> The language specification already defines byte as an alias for
> uint8 and similarly rune as an alias for int32, using the word alias
> as an informal term. It is a goal that the new type declaration
> semantics not introduce a different meaning for alias. That is, it
> should be possible to describe the existing meanings of byte and
> uint8 by saying that they behave as if predefined by:
>
> type byte = uint8
> type rune = int32
So, do that. Seems to work.
Updates #18130
Change-Id: I0740bab3f8fb23e946f3542fdbe819007a99465a
Reviewed-on: https://go-review.googlesource.com/45017 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Ian Lance Taylor [Tue, 6 Jun 2017 22:08:59 +0000 (15:08 -0700)]
runtime: intercept munmap as we do mmap
For cgo programs on linux-amd64 we call the C function mmap.
This supports programs such as the C memory sanitizer that need to
intercept all calls to mmap. It turns out that there are programs that
intercept both mmap and munmap, or that at least expect that if they
intercept mmap, they also intercept munmap. So, if we permit mmap
to be intercepted, also permit munmap to be intercepted.
No test, as it requires two odd things: a C program that intercepts
mmap and munmap, and a Go program that calls munmap.
Brad Fitzpatrick [Tue, 6 Jun 2017 22:05:34 +0000 (22:05 +0000)]
doc: delete go1.8.txt
Fixes #20591
Change-Id: I2a4674a3430c5a4d3c569f3ea654c6ff4d9bf7ee
Reviewed-on: https://go-review.googlesource.com/45015 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Brad Fitzpatrick [Tue, 6 Jun 2017 00:04:12 +0000 (00:04 +0000)]
doc: start of go1.9.html release notes
Many TODOs remain.
Updates #20587
Change-Id: If49854ae4d36346d9e072a386f413cc85c66b62a
Reviewed-on: https://go-review.googlesource.com/45012 Reviewed-by: Ian Lance Taylor <iant@golang.org>
gulyasm [Sat, 7 Jan 2017 22:30:47 +0000 (23:30 +0100)]
math: add doc note about floating point operation
Go doesn't guarantee that the result of floating point operations will
be the same on different architectures. It was not stated in the
documentation, that can lead to confusion.
Fixes #18354
Change-Id: Idb1b4c256fb9a7158a74256136eca3b8ce44476f
Reviewed-on: https://go-review.googlesource.com/34938 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Russ Cox [Wed, 31 May 2017 15:35:29 +0000 (11:35 -0400)]
cmd/link: add -importcfg to specify import resolution
Adds the ability to specify the file location of each imported package,
like in the -importcfg added to cmd/compile in a related CL.
In effect, -importcfg is a generalization of and supersedes -installsuffix
and -L. Of course, those flags will continue to be supported, for
compatibility with other tools.
Having this flag in Go 1.9 will let us try some experiments involving
package management without needing guinea pigs to build a custom
Go toolchain.
This flag also helps with #14271 at some later point.
For #20579.
Change-Id: Ie4c171bcd3aa2faa446ac340e36516f2f9853882
Reviewed-on: https://go-review.googlesource.com/44851
Run-TryBot: Russ Cox <rsc@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Russ Cox [Wed, 31 May 2017 15:19:54 +0000 (11:19 -0400)]
cmd/compile: add -importcfg to specify import resolution
Allows reading -importmap options from a file instead of putting
them all on the command line, and adds the ability to specify the
file location of specific packages. In effect, -importcfg is a generalization
of and supersedes -importmap, -importsuffix, and -I.
Of course, those flags will continue to be supported,
for compatibility with other tools.
Having this flag in Go 1.9 will let us try some experiments involving
package management without needing guinea pigs to build a
custom Go toolchain.
This flag also helps with #14271 at some later point.
For #20579.
Change-Id: If005dbc2b01d8fd16cbfd3687dfbe82499f4bc56
Reviewed-on: https://go-review.googlesource.com/44850
Run-TryBot: Russ Cox <rsc@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
This is actually just a consequence of the fact that if the separator
sep (2nd argument) is not found the string s (1st argument), then the
Split* functions return a length 1 slice with the string s in it.
Document the general case: if sep is not in s, what you get is a len 1
slice with s in it; unless both s and sep are "", in that case you get
an empty slice of length 0.
Fixes #19726
Change-Id: I64c8220b91acd1e5aa1cc1829199e0cd8c47c404
Reviewed-on: https://go-review.googlesource.com/44950 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Daniel Martí <mvdan@mvdan.cc>
Ian Lance Taylor [Fri, 2 Jun 2017 22:56:35 +0000 (15:56 -0700)]
runtime: save and restore CR for ppc64le
C code expects CR2, CR3, and CR4 to be preserved across function calls.
Preserve the entire CR register across function calls in
_rt0_ppc64le_linux_lib and crosscall2. The standard ppc64le call frame
uses 8(R1) as the place to save CR; emulate that.
It's hard to write a reliable test for this as it requires writing C
code that sets CR2, CR3, or CR4 across a call to a Go function.
Change-Id: If39e771a5b574602b848227312e83598fe74eab7
Reviewed-on: https://go-review.googlesource.com/44733
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Carlos Eduardo Seo <cseo@linux.vnet.ibm.com> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
Ian Lance Taylor [Fri, 2 Jun 2017 22:51:18 +0000 (15:51 -0700)]
cmd/internal/obj/ppc64: fix MOVFL REG, CONST
The MOVFL instruction (which external PPC64 docs call mtcrf) can take
either a CR register or a constant. It doesn't make sense to specify
both, as the CR register implies the constant value. Specifying either
a register or a constant is enforced by the implementation in the
asmout method (case 69).
However, the optab was providing a form that specified both a constant
and a CR register, and was not providing a form that specified only a
constant. This CL fixes the optab table to provide a form that takes
only a constant.
No test because I don't know where to write it. The next CL in this
series will use the new instruction format.
Change-Id: I8bb5d3ed60f483b54c341ce613931e126f7d7be6
Reviewed-on: https://go-review.googlesource.com/44732
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
Programs built from N libraries required O(N²) time to do the
deduplication checks, even if there were never any duplicates.
In most programs N is small enough not to worry, but this may
affect large programs.
Noticed by inspection, not any specific bug report.
Fixes #20578.
Change-Id: Ic4108f1058be39da990a79b1e0b8ce95fde44cef
Reviewed-on: https://go-review.googlesource.com/44852
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Russ Cox [Mon, 5 Jun 2017 14:23:42 +0000 (10:23 -0400)]
cmd/go: include Fortran files in build ID computation
Otherwise removing a .f file won't trigger a rebuild.
Noticed by inspection while working on the code.
I don't have a good way to write a test for this,
and I expect the code to change significantly in the next
release, but might as well get it right for Go 1.9.
Change-Id: I3f6f9f71b3a7d4f0be49a47419dac59899959e7c
Reviewed-on: https://go-review.googlesource.com/44855
Run-TryBot: Russ Cox <rsc@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
David Glasser [Fri, 28 Apr 2017 23:40:39 +0000 (16:40 -0700)]
net/http: make Transport retry GetBody requests if nothing written
This is another attempt at the change attempted in
https://golang.org/cl/27117 and rolled back in https://golang.org/cl/34134
The difference between this and the previous attempt is that this version only
retries if the new field GetBody is set on the Request.
Additionally, this allows retries of requests with idempotent methods even if
they have bodies, as long as GetBody is defined.
This also fixes an existing bug where readLoop could make a redundant call to
setReqCanceler for DELETE/POST/PUT/etc requests with no body with zero bytes
written.
This clarifies the existing TestRetryIdempotentRequestsOnError test (and changes
it into a test with 4 subtests). When that test was written, it was in fact
testing "retry idempotent requests" logic, but the logic had changed since then,
and it was actually testing "retry requests with no body when no bytes have been
written". (You can confirm this by changing the existing test from a GET to a
DELETE; it passes without the changes in this CL.) We now test for the no-Body
and GetBody cases for both idempotent and nothing-written-non-idempotent
requests.
Ben Shi [Sat, 27 May 2017 10:00:55 +0000 (10:00 +0000)]
cmd/internal/obj/arm: fix constant decomposition
There are two issues in constant decomposition.
1. A typo in "func immrot2s" blocks "case 107" of []optab be triggered.
2. Though "ADD $0xffff, R0, R0" is decomposed to "ADD $0xff00, R0, R0" and
"ADD $0x00ff, R0, R0" as expected, "ADD $0xffff, R0" still uses the
constant pool, which should be the same as "ADD $0xffff, R0, R0".
This patch fixes them and adds more instruction encoding tests.
Alberto Donizetti [Sat, 3 Jun 2017 19:36:51 +0000 (21:36 +0200)]
encoding/json: clarify unmarshaling behaviour on bad fields
Fixes #19526
Change-Id: Ifaaf454e0e89fdf4309118c2e2e6ac0d0a43c39d
Reviewed-on: https://go-review.googlesource.com/44711 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Austin Clements [Mon, 5 Jun 2017 14:59:57 +0000 (10:59 -0400)]
runtime: expand acceptable PingPongHog factor from 2 to 5
Since TestPingPongHog tests the scheduler, it's ultimately
probabilistic. Currently, it requires the result be at most of factor
of 2 off of the ideal. It turns out this isn't quite enough in
practice, with factors on 1000 iterations on linux/amd64 ranging from
0.48 to 2.5. If the test were failing, we would expect a factor closer
to 1000X, so it's pretty safe to expand the accepted factor from 2 to
5.
Alexander Menzhinsky [Thu, 13 Apr 2017 21:15:26 +0000 (16:15 -0500)]
cmd/go: reject -buildmode=plugin if package is not main
Fixes #17625
Change-Id: I09319b888b547c631a50dbfab9255cc08e8a2426
Reviewed-on: https://go-review.googlesource.com/40691 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Change-Id: I98d265bfa4f5944360d0258d2cb6bbc78010c708
Reviewed-on: https://go-review.googlesource.com/44731 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Robert Griesemer [Fri, 2 Jun 2017 00:51:02 +0000 (17:51 -0700)]
go/printer: don't emit unnecessary //line directives before empty lines
1) Split atLineBegin into its two components: writing of // line directives
and writing of indentation (no functionality changes).
2) Don't call writeLineDirective at the beginning of a line if we're
writing white space - it's not necessary. This is the bug fix.
3) Move testing of the SourcePos mode out of writeLineDirective and
into the (single) caller. Clearer and more efficient.
(Instead of these 3 changes one could also have simply called the
original atLineBegin with position p.out rather than p.pos. This
would have caused atLineBegin to not write a line directive.
Factoring the code seemed like a cleaner and more direct approach.)
Fixes #5945.
Change-Id: Ia8710806b6d3d4e5044116b142c036a4ab5a1764
Reviewed-on: https://go-review.googlesource.com/44651 Reviewed-by: Alan Donovan <adonovan@google.com>
Alberto Donizetti [Fri, 2 Jun 2017 15:07:33 +0000 (17:07 +0200)]
bytes: note that NewBuffer take ownership of its argument
Fixes #19383
Change-Id: Ic84517053ced7794006f6fc65e6f249e97d6cf35
Reviewed-on: https://go-review.googlesource.com/44691 Reviewed-by: Ian Lance Taylor <iant@golang.org>
The test no longer passes as of https://golang.org/cl/43777, which
intentionally disables the feature that this test was checking for.
Nobody noticed since the test is not run in -short mode.
Joe Tsai [Fri, 2 Jun 2017 00:31:24 +0000 (17:31 -0700)]
compress/gzip: clarify behavior of Writer.Close
Fixes #20551
Change-Id: Ia47cae14a26fe5f278ad7209218d083cc50a3ff8
Reviewed-on: https://go-review.googlesource.com/44572 Reviewed-by: Ian Lance Taylor <iant@golang.org>
cmd/compile/internal/gc: speed-up small array comparison
Currently we inline array comparisons for arrays with at most 4 elements.
Compare arrays with small size, but more than 4 elements (e. g. [16]byte)
with larger compares. This provides very slightly smaller binaries,
and results in faster code.
Chris Broadfoot [Wed, 31 May 2017 18:20:16 +0000 (11:20 -0700)]
doc: update bootstrap archive URL
This includes the patch for systems that build PIE executables by
defaul
Updates #20276.
Change-Id: Iecf8dfcf11bc18d397b8075559c37e3610f825cb
Reviewed-on: https://go-review.googlesource.com/44470 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Julio Montes [Mon, 22 May 2017 15:33:31 +0000 (10:33 -0500)]
syscall: allow processes steal a controlling terminal on Linux
The current implementation of forkAndExecInChild for Linux does not allow
spawned processes steal the controlling terminal from a different session
group. This patch passes 1 as the argument to TIOCSCTTY in order to allow
spawned processes steal controlling terminals.
Fixes #20454
Change-Id: I171b8981509d648b07f89bddc1e9d45cb70e00e6
Reviewed-on: https://go-review.googlesource.com/44343 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
runtime: work around NetBSD bug in thread creation
Block signals explicitly during lwp_create since blocking via
the context does not work.
This was originally added in pkgsrc as
http://cvsweb.netbsd.org/bsdweb.cgi/pkgsrc/lang/go/patches/patch-src_runtime_os__netbsd.go?rev=1.1
Fixes #19295.
Change-Id: I431e70c047cff446e9cfc20d356f45df98c63487
Reviewed-on: https://go-review.googlesource.com/41350 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
David Chase [Tue, 30 May 2017 20:06:59 +0000 (16:06 -0400)]
cmd/compile: fix subword store/load elision for amd64, x86, arm
Replacing byteload-of-bytestore-of-x with x is incorrect
when x contains a larger-than-byte value (and so on for
16 and 32-bit load/store pairs). Replace "x" with the
appropriate zero/sign extension of x, which if unnecessary
will be repaired by other rules.
Made logic for arm match x86 and amd64; yields minor extra
optimization, plus I am (much) more confident it's correct,
despite inability to reproduce bug on arm.
Ppc64 lacks this optimization, hence lacks this problem.
See related https://golang.org/cl/37154/
Fixes #20530.
Change-Id: I6af9cac2ad43bee99cafdcb04725ce7e55a43323
Reviewed-on: https://go-review.googlesource.com/44355
Run-TryBot: David Chase <drchase@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
Dmitry Savintsev [Mon, 29 May 2017 20:14:32 +0000 (22:14 +0200)]
cmd/link: fix left-over reference to reflect.c
Replaced ../gc/reflect.c with
cmd/compile/internal/gc/reflect.go.
Fixes #20525
Change-Id: Ibafd36ea446ace5c677df27873a4bbf716a0a9bc
Reviewed-on: https://go-review.googlesource.com/44376 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Change-Id: I0aad9a8b504d673b29d3c26b3717ffa8512e0cdf
Reviewed-on: https://go-review.googlesource.com/44392 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Wei Xiao [Thu, 25 May 2017 05:59:39 +0000 (13:59 +0800)]
runtime: fix stack split at bad time when disable inlining
key32 is called between entersyscallblock and exitsyscall
stack split may occur if disable inlining and the G is preempted
Fix the problem by describing key32 as nosplit function
Fixes #20510
Change-Id: I1f0787995936f34ef0052cf79fde036f1b338865
Reviewed-on: https://go-review.googlesource.com/44390 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Dmitry Savintsev [Mon, 29 May 2017 10:01:45 +0000 (12:01 +0200)]
runtime: fix file reference in comments
cmd/compile/internal/ld/decodesym.go is now
cmd/link/internal/ld/decodesym.go
Change-Id: I16ec5c89aa3507e70676c2b50d70f1fde533a085
Reviewed-on: https://go-review.googlesource.com/44373 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Heschi Kreinick [Wed, 24 May 2017 21:43:08 +0000 (17:43 -0400)]
cmd/internal/dwarf: update to DWARF4, emit frame_base
In preparation for CL 41770, upgrade .debug_info to DWARF4, and emit
DW_AT_frame_base on subprograms. This should make no semantic
difference.
Also fix a long-standing bug/inconsistency in puttattr: it didn't
add the addend to ref_addrs. Previously this didn't matter because it
was only used for types, but now it's used for section offsets into
symbols that have multiple entries.
RELNOTE=yes
Change-Id: Ib10654ac92edfa29c5167c44133648151d70cf76
Reviewed-on: https://go-review.googlesource.com/44210 Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>
Document says:
> If general purpose bit 11 is set, the filename and comment must support The
> Unicode Standard, Version 4.1.0 or greater using the character encoding form
> defined by the UTF-8 storage specification.
Since Go encode the filename to UTF-8, general purpose bit 11 should be set.
Change-Id: Ica4af02b4dc695e9a5c015ae360e70171efb6ee3
Reviewed-on: https://go-review.googlesource.com/39570 Reviewed-by: Joe Tsai <thebrokentoaster@gmail.com>
Run-TryBot: Joe Tsai <thebrokentoaster@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Austin Clements [Wed, 24 May 2017 20:21:01 +0000 (16:21 -0400)]
debug/dwarf: heuristically handle both UNIX and Windows paths
Currently debug/dwarf assumes all paths in line tables will be
UNIX-style paths, which obviously isn't the case for binaries built on
Windows. However, we can't simply switch from the path package to the
filepath package because we don't know that we're running on the same
host type that built the binary and we want this to work even if we're
not. This is essentially the approach taken by GDB, which treats paths
in accordance with the system GDB itself is compiled for. In fact, we
can't even guess the compilation system from the type of the binary
because it may have been cross-compiled.
We fix this by heuristically determining whether paths are UNIX-style
or DOS-style by looking for a drive letter or UNC path. If we see a
DOS-style path, we use appropriate logic for determining whether the
path is absolute and for joining two paths. This is helped by the fact
that we should basically always be starting with an absolute path.
However, it could mistake a relative UNIX-style path that begins with
a directory like "C:" for an absolute DOS-style path. There doesn't
seem to be any way around this.
Fixes #19784.
Change-Id: Ie13b546d2f1dcd8b02e668583a627b571b281588
Reviewed-on: https://go-review.googlesource.com/44017
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Alex Brainman <alex.brainman@gmail.com>
Ben Shi [Tue, 23 May 2017 10:12:36 +0000 (10:12 +0000)]
cmd/asm/internal/asm: fix a bug in ARM assembly encoding test
It is expected to test assembly code for ARMv5, ARMv6 and ARMv7
in cmd/asm/internal/asm/endtoend_test.go. But actually the loop
in "func TestARMEndToEnd(t *testing.T)" runs three times all
for ARMv5.
This patch fixes that bug and adds a new armv6.s which is only tested
with GOARM=6.
Austin Clements [Tue, 23 May 2017 21:07:26 +0000 (17:07 -0400)]
runtime: accept non-monotonic arena allocation on 32-bit
Currently, the heap arena allocator allocates monotonically increasing
addresses. This is fine on 64-bit where we stake out a giant block of
the address space for ourselves and start at the beginning of it, but
on 32-bit the arena starts at address 0 but we start allocating from
wherever the OS feels like giving us memory. We can generally hint the
OS to start us at a low address, but this doesn't always work.
As a result, on 32-bit, if the OS gives us an arena block that's lower
than the current block we're allocating from, we simply say "thanks
but no thanks", return the whole (256MB!) block of memory, and then
take a fallback path that mmaps just the amount of memory we need
(which may be as little as 8K).
We have to do this because mheap_.arena_used is *both* the highest
used address in the arena and the next address we allocate from.
Fix all of this by separating the second role of arena_used out into a
new field called arena_alloc. This lets us accept any arena block the
OS gives us. This also slightly changes the invariants around
arena_end. Previously, we ensured arena_used <= arena_end, but this
was related to arena_used's second role, so the new invariant is
arena_alloc <= arena_end. As a result, we no longer necessarily update
arena_end when we're updating arena_used.
Fixes #20259 properly. (Unlike the original fix, this one should not
be cherry-picked to Go 1.8.)
This is reasonably low risk. I verified several key properties of the
32-bit code path with both 4K and 64K physical pages using a symbolic
model and the change does not materially affect 64-bit (arena_used ==
arena_alloc on 64-bit). The only oddity is that we no longer call
setArenaUsed with racemap == false to indicate that we're creating a
hole in the address space, but this only happened in a 32-bit-only
code path, and the race detector require 64-bit, so this never
mattered anyway.
Change-Id: Ib1334007933e615166bac4159bf357ae06ec6a25
Reviewed-on: https://go-review.googlesource.com/44010
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Rick Hudson <rlh@golang.org>
Emmanuel Odeke [Thu, 25 May 2017 01:36:41 +0000 (19:36 -0600)]
net/http: revert CL 43779
CL 43779/commit 6a6c792eef55eded7fb3165a330ec2b239b83960
broke the builds at tip, and that CL doesn't account for
cases where Redirect is directly invoked with a full URL
that itself has a query string.
Joe Tsai [Wed, 24 May 2017 20:52:56 +0000 (13:52 -0700)]
compress/bzip2: remove dead code in huffman.go
The logic performs a series of shifts, which are useless given
that they are followed by an assignment that overrides the
value of the previous computation.
I suspect (but cannot prove) that this is leftover logic from an
original approach that attempted to store both the Huffman code
and the length within the same variable instead of using two
different variables as it currently does now.
Fixes #17949
Change-Id: Ibf6c807c6cef3b28bfdaf2b68d9bc13503ac21b2
Reviewed-on: https://go-review.googlesource.com/44091 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Keith Randall [Wed, 24 May 2017 18:16:42 +0000 (11:16 -0700)]
cmd/compile: test for moving spills
Test that we really do move spills down to the dominator of
all the uses.
Also add a test where go1.8 would have moved the spill out of
the loop into two exit points, but go1.9 doesn't move the spill.
This is a case where the 1.9 spill moving code does not subsume
the 1.8 spill moving code.
Maybe we resurrect moving-spills-out-of-loops CL to fix this one.
(I suspect it wouldn't be worth the effort, but would be happy
to hear evidence otherwise.)
Update #20472
Change-Id: I7dbf8d65e7f4d675d14e5ecf502887cebda35d2a
Reviewed-on: https://go-review.googlesource.com/44038
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: David Chase <drchase@google.com>
When writing the 'Connection: close' header based on response Close
attribute we also check if it is already in the headers scheduled
to be written and skip if necessary.
net/http: deep copy Request.URL also in Request.WithContext's copy
Despite the previously known behavior of Request.WithContext
shallow copying a request, usage of the request inside server.ServeHTTP
mutates the request's URL. This CL implements deep copying of the URL.
David Chase [Mon, 20 Mar 2017 18:24:00 +0000 (14:24 -0400)]
cmd/dist: ensure android-implies-linux for file names in dist
This is one hurdle to building Go on Android; the runtime does
not build properly because *_linux.go files are excluded from
the "Building go_bootstrap" step when GOOS=android.
Jan Berktold [Thu, 23 Feb 2017 14:53:27 +0000 (15:53 +0100)]
net/http: make Transport respect non lower case Content-Encoding
The existing Transport implementation does not detect gzip encoding
when the Content-Encoding header is not lower-case. This is not
compliant with RFC2616 section 3.5 "All content-coding values are
case-insensitive." and caused issues in the wild.
Brad Fitzpatrick [Wed, 24 May 2017 00:05:04 +0000 (00:05 +0000)]
net/http: update bundled x/net/http2
This updates the bundled copy of x/net/http2 to x/net git rev a8e8f92cd6 for:
http2: remove extra goroutine stack from awaitGracefulShutdown
https://golang.org/cl/43230
http2: Discard DATA frames from the server after the response body is closed
https://golang.org/cl/43810
Fixes #20302
Fixes #18471
Fixes #20448
Change-Id: I00972836deb2fe6049f631ee44901732a641b171
Reviewed-on: https://go-review.googlesource.com/44006 Reviewed-by: Tom Bergan <tombergan@google.com>
Tom Bergan [Tue, 23 May 2017 23:03:21 +0000 (16:03 -0700)]
net/http: Add Server.RegisterOnShutdown
This will be used to allow http2 servers to register a shutdown function
so that net/http.Server.Shutdown will work when the http2 server is
configured via a manual call to http2.ConfigureServer. Currently, Shutdown
only works when the http2 server is configured automatically by the
net/http package.
Updates #20302
Updates #18471
Change-Id: Ifc2b5f3126126a106b49ea4a7e999279852b9cc9
Reviewed-on: https://go-review.googlesource.com/44003
Run-TryBot: Tom Bergan <tombergan@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Austin Clements [Tue, 23 May 2017 21:54:24 +0000 (17:54 -0400)]
runtime: use pselect6 for usleep on linux/386
Commit 4dcba023c6 replaced select with pselect6 on linux/amd64 and
linux/arm, but it turns out the Android emulator uses linux/386. This
makes the equivalent change there, too.
cmd/cgo: support indirect macro expansion for string
Current code cannot handle string #define macros if those macros are
defined via other macros. This CL solve the issue.
Updates #18720
Change-Id: Ibed0773d10db3d545bb246b97e81c0d19e3af3d5
Reviewed-on: https://go-review.googlesource.com/41312 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Brad Fitzpatrick [Tue, 23 May 2017 05:22:25 +0000 (05:22 +0000)]
cmd/go: include GOARM and GO386 in computed build ID
Now:
$ GOARCH=arm GOARM=5 go install -x cmd/go
... followed by:
$ GOARCH=arm GOARM= go install -x cmd/go
... actually does work. Previously the second "go install" would reuse
the cached binaries from the GOARM=5 command and not rebuild.
(Or vice versa from GOARM= to GOARM=5)
Wade Simmons [Tue, 23 May 2017 00:31:17 +0000 (20:31 -0400)]
crypto/rand: only read necessary bytes for Int
We only need to read the number of bytes required to store the value
"max - 1" to generate a random number in the range [0, max).
Before, there was an off-by-one error where an extra byte was read from
the io.Reader for inputs like "256" (right at the boundary for a byte).
There was a similar off-by-one error in the logic for clearing bits and
thus for any input that was a power of 2, there was a 50% chance the
read would continue to be retried as the mask failed to remove a bit.
Austin Clements [Thu, 18 May 2017 18:50:12 +0000 (14:50 -0400)]
runtime: remove unused copies of special stack guards
There are two copies each of the stackPreempt/_StackPreempt and
stackFork/_StackFork constants. Remove the ones left over from C that
are no longer used.
Austin Clements [Thu, 18 May 2017 18:35:53 +0000 (14:35 -0400)]
runtime: fix stackCache=0 debug mode
Setting stackCache to 0 to disable stack caches for debugging hasn't
worked for a long time. It causes stackalloc to fall back to full span
allocation, round sub-page stacks down to 0 pages, and blow up.
Fix this debug mode so it disables the per-P caches, but continues to
use the global stack pools for small stacks, which correctly handle
sub-page stacks. While we're here, rename stackCache to stackNoCache
so it acts like the rest of the stack allocator debug modes where "0"
is the right default value.
Fixes #17291.
Change-Id: If401c41cee3448513cbd7bb2e9334a8efab257a7
Reviewed-on: https://go-review.googlesource.com/43637 Reviewed-by: Keith Randall <khr@golang.org>
Austin Clements [Thu, 18 May 2017 17:59:00 +0000 (13:59 -0400)]
runtime: fix stackFromSystem returning memory
The stackFromSystem debug mode has two problems:
1) It rounds the stack allocation to _PageSize. If the physical page
size is >8K, this can cause unmapping the memory later to either
under-unmap or over-unmap.
2) It doesn't return the rounded-up allocation size to its caller, so
when we later unmap the memory, we may pass the wrong length.
Fix these problems by rounding the size up to the physical page size
and putting that rounded-up size in the returned stack bounds.
Fixes #17289.
Change-Id: I6b854af3b06bb16e3750798397bb5e2a722ec1cb
Reviewed-on: https://go-review.googlesource.com/43636 Reviewed-by: Keith Randall <khr@golang.org>
Austin Clements [Mon, 22 May 2017 19:53:49 +0000 (15:53 -0400)]
runtime: don't corrupt arena bounds on low mmap
If mheap.sysAlloc doesn't have room in the heap arena for an
allocation, it will attempt to map more address space with sysReserve.
sysReserve is given a hint, but can return any unused address range.
Currently, mheap.sysAlloc incorrectly assumes the returned region will
never fall between arena_start and arena_used. If it does,
mheap.sysAlloc will blindly accept the new region as the new
arena_used and arena_end, causing these to decrease and make it so any
Go heap above the new arena_used is no longer considered part of the
Go heap. This assumption *used to be* safe because we had all memory
between arena_start and arena_used mapped, but when we switched to an
arena_start of 0 on 32-bit, it became no longer safe.
Most likely, we've only recently seen this bug occur because we
usually start arena_used just above the binary, which is low in the
address space. Hence, the kernel is very unlikely to give us a region
before arena_used.
Since mheap.sysAlloc is a linear allocator, there's not much we can do
to handle this well. Hence, we fix this problem by simply rejecting
the new region if it isn't after arena_end. In this case, we'll take
the fall-back path and mmap a small region at any address just for the
requested memory.
Fixes #20259.
Change-Id: Ib72e8cd621545002d595c7cade1e817cfe3e5b1e
Reviewed-on: https://go-review.googlesource.com/43870 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Rick Hudson <rlh@golang.org>
Joe Richey joerichey@google.com [Mon, 22 May 2017 21:36:43 +0000 (14:36 -0700)]
crypto/rand: use blocking getrandom call on Linux when supported
By changing getRandomLinux to immediately use the getrandom() syscall
without GRND_NONBLOCK, we now only fall back to reading from
/dev/urandom on Linux if the kernel does not support the getrandom()
syscall. This means reads for crypto/rand will now block if the kernel
has insufficient entropy on Linux kernels after v3.16.
Before, if the kernel had insufficient entropy, it would fall back to
reading from /dev/urandom. This would potentially return predictable
data.
Alex Brainman [Mon, 22 May 2017 07:17:39 +0000 (17:17 +1000)]
os/exec: ignore some pipe write errors on windows
This change is windows version of CL 12152.
It also extends test to cover scenarios reported on issue #20445.
Some source files copied and renamed to make code clearer.