dma, znet->tx_buf_len<<1);
enable_dma(znet->tx_dma);
release_dma_lock(flags);
--
1.7.10.4
^ permalink raw reply related
* tulip Ethernet driver messes up USB keyboard
From: Marti Raudsepp @ 2012-09-14 9:49 UTC (permalink / raw)
To: Kernel hackers, Linux USB list, Linux network list
In-Reply-To: <CABRT9RDdbqz6iscOz7BYwEQPOo8xfstfUZyBqVw44gmPsauLTg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
Hi lists,
After installing an old 100Mbit PCI Ethernet card to my machine, it has
complained a few times about spurious interrupts ("nobody cared") at a
random time of the day. After the oops is reported, my USB keyboard (HP
smart card keyboard) stops working properly -- it has lots of delay, it
misses some keystrokes and also causes "stuck keys". Strangely enough my USB
mouse continues to work without problems. Apparently the USB controller and
the Ethernet card share an interrupt line. Un-/replugging the keyboard does
not help.
So far I have simply rebooted to fix the issue. I haven't been able to
reproduce this on will despite my best attempts.
The Ethernet card is driven by the tulip driver and is listed as "ADMtek
NC100 Network Everywhere Fast Ethernet 10/100 (rev 11)" in lspci. My
motherboard uses the Intel H77 chipset (ASRock H77 Pro4/MVP)
Here are two slightly different examples of the oops report; the effect is
the same:
kernel: irq 16: nobody cared (try booting with the "irqpoll" option)
kernel: Pid: 0, comm: swapper/2 Not tainted 3.5.3-1-ARCH #1
kernel: Call Trace:
kernel: <IRQ> [<ffffffff810d4d3d>] __report_bad_irq+0x3d/0xe0
kernel: [<ffffffff810d5033>] note_interrupt+0x1a3/0x1f0
kernel: [<ffffffff810d292f>] handle_irq_event_percpu+0xbf/0x260
kernel: [<ffffffff810d2b18>] handle_irq_event+0x48/0x70
kernel: [<ffffffff810d5b4a>] handle_fasteoi_irq+0x5a/0x100
kernel: [<ffffffff810160c2>] handle_irq+0x22/0x40
kernel: [<ffffffff81484cea>] do_IRQ+0x5a/0xe0
kernel: [<ffffffff8147c12a>] common_interrupt+0x6a/0x6a
kernel: <EOI> [<ffffffffa02d8f01>] ? acpi_idle_enter_c1+0xda/0x104
[processor]
kernel: [<ffffffffa02d8edc>] ? acpi_idle_enter_c1+0xb5/0x104 [processor]
kernel: [<ffffffff8134f159>] cpuidle_enter+0x19/0x20
kernel: [<ffffffff8134f7a6>] cpuidle_idle_call+0xa6/0x330
kernel: [<ffffffff8101daaf>] cpu_idle+0xbf/0x130
kernel: [<ffffffff8146a19b>] start_secondary+0x203/0x20a
kernel: handlers:
kernel: [<ffffffffa012c0b0>] usb_hcd_irq [usbcore]
kernel: [<ffffffffa023b1c0>] tulip_interrupt [tulip]
kernel: Disabling IRQ #16
kernel: irq 16: nobody cared (try booting with the "irqpoll" option)
kernel: Pid: 0, comm: swapper/0 Not tainted 3.5.3-1-ARCH #1
kernel: Call Trace:
kernel: <IRQ> [<ffffffff810d4d3d>] __report_bad_irq+0x3d/0xe0
kernel: [<ffffffff810d5033>] note_interrupt+0x1a3/0x1f0
kernel: [<ffffffff810d292f>] handle_irq_event_percpu+0xbf/0x260
kernel: [<ffffffff810d2b18>] handle_irq_event+0x48/0x70
kernel: [<ffffffff810d5b4a>] handle_fasteoi_irq+0x5a/0x100
kernel: [<ffffffff810160c2>] handle_irq+0x22/0x40
kernel: [<ffffffff81484cea>] do_IRQ+0x5a/0xe0
kernel: [<ffffffff8147c12a>] common_interrupt+0x6a/0x6a
kernel: <EOI> [<ffffffffa01b6f01>] ? acpi_idle_enter_c1+0xda/0x104
[processor]
kernel: [<ffffffffa01b6edc>] ? acpi_idle_enter_c1+0xb5/0x104 [processor]
kernel: [<ffffffff8134f159>] cpuidle_enter+0x19/0x20
kernel: [<ffffffff8134f7a6>] cpuidle_idle_call+0xa6/0x330
kernel: [<ffffffff8101daaf>] cpu_idle+0xbf/0x130
kernel: [<ffffffff81456fbc>] rest_init+0x80/0x84
kernel: [<ffffffff818bbc35>] start_kernel+0x3c1/0x3ce
kernel: [<ffffffff818bb673>] ? repair_env_string+0x5e/0x5e
kernel: [<ffffffff818bb356>] x86_64_start_reservations+0x131/0x135
kernel: [<ffffffff818bb45a>] x86_64_start_kernel+0x100/0x10f
kernel: handlers:
kernel: [<ffffffffa012c0b0>] usb_hcd_irq [usbcore]
kernel: [<ffffffffa02b41c0>] tulip_interrupt [tulip]
kernel: Disabling IRQ #16
These are after a fresh reboot...
% lspci |grep Eth
03:00.0 Ethernet controller: ADMtek NC100 Network Everywhere Fast Ethernet
10/100 (rev 11)
04:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B
PCI Express Gigabit Ethernet controller (rev 06)
% dmesg |egrep -i 'eth|tulip'
[ 7.868120] tulip: Linux Tulip driver version 1.1.15 (Feb 27, 2007)
[ 7.868146] tulip: tulip_init_one: Enabled WOL support for AN983B
[ 7.870487] tulip0: MII transceiver #1 config 1000 status 786d
advertising 05e1
[ 7.893274] net eth0: ADMtek Comet rev 17 at Port 0xe000,
00:06:25:48:7f:16, IRQ 16
[ 7.902994] systemd-udevd[162]: renamed network interface eth0 to eth1
[ 7.989822] r8169 Gigabit Ethernet driver 2.3LK-NAPI loaded
[ 7.990148] r8169 0000:04:00.0: eth0: RTL8168evl/8111evl at
0xffffc90005798000, bc:5f:f4:44:29:6c, XID 0c900800 IRQ 47
[ 7.990150] r8169 0000:04:00.0: eth0: jumbo features [frames: 9200 bytes,
tx checksumming: ko]
[ 16.032206] r8169 0000:04:00.0: eth0: link down
[ 16.032568] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
[ 18.287417] r8169 0000:04:00.0: eth0: link up
[ 18.287761] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[ 18.437518] tulip 0000:03:00.0: eth1: tulip_stop_rxtx() failed (CSR5
0xfc664010 CSR6 0xff972113)
[ 18.437530] net eth1: Setting full-duplex based on MII#1 link partner
capability of 45e1
Somewhat suspect is the "tulip_stop_rxtx() failed" line. I also get that
line every time I stop the interface (using ip link set eth1 down):
Sep 14 12:40:11 wrx kernel: tulip 0000:03:00.0: eth1: tulip_stop_rxtx()
failed (CSR5 0xfc664010 CSR6 0xff972113)
Sep 14 12:40:11 wrx kernel: net eth1: Setting full-duplex based on MII#1
link partner capability of 45e1
Sep 14 12:40:12 wrx NetworkManager[386]: <info> (eth1): carrier now OFF
(device state 100, deferring action for 4 seconds)
Sep 14 12:40:12 wrx kernel: tulip 0000:03:00.0: eth1: tulip_stop_rxtx()
failed (CSR5 0xfc07c013 CSR6 0xff970111)
% cat /proc/interrupts
CPU0 CPU1 CPU2 CPU3
0: 48 0 0 0 IO-APIC-edge timer
1: 2 1 0 0 IO-APIC-edge i8042
8: 164 5 2 0 IO-APIC-edge rtc0
9: 0 0 0 0 IO-APIC-fasteoi acpi
12: 3 0 1 0 IO-APIC-edge i8042
16: 141 346 8067481 2394 IO-APIC-fasteoi
ehci_hcd:usb1, eth1
23: 319 18623 7406 31659 IO-APIC-fasteoi
ehci_hcd:usb2
40: 0 0 0 0 PCI-MSI-edge PCIe
PME
41: 0 0 0 0 PCI-MSI-edge PCIe
PME
42: 0 0 0 0 PCI-MSI-edge PCIe
PME
43: 22382 61895 31816 10217 PCI-MSI-edge ahci
44: 0 0 0 0 PCI-MSI-edge ahci
45: 0 0 0 0 PCI-MSI-edge
xhci_hcd
46: 12 1 0 0 PCI-MSI-edge mei
47: 6 2 2 32904 PCI-MSI-edge eth0
48: 75136 39014 54183 22566 PCI-MSI-edge i915
49: 267 295 254 57 PCI-MSI-edge
snd_hda_intel
NMI: 40906 21860 26792 17063 Non-maskable interrupts
LOC: 362056 362594 377195 364303 Local timer interrupts
SPU: 0 0 0 0 Spurious interrupts
PMI: 40906 21860 26792 17063 Performance monitoring
interrupts
IWI: 6 3 3 2 IRQ work interrupts
RTR: 3 0 0 0 APIC ICR read retries
RES: 241774 18622 10902 4227 Rescheduling interrupts
CAL: 108 116 119 139 Function call interrupts
TLB: 26279 11870 16395 25314 TLB shootdowns
TRM: 0 0 0 0 Thermal event interrupts
THR: 0 0 0 0 Threshold APIC interrupts
MCE: 0 0 0 0 Machine check exceptions
MCP: 11 11 11 11 Machine check polls
ERR: 0
MIS: 0
Regards,
Marti
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [RFC PATCH net-next v2 0/1] Add support of ECMPv6
From: Vincent Bernat @ 2012-09-14 9:40 UTC (permalink / raw)
To: Nicolas Dichtel; +Cc: yoshfuji, netdev, davem
In-Reply-To: <1347609548-14494-1-git-send-email-nicolas.dichtel@6wind.com>
❦ 14 septembre 2012 09:59 CEST, Nicolas Dichtel <nicolas.dichtel@6wind.com> :
> Here is an example of a command to add an ECMP route:
> $ ip -6 route add 3ffe:304:124:2306::/64 \
> nexthop via fe80::230:1bff:feb4:e05c dev eth0 weight 1 \
> nexthop via fe80::230:1bff:feb4:dd4f dev eth0 weight 1
When displaying ECMP routes, the display is different than for IPv4: we
get two distinct routes instead of an ECMP route (with nexthop
keyword).
With IPv4:
193.252.X.X/26 proto zebra metric 20
nexthop via 193.252.X.X dev bae1 weight 1
nexthop via 193.252.X.X dev bae2 weight 1
With IPv6:
2a01:c9c0:X:X::/64 via fe80::215:17ff:fe85:76b9 dev bae1 metric 11
2a01:c9c0:X:X::/64 via fe80::222:91ff:fe4e:b000 dev bae2 metric 11
If I capture the netlink message from the add command, put it in a file
and use "ip monitor file ...", I see this:
2a01:c9c0:X:X::/64
nexthop via fe80::215:17ff:fe85:76b9 dev if12 weight 1
nexthop via fe80::222:91ff:fe4e:b000 dev if11 weight 1
Therefore, the problem is not in iproute2 which knows how to display
those ECMP routes. I fear that this difference make support in routing
daemons more difficult.
--
Make the coupling between modules visible.
- The Elements of Programming Style (Kernighan & Plauger)
^ permalink raw reply
* Re: [PATCH v2] netfilter: take care of timewait sockets
From: Sami Farin @ 2012-09-14 9:30 UTC (permalink / raw)
To: Eric Dumazet; +Cc: Florian Westphal, Sami Farin, netdev
In-Reply-To: <1346629684.2563.78.camel@edumazet-glaptop>
On Mon, Sep 03, 2012 at 01:48:04 +0200, Eric Dumazet wrote:
> Sami Farin reported crashes in xt_LOG because it assumes skb->sk is a
> full blown socket.
>
> But with TCP early demux, we can have skb->sk pointing to a timewait
> socket.
>
> Same fix is needed in netfnetlink_log
with ip_early_demux=0 and without these patches I did not get panics for
7 days, and with the patches and ip_early_demux=1 it has worked OK for 5 days,
thanks.
--
Do what you love because life is too short for anything else.
^ permalink raw reply
* [PATCH v2 5/7] dt/net/fsl_pq_mdio: Use of_get_child_by_name to get a named child.
From: Srinivas KANDAGATLA @ 2012-09-14 8:19 UTC (permalink / raw)
To: afleming
Cc: davem, netdev, devicetree-discuss, robherring2,
srinivas.kandagatla, grant.likely
From: Srinivas Kandagatla <srinivas.kandagatla@st.com>
As follow-up to "dt: introduce of_get_child_by_name to get child node by
name." patch, This patch removes some of the code duplication in the
driver by replacing it with of_get_child_by_name instead.
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@st.com>
---
drivers/net/ethernet/freescale/fsl_pq_mdio.c | 5 +----
1 files changed, 1 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/freescale/fsl_pq_mdio.c b/drivers/net/ethernet/freescale/fsl_pq_mdio.c
index 9527b28..e710c60 100644
--- a/drivers/net/ethernet/freescale/fsl_pq_mdio.c
+++ b/drivers/net/ethernet/freescale/fsl_pq_mdio.c
@@ -355,10 +355,7 @@ static int fsl_pq_mdio_probe(struct platform_device *ofdev)
goto err_free_irqs;
}
- for_each_child_of_node(np, tbi) {
- if (!strncmp(tbi->type, "tbi-phy", 8))
- break;
- }
+ tbi = of_get_child_by_name(np, "tbi-phy");
if (tbi) {
const u32 *prop = of_get_property(tbi, "reg", NULL);
--
1.7.0.4
^ permalink raw reply related
* [PATCH v2 1/7] dt: introduce of_get_child_by_name to get child node by name.
From: Srinivas KANDAGATLA @ 2012-09-14 8:17 UTC (permalink / raw)
To: davem, robherring2, bergner, devicetree-discuss
Cc: benh, linuxppc-dev, netdev, jassi.brar, afleming,
srinivas.kandagatla, grant.likely
From: Srinivas Kandagatla <srinivas.kandagatla@st.com>
This patch introduces of_get_child_by_name function to get a child node
by its name in a given parent node.
Without this patch each driver code has to iterate the parent and do
a string compare, However having of_get_child_by_name libary function would
avoid code duplication, errors and is more convenient.
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@st.com>
---
drivers/of/base.c | 25 +++++++++++++++++++++++++
include/linux/of.h | 2 ++
2 files changed, 27 insertions(+), 0 deletions(-)
diff --git a/drivers/of/base.c b/drivers/of/base.c
index d4a1c9a..7391f8a 100644
--- a/drivers/of/base.c
+++ b/drivers/of/base.c
@@ -391,6 +391,31 @@ struct device_node *of_get_next_available_child(const struct device_node *node,
EXPORT_SYMBOL(of_get_next_available_child);
/**
+ * of_get_child_by_name - Find the child node by name for a given parent
+ * @node: parent node
+ * @name: child name to look for.
+ *
+ * This function looks for child node for given matching name
+ *
+ * Returns a node pointer if found, with refcount incremented, use
+ * of_node_put() on it when done.
+ * Returns NULL if node is not found.
+ */
+
+struct device_node *of_get_child_by_name(const struct device_node *node,
+ const char *name)
+{
+ struct device_node *child;
+
+ for_each_child_of_node(node, child)
+ if (child->name && (of_node_cmp(child->name, name) == 0)
+ && of_node_get(child))
+ break;
+ return child;
+}
+EXPORT_SYMBOL(of_get_child_by_name);
+
+/**
* of_find_node_by_path - Find a node matching a full OF path
* @path: The full path to match
*
diff --git a/include/linux/of.h b/include/linux/of.h
index 1b11632..7b8e3cd 100644
--- a/include/linux/of.h
+++ b/include/linux/of.h
@@ -192,6 +192,8 @@ extern struct device_node *of_get_next_child(const struct device_node *node,
struct device_node *prev);
extern struct device_node *of_get_next_available_child(
const struct device_node *node, struct device_node *prev);
+extern struct device_node *of_get_child_by_name(const struct device_node *node,
+ const char *name);
#define for_each_child_of_node(parent, child) \
for (child = of_get_next_child(parent, NULL); child != NULL; \
--
1.7.0.4
^ permalink raw reply related
* [PATCH v2 0/7] Introduce of_get_child_by_name.
From: Srinivas KANDAGATLA @ 2012-09-14 8:17 UTC (permalink / raw)
To: davem, robherring2, bergner, devicetree-discuss
Cc: benh, linuxppc-dev, netdev, jassi.brar, afleming,
srinivas.kandagatla, grant.likely
From: Srinivas Kandagatla <srinivas.kandagatla@st.com>
This patch series introduces of_get_child_by_name function to get a
child node by its name in a given parent node and also removes code
duplication in some of the existing driver code by using of_get_child_by_name.
Normally if a driver want to get a child node it would iterate all the nodes
of parent and compare its name and then get it.
This use case is becoming common as device trees are used more, so moving this
functionality to a libary function makes sense.
Having of_get_child_by_name libary function would avoid code duplication,
errors and is more convenient.
Changes from v1:
-rename of_get_child to of_get_child_by_name.
-remove read lock in of_get_child_by_name.
-make use of of_get_child_by_name in the existing kernel code.
Srinivas Kandagatla (7):
dt: introduce of_get_child_by_name to get child node by name.
dt/powerpc: Use of_get_child_by_name to get a named child.
dt/powerpc/powernv: Use of_get_child_by_name to get a named child.
dt/powerpc/sysdev: Use of_get_child_by_name to get a named child.
dt/net/fsl_pq_mdio: Use of_get_child_by_name to get a named child.
dt/s3c64xx/spi: Use of_get_child_by_name to get a named child.
dt/tty/opal: Use of_get_child_by_name to get a named child.
arch/powerpc/kernel/prom.c | 5 +----
arch/powerpc/platforms/powernv/opal.c | 6 ++----
arch/powerpc/sysdev/qe_lib/qe.c | 5 +----
drivers/net/ethernet/freescale/fsl_pq_mdio.c | 5 +----
drivers/of/base.c | 25 +++++++++++++++++++++++++
drivers/spi/spi-s3c64xx.c | 4 +---
drivers/tty/hvc/hvc_opal.c | 9 ++-------
include/linux/of.h | 2 ++
8 files changed, 35 insertions(+), 26 deletions(-)
^ permalink raw reply
* Re[2]: [PATCH V3 0/8] ipvs: IPv6 fragment handling for IPVS
From: Julian Anastasov @ 2012-09-14 8:23 UTC (permalink / raw)
To: Hans Schillstrom
Cc: Jesper Dangaard Brouer, Hans Schillstrom, netdev, Patrick McHardy,
Pablo Neira Ayuso, lvs-devel, Thomas Graf, Wensong Zhang,
netfilter-devel, Simon Horman
In-Reply-To: <ee5vnhe.0891bc4ff9e38f64c4c5c0da6c50af1c@obelix.schillstrom.com>
Hello,
On Thu, 13 Sep 2012, Hans Schillstrom wrote:
> > I also see that we should not send ICMP
> >errors (FRAG NEEDED/TOO BIG) in response to large
> >ICMP error packets but it is not related to your changes,
> >it needs separate change to all transmitters.
>
> Just a curious question, have you seen such big ICMP packets in real life ?
> (if you have normal MTU:s)
No, may be it can happen when PMTU reaches minimum.
Regards
--
Julian Anastasov <ja@ssi.bg>
^ permalink raw reply
* [RFC PATCH net-next v2 1/1] ipv6: add support of ECMP
From: Nicolas Dichtel @ 2012-09-14 7:59 UTC (permalink / raw)
To: yoshfuji; +Cc: bernat, netdev, davem, Nicolas Dichtel
In-Reply-To: <1347609548-14494-1-git-send-email-nicolas.dichtel@6wind.com>
This patch adds the support of equal cost multipath for IPv6.
The patch is based on a previous work from
Luc Saillard <luc.saillard@6wind.com>.
Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
---
include/net/ip6_fib.h | 13 ++++
net/ipv6/Kconfig | 33 ++++++++
net/ipv6/ip6_fib.c | 73 ++++++++++++++++++
net/ipv6/route.c | 209 +++++++++++++++++++++++++++++++++++++++++++++++++-
4 files changed, 325 insertions(+), 3 deletions(-)
diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index cd64cf3..37e502a 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -47,6 +47,10 @@ struct fib6_config {
unsigned long fc_expires;
struct nlattr *fc_mx;
int fc_mx_len;
+#ifdef CONFIG_IPV6_MULTIPATH
+ struct nlattr *fc_mp;
+ int fc_mp_len;
+#endif
struct nl_info fc_nlinfo;
};
@@ -98,6 +102,15 @@ struct rt6_info {
struct fib6_node *rt6i_node;
struct in6_addr rt6i_gateway;
+#ifdef CONFIG_IPV6_MULTIPATH
+ /*
+ * siblings is a list of rt6_info that have the the same metric/weight,
+ * destination, but not the same gateway. nsiblings is just a cache
+ * to speed up lookup.
+ */
+ unsigned int rt6i_nsiblings;
+ struct list_head rt6i_siblings;
+#endif
atomic_t rt6i_ref;
diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index 4f7fe72..e0c92dc 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -266,4 +266,37 @@ config IPV6_PIMSM_V2
Support for IPv6 PIM multicast routing protocol PIM-SMv2.
If unsure, say N.
+config IPV6_MULTIPATH
+ bool "IPv6: equal cost multipath for IPv6 routing"
+ depends on IPV6
+ default y
+ ---help---
+ Enable this option to support ECMP for IPv6.
+
+choice
+ prompt "IPv6: choose Multipath algorithm"
+ depends on IPV6_MULTIPATH
+ default IPV6_MULTIPATH_HASH
+ ---help---
+ Define the method to select route between each possible path.
+ The recommanded algorithm (by RFC4311) is HASH method.
+
+ config IPV6_MULTIPATH_HASH
+ bool "IPv6: MULTIPATH hash/flow algorithm"
+ ---help---
+ Multipath routes are chosen according to hash of packet header to
+ ensure a flow keeps the same route.
+ This algorithm is recommanded by RFC4311.
+
+ config IPV6_MULTIPATH_RR
+ bool "IPv6: MULTIPATH round robin algorithm"
+ ---help---
+ Multipath routes are chosen according to Round Robin.
+
+ config IPV6_MULTIPATH_RANDOM
+ bool "IPv6: MULTIPATH random algorithm"
+ ---help---
+ Multipath routes are chosen in a random fashion.
+endchoice
+
endif # IPV6
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 13690d6..3541e44 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -672,6 +672,10 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
iter->rt6i_idev == rt->rt6i_idev &&
ipv6_addr_equal(&iter->rt6i_gateway,
&rt->rt6i_gateway)) {
+#ifdef CONFIG_IPV6_MULTIPATH
+ if (rt->rt6i_nsiblings)
+ rt->rt6i_nsiblings = 0;
+#endif
if (!(iter->rt6i_flags & RTF_EXPIRES))
return -EEXIST;
if (!(rt->rt6i_flags & RTF_EXPIRES))
@@ -680,6 +684,23 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
rt6_set_expires(iter, rt->dst.expires);
return -EEXIST;
}
+#ifdef CONFIG_IPV6_MULTIPATH
+ /* If we have the same destination and the same metric,
+ * but not the same gateway, then the route we try to
+ * add is sibling to this route, increment our counter
+ * of siblings, and later we will add our route to the
+ * list.
+ * Only static routes (which don't have flag
+ * RTF_EXPIRES) are used for ECMPv6.
+ *
+ * To avoid long list, we only had siblings if the
+ * route have a gateway.
+ */
+ if (rt->rt6i_flags & RTF_GATEWAY &&
+ !(rt->rt6i_flags & RTF_EXPIRES) &&
+ !(iter->rt6i_flags & RTF_EXPIRES))
+ rt->rt6i_nsiblings++;
+#endif
}
if (iter->rt6i_metric > rt->rt6i_metric)
@@ -692,6 +713,43 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
if (ins == &fn->leaf)
fn->rr_ptr = NULL;
+#ifdef CONFIG_IPV6_MULTIPATH
+ /* Link this route to others same route. */
+ if (rt->rt6i_nsiblings) {
+ unsigned int rt6i_nsiblings;
+ struct rt6_info *sibling, *temp_sibling;
+
+ /* Find the first route that have the same metric */
+ sibling = fn->leaf;
+ while (sibling) {
+ if (sibling->rt6i_metric == rt->rt6i_metric) {
+ list_add_tail(&rt->rt6i_siblings,
+ &sibling->rt6i_siblings);
+ break;
+ }
+ sibling = sibling->dst.rt6_next;
+ }
+ /* For each sibling in the list, increment the counter of
+ * siblings. We can check if all the counter are equal.
+ */
+ rt6i_nsiblings = 0;
+ list_for_each_entry_safe(sibling, temp_sibling,
+ &rt->rt6i_siblings,
+ rt6i_siblings) {
+ sibling->rt6i_nsiblings++;
+ if (unlikely(sibling->rt6i_nsiblings !=
+ rt->rt6i_nsiblings)) {
+ pr_err("Wrong number of siblings for route %p (%d)\n",
+ sibling, sibling->rt6i_nsiblings);
+ }
+ rt6i_nsiblings++;
+ }
+ if (unlikely(rt6i_nsiblings != rt->rt6i_nsiblings)) {
+ pr_err("Wrong number of siblings for route %p. I have %d routes, but count %d siblings\n",
+ rt, rt6i_nsiblings, rt->rt6i_nsiblings);
+ }
+ }
+#endif
/*
* insert node
*/
@@ -1197,6 +1255,21 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
if (fn->rr_ptr == rt)
fn->rr_ptr = NULL;
+#ifdef CONFIG_IPV6_MULTIPATH
+ /* Remove this entry from other siblings */
+ if (rt->rt6i_nsiblings) {
+ struct rt6_info *sibling, *next_sibling;
+
+ /* For each siblings, decrement the counter of siblings */
+ list_for_each_entry_safe(sibling, next_sibling,
+ &rt->rt6i_siblings, rt6i_siblings) {
+ sibling->rt6i_nsiblings--;
+ }
+ rt->rt6i_nsiblings = 0;
+ list_del_init(&rt->rt6i_siblings);
+ }
+#endif
+
/* Adjust walkers */
read_lock(&fib6_walker_lock);
FOR_WALKERS(w) {
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 399613b..431f7ad 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -57,6 +57,9 @@
#include <net/xfrm.h>
#include <net/netevent.h>
#include <net/netlink.h>
+#ifdef CONFIG_IPV6_MULTIPATH
+#include <net/nexthop.h>
+#endif
#include <asm/uaccess.h>
@@ -288,6 +291,10 @@ static inline struct rt6_info *ip6_dst_alloc(struct net *net,
memset(dst + 1, 0, sizeof(*rt) - sizeof(*dst));
rt6_init_peer(rt, table ? &table->tb6_peers : net->ipv6.peers);
+#ifdef CONFIG_IPV6_MULTIPATH
+ INIT_LIST_HEAD(&rt->rt6i_siblings);
+ rt->rt6i_nsiblings = 0;
+#endif
}
return rt;
}
@@ -388,6 +395,124 @@ static bool rt6_need_strict(const struct in6_addr *daddr)
(IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL | IPV6_ADDR_LOOPBACK);
}
+#ifdef CONFIG_IPV6_MULTIPATH
+/*
+ * Multipath route selection.
+ */
+
+#ifdef CONFIG_IPV6_MULTIPATH_RANDOM
+/*
+ * Pseudo random candidate function
+ */
+static int rt6_info_hash_randomfn(unsigned int candidate_count)
+{
+ return random32() % candidate_count;
+}
+#endif
+
+#ifdef CONFIG_IPV6_MULTIPATH_RR
+/*
+ * Fake Round Robin candidate function
+ * If we want real RR, we need to add a counter in each route
+ */
+static int rt6_info_hash_falserr(unsigned int candidate_count)
+{
+ static unsigned int seed;
+ seed++;
+ return seed % candidate_count;
+}
+#endif
+
+#ifdef CONFIG_IPV6_MULTIPATH_HASH
+/*
+ * Pseudo random candidate using the src port, and other information
+ * Adapted from fib_info_hashfn()
+ */
+static int rt6_info_hash_nhsfn(unsigned int candidate_count,
+ const struct flowi6 *fl6)
+{
+ unsigned int val = fl6->flowi6_proto;
+
+ val ^= fl6->daddr.s6_addr32[0];
+ val ^= fl6->daddr.s6_addr32[1];
+ val ^= fl6->daddr.s6_addr32[2];
+ val ^= fl6->daddr.s6_addr32[3];
+
+ val ^= fl6->saddr.s6_addr32[0];
+ val ^= fl6->saddr.s6_addr32[1];
+ val ^= fl6->saddr.s6_addr32[2];
+ val ^= fl6->saddr.s6_addr32[3];
+
+ /* Work only if this not encapsulated */
+ switch (fl6->flowi6_proto) {
+ case IPPROTO_UDP:
+ case IPPROTO_TCP:
+ case IPPROTO_SCTP:
+ val ^= fl6->fl6_sport;
+ val ^= fl6->fl6_dport;
+ break;
+
+ case IPPROTO_ICMPV6:
+ val ^= fl6->fl6_icmp_type;
+ val ^= fl6->fl6_icmp_code;
+ break;
+ }
+ /* RFC6438 recommands to use flowlabel */
+ val ^= fl6->flowlabel;
+
+ /* Perhaps, we need to tune, this function? */
+ val = val ^ (val >> 7) ^ (val >> 12);
+ return val % candidate_count;
+}
+#endif
+
+/*
+ * This function return an index used to select (at random, round robin, ...)
+ * a route between any siblings.
+ *
+ * Note: fl6 can be NULL
+ */
+static unsigned int rt6_info_hashfn(const struct rt6_info *rt,
+ const struct flowi6 *fl6)
+{
+ int candidate_count = rt->rt6i_nsiblings + 1;
+
+#if defined(CONFIG_IPV6_MULTIPATH_RR)
+ return rt6_info_hash_falserr(candidate_count);
+#elif defined(CONFIG_IPV6_MULTIPATH_RANDOM)
+ return rt6_info_hash_randomfn(candidate_count);
+#elif defined(CONFIG_IPV6_MULTIPATH_HASH)
+ if (fl6 == NULL)
+ return 0;
+ return rt6_info_hash_nhsfn(candidate_count, fl6);
+#else
+ return 0;
+#endif
+}
+
+static struct rt6_info *rt6_multipath_select(struct rt6_info *match,
+ struct flowi6 *fl6)
+{
+ struct rt6_info *sibling, *next_sibling;
+ int route_choosen;
+
+ route_choosen = rt6_info_hashfn(match, fl6);
+ /* Don't change the route, if route_choosen == 0
+ * (siblings does not include ourself)
+ */
+ if (route_choosen)
+ list_for_each_entry_safe(sibling, next_sibling,
+ &match->rt6i_siblings, rt6i_siblings) {
+ route_choosen--;
+ if (route_choosen == 0) {
+ match = sibling;
+ break;
+ }
+ }
+ return match;
+}
+#endif /* CONFIG_IPV6_MULTIPATH */
+
/*
* Route lookup. Any table->tb6_lock is implied.
*/
@@ -705,6 +830,10 @@ static struct rt6_info *ip6_pol_route_lookup(struct net *net,
restart:
rt = fn->leaf;
rt = rt6_device_match(net, rt, &fl6->saddr, fl6->flowi6_oif, flags);
+#ifdef CONFIG_IPV6_MULTIPATH
+ if (rt->rt6i_nsiblings && fl6->flowi6_oif == 0)
+ rt = rt6_multipath_select(rt, fl6);
+#endif
BACKTRACK(net, &fl6->saddr);
out:
dst_use(&rt->dst, jiffies);
@@ -866,7 +995,10 @@ restart_2:
restart:
rt = rt6_select(fn, oif, strict | reachable);
-
+#ifdef CONFIG_IPV6_MULTIPATH
+ if (rt->rt6i_nsiblings && oif == 0)
+ rt = rt6_multipath_select(rt, fl6);
+#endif
BACKTRACK(net, &fl6->saddr);
if (rt == net->ipv6.ip6_null_entry ||
rt->rt6i_flags & RTF_CACHE)
@@ -2247,6 +2379,9 @@ static const struct nla_policy rtm_ipv6_policy[RTA_MAX+1] = {
[RTA_IIF] = { .type = NLA_U32 },
[RTA_PRIORITY] = { .type = NLA_U32 },
[RTA_METRICS] = { .type = NLA_NESTED },
+#ifdef CONFIG_IPV6_MULTIPATH
+ [RTA_MULTIPATH] = { .len = sizeof(struct rtnexthop) },
+#endif
};
static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh,
@@ -2324,11 +2459,69 @@ static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh,
if (tb[RTA_TABLE])
cfg->fc_table = nla_get_u32(tb[RTA_TABLE]);
+#ifdef CONFIG_IPV6_MULTIPATH
+ if (tb[RTA_MULTIPATH]) {
+ cfg->fc_mp = nla_data(tb[RTA_MULTIPATH]);
+ cfg->fc_mp_len = nla_len(tb[RTA_MULTIPATH]);
+ }
+#endif
+
err = 0;
errout:
return err;
}
+#ifdef CONFIG_IPV6_MULTIPATH
+static int ip6_route_multipath(struct fib6_config *cfg, int add)
+{
+ struct fib6_config r_cfg;
+ struct rtnexthop *rtnh;
+ int remaining;
+ int attrlen;
+ int err = 0, last_err = 0;
+
+beginning:
+ rtnh = (struct rtnexthop *)cfg->fc_mp;
+ remaining = cfg->fc_mp_len;
+
+ /* Parse a Multipath Entry */
+ while (rtnh_ok(rtnh, remaining)) {
+ memcpy(&r_cfg, cfg, sizeof(*cfg));
+ if (rtnh->rtnh_ifindex)
+ r_cfg.fc_ifindex = rtnh->rtnh_ifindex;
+
+ attrlen = rtnh_attrlen(rtnh);
+ if (attrlen > 0) {
+ struct nlattr *nla, *attrs = rtnh_attrs(rtnh);
+
+ nla = nla_find(attrs, attrlen, RTA_GATEWAY);
+ if (nla) {
+ nla_memcpy(&r_cfg.fc_gateway, nla, 16);
+ r_cfg.fc_flags |= RTF_GATEWAY;
+ }
+ }
+ err = add ? ip6_route_add(&r_cfg) : ip6_route_del(&r_cfg);
+ if (err) {
+ last_err = err;
+ /* If we are trying to remove a route, do not stop the
+ * loop when ip6_route_del() fails (because next hop is
+ * already gone), we should try to remove all next hops.
+ */
+ if (add) {
+ /* If add fails, we should try to delete all
+ * next hops that have been already added.
+ */
+ add = 0;
+ goto beginning;
+ }
+ }
+ rtnh = rtnh_next(rtnh, &remaining);
+ }
+
+ return last_err;
+}
+#endif /* CONFIG_IPV6_MULTIPATH */
+
static int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
{
struct fib6_config cfg;
@@ -2338,7 +2531,12 @@ static int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *a
if (err < 0)
return err;
- return ip6_route_del(&cfg);
+#ifdef CONFIG_IPV6_MULTIPATH
+ if (cfg.fc_mp)
+ return ip6_route_multipath(&cfg, 0);
+ else
+#endif
+ return ip6_route_del(&cfg);
}
static int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
@@ -2350,7 +2548,12 @@ static int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *a
if (err < 0)
return err;
- return ip6_route_add(&cfg);
+#ifdef CONFIG_IPV6_MULTIPATH
+ if (cfg.fc_mp)
+ return ip6_route_multipath(&cfg, 1);
+ else
+#endif
+ return ip6_route_add(&cfg);
}
static inline size_t rt6_nlmsg_size(void)
--
1.7.12
^ permalink raw reply related
* [RFC PATCH net-next v2 0/1] Add support of ECMPv6
From: Nicolas Dichtel @ 2012-09-14 7:59 UTC (permalink / raw)
To: yoshfuji; +Cc: bernat, netdev, davem
In-Reply-To: <505058F5.9020707@linux-ipv6.org>
Here is a proposal to add the support of ECMPv6. The previous patch
from Vincent against iproute2 can be used, but a little other patch is needed
too, see http://patchwork.ozlabs.org/patch/183277/
If the kernel patch is approved, I will submit formally the patch for
iproute2.
Here is an example of a command to add an ECMP route:
$ ip -6 route add 3ffe:304:124:2306::/64 \
nexthop via fe80::230:1bff:feb4:e05c dev eth0 weight 1 \
nexthop via fe80::230:1bff:feb4:dd4f dev eth0 weight 1
v2: rename CONFIG_IPV6_MULTIPATH_ROUTE to CONFIG_IPV6_MULTIPATH_HASH
use flowlabel in the hash function
add reference to RFC
fix a small identation issue
remove "If unsure, say N." from the help of CONFIG_IPV6_MULTIPATH
Comments are welcome.
Regards,
Nicolas
^ permalink raw reply
* Re: [net-next.git 3/8 (V2)] stmmac: add the initial tx coalesce schema
From: Giuseppe CAVALLARO @ 2012-09-14 7:36 UTC (permalink / raw)
To: David Miller; +Cc: netdev, bhutchings
In-Reply-To: <20120913.162333.1518469374321928795.davem@davemloft.net>
On 9/13/2012 10:23 PM, David Miller wrote:
> From: Giuseppe CAVALLARO <peppe.cavallaro@st.com>
> Date: Tue, 11 Sep 2012 08:55:09 +0200
>
>> + unsigned long flags;
>> +
>> + spin_lock_irqsave(&priv->tx_lock, flags);
>>
>> - spin_lock(&priv->tx_lock);
>> + priv->xstats.tx_clean++;
>
> You are changing the locking here for the sake of the new timer.
>
> But timers run in software interrupt context, so this change is
> completely unnecessary since NAPI runs in software interrupt context
> as well, and neither timers nor NAPI run in hardware interrupts
> context.
Indeed It can be called by the ISR too in this new implementation.
I have added the spin_lock_irqsave/restore otherwise, testing with
CONFIG_PROVE_LOOKING, I get the following warning on ARM SMP.
[ 8.030000]
[ 8.030000] =================================
[ 8.030000] [ INFO: inconsistent lock state ]
[ 8.030000] 3.4.7_stm24_0302-b2000+ #103 Not tainted
[ 8.030000] ---------------------------------
[ 8.030000] inconsistent {HARDIRQ-ON-W} -> {IN-HARDIRQ-W} usage.
[ 8.030000] swapper/0/1 [HC1[1]:SC0[0]:HE0:SE1] takes:
[ 8.030000] (&(&priv->tx_lock)->rlock){?.-...}, at: [<802651d8>]
stmmac_tx+0x1c/0x388
[ 8.030000] {HARDIRQ-ON-W} state was registered at:
[ 8.030000] [<800562b4>] __lock_acquire+0x638/0x179c
[ 8.030000] [<80057884>] lock_acquire+0x60/0x74
[ 8.030000] [<80428a08>] _raw_spin_lock+0x40/0x50
[ 8.030000] [<802651d8>] stmmac_tx+0x1c/0x388
[ 8.030000] [<80026be0>] run_timer_softirq+0x180/0x23c
[ 8.030000] [<80020ccc>] __do_softirq+0xa0/0x114
[ 8.030000] [<80021204>] irq_exit+0x58/0x7c
[ 8.030000] [<8000dc80>] handle_IRQ+0x7c/0xb8
[ 8.030000] [<80008464>] gic_handle_irq+0x34/0x58
[ 8.030000] [<80429684>] __irq_svc+0x44/0x78
[ 8.030000] [<8001c3f4>] vprintk+0x41c/0x480
[ 8.030000] [<8042097c>] printk+0x18/0x24
[ 8.030000] [<805aef6c>] prepare_namespace+0x1c/0x1a4
[ 8.030000] [<805ae980>] kernel_init+0x1c8/0x20c
[ 8.030000] [<8000deb8>] kernel_thread_exit+0x0/0x8
[ 8.030000] irq event stamp: 254745
[ 8.030000] hardirqs last enabled at (254744): [<80429240>]
_raw_spin_unlock_irqrestore+0x3c/0x6c
[ 8.030000] hardirqs last disabled at (254745): [<80429674>]
__irq_svc+0x34/0x78
[ 8.030000] softirqs last enabled at (254741): [<8035d964>]
dev_queue_xmit+0x6a4/0x724
[ 8.030000] softirqs last disabled at (254737): [<8035d2d4>]
dev_queue_xmit+0x14/0x724
[ 8.030000]
[ 8.030000] other info that might help us debug this:
[ 8.030000] Possible unsafe locking scenario:
[ 8.030000]
[ 8.030000] CPU0
[ 8.030000] ----
[ 8.030000] lock(&(&priv->tx_lock)->rlock);
[ 8.030000] <Interrupt>
[ 8.030000] lock(&(&priv->tx_lock)->rlock);
[ 8.030000]
[ 8.030000] *** DEADLOCK ***
> Therefore, disabling hardware interrupts for this lock is unnecessary
> and will decrease performance.
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
^ permalink raw reply
* Re: [RFC PATCH 4/5] bnx2fc: Use new fcoe_sysfs control interface
From: Bhanu Prakash Gollapudi @ 2012-09-14 7:28 UTC (permalink / raw)
To: Robert Love; +Cc: netdev, gregkh, linux-scsi, devel
In-Reply-To: <20120910225930.13140.97949.stgit@fritz>
On 09/10/2012 03:59 PM, Robert Love wrote:
> Convert bnx2fc to use the new fcoe_sysfs create, delete,
> enable, disable, start and mode.
>
> bnx2fc doesn't support VN2VN. bnx2fc will not initialize
> the set_fcoe_ctlr_mode routine and therefore its instances
> will always be in FABRIC mode. There was previously an
> explicit check for the ctlr's mode, but this is no longer
> needed because not implementing set_fcoe_ctlr_mode implies
> that the ctlr cannot change from the FABRIC mode.
>
> Signed-off-by: Robert Love <robert.w.love@intel.com>
> ---
> drivers/scsi/bnx2fc/bnx2fc_fcoe.c | 98 +++++++++++++++++++++++--------------
> 1 file changed, 60 insertions(+), 38 deletions(-)
>
> diff --git a/drivers/scsi/bnx2fc/bnx2fc_fcoe.c b/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
> index f52f668f..560c8c8 100644
> --- a/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
> +++ b/drivers/scsi/bnx2fc/bnx2fc_fcoe.c
<snip>
> /**
> + * bnx2fc_alloc - Alocate a bnx2fc FCoE interface
> + *
> + * @cdev: The FCoE Controller Device to start
> + *
> + * Called from sysfs.
> + *
> + * Returns: 0 for success
> + */
> +static int bnx2fc_start(struct fcoe_ctlr_device *cdev)
> +{
> + struct fcoe_ctlr *ctlr = fcoe_ctlr_device_priv(cdev);
> + struct fc_lport *lport = ctlr->lp;
> + struct fcoe_port *port = lport_priv(lport);
> + struct bnx2fc_interface *interface = port->priv;
> +
> + lport->boot_time = jiffies;
> +
> + /* Make this master N_port */
> + ctlr->lp = lport;
ctlr->lp should be set in bnx2fc_alloc() as we access it here in the
beginning of this function.
> +
> + if (!bnx2fc_link_ok(lport)) {
> + fcoe_ctlr_link_up(ctlr);
> + fc_host_port_type(lport->host) = FC_PORTTYPE_NPORT;
> + set_bit(ADAPTER_STATE_READY, &interface->hba->adapter_state);
> + }
> +
> + BNX2FC_HBA_DBG(lport, "create: START DISC\n");
> + bnx2fc_start_disc(interface);
I think more changes are required for bnx2fc as fc_lport_init() is
called just before calling fc_fabric_login() - whcih is called during
'start'. Because of this, if we just call 'create' followed by 'destroy'
without calling 'start', lport is not initialized and I expect to see
some panics when destroy is called.
Let me try testing your patches and send you any fixes that are required.
> + interface->enabled = true;
> +
> + return 0;
> +}
> +
> +/**
> * bnx2fc_find_hba_for_cnic - maps cnic instance to bnx2fc hba instance
> *
> * @cnic: Pointer to cnic device instance
> @@ -2271,10 +2292,8 @@ static struct fcoe_transport bnx2fc_transport = {
> .attached = false,
> .list = LIST_HEAD_INIT(bnx2fc_transport.list),
> .match = bnx2fc_match,
> - .create = bnx2fc_create,
> + .alloc = bnx2fc_alloc,
> .destroy = bnx2fc_destroy,
> - .enable = bnx2fc_enable,
> - .disable = bnx2fc_disable,
> };
>
> /**
> @@ -2514,6 +2533,9 @@ module_init(bnx2fc_mod_init);
> module_exit(bnx2fc_mod_exit);
>
> static struct fcoe_sysfs_function_template bnx2fc_fcoe_sysfs_templ = {
> + .set_fcoe_ctlr_start = bnx2fc_start,
> + .set_fcoe_ctlr_enable = bnx2fc_enable,
> + .set_fcoe_ctlr_disable = bnx2fc_disable,
> .get_fcoe_ctlr_mode = fcoe_ctlr_get_fip_mode,
> .get_fcoe_ctlr_link_fail = bnx2fc_ctlr_get_lesb,
> .get_fcoe_ctlr_vlink_fail = bnx2fc_ctlr_get_lesb,
>
>
^ permalink raw reply
* [PATCH net-next 2/2] gre: add GSO support
From: Eric Dumazet @ 2012-09-14 7:25 UTC (permalink / raw)
To: David Miller; +Cc: netdev, Maciej Żenczykowski
From: Eric Dumazet <edumazet@google.com>
Add GSO support to GRE tunnels.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Maciej Żenczykowski <maze@google.com>
---
net/ipv4/ip_gre.c | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index b062a98..f233c1d 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -745,6 +745,10 @@ static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev
__be32 dst;
int mtu;
+ if (skb->ip_summed == CHECKSUM_PARTIAL &&
+ skb_checksum_help(skb))
+ goto tx_error;
+
if (dev->type == ARPHRD_ETHER)
IPCB(skb)->flags = 0;
@@ -1296,6 +1300,11 @@ static void ipgre_dev_free(struct net_device *dev)
free_netdev(dev);
}
+#define GRE_FEATURES (NETIF_F_SG | \
+ NETIF_F_FRAGLIST | \
+ NETIF_F_HIGHDMA | \
+ NETIF_F_HW_CSUM)
+
static void ipgre_tunnel_setup(struct net_device *dev)
{
dev->netdev_ops = &ipgre_netdev_ops;
@@ -1309,6 +1318,9 @@ static void ipgre_tunnel_setup(struct net_device *dev)
dev->addr_len = 4;
dev->features |= NETIF_F_NETNS_LOCAL;
dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
+
+ dev->features |= GRE_FEATURES;
+ dev->hw_features |= GRE_FEATURES;
}
static int ipgre_tunnel_init(struct net_device *dev)
^ permalink raw reply related
* [PATCH net-next 1/2] net: provide a default dev->ethtool_ops
From: Eric Dumazet @ 2012-09-14 7:24 UTC (permalink / raw)
To: David Miller; +Cc: netdev, Maciej Żenczykowski
From: Eric Dumazet <edumazet@google.com>
Instead of forcing device drivers to provide empty ethtool_ops or tweak
net/core/ethtool.c again, we could provide a generic ethtool_ops.
This occurred to me when I wanted to add GSO support to GRE tunnels.
ethtool -k support should be generic for all drivers.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Maciej Żenczykowski <maze@google.com>
---
net/core/dev.c | 5 +++
net/core/ethtool.c | 54 +++++++++++++++++--------------------------
2 files changed, 27 insertions(+), 32 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
index b1e6d63..ff8dcfc 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -6045,6 +6045,11 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
strcpy(dev->name, name);
dev->group = INIT_NETDEV_GROUP;
+ if (!dev->ethtool_ops) {
+ static const struct ethtool_ops default_ethtool_ops;
+
+ dev->ethtool_ops = &default_ethtool_ops;
+ }
return dev;
free_all:
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index cbf033d..aef0162 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -175,7 +175,7 @@ static int __ethtool_get_sset_count(struct net_device *dev, int sset)
if (sset == ETH_SS_FEATURES)
return ARRAY_SIZE(netdev_features_strings);
- if (ops && ops->get_sset_count && ops->get_strings)
+ if (ops->get_sset_count && ops->get_strings)
return ops->get_sset_count(dev, sset);
else
return -EOPNOTSUPP;
@@ -311,7 +311,7 @@ int __ethtool_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
ASSERT_RTNL();
- if (!dev->ethtool_ops || !dev->ethtool_ops->get_settings)
+ if (!dev->ethtool_ops->get_settings)
return -EOPNOTSUPP;
memset(cmd, 0, sizeof(struct ethtool_cmd));
@@ -355,7 +355,7 @@ static noinline_for_stack int ethtool_get_drvinfo(struct net_device *dev,
memset(&info, 0, sizeof(info));
info.cmd = ETHTOOL_GDRVINFO;
- if (ops && ops->get_drvinfo) {
+ if (ops->get_drvinfo) {
ops->get_drvinfo(dev, &info);
} else if (dev->dev.parent && dev->dev.parent->driver) {
strlcpy(info.bus_info, dev_name(dev->dev.parent),
@@ -370,7 +370,7 @@ static noinline_for_stack int ethtool_get_drvinfo(struct net_device *dev,
* this method of obtaining string set info is deprecated;
* Use ETHTOOL_GSSET_INFO instead.
*/
- if (ops && ops->get_sset_count) {
+ if (ops->get_sset_count) {
int rc;
rc = ops->get_sset_count(dev, ETH_SS_TEST);
@@ -383,9 +383,9 @@ static noinline_for_stack int ethtool_get_drvinfo(struct net_device *dev,
if (rc >= 0)
info.n_priv_flags = rc;
}
- if (ops && ops->get_regs_len)
+ if (ops->get_regs_len)
info.regdump_len = ops->get_regs_len(dev);
- if (ops && ops->get_eeprom_len)
+ if (ops->get_eeprom_len)
info.eedump_len = ops->get_eeprom_len(dev);
if (copy_to_user(useraddr, &info, sizeof(info)))
@@ -1275,7 +1275,7 @@ static int ethtool_get_dump_flag(struct net_device *dev,
struct ethtool_dump dump;
const struct ethtool_ops *ops = dev->ethtool_ops;
- if (!dev->ethtool_ops->get_dump_flag)
+ if (!ops->get_dump_flag)
return -EOPNOTSUPP;
if (copy_from_user(&dump, useraddr, sizeof(dump)))
@@ -1299,8 +1299,7 @@ static int ethtool_get_dump_data(struct net_device *dev,
const struct ethtool_ops *ops = dev->ethtool_ops;
void *data = NULL;
- if (!dev->ethtool_ops->get_dump_data ||
- !dev->ethtool_ops->get_dump_flag)
+ if (!ops->get_dump_data || !ops->get_dump_flag)
return -EOPNOTSUPP;
if (copy_from_user(&dump, useraddr, sizeof(dump)))
@@ -1349,7 +1348,7 @@ static int ethtool_get_ts_info(struct net_device *dev, void __user *useraddr)
err = phydev->drv->ts_info(phydev, &info);
- } else if (dev->ethtool_ops && dev->ethtool_ops->get_ts_info) {
+ } else if (ops->get_ts_info) {
err = ops->get_ts_info(dev, &info);
@@ -1410,8 +1409,9 @@ static int ethtool_get_module_eeprom(struct net_device *dev,
modinfo.eeprom_len);
}
-/* The main entry point in this file. Called from net/core/dev.c */
-
+/* The main entry point in this file. Called from net/core/dev.c
+ * with RTNL held.
+ */
int dev_ethtool(struct net *net, struct ifreq *ifr)
{
struct net_device *dev = __dev_get_by_name(net, ifr->ifr_name);
@@ -1419,25 +1419,15 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
u32 ethcmd;
int rc;
u32 old_features;
+ const struct ethtool_ops *ops;
if (!dev || !netif_device_present(dev))
return -ENODEV;
+ ops = dev->ethtool_ops;
if (copy_from_user(ðcmd, useraddr, sizeof(ethcmd)))
return -EFAULT;
- if (!dev->ethtool_ops) {
- /* A few commands do not require any driver support,
- * are unprivileged, and do not change anything, so we
- * can take a shortcut to them. */
- if (ethcmd == ETHTOOL_GDRVINFO)
- return ethtool_get_drvinfo(dev, useraddr);
- else if (ethcmd == ETHTOOL_GET_TS_INFO)
- return ethtool_get_ts_info(dev, useraddr);
- else
- return -EOPNOTSUPP;
- }
-
/* Allow some commands to be done by anyone */
switch (ethcmd) {
case ETHTOOL_GSET:
@@ -1476,8 +1466,8 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
return -EPERM;
}
- if (dev->ethtool_ops->begin) {
- rc = dev->ethtool_ops->begin(dev);
+ if (ops->begin) {
+ rc = ops->begin(dev);
if (rc < 0)
return rc;
}
@@ -1504,11 +1494,11 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
break;
case ETHTOOL_GMSGLVL:
rc = ethtool_get_value(dev, useraddr, ethcmd,
- dev->ethtool_ops->get_msglevel);
+ ops->get_msglevel);
break;
case ETHTOOL_SMSGLVL:
rc = ethtool_set_value_void(dev, useraddr,
- dev->ethtool_ops->set_msglevel);
+ ops->set_msglevel);
break;
case ETHTOOL_GEEE:
rc = ethtool_get_eee(dev, useraddr);
@@ -1570,11 +1560,11 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
break;
case ETHTOOL_GPFLAGS:
rc = ethtool_get_value(dev, useraddr, ethcmd,
- dev->ethtool_ops->get_priv_flags);
+ ops->get_priv_flags);
break;
case ETHTOOL_SPFLAGS:
rc = ethtool_set_value(dev, useraddr,
- dev->ethtool_ops->set_priv_flags);
+ ops->set_priv_flags);
break;
case ETHTOOL_GRXFH:
case ETHTOOL_GRXRINGS:
@@ -1655,8 +1645,8 @@ int dev_ethtool(struct net *net, struct ifreq *ifr)
rc = -EOPNOTSUPP;
}
- if (dev->ethtool_ops->complete)
- dev->ethtool_ops->complete(dev);
+ if (ops->complete)
+ ops->complete(dev);
if (old_features != dev->features)
netdev_features_change(dev);
^ permalink raw reply related
* Re: [PATCH] Xen backend support for paged out grant targets.
From: Ian Campbell @ 2012-09-14 7:19 UTC (permalink / raw)
To: Andres Lagar-Cavilla
Cc: Andres Lagar-Cavilla, xen-devel@xen.lists.org,
Konrad Rzeszutek Wilk, David Vrabel, David Miller,
linux-kernel@vger.kernel.org, netdev@vger.kernel.org
In-Reply-To: <B4C805E1-3390-4002-BECC-7D1AFF2BD56D@gmail.com>
On Thu, 2012-09-13 at 20:45 +0100, Andres Lagar-Cavilla wrote:
> On Sep 13, 2012, at 2:11 PM, Ian Campbell wrote:
>
> > On Thu, 2012-09-13 at 18:28 +0100, Andres Lagar-Cavilla wrote:
> >>
> >> * Add placeholder in array of grant table error descriptions for
> >> unrelated error code we jump over.
> >
> > Why not just define it, it's listed here:
> > http://xenbits.xen.org/docs/unstable/hypercall/include,public,grant_table.h.html#Enum_grant_status
> Well, a) we'd be defining something no one will be using (for the
> moment)
Even if no one in the kernel is using it, having "placeholder" as an
entry in GNTTABOP_error_msgs is just silly, even things which don't
understand GNTST_address_too_big directly could end up looking it up
here.
> b) I would be signing-off on something unrelated.
Lets take this patch instead then.
8<------------------------------------------------
>From cb9daaf3029accb6d5fef58b450a625b27190429 Mon Sep 17 00:00:00 2001
From: Ian Campbell <ian.campbell@citrix.com>
Date: Fri, 14 Sep 2012 08:10:06 +0100
Subject: [PATCH] xen: resynchronise grant table status codes with upstream
Adds GNTST_address_too_big and GNTST_eagain.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
include/xen/interface/grant_table.h | 8 ++++++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/include/xen/interface/grant_table.h b/include/xen/interface/grant_table.h
index a17d844..84a8fbf 100644
--- a/include/xen/interface/grant_table.h
+++ b/include/xen/interface/grant_table.h
@@ -519,7 +519,9 @@ DEFINE_GUEST_HANDLE_STRUCT(gnttab_get_version);
#define GNTST_no_device_space (-7) /* Out of space in I/O MMU. */
#define GNTST_permission_denied (-8) /* Not enough privilege for operation. */
#define GNTST_bad_page (-9) /* Specified page was invalid for op. */
-#define GNTST_bad_copy_arg (-10) /* copy arguments cross page boundary */
+#define GNTST_bad_copy_arg (-10) /* copy arguments cross page boundary. */
+#define GNTST_address_too_big (-11) /* transfer page address too large. */
+#define GNTST_eagain (-12) /* Operation not done; try again. */
#define GNTTABOP_error_msgs { \
"okay", \
@@ -532,7 +534,9 @@ DEFINE_GUEST_HANDLE_STRUCT(gnttab_get_version);
"no spare translation slot in the I/O MMU", \
"permission denied", \
"bad page", \
- "copy arguments cross page boundary" \
+ "copy arguments cross page boundary", \
+ "page address size too large", \
+ "operation not done; try again" \
}
#endif /* __XEN_PUBLIC_GRANT_TABLE_H__ */
--
1.7.10.4
^ permalink raw reply related
* [PATCH] iproute2: bridge: finish removing replace option in man pages
From: John Fastabend @ 2012-09-14 6:50 UTC (permalink / raw)
To: shemminger; +Cc: netdev
This patch finishes removing the replace option from the bridge
man page which I missed in this commit
commit 57b9785de32404da3d2ac5483469b7fcc5a9c9e7
Author: John Fastabend <john.r.fastabend@intel.com>
Date: Mon Aug 27 10:52:31 2012 -0700
iproute2: bridge: remove replace and change options
Also add documentation for "{ self | embedded }" already shown on
the cmd line help msg.
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
---
man/man8/bridge.8 | 19 +++++++++++++------
1 files changed, 13 insertions(+), 6 deletions(-)
diff --git a/man/man8/bridge.8 b/man/man8/bridge.8
index 63d166b..5ce8219 100644
--- a/man/man8/bridge.8
+++ b/man/man8/bridge.8
@@ -22,11 +22,12 @@ bridge \- show / manipulate bridge addresses and devices
\fB\-s\fR[\fItatistics\fR]
.ti -8
-.BR "bridge fdb" " { " add " | " del " | " change " | " replace " } "
+.BR "bridge fdb" " { " add " | " del " } "
.I LLADDR
.B dev
.IR DEV " { "
-.BR local " | " temp " }"
+.BR local " | " temp " } { "
+.BR self " } { " embedded " } "
.ti -8
.BR "bridge fdb" " [ " show " ] [ "
@@ -92,11 +93,9 @@ objects contain known Ethernet addresses on a link.
The corresponding commands display fdb entries, add new entries,
and delete old ones.
-.SS bridge fdb add - add a new neighbor entry
-.SS bridge fdb change - change an existing entry
-.SS bridge fdb replace - add a new entry or change an existing one
+.SS bridge fdb add - add a new fdb entry
-These commands create new neighbor records or update existing ones.
+This command creates a new fdb entry.
.TP
.BI "ADDRESS"
@@ -117,6 +116,14 @@ and is never forwarded.
- the address is a dynamic entry, and will be removed if not used.
.sp
+.B self
+- the address is associated with a software fdb (default)
+.sp
+
+.B embedded
+- the address is associated with an offloaded fdb
+.sp
+
.in -8
.SS bridge fdb delete - delete a forwarding database entry
^ permalink raw reply related
* Re: [RFC PATCH 2/5] libfcoe: Create new libfcoe control interfaces
From: Bhanu Prakash Gollapudi @ 2012-09-14 7:06 UTC (permalink / raw)
To: Robert Love; +Cc: netdev, gregkh, linux-scsi, devel
In-Reply-To: <20120910225919.13140.63240.stgit@fritz>
On 9/10/2012 3:59 PM, Robert Love wrote:
> This patch is the first in a series that will remove
> libfcoe's create, destroy, enable and disable module
> parameters and replace them with interface files in
> the new /sys/bus/fcoe subsystem.
>
> Old layout:
>
> /sys/module/libfcoe/parameters/{create,destroy,enable,disable,vn2vn_create}
>
> New layout:
>
> /sys/bus/fcoe/ctlr_{create,destroy}
> /sys/bus/fcoe/ctlr_X/{enable,disable,start}
>
> This patch moves fcoe drivers to the following
> initialization sequence-
>
> 1) create/alloc
> 2) configure
> 3) start
>
> A control sysfs interface at /sys/bus/fcoe/ctlr_create
> is added. Writing the interface name to this file
> will allocate memory and create a sysfs entry for a
> new fcoe_ctlr_device. The user may then tune the interface in
> any desired way. After configuration the user will
> echo any value into the /sys/bus/fcoe/devices/ctlr_X/start
> interface to proceed with logging in.
>
> VN2VN logins will still use the module parameters.
> A follow up patch to this one will make the 'mode'
> attribute of the fcoe_ctlr_device writable. Which will
> allow a user to change the ctlr's mode to 'VN2VN'.
>
> Signed-off-by: Robert Love <robert.w.love@intel.com>
> ---
> Documentation/ABI/testing/sysfs-bus-fcoe | 43 ++++++++++++
> drivers/scsi/fcoe/fcoe.h | 9 +++
> drivers/scsi/fcoe/fcoe_ctlr.c | 2 -
> drivers/scsi/fcoe/fcoe_sysfs.c | 78 ++++++++++++++++++++++
> drivers/scsi/fcoe/fcoe_transport.c | 105 +++++++++++++++++++++++++++++-
> include/scsi/fcoe_sysfs.h | 4 +
> include/scsi/libfcoe.h | 14 ++++
> 7 files changed, 250 insertions(+), 5 deletions(-)
>
<snip>
> diff --git a/drivers/scsi/fcoe/fcoe_ctlr.c b/drivers/scsi/fcoe/fcoe_ctlr.c
> index bd899bf..ccb92323 100644
> --- a/drivers/scsi/fcoe/fcoe_ctlr.c
> +++ b/drivers/scsi/fcoe/fcoe_ctlr.c
> @@ -147,7 +147,7 @@ static void fcoe_ctlr_map_dest(struct fcoe_ctlr *fip)
> */
> void fcoe_ctlr_init(struct fcoe_ctlr *fip, enum fip_state mode)
> {
> - fcoe_ctlr_set_state(fip, FIP_ST_LINK_WAIT);
> + fcoe_ctlr_set_state(fip, FIP_ST_DISABLED);
Robert, what is the reason for initializing it to DISABLED? Unless the
FIP state is FIP_ST_LINK_WAIT, fcoe_ctlr_link_up() doesnt set
lport->link_up and hence does not allow any FIP/FCoE frames to be sent out.
> fip->mode = mode;
> INIT_LIST_HEAD(&fip->fcfs);
> mutex_init(&fip->ctlr_mutex);
<snip>
> @@ -627,6 +626,108 @@ static int libfcoe_device_notification(struct notifier_block *notifier,
> return NOTIFY_OK;
> }
>
> +ssize_t fcoe_ctlr_create_store(struct bus_type *bus,
> + const char *buf, size_t count)
> +{
> + struct net_device *netdev = NULL;
> + struct fcoe_transport *ft = NULL;
> + struct fcoe_ctlr_device *ctlr_dev = NULL;
> + int rc = -ENODEV;
> + int err;
> +
> + mutex_lock(&ft_mutex);
> +
> + netdev = fcoe_if_to_netdev(buf);
> + if (!netdev) {
> + LIBFCOE_TRANSPORT_DBG("Invalid device %s.\n", buf);
> + rc = -ENODEV;
> + goto out_nodev;
> + }
> +
> + ft = fcoe_netdev_map_lookup(netdev);
> + if (ft) {
> + LIBFCOE_TRANSPORT_DBG("transport %s already has existing "
> + "FCoE instance on %s.\n",
> + ft->name, netdev->name);
> + rc = -EEXIST;
> + goto out_putdev;
> + }
> +
> + ft = fcoe_transport_lookup(netdev);
> + if (!ft) {
> + LIBFCOE_TRANSPORT_DBG("no FCoE transport found for %s.\n",
> + netdev->name);
> + rc = -ENODEV;
> + goto out_putdev;
> + }
> +
> + /* pass to transport create */
> + err = ft->alloc ? ft->alloc(netdev) : -ENODEV;
> + if (err) {
> + fcoe_del_netdev_mapping(netdev);
> + rc = -ENOMEM;
> + goto out_putdev;
> + }
> +
> + err = fcoe_add_netdev_mapping(netdev, ft);
> + if (err) {
> + LIBFCOE_TRANSPORT_DBG("failed to add new netdev mapping "
> + "for FCoE transport %s for %s.\n",
> + ft->name, netdev->name);
> + rc = -ENODEV;
> + goto out_putdev;
> + }
> +
> + LIBFCOE_TRANSPORT_DBG("transport %s %s to create fcoe on %s.\n",
> + ft->name, (ctlr_dev) ? "succeeded" : "failed",
> + netdev->name);
Where is ctlr_dev updated? I guess you're intending to check return
status of ft->alloc() here.
> +
> +out_putdev:
> + dev_put(netdev);
> +out_nodev:
> + mutex_unlock(&ft_mutex);
> + return rc;
> +}
> +
> +ssize_t fcoe_ctlr_destroy_store(struct bus_type *bus,
> + const char *buf, size_t count)
> +{
> + int rc = -ENODEV;
> + struct net_device *netdev = NULL;
> + struct fcoe_transport *ft = NULL;
> +
> + mutex_lock(&ft_mutex);
> +
> + netdev = fcoe_if_to_netdev(buf);
> + if (!netdev) {
> + LIBFCOE_TRANSPORT_DBG("invalid device %s.\n", buf);
> + goto out_nodev;
> + }
> +
> + ft = fcoe_netdev_map_lookup(netdev);
> + if (!ft) {
> + LIBFCOE_TRANSPORT_DBG("no FCoE transport found for %s.\n",
> + netdev->name);
> + goto out_putdev;
> + }
> +
> + /* pass to transport destroy */
> + rc = ft->destroy(netdev);
> + if (rc)
> + goto out_putdev;
> +
> + fcoe_del_netdev_mapping(netdev);
> + LIBFCOE_TRANSPORT_DBG("transport %s %s to destroy fcoe on %s.\n",
> + ft->name, (rc) ? "failed" : "succeeded",
> + netdev->name);
> + rc = count; /* required for successful return */
> +out_putdev:
> + dev_put(netdev);
> +out_nodev:
> + mutex_unlock(&ft_mutex);
> + return rc;
> +}
> +EXPORT_SYMBOL(fcoe_ctlr_destroy_store);
>
> /**
> * fcoe_transport_create() - Create a fcoe interface
> diff --git a/include/scsi/fcoe_sysfs.h b/include/scsi/fcoe_sysfs.h
> index 421ae67..8c5ea70 100644
> --- a/include/scsi/fcoe_sysfs.h
> +++ b/include/scsi/fcoe_sysfs.h
> @@ -36,6 +36,9 @@ struct fcoe_sysfs_function_template {
> void (*get_fcoe_ctlr_fcs_error)(struct fcoe_ctlr_device *);
> void (*get_fcoe_ctlr_mode)(struct fcoe_ctlr_device *);
> void (*set_fcoe_ctlr_mode)(struct fcoe_ctlr_device *);
> + int (*set_fcoe_ctlr_start)(struct fcoe_ctlr_device *);
> + int (*set_fcoe_ctlr_enable)(struct fcoe_ctlr_device *);
> + int (*set_fcoe_ctlr_disable)(struct fcoe_ctlr_device *);
> void (*get_fcoe_fcf_selected)(struct fcoe_fcf_device *);
> void (*get_fcoe_fcf_vlan_id)(struct fcoe_fcf_device *);
> };
> @@ -64,6 +67,7 @@ struct fcoe_ctlr_device {
>
> int fcf_dev_loss_tmo;
> enum fip_conn_type mode;
> + u8 started:1;
>
> /* expected in host order for displaying */
> struct fcoe_fc_els_lesb lesb;
> diff --git a/include/scsi/libfcoe.h b/include/scsi/libfcoe.h
> index 20533cc..b19a489 100644
> --- a/include/scsi/libfcoe.h
> +++ b/include/scsi/libfcoe.h
> @@ -289,8 +289,11 @@ static inline bool is_fip_mode(struct fcoe_ctlr *fip)
> * @attached: whether this transport is already attached
> * @list: list linkage to all attached transports
> * @match: handler to allow the transport driver to match up a given netdev
> + * @alloc: handler to allocate per-instance FCoE structures
> + * (no discovery or login)
> * @create: handler to sysfs entry of create for FCoE instances
> - * @destroy: handler to sysfs entry of destroy for FCoE instances
> + * @destroy: handler to delete per-instance FCoE structures
> + * (frees all memory)
> * @enable: handler to sysfs entry of enable for FCoE instances
> * @disable: handler to sysfs entry of disable for FCoE instances
> */
> @@ -299,6 +302,7 @@ struct fcoe_transport {
> bool attached;
> struct list_head list;
> bool (*match) (struct net_device *device);
> + int (*alloc) (struct net_device *device);
> int (*create) (struct net_device *device, enum fip_state fip_mode);
> int (*destroy) (struct net_device *device);
> int (*enable) (struct net_device *device);
> @@ -375,4 +379,12 @@ struct fcoe_netdev_mapping {
> int fcoe_transport_attach(struct fcoe_transport *ft);
> int fcoe_transport_detach(struct fcoe_transport *ft);
>
> +/* sysfs store handler for ctrl_control interface */
> +ssize_t fcoe_ctlr_create_store(struct bus_type *bus,
> + const char *buf, size_t count);
> +ssize_t fcoe_ctlr_destroy_store(struct bus_type *bus,
> + const char *buf, size_t count);
> +
> #endif /* _LIBFCOE_H */
> +
> +
>
>
^ permalink raw reply
* [PATCH net-next] net: dev: fix incorrect getting net device's name
From: Gao feng @ 2012-09-14 6:58 UTC (permalink / raw)
To: davem; +Cc: ebiederm, eric.dumazet, netdev, Gao feng
When moving a nic from net namespace A to net namespace B,
in dev_change_net_namesapce,we call __dev_get_by_name to
decide if the netns B has the device has the same name.
if the netns B already has the same named device,we call
dev_get_valid_name to try to get a valid name for this nic in
the netns B,but net_device->nd_net still point to netns A now.
this patch fix it.
Signed-off-by: Gao feng <gaofeng@cn.fujitsu.com>
---
net/core/dev.c | 28 ++++++++++++++++++++--------
1 files changed, 20 insertions(+), 8 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
index b1e6d63..381ea68 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -959,18 +959,30 @@ int dev_alloc_name(struct net_device *dev, const char *name)
}
EXPORT_SYMBOL(dev_alloc_name);
-static int dev_get_valid_name(struct net_device *dev, const char *name)
+static int dev_alloc_name_ns(struct net *net,
+ struct net_device *dev,
+ const char *name)
{
- struct net *net;
+ char buf[IFNAMSIZ];
+ int ret;
- BUG_ON(!dev_net(dev));
- net = dev_net(dev);
+ ret = __dev_alloc_name(net, name, buf);
+ if (ret >= 0)
+ strlcpy(dev->name, buf, IFNAMSIZ);
+ return ret;
+}
+
+static int dev_get_valid_name(struct net *net,
+ struct net_device *dev,
+ const char *name)
+{
+ BUG_ON(!net);
if (!dev_valid_name(name))
return -EINVAL;
if (strchr(name, '%'))
- return dev_alloc_name(dev, name);
+ return dev_alloc_name_ns(net, dev, name);
else if (__dev_get_by_name(net, name))
return -EEXIST;
else if (dev->name != name)
@@ -1006,7 +1018,7 @@ int dev_change_name(struct net_device *dev, const char *newname)
memcpy(oldname, dev->name, IFNAMSIZ);
- err = dev_get_valid_name(dev, newname);
+ err = dev_get_valid_name(net, dev, newname);
if (err < 0)
return err;
@@ -5585,7 +5597,7 @@ int register_netdevice(struct net_device *dev)
dev->iflink = -1;
- ret = dev_get_valid_name(dev, dev->name);
+ ret = dev_get_valid_name(net, dev, dev->name);
if (ret < 0)
goto out;
@@ -6229,7 +6241,7 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char
/* We get here if we can't use the current device name */
if (!pat)
goto out;
- if (dev_get_valid_name(dev, pat) < 0)
+ if (dev_get_valid_name(net, dev, pat) < 0)
goto out;
}
--
1.7.7.6
^ permalink raw reply related
* Re: [RFC/RFT 14/15] rtlwifi: Modify files for addition of rtl8723ae
From: Kalle Valo @ 2012-09-14 6:10 UTC (permalink / raw)
To: Larry Finger
Cc: Julian Calaby, linville-2XuSBdqkA4R54TAoqtyWWQ,
linux-wireless-u79uwXL29TY76Z2rM5mHXA,
netdev-u79uwXL29TY76Z2rM5mHXA, chaoming_li-kXabqFNEczNtrwSWzY7KCg
In-Reply-To: <505170C6.8090003-tQ5ms3gMjBLk1uMJSBkQmQ@public.gmane.org>
Larry Finger <Larry.Finger-tQ5ms3gMjBLk1uMJSBkQmQ@public.gmane.org> writes:
>> Is this supposed to be bit #29 - I notice that COMP_USB is also bit 29.
>
> Yes, that is OK. One will only be used for PCI-based drivers, and the
> other is obviously for USB. As nearly all the bits of a 32-bit
> quantity are used, I wanted to save one if possible.
>
> In the final version, I'll code this as
>
> #define COMP_USB BIT(29)
> #define COMP_EASY_CONCURRENT COMP_USB
>
> That way will be more obvious.
You could also add a comment explaining it, just to make sure that
everyone understands it.
--
Kalle Valo
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* [PATCH net-next] ipv6: recursive check rt->dst.from when call rt6_check_expired
From: roy.qing.li @ 2012-09-14 5:54 UTC (permalink / raw)
To: gaofeng, netdev
From: Li RongQing <roy.qing.li@gmail.com>
If dst cache dst_a copies from dst_b, and dst_b copies from dst_c, check
if dst_a is expired or not, we should not end with dst_a->dst.from, dst_b,
we should check dst_c.
CC: Gao feng <gaofeng@cn.fujitsu.com>
Signed-off-by: Li RongQing <roy.qing.li@gmail.com>
---
net/ipv6/route.c | 6 +-----
1 files changed, 1 insertions(+), 5 deletions(-)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 83dafa5..0607ee3 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -369,15 +369,11 @@ static void ip6_dst_ifdown(struct dst_entry *dst, struct net_device *dev,
static bool rt6_check_expired(const struct rt6_info *rt)
{
- struct rt6_info *ort = NULL;
-
if (rt->rt6i_flags & RTF_EXPIRES) {
if (time_after(jiffies, rt->dst.expires))
return true;
} else if (rt->dst.from) {
- ort = (struct rt6_info *) rt->dst.from;
- return (ort->rt6i_flags & RTF_EXPIRES) &&
- time_after(jiffies, ort->dst.expires);
+ return rt6_check_expired((struct rt6_info *) rt->dst.from);
}
return false;
}
--
1.7.4.1
^ permalink raw reply related
* Re: [PATCH] ipconfig: Inform user if carrier is not ready
From: Francois Romieu @ 2012-09-14 5:36 UTC (permalink / raw)
To: Erwan Velu; +Cc: David Miller, netdev
In-Reply-To: <50525758.1090609@gmail.com>
Erwan Velu <erwanaliasr1@gmail.com> :
[...]
> This patch is just adding a simple message every second telling we are
> waiting the carrier to come up.
> ---
> net/ipv4/ipconfig.c | 8 ++++++++
> 1 file changed, 8 insertions(+)
The Signed-off-by: line is missing.
> diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c
> index 67e8a6b..d9f34b7 100644
> --- a/net/ipv4/ipconfig.c
> +++ b/net/ipv4/ipconfig.c
> @@ -205,6 +205,7 @@ static int __init ic_open_devs(void)
> struct net_device *dev;
> unsigned short oflags;
> unsigned long start;
> + unsigned int loops=0;
(nit)
unsigned int loops = 0;
> last = &ic_first_dev;
> rtnl_lock();
> @@ -266,6 +267,13 @@ static int __init ic_open_devs(void)
> if (ic_is_init_dev(dev) && netif_carrier_ok(dev))
> goto have_carrier;
>
> + loops++;
> + /* This loop is blocking the boot process until we get the
> carrier or reach the timeout.
Please split it into 80 cols max lines.
[...]
> + * Every second, we display a short message indicating we
> wait the carrier */
(you can remove this part of the comment)
--
Ueimor
^ permalink raw reply
* Re: [PATCH net-next] r8169: use unlimited DMA burst for TX
From: Francois Romieu @ 2012-09-14 5:19 UTC (permalink / raw)
To: Michal Schmidt; +Cc: netdev, Hayes Wang, Ivan Vecera
In-Reply-To: <1347234926-5263-1-git-send-email-mschmidt@redhat.com>
Michal Schmidt <mschmidt@redhat.com> :
[...]
> Signed-off-by: Michal Schmidt <mschmidt@redhat.com>
Acked-by: Francois Romieu <romieu@fr.zoreil.com>
--
Ueimor
^ 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