A few times now I've git-fetch'd origin and noticed at76_usb.[ch]
missing on origin/master. Strange, wondering if you're aware of any
such type of issues or its just me.
Luis
John W. Linville <[email protected]> writes:
> On Wed, Sep 03, 2008 at 05:34:03PM +0300, Kalle Valo wrote:
>> John W. Linville <[email protected]> writes:
>>
>> > Anyone want to talk me out of pulling-in the at76 parts?
>>
>> Sorry, I don't quite understand what pulling-in means here.
>
> I just mean making sure it is (and continues to be) in
> wireless-testing's master branch.
Ok, thanks for the explanation.
>> But personally I don't mind, choose which one is best in your opinion.
>> If you drop it, I can always resubmit it when it's properly working.
>
> Leaving it in has encouraged people to keep it up-to-date (or at least
> compiling) across mac80211 driver API changes. Whether or not this
> has been worthwhile is, of course, debatable.
>From my perspective it has been worthwhile, but at the same time I
realise how much work it means for you. And as I know how busy you are
I really would not want to cause you any extra work.
> P.S. For the record, I _would_ like to see a working at76 driver
> make it upstream at some point.
Me too. Fortunately starting from September I will have some free time
(yay!) and I can start working with at76 again.
--
Kalle Valo
On Tue, Sep 02, 2008 at 08:03:32PM -0700, Luis R. Rodriguez wrote:
> On Tue, Sep 2, 2008 at 7:59 PM, Kalle Valo <[email protected]> wrote:
> > Luis R. Rodriguez <[email protected]> writes:
> >
> >> A few times now I've git-fetch'd origin and noticed at76_usb.[ch]
> >> missing on origin/master. Strange, wondering if you're aware of any
> >> such type of issues or its just me.
> >
> > Good point, I don't see it either. But at76_usb is quite broken
> > currently, so maybe John decided just to drop it? Good choice, at
> > least for now.
> >
> > And it's not just that, also -wl suffix is missing from the version.
> >
> > My HEAD is currently:
> >
> > commit 18c8adeb0244f823ba78a51e23f813fe68bc3c54
> > Author: Michael Buesch <[email protected]>
> > Date: Thu Aug 28 19:33:40 2008 +0200
> >
> > b43: Rewrite TX power adjustment
>
> I'm at:
>
> commit 908dd0b965e64fcc2eb64152bac65f2876c34e0b
> Author: Senthil Balasubramanian <[email protected]>
> Date: Mon Sep 1 19:58:20 2008 +0530
> ath9: Fix ath_rx_flush_tid() for IRQs disabled kernel warning message.
>
> Reason I find it strange is on one box it went through and I still see
> history for it, and in another I don't. I was going to do a fresh pull
> to test but figure I rather ask as bandwidth is crummy at work.
Crap, I botched my lastest update. I'll pull in the part that adds
-wl to the version for sure. Anyone want to talk me out of pulling-in
the at76 parts?
John
--
John W. Linville
[email protected]
On Wed, Sep 03, 2008 at 09:38:03AM -0400, John W. Linville wrote:
> On Wed, Sep 03, 2008 at 04:24:48PM +0300, Jouni Malinen wrote:
> > I don't see -wl in EXTRAVERSION in kernel.org version (1) or my own
> > clone. Both are at commit 908dd0b965e64fcc2eb64152bac65f2876c34e0b. I
> > noticed this disappearing at one of the updates since I had to add a
> > local patch to be able to update my modules without rebooting..
> The -wl comes from localversion-wireless in the kernel's root
> directory. Has something changed recently to break that mechanism?
Interesting.. I tried it now and it does indeed work, i.e., I get -wl
from localversion-wireless. I have no idea what happened earlier, but
'make modules; sudo make modules_install' ended up installing modules
into 2.6.27-rc4, not 2.6.27-rc4-wl after one of the git tree updates. I
did not look at details at that point and just added -wl to Makefile
EXTRAVERSION. Anyway, it is not needed anymore.
--
Jouni Malinen PGP id EFC895FA
On Wed, Sep 03, 2008 at 05:34:03PM +0300, Kalle Valo wrote:
> John W. Linville <[email protected]> writes:
>
> > Anyone want to talk me out of pulling-in the at76 parts?
>
> Sorry, I don't quite understand what pulling-in means here.
I just mean making sure it is (and continues to be) in
wireless-testing's master branch.
> But personally I don't mind, choose which one is best in your opinion.
> If you drop it, I can always resubmit it when it's properly working.
Leaving it in has encouraged people to keep it up-to-date (or at least
compiling) across mac80211 driver API changes. Whether or not this
has been worthwhile is, of course, debatable.
John
P.S. For the record, I _would_ like to see a working at76 driver
make it upstream at some point.
--
John W. Linville
[email protected]
On Wed, Sep 03, 2008 at 04:24:48PM +0300, Jouni Malinen wrote:
> On Wed, Sep 03, 2008 at 09:04:40AM -0400, John W. Linville wrote:
> > On Wed, Sep 03, 2008 at 08:53:52AM -0400, John W. Linville wrote:
> > > Crap, I botched my lastest update. I'll pull in the part that adds
> > > -wl to the version for sure. Anyone want to talk me out of pulling-in
> > > the at76 parts?
> >
> > Hmmm...upon further review, both my local copy and the one at
> > kernel.org seem to have the proper bits AFAICT...?
>
> I don't see -wl in EXTRAVERSION in kernel.org version (1) or my own
> clone. Both are at commit 908dd0b965e64fcc2eb64152bac65f2876c34e0b. I
> noticed this disappearing at one of the updates since I had to add a
> local patch to be able to update my modules without rebooting..
>
> (1)
> http://git.kernel.org/?p=linux/kernel/git/linville/wireless-testing.git;a=blob_plain;f=Makefile;hb=908dd0b965e64fcc2eb64152bac65f2876c34e0b
The -wl comes from localversion-wireless in the kernel's root
directory. Has something changed recently to break that mechanism?
John
--
John W. Linville
[email protected]
On Wed, Sep 03, 2008 at 08:53:52AM -0400, John W. Linville wrote:
> On Tue, Sep 02, 2008 at 08:03:32PM -0700, Luis R. Rodriguez wrote:
> > On Tue, Sep 2, 2008 at 7:59 PM, Kalle Valo <[email protected]> wrote:
> > > Luis R. Rodriguez <[email protected]> writes:
> > >
> > >> A few times now I've git-fetch'd origin and noticed at76_usb.[ch]
> > >> missing on origin/master. Strange, wondering if you're aware of any
> > >> such type of issues or its just me.
> > >
> > > Good point, I don't see it either. But at76_usb is quite broken
> > > currently, so maybe John decided just to drop it? Good choice, at
> > > least for now.
> > >
> > > And it's not just that, also -wl suffix is missing from the version.
> > >
> > > My HEAD is currently:
> > >
> > > commit 18c8adeb0244f823ba78a51e23f813fe68bc3c54
> > > Author: Michael Buesch <[email protected]>
> > > Date: Thu Aug 28 19:33:40 2008 +0200
> > >
> > > b43: Rewrite TX power adjustment
> >
> > I'm at:
> >
> > commit 908dd0b965e64fcc2eb64152bac65f2876c34e0b
> > Author: Senthil Balasubramanian <[email protected]>
> > Date: Mon Sep 1 19:58:20 2008 +0530
> > ath9: Fix ath_rx_flush_tid() for IRQs disabled kernel warning message.
> >
> > Reason I find it strange is on one box it went through and I still see
> > history for it, and in another I don't. I was going to do a fresh pull
> > to test but figure I rather ask as bandwidth is crummy at work.
>
> Crap, I botched my lastest update. I'll pull in the part that adds
> -wl to the version for sure. Anyone want to talk me out of pulling-in
> the at76 parts?
Hmmm...upon further review, both my local copy and the one at
kernel.org seem to have the proper bits AFAICT...?
--
John W. Linville
[email protected]
On Tue, Sep 2, 2008 at 7:59 PM, Kalle Valo <[email protected]> wrote:
> Luis R. Rodriguez <[email protected]> writes:
>
>> A few times now I've git-fetch'd origin and noticed at76_usb.[ch]
>> missing on origin/master. Strange, wondering if you're aware of any
>> such type of issues or its just me.
>
> Good point, I don't see it either. But at76_usb is quite broken
> currently, so maybe John decided just to drop it? Good choice, at
> least for now.
>
> And it's not just that, also -wl suffix is missing from the version.
>
> My HEAD is currently:
>
> commit 18c8adeb0244f823ba78a51e23f813fe68bc3c54
> Author: Michael Buesch <[email protected]>
> Date: Thu Aug 28 19:33:40 2008 +0200
>
> b43: Rewrite TX power adjustment
I'm at:
commit 908dd0b965e64fcc2eb64152bac65f2876c34e0b
Author: Senthil Balasubramanian <[email protected]>
Date: Mon Sep 1 19:58:20 2008 +0530
ath9: Fix ath_rx_flush_tid() for IRQs disabled kernel warning message.
Reason I find it strange is on one box it went through and I still see
history for it, and in another I don't. I was going to do a fresh pull
to test but figure I rather ask as bandwidth is crummy at work.
Luis
On Wed, Sep 03, 2008 at 09:04:40AM -0400, John W. Linville wrote:
> On Wed, Sep 03, 2008 at 08:53:52AM -0400, John W. Linville wrote:
> > Crap, I botched my lastest update. I'll pull in the part that adds
> > -wl to the version for sure. Anyone want to talk me out of pulling-in
> > the at76 parts?
>
> Hmmm...upon further review, both my local copy and the one at
> kernel.org seem to have the proper bits AFAICT...?
I don't see -wl in EXTRAVERSION in kernel.org version (1) or my own
clone. Both are at commit 908dd0b965e64fcc2eb64152bac65f2876c34e0b. I
noticed this disappearing at one of the updates since I had to add a
local patch to be able to update my modules without rebooting..
(1)
http://git.kernel.org/?p=linux/kernel/git/linville/wireless-testing.git;a=blob_plain;f=Makefile;hb=908dd0b965e64fcc2eb64152bac65f2876c34e0b
However, at76_usb.[ch] seems to be present in both..
--
Jouni Malinen PGP id EFC895FA
Luis R. Rodriguez <[email protected]> writes:
> A few times now I've git-fetch'd origin and noticed at76_usb.[ch]
> missing on origin/master. Strange, wondering if you're aware of any
> such type of issues or its just me.
Good point, I don't see it either. But at76_usb is quite broken
currently, so maybe John decided just to drop it? Good choice, at
least for now.
And it's not just that, also -wl suffix is missing from the version.
My HEAD is currently:
commit 18c8adeb0244f823ba78a51e23f813fe68bc3c54
Author: Michael Buesch <[email protected]>
Date: Thu Aug 28 19:33:40 2008 +0200
b43: Rewrite TX power adjustment
--
Kalle Valo
John W. Linville <[email protected]> writes:
> Anyone want to talk me out of pulling-in the at76 parts?
Sorry, I don't quite understand what pulling-in means here.
But personally I don't mind, choose which one is best in your opinion.
If you drop it, I can always resubmit it when it's properly working.
--
Kalle Valo