Commits · f740d717bd31e1a4b8281072b5cb3dfeb26f3273 · go / golang

08 Dec, 2015 1 commit

crypto/elliptic: resample private keys if out of range. · f740d717

Adam Langley authored Dec 04, 2015

The orders of the curves in crypto/elliptic are all very close to a
power of two. None the less, there is a tiny bias in the private key
selection.

This change makes the distribution uniform by resampling in the case
that a private key is >= to the order of the curve. (It also switches
from using BitSize to Params().N.BitLen() because, although they're the
same value here, the latter is technically the correct thing to do.)

The private key sampling and nonce sampling in crypto/ecdsa don't have
this issue.

Fixes #11082.

Change-Id: Ie2aad563209a529fa1cab522abaf5fd505c7269a
Reviewed-on: https://go-review.googlesource.com/17460Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>

f740d717

07 Dec, 2015 14 commits

doc: add heading IDs to Code of Conduct · aa487e66

Chris Broadfoot authored Dec 07, 2015

Fixes #13514

Change-Id: I3903d3926ed4f5d54cfb77209d93c950b832b933
Reviewed-on: https://go-review.googlesource.com/17511Reviewed-by: Francesc Campoy Flores <campoy@google.com>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

aa487e66

go/parser, go/types: report invalid else branch in if statements · 670642d3

Robert Griesemer authored Dec 04, 2015

- Only accept valid if statement syntax in go/parser.

- Check AST again in go/types since it may have been modified and the
  AST doesn't preclude other statements in the else branch of an if
  statement.

- Removed a test from gofmt which verified that old-style if statements
  permitting any statement in the else branch were correctly reformatted.
  It's been years since we switched to the current syntax; no need to
  support this anymore.

- Added a comment to go/printer.

Fixes #13475.

Change-Id: Id2c8fbcc68b719cd511027d0412a37266cceed6b
Reviewed-on: https://go-review.googlesource.com/17408Reviewed-by: Russ Cox <rsc@golang.org>

670642d3

cmd/go: fix bad shared lib name with buildmode=shared · c5aa53c8

Aleksandr Demakin authored Aug 25, 2015

Use import paths of packages to build a shared lib name.
Use arguments for meta-packages 'std', 'cmd', and 'all'.

Fixes #12236

Change-Id: If274d63301686ef34e198287eb012f9062541ea0
Reviewed-on: https://go-review.googlesource.com/13921Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

c5aa53c8

encoding/gob: document behavior of zero-valued arrays, slices, and maps · 148b13c3

Rob Pike authored Dec 03, 2015

The documentation was inconsistent. It said zero values were not sent, but
that zero-valued elements of arrays and arrays were sent. But which rule
applies if the array is all zero elements, and is therefore itself a zero value?

The answer is: the array is transmitted. In principle the other choice could
be made, but there would be considerable expense and complexity required
to implement this behavior now, not to mention worries about changes of
behavior.

Therefore we just document the situation: Arrays, slices, and maps are
always encoded. It would perhaps be nice to have sorted this out earlier,
but it was a missed opportunity.

Fixes #13378

Change-Id: I8fae345edfa707fcfa7a3e0160d87ff1ac5cc5a2
Reviewed-on: https://go-review.googlesource.com/17394Reviewed-by: Russ Cox <rsc@golang.org>

148b13c3

cmd/compile/internal/gc: fix internal compiler error on invalid declaration · 70da2d0a

Didier Spezia authored Oct 04, 2015

Following an empty import, a declaration involving a ? symbol
generates an internal compiler error when the name of the
symbol (in newname function).

package a
import""
var?

go.go:2: import path is empty
go.go:3: internal compiler error: newname nil

Make sure dclname is not called when the symbol is nil.
The error message is now:

go.go:2: import path is empty
go.go:3: invalid declaration
go.go:4: syntax error: unexpected EOF

This CL was initially meant to be applied to the old parser,
and has been updated to apply to the new parser.

Fixes #11610

Change-Id: I75e07622fb3af1d104e3a38c89d9e128e3b94522
Reviewed-on: https://go-review.googlesource.com/15268Reviewed-by: Russ Cox <rsc@golang.org>

70da2d0a

cmd/compile: avoid converting huge floats to integers · 715f6377

Robert Griesemer authored Dec 03, 2015

Fixes #13471.

Change-Id: I232ad1729343d020254e313cfff182695ad6fc54
Reviewed-on: https://go-review.googlesource.com/17401Reviewed-by: Russ Cox <rsc@golang.org>

715f6377

runtime: best-effort detection of concurrent misuse of maps · 50c50420

Russ Cox authored Dec 07, 2015

If reports like #13062 are really concurrent misuse of maps,
we can detect that, at least some of the time, with a cheap check.

There is an extra pair of memory writes for writing to a map,
but to the same cache line as h.count, which is often being modified anyway,
and there is an extra memory read for reading from a map,
but to the same cache line as h.count, which is always being read anyway.
So the check should be basically invisible and may help reduce the
number of "mysterious runtime crash due to map misuse" reports.

Change-Id: I0e71b0d92eaa3b7bef48bf41b0f5ab790092487e
Reviewed-on: https://go-review.googlesource.com/17501Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Austin Clements <austin@google.com>

50c50420

cmd/compile/internal/gc: fix panic in Type Stringer · 7ebb96a4

Didier Spezia authored Aug 29, 2015

The following code:

func n() {(interface{int})}

generates:

3: interface contains embedded non-interface int
3: type %!v(PANIC=runtime error: invalid memory address or nil pointer dereference) is not an expression

It is because the corresponding symbol (Sym field in Type object)
is nil, resulting in a panic in typefmt.

Just skip the symbol if it is nil, so that the error message becomes:

3: interface contains embedded non-interface int
3: type interface { int } is not an expression

Fixes #11614

Change-Id: I219ae7eb01edca264fad1d4a1bd261d026294b00
Reviewed-on: https://go-review.googlesource.com/14015Reviewed-by: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>

7ebb96a4

cmd/compile: base mapaccess optimizations on algtype · 423b0cc2

Matthew Dempsky authored Dec 07, 2015

algtype already controls the behavior of the normal map access code
paths, so it makes sense to base the decision on which optimized paths
are applicable on it too.

Enables use of optimized paths for key types like [8]byte and struct{s
string}.

Fixes #13271.

Change-Id: I48c52d97abaa7259ad5aba9641ea996a967cd359
Reviewed-on: https://go-review.googlesource.com/17464
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>

423b0cc2

cmd/go: always show current value for GO15VENDOREXPERIMENT · b2492d16

Daniel Theophanes authored Dec 06, 2015

Prior behavior would show empty string when unset. In go1.5 this
would result in "off". In go1.6 this will result in "on". This
change will make empty or "0" off and "1" on for go1.5 and go1.6.
Vendor tools can then rely on this value.

Discussion:
https://groups.google.com/forum/#!topic/golang-dev/oZzcXrlRrkA

Change-Id: I7e145a32e813dfde02dc262a9186c7af28db7b92
Reviewed-on: https://go-review.googlesource.com/17487Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

b2492d16

net/http: sniffWrite test also in http2 mode · 3d704758

Emmanuel Odeke authored Dec 05, 2015

Change-Id: Ifa9f1ed6a3b8d3f7536f2d315259940335b0ee49
Reviewed-on: https://go-review.googlesource.com/17438Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

3d704758

crypto/x509: convert ErrInsecureAlgorithm into a type · c34bc90f

Brad Fitzpatrick authored Dec 03, 2015

Change-Id: I411aeaf0cf75eb8b1c9005b622f664e9f25e4a68
Reviewed-on: https://go-review.googlesource.com/17400Reviewed-by: Russ Cox <rsc@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

c34bc90f

net/url: accept non-ASCII bytes in URL per RFC 3986 · a6869d1c

Russ Cox authored Dec 04, 2015

Fixes #7991.
Fixes #12719.

Change-Id: I5650fa35ec5d49addeda6cc6e7fa93cfbe1cdfc0
Reviewed-on: https://go-review.googlesource.com/17385Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>

a6869d1c

go/doc, syscall: change 'more then' to 'more than' · 84a875ca

andrey mirtchovski authored Dec 07, 2015

This change modifies comments to use the more gramatically correct "more than"
instead of "more then".

Change-Id: Ie3bddcf25eb6b243a21da934f2f3c76a750c083a
Reviewed-on: https://go-review.googlesource.com/17488Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

84a875ca

06 Dec, 2015 2 commits

doc: go1.6.txt: note windows path/filepath.Join behaviour change · b2963a54

Alex Brainman authored Dec 06, 2015

Change-Id: I321eba716319bf88695ac49580837b6254f1279e
Reviewed-on: https://go-review.googlesource.com/17474Reviewed-by: Alex Brainman <alex.brainman@gmail.com>

b2963a54

path/filepath: handle c: as first parameter in Join properly · 51a112fe

Alex Brainman authored Dec 05, 2015

This is CL 11882 brought back to life.

Fixes #11551

Change-Id: I29810183957745442d1e9937f56a66fc9c6cc82a
Reviewed-on: https://go-review.googlesource.com/17470Reviewed-by: Russ Cox <rsc@golang.org>
Run-TryBot: Alex Brainman <alex.brainman@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>

51a112fe

05 Dec, 2015 19 commits

misc/cgo/stdio: reenable tests · 05390a07

Russ Cox authored Dec 05, 2015

The build tags are necessary to keep "go build" in that directory
building only stdio.go, but we have to arrange for test/run.go to
treat them as satisfied.

Fixes #12625.

Change-Id: Iec0cb2fdc2c9b24a4e0530be25e940aa0cc9552e
Reviewed-on: https://go-review.googlesource.com/17454
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

05390a07

runtime: fix integer comparison in signal handling · b9fde860

Russ Cox authored Dec 05, 2015

(sig is unsigned, so sig-1 >= 0 is always true.)

Fixes #11281.

Change-Id: I4b9d784da6e3cc80816f2d2f7228d5d8a237e2d5
Reviewed-on: https://go-review.googlesource.com/17457Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>

b9fde860

runtime: document that NumCPU does not change · c5a94ba2

Russ Cox authored Dec 05, 2015

Fixes #11609.

Change-Id: I3cf64164fde28ebf739706728b84d8ef5b6dc90e
Reviewed-on: https://go-review.googlesource.com/17456Reviewed-by: Ian Lance Taylor <iant@golang.org>

c5a94ba2

mime: fix parsing of empty string attribute value · 025e9b0c

Russ Cox authored Dec 05, 2015

Fixes #11290.

Change-Id: I312f0731077b78a4bed47062eb7fd1ab52bc3dd1
Reviewed-on: https://go-review.googlesource.com/17453Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

025e9b0c

runtime: check and fail early with a message if MMX is not available on 386 · 35e84546

Shenghou Ma authored Oct 17, 2015

Fixes #12970.

Change-Id: Id0026e8274e071d65d47df63d65a93110abbec5d
Reviewed-on: https://go-review.googlesource.com/15998Reviewed-by: Keith Randall <khr@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

35e84546

Revert "syscall: route_freebsd switch routing socket sysctl to use NET_RT_IFLISTL" · 4a291505

Yuval Pavel Zholkover authored Dec 05, 2015

This reverts commit f25f6eab.

Sorry, this was not meant to go in without the ztypes_freebsd_arm.go and the copyFromV9 function.

Change-Id: I4ac2a8a23809ec1b1b9e42992cd0f3c349848f06
Reviewed-on: https://go-review.googlesource.com/17472
Run-TryBot: Mikio Hara <mikioh.mikioh@gmail.com>
Run-TryBot: Alex Brainman <alex.brainman@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

4a291505

cmd/vet: move cmd/vet/whitelist to cmd/vet/internal/whitelist · 5b1a8a52

Russ Cox authored Dec 05, 2015

This was a mistake made when bringing cmd/vet into the main repo.

Fixes #13416.

Change-Id: I03b512ab944577c56085aea06df8ff5e1acc16d7
Reviewed-on: https://go-review.googlesource.com/17455Reviewed-by: Matthew Dempsky <mdempsky@google.com>

5b1a8a52

net: adjust Lookup API test cases · d2ca4518

Mikio Hara authored Dec 04, 2015

This change makes existing Lookup API test cases conform to the new
return value form that all the Lookup APIs except LookupTXT must return
a single or multiple absolute domain names.

Updates #12189.
Fixes #12193.

Change-Id: I03ca09be5bff80e818fbcdc26039daa33d5440a8
Reviewed-on: https://go-review.googlesource.com/17411
Run-TryBot: Mikio Hara <mikioh.mikioh@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

d2ca4518

cmd/compile: reject slice/map/func comparisons against converted nil · 336c9982

Matthew Dempsky authored Dec 04, 2015

Fixes #13480.

Change-Id: Icbf4f83e965e84f7020f56c3f346193f8b91e7bf
Reviewed-on: https://go-review.googlesource.com/17461Reviewed-by: Russ Cox <rsc@golang.org>

336c9982

syscall: route_freebsd switch routing socket sysctl to use NET_RT_IFLISTL · f25f6eab

Yuval Pavel Zholkover authored Sep 19, 2015

Switch IfMsghdr and IfaMsghdr to their 'l' variants, make the IfData layout
to be based on FreeBSD-11.0 (freebsdVersion >= 1100011).
Using freebsdVersion, detect the appropriate layout at runtime and decode
routing socket messages into the new IfData layout.

Fixes #11641

Change-Id: Ic7ec550f00c0d15f46a36f560d835e4f138f61e1
Reviewed-on: https://go-review.googlesource.com/14757Reviewed-by: Russ Cox <rsc@golang.org>

f25f6eab

cmd/link: clean up dwarf.go · 77328b8d

Didier Spezia authored Oct 10, 2015

Try to remove the most visible artefacts resulting from the
C to Go translation. It includes:

- refactoring the find function to eliminate goto and variable declarations
- removing useless variables still having a _ = xxx
- decreasing the number of upfront variable declarations

No semantic changes.

Change-Id: I84d981c48b2d9e22e6b9db5f2a703c80c60249ba
Reviewed-on: https://go-review.googlesource.com/15681Reviewed-by: Russ Cox <rsc@golang.org>

77328b8d

cmd/pprof: exit with non-zero status code on error · 08ab43ec

Mohit Agarwal authored Nov 06, 2015

Set the status code in case of error.

Fixes #11510

Change-Id: If461c30a1f6d2275539f33a2eabd7b19bbfa411d
Reviewed-on: https://go-review.googlesource.com/16718Reviewed-by: Russ Cox <rsc@golang.org>

08ab43ec

cmd/vet: make "-all -shadow" mean all default checks and -shadow · 3fb9e087

Konstantin Shaposhnikov authored Oct 26, 2015

Prior to this change "go tool vet -all -shadow" ran only -shadow check.

Also fix godoc package path in the usage text.

Fixes #13020

Change-Id: I87c60d6b06a02106ae8bff56adb79df032cc4646
Reviewed-on: https://go-review.googlesource.com/16325Reviewed-by: Russ Cox <rsc@golang.org>

3fb9e087

cmd/pprof/internal: map/slice literals janitoring · e0c05fcb

Didier Spezia authored Aug 23, 2015

Simplify slice/map literal expression.
Caught with gofmt -d -s, fixed with gofmt -w -s

Change-Id: I19723900d0649019bf79b9330d68525a68ed69c4
Reviewed-on: https://go-review.googlesource.com/13835Reviewed-by: Russ Cox <rsc@golang.org>

e0c05fcb

net/url: accept empty port after colon in IPv6 literal host · a456e356

Russ Cox authored Dec 04, 2015

Fixes #12200.

Change-Id: I89f2a7326bb9182024c44bf815a06fa48639649d
Reviewed-on: https://go-review.googlesource.com/17384Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

a456e356

misc/cgo/testcshared: use fd 100 instead of fd 10 for back-channel communication · b5e2ec8b

Russ Cox authored Dec 04, 2015

There is a report that fd 10 is already in use when run on some OS X machines.
I don't see how, and I can't reproduce the problem on my own OS X machine,
but it's easy enough to fix.

Fixes #12161.

Change-Id: I73511bdd91258ecda181d60d2829add746d1198b
Reviewed-on: https://go-review.googlesource.com/17451Reviewed-by: Ian Lance Taylor <iant@golang.org>

b5e2ec8b

runtime/cgo: assume Solaris thread stack is at least 1 MB · cd58f44b

Russ Cox authored Dec 04, 2015

When run with "ulimit -s unlimited", the misc/cgo/test test binary
finds a stack size of 0x3000 returned by getcontext, causing the
runtime to try to stay within those bounds and then fault when
called back in the test after 64 kB has been used by C.

I suspect that Solaris is doing something clever like reporting the
current stack size and growing the stack as faults happen.
On all the other systems, getcontext reports the maximum stack size.
And when the ulimit is not unlimited, even Solaris reports the
maximum stack size.

Work around this by assuming that any stack on Solaris must be at least 1 MB.

Fixes #12210.

Change-Id: I0a6ed0afb8a8f50aa1b2486f32b4ae470ab47dbf
Reviewed-on: https://go-review.googlesource.com/17452Reviewed-by: Ian Lance Taylor <iant@golang.org>

cd58f44b

syscall: fix ParseRoutingSockaddr name in docs · 5c248322

Yann Kerhervé authored Dec 04, 2015

Change-Id: I3870150fc8e713f6164371299c029b31f18f250a
Reviewed-on: https://go-review.googlesource.com/17426Reviewed-by: Minux Ma <minux@golang.org>

5c248322

net/http: make TestRedirect also test in http2 mode · c6d660a3

Emmanuel Odeke authored Dec 04, 2015

Change-Id: Idfbe91abc11b2b3b735cd8d11fb1938f9e5c0473
Reviewed-on: https://go-review.googlesource.com/17437Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

c6d660a3

04 Dec, 2015 4 commits

net/http: sniff tests ported to h2 · c2290fee

Emmanuel Odeke authored Dec 04, 2015

Change-Id: Icad6cc130252ac177946b23c12f36d6ba3275bf0
Reviewed-on: https://go-review.googlesource.com/17436
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

c2290fee

net/http: TimeoutHandler test ported to HTTP/2 · 2f2423c8

Emmanuel Odeke authored Dec 04, 2015

Change-Id: I69e62199140e5cf203696af17d4816f3a5f13bd1
Reviewed-on: https://go-review.googlesource.com/17434
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

2f2423c8

net/http: testHeadResponses port for h2 · 983926a7

Emmanuel Odeke authored Dec 04, 2015

Change-Id: I4cf4596a305510cd25df34ecebd11a19266415f1
Reviewed-on: https://go-review.googlesource.com/17435
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

983926a7

archive/tar: properly parse GNU base-256 encoding · 5ebb374f

Joe Tsai authored Dec 02, 2015

Motivation:
* Previous implementation did not detect integer overflow when
parsing a base-256 encoded field.
* Previous implementation did not treat the integer as a two's
complement value as specified by GNU.

The relevant GNU specification says:
<<<
GNU format uses two's-complement base-256 notation to store values
that do not fit into standard ustar range.
>>>

Fixes #12435

Change-Id: I4639bcffac8d12e1cb040b76bd05c9d7bc6c23a8
Reviewed-on: https://go-review.googlesource.com/17424Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

5ebb374f