The ProgInfo is loaded many times during each analysis pass.
Load it once at the beginning (in Flowstart if using that, or explicitly,
as in plive.go) and then refer to the cached copy.

Removes many calls to proginfo.

Makes Prog a little bigger, but the previous CL more than compensates.

Change-Id: If90a12fc6729878fdae10444f9c3bedc8d85026e
Reviewed-on: https://go-review.googlesource.com/7745Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>

An interface{} is more in the spirit of the original union.
By my calculations, on 64-bit systems this reduces
Addr from 120 to 80 bytes, and Prog from 592 to 424 bytes.

Change-Id: I0d7b0981513c2a3c94c9ac76bb4f8816485b5a3c
Reviewed-on: https://go-review.googlesource.com/7744Reviewed-by: Rob Pike <r@golang.org>

We're skating on thin ice, and things are finally starting to melt around here.
(I want to avoid the debugging session that will happen when someone
uses atomicand8 expecting it to be atomic with respect to other operations.)

Change-Id: I254f1582be4eb1f2d7fbba05335a91c6bf0c7f02
Reviewed-on: https://go-review.googlesource.com/7861Reviewed-by: Minux Ma <minux@golang.org>

Change-Id: I847bf32bd0be913fad277c5e657f44df147eee14
Reviewed-on: https://go-review.googlesource.com/7729Reviewed-by: Rob Pike <r@golang.org>
Reviewed-by: Minux Ma <minux@golang.org>

Change-Id: I99aee6dff97a4abcaf5a9cddb505ba90b65667ea
Reviewed-on: https://go-review.googlesource.com/7728Reviewed-by: Rob Pike <r@golang.org>

Fixes #9825.

Change-Id: Id7eeaa14c26201db34db0820371c92a63af485b0
Reviewed-on: https://go-review.googlesource.com/7604Reviewed-by: Rob Pike <r@golang.org>

Fixes #10171.

Change-Id: I1b2e30ebbb2b9d66680008674baa96e550efe1f2
Reviewed-on: https://go-review.googlesource.com/7603Reviewed-by: Adam Langley <agl@golang.org>
Run-TryBot: Adam Langley <agl@golang.org>

I think the file ended up in the order of the typedefs instead of the
order of the actual struct definitions. You can see where some of
the declarations were because some of the comments didn't move.
Put things back in the original order.

Change-Id: I0e3703008278b084b632c917cfb73bc81bdd4f23
Reviewed-on: https://go-review.googlesource.com/7743Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>

This allows gins to let Naddr fill in p.From and p.To directly,
avoiding the zeroing and copying of a temporary.

Change-Id: I96d120afe266e68f94d5e82b00886bf6bd458f85
Reviewed-on: https://go-review.googlesource.com/7742Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>

This way the error messages will show the original file name
in addition to the bootstrap file name, so that you have some
chance of making the correction in the original instead of the copy
(which will be blown away).

Before:
/Users/rsc/g/go/pkg/bootstrap/src/bootstrap/5g/gsubr.go:863: undefined: a

After:
/Users/rsc/g/go/src/cmd/5g/gsubr.go:860[/Users/rsc/g/go/pkg/bootstrap/src/bootstrap/5g/gsubr.go:863]: undefined: a

Change-Id: I8d6006abd9499edb16d9f27fe8b7dc6cae143fca
Reviewed-on: https://go-review.googlesource.com/7741Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

To reduce lock contention in this mode, makes persistent allocation state per-P,
which means at most 64 kB overhead x $GOMAXPROCS, which should be
completely tolerable.

Change-Id: I34ca95e77d7e67130e30822e5a4aff6772b1a1c5
Reviewed-on: https://go-review.googlesource.com/7740Reviewed-by: Rick Hudson <rlh@golang.org>

This reverts commit 42fcc6fe.

Change-Id: If860b7cbff5b5d288c1df1405c1765275dfba7cb
Reviewed-on: https://go-review.googlesource.com/7860Reviewed-by: Russ Cox <rsc@golang.org>

This is a follow-up to review comments on CL 7696.

I believe that this includes the first regular Go test in the compiler.

No functional changes. Passes toolstash -cmp.

Change-Id: Id45f51aa664c5d52ece2a61cd7d8417159ce3cf0
Reviewed-on: https://go-review.googlesource.com/7820Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

The body tag in the pprof template was misplaced.

Change-Id: Icd7948b358f52df1acc7e033ab27a062990ef977
Reviewed-on: https://go-review.googlesource.com/7795Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Accidentally turned on in golang.org/cl/7734.

Change-Id: I8d72c279150a0b93732a2ac41b82fbb3cd7bf9d3
Reviewed-on: https://go-review.googlesource.com/7737Reviewed-by: Burcu Dogan <jbd@google.com>

This CL updates a TODO on a condition excluding a lot of tests on
android, clarifying what needs to be done. Several of the tests should
be turned off, for example anything depending on the Go tool, others
should be enabled. (See #8345, comment 3 for more details.)

Also add iOS, which has the same set of restrictions.

Tested manually on linux/amd64, darwin/amd64, android/arm, darwin/arm.

Updates #8345

Change-Id: I147f0a915426e0e0de9a73f9aea353766156609b
Reviewed-on: https://go-review.googlesource.com/7734Reviewed-by: Burcu Dogan <jbd@google.com>

Change-Id: Ifa71fb443a66eb8d7732f3b0c1408947b583c1f1
Reviewed-on: https://go-review.googlesource.com/7800Reviewed-by: Ian Lance Taylor <iant@golang.org>

CL 7697 caused doasm failures on 386:

runtime/append_test.go:1: doasm: notfound ft=2 tt=20 00112 (runtime/iface_test.go:207)	CMPL	$0, BX 2 20

I think that this should be fixed in liblink,
but in the meantime, work around the problem
by instead generating CMPL BX, $0.

Change-Id: I9c572f8f15fc159507132cf4ace8d7a328a3eb4a
Reviewed-on: https://go-review.googlesource.com/7810Reviewed-by: Keith Randall <khr@golang.org>

Some type assertions of the form _, ok := i.(T) allow efficient inlining.
Such type assertions commonly show up in type switches.
For example, with this optimization, using 6g, the length of
encoding/binary's intDataSize function shrinks from 2224 to 1728 bytes (-22%).

benchmark                    old ns/op     new ns/op     delta
BenchmarkAssertI2E2Blank     4.67          0.82          -82.44%
BenchmarkAssertE2T2Blank     4.38          0.83          -81.05%
BenchmarkAssertE2E2Blank     3.88          0.83          -78.61%
BenchmarkAssertE2E2          14.2          14.4          +1.41%
BenchmarkAssertE2T2          10.3          10.4          +0.97%
BenchmarkAssertI2E2          13.4          13.3          -0.75%

Change-Id: Ie9798c3e85432bb8e0f2c723afc376e233639df7
Reviewed-on: https://go-review.googlesource.com/7697Reviewed-by: Keith Randall <khr@golang.org>

This is preliminary cleanup for another change.

No functional changes. Passes toolstash -cmp.

Change-Id: I11d562fbd6cba5c48d9636f3149e210e5f5308ad
Reviewed-on: https://go-review.googlesource.com/7696Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: Minux Ma <minux@golang.org>

The distinction between gcWorkProducer and gcWork (producer and
consumer) is not serving us as originally intended, so merge these
into just gcWork.

The original intent was to replace the currentwbuf cache with a
gcWorkProducer. However, with gchelpwork (aka mutator assists),
mutators can both produce and consume work, so it will make more sense
to cache a whole gcWork.

Change-Id: I6e633e96db7cb23a64fbadbfc4607e3ad32bcfb3
Reviewed-on: https://go-review.googlesource.com/7733Reviewed-by: Rick Hudson <rlh@golang.org>

Currently markroot fetches the wbuf to fill from the per-M wbuf
cache. The wbuf cache is primarily meant for the write barrier because
it produces very little work on each call. There's little point to
using the cache in mark root, since each call to markroot is likely to
produce a large amount of work (so the slight win on getting it from
the cache instead of from the central wbuf lists doesn't matter), and
markroot does not dispose the wbuf back to the cache (so most markroot
calls won't get anything from the wbuf cache anyway).

Instead, just get the wbuf from the central wbuf lists like other work
producers. This will simplify later changes.

Change-Id: I07a18a4335a41e266a6d70aa3a0911a40babce23
Reviewed-on: https://go-review.googlesource.com/7732Reviewed-by: Rick Hudson <rlh@golang.org>

Currently, the GC's concurrent mark phase runs on the system
stack. There's no need to do this, and running it this way ties up the
entire M and P running the GC by preventing the scheduler from
preempting the GC even during concurrent mark.

Fix this by running concurrent mark on the regular G stack. It's still
non-preemptible because we also set preemptoff around the whole GC
process, but this moves us closer to making it preemptible.

Change-Id: Ia9f1245e299b8c5c513a4b1e3ef13eaa35ac5e73
Reviewed-on: https://go-review.googlesource.com/7730Reviewed-by: Rick Hudson <rlh@golang.org>

"Sync" is not very informative. What's being synchronized and with
whom? Update this comment to explain what we're really doing: enabling
write barriers.

Change-Id: I4f0cbb8771988c7ba4606d566b77c26c64165f0f
Reviewed-on: https://go-review.googlesource.com/7700Reviewed-by: Rick Hudson <rlh@golang.org>

Currently we harvestwbufs the moment we enter the mark phase, even
before starting the world again. Since cached wbufs are only filled
when we're in mark or mark termination, they should all be empty at
this point, making the harvest pointless. Remove the harvest.

We should, but do not currently harvest at the end of the mark phase
when we're running out of work to do.

Change-Id: I5f4ba874f14dd915b8dfbc4ee5bb526eecc2c0b4
Reviewed-on: https://go-review.googlesource.com/7669Reviewed-by: Rick Hudson <rlh@golang.org>

Change-Id: I0ad1a81a235c7c067fea2093bbeac4e06a233c10
Reviewed-on: https://go-review.googlesource.com/7661Reviewed-by: Rick Hudson <rlh@golang.org>

Change-Id: I5a49f56518adf7d64ba8610b51ea1621ad888fc4
Reviewed-on: https://go-review.googlesource.com/7771Reviewed-by: Ian Lance Taylor <iant@golang.org>

Switch statements do a binary search on long runs of constants.
Doing a less-than comparison on a string is much more expensive
than on (say) an int. Use two part comparison for strings:
First compare length, then the strings themselves.

Benchmarks from issue 10000:

benchmark                  old ns/op     new ns/op     delta
BenchmarkIf0               3.36          3.35          -0.30%
BenchmarkIf1               4.45          4.47          +0.45%
BenchmarkIf2               5.22          5.26          +0.77%
BenchmarkIf3               5.56          5.58          +0.36%
BenchmarkIf4               10.5          10.6          +0.95%
BenchmarkIfNewStr0         5.26          5.30          +0.76%
BenchmarkIfNewStr1         7.19          7.15          -0.56%
BenchmarkIfNewStr2         7.23          7.16          -0.97%
BenchmarkIfNewStr3         7.47          7.43          -0.54%
BenchmarkIfNewStr4         12.4          12.2          -1.61%
BenchmarkSwitch0           9.56          4.24          -55.65%
BenchmarkSwitch1           8.64          5.58          -35.42%
BenchmarkSwitch2           9.38          10.1          +7.68%
BenchmarkSwitch3           8.66          5.00          -42.26%
BenchmarkSwitch4           7.99          8.18          +2.38%
BenchmarkSwitchNewStr0     11.3          6.12          -45.84%
BenchmarkSwitchNewStr1     11.1          8.33          -24.95%
BenchmarkSwitchNewStr2     11.0          11.1          +0.91%
BenchmarkSwitchNewStr3     10.3          6.93          -32.72%
BenchmarkSwitchNewStr4     11.0          11.2          +1.82%

Fixes #10000

Change-Id: Ia2fffc32e9843425374c274064f709ec7ee46d80
Reviewed-on: https://go-review.googlesource.com/7698Reviewed-by: Keith Randall <khr@golang.org>

Change-Id: I79b7ed8f7e78e9d35b5e30ef70b98db64bc68a7b
Reviewed-on: https://go-review.googlesource.com/7720Reviewed-by: Ian Lance Taylor <iant@golang.org>

Comment changes only.

Change-Id: I56848814564c4aa0988b451df18bebdfc88d6d94
Reviewed-on: https://go-review.googlesource.com/7721Reviewed-by: Rob Pike <r@golang.org>

One of my earlier versions of finer-grained select locking
failed on this test. If you just naively lock and check channels
one-by-one, it is possible that you skip over ready channels.
Consider that initially c1 is ready and c2 is not. Select checks c2.
Then another goroutine makes c1 not ready and c2 ready (in that order).
Then select checks c1, concludes that no channels are ready and
executes the default case. But there was no point in time when
no channel is ready and so default case must not be executed.

Change-Id: I3594bf1f36cfb120be65e2474794f0562aebcbbd
Reviewed-on: https://go-review.googlesource.com/7550Reviewed-by: Russ Cox <rsc@golang.org>

Change-Id: I216511a4bce431de0a468f618a7a7c4da79e2979
Reviewed-on: https://go-review.googlesource.com/7710Reviewed-by: Ian Lance Taylor <iant@golang.org>

RC4 is frowned upon[1] at this point and major providers are disabling it
by default[2].

Those who still need RC4 support in crypto/tls can enable it by
specifying the CipherSuites slice in crypto/tls.Config explicitly.

Fixes #10094.

[1] https://tools.ietf.org/html/rfc7465
[2] https://blog.cloudflare.com/killing-rc4-the-long-goodbye/

Change-Id: Ia03a456f7e7a4362b706392b0e3c4cc93ce06f9f
Reviewed-on: https://go-review.googlesource.com/7647Reviewed-by: Andrew Gerrand <adg@golang.org>

Just so that we notice in the future if another hash function is added
without updating this utility function, make it panic when passed an
unknown handshake hash function. (Which should never happen.)

Change-Id: I60a6fc01669441523d8c44e8fbe7ed435e7f04c8
Reviewed-on: https://go-review.googlesource.com/7646Reviewed-by: Andrew Gerrand <adg@golang.org>
Reviewed-by: Joël Stemmer <stemmertech@gmail.com>

crypto/rand.Reader doesn't ensure that short reads don't happen. This
change contains a couple of fixups where io.ReadFull wasn't being used
with it.

Change-Id: I3855b81f5890f2e703112eeea804aeba07b6a6b8
Reviewed-on: https://go-review.googlesource.com/7645Reviewed-by: Minux Ma <minux@golang.org>
Reviewed-by: Andrew Gerrand <adg@golang.org>

For example, "GOARCH=sparc go build -compiler=gccgo" should not crash
merely because the architecture character for sparc is not known.

Change-Id: I18912c7f5d90ef8f586592235ec9d6e5053e4bef
Reviewed-on: https://go-review.googlesource.com/7695Reviewed-by: Russ Cox <rsc@golang.org>

Change-Id: I72e8389ec080be8a0119f98df898de6f5510fa4d
Reviewed-on: https://go-review.googlesource.com/7693Reviewed-by: Alan Donovan <adonovan@google.com>

Change-Id: I19e6542e7d79d60e39d62339da51a827c5aa6d3b
Reviewed-on: https://go-review.googlesource.com/7668Reviewed-by: Russ Cox <rsc@golang.org>

The value in question is really a bit pattern
(a pointer with extra bits thrown in),
so treat it as a uintptr instead, avoiding the
generation of a write barrier when there
might not be a p.

Also add the obligatory //go:nowritebarrier.

Change-Id: I4ea097945dd7093a140f4740bcadca3ce7191971
Reviewed-on: https://go-review.googlesource.com/7667Reviewed-by: Rick Hudson <rlh@golang.org>
Reviewed-by: Austin Clements <austin@google.com>

The GC assumes that there will be no asynchronous write barriers when
the world is stopped. This keeps the synchronization between write
barriers and the GC simple. However, currently, there are a few places
in runtime code where this assumption does not hold.
The GC stops the world by collecting all Ps, which stops all user Go
code, but small parts of the runtime can run without a P. For example,
the code that releases a P must still deschedule its G onto a runnable
queue before stopping. Similarly, when a G returns from a long-running
syscall, it must run code to reacquire a P.
Currently, this code can contain write barriers. This can lead to the
GC collecting reachable objects if something like the following
sequence of events happens:
1. GC stops the world by collecting all Ps.
2. G #1 returns from a syscall (for example), tries to install a
pointer to object X, and calls greyobject on X.
3. greyobject on G #1 marks X, but does not yet add it to a write
buffer. At this point, X is effectively black, not grey, even though
it may point to white objects.
4. GC reaches X through some other path and calls greyobject on X, but
greyobject does nothing because X is already marked.
5. GC completes.
6. greyobject on G #1 adds X to a work buffer, but it's too late.
7. Objects that were reachable only through X are incorrectly collected.
To fix this, we check the invariant that no asynchronous write
barriers happen when the world is stopped by checking that write
barriers always have a P, and modify all currently known sources of
these writes to disable the write barrier. In all modified cases this
is safe because the object in question will always be reachable via
some other path.

Some of the trace code was turned off, in particular the
code that traces returning from a syscall. The GC assumes
that as far as the heap is concerned the thread is stopped
when it is in a syscall. Upon returning the trace code
must not do any heap writes for the same reasons discussed
above.

Fixes #10098
Fixes #9953
Fixes #9951
Fixes #9884

May relate to #9610 #9771

Change-Id: Ic2e70b7caffa053e56156838eb8d89503e3c0c8a
Reviewed-on: https://go-review.googlesource.com/7504Reviewed-by: Austin Clements <austin@google.com>