Fixes #18438

Change-Id: I9599c1536d5e8bad7662b8ffa19e9b0746e27e60
Reviewed-on: https://go-review.googlesource.com/44000Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

- spaces are allowed anywhere but the last character of a boundary

Fixes #18768

Change-Id: I36b054462533ff6dfc060e37e7a58777ae4b66fe
Reviewed-on: https://go-review.googlesource.com/35507
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Commit 4dcba023 replaced select with pselect6 on linux/amd64 and
linux/arm, but it turns out the Android emulator uses linux/386. This
makes the equivalent change there, too.

Fixes #20409 more.

Change-Id: If542d6ade06309aab8758d5f5f6edec201ca7670
Reviewed-on: https://go-review.googlesource.com/44011
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

Current code cannot handle string #define macros if those macros are
defined via other macros. This CL solve the issue.

Updates #18720

Change-Id: Ibed0773d10db3d545bb246b97e81c0d19e3af3d5
Reviewed-on: https://go-review.googlesource.com/41312Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Now:
$ GOARCH=arm GOARM=5 go install -x cmd/go
... followed by:
$ GOARCH=arm GOARM= go install -x cmd/go

... actually does work. Previously the second "go install" would reuse
the cached binaries from the GOARM=5 command and not rebuild.
(Or vice versa from GOARM= to GOARM=5)

And do the same for GO386.

Fixes #9737

Change-Id: I9630aab34d06465d5033e6743dfe6592c8247aa0
Reviewed-on: https://go-review.googlesource.com/43855
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

We only need to read the number of bytes required to store the value
"max - 1" to generate a random number in the range [0, max).

Before, there was an off-by-one error where an extra byte was read from
the io.Reader for inputs like "256" (right at the boundary for a byte).
There was a similar off-by-one error in the logic for clearing bits and
thus for any input that was a power of 2, there was a 50% chance the
read would continue to be retried as the mask failed to remove a bit.

Fixes #18165.

Change-Id: I548c1368990e23e365591e77980e9086fafb6518
Reviewed-on: https://go-review.googlesource.com/43891Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

There are two copies each of the stackPreempt/_StackPreempt and
stackFork/_StackFork constants. Remove the ones left over from C that
are no longer used.

Change-Id: I849604c72c11e4a0cb08e45e9817eb3f5a6ce8ba
Reviewed-on: https://go-review.googlesource.com/43638Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Setting stackCache to 0 to disable stack caches for debugging hasn't
worked for a long time. It causes stackalloc to fall back to full span
allocation, round sub-page stacks down to 0 pages, and blow up.

Fix this debug mode so it disables the per-P caches, but continues to
use the global stack pools for small stacks, which correctly handle
sub-page stacks. While we're here, rename stackCache to stackNoCache
so it acts like the rest of the stack allocator debug modes where "0"
is the right default value.

Fixes #17291.

Change-Id: If401c41cee3448513cbd7bb2e9334a8efab257a7
Reviewed-on: https://go-review.googlesource.com/43637Reviewed-by: Keith Randall <khr@golang.org>

The stackFromSystem debug mode has two problems:

1) It rounds the stack allocation to _PageSize. If the physical page
size is >8K, this can cause unmapping the memory later to either
under-unmap or over-unmap.

2) It doesn't return the rounded-up allocation size to its caller, so
when we later unmap the memory, we may pass the wrong length.

Fix these problems by rounding the size up to the physical page size
and putting that rounded-up size in the returned stack bounds.

Fixes #17289.

Change-Id: I6b854af3b06bb16e3750798397bb5e2a722ec1cb
Reviewed-on: https://go-review.googlesource.com/43636Reviewed-by: Keith Randall <khr@golang.org>

Change-Id: I2ed2e8c4890a65288cf3066ebe3c1d9a16fb4c05
Reviewed-on: https://go-review.googlesource.com/43990Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

If mheap.sysAlloc doesn't have room in the heap arena for an
allocation, it will attempt to map more address space with sysReserve.
sysReserve is given a hint, but can return any unused address range.
Currently, mheap.sysAlloc incorrectly assumes the returned region will
never fall between arena_start and arena_used. If it does,
mheap.sysAlloc will blindly accept the new region as the new
arena_used and arena_end, causing these to decrease and make it so any
Go heap above the new arena_used is no longer considered part of the
Go heap. This assumption *used to be* safe because we had all memory
between arena_start and arena_used mapped, but when we switched to an
arena_start of 0 on 32-bit, it became no longer safe.

Most likely, we've only recently seen this bug occur because we
usually start arena_used just above the binary, which is low in the
address space. Hence, the kernel is very unlikely to give us a region
before arena_used.

Since mheap.sysAlloc is a linear allocator, there's not much we can do
to handle this well. Hence, we fix this problem by simply rejecting
the new region if it isn't after arena_end. In this case, we'll take
the fall-back path and mmap a small region at any address just for the
requested memory.

Fixes #20259.

Change-Id: Ib72e8cd621545002d595c7cade1e817cfe3e5b1e
Reviewed-on: https://go-review.googlesource.com/43870Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Rick Hudson <rlh@golang.org>

By changing getRandomLinux to immediately use the getrandom() syscall
without GRND_NONBLOCK, we now only fall back to reading from
/dev/urandom on Linux if the kernel does not support the getrandom()
syscall. This means reads for crypto/rand will now block if the kernel
has insufficient entropy on Linux kernels after v3.16.

Before, if the kernel had insufficient entropy, it would fall back to
reading from /dev/urandom. This would potentially return predictable
data.

Fixes #19274

Change-Id: I1cb081ce2f3096f18ad2820e52ecdbd993dc2afc
Reviewed-on: https://go-review.googlesource.com/43852Reviewed-by: Filippo Valsorda <hi@filippo.io>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

This change is windows version of CL 12152.
It also extends test to cover scenarios reported on issue #20445.
Some source files copied and renamed to make code clearer.

Fixes #20445

Change-Id: Idd2f636f27c6bd5cfe98017ba2df911358263382
Reviewed-on: https://go-review.googlesource.com/43910
Run-TryBot: Alex Brainman <alex.brainman@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Fixes #19478

Change-Id: I9fc186610d79fd003e7b5d88c0955286ebe7d3cf
Reviewed-on: https://go-review.googlesource.com/38634Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Small performance gain:
name              old time/op    new time/op    delta
Itoa-4              95.4ns ± 4%    95.6ns ± 3%    ~     (p=0.256 n=45+46)
Println-4            480ns ± 4%     476ns ± 5%  -0.87%  (p=0.003 n=45+45)
PrintlnNoFlags-4     316ns ± 3%     299ns ± 4%  -5.38%  (p=0.000 n=42+44)

name              old alloc/op   new alloc/op   delta
Itoa-4               0.00B          0.00B         ~     (all equal)
Println-4            21.0B ± 0%     21.0B ± 0%    ~     (all equal)
PrintlnNoFlags-4     21.0B ± 0%     21.0B ± 0%    ~     (all equal)

name              old allocs/op  new allocs/op  delta
Itoa-4                0.00           0.00         ~     (all equal)
Println-4             2.00 ± 0%      2.00 ± 0%    ~     (all equal)
PrintlnNoFlags-4      2.00 ± 0%      2.00 ± 0%    ~     (all equal)

Change-Id: Idcd03609a5a437a69ffa7004a673bf0b8d22e7ad
Reviewed-on: https://go-review.googlesource.com/38056Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>

Fixes #20435.

Change-Id: I15576f36b26d01642c1187325baea82d3077e578
Reviewed-on: https://go-review.googlesource.com/43777Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Due to the fact that -cover injects additional code to the original
source, tests run with -cover will often have incorrect line numbers.

Also includes docs for -list regexp missed by ba8ff87d

Updates #6329

Change-Id: I87f0618ac31e96071bca61055cc17c0cbdee208a
Reviewed-on: https://go-review.googlesource.com/38640Reviewed-by: Rob Pike <r@golang.org>

The jarKey function handles broken PublicSuffixList implementations but
no test verified it.

Change-Id: Ifb76de9e8c3941f3b08d3e43970056e023013457
Reviewed-on: https://go-review.googlesource.com/38357Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

When restoring syms, we must also restore the original Lastlineno.
Bug introduced with https://golang.org/cl/41390/.

Fixes #20415.

Change-Id: Ie81d36279d717e330951b52f42dcee4b0025b9f0
Reviewed-on: https://go-review.googlesource.com/43811Reviewed-by: Matthew Dempsky <mdempsky@google.com>

According to RFC 6255 a cookie value may contain neither spaces " "
nor commas ",". But browsers seem to handle these pretty well and such
values are not uncommon in the wild so we do allow spaces and commas
in cookie values too. Up to now we use the double-quoted wire format
only for cookie values with leading and/or trailing spaces and commas.
Values with internal spaces/commas are sent without the optional double
quotes. This seems to be a problem for some agents.

This CL changes the behaviour for cookie values with spaces or commas:
Such values are always sent in double quotes. This should not have
any impact on existing agents and the increases of data transmitted
is negligible.

Fixes #18627

Change-Id: I575a98d589e048aa39d976a3c984550daaca730a
Reviewed-on: https://go-review.googlesource.com/37328
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Trailers that are not announced in the Trailer must be passed on to
the downstream client.

Rather than iterate over each and find missing trailer values,
this re-adds all trailers to the headers if there is a disparity
between the number of announced trailers and the final number.

This fixes #20437

Change-Id: I867e85f45feff68616a9a9bd6f65f12d73825eb7
Reviewed-on: https://go-review.googlesource.com/43712Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Change-Id: I020cd3d10a441ba4047800fdf4f93433c458398a
Reviewed-on: https://go-review.googlesource.com/43717Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Query and Exec functions on DB first attempt to get a cached
connection before requesting the connection pool to ignore
the cache and get a new connection. This change aligns Stmt to
that behavior as well.

Fixes #20433

Change-Id: Idda5f61927289d7ad0882effa3a50ffc9efd88e6
Reviewed-on: https://go-review.googlesource.com/43790
Run-TryBot: Daniel Theophanes <kardianos@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Allow the memory limit passed into ReadForm to be used as the
memory limit for processing non-file form data as well as file
form data, rather than the existing behaviour of the memory limit
only applying to the file parts and the non-file parts being
arbitrarily limited to 10MB.

This ensures backwards compatibility while still providing the
user with control over the amount of non-file data that can be
processed instead of enforcing an arbitrary 10MB limit.

Change-Id: I53c09eae00147d3ff2d6bdfd4e50949267932c3d
Reviewed-on: https://go-review.googlesource.com/38195Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Ensure that the implicitly created redirect
for
  "/route"
after
  "/route/"
has been registered doesn't lose the query string information.

Fixes #17841.

Change-Id: Ib7df9242fab8c9368a18fc0da678003d6bec63b8
Reviewed-on: https://go-review.googlesource.com/43779
Run-TryBot: Emmanuel Odeke <emm.odeke@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>

Many instructions can not have a .S suffix, such as MULS, SWI, CLZ,
CMP, STREX and others. And so do .P and .W suffixes. Even wrong
assembly code is generated for some instructions with invalid
suffixes.

This patch tries to simplify .S/.W/.P checks. And a wrong assembly
test for arm is added.

fixes #20377

Change-Id: Iba1c99d9e6b7b16a749b4d93ca2102e17c5822fe
Reviewed-on: https://go-review.googlesource.com/43561Reviewed-by: Cherry Zhang <cherryyz@google.com>

Further examples to support the net package.

Updates #5757

Change-Id: I9b65521d211f6c404b9103c1eaf22b0772eb242e
Reviewed-on: https://go-review.googlesource.com/43711Reviewed-by: Mikio Hara <mikioh.mikioh@gmail.com>

This is a followup to https://go-review.googlesource.com/37039.

Updates #19435.

Change-Id: Ia795bd5158d26effa56e897698208ccf73f9e0d2
Reviewed-on: https://go-review.googlesource.com/43693Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

This is a followup to https://go-review.googlesource.com/37038.

Updates #19435.

Change-Id: If3d56bca0e8816d1a169f5cf97f27b20695a9955
Reviewed-on: https://go-review.googlesource.com/43692Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

It was removed in CL 27325.

Fixes #20431

Change-Id: I6842851444186e19029d040f61fdf4f87a3103a6
Reviewed-on: https://go-review.googlesource.com/43771Reviewed-by: Ian Lance Taylor <iant@golang.org>

Change-Id: I72a10cd5dfb863f8219bb3b5b8280c017f523cf4
Reviewed-on: https://go-review.googlesource.com/42856Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>

Instead of just printing the value, print the original node to make the
error more human-friendly. Also print the value if its string form is
different than the original node, to make sure it's obvious what value
was duplicated.

This means that "case '@', '@':", which used to print:

	duplicate case 64 in switch

Will now print:

	duplicate case '@' (value 64) in switch

Factor this logic out into its own function to reuse it in range cases
and any other place where we might want to print a node and its value in
the future.

Also needed to split the errorcheck files because expression switch case
duplicates are now detected earlier, so they stop the compiler before it
gets to generating the AST and detecting the type switch case
duplicates.

Fixes #20112.

Change-Id: I9009b50dec0d0e705e5de9c9ccb08f1dce8a5a99
Reviewed-on: https://go-review.googlesource.com/41852
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>

Android O black-lists the select system call because its libc, Bionic,
does not use this system call. Replace our use of select with pselect6
(which is allowed) on the platforms that support targeting Android.
linux/arm64 already uses pselect6 because there is no select on arm64,
so only linux/amd64 and linux/arm need changing. pselect6 has been
available since Linux 2.6.16, which is before Go's minimum
requirement.

Fixes #20409.

Change-Id: Ic526b5b259a9e01d2f145a1f4d2e76e8c49ce809
Reviewed-on: https://go-review.googlesource.com/43641
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

This change makes {TCP,UDP,IP,Unix}Conn types compliant of
syscall.Conn interface and adds type rawConn as an implementation of
syscall.RawConn interface.

By this change, the long-standing issues regarding unsupported socket
options and system calls can be solved partly and the broken x/net
packages due to https://go-review.googlesource.com/36799 can be
repaired.

Fixes #3661.
Updates #9661.
Updates #19051.
Updates #19435.

Change-Id: Ic996b040418b54f6d043bc70591789d5a5b23270
Reviewed-on: https://go-review.googlesource.com/37039
Run-TryBot: Mikio Hara <mikioh.mikioh@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

This change adds RawControl, RawRead and RawWrite methods to type FD
to make the runtime-integrated network poller work together with a
user-defined function. The methods are used via the net package from
external packages and type FD is considered as an implementation of
syscall.Conn and syscall.RawConn interfaces.

Updates #19435.

Change-Id: I4ad04b10ffddb2b54fa8d70587440960d73c0a2d
Reviewed-on: https://go-review.googlesource.com/37038
Run-TryBot: Mikio Hara <mikioh.mikioh@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>

This change adds Conn and RawConn interfaces which can be used to
manipulate raw network connection end points typically represented as
socket descriptors.

Fixes #19435.

Change-Id: Ide2d28eeab91bfd27473ab47a87bec69950b64c9
Reviewed-on: https://go-review.googlesource.com/37913
Run-TryBot: Mikio Hara <mikioh.mikioh@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>

Change-Id: Ic6d2de92e1f533a9f9a0cd6d7dab463bdafb0e11
Reviewed-on: https://go-review.googlesource.com/43691Reviewed-by: Mikio Hara <mikioh.mikioh@gmail.com>
Run-TryBot: Alex Brainman <alex.brainman@gmail.com>

Change-Id: If5cb80c3c086684ce6c2e8ed9bb23b2a20c8aacd
Reviewed-on: https://go-review.googlesource.com/43690Reviewed-by: Mikio Hara <mikioh.mikioh@gmail.com>
Run-TryBot: Alex Brainman <alex.brainman@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Recent CL 41834 made windows Stat work for all symlinks.
But CL 41834 also made Stat slow.

John Starks sugested
(see https://github.com/golang/go/issues/19922#issuecomment-300031421)
to use GetFileAttributesEx for files and directories instead.
This makes Stat as fast as at go1.9.

I see these improvements on my Windows 7

name       old time/op  new time/op  delta
StatDot    26.5µs ± 1%  20.6µs ± 2%  -22.37%  (p=0.000 n=9+10)
StatFile   22.8µs ± 2%   6.2µs ± 1%  -72.69%  (p=0.000 n=10+10)
StatDir    21.0µs ± 2%   6.1µs ± 3%  -71.12%  (p=0.000 n=10+9)
LstatDot   20.1µs ± 1%  20.7µs ± 6%   +3.37%  (p=0.000 n=9+10)
LstatFile  6.23µs ± 1%  6.36µs ± 8%     ~     (p=0.587 n=9+10)
LstatDir   6.10µs ± 0%  6.14µs ± 4%     ~     (p=0.590 n=9+10)

and on my Windows XP

name         old time/op  new time/op  delta
StatDot-2    20.6µs ± 0%  10.8µs ± 0%  -47.44%  (p=0.000 n=10+10)
StatFile-2   20.2µs ± 0%   7.9µs ± 0%  -60.91%  (p=0.000 n=8+10)
StatDir-2    19.3µs ± 0%   7.6µs ± 0%  -60.51%  (p=0.000 n=10+9)
LstatDot-2   10.8µs ± 0%  10.8µs ± 0%   -0.48%  (p=0.000 n=10+8)
LstatFile-2  7.83µs ± 0%  7.83µs ± 0%     ~     (p=0.844 n=10+8)
LstatDir-2   7.59µs ± 0%  7.56µs ± 0%   -0.46%  (p=0.000 n=10+10)

Updates #19922

Change-Id: Ice1fb5825defb05c79bab4dec0692e0fd1bcfcd5
Reviewed-on: https://go-review.googlesource.com/43071Reviewed-by: Austin Clements <austin@google.com>
Run-TryBot: Alex Brainman <alex.brainman@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>

Change-Id: I4b8f1b61c10f60ddb3687759af0be1641c1f78ce
Reviewed-on: https://go-review.googlesource.com/43111Reviewed-by: Alex Brainman <alex.brainman@gmail.com>