This CL makes reflect require that values be assignable to the target type
in exactly the same places where that is the rule in Go. It also adds
the Implements and AssignableTo methods so that callers can check
the types themselves so as to avoid a panic.
Before this CL, reflect required strict type identity.
This CL expands Call to accept and correctly marshal arbitrary
argument lists for variadic functions; it introduces CallSlice for use
in the case where the slice for the variadic argument is already known.
Fixes#327.
Fixes#1212.
R=r, dsymonds
CC=golang-dev
https://golang.org/cl/4439058
This CL makes it possible to resolve DNS names on OS X
without offending the Application-Level Firewall.
It also means that cross-compiling from one operating
system to another is no longer possible when using
package net, because cgo needs to be able to sniff around
the local C libraries. We could special-case this one use
and check in generated files, but it seems more trouble
than it's worth. Cross compiling is dead anyway.
It is still possible to use either GOARCH=amd64 or GOARCH=386
on typical Linux and OS X x86 systems.
It is also still possible to build GOOS=linux GOARCH=arm on
any system, because arm is for now excluded from this change
(there is no cgo for arm yet).
R=iant, r, mikioh
CC=golang-dev
https://golang.org/cl/4437053
* Accept armored private key blocks
* If an armored block is missing, return an InvalidArgumentError,
rather than ignoring it.
* If every key in a block is skipped due to being unsupported,
return the last unsupported error.
* Include the numeric type of unsupported public keys.
* Don't assume that the self-signature comes immediately after the
user id packet.
R=bradfitzgo
CC=golang-dev
https://golang.org/cl/4434048
This pulls in changes that should have been in 3faf9d0c10c0, but
weren't because x509.go was part of another changelist.
TBR=bradfitzgo
R=bradfitzgo
CC=golang-dev
https://golang.org/cl/4433056
People have a need to verify certificates in situations other than TLS
client handshaking. Thus this CL moves certificate verification into
x509 and expands its abilities.
R=bradfitzgo
CC=golang-dev
https://golang.org/cl/4407046
* Reduces malloc counts during gob encoder/decoder test from 6/6 to 3/5.
The current reflect uses Set to mean two subtly different things.
(1) If you have a reflect.Value v, it might just represent
itself (as in v = reflect.NewValue(42)), in which case calling
v.Set only changed v, not any other data in the program.
(2) If you have a reflect Value v derived from a pointer
or a slice (as in x := []int{42}; v = reflect.NewValue(x).Index(0)),
v represents the value held there. Changing x[0] affects the
value returned by v.Int(), and calling v.Set affects x[0].
This was not really by design; it just happened that way.
The motivation for the new reflect implementation was
to remove mallocs. The use case (1) has an implicit malloc
inside it. If you can do:
v := reflect.NewValue(0)
v.Set(42)
i := v.Int() // i = 42
then that implies that v is referring to some underlying
chunk of memory in order to remember the 42; that is,
NewValue must have allocated some memory.
Almost all the time you are using reflect the goal is to
inspect or to change other data, not to manipulate data
stored solely inside a reflect.Value.
This CL removes use case (1), so that an assignable
reflect.Value must always refer to some other piece of data
in the program. Put another way, removing this case would
make
v := reflect.NewValue(0)
v.Set(42)
as illegal as
0 = 42.
It would also make this illegal:
x := 0
v := reflect.NewValue(x)
v.Set(42)
for the same reason. (Note that right now, v.Set(42) "succeeds"
but does not change the value of x.)
If you really wanted to make v refer to x, you'd start with &x
and dereference it:
x := 0
v := reflect.NewValue(&x).Elem() // v = *&x
v.Set(42)
It's pretty rare, except in tests, to want to use NewValue and then
call Set to change the Value itself instead of some other piece of
data in the program. I haven't seen it happen once yet while
making the tree build with this change.
For the same reasons, reflect.Zero (formerly reflect.MakeZero)
would also return an unassignable, unaddressable value.
This invalidates the (awkward) idiom:
pv := ... some Ptr Value we have ...
v := reflect.Zero(pv.Type().Elem())
pv.PointTo(v)
which, when the API changed, turned into:
pv := ... some Ptr Value we have ...
v := reflect.Zero(pv.Type().Elem())
pv.Set(v.Addr())
In both, it is far from clear what the code is trying to do. Now that
it is possible, this CL adds reflect.New(Type) Value that does the
obvious thing (same as Go's new), so this code would be replaced by:
pv := ... some Ptr Value we have ...
pv.Set(reflect.New(pv.Type().Elem()))
The changes just described can be confusing to think about,
but I believe it is because the old API was confusing - it was
conflating two different kinds of Values - and that the new API
by itself is pretty simple: you can only Set (or call Addr on)
a Value if it actually addresses some real piece of data; that is,
only if it is the result of dereferencing a Ptr or indexing a Slice.
If you really want the old behavior, you'd get it by translating:
v := reflect.NewValue(x)
into
v := reflect.New(reflect.Typeof(x)).Elem()
v.Set(reflect.NewValue(x))
Gofix will not be able to help with this, because whether
and how to change the code depends on whether the original
code meant use (1) or use (2), so the developer has to read
and think about the code.
You can see the effect on packages in the tree in
https://golang.org/cl/4423043/.
R=r
CC=golang-dev
https://golang.org/cl/4435042
NewRequest will save a lot of boilerplate code.
This also updates some docs on Request.Write and
adds some tests.
R=rsc, petar-m, r
CC=golang-dev
https://golang.org/cl/4406047
Ubuntu and/or GNOME have some bug that likes
to set the "http_proxy" environment variable
and forgets to unset it. This is annoying
to debug. Be clear in the error message that
a proxy was in use.
R=rsc
CC=golang-dev
https://golang.org/cl/4409045
. Missing declaration of runtime.brk_();
. Argument v in runtime.SysReserve() is not used;
(I'd prefer a Plan 9-type solution...)
R=golang-dev, r, r2
CC=golang-dev
https://golang.org/cl/4368076
We already had support on the client side. I also changed the name of
the flag in the ServerHello structure to match the name of the same
flag in the ClientHello (ocspStapling).
R=bradfitzgo
CC=golang-dev
https://golang.org/cl/4408044
This fixes our http behavior (even if Handlers forget to
consume a request body, we do it for them before we send
their response header), fixes the racy TestServerExpect,
and adds TestServerConsumesRequestBody.
With GOMAXPROCS>1, the http tests now seem race-free.
R=rsc
CC=golang-dev
https://golang.org/cl/4419042
The list elements are already being allocated out of a
single memory buffer. We can drop the Link* pointer
following and the memory it requires, replacing it with
index operations.
The change also keeps a channel from containing a pointer
back into its own allocation block, which would create a
cycle. Blocks involved in cycles are not guaranteed to be
finalized properly, and channels depend on finalizers to
free OS-level locks on some systems. The self-reference
was keeping channels from being garbage collected.
runtime-gdb.py will need to be updated in order to dump
the content of buffered channels with the new data structure.
Fixes#1676.
R=ken2, r
CC=golang-dev
https://golang.org/cl/4411045
This mostly adds Expect 100-continue tests (from
the perspective of server correctness) that were
missing before.
It also fixes a few missing cases that will
probably never come up in practice, but it's nice
to have handled correctly.
Proper 100-continue client support remains a TODO.
R=rsc, bradfitzwork
CC=golang-dev
https://golang.org/cl/4399044
This CL is only cut-and-paste, moving code around.
Moving it in a separate CL should simplify the diffs in later CLs.
There are three patterns here.
1. A function like
func (v Value) M() (...) {
return v.panicIfNot(K).(*kValue).M()
}
becomes
func (v Value) M() (...) {
vv := v.panicIfNot(K).(*kValue)
// body of (*kValue).M, s/v./vv./g
}
2. A function like
func (v Value) M() (...) {
return v.panicIfNots(kList).(mer).M()
}
becomes
func (v Value) M() (...) {
switch vv := v.panicIfNots(kList).(type) {
case *k1Value:
// body of (*k1Value).M, s/v./vv./g
case *k2Value:
// body of (*k2Value).M, s/v./vv./g
...
}
panic("not reached")
}
3. The rewrite of Value.Set follows 2, but each case
is built from the bodies of (*kValue).SetValue and (*kValue).Set.
func (v *kValue) SetValue(x Value) {
v.Set(x.panicIfNot(K).(*kValue)
}
func (v *kValue) Set(x *kValue) {
... body
}
becomes, in the switch from 2,
case *kValue:
xx := x.panicIfNot(K).(*kValue)
... body, s/v./vv./g; s/x./xx./g
R=r
CC=golang-dev
https://golang.org/cl/4398044
This CL changes the behavior of 'make install' and 'make test'
in the src/cmd directory and the src/pkg directory to have
each recursive make clean up after itself immediately.
It does the same in test/run, removing $F.$A and $A.out
(the common byproducts) between runs.
On machines with slow disks and aggressive kernel caching,
cleaning up immediately can mean that the intermediate
objects never get written to disk.
This change eliminates almost all the disk waiting during
all.bash on my laptop (a Thinkpad X201s with an SSD running Linux).
147.50u 19.95s 277.34r before
148.53u 21.64s 179.59r after
R=golang-dev, r, iant2
CC=golang-dev
https://golang.org/cl/4413042
It matches encoding/line exactly and the tests are copied from there.
If we land this, then encoding/line will get marked as deprecated then
deleted in time.
R=rsc, rog, peterGo
CC=golang-dev
https://golang.org/cl/4389046
Note that declarations.golden is not using spaces for alignment (so
that the alignment tabs are visible) which is why this change affects
the test cases significantly. gofmt uses spaces for alignment (by default)
and only tabs for indentation.
gofmt -w src misc (no changes)
Fixes#1673.
R=iant
CC=golang-dev
https://golang.org/cl/4388044
Type is now an interface that implements all the possible type methods.
Instead of a type switch on a reflect.Type t, switch on t.Kind().
If a method is invoked on the wrong kind of type (for example,
calling t.Field(0) when t.Kind() != Struct), the call panics.
There is one method renaming: t.(*ChanType).Dir() is now t.ChanDir().
Value is now a struct value that implements all the possible value methods.
Instead of a type switch on a reflect.Value v, switch on v.Kind().
If a method is invoked on the wrong kind of value (for example,
calling t.Recv() when t.Kind() != Chan), the call panics.
Since Value is now a struct, not an interface, its zero value
cannot be compared to nil. Instead of v != nil, use v.IsValid().
Instead of other uses of nil as a Value, use Value{}, the zero value.
Many methods have been renamed, most due to signature conflicts:
OLD NEW
v.(*ArrayValue).Elem v.Index
v.(*BoolValue).Get v.Bool
v.(*BoolValue).Set v.SetBool
v.(*ChanType).Dir v.ChanDir
v.(*ChanValue).Get v.Pointer
v.(*ComplexValue).Get v.Complex
v.(*ComplexValue).Overflow v.OverflowComplex
v.(*ComplexValue).Set v.SetComplex
v.(*FloatValue).Get v.Float
v.(*FloatValue).Overflow v.OverflowFloat
v.(*FloatValue).Set v.SetFloat
v.(*FuncValue).Get v.Pointer
v.(*InterfaceValue).Get v.InterfaceData
v.(*IntValue).Get v.Int
v.(*IntValue).Overflow v.OverflowInt
v.(*IntValue).Set v.SetInt
v.(*MapValue).Elem v.MapIndex
v.(*MapValue).Get v.Pointer
v.(*MapValue).Keys v.MapKeys
v.(*MapValue).SetElem v.SetMapIndex
v.(*PtrValue).Get v.Pointer
v.(*SliceValue).Elem v.Index
v.(*SliceValue).Get v.Pointer
v.(*StringValue).Get v.String
v.(*StringValue).Set v.SetString
v.(*UintValue).Get v.Uint
v.(*UintValue).Overflow v.OverflowUint
v.(*UintValue).Set v.SetUint
v.(*UnsafePointerValue).Get v.Pointer
v.(*UnsafePointerValue).Set v.SetPointer
Part of the motivation for this change is to enable a more
efficient implementation of Value, one that does not allocate
memory during most operations. To reduce the size of the CL,
this CL's implementation is a wrapper around the old API.
Later CLs will make the implementation more efficient without
changing the API.
Other CLs to be submitted at the same time as this one
add support for this change to gofix (4343047) and update
the Go source tree (4353043).
R=gri, iant, niemeyer, r, rog, gustavo, r2
CC=golang-dev
https://golang.org/cl/4281055
This CL defines a new, more Go-like representation of
Go types (different structs for different types as
opposed to a single Type node). It also implements
an ast.Importer for object/archive files generated
by the gc compiler tool chain. Besides the individual
type structs, the main difference is the handling of
named types: In the old world, a named type had a
non-nil *Object pointer but otherwise looked no
different from other types. In this new model, named
types have their own representation types.Name. As
a result, resolving cycles is a bit simpler during
construction, at the cost of having to deal with
types.Name nodes explicitly later. It remains to be
seen if this is a good approach. Nevertheless, code
involving types reads more nicely and benefits from
full type checking. Also, the representation seems
to more closely match the spec wording.
Credits: The original version of the gc importer was
written by Evan Shaw (chickencha@gmail.com). The new
version in this CL is based largely on Evan's original
code but contains bug fixes, a few simplifications,
some restructuring, and was adjusted to use the
new type hierarchy. I have added a comprehensive test
that imports all packages found under $GOROOT/pkg (with
a 3s time-out to limit the run-time of the test). Run
gotest -v for details.
The original version of ExportData (exportdata.go) was
written by Russ Cox (rsc@golang.org). The current version
is returning the internal buffer positioned at the beginning
of the export data instead of printing the export data to
stdout.
With the new types package, the existing in-progress
typechecker package is deprecated. I will delete it
once all functionality has been brought over.
R=eds, rog, rsc
CC=golang-dev
https://golang.org/cl/4314054
* tweak mksyscall*.pl to be more gofmt-compatible.
* add mkall.sh -syscalls option.
* add sys/mman.h constants on OS X
R=r, eds, niemeyer
CC=golang-dev
https://golang.org/cl/4369044
Moved the details of how to read a directory
and how to parse the results behind the new
syscall functions ReadDirent and ParseDirent.
Now os needs just one copy of Readdirnames
for the three Unix variants, and it no longer
imports "unsafe".
R=r, r2
CC=golang-dev
https://golang.org/cl/4368048
Not committed to this but it sure makes
the output easier to skim. With this CL:
$ make
install runtime
install sync/atomic
install sync
install unicode
install utf16
install syscall
install os
...
install ../cmd/govet
install ../cmd/goyacc
install ../cmd/hgpatch
$ make test
test archive/tar
test archive/zip
test asn1
test big
test bufio
...
test path
test path/filepath
TEST FAIL reflect
gotest
rm -f _test/reflect.a
6g -o _gotest_.6 deepequal.go type.go value.go
rm -f _test/reflect.a
gopack grc _test/reflect.a _gotest_.6
all_test.go:210: invalid type assertion: reflect.NewValue(tt.i).(*StructValue) (non-interface type reflect.Value on left)
all_test.go:217: cannot type switch on non-interface value v (type reflect.Value)
all_test.go:218: undefined: IntValue
all_test.go:221: cannot use 132 (type int) as type reflect.Value in function argument
all_test.go:223: cannot use 8 (type int) as type reflect.Value in function argument
all_test.go:225: cannot use 16 (type int) as type reflect.Value in function argument
all_test.go:227: cannot use 32 (type int) as type reflect.Value in function argument
all_test.go:229: cannot use 64 (type int) as type reflect.Value in function argument
all_test.go:231: undefined: UintValue
all_test.go:234: cannot use 132 (type int) as type reflect.Value in function argument
all_test.go:234: too many errors
gotest: "/Users/rsc/g/go/bin/6g -I _test -o _xtest_.6 all_test.go tostring_test.go" failed: exit status 1
make[1]: *** [test] Error 2
make: *** [reflect.test] Error 1
R=r, r2
CC=golang-dev
https://golang.org/cl/4343046
- used to be only for standard log, not for user-built.
- there were no getters.
Also rearrange the code a little so we can avoid allocating
a buffer on every call. Logging is expensive but we should
avoid unnecessary cost.
This should have no effect on existing code.
R=rsc
CC=golang-dev
https://golang.org/cl/4363045
The CRT is symmetrical in the case of two variables and I picked a
different form from PKCS#1.
R=golang-dev, rsc1
CC=golang-dev
https://golang.org/cl/4381041
We replace the current Open with:
OpenFile(name, flag, perm) // same as old Open
Open(name) // same as old Open(name, O_RDONLY, 0)
Create(name) // same as old Open(name, O_RDWR|O_TRUNC|O_CREAT, 0666)
This CL includes a gofix module and full code updates: all.bash passes.
(There may be a few comments I missed.)
The interesting packages are:
gofix
os
Everything else is automatically generated except for hand tweaks to:
src/pkg/io/ioutil/ioutil.go
src/pkg/io/ioutil/tempfile.go
src/pkg/crypto/tls/generate_cert.go
src/cmd/goyacc/goyacc.go
src/cmd/goyacc/units.y
R=golang-dev, bradfitzwork, rsc, r2
CC=golang-dev
https://golang.org/cl/4357052
Amazon S3 sends Transfer-Encoding "chunked"
on its 404 responses to HEAD requests for
missing objects.
We weren't ignoring the Transfer-Encoding
and were thus interpretting the subsequent
response headers as a chunk header from the
previous responses body (but a HEAD response
can't have a body)
R=rsc, adg
CC=golang-dev
https://golang.org/cl/4346050
A connection shouldn't be made available
for re-use until its body has been consumed.
(except in the case of pipelining, which isn't
implemented yet)
This CL fixes some issues seen with heavy load
against Amazon S3.
Subtle implementation detail: to prevent a race
with the client requesting a new connection
before previous one is returned, we actually
have to call putIdleConnection _before_ we
return from the final Read/Close call on the
http.Response.Body.
R=rsc, adg
CC=golang-dev
https://golang.org/cl/4351048
The transport readLoop was waiting forever for the client to
read the non-existent body before proceeding to read the next
request.
R=rsc
CC=golang-dev
https://golang.org/cl/4357051
The error will only occur for invalid patterns, but without this
error path there is no way to know that Glob has failed due to
an invalid pattern.
R=rsc
CC=golang-dev
https://golang.org/cl/4346044
Write never writes less than the buffer size and WriteString takes advantage
of the copy built-in to improve write efficiency.
R=rsc, ality, rog
CC=golang-dev
https://golang.org/cl/4344060
According to RFC 3986: "For consistency, URI producers
and normalizers should use uppercase hexadecimal digits
for all percent-encodings." Using lower case characters
makes it incompatible with Google APIs when signing OAuth requests.
R=golang-dev, rsc1, rsc
CC=golang-dev
https://golang.org/cl/4352044
Since Go code can deadlock, this lets a testsuite driver set a
time limit for the test to run. This is simple but imperfect,
in that it only catches deadlocks in Go code, not in the
runtime scheduler.
R=r, rsc, iant2
CC=golang-dev
https://golang.org/cl/4326048
Also fix a bug: precision was in terms of bytes; should be runes.
Fixes#1652.
R=rsc, bradfitzgo, r2, bradfitzwork
CC=golang-dev
https://golang.org/cl/4280086
also: minor fix to parser
Note: gotest won't run the gotype test yet until
it permits TestXXX functions where XXX is empty.
R=r
CC=golang-dev
https://golang.org/cl/4300053
- don't consume '\n' as part of line comment
(otherwise grammars where '\n' are tokens won't
see them after a line comment)
- permit line comments to end in EOF
R=r
CC=golang-dev
https://golang.org/cl/4277089
As a special case, multi-line raw strings (i.e., strings in `` quotes)
were not indented if they were the only token on a line. This heuristic
was meant to improve formatting for multi-line raw strings where sub-
sequent lines are not indented at the level of the surrounding code.
Multiple people have complained about this. Removing the heuristic
again because it makes the formatting more regular, easier to under-
stand, and simplifies the implementation.
- manual changes to ebnf/ebnf_test.go for readability
- gofmt -w src misc
Fixes#1643.
R=r, rsc
CC=golang-dev
https://golang.org/cl/4307045
I'm in two minds as to whether this should be a function of gotest.
Tests that can flake out like this should be rare enough that we
needn't add more mechanism.
R=r
CC=golang-dev
https://golang.org/cl/4335042
It was left in netFD.connect() by an oversight (as the name
implies, bind has no business being in connect). As a result
of this change and by only calling netFD.connect() when ra
isn't nil it becomes simpler with less code duplication.
Additionally, if netFD.connect() fails, set sysfd to -1 to
avoid finalizers (e.g. on windows) calling shutdown on a
closed and possibly reopened socket that just happened to
share the same descriptor.
R=golang-dev, rsc1, rsc
CC=golang-dev
https://golang.org/cl/4328043
This changeset makes it possible for crypto/x509 to parse
certificates that include the 'Extended Key Usage' extension
with the critical bit set.
R=agl1
CC=golang-dev
https://golang.org/cl/4277075
* Adds support for GENERAL STRING
* Adds support for APPLICATION tagged values.
* Add UnmarshalWithParams to set parameters for the top-level
structure
R=golang-dev, rsc1, r
CC=golang-dev
https://golang.org/cl/4291075
Refactored bind/connect from sock.go into netFD.connect(), as
a consequence newFD() doesn't accept laddr/raddr anymore, and
expects an (optional) call to netFD.connect() followed by a
call to netFD.setAddr().
Windows code is updated, but still uses blocking connect,
since otherwise it needs support for ConnectEx syscall.
R=brainman, rsc
CC=golang-dev
https://golang.org/cl/4303060
Drop laddr argument from Dial.
Drop cname return from LookupHost.
Add LookupIP, LookupCNAME, ParseCIDR, IP.Equal.
Export SplitHostPort, JoinHostPort.
Add AAAA (IPv6) support to host lookups.
Preparations for implementing some of the
lookups using cgo.
ParseCIDR and IP.Equal are logically new in this CL
but accidentally snuck into an earlier CL about unused
labels that was in the same client.
In crypto/tls, drop laddr from Dial to match net.
R=golang-dev, dsymonds, adg, rh
CC=golang-dev
https://golang.org/cl/4244055
With gccgo some operating systems require using select rather
than epoll or kevent. Using select means that we have to wake
up the polling thread each time we add a new file descriptor.
This implements that in the generic code rather than adding
another wakeup channel, even though nothing in the current net
package uses the capability.
R=rsc, iant2
CC=golang-dev
https://golang.org/cl/4284069
NewPackage creates an ast.Package node from
a set of package files and resolves unresolved
identifiers.
Also:
- Changed semantics of Scope.Insert: If an
object is inserted w/o errors, the result
is nil (before it was obj).
- Fixed an identifier resolution bug in the
parser: map keys must not be resolved.
gotype runs through several go/* packages
and successfully resolves all (non-field/method)
identifiers.
R=rog, rsc
CC=golang-dev
https://golang.org/cl/4298044
in gdb, 'info goroutines' and 'goroutine <n> <cmd> were crashing
because the 'g' and 'm' structures had changed a bit.
R=rsc
CC=golang-dev
https://golang.org/cl/4289077
On darwin amd64 it was impossible to create more that ~132 threads. While
investigating I noticed that go consumes almost 1TB of virtual memory per
OS thread and the reason for such a small limit of OS thread was because
process was running out of virtual memory. While looking at bsdthread_create
I noticed that on amd64 it wasn't using PTHREAD_START_CUSTOM.
If you look at http://fxr.watson.org/fxr/source/bsd/kern/pthread_synch.c?v=xnu-1228
you will see that in that case darwin will use stack pointer as stack size,
allocating huge amounts of memory for stack. This change fixes the issue
and allows for creation of up to 2560 OS threads (which appears to be some
Mac OS X limit) with relatively small virtual memory consumption.
R=rsc
CC=golang-dev
https://golang.org/cl/4289075
New make target "testshort" runs "gotest -test.short" and is invoked
by run.bash, which is invoked by all.bash.
Use -test.short to make one package (crypto ecdsa) run much faster.
More changes to come.
Once this is in, I will update the long-running tests to use the new flag.
R=rsc
CC=golang-dev
https://golang.org/cl/4317043
Fixes#1641.
Actually it side steps the real issue, which is that the
setitimer(2) implementation on OS X is not useful for
profiling of multi-threaded programs. I filed the below
using the Apple Bug Reporter.
/*
Filed as Apple Bug Report #9177434.
This program creates a new pthread that loops, wasting cpu time.
In the main pthread, it sleeps on a condition that will never come true.
Before doing so it sets up an interval timer using ITIMER_PROF.
The handler prints a message saying which thread it is running on.
POSIX does not specify which thread should receive the signal, but
in order to be useful in a user-mode self-profiler like pprof or gprof
http://code.google.com/p/google-perftoolshttp://www.delorie.com/gnu/docs/binutils/gprof_25.html
it is important that the thread that receives the signal is the one
whose execution caused the timer to expire.
Linux and FreeBSD handle this by sending the signal to the process's
queue but delivering it to the current thread if possible:
http://lxr.linux.no/linux+v2.6.38/kernel/signal.c#L802
807 /*
808 * Now find a thread we can wake up to take the signal off the queue.
809 *
810 * If the main thread wants the signal, it gets first crack.
811 * Probably the least surprising to the average bear.
812 * /
http://fxr.watson.org/fxr/source/kern/kern_sig.c?v=FREEBSD8;im=bigexcerpts#L1907
1914 /*
1915 * Check if current thread can handle the signal without
1916 * switching context to another thread.
1917 * /
On those operating systems, this program prints:
$ ./a.out
signal on cpu-chewing looper thread
signal on cpu-chewing looper thread
signal on cpu-chewing looper thread
signal on cpu-chewing looper thread
signal on cpu-chewing looper thread
signal on cpu-chewing looper thread
signal on cpu-chewing looper thread
signal on cpu-chewing looper thread
signal on cpu-chewing looper thread
signal on cpu-chewing looper thread
$
The OS X kernel does not have any such preference. Its get_signalthread
does not prefer current_thread(), in contrast to the other two systems,
so the signal gets delivered to the first thread in the list that is able to
handle it, which ends up being the main thread in this experiment.
http://fxr.watson.org/fxr/source/bsd/kern/kern_sig.c?v=xnu-1456.1.26;im=excerpts#L1666
$ ./a.out
signal on sleeping main thread
signal on sleeping main thread
signal on sleeping main thread
signal on sleeping main thread
signal on sleeping main thread
signal on sleeping main thread
signal on sleeping main thread
signal on sleeping main thread
signal on sleeping main thread
signal on sleeping main thread
$
The fix is to make get_signalthread use the same heuristic as
Linux and FreeBSD, namely to use current_thread() if possible
before scanning the process thread list.
*/
#include <sys/time.h>
#include <sys/signal.h>
#include <pthread.h>
#include <unistd.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
static void handler(int);
static void* looper(void*);
static pthread_t pmain, ploop;
int
main(void)
{
struct itimerval it;
struct sigaction sa;
pthread_cond_t cond;
pthread_mutex_t mu;
memset(&sa, 0, sizeof sa);
sa.sa_handler = handler;
sa.sa_flags = SA_RESTART;
memset(&sa.sa_mask, 0xff, sizeof sa.sa_mask);
sigaction(SIGPROF, &sa, 0);
pmain = pthread_self();
pthread_create(&ploop, 0, looper, 0);
memset(&it, 0, sizeof it);
it.it_interval.tv_usec = 10000;
it.it_value = it.it_interval;
setitimer(ITIMER_PROF, &it, 0);
pthread_mutex_init(&mu, 0);
pthread_mutex_lock(&mu);
pthread_cond_init(&cond, 0);
for(;;)
pthread_cond_wait(&cond, &mu);
return 0;
}
static void
handler(int sig)
{
static int nsig;
pthread_t p;
p = pthread_self();
if(p == pmain)
printf("signal on sleeping main thread\n");
else if(p == ploop)
printf("signal on cpu-chewing looper thread\n");
else
printf("signal on %p\n", (void*)p);
if(++nsig >= 10)
exit(0);
}
static void*
looper(void *v)
{
for(;;);
}
R=r
CC=golang-dev
https://golang.org/cl/4273113
Also fix comment.
The only caller of chanrecv initializes the value to false, so
this patch makes no difference at present. But it seems like
the right thing to do.
R=rsc
CC=golang-dev
https://golang.org/cl/4312053
Correctly distinguish between lhs and rhs identifiers
and resolve/declare them accordingly.
Collect field and method names in respective scopes
(will be available after some minor AST API changes).
Also collect imports since it's useful to have that
list directly w/o having to re-traverse the AST
(will also be available after some minor AST API changes).
No external API changes in this CL.
R=rsc, rog
CC=golang-dev
https://golang.org/cl/4271061
The top level bytes.Buffer is always there and can be re-used.
Rpc goes from 83 to 79 mallocs per round trip.
R=rsc
CC=golang-dev
https://golang.org/cl/4271062
- StartProcess will work with relative (to attr.Dir, not
current directory) executable filenames
- StartProcess will only work if executable filename points
to the real file, it will not search for executable in the
$PATH list and others (see CreateProcess manual for details)
- StartProcess argv strings can contain any characters
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/4306041
This patch adds a connection cache and keep-alive
support to Transport, which is used by the
HTTP client.
It's also structured such that it's easy to add
HTTP pipelining in the future.
R=rsc, petar-m, bradfitzwork, r
CC=golang-dev
https://golang.org/cl/4272045
Revert changes to printer.Config. Pass in the
nodeSizes map trough an internal helper function.
R=golang-dev, rsc1
CC=golang-dev
https://golang.org/cl/4309042
Use memoization to avoid repeated recomputation of nested
node sizes. Speeds up testdata/slow.input by several orders
of magnitude.
- added respective test case
- added timeout to test code
- deleted some unrelated unused code
Fixes#1628.
R=rsc, r
CC=golang-dev
https://golang.org/cl/4274075
These timeouts are breaking tests in very slow
systems every once in a while. I've noticed
problems when compiling the Ubuntu packages for
arm, specifically.
R=golang-dev, adg
CC=golang-dev
https://golang.org/cl/4291058
Also in the common case avoid unnecessary buffering in
the channel.
Removes 13 allocations per round trip. Now at 86, down from
144 a week ago.
R=rsc, bradfitzgo, r2, rsc1
CC=golang-dev
https://golang.org/cl/4277060
The scanner returns slices into the original source
for token values. If those slices are making it into
the AST and from there into other long-living data
structures (e.g. godoc search), references to the
original source are kept around involuntarily.
For the current godoc and source tree, this change reduces
memory consumption after indexing and before GC by ~92MB
or almost 30%, and by ~10MB after GC (or about 6%).
R=rsc
CC=golang-dev
https://golang.org/cl/4273072
In conjunction with the non-blocking system call CL, this
gives about an 8% performance improvement on a client/server
test running on my local machine.
R=rsc, iant2
CC=golang-dev
https://golang.org/cl/4272057
- just an oversight; we were reallocating a buffer.
- use unsafe to avoid allocating storage for a string twice.
R=rsc
CC=golang-dev
https://golang.org/cl/4290056
Permit system calls to be designated as non-blocking, meaning
that we simply call them without involving the scheduler.
This change by itself is mostly performance neutral. In
combination with a following change to the net package there
is a performance advantage.
R=rsc, dfc, r2, iant2, rsc1
CC=golang-dev
https://golang.org/cl/4278055
- use enc.err and dec.err instead of return values in deferred error catcher
- replace io.WriteString with buffer.WriteString
now at:
mallocs per encode of type Bench: 7
mallocs per decode of type Bench: 8
R=rsc
CC=golang-dev
https://golang.org/cl/4277057
This just returns a ClientConn suitable for writing
proxy requests. To be used in Transport.
R=rsc, petar-m
CC=golang-dev
https://golang.org/cl/4290052
Dependency on bufio crept in during last CL; this breaks the cycle.
Also add a missing '-' to the documentation.
R=rsc
CC=golang-dev
https://golang.org/cl/4274061
There is some disagreement about how to deal with hash values larger
than the curve order size. We choose to follow OpenSSL's lead here.
R=bradfitzgo, r
CC=golang-dev
https://golang.org/cl/4273059
If braces don't have position information for a composite
literal, don't assume alignment of key:value pairs under
the (wrong) assumption that there may be multiple lines.
R=rsc
CC=golang-dev
https://golang.org/cl/4297043
On my mac:
mallocs per rpc round trip: 144
rpc.BenchmarkEndToEnd 10000 228244 ns/op
Room for improvement.
R=rsc
CC=golang-dev
https://golang.org/cl/4274058
This reduces the number of writes by 2 (1 client, 1 server) on each round trip.
A simple test shows 24% higher throughput.
R=rsc
CC=golang-dev
https://golang.org/cl/4279057
The loop always makes an extra system call. It only makes a
difference if more than 100 goroutines started waiting for
something to happen on a network file descriptor since the
last time the pipe was drained, which is unlikely since we
will be woken up the first time a goroutine starts waiting.
If we don't drain the pipe this time, we'll be woken up again
right away and can drain again.
R=rsc
CC=golang-dev
https://golang.org/cl/4275042
Add a benchmark.
BenchmarkEndToEndPipe gives 14.3microseconds/op before,
13.1microseconds/op after, or about 76e3 round trips per second
through the kernel.
With a bytes buffer, and therefore no system calls for I/O, the
numbers go to 7.3microseconds/op, or about 137e3 round trips
per second.
R=rsc
CC=golang-dev
https://golang.org/cl/4279045
Transport.Do -> RoundTripper.RoundTrip
This makes way for a subsequent CL to export the
currently private RoundTripper implementation
as struct Transport.
R=rsc, r
CC=golang-dev
https://golang.org/cl/4286043
