2 compile errors: rt2x00firmware.c and mesh.c

2 compile errors: rt2x00firmware.c and mesh.c

[Franz Apeltauer]

Hi,

after downloading the latest
http://linuxwireless.org/download/compat-wireless-2.6/compat-wireless-2=
=2E6.tar.bz2
today (2.3.2008) I got these errors during make:

   CC [M]
=2E../compat-wireless-2.6/drivers/net/wireless/rt2x00/rt2x00firmware.o
=2E../compat-wireless-2.6/drivers/net/wireless/rt2x00/rt2x00firmware.c:=
116:
=46ehler: Redefinition von =BBrt2x00lib_load_firmware=AB
=3D Error: redefiniton of  =BBrt2x00lib_load_firmware=AB
=2E../compat-wireless-2.6/drivers/net/wireless/rt2x00/rt2x00lib.h:129:
=46ehler: Vorherige Definition von =BBrt2x00lib_load_firmware=AB war hi=
er
=3D Error: previous definition of  =BBrt2x00lib_load_firmware=AB was he=
re
=2E../compat-wireless-2.6/drivers/net/wireless/rt2x00/rt2x00firmware.c:=
138:
=46ehler: Redefinition von =BBrt2x00lib_free_firmware=AB
=2E../compat-wireless-2.6/drivers/net/wireless/rt2x00/rt2x00lib.h:133:
=46ehler: Vorherige Definition von =BBrt2x00lib_free_firmware=AB war hi=
er
make[4]: ***
[.../compat-wireless-2.6/drivers/net/wireless/rt2x00/rt2x00firmware.o]
=46ehler 1

My environment (Ubuntu gutsy):
kernel: 2.6.22-14-generic #1 SMP Tue Feb 12 07:42:25 UTC 2008 i686 GNU/=
Linux
cc: gcc-Version 4.1.3 20070929 (prerelease) (Ubuntu 4.1.2-16ubuntu2)
wlan card: MSI PC54G2 ... RaLink rt2500 chip


When I tried to "export CONFIG_CRC_ITU_T=3Dy" before "make",=20
rt2x00firmware.c compiled fine. (Probably this is not a good idea. I=20
just wanted to see, if it compiles then.)

-----------------
=2E.. but a bit later I got this error:

   CC [M]  .../compat-wireless-2.6/net/mac80211/mesh.o
=2E../compat-wireless-2.6/net/mac80211/mesh.c: In Funktion =BBieee80211=
s_init=AB:
=2E../compat-wireless-2.6/net/mac80211/mesh.c:28:
=46ehler: Zu wenige Argumente f=FCr Funktion =BBkmem_cache_create=AB
=3D Error: Not enough arguments for function =BBkmem_cache_create=AB

In /usr/src/linux-headers-2.6.22-14/include/linux/slab.h I found:
    struct kmem_cache *kmem_cache_create(const char *, size_t, size_t,
		unsigned long,
		void (*)(void *, struct kmem_cache *, unsigned long),
		void (*)(void *, struct kmem_cache *, unsigned long));
i.e. the old, 6-parameter version.

So, IMHO the #define kmem_cache_create in
- compat-wireless-2.6/include/net/compat.h, line 120 and
- compat-wireless-2.6/compat/compat.h, line 120
should be moved 2 lines downwards into the block named
/* Compat work for 2.6.22 and 2.6.23 */
Then it compiles fine.

Best regards,
=46ranz
--
To unsubscribe from this list: send the line "unsubscribe linux-wireles=
s" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 1338 bytes --]


>    CC [M]
> .../compat-wireless-2.6/drivers/net/wireless/rt2x00/rt2x00firmware.o
> .../compat-wireless-2.6/drivers/net/wireless/rt2x00/rt2x00firmware.c:116:
> Fehler: Redefinition von »rt2x00lib_load_firmware«
> = Error: redefiniton of  »rt2x00lib_load_firmware«
> 
> When I tried to "export CONFIG_CRC_ITU_T=y" before "make", 
> rt2x00firmware.c compiled fine. (Probably this is not a good idea. I 
> just wanted to see, if it compiles then.)

That's odd because the latest config.mk should check for CRC_ITU_T
before... Ivo could there be a problem with rt2x00 when crc-itu-t isn't
available?

> In /usr/src/linux-headers-2.6.22-14/include/linux/slab.h I found:
>     struct kmem_cache *kmem_cache_create(const char *, size_t, size_t,
> 		unsigned long,
> 		void (*)(void *, struct kmem_cache *, unsigned long),
> 		void (*)(void *, struct kmem_cache *, unsigned long));
> i.e. the old, 6-parameter version.
> 
> So, IMHO the #define kmem_cache_create in
> - compat-wireless-2.6/include/net/compat.h, line 120 and
> - compat-wireless-2.6/compat/compat.h, line 120
> should be moved 2 lines downwards into the block named
> /* Compat work for 2.6.22 and 2.6.23 */
> Then it compiles fine.

Ok, indeed, it should be moved then, I was working on .21 when I noticed
that needed to be done.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 828 bytes --]

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Luis R. Rodriguez]

On Mon, Mar 03, 2008 at 01:32:11AM +0100, Johannes Berg wrote:
>=20
> >    CC [M]
> > .../compat-wireless-2.6/drivers/net/wireless/rt2x00/rt2x00firmware.=
o
> > .../compat-wireless-2.6/drivers/net/wireless/rt2x00/rt2x00firmware.=
c:116:
> > Fehler: Redefinition von =BBrt2x00lib_load_firmware=AB
> > =3D Error: redefiniton of  =BBrt2x00lib_load_firmware=AB
> >=20
> > When I tried to "export CONFIG_CRC_ITU_T=3Dy" before "make",=20
> > rt2x00firmware.c compiled fine. (Probably this is not a good idea. =
I=20
> > just wanted to see, if it compiles then.)
>=20
> That's odd because the latest config.mk should check for CRC_ITU_T
> before... Ivo could there be a problem with rt2x00 when crc-itu-t isn=
't
> available?

=46or now I'll disable rt2x00 in compat-wireless.

> > In /usr/src/linux-headers-2.6.22-14/include/linux/slab.h I found:
> >     struct kmem_cache *kmem_cache_create(const char *, size_t, size=
_t,
> > 		unsigned long,
> > 		void (*)(void *, struct kmem_cache *, unsigned long),
> > 		void (*)(void *, struct kmem_cache *, unsigned long));
> > i.e. the old, 6-parameter version.
> >=20
> > So, IMHO the #define kmem_cache_create in
> > - compat-wireless-2.6/include/net/compat.h, line 120 and
> > - compat-wireless-2.6/compat/compat.h, line 120
> > should be moved 2 lines downwards into the block named
> > /* Compat work for 2.6.22 and 2.6.23 */
> > Then it compiles fine.
>=20
> Ok, indeed, it should be moved then, I was working on .21 when I noti=
ced
> that needed to be done.

OK -- this required a just slightly different approach to handle the
different kernel versions. Below is the approach I took. If you
apply this patch manually don't forget to

cp compat/compat.h include/net/compat.h

as this is what the autogeneration tool does.

=46orced a new tarball out with new changes applied, would appreciate
some testing on 2.6.21, 2.6.22. You can get the tarball from the
usual place:

http://linuxwireless.org/download/compat-wireless-2.6/compat-wireless-2=
=2E6.tar.bz2

=46or those cloning please pull latest changes in.

diff --git a/compat/compat.h b/compat/compat.h
index ec966de..4ae26ac 100644
--- a/compat/compat.h
+++ b/compat/compat.h
@@ -130,23 +130,36 @@ static inline void pci_clear_mwi(struct pci_dev *=
dev)
 #define list_first_entry(ptr, type, member) \
         list_entry((ptr)->next, type, member)
=20
+#endif /* (LINUX_VERSION_CODE < KERNEL_VERSION(2,6,22)) */
+
+/* Compat work for 2.6.21, 2.6.22 and 2.6.23 */
+#if (LINUX_VERSION_CODE < KERNEL_VERSION(2,6,24))
+
 /*
  * Force link bug if constructor is used, can't be done compatibly
  * because constructor arguments were swapped since then!
  */
 extern void __incompatible_kmem_cache_create(void);
=20
+/* 2.6.21 and 2.6.22 kmem_cache_create() takes 6 arguments */
+#if (LINUX_VERSION_CODE < KERNEL_VERSION(2,6,23))
 #define kmem_cache_create(name, objsize, align, flags, ctor) 	\
 	({							\
 		if (ctor) __incompatible_kmem_cache_create();	\
 		kmem_cache_create((name), (objsize), (align),	\
 				  (flags), NULL, NULL);		\
 	})
-
 #endif
=20
-/* Compat work for 2.6.22 and 2.6.23 */
-#if (LINUX_VERSION_CODE < KERNEL_VERSION(2,6,24))
+/* 2.6.23 kmem_cache_create() takes 5 arguments */
+#if (LINUX_VERSION_CODE =3D=3D KERNEL_VERSION(2,6,23))
+#define kmem_cache_create(name, objsize, align, flags, ctor) 	\
+	({							\
+		if (ctor) __incompatible_kmem_cache_create();	\
+		kmem_cache_create((name), (objsize), (align),	\
+				  (flags), NULL);		\
+	})
+#endif
=20
 /* From include/linux/mod_devicetable.h */
=20
@@ -254,7 +267,7 @@ int compat_is_pcie(struct pci_dev *pdev);
=20
 #endif /* (LINUX_VERSION_CODE < KERNEL_VERSION(2,6,24)) */
=20
-/* Compat work for 2.6.22 */
+/* Compat work for kernels <=3D 2.6.22 */
 #if (LINUX_VERSION_CODE < KERNEL_VERSION(2,6,23))
=20
 /* dev_mc_list was replaced with dev_addr_list as of 2.6.23 */
--
To unsubscribe from this list: send the line "unsubscribe linux-wireles=
s" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Franz Apeltauer]

Luis R. Rodriguez schrieb:
> Forced a new tarball out with new changes applied, would appreciate
> some testing on 2.6.21, 2.6.22. You can get the tarball from the
> usual place:
> http://linuxwireless.org/download/compat-wireless-2.6/compat-wireless-2.6.tar.bz2

Thank you very much!
Now everything compiles fine on 2.6.22.

I cannot send you test results yet. It looks like I have to read some 
more documentation first. For now I can only tell you this:

a) Test on a PC with a Broadcom 4306 rev 03:
- Downloaded and installed the firmware.
- The module b43legacy loads without errors, but nothing else happens.
   ifconfig doesn't show any new interface, like eth0 or wlan0.
# modprobe b43legacy
=> messages: Mar  5 23:14:02 af-nb-3 kernel: [10519.728000] Broadcom 
43xx driver loaded [ Features: PID, Firmware-ID: FW10 ]
=> dmesg: [10519.728000] Broadcom 43xx driver loaded [ Features: PID, 
Firmware-ID: FW10 ]

Maybe this is important: dmesg says "Firmware-ID: FW10", but there is no 
ucode10.fw in /lib/firmware/b43legacy, only 2, 4, 5, 11 (downloaded 
today, as described in the wiki).

# lsmod | grep 43
b43legacy             102040  0
ssb                    36484  1 b43legacy
mac80211              220020  1 b43legacy

# lspci
00:00.0 Host bridge: VIA Technologies, Inc. VT8378 [KM400/A] Chipset 
Host Bridge
00:01.0 PCI bridge: VIA Technologies, Inc. VT8237 PCI Bridge
00:09.0 Network controller: Broadcom Corporation BCM4306 802.11b/g 
Wireless LAN Controller (rev 03)
...


b) Test on a PC with MSI PC54G2 ... RaLink rt2500
You wrote "For now I'll disable rt2x00 in compat-wireless."
So, the result of this test seems to be unimportant (it tried to load 
the old driver).


If you would like to see the output of any test, that I can do for you, 
just let me know.

Greetings,
Franz

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 278 bytes --]


> a) Test on a PC with a Broadcom 4306 rev 03:
> - Downloaded and installed the firmware.
> - The module b43legacy loads without errors, but nothing else happens.
>    ifconfig doesn't show any new interface, like eth0 or wlan0.

Well, yeah, you need b43...

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 828 bytes --]

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Franz Apeltauer]

Johannes Berg schrieb:
>> a) Test on a PC with a Broadcom 4306 rev 03:
>> - Downloaded and installed the firmware.
>> - The module b43legacy loads without errors, but nothing else happens.
>>    ifconfig doesn't show any new interface, like eth0 or wlan0.
> 
> Well, yeah, you need b43...
> 
> johannes

I'm sorry, almost the same result:
# rmmod b43legacy
# modprobe b43
=> messages: Mar  6 00:18:36 af-nb-3 kernel: [14393.692000] Broadcom 
43xx driver loaded [ Features: PML, Firmware-ID: FW13 ]
=> dmesg: [14393.692000] Broadcom 43xx driver loaded [ Features: PML, 
Firmware-ID: FW13 ]

Again: No new interfaces.

Franz

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 425 bytes --]


> I'm sorry, almost the same result:
> # rmmod b43legacy
> # modprobe b43
> => messages: Mar  6 00:18:36 af-nb-3 kernel: [14393.692000] Broadcom 
> 43xx driver loaded [ Features: PML, Firmware-ID: FW13 ]
> => dmesg: [14393.692000] Broadcom 43xx driver loaded [ Features: PML, 
> Firmware-ID: FW13 ]

Huh, ok, I see.

Edit compat's compat.mk and add

SSB_B43_PCI_BRIDGE=y

to the end and rebuild.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 828 bytes --]

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Franz Apeltauer]

[-- Attachment #1: Type: text/plain, Size: 581 bytes --]

Johannes Berg schrieb:
>> I'm sorry, almost the same result:
>> # rmmod b43legacy
>> # modprobe b43
>> => messages: Mar  6 00:18:36 af-nb-3 kernel: [14393.692000] Broadcom 
>> 43xx driver loaded [ Features: PML, Firmware-ID: FW13 ]
>> => dmesg: [14393.692000] Broadcom 43xx driver loaded [ Features: PML, 
>> Firmware-ID: FW13 ]
> 
> Huh, ok, I see.
> 
> Edit compat's compat.mk and add
> 
> SSB_B43_PCI_BRIDGE=y
> 
> to the end and rebuild.
> 
> johannes

Loaded today's tarball (10.3.).
Again: no new devices.
I cannot see, what's wrong. (Attachment: "make load", dmesg).

Franz

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: load.log --]
[-- Type: text/x-log; name="load.log", Size: 7504 bytes --]

# make load
Loading ipw2100...
Loading ipw2200...
Loading libertas_cs...
Loading usb8xxx...
Loading p54pci...
Loading p54usb...
Loading adm8211...
Loading zd1211rw...
Loading rtl8180...
Loading rtl8187...
Loading p54pci...
Loading p54usb...
Loading iwl3945...
Loading iwl4965...
Loading rtl8180...
Loading rtl8187...
Loading rtl8180...
Loading rtl8187...
Loading rt2400pci...
Loading rt2500pci...
Loading rt61pci...
Loading rt2500usb...
WARNING: Error inserting rt2x00usb (/lib/modules/2.6.22-14-generic/ubuntu/wireless/rt2x00/rt2x00usb.ko): Unknown symbol in module, or unknown parameter (see dmesg)
FATAL: Error inserting rt2500usb (/lib/modules/2.6.22-14-generic/updates/drivers/net/wireless/rt2x00/rt2500usb.ko): Unknown symbol in module, or unknown parameter (see dmesg)
Loading rt73usb...
WARNING: Error inserting rt2x00usb (/lib/modules/2.6.22-14-generic/ubuntu/wireless/rt2x00/rt2x00usb.ko): Unknown symbol in module, or unknown parameter (see dmesg)
FATAL: Error inserting rt73usb (/lib/modules/2.6.22-14-generic/updates/drivers/net/wireless/rt2x00/rt73usb.ko): Unknown symbol in module, or unknown parameter (see dmesg)
Loading rndis_wlan...
Module ath_pci not detected -- this is fine
ath5k loaded successfully
Module bcm43xx not detected -- this is fine
b43 loaded successfully
b43legacy loaded successfully

# dmesg -c
[ 3862.036000] ieee80211_crypt: registered algorithm 'NULL'
[ 3862.044000] ieee80211: 802.11 data/management/control stack, git-1.1.13
[ 3862.044000] ieee80211: Copyright (C) 2004-2005 Intel Corporation <jketreno@linux.intel.com>
[ 3862.052000] ipw2100: Intel(R) PRO/Wireless 2100 Network Driver, git-1.2.2
[ 3862.052000] ipw2100: Copyright(c) 2003-2006 Intel Corporation
[ 3862.116000] ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 1.2.2kmprq
[ 3862.116000] ipw2200: Copyright(c) 2003-2006 Intel Corporation
[ 3862.600000] usbcore: registered new interface driver usb8xxx
[ 3862.692000] usbcore: registered new interface driver prism54usb
[ 3862.804000] usbcore: registered new interface driver zd1211rw
[ 3862.904000] usbcore: registered new interface driver rtl8187
[ 3862.964000] iwl3945: Intel(R) PRO/Wireless 3945ABG/BG Network Connection driver for Linux, 1.2.26k
[ 3862.964000] iwl3945: Copyright(c) 2003-2007 Intel Corporation
[ 3863.008000] iwl4965: Intel(R) Wireless WiFi Link 4965AGN driver for Linux, 1.2.26k
[ 3863.008000] iwl4965: Copyright(c) 2003-2008 Intel Corporation
[ 3863.200000] rt2x00usb: disagrees about version of symbol rt2x00lib_suspend
[ 3863.200000] rt2x00usb: Unknown symbol rt2x00lib_suspend
[ 3863.200000] rt2x00usb: disagrees about version of symbol rt2x00lib_probe_dev
[ 3863.200000] rt2x00usb: Unknown symbol rt2x00lib_probe_dev
[ 3863.200000] rt2x00usb: disagrees about version of symbol ieee80211_free_hw
[ 3863.200000] rt2x00usb: Unknown symbol ieee80211_free_hw
[ 3863.200000] rt2x00usb: disagrees about version of symbol ieee80211_alloc_hw
[ 3863.200000] rt2x00usb: Unknown symbol ieee80211_alloc_hw
[ 3863.200000] rt2x00usb: disagrees about version of symbol ieee80211_wake_queue
[ 3863.200000] rt2x00usb: Unknown symbol ieee80211_wake_queue
[ 3863.200000] rt2x00usb: disagrees about version of symbol rt2x00lib_rxdone
[ 3863.200000] rt2x00usb: Unknown symbol rt2x00lib_rxdone
[ 3863.200000] rt2x00usb: disagrees about version of symbol rt2x00lib_remove_dev
[ 3863.200000] rt2x00usb: Unknown symbol rt2x00lib_remove_dev
[ 3863.200000] rt2x00usb: disagrees about version of symbol rt2x00lib_txdone
[ 3863.200000] rt2x00usb: Unknown symbol rt2x00lib_txdone
[ 3863.200000] rt2x00usb: disagrees about version of symbol rt2x00lib_write_tx_desc
[ 3863.200000] rt2x00usb: Unknown symbol rt2x00lib_write_tx_desc
[ 3863.200000] rt2x00usb: disagrees about version of symbol ieee80211_stop_queue
[ 3863.200000] rt2x00usb: Unknown symbol ieee80211_stop_queue
[ 3863.200000] rt2x00usb: disagrees about version of symbol rt2x00lib_resume
[ 3863.200000] rt2x00usb: Unknown symbol rt2x00lib_resume
[ 3863.232000] rt2500usb: Unknown symbol rt2x00usb_init_rxentry
[ 3863.232000] rt2500usb: Unknown symbol rt2x00usb_disable_radio
[ 3863.232000] rt2500usb: Unknown symbol rt2x00usb_init_txentry
[ 3863.232000] rt2500usb: Unknown symbol rt2x00usb_vendor_request_buff
[ 3863.232000] rt2500usb: Unknown symbol rt2x00usb_write_tx_data
[ 3863.232000] rt2500usb: Unknown symbol rt2x00usb_vendor_request
[ 3863.232000] rt2500usb: Unknown symbol rt2x00usb_probe
[ 3863.232000] rt2500usb: Unknown symbol rt2x00usb_suspend
[ 3863.236000] rt2500usb: Unknown symbol rt2x00usb_disconnect
[ 3863.236000] rt2500usb: Unknown symbol rt2x00usb_vendor_req_buff_lock
[ 3863.236000] rt2500usb: Unknown symbol rt2x00usb_resume
[ 3863.236000] rt2500usb: Unknown symbol rt2x00usb_uninitialize
[ 3863.236000] rt2500usb: Unknown symbol rt2x00usb_initialize
[ 3863.244000] rt2x00usb: disagrees about version of symbol rt2x00lib_suspend
[ 3863.244000] rt2x00usb: Unknown symbol rt2x00lib_suspend
[ 3863.244000] rt2x00usb: disagrees about version of symbol rt2x00lib_probe_dev
[ 3863.244000] rt2x00usb: Unknown symbol rt2x00lib_probe_dev
[ 3863.244000] rt2x00usb: disagrees about version of symbol ieee80211_free_hw
[ 3863.244000] rt2x00usb: Unknown symbol ieee80211_free_hw
[ 3863.244000] rt2x00usb: disagrees about version of symbol ieee80211_alloc_hw
[ 3863.244000] rt2x00usb: Unknown symbol ieee80211_alloc_hw
[ 3863.244000] rt2x00usb: disagrees about version of symbol ieee80211_wake_queue
[ 3863.244000] rt2x00usb: Unknown symbol ieee80211_wake_queue
[ 3863.244000] rt2x00usb: disagrees about version of symbol rt2x00lib_rxdone
[ 3863.244000] rt2x00usb: Unknown symbol rt2x00lib_rxdone
[ 3863.244000] rt2x00usb: disagrees about version of symbol rt2x00lib_remove_dev
[ 3863.244000] rt2x00usb: Unknown symbol rt2x00lib_remove_dev
[ 3863.244000] rt2x00usb: disagrees about version of symbol rt2x00lib_txdone
[ 3863.244000] rt2x00usb: Unknown symbol rt2x00lib_txdone
[ 3863.244000] rt2x00usb: disagrees about version of symbol rt2x00lib_write_tx_desc
[ 3863.244000] rt2x00usb: Unknown symbol rt2x00lib_write_tx_desc
[ 3863.244000] rt2x00usb: disagrees about version of symbol ieee80211_stop_queue
[ 3863.244000] rt2x00usb: Unknown symbol ieee80211_stop_queue
[ 3863.244000] rt2x00usb: disagrees about version of symbol rt2x00lib_resume
[ 3863.244000] rt2x00usb: Unknown symbol rt2x00lib_resume
[ 3863.252000] rt73usb: Unknown symbol rt2x00usb_init_rxentry
[ 3863.252000] rt73usb: Unknown symbol rt2x00usb_disable_radio
[ 3863.252000] rt73usb: Unknown symbol rt2x00usb_init_txentry
[ 3863.252000] rt73usb: Unknown symbol rt2x00usb_vendor_request_buff
[ 3863.252000] rt73usb: Unknown symbol rt2x00usb_write_tx_data
[ 3863.252000] rt73usb: Unknown symbol rt2x00usb_vendor_request
[ 3863.252000] rt73usb: Unknown symbol rt2x00usb_probe
[ 3863.256000] rt73usb: Unknown symbol rt2x00usb_suspend
[ 3863.256000] rt73usb: Unknown symbol rt2x00usb_disconnect
[ 3863.256000] rt73usb: Unknown symbol rt2x00usb_vendor_req_buff_lock
[ 3863.256000] rt73usb: Unknown symbol rt2x00usb_resume
[ 3863.256000] rt73usb: Unknown symbol rt2x00usb_uninitialize
[ 3863.256000] rt73usb: Unknown symbol rt2x00usb_initialize
[ 3863.320000] usbcore: registered new interface driver cdc_ether
[ 3863.332000] usbcore: registered new interface driver rndis_host
[ 3863.340000] usbcore: registered new interface driver rndis_wlan
[ 3864.096000] Broadcom 43xx driver loaded [ Features: PML, Firmware-ID: FW13 ]
[ 3864.116000] Broadcom 43xx-legacy driver loaded [ Features: PID, Firmware-ID: FW10 ]
#

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Luis R. Rodriguez]

On Wed, Mar 5, 2008 at 6:12 PM, Franz Apeltauer <franz.apeltauer@gmx.at> wrote:
> Luis R. Rodriguez schrieb:
>
> > Forced a new tarball out with new changes applied, would appreciate
>  > some testing on 2.6.21, 2.6.22. You can get the tarball from the
>  > usual place:
>  > http://linuxwireless.org/download/compat-wireless-2.6/compat-wireless-2.6.tar.bz2
>
>  Thank you very much!
>  Now everything compiles fine on 2.6.22.
>
>  I cannot send you test results yet. It looks like I have to read some
>  more documentation first. For now I can only tell you this:
>
>  a) Test on a PC with a Broadcom 4306 rev 03:
>  - Downloaded and installed the firmware.
>  - The module b43legacy loads without errors, but nothing else happens.
>    ifconfig doesn't show any new interface, like eth0 or wlan0.
>  # modprobe b43legacy
>  => messages: Mar  5 23:14:02 af-nb-3 kernel: [10519.728000] Broadcom
>  43xx driver loaded [ Features: PID, Firmware-ID: FW10 ]
>  => dmesg: [10519.728000] Broadcom 43xx driver loaded [ Features: PID,
>  Firmware-ID: FW10 ]
>
>  Maybe this is important: dmesg says "Firmware-ID: FW10", but there is no
>  ucode10.fw in /lib/firmware/b43legacy, only 2, 4, 5, 11 (downloaded
>  today, as described in the wiki).
>
>  # lsmod | grep 43
>  b43legacy             102040  0
>  ssb                    36484  1 b43legacy
>  mac80211              220020  1 b43legacy
>
>  # lspci
>  00:00.0 Host bridge: VIA Technologies, Inc. VT8378 [KM400/A] Chipset
>  Host Bridge
>  00:01.0 PCI bridge: VIA Technologies, Inc. VT8237 PCI Bridge
>  00:09.0 Network controller: Broadcom Corporation BCM4306 802.11b/g
>  Wireless LAN Controller (rev 03)
>  ...
>
>
>  b) Test on a PC with MSI PC54G2 ... RaLink rt2500
>  You wrote "For now I'll disable rt2x00 in compat-wireless."
>  So, the result of this test seems to be unimportant (it tried to load
>  the old driver).

You can test this now, I've enabled rt2x00 drivers by fixing
dependencies on config.mk. Forced re-generation of new tarball based
on latest updates too.

  Luis

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Franz Apeltauer]

Luis R. Rodriguez schrieb:
>>  b) Test on a PC with MSI PC54G2 ... RaLink rt2500
> 
> You can test this now, I've enabled rt2x00 drivers by fixing
> dependencies on config.mk. Forced re-generation of new tarball based
> on latest updates too.
> 
>   Luis

Loaded today's tarball (10.3.).
The "redefinition of rt2x00lib_load_firmware" error is back.

I believe, that there should be 2 additional lines in rt2x00firmware.c:
a) line #115, before "int rt2x00lib_load_firmware":
     #ifdef CONFIG_RT2X00_LIB_FIRMWARE
b) at the end of the file:
     #endif /* CONFIG_RT2X00_LIB_FIRMWARE */
to be consistent with #ifdef/#else/#endif in rt2x00lib.h lines 124, 127, 
135.
Then it compiles fine. install, unload, load.
... and now I see wlan0 and wmaster0!  Yippee!

Thank you very much!
Franz

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Ivo van Doorn]

On Monday 10 March 2008, Franz Apeltauer wrote:
> Luis R. Rodriguez schrieb:
> >>  b) Test on a PC with MSI PC54G2 ... RaLink rt2500
> > 
> > You can test this now, I've enabled rt2x00 drivers by fixing
> > dependencies on config.mk. Forced re-generation of new tarball based
> > on latest updates too.
> > 
> >   Luis
> 
> Loaded today's tarball (10.3.).
> The "redefinition of rt2x00lib_load_firmware" error is back.
> 
> I believe, that there should be 2 additional lines in rt2x00firmware.c:
> a) line #115, before "int rt2x00lib_load_firmware":
>      #ifdef CONFIG_RT2X00_LIB_FIRMWARE
> b) at the end of the file:
>      #endif /* CONFIG_RT2X00_LIB_FIRMWARE */
> to be consistent with #ifdef/#else/#endif in rt2x00lib.h lines 124, 127, 
> 135.
> Then it compiles fine. install, unload, load.
> ... and now I see wlan0 and wmaster0!  Yippee!

How about not linking rt2x00firmware.c at all when RT2X00_LIB_FIRMWARE
is not set? That should be a clean solution that can be made in the Makefile...

Ivo

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Luis R. Rodriguez]

On Mon, Mar 10, 2008 at 11:21:35PM +0100, Ivo van Doorn wrote:
> On Monday 10 March 2008, Franz Apeltauer wrote:
> > Luis R. Rodriguez schrieb:
> > >>  b) Test on a PC with MSI PC54G2 ... RaLink rt2500
> > > 
> > > You can test this now, I've enabled rt2x00 drivers by fixing
> > > dependencies on config.mk. Forced re-generation of new tarball based
> > > on latest updates too.
> > > 
> > >   Luis
> > 
> > Loaded today's tarball (10.3.).
> > The "redefinition of rt2x00lib_load_firmware" error is back.
> > 
> > I believe, that there should be 2 additional lines in rt2x00firmware.c:
> > a) line #115, before "int rt2x00lib_load_firmware":
> >      #ifdef CONFIG_RT2X00_LIB_FIRMWARE
> > b) at the end of the file:
> >      #endif /* CONFIG_RT2X00_LIB_FIRMWARE */
> > to be consistent with #ifdef/#else/#endif in rt2x00lib.h lines 124, 127, 
> > 135.
> > Then it compiles fine. install, unload, load.
> > ... and now I see wlan0 and wmaster0!  Yippee!
> 
> How about not linking rt2x00firmware.c at all when RT2X00_LIB_FIRMWARE
> is not set? That should be a clean solution that can be made in the Makefile...

Actually Ivo, I think you're fine, this is an issue with
compat-wireless. I've applied the patch below to correct this.

The issue seems to have come from scripts/gen-compat-autoconf.sh, in the 
the "skip vars that weren't actually set due to dependencies"
section.

Frankz, I generated a new tarball with this applied. You can either
patch your local tarball with this patch or just download the
tarball again.

  Luis

commit d4097ab3593a721fc1d99823d2442464debf7f09
Author: Luis R. Rodriguez <mcgrof@winlab.rutgers.edu>
Date:   Mon Mar 10 19:20:33 2008 -0400

    Fix rt2x00 firmware build issue, fix typo with IPW2200 radiotap options
    
    Signed-off-by: Luis R. Rodriguez <mcgrof@winlab.rutgers.edu>

diff --git a/config.mk b/config.mk
index a146854..b9d537d 100644
--- a/config.mk
+++ b/config.mk
@@ -59,11 +59,11 @@ CONFIG_B43LEGACY_DMA_AND_PIO_MODE=y
 
 # The Intel ipws
 CONFIG_IPW2100=m
-IPW2100_MONITOR=y
+CONFIG_IPW2100_MONITOR=y
 CONFIG_IPW2200=m
-IPW2200_MONITOR=y
-IPW2200_RADIOTAP=y
-IPW2200_PROMISCUOUS=y
+CONFIG_IPW2200_MONITOR=y
+CONFIG_IPW2200_RADIOTAP=y
+CONFIG_IPW2200_PROMISCUOUS=y
 # The above enables use a second interface prefixed 'rtap'.
 #           Example usage:
 #
@@ -76,7 +76,7 @@ IPW2200_PROMISCUOUS=y
 # it on via sysfs:
 #
 # % echo 1 > /sys/bus/pci/drivers/ipw2200/*/rtap_iface
-IPW2200_QOS=y
+CONFIG_IPW2200_QOS=y
 
 NEED_IEEE80211=y
 
@@ -113,9 +113,12 @@ CONFIG_EEPROM_93CX6=m
 ifneq ($(CONFIG_USB),)
 CONFIG_ZD1211RW=m
 
-# Sorry, it uses cancel_work_sync which is new and can't be done in compat...
+# Stuff here things which depend on kernel versions for USB
+ifeq ($(shell test -e $(KLIB_BUILD)/Makefile && echo yes),yes)
 ifeq ($(shell test $(shell sed 's/^SUBLEVEL = //;t;d' < $(KLIB_BUILD)/Makefile) -gt 21 && echo yes),yes)
 
+# Sorry, rndis_wlan uses cancel_work_sync which is new and can't be done in compat...
+
 # Wireless RNDIS USB support (RTL8185 802.11g) A-Link WL54PC
 # All of these devices are based on Broadcom 4320 chip which
 # is only wireless RNDIS chip known to date.
@@ -125,12 +128,14 @@ CONFIG_USB_NET_RNDIS_HOST=m
 CONFIG_USB_NET_RNDIS_WLAN=m
 
 endif
+endif
 
 CONFIG_P54_USB=m
 CONFIG_RTL8187=m
 
 # RT2500USB does not require firmware
 CONFIG_RT2500USB=m
+CONFIG_RT2X00_LIB_USB=m
 NEED_RT2X00=y
 # RT73USB requires firmware
 ifneq ($(CONFIG_CRC_ITU_T),)
@@ -144,12 +149,12 @@ endif # end of USB driver list
 ifeq ($(NEED_RT2X00),y)
 CONFIG_RT2X00=m
 CONFIG_RT2X00_LIB=m
+# CONFIG_RT2X00_LIB_DEBUGFS is not set
+# CONFIG_RT2X00_DEBUG is not set
 endif
 
 ifeq ($(NEED_RT2X00_FIRMWARE),y)
 CONFIG_RT2X00_LIB_FIRMWARE=y
-# CONFIG_RT2X00_LIB_DEBUGFS is not set
-# CONFIG_RT2X00_DEBUG is not set
 endif
 
 # p54
diff --git a/scripts/gen-compat-autoconf.sh b/scripts/gen-compat-autoconf.sh
index 4c97835..173431f 100755
--- a/scripts/gen-compat-autoconf.sh
+++ b/scripts/gen-compat-autoconf.sh
@@ -127,9 +127,9 @@ for i in $(grep '^CONFIG_' $COMPAT_CONFIG); do
 	VALUE=$(echo $i | cut -d"=" -f 2)
 
 	# skip vars that weren't actually set due to dependencies
-	if [ "${!VAR}" = "" ] ; then
-		continue
-	fi
+	#if [ "${!VAR}" = "" ] ; then
+	#	continue
+	#fi
 
 	# Handle core kernel module depenencies here.
 	case $VAR in

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Franz Apeltauer]

Luis R. Rodriguez schrieb:
> Frankz, I generated a new tarball with this applied. You can either
> patch your local tarball with this patch or just download the
> tarball again.
> 
>   Luis
> 
> commit d4097ab3593a721fc1d99823d2442464debf7f09
> Author: Luis R. Rodriguez <mcgrof@winlab.rutgers.edu>
> Date:   Mon Mar 10 19:20:33 2008 -0400

Loaded the new tarball.
It compiles fine.

Thanks!
Franz

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Luis R. Rodriguez]

On Mon, Mar 10, 2008 at 8:07 PM, Franz Apeltauer <franz.apeltauer@gmx.at> wrote:
> Luis R. Rodriguez schrieb:
>
> > Frankz, I generated a new tarball with this applied. You can either
>  > patch your local tarball with this patch or just download the
>  > tarball again.
>  >
>  >   Luis
>  >
>  > commit d4097ab3593a721fc1d99823d2442464debf7f09
>  > Author: Luis R. Rodriguez <mcgrof@winlab.rutgers.edu>
>  > Date:   Mon Mar 10 19:20:33 2008 -0400
>
>  Loaded the new tarball.
>  It compiles fine.

And it works too I hope? :)

  Luis

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Hauke Mehrtens]

[-- Attachment #1: Type: text/plain, Size: 904 bytes --]

Luis R. Rodriguez schrieb:
> On Mon, Mar 10, 2008 at 8:07 PM, Franz Apeltauer
<franz.apeltauer@gmx.at> wrote:
>> Luis R. Rodriguez schrieb:
>>
>>> Frankz, I generated a new tarball with this applied. You can either
>>  > patch your local tarball with this patch or just download the
>>  > tarball again.
>>  >
>>  >   Luis
>>  >
>>  > commit d4097ab3593a721fc1d99823d2442464debf7f09
>>  > Author: Luis R. Rodriguez <mcgrof@winlab.rutgers.edu>
>>  > Date:   Mon Mar 10 19:20:33 2008 -0400
>>
>>  Loaded the new tarball.
>>  It compiles fine.
>
> And it works too I hope? :)
>
>   Luis

On my system the tarball is compiling now, on Sunday it does not so,
with the error mentioned here.

It tries to find the .config for the kernel at /lib/modules/`uname
-r`/.config, but it is only at /lib/modules/`uname -r`/build/.config on
Ubuntu and Fedora systems.

--
Hauke Mehrtens


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 890 bytes --]

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Luis R. Rodriguez]

On Tue, Mar 11, 2008 at 09:06:33PM +0100, Hauke Mehrtens wrote:
> Luis R. Rodriguez schrieb:
> > On Mon, Mar 10, 2008 at 8:07 PM, Franz Apeltauer
> <franz.apeltauer@gmx.at> wrote:
> >> Luis R. Rodriguez schrieb:
> >>
> >>> Frankz, I generated a new tarball with this applied. You can either
> >>  > patch your local tarball with this patch or just download the
> >>  > tarball again.
> >>  >
> >>  >   Luis
> >>  >
> >>  > commit d4097ab3593a721fc1d99823d2442464debf7f09
> >>  > Author: Luis R. Rodriguez <mcgrof@winlab.rutgers.edu>
> >>  > Date:   Mon Mar 10 19:20:33 2008 -0400
> >>
> >>  Loaded the new tarball.
> >>  It compiles fine.
> >
> > And it works too I hope? :)
> >
> >   Luis
> 
> On my system the tarball is compiling now, on Sunday it does not so,
> with the error mentioned here.
> 
> It tries to find the .config for the kernel at /lib/modules/`uname
> -r`/.config, but it is only at /lib/modules/`uname -r`/build/.config on
> Ubuntu and Fedora systems.

Fixed, you can either apply this patch or download the new tarball
which I just regenerated.

	Luis

diff --git a/config.mk b/config.mk
index b9d537d..f6121e9 100644
--- a/config.mk
+++ b/config.mk
@@ -4,13 +4,13 @@ export
 ## Make sure to have each variable declaration start
 ## in the first column, no whitespace allowed.
 
-ifeq ($(wildcard $(KLIB)/.config),)
+ifeq ($(wildcard $(KLIB_BUILD)/.config),)
 # These will be ignored by compat autoconf
  CONFIG_PCI=y
  CONFIG_USB=y
  CONFIG_PCMCIA=y
 else
-include $(KLIB)/.config
+include $(KLIB_BUILD)/.config
 endif
 
 # Wireless subsystem stuff

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Franz Apeltauer]

Luis R. Rodriguez schrieb:
>>  Loaded the new tarball.
>>  It compiles fine.
> 
> And it works too I hope? :)
> 
>   Luis
> 
Unfortunately not. I am still trying to find out, if it is my fault or not.

It seems, that it only allows Managed mode.
"iwconfig wlan0 mode Ad-Hoc" and "iwconfig wlan0 mode Master" both give 
me this error:
Error for wireless request "Set Mode" (8B06) :
     SET failed on device wlan0 ; Invalid argument.

I intended to use an ad-hoc connection between 2 PCs.
Since I have no access point I cannot test much at the moment.

Is it supposed to provide ad-hoc and master mode on rt2500 (pci)?
Probably not yet, because it is not in the list 
(http://linuxwireless.org/en/users/Drivers). Right?

Franz

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Ivo van Doorn]

On Tuesday 11 March 2008, Franz Apeltauer wrote:
> Luis R. Rodriguez schrieb:
> >>  Loaded the new tarball.
> >>  It compiles fine.
> > 
> > And it works too I hope? :)
> > 
> >   Luis
> > 
> Unfortunately not. I am still trying to find out, if it is my fault or not.
> 
> It seems, that it only allows Managed mode.
> "iwconfig wlan0 mode Ad-Hoc" and "iwconfig wlan0 mode Master" both give 
> me this error:
> Error for wireless request "Set Mode" (8B06) :
>      SET failed on device wlan0 ; Invalid argument.
> 
> I intended to use an ad-hoc connection between 2 PCs.
> Since I have no access point I cannot test much at the moment.
> 
> Is it supposed to provide ad-hoc and master mode on rt2500 (pci)?
> Probably not yet, because it is not in the list 
> (http://linuxwireless.org/en/users/Drivers). Right?

rt2500pci should work with adhoc and master mode, I have to note that
with versions up to 2.6.25 this is disabled due to lack of testing.
I don't know which rt2x00 version is packaged in the tarball, but if it
is older then 2.1.0 then adhoc and master mode is disabled in the driver.

Also note that although it is currently enabled, I still have not
received test reports about it, so I cannot claim that it is working.

Ivo

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Franz Apeltauer]

Ivo van Doorn schrieb:
> rt2500pci should work with adhoc and master mode, I have to note that
> with versions up to 2.6.25 this is disabled due to lack of testing.
> I don't know which rt2x00 version is packaged in the tarball, but if it
> is older then 2.1.0 then adhoc and master mode is disabled in the driver.
> 
> Also note that although it is currently enabled, I still have not
> received test reports about it, so I cannot claim that it is working.
> 
> Ivo
> 
Found in rt2x00.h: #define DRV_VERSION    "2.1.3"

I am far away from being a wlan expert or kernel hacker.
But if that helps, I am willing to run any test you want on
Ubuntu gutsy, kernel 2.6.22.

Franz

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Ivo van Doorn]

On Wednesday 12 March 2008, Franz Apeltauer wrote:
> Ivo van Doorn schrieb:
> > rt2500pci should work with adhoc and master mode, I have to note that
> > with versions up to 2.6.25 this is disabled due to lack of testing.
> > I don't know which rt2x00 version is packaged in the tarball, but if it
> > is older then 2.1.0 then adhoc and master mode is disabled in the driver.
> > 
> > Also note that although it is currently enabled, I still have not
> > received test reports about it, so I cannot claim that it is working.
> > 
> > Ivo
> > 
> Found in rt2x00.h: #define DRV_VERSION    "2.1.3"

Hmm, thats odd. It should at least support adhoc.
(Master mode depends on mac80211, regular mac80211 versions found
in the kernel don't support it yet, and they need patches from
Johannes Berg).

Does 'dmesg' indicate anything?

> I am far away from being a wlan expert or kernel hacker.
> But if that helps, I am willing to run any test you want on
> Ubuntu gutsy, kernel 2.6.22.

Ivo

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Franz Apeltauer]

[-- Attachment #1: Type: text/plain, Size: 798 bytes --]

Ivo van Doorn schrieb:
> Hmm, thats odd. It should at least support adhoc.
> (Master mode depends on mac80211, regular mac80211 versions found
> in the kernel don't support it yet, and they need patches from
> Johannes Berg).
> 
> Does 'dmesg' indicate anything?
> 

Loaded the current tarball.
There are not many error or warning messages in dmeg output anymore.
Like before I found "ADDRCONF(NETDEV_UP): wlan0: link is not ready"
but I thought, that this probably tells me, that it could not "see" any 
access point (which is true).

dmesg output etc. is attached.

It seems, that ad-hoc is not needed by many people. And for me wlan is 
nice to have, not a must. So, I do not want to waste your time any 
longer to get this 'special' case runnning.

Thank you all very much for your help!
Franz

[-- Attachment #2: dmesg2.txt --]
[-- Type: text/plain, Size: 9901 bytes --]

# make load
Unloading rt2500pci...
Loading ipw2100...
Loading ipw2200...
Loading libertas_cs...
Loading usb8xxx...
Loading p54pci...
Loading p54usb...
Loading adm8211...
Loading zd1211rw...
Loading rtl8180...
Loading rtl8187...
Loading p54pci...
Loading p54usb...
Loading iwl3945...
Loading iwl4965...
Loading rtl8180...
Loading rtl8187...
Loading rtl8180...
Loading rtl8187...
Loading rt2400pci...
Loading rt2500pci...
Loading rt61pci...
Loading rt2500usb...
Loading rt73usb...
Loading rndis_wlan...
Loading at76_usb...
Module ath_pci not detected -- this is fine
ath5k loaded successfully
Disabling bcm43xx ...   [OK]    Module disabled:
/lib/modules/2.6.22-14-generic/kernel/drivers/net/wireless/bcm43xx/bcm43xx.ko
b43 loaded successfully
b43legacy loaded successfully


# dmesg -c
[15903.292833] dev_mc_discard: multicast leakage! dmi_users=2
[15903.327793] dev_mc_discard: multicast leakage! dmi_users=2
[15903.385363] ACPI: PCI interrupt for device 0000:00:06.0 disabled
[15904.586329] ieee80211_crypt: registered algorithm 'NULL'
[15904.595361] ieee80211: 802.11 data/management/control stack, git-1.1.13
[15904.595368] ieee80211: Copyright (C) 2004-2005 Intel Corporation <jketreno@linux.intel.com>
[15904.602865] ipw2100: Intel(R) PRO/Wireless 2100 Network Driver, git-1.2.2
[15904.602872] ipw2100: Copyright(c) 2003-2006 Intel Corporation
[15904.681788] ipw2200: Intel(R) PRO/Wireless 2200/2915 Network Driver, 1.2.2kmprq
[15904.681796] ipw2200: Copyright(c) 2003-2006 Intel Corporation
[15904.867622] usbcore: registered new interface driver usb8xxx
[15905.049941] usbcore: registered new interface driver prism54usb
[15905.213550] usbcore: registered new interface driver zd1211rw
[15905.351793] usbcore: registered new interface driver rtl8187
[15905.549230] iwl3945: Intel(R) PRO/Wireless 3945ABG/BG Network Connection driver for Linux, 1.2.26k
[15905.549237] iwl3945: Copyright(c) 2003-2008 Intel Corporation
[15905.625621] iwl4965: Intel(R) Wireless WiFi Link 4965AGN driver for Linux, 1.2.26k
[15905.625629] iwl4965: Copyright(c) 2003-2008 Intel Corporation
[15905.994840] ACPI: PCI Interrupt 0000:00:06.0[A] -> GSI 17 (level, low) -> IRQ 18
[15906.008582] phy0: Selected rate control algorithm 'pid'
[15906.107488] ADDRCONF(NETDEV_UP): wlan0: link is not ready
[15906.205649] usbcore: registered new interface driver rt2500usb
[15906.266356] usbcore: registered new interface driver rt73usb
[15906.334440] usbcore: registered new interface driver cdc_ether
[15906.342936] usbcore: registered new interface driver rndis_host
[15906.352212] usbcore: registered new interface driver rndis_wlan
[15906.409783] Atmel at76x USB Wireless LAN Driver 0.17 loading
[15906.413100] usbcore: registered new interface driver at76_usb
[15913.730773] Broadcom 43xx driver loaded [ Features: PML, Firmware-ID: FW13 ]
[15913.762510] Broadcom 43xx-legacy driver loaded [ Features: PID, Firmware-ID: FW10 ]



# ifdown wlan0
ifdown: interface wlan0 not configured
# dmesg -c



# ifup wlan0
Error for wireless request "Set Mode" (8B06) :
    SET failed on device wlan0 ; Device or resource busy.
# dmesg -c



# iwconfig wlan0 mode Ad-Hoc
Error for wireless request "Set Mode" (8B06) :
    SET failed on device wlan0 ; Device or resource busy.
# dmesg -c



# lsmod
Module                  Size  Used by
b43legacy             102040  0
b43                   149532  0
ssb                    40580  2 b43legacy,b43
led_class               5380  1 b43
ath5k                  96512  0
at76_usb               40484  0
rndis_wlan             25224  0
rndis_host              9984  1 rndis_wlan
cdc_ether               7168  1 rndis_host
usbnet                 19976  3 rndis_wlan,rndis_host,cdc_ether
rt73usb                28416  0
rt2500usb              25984  0
rt2x00usb              14464  2 rt73usb,rt2500usb
rt61pci                24704  0
rt2500pci              21248  0
rt2400pci              17536  0
rt2x00pci              11648  3 rt61pci,rt2500pci,rt2400pci
rt2x00lib              24064  7 rt73usb,rt2500usb,rt2x00usb,rt61pci,rt2500pci,rt2400pci,rt2x00pci
iwl4965               116340  0
iwlcore                 4736  1 iwl4965
iwl3945                88948  0
rtl8187                36992  0
rtl8180                31872  0
zd1211rw               50052  0
adm8211                28288  0
eeprom_93cx6            3456  6 rt61pci,rt2500pci,rt2400pci,rtl8187,rtl8180,adm8211
p54usb                 16768  0
p54pci                 13312  0
p54common              14080  2 p54usb,p54pci
mac80211              220916  16 b43legacy,b43,ath5k,at76_usb,rt2x00usb,rt2x00pci,rt2x00lib,iwl4965,iwl3945,rtl8187,rtl8180,zd1211rw,adm8211,p54usb,p54pci,p54common
usb8xxx                13056  0
libertas_cs            11008  0
pcmcia                 41388  3 b43,ssb,libertas_cs
libertas               81672  2 usb8xxx,libertas_cs
pcmcia_core            40980  4 b43,ssb,libertas_cs,pcmcia
ipw2200               146248  0
ipw2100                73008  0
ieee80211              35784  3 libertas,ipw2200,ipw2100
ieee80211_crypt         7296  1 ieee80211
cfg80211               36576  15 ssb,ath5k,rt2x00lib,iwl4965,iwl3945,rtl8187,rtl8180,adm8211,p54usb,p54pci,mac80211,libertas,ipw2200,ipw2100,ieee80211
nfsd                  220912  13
exportfs                7040  1 nfsd
lockd                  67592  2 nfsd
sunrpc                172412  8 nfsd,lockd
ppp_async              13184  1
ppp_generic            29332  5 ppp_async
slhc                    7552  1 ppp_generic
vmnet                  39460  5
vmmon                 113900  0
binfmt_misc            12936  1
rfcomm                 42136  2
l2cap                  26240  11 rfcomm
bluetooth              57060  4 rfcomm,l2cap
radeon                125472  2
drm                    83348  3 radeon
ppdev                  10244  0
cpufreq_stats           7232  0
cpufreq_userspace       5280  0
cpufreq_powersave       2688  0
cpufreq_ondemand        9612  0
freq_table              5792  2 cpufreq_stats,cpufreq_ondemand
cpufreq_conservative     8072  0
sbs                    19592  0
ac                      6148  0
button                  8976  0
video                  18060  0
dock                   10656  0
container               5504  0
battery                11012  0
nls_iso8859_1           5120  1
nls_cp437               6784  1
vfat                   14080  1
fat                    54300  1 vfat
sbp2                   24072  0
lp                     12580  0
loop                   19076  0
snd_via82xx            29336  1
gameport               16776  1 snd_via82xx
snd_ac97_codec        100644  1 snd_via82xx
ac97_bus                3200  1 snd_ac97_codec
snd_pcm_oss            44672  0
snd_mixer_oss          17664  1 snd_pcm_oss
snd_pcm                80388  3 snd_via82xx,snd_ac97_codec,snd_pcm_oss
snd_page_alloc         11400  2 snd_via82xx,snd_pcm
snd_mpu401_uart         9600  1 snd_via82xx
snd_seq_dummy           4740  0
snd_seq_oss            33152  0
snd_seq_midi            9600  0
parport_pc             37412  1
parport                37448  3 ppdev,lp,parport_pc
sg                     36764  0
snd_rawmidi            25728  2 snd_mpu401_uart,snd_seq_midi
snd_seq_midi_event      8448  2 snd_seq_oss,snd_seq_midi
snd_seq                53232  6 snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_seq_midi_event
snd_timer              24324  2 snd_pcm,snd_seq
snd_seq_device          9228  5 snd_seq_dummy,snd_seq_oss,snd_seq_midi,snd_rawmidi,snd_seq
pcspkr                  4224  0
sd_mod                 30336  0
i2c_viapro             10004  0
snd                    54660  13 snd_via82xx,snd_ac97_codec,snd_pcm_oss,snd_mixer_oss,snd_pcm,snd_mpu401_uart,snd_seq_oss,snd_rawmidi,snd_seq,snd_timer,snd_seq_device
arc4                    2944  2
ecb                     4608  2
blkcipher               7556  1 ecb
i2c_core               26112  1 i2c_viapro
soundcore               8800  1 snd
shpchp                 34580  0
crc_itu_t               3072  2 rt73usb,rt61pci
crc_ccitt               3072  1 ppp_async
pci_hotplug            32704  1 shpchp
via_agp                11264  1
agpgart                35016  2 drm,via_agp
ipv6                  273892  10
evdev                  11136  4
ipt_MASQUERADE          4608  2
xt_mark                 2816  1
iptable_mangle          3840  1
iptable_nat             8708  1
nf_nat                 20140  2 ipt_MASQUERADE,iptable_nat
nf_conntrack_ipv4      19724  2 iptable_nat
nf_conntrack           65288  4 ipt_MASQUERADE,iptable_nat,nf_nat,nf_conntrack_ipv4
nfnetlink               6936  3 nf_nat,nf_conntrack_ipv4,nf_conntrack
ip_tables              13924  2 iptable_mangle,iptable_nat
x_tables               16260  4 ipt_MASQUERADE,xt_mark,iptable_nat,ip_tables
ext3                  133896  2
jbd                    60456  1 ext3
mbcache                 9732  1 ext3
ide_cd                 32672  0
cdrom                  37536  1 ide_cd
ide_disk               18560  5
usbhid                 29536  0
hid                    28928  1 usbhid
floppy                 60004  0
ehci_hcd               36492  0
sata_via               12548  0
via82cxxx              10372  0 [permanent]
ide_core              116804  3 ide_cd,ide_disk,via82cxxx
uhci_hcd               26640  0
usbcore               138632  16 at76_usb,rndis_wlan,rndis_host,cdc_ether,usbnet,rt73usb,rt2500usb,rt2x00usb,rtl8187,zd1211rw,p54usb,usb8xxx,usbhid,ehci_hcd,uhci_hcd
ata_generic             8452  0
libata                125168  2 sata_via,ata_generic
scsi_mod              147084  4 sbp2,sg,sd_mod,libata
ohci1394               36528  0
ieee1394               96312  2 sbp2,ohci1394
tg3                   110980  0
3c59x                  46632  0
mii                     6528  2 usbnet,3c59x
thermal                14344  1
processor              32072  1 thermal
fan                     5764  1
fuse                   47124  1
apparmor               40728  0
commoncap               8320  1 apparmor
#

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Luis R. Rodriguez]

On Sat, Mar 15, 2008 at 8:33 PM, Franz Apeltauer <franz.apeltauer@gmx.at> wrote:
> Ivo van Doorn schrieb:
>
> > Hmm, thats odd. It should at least support adhoc.
>  > (Master mode depends on mac80211, regular mac80211 versions found
>  > in the kernel don't support it yet, and they need patches from
>  > Johannes Berg).
>  >
>  > Does 'dmesg' indicate anything?
>  >
>
>  Loaded the current tarball.
>  There are not many error or warning messages in dmeg output anymore.
>  Like before I found "ADDRCONF(NETDEV_UP): wlan0: link is not ready"
>  but I thought, that this probably tells me, that it could not "see" any
>  access point (which is true).

This message comes out when your device has no queue dicipline
attached to it yet. For new wireless drivers (mac80211) this pops up
when you are not associated as you have no communication link
established yet.

>  dmesg output etc. is attached.
>
>  It seems, that ad-hoc is not needed by many people. And for me wlan is
>  nice to have, not a must. So, I do not want to waste your time any
>  longer to get this 'special' case runnning.
>
>  Thank you all very much for your help!
>  Franz
>
> # make load

You can avoid using make load by just doing:

make unload
modprobe foo

where foo is the wireless module you *know* you need.

  Luis

Re: 2 compile errors: rt2x00firmware.c and mesh.c

[Ivo Van Doorn]

Hi,

>  >    CC [M]
>  > .../compat-wireless-2.6/drivers/net/wireless/rt2x00/rt2x00firmware=
=2Eo
>  > .../compat-wireless-2.6/drivers/net/wireless/rt2x00/rt2x00firmware=
=2Ec:116:
>  > Fehler: Redefinition von =BBrt2x00lib_load_firmware=AB
>  > =3D Error: redefiniton of  =BBrt2x00lib_load_firmware=AB
>  >
>
> > When I tried to "export CONFIG_CRC_ITU_T=3Dy" before "make",
>  > rt2x00firmware.c compiled fine. (Probably this is not a good idea.=
 I
>  > just wanted to see, if it compiles then.)
>
>  That's odd because the latest config.mk should check for CRC_ITU_T
>  before... Ivo could there be a problem with rt2x00 when crc-itu-t is=
n't
>  available?

Yes, but only for rt61 and rt73. The crc-itu-t is used for firmware
crc validation.
(Note that rt2x00firmware.o should only be compiled when rt61 and rt73 =
are
being compiled).

There currently is also a dependency on crc-ccitt, but that is for the
upcoming rt2800 drivers (Yes, I could do the same as the legacy driver
and use crc-itu-t, but I dislike the idea of bit-reverting each firmwar=
e word
before pushing it through the crc algorithm.).

>  > In /usr/src/linux-headers-2.6.22-14/include/linux/slab.h I found:
>  >     struct kmem_cache *kmem_cache_create(const char *, size_t, siz=
e_t,
>  >               unsigned long,
>  >               void (*)(void *, struct kmem_cache *, unsigned long)=
,
>  >               void (*)(void *, struct kmem_cache *, unsigned long)=
);
>  > i.e. the old, 6-parameter version.
>  >
>  > So, IMHO the #define kmem_cache_create in
>  > - compat-wireless-2.6/include/net/compat.h, line 120 and
>  > - compat-wireless-2.6/compat/compat.h, line 120
>  > should be moved 2 lines downwards into the block named
>  > /* Compat work for 2.6.22 and 2.6.23 */
>  > Then it compiles fine.
>
>  Ok, indeed, it should be moved then, I was working on .21 when I not=
iced
>  that needed to be done.
>
>  johannes
>
--
To unsubscribe from this list: send the line "unsubscribe linux-wireles=
s" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html