Plus:
- calibration "test" - include in tests with gotest -calibrate
- basic Mul benchmark
- extra multiplication tests
- various cleanups
This change improves multiplication speed of numbers >= 30 words
in length (current threshold; found empirically with calibrate):
The multiplication benchmark (multiplication of a variety of long numbers)
improves by ~35%, individual multiplies can be significantly faster.
gotest -benchmarks=Mul
big.BenchmarkMul 500 6829290 ns/op (w/ Karatsuba)
big.BenchmarkMul 100 10600760 ns/op
There's no impact on pidigits for -n=10000 or -n=20000
because the operands are are too small.
R=rsc
CC=golang-dev
https://golang.org/cl/1004042
Rather than confuse people by linking to a spec we
don't fully support, let them just use the package docs.
R=r, rsc
CC=golang-dev
https://golang.org/cl/1016041
This matches the old JSON package behavior.
All lowercase names are not as standard as I believed,
and it seems less surprising to need to write
type T struct { Field string "field" }
to get lower case (behavior after this CL) than it does to need
to write
type T struct { Field string "Field" }
to preserve the case (behavior before this CL).
Also test and fix unmarshal into non-nil interface
value or pointer.
Fixes#744.
R=r
CC=golang-dev
https://golang.org/cl/1013041
Adds two more methods, Timeout and Temporary.
Implemented by os.Errno too. The intent is to make
the checks for os.EAGAIN a little less clunky.
It should also let us clean up a bug that Mike Solomon
pointed out: if a network server gets an "out of file descriptors"
error from Accept, the listener should not stop.
It will be able to check this because that error would
have Temporary() == true.
Also clean up some underscore names.
Fixes#442.
R=r
CC=golang-dev, msolo
https://golang.org/cl/957045
There was a bug in makeN that caused lots of unnecessary
allocations. Fixing this exposed a few bugs in other
functions which worked when makeN allocated a new slice, but
failed when it simply resized an existing slice. The result
is a pretty big performance improvement. When running
pidigits, here were the numbers I got on amd64:
Before this change:
pidigits 10000
gcc -O2 pidigits.c -lgmp 2.09u 0.02s 2.11r
gc pidigits 12.68u 0.04s 12.72r
gc_B pidigits 12.71u 0.03s 12.75r
After:
pidigits 10000
gcc -O2 pidigits.c -lgmp 2.09u 0.00s 2.10r
gc pidigits 6.82u 0.00s 6.85r
gc_B pidigits 6.55u 0.01s 6.59r
R=rsc, gri
CC=golang-dev
https://golang.org/cl/953042
Due to page boundary rounding, the header would have
been loaded as part of the text segment already, but this
change placates the "paxctl" tool on so-called hardened
Linux distributions (as if normal distributions weren't already
hard enough to use).
R=r
CC=golang-dev
https://golang.org/cl/954041
SETSID does return an errno - any reason why it has been done this
way in zsyscall_linux_* ? Otherwise it should be the same as darwin.
From SETSID(2) on my Linux box:
ERRORS
On error, -1 is returned, and errno is set.
Fixes#730
R=rsc
CC=golang-dev
https://golang.org/cl/878047
This is the first of probably four separate CLs
for the new implementation of the json package.
The scanner is the core of the new implementation.
The other CLs would be the new decoder,
the new encoder, and support for JSON streams.
R=r
CC=golang-dev
https://golang.org/cl/802051
$ godoc xml | grep Copy\(\)
func (c CharData) Copy() CharData
func (c Comment) Copy() Comment
func (d Directive) Copy() Directive
func (p ProcInst) Copy() ProcInst
func (e StartElement) Copy() StartElement
--------------------------------------------
$ godoc -src xml | grep Copy\(\)
func (c CharData) Copy() CharData
--------------------------------------------
$ godoc -src xml Copy
func (c CharData) Copy() CharData { return CharData(makeCopy(c)) }
--------------------------------------------
The command "godoc -src pkg_name" should output the interface of the named package, but it excludes all duplicate entries. Also the command "godoc -src pkg_name method_name" will output the source code only for one method even if there are more of them with the same name in the same package. This patch set fixes this issue.
R=gri
CC=golang-dev
https://golang.org/cl/883051
fmt.Printf("%b", int8(-1)) prints 64 ones instead of 8.
This happens only for signed integers (int8, in16 and int32). I guess it's because of the way the conversion between integer types works. From go spec: "Conversions between integer types. If the value is a signed quantity, it is sign extended to implicit infinite precision ....". And there are several conversions to int64 and uint64 in the fmt package. This pathch solves only half of the problem. On a 32 bit system, an fmt.Printf("%b", int(-1)) should still print 64 ones.
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/891049
This is required to make cgo export work on Darwin. Note that
this corrects the stack alignment when calling initcgo to that
required by gcc on amd64.
R=rsc
CC=golang-dev
https://golang.org/cl/907041
The new //export comment marks a Go function as callable from
C. The syntax is "//export NAME" where NAME is the name of
the function as seen from C. If such a comment is seen, cgo
will generate two new files: _cgo_export.h and _cgo_export.c.
The _cgo_export.h file provides declarations which C code may
use to call Go functions. The _cgo_export.c file contains
wrappers, and is to be compiled with gcc.
The changes to Make.pkg support using this from a Go Makefile,
though it could probably be more convenient.
R=rsc
CC=golang-dev
https://golang.org/cl/853042
These functions are used to call from a C function back to a
Go function. This only includes 386 support.
R=rsc
CC=golang-dev
https://golang.org/cl/834045
tested on linux/amd64, linux/386, linux/arm, darwin/amd64, darwin/386.
freebsd untested; will finish in a separate CL.
for now all the panics are errorStrings.
richer structures can be added as necessary
once the mechanism is shaked out.
R=r
CC=golang-dev
https://golang.org/cl/906041
Could not take a signal on threads other than the main thread.
If you look at the spinning binary with dtrace, you can see a
fault happening over and over:
$ dtrace -n '
fbt::user_trap:entry /execname=="boot32" && self->count < 10/
{
self->count++;
printf("%s %x %x %x %x", probefunc, arg1, arg2, arg3, arg4);
stack();
tracemem(arg4, 256);
}'
dtrace: description 'fbt::user_trap:entry ' matched 1 probe
CPU ID FUNCTION:NAME
1 17015 user_trap:entry user_trap 0 10 79af0a0 79af0a0
mach_kernel`lo_alltraps+0x12a
0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef
0: 0e 00 00 00 37 00 00 00 00 00 00 00 1f 00 00 00 ....7...........
10: 1f 00 00 00 a8 33 00 00 00 00 00 01 00 00 00 00 .....3..........
20: 98 ba dc fe 07 09 00 00 00 00 00 00 98 ba dc fe ................
30: 06 00 00 00 0d 00 00 00 34 00 00 00 9e 1c 00 00 ........4.......
40: 17 00 00 00 00 02 00 00 ac 30 00 00 1f 00 00 00 .........0......
50: 00 00 00 00 00 00 00 00 0d 00 00 00 e0 e6 29 00 ..............).
60: 34 00 00 00 00 00 00 00 9e 1c 00 00 00 00 00 00 4...............
70: 17 00 00 00 00 00 00 00 00 02 00 00 00 00 00 00 ................
80: ac 30 00 00 00 00 00 00 1f 00 00 00 00 00 00 00 .0..............
90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
a0: 48 00 00 00 10 00 00 00 85 00 00 00 a0 f2 29 00 H.............).
b0: 69 01 00 02 00 00 00 00 e6 93 04 82 ff 7f 00 00 i...............
c0: 2f 00 00 00 00 00 00 00 06 02 00 00 00 00 00 00 /...............
d0: 78 ee 42 01 01 00 00 00 1f 00 00 00 00 00 00 00 x.B.............
e0: 00 ed 9a 07 00 00 00 00 00 00 00 00 00 00 00 00 ................
f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
...
The memory dump shows a 32-bit exception frame:
x86_saved_state32
gs = 0x37
fs = 0
es = 0x1f
ds = 0x1f
edi = 0x33a8
esi = 0x01000000
ebp = 0
cr2 = 0xfedcba98
ebx = 0x0907
edx = 0
ecx = 0xfedcba98
eax = 0x06
trapno = 0x0d
err = 0x34
eip = 0x1c9e
cs = 0x17
efl = 0x0200
uesp = 0x30ac
ss = 0x1f
The cr2 of 0xfedcba98 is the address that the new thread read
to cause the fault, but note that the trap is now a GP fault with
error code 0x34, meaning it's moved past the cr2 problem and on
to an invaild segment selector. The 0x34 is suspiciously similar
to the 0x37 in gs, and sure enough, OS X forces gs to have
that value in the signal handler, and if your thread hasn't set
up that segment (known as USER_CTHREAD), you'll fault on the IRET
into the signal handler and never be able to handle a signal.
The kernel bug is that it forces segment 0x37 without making sure
it is a valid segment. Leopard also forced 0x37 but had the courtesy
to set it up first.
Since OS X requires us to set up that segment (using the
thread_fast_set_cthread_self system call), we might as well
use it instead of the more complicated i386_set_ldt call to
set up our per-OS thread storage.
Also add some more zeros to bsdthread_register for new arguments
in Snow Leopard (apparently unnecessary, but being careful).
Fixes#510.
R=r
CC=golang-dev
https://golang.org/cl/824046
data just read from the channel.
this will make it easier to
recognize when to garbage
collect and finalize.
R=rsc
CC=golang-dev
https://golang.org/cl/882043
The cycle is *netFD -> cw chanl *netFD in struct ->
same *netFD in channel read buffer.
Because channels are finalized, the cycle makes them
uncollectable. A better fix is to make channels not
finalized anymore, and that will happen, but this is
an easy, reasonable workaround until then.
Another good fix would be to zero the channel receive
buffer entry after the receive. That too will happen.
R=r
CC=golang-dev
https://golang.org/cl/875043
1. Fix bug in GOMAXPROCS when trying to cut number of procs
Race could happen on any system but was
manifesting only on Xen hosted Linux.
2. Fix recover on ARM, where FP != caller SP.
R=r
CC=golang-dev
https://golang.org/cl/880043
Instead of calling the package scope Lex function,
Parse now takes an argument which is used to
do the lexing.
I reverted to having the generated switch
code inside Parse rather than a separate function because
the function needs 7 arguments or a context structure,
which seems unnecessary.
I used yyrun(), not the original $A so that
it's possible to run the backquoted code through gofmt.
R=rsc, ken2, ken3
CC=golang-dev
https://golang.org/cl/879041
This is a documentation enhancement only, without any code
change.
The rationale for documenting this precisely is that Unix
programmers who "know" that errno's value is undefined after
a successful system call may be surprised otherwise and
search to be sure that a zero errno may be relied upon after
successful calls.
R=r, rsc1, rsc
CC=golang-dev
https://golang.org/cl/812044
log: add test of Caller
New regexp in log test is picky and will require some
maintenance, but it catches off-by-one mistakes too.
Fixes#710.
R=gri
CC=esko.luontola, golang-dev
https://golang.org/cl/887043
* correct symbol table size
* do not reorder functions in output
* traceback
* signal handling
* use same code for go + defer
* handle leaf functions in symbol table
R=kaib, dpx
CC=golang-dev
https://golang.org/cl/884041
* adds pass 3 to dequeue from channels eagerly
various other cleanup/churn:
* use switch on cas->send in each pass to
factor out common code.
* longer goto labels, commented at target
* be more agressive about can't happen:
throw instead of print + cope.
* use "select" instead of "selectgo" in errors
* use printf for debug prints when possible
R=ken2, ken3
CC=golang-dev, r
https://golang.org/cl/875041
factor out environment variable checks.
infer $GOROOT etc during build if not set.
it's still necessary to set them for yourself
to use the standard Makefiles.
when running all.bash, don't recompile all the
go packages in run.bash, since make.bash already did.
R=r
CC=golang-dev
https://golang.org/cl/609042
in character and string case mapping routines.
Add a custom mapper for Turkish and Azeri.
A more general solution for deriving the case information from Unicode's
SpecialCasing.txt will require more work.
Fixes#703.
R=rsc, rsc1
CC=golang-dev, mdakin
https://golang.org/cl/824043
instead use pure substring matching to find template values.
this makes stdZulu unnecessary and allows formats
like "20060102 030405" (used in some internet protocols).
this makes Parse not handle years < 0000 or > 9999 anymore.
that seems like an okay price to pay, trading hypothetical
functionality for real functionality.
also changed the comments on the Time struct to use the
same reference date as the format and parse routines.
R=r
CC=golang-dev
https://golang.org/cl/833045
main semantic change is to enforce single argument to panic.
runtime: change to 1-argument panic.
use String method on argument if it has one.
R=ken2, r
CC=golang-dev
https://golang.org/cl/812043
note that sortmain.go has been run through hg gofmt;
only the formatting of the day initializers changed.
i'm happy to revert that formatting if you'd prefer.
stop on error in doc/progs/run
R=r
CC=golang-dev
https://golang.org/cl/850041
the set() method and add() functions. But we rename add() to Var() for consistency.
Also rename FlagValue to Value for simplicity.
Also, delete the check for multiple settings for a flag. This makes it possible to
define a flag that collects values, such as into a slice of strings.
type flagVar []string
func (f *flagVar) String() string {
return fmt.Sprint(v)
}
func (f *flagVar) Set(value string) bool {
if v == nil {
v = make(flagVar, 1)
} else {
nv := make(flagVar, len(v)+1)
copy(nv, v)
v = nv
}
v[len(v)-1] = value
return true
}
var v flagVar
func main() {
flag.Var(&v, "testV", "multiple values build []string")
flag.Parse()
fmt.Printf("v = %v\n", v)
}
R=rsc
CC=golang-dev
https://golang.org/cl/842041
This is a modified version of the open source pprof
from code.google.com/p/google-perftools.
That version is likely to catch up to this one,
but it's still useful to ship our own copy since
we only need the one script from that project,
not all the C++ libraries.
R=r
CC=golang-dev
https://golang.org/cl/783041
In this change I'd like to combine the common code that is
present in syscall_darwin.go and syscall_freebsd.go. I
have three reasons for wanting to do this now:
1. reducing code duplication is nearly always good :-)
2. the duplication will get worse if I duplicate this code
a third time for the NetBSD port I'm working on, which
I need to do almost immediately
3. by making this change all in one lump and ignoring any
commonality with the syscall_linux*.go files the diff
is long but, I think, readable
In future it may be possible to cherry pick functions that
also apply to Linux and put them in (say) syscall_unix.go,
and of course some functions may diverge in future and have
to move out to OS or architecture specific files, but today
I want just the low hanging fruit.
Tested and passed on:
Darwin (Snow Leopard, 10.6): amd64 and 386
FreeBSD (8.0-RELEASE): 386 only(*)
(*) All my virtualisation software has stopped playing nice
with FreeBSD for the moment, so I don't have facilities to
test the amd64 port. As the OS X port is OK and the diff
looks all right to my eyes I shall keep my fingers crossed.
If someone with a FreeBSD/amd64 system cares to test and
report I would be appreciative.
2010-03-27 update: I have replaced my virtualisation software, and have working FreeBSD/i386 and FreeBSD/amd64 virtual machines again.
As I hoped (and expected -- programmers are optimists :-) the code built and passed all but the two currently known to fail tests on FreeBSD/amd64. I rechecked FreeBSD/i386 too: same results.
R=rsc
CC=golang-dev
https://golang.org/cl/751041
- new heuristic: if both the opening { and closing } braces are on the
same line, and the function body doesn't contain comments or is other-
wise too long (e.g. signature too long), it is formatted as a one-line
function
- related cleanups along the way
- gofmt -w src misc led to no additional changes as expected
R=rsc, rsc1
CC=golang-dev, ken2, r
https://golang.org/cl/758041
up to date.
No functional change, but when these files are regenerated
they change, leading to noisy diffs when working in the
syscall package.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/705043
- change the various url-xxx formatters to return a relative URL path
- make the leading '/' for URLs explicit in the template
- on the way change some |html formatters to |html-esc
(html should only be used for formatting AST nodes)
R=rsc, r
CC=golang-dev
https://golang.org/cl/740041
Instead of returning the index lookup result via
RPC which has to be corrected for the client,
simply render it on the server and return the
final output.
R=rsc, r
CC=golang-dev
https://golang.org/cl/669041
and highlight special words, if provided. Also:
- related cleanups in src/pkg/go/doc/comment.go
- fix typos in src/cmd/goinstall/doc.go
Fixes#672.
R=rsc
CC=adg, golang-dev
https://golang.org/cl/601042
The command-line search is using a running webserver
as index server; i.e., the search result is reflecting
the index at the server. See the documentation for
details.
Usage: godoc -q query1 query2 ...
Known issue: Results don't show the all-important
line numbers yet due to the way the index is organized.
Next CL.
R=rsc, r
CC=golang-dev
https://golang.org/cl/648041
- mkall now generates syscall stabs automatically
- we can call unicode versions of winapi now
- introduce GetErrstr function to fetch error text given errno
- general cleanup
R=rsc
CC=golang-dev
https://golang.org/cl/562041
for identifiers in Go source code
- at the moment just show identifier kind (var, func, etc.) and name
(eventually should show declaration, type, etc.)
- JavaScript parts by adg
R=rsc
CC=adg, golang-dev
https://golang.org/cl/578042
a /*-style comment at the end of a file
Some minor cleanups/typo fixes along the way.
Fixes an issue where that newline was removed
after applying gofmt.
R=r
CC=golang-dev
https://golang.org/cl/476043
visible only when enabling internal debug mode:
- in rare cases expression depth can underflow
- when printing a single labeled statement, indentation
may underflow if not setup correctly
R=rsc
CC=golang-dev
https://golang.org/cl/484041
- on the commandline: godoc -x big
- in a webpage: provide form parameter ?m=src
Known issues:
- Positioning of comments incorrect in several cases. Separate CL.
- Need a link/menu to switch between different modes of presentation
in the web view.
R=rsc
CC=golang-dev
https://golang.org/cl/376041
2. add complex algorithm for map/chan
3. test for use of complex in
array, slice, field, chan, map,
field, pointer.
R=rsc
CC=golang-dev
https://golang.org/cl/384041
trivial stuff
lex.c: these prototypes are in a.h
asm.c: unused variables
arm-pass.txt deal w/ sieve.go rename and addition of sieve2.go
R=kaib, rsc
CC=golang-dev
https://golang.org/cl/244041
This keeps fragmentation from delaying
garbage collections (and causing more fragmentation).
Cuts fresh godoc (with indexes) from 261M to 166M (120M live).
Cuts toy wc program from 50M to 8M.
Fixes#647.
R=r, cw
CC=golang-dev
https://golang.org/cl/257041
Previously we would require safe primes for our RSA key generation.
Since this took rather a long time, this removes the requirement that
the primes be safe.
OpenSSL doesn't use safe primes for RSA key generation either
(openssl-0.9.8l/crypto/rsa/rsa_gen.c:122)
Fixes#649.
R=rsc
CC=golang-dev
https://golang.org/cl/253041
Use hardware sqrt for faster hypot; preserve software-only
hypot as hypotGo (like sqrtGo); enable benchmarking of
hypotGo.
R=rsc
CC=golang-dev
https://golang.org/cl/229049
probablyPrime would return false negatives in some cases.
This code has now been tested against GMP for several million iterations without issues.
Fixes#638.
R=rsc
CC=golang-dev
https://golang.org/cl/252041
- only manual changes are in src/pkg/go/printer/nodes.go
- use a heuristic to determine "outliers" such that not entire composites are
forced to align with them
- improves several places that were not unligned before due too simple heuristic
- unalignes some cases that contain "outliers"
- gofmt -w src misc
Fixes#644.
R=rsc, r
CC=golang-dev
https://golang.org/cl/241041
Set defaults to the same values as various C DNS resolver libraries.
Use the timeout value from the configuration.
Fixes#632.
R=rsc
CC=golang-dev
https://golang.org/cl/223094
This permits more flexibility with cgo and swig in cases where
the program is run on a machine other than the one on which it
is built. Rather than storing the absolute path to the shared
library in the DT_NEEDED entry, we can store just the name,
and let the dynamic linker find it using DT_RUNPATH or the
LD_LIBRARY_PATH environment variable.
R=rsc
CC=golang-dev
https://golang.org/cl/223068
Background: This didn't matter until recently, because '\n' don't appear as token starts
in source code and thus the exact position was irrelevant (and set as was easiest in the
code). With auto semicolon insertion, a virtual semicolon may be inserted when a '\n' is
seen. The position of the semicolon is the position of the '\n'. Without this fix, these
semicolons appeared on the next line instead of the line where they were inserted. This
affected the association of comments to declarations in the parser. As a result, some
lead comments where considered line comments, not collected in the ast, and not shown in
godoc pages. (This affected only godoc pages, not gofmt-formatted programs).
Fixes#592.
R=rsc
CC=golang-dev
https://golang.org/cl/224068
- gofmt -w src misc
- improves several lists and fixes minor degradation introduced
with the fix for issue 628
- removed some dead code (stringList)
R=rsc
CC=golang-dev
https://golang.org/cl/223058
- applied gofmt to src and misc
Note: This fix improved formatting of src/pkg/math/all_test.go but leads
to a degradation in src/pkg/exp/4s/xs.go. The latter happened to "work"
before accidentally. Fixing the alignment in that case in general will
be a separate CL.
Fixes#628.
R=rsc
CC=golang-dev
https://golang.org/cl/223054
path.Match() errors out when testing "*x" against "xxx"
because it matches the star non-greedily. Ensure that
the last chunk consumes the rest of the name.
R=r, rsc
CC=golang-dev
https://golang.org/cl/223050
- always include position information about opening/closing parens/braces
- replace uses of []*ast.Field with *ast.FieldList
Fixes#473.
R=rsc
CC=golang-dev
https://golang.org/cl/223043
- remove several TODOs
- as a side-effect, comment stylers are now used always and comments
will be properly colored in godoc pkg documentation pages (and not
only when looking at source text)
R=rsc
CC=golang-dev
https://golang.org/cl/222041
- go/ast: removed StringList (not needed anymore)
- go/ast: changed import path and field list tag to a single string
- updated all dependencies
R=rsc
CC=golang-dev
https://golang.org/cl/217056
Compliance issue addressed here: POST requests carrying form data are required
to use "identity" transfer encoding by common nginx and apache server configurations,
e.g. wordpress.com (and many others). So, Request needed to be able to send
non-chunked encodings.
Thus, Request is extended to support identity and chunked encodings, like
Response. Since the Read() and Write() logic are shared by both (and are
quite long), it is exported in a separate file transfer.go.
R=rsc
CC=golang-dev
https://golang.org/cl/217048
Example use: godoc -path=/home/user1:/home/build/foo -http=:6666
will start a local godoc that maps urls starting with /pkg/user1 or
/pkg/foo to the respective roots specified in the path.
Missing: Handling of overlapping package directories, multiple
packages per directory.
R=rsc
CC=golang-dev
https://golang.org/cl/206078
Try to avoid infinite recursion if String fails due to
printing a bad type.
Add test for String method with named basic types.
R=r
CC=golang-dev
https://golang.org/cl/207102
because free needs to mark the block as freed to
coordinate with the garbage collector.
(in C++ free can blindly put the block on the free list,
no questions asked, so the cache saves some work.)
R=iant
CC=golang-dev
https://golang.org/cl/206069
not trying to be secure, just not repetitive
(sending with the same id repeatedly makes
some resolvers stop replying.)
eventually we'll replace this with linking against
the system's native resolver.
R=p
CC=golang-dev
https://golang.org/cl/207051
* specialize sweepspan as sweepspan0 and sweepspan1.
* in sweepspan1, inline "free" to avoid expensive mlookup.
R=iant
CC=golang-dev
https://golang.org/cl/206060
Added tests and benchmarks for Exp2 (special cases same
as Exp). Log1p also enhances speed of inverse hyperbolics.
R=rsc
CC=golang-dev
https://golang.org/cl/206058
Add functions, tests and benchmarks. Fix typos in comments
in expm1 and hypot_386. Fix Acosh domain error in benchmark
test.
R=rsc
CC=golang-dev
https://golang.org/cl/204069
Averaged times (in microseconds) for drawing an 800x600 rectangle
are listed below. The summary is: around a 100x improvement.
draw.Draw call times were typically linear in the number of pixels
touched (i.e. drawing an 800x600 rectangle took 100x as much time as
drawing an 80x60 rectangle).
Before this change, there was only the general-but-slow code path.
When drawing any src with a 50%-opaque mask: 237300 us
When drawing any src with a nil mask: 50100 us
After this change, the 50%-opaque mask case is unchanged.
For an *image.RGBA dst and nil mask and...
...a uniform color (i.e. an image.ColorImage) src: 282 us
...another *image.RGBA src: 615 us.
For the curious, an intermediate implementation detected the special
cases but used simple nested for loops instead of the built-in copy
function. The respective times (compared to 282 and 615 for the
final implementation, or 50100 for the original) were 3110 and 3573.
Times were measured with 8g/8l on my laptop. I haven't tried gccgo
or other architectures.
R=r, rsc
CC=golang-dev
https://golang.org/cl/201048
Go source outside $GOROOT.
If the argument is a path starting with / or ., disregard $GOROOT.
Also, disable the check for package name matching the directory,
which is counterproductive in this case.
Apologies for the violence to the code but we need some help documenting
Go code outside the standard repository.
R=gri
CC=golang-dev
https://golang.org/cl/201064
Marshalling:
* Fixes several silly bugs.
* Support the RawContents type.
* Support the RawValue type.
* Recurse into nested interface{}.
Both directions:
* Better handling of SETs. You can now tag an element in a
structure with "set" to get the correct tag for serialisation.
* For types that aren't in a structure, you can now name them
with "SET" on the end.
* SETs are no longer implicitly treated as SEQUENCEs.
R=rsc
CC=golang-dev
https://golang.org/cl/201049
Make sure to print a time zone when formatting even if none is defined.
Add a comment introducing lookupTimezone (not lookupTimeZone).
Fixes isse 577.
R=rsc
CC=golang-dev
https://golang.org/cl/196090
Added special cases to comments for asin.go and fabs.go.
Added Trunc() to floor.go and floor_386.s. Fixed formatting
error in hypot_386.s Added new functions Acosh, Asinh,
Atanh, Copysign, Erf, Erfc, Expm1, and Log1p. Added
386 FPU version of Fmod. Added tests, benchmarks, and
precision to expected results in all_test.go. Edited
makefile so it all compiles.
R=rsc
CC=golang-dev
https://golang.org/cl/195052
* renamed channels to say what gets sent
* use channel closed status instead of racy check of boolean
R=nigeltao_golang
CC=golang-dev
https://golang.org/cl/196065
In particular, add field-value tokenizer which respects quoting rules.
The code is intended for use in tokenizing the Transfer-Encoding and Trailer fields.
The lexing function is not connected to the main parsing code yet (in the next CL).
R=rsc
CC=golang-dev
https://golang.org/cl/190085
(1) http.Response must close resp.Body after writing.
(2) Case when resp.Body != nil and resp.ContentLength = 0 should not be
treated as an error in Response.Write, because this is what
ReadResponse often returns.
(3) Changed body.th to body.hdr for readability.
R=rsc
CC=golang-dev
https://golang.org/cl/194084
Seems to be enabled spuriously during godoc (can't see why),
producing errors like:
parser.parseDir: src/pkg/http/server.go:159:16: 'Write' declared already at src/pkg/http/request.go:140:21 (and 4 more errors)
R=r
CC=golang-dev
https://golang.org/cl/194119
- provide scope to parse functions; if non-nil, parser uses the scope
to declare and lookup identifiers
- resolve forward references where possible
R=rsc
CC=golang-dev
https://golang.org/cl/194098
* example-based syntax errors (go.errors)
* enable bison's more specific errors
and translate grammar token names into
tokens like ++
* test cases
R=ken2, r, ken3
CC=golang-dev
https://golang.org/cl/194085
Hostname reads the file /proc/sys/kernel/hostname to determine
the value it returns. Some people set this to a Fully Qualified
Doamin Name. At least one implementation of /bin/hostname
truncates the name it gets (often from the "uname" system call)
at the first dot unless it is given a "-f" flag. This change makes
the unit test also truncate at the first dot and checks if the strings
then match. This seems more portable than adding an extra flag
to the called /bin/hostname program.
R=rsc
CC=golang-dev
https://golang.org/cl/181097
If your $PATH does not include $GOBIN then
the build fails at the first *.a file, which
is probably when running the tests for tar
R=rsc
CC=golang-dev
https://golang.org/cl/193066
Added 386 FPU version of Hypot; modified all_test.go to test
Hypot with large arguments. Also edited sqrt.go to remove
Sqrt(0) as a special case.
R=rsc
CC=golang-dev
https://golang.org/cl/186180
bonus: type switch now detects multiple uses of identical interface types.
bonus: interface types are now order-independent, following the spec.
R=ken2
CC=golang-dev
https://golang.org/cl/194053
eliminates spurious multiple initialization errors.
give more information in the multiple init errors that remain.
Fixes#87.
R=r
CC=golang-dev
https://golang.org/cl/194052
5g/6g/8g: add import statements to export metadata, mapping package path to package name.
recognize "" as the path of the package in export metadata.
use "" as the path of the package in object symbol names.
5c/6c/8c, 5a/6a/8a: rewrite leading . to "". so that ·Sin means Sin in this package.
5l/6l/8l: rewrite "" in symbol names as object files are read.
gotest: handle new symbol names.
gopack: handle new import lines in export metadata.
Collectively, these changes eliminate the assumption of a global
name space in the object file formats. Higher level pieces such as
reflect and the computation of type hashes still depend on the
assumption; we're not done yet.
R=ken2, r, ken3
CC=golang-dev
https://golang.org/cl/186263
Other code fixing:
- Fixed bugs in get32.
- Fix code for parsing display string (as a new function).
- Fix code for connecting to X server. The old code only work
if the server is listening to TCP port, otherwise it doesn't
work (at least in my PC).
R=nigeltao_golang, rsc, jhh
CC=golang-dev
https://golang.org/cl/183111
* switch to real dot (.) instead of center dot (·) everywhere in object files.
before it was half and half depending on where in the name it appeared.
* in 6c/6a/etc identifiers, · can still be used but turns into . immediately.
* in export metadata, replace package identifiers with quoted strings
(still package names, not paths).
R=ken2, r
CC=golang-dev
https://golang.org/cl/190076
need better management of mach semaphores eventually
but this avoids allocating them for uncontended Locks.
R=r
CC=agl1, golang-dev
https://golang.org/cl/190079
limitations:
poor error handling
teardown not done
exporter must send, importer must receive
testing is rudimentary at best
R=rsc
CC=golang-dev
https://golang.org/cl/186234
Was trying to interpret raw body as chunked body.
Add test for ReadResponse.
Fixes#544.
R=r, petar-m
CC=golang-dev, shadowice
https://golang.org/cl/190068
Wrong position information for identifier
could lead to destructive reformatting of
source via hg gofmt.
R=rsc
CC=golang-dev
https://golang.org/cl/189100
- Identifiers refer now to the language entity (Object)
that they denote. At the moment this is at best an
approximation.
- Initial data structures for language entities (Objects)
and expression types (Type) independent of the actual
type notations.
- Initial support for declaring and looking up identifiers.
- Updated various dependent files and added support functions.
- Extensively tested to avoid breakage. This is an AST change.
R=rsc
CC=golang-dev, rog
https://golang.org/cl/189080
* move memory code into $GOOS-specific directory.
* allow printing of static strings < 256 bytes.
(dynamic strings will bump maxstring as they are allocated.)
* use cgo2c for runtime.mal.
R=r, dho
CC=golang-dev
https://golang.org/cl/186143
by definition, they know what they are building for.
makes it easier to switch architectures when testing.
% 6g x.go
% 6l x.6
% 6.out
"Wed Jan 13 10:57:46 EST 2010"
% 8g x.go
% 8l x.8
% 8.out
"Wed Jan 13 10:57:46 EST 2010"
% echo $GOARCH
%
R=rsc
CC=golang-dev
https://golang.org/cl/186116
Emphasis on minimal interface and fast scanning.
Recognizes all Go literals by default. Easily
configurable to recognize different whitespace
characters and tokens.
Provides detailed position information for each
token.
R=rsc, r
CC=golang-dev
https://golang.org/cl/181160
Added special case tests to all_test.go for Fmod. Fixed Fmod [hung
for Fmod(+/-Inf, <finite>)]. Also added test for Ceil in all_test.go.
R=rsc
CC=golang-dev
https://golang.org/cl/186076
now printed using the semicolon-free style.
Removed NoSemis flag and mechanism dealing with
optional semicolons from go/printer.
Updated all go/printer output files using the
semi-colon free style.
Changes have no formatting impact on checked in
go code under src and misc.
R=rsc, r
CC=golang-dev
https://golang.org/cl/184068
Addresses some cases and respects position of
closing parenthesis; but doesn't work yet in
all cases. Heuristic; will need to be replaced
by a real solution eventually. Does not cause
existing code to be reformatted.
R=r
CC=golang-dev, rsc
https://golang.org/cl/186043
Using FUCOMIP instead of the FUCOMP-FSTSW-SAHF sequence gives better performance and saves code space.
R=rsc
CC=golang-dev
https://golang.org/cl/183139
The model is that formatString is a a representation of a standard time,
and that Format converts the time to that representation.
Standard representaitons are defined for ANSIC, RFC850, RFC1123, and ISO8601.
There's also a humane Kitchen fomat: 3:04PM.
R=rsc, benolive, cw
CC=golang-dev
https://golang.org/cl/181130
used only for debugging, debug.go is not normally part of the package source.
also add a dump program to call it.
R=rsc
CC=golang-dev
https://golang.org/cl/183075
tabs after an empty line where not converted.
Also, made it more robust in the presence of
(unexpected) ' ' and '\v' chars in indentation
mode.
R=r
CC=golang-dev
https://golang.org/cl/181085
(I was looking at this code accidentally because of some gofmt
issues and thought that one could write this more effectively.
You may have deliberately chosen not to use ranges here to make
the index range clearer. Just let me know.)
R=agl, agl1
CC=golang-dev
https://golang.org/cl/181084
Listener contains private members and 6g now enforces that private
members cannot be assigned outside of their package.
R=rsc
CC=golang-dev
https://golang.org/cl/183073
This was convenient for me to have without being forced
to parse the regexp myself. I'd understand if it's not
really wanted, but I also think that some meta information
about compiled regexps would be fine.
R=r, rsc
CC=golang-dev
https://golang.org/cl/183044
for now, it's amd64 and 386 only but it's trivial to add more.
Fixes#385.
(why couldn't it have been issue 386?)
tested for amd64 and 386 on darwin.
R=rsc
CC=golang-dev
https://golang.org/cl/182043
It's expected to be shared between all files so that all types are output.
Fixes bug reported on mailing list by Peter Froehlich.
R=rsc, phf
CC=golang-dev
https://golang.org/cl/183043
as it is not needed anymore (only one impl.
of vector package).
Makefile, vector_test.go, and nogen_test.go
were modified manually (find/replace), the
other files (intvector_test.go, strinvector_test.go
are generated).
Runs all tests.
R=r
https://golang.org/cl/182041
Manual changes to the following files:
src/pkg/Makefile
src/pkg/exp/vector/Makefile (now: src/pkg/container/vector/Makefile)
R=rsc, r
CC=golang-dev
https://golang.org/cl/181041
- use an interface {Get()}
- implement Get for maps, slices
- for slices, retrieves the address of the end of the array, which will give the
same value for every slice of the same array.
R=rsc
CC=golang-dev
https://golang.org/cl/179129
(Thanks to ken and rsc for pointing this out)
rsc:
ken pointed out that there's a race in the new
one-lock-per-channel code. the issue is that
if one goroutine has gone to sleep doing
select {
case <-c1:
case <-c2:
}
and then two more goroutines try to send
on c1 and c2 simultaneously, the way that
the code makes sure only one wins is the
selgen field manipulation in dequeue:
// if sgp is stale, ignore it
if(sgp->selgen != sgp->g->selgen) {
//prints("INVALID PSEUDOG POINTER\n");
freesg(c, sgp);
goto loop;
}
// invalidate any others
sgp->g->selgen++;
but because the global lock is gone both
goroutines will be fiddling with sgp->g->selgen
at the same time.
This results in a 7% slowdown in the single threaded case for a
ping-pong microbenchmark.
Since the cas predominantly succeeds, adding a simple check first
didn't make any difference.
R=rsc
CC=golang-dev
https://golang.org/cl/180068
Allows stand-alone types (e.g. []int as patterns) and doesn't require
a semicolon at the end (which are now mandatory terminators).
- Fix a matcher bug.
R=rsc
CC=golang-dev
https://golang.org/cl/179088
This is not a complete JPEG implementation (e.g. it does not handle
progressive JPEGs or restart markers), but I was able to take a photo
with my phone, and view the resultant JPEG in pure Go.
The decoder is simple, but slow. The Huffman decoder in particular
should be easily improvable, but optimization is left to future
changelists. Being able to inline functions in the inner loop should
also help performance.
The output is not pixel-for-pixel identical to libjpeg, although
identical behavior isn't necessarily a goal, since JPEG is a lossy
codec. There are at least two reasons for the discrepancy.
First, the inverse DCT algorithm used is the same as Plan9's
src/cmd/jpg, which has different rounding errors from libjpeg's
default IDCT implementation. Note that libjpeg actually has three
different IDCT implementations: one floating point, and two fixed
point. Out of those four, Plan9's seemed the simplest to understand,
partly because it has no #ifdef's or C macros.
Second, for 4:2:2 or 4:2:0 chroma sampling, this implementation does
nearest neighbor upsampling, compared to libjpeg's triangle filter
(e.g. see h2v1_fancy_upsample in jdsample.c).
The difference from the first reason is typically zero, but sometimes
1 (out of 256) in YCbCr space, or double that in RGB space. The
difference from the second reason can be as large as 8/256 in YCbCr
space, in regions of steep chroma gradients. Informal eyeballing
suggests that the net difference is typically imperceptible, though.
R=r
CC=golang-dev, rsc
https://golang.org/cl/164056
into consts in the resulting Go source. Previously known as issue 161047,
which I deleted accidentally. Fixes issue 207.
R=rsc
https://golang.org/cl/166059
* better error for lookup of unexported field
* do not assign "ideal string" type to typed string literal
* do not confuse methods and fields during interface check
Fixes#410.
Fixes#411.
Fixes#426.
R=ken2
https://golang.org/cl/179069
parsing and printing to new syntax.
Use -oldparser to parse the old syntax,
use -oldprinter to print the old syntax.
2) Change default gofmt formatting settings
to use tabs for indentation only and to use
spaces for alignment. This will make the code
alignment insensitive to an editor's tabwidth.
Use -spaces=false to use tabs for alignment.
3) Manually changed src/exp/parser/parser_test.go
so that it doesn't try to parse the parser's
source files using the old syntax (they have
new syntax now).
4) gofmt -w src misc test/bench
5th and last set of files.
R=rsc
CC=golang-dev
https://golang.org/cl/180050
parsing and printing to new syntax.
Use -oldparser to parse the old syntax,
use -oldprinter to print the old syntax.
2) Change default gofmt formatting settings
to use tabs for indentation only and to use
spaces for alignment. This will make the code
alignment insensitive to an editor's tabwidth.
Use -spaces=false to use tabs for alignment.
3) Manually changed src/exp/parser/parser_test.go
so that it doesn't try to parse the parser's
source files using the old syntax (they have
new syntax now).
4) gofmt -w src misc test/bench
4th set of files.
R=rsc
CC=golang-dev
https://golang.org/cl/180049
parsing and printing to new syntax.
Use -oldparser to parse the old syntax,
use -oldprinter to print the old syntax.
2) Change default gofmt formatting settings
to use tabs for indentation only and to use
spaces for alignment. This will make the code
alignment insensitive to an editor's tabwidth.
Use -spaces=false to use tabs for alignment.
3) Manually changed src/exp/parser/parser_test.go
so that it doesn't try to parse the parser's
source files using the old syntax (they have
new syntax now).
4) gofmt -w src misc test/bench
3rd set of files.
R=rsc
CC=golang-dev
https://golang.org/cl/180048
parsing and printing to new syntax.
Use -oldparser to parse the old syntax,
use -oldprinter to print the old syntax.
2) Change default gofmt formatting settings
to use tabs for indentation only and to use
spaces for alignment. This will make the code
alignment insensitive to an editor's tabwidth.
Use -spaces=false to use tabs for alignment.
3) Manually changed src/exp/parser/parser_test.go
so that it doesn't try to parse the parser's
source files using the old syntax (they have
new syntax now).
4) gofmt -w src misc test/bench
1st set of files.
R=rsc
CC=agl, golang-dev, iant, ken2, r
https://golang.org/cl/180047
parsing and printing to new syntax.
Use -oldparser to parse the old syntax,
use -oldprinter to print the old syntax.
2) Change default gofmt formatting settings
to use tabs for indentation only and to use
spaces for alignment. This will make the code
alignment insensitive to an editor's tabwidth.
Use -spaces=false to use tabs for alignment.
3) Manually changed src/exp/parser/parser_test.go
so that it doesn't try to parse the parser's
source files using the old syntax (they have
new syntax now).
4) gofmt -w src misc test/bench
2nd set of files.
R=rsc
CC=golang-dev
https://golang.org/cl/179067
mustgetc reports unexpected EOF as SyntaxError. using
mustgetc seems to be a better approach than letting the
caller handle unexpected EOF every time.
name: the second if statement should explicitly return
ok==false.
R=rsc
https://golang.org/cl/174083
1. If all data is exhausted using Read then a following Next will
fail as if it saw EOF. (Test case added.)
2. Seeking isn't always possible (i.e. sockets and pipes). Fallback
to read. (Test case added.)
3. Fix to readHeader (cleaner fix pointed out by rsc).
(TestReader modified.)
4. When Read has consumed all the data, don't try to read 0 bytes from reader.
In cases where tr.nb is zero we attempt to read zero bytes and thus
never see an EOF (this is most easily seen when the 'tar source' is
something like bytes.Buffer{} as opposed to os.File).
5. If write is used to the point of ErrWriteTooLong, allow additional file entries.
6. Make close work as expected. That is any further Write or
WriteHeader attempts will result in ErrWriteAfterClose.
Fixes#419.
R=rsc, dsymonds1
https://golang.org/cl/162062
Doing rm -rf /pkg/.. blindly isn't nice. It could have
unintended consequences.
Secondly set bash to abort on (unexpected) errors.
R=dho, rsc
CC=golang-dev
https://golang.org/cl/176056
- no need to replace comments for stand-alone blocks
- always print string concatenations with interspersed "+"
(remove option)
- minor cleanups
R=rsc
https://golang.org/cl/174076
This change removes the necessity to have GOBIN in $PATH,
and also doesn't assume that the build is being run from
$GOROOT/src. This is a minimal set of necessary changes
to get Go to build happily from the FreeBSD ports
collection.
R=rsc
CC=golang-dev
https://golang.org/cl/171044
- oldparser parse old syntax (required semicolons)
- oldprinter print old syntax (required semicolons)
By default, these flags are enabled for now.
Setting -oldparser=false has no effect until go/parser is changed
to accept the new syntax.
Enabled exp/parser in Makefile; update dependent exp/eval.
R=rsc
https://golang.org/cl/174051
UTF-8 string, Yconv() converts it into an octal sequence. If the
string converted to more than 30 bytes, the str buffer would
overflow. For example, 4 Greek runes became 32 bytes, 3 Hiragana
runes became 36 bytes, and 2 Gothic runes became 32 bytes. In
8l, 6l and 5l the function is Sconv(). For some reason, only 5l uses
the constant STRINGSZ (defined as 200) for the buffer size.
R=rsc
https://golang.org/cl/168045
FreeBSD was passing stk as the new thread's stack base, while
stk is the top of the stack in go. The added check should cause
a trap if this ever comes up in any new ports, or regresses
in current ones.
R=rsc
CC=golang-dev
https://golang.org/cl/167055
nodes in the tree are nested with respect to one another.
a simple change to the Visitor interface makes it possible
to do this (for example to maintain a current node-depth, or a
knowledge of the name of the current function).
Visit(nil) is called at the end of a node's children;
this make possible the channel-based interface below,
amongst other possibilities.
It is still just as simple to get the original behaviour - just
return the same Visitor from Visit.
Here are a couple of possible Visitor types.
// closure-based
type FVisitor func(n interface{}) FVisitor
func (f FVisitor) Visit(n interface{}) Visitor {
return f(n);
}
// channel-based
type CVisitor chan Visit;
type Visit struct {
node interface{};
reply chan CVisitor;
};
func (v CVisitor) Visit(n interface{}) Visitor
{
if n == nil {
close(v);
} else {
reply := make(chan CVisitor);
v <- Visit{n, reply};
r := <-reply;
if r == nil {
return nil;
}
return r;
}
return nil;
}
R=gri
CC=rsc
https://golang.org/cl/166047
Roughly 33% faster for simple cases, probably more for complex ones.
Before:
mallocs per Sprintf(""): 4
mallocs per Sprintf("xxx"): 6
mallocs per Sprintf("%x"): 10
mallocs per Sprintf("%x %x"): 12
Now:
mallocs per Sprintf(""): 2
mallocs per Sprintf("xxx"): 3
mallocs per Sprintf("%x"): 5
mallocs per Sprintf("%x %x"): 7
Speed improves because of avoiding mallocs and also by sharing a bytes.Buffer
between print.go and format.go rather than copying the data back after each
printed item.
Before:
fmt_test.BenchmarkSprintfEmpty 1000000 1346 ns/op
fmt_test.BenchmarkSprintfString 500000 3461 ns/op
fmt_test.BenchmarkSprintfInt 500000 3671 ns/op
Now:
fmt_test.BenchmarkSprintfEmpty 2000000 995 ns/op
fmt_test.BenchmarkSprintfString 1000000 2745 ns/op
fmt_test.BenchmarkSprintfInt 1000000 2391 ns/op
fmt_test.BenchmarkSprintfIntInt 500000 3751 ns/op
I believe there is more to get but this is a good milestone.
R=rsc
CC=golang-dev, hong
https://golang.org/cl/166076
For 386 we use the [f]statfs64 system call, which takes three
parameters: the filename, the size of the statfs64 structure,
and a pointer to the structure itself.
R=rsc
https://golang.org/cl/166073
On a microbenchmark that ping-pongs on lots of channels, this makes
the multithreaded case about 20% faster and the uniprocessor case
about 1% slower. (Due to cache effects, I expect.)
R=rsc, agl
CC=golang-dev
https://golang.org/cl/166043
Use them in Copy and Copyn.
Speed up ReadFile by using ReadFrom and avoiding Copy altogether (a minor win).
R=rsc, gri
CC=golang-dev
https://golang.org/cl/166041
tabs for indentation even if -spaces is set.
Changes to gofmt:
- added -tabindent flag
- don't recompute parser and printer mode repeatedly
Changes to go/printer:
- provide new printing mode TabIndent
Changes to tabwriter:
- implement new mode TabIndent to use tabs independent
of the actual padding character for leading empty columns
- distinguish between minimal cell width and tab width
(tabwidth is only used if the output contains tabs,
minwidth and padding are always considered)
- fixed and added more comments
- some additional factoring
By default, -tabindent is disabled and the default gofmt
behavior is unchanged. By setting -spaces and -tabindent,
gofmt will use tabs for indentation but do any other
alignment with spaces. This permits a user to change the
visible indentation by simply changing the editor's tab
width and the code will remain properly aligned without
the need to rerun gofmt.
R=rsc
https://golang.org/cl/163068
How to reproduce:
$ mkdir /tmp/foo
$ cp /dev/null /tmp/foo/bar.go
$ chmod -r /tmp/foo/bar.go
$ gofmt /tmp/foo
open /tmp/foo/bar.go: permission denied
$ echo $? # should echo 2
0
$
Maybe you need to put a call to time.Sleep at the beginning of report().
R=gri
CC=golang-dev
https://golang.org/cl/164073
1) need to send slice and array types (was only sending element types)
2) compatibleType needs to use decoder's type map
R=rsc
CC=golang-dev
https://golang.org/cl/164062
Opening /dev/stdin can sometimes fail. For example, in the acme editor,
executing "Edit ,|gofmt" fails with:
open /dev/stdin: no such device or address
Executing "Edit ,|ls -l /dev/stdin /proc/self/fd/0" gives:
lrwxrwxrwx 1 root root 15 2009-09-07 02:17 /dev/stdin -> /proc/self/fd/0
lrwx------ 1 fhs users 64 2009-11-26 22:05 /proc/self/fd/0 -> socket:[5528230]
(This is my first change, and I've signed the individual contributor license agreement.)
R=rsc, gri
CC=golang-dev
https://golang.org/cl/162041
- simplified dealing with parse errors: no need to intersperse them in the source
- improve visibility of highlighted identifiers by showing them in bold
R=rsc
https://golang.org/cl/163051
The python script needs a checkout of xcb/proto to generate
an xproto.go file, which together with xgb.go provide functions
to access all of the core X11 protocol requests. I have included the
generated file.
Extensions and authentication methods are not implemented.
R=r, rsc, nigeltao_golang
https://golang.org/cl/162053
this is the exact same thing issue #115 is about. fix makefiles to use relative
path to work in the case we have whitespaces as part of GOROOT.
R=rsc
https://golang.org/cl/162055
Ideally, the C name would come from the typedef or pointer that
references the Size<0 type, but we can't easily generate this without
performing a look-ahead to see if any referencing type will become opaque.
Fixes#334.
Fixes#281.
R=rsc
https://golang.org/cl/161056
This provides an experimental X11 backend for the exp/draw interface.
It does not aim to provide a complete implementation of the X11 client protocol.
This works for me (Ubuntu Hardy 8.04, GOARCH=386). Your mileage my vary.
R=r, rsc, r1
CC=golang-dev
https://golang.org/cl/156109
before this change, if pkg/Make.deps is missing or broken, clean.bash fails and the build dies
but not until much later.
add freebsd to error message about valid values of $GOOS
TODO: would be nice if this process exited when an error occurred. subshells make it hard
R=rsc
CC=golang-dev
https://golang.org/cl/160065
the bash scripts and makefiles for building go didn't take into account
the fact $GOROOT / $GOBIN could both be directories containing whitespaces,
and was not possible to build it in such a situation.
this commit adjusts the various makefiles/scripts to make it aware of that
possibility, and now it builds successfully when using a path with whitespaces
as well.
Fixes#115.
R=rsc, dsymonds1
https://golang.org/cl/157067
getquoted() currently checks for whitespaces and returns nil
if it finds one. this prevents us from having go in a path
containing whitespaces, as the #pragma dynld directives are
processed through the said function.
this commit makes getquoted() accept whitespaces, and this is
also needed for solving issue #115.
R=rsc
https://golang.org/cl/157066
a little slow, but usable (speed unchanged when not using -r)
tweak go/printer to handle nodes without line numbers
more gracefully in a couple cases.
R=gri
https://golang.org/cl/156103
Meant as illustration of the Go pattern that is using
goroutines and channels to handle exceptional situations.
Note: There is no need for "Finally" since the
"try block" (the function f supplied to Try)
cannot do a Smalltalk-style non-local return
and terminate the function surrounding Try.
Replaces CL 157083.
R=r, rsc
https://golang.org/cl/157087
* add runtime sliceslice1 for x[lo:]
* remove runtime arraytoslice, rewriting &arr into arr[0:len(arr)].
* port cgen_inline into 8g, 5g.
* use native memmove in maps
R=ken2
https://golang.org/cl/157106
Allows the developer to pass a map either by itself for
evaluation, or inside a struct. Access to data inside
maps is identical to the current system for structs, ie.
-Psuedocode-
mp map[string]string = {
"header" : "A fantastic header!",
"footer" : "A not-so-fantastic footer!",
}
template.Execute(mp)
...can be accessed using {header} and {footer} in
the template. Similarly, for maps inside structs:
type s struct {
mp map[string]string,
}
s1 = new s
s1.mp["header"] = "A fantastic header!";
template.Execute(s1)
...is accessed using {mp.header}. Multi-maps, ie.
map[string](map[string]string) and maps of structs
containing more maps are unsupported, but then, I'm
not even sure if that's supported by the language.
Map elements can be of any type that can be written
by the formatters. Keys should really only be strings.
Fixes#259.
R=r, rsc
https://golang.org/cl/157088
There's no functional change here. io gives the Read and Write methods byte slice arguments, but tar called them uint8. It's the same thing, but I think this is clearer and it matches what other packages do.
R=rsc
CC=golang-dev
https://golang.org/cl/157099
1) if char class contains a single character, make it a single character.
(this is used to quote, e.g. [.] rather than \.
2) if regexp begins with ordinary text substring, use plain string match to start engine
R=rsc
CC=golang-dev
https://golang.org/cl/157095
No benchmarks are run unless the --benchmarks=<regexp> flag
is specified on the gotest command line. This change includes
sample benchmarks for regexp.
% gotest --benchmarks=.*
(standard test output redacted)
testing.BenchmarkSimpleMatch 200000 7799 ns/op
testing.BenchmarkUngroupedMatch 20000 76898 ns/op
testing.BenchmarkGroupedMatch 50000 38148 ns/op
R=r, rsc
https://golang.org/cl/154173
also pick off the special case in strings.Index. don't want strings.IndexByte
because the call site will very rarely need to allocate and we can handle the
test in the code itself. bytes.IndexByte can avoid a common allocation.
R=rsc
CC=golang-dev
https://golang.org/cl/156091
Previously a netFd could be queued for reading/writing in the channel,
but close(2)'ed before pollServer got to it. In this case, the kernel
would consider the descriptor closed and the attempt to add it to the
epoll set would fail and panic.
This patch makes Close a roundtrip to the pollServer, although the
actual close(2) still occurs elsewhere to avoid blocking the
pollServer.
Fixes#143.
R=rsc
CC=golang-dev
https://golang.org/cl/152130
7x speedup on big and crypto/rsa unit tests.
also dropped useAsm in favor of making the
asm stubs jump to the Go versions.
R=agl1
CC=golang-dev, gri
https://golang.org/cl/157062
In thread.c, we need to cast to whatever the native
size of intptr is on the system, but we only have
uintptr available. They're the same size, but can't
do signed casts without this one :).
R=rsc
CC=golang-dev
https://golang.org/cl/156073
* move memmove to arch-specific subdirectories
* add memmove for arm
* add copyright notices marking them as copied from Inferno
R=ken2
https://golang.org/cl/156061
* add Marshal
* add BitString.RightAlign
* change to using a *time.Time (from time.Time) since that's what
the time package uses.
* return the unparsed data from Unmarshal.
R=rsc
CC=golang-dev
https://golang.org/cl/156047
the signal handling stack is a different size than
the normal stack, so it cannot be allocated using
the backup stack allocator.
Fixes#250.
R=agl1
CC=golang-dev
https://golang.org/cl/157044
Error information is carried from RPC server to client in the string
'Error' field of rpc.Response. An empty string is sent in the success
case. This empty string was being returned to the caller (of Client.Call
or Client.Go), resulting in a non-nil error response.
This change detects an empty-string Response.Error at the client, and
translates it into a nil value in Call.Error.
Tests updated to check error return in success cases.
R=r, rsc
https://golang.org/cl/154159
Currently, -I switch can't deal with a path containing spaces.
This commit simplify setinclude(), by removing the special case
of a string that had spaces. After this change, setinclude() will
merely add the given directories to the include path, if it does
not yet exist, and this approach works.
Will be needed for solving issue 115.
R=agl1, rsc, iant2, r
https://golang.org/cl/155059
cgo/libmach remain unimplemented. However, compilers, runtime,
and packages are 100%. I still need to go through and implement
missing syscalls (at least make sure they're all listed), but
for all shipped functionality, this is done. Ship! ;)
R=rsc, VenkateshSrinivas
https://golang.org/cl/152142
package documentation using templates. The line tag interferes
with the anchor tag introduces by the template.
This fixes an an issue where some headers lost their ability
to link to the respective source code.
R=rsc
https://golang.org/cl/154166
We are dealing with the multiplicative group ℤ/pqℤ. Multiples of
either p or q are not members of the group since they cannot have an
inverse. (Such numbers are 0 in the subgroup ℤ/pℤ.)
With p and q of typical size (> 512 bits), the probability of a random
blind [1..pq-1] being a multiple of p or q is negligible. However, in
the unit tests, much smaller sizes are used and the event could occur.
This change checks the result of the ext GCD and deals with this case.
It also increases the size of p and q in the unit test as a large
number of the keys selected were p, q = 227,169.
R=rsc
CC=golang-dev
https://golang.org/cl/154141
The ByteOrder.Put* methods are already available, this change uses
them to implement the Write function.
R=golang-dev, agl1, rsc, r
https://golang.org/cl/152141
performance hit of about 20% but more intuitive results for submatches.
we need a good regexp package at some point.
Fixes#110.
R=rsc
CC=golang-dev
https://golang.org/cl/152131
Commented both flush methods so people know what they are looking at.
This is a necessary fix for streaming and long polling HTTP services.
Fixes#93.
R=r, rsc, david.titarenco
https://golang.org/cl/154099
SELinux will cause mmap to fail when we request w+x memory unless the
user has configured their policies. We have a warning in make.bash,
but it's quite likely that the policy will be reset at some point and
then all their binaries start failing.
This patch prints a warning on Linux when mmap fails with EACCES.
R=rsc
CC=golang-dev
https://golang.org/cl/152086
(Making it work correctly with -spaces is a bit of work and the output
won't make much sense as it is intended as input to tabwriter.)
Fixes#100.
R=rsc
https://golang.org/cl/154102
The default SELinux policy on Fedora 12 (at least) disallows stack
pages to be +x. This causes all binaries written by 6g/6l to segfault
immedately. The 'true' way to fix this issue is to mark binaries with
the correct type. However, that assumes that 6l is going to detect
SELinux, figure out the correct type for the current distribution and
set the type (without libselinux).
For now we'll warn users and point them towards the way to enable
execstack for the whole system.
http://code.google.com/p/go/issues/detail?id=47
R=rsc
CC=golang-dev
http://codereview.prom.corp.google.com/1026041
1. allocate an extra byte for use by write.
2. throw away the code trying to translate
uid and gid to names. i suspect it of causing
seg faults in some situations, and it is
not needed by the compilers.
Fixes#48. (I hope.)
R=r
https://golang.org/cl/152077
Turn methods that don't store the result in their receiver into
functions in order to preserve the convention.
Re-jig Exp and Div by moving their guts into nat.go.
Add ProbablyPrime to perform Miller-Rabin primality tests.
crypto/rsa: reenable key generation since we now have ProbablyPrime.
R=gri
CC=go-dev
http://codereview.prom.corp.google.com/1024038
if suffixes don't work, check for name directly.
also fixes short names like bit.ly when ndots>1.
tested by tossing domain and search lines from /etc/resolv.conf
Fixes#2.
R=agl, agl1
CC=golang-dev
https://golang.org/cl/152048
too hard to make it the same everywhere.
still checking for non-zero exit status though.
disable core files while running tests
R=r
http://go/go-review/1026037
a node fits on one line:
- for purposes of measuring the node size in text,
don't generate html or use a styler that could
generate html as it will lead to overly large
sizes
A consequence of this bug is that source code displayed
with godoc may show functions that fit on one line in
the source on multiple lines.
This change causes no difference to the gofmt formatting
of any files in src or misc.
R=rsc
http://go/go-review/1026034
- When providing alternative spellings to a query, do not
prefix it with a package qualifier as the suggestion may
not have any results. Correctly filtering is quite a bit
of work, and clicking the alternative spelling will always
also show the qualified hits if they exist (but also others).
Seems good enough for now.
- Give user feedback when the query syntax was wrong.
- Package names in search results are now links to the respective
package documentation.
- Experimented with excluding main packages and test files
from index with inconclusive results. Code is present and
can be enabled by changing a flag in the source. This needs
some more work.
R=rsc
CC=r, dsymonds
http://go/go-review/1026033
HTML vs Html, URL vs Url, HTTP vs Http,
current source is 6:1 in favor of the former,
so change instances of the latter.
R=r
CC=go-dev
http://go/go-review/1024026
- introduced a new run per file containing all spots belonging
to the same kind (e.g. var decl, const decl, etc.)
- more comments, better index.go file organization
R=rsc
http://go/go-review/1026028
(this was surprisingly hard to get right in HTML)
- show modification times in source directory listings
- various tweaks
R=rsc
http://go/go-review/1024024
reflect is a little more detailed than some because it affords an opportunity
to explain how to approach the library.
R=gri, rsc
CC=go-dev
http://go/go-review/1026026
- formatting in dirs.html is crude, needs better html (open to suggestions),
but shows the synopsis
- many package comments should probably be adjusted such that the first
sentence is more concise
R=rsc, iant
http://go/go-review/1025014
- blank padding around composite literal contents for a less dense look
(most if not all composite literals were written in that style before gofmt
ran through them)
- corresponding (internal) flag: compositeLitBlank
- don't print a semi after a one-statement statement list (as preferred by r)
- corresponding (internal flag): fewerSemis
- the number of changes in nodes.go is huge because of the removed semis;
hg mail requires that I gofmt the file before
With both flags set to false, this CL causes no gofmt formatting change. We
can turn them on if we like it (and then remove the flags). Will submit with
flags disabled.
R=rsc, r
http://go/go-review/1025015
- enabled for function declarations (not just function literals)
- applied gofmt -w $GOROOT/src
(look for instance at src/pkg/debug/elf/elf.go)
R=r, rsc
CC=go-dev
http://go/go-review/1026006
both 64 and 32 bit platforms.
Also, uncomment some tests that were still commented out from
debugging.
R=gri, rsc
CC=go-dev
http://go/go-review/1026003
- this ensures better comment formatting in template.go and codec_test.go
- it affects only 2 files of all files that have been gofmt'ed already,
see separate CL (it fixes the same issue in those files)
R=rsc
http://go/go-review/1023002
Make RSA and X509 build by using big. (This involves commenting out
key generation for now since I haven't written Miller-Rabin in big
yet.)
Add entries to the Makefile.
R=rsc
CC=go-dev
http://go/go-review/1022005
(With hindsight, record_process might have been designed wrong, but it
works for now. It'll get redrawn when client support is added.)
R=rsc
CC=r
http://go/go-review/1018032
(the one-line struct types used in composite literals will become
one line again in another cleanup round; don't worry about them now)
R=r
http://go/go-review/1016056
- slightly better output in general
- extra indentation where we could do without
(however that seems better for now that not having the indentation
where it is needed)
- no information is lost, so a future better approach can fix these
places up again
R=rsc
http://go/go-review/1017050
- preserve (some) indentation of comment text for /*-style comments
even if the first comment line does not contain any text that might
suggest the "correct" indentation
- enabled because otherwise existing larger comments get re-formatted
(this will not introduce a lot of changes since comments of this
kind - until now - were not changed with respect to indentation)
R=rsc
http://go/go-review/1016047
picked up by gofmt (they should not be modified) or godoc
(testdata should not be considered a potential package
directory)
R=rsc
http://go/go-review/1018045
- at the moment this also affects single-line function declarations
because we have disabled them at the moment (but not single-line
function literals)
R=rsc
http://go/go-review/1016040
For testing it can be useful to use a null random source (one which
always returns zero) to remove non-determinism from the tests.
However, when performing RSA blinding, the random blind ends up being
zero and it's hard to reverse a multiplication by zero.
R=rsc
CC=go-dev
http://go/go-review/1018033
(which I uploaded at home and thus can't upload from here).
CL 1018027 was reviewed.
- added comments to scope.go
- commented out some code that is not yet needed
(and which showed up prominently in the documentation)
R=rsc
http://go/go-review/1017017
- start new sections if a field/method declaration spans multiple lines;
this avoids tabs from the previous line affecting the next field/method
R=rsc
http://go/go-review/1017015
(No visual change yet since the comments for
exported code snippets - as shown on the pkg
page - are not yet formatted using the standard
comment intersperse mechanism in the printer.
This is a TODO in the printer.)
R=rsc
http://go/go-review/1016031
- nodes.go implements ast node formatting
- printer.go implements the core printing functionality
and public interface
No code changes except for updating the import clauses
and adding a top-level comment to nodes.go.
R=rsc
http://go/go-review/1016026
added ReadFrom/WriteTo for packet protocols like UDP.
simplified the net.Conn interface.
added new net.PacketConn interface for packet protocols.
implemented proper UDP listener.
cleaned up LocalAddr/RemoteAddr methods - cache in netFD.
threw away various unused methods.
an interface change:
introduced net.Addr as a network address interface,
to avoid conversion of UDP host:port to string and
back for every ReadFrom/WriteTo sequence.
another interface change:
since signature of Listener.Accept was changing anyway,
dropped the middle return value, because it is available
as c.RemoteAddr(). (the Accept signature predates the
existence of that method.)
Dial and Listen still accept strings, but the proto-specific
versions DialTCP, ListenUDP, etc. take net.Addr instead.
because the generic Dial didn't change and because
no one calls Accept directly (only indirectly via the http
server), very little code will be affected by these interface
changes.
design comments welcome.
R=p
CC=go-dev, r
http://go/go-review/1018017
Firstly, with -Werror, GCC switched to printing warnings starting
with "error:". Widening the string matches solves this as the messages
are otherwise unchanged.
Secondly, GCC 4.4 outputs DWARF sections with with NUL bytes in all
the offsets and requires the relocation section for .debug_info to be
processed in order to result in valid DWARF data. Thus we add minimal
handling for relocation sections, which is sufficient for our needs.
BUG=1
Fixes#1.
R=rsc, iant
CC=go-dev
http://go/go-review/1017003
just a single directory
- all pieces present but not well integrated
- directory tree served at the moment under /tree
R=rsc
http://go/go-review/1018016
(previously not shown in godoc)
- simplify parsing of struct types (match code structure for parsing interface types)
R=rsc, r
http://go/go-review/1016019
* rename PORT.sh -> mkall.sh (hopefully more obvious),
change behavior: run commands by default.
* pull more constants out of #defines automatically,
instead of editing large lists by hand.
* add Recvfrom, Sendto
add os.O_EXCL.
R=r
http://go/go-review/1017009
allow Makefiles using Make.pkg to specify a
list of directories that should be installed
before trying to build the package.
this is a stopgap for small package trees
maintained outside the standard tree.
R=r
http://go/go-review/1016012
- print comments line by line, strip common prefix but do not
modify comment contents otherwise
- align comments with subsequent keyword if indicated (e.g. case labels)
- terminate "column section" after multi-line expressions for better alignment
R=rsc
http://go/go-review/1017002