R=r
CC=golang-dev
https://golang.org/cl/4272058

Permit system calls to be designated as non-blocking, meaning
that we simply call them without involving the scheduler.

This change by itself is mostly performance neutral.  In
combination with a following change to the net package there
is a performance advantage.

R=rsc, dfc, r2, iant2, rsc1
CC=golang-dev
https://golang.org/cl/4278055

R=ken2
CC=golang-dev
https://golang.org/cl/4278056

- use enc.err and dec.err instead of return values in deferred error catcher
- replace io.WriteString with buffer.WriteString

now at:
mallocs per encode of type Bench: 7
mallocs per decode of type Bench: 8

R=rsc
CC=golang-dev
https://golang.org/cl/4277057

R=brainman, rsc, rsc1
CC=golang-dev
https://golang.org/cl/4249064

R=rsc
CC=golang-dev
https://golang.org/cl/4271048

gobuilder: recognize CLs of the form weekly.DATE

R=rsc, r
CC=golang-dev
https://golang.org/cl/4282052

This just returns a ClientConn suitable for writing
proxy requests.  To be used in Transport.

R=rsc, petar-m
CC=golang-dev
https://golang.org/cl/4290052

(per rsc's suggestion)

R=rsc
CC=golang-dev
https://golang.org/cl/4276057

R=gri
CC=golang-dev
https://golang.org/cl/4280048

The -test.run and -test.bench flags were compilng the regexp for ever test
function, which was mucking up memory profiles.   Add a simple wrapper
to save the compiled state so that the regexp is compiled only once for
each flag.

R=rsc
CC=golang-dev
https://golang.org/cl/4274063

Dependency on bufio crept in during last CL; this breaks the cycle.
Also add a missing '-' to the documentation.

R=rsc
CC=golang-dev
https://golang.org/cl/4274061

These allow a test to generate memory profiles automatically.

R=golang-dev, rsc1
CC=golang-dev
https://golang.org/cl/4273064

R=bradfitzgo, nsz
CC=golang-dev
https://golang.org/cl/4280041

There is some disagreement about how to deal with hash values larger
than the curve order size. We choose to follow OpenSSL's lead here.

R=bradfitzgo, r
CC=golang-dev
https://golang.org/cl/4273059

R=r
CC=golang-dev
https://golang.org/cl/4281052

R=r
CC=golang-dev
https://golang.org/cl/4289048

R=r
CC=golang-dev
https://golang.org/cl/4272052

Avoids 3 mallocs in a round trip encoding/decoding a struct.

R=rsc, rsc1
CC=golang-dev
https://golang.org/cl/4278052

R=golang-dev, adg
CC=golang-dev
https://golang.org/cl/4290049

R=adg
CC=golang-dev
https://golang.org/cl/4275054

R=adg
CC=golang-dev
https://golang.org/cl/4274059

If braces don't have position information for a composite
literal, don't assume alignment of key:value pairs under
the (wrong) assumption that there may be multiple lines.

R=rsc
CC=golang-dev
https://golang.org/cl/4297043

Add malloc counts for encode and decode.

R=rsc
CC=golang-dev
https://golang.org/cl/4290048

R=r
CC=golang-dev
https://golang.org/cl/4273060

R=rsc, dsymonds
CC=golang-dev
https://golang.org/cl/4290047

On my mac:
mallocs per rpc round trip: 144
rpc.BenchmarkEndToEnd	   10000	    228244 ns/op

Room for improvement.

R=rsc
CC=golang-dev
https://golang.org/cl/4274058

This makes it possible to build Windows binaries on non-Windows OSes.

R=golang-dev, rsc1, rsc
CC=golang-dev
https://golang.org/cl/4271046

The Windows code is untested.

R=rsc, gri, brainman, rsc1
CC=golang-dev
https://golang.org/cl/4253052

R=bradfitzgo, dsymonds, r, gri, adg
CC=golang-dev
https://golang.org/cl/4282044

R=ken2
CC=golang-dev
https://golang.org/cl/4287047

R=golang-dev, gri
CC=golang-dev
https://golang.org/cl/4274052

Swapping the goroutines lets them reuse the
communication completion on v instead of
needing a second channel (done).

R=gri
CC=golang-dev
https://golang.org/cl/4287045

Also change labelled examples to use gofmt formatting.

R=gri, r, jnml
CC=golang-dev
https://golang.org/cl/4287046

This matches the real ResponseWriter behavior.

R=rsc
CC=golang-dev
https://golang.org/cl/4291048

R=agl, agl1
CC=golang-dev
https://golang.org/cl/4297041

This reduces the number of writes by 2 (1 client, 1 server) on each round trip.
A simple test shows 24% higher throughput.

R=rsc
CC=golang-dev
https://golang.org/cl/4279057

R=golang-dev, r, dsymonds, rsc1
CC=golang-dev
https://golang.org/cl/4277046

R=ken2
CC=golang-dev
https://golang.org/cl/4279053

This change records more metadata about what
influenced the creation of the object file.
Specifically, if a package imports, say, "fmt" but does not
need to describe any fmt types in its own export data,
that package's object file did not mention the dependency
on "fmt" before.  Now it does.

Listing the import is purely informational.
It has no effect on which files are opened or consulted
when importing a package.

Import lines are marked indirect when they are needed
to explain the API but were not imported directly.
For example http imports crypto/tls and exports
a struct with a field of type tls.ConnectionState,
which contains an x509.Certificate.  Since http does
not import x509 but needs to explain the x509.Certificate
type in its export data, the import of x509 is marked
as indirect.  These import lines were always present;
marking them with the indirect comment makes clear
which were imported directly and which are incidental.

R=ken2
CC=golang-dev
https://golang.org/cl/4295048