tabs after an empty line where not converted.

Also, made it more robust in the presence of
(unexpected) ' ' and '\v' chars in indentation
mode.

R=r
CC=golang-dev
https://golang.org/cl/181085

Use uint index variables in some cases instead
of int to enable strength reduction; this makes
it possible for the compiler to reduce % into
masks.

Old code: 6g -S md4.go md4block.go | grep "md4block.go:44"
0471 (md4block.go:44) MOVL    AX,BX
0472 (md4block.go:44) MOVL    AX,BP
0473 (md4block.go:44) MOVL    AX,R8
0474 (md4block.go:44) SARL    $31,R8
0475 (md4block.go:44) SHRL    $30,R8
0476 (md4block.go:44) ADDL    R8,BP
0477 (md4block.go:44) SARL    $2,BP
0478 (md4block.go:44) IMULL   $4,BP
0479 (md4block.go:44) SUBL    BP,BX
0480 (md4block.go:44) MOVLQSX BX,BX
0481 (md4block.go:44) LEAQ    shift1+0(SB),BP
0482 (md4block.go:44) CMPL    BX,8(BP)
0483 (md4block.go:44) JCS     ,485
0484 (md4block.go:44) CALL    ,runtime.throwindex+0(SB)
0485 (md4block.go:44) MOVQ    (BP),BP
0486 (md4block.go:44) MOVL    (BP)(BX*4),DI

New code: 6g -S md4.go md4block.go | grep "md4block.go:44"
0471 (md4block.go:44) MOVL    AX,BX
0472 (md4block.go:44) ANDL    $3,BX
0473 (md4block.go:44) MOVLQZX BX,BX
0474 (md4block.go:44) LEAQ    shift1+0(SB),BP
0475 (md4block.go:44) CMPL    BX,8(BP)
0476 (md4block.go:44) JCS     ,478
0477 (md4block.go:44) CALL    ,runtime.throwindex+0(SB)
0478 (md4block.go:44) MOVQ    (BP),BP
0479 (md4block.go:44) MOVL    (BP)(BX*4),DI

R=agl, agl1
CC=golang-dev
https://golang.org/cl/181086

(I was looking at this code accidentally because of some gofmt
issues and thought that one could write this more effectively.
You may have deliberately chosen not to use ranges here to make
the index range clearer. Just let me know.)

R=agl, agl1
CC=golang-dev
https://golang.org/cl/181084

Fixes #457.

R=r
CC=golang-dev
https://golang.org/cl/181083

Fixes #459.

R=rsc, imkrasin, sonia
CC=golang-dev
https://golang.org/cl/181073

R=agl, agl1
CC=golang-dev
https://golang.org/cl/181080

Listener contains private members and 6g now enforces that private
members cannot be assigned outside of their package.

R=rsc
CC=golang-dev
https://golang.org/cl/183073

reported by erik quanstrom.

R=ken2
https://golang.org/cl/181071

also a way to run code when object is deleted.
both of these are possibilities, not certainties.

R=rsc, iant
CC=golang-dev
https://golang.org/cl/181057

a possible interface for more granular sleep times.

Fixes issue #260.

R=rsc, r
CC=golang-dev
https://golang.org/cl/181058

(i thought these were legal in <pre> blocks)

R=rsc
CC=golang-dev, pwil3058
https://golang.org/cl/181055

R=dsymonds1
CC=golang-dev
https://golang.org/cl/183049

gccgo currently miscompiles this test.

R=rsc
CC=golang-dev
https://golang.org/cl/181050

Without this change, the spec parses <-c.(type) as (<-c).(type)
but parses <-c.(int) as <-(c.(int)).

With this change, the former parses as <-(c.(type)).

All three parsers already implement this rule, because
they look for the "type" word during parsing of a standard
type assertion.  This change merely brings the
spec in line with the implementations.

http://code.google.com/p/go/source/browse/src/cmd/gc/go.y#801
http://code.google.com/p/go/source/browse/src/pkg/go/parser/parser.go#900
http://gcc.gnu.org/viewcvs/branches/gccgo/gcc/go/parse.cc?revision=155389&view=markup#l2469

Pointed out by Brett Kail on golang-nuts.

R=gri
CC=golang-dev
https://golang.org/cl/182044

This was convenient for me to have without being forced
to parse the regexp myself. I'd understand if it's not
really wanted, but I also think that some meta information
about compiled regexps would be fine.

R=r, rsc
CC=golang-dev
https://golang.org/cl/183044

R=rsc
CC=golang-dev
https://golang.org/cl/181049

Fixes #309.

R=rsc
CC=golang-dev
https://golang.org/cl/181044

for now, it's amd64 and 386 only but it's trivial to add more.

Fixes #385.

(why couldn't it have been issue 386?)

tested for amd64 and 386 on darwin.

R=rsc
CC=golang-dev
https://golang.org/cl/182043

Add support for TLS extensions in general and Next Protocol
Negotiation in particular.

R=rsc
CC=golang-dev
https://golang.org/cl/181045

It's expected to be shared between all files so that all types are output.
Fixes bug reported on mailing list by Peter Froehlich.

R=rsc, phf
CC=golang-dev
https://golang.org/cl/183043

R=r, rsc
CC=golang-dev
https://golang.org/cl/183042

Fixes #452.

R=gri
CC=golang-dev, dougfelt
https://golang.org/cl/181043

as it is not needed anymore (only one impl.
of vector package).

Makefile, vector_test.go, and nogen_test.go
were modified manually (find/replace), the
other files (intvector_test.go, strinvector_test.go
are generated).

Runs all tests.

R=r
https://golang.org/cl/182041

Manual changes to the following files:
src/pkg/Makefile
src/pkg/exp/vector/Makefile (now: src/pkg/container/vector/Makefile)

R=rsc, r
CC=golang-dev
https://golang.org/cl/181041

R=r
CC=golang-dev
https://golang.org/cl/180108

R=r
CC=golang-dev
https://golang.org/cl/179130

R=r
CC=golang-dev
https://golang.org/cl/180118

- use an interface {Get()}
- implement Get for maps, slices
- for slices, retrieves the address of the end of the array, which will give the
	same value for every slice of the same array.

R=rsc
CC=golang-dev
https://golang.org/cl/179129

R=gri
CC=golang-dev
https://golang.org/cl/179128

(Also fix case sensitivity in test for PTR inside fmt_test.go)
Fixes #441.

R=rsc, iant
CC=golang-dev
https://golang.org/cl/180112

R=r
CC=golang-dev
https://golang.org/cl/179120

R=gri
CC=rsc
https://golang.org/cl/178048

R=r, rsc
CC=golang-dev
https://golang.org/cl/179118

Fixes #436.

R=ken2
https://golang.org/cl/180105

R=dho
CC=golang-dev
https://golang.org/cl/180099

gccgo currently fails this test.

R=rsc
CC=golang-dev
https://golang.org/cl/179108

R=ken2
https://golang.org/cl/180092

corrects a common misunderstanding about NewBuffer.

R=rsc
CC=golang-dev
https://golang.org/cl/179106

(Thanks to ken and rsc for pointing this out)

rsc:
	ken pointed out that there's a race in the new
	one-lock-per-channel code.  the issue is that
	if one goroutine has gone to sleep doing

	select {
	case <-c1:
	case <-c2:
	}

	and then two more goroutines try to send
	on c1 and c2 simultaneously, the way that
	the code makes sure only one wins is the
	selgen field manipulation in dequeue:

	       // if sgp is stale, ignore it
	       if(sgp->selgen != sgp->g->selgen) {
		       //prints("INVALID PSEUDOG POINTER\n");
		       freesg(c, sgp);
		       goto loop;
	       }

	       // invalidate any others
	       sgp->g->selgen++;

	but because the global lock is gone both
	goroutines will be fiddling with sgp->g->selgen
	at the same time.

This results in a 7% slowdown in the single threaded case for a
ping-pong microbenchmark.

Since the cas predominantly succeeds, adding a simple check first
didn't make any difference.

R=rsc
CC=golang-dev
https://golang.org/cl/180068

now that the parser doesn't do this test anymore

R=rsc
CC=golang-dev
https://golang.org/cl/179105