The alternative is to record enough information that the
trap handler know which registers contain cached globals
and can flush the registers back to their original locations.
That's significantly more work.
This only affects globals that have been written to.
Code that reads from a global should continue to registerize
as well as before.
Fixes#1304.
R=ken2
CC=golang-dev
https://golang.org/cl/5687046
They are portability problems and the options are almost always zero in practice anyway.
R=golang-dev, dsymonds, r, bradfitz
CC=golang-dev
https://golang.org/cl/5688046
-- add driver.Value type and documentation,
convert from interface{} to Value where
appropriate.
-- don't say "subset" anywhere,
-- SubsetValuer -> Valuer
-- SubsetValue -> Value
-- IsParameterSubsetType -> IsValue
-- IsScanSubsetType -> IsScanValue
Fixes#2842
R=golang-dev, r, rsc
CC=golang-dev
https://golang.org/cl/5674084
While we're here, get rid of the old names for the escaping functions.
Fixes#3073.
R=golang-dev, dsymonds, rsc
CC=golang-dev
https://golang.org/cl/5685049
My theory is that the call to f() allocates, which triggers
a garbage collection, which itself may do some allocation,
which is being counted. Running a garbage collection
before starting the test should avoid this problem.
Fixes#2894 (I hope).
R=golang-dev, bradfitz, nigeltao
CC=golang-dev
https://golang.org/cl/5685046
ARM doesn't have the concept of scale, so I renamed the field
Addr.scale to Addr.flag to better reflect its true meaning.
R=rsc
CC=golang-dev
https://golang.org/cl/5687044
* disallow embedding of C type (Fixes issue 2552)
* detect 0-length array (Fixes issue 2806)
* use typedefs when possible, to avoid attribute((unavailable)) (Fixes issue 2888)
* print Go types constructed from C types using original C types (Fixes issue 2612)
This fix changes _cgo_export.h to repeat the preamble from import "C".
Otherwise the fix to issue 2612 is impossible, since it cannot refer to
types that have not been defined. If people are using //export and
putting non-header information in the preamble, they will need to
refactor their code.
R=golang-dev, r, r
CC=golang-dev
https://golang.org/cl/5672080
morebuf holds a pc/sp from the last stack split or
reflect.call or panic/recover. If the pc is a closure,
the reference will keep it from being collected.
moreargp holds a pointer to the arguments from the
last stack split or reflect.call or panic/recover.
Normally it is a stack pointer and thus not of interest,
but in the case of reflect.call it is an allocated argument
list and holds up the arguments to the call.
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/5674109
The garbage collector can avoid scanning this section, with
reduces collection time as well as the number of false positives.
Helps a little bit with issue 909, but certainly does not solve it.
R=ken2
CC=golang-dev
https://golang.org/cl/5671099
We should, after Go 1, make them work the same as
package xml, that is, make them appear in the outer
struct. For now turn them off so that people do not
depend on the old behavior.
Fixing them is issue 3069.
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/5656102
The m->cret word holds the C return value when returning
across a stack split boundary. It was not being cleared after
use, which means that the return value (if a C function)
or else the value of AX/R0 at the time of the last stack unsplit
was being kept alive longer than necessary. Clear it.
I think the effect here should be very small, but worth fixing
anyway.
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/5677092
When we build the encode engine for a recursive type, we
mustn't disregard the indirections or we can try to reuse an
engine at the wrong indirection level.
Fixes#3026.
R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/5675087
Fix by setting the element type if we discover it's zero while building.
We could have fixed this better with foresight by doing the id setting in a
different sequence, but doing that now would break binary compatibility.
Fixes#2995.
R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/5675083
I am making a unilateral decision here. I could also settle for returning 0,
as long it's documented, but I argue that it's equivalent to an index
out of bounds.
Fixes#2892.
R=golang-dev, dsymonds, rsc
CC=golang-dev
https://golang.org/cl/5676079
1, strip last path separator from $GOROOT
The user might define GOROOT=/path/to/go/, but then the dir
check in defaulttarg() will always complain the current dir
is not within $GOROOT/src/.
2, resolve symlinks in the default goroot
Or if getcwd() returns a fully-resolved path, the check in
defaulttarg() will always fail.
R=rsc
CC=golang-dev
https://golang.org/cl/5649073
The panic happens if -benchtime flag is specified:
go test -bench=EndToEndAsyncHTTP -benchtime=120
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/5677075
Convert cryptotype to general go1rename fix.
Add os.Exec -> syscall.Exec fix along with new
URL fixes.
Fixes#2946.
R=golang-dev, r, dsymonds
CC=golang-dev
https://golang.org/cl/5672072
We want to be able to implement good Windows support
after Go 1. Right now Windows tries to use Unix domain
sockets, and I'd rather just have it not be available.
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/5671076
The set of errors forwarded by the os package varied with system and
was therefore non-portable.
Three helpers added for portable error checking: IsExist, IsNotExist, and IsPermission.
One or two more may need to come, but let's keep the set very small to discourage
thinking about errors that way.
R=mikioh.mikioh, gustavo, r, rsc
CC=golang-dev
https://golang.org/cl/5672047
It was being skipped due to an oversight.
Also adjust naming parameters for map type construction - makes debugging easier.
Prelude to issue 3026.
R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/5674071
Clang 3.1 has more warnings enabled by default than GCC.
Combined with -Werror, they cause the build to fail
unnecessarily. if the name of our compiler ends in "clang",
add the necessary extra -Wno options. Ideally we would add
these flags unconditionally, as GCC is supposed to ignore
unknown -Wno flags, but apple's llvm-gcc doesn't.
Fixes#2878.
R=rsc, dave
CC=golang-dev
https://golang.org/cl/5673055
Periodically browse MHeap's freelists for long unused spans and release them if any.
Current hardcoded settings:
- GC is forced if none occured over the last 2 minutes.
- spans are handed back after 5 minutes of uselessness.
SysUnused (for Unix) is a wrapper on madvise MADV_DONTNEED on Linux and MADV_FREE on BSDs.
R=rsc, dvyukov, remyoudompheng
CC=golang-dev
https://golang.org/cl/5451057
An unindented XML example is hard to follow. MarshalIndent
allows moving the example over to a test file (and fixing it).
R=golang-dev, r, gustavo, r, rsc
CC=golang-dev
https://golang.org/cl/5674050
It depended on old behavior of functions in structs.
Solved by adding a boolean method to check .Run != nil.
R=golang-dev, adg, r, rsc
CC=golang-dev
https://golang.org/cl/5674062
go/doc: move Examples to go/ast
cmd/go: use go/doc to read examples
src/pkg: update examples to use new convention
This is to make whole file examples more readable. When presented as a
complete function, preceding an Example with its output is confusing.
The new convention is to put the expected output in the final comment
of the example, preceded by the string "output:" (case insensitive).
An idiomatic example looks like this:
// This example demonstrates Foo by doing bar and quux.
func ExampleFoo() {
// example body that does bar and quux
// Output:
// example output
}
R=rsc, gri
CC=golang-dev
https://golang.org/cl/5673053
Whoops. Consume the body of the first request
before making the subsequent /quit request.
R=golang-dev, untheoretic
CC=golang-dev
https://golang.org/cl/5674054
Also: Simplified handling of selector expressions. As a result, complicated
multi-line expressions containing selectors and calls/indices with arguments
broken accross lines don't get indented the same way as before, but the change
is minimal (see tests) and there's no such code in the std library. It seems
a worthwhile compromise given the much simpler code.
Applied gofmt -w $GOROOT/src $GOROOT/misc .
Fixes#1847.
R=rsc
CC=golang-dev
https://golang.org/cl/5675062
Makes it possible to run
GOARCH=amd64 make.bash
GOARCH=386 make.bash --no-clean
to avoid deleting some of the work done by the first one.
R=golang-dev
CC=golang-dev
https://golang.org/cl/5673056
Changeset 36c9c7810f14 broke support for grsec-patched kernels.
Those do not give back the address requested without MAP_FIXED,
so when verifying an mmap without this flag for success, the
resulting address must not be compared against the requested
address since it may have succeeded at a different location.
R=golang-dev, rsc, gustavo, iant
CC=golang-dev
https://golang.org/cl/5650072
Apparently some distros don't let you ptrace attach
to your own existing processes.
Run strace on the child directly, instead, which
reportedly is more often allowed, and makes the
code simpler too.
R=golang-dev, n13m3y3r
CC=golang-dev
https://golang.org/cl/5675050
Before this, make.bash --dist-tool would fail,
because $GOROOT/bin/tool was being mkdir'd
instead of $GOROOT/pkg/tool/linux_amd64.
R=rsc
CC=golang-dev
https://golang.org/cl/5666046
- if a package path leads to subdirectories only,
show command instead, if any
- to force documentation for a command, use the
cmd/ prefix, as in: godoc cmd/go
(note that for the go command, the prefix is
not required since there is no actual go library
package at the moment)
Fixes#3012.
R=rsc
CC=golang-dev
https://golang.org/cl/5665049
* add -work option to save temporary files (Fixes issue 2980)
* fix go test -i to work with cgo packages (Fixes issue 2936)
* do not overwrite/remove empty directories or non-object
files during build (Fixes issue 2829)
* remove package main vs package non-main heuristic:
a directory must contain only one package (Fixes issue 2864)
* to make last item workable, ignore +build tags for files
named on command line: go build x.go builds x.go even
if it says // +build ignore.
* add // +build ignore tags to helper programs
R=golang-dev, r, r
CC=golang-dev
https://golang.org/cl/5674043
before after
go test -short time 4.144s 1.215s
go test -short runtime 1.315s 0.351s
go test -short -cpu=1,2,4 runtime 4.376s 1.266s
Partially solves issue 3015.
R=golang-dev, r
CC=golang-dev, remy
https://golang.org/cl/5673045
This lets us get rid of the OS-dependent implementations
of SyscallError. The name "Err" was chosen to match the
PathError type.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/5651084
This change shows all the information present
in the Time value (now including fractional seconds)
and also arranges the fields so that, within a single time zone,
string comparison and time comparison agree.
R=golang-dev, rogpeppe, r
CC=golang-dev
https://golang.org/cl/5654078
It is possible that Linux and Windows copy the FP control word
from the parent thread when creating a new thread. Empirically,
Darwin does not. Reset the FP control world in all cases.
Enable the floating-point strconv test.
Fixes#2917 (again).
R=golang-dev, r, iant
CC=golang-dev
https://golang.org/cl/5660047
Once we've evicted all the blocked I/O, the ref count
should go to zero quickly, so it should be safe to
postpone the close(2) until then.
Fixes#1898.
Fixes#2116.
Fixes#2122.
R=golang-dev, mikioh.mikioh, bradfitz, fullung, iant
CC=golang-dev
https://golang.org/cl/5649076
Filed issue 3016 to fix this, but I really want
to see a "ok" in the Windows column so we
know what is and is not working.
R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/5658050
Now with a bit more paranoia and lower number of requests
to keep it under the default OS X 256 fd limit.
R=golang-dev, dsymonds, rsc
CC=golang-dev
https://golang.org/cl/5659051
It's not as pretty, but it deletes some irrelevant information from the
printout and avoids a dependency.
It also means the test binary will stop if a test panics. That's a feature,
not a bug.
Any output printed by the test appears before the panic traceback.
before:
--- FAIL: TestPanic (0.00 seconds)
fmt_test.go:19: HI
testing.go:257: runtime error: index out of range
/Users/r/go/src/pkg/testing/testing.go:257 (0x23998)
_func_003: t.Logf("%s\n%s", err, debug.Stack())
/Users/r/go/src/pkg/runtime/proc.c:1388 (0x10d2d)
panic: reflect·call(d->fn, d->args, d->siz);
/Users/r/go/src/pkg/runtime/runtime.c:128 (0x119b0)
panicstring: runtime·panic(err);
/Users/r/go/src/pkg/runtime/runtime.c:85 (0x11857)
panicindex: runtime·panicstring("index out of range");
/Users/r/go/src/pkg/fmt/fmt_test.go:21 (0x23d72)
TestPanic: a[10]=1
/Users/r/go/src/pkg/testing/testing.go:264 (0x21b75)
tRunner: test.F(t)
/Users/r/go/src/pkg/runtime/proc.c:258 (0xee9e)
goexit: runtime·goexit(void)
FAIL
after:
--- FAIL: TestPanic (0.00 seconds)
fmt_test.go:19: HI
panic: runtime error: index out of range [recovered]
panic: (*testing.T) (0xec3b0,0xf8400001c0)
goroutine 2 [running]:
testing._func_003(0x21f5fa8, 0x21f5100, 0x21f5fb8, 0x21f5e88)
/Users/r/go/src/pkg/testing/testing.go:259 +0x108
----- stack segment boundary -----
fmt_test.TestPanic(0xf8400001c0, 0x27603728)
/Users/r/go/src/pkg/fmt/fmt_test.go:21 +0x6b
testing.tRunner(0xf8400001c0, 0x18edb8, 0x0, 0x0)
/Users/r/go/src/pkg/testing/testing.go:264 +0x6f
created by testing.RunTests
/Users/r/go/src/pkg/testing/testing.go:343 +0x76e
goroutine 1 [chan receive]:
testing.RunTests(0x2000, 0x18edb8, 0x2400000024, 0x100000001, 0x200000001, ...)
/Users/r/go/src/pkg/testing/testing.go:344 +0x791
testing.Main(0x2000, 0x18edb8, 0x2400000024, 0x188a58, 0x800000008, ...)
/Users/r/go/src/pkg/testing/testing.go:275 +0x62
main.main()
/var/folders/++/+++Fn+++6+0++4RjPqRgNE++2Qk/-Tmp-/go-build743922747/fmt/_test/_testmain.go:129 +0x91
exit status 2
R=rsc, dsymonds
CC=golang-dev
https://golang.org/cl/5658048
This CL makes it possible to run make.bash with
GOOS and GOARCH set to something other than
the native host GOOS and GOARCH.
As part of the CL, the tool directory moves from bin/tool/
to pkg/tool/goos_goarch where goos and goarch are
the values for the host system (running the build), not
the target. pkg/ is not technically appropriate, but C objects
are there now tool (pkg/obj/) so this puts all the generated
binaries in one place (rm -rf $GOROOT/pkg cleans everything).
Including goos_goarch in the name allows different systems
to share a single $GOROOT on a shared file system.
Fixes#2920.
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/5645093
Generates an infinite stream (at least >1GB) of:
=== RUN TestTransportPersistConnLeak
2012/02/13 22:20:19 http: Accept error: accept tcp 127.0.0.1:63972:
too many open files
2012/02/13 22:20:19 http: Accept error: accept tcp 127.0.0.1:63972:
too many open files
2012/02/13 22:20:19 http: Accept error: accept tcp 127.0.0.1:63972:
too many open files
2012/02/13 22:20:19 http: Accept error: accept tcp 127.0.0.1:63972:
too many open files
2012/02/13 22:20:19 http: Accept error: accept tcp 127.0.0.1:63972:
too many open files
2012/02/13 22:20:19 http: Accept error: accept tcp 127.0.0.1:63972:
too many open files
2012/02/13 22:20:19 http: Accept error: accept tcp 127.0.0.1:63972:
too many open files
2012/02/13 22:20:19 http: Accept error: accept tcp 127.0.0.1:63972:
too many open files
2012/02/13 22:20:19 http: Accept error: accept tcp 127.0.0.1:63972:
too many open files
2012/02/13 22:20:19 http: Accept error: accept tcp 127.0.0.1:63972:
too many open files
R=bradfitz
CC=golang-dev
https://golang.org/cl/5661052
1, IMO, the fatal error "regfree: not a register" from 5g when
compiling runtime/debug.go is due to gcc miscompile, it doesn't
show up when compiled with -O0. But I still haven't thought of
a way to fix this, should all ARM builds be built with -O0?
2, fixed mksysnum_linux.pl, so zsysnum_linux_arm.go no longer
needs to be hand-generated.
3, regen all in pkg syscall for Linux/ARM on Debian 6.0
This CL is somewhat big, I'd like to split it if necessary.
R=rsc, dave
CC=golang-dev
https://golang.org/cl/5659044
This was an implementation detail that snuck into the public interface.
*Writer.Create gives you an io.Writer, the *Writer itself was never
meant to be written to.
R=golang-dev, dsymonds, r
CC=golang-dev
https://golang.org/cl/5654076
Makes files like src/pkg/syscall/ztypes_linux_amd64.go easier to read.
(The copy that is checked in predates the //line output mode,
so this also preserves the status quo.)
R=golang-dev, iant, gri
CC=golang-dev
https://golang.org/cl/5655068
Also update build to be able to run mkbuiltin again.
The export form has changed a little, so builtin.c has
more diffs than unsafe.go.
In CL 5650069, I just edited the documentation, a rarely
successful method of change.
R=ken2
CC=golang-dev
https://golang.org/cl/5662043
Restore package os/signal, with new API:
Notify replaces Incoming, allowing clients
to ask for certain signals only. Also, signals
go to everyone who asks, not just one client.
This could plausibly move into package os now
that there are no magic side effects as a result
of the import.
Update runtime for new API: move common Unix
signal handling code into signal_unix.c.
(It's so easy to do this now that we don't have
to edit Makefiles!)
Tested on darwin,linux 386,amd64.
Fixes#1266.
R=r, dsymonds, bradfitz, iant, borman
CC=golang-dev
https://golang.org/cl/3749041
1) Remove the Reset() member in crypto/aes and crypto/des (and
document the change).
2) Turn several empty error structures into vars. Any remaining error
structures are either non-empty, or will probably become so in the
future.
3) Implement SetWriteDeadline for TLS sockets. At the moment, the TLS
status cannot be reused after a Write error, which is probably fine
for most uses.
4) Make crypto/aes and crypto/des return a cipher.Block.
R=rsc, r
CC=golang-dev
https://golang.org/cl/5625045
one trie lookup per rune is needed. See forminfo.go for a description
of the new format. Also included leading and trailing canonical
combining class in decomposition information. This will often avoid
additional trie lookups.
R=r, r
CC=golang-dev
https://golang.org/cl/5616071
This is not the finished product,
but a good checkpoint from which to
proceed with further development.
R=golang-dev, dsymonds, r
CC=golang-dev
https://golang.org/cl/5571061
8g/cgen.c
print format type mismatch
8l/asm.c
resoff set and not used
gc/pgen.c
misleading comparison INT > 0x80000000
gc/reflect.c
dalgsym must be static to match forward declaration
gc/subr.c
assumed_equal set and not used
hashmem's second argument is not used
gc/walk.c
duplicated (unreachable) code
R=rsc
CC=golang-dev
https://golang.org/cl/5651079
hammerCompareAndSwapPointer64 was only passing on
little-endian systems. hammerCompareAndSwapPointer32 was
writing 8 bytes to a uint32 value on the heap.
R=rsc, dvyukov
CC=golang-dev
https://golang.org/cl/5654065
You could argue for changing all the others, but
Linux is outvoted, and the only time it matters
is when newfd==-1, in which case you can call Dup.
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/5650073
Fixes#2776.
There was a previous attempt at CL 5592043 but that
seems to have stalled. This one is simpler, and more up to date
(correct handling of spdy, for example).
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/5645091
The Date-Modified header truncates sub-second precision, so
use mtime < t+1s instead of mtime <= t to check for unmodified.
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/5655052
Among other things, this avoids putting a testing.go:nnn:
prefix on every line of the stack trace.
R=golang-dev, r, dsymonds, r
CC=golang-dev
https://golang.org/cl/5651081
unsafe: delete Typeof, Reflect, Unreflect, New, NewArray
Part of issue 2955 and issue 2968.
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/5650069
Cannot reproduce the failure locally,
but add explicit test in case some other
machine can.
Fixes#2917 (for now).
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/5651071
I thought that -timeout was per-test, but it is for the
whole program execution, so cmd/go can adjust its timer
(also for whole program execution) accordingly.
Fixes#2993.
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/5650070