Cherry Zhang [Sat, 30 Jun 2018 03:09:34 +0000 (23:09 -0400)]
misc/wasm: make sure value ref id is unique
For each Javascript object that returns to Go as a js.Value, we
associate the ref id to it. But if this ref id is copied or
inherited to other object, it would mess up the ref-object
mapping.
In storeValue, make sure the object is indeed the one we are
storing. Otherwise allocate a new ref id.
Rob Pike [Sat, 30 Jun 2018 22:03:36 +0000 (08:03 +1000)]
testing/cover: improve comments on CoverBlock
The previous CL (https://go-review.googlesource.com/c/go/+/96756)
added comments that didn't really say much, but there is something
so say: what the units are and that they are indexed starting at 1.
Add a more helpful comment on the type, and also follow proper
style by using initial capitals and a period.
Change-Id: Id19cd5f392faf7c7bac034073f276cc770589075
Reviewed-on: https://go-review.googlesource.com/121875 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Alex Myasoedov [Sat, 18 Nov 2017 17:49:54 +0000 (12:49 -0500)]
regexp: examples for Regexp.FindIndex and Regexp.FindAllSubmatchIndex methods
This commit adds examples that demonstrate usage in a practical way.
Change-Id: I105baf610764c14a2c247cfc0b0c06f27888d377
Reviewed-on: https://go-review.googlesource.com/78635 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Ian Lance Taylor [Tue, 26 Jun 2018 21:40:51 +0000 (14:40 -0700)]
cmd/cgo: handle GCC 8 change in errors about constant initializers
Before GCC 8 C code like
const unsigned long long int neg = (const unsigned long long) -1;
void f(void) { static const double x = (neg); }
would get an error "initializer element is not constant". In GCC 8 and
later it does not.
Because a value like neg, above, can not be used as a general integer
constant, this causes cgo to conclude that it is a floating point
constant. The way that cgo handles floating point values then causes
it to get the wrong value for it: 18446744073709551615 rather than -1.
These are of course the same value when converted to int64, but Go
does not permit that kind of conversion for an out-of-range constant.
This CL side-steps the problem by treating floating point constants
with integer type as they would up being treated before GCC 8: as
variables rather than constants.
Fixes #26066
Change-Id: I6f2f9ac2fa8a4b8218481b474f0b539758eb3b79
Reviewed-on: https://go-review.googlesource.com/121035
Run-TryBot: Ian Lance Taylor <iant@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Brad Fitzpatrick [Wed, 20 Jun 2018 19:36:14 +0000 (19:36 +0000)]
time: clarify Unix, UnixNano, and In a bit
Fixes #23316
Change-Id: Ia1758b406d369bbfaace0bdfea02cd6f40735b65
Reviewed-on: https://go-review.googlesource.com/120060 Reviewed-by: Ian Lance Taylor <iant@golang.org>
http2: make Server send GOAWAY if Handler sets "Connection: close" header
https://golang.org/cl/121415
Fixes #20977
Change-Id: I9b8659b5191409ed007e2d911913763bcbabb7cc
Reviewed-on: https://go-review.googlesource.com/121695 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Ian Lance Taylor [Thu, 28 Jun 2018 23:45:28 +0000 (16:45 -0700)]
runtime: throw if the runtime panics with out of bounds index
If the runtime code panics due to a bad index or slice expression,
then throw instead of panicing. This will skip calls to recover and dump
the entire runtime stack trace. The runtime should never panic due to
an out of bounds index, and this will help with debugging if it does.
Austin Clements [Fri, 29 Jun 2018 18:56:48 +0000 (14:56 -0400)]
runtime: remap stack spans with MAP_STACK on OpenBSD
OpenBSD 6.4 is going to start requiring that the SP points to memory
that was mapped with MAP_STACK on system call entry, traps, and when
switching to the alternate signal stack [1]. Currently, Go doesn't map
any memory MAP_STACK, so the kernel quickly kills Go processes.
Fix this by remapping the memory that backs stack spans with
MAP_STACK, and re-remapping it without MAP_STACK when it's returned to
the heap.
Daniel Martí [Fri, 29 Jun 2018 15:59:04 +0000 (16:59 +0100)]
os: treat "${}" in Expand like in Go 1.10
CL 103055 made it so that invalid parameter expansions, like "$|", did
not make the dollar sign silently disappear.
A few edge cases were not taken into account, such as "${}" and "${",
which were now printing just "$". For consistency and to not break
existing programs, go back to eating up the characters when invalid
syntax is encountered.
For completeness, add a "$" test case too, even though its behavior is
unchanged by this CL.
Fixes #26135.
Change-Id: I5d25db9a8356dc6047a8502e318355113a99b247
Reviewed-on: https://go-review.googlesource.com/121636
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
Run-TryBot: Ian Lance Taylor <iant@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
bill_ofarrell [Thu, 28 Jun 2018 22:39:37 +0000 (18:39 -0400)]
bytes, strings: fix comparison of long byte slices on s390x
The existing implementation of bytes.Compare on s390x doesn't work properly for slices longer
than 256 elements. This change fixes that. Added tests for long strings and slices of bytes.
Brad Fitzpatrick [Fri, 29 Jun 2018 15:00:52 +0000 (15:00 +0000)]
net/http: update docs on Transport.DisableKeepAlives
Be super explicit that HTTP keep-alives != TCP keep-alives.
Fixes #26128
Change-Id: I77d74a6fe077259d996543f901a58aa3e49c1093
Reviewed-on: https://go-review.googlesource.com/121616 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Daniel Martí [Fri, 29 Jun 2018 16:46:17 +0000 (17:46 +0100)]
strings: add note for new Go developers to TrimLeft and TrimRight
If one quickly looks at the strings package godoc, reading the name
TrimLeft, one might think it removes a prefix from the string.
The function's godoc does explain its purpose, but it's apparent that it
is not clear enough, as there have been numerous raised issues about
this confusion: #12771 #14657 #18160 #19371 #20085 #25328 #26119. These
questions are also frequent elsewhere on the internet.
Add a very short paragraph to the godoc, to hopefully point new Go
developers in the right direction faster. Do the same thing for
TrimRight and TrimSuffix.
David Chase [Thu, 28 Jun 2018 20:22:21 +0000 (16:22 -0400)]
cmd/compile: make OpAddr depend on VarDef in storeOrder
Given a carefully constructed input, writebarrier would
split a block with the OpAddr in the first half and the
VarDef in the second half which ultimately leads to a
compiler crash because the scheduler is no longer able
to put them in the proper order.
To fix, recognize the implicit dependence of OpAddr on
the VarDef of the same symbol if any exists.
This fix was chosen over making OpAddr take a memory
operand to make the dependence explicit, because this
change is less invasive at this late part of the 1.11
release cycle.
Cherry Zhang [Fri, 29 Jun 2018 00:34:05 +0000 (20:34 -0400)]
cmd/compile: check SSAability in handling of INDEX of 1-element array
SSA can handle 1-element array, but only when the element type
is SSAable. When building SSA for INDEX of 1-element array, we
did not check the element type is SSAable. And when it's not,
it resulted in an unhandled SSA op.
Daniel Martí [Thu, 28 Jun 2018 16:18:01 +0000 (17:18 +0100)]
cmd/vet: don't run buildtag check when in vetxonly mode
The check was running in the loop that read source files in, much before
any of the other checks ran. Vetxonly makes vet exit early, but after
all the source files have been read.
To fix this, simply run the buildtag check along with all the other
checks that get run on specific syntax tree nodes.
Add a cmd/go test with go test -a, to ensure that the issue as reported
is fixed.
Fixes #26102.
Change-Id: If6e3b9418ffa8166c0f982668b0d10872283776a
Reviewed-on: https://go-review.googlesource.com/121395
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Ian Lance Taylor [Thu, 28 Jun 2018 22:44:41 +0000 (15:44 -0700)]
reflect: remove struct tags from unexported types
Before CL 4281055 in 2011, the reflect package was quite different.
rtype, then called commonType, was embedded in exported structs with
names like StructType. In order to avoid accidental conversions
between pointers to these public structs, which sometimes had
identical fields, the embedded commonType fields were tagged.
In CL 4281055 the formerly public structs were unexported, and all
access was done through the Type interface. At that point the field
tags in the reflect structs were no longer useful.
In Go 1.8 the language was changed to ignore struct field tags when
converting between types. This made the field tags in the reflect
structs doubly useless.
This CL simply removes them.
Fixes #20914
Change-Id: I9af4d6d0709276a91a6b6ee5323cad9dcd0cd0a0
Reviewed-on: https://go-review.googlesource.com/121475
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Brad Fitzpatrick [Thu, 28 Jun 2018 21:01:45 +0000 (21:01 +0000)]
net/http: make Server.Shutdown treat new connections as idle after 5 seconds
The Server distinguishes "new" vs "idle" connections. A TCP connection
from which no bytes have yet been written is "new". A connection that
has previously served a request and is in "keep-alive" state while
waiting for a second or further request is "idle".
The graceful Server.Shutdown historically only shut down "idle"
connections, with the assumption that a "new" connection was about to
read its request and would then shut down on its own afterwards.
But apparently some clients spin up connections and don't end up using
them, so we have something that's "new" to us, but browsers or other
clients are treating as "idle" to them.
This CL tweaks our heuristic to treat a StateNew connection as
StateIdle if it's been stuck in StateNew for over 5 seconds.
Fixes #22682
Change-Id: I01ba59a6ab67755ca5ab567041b1f54aa7b7da6f
Reviewed-on: https://go-review.googlesource.com/121419
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Russ Cox [Thu, 28 Jun 2018 04:05:46 +0000 (00:05 -0400)]
cmd/vet: fix ironic misuse of fmt.Sprintf
Move badf helper into top-level function so that prints from buildtag.go
are once again themselves printf-format-checked by vet.
Also, fix implementation, which was missing a ... in the Sprintf call and
produced messages like:
/Users/rsc/x_test.go:1: +build comment must appear before package clause and be followed by a blank line%!(EXTRA []interface {}=[])
Dmitri Shuralyov [Wed, 27 Jun 2018 19:37:19 +0000 (15:37 -0400)]
internal/syscall/unix: add build constraint to nonblocking_js.go
The intention was for this file to be constrained to both js and wasm,
but the build constraint was missing, causing it to be constrained only
to js because of the _js suffix in the filename.
Add a js,wasm build constraint. The js part is redundant, but specified
anyway to make it more visible and consistent with other similar files.
This issue was spotted while working on GopherJS, because it was causing
a conflict there (both nonblocking.go and nonblocking_js.go files were
being matched).
Peter Wu [Wed, 22 Nov 2017 19:27:20 +0000 (19:27 +0000)]
crypto/tls: add RSASSA-PSS support for handshake messages
This adds support for RSASSA-PSS signatures in handshake messages as
required by TLS 1.3. Even if TLS 1.2 is negotiated, it must support PSS
when advertised in the Client Hello (this will be done later as the
testdata will change).
Ian Lance Taylor [Wed, 27 Jun 2018 18:05:09 +0000 (11:05 -0700)]
os: when looping in RemoveAll, close and re-open directory
On some systems removing files can cause a directory to be re-shuffled,
so simply continuing to read files can cause us to miss some.
Close and re-open the directory when looping, to avoid that.
Read more files each time through the loop, to reduce the chance of
having to re-open.
Fixes #20841
Change-Id: I98a14774ca63786ad05ba5000cbdb01ad2884332
Reviewed-on: https://go-review.googlesource.com/121255
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
The current implementation does not generate wrappers for methods of
embedded non-interface types. We can only skip the wrapper if
kindDirectIface of the generated struct type matches kindDirectIface
of the embedded type. Panic if that is not the case.
It would be better to actually generate wrappers, but that can be done
later.
Richard Musiol [Wed, 27 Jun 2018 15:36:24 +0000 (17:36 +0200)]
syscall/js: add TypedArrayOf
The new function js.TypedArrayOf returns a JavaScript typed array for
a given slice.
https://developer.mozilla.org/en-US/docs/Web/JavaScript/Typed_arrays
This change also changes js.ValueOf to not accept a []byte any more.
Grégoire Delattre [Tue, 19 Dec 2017 18:42:12 +0000 (19:42 +0100)]
net: make concurrent resolver lookups independent
The current resolver uses a global lookupGroup which merges LookupIPAddr
calls together for lookups for the same hostname if used concurrently.
As a result only one of the resolvers is actually used to perform the
DNS lookup but the result is shared by all the resolvers.
This commit limits the scope of the lookupGroup to the resolver itself
allowing each resolver to make its own requests without sharing the
result with other resolvers.
Brad Fitzpatrick [Wed, 27 Jun 2018 16:50:39 +0000 (16:50 +0000)]
net/http/httptrace: add clarification never added to CL 67430
Updates #19761
Change-Id: Iac3bd4c40002f8e348452b50bff54dee3210d447
Reviewed-on: https://go-review.googlesource.com/121236 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Ian Lance Taylor [Wed, 27 Jun 2018 00:14:43 +0000 (17:14 -0700)]
mime/quotedprintable: accept bytes >= 0x80
RFC 2045 doesn't permit non-ASCII bytes, but some systems send them
anyhow. With this change, we accept them. This does make it harder to
validate quotedprintable data, but on balance this seems like the best
approach given the existence of systems that generate invalid data.
Ian Lance Taylor [Wed, 27 Jun 2018 14:08:41 +0000 (07:08 -0700)]
net: limit concurrent threads to limit on file descriptors
At least on Darwin, if getaddrinfo can't open a file descriptor it
returns EAI_NONAME ("no such host") rather than a meaningful error.
Limit the number of concurrent getaddrinfo calls to the number of file
descriptors we can open, to make that meaningless error less likely.
We don't apply the same limit to Go lookups, because for that we will
return a meaningful "too many open files" error.
Meir Fischer [Sun, 8 Oct 2017 19:25:28 +0000 (15:25 -0400)]
net/http/httptrace: expose request headers for http/1.1
Some headers, which are set or modified by the http library,
are not written to the standard http.Request.Header and are
not included as part of http.Response.Request.Header.
Exposing all headers alleviates this problem.
This is not a complete solution to 19761 since it does not have http/2 support.
Cherry Zhang [Sat, 23 Jun 2018 17:23:52 +0000 (13:23 -0400)]
cmd/compile: fold offset into address on Wasm
On Wasm, the offset was not folded into LoweredAddr, so it was
not rematerializeable. This led to the address-taken operation
in some cases generated too early, before the local variable
becoming live. The liveness code thinks the variable live when
the address is taken, then backs it up to live at function
entry, then complains about it, because nothing other than
arguments should be live on entry.
This CL folds the offset into the address operation, so it is
rematerializeable and so generated right before use, after the
variable actually becomes live.
It might be possible to relax the liveness code not to think a
variable live when its address being taken, but until the address
actually being used. But it would be quite complicated. As we're
late in Go 1.11 freeze, it would be better not to do it. Also,
I think the address operation is rematerializeable now on all
architectures, so this is probably less necessary.
This may also be a slight optimization, as the address+offset is
now rematerializeable, which can be generated on the Wasm stack,
without using any "registers" which are emulated by local
variables on Wasm. I don't know how to do benchmarks on Wasm. At
least, cmd/go binary size shrinks 9K.
LE Manh Cuong [Tue, 26 Jun 2018 06:48:05 +0000 (13:48 +0700)]
cmd/link: document limitation of -X
Fixes #26042
Change-Id: Ica16f14a65c03659a19926852cca5e554c99baf1
Reviewed-on: https://go-review.googlesource.com/120935 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Ian Lance Taylor [Tue, 26 Jun 2018 22:57:35 +0000 (15:57 -0700)]
mime/multipart: restore 1.9 handling of missing/empty form-data file name
Revert the code changes of CL 96975 and CL 70931, but keep the tests,
appropriately modified for the code changes. This restores the 1.9
handling of form-data entries with missing or empty file names.
Changing the handling of this simply confused existing programs for no
useful benefit. Go back to the old behavior.
Updates #19183
Fixes #24041
Change-Id: I4ebc32433911e6360b9fd79d8f63a6d884822e0e
Reviewed-on: https://go-review.googlesource.com/121055
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Robert Griesemer [Mon, 25 Jun 2018 23:21:59 +0000 (16:21 -0700)]
go/types: rename NewInterface2 to NewInterfaceType
NewInterface2 was introduced with https://go-review.googlesource.com/114317
which fixed #25301. Changed the name to NewInterfaceType to better match
Go naming styles, per discussion with @josharian, @iant, et al.
Change-Id: Ifa4708a5efd4f708295b33c3d20fdc5812e1b4fc
Reviewed-on: https://go-review.googlesource.com/120875 Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Richard Musiol [Sun, 24 Jun 2018 15:23:38 +0000 (17:23 +0200)]
syscall/js: use stable references to JavaScript values
This commit changes how JavaScript values are referenced by Go code.
After this change, a JavaScript value is always represented by the same
ref, even if passed multiple times from JavaScript to Go. This allows
Go's == operator to work as expected on js.Value (strict equality).
Additionally, the performance of some operations of the syscall/js
package got improved by saving additional roundtrips to JavaScript code.
Vladimir Kuzmin [Thu, 21 Jun 2018 05:19:56 +0000 (22:19 -0700)]
cmd/compile: map delete should clear value always
Map delete must clear value every time because
newly added map optimizations of compound-assignment
operators (CL #91557) rely on this behavior of map delete.
It slows down map delete operation for non-reference types:
Andrew [Mon, 25 Jun 2018 20:24:37 +0000 (20:24 +0000)]
cmd/dist: increase timeout scale to 3 for windows
cmd/go can sometimes take up to 400s on windows due
to various issues (disk I/O on builders being the
latest cause). Increase the timeout scale to account
for this.
David Chase [Mon, 25 Jun 2018 19:26:45 +0000 (15:26 -0400)]
cmd/compile: avoid remainder in loopbce when increment=0
For non-unit increment, loopbce checks to see if the
increment evenly divides the difference between (constant)
loop start and end. This test panics when the increment
is zero.
Fix: check for zero, if found, don't optimize the loop.
Also added missing copyright notice to loopbce.go.
Guilherme Goncalves [Sat, 23 Jun 2018 22:32:26 +0000 (22:32 +0000)]
net/http: document and test behavior of ServeMux with ports
Beginning on Go 1.9, ServeMux has been dropping the port number from the Host
header and in the path pattern. This commit explicitly mentions the change in
behavior and adds a simple test case to ensure consistency.
Nikhil Benesch [Sat, 23 Jun 2018 05:15:19 +0000 (01:15 -0400)]
runtime: respect timeout in semasleep on Darwin
semasleep on Darwin was refactored in https://golang.org/cl/118736 to
use the pthread_cond_timedwait function from libc. The new code
incorrectly assumed that pthread_cond_timedwait took a timeout relative
to the current time, when it in fact it takes a timeout specified in
absolute time. semasleep thus specified a timeout well in the past,
causing it to immediately exceed the timeout and spin hot. This was the
source of a large performance hit to CockroachDB (#26019).
Adjust semasleep to instead call pthread_cond_timedwait_relative_np,
which properly interprets its timeout parameter as relative to the
current time.
pthread_cond_timedwait_relative_np is non-portable, but using
pthread_cond_timedwait correctly would require two calls to
gettimeofday: one in the runtime package to convert the relative timeout
to absolute time, then another in the pthread library to convert back to
a relative offset [0], as the Darwin kernel expects a relative offset.
Heschi Kreinick [Wed, 20 Jun 2018 20:45:25 +0000 (16:45 -0400)]
cmd/link: support DWARF compression on Darwin
We want to compress DWARF even on macOS, but the native toolchain isn't
going to understand it. Add a flag that can be used to disable
compression, then add Darwin to the whitelist used during internal
linking.
Unlike GNU ld, the Darwin linker doesn't have a handy linker flag to do
compression. But since we're already doing surgery to put the DWARF in
the output executable in the first place, compressing it at the same
time isn't unduly difficult. This does have the slightly odd effect of
compressing some Apple proprietary debug sections, which absolutely
nothing will understand. Leaving them uncompressed didn't make much
sense, though, since I doubt they're useful without (say) __debug_info.
Brad Fitzpatrick [Fri, 22 Jun 2018 17:36:12 +0000 (17:36 +0000)]
cmd/dist: skip non-std tests on js/wasm
After the std tests, most of the rest of the tests aren't applicable
to js/wasm. (anything with -cpu=>1, cgo, etc)
Skip them all for now. We can incrementally re-enable them over time
as the js/wasm port is fleshed out. But let's get the builder column
black again so we can enable trybots and keep it black.
Matthew Dempsky [Thu, 21 Jun 2018 23:12:17 +0000 (16:12 -0700)]
cmd/compile: fix compile failure for lazily resolved shadowed types
If expanding an inline function body required lazily expanding a
package-scoped type whose identifier was shadowed within the function
body, the lazy expansion would instead overwrite the local symbol
definition instead of the package-scoped symbol. This was due to
importsym using s.Def instead of s.PkgDef.
Unfortunately, this is yet another consequence of the current awkward
scope handling code.
Passes toolstash-check.
Fixes #25984.
Change-Id: Ia7033e1749a883e6e979c854d4b12b0b28083dd8
Reviewed-on: https://go-review.googlesource.com/120456
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: David Chase <drchase@google.com> Reviewed-by: Robert Griesemer <gri@golang.org>
Dmitry Vyukov [Fri, 22 Jun 2018 09:00:13 +0000 (11:00 +0200)]
sync: fix deficiency in RWMutex race annotations
Remove unnecessary race.Release annotation from Unlock.
For RWMutex we want to establish the following happens-before (HB) edges:
1. Between Unlock and the subsequent Lock.
2. Between Unlock and the subsequent RLock.
3. Between batch of RUnlock's and the subsequent Lock.
1 is provided by Release(&rw.readerSem) in Unlock and Acquire(&rw.readerSem) in Lock.
2 is provided by Release(&rw.readerSem) in Unlock and Acquire(&rw.readerSem) in RLock.
3 is provided by ReleaseMerge(&rw.writerSem) in RUnlock in Acquire(&rw.writerSem) in Lock,
since we want to establish HB between a batch of RUnlock's this uses ReleaseMerge instead of Release.
Release(&rw.writerSem) in Unlock is simply not needed.
FWIW this is also how C++ tsan handles mutexes, not a proof but at least something.
Making 2 implementations consistent also simplifies any kind of reasoning against both of them.
Since this only affects performance, a reasonable test is not possible.
Everything should just continue to work but slightly faster.
Credit for discovering this goes to Jamie Liu.
Change-Id: Ice37d29ecb7a5faed3f7781c38dd32c7469b2735
Reviewed-on: https://go-review.googlesource.com/120495
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Zhou Peng [Fri, 22 Jun 2018 05:31:06 +0000 (05:31 +0000)]
runtime: fix comments style typo
Code comments should have a space between comments characters and
actual words.
Change-Id: I6274baf1fc09b37a32ec6c69ddbb8edca9eb5469
Reviewed-on: https://go-review.googlesource.com/120475 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Daniel Theophanes [Thu, 21 Jun 2018 17:41:26 +0000 (10:41 -0700)]
text/template/parse: undo breaking API changes
golang.org/cl/84480 altered the API for the parse package for
clarity and consistency. However, the changes also broke the
API for consumers of the package. This CL reverts the API
to the previous spelling, adding only a single new exported
symbol.
Fixes #25968
Change-Id: Ieb81054b61eeac7df3bc3864ef446df43c26b80f
Reviewed-on: https://go-review.googlesource.com/120355 Reviewed-by: Daniel Martí <mvdan@mvdan.cc> Reviewed-by: Rob Pike <r@golang.org>
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Austin Clements [Thu, 21 Jun 2018 20:56:13 +0000 (16:56 -0400)]
runtime: avoid recursive panic on bad lock count
Currently, if lock or unlock calls throw because the g.m.lock count is
corrupted, we're unlikely to get a stack trace because startpanic_m
will itself attempt to acquire a lock, causing a recursive failure.
Avoid this by forcing the g.m.locks count to a sane value if it's
currently bad.
This might be enough to get a stack trace from #25128.
Change-Id: I52d7bd4717ffae94a821f4249585f3eb6cd5aa41
Reviewed-on: https://go-review.googlesource.com/120416 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Matthew Dempsky [Thu, 21 Jun 2018 22:35:32 +0000 (15:35 -0700)]
cmd/compile: fix recursive inimport handling
expandDecl can be called recursively, so it's not an appropriate place
to clean inimport. Instead, move this up to resolve, along with an
appropriate recursion check.
Passes toolstash-check.
Change-Id: I138d37b057dcc6525c780b4b3fbaa5e97f99655b
Reviewed-on: https://go-review.googlesource.com/120455
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Terin Stock [Wed, 20 Jun 2018 21:00:04 +0000 (14:00 -0700)]
flag: add a Value example
Change-Id: I579cc9f4f8e5be5fd6447a99614797ab7bc53611
Reviewed-on: https://go-review.googlesource.com/120175
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Paul Jolly [Thu, 21 Jun 2018 17:30:04 +0000 (18:30 +0100)]
misc/wasm: fix permissions on wasm_exec.js
Currently wasm_exec.js is executable (0755) yet has no interpreter.
Indeed wasm_exec.js is only ever used as an argument to Node or loaded
via a <script> tag in a browser-loaded HTML file. Hence the execute
mode bits are superfluous and simply serve to clutter your PATH if
$GOROOT/misc/wasm is on your PATH (as is required if you want to run go
test syscall/js).
Change-Id: I279e2457094f8a12b9bf380ad7f1a9f47b22fc96
Reviewed-on: https://go-review.googlesource.com/120435
Run-TryBot: Paul Jolly <paul@myitcv.org.uk> Reviewed-by: Daniel Martí <mvdan@mvdan.cc> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Andrei Tudor Călin [Thu, 21 Jun 2018 17:11:34 +0000 (19:11 +0200)]
internal/poll: use more fine-grained locking in Splice
The previous code acquired a read lock on src and a write lock on
dst for the entire duration of Splice. This resulted in deadlock,
in a situation akin to the following:
Splice is blocking, waiting to read from src.
The caller tries to close dst from another goroutine, but Close on
dst blocks in runtime.semacquire, because Splice is still holding a
write lock on it, and the Close didn't unblock any I/O.
The caller cannot unblock the read side of Splice through other means,
because they are stuck waiting in dst.Close().
Use more fine-grained locking instead: acquire the read lock on src
just before trying to splice from the source socket to the pipe,
and acquire the write lock on dst just before trying to splice from
the pipe to the destination socket.
Fixes #25985
Change-Id: I264c91c7a69bb6c5e28610e2bd801244804cf86d
Reviewed-on: https://go-review.googlesource.com/120317
Run-TryBot: Aram Hăvărneanu <aram@mgk.ro>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Wei Xiao [Fri, 3 Nov 2017 02:05:28 +0000 (02:05 +0000)]
cmd/compile: improve atomic add intrinsics with ARMv8.1 new instruction
ARMv8.1 has added new instruction (LDADDAL) for atomic memory operations. This
CL improves existing atomic add intrinsics with the new instruction. Since the
new instruction is only guaranteed to be present after ARMv8.1, we guard its
usage with a conditional on CPU feature.
Performance result on ARMv8.1 machine:
name old time/op new time/op delta
Xadd-224 1.05µs ± 6% 0.02µs ± 4% -98.06% (p=0.000 n=10+8)
Xadd64-224 1.05µs ± 3% 0.02µs ±13% -98.10% (p=0.000 n=9+10)
[Geo mean] 1.05µs 0.02µs -98.08%
Performance result on ARMv8.0 machine:
name old time/op new time/op delta
Xadd-46 538ns ± 1% 541ns ± 1% +0.62% (p=0.000 n=9+9)
Xadd64-46 505ns ± 1% 508ns ± 0% +0.48% (p=0.003 n=9+8)
[Geo mean] 521ns 524ns +0.55%
Emmanuel T Odeke [Mon, 21 May 2018 19:57:53 +0000 (12:57 -0700)]
net/http: avoid deferred unlock in ServeMux.shouldRedirect
CL 96575 introduced concurrency protection for
ServeMux.shouldRedirect with a read lock and deferred unlock.
However, the change produced a noticeable regression.
Instead add the suffix "RLocked" to the function name to
declare that we should hold the read lock as a pre-requisite
before calling it, hence avoiding the defer altogether.
Benchmarks:
name old time/op new time/op delta
ServeMux-8 63.3µs ± 0% 54.6µs ± 0% -13.74% (p=0.000 n=9+9)
ServeMux_SkipServe-8 41.4µs ± 2% 32.7µs ± 1% -21.05% (p=0.000 n=10+10)
name old alloc/op new alloc/op delta
ServeMux-8 17.3kB ± 0% 17.3kB ± 0% ~ (all equal)
ServeMux_SkipServe-8 0.00B 0.00B ~ (all equal)
name old allocs/op new allocs/op delta
ServeMux-8 360 ± 0% 360 ± 0% ~ (all equal)
ServeMux_SkipServe-8 0.00 0.00 ~ (all equal)
Tobias Klauser [Wed, 20 Jun 2018 13:41:37 +0000 (15:41 +0200)]
syscall: check faccessat flags parameter on Linux
Port CL 119495 from golang.org/x/sys/unix to the syscall package.
Currently Linux faccessat(2) syscall implementation doesn't support the
flags parameter. As per the discussion in #25845, permit the same flags
as glibc [1].
Brad Fitzpatrick [Wed, 20 Jun 2018 18:25:54 +0000 (18:25 +0000)]
mime: change *.js mime type to application/javascript, not x-javascript
We delayed doing this for 4 years for fear that it might break something,
but it was standardized (RFC 4329) 12 years ago, and the default in Debian
and other places is correct:
Agniva De Sarker [Tue, 19 Jun 2018 14:57:08 +0000 (20:27 +0530)]
cmd/go/internal: add a note about GOCACHE=off
Fixes #25928
Change-Id: I1401ecc54af26eeeee648bb8eeb5d2d3566fa60c
Reviewed-on: https://go-review.googlesource.com/119695 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Peter Wu [Wed, 22 Nov 2017 18:25:20 +0000 (18:25 +0000)]
crypto/tls: consolidate signatures handling in SKE and CV
ServerKeyExchange and CertificateVerify can share the same logic for
picking a signature algorithm (based on the certificate public key and
advertised algorithms), selecting a hash algorithm (depending on TLS
version) and signature verification.
Refactor the code to achieve code reuse, have common error checking
(especially for intersecting supported signature algorithms) and to
prepare for addition of new signature algorithms. Code should be easier
to read since version-dependent logic is concentrated at one place.
Ian Lance Taylor [Tue, 19 Jun 2018 23:53:18 +0000 (16:53 -0700)]
go/internal/gccgoimporter: read export data from archives
When used with the go tool, gccgo will normally generate archive files.
This change teaches the gccgoimporter package how to read the export
data from an archive.
This is needed by, for example, cmd/vet, when typechecking packages.
Change-Id: I21267949a7808cd81c0042af425c774a4ff7d82f
Reviewed-on: https://go-review.googlesource.com/119895
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>