2005-03-23 01:46:48

by Jeff Garzik

[permalink] [raw]
Subject: 2.6.x wireless update and status

BK users:

bk pull bk://gkernel.bkbits.net/wireless-2.6

Patch:
http://www.kernel.org/pub/linux/kernel/people/jgarzik/patchkits/2.6/2.6.12-rc1-bk1-wireless1.patch.bz2

This will update the following files:

drivers/net/wireless/ieee802_11.h | 78
MAINTAINERS | 7
drivers/net/wireless/Kconfig | 2
drivers/net/wireless/Makefile | 2
drivers/net/wireless/atmel.c | 62
drivers/net/wireless/hostap/Kconfig | 104
drivers/net/wireless/hostap/Makefile | 8
drivers/net/wireless/hostap/hostap.c | 1205 +++++++
drivers/net/wireless/hostap/hostap.h | 57
drivers/net/wireless/hostap/hostap_80211.h | 107
drivers/net/wireless/hostap/hostap_80211_rx.c | 1084 ++++++
drivers/net/wireless/hostap/hostap_80211_tx.c | 522 +++
drivers/net/wireless/hostap/hostap_ap.c | 3286 +++++++++++++++++++
drivers/net/wireless/hostap/hostap_ap.h | 272 +
drivers/net/wireless/hostap/hostap_common.h | 557 +++
drivers/net/wireless/hostap/hostap_config.h | 86
drivers/net/wireless/hostap/hostap_crypt.c | 167
drivers/net/wireless/hostap/hostap_crypt.h | 50
drivers/net/wireless/hostap/hostap_crypt_ccmp.c | 486 ++
drivers/net/wireless/hostap/hostap_crypt_tkip.c | 696 ++++
drivers/net/wireless/hostap/hostap_crypt_wep.c | 281 +
drivers/net/wireless/hostap/hostap_cs.c | 929 +++++
drivers/net/wireless/hostap/hostap_download.c | 766 ++++
drivers/net/wireless/hostap/hostap_hw.c | 3631 +++++++++++++++++++++
drivers/net/wireless/hostap/hostap_info.c | 478 ++
drivers/net/wireless/hostap/hostap_ioctl.c | 4116 ++++++++++++++++++++++++
drivers/net/wireless/hostap/hostap_pci.c | 453 ++
drivers/net/wireless/hostap/hostap_plx.c | 612 +++
drivers/net/wireless/hostap/hostap_proc.c | 466 ++
drivers/net/wireless/hostap/hostap_wlan.h | 1075 ++++++
drivers/net/wireless/orinoco.c | 11
drivers/net/wireless/wl3501.h | 4
include/linux/wireless.h | 283 +
include/net/ieee80211.h | 887 +++++
include/net/ieee80211_crypt.h | 86
net/Kconfig | 2
net/Makefile | 1
net/core/wireless.c | 74
net/ieee80211/Kconfig | 67
net/ieee80211/Makefile | 11
net/ieee80211/ieee80211_crypt.c | 259 +
net/ieee80211/ieee80211_crypt_ccmp.c | 470 ++
net/ieee80211/ieee80211_crypt_tkip.c | 708 ++++
net/ieee80211/ieee80211_crypt_wep.c | 272 +
net/ieee80211/ieee80211_module.c | 268 +
net/ieee80211/ieee80211_rx.c | 1206 +++++++
net/ieee80211/ieee80211_tx.c | 448 ++
net/ieee80211/ieee80211_wx.c | 471 ++
48 files changed, 27052 insertions(+), 121 deletions(-)

through these ChangeSets:

<jketreno:linux.intel.com>:
o ieee80211 subsystem

<mgalgoci:parcelfarce.linux.theplanet.co.uk>:
o wireless-2.6 cleanup

Adrian Bunk:
o net/ieee80211/Kconfig: don't describe what gets selected

Alexander Viro:
o hostap __user annotations

Fran?ois Romieu:
o ieee80211: offset_in_page() removal
o ieee80211: C99 initialization for eap_types
o ieee80211: failure of ieee80211_crypto_init()

Jean Tourrilhes:
o Wireless Extensions 18 (aka WPA)

Jeff Garzik:
o [wireless hostap] update for new pci_save_state()

Joshua Kwan:
o hostap: fix Kconfig typos and missing select CRYPTO

Jouni Malinen:
o hostap: Update to use the new WE18 proposal
o hostap: Filter disconnect events
o hostap: Improved suspend
o hostap: Rate limiting for debug messages
o hostap: Clear station statistic on accounting start
o hostap: Fix procfs rmdir after ifname change
o hostap: Add support for multi-func SanDisk ConnectPlus
o hostap: Disable interrupts before Genesis mode
o hostap: Include asm/io.h
o hostap: Clear station statistic on accounting start
o Host AP: Replaced MODULE_PARM with module_param*
o Host AP: Replaced direct dev->priv references with netdev_priv(dev)
o Host AP: Updated to use Linux wireless extensions v17
o Host AP: pci_register_driver() return value changes
o Host AP: Fix netif_carrier_off() in non-client modes
o Host AP: Fix PRISM2_IO_DEBUG
o Host AP: Use void __iomem * with {read,write}{b,w}
o Host AP: Fix card enabling after firmware download
o Host AP: Do not bridge packets to unauthorized ports
o Host AP: Fix compilation with PRISM2_NO_STATION_MODES defined
o Host AP: Prevent STAs from associating using AP address
o Host AP: Fix hw address changing for wifi# interface
o Host AP: Remove ioctl debug messages
o Host AP: Ignore (Re)AssocResp messages silently
o Host AP: Fix interface packet counters
o Host AP: Disable EAPOL TX/RX debug messages
o fix hostap crypto bugs
o Add HostAP wireless driver


Attachments:
changelog.txt (5.04 kB)

2005-03-23 05:52:54

by Jouni Malinen

[permalink] [raw]
Subject: Re: 2.6.x wireless update and status

On Tue, Mar 22, 2005 at 08:46:17PM -0500, Jeff Garzik wrote:

> Just updated the wireless-2.6 queue to include a HostAP update, and to
> add Wireless Extensions 18 (WPA). See attached for BK info, patch info,
> and changelog.

Thanks!

> Moving forward, the next "todo" for kernel wireless hackers is to get
> ieee80211 common code lib into shape, namely:
> * Merge Intel ipw drivers, which use ieee80211
> * Update HostAP to use ieee80211
> * Merge/convert other drivers to use ieee80211?

I'll be working on HostAP driver next; and ieee80211 code of course at
the same time, since it is likely to need some changes for this. As far
as other drivers are concerned, I'd like to see Atheros cards working
with the generic ieee80211 code. They would be a good test target since
they are an example of design where very large part of functionality is
in the driver/network stack (no firmware used). This would be a good
test to verify that the 802.11 code is generic enough for such a design.

> There is one minor point of contention so far. Jouni stated he prefers
> that HostAP go upstream before it gets updated to use ieee80211. I
> respectfully disagree, and prefer that HostAP is updated -first- to use
> the ieee80211 lib, before going upstream.

I think we can resolve this quite easily. The main reason for the other
order was in trying to save my time by not having to work in more than
one active development tree at the same time. This is kind of required
since designing and maintaining an IEEE 802.11 stack would really be a
full-time job and unfortunately, I have not yet managed to reach this
goal in a way that would allow me to use all my time on open source
development.

BK did not really work that well for me, but it looks like I'm having
better luck with quilt as far as the amount of time needed for
organizing changes to wireless-2.6 is concerned. In addition, the total
number of changes to the driver code has been quite small lately, so I'm
beginning to be more open to moving all future development into
wireless-2.6 tree and just keeping my current CVS repository as a
backwards compatible (2.2/2.4/2.6 kernels), stable version that wouldn't
get any more new features. This would allow the order that you prefer.
In addition, if someone really wants to get the new features I may be
adding, these should be available through wireless-2.6 tree (and -mm for
that matter).

--
Jouni Malinen PGP id EFC895FA

2005-03-24 00:20:34

by Michael Wu

[permalink] [raw]
Subject: Re: 2.6.x wireless update and status

On Wednesday 23 March 2005 12:52 am, Jouni Malinen wrote:
> > Moving forward, the next "todo" for kernel wireless hackers is to get
> > ieee80211 common code lib into shape, namely:
> > * Merge Intel ipw drivers, which use ieee80211
> > * Update HostAP to use ieee80211
> > * Merge/convert other drivers to use ieee80211?
>
> I'll be working on HostAP driver next; and ieee80211 code of course at
> the same time, since it is likely to need some changes for this. As far
> as other drivers are concerned, I'd like to see Atheros cards working
> with the generic ieee80211 code. They would be a good test target since
> they are an example of design where very large part of functionality is
> in the driver/network stack (no firmware used). This would be a good
> test to verify that the 802.11 code is generic enough for such a design.
>
The ADM8211 would be a good test too. So far, the main thing I see missing
from the 802.11 code is software scanning/authentication/association. Is all
of that being moved to userspace? I prefer having a little bit of that in the
kernel so minimal managed, adhoc, and WEP features can be used without a
daemon.

-Michael Wu


Attachments:
(No filename) (1.14 kB)
(No filename) (189.00 B)
Download all attachments