Thanks to Dustin Sallings for exposing the most frustrating
bug ever, and for providing repro cases (which formed the
basis of the new tests in this CL), and to Dave Cheney and
Dmitry Vyukov for help debugging and fixing.

This CL depends on submited pollster CLs ffd1e075c260 (Unix)
and 14b544194509 (Windows), as well as unsubmitted 6852085.
Some operating systems (OpenBSD, NetBSD, ?) may still require
more pollster work, fixing races (Issue 4434 and
http://goo.gl/JXB6W).

Tested on linux-amd64 and darwin-amd64, both with GOMAXPROCS 1
and 4 (all combinations of which previously failed differently)

Fixes #4191
Update #4434 (related fallout from this bug)

R=dave, bradfitz, dsallings, rsc, fullung
CC=golang-dev
https://golang.org/cl/6851061

Bools from comparisons can be assigned to all bool types, but this idealness would propagate through logical operators when the result should have been lowered to a non-ideal form.

Fixes #3924.

R=golang-dev, remyoudompheng, r, rsc, mtj
CC=golang-dev
https://golang.org/cl/6855061

Cleanups introduced originally by now abandoned
https://golang.org/cl/6846078/ .

Includes additional test cases for 'if' and 'for'
statements.

No formatting changes.

R=rsc
CC=golang-dev
https://golang.org/cl/6855096

Also: Can set base indentation in printer.Config: all code
is going to be indented by at least that amount (except for
raw string literals spanning multiple lines, since their
values must not change).

R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/6847086

R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/6849105

The stack overflow checker in the linker uses the spadj field
to determine whether stack space will be large enough or not.
When spadj=0, the checker treats the function as a nosplit
and emits an error although the program is correct.

Also enable the stack checker in 8l.

Fixes #4316.

R=rsc, golang-dev
CC=golang-dev
https://golang.org/cl/6855088

also:

- composite literal checking close to complete
- cleaned up parameter, method, field checking
- don't let panics escape type checker
- more TODOs eliminated

R=rsc
CC=golang-dev
https://golang.org/cl/6816083

Since p.M is rejected (must call methods), use p.M(),
to keep the examples compiling.

Fixes #4441.

R=gri
CC=golang-dev
https://golang.org/cl/6854096

The 8l linker automatically inserts XCHG instructions
to support otherwise impossible byte registers
(only available on AX, BX, CX, DX).

Sometimes AX or DX is needed (for MUL and DIV) so
we need to avoid clobbering them.

R=golang-dev, dave, iant, iant, rsc
CC=golang-dev
https://golang.org/cl/6846057

Update #4415.

R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/6856080

This requires rebasing the block-mode test scripts.
I used GnuTLS version 3.1.4.

R=agl
CC=golang-dev
https://golang.org/cl/6844073

Currently the test exposes data races on deadline vars.

R=golang-dev, dave, bradfitz, rsc
CC=golang-dev
https://golang.org/cl/6845091

R=golang-dev
CC=golang-dev
https://golang.org/cl/6856097

Fixes #4424.
Fixes #4425.

R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/6849093

Fixes #4429.

R=golang-dev, remyoudompheng, daniel.morsing, rsc
CC=golang-dev
https://golang.org/cl/6850097

Fixes #4402.

R=remyoudompheng, shivakumar.gn, random0x00, rsc
CC=golang-dev
https://golang.org/cl/6842063

Failed with gccgo.

R=golang-dev, remyoudompheng, rsc
CC=golang-dev
https://golang.org/cl/6855053

add necessary newlines.

R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/6847067

This CL starts to introduce IPv6 scoped addressing capability
into the net package.

The Public API changes are:
+pkg net, type IPAddr struct, Zone string
+pkg net, type IPNet struct, Zone string
+pkg net, type TCPAddr struct, Zone string
+pkg net, type UDPAddr struct, Zone string

Update #4234.

R=rsc, bradfitz, iant
CC=golang-dev
https://golang.org/cl/6849045

R=minux.ma, jsing
CC=golang-dev
https://golang.org/cl/6855080

while we're at it, also fix a return value stutter in ParseForm.

R=bradfitz
CC=golang-dev
https://golang.org/cl/6847074

R=golang-dev, bradfitz, minux.ma, dave
CC=golang-dev
https://golang.org/cl/6842081

R=golang-dev, minux.ma
CC=golang-dev
https://golang.org/cl/6854079

Check the return value from malloc - do not assume that we were
allocated memory just because we asked for it.

Update #4415.

R=minux.ma, daniel.morsing, remyoudompheng, rsc
CC=golang-dev
https://golang.org/cl/6782100

If the a network read would block, and a packet arrived just before the timeout expired, then the number of bytes from the previous (blocking) read, -1, would be returned.

This change restores the previous logic, where n would be unconditionally set to 0 if err != nil, but was skipped due to a change in CL 6851096.

The test for this change is CL 6851061.

R=bradfitz, mikioh.mikioh, dvyukov, rsc
CC=golang-dev
https://golang.org/cl/6852085

Should make BSDs more reliable. (they seem to reuse ports
quicker than Linux)

Tested by hand with local modifications to force reuse on
Linux. (net/http tests failed before, pass now) Details in the
issue.

Fixes #4436

R=golang-dev, minux.ma
CC=golang-dev
https://golang.org/cl/6847101

Fixes #4421

R=golang-dev, dave, minux.ma, mchaten, rsc
CC=golang-dev
https://golang.org/cl/6855083

Make use of carry clear condition instead of low or same.

R=minux.ma, jsing, dave
CC=golang-dev
https://golang.org/cl/6844080

The tests verify that deadlines are "persistent",
read/write deadlines do not interfere, can be reset,
read deadline can be set with both SetDeadline()
and SetReadDeadline(), etc.

R=golang-dev, bradfitz, dave
CC=golang-dev
https://golang.org/cl/6850070

R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/6856084

The fix for issue 4403 may include more calls to time.Now().UnixNano(). I was concerned that if this function allocated it would cause additional garbage on the heap. It turns out that it doesn't, which is a nice surprise.

Also add benchmark for Now().UnixNano()

R=bradfitz, minux.ma
CC=golang-dev
https://golang.org/cl/6849097

R=golang-dev, mikioh.mikioh, minux.ma, bradfitz
CC=golang-dev
https://golang.org/cl/6851098

R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/6856078

Fixes #4432.

R=golang-dev, minux.ma
CC=golang-dev
https://golang.org/cl/6858064

Otherwise a fast sender or receiver can make sockets always
readable or writable, preventing deadline checks from ever
occuring.

Update #4191 (fixes it with other CL, coming separately)
Fixes #4403

R=golang-dev, alex.brainman, dave, mikioh.mikioh
CC=golang-dev
https://golang.org/cl/6851096

madvise was missing so implement it in assembler. This change
needs to be extended to the other BSD variantes (Net and Open)

Without this change the scavenger will attempt to pass memory back
to the operating system when it has become idle, but the memory is
not returned and for long running Go processes the total memory used
can grow until OOM occurs.

I have only been able to test the code on FreeBSD AMD64. The ARM
platforms needs testing.

R=golang-dev, mikioh.mikioh, dave, jgc, minux.ma
CC=golang-dev
https://golang.org/cl/6850081

R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/6845077

R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/6846095

Fixes #4426.

R=dvyukov
CC=golang-dev
https://golang.org/cl/6851099

R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/6851100