* Re: [PATCH] netdev: bfin_mac: drop experimental markings on RMII support
From: David Miller @ 2010-02-17 4:20 UTC (permalink / raw)
To: vapier; +Cc: netdev, uclinux-dist-devel
In-Reply-To: <1266280370-18670-1-git-send-email-vapier@gentoo.org>
From: Mike Frysinger <vapier@gentoo.org>
Date: Mon, 15 Feb 2010 19:32:50 -0500
> The code has been around for a long time now and is known to work on a
> bunch of different parts/boards.
>
> Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Applied to net-next-2.6
^ permalink raw reply
* TEST, Ignore!
From: Tadepalli, Hari K @ 2010-02-17 3:49 UTC (permalink / raw)
To: netdev@vger.kernel.org
Newbie testing, ...
- Hari Tadepalli
____________________________________
Embedded Communications Group
Intel Corporation, Chandler, AZ
^ permalink raw reply
* Recent change to net-next broke KVM bridging
From: Stephen Hemminger @ 2010-02-17 1:36 UTC (permalink / raw)
To: David Miller, Arnd Bergmann; +Cc: netdev
Something in net-next tree broke bridging of virtual nets.
My local VM's can no longer access external networks.
--
^ permalink raw reply
* [PATCH 2/2] Staging: Octeon: Remove /proc/octeon_ethernet_stats
From: David Daney @ 2010-02-17 1:25 UTC (permalink / raw)
To: ralf, linux-mips, netdev, gregkh; +Cc: David Daney
In-Reply-To: <4B7B4540.1010700@caviumnetworks.com>
This file shouldn't be in /proc, so we remove it.
Signed-off-by: David Daney <ddaney@caviumnetworks.com>
---
drivers/staging/octeon/Makefile | 1 -
drivers/staging/octeon/ethernet-proc.c | 144 --------------------------------
drivers/staging/octeon/ethernet-proc.h | 29 -------
drivers/staging/octeon/ethernet.c | 4 -
4 files changed, 0 insertions(+), 178 deletions(-)
delete mode 100644 drivers/staging/octeon/ethernet-proc.c
delete mode 100644 drivers/staging/octeon/ethernet-proc.h
diff --git a/drivers/staging/octeon/Makefile b/drivers/staging/octeon/Makefile
index c0a583c..87447c1 100644
--- a/drivers/staging/octeon/Makefile
+++ b/drivers/staging/octeon/Makefile
@@ -14,7 +14,6 @@ obj-${CONFIG_OCTEON_ETHERNET} := octeon-ethernet.o
octeon-ethernet-objs := ethernet.o
octeon-ethernet-objs += ethernet-mdio.o
octeon-ethernet-objs += ethernet-mem.o
-octeon-ethernet-objs += ethernet-proc.o
octeon-ethernet-objs += ethernet-rgmii.o
octeon-ethernet-objs += ethernet-rx.o
octeon-ethernet-objs += ethernet-sgmii.o
diff --git a/drivers/staging/octeon/ethernet-proc.c b/drivers/staging/octeon/ethernet-proc.c
deleted file mode 100644
index 16308d4..0000000
--- a/drivers/staging/octeon/ethernet-proc.c
+++ /dev/null
@@ -1,144 +0,0 @@
-/**********************************************************************
- * Author: Cavium Networks
- *
- * Contact: support@caviumnetworks.com
- * This file is part of the OCTEON SDK
- *
- * Copyright (c) 2003-2007 Cavium Networks
- *
- * This file is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License, Version 2, as
- * published by the Free Software Foundation.
- *
- * This file is distributed in the hope that it will be useful, but
- * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
- * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
- * NONINFRINGEMENT. See the GNU General Public License for more
- * details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this file; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
- * or visit http://www.gnu.org/licenses/.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium Networks for more information
-**********************************************************************/
-#include <linux/kernel.h>
-#include <linux/seq_file.h>
-#include <linux/proc_fs.h>
-#include <net/dst.h>
-
-#include <asm/octeon/octeon.h>
-
-#include "octeon-ethernet.h"
-#include "ethernet-defines.h"
-
-#include "cvmx-helper.h"
-#include "cvmx-pip.h"
-
-/**
- * User is reading /proc/octeon_ethernet_stats
- *
- * @m:
- * @v:
- * Returns
- */
-static int cvm_oct_stats_show(struct seq_file *m, void *v)
-{
- struct octeon_ethernet *priv;
- int port;
-
- for (port = 0; port < TOTAL_NUMBER_OF_PORTS; port++) {
-
- if (cvm_oct_device[port]) {
- priv = netdev_priv(cvm_oct_device[port]);
-
- seq_printf(m, "\nOcteon Port %d (%s)\n", port,
- cvm_oct_device[port]->name);
- seq_printf(m,
- "rx_packets: %12lu\t"
- "tx_packets: %12lu\n",
- priv->stats.rx_packets,
- priv->stats.tx_packets);
- seq_printf(m,
- "rx_bytes: %12lu\t"
- "tx_bytes: %12lu\n",
- priv->stats.rx_bytes, priv->stats.tx_bytes);
- seq_printf(m,
- "rx_errors: %12lu\t"
- "tx_errors: %12lu\n",
- priv->stats.rx_errors,
- priv->stats.tx_errors);
- seq_printf(m,
- "rx_dropped: %12lu\t"
- "tx_dropped: %12lu\n",
- priv->stats.rx_dropped,
- priv->stats.tx_dropped);
- seq_printf(m,
- "rx_length_errors: %12lu\t"
- "tx_aborted_errors: %12lu\n",
- priv->stats.rx_length_errors,
- priv->stats.tx_aborted_errors);
- seq_printf(m,
- "rx_over_errors: %12lu\t"
- "tx_carrier_errors: %12lu\n",
- priv->stats.rx_over_errors,
- priv->stats.tx_carrier_errors);
- seq_printf(m,
- "rx_crc_errors: %12lu\t"
- "tx_fifo_errors: %12lu\n",
- priv->stats.rx_crc_errors,
- priv->stats.tx_fifo_errors);
- seq_printf(m,
- "rx_frame_errors: %12lu\t"
- "tx_heartbeat_errors: %12lu\n",
- priv->stats.rx_frame_errors,
- priv->stats.tx_heartbeat_errors);
- seq_printf(m,
- "rx_fifo_errors: %12lu\t"
- "tx_window_errors: %12lu\n",
- priv->stats.rx_fifo_errors,
- priv->stats.tx_window_errors);
- seq_printf(m,
- "rx_missed_errors: %12lu\t"
- "multicast: %12lu\n",
- priv->stats.rx_missed_errors,
- priv->stats.multicast);
- }
- }
-
- return 0;
-}
-
-/**
- * /proc/octeon_ethernet_stats was openned. Use the single_open iterator
- *
- * @inode:
- * @file:
- * Returns
- */
-static int cvm_oct_stats_open(struct inode *inode, struct file *file)
-{
- return single_open(file, cvm_oct_stats_show, NULL);
-}
-
-static const struct file_operations cvm_oct_stats_operations = {
- .open = cvm_oct_stats_open,
- .read = seq_read,
- .llseek = seq_lseek,
- .release = single_release,
-};
-
-void cvm_oct_proc_initialize(void)
-{
- struct proc_dir_entry *entry =
- create_proc_entry("octeon_ethernet_stats", 0, NULL);
- if (entry)
- entry->proc_fops = &cvm_oct_stats_operations;
-}
-
-void cvm_oct_proc_shutdown(void)
-{
- remove_proc_entry("octeon_ethernet_stats", NULL);
-}
diff --git a/drivers/staging/octeon/ethernet-proc.h b/drivers/staging/octeon/ethernet-proc.h
deleted file mode 100644
index 82c7d9f..0000000
--- a/drivers/staging/octeon/ethernet-proc.h
+++ /dev/null
@@ -1,29 +0,0 @@
-/*********************************************************************
- * Author: Cavium Networks
- *
- * Contact: support@caviumnetworks.com
- * This file is part of the OCTEON SDK
- *
- * Copyright (c) 2003-2007 Cavium Networks
- *
- * This file is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License, Version 2, as
- * published by the Free Software Foundation.
- *
- * This file is distributed in the hope that it will be useful, but
- * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
- * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
- * NONINFRINGEMENT. See the GNU General Public License for more
- * details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this file; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
- * or visit http://www.gnu.org/licenses/.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium Networks for more information
-*********************************************************************/
-
-void cvm_oct_proc_initialize(void);
-void cvm_oct_proc_shutdown(void);
diff --git a/drivers/staging/octeon/ethernet.c b/drivers/staging/octeon/ethernet.c
index 45cb4c7..02b6367 100644
--- a/drivers/staging/octeon/ethernet.c
+++ b/drivers/staging/octeon/ethernet.c
@@ -42,8 +42,6 @@
#include "ethernet-tx.h"
#include "ethernet-mdio.h"
#include "ethernet-util.h"
-#include "ethernet-proc.h"
-
#include "cvmx-pip.h"
#include "cvmx-pko.h"
@@ -621,7 +619,6 @@ static int __init cvm_oct_init_module(void)
return -ENOMEM;
}
- cvm_oct_proc_initialize();
cvm_oct_configure_common_hw();
cvmx_helper_initialize_packet_io_global();
@@ -828,7 +825,6 @@ static void __exit cvm_oct_cleanup_module(void)
destroy_workqueue(cvm_oct_poll_queue);
cvmx_pko_shutdown();
- cvm_oct_proc_shutdown();
cvmx_ipd_free_ptr();
--
1.6.6
^ permalink raw reply related
* [PATCH 1/2] Staging: Octeon: Reformat a bunch of comments.
From: David Daney @ 2010-02-17 1:25 UTC (permalink / raw)
To: ralf, linux-mips, netdev, gregkh; +Cc: David Daney
In-Reply-To: <4B7B4540.1010700@caviumnetworks.com>
Many of the comments didn't follow kerneldoc guidlines.
Signed-off-by: David Daney <ddaney@caviumnetworks.com>
---
drivers/staging/octeon/ethernet-mdio.c | 6 ++--
drivers/staging/octeon/ethernet-mem.c | 16 +++++-----
drivers/staging/octeon/ethernet-rx.c | 17 ++++++-----
drivers/staging/octeon/ethernet-tx.c | 14 +++++-----
drivers/staging/octeon/ethernet-util.h | 13 +++------
drivers/staging/octeon/ethernet.c | 42 +++++++++--------------------
drivers/staging/octeon/octeon-ethernet.h | 7 -----
7 files changed, 44 insertions(+), 71 deletions(-)
diff --git a/drivers/staging/octeon/ethernet-mdio.c b/drivers/staging/octeon/ethernet-mdio.c
index 05a5cc0..7e0be8d 100644
--- a/drivers/staging/octeon/ethernet-mdio.c
+++ b/drivers/staging/octeon/ethernet-mdio.c
@@ -96,11 +96,11 @@ const struct ethtool_ops cvm_oct_ethtool_ops = {
};
/**
- * IOCTL support for PHY control
- *
+ * cvm_oct_ioctl - IOCTL support for PHY control
* @dev: Device to change
* @rq: the request
* @cmd: the command
+ *
* Returns Zero on success
*/
int cvm_oct_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
@@ -153,7 +153,7 @@ static void cvm_oct_adjust_link(struct net_device *dev)
/**
- * Setup the PHY
+ * cvm_oct_phy_setup_device - setup the PHY
*
* @dev: Device to setup
*
diff --git a/drivers/staging/octeon/ethernet-mem.c b/drivers/staging/octeon/ethernet-mem.c
index 53ed2f7..00cc91d 100644
--- a/drivers/staging/octeon/ethernet-mem.c
+++ b/drivers/staging/octeon/ethernet-mem.c
@@ -34,11 +34,12 @@
#include "cvmx-fpa.h"
/**
- * Fill the supplied hardware pool with skbuffs
- *
+ * cvm_oct_fill_hw_skbuff - fill the supplied hardware pool with skbuffs
* @pool: Pool to allocate an skbuff for
* @size: Size of the buffer needed for the pool
* @elements: Number of buffers to allocate
+ *
+ * Returns the actual number of buffers allocated.
*/
static int cvm_oct_fill_hw_skbuff(int pool, int size, int elements)
{
@@ -62,8 +63,7 @@ static int cvm_oct_fill_hw_skbuff(int pool, int size, int elements)
}
/**
- * Free the supplied hardware pool of skbuffs
- *
+ * cvm_oct_free_hw_skbuff- free hardware pool skbuffs
* @pool: Pool to allocate an skbuff for
* @size: Size of the buffer needed for the pool
* @elements: Number of buffers to allocate
@@ -91,11 +91,12 @@ static void cvm_oct_free_hw_skbuff(int pool, int size, int elements)
}
/**
- * This function fills a hardware pool with memory.
- *
+ * cvm_oct_fill_hw_memory - fill a hardware pool with memory.
* @pool: Pool to populate
* @size: Size of each buffer in the pool
* @elements: Number of buffers to allocate
+ *
+ * Returns the actual number of buffers allocated.
*/
static int cvm_oct_fill_hw_memory(int pool, int size, int elements)
{
@@ -129,8 +130,7 @@ static int cvm_oct_fill_hw_memory(int pool, int size, int elements)
}
/**
- * Free memory previously allocated with cvm_oct_fill_hw_memory
- *
+ * cvm_oct_free_hw_memory - Free memory allocated by cvm_oct_fill_hw_memory
* @pool: FPA pool to free
* @size: Size of each buffer in the pool
* @elements: Number of buffers that should be in the pool
diff --git a/drivers/staging/octeon/ethernet-rx.c b/drivers/staging/octeon/ethernet-rx.c
index b2e6ab6..cb38f9e 100644
--- a/drivers/staging/octeon/ethernet-rx.c
+++ b/drivers/staging/octeon/ethernet-rx.c
@@ -124,8 +124,9 @@ static void cvm_oct_no_more_work(void)
}
/**
- * Interrupt handler. The interrupt occurs whenever the POW
- * has packets in our group.
+ * cvm_oct_do_interrupt - interrupt handler.
+ *
+ * The interrupt occurs whenever the POW has packets in our group.
*
*/
static irqreturn_t cvm_oct_do_interrupt(int cpl, void *dev_id)
@@ -138,10 +139,9 @@ static irqreturn_t cvm_oct_do_interrupt(int cpl, void *dev_id)
}
/**
- * This is called on receive errors, and determines if the packet
- * can be dropped early-on in cvm_oct_tasklet_rx().
- *
+ * cvm_oct_check_rcv_error - process receive errors
* @work: Work queue entry pointing to the packet.
+ *
* Returns Non-zero if the packet can be dropped, zero otherwise.
*/
static inline int cvm_oct_check_rcv_error(cvmx_wqe_t *work)
@@ -224,10 +224,11 @@ static inline int cvm_oct_check_rcv_error(cvmx_wqe_t *work)
}
/**
- * The NAPI poll function.
- *
+ * cvm_oct_napi_poll - the NAPI poll function.
* @napi: The NAPI instance, or null if called from cvm_oct_poll_controller
* @budget: Maximum number of packets to receive.
+ *
+ * Returns the number of packets processed.
*/
static int cvm_oct_napi_poll(struct napi_struct *napi, int budget)
{
@@ -484,7 +485,7 @@ static int cvm_oct_napi_poll(struct napi_struct *napi, int budget)
#ifdef CONFIG_NET_POLL_CONTROLLER
/**
- * This is called when the kernel needs to manually poll the
+ * cvm_oct_poll_controller - poll for receive packets
* device.
*
* @dev: Device to poll. Unused
diff --git a/drivers/staging/octeon/ethernet-tx.c b/drivers/staging/octeon/ethernet-tx.c
index 5175247..afc2b73 100644
--- a/drivers/staging/octeon/ethernet-tx.c
+++ b/drivers/staging/octeon/ethernet-tx.c
@@ -137,11 +137,11 @@ void cvm_oct_free_tx_skbs(struct net_device *dev)
}
/**
- * Packet transmit
- *
+ * cvm_oct_xmit - transmit a packet
* @skb: Packet to send
* @dev: Device info structure
- * Returns Always returns zero
+ *
+ * Returns Always returns NETDEV_TX_OK
*/
int cvm_oct_xmit(struct sk_buff *skb, struct net_device *dev)
{
@@ -510,10 +510,10 @@ skip_xmit:
}
/**
- * Packet transmit to the POW
- *
+ * cvm_oct_xmit_pow - transmit a packet to the POW
* @skb: Packet to send
* @dev: Device info structure
+
* Returns Always returns zero
*/
int cvm_oct_xmit_pow(struct sk_buff *skb, struct net_device *dev)
@@ -661,9 +661,9 @@ int cvm_oct_xmit_pow(struct sk_buff *skb, struct net_device *dev)
}
/**
- * This function frees all skb that are currently queued for TX.
- *
+ * cvm_oct_tx_shutdown_dev - free all skb that are currently queued for TX.
* @dev: Device being shutdown
+ *
*/
void cvm_oct_tx_shutdown_dev(struct net_device *dev)
{
diff --git a/drivers/staging/octeon/ethernet-util.h b/drivers/staging/octeon/ethernet-util.h
index 37b6659..2346756 100644
--- a/drivers/staging/octeon/ethernet-util.h
+++ b/drivers/staging/octeon/ethernet-util.h
@@ -30,10 +30,9 @@
} while (0)
/**
- * Given a packet data address, return a pointer to the
- * beginning of the packet buffer.
- *
+ * cvm_oct_get_buffer_ptr - convert packet data address to pointer
* @packet_ptr: Packet data hardware address
+ *
* Returns Packet buffer pointer
*/
static inline void *cvm_oct_get_buffer_ptr(union cvmx_buf_ptr packet_ptr)
@@ -43,9 +42,7 @@ static inline void *cvm_oct_get_buffer_ptr(union cvmx_buf_ptr packet_ptr)
}
/**
- * Given an IPD/PKO port number, return the logical interface it is
- * on.
- *
+ * INTERFACE - convert IPD port to locgical interface
* @ipd_port: Port to check
*
* Returns Logical interface
@@ -65,9 +62,7 @@ static inline int INTERFACE(int ipd_port)
}
/**
- * Given an IPD/PKO port number, return the port's index on a
- * logical interface.
- *
+ * INDEX - convert IPD/PKO port number to the port's interface index
* @ipd_port: Port to check
*
* Returns Index into interface port list
diff --git a/drivers/staging/octeon/ethernet.c b/drivers/staging/octeon/ethernet.c
index 5ee60ab..45cb4c7 100644
--- a/drivers/staging/octeon/ethernet.c
+++ b/drivers/staging/octeon/ethernet.c
@@ -176,9 +176,6 @@ static void cvm_oct_periodic_worker(struct work_struct *work)
queue_delayed_work(cvm_oct_poll_queue, &priv->port_periodic_work, HZ);
}
-/**
- * Configure common hardware for all interfaces
- */
static __init void cvm_oct_configure_common_hw(void)
{
/* Setup the FPA */
@@ -198,10 +195,10 @@ static __init void cvm_oct_configure_common_hw(void)
}
/**
- * Free a work queue entry received in a intercept callback.
+ * cvm_oct_free_work- Free a work queue entry
+ *
+ * @work_queue_entry: Work queue entry to free
*
- * @work_queue_entry:
- * Work queue entry to free
* Returns Zero on success, Negative on failure.
*/
int cvm_oct_free_work(void *work_queue_entry)
@@ -228,9 +225,9 @@ int cvm_oct_free_work(void *work_queue_entry)
EXPORT_SYMBOL(cvm_oct_free_work);
/**
- * Get the low level ethernet statistics
- *
+ * cvm_oct_common_get_stats - get the low level ethernet statistics
* @dev: Device to get the statistics from
+ *
* Returns Pointer to the statistics
*/
static struct net_device_stats *cvm_oct_common_get_stats(struct net_device *dev)
@@ -274,8 +271,7 @@ static struct net_device_stats *cvm_oct_common_get_stats(struct net_device *dev)
}
/**
- * Change the link MTU. Unimplemented
- *
+ * cvm_oct_common_change_mtu - change the link MTU
* @dev: Device to change
* @new_mtu: The new MTU
*
@@ -339,8 +335,7 @@ static int cvm_oct_common_change_mtu(struct net_device *dev, int new_mtu)
}
/**
- * Set the multicast list. Currently unimplemented.
- *
+ * cvm_oct_common_set_multicast_list - set the multicast list
* @dev: Device to work on
*/
static void cvm_oct_common_set_multicast_list(struct net_device *dev)
@@ -395,10 +390,10 @@ static void cvm_oct_common_set_multicast_list(struct net_device *dev)
}
/**
- * Set the hardware MAC address for a device
- *
- * @dev: Device to change the MAC address for
- * @addr: Address structure to change it too. MAC address is addr + 2.
+ * cvm_oct_common_set_mac_address - set the hardware MAC address for a device
+ * @dev: The device in question.
+ * @addr: Address structure to change it too.
+
* Returns Zero on success
*/
static int cvm_oct_common_set_mac_address(struct net_device *dev, void *addr)
@@ -445,9 +440,9 @@ static int cvm_oct_common_set_mac_address(struct net_device *dev, void *addr)
}
/**
- * Per network device initialization
- *
+ * cvm_oct_common_init - per network device initialization
* @dev: Device to initialize
+ *
* Returns Zero on success
*/
int cvm_oct_common_init(struct net_device *dev)
@@ -603,12 +598,6 @@ static const struct net_device_ops cvm_oct_pow_netdev_ops = {
extern void octeon_mdiobus_force_mod_depencency(void);
-/**
- * Module/ driver initialization. Creates the linux network
- * devices.
- *
- * Returns Zero on success
- */
static int __init cvm_oct_init_module(void)
{
int num_interfaces;
@@ -802,11 +791,6 @@ static int __init cvm_oct_init_module(void)
return 0;
}
-/**
- * Module / driver shutdown
- *
- * Returns Zero on success
- */
static void __exit cvm_oct_cleanup_module(void)
{
int port;
diff --git a/drivers/staging/octeon/octeon-ethernet.h b/drivers/staging/octeon/octeon-ethernet.h
index db2a3cc..d581925 100644
--- a/drivers/staging/octeon/octeon-ethernet.h
+++ b/drivers/staging/octeon/octeon-ethernet.h
@@ -61,13 +61,6 @@ struct octeon_ethernet {
struct work_struct port_work; /* may be unused. */
};
-/**
- * Free a work queue entry received in a intercept callback.
- *
- * @work_queue_entry:
- * Work queue entry to free
- * Returns Zero on success, Negative on failure.
- */
int cvm_oct_free_work(void *work_queue_entry);
extern int cvm_oct_rgmii_init(struct net_device *dev);
--
1.6.6
^ permalink raw reply related
* [PATCH 0/2] Staging: Clean up octeon Ethernet some.
From: David Daney @ 2010-02-17 1:24 UTC (permalink / raw)
To: Ralf Baechle, linux-mips, Netdev, Greg Kroah-Hartman
These two patches clean up some comments and get rid of the code that
creates a non-standard file in /proc
As with the previous octeon_ethernet patches, these can probably go
via Ralf's linux-mips.org tree.
I will reply with the two patches.
David Daney (2):
Staging: Octeon: Reformat a bunch of comments.
Staging: Octeon: Remove /proc/octeon_ethernet_stats
drivers/staging/octeon/Makefile | 1 -
drivers/staging/octeon/ethernet-mdio.c | 6 +-
drivers/staging/octeon/ethernet-mem.c | 16 ++--
drivers/staging/octeon/ethernet-proc.c | 144
------------------------------
drivers/staging/octeon/ethernet-proc.h | 29 ------
drivers/staging/octeon/ethernet-rx.c | 17 ++--
drivers/staging/octeon/ethernet-tx.c | 14 ++--
drivers/staging/octeon/ethernet-util.h | 13 +--
drivers/staging/octeon/ethernet.c | 46 +++-------
drivers/staging/octeon/octeon-ethernet.h | 7 --
10 files changed, 44 insertions(+), 249 deletions(-)
delete mode 100644 drivers/staging/octeon/ethernet-proc.c
delete mode 100644 drivers/staging/octeon/ethernet-proc.h
^ permalink raw reply
* [PATCH net-next 2/2] percpu: add __percpu sparse annotations to net drivers
From: Tejun Heo @ 2010-02-17 1:21 UTC (permalink / raw)
To: David S. Miller, lkml, netdev, Stephen Rothwell
In-Reply-To: <4B7B445A.1080108@kernel.org>
Add __percpu sparse annotations to net drivers.
These annotations are to make sparse consider percpu variables to be
in a different address space and warn if accessed without going
through percpu accessors. This patch doesn't affect normal builds.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: David S. Miller <davem@davemloft.net>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Arnd Bergmann <arnd@arndb.de>
---
drivers/net/chelsio/sge.c | 2 +-
drivers/net/loopback.c | 16 +++++++++-------
drivers/net/veth.c | 4 ++--
include/linux/if_macvlan.h | 2 +-
4 files changed, 13 insertions(+), 11 deletions(-)
Index: net-next/drivers/net/chelsio/sge.c
===================================================================
--- net-next.orig/drivers/net/chelsio/sge.c
+++ net-next/drivers/net/chelsio/sge.c
@@ -267,7 +267,7 @@ struct sge {
struct sk_buff *espibug_skb[MAX_NPORTS];
u32 sge_control; /* shadow value of sge control reg */
struct sge_intr_counts stats;
- struct sge_port_stats *port_stats[MAX_NPORTS];
+ struct sge_port_stats __percpu *port_stats[MAX_NPORTS];
struct sched *tx_sched;
struct cmdQ cmdQ[SGE_CMDQ_N] ____cacheline_aligned_in_smp;
};
Index: net-next/drivers/net/loopback.c
===================================================================
--- net-next.orig/drivers/net/loopback.c
+++ net-next/drivers/net/loopback.c
@@ -72,7 +72,8 @@ struct pcpu_lstats {
static netdev_tx_t loopback_xmit(struct sk_buff *skb,
struct net_device *dev)
{
- struct pcpu_lstats *pcpu_lstats, *lb_stats;
+ struct pcpu_lstats __percpu *pcpu_lstats;
+ struct pcpu_lstats *lb_stats;
int len;
skb_orphan(skb);
@@ -80,7 +81,7 @@ static netdev_tx_t loopback_xmit(struct
skb->protocol = eth_type_trans(skb, dev);
/* it's OK to use per_cpu_ptr() because BHs are off */
- pcpu_lstats = dev->ml_priv;
+ pcpu_lstats = (void __percpu __force *)dev->ml_priv;
lb_stats = this_cpu_ptr(pcpu_lstats);
len = skb->len;
@@ -95,14 +96,14 @@ static netdev_tx_t loopback_xmit(struct
static struct net_device_stats *loopback_get_stats(struct net_device *dev)
{
- const struct pcpu_lstats *pcpu_lstats;
+ const struct pcpu_lstats __percpu *pcpu_lstats;
struct net_device_stats *stats = &dev->stats;
unsigned long bytes = 0;
unsigned long packets = 0;
unsigned long drops = 0;
int i;
- pcpu_lstats = dev->ml_priv;
+ pcpu_lstats = (void __percpu __force *)dev->ml_priv;
for_each_possible_cpu(i) {
const struct pcpu_lstats *lb_stats;
@@ -135,19 +136,20 @@ static const struct ethtool_ops loopback
static int loopback_dev_init(struct net_device *dev)
{
- struct pcpu_lstats *lstats;
+ struct pcpu_lstats __percpu *lstats;
lstats = alloc_percpu(struct pcpu_lstats);
if (!lstats)
return -ENOMEM;
- dev->ml_priv = lstats;
+ dev->ml_priv = (void __force *)lstats;
return 0;
}
static void loopback_dev_free(struct net_device *dev)
{
- struct pcpu_lstats *lstats = dev->ml_priv;
+ struct pcpu_lstats __percpu *lstats =
+ (void __percpu __force *)dev->ml_priv;
free_percpu(lstats);
free_netdev(dev);
Index: net-next/drivers/net/veth.c
===================================================================
--- net-next.orig/drivers/net/veth.c
+++ net-next/drivers/net/veth.c
@@ -34,7 +34,7 @@ struct veth_net_stats {
struct veth_priv {
struct net_device *peer;
- struct veth_net_stats *stats;
+ struct veth_net_stats __percpu *stats;
unsigned ip_summed;
};
@@ -263,7 +263,7 @@ static int veth_change_mtu(struct net_de
static int veth_dev_init(struct net_device *dev)
{
- struct veth_net_stats *stats;
+ struct veth_net_stats __percpu *stats;
struct veth_priv *priv;
stats = alloc_percpu(struct veth_net_stats);
Index: net-next/include/linux/if_macvlan.h
===================================================================
--- net-next.orig/include/linux/if_macvlan.h
+++ net-next/include/linux/if_macvlan.h
@@ -30,7 +30,7 @@ struct macvlan_dev {
struct hlist_node hlist;
struct macvlan_port *port;
struct net_device *lowerdev;
- struct macvlan_rx_stats *rx_stats;
+ struct macvlan_rx_stats __percpu *rx_stats;
enum macvlan_mode mode;
int (*receive)(struct sk_buff *skb);
int (*forward)(struct net_device *dev, struct sk_buff *skb);
^ permalink raw reply
* [PATCH net-next 1/2] percpu: add __percpu sparse annotations to net
From: Tejun Heo @ 2010-02-17 1:20 UTC (permalink / raw)
To: David S. Miller, lkml, netdev, Stephen Rothwell
Add __percpu sparse annotations to net.
These annotations are to make sparse consider percpu variables to be
in a different address space and warn if accessed without going
through percpu accessors. This patch doesn't affect normal builds.
The macro and type tricks around snmp stats make things a bit
interesting. DEFINE/DECLARE_SNMP_STAT() macros mark the target field
as __percpu and SNMP_UPD_PO_STATS() macro is updated accordingly. All
snmp_mib_*() users which used to cast the argument to (void **) are
updated to cast it to (void __percpu **).
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: David S. Miller <davem@davemloft.net>
Cc: Patrick McHardy <kaber@trash.net>
Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Cc: Vlad Yasevich <vladislav.yasevich@hp.com>
Cc: netdev@vger.kernel.org
---
David, these two patches require the following commit from mainline to
define dummy __percpu.
2938429501b73f6aeb312236eac7ed0416a07cd5
Thank you.
include/net/ip.h | 6 ++---
include/net/ipcomp.h | 2 -
include/net/neighbour.h | 2 -
include/net/netns/conntrack.h | 2 -
include/net/netns/core.h | 2 -
include/net/route.h | 2 -
include/net/snmp.h | 12 +++++++---
include/net/tcp.h | 2 -
net/8021q/vlan.h | 2 -
net/dccp/proto.c | 5 ++--
net/ipv4/af_inet.c | 46 +++++++++++++++++++++---------------------
net/ipv4/proc.c | 28 ++++++++++++-------------
net/ipv4/route.c | 2 -
net/ipv4/tcp.c | 21 ++++++++++---------
net/ipv6/addrconf.c | 24 ++++++++++-----------
net/ipv6/af_inet6.c | 28 ++++++++++++-------------
net/ipv6/proc.c | 23 +++++++++++----------
net/sctp/proc.c | 2 -
net/sctp/protocol.c | 5 ++--
net/xfrm/xfrm_ipcomp.c | 16 +++++++-------
net/xfrm/xfrm_policy.c | 6 ++---
net/xfrm/xfrm_proc.c | 3 +-
22 files changed, 127 insertions(+), 114 deletions(-)
Index: net-next/include/net/ip.h
===================================================================
--- net-next.orig/include/net/ip.h
+++ net-next/include/net/ip.h
@@ -174,9 +174,9 @@ extern struct ipv4_config ipv4_config;
#define NET_ADD_STATS_BH(net, field, adnd) SNMP_ADD_STATS_BH((net)->mib.net_statistics, field, adnd)
#define NET_ADD_STATS_USER(net, field, adnd) SNMP_ADD_STATS_USER((net)->mib.net_statistics, field, adnd)
-extern unsigned long snmp_fold_field(void *mib[], int offt);
-extern int snmp_mib_init(void *ptr[2], size_t mibsize);
-extern void snmp_mib_free(void *ptr[2]);
+extern unsigned long snmp_fold_field(void __percpu *mib[], int offt);
+extern int snmp_mib_init(void __percpu *ptr[2], size_t mibsize);
+extern void snmp_mib_free(void __percpu *ptr[2]);
extern struct local_ports {
seqlock_t lock;
Index: net-next/include/net/ipcomp.h
===================================================================
--- net-next.orig/include/net/ipcomp.h
+++ net-next/include/net/ipcomp.h
@@ -9,7 +9,7 @@ struct crypto_comp;
struct ipcomp_data {
u16 threshold;
- struct crypto_comp **tfms;
+ struct crypto_comp * __percpu *tfms;
};
struct ip_comp_hdr;
Index: net-next/include/net/neighbour.h
===================================================================
--- net-next.orig/include/net/neighbour.h
+++ net-next/include/net/neighbour.h
@@ -164,7 +164,7 @@ struct neigh_table {
rwlock_t lock;
unsigned long last_rand;
struct kmem_cache *kmem_cachep;
- struct neigh_statistics *stats;
+ struct neigh_statistics __percpu *stats;
struct neighbour **hash_buckets;
unsigned int hash_mask;
__u32 hash_rnd;
Index: net-next/include/net/netns/conntrack.h
===================================================================
--- net-next.orig/include/net/netns/conntrack.h
+++ net-next/include/net/netns/conntrack.h
@@ -17,7 +17,7 @@ struct netns_ct {
struct hlist_head *expect_hash;
struct hlist_nulls_head unconfirmed;
struct hlist_nulls_head dying;
- struct ip_conntrack_stat *stat;
+ struct ip_conntrack_stat __percpu *stat;
int sysctl_events;
unsigned int sysctl_events_retry_timeout;
int sysctl_acct;
Index: net-next/include/net/netns/core.h
===================================================================
--- net-next.orig/include/net/netns/core.h
+++ net-next/include/net/netns/core.h
@@ -10,7 +10,7 @@ struct netns_core {
int sysctl_somaxconn;
- struct prot_inuse *inuse;
+ struct prot_inuse __percpu *inuse;
};
#endif
Index: net-next/include/net/route.h
===================================================================
--- net-next.orig/include/net/route.h
+++ net-next/include/net/route.h
@@ -101,7 +101,7 @@ struct rt_cache_stat {
unsigned int out_hlist_search;
};
-extern struct ip_rt_acct *ip_rt_acct;
+extern struct ip_rt_acct __percpu *ip_rt_acct;
struct in_device;
extern int ip_rt_init(void);
Index: net-next/include/net/snmp.h
===================================================================
--- net-next.orig/include/net/snmp.h
+++ net-next/include/net/snmp.h
@@ -129,9 +129,9 @@ struct linux_xfrm_mib {
* nonlocked_atomic_inc() primitives -AK
*/
#define DEFINE_SNMP_STAT(type, name) \
- __typeof__(type) *name[2]
+ __typeof__(type) __percpu *name[2]
#define DECLARE_SNMP_STAT(type, name) \
- extern __typeof__(type) *name[2]
+ extern __typeof__(type) __percpu *name[2]
#define SNMP_STAT_BHPTR(name) (name[0])
#define SNMP_STAT_USRPTR(name) (name[1])
@@ -148,9 +148,13 @@ struct linux_xfrm_mib {
__this_cpu_add(mib[0]->mibs[field], addend)
#define SNMP_ADD_STATS_USER(mib, field, addend) \
this_cpu_add(mib[1]->mibs[field], addend)
+/*
+ * Use "__typeof__(*mib[0]) *ptr" instead of "__typeof__(mib[0]) ptr"
+ * to make @ptr a non-percpu pointer.
+ */
#define SNMP_UPD_PO_STATS(mib, basefield, addend) \
do { \
- __typeof__(mib[0]) ptr; \
+ __typeof__(*mib[0]) *ptr; \
preempt_disable(); \
ptr = this_cpu_ptr((mib)[!in_softirq()]); \
ptr->mibs[basefield##PKTS]++; \
@@ -159,7 +163,7 @@ struct linux_xfrm_mib {
} while (0)
#define SNMP_UPD_PO_STATS_BH(mib, basefield, addend) \
do { \
- __typeof__(mib[0]) ptr = \
+ __typeof__(*mib[0]) *ptr = \
__this_cpu_ptr((mib)[!in_softirq()]); \
ptr->mibs[basefield##PKTS]++; \
ptr->mibs[basefield##OCTETS] += addend;\
Index: net-next/include/net/tcp.h
===================================================================
--- net-next.orig/include/net/tcp.h
+++ net-next/include/net/tcp.h
@@ -1189,7 +1189,7 @@ extern int tcp_v4_md5_do_del(struct so
#define tcp_twsk_md5_key(twsk) NULL
#endif
-extern struct tcp_md5sig_pool **tcp_alloc_md5sig_pool(struct sock *);
+extern struct tcp_md5sig_pool * __percpu *tcp_alloc_md5sig_pool(struct sock *);
extern void tcp_free_md5sig_pool(void);
extern struct tcp_md5sig_pool *__tcp_get_md5sig_pool(int cpu);
Index: net-next/net/8021q/vlan.h
===================================================================
--- net-next.orig/net/8021q/vlan.h
+++ net-next/net/8021q/vlan.h
@@ -61,7 +61,7 @@ struct vlan_dev_info {
struct proc_dir_entry *dent;
unsigned long cnt_inc_headroom_on_tx;
unsigned long cnt_encap_on_xmit;
- struct vlan_rx_stats *vlan_rx_stats;
+ struct vlan_rx_stats __percpu *vlan_rx_stats;
};
static inline struct vlan_dev_info *vlan_dev_info(const struct net_device *dev)
Index: net-next/net/dccp/proto.c
===================================================================
--- net-next.orig/net/dccp/proto.c
+++ net-next/net/dccp/proto.c
@@ -1005,12 +1005,13 @@ EXPORT_SYMBOL_GPL(dccp_shutdown);
static inline int dccp_mib_init(void)
{
- return snmp_mib_init((void**)dccp_statistics, sizeof(struct dccp_mib));
+ return snmp_mib_init((void __percpu **)dccp_statistics,
+ sizeof(struct dccp_mib));
}
static inline void dccp_mib_exit(void)
{
- snmp_mib_free((void**)dccp_statistics);
+ snmp_mib_free((void __percpu **)dccp_statistics);
}
static int thash_entries;
Index: net-next/net/ipv4/af_inet.c
===================================================================
--- net-next.orig/net/ipv4/af_inet.c
+++ net-next/net/ipv4/af_inet.c
@@ -1385,7 +1385,7 @@ int inet_ctl_sock_create(struct sock **s
}
EXPORT_SYMBOL_GPL(inet_ctl_sock_create);
-unsigned long snmp_fold_field(void *mib[], int offt)
+unsigned long snmp_fold_field(void __percpu *mib[], int offt)
{
unsigned long res = 0;
int i;
@@ -1398,7 +1398,7 @@ unsigned long snmp_fold_field(void *mib[
}
EXPORT_SYMBOL_GPL(snmp_fold_field);
-int snmp_mib_init(void *ptr[2], size_t mibsize)
+int snmp_mib_init(void __percpu *ptr[2], size_t mibsize)
{
BUG_ON(ptr == NULL);
ptr[0] = __alloc_percpu(mibsize, __alignof__(unsigned long long));
@@ -1416,7 +1416,7 @@ err0:
}
EXPORT_SYMBOL_GPL(snmp_mib_init);
-void snmp_mib_free(void *ptr[2])
+void snmp_mib_free(void __percpu *ptr[2])
{
BUG_ON(ptr == NULL);
free_percpu(ptr[0]);
@@ -1460,25 +1460,25 @@ static const struct net_protocol icmp_pr
static __net_init int ipv4_mib_init_net(struct net *net)
{
- if (snmp_mib_init((void **)net->mib.tcp_statistics,
+ if (snmp_mib_init((void __percpu **)net->mib.tcp_statistics,
sizeof(struct tcp_mib)) < 0)
goto err_tcp_mib;
- if (snmp_mib_init((void **)net->mib.ip_statistics,
+ if (snmp_mib_init((void __percpu **)net->mib.ip_statistics,
sizeof(struct ipstats_mib)) < 0)
goto err_ip_mib;
- if (snmp_mib_init((void **)net->mib.net_statistics,
+ if (snmp_mib_init((void __percpu **)net->mib.net_statistics,
sizeof(struct linux_mib)) < 0)
goto err_net_mib;
- if (snmp_mib_init((void **)net->mib.udp_statistics,
+ if (snmp_mib_init((void __percpu **)net->mib.udp_statistics,
sizeof(struct udp_mib)) < 0)
goto err_udp_mib;
- if (snmp_mib_init((void **)net->mib.udplite_statistics,
+ if (snmp_mib_init((void __percpu **)net->mib.udplite_statistics,
sizeof(struct udp_mib)) < 0)
goto err_udplite_mib;
- if (snmp_mib_init((void **)net->mib.icmp_statistics,
+ if (snmp_mib_init((void __percpu **)net->mib.icmp_statistics,
sizeof(struct icmp_mib)) < 0)
goto err_icmp_mib;
- if (snmp_mib_init((void **)net->mib.icmpmsg_statistics,
+ if (snmp_mib_init((void __percpu **)net->mib.icmpmsg_statistics,
sizeof(struct icmpmsg_mib)) < 0)
goto err_icmpmsg_mib;
@@ -1486,30 +1486,30 @@ static __net_init int ipv4_mib_init_net(
return 0;
err_icmpmsg_mib:
- snmp_mib_free((void **)net->mib.icmp_statistics);
+ snmp_mib_free((void __percpu **)net->mib.icmp_statistics);
err_icmp_mib:
- snmp_mib_free((void **)net->mib.udplite_statistics);
+ snmp_mib_free((void __percpu **)net->mib.udplite_statistics);
err_udplite_mib:
- snmp_mib_free((void **)net->mib.udp_statistics);
+ snmp_mib_free((void __percpu **)net->mib.udp_statistics);
err_udp_mib:
- snmp_mib_free((void **)net->mib.net_statistics);
+ snmp_mib_free((void __percpu **)net->mib.net_statistics);
err_net_mib:
- snmp_mib_free((void **)net->mib.ip_statistics);
+ snmp_mib_free((void __percpu **)net->mib.ip_statistics);
err_ip_mib:
- snmp_mib_free((void **)net->mib.tcp_statistics);
+ snmp_mib_free((void __percpu **)net->mib.tcp_statistics);
err_tcp_mib:
return -ENOMEM;
}
static __net_exit void ipv4_mib_exit_net(struct net *net)
{
- snmp_mib_free((void **)net->mib.icmpmsg_statistics);
- snmp_mib_free((void **)net->mib.icmp_statistics);
- snmp_mib_free((void **)net->mib.udplite_statistics);
- snmp_mib_free((void **)net->mib.udp_statistics);
- snmp_mib_free((void **)net->mib.net_statistics);
- snmp_mib_free((void **)net->mib.ip_statistics);
- snmp_mib_free((void **)net->mib.tcp_statistics);
+ snmp_mib_free((void __percpu **)net->mib.icmpmsg_statistics);
+ snmp_mib_free((void __percpu **)net->mib.icmp_statistics);
+ snmp_mib_free((void __percpu **)net->mib.udplite_statistics);
+ snmp_mib_free((void __percpu **)net->mib.udp_statistics);
+ snmp_mib_free((void __percpu **)net->mib.net_statistics);
+ snmp_mib_free((void __percpu **)net->mib.ip_statistics);
+ snmp_mib_free((void __percpu **)net->mib.tcp_statistics);
}
static __net_initdata struct pernet_operations ipv4_mib_ops = {
Index: net-next/net/ipv4/proc.c
===================================================================
--- net-next.orig/net/ipv4/proc.c
+++ net-next/net/ipv4/proc.c
@@ -280,7 +280,7 @@ static void icmpmsg_put(struct seq_file
count = 0;
for (i = 0; i < ICMPMSG_MIB_MAX; i++) {
- val = snmp_fold_field((void **) net->mib.icmpmsg_statistics, i);
+ val = snmp_fold_field((void __percpu **) net->mib.icmpmsg_statistics, i);
if (val) {
type[count] = i;
vals[count++] = val;
@@ -307,18 +307,18 @@ static void icmp_put(struct seq_file *se
for (i=0; icmpmibmap[i].name != NULL; i++)
seq_printf(seq, " Out%s", icmpmibmap[i].name);
seq_printf(seq, "\nIcmp: %lu %lu",
- snmp_fold_field((void **) net->mib.icmp_statistics, ICMP_MIB_INMSGS),
- snmp_fold_field((void **) net->mib.icmp_statistics, ICMP_MIB_INERRORS));
+ snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_INMSGS),
+ snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_INERRORS));
for (i=0; icmpmibmap[i].name != NULL; i++)
seq_printf(seq, " %lu",
- snmp_fold_field((void **) net->mib.icmpmsg_statistics,
+ snmp_fold_field((void __percpu **) net->mib.icmpmsg_statistics,
icmpmibmap[i].index));
seq_printf(seq, " %lu %lu",
- snmp_fold_field((void **) net->mib.icmp_statistics, ICMP_MIB_OUTMSGS),
- snmp_fold_field((void **) net->mib.icmp_statistics, ICMP_MIB_OUTERRORS));
+ snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_OUTMSGS),
+ snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_OUTERRORS));
for (i=0; icmpmibmap[i].name != NULL; i++)
seq_printf(seq, " %lu",
- snmp_fold_field((void **) net->mib.icmpmsg_statistics,
+ snmp_fold_field((void __percpu **) net->mib.icmpmsg_statistics,
icmpmibmap[i].index | 0x100));
}
@@ -341,7 +341,7 @@ static int snmp_seq_show(struct seq_file
for (i = 0; snmp4_ipstats_list[i].name != NULL; i++)
seq_printf(seq, " %lu",
- snmp_fold_field((void **)net->mib.ip_statistics,
+ snmp_fold_field((void __percpu **)net->mib.ip_statistics,
snmp4_ipstats_list[i].entry));
icmp_put(seq); /* RFC 2011 compatibility */
@@ -356,11 +356,11 @@ static int snmp_seq_show(struct seq_file
/* MaxConn field is signed, RFC 2012 */
if (snmp4_tcp_list[i].entry == TCP_MIB_MAXCONN)
seq_printf(seq, " %ld",
- snmp_fold_field((void **)net->mib.tcp_statistics,
+ snmp_fold_field((void __percpu **)net->mib.tcp_statistics,
snmp4_tcp_list[i].entry));
else
seq_printf(seq, " %lu",
- snmp_fold_field((void **)net->mib.tcp_statistics,
+ snmp_fold_field((void __percpu **)net->mib.tcp_statistics,
snmp4_tcp_list[i].entry));
}
@@ -371,7 +371,7 @@ static int snmp_seq_show(struct seq_file
seq_puts(seq, "\nUdp:");
for (i = 0; snmp4_udp_list[i].name != NULL; i++)
seq_printf(seq, " %lu",
- snmp_fold_field((void **)net->mib.udp_statistics,
+ snmp_fold_field((void __percpu **)net->mib.udp_statistics,
snmp4_udp_list[i].entry));
/* the UDP and UDP-Lite MIBs are the same */
@@ -382,7 +382,7 @@ static int snmp_seq_show(struct seq_file
seq_puts(seq, "\nUdpLite:");
for (i = 0; snmp4_udp_list[i].name != NULL; i++)
seq_printf(seq, " %lu",
- snmp_fold_field((void **)net->mib.udplite_statistics,
+ snmp_fold_field((void __percpu **)net->mib.udplite_statistics,
snmp4_udp_list[i].entry));
seq_putc(seq, '\n');
@@ -419,7 +419,7 @@ static int netstat_seq_show(struct seq_f
seq_puts(seq, "\nTcpExt:");
for (i = 0; snmp4_net_list[i].name != NULL; i++)
seq_printf(seq, " %lu",
- snmp_fold_field((void **)net->mib.net_statistics,
+ snmp_fold_field((void __percpu **)net->mib.net_statistics,
snmp4_net_list[i].entry));
seq_puts(seq, "\nIpExt:");
@@ -429,7 +429,7 @@ static int netstat_seq_show(struct seq_f
seq_puts(seq, "\nIpExt:");
for (i = 0; snmp4_ipextstats_list[i].name != NULL; i++)
seq_printf(seq, " %lu",
- snmp_fold_field((void **)net->mib.ip_statistics,
+ snmp_fold_field((void __percpu **)net->mib.ip_statistics,
snmp4_ipextstats_list[i].entry));
seq_putc(seq, '\n');
Index: net-next/net/ipv4/route.c
===================================================================
--- net-next.orig/net/ipv4/route.c
+++ net-next/net/ipv4/route.c
@@ -3334,7 +3334,7 @@ static __net_initdata struct pernet_oper
#ifdef CONFIG_NET_CLS_ROUTE
-struct ip_rt_acct *ip_rt_acct __read_mostly;
+struct ip_rt_acct __percpu *ip_rt_acct __read_mostly;
#endif /* CONFIG_NET_CLS_ROUTE */
static __initdata unsigned long rhash_entries;
Index: net-next/net/ipv4/tcp.c
===================================================================
--- net-next.orig/net/ipv4/tcp.c
+++ net-next/net/ipv4/tcp.c
@@ -2788,10 +2788,10 @@ EXPORT_SYMBOL(tcp_gro_complete);
#ifdef CONFIG_TCP_MD5SIG
static unsigned long tcp_md5sig_users;
-static struct tcp_md5sig_pool **tcp_md5sig_pool;
+static struct tcp_md5sig_pool * __percpu *tcp_md5sig_pool;
static DEFINE_SPINLOCK(tcp_md5sig_pool_lock);
-static void __tcp_free_md5sig_pool(struct tcp_md5sig_pool **pool)
+static void __tcp_free_md5sig_pool(struct tcp_md5sig_pool * __percpu *pool)
{
int cpu;
for_each_possible_cpu(cpu) {
@@ -2808,7 +2808,7 @@ static void __tcp_free_md5sig_pool(struc
void tcp_free_md5sig_pool(void)
{
- struct tcp_md5sig_pool **pool = NULL;
+ struct tcp_md5sig_pool * __percpu *pool = NULL;
spin_lock_bh(&tcp_md5sig_pool_lock);
if (--tcp_md5sig_users == 0) {
@@ -2822,10 +2822,11 @@ void tcp_free_md5sig_pool(void)
EXPORT_SYMBOL(tcp_free_md5sig_pool);
-static struct tcp_md5sig_pool **__tcp_alloc_md5sig_pool(struct sock *sk)
+static struct tcp_md5sig_pool * __percpu *
+__tcp_alloc_md5sig_pool(struct sock *sk)
{
int cpu;
- struct tcp_md5sig_pool **pool;
+ struct tcp_md5sig_pool * __percpu *pool;
pool = alloc_percpu(struct tcp_md5sig_pool *);
if (!pool)
@@ -2852,9 +2853,9 @@ out_free:
return NULL;
}
-struct tcp_md5sig_pool **tcp_alloc_md5sig_pool(struct sock *sk)
+struct tcp_md5sig_pool * __percpu *tcp_alloc_md5sig_pool(struct sock *sk)
{
- struct tcp_md5sig_pool **pool;
+ struct tcp_md5sig_pool * __percpu *pool;
int alloc = 0;
retry:
@@ -2873,7 +2874,9 @@ retry:
if (alloc) {
/* we cannot hold spinlock here because this may sleep. */
- struct tcp_md5sig_pool **p = __tcp_alloc_md5sig_pool(sk);
+ struct tcp_md5sig_pool * __percpu *p;
+
+ p = __tcp_alloc_md5sig_pool(sk);
spin_lock_bh(&tcp_md5sig_pool_lock);
if (!p) {
tcp_md5sig_users--;
@@ -2897,7 +2900,7 @@ EXPORT_SYMBOL(tcp_alloc_md5sig_pool);
struct tcp_md5sig_pool *__tcp_get_md5sig_pool(int cpu)
{
- struct tcp_md5sig_pool **p;
+ struct tcp_md5sig_pool * __percpu *p;
spin_lock_bh(&tcp_md5sig_pool_lock);
p = tcp_md5sig_pool;
if (p)
Index: net-next/net/ipv6/addrconf.c
===================================================================
--- net-next.orig/net/ipv6/addrconf.c
+++ net-next/net/ipv6/addrconf.c
@@ -278,31 +278,31 @@ static void addrconf_mod_timer(struct in
static int snmp6_alloc_dev(struct inet6_dev *idev)
{
- if (snmp_mib_init((void **)idev->stats.ipv6,
+ if (snmp_mib_init((void __percpu **)idev->stats.ipv6,
sizeof(struct ipstats_mib)) < 0)
goto err_ip;
- if (snmp_mib_init((void **)idev->stats.icmpv6,
+ if (snmp_mib_init((void __percpu **)idev->stats.icmpv6,
sizeof(struct icmpv6_mib)) < 0)
goto err_icmp;
- if (snmp_mib_init((void **)idev->stats.icmpv6msg,
+ if (snmp_mib_init((void __percpu **)idev->stats.icmpv6msg,
sizeof(struct icmpv6msg_mib)) < 0)
goto err_icmpmsg;
return 0;
err_icmpmsg:
- snmp_mib_free((void **)idev->stats.icmpv6);
+ snmp_mib_free((void __percpu **)idev->stats.icmpv6);
err_icmp:
- snmp_mib_free((void **)idev->stats.ipv6);
+ snmp_mib_free((void __percpu **)idev->stats.ipv6);
err_ip:
return -ENOMEM;
}
static void snmp6_free_dev(struct inet6_dev *idev)
{
- snmp_mib_free((void **)idev->stats.icmpv6msg);
- snmp_mib_free((void **)idev->stats.icmpv6);
- snmp_mib_free((void **)idev->stats.ipv6);
+ snmp_mib_free((void __percpu **)idev->stats.icmpv6msg);
+ snmp_mib_free((void __percpu **)idev->stats.icmpv6);
+ snmp_mib_free((void __percpu **)idev->stats.ipv6);
}
/* Nobody refers to this device, we may destroy it. */
@@ -3766,8 +3766,8 @@ static inline size_t inet6_if_nlmsg_size
);
}
-static inline void __snmp6_fill_stats(u64 *stats, void **mib, int items,
- int bytes)
+static inline void __snmp6_fill_stats(u64 *stats, void __percpu **mib,
+ int items, int bytes)
{
int i;
int pad = bytes - sizeof(u64) * items;
@@ -3786,10 +3786,10 @@ static void snmp6_fill_stats(u64 *stats,
{
switch(attrtype) {
case IFLA_INET6_STATS:
- __snmp6_fill_stats(stats, (void **)idev->stats.ipv6, IPSTATS_MIB_MAX, bytes);
+ __snmp6_fill_stats(stats, (void __percpu **)idev->stats.ipv6, IPSTATS_MIB_MAX, bytes);
break;
case IFLA_INET6_ICMP6STATS:
- __snmp6_fill_stats(stats, (void **)idev->stats.icmpv6, ICMP6_MIB_MAX, bytes);
+ __snmp6_fill_stats(stats, (void __percpu **)idev->stats.icmpv6, ICMP6_MIB_MAX, bytes);
break;
}
}
Index: net-next/net/ipv6/af_inet6.c
===================================================================
--- net-next.orig/net/ipv6/af_inet6.c
+++ net-next/net/ipv6/af_inet6.c
@@ -971,41 +971,41 @@ static void ipv6_packet_cleanup(void)
static int __net_init ipv6_init_mibs(struct net *net)
{
- if (snmp_mib_init((void **)net->mib.udp_stats_in6,
+ if (snmp_mib_init((void __percpu **)net->mib.udp_stats_in6,
sizeof (struct udp_mib)) < 0)
return -ENOMEM;
- if (snmp_mib_init((void **)net->mib.udplite_stats_in6,
+ if (snmp_mib_init((void __percpu **)net->mib.udplite_stats_in6,
sizeof (struct udp_mib)) < 0)
goto err_udplite_mib;
- if (snmp_mib_init((void **)net->mib.ipv6_statistics,
+ if (snmp_mib_init((void __percpu **)net->mib.ipv6_statistics,
sizeof(struct ipstats_mib)) < 0)
goto err_ip_mib;
- if (snmp_mib_init((void **)net->mib.icmpv6_statistics,
+ if (snmp_mib_init((void __percpu **)net->mib.icmpv6_statistics,
sizeof(struct icmpv6_mib)) < 0)
goto err_icmp_mib;
- if (snmp_mib_init((void **)net->mib.icmpv6msg_statistics,
+ if (snmp_mib_init((void __percpu **)net->mib.icmpv6msg_statistics,
sizeof(struct icmpv6msg_mib)) < 0)
goto err_icmpmsg_mib;
return 0;
err_icmpmsg_mib:
- snmp_mib_free((void **)net->mib.icmpv6_statistics);
+ snmp_mib_free((void __percpu **)net->mib.icmpv6_statistics);
err_icmp_mib:
- snmp_mib_free((void **)net->mib.ipv6_statistics);
+ snmp_mib_free((void __percpu **)net->mib.ipv6_statistics);
err_ip_mib:
- snmp_mib_free((void **)net->mib.udplite_stats_in6);
+ snmp_mib_free((void __percpu **)net->mib.udplite_stats_in6);
err_udplite_mib:
- snmp_mib_free((void **)net->mib.udp_stats_in6);
+ snmp_mib_free((void __percpu **)net->mib.udp_stats_in6);
return -ENOMEM;
}
static void ipv6_cleanup_mibs(struct net *net)
{
- snmp_mib_free((void **)net->mib.udp_stats_in6);
- snmp_mib_free((void **)net->mib.udplite_stats_in6);
- snmp_mib_free((void **)net->mib.ipv6_statistics);
- snmp_mib_free((void **)net->mib.icmpv6_statistics);
- snmp_mib_free((void **)net->mib.icmpv6msg_statistics);
+ snmp_mib_free((void __percpu **)net->mib.udp_stats_in6);
+ snmp_mib_free((void __percpu **)net->mib.udplite_stats_in6);
+ snmp_mib_free((void __percpu **)net->mib.ipv6_statistics);
+ snmp_mib_free((void __percpu **)net->mib.icmpv6_statistics);
+ snmp_mib_free((void __percpu **)net->mib.icmpv6msg_statistics);
}
static int __net_init inet6_net_init(struct net *net)
Index: net-next/net/ipv6/proc.c
===================================================================
--- net-next.orig/net/ipv6/proc.c
+++ net-next/net/ipv6/proc.c
@@ -136,7 +136,7 @@ static const struct snmp_mib snmp6_udpli
SNMP_MIB_SENTINEL
};
-static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void **mib)
+static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void __percpu **mib)
{
char name[32];
int i;
@@ -170,7 +170,7 @@ static void snmp6_seq_show_icmpv6msg(str
return;
}
-static void snmp6_seq_show_item(struct seq_file *seq, void **mib,
+static void snmp6_seq_show_item(struct seq_file *seq, void __percpu **mib,
const struct snmp_mib *itemlist)
{
int i;
@@ -183,14 +183,15 @@ static int snmp6_seq_show(struct seq_fil
{
struct net *net = (struct net *)seq->private;
- snmp6_seq_show_item(seq, (void **)net->mib.ipv6_statistics,
+ snmp6_seq_show_item(seq, (void __percpu **)net->mib.ipv6_statistics,
snmp6_ipstats_list);
- snmp6_seq_show_item(seq, (void **)net->mib.icmpv6_statistics,
+ snmp6_seq_show_item(seq, (void __percpu **)net->mib.icmpv6_statistics,
snmp6_icmp6_list);
- snmp6_seq_show_icmpv6msg(seq, (void **)net->mib.icmpv6msg_statistics);
- snmp6_seq_show_item(seq, (void **)net->mib.udp_stats_in6,
+ snmp6_seq_show_icmpv6msg(seq,
+ (void __percpu **)net->mib.icmpv6msg_statistics);
+ snmp6_seq_show_item(seq, (void __percpu **)net->mib.udp_stats_in6,
snmp6_udp6_list);
- snmp6_seq_show_item(seq, (void **)net->mib.udplite_stats_in6,
+ snmp6_seq_show_item(seq, (void __percpu **)net->mib.udplite_stats_in6,
snmp6_udplite6_list);
return 0;
}
@@ -213,9 +214,11 @@ static int snmp6_dev_seq_show(struct seq
struct inet6_dev *idev = (struct inet6_dev *)seq->private;
seq_printf(seq, "%-32s\t%u\n", "ifIndex", idev->dev->ifindex);
- snmp6_seq_show_item(seq, (void **)idev->stats.ipv6, snmp6_ipstats_list);
- snmp6_seq_show_item(seq, (void **)idev->stats.icmpv6, snmp6_icmp6_list);
- snmp6_seq_show_icmpv6msg(seq, (void **)idev->stats.icmpv6msg);
+ snmp6_seq_show_item(seq, (void __percpu **)idev->stats.ipv6,
+ snmp6_ipstats_list);
+ snmp6_seq_show_item(seq, (void __percpu **)idev->stats.icmpv6,
+ snmp6_icmp6_list);
+ snmp6_seq_show_icmpv6msg(seq, (void __percpu **)idev->stats.icmpv6msg);
return 0;
}
Index: net-next/net/sctp/proc.c
===================================================================
--- net-next.orig/net/sctp/proc.c
+++ net-next/net/sctp/proc.c
@@ -83,7 +83,7 @@ static int sctp_snmp_seq_show(struct seq
for (i = 0; sctp_snmp_list[i].name != NULL; i++)
seq_printf(seq, "%-32s\t%ld\n", sctp_snmp_list[i].name,
- snmp_fold_field((void **)sctp_statistics,
+ snmp_fold_field((void __percpu **)sctp_statistics,
sctp_snmp_list[i].entry));
return 0;
Index: net-next/net/sctp/protocol.c
===================================================================
--- net-next.orig/net/sctp/protocol.c
+++ net-next/net/sctp/protocol.c
@@ -996,12 +996,13 @@ int sctp_register_pf(struct sctp_pf *pf,
static inline int init_sctp_mibs(void)
{
- return snmp_mib_init((void**)sctp_statistics, sizeof(struct sctp_mib));
+ return snmp_mib_init((void __percpu **)sctp_statistics,
+ sizeof(struct sctp_mib));
}
static inline void cleanup_sctp_mibs(void)
{
- snmp_mib_free((void**)sctp_statistics);
+ snmp_mib_free((void __percpu **)sctp_statistics);
}
static void sctp_v4_pf_init(void)
Index: net-next/net/xfrm/xfrm_ipcomp.c
===================================================================
--- net-next.orig/net/xfrm/xfrm_ipcomp.c
+++ net-next/net/xfrm/xfrm_ipcomp.c
@@ -30,12 +30,12 @@
struct ipcomp_tfms {
struct list_head list;
- struct crypto_comp **tfms;
+ struct crypto_comp * __percpu *tfms;
int users;
};
static DEFINE_MUTEX(ipcomp_resource_mutex);
-static void **ipcomp_scratches;
+static void * __percpu *ipcomp_scratches;
static int ipcomp_scratch_users;
static LIST_HEAD(ipcomp_tfms_list);
@@ -200,7 +200,7 @@ EXPORT_SYMBOL_GPL(ipcomp_output);
static void ipcomp_free_scratches(void)
{
int i;
- void **scratches;
+ void * __percpu *scratches;
if (--ipcomp_scratch_users)
return;
@@ -215,10 +215,10 @@ static void ipcomp_free_scratches(void)
free_percpu(scratches);
}
-static void **ipcomp_alloc_scratches(void)
+static void * __percpu *ipcomp_alloc_scratches(void)
{
int i;
- void **scratches;
+ void * __percpu *scratches;
if (ipcomp_scratch_users++)
return ipcomp_scratches;
@@ -239,7 +239,7 @@ static void **ipcomp_alloc_scratches(voi
return scratches;
}
-static void ipcomp_free_tfms(struct crypto_comp **tfms)
+static void ipcomp_free_tfms(struct crypto_comp * __percpu *tfms)
{
struct ipcomp_tfms *pos;
int cpu;
@@ -267,10 +267,10 @@ static void ipcomp_free_tfms(struct cryp
free_percpu(tfms);
}
-static struct crypto_comp **ipcomp_alloc_tfms(const char *alg_name)
+static struct crypto_comp * __percpu *ipcomp_alloc_tfms(const char *alg_name)
{
struct ipcomp_tfms *pos;
- struct crypto_comp **tfms;
+ struct crypto_comp * __percpu *tfms;
int cpu;
/* This can be any valid CPU ID so we don't need locking. */
Index: net-next/net/xfrm/xfrm_policy.c
===================================================================
--- net-next.orig/net/xfrm/xfrm_policy.c
+++ net-next/net/xfrm/xfrm_policy.c
@@ -2428,19 +2428,19 @@ static int __net_init xfrm_statistics_in
{
int rv;
- if (snmp_mib_init((void **)net->mib.xfrm_statistics,
+ if (snmp_mib_init((void __percpu **)net->mib.xfrm_statistics,
sizeof(struct linux_xfrm_mib)) < 0)
return -ENOMEM;
rv = xfrm_proc_init(net);
if (rv < 0)
- snmp_mib_free((void **)net->mib.xfrm_statistics);
+ snmp_mib_free((void __percpu **)net->mib.xfrm_statistics);
return rv;
}
static void xfrm_statistics_fini(struct net *net)
{
xfrm_proc_fini(net);
- snmp_mib_free((void **)net->mib.xfrm_statistics);
+ snmp_mib_free((void __percpu **)net->mib.xfrm_statistics);
}
#else
static int __net_init xfrm_statistics_init(struct net *net)
Index: net-next/net/xfrm/xfrm_proc.c
===================================================================
--- net-next.orig/net/xfrm/xfrm_proc.c
+++ net-next/net/xfrm/xfrm_proc.c
@@ -50,7 +50,8 @@ static int xfrm_statistics_seq_show(stru
int i;
for (i=0; xfrm_mib_list[i].name; i++)
seq_printf(seq, "%-24s\t%lu\n", xfrm_mib_list[i].name,
- snmp_fold_field((void **)net->mib.xfrm_statistics,
+ snmp_fold_field((void __percpu **)
+ net->mib.xfrm_statistics,
xfrm_mib_list[i].entry));
return 0;
}
^ permalink raw reply
* [GIT]: Networking
From: David Miller @ 2010-02-17 0:46 UTC (permalink / raw)
To: torvalds; +Cc: akpm, netdev, linux-kernel
1) When TCP timestamps are disabled, we can end up with an
->srtt of zero after the three way handshake, and if the
right series of losses occur afterwards we end up in an
ACK war. Fix from Damian Lukowski.
2) Fix 1G flow control and WOL setup in IXGBE driver, from Don
Skidmore.
3) Type in SSB layer causes SDIOHOST case to not hookup parent
bus/device pointers properly. Fix from Michael Buesch.
4) Fix locking regressions wrt. ->poll() and changing IRQ coalesce
settings in VIA Velocity driver. From Simon Kagstrom.
5) Looking at the GRO setting by normal users is restricted
unnecessarily. From Stephen Hemminger.
6) Couple of wireless stragglers including a B43 throughput
regression fix from Larry Finger.
7) We pretend to handle IGMPv3 reports, but really the structure
is different so such handling is wrong. Fix from Herbert Xu.
8) be2net driver fix to properly initialize the version field to
zero in request headers. From Ajit Khaparde.
9) IPCOMP can double release an object on errors, from Herbert Xu.
10) If an error happens when cloning an XFRM state we'll leak
the module reference because we directly kfree() instead of
doing a put on the object. Also from Herbert Xu.
Please pull, thanks a lot!
The following changes since commit 88626272549b94310975932a9970d11a9b362a16:
Linus Torvalds (1):
Merge git://git.kernel.org/.../agk/linux-2.6-dm
are available in the git repository at:
master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6.git master
Ajit Khaparde (1):
be2net: set proper value to version field in req hdr
Benoit Papillault (1):
mac80211: Fix probe request filtering in IBSS mode
Damian Lukowski (1):
tcp: fix ICMP-RTO war
David S. Miller (2):
Merge branch 'master' of git://git.kernel.org/.../linville/wireless-2.6
Merge branch 'master' of ssh://master.kernel.org/.../linville/wireless-2.6
Don Skidmore (2):
ixgbe: Fix - Do not allow Rx FC on 82598 at 1G due to errata
ixgbe: fix WOL register setup for 82599
Herbert Xu (3):
inet: Remove bogus IGMPv3 report handling
ipcomp: Avoid duplicate calls to ipcomp_destroy
xfrm: Fix xfrm_state_clone leak
Jelle Martijn Kok (1):
rtl8187: Add new device ID
Johannes Berg (1):
mac80211: fix deferred hardware scan requests
Juuso Oikarinen (1):
mac80211: fix handling of null-rate control in rate_control_get_rate
Larry Finger (1):
b43: Fix throughput regression
Michael Buesch (1):
ssb: Fix CONFIG_SSB_SDIOHOST typo
Roel Kluin (1):
iwmc3200wifi: Test of wrong pointer after kzalloc in iwm_mlme_update_bss_table()
Shanyu Zhao (1):
iwlwifi: fix AMSDU Rx after paged Rx patch
Simon Kagstrom (3):
via-velocity: Remove unused IRQ status parameter from rx_srv and tx_srv
via-velocity: Take spinlock on set coalesce
via-velocity: Fix races on shared interrupts
Steve Hodgson (1):
sfc: Fix SFE4002 initialisation
Sujith (1):
ath9k: Fix sequence numbers for PAE frames
Wey-Yi Guy (1):
iwlwifi: Fix to set correct ht configuration
stephen hemminger (1):
ethtool: allow non-admin user to read GRO settings.
drivers/net/benet/be_cmds.c | 1 +
drivers/net/ixgbe/ixgbe_82598.c | 22 +++++++++++++++
drivers/net/ixgbe/ixgbe_main.c | 4 +++
drivers/net/sfc/efx.c | 1 +
drivers/net/sfc/qt202x_phy.c | 2 +-
drivers/net/via-velocity.c | 41 +++++++++++++--------------
drivers/net/wireless/ath/ath9k/xmit.c | 4 +-
drivers/net/wireless/b43/b43.h | 1 +
drivers/net/wireless/b43/main.c | 13 +++++++--
drivers/net/wireless/iwlwifi/iwl-core.c | 1 +
drivers/net/wireless/iwlwifi/iwl-rx.c | 5 +++-
drivers/net/wireless/iwmc3200wifi/rx.c | 2 +-
drivers/net/wireless/rtl818x/rtl8187_dev.c | 1 +
drivers/ssb/main.c | 3 +-
net/core/ethtool.c | 1 +
net/ipv4/igmp.c | 2 +-
net/ipv4/ipcomp.c | 6 +---
net/ipv4/tcp_input.c | 6 +---
net/ipv6/ipcomp6.c | 6 +---
net/mac80211/ibss.c | 2 +-
net/mac80211/rate.c | 3 ++
net/mac80211/scan.c | 18 +++++++-----
net/xfrm/xfrm_state.c | 12 ++------
23 files changed, 93 insertions(+), 64 deletions(-)
^ permalink raw reply
* (unknown)
From: Vibhav Sreekanti @ 2010-02-17 0:10 UTC (permalink / raw)
To: netdev
subscribe
^ permalink raw reply
* Re: [net-next PATCH] via-velocity: Enable scatter/gather IO by default
From: David Miller @ 2010-02-17 0:20 UTC (permalink / raw)
To: simon.kagstrom; +Cc: netdev, krkumar2
In-Reply-To: <20100211163955.1b1dbe2c@marrow.netinsight.se>
From: Simon Kagstrom <simon.kagstrom@netinsight.net>
Date: Thu, 11 Feb 2010 16:39:55 +0100
> Reduces CPU utilization significantly with sendfile for example.
>
> Signed-off-by: Simon Kagstrom <simon.kagstrom@netinsight.net>
I think you need to have a look at this patch before making
a change like this:
commit 83c98a8cd04dd0f848574370594886ba3bf56750
Author: Dave Jones <davej@redhat.com>
Date: Tue Jul 21 09:15:49 2009 +0000
Remove unused zero-copy code from velocity NIC driver.
This code hasn't been enabled in forever.
Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/drivers/net/via-velocity.c b/drivers/net/via-velocity.c
index 42365e5..4ebe2ca 100644
--- a/drivers/net/via-velocity.c
+++ b/drivers/net/via-velocity.c
@@ -976,9 +976,6 @@ static int __devinit velocity_found1(struct pci_dev *pdev, const struct pci_devi
dev->netdev_ops = &velocity_netdev_ops;
dev->ethtool_ops = &velocity_ethtool_ops;
-#ifdef VELOCITY_ZERO_COPY_SUPPORT
- dev->features |= NETIF_F_SG;
-#endif
dev->features |= NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_FILTER |
NETIF_F_HW_VLAN_RX;
@@ -1849,11 +1846,7 @@ static void velocity_free_tx_buf(struct velocity_info *vptr, struct velocity_td_
pktlen = max_t(unsigned int, skb->len, ETH_ZLEN);
for (i = 0; i < tdinfo->nskb_dma; i++) {
-#ifdef VELOCITY_ZERO_COPY_SUPPORT
- pci_unmap_single(vptr->pdev, tdinfo->skb_dma[i], le16_to_cpu(td->tdesc1.len), PCI_DMA_TODEVICE);
-#else
pci_unmap_single(vptr->pdev, tdinfo->skb_dma[i], pktlen, PCI_DMA_TODEVICE);
-#endif
tdinfo->skb_dma[i] = 0;
}
}
@@ -2095,13 +2088,6 @@ static int velocity_xmit(struct sk_buff *skb, struct net_device *dev)
len = cpu_to_le16(pktlen);
-#ifdef VELOCITY_ZERO_COPY_SUPPORT
- if (skb_shinfo(skb)->nr_frags > 6 && __skb_linearize(skb)) {
- kfree_skb(skb);
- return NETDEV_TX_OK;
- }
-#endif
-
spin_lock_irqsave(&vptr->lock, flags);
index = vptr->tx.curr[qnum];
@@ -2111,59 +2097,18 @@ static int velocity_xmit(struct sk_buff *skb, struct net_device *dev)
td_ptr->tdesc1.TCR = TCR0_TIC;
td_ptr->td_buf[0].size &= ~TD_QUEUE;
-#ifdef VELOCITY_ZERO_COPY_SUPPORT
- if (skb_shinfo(skb)->nr_frags > 0) {
- int nfrags = skb_shinfo(skb)->nr_frags;
- tdinfo->skb = skb;
- if (nfrags > 6) {
- skb_copy_from_linear_data(skb, tdinfo->buf, skb->len);
- tdinfo->skb_dma[0] = tdinfo->buf_dma;
- td_ptr->tdesc0.len = len;
- td_ptr->tx.buf[0].pa_low = cpu_to_le32(tdinfo->skb_dma[0]);
- td_ptr->tx.buf[0].pa_high = 0;
- td_ptr->tx.buf[0].size = len; /* queue is 0 anyway */
- tdinfo->nskb_dma = 1;
- } else {
- int i = 0;
- tdinfo->nskb_dma = 0;
- tdinfo->skb_dma[i] = pci_map_single(vptr->pdev, skb->data,
- skb_headlen(skb), PCI_DMA_TODEVICE);
-
- td_ptr->tdesc0.len = len;
-
- /* FIXME: support 48bit DMA later */
- td_ptr->tx.buf[i].pa_low = cpu_to_le32(tdinfo->skb_dma);
- td_ptr->tx.buf[i].pa_high = 0;
- td_ptr->tx.buf[i].size = cpu_to_le16(skb_headlen(skb));
-
- for (i = 0; i < nfrags; i++) {
- skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
- void *addr = (void *)page_address(frag->page) + frag->page_offset;
-
- tdinfo->skb_dma[i + 1] = pci_map_single(vptr->pdev, addr, frag->size, PCI_DMA_TODEVICE);
-
- td_ptr->tx.buf[i + 1].pa_low = cpu_to_le32(tdinfo->skb_dma[i + 1]);
- td_ptr->tx.buf[i + 1].pa_high = 0;
- td_ptr->tx.buf[i + 1].size = cpu_to_le16(frag->size);
- }
- tdinfo->nskb_dma = i - 1;
- }
+ /*
+ * Map the linear network buffer into PCI space and
+ * add it to the transmit ring.
+ */
+ tdinfo->skb = skb;
+ tdinfo->skb_dma[0] = pci_map_single(vptr->pdev, skb->data, pktlen, PCI_DMA_TODEVICE);
+ td_ptr->tdesc0.len = len;
+ td_ptr->td_buf[0].pa_low = cpu_to_le32(tdinfo->skb_dma[0]);
+ td_ptr->td_buf[0].pa_high = 0;
+ td_ptr->td_buf[0].size = len;
+ tdinfo->nskb_dma = 1;
- } else
-#endif
- {
- /*
- * Map the linear network buffer into PCI space and
- * add it to the transmit ring.
- */
- tdinfo->skb = skb;
- tdinfo->skb_dma[0] = pci_map_single(vptr->pdev, skb->data, pktlen, PCI_DMA_TODEVICE);
- td_ptr->tdesc0.len = len;
- td_ptr->td_buf[0].pa_low = cpu_to_le32(tdinfo->skb_dma[0]);
- td_ptr->td_buf[0].pa_high = 0;
- td_ptr->td_buf[0].size = len;
- tdinfo->nskb_dma = 1;
- }
td_ptr->tdesc1.cmd = TCPLS_NORMAL + (tdinfo->nskb_dma + 1) * 16;
if (vptr->vlgrp && vlan_tx_tag_present(skb)) {
^ permalink raw reply related
* Re: [PATCH][KS8695] Mark network interface as running on ifconfig up
From: David Miller @ 2010-02-17 0:06 UTC (permalink / raw)
To: yegor_sub1; +Cc: linux-arm-kernel, avictor.za, netdev
In-Reply-To: <4B73E62B.3030305@visionsystems.de>
From: Yegor Yefremov <yegor_sub1@visionsystems.de>
Date: Thu, 11 Feb 2010 12:12:43 +0100
> [KS8695] Mark network interface as running on ifconfig up
>
> Without netif_carrier_on() network interface will not transmit any packets
> after ifconfig down and subsequent ifconfig up.
>
> Signed-off-by: Yegor Yefremov <yegorslists@googlemail.com>
>
> Index: linux-2.6.33-rc7/drivers/net/arm/ks8695net.c
> ===================================================================
> --- linux-2.6.33-rc7.orig/drivers/net/arm/ks8695net.c
> +++ linux-2.6.33-rc7/drivers/net/arm/ks8695net.c
> @@ -1371,6 +1371,7 @@ ks8695_open(struct net_device *ndev)
>
> napi_enable(&ksp->napi);
> netif_start_queue(ndev);
> + netif_carrier_on(ndev);
>
> return 0;
> }
>
Only ks8695_link_irq() should be setting the carrier state.
If necessary, reset the PHY on device open so that the link up
interrupt arrives and the code handling that event can set the carrier
status properly.
This is how every other driver handles this situation.
^ permalink raw reply
* Re: [PATCH v2 00/41] CAPI: Major rework, tons of bug fixes
From: David Miller @ 2010-02-17 0:02 UTC (permalink / raw)
To: jan.kiszka
Cc: isdn, linux-kernel, i4ldeveloper, isdn4linux, netdev, alan,
marcel
In-Reply-To: <cover.1265659933.git.jan.kiszka@web.de>
From: Jan Kiszka <jan.kiszka@web.de>
Date: Mon, 8 Feb 2010 21:12:04 +0100
> Here is the second take of my CAPI rework. I tried to address all
> feedback on v1, so this one comes with the following changes:
>
> o rebased over net-next
> o reworked locking of the NCCI TTY using latest and greatest tty_port
> features and kref, hopefully in the right way
> o more fine-grained patch steps for the tricky TTY rework
> o dynamic major for NCCI TTYs, ie. CAPI no longer claims 191
> o dynamic TTY minor registrations, completely obsoleting capifs
> o schedule capifs for removal
> o some small additional cleanups
>
> I kept my capifs fixes though this thing should die in the future. The
> work is done, and people may still use it due to their distro
> configuration (and the hard-wired loading by capiinit - needs fixing).
Thank you for all of your hard work.
I've applied your entire series to net-next-2.6
Thanks again!
^ permalink raw reply
* Re: [net-next-2.6 PATCH 2/2] net neigh: Decouple per interface neighbour table controls from binary sysctls
From: David Miller @ 2010-02-16 23:55 UTC (permalink / raw)
To: ebiederm; +Cc: netdev
In-Reply-To: <m1bpfskkwo.fsf@fess.ebiederm.org>
From: ebiederm@xmission.com (Eric W. Biederman)
Date: Sun, 14 Feb 2010 05:27:03 -0800
>
> Stop computing the number of neighbour table settings we have by
> counting the number of binary sysctls. This behaviour was silly
> and meant that we could not add another neighbour table setting
> without also adding another binary sysctl.
>
> Don't pass the binary sysctl path for neighour table entries
> into neigh_sysctl_register. These parameters are no longer
> used and so are just dead code.
>
> Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Applied.
I bet we can do even better and make it such that even
this NEIGH_VARS_MAX thing isn't needed.
Anyways, thanks for doing this work Eric.
^ permalink raw reply
* Re: [net-next-2.6 PATCH 1/2] net ipv4: Decouple ipv4 interface parameters from binary sysctl numbers
From: David Miller @ 2010-02-16 23:55 UTC (permalink / raw)
To: ebiederm; +Cc: netdev
In-Reply-To: <m1fx54kkyo.fsf@fess.ebiederm.org>
From: ebiederm@xmission.com (Eric W. Biederman)
Date: Sun, 14 Feb 2010 05:25:51 -0800
>
> Stop using the binary sysctl enumeartion in sysctl.h as an index into
> a per interface array. This leads to unnecessary binary sysctl number
> allocation, and a fragility in data structure and implementation
> because of unnecessary coupling.
>
> Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Applied.
^ permalink raw reply
* Re: [PATCH net-next 0/7] drivers/net: Use netif_<level> macros
From: David Miller @ 2010-02-16 23:46 UTC (permalink / raw)
To: joe
Cc: e1000-devel, netdev, bruce.w.allan, jesse.brandeburg,
linux-kernel, john.ronciak, jeffrey.t.kirsher, pcnet32,
shemminger, andy, baum
In-Reply-To: <cover.1266258417.git.joe@perches.com>
From: Joe Perches <joe@perches.com>
Date: Mon, 15 Feb 2010 10:34:19 -0800
> And a few checkpatch cleanups
All applied to net-next-2.6, thanks Joe.
------------------------------------------------------------------------------
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev
_______________________________________________
E1000-devel mailing list
E1000-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired
^ permalink raw reply
* Re: [PATCH net-next-2.6 v2 02/12] net-caif: add CAIF socket and configuration headers
From: David Miller @ 2010-02-16 23:40 UTC (permalink / raw)
To: sjur.brandeland
Cc: netdev, stefano.babic, randy.dunlap, daniel.martensson, kaber,
marcel
In-Reply-To: <1266328017-28406-3-git-send-email-sjur.brandeland@stericsson.com>
From: sjur.brandeland@stericsson.com
Date: Tue, 16 Feb 2010 14:46:47 +0100
> +#ifdef __cplusplus
> +extern "C" {
> +#endif
Sorry, we really don't make amends for C++ to use the kernel
headers in this way.
If you look at other headers defining user visible datastructures
there is really no precendence for doing what you're doing here.
Please remove these __cplusplus things and resubmit your patch
set, thank you.
^ permalink raw reply
* [PATCH]: Sparc netfilter build fix
From: David Miller @ 2010-02-16 23:30 UTC (permalink / raw)
To: kaber; +Cc: netdev, netfilter-devel
I just checked in the following patch to fix the build
after this morning's netfilter merge into net-next-2.6
Just FYI...
sparc64: Kill bogus ip_tables.h include.
Fixes the following build failure:
CC arch/sparc/kernel/sys_sparc32.o
In file included from include/linux/netfilter_ipv4/ip_tables.h:28,
from arch/sparc/kernel/sys_sparc32.c:46:
include/linux/netfilter/x_tables.h:525: error: expected declaration specifiers or ‘...’ before ‘nf_hookfn’
Signed-off-by: David S. Miller <davem@davemloft.net>
---
arch/sparc/kernel/sys_sparc32.c | 1 -
1 files changed, 0 insertions(+), 1 deletions(-)
diff --git a/arch/sparc/kernel/sys_sparc32.c b/arch/sparc/kernel/sys_sparc32.c
index dc0ac19..daded3b 100644
--- a/arch/sparc/kernel/sys_sparc32.c
+++ b/arch/sparc/kernel/sys_sparc32.c
@@ -43,7 +43,6 @@
#include <linux/security.h>
#include <linux/compat.h>
#include <linux/vfs.h>
-#include <linux/netfilter_ipv4/ip_tables.h>
#include <linux/ptrace.h>
#include <asm/types.h>
--
1.6.6.1
^ permalink raw reply related
* Re: [PATCH -next 6/6] bnx2: Update firmwares and update version to 2.0.8.
From: David Miller @ 2010-02-16 23:20 UTC (permalink / raw)
To: mchan; +Cc: netdev
In-Reply-To: <1266298932-28129-6-git-send-email-mchan@broadcom.com>
From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 15 Feb 2010 21:42:12 -0800
> - Increase FTQ depth to 256 to ehnabce performance.
> - Fix RV2P context corruption on 5709 when flow control is enabled.
>
> Signed-off-by: Michael Chan <mchan@broadcom.com>
Also applied, thanks!
^ permalink raw reply
* Re: [PATCH -next 5/6] bnx2: Fix bug when saving statistics.
From: David Miller @ 2010-02-16 23:20 UTC (permalink / raw)
To: mchan; +Cc: netdev
In-Reply-To: <1266298932-28129-5-git-send-email-mchan@broadcom.com>
From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 15 Feb 2010 21:42:11 -0800
> From: Patrick Rabau <pr2345@gmail.com>
>
> This fixes the problem of dropping the carry when adding 2 32-bit values.
> Switch to use array indexing for better readability.
>
> Reported by and fix provided by Patrick Rabau.
>
> Signed-off-by: Michael Chan <mchan@broadcom.com>
> Signed-off-by: Benjamin Li <benli@broadcom.com>
Applied.
^ permalink raw reply
* Re: [PATCH -next 4/6] bnx2: Allow user-specified multiple advertisement speed values.
From: David Miller @ 2010-02-16 23:20 UTC (permalink / raw)
To: mchan; +Cc: netdev
In-Reply-To: <1266298932-28129-4-git-send-email-mchan@broadcom.com>
From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 15 Feb 2010 21:42:10 -0800
> Remove unnecessary code that works around older versions of ethtool
> that can pass down invalid advertisement speed values. This old
> code prevents the user from specifying multiple advertisement values.
> The new code uses simple masking to mask out invalid advertisment bits.
>
> Reported-by: Vlad Yasevich <vladislav.yasevich@hp.com>
> Signed-off-by: Michael Chan <mchan@broadcom.com>
> Signed-off-by: Benjamin Li <benli@broadcom.com>
Applied.
^ permalink raw reply
* Re: [PATCH -next 3/6] bnx2: Adjust flow control water marks.
From: David Miller @ 2010-02-16 23:20 UTC (permalink / raw)
To: mchan; +Cc: netdev
In-Reply-To: <1266298932-28129-3-git-send-email-mchan@broadcom.com>
From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 15 Feb 2010 21:42:09 -0800
> The current water marks are too high and can cause unnecessary flow
> control frames.
>
> Signed-off-by: Michael Chan <mchan@broadcom.com>
> Signed-off-by: Benjamin Li <benli@broadcom.com>
Applied.
^ permalink raw reply
* Re: [PATCH -next 2/6] bnx2: Need to call cnic_setup_cnic_irq_info() after MTU change.
From: David Miller @ 2010-02-16 23:20 UTC (permalink / raw)
To: mchan; +Cc: netdev
In-Reply-To: <1266298932-28129-2-git-send-email-mchan@broadcom.com>
From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 15 Feb 2010 21:42:08 -0800
> New status blocks are allocated during MTU change so we need to
> update this information for the cnic driver.
>
> Signed-off-by: Michael Chan <mchan@broadcom.com>
> Signed-off-by: Benjamin Li <benli@broadcom.com>
Applied.
^ permalink raw reply
* Re: [PATCH -next 1/6] bnx2: Check BNX2_FLAG_USING_MSIX flag when setting up MSIX.
From: David Miller @ 2010-02-16 23:20 UTC (permalink / raw)
To: mchan; +Cc: netdev
In-Reply-To: <1266298932-28129-1-git-send-email-mchan@broadcom.com>
From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 15 Feb 2010 21:42:07 -0800
> Checking the flag is more correct than checking bp->irq_nvecs. By
> accident it is not a problem because we always have more than 1
> vectors when using MSIX mode.
>
> Signed-off-by: Michael Chan <mchan@broadcom.com>
> Signed-off-by: Benjamin Li <benli@broadcom.com>
Applied.
^ permalink raw reply
* Re: [PATCH] atl1c: Add support for Atheros AR8152 and AR8152
From: David Miller @ 2010-02-16 23:16 UTC (permalink / raw)
To: lrodriguez
Cc: netdev, linux-kernel, espy, ken.wu, graham.richards, scott.tan,
mcgrof, chris.snook
In-Reply-To: <1265934812-14157-1-git-send-email-lrodriguez@atheros.com>
From: "Luis R. Rodriguez" <lrodriguez@atheros.com>
Date: Thu, 11 Feb 2010 19:33:32 -0500
> AR8151 is a Gigabit Ethernet device. AR8152 devices are
> Fast Ethernet devices, there are two revisions, a 1.0
> and a 2.0 revision.
>
> This has been tested against these devices:
>
> Driver Model-name vendor:device Type
> atl1c AR8131 1969:1063 Gigabit Ethernet
> atl1c AR8132 1969:1062 Fast Ethernet
> atl1c AR8151(v1.0) 1969:1073 Gigabit Ethernet
> atl1c AR8152(v1.1) 1969:2060 Fast Ethernet
>
> This device has no hardware available yet so it goes untested,
> but it should work:
>
> atl1c AR8152(v2.0) 1969:2062 Fast Ethernet
>
> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Applied to net-next-2.6, thanks.
^ 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