* [PATCH net-next V2] tun/macvtap: use consume_skb() instead of kfree_skb() when needed
From: Jason Wang @ 2014-11-26 7:43 UTC (permalink / raw)
To: davem, netdev, linux-kernel; +Cc: mst, Jason Wang, Eric Dumazet
To be more friendly with drop monitor, we should only call kfree_skb() when
the packets were dropped and use consume_skb() in other cases.
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
Changes from V1:
- check the return value of tun/macvtap_put_user()
---
drivers/net/macvtap.c | 5 ++++-
drivers/net/tun.c | 5 ++++-
2 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
index 42a80d3..c171ab6 100644
--- a/drivers/net/macvtap.c
+++ b/drivers/net/macvtap.c
@@ -862,7 +862,10 @@ static ssize_t macvtap_do_read(struct macvtap_queue *q,
}
iov_iter_init(&iter, READ, iv, segs, len);
ret = macvtap_put_user(q, skb, &iter);
- kfree_skb(skb);
+ if (ret < 0)
+ kfree_skb(skb);
+ else
+ consume_skb(skb);
break;
}
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index ac53a73..a21c130 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -1363,7 +1363,10 @@ static ssize_t tun_do_read(struct tun_struct *tun, struct tun_file *tfile,
iov_iter_init(&iter, READ, iv, segs, len);
ret = tun_put_user(tun, tfile, skb, &iter);
- kfree_skb(skb);
+ if (ret < 0)
+ kfree_skb(skb);
+ else
+ consume_skb(skb);
return ret;
}
--
1.9.1
^ permalink raw reply related
* Re: [PATCH v2 net-next 3/3] tuntap: reduce the size of tun_struct by using flex array.
From: Pankaj Gupta @ 2014-11-26 7:40 UTC (permalink / raw)
To: David Miller
Cc: linux-kernel, netdev, jasowang, dgibson, vfalico, edumazet,
vyasevic, hkchu, xemul, therbert, bhutchings, xii, stephen, jiri,
sergei shtylyov, Michael S. Tsirkin
In-Reply-To: <20141125.132717.122094133335261929.davem@davemloft.net>
>
> From: Pankaj Gupta <pagupta@redhat.com>
> Date: Tue, 25 Nov 2014 00:04:04 +0530
>
> > This patch switches to flex array to implement the flow caches, it brings
> > several advantages:
> >
> > - Reduce the size of the tun_struct structure, which allows us to increase
> > the
> > upper limit of queues in future.
> > - Avoid higher order memory allocation. It will be useful when switching to
> > pure hashing in flow cache which may demand a larger size array in
> > future.
> >
> > After this patch, the size of tun_struct on x86_64 reduced from 8512 to
> > 328
> >
> > Signed-off-by: Jason Wang <jasowang@redhat.com>
> > Signed-off-by: Pankaj Gupta <pagupta@redhat.com>
> > Reviewed-by: David Gibson <dgibson@redhat.com>
>
> I see no reason to use flex arrays for this, you are preallocaing the
> memory so if anything flex array is adding an unnecessary level of
> redirection for every access in return for no real gains.
>
> Just allocate the thing normally using kzalloc() or whatever.
I agree. Will do the changes and submit v3.
> --
> 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 v3 09/17] bridge: add API to notify bridge driver of learned FBD on offloaded device
From: Arad, Ronen @ 2014-11-26 7:37 UTC (permalink / raw)
To: netdev@vger.kernel.org
> >>>
> >>> Is there any case where this fdb entry gets re-used and is no
> >>> longer added by an external learning? Should we clear this flag somewhere?
> >>
> >> Once the FDB entry is marked "added_by_external_learn" it stays
> >> marked as such until removed by aging cleanup process (or flushed
> >> due to interface down, etc). If aged out (and now deleted), the
> >> FDB entry may come back either by SW learn or by HW learn. If SW
> >> learn comes first, and then HW learn, HW learn will override and
> >> mark the existing FDB entry "added_by_external_learn". So there is
> >> take-over by HW but no give-back to SW. And there is no explicit
> >> clearing of the mark short of deleting the FDB entry. The mark is
> >> mostly for letting user's know which FDB entries where learned by
> >> HW and synced to the bridge's FDB.
> >
> > Thanks, makes sense now. This is probably obvious in this context,
> > but maybe it would not hurt to come up with a documentation that
> > describe the offload API, FDB entry lifetime and HW/SW ownership etc...
>
> I have an updated Documentation/networking/switchdev.txt that covers
> the swdev APIs and usage and notes, but Jiri is being stingy with it.
> Will get this out, either in v4 or follow-on patches. There is enough
> going on just with L2 offload that we're going to need some good
> documentation to guide implementers.
> --
To control the lifetime of an externally learned FDB entry, the bridge shall provide an API for the switch driver to update the freshness of externally learned entries. Otherwise, the bridge aging will age entries which are currently or frequently used by the HW.
Is this covered in the updated document?
Is this functionality planned for v4?
^ permalink raw reply
* RE: [patch net-next v3 02/17] net: make vid as a parameter for ndo_fdb_add/ndo_fdb_del
From: Arad, Ronen @ 2014-11-26 7:35 UTC (permalink / raw)
To: netdev@vger.kernel.org
>@@ -2312,7 +2313,7 @@ errout:
> int ndo_dflt_fdb_add(struct ndmsg *ndm,
> struct nlattr *tb[],
> struct net_device *dev,
>- const unsigned char *addr,
>+ const unsigned char *addr, u16 vid,
> u16 flags)
> {
> int err = -EINVAL;
>@@ -2409,7 +2439,7 @@ out:
> int ndo_dflt_fdb_del(struct ndmsg *ndm,
> struct nlattr *tb[],
> struct net_device *dev,
>- const unsigned char *addr)
>+ const unsigned char *addr, u16 vid)
> {
> int err = -EINVAL;
What is the expected role of vid in ndo_dflt_fdb_add/ndo_dflt_fdb_del?
It is ignored in both functions.
Will the vid be passed to dev_uc_add/dev_uc_add_excl/dev_uc_del and similar dev_mc_xxx functions?
Will the device address lists become VLAN-aware?
^ permalink raw reply
* Re: Multiple DSA switch on shared MII
From: Rajib Karmakar @ 2014-11-26 7:33 UTC (permalink / raw)
To: Florian Fainelli; +Cc: netdev
In-Reply-To: <CAGVrzcaG_uYpmmgC252ObHJ9Cmt91-BUtB3uOfPJHV6RhkRU1Q@mail.gmail.com>
Hello Florian,
Thanks for your reply.
On Wed, Nov 26, 2014 at 10:52 AM, Florian Fainelli <f.fainelli@gmail.com> wrote:
> 2014-11-25 20:52 GMT-08:00 Rajib Karmakar <rajibkit@gmail.com>:
>> Hello,
>>
>> I am developing a DSA driver for Marvell 6172 and need to create 2 dsa
>> switch chip (one for WAN and one for LAN).
>
> This is not typically how it is designed to work, you would register
> one dsa switch chip with the ports assignment in this data structure.
> Right now, DSA does not support a dual Ethernet MAC configuration,
> although you could probably do per-port VLAN membership and create two
> default VLANs to allow that.
>
>>
>> My device has 2 MACs and one shared mii_bus. I have added two
>> dsa_platform_data structures and registered them but cannot probe as
>> dsa_probe tries mdio_register() on the same bus and fails.
>>
>> Is it possible to create two DSA switch on same (shared) mii?
>
> Not in its current form, and I am not exactly sure how we would support that.
Yes, not with the current DSA implementation, but I can manage to
solve this by a small (ugly) patch - renamed the slave mii bus as
"<master_bus->id>:<pd->sw_addr>:<platform_device->id>" instead of
"<master_bus->id>:<pd->sw_addr>". Though I am not yet sure enough if
this could have any negative impact or not.
Comments please.
>
>>
>> Regards,
>> Rajib
>> --
>> 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
>
>
>
> --
> Florian
----
diff -rup org/net/dsa/dsa.c mod/net/dsa/dsa.c
--- org/net/dsa/dsa.c
+++ mod/net/dsa/dsa.c
@@ -68,7 +68,7 @@ dsa_switch_probe(struct mii_bus *bus, in
/* basic switch operations **************************************************/
static struct dsa_switch *
-dsa_switch_setup(struct dsa_switch_tree *dst, int index,
+dsa_switch_setup(struct dsa_switch_tree *dst, int index, int id,
struct device *parent, struct mii_bus *bus)
{
struct dsa_chip_data *pd = dst->pd->chip + index;
@@ -156,7 +156,7 @@ dsa_switch_setup(struct dsa_switch_tree
ret = -ENOMEM;
goto out;
}
- dsa_slave_mii_bus_init(ds);
+ dsa_slave_mii_bus_init(ds, id);
ret = mdiobus_register(ds->slave_mii_bus);
if (ret < 0)
@@ -349,7 +349,7 @@ static int dsa_probe(struct platform_dev
continue;
}
- ds = dsa_switch_setup(dst, i, &pdev->dev, bus);
+ ds = dsa_switch_setup(dst, i, pdev->id, &pdev->dev, bus);
if (IS_ERR(ds)) {
printk(KERN_ERR "%s[%d]: couldn't create dsa switch "
"instance (error %ld)\n", dev->name, i,
diff -rup org/net/dsa/dsa_priv.h mod/net/dsa/dsa_priv.h
--- org/net/dsa/dsa_priv.h
+++ mod/net/dsa/dsa_priv.h
@@ -163,7 +163,7 @@ void register_switch_driver(struct dsa_s
void unregister_switch_driver(struct dsa_switch_driver *type);
/* slave.c */
-void dsa_slave_mii_bus_init(struct dsa_switch *ds);
+void dsa_slave_mii_bus_init(struct dsa_switch *ds, int id);
struct net_device *dsa_slave_create(struct dsa_switch *ds,
struct device *parent,
int port, char *name);
diff -rup org/net/dsa/slave.c mod/net/dsa/slave.c
--- org/net/dsa/slave.c
+++ mod/net/dsa/slave.c
@@ -35,14 +35,14 @@ static int dsa_slave_phy_write(struct mi
return 0;
}
-void dsa_slave_mii_bus_init(struct dsa_switch *ds)
+void dsa_slave_mii_bus_init(struct dsa_switch *ds, int id)
{
ds->slave_mii_bus->priv = (void *)ds;
ds->slave_mii_bus->name = "dsa slave smi";
ds->slave_mii_bus->read = dsa_slave_phy_read;
ds->slave_mii_bus->write = dsa_slave_phy_write;
- snprintf(ds->slave_mii_bus->id, MII_BUS_ID_SIZE, "%s:%.2x",
- ds->master_mii_bus->id, ds->pd->sw_addr);
+ snprintf(ds->slave_mii_bus->id, MII_BUS_ID_SIZE, "%s:%.2x:%.1x",
+ ds->master_mii_bus->id, ds->pd->sw_addr, id);
ds->slave_mii_bus->parent = &ds->master_mii_bus->dev;
}
^ permalink raw reply
* Re: [PATCH 1/2] virito: introduce methods of fixing device features
From: Jason Wang @ 2014-11-26 7:09 UTC (permalink / raw)
To: Rusty Russell; +Cc: netdev, virtualization, linux-kernel, mst
In-Reply-To: <87sih9xv5u.fsf@rustcorp.com.au>
----- Original Message -----
> Jason Wang <jasowang@redhat.com> writes:
> > Buggy host may advertised buggy host features (a usual case is that host
> > advertise a feature whose dependencies were missed). In this case, driver
> > should detect and disable the buggy features by itself.
>
> Sorry, I've been focussing elsewhere.
>
> I would really prefer that drivers offer a "feature_depends" table,
> which can indicate that feature A depends on feature B, and have the
> core iterate, complain and fixup as necessary.
>
> Is that expressive enough, or do we need more?
>
> Thanks,
> Rusty.
I suggested something like this in https://lkml.org/lkml/2014/11/19/132, but
Michael does not like it ...
^ permalink raw reply
* Re: [PATCH net-next v4] ipvlan: Initial check-in of the IPVLAN driver.
From: Toshiaki Makita @ 2014-11-26 6:41 UTC (permalink / raw)
To: Mahesh Bandewar, netdev
Cc: Eric Dumazet, Maciej Zenczykowski, Laurent Chavey, Tim Hockin,
David Miller, Brandon Philips, Pavel Emelianov
In-Reply-To: <1416812866-13401-1-git-send-email-maheshb@google.com>
Hi Mahesh,
I found that deleting the last ipvlan device triggers WARN_ON() in
rtmsg_ifinfo().
ipvlan_nl_fillinfo() seems to return -EINVAL in that case.
> +static int ipvlan_nl_fillinfo(struct sk_buff *skb,
> + const struct net_device *dev)
> +{
> + struct ipvl_dev *ipvlan = netdev_priv(dev);
> + struct ipvl_port *port = ipvlan_port_get_rtnl(ipvlan->phy_dev);
> + int ret = -EINVAL;
> +
> + if (!port)
> + goto err;
> +
> + ret = -EMSGSIZE;
> + if (nla_put_u16(skb, IFLA_IPVLAN_MODE, port->mode))
> + goto err;
> +
> + return 0;
> +
> +err:
> + return ret;
> +}
rollback_registered_many() calls rtmsg_ifinfo() after calling ndo_uninit().
ndo_uninit() (ipvlan_uninit() -> ipvlan_port_destroy() ->
netdev_rx_handler_unregister()) sets rx_handler_data into NULL.
So, we cannot dereference "port" in ipvlan_nl_fillinfo().
Maybe "mode" should belong to struct ipvl_dev?
Thanks,
Toshiaki Makita
^ permalink raw reply
* Re: [PATCH net-next] tun/macvtap: use consume_skb() instead of kfree_skb() in tun/macvtap_do_read()
From: Jason Wang @ 2014-11-26 6:41 UTC (permalink / raw)
To: Eric Dumazet; +Cc: davem, netdev, linux-kernel, mst
In-Reply-To: <1416983160.29427.52.camel@edumazet-glaptop2.roam.corp.google.com>
On 11/26/2014 02:26 PM, Eric Dumazet wrote:
> On Wed, 2014-11-26 at 14:08 +0800, Jason Wang wrote:
>> To be more friendly with drop monitor, we should use comsume_skb() instead
>> of kfree_skb() in tun/macvtap_do_read(). Otherwise, the packets will be
>> marked as dropped.
>>
>> Signed-off-by: Jason Wang <jasowang@redhat.com>
>> ---
>> drivers/net/macvtap.c | 2 +-
>> drivers/net/tun.c | 2 +-
>> 2 files changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
>> index 42a80d3..74d1b23 100644
>> --- a/drivers/net/macvtap.c
>> +++ b/drivers/net/macvtap.c
>> @@ -862,7 +862,7 @@ static ssize_t macvtap_do_read(struct macvtap_queue *q,
>> }
>> iov_iter_init(&iter, READ, iv, segs, len);
>> ret = macvtap_put_user(q, skb, &iter);
>> - kfree_skb(skb);
>> + consume_skb(skb);
>> break;
>> }
>>
>> diff --git a/drivers/net/tun.c b/drivers/net/tun.c
>> index ac53a73..d7edeaf 100644
>> --- a/drivers/net/tun.c
>> +++ b/drivers/net/tun.c
>> @@ -1363,7 +1363,7 @@ static ssize_t tun_do_read(struct tun_struct *tun, struct tun_file *tfile,
>>
>> iov_iter_init(&iter, READ, iv, segs, len);
>> ret = tun_put_user(tun, tfile, skb, &iter);
>> - kfree_skb(skb);
>> + consume_skb(skb);
>>
>> return ret;
>> }
>
> These patches keep going on, but they are wrong.
>
> If you care about drop monitor, then you want to not hide drops but
> precisely report them.
>
> Surely tun_put_user() can return an error, and then packet _is_ dropped.
True, I miss this. Will post V2. Thanks
^ permalink raw reply
* Re: [PATCH net-next] tun/macvtap: use consume_skb() instead of kfree_skb() in tun/macvtap_do_read()
From: Eric Dumazet @ 2014-11-26 6:26 UTC (permalink / raw)
To: Jason Wang; +Cc: davem, netdev, linux-kernel, mst
In-Reply-To: <1416982084-12212-1-git-send-email-jasowang@redhat.com>
On Wed, 2014-11-26 at 14:08 +0800, Jason Wang wrote:
> To be more friendly with drop monitor, we should use comsume_skb() instead
> of kfree_skb() in tun/macvtap_do_read(). Otherwise, the packets will be
> marked as dropped.
>
> Signed-off-by: Jason Wang <jasowang@redhat.com>
> ---
> drivers/net/macvtap.c | 2 +-
> drivers/net/tun.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
> index 42a80d3..74d1b23 100644
> --- a/drivers/net/macvtap.c
> +++ b/drivers/net/macvtap.c
> @@ -862,7 +862,7 @@ static ssize_t macvtap_do_read(struct macvtap_queue *q,
> }
> iov_iter_init(&iter, READ, iv, segs, len);
> ret = macvtap_put_user(q, skb, &iter);
> - kfree_skb(skb);
> + consume_skb(skb);
> break;
> }
>
> diff --git a/drivers/net/tun.c b/drivers/net/tun.c
> index ac53a73..d7edeaf 100644
> --- a/drivers/net/tun.c
> +++ b/drivers/net/tun.c
> @@ -1363,7 +1363,7 @@ static ssize_t tun_do_read(struct tun_struct *tun, struct tun_file *tfile,
>
> iov_iter_init(&iter, READ, iv, segs, len);
> ret = tun_put_user(tun, tfile, skb, &iter);
> - kfree_skb(skb);
> + consume_skb(skb);
>
> return ret;
> }
These patches keep going on, but they are wrong.
If you care about drop monitor, then you want to not hide drops but
precisely report them.
Surely tun_put_user() can return an error, and then packet _is_ dropped.
^ permalink raw reply
* [PATCH net-next] tun/macvtap: use consume_skb() instead of kfree_skb() in tun/macvtap_do_read()
From: Jason Wang @ 2014-11-26 6:08 UTC (permalink / raw)
To: davem, netdev, linux-kernel; +Cc: mst, Jason Wang
To be more friendly with drop monitor, we should use comsume_skb() instead
of kfree_skb() in tun/macvtap_do_read(). Otherwise, the packets will be
marked as dropped.
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
drivers/net/macvtap.c | 2 +-
drivers/net/tun.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
index 42a80d3..74d1b23 100644
--- a/drivers/net/macvtap.c
+++ b/drivers/net/macvtap.c
@@ -862,7 +862,7 @@ static ssize_t macvtap_do_read(struct macvtap_queue *q,
}
iov_iter_init(&iter, READ, iv, segs, len);
ret = macvtap_put_user(q, skb, &iter);
- kfree_skb(skb);
+ consume_skb(skb);
break;
}
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index ac53a73..d7edeaf 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -1363,7 +1363,7 @@ static ssize_t tun_do_read(struct tun_struct *tun, struct tun_file *tfile,
iov_iter_init(&iter, READ, iv, segs, len);
ret = tun_put_user(tun, tfile, skb, &iter);
- kfree_skb(skb);
+ consume_skb(skb);
return ret;
}
--
1.9.1
^ permalink raw reply related
* [PATCH net-next] ipvlan: fix sparse warnings
From: Mahesh Bandewar @ 2014-11-26 5:24 UTC (permalink / raw)
To: netdev; +Cc: David Miller, Mahesh Bandewar
Fix sparse warnings reported by kbuild robot
drivers/net/ipvlan/ipvlan_main.c:172:13: warning: symbol 'ipvlan_start_xmit' was not declared. Should it be static?
drivers/net/ipvlan/ipvlan_main.c:256:33: warning: incorrect type in initializer (different address spaces)
drivers/net/ipvlan/ipvlan_main.c:256:33: expected void const [noderef] <asn:3>*__vpp_verify
drivers/net/ipvlan/ipvlan_main.c:256:33: got struct ipvl_pcpu_stats *<noident>
drivers/net/ipvlan/ipvlan_main.c:544:5: warning: symbol 'ipvlan_link_register' was not declared. Should it be static
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Mahesh Bandewar <maheshb@google.com>
---
drivers/net/ipvlan/ipvlan_main.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c
index c3df84bd2857..96b71b0d78f6 100644
--- a/drivers/net/ipvlan/ipvlan_main.c
+++ b/drivers/net/ipvlan/ipvlan_main.c
@@ -169,7 +169,8 @@ static int ipvlan_stop(struct net_device *dev)
return 0;
}
-netdev_tx_t ipvlan_start_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t ipvlan_start_xmit(struct sk_buff *skb,
+ struct net_device *dev)
{
const struct ipvl_dev *ipvlan = netdev_priv(dev);
int skblen = skb->len;
@@ -541,7 +542,7 @@ static struct rtnl_link_ops ipvlan_link_ops = {
.dellink = ipvlan_link_delete,
};
-int ipvlan_link_register(struct rtnl_link_ops *ops)
+static int ipvlan_link_register(struct rtnl_link_ops *ops)
{
return rtnl_link_register(ops);
}
--
2.2.0.rc0.207.ga3a616c
^ permalink raw reply related
* Re: Multiple DSA switch on shared MII
From: Florian Fainelli @ 2014-11-26 5:22 UTC (permalink / raw)
To: Rajib Karmakar; +Cc: netdev
In-Reply-To: <CAOi_9k8NwpVFxg=9VOagVt3m6hn2uLKuCZ_0-+4PciJNRWuTmA@mail.gmail.com>
2014-11-25 20:52 GMT-08:00 Rajib Karmakar <rajibkit@gmail.com>:
> Hello,
>
> I am developing a DSA driver for Marvell 6172 and need to create 2 dsa
> switch chip (one for WAN and one for LAN).
This is not typically how it is designed to work, you would register
one dsa switch chip with the ports assignment in this data structure.
Right now, DSA does not support a dual Ethernet MAC configuration,
although you could probably do per-port VLAN membership and create two
default VLANs to allow that.
>
> My device has 2 MACs and one shared mii_bus. I have added two
> dsa_platform_data structures and registered them but cannot probe as
> dsa_probe tries mdio_register() on the same bus and fails.
>
> Is it possible to create two DSA switch on same (shared) mii?
Not in its current form, and I am not exactly sure how we would support that.
>
> Regards,
> Rajib
> --
> 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
--
Florian
^ permalink raw reply
* [PATCH net-next 3/3] net: bcmgenet: support restarting auto-negotiation
From: Florian Fainelli @ 2014-11-26 5:16 UTC (permalink / raw)
To: netdev; +Cc: davem, pgynther, Florian Fainelli
In-Reply-To: <1416978996-24808-1-git-send-email-f.fainelli@gmail.com>
Hook a nway_reset ethtool callback to allow restarting the
auto-negotiation process when asked to. We defer to the PHY library call
to do the heavy lifting.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/ethernet/broadcom/genet/bcmgenet.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index bf44e0e23799..f2fadb053d52 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -799,6 +799,13 @@ static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_eee *e)
return phy_ethtool_set_eee(priv->phydev, e);
}
+static int bcmgenet_nway_reset(struct net_device *dev)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+
+ return genphy_restart_aneg(priv->phydev);
+}
+
/* standard ethtool support functions. */
static struct ethtool_ops bcmgenet_ethtool_ops = {
.get_strings = bcmgenet_get_strings,
@@ -814,6 +821,7 @@ static struct ethtool_ops bcmgenet_ethtool_ops = {
.set_wol = bcmgenet_set_wol,
.get_eee = bcmgenet_get_eee,
.set_eee = bcmgenet_set_eee,
+ .nway_reset = bcmgenet_nway_reset,
};
/* Power down the unimac, based on mode. */
--
2.1.0
^ permalink raw reply related
* [PATCH net-next 2/3] net: bcmgenet: add EEE support
From: Florian Fainelli @ 2014-11-26 5:16 UTC (permalink / raw)
To: netdev; +Cc: davem, pgynther, Florian Fainelli
In-Reply-To: <1416978996-24808-1-git-send-email-f.fainelli@gmail.com>
Allow enabling and disabling EEE using the designated ethtool getters
and setters. GENET allows controlling EEE at the UniMAC, RBUF and TBUF
levels. We also take care of restoring EEE after a suspend/resume cycle
if it was enabled prior to suspending.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/ethernet/broadcom/genet/bcmgenet.c | 96 ++++++++++++++++++++++++++
drivers/net/ethernet/broadcom/genet/bcmgenet.h | 4 ++
2 files changed, 100 insertions(+)
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index fcbf1255ae5a..bf44e0e23799 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -714,6 +714,91 @@ static void bcmgenet_get_ethtool_stats(struct net_device *dev,
}
}
+static void bcmgenet_eee_enable_set(struct net_device *dev, bool enable)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ u32 off = priv->hw_params->tbuf_offset + TBUF_ENERGY_CTRL;
+ u32 reg;
+
+ if (enable && !priv->clk_eee_enabled) {
+ clk_prepare_enable(priv->clk_eee);
+ priv->clk_eee_enabled = true;
+ }
+
+ reg = bcmgenet_umac_readl(priv, UMAC_EEE_CTRL);
+ if (enable)
+ reg |= EEE_EN;
+ else
+ reg &= ~EEE_EN;
+ bcmgenet_umac_writel(priv, reg, UMAC_EEE_CTRL);
+
+ /* Enable EEE and switch to a 27Mhz clock automatically */
+ reg = __raw_readl(priv->base + off);
+ if (enable)
+ reg |= TBUF_EEE_EN | TBUF_PM_EN;
+ else
+ reg &= ~(TBUF_EEE_EN | TBUF_PM_EN);
+ __raw_writel(reg, priv->base + off);
+
+ /* Do the same for thing for RBUF */
+ reg = bcmgenet_rbuf_readl(priv, RBUF_ENERGY_CTRL);
+ if (enable)
+ reg |= RBUF_EEE_EN | RBUF_PM_EN;
+ else
+ reg &= ~(RBUF_EEE_EN | RBUF_PM_EN);
+ bcmgenet_rbuf_writel(priv, reg, RBUF_ENERGY_CTRL);
+
+ if (!enable && priv->clk_eee_enabled) {
+ clk_disable_unprepare(priv->clk_eee);
+ priv->clk_eee_enabled = false;
+ }
+
+ priv->eee.eee_enabled = enable;
+ priv->eee.eee_active = enable;
+}
+
+static int bcmgenet_get_eee(struct net_device *dev, struct ethtool_eee *e)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ struct ethtool_eee *p = &priv->eee;
+
+ if (GENET_IS_V1(priv))
+ return -EOPNOTSUPP;
+
+ e->eee_enabled = p->eee_enabled;
+ e->eee_active = p->eee_active;
+ e->tx_lpi_timer = bcmgenet_umac_readl(priv, UMAC_EEE_LPI_TIMER);
+
+ return phy_ethtool_get_eee(priv->phydev, e);
+}
+
+static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_eee *e)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ struct ethtool_eee *p = &priv->eee;
+ int ret = 0;
+
+ if (GENET_IS_V1(priv))
+ return -EOPNOTSUPP;
+
+ p->eee_enabled = e->eee_enabled;
+
+ if (!p->eee_enabled) {
+ bcmgenet_eee_enable_set(dev, false);
+ } else {
+ ret = phy_init_eee(priv->phydev, 0);
+ if (ret) {
+ netif_err(priv, hw, dev, "EEE initialization failed\n");
+ return ret;
+ }
+
+ bcmgenet_umac_writel(priv, e->tx_lpi_timer, UMAC_EEE_LPI_TIMER);
+ bcmgenet_eee_enable_set(dev, true);
+ }
+
+ return phy_ethtool_set_eee(priv->phydev, e);
+}
+
/* standard ethtool support functions. */
static struct ethtool_ops bcmgenet_ethtool_ops = {
.get_strings = bcmgenet_get_strings,
@@ -727,6 +812,8 @@ static struct ethtool_ops bcmgenet_ethtool_ops = {
.set_msglevel = bcmgenet_set_msglevel,
.get_wol = bcmgenet_get_wol,
.set_wol = bcmgenet_set_wol,
+ .get_eee = bcmgenet_get_eee,
+ .set_eee = bcmgenet_set_eee,
};
/* Power down the unimac, based on mode. */
@@ -2585,6 +2672,12 @@ static int bcmgenet_probe(struct platform_device *pdev)
if (IS_ERR(priv->clk_wol))
dev_warn(&priv->pdev->dev, "failed to get enet-wol clock\n");
+ priv->clk_eee = devm_clk_get(&priv->pdev->dev, "enet-eee");
+ if (IS_ERR(priv->clk_eee)) {
+ dev_warn(&priv->pdev->dev, "failed to get enet-eee clock\n");
+ priv->clk_eee = NULL;
+ }
+
err = reset_umac(priv);
if (err)
goto err_clk_disable;
@@ -2735,6 +2828,9 @@ static int bcmgenet_resume(struct device *d)
phy_resume(priv->phydev);
+ if (priv->eee.eee_enabled)
+ bcmgenet_eee_enable_set(dev, true);
+
bcmgenet_netif_start(dev);
return 0;
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.h b/drivers/net/ethernet/broadcom/genet/bcmgenet.h
index ea6f0ab566bf..b36ddec0cc0a 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h
@@ -573,6 +573,8 @@ struct bcmgenet_priv {
struct device_node *phy_dn;
struct mii_bus *mii_bus;
u16 gphy_rev;
+ struct clk *clk_eee;
+ bool clk_eee_enabled;
/* PHY device variables */
int old_link;
@@ -609,6 +611,8 @@ struct bcmgenet_priv {
u32 wolopts;
struct bcmgenet_mib_counters mib;
+
+ struct ethtool_eee eee;
};
#define GENET_IO_MACRO(name, offset) \
--
2.1.0
^ permalink raw reply related
* [PATCH net-next 1/3] net: bcmgenet: add register definitions for EEE
From: Florian Fainelli @ 2014-11-26 5:16 UTC (permalink / raw)
To: netdev; +Cc: davem, pgynther, Florian Fainelli
In-Reply-To: <1416978996-24808-1-git-send-email-f.fainelli@gmail.com>
Add register definitions to control EEE in the UniMAC, RBUF and TBUF
register ranges.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
---
drivers/net/ethernet/broadcom/genet/bcmgenet.h | 22 ++++++++++++++++++++++
1 file changed, 22 insertions(+)
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.h b/drivers/net/ethernet/broadcom/genet/bcmgenet.h
index c4ca7282a601..ea6f0ab566bf 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h
@@ -185,6 +185,21 @@ struct bcmgenet_mib_counters {
#define UMAC_MAC1 0x010
#define UMAC_MAX_FRAME_LEN 0x014
+#define UMAC_EEE_CTRL 0x064
+#define EN_LPI_RX_PAUSE (1 << 0)
+#define EN_LPI_TX_PFC (1 << 1)
+#define EN_LPI_TX_PAUSE (1 << 2)
+#define EEE_EN (1 << 3)
+#define RX_FIFO_CHECK (1 << 4)
+#define EEE_TX_CLK_DIS (1 << 5)
+#define DIS_EEE_10M (1 << 6)
+#define LP_IDLE_PREDICTION_MODE (1 << 7)
+
+#define UMAC_EEE_LPI_TIMER 0x068
+#define UMAC_EEE_WAKE_TIMER 0x06C
+#define UMAC_EEE_REF_COUNT 0x070
+#define EEE_REFERENCE_COUNT_MASK 0xffff
+
#define UMAC_TX_FLUSH 0x334
#define UMAC_MIB_START 0x400
@@ -232,6 +247,10 @@ struct bcmgenet_mib_counters {
#define RBUF_RXCHK_EN (1 << 0)
#define RBUF_SKIP_FCS (1 << 4)
+#define RBUF_ENERGY_CTRL 0x9c
+#define RBUF_EEE_EN (1 << 0)
+#define RBUF_PM_EN (1 << 1)
+
#define RBUF_TBUF_SIZE_CTRL 0xb4
#define RBUF_HFB_CTRL_V1 0x38
@@ -247,6 +266,9 @@ struct bcmgenet_mib_counters {
#define TBUF_CTRL 0x00
#define TBUF_BP_MC 0x0C
+#define TBUF_ENERGY_CTRL 0x14
+#define TBUF_EEE_EN (1 << 0)
+#define TBUF_PM_EN (1 << 1)
#define TBUF_CTRL_V1 0x80
#define TBUF_BP_MC_V1 0xA0
--
2.1.0
^ permalink raw reply related
* [PATCH net-next 0/3] net: bcmgenet: EEE support
From: Florian Fainelli @ 2014-11-26 5:16 UTC (permalink / raw)
To: netdev; +Cc: davem, pgynther, Florian Fainelli
Hi David,
This patch series add EEE support to the Broadcom GENET driver, first patch
basically adds register definitions for the hardware, the second patch does the
actual implementation and hooks the {get,set}_eee ethtool callbacks, the last
patch adds auto-negotiation restart capability.
Thanks!
Florian Fainelli (3):
net: bcmgenet: add register definitions for EEE
net: bcmgenet: add EEE support
net: bcmgenet: support restarting auto-negotiation
drivers/net/ethernet/broadcom/genet/bcmgenet.c | 104 +++++++++++++++++++++++++
drivers/net/ethernet/broadcom/genet/bcmgenet.h | 26 +++++++
2 files changed, 130 insertions(+)
--
2.1.0
^ permalink raw reply
* Multiple DSA switch on shared MII
From: Rajib Karmakar @ 2014-11-26 4:52 UTC (permalink / raw)
To: netdev
Hello,
I am developing a DSA driver for Marvell 6172 and need to create 2 dsa
switch chip (one for WAN and one for LAN).
My device has 2 MACs and one shared mii_bus. I have added two
dsa_platform_data structures and registered them but cannot probe as
dsa_probe tries mdio_register() on the same bus and fails.
Is it possible to create two DSA switch on same (shared) mii?
Regards,
Rajib
^ permalink raw reply
* Re: [patch net-next v3 04/17] net: introduce generic switch devices support
From: Scott Feldman @ 2014-11-26 4:18 UTC (permalink / raw)
To: Jamal Hadi Salim
Cc: Thomas Graf, Jiri Pirko, Netdev, David S. Miller,
nhorman@tuxdriver.com, Andy Gospodarek, dborkman@redhat.com,
ogerlitz@mellanox.com, jesse@nicira.com, pshelar@nicira.com,
azhou@nicira.com, ben@decadent.org.uk, stephen@networkplumber.org,
Kirsher, Jeffrey T, vyasevic@redhat.com, Cong Wang,
Fastabend, John R, Eric Dumazet, Florian Fainelli, Roopa Prabhu,
John Linville
In-Reply-To: <54754A10.6020601@mojatatu.com>
On Tue, Nov 25, 2014 at 5:33 PM, Jamal Hadi Salim <jhs@mojatatu.com> wrote:
> On 11/25/14 16:54, Thomas Graf wrote:
>>
>> On 11/25/14 at 12:08pm, Jamal Hadi Salim wrote:
>
>
>> It would definitely help if you could expose some more details on the
>> "some network processor" you have. We're all very eager ;-)
>>
>
> Well, this thing doesnt run ovs ;-> (/me runs). If you come
> to netdev i may let you play with it ;-> Its a humongous device
> (think multi 100G ports).
>
> On a serious note: Even if you took what Simon/Netronome has
> (yes, I know they use ovs;->) - there is really no need for a switch
> abstraction *at all* if all you want to is hang a packet
> processing graph that ingresses at a port and egress at another port.
> As you know, Linux supports it just fine with tc.
You have a pointer to the kernel driver for that HW? Can you show how
you're using Linux tc netlink msg in kernel to program HW? I'd like
to see the in-kernel API.
^ permalink raw reply
* Re: [PATCH 0/4] support GMAC driver for RK3288
From: Kever Yang @ 2014-11-26 4:05 UTC (permalink / raw)
To: Roger Chen, heiko
Cc: peppe.cavallaro, netdev, linux-kernel, linux-rockchip, mark.yao,
eddie.cai
In-Reply-To: <1416906460-26078-1-git-send-email-roger.chen@rock-chips.com>
Hi Roger,
Please use the --in-reply-to option for patches other than 0/4
with the git send-email.
On 11/25/2014 05:07 PM, Roger Chen wrote:
> Roger Chen (4):
> patch1: add driver for Rockchip RK3288 SoCs integrated GMAC
> patch2: modify CRU config for Rockchip RK3288 SoCs integrated GMAC
> patch3: dts: add gmac info for rk3288
> patch4: dts: enable gmac on RK3288 evb board
>
> tested on rk3288 SDK board:
> busybox ifconfig eth0 up
> busybox ifconfig eth0 192.168.1.111
> ping 192.168.1.1
>
^ permalink raw reply
* Re: [net-next PATCH 2/5] ethernet/intel: Use eth_skb_pad helper
From: Eric Dumazet @ 2014-11-26 4:01 UTC (permalink / raw)
To: David Miller
Cc: alexander.duyck, alexander.h.duyck, netdev, jeffrey.t.kirsher
In-Reply-To: <20141125.221900.653470310152197801.davem@davemloft.net>
On Tue, 2014-11-25 at 22:19 -0500, David Miller wrote:
> From: Eric Dumazet <eric.dumazet@gmail.com>
> Date: Tue, 25 Nov 2014 17:43:05 -0800
>
> > I believe I finally have an idea why we had various + 15 in skb
> > allocations in TCP stack !
>
> It was so that you could do one level of tunneling with "for
> free". Or that is my recollection.
>
> Those + 15 existed way before any of these padto() calls even
> existed.
Well, tunneling is added in front of the packet. Thats why we use
MAX_TCP_HEADER.
The +15 is in fact because TCP stack wanted to make sure the eventual
padding (needing tailroom, not headroom) was possible...
Note that ack packets never used the +15, but other packets did.
^ permalink raw reply
* Re: [patch net-next v3 02/17] net: make vid as a parameter for ndo_fdb_add/ndo_fdb_del
From: Scott Feldman @ 2014-11-26 3:59 UTC (permalink / raw)
To: Jamal Hadi Salim
Cc: John Fastabend, Jiri Pirko, Netdev, David S. Miller,
nhorman@tuxdriver.com, Andy Gospodarek, Thomas Graf,
dborkman@redhat.com, ogerlitz@mellanox.com, jesse@nicira.com,
pshelar@nicira.com, azhou@nicira.com, ben@decadent.org.uk,
stephen@networkplumber.org, Kirsher, Jeffrey T,
vyasevic@redhat.com, Cong Wang, Eric Dumazet, Florian Fainelli,
Roopa Prabhu, John Linville
In-Reply-To: <547546C5.3060207@mojatatu.com>
On Tue, Nov 25, 2014 at 5:19 PM, Jamal Hadi Salim <jhs@mojatatu.com> wrote:
> On 11/25/14 21:36, Scott Feldman wrote:
>>
>> On Tue, Nov 25, 2014 at 6:50 AM, Jamal Hadi Salim <jhs@mojatatu.com>
>> wrote:
>>>
>>> On 11/25/14 11:30, John Fastabend wrote:
>
>
>>> Ok, guess i am gonna have to go stare at the code some more.
>>> I thought we returned one of the error codes?
>>> A bitmask would work for a single entry - because you have two
>>> options add to h/ware and/or s/ware. So response is easy to encode.
>>> But if i have 1000 and they are sparsely populated (think an indexed
>>> table and i have indices 1, 23, 45, etc), then a bitmask would be
>>> hard to use.
>>
>>
>> I'm confused by this discussion.
>
>
> This is about the policy which states "install as many as you can, dont
> worry about failures". In such a case, how do you tell user space back
> "oh, btw you know your request #1, #23, and 45 went ok, but nothing else
> worked". A simple return code wont work. You could return a code to
> say "some worked". At which case user space could dump and find out only
> #1, #23 and #45 worked.
You request for what? That's my confusion. Are you trying to install
FDB entry into both SW and HW at same time? And then do a bunch in a
batch? I'm saying use MASTER for SW and SELF for HW in two steps, if
you want FDB entry installed in both Sw and HW. Check your return
code each step. Batch all to HW first, then batch all that PASSED to
SW. I don't even know really why you're trying to install to both HW
and SW. Install it to HW and be done. fdb_dump will set HW entries
via SELF.
> Your question below is a different context; Some people may want
> the policy where whats in hardware
> a) gets to be seen in software and b) allow for destination lookup
> failures in hardware to show up in the kernel, refresh the fdb in the
> kernel via learning
> and c) whats in s.ware gets synced to hardware just because there's
> space in the hardware
> I dont want any of the above;-> Which would work if we had policy knobs.
> Learning, flooding, sync from hardware. Speaking of the last one:
> Where is my cookie Scott? I want my cookie.
Ah, Jamal, look again at patches 13-17/17 in last v3 set. That was a
big steaming snickerdoodle just for you! Now you can push policy
knobs down to port driver and or bridge to fine tune what ever you
want. You'll find knobs for learning, flooding, learning sync to hw,
etc. I thought you even ACKed some of these. a) above knob is 14/17
patch, b) above is using existing learning knob on bridge, c) above I
don't get...no point in syncing that direction.
> cheers,
> jamal
>
>
>
>> Do I have this right: You want to
>> send 1000 RTM_NEWNEIGHs to PF_BRIDGE with both NTF_MASTER and NTF_SELF
>> set such that 1000 new FBD entries are installed in both (SW) the
>> bridge's FDB and (HW) the port driver's FDB. My first confusion is
>> why do you want these FBD entries in bridge's FDB? We're offloading
>> the switching to HW so HW should be handling fwd plane. If ctrl pkt
>> make it to SW, it can learn those FDB entries;
>> no need for manual
>> install of FDB entry in SW. It seems to me you only want to use
>> NTF_SELF to install the FDB entry in HW using the port driver. And an
>> error code is returned for that install. Since there is only one
>> target (NTF_SELF) there is no need for bitmask return.
>
>
>
>
^ permalink raw reply
* [PATCH net-next 02/11] tipc: remove protocol message queue
From: Ying Xue @ 2014-11-26 3:41 UTC (permalink / raw)
To: davem
Cc: jon.maloy, Paul.Gortmaker, erik.hugne, richard.alpe, netdev,
tipc-discussion
In-Reply-To: <1416973315-6047-1-git-send-email-ying.xue@windriver.com>
TIPC protocol message queue is intended to save one protocol message
when bearer is congested so that the message stored in the queue can
be immediately transmitted when bearer congestion is released. However,
as now the protocol queue has no mission any more with the removal of
bearer congestion mechanism, it should be removed.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
---
net/tipc/link.c | 22 ----------------------
net/tipc/link.h | 2 --
2 files changed, 24 deletions(-)
diff --git a/net/tipc/link.c b/net/tipc/link.c
index 4738cb1..450ed0c 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -421,8 +421,6 @@ void tipc_link_purge_queues(struct tipc_link *l_ptr)
kfree_skb_list(l_ptr->oldest_deferred_in);
kfree_skb_list(l_ptr->first_out);
tipc_link_reset_fragments(l_ptr);
- kfree_skb(l_ptr->proto_msg_queue);
- l_ptr->proto_msg_queue = NULL;
}
void tipc_link_reset(struct tipc_link *l_ptr)
@@ -455,8 +453,6 @@ void tipc_link_reset(struct tipc_link *l_ptr)
/* Clean up all queues: */
link_release_outqueue(l_ptr);
- kfree_skb(l_ptr->proto_msg_queue);
- l_ptr->proto_msg_queue = NULL;
kfree_skb_list(l_ptr->oldest_deferred_in);
if (!skb_queue_empty(&l_ptr->waiting_sks)) {
skb_queue_splice_init(&l_ptr->waiting_sks, &owner->waiting_sks);
@@ -904,18 +900,6 @@ static u32 tipc_link_push_packet(struct tipc_link *l_ptr)
return 0;
}
- /* Send deferred protocol message, if any: */
- buf = l_ptr->proto_msg_queue;
- if (buf) {
- msg_set_ack(buf_msg(buf), mod(l_ptr->next_in_no - 1));
- msg_set_bcast_ack(buf_msg(buf), l_ptr->owner->bclink.last_in);
- tipc_bearer_send(l_ptr->bearer_id, buf, &l_ptr->media_addr);
- l_ptr->unacked_window = 0;
- kfree_skb(buf);
- l_ptr->proto_msg_queue = NULL;
- return 0;
- }
-
/* Send one deferred data message, if send window not full: */
buf = l_ptr->next_out;
if (buf) {
@@ -1446,12 +1430,6 @@ void tipc_link_proto_xmit(struct tipc_link *l_ptr, u32 msg_typ, int probe_msg,
u32 msg_size = sizeof(l_ptr->proto_msg);
int r_flag;
- /* Discard any previous message that was deferred due to congestion */
- if (l_ptr->proto_msg_queue) {
- kfree_skb(l_ptr->proto_msg_queue);
- l_ptr->proto_msg_queue = NULL;
- }
-
/* Don't send protocol message during link changeover */
if (l_ptr->exp_msg_count)
return;
diff --git a/net/tipc/link.h b/net/tipc/link.h
index f463e7b..fb3f99b 100644
--- a/net/tipc/link.h
+++ b/net/tipc/link.h
@@ -130,7 +130,6 @@ struct tipc_stats {
* @oldest_deferred_in: ptr to first inbound message in queue
* @newest_deferred_in: ptr to last inbound message in queue
* @unacked_window: # of inbound messages rx'd without ack'ing back to peer
- * @proto_msg_queue: ptr to (single) outbound control message
* @retransm_queue_size: number of messages to retransmit
* @retransm_queue_head: sequence number of first message to retransmit
* @next_out: ptr to first unsent outbound message in queue
@@ -191,7 +190,6 @@ struct tipc_link {
u32 unacked_window;
/* Congestion handling */
- struct sk_buff *proto_msg_queue;
u32 retransm_queue_size;
u32 retransm_queue_head;
struct sk_buff *next_out;
--
1.7.9.5
^ permalink raw reply related
* [PATCH net-next 00/11] standardize TIPC SKB queue operations
From: Ying Xue @ 2014-11-26 3:41 UTC (permalink / raw)
To: davem
Cc: jon.maloy, Paul.Gortmaker, erik.hugne, richard.alpe, netdev,
tipc-discussion
Now the following SKB queues are created and maintained within internal
TIPC stack:
- link transmission queue
- link deferred queue
- link receive queue
- socket outgoing packet chain
- name table outgoing packet chain
In order to manage above queues, TIPC stack declares a sk_buff pointer
for each queue to record its head, and directly modifies "prev" and
"next" SKB pointers of SKB structure when inserting or deleting a SKB
to or from the queue. As these operations are pretty complex, they
easily involve fatal mistakes. If these sk_buff pointers are replaced
with sk_buff_head instances as queue heads and corresponding generic
SKB list APIs are used to manage them, the entire TIPC code would
become quite clean and readable. But before make the change, we need
to clean up below redundant functionalities:
- remove node subscribe infrastructure
- remove protocol message queue
- remove retransmission queue
- clean up process of pushing packets in link layer
Ying Xue (11):
tipc: remove node subscription infrastructure
tipc: remove protocol message queue
tipc: remove retransmission queue
tipc: clean up the process of link pushing packets
tipc: eliminate two pseudo message types of BUNDLE_OPEN and
BUNDLE_CLOSED
tipc: remove unused between routine
tipc: use skb_queue_walk_safe marco to simplify link_prepare_wakeup
routine
tipc: use generic SKB list APIs to manage link transmission queue
tipc: use generic SKB list APIs to manage deferred queue of link
tipc: use generic SKB list APIs to manage link receive queue
tipc: use generic SKB list APIs to manage TIPC outgoing packet chains
net/tipc/Makefile | 4 +-
net/tipc/bcast.c | 109 +++++-----
net/tipc/bcast.h | 2 +-
net/tipc/bearer.h | 2 +-
net/tipc/core.h | 1 +
net/tipc/link.c | 514 ++++++++++++++++++++----------------------------
net/tipc/link.h | 48 ++---
net/tipc/msg.c | 125 ++++++------
net/tipc/msg.h | 16 +-
net/tipc/name_distr.c | 98 ++++++---
net/tipc/name_distr.h | 1 +
net/tipc/name_table.c | 2 +-
net/tipc/name_table.h | 6 +-
net/tipc/node.c | 10 +-
net/tipc/node.h | 12 +-
net/tipc/node_subscr.c | 96 ---------
net/tipc/node_subscr.h | 63 ------
net/tipc/socket.c | 127 ++++++------
18 files changed, 496 insertions(+), 740 deletions(-)
delete mode 100644 net/tipc/node_subscr.c
delete mode 100644 net/tipc/node_subscr.h
--
1.7.9.5
^ permalink raw reply
* [PATCH net-next 11/11] tipc: use generic SKB list APIs to manage TIPC outgoing packet chains
From: Ying Xue @ 2014-11-26 3:41 UTC (permalink / raw)
To: davem; +Cc: jon.maloy, netdev, Paul.Gortmaker, tipc-discussion
In-Reply-To: <1416973315-6047-1-git-send-email-ying.xue@windriver.com>
Use standard SKB list APIs associated with struct sk_buff_head to
manage socket outgoing packet chain and name table outgoing packet
chain, having relevant code simpler and more readable.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
---
net/tipc/bcast.c | 20 ++++----
net/tipc/bcast.h | 2 +-
net/tipc/link.c | 98 ++++++++++++++++++++++++--------------
net/tipc/link.h | 5 +-
net/tipc/msg.c | 74 ++++++++++++++--------------
net/tipc/msg.h | 6 +--
net/tipc/name_distr.c | 46 +++++++++---------
net/tipc/socket.c | 127 +++++++++++++++++++++++++------------------------
8 files changed, 203 insertions(+), 175 deletions(-)
diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c
index 7b238b1..f0761c7 100644
--- a/net/tipc/bcast.c
+++ b/net/tipc/bcast.c
@@ -398,20 +398,20 @@ static void bclink_peek_nack(struct tipc_msg *msg)
/* tipc_bclink_xmit - broadcast buffer chain to all nodes in cluster
* and to identified node local sockets
- * @buf: chain of buffers containing message
+ * @list: chain of buffers containing message
* Consumes the buffer chain, except when returning -ELINKCONG
* Returns 0 if success, otherwise errno: -ELINKCONG,-EHOSTUNREACH,-EMSGSIZE
*/
-int tipc_bclink_xmit(struct sk_buff *buf)
+int tipc_bclink_xmit(struct sk_buff_head *list)
{
int rc = 0;
int bc = 0;
- struct sk_buff *clbuf;
+ struct sk_buff *skb;
/* Prepare clone of message for local node */
- clbuf = tipc_msg_reassemble(buf);
- if (unlikely(!clbuf)) {
- kfree_skb_list(buf);
+ skb = tipc_msg_reassemble(list);
+ if (unlikely(!skb)) {
+ __skb_queue_purge(list);
return -EHOSTUNREACH;
}
@@ -419,7 +419,7 @@ int tipc_bclink_xmit(struct sk_buff *buf)
if (likely(bclink)) {
tipc_bclink_lock();
if (likely(bclink->bcast_nodes.count)) {
- rc = __tipc_link_xmit(bcl, buf);
+ rc = __tipc_link_xmit(bcl, list);
if (likely(!rc)) {
u32 len = skb_queue_len(&bcl->outqueue);
@@ -433,13 +433,13 @@ int tipc_bclink_xmit(struct sk_buff *buf)
}
if (unlikely(!bc))
- kfree_skb_list(buf);
+ __skb_queue_purge(list);
/* Deliver message clone */
if (likely(!rc))
- tipc_sk_mcast_rcv(clbuf);
+ tipc_sk_mcast_rcv(skb);
else
- kfree_skb(clbuf);
+ kfree_skb(skb);
return rc;
}
diff --git a/net/tipc/bcast.h b/net/tipc/bcast.h
index 443de08..644d791 100644
--- a/net/tipc/bcast.h
+++ b/net/tipc/bcast.h
@@ -100,7 +100,7 @@ int tipc_bclink_reset_stats(void);
int tipc_bclink_set_queue_limits(u32 limit);
void tipc_bcbearer_sort(struct tipc_node_map *nm_ptr, u32 node, bool action);
uint tipc_bclink_get_mtu(void);
-int tipc_bclink_xmit(struct sk_buff *buf);
+int tipc_bclink_xmit(struct sk_buff_head *list);
void tipc_bclink_wakeup_users(void);
int tipc_nl_add_bc_link(struct tipc_nl_msg *msg);
diff --git a/net/tipc/link.c b/net/tipc/link.c
index 0e04508..34bf15c 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -664,9 +664,10 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event)
* - For all other messages we discard the buffer and return -EHOSTUNREACH
* - For TIPC internal messages we also reset the link
*/
-static int tipc_link_cong(struct tipc_link *link, struct sk_buff *buf)
+static int tipc_link_cong(struct tipc_link *link, struct sk_buff_head *list)
{
- struct tipc_msg *msg = buf_msg(buf);
+ struct sk_buff *skb = skb_peek(list);
+ struct tipc_msg *msg = buf_msg(skb);
uint imp = tipc_msg_tot_importance(msg);
u32 oport = msg_tot_origport(msg);
@@ -679,28 +680,29 @@ static int tipc_link_cong(struct tipc_link *link, struct sk_buff *buf)
goto drop;
if (unlikely(msg_reroute_cnt(msg)))
goto drop;
- if (TIPC_SKB_CB(buf)->wakeup_pending)
+ if (TIPC_SKB_CB(skb)->wakeup_pending)
return -ELINKCONG;
- if (link_schedule_user(link, oport, TIPC_SKB_CB(buf)->chain_sz, imp))
+ if (link_schedule_user(link, oport, skb_queue_len(list), imp))
return -ELINKCONG;
drop:
- kfree_skb_list(buf);
+ __skb_queue_purge(list);
return -EHOSTUNREACH;
}
/**
* __tipc_link_xmit(): same as tipc_link_xmit, but destlink is known & locked
* @link: link to use
- * @skb: chain of buffers containing message
+ * @list: chain of buffers containing message
+ *
* Consumes the buffer chain, except when returning -ELINKCONG
* Returns 0 if success, otherwise errno: -ELINKCONG, -EMSGSIZE (plain socket
* user data messages) or -EHOSTUNREACH (all other messages/senders)
* Only the socket functions tipc_send_stream() and tipc_send_packet() need
* to act on the return value, since they may need to do more send attempts.
*/
-int __tipc_link_xmit(struct tipc_link *link, struct sk_buff *skb)
+int __tipc_link_xmit(struct tipc_link *link, struct sk_buff_head *list)
{
- struct tipc_msg *msg = buf_msg(skb);
+ struct tipc_msg *msg = buf_msg(skb_peek(list));
uint psz = msg_size(msg);
uint sndlim = link->queue_limit[0];
uint imp = tipc_msg_tot_importance(msg);
@@ -710,21 +712,21 @@ int __tipc_link_xmit(struct tipc_link *link, struct sk_buff *skb)
uint bc_last_in = link->owner->bclink.last_in;
struct tipc_media_addr *addr = &link->media_addr;
struct sk_buff_head *outqueue = &link->outqueue;
- struct sk_buff *next;
+ struct sk_buff *skb, *tmp;
/* Match queue limits against msg importance: */
if (unlikely(skb_queue_len(outqueue) >= link->queue_limit[imp]))
- return tipc_link_cong(link, skb);
+ return tipc_link_cong(link, list);
/* Has valid packet limit been used ? */
if (unlikely(psz > mtu)) {
- kfree_skb_list(skb);
+ __skb_queue_purge(list);
return -EMSGSIZE;
}
/* Prepare each packet for sending, and add to outqueue: */
- while (skb) {
- next = skb->next;
+ skb_queue_walk_safe(list, skb, tmp) {
+ __skb_unlink(skb, list);
msg = buf_msg(skb);
msg_set_word(msg, 2, ((ack << 16) | mod(seqno)));
msg_set_bcast_ack(msg, bc_last_in);
@@ -736,7 +738,6 @@ int __tipc_link_xmit(struct tipc_link *link, struct sk_buff *skb)
link->unacked_window = 0;
} else if (tipc_msg_bundle(outqueue, skb, mtu)) {
link->stats.sent_bundled++;
- skb = next;
continue;
} else if (tipc_msg_make_bundle(outqueue, skb, mtu,
link->addr)) {
@@ -750,22 +751,43 @@ int __tipc_link_xmit(struct tipc_link *link, struct sk_buff *skb)
link->next_out = skb;
}
seqno++;
- skb = next;
}
link->next_out_no = seqno;
return 0;
}
+static void skb2list(struct sk_buff *skb, struct sk_buff_head *list)
+{
+ __skb_queue_head_init(list);
+ __skb_queue_tail(list, skb);
+}
+
+static int __tipc_link_xmit_skb(struct tipc_link *link, struct sk_buff *skb)
+{
+ struct sk_buff_head head;
+
+ skb2list(skb, &head);
+ return __tipc_link_xmit(link, &head);
+}
+
+int tipc_link_xmit_skb(struct sk_buff *skb, u32 dnode, u32 selector)
+{
+ struct sk_buff_head head;
+
+ skb2list(skb, &head);
+ return tipc_link_xmit(&head, dnode, selector);
+}
+
/**
* tipc_link_xmit() is the general link level function for message sending
- * @buf: chain of buffers containing message
+ * @list: chain of buffers containing message
* @dsz: amount of user data to be sent
* @dnode: address of destination node
* @selector: a number used for deterministic link selection
* Consumes the buffer chain, except when returning -ELINKCONG
* Returns 0 if success, otherwise errno: -ELINKCONG,-EHOSTUNREACH,-EMSGSIZE
*/
-int tipc_link_xmit(struct sk_buff *buf, u32 dnode, u32 selector)
+int tipc_link_xmit(struct sk_buff_head *list, u32 dnode, u32 selector)
{
struct tipc_link *link = NULL;
struct tipc_node *node;
@@ -776,17 +798,22 @@ int tipc_link_xmit(struct sk_buff *buf, u32 dnode, u32 selector)
tipc_node_lock(node);
link = node->active_links[selector & 1];
if (link)
- rc = __tipc_link_xmit(link, buf);
+ rc = __tipc_link_xmit(link, list);
tipc_node_unlock(node);
}
if (link)
return rc;
- if (likely(in_own_node(dnode)))
- return tipc_sk_rcv(buf);
+ if (likely(in_own_node(dnode))) {
+ /* As a node local message chain never contains more than one
+ * buffer, we just need to dequeue one SKB buffer from the
+ * head list.
+ */
+ return tipc_sk_rcv(__skb_dequeue(list));
+ }
+ __skb_queue_purge(list);
- kfree_skb_list(buf);
return rc;
}
@@ -800,17 +827,17 @@ int tipc_link_xmit(struct sk_buff *buf, u32 dnode, u32 selector)
*/
static void tipc_link_sync_xmit(struct tipc_link *link)
{
- struct sk_buff *buf;
+ struct sk_buff *skb;
struct tipc_msg *msg;
- buf = tipc_buf_acquire(INT_H_SIZE);
- if (!buf)
+ skb = tipc_buf_acquire(INT_H_SIZE);
+ if (!skb)
return;
- msg = buf_msg(buf);
+ msg = buf_msg(skb);
tipc_msg_init(msg, BCAST_PROTOCOL, STATE_MSG, INT_H_SIZE, link->addr);
msg_set_last_bcast(msg, link->owner->bclink.acked);
- __tipc_link_xmit(link, buf);
+ __tipc_link_xmit_skb(link, skb);
}
/*
@@ -1053,8 +1080,7 @@ void tipc_rcv(struct sk_buff *skb, struct tipc_bearer *b_ptr)
u32 ackd;
u32 released;
- __skb_queue_head_init(&head);
- __skb_queue_tail(&head, skb);
+ skb2list(skb, &head);
while ((skb = __skb_dequeue(&head))) {
/* Ensure message is well-formed */
@@ -1573,7 +1599,7 @@ static void tipc_link_tunnel_xmit(struct tipc_link *l_ptr,
u32 selector)
{
struct tipc_link *tunnel;
- struct sk_buff *buf;
+ struct sk_buff *skb;
u32 length = msg_size(msg);
tunnel = l_ptr->owner->active_links[selector & 1];
@@ -1582,14 +1608,14 @@ static void tipc_link_tunnel_xmit(struct tipc_link *l_ptr,
return;
}
msg_set_size(tunnel_hdr, length + INT_H_SIZE);
- buf = tipc_buf_acquire(length + INT_H_SIZE);
- if (!buf) {
+ skb = tipc_buf_acquire(length + INT_H_SIZE);
+ if (!skb) {
pr_warn("%sunable to send tunnel msg\n", link_co_err);
return;
}
- skb_copy_to_linear_data(buf, tunnel_hdr, INT_H_SIZE);
- skb_copy_to_linear_data_offset(buf, INT_H_SIZE, msg, length);
- __tipc_link_xmit(tunnel, buf);
+ skb_copy_to_linear_data(skb, tunnel_hdr, INT_H_SIZE);
+ skb_copy_to_linear_data_offset(skb, INT_H_SIZE, msg, length);
+ __tipc_link_xmit_skb(tunnel, skb);
}
@@ -1620,7 +1646,7 @@ void tipc_link_failover_send_queue(struct tipc_link *l_ptr)
if (skb) {
skb_copy_to_linear_data(skb, &tunnel_hdr, INT_H_SIZE);
msg_set_size(&tunnel_hdr, INT_H_SIZE);
- __tipc_link_xmit(tunnel, skb);
+ __tipc_link_xmit_skb(tunnel, skb);
} else {
pr_warn("%sunable to send changeover msg\n",
link_co_err);
@@ -1691,7 +1717,7 @@ void tipc_link_dup_queue_xmit(struct tipc_link *l_ptr,
skb_copy_to_linear_data(outskb, &tunnel_hdr, INT_H_SIZE);
skb_copy_to_linear_data_offset(outskb, INT_H_SIZE, skb->data,
length);
- __tipc_link_xmit(tunnel, outskb);
+ __tipc_link_xmit_skb(tunnel, outskb);
if (!tipc_link_is_up(l_ptr))
return;
}
diff --git a/net/tipc/link.h b/net/tipc/link.h
index de7b883..55812e8 100644
--- a/net/tipc/link.h
+++ b/net/tipc/link.h
@@ -213,8 +213,9 @@ struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area,
void tipc_link_reset_all(struct tipc_node *node);
void tipc_link_reset(struct tipc_link *l_ptr);
void tipc_link_reset_list(unsigned int bearer_id);
-int tipc_link_xmit(struct sk_buff *buf, u32 dest, u32 selector);
-int __tipc_link_xmit(struct tipc_link *link, struct sk_buff *buf);
+int tipc_link_xmit_skb(struct sk_buff *skb, u32 dest, u32 selector);
+int tipc_link_xmit(struct sk_buff_head *list, u32 dest, u32 selector);
+int __tipc_link_xmit(struct tipc_link *link, struct sk_buff_head *list);
u32 tipc_link_get_max_pkt(u32 dest, u32 selector);
void tipc_link_bundle_rcv(struct sk_buff *buf);
void tipc_link_proto_xmit(struct tipc_link *l_ptr, u32 msg_typ, int prob,
diff --git a/net/tipc/msg.c b/net/tipc/msg.c
index ce7514a..5b06597 100644
--- a/net/tipc/msg.c
+++ b/net/tipc/msg.c
@@ -166,11 +166,12 @@ err:
* @offset: Posision in iov to start copying from
* @dsz: Total length of user data
* @pktmax: Max packet size that can be used
- * @chain: Buffer or chain of buffers to be returned to caller
+ * @list: Buffer or chain of buffers to be returned to caller
+ *
* Returns message data size or errno: -ENOMEM, -EFAULT
*/
-int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m,
- int offset, int dsz, int pktmax , struct sk_buff **chain)
+int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m, int offset,
+ int dsz, int pktmax, struct sk_buff_head *list)
{
int mhsz = msg_hdr_sz(mhdr);
int msz = mhsz + dsz;
@@ -179,22 +180,22 @@ int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m,
int pktrem = pktmax;
int drem = dsz;
struct tipc_msg pkthdr;
- struct sk_buff *buf, *prev;
+ struct sk_buff *skb;
char *pktpos;
int rc;
- uint chain_sz = 0;
+
msg_set_size(mhdr, msz);
/* No fragmentation needed? */
if (likely(msz <= pktmax)) {
- buf = tipc_buf_acquire(msz);
- *chain = buf;
- if (unlikely(!buf))
+ skb = tipc_buf_acquire(msz);
+ if (unlikely(!skb))
return -ENOMEM;
- skb_copy_to_linear_data(buf, mhdr, mhsz);
- pktpos = buf->data + mhsz;
- TIPC_SKB_CB(buf)->chain_sz = 1;
- if (!dsz || !memcpy_fromiovecend(pktpos, m->msg_iov, offset, dsz))
+ __skb_queue_tail(list, skb);
+ skb_copy_to_linear_data(skb, mhdr, mhsz);
+ pktpos = skb->data + mhsz;
+ if (!dsz || !memcpy_fromiovecend(pktpos, m->msg_iov, offset,
+ dsz))
return dsz;
rc = -EFAULT;
goto error;
@@ -207,15 +208,15 @@ int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m,
msg_set_fragm_no(&pkthdr, pktno);
/* Prepare first fragment */
- *chain = buf = tipc_buf_acquire(pktmax);
- if (!buf)
+ skb = tipc_buf_acquire(pktmax);
+ if (!skb)
return -ENOMEM;
- chain_sz = 1;
- pktpos = buf->data;
- skb_copy_to_linear_data(buf, &pkthdr, INT_H_SIZE);
+ __skb_queue_tail(list, skb);
+ pktpos = skb->data;
+ skb_copy_to_linear_data(skb, &pkthdr, INT_H_SIZE);
pktpos += INT_H_SIZE;
pktrem -= INT_H_SIZE;
- skb_copy_to_linear_data_offset(buf, INT_H_SIZE, mhdr, mhsz);
+ skb_copy_to_linear_data_offset(skb, INT_H_SIZE, mhdr, mhsz);
pktpos += mhsz;
pktrem -= mhsz;
@@ -238,28 +239,25 @@ int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m,
pktsz = drem + INT_H_SIZE;
else
pktsz = pktmax;
- prev = buf;
- buf = tipc_buf_acquire(pktsz);
- if (!buf) {
+ skb = tipc_buf_acquire(pktsz);
+ if (!skb) {
rc = -ENOMEM;
goto error;
}
- chain_sz++;
- prev->next = buf;
+ __skb_queue_tail(list, skb);
msg_set_type(&pkthdr, FRAGMENT);
msg_set_size(&pkthdr, pktsz);
msg_set_fragm_no(&pkthdr, ++pktno);
- skb_copy_to_linear_data(buf, &pkthdr, INT_H_SIZE);
- pktpos = buf->data + INT_H_SIZE;
+ skb_copy_to_linear_data(skb, &pkthdr, INT_H_SIZE);
+ pktpos = skb->data + INT_H_SIZE;
pktrem = pktsz - INT_H_SIZE;
} while (1);
- TIPC_SKB_CB(*chain)->chain_sz = chain_sz;
- msg_set_type(buf_msg(buf), LAST_FRAGMENT);
+ msg_set_type(buf_msg(skb), LAST_FRAGMENT);
return dsz;
error:
- kfree_skb_list(*chain);
- *chain = NULL;
+ __skb_queue_purge(list);
+ __skb_queue_head_init(list);
return rc;
}
@@ -430,22 +428,23 @@ int tipc_msg_eval(struct sk_buff *buf, u32 *dnode)
/* tipc_msg_reassemble() - clone a buffer chain of fragments and
* reassemble the clones into one message
*/
-struct sk_buff *tipc_msg_reassemble(struct sk_buff *chain)
+struct sk_buff *tipc_msg_reassemble(struct sk_buff_head *list)
{
- struct sk_buff *buf = chain;
- struct sk_buff *frag = buf;
+ struct sk_buff *skb;
+ struct sk_buff *frag = NULL;
struct sk_buff *head = NULL;
int hdr_sz;
/* Copy header if single buffer */
- if (!buf->next) {
- hdr_sz = skb_headroom(buf) + msg_hdr_sz(buf_msg(buf));
- return __pskb_copy(buf, hdr_sz, GFP_ATOMIC);
+ if (skb_queue_len(list) == 1) {
+ skb = skb_peek(list);
+ hdr_sz = skb_headroom(skb) + msg_hdr_sz(buf_msg(skb));
+ return __pskb_copy(skb, hdr_sz, GFP_ATOMIC);
}
/* Clone all fragments and reassemble */
- while (buf) {
- frag = skb_clone(buf, GFP_ATOMIC);
+ skb_queue_walk(list, skb) {
+ frag = skb_clone(skb, GFP_ATOMIC);
if (!frag)
goto error;
frag->next = NULL;
@@ -453,7 +452,6 @@ struct sk_buff *tipc_msg_reassemble(struct sk_buff *chain)
break;
if (!head)
goto error;
- buf = buf->next;
}
return frag;
error:
diff --git a/net/tipc/msg.h b/net/tipc/msg.h
index 53e425f..d5c83d7 100644
--- a/net/tipc/msg.h
+++ b/net/tipc/msg.h
@@ -739,9 +739,9 @@ bool tipc_msg_bundle(struct sk_buff_head *list, struct sk_buff *skb, u32 mtu);
bool tipc_msg_make_bundle(struct sk_buff_head *list, struct sk_buff *skb,
u32 mtu, u32 dnode);
-int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m,
- int offset, int dsz, int mtu , struct sk_buff **chain);
+int tipc_msg_build(struct tipc_msg *mhdr, struct msghdr *m, int offset,
+ int dsz, int mtu, struct sk_buff_head *list);
-struct sk_buff *tipc_msg_reassemble(struct sk_buff *chain);
+struct sk_buff *tipc_msg_reassemble(struct sk_buff_head *list);
#endif
diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c
index 6c2638d..56248db 100644
--- a/net/tipc/name_distr.c
+++ b/net/tipc/name_distr.c
@@ -114,9 +114,9 @@ static struct sk_buff *named_prepare_buf(u32 type, u32 size, u32 dest)
return buf;
}
-void named_cluster_distribute(struct sk_buff *buf)
+void named_cluster_distribute(struct sk_buff *skb)
{
- struct sk_buff *obuf;
+ struct sk_buff *oskb;
struct tipc_node *node;
u32 dnode;
@@ -127,15 +127,15 @@ void named_cluster_distribute(struct sk_buff *buf)
continue;
if (!tipc_node_active_links(node))
continue;
- obuf = skb_copy(buf, GFP_ATOMIC);
- if (!obuf)
+ oskb = skb_copy(skb, GFP_ATOMIC);
+ if (!oskb)
break;
- msg_set_destnode(buf_msg(obuf), dnode);
- tipc_link_xmit(obuf, dnode, dnode);
+ msg_set_destnode(buf_msg(oskb), dnode);
+ tipc_link_xmit_skb(oskb, dnode, dnode);
}
rcu_read_unlock();
- kfree_skb(buf);
+ kfree_skb(skb);
}
/**
@@ -190,15 +190,15 @@ struct sk_buff *tipc_named_withdraw(struct publication *publ)
/**
* named_distribute - prepare name info for bulk distribution to another node
- * @msg_list: list of messages (buffers) to be returned from this function
+ * @list: list of messages (buffers) to be returned from this function
* @dnode: node to be updated
* @pls: linked list of publication items to be packed into buffer chain
*/
-static void named_distribute(struct list_head *msg_list, u32 dnode,
+static void named_distribute(struct sk_buff_head *list, u32 dnode,
struct publ_list *pls)
{
struct publication *publ;
- struct sk_buff *buf = NULL;
+ struct sk_buff *skb = NULL;
struct distr_item *item = NULL;
uint dsz = pls->size * ITEM_SIZE;
uint msg_dsz = (tipc_node_get_mtu(dnode, 0) / ITEM_SIZE) * ITEM_SIZE;
@@ -207,15 +207,15 @@ static void named_distribute(struct list_head *msg_list, u32 dnode,
list_for_each_entry(publ, &pls->list, local_list) {
/* Prepare next buffer: */
- if (!buf) {
+ if (!skb) {
msg_rem = min_t(uint, rem, msg_dsz);
rem -= msg_rem;
- buf = named_prepare_buf(PUBLICATION, msg_rem, dnode);
- if (!buf) {
+ skb = named_prepare_buf(PUBLICATION, msg_rem, dnode);
+ if (!skb) {
pr_warn("Bulk publication failure\n");
return;
}
- item = (struct distr_item *)msg_data(buf_msg(buf));
+ item = (struct distr_item *)msg_data(buf_msg(skb));
}
/* Pack publication into message: */
@@ -225,8 +225,8 @@ static void named_distribute(struct list_head *msg_list, u32 dnode,
/* Append full buffer to list: */
if (!msg_rem) {
- list_add_tail((struct list_head *)buf, msg_list);
- buf = NULL;
+ __skb_queue_tail(list, skb);
+ skb = NULL;
}
}
}
@@ -236,18 +236,16 @@ static void named_distribute(struct list_head *msg_list, u32 dnode,
*/
void tipc_named_node_up(u32 dnode)
{
- LIST_HEAD(msg_list);
- struct sk_buff *buf_chain;
+ struct sk_buff_head head;
+
+ __skb_queue_head_init(&head);
read_lock_bh(&tipc_nametbl_lock);
- named_distribute(&msg_list, dnode, &publ_cluster);
- named_distribute(&msg_list, dnode, &publ_zone);
+ named_distribute(&head, dnode, &publ_cluster);
+ named_distribute(&head, dnode, &publ_zone);
read_unlock_bh(&tipc_nametbl_lock);
- /* Convert circular list to linear list and send: */
- buf_chain = (struct sk_buff *)msg_list.next;
- ((struct sk_buff *)msg_list.prev)->next = NULL;
- tipc_link_xmit(buf_chain, dnode, dnode);
+ tipc_link_xmit(&head, dnode, dnode);
}
static void tipc_publ_subscribe(struct publication *publ, u32 addr)
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index 341fbd1..9658d9b 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -244,12 +244,12 @@ static void tsk_advance_rx_queue(struct sock *sk)
*/
static void tsk_rej_rx_queue(struct sock *sk)
{
- struct sk_buff *buf;
+ struct sk_buff *skb;
u32 dnode;
- while ((buf = __skb_dequeue(&sk->sk_receive_queue))) {
- if (tipc_msg_reverse(buf, &dnode, TIPC_ERR_NO_PORT))
- tipc_link_xmit(buf, dnode, 0);
+ while ((skb = __skb_dequeue(&sk->sk_receive_queue))) {
+ if (tipc_msg_reverse(skb, &dnode, TIPC_ERR_NO_PORT))
+ tipc_link_xmit_skb(skb, dnode, 0);
}
}
@@ -462,7 +462,7 @@ static int tipc_release(struct socket *sock)
{
struct sock *sk = sock->sk;
struct tipc_sock *tsk;
- struct sk_buff *buf;
+ struct sk_buff *skb;
u32 dnode;
/*
@@ -481,11 +481,11 @@ static int tipc_release(struct socket *sock)
*/
dnode = tsk_peer_node(tsk);
while (sock->state != SS_DISCONNECTING) {
- buf = __skb_dequeue(&sk->sk_receive_queue);
- if (buf == NULL)
+ skb = __skb_dequeue(&sk->sk_receive_queue);
+ if (skb == NULL)
break;
- if (TIPC_SKB_CB(buf)->handle != NULL)
- kfree_skb(buf);
+ if (TIPC_SKB_CB(skb)->handle != NULL)
+ kfree_skb(skb);
else {
if ((sock->state == SS_CONNECTING) ||
(sock->state == SS_CONNECTED)) {
@@ -493,8 +493,8 @@ static int tipc_release(struct socket *sock)
tsk->connected = 0;
tipc_node_remove_conn(dnode, tsk->ref);
}
- if (tipc_msg_reverse(buf, &dnode, TIPC_ERR_NO_PORT))
- tipc_link_xmit(buf, dnode, 0);
+ if (tipc_msg_reverse(skb, &dnode, TIPC_ERR_NO_PORT))
+ tipc_link_xmit_skb(skb, dnode, 0);
}
}
@@ -502,12 +502,12 @@ static int tipc_release(struct socket *sock)
tipc_sk_ref_discard(tsk->ref);
k_cancel_timer(&tsk->timer);
if (tsk->connected) {
- buf = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, TIPC_CONN_MSG,
+ skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, TIPC_CONN_MSG,
SHORT_H_SIZE, 0, dnode, tipc_own_addr,
tsk_peer_port(tsk),
tsk->ref, TIPC_ERR_NO_PORT);
- if (buf)
- tipc_link_xmit(buf, dnode, tsk->ref);
+ if (skb)
+ tipc_link_xmit_skb(skb, dnode, tsk->ref);
tipc_node_remove_conn(dnode, tsk->ref);
}
k_term_timer(&tsk->timer);
@@ -712,7 +712,7 @@ static int tipc_sendmcast(struct socket *sock, struct tipc_name_seq *seq,
{
struct sock *sk = sock->sk;
struct tipc_msg *mhdr = &tipc_sk(sk)->phdr;
- struct sk_buff *buf;
+ struct sk_buff_head head;
uint mtu;
int rc;
@@ -727,12 +727,13 @@ static int tipc_sendmcast(struct socket *sock, struct tipc_name_seq *seq,
new_mtu:
mtu = tipc_bclink_get_mtu();
- rc = tipc_msg_build(mhdr, msg, 0, dsz, mtu, &buf);
+ __skb_queue_head_init(&head);
+ rc = tipc_msg_build(mhdr, msg, 0, dsz, mtu, &head);
if (unlikely(rc < 0))
return rc;
do {
- rc = tipc_bclink_xmit(buf);
+ rc = tipc_bclink_xmit(&head);
if (likely(rc >= 0)) {
rc = dsz;
break;
@@ -744,7 +745,7 @@ new_mtu:
tipc_sk(sk)->link_cong = 1;
rc = tipc_wait_for_sndmsg(sock, &timeo);
if (rc)
- kfree_skb_list(buf);
+ __skb_queue_purge(&head);
} while (!rc);
return rc;
}
@@ -906,7 +907,8 @@ static int tipc_sendmsg(struct kiocb *iocb, struct socket *sock,
struct tipc_sock *tsk = tipc_sk(sk);
struct tipc_msg *mhdr = &tsk->phdr;
u32 dnode, dport;
- struct sk_buff *buf;
+ struct sk_buff_head head;
+ struct sk_buff *skb;
struct tipc_name_seq *seq = &dest->addr.nameseq;
u32 mtu;
long timeo;
@@ -981,13 +983,15 @@ static int tipc_sendmsg(struct kiocb *iocb, struct socket *sock,
new_mtu:
mtu = tipc_node_get_mtu(dnode, tsk->ref);
- rc = tipc_msg_build(mhdr, m, 0, dsz, mtu, &buf);
+ __skb_queue_head_init(&head);
+ rc = tipc_msg_build(mhdr, m, 0, dsz, mtu, &head);
if (rc < 0)
goto exit;
do {
- TIPC_SKB_CB(buf)->wakeup_pending = tsk->link_cong;
- rc = tipc_link_xmit(buf, dnode, tsk->ref);
+ skb = skb_peek(&head);
+ TIPC_SKB_CB(skb)->wakeup_pending = tsk->link_cong;
+ rc = tipc_link_xmit(&head, dnode, tsk->ref);
if (likely(rc >= 0)) {
if (sock->state != SS_READY)
sock->state = SS_CONNECTING;
@@ -1001,7 +1005,7 @@ new_mtu:
tsk->link_cong = 1;
rc = tipc_wait_for_sndmsg(sock, &timeo);
if (rc)
- kfree_skb_list(buf);
+ __skb_queue_purge(&head);
} while (!rc);
exit:
if (iocb)
@@ -1058,7 +1062,7 @@ static int tipc_send_stream(struct kiocb *iocb, struct socket *sock,
struct sock *sk = sock->sk;
struct tipc_sock *tsk = tipc_sk(sk);
struct tipc_msg *mhdr = &tsk->phdr;
- struct sk_buff *buf;
+ struct sk_buff_head head;
DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
u32 ref = tsk->ref;
int rc = -EINVAL;
@@ -1093,12 +1097,13 @@ static int tipc_send_stream(struct kiocb *iocb, struct socket *sock,
next:
mtu = tsk->max_pkt;
send = min_t(uint, dsz - sent, TIPC_MAX_USER_MSG_SIZE);
- rc = tipc_msg_build(mhdr, m, sent, send, mtu, &buf);
+ __skb_queue_head_init(&head);
+ rc = tipc_msg_build(mhdr, m, sent, send, mtu, &head);
if (unlikely(rc < 0))
goto exit;
do {
if (likely(!tsk_conn_cong(tsk))) {
- rc = tipc_link_xmit(buf, dnode, ref);
+ rc = tipc_link_xmit(&head, dnode, ref);
if (likely(!rc)) {
tsk->sent_unacked++;
sent += send;
@@ -1116,7 +1121,7 @@ next:
}
rc = tipc_wait_for_sndpkt(sock, &timeo);
if (rc)
- kfree_skb_list(buf);
+ __skb_queue_purge(&head);
} while (!rc);
exit:
if (iocb)
@@ -1261,20 +1266,20 @@ static int tipc_sk_anc_data_recv(struct msghdr *m, struct tipc_msg *msg,
static void tipc_sk_send_ack(struct tipc_sock *tsk, uint ack)
{
- struct sk_buff *buf = NULL;
+ struct sk_buff *skb = NULL;
struct tipc_msg *msg;
u32 peer_port = tsk_peer_port(tsk);
u32 dnode = tsk_peer_node(tsk);
if (!tsk->connected)
return;
- buf = tipc_msg_create(CONN_MANAGER, CONN_ACK, INT_H_SIZE, 0, dnode,
+ skb = tipc_msg_create(CONN_MANAGER, CONN_ACK, INT_H_SIZE, 0, dnode,
tipc_own_addr, peer_port, tsk->ref, TIPC_OK);
- if (!buf)
+ if (!skb)
return;
- msg = buf_msg(buf);
+ msg = buf_msg(skb);
msg_set_msgcnt(msg, ack);
- tipc_link_xmit(buf, dnode, msg_link_selector(msg));
+ tipc_link_xmit_skb(skb, dnode, msg_link_selector(msg));
}
static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop)
@@ -1729,20 +1734,20 @@ static int filter_rcv(struct sock *sk, struct sk_buff *buf)
/**
* tipc_backlog_rcv - handle incoming message from backlog queue
* @sk: socket
- * @buf: message
+ * @skb: message
*
* Caller must hold socket lock, but not port lock.
*
* Returns 0
*/
-static int tipc_backlog_rcv(struct sock *sk, struct sk_buff *buf)
+static int tipc_backlog_rcv(struct sock *sk, struct sk_buff *skb)
{
int rc;
u32 onode;
struct tipc_sock *tsk = tipc_sk(sk);
- uint truesize = buf->truesize;
+ uint truesize = skb->truesize;
- rc = filter_rcv(sk, buf);
+ rc = filter_rcv(sk, skb);
if (likely(!rc)) {
if (atomic_read(&tsk->dupl_rcvcnt) < TIPC_CONN_OVERLOAD_LIMIT)
@@ -1750,25 +1755,25 @@ static int tipc_backlog_rcv(struct sock *sk, struct sk_buff *buf)
return 0;
}
- if ((rc < 0) && !tipc_msg_reverse(buf, &onode, -rc))
+ if ((rc < 0) && !tipc_msg_reverse(skb, &onode, -rc))
return 0;
- tipc_link_xmit(buf, onode, 0);
+ tipc_link_xmit_skb(skb, onode, 0);
return 0;
}
/**
* tipc_sk_rcv - handle incoming message
- * @buf: buffer containing arriving message
+ * @skb: buffer containing arriving message
* Consumes buffer
* Returns 0 if success, or errno: -EHOSTUNREACH
*/
-int tipc_sk_rcv(struct sk_buff *buf)
+int tipc_sk_rcv(struct sk_buff *skb)
{
struct tipc_sock *tsk;
struct sock *sk;
- u32 dport = msg_destport(buf_msg(buf));
+ u32 dport = msg_destport(buf_msg(skb));
int rc = TIPC_OK;
uint limit;
u32 dnode;
@@ -1776,7 +1781,7 @@ int tipc_sk_rcv(struct sk_buff *buf)
/* Validate destination and message */
tsk = tipc_sk_get(dport);
if (unlikely(!tsk)) {
- rc = tipc_msg_eval(buf, &dnode);
+ rc = tipc_msg_eval(skb, &dnode);
goto exit;
}
sk = &tsk->sk;
@@ -1785,12 +1790,12 @@ int tipc_sk_rcv(struct sk_buff *buf)
spin_lock_bh(&sk->sk_lock.slock);
if (!sock_owned_by_user(sk)) {
- rc = filter_rcv(sk, buf);
+ rc = filter_rcv(sk, skb);
} else {
if (sk->sk_backlog.len == 0)
atomic_set(&tsk->dupl_rcvcnt, 0);
- limit = rcvbuf_limit(sk, buf) + atomic_read(&tsk->dupl_rcvcnt);
- if (sk_add_backlog(sk, buf, limit))
+ limit = rcvbuf_limit(sk, skb) + atomic_read(&tsk->dupl_rcvcnt);
+ if (sk_add_backlog(sk, skb, limit))
rc = -TIPC_ERR_OVERLOAD;
}
spin_unlock_bh(&sk->sk_lock.slock);
@@ -1798,10 +1803,10 @@ int tipc_sk_rcv(struct sk_buff *buf)
if (likely(!rc))
return 0;
exit:
- if ((rc < 0) && !tipc_msg_reverse(buf, &dnode, -rc))
+ if ((rc < 0) && !tipc_msg_reverse(skb, &dnode, -rc))
return -EHOSTUNREACH;
- tipc_link_xmit(buf, dnode, 0);
+ tipc_link_xmit_skb(skb, dnode, 0);
return (rc < 0) ? -EHOSTUNREACH : 0;
}
@@ -2059,7 +2064,7 @@ static int tipc_shutdown(struct socket *sock, int how)
{
struct sock *sk = sock->sk;
struct tipc_sock *tsk = tipc_sk(sk);
- struct sk_buff *buf;
+ struct sk_buff *skb;
u32 dnode;
int res;
@@ -2074,23 +2079,23 @@ static int tipc_shutdown(struct socket *sock, int how)
restart:
/* Disconnect and send a 'FIN+' or 'FIN-' message to peer */
- buf = __skb_dequeue(&sk->sk_receive_queue);
- if (buf) {
- if (TIPC_SKB_CB(buf)->handle != NULL) {
- kfree_skb(buf);
+ skb = __skb_dequeue(&sk->sk_receive_queue);
+ if (skb) {
+ if (TIPC_SKB_CB(skb)->handle != NULL) {
+ kfree_skb(skb);
goto restart;
}
- if (tipc_msg_reverse(buf, &dnode, TIPC_CONN_SHUTDOWN))
- tipc_link_xmit(buf, dnode, tsk->ref);
+ if (tipc_msg_reverse(skb, &dnode, TIPC_CONN_SHUTDOWN))
+ tipc_link_xmit_skb(skb, dnode, tsk->ref);
tipc_node_remove_conn(dnode, tsk->ref);
} else {
dnode = tsk_peer_node(tsk);
- buf = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE,
+ skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE,
TIPC_CONN_MSG, SHORT_H_SIZE,
0, dnode, tipc_own_addr,
tsk_peer_port(tsk),
tsk->ref, TIPC_CONN_SHUTDOWN);
- tipc_link_xmit(buf, dnode, tsk->ref);
+ tipc_link_xmit_skb(skb, dnode, tsk->ref);
}
tsk->connected = 0;
sock->state = SS_DISCONNECTING;
@@ -2119,7 +2124,7 @@ static void tipc_sk_timeout(unsigned long ref)
{
struct tipc_sock *tsk;
struct sock *sk;
- struct sk_buff *buf = NULL;
+ struct sk_buff *skb = NULL;
u32 peer_port, peer_node;
tsk = tipc_sk_get(ref);
@@ -2137,20 +2142,20 @@ static void tipc_sk_timeout(unsigned long ref)
if (tsk->probing_state == TIPC_CONN_PROBING) {
/* Previous probe not answered -> self abort */
- buf = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, TIPC_CONN_MSG,
+ skb = tipc_msg_create(TIPC_CRITICAL_IMPORTANCE, TIPC_CONN_MSG,
SHORT_H_SIZE, 0, tipc_own_addr,
peer_node, ref, peer_port,
TIPC_ERR_NO_PORT);
} else {
- buf = tipc_msg_create(CONN_MANAGER, CONN_PROBE, INT_H_SIZE,
+ skb = tipc_msg_create(CONN_MANAGER, CONN_PROBE, INT_H_SIZE,
0, peer_node, tipc_own_addr,
peer_port, ref, TIPC_OK);
tsk->probing_state = TIPC_CONN_PROBING;
k_start_timer(&tsk->timer, tsk->probing_interval);
}
bh_unlock_sock(sk);
- if (buf)
- tipc_link_xmit(buf, peer_node, ref);
+ if (skb)
+ tipc_link_xmit_skb(skb, peer_node, ref);
exit:
tipc_sk_put(tsk);
}
--
1.7.9.5
------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=157005751&iu=/4140/ostg.clktrk
^ permalink raw reply related
* [PATCH net-next 10/11] tipc: use generic SKB list APIs to manage link receive queue
From: Ying Xue @ 2014-11-26 3:41 UTC (permalink / raw)
To: davem; +Cc: jon.maloy, netdev, Paul.Gortmaker, tipc-discussion
In-Reply-To: <1416973315-6047-1-git-send-email-ying.xue@windriver.com>
Use standard SKB list APIs associated with struct sk_buff_head to
manage link's receive queue to simplify its relevant code cemplexity.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
---
net/tipc/bearer.h | 2 +-
net/tipc/link.c | 85 +++++++++++++++++++++++------------------------------
2 files changed, 37 insertions(+), 50 deletions(-)
diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h
index b1d9052..2c1230a 100644
--- a/net/tipc/bearer.h
+++ b/net/tipc/bearer.h
@@ -165,7 +165,7 @@ extern struct tipc_bearer __rcu *bearer_list[];
* TIPC routines available to supported media types
*/
-void tipc_rcv(struct sk_buff *buf, struct tipc_bearer *tb_ptr);
+void tipc_rcv(struct sk_buff *skb, struct tipc_bearer *tb_ptr);
int tipc_enable_bearer(const char *bearer_name, u32 disc_domain, u32 priority);
int tipc_disable_bearer(const char *name);
diff --git a/net/tipc/link.c b/net/tipc/link.c
index d9c2310..0e04508 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -966,29 +966,17 @@ void tipc_link_retransmit(struct tipc_link *l_ptr, struct sk_buff *skb,
}
}
-/**
- * link_insert_deferred_queue - insert deferred messages back into receive chain
- */
-static struct sk_buff *link_insert_deferred_queue(struct tipc_link *l_ptr,
- struct sk_buff *buf)
+static void link_retrieve_defq(struct tipc_link *link,
+ struct sk_buff_head *list)
{
- struct sk_buff_head head;
- struct sk_buff *skb = NULL;
u32 seq_no;
- if (skb_queue_empty(&l_ptr->deferred_queue))
- return buf;
-
- seq_no = buf_seqno(skb_peek(&l_ptr->deferred_queue));
- if (seq_no == mod(l_ptr->next_in_no)) {
- __skb_queue_head_init(&head);
- skb_queue_splice_tail_init(&l_ptr->deferred_queue, &head);
- skb = head.next;
- skb->prev = NULL;
- head.prev->next = buf;
- head.prev->prev = NULL;
- }
- return skb;
+ if (skb_queue_empty(&link->deferred_queue))
+ return;
+
+ seq_no = buf_seqno(skb_peek(&link->deferred_queue));
+ if (seq_no == mod(link->next_in_no))
+ skb_queue_splice_tail_init(&link->deferred_queue, list);
}
/**
@@ -1048,43 +1036,43 @@ static int link_recv_buf_validate(struct sk_buff *buf)
/**
* tipc_rcv - process TIPC packets/messages arriving from off-node
- * @head: pointer to message buffer chain
+ * @skb: TIPC packet
* @b_ptr: pointer to bearer message arrived on
*
* Invoked with no locks held. Bearer pointer must point to a valid bearer
* structure (i.e. cannot be NULL), but bearer can be inactive.
*/
-void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr)
+void tipc_rcv(struct sk_buff *skb, struct tipc_bearer *b_ptr)
{
- while (head) {
- struct tipc_node *n_ptr;
- struct tipc_link *l_ptr;
- struct sk_buff *buf = head;
- struct sk_buff *skb1, *tmp;
- struct tipc_msg *msg;
- u32 seq_no;
- u32 ackd;
- u32 released;
+ struct sk_buff_head head;
+ struct tipc_node *n_ptr;
+ struct tipc_link *l_ptr;
+ struct sk_buff *skb1, *tmp;
+ struct tipc_msg *msg;
+ u32 seq_no;
+ u32 ackd;
+ u32 released;
- head = head->next;
- buf->next = NULL;
+ __skb_queue_head_init(&head);
+ __skb_queue_tail(&head, skb);
+ while ((skb = __skb_dequeue(&head))) {
/* Ensure message is well-formed */
- if (unlikely(!link_recv_buf_validate(buf)))
+ if (unlikely(!link_recv_buf_validate(skb)))
goto discard;
/* Ensure message data is a single contiguous unit */
- if (unlikely(skb_linearize(buf)))
+ if (unlikely(skb_linearize(skb)))
goto discard;
/* Handle arrival of a non-unicast link message */
- msg = buf_msg(buf);
+ msg = buf_msg(skb);
if (unlikely(msg_non_seq(msg))) {
if (msg_user(msg) == LINK_CONFIG)
- tipc_disc_rcv(buf, b_ptr);
+ tipc_disc_rcv(skb, b_ptr);
else
- tipc_bclink_rcv(buf);
+ tipc_bclink_rcv(skb);
continue;
}
@@ -1145,8 +1133,8 @@ void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr)
/* Process the incoming packet */
if (unlikely(!link_working_working(l_ptr))) {
if (msg_user(msg) == LINK_PROTOCOL) {
- tipc_link_proto_rcv(l_ptr, buf);
- head = link_insert_deferred_queue(l_ptr, head);
+ tipc_link_proto_rcv(l_ptr, skb);
+ link_retrieve_defq(l_ptr, &head);
tipc_node_unlock(n_ptr);
continue;
}
@@ -1156,8 +1144,7 @@ void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr)
if (link_working_working(l_ptr)) {
/* Re-insert buffer in front of queue */
- buf->next = head;
- head = buf;
+ __skb_queue_head(&head, skb);
tipc_node_unlock(n_ptr);
continue;
}
@@ -1166,33 +1153,33 @@ void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr)
/* Link is now in state WORKING_WORKING */
if (unlikely(seq_no != mod(l_ptr->next_in_no))) {
- link_handle_out_of_seq_msg(l_ptr, buf);
- head = link_insert_deferred_queue(l_ptr, head);
+ link_handle_out_of_seq_msg(l_ptr, skb);
+ link_retrieve_defq(l_ptr, &head);
tipc_node_unlock(n_ptr);
continue;
}
l_ptr->next_in_no++;
if (unlikely(!skb_queue_empty(&l_ptr->deferred_queue)))
- head = link_insert_deferred_queue(l_ptr, head);
+ link_retrieve_defq(l_ptr, &head);
if (unlikely(++l_ptr->unacked_window >= TIPC_MIN_LINK_WIN)) {
l_ptr->stats.sent_acks++;
tipc_link_proto_xmit(l_ptr, STATE_MSG, 0, 0, 0, 0, 0);
}
- if (tipc_link_prepare_input(l_ptr, &buf)) {
+ if (tipc_link_prepare_input(l_ptr, &skb)) {
tipc_node_unlock(n_ptr);
continue;
}
tipc_node_unlock(n_ptr);
- msg = buf_msg(buf);
- if (tipc_link_input(l_ptr, buf) != 0)
+
+ if (tipc_link_input(l_ptr, skb) != 0)
goto discard;
continue;
unlock_discard:
tipc_node_unlock(n_ptr);
discard:
- kfree_skb(buf);
+ kfree_skb(skb);
}
}
--
1.7.9.5
------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=157005751&iu=/4140/ostg.clktrk
^ 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