-J. Bruce Fields has nfsd fixes, James Bottomley
has a few high code motion patches for HPPA, Greg
KH has some patches for -rc2, namely USB and TTY,
and Guenter Roeck has hwmon fixes, also for -rc2.
-Chris Zankel has important changes for the Xtensa tree,
H. Peter Anvin has a pull request for misc x86 fixes,
and John W. Linville updates wireless:
This is a batch of fixes intended for the 3.7 stream.
Dan Carpenter brings a fix for a simple signedness bug that could
prevent the proper termination of a loop.
Felix Fietkau found a few more places that need to use
ieee80211_free_txskb for properly releasing SKBs used by mac80211.
Franky Lin offers a pair of brcmfmac fixes, both fixing simple state
Hante Meuleman corrects an error reporting case that wasn’t handling
all types of errors properly.
Johan Hedberg offers a fix for an issue discovered at the Bluetooth
UnPlug Fest. Gustavo says “the patch fixes a failure to pair with
devices that support the LE Secure Connections feature.”
Johannes Berg sends an iwlwifi fix to handle a message type that
is too large for the normal command mechanism. He also provides a
mac80211 fix to use HT20 channels when HT40 channels are not permitted.
Jouni Malinen offers a mac80211 fix for a masking error that was
incorrectly marking some frames.
Piotr Haber provides a fix to make sure bcma devices are unregistered
Stanislav Yakovlev gives us a fix for a panic in the ipw2200 driver.
Stanislaw Gruszka sends a pair of fixes: one prevents a mismatch on
connection states between cfg80211 and mac80211; the other prevents
some frame corruption related to handling encryption.
-Ingo Molnar has a perf-related pull request, Catalin Marinas
updates arm64, and Linus announces -rc2:
For the last few months, it’s almost become a habit of mine to make
the occasional release from the airport while flying out somewhere.
And this is another such one. Go free PDX airport wireless!
Anyway, it’s been roughly a week, and -rc2 is out. The most noticeable
thing tends to be fixing various fallout issues – there’s lots of
patches to finish up (and fix the fallout) from the UAPI include file
reorganization, for example, but there’s also some changes to how
module signing is done etc etc.
In pure lines, the uapi stuff (even with rename detection) dwarfs
everything else, and the diffstat of -rc2 is very unusual. Instead of
the normal “65% drivers, random everything else”, it’s “50% arch
patches (largely uapi header files), 30% include (more uapi header
file changes, and 20% everything else”.
But the appended shortlog is fairly readable, because the uapi changes
don’t dominate there. Nothing really particular stands out. DRM and
USB driver changes, and various random fixes all over..