cmd/go: skip vet when package cannot be build under "go test/vet"
If the the package cannot be built,
"go test" and "go vet" should not run the "vet" tool.
In that case only errors from the compilers will be displayed.
Fixes #26125
Change-Id: I5da6ba64bae5f44feaf5bd4e765eea85533cddd4
Reviewed-on: https://go-review.googlesource.com/123938
Run-TryBot: Yury Smolsky <yury@smolsky.by>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Bryan C. Mills <bcmills@google.com> Reviewed-by: Ian Lance Taylor <iant@golang.org>
David du Colombier [Thu, 19 Jul 2018 08:41:54 +0000 (10:41 +0200)]
cmd/go: fix TestScript on Plan 9
CL 123577 added TestScript. The install_rebuild_gopath
test was failing on Plan 9 because it defines a GOPATH
using the ':' separator, while Plan 9 expects the '\000'
separator in environment variables.
This change fixes the script engine by defining a new
":" environment variable set to OS-specific path list
separator.
The install_rebuild_gopath test has been updated to use
"${:}" instead of ":".
Fixes #26421.
Change-Id: I58a97f882cdb48cc0836398b0d98a80ea58041ba
Reviewed-on: https://go-review.googlesource.com/124435
Run-TryBot: David du Colombier <0intro@gmail.com> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Richard Musiol [Sat, 14 Jul 2018 10:19:36 +0000 (12:19 +0200)]
syscall/js: show goroutine stack traces on deadlock
When using callbacks, it is not necessarily a deadlock if there is no
runnable goroutine, since a callback might still be pending. If there
is no callback pending, Node.js simply exits with exit code zero,
which is not desired if the Go program is still considered running.
This is why an explicit check on exit is used to trigger the "deadlock"
error. This CL makes it so this is Go's normal "deadlock" error, which
includes the stack traces of all goroutines.
cmd/go: warn about non-use of go.mod in legacy go get
It's important for a smooth transition for non-module users
not to change operation in GOPATH/src by default in Go 1.11,
even if go.mod exists in a downloaded dependency.
Even so, users create go.mod and then are confused about
why 'go get' commands seem to behave oddly, when in fact
they are getting the old 'go get'.
Try to split the difference by printing a warning in 'go get'
when run in a tree that would normally be considered a
module if only it were outside GOPATH/src.
Change-Id: I55a1cbef127f3f36de54a8d7b93e1fc64bf0a708
Reviewed-on: https://go-review.googlesource.com/124859 Reviewed-by: Bryan C. Mills <bcmills@google.com>
cmd/go/internal/get: fix "mod over non-mod" preference for meta tags
If there was a mod and non-mod meta tag for a given prefix,
the meta tag extractor was already dropping the non-mod meta tag.
But we might have mod and non-mod meta tags with different
prefixes, in which case the mod tag should prevail when both match.
Fixes #26200.
Change-Id: I17ab361338e270b9fa03999ad1954f2bbe0f5017
Reviewed-on: https://go-review.googlesource.com/124714 Reviewed-by: Bryan C. Mills <bcmills@google.com>
go.sum accumulates cruft as modules are added and removed as
direct and indirect dependencies. Instead of exposing all that cruft,
let "go mod -sync" clean it out.
Fixes #26381.
Change-Id: I7c9534cf7cc4579f7f82646d00ff691c87a13c4a
Reviewed-on: https://go-review.googlesource.com/124713 Reviewed-by: Bryan C. Mills <bcmills@google.com>
People are (understandably) confused by creating go.mod files in GOPATH/src
and then having the go command not use modules in those directories.
We can't change that behavior (or we'll break non-module users of GOPATH)
but we can force 'go mod' (including 'go mod -init') to fail loudly in that case.
If this is not enough, the next step would be to print a warning every time
the go command is run in a GOPATH/src directory with a go.mod but
module mode hasn't triggered. But that will annoy all the non-module users.
Hopefully anyone confused will eventually run a 'go mod' command of
some kind, which will fail loudly.
Fixes #26365.
Change-Id: I8c5fe987fbc3f8d2eceb1138e6862a391ade150c
Reviewed-on: https://go-review.googlesource.com/124708 Reviewed-by: Bryan C. Mills <bcmills@google.com>
These were intentionally chosen to be valid semantic versions
that sort below any explicitly-chosen semantic version (even v0.0.0),
so that they could be used before anything was tagged but after
that would essentially only be useful in replace statements
(because the max operation during MVS would always prefer
a tagged version).
Then we changed the go command to accept hashes on the
command line, so that you can say
go get github.com/my/proj@abcdef
and it will download and use v0.0.0-yyyymmddhhmmss-abcdef123456.
If you were using v1.10.1 before and this commit is just little bit
newer than that commit, calling it v0.0.0-xxx is confusing but
also harmful: the go command sees the change from v1.10.1 to
the v0.0.0 pseudoversion as a downgrade, and it downgrades other
modules in the build. In particular if some other module has
a requirement of github.com/my/proj v1.9.0 (or later), the
pseudo-version appears to be before that, so go get would
downgrade that module too. It might even remove it entirely,
if every available version needs a post-v0.0.0 version of my/proj.
This CL introduces new pseudo-version forms that can be used
to slot in after the most recent explicit tag before the commit.
If the most recent tagged commit before abcdef is v1.10.1,
then now we will use
This has the right properties for downgrades and the like,
since it is after v1.10.1 but before almost any possible
successor, such as v1.10.2, v1.10.2-1, or v1.10.2-pre.
This CL also uses those pseudo-version forms as appropriate
when mapping a hash to a pseudo-version. This fixes the
downgrade problem.
Overall, this CL reflects our growing recognition of pseudo-versions
as being like "untagged prereleases".
Issue #26150 was about documenting best practices for how
to work around this kind of accidental downgrade problem
with additional steps. Now there are no additional steps:
the problem is avoided by default.
Fixes #26150.
Change-Id: I402feeccb93e8e937bafcaa26402d88572e9b14c
Reviewed-on: https://go-review.googlesource.com/124515 Reviewed-by: Bryan C. Mills <bcmills@google.com>
cmd/go/internal/module: add new +incompatible version build annotation
Repos written before the introduction of semantic import versioning
introduced tags like v2.0.0, v3.0.0, and so on, expecting that
(1) the import path would remain unchanged, and perhaps also
(2) there would be at most one copy of the package in a build.
We've always accommodated these by mapping them into the
v0/v1 version range, so that if you ran
go get k8s.io/client-go@v8.0.0
it would not complain about v8.x.x being a non-v1 version and
instead would map that version to a pseudo-version in go.mod:
The pseudo-version fails to capture two important facts: first,
that this really is the v8.0.0 tag, and second, that it should be
preferred over any earlier v1 tags.
A related problem is that running "go get k8s.io/client-go"
with no version will choose the latest v1 tag (v1.5.1), which
is obsolete.
This CL introduces a new version suffix +incompatible that
indicates that the tag should be considered an (incompatible)
extension of the v1 version sequence instead of part of its
own major version with its own versioned module path.
The requirement above can now be written:
require k8s.io/client-go v8.0.0+incompatible
(The +metadata suffix is a standard part of semantic versioning,
and that suffix is ignored when comparing two versions for
precedence or equality. As part of canonicalizing versions
recorded in go.mod, the go command has always stripped all
such suffixes. It still strips nearly all: only +incompatible is
preserved now.)
In addition to recognizing the +incompatible, the code that
maps a commit hash to a version will use that form when
appropriate, so that
go get k8s.io/client-go@7d04d0
will choose k8s.io/client-go@v8.0.0+incompatible.
Also, the code that computes the list of available versions from
a given source code repository also maps old tags to +incompatible
versions, for any tagged commit in which a go.mod file does not exist.
Therefore
will now choose v8.0.0+incompatible as the meaning of "latest tagged version".
The extraction of +incompatible versions from source code repos
depends on a codehost.Repo method ReadFileRevs, to do a bulk read
of multiple revisions of a file. That method is only implemented for git in this CL.
Future CLs will need to add support for that method to the other repository
implementations.
TSAN for Go only supports heap address in the range [0x00c000000000,
0x00e000000000). However, we currently create heap hints of the form
0xXXc000000000 for XX between 0x00 and 0x7f. Even for XX=0x01, this
hint is outside TSAN's supported heap address range.
Fix this by creating a slightly different set of hints in race mode,
all of which fall inside TSAN's heap address range.
This should fix TestArenaCollision flakes. That test forces the
runtime to use later heap hints. Currently, this always results in
TSAN "failed to allocate" failures on Windows (which happens to have a
slightly more constrained TSAN layout than non-Windows). Most of the
time we don't notice these failures, but sometimes it crashes TSAN,
leading to a test failure.
David Chase [Wed, 18 Jul 2018 15:23:56 +0000 (11:23 -0400)]
cmd/compile: avoid compressed dwarf when testing for gdb on OSX
Until we figure out how to deal with gdb on Darwin (doesn't
read compressed DWARF from binaries), avoid compressing
DWARF in that case so that the test will still yield meaningful
results.
This is also reported to be a problem for Windows.
Problem also exists for lldb, but this test doesn't check
lldb.
Rob Pike [Wed, 18 Jul 2018 05:10:04 +0000 (15:10 +1000)]
doc: update the Origins section of the FAQ
Completely replace the opener, which had become not only stale
but bad, expand the discussion of the gopher, and generally provide
prose more connected to the present than to the programming world
of 2007.
Fixes #26107
Change-Id: I5e72f0c81e71d1237fe142dc26114991329a6996
Reviewed-on: https://go-review.googlesource.com/124616 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Ian Lance Taylor [Wed, 18 Jul 2018 22:29:32 +0000 (15:29 -0700)]
doc: fill in final standard library TODOs in go1.11.html
Change-Id: Ic1ff580573711a6c91c1d5e3eb019a298a2fec49
Reviewed-on: https://go-review.googlesource.com/124837 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Daniel Martí [Tue, 17 Jul 2018 13:38:52 +0000 (14:38 +0100)]
cmd/compile: fix crash on invalid struct literal
If one tries to use promoted fields in a struct literal, the compiler
errors correctly. However, if the embedded fields are of struct pointer
type, the field.Type.Sym.Name expression below panics.
This is because field.Type.Sym is nil in that case. We can simply use
field.Sym.Name in this piece of code though, as it only concerns
embedded fields, in which case what we are after is the field name.
Added a test mirroring fixedbugs/issue23609.go, but with pointer types.
Fixes #26416.
Change-Id: Ia46ce62995c9e1653f315accb99d592aff2f285e
Reviewed-on: https://go-review.googlesource.com/124395
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Emmanuel Odeke <emm.odeke@gmail.com> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Ian Lance Taylor [Wed, 18 Jul 2018 20:19:04 +0000 (13:19 -0700)]
doc: update go1.11 release notes
Change-Id: I806d411c075cdc66322112b6ee5e50f58462bc6b
Reviewed-on: https://go-review.googlesource.com/124776 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Jack [Wed, 18 Jul 2018 19:12:14 +0000 (19:12 +0000)]
filepath: updates doc to give case where WalkFunc info arg may be nil
If a filepath.WalkFunc is called with an non-nil err argument, it's possible
that the info argument will be nil. The comment above filepath.WalkFunc now
reflects this.
Fixes #26425
Change-Id: Ib9963b3344587d2993f1698c5a801f2d1286856b
GitHub-Last-Rev: 553fc266b570d0c47efe12b3b670f88112e3b334
GitHub-Pull-Request: golang/go#26435
Reviewed-on: https://go-review.googlesource.com/124635 Reviewed-by: Ian Lance Taylor <iant@golang.org>
doc: fix a link in go1.11.html and flesh out a few sections
Change-Id: Ic5b9ccb2772534cee77ffcaeee617c7d5edfb6fd
Reviewed-on: https://go-review.googlesource.com/124715 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Hana (Hyang-Ah) Kim [Wed, 18 Jul 2018 07:06:21 +0000 (03:06 -0400)]
doc/go1.11: update runtime/pprof change note
Mention the change in the behavior of go test -memprofile.
Change-Id: I0384f058298bd8fcfd2d97996464d46b4e419938
Reviewed-on: https://go-review.googlesource.com/124656 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Alberto Donizetti [Wed, 18 Jul 2018 08:53:44 +0000 (10:53 +0200)]
doc/go1.11: add space in Go version name
Missed in CL 124516.
Change-Id: I6488196c8392987d69eca832ab4969aaafe1a26c
Reviewed-on: https://go-review.googlesource.com/124658 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Elias Naur [Wed, 18 Jul 2018 07:17:19 +0000 (09:17 +0200)]
doc/go1.11: mention the libSystem.so change for iOS
The change to make the runtime use libSystem.so macOS instead of
direct kernel calls applies to iOS as well.
Change-Id: I97ea86452ac5f7433aea58bbd3ff53a2eb2835e0
Reviewed-on: https://go-review.googlesource.com/124657 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Ian Lance Taylor [Wed, 18 Jul 2018 00:36:15 +0000 (17:36 -0700)]
cmd/cgo: don't report inconsistency error for incomplete typedef
In CLs 122575 and 123177 the cgo tool started explicitly looking up
typedefs. When there are two Go files using import "C", and the first
one has an incomplete typedef and the second one has a complete
version of the same typedef, then we will now record a version of the
first typedef which will not match the recorded version of the second
typedef, producing an "inconsistent definitions" error. Fix this by
silently merging incomplete typedefs with complete ones.
Fixes #26430
Change-Id: I9e629228783b866dd29b5c3a31acd48f6e410a2d
Reviewed-on: https://go-review.googlesource.com/124575
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
cmd/go/internal/modfetch: restrict file names in zip files, avoid case-insensitive collisions
Within the zip file for a given module, disallow names that are invalid
on various operating systems (mostly Windows), and disallow
having two different paths that are case-fold-equivalent.
Disallowing different case-fold-equivalent paths means the
zip file content is safe for case-insensitive file systems.
There is more we could do to relax the rules later, but I think
this should be enough to avoid digging a hole in the early days
of modules that's hard to climb out of later.
In tests on my repo test corpus, the repos now rejected are:
The : and ? are reserved on Windows,
and the : is half-reserved (and quite confusing) on macOS.
The ☺ is perhaps an overreach, but I am not convinced
that allowing all of category So is safe; certainly Sk is not.
cmd/go/internal/modfetch: do not rely on file system for case sensitivity
Over time there may exist two modules with names that differ only in case.
On systems with case-insensitive file systems, we need to make sure those
modules do not collide in the download cache.
Do this by using the new "safe encoding" for file system paths as well as
proxy paths.
Module paths, like import paths, are case-sensitive, for better or worse.
But not all file systems distinguish file paths with different cases.
If we are going to use module paths to construct file system paths,
we must apply an encoding that distinguishes case without relying
upon the file system to do it.
This CL defines that encoding, the "safe module path encoding".
Module paths today are ASCII-only with limited punctuation,
so the safe module path encoding is to convert the whole path
to lower case and insert an ! before every formerly upper-case letter:
github.com/Sirupsen/logrus is stored as github.com/!sirupsen/logrus.
Although this CL defines the encoding, it does not change the rest
of the go command to use the encoding. That will be done in
follow-up CLs.
path/filepath: make Walk example runnable in the playground
Relates to #9679
Change-Id: I68951f664d2a03812dae309c580c181869d8af21
Reviewed-on: https://go-review.googlesource.com/122237
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
cmd/cgo: update JNI's jobject to uintptr check for newer Android NDKs
In Android's NDK16, jobject is now declared as:
#ifdef __cplusplus
class _jobject {};
typedef _jobject* jobject;
#else /* not __cplusplus */
typedef void* jobject;
#endif
This makes the jobject to uintptr check fail because it expects the
following definition:
struct _jobject;
typedef struct _jobject *jobject;
Update the type check to handle that new type definition in both C and
C++ modes.
Fixes #26213
Change-Id: Ic36d4a5176526998d2d5e4e404f8943961141f7a
GitHub-Last-Rev: 42037c3c584579c2b3281c25372b830e864e7aec
GitHub-Pull-Request: golang/go#26221
Reviewed-on: https://go-review.googlesource.com/122217
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
Alberto Donizetti [Thu, 28 Jun 2018 18:22:13 +0000 (20:22 +0200)]
doc/go1.11: explain new vet typechecking behaviour in release notes
Since Go1.10, go test runs vet on the tests before executing them.
Moreover, the vet tool typechecks the package under analysis with
go/types before running. In Go1.10, a typechecking failure just caused
a warning to be printed. In Go1.11, a typechecking failure will cause
vet to exit with a fatal error (see Issue #21287).
This means that starting with Go1.11, tests that don't typecheck will
fail immediately. This would not normally be an issue, since a test
that doesn't typecheck shouldn't even compile, and it should already
be broken.
Unfortunately, there's a bug in gc that makes it accept programs with
unused variables inside a closure (Issue #3059). This means that a
test with an unused variable inside a closure, that compiled and
passed in Go1.10, will fail in the typechecking step of vet starting
with Go1.11.
Explain this in the 1.11 release notes.
Fixes #26109
Change-Id: I970c1033ab6bc985d8c64bd24f56e854af155f96
Reviewed-on: https://go-review.googlesource.com/121455 Reviewed-by: Ian Lance Taylor <iant@golang.org>
The original CL skipped the lldb test if it couldn't read compressed
DWARF, but lldb can never read compressed DWARF, so this effectively
disabled this test unconditionally.
The previous commit disabled DWARF compression for this test, so the
test now works on its own merits again. This CL reverts the change to
skip the test so we don't simply mask lldb failures.
lldb doesn't support compressed DWARF, so right now we're just always
skipping the lldb test. This CL makes the test run again by disabling
compressed DWARF just for this test.
Michael Munday [Tue, 17 Jul 2018 07:39:30 +0000 (08:39 +0100)]
cmd/compile: keep autos if their address reaches a control value
Autos must be kept if their address reaches the control value of a
block. We didn't see this before because it is rare for an auto's
address to reach a control value without also reaching a phi or
being written to memory. We can probably optimize away the
comparisons that lead to this scenario since autos cannot alias
with pointers from elsewhere, however for now we take the
conservative approach and just ensure the auto is properly
initialised if its address reaches a control value.
Fixes #26407.
Change-Id: I02265793f010a9e001c3e1a5397c290c6769d4de
Reviewed-on: https://go-review.googlesource.com/124335 Reviewed-by: David Chase <drchase@google.com>
The original cmd/go tests were tiny shell scripts
written against a library of shell functions.
They were okay to write but difficult to run:
you couldn't select individual tests (with -run)
they didn't run on Windows, they were slow, and so on.
CL 10464 introduced go_test.go's testgo framework
and later CLs translated the test shell script over to
individual go tests. This let us run tests selectively,
run tests on Windows, run tests in parallel, isolate
different tests, and so on. It was a big advance.
The tests had always been awkward to write.
Here was the first test in test.bash:
TEST 'file:line in error messages'
# Test that error messages have file:line information at beginning of
# the line. Also test issue 4917: that the error is on stderr.
d=$(TMPDIR=/var/tmp mktemp -d -t testgoXXX)
fn=$d/err.go
echo "package main" > $fn
echo 'import "bar"' >> $fn
./testgo run $fn 2>$d/err.out || true
if ! grep -q "^$fn:" $d/err.out; then
echo "missing file:line in error message"
cat $d/err.out
ok=false
fi
rm -r $d
This CL introduces a new facility meant as a successor to the testgo
approach that brings back the style of writing tests as little scripts,
but they are now scripts in a built-for-purpose shell-like language,
not bash itself. In this new form, the test above is a single file,
testdata/script/fileline.txt:
# look for short, relative file:line in error message
! go run ../../gopath/x/y/z/err.go
stderr ^..[\\/]x[\\/]y[\\/]z[\\/]err.go:
-- ../x/y/z/err.go --
package main; import "bar"
The file is a txtar text archive (see CL 123359) in which the leading comment
is the test script and the files are the initial state of the temporary file
system where the script runs.
Each script runs as a subtest, so that they can still be selected individually.
The scripts are kept isolated from each other by default,
so all script subtests are treated as parallel tests, for the
testing package to run in parallel. Even for the 15 tests in
this CL, that cuts the time for TestScript from 5.5s to 2.5s.
The scripts do not have access to the cmd/go source directory,
nor to cmd/go/testdata, so they are prevented from creating temporary
files in those places or modifying existing ones. (Many existing tests
scribble in testdata, unfortunately, especially testdata/pkg when
they run builds with GOPATH=testdata.)
This CL introduces the script facility and converts 15 tests.
The txtar archive form will allow us to delete the large trees of trivial
files in testdata; a few are deleted in this CL.
See testdata/script/README for details and a larger conversion example.
As part of converting testdata/script/test_badtest.txt,
I discovered that 'go test' was incorrectly printing a FAIL line
to stderr (not stdout) in one corner case. This CL fixes that
to keep the test passing.
Future CLs will convert more tests.
Change-Id: I11aa9e18dd2d4c7dcd8e310dbdc6a1ea5f7e54c1
Reviewed-on: https://go-review.googlesource.com/123577
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Rob Pike [Mon, 16 Jul 2018 10:32:05 +0000 (20:32 +1000)]
doc: update Usage section of the FAQ
This is close to a complete rewrite, as the content was pretty old.
The CL includes links to the Wiki for information about companies
using Go, a new section about IDEs and editors¹, and a restatement
of the foreign function interface story. It also modernizes and
expands a little on the use of Go inside Google.
¹ Ed is the standard editor.
Change-Id: I5e54aafa53d00d86297b2691960a376b40f6225b
Reviewed-on: https://go-review.googlesource.com/123922 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Rob Pike [Mon, 16 Jul 2018 21:22:41 +0000 (07:22 +1000)]
doc: rewrite run-on sentence in garbage collection discussion
Change-Id: I60cb7010448757ca4c7a2973bee2277b3d5fc439
Reviewed-on: https://go-review.googlesource.com/124175 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Ian Lance Taylor [Tue, 17 Jul 2018 02:09:24 +0000 (19:09 -0700)]
cmd/go: fix handling of vet.cfg with buggyInstall
The vet action assumes that a.Deps[0] is the compilation action for
which vet information should be generated. However, when using
-linkshared, the action graph is built with a ModeBuggyInstall action
to install the shared library built from the compilation action.
Adjust the set up of the vet action accordingly. Also don't clean up
the working directory after completing the buggy install.
Updates #26400
Change-Id: Ia51f9f6b8cde5614a6f2e41b6207478951547770
Reviewed-on: https://go-review.googlesource.com/124275
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Bryan C. Mills <bcmills@google.com>
cmd/go/internal/cache: squelch cache init warnings when $HOME is /
Docker sets $HOME to / when running with a UID that doesn't exist within
the container. This not uncommon on CI servers.
Fixes #26280
Change-Id: Ic7ff62b41403fe6e7c0cef12814667ef73f6c954
Reviewed-on: https://go-review.googlesource.com/122487
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Joe Tsai [Mon, 16 Jul 2018 20:05:25 +0000 (13:05 -0700)]
archive/zip: fix regression when writing directories
Several adjustments:
1) When encoding the FileHeader for a directory, explicitly set all of the sizes
to zero regardless of their prior values. These values are currently populated
by FileInfoHeader as it calls os.FileInfo.Size regardless of whether the file is
a directory or not. We avoid fixing FileInfoHeader now as it is too late in the
release cycle (see #24082).
We silently adjust slightly wrong FileHeader fields as opposed to returning
an error because the CreateHeader method already does such mutations
(e.g., for UTF-8 detection, data descriptor, etc).
2) Have dirWriter.Write only return an error if some number of bytes are written.
Some code still call Write for both normal files and directories, but just pass
an empty []byte to Write for directories.
When x509ignoreCN=1 is present in GODEBUG, ignore the deprecated Common
Name field. This will let people test a behavior we might make the
default in the future, and lets a final class of certificates avoid the
NameConstraintsWithoutSANs error.
crypto/x509: ignore Common Name when it does not parse as a hostname
The Common Name is used as a hostname when there are no Subject
Alternative Names, but it is not restricted by name constraints. To
protect against a name constraints bypass, we used to require SANs for
constrained chains. See the NameConstraintsWithoutSANs error.
This change ignores the CN when it does not look like a hostname, so we
can avoid returning NameConstraintsWithoutSANs.
This makes it possible to validate certificates with non-hostname CN
against chains that use name constraints to disallow all names, like the
Estonian IDs.
Updates #24151
Change-Id: I798d797990720a01ad9b5a13336756cc472ebf44
Reviewed-on: https://go-review.googlesource.com/123355 Reviewed-by: Adam Langley <agl@golang.org>
Keith Randall [Mon, 16 Jul 2018 17:45:25 +0000 (10:45 -0700)]
misc/cgo: fix darwin test, again
TARGET_OS_OSX is the right macro, but it also was only introduced
in 1.12. For 1.11 and earlier a reasonable substitution is
TARGET_OS_IPHONE == 0.
Update #24161
Update #26355
Change-Id: I5f43c463d14fada9ed1d83cc684c7ea05d94c5f3
Reviewed-on: https://go-review.googlesource.com/124075
Run-TryBot: Keith Randall <khr@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Ian Lance Taylor [Fri, 13 Jul 2018 16:57:17 +0000 (09:57 -0700)]
cmd/go: make TestNewReleaseRebuildsStalePackagesInGOPATH pass again
The test TestNewReleaseRebuildsStalePackagesInGOPATH is not run in
short mode, so people tend to not notice when it fails. It was failing
due to the build cache. Make it pass again by 1) changing it to modify
the package in a way visible to the compiler, so that the change is
not hidden by caching; 2) accepting "not installed but available in
build cache" as always being a valid reason for a stale package, as go
list does not try to figure out an underlying reason for why a package
is stale when it finds it in the build cache but not installed.
Updates #24436
Change-Id: Iaeaa298f153451ec913a653dd4e6da79a33055bb
Reviewed-on: https://go-review.googlesource.com/123815 Reviewed-by: Daniel Martí <mvdan@mvdan.cc> Reviewed-by: Bryan C. Mills <bcmills@google.com>
Hana (Hyang-Ah) Kim [Fri, 13 Jul 2018 20:11:24 +0000 (16:11 -0400)]
runtime/pprof: add a fake mapping when /proc/self/maps is unavailable
Profile's Mapping field is currently populated by reading /proc/self/maps.
On systems where /proc/self/maps is not available, the profile generated
by Go's runtime will not have any Mapping entry. Pprof command then adds
a fake entry and links all Location entries in the profile with the fake
entry to be used during symbolization.
https://github.com/google/pprof/blob/a8644067d5a3c9a6386e7c88fa4a3d9d37877ca3/internal/driver/fetch.go#L437
The fake entry is not enough to suppress the error or warning messages
pprof command produces. We need to tell pprof that Location entries are
symbolized already by Go runtime and pprof does not have to attempt to
perform further symbolization.
In #25743, we made Go runtime mark Mapping entries with HasFunctions=true
when all Location entries from the Mapping entries are successfully
symbolized. This change makes the Go runtime add a fake mapping entry,
otherwise the pprof command tool would add, and set the HasFunctions=true
following the same logic taken when the real mapping information is
available.
Updates #19790.
Fixes #26255. Tested pprof doesn't report the error message any more
for pure Go program.
Change-Id: Ib12b62e15073f5d6c80967e44b3e8709277c11bd
Reviewed-on: https://go-review.googlesource.com/123779
Run-TryBot: Hyang-Ah Hana Kim <hyangah@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Rob Pike [Sat, 14 Jul 2018 06:46:57 +0000 (16:46 +1000)]
doc: update Design and Types sections of the FAQ
Update #26107.
Change-Id: I8bfa5b01ce953c53f7fd7a866d0ece61ba04c618
Reviewed-on: https://go-review.googlesource.com/123919 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Rob Pike [Sat, 14 Jul 2018 04:52:31 +0000 (14:52 +1000)]
doc: update Values, Writing Code, and Pointers and Allocation sections of the FAQ
Significant surgery done to the Versioning section, bringing it closer to
modern thinking.
Also add a question about constants.
Update #26107.
Change-Id: Icf70b7228503c6baaeab0b95ee3e6bee921575aa
Reviewed-on: https://go-review.googlesource.com/123918 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Rob Pike [Sat, 14 Jul 2018 04:06:28 +0000 (14:06 +1000)]
doc: update Implementation and Performance sections of the FAQ
Changes are mostly about making more about now than about the past,
changing some verb tenses, and mentioning gollvm (which should
be pronounced "gollum" if you ask me).
Update #26107
Change-Id: I6c14f42b9fc2684259d4ba8bc149d7ec9bb83d15
Reviewed-on: https://go-review.googlesource.com/123917 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Rob Pike [Thu, 12 Jul 2018 23:08:13 +0000 (09:08 +1000)]
doc: rearrange the description of GOMAXPROCS
The old text was written when it was only 1 by default, which
changed a long time ago.
Also add a note that GOMAXPROCS does not limit the total
number of threads.
Change-Id: I104ccd7266d11335320a4d7f5671fb09ed641f88
Reviewed-on: https://go-review.googlesource.com/123916 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Keith Randall [Sun, 15 Jul 2018 00:18:45 +0000 (17:18 -0700)]
misc/cgo: fix test on iOS
The test in CL 123715 doesn't work on iOS, it needs to use a different
version scheme to determine whether SecKeyAlgorithm and friends exist.
Restrict the old version test to OSX only.
The same problem occurs on iOS: the functions tested don't exist before
iOS 10. But we don't have builders below iOS 10, so it isn't a big issue.
If we ever get older builders, or someone wants to run all.bash on an
old iOS, they'll need to figure out the right incantation.