This keeps fragmentation from delaying
garbage collections (and causing more fragmentation).
Cuts fresh godoc (with indexes) from 261M to 166M (120M live).
Cuts toy wc program from 50M to 8M.
Fixes#647.
R=r, cw
CC=golang-dev
https://golang.org/cl/257041
Previously we would require safe primes for our RSA key generation.
Since this took rather a long time, this removes the requirement that
the primes be safe.
OpenSSL doesn't use safe primes for RSA key generation either
(openssl-0.9.8l/crypto/rsa/rsa_gen.c:122)
Fixes#649.
R=rsc
CC=golang-dev
https://golang.org/cl/253041
Use hardware sqrt for faster hypot; preserve software-only
hypot as hypotGo (like sqrtGo); enable benchmarking of
hypotGo.
R=rsc
CC=golang-dev
https://golang.org/cl/229049
probablyPrime would return false negatives in some cases.
This code has now been tested against GMP for several million iterations without issues.
Fixes#638.
R=rsc
CC=golang-dev
https://golang.org/cl/252041
- only manual changes are in src/pkg/go/printer/nodes.go
- use a heuristic to determine "outliers" such that not entire composites are
forced to align with them
- improves several places that were not unligned before due too simple heuristic
- unalignes some cases that contain "outliers"
- gofmt -w src misc
Fixes#644.
R=rsc, r
CC=golang-dev
https://golang.org/cl/241041
Set defaults to the same values as various C DNS resolver libraries.
Use the timeout value from the configuration.
Fixes#632.
R=rsc
CC=golang-dev
https://golang.org/cl/223094
Background: This didn't matter until recently, because '\n' don't appear as token starts
in source code and thus the exact position was irrelevant (and set as was easiest in the
code). With auto semicolon insertion, a virtual semicolon may be inserted when a '\n' is
seen. The position of the semicolon is the position of the '\n'. Without this fix, these
semicolons appeared on the next line instead of the line where they were inserted. This
affected the association of comments to declarations in the parser. As a result, some
lead comments where considered line comments, not collected in the ast, and not shown in
godoc pages. (This affected only godoc pages, not gofmt-formatted programs).
Fixes#592.
R=rsc
CC=golang-dev
https://golang.org/cl/224068
- gofmt -w src misc
- improves several lists and fixes minor degradation introduced
with the fix for issue 628
- removed some dead code (stringList)
R=rsc
CC=golang-dev
https://golang.org/cl/223058
- applied gofmt to src and misc
Note: This fix improved formatting of src/pkg/math/all_test.go but leads
to a degradation in src/pkg/exp/4s/xs.go. The latter happened to "work"
before accidentally. Fixing the alignment in that case in general will
be a separate CL.
Fixes#628.
R=rsc
CC=golang-dev
https://golang.org/cl/223054
path.Match() errors out when testing "*x" against "xxx"
because it matches the star non-greedily. Ensure that
the last chunk consumes the rest of the name.
R=r, rsc
CC=golang-dev
https://golang.org/cl/223050
- always include position information about opening/closing parens/braces
- replace uses of []*ast.Field with *ast.FieldList
Fixes#473.
R=rsc
CC=golang-dev
https://golang.org/cl/223043
- remove several TODOs
- as a side-effect, comment stylers are now used always and comments
will be properly colored in godoc pkg documentation pages (and not
only when looking at source text)
R=rsc
CC=golang-dev
https://golang.org/cl/222041
- go/ast: removed StringList (not needed anymore)
- go/ast: changed import path and field list tag to a single string
- updated all dependencies
R=rsc
CC=golang-dev
https://golang.org/cl/217056
Compliance issue addressed here: POST requests carrying form data are required
to use "identity" transfer encoding by common nginx and apache server configurations,
e.g. wordpress.com (and many others). So, Request needed to be able to send
non-chunked encodings.
Thus, Request is extended to support identity and chunked encodings, like
Response. Since the Read() and Write() logic are shared by both (and are
quite long), it is exported in a separate file transfer.go.
R=rsc
CC=golang-dev
https://golang.org/cl/217048
Try to avoid infinite recursion if String fails due to
printing a bad type.
Add test for String method with named basic types.
R=r
CC=golang-dev
https://golang.org/cl/207102
because free needs to mark the block as freed to
coordinate with the garbage collector.
(in C++ free can blindly put the block on the free list,
no questions asked, so the cache saves some work.)
R=iant
CC=golang-dev
https://golang.org/cl/206069
not trying to be secure, just not repetitive
(sending with the same id repeatedly makes
some resolvers stop replying.)
eventually we'll replace this with linking against
the system's native resolver.
R=p
CC=golang-dev
https://golang.org/cl/207051
* specialize sweepspan as sweepspan0 and sweepspan1.
* in sweepspan1, inline "free" to avoid expensive mlookup.
R=iant
CC=golang-dev
https://golang.org/cl/206060
Added tests and benchmarks for Exp2 (special cases same
as Exp). Log1p also enhances speed of inverse hyperbolics.
R=rsc
CC=golang-dev
https://golang.org/cl/206058
Add functions, tests and benchmarks. Fix typos in comments
in expm1 and hypot_386. Fix Acosh domain error in benchmark
test.
R=rsc
CC=golang-dev
https://golang.org/cl/204069
Averaged times (in microseconds) for drawing an 800x600 rectangle
are listed below. The summary is: around a 100x improvement.
draw.Draw call times were typically linear in the number of pixels
touched (i.e. drawing an 800x600 rectangle took 100x as much time as
drawing an 80x60 rectangle).
Before this change, there was only the general-but-slow code path.
When drawing any src with a 50%-opaque mask: 237300 us
When drawing any src with a nil mask: 50100 us
After this change, the 50%-opaque mask case is unchanged.
For an *image.RGBA dst and nil mask and...
...a uniform color (i.e. an image.ColorImage) src: 282 us
...another *image.RGBA src: 615 us.
For the curious, an intermediate implementation detected the special
cases but used simple nested for loops instead of the built-in copy
function. The respective times (compared to 282 and 615 for the
final implementation, or 50100 for the original) were 3110 and 3573.
Times were measured with 8g/8l on my laptop. I haven't tried gccgo
or other architectures.
R=r, rsc
CC=golang-dev
https://golang.org/cl/201048
Marshalling:
* Fixes several silly bugs.
* Support the RawContents type.
* Support the RawValue type.
* Recurse into nested interface{}.
Both directions:
* Better handling of SETs. You can now tag an element in a
structure with "set" to get the correct tag for serialisation.
* For types that aren't in a structure, you can now name them
with "SET" on the end.
* SETs are no longer implicitly treated as SEQUENCEs.
R=rsc
CC=golang-dev
https://golang.org/cl/201049
Make sure to print a time zone when formatting even if none is defined.
Add a comment introducing lookupTimezone (not lookupTimeZone).
Fixes isse 577.
R=rsc
CC=golang-dev
https://golang.org/cl/196090
Added special cases to comments for asin.go and fabs.go.
Added Trunc() to floor.go and floor_386.s. Fixed formatting
error in hypot_386.s Added new functions Acosh, Asinh,
Atanh, Copysign, Erf, Erfc, Expm1, and Log1p. Added
386 FPU version of Fmod. Added tests, benchmarks, and
precision to expected results in all_test.go. Edited
makefile so it all compiles.
R=rsc
CC=golang-dev
https://golang.org/cl/195052
* renamed channels to say what gets sent
* use channel closed status instead of racy check of boolean
R=nigeltao_golang
CC=golang-dev
https://golang.org/cl/196065
In particular, add field-value tokenizer which respects quoting rules.
The code is intended for use in tokenizing the Transfer-Encoding and Trailer fields.
The lexing function is not connected to the main parsing code yet (in the next CL).
R=rsc
CC=golang-dev
https://golang.org/cl/190085
(1) http.Response must close resp.Body after writing.
(2) Case when resp.Body != nil and resp.ContentLength = 0 should not be
treated as an error in Response.Write, because this is what
ReadResponse often returns.
(3) Changed body.th to body.hdr for readability.
R=rsc
CC=golang-dev
https://golang.org/cl/194084
Seems to be enabled spuriously during godoc (can't see why),
producing errors like:
parser.parseDir: src/pkg/http/server.go:159:16: 'Write' declared already at src/pkg/http/request.go:140:21 (and 4 more errors)
R=r
CC=golang-dev
https://golang.org/cl/194119
- provide scope to parse functions; if non-nil, parser uses the scope
to declare and lookup identifiers
- resolve forward references where possible
R=rsc
CC=golang-dev
https://golang.org/cl/194098
Hostname reads the file /proc/sys/kernel/hostname to determine
the value it returns. Some people set this to a Fully Qualified
Doamin Name. At least one implementation of /bin/hostname
truncates the name it gets (often from the "uname" system call)
at the first dot unless it is given a "-f" flag. This change makes
the unit test also truncate at the first dot and checks if the strings
then match. This seems more portable than adding an extra flag
to the called /bin/hostname program.
R=rsc
CC=golang-dev
https://golang.org/cl/181097
Added 386 FPU version of Hypot; modified all_test.go to test
Hypot with large arguments. Also edited sqrt.go to remove
Sqrt(0) as a special case.
R=rsc
CC=golang-dev
https://golang.org/cl/186180
bonus: type switch now detects multiple uses of identical interface types.
bonus: interface types are now order-independent, following the spec.
R=ken2
CC=golang-dev
https://golang.org/cl/194053
5g/6g/8g: add import statements to export metadata, mapping package path to package name.
recognize "" as the path of the package in export metadata.
use "" as the path of the package in object symbol names.
5c/6c/8c, 5a/6a/8a: rewrite leading . to "". so that ·Sin means Sin in this package.
5l/6l/8l: rewrite "" in symbol names as object files are read.
gotest: handle new symbol names.
gopack: handle new import lines in export metadata.
Collectively, these changes eliminate the assumption of a global
name space in the object file formats. Higher level pieces such as
reflect and the computation of type hashes still depend on the
assumption; we're not done yet.
R=ken2, r, ken3
CC=golang-dev
https://golang.org/cl/186263
Other code fixing:
- Fixed bugs in get32.
- Fix code for parsing display string (as a new function).
- Fix code for connecting to X server. The old code only work
if the server is listening to TCP port, otherwise it doesn't
work (at least in my PC).
R=nigeltao_golang, rsc, jhh
CC=golang-dev
https://golang.org/cl/183111
need better management of mach semaphores eventually
but this avoids allocating them for uncontended Locks.
R=r
CC=agl1, golang-dev
https://golang.org/cl/190079
limitations:
poor error handling
teardown not done
exporter must send, importer must receive
testing is rudimentary at best
R=rsc
CC=golang-dev
https://golang.org/cl/186234
Was trying to interpret raw body as chunked body.
Add test for ReadResponse.
Fixes#544.
R=r, petar-m
CC=golang-dev, shadowice
https://golang.org/cl/190068
Wrong position information for identifier
could lead to destructive reformatting of
source via hg gofmt.
R=rsc
CC=golang-dev
https://golang.org/cl/189100
- Identifiers refer now to the language entity (Object)
that they denote. At the moment this is at best an
approximation.
- Initial data structures for language entities (Objects)
and expression types (Type) independent of the actual
type notations.
- Initial support for declaring and looking up identifiers.
- Updated various dependent files and added support functions.
- Extensively tested to avoid breakage. This is an AST change.
R=rsc
CC=golang-dev, rog
https://golang.org/cl/189080
* move memory code into $GOOS-specific directory.
* allow printing of static strings < 256 bytes.
(dynamic strings will bump maxstring as they are allocated.)
* use cgo2c for runtime.mal.
R=r, dho
CC=golang-dev
https://golang.org/cl/186143
Emphasis on minimal interface and fast scanning.
Recognizes all Go literals by default. Easily
configurable to recognize different whitespace
characters and tokens.
Provides detailed position information for each
token.
R=rsc, r
CC=golang-dev
https://golang.org/cl/181160
Added special case tests to all_test.go for Fmod. Fixed Fmod [hung
for Fmod(+/-Inf, <finite>)]. Also added test for Ceil in all_test.go.
R=rsc
CC=golang-dev
https://golang.org/cl/186076
now printed using the semicolon-free style.
Removed NoSemis flag and mechanism dealing with
optional semicolons from go/printer.
Updated all go/printer output files using the
semi-colon free style.
Changes have no formatting impact on checked in
go code under src and misc.
R=rsc, r
CC=golang-dev
https://golang.org/cl/184068
Addresses some cases and respects position of
closing parenthesis; but doesn't work yet in
all cases. Heuristic; will need to be replaced
by a real solution eventually. Does not cause
existing code to be reformatted.
R=r
CC=golang-dev, rsc
https://golang.org/cl/186043
The model is that formatString is a a representation of a standard time,
and that Format converts the time to that representation.
Standard representaitons are defined for ANSIC, RFC850, RFC1123, and ISO8601.
There's also a humane Kitchen fomat: 3:04PM.
R=rsc, benolive, cw
CC=golang-dev
https://golang.org/cl/181130
used only for debugging, debug.go is not normally part of the package source.
also add a dump program to call it.
R=rsc
CC=golang-dev
https://golang.org/cl/183075
(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
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
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
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
- 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
(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
This is not a complete JPEG implementation (e.g. it does not handle
progressive JPEGs or restart markers), but I was able to take a photo
with my phone, and view the resultant JPEG in pure Go.
The decoder is simple, but slow. The Huffman decoder in particular
should be easily improvable, but optimization is left to future
changelists. Being able to inline functions in the inner loop should
also help performance.
The output is not pixel-for-pixel identical to libjpeg, although
identical behavior isn't necessarily a goal, since JPEG is a lossy
codec. There are at least two reasons for the discrepancy.
First, the inverse DCT algorithm used is the same as Plan9's
src/cmd/jpg, which has different rounding errors from libjpeg's
default IDCT implementation. Note that libjpeg actually has three
different IDCT implementations: one floating point, and two fixed
point. Out of those four, Plan9's seemed the simplest to understand,
partly because it has no #ifdef's or C macros.
Second, for 4:2:2 or 4:2:0 chroma sampling, this implementation does
nearest neighbor upsampling, compared to libjpeg's triangle filter
(e.g. see h2v1_fancy_upsample in jdsample.c).
The difference from the first reason is typically zero, but sometimes
1 (out of 256) in YCbCr space, or double that in RGB space. The
difference from the second reason can be as large as 8/256 in YCbCr
space, in regions of steep chroma gradients. Informal eyeballing
suggests that the net difference is typically imperceptible, though.
R=r
CC=golang-dev, rsc
https://golang.org/cl/164056
parsing and printing to new syntax.
Use -oldparser to parse the old syntax,
use -oldprinter to print the old syntax.
2) Change default gofmt formatting settings
to use tabs for indentation only and to use
spaces for alignment. This will make the code
alignment insensitive to an editor's tabwidth.
Use -spaces=false to use tabs for alignment.
3) Manually changed src/exp/parser/parser_test.go
so that it doesn't try to parse the parser's
source files using the old syntax (they have
new syntax now).
4) gofmt -w src misc test/bench
5th and last set of files.
R=rsc
CC=golang-dev
https://golang.org/cl/180050
parsing and printing to new syntax.
Use -oldparser to parse the old syntax,
use -oldprinter to print the old syntax.
2) Change default gofmt formatting settings
to use tabs for indentation only and to use
spaces for alignment. This will make the code
alignment insensitive to an editor's tabwidth.
Use -spaces=false to use tabs for alignment.
3) Manually changed src/exp/parser/parser_test.go
so that it doesn't try to parse the parser's
source files using the old syntax (they have
new syntax now).
4) gofmt -w src misc test/bench
4th set of files.
R=rsc
CC=golang-dev
https://golang.org/cl/180049
parsing and printing to new syntax.
Use -oldparser to parse the old syntax,
use -oldprinter to print the old syntax.
2) Change default gofmt formatting settings
to use tabs for indentation only and to use
spaces for alignment. This will make the code
alignment insensitive to an editor's tabwidth.
Use -spaces=false to use tabs for alignment.
3) Manually changed src/exp/parser/parser_test.go
so that it doesn't try to parse the parser's
source files using the old syntax (they have
new syntax now).
4) gofmt -w src misc test/bench
3rd set of files.
R=rsc
CC=golang-dev
https://golang.org/cl/180048
parsing and printing to new syntax.
Use -oldparser to parse the old syntax,
use -oldprinter to print the old syntax.
2) Change default gofmt formatting settings
to use tabs for indentation only and to use
spaces for alignment. This will make the code
alignment insensitive to an editor's tabwidth.
Use -spaces=false to use tabs for alignment.
3) Manually changed src/exp/parser/parser_test.go
so that it doesn't try to parse the parser's
source files using the old syntax (they have
new syntax now).
4) gofmt -w src misc test/bench
1st set of files.
R=rsc
CC=agl, golang-dev, iant, ken2, r
https://golang.org/cl/180047
parsing and printing to new syntax.
Use -oldparser to parse the old syntax,
use -oldprinter to print the old syntax.
2) Change default gofmt formatting settings
to use tabs for indentation only and to use
spaces for alignment. This will make the code
alignment insensitive to an editor's tabwidth.
Use -spaces=false to use tabs for alignment.
3) Manually changed src/exp/parser/parser_test.go
so that it doesn't try to parse the parser's
source files using the old syntax (they have
new syntax now).
4) gofmt -w src misc test/bench
2nd set of files.
R=rsc
CC=golang-dev
https://golang.org/cl/179067
mustgetc reports unexpected EOF as SyntaxError. using
mustgetc seems to be a better approach than letting the
caller handle unexpected EOF every time.
name: the second if statement should explicitly return
ok==false.
R=rsc
https://golang.org/cl/174083
1. If all data is exhausted using Read then a following Next will
fail as if it saw EOF. (Test case added.)
2. Seeking isn't always possible (i.e. sockets and pipes). Fallback
to read. (Test case added.)
3. Fix to readHeader (cleaner fix pointed out by rsc).
(TestReader modified.)
4. When Read has consumed all the data, don't try to read 0 bytes from reader.
In cases where tr.nb is zero we attempt to read zero bytes and thus
never see an EOF (this is most easily seen when the 'tar source' is
something like bytes.Buffer{} as opposed to os.File).
5. If write is used to the point of ErrWriteTooLong, allow additional file entries.
6. Make close work as expected. That is any further Write or
WriteHeader attempts will result in ErrWriteAfterClose.
Fixes#419.
R=rsc, dsymonds1
https://golang.org/cl/162062
- no need to replace comments for stand-alone blocks
- always print string concatenations with interspersed "+"
(remove option)
- minor cleanups
R=rsc
https://golang.org/cl/174076
This change removes the necessity to have GOBIN in $PATH,
and also doesn't assume that the build is being run from
$GOROOT/src. This is a minimal set of necessary changes
to get Go to build happily from the FreeBSD ports
collection.
R=rsc
CC=golang-dev
https://golang.org/cl/171044
- oldparser parse old syntax (required semicolons)
- oldprinter print old syntax (required semicolons)
By default, these flags are enabled for now.
Setting -oldparser=false has no effect until go/parser is changed
to accept the new syntax.
Enabled exp/parser in Makefile; update dependent exp/eval.
R=rsc
https://golang.org/cl/174051
FreeBSD was passing stk as the new thread's stack base, while
stk is the top of the stack in go. The added check should cause
a trap if this ever comes up in any new ports, or regresses
in current ones.
R=rsc
CC=golang-dev
https://golang.org/cl/167055
nodes in the tree are nested with respect to one another.
a simple change to the Visitor interface makes it possible
to do this (for example to maintain a current node-depth, or a
knowledge of the name of the current function).
Visit(nil) is called at the end of a node's children;
this make possible the channel-based interface below,
amongst other possibilities.
It is still just as simple to get the original behaviour - just
return the same Visitor from Visit.
Here are a couple of possible Visitor types.
// closure-based
type FVisitor func(n interface{}) FVisitor
func (f FVisitor) Visit(n interface{}) Visitor {
return f(n);
}
// channel-based
type CVisitor chan Visit;
type Visit struct {
node interface{};
reply chan CVisitor;
};
func (v CVisitor) Visit(n interface{}) Visitor
{
if n == nil {
close(v);
} else {
reply := make(chan CVisitor);
v <- Visit{n, reply};
r := <-reply;
if r == nil {
return nil;
}
return r;
}
return nil;
}
R=gri
CC=rsc
https://golang.org/cl/166047
Roughly 33% faster for simple cases, probably more for complex ones.
Before:
mallocs per Sprintf(""): 4
mallocs per Sprintf("xxx"): 6
mallocs per Sprintf("%x"): 10
mallocs per Sprintf("%x %x"): 12
Now:
mallocs per Sprintf(""): 2
mallocs per Sprintf("xxx"): 3
mallocs per Sprintf("%x"): 5
mallocs per Sprintf("%x %x"): 7
Speed improves because of avoiding mallocs and also by sharing a bytes.Buffer
between print.go and format.go rather than copying the data back after each
printed item.
Before:
fmt_test.BenchmarkSprintfEmpty 1000000 1346 ns/op
fmt_test.BenchmarkSprintfString 500000 3461 ns/op
fmt_test.BenchmarkSprintfInt 500000 3671 ns/op
Now:
fmt_test.BenchmarkSprintfEmpty 2000000 995 ns/op
fmt_test.BenchmarkSprintfString 1000000 2745 ns/op
fmt_test.BenchmarkSprintfInt 1000000 2391 ns/op
fmt_test.BenchmarkSprintfIntInt 500000 3751 ns/op
I believe there is more to get but this is a good milestone.
R=rsc
CC=golang-dev, hong
https://golang.org/cl/166076
For 386 we use the [f]statfs64 system call, which takes three
parameters: the filename, the size of the statfs64 structure,
and a pointer to the structure itself.
R=rsc
https://golang.org/cl/166073
On a microbenchmark that ping-pongs on lots of channels, this makes
the multithreaded case about 20% faster and the uniprocessor case
about 1% slower. (Due to cache effects, I expect.)
R=rsc, agl
CC=golang-dev
https://golang.org/cl/166043
Use them in Copy and Copyn.
Speed up ReadFile by using ReadFrom and avoiding Copy altogether (a minor win).
R=rsc, gri
CC=golang-dev
https://golang.org/cl/166041
tabs for indentation even if -spaces is set.
Changes to gofmt:
- added -tabindent flag
- don't recompute parser and printer mode repeatedly
Changes to go/printer:
- provide new printing mode TabIndent
Changes to tabwriter:
- implement new mode TabIndent to use tabs independent
of the actual padding character for leading empty columns
- distinguish between minimal cell width and tab width
(tabwidth is only used if the output contains tabs,
minwidth and padding are always considered)
- fixed and added more comments
- some additional factoring
By default, -tabindent is disabled and the default gofmt
behavior is unchanged. By setting -spaces and -tabindent,
gofmt will use tabs for indentation but do any other
alignment with spaces. This permits a user to change the
visible indentation by simply changing the editor's tab
width and the code will remain properly aligned without
the need to rerun gofmt.
R=rsc
https://golang.org/cl/163068
1) need to send slice and array types (was only sending element types)
2) compatibleType needs to use decoder's type map
R=rsc
CC=golang-dev
https://golang.org/cl/164062