* Re: [PATCH net-next-2.6 v5 5/5] sctp: Add ADD/DEL ASCONF handling at the receiver
From: Wei Yongjun @ 2011-04-22 4:05 UTC (permalink / raw)
To: Michio Honda; +Cc: netdev, lksctp-developers
In-Reply-To: <8F7C47DA-A0A2-48BE-84A7-1D073B5617AE@sfc.wide.ad.jp>
> This patch fixes the problem that the original code cannot delete the remote address where the corresponding transport is currently directed, even when the ASCONF is sent from the other address (this situation happens when the single-homed sender transmits ASCONF with ADD and DEL.)
we should apply this patch first before patch 4, so patch 5 and
patch 4 should change the order in the patchset.
> Signed-off-by: Michio Honda <micchie@sfc.wide.ad.jp>
> ---
> diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
> index de98665..a9f25d7 100644
> --- a/net/sctp/sm_make_chunk.c
> +++ b/net/sctp/sm_make_chunk.c
> @@ -2990,7 +2990,7 @@ static __be16 sctp_process_asconf_param(struct sctp_association *asoc,
> * an Error Cause TLV set to the new error code 'Request to
> * Delete Source IP Address'
> */
> - if (sctp_cmp_addr_exact(sctp_source(asconf), &addr))
> + if (sctp_cmp_addr_exact(&asconf->source, &addr))
> return SCTP_ERROR_DEL_SRC_IP;
>
> /* Section 4.2.2
>
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
^ permalink raw reply
* Re: [PATCH net-next-2.6 v4 4/5] sctp: Add ASCONF operation on the single-homed host
From: Wei Yongjun @ 2011-04-22 4:10 UTC (permalink / raw)
To: Michio Honda; +Cc: netdev, lksctp-developers
In-Reply-To: <4DB0FD45.2050709@cn.fujitsu.com>
>
> Since the sender MUST NOT use the new IP address as a source for ANY SCTP
> packet except on carrying an ASCONF Chunk. And ASCONF chunk can be bundled.
> How about this change. If so, you do not need change to sctp_outq_tail();
>
> diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
> index 1c88c89..bd6cc9c 100644
> --- a/net/sctp/outqueue.c
> +++ b/net/sctp/outqueue.c
> @@ -754,6 +754,13 @@ static int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
> */
>
> list_for_each_entry_safe(chunk, tmp, &q->control_chunk_list, list) {
> + /* RFC 5061, 5.3
> + * F1) This ...
> + */
> + if (q->asoc->src_out_of_asoc_ok &&
> + chunk->chunk_hdr->type != SCTP_CID_ASCONF)
SCTP_CID_ASCONF_ACK should be also allowed, the peer may
send ASCONF to do the same thing at the same time.
> + continue;
> +
> list_del_init(&chunk->list);
>
> /* Pick the right transport to use. */
> @@ -881,6 +888,9 @@ static int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout)
> }
> }
>
> + if (q->asoc->src_out_of_asoc_ok)
> + goto sctp_flush_out;
> +
> /* Is it OK to send data chunks? */
> switch (asoc->state) {
> case SCTP_STATE_COOKIE_ECHOED:
>
>
>
^ permalink raw reply
* Re: [PATCH net-next-2.6 3/3] bonding,ipv4,ipv6,vlan: Handle NETDEV_BONDING_FAILOVER like NETDEV_NOTIFY_PEERS
From: David Miller @ 2011-04-22 4:13 UTC (permalink / raw)
To: fubar; +Cc: bhutchings, andy, kaber, netdev, brian.haley
In-Reply-To: <26398.1303240321@death>
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Tue, 19 Apr 2011 12:12:01 -0700
> Ben Hutchings <bhutchings@solarflare.com> wrote:
>
>>Why would we activate a slave without link up? Perhaps if the previous
>>active slave is removed?
>
> It's special sauce for Infiniband; I don't recall the details
> except that the submitter said that without it the initial gratuitous
> ARP could be lost. I didn't (and still don't) have IB hardware to test
> this on.
I vaguely remember this IB has too, but also forget the details.
If someone would get to the bottom of this and add a nice big comment
to the code, to prevent such difficulties in remembering exactly why we
do this in the future, I would very much appreciate it.
^ permalink raw reply
* Re: [PATCH] bonding: fix bridged bonds in 802.3ad mode
From: David Miller @ 2011-04-22 4:19 UTC (permalink / raw)
To: benjamin.poirier; +Cc: bhutchings, jbohac, netdev, shemminger, fubar, andy
In-Reply-To: <4DB096E0.4010407@polymtl.ca>
From: Benjamin Poirier <benjamin.poirier@polymtl.ca>
Date: Thu, 21 Apr 2011 16:43:12 -0400
> On 21/04/11 03:08 PM, Ben Hutchings wrote:
>> It seems to me that 1e253c3b8a1aeed51eef6fc366812f219b97de65 is bogus
>
> You bet, it's rubbish ;)
I've reverted it from net-2.6 and will send the revert to -stable
too.
Thanks.
^ permalink raw reply
* Re: [PATCH v5] net: bnx2x: convert to hw_features
From: Eric Dumazet @ 2011-04-22 4:51 UTC (permalink / raw)
To: Michał Mirosław; +Cc: netdev, Vladislav Zolotarov, Eilon Greenstein
In-Reply-To: <20110421224134.GB7888@rere.qmqm.pl>
Le vendredi 22 avril 2011 à 00:41 +0200, Michał Mirosław a écrit :
> On Thu, Apr 21, 2011 at 04:52:11PM +0200, Eric Dumazet wrote:
> > Le mardi 12 avril 2011 à 21:38 +0200, Michał Mirosław a écrit :
> > > Since ndo_fix_features callback is postponing features change when
> > > bp->recovery_state != BNX2X_RECOVERY_DONE, netdev_update_features()
> > > has to be called again when this condition changes. Previously,
> > > ethtool_ops->set_flags callback returned -EBUSY in that case
> > > (it's not possible in the new model).
> > >
> > > Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
> > >
> > > v5: - don't delay set_features, as it's rtnl_locked - same as recovery process
> > > v4: - complete bp->rx_csum -> NETIF_F_RXCSUM conversion
> > > - add check for failed ndo_set_features in ndo_open callback
> > > v3: - include NETIF_F_LRO in hw_features
> > > - don't call netdev_update_features() if bnx2x_nic_load() failed
> > > v2: - comment in ndo_fix_features callback
> > > ---
> > I am not sure its related to these changes, but I now have in
> > net-next-2.6 :
>
> > [ 23.674263] ------------[ cut here ]------------
> > [ 23.674266] WARNING: at net/core/dev.c:1318 dev_disable_lro+0x83/0x90()
> > [ 23.674270] Hardware name: ProLiant BL460c G6
> > [ 23.674273] Modules linked in: tg3 libphy sg
> > [ 23.674280] Pid: 3070, comm: sysctl Tainted: G W 2.6.39-rc2-01242-g3ef22b9-dirty #669
> > [ 23.674282] Call Trace:
> > [ 23.674285] [<ffffffff813b94f3>] ? dev_disable_lro+0x83/0x90
> > [ 23.674291] [<ffffffff81042c9b>] warn_slowpath_common+0x8b/0xc0
> > [ 23.674298] [<ffffffff81042ce5>] warn_slowpath_null+0x15/0x20
> > [ 23.674304] [<ffffffff813b94f3>] dev_disable_lro+0x83/0x90
> > [ 23.674309] [<ffffffff81429789>] devinet_sysctl_forward+0x199/0x210
> [...]
>
> Hmm. Looks like something is not allowing to disable LRO. Please check with
> following patch so we can be sure which driver causes this.
>
> Best Regards,
> Michał Mirosław
Yes, obviously, and I suggested roughly the same patch some time ago.
^ permalink raw reply
* Re: [PATCH] net: make WARN_ON in dev_disable_lro() useful
From: Eric Dumazet @ 2011-04-22 5:11 UTC (permalink / raw)
To: Michał Mirosław; +Cc: netdev
In-Reply-To: <20110421224215.9B8AC13909@rere.qmqm.pl>
Le vendredi 22 avril 2011 à 00:42 +0200, Michał Mirosław a écrit :
> Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
> ---
> net/core/dev.c | 3 ++-
> 1 files changed, 2 insertions(+), 1 deletions(-)
>
> diff --git a/net/core/dev.c b/net/core/dev.c
> index 3871bf6..3421184 100644
> --- a/net/core/dev.c
> +++ b/net/core/dev.c
> @@ -1315,7 +1315,8 @@ void dev_disable_lro(struct net_device *dev)
> return;
>
> __ethtool_set_flags(dev, flags & ~ETH_FLAG_LRO);
> - WARN_ON(dev->features & NETIF_F_LRO);
> + if (unlikely(dev->features & NETIF_F_LRO))
> + netdev_WARN(dev, "failed to disable LRO!\n");
> }
> EXPORT_SYMBOL(dev_disable_lro);
>
Yes, as suggested one month ago
http://permalink.gmane.org/gmane.linux.network/189951
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
BTW, using unlikely() or likely() in slowpath is not exactly useful,
unless its integrated in macro (WARN_... )
^ permalink raw reply
* Re: [PATCH v4] net: fix hw_features ethtool_ops->set_flags compatibility
From: Eric Dumazet @ 2011-04-22 5:16 UTC (permalink / raw)
To: Michał Mirosław
Cc: netdev, Ben Hutchings, Vladislav Zolotarov, Eilon Greenstein
In-Reply-To: <20110421235921.89D1D13909@rere.qmqm.pl>
Le vendredi 22 avril 2011 à 01:59 +0200, Michał Mirosław a écrit :
> __ethtool_set_flags() was not taking into account features set but not
> user-toggleable.
>
> Since GFLAGS returns masked dev->features, EINVAL is returned when
> passed flags differ to it, and not to wanted_features.
>
> Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
> ---
> This should be the simpler way Ben was talking about.
>
> net/core/ethtool.c | 4 ++--
> 1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/net/core/ethtool.c b/net/core/ethtool.c
> index 13d79f5..d8b1a8d 100644
> --- a/net/core/ethtool.c
> +++ b/net/core/ethtool.c
> @@ -546,12 +546,12 @@ int __ethtool_set_flags(struct net_device *dev, u32 data)
> }
>
> /* allow changing only bits set in hw_features */
> - changed = (data ^ dev->wanted_features) & flags_dup_features;
> + changed = (data ^ dev->features) & flags_dup_features;
> if (changed & ~dev->hw_features)
> return (changed & dev->hw_features) ? -EINVAL : -EOPNOTSUPP;
>
> dev->wanted_features =
> - (dev->wanted_features & ~changed) | data;
> + (dev->wanted_features & ~changed) | (data & dev->hw_features);
>
> __netdev_update_features(dev);
>
Reported-by: Eric Dumazet <eric.dumazet@gmail.com>
Tested-by: Eric Dumazet <eric.dumazet@gmail.com>
Thanks, this solves the bnx2x problem !
^ permalink raw reply
* Re: [PATCH v4] net: fix hw_features ethtool_ops->set_flags compatibility
From: Eric Dumazet @ 2011-04-22 5:27 UTC (permalink / raw)
To: Michał Mirosław
Cc: netdev, Ben Hutchings, Vladislav Zolotarov, Eilon Greenstein
In-Reply-To: <1303449402.13893.6.camel@edumazet-laptop>
Le vendredi 22 avril 2011 à 07:16 +0200, Eric Dumazet a écrit :
> Reported-by: Eric Dumazet <eric.dumazet@gmail.com>
> Tested-by: Eric Dumazet <eric.dumazet@gmail.com>
>
> Thanks, this solves the bnx2x problem !
>
And already in net-next-2.6 ;)
^ permalink raw reply
* Re: [PATCH] netfilter/IPv6: fix DSCP mangle code
From: David Miller @ 2011-04-22 7:08 UTC (permalink / raw)
To: fernando; +Cc: netdev, yoshfuji, jengelh, netfilter-devel
In-Reply-To: <1303455759.2023.39.camel@nausicaa>
From: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
Date: Fri, 22 Apr 2011 16:02:39 +0900
> The mask indicates the bits one wants to zero out, so it needs to be
> inverted before applying to the original TOS field.
>
> Signed-off-by: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
> ---
Netfilter patches should be sent to the netfilter developer
list, CC:'d
> diff -urNp linux-2.6.37-orig/net/netfilter/xt_DSCP.c linux-2.6.37/net/netfilter/xt_DSCP.c
> --- linux-2.6.37-orig/net/netfilter/xt_DSCP.c 2011-01-05 09:50:19.000000000 +0900
> +++ linux-2.6.37/net/netfilter/xt_DSCP.c 2011-04-21 16:01:25.801890733 +0900
> @@ -99,7 +99,7 @@ tos_tg6(struct sk_buff *skb, const struc
> u_int8_t orig, nv;
>
> orig = ipv6_get_dsfield(iph);
> - nv = (orig & info->tos_mask) ^ info->tos_value;
> + nv = (orig & ~info->tos_mask) ^ info->tos_value;
>
> if (orig != nv) {
> if (!skb_make_writable(skb, sizeof(struct iphdr)))
>
>
^ permalink raw reply
* [PATCH] netfilter/IPv6: initialize TOS field in REJECT target module
From: Fernando Luis Vazquez Cao @ 2011-04-22 7:19 UTC (permalink / raw)
To: netfilter-devel; +Cc: netdev, yoshfuji, jengelh, davem
In-Reply-To: <1303455759.2023.39.camel@nausicaa>
The IPv6 header is not zeroed out in alloc_skb so we must initialize it
properly unless we want to see IPv6 packets with random TOS fields
floating around. The current implementation resets the flow label
but this could be changed if deemed necessary.
We stumbled upon this issue when trying to apply a mangle rule to
the RST packet generated by the REJECT target module.
Signed-off-by: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
---
diff -urNp linux-2.6.38.4-orig/net/ipv6/netfilter/ip6t_REJECT.c linux-2.6.38.4/net/ipv6/netfilter/ip6t_REJECT.c
--- linux-2.6.38.4-orig/net/ipv6/netfilter/ip6t_REJECT.c 2011-03-15 10:20:32.000000000 +0900
+++ linux-2.6.38.4/net/ipv6/netfilter/ip6t_REJECT.c 2011-04-22 15:00:30.695373543 +0900
@@ -45,6 +45,8 @@ static void send_reset(struct net *net,
int tcphoff, needs_ack;
const struct ipv6hdr *oip6h = ipv6_hdr(oldskb);
struct ipv6hdr *ip6h;
+#define DEFAULT_TOS_VALUE 0x0U
+ const __u8 tclass = DEFAULT_TOS_VALUE;
struct dst_entry *dst = NULL;
u8 proto;
struct flowi fl;
@@ -123,7 +125,7 @@ static void send_reset(struct net *net,
skb_put(nskb, sizeof(struct ipv6hdr));
skb_reset_network_header(nskb);
ip6h = ipv6_hdr(nskb);
- ip6h->version = 6;
+ *(__be32 *)ip6h = htonl(0x60000000 | (tclass << 20));
ip6h->hop_limit = ip6_dst_hoplimit(dst);
ip6h->nexthdr = IPPROTO_TCP;
ipv6_addr_copy(&ip6h->saddr, &oip6h->daddr);
^ permalink raw reply
* [PATCH] netfilter/IPv6: fix DSCP mangle code
From: Fernando Luis Vazquez Cao @ 2011-04-22 7:02 UTC (permalink / raw)
To: netdev; +Cc: davem, yoshfuji, jengelh
The mask indicates the bits one wants to zero out, so it needs to be
inverted before applying to the original TOS field.
Signed-off-by: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
---
diff -urNp linux-2.6.37-orig/net/netfilter/xt_DSCP.c linux-2.6.37/net/netfilter/xt_DSCP.c
--- linux-2.6.37-orig/net/netfilter/xt_DSCP.c 2011-01-05 09:50:19.000000000 +0900
+++ linux-2.6.37/net/netfilter/xt_DSCP.c 2011-04-21 16:01:25.801890733 +0900
@@ -99,7 +99,7 @@ tos_tg6(struct sk_buff *skb, const struc
u_int8_t orig, nv;
orig = ipv6_get_dsfield(iph);
- nv = (orig & info->tos_mask) ^ info->tos_value;
+ nv = (orig & ~info->tos_mask) ^ info->tos_value;
if (orig != nv) {
if (!skb_make_writable(skb, sizeof(struct iphdr)))
^ permalink raw reply
* Re: [PATCH] netfilter/IPv6: fix DSCP mangle code
From: Fernando Luis Vazquez Cao @ 2011-04-22 7:29 UTC (permalink / raw)
To: David Miller; +Cc: netdev, yoshfuji, jengelh, netfilter-devel
In-Reply-To: <20110422.000838.193727308.davem@davemloft.net>
On Fri, 2011-04-22 at 00:08 -0700, David Miller wrote:
> From: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
> Date: Fri, 22 Apr 2011 16:02:39 +0900
>
> > The mask indicates the bits one wants to zero out, so it needs to be
> > inverted before applying to the original TOS field.
> >
> > Signed-off-by: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
> > ---
>
> Netfilter patches should be sent to the netfilter developer
> list, CC:'d
Thank you for the heads-up, David.
By the way, I have just sent one more patch that fixes what I think is
another netfilter bug. If everyone is ok with these two patches it would
be great if we could get them merged for the next -rc release (we hit
these bugs in our production systems).
Regards,
Fernando
^ permalink raw reply
* Re: [PATCH] netfilter/IPv6: initialize TOS field in REJECT target module
From: Eric Dumazet @ 2011-04-22 7:34 UTC (permalink / raw)
To: Fernando Luis Vazquez Cao
Cc: netfilter-devel, netdev, yoshfuji, jengelh, davem
In-Reply-To: <1303456743.2023.55.camel@nausicaa>
Le vendredi 22 avril 2011 à 16:19 +0900, Fernando Luis Vazquez Cao a
écrit :
> The IPv6 header is not zeroed out in alloc_skb so we must initialize it
> properly unless we want to see IPv6 packets with random TOS fields
> floating around. The current implementation resets the flow label
> but this could be changed if deemed necessary.
>
> We stumbled upon this issue when trying to apply a mangle rule to
> the RST packet generated by the REJECT target module.
>
> Signed-off-by: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
> ---
>
> diff -urNp linux-2.6.38.4-orig/net/ipv6/netfilter/ip6t_REJECT.c linux-2.6.38.4/net/ipv6/netfilter/ip6t_REJECT.c
> --- linux-2.6.38.4-orig/net/ipv6/netfilter/ip6t_REJECT.c 2011-03-15 10:20:32.000000000 +0900
> +++ linux-2.6.38.4/net/ipv6/netfilter/ip6t_REJECT.c 2011-04-22 15:00:30.695373543 +0900
> @@ -45,6 +45,8 @@ static void send_reset(struct net *net,
> int tcphoff, needs_ack;
> const struct ipv6hdr *oip6h = ipv6_hdr(oldskb);
> struct ipv6hdr *ip6h;
> +#define DEFAULT_TOS_VALUE 0x0U
> + const __u8 tclass = DEFAULT_TOS_VALUE;
> struct dst_entry *dst = NULL;
> u8 proto;
> struct flowi fl;
> @@ -123,7 +125,7 @@ static void send_reset(struct net *net,
> skb_put(nskb, sizeof(struct ipv6hdr));
> skb_reset_network_header(nskb);
> ip6h = ipv6_hdr(nskb);
> - ip6h->version = 6;
> + *(__be32 *)ip6h = htonl(0x60000000 | (tclass << 20));
It seems fine but I am wondering where is payload_len set ?
^ permalink raw reply
* Re: [PATCH] netfilter/IPv6: initialize TOS field in REJECT target module
From: Fernando Luis Vazquez Cao @ 2011-04-22 7:47 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netfilter-devel, netdev, yoshfuji, jengelh, davem
In-Reply-To: <1303457656.3134.12.camel@edumazet-laptop>
On Fri, 2011-04-22 at 09:34 +0200, Eric Dumazet wrote:
> Le vendredi 22 avril 2011 à 16:19 +0900, Fernando Luis Vazquez Cao a
> écrit :
> > The IPv6 header is not zeroed out in alloc_skb so we must initialize it
> > properly unless we want to see IPv6 packets with random TOS fields
> > floating around. The current implementation resets the flow label
> > but this could be changed if deemed necessary.
> >
> > We stumbled upon this issue when trying to apply a mangle rule to
> > the RST packet generated by the REJECT target module.
> >
> > Signed-off-by: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
> > ---
> >
> > diff -urNp linux-2.6.38.4-orig/net/ipv6/netfilter/ip6t_REJECT.c linux-2.6.38.4/net/ipv6/netfilter/ip6t_REJECT.c
> > --- linux-2.6.38.4-orig/net/ipv6/netfilter/ip6t_REJECT.c 2011-03-15 10:20:32.000000000 +0900
> > +++ linux-2.6.38.4/net/ipv6/netfilter/ip6t_REJECT.c 2011-04-22 15:00:30.695373543 +0900
> > @@ -45,6 +45,8 @@ static void send_reset(struct net *net,
> > int tcphoff, needs_ack;
> > const struct ipv6hdr *oip6h = ipv6_hdr(oldskb);
> > struct ipv6hdr *ip6h;
> > +#define DEFAULT_TOS_VALUE 0x0U
> > + const __u8 tclass = DEFAULT_TOS_VALUE;
> > struct dst_entry *dst = NULL;
> > u8 proto;
> > struct flowi fl;
> > @@ -123,7 +125,7 @@ static void send_reset(struct net *net,
> > skb_put(nskb, sizeof(struct ipv6hdr));
> > skb_reset_network_header(nskb);
> > ip6h = ipv6_hdr(nskb);
> > - ip6h->version = 6;
> > + *(__be32 *)ip6h = htonl(0x60000000 | (tclass << 20));
>
> It seems fine but I am wondering where is payload_len set ?
Hi Eric,
Thank you for the review, I really appreciate it.
payload_len is set in __ip6_local_out() which is called from
ip6_local_out.
Could I get your acked-by for the two patches I sent?
Thanks,
Fernando
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH] netfilter/IPv6: initialize TOS field in REJECT target module
From: Eric Dumazet @ 2011-04-22 8:06 UTC (permalink / raw)
To: Fernando Luis Vazquez Cao
Cc: netfilter-devel, netdev, yoshfuji, jengelh, davem
In-Reply-To: <1303458454.9791.13.camel@nausicaa>
Le vendredi 22 avril 2011 à 16:47 +0900, Fernando Luis Vazquez Cao a
écrit :
> Hi Eric,
>
> Thank you for the review, I really appreciate it.
>
> payload_len is set in __ip6_local_out() which is called from
> ip6_local_out.
>
Thats right, thanks !
> Could I get your acked-by for the two patches I sent?
>
Sure
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH] netfilter/IPv6: initialize TOS field in REJECT target module
From: Fernando Luis Vazquez Cao @ 2011-04-22 8:11 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netfilter-devel, netdev, yoshfuji, jengelh, davem
In-Reply-To: <1303459560.3134.14.camel@edumazet-laptop>
On Fri, 2011-04-22 at 10:06 +0200, Eric Dumazet wrote:
> Le vendredi 22 avril 2011 à 16:47 +0900, Fernando Luis Vazquez Cao a
> écrit :
> > Could I get your acked-by for the two patches I sent?
> >
>
> Sure
>
> Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Thank you!
Should we send these two patches to -stable too?
^ permalink raw reply
* Re: [PATCH] netfilter/IPv6: initialize TOS field in REJECT target module
From: Eric Dumazet @ 2011-04-22 8:37 UTC (permalink / raw)
To: Fernando Luis Vazquez Cao
Cc: netfilter-devel, netdev, yoshfuji, jengelh, davem
In-Reply-To: <1303459875.9791.19.camel@nausicaa>
Le vendredi 22 avril 2011 à 17:11 +0900, Fernando Luis Vazquez Cao a
écrit :
> Thank you!
>
> Should we send these two patches to -stable too?
>
David takes care of stable submissions for netdev stuff, thanks.
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH V14 4/4] ptp: Added a clock driver for the National Semiconductor PHYTER.
From: Richard Cochran @ 2011-04-22 9:21 UTC (permalink / raw)
To: Ben Hutchings
Cc: linux-kernel, linux-api, netdev, devicetree-discuss,
linux-arm-kernel, linuxppc-dev, Alan Cox, Arnd Bergmann,
Christoph Lameter, David Miller, John Stultz, Krzysztof Halasa,
Peter Zijlstra, Rodolfo Giometti, Thomas Gleixner,
Benjamin Herrenschmidt, Mike Frysinger, Paul Mackerras,
Russell King
In-Reply-To: <1303160245.2857.60.camel@bwh-desktop>
On Mon, Apr 18, 2011 at 09:57:25PM +0100, Ben Hutchings wrote:
> Would be better without the magic numbers.
Yes, of course, you are right, and also about patch #3.
Please take a look at v15 and see if it improves things.
Thanks,
Richard
^ permalink raw reply
* Re: [PATCH net-next-2.6] veth: remove unneeded ifname code from veth_newlink()
From: Michał Mirosław @ 2011-04-22 9:43 UTC (permalink / raw)
To: Jiri Pirko; +Cc: netdev, davem, xemul
In-Reply-To: <20110124154546.GA2689@psychotron.brq.redhat.com>
2011/1/24 Jiri Pirko <jpirko@redhat.com>:
> The code is not needed because tb[IFLA_IFNAME] is already
> processed in rtnl_newlink(). Remove this redundancy.
>
> Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Hi. This patch broke creation of veth devices. Reverting it fixes the issue.
Symptoms:
icybox:~# ip link add type veth
RTNETLINK answers: File exists
icybox:~# ip link add type veth peer veth1
icybox:~# ip addr
[...]
56: D: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
link/ether e6:57:58:52:03:50 brd ff:ff:ff:ff:ff:ff
57: veth0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
link/ether c2:c2:a2:d5:d5:3a brd ff:ff:ff:ff:ff:ff
Best Regards,
Michał Mirosław
^ permalink raw reply
* Re: [PATCH net-next-2.6] veth: remove unneeded ifname code from veth_newlink()
From: Michał Mirosław @ 2011-04-22 9:44 UTC (permalink / raw)
To: Jiri Pirko; +Cc: netdev, davem, xemul
In-Reply-To: <BANLkTi=x8xi7-g3SZdva7Ndg0a4EY1NK8g@mail.gmail.com>
W dniu 22 kwietnia 2011 11:43 użytkownik Michał Mirosław
<mirqus@gmail.com> napisał:
> 2011/1/24 Jiri Pirko <jpirko@redhat.com>:
>> The code is not needed because tb[IFLA_IFNAME] is already
>> processed in rtnl_newlink(). Remove this redundancy.
> Hi. This patch broke creation of veth devices. Reverting it fixes the issue.
This is commit 84c49d8c3e4abefb0a41a77b25aa37ebe8d6b743.
Best Regards,
Michał Mirosław
^ permalink raw reply
* Re: [PATCH net-next-2.6] veth: remove unneeded ifname code from veth_newlink()
From: Michał Mirosław @ 2011-04-22 9:49 UTC (permalink / raw)
To: Jiri Pirko; +Cc: netdev, davem, xemul
In-Reply-To: <BANLkTi=x8xi7-g3SZdva7Ndg0a4EY1NK8g@mail.gmail.com>
W dniu 22 kwietnia 2011 11:43 użytkownik Michał Mirosław
<mirqus@gmail.com> napisał:
> 2011/1/24 Jiri Pirko <jpirko@redhat.com>:
>> The code is not needed because tb[IFLA_IFNAME] is already
>> processed in rtnl_newlink(). Remove this redundancy.
> Hi. This patch broke creation of veth devices. Reverting it fixes the issue.
>
> Symptoms:
>
> icybox:~# ip link add type veth
> RTNETLINK answers: File exists
> icybox:~# ip link add type veth peer veth1
> icybox:~# ip addr
> [...]
> 56: D: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
> link/ether e6:57:58:52:03:50 brd ff:ff:ff:ff:ff:ff
> 57: veth0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
> link/ether c2:c2:a2:d5:d5:3a brd ff:ff:ff:ff:ff:ff
Hmm. I was too quick. Reverting it fixes only first problem - "ip link
add type veth" adds new veth pair now, but "ip link add type veth peer
veth1" gives the same result as above.
Best Regards,
Michał Mirosław
^ permalink raw reply
* [PATCH V15 0/4] ptp: IEEE 1588 hardware clock support
From: Richard Cochran @ 2011-04-22 10:02 UTC (permalink / raw)
To: linux-kernel
Cc: linux-api, netdev, devicetree-discuss, linux-arm-kernel,
linuxppc-dev, Alan Cox, Arnd Bergmann, Ben Hutchings,
Benjamin Herrenschmidt, Christoph Lameter, David Miller,
John Stultz, Krzysztof Halasa, Mike Frysinger, Paul Mackerras,
Peter Zijlstra, Rodolfo Giometti, Russell King, Thomas Gleixner
Here is V15 of the PHC patch series. I had been a bit lazy in adding
the acks that have accumulated since V13, and so now I have added them
into the commit message, unless the particular patch changed
substantially since the ack.
* Introduction
The aim of this patch set is to add support for PTP Hardware Clocks
(PHCs) into the Linux kernel. The patch series builds upon the
dynamic posix clock work appearing in kernel version 2.6.39.
Support for obtaining timestamps from a PHC already exists via the
SO_TIMESTAMPING socket option, integrated in kernel version 2.6.30.
This patch set completes the picture by allow user space programs to
adjust the PHC and to control its ancillary features.
* Patches for ptpd on sf.net
https://sourceforge.net/tracker/?group_id=139814&atid=744634
3225599 [PATCH 1/3] Convert to POSIX clock API.
3225603 [PATCH 2/3] Adapted to use the Linux PTP Hardware Clock API.
3225607 [PATCH 3/3] Adapted to use the newer SO_TIMESTAMPING Linux API.
* PHC Patch ChangeLog
** v15
*** general
- allow test program to set system time from PTP HW clock
*** ixp46x and phyter packet matching
- check packet bounds
- handle variable ipv4 header length
- replace magic numbers with macros
* Why all the CCs?
- One driver is for PowerPC, and adds device tree stuff.
- One driver is for the ARM Xscale IXP465.
* Previous Discussions
- [V14] http://lkml.org/lkml/2011/4/18/16
- [V13] http://lkml.org/lkml/2011/3/27/2
- [V12] http://lkml.org/lkml/2011/2/28/53
- [V11] http://lkml.org/lkml/2011/2/23/107
- [V10] http://lkml.org/lkml/2011/1/27/71
- [V9] http://lkml.org/lkml/2011/1/13/65
- [V8] http://lkml.org/lkml/2010/12/31/128
- [V7] http://lkml.org/lkml/2010/12/16/195
- [V6] http://lkml.org/lkml/2010/9/23/310
- [V5] http://lkml.org/lkml/2010/8/16/90
- Thomas Gleixner: Rework of the PTP support series core code
http://lkml.org/lkml/2011/2/1/137
- Dynamic clock devices [RFC]
http://lkml.org/lkml/2010/11/4/290
- POSIX clock tuning syscall with dynamic clock ids
http://lkml.org/lkml/2010/9/3/119
- POSIX clock tuning syscall with static clock ids
http://lkml.org/lkml/2010/8/23/49
- Versions 1-4 appeared on the netdev list.
Richard Cochran (4):
ptp: Added a brand new class driver for ptp clocks.
ptp: Added a clock that uses the eTSEC found on the MPC85xx.
ptp: Added a clock driver for the IXP46x.
ptp: Added a clock driver for the National Semiconductor PHYTER.
Documentation/ABI/testing/sysfs-ptp | 98 ++
.../devicetree/bindings/net/fsl-tsec-phy.txt | 54 +
Documentation/ptp/ptp.txt | 89 ++
Documentation/ptp/testptp.c | 381 +++++++
Documentation/ptp/testptp.mk | 33 +
arch/arm/mach-ixp4xx/include/mach/ixp46x_ts.h | 78 ++
arch/powerpc/boot/dts/mpc8313erdb.dts | 13 +
arch/powerpc/boot/dts/mpc8572ds.dts | 13 +
arch/powerpc/boot/dts/p2020ds.dts | 13 +
arch/powerpc/boot/dts/p2020rdb.dts | 13 +
drivers/Kconfig | 2 +
drivers/Makefile | 1 +
drivers/net/Makefile | 1 +
drivers/net/arm/ixp4xx_eth.c | 195 ++++-
drivers/net/gianfar_ptp.c | 588 +++++++++++
drivers/net/phy/Makefile | 1 +
drivers/net/phy/dp83640.c | 1100 ++++++++++++++++++++
drivers/net/phy/dp83640_reg.h | 267 +++++
drivers/ptp/Kconfig | 75 ++
drivers/ptp/Makefile | 7 +
drivers/ptp/ptp_chardev.c | 159 +++
drivers/ptp/ptp_clock.c | 343 ++++++
drivers/ptp/ptp_ixp46x.c | 332 ++++++
drivers/ptp/ptp_private.h | 92 ++
drivers/ptp/ptp_sysfs.c | 230 ++++
include/linux/Kbuild | 1 +
include/linux/ptp_classify.h | 7 +
include/linux/ptp_clock.h | 84 ++
include/linux/ptp_clock_kernel.h | 139 +++
29 files changed, 4406 insertions(+), 3 deletions(-)
create mode 100644 Documentation/ABI/testing/sysfs-ptp
create mode 100644 Documentation/ptp/ptp.txt
create mode 100644 Documentation/ptp/testptp.c
create mode 100644 Documentation/ptp/testptp.mk
create mode 100644 arch/arm/mach-ixp4xx/include/mach/ixp46x_ts.h
create mode 100644 drivers/net/gianfar_ptp.c
create mode 100644 drivers/net/phy/dp83640.c
create mode 100644 drivers/net/phy/dp83640_reg.h
create mode 100644 drivers/ptp/Kconfig
create mode 100644 drivers/ptp/Makefile
create mode 100644 drivers/ptp/ptp_chardev.c
create mode 100644 drivers/ptp/ptp_clock.c
create mode 100644 drivers/ptp/ptp_ixp46x.c
create mode 100644 drivers/ptp/ptp_private.h
create mode 100644 drivers/ptp/ptp_sysfs.c
create mode 100644 include/linux/ptp_clock.h
create mode 100644 include/linux/ptp_clock_kernel.h
^ permalink raw reply
* [PATCH V15 1/4] ptp: Added a brand new class driver for ptp clocks.
From: Richard Cochran @ 2011-04-22 10:03 UTC (permalink / raw)
To: linux-kernel-u79uwXL29TY76Z2rM5mHXA
Cc: linux-api-u79uwXL29TY76Z2rM5mHXA, netdev-u79uwXL29TY76Z2rM5mHXA,
devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
linuxppc-dev-uLR06cmDAlY/bJ5BZ2RsiQ, Alan Cox, Arnd Bergmann,
Ben Hutchings, Benjamin Herrenschmidt, Christoph Lameter,
David Miller, John Stultz, Krzysztof Halasa, Mike Frysinger,
Paul Mackerras, Peter Zijlstra, Rodolfo Giometti, Russell King,
Thomas Gleixner
In-Reply-To: <cover.1303466102.git.richard.cochran-3mrvs1K0uXizZXS1Dc/lvw@public.gmane.org>
This patch adds an infrastructure for hardware clocks that implement
IEEE 1588, the Precision Time Protocol (PTP). A class driver offers a
registration method to particular hardware clock drivers. Each clock is
presented as a standard POSIX clock.
The ancillary clock features are exposed in two different ways, via
the sysfs and by a character device.
Signed-off-by: Richard Cochran <richard.cochran-3mrvs1K0uXizZXS1Dc/lvw@public.gmane.org>
Acked-by: Arnd Bergmann <arnd-r2nGTMty4D4@public.gmane.org>
Acked-by: David S. Miller <davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
---
Documentation/ABI/testing/sysfs-ptp | 98 +++++++++
Documentation/ptp/ptp.txt | 89 ++++++++
Documentation/ptp/testptp.c | 381 +++++++++++++++++++++++++++++++++++
Documentation/ptp/testptp.mk | 33 +++
drivers/Kconfig | 2 +
drivers/Makefile | 1 +
drivers/ptp/Kconfig | 30 +++
drivers/ptp/Makefile | 6 +
drivers/ptp/ptp_chardev.c | 159 +++++++++++++++
drivers/ptp/ptp_clock.c | 343 +++++++++++++++++++++++++++++++
drivers/ptp/ptp_private.h | 92 +++++++++
drivers/ptp/ptp_sysfs.c | 230 +++++++++++++++++++++
include/linux/Kbuild | 1 +
include/linux/ptp_classify.h | 7 +
include/linux/ptp_clock.h | 84 ++++++++
include/linux/ptp_clock_kernel.h | 139 +++++++++++++
16 files changed, 1695 insertions(+), 0 deletions(-)
create mode 100644 Documentation/ABI/testing/sysfs-ptp
create mode 100644 Documentation/ptp/ptp.txt
create mode 100644 Documentation/ptp/testptp.c
create mode 100644 Documentation/ptp/testptp.mk
create mode 100644 drivers/ptp/Kconfig
create mode 100644 drivers/ptp/Makefile
create mode 100644 drivers/ptp/ptp_chardev.c
create mode 100644 drivers/ptp/ptp_clock.c
create mode 100644 drivers/ptp/ptp_private.h
create mode 100644 drivers/ptp/ptp_sysfs.c
create mode 100644 include/linux/ptp_clock.h
create mode 100644 include/linux/ptp_clock_kernel.h
diff --git a/Documentation/ABI/testing/sysfs-ptp b/Documentation/ABI/testing/sysfs-ptp
new file mode 100644
index 0000000..d40d2b5
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-ptp
@@ -0,0 +1,98 @@
+What: /sys/class/ptp/
+Date: September 2010
+Contact: Richard Cochran <richardcochran-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+Description:
+ This directory contains files and directories
+ providing a standardized interface to the ancillary
+ features of PTP hardware clocks.
+
+What: /sys/class/ptp/ptpN/
+Date: September 2010
+Contact: Richard Cochran <richardcochran-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+Description:
+ This directory contains the attributes of the Nth PTP
+ hardware clock registered into the PTP class driver
+ subsystem.
+
+What: /sys/class/ptp/ptpN/clock_name
+Date: September 2010
+Contact: Richard Cochran <richardcochran-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+Description:
+ This file contains the name of the PTP hardware clock
+ as a human readable string.
+
+What: /sys/class/ptp/ptpN/max_adjustment
+Date: September 2010
+Contact: Richard Cochran <richardcochran-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+Description:
+ This file contains the PTP hardware clock's maximum
+ frequency adjustment value (a positive integer) in
+ parts per billion.
+
+What: /sys/class/ptp/ptpN/n_alarms
+Date: September 2010
+Contact: Richard Cochran <richardcochran-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+Description:
+ This file contains the number of periodic or one shot
+ alarms offer by the PTP hardware clock.
+
+What: /sys/class/ptp/ptpN/n_external_timestamps
+Date: September 2010
+Contact: Richard Cochran <richardcochran-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+Description:
+ This file contains the number of external timestamp
+ channels offered by the PTP hardware clock.
+
+What: /sys/class/ptp/ptpN/n_periodic_outputs
+Date: September 2010
+Contact: Richard Cochran <richardcochran-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+Description:
+ This file contains the number of programmable periodic
+ output channels offered by the PTP hardware clock.
+
+What: /sys/class/ptp/ptpN/pps_avaiable
+Date: September 2010
+Contact: Richard Cochran <richardcochran-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+Description:
+ This file indicates whether the PTP hardware clock
+ supports a Pulse Per Second to the host CPU. Reading
+ "1" means that the PPS is supported, while "0" means
+ not supported.
+
+What: /sys/class/ptp/ptpN/extts_enable
+Date: September 2010
+Contact: Richard Cochran <richardcochran-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+Description:
+ This write-only file enables or disables external
+ timestamps. To enable external timestamps, write the
+ channel index followed by a "1" into the file.
+ To disable external timestamps, write the channel
+ index followed by a "0" into the file.
+
+What: /sys/class/ptp/ptpN/fifo
+Date: September 2010
+Contact: Richard Cochran <richardcochran-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+Description:
+ This file provides timestamps on external events, in
+ the form of three integers: channel index, seconds,
+ and nanoseconds.
+
+What: /sys/class/ptp/ptpN/period
+Date: September 2010
+Contact: Richard Cochran <richardcochran-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+Description:
+ This write-only file enables or disables periodic
+ outputs. To enable a periodic output, write five
+ integers into the file: channel index, start time
+ seconds, start time nanoseconds, period seconds, and
+ period nanoseconds. To disable a periodic output, set
+ all the seconds and nanoseconds values to zero.
+
+What: /sys/class/ptp/ptpN/pps_enable
+Date: September 2010
+Contact: Richard Cochran <richardcochran-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+Description:
+ This write-only file enables or disables delivery of
+ PPS events to the Linux PPS subsystem. To enable PPS
+ events, write a "1" into the file. To disable events,
+ write a "0" into the file.
diff --git a/Documentation/ptp/ptp.txt b/Documentation/ptp/ptp.txt
new file mode 100644
index 0000000..ae8fef8
--- /dev/null
+++ b/Documentation/ptp/ptp.txt
@@ -0,0 +1,89 @@
+
+* PTP hardware clock infrastructure for Linux
+
+ This patch set introduces support for IEEE 1588 PTP clocks in
+ Linux. Together with the SO_TIMESTAMPING socket options, this
+ presents a standardized method for developing PTP user space
+ programs, synchronizing Linux with external clocks, and using the
+ ancillary features of PTP hardware clocks.
+
+ A new class driver exports a kernel interface for specific clock
+ drivers and a user space interface. The infrastructure supports a
+ complete set of PTP hardware clock functionality.
+
+ + Basic clock operations
+ - Set time
+ - Get time
+ - Shift the clock by a given offset atomically
+ - Adjust clock frequency
+
+ + Ancillary clock features
+ - One short or periodic alarms, with signal delivery to user program
+ - Time stamp external events
+ - Period output signals configurable from user space
+ - Synchronization of the Linux system time via the PPS subsystem
+
+** PTP hardware clock kernel API
+
+ A PTP clock driver registers itself with the class driver. The
+ class driver handles all of the dealings with user space. The
+ author of a clock driver need only implement the details of
+ programming the clock hardware. The clock driver notifies the class
+ driver of asynchronous events (alarms and external time stamps) via
+ a simple message passing interface.
+
+ The class driver supports multiple PTP clock drivers. In normal use
+ cases, only one PTP clock is needed. However, for testing and
+ development, it can be useful to have more than one clock in a
+ single system, in order to allow performance comparisons.
+
+** PTP hardware clock user space API
+
+ The class driver also creates a character device for each
+ registered clock. User space can use an open file descriptor from
+ the character device as a POSIX clock id and may call
+ clock_gettime, clock_settime, and clock_adjtime. These calls
+ implement the basic clock operations.
+
+ User space programs may control the clock using standardized
+ ioctls. A program may query, enable, configure, and disable the
+ ancillary clock features. User space can receive time stamped
+ events via blocking read() and poll(). One shot and periodic
+ signals may be configured via the POSIX timer_settime() system
+ call.
+
+** Writing clock drivers
+
+ Clock drivers include include/linux/ptp_clock_kernel.h and register
+ themselves by presenting a 'struct ptp_clock_info' to the
+ registration method. Clock drivers must implement all of the
+ functions in the interface. If a clock does not offer a particular
+ ancillary feature, then the driver should just return -EOPNOTSUPP
+ from those functions.
+
+ Drivers must ensure that all of the methods in interface are
+ reentrant. Since most hardware implementations treat the time value
+ as a 64 bit integer accessed as two 32 bit registers, drivers
+ should use spin_lock_irqsave/spin_unlock_irqrestore to protect
+ against concurrent access. This locking cannot be accomplished in
+ class driver, since the lock may also be needed by the clock
+ driver's interrupt service routine.
+
+** Supported hardware
+
+ + Freescale eTSEC gianfar
+ - 2 Time stamp external triggers, programmable polarity (opt. interrupt)
+ - 2 Alarm registers (optional interrupt)
+ - 3 Periodic signals (optional interrupt)
+
+ + National DP83640
+ - 6 GPIOs programmable as inputs or outputs
+ - 6 GPIOs with dedicated functions (LED/JTAG/clock) can also be
+ used as general inputs or outputs
+ - GPIO inputs can time stamp external triggers
+ - GPIO outputs can produce periodic signals
+ - 1 interrupt pin
+
+ + Intel IXP465
+ - Auxiliary Slave/Master Mode Snapshot (optional interrupt)
+ - Target Time (optional interrupt)
diff --git a/Documentation/ptp/testptp.c b/Documentation/ptp/testptp.c
new file mode 100644
index 0000000..f59ded0
--- /dev/null
+++ b/Documentation/ptp/testptp.c
@@ -0,0 +1,381 @@
+/*
+ * PTP 1588 clock support - User space test program
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+#include <errno.h>
+#include <fcntl.h>
+#include <math.h>
+#include <signal.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/ioctl.h>
+#include <sys/mman.h>
+#include <sys/stat.h>
+#include <sys/time.h>
+#include <sys/timex.h>
+#include <sys/types.h>
+#include <time.h>
+#include <unistd.h>
+
+#include <linux/ptp_clock.h>
+
+#define DEVICE "/dev/ptp0"
+
+#ifndef ADJ_SETOFFSET
+#define ADJ_SETOFFSET 0x0100
+#endif
+
+#ifndef CLOCK_INVALID
+#define CLOCK_INVALID -1
+#endif
+
+/* When glibc offers the syscall, this will go away. */
+#include <sys/syscall.h>
+static int clock_adjtime(clockid_t id, struct timex *tx)
+{
+ return syscall(__NR_clock_adjtime, id, tx);
+}
+
+static clockid_t get_clockid(int fd)
+{
+#define CLOCKFD 3
+#define FD_TO_CLOCKID(fd) ((~(clockid_t) (fd) << 3) | CLOCKFD)
+
+ return FD_TO_CLOCKID(fd);
+}
+
+static void handle_alarm(int s)
+{
+ printf("received signal %d\n", s);
+}
+
+static int install_handler(int signum, void (*handler)(int))
+{
+ struct sigaction action;
+ sigset_t mask;
+
+ /* Unblock the signal. */
+ sigemptyset(&mask);
+ sigaddset(&mask, signum);
+ sigprocmask(SIG_UNBLOCK, &mask, NULL);
+
+ /* Install the signal handler. */
+ action.sa_handler = handler;
+ action.sa_flags = 0;
+ sigemptyset(&action.sa_mask);
+ sigaction(signum, &action, NULL);
+
+ return 0;
+}
+
+static long ppb_to_scaled_ppm(int ppb)
+{
+ /*
+ * The 'freq' field in the 'struct timex' is in parts per
+ * million, but with a 16 bit binary fractional field.
+ * Instead of calculating either one of
+ *
+ * scaled_ppm = (ppb / 1000) << 16 [1]
+ * scaled_ppm = (ppb << 16) / 1000 [2]
+ *
+ * we simply use double precision math, in order to avoid the
+ * truncation in [1] and the possible overflow in [2].
+ */
+ return (long) (ppb * 65.536);
+}
+
+static void usage(char *progname)
+{
+ fprintf(stderr,
+ "usage: %s [options]\n"
+ " -a val request a one-shot alarm after 'val' seconds\n"
+ " -A val request a periodic alarm every 'val' seconds\n"
+ " -c query the ptp clock's capabilities\n"
+ " -d name device to open\n"
+ " -e val read 'val' external time stamp events\n"
+ " -f val adjust the ptp clock frequency by 'val' ppb\n"
+ " -g get the ptp clock time\n"
+ " -h prints this message\n"
+ " -p val enable output with a period of 'val' nanoseconds\n"
+ " -P val enable or disable (val=1|0) the system clock PPS\n"
+ " -s set the ptp clock time from the system time\n"
+ " -S set the system time from the ptp clock time\n"
+ " -t val shift the ptp clock time by 'val' seconds\n",
+ progname);
+}
+
+int main(int argc, char *argv[])
+{
+ struct ptp_clock_caps caps;
+ struct ptp_extts_event event;
+ struct ptp_extts_request extts_request;
+ struct ptp_perout_request perout_request;
+ struct timespec ts;
+ struct timex tx;
+
+ static timer_t timerid;
+ struct itimerspec timeout;
+ struct sigevent sigevent;
+
+ char *progname;
+ int c, cnt, fd;
+
+ char *device = DEVICE;
+ clockid_t clkid;
+ int adjfreq = 0x7fffffff;
+ int adjtime = 0;
+ int capabilities = 0;
+ int extts = 0;
+ int gettime = 0;
+ int oneshot = 0;
+ int periodic = 0;
+ int perout = -1;
+ int pps = -1;
+ int settime = 0;
+
+ progname = strrchr(argv[0], '/');
+ progname = progname ? 1+progname : argv[0];
+ while (EOF != (c = getopt(argc, argv, "a:A:cd:e:f:ghp:P:sSt:v"))) {
+ switch (c) {
+ case 'a':
+ oneshot = atoi(optarg);
+ break;
+ case 'A':
+ periodic = atoi(optarg);
+ break;
+ case 'c':
+ capabilities = 1;
+ break;
+ case 'd':
+ device = optarg;
+ break;
+ case 'e':
+ extts = atoi(optarg);
+ break;
+ case 'f':
+ adjfreq = atoi(optarg);
+ break;
+ case 'g':
+ gettime = 1;
+ break;
+ case 'p':
+ perout = atoi(optarg);
+ break;
+ case 'P':
+ pps = atoi(optarg);
+ break;
+ case 's':
+ settime = 1;
+ break;
+ case 'S':
+ settime = 2;
+ break;
+ case 't':
+ adjtime = atoi(optarg);
+ break;
+ case 'h':
+ usage(progname);
+ return 0;
+ case '?':
+ default:
+ usage(progname);
+ return -1;
+ }
+ }
+
+ fd = open(device, O_RDWR);
+ if (fd < 0) {
+ fprintf(stderr, "opening %s: %s\n", device, strerror(errno));
+ return -1;
+ }
+
+ clkid = get_clockid(fd);
+ if (CLOCK_INVALID == clkid) {
+ fprintf(stderr, "failed to read clock id\n");
+ return -1;
+ }
+
+ if (capabilities) {
+ if (ioctl(fd, PTP_CLOCK_GETCAPS, &caps)) {
+ perror("PTP_CLOCK_GETCAPS");
+ } else {
+ printf("capabilities:\n"
+ " %d maximum frequency adjustment (ppb)\n"
+ " %d programmable alarms\n"
+ " %d external time stamp channels\n"
+ " %d programmable periodic signals\n"
+ " %d pulse per second\n",
+ caps.max_adj,
+ caps.n_alarm,
+ caps.n_ext_ts,
+ caps.n_per_out,
+ caps.pps);
+ }
+ }
+
+ if (0x7fffffff != adjfreq) {
+ memset(&tx, 0, sizeof(tx));
+ tx.modes = ADJ_FREQUENCY;
+ tx.freq = ppb_to_scaled_ppm(adjfreq);
+ if (clock_adjtime(clkid, &tx)) {
+ perror("clock_adjtime");
+ } else {
+ puts("frequency adjustment okay");
+ }
+ }
+
+ if (adjtime) {
+ memset(&tx, 0, sizeof(tx));
+ tx.modes = ADJ_SETOFFSET;
+ tx.time.tv_sec = adjtime;
+ tx.time.tv_usec = 0;
+ if (clock_adjtime(clkid, &tx) < 0) {
+ perror("clock_adjtime");
+ } else {
+ puts("time shift okay");
+ }
+ }
+
+ if (gettime) {
+ if (clock_gettime(clkid, &ts)) {
+ perror("clock_gettime");
+ } else {
+ printf("clock time: %ld.%09ld or %s",
+ ts.tv_sec, ts.tv_nsec, ctime(&ts.tv_sec));
+ }
+ }
+
+ if (settime == 1) {
+ clock_gettime(CLOCK_REALTIME, &ts);
+ if (clock_settime(clkid, &ts)) {
+ perror("clock_settime");
+ } else {
+ puts("set time okay");
+ }
+ }
+
+ if (settime == 2) {
+ clock_gettime(clkid, &ts);
+ if (clock_settime(CLOCK_REALTIME, &ts)) {
+ perror("clock_settime");
+ } else {
+ puts("set time okay");
+ }
+ }
+
+ if (extts) {
+ memset(&extts_request, 0, sizeof(extts_request));
+ extts_request.index = 0;
+ extts_request.flags = PTP_ENABLE_FEATURE;
+ if (ioctl(fd, PTP_EXTTS_REQUEST, &extts_request)) {
+ perror("PTP_EXTTS_REQUEST");
+ extts = 0;
+ } else {
+ puts("external time stamp request okay");
+ }
+ for (; extts; extts--) {
+ cnt = read(fd, &event, sizeof(event));
+ if (cnt != sizeof(event)) {
+ perror("read");
+ break;
+ }
+ printf("event index %u at %lld.%09u\n", event.index,
+ event.t.sec, event.t.nsec);
+ fflush(stdout);
+ }
+ /* Disable the feature again. */
+ extts_request.flags = 0;
+ if (ioctl(fd, PTP_EXTTS_REQUEST, &extts_request)) {
+ perror("PTP_EXTTS_REQUEST");
+ }
+ }
+
+ if (oneshot) {
+ install_handler(SIGALRM, handle_alarm);
+ /* Create a timer. */
+ sigevent.sigev_notify = SIGEV_SIGNAL;
+ sigevent.sigev_signo = SIGALRM;
+ if (timer_create(clkid, &sigevent, &timerid)) {
+ perror("timer_create");
+ return -1;
+ }
+ /* Start the timer. */
+ memset(&timeout, 0, sizeof(timeout));
+ timeout.it_value.tv_sec = oneshot;
+ if (timer_settime(timerid, 0, &timeout, NULL)) {
+ perror("timer_settime");
+ return -1;
+ }
+ pause();
+ timer_delete(timerid);
+ }
+
+ if (periodic) {
+ install_handler(SIGALRM, handle_alarm);
+ /* Create a timer. */
+ sigevent.sigev_notify = SIGEV_SIGNAL;
+ sigevent.sigev_signo = SIGALRM;
+ if (timer_create(clkid, &sigevent, &timerid)) {
+ perror("timer_create");
+ return -1;
+ }
+ /* Start the timer. */
+ memset(&timeout, 0, sizeof(timeout));
+ timeout.it_interval.tv_sec = periodic;
+ timeout.it_value.tv_sec = periodic;
+ if (timer_settime(timerid, 0, &timeout, NULL)) {
+ perror("timer_settime");
+ return -1;
+ }
+ while (1) {
+ pause();
+ }
+ timer_delete(timerid);
+ }
+
+ if (perout >= 0) {
+ if (clock_gettime(clkid, &ts)) {
+ perror("clock_gettime");
+ return -1;
+ }
+ memset(&perout_request, 0, sizeof(perout_request));
+ perout_request.index = 0;
+ perout_request.start.sec = ts.tv_sec + 2;
+ perout_request.start.nsec = 0;
+ perout_request.period.sec = 0;
+ perout_request.period.nsec = perout;
+ if (ioctl(fd, PTP_PEROUT_REQUEST, &perout_request)) {
+ perror("PTP_PEROUT_REQUEST");
+ } else {
+ puts("periodic output request okay");
+ }
+ }
+
+ if (pps != -1) {
+ int enable = pps ? 1 : 0;
+ if (ioctl(fd, PTP_ENABLE_PPS, enable)) {
+ perror("PTP_ENABLE_PPS");
+ } else {
+ puts("pps for system time request okay");
+ }
+ }
+
+ close(fd);
+ return 0;
+}
diff --git a/Documentation/ptp/testptp.mk b/Documentation/ptp/testptp.mk
new file mode 100644
index 0000000..4ef2d97
--- /dev/null
+++ b/Documentation/ptp/testptp.mk
@@ -0,0 +1,33 @@
+# PTP 1588 clock support - User space test program
+#
+# Copyright (C) 2010 OMICRON electronics GmbH
+#
+# 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.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+
+CC = $(CROSS_COMPILE)gcc
+INC = -I$(KBUILD_OUTPUT)/usr/include
+CFLAGS = -Wall $(INC)
+LDLIBS = -lrt
+PROGS = testptp
+
+all: $(PROGS)
+
+testptp: testptp.o
+
+clean:
+ rm -f testptp.o
+
+distclean: clean
+ rm -f $(PROGS)
diff --git a/drivers/Kconfig b/drivers/Kconfig
index 177c7d1..6fd7446 100644
--- a/drivers/Kconfig
+++ b/drivers/Kconfig
@@ -54,6 +54,8 @@ source "drivers/spi/Kconfig"
source "drivers/pps/Kconfig"
+source "drivers/ptp/Kconfig"
+
source "drivers/gpio/Kconfig"
source "drivers/w1/Kconfig"
diff --git a/drivers/Makefile b/drivers/Makefile
index 3f135b6..71ebbf9 100644
--- a/drivers/Makefile
+++ b/drivers/Makefile
@@ -76,6 +76,7 @@ obj-$(CONFIG_I2O) += message/
obj-$(CONFIG_RTC_LIB) += rtc/
obj-y += i2c/ media/
obj-$(CONFIG_PPS) += pps/
+obj-$(CONFIG_PTP_1588_CLOCK) += ptp/
obj-$(CONFIG_W1) += w1/
obj-$(CONFIG_POWER_SUPPLY) += power/
obj-$(CONFIG_HWMON) += hwmon/
diff --git a/drivers/ptp/Kconfig b/drivers/ptp/Kconfig
new file mode 100644
index 0000000..70d4bb1
--- /dev/null
+++ b/drivers/ptp/Kconfig
@@ -0,0 +1,30 @@
+#
+# PTP clock support configuration
+#
+
+menu "PTP clock support"
+
+comment "Enable Device Drivers -> PPS to see the PTP clock options."
+ depends on PPS=n
+
+config PTP_1588_CLOCK
+ tristate "PTP clock support"
+ depends on EXPERIMENTAL
+ depends on PPS
+ help
+ The IEEE 1588 standard defines a method to precisely
+ synchronize distributed clocks over Ethernet networks. The
+ standard defines a Precision Time Protocol (PTP), which can
+ be used to achieve synchronization within a few dozen
+ microseconds. In addition, with the help of special hardware
+ time stamping units, it can be possible to achieve
+ synchronization to within a few hundred nanoseconds.
+
+ This driver adds support for PTP clocks as character
+ devices. If you want to use a PTP clock, then you should
+ also enable at least one clock driver as well.
+
+ To compile this driver as a module, choose M here: the module
+ will be called ptp.
+
+endmenu
diff --git a/drivers/ptp/Makefile b/drivers/ptp/Makefile
new file mode 100644
index 0000000..480e2af
--- /dev/null
+++ b/drivers/ptp/Makefile
@@ -0,0 +1,6 @@
+#
+# Makefile for PTP 1588 clock support.
+#
+
+ptp-y := ptp_clock.o ptp_chardev.o ptp_sysfs.o
+obj-$(CONFIG_PTP_1588_CLOCK) += ptp.o
diff --git a/drivers/ptp/ptp_chardev.c b/drivers/ptp/ptp_chardev.c
new file mode 100644
index 0000000..a8d03ae
--- /dev/null
+++ b/drivers/ptp/ptp_chardev.c
@@ -0,0 +1,159 @@
+/*
+ * PTP 1588 clock support - character device implementation.
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+#include <linux/module.h>
+#include <linux/posix-clock.h>
+#include <linux/poll.h>
+#include <linux/sched.h>
+
+#include "ptp_private.h"
+
+int ptp_open(struct posix_clock *pc, fmode_t fmode)
+{
+ return 0;
+}
+
+long ptp_ioctl(struct posix_clock *pc, unsigned int cmd, unsigned long arg)
+{
+ struct ptp_clock_caps caps;
+ struct ptp_clock_request req;
+ struct ptp_clock *ptp = container_of(pc, struct ptp_clock, clock);
+ struct ptp_clock_info *ops = ptp->info;
+ int enable, err = 0;
+
+ switch (cmd) {
+
+ case PTP_CLOCK_GETCAPS:
+ memset(&caps, 0, sizeof(caps));
+ caps.max_adj = ptp->info->max_adj;
+ caps.n_alarm = ptp->info->n_alarm;
+ caps.n_ext_ts = ptp->info->n_ext_ts;
+ caps.n_per_out = ptp->info->n_per_out;
+ caps.pps = ptp->info->pps;
+ err = copy_to_user((void __user *)arg, &caps, sizeof(caps));
+ break;
+
+ case PTP_EXTTS_REQUEST:
+ if (copy_from_user(&req.extts, (void __user *)arg,
+ sizeof(req.extts))) {
+ err = -EFAULT;
+ break;
+ }
+ if (req.extts.index >= ops->n_ext_ts) {
+ err = -EINVAL;
+ break;
+ }
+ req.type = PTP_CLK_REQ_EXTTS;
+ enable = req.extts.flags & PTP_ENABLE_FEATURE ? 1 : 0;
+ err = ops->enable(ops, &req, enable);
+ break;
+
+ case PTP_PEROUT_REQUEST:
+ if (copy_from_user(&req.perout, (void __user *)arg,
+ sizeof(req.perout))) {
+ err = -EFAULT;
+ break;
+ }
+ if (req.perout.index >= ops->n_per_out) {
+ err = -EINVAL;
+ break;
+ }
+ req.type = PTP_CLK_REQ_PEROUT;
+ enable = req.perout.period.sec || req.perout.period.nsec;
+ err = ops->enable(ops, &req, enable);
+ break;
+
+ case PTP_ENABLE_PPS:
+ if (!capable(CAP_SYS_TIME))
+ return -EPERM;
+ req.type = PTP_CLK_REQ_PPS;
+ enable = arg ? 1 : 0;
+ err = ops->enable(ops, &req, enable);
+ break;
+
+ default:
+ err = -ENOTTY;
+ break;
+ }
+ return err;
+}
+
+unsigned int ptp_poll(struct posix_clock *pc, struct file *fp, poll_table *wait)
+{
+ struct ptp_clock *ptp = container_of(pc, struct ptp_clock, clock);
+
+ poll_wait(fp, &ptp->tsev_wq, wait);
+
+ return queue_cnt(&ptp->tsevq) ? POLLIN : 0;
+}
+
+ssize_t ptp_read(struct posix_clock *pc,
+ uint rdflags, char __user *buf, size_t cnt)
+{
+ struct ptp_clock *ptp = container_of(pc, struct ptp_clock, clock);
+ struct timestamp_event_queue *queue = &ptp->tsevq;
+ struct ptp_extts_event event[PTP_BUF_TIMESTAMPS];
+ unsigned long flags;
+ size_t qcnt, i;
+
+ if (cnt % sizeof(struct ptp_extts_event) != 0)
+ return -EINVAL;
+
+ if (cnt > sizeof(event))
+ cnt = sizeof(event);
+
+ cnt = cnt / sizeof(struct ptp_extts_event);
+
+ if (mutex_lock_interruptible(&ptp->tsevq_mux))
+ return -ERESTARTSYS;
+
+ if (wait_event_interruptible(ptp->tsev_wq,
+ ptp->defunct || queue_cnt(queue))) {
+ mutex_unlock(&ptp->tsevq_mux);
+ return -ERESTARTSYS;
+ }
+
+ if (ptp->defunct)
+ return -ENODEV;
+
+ spin_lock_irqsave(&queue->lock, flags);
+
+ qcnt = queue_cnt(queue);
+
+ if (cnt > qcnt)
+ cnt = qcnt;
+
+ for (i = 0; i < cnt; i++) {
+ event[i] = queue->buf[queue->head];
+ queue->head = (queue->head + 1) % PTP_MAX_TIMESTAMPS;
+ }
+
+ spin_unlock_irqrestore(&queue->lock, flags);
+
+ cnt = cnt * sizeof(struct ptp_extts_event);
+
+ mutex_unlock(&ptp->tsevq_mux);
+
+ if (copy_to_user(buf, event, cnt)) {
+ mutex_unlock(&ptp->tsevq_mux);
+ return -EFAULT;
+ }
+
+ return cnt;
+}
diff --git a/drivers/ptp/ptp_clock.c b/drivers/ptp/ptp_clock.c
new file mode 100644
index 0000000..cf3f999
--- /dev/null
+++ b/drivers/ptp/ptp_clock.c
@@ -0,0 +1,343 @@
+/*
+ * PTP 1588 clock support
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+#include <linux/bitops.h>
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/posix-clock.h>
+#include <linux/pps_kernel.h>
+#include <linux/slab.h>
+#include <linux/syscalls.h>
+#include <linux/uaccess.h>
+
+#include "ptp_private.h"
+
+#define PTP_MAX_ALARMS 4
+#define PTP_MAX_CLOCKS 8
+#define PTP_PPS_DEFAULTS (PPS_CAPTUREASSERT | PPS_OFFSETASSERT)
+#define PTP_PPS_EVENT PPS_CAPTUREASSERT
+#define PTP_PPS_MODE (PTP_PPS_DEFAULTS | PPS_CANWAIT | PPS_TSFMT_TSPEC)
+
+/* private globals */
+
+static dev_t ptp_devt;
+static struct class *ptp_class;
+
+static DECLARE_BITMAP(ptp_clocks_map, PTP_MAX_CLOCKS);
+static DEFINE_MUTEX(ptp_clocks_mutex); /* protects 'ptp_clocks_map' */
+
+/* time stamp event queue operations */
+
+static inline int queue_free(struct timestamp_event_queue *q)
+{
+ return PTP_MAX_TIMESTAMPS - queue_cnt(q) - 1;
+}
+
+static void enqueue_external_timestamp(struct timestamp_event_queue *queue,
+ struct ptp_clock_event *src)
+{
+ struct ptp_extts_event *dst;
+ unsigned long flags;
+ s64 seconds;
+ u32 remainder;
+
+ seconds = div_u64_rem(src->timestamp, 1000000000, &remainder);
+
+ spin_lock_irqsave(&queue->lock, flags);
+
+ dst = &queue->buf[queue->tail];
+ dst->index = src->index;
+ dst->t.sec = seconds;
+ dst->t.nsec = remainder;
+
+ if (!queue_free(queue))
+ queue->head = (queue->head + 1) % PTP_MAX_TIMESTAMPS;
+
+ queue->tail = (queue->tail + 1) % PTP_MAX_TIMESTAMPS;
+
+ spin_unlock_irqrestore(&queue->lock, flags);
+}
+
+static s32 scaled_ppm_to_ppb(long ppm)
+{
+ /*
+ * The 'freq' field in the 'struct timex' is in parts per
+ * million, but with a 16 bit binary fractional field.
+ *
+ * We want to calculate
+ *
+ * ppb = scaled_ppm * 1000 / 2^16
+ *
+ * which simplifies to
+ *
+ * ppb = scaled_ppm * 125 / 2^13
+ */
+ s64 ppb = 1 + ppm;
+ ppb *= 125;
+ ppb >>= 13;
+ return (s32) ppb;
+}
+
+/* posix clock implementation */
+
+static int ptp_clock_getres(struct posix_clock *pc, struct timespec *tp)
+{
+ return 1; /* always round timer functions to one nanosecond */
+}
+
+static int ptp_clock_settime(struct posix_clock *pc, const struct timespec *tp)
+{
+ struct ptp_clock *ptp = container_of(pc, struct ptp_clock, clock);
+ return ptp->info->settime(ptp->info, tp);
+}
+
+static int ptp_clock_gettime(struct posix_clock *pc, struct timespec *tp)
+{
+ struct ptp_clock *ptp = container_of(pc, struct ptp_clock, clock);
+ return ptp->info->gettime(ptp->info, tp);
+}
+
+static int ptp_clock_adjtime(struct posix_clock *pc, struct timex *tx)
+{
+ struct ptp_clock *ptp = container_of(pc, struct ptp_clock, clock);
+ struct ptp_clock_info *ops;
+ int err = -EOPNOTSUPP;
+
+ ops = ptp->info;
+
+ if (tx->modes & ADJ_SETOFFSET) {
+ struct timespec ts;
+ ktime_t kt;
+ s64 delta;
+
+ ts.tv_sec = tx->time.tv_sec;
+ ts.tv_nsec = tx->time.tv_usec;
+
+ if (!(tx->modes & ADJ_NANO))
+ ts.tv_nsec *= 1000;
+
+ if ((unsigned long) ts.tv_nsec >= NSEC_PER_SEC)
+ return -EINVAL;
+
+ kt = timespec_to_ktime(ts);
+ delta = ktime_to_ns(kt);
+ err = ops->adjtime(ops, delta);
+
+ } else if (tx->modes & ADJ_FREQUENCY) {
+
+ err = ops->adjfreq(ops, scaled_ppm_to_ppb(tx->freq));
+ }
+
+ return err;
+}
+
+static struct posix_clock_operations ptp_clock_ops = {
+ .owner = THIS_MODULE,
+ .clock_adjtime = ptp_clock_adjtime,
+ .clock_gettime = ptp_clock_gettime,
+ .clock_getres = ptp_clock_getres,
+ .clock_settime = ptp_clock_settime,
+ .ioctl = ptp_ioctl,
+ .open = ptp_open,
+ .poll = ptp_poll,
+ .read = ptp_read,
+};
+
+static void delete_ptp_clock(struct posix_clock *pc)
+{
+ struct ptp_clock *ptp = container_of(pc, struct ptp_clock, clock);
+
+ mutex_destroy(&ptp->tsevq_mux);
+
+ /* Remove the clock from the bit map. */
+ mutex_lock(&ptp_clocks_mutex);
+ clear_bit(ptp->index, ptp_clocks_map);
+ mutex_unlock(&ptp_clocks_mutex);
+
+ kfree(ptp);
+}
+
+/* public interface */
+
+struct ptp_clock *ptp_clock_register(struct ptp_clock_info *info)
+{
+ struct ptp_clock *ptp;
+ int err = 0, index, major = MAJOR(ptp_devt);
+
+ if (info->n_alarm > PTP_MAX_ALARMS)
+ return ERR_PTR(-EINVAL);
+
+ /* Find a free clock slot and reserve it. */
+ err = -EBUSY;
+ mutex_lock(&ptp_clocks_mutex);
+ index = find_first_zero_bit(ptp_clocks_map, PTP_MAX_CLOCKS);
+ if (index < PTP_MAX_CLOCKS)
+ set_bit(index, ptp_clocks_map);
+ else
+ goto no_slot;
+
+ /* Initialize a clock structure. */
+ err = -ENOMEM;
+ ptp = kzalloc(sizeof(struct ptp_clock), GFP_KERNEL);
+ if (ptp == NULL)
+ goto no_memory;
+
+ ptp->clock.ops = ptp_clock_ops;
+ ptp->clock.release = delete_ptp_clock;
+ ptp->info = info;
+ ptp->devid = MKDEV(major, index);
+ ptp->index = index;
+ spin_lock_init(&ptp->tsevq.lock);
+ mutex_init(&ptp->tsevq_mux);
+ init_waitqueue_head(&ptp->tsev_wq);
+
+ /* Create a new device in our class. */
+ ptp->dev = device_create(ptp_class, NULL, ptp->devid, ptp,
+ "ptp%d", ptp->index);
+ if (IS_ERR(ptp->dev))
+ goto no_device;
+
+ dev_set_drvdata(ptp->dev, ptp);
+
+ err = ptp_populate_sysfs(ptp);
+ if (err)
+ goto no_sysfs;
+
+ /* Register a new PPS source. */
+ if (info->pps) {
+ struct pps_source_info pps;
+ memset(&pps, 0, sizeof(pps));
+ snprintf(pps.name, PPS_MAX_NAME_LEN, "ptp%d", index);
+ pps.mode = PTP_PPS_MODE;
+ pps.owner = info->owner;
+ ptp->pps_source = pps_register_source(&pps, PTP_PPS_DEFAULTS);
+ if (!ptp->pps_source) {
+ pr_err("failed to register pps source\n");
+ goto no_pps;
+ }
+ }
+
+ /* Create a posix clock. */
+ err = posix_clock_register(&ptp->clock, ptp->devid);
+ if (err) {
+ pr_err("failed to create posix clock\n");
+ goto no_clock;
+ }
+
+ mutex_unlock(&ptp_clocks_mutex);
+ return ptp;
+
+no_clock:
+ if (ptp->pps_source)
+ pps_unregister_source(ptp->pps_source);
+no_pps:
+ ptp_cleanup_sysfs(ptp);
+no_sysfs:
+ device_destroy(ptp_class, ptp->devid);
+no_device:
+ mutex_destroy(&ptp->tsevq_mux);
+ kfree(ptp);
+no_memory:
+ clear_bit(index, ptp_clocks_map);
+no_slot:
+ mutex_unlock(&ptp_clocks_mutex);
+ return ERR_PTR(err);
+}
+EXPORT_SYMBOL(ptp_clock_register);
+
+int ptp_clock_unregister(struct ptp_clock *ptp)
+{
+ ptp->defunct = 1;
+ wake_up_interruptible(&ptp->tsev_wq);
+
+ /* Release the clock's resources. */
+ if (ptp->pps_source)
+ pps_unregister_source(ptp->pps_source);
+ ptp_cleanup_sysfs(ptp);
+ device_destroy(ptp_class, ptp->devid);
+
+ posix_clock_unregister(&ptp->clock);
+ return 0;
+}
+EXPORT_SYMBOL(ptp_clock_unregister);
+
+void ptp_clock_event(struct ptp_clock *ptp, struct ptp_clock_event *event)
+{
+ struct pps_event_time evt;
+
+ switch (event->type) {
+
+ case PTP_CLOCK_ALARM:
+ break;
+
+ case PTP_CLOCK_EXTTS:
+ enqueue_external_timestamp(&ptp->tsevq, event);
+ wake_up_interruptible(&ptp->tsev_wq);
+ break;
+
+ case PTP_CLOCK_PPS:
+ pps_get_ts(&evt);
+ pps_event(ptp->pps_source, &evt, PTP_PPS_EVENT, NULL);
+ break;
+ }
+}
+EXPORT_SYMBOL(ptp_clock_event);
+
+/* module operations */
+
+static void __exit ptp_exit(void)
+{
+ class_destroy(ptp_class);
+ unregister_chrdev_region(ptp_devt, PTP_MAX_CLOCKS);
+}
+
+static int __init ptp_init(void)
+{
+ int err;
+
+ ptp_class = class_create(THIS_MODULE, "ptp");
+ if (IS_ERR(ptp_class)) {
+ pr_err("ptp: failed to allocate class\n");
+ return PTR_ERR(ptp_class);
+ }
+
+ err = alloc_chrdev_region(&ptp_devt, 0, PTP_MAX_CLOCKS, "ptp");
+ if (err < 0) {
+ pr_err("ptp: failed to allocate device region\n");
+ goto no_region;
+ }
+
+ ptp_class->dev_attrs = ptp_dev_attrs;
+ pr_info("PTP clock support registered\n");
+ return 0;
+
+no_region:
+ class_destroy(ptp_class);
+ return err;
+}
+
+subsys_initcall(ptp_init);
+module_exit(ptp_exit);
+
+MODULE_AUTHOR("Richard Cochran <richard.cochran-3mrvs1K0uXizZXS1Dc/lvw@public.gmane.org>");
+MODULE_DESCRIPTION("PTP clocks support");
+MODULE_LICENSE("GPL");
diff --git a/drivers/ptp/ptp_private.h b/drivers/ptp/ptp_private.h
new file mode 100644
index 0000000..4d5b508
--- /dev/null
+++ b/drivers/ptp/ptp_private.h
@@ -0,0 +1,92 @@
+/*
+ * PTP 1588 clock support - private declarations for the core module.
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+#ifndef _PTP_PRIVATE_H_
+#define _PTP_PRIVATE_H_
+
+#include <linux/cdev.h>
+#include <linux/device.h>
+#include <linux/mutex.h>
+#include <linux/posix-clock.h>
+#include <linux/ptp_clock.h>
+#include <linux/ptp_clock_kernel.h>
+#include <linux/time.h>
+
+#define PTP_MAX_TIMESTAMPS 128
+#define PTP_BUF_TIMESTAMPS 30
+
+struct timestamp_event_queue {
+ struct ptp_extts_event buf[PTP_MAX_TIMESTAMPS];
+ int head;
+ int tail;
+ spinlock_t lock;
+};
+
+struct ptp_clock {
+ struct posix_clock clock;
+ struct device *dev;
+ struct ptp_clock_info *info;
+ dev_t devid;
+ int index; /* index into clocks.map */
+ struct pps_device *pps_source;
+ struct timestamp_event_queue tsevq; /* simple fifo for time stamps */
+ struct mutex tsevq_mux; /* one process at a time reading the fifo */
+ wait_queue_head_t tsev_wq;
+ int defunct; /* tells readers to go away when clock is being removed */
+};
+
+/*
+ * The function queue_cnt() is safe for readers to call without
+ * holding q->lock. Readers use this function to verify that the queue
+ * is nonempty before proceeding with a dequeue operation. The fact
+ * that a writer might concurrently increment the tail does not
+ * matter, since the queue remains nonempty nonetheless.
+ */
+static inline int queue_cnt(struct timestamp_event_queue *q)
+{
+ int cnt = q->tail - q->head;
+ return cnt < 0 ? PTP_MAX_TIMESTAMPS + cnt : cnt;
+}
+
+/*
+ * see ptp_chardev.c
+ */
+
+long ptp_ioctl(struct posix_clock *pc,
+ unsigned int cmd, unsigned long arg);
+
+int ptp_open(struct posix_clock *pc, fmode_t fmode);
+
+ssize_t ptp_read(struct posix_clock *pc,
+ uint flags, char __user *buf, size_t cnt);
+
+uint ptp_poll(struct posix_clock *pc,
+ struct file *fp, poll_table *wait);
+
+/*
+ * see ptp_sysfs.c
+ */
+
+extern struct device_attribute ptp_dev_attrs[];
+
+int ptp_cleanup_sysfs(struct ptp_clock *ptp);
+
+int ptp_populate_sysfs(struct ptp_clock *ptp);
+
+#endif
diff --git a/drivers/ptp/ptp_sysfs.c b/drivers/ptp/ptp_sysfs.c
new file mode 100644
index 0000000..2f93926
--- /dev/null
+++ b/drivers/ptp/ptp_sysfs.c
@@ -0,0 +1,230 @@
+/*
+ * PTP 1588 clock support - sysfs interface.
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+#include <linux/capability.h>
+
+#include "ptp_private.h"
+
+static ssize_t clock_name_show(struct device *dev,
+ struct device_attribute *attr, char *page)
+{
+ struct ptp_clock *ptp = dev_get_drvdata(dev);
+ return snprintf(page, PAGE_SIZE-1, "%s\n", ptp->info->name);
+}
+
+#define PTP_SHOW_INT(name) \
+static ssize_t name##_show(struct device *dev, \
+ struct device_attribute *attr, char *page) \
+{ \
+ struct ptp_clock *ptp = dev_get_drvdata(dev); \
+ return snprintf(page, PAGE_SIZE-1, "%d\n", ptp->info->name); \
+}
+
+PTP_SHOW_INT(max_adj);
+PTP_SHOW_INT(n_alarm);
+PTP_SHOW_INT(n_ext_ts);
+PTP_SHOW_INT(n_per_out);
+PTP_SHOW_INT(pps);
+
+#define PTP_RO_ATTR(_var, _name) { \
+ .attr = { .name = __stringify(_name), .mode = 0444 }, \
+ .show = _var##_show, \
+}
+
+struct device_attribute ptp_dev_attrs[] = {
+ PTP_RO_ATTR(clock_name, clock_name),
+ PTP_RO_ATTR(max_adj, max_adjustment),
+ PTP_RO_ATTR(n_alarm, n_alarms),
+ PTP_RO_ATTR(n_ext_ts, n_external_timestamps),
+ PTP_RO_ATTR(n_per_out, n_periodic_outputs),
+ PTP_RO_ATTR(pps, pps_available),
+ __ATTR_NULL,
+};
+
+static ssize_t extts_enable_store(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t count)
+{
+ struct ptp_clock *ptp = dev_get_drvdata(dev);
+ struct ptp_clock_info *ops = ptp->info;
+ struct ptp_clock_request req = { .type = PTP_CLK_REQ_EXTTS };
+ int cnt, enable;
+ int err = -EINVAL;
+
+ cnt = sscanf(buf, "%u %d", &req.extts.index, &enable);
+ if (cnt != 2)
+ goto out;
+ if (req.extts.index >= ops->n_ext_ts)
+ goto out;
+
+ err = ops->enable(ops, &req, enable ? 1 : 0);
+ if (err)
+ goto out;
+
+ return count;
+out:
+ return err;
+}
+
+static ssize_t extts_fifo_show(struct device *dev,
+ struct device_attribute *attr, char *page)
+{
+ struct ptp_clock *ptp = dev_get_drvdata(dev);
+ struct timestamp_event_queue *queue = &ptp->tsevq;
+ struct ptp_extts_event event;
+ unsigned long flags;
+ size_t qcnt;
+ int cnt = 0;
+
+ memset(&event, 0, sizeof(event));
+
+ if (mutex_lock_interruptible(&ptp->tsevq_mux))
+ return -ERESTARTSYS;
+
+ spin_lock_irqsave(&queue->lock, flags);
+ qcnt = queue_cnt(queue);
+ if (qcnt) {
+ event = queue->buf[queue->head];
+ queue->head = (queue->head + 1) % PTP_MAX_TIMESTAMPS;
+ }
+ spin_unlock_irqrestore(&queue->lock, flags);
+
+ if (!qcnt)
+ goto out;
+
+ cnt = snprintf(page, PAGE_SIZE, "%u %lld %u\n",
+ event.index, event.t.sec, event.t.nsec);
+out:
+ mutex_unlock(&ptp->tsevq_mux);
+ return cnt;
+}
+
+static ssize_t period_store(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t count)
+{
+ struct ptp_clock *ptp = dev_get_drvdata(dev);
+ struct ptp_clock_info *ops = ptp->info;
+ struct ptp_clock_request req = { .type = PTP_CLK_REQ_PEROUT };
+ int cnt, enable, err = -EINVAL;
+
+ cnt = sscanf(buf, "%u %lld %u %lld %u", &req.perout.index,
+ &req.perout.start.sec, &req.perout.start.nsec,
+ &req.perout.period.sec, &req.perout.period.nsec);
+ if (cnt != 5)
+ goto out;
+ if (req.perout.index >= ops->n_per_out)
+ goto out;
+
+ enable = req.perout.period.sec || req.perout.period.nsec;
+ err = ops->enable(ops, &req, enable);
+ if (err)
+ goto out;
+
+ return count;
+out:
+ return err;
+}
+
+static ssize_t pps_enable_store(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t count)
+{
+ struct ptp_clock *ptp = dev_get_drvdata(dev);
+ struct ptp_clock_info *ops = ptp->info;
+ struct ptp_clock_request req = { .type = PTP_CLK_REQ_PPS };
+ int cnt, enable;
+ int err = -EINVAL;
+
+ if (!capable(CAP_SYS_TIME))
+ return -EPERM;
+
+ cnt = sscanf(buf, "%d", &enable);
+ if (cnt != 1)
+ goto out;
+
+ err = ops->enable(ops, &req, enable ? 1 : 0);
+ if (err)
+ goto out;
+
+ return count;
+out:
+ return err;
+}
+
+static DEVICE_ATTR(extts_enable, 0220, NULL, extts_enable_store);
+static DEVICE_ATTR(fifo, 0444, extts_fifo_show, NULL);
+static DEVICE_ATTR(period, 0220, NULL, period_store);
+static DEVICE_ATTR(pps_enable, 0220, NULL, pps_enable_store);
+
+int ptp_cleanup_sysfs(struct ptp_clock *ptp)
+{
+ struct device *dev = ptp->dev;
+ struct ptp_clock_info *info = ptp->info;
+
+ if (info->n_ext_ts) {
+ device_remove_file(dev, &dev_attr_extts_enable);
+ device_remove_file(dev, &dev_attr_fifo);
+ }
+ if (info->n_per_out)
+ device_remove_file(dev, &dev_attr_period);
+
+ if (info->pps)
+ device_remove_file(dev, &dev_attr_pps_enable);
+
+ return 0;
+}
+
+int ptp_populate_sysfs(struct ptp_clock *ptp)
+{
+ struct device *dev = ptp->dev;
+ struct ptp_clock_info *info = ptp->info;
+ int err;
+
+ if (info->n_ext_ts) {
+ err = device_create_file(dev, &dev_attr_extts_enable);
+ if (err)
+ goto out1;
+ err = device_create_file(dev, &dev_attr_fifo);
+ if (err)
+ goto out2;
+ }
+ if (info->n_per_out) {
+ err = device_create_file(dev, &dev_attr_period);
+ if (err)
+ goto out3;
+ }
+ if (info->pps) {
+ err = device_create_file(dev, &dev_attr_pps_enable);
+ if (err)
+ goto out4;
+ }
+ return 0;
+out4:
+ if (info->n_per_out)
+ device_remove_file(dev, &dev_attr_period);
+out3:
+ if (info->n_ext_ts)
+ device_remove_file(dev, &dev_attr_fifo);
+out2:
+ if (info->n_ext_ts)
+ device_remove_file(dev, &dev_attr_extts_enable);
+out1:
+ return err;
+}
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 75cf611..4585836 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -302,6 +302,7 @@ header-y += ppp-comp.h
header-y += ppp_defs.h
header-y += pps.h
header-y += prctl.h
+header-y += ptp_clock.h
header-y += ptrace.h
header-y += qnx4_fs.h
header-y += qnxtypes.h
diff --git a/include/linux/ptp_classify.h b/include/linux/ptp_classify.h
index 943a85a..e07e274 100644
--- a/include/linux/ptp_classify.h
+++ b/include/linux/ptp_classify.h
@@ -25,6 +25,7 @@
#include <linux/if_ether.h>
#include <linux/if_vlan.h>
+#include <linux/ip.h>
#include <linux/filter.h>
#ifdef __KERNEL__
#include <linux/in.h>
@@ -58,6 +59,12 @@
#define OFF_NEXT 6
#define OFF_UDP_DST 2
+#define OFF_PTP_SOURCE_UUID 22 /* PTPv1 only */
+#define OFF_PTP_SEQUENCE_ID 30
+#define OFF_PTP_CONTROL 32 /* PTPv1 only */
+
+#define IPV4_HLEN(data) (((struct iphdr *)(data + OFF_IHL))->ihl << 2)
+
#define IP6_HLEN 40
#define UDP_HLEN 8
diff --git a/include/linux/ptp_clock.h b/include/linux/ptp_clock.h
new file mode 100644
index 0000000..94e981f
--- /dev/null
+++ b/include/linux/ptp_clock.h
@@ -0,0 +1,84 @@
+/*
+ * PTP 1588 clock support - user space interface
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef _PTP_CLOCK_H_
+#define _PTP_CLOCK_H_
+
+#include <linux/ioctl.h>
+#include <linux/types.h>
+
+/* PTP_xxx bits, for the flags field within the request structures. */
+#define PTP_ENABLE_FEATURE (1<<0)
+#define PTP_RISING_EDGE (1<<1)
+#define PTP_FALLING_EDGE (1<<2)
+
+/*
+ * struct ptp_clock_time - represents a time value
+ *
+ * The sign of the seconds field applies to the whole value. The
+ * nanoseconds field is always unsigned. The reserved field is
+ * included for sub-nanosecond resolution, should the demand for
+ * this ever appear.
+ *
+ */
+struct ptp_clock_time {
+ __s64 sec; /* seconds */
+ __u32 nsec; /* nanoseconds */
+ __u32 reserved;
+};
+
+struct ptp_clock_caps {
+ int max_adj; /* Maximum frequency adjustment in parts per billon. */
+ int n_alarm; /* Number of programmable alarms. */
+ int n_ext_ts; /* Number of external time stamp channels. */
+ int n_per_out; /* Number of programmable periodic signals. */
+ int pps; /* Whether the clock supports a PPS callback. */
+ int rsv[15]; /* Reserved for future use. */
+};
+
+struct ptp_extts_request {
+ unsigned int index; /* Which channel to configure. */
+ unsigned int flags; /* Bit field for PTP_xxx flags. */
+ unsigned int rsv[2]; /* Reserved for future use. */
+};
+
+struct ptp_perout_request {
+ struct ptp_clock_time start; /* Absolute start time. */
+ struct ptp_clock_time period; /* Desired period, zero means disable. */
+ unsigned int index; /* Which channel to configure. */
+ unsigned int flags; /* Reserved for future use. */
+ unsigned int rsv[4]; /* Reserved for future use. */
+};
+
+#define PTP_CLK_MAGIC '='
+
+#define PTP_CLOCK_GETCAPS _IOR(PTP_CLK_MAGIC, 1, struct ptp_clock_caps)
+#define PTP_EXTTS_REQUEST _IOW(PTP_CLK_MAGIC, 2, struct ptp_extts_request)
+#define PTP_PEROUT_REQUEST _IOW(PTP_CLK_MAGIC, 3, struct ptp_perout_request)
+#define PTP_ENABLE_PPS _IOW(PTP_CLK_MAGIC, 4, int)
+
+struct ptp_extts_event {
+ struct ptp_clock_time t; /* Time event occured. */
+ unsigned int index; /* Which channel produced the event. */
+ unsigned int flags; /* Reserved for future use. */
+ unsigned int rsv[2]; /* Reserved for future use. */
+};
+
+#endif
diff --git a/include/linux/ptp_clock_kernel.h b/include/linux/ptp_clock_kernel.h
new file mode 100644
index 0000000..dd2e44f
--- /dev/null
+++ b/include/linux/ptp_clock_kernel.h
@@ -0,0 +1,139 @@
+/*
+ * PTP 1588 clock support
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef _PTP_CLOCK_KERNEL_H_
+#define _PTP_CLOCK_KERNEL_H_
+
+#include <linux/ptp_clock.h>
+
+
+struct ptp_clock_request {
+ enum {
+ PTP_CLK_REQ_EXTTS,
+ PTP_CLK_REQ_PEROUT,
+ PTP_CLK_REQ_PPS,
+ } type;
+ union {
+ struct ptp_extts_request extts;
+ struct ptp_perout_request perout;
+ };
+};
+
+/**
+ * struct ptp_clock_info - decribes a PTP hardware clock
+ *
+ * @owner: The clock driver should set to THIS_MODULE.
+ * @name: A short name to identify the clock.
+ * @max_adj: The maximum possible frequency adjustment, in parts per billon.
+ * @n_alarm: The number of programmable alarms.
+ * @n_ext_ts: The number of external time stamp channels.
+ * @n_per_out: The number of programmable periodic signals.
+ * @pps: Indicates whether the clock supports a PPS callback.
+ *
+ * clock operations
+ *
+ * @adjfreq: Adjusts the frequency of the hardware clock.
+ * parameter delta: Desired period change in parts per billion.
+ *
+ * @adjtime: Shifts the time of the hardware clock.
+ * parameter delta: Desired change in nanoseconds.
+ *
+ * @gettime: Reads the current time from the hardware clock.
+ * parameter ts: Holds the result.
+ *
+ * @settime: Set the current time on the hardware clock.
+ * parameter ts: Time value to set.
+ *
+ * @enable: Request driver to enable or disable an ancillary feature.
+ * parameter request: Desired resource to enable or disable.
+ * parameter on: Caller passes one to enable or zero to disable.
+ *
+ * Drivers should embed their ptp_clock_info within a private
+ * structure, obtaining a reference to it using container_of().
+ *
+ * The callbacks must all return zero on success, non-zero otherwise.
+ */
+
+struct ptp_clock_info {
+ struct module *owner;
+ char name[16];
+ s32 max_adj;
+ int n_alarm;
+ int n_ext_ts;
+ int n_per_out;
+ int pps;
+ int (*adjfreq)(struct ptp_clock_info *ptp, s32 delta);
+ int (*adjtime)(struct ptp_clock_info *ptp, s64 delta);
+ int (*gettime)(struct ptp_clock_info *ptp, struct timespec *ts);
+ int (*settime)(struct ptp_clock_info *ptp, const struct timespec *ts);
+ int (*enable)(struct ptp_clock_info *ptp,
+ struct ptp_clock_request *request, int on);
+};
+
+struct ptp_clock;
+
+/**
+ * ptp_clock_register() - register a PTP hardware clock driver
+ *
+ * @info: Structure describing the new clock.
+ */
+
+extern struct ptp_clock *ptp_clock_register(struct ptp_clock_info *info);
+
+/**
+ * ptp_clock_unregister() - unregister a PTP hardware clock driver
+ *
+ * @ptp: The clock to remove from service.
+ */
+
+extern int ptp_clock_unregister(struct ptp_clock *ptp);
+
+
+enum ptp_clock_events {
+ PTP_CLOCK_ALARM,
+ PTP_CLOCK_EXTTS,
+ PTP_CLOCK_PPS,
+};
+
+/**
+ * struct ptp_clock_event - decribes a PTP hardware clock event
+ *
+ * @type: One of the ptp_clock_events enumeration values.
+ * @index: Identifies the source of the event.
+ * @timestamp: When the event occured.
+ */
+
+struct ptp_clock_event {
+ int type;
+ int index;
+ u64 timestamp;
+};
+
+/**
+ * ptp_clock_event() - notify the PTP layer about an event
+ *
+ * @ptp: The clock obtained from ptp_clock_register().
+ * @event: Message structure describing the event.
+ */
+
+extern void ptp_clock_event(struct ptp_clock *ptp,
+ struct ptp_clock_event *event);
+
+#endif
--
1.7.0.4
^ permalink raw reply related
* [PATCH V15 2/4] ptp: Added a clock that uses the eTSEC found on the MPC85xx.
From: Richard Cochran @ 2011-04-22 10:03 UTC (permalink / raw)
To: linux-kernel-u79uwXL29TY76Z2rM5mHXA
Cc: linux-api-u79uwXL29TY76Z2rM5mHXA, netdev-u79uwXL29TY76Z2rM5mHXA,
devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
linuxppc-dev-uLR06cmDAlY/bJ5BZ2RsiQ, Alan Cox, Arnd Bergmann,
Ben Hutchings, Benjamin Herrenschmidt, Christoph Lameter,
David Miller, John Stultz, Krzysztof Halasa, Mike Frysinger,
Paul Mackerras, Peter Zijlstra, Rodolfo Giometti, Russell King,
Thomas Gleixner
In-Reply-To: <cover.1303466102.git.richard.cochran-3mrvs1K0uXizZXS1Dc/lvw@public.gmane.org>
The eTSEC includes a PTP clock with quite a few features. This patch adds
support for the basic clock adjustment functions, plus two external time
stamps, one alarm, and the PPS callback.
Signed-off-by: Richard Cochran <richard.cochran-3mrvs1K0uXizZXS1Dc/lvw@public.gmane.org>
Acked-by: David S. Miller <davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
Acked-by: John Stultz <john.stultz-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
---
.../devicetree/bindings/net/fsl-tsec-phy.txt | 54 ++
arch/powerpc/boot/dts/mpc8313erdb.dts | 13 +
arch/powerpc/boot/dts/mpc8572ds.dts | 13 +
arch/powerpc/boot/dts/p2020ds.dts | 13 +
arch/powerpc/boot/dts/p2020rdb.dts | 13 +
drivers/net/Makefile | 1 +
drivers/net/gianfar_ptp.c | 588 ++++++++++++++++++++
drivers/ptp/Kconfig | 13 +
8 files changed, 708 insertions(+), 0 deletions(-)
create mode 100644 drivers/net/gianfar_ptp.c
diff --git a/Documentation/devicetree/bindings/net/fsl-tsec-phy.txt b/Documentation/devicetree/bindings/net/fsl-tsec-phy.txt
index edb7ae1..2c6be03 100644
--- a/Documentation/devicetree/bindings/net/fsl-tsec-phy.txt
+++ b/Documentation/devicetree/bindings/net/fsl-tsec-phy.txt
@@ -74,3 +74,57 @@ Example:
interrupt-parent = <&mpic>;
phy-handle = <&phy0>
};
+
+* Gianfar PTP clock nodes
+
+General Properties:
+
+ - compatible Should be "fsl,etsec-ptp"
+ - reg Offset and length of the register set for the device
+ - interrupts There should be at least two interrupts. Some devices
+ have as many as four PTP related interrupts.
+
+Clock Properties:
+
+ - fsl,tclk-period Timer reference clock period in nanoseconds.
+ - fsl,tmr-prsc Prescaler, divides the output clock.
+ - fsl,tmr-add Frequency compensation value.
+ - fsl,tmr-fiper1 Fixed interval period pulse generator.
+ - fsl,tmr-fiper2 Fixed interval period pulse generator.
+ - fsl,max-adj Maximum frequency adjustment in parts per billion.
+
+ These properties set the operational parameters for the PTP
+ clock. You must choose these carefully for the clock to work right.
+ Here is how to figure good values:
+
+ TimerOsc = system clock MHz
+ tclk_period = desired clock period nanoseconds
+ NominalFreq = 1000 / tclk_period MHz
+ FreqDivRatio = TimerOsc / NominalFreq (must be greater that 1.0)
+ tmr_add = ceil(2^32 / FreqDivRatio)
+ OutputClock = NominalFreq / tmr_prsc MHz
+ PulseWidth = 1 / OutputClock microseconds
+ FiperFreq1 = desired frequency in Hz
+ FiperDiv1 = 1000000 * OutputClock / FiperFreq1
+ tmr_fiper1 = tmr_prsc * tclk_period * FiperDiv1 - tclk_period
+ max_adj = 1000000000 * (FreqDivRatio - 1.0) - 1
+
+ The calculation for tmr_fiper2 is the same as for tmr_fiper1. The
+ driver expects that tmr_fiper1 will be correctly set to produce a 1
+ Pulse Per Second (PPS) signal, since this will be offered to the PPS
+ subsystem to synchronize the Linux clock.
+
+Example:
+
+ ptp_clock@24E00 {
+ compatible = "fsl,etsec-ptp";
+ reg = <0x24E00 0xB0>;
+ interrupts = <12 0x8 13 0x8>;
+ interrupt-parent = < &ipic >;
+ fsl,tclk-period = <10>;
+ fsl,tmr-prsc = <100>;
+ fsl,tmr-add = <0x999999A4>;
+ fsl,tmr-fiper1 = <0x3B9AC9F6>;
+ fsl,tmr-fiper2 = <0x00018696>;
+ fsl,max-adj = <659999998>;
+ };
diff --git a/arch/powerpc/boot/dts/mpc8313erdb.dts b/arch/powerpc/boot/dts/mpc8313erdb.dts
index 183f2aa..502e17c 100644
--- a/arch/powerpc/boot/dts/mpc8313erdb.dts
+++ b/arch/powerpc/boot/dts/mpc8313erdb.dts
@@ -208,6 +208,19 @@
sleep = <&pmc 0x00300000>;
};
+ ptp_clock@24E00 {
+ compatible = "fsl,etsec-ptp";
+ reg = <0x24E00 0xB0>;
+ interrupts = <12 0x8 13 0x8>;
+ interrupt-parent = < &ipic >;
+ fsl,tclk-period = <10>;
+ fsl,tmr-prsc = <100>;
+ fsl,tmr-add = <0x999999A4>;
+ fsl,tmr-fiper1 = <0x3B9AC9F6>;
+ fsl,tmr-fiper2 = <0x00018696>;
+ fsl,max-adj = <659999998>;
+ };
+
enet0: ethernet@24000 {
#address-cells = <1>;
#size-cells = <1>;
diff --git a/arch/powerpc/boot/dts/mpc8572ds.dts b/arch/powerpc/boot/dts/mpc8572ds.dts
index cafc128..f6c04d2 100644
--- a/arch/powerpc/boot/dts/mpc8572ds.dts
+++ b/arch/powerpc/boot/dts/mpc8572ds.dts
@@ -324,6 +324,19 @@
};
};
+ ptp_clock@24E00 {
+ compatible = "fsl,etsec-ptp";
+ reg = <0x24E00 0xB0>;
+ interrupts = <68 2 69 2 70 2 71 2>;
+ interrupt-parent = < &mpic >;
+ fsl,tclk-period = <5>;
+ fsl,tmr-prsc = <200>;
+ fsl,tmr-add = <0xAAAAAAAB>;
+ fsl,tmr-fiper1 = <0x3B9AC9FB>;
+ fsl,tmr-fiper2 = <0x3B9AC9FB>;
+ fsl,max-adj = <499999999>;
+ };
+
enet0: ethernet@24000 {
#address-cells = <1>;
#size-cells = <1>;
diff --git a/arch/powerpc/boot/dts/p2020ds.dts b/arch/powerpc/boot/dts/p2020ds.dts
index 1101914..3956daf 100644
--- a/arch/powerpc/boot/dts/p2020ds.dts
+++ b/arch/powerpc/boot/dts/p2020ds.dts
@@ -336,6 +336,19 @@
phy_type = "ulpi";
};
+ ptp_clock@24E00 {
+ compatible = "fsl,etsec-ptp";
+ reg = <0x24E00 0xB0>;
+ interrupts = <68 2 69 2 70 2>;
+ interrupt-parent = < &mpic >;
+ fsl,tclk-period = <5>;
+ fsl,tmr-prsc = <200>;
+ fsl,tmr-add = <0xCCCCCCCD>;
+ fsl,tmr-fiper1 = <0x3B9AC9FB>;
+ fsl,tmr-fiper2 = <0x0001869B>;
+ fsl,max-adj = <249999999>;
+ };
+
enet0: ethernet@24000 {
#address-cells = <1>;
#size-cells = <1>;
diff --git a/arch/powerpc/boot/dts/p2020rdb.dts b/arch/powerpc/boot/dts/p2020rdb.dts
index e2d48fd..6c55f24 100644
--- a/arch/powerpc/boot/dts/p2020rdb.dts
+++ b/arch/powerpc/boot/dts/p2020rdb.dts
@@ -396,6 +396,19 @@
phy_type = "ulpi";
};
+ ptp_clock@24E00 {
+ compatible = "fsl,etsec-ptp";
+ reg = <0x24E00 0xB0>;
+ interrupts = <68 2 69 2 70 2>;
+ interrupt-parent = < &mpic >;
+ fsl,tclk-period = <5>;
+ fsl,tmr-prsc = <200>;
+ fsl,tmr-add = <0xCCCCCCCD>;
+ fsl,tmr-fiper1 = <0x3B9AC9FB>;
+ fsl,tmr-fiper2 = <0x0001869B>;
+ fsl,max-adj = <249999999>;
+ };
+
enet0: ethernet@24000 {
#address-cells = <1>;
#size-cells = <1>;
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index 01b604a..97a0cef 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -31,6 +31,7 @@ obj-$(CONFIG_ATL2) += atlx/
obj-$(CONFIG_ATL1E) += atl1e/
obj-$(CONFIG_ATL1C) += atl1c/
obj-$(CONFIG_GIANFAR) += gianfar_driver.o
+obj-$(CONFIG_PTP_1588_CLOCK_GIANFAR) += gianfar_ptp.o
obj-$(CONFIG_TEHUTI) += tehuti.o
obj-$(CONFIG_ENIC) += enic/
obj-$(CONFIG_JME) += jme.o
diff --git a/drivers/net/gianfar_ptp.c b/drivers/net/gianfar_ptp.c
new file mode 100644
index 0000000..d8e1753
--- /dev/null
+++ b/drivers/net/gianfar_ptp.c
@@ -0,0 +1,588 @@
+/*
+ * PTP 1588 clock using the eTSEC
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+#include <linux/device.h>
+#include <linux/hrtimer.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+#include <linux/timex.h>
+#include <linux/io.h>
+
+#include <linux/ptp_clock_kernel.h>
+
+#include "gianfar.h"
+
+/*
+ * gianfar ptp registers
+ * Generated by regen.tcl on Thu May 13 01:38:57 PM CEST 2010
+ */
+struct gianfar_ptp_registers {
+ u32 tmr_ctrl; /* Timer control register */
+ u32 tmr_tevent; /* Timestamp event register */
+ u32 tmr_temask; /* Timer event mask register */
+ u32 tmr_pevent; /* Timestamp event register */
+ u32 tmr_pemask; /* Timer event mask register */
+ u32 tmr_stat; /* Timestamp status register */
+ u32 tmr_cnt_h; /* Timer counter high register */
+ u32 tmr_cnt_l; /* Timer counter low register */
+ u32 tmr_add; /* Timer drift compensation addend register */
+ u32 tmr_acc; /* Timer accumulator register */
+ u32 tmr_prsc; /* Timer prescale */
+ u8 res1[4];
+ u32 tmroff_h; /* Timer offset high */
+ u32 tmroff_l; /* Timer offset low */
+ u8 res2[8];
+ u32 tmr_alarm1_h; /* Timer alarm 1 high register */
+ u32 tmr_alarm1_l; /* Timer alarm 1 high register */
+ u32 tmr_alarm2_h; /* Timer alarm 2 high register */
+ u32 tmr_alarm2_l; /* Timer alarm 2 high register */
+ u8 res3[48];
+ u32 tmr_fiper1; /* Timer fixed period interval */
+ u32 tmr_fiper2; /* Timer fixed period interval */
+ u32 tmr_fiper3; /* Timer fixed period interval */
+ u8 res4[20];
+ u32 tmr_etts1_h; /* Timestamp of general purpose external trigger */
+ u32 tmr_etts1_l; /* Timestamp of general purpose external trigger */
+ u32 tmr_etts2_h; /* Timestamp of general purpose external trigger */
+ u32 tmr_etts2_l; /* Timestamp of general purpose external trigger */
+};
+
+/* Bit definitions for the TMR_CTRL register */
+#define ALM1P (1<<31) /* Alarm1 output polarity */
+#define ALM2P (1<<30) /* Alarm2 output polarity */
+#define FS (1<<28) /* FIPER start indication */
+#define PP1L (1<<27) /* Fiper1 pulse loopback mode enabled. */
+#define PP2L (1<<26) /* Fiper2 pulse loopback mode enabled. */
+#define TCLK_PERIOD_SHIFT (16) /* 1588 timer reference clock period. */
+#define TCLK_PERIOD_MASK (0x3ff)
+#define RTPE (1<<15) /* Record Tx Timestamp to PAL Enable. */
+#define FRD (1<<14) /* FIPER Realignment Disable */
+#define ESFDP (1<<11) /* External Tx/Rx SFD Polarity. */
+#define ESFDE (1<<10) /* External Tx/Rx SFD Enable. */
+#define ETEP2 (1<<9) /* External trigger 2 edge polarity */
+#define ETEP1 (1<<8) /* External trigger 1 edge polarity */
+#define COPH (1<<7) /* Generated clock output phase. */
+#define CIPH (1<<6) /* External oscillator input clock phase */
+#define TMSR (1<<5) /* Timer soft reset. */
+#define BYP (1<<3) /* Bypass drift compensated clock */
+#define TE (1<<2) /* 1588 timer enable. */
+#define CKSEL_SHIFT (0) /* 1588 Timer reference clock source */
+#define CKSEL_MASK (0x3)
+
+/* Bit definitions for the TMR_TEVENT register */
+#define ETS2 (1<<25) /* External trigger 2 timestamp sampled */
+#define ETS1 (1<<24) /* External trigger 1 timestamp sampled */
+#define ALM2 (1<<17) /* Current time = alarm time register 2 */
+#define ALM1 (1<<16) /* Current time = alarm time register 1 */
+#define PP1 (1<<7) /* periodic pulse generated on FIPER1 */
+#define PP2 (1<<6) /* periodic pulse generated on FIPER2 */
+#define PP3 (1<<5) /* periodic pulse generated on FIPER3 */
+
+/* Bit definitions for the TMR_TEMASK register */
+#define ETS2EN (1<<25) /* External trigger 2 timestamp enable */
+#define ETS1EN (1<<24) /* External trigger 1 timestamp enable */
+#define ALM2EN (1<<17) /* Timer ALM2 event enable */
+#define ALM1EN (1<<16) /* Timer ALM1 event enable */
+#define PP1EN (1<<7) /* Periodic pulse event 1 enable */
+#define PP2EN (1<<6) /* Periodic pulse event 2 enable */
+
+/* Bit definitions for the TMR_PEVENT register */
+#define TXP2 (1<<9) /* PTP transmitted timestamp im TXTS2 */
+#define TXP1 (1<<8) /* PTP transmitted timestamp in TXTS1 */
+#define RXP (1<<0) /* PTP frame has been received */
+
+/* Bit definitions for the TMR_PEMASK register */
+#define TXP2EN (1<<9) /* Transmit PTP packet event 2 enable */
+#define TXP1EN (1<<8) /* Transmit PTP packet event 1 enable */
+#define RXPEN (1<<0) /* Receive PTP packet event enable */
+
+/* Bit definitions for the TMR_STAT register */
+#define STAT_VEC_SHIFT (0) /* Timer general purpose status vector */
+#define STAT_VEC_MASK (0x3f)
+
+/* Bit definitions for the TMR_PRSC register */
+#define PRSC_OCK_SHIFT (0) /* Output clock division/prescale factor. */
+#define PRSC_OCK_MASK (0xffff)
+
+
+#define DRIVER "gianfar_ptp"
+#define DEFAULT_CKSEL 1
+#define N_ALARM 1 /* first alarm is used internally to reset fipers */
+#define N_EXT_TS 2
+#define REG_SIZE sizeof(struct gianfar_ptp_registers)
+
+struct etsects {
+ struct gianfar_ptp_registers *regs;
+ spinlock_t lock; /* protects regs */
+ struct ptp_clock *clock;
+ struct ptp_clock_info caps;
+ struct resource *rsrc;
+ int irq;
+ u64 alarm_interval; /* for periodic alarm */
+ u64 alarm_value;
+ u32 tclk_period; /* nanoseconds */
+ u32 tmr_prsc;
+ u32 tmr_add;
+ u32 cksel;
+ u32 tmr_fiper1;
+ u32 tmr_fiper2;
+};
+
+/*
+ * Register access functions
+ */
+
+/* Caller must hold etsects->lock. */
+static u64 tmr_cnt_read(struct etsects *etsects)
+{
+ u64 ns;
+ u32 lo, hi;
+
+ lo = gfar_read(&etsects->regs->tmr_cnt_l);
+ hi = gfar_read(&etsects->regs->tmr_cnt_h);
+ ns = ((u64) hi) << 32;
+ ns |= lo;
+ return ns;
+}
+
+/* Caller must hold etsects->lock. */
+static void tmr_cnt_write(struct etsects *etsects, u64 ns)
+{
+ u32 hi = ns >> 32;
+ u32 lo = ns & 0xffffffff;
+
+ gfar_write(&etsects->regs->tmr_cnt_l, lo);
+ gfar_write(&etsects->regs->tmr_cnt_h, hi);
+}
+
+/* Caller must hold etsects->lock. */
+static void set_alarm(struct etsects *etsects)
+{
+ u64 ns;
+ u32 lo, hi;
+
+ ns = tmr_cnt_read(etsects) + 1500000000ULL;
+ ns = div_u64(ns, 1000000000UL) * 1000000000ULL;
+ ns -= etsects->tclk_period;
+ hi = ns >> 32;
+ lo = ns & 0xffffffff;
+ gfar_write(&etsects->regs->tmr_alarm1_l, lo);
+ gfar_write(&etsects->regs->tmr_alarm1_h, hi);
+}
+
+/* Caller must hold etsects->lock. */
+static void set_fipers(struct etsects *etsects)
+{
+ u32 tmr_ctrl = gfar_read(&etsects->regs->tmr_ctrl);
+
+ gfar_write(&etsects->regs->tmr_ctrl, tmr_ctrl & (~TE));
+ gfar_write(&etsects->regs->tmr_prsc, etsects->tmr_prsc);
+ gfar_write(&etsects->regs->tmr_fiper1, etsects->tmr_fiper1);
+ gfar_write(&etsects->regs->tmr_fiper2, etsects->tmr_fiper2);
+ set_alarm(etsects);
+ gfar_write(&etsects->regs->tmr_ctrl, tmr_ctrl|TE);
+}
+
+/*
+ * Interrupt service routine
+ */
+
+static irqreturn_t isr(int irq, void *priv)
+{
+ struct etsects *etsects = priv;
+ struct ptp_clock_event event;
+ u64 ns;
+ u32 ack = 0, lo, hi, mask, val;
+
+ val = gfar_read(&etsects->regs->tmr_tevent);
+
+ if (val & ETS1) {
+ ack |= ETS1;
+ hi = gfar_read(&etsects->regs->tmr_etts1_h);
+ lo = gfar_read(&etsects->regs->tmr_etts1_l);
+ event.type = PTP_CLOCK_EXTTS;
+ event.index = 0;
+ event.timestamp = ((u64) hi) << 32;
+ event.timestamp |= lo;
+ ptp_clock_event(etsects->clock, &event);
+ }
+
+ if (val & ETS2) {
+ ack |= ETS2;
+ hi = gfar_read(&etsects->regs->tmr_etts2_h);
+ lo = gfar_read(&etsects->regs->tmr_etts2_l);
+ event.type = PTP_CLOCK_EXTTS;
+ event.index = 1;
+ event.timestamp = ((u64) hi) << 32;
+ event.timestamp |= lo;
+ ptp_clock_event(etsects->clock, &event);
+ }
+
+ if (val & ALM2) {
+ ack |= ALM2;
+ if (etsects->alarm_value) {
+ event.type = PTP_CLOCK_ALARM;
+ event.index = 0;
+ event.timestamp = etsects->alarm_value;
+ ptp_clock_event(etsects->clock, &event);
+ }
+ if (etsects->alarm_interval) {
+ ns = etsects->alarm_value + etsects->alarm_interval;
+ hi = ns >> 32;
+ lo = ns & 0xffffffff;
+ spin_lock(&etsects->lock);
+ gfar_write(&etsects->regs->tmr_alarm2_l, lo);
+ gfar_write(&etsects->regs->tmr_alarm2_h, hi);
+ spin_unlock(&etsects->lock);
+ etsects->alarm_value = ns;
+ } else {
+ gfar_write(&etsects->regs->tmr_tevent, ALM2);
+ spin_lock(&etsects->lock);
+ mask = gfar_read(&etsects->regs->tmr_temask);
+ mask &= ~ALM2EN;
+ gfar_write(&etsects->regs->tmr_temask, mask);
+ spin_unlock(&etsects->lock);
+ etsects->alarm_value = 0;
+ etsects->alarm_interval = 0;
+ }
+ }
+
+ if (val & PP1) {
+ ack |= PP1;
+ event.type = PTP_CLOCK_PPS;
+ ptp_clock_event(etsects->clock, &event);
+ }
+
+ if (ack) {
+ gfar_write(&etsects->regs->tmr_tevent, ack);
+ return IRQ_HANDLED;
+ } else
+ return IRQ_NONE;
+}
+
+/*
+ * PTP clock operations
+ */
+
+static int ptp_gianfar_adjfreq(struct ptp_clock_info *ptp, s32 ppb)
+{
+ u64 adj;
+ u32 diff, tmr_add;
+ int neg_adj = 0;
+ struct etsects *etsects = container_of(ptp, struct etsects, caps);
+
+ if (ppb < 0) {
+ neg_adj = 1;
+ ppb = -ppb;
+ }
+ tmr_add = etsects->tmr_add;
+ adj = tmr_add;
+ adj *= ppb;
+ diff = div_u64(adj, 1000000000ULL);
+
+ tmr_add = neg_adj ? tmr_add - diff : tmr_add + diff;
+
+ gfar_write(&etsects->regs->tmr_add, tmr_add);
+
+ return 0;
+}
+
+static int ptp_gianfar_adjtime(struct ptp_clock_info *ptp, s64 delta)
+{
+ s64 now;
+ unsigned long flags;
+ struct etsects *etsects = container_of(ptp, struct etsects, caps);
+
+ spin_lock_irqsave(&etsects->lock, flags);
+
+ now = tmr_cnt_read(etsects);
+ now += delta;
+ tmr_cnt_write(etsects, now);
+
+ spin_unlock_irqrestore(&etsects->lock, flags);
+
+ set_fipers(etsects);
+
+ return 0;
+}
+
+static int ptp_gianfar_gettime(struct ptp_clock_info *ptp, struct timespec *ts)
+{
+ u64 ns;
+ u32 remainder;
+ unsigned long flags;
+ struct etsects *etsects = container_of(ptp, struct etsects, caps);
+
+ spin_lock_irqsave(&etsects->lock, flags);
+
+ ns = tmr_cnt_read(etsects);
+
+ spin_unlock_irqrestore(&etsects->lock, flags);
+
+ ts->tv_sec = div_u64_rem(ns, 1000000000, &remainder);
+ ts->tv_nsec = remainder;
+ return 0;
+}
+
+static int ptp_gianfar_settime(struct ptp_clock_info *ptp,
+ const struct timespec *ts)
+{
+ u64 ns;
+ unsigned long flags;
+ struct etsects *etsects = container_of(ptp, struct etsects, caps);
+
+ ns = ts->tv_sec * 1000000000ULL;
+ ns += ts->tv_nsec;
+
+ spin_lock_irqsave(&etsects->lock, flags);
+
+ tmr_cnt_write(etsects, ns);
+ set_fipers(etsects);
+
+ spin_unlock_irqrestore(&etsects->lock, flags);
+
+ return 0;
+}
+
+static int ptp_gianfar_enable(struct ptp_clock_info *ptp,
+ struct ptp_clock_request *rq, int on)
+{
+ struct etsects *etsects = container_of(ptp, struct etsects, caps);
+ unsigned long flags;
+ u32 bit, mask;
+
+ switch (rq->type) {
+ case PTP_CLK_REQ_EXTTS:
+ switch (rq->extts.index) {
+ case 0:
+ bit = ETS1EN;
+ break;
+ case 1:
+ bit = ETS2EN;
+ break;
+ default:
+ return -EINVAL;
+ }
+ spin_lock_irqsave(&etsects->lock, flags);
+ mask = gfar_read(&etsects->regs->tmr_temask);
+ if (on)
+ mask |= bit;
+ else
+ mask &= ~bit;
+ gfar_write(&etsects->regs->tmr_temask, mask);
+ spin_unlock_irqrestore(&etsects->lock, flags);
+ return 0;
+
+ case PTP_CLK_REQ_PPS:
+ spin_lock_irqsave(&etsects->lock, flags);
+ mask = gfar_read(&etsects->regs->tmr_temask);
+ if (on)
+ mask |= PP1EN;
+ else
+ mask &= ~PP1EN;
+ gfar_write(&etsects->regs->tmr_temask, mask);
+ spin_unlock_irqrestore(&etsects->lock, flags);
+ return 0;
+
+ default:
+ break;
+ }
+
+ return -EOPNOTSUPP;
+}
+
+static struct ptp_clock_info ptp_gianfar_caps = {
+ .owner = THIS_MODULE,
+ .name = "gianfar clock",
+ .max_adj = 512000,
+ .n_alarm = N_ALARM,
+ .n_ext_ts = N_EXT_TS,
+ .n_per_out = 0,
+ .pps = 1,
+ .adjfreq = ptp_gianfar_adjfreq,
+ .adjtime = ptp_gianfar_adjtime,
+ .gettime = ptp_gianfar_gettime,
+ .settime = ptp_gianfar_settime,
+ .enable = ptp_gianfar_enable,
+};
+
+/* OF device tree */
+
+static int get_of_u32(struct device_node *node, char *str, u32 *val)
+{
+ int plen;
+ const u32 *prop = of_get_property(node, str, &plen);
+
+ if (!prop || plen != sizeof(*prop))
+ return -1;
+ *val = *prop;
+ return 0;
+}
+
+static int gianfar_ptp_probe(struct platform_device *dev)
+{
+ struct device_node *node = dev->dev.of_node;
+ struct etsects *etsects;
+ struct timespec now;
+ int err = -ENOMEM;
+ u32 tmr_ctrl;
+ unsigned long flags;
+
+ etsects = kzalloc(sizeof(*etsects), GFP_KERNEL);
+ if (!etsects)
+ goto no_memory;
+
+ err = -ENODEV;
+
+ etsects->caps = ptp_gianfar_caps;
+ etsects->cksel = DEFAULT_CKSEL;
+
+ if (get_of_u32(node, "fsl,tclk-period", &etsects->tclk_period) ||
+ get_of_u32(node, "fsl,tmr-prsc", &etsects->tmr_prsc) ||
+ get_of_u32(node, "fsl,tmr-add", &etsects->tmr_add) ||
+ get_of_u32(node, "fsl,tmr-fiper1", &etsects->tmr_fiper1) ||
+ get_of_u32(node, "fsl,tmr-fiper2", &etsects->tmr_fiper2) ||
+ get_of_u32(node, "fsl,max-adj", &etsects->caps.max_adj)) {
+ pr_err("device tree node missing required elements\n");
+ goto no_node;
+ }
+
+ etsects->irq = platform_get_irq(dev, 0);
+
+ if (etsects->irq == NO_IRQ) {
+ pr_err("irq not in device tree\n");
+ goto no_node;
+ }
+ if (request_irq(etsects->irq, isr, 0, DRIVER, etsects)) {
+ pr_err("request_irq failed\n");
+ goto no_node;
+ }
+
+ etsects->rsrc = platform_get_resource(dev, IORESOURCE_MEM, 0);
+ if (!etsects->rsrc) {
+ pr_err("no resource\n");
+ goto no_resource;
+ }
+ if (request_resource(&ioport_resource, etsects->rsrc)) {
+ pr_err("resource busy\n");
+ goto no_resource;
+ }
+
+ spin_lock_init(&etsects->lock);
+
+ etsects->regs = ioremap(etsects->rsrc->start,
+ 1 + etsects->rsrc->end - etsects->rsrc->start);
+ if (!etsects->regs) {
+ pr_err("ioremap ptp registers failed\n");
+ goto no_ioremap;
+ }
+ getnstimeofday(&now);
+ ptp_gianfar_settime(&etsects->caps, &now);
+
+ tmr_ctrl =
+ (etsects->tclk_period & TCLK_PERIOD_MASK) << TCLK_PERIOD_SHIFT |
+ (etsects->cksel & CKSEL_MASK) << CKSEL_SHIFT;
+
+ spin_lock_irqsave(&etsects->lock, flags);
+
+ gfar_write(&etsects->regs->tmr_ctrl, tmr_ctrl);
+ gfar_write(&etsects->regs->tmr_add, etsects->tmr_add);
+ gfar_write(&etsects->regs->tmr_prsc, etsects->tmr_prsc);
+ gfar_write(&etsects->regs->tmr_fiper1, etsects->tmr_fiper1);
+ gfar_write(&etsects->regs->tmr_fiper2, etsects->tmr_fiper2);
+ set_alarm(etsects);
+ gfar_write(&etsects->regs->tmr_ctrl, tmr_ctrl|FS|RTPE|TE);
+
+ spin_unlock_irqrestore(&etsects->lock, flags);
+
+ etsects->clock = ptp_clock_register(&etsects->caps);
+ if (IS_ERR(etsects->clock)) {
+ err = PTR_ERR(etsects->clock);
+ goto no_clock;
+ }
+
+ dev_set_drvdata(&dev->dev, etsects);
+
+ return 0;
+
+no_clock:
+no_ioremap:
+ release_resource(etsects->rsrc);
+no_resource:
+ free_irq(etsects->irq, etsects);
+no_node:
+ kfree(etsects);
+no_memory:
+ return err;
+}
+
+static int gianfar_ptp_remove(struct platform_device *dev)
+{
+ struct etsects *etsects = dev_get_drvdata(&dev->dev);
+
+ gfar_write(&etsects->regs->tmr_temask, 0);
+ gfar_write(&etsects->regs->tmr_ctrl, 0);
+
+ ptp_clock_unregister(etsects->clock);
+ iounmap(etsects->regs);
+ release_resource(etsects->rsrc);
+ free_irq(etsects->irq, etsects);
+ kfree(etsects);
+
+ return 0;
+}
+
+static struct of_device_id match_table[] = {
+ { .compatible = "fsl,etsec-ptp" },
+ {},
+};
+
+static struct platform_driver gianfar_ptp_driver = {
+ .driver = {
+ .name = "gianfar_ptp",
+ .of_match_table = match_table,
+ .owner = THIS_MODULE,
+ },
+ .probe = gianfar_ptp_probe,
+ .remove = gianfar_ptp_remove,
+};
+
+/* module operations */
+
+static int __init ptp_gianfar_init(void)
+{
+ return platform_driver_register(&gianfar_ptp_driver);
+}
+
+module_init(ptp_gianfar_init);
+
+static void __exit ptp_gianfar_exit(void)
+{
+ platform_driver_unregister(&gianfar_ptp_driver);
+}
+
+module_exit(ptp_gianfar_exit);
+
+MODULE_AUTHOR("Richard Cochran <richard.cochran-3mrvs1K0uXizZXS1Dc/lvw@public.gmane.org>");
+MODULE_DESCRIPTION("PTP clock using the eTSEC");
+MODULE_LICENSE("GPL");
diff --git a/drivers/ptp/Kconfig b/drivers/ptp/Kconfig
index 70d4bb1..12eb844 100644
--- a/drivers/ptp/Kconfig
+++ b/drivers/ptp/Kconfig
@@ -27,4 +27,17 @@ config PTP_1588_CLOCK
To compile this driver as a module, choose M here: the module
will be called ptp.
+config PTP_1588_CLOCK_GIANFAR
+ tristate "Freescale eTSEC as PTP clock"
+ depends on PTP_1588_CLOCK
+ depends on GIANFAR
+ help
+ This driver adds support for using the eTSEC as a PTP
+ clock. This clock is only useful if your PTP programs are
+ getting hardware time stamps on the PTP Ethernet packets
+ using the SO_TIMESTAMPING API.
+
+ To compile this driver as a module, choose M here: the module
+ will be called gianfar_ptp.
+
endmenu
--
1.7.0.4
^ permalink raw reply related
* [PATCH V15 3/4] ptp: Added a clock driver for the IXP46x.
From: Richard Cochran @ 2011-04-22 10:04 UTC (permalink / raw)
To: linux-kernel
Cc: linux-api, netdev, devicetree-discuss, linux-arm-kernel,
linuxppc-dev, Alan Cox, Arnd Bergmann, Ben Hutchings,
Benjamin Herrenschmidt, Christoph Lameter, David Miller,
John Stultz, Krzysztof Halasa, Mike Frysinger, Paul Mackerras,
Peter Zijlstra, Rodolfo Giometti, Russell King, Thomas Gleixner
In-Reply-To: <cover.1303466102.git.richard.cochran@omicron.at>
This patch adds a driver for the hardware time stamping unit found on the
IXP465. The basic clock operations and an external trigger are implemented.
Signed-off-by: Richard Cochran <richard.cochran@omicron.at>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: David S. Miller <davem@davemloft.net>
Acked-by: John Stultz <john.stultz@linaro.org>
Acked-by: Krzysztof Halasa <khc@pm.waw.pl>
---
arch/arm/mach-ixp4xx/include/mach/ixp46x_ts.h | 78 ++++++
drivers/net/arm/ixp4xx_eth.c | 195 ++++++++++++++-
drivers/ptp/Kconfig | 13 +
drivers/ptp/Makefile | 1 +
drivers/ptp/ptp_ixp46x.c | 332 +++++++++++++++++++++++++
5 files changed, 616 insertions(+), 3 deletions(-)
create mode 100644 arch/arm/mach-ixp4xx/include/mach/ixp46x_ts.h
create mode 100644 drivers/ptp/ptp_ixp46x.c
diff --git a/arch/arm/mach-ixp4xx/include/mach/ixp46x_ts.h b/arch/arm/mach-ixp4xx/include/mach/ixp46x_ts.h
new file mode 100644
index 0000000..292d55e
--- /dev/null
+++ b/arch/arm/mach-ixp4xx/include/mach/ixp46x_ts.h
@@ -0,0 +1,78 @@
+/*
+ * PTP 1588 clock using the IXP46X
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef _IXP46X_TS_H_
+#define _IXP46X_TS_H_
+
+#define DEFAULT_ADDEND 0xF0000029
+#define TICKS_NS_SHIFT 4
+
+struct ixp46x_channel_ctl {
+ u32 ch_control; /* 0x40 Time Synchronization Channel Control */
+ u32 ch_event; /* 0x44 Time Synchronization Channel Event */
+ u32 tx_snap_lo; /* 0x48 Transmit Snapshot Low Register */
+ u32 tx_snap_hi; /* 0x4C Transmit Snapshot High Register */
+ u32 rx_snap_lo; /* 0x50 Receive Snapshot Low Register */
+ u32 rx_snap_hi; /* 0x54 Receive Snapshot High Register */
+ u32 src_uuid_lo; /* 0x58 Source UUID0 Low Register */
+ u32 src_uuid_hi; /* 0x5C Sequence Identifier/Source UUID0 High */
+};
+
+struct ixp46x_ts_regs {
+ u32 control; /* 0x00 Time Sync Control Register */
+ u32 event; /* 0x04 Time Sync Event Register */
+ u32 addend; /* 0x08 Time Sync Addend Register */
+ u32 accum; /* 0x0C Time Sync Accumulator Register */
+ u32 test; /* 0x10 Time Sync Test Register */
+ u32 unused; /* 0x14 */
+ u32 rsystime_lo; /* 0x18 RawSystemTime_Low Register */
+ u32 rsystime_hi; /* 0x1C RawSystemTime_High Register */
+ u32 systime_lo; /* 0x20 SystemTime_Low Register */
+ u32 systime_hi; /* 0x24 SystemTime_High Register */
+ u32 trgt_lo; /* 0x28 TargetTime_Low Register */
+ u32 trgt_hi; /* 0x2C TargetTime_High Register */
+ u32 asms_lo; /* 0x30 Auxiliary Slave Mode Snapshot Low */
+ u32 asms_hi; /* 0x34 Auxiliary Slave Mode Snapshot High */
+ u32 amms_lo; /* 0x38 Auxiliary Master Mode Snapshot Low */
+ u32 amms_hi; /* 0x3C Auxiliary Master Mode Snapshot High */
+
+ struct ixp46x_channel_ctl channel[3];
+};
+
+/* 0x00 Time Sync Control Register Bits */
+#define TSCR_AMM (1<<3)
+#define TSCR_ASM (1<<2)
+#define TSCR_TTM (1<<1)
+#define TSCR_RST (1<<0)
+
+/* 0x04 Time Sync Event Register Bits */
+#define TSER_SNM (1<<3)
+#define TSER_SNS (1<<2)
+#define TTIPEND (1<<1)
+
+/* 0x40 Time Synchronization Channel Control Register Bits */
+#define MASTER_MODE (1<<0)
+#define TIMESTAMP_ALL (1<<1)
+
+/* 0x44 Time Synchronization Channel Event Register Bits */
+#define TX_SNAPSHOT_LOCKED (1<<0)
+#define RX_SNAPSHOT_LOCKED (1<<1)
+
+#endif
diff --git a/drivers/net/arm/ixp4xx_eth.c b/drivers/net/arm/ixp4xx_eth.c
index 9eb9b98..de51e84 100644
--- a/drivers/net/arm/ixp4xx_eth.c
+++ b/drivers/net/arm/ixp4xx_eth.c
@@ -30,9 +30,12 @@
#include <linux/etherdevice.h>
#include <linux/io.h>
#include <linux/kernel.h>
+#include <linux/net_tstamp.h>
#include <linux/phy.h>
#include <linux/platform_device.h>
+#include <linux/ptp_classify.h>
#include <linux/slab.h>
+#include <mach/ixp46x_ts.h>
#include <mach/npe.h>
#include <mach/qmgr.h>
@@ -67,6 +70,10 @@
#define RXFREE_QUEUE(port_id) (NPE_ID(port_id) + 26)
#define TXDONE_QUEUE 31
+#define PTP_SLAVE_MODE 1
+#define PTP_MASTER_MODE 2
+#define PORT2CHANNEL(p) NPE_ID(p->id)
+
/* TX Control Registers */
#define TX_CNTRL0_TX_EN 0x01
#define TX_CNTRL0_HALFDUPLEX 0x02
@@ -171,6 +178,8 @@ struct port {
int id; /* logical port ID */
int speed, duplex;
u8 firmware[4];
+ int hwts_tx_en;
+ int hwts_rx_en;
};
/* NPE message structure */
@@ -246,6 +255,172 @@ static int ports_open;
static struct port *npe_port_tab[MAX_NPES];
static struct dma_pool *dma_pool;
+static struct sock_filter ptp_filter[] = {
+ PTP_FILTER
+};
+
+static int ixp_ptp_match(struct sk_buff *skb, u16 uid_hi, u32 uid_lo, u16 seqid)
+{
+ u8 *data = skb->data;
+ unsigned int offset;
+ u16 *hi, *id;
+ u32 lo;
+
+ if (sk_run_filter(skb, ptp_filter) != PTP_CLASS_V1_IPV4)
+ return 0;
+
+ offset = ETH_HLEN + IPV4_HLEN(data) + UDP_HLEN;
+
+ if (skb->len < offset + OFF_PTP_SEQUENCE_ID + sizeof(seqid))
+ return 0;
+
+ hi = (u16 *)(data + offset + OFF_PTP_SOURCE_UUID);
+ id = (u16 *)(data + offset + OFF_PTP_SEQUENCE_ID);
+
+ memcpy(&lo, &hi[1], sizeof(lo));
+
+ return (uid_hi == ntohs(*hi) &&
+ uid_lo == ntohl(lo) &&
+ seqid == ntohs(*id));
+}
+
+static void ixp_rx_timestamp(struct port *port, struct sk_buff *skb)
+{
+ struct skb_shared_hwtstamps *shhwtstamps;
+ struct ixp46x_ts_regs *regs;
+ u64 ns;
+ u32 ch, hi, lo, val;
+ u16 uid, seq;
+
+ if (!port->hwts_rx_en)
+ return;
+
+ ch = PORT2CHANNEL(port);
+
+ regs = (struct ixp46x_ts_regs __iomem *) IXP4XX_TIMESYNC_BASE_VIRT;
+
+ val = __raw_readl(®s->channel[ch].ch_event);
+
+ if (!(val & RX_SNAPSHOT_LOCKED))
+ return;
+
+ lo = __raw_readl(®s->channel[ch].src_uuid_lo);
+ hi = __raw_readl(®s->channel[ch].src_uuid_hi);
+
+ uid = hi & 0xffff;
+ seq = (hi >> 16) & 0xffff;
+
+ if (!ixp_ptp_match(skb, htons(uid), htonl(lo), htons(seq)))
+ goto out;
+
+ lo = __raw_readl(®s->channel[ch].rx_snap_lo);
+ hi = __raw_readl(®s->channel[ch].rx_snap_hi);
+ ns = ((u64) hi) << 32;
+ ns |= lo;
+ ns <<= TICKS_NS_SHIFT;
+
+ shhwtstamps = skb_hwtstamps(skb);
+ memset(shhwtstamps, 0, sizeof(*shhwtstamps));
+ shhwtstamps->hwtstamp = ns_to_ktime(ns);
+out:
+ __raw_writel(RX_SNAPSHOT_LOCKED, ®s->channel[ch].ch_event);
+}
+
+static void ixp_tx_timestamp(struct port *port, struct sk_buff *skb)
+{
+ struct skb_shared_hwtstamps shhwtstamps;
+ struct ixp46x_ts_regs *regs;
+ struct skb_shared_info *shtx;
+ u64 ns;
+ u32 ch, cnt, hi, lo, val;
+
+ shtx = skb_shinfo(skb);
+ if (unlikely(shtx->tx_flags & SKBTX_HW_TSTAMP && port->hwts_tx_en))
+ shtx->tx_flags |= SKBTX_IN_PROGRESS;
+ else
+ return;
+
+ ch = PORT2CHANNEL(port);
+
+ regs = (struct ixp46x_ts_regs __iomem *) IXP4XX_TIMESYNC_BASE_VIRT;
+
+ /*
+ * This really stinks, but we have to poll for the Tx time stamp.
+ * Usually, the time stamp is ready after 4 to 6 microseconds.
+ */
+ for (cnt = 0; cnt < 100; cnt++) {
+ val = __raw_readl(®s->channel[ch].ch_event);
+ if (val & TX_SNAPSHOT_LOCKED)
+ break;
+ udelay(1);
+ }
+ if (!(val & TX_SNAPSHOT_LOCKED)) {
+ shtx->tx_flags &= ~SKBTX_IN_PROGRESS;
+ return;
+ }
+
+ lo = __raw_readl(®s->channel[ch].tx_snap_lo);
+ hi = __raw_readl(®s->channel[ch].tx_snap_hi);
+ ns = ((u64) hi) << 32;
+ ns |= lo;
+ ns <<= TICKS_NS_SHIFT;
+
+ memset(&shhwtstamps, 0, sizeof(shhwtstamps));
+ shhwtstamps.hwtstamp = ns_to_ktime(ns);
+ skb_tstamp_tx(skb, &shhwtstamps);
+
+ __raw_writel(TX_SNAPSHOT_LOCKED, ®s->channel[ch].ch_event);
+}
+
+static int hwtstamp_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
+{
+ struct hwtstamp_config cfg;
+ struct ixp46x_ts_regs *regs;
+ struct port *port = netdev_priv(netdev);
+ int ch;
+
+ if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg)))
+ return -EFAULT;
+
+ if (cfg.flags) /* reserved for future extensions */
+ return -EINVAL;
+
+ ch = PORT2CHANNEL(port);
+ regs = (struct ixp46x_ts_regs __iomem *) IXP4XX_TIMESYNC_BASE_VIRT;
+
+ switch (cfg.tx_type) {
+ case HWTSTAMP_TX_OFF:
+ port->hwts_tx_en = 0;
+ break;
+ case HWTSTAMP_TX_ON:
+ port->hwts_tx_en = 1;
+ break;
+ default:
+ return -ERANGE;
+ }
+
+ switch (cfg.rx_filter) {
+ case HWTSTAMP_FILTER_NONE:
+ port->hwts_rx_en = 0;
+ break;
+ case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
+ port->hwts_rx_en = PTP_SLAVE_MODE;
+ __raw_writel(0, ®s->channel[ch].ch_control);
+ break;
+ case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
+ port->hwts_rx_en = PTP_MASTER_MODE;
+ __raw_writel(MASTER_MODE, ®s->channel[ch].ch_control);
+ break;
+ default:
+ return -ERANGE;
+ }
+
+ /* Clear out any old time stamps. */
+ __raw_writel(TX_SNAPSHOT_LOCKED | RX_SNAPSHOT_LOCKED,
+ ®s->channel[ch].ch_event);
+
+ return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0;
+}
static int ixp4xx_mdio_cmd(struct mii_bus *bus, int phy_id, int location,
int write, u16 cmd)
@@ -573,6 +748,7 @@ static int eth_poll(struct napi_struct *napi, int budget)
debug_pkt(dev, "eth_poll", skb->data, skb->len);
+ ixp_rx_timestamp(port, skb);
skb->protocol = eth_type_trans(skb, dev);
dev->stats.rx_packets++;
dev->stats.rx_bytes += skb->len;
@@ -679,14 +855,12 @@ static int eth_xmit(struct sk_buff *skb, struct net_device *dev)
return NETDEV_TX_OK;
}
memcpy_swab32(mem, (u32 *)((int)skb->data & ~3), bytes / 4);
- dev_kfree_skb(skb);
#endif
phys = dma_map_single(&dev->dev, mem, bytes, DMA_TO_DEVICE);
if (dma_mapping_error(&dev->dev, phys)) {
-#ifdef __ARMEB__
dev_kfree_skb(skb);
-#else
+#ifndef __ARMEB__
kfree(mem);
#endif
dev->stats.tx_dropped++;
@@ -728,6 +902,13 @@ static int eth_xmit(struct sk_buff *skb, struct net_device *dev)
#if DEBUG_TX
printk(KERN_DEBUG "%s: eth_xmit end\n", dev->name);
#endif
+
+ ixp_tx_timestamp(port, skb);
+ skb_tx_timestamp(skb);
+
+#ifndef __ARMEB__
+ dev_kfree_skb(skb);
+#endif
return NETDEV_TX_OK;
}
@@ -783,6 +964,9 @@ static int eth_ioctl(struct net_device *dev, struct ifreq *req, int cmd)
if (!netif_running(dev))
return -EINVAL;
+ if (cpu_is_ixp46x() && cmd == SIOCSHWTSTAMP)
+ return hwtstamp_ioctl(dev, req, cmd);
+
return phy_mii_ioctl(port->phydev, req, cmd);
}
@@ -1171,6 +1355,11 @@ static int __devinit eth_init_one(struct platform_device *pdev)
char phy_id[MII_BUS_ID_SIZE + 3];
int err;
+ if (ptp_filter_init(ptp_filter, ARRAY_SIZE(ptp_filter))) {
+ pr_err("ixp4xx_eth: bad ptp filter\n");
+ return -EINVAL;
+ }
+
if (!(dev = alloc_etherdev(sizeof(struct port))))
return -ENOMEM;
diff --git a/drivers/ptp/Kconfig b/drivers/ptp/Kconfig
index 12eb844..fcfafd0 100644
--- a/drivers/ptp/Kconfig
+++ b/drivers/ptp/Kconfig
@@ -40,4 +40,17 @@ config PTP_1588_CLOCK_GIANFAR
To compile this driver as a module, choose M here: the module
will be called gianfar_ptp.
+config PTP_1588_CLOCK_IXP46X
+ tristate "Intel IXP46x as PTP clock"
+ depends on PTP_1588_CLOCK
+ depends on IXP4XX_ETH
+ help
+ This driver adds support for using the IXP46X as a PTP
+ clock. This clock is only useful if your PTP programs are
+ getting hardware time stamps on the PTP Ethernet packets
+ using the SO_TIMESTAMPING API.
+
+ To compile this driver as a module, choose M here: the module
+ will be called ptp_ixp46x.
+
endmenu
diff --git a/drivers/ptp/Makefile b/drivers/ptp/Makefile
index 480e2af..f6933e8 100644
--- a/drivers/ptp/Makefile
+++ b/drivers/ptp/Makefile
@@ -4,3 +4,4 @@
ptp-y := ptp_clock.o ptp_chardev.o ptp_sysfs.o
obj-$(CONFIG_PTP_1588_CLOCK) += ptp.o
+obj-$(CONFIG_PTP_1588_CLOCK_IXP46X) += ptp_ixp46x.o
diff --git a/drivers/ptp/ptp_ixp46x.c b/drivers/ptp/ptp_ixp46x.c
new file mode 100644
index 0000000..803d665
--- /dev/null
+++ b/drivers/ptp/ptp_ixp46x.c
@@ -0,0 +1,332 @@
+/*
+ * PTP 1588 clock using the IXP46X
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ *
+ * 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.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/gpio.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/irq.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+
+#include <linux/ptp_clock_kernel.h>
+#include <mach/ixp46x_ts.h>
+
+#define DRIVER "ptp_ixp46x"
+#define N_EXT_TS 2
+#define MASTER_GPIO 8
+#define MASTER_IRQ 25
+#define SLAVE_GPIO 7
+#define SLAVE_IRQ 24
+
+struct ixp_clock {
+ struct ixp46x_ts_regs *regs;
+ struct ptp_clock *ptp_clock;
+ struct ptp_clock_info caps;
+ int exts0_enabled;
+ int exts1_enabled;
+};
+
+DEFINE_SPINLOCK(register_lock);
+
+/*
+ * Register access functions
+ */
+
+static u64 ixp_systime_read(struct ixp46x_ts_regs *regs)
+{
+ u64 ns;
+ u32 lo, hi;
+
+ lo = __raw_readl(®s->systime_lo);
+ hi = __raw_readl(®s->systime_hi);
+
+ ns = ((u64) hi) << 32;
+ ns |= lo;
+ ns <<= TICKS_NS_SHIFT;
+
+ return ns;
+}
+
+static void ixp_systime_write(struct ixp46x_ts_regs *regs, u64 ns)
+{
+ u32 hi, lo;
+
+ ns >>= TICKS_NS_SHIFT;
+ hi = ns >> 32;
+ lo = ns & 0xffffffff;
+
+ __raw_writel(lo, ®s->systime_lo);
+ __raw_writel(hi, ®s->systime_hi);
+}
+
+/*
+ * Interrupt service routine
+ */
+
+static irqreturn_t isr(int irq, void *priv)
+{
+ struct ixp_clock *ixp_clock = priv;
+ struct ixp46x_ts_regs *regs = ixp_clock->regs;
+ struct ptp_clock_event event;
+ u32 ack = 0, lo, hi, val;
+
+ val = __raw_readl(®s->event);
+
+ if (val & TSER_SNS) {
+ ack |= TSER_SNS;
+ if (ixp_clock->exts0_enabled) {
+ hi = __raw_readl(®s->asms_hi);
+ lo = __raw_readl(®s->asms_lo);
+ event.type = PTP_CLOCK_EXTTS;
+ event.index = 0;
+ event.timestamp = ((u64) hi) << 32;
+ event.timestamp |= lo;
+ event.timestamp <<= TICKS_NS_SHIFT;
+ ptp_clock_event(ixp_clock->ptp_clock, &event);
+ }
+ }
+
+ if (val & TSER_SNM) {
+ ack |= TSER_SNM;
+ if (ixp_clock->exts1_enabled) {
+ hi = __raw_readl(®s->amms_hi);
+ lo = __raw_readl(®s->amms_lo);
+ event.type = PTP_CLOCK_EXTTS;
+ event.index = 1;
+ event.timestamp = ((u64) hi) << 32;
+ event.timestamp |= lo;
+ event.timestamp <<= TICKS_NS_SHIFT;
+ ptp_clock_event(ixp_clock->ptp_clock, &event);
+ }
+ }
+
+ if (val & TTIPEND)
+ ack |= TTIPEND; /* this bit seems to be always set */
+
+ if (ack) {
+ __raw_writel(ack, ®s->event);
+ return IRQ_HANDLED;
+ } else
+ return IRQ_NONE;
+}
+
+/*
+ * PTP clock operations
+ */
+
+static int ptp_ixp_adjfreq(struct ptp_clock_info *ptp, s32 ppb)
+{
+ u64 adj;
+ u32 diff, addend;
+ int neg_adj = 0;
+ struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
+ struct ixp46x_ts_regs *regs = ixp_clock->regs;
+
+ if (ppb < 0) {
+ neg_adj = 1;
+ ppb = -ppb;
+ }
+ addend = DEFAULT_ADDEND;
+ adj = addend;
+ adj *= ppb;
+ diff = div_u64(adj, 1000000000ULL);
+
+ addend = neg_adj ? addend - diff : addend + diff;
+
+ __raw_writel(addend, ®s->addend);
+
+ return 0;
+}
+
+static int ptp_ixp_adjtime(struct ptp_clock_info *ptp, s64 delta)
+{
+ s64 now;
+ unsigned long flags;
+ struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
+ struct ixp46x_ts_regs *regs = ixp_clock->regs;
+
+ spin_lock_irqsave(®ister_lock, flags);
+
+ now = ixp_systime_read(regs);
+ now += delta;
+ ixp_systime_write(regs, now);
+
+ spin_unlock_irqrestore(®ister_lock, flags);
+
+ return 0;
+}
+
+static int ptp_ixp_gettime(struct ptp_clock_info *ptp, struct timespec *ts)
+{
+ u64 ns;
+ u32 remainder;
+ unsigned long flags;
+ struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
+ struct ixp46x_ts_regs *regs = ixp_clock->regs;
+
+ spin_lock_irqsave(®ister_lock, flags);
+
+ ns = ixp_systime_read(regs);
+
+ spin_unlock_irqrestore(®ister_lock, flags);
+
+ ts->tv_sec = div_u64_rem(ns, 1000000000, &remainder);
+ ts->tv_nsec = remainder;
+ return 0;
+}
+
+static int ptp_ixp_settime(struct ptp_clock_info *ptp,
+ const struct timespec *ts)
+{
+ u64 ns;
+ unsigned long flags;
+ struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
+ struct ixp46x_ts_regs *regs = ixp_clock->regs;
+
+ ns = ts->tv_sec * 1000000000ULL;
+ ns += ts->tv_nsec;
+
+ spin_lock_irqsave(®ister_lock, flags);
+
+ ixp_systime_write(regs, ns);
+
+ spin_unlock_irqrestore(®ister_lock, flags);
+
+ return 0;
+}
+
+static int ptp_ixp_enable(struct ptp_clock_info *ptp,
+ struct ptp_clock_request *rq, int on)
+{
+ struct ixp_clock *ixp_clock = container_of(ptp, struct ixp_clock, caps);
+
+ switch (rq->type) {
+ case PTP_CLK_REQ_EXTTS:
+ switch (rq->extts.index) {
+ case 0:
+ ixp_clock->exts0_enabled = on ? 1 : 0;
+ break;
+ case 1:
+ ixp_clock->exts1_enabled = on ? 1 : 0;
+ break;
+ default:
+ return -EINVAL;
+ }
+ return 0;
+ default:
+ break;
+ }
+
+ return -EOPNOTSUPP;
+}
+
+static struct ptp_clock_info ptp_ixp_caps = {
+ .owner = THIS_MODULE,
+ .name = "IXP46X timer",
+ .max_adj = 66666655,
+ .n_ext_ts = N_EXT_TS,
+ .pps = 0,
+ .adjfreq = ptp_ixp_adjfreq,
+ .adjtime = ptp_ixp_adjtime,
+ .gettime = ptp_ixp_gettime,
+ .settime = ptp_ixp_settime,
+ .enable = ptp_ixp_enable,
+};
+
+/* module operations */
+
+static struct ixp_clock ixp_clock;
+
+static int setup_interrupt(int gpio)
+{
+ int irq;
+
+ gpio_line_config(gpio, IXP4XX_GPIO_IN);
+
+ irq = gpio_to_irq(gpio);
+
+ if (NO_IRQ == irq)
+ return NO_IRQ;
+
+ if (irq_set_irq_type(irq, IRQF_TRIGGER_FALLING)) {
+ pr_err("cannot set trigger type for irq %d\n", irq);
+ return NO_IRQ;
+ }
+
+ if (request_irq(irq, isr, 0, DRIVER, &ixp_clock)) {
+ pr_err("request_irq failed for irq %d\n", irq);
+ return NO_IRQ;
+ }
+
+ return irq;
+}
+
+static void __exit ptp_ixp_exit(void)
+{
+ free_irq(MASTER_IRQ, &ixp_clock);
+ free_irq(SLAVE_IRQ, &ixp_clock);
+ ptp_clock_unregister(ixp_clock.ptp_clock);
+}
+
+static int __init ptp_ixp_init(void)
+{
+ if (!cpu_is_ixp46x())
+ return -ENODEV;
+
+ ixp_clock.regs =
+ (struct ixp46x_ts_regs __iomem *) IXP4XX_TIMESYNC_BASE_VIRT;
+
+ ixp_clock.caps = ptp_ixp_caps;
+
+ ixp_clock.ptp_clock = ptp_clock_register(&ixp_clock.caps);
+
+ if (IS_ERR(ixp_clock.ptp_clock))
+ return PTR_ERR(ixp_clock.ptp_clock);
+
+ __raw_writel(DEFAULT_ADDEND, &ixp_clock.regs->addend);
+ __raw_writel(1, &ixp_clock.regs->trgt_lo);
+ __raw_writel(0, &ixp_clock.regs->trgt_hi);
+ __raw_writel(TTIPEND, &ixp_clock.regs->event);
+
+ if (MASTER_IRQ != setup_interrupt(MASTER_GPIO)) {
+ pr_err("failed to setup gpio %d as irq\n", MASTER_GPIO);
+ goto no_master;
+ }
+ if (SLAVE_IRQ != setup_interrupt(SLAVE_GPIO)) {
+ pr_err("failed to setup gpio %d as irq\n", SLAVE_GPIO);
+ goto no_slave;
+ }
+
+ return 0;
+no_slave:
+ free_irq(MASTER_IRQ, &ixp_clock);
+no_master:
+ ptp_clock_unregister(ixp_clock.ptp_clock);
+ return -ENODEV;
+}
+
+module_init(ptp_ixp_init);
+module_exit(ptp_ixp_exit);
+
+MODULE_AUTHOR("Richard Cochran <richard.cochran@omicron.at>");
+MODULE_DESCRIPTION("PTP clock using the IXP46X timer");
+MODULE_LICENSE("GPL");
--
1.7.0.4
^ 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