mbox series

pull-request: wireless-2023-10-24

Message ID 20231024103540.19198-2-johannes@sipsolutions.net
State New
Headers show
Series pull-request: wireless-2023-10-24 | expand

Pull-request

git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless.git tags/wireless-2023-10-24

Message

Johannes Berg Oct. 24, 2023, 10:35 a.m. UTC
Hi,

We have a couple of last-minute fixes for some issues.

Note that this introduces a merge conflict with -next,
which Stephen reported and (correctly) resolved here:
https://lore.kernel.org/linux-wireless/20231024112424.7de86457@canb.auug.org.au/
Basically just context - use the ieee80211_is_protected_dual_of_public_action()
check from this pull request, and the return code
RX_DROP_U_UNPROT_UNICAST_PUB_ACTION from -next.

Please pull and let us know if there's any problem.

Thanks,
johannes



The following changes since commit f2ac54ebf85615a6d78f5eb213a8bbeeb17ebe5d:

  net: rfkill: reduce data->mtx scope in rfkill_fop_open (2023-10-11 16:55:10 +0200)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless.git tags/wireless-2023-10-24

for you to fetch changes up to 91535613b6090fc968c601d11d4e2f16b333713c:

  wifi: mac80211: don't drop all unprotected public action frames (2023-10-23 13:25:30 +0200)

----------------------------------------------------------------
Three more fixes:
 - don't drop all unprotected public action frames since
   some don't have a protected dual
 - fix pointer confusion in scanning code
 - fix warning in some connections with multiple links

----------------------------------------------------------------
Avraham Stern (1):
      wifi: mac80211: don't drop all unprotected public action frames

Ben Greear (1):
      wifi: cfg80211: pass correct pointer to rdev_inform_bss()

Johannes Berg (1):
      wifi: cfg80211: fix assoc response warning on failed links

 include/linux/ieee80211.h | 29 +++++++++++++++++++++++++++++
 net/mac80211/rx.c         |  3 +--
 net/wireless/mlme.c       |  3 ++-
 net/wireless/scan.c       |  2 +-
 4 files changed, 33 insertions(+), 4 deletions(-)

Comments

Johannes Berg Oct. 24, 2023, 8:25 p.m. UTC | #1
Thanks Jakub.

> > Note that this introduces a merge conflict with -next,
> > which Stephen reported and (correctly) resolved here:
> > https://lore.kernel.org/linux-wireless/20231024112424.7de86457@canb.auug.org.au/
> > Basically just context - use the ieee80211_is_protected_dual_of_public_action()
> > check from this pull request, and the return code
> > RX_DROP_U_UNPROT_UNICAST_PUB_ACTION from -next.

Are you planning to merge net into net-next really soon for some reason?

If not, I can resolve this conflict and we'll include it in the next
(and last) wireless-next pull request, which will be going out Thursday
morning (Europe time.)

johannes
Johannes Berg Oct. 24, 2023, 8:54 p.m. UTC | #2
On Tue, 2023-10-24 at 13:52 -0700, Jakub Kicinski wrote:
> On Tue, 24 Oct 2023 22:25:18 +0200 Johannes Berg wrote:
> > Are you planning to merge net into net-next really soon for some reason?
> 
> Submitting on Wed did cross my mind, but there's no solid plan.
> Unless that changes, Paolo will submit net on Thursday, EU time.
> And we'll cross-merge once Linux pulls. 

OK, sounds good.

> > If not, I can resolve this conflict and we'll include it in the next
> > (and last) wireless-next pull request, which will be going out Thursday
> > morning (Europe time.)
> 
> Sounds good! But do you need to do the resolution to put something 
> on top? Otherwise we can deal with the conflict when pulling.

No, not really, nothing left that's not in wireless-next already (I
think), except maybe some tiny cleanups.

Just trying to make it easier for you, even if it's really not a complex
conflict :)

johannes
Johannes Berg Oct. 24, 2023, 9:03 p.m. UTC | #3
On Tue, 2023-10-24 at 14:01 -0700, Jakub Kicinski wrote:
> On Tue, 24 Oct 2023 22:54:50 +0200 Johannes Berg wrote:
> > > > If not, I can resolve this conflict and we'll include it in the next
> > > > (and last) wireless-next pull request, which will be going out Thursday
> > > > morning (Europe time.)  
> > > 
> > > Sounds good! But do you need to do the resolution to put something 
> > > on top? Otherwise we can deal with the conflict when pulling.  
> > 
> > No, not really, nothing left that's not in wireless-next already (I
> > think), except maybe some tiny cleanups.
> > 
> > Just trying to make it easier for you, even if it's really not a complex
> > conflict :)
> 
> I think "Linus rules" would dictate that cross-merges to hide conflicts
> are a bad thing. We don't have much to win so let's stick to that :)

Fair enough :)

> Hopefully I can deal with the resolution, but if you want to be 100%
> sure - you can drop a git-rerere resolution somewhere I can fetch it.

No need I think, just the return codes changed in the -next version for
better skb drop reasons :)

johannes