Just saw a few dragonfly failures here.

I'm tempted to preemptively add plan9 here too, but I'll wait until
I see it fail.

Change-Id: Ic99fc088dbfd1aa21f509148aee98ccfe7f640bf
Reviewed-on: https://go-review.googlesource.com/18306Reviewed-by: Russ Cox <rsc@golang.org>

Fixes #12790.

Change-Id: I0f231d198c76632c23692fc1337b57cfeafaf4c7
Reviewed-on: https://go-review.googlesource.com/18338Reviewed-by: Andrew Gerrand <adg@golang.org>

Fixes #13269.

Change-Id: I960d1825bda9d8873c2a9005872c45e4c7d30111
Reviewed-on: https://go-review.googlesource.com/18339Reviewed-by: Ian Lance Taylor <iant@golang.org>

Fixes #12059.

Change-Id: Ib5caf8133cd3ed888f9102dfbfeca11c506f3b5b
Reviewed-on: https://go-review.googlesource.com/18337Reviewed-by: Ian Lance Taylor <iant@golang.org>

This time with a test.
Also adjust another test to skip when hg is not present,
and delete no longer needed fixDetachedHead code.

Fixes #9032 (again).

Change-Id: I481717409e1d44b524f83c70a8dc377699d1a2a5
Reviewed-on: https://go-review.googlesource.com/18334Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Fixes #12542.

Change-Id: Icd0fa84d891d6b1feab9b4d4dd319cdf1e6d6c48
Reviewed-on: https://go-review.googlesource.com/18336Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

But also cache the previous parsed form and don't reread if the
size and modification time are both unchanged from before.

On systems with stable /etc/hosts this should result in more stat calls
but only a single parsing of /etc/hosts.

On systems with variable /etc/hosts files (like some Docker systems)
this should result in quicker adoption of changes.

Fixes #13340.

Change-Id: Iba93b204be73d6d903cd17c58038a4fcfd0952b9
Reviewed-on: https://go-review.googlesource.com/18258Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Add a couple more cases where we convert random network I/O errors
into errRequestCanceled if the request was forcefully aborted.

It failed ~1/1000 times without -race, or very easily with -race.
(due to -race randomizing some scheduling)

Fixes #11894

Change-Id: Ib1c123ce1eebdd88642da28a5948ca4f30581907
Reviewed-on: https://go-review.googlesource.com/18287Reviewed-by: Russ Cox <rsc@golang.org>

This shouldn't need to exist in general, but in practice I want something
like this a few times per year.

Change-Id: I9c220e58be44b7726f75d776f714212c570cf8bb
Reviewed-on: https://go-review.googlesource.com/18286Reviewed-by: Russ Cox <rsc@golang.org>

The old link died; replace with an archive.org copy.

Fixes #13345.

Change-Id: Ic4a7fdcf258e1ff3b4a02ecb4f237ae7db2686c7
Reviewed-on: https://go-review.googlesource.com/18335Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Adds a test that both http1 and http2's Transport send a default
User-Agent, with the same behavior.

Updates bundled http2 to golang.org/x/net git rev 1ade16a545 (for
https://go-review.googlesource.com/18285)

The http1 behavior changes slightly: if req.Header["User-Agent"] is
defined at all, even if it's nil or a zero-length slice, then the
User-Agent header is omitted. This is a slight behavior change for
http1, but is consistent with how http1 & http2 do optional headers
elsewhere (such as "Date", "Content-Type"). The old behavior (set it
explicitly to "", aka []string{""}) still works as before. And now
there are even tests.

Fixes #13685

Change-Id: I5786a6913b560de4a5f1f90e595fe320ff567adf
Reviewed-on: https://go-review.googlesource.com/18284
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

The textual export data generated by gc sometimes contains forward
references of packages. In rare cases such forward-referenced packages
were not created when needed because no package name was present.

Create unnamed packages in this case and set the name later when it
becomes known.

Fixes #13566.

Change-Id: I193e0ec712e874030b194ab8ecb3fca140f7997a
Reviewed-on: https://go-review.googlesource.com/18301Reviewed-by: Russ Cox <rsc@golang.org>
Reviewed-by: Alan Donovan <adonovan@google.com>

This was supposed to be in CL 18204 but I submitted from the web
instead of my computer and lost this final edit.

Change-Id: I41598e936bb088d77f5e44752eda74222a4208c7
Reviewed-on: https://go-review.googlesource.com/18310Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

This was supposed to be in CL 18205 but I submitted via the web
instead of from my computer, so it got lost.
May deflake some things.

Change-Id: I880fb74b5943b8a17f952a82639c60126701187a
Reviewed-on: https://go-review.googlesource.com/18259Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Change-Id: I8ee338c1244fc4e2fb75deec752a7f83239c33ea
Reviewed-on: https://go-review.googlesource.com/18257Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Fixes #13425.

Change-Id: I82f49d07841e89fc4f3fde9ea41e710a634579d0
Reviewed-on: https://go-review.googlesource.com/18313Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Fixes #13781.

Change-Id: Icfac8f2bfc3a4106f646409cfdc053df1e0cd76a
Reviewed-on: https://go-review.googlesource.com/18314Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

Fixes #13693.

Change-Id: Ieb89d8a8874ee8c6b304de99dceb07193c26b60e
Reviewed-on: https://go-review.googlesource.com/18315Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

We don't use these for benchmarking anymore.
Now we have the go1 dir and the benchmarks subrepo.
Some have problematic copyright notices, so move out of main repo.

Preserved in golang.org/x/exp/shootout.

Fixes #12688.
Fixes #13584.

Change-Id: Ic0b71191ca1a286d33d7813aca94bab1617a1c82
Reviewed-on: https://go-review.googlesource.com/18320Reviewed-by: Ian Lance Taylor <iant@golang.org>

This reverts commit ab096d58.

Change-Id: Icf366aa43acc41b4f8474edae0297e554368bf14
Reviewed-on: https://go-review.googlesource.com/18321Reviewed-by: Russ Cox <rsc@golang.org>

Fixes #13725.

Change-Id: I5fe46851b238fc9ab301da8f8fc37bd1b7871748
Reviewed-on: https://go-review.googlesource.com/18316Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Do not check DaylightName if DynamicDaylightTimeDisabled is false.

Fixes #13502

Change-Id: I03e01ec0afdeb0037a5d3ee243674e5a6b559da5
Reviewed-on: https://go-review.googlesource.com/17998Reviewed-by: Yasuhiro MATSUMOTO <mattn.jp@gmail.com>
Reviewed-by: Russ Cox <rsc@golang.org>

Simply checking the exit code of `git rev-parse --git-dir` should
suffice here, but that requires deviating from the infrastructure
provided by `run`, so I've left that for a future change.

Fixes #11211.

Change-Id: I7cbad86a8a06578f52f66f734f5447b597ddc962
Reviewed-on: https://go-review.googlesource.com/18213Reviewed-by: Russ Cox <rsc@golang.org>

Fixes #13821.

Change-Id: I4a28a92d137edac3061537af25ac9d7aba411a66
Reviewed-on: https://go-review.googlesource.com/18262Reviewed-by: Russ Cox <rsc@golang.org>

Since Stop was introduced, it would revert to the system default for the
signal, rather than to the default Go behavior.  Change it to revert to
the default Go behavior.

Change-Id: I345467ece0e49e31b2806d6fce2f1937b17905a6
Reviewed-on: https://go-review.googlesource.com/18229Reviewed-by: Russ Cox <rsc@golang.org>

Mention that:
- connection pooling is enabled by default,
- the Transport is safe for concurrent use, and
- the Client type should be used for high-level stuff.

Change-Id: Idfd8cc852e733c44211e77cf0e22720b1fdca39b
Reviewed-on: https://go-review.googlesource.com/18273Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Fixes #13833

Change-Id: If0bd5f7dcfc39d34680d11eb998050f0900d5a26
Reviewed-on: https://go-review.googlesource.com/18283Reviewed-by: Russ Cox <rsc@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>

Update net/http's copy of http2 (sync as of x/net git rev 961116aee,
aka https://golang.org/cl/18266)

Also adds some CONNECT tests for #13717 (mostly a copy of http2's
version of test, but in the main repo it also tests that http1 behaves
the same)

Fixes #13668
Fixes #13717

Change-Id: I7db93fe0b7c42bd17a43ef32953f2d20620dd3ea
Reviewed-on: https://go-review.googlesource.com/18269
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Also reference the new Transport.ExpectContinueTimeout after the
mention of 100-continue.

Fixes #13721

Change-Id: I3445c011ed20f29128092c801c7a4bb4dd2b8351
Reviewed-on: https://go-review.googlesource.com/18281Reviewed-by: Andrew Gerrand <adg@golang.org>

Update #11845.

Change-Id: I1c248dc48abc62e51836b9ba50d6deb89706c730
Reviewed-on: https://go-review.googlesource.com/18226Reviewed-by: Russ Cox <rsc@golang.org>

Avoids an msan error when runtime/cgo is explicitly rebuilt with
-fsanitize=memory.

Fixes #13815.

Change-Id: I70308034011fb308b63585bcd40b0d1e62ec93ef
Reviewed-on: https://go-review.googlesource.com/18263Reviewed-by: Russ Cox <rsc@golang.org>

Thanks to Kevin Kirsche (github kkirsche).

Change-Id: Ia0017371f56065a5e88d1ebb800a6489136ee9b1
Reviewed-on: https://go-review.googlesource.com/18280Reviewed-by: Andrew Gerrand <adg@golang.org>

SEC-1 says: “The component privateKey is the private key defined to be
the octet string of length ⌊log₂(n)/8⌋ (where n is the order of the
curve)”.

Previously the code for parsing ECC private keys would panic (on
non-amd64) when the private was too long. It would also pass a too-short
private key to crypto/elliptic, possibly resulting in undesirable
behaviour.

This change makes the parsing function handle both too much and too
little padding because GnuTLS does the former and OpenSSL did the latter
until 30cd4ff294252c4b6a4b69cbef6a5b4117705d22. It also causes
serialisation to pad private keys correctly.

Fixes #13699

Change-Id: If9c2faeaeb45af8a4d7770d784f3d2633e7f8290
Reviewed-on: https://go-review.googlesource.com/18094
Run-TryBot: Adam Langley <agl@golang.org>
Reviewed-by: Adam Langley <agl@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>

Also add relevant test.

Fixes #13343.

Change-Id: Ib1e65af1d643d501de89adee3618eddbf6c69c9e
Reviewed-on: https://go-review.googlesource.com/18159Reviewed-by: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Change-Id: I45b026f4b4bf23b222a81669280cbe245048022b
Reviewed-on: https://go-review.googlesource.com/18212Reviewed-by: Russ Cox <rsc@golang.org>

Before golang.org/cl/13921, "go install -buildmode=shared prefix/..." created a
file called "libprefix.so", which was obviously a problem when prefix was
something like "." or "../".  However, now it expands the ... into all the
matched packages, joins them with -, which can clearly be a very long name
indeed. Because I plan to build shared libraries for Ubuntu by running commands
exactly like "go install -buildmode=shared prefix/...", this special cases this
to produce the old behaviour (but de-relativises prefix first).

Fixes #13714

Change-Id: I4fd8d4934279f9a18cc70a13e4ef3e23f6abcb6e
Reviewed-on: https://go-review.googlesource.com/18114Reviewed-by: Ian Lance Taylor <iant@golang.org>

parseBase128Int compares |shifted| with four, seemingly to ensure the result
fits in an int32 on 32-bit platforms where int is 32-bit. However, there is an
off-by-one in this logic, so it actually allows five shifts, making the maximum
tag number or OID component 2^35-1.

Fix this so the maximum is 2^28-1 which should be plenty for OID components and
tag numbers while not overflowing on 32-bit platforms.

Change-Id: If825b30cc53a0fc08e68ea1a24d265e7eb1a13a4
Reviewed-on: https://go-review.googlesource.com/18225Reviewed-by: Adam Langley <agl@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>

These are the easy, automated cases. There were some more where we
need to fight Gerrit and the CLA system to extract the appropriate
metadata.

Updates #12042

Change-Id: Id63ae635ee7efeec4cd372c7d85bb5b1f557951b
Reviewed-on: https://go-review.googlesource.com/18264Reviewed-by: Ian Lance Taylor <iant@golang.org>

Try to reduce feature request bug reports.

Change-Id: I713bb715d25d23e084b054aea8e1c3197dde90d4
Reviewed-on: https://go-review.googlesource.com/18222Reviewed-by: Ian Lance Taylor <iant@golang.org>

In the beginning, there was no way to cancel an HTTP request.

We later added Transport.CancelRequest to cancel an in-flight HTTP
request by breaking its underlying TCP connection, but it was hard to
use correctly and didn't work in all cases. And its error messages
were terrible. Some of those issues were fixed over time, but the most
unfixable problem was that it didn't compose well. All RoundTripper
implementations had to choose to whether to implement CancelRequest
and both decisions had negative consequences.

In Go 1.5 we added Request.Cancel, which composed well, worked in all
phases, had nice error messages, etc. But we forgot to use it in the
implementation of Client.Timeout (a timeout which spans multiple
requests and reading request bodies).

In Go 1.6 (upcoming), we added HTTP/2 support, but now Client.Timeout
didn't work because the http2.Transport didn't have a CancelRequest
method.

Rather than add a CancelRequest method to http2, officially deprecate
it and update the only caller (Client, for Client.Cancel) to use
Request.Cancel instead.

The http2 Client timeout tests are enabled now.

For compatibility, we still use CancelRequest in Client if we don't
recognize the RoundTripper type. But documentation has been updated to
tell people that CancelRequest is deprecated.

Fixes #13540

Change-Id: I15546b90825bb8b54905e17563eca55ea2642075
Reviewed-on: https://go-review.googlesource.com/18260Reviewed-by: Andrew Gerrand <adg@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>