Added goroutine; got simpler.

Fixes deadlock when doing Read+Close
or Write+Close on same end.

R=r, cw
CC=golang-dev
https://golang.org/cl/994043

R=rsc
CC=golang-dev
https://golang.org/cl/949043

Fixes #734.

R=rsc, r
CC=golang-dev
https://golang.org/cl/975042

R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/988042

R=rsc, rsc1
CC=golang-dev
https://golang.org/cl/878046

R=rsc
CC=golang-dev
https://golang.org/cl/897042

R=rsc
CC=golang-dev
https://golang.org/cl/1014041

R=rsc
CC=golang-dev
https://golang.org/cl/968043

R=r, rsc
CC=golang-dev
https://golang.org/cl/978042

R=rsc
CC=golang-dev
https://golang.org/cl/937042

R=ken2, r, ken3
CC=golang-dev
https://golang.org/cl/1006041

R=adg, gri
CC=golang-dev, r
https://golang.org/cl/1008042

Depends on CL 957045, 980043, 1004043.
Fixes #715.

R=agl1, agl
CC=golang-dev
https://golang.org/cl/943043

Adds two more methods, Timeout and Temporary.
Implemented by os.Errno too.  The intent is to make
the checks for os.EAGAIN a little less clunky.
It should also let us clean up a bug that Mike Solomon
pointed out: if a network server gets an "out of file descriptors"
error from Accept, the listener should not stop.
It will be able to check this because that error would
have Temporary() == true.

Also clean up some underscore names.

Fixes #442.

R=r
CC=golang-dev, msolo
https://golang.org/cl/957045

R=rsc, r
CC=golang-dev
https://golang.org/cl/1010041

R=r
CC=golang-dev
https://golang.org/cl/1004043

R=r
CC=golang-dev
https://golang.org/cl/980043

R=r
CC=golang-dev
https://golang.org/cl/1009041

R=r
CC=golang-dev
https://golang.org/cl/982043

More in line with other regexp packages
and egrep; accommodates overzealous escapers.

R=r
CC=golang-dev
https://golang.org/cl/1008041

R=ken2
CC=golang-dev
https://golang.org/cl/1007041

Enable package tests for Native Client build.

R=r
CC=golang-dev
https://golang.org/cl/957042

R=rsc
CC=golang-dev
https://golang.org/cl/970042

R=rsc
CC=golang-dev
https://golang.org/cl/962043

Changed most of the functions in nat.go to methods on nat.

R=gri
CC=golang-dev
https://golang.org/cl/976041

Uid, Gid become int.
File size info becomes int64.
Times become int64.

R=rsc, cw
CC=golang-dev
https://golang.org/cl/968042

Everything was linking to "Run_time_panics", but the id was "Run_time_errors".

R=golang-dev, r
CC=golang-dev
https://golang.org/cl/961042

R=golang-dev, adg
CC=golang-dev
https://golang.org/cl/917042

R=rsc, gri
CC=golang-dev
https://golang.org/cl/967041

R=r
CC=golang-dev
https://golang.org/cl/953041

Fixes #733.

R=r
CC=golang-dev
https://golang.org/cl/958041

R=r
CC=golang-dev
https://golang.org/cl/971041

R=r
CC=golang-dev
https://golang.org/cl/970041

R=r
CC=golang-dev
https://golang.org/cl/961041

run reverse-complement only once. (garbage is no longer an issue)

R=rsc
CC=golang-dev
https://golang.org/cl/968041

R=gri, eds
CC=golang-dev
https://golang.org/cl/947042

R=rsc
CC=golang-dev
https://golang.org/cl/940041

This gives about a 6% performance improvement to pidigits.
Thanks to Russ for the suggestion.

R=rsc, gri
CC=golang-dev
https://golang.org/cl/957041

There was a bug in makeN that caused lots of unnecessary
allocations.  Fixing this exposed a few bugs in other
functions which worked when makeN allocated a new slice, but
failed when it simply resized an existing slice.  The result
is a pretty big performance improvement.  When running
pidigits, here were the numbers I got on amd64:

Before this change:
pidigits 10000
        gcc -O2 pidigits.c -lgmp        2.09u 0.02s 2.11r
        gc pidigits     12.68u 0.04s 12.72r
        gc_B pidigits   12.71u 0.03s 12.75r

After:
pidigits 10000
        gcc -O2 pidigits.c -lgmp        2.09u 0.00s 2.10r
        gc pidigits     6.82u 0.00s 6.85r
        gc_B pidigits   6.55u 0.01s 6.59r

R=rsc, gri
CC=golang-dev
https://golang.org/cl/953042

R=r, cw
CC=golang-dev
https://golang.org/cl/895043