Oliver Tonnhofer [Tue, 27 Sep 2016 13:24:00 +0000 (15:24 +0200)]
image/png: improve compression by skipping filter for paletted images
Compression of paletted images is more efficient if they are not filtered.
This patch skips filtering for cbP8 images.
The improvements are demonstrated at https://github.com/olt/compressbench
This optimizes deferproc and deferreturn in various ways.
The most important optimization is that it more carefully arranges to
prevent preemption or stack growth. Currently we do this by switching
to the system stack on every deferproc and every deferreturn. While we
need to be on the system stack for the slow path of allocating and
freeing defers, in the common case we can fit in the nosplit stack.
Hence, this change pushes the system stack switch down into the slow
paths and makes everything now exposed to the user stack nosplit. This
also eliminates the need for various acquirem/releasem pairs, since we
are now preventing preemption by preventing stack split checks.
As another smaller optimization, we special case the common cases of
zero-sized and pointer-sized defer frames to respectively skip the
copy and perform the copy in line instead of calling memmove.
This speeds up the runtime defer benchmark by 42%:
name old time/op new time/op delta
Defer-4 75.1ns ± 1% 43.3ns ± 1% -42.31% (p=0.000 n=8+10)
In reality, this speeds up defer by about 2.2X. The two benchmarks
below compare a Lock/defer Unlock pair (DeferLock) with a Lock/Unlock
pair (NoDeferLock). NoDeferLock establishes a baseline cost, so these
two benchmarks together show that this change reduces the overhead of
defer from 61.4ns to 27.9ns.
name old time/op new time/op delta
DeferLock-4 77.4ns ± 1% 43.9ns ± 1% -43.31% (p=0.000 n=10+10)
NoDeferLock-4 16.0ns ± 0% 15.9ns ± 0% -0.39% (p=0.000 n=9+8)
This also shaves 34ns off cgo calls:
name old time/op new time/op delta
CgoNoop-4 122ns ± 1% 88.3ns ± 1% -27.72% (p=0.000 n=8+9)
Updates #14939, #16051.
Change-Id: I2baa0dea378b7e4efebbee8fca919a97d5e15f38
Reviewed-on: https://go-review.googlesource.com/29656 Reviewed-by: Keith Randall <khr@golang.org>
The big documentation comment at the top of malloc.go has gotten
woefully out of date. Update it.
Change-Id: Ibdb1bdcfdd707a6dc9db79d0633a36a28882301b
Reviewed-on: https://go-review.googlesource.com/29731 Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com> Reviewed-by: Rick Hudson <rlh@golang.org>
This documents all fields in MemStats and more clearly documents where
mstats differs from MemStats.
Fixes #15849.
Change-Id: Ie09374bcdb3a5fdd2d25fe4bba836aaae92cb1dd
Reviewed-on: https://go-review.googlesource.com/28972 Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>
We used to compute an estimate of the reachable heap size that was
different from the marked heap size. This ultimately caused more
problems than it solved, so we pulled it out, but memstats still has
both heap_reachable and heap_marked, and there are some leftover TODOs
about the problems with this estimate.
Clean this up by eliminating heap_reachable in favor of heap_marked
and deleting the stale TODOs.
Back in Go 1.4, memstats.next_gc was both the heap size at which GC
would trigger, and the size GC kept the heap under. When we switched
to concurrent GC in Go 1.5, we got somewhat confused and made this
variable the trigger heap size, while gcController.heapGoal became the
goal heap size.
memstats.next_gc is exposed to the user via MemStats.NextGC, while
gcController.heapGoal is not. This is unfortunate because 1) the heap
goal is far more useful for diagnostics, and 2) the trigger heap size
is just part of the GC trigger heuristic, which means it wouldn't be
useful to an application even if it tried to use it.
We never noticed this mess because MemStats.NextGC is practically
undocumented. Now that we're trying to document MemStats, it became
clear that this field had diverged from its original usefulness.
Clean up this mess by shuffling things back around so that next_gc is
the goal heap size and the new (unexposed) memstats.gc_trigger field
is the trigger heap size. This eliminates gcController.heapGoal.
Updates #15849.
Change-Id: I2cbbd43b1d78bdf613cb43f53488bd63913189b7
Reviewed-on: https://go-review.googlesource.com/29270
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com> Reviewed-by: Rick Hudson <rlh@golang.org>
Ian Lance Taylor [Mon, 26 Sep 2016 18:14:41 +0000 (11:14 -0700)]
runtime: merge setting new signal mask in minit
All the variants that sets the new signal mask in minit do the same
thing, so merge them. This requires an OS-specific sigdelset function;
the function already exists for linux, and is now added for other OS's.
Change-Id: Ie96f6f02e2cf09c43005085985a078bd9581f670
Reviewed-on: https://go-review.googlesource.com/29771
Run-TryBot: Ian Lance Taylor <iant@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Ian Lance Taylor [Mon, 26 Sep 2016 04:33:27 +0000 (21:33 -0700)]
runtime: unify sigtrampgo
Combine the various versions of sigtrampgo into a single function in
signal_unix.go. This requires defining a fixsigcode method on sigctxt
for all operating systems; it only does something on Darwin. This also
requires changing the darwin/amd64 signal handler to call sigreturn
itself, rather than relying on sigtrampgo to call sigreturn for it. We
can then drop the Darwin sigreturn function, as it is no longer used.
Change-Id: I5a0b9d2d2c141957e151b41e694efeb20e4b4b9a
Reviewed-on: https://go-review.googlesource.com/29761
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Alberto Donizetti [Sat, 7 May 2016 19:34:04 +0000 (21:34 +0200)]
cmd/compile: fix bogus "fallthrough statement out of place"
When processing a fallthrough, the casebody function in swt.go
checks that the last statement has indeed Op == OXFALL (not-processed
fallthrough) before setting it to OFALL (processed fallthrough).
Unfortunately, sometimes the fallthrough statement won't be in the
last node. For example, in
case 0:
return func() int {return 1}()
fallthrough
with an OVARKILL node in the last position. casebody will find that
last.Op != OXFALL, won't mark the fallthrough as processed, and the
fallthrough line will cause a "fallthrough statement out of place" error.
To fix this, we change casebody so that it searches for the fallthrough
statement backwards in the statements list, without assuming that it'll
be in the last position.
Fixes #13262
Change-Id: I366c6caa7fd7442d365bd7a08cc66a552212d9b2
Reviewed-on: https://go-review.googlesource.com/22921
Run-TryBot: Quentin Smith <quentin@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Quentin Smith <quentin@golang.org>
Ian Lance Taylor [Sun, 25 Sep 2016 20:38:54 +0000 (13:38 -0700)]
runtime: unify handling of alternate signal stack
Change all Unix systems to use stackt for the alternate signal
stack (some were using sigaltstackt). Add OS-specific setSignalstackSP
function to handle different types for ss_sp field, and unify all
OS-specific signalstack functions into one. Unify handling of alternate
signal stack in OS-specific minit and sigtrampgo functions via new
functions minitSignalstack and setGsignalStack.
Change-Id: Idc316dc69b1dd725717acdf61a1cd8b9f33ed174
Reviewed-on: https://go-review.googlesource.com/29757
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Currently raceSymbolizeCode uses funcline, which is internal runtime
function which crashes on incorrect PCs. Use FileLine instead,
it is public and does not crash on invalid data.
Note: FileLine returns "?" file on failure. That string is not NUL-terminated,
so we need to additionally check what FileLine returns.
Fixes #17190
Change-Id: Ic6fbd4f0e68ddd52e9b2dd25e625b50adcb69a98
Reviewed-on: https://go-review.googlesource.com/29714
Run-TryBot: Dmitry Vyukov <dvyukov@google.com> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Which is not useful. Moreover we never override source info,
so subsequent source code uses the same source info.
Moreover, empty file name makes compile emit no source debug info at all.
Update #17190
Change-Id: I7ae6fa4964520d7665743d340419b787df0b51e8
Reviewed-on: https://go-review.googlesource.com/29713
Run-TryBot: Dmitry Vyukov <dvyukov@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
PC passed to racegostart is expected to be a return PC
of the go statement. Race runtime will subtract 1 from the PC
before symbolization. Passing start PC of a function is wrong.
Add sys.PCQuantum to the function start PC.
Update #17190
Change-Id: Ia504c49e79af84ed4ea360c2aea472b370ea8bf5
Reviewed-on: https://go-review.googlesource.com/29712
Run-TryBot: Dmitry Vyukov <dvyukov@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Ian Lance Taylor [Sat, 24 Sep 2016 05:05:51 +0000 (22:05 -0700)]
runtime: merge Unix sighandler functions
Replace all the Unix sighandler functions with a single instance.
Push the relatively small amount of processor-specific code into five
methods on sigctxt: sigpc, sigsp, siglr, fault, preparePanic.
(Some processors already had a fault method.)
Change-Id: Ib459412ff8f7e0f5ad06bfd43eb827c8b196fc32
Reviewed-on: https://go-review.googlesource.com/29752
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: David Crawshaw <crawshaw@golang.org>
Ian Lance Taylor [Sat, 24 Sep 2016 00:54:51 +0000 (17:54 -0700)]
runtime: unify some signal handling functions
Unify the OS-specific versions of msigsave, msigrestore, sigblock,
updatesigmask, and unblocksig into single versions in signal_unix.go.
To do this, make sigprocmask work the same way on all systems, which
required adding a definition of sigprocmask for linux and openbsd.
Also add a single OS-specific function sigmaskToSigset.
Change-Id: I7cbf75131dddb57eeefe648ef845b0791404f785
Reviewed-on: https://go-review.googlesource.com/29689
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: David Crawshaw <crawshaw@golang.org>
David Crawshaw [Mon, 19 Sep 2016 18:13:07 +0000 (14:13 -0400)]
cmd/link: plugin support on darwin/amd64
This CL turns some special section marker symbols into real symbols
laid out in the sections they mark. This is to deal with the fact
that dyld on OS X resolves the section marker symbols in any dlopen-ed
Go program to the original section marker symbols in the host program.
More details in a comment in cmd/link/internal/ld/data.go.
Change-Id: Ie9451cfbf06d0bdcccb9959219c791b829f3f771
Reviewed-on: https://go-review.googlesource.com/29394 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: David Crawshaw <crawshaw@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Keith Randall [Mon, 19 Sep 2016 21:35:41 +0000 (14:35 -0700)]
cmd/compile: fix 4-byte unaligned load rules
The 2-byte rule was firing before the 4-byte rule, preventing
the 4-byte rule from firing. Update the 4-byte rule to use
the results of the 2-byte rule instead.
Add some tests to make sure we don't regress again.
math, cmd/internal/obj/ppc64: improve floor, ceil, trunc with asm
This adds the instructions frim, frip, and friz to the ppc64x
assembler for use in implementing the math.Floor, math.Ceil, and
math.Trunc functions to improve performance.
David Crawshaw [Fri, 23 Sep 2016 04:02:38 +0000 (00:02 -0400)]
runtime: use sched_yield instead of pthread_yield
Attempt to fix the linux-amd64-clang builder, which broke
with CL 29472.
Turns out pthread_yield is a non-portable Linux function, and
should have #define _GNU_SOURCE before #include <pthread.h>.
GCC doesn't complain about this, but Clang does:
./raceprof.go:44:3: warning: implicit declaration of function 'pthread_yield' is invalid in C99 [-Wimplicit-function-declaration]
(Though the error, while explicable, certainly could be clearer.)
There is a portable POSIX equivalent, sched_yield, so this
CL uses it instead.
David Crawshaw [Mon, 19 Sep 2016 18:08:21 +0000 (14:08 -0400)]
runtime: check plugin-loaded moduledata addresses
Inspired by difficulties with plugin support on darwin.
Change-Id: I2cef8410837946454e75d00e94e46791f03f2267
Reviewed-on: https://go-review.googlesource.com/29391 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Ian Lance Taylor [Wed, 21 Sep 2016 16:44:40 +0000 (09:44 -0700)]
cmd/compile: don't instrument copy and append in runtime
Instrumenting copy and append for the race detector changes them to call
different functions. In the runtime package the alternate functions are
not marked as nosplit. This caused a crash in the SIGPROF handler when
invoked on a non-Go thread in a program built with the race detector. In
some cases the handler can call copy, the race detector changed that to
a call to a non-nosplit function, the function tried to check the stack
guard, and crashed because it was running on a non-Go thread. The
SIGPROF handler is written carefully to avoid such problems, but hidden
function calls are difficult to avoid.
Fix this by changing the compiler to not instrument copy and append when
compiling the runtime package. Change the runtime package to add
explicit race checks for the only code I could find where copy is used
to write to user data (append is never used).
Change-Id: I11078a66c0aaa459a7d2b827b49f4147922050af
Reviewed-on: https://go-review.googlesource.com/29472
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Keith Randall [Tue, 20 Sep 2016 23:34:30 +0000 (16:34 -0700)]
cmd/compile: fix type of static closure pointer
var x *X = ...
defer x.foo()
As part of the defer, we need to calculate &(*X).foo·f. This expression
is the address of the static closure that will call (*X).foo when a
pointer to that closure is used in a call/defer/go. This pointer is not
currently properly typed in SSA. It is a pointer type, but the base
type is nil, not a proper type.
This turns out not to be a problem currently because we never use the
type of these SSA values. But I'm trying to change that (to be able to
spill them) in CL 28391. To fix, use uint8 as the fake type of the
closure.
Change-Id: Ieee388089c9af398ed772ee8c815122c347cb633
Reviewed-on: https://go-review.googlesource.com/29444
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Add an "errorcheckwithauto" action which performs error check
including lines with auto-generated functions (excluded by
default). Comment "// ERRORAUTO" matches these lines.
cmd/compile: ensure args are live in tail calls for LR machines
On link-register machines we uses RET (sym), instead of JMP (sym),
for tail call (so the assembler knows and may rewrite it to
restore link register if necessary). Add RET to the analysis.
Fixes #17186.
Fixes #16016 on link-register machines.
Adam Langley [Wed, 14 Sep 2016 18:50:36 +0000 (11:50 -0700)]
crypto/tls: fix deadlock when racing to complete handshake.
After renegotiation support was added (af125a5193c) it's possible for a
Write to block on a Read when racing to complete the handshake:
1. The Write determines that a handshake is needed and tries to
take the neccesary locks in the correct order.
2. The Read also determines that a handshake is needed and wins
the race to take the locks.
3. The Read goroutine completes the handshake and wins a race
to unlock and relock c.in, which it'll hold when waiting for
more network data.
If the application-level protocol requires the Write to complete before
data can be read then the system as a whole will deadlock.
Unfortunately it doesn't appear possible to reverse the locking order of
c.in and handshakeMutex because we might read a renegotiation request at
any point and need to be able to do a handshake without unlocking.
So this change adds a sync.Cond that indicates that a goroutine has
committed to doing a handshake. Other interested goroutines can wait on
that Cond when needed.
The test for this isn't great. I was able to reproduce the deadlock with
it only when building with -race. (Because -race happened to alter the
timing just enough.)
Fixes #17101.
Change-Id: I4e8757f7b82a84e46c9963a977d089f0fb675495
Reviewed-on: https://go-review.googlesource.com/29164 Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Kale Blankenship [Thu, 22 Sep 2016 02:03:06 +0000 (19:03 -0700)]
net/url: prefix relative paths containing ":" in the first segment with "./"
This change modifies URL.String to prepend "./" to a relative URL which
contains a colon in the first path segment.
Per RFC 3986 §4.2:
> A path segment that contains a colon character (e.g., "this:that")
> cannot be used as the first segment of a relative-path reference, as
> it would be mistaken for a scheme name. Such a segment must be
> preceded by a dot-segment (e.g., "./this:that") to make a relative-
> path reference.
https://go-review.googlesource.com/27440 corrects the behavior for http.FileServer,
but URL.String will still return an invalid URL. This CL reverts the changes to
http.FileServer as they are unnecessary with this fix.
runtime: revert CL 18835; don't install new signal stack unconditionally on dragonfly
This change reverts CL 18835 which is a workaroud for older DragonFly
BSD kernels, and fixes #14051, #14052 and #14067 in a more general way
the same as other platforms except NetBSD.
This change also bumps the minimum required version of DragonFly BSD
kernel to 4.4.4.
Fixes #16329.
Change-Id: I0b44b6afa675f5ed9523914226bd9ec4809ba5ae
Reviewed-on: https://go-review.googlesource.com/29491 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Lynn Boger [Thu, 25 Aug 2016 16:07:33 +0000 (11:07 -0500)]
cmd/link: split large elf text sections on ppc64x
Some applications built with Go on ppc64x with external linking
can fail to link with relocation truncation errors if the elf
text section that is generated is larger than 2^26 bytes and that
section contains a call instruction (bl) which calls a function
beyond the limit addressable by the 24 bit field in the
instruction.
This solution consists of generating multiple text sections where
each is small enough to allow the GNU linker to resolve the calls
by generating long branch code where needed. Other changes were added
to handle differences in processing when multiple text sections exist.
Some adjustments were required to the computation of a method's address
when using the method offset table when there are multiple text sections.
The number of possible section headers was increased to allow for up
to 128 text sections. A test case was also added.
Fixes #15823.
Change-Id: If8117b0e0afb058cbc072258425a35aef2363c92
Reviewed-on: https://go-review.googlesource.com/27790
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
net: add Resolver type, Dialer.Resolver, and DefaultResolver
The new Resolver type (a struct) has 9 Lookup methods, all taking a
context.Context.
There's now a new DefaultResolver global, like http's
DefaultTransport and DefaultClient.
net.Dialer now has an optional Resolver field to set the Resolver.
This also does finishes some resolver cleanup internally, deleting
lookupIPMerge and renaming lookupIPContext into Resolver.LookupIPAddr.
The Resolver currently doesn't let you tweak much, but it's a struct
specifically so we can add knobs in the future. Currently I just added
a bool to force the pure Go resolver. In the future we could let
people provide an interface to implement the methods, or add a Timeout
time.Duration, which would wrap all provided contexts in a
context.WithTimeout.
Fixes #16672
Change-Id: I7ba1f886704f06def7b6b5c4da9809db51bc1495
Reviewed-on: https://go-review.googlesource.com/29440
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Austin Clements [Mon, 9 May 2016 19:12:07 +0000 (15:12 -0400)]
runtime: remove unnecessary atomics from heapBitSetType
These used to be necessary when racing with updates to the mark bit,
but since the mark bit is no longer in the bitmap and the checkmark is
only updated with the world stopped, we can now always use regular
writes to update the type information in the heap bitmap.
Somewhat surprisingly, this has basically no overall performance
effect beyond the usual noise, but it does clean up the code.
Yasuhiro Matsumoto [Wed, 24 Aug 2016 01:34:16 +0000 (10:34 +0900)]
os: use GetConsoleCP() instead of GetACP()
It is possible (and common) for Windows systems to use a different codepage
for console applications from that used on normal windowed application
(called ANSI codepage); for instance, most of the western Europe uses
CP850 for console (for backward compatibility with MS-DOS), while
windowed applications use a different codepage depending on the country
(eg: CP1252 aka Latin-1). The usage being changed with this commit is
specifically related to decoding input coming from the console, so the
previous usage of the ANSI codepage was wrong.
Also fixes an issue that previous did convert bytes as NFD. Go is
designed to handle single Unicode code point. This fix change behaivor
to NFC.
Change-Id: I939f087df133710495fdf6f09040051cb9b176d7
Reviewed-on: https://go-review.googlesource.com/29442 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Austin Clements [Thu, 25 Aug 2016 15:50:50 +0000 (11:50 -0400)]
runtime: consistency check for G rescan position
Issue #17099 shows a failure that indicates we rescanned a stack twice
concurrently during mark termination, which suggests that the rescan
list became inconsistent. Add a simple check when we dequeue something
from the rescan list that it claims to be at the index where we found
it.
Austin Clements [Thu, 25 Aug 2016 15:52:24 +0000 (11:52 -0400)]
runtime: report GCSys and OtherSys in heap profile
The comment block at the end of the heap profile includes *almost*
everything from MemStats. Add the missing fields. These are useful for
debugging RSS that has gone to GC-internal data structures.
On ARM, DIV, DIVU, MOD, MODU are pseudo instructions that makes
runtime calls _div/_udiv/_mod/_umod, which themselves are wrappers
of udiv. The udiv function does the real thing.
Instead of generating these pseudo instructions, call to udiv
directly. This removes one layer of wrappers (which has an awkward
way of passing argument), and also allows combining DIV and MOD
if both results are needed.
Change-Id: I118afc3986db3a1daabb5c1e6e57430888c91817
Reviewed-on: https://go-review.googlesource.com/29390 Reviewed-by: David Chase <drchase@google.com>
David Crawshaw [Mon, 19 Sep 2016 18:11:20 +0000 (14:11 -0400)]
cmd/internal/obj, cmd/link: darwin dynlink support
This makes it possible for cmd/compile, when run with -dynlink on
darwin/amd64, to generate TLS_LE relocations which the linker then
turns into the appropriate PC-relative GOT load.
Change-Id: I1a71da432608bdb108ff66c22de600100209c873
Reviewed-on: https://go-review.googlesource.com/29393 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: David Crawshaw <crawshaw@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
David Crawshaw [Sat, 17 Sep 2016 13:39:33 +0000 (09:39 -0400)]
cmd/link: replace ld.Link.Diag with ld.Errorf
Instead of using ctxt.Cursym, Errorf takes an explicit *Symbol
parameter. This removes most uses of Cursym and means the *Link
context object is needed in fewer parts of the linker.
All transformations done manually, as wiring Cursym is tricky.
Change-Id: Ief88b00b73904224675c0035684c3a84c19249d7
Reviewed-on: https://go-review.googlesource.com/29369 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Michael Hudson-Doyle <michael.hudson@canonical.com>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Keith Randall [Mon, 19 Sep 2016 22:26:48 +0000 (15:26 -0700)]
doc: mention KeepAlive & input args change
Change-Id: Icfb38f492ae71432858b7104fcba18a9ba835192
Reviewed-on: https://go-review.googlesource.com/29410 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Rob Pike [Sun, 18 Sep 2016 04:35:42 +0000 (14:35 +1000)]
time: allow long fractions in ParseDuration
The code scanned for an integer after a decimal point, which
meant things could overflow if the number was very precise
(0.1234123412341234123412342134s). This fix changes the
parser to stop adding precision once we run out of bits, rather
than trigger an erroneous overflow.
We could parse durations using floating-point arithmetic,
but since the type is int64 and float64 has only has 53 bits
of precision, that would be imprecise.
Carlos Eduardo Seo [Tue, 12 Apr 2016 21:38:00 +0000 (18:38 -0300)]
cmd/asm, cmd/internal/obj/ppc64: add ppc64 vector registers and instructions
The current implementation for Power architecture does not include the vector
(Altivec) registers. This adds the 32 VMX registers and the most commonly used
instructions: X-form loads/stores; VX-form logical operations, add/sub,
rotate/shift, count, splat, SHA Sigma and AES cipher; VC-form compare; and
VA-form permute, shift, add/sub and select.
Fixes #15619
Change-Id: I544b990631726e8fdfcce8ecca0aeeb72faae9aa
Reviewed-on: https://go-review.googlesource.com/25600
Run-TryBot: David Chase <drchase@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com> Reviewed-by: David Chase <drchase@google.com>
Keith Randall [Fri, 16 Sep 2016 20:50:18 +0000 (13:50 -0700)]
cmd/compile: cache CFG-dependent computations
We compute a lot of stuff based off the CFG: postorder traversal,
dominators, dominator tree, loop nest. Multiple phases use this
information and we end up recomputing some of it. Add a cache
for this information so if the CFG hasn't changed, we can reuse
the previous computation.
Change-Id: I9b5b58af06830bd120afbee9cfab395a0a2f74b2
Reviewed-on: https://go-review.googlesource.com/29356 Reviewed-by: David Chase <drchase@google.com>