* Re: [patch] korina: use resource_size()
From: David Miller @ 2010-03-23 1:33 UTC (permalink / raw)
To: error27; +Cc: netdev, kernel-janitors, n0-1, jpirko, shemminger, joe
In-Reply-To: <20100322121145.GK21571@bicker>
From: Dan Carpenter <error27@gmail.com>
Date: Mon, 22 Mar 2010 15:11:45 +0300
> This size calculation is wrong. It should be end - start + 1.
> Use resource_size() to calculate it correctly.
>
> Signed-off-by: Dan Carpenter <error27@gmail.com>
Applied.
^ permalink raw reply
* Re: [net-next-2.6 PATCH 3/3] igb: add support for Intel I350 Gigabit Network Connection
From: David Miller @ 2010-03-23 1:30 UTC (permalink / raw)
To: jeffrey.t.kirsher; +Cc: netdev, gospo, alexander.h.duyck, james.r.hearn
In-Reply-To: <20100323000804.19939.27632.stgit@localhost.localdomain>
From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Mon, 22 Mar 2010 17:08:06 -0700
> From: Alexander Duyck <alexander.h.duyck@intel.com>
>
> This patch adds support for the the I350 Gigabit network connection which
> is the follow-on part to the 82580.
>
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
> CC: James Hearn <james.r.hearn@intel.com>
Applied.
^ permalink raw reply
* Re: [net-next-2.6 PATCH 2/3] igb: cleanup usage of virtualization registers
From: David Miller @ 2010-03-23 1:30 UTC (permalink / raw)
To: jeffrey.t.kirsher; +Cc: netdev, gospo, alexander.h.duyck
In-Reply-To: <20100323000744.19939.65933.stgit@localhost.localdomain>
From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Mon, 22 Mar 2010 17:07:46 -0700
> From: Alexander Duyck <alexander.h.duyck@intel.com>
>
> The igb driver was incorrectly attempting to write to registers that do not
> exist on 82580 hardware. This wasn't causing any issues that I can tell,
> but it is not recommended behavior either so I have changed this so we are
> determining which registers to write to based on mac type.
>
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Applied.
^ permalink raw reply
* Re: [net-next-2.6 PATCH 1/3] igb: Do not overwrite mdicnfg register when accessing 82580 phy
From: David Miller @ 2010-03-23 1:30 UTC (permalink / raw)
To: jeffrey.t.kirsher; +Cc: netdev, gospo, alexander.h.duyck
In-Reply-To: <20100323000724.19939.17642.stgit@localhost.localdomain>
From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Mon, 22 Mar 2010 17:07:26 -0700
> From: Alexander Duyck <alexander.h.duyck@intel.com>
>
> This change removes the extra configuration we were doing on the mdicnfg
> register which should be set by EEPROM and which we should not need to
> write again afterwards. This code was a holdover from some earlier
> development work that was being done on a board with an incomplete EEPROM
> and is no longer needed now that the production EEPROMs are in place.
>
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Applied.
^ permalink raw reply
* Re: [net-next-2.6 PATCH] igb: add IntMode module parameter
From: David Miller @ 2010-03-23 1:28 UTC (permalink / raw)
To: jeffrey.t.kirsher; +Cc: netdev, gospo, nicholasx.d.nunley
In-Reply-To: <20100323000519.19902.14802.stgit@localhost.localdomain>
From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Mon, 22 Mar 2010 17:06:20 -0700
> From: Nick Nunley <nicholasx.d.nunley@intel.com>
>
> This patch adds the IntMode module parameter to igb. This
> allows user selection of interrupt mode (MSI-X, MSI, legacy)
> on driver load.
>
> Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
And IGB is special and needs this for what reason?
If there are real interrupt issues then:
1) Diagnosing via verification of the problem can be simply performed
by asking the user to pass the generic kernel boot option to
globally disable MSI and friends.
2) Specifically disabling it for buggy parts belongs in the
driver or in the PCI quirk tables.
So from just about any angle, this module option makes no sense.
So I'm not applying it, sorry.
^ permalink raw reply
* Re: pull request: wireless-next-2.6 2010-03-19
From: David Miller @ 2010-03-23 1:16 UTC (permalink / raw)
To: linville-2XuSBdqkA4R54TAoqtyWWQ
Cc: linux-wireless-u79uwXL29TY76Z2rM5mHXA,
netdev-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <20100319204507.GF9552-2XuSBdqkA4R54TAoqtyWWQ@public.gmane.org>
From: "John W. Linville" <linville-2XuSBdqkA4R54TAoqtyWWQ@public.gmane.org>
Date: Fri, 19 Mar 2010 16:45:08 -0400
> Here is the first big dump of wireless bit for the 2.6.35 cycle...
>
> Included are the usual driver updates, etc. In particular, the Intel
> and Nokia guys have been busy as usual. :-)
Pulled, I'll push out to kernel.org after some build smoke
tests.
Thanks!
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: rps: Fix build with CONFIG_SYSFS enabled
From: David Miller @ 2010-03-23 1:07 UTC (permalink / raw)
To: randy.dunlap; +Cc: therbert, netdev
In-Reply-To: <4BA7E5BF.1080805@oracle.com>
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Mon, 22 Mar 2010 14:48:47 -0700
> Dave, this patch and Eric's patch both build OK. Your choice...
Yep, I took Tom's, thanks for testing Randy.
^ permalink raw reply
* Re: rps: Fix build with CONFIG_SYSFS enabled
From: David Miller @ 2010-03-23 1:07 UTC (permalink / raw)
To: therbert; +Cc: netdev
In-Reply-To: <alpine.DEB.1.00.1003221355380.25849@pokey.mtv.corp.google.com>
From: Tom Herbert <therbert@google.com>
Date: Mon, 22 Mar 2010 13:57:56 -0700 (PDT)
> Fix build with CONFIG_SYSFS not enabled.
>
> Signed-off-by: Tom Herbert <therbert@google.com>
Applied, thanks everyone.
^ permalink raw reply
* Re: linux-next: Tree for March 22 (net-sysfs.c)
From: David Miller @ 2010-03-23 1:05 UTC (permalink / raw)
To: eric.dumazet
Cc: randy.dunlap, sfr, linux-next, linux-kernel, netdev, therbert
In-Reply-To: <1269290922.3043.9.camel@edumazet-laptop>
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 22 Mar 2010 21:48:42 +0100
> [PATCH net-next-2.6] net: RPS depends on CONFIG_SYSFS
>
> Randy Dunlap found net/core/net-sysfs.c could not compile if
> CONFIG_SYSFS=n
>
> Defines CONFIG_RPS in net/Kconfig to let user disable RPS if wanted.
>
> Reported-by: Randy Dunlap <randy.dunlap@oracle.com>
> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
This is interesting but I'd like to be able to use some
of the RPS infrastructure unconditionally so I'm not ready
to apply something like this yet.
So I'll add Tom's fix for now, thanks Eric!
^ permalink raw reply
* Re: [RFC Patch 3/3] bonding: make bonding support netpoll
From: Andy Gospodarek @ 2010-03-23 0:56 UTC (permalink / raw)
To: Amerigo Wang
Cc: linux-kernel, netdev, bridge, Andy Gospodarek, Neil Horman,
Stephen Hemminger, bonding-devel, Jay Vosburgh, David Miller
In-Reply-To: <20100322082123.4967.77433.sendpatchset@localhost.localdomain>
On Mon, Mar 22, 2010 at 04:17:40AM -0400, Amerigo Wang wrote:
>
> Based on Andy's work, but I modify a lot.
>
> Similar to the patch for bridge, this patch does:
>
> 1) implement the 4 methods to support netpoll for bonding;
>
> 2) modify netpoll during forwarding packets in bonding;
>
> 3) disable netpoll support of bridge when a netpoll-unabled device
> is added to bonding;
>
> 4) enable netpoll support when all underlying devices support netpoll.
>
> Cc: Andy Gospodarek <gospo@redhat.com>
> Cc: Neil Horman <nhorman@tuxdriver.com>
> Cc: Jay Vosburgh <fubar@us.ibm.com>
> Cc: David Miller <davem@davemloft.net>
> Signed-off-by: WANG Cong <amwang@redhat.com>
>
How much testing was done on this?
One of the potential problems with this code is how gracefully the
system can handle tear-down of interfaces or removal of the bonding
module when netconsole is active. Was that tested heavily?
>
> ---
> Index: linux-2.6/drivers/net/bonding/bond_main.c
> ===================================================================
> --- linux-2.6.orig/drivers/net/bonding/bond_main.c
> +++ linux-2.6/drivers/net/bonding/bond_main.c
> @@ -59,6 +59,7 @@
> #include <linux/uaccess.h>
> #include <linux/errno.h>
> #include <linux/netdevice.h>
> +#include <linux/netpoll.h>
> #include <linux/inetdevice.h>
> #include <linux/igmp.h>
> #include <linux/etherdevice.h>
> @@ -430,7 +431,17 @@ int bond_dev_queue_xmit(struct bonding *
> }
>
> skb->priority = 1;
> - dev_queue_xmit(skb);
> +#ifdef CONFIG_NET_POLL_CONTROLLER
> + if (bond->dev->priv_flags & IFF_IN_NETPOLL) {
> + bond->dev->npinfo->netpoll->dev = skb->dev;
> + if (!slave_dev->npinfo)
> + slave_dev->npinfo = bond->dev->npinfo;
> + slave_dev->priv_flags |= IFF_IN_NETPOLL;
> + netpoll_send_skb(bond->dev->npinfo->netpoll, skb);
> + slave_dev->priv_flags &= ~IFF_IN_NETPOLL;
> + } else
> +#endif
> + dev_queue_xmit(skb);
>
> return 0;
> }
> @@ -1324,6 +1335,87 @@ static void bond_detach_slave(struct bon
> bond->slave_cnt--;
> }
>
> +#ifdef CONFIG_NET_POLL_CONTROLLER
> +static bool slaves_support_netpoll(struct net_device *bond_dev)
> +{
> + struct bonding *bond = netdev_priv(bond_dev);
> + struct slave *slave;
> + int i = 0;
> + bool ret = true;
> +
> + read_lock_bh(&bond->lock);
> + bond_for_each_slave(bond, slave, i) {
> + if ((slave->dev->priv_flags & IFF_DISABLE_NETPOLL)
> + || !slave->dev->netdev_ops->ndo_poll_controller)
> + ret = false;
> + }
> + read_unlock_bh(&bond->lock);
> + return i != 0 && ret;
> +}
> +
> +static void bond_poll_controller(struct net_device *bond_dev)
> +{
> + struct bonding *bond = netdev_priv(bond_dev);
> + struct slave *slave;
> + int i;
> +
> + read_lock(&bond->lock);
> + bond_for_each_slave(bond, slave, i) {
> + if (slave->dev->netdev_ops->ndo_poll_controller)
> + netpoll_poll_dev(slave->dev);
> + }
> + read_unlock(&bond->lock);
> +}
> +
> +static void bond_netpoll_setup(struct net_device *bond_dev,
> + struct netpoll_info *npinfo)
> +{
> + struct bonding *bond = netdev_priv(bond_dev);
> + struct slave *slave;
> + int i;
> +
> + write_lock_bh(&bond->lock);
> + bond_for_each_slave(bond, slave, i) {
> + if (slave->dev)
> + slave->dev->npinfo = npinfo;
> + }
> + write_unlock_bh(&bond->lock);
> +}
> +
> +static void bond_netpoll_cleanup(struct net_device *bond_dev)
> +{
> + struct bonding *bond = netdev_priv(bond_dev);
> + struct slave *slave;
> + const struct net_device_ops *ops;
> + int i;
> +
> + write_lock_bh(&bond->lock);
> + bond_dev->npinfo = NULL;
> + bond_for_each_slave(bond, slave, i) {
> + if (slave->dev) {
> + ops = slave->dev->netdev_ops;
> + if (ops->ndo_netpoll_cleanup)
> + ops->ndo_netpoll_cleanup(slave->dev);
> + else
> + slave->dev->npinfo = NULL;
> + }
> + }
> + write_unlock_bh(&bond->lock);
> +}
> +
> +static int bond_netpoll_xmit(struct netpoll *np, struct sk_buff *skb,
> + struct net_device *dev)
> +{
> + int ret;
> +
> + dev->priv_flags |= IFF_IN_NETPOLL;
> + ret = dev->netdev_ops->ndo_start_xmit(skb, dev);
> + np->dev = dev;
> + dev->priv_flags &= ~IFF_IN_NETPOLL;
> + return ret;
> +}
> +#endif
> +
> /*---------------------------------- IOCTL ----------------------------------*/
>
> static int bond_sethwaddr(struct net_device *bond_dev,
> @@ -1741,6 +1833,18 @@ int bond_enslave(struct net_device *bond
> new_slave->state == BOND_STATE_ACTIVE ? "n active" : " backup",
> new_slave->link != BOND_LINK_DOWN ? "n up" : " down");
>
> +#ifdef CONFIG_NET_POLL_CONTROLLER
> + if (slaves_support_netpoll(bond_dev)) {
> + bond_dev->priv_flags &= ~IFF_DISABLE_NETPOLL;
> + if (bond_dev->npinfo)
> + slave_dev->npinfo = bond_dev->npinfo;
> + } else if (!(bond_dev->priv_flags & IFF_DISABLE_NETPOLL)) {
> + bond_dev->priv_flags |= IFF_DISABLE_NETPOLL;
> + pr_info("New slave device %s does not support netpoll\n",
> + slave_dev->name);
> + pr_info("Disabling netpoll support for %s\n", bond_dev->name);
> + }
> +#endif
> /* enslave is successful */
> return 0;
>
> @@ -1924,6 +2028,15 @@ int bond_release(struct net_device *bond
>
> netdev_set_master(slave_dev, NULL);
>
> +#ifdef CONFIG_NET_POLL_CONTROLLER
> + if (slaves_support_netpoll(bond_dev))
> + bond_dev->priv_flags &= ~IFF_DISABLE_NETPOLL;
> + if (slave_dev->netdev_ops->ndo_netpoll_cleanup)
> + slave_dev->netdev_ops->ndo_netpoll_cleanup(slave_dev);
> + else
> + slave_dev->npinfo = NULL;
> +#endif
> +
> /* close slave before restoring its mac address */
> dev_close(slave_dev);
>
> @@ -2032,6 +2145,9 @@ static int bond_release_all(struct net_d
>
> netdev_set_master(slave_dev, NULL);
>
> +#ifdef CONFIG_NET_POLL_CONTROLLER
> + slave_dev->npinfo = NULL;
> +#endif
> /* close slave before restoring its mac address */
> dev_close(slave_dev);
>
> @@ -4424,6 +4540,12 @@ static const struct net_device_ops bond_
> .ndo_vlan_rx_register = bond_vlan_rx_register,
> .ndo_vlan_rx_add_vid = bond_vlan_rx_add_vid,
> .ndo_vlan_rx_kill_vid = bond_vlan_rx_kill_vid,
> +#ifdef CONFIG_NET_POLL_CONTROLLER
> + .ndo_netpoll_setup = bond_netpoll_setup,
> + .ndo_netpoll_xmit = bond_netpoll_xmit,
> + .ndo_netpoll_cleanup = bond_netpoll_cleanup,
> + .ndo_poll_controller = bond_poll_controller,
> +#endif
> };
>
> static void bond_setup(struct net_device *bond_dev)
^ permalink raw reply
* [PATCH 15/18] net: Provide config option for enabling NAPI tracing
From: Jan Kara @ 2010-03-23 0:32 UTC (permalink / raw)
To: LKML; +Cc: Jan Kara, David S. Miller, netdev
In-Reply-To: <1269304340-25372-1-git-send-email-jack@suse.cz>
CC: "David S. Miller" <davem@davemloft.net>
CC: netdev@vger.kernel.org
Signed-off-by: Jan Kara <jack@suse.cz>
---
include/trace/events/napi.h | 2 ++
kernel/trace/Kconfig | 9 +++++++++
2 files changed, 11 insertions(+), 0 deletions(-)
diff --git a/include/trace/events/napi.h b/include/trace/events/napi.h
index a8989c4..70cb4dd 100644
--- a/include/trace/events/napi.h
+++ b/include/trace/events/napi.h
@@ -2,6 +2,8 @@
#define _TRACE_NAPI_H_
#include <linux/netdevice.h>
+#undef TRACE_CONFIG
+#define TRACE_CONFIG CONFIG_NAPI_TRACER
#include <linux/tracepoint.h>
DECLARE_TRACE(napi_poll,
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index a02d577..88b556c 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -508,6 +508,15 @@ config SKB_TRACER
If unsure, say N.
+config NAPI_TRACER
+ bool "NAPI tracing"
+ depends on NET
+ select GENERIC_TRACER
+ help
+ Provide tracing for NAPI functions.
+
+ If unsure, say N.
+
config KPROBE_EVENT
depends on KPROBES
depends on HAVE_REGS_AND_STACK_ACCESS_API
--
1.6.4.2
^ permalink raw reply related
* [PATCH 14/18] net: Provide config option for enabling SKB tracing
From: Jan Kara @ 2010-03-23 0:32 UTC (permalink / raw)
To: LKML; +Cc: Jan Kara, David S. Miller, netdev
In-Reply-To: <1269304340-25372-1-git-send-email-jack@suse.cz>
CC: "David S. Miller" <davem@davemloft.net>
CC: netdev@vger.kernel.org
Signed-off-by: Jan Kara <jack@suse.cz>
---
include/trace/events/skb.h | 2 ++
kernel/trace/Kconfig | 10 ++++++++++
2 files changed, 12 insertions(+), 0 deletions(-)
diff --git a/include/trace/events/skb.h b/include/trace/events/skb.h
index 4b2be6d..cd340df 100644
--- a/include/trace/events/skb.h
+++ b/include/trace/events/skb.h
@@ -1,5 +1,7 @@
#undef TRACE_SYSTEM
+#undef TRACE_CONFIG
#define TRACE_SYSTEM skb
+#define TRACE_CONFIG CONFIG_SKB_TRACER
#if !defined(_TRACE_SKB_H) || defined(TRACE_HEADER_MULTI_READ)
#define _TRACE_SKB_H
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index a48ff49..a02d577 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -498,6 +498,16 @@ config JBD2_TRACER
If unsure, say N.
+config SKB_TRACER
+ bool "SKB tracing"
+ depends on NET
+ select GENERIC_TRACER
+ help
+ Provide tracing for copying datagrams into iovecs and freeing
+ skbs.
+
+ If unsure, say N.
+
config KPROBE_EVENT
depends on KPROBES
depends on HAVE_REGS_AND_STACK_ACCESS_API
--
1.6.4.2
^ permalink raw reply related
* [net-next-2.6 PATCH 3/3] igb: add support for Intel I350 Gigabit Network Connection
From: Jeff Kirsher @ 2010-03-23 0:08 UTC (permalink / raw)
To: davem; +Cc: netdev, gospo, Alexander Duyck, Jeff Kirsher, James Hearn
In-Reply-To: <20100323000724.19939.17642.stgit@localhost.localdomain>
From: Alexander Duyck <alexander.h.duyck@intel.com>
This patch adds support for the the I350 Gigabit network connection which
is the follow-on part to the 82580.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
CC: James Hearn <james.r.hearn@intel.com>
---
drivers/net/igb/e1000_82575.c | 13 +++++++++-
drivers/net/igb/e1000_82575.h | 7 +++--
drivers/net/igb/e1000_defines.h | 1 +
drivers/net/igb/e1000_hw.h | 5 ++++
drivers/net/igb/igb_ethtool.c | 50 +++++++++++++++++++++++++++++++++++++++
drivers/net/igb/igb_main.c | 12 +++++++++
6 files changed, 83 insertions(+), 5 deletions(-)
diff --git a/drivers/net/igb/e1000_82575.c b/drivers/net/igb/e1000_82575.c
index ea944f5..430631f 100644
--- a/drivers/net/igb/e1000_82575.c
+++ b/drivers/net/igb/e1000_82575.c
@@ -105,6 +105,12 @@ static s32 igb_get_invariants_82575(struct e1000_hw *hw)
case E1000_DEV_ID_82580_COPPER_DUAL:
mac->type = e1000_82580;
break;
+ case E1000_DEV_ID_I350_COPPER:
+ case E1000_DEV_ID_I350_FIBER:
+ case E1000_DEV_ID_I350_SERDES:
+ case E1000_DEV_ID_I350_SGMII:
+ mac->type = e1000_i350;
+ break;
default:
return -E1000_ERR_MAC_INIT;
break;
@@ -154,8 +160,10 @@ static s32 igb_get_invariants_82575(struct e1000_hw *hw)
mac->rar_entry_count = E1000_RAR_ENTRIES_82576;
if (mac->type == e1000_82580)
mac->rar_entry_count = E1000_RAR_ENTRIES_82580;
+ if (mac->type == e1000_i350)
+ mac->rar_entry_count = E1000_RAR_ENTRIES_I350;
/* reset */
- if (mac->type == e1000_82580)
+ if (mac->type >= e1000_82580)
mac->ops.reset_hw = igb_reset_hw_82580;
else
mac->ops.reset_hw = igb_reset_hw_82575;
@@ -226,7 +234,7 @@ static s32 igb_get_invariants_82575(struct e1000_hw *hw)
phy->ops.reset = igb_phy_hw_reset_sgmii_82575;
phy->ops.read_reg = igb_read_phy_reg_sgmii_82575;
phy->ops.write_reg = igb_write_phy_reg_sgmii_82575;
- } else if (hw->mac.type == e1000_82580) {
+ } else if (hw->mac.type >= e1000_82580) {
phy->ops.reset = igb_phy_hw_reset;
phy->ops.read_reg = igb_read_phy_reg_82580;
phy->ops.write_reg = igb_write_phy_reg_82580;
@@ -262,6 +270,7 @@ static s32 igb_get_invariants_82575(struct e1000_hw *hw)
phy->ops.set_d3_lplu_state = igb_set_d3_lplu_state;
break;
case I82580_I_PHY_ID:
+ case I350_I_PHY_ID:
phy->type = e1000_phy_82580;
phy->ops.force_speed_duplex = igb_phy_force_speed_duplex_82580;
phy->ops.get_cable_length = igb_get_cable_length_82580;
diff --git a/drivers/net/igb/e1000_82575.h b/drivers/net/igb/e1000_82575.h
index fbe1c99..c1cad8a 100644
--- a/drivers/net/igb/e1000_82575.h
+++ b/drivers/net/igb/e1000_82575.h
@@ -38,9 +38,10 @@ extern void igb_rx_fifo_flush_82575(struct e1000_hw *hw);
(ID_LED_DEF1_DEF2 << 4) | \
(ID_LED_OFF1_ON2))
-#define E1000_RAR_ENTRIES_82575 16
-#define E1000_RAR_ENTRIES_82576 24
-#define E1000_RAR_ENTRIES_82580 24
+#define E1000_RAR_ENTRIES_82575 16
+#define E1000_RAR_ENTRIES_82576 24
+#define E1000_RAR_ENTRIES_82580 24
+#define E1000_RAR_ENTRIES_I350 32
#define E1000_SW_SYNCH_MB 0x00000100
#define E1000_STAT_DEV_RST_SET 0x00100000
diff --git a/drivers/net/igb/e1000_defines.h b/drivers/net/igb/e1000_defines.h
index fe6cf1b..31d24e0 100644
--- a/drivers/net/igb/e1000_defines.h
+++ b/drivers/net/igb/e1000_defines.h
@@ -629,6 +629,7 @@
#define M88E1111_I_PHY_ID 0x01410CC0
#define IGP03E1000_E_PHY_ID 0x02A80390
#define I82580_I_PHY_ID 0x015403A0
+#define I350_I_PHY_ID 0x015403B0
#define M88_VENDOR 0x0141
/* M88E1000 Specific Registers */
diff --git a/drivers/net/igb/e1000_hw.h b/drivers/net/igb/e1000_hw.h
index 82a533f..593d5fa 100644
--- a/drivers/net/igb/e1000_hw.h
+++ b/drivers/net/igb/e1000_hw.h
@@ -53,6 +53,10 @@ struct e1000_hw;
#define E1000_DEV_ID_82580_SERDES 0x1510
#define E1000_DEV_ID_82580_SGMII 0x1511
#define E1000_DEV_ID_82580_COPPER_DUAL 0x1516
+#define E1000_DEV_ID_I350_COPPER 0x1521
+#define E1000_DEV_ID_I350_FIBER 0x1522
+#define E1000_DEV_ID_I350_SERDES 0x1523
+#define E1000_DEV_ID_I350_SGMII 0x1524
#define E1000_REVISION_2 2
#define E1000_REVISION_4 4
@@ -72,6 +76,7 @@ enum e1000_mac_type {
e1000_82575,
e1000_82576,
e1000_82580,
+ e1000_i350,
e1000_num_macs /* List is 1-based, so subtract 1 for true count. */
};
diff --git a/drivers/net/igb/igb_ethtool.c b/drivers/net/igb/igb_ethtool.c
index a4cead1..1d4ee41 100644
--- a/drivers/net/igb/igb_ethtool.c
+++ b/drivers/net/igb/igb_ethtool.c
@@ -901,6 +901,49 @@ struct igb_reg_test {
#define TABLE64_TEST_LO 5
#define TABLE64_TEST_HI 6
+/* i350 reg test */
+static struct igb_reg_test reg_test_i350[] = {
+ { E1000_FCAL, 0x100, 1, PATTERN_TEST, 0xFFFFFFFF, 0xFFFFFFFF },
+ { E1000_FCAH, 0x100, 1, PATTERN_TEST, 0x0000FFFF, 0xFFFFFFFF },
+ { E1000_FCT, 0x100, 1, PATTERN_TEST, 0x0000FFFF, 0xFFFFFFFF },
+ { E1000_VET, 0x100, 1, PATTERN_TEST, 0xFFFF0000, 0xFFFF0000 },
+ { E1000_RDBAL(0), 0x100, 4, PATTERN_TEST, 0xFFFFFF80, 0xFFFFFFFF },
+ { E1000_RDBAH(0), 0x100, 4, PATTERN_TEST, 0xFFFFFFFF, 0xFFFFFFFF },
+ { E1000_RDLEN(0), 0x100, 4, PATTERN_TEST, 0x000FFFF0, 0x000FFFFF },
+ { E1000_RDBAL(4), 0x40, 4, PATTERN_TEST, 0xFFFFFF80, 0xFFFFFFFF },
+ { E1000_RDBAH(4), 0x40, 4, PATTERN_TEST, 0xFFFFFFFF, 0xFFFFFFFF },
+ { E1000_RDLEN(4), 0x40, 4, PATTERN_TEST, 0x000FFFF0, 0x000FFFFF },
+ /* RDH is read-only for i350, only test RDT. */
+ { E1000_RDT(0), 0x100, 4, PATTERN_TEST, 0x0000FFFF, 0x0000FFFF },
+ { E1000_RDT(4), 0x40, 4, PATTERN_TEST, 0x0000FFFF, 0x0000FFFF },
+ { E1000_FCRTH, 0x100, 1, PATTERN_TEST, 0x0000FFF0, 0x0000FFF0 },
+ { E1000_FCTTV, 0x100, 1, PATTERN_TEST, 0x0000FFFF, 0x0000FFFF },
+ { E1000_TIPG, 0x100, 1, PATTERN_TEST, 0x3FFFFFFF, 0x3FFFFFFF },
+ { E1000_TDBAL(0), 0x100, 4, PATTERN_TEST, 0xFFFFFF80, 0xFFFFFFFF },
+ { E1000_TDBAH(0), 0x100, 4, PATTERN_TEST, 0xFFFFFFFF, 0xFFFFFFFF },
+ { E1000_TDLEN(0), 0x100, 4, PATTERN_TEST, 0x000FFFF0, 0x000FFFFF },
+ { E1000_TDBAL(4), 0x40, 4, PATTERN_TEST, 0xFFFFFF80, 0xFFFFFFFF },
+ { E1000_TDBAH(4), 0x40, 4, PATTERN_TEST, 0xFFFFFFFF, 0xFFFFFFFF },
+ { E1000_TDLEN(4), 0x40, 4, PATTERN_TEST, 0x000FFFF0, 0x000FFFFF },
+ { E1000_TDT(0), 0x100, 4, PATTERN_TEST, 0x0000FFFF, 0x0000FFFF },
+ { E1000_TDT(4), 0x40, 4, PATTERN_TEST, 0x0000FFFF, 0x0000FFFF },
+ { E1000_RCTL, 0x100, 1, SET_READ_TEST, 0xFFFFFFFF, 0x00000000 },
+ { E1000_RCTL, 0x100, 1, SET_READ_TEST, 0x04CFB0FE, 0x003FFFFB },
+ { E1000_RCTL, 0x100, 1, SET_READ_TEST, 0x04CFB0FE, 0xFFFFFFFF },
+ { E1000_TCTL, 0x100, 1, SET_READ_TEST, 0xFFFFFFFF, 0x00000000 },
+ { E1000_RA, 0, 16, TABLE64_TEST_LO,
+ 0xFFFFFFFF, 0xFFFFFFFF },
+ { E1000_RA, 0, 16, TABLE64_TEST_HI,
+ 0xC3FFFFFF, 0xFFFFFFFF },
+ { E1000_RA2, 0, 16, TABLE64_TEST_LO,
+ 0xFFFFFFFF, 0xFFFFFFFF },
+ { E1000_RA2, 0, 16, TABLE64_TEST_HI,
+ 0xC3FFFFFF, 0xFFFFFFFF },
+ { E1000_MTA, 0, 128, TABLE32_TEST,
+ 0xFFFFFFFF, 0xFFFFFFFF },
+ { 0, 0, 0, 0 }
+};
+
/* 82580 reg test */
static struct igb_reg_test reg_test_82580[] = {
{ E1000_FCAL, 0x100, 1, PATTERN_TEST, 0xFFFFFFFF, 0xFFFFFFFF },
@@ -1076,6 +1119,10 @@ static int igb_reg_test(struct igb_adapter *adapter, u64 *data)
u32 i, toggle;
switch (adapter->hw.mac.type) {
+ case e1000_i350:
+ test = reg_test_i350;
+ toggle = 0x7FEFF3FF;
+ break;
case e1000_82580:
test = reg_test_82580;
toggle = 0x7FEFF3FF;
@@ -1237,6 +1284,9 @@ static int igb_intr_test(struct igb_adapter *adapter, u64 *data)
case e1000_82580:
ics_mask = 0x77DCFED5;
break;
+ case e1000_i350:
+ ics_mask = 0x77DCFED5;
+ break;
default:
ics_mask = 0x7FFFFFFF;
break;
diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index 61029e8..d732280 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -61,6 +61,10 @@ static const struct e1000_info *igb_info_tbl[] = {
};
static DEFINE_PCI_DEVICE_TABLE(igb_pci_tbl) = {
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_I350_COPPER), board_82575 },
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_I350_FIBER), board_82575 },
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_I350_SERDES), board_82575 },
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_I350_SGMII), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82580_COPPER), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82580_FIBER), board_82575 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82580_SERDES), board_82575 },
@@ -327,6 +331,7 @@ static void igb_cache_ring_register(struct igb_adapter *adapter)
}
case e1000_82575:
case e1000_82580:
+ case e1000_i350:
default:
for (; i < adapter->num_rx_queues; i++)
adapter->rx_ring[i]->reg_idx = rbase_offset + i;
@@ -470,6 +475,7 @@ static void igb_assign_vector(struct igb_q_vector *q_vector, int msix_vector)
q_vector->eims_value = 1 << msix_vector;
break;
case e1000_82580:
+ case e1000_i350:
/* 82580 uses the same table-based approach as 82576 but has fewer
entries as a result we carry over for queues greater than 4. */
if (rx_queue > IGB_N0_QUEUE) {
@@ -550,6 +556,7 @@ static void igb_configure_msix(struct igb_adapter *adapter)
case e1000_82576:
case e1000_82580:
+ case e1000_i350:
/* Turn on MSI-X capability first, or our settings
* won't stick. And it will take days to debug. */
wr32(E1000_GPIE, E1000_GPIE_MSIX_MODE |
@@ -1271,6 +1278,7 @@ void igb_reset(struct igb_adapter *adapter)
* To take effect CTRL.RST is required.
*/
switch (mac->type) {
+ case e1000_i350:
case e1000_82580:
pba = rd32(E1000_RXPBS);
pba = igb_rxpbs_adjust_82580(pba);
@@ -1847,6 +1855,7 @@ static void igb_init_hw_timer(struct igb_adapter *adapter)
struct e1000_hw *hw = &adapter->hw;
switch (hw->mac.type) {
+ case e1000_i350:
case e1000_82580:
memset(&adapter->cycles, 0, sizeof(adapter->cycles));
adapter->cycles.read = igb_read_clock;
@@ -2362,6 +2371,7 @@ static void igb_setup_mrqc(struct igb_adapter *adapter)
if (adapter->vfs_allocated_count) {
/* 82575 and 82576 supports 2 RSS queues for VMDq */
switch (hw->mac.type) {
+ case e1000_i350:
case e1000_82580:
num_rx_queues = 1;
shift = 0;
@@ -6173,6 +6183,8 @@ static void igb_vmm_control(struct igb_adapter *adapter)
reg = rd32(E1000_RPLOLR);
reg |= E1000_RPLOLR_STRVLAN;
wr32(E1000_RPLOLR, reg);
+ case e1000_i350:
+ /* none of the above registers are supported by i350 */
break;
}
^ permalink raw reply related
* [net-next-2.6 PATCH 2/3] igb: cleanup usage of virtualization registers
From: Jeff Kirsher @ 2010-03-23 0:07 UTC (permalink / raw)
To: davem; +Cc: netdev, gospo, Alexander Duyck, Jeff Kirsher
In-Reply-To: <20100323000724.19939.17642.stgit@localhost.localdomain>
From: Alexander Duyck <alexander.h.duyck@intel.com>
The igb driver was incorrectly attempting to write to registers that do not
exist on 82580 hardware. This wasn't causing any issues that I can tell,
but it is not recommended behavior either so I have changed this so we are
determining which registers to write to based on mac type.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/igb/igb_main.c | 28 ++++++++++++++++------------
1 files changed, 16 insertions(+), 12 deletions(-)
diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index 83431e9..61029e8 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -6158,19 +6158,23 @@ static void igb_vmm_control(struct igb_adapter *adapter)
struct e1000_hw *hw = &adapter->hw;
u32 reg;
- /* replication is not supported for 82575 */
- if (hw->mac.type == e1000_82575)
+ switch (hw->mac.type) {
+ case e1000_82575:
+ default:
+ /* replication is not supported for 82575 */
return;
-
- /* enable replication vlan tag stripping */
- reg = rd32(E1000_RPLOLR);
- reg |= E1000_RPLOLR_STRVLAN;
- wr32(E1000_RPLOLR, reg);
-
- /* notify HW that the MAC is adding vlan tags */
- reg = rd32(E1000_DTXCTL);
- reg |= E1000_DTXCTL_VLAN_ADDED;
- wr32(E1000_DTXCTL, reg);
+ case e1000_82576:
+ /* notify HW that the MAC is adding vlan tags */
+ reg = rd32(E1000_DTXCTL);
+ reg |= E1000_DTXCTL_VLAN_ADDED;
+ wr32(E1000_DTXCTL, reg);
+ case e1000_82580:
+ /* enable replication vlan tag stripping */
+ reg = rd32(E1000_RPLOLR);
+ reg |= E1000_RPLOLR_STRVLAN;
+ wr32(E1000_RPLOLR, reg);
+ break;
+ }
if (adapter->vfs_allocated_count) {
igb_vmdq_set_loopback_pf(hw, true);
^ permalink raw reply related
* [net-next-2.6 PATCH 1/3] igb: Do not overwrite mdicnfg register when accessing 82580 phy
From: Jeff Kirsher @ 2010-03-23 0:07 UTC (permalink / raw)
To: davem; +Cc: netdev, gospo, Alexander Duyck, Jeff Kirsher
From: Alexander Duyck <alexander.h.duyck@intel.com>
This change removes the extra configuration we were doing on the mdicnfg
register which should be set by EEPROM and which we should not need to
write again afterwards. This code was a holdover from some earlier
development work that was being done on a board with an incomplete EEPROM
and is no longer needed now that the production EEPROMs are in place.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/igb/e1000_82575.c | 20 --------------------
1 files changed, 0 insertions(+), 20 deletions(-)
diff --git a/drivers/net/igb/e1000_82575.c b/drivers/net/igb/e1000_82575.c
index 0bc990e..ea944f5 100644
--- a/drivers/net/igb/e1000_82575.c
+++ b/drivers/net/igb/e1000_82575.c
@@ -1446,7 +1446,6 @@ void igb_vmdq_set_replication_pf(struct e1000_hw *hw, bool enable)
**/
static s32 igb_read_phy_reg_82580(struct e1000_hw *hw, u32 offset, u16 *data)
{
- u32 mdicnfg = 0;
s32 ret_val;
@@ -1454,15 +1453,6 @@ static s32 igb_read_phy_reg_82580(struct e1000_hw *hw, u32 offset, u16 *data)
if (ret_val)
goto out;
- /*
- * We config the phy address in MDICNFG register now. Same bits
- * as before. The values in MDIC can be written but will be
- * ignored. This allows us to call the old function after
- * configuring the PHY address in the new register
- */
- mdicnfg = (hw->phy.addr << E1000_MDIC_PHY_SHIFT);
- wr32(E1000_MDICNFG, mdicnfg);
-
ret_val = igb_read_phy_reg_mdic(hw, offset, data);
hw->phy.ops.release(hw);
@@ -1481,7 +1471,6 @@ out:
**/
static s32 igb_write_phy_reg_82580(struct e1000_hw *hw, u32 offset, u16 data)
{
- u32 mdicnfg = 0;
s32 ret_val;
@@ -1489,15 +1478,6 @@ static s32 igb_write_phy_reg_82580(struct e1000_hw *hw, u32 offset, u16 data)
if (ret_val)
goto out;
- /*
- * We config the phy address in MDICNFG register now. Same bits
- * as before. The values in MDIC can be written but will be
- * ignored. This allows us to call the old function after
- * configuring the PHY address in the new register
- */
- mdicnfg = (hw->phy.addr << E1000_MDIC_PHY_SHIFT);
- wr32(E1000_MDICNFG, mdicnfg);
-
ret_val = igb_write_phy_reg_mdic(hw, offset, data);
hw->phy.ops.release(hw);
^ permalink raw reply related
* [net-next-2.6 PATCH] igb: add IntMode module parameter
From: Jeff Kirsher @ 2010-03-23 0:06 UTC (permalink / raw)
To: davem; +Cc: netdev, gospo, Nicholas Nunley, Jeff Kirsher
From: Nick Nunley <nicholasx.d.nunley@intel.com>
This patch adds the IntMode module parameter to igb. This
allows user selection of interrupt mode (MSI-X, MSI, legacy)
on driver load.
Signed-off-by: Nicholas Nunley <nicholasx.d.nunley@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/igb/Makefile | 3 +
drivers/net/igb/igb.h | 7 ++
drivers/net/igb/igb_main.c | 125 ++++++++++++++++++++++++----------------
drivers/net/igb/igb_param.c | 133 +++++++++++++++++++++++++++++++++++++++++++
4 files changed, 215 insertions(+), 53 deletions(-)
create mode 100644 drivers/net/igb/igb_param.c
diff --git a/drivers/net/igb/Makefile b/drivers/net/igb/Makefile
index 8372cb9..049255e 100644
--- a/drivers/net/igb/Makefile
+++ b/drivers/net/igb/Makefile
@@ -33,5 +33,6 @@
obj-$(CONFIG_IGB) += igb.o
igb-objs := igb_main.o igb_ethtool.o e1000_82575.o \
- e1000_mac.o e1000_nvm.o e1000_phy.o e1000_mbx.o
+ e1000_mac.o e1000_nvm.o e1000_phy.o e1000_mbx.o \
+ igb_param.o
diff --git a/drivers/net/igb/igb.h b/drivers/net/igb/igb.h
index a177570..78a9c28 100644
--- a/drivers/net/igb/igb.h
+++ b/drivers/net/igb/igb.h
@@ -43,6 +43,11 @@ struct igb_adapter;
/* ((1000000000ns / (6000ints/s * 1024ns)) << 2 = 648 */
#define IGB_START_ITR 648
+/* Interrupt modes, as used by the IntMode paramter */
+#define IGB_INT_MODE_LEGACY 0
+#define IGB_INT_MODE_MSI 1
+#define IGB_INT_MODE_MSIX 2
+
/* TX/RX descriptor defines */
#define IGB_DEFAULT_TXD 256
#define IGB_MIN_TXD 80
@@ -314,6 +319,7 @@ struct igb_adapter {
u16 rx_ring_count;
unsigned int vfs_allocated_count;
struct vf_data_storage *vf_data;
+ int int_mode;
u32 rss_queues;
};
@@ -358,6 +364,7 @@ extern void igb_alloc_rx_buffers_adv(struct igb_ring *, int);
extern void igb_update_stats(struct igb_adapter *);
extern bool igb_has_link(struct igb_adapter *adapter);
extern void igb_set_ethtool_ops(struct net_device *);
+extern void igb_check_options(struct igb_adapter *);
extern void igb_power_up_link(struct igb_adapter *);
static inline s32 igb_reset_phy(struct e1000_hw *hw)
diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index 45a0e4f..83431e9 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -679,6 +679,7 @@ static void igb_clear_interrupt_scheme(struct igb_adapter *adapter)
**/
static void igb_set_interrupt_capability(struct igb_adapter *adapter)
{
+ struct pci_dev *pdev = adapter->pdev;
int err;
int numvecs, i;
@@ -686,60 +687,74 @@ static void igb_set_interrupt_capability(struct igb_adapter *adapter)
adapter->num_rx_queues = adapter->rss_queues;
adapter->num_tx_queues = adapter->rss_queues;
- /* start with one vector for every rx queue */
- numvecs = adapter->num_rx_queues;
-
- /* if tx handler is separate add 1 for every tx queue */
- if (!(adapter->flags & IGB_FLAG_QUEUE_PAIRS))
- numvecs += adapter->num_tx_queues;
-
- /* store the number of vectors reserved for queues */
- adapter->num_q_vectors = numvecs;
-
- /* add 1 vector for link status interrupts */
- numvecs++;
- adapter->msix_entries = kcalloc(numvecs, sizeof(struct msix_entry),
- GFP_KERNEL);
- if (!adapter->msix_entries)
- goto msi_only;
-
- for (i = 0; i < numvecs; i++)
- adapter->msix_entries[i].entry = i;
-
- err = pci_enable_msix(adapter->pdev,
- adapter->msix_entries,
- numvecs);
- if (err == 0)
- goto out;
-
- igb_reset_interrupt_capability(adapter);
-
- /* If we can't do MSI-X, try MSI */
-msi_only:
+ switch (adapter->int_mode) {
+ case IGB_INT_MODE_MSIX:
+ /* start with one vector for every rx queue */
+ numvecs = adapter->num_rx_queues;
+
+ /* if tx handler is seperate add 1 for every tx queue */
+ if (!(adapter->flags & IGB_FLAG_QUEUE_PAIRS))
+ numvecs += adapter->num_tx_queues;
+
+ /* store the number of vectors reserved for queues */
+ adapter->num_q_vectors = numvecs;
+
+ /* add 1 vecotr for link status interrupts */
+ numvecs++;
+ adapter->msix_entries = kcalloc(numvecs,
+ sizeof(struct msix_entry),
+ GFP_KERNEL);
+ if (adapter->msix_entries) {
+ for (i = 0; i < numvecs; i++)
+ adapter->msix_entries[i].entry = i;
+
+ err = pci_enable_msix(pdev,
+ adapter->msix_entries, numvecs);
+ if (err == 0)
+ break;
+ }
+ /* MSI-X failed, so fall through and try MSI */
+ dev_warn(&pdev->dev, "Failed to initialize MSI-X interrupts. "
+ "Falling back to MSI interrupts.\n");
+ igb_reset_interrupt_capability(adapter);
+ case IGB_INT_MODE_MSI:
+ if (!pci_enable_msi(pdev))
+ adapter->flags |= IGB_FLAG_HAS_MSI;
+ else
+ dev_warn(&pdev->dev, "Failed to initialize MSI "
+ "interrupts. Falling back to legacy "
+ "interrupts.\n");
+ /* Fall through */
+ case IGB_INT_MODE_LEGACY:
#ifdef CONFIG_PCI_IOV
- /* disable SR-IOV for non MSI-X configurations */
- if (adapter->vf_data) {
- struct e1000_hw *hw = &adapter->hw;
- /* disable iov and allow time for transactions to clear */
- pci_disable_sriov(adapter->pdev);
- msleep(500);
-
- kfree(adapter->vf_data);
- adapter->vf_data = NULL;
- wr32(E1000_IOVCTL, E1000_IOVCTL_REUSE_VFQ);
- msleep(100);
- dev_info(&adapter->pdev->dev, "IOV Disabled\n");
- }
+ /* disable SR-IOV for non MSI-X configurations */
+ if (adapter->vf_data) {
+ struct e1000_hw *hw = &adapter->hw;
+ /*
+ * disable iov and allow time for transactions
+ * to clear
+ */
+ pci_disable_sriov(adapter->pdev);
+ msleep(500);
+
+ kfree(adapter->vf_data);
+ adapter->vf_data = NULL;
+ wr32(E1000_IOVCTL, E1000_IOVCTL_REUSE_VFQ);
+ msleep(100);
+ dev_info(&adapter->pdev->dev, "IOV Disabled\n");
+ }
#endif
- adapter->vfs_allocated_count = 0;
- adapter->rss_queues = 1;
- adapter->flags |= IGB_FLAG_QUEUE_PAIRS;
- adapter->num_rx_queues = 1;
- adapter->num_tx_queues = 1;
- adapter->num_q_vectors = 1;
- if (!pci_enable_msi(adapter->pdev))
- adapter->flags |= IGB_FLAG_HAS_MSI;
-out:
+ /* disable advanced features and set number of queues to 1 */
+ adapter->vfs_allocated_count = 0;
+ adapter->rss_queues = 1;
+ adapter->flags |= IGB_FLAG_QUEUE_PAIRS;
+ adapter->num_rx_queues = 1;
+ adapter->num_tx_queues = 1;
+ adapter->num_q_vectors = 1;
+ /* Don't do anything; this is system default */
+ break;
+ }
+
/* Notify the stack of the (possibly) reduced Tx Queue count. */
adapter->netdev->real_num_tx_queues = adapter->num_tx_queues;
return;
@@ -1411,6 +1426,7 @@ static int __devinit igb_probe(struct pci_dev *pdev,
const struct e1000_info *ei = igb_info_tbl[ent->driver_data];
unsigned long mmio_start, mmio_len;
int err, pci_using_dac;
+ static int cards_found;
u16 eeprom_apme_mask = IGB_EEPROM_APME;
u32 part_num;
@@ -1477,6 +1493,8 @@ static int __devinit igb_probe(struct pci_dev *pdev,
strncpy(netdev->name, pci_name(pdev), sizeof(netdev->name) - 1);
+ adapter->bd_number = cards_found;
+
netdev->mem_start = mmio_start;
netdev->mem_end = mmio_start + mmio_len;
@@ -1675,6 +1693,7 @@ static int __devinit igb_probe(struct pci_dev *pdev,
(adapter->flags & IGB_FLAG_HAS_MSI) ? "MSI" : "legacy",
adapter->num_rx_queues, adapter->num_tx_queues);
+ cards_found++;
return 0;
err_register:
@@ -1945,6 +1964,8 @@ static int __devinit igb_sw_init(struct igb_adapter *adapter)
adapter->max_frame_size = netdev->mtu + ETH_HLEN + ETH_FCS_LEN;
adapter->min_frame_size = ETH_ZLEN + ETH_FCS_LEN;
+ igb_check_options(adapter);
+
#ifdef CONFIG_PCI_IOV
if (hw->mac.type == e1000_82576)
adapter->vfs_allocated_count = max_vfs;
diff --git a/drivers/net/igb/igb_param.c b/drivers/net/igb/igb_param.c
new file mode 100644
index 0000000..af9d07f
--- /dev/null
+++ b/drivers/net/igb/igb_param.c
@@ -0,0 +1,133 @@
+/*******************************************************************************
+
+ Intel(R) Gigabit Ethernet Linux driver
+ Copyright(c) 2007-2010 Intel Corporation.
+
+ This program is free software; you can redistribute it and/or modify it
+ under the terms and conditions of the GNU General Public License,
+ version 2, as published by the Free Software Foundation.
+
+ This program is distributed in the hope 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.,
+ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+
+ The full GNU General Public License is included in this distribution in
+ the file called "COPYING".
+
+ Contact Information:
+ e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+
+*******************************************************************************/
+
+#include <linux/netdevice.h>
+#include <linux/pci.h>
+
+#include "igb.h"
+
+/*
+ * This is the only thing that needs to be changed to adjust the
+ * maximum number of ports that the driver can manage.
+ */
+
+#define IGB_MAX_NIC 32
+
+#define OPTION_UNSET -1
+
+/*
+ * All parameters are treated the same, as an integer array of values.
+ * This macro just reduces the need to repeat the same declaration code
+ * over and over (plus this helps to avoid typo bugs).
+ */
+
+#define IGB_PARAM_INIT { [0 ... IGB_MAX_NIC] = OPTION_UNSET }
+#define IGB_PARAM(X, desc) \
+ static int __devinitdata X[IGB_MAX_NIC+1] \
+ = IGB_PARAM_INIT; \
+ static unsigned int num_##X; \
+ module_param_array_named(X, X, int, &num_##X, 0); \
+ MODULE_PARM_DESC(X, desc);
+
+/* IntMode (Interrupt Mode)
+ *
+ * Valid Range: 0 - 2
+ *
+ * Default Value: 2 (MSI-X)
+ */
+IGB_PARAM(IntMode, "Interrupt Mode");
+#define MAX_INTMODE IGB_INT_MODE_MSIX
+#define MIN_INTMODE IGB_INT_MODE_LEGACY
+
+struct igb_option {
+ const char *name;
+ const char *err;
+ int def;
+ int min;
+ int max;
+};
+
+static int __devinit igb_validate_option(unsigned int *value,
+ const struct igb_option *opt,
+ struct igb_adapter *adapter)
+{
+ struct pci_dev *pdev = adapter->pdev;
+
+ if (*value == OPTION_UNSET) {
+ *value = opt->def;
+ return 0;
+ }
+
+ if (*value >= opt->min && *value <= opt->max) {
+ dev_info(&pdev->dev, "%s set to %d\n", opt->name, *value);
+ return 0;
+ } else {
+ dev_info(&pdev->dev, "Invalid %s value specified (%d) %s\n",
+ opt->name, *value, opt->err);
+ *value = opt->def;
+ return -1;
+ }
+}
+
+/**
+ * igb_check_options - Range Checking for Command Line Parameters
+ * @adapter: board private structure
+ *
+ * This routine checks all command line parameters for valid user
+ * input. If an invalid value is given, or if no user specified
+ * value exists, a default value is used. The final value is stored
+ * in a variable in the adapter structure.
+ **/
+void __devinit igb_check_options(struct igb_adapter *adapter)
+{
+ struct pci_dev *pdev = adapter->pdev;
+ int bd = adapter->bd_number;
+
+ if (bd >= IGB_MAX_NIC) {
+ dev_notice(&pdev->dev, "Warning: no configuration for "
+ "board #%d\n", bd);
+ dev_notice(&pdev->dev, "Using defaults for all values\n");
+ }
+
+ { /* Interrupt Mode */
+ struct igb_option opt = {
+ .name = "Interrupt Mode",
+ .err = "defaulting to 2 (MSI-X)",
+ .def = IGB_INT_MODE_MSIX,
+ .min = MIN_INTMODE,
+ .max = MAX_INTMODE
+ };
+
+ if (num_IntMode > bd) {
+ unsigned int int_mode = IntMode[bd];
+ igb_validate_option(&int_mode, &opt, adapter);
+ adapter->int_mode = int_mode;
+ } else {
+ adapter->int_mode = opt.def;
+ }
+ }
+}
^ permalink raw reply related
* 2.6.34-rc1 regression: rmmod hangs on modules which interfaces have ipv6 address
From: Tantilov, Emil S @ 2010-03-23 0:02 UTC (permalink / raw)
To: netdev@vger.kernel.org; +Cc: David Miller
Steps to reproduce:
1. Load driver
2. Bring interfaces up with IPv6 address
3. rmmod the driver
rmmod fails with "waiting for device to become free. Usage count=1"
Bisecting turned up this patch:
http://git.kernel.org/?p=linux/kernel/git/davem/net-next-2.6.git;a=commitdiff;h=3e81c6da39a265e11ef48f52bd15bf7ca0068c75
Although I can't confirm this since without it I am not able to assign IPv6 addresses. So it could be caused by whatever this commit is trying to fix.
Thanks,
Emil
^ permalink raw reply
* Re: Unable to create more than 1 guest virtio-net device using vhost-net backend
From: Sridhar Samudrala @ 2010-03-22 23:41 UTC (permalink / raw)
To: Michael S. Tsirkin
Cc: Avi Kivity, netdev, kvm@vger.kernel.org, gleb, David Stevens
In-Reply-To: <20100322181623.GA20188@redhat.com>
On Mon, 2010-03-22 at 20:16 +0200, Michael S. Tsirkin wrote:
> On Sun, Mar 21, 2010 at 01:58:29PM +0200, Avi Kivity wrote:
> > On 03/21/2010 01:34 PM, Michael S. Tsirkin wrote:
> >> On Sun, Mar 21, 2010 at 12:29:31PM +0200, Avi Kivity wrote:
> >>
> >>> On 03/21/2010 12:15 PM, Michael S. Tsirkin wrote:
> >>>
> >>>>>> Nothing easy that I can see. Each device needs 2 of these. Avi, Gleb,
> >>>>>> any objections to increasing the limit to say 16? That would give us
> >>>>>> 5 more devices to the limit of 6 per guest.
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>> Increase it to 200, then.
> >>>>>
> >>>>>
> >>>> OK. I think we'll also need a smarter allocator
> >>>> than bus->dev_count++ than we now have. Right?
> >>>>
> >>>>
> >>> No, why?
> >>>
> >> We'll run into problems if devices are created/removed in random order,
> >> won't we?
> >>
> >
> > unregister_dev() takes care of it.
> >
> >>> Eventually we'll want faster scanning than the linear search we employ
> >>> now, though.
> >>>
> >> Yes I suspect with 200 entries we will :). Let's just make it 16 for
> >> now?
> >>
> >
> > Let's make it 200 and fix the performance problems later. Making it 16
> > is just asking for trouble.
>
> I did this and performance with vhost seems to become much more noisy,
> and drop by about 10% on average, even though in practice only
> a single device is created. Still trying to figure it out ...
> Any idea?
I am not sure if this 10% variation is due to the increase of NR_IO_BUS_DEVS.
In our testing, we do see variations of 10% or higher between multiple netperf
instances with the same setup/configuration when using virtio/vhost.
Thanks
Sridhar
^ permalink raw reply
* Re: [RFC Patch 3/3] bonding: make bonding support netpoll
From: Jay Vosburgh @ 2010-03-22 23:36 UTC (permalink / raw)
To: Matt Mackall
Cc: Amerigo Wang, linux-kernel, netdev, bridge, Andy Gospodarek,
Neil Horman, Stephen Hemminger, bonding-devel, David Miller
In-Reply-To: <1269297485.3552.26.camel@calx>
Matt Mackall <mpm@selenic.com> wrote:
>On Mon, 2010-03-22 at 04:17 -0400, Amerigo Wang wrote:
>> Based on Andy's work, but I modify a lot.
>>
>> Similar to the patch for bridge, this patch does:
>>
>> 1) implement the 4 methods to support netpoll for bonding;
>>
>> 2) modify netpoll during forwarding packets in bonding;
>>
>> 3) disable netpoll support of bridge when a netpoll-unabled device
>> is added to bonding;
>>
>> 4) enable netpoll support when all underlying devices support netpoll.
>
>Again, not sure if this is the right policy. Seems to me that on a
>bonding device we should simply pick an interface to send netpoll
>messages on, without reference to balancing, etc. Thus, if any of the
>bonded devices supports polling, it should work.
For some of the modes, the above is pretty straighforward.
Others, 802.3ad and balance-alb, are a bit more complicated.
The risk is that the network peers and switches may see the same
MAC address on multiple ports, or different MAC addresses for the same
IP address.
To implement the above suggestion, I think a "current netpoll
slave" would have to be tracked, and kept up to date (as slaves become
active or inactive, etc). Reusing the existing "current active slave"
won't work for the case that the active slave is not netpoll-capable,
but a different slave is; also, not all modes use the current active
slave.
In 802.3ad, the "current netpoll slave" selector will have to
poke into the aggregator status to choose the netpoll slave. It's not a
simple matter of picking one and then sticking with it forever; if the
aggregator containing the netpoll slave is deactivated, then peers may
not receive the traffic, etc.
In the active-backup mode, only the active slave can send or
receive, so if it's not netpoll capable, but a backup slave is, you're
still out of luck (unless netpoll awareness is added to the "best slave"
selection logic, and even then it'd have to be a secondary criteria).
Or, the inactive slave can be transmitted on, but if the same MAC comes
out of the active and a backup slave, it can confuse switches.
In one mode (balance-alb), slaves keep their own MAC addresses,
and are matched with peers. Bypassing the balance algorithm could again
confuse peers or switches, who could see two MAC addresses for the same
IP address, if netpoll traffic goes out a different slave than the
balance algorithm picks for the same destination.
I think, then, the question becomes: is this extra complexity
worth it to cover the cases of netpoll over bonding wherein one or more
slaves don't support netpoll?
How many network drivers don't support netpoll nowadays?
-J
---
-Jay Vosburgh, IBM Linux Technology Center, fubar@us.ibm.com
^ permalink raw reply
* Re: [Uclinux-dist-devel] [PATCH] can: bfin_can: switch to common Blackfin can header
From: David Miller @ 2010-03-22 23:36 UTC (permalink / raw)
To: vapier.adi
Cc: socketcan-core, netdev, uclinux-dist-devel, oliver.hartkopp,
urs.thuermann
In-Reply-To: <8bd0f97a1003221409w24437f85ke27f4338e82ba44e@mail.gmail.com>
From: Mike Frysinger <vapier.adi@gmail.com>
Date: Mon, 22 Mar 2010 17:09:55 -0400
> On Mon, Mar 22, 2010 at 13:17, David Miller wrote:
>> From: Mike Frysinger <vapier.adi@gmail.com>
>> Do you still feel this way after people other than me also asked you
>> to do the same exact thing for me? Do you think they are making
>> arbitrary requests as well?
>
> the only other person to say anything did not say "you now need to
> resend your patch". it was "please wait until Linus has pulled things
> and then Dave has pulled things". which was not relevant at the point
> it was sent.
Wolfgang told you to "just resubmit the patch", just like I asked you
to, when you said "the code is there just apply the old one."
^ permalink raw reply
* Re: [Bug 15582] New: BUG: unable to handle kernel NULL pointer dereference at 0000000000000028
From: Andrew Morton @ 2010-03-22 23:14 UTC (permalink / raw)
To: Patrick McHardy, Jeff Kirsher, Jesse Brandeburg, Bruce Allan,
Alex Duyck <alexa
Cc: bugzilla-daemon, netdev, stivi
In-Reply-To: <bug-15582-27@http.bugzilla.kernel.org/>
(switched to email. Please respond via emailed reply-to-all, not via the
bugzilla web interface).
On Fri, 19 Mar 2010 12:01:10 GMT
bugzilla-daemon@bugzilla.kernel.org wrote:
> http://bugzilla.kernel.org/show_bug.cgi?id=15582
>
> Summary: BUG: unable to handle kernel NULL pointer dereference
> at 0000000000000028
A bug in igb or the vlan code, I guess.
> Product: Memory Management
> Version: 2.5
> Kernel Version: 2.6.33.1
> Platform: All
> OS/Version: Linux
> Tree: Mainline
> Status: NEW
> Severity: normal
> Priority: P1
> Component: Other
> AssignedTo: akpm@linux-foundation.org
> ReportedBy: stivi@kity.pl
> Regression: No
>
>
> Distribution:Debian Lenny
>
> Hardware Environment:
> INTEL Server Board S5520HC
> Intel(R) Xeon(R) CPU X5560 @ 2.80GHz
> 2x RAID bus controller 3ware Inc 9690SA-8I
>
> Software Environment: squid (multi instances)
>
> Problem Description:
>
> The following error occurs whenever i halt the machine:
> [ 439.629361] BUG: unable to handle kernel NULL pointer dereference at
> 0000000000000028
> [ 439.723410] IP: [<ffffffff812462aa>] vlan_gro_common+0xd7/0x190
> [ 439.794374] PGD 0
> [ 439.818655] Oops: 0000 [#1] SMP
> [ 439.857638] last sysfs file:
> /sys/devices/pci0000:00/0000:00:09.0/0000:05:00.0/host1/target1:0:0/1:0:0:0/block/sdc/queue/nr_requests
> [ 440.000143] CPU 7
> [ 440.024385] Pid: 0, comm: swapper Not tainted 2.6.33.1-univ #1
> S5520HC/S5520HC
> [ 440.110817] RIP: 0010:[<ffffffff812462aa>] [<ffffffff812462aa>]
> vlan_gro_common+0xd7/0x190
> [ 440.210919] RSP: 0018:ffff8800283c3d10 EFLAGS: 00010203
> [ 440.274472] RAX: 0000000000000001 RBX: ffff88066d5fd818 RCX:
> ffff8806683aa680
> [ 440.359798] RDX: 00000000000003e7 RSI: 0000000000000000 RDI:
> ffff88066d2b2000
> [ 440.445124] RBP: 00000000000003e7 R08: ffff88066c5ea000 R09:
> ffff8806683aa680
> [ 440.530488] R10: 00000000000003e7 R11: 000000000000040b R12:
> 0000000000000000
> [ 440.615816] R13: ffff88066d5fd818 R14: 000000000000003e R15:
> 000000000000040c
> [ 440.701169] FS: 0000000000000000(0000) GS:ffff8800283c0000(0000)
> knlGS:0000000000000000
> [ 440.797986] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> [ 440.866705] CR2: 0000000000000028 CR3: 0000000001361000 CR4:
> 00000000000006e0
> [ 440.952036] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
> 0000000000000000
> [ 441.037364] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
> 0000000000000400
> [ 441.122690] Process swapper (pid: 0, threadinfo ffff88066fa82000, task
> ffff88066fa7acd0)
> [ 441.219462] Stack:
> [ 441.243535] ffff8806683aa680 ffffffff812466c7 0000000000000246
> ffff8806683aa680
> [ 441.330462] <0> 000000000000003e ffff88066d5fd680 0000000000000008
> ffffffffa00b1871
> [ 441.422868] <0> 0000000000000000 0000000000000000 0000000000000000
> 0000000000000000
> [ 441.517692] Call Trace:
> [ 441.546959] <IRQ>
> [ 441.572233] [<ffffffff812466c7>] ? vlan_gro_receive+0x6e/0x83
> [ 441.641993] [<ffffffffa00b1871>] ? igb_poll+0x741/0xe80 [igb]
> [ 441.711790] [<ffffffff811e07ee>] ? net_rx_action+0xa8/0x1a1
> [ 441.779478] [<ffffffff81039ffe>] ? __do_softirq+0xd7/0x195
> [ 441.846122] [<ffffffff8100370c>] ? call_softirq+0x1c/0x28
> [ 441.911727] [<ffffffff81005325>] ? do_softirq+0x31/0x63
> [ 441.975300] [<ffffffff81039df4>] ? irq_exit+0x36/0x78
> [ 442.036746] [<ffffffff81004a26>] ? do_IRQ+0xa7/0xbd
> [ 442.096119] [<ffffffff81258ed3>] ? ret_from_intr+0x0/0xa
> [ 442.160685] <EOI>
> [ 442.185957] [<ffffffff811cad43>] ? poll_idle+0x1b/0x55
> [ 442.248444] [<ffffffff811cad32>] ? poll_idle+0xa/0x55
> [ 442.309890] [<ffffffff811cb06a>] ? cpuidle_idle_call+0x8e/0xe8
> [ 442.380683] [<ffffffff81001cd4>] ? cpu_idle+0x53/0x8b
> [ 442.442133] Code: 41 81 79 7e 88 09 0f 85 d1 00 00 00 44 89 d2 44 89 d0 66
> 81 e2 ff 0f 80 cc 10 66 41 89 81 b8 00 00 00 89 d0 66 c1 e8 09 0f b7 c0 <48> 8b
> 4c c6 20 31 c0 48 85 c9 74 0c 48 89 d0 25 ff$
> [ 442.678920] RIP [<ffffffff812462aa>] vlan_gro_common+0xd7/0x190
> [ 442.750922] RSP <ffff8800283c3d10>
> [ 442.792645] CR2: 0000000000000028
> [ 442.832293] ---[ end trace 9245d00ed2188cae ]---
> [ 442.887516] Kernel panic - not syncing: Fatal exception in interrupt
> [ 442.963506] Pid: 0, comm: swapper Tainted: G D 2.6.33.1-univ #1
> [ 443.042610] Call Trace:
> [ 443.071875] <IRQ> [<ffffffff812568b7>] ? panic+0x86/0x145
> [ 443.138593] [<ffffffff81039e06>] ? irq_exit+0x48/0x78
> [ 443.200050] [<ffffffff81258ed3>] ? ret_from_intr+0x0/0xa
> [ 443.264661] [<ffffffff8103554c>] ? kmsg_dump+0x99/0x124
> [ 443.328202] [<ffffffff81006416>] ? oops_end+0x9f/0xac
> [ 443.389656] [<ffffffff8101e9af>] ? no_context+0x1f2/0x201
> [ 443.455263] [<ffffffffa014cd20>] ? bond_dev_queue_xmit+0x14c/0x169
> [bonding]
> [ 443.540611] [<ffffffff8101eb65>] ? __bad_area_nosemaphore+0x1a7/0x1cb
> [ 443.618679] [<ffffffff811e0d8f>] ? dev_hard_start_xmit+0x221/0x2dd
> [ 443.693630] [<ffffffff811e134d>] ? dev_queue_xmit+0x401/0x433
> [ 443.763397] [<ffffffff812590df>] ? page_fault+0x1f/0x30
> [ 443.826927] [<ffffffff812462aa>] ? vlan_gro_common+0xd7/0x190
> [ 443.896686] [<ffffffff812466c7>] ? vlan_gro_receive+0x6e/0x83
> [ 443.966485] [<ffffffffa00b1871>] ? igb_poll+0x741/0xe80 [igb]
> [ 444.036246] [<ffffffff811e07ee>] ? net_rx_action+0xa8/0x1a1
> [ 444.103927] [<ffffffff81039ffe>] ? __do_softirq+0xd7/0x195
> [ 444.170569] [<ffffffff8100370c>] ? call_softirq+0x1c/0x28
> [ 444.236176] [<ffffffff81005325>] ? do_softirq+0x31/0x63
> [ 444.299704] [<ffffffff81039df4>] ? irq_exit+0x36/0x78
> [ 444.361157] [<ffffffff81004a26>] ? do_IRQ+0xa7/0xbd
> [ 444.420533] [<ffffffff81258ed3>] ? ret_from_intr+0x0/0xa
> [ 444.485101] <EOI> [<ffffffff811cad43>] ? poll_idle+0x1b/0x55
> [ 444.555036] [<ffffffff811cad32>] ? poll_idle+0xa/0x55
> [ 444.616486] [<ffffffff811cb06a>] ? cpuidle_idle_call+0x8e/0xe8
> [ 444.687283] [<ffffffff81001cd4>] ? cpu_idle+0x53/0x8b
>
> Steps to reproduce: Whenever i halt the machine.
>
>
> Igb driver is replaced by:
> http://downloadcenter.intel.com/detail_desc.aspx?agr=Y&DwnldID=13663
>
> This is the same machine from bugs:
> https://bugzilla.kernel.org/show_bug.cgi?id=15148,
> https://bugzilla.kernel.org/show_bug.cgi?id=15581.
>
^ permalink raw reply
* [PATCH] kaweth: New USB ID 07c9:b010 Allied Telesyn AT-USB10
From: Andreas Hartmann @ 2010-03-22 23:07 UTC (permalink / raw)
To: Greg Kroah-Hartman; +Cc: linux-usb, netdev, linux-kernel, Andrew Morton
New USB ID 07c9:b010 Allied Telesyn AT-USB10
Signed-off-by: Andreas Hartmann <andi@ahash.de>
---
This USB Ethernet adapter has the KL5KUSB101B chip and is so far unknown to the
kaweth driver. The device was found working with the patch applied.
--- linux-2.6.34-rc2/drivers/net/usb/kaweth.c.orig 2010-03-20 02:17:57.000000000 +0100
+++ linux-2.6.34-rc2/drivers/net/usb/kaweth.c 2010-03-22 23:30:31.000000000 +0100
@@ -145,6 +145,7 @@ static struct usb_device_id usb_klsi_tab
{ USB_DEVICE(0x0707, 0x0100) }, /* SMC 2202USB */
{ USB_DEVICE(0x07aa, 0x0001) }, /* Correga K.K. */
{ USB_DEVICE(0x07b8, 0x4000) }, /* D-Link DU-E10 */
+ { USB_DEVICE(0x07c9, 0xb010) }, /* Allied Telesyn AT-USB10 */
{ USB_DEVICE(0x0846, 0x1001) }, /* NetGear EA-101 */
{ USB_DEVICE(0x0846, 0x1002) }, /* NetGear EA-101 */
{ USB_DEVICE(0x085a, 0x0008) }, /* PortGear Ethernet Adapter */
^ permalink raw reply
* Re: [RFC Patch 3/3] bonding: make bonding support netpoll
From: Matt Mackall @ 2010-03-22 22:38 UTC (permalink / raw)
To: Amerigo Wang
Cc: linux-kernel, netdev, bridge, Andy Gospodarek, Neil Horman,
Stephen Hemminger, bonding-devel, Jay Vosburgh, David Miller
In-Reply-To: <20100322082123.4967.77433.sendpatchset@localhost.localdomain>
On Mon, 2010-03-22 at 04:17 -0400, Amerigo Wang wrote:
> Based on Andy's work, but I modify a lot.
>
> Similar to the patch for bridge, this patch does:
>
> 1) implement the 4 methods to support netpoll for bonding;
>
> 2) modify netpoll during forwarding packets in bonding;
>
> 3) disable netpoll support of bridge when a netpoll-unabled device
> is added to bonding;
>
> 4) enable netpoll support when all underlying devices support netpoll.
Again, not sure if this is the right policy. Seems to me that on a
bonding device we should simply pick an interface to send netpoll
messages on, without reference to balancing, etc. Thus, if any of the
bonded devices supports polling, it should work.
Hopefully someone more familiar with the goals and philosophy of the
bonding code can comment further.
--
http://selenic.com : development and support for Mercurial and Linux
^ permalink raw reply
* Re: [RFC Patch 2/3] bridge: make bridge support netpoll
From: Matt Mackall @ 2010-03-22 22:35 UTC (permalink / raw)
To: Amerigo Wang
Cc: linux-kernel, netdev, bridge, Andy Gospodarek, Neil Horman,
Stephen Hemminger, bonding-devel, Jay Vosburgh, David Miller
In-Reply-To: <20100322082112.4967.5504.sendpatchset@localhost.localdomain>
On Mon, 2010-03-22 at 04:17 -0400, Amerigo Wang wrote:
> Based on the previous patch, make bridge support netpoll by:
>
> 1) implement the 4 methods to support netpoll for bridge;
>
> 2) modify netpoll during forwarding packets in bridge;
>
> 3) disable netpoll support of bridge when a netpoll-unabled device
> is added to bridge;
Not sure if this is the right thing to do. Shouldn't we simply enable
polling on all devices that support it and warn about the others (aka
best effort)?
--
http://selenic.com : development and support for Mercurial and Linux
^ permalink raw reply
* Re: [RFC Patch 1/3] netpoll: add generic support for bridge and bonding devices
From: Matt Mackall @ 2010-03-22 22:31 UTC (permalink / raw)
To: Amerigo Wang
Cc: linux-kernel, netdev, bridge, Andy Gospodarek, Neil Horman,
Stephen Hemminger, bonding-devel, Jay Vosburgh, David Miller,
Jeff Moyer
In-Reply-To: <20100322082059.4967.63492.sendpatchset@localhost.localdomain>
On Mon, 2010-03-22 at 04:17 -0400, Amerigo Wang wrote:
> This whole patchset is for adding netpoll support to bridge and bonding
> devices. I already tested it for bridge, bonding, bridge over bonding,
> and bonding over bridge. It looks fine now.
Ages ago, Jeff Moyer took a run at this, added him to the cc: on the off
chance he still cares.
> Please comment.
>
>
> To make bridge and bonding support netpoll, we need to adjust
> some netpoll generic code. This patch does the following things:
>
> 1) introduce two new priv_flags for struct net_device:
> IFF_IN_NETPOLL which identifies we are processing a netpoll;
> IFF_DISABLE_NETPOLL is used to disable netpoll support for a device
> at run-time;
This one is a little worrisome. I've tried to keep the netpoll code
restricted to as tight an area as possible. Adding new flags like these
that random drivers might try to fiddle with seems like a good way for a
driver writer to get in trouble. Also flag space is filling up.
> 2) introduce three new methods for netdev_ops:
> ->ndo_netpoll_setup() is used to setup netpoll for a device;
> ->ndo_netpoll_xmit() is used to transmit netpoll requests;
> ->ndo_netpoll_cleanup() is used to clean up netpoll when a device is
> removed.
Seems like a lot of interface for something to be used by only a couple
core drivers. Hopefully Dave has an opinion here.
> 3) introduce netpoll_poll_dev() which takes a struct net_device * parameter;
>
> 4) export netpoll_send_skb() and netpoll_poll_dev() which will be used later;
>
> 5) hide a pointer to struct netpoll in struct netpoll_info, ditto.
>
> Cc: David Miller <davem@davemloft.net>
> Cc: Neil Horman <nhorman@tuxdriver.com>
> Signed-off-by: WANG Cong <amwang@redhat.com>
>
> ---
> Index: linux-2.6/include/linux/if.h
> ===================================================================
> --- linux-2.6.orig/include/linux/if.h
> +++ linux-2.6/include/linux/if.h
> @@ -71,6 +71,8 @@
> * release skb->dst
> */
> #define IFF_DONT_BRIDGE 0x800 /* disallow bridging this ether dev */
> +#define IFF_IN_NETPOLL 0x1000 /* whether we are processing netpoll */
> +#define IFF_DISABLE_NETPOLL 0x2000 /* disable netpoll at run-time */
>
> #define IF_GET_IFACE 0x0001 /* for querying only */
> #define IF_GET_PROTO 0x0002
> Index: linux-2.6/include/linux/netdevice.h
> ===================================================================
> --- linux-2.6.orig/include/linux/netdevice.h
> +++ linux-2.6/include/linux/netdevice.h
> @@ -530,6 +530,8 @@ struct netdev_queue {
> unsigned long tx_dropped;
> } ____cacheline_aligned_in_smp;
>
> +struct netpoll;
> +struct netpoll_info;
>
> /*
> * This structure defines the management hooks for network devices.
> @@ -667,6 +669,12 @@ struct net_device_ops {
> unsigned short vid);
> #ifdef CONFIG_NET_POLL_CONTROLLER
> void (*ndo_poll_controller)(struct net_device *dev);
> + void (*ndo_netpoll_setup)(struct net_device *dev,
> + struct netpoll_info *npinfo);
> + int (*ndo_netpoll_xmit)(struct netpoll *np,
> + struct sk_buff *skb,
> + struct net_device *dev);
> + void (*ndo_netpoll_cleanup)(struct net_device *dev);
> #endif
> int (*ndo_set_vf_mac)(struct net_device *dev,
> int queue, u8 *mac);
> Index: linux-2.6/include/linux/netpoll.h
> ===================================================================
> --- linux-2.6.orig/include/linux/netpoll.h
> +++ linux-2.6/include/linux/netpoll.h
> @@ -36,8 +36,11 @@ struct netpoll_info {
> struct sk_buff_head txq;
>
> struct delayed_work tx_work;
> +
> + struct netpoll *netpoll;
> };
>
> +void netpoll_poll_dev(struct net_device *dev);
> void netpoll_poll(struct netpoll *np);
> void netpoll_send_udp(struct netpoll *np, const char *msg, int len);
> void netpoll_print_options(struct netpoll *np);
> @@ -47,6 +50,7 @@ int netpoll_trap(void);
> void netpoll_set_trap(int trap);
> void netpoll_cleanup(struct netpoll *np);
> int __netpoll_rx(struct sk_buff *skb);
> +void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb);
>
>
> #ifdef CONFIG_NETPOLL
> Index: linux-2.6/net/core/netpoll.c
> ===================================================================
> --- linux-2.6.orig/net/core/netpoll.c
> +++ linux-2.6/net/core/netpoll.c
> @@ -178,9 +178,8 @@ static void service_arp_queue(struct net
> }
> }
>
> -void netpoll_poll(struct netpoll *np)
> +void netpoll_poll_dev(struct net_device *dev)
> {
> - struct net_device *dev = np->dev;
> const struct net_device_ops *ops;
>
> if (!dev || !netif_running(dev))
> @@ -200,6 +199,13 @@ void netpoll_poll(struct netpoll *np)
> zap_completion_queue();
> }
>
> +void netpoll_poll(struct netpoll *np)
> +{
> + if (!np->dev)
> + return;
> + netpoll_poll_dev(np->dev);
> +}
> +
> static void refill_skbs(void)
> {
> struct sk_buff *skb;
> @@ -281,7 +287,7 @@ static int netpoll_owner_active(struct n
> return 0;
> }
>
> -static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb)
> +void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb)
> {
> int status = NETDEV_TX_BUSY;
> unsigned long tries;
> @@ -307,7 +313,10 @@ static void netpoll_send_skb(struct netp
> tries > 0; --tries) {
> if (__netif_tx_trylock(txq)) {
> if (!netif_tx_queue_stopped(txq)) {
> - status = ops->ndo_start_xmit(skb, dev);
> + if (ops->ndo_netpoll_xmit)
> + status = ops->ndo_netpoll_xmit(np, skb, dev);
> + else
> + status = ops->ndo_start_xmit(skb, dev);
> if (status == NETDEV_TX_OK)
> txq_trans_update(txq);
> }
> @@ -752,7 +761,10 @@ int netpoll_setup(struct netpoll *np)
> atomic_inc(&npinfo->refcnt);
> }
>
> - if (!ndev->netdev_ops->ndo_poll_controller) {
> + npinfo->netpoll = np;
> +
> + if (ndev->priv_flags & IFF_DISABLE_NETPOLL
> + || !ndev->netdev_ops->ndo_poll_controller) {
> printk(KERN_ERR "%s: %s doesn't support polling, aborting.\n",
> np->name, np->dev_name);
> err = -ENOTSUPP;
> @@ -830,6 +842,9 @@ int netpoll_setup(struct netpoll *np)
> /* last thing to do is link it to the net device structure */
> ndev->npinfo = npinfo;
>
> + if (ndev->netdev_ops->ndo_netpoll_setup)
> + ndev->netdev_ops->ndo_netpoll_setup(ndev, npinfo);
> +
> /* avoid racing with NAPI reading npinfo */
> synchronize_rcu();
>
> @@ -904,6 +919,7 @@ void netpoll_set_trap(int trap)
> atomic_dec(&trapped);
> }
>
> +EXPORT_SYMBOL(netpoll_send_skb);
> EXPORT_SYMBOL(netpoll_set_trap);
> EXPORT_SYMBOL(netpoll_trap);
> EXPORT_SYMBOL(netpoll_print_options);
> @@ -911,4 +927,5 @@ EXPORT_SYMBOL(netpoll_parse_options);
> EXPORT_SYMBOL(netpoll_setup);
> EXPORT_SYMBOL(netpoll_cleanup);
> EXPORT_SYMBOL(netpoll_send_udp);
> +EXPORT_SYMBOL(netpoll_poll_dev);
> EXPORT_SYMBOL(netpoll_poll);
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
--
http://selenic.com : development and support for Mercurial and Linux
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox