[linux-next] next-20101207: Better now, but modinfo ath5k

[linux-next] next-20101207: Better now, but modinfo ath5k

[Sedat Dilek]

Hi,

I just compiled linux-next (next-20101207) which also includes
wireless-next-2.6 (master-2010-12-06) [1].

I have already reported on linux-wireless (short: l-w) ML that after
especially the merge of "AHB Bus support" patchset (see [2]) my ath5k
was not working as expected in daily use (also modinfo).

I am not sure if it is the driver but reverting the 9 commits in my
local linux-next GIT seemed to "stabilize" a bit the Internet
connection.
The patchset from Nick (see below) definitely helped to stabilize ath5k.

What means "daily use"?
Quassel IRC-client (around 20 channels), Iceweasel (Debian's answer to
firefox) with sometimes 20 tabs open and radio live-stream in VLC.

Here and there I have still audio-dropouts or when typing emails in
web-based email-service, it stalls when "stressing" with for example
parallelly creating a tarball (that's normally no huge workload) the
probability for audio-dropouts increases.

As I said I am not sure what is causing these hangs, it's ath5k
driver? Is it sth. else?
How to dig into it?

Unfortunately, modinfo does not list aliases with pci-id matching (I
have already reported to l-w ML, see below).
With reverting "AHB Bus support" patchset modinfo works as expected.

As a workaround I added ath5k to /etc/modules (see below), otherwise I
have to stop networking, unload + reload ath5k and start networking
again.

I also added an iw station dump (see below) and outputs of filtered
lspci and dmesg commands.

Not sure if this is helpful to mention, I use crda and wireless-regdb
(see below).

BTW, I have these two patches in addition to linux-next:
  (+) OK   mac80211-fix/mac80211-fix-a-compiler-warning.patch
  (+) OK   ath5k-fix/ath5k-Use-capabilities-information-for-the-number-of-TX-queues.patch

Any ideas/help appreciated, Thanks in advance.

Kind Regards,
- Sedat -

[1] http://git.kernel.org/?p=linux/kernel/git/linville/wireless-next-2.6.git
[2] http://wireless.kernel.org/en/users/Drivers/ath5k

P.S.: Patchset from Nick Kossifidis

$ ls -l ath5k-fix/*-6-*
-rw-r--r-- 1 sd sd 3439  3. Dez 06:52
ath5k-fix/1-6-ath5k-Always-write-tx-powertable-on-hw.patch
-rw-r--r-- 1 sd sd 1725  3. Dez 06:51
ath5k-fix/2-6-ath5k-Always-free-tx-buffers-before-reset.patch
-rw-r--r-- 1 sd sd 2132  3. Dez 06:52
ath5k-fix/3-6-ath5k-Disable-ANI-during-reset.patch
-rw-r--r-- 1 sd sd 1360  3. Dez 06:52
ath5k-fix/4-6-ath5k-Fix-reporting-of-RX-dma-stop-failure.patch
-rw-r--r-- 1 sd sd 1735  3. Dez 06:52
ath5k-fix/5-6-ath5k-Update-version-string.patch
-rw-r--r-- 1 sd sd 1744  3. Dez 06:53
ath5k-fix/6-6-ath5k-Include-tx-ack-reporting-on-hw-flags.patch

# modinfo ath5k
filename:
/lib/modules/2.6.37-rc4-686/kernel/drivers/net/wireless/ath/ath5k/ath5k.ko
version:        0.6.0 (EXPERIMENTAL)
license:        Dual BSD/GPL
description:    Support for 5xxx series of Atheros 802.11 wireless LAN cards.
author:         Nick Kossifidis
author:         Jiri Slaby
srcversion:     49DBF0BE975A1F9917151F4
depends:        mac80211,cfg80211,ath
vermagic:       2.6.37-rc4-686 SMP mod_unload modversions 686
parm:           debug:uint
parm:           nohwcrypt:Disable hardware encryption. (bool)
parm:           all_channels:Expose all channels the device can use. (bool)

# grep ath5k /etc/modules
ath5k

# iw wlan0 station dump
Station 00:04:0e:e4:00:3d (on wlan0)
        inactive time:  48 ms
        rx bytes:       60460696
        rx packets:     95331
        tx bytes:       5034456
        tx packets:     50165
        signal:         -50 dBm
        tx bitrate:     36.0 MBit/s

# lspci -nnvv | grep -i ath
02:02.0 Ethernet controller [0200]: Atheros Communications Inc. AR5212
802.11abg NIC [168c:1014] (rev 01)
        Kernel driver in use: ath5k

# dmesg | egrep -i 'ath|80211|wlan'
[   14.444118] cfg80211: Calling CRDA to update world regulatory domain
[   14.725757] ath5k 0000:02:02.0: PCI INT A -> Link[LNKC] -> GSI 11
(level, low) -> IRQ 11
[   14.736940] ath5k 0000:02:02.0: registered as 'phy0'
[   15.036636] ath: EEPROM regdomain: 0x61
[   15.036640] ath: EEPROM indicates we should expect a direct regpair map
[   15.036646] ath: Country alpha2 being used: 00
[   15.036648] ath: Regpair used: 0x61
[   15.665415] cfg80211: World regulatory domain updated:
[   15.676274] cfg80211:     (start_freq - end_freq @ bandwidth),
(max_antenna_gain, max_eirp)
[   15.687303] cfg80211:     (2402000 KHz - 2472000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[   15.698413] cfg80211:     (2457000 KHz - 2482000 KHz @ 20000 KHz),
(300 mBi, 2000 mBm)
[   15.709461] cfg80211:     (2474000 KHz - 2494000 KHz @ 20000 KHz),
(300 mBi, 2000 mBm)
[   15.720489] cfg80211:     (5170000 KHz - 5250000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[   15.731526] cfg80211:     (5735000 KHz - 5835000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[   15.754490] ieee80211 phy0: Selected rate control algorithm 'minstrel_ht'
[   15.755482] Registered led device: ath5k-phy0::rx
[   15.755516] Registered led device: ath5k-phy0::tx
[   15.755529] ath5k phy0: Atheros AR5212 chip found (MAC: 0x56, PHY: 0x41)
[   15.766686] ath5k phy0: RF5111 5GHz radio found (0x17)
[   15.777870] ath5k phy0: RF2111 2GHz radio found (0x23)
[   17.374647] ADDRCONF(NETDEV_UP): wlan0: link is not ready
[   22.596104] wlan0: authenticate with 00:04:0e:e4:00:3d (try 1)
[   22.597531] wlan0: authenticated
[   22.597549] wlan0: associate with 00:04:0e:e4:00:3d (try 1)
[   22.602095] wlan0: RX AssocResp from 00:04:0e:e4:00:3d (capab=0x411
status=0 aid=1)
[   22.602100] wlan0: associated
[   22.602986] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
[   33.312034] wlan0: no IPv6 routers present
[  215.317105] cfg80211: Calling CRDA to update world regulatory domain
[  215.346333] cfg80211: World regulatory domain updated:
[  215.346341] cfg80211:     (start_freq - end_freq @ bandwidth),
(max_antenna_gain, max_eirp)
[  215.346346] cfg80211:     (2402000 KHz - 2472000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[  215.346350] cfg80211:     (2457000 KHz - 2482000 KHz @ 20000 KHz),
(300 mBi, 2000 mBm)
[  215.346354] cfg80211:     (2474000 KHz - 2494000 KHz @ 20000 KHz),
(300 mBi, 2000 mBm)
[  215.346358] cfg80211:     (5170000 KHz - 5250000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[  215.346362] cfg80211:     (5735000 KHz - 5835000 KHz @ 40000 KHz),
(300 mBi, 2000 mBm)
[  219.627199] wlan0: authenticate with 00:04:0e:e4:00:3d (try 1)
[  219.629321] wlan0: authenticated
[  219.629365] wlan0: associate with 00:04:0e:e4:00:3d (try 1)
[  219.635825] wlan0: RX AssocResp from 00:04:0e:e4:00:3d (capab=0x411
status=0 aid=1)
[  219.635836] wlan0: associated

# dpkg -l | egrep 'crda|wireless-regdb' | cut -c-80
ii  crda                                            1.1.1-0~svnr1515.3
ii  wireless-regdb                                  2010.11.24-1

Re: [linux-next] next-20101207: Better now, but modinfo ath5k

[Sedat Dilek]

On Tue, Dec 7, 2010 at 12:24 PM, Sedat Dilek <sedat.dilek@googlemail.com> wrote:
> Hi,
>
> I just compiled linux-next (next-20101207) which also includes
> wireless-next-2.6 (master-2010-12-06) [1].
>
> I have already reported on linux-wireless (short: l-w) ML that after
> especially the merge of "AHB Bus support" patchset (see [2]) my ath5k
> was not working as expected in daily use (also modinfo).
>
> I am not sure if it is the driver but reverting the 9 commits in my
> local linux-next GIT seemed to "stabilize" a bit the Internet
> connection.
> The patchset from Nick (see below) definitely helped to stabilize ath5k.
>
> What means "daily use"?
> Quassel IRC-client (around 20 channels), Iceweasel (Debian's answer to
> firefox) with sometimes 20 tabs open and radio live-stream in VLC.
>
> Here and there I have still audio-dropouts or when typing emails in
> web-based email-service, it stalls when "stressing" with for example
> parallelly creating a tarball (that's normally no huge workload) the
> probability for audio-dropouts increases.
>
> As I said I am not sure what is causing these hangs, it's ath5k
> driver? Is it sth. else?
> How to dig into it?
>
> Unfortunately, modinfo does not list aliases with pci-id matching (I
> have already reported to l-w ML, see below).
> With reverting "AHB Bus support" patchset modinfo works as expected.
>
> As a workaround I added ath5k to /etc/modules (see below), otherwise I
> have to stop networking, unload + reload ath5k and start networking
> again.
>
> I also added an iw station dump (see below) and outputs of filtered
> lspci and dmesg commands.
>
> Not sure if this is helpful to mention, I use crda and wireless-regdb
> (see below).
>
> BTW, I have these two patches in addition to linux-next:
>  (+) OK   mac80211-fix/mac80211-fix-a-compiler-warning.patch
>  (+) OK   ath5k-fix/ath5k-Use-capabilities-information-for-the-number-of-TX-queues.patch
>
> Any ideas/help appreciated, Thanks in advance.
>
> Kind Regards,
> - Sedat -
>
> [1] http://git.kernel.org/?p=linux/kernel/git/linville/wireless-next-2.6.git
> [2] http://wireless.kernel.org/en/users/Drivers/ath5k
>
> P.S.: Patchset from Nick Kossifidis
>
> $ ls -l ath5k-fix/*-6-*
> -rw-r--r-- 1 sd sd 3439  3. Dez 06:52
> ath5k-fix/1-6-ath5k-Always-write-tx-powertable-on-hw.patch
> -rw-r--r-- 1 sd sd 1725  3. Dez 06:51
> ath5k-fix/2-6-ath5k-Always-free-tx-buffers-before-reset.patch
> -rw-r--r-- 1 sd sd 2132  3. Dez 06:52
> ath5k-fix/3-6-ath5k-Disable-ANI-during-reset.patch
> -rw-r--r-- 1 sd sd 1360  3. Dez 06:52
> ath5k-fix/4-6-ath5k-Fix-reporting-of-RX-dma-stop-failure.patch
> -rw-r--r-- 1 sd sd 1735  3. Dez 06:52
> ath5k-fix/5-6-ath5k-Update-version-string.patch
> -rw-r--r-- 1 sd sd 1744  3. Dez 06:53
> ath5k-fix/6-6-ath5k-Include-tx-ack-reporting-on-hw-flags.patch
>
> # modinfo ath5k
> filename:
> /lib/modules/2.6.37-rc4-686/kernel/drivers/net/wireless/ath/ath5k/ath5k.ko
> version:        0.6.0 (EXPERIMENTAL)
> license:        Dual BSD/GPL
> description:    Support for 5xxx series of Atheros 802.11 wireless LAN cards.
> author:         Nick Kossifidis
> author:         Jiri Slaby
> srcversion:     49DBF0BE975A1F9917151F4
> depends:        mac80211,cfg80211,ath
> vermagic:       2.6.37-rc4-686 SMP mod_unload modversions 686
> parm:           debug:uint
> parm:           nohwcrypt:Disable hardware encryption. (bool)
> parm:           all_channels:Expose all channels the device can use. (bool)
>
> # grep ath5k /etc/modules
> ath5k
>
> # iw wlan0 station dump
> Station 00:04:0e:e4:00:3d (on wlan0)
>        inactive time:  48 ms
>        rx bytes:       60460696
>        rx packets:     95331
>        tx bytes:       5034456
>        tx packets:     50165
>        signal:         -50 dBm
>        tx bitrate:     36.0 MBit/s
>
> # lspci -nnvv | grep -i ath
> 02:02.0 Ethernet controller [0200]: Atheros Communications Inc. AR5212
> 802.11abg NIC [168c:1014] (rev 01)
>        Kernel driver in use: ath5k
>
> # dmesg | egrep -i 'ath|80211|wlan'
> [   14.444118] cfg80211: Calling CRDA to update world regulatory domain
> [   14.725757] ath5k 0000:02:02.0: PCI INT A -> Link[LNKC] -> GSI 11
> (level, low) -> IRQ 11
> [   14.736940] ath5k 0000:02:02.0: registered as 'phy0'
> [   15.036636] ath: EEPROM regdomain: 0x61
> [   15.036640] ath: EEPROM indicates we should expect a direct regpair map
> [   15.036646] ath: Country alpha2 being used: 00
> [   15.036648] ath: Regpair used: 0x61
> [   15.665415] cfg80211: World regulatory domain updated:
> [   15.676274] cfg80211:     (start_freq - end_freq @ bandwidth),
> (max_antenna_gain, max_eirp)
> [   15.687303] cfg80211:     (2402000 KHz - 2472000 KHz @ 40000 KHz),
> (300 mBi, 2000 mBm)
> [   15.698413] cfg80211:     (2457000 KHz - 2482000 KHz @ 20000 KHz),
> (300 mBi, 2000 mBm)
> [   15.709461] cfg80211:     (2474000 KHz - 2494000 KHz @ 20000 KHz),
> (300 mBi, 2000 mBm)
> [   15.720489] cfg80211:     (5170000 KHz - 5250000 KHz @ 40000 KHz),
> (300 mBi, 2000 mBm)
> [   15.731526] cfg80211:     (5735000 KHz - 5835000 KHz @ 40000 KHz),
> (300 mBi, 2000 mBm)
> [   15.754490] ieee80211 phy0: Selected rate control algorithm 'minstrel_ht'
> [   15.755482] Registered led device: ath5k-phy0::rx
> [   15.755516] Registered led device: ath5k-phy0::tx
> [   15.755529] ath5k phy0: Atheros AR5212 chip found (MAC: 0x56, PHY: 0x41)
> [   15.766686] ath5k phy0: RF5111 5GHz radio found (0x17)
> [   15.777870] ath5k phy0: RF2111 2GHz radio found (0x23)
> [   17.374647] ADDRCONF(NETDEV_UP): wlan0: link is not ready
> [   22.596104] wlan0: authenticate with 00:04:0e:e4:00:3d (try 1)
> [   22.597531] wlan0: authenticated
> [   22.597549] wlan0: associate with 00:04:0e:e4:00:3d (try 1)
> [   22.602095] wlan0: RX AssocResp from 00:04:0e:e4:00:3d (capab=0x411
> status=0 aid=1)
> [   22.602100] wlan0: associated
> [   22.602986] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
> [   33.312034] wlan0: no IPv6 routers present
> [  215.317105] cfg80211: Calling CRDA to update world regulatory domain
> [  215.346333] cfg80211: World regulatory domain updated:
> [  215.346341] cfg80211:     (start_freq - end_freq @ bandwidth),
> (max_antenna_gain, max_eirp)
> [  215.346346] cfg80211:     (2402000 KHz - 2472000 KHz @ 40000 KHz),
> (300 mBi, 2000 mBm)
> [  215.346350] cfg80211:     (2457000 KHz - 2482000 KHz @ 20000 KHz),
> (300 mBi, 2000 mBm)
> [  215.346354] cfg80211:     (2474000 KHz - 2494000 KHz @ 20000 KHz),
> (300 mBi, 2000 mBm)
> [  215.346358] cfg80211:     (5170000 KHz - 5250000 KHz @ 40000 KHz),
> (300 mBi, 2000 mBm)
> [  215.346362] cfg80211:     (5735000 KHz - 5835000 KHz @ 40000 KHz),
> (300 mBi, 2000 mBm)
> [  219.627199] wlan0: authenticate with 00:04:0e:e4:00:3d (try 1)
> [  219.629321] wlan0: authenticated
> [  219.629365] wlan0: associate with 00:04:0e:e4:00:3d (try 1)
> [  219.635825] wlan0: RX AssocResp from 00:04:0e:e4:00:3d (capab=0x411
> status=0 aid=1)
> [  219.635836] wlan0: associated
>
> # dpkg -l | egrep 'crda|wireless-regdb' | cut -c-80
> ii  crda                                            1.1.1-0~svnr1515.3
> ii  wireless-regdb                                  2010.11.24-1
>

Might also be interesting to know:

# wpa_cli -i wlan0 status
bssid=00:04:0e:e4:00:3d
ssid=myCastle-WLAN WPA (Wireless LAN)
id=0
pairwise_cipher=CCMP
group_cipher=TKIP
key_mgmt=WPA2-PSK
wpa_state=COMPLETED
ip_address=192.168.178.25

- Sedat -

Re: [linux-next] next-20101207: Better now, but modinfo ath5k

[John W. Linville]

On Tue, Dec 07, 2010 at 12:24:14PM +0100, Sedat Dilek wrote:

> I have already reported on linux-wireless (short: l-w) ML that after
> especially the merge of "AHB Bus support" patchset (see [2]) my ath5k
> was not working as expected in daily use (also modinfo).
> 
> I am not sure if it is the driver but reverting the 9 commits in my
> local linux-next GIT seemed to "stabilize" a bit the Internet
> connection.
> The patchset from Nick (see below) definitely helped to stabilize ath5k.

I imagine that these are the "9 commits" you mean:

> git rev-list c30ae13..4cebb34
4cebb34caa5122216a1e2451eae9e0fc47ec2589
a0b907ee2a71052fefdf6151764095f3f97b3275
c31b5c9c806b1fbdc9e98885d897664a0d482989
e7aecd327d80b2f156b54769013aaccb2a20645a
8efa5d7d6ad307ae2d220def37ca89594062c40d
4aa5d783c9e1c72e4950ff34f388077ccecac74a
e5b046d86fac609f636d127a38de94a175c7e83b
132b1c3ee38ea6fa0501004fd0f19acb554e5a44
aeae4ac9090462ea38387dcdbac4f01b944af6a4

Most of those look fairly harmless so long as you don't define
CONFIG_ATHEROS_AR231X.  Just to narrow things down...if you only revert
the first 8 (i.e. leave aeae4ac alone), do you still have problems?
Or does it still "stabilize"?

That patch converts the driver to the generic DMA API.  It seems
OK at first glance, but I suspect it is the only one likely to be
effecting your situation.

John
-- 
John W. Linville		Someday the world will need a hero, and you
linville@tuxdriver.com			might be all we have.  Be ready.

Re: [linux-next] next-20101207: Better now, but modinfo ath5k

[Sedat Dilek]

On Tue, Dec 7, 2010 at 3:26 PM, John W. Linville <linville@tuxdriver.com> wrote:
> On Tue, Dec 07, 2010 at 12:24:14PM +0100, Sedat Dilek wrote:
>
>> I have already reported on linux-wireless (short: l-w) ML that after
>> especially the merge of "AHB Bus support" patchset (see [2]) my ath5k
>> was not working as expected in daily use (also modinfo).
>>
>> I am not sure if it is the driver but reverting the 9 commits in my
>> local linux-next GIT seemed to "stabilize" a bit the Internet
>> connection.
>> The patchset from Nick (see below) definitely helped to stabilize ath5k.
>
> I imagine that these are the "9 commits" you mean:
>
>> git rev-list c30ae13..4cebb34
> 4cebb34caa5122216a1e2451eae9e0fc47ec2589
> a0b907ee2a71052fefdf6151764095f3f97b3275
> c31b5c9c806b1fbdc9e98885d897664a0d482989
> e7aecd327d80b2f156b54769013aaccb2a20645a
> 8efa5d7d6ad307ae2d220def37ca89594062c40d
> 4aa5d783c9e1c72e4950ff34f388077ccecac74a
> e5b046d86fac609f636d127a38de94a175c7e83b
> 132b1c3ee38ea6fa0501004fd0f19acb554e5a44
> aeae4ac9090462ea38387dcdbac4f01b944af6a4
>
> Most of those look fairly harmless so long as you don't define
> CONFIG_ATHEROS_AR231X.  Just to narrow things down...if you only revert
> the first 8 (i.e. leave aeae4ac alone), do you still have problems?
> Or does it still "stabilize"?
>
> That patch converts the driver to the generic DMA API.  It seems
> OK at first glance, but I suspect it is the only one likely to be
> effecting your situation.
>
> John
> --
> John W. Linville                Someday the world will need a hero, and you
> linville@tuxdriver.com                  might be all we have.  Be ready.
>

Before jumping to another problem, I had a quick look into the
appropriate Kconfig file, as you say I have not CONFIG_ATHEROS_AR231X
set and that's why AHB support is not selected/built, instead

  CC [M]  drivers/net/wireless/ath/ath5k/pci.o

is built.

Unfortunately, I can't cleanly revert. Sorry, I can't help.

- Sedat -

sd@tbox:~/src/linux-2.6/linux-2.6.37-rc4/debian/build/source_i386_none$
LC_ALL=C quilt -f push
Applying patch 0001-Revert-Add-ath5k-Use-capabilities-information-for-th.patch
patching file drivers/net/wireless/ath/ath5k/base.c
patching file drivers/net/wireless/ath/ath5k/qcu.c

Now at patch 0001-Revert-Add-ath5k-Use-capabilities-information-for-th.patch
sd@tbox:~/src/linux-2.6/linux-2.6.37-rc4/debian/build/source_i386_none$
LC_ALL=C quilt -f push
Applying patch revert-ath5k-patches/0001-Revert-ath5k-Fix-reset-and-interrupts-for-AHB-type-o.patch
patching file drivers/net/wireless/ath/ath5k/base.c
patching file drivers/net/wireless/ath/ath5k/reset.c

Now at patch revert-ath5k-patches/0001-Revert-ath5k-Fix-reset-and-interrupts-for-AHB-type-o.patch
sd@tbox:~/src/linux-2.6/linux-2.6.37-rc4/debian/build/source_i386_none$
LC_ALL=C quilt -f push
Applying patch revert-ath5k-patches/0002-Revert-ath5k-Add-AHB-bus-support.patch
patching file drivers/net/wireless/ath/ath5k/Kconfig
patching file drivers/net/wireless/ath/ath5k/Makefile
patching file drivers/net/wireless/ath/ath5k/ahb.c
patching file drivers/net/wireless/ath/ath5k/ath5k.h
patching file drivers/net/wireless/ath/ath5k/base.c
patching file drivers/net/wireless/ath/ath5k/led.c
patching file drivers/net/wireless/ath/ath5k/reg.h

Now at patch revert-ath5k-patches/0002-Revert-ath5k-Add-AHB-bus-support.patch
sd@tbox:~/src/linux-2.6/linux-2.6.37-rc4/debian/build/source_i386_none$
LC_ALL=C quilt -f push
Applying patch revert-ath5k-patches/0003-Revert-ath5k-Add-initial-registers-values-for-radio-.patch
patching file drivers/net/wireless/ath/ath5k/initvals.c

Now at patch revert-ath5k-patches/0003-Revert-ath5k-Add-initial-registers-values-for-radio-.patch
sd@tbox:~/src/linux-2.6/linux-2.6.37-rc4/debian/build/source_i386_none$
LC_ALL=C quilt -f push
Applying patch revert-ath5k-patches/0004-Revert-ath5k-Add-a-function-to-read-chipset-s-MAC-re.patch
patching file drivers/net/wireless/ath/ath5k/ath5k.h
patching file drivers/net/wireless/ath/ath5k/attach.c
patching file drivers/net/wireless/ath/ath5k/pci.c

Now at patch revert-ath5k-patches/0004-Revert-ath5k-Add-a-function-to-read-chipset-s-MAC-re.patch
sd@tbox:~/src/linux-2.6/linux-2.6.37-rc4/debian/build/source_i386_none$
LC_ALL=C quilt -f push
Applying patch revert-ath5k-patches/0005-Revert-ath5k-Check-if-pci-pdev-struct-is-initialized.patch
patching file drivers/net/wireless/ath/ath5k/attach.c
patching file drivers/net/wireless/ath/ath5k/led.c
patching file drivers/net/wireless/ath/ath5k/reset.c
patching file drivers/net/wireless/ath/ath5k/sysfs.c

Now at patch revert-ath5k-patches/0005-Revert-ath5k-Check-if-pci-pdev-struct-is-initialized.patch
sd@tbox:~/src/linux-2.6/linux-2.6.37-rc4/debian/build/source_i386_none$
LC_ALL=C quilt -f push
Applying patch revert-ath5k-patches/0006-Revert-ath5k-Use-generic-eeprom-read-from-common-ath.patch
patching file drivers/net/wireless/ath/ath5k/ath5k.h
patching file drivers/net/wireless/ath/ath5k/eeprom.c
patching file drivers/net/wireless/ath/ath5k/eeprom.h
patching file drivers/net/wireless/ath/ath5k/pci.c

Now at patch revert-ath5k-patches/0006-Revert-ath5k-Use-generic-eeprom-read-from-common-ath.patch
sd@tbox:~/src/linux-2.6/linux-2.6.37-rc4/debian/build/source_i386_none$
LC_ALL=C quilt -f push
Applying patch revert-ath5k-patches/0007-Revert-ath5k-Move-PCI-bus-functions-to-separate-file.patch
patching file drivers/net/wireless/ath/ath5k/Makefile
patching file drivers/net/wireless/ath/ath5k/ath5k.h
patching file drivers/net/wireless/ath/ath5k/base.c
Hunk #5 succeeded at 3636 (offset 9 lines).
patching file drivers/net/wireless/ath/ath5k/pci.c

Now at patch revert-ath5k-patches/0007-Revert-ath5k-Move-PCI-bus-functions-to-separate-file.patch
sd@tbox:~/src/linux-2.6/linux-2.6.37-rc4/debian/build/source_i386_none$
LC_ALL=C quilt -f push
Applying patch revert-ath5k-patches/0008-Revert-ath5k-Introduce-ath5k_init_softc-function-as-.patch
patching file drivers/net/wireless/ath/ath5k/ath5k.h
patching file drivers/net/wireless/ath/ath5k/attach.c
patching file drivers/net/wireless/ath/ath5k/base.c
Hunk #4 FAILED at 2338.
Hunk #5 succeeded at 2530 (offset 1 line).
Hunk #6 succeeded at 2736 (offset 9 lines).
Hunk #7 succeeded at 2745 (offset 9 lines).
Hunk #8 succeeded at 2886 (offset 9 lines).
Hunk #9 succeeded at 2904 (offset 9 lines).
Hunk #10 succeeded at 2916 (offset 9 lines).
Hunk #11 succeeded at 2938 (offset 9 lines).
Hunk #12 succeeded at 3583 (offset 9 lines).
Hunk #13 succeeded at 3617 (offset 9 lines).
Hunk #14 succeeded at 3713 (offset 9 lines).
Hunk #15 succeeded at 3868 (offset 9 lines).
1 out of 15 hunks FAILED -- saving rejects to file
drivers/net/wireless/ath/ath5k/base.c.rej
patching file drivers/net/wireless/ath/ath5k/base.h
patching file drivers/net/wireless/ath/ath5k/eeprom.c
Applied patch revert-ath5k-patches/0008-Revert-ath5k-Introduce-ath5k_init_softc-function-as-.patch
(forced; needs refresh)
- EOT -

Re: [linux-next] next-20101207: Better now, but modinfo ath5k

[John W. Linville]

On Tue, Dec 07, 2010 at 04:57:08PM +0100, Sedat Dilek wrote:
> On Tue, Dec 7, 2010 at 3:26 PM, John W. Linville <linville@tuxdriver.com> wrote:
> > On Tue, Dec 07, 2010 at 12:24:14PM +0100, Sedat Dilek wrote:
> >
> >> I have already reported on linux-wireless (short: l-w) ML that after
> >> especially the merge of "AHB Bus support" patchset (see [2]) my ath5k
> >> was not working as expected in daily use (also modinfo).
> >>
> >> I am not sure if it is the driver but reverting the 9 commits in my
> >> local linux-next GIT seemed to "stabilize" a bit the Internet
> >> connection.
> >> The patchset from Nick (see below) definitely helped to stabilize ath5k.
> >
> > I imagine that these are the "9 commits" you mean:
> >
> >> git rev-list c30ae13..4cebb34
> > 4cebb34caa5122216a1e2451eae9e0fc47ec2589
> > a0b907ee2a71052fefdf6151764095f3f97b3275
> > c31b5c9c806b1fbdc9e98885d897664a0d482989
> > e7aecd327d80b2f156b54769013aaccb2a20645a
> > 8efa5d7d6ad307ae2d220def37ca89594062c40d
> > 4aa5d783c9e1c72e4950ff34f388077ccecac74a
> > e5b046d86fac609f636d127a38de94a175c7e83b
> > 132b1c3ee38ea6fa0501004fd0f19acb554e5a44
> > aeae4ac9090462ea38387dcdbac4f01b944af6a4
> >
> > Most of those look fairly harmless so long as you don't define
> > CONFIG_ATHEROS_AR231X.  Just to narrow things down...if you only revert
> > the first 8 (i.e. leave aeae4ac alone), do you still have problems?
> > Or does it still "stabilize"?
> >
> > That patch converts the driver to the generic DMA API.  It seems
> > OK at first glance, but I suspect it is the only one likely to be
> > effecting your situation.
> >
> > John
> > --
> > John W. Linville                Someday the world will need a hero, and you
> > linville@tuxdriver.com                  might be all we have.  Be ready.
> >
> 
> Before jumping to another problem, I had a quick look into the
> appropriate Kconfig file, as you say I have not CONFIG_ATHEROS_AR231X
> set and that's why AHB support is not selected/built, instead
> 
>   CC [M]  drivers/net/wireless/ath/ath5k/pci.o
> 
> is built.
> 
> Unfortunately, I can't cleanly revert. Sorry, I can't help.

I don't quite understand -- you said you were reverting them before,
but now I ask you to revert one _less_ patch (i.e. the last one to
be reverted anyway) and you can't do it?

John
-- 
John W. Linville		Someday the world will need a hero, and you
linville@tuxdriver.com			might be all we have.  Be ready.

Re: [linux-next] next-20101207: Better now, but modinfo ath5k

[Sedat Dilek]

On Tue, Dec 7, 2010 at 4:59 PM, John W. Linville <linville@tuxdriver.com> wrote:
> On Tue, Dec 07, 2010 at 04:57:08PM +0100, Sedat Dilek wrote:
>> On Tue, Dec 7, 2010 at 3:26 PM, John W. Linville <linville@tuxdriver.com> wrote:
>> > On Tue, Dec 07, 2010 at 12:24:14PM +0100, Sedat Dilek wrote:
>> >
>> >> I have already reported on linux-wireless (short: l-w) ML that after
>> >> especially the merge of "AHB Bus support" patchset (see [2]) my ath5k
>> >> was not working as expected in daily use (also modinfo).
>> >>
>> >> I am not sure if it is the driver but reverting the 9 commits in my
>> >> local linux-next GIT seemed to "stabilize" a bit the Internet
>> >> connection.
>> >> The patchset from Nick (see below) definitely helped to stabilize ath5k.
>> >
>> > I imagine that these are the "9 commits" you mean:
>> >
>> >> git rev-list c30ae13..4cebb34
>> > 4cebb34caa5122216a1e2451eae9e0fc47ec2589
>> > a0b907ee2a71052fefdf6151764095f3f97b3275
>> > c31b5c9c806b1fbdc9e98885d897664a0d482989
>> > e7aecd327d80b2f156b54769013aaccb2a20645a
>> > 8efa5d7d6ad307ae2d220def37ca89594062c40d
>> > 4aa5d783c9e1c72e4950ff34f388077ccecac74a
>> > e5b046d86fac609f636d127a38de94a175c7e83b
>> > 132b1c3ee38ea6fa0501004fd0f19acb554e5a44
>> > aeae4ac9090462ea38387dcdbac4f01b944af6a4
>> >
>> > Most of those look fairly harmless so long as you don't define
>> > CONFIG_ATHEROS_AR231X.  Just to narrow things down...if you only revert
>> > the first 8 (i.e. leave aeae4ac alone), do you still have problems?
>> > Or does it still "stabilize"?
>> >
>> > That patch converts the driver to the generic DMA API.  It seems
>> > OK at first glance, but I suspect it is the only one likely to be
>> > effecting your situation.
>> >
>> > John
>> > --
>> > John W. Linville                Someday the world will need a hero, and you
>> > linville@tuxdriver.com                  might be all we have.  Be ready.
>> >
>>
>> Before jumping to another problem, I had a quick look into the
>> appropriate Kconfig file, as you say I have not CONFIG_ATHEROS_AR231X
>> set and that's why AHB support is not selected/built, instead
>>
>>   CC [M]  drivers/net/wireless/ath/ath5k/pci.o
>>
>> is built.
>>
>> Unfortunately, I can't cleanly revert. Sorry, I can't help.
>
> I don't quite understand -- you said you were reverting them before,
> but now I ask you to revert one _less_ patch (i.e. the last one to
> be reverted anyway) and you can't do it?
>
> John
> --
> John W. Linville                Someday the world will need a hero, and you
> linville@tuxdriver.com                  might be all we have.  Be ready.
>

After AHB-support patches went into linux-next (next-20101207), there
are now new ath5k patches touching
"drivers/net/wireless/ath/ath5k/base.c" on top.
For example the patchset from Nick, this would mean revert them, too.
Doing so, won't give you a clear answer.

BTW, 0008 patch has a big change:

 drivers/net/wireless/ath/ath5k/base.c   |  344 +++++++++++++------------------

Another solution would be to take a previous linux-next where those
patches were not in.
As you say by yourself 0001..0008 do not hurt, it's 0009 patch (first
in patchset) you are speculating on.

As I have archived my "compiled" build-trees, I can look at them.
AFAICS linux-next (next-20101203) should have wireless-next-2.6
(master-2010-12-02) [1] included.
Also, I have to find the fitting linux-image binary and install it,
hope I have not purged it from hdd.

- Sedat -

[1] http://git.kernel.org/?p=linux/kernel/git/linville/wireless-next-2.6.git;a=shortlog;h=refs/tags/master-2010-12-02

$ ls revert-ath5k-patches/000*
revert-ath5k-patches/0001-Revert-ath5k-Fix-reset-and-interrupts-for-AHB-type-o.patch
revert-ath5k-patches/0002-Revert-ath5k-Add-AHB-bus-support.patch
revert-ath5k-patches/0003-Revert-ath5k-Add-initial-registers-values-for-radio-.patch
revert-ath5k-patches/0004-Revert-ath5k-Add-a-function-to-read-chipset-s-MAC-re.patch
revert-ath5k-patches/0005-Revert-ath5k-Check-if-pci-pdev-struct-is-initialized.patch
revert-ath5k-patches/0006-Revert-ath5k-Use-generic-eeprom-read-from-common-ath.patch
revert-ath5k-patches/0007-Revert-ath5k-Move-PCI-bus-functions-to-separate-file.patch
revert-ath5k-patches/0008-Revert-ath5k-Introduce-ath5k_init_softc-function-as-.patch
revert-ath5k-patches/0009-Revert-ath5k-Use-Generic-DMA-for-later-support-of-AH.patch

Re: [linux-next] next-20101207: Better now, but modinfo ath5k

[Sedat Dilek]

[-- Attachment #1: Type: text/plain, Size: 5567 bytes --]

On Tue, Dec 7, 2010 at 4:59 PM, John W. Linville <linville@tuxdriver.com> wrote:
> On Tue, Dec 07, 2010 at 04:57:08PM +0100, Sedat Dilek wrote:
>> On Tue, Dec 7, 2010 at 3:26 PM, John W. Linville <linville@tuxdriver.com> wrote:
>> > On Tue, Dec 07, 2010 at 12:24:14PM +0100, Sedat Dilek wrote:
>> >
>> >> I have already reported on linux-wireless (short: l-w) ML that after
>> >> especially the merge of "AHB Bus support" patchset (see [2]) my ath5k
>> >> was not working as expected in daily use (also modinfo).
>> >>
>> >> I am not sure if it is the driver but reverting the 9 commits in my
>> >> local linux-next GIT seemed to "stabilize" a bit the Internet
>> >> connection.
>> >> The patchset from Nick (see below) definitely helped to stabilize ath5k.
>> >
>> > I imagine that these are the "9 commits" you mean:
>> >
>> >> git rev-list c30ae13..4cebb34
>> > 4cebb34caa5122216a1e2451eae9e0fc47ec2589
>> > a0b907ee2a71052fefdf6151764095f3f97b3275
>> > c31b5c9c806b1fbdc9e98885d897664a0d482989
>> > e7aecd327d80b2f156b54769013aaccb2a20645a
>> > 8efa5d7d6ad307ae2d220def37ca89594062c40d
>> > 4aa5d783c9e1c72e4950ff34f388077ccecac74a
>> > e5b046d86fac609f636d127a38de94a175c7e83b
>> > 132b1c3ee38ea6fa0501004fd0f19acb554e5a44
>> > aeae4ac9090462ea38387dcdbac4f01b944af6a4
>> >
>> > Most of those look fairly harmless so long as you don't define
>> > CONFIG_ATHEROS_AR231X.  Just to narrow things down...if you only revert
>> > the first 8 (i.e. leave aeae4ac alone), do you still have problems?
>> > Or does it still "stabilize"?
>> >
>> > That patch converts the driver to the generic DMA API.  It seems
>> > OK at first glance, but I suspect it is the only one likely to be
>> > effecting your situation.
>> >
>> > John
>> > --
>> > John W. Linville                Someday the world will need a hero, and you
>> > linville@tuxdriver.com                  might be all we have.  Be ready.
>> >
>>
>> Before jumping to another problem, I had a quick look into the
>> appropriate Kconfig file, as you say I have not CONFIG_ATHEROS_AR231X
>> set and that's why AHB support is not selected/built, instead
>>
>>   CC [M]  drivers/net/wireless/ath/ath5k/pci.o
>>
>> is built.
>>
>> Unfortunately, I can't cleanly revert. Sorry, I can't help.
>
> I don't quite understand -- you said you were reverting them before,
> but now I ask you to revert one _less_ patch (i.e. the last one to
> be reverted anyway) and you can't do it?
>
> John
> --
> John W. Linville                Someday the world will need a hero, and you
> linville@tuxdriver.com                  might be all we have.  Be ready.
>

I tested with a linux-next (next-20101206) with revert-at5k-patches.
Unfortunately, Nick's patchset is needed for a stable WLAN connection,
several ReAssoc's etc.
So, I can't give a comment to the effects of the AHB patchset on my AR5212.

But... I have found the culprit patch which makes modinfo no more list
the aliases/pci-id (see [1]):

commit e5b046d86fac609f636d127a38de94a175c7e83b
"ath5k: Move PCI bus functions to separate file."

- Sedat -

[1] http://git.kernel.org/?p=linux/kernel/git/linville/wireless-next-2.6.git;a=commit;h=e5b046d86fac609f636d127a38de94a175c7e83b

$ ~/src/linux-2.6/linux-2.6.37-rc4/debian/build/source_i386_none

$ cat .pc/applied-patches
revert-ath5k-patches/0001-Revert-ath5k-Fix-reset-and-interrupts-for-AHB-type-o.patch
revert-ath5k-patches/0002-Revert-ath5k-Add-AHB-bus-support.patch
revert-ath5k-patches/0003-Revert-ath5k-Add-initial-registers-values-for-radio-.patch
revert-ath5k-patches/0004-Revert-ath5k-Add-a-function-to-read-chipset-s-MAC-re.patch
revert-ath5k-patches/0005-Revert-ath5k-Check-if-pci-pdev-struct-is-initialized.patch
revert-ath5k-patches/0006-Revert-ath5k-Use-generic-eeprom-read-from-common-ath.patch
revert-ath5k-patches/0007-Revert-ath5k-Move-PCI-bus-functions-to-separate-file.patch

# modinfo ath5k
filename:
/lib/modules/2.6.37-rc4-686/kernel/drivers/net/wireless/ath/ath5k/ath5k.ko
version:        0.6.0 (EXPERIMENTAL)
license:        Dual BSD/GPL
description:    Support for 5xxx series of Atheros 802.11 wireless LAN
cards.
author:         Nick Kossifidis
author:         Jiri Slaby
srcversion:     666105BAEE881F2C2F10CE6
alias:          pci:v0000168Cd0000001Dsv*sd*bc*sc*i*
alias:          pci:v0000168Cd0000001Csv*sd*bc*sc*i*
alias:          pci:v0000168Cd0000001Bsv*sd*bc*sc*i*
alias:          pci:v0000168Cd0000001Asv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000019sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000018sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000017sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000016sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000015sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000014sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00001014sv*sd*bc*sc*i*
alias:          pci:v000010B7d00000013sv*sd*bc*sc*i*
alias:          pci:v0000A727d00000013sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000013sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000012sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000011sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000007sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000207sv*sd*bc*sc*i*
depends:        mac80211,cfg80211,ath
vermagic:       2.6.37-rc4-686 SMP mod_unload modversions 686
parm:           debug:uint
parm:           nohwcrypt:Disable hardware encryption. (bool)
parm:           all_channels:Expose all channels the device can use. (bool)

[-- Attachment #2: 0007-Revert-ath5k-Move-PCI-bus-functions-to-separate-file.patch --]
[-- Type: text/x-diff, Size: 18973 bytes --]

From e2f79441224cf9cd1f9e399ca774040f1ece72fa Mon Sep 17 00:00:00 2001
From: Sedat Dilek <sedat.dilek@gmail.com>
Date: Fri, 3 Dec 2010 23:19:54 +0100
Subject: [PATCH 7/9] Revert "ath5k: Move PCI bus functions to separate file."

This reverts commit e5b046d86fac609f636d127a38de94a175c7e83b.
---
 drivers/net/wireless/ath/ath5k/Makefile |    1 -
 drivers/net/wireless/ath/ath5k/ath5k.h  |    3 -
 drivers/net/wireless/ath/ath5k/base.c   |  267 +++++++++++++++++++++++++++++-
 drivers/net/wireless/ath/ath5k/pci.c    |  284 -------------------------------
 4 files changed, 266 insertions(+), 289 deletions(-)
 delete mode 100644 drivers/net/wireless/ath/ath5k/pci.c

diff --git a/drivers/net/wireless/ath/ath5k/Makefile b/drivers/net/wireless/ath/ath5k/Makefile
index dae881c..2242a14 100644
--- a/drivers/net/wireless/ath/ath5k/Makefile
+++ b/drivers/net/wireless/ath/ath5k/Makefile
@@ -15,5 +15,4 @@ ath5k-y				+= rfkill.o
 ath5k-y				+= ani.o
 ath5k-y				+= sysfs.o
 ath5k-$(CONFIG_ATH5K_DEBUG)	+= debug.o
-ath5k-y				+= pci.o
 obj-$(CONFIG_ATH5K)		+= ath5k.o
diff --git a/drivers/net/wireless/ath/ath5k/ath5k.h b/drivers/net/wireless/ath/ath5k/ath5k.h
index ee3c0af..a74f448 100644
--- a/drivers/net/wireless/ath/ath5k/ath5k.h
+++ b/drivers/net/wireless/ath/ath5k/ath5k.h
@@ -1145,7 +1145,6 @@ struct ath5k_hw {
 /*
  * Prototypes
  */
-extern const struct ieee80211_ops ath5k_hw_ops;
 
 /* Initialization and detach functions */
 int ath5k_init_softc(struct ath5k_softc *sc, const struct ath_bus_ops *bus_ops);
@@ -1156,8 +1155,6 @@ void ath5k_hw_deinit(struct ath5k_hw *ah);
 int ath5k_sysfs_register(struct ath5k_softc *sc);
 void ath5k_sysfs_unregister(struct ath5k_softc *sc);
 
-/*Chip id helper function */
-const char *ath5k_chip_name(enum ath5k_srev_type type, u_int16_t val);
 
 /* LED functions */
 int ath5k_init_leds(struct ath5k_softc *sc);
diff --git a/drivers/net/wireless/ath/ath5k/base.c b/drivers/net/wireless/ath/ath5k/base.c
index 9fcc4e1..b11ea3d 100644
--- a/drivers/net/wireless/ath/ath5k/base.c
+++ b/drivers/net/wireless/ath/ath5k/base.c
@@ -47,6 +47,8 @@
 #include <linux/io.h>
 #include <linux/netdevice.h>
 #include <linux/cache.h>
+#include <linux/pci.h>
+#include <linux/pci-aspm.h>
 #include <linux/ethtool.h>
 #include <linux/uaccess.h>
 #include <linux/slab.h>
@@ -85,6 +87,30 @@ static int ath5k_beacon_update(struct ieee80211_hw *hw,
 		struct ieee80211_vif *vif);
 static void ath5k_beacon_update_timers(struct ath5k_softc *sc, u64 bc_tsf);
 
+/* Known PCI ids */
+static DEFINE_PCI_DEVICE_TABLE(ath5k_pci_id_table) = {
+	{ PCI_VDEVICE(ATHEROS, 0x0207) }, /* 5210 early */
+	{ PCI_VDEVICE(ATHEROS, 0x0007) }, /* 5210 */
+	{ PCI_VDEVICE(ATHEROS, 0x0011) }, /* 5311 - this is on AHB bus !*/
+	{ PCI_VDEVICE(ATHEROS, 0x0012) }, /* 5211 */
+	{ PCI_VDEVICE(ATHEROS, 0x0013) }, /* 5212 */
+	{ PCI_VDEVICE(3COM_2,  0x0013) }, /* 3com 5212 */
+	{ PCI_VDEVICE(3COM,    0x0013) }, /* 3com 3CRDAG675 5212 */
+	{ PCI_VDEVICE(ATHEROS, 0x1014) }, /* IBM minipci 5212 */
+	{ PCI_VDEVICE(ATHEROS, 0x0014) }, /* 5212 combatible */
+	{ PCI_VDEVICE(ATHEROS, 0x0015) }, /* 5212 combatible */
+	{ PCI_VDEVICE(ATHEROS, 0x0016) }, /* 5212 combatible */
+	{ PCI_VDEVICE(ATHEROS, 0x0017) }, /* 5212 combatible */
+	{ PCI_VDEVICE(ATHEROS, 0x0018) }, /* 5212 combatible */
+	{ PCI_VDEVICE(ATHEROS, 0x0019) }, /* 5212 combatible */
+	{ PCI_VDEVICE(ATHEROS, 0x001a) }, /* 2413 Griffin-lite */
+	{ PCI_VDEVICE(ATHEROS, 0x001b) }, /* 5413 Eagle */
+	{ PCI_VDEVICE(ATHEROS, 0x001c) }, /* PCI-E cards */
+	{ PCI_VDEVICE(ATHEROS, 0x001d) }, /* 2417 Nala */
+	{ 0 }
+};
+MODULE_DEVICE_TABLE(pci, ath5k_pci_id_table);
+
 /* Known SREVs */
 static const struct ath5k_srev_name srev_names[] = {
 	{ "5210",	AR5K_VERSION_MAC,	AR5K_SREV_AR5210 },
@@ -167,6 +193,32 @@ static const struct ieee80211_rate ath5k_rates[] = {
 	/* XR missing */
 };
 
+/* return bus cachesize in 4B word units */
+static void ath5k_pci_read_cachesize(struct ath_common *common, int *csz)
+{
+	struct ath5k_softc *sc = (struct ath5k_softc *) common->priv;
+	u8 u8tmp;
+
+	pci_read_config_byte(sc->pdev, PCI_CACHE_LINE_SIZE, &u8tmp);
+	*csz = (int)u8tmp;
+
+	/*
+	 * This check was put in to avoid "unplesant" consequences if
+	 * the bootrom has not fully initialized all PCI devices.
+	 * Sometimes the cache line size register is not set
+	 */
+
+	if (*csz == 0)
+		*csz = L1_CACHE_BYTES >> 2;   /* Use the default size */
+}
+
+/* Common ath_bus_opts structure */
+static const struct ath_bus_ops ath_pci_bus_ops = {
+	.ath_bus_type = ATH_PCI,
+	.read_cachesize = ath5k_pci_read_cachesize,
+};
+
+
 static inline void ath5k_txbuf_free_skb(struct ath5k_softc *sc,
 				struct ath5k_buf *bf)
 {
@@ -209,7 +261,7 @@ static inline u64 ath5k_extend_tsf(struct ath5k_hw *ah, u32 rstamp)
 	return (tsf & ~0x7fff) | rstamp;
 }
 
-const char *
+static const char *
 ath5k_chip_name(enum ath5k_srev_type type, u_int16_t val)
 {
 	const char *name = "xxxxx";
@@ -3575,3 +3627,216 @@ const struct ieee80211_ops ath5k_hw_ops = {
 	.set_antenna	= ath5k_set_antenna,
 	.get_antenna	= ath5k_get_antenna,
 };
+
+/********************\
+* PCI Initialization *
+\********************/
+
+static int __devinit
+ath5k_pci_probe(struct pci_dev *pdev,
+		const struct pci_device_id *id)
+{
+	void __iomem *mem;
+	struct ath5k_softc *sc;
+	struct ieee80211_hw *hw;
+	int ret;
+	u8 csz;
+
+	/*
+	 * L0s needs to be disabled on all ath5k cards.
+	 *
+	 * For distributions shipping with CONFIG_PCIEASPM (this will be enabled
+	 * by default in the future in 2.6.36) this will also mean both L1 and
+	 * L0s will be disabled when a pre 1.1 PCIe device is detected. We do
+	 * know L1 works correctly even for all ath5k pre 1.1 PCIe devices
+	 * though but cannot currently undue the effect of a blacklist, for
+	 * details you can read pcie_aspm_sanity_check() and see how it adjusts
+	 * the device link capability.
+	 *
+	 * It may be possible in the future to implement some PCI API to allow
+	 * drivers to override blacklists for pre 1.1 PCIe but for now it is
+	 * best to accept that both L0s and L1 will be disabled completely for
+	 * distributions shipping with CONFIG_PCIEASPM rather than having this
+	 * issue present. Motivation for adding this new API will be to help
+	 * with power consumption for some of these devices.
+	 */
+	pci_disable_link_state(pdev, PCIE_LINK_STATE_L0S);
+
+	ret = pci_enable_device(pdev);
+	if (ret) {
+		dev_err(&pdev->dev, "can't enable device\n");
+		goto err;
+	}
+
+	/* XXX 32-bit addressing only */
+	ret = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
+	if (ret) {
+		dev_err(&pdev->dev, "32-bit DMA not available\n");
+		goto err_dis;
+	}
+
+	/*
+	 * Cache line size is used to size and align various
+	 * structures used to communicate with the hardware.
+	 */
+	pci_read_config_byte(pdev, PCI_CACHE_LINE_SIZE, &csz);
+	if (csz == 0) {
+		/*
+		 * Linux 2.4.18 (at least) writes the cache line size
+		 * register as a 16-bit wide register which is wrong.
+		 * We must have this setup properly for rx buffer
+		 * DMA to work so force a reasonable value here if it
+		 * comes up zero.
+		 */
+		csz = L1_CACHE_BYTES >> 2;
+		pci_write_config_byte(pdev, PCI_CACHE_LINE_SIZE, csz);
+	}
+	/*
+	 * The default setting of latency timer yields poor results,
+	 * set it to the value used by other systems.  It may be worth
+	 * tweaking this setting more.
+	 */
+	pci_write_config_byte(pdev, PCI_LATENCY_TIMER, 0xa8);
+
+	/* Enable bus mastering */
+	pci_set_master(pdev);
+
+	/*
+	 * Disable the RETRY_TIMEOUT register (0x41) to keep
+	 * PCI Tx retries from interfering with C3 CPU state.
+	 */
+	pci_write_config_byte(pdev, 0x41, 0);
+
+	ret = pci_request_region(pdev, 0, "ath5k");
+	if (ret) {
+		dev_err(&pdev->dev, "cannot reserve PCI memory region\n");
+		goto err_dis;
+	}
+
+	mem = pci_iomap(pdev, 0, 0);
+	if (!mem) {
+		dev_err(&pdev->dev, "cannot remap PCI memory region\n") ;
+		ret = -EIO;
+		goto err_reg;
+	}
+
+	/*
+	 * Allocate hw (mac80211 main struct)
+	 * and hw->priv (driver private data)
+	 */
+	hw = ieee80211_alloc_hw(sizeof(*sc), &ath5k_hw_ops);
+	if (hw == NULL) {
+		dev_err(&pdev->dev, "cannot allocate ieee80211_hw\n");
+		ret = -ENOMEM;
+		goto err_map;
+	}
+
+	dev_info(&pdev->dev, "registered as '%s'\n", wiphy_name(hw->wiphy));
+
+	sc = hw->priv;
+	sc->hw = hw;
+	sc->pdev = pdev;
+	sc->dev = &pdev->dev;
+	sc->irq = pdev->irq;
+	sc->devid = id->device;
+	sc->iobase = mem; /* So we can unmap it on detach */
+
+	/* Initialize */
+	ret = ath5k_init_softc(sc, &ath_pci_bus_ops);
+	if (ret) {
+		goto err_free;
+	}
+
+	/* Set private data */
+	pci_set_drvdata(pdev, hw);
+
+	return 0;
+err_free:
+	ieee80211_free_hw(hw);
+err_map:
+	pci_iounmap(pdev, mem);
+err_reg:
+	pci_release_region(pdev, 0);
+err_dis:
+	pci_disable_device(pdev);
+err:
+	return ret;
+}
+
+static void __devexit
+ath5k_pci_remove(struct pci_dev *pdev)
+{
+	struct ieee80211_hw *hw = pci_get_drvdata(pdev);
+	struct ath5k_softc *sc = hw->priv;
+
+	ath5k_deinit_softc(sc);
+	pci_iounmap(pdev, sc->iobase);
+	pci_release_region(pdev, 0);
+	pci_disable_device(pdev);
+	ieee80211_free_hw(hw);
+}
+
+#ifdef CONFIG_PM_SLEEP
+static int ath5k_pci_suspend(struct device *dev)
+{
+	struct ath5k_softc *sc = pci_get_drvdata(to_pci_dev(dev));
+
+	ath5k_led_off(sc);
+	return 0;
+}
+
+static int ath5k_pci_resume(struct device *dev)
+{
+	struct pci_dev *pdev = to_pci_dev(dev);
+	struct ath5k_softc *sc = pci_get_drvdata(pdev);
+
+	/*
+	 * Suspend/Resume resets the PCI configuration space, so we have to
+	 * re-disable the RETRY_TIMEOUT register (0x41) to keep
+	 * PCI Tx retries from interfering with C3 CPU state
+	 */
+	pci_write_config_byte(pdev, 0x41, 0);
+
+	ath5k_led_enable(sc);
+	return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(ath5k_pm_ops, ath5k_pci_suspend, ath5k_pci_resume);
+#define ATH5K_PM_OPS	(&ath5k_pm_ops)
+#else
+#define ATH5K_PM_OPS	NULL
+#endif /* CONFIG_PM_SLEEP */
+
+static struct pci_driver ath5k_pci_driver = {
+	.name		= KBUILD_MODNAME,
+	.id_table	= ath5k_pci_id_table,
+	.probe		= ath5k_pci_probe,
+	.remove		= __devexit_p(ath5k_pci_remove),
+	.driver.pm	= ATH5K_PM_OPS,
+};
+
+/*
+ * Module init/exit functions
+ */
+static int __init
+init_ath5k_pci(void)
+{
+	int ret;
+
+	ret = pci_register_driver(&ath5k_pci_driver);
+	if (ret) {
+		printk(KERN_ERR "ath5k_pci: can't register pci driver\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static void __exit
+exit_ath5k_pci(void)
+{
+	pci_unregister_driver(&ath5k_pci_driver);
+}
+
+module_init(init_ath5k_pci);
+module_exit(exit_ath5k_pci);
diff --git a/drivers/net/wireless/ath/ath5k/pci.c b/drivers/net/wireless/ath/ath5k/pci.c
deleted file mode 100644
index 9f7d3ca..0000000
--- a/drivers/net/wireless/ath/ath5k/pci.c
+++ /dev/null
@@ -1,284 +0,0 @@
-/*
- * Copyright (c) 2008-2009 Atheros Communications Inc.
- *
- * Permission to use, copy, modify, and/or distribute this software for any
- * purpose with or without fee is hereby granted, provided that the above
- * copyright notice and this permission notice appear in all copies.
- *
- * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
- * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
- * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
- * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
- * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
- * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
- * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
- */
-
-#include <linux/nl80211.h>
-#include <linux/pci.h>
-#include <linux/pci-aspm.h>
-#include "../ath.h"
-#include "ath5k.h"
-#include "debug.h"
-#include "base.h"
-#include "reg.h"
-
-/* Known PCI ids */
-static DEFINE_PCI_DEVICE_TABLE(ath5k_pci_id_table) = {
-	{ PCI_VDEVICE(ATHEROS, 0x0207) }, /* 5210 early */
-	{ PCI_VDEVICE(ATHEROS, 0x0007) }, /* 5210 */
-	{ PCI_VDEVICE(ATHEROS, 0x0011) }, /* 5311 - this is on AHB bus !*/
-	{ PCI_VDEVICE(ATHEROS, 0x0012) }, /* 5211 */
-	{ PCI_VDEVICE(ATHEROS, 0x0013) }, /* 5212 */
-	{ PCI_VDEVICE(3COM_2,  0x0013) }, /* 3com 5212 */
-	{ PCI_VDEVICE(3COM,    0x0013) }, /* 3com 3CRDAG675 5212 */
-	{ PCI_VDEVICE(ATHEROS, 0x1014) }, /* IBM minipci 5212 */
-	{ PCI_VDEVICE(ATHEROS, 0x0014) }, /* 5212 combatible */
-	{ PCI_VDEVICE(ATHEROS, 0x0015) }, /* 5212 combatible */
-	{ PCI_VDEVICE(ATHEROS, 0x0016) }, /* 5212 combatible */
-	{ PCI_VDEVICE(ATHEROS, 0x0017) }, /* 5212 combatible */
-	{ PCI_VDEVICE(ATHEROS, 0x0018) }, /* 5212 combatible */
-	{ PCI_VDEVICE(ATHEROS, 0x0019) }, /* 5212 combatible */
-	{ PCI_VDEVICE(ATHEROS, 0x001a) }, /* 2413 Griffin-lite */
-	{ PCI_VDEVICE(ATHEROS, 0x001b) }, /* 5413 Eagle */
-	{ PCI_VDEVICE(ATHEROS, 0x001c) }, /* PCI-E cards */
-	{ PCI_VDEVICE(ATHEROS, 0x001d) }, /* 2417 Nala */
-	{ 0 }
-};
-
-/* return bus cachesize in 4B word units */
-static void ath5k_pci_read_cachesize(struct ath_common *common, int *csz)
-{
-	struct ath5k_softc *sc = (struct ath5k_softc *) common->priv;
-	u8 u8tmp;
-
-	pci_read_config_byte(sc->pdev, PCI_CACHE_LINE_SIZE, &u8tmp);
-	*csz = (int)u8tmp;
-
-	/*
-	 * This check was put in to avoid "unplesant" consequences if
-	 * the bootrom has not fully initialized all PCI devices.
-	 * Sometimes the cache line size register is not set
-	 */
-
-	if (*csz == 0)
-		*csz = L1_CACHE_BYTES >> 2;   /* Use the default size */
-}
-
-/* Common ath_bus_opts structure */
-static const struct ath_bus_ops ath_pci_bus_ops = {
-	.ath_bus_type = ATH_PCI,
-	.read_cachesize = ath5k_pci_read_cachesize,
-};
-
-/********************\
-* PCI Initialization *
-\********************/
-
-static int __devinit
-ath5k_pci_probe(struct pci_dev *pdev,
-		const struct pci_device_id *id)
-{
-	void __iomem *mem;
-	struct ath5k_softc *sc;
-	struct ieee80211_hw *hw;
-	int ret;
-	u8 csz;
-
-	/*
-	 * L0s needs to be disabled on all ath5k cards.
-	 *
-	 * For distributions shipping with CONFIG_PCIEASPM (this will be enabled
-	 * by default in the future in 2.6.36) this will also mean both L1 and
-	 * L0s will be disabled when a pre 1.1 PCIe device is detected. We do
-	 * know L1 works correctly even for all ath5k pre 1.1 PCIe devices
-	 * though but cannot currently undue the effect of a blacklist, for
-	 * details you can read pcie_aspm_sanity_check() and see how it adjusts
-	 * the device link capability.
-	 *
-	 * It may be possible in the future to implement some PCI API to allow
-	 * drivers to override blacklists for pre 1.1 PCIe but for now it is
-	 * best to accept that both L0s and L1 will be disabled completely for
-	 * distributions shipping with CONFIG_PCIEASPM rather than having this
-	 * issue present. Motivation for adding this new API will be to help
-	 * with power consumption for some of these devices.
-	 */
-	pci_disable_link_state(pdev, PCIE_LINK_STATE_L0S);
-
-	ret = pci_enable_device(pdev);
-	if (ret) {
-		dev_err(&pdev->dev, "can't enable device\n");
-		goto err;
-	}
-
-	/* XXX 32-bit addressing only */
-	ret = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
-	if (ret) {
-		dev_err(&pdev->dev, "32-bit DMA not available\n");
-		goto err_dis;
-	}
-
-	/*
-	 * Cache line size is used to size and align various
-	 * structures used to communicate with the hardware.
-	 */
-	pci_read_config_byte(pdev, PCI_CACHE_LINE_SIZE, &csz);
-	if (csz == 0) {
-		/*
-		 * Linux 2.4.18 (at least) writes the cache line size
-		 * register as a 16-bit wide register which is wrong.
-		 * We must have this setup properly for rx buffer
-		 * DMA to work so force a reasonable value here if it
-		 * comes up zero.
-		 */
-		csz = L1_CACHE_BYTES >> 2;
-		pci_write_config_byte(pdev, PCI_CACHE_LINE_SIZE, csz);
-	}
-	/*
-	 * The default setting of latency timer yields poor results,
-	 * set it to the value used by other systems.  It may be worth
-	 * tweaking this setting more.
-	 */
-	pci_write_config_byte(pdev, PCI_LATENCY_TIMER, 0xa8);
-
-	/* Enable bus mastering */
-	pci_set_master(pdev);
-
-	/*
-	 * Disable the RETRY_TIMEOUT register (0x41) to keep
-	 * PCI Tx retries from interfering with C3 CPU state.
-	 */
-	pci_write_config_byte(pdev, 0x41, 0);
-
-	ret = pci_request_region(pdev, 0, "ath5k");
-	if (ret) {
-		dev_err(&pdev->dev, "cannot reserve PCI memory region\n");
-		goto err_dis;
-	}
-
-	mem = pci_iomap(pdev, 0, 0);
-	if (!mem) {
-		dev_err(&pdev->dev, "cannot remap PCI memory region\n") ;
-		ret = -EIO;
-		goto err_reg;
-	}
-
-	/*
-	 * Allocate hw (mac80211 main struct)
-	 * and hw->priv (driver private data)
-	 */
-	hw = ieee80211_alloc_hw(sizeof(*sc), &ath5k_hw_ops);
-	if (hw == NULL) {
-		dev_err(&pdev->dev, "cannot allocate ieee80211_hw\n");
-		ret = -ENOMEM;
-		goto err_map;
-	}
-
-	dev_info(&pdev->dev, "registered as '%s'\n", wiphy_name(hw->wiphy));
-
-	sc = hw->priv;
-	sc->hw = hw;
-	sc->pdev = pdev;
-	sc->dev = &pdev->dev;
-	sc->irq = pdev->irq;
-	sc->devid = id->device;
-	sc->iobase = mem; /* So we can unmap it on detach */
-
-	/* Initialize */
-	ret = ath5k_init_softc(sc, &ath_pci_bus_ops);
-	if (ret)
-		goto err_free;
-
-	/* Set private data */
-	pci_set_drvdata(pdev, hw);
-
-	return 0;
-err_free:
-	ieee80211_free_hw(hw);
-err_map:
-	pci_iounmap(pdev, mem);
-err_reg:
-	pci_release_region(pdev, 0);
-err_dis:
-	pci_disable_device(pdev);
-err:
-	return ret;
-}
-
-static void __devexit
-ath5k_pci_remove(struct pci_dev *pdev)
-{
-	struct ieee80211_hw *hw = pci_get_drvdata(pdev);
-	struct ath5k_softc *sc = hw->priv;
-
-	ath5k_deinit_softc(sc);
-	pci_iounmap(pdev, sc->iobase);
-	pci_release_region(pdev, 0);
-	pci_disable_device(pdev);
-	ieee80211_free_hw(hw);
-}
-
-#ifdef CONFIG_PM_SLEEP
-static int ath5k_pci_suspend(struct device *dev)
-{
-	struct ath5k_softc *sc = pci_get_drvdata(to_pci_dev(dev));
-
-	ath5k_led_off(sc);
-	return 0;
-}
-
-static int ath5k_pci_resume(struct device *dev)
-{
-	struct pci_dev *pdev = to_pci_dev(dev);
-	struct ath5k_softc *sc = pci_get_drvdata(pdev);
-
-	/*
-	 * Suspend/Resume resets the PCI configuration space, so we have to
-	 * re-disable the RETRY_TIMEOUT register (0x41) to keep
-	 * PCI Tx retries from interfering with C3 CPU state
-	 */
-	pci_write_config_byte(pdev, 0x41, 0);
-
-	ath5k_led_enable(sc);
-	return 0;
-}
-
-static SIMPLE_DEV_PM_OPS(ath5k_pm_ops, ath5k_pci_suspend, ath5k_pci_resume);
-#define ATH5K_PM_OPS	(&ath5k_pm_ops)
-#else
-#define ATH5K_PM_OPS	NULL
-#endif /* CONFIG_PM_SLEEP */
-
-static struct pci_driver ath5k_pci_driver = {
-	.name		= KBUILD_MODNAME,
-	.id_table	= ath5k_pci_id_table,
-	.probe		= ath5k_pci_probe,
-	.remove		= __devexit_p(ath5k_pci_remove),
-	.driver.pm	= ATH5K_PM_OPS,
-};
-
-/*
- * Module init/exit functions
- */
-static int __init
-init_ath5k_pci(void)
-{
-	int ret;
-
-	ret = pci_register_driver(&ath5k_pci_driver);
-	if (ret) {
-		printk(KERN_ERR "ath5k_pci: can't register pci driver\n");
-		return ret;
-	}
-
-	return 0;
-}
-
-static void __exit
-exit_ath5k_pci(void)
-{
-	pci_unregister_driver(&ath5k_pci_driver);
-}
-
-module_init(init_ath5k_pci);
-module_exit(exit_ath5k_pci);
-- 
1.7.2.3

Re: [linux-next] next-20101207: Better now, but modinfo ath5k

[Sedat Dilek]

On Tue, Dec 7, 2010 at 7:13 PM, Sedat Dilek <sedat.dilek@googlemail.com> wrote:
> On Tue, Dec 7, 2010 at 4:59 PM, John W. Linville <linville@tuxdriver.com> wrote:
>> On Tue, Dec 07, 2010 at 04:57:08PM +0100, Sedat Dilek wrote:
>>> On Tue, Dec 7, 2010 at 3:26 PM, John W. Linville <linville@tuxdriver.com> wrote:
>>> > On Tue, Dec 07, 2010 at 12:24:14PM +0100, Sedat Dilek wrote:
>>> >
>>> >> I have already reported on linux-wireless (short: l-w) ML that after
>>> >> especially the merge of "AHB Bus support" patchset (see [2]) my ath5k
>>> >> was not working as expected in daily use (also modinfo).
>>> >>
>>> >> I am not sure if it is the driver but reverting the 9 commits in my
>>> >> local linux-next GIT seemed to "stabilize" a bit the Internet
>>> >> connection.
>>> >> The patchset from Nick (see below) definitely helped to stabilize ath5k.
>>> >
>>> > I imagine that these are the "9 commits" you mean:
>>> >
>>> >> git rev-list c30ae13..4cebb34
>>> > 4cebb34caa5122216a1e2451eae9e0fc47ec2589
>>> > a0b907ee2a71052fefdf6151764095f3f97b3275
>>> > c31b5c9c806b1fbdc9e98885d897664a0d482989
>>> > e7aecd327d80b2f156b54769013aaccb2a20645a
>>> > 8efa5d7d6ad307ae2d220def37ca89594062c40d
>>> > 4aa5d783c9e1c72e4950ff34f388077ccecac74a
>>> > e5b046d86fac609f636d127a38de94a175c7e83b
>>> > 132b1c3ee38ea6fa0501004fd0f19acb554e5a44
>>> > aeae4ac9090462ea38387dcdbac4f01b944af6a4
>>> >
>>> > Most of those look fairly harmless so long as you don't define
>>> > CONFIG_ATHEROS_AR231X.  Just to narrow things down...if you only revert
>>> > the first 8 (i.e. leave aeae4ac alone), do you still have problems?
>>> > Or does it still "stabilize"?
>>> >
>>> > That patch converts the driver to the generic DMA API.  It seems
>>> > OK at first glance, but I suspect it is the only one likely to be
>>> > effecting your situation.
>>> >
>>> > John
>>> > --
>>> > John W. Linville                Someday the world will need a hero, and you
>>> > linville@tuxdriver.com                  might be all we have.  Be ready.
>>> >
>>>
>>> Before jumping to another problem, I had a quick look into the
>>> appropriate Kconfig file, as you say I have not CONFIG_ATHEROS_AR231X
>>> set and that's why AHB support is not selected/built, instead
>>>
>>>   CC [M]  drivers/net/wireless/ath/ath5k/pci.o
>>>
>>> is built.
>>>
>>> Unfortunately, I can't cleanly revert. Sorry, I can't help.
>>
>> I don't quite understand -- you said you were reverting them before,
>> but now I ask you to revert one _less_ patch (i.e. the last one to
>> be reverted anyway) and you can't do it?
>>
>> John
>> --
>> John W. Linville                Someday the world will need a hero, and you
>> linville@tuxdriver.com                  might be all we have.  Be ready.
>>
>
> I tested with a linux-next (next-20101206) with revert-at5k-patches.
> Unfortunately, Nick's patchset is needed for a stable WLAN connection,
> several ReAssoc's etc.
> So, I can't give a comment to the effects of the AHB patchset on my AR5212.
>
> But... I have found the culprit patch which makes modinfo no more list
> the aliases/pci-id (see [1]):
>
> commit e5b046d86fac609f636d127a38de94a175c7e83b
> "ath5k: Move PCI bus functions to separate file."
>
> - Sedat -
>
> [1] http://git.kernel.org/?p=linux/kernel/git/linville/wireless-next-2.6.git;a=commit;h=e5b046d86fac609f636d127a38de94a175c7e83b
>
> $ ~/src/linux-2.6/linux-2.6.37-rc4/debian/build/source_i386_none
>
> $ cat .pc/applied-patches
> revert-ath5k-patches/0001-Revert-ath5k-Fix-reset-and-interrupts-for-AHB-type-o.patch
> revert-ath5k-patches/0002-Revert-ath5k-Add-AHB-bus-support.patch
> revert-ath5k-patches/0003-Revert-ath5k-Add-initial-registers-values-for-radio-.patch
> revert-ath5k-patches/0004-Revert-ath5k-Add-a-function-to-read-chipset-s-MAC-re.patch
> revert-ath5k-patches/0005-Revert-ath5k-Check-if-pci-pdev-struct-is-initialized.patch
> revert-ath5k-patches/0006-Revert-ath5k-Use-generic-eeprom-read-from-common-ath.patch
> revert-ath5k-patches/0007-Revert-ath5k-Move-PCI-bus-functions-to-separate-file.patch
>
> # modinfo ath5k
> filename:
> /lib/modules/2.6.37-rc4-686/kernel/drivers/net/wireless/ath/ath5k/ath5k.ko
> version:        0.6.0 (EXPERIMENTAL)
> license:        Dual BSD/GPL
> description:    Support for 5xxx series of Atheros 802.11 wireless LAN
> cards.
> author:         Nick Kossifidis
> author:         Jiri Slaby
> srcversion:     666105BAEE881F2C2F10CE6
> alias:          pci:v0000168Cd0000001Dsv*sd*bc*sc*i*
> alias:          pci:v0000168Cd0000001Csv*sd*bc*sc*i*
> alias:          pci:v0000168Cd0000001Bsv*sd*bc*sc*i*
> alias:          pci:v0000168Cd0000001Asv*sd*bc*sc*i*
> alias:          pci:v0000168Cd00000019sv*sd*bc*sc*i*
> alias:          pci:v0000168Cd00000018sv*sd*bc*sc*i*
> alias:          pci:v0000168Cd00000017sv*sd*bc*sc*i*
> alias:          pci:v0000168Cd00000016sv*sd*bc*sc*i*
> alias:          pci:v0000168Cd00000015sv*sd*bc*sc*i*
> alias:          pci:v0000168Cd00000014sv*sd*bc*sc*i*
> alias:          pci:v0000168Cd00001014sv*sd*bc*sc*i*
> alias:          pci:v000010B7d00000013sv*sd*bc*sc*i*
> alias:          pci:v0000A727d00000013sv*sd*bc*sc*i*
> alias:          pci:v0000168Cd00000013sv*sd*bc*sc*i*
> alias:          pci:v0000168Cd00000012sv*sd*bc*sc*i*
> alias:          pci:v0000168Cd00000011sv*sd*bc*sc*i*
> alias:          pci:v0000168Cd00000007sv*sd*bc*sc*i*
> alias:          pci:v0000168Cd00000207sv*sd*bc*sc*i*
> depends:        mac80211,cfg80211,ath
> vermagic:       2.6.37-rc4-686 SMP mod_unload modversions 686
> parm:           debug:uint
> parm:           nohwcrypt:Disable hardware encryption. (bool)
> parm:           all_channels:Expose all channels the device can use. (bool)
>

I have found the cause for the modinfo problem, here is a fix (I will
send later a separate patch):

$ diff -Naur drivers/net/wireless/ath/ath5k/pci.c.orig
drivers/net/wireless/ath/ath5k/pci.c
--- drivers/net/wireless/ath/ath5k/pci.c.orig   2010-12-07
19:34:53.999334165 +0100
+++ drivers/net/wireless/ath/ath5k/pci.c        2010-12-07
22:25:20.387028188 +0100
@@ -45,6 +45,7 @@
        { PCI_VDEVICE(ATHEROS, 0x001d) }, /* 2417 Nala */
        { 0 }
 };
+MODULE_DEVICE_TABLE(pci, ath5k_pci_id_table);

 /* return bus cachesize in 4B word units */
 static void ath5k_pci_read_cachesize(struct ath_common *common, int *csz)

# modinfo ath5k
filename:
/lib/modules/2.6.37-rc4-686/kernel/drivers/net/wireless/ath/ath5k/ath5k.ko
version:        0.6.0 (EXPERIMENTAL)
license:        Dual BSD/GPL
description:    Support for 5xxx series of Atheros 802.11 wireless LAN
cards.
author:         Nick Kossifidis
author:         Jiri Slaby
srcversion:     1B224A604952CD0597CA568
alias:          pci:v0000168Cd0000001Dsv*sd*bc*sc*i*
alias:          pci:v0000168Cd0000001Csv*sd*bc*sc*i*
alias:          pci:v0000168Cd0000001Bsv*sd*bc*sc*i*
alias:          pci:v0000168Cd0000001Asv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000019sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000018sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000017sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000016sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000015sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000014sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00001014sv*sd*bc*sc*i*
alias:          pci:v000010B7d00000013sv*sd*bc*sc*i*
alias:          pci:v0000A727d00000013sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000013sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000012sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000011sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000007sv*sd*bc*sc*i*
alias:          pci:v0000168Cd00000207sv*sd*bc*sc*i*
depends:        mac80211,cfg80211,ath
vermagic:       2.6.37-rc4-686 SMP mod_unload modversions 686
parm:           debug:uint
parm:           nohwcrypt:Disable hardware encryption. (bool)
parm:           all_channels:Expose all channels the device can use. (bool)