This change implements Payne-Hanek range reduction by Pi/4
to properly calculate trigonometric functions of huge arguments.
The implementation is based on:
"ARGUMENT REDUCTION FOR HUGE ARGUMENTS: Good to the Last Bit"
K. C. Ng et al, March 24, 1992
The major difference with the reference is that the simulated
multi-precision calculation of x*B is implemented using 64-bit
integer arithmetic rather than floating point to ease extraction
of the relevant bits of 4/Pi.
The assembly implementations for 386 were removed since the trigonometric
instructions only use a 66-bit representation of Pi internally for
reduction. It is not possible to use these instructions and maintain
accuracy without a prior accurate reduction in software as recommended
by Intel.
Fixes#6794
Change-Id: I31bf1369e0578891d738c5473447fe9b10560196
Reviewed-on: https://go-review.googlesource.com/c/153059
Reviewed-by: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Updates #26148
Change-Id: Ide1fe821cc061a08488df9d40878131f37f894c9
Reviewed-on: https://go-review.googlesource.com/c/153844
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Update golang.org/x/sys/unix to revision b05ddf57801d2239d6ab0ee35f9d981e0420f4ac.
Changes exist in upstream golang.org/x/sys/unix, which allow for code to work and
tests to pass on openbsd/arm.
Change-Id: Iecc8598681a23cb0466f94c914f0e605a6fc64d7
Reviewed-on: https://go-review.googlesource.com/c/153838
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
The OpenBSD armv7 port requires 64-bit alignment for cmsgs.
Rework the cmsg alignment code to facilitate this.
Change-Id: I52cf55a8a4cda46c6ef35b0f694862b842028b42
Reviewed-on: https://go-review.googlesource.com/c/153837
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Change-Id: I2ff29aa9909be3062fcd5f65af261f5d8c46fbc1
Reviewed-on: https://go-review.googlesource.com/c/153843
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Context.BuildTags is not set when you read go/build.Default.BuildTags.
It's only used by (*BuildTags).Import, etc.
Fixes: #27320
Change-Id: I97e5f1923c410b48f70be8c15938a7e04a178e3f
Reviewed-on: https://go-review.googlesource.com/c/131975
Reviewed-by: Ian Lance Taylor <iant@golang.org>
TestCPUProfileLabel was failing on windows/arm because the link register
was not being passed to sigprof(). The link register is required to
generate a correct traceback. With this change, all tests in runtime.pprof
are now passing.
Updates #26148
Change-Id: Ia693b34278dc08a98023751ff1a922d9eee8fdd6
Reviewed-on: https://go-review.googlesource.com/c/153839
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Allow GOCACHE=off only for operations that never actually write
anything to the cache (in which case the GOCACHE setting should not
matter at all).
Fixes#29127
Change-Id: I733d02cd2fbcf3671f5adcfb73522865d131e360
Reviewed-on: https://go-review.googlesource.com/c/153462
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
The changes added in https://golang.org/cl/151997 to fix problems when
reading older export data introduced the ability to add "fixups" to
handle references to a type whose definition has not yet been
finalized. It turns out we need to allow for fixups even for more
recent export data (V2 and V3); this patch removes a version guard for
the fixup generation logic.
Fixes#29198.
Change-Id: I82136ac45b53e4a59c05ff0879ac6bb545d0ff31
Reviewed-on: https://go-review.googlesource.com/c/153821
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
It's similar to what cmd/go/internal/web package does
when cfg.BuildV is set. The web2 package is what
cmd/go/internal/modfetch uses, so this change allows us
to trace web requests go get command in module mode
sends for modfetch.
Change-Id: If387efd8a8698c816bf267d1e6c6766fd357c298
Reviewed-on: https://go-review.googlesource.com/c/153640
Run-TryBot: Hyang-Ah Hana Kim <hyangah@gmail.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
golang.org/cl/147117 increased the number of arguments permitted
by Proc.Call on Windows, but the doc comment was never updated.
Change-Id: Iea5eb9e0aafbc1025d5fcb8665d028b2254c183a
Reviewed-on: https://go-review.googlesource.com/c/153825
Reviewed-by: Channing Kimble-Brown <ckimblebrown@google.com>
Reviewed-by: Andrew Bonventre <andybons@golang.org>
Tested manually.
Before:
$ go mod init golang.org/issue/scratch
go: creating new go.mod: module golang.org/issue/scratch
$ go1.11.2 mod download github.com/rogpeppe/test2@latest
go: finding github.com/rogpeppe/test2 v0.0.11
$ find $GOPATH -name goodbye
/tmp/tmp.Y8a8UzX3zD/_gopath/pkg/mod/github.com/rogpeppe/test2@v0.0.11/tests/goodbye
$ cat $(find $GOPATH -name goodbye)
hello
After:
$ go mod init golang.org/issue/scratch
go: creating new go.mod: module golang.org/issue/scratch
$ go mod download github.com/rogpeppe/test2@latest
go: finding github.com/rogpeppe/test2 v0.0.11
$ find $GOPATH -name goodbye
$ find $GOPATH -name hello
/tmp/tmp.Zo0jhfLaRs/_gopath/pkg/mod/github.com/rogpeppe/test2@v0.0.11/tests/hello
A proper regression test would require one of:
• a new entry in the vcs-test server (feasible but tedious, and not easily updated by open-source contributors), or
• a way to set up an HTTPS proxy in a script_test, or
• a way to explicitly populate the module cache from the contents of a local repository (#28835).
Fixes#27093
Updates #28835
Change-Id: I72702a7e791f8815965f0f87c82a30df4d6f0151
Reviewed-on: https://go-review.googlesource.com/c/153819
Run-TryBot: Bryan C. Mills <bcmills@google.com>
Reviewed-by: Jay Conrod <jayconrod@google.com>
This change encodes the current behavior in mod_clean_cache.txt.
A fix for that behavior will probably have to wait for 1.13.
Updates #28680
Change-Id: I216b5a783971309cc75187502bddccc58c3a9c35
Reviewed-on: https://go-review.googlesource.com/c/153818
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
By combining the load+op, we may force the op to happen earlier in
the store chain. That might force the SymAddr operation earlier, and
in particular earlier than its corresponding VarDef. That leads to
an invalid schedule, so avoid that.
This is kind of a hack to work around the issue presented. I think
the underlying problem, that LEAQ is not directly ordered with respect
to its vardef, is the real problem. The benefit of this CL is that
it fixes the immediate issue, is small, and obviously won't break
anything. A real fix for this issue is much more invasive.
The go binary is unchanged in size.
This situation just doesn't occur very often.
Fixes#28445
Change-Id: I13a765e13f075d5b6808a355ef3c43cdd7cd47b6
Reviewed-on: https://go-review.googlesource.com/c/153641
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Both the encoding/xml and encoding/json packages support custom
marshalers for JSON and XML, as well as the basic encoding.TextMarshaler
and encoding.TextUnmarshaler interfaces, but the docs and examples for
these are missing.
There are docs for how to use encoding.TextMarshaler and
encoding.TextUnmarshaler in encoding/json, but not encoding/xml. There
are no examples for how to use them with either json or xml. This commit
includes docs for encoding/xml and examples for both encoding/json and
encoding/xml.
There is an example using custom marshalers MarshalJSON and
UnmarshalJSON in encoding/json, but not MarshalXML and UnmarshalXML in
encoding/json. These docs are more so necessary for encoding/xml because
the complexities of XML documents is significantly greater than JSON
documents which more often leads to the need for custom marshaling. The
encoding/json package includes an example of how to write a custom
marshaler, and this commit includes the same example for the xml
package.
All examples are mirrored off the existing custom marshaler example in
encoding/json.
Fixes#6859
Change-Id: Ic93abc27c0b4d5e48dea6ede4e20b1bedca4ab39
Reviewed-on: https://go-review.googlesource.com/c/76350
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
This change avoids creating zero length location lists by
repairing an overly aggressive change in CL146718
and by explicitly checking for and filtering out any
zero-length lists that are detected (building
compiler+runtime creates a single one).
Updates #28486.
Change-Id: I01c571fee2376474c7f3038e801bd58fd9e0b820
Reviewed-on: https://go-review.googlesource.com/c/150097
Run-TryBot: David Chase <drchase@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Heschi Kreinick <heschi@google.com>
For now, this CL adds as a first benchmark the formatting of
a 10,000 element array literal. It is easy to add additional
test cases as we see fit.
name time/op
Format/array1-10000-4 26.7ms ± 7%
name speed
Format/array1-10000-4 2.43MB/s ± 6%
name alloc/op
Format/array1-10000-4 5.52MB ± 0%
name allocs/op
Format/array1-10000-4 119k ± 0%
Updates #26528.
Change-Id: Ic8ec8f70160d122b877740412d4d4406f5f4b345
Reviewed-on: https://go-review.googlesource.com/c/153642
Reviewed-by: Ian Lance Taylor <iant@golang.org>
This test is currently too flakey on openbsd/arm - ignore failures for the time
being.
Change-Id: Ia334d188f505167e691177ebe2c7a2df54bf556a
Reviewed-on: https://go-review.googlesource.com/c/153579
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Regenerate ztypes for openbsd/arm - most of the changes relate to the OpenBSD armv7 port
switching to EABI in September 2016.
Also use signed char when generating openbsd/arm ztypes, to avoid inconsistencies between
architectures impacting MI code.
Change-Id: I9d2e19c1ac045922e270896861c830f94fc59c10
Reviewed-on: https://go-review.googlesource.com/c/153578
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Otherwise we can fail to remove a unreadable empty directory.
Fixes#29178
Change-Id: I43d5c89fce57a86626abe2a1c2bbf145716e087b
Reviewed-on: https://go-review.googlesource.com/c/153720
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Use smaller heap on Windows/ARM, which generally does not have
page file enabled and therefore has limited virtual address space.
Updates #26148
Signed-off-by: Jordan Rhee <jordanrh@microsoft.com>
Change-Id: I4030be24a10fab7b9b659e3736b7e83f10710bfa
Reviewed-on: https://go-review.googlesource.com/c/153719
Run-TryBot: Hyang-Ah Hana Kim <hyangah@gmail.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Fix profiling handler to get the correct g for the m being profiled.
Store a pointer to the TLS slot holding g in the thread's m. This
enables the profiling handler to get the current g for the thread,
even if the thread is executing external code or system code.
Updates #26148
Signed-off-by: Jordan Rhee <jordanrh@microsoft.com>
Change-Id: Ie061284c12341c76c7d96cc0c2d5bac969230829
Reviewed-on: https://go-review.googlesource.com/c/153718
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Take advantage of the new /*line*/ comments.
Fixes#26745
Change-Id: I8098642e0f11f7418fe81b9a08dbe07671f930fe
Reviewed-on: https://go-review.googlesource.com/c/151598
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
Use the actual loader result in findModule instead of making
assumptions about nesting in the build list.
As a side-effect, this produces much clearer error messages for
packages that (for one reason or another) failed to load.
Adjust the package and module path outside a module to
"command-line-arguments". That string already appears in the output of
a number of (module-mode and GOPATH-mode) commands for file arguments,
and as far as I can tell operation outside a module is currently the
only case in which the module path of a package is not actually a
prefix of the import path.
Fixes#28011Fixes#27099Fixes#28943
Updates #27102
Updates #28459
Updates #27063
Change-Id: I61d5556df7b1b7d1efdaffa892f0e3e95b612d87
Reviewed-on: https://go-review.googlesource.com/c/153459
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
When functions are inlined, for instructions in the inlined body, does
-S print the location of the call, or the location of the body? Right
now, we do the former. I'd like to do the latter by default, it makes
much more sense when reading disassembly. With mid-stack inlining
enabled in more cases, this quandry will come up more often.
The original behavior is still available with -S=2. Some tests
use this mode (so they can find assembly generated by a particular
source line).
This helped me with understanding what the compiler was doing
while fixing #29007.
Change-Id: Id14a3a41e1b18901e7c5e460aa4caf6d940ed064
Reviewed-on: https://go-review.googlesource.com/c/153241
Reviewed-by: David Chase <drchase@google.com>
[root] reports whether the current effective user ID is 0.
Updates #29127
Change-Id: I9ef42f1271ea669689011e7ceff4d918c0cecb6b
Reviewed-on: https://go-review.googlesource.com/c/153637
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
This commit adds support to read XCOFF files and AIX big archives in
go/internal/gccgoimporter.
Fixes: #29113
Change-Id: Id84d40358ff98fae5a576d1ebdd65980896365b9
Reviewed-on: https://go-review.googlesource.com/c/152720
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
This commit adds support to read AIX big archive inside internal/xcoff
package.
Change-Id: I4317b40824b24312a69c918dfc6438dc3aff7be7
Reviewed-on: https://go-review.googlesource.com/c/153398
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
The OpenBSD arm port switched to EABI in September 2016 - this revises the layout
of the runtime definitions to match what the kernel currently uses.
Change-Id: I1bca7de56979f576862a7c280631e835f7ae4278
Reviewed-on: https://go-review.googlesource.com/c/153577
Reviewed-by: Ian Lance Taylor <iant@golang.org>
The set of bad pointer typedefs changes as we see more typedefs, so
avoid looking in the cache when we find one.
Fixes#29175
Change-Id: Idd82289bdd8628d11a983fa5ec96517e3a5bcbf1
Reviewed-on: https://go-review.googlesource.com/c/153597
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Saves about 35% on total test time on my laptop.
Fixes#26471
Change-Id: I15b28b1bc00f889934d577dc7996864bbab10105
Reviewed-on: https://go-review.googlesource.com/c/153499
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
This avoids a warning from old versions of the GNU linker or glibc.
No test because these old versions are not readily available.
I tested this by hand on CentOS 6.
Fixes#28722
Change-Id: I16640c9b83a79f759ec68fac64874803e74fbbfb
Reviewed-on: https://go-review.googlesource.com/c/153257
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
CL 122575 and its successors introduced a loop calling loadDWARF,
whereas before we only called it once. Pass a single typeConv to each
call, rather than creating a new one in loadDWARF itself. Change the
maps from dwarf.Type to use string keys rather than dwarf.Type keys,
since when the DWARF is reloaded the dwarf.Type pointers will be
different. These changes permit typeConv.Type to return a consistent
value for a given DWARF type, avoiding spurious type conversion errors
due to typedefs loaded after the first loop iteration.
Fixes#27340
Change-Id: Ic33467bbfca4c54e95909621b35ba2a58216d96e
Reviewed-on: https://go-review.googlesource.com/c/152762
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
When converting a method to a function, like this:
type T ...
func (t T) foo() {
}
var t T
f := t.foo
We need to build a wrapper function for the partially evaluated
method. Currently that wrapper function gets the line number of
the first place where t.foo appears. Instead it should have the
line number of where foo is declared.
Fixes#26839
Change-Id: I7dbe2094e53d5d336f329273f10f8430e0af544e
Reviewed-on: https://go-review.googlesource.com/c/153498
Run-TryBot: Keith Randall <khr@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
Without this, each additional C frame found via SetCgoTraceback will
cause a frame to be dropped from the bottom of the traceback stack.
Fixes#29034
Change-Id: I90aa6b2a1dced90c69b64c5dd565fe64a25724a3
Reviewed-on: https://go-review.googlesource.com/c/151917
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
At the moment, the cover tool does not check that the argument of -var
is a valid identifier. Hence, it could generate a file that fails to
compile afterwards.
Updates #25280
Change-Id: I6eb1872736377680900a18a4a28ba002ab5ea8ca
Reviewed-on: https://go-review.googlesource.com/c/120316
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Replace the 128-bit multiplication in 4 parts with bits.Mul64
and two single-width multiplications. This simplifies the code
and increases throughput by ~50% on amd64.
name old time/op new time/op delta
Fnv128KB-4 9.64µs ± 0% 6.09µs ± 0% -36.89% (p=0.016 n=4+5)
Fnv128aKB-4 9.11µs ± 0% 6.17µs ± 5% -32.32% (p=0.008 n=5+5)
name old speed new speed delta
Fnv128KB-4 106MB/s ± 0% 168MB/s ± 0% +58.44% (p=0.016 n=4+5)
Fnv128aKB-4 112MB/s ± 0% 166MB/s ± 5% +47.85% (p=0.008 n=5+5)
Change-Id: Id752f2a20ea3de23a41e08db89eecf2bb60b7e6d
Reviewed-on: https://go-review.googlesource.com/c/133936
Run-TryBot: Matt Layher <mdlayher@gmail.com>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Matt Layher <mdlayher@gmail.com>
Reviewed-by: Robert Griesemer <gri@golang.org>
I think we ought to make these tests hermetic, but in the meantime we
should at least make them pass.
Fixes#27692
Updates #28856
Change-Id: Ia78fa60e998dea3c871f640ffa2ece67b054f866
Reviewed-on: https://go-review.googlesource.com/c/153460
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
Since CL 148517, several commands (including list and get) work when
GO111MODULE=on even when no go.mod file is present. This broke an
assumption made by "fix" and "generate" which caused panics when run
with a list of .go files (whether or not the command was run inside a
module).
This change fixes those assumptions and adds test cases for other
commands run outside modules.
Fixes#29097
Change-Id: I7927559769c5d4617d73eb63f3b17e2f26d8c219
Reviewed-on: https://go-review.googlesource.com/c/153158
Reviewed-by: Bryan C. Mills <bcmills@google.com>
