Currently cfg80211 fails to create a "phy80211" symlink in sysfs from
the net device to the wiphy device. The latter needs to be registered
first.
Compile-tested only.
Reported-by: Cesare Leonardi <[email protected]>
Signed-off-by: Ben Hutchings <[email protected]>
---
drivers/net/wireless/ipw2x00/ipw2200.c | 23 ++++++++++++-----------
1 files changed, 12 insertions(+), 11 deletions(-)
diff --git a/drivers/net/wireless/ipw2x00/ipw2200.c b/drivers/net/wireless/ipw2x00/ipw2200.c
index 018a8de..737dfaf 100644
--- a/drivers/net/wireless/ipw2x00/ipw2200.c
+++ b/drivers/net/wireless/ipw2x00/ipw2200.c
@@ -11839,16 +11839,17 @@ static int __devinit ipw_pci_probe(struct pci_dev *pdev,
}
mutex_unlock(&priv->mutex);
- err = register_netdev(net_dev);
+
+ err = ipw_wdev_init(net_dev);
if (err) {
- IPW_ERROR("failed to register network device\n");
+ IPW_ERROR("failed to register wireless device\n");
goto out_remove_sysfs;
}
- err = ipw_wdev_init(net_dev);
+ err = register_netdev(net_dev);
if (err) {
- IPW_ERROR("failed to register wireless device\n");
- goto out_unregister_netdev;
+ IPW_ERROR("failed to register network device\n");
+ goto out_unregister_wiphy;
}
#ifdef CONFIG_IPW2200_PROMISCUOUS
@@ -11857,10 +11858,8 @@ static int __devinit ipw_pci_probe(struct pci_dev *pdev,
if (err) {
IPW_ERROR("Failed to register promiscuous network "
"device (error %d).\n", err);
- wiphy_unregister(priv->ieee->wdev.wiphy);
- kfree(priv->ieee->a_band.channels);
- kfree(priv->ieee->bg_band.channels);
- goto out_unregister_netdev;
+ unregister_netdev(priv->net_dev);
+ goto out_unregister_wiphy;
}
}
#endif
@@ -11872,8 +11871,10 @@ static int __devinit ipw_pci_probe(struct pci_dev *pdev,
return 0;
- out_unregister_netdev:
- unregister_netdev(priv->net_dev);
+ out_unregister_wiphy:
+ wiphy_unregister(priv->ieee->wdev.wiphy);
+ kfree(priv->ieee->a_band.channels);
+ kfree(priv->ieee->bg_band.channels);
out_remove_sysfs:
sysfs_remove_group(&pdev->dev.kobj, &ipw_attribute_group);
out_release_irq:
--
1.7.8.2
On Sun, 2012-01-22 at 04:32 +0100, Stefan Lippers-Hollmann wrote:
> Hi
>
> On Sunday 22 January 2012, Ben Hutchings wrote:
> > Currently cfg80211 fails to create a "phy80211" symlink in sysfs from
> > the net device to the wiphy device. The latter needs to be registered
> > first.
> >
> > Compile-tested only.
> >
> > Reported-by: Cesare Leonardi <[email protected]>
> > Signed-off-by: Ben Hutchings <[email protected]>
> > ---
> > drivers/net/wireless/ipw2x00/ipw2200.c | 23 ++++++++++++-----------
> > 1 files changed, 12 insertions(+), 11 deletions(-)
>
> I've tested this patch on my ipw2200 (inside an Acer TravelMate 292LMi).
> While the original
[...]
> is almost cosmetic, without any apparent ill-effects, applying that
> patch to kernel 3.2.1 (not exactly Debian's - e.g. I need acerhk/
> dritek to disable rfkill for that card, but it should be close enough)
> results in this trace:
>
> [ 5.708938] libipw: 802.11 data/management/control stack, git-1.1.13
> [ 5.708942] libipw: Copyright (C) 2004-2005 Intel Corporation <[email protected]>
> […]
> [ 5.785178] ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 1.2.2kmprq
> [ 5.785184] ipw2200: Copyright(c) 2003-2006 Intel Corporation
> [ 5.785323] ipw2200 0000:02:02.0: power state changed by ACPI to D0
> [ 5.785330] ipw2200 0000:02:02.0: power state changed by ACPI to D0
> [ 5.785344] ipw2200 0000:02:02.0: PCI INT A -> Link[LNKC] -> GSI 10 (level, low) -> IRQ 10
> [ 5.785370] ipw2200: Detected Intel PRO/Wireless 2200BG Network Connection
> [ 5.785416] ------------[ cut here ]------------
> [ 5.785435] WARNING: at /tmp/buildd/linux-aptosid-3.2/debian/build/source_i386_none/net/wireless/core.c:562 wiphy_register+0x45/0x38d [cfg80211]()
[...]
Which is:
if (!have_band) {
WARN_ON(1);
return -EINVAL;
}
and have_band is set if any element of wiphy->bands[] is non-NULL.
ipw_wdev_init() clearly does initialise wiphy->bands... except that it
seems to depend on the regulatory area having been determined already,
which I suppose must be done during the callback from netdev
registration (ipw_net_init()). Judging by the comment about this in
ipw2100.c, it's probably important that initialisation is done in this
order.
So maybe cfg80211 should instead tolerate registration in this order?
Ben.
--
Ben Hutchings
Knowledge is power. France is bacon.
Hi
On Sunday 22 January 2012, Ben Hutchings wrote:
> Currently cfg80211 fails to create a "phy80211" symlink in sysfs from
> the net device to the wiphy device. The latter needs to be registered
> first.
>
> Compile-tested only.
>
> Reported-by: Cesare Leonardi <[email protected]>
> Signed-off-by: Ben Hutchings <[email protected]>
> ---
> drivers/net/wireless/ipw2x00/ipw2200.c | 23 ++++++++++++-----------
> 1 files changed, 12 insertions(+), 11 deletions(-)
I've tested this patch on my ipw2200 (inside an Acer TravelMate 292LMi).
While the original
[ 5.525347] libipw: 802.11 data/management/control stack, git-1.1.13
[ 5.525352] libipw: Copyright (C) 2004-2005 Intel Corporation <[email protected]>
[…]
[ 5.787605] ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 1.2.2kmprq
[ 5.787610] ipw2200: Copyright(c) 2003-2006 Intel Corporation
[ 5.787742] ipw2200 0000:02:02.0: power state changed by ACPI to D0
[ 5.787749] ipw2200 0000:02:02.0: power state changed by ACPI to D0
[ 5.787762] ipw2200 0000:02:02.0: PCI INT A -> Link[LNKC] -> GSI 10 (level, low) -> IRQ 10
[ 5.787789] ipw2200: Detected Intel PRO/Wireless 2200BG Network Connection
[ 6.120089] cfg80211: failed to add phy80211 symlink to netdev!
[ 6.120373] ipw2200: Detected geography ZZG (13 802.11bg channels, 0 802.11a channels)
[ 6.138818] udevd[252]: renamed network interface eth1 to wlan0
[…]
[ 8.744754] acerhk: registered input device
[ 8.744878] input: Acer hotkey driver as /devices/virtual/input/input7
[ 8.747175] acerhk: forced laptop series to 290
[ 8.747184] acerhk: enabling dritek keyboard extension
[ 8.747704] acerhk: bios routine found at 0xc00fdf22
[ 8.747706] Acer Travelmate hotkey driver v0.5.35
[…]
[ 15.107990] lib80211_crypt: registered algorithm 'CCMP'
[ 15.276082] Intel AES-NI instructions are not detected.
[ 24.386037] wlan0: no IPv6 routers present
[wlan available]
is almost cosmetic, without any apparent ill-effects, applying that
patch to kernel 3.2.1 (not exactly Debian's - e.g. I need acerhk/
dritek to disable rfkill for that card, but it should be close enough)
results in this trace:
[ 5.708938] libipw: 802.11 data/management/control stack, git-1.1.13
[ 5.708942] libipw: Copyright (C) 2004-2005 Intel Corporation <[email protected]>
[…]
[ 5.785178] ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 1.2.2kmprq
[ 5.785184] ipw2200: Copyright(c) 2003-2006 Intel Corporation
[ 5.785323] ipw2200 0000:02:02.0: power state changed by ACPI to D0
[ 5.785330] ipw2200 0000:02:02.0: power state changed by ACPI to D0
[ 5.785344] ipw2200 0000:02:02.0: PCI INT A -> Link[LNKC] -> GSI 10 (level, low) -> IRQ 10
[ 5.785370] ipw2200: Detected Intel PRO/Wireless 2200BG Network Connection
[ 5.785416] ------------[ cut here ]------------
[ 5.785435] WARNING: at /tmp/buildd/linux-aptosid-3.2/debian/build/source_i386_none/net/wireless/core.c:562 wiphy_register+0x45/0x38d [cfg80211]()
[ 5.785440] Hardware name: TravelMate 290 \xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff\xffffffff
[ 5.785443] Modules linked in: ipw2200(+) iTCO_wdt libipw joydev drm snd_seq snd_timer snd_seq_device iTCO_vendor_support yenta_socket snd intel_agp i2c_i801 pcmcia_rsrc cfg80211 soundcore parport_pc psmouse parport rng_core snd_page_alloc serio_raw pcspkr i2c_algo_bit intel_gtt pcmcia_core evdev irda crc_ccitt rfkill lib80211 processor container ac battery shpchp pci_hotplug button ext4 mbcache jbd2 crc16 dm_mod sd_mod sr_mod crc_t10dif cdrom ata_generic pata_acpi ata_piix libata scsi_mod firewire_ohci firewire_core crc_itu_t 8139too 8139cp mii uhci_hcd ehci_hcd usbcore usb_common [last unloaded: scsi_wait_scan]
[ 5.785512] Pid: 328, comm: modprobe Not tainted 3.2-1.slh.4-aptosid-686 #1
[ 5.785515] Call Trace:
[ 5.785527] [<c012eaf4>] ? warn_slowpath_common+0x7c/0x8f
[ 5.785538] [<e0ff0b3e>] ? wiphy_register+0x45/0x38d [cfg80211]
[ 5.785549] [<e0ff0b3e>] ? wiphy_register+0x45/0x38d [cfg80211]
[ 5.785554] [<c012eb22>] ? warn_slowpath_null+0x1b/0x1f
[ 5.785565] [<e0ff0b3e>] ? wiphy_register+0x45/0x38d [cfg80211]
[ 5.785574] [<c01f89d7>] ? internal_create_group+0xf5/0xff
[ 5.785589] [<e0a2de1c>] ? ipw_pci_probe+0xa9a/0xbd0 [ipw2200]
[ 5.785597] [<c01519f4>] ? arch_local_irq_save+0xf/0x14
[ 5.785605] [<c0252986>] ? pci_device_probe+0x53/0x9a
[ 5.785612] [<c02c2820>] ? driver_probe_device+0x94/0x124
[ 5.785617] [<c0252871>] ? pci_match_id+0x15/0x34
[ 5.785622] [<c02c28f0>] ? __driver_attach+0x40/0x5b
[ 5.785626] [<c02c1d81>] ? bus_for_each_dev+0x37/0x60
[ 5.785631] [<c02c25aa>] ? driver_attach+0x17/0x1a
[ 5.785636] [<c02c28b0>] ? driver_probe_device+0x124/0x124
[ 5.785641] [<c02c22c4>] ? bus_add_driver+0x92/0x1d1
[ 5.785648] [<e099d000>] ? 0xe099cfff
[ 5.785652] [<c02c2cb8>] ? driver_register+0x7d/0xd4
[ 5.785661] [<c017cd50>] ? jump_label_module_notify+0xec/0x167
[ 5.785666] [<e099d000>] ? 0xe099cfff
[ 5.785670] [<c0253017>] ? __pci_register_driver+0x32/0x87
[ 5.785675] [<e099d000>] ? 0xe099cfff
[ 5.785687] [<e099d02e>] ? ipw_init+0x2e/0x72 [ipw2200]
[ 5.785693] [<c0101173>] ? do_one_initcall+0x7d/0x132
[ 5.785699] [<c0145016>] ? __blocking_notifier_call_chain+0x47/0x4f
[ 5.785705] [<c0154a73>] ? sys_init_module+0x13a4/0x159c
[ 5.785718] [<c03a639f>] ? sysenter_do_call+0x12/0x28
[ 5.785722] ---[ end trace d5a4597cc4a1d28f ]---
[ 5.785724] ipw2200: failed to register wireless device
[ 5.785795] ipw2200 0000:02:02.0: PCI INT A disabled
[ 5.785820] ipw2200: probe of 0000:02:02.0 failed with error -5
[acerhk is loaded much later than ipw2200 through /etc/modules, so it
can't interfere at this stage]
I can try to give it some further checking tomorrow.
Regards
Stefan Lippers-Hollmann
Currently cfg80211 fails to create a "phy80211" symlink in sysfs from
the net device to the wiphy device. The latter needs to be registered
first.
Compile-tested only.
Signed-off-by: Ben Hutchings <[email protected]>
---
This appears to have the same bug as ipw2200.
Ben.
drivers/net/wireless/ipw2x00/ipw2100.c | 24 +++++++++++++-----------
1 files changed, 13 insertions(+), 11 deletions(-)
diff --git a/drivers/net/wireless/ipw2x00/ipw2100.c b/drivers/net/wireless/ipw2x00/ipw2100.c
index a0e5c21..a5437e7 100644
--- a/drivers/net/wireless/ipw2x00/ipw2100.c
+++ b/drivers/net/wireless/ipw2x00/ipw2100.c
@@ -1956,10 +1956,8 @@ static int ipw2100_wdev_init(struct net_device *dev)
}
set_wiphy_dev(wdev->wiphy, &priv->pci_dev->dev);
- if (wiphy_register(wdev->wiphy)) {
- ipw2100_down(priv);
+ if (wiphy_register(wdev->wiphy))
return -EIO;
- }
return 0;
}
@@ -6337,6 +6335,11 @@ static int ipw2100_pci_init_one(struct pci_dev *pci_dev,
printk(KERN_INFO DRV_NAME
": Detected Intel PRO/Wireless 2100 Network Connection\n");
+ err = ipw2100_wdev_init(dev);
+ if (err)
+ goto fail;
+ registered = 1;
+
/* Bring up the interface. Pre 0.46, after we registered the
* network device we would call ipw2100_up. This introduced a race
* condition with newer hotplug configurations (network was coming
@@ -6353,11 +6356,7 @@ static int ipw2100_pci_init_one(struct pci_dev *pci_dev,
"Error calling register_netdev.\n");
goto fail;
}
- registered = 1;
-
- err = ipw2100_wdev_init(dev);
- if (err)
- goto fail;
+ registered = 2;
mutex_lock(&priv->action_mutex);
@@ -6396,13 +6395,16 @@ static int ipw2100_pci_init_one(struct pci_dev *pci_dev,
fail_unlock:
mutex_unlock(&priv->action_mutex);
- wiphy_unregister(priv->ieee->wdev.wiphy);
- kfree(priv->ieee->bg_band.channels);
fail:
if (dev) {
- if (registered)
+ if (registered >= 2)
unregister_netdev(dev);
+ if (registered) {
+ wiphy_unregister(priv->ieee->wdev.wiphy);
+ kfree(priv->ieee->bg_band.channels);
+ }
+
ipw2100_hw_stop_adapter(priv);
ipw2100_disable_interrupts(priv);
--
1.7.8.2
Hi Ben
On Sun, Jan 22, 2012 at 03:09:35AM +0000, Ben Hutchings wrote:
> Currently cfg80211 fails to create a "phy80211" symlink in sysfs from
> the net device to the wiphy device. The latter needs to be registered
> first.
>
> Compile-tested only.
>
> Reported-by: Cesare Leonardi <[email protected]>
> Signed-off-by: Ben Hutchings <[email protected]>
This problem was created by my changes (commit
7cabafcea793c003503a118da58da358b0692930). I was aware of remove
symlink, and I'm sorry for that, but I think this symlink is not
necessary, i.e: dereferencing device/ieee80211/phy0 give the same
data:
[root@dhcp-27-172 wlan0]# readlink -f phy80211
/sys/devices/pci0000:00/0000:00:1c.1/0000:03:00.0/ieee80211/phy0
[root@dhcp-27-172 wlan0]# readlink -f device
/sys/devices/pci0000:00/0000:00:1c.1/0000:03:00.0
Of course user space application may relay on the link, but well
fixing that in kernel is not so easy as it looks at first glace.
Reverting commits causing this problem, will remove fix for
another quite serious issue. Changing ipw2x00 or mac80211 is
risky, as that initalization code is kinda fragile. The best way
for fix, is change applications that relay on that link.
Thanks
Stanislaw