readyExecute passes a closure to mcall that captures an argument to
readyExecute. Since mcall is marked noescape, this closure lives on
the stack of the calling goroutine. However, the closure puts the
calling goroutine on the run queue (and switches to a new
goroutine). If the calling goroutine gets scheduled before the mcall
returns, this stack-allocated closure will become invalid while it's
still executing. One consequence of this we've observed is that the
captured gp variable can get overwritten before the call to
execute(gp), causing execute(gp) to segfault.

Fix this by passing the currently captured gp variable through a field
in the calling goroutine's g struct so that the func is no longer a
closure.

To prevent problems like this in the future, this change also removes
the go:noescape annotation from mcall. Due to a compiler bug, this
will currently cause a func closure passed to mcall to be implicitly
allocated rather than refusing the implicit allocation. However, this
is okay because there are no other closures passed to mcall right now
and the compiler bug will be fixed shortly.

Fixes #10428.

Change-Id: I49b48b85de5643323b89e9eaa4df63854e968c32
Reviewed-on: https://go-review.googlesource.com/8866
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: David Chase <drchase@google.com>
Reviewed-by: Russ Cox <rsc@golang.org>

Change-Id: I8029640fe74731d7847390f371af37b85dec96cc
Reviewed-on: https://go-review.googlesource.com/9006Reviewed-by: Rob Pike <r@golang.org>

Trivial to do, but overlooked for 1.4, which is good because I prefer
the new design, which is just to match against the source code of
the line rather than the command word alone.

Change-Id: Idcf7c4479e97bb7cd732f0d058012321b6057628
Reviewed-on: https://go-review.googlesource.com/9005Reviewed-by: Russ Cox <rsc@golang.org>

Change-Id: Id1ecad2565edd5937a764dcf56b89263353f47c2
Reviewed-on: https://go-review.googlesource.com/8947Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>

Fixes #10319

Change-Id: I96015b0e1dff30a72de11fea3837638b5c672891
Reviewed-on: https://go-review.googlesource.com/8501Reviewed-by: Russ Cox <rsc@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>

Followed the same test pattern in misc/cgo/testcarchive.

Change-Id: I2f863b5c24a28f0b38b0128ed3e8a92c17fb5b9f
Reviewed-on: https://go-review.googlesource.com/8985Reviewed-by: David Crawshaw <crawshaw@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

Change-Id: I0d2b909ceeba9da86c77de4ce23d7c6063f8f10f
Reviewed-on: https://go-review.googlesource.com/9017Reviewed-by: Minux Ma <minux@golang.org>

Updates #10476

Change-Id: Ic4414f669104905c6004835be5cf0fa873553ea6
Reviewed-on: https://go-review.googlesource.com/8962Reviewed-by: Russ Cox <rsc@golang.org>

Repeat of CL 8951.

Change-Id: I5430e4a9eb5d8b7d0e3963657092bede67439056
Reviewed-on: https://go-review.googlesource.com/9003Reviewed-by: Rob Pike <r@golang.org>

Previously we started the Go runtime from a JNI function call, which
eventually called the program's main function. Now the runtime is
initialized by an ELF initialization function as a c-shared library,
and the program's main function is not called. So now we export main
so it can be called from JNI.

This is necessary for all-Go apps because unlike a normal shared
library, the program loading the library is not written by or known
to the programmer. As far as they are concerned, the .so is
everything. In fact the same code is compiled for iOS as a normal Go
program.

Change-Id: I61c6a92243240ed229342362231b1bfc7ca526ba
Reviewed-on: https://go-review.googlesource.com/9015Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>

Change-Id: Ie7f85873978adf3fd5c739176f501ca219592824
Reviewed-on: https://go-review.googlesource.com/9011Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

Instead of only checking packages that are already listed in pkgDeps,
apply deps checks to all standard library packages.

To avoid slowing testing down too much, instead of running "go list
std" in a subprocess like cmd/api or cmd/dist, this test manually
walks the GOROOT src directory to enumerate packages.

Timings on an HP Z620 using linux/amd64:

    	short	full
before	0.092s	4.880s
after	0.137s	5.104s

Additionally, a handful of packages that were previously unchecked are
now listed, along with their current dependencies.  These should
probably eventually be moved elsewhere and assigned appropriate
allowable-dependency sets.  For now, they've been grandfathered in by
simply assigning them their current dependencies, so that followup CLs
can review them individually as appropriate.

Fixes #10475.

Change-Id: I83ffd8ff329092f664bf3e3f2c9e3dad8e77ac02
Reviewed-on: https://go-review.googlesource.com/9001
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Russ Cox <rsc@golang.org>

Change-Id: I223a4bd6e3ee31324b46ac79a4022e40f1868491
Reviewed-on: https://go-review.googlesource.com/8995Reviewed-by: Matthew Dempsky <mdempsky@google.com>

This cleanup is in anticipation of implementing
jump-free booleans (CL 2284) and zero-aware
comparisons (issue 10381).

No functional changes. Passes toolstash -cmp.

Change-Id: I50f394c60fa2927e177d7fc85b75085060a9e912
Reviewed-on: https://go-review.googlesource.com/8738Reviewed-by: Russ Cox <rsc@golang.org>

This memory is untyped and can't be used anymore.
The next version of SWIG won't need it.

Change-Id: I592b287c5f5186975ee09a9b28d8efe3b57134e7
Reviewed-on: https://go-review.googlesource.com/8956Reviewed-by: Ian Lance Taylor <iant@golang.org>

This change makes TestDialError, TestListenError work without any
external dependency, enables them by default, and removes unnecessary
-run_error_test flag for fixing #4856.

Also fixes inconsistent error values on Dial, Listen partially as a
first stab.

Updates #4856.

Change-Id: Ie10c151ae06759085f352c7db2ca45107a81914f
Reviewed-on: https://go-review.googlesource.com/8903Reviewed-by: Ian Lance Taylor <iant@golang.org>

In followup changes, we'll move OpError around from the netFD layer to
the Conn layer for fixing #4856. Before doing that, this change makes
netFD of Plan 9 match netFD for POSIX platforms to avoid conflict.

Change-Id: Iea7632716d48722a1758e52effefec964a3a9442
Reviewed-on: https://go-review.googlesource.com/8990Reviewed-by: David du Colombier <0intro@gmail.com>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

red, green and blue.

Fixes #9902

Change-Id: Ibffd0aa2f98996170e39a919296f69e9d5c71545
Reviewed-on: https://go-review.googlesource.com/8907Reviewed-by: Rob Pike <r@golang.org>

Change-Id: I3b3f80791a1db4c2b7318f81a115972cd2237f08
Signed-off-by: Shenghou Ma <minux@golang.org>
Reviewed-on: https://go-review.googlesource.com/8787Reviewed-by: David Crawshaw <crawshaw@golang.org>

I said I removed this from cl/8711 in response to your comment, but
apparently I did not.

misc/cgo/testcarchive continues to pass on darwin/amd64.

Change-Id: I6410782f2a78bf117741628fb71cac56e289b590
Reviewed-on: https://go-review.googlesource.com/9010Reviewed-by: Ian Lance Taylor <iant@golang.org>

For some reason the absense of an implementation does not stop arm64
binaries being built. However it comes up with -buildmode=c-archive.

Change-Id: Ic0db5fd8fb4fe8252b5aa320818df0c7aec3db8f
Reviewed-on: https://go-review.googlesource.com/8989Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Change-Id: I749fd91cd3c7581cdcc97a15e8eeee0c20f0b259
Reviewed-on: https://go-review.googlesource.com/8805Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>

Change-Id: Ib227aa3e14d01a0ab1ad9e53d107858e045d1c42
Reviewed-on: https://go-review.googlesource.com/8984Reviewed-by: Ian Lance Taylor <iant@golang.org>

Change-Id: I8b2063e65d3454a694a789c1682dacfe0bea2e19
Reviewed-on: https://go-review.googlesource.com/8965Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

- Multiple GetCertificate tests shared the same name and were
  overwriting each other, each test now has a unique name.
- expectAlert was not implemented in the data updater, the single
  test that used it has been replaced with a ClientHello failure
  test.

Fixes #10470

Change-Id: I500738f6302ffa863d7ee45d85fa8773155e0614
Reviewed-on: https://go-review.googlesource.com/8959Reviewed-by: Adam Langley <agl@golang.org>
Run-TryBot: Adam Langley <agl@golang.org>

Change-Id: I8e912ff9327a4163b63b8c628aa3546e86ddcc02
Reviewed-on: https://go-review.googlesource.com/8983Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>

This reverts commit 3e6b7f35.

Change-Id: Icad2b235b9880729fbdf80d7dfd71e56df4b1231
Reviewed-on: https://go-review.googlesource.com/8944Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Change-Id: I3b3f80791a1db4c2b7318f81a115972cd2237f06
Signed-off-by: Shenghou Ma <minux@golang.org>
Reviewed-on: https://go-review.googlesource.com/8785Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Fixes #7075

Change-Id: I57fcebaaef768b00263b6b91211b1434f6a4b229
Reviewed-on: https://go-review.googlesource.com/8919Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>

Change-Id: I86530a4fd240f3e056e3277355d2965f6863b852
Reviewed-on: https://go-review.googlesource.com/8951Reviewed-by: Rob Pike <r@golang.org>

It is not needed right now, but it will allow more flexibility in
the future.

Fixes #10472

Change-Id: I2eaea70abeca5ed10f89b0b2dfdabdac376a0a41
Reviewed-on: https://go-review.googlesource.com/8964Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

The ptrto field of the type data cannot be relied on when dynamic linking: a
type T may be defined in a module that makes no use of pointers to that type,
but another module can contain a package that imports the first one and does use
*T pointers.  The second module will end up defining type data for *T and a
type.*T symbol pointing at it. It's important that calling .PtrTo() on the
refect.Type for T returns this type data and not some synthesized object, so we
need reflect to be able to find it!

Fortunately, the reflect package already has a mechanism for doing this sort of
thing: ChanOf/MapOf/etc look for pre-existing type data by name.  So this change
just extends PtrTo() to consult this too, and changes the compiler to include
pointer types in the data consulted when compiling for dynamic linking.

Change-Id: I3773c066fd0679a62e9fc52a84bf64f1d67662b7
Reviewed-on: https://go-review.googlesource.com/8232Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>

This is a follow-up to CL 8910.
This is the version that I have tested and which works
when appID and teamID are not the same (which they appear
to be for the builder).

I am unsure how I submitted it with the wrong code.

Change-Id: I186e34e91953d082b507390c1cd2042e5419c4c9
Reviewed-on: https://go-review.googlesource.com/8943Reviewed-by: David Crawshaw <crawshaw@golang.org>

7g and 9g disagree with componentgen
about what type len and cap have.

This results in an etype mismatch,
which inhibits registerization.

Fixing this results in 7406 more registerizations
while building the stdlib.
There are still 1512 missed opportunities.

This should improve the performance benefit
to 7g of enabling componentgen (CL 8636).

This CL reduces the size of godoc by 203k (-1.177%).

This was discovered by using the diagnostics
added in CL 8732 and running:

GOARCH=arm64 GOOS=linux go build -gcflags="-d registerization" std

See CL 91850043 for similar earlier fixes for 6g and 8g.

Change-Id: I57f478228a000ad7529d4136bad94a51343c4daa
Reviewed-on: https://go-review.googlesource.com/8733Reviewed-by: Dave Cheney <dave@cheney.net>
Reviewed-by: Keith Randall <khr@golang.org>

We can use CBNZ instruction and make it one instruction shorter.

Saves 66kB in godoc.

Change-Id: Ie71fe7cf31e7f73644ee926f4f9624c009c3eb1a
Reviewed-on: https://go-review.googlesource.com/8634Reviewed-by: Minux Ma <minux@golang.org>

Optimize the sequence:
	MOV $imm, Rt
	ADD Rt, Rs, Rd

into:
	ADD $imm, Rs, Rd

Saves 66k in godoc.

Change-Id: I27b4aaa0ec80a59472fe2e5816efdf3db9c901ee
Reviewed-on: https://go-review.googlesource.com/8632Reviewed-by: Minux Ma <minux@golang.org>

Change-Id: I3b3f80791a1db4c2b7318f81a115972cd2237f03
Signed-off-by: Shenghou Ma <minux@golang.org>
Reviewed-on: https://go-review.googlesource.com/8782Reviewed-by: David Crawshaw <crawshaw@golang.org>

Because there is no libgcc.

Change-Id: I3b3f80791a1db4c2b7318f81a115972cd2237f07
Signed-off-by: Shenghou Ma <minux@golang.org>
Reviewed-on: https://go-review.googlesource.com/8786Reviewed-by: David Crawshaw <crawshaw@golang.org>

Change-Id: I402629b154ae36c879a8d5088cb0f71b075f31dc
Reviewed-on: https://go-review.googlesource.com/8958Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>

Fixes #10116.

Change-Id: I3b3f80791a1db4c2b7318f81a115972cd2237f05
Signed-off-by: Shenghou Ma <minux@golang.org>
Reviewed-on: https://go-review.googlesource.com/8784Reviewed-by: David Crawshaw <crawshaw@golang.org>