Ian Lance Taylor [Thu, 9 Jan 2014 05:40:33 +0000 (21:40 -0800)]
runtime: change errorCString to a struct
This prevents callers from using reflect to create a new
instance of errorCString with an arbitrary value and calling
the Error method to examine arbitrary memory.
Rob Pike [Wed, 8 Jan 2014 23:56:40 +0000 (15:56 -0800)]
cmd/nm: add -sort=size
When printing the size, we often want to sort on that key.
Because it's used when looking for large things, make the
sort go from largest to smallest.
Perfect recreation of CL 45150044, which was lost to some blunder.
David du Colombier [Wed, 8 Jan 2014 22:34:11 +0000 (23:34 +0100)]
libmach: use Go's ureg headers on Plan 9
The CL 49090043 renamed Ureg structures to Ureg386,
UregArm and UregAmd64. This broke build on Plan 9,
since ureg_x86.h includes /386/include/ureg.h, which
declares a structure named Ureg instead of Ureg386.
Michael Hudson-Doyle [Wed, 8 Jan 2014 04:53:16 +0000 (23:53 -0500)]
cmd/go: test: do not put object files where later steps will find them
When recompiling a package whose basename is the name of a standard
package for testing with gccgo, a .o file with the basename of the
package being tested was being placed in the _test/ directory where the
compilation of the test binary then found it when looking for the
standard library package.
This change puts the object files in a separate directory.
Keith Randall [Tue, 7 Jan 2014 21:45:50 +0000 (13:45 -0800)]
runtime: use special records hung off the MSpan to
record finalizers and heap profile info. Enables
removing the special bit from the heap bitmap. Also
provides a generic mechanism for annotating occasional
heap objects.
finalizers
overhead per obj
old 680 B 80 B avg
new 16 B/span 48 B
profile
overhead per obj
old 32KB 24 B + hash tables
new 16 B/span 24 B
Brad Fitzpatrick [Tue, 7 Jan 2014 18:40:56 +0000 (10:40 -0800)]
net/http: fix data race when sharing request body between client and server
A server Handler (e.g. a proxy) can receive a Request, and
then turn around and give a copy of that Request.Body out to
the Transport. So then two goroutines own that Request.Body
(the server and the http client), and both think they can
close it on failure. Therefore, all incoming server requests
bodies (always *http.body from transfer.go) need to be
thread-safe.
Fixes #6995
R=golang-codereviews, r
CC=golang-codereviews
https://golang.org/cl/46570043
Russ Cox [Tue, 7 Jan 2014 01:43:44 +0000 (20:43 -0500)]
cmd/gc: use 100x less memory for []byte("string")
[]byte("string") was simplifying to
[]byte{0: 0x73, 1: 0x74, 2: 0x72, 3: 0x69, 4: 0x6e, 5: 0x67},
but that latter form takes up much more memory in the compiler.
Preserve the string form and recognize it to turn global variables
initialized this way into linker-initialized data.
Reduces the compiler memory footprint for a large []byte initialized
this way from approximately 10 kB/B to under 100 B/B.
Adam Langley [Mon, 6 Jan 2014 21:11:58 +0000 (16:11 -0500)]
crypto/rsa: support unpadded signatures.
Usually when a message is signed it's first hashed because RSA has low
limits on the size of messages that it can sign. However, some
protocols sign short messages directly. This isn't a great idea because
the messages that can be signed suddenly depend on the size of the RSA
key, but several people on golang-nuts have requested support for
this and it's very easy to do.
Brad Fitzpatrick [Mon, 6 Jan 2014 18:43:56 +0000 (10:43 -0800)]
archive/zip: fix bug reading zip64 files
ZIP64 Extra records are variably sized, but we weren't capping
our reading of the extra fields at its previously-declared
size.
No test because I don't know how to easily create such files
and don't feel like manually construction one. But all
existing tests pass, and this is "obviously correct" (queue
laughter).
Matthew Cottingham [Mon, 6 Jan 2014 18:36:04 +0000 (10:36 -0800)]
net/http: Add more call order tests for request form parsing.
Adds tests for branches handling call ordering which
were shown to be untested by the cover tool.
This is part of the refactoring of form parsing discussed
in CL 44040043. These tests may need to be changed later but
should help lock in the current behaviour.
Volker Dobler [Mon, 6 Jan 2014 18:00:58 +0000 (10:00 -0800)]
net/http: remove todos from cookie code
The Domain and Path field of a parsed cookie have been
the unprocessed wire data since Go 1.0; this seems to
be okay for most applications so let's keep it.
Returning the unprocessed wire data makes it easy to
handle nonstandard or even broken clients without
consulting Raw or Unparsed of a cookie.
The RFC 6265 parsing rules for domain and path are
currently buried in net/http/cookiejar but could be
exposed in net/http if necessary.
Joel Sing [Sun, 5 Jan 2014 14:34:56 +0000 (01:34 +1100)]
crypto/sha1, crypto/sha256, crypto/sha512: use copy for partial block
Use copy rather than a hand rolled loop when moving a partial input
block to the scratch area. This results in a reasonable performance
gain when partial blocks are written.
Benchmarks on Intel(R) Xeon(R) CPU X5650 @ 2.67GHz with Go amd64:
Rob Pike [Sat, 4 Jan 2014 18:52:59 +0000 (10:52 -0800)]
spec: s/and/or/ for correctness and parallelism
No change to the meaning, just bad writing found by
Doug McIlroy.
Let's start the new year off with a bang.
R=golang-codereviews, bradfitz, dave
CC=golang-codereviews
https://golang.org/cl/47110044
Joel Sing [Fri, 3 Jan 2014 13:29:20 +0000 (00:29 +1100)]
syscall: handle varied path lengths for unix sockets
Most BSDs include the trailing NUL character of the socket path in the
length, however some do not (such as NetBSD 6.99). Handle this by only
subtracting the family and length bytes from the returned length, then
scanning the path and removing any terminating NUL bytes.
David Symonds [Thu, 2 Jan 2014 23:10:54 +0000 (10:10 +1100)]
runtime: add a test for randomised map iteration order.
Technically the spec does not guarantee that the iteration order is random,
but it is a property that we have consciously pursued, and so it seems
right to verify that our implementation does indeed randomise.
Keith Randall [Mon, 30 Dec 2013 20:14:27 +0000 (12:14 -0800)]
cmd/6g: inline copies of up to 4 words in length.
This change fixes a serious performance regression
with reflect.Value growing to 4 words instead of 3.
The json benchmark was ~50% slower, with this change
it is ~5% slower (and the binary is 0.5% larger).
Longer term, we probably need to rethink our copy
generation. Using REP is really expensive time-wise.
But inlining the copy grows the binary.
R=golang-codereviews, r
CC=golang-codereviews
https://golang.org/cl/44990043
Ian Lance Taylor [Sat, 28 Dec 2013 17:37:54 +0000 (09:37 -0800)]
net: work around Solaris connect issue when server closes socket
On Solaris, if you do a in-progress connect, and then the
server accepts and closes the socket, the client's later
attempt to complete the connect will fail with EINVAL. Handle
this case by assuming that the connect succeeded. This code
is weird enough that it is implemented as Solaris-only so that
it doesn't hide a real error on a different OS.
Update #6828
R=golang-codereviews, bradfitz, dave
CC=golang-codereviews
https://golang.org/cl/46160043
Ian Lance Taylor [Fri, 27 Dec 2013 16:49:47 +0000 (08:49 -0800)]
net: use DialTimeout in TestSelfConnect
This avoids problems with systems that take a long time to
find out nothing is listening, while still testing for the
self-connect misfeature since a self-connect should be fast.
With this we may be able to remove the test for non-Linux
systems.
Tested (on GNU/Linux) by editing selfConnect in
tcpsock_posix.go to always return false and verifying that
TestSelfConnect then fails with and without this change.
John Newlin [Thu, 26 Dec 2013 19:52:14 +0000 (11:52 -0800)]
net/http: Release reference to chunkWriter's bufio.Writer on hijack
When a connection is hijacked, release the reference to the bufio.Writer
that is used with the chunkWriter. The chunkWriter is not used after
the connection is hijacked.
Also add a test to check that double Hijack calls do something sensible.
benchmark old ns/op new ns/op delta
BenchmarkServerHijack 24137 20629 -14.53%
benchmark old allocs new allocs delta
BenchmarkServerHijack 21 19 -9.52%
benchmark old bytes new bytes delta
BenchmarkServerHijack 11774 9667 -17.90%
Russ Cox [Sat, 21 Dec 2013 00:14:42 +0000 (19:14 -0500)]
cmd/gc: do not compute dead value maps if they will not be used
Reduces 6g big.go memory usage from 251 MB to 242 MB.
Reduces 6g slow.go memory usage from 529 MB to 453 MB.
Mostly a memory savings; 6g slow.go is only about 5% faster.
The test programs are at
https://rsc.googlecode.com/hg/testdata/big.go (36k lines, 276kB)
https://rsc.googlecode.com/hg/testdata/slow.go (7k lines, 352kB)
Russ Cox [Fri, 20 Dec 2013 19:24:48 +0000 (14:24 -0500)]
cmd/gc: address 1½ liveness bottlenecks
As much as 7x speedup on some programs, cuts all.bash time by 20%.
Change splicebefore function from O(n) to O(1).
The approach was suggested by Carl during the code's review
but apparently did not make it into the tree.
It makes a huge difference on huge programs.
Make twobitwalktype1 slightly faster by using & instead of %.
Really it needs to be cached; left a note to that effect.
(Not a complete fix, hence the ½.)
big.go (output of test/chan/select5.go)
47.53u 0.50s 48.14r before this CL
7.09u 0.47s 7.59r with splicebefore change (6.7x speedup)
6.15u 0.42s 6.59r with twobitwalktype1 change (1.15x speedup; total 7.7x)
slow.go (variant of program in go.text, by mpvl)
77.75u 2.11s 80.03r before this CL
24.40u 1.97s 26.44r with splicebefore change (3.2x speedup)
18.12u 2.19s 20.38r with twobitwalktype1 change (1.35x speedup; total 4.3x)
test/run
150.63u 49.57s 81.08r before this CL
88.01u 45.60s 46.65r after this CL (1.7x speedup)
all.bash
369.70u 115.64s 256.21r before this CL
298.52u 110.35s 214.67r after this CL (1.24x speedup)
The test programs are at
https://rsc.googlecode.com/hg/testdata/big.go (36k lines, 276kB)
https://rsc.googlecode.com/hg/testdata/slow.go (7k lines, 352kB)
R=golang-codereviews, gobot, r
CC=cshapiro, golang-codereviews
https://golang.org/cl/43210045
Russ Cox [Fri, 20 Dec 2013 19:24:39 +0000 (14:24 -0500)]
cmd/gc: bypass DATA instruction for data initialized to integer constant
Eventually we will want to bypass DATA for everything,
but the relocations are not standardized well enough across
architectures to make that possible.
This did not help as much as I expected, but it is definitely better.
It shaves maybe 1-2% off all.bash depending on how much you
trust the timings of a single run: