Libc on Darwin/ARM has a buggy implementation of ldexp and frexp
that could not handle denormal numbers.

Also disable VFP runfast (flush-to-zero) mode so that the gc
compiler can correctly handle denormal constants used in math and
strconv tests.

Change-Id: Ie64220b882f414e0b37f406f38181c3586104d46
Reviewed-on: https://go-review.googlesource.com/2119Reviewed-by: David Crawshaw <crawshaw@golang.org>
Reviewed-by: Dave Cheney <dave@cheney.net>

In the old code, liblink, cmd/ld and runtime all have code determine
whether runtime.tlsg is an actual variable or a placeholder for TLS
relocation. This change consolidate them into one: the runtime/tls_arm.s
will ultimately determine the type of that variable.

Change-Id: I3b3f80791a1db4c2b7318f81a115972cd2237e43
Reviewed-on: https://go-review.googlesource.com/2118Reviewed-by: David Crawshaw <crawshaw@golang.org>
Reviewed-by: Dave Cheney <dave@cheney.net>

Fixes #9742.

Change-Id: Ifedf7ff9465bc49534b708d414c8e435ee9ce6cd
Reviewed-on: https://go-review.googlesource.com/3970Reviewed-by: Ian Lance Taylor <iant@golang.org>

The main issue is that the misc/cgo/{stdio,life} tests are silently
getting skipped when invoked from run.bash.

run.go should ignore any build tags after the first blank line in
source file. It already checks for test actions only upto the first
blank line. Build tags must be specified in the same block.

See http://golang.org/cl/3675 for background.

Change-Id: Id8abf000119e3335f7250d8ef34aac7811fc9dff
Reviewed-on: https://go-review.googlesource.com/3812Reviewed-by: Minux Ma <minux@golang.org>

issue9355 generated a file a.[568] in test/ directory and left it there.
For tests like these, it is best to chdir to a test specific directory
before generating any temporary files, since the tests are running
in parallel and might otherwise race with each other for the same files.

Change-Id: I58d96256d4d8ee3fda70d81077f19006064a7425
Reviewed-on: https://go-review.googlesource.com/3813Reviewed-by: Ian Lance Taylor <iant@golang.org>

PalettedColorModel was renamed to color.Palette
over three years ago by https://golang.org/cl/5132048.

Change-Id: I0204ade10eabff45620fda2990fed428c65d871e
Reviewed-on: https://go-review.googlesource.com/3305Reviewed-by: Minux Ma <minux@golang.org>
Reviewed-by: Nigel Tao <nigeltao@golang.org>

and empty keys. Also reject malformed (quoted) values.

See also https://go-review.googlesource.com/3952

Change-Id: Ice6de33b09f9904b28e410a680a90aa6c8c76fed
Reviewed-on: https://go-review.googlesource.com/3953Reviewed-by: Rob Pike <r@golang.org>

Change-Id: I7514e643a6a81ca715adbf6f8d19d3b2dd43fe6c
Reviewed-on: https://go-review.googlesource.com/3810Reviewed-by: Alan Donovan <adonovan@google.com>

Change-Id: Ica419a1215ca33dc1cff1e9e4137f204591e3cee
Reviewed-on: https://go-review.googlesource.com/3942Reviewed-by: Alan Donovan <adonovan@google.com>

Change-Id: Ia30886569141ca2e0321bea6ee1d5c9e0f79d6f9
Reviewed-on: https://go-review.googlesource.com/3941Reviewed-by: Alan Donovan <adonovan@google.com>

Change-Id: I4329c44b829fcd77e4f1a1d45904f0f8a280a595
Reviewed-on: https://go-review.googlesource.com/3940Reviewed-by: Alan Donovan <adonovan@google.com>

Change-Id: Ida064ff422d077d508ef94f52fc813daa6277a99
Reviewed-on: https://go-review.googlesource.com/3933Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

This was working when 5a was built on x86 because REG_R0 = 32,
and a 32-bit shift on x86 uses only the low 32 bits of the shift count.

On ARM, the shift clamping is different.

Moving to Go will avoid these differing shift semantics.
I tripped over and fixed this bug in new5a the same way earlier tonight.

Change-Id: Id56aa0bb1830ccf250960f843e0acb8a0409e87d
Reviewed-on: https://go-review.googlesource.com/3961Reviewed-by: Dave Cheney <dave@cheney.net>

Change-Id: I6ee7aa76673e51576b5a84c512b22a0f8af5b02f
Reviewed-on: https://go-review.googlesource.com/3966Reviewed-by: Dmitry Vyukov <dvyukov@google.com>

These benchmarks are only for functions commonly used in loops. The
other functions are typically used for inspection or setup and thus are
not performance sensitive.

Change-Id: I8d0a0ba2d8234ecacb40fa3aa9077bf93c8fe89c
Reviewed-on: https://go-review.googlesource.com/3680Reviewed-by: Dmitry Vyukov <dvyukov@google.com>

This reverts commit da4abda2.

Change-Id: Ifd09b3dae0af0c7cef5fbbc332c63e78dc90d6b1
Reviewed-on: https://go-review.googlesource.com/3960Reviewed-by: Russ Cox <rsc@golang.org>

Fixes #9756.

Change-Id: If4ee6fe10f8f90294ff9c5e7480371494094b111
Signed-off-by: Shenghou Ma <minux@golang.org>
Reviewed-on: https://go-review.googlesource.com/3740Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Alex Brainman <alex.brainman@gmail.com>

Fixes #9759.

Change-Id: I263f1251b9401371231374551c4f71c70cb6e359
Signed-off-by: Shenghou Ma <minux@golang.org>
Reviewed-on: https://go-review.googlesource.com/3931Reviewed-by: Dave Cheney <dave@cheney.net>

Followup to https://go-review.googlesource.com/3910

We only need 1000 iteratinons.

Change-Id: Ib63ae53105176abec77bad9609d638aeda7bcd61
Reviewed-on: https://go-review.googlesource.com/3901Reviewed-by: Minux Ma <minux@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

Change-Id: I10efa3bc8bc7f41100feabe17837f805a42d7eb6
Reviewed-on: https://go-review.googlesource.com/3842Reviewed-by: Alan Donovan <adonovan@google.com>

(For zero values the strconv %b format prints the bias-adjusted exponent;
there's no bias in Float.)

Change-Id: I6f4dda9c3a50d02eac375cfe2c927c1540aae865
Reviewed-on: https://go-review.googlesource.com/3841Reviewed-by: Alan Donovan <adonovan@google.com>

Change-Id: Id508ca2f6c087861e8c6bc536bc39e54dce09825
Reviewed-on: https://go-review.googlesource.com/3840Reviewed-by: Alan Donovan <adonovan@google.com>

In android-L, logging is done through the logd daemon.
If logd daemon is available, send logging to logd.
Otherwise, fallback to the legacy mechanism (/dev/log files).

This change adds access/socket/connect calls to interact with the logd.

Fixes golang/go#9398.

Change-Id: I3c52b81b451f5862107d7c675f799fc85548486d
Reviewed-on: https://go-review.googlesource.com/3350Reviewed-by: David Crawshaw <crawshaw@golang.org>

Change-Id: I00f18a8c62a8d63822d7e07fee1ac5f088cc9587
Reviewed-on: https://go-review.googlesource.com/3866Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>

This reverts commit 80590711.

Conflicts:
	src/runtime/proc1.go (resolved by hand)

Change-Id: Ic62e3e802f0007ff9012b37bc5fd41fcafc153d0
Reviewed-on: https://go-review.googlesource.com/3885Reviewed-by: Dmitry Vyukov <dvyukov@google.com>

Add Addr-checking for all Progs on input to liblink, in liblink/pass.c,
including requiring use of TYPE_ADDR, not TYPE_CONST.
Update compilers and assemblers to satisfy checks.

Change-Id: Idac36b9f6805f0451cb541d2338992ca5eaf3963
Reviewed-on: https://go-review.googlesource.com/3801Reviewed-by: Austin Clements <austin@google.com>

Generating array types like [4]int would fail even though the int type
is generatable. Allow generating values of array types when the inner
type is generatable.

Change-Id: I7d71b3c18edb3737e2fec1ddf5e36c9dc8401971
Reviewed-on: https://go-review.googlesource.com/3865Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>

cc is no more.

Change-Id: I8d1bc0d2e471cd9357274204c9bc1fa67cbc272d
Reviewed-on: https://go-review.googlesource.com/3833Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Andrew Gerrand <adg@golang.org>

Change-Id: I274c438ab168278dfbb34ada3ed9a7f7c66b66ef
Reviewed-on: https://go-review.googlesource.com/3834Reviewed-by: Minux Ma <minux@golang.org>

We don't need placeholders for the old built-in poll server any more.

Change-Id: I3a510aec6a30bc2ac97676c400177cdfe557b8dc
Reviewed-on: https://go-review.googlesource.com/3863Reviewed-by: Dmitry Vyukov <dvyukov@google.com>

Change-Id: I3e280ca7d922f6ab14b2477361327ed076a95779
Reviewed-on: https://go-review.googlesource.com/3743Reviewed-by: Keith Randall <khr@golang.org>

The issue #8432 has been marked as an issue for golang.org/x/net.

Change-Id: Ia39abd99b685c820ea6169ee6505b16028e7e77f
Reviewed-on: https://go-review.googlesource.com/3836Reviewed-by: Ian Lance Taylor <iant@golang.org>

The unbounded list-based defer pool can grow infinitely.
This can happen if a goroutine routinely allocates a defer;
then blocks on one P; and then unblocked, scheduled and
frees the defer on another P.
The scenario was reported on golang-nuts list.

We've been here several times. Any unbounded local caches
are bad and grow to infinite size. This change introduces
central defer pool; local pools become fixed-size
with the only purpose of amortizing accesses to the
central pool.

Change-Id: Iadcfb113ccecf912e1b64afc07926f0de9de2248
Reviewed-on: https://go-review.googlesource.com/3741Reviewed-by: Keith Randall <khr@golang.org>

Using benchmark from the issue:

benchmark                    old ns/op     new ns/op     delta
BenchmarkRangeStringCast     2162          1152          -46.72%

benchmark                    old allocs     new allocs     delta
BenchmarkRangeStringCast     1              0              -100.00%

Fixes #2204

Change-Id: I92c5edd2adca4a7b6fba00713a581bf49dc59afe
Reviewed-on: https://go-review.googlesource.com/3790Reviewed-by: Keith Randall <khr@golang.org>

Before 3c0fee10, runtime.gogo was just long enough to align to 64 bytes
on OSs with short get_tls implementations and 80 bytes on OSs with
longer get_tls implementations (Windows, Solaris, and Plan 9).
3c0fee10 added a few instructions, which pushed it to 80 on most OSs,
including Windows and Plan 9, and 96 on Solaris.

Fixes #9770.

Change-Id: Ie84810657c14ab16dce9f0e0a932955251b0bf33
Reviewed-on: https://go-review.googlesource.com/3850Reviewed-by: Alex Brainman <alex.brainman@gmail.com>

golang.org/cl/144110044 made _ consts treated
as exported as a small, safe fix for #5397.
It also introduced issue #9615.

golang.org/cl/2091 then fixed the underlying issue,
which was missing type information when the type
was specified only for _.

This cl reverts the original fix.

Fixes #9615.

Change-Id: I4815ad8292bb5bec18beb8c131b48949d9af8876
Reviewed-on: https://go-review.googlesource.com/3832Reviewed-by: Robert Griesemer <gri@golang.org>

Generalizes PRF calculation for TLS 1.2 to support arbitrary hashes (SHA-384 instead of SHA-256).
Testdata were all updated to correspond with the new cipher suites in the handshake.

Change-Id: I3d9fc48c19d1043899e38255a53c80dc952ee08f
Reviewed-on: https://go-review.googlesource.com/3265Reviewed-by: Adam Langley <agl@golang.org>

Follow-up to cl 2682.

Change-Id: I06a589e782af821e4fa46a7c34d9a549de47fa60
Reviewed-on: https://go-review.googlesource.com/3830Reviewed-by: Dave Cheney <dave@cheney.net>

Change-Id: Ic8a01d06ffc6c80beaf6dee9e7242dcbb7425e71
Reviewed-on: https://go-review.googlesource.com/3822Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Change-Id: Ifd983cf50b1ccbe64a45d3b3ad171ef67f608033
Reviewed-on: https://go-review.googlesource.com/3831Reviewed-by: Dave Cheney <dave@cheney.net>