* Re: [PATCH v2 net-next] net: ipv6: Add a sysctl to make optimistic addresses useful candidates
From: Hannes Frederic Sowa @ 2014-10-29 14:37 UTC (permalink / raw)
To: Erik Kline; +Cc: netdev, davem, ben, lorenzo
In-Reply-To: <1414487474-18201-1-git-send-email-ek@google.com>
On Di, 2014-10-28 at 18:11 +0900, Erik Kline wrote:
> Add a sysctl that causes an interface's optimistic addresses
> to be considered equivalent to other non-deprecated addresses
> for source address selection purposes. Preferred addresses
> will still take precedence over optimistic addresses, subject
> to other ranking in the source address selection algorithm.
>
> This is useful where different interfaces are connected to
> different networks from different ISPs (e.g., a cell network
> and a home wifi network).
>
> The current behaviour complies with RFC 3484/6724, and it
> makes sense if the host has only one interface, or has
> multiple interfaces on the same network (same or cooperating
> administrative domain(s), but not in the multiple distinct
> networks case.
>
> For example, if a mobile device has an IPv6 address on an LTE
> network and then connects to IPv6-enabled wifi, while the wifi
> IPv6 address is undergoing DAD, IPv6 connections will try use
> the wifi default route with the LTE IPv6 address, and will get
> stuck until they time out.
>
> Also, because optimistic nodes can receive frames, issue
> an RTM_NEWADDR as soon as DAD starts (with the IFA_F_OPTIMSTIC
> flag appropriately set). A second RTM_NEWADDR is sent if DAD
> completes (the address flags have changed), otherwise an
> RTM_DELADDR is sent.
>
> Also: add an entry in ip-sysctl.txt for optimistic_dad.
>
> Signed-off-by: Erik Kline <ek@google.com>
> ---
>
> [...]
>
> +static inline bool ipv6_use_optimistic_addr(struct inet6_dev *idev)
> +{
> +#ifdef CONFIG_IPV6_OPTIMISTIC_DAD
> + return idev && idev->cnf.optimistic_dad && idev->cnf.use_optimistic;
Just a small nit: is this idev != NULL check necessary?
Otherwise:
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Thanks,
Hannes
^ permalink raw reply
* [RFC] use smp_load_acquire()/smp_store_release()
From: Eric Dumazet @ 2014-10-29 14:49 UTC (permalink / raw)
To: Alexander Duyck; +Cc: netdev, Stephen Ko
Hi Alexander
The memory barriers added in commit
b37c0fbe3f6dfba1f8ad2aed47fb40578a254635
("net: Add memory barriers to prevent possible race in byte queue
limits")
have heavy cost.
It seems we could use smp_load_acquire() and smp_store_release()
instead ?
I'll post a patch later today. I would be interested if someone was able
to test it, as your commit apparently was tested and known to fix a
reproducible race.
Thanks !
^ permalink raw reply
* Re: some failures with vxlan offloads..
From: Tom Herbert @ 2014-10-29 14:59 UTC (permalink / raw)
To: Or Gerlitz; +Cc: netdev@vger.kernel.org
In-Reply-To: <54508020.1040305@mellanox.com>
On Tue, Oct 28, 2014 at 10:50 PM, Or Gerlitz <ogerlitz@mellanox.com> wrote:
> On 10/28/2014 5:36 PM, Tom Herbert wrote:
>>>
>>> I wonder if we have another bug somewhere... when both sides were
>>> offloaded,
>>> >it works even with the mlx4 bug, canyou explain that?is it possible that
>>> > the
>>> >GRO stack somehow covers on the bug when both sides are offloaded and
>>> >GRO/VXLAN comes into play?
>>>
>> Look at the receive side. As I mentioned, if the device is returning
>> checksum-unnecessary and setting csum_level to 1 (inner checksum was
>> validated) then stack won't try to verify the outer checksum. So in
>> this case if outer checksum is incorrect nobody complains about it.
>
>
> OK, I'll look there. Anything that should worries us at that stack trace I
> sent in my initial email of this thread, or you think this is related to the
> mlx4 driver checksum bug?
>
The trace doesn't seem like it would be related to a checksum bug. Do
you only see this with offload enabled?
> Or.
>
>>
>
^ permalink raw reply
* Re: [PATCH net-next] net: introduce napi_schedule_irqoff()
From: Alexei Starovoitov @ 2014-10-29 15:26 UTC (permalink / raw)
To: Eric Dumazet; +Cc: David Miller, netdev
In-Reply-To: <1414563627.631.75.camel@edumazet-glaptop2.roam.corp.google.com>
On Tue, Oct 28, 2014 at 11:20 PM, Eric Dumazet <eric.dumazet@gmail.com> wrote:
> On Tue, 2014-10-28 at 22:13 -0700, Alexei Starovoitov wrote:
>
>> tried 50 parallel netperf -t TCP_RR over ixgbe
>> and perf top were tcp stack bits, qdisc locks and netperf itself.
>> What do you see?
>
> You are kidding right ?
>
> If you save 30 nsec ( 2 * 15 nsec) per transaction, and rtt is about 20
> usec, its a 0.15 % gain. Not bad for a trivial patch.
agreed.
I wasn't arguing against the patch at all. Was just curious
what performance gain we'll see. 0.15% is tiny and some might
say the code bloat is not worth it, but imo it's a good one. ack.
> Every atomic op we remove/avoid, every irq masking unmasking we remove,
> every cache line miss or extra bus transaction we remove, TLB miss, is
> the path for better latency.
yes. We're saying the same thing.
^ permalink raw reply
* Re: [PATCH] mac80211_hwsim: release driver when ieee80211_register_hw fails
From: Johannes Berg @ 2014-10-29 15:31 UTC (permalink / raw)
To: Junjie Mao
Cc: Martin Pitt, Fengguang Wu, linux-wireless, netdev, linux-kernel
In-Reply-To: <1414459907-7509-1-git-send-email-eternal.n08@gmail.com>
On Tue, 2014-10-28 at 09:31 +0800, Junjie Mao wrote:
> The driver is not released when ieee80211_register_hw fails in
> mac80211_hwsim_create_radio, leading to the access to the unregistered (and
> possibly freed) device in platform_driver_unregister:
Applied.
johannes
^ permalink raw reply
* Re: [patch] SUNRPC: off by one in BUG_ON()
From: J. Bruce Fields @ 2014-10-29 15:38 UTC (permalink / raw)
To: Dan Carpenter
Cc: Trond Myklebust, David S. Miller,
linux-nfs-u79uwXL29TY76Z2rM5mHXA, netdev-u79uwXL29TY76Z2rM5mHXA,
kernel-janitors-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <20141029084416.GC8939@mwanda>
On Wed, Oct 29, 2014 at 11:44:16AM +0300, Dan Carpenter wrote:
> The m->pool_to[] array has "maxpools" number of elements. It's
> allocated in svc_pool_map_alloc_arrays() which we called earlier in the
> function. This test should be >= instead of >.
>
> Signed-off-by: Dan Carpenter <dan.carpenter-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>
> ---
> This is very old code, but hopefully the off by one doesn't affect
> runtime.
Yeah, doesn't look like a big deal, but thanks, applying for 3.19.--b.
>
> diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
> index ca8a795..349c98f 100644
> --- a/net/sunrpc/svc.c
> +++ b/net/sunrpc/svc.c
> @@ -189,7 +189,7 @@ svc_pool_map_init_percpu(struct svc_pool_map *m)
> return err;
>
> for_each_online_cpu(cpu) {
> - BUG_ON(pidx > maxpools);
> + BUG_ON(pidx >= maxpools);
> m->to_pool[cpu] = pidx;
> m->pool_to[pidx] = cpu;
> pidx++;
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" 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: some failures with vxlan offloads..
From: Or Gerlitz @ 2014-10-29 15:56 UTC (permalink / raw)
To: Tom Herbert; +Cc: netdev@vger.kernel.org
In-Reply-To: <CA+mtBx9TCtHgW=TegUgjZ6Se=mP+0CNiNHYmbYHPd1cJJGYoBQ@mail.gmail.com>
On 10/29/2014 4:59 PM, Tom Herbert wrote:
>> >OK, I'll look there. Anything that should worries us at that stack trace I
>> >sent in my initial email of this thread, or you think this is related to the
>> >mlx4 driver checksum bug?
>> >
> The trace doesn't seem like it would be related to a checksum bug. Do you only see this with offload enabled?
no, it happened on the server side of configuration #4 in my original
email, which is offloaded client and non-offloaded server.
Or.
^ permalink raw reply
* e1000_netpoll(): disable_irq() triggers might_sleep() on linux-next
From: Sabrina Dubroca @ 2014-10-29 15:56 UTC (permalink / raw)
To: netdev; +Cc: linux-kernel, peterz, jeffrey.t.kirsher
commit e22b886a8a43b ("sched/wait: Add might_sleep() checks") included
in today's linux-next added a check that fires on e1000 with netpoll:
BUG: sleeping function called from invalid context at kernel/irq/manage.c:104
in_atomic(): 1, irqs_disabled(): 1, pid: 1, name: systemd
no locks held by systemd/1.
irq event stamp: 10102965
hardirqs last enabled at (10102965): [<ffffffff810cbafd>] vprintk_emit+0x2dd/0x6a0
hardirqs last disabled at (10102964): [<ffffffff810cb897>] vprintk_emit+0x77/0x6a0
softirqs last enabled at (10102342): [<ffffffff810666aa>] __do_softirq+0x27a/0x6f0
softirqs last disabled at (10102337): [<ffffffff81066e86>] irq_exit+0x56/0xe0
Preemption disabled at:[<ffffffff817de50d>] printk_emit+0x31/0x33
CPU: 1 PID: 1 Comm: systemd Not tainted 3.18.0-rc2-next-20141029-dirty #222
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140617_173321-var-lib-archbuild-testing-x86_64-tobias 04/01/2014
ffffffff81a82291 ffff88001e743978 ffffffff817df31d 0000000000000000
0000000000000000 ffff88001e7439a8 ffffffff8108dfa2 ffff88001e7439a8
ffffffff81a82291 0000000000000068 0000000000000000 ffff88001e7439d8
Call Trace:
[<ffffffff817df31d>] dump_stack+0x4f/0x7c
[<ffffffff8108dfa2>] ___might_sleep+0x182/0x2b0
[<ffffffff8108e10a>] __might_sleep+0x3a/0xc0
[<ffffffff810ce358>] synchronize_irq+0x38/0xa0
[<ffffffff810ce633>] ? __disable_irq_nosync+0x43/0x70
[<ffffffff810ce690>] disable_irq+0x20/0x30
[<ffffffff815d7253>] e1000_netpoll+0x23/0x60
[<ffffffff81678d02>] netpoll_poll_dev+0x72/0x3a0
[<ffffffff817e9993>] ? _raw_spin_trylock+0x73/0x90
[<ffffffff8167920f>] ? netpoll_send_skb_on_dev+0x1df/0x2e0
[<ffffffff816791e7>] netpoll_send_skb_on_dev+0x1b7/0x2e0
[<ffffffff816795f3>] netpoll_send_udp+0x2e3/0x490
[<ffffffff815d1f61>] ? write_msg+0x51/0x140
[<ffffffff815d1fdf>] write_msg+0xcf/0x140
[<ffffffff810cadbb>] call_console_drivers.constprop.22+0x13b/0x2a0
[<ffffffff810cb6bd>] console_unlock+0x39d/0x500
[<ffffffff810cbb3e>] ? vprintk_emit+0x31e/0x6a0
[<ffffffff810cbb5c>] vprintk_emit+0x33c/0x6a0
[<ffffffff811a6c6e>] ? might_fault+0x5e/0xc0
[<ffffffff817de50d>] printk_emit+0x31/0x33
[<ffffffff810cbfad>] devkmsg_write+0xbd/0x110
[<ffffffff811f24d5>] do_iter_readv_writev+0x65/0xa0
[<ffffffff811f3b72>] do_readv_writev+0xe2/0x290
[<ffffffff810cbef0>] ? vprintk+0x30/0x30
[<ffffffff810d499d>] ? rcu_read_lock_held+0x6d/0x70
[<ffffffff812142d6>] ? __fget_light+0xc6/0xd0
[<ffffffff811f3dac>] vfs_writev+0x3c/0x50
[<ffffffff811f3eed>] SyS_writev+0x4d/0xe0
[<ffffffff817ea16d>] system_call_fastpath+0x16/0x1b
I'm able to reproduce it consistently by sending a lot of packets from
that interface while writing to /dev/kmsg with netconsole
enabled. Just writing to /dev/kmsg isn't enough.
# with ping -f or pktgen running
for i in `seq 1 20` ; do echo '............................................' > /dev/kmsg ; done
--
Sabrina
^ permalink raw reply
* Fw: [Bug 87111] New: hlist_for_each_entry_rcu() returns invalid pointer causing kernel to OOPS
From: Stephen Hemminger @ 2014-10-29 15:57 UTC (permalink / raw)
To: netdev
Begin forwarded message:
Date: Wed, 29 Oct 2014 07:16:13 -0700
From: "bugzilla-daemon@bugzilla.kernel.org" <bugzilla-daemon@bugzilla.kernel.org>
To: "stephen@networkplumber.org" <stephen@networkplumber.org>
Subject: [Bug 87111] New: hlist_for_each_entry_rcu() returns invalid pointer causing kernel to OOPS
https://bugzilla.kernel.org/show_bug.cgi?id=87111
Bug ID: 87111
Summary: hlist_for_each_entry_rcu() returns invalid pointer
causing kernel to OOPS
Product: Networking
Version: 2.5
Kernel Version: 2.6.32.24
Hardware: x86-64
OS: Linux
Tree: Mainline
Status: NEW
Severity: high
Priority: P1
Component: IPV4
Assignee: shemminger@linux-foundation.org
Reporter: jith131986@gmail.com
Regression: No
Created attachment 155781
--> https://bugzilla.kernel.org/attachment.cgi?id=155781&action=edit
nf_nat.ko objdump for analysing IP and offset to see exact line where kernel
panic happened
In my setup linux stack is only used for layer 2 network services. when layer 2
packet is recieved by linux for layer 2 functionality, in nf_nat kernel module
hlist_for_each_entry_rcu()(where IP points) function return an invalid pointer
resulting in Oops panic. I have attached panic dump and nf_nat.ko objdump for
further analysis.
Would like to know the issue is seen/reported before and fixed ?. If not is it
possible to get cause or solution for the same.
Pasting the panic dump below and attaching nf_nat.ko objdump
<1>BUG: unable to handle kernel NULL pointer dereference at 000000000000003e
<1>IP: [<ffffffffa003794b>] nf_nat_setup_info+0x1ab/0x740 [nf_nat]
<6>PGD 641576067 PUD 7dd9f3067 PMD 0
<0>Oops: 0000 [#1] PREEMPT SMP
<0>last sysfs file:
/sys/devices/pci0000:00/0000:00:1d.7/usb2/2-1/2-1:1.0/host5/scsi_host/host5/proc_name
<6>CPU 3
<6>Modules linked in: bridge stp llc ixgbe igb ftdi_sio usbserial xt_connlimit
xt_tcpudp xt_mark iptable_nat nf_nat nf_conntrack_ipv4 nf_defrag_ipv4
nf_conntrack iptable_filter ip_tables x_tables
<6>Pid: 0, comm: swapper Tainted: P W 2.6.32.24 #1 S5520UR
<6>RIP: e030:[<ffffffffa003794b>] [<ffffffffa003794b>]
nf_nat_setup_info+0x1ab/0x740 [nf_nat]
<6>RSP: e02b:ffff88002808d910 EFLAGS: 00010282
<6>RAX: 0000000000000000 RBX: ffff880381313b58 RCX: 0000000000000000
<6>RDX: 0000000000000018 RSI: 000000007049f4f6 RDI: ffff88002808d9b0
<6>RBP: ffff88002808da10 R08: ffffffff81393e80 R09: ffffffffa0040790
<6>R10: 0000000000004000 R11: 000000000000002c R12: ffff88002808da20
<6>R13: ffff8807fc8ebfd8 R14: ffff880396c3bb70 R15: 0000000000000000
<6>FS: 00007fde2cd296f0(0000) GS:ffff88002808a000(0000) knlGS:0000000000000000
<6>CS: e033 DS: 002b ES: 002b CR0: 000000008005003b
<6>CR2: 000000000000003e CR3: 000000079ab27000 CR4: 0000000000002660
<6>DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
<6>DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
<6>Process swapper (pid: 0, threadinfo ffff8807fc8ea000, task ffff8807fc8da050)
<0>Stack:
<6> 0000000000000000 ffff88002808d980 ffff88002808da2c ffff88002808da2e
<6><0> ffff8807fc8ea000 ffff8807fc8ebfd8 0000000000000100 0000000000000100
<6><0> 0000000000000000 0000000000010001 00000000002ace3f ffff88002809a720
<0>Call Trace:
<0> <IRQ>
<6> [<ffffffff81048b07>] ? local_bh_enable+0x77/0xc0
<6> [<ffffffffa0009945>] ? ipt_do_table+0x2a5/0x3e0 [ip_tables]
<6> [<ffffffffa00400cf>] alloc_null_binding+0x3f/0x70 [iptable_nat]
<6> [<ffffffffa00402fb>] nf_nat_rule_find+0x1fb/0x390 [iptable_nat]
<6> [<ffffffff8138ca3f>] nf_iterate+0x5f/0x90
<6> [<ffffffff81393e80>] ? ip_local_deliver_finish+0x0/0x1e0
<6> [<ffffffff8138cdb0>] nf_hook_slow+0xb0/0x110
<6> [<ffffffff81393e80>] ? ip_local_deliver_finish+0x0/0x1e0
<6> [<ffffffff81394559>] ip_local_deliver+0x69/0x90
<6> [<ffffffff81393ba6>] ip_rcv_finish+0x146/0x420
<6> [<ffffffff8139440d>] ip_rcv+0x27d/0x360
<6> [<ffffffff81371747>] netif_receive_skb+0x2b7/0x390
<6> [<ffffffffa12cce50>] br_handle_frame_finish+0x130/0x170 [bridge]
<6> [<ffffffffa12d1458>] br_netfilter_fini+0x6a8/0x810 [bridge]
<6> [<ffffffff8138cdb0>] ? nf_hook_slow+0xb0/0x110
<6> [<ffffffffa12d1270>] ? br_netfilter_fini+0x4c0/0x810 [bridge]
<6> [<ffffffffa12d2389>] nf_bridge_copy_header+0xdc9/0x10e0 [bridge]
<6> [<ffffffff8138ca3f>] nf_iterate+0x5f/0x90
<6> [<ffffffffa12ccd20>] ? br_handle_frame_finish+0x0/0x170 [bridge]
<6> [<ffffffff8138cdb0>] nf_hook_slow+0xb0/0x110
<6> [<ffffffffa12ccd20>] ? br_handle_frame_finish+0x0/0x170 [bridge]
<6> [<ffffffffa12ccfe6>] br_handle_frame+0x156/0x2b0 [bridge]
<6> [<ffffffff813f2ab8>] ? vlan_skb_recv+0x1a8/0x2f0
<6> [<ffffffff81371699>] netif_receive_skb+0x209/0x390
<6> [<ffffffff81374d79>] process_backlog+0x89/0xc0
<6> [<ffffffff81374b7f>] net_rx_action+0x7f/0x160
<6> [<ffffffffa0078165>] ? igb_reinit_locked+0x1995/0x2900 [igb]
<6> [<ffffffff810484f8>] __do_softirq+0xa8/0x130
<6> [<ffffffff810755a8>] ? handle_level_irq+0xe8/0x130
<6> [<ffffffff81014efc>] call_softirq+0x1c/0x30
<6> [<ffffffff81016765>] do_softirq+0x65/0xa0
<6> [<ffffffff81048358>] irq_exit+0x48/0x50
<6> [<ffffffff81228ddd>] xen_evtchn_do_upcall+0x3d/0x60
<6> [<ffffffff81014f4e>] xen_do_hypervisor_callback+0x1e/0x30
<0> <EOI>
<6> [<ffffffff810093aa>] ? hypercall_page+0x3aa/0x1010
<6> [<ffffffff810093aa>] ? hypercall_page+0x3aa/0x1010
<6> [<ffffffff8100f8d0>] ? xen_safe_halt+0x10/0x20
<6> [<ffffffff8100c4d5>] ? xen_idle+0x45/0x70
<6> [<ffffffff81012d78>] ? cpu_idle+0x58/0x90
<6> [<ffffffff810101c9>] ? xen_irq_enable_direct_end+0x0/0x7
<6> [<ffffffff8140a86e>] ? cpu_bringup_and_idle+0xe/0x10
<0>Code: ff ff ff 49 8d 44 24 0c 48 89 85 10 ff ff ff eb 0c 48 8b 1b 48 85 db
0f 84 f1 00 00 00 48 8b 4b 20 48 8b 03 48 8d 51 18 0f 18 08 <0f> b6 42 26 3a 45
c6 75 dd 8b 02 3b 45 a0 75 d6 0f b7 42 10 66
<1>RIP [<ffffffffa003794b>] nf_nat_setup_info+0x1ab/0x740 [nf_nat]
<6> RSP <ffff88002808d910>
<0>CR2: 000000000000003e
WARN paging error trying to follow 0x0000000000000000 - level 2, cr3
000000058ea67000
--
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply
* Re: [PATCH 1/1 net-next] cfg80211: fix set but not used warning in nl80211_channel_switch()
From: Johannes Berg @ 2014-10-29 16:04 UTC (permalink / raw)
To: Fabian Frederick
Cc: linux-kernel, John W. Linville, David S. Miller, linux-wireless,
netdev
In-Reply-To: <1414252655-20506-1-git-send-email-fabf@skynet.be>
On Sat, 2014-10-25 at 17:57 +0200, Fabian Frederick wrote:
> radar_detect_width is unused since commit 97dc94f1d933
> ("cfg80211: remove channel_switch combination check")
Applied, thanks.
johannes
^ permalink raw reply
* [patch] iwlwifi: cleanup a mask shift in iwlagn_bt_traffic_is_sco()
From: Dan Carpenter @ 2014-10-29 16:08 UTC (permalink / raw)
To: Johannes Berg
Cc: Emmanuel Grumbach, Intel Linux Wireless, John W. Linville,
Paul Gortmaker, Larry Finger, linux-wireless, netdev,
linux-kernel, kernel-janitors
The shift operation is higher precedence so the code is wrong and it
sets of a static checker warning. But it doesn't affect real life
because BT_UART_MSG_FRAME3SCOESCO_POS is zero so the shift is a no-op.
I have re-written it in normal style and with parenthesis as a cleanup
and to silence the static checker warning.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
diff --git a/drivers/net/wireless/iwlwifi/dvm/lib.c b/drivers/net/wireless/iwlwifi/dvm/lib.c
index 2191621..065d3d5 100644
--- a/drivers/net/wireless/iwlwifi/dvm/lib.c
+++ b/drivers/net/wireless/iwlwifi/dvm/lib.c
@@ -418,7 +418,7 @@ void iwlagn_bt_adjust_rssi_monitor(struct iwl_priv *priv, bool rssi_ena)
static bool iwlagn_bt_traffic_is_sco(struct iwl_bt_uart_msg *uart_msg)
{
- return BT_UART_MSG_FRAME3SCOESCO_MSK & uart_msg->frame3 >>
+ return (uart_msg->frame3 & BT_UART_MSG_FRAME3SCOESCO_MSK) >>
BT_UART_MSG_FRAME3SCOESCO_POS;
}
^ permalink raw reply related
* [patch] Bluetooth: 6lowpan: use after free in disconnect_devices()
From: Dan Carpenter @ 2014-10-29 16:10 UTC (permalink / raw)
To: Marcel Holtmann
Cc: Gustavo Padovan, Johan Hedberg, David S. Miller, linux-bluetooth,
netdev, linux-kernel, kernel-janitors
This was accidentally changed from list_for_each_entry_safe() to
list_for_each_entry() so now it has a use after free bug. I've changed
it back.
Fixes: 90305829635d ('Bluetooth: 6lowpan: Converting rwlocks to use RCU')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
diff --git a/net/bluetooth/6lowpan.c b/net/bluetooth/6lowpan.c
index 7254bdd..eef298d 100644
--- a/net/bluetooth/6lowpan.c
+++ b/net/bluetooth/6lowpan.c
@@ -1383,7 +1383,7 @@ static const struct file_operations lowpan_control_fops = {
static void disconnect_devices(void)
{
- struct lowpan_dev *entry, *new_dev;
+ struct lowpan_dev *entry, *tmp, *new_dev;
struct list_head devices;
INIT_LIST_HEAD(&devices);
@@ -1408,7 +1408,7 @@ static void disconnect_devices(void)
rcu_read_unlock();
- list_for_each_entry(entry, &devices, list) {
+ list_for_each_entry_safe(entry, tmp, &devices, list) {
ifdown(entry->netdev);
BT_DBG("Unregistering netdev %s %p",
entry->netdev->name, entry->netdev);
^ permalink raw reply related
* Re: [patch] iwlwifi: cleanup a mask shift in iwlagn_bt_traffic_is_sco()
From: Emmanuel Grumbach @ 2014-10-29 16:16 UTC (permalink / raw)
To: Dan Carpenter
Cc: Johannes Berg, Emmanuel Grumbach, Intel Linux Wireless,
John W. Linville, Paul Gortmaker, Larry Finger, linux-wireless,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
kernel-janitors
In-Reply-To: <20141029160827.GD5290@mwanda>
On Wed, Oct 29, 2014 at 6:08 PM, Dan Carpenter <dan.carpenter@oracle.com> wrote:
> The shift operation is higher precedence so the code is wrong and it
> sets of a static checker warning. But it doesn't affect real life
> because BT_UART_MSG_FRAME3SCOESCO_POS is zero so the shift is a no-op.
>
> I have re-written it in normal style and with parenthesis as a cleanup
> and to silence the static checker warning.
>
> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
>
In my tree already - I got it from Joe.
https://git.kernel.org/cgit/linux/kernel/git/iwlwifi/iwlwifi-next.git/commit/?id=50f6635afe565a0e1c5ab78f040294fe1dc41de0
^ permalink raw reply
* Re: [RFC] use smp_load_acquire()/smp_store_release()
From: Alexander Duyck @ 2014-10-29 16:16 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev, jeffrey.t.kirsher
In-Reply-To: <1414594159.631.85.camel@edumazet-glaptop2.roam.corp.google.com>
On 10/29/2014 07:49 AM, Eric Dumazet wrote:
> Hi Alexander
>
> The memory barriers added in commit
> b37c0fbe3f6dfba1f8ad2aed47fb40578a254635
> ("net: Add memory barriers to prevent possible race in byte queue
> limits")
>
> have heavy cost.
>
> It seems we could use smp_load_acquire() and smp_store_release()
> instead ?
>
> I'll post a patch later today. I would be interested if someone was able
> to test it, as your commit apparently was tested and known to fix a
> reproducible race.
>
> Thanks !
Unfortunately Stephen left Intel before I did, so we will need to find
someone else in the validation team to test this if possible. I have
added Jeff to the CC so that he can give the appropriate validation
people a heads up that this patch might be coming.
As I recall what was seen was random Tx hangs on systems with the
original BQL code when interfaces were stressed. It has been a while so
I don't recall the exact set-up for all of it. Also some less
used/tested architectures such as PowerPC can be more susceptible to
synchronization issues such as these as the memory model is more weakly
ordered.
I'm wondering where you are seeing the barrier show up? In
netdev_tx_send_queue you should only hit the barrier if you actually are
triggering the XOFF condition, and in netdev_tx_completed_queue the
barrier should be coalesced in amongst a number of frames reducing the cost.
My concern with this would be that we are actually syncronizing multiple
things, the __QUEUE_STATE_STACK_XOFF flag, dql->adj_limit, and
dql->num_queued, and we might be trading off reducing the cost on x86 to
result in it being increased on other architectures as we may have to
actually add additional synchronization as I suspect we would need to
use acquire/release on both adj_limit and num_queued.
Thanks,
Alex
^ permalink raw reply
* Re: [PATCH] netdev: Fix sleeping inside wait event
From: Cong Wang @ 2014-10-29 16:29 UTC (permalink / raw)
To: Peter Zijlstra
Cc: Fengguang Wu, LKP, LKML, oleg@redhat.com, Eric W. Biederman,
David Miller, Linux Kernel Network Developers
In-Reply-To: <20141029161657.GF3337@twins.programming.kicks-ass.net>
(Adding netdev@...)
On Wed, Oct 29, 2014 at 9:16 AM, Peter Zijlstra <peterz@infradead.org> wrote:
>
> Dave, this relies on bits currently in tip/sched/core, if you're ok I'll
> merge it through that tree.
>
> ---
> Subject: netdev: Fix sleeping inside wait event
> From: Peter Zijlstra <peterz@infradead.org>
> Date: Wed Oct 29 17:04:56 CET 2014
>
> rtnl_lock_unregistering() takes rtnl_lock() -- a mutex -- inside a
> wait loop. The wait loop relies on current->state to function, but so
> does mutex_lock(), nesting them makes for the inner to destroy the
> outer state.
>
While you are on it, please fix rtnl_lock_unregistering_all() too?
Thanks!
> Fix this using the new wait_woken() bits.
>
> Cc: Oleg Nesterov <oleg@redhat.com>
> Cc: Eric Biederman <ebiederm@xmission.com>
> Cc: David Miller <davem@davemloft.net>
> Reported-by: Fengguang Wu <fengguang.wu@intel.com>
> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> ---
> net/core/dev.c | 10 +++++-----
> 1 file changed, 5 insertions(+), 5 deletions(-)
>
> --- a/net/core/dev.c
> +++ b/net/core/dev.c
> @@ -7196,11 +7196,10 @@ static void __net_exit rtnl_lock_unregis
> */
> struct net *net;
> bool unregistering;
> - DEFINE_WAIT(wait);
> + DEFINE_WAIT_FUNC(wait, woken_wake_function);
>
> + add_wait_queue(&netdev_unregistering_wq, &wait);
> for (;;) {
> - prepare_to_wait(&netdev_unregistering_wq, &wait,
> - TASK_UNINTERRUPTIBLE);
> unregistering = false;
> rtnl_lock();
> list_for_each_entry(net, net_list, exit_list) {
> @@ -7212,9 +7211,10 @@ static void __net_exit rtnl_lock_unregis
> if (!unregistering)
> break;
> __rtnl_unlock();
> - schedule();
> +
> + wait_woken(&wait, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
> }
> - finish_wait(&netdev_unregistering_wq, &wait);
> + remove_wait_queue(&netdev_unregistering_wq, &wait);
> }
>
> static void __net_exit default_device_exit_batch(struct list_head *net_list)
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
^ permalink raw reply
* Re: [PATCH] PPC: bpf_jit_comp: add SKF_AD_PKTTYPE instruction
From: Alexei Starovoitov @ 2014-10-29 17:08 UTC (permalink / raw)
To: Denis Kirjanov; +Cc: linuxppc-dev, Matt Evans, netdev@vger.kernel.org
In-Reply-To: <CAOJe8K0t3G-bHm_24GjrTp9mmKnYZS1_bdGgrwQBLjC_s5is6w@mail.gmail.com>
On Wed, Oct 29, 2014 at 2:21 AM, Denis Kirjanov <kda@linux-powerpc.org> wrote:
> Any feedback from PPC folks?
not a ppc guy, but looks reasonable to me.
What lib/test_bpf says? Like performance difference before/after
for LD_PKTTYPE test...
^ permalink raw reply
* Re: [PATCH] netdev: Fix sleeping inside wait event
From: Peter Zijlstra @ 2014-10-29 17:13 UTC (permalink / raw)
To: Cong Wang
Cc: Fengguang Wu, LKP, LKML, oleg@redhat.com, Eric W. Biederman,
David Miller, Linux Kernel Network Developers
In-Reply-To: <CAM_iQpXTyBT_4TrnYAoAuAE_jMto=2B=cutFggpHRMhvSrG7qg@mail.gmail.com>
On Wed, Oct 29, 2014 at 09:29:55AM -0700, Cong Wang wrote:
> (Adding netdev@...)
>
> On Wed, Oct 29, 2014 at 9:16 AM, Peter Zijlstra <peterz@infradead.org> wrote:
> >
> > Dave, this relies on bits currently in tip/sched/core, if you're ok I'll
> > merge it through that tree.
> >
> > ---
> > Subject: netdev: Fix sleeping inside wait event
> > From: Peter Zijlstra <peterz@infradead.org>
> > Date: Wed Oct 29 17:04:56 CET 2014
> >
> > rtnl_lock_unregistering() takes rtnl_lock() -- a mutex -- inside a
> > wait loop. The wait loop relies on current->state to function, but so
> > does mutex_lock(), nesting them makes for the inner to destroy the
> > outer state.
> >
>
> While you are on it, please fix rtnl_lock_unregistering_all() too?
Ah, that's hidden someplace else, sure I can do that. Thanks for
pointing it out.
^ permalink raw reply
* Re: [PATCH] netdev: Fix sleeping inside wait event
From: Peter Zijlstra @ 2014-10-29 17:31 UTC (permalink / raw)
To: Cong Wang
Cc: Fengguang Wu, LKP, LKML, oleg@redhat.com, Eric W. Biederman,
David Miller, Linux Kernel Network Developers
In-Reply-To: <20141029171345.GO12706@worktop.programming.kicks-ass.net>
On Wed, Oct 29, 2014 at 06:13:45PM +0100, Peter Zijlstra wrote:
> On Wed, Oct 29, 2014 at 09:29:55AM -0700, Cong Wang wrote:
> > While you are on it, please fix rtnl_lock_unregistering_all() too?
>
> Ah, that's hidden someplace else, sure I can do that. Thanks for
> pointing it out.
Here goes..
---
Subject: netdev: Fix sleeping inside wait event
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed Oct 29 17:04:56 CET 2014
rtnl_lock_unregistering*() take rtnl_lock() -- a mutex -- inside a
wait loop. The wait loop relies on current->state to function, but so
does mutex_lock(), nesting them makes for the inner to destroy the
outer state.
Fix this using the new wait_woken() bits.
Cc: David Miller <davem@davemloft.net>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
net/core/dev.c | 10 +++++-----
net/core/rtnetlink.c | 10 +++++-----
2 files changed, 10 insertions(+), 10 deletions(-)
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -7196,11 +7196,10 @@ static void __net_exit rtnl_lock_unregis
*/
struct net *net;
bool unregistering;
- DEFINE_WAIT(wait);
+ DEFINE_WAIT_FUNC(wait, woken_wake_function);
+ add_wait_queue(&netdev_unregistering_wq, &wait);
for (;;) {
- prepare_to_wait(&netdev_unregistering_wq, &wait,
- TASK_UNINTERRUPTIBLE);
unregistering = false;
rtnl_lock();
list_for_each_entry(net, net_list, exit_list) {
@@ -7212,9 +7211,10 @@ static void __net_exit rtnl_lock_unregis
if (!unregistering)
break;
__rtnl_unlock();
- schedule();
+
+ wait_woken(&wait, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
}
- finish_wait(&netdev_unregistering_wq, &wait);
+ remove_wait_queue(&netdev_unregistering_wq, &wait);
}
static void __net_exit default_device_exit_batch(struct list_head *net_list)
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -365,11 +365,10 @@ static void rtnl_lock_unregistering_all(
{
struct net *net;
bool unregistering;
- DEFINE_WAIT(wait);
+ DEFINE_WAIT_FUNC(wait, woken_wake_function);
+ add_wait_queue(&netdev_unregistering_wq, &wait);
for (;;) {
- prepare_to_wait(&netdev_unregistering_wq, &wait,
- TASK_UNINTERRUPTIBLE);
unregistering = false;
rtnl_lock();
for_each_net(net) {
@@ -381,9 +380,10 @@ static void rtnl_lock_unregistering_all(
if (!unregistering)
break;
__rtnl_unlock();
- schedule();
+
+ wait_woken(&wait, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
}
- finish_wait(&netdev_unregistering_wq, &wait);
+ remove_wait_queue(&netdev_unregistering_wq, &wait);
}
/**
^ permalink raw reply
* Re: [PATCH] netdev: Fix sleeping inside wait event
From: David Miller @ 2014-10-29 17:38 UTC (permalink / raw)
To: peterz
Cc: xiyou.wangcong, fengguang.wu, lkp, linux-kernel, oleg, ebiederm,
netdev
In-Reply-To: <20141029173110.GE15602@worktop.programming.kicks-ass.net>
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 29 Oct 2014 18:31:10 +0100
> On Wed, Oct 29, 2014 at 06:13:45PM +0100, Peter Zijlstra wrote:
>> On Wed, Oct 29, 2014 at 09:29:55AM -0700, Cong Wang wrote:
>> > While you are on it, please fix rtnl_lock_unregistering_all() too?
>>
>> Ah, that's hidden someplace else, sure I can do that. Thanks for
>> pointing it out.
>
> Here goes..
For this as well:
Acked-by: David S. Miller <davem@davemloft.net>
^ permalink raw reply
* [PATCH v3 01/15] net: dsa: Don't set skb->protocol on outgoing tagged packets
From: Guenter Roeck @ 2014-10-29 17:44 UTC (permalink / raw)
To: netdev
Cc: David S. Miller, Florian Fainelli, Andrew Lunn, linux-kernel,
Guenter Roeck
In-Reply-To: <1414604707-22407-1-git-send-email-linux@roeck-us.net>
Setting skb->protocol to a private protocol type may result in warning
messages such as
e1000e 0000:00:19.0 em1: checksum_partial proto=dada!
This happens if the L3 protocol is IP or IPv6 and skb->ip_summed is set
to CHECKSUM_PARTIAL. Looking through the code, it appears that changing
skb->protocol for transmitted packets is not necessary and may actually
be harmful. For example, it prevents purposely unmodified (from a DSA
perspective) network drivers from properly setting up their transmit
checksum offload pointers since they inspect skb->protocol to set up the
IPv4 header or IPv6 header pointers. So don't unnecessarily change the
protocol field.
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
v3:
- Updated description
v2:
- No change
net/dsa/tag_dsa.c | 2 --
net/dsa/tag_edsa.c | 2 --
net/dsa/tag_trailer.c | 2 --
3 files changed, 6 deletions(-)
diff --git a/net/dsa/tag_dsa.c b/net/dsa/tag_dsa.c
index ce90c8b..2dab270 100644
--- a/net/dsa/tag_dsa.c
+++ b/net/dsa/tag_dsa.c
@@ -63,8 +63,6 @@ static netdev_tx_t dsa_xmit(struct sk_buff *skb, struct net_device *dev)
dsa_header[3] = 0x00;
}
- skb->protocol = htons(ETH_P_DSA);
-
skb->dev = p->parent->dst->master_netdev;
dev_queue_xmit(skb);
diff --git a/net/dsa/tag_edsa.c b/net/dsa/tag_edsa.c
index 94fcce7..9aeda59 100644
--- a/net/dsa/tag_edsa.c
+++ b/net/dsa/tag_edsa.c
@@ -76,8 +76,6 @@ static netdev_tx_t edsa_xmit(struct sk_buff *skb, struct net_device *dev)
edsa_header[7] = 0x00;
}
- skb->protocol = htons(ETH_P_EDSA);
-
skb->dev = p->parent->dst->master_netdev;
dev_queue_xmit(skb);
diff --git a/net/dsa/tag_trailer.c b/net/dsa/tag_trailer.c
index 115fdca..e268f9d 100644
--- a/net/dsa/tag_trailer.c
+++ b/net/dsa/tag_trailer.c
@@ -57,8 +57,6 @@ static netdev_tx_t trailer_xmit(struct sk_buff *skb, struct net_device *dev)
trailer[2] = 0x10;
trailer[3] = 0x00;
- nskb->protocol = htons(ETH_P_TRAILER);
-
nskb->dev = p->parent->dst->master_netdev;
dev_queue_xmit(nskb);
--
1.9.1
^ permalink raw reply related
* [PATCH v3 04/15] net: dsa: Add support for Marvell 88E6352
From: Guenter Roeck @ 2014-10-29 17:44 UTC (permalink / raw)
To: netdev
Cc: David S. Miller, Florian Fainelli, Andrew Lunn, linux-kernel,
Guenter Roeck
In-Reply-To: <1414604707-22407-1-git-send-email-linux@roeck-us.net>
Marvell 88E6352 is mostly compatible to MV88E6123/61/65,
but requires indirect phy access. Also, its configuration
registers are a bit different.
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
v3:
- No change
v2:
- No change
MAINTAINERS | 5 +
drivers/net/dsa/Kconfig | 8 +
drivers/net/dsa/Makefile | 3 +
drivers/net/dsa/mv88e6352.c | 464 ++++++++++++++++++++++++++++++++++++++++++++
drivers/net/dsa/mv88e6xxx.c | 3 +
drivers/net/dsa/mv88e6xxx.h | 7 +
6 files changed, 490 insertions(+)
create mode 100644 drivers/net/dsa/mv88e6352.c
diff --git a/MAINTAINERS b/MAINTAINERS
index dab92a7..91db18e 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5860,6 +5860,11 @@ M: Russell King <rmk+kernel@arm.linux.org.uk>
S: Maintained
F: drivers/gpu/drm/armada/
+MARVELL 88E6352 DSA support
+M: Guenter Roeck <linux@roeck-us.net>
+S: Maintained
+F: drivers/net/dsa/mv88e6352.c
+
MARVELL GIGABIT ETHERNET DRIVERS (skge/sky2)
M: Mirko Lindner <mlindner@marvell.com>
M: Stephen Hemminger <stephen@networkplumber.org>
diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig
index 9234d80..0987c33 100644
--- a/drivers/net/dsa/Kconfig
+++ b/drivers/net/dsa/Kconfig
@@ -45,6 +45,14 @@ config NET_DSA_MV88E6171
This enables support for the Marvell 88E6171 ethernet switch
chip.
+config NET_DSA_MV88E6352
+ tristate "Marvell 88E6352 ethernet switch chip support"
+ select NET_DSA
+ select NET_DSA_MV88E6XXX
+ select NET_DSA_TAG_EDSA
+ ---help---
+ This enables support for the Marvell 88E6352 ethernet switch chip.
+
config NET_DSA_BCM_SF2
tristate "Broadcom Starfighter 2 Ethernet switch support"
depends on HAS_IOMEM
diff --git a/drivers/net/dsa/Makefile b/drivers/net/dsa/Makefile
index 23a90de..e2d51c4 100644
--- a/drivers/net/dsa/Makefile
+++ b/drivers/net/dsa/Makefile
@@ -7,6 +7,9 @@ endif
ifdef CONFIG_NET_DSA_MV88E6131
mv88e6xxx_drv-y += mv88e6131.o
endif
+ifdef CONFIG_NET_DSA_MV88E6352
+mv88e6xxx_drv-y += mv88e6352.o
+endif
ifdef CONFIG_NET_DSA_MV88E6171
mv88e6xxx_drv-y += mv88e6171.o
endif
diff --git a/drivers/net/dsa/mv88e6352.c b/drivers/net/dsa/mv88e6352.c
new file mode 100644
index 0000000..43a5826
--- /dev/null
+++ b/drivers/net/dsa/mv88e6352.c
@@ -0,0 +1,464 @@
+/*
+ * net/dsa/mv88e6352.c - Marvell 88e6352 switch chip support
+ *
+ * Copyright (c) 2014 Guenter Roeck
+ *
+ * Derived from mv88e6123_61_65.c
+ * Copyright (c) 2008-2009 Marvell Semiconductor
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/delay.h>
+#include <linux/jiffies.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/platform_device.h>
+#include <linux/phy.h>
+#include <net/dsa.h>
+#include "mv88e6xxx.h"
+
+static int mv88e6352_phy_wait(struct dsa_switch *ds)
+{
+ unsigned long timeout = jiffies + HZ / 10;
+
+ while (time_before(jiffies, timeout)) {
+ int ret;
+
+ ret = REG_READ(REG_GLOBAL2, 0x18);
+ if (ret < 0)
+ return ret;
+
+ if (!(ret & 0x8000))
+ return 0;
+
+ usleep_range(1000, 2000);
+ }
+ return -ETIMEDOUT;
+}
+
+static int __mv88e6352_phy_read(struct dsa_switch *ds, int addr, int regnum)
+{
+ int ret;
+
+ REG_WRITE(REG_GLOBAL2, 0x18, 0x9800 | (addr << 5) | regnum);
+
+ ret = mv88e6352_phy_wait(ds);
+ if (ret < 0)
+ return ret;
+
+ return REG_READ(REG_GLOBAL2, 0x19);
+}
+
+static int __mv88e6352_phy_write(struct dsa_switch *ds, int addr, int regnum,
+ u16 val)
+{
+ REG_WRITE(REG_GLOBAL2, 0x19, val);
+ REG_WRITE(REG_GLOBAL2, 0x18, 0x9400 | (addr << 5) | regnum);
+
+ return mv88e6352_phy_wait(ds);
+}
+
+static char *mv88e6352_probe(struct device *host_dev, int sw_addr)
+{
+ struct mii_bus *bus = dsa_host_dev_to_mii_bus(host_dev);
+ int ret;
+
+ if (bus == NULL)
+ return NULL;
+
+ ret = __mv88e6xxx_reg_read(bus, sw_addr, REG_PORT(0), 0x03);
+ if (ret >= 0) {
+ if (ret == 0x3521)
+ return "Marvell 88E6352 (A0)";
+ if (ret == 0x3522)
+ return "Marvell 88E6352 (A1)";
+ if ((ret & 0xfff0) == 0x3520)
+ return "Marvell 88E6352";
+ }
+
+ return NULL;
+}
+
+static int mv88e6352_switch_reset(struct dsa_switch *ds)
+{
+ unsigned long timeout;
+ int ret;
+ int i;
+
+ /* Set all ports to the disabled state. */
+ for (i = 0; i < 7; i++) {
+ ret = REG_READ(REG_PORT(i), 0x04);
+ REG_WRITE(REG_PORT(i), 0x04, ret & 0xfffc);
+ }
+
+ /* Wait for transmit queues to drain. */
+ usleep_range(2000, 4000);
+
+ /* Reset the switch. Keep PPU active (bit 14, undocumented).
+ * The PPU needs to be active to support indirect phy register
+ * accesses through global registers 0x18 and 0x19.
+ */
+ REG_WRITE(REG_GLOBAL, 0x04, 0xc000);
+
+ /* Wait up to one second for reset to complete. */
+ timeout = jiffies + 1 * HZ;
+ while (time_before(jiffies, timeout)) {
+ ret = REG_READ(REG_GLOBAL, 0x00);
+ if ((ret & 0x8800) == 0x8800)
+ break;
+ usleep_range(1000, 2000);
+ }
+ if (time_after(jiffies, timeout))
+ return -ETIMEDOUT;
+
+ return 0;
+}
+
+static int mv88e6352_setup_global(struct dsa_switch *ds)
+{
+ int ret;
+ int i;
+
+ /* Discard packets with excessive collisions,
+ * mask all interrupt sources, enable PPU (bit 14, undocumented).
+ */
+ REG_WRITE(REG_GLOBAL, 0x04, 0x6000);
+
+ /* Set the default address aging time to 5 minutes, and
+ * enable address learn messages to be sent to all message
+ * ports.
+ */
+ REG_WRITE(REG_GLOBAL, 0x0a, 0x0148);
+
+ /* Configure the priority mapping registers. */
+ ret = mv88e6xxx_config_prio(ds);
+ if (ret < 0)
+ return ret;
+
+ /* Configure the upstream port, and configure the upstream
+ * port as the port to which ingress and egress monitor frames
+ * are to be sent.
+ */
+ REG_WRITE(REG_GLOBAL, 0x1a, (dsa_upstream_port(ds) * 0x1110));
+
+ /* Disable remote management for now, and set the switch's
+ * DSA device number.
+ */
+ REG_WRITE(REG_GLOBAL, 0x1c, ds->index & 0x1f);
+
+ /* Send all frames with destination addresses matching
+ * 01:80:c2:00:00:2x to the CPU port.
+ */
+ REG_WRITE(REG_GLOBAL2, 0x02, 0xffff);
+
+ /* Send all frames with destination addresses matching
+ * 01:80:c2:00:00:0x to the CPU port.
+ */
+ REG_WRITE(REG_GLOBAL2, 0x03, 0xffff);
+
+ /* Disable the loopback filter, disable flow control
+ * messages, disable flood broadcast override, disable
+ * removing of provider tags, disable ATU age violation
+ * interrupts, disable tag flow control, force flow
+ * control priority to the highest, and send all special
+ * multicast frames to the CPU at the highest priority.
+ */
+ REG_WRITE(REG_GLOBAL2, 0x05, 0x00ff);
+
+ /* Program the DSA routing table. */
+ for (i = 0; i < 32; i++) {
+ int nexthop = 0x1f;
+
+ if (i != ds->index && i < ds->dst->pd->nr_chips)
+ nexthop = ds->pd->rtable[i] & 0x1f;
+
+ REG_WRITE(REG_GLOBAL2, 0x06, 0x8000 | (i << 8) | nexthop);
+ }
+
+ /* Clear all trunk masks. */
+ for (i = 0; i < 8; i++)
+ REG_WRITE(REG_GLOBAL2, 0x07, 0x8000 | (i << 12) | 0x7f);
+
+ /* Clear all trunk mappings. */
+ for (i = 0; i < 16; i++)
+ REG_WRITE(REG_GLOBAL2, 0x08, 0x8000 | (i << 11));
+
+ /* Disable ingress rate limiting by resetting all ingress
+ * rate limit registers to their initial state.
+ */
+ for (i = 0; i < 7; i++)
+ REG_WRITE(REG_GLOBAL2, 0x09, 0x9000 | (i << 8));
+
+ /* Initialise cross-chip port VLAN table to reset defaults. */
+ REG_WRITE(REG_GLOBAL2, 0x0b, 0x9000);
+
+ /* Clear the priority override table. */
+ for (i = 0; i < 16; i++)
+ REG_WRITE(REG_GLOBAL2, 0x0f, 0x8000 | (i << 8));
+
+ /* @@@ initialise AVB (22/23) watchdog (27) sdet (29) registers */
+
+ return 0;
+}
+
+static int mv88e6352_setup_port(struct dsa_switch *ds, int p)
+{
+ int addr = REG_PORT(p);
+ u16 val;
+
+ /* MAC Forcing register: don't force link, speed, duplex
+ * or flow control state to any particular values on physical
+ * ports, but force the CPU port and all DSA ports to 1000 Mb/s
+ * full duplex.
+ */
+ if (dsa_is_cpu_port(ds, p) || ds->dsa_port_mask & (1 << p))
+ REG_WRITE(addr, 0x01, 0x003e);
+ else
+ REG_WRITE(addr, 0x01, 0x0003);
+
+ /* Do not limit the period of time that this port can be
+ * paused for by the remote end or the period of time that
+ * this port can pause the remote end.
+ */
+ REG_WRITE(addr, 0x02, 0x0000);
+
+ /* Port Control: disable Drop-on-Unlock, disable Drop-on-Lock,
+ * disable Header mode, enable IGMP/MLD snooping, disable VLAN
+ * tunneling, determine priority by looking at 802.1p and IP
+ * priority fields (IP prio has precedence), and set STP state
+ * to Forwarding.
+ *
+ * If this is the CPU link, use DSA or EDSA tagging depending
+ * on which tagging mode was configured.
+ *
+ * If this is a link to another switch, use DSA tagging mode.
+ *
+ * If this is the upstream port for this switch, enable
+ * forwarding of unknown unicasts and multicasts.
+ */
+ val = 0x0433;
+ if (dsa_is_cpu_port(ds, p)) {
+ if (ds->dst->tag_protocol == DSA_TAG_PROTO_EDSA)
+ val |= 0x3300;
+ else
+ val |= 0x0100;
+ }
+ if (ds->dsa_port_mask & (1 << p))
+ val |= 0x0100;
+ if (p == dsa_upstream_port(ds))
+ val |= 0x000c;
+ REG_WRITE(addr, 0x04, val);
+
+ /* Port Control 1: disable trunking. Also, if this is the
+ * CPU port, enable learn messages to be sent to this port.
+ */
+ REG_WRITE(addr, 0x05, dsa_is_cpu_port(ds, p) ? 0x8000 : 0x0000);
+
+ /* Port based VLAN map: give each port its own address
+ * database, allow the CPU port to talk to each of the 'real'
+ * ports, and allow each of the 'real' ports to only talk to
+ * the upstream port.
+ */
+ val = (p & 0xf) << 12;
+ if (dsa_is_cpu_port(ds, p))
+ val |= ds->phys_port_mask;
+ else
+ val |= 1 << dsa_upstream_port(ds);
+ REG_WRITE(addr, 0x06, val);
+
+ /* Default VLAN ID and priority: don't set a default VLAN
+ * ID, and set the default packet priority to zero.
+ */
+ REG_WRITE(addr, 0x07, 0x0000);
+
+ /* Port Control 2: don't force a good FCS, set the maximum
+ * frame size to 10240 bytes, don't let the switch add or
+ * strip 802.1q tags, don't discard tagged or untagged frames
+ * on this port, do a destination address lookup on all
+ * received packets as usual, disable ARP mirroring and don't
+ * send a copy of all transmitted/received frames on this port
+ * to the CPU.
+ */
+ REG_WRITE(addr, 0x08, 0x2080);
+
+ /* Egress rate control: disable egress rate control. */
+ REG_WRITE(addr, 0x09, 0x0001);
+
+ /* Egress rate control 2: disable egress rate control. */
+ REG_WRITE(addr, 0x0a, 0x0000);
+
+ /* Port Association Vector: when learning source addresses
+ * of packets, add the address to the address database using
+ * a port bitmap that has only the bit for this port set and
+ * the other bits clear.
+ */
+ REG_WRITE(addr, 0x0b, 1 << p);
+
+ /* Port ATU control: disable limiting the number of address
+ * database entries that this port is allowed to use.
+ */
+ REG_WRITE(addr, 0x0c, 0x0000);
+
+ /* Priority Override: disable DA, SA and VTU priority override. */
+ REG_WRITE(addr, 0x0d, 0x0000);
+
+ /* Port Ethertype: use the Ethertype DSA Ethertype value. */
+ REG_WRITE(addr, 0x0f, ETH_P_EDSA);
+
+ /* Tag Remap: use an identity 802.1p prio -> switch prio
+ * mapping.
+ */
+ REG_WRITE(addr, 0x18, 0x3210);
+
+ /* Tag Remap 2: use an identity 802.1p prio -> switch prio
+ * mapping.
+ */
+ REG_WRITE(addr, 0x19, 0x7654);
+
+ return 0;
+}
+
+static int mv88e6352_setup(struct dsa_switch *ds)
+{
+ struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+ int ret;
+ int i;
+
+ mutex_init(&ps->smi_mutex);
+ mutex_init(&ps->stats_mutex);
+ mutex_init(&ps->phy_mutex);
+
+ ps->id = REG_READ(REG_PORT(0), 0x03) & 0xfff0;
+
+ ret = mv88e6352_switch_reset(ds);
+ if (ret < 0)
+ return ret;
+
+ /* @@@ initialise vtu and atu */
+
+ ret = mv88e6352_setup_global(ds);
+ if (ret < 0)
+ return ret;
+
+ for (i = 0; i < 7; i++) {
+ ret = mv88e6352_setup_port(ds, i);
+ if (ret < 0)
+ return ret;
+ }
+
+ return 0;
+}
+
+static int mv88e6352_port_to_phy_addr(int port)
+{
+ if (port >= 0 && port <= 4)
+ return port;
+ return -EINVAL;
+}
+
+static int
+mv88e6352_phy_read(struct dsa_switch *ds, int port, int regnum)
+{
+ struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+ int addr = mv88e6352_port_to_phy_addr(port);
+ int ret;
+
+ if (addr < 0)
+ return addr;
+
+ mutex_lock(&ps->phy_mutex);
+ ret = __mv88e6352_phy_read(ds, addr, regnum);
+ mutex_unlock(&ps->phy_mutex);
+
+ return ret;
+}
+
+static int
+mv88e6352_phy_write(struct dsa_switch *ds, int port, int regnum, u16 val)
+{
+ struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+ int addr = mv88e6352_port_to_phy_addr(port);
+ int ret;
+
+ if (addr < 0)
+ return addr;
+
+ mutex_lock(&ps->phy_mutex);
+ ret = __mv88e6352_phy_write(ds, addr, regnum, val);
+ mutex_unlock(&ps->phy_mutex);
+
+ return ret;
+}
+
+static struct mv88e6xxx_hw_stat mv88e6352_hw_stats[] = {
+ { "in_good_octets", 8, 0x00, },
+ { "in_bad_octets", 4, 0x02, },
+ { "in_unicast", 4, 0x04, },
+ { "in_broadcasts", 4, 0x06, },
+ { "in_multicasts", 4, 0x07, },
+ { "in_pause", 4, 0x16, },
+ { "in_undersize", 4, 0x18, },
+ { "in_fragments", 4, 0x19, },
+ { "in_oversize", 4, 0x1a, },
+ { "in_jabber", 4, 0x1b, },
+ { "in_rx_error", 4, 0x1c, },
+ { "in_fcs_error", 4, 0x1d, },
+ { "out_octets", 8, 0x0e, },
+ { "out_unicast", 4, 0x10, },
+ { "out_broadcasts", 4, 0x13, },
+ { "out_multicasts", 4, 0x12, },
+ { "out_pause", 4, 0x15, },
+ { "excessive", 4, 0x11, },
+ { "collisions", 4, 0x1e, },
+ { "deferred", 4, 0x05, },
+ { "single", 4, 0x14, },
+ { "multiple", 4, 0x17, },
+ { "out_fcs_error", 4, 0x03, },
+ { "late", 4, 0x1f, },
+ { "hist_64bytes", 4, 0x08, },
+ { "hist_65_127bytes", 4, 0x09, },
+ { "hist_128_255bytes", 4, 0x0a, },
+ { "hist_256_511bytes", 4, 0x0b, },
+ { "hist_512_1023bytes", 4, 0x0c, },
+ { "hist_1024_max_bytes", 4, 0x0d, },
+};
+
+static void
+mv88e6352_get_strings(struct dsa_switch *ds, int port, uint8_t *data)
+{
+ mv88e6xxx_get_strings(ds, ARRAY_SIZE(mv88e6352_hw_stats),
+ mv88e6352_hw_stats, port, data);
+}
+
+static void
+mv88e6352_get_ethtool_stats(struct dsa_switch *ds, int port, uint64_t *data)
+{
+ mv88e6xxx_get_ethtool_stats(ds, ARRAY_SIZE(mv88e6352_hw_stats),
+ mv88e6352_hw_stats, port, data);
+}
+
+static int mv88e6352_get_sset_count(struct dsa_switch *ds)
+{
+ return ARRAY_SIZE(mv88e6352_hw_stats);
+}
+
+struct dsa_switch_driver mv88e6352_switch_driver = {
+ .tag_protocol = DSA_TAG_PROTO_EDSA,
+ .priv_size = sizeof(struct mv88e6xxx_priv_state),
+ .probe = mv88e6352_probe,
+ .setup = mv88e6352_setup,
+ .set_addr = mv88e6xxx_set_addr_indirect,
+ .phy_read = mv88e6352_phy_read,
+ .phy_write = mv88e6352_phy_write,
+ .poll_link = mv88e6xxx_poll_link,
+ .get_strings = mv88e6352_get_strings,
+ .get_ethtool_stats = mv88e6352_get_ethtool_stats,
+ .get_sset_count = mv88e6352_get_sset_count,
+};
+
+MODULE_ALIAS("platform:mv88e6352");
diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index a6c90cf..8e1090b 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -507,6 +507,9 @@ static int __init mv88e6xxx_init(void)
#if IS_ENABLED(CONFIG_NET_DSA_MV88E6123_61_65)
register_switch_driver(&mv88e6123_61_65_switch_driver);
#endif
+#if IS_ENABLED(CONFIG_NET_DSA_MV88E6352)
+ register_switch_driver(&mv88e6352_switch_driver);
+#endif
#if IS_ENABLED(CONFIG_NET_DSA_MV88E6171)
register_switch_driver(&mv88e6171_switch_driver);
#endif
diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h
index 5e5145a..c0ce133 100644
--- a/drivers/net/dsa/mv88e6xxx.h
+++ b/drivers/net/dsa/mv88e6xxx.h
@@ -37,6 +37,12 @@ struct mv88e6xxx_priv_state {
*/
struct mutex stats_mutex;
+ /* This mutex serializes phy access for chips with
+ * indirect phy addressing. It is unused for chips
+ * with direct phy access.
+ */
+ struct mutex phy_mutex;
+
int id; /* switch product id */
};
@@ -70,6 +76,7 @@ void mv88e6xxx_get_ethtool_stats(struct dsa_switch *ds,
extern struct dsa_switch_driver mv88e6131_switch_driver;
extern struct dsa_switch_driver mv88e6123_61_65_switch_driver;
+extern struct dsa_switch_driver mv88e6352_switch_driver;
extern struct dsa_switch_driver mv88e6171_switch_driver;
#define REG_READ(addr, reg) \
--
1.9.1
^ permalink raw reply related
* [PATCH v3 06/15] net: dsa: Add support for reporting switch chip temperatures
From: Guenter Roeck @ 2014-10-29 17:44 UTC (permalink / raw)
To: netdev
Cc: David S. Miller, Florian Fainelli, Andrew Lunn, linux-kernel,
Guenter Roeck
In-Reply-To: <1414604707-22407-1-git-send-email-linux@roeck-us.net>
Some switches provide chip temperature data.
Add support for reporting it through the hwmon subsystem.
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
v3:
- No change
v2:
- Updated headline to reflect what is reported, not how.
- Make added functionality optional with new Kconfig flag
- Register with hwmon subsystem as virtual device (no parent)
- Generate hwmon 'name' attribute from sanitized master network device
name plus _dsaX, where X is the DSA switch index and 'sanitized'
means to pick all characters and digits from the network device name
- Do not use devm_ function to register hwmon device, since the hwmon
device is not associated with a device and thus can not be auto-removed.
include/net/dsa.h | 16 +++++++
net/dsa/Kconfig | 11 +++++
net/dsa/dsa.c | 131 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 158 insertions(+)
diff --git a/include/net/dsa.h b/include/net/dsa.h
index b765592..55e75e7 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -139,6 +139,14 @@ struct dsa_switch {
*/
struct device *master_dev;
+#ifdef CONFIG_NET_DSA_HWMON
+ /*
+ * Hardware monitoring information
+ */
+ char hwmon_name[IFNAMSIZ + 8];
+ struct device *hwmon_dev;
+#endif
+
/*
* Slave mii_bus and devices for the individual ports.
*/
@@ -242,6 +250,14 @@ struct dsa_switch_driver {
struct ethtool_eee *e);
int (*get_eee)(struct dsa_switch *ds, int port,
struct ethtool_eee *e);
+
+#ifdef CONFIG_NET_DSA_HWMON
+ /* Hardware monitoring */
+ int (*get_temp)(struct dsa_switch *ds, int *temp);
+ int (*get_temp_limit)(struct dsa_switch *ds, int *temp);
+ int (*set_temp_limit)(struct dsa_switch *ds, int temp);
+ int (*get_temp_alarm)(struct dsa_switch *ds, bool *alarm);
+#endif
};
void register_switch_driver(struct dsa_switch_driver *type);
diff --git a/net/dsa/Kconfig b/net/dsa/Kconfig
index a585fd6..5f8ac40 100644
--- a/net/dsa/Kconfig
+++ b/net/dsa/Kconfig
@@ -11,6 +11,17 @@ config NET_DSA
if NET_DSA
+config NET_DSA_HWMON
+ bool "Distributed Switch Architecture HWMON support"
+ default y
+ depends on HWMON && !(NET_DSA=y && HWMON=m)
+ ---help---
+ Say Y if you want to expose thermal sensor data on switches supported
+ by the Distributed Switch Architecture.
+
+ Some of those switches contain thermal sensors. This data is available
+ via the hwmon sysfs interface and exposes the onboard sensors.
+
# tagging formats
config NET_DSA_TAG_BRCM
bool
diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index 22f34cf..5edbbca 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -9,6 +9,9 @@
* (at your option) any later version.
*/
+#include <linux/ctype.h>
+#include <linux/device.h>
+#include <linux/hwmon.h>
#include <linux/list.h>
#include <linux/platform_device.h>
#include <linux/slab.h>
@@ -17,6 +20,7 @@
#include <linux/of.h>
#include <linux/of_mdio.h>
#include <linux/of_platform.h>
+#include <linux/sysfs.h>
#include "dsa_priv.h"
char dsa_driver_version[] = "0.1";
@@ -71,6 +75,104 @@ dsa_switch_probe(struct device *host_dev, int sw_addr, char **_name)
return ret;
}
+/* hwmon support ************************************************************/
+
+#ifdef CONFIG_NET_DSA_HWMON
+
+static ssize_t temp1_input_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct dsa_switch *ds = dev_get_drvdata(dev);
+ int temp, ret;
+
+ ret = ds->drv->get_temp(ds, &temp);
+ if (ret < 0)
+ return ret;
+
+ return sprintf(buf, "%d\n", temp * 1000);
+}
+static DEVICE_ATTR_RO(temp1_input);
+
+static ssize_t temp1_max_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct dsa_switch *ds = dev_get_drvdata(dev);
+ int temp, ret;
+
+ ret = ds->drv->get_temp_limit(ds, &temp);
+ if (ret < 0)
+ return ret;
+
+ return sprintf(buf, "%d\n", temp * 1000);
+}
+
+static ssize_t temp1_max_store(struct device *dev,
+ struct device_attribute *attr, const char *buf,
+ size_t count)
+{
+ struct dsa_switch *ds = dev_get_drvdata(dev);
+ int temp, ret;
+
+ ret = kstrtoint(buf, 0, &temp);
+ if (ret < 0)
+ return ret;
+
+ ret = ds->drv->set_temp_limit(ds, DIV_ROUND_CLOSEST(temp, 1000));
+ if (ret < 0)
+ return ret;
+
+ return count;
+}
+static DEVICE_ATTR(temp1_max, S_IRUGO, temp1_max_show, temp1_max_store);
+
+static ssize_t temp1_max_alarm_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct dsa_switch *ds = dev_get_drvdata(dev);
+ bool alarm;
+ int ret;
+
+ ret = ds->drv->get_temp_alarm(ds, &alarm);
+ if (ret < 0)
+ return ret;
+
+ return sprintf(buf, "%d\n", alarm);
+}
+static DEVICE_ATTR_RO(temp1_max_alarm);
+
+static struct attribute *dsa_hwmon_attrs[] = {
+ &dev_attr_temp1_input.attr, /* 0 */
+ &dev_attr_temp1_max.attr, /* 1 */
+ &dev_attr_temp1_max_alarm.attr, /* 2 */
+ NULL
+};
+
+static umode_t dsa_hwmon_attrs_visible(struct kobject *kobj,
+ struct attribute *attr, int index)
+{
+ struct device *dev = container_of(kobj, struct device, kobj);
+ struct dsa_switch *ds = dev_get_drvdata(dev);
+ struct dsa_switch_driver *drv = ds->drv;
+ umode_t mode = attr->mode;
+
+ if (index == 1) {
+ if (!drv->get_temp_limit)
+ mode = 0;
+ else if (drv->set_temp_limit)
+ mode |= S_IWUSR;
+ } else if (index == 2 && !drv->get_temp_alarm) {
+ mode = 0;
+ }
+ return mode;
+}
+
+static const struct attribute_group dsa_hwmon_group = {
+ .attrs = dsa_hwmon_attrs,
+ .is_visible = dsa_hwmon_attrs_visible,
+};
+__ATTRIBUTE_GROUPS(dsa_hwmon);
+
+#endif /* CONFIG_NET_DSA_HWMON */
/* basic switch operations **************************************************/
static struct dsa_switch *
@@ -225,6 +327,31 @@ dsa_switch_setup(struct dsa_switch_tree *dst, int index,
ds->ports[i] = slave_dev;
}
+#ifdef CONFIG_NET_DSA_HWMON
+ /* If the switch provides a temperature sensor,
+ * register with hardware monitoring subsystem.
+ * Treat registration error as non-fatal and ignore it.
+ */
+ if (drv->get_temp) {
+ const char *netname = netdev_name(dst->master_netdev);
+ char hname[IFNAMSIZ + 1];
+ int i, j;
+
+ /* Create valid hwmon 'name' attribute */
+ for (i = j = 0; i < IFNAMSIZ && netname[i]; i++) {
+ if (isalnum(netname[i]))
+ hname[j++] = netname[i];
+ }
+ hname[j] = '\0';
+ scnprintf(ds->hwmon_name, sizeof(ds->hwmon_name), "%s_dsa%d",
+ hname, index);
+ ds->hwmon_dev = hwmon_device_register_with_groups(NULL,
+ ds->hwmon_name, ds, dsa_hwmon_groups);
+ if (IS_ERR(ds->hwmon_dev))
+ ds->hwmon_dev = NULL;
+ }
+#endif /* CONFIG_NET_DSA_HWMON */
+
return ds;
out_free:
@@ -236,6 +363,10 @@ out:
static void dsa_switch_destroy(struct dsa_switch *ds)
{
+#ifdef CONFIG_NET_DSA_HWMON
+ if (ds->hwmon_dev)
+ hwmon_device_unregister(ds->hwmon_dev);
+#endif
}
#ifdef CONFIG_PM_SLEEP
--
1.9.1
^ permalink raw reply related
* [PATCH v3 07/15] net: dsa/mv88e6352: Report chip temperature
From: Guenter Roeck @ 2014-10-29 17:44 UTC (permalink / raw)
To: netdev
Cc: David S. Miller, Florian Fainelli, Andrew Lunn, linux-kernel,
Guenter Roeck
In-Reply-To: <1414604707-22407-1-git-send-email-linux@roeck-us.net>
MV88E6352 supports reading the chip temperature from two PHY registers,
6:26 and 6:27. Report it using the more accurate register 6:27.
Also report temperature limit and alarm.
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
v3:
- No change
v2:
- Make new functionality optional, depending on CONFIG_NET_DSA_HWMON
drivers/net/dsa/mv88e6352.c | 101 ++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 101 insertions(+)
diff --git a/drivers/net/dsa/mv88e6352.c b/drivers/net/dsa/mv88e6352.c
index f17364f..744e6fa 100644
--- a/drivers/net/dsa/mv88e6352.c
+++ b/drivers/net/dsa/mv88e6352.c
@@ -325,6 +325,101 @@ static int mv88e6352_setup_port(struct dsa_switch *ds, int p)
return 0;
}
+#ifdef CONFIG_NET_DSA_HWMON
+
+static int mv88e6352_phy_page_read(struct dsa_switch *ds,
+ int port, int page, int reg)
+{
+ struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+ int ret;
+
+ mutex_lock(&ps->phy_mutex);
+ ret = __mv88e6352_phy_write(ds, port, 0x16, page);
+ if (ret < 0)
+ goto error;
+ ret = __mv88e6352_phy_read(ds, port, reg);
+error:
+ __mv88e6352_phy_write(ds, port, 0x16, 0x0);
+ mutex_unlock(&ps->phy_mutex);
+ return ret;
+}
+
+static int mv88e6352_phy_page_write(struct dsa_switch *ds,
+ int port, int page, int reg, int val)
+{
+ struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+ int ret;
+
+ mutex_lock(&ps->phy_mutex);
+ ret = __mv88e6352_phy_write(ds, port, 0x16, page);
+ if (ret < 0)
+ goto error;
+
+ ret = __mv88e6352_phy_write(ds, port, reg, val);
+error:
+ __mv88e6352_phy_write(ds, port, 0x16, 0x0);
+ mutex_unlock(&ps->phy_mutex);
+ return ret;
+}
+
+static int mv88e6352_get_temp(struct dsa_switch *ds, int *temp)
+{
+ int ret;
+
+ *temp = 0;
+
+ ret = mv88e6352_phy_page_read(ds, 0, 6, 27);
+ if (ret < 0)
+ return ret;
+
+ *temp = (ret & 0xff) - 25;
+
+ return 0;
+}
+
+static int mv88e6352_get_temp_limit(struct dsa_switch *ds, int *temp)
+{
+ int ret;
+
+ *temp = 0;
+
+ ret = mv88e6352_phy_page_read(ds, 0, 6, 26);
+ if (ret < 0)
+ return ret;
+
+ *temp = (((ret >> 8) & 0x1f) * 5) - 25;
+
+ return 0;
+}
+
+static int mv88e6352_set_temp_limit(struct dsa_switch *ds, int temp)
+{
+ int ret;
+
+ ret = mv88e6352_phy_page_read(ds, 0, 6, 26);
+ if (ret < 0)
+ return ret;
+ temp = clamp_val(DIV_ROUND_CLOSEST(temp, 5) + 5, 0, 0x1f);
+ return mv88e6352_phy_page_write(ds, 0, 6, 26,
+ (ret & 0xe0ff) | (temp << 8));
+}
+
+static int mv88e6352_get_temp_alarm(struct dsa_switch *ds, bool *alarm)
+{
+ int ret;
+
+ *alarm = false;
+
+ ret = mv88e6352_phy_page_read(ds, 0, 6, 26);
+ if (ret < 0)
+ return ret;
+
+ *alarm = !!(ret & 0x40);
+
+ return 0;
+}
+#endif /* CONFIG_NET_DSA_HWMON */
+
static int mv88e6352_setup(struct dsa_switch *ds)
{
struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
@@ -461,6 +556,12 @@ struct dsa_switch_driver mv88e6352_switch_driver = {
.get_strings = mv88e6352_get_strings,
.get_ethtool_stats = mv88e6352_get_ethtool_stats,
.get_sset_count = mv88e6352_get_sset_count,
+#ifdef CONFIG_NET_DSA_HWMON
+ .get_temp = mv88e6352_get_temp,
+ .get_temp_limit = mv88e6352_get_temp_limit,
+ .set_temp_limit = mv88e6352_set_temp_limit,
+ .get_temp_alarm = mv88e6352_get_temp_alarm,
+#endif
};
MODULE_ALIAS("platform:mv88e6352");
--
1.9.1
^ permalink raw reply related
* [PATCH v3 09/15] net: dsa: Add support for switch EEPROM access
From: Guenter Roeck @ 2014-10-29 17:45 UTC (permalink / raw)
To: netdev
Cc: David S. Miller, Florian Fainelli, Andrew Lunn, linux-kernel,
Guenter Roeck
In-Reply-To: <1414604707-22407-1-git-send-email-linux@roeck-us.net>
On some chips it is possible to access the switch eeprom.
Add infrastructure support for it.
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
v3:
- Fix bug seen if devicetree is enabled; eeprom-length property is
attached to switch devicetree node, not to dsa node.
v2:
- Add support for configuring switch EEPROM size through platform data
or devicetree data
- Do not compare new pointers against NULL
- Check if get_eeprom pointer is set before calling it
include/net/dsa.h | 10 ++++++++++
net/dsa/dsa.c | 4 ++++
net/dsa/slave.c | 41 +++++++++++++++++++++++++++++++++++++++++
3 files changed, 55 insertions(+)
diff --git a/include/net/dsa.h b/include/net/dsa.h
index 55e75e7..37856a2 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -38,6 +38,9 @@ struct dsa_chip_data {
struct device *host_dev;
int sw_addr;
+ /* set to size of eeprom if supported by the switch */
+ int eeprom_len;
+
/* Device tree node pointer for this specific switch chip
* used during switch setup in case additional properties
* and resources needs to be used
@@ -258,6 +261,13 @@ struct dsa_switch_driver {
int (*set_temp_limit)(struct dsa_switch *ds, int temp);
int (*get_temp_alarm)(struct dsa_switch *ds, bool *alarm);
#endif
+
+ /* EEPROM access */
+ int (*get_eeprom_len)(struct dsa_switch *ds);
+ int (*get_eeprom)(struct dsa_switch *ds,
+ struct ethtool_eeprom *eeprom, u8 *data);
+ int (*set_eeprom)(struct dsa_switch *ds,
+ struct ethtool_eeprom *eeprom, u8 *data);
};
void register_switch_driver(struct dsa_switch_driver *type);
diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index 5edbbca..b51ef592 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -575,6 +575,7 @@ static int dsa_of_probe(struct platform_device *pdev)
const char *port_name;
int chip_index, port_index;
const unsigned int *sw_addr, *port_reg;
+ u32 eeprom_len;
int ret;
mdio = of_parse_phandle(np, "dsa,mii-bus", 0);
@@ -626,6 +627,9 @@ static int dsa_of_probe(struct platform_device *pdev)
if (cd->sw_addr > PHY_MAX_ADDR)
continue;
+ if (!of_property_read_u32(np, "eeprom-length", &eeprom_len))
+ cd->eeprom_len = eeprom_len;
+
for_each_available_child_of_node(child, port) {
port_reg = of_get_property(port, "reg", NULL);
if (!port_reg)
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 6d18174..ff2fbe7 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -271,6 +271,44 @@ static u32 dsa_slave_get_link(struct net_device *dev)
return -EOPNOTSUPP;
}
+static int dsa_slave_get_eeprom_len(struct net_device *dev)
+{
+ struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_switch *ds = p->parent;
+
+ if (ds->pd->eeprom_len)
+ return ds->pd->eeprom_len;
+
+ if (ds->drv->get_eeprom_len)
+ return ds->drv->get_eeprom_len(ds);
+
+ return 0;
+}
+
+static int dsa_slave_get_eeprom(struct net_device *dev,
+ struct ethtool_eeprom *eeprom, u8 *data)
+{
+ struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_switch *ds = p->parent;
+
+ if (ds->drv->get_eeprom)
+ return ds->drv->get_eeprom(ds, eeprom, data);
+
+ return -EOPNOTSUPP;
+}
+
+static int dsa_slave_set_eeprom(struct net_device *dev,
+ struct ethtool_eeprom *eeprom, u8 *data)
+{
+ struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_switch *ds = p->parent;
+
+ if (ds->drv->set_eeprom)
+ return ds->drv->set_eeprom(ds, eeprom, data);
+
+ return -EOPNOTSUPP;
+}
+
static void dsa_slave_get_strings(struct net_device *dev,
uint32_t stringset, uint8_t *data)
{
@@ -387,6 +425,9 @@ static const struct ethtool_ops dsa_slave_ethtool_ops = {
.get_drvinfo = dsa_slave_get_drvinfo,
.nway_reset = dsa_slave_nway_reset,
.get_link = dsa_slave_get_link,
+ .get_eeprom_len = dsa_slave_get_eeprom_len,
+ .get_eeprom = dsa_slave_get_eeprom,
+ .set_eeprom = dsa_slave_set_eeprom,
.get_strings = dsa_slave_get_strings,
.get_ethtool_stats = dsa_slave_get_ethtool_stats,
.get_sset_count = dsa_slave_get_sset_count,
--
1.9.1
^ permalink raw reply related
* [PATCH v3 11/15] net: dsa/mv88e6352: Implement EEPROM access functions
From: Guenter Roeck @ 2014-10-29 17:45 UTC (permalink / raw)
To: netdev
Cc: David S. Miller, Florian Fainelli, Andrew Lunn, linux-kernel,
Guenter Roeck
In-Reply-To: <1414604707-22407-1-git-send-email-linux@roeck-us.net>
MV88E6352 supports read and write access to its configuration eeprom.
There is no means to detect if an EEPROM is connected to the switch.
Also, the switch supports EEPROMs with different sizes, but can not detect
or report the type or size of connected EEPROMs. Therefore, do not implement
the get_eeprom_len callback but depend on platform or devicetree data to
provide information about EEPROM presence and size.
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
v3:
- No change
v2:
- EEPROM length must now be provided through platform or devicetree data
drivers/net/dsa/mv88e6352.c | 222 +++++++++++++++++++++++++++++++++++++++++++-
drivers/net/dsa/mv88e6xxx.h | 5 +
2 files changed, 224 insertions(+), 3 deletions(-)
diff --git a/drivers/net/dsa/mv88e6352.c b/drivers/net/dsa/mv88e6352.c
index 744e6fa..8a956f9 100644
--- a/drivers/net/dsa/mv88e6352.c
+++ b/drivers/net/dsa/mv88e6352.c
@@ -22,18 +22,18 @@
#include <net/dsa.h>
#include "mv88e6xxx.h"
-static int mv88e6352_phy_wait(struct dsa_switch *ds)
+static int mv88e6352_wait(struct dsa_switch *ds, int reg, u16 mask)
{
unsigned long timeout = jiffies + HZ / 10;
while (time_before(jiffies, timeout)) {
int ret;
- ret = REG_READ(REG_GLOBAL2, 0x18);
+ ret = REG_READ(REG_GLOBAL2, reg);
if (ret < 0)
return ret;
- if (!(ret & 0x8000))
+ if (!(ret & mask))
return 0;
usleep_range(1000, 2000);
@@ -41,6 +41,21 @@ static int mv88e6352_phy_wait(struct dsa_switch *ds)
return -ETIMEDOUT;
}
+static inline int mv88e6352_phy_wait(struct dsa_switch *ds)
+{
+ return mv88e6352_wait(ds, 0x18, 0x8000);
+}
+
+static inline int mv88e6352_eeprom_load_wait(struct dsa_switch *ds)
+{
+ return mv88e6352_wait(ds, 0x14, 0x0800);
+}
+
+static inline int mv88e6352_eeprom_busy_wait(struct dsa_switch *ds)
+{
+ return mv88e6352_wait(ds, 0x14, 0x8000);
+}
+
static int __mv88e6352_phy_read(struct dsa_switch *ds, int addr, int regnum)
{
int ret;
@@ -429,6 +444,7 @@ static int mv88e6352_setup(struct dsa_switch *ds)
mutex_init(&ps->smi_mutex);
mutex_init(&ps->stats_mutex);
mutex_init(&ps->phy_mutex);
+ mutex_init(&ps->eeprom_mutex);
ps->id = REG_READ(REG_PORT(0), 0x03) & 0xfff0;
@@ -525,6 +541,204 @@ static struct mv88e6xxx_hw_stat mv88e6352_hw_stats[] = {
{ "hist_1024_max_bytes", 4, 0x0d, },
};
+static int mv88e6352_read_eeprom_word(struct dsa_switch *ds, int addr)
+{
+ struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+ int ret;
+
+ mutex_lock(&ps->eeprom_mutex);
+
+ ret = mv88e6xxx_reg_write(ds, REG_GLOBAL2, 0x14,
+ 0xc000 | (addr & 0xff));
+ if (ret < 0)
+ goto error;
+
+ ret = mv88e6352_eeprom_busy_wait(ds);
+ if (ret < 0)
+ goto error;
+
+ ret = mv88e6xxx_reg_read(ds, REG_GLOBAL2, 0x15);
+error:
+ mutex_unlock(&ps->eeprom_mutex);
+ return ret;
+}
+
+static int mv88e6352_get_eeprom(struct dsa_switch *ds,
+ struct ethtool_eeprom *eeprom, u8 *data)
+{
+ int offset;
+ int len;
+ int ret;
+
+ offset = eeprom->offset;
+ len = eeprom->len;
+ eeprom->len = 0;
+
+ eeprom->magic = 0xc3ec4951;
+
+ ret = mv88e6352_eeprom_load_wait(ds);
+ if (ret < 0)
+ return ret;
+
+ if (offset & 1) {
+ int word;
+
+ word = mv88e6352_read_eeprom_word(ds, offset >> 1);
+ if (word < 0)
+ return word;
+
+ *data++ = (word >> 8) & 0xff;
+
+ offset++;
+ len--;
+ eeprom->len++;
+ }
+
+ while (len >= 2) {
+ int word;
+
+ word = mv88e6352_read_eeprom_word(ds, offset >> 1);
+ if (word < 0)
+ return word;
+
+ *data++ = word & 0xff;
+ *data++ = (word >> 8) & 0xff;
+
+ offset += 2;
+ len -= 2;
+ eeprom->len += 2;
+ }
+
+ if (len) {
+ int word;
+
+ word = mv88e6352_read_eeprom_word(ds, offset >> 1);
+ if (word < 0)
+ return word;
+
+ *data++ = word & 0xff;
+
+ offset++;
+ len--;
+ eeprom->len++;
+ }
+
+ return 0;
+}
+
+static int mv88e6352_eeprom_is_readonly(struct dsa_switch *ds)
+{
+ int ret;
+
+ ret = mv88e6xxx_reg_read(ds, REG_GLOBAL2, 0x14);
+ if (ret < 0)
+ return ret;
+
+ if (!(ret & 0x0400))
+ return -EROFS;
+
+ return 0;
+}
+
+static int mv88e6352_write_eeprom_word(struct dsa_switch *ds, int addr,
+ u16 data)
+{
+ struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+ int ret;
+
+ mutex_lock(&ps->eeprom_mutex);
+
+ ret = mv88e6xxx_reg_write(ds, REG_GLOBAL2, 0x15, data);
+ if (ret < 0)
+ goto error;
+
+ ret = mv88e6xxx_reg_write(ds, REG_GLOBAL2, 0x14,
+ 0xb000 | (addr & 0xff));
+ if (ret < 0)
+ goto error;
+
+ ret = mv88e6352_eeprom_busy_wait(ds);
+error:
+ mutex_unlock(&ps->eeprom_mutex);
+ return ret;
+}
+
+static int mv88e6352_set_eeprom(struct dsa_switch *ds,
+ struct ethtool_eeprom *eeprom, u8 *data)
+{
+ int offset;
+ int ret;
+ int len;
+
+ if (eeprom->magic != 0xc3ec4951)
+ return -EINVAL;
+
+ ret = mv88e6352_eeprom_is_readonly(ds);
+ if (ret)
+ return ret;
+
+ offset = eeprom->offset;
+ len = eeprom->len;
+ eeprom->len = 0;
+
+ ret = mv88e6352_eeprom_load_wait(ds);
+ if (ret < 0)
+ return ret;
+
+ if (offset & 1) {
+ int word;
+
+ word = mv88e6352_read_eeprom_word(ds, offset >> 1);
+ if (word < 0)
+ return word;
+
+ word = (*data++ << 8) | (word & 0xff);
+
+ ret = mv88e6352_write_eeprom_word(ds, offset >> 1, word);
+ if (ret < 0)
+ return ret;
+
+ offset++;
+ len--;
+ eeprom->len++;
+ }
+
+ while (len >= 2) {
+ int word;
+
+ word = *data++;
+ word |= *data++ << 8;
+
+ ret = mv88e6352_write_eeprom_word(ds, offset >> 1, word);
+ if (ret < 0)
+ return ret;
+
+ offset += 2;
+ len -= 2;
+ eeprom->len += 2;
+ }
+
+ if (len) {
+ int word;
+
+ word = mv88e6352_read_eeprom_word(ds, offset >> 1);
+ if (word < 0)
+ return word;
+
+ word = (word & 0xff00) | *data++;
+
+ ret = mv88e6352_write_eeprom_word(ds, offset >> 1, word);
+ if (ret < 0)
+ return ret;
+
+ offset++;
+ len--;
+ eeprom->len++;
+ }
+
+ return 0;
+}
+
static void
mv88e6352_get_strings(struct dsa_switch *ds, int port, uint8_t *data)
{
@@ -562,6 +776,8 @@ struct dsa_switch_driver mv88e6352_switch_driver = {
.set_temp_limit = mv88e6352_set_temp_limit,
.get_temp_alarm = mv88e6352_get_temp_alarm,
#endif
+ .get_eeprom = mv88e6352_get_eeprom,
+ .set_eeprom = mv88e6352_set_eeprom,
};
MODULE_ALIAS("platform:mv88e6352");
diff --git a/drivers/net/dsa/mv88e6xxx.h b/drivers/net/dsa/mv88e6xxx.h
index c0ce133..29feed0 100644
--- a/drivers/net/dsa/mv88e6xxx.h
+++ b/drivers/net/dsa/mv88e6xxx.h
@@ -43,6 +43,11 @@ struct mv88e6xxx_priv_state {
*/
struct mutex phy_mutex;
+ /* This mutex serializes eeprom access for chips with
+ * eeprom support.
+ */
+ struct mutex eeprom_mutex;
+
int id; /* switch product id */
};
--
1.9.1
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox