R=golang-dev, rsc
CC=golang-dev, remy
https://golang.org/cl/6554062

R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/6553067

R=rsc
CC=golang-dev
https://golang.org/cl/6543069

Fixes #4143.

R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/6541064

R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/6541062

Was describing an old implementation.

R=ken2
CC=golang-dev
https://golang.org/cl/6553066

Using offsets from Tos is cumbersome and we've had problems
in the past. Since it's only being used to grab the PID, we'll just
get that from the default TLS instead.

R=rsc, rminnich, npe
CC=golang-dev
https://golang.org/cl/6543049

Pointed out by James Chalfant after submit of CL 65000655.

TBR=golang-dev
CC=golang-dev
https://golang.org/cl/6543067

The code was assuming that int = 32 bits. Don't.

Update #2188.

R=agl
CC=golang-dev
https://golang.org/cl/6543063

I'm not exactly sure why there's a performance gain, but it seems like
an easy win. Maybe it's a cache line thing. Maybe it's that
unsafe.Sizeof(decompressor{}) drops to below unmappedzero, so that
checkref/checkoffset don't need to insert TESTB instructions. Maybe
it's less noise for the conservative garbage collector. Maybe it's
something else.

compress/flate benchmarks:
BenchmarkDecodeDigitsSpeed1e4          378628       349906   -7.59%
BenchmarkDecodeDigitsSpeed1e5         3481976      3204898   -7.96%
BenchmarkDecodeDigitsSpeed1e6        34419500     31750660   -7.75%
BenchmarkDecodeDigitsDefault1e4        362317       335562   -7.38%
BenchmarkDecodeDigitsDefault1e5       3290032      3107624   -5.54%
BenchmarkDecodeDigitsDefault1e6      30542540     28937480   -5.26%
BenchmarkDecodeDigitsCompress1e4       362803       335158   -7.62%
BenchmarkDecodeDigitsCompress1e5      3294512      3114526   -5.46%
BenchmarkDecodeDigitsCompress1e6     30514940     28927090   -5.20%
BenchmarkDecodeTwainSpeed1e4           412818       389521   -5.64%
BenchmarkDecodeTwainSpeed1e5          3475780      3288908   -5.38%
BenchmarkDecodeTwainSpeed1e6         33629640     31931420   -5.05%
BenchmarkDecodeTwainDefault1e4         369736       348850   -5.65%
BenchmarkDecodeTwainDefault1e5        2861050      2721383   -4.88%
BenchmarkDecodeTwainDefault1e6       27120120     25862050   -4.64%
BenchmarkDecodeTwainCompress1e4        372057       350822   -5.71%
BenchmarkDecodeTwainCompress1e5       2855109      2718664   -4.78%
BenchmarkDecodeTwainCompress1e6      26987010     26336030   -2.41%

image/png benchmarks:
BenchmarkDecodeGray               1841839      1802251   -2.15%
BenchmarkDecodeNRGBAGradient      7115318      6933280   -2.56%
BenchmarkDecodeNRGBAOpaque        6135892      6013284   -2.00%
BenchmarkDecodePaletted           1153313      1114302   -3.38%
BenchmarkDecodeRGB                5619404      5511190   -1.93%

R=rsc, r
CC=golang-dev
https://golang.org/cl/6533048

various implementation of collation.  The tool provides commands for soring,
regressing one implementation against another, and benchmarking.
Currently it includes collation implementations for the Go collator, ICU,
and one using Darwin's CoreFoundation framework.
To avoid building this tool in the default build, the colcmp tag has been
added to all files. This allows other tools/colcmp in this directory (e.g. it may make
sense to move maketables here) to be put in this directory as well.

R=r, rsc, mpvl
CC=golang-dev
https://golang.org/cl/6496118

Minor tweaks to avoid assuming that int is always 32 bits.

Update #2188.

R=golang-dev, r
CC=golang-dev
https://golang.org/cl/6553062

Can tell this doesn't get run very often, but it is still important
for when you've broken everything else.

R=golang-dev, r
CC=golang-dev
https://golang.org/cl/6547065

This CL fixes code that incorrectly assumes that int is 32 bits wide.
Specifically, the socketpair system call expects a pointer to a pair
of int32s, not a pair of ints. Fix this inside the wrappers without
changing the APIs.

Update #2188.

R=golang-dev, r
CC=golang-dev
https://golang.org/cl/6552063

Change the grammar so that field access is a proper operator.
This introduces a new node, ChainNode, into the public (but
actually internal) API of text/template/parse. For
compatibility, we only use the new node type for the specific
construct, which was not parseable before. Therefore this
should be backward-compatible.

Before, .X.Y was a token in the lexer; this CL breaks it out
into .Y applied to .X. But for compatibility we mush them
back together before delivering. One day we might remove
that hack; it's the simple TODO in parse.go/operand.

This change also provides grammatical distinction between
        f
and
        (f)
which might permit function values later, but not now.

Fixes #3999.

R=golang-dev, dsymonds, gri, rsc, mikesamuel
CC=golang-dev
https://golang.org/cl/6494119

The code was trying to interpret a uint32 n as a signed quantity
and then store it in an int. For this, int(n) currently works, but when
int becomes 64 bits one must write int(int32(n)) instead, to get
the 32-bit sign extension.

Update #2188.

R=golang-dev, nigeltao
CC=golang-dev
https://golang.org/cl/6551068

Fixes #3917.

R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/6553061

After each line, ast.Print would do a zero-length write,
which would hit the boundary condition on Plan 9 when
reading over pipes (since message boundaries are
preserved). This change makes sure we only do positive-
length writes.

R=rsc, rminnich, dave, r
CC=golang-dev
https://golang.org/cl/6558046

This was misleading and often confusing for new users.

Leaving the actual weekly release history page for historical reference.

R=golang-dev, rsc, adg
CC=golang-dev
https://golang.org/cl/6555053

Today, if run.go doesn't understand a test header line it just ignores
the test, making it too easy to write or edit tests that are not actually
being run.

- expand errorcheck to accept flags, so that bounds.go and escape*.go can run.
- create a whitelist of skippable tests in run.go; skipping others is an error.
- mark all skipped tests at top of file.

Update #4139.

R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/6549054

Revision 63f7abcae015 introduced a bug caused by
code assuming registers started at X5, not X0.

Fixes #4138.

R=rsc
CC=golang-dev, remy
https://golang.org/cl/6558043

R=rsc, r
CC=golang-dev
https://golang.org/cl/6552061

When I was a lad (and well past), users were in /usr;
now they're somewhere else, I'm told.
Kids today.

Fixes #4122.

R=golang-dev, dsymonds, rsc
CC=golang-dev
https://golang.org/cl/6552060

Fixes #3670.

R=ken2
CC=golang-dev
https://golang.org/cl/6542058

- remove ambiguous 'it'
- use a lower-case PackageName in the example

R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/6552057

The width was not being set on the address, which meant
that the optimizer could not find variables that overlapped
with it and mark them as having had their address taken.
This let to the compiler believing variables had been set
but never used and then optimizing away the set.

Fixes #4129.

R=ken2
CC=golang-dev
https://golang.org/cl/6552059

Fixes #4047.

R=iant, r
CC=golang-dev
https://golang.org/cl/6500065

Use the debug data for enums on windows.

Fixes #4120.

R=alex.brainman
CC=golang-dev
https://golang.org/cl/6545047

Add support for processing the COFF symbol table.

R=alex.brainman
CC=golang-dev
https://golang.org/cl/6551045

Fixes #4082.

R=dsymonds
CC=golang-dev
https://golang.org/cl/6545055

Fixes #4114.

R=golang-dev, iant, rsc, iant, devon.odell
CC=golang-dev
https://golang.org/cl/6553050

R=rsc, r
CC=gobot, golang-dev
https://golang.org/cl/6532049

These talks have been moved to the go.talks sub-repository:
        https://code.google.com/p/go.talks

R=rsc, r
CC=gobot, golang-dev
https://golang.org/cl/6529052

R=rsc, golang-dev
CC=golang-dev
https://golang.org/cl/6528044

Fixes #2446.

R=ken2
CC=golang-dev
https://golang.org/cl/6557044

R=rsc
CC=golang-dev
https://golang.org/cl/6554043

We only pass lower 32 bits of file offset to asm routine.

R=r, dave, rsc
CC=golang-dev
https://golang.org/cl/6499118

R=golang-dev, r
CC=golang-dev
https://golang.org/cl/6548046

*** There is an API change here: the introduction of the
LatinOffset int in the RangeTable struct. ***

* Avoid checking Latin range multiple times for non-Latin runes.
* Use linear search when it is faster than binary search.

go test -calibrate runs the calibration for where the linear/binary
crossover should be.

benchmark                       old MB/s     new MB/s  speedup
BenchmarkFields            36.27        41.43    1.14x
BenchmarkFieldsFunc        36.23        41.38    1.14x

The speedup here is evenly split between the linear scans
and the LatinOffset change. Both are about 1.07x.

R=r
CC=bradfitz, gobot, golang-dev
https://golang.org/cl/6526048

OpenBSD now requires ELF binaries to have a PT_NOTE that identifies
it as an OpenBSD binary. Refactor the existing NetBSD ELF signature
code and implement support for OpenBSD ELF signatures.

R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/6489131