[git patches] new wireless stuffs

If this is post-2.6.13 material, that's fine.

I've been getting tired of the slow movement of wireless.  A little bit 
of that is my fault, certainly.  I've also been wanting to get these net 
drivers out to Linux users.

This adds HostAP, ipw2100, ipw2200, and the generic ieee80211 code.  The 
only -changes- in this set are cosmetic.

Further work is pending from SuSE and Intel [*poke* *poke* to them], but 
these should be working, and have been in -mm for quite a while.

One big thing I'm still hoping for is that someone will merge HostAP 
(where ieee80211 code came from) with the ieee80211 generic code.  The 
HostAP maintainer has been unwilling to do it, even though he has been 
good about keeping HostAP updated, so hopefully a volunteer will step up.

Please pull from branch 'ieee80211-wifi' of
rsync://rsync.kernel.org/pub/scm/linux/kernel/git/jgarzik/netdev-2.6.git

to obtain the changes described by the attached diffstat/changelog.  The 
file additions were too large to attach, so I only attached the changes 
to existing files.  Existing drivers merely had some struct renames (for 
new ieee80211 header), and some 'add "static" attribute' changes.

	Jeff

On Sat, Jul 30, 2005 at 07:24:46PM -0400, Jeff Garzik wrote:

One big thing I'm still hoping for is that someone will merge HostAP 
(where ieee80211 code came from) with the ieee80211 generic code.  The 
HostAP maintainer has been unwilling to do it, even though he has been 
good about keeping HostAP updated, so hopefully a volunteer will step up.

I would say this has been more due to lack of time than unwillingness on
my part. Finally, I got git set up yesterday and got back to working
with the wireless-2.6/netdev-2.6 code and I'm planning on working with
the Host AP and ieee80211 code merge. I would expect the client side
functionality to be relatively easy merge, but the AP side functionality
may require considerable amount of work due to the current ieee80211
code being more focused on the client side.

-- 
Jouni Malinen                                            PGP id EFC895FA

On Sat, Jul 30, 2005 at 07:24:46PM -0400, Jeff Garzik wrote:

If this is post-2.6.13 material, that's fine.

I've been getting tired of the slow movement of wireless.  A little bit 
of that is my fault, certainly.  I've also been wanting to get these net 
drivers out to Linux users.

This adds HostAP, ipw2100, ipw2200, and the generic ieee80211 code.  The 
only -changes- in this set are cosmetic.

Any chance we can give the hostap driver a better name?  hostap describes
the AP mode it can operate on, but it should really be named after the hardware
it supports in some way.

Hi!

If this is post-2.6.13 material, that's fine.

I've been getting tired of the slow movement of wireless.  A little bit 
of that is my fault, certainly.  I've also been wanting to get these net 
drivers out to Linux users.

This adds HostAP, ipw2100, ipw2200, and the generic ieee80211 code.  The 
only -changes- in this set are cosmetic.

Further work is pending from SuSE and Intel [*poke* *poke* to them], but 
these should be working, and have been in -mm for quite a while.

One big thing I'm still hoping for is that someone will merge HostAP 
(where ieee80211 code came from) with the ieee80211 generic code.  The 
HostAP maintainer has been unwilling to do it, even though he has been 
good about keeping HostAP updated, so hopefully a volunteer will step up.

Please pull from branch 'ieee80211-wifi' of
rsync://rsync.kernel.org/pub/scm/linux/kernel/git/jgarzik/netdev-2.6.git

to obtain the changes described by the attached diffstat/changelog.  The 
file additions were too large to attach, so I only attached the changes 
to existing files.  Existing drivers merely had some struct renames (for 
new ieee80211 header), and some 'add "static" attribute' changes.

Wow, thanks!
								Pavel

-- 
if you have sharp zaurus hardware you don't need... you know my address

Christoph Hellwig wrote:

Any chance we can give the hostap driver a better name?  hostap describes
the AP mode it can operate on, but it should really be named after the hardware
it supports in some way.

If the driver gets rewritten, sure.  Otherwise, users are already 
familiar with the name.

	Jeff