For some reason this week actually felt very busy, but the rc5 numbers
show otherwise. It's all small and calm, and things are progressing
nicely.
I think the "it felt busy" was partly due to me stressing out over a
nasty VM bug that turned out to have a trivial two-liner fix. But
there were also a fair amount of email threads for future stuff, so
that probably also made me feel last week was busier than the actual
rc5 tree shows.
Anyway, of what little happened in rc5 (see appended shortlog for
details), it's just a fairly random collection of smallish fixes all
over. About a third drivers (nothing in particular stands out - rdma,
usb, ata, mmc, sound) with the rest being some tooling (mostly perf),
some arch updates, some filesystem stuff (mostly reiserfs), some arch
fixlets (mips, arm[64], x86) and some misc core kernel (tracing, VM
fixes, timers, yadda yadda).
And we had a few more interface cleanups for rseq.
Go out and test. The VM thing that annoyed me so much this week isn't
actually triggerable for most people (and only happened in rc3 and
rc4), but if you hit a "kernel BUG in mm/page_alloc.c" during earlier
rc testing, it's fixed (although for most people, if it hit you, it
probably just showed as unexplained hung machine under moderate memory
pressure).
Linus
