* Re: [PATCH net-next 2/2] l2tp: add peer_offset parameter
From: Guillaume Nault @ 2018-01-02 17:50 UTC (permalink / raw)
To: James Chapman; +Cc: Lorenzo Bianconi, davem, netdev, liuhangbin
In-Reply-To: <27e0a7c8-b95b-04eb-d808-bcae67e417b1@katalix.com>
On Fri, Dec 29, 2017 at 06:53:56PM +0000, James Chapman wrote:
> On 28/12/17 19:45, Guillaume Nault wrote:
> > Here we have an option that:
> > * creates invalid packets (AFAIK),
> > * is buggy and leaks memory on the network,
> > * doesn't seem to have any use case (even the manpage
> > says "This is hardly ever used").
> >
> > So I'm sorry, but I don't see the point in expanding this option to
> > allow even stranger setups. If there's a use case, then fine.
> > Otherwise, let's just acknowledge that the "peer_offset" option of
> > iproute2 is a noop (and maybe remove it from the manpage).
> >
> > And the kernel "offset" option needs to be fixed. Actually, I wouldn't
> > mind if it was converted to be a noop, or even rejected. L2TP already
> > has its share of unused features that complicate the code and hamper
> > evolution and bug fixing. As I said earlier, if it's buggy, unused and
> > can't even produce valid packets, then why bothering with it?
> >
> > But that's just my point of view. James, do you have an opinion on
> > this?
>
> I agree, Guillaume.
>
> The L2TPv3 protocol RFC dropped the configurable offset of L2TPv2 - instead,
> the Layer-2-Specific-Sublayer is supposed to handle any transport-specific
> data alignment requirements.
>
Yes, and AFAIK, no RFC has ever defined an L2TPv3 sublayer using offsets.
> I think a configurable offset has found its way
> into iproute2 l2tp commands by mistake, perhaps because the netlink API
> defines an attribute for it, but which was only intended for use with
> L2TPv2.
>
Makes sense, however L2TP_ATTR_OFFSET seems to be a noop for L2TPv2 in
the current implementation.
> For L2TPv2, we only configure the offset for transmitted packets. In
> received packets, the offset (if present) is obtained from the L2TPv2 header
> in each received packet. There is no need to add a peer-offset netlink
> attribute to set the offset expected in received packets.
>
Agreed for Rx side. I also agree on the theory for Tx, but in the current
implementation, l2tp_build_l2tpv2_header() doesn't take the session's
"offset" field into account. So, unless I've missed something,
L2TP_ATTR_OFFSET is already a noop for L2TPv2.
Not sure if it's worth handling this feature of L2TPv2. The Linux
implementation has been there for so long, and nobody ever complained
that there was no way to define an offset on Tx.
^ permalink raw reply
* Re: [bpf-next V3 PATCH 13/14] bpf: finally expose xdp_rxq_info to XDP bpf-programs
From: David Ahern @ 2018-01-02 17:47 UTC (permalink / raw)
To: Jesper Dangaard Brouer, Daniel Borkmann, Alexei Starovoitov
Cc: netdev, gospo, bjorn.topel, michael.chan
In-Reply-To: <151471812024.30703.13086182080575474605.stgit@firesoul>
On 12/31/17 4:02 AM, Jesper Dangaard Brouer wrote:
> Now all XDP driver have been updated to setup xdp_rxq_info and assign
> this to xdp_buff->rxq. Thus, it is now safe to enable access to some
> of the xdp_rxq_info struct members.
>
> This patch extend xdp_md and expose UAPI to userspace for
> ingress_ifindex and rx_queue_index. Access happens via bpf
> instruction rewrite, that load data directly from struct xdp_rxq_info.
>
> * ingress_ifindex map to xdp_rxq_info->dev->ifindex
> * rx_queue_index map to xdp_rxq_info->queue_index
>
> Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
> ---
> include/uapi/linux/bpf.h | 3 +++
> net/core/filter.c | 19 +++++++++++++++++++
> 2 files changed, 22 insertions(+)
>
> diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
> index 69eabfcb9bdb..a6000a95d40e 100644
> --- a/include/uapi/linux/bpf.h
> +++ b/include/uapi/linux/bpf.h
> @@ -899,6 +899,9 @@ struct xdp_md {
> __u32 data;
> __u32 data_end;
> __u32 data_meta;
> + /* Below access go though struct xdp_rxq_info */
> + __u32 ingress_ifindex; /* rxq->dev->ifindex */
> + __u32 rx_queue_index; /* rxq->queue_index */
> };
>
> enum sk_action {
At some point we need to tackle the network namespace context for the
ifindex.
^ permalink raw reply
* Re: [RFC PATCH net-next 13/19] ipv6: Flush all sibling routes upon NETDEV_UNREGISTER
From: David Ahern @ 2018-01-02 17:42 UTC (permalink / raw)
To: Ido Schimmel, netdev; +Cc: davem, roopa, nicolas.dichtel, mlxsw
In-Reply-To: <20171231161513.25785-14-idosch@mellanox.com>
On 12/31/17 9:15 AM, Ido Schimmel wrote:
> IPv4 and IPv6 react differently to a netdev being unregistered. In IPv4,
> in case the netdev is used as a nexthop device in a multipath route, the
> entire route is flushed.
>
> However, IPv6 only removes the nexthops associated with the unregistered
> netdev.
>
> Align IPv4 and IPv6 and flush all the sibling routes when a nexthop
> device is unregistered.
>
> Signed-off-by: Ido Schimmel <idosch@mellanox.com>
> ---
> net/ipv6/route.c | 44 +++++++++++++++++++++++++++++++++++++++-----
> 1 file changed, 39 insertions(+), 5 deletions(-)
not so sure about this one.
When we get to nexthops as separate objects, we can bring in consistency
by allowing ipv4 routes to just drop a single nexthop in the route
versus the behavior today.
^ permalink raw reply
* [PATCH iproute2] man: fix small formatting errors
From: Luca Boccassi @ 2018-01-02 17:42 UTC (permalink / raw)
To: netdev
Lintian detected the following formatting errors:
man/man8/devlink-sb.8.gz 230: warning: macro `b' not defined
man/man8/ip-link.8.gz 1243: warning: macro `in-8' not defined
(possibly missing space after `in')
man/man8/tc-u32.8.gz `R' is a string (producing the registered sign),
not a macro.
Signed-off-by: Luca Boccassi <bluca@debian.org>
---
man/man8/devlink-sb.8 | 2 +-
man/man8/ip-link.8.in | 2 +-
man/man8/tc-u32.8 | 6 ++----
3 files changed, 4 insertions(+), 6 deletions(-)
diff --git a/man/man8/devlink-sb.8 b/man/man8/devlink-sb.8
index ffb5553e..1882833a 100644
--- a/man/man8/devlink-sb.8
+++ b/man/man8/devlink-sb.8
@@ -227,7 +227,7 @@ This command is used to browse shared buffer occupancy values. Values are showed
current_value/max_value
.in -16
Note that before showing values, one has to issue
-.b occupancy snapshot
+.B occupancy snapshot
command first.
.PP
diff --git a/man/man8/ip-link.8.in b/man/man8/ip-link.8.in
index 9ddf0ffb..ff6bb9a2 100644
--- a/man/man8/ip-link.8.in
+++ b/man/man8/ip-link.8.in
@@ -1275,7 +1275,7 @@ or disable
arptables hooks on the bridge.
-.in-8
+.in -8
.TP
MACsec Type Support
diff --git a/man/man8/tc-u32.8 b/man/man8/tc-u32.8
index e9475a93..2bf2e3e9 100644
--- a/man/man8/tc-u32.8
+++ b/man/man8/tc-u32.8
@@ -29,10 +29,8 @@ u32 \- universal 32bit traffic control filter
.IR HANDLE " ] [ "
.B indev
.IR ifname " ] [ "
-.B skip_hw
-.R "|"
-.B skip_sw
-.R " ] [ "
+.BR skip_hw " | "
+.BR skip_sw " ] [ "
.BR help " ]"
.ti -8
--
2.14.2
^ permalink raw reply related
* Re: [patch net] mlxsw: spectrum_router: Fix NULL pointer deref
From: David Miller @ 2018-01-02 17:39 UTC (permalink / raw)
To: jiri; +Cc: netdev, idosch, mlxsw
In-Reply-To: <20171225080254.GA1885@nanopsycho>
From: Jiri Pirko <jiri@resnulli.us>
Date: Mon, 25 Dec 2017 09:02:54 +0100
> Mon, Dec 25, 2017 at 08:57:35AM CET, jiri@resnulli.us wrote:
>>From: Ido Schimmel <idosch@mellanox.com>
>>
>>When we remove the neighbour associated with a nexthop we should always
>>refuse to write the nexthop to the adjacency table. Regardless if it is
>>already present in the table or not.
>>
>>Otherwise, we risk dereferencing the NULL pointer that was set instead
>>of the neighbour.
>>
>>Fixes: a7ff87acd995 ("mlxsw: spectrum_router: Implement next-hop routing")
>>Signed-off-by: Ido Schimmel <idosch@mellanox.com>
>>Reported-by: Alexander Petrovskiy <alexpe@mellanox.com>
>>Signed-off-by: Jiri Pirko <jiri@mellanox.com>
>
> Dave, could you please queue this up for 4.14.y together
> with "mlxsw: spectrum: Relax sanity checks during enslavement".
Both applied and queued up for -stable.
^ permalink raw reply
* Re: [RFC PATCH net-next 16/19] ipv6: Flush multipath routes when all siblings are dead
From: David Ahern @ 2018-01-02 17:38 UTC (permalink / raw)
To: Ido Schimmel, netdev; +Cc: davem, roopa, nicolas.dichtel, mlxsw
In-Reply-To: <20171231161513.25785-17-idosch@mellanox.com>
On 12/31/17 9:15 AM, Ido Schimmel wrote:
> By default, IPv6 deletes nexthops from a multipath route when the
> nexthop device is put administratively down. This differs from IPv4
> where the nexthops are kept, but marked with the RTNH_F_DEAD flag. A
> multipath route is flushed when all of its nexthops become dead.
>
> Align IPv6 with IPv4 and have it conform to the same guidelines.
>
> In case the multipath route needs to be flushed, its siblings are
> flushed one by one. Otherwise, the nexthops are marked with the
> appropriate flags and the tree walker is instructed to skip all the
> siblings.
>
> As explained in previous patches, care is taken to update the sernum of
> the affected tree nodes, so as to prevent the use of wrong dst entries.
>
> Signed-off-by: Ido Schimmel <idosch@mellanox.com>
> ---
> net/ipv6/route.c | 53 ++++++++++++++++++++++++++++++++++++++++++++++-------
> 1 file changed, 46 insertions(+), 7 deletions(-)
>
> diff --git a/net/ipv6/route.c b/net/ipv6/route.c
> index ab95e07b5e74..9b142eb83f3c 100644
> --- a/net/ipv6/route.c
> +++ b/net/ipv6/route.c
> @@ -3486,8 +3486,10 @@ static int fib6_ifup(struct rt6_info *rt, void *p_arg)
> const struct arg_netdev_event *arg = p_arg;
> const struct net *net = dev_net(arg->dev);
>
> - if (rt != net->ipv6.ip6_null_entry && rt->dst.dev == arg->dev)
> + if (rt != net->ipv6.ip6_null_entry && rt->dst.dev == arg->dev) {
> rt->rt6i_nh_flags &= ~arg->nh_flags;
> + fib6_update_sernum_upto_root(dev_net(rt->dst.dev), rt);
> + }
>
> return 0;
> }
> @@ -3528,6 +3530,35 @@ static void rt6_multipath_flush(struct rt6_info *rt)
> iter->should_flush = 1;
> }
>
> +static unsigned int rt6_multipath_dead_count(const struct rt6_info *rt,
> + const struct net_device *down_dev)
> +{
> + struct rt6_info *iter;
> + unsigned int dead = 0;
> +
> + if (rt->dst.dev == down_dev || rt->rt6i_nh_flags & RTNH_F_DEAD)
> + dead++;
> + list_for_each_entry(iter, &rt->rt6i_siblings, rt6i_siblings)
> + if (iter->dst.dev == down_dev ||
> + iter->rt6i_nh_flags & RTNH_F_DEAD)
> + dead++;
> +
> + return dead;
> +}
> +
> +static void rt6_multipath_nh_flags_set(struct rt6_info *rt,
> + const struct net_device *dev,
> + unsigned int nh_flags)
> +{
> + struct rt6_info *iter;
> +
> + if (rt->dst.dev == dev)
> + rt->rt6i_nh_flags |= nh_flags;
> + list_for_each_entry(iter, &rt->rt6i_siblings, rt6i_siblings)
> + if (iter->dst.dev == dev)
> + iter->rt6i_nh_flags |= nh_flags;
> +}
> +
> /* called with write lock held for table with rt */
> static int fib6_ifdown(struct rt6_info *rt, void *p_arg)
> {
> @@ -3550,13 +3581,21 @@ static int fib6_ifdown(struct rt6_info *rt, void *p_arg)
> }
> return -2;
> case NETDEV_DOWN:
> - if (rt->dst.dev != dev)
> - break;
> - if (rt->rt6i_nsiblings == 0 ||
> - !rt->rt6i_idev->cnf.ignore_routes_with_linkdown)
> + if (rt->should_flush)
> return -1;
> - rt->rt6i_nh_flags |= (RTNH_F_DEAD | RTNH_F_LINKDOWN);
> - break;
> + if (!rt->rt6i_nsiblings)
> + return rt->dst.dev == dev ? -1 : 0;
> + if (rt6_multipath_uses_dev(rt, dev)) {
> + if (rt->rt6i_nsiblings + 1 ==
> + rt6_multipath_dead_count(rt, dev)) {
I'd prefer a tmp variable to make that line more readable and still
within the 80 column guideline.
unsigned int count;
count = rt6_multipath_dead_count(rt, dev);
if (rt->rt6i_nsiblings + 1 == count) {
> + rt6_multipath_flush(rt);
> + return -1;
> + }
> + rt6_multipath_nh_flags_set(rt, dev, RTNH_F_DEAD |
> + RTNH_F_LINKDOWN);
> + fib6_update_sernum(rt);
> + }
> + return -2;
> case NETDEV_CHANGE:
> if (rt->dst.dev != dev ||
> rt->rt6i_flags & (RTF_LOCAL | RTF_ANYCAST))
>
^ permalink raw reply
* Re: [PATCH net] ip6_tunnel: allow ip6gre dev mtu to be set below 1280
From: David Miller @ 2018-01-02 17:36 UTC (permalink / raw)
To: lucien.xin; +Cc: netdev
In-Reply-To: <2593c4a11b5ea0cf4efbec5517b7adec72ab4c17.1514184312.git.lucien.xin@gmail.com>
From: Xin Long <lucien.xin@gmail.com>
Date: Mon, 25 Dec 2017 14:45:12 +0800
> Commit 582442d6d5bc ("ipv6: Allow the MTU of ipip6 tunnel to be set
> below 1280") fixed a mtu setting issue. It works for ipip6 tunnel.
>
> But ip6gre dev updates the mtu also with ip6_tnl_change_mtu. Since
> the inner packet over ip6gre can be ipv4 and it's mtu should also
> be allowed to set below 1280, the same issue also exists on ip6gre.
>
> This patch is to fix it by simply changing to check if parms.proto
> is IPPROTO_IPV6 in ip6_tnl_change_mtu instead, to make ip6gre to
> go to 'else' branch.
>
> Signed-off-by: Xin Long <lucien.xin@gmail.com>
Applied.
^ permalink raw reply
* Re: [PATCH net] geneve: update skb dst pmtu on tx path
From: David Miller @ 2018-01-02 17:36 UTC (permalink / raw)
To: lucien.xin; +Cc: netdev
In-Reply-To: <fe176067f43de7928afcbc9810b13acb26f7adf0.1514184238.git.lucien.xin@gmail.com>
From: Xin Long <lucien.xin@gmail.com>
Date: Mon, 25 Dec 2017 14:43:58 +0800
> Commit a93bf0ff4490 ("vxlan: update skb dst pmtu on tx path") has fixed
> a performance issue caused by the change of lower dev's mtu for vxlan.
>
> The same thing needs to be done for geneve as well.
>
> Note that geneve cannot adjust it's mtu according to lower dev's mtu
> when creating it. The performance is very low later when netperfing
> over it without fixing the mtu manually. This patch could also avoid
> this issue.
>
> Signed-off-by: Xin Long <lucien.xin@gmail.com>
Applied.
^ permalink raw reply
* Re: [PATCH] ip6_tunnel: disable dst caching if tunnel is dual-stack
From: David Miller @ 2018-01-02 17:33 UTC (permalink / raw)
To: elicooper; +Cc: netdev
In-Reply-To: <20171225024349.4879-1-elicooper@gmx.com>
From: Eli Cooper <elicooper@gmx.com>
Date: Mon, 25 Dec 2017 10:43:49 +0800
> When an ip6_tunnel is in mode 'any', where the transport layer
> protocol can be either 4 or 41, dst_cache must be disabled.
>
> This is because xfrm policies might apply to only one of the two
> protocols. Caching dst would cause xfrm policies for one protocol
> incorrectly used for the other.
>
> Cc: stable@vger.kernel.org
> Signed-off-by: Eli Cooper <elicooper@gmx.com>
Please do not CC: stable on networking changes, as per the netdev
FAQ I queue these up by hand.
> - } else if (!(t->parms.flags &
> + } else if (t->parms.proto != 0 && !(t->parms.flags &
> (IP6_TNL_F_USE_ORIG_TCLASS | IP6_TNL_F_USE_ORIG_FWMARK))) {
When you adjust the indentation of an inner-expression, you must reindent the
subsequent lines that are also part of that inner-expression.
I've fixed up both of these issues and applied your patch, but please
take care of this yourself next time.
Thanks.
^ permalink raw reply
* Re: [PATCH] sky2: Replace mdelay with msleep in sky2_vpd_wait
From: David Miller @ 2018-01-02 17:27 UTC (permalink / raw)
To: baijiaju1990; +Cc: mlindner, stephen, shemminger, netdev, linux-kernel
In-Reply-To: <1514632187-14849-1-git-send-email-baijiaju1990@gmail.com>
From: Jia-Ju Bai <baijiaju1990@gmail.com>
Date: Sat, 30 Dec 2017 19:09:47 +0800
> sky2_vpd_wait is not called in an interrupt handler nor holding a spinlock.
> The function mdelay in it can be replaced with msleep, to reduce busy wait.
>
> Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Applied.
^ permalink raw reply
* Re: [PATCH v2] net: sh-eth: Add flag to determine the type of TSU register
From: Sergei Shtylyov @ 2018-01-02 17:27 UTC (permalink / raw)
To: Nobuhiro Iwamatsu, netdev; +Cc: yoshihiro.shimoda.uh
In-Reply-To: <1367903291-32323-1-git-send-email-nobuhiro.iwamatsu.yj@renesas.com>
Hello!
On 05/07/2013 09:08 AM, Nobuhiro Iwamatsu wrote:
> Some sh-eth devices may have two ether devices inside.
> And the function of TSU is accessed from each ether device.
> In this case, sh-eth need to remap address using devm_ioremap(),
> without using devm_ioremap_resource().
> tsu_shared_reg of sh_eth_cpu_data is used for this control.
>
> Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
> ---
> V2:
> - Change variable name from tsu_multi_reg to tsu_shared_reg.
> - Update commit message.
>
> drivers/net/ethernet/renesas/sh_eth.c | 11 ++++++++++-
> drivers/net/ethernet/renesas/sh_eth.h | 1 +
> 2 files changed, 11 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
> index 33dc6f2..489be0e 100644
> --- a/drivers/net/ethernet/renesas/sh_eth.c
> +++ b/drivers/net/ethernet/renesas/sh_eth.c
[...]
> @@ -2750,7 +2754,12 @@ static int sh_eth_drv_probe(struct platform_device *pdev)
> ret = -ENODEV;
> goto out_release;
> }
> - mdp->tsu_addr = devm_ioremap_resource(&pdev->dev, rtsu);
> +
> + if (mdp->cd->tsu_shared_reg)
> + mdp->tsu_addr = devm_ioremap(&pdev->dev, rtsu->start,
> + resource_size(rtsu));
I now think there's no need for a special flag -- we can just use 'devno'
(at least for the pure platform devices). The DT probing case remains unsolved
though (might make sense to look at the resource #0, whether it has 0x800 set
or not)...
> + else
> + mdp->tsu_addr = devm_ioremap_resource(&pdev->dev, rtsu);
> if (IS_ERR(mdp->tsu_addr)) {
> ret = PTR_ERR(mdp->tsu_addr);
> goto out_release;
MBR, Sergei
^ permalink raw reply
* Re: [PATCH 00/10] Convert mvneta to phylink
From: Russell King - ARM Linux @ 2018-01-02 17:27 UTC (permalink / raw)
To: Andrew Lunn, Florian Fainelli, Thomas Petazzoni; +Cc: netdev
In-Reply-To: <20180102172242.GE28752@n2100.armlinux.org.uk>
On Tue, Jan 02, 2018 at 05:22:42PM +0000, Russell King - ARM Linux wrote:
> Hi,
>
> This series converts mvneta to use phylink, which is necessary to
> support the SFP cages on SolidRun's Clearfog platform. This series just
> converts mvneta without adding the DT parts - having discussed with
> Andrew, we believe we're too close to the merge window to submit that
> patch.
>
> I've split the "net: mvneta: convert to phylink" patch up to make it
> easier to review, and in doing so, spotted some minor corner cases that
> needed to be fixed along the way.
>
> This series depends on the previously merged phylink patches in netdev,
> along with the recently reviewed 7 patch series "Resolve races in phy
> accessors" without which, the race described in patch 5 of that series
> is very evident when triggering a dummy hibernate cycle.
>
> This series also illustrates how to convert mvpp2 to phylink.
>
> mvneta is the only user of the fixed_phy_update_state() API, and this
> becomes redundant with the conversion.
>
> It would be good to get this series not only reviewed, but also
> independently tested to ensure that I haven't missed anything - I only
> have the Clearfog platform to test on, and that doesn't support all the
> different interface modes that mvneta supports.
>
> A particularly interesting side effect of this series is that DSA
> switches no longer need the "CPU" port and DSA facing MAC ethernet
> instance to be marked as a fixed link anymore with mvneta - we can use
> 1000BaseX mode, and the DSA to CPU link will use the 802.3z negotiation
> to determine the link properties without needing the link parameters to
> be explicitly stated in DT - that is a subject of a future patch.
>
> drivers/net/ethernet/marvell/Kconfig | 2 +-
> drivers/net/ethernet/marvell/mvneta.c | 687 ++++++++++++++++++++--------------
> drivers/net/phy/fixed_phy.c | 31 --
> include/linux/phy_fixed.h | 9 -
> 4 files changed, 405 insertions(+), 324 deletions(-)
Of course, this series is for net-next, not net... forgot to add that
annotation, sorry.
--
RMK's Patch system: http://www.armlinux.org.uk/developer/patches/
FTTC broadband for 0.8mile line in suburbia: sync at 8.8Mbps down 630kbps up
According to speedtest.net: 8.21Mbps down 510kbps up
^ permalink raw reply
* [PATCH 10/10] net: phy: fixed-phy: remove fixed_phy_update_state()
From: Russell King @ 2018-01-02 17:25 UTC (permalink / raw)
To: Andrew Lunn, Florian Fainelli, Thomas Petazzoni; +Cc: netdev
In-Reply-To: <20180102172242.GE28752@n2100.armlinux.org.uk>
mvneta is the only user of fixed_phy_update_state(), which has been
converted to use phylink instead. Remove fixed_phy_update_state().
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
drivers/net/phy/fixed_phy.c | 31 -------------------------------
include/linux/phy_fixed.h | 9 ---------
2 files changed, 40 deletions(-)
diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c
index eb5167210681..001fe1df7557 100644
--- a/drivers/net/phy/fixed_phy.c
+++ b/drivers/net/phy/fixed_phy.c
@@ -115,37 +115,6 @@ int fixed_phy_set_link_update(struct phy_device *phydev,
}
EXPORT_SYMBOL_GPL(fixed_phy_set_link_update);
-int fixed_phy_update_state(struct phy_device *phydev,
- const struct fixed_phy_status *status,
- const struct fixed_phy_status *changed)
-{
- struct fixed_mdio_bus *fmb = &platform_fmb;
- struct fixed_phy *fp;
-
- if (!phydev || phydev->mdio.bus != fmb->mii_bus)
- return -EINVAL;
-
- list_for_each_entry(fp, &fmb->phys, node) {
- if (fp->addr == phydev->mdio.addr) {
- write_seqcount_begin(&fp->seqcount);
-#define _UPD(x) if (changed->x) \
- fp->status.x = status->x
- _UPD(link);
- _UPD(speed);
- _UPD(duplex);
- _UPD(pause);
- _UPD(asym_pause);
-#undef _UPD
- fixed_phy_update(fp);
- write_seqcount_end(&fp->seqcount);
- return 0;
- }
- }
-
- return -ENOENT;
-}
-EXPORT_SYMBOL(fixed_phy_update_state);
-
int fixed_phy_add(unsigned int irq, int phy_addr,
struct fixed_phy_status *status,
int link_gpio)
diff --git a/include/linux/phy_fixed.h b/include/linux/phy_fixed.h
index cf6392de6eb0..ee54453a40a0 100644
--- a/include/linux/phy_fixed.h
+++ b/include/linux/phy_fixed.h
@@ -24,9 +24,6 @@ extern void fixed_phy_unregister(struct phy_device *phydev);
extern int fixed_phy_set_link_update(struct phy_device *phydev,
int (*link_update)(struct net_device *,
struct fixed_phy_status *));
-extern int fixed_phy_update_state(struct phy_device *phydev,
- const struct fixed_phy_status *status,
- const struct fixed_phy_status *changed);
#else
static inline int fixed_phy_add(unsigned int irq, int phy_id,
struct fixed_phy_status *status,
@@ -50,12 +47,6 @@ static inline int fixed_phy_set_link_update(struct phy_device *phydev,
{
return -ENODEV;
}
-static inline int fixed_phy_update_state(struct phy_device *phydev,
- const struct fixed_phy_status *status,
- const struct fixed_phy_status *changed)
-{
- return -ENODEV;
-}
#endif /* CONFIG_FIXED_PHY */
#endif /* __PHY_FIXED_H */
--
2.7.4
^ permalink raw reply related
* [PATCH 09/10] net: mvneta: add module EEPROM reading support
From: Russell King @ 2018-01-02 17:25 UTC (permalink / raw)
To: Andrew Lunn, Florian Fainelli, Thomas Petazzoni; +Cc: netdev
In-Reply-To: <20180102172242.GE28752@n2100.armlinux.org.uk>
Add support for reading the SFF module's EEPROM via the ethtool API.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
drivers/net/ethernet/marvell/mvneta.c | 18 ++++++++++++++++++
1 file changed, 18 insertions(+)
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 6a018cfa36f6..25e9a551cc8c 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -4073,6 +4073,22 @@ static int mvneta_ethtool_set_wol(struct net_device *dev,
return ret;
}
+static int mvneta_ethtool_get_module_info(struct net_device *dev,
+ struct ethtool_modinfo *modinfo)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+
+ return phylink_ethtool_get_module_info(pp->phylink, modinfo);
+}
+
+static int mvneta_ethtool_get_module_eeprom(struct net_device *dev,
+ struct ethtool_eeprom *ee, u8 *buf)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+
+ return phylink_ethtool_get_module_eeprom(pp->phylink, ee, buf);
+}
+
static int mvneta_ethtool_get_eee(struct net_device *dev,
struct ethtool_eee *eee)
{
@@ -4147,6 +4163,8 @@ static const struct ethtool_ops mvneta_eth_tool_ops = {
.set_link_ksettings = mvneta_ethtool_set_link_ksettings,
.get_wol = mvneta_ethtool_get_wol,
.set_wol = mvneta_ethtool_set_wol,
+ .get_module_info = mvneta_ethtool_get_module_info,
+ .get_module_eeprom = mvneta_ethtool_get_module_eeprom,
.get_eee = mvneta_ethtool_get_eee,
.set_eee = mvneta_ethtool_set_eee,
};
--
2.7.4
^ permalink raw reply related
* [PATCH 08/10] net: mvneta: disable MVNETA_CAUSE_PSC_SYNC_CHANGE interrupt
From: Russell King @ 2018-01-02 17:25 UTC (permalink / raw)
To: Andrew Lunn, Florian Fainelli, Thomas Petazzoni; +Cc: netdev
In-Reply-To: <20180102172242.GE28752@n2100.armlinux.org.uk>
The PSC sync change interrupt can fire multiple times while the link is
down, which is caused by noise on the serdes lines. As this isn't
information we make use of, it's pointless having the interrupt enabled.
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
---
drivers/net/ethernet/marvell/mvneta.c | 12 ++++--------
1 file changed, 4 insertions(+), 8 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 44d665887b50..6a018cfa36f6 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -2704,8 +2704,7 @@ static int mvneta_poll(struct napi_struct *napi, int budget)
mvreg_write(pp, MVNETA_INTR_MISC_CAUSE, 0);
if (cause_misc & (MVNETA_CAUSE_PHY_STATUS_CHANGE |
- MVNETA_CAUSE_LINK_CHANGE |
- MVNETA_CAUSE_PSC_SYNC_CHANGE))
+ MVNETA_CAUSE_LINK_CHANGE))
mvneta_link_change(pp);
}
@@ -3044,8 +3043,7 @@ static void mvneta_start_dev(struct mvneta_port *pp)
mvreg_write(pp, MVNETA_INTR_MISC_MASK,
MVNETA_CAUSE_PHY_STATUS_CHANGE |
- MVNETA_CAUSE_LINK_CHANGE |
- MVNETA_CAUSE_PSC_SYNC_CHANGE);
+ MVNETA_CAUSE_LINK_CHANGE);
phylink_start(pp->phylink);
netif_tx_start_all_queues(pp->dev);
@@ -3570,8 +3568,7 @@ static int mvneta_cpu_online(unsigned int cpu, struct hlist_node *node)
on_each_cpu(mvneta_percpu_unmask_interrupt, pp, true);
mvreg_write(pp, MVNETA_INTR_MISC_MASK,
MVNETA_CAUSE_PHY_STATUS_CHANGE |
- MVNETA_CAUSE_LINK_CHANGE |
- MVNETA_CAUSE_PSC_SYNC_CHANGE);
+ MVNETA_CAUSE_LINK_CHANGE);
netif_tx_start_all_queues(pp->dev);
spin_unlock(&pp->lock);
return 0;
@@ -3612,8 +3609,7 @@ static int mvneta_cpu_dead(unsigned int cpu, struct hlist_node *node)
on_each_cpu(mvneta_percpu_unmask_interrupt, pp, true);
mvreg_write(pp, MVNETA_INTR_MISC_MASK,
MVNETA_CAUSE_PHY_STATUS_CHANGE |
- MVNETA_CAUSE_LINK_CHANGE |
- MVNETA_CAUSE_PSC_SYNC_CHANGE);
+ MVNETA_CAUSE_LINK_CHANGE);
netif_tx_start_all_queues(pp->dev);
return 0;
}
--
2.7.4
^ permalink raw reply related
* [PATCH 07/10] net: mvneta: add EEE support
From: Russell King @ 2018-01-02 17:25 UTC (permalink / raw)
To: Andrew Lunn, Florian Fainelli, Thomas Petazzoni; +Cc: netdev
In-Reply-To: <20180102172242.GE28752@n2100.armlinux.org.uk>
Add support for EEE to mvneta.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
drivers/net/ethernet/marvell/mvneta.c | 99 +++++++++++++++++++++++++++++++++--
1 file changed, 94 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 0b8701d3ba38..44d665887b50 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -244,6 +244,12 @@
#define MVNETA_TXQ_TOKEN_SIZE_REG(q) (0x3e40 + ((q) << 2))
#define MVNETA_TXQ_TOKEN_SIZE_MAX 0x7fffffff
+#define MVNETA_LPI_CTRL_0 0x2cc0
+#define MVNETA_LPI_CTRL_1 0x2cc4
+#define MVNETA_LPI_REQUEST_ENABLE BIT(0)
+#define MVNETA_LPI_CTRL_2 0x2cc8
+#define MVNETA_LPI_STATUS 0x2ccc
+
#define MVNETA_CAUSE_TXQ_SENT_DESC_ALL_MASK 0xff
/* Descriptor ring Macros */
@@ -320,6 +326,11 @@
#define MVNETA_RX_GET_BM_POOL_ID(rxd) \
(((rxd)->status & MVNETA_RXD_BM_POOL_MASK) >> MVNETA_RXD_BM_POOL_SHIFT)
+enum {
+ ETHTOOL_STAT_EEE_WAKEUP,
+ ETHTOOL_MAX_STATS,
+};
+
struct mvneta_statistic {
unsigned short offset;
unsigned short type;
@@ -328,6 +339,7 @@ struct mvneta_statistic {
#define T_REG_32 32
#define T_REG_64 64
+#define T_SW 1
static const struct mvneta_statistic mvneta_statistics[] = {
{ 0x3000, T_REG_64, "good_octets_received", },
@@ -362,6 +374,7 @@ static const struct mvneta_statistic mvneta_statistics[] = {
{ 0x304c, T_REG_32, "broadcast_frames_sent", },
{ 0x3054, T_REG_32, "fc_sent", },
{ 0x300c, T_REG_32, "internal_mac_transmit_err", },
+ { ETHTOOL_STAT_EEE_WAKEUP, T_SW, "eee_wakeup_errors", },
};
struct mvneta_pcpu_stats {
@@ -424,6 +437,10 @@ struct mvneta_port {
struct mvneta_bm_pool *pool_short;
int bm_win_id;
+ bool eee_enabled;
+ bool eee_active;
+ bool tx_lpi_enabled;
+
u64 ethtool_stats[ARRAY_SIZE(mvneta_statistics)];
u32 indir[MVNETA_RSS_LU_TABLE_SIZE];
@@ -3369,6 +3386,18 @@ static void mvneta_mac_config(struct net_device *ndev, unsigned int mode,
}
}
+static void mvneta_set_eee(struct mvneta_port *pp, bool enable)
+{
+ u32 lpi_ctl1;
+
+ lpi_ctl1 = mvreg_read(pp, MVNETA_LPI_CTRL_1);
+ if (enable)
+ lpi_ctl1 |= MVNETA_LPI_REQUEST_ENABLE;
+ else
+ lpi_ctl1 &= ~MVNETA_LPI_REQUEST_ENABLE;
+ mvreg_write(pp, MVNETA_LPI_CTRL_1, lpi_ctl1);
+}
+
static void mvneta_mac_link_down(struct net_device *ndev, unsigned int mode)
{
struct mvneta_port *pp = netdev_priv(ndev);
@@ -3382,6 +3411,9 @@ static void mvneta_mac_link_down(struct net_device *ndev, unsigned int mode)
val |= MVNETA_GMAC_FORCE_LINK_DOWN;
mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val);
}
+
+ pp->eee_active = false;
+ mvneta_set_eee(pp, false);
}
static void mvneta_mac_link_up(struct net_device *ndev, unsigned int mode,
@@ -3398,6 +3430,11 @@ static void mvneta_mac_link_up(struct net_device *ndev, unsigned int mode,
}
mvneta_port_up(pp);
+
+ if (phy && pp->eee_enabled) {
+ pp->eee_active = phy_init_eee(phy, 0) >= 0;
+ mvneta_set_eee(pp, pp->eee_active && pp->tx_lpi_enabled);
+ }
}
static const struct phylink_mac_ops mvneta_phylink_ops = {
@@ -3859,26 +3896,35 @@ static void mvneta_ethtool_update_stats(struct mvneta_port *pp)
{
const struct mvneta_statistic *s;
void __iomem *base = pp->base;
- u32 high, low, val;
- u64 val64;
+ u32 high, low;
+ u64 val;
int i;
for (i = 0, s = mvneta_statistics;
s < mvneta_statistics + ARRAY_SIZE(mvneta_statistics);
s++, i++) {
+ val = 0;
+
switch (s->type) {
case T_REG_32:
val = readl_relaxed(base + s->offset);
- pp->ethtool_stats[i] += val;
break;
case T_REG_64:
/* Docs say to read low 32-bit then high */
low = readl_relaxed(base + s->offset);
high = readl_relaxed(base + s->offset + 4);
- val64 = (u64)high << 32 | low;
- pp->ethtool_stats[i] += val64;
+ val = (u64)high << 32 | low;
+ break;
+ case T_SW:
+ switch (s->offset) {
+ case ETHTOOL_STAT_EEE_WAKEUP:
+ val = phylink_get_eee_err(pp->phylink);
+ break;
+ }
break;
}
+
+ pp->ethtool_stats[i] += val;
}
}
@@ -4031,6 +4077,47 @@ static int mvneta_ethtool_set_wol(struct net_device *dev,
return ret;
}
+static int mvneta_ethtool_get_eee(struct net_device *dev,
+ struct ethtool_eee *eee)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+ u32 lpi_ctl0;
+
+ lpi_ctl0 = mvreg_read(pp, MVNETA_LPI_CTRL_0);
+
+ eee->eee_enabled = pp->eee_enabled;
+ eee->eee_active = pp->eee_active;
+ eee->tx_lpi_enabled = pp->tx_lpi_enabled;
+ eee->tx_lpi_timer = (lpi_ctl0) >> 8; // * scale;
+
+ return phylink_ethtool_get_eee(pp->phylink, eee);
+}
+
+static int mvneta_ethtool_set_eee(struct net_device *dev,
+ struct ethtool_eee *eee)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+ u32 lpi_ctl0;
+
+ /* The Armada 37x documents do not give limits for this other than
+ * it being an 8-bit register. */
+ if (eee->tx_lpi_enabled &&
+ (eee->tx_lpi_timer < 0 || eee->tx_lpi_timer > 255))
+ return -EINVAL;
+
+ lpi_ctl0 = mvreg_read(pp, MVNETA_LPI_CTRL_0);
+ lpi_ctl0 &= ~(0xff << 8);
+ lpi_ctl0 |= eee->tx_lpi_timer << 8;
+ mvreg_write(pp, MVNETA_LPI_CTRL_0, lpi_ctl0);
+
+ pp->eee_enabled = eee->eee_enabled;
+ pp->tx_lpi_enabled = eee->tx_lpi_enabled;
+
+ mvneta_set_eee(pp, eee->tx_lpi_enabled && eee->eee_enabled);
+
+ return phylink_ethtool_set_eee(pp->phylink, eee);
+}
+
static const struct net_device_ops mvneta_netdev_ops = {
.ndo_open = mvneta_open,
.ndo_stop = mvneta_stop,
@@ -4064,6 +4151,8 @@ static const struct ethtool_ops mvneta_eth_tool_ops = {
.set_link_ksettings = mvneta_ethtool_set_link_ksettings,
.get_wol = mvneta_ethtool_get_wol,
.set_wol = mvneta_ethtool_set_wol,
+ .get_eee = mvneta_ethtool_get_eee,
+ .set_eee = mvneta_ethtool_set_eee,
};
/* Initialize hw */
--
2.7.4
^ permalink raw reply related
* [PATCH 06/10] net: mvneta: add flow control support
From: Russell King @ 2018-01-02 17:24 UTC (permalink / raw)
To: Andrew Lunn, Florian Fainelli, Thomas Petazzoni; +Cc: netdev
In-Reply-To: <20180102172242.GE28752@n2100.armlinux.org.uk>
Add support for flow control to mvneta.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
drivers/net/ethernet/marvell/mvneta.c | 32 ++++++++++++++++++++++++++++++++
1 file changed, 32 insertions(+)
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 73643dbd5cb0..0b8701d3ba38 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -3211,6 +3211,8 @@ static void mvneta_validate(struct net_device *ndev, unsigned long *supported,
phylink_set(mask, Autoneg);
phylink_set_port_modes(mask);
+ /* Asymmetric pause is unsupported */
+ phylink_set(mask, Pause);
/* Half-duplex at speeds higher than 100Mbit is unsupported */
phylink_set(mask, 1000baseT_Full);
phylink_set(mask, 1000baseX_Full);
@@ -3249,6 +3251,10 @@ static int mvneta_mac_link_state(struct net_device *ndev,
state->duplex = !!(gmac_stat & MVNETA_GMAC_FULL_DUPLEX);
state->pause = 0;
+ if (gmac_stat & MVNETA_GMAC_RX_FLOW_CTRL_ENABLE)
+ state->pause |= MLO_PAUSE_RX;
+ if (gmac_stat & MVNETA_GMAC_TX_FLOW_CTRL_ENABLE)
+ state->pause |= MLO_PAUSE_TX;
return 1;
}
@@ -3298,6 +3304,11 @@ static void mvneta_mac_config(struct net_device *ndev, unsigned int mode,
phy_interface_mode_is_8023z(state->interface))
new_ctrl2 |= MVNETA_GMAC2_PCS_ENABLE;
+ if (phylink_test(state->advertising, Pause))
+ new_an |= MVNETA_GMAC_ADVERT_SYM_FLOW_CTRL;
+ if (state->pause & MLO_PAUSE_TXRX_MASK)
+ new_an |= MVNETA_GMAC_CONFIG_FLOW_CTRL;
+
if (!phylink_autoneg_inband(mode)) {
/* Phy or fixed speed */
if (state->duplex)
@@ -3326,6 +3337,9 @@ static void mvneta_mac_config(struct net_device *ndev, unsigned int mode,
MVNETA_GMAC_CONFIG_GMII_SPEED |
/* The MAC only supports FD mode */
MVNETA_GMAC_CONFIG_FULL_DUPLEX;
+
+ if (state->pause & MLO_PAUSE_AN && state->an_enabled)
+ new_an |= MVNETA_GMAC_AN_FLOW_CTRL_EN;
}
/* Armada 370 documentation says we can only change the port mode
@@ -3813,6 +3827,22 @@ static int mvneta_ethtool_set_ringparam(struct net_device *dev,
return 0;
}
+static void mvneta_ethtool_get_pauseparam(struct net_device *dev,
+ struct ethtool_pauseparam *pause)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+
+ phylink_ethtool_get_pauseparam(pp->phylink, pause);
+}
+
+static int mvneta_ethtool_set_pauseparam(struct net_device *dev,
+ struct ethtool_pauseparam *pause)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
+
+ return phylink_ethtool_set_pauseparam(pp->phylink, pause);
+}
+
static void mvneta_ethtool_get_strings(struct net_device *netdev, u32 sset,
u8 *data)
{
@@ -4021,6 +4051,8 @@ static const struct ethtool_ops mvneta_eth_tool_ops = {
.get_drvinfo = mvneta_ethtool_get_drvinfo,
.get_ringparam = mvneta_ethtool_get_ringparam,
.set_ringparam = mvneta_ethtool_set_ringparam,
+ .get_pauseparam = mvneta_ethtool_get_pauseparam,
+ .set_pauseparam = mvneta_ethtool_set_pauseparam,
.get_strings = mvneta_ethtool_get_strings,
.get_ethtool_stats = mvneta_ethtool_get_stats,
.get_sset_count = mvneta_ethtool_get_sset_count,
--
2.7.4
^ permalink raw reply related
* [PATCH 05/10] net: mvneta: add 1000BaseX support
From: Russell King @ 2018-01-02 17:24 UTC (permalink / raw)
To: Andrew Lunn, Florian Fainelli, Thomas Petazzoni; +Cc: netdev
In-Reply-To: <20180102172242.GE28752@n2100.armlinux.org.uk>
Add support for 1000BaseX link modes.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
drivers/net/ethernet/marvell/mvneta.c | 59 +++++++++++++++++++++++++++++------
1 file changed, 50 insertions(+), 9 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index ef54a8fc9515..73643dbd5cb0 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -189,6 +189,7 @@
#define MVNETA_GMAC_CTRL_0 0x2c00
#define MVNETA_GMAC_MAX_RX_SIZE_SHIFT 2
#define MVNETA_GMAC_MAX_RX_SIZE_MASK 0x7ffc
+#define MVNETA_GMAC0_PORT_1000BASE_X BIT(1)
#define MVNETA_GMAC0_PORT_ENABLE BIT(0)
#define MVNETA_GMAC_CTRL_2 0x2c08
#define MVNETA_GMAC2_INBAND_AN_ENABLE BIT(0)
@@ -210,9 +211,13 @@
#define MVNETA_GMAC_FORCE_LINK_DOWN BIT(0)
#define MVNETA_GMAC_FORCE_LINK_PASS BIT(1)
#define MVNETA_GMAC_INBAND_AN_ENABLE BIT(2)
+#define MVNETA_GMAC_AN_BYPASS_ENABLE BIT(3)
+#define MVNETA_GMAC_INBAND_RESTART_AN BIT(4)
#define MVNETA_GMAC_CONFIG_MII_SPEED BIT(5)
#define MVNETA_GMAC_CONFIG_GMII_SPEED BIT(6)
#define MVNETA_GMAC_AN_SPEED_EN BIT(7)
+#define MVNETA_GMAC_CONFIG_FLOW_CTRL BIT(8)
+#define MVNETA_GMAC_ADVERT_SYM_FLOW_CTRL BIT(9)
#define MVNETA_GMAC_AN_FLOW_CTRL_EN BIT(11)
#define MVNETA_GMAC_CONFIG_FULL_DUPLEX BIT(12)
#define MVNETA_GMAC_AN_DUPLEX_EN BIT(13)
@@ -3192,10 +3197,11 @@ static void mvneta_validate(struct net_device *ndev, unsigned long *supported,
{
__ETHTOOL_DECLARE_LINK_MODE_MASK(mask) = { 0, };
- /* We only support QSGMII, SGMII and RGMII modes */
+ /* We only support QSGMII, SGMII, 802.3z and RGMII modes */
if (state->interface != PHY_INTERFACE_MODE_NA &&
state->interface != PHY_INTERFACE_MODE_QSGMII &&
state->interface != PHY_INTERFACE_MODE_SGMII &&
+ !phy_interface_mode_is_8023z(state->interface) &&
!phy_interface_mode_is_rgmii(state->interface)) {
bitmap_zero(supported, __ETHTOOL_LINK_MODE_MASK_NBITS);
return;
@@ -3208,10 +3214,14 @@ static void mvneta_validate(struct net_device *ndev, unsigned long *supported,
/* Half-duplex at speeds higher than 100Mbit is unsupported */
phylink_set(mask, 1000baseT_Full);
phylink_set(mask, 1000baseX_Full);
- phylink_set(mask, 10baseT_Half);
- phylink_set(mask, 10baseT_Full);
- phylink_set(mask, 100baseT_Half);
- phylink_set(mask, 100baseT_Full);
+
+ if (!phy_interface_mode_is_8023z(state->interface)) {
+ /* 10M and 100M are only supported in non-802.3z mode */
+ phylink_set(mask, 10baseT_Half);
+ phylink_set(mask, 10baseT_Full);
+ phylink_set(mask, 100baseT_Half);
+ phylink_set(mask, 100baseT_Full);
+ }
bitmap_and(supported, supported, mask,
__ETHTOOL_LINK_MODE_MASK_NBITS);
@@ -3243,14 +3253,27 @@ static int mvneta_mac_link_state(struct net_device *ndev,
return 1;
}
+static void mvneta_mac_an_restart(struct net_device *ndev)
+{
+ struct mvneta_port *pp = netdev_priv(ndev);
+ u32 gmac_an = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
+
+ mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG,
+ gmac_an | MVNETA_GMAC_INBAND_RESTART_AN);
+ mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG,
+ gmac_an & ~MVNETA_GMAC_INBAND_RESTART_AN);
+}
+
static void mvneta_mac_config(struct net_device *ndev, unsigned int mode,
const struct phylink_link_state *state)
{
struct mvneta_port *pp = netdev_priv(ndev);
+ u32 new_ctrl0, gmac_ctrl0 = mvreg_read(pp, MVNETA_GMAC_CTRL_0);
u32 new_ctrl2, gmac_ctrl2 = mvreg_read(pp, MVNETA_GMAC_CTRL_2);
u32 new_clk, gmac_clk = mvreg_read(pp, MVNETA_GMAC_CLOCK_DIVIDER);
u32 new_an, gmac_an = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
+ new_ctrl0 = gmac_ctrl0 & ~MVNETA_GMAC0_PORT_1000BASE_X;
new_ctrl2 = gmac_ctrl2 & ~(MVNETA_GMAC2_INBAND_AN_ENABLE |
MVNETA_GMAC2_PORT_RESET);
new_clk = gmac_clk & ~MVNETA_GMAC_1MS_CLOCK_ENABLE;
@@ -3259,6 +3282,8 @@ static void mvneta_mac_config(struct net_device *ndev, unsigned int mode,
MVNETA_GMAC_CONFIG_MII_SPEED |
MVNETA_GMAC_CONFIG_GMII_SPEED |
MVNETA_GMAC_AN_SPEED_EN |
+ MVNETA_GMAC_ADVERT_SYM_FLOW_CTRL |
+ MVNETA_GMAC_CONFIG_FLOW_CTRL |
MVNETA_GMAC_AN_FLOW_CTRL_EN |
MVNETA_GMAC_CONFIG_FULL_DUPLEX |
MVNETA_GMAC_AN_DUPLEX_EN);
@@ -3269,7 +3294,8 @@ static void mvneta_mac_config(struct net_device *ndev, unsigned int mode,
new_ctrl2 |= MVNETA_GMAC2_PORT_RGMII;
if (state->interface == PHY_INTERFACE_MODE_QSGMII ||
- state->interface == PHY_INTERFACE_MODE_SGMII)
+ state->interface == PHY_INTERFACE_MODE_SGMII ||
+ phy_interface_mode_is_8023z(state->interface))
new_ctrl2 |= MVNETA_GMAC2_PCS_ENABLE;
if (!phylink_autoneg_inband(mode)) {
@@ -3281,7 +3307,7 @@ static void mvneta_mac_config(struct net_device *ndev, unsigned int mode,
new_an |= MVNETA_GMAC_CONFIG_GMII_SPEED;
else if (state->speed == SPEED_100)
new_an |= MVNETA_GMAC_CONFIG_MII_SPEED;
- } else {
+ } else if (state->interface == PHY_INTERFACE_MODE_SGMII) {
/* SGMII mode receives the state from the PHY */
new_ctrl2 |= MVNETA_GMAC2_INBAND_AN_ENABLE;
new_clk |= MVNETA_GMAC_1MS_CLOCK_ENABLE;
@@ -3290,18 +3316,31 @@ static void mvneta_mac_config(struct net_device *ndev, unsigned int mode,
MVNETA_GMAC_INBAND_AN_ENABLE |
MVNETA_GMAC_AN_SPEED_EN |
MVNETA_GMAC_AN_DUPLEX_EN;
+ } else {
+ /* 802.3z negotiation - only 1000base-X */
+ new_ctrl0 |= MVNETA_GMAC0_PORT_1000BASE_X;
+ new_clk |= MVNETA_GMAC_1MS_CLOCK_ENABLE;
+ new_an = (new_an & ~(MVNETA_GMAC_FORCE_LINK_DOWN |
+ MVNETA_GMAC_FORCE_LINK_PASS)) |
+ MVNETA_GMAC_INBAND_AN_ENABLE |
+ MVNETA_GMAC_CONFIG_GMII_SPEED |
+ /* The MAC only supports FD mode */
+ MVNETA_GMAC_CONFIG_FULL_DUPLEX;
}
/* Armada 370 documentation says we can only change the port mode
* and in-band enable when the link is down, so force it down
* while making these changes. We also do this for GMAC_CTRL2 */
- if ((new_ctrl2 ^ gmac_ctrl2) & MVNETA_GMAC2_INBAND_AN_ENABLE ||
+ if ((new_ctrl0 ^ gmac_ctrl0) & MVNETA_GMAC0_PORT_1000BASE_X ||
+ (new_ctrl2 ^ gmac_ctrl2) & MVNETA_GMAC2_INBAND_AN_ENABLE ||
(new_an ^ gmac_an) & MVNETA_GMAC_INBAND_AN_ENABLE) {
mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG,
(gmac_an & ~MVNETA_GMAC_FORCE_LINK_PASS) |
MVNETA_GMAC_FORCE_LINK_DOWN);
}
+ if (new_ctrl0 != gmac_ctrl0)
+ mvreg_write(pp, MVNETA_GMAC_CTRL_0, new_ctrl0);
if (new_ctrl2 != gmac_ctrl2)
mvreg_write(pp, MVNETA_GMAC_CTRL_2, new_ctrl2);
if (new_clk != gmac_clk)
@@ -3350,6 +3389,7 @@ static void mvneta_mac_link_up(struct net_device *ndev, unsigned int mode,
static const struct phylink_mac_ops mvneta_phylink_ops = {
.validate = mvneta_validate,
.mac_link_state = mvneta_mac_link_state,
+ .mac_an_restart = mvneta_mac_an_restart,
.mac_config = mvneta_mac_config,
.mac_link_down = mvneta_mac_link_down,
.mac_link_up = mvneta_mac_link_up,
@@ -4096,7 +4136,8 @@ static int mvneta_port_power_up(struct mvneta_port *pp, int phy_mode)
if (phy_mode == PHY_INTERFACE_MODE_QSGMII)
mvreg_write(pp, MVNETA_SERDES_CFG, MVNETA_QSGMII_SERDES_PROTO);
- else if (phy_mode == PHY_INTERFACE_MODE_SGMII)
+ else if (phy_mode == PHY_INTERFACE_MODE_SGMII ||
+ phy_mode == PHY_INTERFACE_MODE_1000BASEX)
mvreg_write(pp, MVNETA_SERDES_CFG, MVNETA_SGMII_SERDES_PROTO);
else if (!phy_interface_mode_is_rgmii(phy_mode))
return -EINVAL;
--
2.7.4
^ permalink raw reply related
* [PATCH 04/10] net: mvneta: move port configuration
From: Russell King @ 2018-01-02 17:24 UTC (permalink / raw)
To: Andrew Lunn, Florian Fainelli, Thomas Petazzoni; +Cc: netdev
In-Reply-To: <20180102172242.GE28752@n2100.armlinux.org.uk>
Move the port configuration and release of reset to mvneta_mac_config()
along side the rest of the port mode configuration.
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
---
drivers/net/ethernet/marvell/mvneta.c | 51 ++++++++++++++---------------------
1 file changed, 20 insertions(+), 31 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index fabe17bd39f9..ef54a8fc9515 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -3251,7 +3251,8 @@ static void mvneta_mac_config(struct net_device *ndev, unsigned int mode,
u32 new_clk, gmac_clk = mvreg_read(pp, MVNETA_GMAC_CLOCK_DIVIDER);
u32 new_an, gmac_an = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
- new_ctrl2 = gmac_ctrl2 & ~MVNETA_GMAC2_INBAND_AN_ENABLE;
+ new_ctrl2 = gmac_ctrl2 & ~(MVNETA_GMAC2_INBAND_AN_ENABLE |
+ MVNETA_GMAC2_PORT_RESET);
new_clk = gmac_clk & ~MVNETA_GMAC_1MS_CLOCK_ENABLE;
new_an = gmac_an & ~(MVNETA_GMAC_INBAND_AN_ENABLE |
MVNETA_GMAC_INBAND_RESTART_AN |
@@ -3262,6 +3263,15 @@ static void mvneta_mac_config(struct net_device *ndev, unsigned int mode,
MVNETA_GMAC_CONFIG_FULL_DUPLEX |
MVNETA_GMAC_AN_DUPLEX_EN);
+ /* Even though it might look weird, when we're configured in
+ * SGMII or QSGMII mode, the RGMII bit needs to be set.
+ */
+ new_ctrl2 |= MVNETA_GMAC2_PORT_RGMII;
+
+ if (state->interface == PHY_INTERFACE_MODE_QSGMII ||
+ state->interface == PHY_INTERFACE_MODE_SGMII)
+ new_ctrl2 |= MVNETA_GMAC2_PCS_ENABLE;
+
if (!phylink_autoneg_inband(mode)) {
/* Phy or fixed speed */
if (state->duplex)
@@ -3298,6 +3308,12 @@ static void mvneta_mac_config(struct net_device *ndev, unsigned int mode,
mvreg_write(pp, MVNETA_GMAC_CLOCK_DIVIDER, new_clk);
if (new_an != gmac_an)
mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, new_an);
+
+ if (gmac_ctrl2 & MVNETA_GMAC2_PORT_RESET) {
+ while ((mvreg_read(pp, MVNETA_GMAC_CTRL_2) &
+ MVNETA_GMAC2_PORT_RESET) != 0)
+ continue;
+ }
}
static void mvneta_mac_link_down(struct net_device *ndev, unsigned int mode)
@@ -4075,42 +4091,15 @@ static void mvneta_conf_mbus_windows(struct mvneta_port *pp,
/* Power up the port */
static int mvneta_port_power_up(struct mvneta_port *pp, int phy_mode)
{
- u32 ctrl;
-
/* MAC Cause register should be cleared */
mvreg_write(pp, MVNETA_UNIT_INTR_CAUSE, 0);
- ctrl = mvreg_read(pp, MVNETA_GMAC_CTRL_2);
-
- /* Even though it might look weird, when we're configured in
- * SGMII or QSGMII mode, the RGMII bit needs to be set.
- */
- switch(phy_mode) {
- case PHY_INTERFACE_MODE_QSGMII:
+ if (phy_mode == PHY_INTERFACE_MODE_QSGMII)
mvreg_write(pp, MVNETA_SERDES_CFG, MVNETA_QSGMII_SERDES_PROTO);
- ctrl |= MVNETA_GMAC2_PCS_ENABLE | MVNETA_GMAC2_PORT_RGMII;
- break;
- case PHY_INTERFACE_MODE_SGMII:
+ else if (phy_mode == PHY_INTERFACE_MODE_SGMII)
mvreg_write(pp, MVNETA_SERDES_CFG, MVNETA_SGMII_SERDES_PROTO);
- ctrl |= MVNETA_GMAC2_PCS_ENABLE | MVNETA_GMAC2_PORT_RGMII;
- break;
- case PHY_INTERFACE_MODE_RGMII:
- case PHY_INTERFACE_MODE_RGMII_ID:
- case PHY_INTERFACE_MODE_RGMII_RXID:
- case PHY_INTERFACE_MODE_RGMII_TXID:
- ctrl |= MVNETA_GMAC2_PORT_RGMII;
- break;
- default:
+ else if (!phy_interface_mode_is_rgmii(phy_mode))
return -EINVAL;
- }
-
- /* Cancel Port Reset */
- ctrl &= ~MVNETA_GMAC2_PORT_RESET;
- mvreg_write(pp, MVNETA_GMAC_CTRL_2, ctrl);
-
- while ((mvreg_read(pp, MVNETA_GMAC_CTRL_2) &
- MVNETA_GMAC2_PORT_RESET) != 0)
- continue;
return 0;
}
--
2.7.4
^ permalink raw reply related
* [PATCH 03/10] net: mvneta: convert to phylink
From: Russell King @ 2018-01-02 17:24 UTC (permalink / raw)
To: Andrew Lunn, Florian Fainelli, Thomas Petazzoni; +Cc: netdev
In-Reply-To: <20180102172242.GE28752@n2100.armlinux.org.uk>
Convert mvneta to use phylink, which models the MAC to PHY link in
a generic, reusable form.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
- remove unused sync status
---
drivers/net/ethernet/marvell/Kconfig | 2 +-
drivers/net/ethernet/marvell/mvneta.c | 417 ++++++++++++++--------------------
2 files changed, 176 insertions(+), 243 deletions(-)
diff --git a/drivers/net/ethernet/marvell/Kconfig b/drivers/net/ethernet/marvell/Kconfig
index da6fb825afea..ebe5c9148935 100644
--- a/drivers/net/ethernet/marvell/Kconfig
+++ b/drivers/net/ethernet/marvell/Kconfig
@@ -60,7 +60,7 @@ config MVNETA
depends on ARCH_MVEBU || COMPILE_TEST
depends on HAS_DMA
select MVMDIO
- select FIXED_PHY
+ select PHYLINK
---help---
This driver supports the network interface units in the
Marvell ARMADA XP, ARMADA 370, ARMADA 38x and
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 3dfe4b3edef2..fabe17bd39f9 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -28,7 +28,7 @@
#include <linux/of_mdio.h>
#include <linux/of_net.h>
#include <linux/phy.h>
-#include <linux/phy_fixed.h>
+#include <linux/phylink.h>
#include <linux/platform_device.h>
#include <linux/skbuff.h>
#include <net/hwbm.h>
@@ -204,6 +204,8 @@
#define MVNETA_GMAC_TX_FLOW_CTRL_ENABLE BIT(5)
#define MVNETA_GMAC_RX_FLOW_CTRL_ACTIVE BIT(6)
#define MVNETA_GMAC_TX_FLOW_CTRL_ACTIVE BIT(7)
+#define MVNETA_GMAC_AN_COMPLETE BIT(11)
+#define MVNETA_GMAC_SYNC_OK BIT(14)
#define MVNETA_GMAC_AUTONEG_CONFIG 0x2c0c
#define MVNETA_GMAC_FORCE_LINK_DOWN BIT(0)
#define MVNETA_GMAC_FORCE_LINK_PASS BIT(1)
@@ -407,14 +409,10 @@ struct mvneta_port {
u16 tx_ring_size;
u16 rx_ring_size;
- struct mii_bus *mii_bus;
phy_interface_t phy_interface;
- struct device_node *phy_node;
- unsigned int link;
- unsigned int duplex;
- unsigned int speed;
+ struct device_node *dn;
unsigned int tx_csum_limit;
- unsigned int use_inband_status:1;
+ struct phylink *phylink;
struct mvneta_bm *bm_priv;
struct mvneta_bm_pool *pool_long;
@@ -1214,10 +1212,6 @@ static void mvneta_port_disable(struct mvneta_port *pp)
val &= ~MVNETA_GMAC0_PORT_ENABLE;
mvreg_write(pp, MVNETA_GMAC_CTRL_0, val);
- pp->link = 0;
- pp->duplex = -1;
- pp->speed = 0;
-
udelay(200);
}
@@ -1277,44 +1271,6 @@ static void mvneta_set_other_mcast_table(struct mvneta_port *pp, int queue)
mvreg_write(pp, MVNETA_DA_FILT_OTH_MCAST + offset, val);
}
-static void mvneta_set_autoneg(struct mvneta_port *pp, int enable)
-{
- u32 val;
-
- if (enable) {
- val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
- val &= ~(MVNETA_GMAC_FORCE_LINK_PASS |
- MVNETA_GMAC_FORCE_LINK_DOWN |
- MVNETA_GMAC_AN_FLOW_CTRL_EN);
- val |= MVNETA_GMAC_INBAND_AN_ENABLE |
- MVNETA_GMAC_AN_SPEED_EN |
- MVNETA_GMAC_AN_DUPLEX_EN;
- mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val);
-
- val = mvreg_read(pp, MVNETA_GMAC_CLOCK_DIVIDER);
- val |= MVNETA_GMAC_1MS_CLOCK_ENABLE;
- mvreg_write(pp, MVNETA_GMAC_CLOCK_DIVIDER, val);
-
- val = mvreg_read(pp, MVNETA_GMAC_CTRL_2);
- val |= MVNETA_GMAC2_INBAND_AN_ENABLE;
- mvreg_write(pp, MVNETA_GMAC_CTRL_2, val);
- } else {
- val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
- val &= ~(MVNETA_GMAC_INBAND_AN_ENABLE |
- MVNETA_GMAC_AN_SPEED_EN |
- MVNETA_GMAC_AN_DUPLEX_EN);
- mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val);
-
- val = mvreg_read(pp, MVNETA_GMAC_CLOCK_DIVIDER);
- val &= ~MVNETA_GMAC_1MS_CLOCK_ENABLE;
- mvreg_write(pp, MVNETA_GMAC_CLOCK_DIVIDER, val);
-
- val = mvreg_read(pp, MVNETA_GMAC_CTRL_2);
- val &= ~MVNETA_GMAC2_INBAND_AN_ENABLE;
- mvreg_write(pp, MVNETA_GMAC_CTRL_2, val);
- }
-}
-
static void mvneta_percpu_unmask_interrupt(void *arg)
{
struct mvneta_port *pp = arg;
@@ -1467,7 +1423,6 @@ static void mvneta_defaults_set(struct mvneta_port *pp)
val &= ~MVNETA_PHY_POLLING_ENABLE;
mvreg_write(pp, MVNETA_UNIT_CONTROL, val);
- mvneta_set_autoneg(pp, pp->use_inband_status);
mvneta_set_ucast_table(pp, -1);
mvneta_set_special_mcast_table(pp, -1);
mvneta_set_other_mcast_table(pp, -1);
@@ -2692,26 +2647,11 @@ static irqreturn_t mvneta_percpu_isr(int irq, void *dev_id)
return IRQ_HANDLED;
}
-static int mvneta_fixed_link_update(struct mvneta_port *pp,
- struct phy_device *phy)
+static void mvneta_link_change(struct mvneta_port *pp)
{
- struct fixed_phy_status status;
- struct fixed_phy_status changed = {};
u32 gmac_stat = mvreg_read(pp, MVNETA_GMAC_STATUS);
- status.link = !!(gmac_stat & MVNETA_GMAC_LINK_UP);
- if (gmac_stat & MVNETA_GMAC_SPEED_1000)
- status.speed = SPEED_1000;
- else if (gmac_stat & MVNETA_GMAC_SPEED_100)
- status.speed = SPEED_100;
- else
- status.speed = SPEED_10;
- status.duplex = !!(gmac_stat & MVNETA_GMAC_FULL_DUPLEX);
- changed.link = 1;
- changed.speed = 1;
- changed.duplex = 1;
- fixed_phy_update_state(phy, &status, &changed);
- return 0;
+ phylink_mac_change(pp->phylink, !!(gmac_stat & MVNETA_GMAC_LINK_UP));
}
/* NAPI handler
@@ -2727,7 +2667,6 @@ static int mvneta_poll(struct napi_struct *napi, int budget)
u32 cause_rx_tx;
int rx_queue;
struct mvneta_port *pp = netdev_priv(napi->dev);
- struct net_device *ndev = pp->dev;
struct mvneta_pcpu_port *port = this_cpu_ptr(pp->ports);
if (!netif_running(pp->dev)) {
@@ -2741,12 +2680,11 @@ static int mvneta_poll(struct napi_struct *napi, int budget)
u32 cause_misc = mvreg_read(pp, MVNETA_INTR_MISC_CAUSE);
mvreg_write(pp, MVNETA_INTR_MISC_CAUSE, 0);
- if (pp->use_inband_status && (cause_misc &
- (MVNETA_CAUSE_PHY_STATUS_CHANGE |
- MVNETA_CAUSE_LINK_CHANGE |
- MVNETA_CAUSE_PSC_SYNC_CHANGE))) {
- mvneta_fixed_link_update(pp, ndev->phydev);
- }
+
+ if (cause_misc & (MVNETA_CAUSE_PHY_STATUS_CHANGE |
+ MVNETA_CAUSE_LINK_CHANGE |
+ MVNETA_CAUSE_PSC_SYNC_CHANGE))
+ mvneta_link_change(pp);
}
/* Release Tx descriptors */
@@ -3060,7 +2998,6 @@ static int mvneta_setup_txqs(struct mvneta_port *pp)
static void mvneta_start_dev(struct mvneta_port *pp)
{
int cpu;
- struct net_device *ndev = pp->dev;
mvneta_max_rx_size_set(pp, pp->pkt_size);
mvneta_txq_max_tx_size_set(pp, pp->pkt_size);
@@ -3088,16 +3025,15 @@ static void mvneta_start_dev(struct mvneta_port *pp)
MVNETA_CAUSE_LINK_CHANGE |
MVNETA_CAUSE_PSC_SYNC_CHANGE);
- phy_start(ndev->phydev);
+ phylink_start(pp->phylink);
netif_tx_start_all_queues(pp->dev);
}
static void mvneta_stop_dev(struct mvneta_port *pp)
{
unsigned int cpu;
- struct net_device *ndev = pp->dev;
- phy_stop(ndev->phydev);
+ phylink_stop(pp->phylink);
if (!pp->neta_armada3700) {
for_each_online_cpu(cpu) {
@@ -3251,55 +3187,141 @@ static int mvneta_set_mac_addr(struct net_device *dev, void *addr)
return 0;
}
-static void mvneta_mac_config(struct net_device *ndev)
+static void mvneta_validate(struct net_device *ndev, unsigned long *supported,
+ struct phylink_link_state *state)
+{
+ __ETHTOOL_DECLARE_LINK_MODE_MASK(mask) = { 0, };
+
+ /* We only support QSGMII, SGMII and RGMII modes */
+ if (state->interface != PHY_INTERFACE_MODE_NA &&
+ state->interface != PHY_INTERFACE_MODE_QSGMII &&
+ state->interface != PHY_INTERFACE_MODE_SGMII &&
+ !phy_interface_mode_is_rgmii(state->interface)) {
+ bitmap_zero(supported, __ETHTOOL_LINK_MODE_MASK_NBITS);
+ return;
+ }
+
+ /* Allow all the expected bits */
+ phylink_set(mask, Autoneg);
+ phylink_set_port_modes(mask);
+
+ /* Half-duplex at speeds higher than 100Mbit is unsupported */
+ phylink_set(mask, 1000baseT_Full);
+ phylink_set(mask, 1000baseX_Full);
+ phylink_set(mask, 10baseT_Half);
+ phylink_set(mask, 10baseT_Full);
+ phylink_set(mask, 100baseT_Half);
+ phylink_set(mask, 100baseT_Full);
+
+ bitmap_and(supported, supported, mask,
+ __ETHTOOL_LINK_MODE_MASK_NBITS);
+ bitmap_and(state->advertising, state->advertising, mask,
+ __ETHTOOL_LINK_MODE_MASK_NBITS);
+}
+
+static int mvneta_mac_link_state(struct net_device *ndev,
+ struct phylink_link_state *state)
{
struct mvneta_port *pp = netdev_priv(ndev);
- struct phy_device *phydev = ndev->phydev;
- u32 val;
+ u32 gmac_stat;
- if ((pp->speed != phydev->speed) ||
- (pp->duplex != phydev->duplex)) {
- val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
- val &= ~(MVNETA_GMAC_CONFIG_MII_SPEED |
- MVNETA_GMAC_CONFIG_GMII_SPEED |
- MVNETA_GMAC_CONFIG_FULL_DUPLEX);
+ gmac_stat = mvreg_read(pp, MVNETA_GMAC_STATUS);
- if (phydev->duplex)
- val |= MVNETA_GMAC_CONFIG_FULL_DUPLEX;
+ if (gmac_stat & MVNETA_GMAC_SPEED_1000)
+ state->speed = SPEED_1000;
+ else if (gmac_stat & MVNETA_GMAC_SPEED_100)
+ state->speed = SPEED_100;
+ else
+ state->speed = SPEED_10;
- if (phydev->speed == SPEED_1000)
- val |= MVNETA_GMAC_CONFIG_GMII_SPEED;
- else if (phydev->speed == SPEED_100)
- val |= MVNETA_GMAC_CONFIG_MII_SPEED;
+ state->an_complete = !!(gmac_stat & MVNETA_GMAC_AN_COMPLETE);
+ state->link = !!(gmac_stat & MVNETA_GMAC_LINK_UP);
+ state->duplex = !!(gmac_stat & MVNETA_GMAC_FULL_DUPLEX);
- mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val);
+ state->pause = 0;
+
+ return 1;
+}
+
+static void mvneta_mac_config(struct net_device *ndev, unsigned int mode,
+ const struct phylink_link_state *state)
+{
+ struct mvneta_port *pp = netdev_priv(ndev);
+ u32 new_ctrl2, gmac_ctrl2 = mvreg_read(pp, MVNETA_GMAC_CTRL_2);
+ u32 new_clk, gmac_clk = mvreg_read(pp, MVNETA_GMAC_CLOCK_DIVIDER);
+ u32 new_an, gmac_an = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
+
+ new_ctrl2 = gmac_ctrl2 & ~MVNETA_GMAC2_INBAND_AN_ENABLE;
+ new_clk = gmac_clk & ~MVNETA_GMAC_1MS_CLOCK_ENABLE;
+ new_an = gmac_an & ~(MVNETA_GMAC_INBAND_AN_ENABLE |
+ MVNETA_GMAC_INBAND_RESTART_AN |
+ MVNETA_GMAC_CONFIG_MII_SPEED |
+ MVNETA_GMAC_CONFIG_GMII_SPEED |
+ MVNETA_GMAC_AN_SPEED_EN |
+ MVNETA_GMAC_AN_FLOW_CTRL_EN |
+ MVNETA_GMAC_CONFIG_FULL_DUPLEX |
+ MVNETA_GMAC_AN_DUPLEX_EN);
+
+ if (!phylink_autoneg_inband(mode)) {
+ /* Phy or fixed speed */
+ if (state->duplex)
+ new_an |= MVNETA_GMAC_CONFIG_FULL_DUPLEX;
+
+ if (state->speed == SPEED_1000)
+ new_an |= MVNETA_GMAC_CONFIG_GMII_SPEED;
+ else if (state->speed == SPEED_100)
+ new_an |= MVNETA_GMAC_CONFIG_MII_SPEED;
+ } else {
+ /* SGMII mode receives the state from the PHY */
+ new_ctrl2 |= MVNETA_GMAC2_INBAND_AN_ENABLE;
+ new_clk |= MVNETA_GMAC_1MS_CLOCK_ENABLE;
+ new_an = (new_an & ~(MVNETA_GMAC_FORCE_LINK_DOWN |
+ MVNETA_GMAC_FORCE_LINK_PASS)) |
+ MVNETA_GMAC_INBAND_AN_ENABLE |
+ MVNETA_GMAC_AN_SPEED_EN |
+ MVNETA_GMAC_AN_DUPLEX_EN;
+ }
- pp->duplex = phydev->duplex;
- pp->speed = phydev->speed;
+ /* Armada 370 documentation says we can only change the port mode
+ * and in-band enable when the link is down, so force it down
+ * while making these changes. We also do this for GMAC_CTRL2 */
+ if ((new_ctrl2 ^ gmac_ctrl2) & MVNETA_GMAC2_INBAND_AN_ENABLE ||
+ (new_an ^ gmac_an) & MVNETA_GMAC_INBAND_AN_ENABLE) {
+ mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG,
+ (gmac_an & ~MVNETA_GMAC_FORCE_LINK_PASS) |
+ MVNETA_GMAC_FORCE_LINK_DOWN);
}
+
+ if (new_ctrl2 != gmac_ctrl2)
+ mvreg_write(pp, MVNETA_GMAC_CTRL_2, new_ctrl2);
+ if (new_clk != gmac_clk)
+ mvreg_write(pp, MVNETA_GMAC_CLOCK_DIVIDER, new_clk);
+ if (new_an != gmac_an)
+ mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, new_an);
}
-static void mvneta_mac_link_down(struct net_device *ndev, bool autoneg)
+static void mvneta_mac_link_down(struct net_device *ndev, unsigned int mode)
{
struct mvneta_port *pp = netdev_priv(ndev);
u32 val;
- if (!autoneg) {
+ mvneta_port_down(pp);
+
+ if (!phylink_autoneg_inband(mode)) {
val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
val &= ~MVNETA_GMAC_FORCE_LINK_PASS;
val |= MVNETA_GMAC_FORCE_LINK_DOWN;
mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val);
}
-
- mvneta_port_down(pp);
}
-static void mvneta_mac_link_up(struct net_device *ndev, bool autoneg)
+static void mvneta_mac_link_up(struct net_device *ndev, unsigned int mode,
+ struct phy_device *phy)
{
struct mvneta_port *pp = netdev_priv(ndev);
u32 val;
- if (!autoneg) {
+ if (!phylink_autoneg_inband(mode)) {
val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
val &= ~MVNETA_GMAC_FORCE_LINK_DOWN;
val |= MVNETA_GMAC_FORCE_LINK_PASS;
@@ -3309,64 +3331,31 @@ static void mvneta_mac_link_up(struct net_device *ndev, bool autoneg)
mvneta_port_up(pp);
}
-static void mvneta_adjust_link(struct net_device *ndev)
-{
- struct mvneta_port *pp = netdev_priv(ndev);
- struct phy_device *phydev = ndev->phydev;
- int status_change = 0;
-
- if (phydev->link)
- mvneta_mac_config(ndev);
-
- if (phydev->link != pp->link) {
- if (!phydev->link) {
- pp->duplex = -1;
- pp->speed = 0;
- }
-
- pp->link = phydev->link;
- status_change = 1;
- }
-
- if (status_change) {
- if (phydev->link)
- mvneta_mac_link_down(ndev, pp->use_inband_status);
- else
- mvneta_mac_link_up(ndev, pp->use_inband_status);
- phy_print_status(phydev);
- }
-}
+static const struct phylink_mac_ops mvneta_phylink_ops = {
+ .validate = mvneta_validate,
+ .mac_link_state = mvneta_mac_link_state,
+ .mac_config = mvneta_mac_config,
+ .mac_link_down = mvneta_mac_link_down,
+ .mac_link_up = mvneta_mac_link_up,
+};
static int mvneta_mdio_probe(struct mvneta_port *pp)
{
- struct phy_device *phy_dev;
struct ethtool_wolinfo wol = { .cmd = ETHTOOL_GWOL };
+ int err = phylink_of_phy_connect(pp->phylink, pp->dn, 0);
- phy_dev = of_phy_connect(pp->dev, pp->phy_node, mvneta_adjust_link, 0,
- pp->phy_interface);
- if (!phy_dev) {
- netdev_err(pp->dev, "could not find the PHY\n");
- return -ENODEV;
- }
+ if (err)
+ netdev_err(pp->dev, "could not attach PHY: %d\n", err);
- phy_ethtool_get_wol(phy_dev, &wol);
+ phylink_ethtool_get_wol(pp->phylink, &wol);
device_set_wakeup_capable(&pp->dev->dev, !!wol.supported);
- phy_dev->supported &= PHY_GBIT_FEATURES;
- phy_dev->advertising = phy_dev->supported;
-
- pp->link = 0;
- pp->duplex = 0;
- pp->speed = 0;
-
- return 0;
+ return err;
}
static void mvneta_mdio_remove(struct mvneta_port *pp)
{
- struct net_device *ndev = pp->dev;
-
- phy_disconnect(ndev->phydev);
+ phylink_disconnect_phy(pp->phylink);
}
/* Electing a CPU must be done in an atomic way: it should be done
@@ -3645,10 +3634,9 @@ static int mvneta_stop(struct net_device *dev)
static int mvneta_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
{
- if (!dev->phydev)
- return -ENOTSUPP;
+ struct mvneta_port *pp = netdev_priv(dev);
- return phy_mii_ioctl(dev->phydev, ifr, cmd);
+ return phylink_mii_ioctl(pp->phylink, ifr, cmd);
}
/* Ethtool methods */
@@ -3659,44 +3647,25 @@ mvneta_ethtool_set_link_ksettings(struct net_device *ndev,
const struct ethtool_link_ksettings *cmd)
{
struct mvneta_port *pp = netdev_priv(ndev);
- struct phy_device *phydev = ndev->phydev;
-
- if (!phydev)
- return -ENODEV;
-
- if ((cmd->base.autoneg == AUTONEG_ENABLE) != pp->use_inband_status) {
- u32 val;
-
- mvneta_set_autoneg(pp, cmd->base.autoneg == AUTONEG_ENABLE);
- if (cmd->base.autoneg == AUTONEG_DISABLE) {
- val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
- val &= ~(MVNETA_GMAC_CONFIG_MII_SPEED |
- MVNETA_GMAC_CONFIG_GMII_SPEED |
- MVNETA_GMAC_CONFIG_FULL_DUPLEX);
-
- if (phydev->duplex)
- val |= MVNETA_GMAC_CONFIG_FULL_DUPLEX;
-
- if (phydev->speed == SPEED_1000)
- val |= MVNETA_GMAC_CONFIG_GMII_SPEED;
- else if (phydev->speed == SPEED_100)
- val |= MVNETA_GMAC_CONFIG_MII_SPEED;
+ return phylink_ethtool_ksettings_set(pp->phylink, cmd);
+}
- mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val);
- }
+/* Get link ksettings for ethtools */
+static int
+mvneta_ethtool_get_link_ksettings(struct net_device *ndev,
+ struct ethtool_link_ksettings *cmd)
+{
+ struct mvneta_port *pp = netdev_priv(ndev);
- pp->use_inband_status = (cmd->base.autoneg == AUTONEG_ENABLE);
- netdev_info(pp->dev, "autoneg status set to %i\n",
- pp->use_inband_status);
+ return phylink_ethtool_ksettings_get(pp->phylink, cmd);
+}
- if (netif_running(ndev)) {
- mvneta_port_down(pp);
- mvneta_port_up(pp);
- }
- }
+static int mvneta_ethtool_nway_reset(struct net_device *dev)
+{
+ struct mvneta_port *pp = netdev_priv(dev);
- return phy_ethtool_ksettings_set(ndev->phydev, cmd);
+ return phylink_ethtool_nway_reset(pp->phylink);
}
/* Set interrupt coalescing for ethtools */
@@ -3958,22 +3927,18 @@ static int mvneta_ethtool_get_rxfh(struct net_device *dev, u32 *indir, u8 *key,
static void mvneta_ethtool_get_wol(struct net_device *dev,
struct ethtool_wolinfo *wol)
{
- wol->supported = 0;
- wol->wolopts = 0;
+ struct mvneta_port *pp = netdev_priv(dev);
- if (dev->phydev)
- phy_ethtool_get_wol(dev->phydev, wol);
+ phylink_ethtool_get_wol(pp->phylink, wol);
}
static int mvneta_ethtool_set_wol(struct net_device *dev,
struct ethtool_wolinfo *wol)
{
+ struct mvneta_port *pp = netdev_priv(dev);
int ret;
- if (!dev->phydev)
- return -EOPNOTSUPP;
-
- ret = phy_ethtool_set_wol(dev->phydev, wol);
+ ret = phylink_ethtool_set_wol(pp->phylink, wol);
if (!ret)
device_set_wakeup_enable(&dev->dev, !!wol->wolopts);
@@ -3993,7 +3958,7 @@ static const struct net_device_ops mvneta_netdev_ops = {
};
static const struct ethtool_ops mvneta_eth_tool_ops = {
- .nway_reset = phy_ethtool_nway_reset,
+ .nway_reset = mvneta_ethtool_nway_reset,
.get_link = ethtool_op_get_link,
.set_coalesce = mvneta_ethtool_set_coalesce,
.get_coalesce = mvneta_ethtool_get_coalesce,
@@ -4007,7 +3972,7 @@ static const struct ethtool_ops mvneta_eth_tool_ops = {
.get_rxnfc = mvneta_ethtool_get_rxnfc,
.get_rxfh = mvneta_ethtool_get_rxfh,
.set_rxfh = mvneta_ethtool_set_rxfh,
- .get_link_ksettings = phy_ethtool_get_link_ksettings,
+ .get_link_ksettings = mvneta_ethtool_get_link_ksettings,
.set_link_ksettings = mvneta_ethtool_set_link_ksettings,
.get_wol = mvneta_ethtool_get_wol,
.set_wol = mvneta_ethtool_set_wol,
@@ -4155,14 +4120,13 @@ static int mvneta_probe(struct platform_device *pdev)
{
struct resource *res;
struct device_node *dn = pdev->dev.of_node;
- struct device_node *phy_node;
struct device_node *bm_node;
struct mvneta_port *pp;
struct net_device *dev;
+ struct phylink *phylink;
const char *dt_mac_addr;
char hw_mac_addr[ETH_ALEN];
const char *mac_from;
- const char *managed;
int tx_csum_limit;
int phy_mode;
int err;
@@ -4178,31 +4142,18 @@ static int mvneta_probe(struct platform_device *pdev)
goto err_free_netdev;
}
- phy_node = of_parse_phandle(dn, "phy", 0);
- if (!phy_node) {
- if (!of_phy_is_fixed_link(dn)) {
- dev_err(&pdev->dev, "no PHY specified\n");
- err = -ENODEV;
- goto err_free_irq;
- }
-
- err = of_phy_register_fixed_link(dn);
- if (err < 0) {
- dev_err(&pdev->dev, "cannot register fixed PHY\n");
- goto err_free_irq;
- }
-
- /* In the case of a fixed PHY, the DT node associated
- * to the PHY is the Ethernet MAC DT node.
- */
- phy_node = of_node_get(dn);
- }
-
phy_mode = of_get_phy_mode(dn);
if (phy_mode < 0) {
dev_err(&pdev->dev, "incorrect phy-mode\n");
err = -EINVAL;
- goto err_put_phy_node;
+ goto err_free_irq;
+ }
+
+ phylink = phylink_create(dev, pdev->dev.fwnode, phy_mode,
+ &mvneta_phylink_ops);
+ if (IS_ERR(phylink)) {
+ err = PTR_ERR(phylink);
+ goto err_free_irq;
}
dev->tx_queue_len = MVNETA_MAX_TXD;
@@ -4213,12 +4164,9 @@ static int mvneta_probe(struct platform_device *pdev)
pp = netdev_priv(dev);
spin_lock_init(&pp->lock);
- pp->phy_node = phy_node;
+ pp->phylink = phylink;
pp->phy_interface = phy_mode;
-
- err = of_property_read_string(dn, "managed", &managed);
- pp->use_inband_status = (err == 0 &&
- strcmp(managed, "in-band-status") == 0);
+ pp->dn = dn;
pp->rxq_def = rxq_def;
@@ -4240,7 +4188,7 @@ static int mvneta_probe(struct platform_device *pdev)
pp->clk = devm_clk_get(&pdev->dev, NULL);
if (IS_ERR(pp->clk)) {
err = PTR_ERR(pp->clk);
- goto err_put_phy_node;
+ goto err_free_phylink;
}
clk_prepare_enable(pp->clk);
@@ -4377,14 +4325,6 @@ static int mvneta_probe(struct platform_device *pdev)
platform_set_drvdata(pdev, pp->dev);
- if (pp->use_inband_status) {
- struct phy_device *phy = of_phy_find_device(dn);
-
- mvneta_fixed_link_update(pp, phy);
-
- put_device(&phy->mdio.dev);
- }
-
return 0;
err_netdev:
@@ -4401,10 +4341,9 @@ static int mvneta_probe(struct platform_device *pdev)
err_clk:
clk_disable_unprepare(pp->clk_bus);
clk_disable_unprepare(pp->clk);
-err_put_phy_node:
- of_node_put(phy_node);
- if (of_phy_is_fixed_link(dn))
- of_phy_deregister_fixed_link(dn);
+err_free_phylink:
+ if (pp->phylink)
+ phylink_destroy(pp->phylink);
err_free_irq:
irq_dispose_mapping(dev->irq);
err_free_netdev:
@@ -4416,7 +4355,6 @@ static int mvneta_probe(struct platform_device *pdev)
static int mvneta_remove(struct platform_device *pdev)
{
struct net_device *dev = platform_get_drvdata(pdev);
- struct device_node *dn = pdev->dev.of_node;
struct mvneta_port *pp = netdev_priv(dev);
unregister_netdev(dev);
@@ -4424,10 +4362,8 @@ static int mvneta_remove(struct platform_device *pdev)
clk_disable_unprepare(pp->clk);
free_percpu(pp->ports);
free_percpu(pp->stats);
- if (of_phy_is_fixed_link(dn))
- of_phy_deregister_fixed_link(dn);
irq_dispose_mapping(dev->irq);
- of_node_put(pp->phy_node);
+ phylink_destroy(pp->phylink);
free_netdev(dev);
if (pp->bm_priv) {
@@ -4481,9 +4417,6 @@ static int mvneta_resume(struct device *device)
return err;
}
- if (pp->use_inband_status)
- mvneta_fixed_link_update(pp, dev->phydev);
-
netif_device_attach(dev);
rtnl_lock();
if (netif_running(dev)) {
--
2.7.4
^ permalink raw reply related
* [PATCH 02/10] net: mvneta: prepare to convert to phylink
From: Russell King @ 2018-01-02 17:24 UTC (permalink / raw)
To: Andrew Lunn, Florian Fainelli, Thomas Petazzoni; +Cc: netdev
In-Reply-To: <20180102172242.GE28752@n2100.armlinux.org.uk>
Prepare to convert mvneta to phylink by splitting the adjust_link
function into its consituent parts.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
drivers/net/ethernet/marvell/mvneta.c | 101 ++++++++++++++++++++--------------
1 file changed, 60 insertions(+), 41 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index e08bd20352cb..3dfe4b3edef2 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -3251,37 +3251,73 @@ static int mvneta_set_mac_addr(struct net_device *dev, void *addr)
return 0;
}
-static void mvneta_adjust_link(struct net_device *ndev)
+static void mvneta_mac_config(struct net_device *ndev)
{
struct mvneta_port *pp = netdev_priv(ndev);
struct phy_device *phydev = ndev->phydev;
- int status_change = 0;
+ u32 val;
- if (phydev->link) {
- if ((pp->speed != phydev->speed) ||
- (pp->duplex != phydev->duplex)) {
- u32 val;
+ if ((pp->speed != phydev->speed) ||
+ (pp->duplex != phydev->duplex)) {
+ val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
+ val &= ~(MVNETA_GMAC_CONFIG_MII_SPEED |
+ MVNETA_GMAC_CONFIG_GMII_SPEED |
+ MVNETA_GMAC_CONFIG_FULL_DUPLEX);
- val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
- val &= ~(MVNETA_GMAC_CONFIG_MII_SPEED |
- MVNETA_GMAC_CONFIG_GMII_SPEED |
- MVNETA_GMAC_CONFIG_FULL_DUPLEX);
+ if (phydev->duplex)
+ val |= MVNETA_GMAC_CONFIG_FULL_DUPLEX;
- if (phydev->duplex)
- val |= MVNETA_GMAC_CONFIG_FULL_DUPLEX;
+ if (phydev->speed == SPEED_1000)
+ val |= MVNETA_GMAC_CONFIG_GMII_SPEED;
+ else if (phydev->speed == SPEED_100)
+ val |= MVNETA_GMAC_CONFIG_MII_SPEED;
- if (phydev->speed == SPEED_1000)
- val |= MVNETA_GMAC_CONFIG_GMII_SPEED;
- else if (phydev->speed == SPEED_100)
- val |= MVNETA_GMAC_CONFIG_MII_SPEED;
+ mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val);
- mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val);
+ pp->duplex = phydev->duplex;
+ pp->speed = phydev->speed;
+ }
+}
- pp->duplex = phydev->duplex;
- pp->speed = phydev->speed;
- }
+static void mvneta_mac_link_down(struct net_device *ndev, bool autoneg)
+{
+ struct mvneta_port *pp = netdev_priv(ndev);
+ u32 val;
+
+ if (!autoneg) {
+ val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
+ val &= ~MVNETA_GMAC_FORCE_LINK_PASS;
+ val |= MVNETA_GMAC_FORCE_LINK_DOWN;
+ mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val);
}
+ mvneta_port_down(pp);
+}
+
+static void mvneta_mac_link_up(struct net_device *ndev, bool autoneg)
+{
+ struct mvneta_port *pp = netdev_priv(ndev);
+ u32 val;
+
+ if (!autoneg) {
+ val = mvreg_read(pp, MVNETA_GMAC_AUTONEG_CONFIG);
+ val &= ~MVNETA_GMAC_FORCE_LINK_DOWN;
+ val |= MVNETA_GMAC_FORCE_LINK_PASS;
+ mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG, val);
+ }
+
+ mvneta_port_up(pp);
+}
+
+static void mvneta_adjust_link(struct net_device *ndev)
+{
+ struct mvneta_port *pp = netdev_priv(ndev);
+ struct phy_device *phydev = ndev->phydev;
+ int status_change = 0;
+
+ if (phydev->link)
+ mvneta_mac_config(ndev);
+
if (phydev->link != pp->link) {
if (!phydev->link) {
pp->duplex = -1;
@@ -3293,27 +3329,10 @@ static void mvneta_adjust_link(struct net_device *ndev)
}
if (status_change) {
- if (phydev->link) {
- if (!pp->use_inband_status) {
- u32 val = mvreg_read(pp,
- MVNETA_GMAC_AUTONEG_CONFIG);
- val &= ~MVNETA_GMAC_FORCE_LINK_DOWN;
- val |= MVNETA_GMAC_FORCE_LINK_PASS;
- mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG,
- val);
- }
- mvneta_port_up(pp);
- } else {
- if (!pp->use_inband_status) {
- u32 val = mvreg_read(pp,
- MVNETA_GMAC_AUTONEG_CONFIG);
- val &= ~MVNETA_GMAC_FORCE_LINK_PASS;
- val |= MVNETA_GMAC_FORCE_LINK_DOWN;
- mvreg_write(pp, MVNETA_GMAC_AUTONEG_CONFIG,
- val);
- }
- mvneta_port_down(pp);
- }
+ if (phydev->link)
+ mvneta_mac_link_down(ndev, pp->use_inband_status);
+ else
+ mvneta_mac_link_up(ndev, pp->use_inband_status);
phy_print_status(phydev);
}
}
--
2.7.4
^ permalink raw reply related
* [PATCH 01/10] net: mvneta: ensure PM paths take the rtnl lock
From: Russell King @ 2018-01-02 17:24 UTC (permalink / raw)
To: Andrew Lunn, Florian Fainelli, Thomas Petazzoni; +Cc: netdev
In-Reply-To: <20180102172242.GE28752@n2100.armlinux.org.uk>
The netdev core always ensures that the rtnl lock is held while calling
the ndo_open() and ndo_stop() methods. However, the suspend/resume paths
do not hold the rtnl lock. phylink will expect the rtnl lock to be held
when the MAC driver calls it, so we end up with kernel warnings. Take
the lock to ensure that these functions are called in a consistent
manner.
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
---
drivers/net/ethernet/marvell/mvneta.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index a539263cd79c..e08bd20352cb 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -4426,8 +4426,10 @@ static int mvneta_suspend(struct device *device)
struct net_device *dev = dev_get_drvdata(device);
struct mvneta_port *pp = netdev_priv(dev);
+ rtnl_lock();
if (netif_running(dev))
mvneta_stop(dev);
+ rtnl_unlock();
netif_device_detach(dev);
clk_disable_unprepare(pp->clk_bus);
clk_disable_unprepare(pp->clk);
@@ -4464,10 +4466,12 @@ static int mvneta_resume(struct device *device)
mvneta_fixed_link_update(pp, dev->phydev);
netif_device_attach(dev);
+ rtnl_lock();
if (netif_running(dev)) {
mvneta_open(dev);
mvneta_set_rx_mode(dev);
}
+ rtnl_unlock();
return 0;
}
--
2.7.4
^ permalink raw reply related
* Re: general protection fault in skb_segment
From: Willem de Bruijn @ 2018-01-02 17:23 UTC (permalink / raw)
To: Marcelo Ricardo Leitner
Cc: syzbot, David Miller, LKML, linux-sctp, Network Development,
Neil Horman, syzkaller-bugs, Vladislav Yasevich
In-Reply-To: <CAF=yD-K9M1yps+SogVLSVgoi+ngKzKrwWH+xMr4-m+v-pqyXXQ@mail.gmail.com>
> Actually, changes just to inet_gso_segment and ipv6_gso_segment
> will suffice:
>
> bool udpfrag = false, fixedid = false, gso_partial, encap;
> struct sk_buff *segs = ERR_PTR(-EINVAL);
> + unsigned int offset = 0, gso_type;
> const struct net_offload *ops;
> - unsigned int offset = 0;
> struct iphdr *iph;
> int proto, tot_len;
> int nhoff;
> @@ -1258,6 +1258,22 @@ struct sk_buff *inet_gso_segment(struct sk_buff *skb,
>
> skb_reset_transport_header(skb);
>
> + gso_type = skb_shinfo(skb)->gso_type;
> + if (gso_type & SKB_GSO_DODGY) {
> + switch (gso_type & (SKB_GSO_TCPV4 | SKB_GSO_UDP)) {
> + case SKB_GSO_TCPV4:
> + if (proto != IPPROTO_TCP)
> + goto out;
> + break;
> + case SKB_GSO_UDP:
> + if (proto != IPPROTO_UDP)
> + goto out;
> + break;
> + default:
> + goto out;
> + }
> + }
>
> and analogous for IPv6. For a real patch I would deduplicate this
> logic between them and move it to a separate helper function
> (in a header file, then).
This approach would also need an skb->protocol check either in
virtio_net_hdr_to_skb or skb_mac_gso_segment.
^ permalink raw reply
* [PATCH 00/10] Convert mvneta to phylink
From: Russell King - ARM Linux @ 2018-01-02 17:22 UTC (permalink / raw)
To: Andrew Lunn, Florian Fainelli, Thomas Petazzoni; +Cc: netdev
Hi,
This series converts mvneta to use phylink, which is necessary to
support the SFP cages on SolidRun's Clearfog platform. This series just
converts mvneta without adding the DT parts - having discussed with
Andrew, we believe we're too close to the merge window to submit that
patch.
I've split the "net: mvneta: convert to phylink" patch up to make it
easier to review, and in doing so, spotted some minor corner cases that
needed to be fixed along the way.
This series depends on the previously merged phylink patches in netdev,
along with the recently reviewed 7 patch series "Resolve races in phy
accessors" without which, the race described in patch 5 of that series
is very evident when triggering a dummy hibernate cycle.
This series also illustrates how to convert mvpp2 to phylink.
mvneta is the only user of the fixed_phy_update_state() API, and this
becomes redundant with the conversion.
It would be good to get this series not only reviewed, but also
independently tested to ensure that I haven't missed anything - I only
have the Clearfog platform to test on, and that doesn't support all the
different interface modes that mvneta supports.
A particularly interesting side effect of this series is that DSA
switches no longer need the "CPU" port and DSA facing MAC ethernet
instance to be marked as a fixed link anymore with mvneta - we can use
1000BaseX mode, and the DSA to CPU link will use the 802.3z negotiation
to determine the link properties without needing the link parameters to
be explicitly stated in DT - that is a subject of a future patch.
drivers/net/ethernet/marvell/Kconfig | 2 +-
drivers/net/ethernet/marvell/mvneta.c | 687 ++++++++++++++++++++--------------
drivers/net/phy/fixed_phy.c | 31 --
include/linux/phy_fixed.h | 9 -
4 files changed, 405 insertions(+), 324 deletions(-)
--
RMK's Patch system: http://www.armlinux.org.uk/developer/patches/
FTTC broadband for 0.8mile line in suburbia: sync at 8.8Mbps down 630kbps up
According to speedtest.net: 8.21Mbps down 510kbps up
^ permalink raw reply
* Re: [net-next: PATCH v2 5/5] net: mvpp2: enable ACPI support in the driver
From: Florian Fainelli @ 2018-01-02 17:22 UTC (permalink / raw)
To: Marcin Wojtas, Andrew Lunn
Cc: linux-kernel, linux-arm-kernel, netdev, linux-acpi,
Graeme Gregory, David S. Miller, Russell King - ARM Linux,
Rafael J. Wysocki, Antoine Ténart, Thomas Petazzoni,
Gregory Clément, Ezequiel Garcia, nadavh,
Neta Zur Hershkovits, Ard Biesheuvel, Grzegorz Jaszczyk,
Tomasz Nowicki
In-Reply-To: <CAPv3WKdD0HwdFVOBx-yeTCfcMF7DXxxFaUt+Zm06oxuVuty8Rg@mail.gmail.com>
On January 2, 2018 7:05:35 AM PST, Marcin Wojtas <mw@semihalf.com> wrote:
>2018-01-02 15:08 GMT+01:00 Andrew Lunn <andrew@lunn.ch>:
>>> Indeed in of_mdio_bus_register_phy, there is of_irq_get. This is
>more
>>> a discussion for a MDIO bus / ACPI patchset, but we either find a
>way
>>> to use IRQs with ACPI obtained from child nodes or for this world
>the
>>> functionality will be limited (at least for the beginning).
>>
>> Hi Marcin
>>
>> What i want to avoid is adding something which partially works, and
>> then have to throw it all away and start again in order to add full
>> support.
>>
>> If ACPI really limits interrupts to devices, maybe we need a totally
>> different representation of MDIO and PHYs in ACPI to what it used in
>> device tree? The same may be true for the Ethernet ports of the
>mvpp2?
>> They might have to be represented as real devices, not children of a
>> device? Maybe trying to map DT to ACPI on a one-to-one basis is the
>> wrong approach?
>>
>
>In terms of PP2 controller, I'd prefer to keep as much as possible to
>describing how real hardware looks like, i.e. single common controller
>with multiple ports as its children. Those considerations are
>reflected in the DT description shape and how the driver enumerates,
>which was part of the design of the initial support. Bending the
>driver (huge amount of shared initialization and resources) to
>multiple instances just for the sake of possible avoidance of IRQ
>description in ACPI is IMO a huge and unnecessary overkill.
True, although keep in mind that Device Tree, as implemented in Linux allows for a lot of flexibility in how parent/child nodes are represented and backed or not by a corresponding struct device. I suspect ACPI is much less permissive than that and we might want to have a struct device for the whole mvpp2 controller as well as for the child ports within that controller (something you could today with device tree too, just it would be more overhead). This does not necessarily have to influence the representation within the description language but we should not be biased by how the current implementation using Device Tree has shaped both representation and implementation.
--
Florian
^ 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