* ipoib crash when booting with ipautoconfig
From: Yinghai Lu @ 2011-07-25 21:39 UTC (permalink / raw)
To: David Miller, NetDev
[ 88.946112] BUG: unable to handle kernel NULL pointer dereference
at 0000000000000040
[ 88.954153] IP: [<ffffffff81a61ee1>] ipoib_start_xmit+0x5c/0x31a
[ 88.960196] PGD 0
[ 88.962235] Oops: 0000 [#1] SMP
[ 88.965500] CPU 0
[ 88.967343] Modules linked in:
[ 88.970601]
[ 88.972094] Pid: 1, comm: swapper Not tainted
3.0.0-tip-yh-04912-gf572f66-dirty #1148
[ 88.984053] RIP: 0010:[<ffffffff81a61ee1>] [<ffffffff81a61ee1>]
ipoib_start_xmit+0x5c/0x31a
[ 88.992497] RSP: 0018:ffff88046dd37cb0 EFLAGS: 00010246
[ 88.997805] RAX: 0000000000000000 RBX: ffff88045a5af380 RCX: 0000000060014c20
[ 89.004933] RDX: 0000000000000000 RSI: ffff880464768000 RDI: ffff88045a5af380
[ 89.012060] RBP: ffff88046dd37cf0 R08: 0000000000001d1c R09: 000000000039e295
[ 89.019188] R10: 0000000000000000 R11: ffff88046dd38708 R12: ffff880464768000
[ 89.026317] R13: 0000000000000000 R14: ffff8804647689c0 R15: ffff880464768000
[ 89.033447] FS: 0000000000000000(0000) GS:ffff88047de00000(0000)
knlGS:0000000000000000
[ 89.041524] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 89.047268] CR2: 0000000000000040 CR3: 00000000023fc000 CR4: 00000000000406f0
[ 89.054394] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 89.061523] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[ 89.068652] Process swapper (pid: 1, threadinfo ffff88046dd36000,
task ffff88046dd38000)
[ 89.076729] Stack:
[ 89.078741] 0000000000000000 ffff88046dd38000 0000000000000246
ffff88045a5af380
[ 89.086189] 0000000060014c23 0000000000000000 0000000000000258
ffff880464768000
[ 89.093645] ffff88046dd37d50 ffffffff81b15a31 ffff880670687e98
ffff880670687e80
[ 89.101093] Call Trace:
[ 89.103549] [<ffffffff81b15a31>] dev_hard_start_xmit+0x2a7/0x435
[ 89.109658] [<ffffffff81b2984d>] sch_direct_xmit+0x74/0x207
[ 89.115317] [<ffffffff81b15db9>] dev_queue_xmit+0x1fa/0x4aa
[ 89.120981] [<ffffffff81b15bbf>] ? dev_hard_start_xmit+0x435/0x435
[ 89.127267] [<ffffffff81b0abd2>] ? __alloc_skb+0x83/0x141
[ 89.132774] [<ffffffff82754586>] ic_bootp_send_if+0x2b4/0x2d2
[ 89.138597] [<ffffffff82754709>] ic_dynamic+0x165/0x315
[ 89.143913] [<ffffffff8275508e>] ip_auto_config+0xf2/0x2c0
[ 89.149484] [<ffffffff82754f9c>] ? root_nfs_parse_addr+0xb5/0xb5
[ 89.155604] [<ffffffff810002cf>] do_one_initcall+0x57/0x134
[ 89.161278] [<ffffffff82708f85>] kernel_init+0x11f/0x1a3
[ 89.166699] [<ffffffff81c471d4>] kernel_thread_helper+0x4/0x10
[ 89.172640] [<ffffffff81c3e61d>] ? retint_restore_args+0xe/0xe
[ 89.178550] [<ffffffff82708e66>] ? start_kernel+0x3c9/0x3c9
[ 89.184212] [<ffffffff81c471d0>] ? gs_change+0xb/0xb
[ 89.189261] Code: 1d 48 c7 c7 60 11 42 82 e8 e0 d7 64 ff 85 c0 75
0d e8 91 5e 63 ff 85 c0 0f 84 89 02 00 00 48 8b 43 58 48 89 c2 48 83
e2 fe a8 01 <4c> 8b 7a 40 74 26 e8 3c 5e 63 ff 85 c0 74 1d 48 c7 c7 60
11 42
[ 89.209206] RIP [<ffffffff81a61ee1>] ipoib_start_xmit+0x5c/0x31a
[ 89.215311] RSP <ffff88046dd37cb0>
[ 89.218792] CR2: 0000000000000040
[ 89.222287] ---[ end trace 88d8b543880f18a5 ]---
^ permalink raw reply
* Re: [PATCH] net/smsc911x: add device tree probe support
From: Grant Likely @ 2011-07-25 21:37 UTC (permalink / raw)
To: Shawn Guo
Cc: netdev, devicetree-discuss, linux-arm-kernel, patches,
Steve Glendinning, David S. Miller
In-Reply-To: <1311587040-8988-1-git-send-email-shawn.guo@linaro.org>
On Mon, Jul 25, 2011 at 05:44:00PM +0800, Shawn Guo wrote:
> It adds device tree probe support for smsc911x driver.
>
> Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
> Cc: Grant Likely <grant.likely@secretlab.ca>
> Cc: Steve Glendinning <steve.glendinning@smsc.com>
> Cc: David S. Miller <davem@davemloft.net>
> ---
> Documentation/devicetree/bindings/net/smsc.txt | 34 +++++++
> drivers/net/smsc911x.c | 123 +++++++++++++++++++-----
> 2 files changed, 132 insertions(+), 25 deletions(-)
> create mode 100644 Documentation/devicetree/bindings/net/smsc.txt
>
> diff --git a/Documentation/devicetree/bindings/net/smsc.txt b/Documentation/devicetree/bindings/net/smsc.txt
> new file mode 100644
> index 0000000..1920695
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/net/smsc.txt
> @@ -0,0 +1,34 @@
> +* Smart Mixed-Signal Connectivity (SMSC) LAN Controller
> +
> +Required properties:
> +- compatible : Should be "smsc,lan<model>""smsc,lan"
Drop "smsc,lan". That's far too generic.
> +- reg : Address and length of the io space for SMSC LAN
> +- smsc-int-gpios : Should specify the GPIO for SMSC LAN interrupt line
This looks broken. Shouldn't this be specified as a normal
"interrupts" property?
> +- phy-mode : String, operation mode of the PHY interface.
> + Supported values are: "mii", "gmii", "sgmii", "tbi", "rmii",
> + "rgmii", "rgmii-id", "rgmii-rxid", "rgmii-txid", "rtbi", "smii".
> +
> +Optional properties:
> +- smsc,irq-active-high : Indicates the IRQ polarity is active-low
> +- smsc,irq-push-pull : Indicates the IRQ type is push-pull
> +- smsc,register-needs-shift : Indicates the register access needs shift
> +- smsc,access-in-32bit : Indicates the access to controller is in 32-bit
> + mode
Currently, reg-io-width and reg-shift are being used to manipulate
register access on ns16550 serial ports. The same thing can be used
here. See bindings/tty/serial/of-serial.txt
> +- smsc,force-internal-phy : Forces SMSC LAN controller to use
> + internal PHY
> +- smsc,force-external-phy : Forces SMSC LAN controller to use
> + external PHY
I would expect using an external phy would also expect a phy-device
property to connect to the phy node.
> +- smsc,save-mac-address : Indicates that mac address needs to be saved
> + before resetting the controller
> +- local-mac-address : 6 bytes, mac address
> +
> +Examples:
> +
> +lan9220@f4000000 {
> + compatible = "smsc,lan9220", "smsc,lan";
> + reg = <0xf4000000 0x2000000>;
> + phy-mode = "mii";
> + smsc-int-gpios = <&gpio1 31 0>; /* GPIO2_31 */
> + smsc,irq-push-pull;
> + smsc,access-in-32bit;
> +};
> diff --git a/drivers/net/smsc911x.c b/drivers/net/smsc911x.c
> index b9016a3..0097048 100644
> --- a/drivers/net/smsc911x.c
> +++ b/drivers/net/smsc911x.c
> @@ -53,6 +53,10 @@
> #include <linux/phy.h>
> #include <linux/smsc911x.h>
> #include <linux/device.h>
> +#include <linux/of.h>
> +#include <linux/of_device.h>
> +#include <linux/of_gpio.h>
> +#include <linux/of_net.h>
> #include "smsc911x.h"
>
> #define SMSC_CHIPNAME "smsc911x"
> @@ -2095,25 +2099,67 @@ static const struct smsc911x_ops shifted_smsc911x_ops = {
> .tx_writefifo = smsc911x_tx_writefifo_shift,
> };
>
> +#ifdef CONFIG_OF
> +static int __devinit smsc911x_probe_config_dt(
> + struct smsc911x_platform_config *config,
> + struct device_node *np)
> +{
> + const char *mac;
> +
> + if (!np)
> + return -ENODEV;
> +
> + config->phy_interface = of_get_phy_mode(np);
> +
> + mac = of_get_mac_address(np);
> + if (mac)
> + memcpy(config->mac, mac, ETH_ALEN);
> +
> + if (of_get_property(np, "smsc,irq-active-high", NULL))
> + config->irq_polarity = SMSC911X_IRQ_POLARITY_ACTIVE_HIGH;
> +
> + if (of_get_property(np, "smsc,irq-push-pull", NULL))
> + config->irq_type = SMSC911X_IRQ_TYPE_PUSH_PULL;
> +
> + if (of_get_property(np, "smsc,register-needs-shift", NULL))
> + config->shift = 1;
> +
> + if (of_get_property(np, "smsc,access-in-32bit", NULL))
> + config->flags |= SMSC911X_USE_32BIT;
> +
> + if (of_get_property(np, "smsc,force-internal-phy", NULL))
> + config->flags |= SMSC911X_FORCE_INTERNAL_PHY;
> +
> + if (of_get_property(np, "smsc,force-external-phy", NULL))
> + config->flags |= SMSC911X_FORCE_EXTERNAL_PHY;
> +
> + if (of_get_property(np, "smsc,save-mac-address", NULL))
> + config->flags |= SMSC911X_SAVE_MAC_ADDRESS;
> +
> + return 0;
> +}
> +#else
> +static inline int smsc911x_probe_config_dt(
> + struct smsc911x_platform_config *config,
> + struct device_node *np)
> +{
> + return -ENODEV;
> +}
> +#endif /* CONFIG_OF */
> +
> static int __devinit smsc911x_drv_probe(struct platform_device *pdev)
> {
> + struct device_node *np = pdev->dev.of_node;
> struct net_device *dev;
> struct smsc911x_data *pdata;
> struct smsc911x_platform_config *config = pdev->dev.platform_data;
> struct resource *res, *irq_res;
> unsigned int intcfg = 0;
> - int res_size, irq_flags;
> - int retval;
> + int irq_gpio, res_size, irq_flags = 0;
> + int retval = 0;
>
> pr_info("Driver version %s\n", SMSC_DRV_VERSION);
>
> - /* platform data specifies irq & dynamic bus configuration */
> - if (!pdev->dev.platform_data) {
> - pr_warn("platform_data not provided\n");
> - retval = -ENODEV;
> - goto out_0;
> - }
> -
> res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
> "smsc911x-memory");
> if (!res)
> @@ -2125,13 +2171,6 @@ static int __devinit smsc911x_drv_probe(struct platform_device *pdev)
> }
> res_size = resource_size(res);
>
> - irq_res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
> - if (!irq_res) {
> - pr_warn("Could not allocate irq resource\n");
> - retval = -ENODEV;
> - goto out_0;
> - }
> -
This should still work for the device-tree situation. Why remove it?
> if (!request_mem_region(res->start, res_size, SMSC_CHIPNAME)) {
> retval = -EBUSY;
> goto out_0;
> @@ -2148,26 +2187,53 @@ static int __devinit smsc911x_drv_probe(struct platform_device *pdev)
>
> pdata = netdev_priv(dev);
>
> - dev->irq = irq_res->start;
> - irq_flags = irq_res->flags & IRQF_TRIGGER_MASK;
> - pdata->ioaddr = ioremap_nocache(res->start, res_size);
> -
> - /* copy config parameters across to pdata */
> - memcpy(&pdata->config, config, sizeof(pdata->config));
> + if (np) {
> + irq_gpio = of_get_named_gpio(np, "smsc-int-gpios", 0);
> + retval = gpio_request_one(irq_gpio, GPIOF_IN, "smsc-int-gpio");
> + if (!retval)
> + dev->irq = gpio_to_irq(irq_gpio);
Yeah, that's definitely the wrong way to handle this. If the
device it wired to a gpio controller, then the gpio controller also
need to be an interrupt controller to ensure that it can map interrupt
numbers.
> + } else {
> + irq_res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
> + if (irq_res) {
> + dev->irq = irq_res->start;
> + irq_flags = irq_res->flags & IRQF_TRIGGER_MASK;
> + } else {
> + retval = -ENODEV;
> + }
> + }
>
> - pdata->dev = dev;
> - pdata->msg_enable = ((1 << debug) - 1);
> + if (retval) {
> + SMSC_WARN(pdata, probe, "Error smsc911x irq not found");
> + retval = -EINVAL;
> + goto out_free_netdev_2;
> + }
>
> + pdata->ioaddr = ioremap_nocache(res->start, res_size);
> if (pdata->ioaddr == NULL) {
> SMSC_WARN(pdata, probe, "Error smsc911x base address invalid");
> retval = -ENOMEM;
> goto out_free_netdev_2;
> }
>
> + pdata->dev = dev;
> + pdata->msg_enable = ((1 << debug) - 1);
> +
> + retval = smsc911x_probe_config_dt(&pdata->config, np);
> + if (retval && config) {
> + /* copy config parameters across to pdata */
> + memcpy(&pdata->config, config, sizeof(pdata->config));
> + retval = 0;
> + }
> +
> + if (retval) {
> + SMSC_WARN(pdata, probe, "Error smsc911x config not found");
> + goto out_unmap_io_3;
> + }
> +
> /* assume standard, non-shifted, access to HW registers */
> pdata->ops = &standard_smsc911x_ops;
> /* apply the right access if shifting is needed */
> - if (config->shift)
> + if (pdata->config.shift)
> pdata->ops = &shifted_smsc911x_ops;
>
> retval = smsc911x_init(dev);
> @@ -2314,6 +2380,12 @@ static const struct dev_pm_ops smsc911x_pm_ops = {
> #define SMSC911X_PM_OPS NULL
> #endif
>
> +static const struct of_device_id smsc_dt_ids[] = {
> + { .compatible = "smsc,lan", },
As mentioned above, "smsc,lan" is far too generic.
> + { /* sentinel */ }
> +};
> +MODULE_DEVICE_TABLE(of, smsc_dt_ids);
> +
> static struct platform_driver smsc911x_driver = {
> .probe = smsc911x_drv_probe,
> .remove = __devexit_p(smsc911x_drv_remove),
> @@ -2321,6 +2393,7 @@ static struct platform_driver smsc911x_driver = {
> .name = SMSC_CHIPNAME,
> .owner = THIS_MODULE,
> .pm = SMSC911X_PM_OPS,
> + .of_match_table = smsc_dt_ids,
> },
> };
>
> --
> 1.7.4.1
>
>
^ permalink raw reply
* Re: [PATCH 0/7] More sane neigh infrastructure
From: David Miller @ 2011-07-25 21:10 UTC (permalink / raw)
To: roland; +Cc: linux-rdma, netdev
In-Reply-To: <CAG4TOxOv1WsgfRXxrBWWq-+9ee3apEE0B_Onn+JB=tEEuOhhxg@mail.gmail.com>
From: Roland Dreier <roland@kernel.org>
Date: Mon, 25 Jul 2011 09:34:16 -0700
> So I don't think we can use custom neigh_ops with a new solict method
> the way clip does -- we actually want to let the normal stack do ARP
> or ND, but then extend the process by another message/response step.
> I'm sure this is possible within your scheme but I'm not sure I
> understand what the "right" way is.
So call the normal ARP neigh solicit stuff in your neigh ops, and do
your local stuff there as well.
See if you can make it work.
^ permalink raw reply
* info
From: Western Union® @ 2011-07-25 20:50 UTC (permalink / raw)
You have a transfer of 750,000.00 pounds from Western Union® For more information
(Contact This Office Email: western.uniondpt@w.cn) Mr.Frank Ban
^ permalink raw reply
* Re: [PATCH 1/2] net: add IFF_SKB_TX_SHARED flag to priv_flags
From: Eric Dumazet @ 2011-07-25 20:11 UTC (permalink / raw)
To: Neil Horman; +Cc: netdev, Robert Olsson, Alexey Dobriyan, David S. Miller
In-Reply-To: <1311623120-26839-2-git-send-email-nhorman@tuxdriver.com>
Le lundi 25 juillet 2011 à 15:45 -0400, Neil Horman a écrit :
> Pktgen attempts to transmit shared skbs to net devices, which can't be used by
> some drivers as they keep state information in skbs. This patch adds a flag
> marking drivers as being able to handle shared skbs in their tx path. Drivers
> are defaulted to being unable to do so, but calling ether_setup enables this
> flag, as 90% of the drivers calling ether_setup touch real hardware and can
> handle shared skbs. A subsequent patch will audit drivers to ensure that the
> flag is set properly
>
> Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
> Reported-by: Jiri Pirko <jpirko@redhat.com>
> CC: Robert Olsson <robert.olsson@its.uu.se>
> CC: Eric Dumazet <eric.dumazet@gmail.com>
> CC: Alexey Dobriyan <adobriyan@gmail.com>
> CC: David S. Miller <davem@davemloft.net>
> ---
> include/linux/if.h | 2 ++
> net/core/pktgen.c | 8 +++++---
> net/ethernet/eth.c | 1 +
> 3 files changed, 8 insertions(+), 3 deletions(-)
>
> diff --git a/include/linux/if.h b/include/linux/if.h
> index 3bc63e6..03489ca 100644
> --- a/include/linux/if.h
> +++ b/include/linux/if.h
> @@ -76,6 +76,8 @@
> #define IFF_BRIDGE_PORT 0x4000 /* device used as bridge port */
> #define IFF_OVS_DATAPATH 0x8000 /* device used as Open vSwitch
> * datapath port */
> +#define IFF_TX_SKB_SHARING 0x10000 /* The interface supports sharing
> + * skbs on transmit */
>
> #define IF_GET_IFACE 0x0001 /* for querying only */
> #define IF_GET_PROTO 0x0002
> diff --git a/net/core/pktgen.c b/net/core/pktgen.c
> index f76079c..53f3f15 100644
> --- a/net/core/pktgen.c
> +++ b/net/core/pktgen.c
> @@ -1070,7 +1070,9 @@ static ssize_t pktgen_if_write(struct file *file,
> len = num_arg(&user_buffer[i], 10, &value);
> if (len < 0)
> return len;
> -
> + if ((len > 0) &&
if ((value > 0) ...
> + (!(pkt_dev->odev->priv_flags & IFF_TX_SKB_SHARING)))
> + return -ENOTSUPP;
> i += len;
> pkt_dev->clone_skb = value;
>
^ permalink raw reply
* [PATCH] cxgb3i: ref count cdev access to prevent modification while in use
From: Neil Horman @ 2011-07-25 19:56 UTC (permalink / raw)
To: netdev; +Cc: Neil Horman, Divy Le Ray, Steve Wise, David S. Miller
This oops was reported recently:
d:mon> e
cpu 0xd: Vector: 300 (Data Access) at [c0000000fd4c7120]
pc: d00000000076f194: .t3_l2t_get+0x44/0x524 [cxgb3]
lr: d000000000b02108: .init_act_open+0x150/0x3d4 [cxgb3i]
sp: c0000000fd4c73a0
msr: 8000000000009032
dar: 0
dsisr: 40000000
current = 0xc0000000fd640d40
paca = 0xc00000000054ff80
pid = 5085, comm = iscsid
d:mon> t
[c0000000fd4c7450] d000000000b02108 .init_act_open+0x150/0x3d4 [cxgb3i]
[c0000000fd4c7500] d000000000e45378 .cxgbi_ep_connect+0x784/0x8e8 [libcxgbi]
[c0000000fd4c7650] d000000000db33f0 .iscsi_if_rx+0x71c/0xb18
[scsi_transport_iscsi2]
[c0000000fd4c7740] c000000000370c9c .netlink_data_ready+0x40/0xa4
[c0000000fd4c77c0] c00000000036f010 .netlink_sendskb+0x4c/0x9c
[c0000000fd4c7850] c000000000370c18 .netlink_sendmsg+0x358/0x39c
[c0000000fd4c7950] c00000000033be24 .sock_sendmsg+0x114/0x1b8
[c0000000fd4c7b50] c00000000033d208 .sys_sendmsg+0x218/0x2ac
[c0000000fd4c7d70] c00000000033f55c .sys_socketcall+0x228/0x27c
[c0000000fd4c7e30] c0000000000086a4 syscall_exit+0x0/0x40
--- Exception: c01 (System Call) at 00000080da560cfc
The root cause was an EEH error, which sent us down the offload_close path in
the cxgb3 driver, which in turn sets cdev->lldev to NULL, without regard for
upper layer driver (like the cxgbi drivers) which might have execution contexts
in the middle of its use. The result is the oops above, when t3_l2t_get attempts
to dereference cdev->lldev right after the EEH error handler sets it to NULL.
The fix is to reference count the cdev structure. When an EEH error occurs, the
shutdown path:
t3_adapter_error->offload_close->cxgb3i_remove_clients->cxgb3i_dev_close
will now block until such time as the cdev pointer has a use count of zero.
This coupled with the fact that lookups will now skip finding any registered
cdev's in cxgbi_device_find_by_[lldev|netdev] with the CXGBI_FLAG_ADAPTER_RESET
bit set ensures that on an EEH, the setting of lldev to NULL in offload_close
will only happen after there are no longer any active users of the data
structure.
This has been tested by the reporter and shown to fix the reproted oops
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
CC: Divy Le Ray <divy@chelsio.com>
CC: Steve Wise <swise@chelsio.com>
CC: "David S. Miller" <davem@davemloft.net>
---
drivers/scsi/cxgbi/cxgb3i/cxgb3i.c | 8 +++++++-
drivers/scsi/cxgbi/libcxgbi.c | 9 +++++++++
drivers/scsi/cxgbi/libcxgbi.h | 3 +++
3 files changed, 19 insertions(+), 1 deletions(-)
diff --git a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c
index abc7b12..7d752cd 100644
--- a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c
+++ b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c
@@ -1301,9 +1301,13 @@ static void cxgb3i_dev_close(struct t3cdev *t3dev)
if (!cdev || cdev->flags & CXGBI_FLAG_ADAPTER_RESET) {
pr_info("0x%p close, f 0x%x.\n", cdev, cdev ? cdev->flags : 0);
+ if (cdev)
+ cdev_put(cdev);
+ while (cdev && atomic_read(&cdev->use_count) != 0)
+ msleep(1);
return;
}
-
+ cdev_put(cdev);
cxgbi_device_unregister(cdev);
}
@@ -1318,6 +1322,7 @@ static void cxgb3i_dev_open(struct t3cdev *t3dev)
int i, err;
if (cdev) {
+ cdev_put(cdev);
pr_info("0x%p, updating.\n", cdev);
return;
}
@@ -1390,6 +1395,7 @@ static void cxgb3i_dev_event_handler(struct t3cdev *t3dev, u32 event, u32 port)
cdev->flags &= ~CXGBI_FLAG_ADAPTER_RESET;
break;
}
+ cdev_put(cdev);
}
/**
diff --git a/drivers/scsi/cxgbi/libcxgbi.c b/drivers/scsi/cxgbi/libcxgbi.c
index 77ac217..eb5625d 100644
--- a/drivers/scsi/cxgbi/libcxgbi.c
+++ b/drivers/scsi/cxgbi/libcxgbi.c
@@ -181,6 +181,9 @@ struct cxgbi_device *cxgbi_device_find_by_lldev(void *lldev)
mutex_lock(&cdev_mutex);
list_for_each_entry_safe(cdev, tmp, &cdev_list, list_head) {
if (cdev->lldev == lldev) {
+ if (cdev->flags & CXGBI_FLAG_ADAPTER_RESET)
+ continue;
+ cdev_hold(cdev);
mutex_unlock(&cdev_mutex);
return cdev;
}
@@ -210,7 +213,10 @@ static struct cxgbi_device *cxgbi_device_find_by_netdev(struct net_device *ndev,
list_for_each_entry_safe(cdev, tmp, &cdev_list, list_head) {
for (i = 0; i < cdev->nports; i++) {
if (ndev == cdev->ports[i]) {
+ if (cdev->flags & CXGBI_FLAG_ADAPTER_RESET)
+ continue;
cdev->hbas[i]->vdev = vdev;
+ cdev_hold(cdev);
mutex_unlock(&cdev_mutex);
if (port)
*port = i;
@@ -542,6 +548,8 @@ rel_rt:
if (csk)
cxgbi_sock_closed(csk);
err_out:
+ if (cdev)
+ cdev_put(cdev);
return ERR_PTR(err);
}
@@ -2491,6 +2499,7 @@ struct iscsi_endpoint *cxgbi_ep_connect(struct Scsi_Host *shost,
return ep;
release_conn:
+ cdev_put(&csk->cdev);
cxgbi_sock_put(csk);
cxgbi_sock_closed(csk);
err_out:
diff --git a/drivers/scsi/cxgbi/libcxgbi.h b/drivers/scsi/cxgbi/libcxgbi.h
index 9267844..aad1749 100644
--- a/drivers/scsi/cxgbi/libcxgbi.h
+++ b/drivers/scsi/cxgbi/libcxgbi.h
@@ -514,6 +514,7 @@ struct cxgbi_device {
unsigned int flags;
struct net_device **ports;
void *lldev;
+ atomic_t use_count;
struct cxgbi_hba **hbas;
const unsigned short *mtus;
unsigned char nmtus;
@@ -557,6 +558,8 @@ struct cxgbi_device {
void *dd_data;
};
#define cxgbi_cdev_priv(cdev) ((cdev)->dd_data)
+#define cdev_hold(x) do {atomic_inc(&x->use_count);} while(0)
+#define cdev_put(x) do {atomic_dec(&x->use_count);} while(0)
struct cxgbi_conn {
struct cxgbi_endpoint *cep;
--
1.7.3.4
^ permalink raw reply related
* Confirm Your E-Mail Details
From: Information Technology @ 2011-07-25 19:06 UTC (permalink / raw)
To: netdev
Confirm Your E-Mail Details
Update Your E-Mail Details Account
Dear Email User
This message is from Information Technology Services of This EMAIL to all our Staff. We are currently upgrading our database and e-mail center and this is our final notification to you.we have sent several messages to you without response.
We are deleting all unused Mail account to create space for new accounts. In order not to be suspended, you will have to update your account by providing the information listed below:
Confirm Your E-Mail Details..
Email.......................
User name: ..................
Password:..............
Re Confirm Password:.............
If you fail to confirm your continuous usage of our services by confirming your email password now, your account will be disable and you will not be able to access your email.
You should immediately reply this email: upgrading2010@mail.md and enter your password in the above password column.
Thanks for your understanding.
Regard,
IT Services
^ permalink raw reply
* Realtek 8139 Flow Control?
From: Sven Anders @ 2011-07-25 19:42 UTC (permalink / raw)
To: netdev; +Cc: becker, jgarzik
[-- Attachment #1: Type: text/plain, Size: 946 bytes --]
Hello!
I have just a short question.
We have appliances with Realtek 8139C/8139C+ (rev 10) chipsets (with the
PCI IDs: 10ec:8139). We are using the 8139too driver (version: 0.9.28)
According the datasheet the chipset supports Flow Control (IEEE 802.3x).
I want to know, if the support for enabling flow control is missing in
the driver by purpose or is only not implemented due to missing time?
Regards
Sven
--
Sven Anders <anders@anduras.de> () UTF-8 Ribbon Campaign
/\ Support plain text e-mail
ANDURAS intranet security AG
Messestraße 3 - 94036 Passau - Germany
Web: www.anduras.de - Tel: +49 (0)851-4 90 50-0 - Fax: +49 (0)851-4 90 50-55
Rechtsform: Aktiengesellschaft - Sitz: Passau - Amtsgericht: Passau HRB 6032
Mitglieder des Vorstands: Dipl.-Inf. Sven Anders, Dipl.-Inf. Marcus Junker
Vorsitzender des Aufsichtsrats: RA Mark Peters
[-- Attachment #2: anders.vcf --]
[-- Type: text/x-vcard, Size: 339 bytes --]
begin:vcard
fn:Sven Anders
n:Anders;Sven
org:ANDURAS AG;Research and Development
adr;quoted-printable:;;Messestra=C3=9Fe 3;Passau;Bavaria;94036;Germany
email;internet:anders@anduras.de
title:Dipl. Inf.
tel;work:++49 (0)851 / 490 50 -0
tel;fax:++49 (0)851 / 590 50 - 55
x-mozilla-html:FALSE
url:http://www.anduras.de
version:2.1
end:vcard
^ permalink raw reply
* [PATCH 2/2] net: Audit drivers to identify those needing IFF_TX_SKB_SHARING cleared
From: Neil Horman @ 2011-07-25 19:45 UTC (permalink / raw)
To: netdev
Cc: Neil Horman, Karsten Keil, David S. Miller, Jay Vosburgh,
Andy Gospodarek, Patrick McHardy, Krzysztof Halasa,
John W. Linville, Greg Kroah-Hartman, Marcel Holtmann,
Johannes Berg
In-Reply-To: <1311623120-26839-1-git-send-email-nhorman@tuxdriver.com>
After the last patch, We are left in a state in which only drivers calling
ether_setup have IFF_TX_SKB_SHARING set (we assume that drivers touching real
hardware call ether_setup for their net_devices and don't hold any state in
their skbs. There are a handful of drivers that violate this assumption of
course, and need to be fixed up. This patch identifies those drivers, and marks
them as not being able to support the safe transmission of skbs by clearning the
IFF_TX_SKB_SHARING flag in priv_flags
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
CC: Karsten Keil <isdn@linux-pingi.de>
CC: "David S. Miller" <davem@davemloft.net>
CC: Jay Vosburgh <fubar@us.ibm.com>
CC: Andy Gospodarek <andy@greyhouse.net>
CC: Patrick McHardy <kaber@trash.net>
CC: Krzysztof Halasa <khc@pm.waw.pl>
CC: "John W. Linville" <linville@tuxdriver.com>
CC: Greg Kroah-Hartman <gregkh@suse.de>
CC: Marcel Holtmann <marcel@holtmann.org>
CC: Johannes Berg <johannes@sipsolutions.net>
---
drivers/isdn/i4l/isdn_net.c | 3 +++
drivers/net/bonding/bond_main.c | 6 ++++--
drivers/net/ifb.c | 2 +-
drivers/net/macvlan.c | 2 +-
drivers/net/tun.c | 1 +
drivers/net/veth.c | 2 ++
drivers/net/wan/hdlc_fr.c | 5 +++--
drivers/net/wireless/airo.c | 1 +
drivers/net/wireless/hostap/hostap_main.c | 1 +
drivers/staging/ath6kl/os/linux/ar6000_drv.c | 1 +
net/8021q/vlan_dev.c | 2 +-
net/bluetooth/bnep/netdev.c | 1 +
net/l2tp/l2tp_eth.c | 2 +-
net/mac80211/iface.c | 1 +
14 files changed, 22 insertions(+), 8 deletions(-)
diff --git a/drivers/isdn/i4l/isdn_net.c b/drivers/isdn/i4l/isdn_net.c
index 48e9cc0..1f73d7f 100644
--- a/drivers/isdn/i4l/isdn_net.c
+++ b/drivers/isdn/i4l/isdn_net.c
@@ -2532,6 +2532,9 @@ static void _isdn_setup(struct net_device *dev)
/* Setup the generic properties */
dev->flags = IFF_NOARP|IFF_POINTOPOINT;
+
+ /* isdn prepends a header in the tx path, can't share skbs */
+ dev->priv_flags &= ~IFF_TX_SKB_SHARING;
dev->header_ops = NULL;
dev->netdev_ops = &isdn_netdev_ops;
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 02842d0..df21e84 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1557,8 +1557,10 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
if (slave_dev->type != ARPHRD_ETHER)
bond_setup_by_slave(bond_dev, slave_dev);
- else
+ else {
ether_setup(bond_dev);
+ bond_dev->priv_flags &= ~IFF_TX_SKB_SHARING;
+ }
netdev_bonding_change(bond_dev,
NETDEV_POST_TYPE_CHANGE);
@@ -4330,7 +4332,7 @@ static void bond_setup(struct net_device *bond_dev)
bond_dev->tx_queue_len = 0;
bond_dev->flags |= IFF_MASTER|IFF_MULTICAST;
bond_dev->priv_flags |= IFF_BONDING;
- bond_dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
+ bond_dev->priv_flags &= ~(IFF_XMIT_DST_RELEASE | IFF_TX_SKB_SHARING);
/* At first, we block adding VLANs. That's the only way to
* prevent problems that occur when adding VLANs over an
diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index 6e82dd3..46b5f5f 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -183,7 +183,7 @@ static void ifb_setup(struct net_device *dev)
dev->flags |= IFF_NOARP;
dev->flags &= ~IFF_MULTICAST;
- dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
+ dev->priv_flags &= ~(IFF_XMIT_DST_RELEASE | IFF_TX_SKB_SHARING);
random_ether_addr(dev->dev_addr);
}
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index ba631fc..05172c3 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -572,7 +572,7 @@ void macvlan_common_setup(struct net_device *dev)
{
ether_setup(dev);
- dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
+ dev->priv_flags &= ~(IFF_XMIT_DST_RELEASE | IFF_TX_SKB_SHARING);
dev->netdev_ops = &macvlan_netdev_ops;
dev->destructor = free_netdev;
dev->header_ops = &macvlan_hard_header_ops,
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 9a6b382..71f3d1a 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -528,6 +528,7 @@ static void tun_net_init(struct net_device *dev)
dev->netdev_ops = &tap_netdev_ops;
/* Ethernet TAP Device */
ether_setup(dev);
+ dev->priv_flags &= ~IFF_TX_SKB_SHARING;
random_ether_addr(dev->dev_addr);
diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index 7f78db7..5b23767 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -263,6 +263,8 @@ static void veth_setup(struct net_device *dev)
{
ether_setup(dev);
+ dev->priv_flags &= ~IFF_TX_SKB_SHARING;
+
dev->netdev_ops = &veth_netdev_ops;
dev->ethtool_ops = &veth_ethtool_ops;
dev->features |= NETIF_F_LLTX;
diff --git a/drivers/net/wan/hdlc_fr.c b/drivers/net/wan/hdlc_fr.c
index b25c922..eb20281 100644
--- a/drivers/net/wan/hdlc_fr.c
+++ b/drivers/net/wan/hdlc_fr.c
@@ -1074,9 +1074,10 @@ static int fr_add_pvc(struct net_device *frad, unsigned int dlci, int type)
used = pvc_is_used(pvc);
- if (type == ARPHRD_ETHER)
+ if (type == ARPHRD_ETHER) {
dev = alloc_netdev(0, "pvceth%d", ether_setup);
- else
+ dev->priv_flags &= ~IFF_TX_SKB_SHARING;
+ } else
dev = alloc_netdev(0, "pvc%d", pvc_setup);
if (!dev) {
diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c
index 55cf71f..e1b3e3c 100644
--- a/drivers/net/wireless/airo.c
+++ b/drivers/net/wireless/airo.c
@@ -2823,6 +2823,7 @@ static struct net_device *_init_airo_card( unsigned short irq, int port,
dev->wireless_data = &ai->wireless_data;
dev->irq = irq;
dev->base_addr = port;
+ dev->priv_flags &= ~IFF_TX_SKB_SHARING;
SET_NETDEV_DEV(dev, dmdev);
diff --git a/drivers/net/wireless/hostap/hostap_main.c b/drivers/net/wireless/hostap/hostap_main.c
index d508482..89a116f 100644
--- a/drivers/net/wireless/hostap/hostap_main.c
+++ b/drivers/net/wireless/hostap/hostap_main.c
@@ -855,6 +855,7 @@ void hostap_setup_dev(struct net_device *dev, local_info_t *local,
iface = netdev_priv(dev);
ether_setup(dev);
+ dev->priv_flags &= ~IFF_TX_SKB_SHARING;
/* kernel callbacks */
if (iface) {
diff --git a/drivers/staging/ath6kl/os/linux/ar6000_drv.c b/drivers/staging/ath6kl/os/linux/ar6000_drv.c
index 48dd9e3..8ff5289 100644
--- a/drivers/staging/ath6kl/os/linux/ar6000_drv.c
+++ b/drivers/staging/ath6kl/os/linux/ar6000_drv.c
@@ -6179,6 +6179,7 @@ int ar6000_create_ap_interface(struct ar6_softc *ar, char *ap_ifname)
ether_setup(dev);
init_netdev(dev, ap_ifname);
+ dev->priv_flags &= ~IFF_TX_SKB_SHARING;
if (register_netdev(dev)) {
AR_DEBUG_PRINTF(ATH_DEBUG_ERR,("ar6000_create_ap_interface: register_netdev failed\n"));
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 934e221..9d40a07 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -695,7 +695,7 @@ void vlan_setup(struct net_device *dev)
ether_setup(dev);
dev->priv_flags |= IFF_802_1Q_VLAN;
- dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
+ dev->priv_flags &= ~(IFF_XMIT_DST_RELEASE | IFF_TX_SKB_SHARING);
dev->tx_queue_len = 0;
dev->netdev_ops = &vlan_netdev_ops;
diff --git a/net/bluetooth/bnep/netdev.c b/net/bluetooth/bnep/netdev.c
index 8c100c9..d4f5dff 100644
--- a/net/bluetooth/bnep/netdev.c
+++ b/net/bluetooth/bnep/netdev.c
@@ -231,6 +231,7 @@ void bnep_net_setup(struct net_device *dev)
dev->addr_len = ETH_ALEN;
ether_setup(dev);
+ dev->priv_flags &= ~IFF_TX_SKB_SHARING;
dev->netdev_ops = &bnep_netdev_ops;
dev->watchdog_timeo = HZ * 2;
diff --git a/net/l2tp/l2tp_eth.c b/net/l2tp/l2tp_eth.c
index a8193f5..d2726a7 100644
--- a/net/l2tp/l2tp_eth.c
+++ b/net/l2tp/l2tp_eth.c
@@ -103,7 +103,7 @@ static struct net_device_ops l2tp_eth_netdev_ops = {
static void l2tp_eth_dev_setup(struct net_device *dev)
{
ether_setup(dev);
-
+ dev->priv_flags &= ~IFF_TX_SKB_SHARING;
dev->netdev_ops = &l2tp_eth_netdev_ops;
dev->destructor = free_netdev;
}
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index cd5fb40..556e7e6 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -698,6 +698,7 @@ static const struct net_device_ops ieee80211_monitorif_ops = {
static void ieee80211_if_setup(struct net_device *dev)
{
ether_setup(dev);
+ dev->priv_flags &= ~IFF_TX_SKB_SHARING;
dev->netdev_ops = &ieee80211_dataif_ops;
dev->destructor = free_netdev;
}
--
1.7.6
^ permalink raw reply related
* [PATCH 1/2] net: add IFF_SKB_TX_SHARED flag to priv_flags
From: Neil Horman @ 2011-07-25 19:45 UTC (permalink / raw)
To: netdev
Cc: Neil Horman, Robert Olsson, Eric Dumazet, Alexey Dobriyan,
David S. Miller
In-Reply-To: <1311623120-26839-1-git-send-email-nhorman@tuxdriver.com>
Pktgen attempts to transmit shared skbs to net devices, which can't be used by
some drivers as they keep state information in skbs. This patch adds a flag
marking drivers as being able to handle shared skbs in their tx path. Drivers
are defaulted to being unable to do so, but calling ether_setup enables this
flag, as 90% of the drivers calling ether_setup touch real hardware and can
handle shared skbs. A subsequent patch will audit drivers to ensure that the
flag is set properly
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Reported-by: Jiri Pirko <jpirko@redhat.com>
CC: Robert Olsson <robert.olsson@its.uu.se>
CC: Eric Dumazet <eric.dumazet@gmail.com>
CC: Alexey Dobriyan <adobriyan@gmail.com>
CC: David S. Miller <davem@davemloft.net>
---
include/linux/if.h | 2 ++
net/core/pktgen.c | 8 +++++---
net/ethernet/eth.c | 1 +
3 files changed, 8 insertions(+), 3 deletions(-)
diff --git a/include/linux/if.h b/include/linux/if.h
index 3bc63e6..03489ca 100644
--- a/include/linux/if.h
+++ b/include/linux/if.h
@@ -76,6 +76,8 @@
#define IFF_BRIDGE_PORT 0x4000 /* device used as bridge port */
#define IFF_OVS_DATAPATH 0x8000 /* device used as Open vSwitch
* datapath port */
+#define IFF_TX_SKB_SHARING 0x10000 /* The interface supports sharing
+ * skbs on transmit */
#define IF_GET_IFACE 0x0001 /* for querying only */
#define IF_GET_PROTO 0x0002
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index f76079c..53f3f15 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -1070,7 +1070,9 @@ static ssize_t pktgen_if_write(struct file *file,
len = num_arg(&user_buffer[i], 10, &value);
if (len < 0)
return len;
-
+ if ((len > 0) &&
+ (!(pkt_dev->odev->priv_flags & IFF_TX_SKB_SHARING)))
+ return -ENOTSUPP;
i += len;
pkt_dev->clone_skb = value;
@@ -3555,7 +3557,6 @@ static int pktgen_add_device(struct pktgen_thread *t, const char *ifname)
pkt_dev->min_pkt_size = ETH_ZLEN;
pkt_dev->max_pkt_size = ETH_ZLEN;
pkt_dev->nfrags = 0;
- pkt_dev->clone_skb = pg_clone_skb_d;
pkt_dev->delay = pg_delay_d;
pkt_dev->count = pg_count_d;
pkt_dev->sofar = 0;
@@ -3563,7 +3564,6 @@ static int pktgen_add_device(struct pktgen_thread *t, const char *ifname)
pkt_dev->udp_src_max = 9;
pkt_dev->udp_dst_min = 9;
pkt_dev->udp_dst_max = 9;
-
pkt_dev->vlan_p = 0;
pkt_dev->vlan_cfi = 0;
pkt_dev->vlan_id = 0xffff;
@@ -3575,6 +3575,8 @@ static int pktgen_add_device(struct pktgen_thread *t, const char *ifname)
err = pktgen_setup_dev(pkt_dev, ifname);
if (err)
goto out1;
+ if (pkt_dev->odev->priv_flags & IFF_TX_SKB_SHARING)
+ pkt_dev->clone_skb = pg_clone_skb_d;
pkt_dev->entry = proc_create_data(ifname, 0600, pg_proc_dir,
&pktgen_if_fops, pkt_dev);
diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c
index 5cffb63..4866330 100644
--- a/net/ethernet/eth.c
+++ b/net/ethernet/eth.c
@@ -339,6 +339,7 @@ void ether_setup(struct net_device *dev)
dev->addr_len = ETH_ALEN;
dev->tx_queue_len = 1000; /* Ethernet wants good queues */
dev->flags = IFF_BROADCAST|IFF_MULTICAST;
+ dev->priv_flags = IFF_TX_SKB_SHARING;
memset(dev->broadcast, 0xFF, ETH_ALEN);
--
1.7.6
^ permalink raw reply related
* [PATCH 0/2] pktgen: Clone skb to avoid corruption of skbs in ndo_start_xmit methods (v2)
From: Neil Horman @ 2011-07-25 19:45 UTC (permalink / raw)
To: netdev
In-Reply-To: <1311105179-26408-1-git-send-email-nhorman@tuxdriver.com>
Ok, after considering all your comments, Dave suggested this as an alternate
approach:
1) We create a new priv_flag, IFF_SKB_TX_SHARED, to identify drivers capable of
handling shared skbs. Default is to not set this flag
2) Modify ether_setup to enable this flag, under the assumption that any driver
calling this function is initalizing a real ethernet device and as such can
handle shared skbs since they don't tend to store state in the skb struct.
Pktgen can then query this flag when a user script attempts to issue the
clone_skb command and decide if it is to be alowed or not.
3) Audit the network drivers calling ether_setup to identify any code doing so
that can't actualy handle shared skbs and manually disable the new flag. There
are about 10 drivers in this category.
Thoughts/reviews aprpeciated.
Neil
^ permalink raw reply
* Re: [lm-sensors] [PATCH 01/34] System Firmware Interface
From: Jean Delvare @ 2011-07-25 19:03 UTC (permalink / raw)
To: Prarit Bhargava
Cc: linux-kernel-u79uwXL29TY76Z2rM5mHXA,
linux-ia64-u79uwXL29TY76Z2rM5mHXA,
linux-pci-u79uwXL29TY76Z2rM5mHXA,
dri-devel-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW,
platform-driver-x86-u79uwXL29TY76Z2rM5mHXA,
grant.likely-s3s/WqlpOiPyB63q8FvJNQ,
linux-ide-u79uwXL29TY76Z2rM5mHXA,
linux-i2c-u79uwXL29TY76Z2rM5mHXA,
device-drivers-devel-ZG0+EudsQA8dtHy/vicBwGD2FQJk+8+b,
abelay-3s7WtUTddSA, eric.piel-VkQ1JFuSMpfAbQlEx87xDw,
x86-DgEjT+Ai2ygdnm+yROfE0A, lm-sensors-GZX6beZjE8VD60Wz+7aTrA,
linux-acpi-u79uwXL29TY76Z2rM5mHXA,
linux-input-u79uwXL29TY76Z2rM5mHXA,
linux-media-u79uwXL29TY76Z2rM5mHXA,
johnpol-9fLWQ3dKdXwox3rIn2DAYQ,
linux-watchdog-u79uwXL29TY76Z2rM5mHXA,
rtc-linux-/JYPxA39Uh5TLH3MbocFFw, dz-8fiUuRrzOP0dnm+yROfE0A,
openipmi-developer-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f,
evel-gWbeCf7V1WCQmaza687I9mD2FQJk+8+b,
netdev-u79uwXL29TY76Z2rM5mHXA, linux-usb-u79uwXL29TY76Z2rM5mHXA,
rpurdie-Fm38FmjxZ/leoWH0uzbU5w,
linux-crypto-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1310994528-26276-2-git-send-email-prarit-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
On Mon, 18 Jul 2011 09:08:15 -0400, Prarit Bhargava wrote:
> This patch introduces a general System Firmware interface to the kernel, called
> sysfw.
>
> Inlcluded in this interface is the ability to search a standard set of fields,
> sysfw_lookup(). The fields are currently based upon the x86 and ia64 SMBIOS
> fields but exapandable to fields that other arches may introduce. Also
> included is the ability to search and match against those fields, and run
> a callback function against the matches, sysfw_callback().
>
> Modify module code to use sysfw instead of old DMI interface.
This is a HUGE patch set. You'd need to have a good reason for such a
big and intrusive change, yet I see no such reason explained. I
understand that we _can_ abstract system information interfaces, but
just because we can doesn't mean we have to. I would at least wait for
a second DMI-like interface to be widely implemented and support before
any attempt to abstract, otherwise your design is bound to be missing
the target. And even then, you'd still need to convince me that there
is a need for a unified interface to access both backends at once. I
would guess that you know what backend is present on a system when you
try to identify it.
At this point, I see the work needed to review your patches, the risk
of regressions due to the large size of the patch set, but I don't see
any immediate benefit. Thus I am not going to look into it at all,
sorry.
--
Jean Delvare
^ permalink raw reply
* [PATCH] net: fix eth.c kernel-doc warning
From: Randy Dunlap @ 2011-07-25 18:41 UTC (permalink / raw)
To: netdev; +Cc: davem
From: Randy Dunlap <rdunlap@xenotime.net>
Fix new kernel-doc warning in eth.c:
Warning(net/ethernet/eth.c:237): No description found for parameter 'type'
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
---
net/ethernet/eth.c | 1 +
1 file changed, 1 insertion(+)
--- linux-3.0-git4.orig/net/ethernet/eth.c
+++ linux-3.0-git4/net/ethernet/eth.c
@@ -231,6 +231,7 @@ EXPORT_SYMBOL(eth_header_parse);
* eth_header_cache - fill cache entry from neighbour
* @neigh: source neighbour
* @hh: destination cache entry
+ * @type: Ethernet type field
* Create an Ethernet header template from the neighbour.
*/
int eth_header_cache(const struct neighbour *neigh, struct hh_cache *hh, __be16 type)
^ permalink raw reply
* Re: write() udp socket
From: Rick Jones @ 2011-07-25 17:38 UTC (permalink / raw)
To: Huajun Li; +Cc: ZHOU Xiaobo, netdev
In-Reply-To: <CA+v9cxZAF27mXc6aCEPyOdqbeLved3_xibvDS4TQugHR=7owPg@mail.gmail.com>
On 07/24/2011 01:33 AM, Huajun Li wrote:
> 2011/7/23 ZHOU Xiaobo<xb.zhou@qq.com>:
>> question No1:
>> When I call
>> ssize_t write(int fd, const void *buf, size_t count);
>>
>>
>> on a nonblocking UDP socket, is the return value always equal to 'count'?
>>
>>
>
> I don't think so. The function may be interrupt by signal or return
> due to other reason, so the return value only represents the size it
> writes successfully to the fd.
I believe it should either appaear to succeed or fail. write() best not
be sending partial UDP datagrams. That would be "bad."
^ permalink raw reply
* Attention
From: MONEYGRAM CENTER @ 2011-07-25 17:18 UTC (permalink / raw)
My associate has helped me to send your first payment of $5000 USD to you
as instructed by the Malaysian Government and Mr. David Cameron the United
Kingdom prime minister after the last G20 meeting that was held in
Malaysia, making you one of the beneficiaries.Here is the information
below.
MONEY TRANSFER REFERENCE: 2116-3297
SENDER'S NAME: Patrick Lee Chun
AMOUNT: US$5000
We will keep sending you $5000 USD twice a week until the FULL payment of
($820000.00 United State Dollars)is completed.
A certificate will be made to change the Receivers Name to your name as
stated by the Malaysian Government,reconfirm your {1}Full Names {2}address
{3}Mobile Number
via Email to: center.moneygram@yahoo.com.my to proceed.
Note: You cannot pickup the money until the certificate is obtained by you.
Best Regards
Money Gram Agent
Mr. Allan Davis.
Tel: +(60)1-6240-6207
For more info: www.g20.org
I wrote to know if this is your valid email. Please, let me know if this
email is valid.I have an information to pass to you.
My valid Email is: center.moneygram@yahoo.com.my
Money Gram Agent
Mr. Allan Davis.
Tel: +(60)1-6240-6207
For more info: www.g20.org
^ permalink raw reply
* Re: [PATCH] net: Fix security_socket_sendmsg() bypass problem.
From: Casey Schaufler @ 2011-07-25 17:00 UTC (permalink / raw)
To: Tetsuo Handa
Cc: anton, mjt, davem, netdev, linux-security-module, Casey Schaufler
In-Reply-To: <201107260143.CGH18263.FOOSVMOQFJFLHt@I-love.SAKURA.ne.jp>
On 7/25/2011 9:43 AM, Tetsuo Handa wrote:
> Casey Schaufler wrote:
>>> OK. Then, the question is how to reduce performance loss by redundant
>>> security_socket_sendmsg() calls.
>> Not to be splitting hairs, but if the packets are headed to
>> different destinations the calls to security_socket_sendmsg()
>> are not redundant, they are necessary and appropriate. What
>> you have with sendmmsg() is an optimization that sacrifices
>> correctness for performance.
> Excuse me, but this thread is not trying to remove necessary and appropriate
> security_socket_sendmsg() calls. Linux 3.0 was released without necessary and
> appropriate security_socket_sendmsg() calls, and I'm trying to correct it (via
> msg11504.html or msg11510.html) for Linux 3.0.x stable release.
I understand. Sorry if I did a poor job of jumping into
the thread.
>> I fear that you are going to find that the work you have
>> to do to reduce the number of calls is going to outweigh
>> the benefits of your optimization, as has been pointed out
>> earlier.
> I fear it too. Unless many dozens (maybe some hundreds) of packets are sent by
> sendmmsg(), msg11504.html might show better performance than msg11510.html .
> But I don't have a machine to benchmark.
Is there some chance that the original authors could step up
to help with the benchmarking effort on this repair? Having been
on the end where I introduced problems more than once, I have a
good understanding of the principle "you broke it, you bought it".
^ permalink raw reply
* Re: IPv6: autoconfiguration and suspend/resume or link down/up
From: Stephen Hemminger @ 2011-07-25 16:55 UTC (permalink / raw)
To: Herbert Xu, Stephen Hemminger
Cc: Nicolas de Pesloüan, David Miller, jbohac, netdev
who manages link wit sriov? I assume it is up to the guest. And it is not really safe to assume that network is the same after migration. It makes sense to do DAD again.
Herbert Xu <herbert@gondor.apana.org.au> wrote:
>On Sun, Jul 24, 2011 at 08:26:20PM -0700, Stephen Hemminger wrote:
>>
>> Since virtual machines should be using virtio network devices, shouldn't
>> the suspend/resume in that device just work. It doesn't need to drop the link.
>
>The VM may also be using SRIOV.
>
>Cheers,
>--
>Email: Herbert Xu <herbert@gondor.apana.org.au>
>Home Page: http://gondor.apana.org.au/~herbert/
>PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
^ permalink raw reply
* Re: IPv6: autoconfiguration and suspend/resume or link down/up
From: Stephen Hemminger @ 2011-07-19 19:42 UTC (permalink / raw)
To: Jiri Bohac, netdev; +Cc: Herbert Xu, David S. Miller, stephen hemminger
bridge forwarding table; route cache; and neighbor table could have same problem. I thought carrier is supposed to toggle on suspend or hibernate
Jiri Bohac <jbohac@suse.cz> wrote:
>Hi,
>
>I came over a surprising behaviour with IPv6 autoconfiguration,
>which I think is a bug, but I would first like to hear other
>people's opinions before trying to fix this:
>
>Problem 1: all the address/route lifetimes are kept in jiffies
>and jiffies don't get incremented on resume. So when a
>route/address lifetime is 30 minutes and the system resumes after
>1 hour, the route/address should be considered expired, but it is
>not.
>
>Problem 2: when a system is moved to a new network a RS is not
>sent. Thus, IPv6 does not autoconfigure until the router sends a
>periodic RA. This can occur both while the system is alive and
>while it is suspended. I think the autoconfigured state should be
>discarded when the kernel suspects the system could have been
>moved to a different network.
>
>When the cable is unplugged and plugged in again, we already get
>notified through linkwatch -> netdev_state_change ->
> -> call_netdevice_notifiers(NETDEV_CHANGE, ...)
>However, if the device has already been autoconfigured,
>addrconf_notify() only handles this event by printing a
>message.
>
>So my idea was to:
>- handle link up/down in addrconf_notify() similarly to
> NETDEV_UP/NETDEV_DOWN
>
>- on suspend, faking a link down event; on resume, faking a link up event
> (or better, having a special event type for suspend/resume)
>
>This would cause autoconfiguration to be restarted on resume as
>well as cable plug/unplug, solving both the above problems.
>
>Or do we want to completely rely on userspace tools
>(networkmanager/ifplug) and expect them to do NETDEV_DOWN on
>unplug/suspend and NETDEV_UP on plug/resume?
>
>Any thoughts?
>
>--
>Jiri Bohac <jbohac@suse.cz>
>SUSE Labs, SUSE CZ
>
^ permalink raw reply
* Re: [PATCH] net: Fix security_socket_sendmsg() bypass problem.
From: Tetsuo Handa @ 2011-07-25 16:43 UTC (permalink / raw)
To: casey; +Cc: anton, mjt, davem, netdev, linux-security-module
In-Reply-To: <4E2D8F4D.2000009@schaufler-ca.com>
Casey Schaufler wrote:
> > OK. Then, the question is how to reduce performance loss by redundant
> > security_socket_sendmsg() calls.
>
> Not to be splitting hairs, but if the packets are headed to
> different destinations the calls to security_socket_sendmsg()
> are not redundant, they are necessary and appropriate. What
> you have with sendmmsg() is an optimization that sacrifices
> correctness for performance.
Excuse me, but this thread is not trying to remove necessary and appropriate
security_socket_sendmsg() calls. Linux 3.0 was released without necessary and
appropriate security_socket_sendmsg() calls, and I'm trying to correct it (via
msg11504.html or msg11510.html) for Linux 3.0.x stable release.
> I fear that you are going to find that the work you have
> to do to reduce the number of calls is going to outweigh
> the benefits of your optimization, as has been pointed out
> earlier.
I fear it too. Unless many dozens (maybe some hundreds) of packets are sent by
sendmmsg(), msg11504.html might show better performance than msg11510.html .
But I don't have a machine to benchmark.
^ permalink raw reply
* Re: [PATCH 0/7] More sane neigh infrastructure
From: Roland Dreier @ 2011-07-25 16:34 UTC (permalink / raw)
To: David Miller; +Cc: linux-rdma, netdev
In-Reply-To: <20110725.030109.1723861338142084129.davem@davemloft.net>
On Mon, Jul 25, 2011 at 3:01 AM, David Miller <davem@davemloft.net> wrote:
> Devices provide up to three things:
>
> 1) netdev->neighpriv_len, length of per-neighbour device private
> state, accessible via neighbour_priv(neigh)
>
> 2) net_device_ops->ndo_neigh_construct(), invoked right after
> neigh_tbl->constructor(), can fail
Hey Dave,
I'll definitely look at converting IPoIB over to using this stuff.
Would love to get rid of all the dicy handling of ipoib_neigh lifetime
that we currently have. However, I have a question about what the
intention for ndo_neigh_construct() is in the IPoIB case.
As we talked about, IPoIB has to trigger a path lookup to the subnet
manager (SM) when it gets a remote port ID. However the SM is a
remote entity, so this lookup means we send a message and then
asynchronously wait for it to complete (or possibly timeout), just
like the ARP itself. But this is done after we get the port ID via
normal RFC 826 ARP (with an address format as specified by RFC 4391).
So I don't think we can use custom neigh_ops with a new solict method
the way clip does -- we actually want to let the normal stack do ARP
or ND, but then extend the process by another message/response step.
I'm sure this is possible within your scheme but I'm not sure I
understand what the "right" way is.
Thanks!
Roland
^ permalink raw reply
* Re: [PATCH] net: Fix security_socket_sendmsg() bypass problem.
From: Casey Schaufler @ 2011-07-25 15:44 UTC (permalink / raw)
To: Tetsuo Handa
Cc: anton, mjt, davem, netdev, linux-security-module, Casey Schaufler
In-Reply-To: <201107252215.GBG95887.OQVMFOOJLSFFHt@I-love.SAKURA.ne.jp>
On 7/25/2011 6:15 AM, Tetsuo Handa wrote:
> Anton Blanchard wrote:
>>>> When I saw recvmmsg()/sendmmsg() here, my first thought was an
>>>> authoritative DNS server which can read several requests at a
>>>> time and answer them all at once too - this way it all will go
>>>> to different addresses.
>>> I don't know what application wants sendmmsg(). Since users can send
>>> up to UIO_MAXIOV (= 1024) "struct iovec" blocks using sendmsg(), they
>>> will use sendmsg() rather than sendmmsg() if the destination address
>>> are the same.
>> But if an application needs to maintain packet boundaries, then sendmsg
>> isn't going to help is it?
> Well, such application might want to use RDM or SeqPacket... but your point is
> to maintain packet boundaries. You are assuming that sendmmsg() will be used
> for sending as much data as possible while preserving packet boundaries.
>
> OK. Then, the question is how to reduce performance loss by redundant
> security_socket_sendmsg() calls.
Not to be splitting hairs, but if the packets are headed to
different destinations the calls to security_socket_sendmsg()
are not redundant, they are necessary and appropriate. What
you have with sendmmsg() is an optimization that sacrifices
correctness for performance.
> If sendmmsg() likely contains single (or few)
> destination(s), trying to optimize security_socket_sendmsg() calls by comparing
> destination address (as proposed at
> http://www.spinics.net/linux/fedora/linux-security-module/msg11510.html
> ) would help. Otherwise, no optimization (as proposed at
> http://www.spinics.net/linux/fedora/linux-security-module/msg11504.html
> ) would be better. Which approach do you like?
I fear that you are going to find that the work you have
to do to reduce the number of calls is going to outweigh
the benefits of your optimization, as has been pointed out
earlier. My recommendation is that the sendmmsg() interface
is ill conceived and that you should look for alternative
ways to improve the performance of the use case.
^ permalink raw reply
* Pick Up Your $5000.00USD,
From: WESTERN UNION OFFICE @ 2011-07-25 10:01 UTC (permalink / raw)
How are you today?
I write to inform you that we have already sent you $5,000.00USD
through Western union as we have been given the mandate to transfer
your full compensation payment of $1.800,000.00USD via western union
by this government.
I called to give you the information through phone as internet hackers
were many but i cannot reach you yesterday even this morning,So I
decided to email you the (MTCN) and sender name so that you can pick
up this $5,000.00USD to enable us send another $5,000.00USD by
tomorrow as you knows we will be sending you only $5,000.00USD per
day.Please pick up this information and run to any western union
(OUTLET) in your country and pick up this $5,000.00USD and send us an
email back,so that we can send another $5,000.00USD by tomorrow.
Manager: Mr Frank Amos
email me on:western-money677@hotmail.com
call us on: +234-7031908911
once you picked up this $5000.00USD today.
Here is the western union information to pick up the $5000.00USD,
MTCN :___________MTCN 9500834460
first name: ______Appoline
Second Name: ______Ouedraoge
Text Question: ___________When
Answer: ___________________2Hours ago
Amount:______________________ $5,000.00 United State Dollars
I am waiting for your E-mail once you pick up $5000.00USD,
Thanks
Mr Frank Amos.
^ permalink raw reply
* [PATCH net-next] net: Convert struct net_device uc_promisc to bool
From: Joe Perches @ 2011-07-25 14:41 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, linux-kernel
No need to use int, its uses are boolean.
May save a few bytes one day.
Signed-off-by: Joe Perches <joe@perches.com>
---
include/linux/netdevice.h | 2 +-
net/core/dev.c | 4 ++--
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 34f3abc..1d92acc 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1132,7 +1132,7 @@ struct net_device {
spinlock_t addr_list_lock;
struct netdev_hw_addr_list uc; /* Unicast mac addresses */
struct netdev_hw_addr_list mc; /* Multicast mac addresses */
- int uc_promisc;
+ bool uc_promisc;
unsigned int promiscuity;
unsigned int allmulti;
diff --git a/net/core/dev.c b/net/core/dev.c
index 9444c5c..17d67b5 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4497,10 +4497,10 @@ void __dev_set_rx_mode(struct net_device *dev)
*/
if (!netdev_uc_empty(dev) && !dev->uc_promisc) {
__dev_set_promiscuity(dev, 1);
- dev->uc_promisc = 1;
+ dev->uc_promisc = true;
} else if (netdev_uc_empty(dev) && dev->uc_promisc) {
__dev_set_promiscuity(dev, -1);
- dev->uc_promisc = 0;
+ dev->uc_promisc = false;
}
if (ops->ndo_set_multicast_list)
--
1.7.6.131.g99019
^ permalink raw reply related
* Re: [bisected regression] Partial breakage of forcedeth driver
From: Jiri Pirko @ 2011-07-25 14:36 UTC (permalink / raw)
To: walt; +Cc: netdev
In-Reply-To: <4E2CA683.4010005@gmail.com>
Mon, Jul 25, 2011 at 01:10:59AM CEST, w41ter@gmail.com wrote:
>Hi gang.
>
>commit 3326c784c9f492e988617d93f647ae0cfd4c8d09
>Author: Jiri Pirko <jpirko@>
>Date: Wed Jul 20 04:54:38 2011 +0000
>
> forcedeth: do vlan cleanup
>
> - unify vlan and nonvlan rx path
> - kill np->vlangrp and nv_vlan_rx_register
> - allow to turn on/off rx vlan accel via ethtool (set_features)
>
> Signed-off-by: Jiri Pirko <jpirko@>
> Signed-off-by: David S. Miller <davem@>
>
>This commit causes networking trouble for my nForce-based motherboard.
>(Details happily supplied if needed.)
>
>I say 'partial' breakage because networking is not completely dead,
>just limping very slowly :)
>
>The simplest test is to ping any host (in my LAN or on the internet,
>makes no difference). The replies to my ping come at very irregular
>intervals, ranging from normal to 5-10 seconds or so, seemingly at
>random AFAICT.
>
>I'm happy to try any patches, tests, whatever, to help fix this.
>
>Many thanks!
Same is happening on machine I borrowed. ccing list.
Will look at it very soon (no time today).
jirka
^ permalink raw reply
* [PATCH 7/7] dccp ccid-2: check Ack Ratio when reducing cwnd
From: Gerrit Renker @ 2011-07-25 13:36 UTC (permalink / raw)
To: davem; +Cc: dccp, netdev, Samuel Jero
In-Reply-To: <test_tree_patch_set_update_2011-07-25>
From: Samuel Jero <sj323707@ohio.edu>
This patch causes CCID-2 to check the Ack Ratio after reducing the congestion
window. If the Ack Ratio is greater than the congestion window, it is
reduced. This prevents timeouts caused by an Ack Ratio larger than the
congestion window.
In this situation, we choose to set the Ack Ratio to half the congestion window
(or one if that's zero) so that if we loose one ack we don't trigger a timeout.
Signed-off-by: Samuel Jero <sj323707@ohio.edu>
Acked-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
---
net/dccp/ccids/ccid2.c | 26 +++++++++++++++++++++++---
1 files changed, 23 insertions(+), 3 deletions(-)
--- a/net/dccp/ccids/ccid2.c
+++ b/net/dccp/ccids/ccid2.c
@@ -101,6 +101,24 @@ static void ccid2_change_l_ack_ratio(struct sock *sk, u32 val)
min_t(u32, val, DCCPF_ACK_RATIO_MAX));
}
+static void ccid2_check_l_ack_ratio(struct sock *sk)
+{
+ struct ccid2_hc_tx_sock *hc = ccid2_hc_tx_sk(sk);
+
+ /*
+ * After a loss, idle period, application limited period, or RTO we
+ * need to check that the ack ratio is still less than the congestion
+ * window. Otherwise, we will send an entire congestion window of
+ * packets and got no response because we haven't sent ack ratio
+ * packets yet.
+ * If the ack ratio does need to be reduced, we reduce it to half of
+ * the congestion window (or 1 if that's zero) instead of to the
+ * congestion window. This prevents problems if one ack is lost.
+ */
+ if (dccp_feat_nn_get(sk, DCCPF_ACK_RATIO) > hc->tx_cwnd)
+ ccid2_change_l_ack_ratio(sk, hc->tx_cwnd/2 ? : 1U);
+}
+
static void ccid2_change_l_seq_window(struct sock *sk, u64 val)
{
dccp_feat_signal_nn_change(sk, DCCPF_SEQUENCE_WINDOW,
@@ -187,6 +205,8 @@ static void ccid2_cwnd_application_limited(struct sock *sk, const u32 now)
}
hc->tx_cwnd_used = 0;
hc->tx_cwnd_stamp = now;
+
+ ccid2_check_l_ack_ratio(sk);
}
/* This borrows the code of tcp_cwnd_restart() */
@@ -205,6 +225,8 @@ static void ccid2_cwnd_restart(struct sock *sk, const u32 now)
hc->tx_cwnd_stamp = now;
hc->tx_cwnd_used = 0;
+
+ ccid2_check_l_ack_ratio(sk);
}
static void ccid2_hc_tx_packet_sent(struct sock *sk, unsigned int len)
@@ -461,9 +483,7 @@ static void ccid2_congestion_event(struct sock *sk, struct ccid2_seq *seqp)
hc->tx_cwnd = hc->tx_cwnd / 2 ? : 1U;
hc->tx_ssthresh = max(hc->tx_cwnd, 2U);
- /* Avoid spurious timeouts resulting from Ack Ratio > cwnd */
- if (dccp_sk(sk)->dccps_l_ack_ratio > hc->tx_cwnd)
- ccid2_change_l_ack_ratio(sk, hc->tx_cwnd);
+ ccid2_check_l_ack_ratio(sk);
}
static int ccid2_hc_tx_parse_options(struct sock *sk, u8 packet_type,
^ 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