]> Cypherpunks repositories - gostls13.git/log
gostls13.git
4 years ago[dev.regabi] cmd/compile: reorganize ir/fmt.go
Russ Cox [Sat, 5 Dec 2020 05:02:46 +0000 (00:02 -0500)]
[dev.regabi] cmd/compile: reorganize ir/fmt.go

This code is a few layer of abstraction stacked up on top
of each other, and they're hard to see all at the same time
because the file is pretty mixed up. As much as I try to avoid
code rearrangement to keep history, this one is long overdue.

A followup CL will cut out some of the layers, and the diff will be
much clearer what's going on with the code ordered with
callers near callees, as it is now.

Passes buildall w/ toolstash -cmp.

Change-Id: Iffc49d43cf4be9fab47e2dd59a5f98930573350f
Reviewed-on: https://go-review.googlesource.com/c/go/+/275773
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: remove "short" node header mode
Russ Cox [Fri, 4 Dec 2020 23:40:24 +0000 (18:40 -0500)]
[dev.regabi] cmd/compile: remove "short" node header mode

This is unreachable code - the only way short can be true is
if verb == 'S', but jconv is only called when verb == 'j'.
Simplify by removing.

Passes buildall w/ toolstash -cmp.

Change-Id: I27bd38319f72215069e940b320b5c82608e2651a
Reviewed-on: https://go-review.googlesource.com/c/go/+/275772
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: arrange for typecheck1 to end in switch
Russ Cox [Fri, 4 Dec 2020 16:37:54 +0000 (11:37 -0500)]
[dev.regabi] cmd/compile: arrange for typecheck1 to end in switch

Ending typecheck1 in the switch makes it safe for each case
to do an appropriate type assertion. The main change is dropping
the computation of "ok" and using the syntax nodes themselves
to decide what's OK.

Passes buildall w/ toolstash -cmp.

Change-Id: I2a1873a51e3f1194d74bb87a6653cb9857a02a1b
Reviewed-on: https://go-review.googlesource.com/c/go/+/275444
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] test: add exhaustive test of evaluated but not used
Russ Cox [Fri, 4 Dec 2020 17:40:39 +0000 (12:40 -0500)]
[dev.regabi] test: add exhaustive test of evaluated but not used

Change-Id: I49db03c88b7595f1ea593df568244ad6aad3b024
Reviewed-on: https://go-review.googlesource.com/c/go/+/275443
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: generate Node methods using program
Russ Cox [Fri, 4 Dec 2020 15:17:50 +0000 (10:17 -0500)]
[dev.regabi] cmd/compile: generate Node methods using program

Add Node method generator by Matthew Dempsky,
lightly adapted to account for a few special cases.
No more writing these by hand.

Change-Id: I6933b895df666928b851bddf81b994799c0c97f7
Reviewed-on: https://go-review.googlesource.com/c/go/+/275434
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: silence errors about missing blank methods
Matthew Dempsky [Thu, 3 Dec 2020 22:00:19 +0000 (14:00 -0800)]
[dev.regabi] cmd/compile: silence errors about missing blank methods

If an interface contains a blank method, that's already an error. No
need for useless follow-up error messages about not implementing them.

Fixes #42964.

Change-Id: I5bf53a8f27d75d4c86c61588c5e2e3e95563d320
Reviewed-on: https://go-review.googlesource.com/c/go/+/275294
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
4 years ago[dev.regabi] cmd/compile: change iexport to avoid map[ir.Node]
Matthew Dempsky [Sun, 6 Dec 2020 20:14:46 +0000 (12:14 -0800)]
[dev.regabi] cmd/compile: change iexport to avoid map[ir.Node]

In the past, we had a lot of trouble with misusing *types.Sym
throughout the frontend, so I tried to push us towards always passing
around ONAMEs instead. But for constructing and writing out the symbol
indexes for the indexed export data, keying by *types.Sym is exactly
what we want.

Passes buildall w/ toolstash -cmp.

Updates #42982.

Change-Id: Idd8f1fb057d75a52a34ebc7788d9332fb49caf8d
Reviewed-on: https://go-review.googlesource.com/c/go/+/275755
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
4 years ago[dev.regabi] cmd/compile: change nowritebarrierrec to use map[*ir.Func]
Matthew Dempsky [Sun, 6 Dec 2020 00:33:59 +0000 (16:33 -0800)]
[dev.regabi] cmd/compile: change nowritebarrierrec to use map[*ir.Func]

All of the uses were already using *ir.Func index operands, so only
needs the map type itself updated.

Passes buildall w/ toolstash -cmp.

Updates #42982.

Change-Id: I568d8601f3eb077e07e887f2071aa1a2667d803c
Reviewed-on: https://go-review.googlesource.com/c/go/+/275754
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>

4 years ago[dev.regabi] cmd/compile: replace NodeQueue with NameQueue
Matthew Dempsky [Sun, 6 Dec 2020 20:29:42 +0000 (12:29 -0800)]
[dev.regabi] cmd/compile: replace NodeQueue with NameQueue

Similar to the previous CL, the only two users of NodeQueue only
needed it for tracking objects, not arbitrary AST nodes. So change
it's signature to use *Name instead of Node.

This does require a tweak to the nowritebarrierrec checker, because
previously it was pushing the ODCLFUNC *Func pointers into the queue,
whereas now we push the ONAME/PFUNC *Name pointers instead. However,
it's trivial and safe to flip between them.

Also, this changes a handful of export-related code from Node to
*Name, to avoid introducing type assertions within iexport.go.

Passes buildall w/ toolstash -cmp.

Updates #42982.

Change-Id: I867f9752121509fc3da753978c6a41d5015bc0ce
Reviewed-on: https://go-review.googlesource.com/c/go/+/275753
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years ago[dev.regabi] cmd/compile: change NodeSet to NameSet
Matthew Dempsky [Sat, 5 Dec 2020 23:49:03 +0000 (15:49 -0800)]
[dev.regabi] cmd/compile: change NodeSet to NameSet

The only user of NodeSet (computing initialization dependencies) only
needs to store *Names in this structure. So change its definition to
match that need, and update the code in initorder.go accordingly.

Passes buildall w/ toolstash -cmp.

Updates #42982.

Change-Id: I181a8aaf9bc71e88f4ac009c4f381a718080e48f
Reviewed-on: https://go-review.googlesource.com/c/go/+/275752
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years ago[dev.regabi] cmd/compile: replace ir.Node with *ir.Name in Order
Cuong Manh Le [Fri, 4 Dec 2020 04:38:47 +0000 (11:38 +0700)]
[dev.regabi] cmd/compile: replace ir.Node with *ir.Name in Order

Passes buildall w/ toolstash -cmp.

Updates #42982

Change-Id: I7121c37f72ccbc141a7dd17fba1753f2c6289908
Reviewed-on: https://go-review.googlesource.com/c/go/+/275353
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: replace ir.Node with *ir.Name in Liveness
Cuong Manh Le [Fri, 4 Dec 2020 04:07:25 +0000 (11:07 +0700)]
[dev.regabi] cmd/compile: replace ir.Node with *ir.Name in Liveness

Passes buildall w/ toolstash -cmp.

Updates #42982

Change-Id: Iad8df321adfd576da070c13ed16a9651d4e59ad8
Reviewed-on: https://go-review.googlesource.com/c/go/+/275352
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: rewrite code to use DeclaredBy
Matthew Dempsky [Fri, 4 Dec 2020 00:57:56 +0000 (16:57 -0800)]
[dev.regabi] cmd/compile: rewrite code to use DeclaredBy

Passes buildall w/ toolstash -cmp.

Updates #42990.

[git-generate]
cd src/cmd/compile/internal/gc
rf '
ex {
  import "cmd/compile/internal/ir"
  var x, stmt ir.Node
  x.Name() != nil && x.Name().Defn == stmt ->  ir.DeclaredBy(x, stmt)
  x.Name() == nil || x.Name().Defn != stmt -> !ir.DeclaredBy(x, stmt)
}
'

Change-Id: I222a757296dbcb5d0889d617d221a9d7319f2d74
Reviewed-on: https://go-review.googlesource.com/c/go/+/275306
Reviewed-by: Russ Cox <rsc@golang.org>
Trust: Matthew Dempsky <mdempsky@google.com>

4 years ago[dev.regabi] cmd/compile: add SameSource, Uses, and DeclaredBy helpers
Matthew Dempsky [Fri, 4 Dec 2020 00:52:45 +0000 (16:52 -0800)]
[dev.regabi] cmd/compile: add SameSource, Uses, and DeclaredBy helpers

Currently, because we use the same *Name to represent both declaration
and uses of an object, it's ambiguous what "n1 == n2" means when
comparing two Node values. It can mean any of: Are these the same
syntactic element? Is n1 a use of declared variable n2? Are n1 and n2
both uses of the same declared variable?

We'd like to introduce a new IdentExpr node to replace use of Name
within the AST, but that means those three cases need to be handled
differently. The first case needs to stay "n1 == n2", but the other
cases need to become "n1.Name() == n2" and "n1.Name() == n2.Name()",
respectively. ("n1.Name() == n2.Name()" also currently works for the
second case, but eventually we'll want to get rid of the Name.Name
method.)

This CL introduces helper functions SameSource and Uses to handle
these cases. It also introduces DeclaredBy, which is another somewhat
common case that the next CL introduces uses of.

Passes buildall w/ toolstash -cmp.

Updates #42990.

Change-Id: Ia816c124446e9067645d5820a8163f295968794f
Reviewed-on: https://go-review.googlesource.com/c/go/+/275305
Reviewed-by: Russ Cox <rsc@golang.org>
Trust: Matthew Dempsky <mdempsky@google.com>

4 years ago[dev.regabi] cmd/compile: remove DeepCopyNode interface
Russ Cox [Fri, 4 Dec 2020 05:30:53 +0000 (00:30 -0500)]
[dev.regabi] cmd/compile: remove DeepCopyNode interface

The only reason for the DeepCopyNode interface was to
allow the type syntaxes to avoid being constrained by
Left, Right etc. methods. Now those are gone, so the
general traversal methods they implement (doChildren, editChildren)
do the right thing for DeepCopy.

Passes buildall w/ toolstash -cmp.

Change-Id: I54672c011114a95efabff32dbcf02e6071f91b9e
Reviewed-on: https://go-review.googlesource.com/c/go/+/275379
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: avoid general traversal in deadcode
Russ Cox [Fri, 4 Dec 2020 04:56:16 +0000 (23:56 -0500)]
[dev.regabi] cmd/compile: avoid general traversal in deadcode

deadcode is trying to walk the statements it can find,
but it can sweep in other nodes too. Stop doing that:
only walk known statements containing statements.

Otherwise, if we put panics in expression accessors that
shouldn't be used anymore, deadcode can trip them.

deadcode would be a good candidate to rewrite using
EditChildren, but that would certainly cause toolstash
changes, since deadcode is so ad-hoc about exactly
which parts of the function it looks at. For now just
remove the general traversal and leave as is.

Passes buildall w/ toolstash -cmp.

Change-Id: I06481eb87350905597600203c4fa724d55645b46
Reviewed-on: https://go-review.googlesource.com/c/go/+/275377
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: implement editChildren for nodes
Russ Cox [Fri, 4 Dec 2020 02:29:23 +0000 (21:29 -0500)]
[dev.regabi] cmd/compile: implement editChildren for nodes

Put each node in charge of its EditChildren implementation.
This removes the final generic use of Left, SetLeft, Right, SetRight,
and so on in package ir.

Passes buildall w/ toolstash -cmp.

Change-Id: I9821cc20f5b91cc9b44eb1f386cc82f20cd6770c
Reviewed-on: https://go-review.googlesource.com/c/go/+/275376
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: implement doChildren for nodes
Russ Cox [Fri, 4 Dec 2020 02:02:19 +0000 (21:02 -0500)]
[dev.regabi] cmd/compile: implement doChildren for nodes

Put each node in charge of its DoChildren implementation.
This removes a generic use of Left, Right, and so on
in func DoChildren, heading toward removing those even from
being used in package ir.

Passes buildall w/ toolstash -cmp.

Change-Id: Ibdf56f36801217cf24549e063da0078c1820a56b
Reviewed-on: https://go-review.googlesource.com/c/go/+/275375
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: implement copy for nodes
Russ Cox [Thu, 3 Dec 2020 23:43:18 +0000 (18:43 -0500)]
[dev.regabi] cmd/compile: implement copy for nodes

Put each node in charge of making copies of its own slices.
This removes a generic use of Body, SetBody, and so on
in func Copy, heading toward removing those even from
being used in package ir.

Passes buildall w/ toolstash -cmp.

Change-Id: I249b7fe54cf72e9d2f0467b10f3f257abf9b29b9
Reviewed-on: https://go-review.googlesource.com/c/go/+/275374
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: use ir.EditChildren for inline rewriting
Russ Cox [Thu, 3 Dec 2020 19:06:41 +0000 (14:06 -0500)]
[dev.regabi] cmd/compile: use ir.EditChildren for inline rewriting

This CL rephrases the general inlining rewriter in terms of ir.EditChildren.
It is the final part of the code that was processing arbitrary nodes using
Left, SetLeft, and so on. After this CL, there should be none left except
for the implementations of DoChildren and EditChildren, which fall next.

Passes buildall w/ toolstash -cmp.

Change-Id: I9c36053360cd040710716f0b39397a80114be713
Reviewed-on: https://go-review.googlesource.com/c/go/+/275373
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: use ir.Find for "search" traversals
Russ Cox [Thu, 3 Dec 2020 01:18:47 +0000 (20:18 -0500)]
[dev.regabi] cmd/compile: use ir.Find for "search" traversals

This CL converts all the generic searching traversal to use ir.Find
instead of relying on direct access to Left, Right, and so on.

Passes buildall w/ toolstash -cmp.

Change-Id: I4d951aef630c00bf333f24be79565cc564694d04
Reviewed-on: https://go-review.googlesource.com/c/go/+/275372
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: introduce IR visitors
Russ Cox [Thu, 3 Dec 2020 17:57:38 +0000 (12:57 -0500)]
[dev.regabi] cmd/compile: introduce IR visitors

This CL introduces the general visitor functionality that will replace
the Left, SetLeft, Right, SetRight, etc methods in the Node interface.

For now, the CL defines the functionality in terms of those methods,
but eventually the Nodes themselves will implement DoChildren
and EditChildren and be relieved of implementing Left, SetLeft, and so on.

The CL also updates Inspect (which moved to visit.go) and DeepCopy
to use the new functionality.

The Find helper is not used in this CL but will be used in a future one.

Passes buildall w/ toolstash -cmp.

Change-Id: Id0eea654a884ab3ea25f48bd8bdd71712b5dcb44
Reviewed-on: https://go-review.googlesource.com/c/go/+/275311
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: replace inlcopy with ir.DeepCopy
Russ Cox [Thu, 3 Dec 2020 03:54:33 +0000 (22:54 -0500)]
[dev.regabi] cmd/compile: replace inlcopy with ir.DeepCopy

Now inlcopy and ir.DeepCopy are semantically the same,
so drop the inlcopy implementation.

Passes buildall w/ toolstash -cmp.

Change-Id: Id2abb39a412a8e57167a29be5ecf76e990dc9d3d
Reviewed-on: https://go-review.googlesource.com/c/go/+/275310
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: adjustments to Copy and DeepCopy
Russ Cox [Thu, 3 Dec 2020 03:42:24 +0000 (22:42 -0500)]
[dev.regabi] cmd/compile: adjustments to Copy and DeepCopy

DeepCopy is not called DeepSepCopy, so it should use Copy, not SepCopy.

Also, the old gc.treecopy, which became ir.DeepCopy, only copied
the Left, Right, and List fields - not Init, Rlist, Body - and I didn't
notice when I moved it over. A general utility function should of
course copy the whole node, so do that.

Finally, the semantics of Copy should not depend on whether a
particular child node is held directly in a field or in a slice,
so make Copy duplicate the slice backing arrays as well.
(Logically, those backing arrays are part of the node storage.)

Passes buildall w/ toolstash -cmp.

Change-Id: I18fbe3f2b40078f566ed6370684d5585052b36a1
Reviewed-on: https://go-review.googlesource.com/c/go/+/275309
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: swap inlining order of if then vs else blocks
Russ Cox [Thu, 3 Dec 2020 20:40:46 +0000 (15:40 -0500)]
[dev.regabi] cmd/compile: swap inlining order of if then vs else blocks

The upcoming general iterators will process nodes in
source code order, meaning that the "then" block comes
before the "else" block. But for an if node, "then" is Body
while "else" is Rlist, and the inliner processes Rlist first.

The order of processing changes the order of inlining decisions,
which can affect which functions are inlined, but in general
won't affect much. (It's not like we know that we should prefer
to inline functions in else bodies over then bodies.)

Swapping these is not safe for toolstash -cmp.
Doing it in a separate CL lets the upcoming CLs all be toolstash-safe.

Change-Id: Id16172849239b0564930d2bbff1260ad6d03d5ab
Reviewed-on: https://go-review.googlesource.com/c/go/+/275308
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: eliminate more SetOrig
Matthew Dempsky [Thu, 3 Dec 2020 19:56:29 +0000 (11:56 -0800)]
[dev.regabi] cmd/compile: eliminate more SetOrig

This CL consolidates and cleans up fmt.go's logic for skipping past
Nodes introduced during typechecking. This allows eliminating SetOrig
on ConvExpr and Name. Also changes ConstExpr.SetOrig to a panic for
good measure.

The only remaining SetOrig uses now are for rewriting multi-value
"f(g())" calls and "return g()" statements, and type-checking
composite literals. It should be possible to eliminate both of those
as well.

Passes buildall w/ toolstash -cmp.

Change-Id: I478aea1a17dfb7a784293b930bf9081637eb2d7a
Reviewed-on: https://go-review.googlesource.com/c/go/+/275179
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
4 years ago[dev.regabi] cmd/compile: store types.Field on {Selector,CallPart}Expr
Matthew Dempsky [Thu, 3 Dec 2020 05:38:20 +0000 (21:38 -0800)]
[dev.regabi] cmd/compile: store types.Field on {Selector,CallPart}Expr

It's useful to have quick access to the types.Field that a given
selector or method value expression refer to. Previously we abused Opt
for this, but couldn't do that for OCALLPART because escape analysis
uses Opt.

Now that we have more flexibility, we can simply add additional
pointer fields for this. This also allows getting rid of an unneeded
ONAME node for OCALLPART.

Passes buildall w/ toolstash -cmp.

Change-Id: I980d7bdb19abfd0b6f58a232876861b88dee1e47
Reviewed-on: https://go-review.googlesource.com/c/go/+/275034
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
Trust: Matthew Dempsky <mdempsky@google.com>

4 years ago[dev.regabi] cmd/compile: add ConstExpr
Matthew Dempsky [Thu, 3 Dec 2020 03:26:56 +0000 (19:26 -0800)]
[dev.regabi] cmd/compile: add ConstExpr

Currently, we represent constant-folded expressions with Name, which
is suboptimal because Name has a lot of fields to support declared
names (which are irrelevant to constant-folded expressions), while
constant expressions are fairly common.

This CL introduces a new lightweight ConstExpr type that simply wraps
an existing expression and associates it with a value.

Passes buildall w/ toolstash -cmp.

name                      old time/op       new time/op       delta
Template                        252ms ± 3%        254ms ± 1%     ~     (p=0.821 n=12+10)
Unicode                         120ms ± 2%        107ms ± 7%  -11.09%  (p=0.000 n=12+12)
GoTypes                         918ms ± 2%        918ms ± 1%     ~     (p=0.974 n=12+10)
Compiler                        5.19s ± 1%        5.18s ± 0%     ~     (p=0.190 n=12+11)
SSA                             12.4s ± 1%        12.3s ± 1%     ~     (p=0.283 n=10+12)
Flate                           152ms ± 2%        148ms ± 4%   -2.68%  (p=0.007 n=10+12)
GoParser                        212ms ± 1%        211ms ± 2%     ~     (p=0.674 n=10+12)
Reflect                         543ms ± 3%        542ms ± 3%     ~     (p=0.799 n=12+12)
Tar                             224ms ± 2%        225ms ± 2%     ~     (p=0.378 n=12+12)
XML                             292ms ± 1%        299ms ± 3%   +2.18%  (p=0.006 n=10+12)

name                      old user-time/op  new user-time/op  delta
Template                        243ms ± 4%        244ms ± 5%     ~     (p=0.887 n=12+12)
Unicode                         112ms ± 6%        100ms ±10%  -10.76%  (p=0.000 n=12+12)
GoTypes                         898ms ± 3%        895ms ± 3%     ~     (p=0.671 n=12+12)
Compiler                        5.10s ± 1%        5.08s ± 1%     ~     (p=0.104 n=12+11)
SSA                             12.2s ± 2%        12.1s ± 1%     ~     (p=0.487 n=11+12)
Flate                           144ms ± 6%        145ms ± 5%     ~     (p=0.695 n=12+11)
GoParser                        205ms ± 5%        204ms ± 3%     ~     (p=0.514 n=12+12)
Reflect                         528ms ± 3%        531ms ± 4%     ~     (p=0.630 n=12+12)
Tar                             218ms ± 4%        219ms ± 3%     ~     (p=0.843 n=12+12)
XML                             284ms ± 5%        291ms ± 5%     ~     (p=0.069 n=11+12)

name                      old alloc/op      new alloc/op      delta
Template                       37.0MB ± 0%       36.7MB ± 0%   -0.72%  (p=0.000 n=12+12)
Unicode                        31.9MB ± 0%       29.5MB ± 0%   -7.60%  (p=0.000 n=12+12)
GoTypes                         119MB ± 0%        118MB ± 0%   -0.40%  (p=0.000 n=12+12)
Compiler                        629MB ± 0%        626MB ± 0%   -0.36%  (p=0.000 n=11+12)
SSA                            1.45GB ± 0%       1.43GB ± 0%   -0.78%  (p=0.000 n=12+12)
Flate                          22.2MB ± 0%       21.9MB ± 0%   -1.12%  (p=0.000 n=12+12)
GoParser                       29.4MB ± 0%       29.3MB ± 0%   -0.36%  (p=0.000 n=12+12)
Reflect                        76.1MB ± 0%       75.8MB ± 0%   -0.38%  (p=0.000 n=12+12)
Tar                            33.4MB ± 0%       33.2MB ± 0%   -0.61%  (p=0.000 n=12+12)
XML                            43.2MB ± 0%       42.8MB ± 0%   -1.03%  (p=0.000 n=11+12)

name                      old allocs/op     new allocs/op     delta
Template                         375k ± 0%         375k ± 0%     ~     (p=0.854 n=12+12)
Unicode                          300k ± 0%         300k ± 0%     ~     (p=0.766 n=12+12)
GoTypes                         1.30M ± 0%        1.30M ± 0%     ~     (p=0.272 n=12+12)
Compiler                        5.89M ± 0%        5.89M ± 0%     ~     (p=0.478 n=12+12)
SSA                             14.0M ± 0%        14.0M ± 0%     ~     (p=0.266 n=12+12)
Flate                            226k ± 0%         226k ± 0%     ~     (p=0.898 n=12+12)
GoParser                         313k ± 0%         313k ± 0%   -0.01%  (p=0.042 n=12+11)
Reflect                          971k ± 0%         971k ± 0%     ~     (p=0.080 n=12+12)
Tar                              342k ± 0%         342k ± 0%     ~     (p=0.600 n=12+12)
XML                              416k ± 0%         416k ± 0%     ~     (p=0.217 n=11+12)

name                      old maxRSS/op     new maxRSS/op     delta
Template                        43.1M ± 5%        42.5M ± 5%     ~     (p=0.086 n=12+12)
Unicode                         49.4M ± 2%        47.0M ± 2%   -4.88%  (p=0.000 n=12+12)
GoTypes                         85.3M ± 2%        84.6M ± 2%   -0.84%  (p=0.047 n=11+11)
Compiler                         394M ± 3%         386M ± 2%   -1.97%  (p=0.000 n=10+11)
SSA                              847M ± 4%         821M ± 2%   -2.98%  (p=0.000 n=11+12)
Flate                           36.0M ± 7%        35.2M ± 7%     ~     (p=0.128 n=12+12)
GoParser                        39.4M ± 7%        39.5M ± 4%     ~     (p=0.413 n=12+11)
Reflect                         64.0M ± 3%        63.6M ± 3%     ~     (p=0.413 n=11+12)
Tar                             43.3M ± 5%        43.3M ± 5%     ~     (p=0.503 n=12+12)
XML                             47.6M ± 4%        46.4M ± 2%   -2.46%  (p=0.013 n=11+12)

Change-Id: If5781be346351c30b2228807211b5e57f777c506
Reviewed-on: https://go-review.googlesource.com/c/go/+/275033
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
Trust: Matthew Dempsky <mdempsky@google.com>

4 years ago[dev.regabi] cmd/compile: refactoring prep for ConstExpr
Matthew Dempsky [Thu, 3 Dec 2020 07:55:42 +0000 (23:55 -0800)]
[dev.regabi] cmd/compile: refactoring prep for ConstExpr

The next CL adds ConstExpr, which is a more memory efficient
representation for constant expressions than Name. However, currently
a bunch of Val helper methods are defined on Name. This CL changes
them into standalone functions that work with any Node.Val
implementation.

There's also an existing standalone function named Int64Val, which
takes a Type argument to specify what type of integer is expected. So
to avoid collisions, this CL renames it to IntVal.

Passes buildall w/ toolstash -cmp.

[git-generate]
cd src/cmd/compile/internal/ir
rf 'mv Int64Val IntVal'
sed -i -E -e 's/\(n \*Name\) (CanInt64|((I|Ui)nt64|Bool|String)Val)\(/\1(n Node/' name.go

cd ../gc
rf '
ex {
  import "cmd/compile/internal/ir"
  var n ir.Node
  n.CanInt64() -> ir.CanInt64(n)
  n.Int64Val() -> ir.Int64Val(n)
  n.Uint64Val() -> ir.Uint64Val(n)
  n.BoolVal() -> ir.BoolVal(n)
  n.StringVal() -> ir.StringVal(n)
}
'

cd ../ir
rf '
mv CanInt64 Int64Val Uint64Val BoolVal StringVal val.go
rm Node.CanInt64 Node.Int64Val Node.Uint64Val Node.BoolVal Node.StringVal
'

Change-Id: I003140bda1690d770fd608bdd087e6d4ff00fb1f
Reviewed-on: https://go-review.googlesource.com/c/go/+/275032
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
4 years ago[dev.regabi] cmd/compile: rename addinit(n, init) to initExpr(init, n)
Russ Cox [Wed, 2 Dec 2020 22:03:18 +0000 (17:03 -0500)]
[dev.regabi] cmd/compile: rename addinit(n, init) to initExpr(init, n)

Recreated manually to push below some CLs it depended on.

Change-Id: I1b3316fcdce39cbb33e5cbb471f5cd1cd2efc1f5
Reviewed-on: https://go-review.googlesource.com/c/go/+/274599
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] all: merge master (d0c0dc682c1f) into dev.regabi
Russ Cox [Thu, 3 Dec 2020 17:33:12 +0000 (12:33 -0500)]
[dev.regabi] all: merge master (d0c0dc682c1f) into dev.regabi

Change-Id: Ia54d7306ca7550b8d5623f505070558d275faa23

4 years ago[dev.regabi] cmd/compile: handle OCONVNOP better in ssa
Russ Cox [Tue, 1 Dec 2020 17:02:16 +0000 (12:02 -0500)]
[dev.regabi] cmd/compile: handle OCONVNOP better in ssa

This CL improves handling of OCONVNOP nodes during ssa generation,
so it is not toolstash safe.

An OCONVNOP wrapper is necessary for the "for" condition of
certain compiled range loops, and the boolean evaluator was
not looking through them properly, generating unnecessary
temporaries. That change saved 8k of the (13 MB) go binary.

The other changes just streamline the handling of OCONVNOP
to be more like what OSTMTEXPR will be like. They have no
effect on output size but do tweak the ssa graph a little, which
causes different register decisions and therefore different output.

Change-Id: I9e1dcd413b60944e21554c3e3f2bdc9adcee7634
Reviewed-on: https://go-review.googlesource.com/c/go/+/274598
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
4 years agodoc/go1.16: document os package changes
Tobias Klauser [Thu, 3 Dec 2020 10:07:27 +0000 (11:07 +0100)]
doc/go1.16: document os package changes

For #39444
For #40700
Fixes #42908

Change-Id: Idae35adecd79e9d7d207f9d78cb009a980e5c8a9
Reviewed-on: https://go-review.googlesource.com/c/go/+/274477
Trust: Tobias Klauser <tobias.klauser@gmail.com>
Reviewed-by: Russ Cox <rsc@golang.org>
4 years ago[dev.regabi] cmd/compile: remove okAs
Cuong Manh Le [Wed, 2 Dec 2020 07:40:48 +0000 (14:40 +0700)]
[dev.regabi] cmd/compile: remove okAs

The check for blank in okAs is redundant with what its callers already
done, so just inline the conversion in callers side instead.

Passes toolstash-check.

Change-Id: I606105e2d2cf8e80214722a13c3101c464d20d82
Reviewed-on: https://go-review.googlesource.com/c/go/+/274793
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: remove extra typ field in Name struct
Dan Scales [Wed, 2 Dec 2020 04:51:18 +0000 (20:51 -0800)]
[dev.regabi] cmd/compile: remove extra typ field in Name struct

Noticed the typ field was duplicated, since it is also in miniExpr inside Name.

Also clarified the comments for Func, now that it is actually the ODCLFUNC node.

Change-Id: Ia483a0ad34bb409cd92c43d4ae0a6852f9e4f644
Reviewed-on: https://go-review.googlesource.com/c/go/+/274619
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Dan Scales <danscales@google.com>
Reviewed-by: Russ Cox <rsc@golang.org>
4 years agodoc/go1.16: document new behavior of asn1.Unmarshal on invalid argument
KimMachineGun [Sun, 29 Nov 2020 08:18:12 +0000 (08:18 +0000)]
doc/go1.16: document new behavior of asn1.Unmarshal on invalid argument

For #41509

Change-Id: Ie761c428710d15848cb80ffd2d85de747113f2d4
GitHub-Last-Rev: 05541624593d945d82b6f4cfae1461654eabea7b
GitHub-Pull-Request: golang/go#42315
Reviewed-on: https://go-review.googlesource.com/c/go/+/267057
Trust: Dmitri Shuralyov <dmitshur@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
4 years agodoc/go1.16: add encoding/json note for tag change
Joe Tsai [Tue, 1 Dec 2020 22:59:23 +0000 (14:59 -0800)]
doc/go1.16: add encoding/json note for tag change

For #40700
Fixes #42898

Change-Id: I652657ff8d6cce20bf868f0b1101d723d3f704d1
Reviewed-on: https://go-review.googlesource.com/c/go/+/274614
Trust: Joe Tsai <joetsai@digital-static.net>
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
4 years agodoc/go1.16: add crypto/tls Config.Clone note
Roland Shoemaker [Tue, 1 Dec 2020 21:02:44 +0000 (13:02 -0800)]
doc/go1.16: add crypto/tls Config.Clone note

For #40700
Fixes #42896

Change-Id: I842c9d60b18abe2ee061c6705a5c7ba62b224d77
Reviewed-on: https://go-review.googlesource.com/c/go/+/274613
Trust: Roland Shoemaker <roland@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
4 years ago[dev.regabi] cmd/compile: use explicit block statements for init
Russ Cox [Tue, 1 Dec 2020 04:49:25 +0000 (23:49 -0500)]
[dev.regabi] cmd/compile: use explicit block statements for init

For statements like goto that don't need an init, use an
explicit block statement instead of forcing them to have one.

There is also one call to addinit that is being replaced with
a block. That call is the source of much of my confusion
regarding init statements: walkstmt calls addinit on a statement,
whereas all the other uses of addinit are on expressions.

After this CL, they're all expressions.

Passes buildall w/ toolstash -cmp.

Change-Id: Ifdef9d318c236dc1a7567f9e9ef4a6bedd3fe81f
Reviewed-on: https://go-review.googlesource.com/c/go/+/274597
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years agogo/parser: ignore subdirectories in ParseDir
Carlos Alexandro Becker [Wed, 2 Dec 2020 21:04:13 +0000 (21:04 +0000)]
go/parser: ignore subdirectories in ParseDir

Issue and PR on GoReleaser:
- https://github.com/goreleaser/goreleaser/issues/1897
- https://github.com/goreleaser/goreleaser/pull/1899

Fixes #42951.

Change-Id: Ia0d6018e0bad59cd60cd600188c368c431032a4b
GitHub-Last-Rev: be59d85fe2d473f4dfd828a244023c4064d6e31f
GitHub-Pull-Request: golang/go#42581
Reviewed-on: https://go-review.googlesource.com/c/go/+/269897
Trust: Robert Griesemer <gri@golang.org>
Trust: Dmitri Shuralyov <dmitshur@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
4 years agocrypto/ed25519/internal/edwards25519: fix typo in comments
Filippo Valsorda [Wed, 25 Nov 2020 01:40:32 +0000 (02:40 +0100)]
crypto/ed25519/internal/edwards25519: fix typo in comments

Change-Id: I8133762d53d9e5d3cc13e0f97b9679a3248a7f0f
Reviewed-on: https://go-review.googlesource.com/c/go/+/273087
Trust: Filippo Valsorda <filippo@golang.org>
Reviewed-by: Katie Hockman <katie@golang.org>
4 years ago[dev.regabi] cmd/compile: delete OEMPTY
Russ Cox [Tue, 1 Dec 2020 02:59:44 +0000 (21:59 -0500)]
[dev.regabi] cmd/compile: delete OEMPTY

Not toolstash -cmp safe, so split into its own CL.

Change-Id: I523227514a95e19c9afe17556d754a03164bce76
Reviewed-on: https://go-review.googlesource.com/c/go/+/274595
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: merge OBLOCK and OEMPTY
Russ Cox [Tue, 1 Dec 2020 02:56:24 +0000 (21:56 -0500)]
[dev.regabi] cmd/compile: merge OBLOCK and OEMPTY

OEMPTY is an empty *statement*, but it confusingly
gets handled as an expression in a few places.
More confusingly, OEMPTY often has an init list,
making it not empty at all. Replace uses and analysis
of OEMPTY with OBLOCK instead.

Passes buildall w/ toolstash -cmp.

Change-Id: I8d4fcef151e4f441fa19b1b96da5272d778131d6
Reviewed-on: https://go-review.googlesource.com/c/go/+/274594
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years agocmd/go: fix TestNewReleaseRebuildsStalePackagesInGOPATH
Russ Cox [Wed, 2 Dec 2020 17:56:29 +0000 (12:56 -0500)]
cmd/go: fix TestNewReleaseRebuildsStalePackagesInGOPATH

Broken during CL 267719.

Change-Id: If5acb8231d3053c0e714a79c02cb56eaba6e74e6
Reviewed-on: https://go-review.googlesource.com/c/go/+/274854
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
4 years agocmd/go: stop tests from using network during -short
Russ Cox [Wed, 2 Dec 2020 15:14:53 +0000 (10:14 -0500)]
cmd/go: stop tests from using network during -short

It turned out that "go get" was using the network to look up
https://github.com?go-get=1 while resolving github.com/google/go-cmp,
and that is not the fastest page to load.
Stop that lookup by adjusting the path prefixes in the vcs table.

It also turned out that "go get" was using the network to look up
https://rsc.io?go-get=1 while resolving https://rsc.io/nonexist.svn.
That's a bit more defensible maybe, since rsc.io is not a known VCS host.
But for tests we really want to avoid the network entirely, so this CL
adds a special case in repoRootFromVCSPaths that returns a hard error
for plain "rsc.io" instead of doing the web fetch.

To keep us honest in the future, I added two automatically-set env
variables TESTGONETWORK=panic and TESTGOVCS=panic.
These cause the go command to panic rather than make a network request
or invoke a VCS command.

go test -short cmd/go now passes with these checks.

This reduced the time spent in go test -short cmd/go on my
Google workstation from 154s to 30s. (Yay network firewalls.)

Change-Id: I49207fca7f901fa011765fb984dc9cec8b691f11
Reviewed-on: https://go-review.googlesource.com/c/go/+/274441
Trust: Russ Cox <rsc@golang.org>
Trust: Jay Conrod <jayconrod@google.com>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Jay Conrod <jayconrod@google.com>
4 years agoos: add ReadFile, WriteFile, CreateTemp (was TempFile), MkdirTemp (was TempDir) from...
Russ Cox [Thu, 29 Oct 2020 17:51:20 +0000 (13:51 -0400)]
os: add ReadFile, WriteFile, CreateTemp (was TempFile), MkdirTemp (was TempDir) from io/ioutil

io/ioutil was a poorly defined collection of helpers.
Proposal #40025 moved out the generic I/O helpers to io.
This CL for proposal #42026 moves the OS-specific helpers to os,
making the entire io/ioutil package deprecated.

For #42026.

Change-Id: I018bcb2115ef2ff1bc7ca36a9247eda429af21ad
Reviewed-on: https://go-review.googlesource.com/c/go/+/266364
Trust: Russ Cox <rsc@golang.org>
Trust: Emmanuel Odeke <emmanuel@orijtech.com>
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
4 years agodoc: update signal.Notify example to use buffered channel
Cuong Manh Le [Tue, 1 Dec 2020 15:23:42 +0000 (22:23 +0700)]
doc: update signal.Notify example to use buffered channel

This if follow up of CL 274332.

Updates #9399.

Change-Id: Ic6dd534dc18227a799cbb9577979f2285596b825
Reviewed-on: https://go-review.googlesource.com/c/go/+/274393
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
4 years agocmd/go: fix unbuffered channel passed to signal.Notify
Cuong Manh Le [Tue, 1 Dec 2020 05:54:24 +0000 (12:54 +0700)]
cmd/go: fix unbuffered channel passed to signal.Notify

Unbuffered channels passed into signal.Notify can be lost
as the docs for signal.Notify caution with:

    Package signal will not block sending to c: the caller must ensure
    that c has sufficient buffer space to keep up with the expected signal
    rate. For a channel used for notification of just one signal value,
    a buffer of size 1 is sufficient.

Found by a static analyzer from Orijtech, Inc. called "sigchanyzer", but
it'll be donated to the Go project soon.

Updates #9399.

Change-Id: Ia0690e447582da028694ed65ace7b97961997b84
Reviewed-on: https://go-review.googlesource.com/c/go/+/274332
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Trust: Emmanuel Odeke <emmanuel@orijtech.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agoall: update to use filepath.WalkDir instead of filepath.Walk
Russ Cox [Wed, 4 Nov 2020 23:20:17 +0000 (18:20 -0500)]
all: update to use filepath.WalkDir instead of filepath.Walk

Now that filepath.WalkDir is available, it is more efficient
and should be used in place of filepath.Walk.
Update the tree to reflect best practices.

As usual, the code compiled with Go 1.4 during bootstrap is excluded.
(In this CL, that's only cmd/dist.)

For #42027.

Change-Id: Ib0f7b1e43e50b789052f9835a63ced701d8c411c
Reviewed-on: https://go-review.googlesource.com/c/go/+/267719
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Rob Pike <r@golang.org>
4 years agocmd/asm, cmd/internal/obj/riscv: fix branch pseudo-instructions
Quey-Liang Kao [Sat, 21 Nov 2020 14:48:55 +0000 (22:48 +0800)]
cmd/asm, cmd/internal/obj/riscv: fix branch pseudo-instructions

Pseudo branch instructions BGT, BGTU, BLE, and BLEU implemented In
CL 226397 were translated inconsistently compared to other ones due
to the inversion of registers. For instance, while "BLT a, b" generates
"jump if a < b", "BLE a, b" generates "jump if b <= a."

This CL fixes the translation in the assembler and the tests.

Change-Id: Ia757be73e848734ca5b3a790e081f7c4f98c30f2
Reviewed-on: https://go-review.googlesource.com/c/go/+/271911
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Reviewed-by: Joel Sing <joel@sing.id.au>
Run-TryBot: Joel Sing <joel@sing.id.au>

4 years ago[dev.regabi] cmd/compile: add ir.NewDeclNameAt
Matthew Dempsky [Wed, 2 Dec 2020 07:55:03 +0000 (23:55 -0800)]
[dev.regabi] cmd/compile: add ir.NewDeclNameAt

This allows directly creating an ONONAME, which is a primordial Name
before having its Op initialized. Then after an Op is assigned, we
never allow it to be reassigned.

Passes buildall w/ toolstash -cmp.

Change-Id: Ibc2f413dc68c0af6a96abfe653c25ce31b184287
Reviewed-on: https://go-review.googlesource.com/c/go/+/274620
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
4 years ago[dev.regabi] cmd/compile: add Pkg parameter to type constructors
Matthew Dempsky [Wed, 2 Dec 2020 07:05:53 +0000 (23:05 -0800)]
[dev.regabi] cmd/compile: add Pkg parameter to type constructors

Allows getting rid of the SetPkg method and also addresses a
long-standing TODO in the exporter. Suggested by rsc@.

Passes buildall w/ toolstash -cmp.

Change-Id: Ib294f75f1350572efb2e0d993d49efef884de3d4
Reviewed-on: https://go-review.googlesource.com/c/go/+/274440
Trust: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
4 years ago[dev.regabi] cmd/compile: comment out //go:linkname warning
Matthew Dempsky [Wed, 2 Dec 2020 06:11:56 +0000 (22:11 -0800)]
[dev.regabi] cmd/compile: comment out //go:linkname warning

It's noisy and not doing any harm, and we still have an entire release
cycle to revisit and address the issue properly.

Updates #42938

Change-Id: I1de5cfb495a8148c9c08b215deba38f2617fb467
Reviewed-on: https://go-review.googlesource.com/c/go/+/274732
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
4 years ago[dev.regabi] cmd/compile: add Interface, Signature, and Struct constructors
Matthew Dempsky [Tue, 1 Dec 2020 21:07:35 +0000 (13:07 -0800)]
[dev.regabi] cmd/compile: add Interface, Signature, and Struct constructors

This CL adds the remaining constructors needed to abstract away
construction of Types, and updates the compiler to use them
throughout. There's now just a couple uses within test cases to
remove.

While at it, I also replace the Func.Outnamed field with a simple
helper function, which reduces the size of function types somewhat.

Passes toolstash/buildall.

Change-Id: If1aa1095c98ae34b00380d0b3531bd63c10ce885
Reviewed-on: https://go-review.googlesource.com/c/go/+/274713
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
4 years ago[dev.regabi] cmd/compile: tweak hash bucket type descriptor
Matthew Dempsky [Tue, 1 Dec 2020 21:38:26 +0000 (13:38 -0800)]
[dev.regabi] cmd/compile: tweak hash bucket type descriptor

There's no need for the bucket type to be precise. The compiler
doesn't actually generate code that references these fields; it just
needs it for size and GC bitmap calculations.

However, changing the type field does alter the runtime type
descriptor and relocations emitted by the compiler, so this change
isn't safe for toolstash.

Change-Id: Icf79d6c4326515889b13435a575d618e3bbfbcd7
Reviewed-on: https://go-review.googlesource.com/c/go/+/274712
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
4 years agotest: match gofrontend error messages
Ian Lance Taylor [Wed, 2 Dec 2020 02:14:40 +0000 (18:14 -0800)]
test: match gofrontend error messages

The gofrontend code doesn't distinguish semicolon and newline,
and it doesn't have special treatment for EOF.

syntax/semi6.go:9:47: error: unexpected semicolon or newline in type declaration
syntax/semi6.go:11:62: error: unexpected semicolon or newline in type declaration

Change-Id: I9996b59a4fc78ad1935e779f354ddf75c0fb44e0
Reviewed-on: https://go-review.googlesource.com/c/go/+/274692
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agocmd/link: invalidate kernel cache on darwin
Cherry Zhang [Sun, 22 Nov 2020 16:43:29 +0000 (11:43 -0500)]
cmd/link: invalidate kernel cache on darwin

Apparently, the darwin kernel may cache the code signature at
mmap. When we mmap the output buffer, it doesn't have a code
signature (as we haven't generated one). Invalidate the kernel
cache after writing the file.

See https://github.com/golang/go/issues/42684#issuecomment-731704900
for more information.

Updates #38485.
Fixes #42684.

Change-Id: Iac2ef756ca1454c856944423e5040b8e17a6b420
Reviewed-on: https://go-review.googlesource.com/c/go/+/272258
Trust: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
4 years agocmd/internal/buildid: update Mach-O code signature when rewriting buildid
Cherry Zhang [Sun, 22 Nov 2020 02:24:57 +0000 (21:24 -0500)]
cmd/internal/buildid: update Mach-O code signature when rewriting buildid

As the code signature contains hashes of the entire file (except
the signature itself), rewriting buildid will invalidate the
signature. This CL makes it regenerate the signature when
rewriting the buildid. It only does it when the file already has
a code signature, with proper size (darwin/arm64 binaries
generated by the Go linker should have).

Updates #38485, #42684.

Change-Id: I082d9e5808b0ee6a35f9c362d7262aadd9113c81
Reviewed-on: https://go-review.googlesource.com/c/go/+/272257
Trust: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
Reviewed-by: Than McIntosh <thanm@google.com>
4 years agocmd/link: code-sign on darwin/arm64
Cherry Zhang [Sun, 22 Nov 2020 01:11:03 +0000 (20:11 -0500)]
cmd/link: code-sign on darwin/arm64

This CL lets the linker code-sign output binaries on
darwin/arm64, as the kernel requires binaries must be signed in
order to run.

This signature will likely be invalidated when we stamp the
buildid after linking. We still do it in the linker, for
- plain "go tool link" works.
- the linker generates the LC_CODE_SIGNATURE load command with
  the right size and offset, so we don't need to update it when
  stamping the buildid.

Updates #38485, #42684.

Change-Id: Ia306328906d73217221ba31093fe61a935a46122
Reviewed-on: https://go-review.googlesource.com/c/go/+/272256
Trust: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
Reviewed-by: Than McIntosh <thanm@google.com>
4 years agocmd/compile: do not assume TST and TEQ set V on arm
Jason A. Donenfeld [Mon, 30 Nov 2020 09:41:46 +0000 (10:41 +0100)]
cmd/compile: do not assume TST and TEQ set V on arm

These replacement rules assume that TST and TEQ set V. But TST and
TEQ do not set V. This is a problem because instructions like LT are
actually checking for N!=V. But with TST and TEQ not setting V, LT
doesn't do anything meaningful. It's possible to construct trivial
miscompilations from this, such as:

    package main

    var x = [4]int32{-0x7fffffff, 0x7fffffff, 2, 4}

    func main() {
        if x[0] > x[1] {
            panic("fail 1")
        }
        if x[2]&x[3] < 0 {
            panic("fail 2") // Fails here
        }
    }

That first comparison sets V, via the CMP that subtracts the values
causing the overflow. Then the second comparison operation thinks that
it uses the result of TST, when it actually uses the V from CMP.

Before this fix:

    TST             R0, R1
    BLT             loc_6C164

After this fix:

    TST             R0, R1
    BMI             loc_6C164

The BMI instruction checks the N flag, which TST sets.  This commit
fixes the issue by using [LG][TE]noov instead of vanilla [LG][TE], and
also adds a test case for the direct issue.

Fixes #42876.

Change-Id: I13c62c88d18574247ad002b671b38d2d0b0fc6fa
Reviewed-on: https://go-review.googlesource.com/c/go/+/274026
Run-TryBot: Jason A. Donenfeld <Jason@zx2c4.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Trust: Jason A. Donenfeld <Jason@zx2c4.com>

4 years agoencoding/json: revert "add "json: " prefix to SyntaxError messages"
Kevin Burke [Fri, 27 Nov 2020 21:55:27 +0000 (13:55 -0800)]
encoding/json: revert "add "json: " prefix to SyntaxError messages"

This reverts commit 6af088bfc66c13143c9ef46b4cf0805df77a8fbe.

Reason for revert: Broke many tests inside Google which implies many
tests were broken outside of Google as well. The tests may be brittle
but still would require work to change and it's not clear it's worth
the benefit.

Updates #36221
Fixes #42675

Change-Id: Id3a14eb37e7119f5abe50e80dfbf120fdc44db72
Reviewed-on: https://go-review.googlesource.com/c/go/+/273747
Run-TryBot: Joe Tsai <thebrokentoaster@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
Reviewed-by: Joe Tsai <thebrokentoaster@gmail.com>
Trust: Joe Tsai <thebrokentoaster@gmail.com>

4 years ago[dev.regabi] cmd/compile: cleanup some leftover cruft
Matthew Dempsky [Tue, 1 Dec 2020 19:37:30 +0000 (11:37 -0800)]
[dev.regabi] cmd/compile: cleanup some leftover cruft

Just clearing away some scaffolding artifacts from previous
refactorings.

[git-generate]
cd src/cmd/compile/internal/gc
rf '
ex {
  import "cmd/compile/internal/ir"
  import "cmd/compile/internal/types"

  var n *ir.Name; n.Name() -> n
  var f *ir.Func; f.Func() -> f

  var o types.Object
  ir.AsNode(o).Sym() -> o.Sym()
  ir.AsNode(o).Type() -> o.Type()
  ir.AsNode(o).(*ir.Name) -> o.(*ir.Name)
  ir.AsNode(o).(*ir.Func) -> o.(*ir.Func)

  var x ir.Node
  ir.AsNode(o) != x -> o != x
}
'

Change-Id: I946ec344bd7ee274900a392da53b95308ceaade4
Reviewed-on: https://go-review.googlesource.com/c/go/+/274592
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
4 years agocmd/internal/buildid: exclude Mach-O code signature in hash calculation
Cherry Zhang [Sat, 21 Nov 2020 22:43:16 +0000 (17:43 -0500)]
cmd/internal/buildid: exclude Mach-O code signature in hash calculation

The code signature contains hashes of the entire file (except the
signature itself), including the buildid. Therefore, the buildid
cannot depend on the signature. Otherwise updating buildid will
invalidate the signature, and vice versa. As we cannot change the
code-signing algorithm, we can only change buildid calculation.

This CL changes the buildid calculation to exclude the Mach-O
code signature. So updating code signature after stamping the
buildid will not invalidate the buildid.

Updates #38485, #42684.

Change-Id: I8a9e2e25ca9dc00d9556d13b81652f43bbf6a084
Reviewed-on: https://go-review.googlesource.com/c/go/+/272255
Trust: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
Reviewed-by: Than McIntosh <thanm@google.com>
4 years agocmd/internal/codesign: new package
Cherry Zhang [Sun, 22 Nov 2020 02:33:18 +0000 (21:33 -0500)]
cmd/internal/codesign: new package

On macOS/ARM64, the kernel requires that binaries must have a
valid code signature to run. The C toolchain code-signs the
binary at link time. We do the same.

It is more subtle for Go because we stamp the buildid after
linking. As the signature contains hashes of the entire file
(except the signature itself), we must (re)generate the signature
after stamping the buildid.

This CL adds a new codesign package, which provides
functionality to generate the code signature. It is a separate
internal package so it can be used both in the linker and by the
go command. The next CLs will add code-signing to the linker and
the go command.

Updates #38485, #42684.

Change-Id: Id46801a6665beebaab0eb413ff2e64c5b9467059
Reviewed-on: https://go-review.googlesource.com/c/go/+/272254
Trust: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
Reviewed-by: Than McIntosh <thanm@google.com>
4 years agocmd: update golang.org/x/mod to v0.4.0
Jay Conrod [Tue, 1 Dec 2020 20:10:03 +0000 (15:10 -0500)]
cmd: update golang.org/x/mod to v0.4.0

CL 269357 is the only difference between the pseudo-version we were
using and v0.4.0.

Change-Id: If15326bda51e04b47130429b818bfe2facaee03d
Reviewed-on: https://go-review.googlesource.com/c/go/+/274593
Trust: Jay Conrod <jayconrod@google.com>
Run-TryBot: Jay Conrod <jayconrod@google.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agocmd/go: don't print deprecation notice for 'go get exe'
Jay Conrod [Tue, 1 Dec 2020 18:51:17 +0000 (13:51 -0500)]
cmd/go: don't print deprecation notice for 'go get exe'

It's difficult for module authors to provide installation instructions
that work in both Go 1.15 and 1.16. We'll wait until 1.17 to print a
deprecation warning for installing executables with 'go get'.

Fixes #42885

Change-Id: I835b447e83e760f48fd664e8a117749e0cb59f83
Reviewed-on: https://go-review.googlesource.com/c/go/+/274552
Trust: Jay Conrod <jayconrod@google.com>
Run-TryBot: Bryan C. Mills <bcmills@google.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agonet/http: allow upgrading non keepalive connections
Anmol Sethi [Sat, 24 Oct 2020 00:40:41 +0000 (00:40 +0000)]
net/http: allow upgrading non keepalive connections

If one was using http.Transport with DisableKeepAlives and trying
to upgrade a connection against net/http's Server, the Server
would not allow a "Connection: Upgrade" header to be written
and instead override it to "Connection: Close" which would
break the handshake.

This change ensures net/http's Server does not override the
connection header for successful protocol switch responses.

Fixes #36381.

Change-Id: I882aad8539e6c87ff5f37c20e20b3a7fa1a30357
GitHub-Last-Rev: dc0de83201dc26236527b68bd49dffc53dd0389b
GitHub-Pull-Request: golang/go#36382
Reviewed-on: https://go-review.googlesource.com/c/go/+/213277
Trust: Emmanuel Odeke <emmanuel@orijtech.com>
Trust: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
4 years agonet/http: ignore connection closes once done with the connection
Michael Fraenkel [Sat, 26 Sep 2020 15:20:16 +0000 (09:20 -0600)]
net/http: ignore connection closes once done with the connection

Once the connection is put back into the idle pool, the request should
not take any action if the connection is closed.

Fixes #41600

Change-Id: I5e4ddcdc03cd44f5197ecfbe324638604961de84
Reviewed-on: https://go-review.googlesource.com/c/go/+/257818
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Trust: Damien Neil <dneil@google.com>

4 years agodoc/go1.16: add runtime/debug changes to release notes
Alberto Donizetti [Tue, 1 Dec 2020 13:03:54 +0000 (14:03 +0100)]
doc/go1.16: add runtime/debug changes to release notes

For #40700
Fixes #42912

Change-Id: Ifd36950136db1fc93a8de76a2717a473210418b1
Reviewed-on: https://go-review.googlesource.com/c/go/+/274473
Trust: Alberto Donizetti <alb.donizetti@gmail.com>
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
4 years agodoc/go1.16: add text/template changes to release notes
Alberto Donizetti [Tue, 1 Dec 2020 12:54:53 +0000 (13:54 +0100)]
doc/go1.16: add text/template changes to release notes

For #40700
Fixes #42914

Change-Id: I673d86a946c362e28bfbf35fab2c60ebfbd8bda2
Reviewed-on: https://go-review.googlesource.com/c/go/+/274472
Trust: Alberto Donizetti <alb.donizetti@gmail.com>
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
4 years ago[dev.regabi] cmd/compile: first pass at abstracting Type
Matthew Dempsky [Tue, 1 Dec 2020 11:25:29 +0000 (03:25 -0800)]
[dev.regabi] cmd/compile: first pass at abstracting Type

Passes toolstash/buildall.

[git-generate]
cd src/cmd/compile/internal/ssa
rf '
ex . ../ir ../gc {
  import "cmd/compile/internal/types"
  var t *types.Type
  t.Etype -> t.Kind()
  t.Sym -> t.GetSym()
  t.Orig -> t.Underlying()
}
'

cd ../types
rf '
mv EType Kind
mv IRNode Object

mv Type.Etype Type.kind
mv Type.Sym Type.sym
mv Type.Orig Type.underlying
mv Type.Cache Type.cache

mv Type.GetSym Type.Sym

mv Bytetype ByteType
mv Runetype RuneType
mv Errortype ErrorType
'

cd ../gc
sed -i 's/Bytetype/ByteType/; s/Runetype/RuneType/' mkbuiltin.go

git codereview gofmt
go install cmd/compile/internal/...
go test cmd/compile -u || go test cmd/compile

Change-Id: Ibecb2d7100d3318a49238eb4a78d70acb49eedca
Reviewed-on: https://go-review.googlesource.com/c/go/+/274437
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Russ Cox <rsc@golang.org>
Trust: Matthew Dempsky <mdempsky@google.com>

4 years agodoc/go1.16: add multiple release notes for the go command
Jay Conrod [Tue, 1 Dec 2020 16:10:40 +0000 (11:10 -0500)]
doc/go1.16: add multiple release notes for the go command

Added notes for:

* go test -c and -i flags used with unknown flags
* GO111MODULE=on by default
* GOVCS
* Dropped requirements on excluded versions

Removed TODOs for documentation on the retract directive and
'go install pkg@version'. These pages will be written after the beta.

Change-Id: Ic9877a62f908be177a6035a039b72e969e7b7f22
Reviewed-on: https://go-review.googlesource.com/c/go/+/274438
Trust: Jay Conrod <jayconrod@google.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years ago[dev.regabi] cmd/compile: only save ONAMEs on Curfn.Dcl
Matthew Dempsky [Tue, 1 Dec 2020 11:52:20 +0000 (03:52 -0800)]
[dev.regabi] cmd/compile: only save ONAMEs on Curfn.Dcl

There's not really any use to tracking function-scoped constants and
types on Curfn.Dcl, and there's sloppy code that assumes all of the
declarations are variables (e.g., cmpstackvarlt).

Change-Id: I5d10dc681dac2c161c7b73ba808403052ca0608e
Reviewed-on: https://go-review.googlesource.com/c/go/+/274436
Reviewed-by: Russ Cox <rsc@golang.org>
Trust: Matthew Dempsky <mdempsky@google.com>

4 years ago[dev.regabi] cmd/compile: add NewBasic and cleanup universe
Matthew Dempsky [Tue, 1 Dec 2020 10:58:41 +0000 (02:58 -0800)]
[dev.regabi] cmd/compile: add NewBasic and cleanup universe

This CL introduces types.NewBasic, for creating the predeclared
universal types, and reorganizes how the universe is initialized so
that all predeclared types are uniformly constructed.

There are now a bunch of Type fields that are no longer assigned
outside of the package, so this CL also introduces some new accessor
methods that a subsequent CL will mechanically introduce uses of.

Change-Id: Ie7996c3d5f1ca46cd5bfe45ecc91ebfa6a7b6c7d
Reviewed-on: https://go-review.googlesource.com/c/go/+/274435
Trust: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Russ Cox <rsc@golang.org>
4 years ago[dev.regabi] cmd/compile: add NewNamed
Matthew Dempsky [Tue, 1 Dec 2020 09:42:47 +0000 (01:42 -0800)]
[dev.regabi] cmd/compile: add NewNamed

The start of abstracting away Type fields. This adds a new constructor
for named types, styled after go/types.NewNamed. Along with helper
methods for SetNod and Pos, this allows hiding Nod.

Change-Id: Ica107034b6346c7b523bf6ae2a34009e350a9aa8
Reviewed-on: https://go-review.googlesource.com/c/go/+/274434
Trust: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Russ Cox <rsc@golang.org>
4 years ago[dev.regabi] cmd/compile: move setUnderlying to package types
Matthew Dempsky [Tue, 1 Dec 2020 09:31:29 +0000 (01:31 -0800)]
[dev.regabi] cmd/compile: move setUnderlying to package types

Now that setUnderlying is decoupled from Nodes, it can be moved into
package types, where it really belongs.

[git-generate]
cd src/cmd/compile/internal/gc
rf '
mv setUnderlying SetUnderlying
mv SetUnderlying typex.go
mv typex.go cmd/compile/internal/types
'

cd ../types
rf '
mv typex.go type.go
'

Change-Id: I76e2d4d8a6df599f24a731c4d8e5774ec83a119c
Reviewed-on: https://go-review.googlesource.com/c/go/+/274433
Trust: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Russ Cox <rsc@golang.org>
4 years ago[dev.regabi] cmd/compile: cleanup type-checking of defined types
Matthew Dempsky [Tue, 1 Dec 2020 09:01:59 +0000 (01:01 -0800)]
[dev.regabi] cmd/compile: cleanup type-checking of defined types

The code for type-checking defined types was scattered between
typecheckdef, typecheckdeftype, and setUnderlying. There was redundant
work between them, and setUnderlying also needed to redo a lot of work
because of its brute-force solution of just copying all Type fields.

This CL reorders things so as many of the defined type's fields are
set in advance (in typecheckdeftype), and then setUnderlying only
copies over the details actually needed from the underlying type.

Incidentally, this evidently improves our error handling for an
existing test case, by allowing us to report an additional error.

Passes toolstash/buildall.

Change-Id: Id59a24341e7e960edd1f7366c3e2356da91b9fe7
Reviewed-on: https://go-review.googlesource.com/c/go/+/274432
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
4 years ago[dev.regabi] cmd/compile: process //go:linknames after declarations
Matthew Dempsky [Sun, 29 Nov 2020 22:06:17 +0000 (14:06 -0800)]
[dev.regabi] cmd/compile: process //go:linknames after declarations

Allows emitting errors about ineffectual //go:linkname directives.

In particular, this exposed: a typo in os2_aix.go; redundant (but
harmless) directives for libc_pipe in both os3_solaris.go and
syscall2_solaris.go; and a bunch of useless //go:linkname directives
in macOS wrapper code.

However, because there's also ineffectual directives in the vendored
macOS code from x/sys, that can't be an error just yet. So instead we
print a warning (including a heads up that it will be promoted to an
error in Go 1.17) to prevent backsliding while we fix and re-vendor
that code.

Passes toolstash-check.

Change-Id: I59badeab5df0d8b3abfd14c6066e9bb00e840f73
Reviewed-on: https://go-review.googlesource.com/c/go/+/273986
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
Trust: Matthew Dempsky <mdempsky@google.com>

4 years ago[dev.regabi] cmd/compile: unexport Node.RawCopy
Russ Cox [Tue, 1 Dec 2020 02:20:45 +0000 (21:20 -0500)]
[dev.regabi] cmd/compile: unexport Node.RawCopy

RawCopy breaks the invariant that ir.Orig depends on for
allowing nodes to omit keeping their own orig fields.
Avoid surprises by unexporting it.

The only use in package gc was removed in the previous CL.
This one is a straight global search and replace RawCopy -> rawCopy.

Change-Id: Ia99c0f4665bf7ed4f878cc44456d5fbdf33bab8d
Reviewed-on: https://go-review.googlesource.com/c/go/+/274293
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years ago[dev.regabi] cmd/compile: use ir.Copy instead of direct use of RawCopy
Russ Cox [Tue, 1 Dec 2020 02:18:48 +0000 (21:18 -0500)]
[dev.regabi] cmd/compile: use ir.Copy instead of direct use of RawCopy

The ONIL export bug happened because the logic about
maintaining an “implicit” orig pointer in the comments
around ir.Orig only applies to Copy and SepCopy, not to
direct use of RawCopy. I'd forgotten those could exist.

The sole direct use of RawCopy was for the OLITERAL/ONIL case.
The ONIL is now provably indistinguishable from Copy, since
NilExpr does not have an explicit Orig field, so for NilExpr
RawCopy and Copy are the same.
The OLITERAL is not, but we can reconstruct the effect
with Copy+SetOrig to be explicit that we need the orig link.

The next CL will unexport RawCopy.

Also fix a typo in MapType doc comment.

Passes buildall w/ toolstash -cmp.

Change-Id: I876a85ff188e6d1cd4c0dfa385be32482e0de0d4
Reviewed-on: https://go-review.googlesource.com/c/go/+/274292
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years ago[dev.regabi] cmd/compile: improve findTypeLoop
Matthew Dempsky [Tue, 1 Dec 2020 04:34:25 +0000 (20:34 -0800)]
[dev.regabi] cmd/compile: improve findTypeLoop

When checking if a defined type is part of a type loop, we can
short-circuit if it was defined in another package. We can assume any
package we import already successfully compiled, so any types it
contains cannot be part of a type loop.

This also allows us to simplify the logic for recursing into the type
used in the type declaration, because any defined type from this
package will have a properly setup node.

Change-Id: Ic024814d95533afd9e59f2103c8ddb22bd87e900
Reviewed-on: https://go-review.googlesource.com/c/go/+/274294
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agocmd/internal/obj/riscv: add tests for BGE/BGEU/BLT/BLTU
Joel Sing [Sun, 22 Nov 2020 00:52:55 +0000 (11:52 +1100)]
cmd/internal/obj/riscv: add tests for BGE/BGEU/BLT/BLTU

Add tests for BGE/BGEU/BLT/BLTU branch instructions. Also add pure Go variants
of these to ensure that the test data, Go and assembly all match up.

Change-Id: I84c68605e116a4e57f6c5c765bf0aaecab84b675
Reviewed-on: https://go-review.googlesource.com/c/go/+/271913
Trust: Joel Sing <joel@sing.id.au>
Reviewed-by: Quey-Liang Kao <s101062801@m101.nthu.edu.tw>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agocmd/link/internal/amd64: always generate R_X86_64_PLT32 for SDYNIMPORT calls
Joel Sing [Thu, 26 Nov 2020 15:06:08 +0000 (02:06 +1100)]
cmd/link/internal/amd64: always generate R_X86_64_PLT32 for SDYNIMPORT calls

Currently, in the non-DynlinkingGo case with external linking, we generate a
R_X86_64_GOTPCREL relocation for the imported symbol. This results in the
external linker turning this into a R_X86_64_GLOB_DAT relocation, rather
than a R_X86_64_JUMP_SLOT. Always generate R_X86_64_PLT32 for SDYNIMPORT
calls so that these calls work correctly.

Update #36435
Fixes #42671

Change-Id: I8a28884b7853cb4135053ed817bedc919482f4ad
Reviewed-on: https://go-review.googlesource.com/c/go/+/270377
Trust: Joel Sing <joel@sing.id.au>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
4 years agodoc/go1.16: add crypto/x509 note about Verify on Windows
Roland Shoemaker [Tue, 1 Dec 2020 00:32:41 +0000 (16:32 -0800)]
doc/go1.16: add crypto/x509 note about Verify on Windows

Updates #42897

Change-Id: Ice25922475405aca3cf2cb1c163462f223ede736
Reviewed-on: https://go-review.googlesource.com/c/go/+/274239
Trust: Roland Shoemaker <roland@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
4 years ago[dev.regabi] cmd/compile: add OSTMTEXPR Op
Russ Cox [Mon, 30 Nov 2020 04:06:02 +0000 (23:06 -0500)]
[dev.regabi] cmd/compile: add OSTMTEXPR Op

This CL only adds the new constant, which is not safe for toolstash -cmp.

Change-Id: I774463a0ab5f57113d67a8888b6ac787be68510c
Reviewed-on: https://go-review.googlesource.com/c/go/+/274110
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: add custom expression Node implementations
Russ Cox [Mon, 30 Nov 2020 02:25:47 +0000 (21:25 -0500)]
[dev.regabi] cmd/compile: add custom expression Node implementations

These are fairly rote implementations of structs appropriate to
each Op (or group of Ops).

The names of these are unknown except to ir.NodAt for now.
A later, automated change will introduce direct use of the types
throughout package gc.

(This CL is expressions; the previous one was statements.)

This is the last of the Ops that were previously handled by the
generic node struct, so that struct and its methods can be
and are deleted in this CL.

Passes buildall w/ toolstash -cmp.

Change-Id: I1703f35f24dcd3f7c5782a278e53c3fe04e87c37
Reviewed-on: https://go-review.googlesource.com/c/go/+/274109
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years agocmd/compile: do not mark OpSP, OpSB pos for debugging
Cuong Manh Le [Thu, 26 Nov 2020 16:28:11 +0000 (23:28 +0700)]
cmd/compile: do not mark OpSP, OpSB pos for debugging

Fixes #42801

Change-Id: I2080ecacc109479f5820035401ce2b26d72e2ef2
Reviewed-on: https://go-review.googlesource.com/c/go/+/273506
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Reviewed-by: David Chase <drchase@google.com>
4 years ago[dev.regabi] cmd/compile: remove types.InitSyms
Matthew Dempsky [Mon, 30 Nov 2020 08:01:26 +0000 (00:01 -0800)]
[dev.regabi] cmd/compile: remove types.InitSyms

It's not types's responsibility to understand how package
initialization is implemented. Instead, have gc keep track of the
order that packages were imported, and then look for inittask
declarations.

Also, use resolve to force importing of the inittask's export data, so
that we can get the appropriate linker symbol index. (This is also why
this CL doesn't satisfy "toolstash -cmp".)

Change-Id: I5b706497d4a8d1c4439178863b4a8dba4da0f5a9
Reviewed-on: https://go-review.googlesource.com/c/go/+/274006
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
4 years ago[dev.regabi] cmd/compile: fix -h
Russ Cox [Mon, 30 Nov 2020 14:34:34 +0000 (09:34 -0500)]
[dev.regabi] cmd/compile: fix -h

The compile -h flag is *meant* to panic, so you can see the stack
trace where the error is being printed. Make it do that again.

Change-Id: Ieb0042863582d7a4c5d08d2f866a144962915b06
Reviewed-on: https://go-review.googlesource.com/c/go/+/274116
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: add custom statement Node implementations
Russ Cox [Mon, 30 Nov 2020 02:23:47 +0000 (21:23 -0500)]
[dev.regabi] cmd/compile: add custom statement Node implementations

These are fairly rote implementations of structs appropriate to
each Op (or group of Ops).

The names of these are unknown except to ir.NodAt for now.
A later, automated change will introduce direct use of the types
throughout package gc.

Passes buildall w/ toolstash -cmp.

Change-Id: Ie9835fcd2b214fda5b2149e187af369d76534487
Reviewed-on: https://go-review.googlesource.com/c/go/+/274108
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: clean up ONEW node
Russ Cox [Mon, 30 Nov 2020 19:16:39 +0000 (14:16 -0500)]
[dev.regabi] cmd/compile: clean up ONEW node

The list is no longer needed and can be deleted.
Doing so reduces the inlining cost of any function containing
an explicit call to new by 1 point, so this change is not
toolstash -cmp safe.

Change-Id: Id29e115d68e466a353708ab4b8c1021e9c85a628
Reviewed-on: https://go-review.googlesource.com/c/go/+/274132
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: clean up in preparation for expression Nodes
Russ Cox [Mon, 30 Nov 2020 02:11:07 +0000 (21:11 -0500)]
[dev.regabi] cmd/compile: clean up in preparation for expression Nodes

Using expression nodes restricts the set of valid SetOp operations,
because you can't SetOp across representation. Rewrite various code
to avoid crossing those as-yet-unintroduced boundaries.

This also includes choosing a single representation for any given Op.
For example, OCLOSE starts out as an OCALL, so it starts with a List
of one node and then moves that node to Left. That's no good with
real data structures, so the code picks a single canonical implementation
and prepares it during the conversion from one Op to the next.
In this case, the conversion of an OCALL to an OCLOSE now creates
a new node with Left initialized from the start. This pattern repeats.

Passes buildall w/ toolstash -cmp.

Change-Id: I55a0872c614d883cac9d64976c46aeeaa639e25d
Reviewed-on: https://go-review.googlesource.com/c/go/+/274107
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: clean up Order.copyExpr TODO
Russ Cox [Sun, 29 Nov 2020 16:16:20 +0000 (11:16 -0500)]
[dev.regabi] cmd/compile: clean up Order.copyExpr TODO

Just a little cleaner to read.

Passes buildall w/ toolstash -cmp.

Change-Id: I27b9f09bf6756f74f1c01794444518ded1a7d625
Reviewed-on: https://go-review.googlesource.com/c/go/+/274106
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years ago[dev.regabi] cmd/compile: clean up in preparation for statement Nodes
Russ Cox [Sat, 28 Nov 2020 20:28:18 +0000 (15:28 -0500)]
[dev.regabi] cmd/compile: clean up in preparation for statement Nodes

Using statement nodes restricts the set of valid SetOp operations,
because you can't SetOp across representation. Rewrite various
code to avoid crossing those as-yet-unintroduced boundaries.

In particular, code like

        x, y := v.(T)
        x, y := f()
        x, y := m[k]
        x, y := <-c

starts out with Op = OAS2, and then it turns into a specific Op
OAS2DOTTYPE, OAS2FUNC, OAS2MAPR, OAS2RECV, and then
later in walk is lowered to an OAS2 again.

In the middle, the specific forms move the right-hand side from
n.Rlist().First() to n.Right(), and then the conversion to OAS2 moves
it back. This is unnecessary and makes it hard for these all to
share an underlying Node implementation.

This CL changes these specific forms to leave the right-hand side
in n.Rlist().First().

Similarly, OSELRECV2 is really just a temporary form of OAS2.
This CL changes it to use same fields too.

Finally, this CL fixes the printing of OAS2 nodes in ir/fmt.go,
which formerly printed n.Right() instead of n.Rlist().
This results in a (correct!) update to cmd/compile/internal/logopt's
expected output: ~R0 = <N> becomes ~R0 = &y.b.

Passes buildall w/ toolstash -cmp.

Change-Id: I164aa2e17dc55bfb292024de53d7d250192ad64a
Reviewed-on: https://go-review.googlesource.com/c/go/+/274105
Trust: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
4 years agoruntime: mlock signal stack on macOS/ARM64
Cherry Zhang [Fri, 27 Nov 2020 05:42:41 +0000 (00:42 -0500)]
runtime: mlock signal stack on macOS/ARM64

Apparently, the macOS ARM64 kernel has a bug where when a signal
arrives and the signal stack is not currently faulted in, it may
kill the program with a SIGILL. Work around it by mlock the
signal stacks.

Fixes #42774.

Change-Id: I99a4b3fdb6d8af1c945725ddc2c25568d81c510a
Reviewed-on: https://go-review.googlesource.com/c/go/+/273686
Trust: Cherry Zhang <cherryyz@google.com>
Reviewed-by: Austin Clements <austin@google.com>
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years ago[dev.regabi] cmd/compile: simplify export data representation of nil
Matthew Dempsky [Mon, 30 Nov 2020 21:50:05 +0000 (13:50 -0800)]
[dev.regabi] cmd/compile: simplify export data representation of nil

The handling of ONIL and Orig has been a mess for a while, and dates
back to how fmt.go used to print out typed nils. That hasn't applied
for a while, but we've kept dragging it along to appease toolstash
with the intention of someday finally removing it.

Today is that day.

Change-Id: I9a441628e53068ab1993cd2b67b977574d8117b7
Reviewed-on: https://go-review.googlesource.com/c/go/+/274212
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Russ Cox <rsc@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agocmd/go: fix infinite loop in modload.keepSums
Jay Conrod [Mon, 30 Nov 2020 20:46:33 +0000 (15:46 -0500)]
cmd/go: fix infinite loop in modload.keepSums

Fixes #42891

Change-Id: I0cce4204a1c4959b896188a2ab3719c0507f95e6
Reviewed-on: https://go-review.googlesource.com/c/go/+/274172
Run-TryBot: Jay Conrod <jayconrod@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Michael Matloob <matloob@golang.org>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Trust: Jay Conrod <jayconrod@google.com>

4 years agonet: add note about disabling loopback in ListenMulticastUDP()
Andy Pan [Sat, 21 Nov 2020 06:48:26 +0000 (14:48 +0800)]
net: add note about disabling loopback in ListenMulticastUDP()

Fixes #41752

Change-Id: I83520d2303e5fd2e5f6329f092b40e73c13771a1
Reviewed-on: https://go-review.googlesource.com/c/go/+/271908
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: Dmitri Shuralyov <dmitshur@golang.org>

4 years agocmd/go: remove trailing whitespace from test script
Daniel Martí [Mon, 23 Nov 2020 05:47:51 +0000 (05:47 +0000)]
cmd/go: remove trailing whitespace from test script

Noticed while skimming through recent master commits.

Change-Id: I42a99ea7d71c05fc5b6107627105375a21920f5e
Reviewed-on: https://go-review.googlesource.com/c/go/+/271990
Trust: Daniel Martí <mvdan@mvdan.cc>
Reviewed-by: Jay Conrod <jayconrod@google.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>

4 years agotest: update gofrontend expected errors
Ian Lance Taylor [Sun, 29 Nov 2020 03:10:57 +0000 (19:10 -0800)]
test: update gofrontend expected errors

This matches the error messages after CL 273890.

syntax/semi4.go:11:9: error: unexpected semicolon or newline, expecting ‘{’ after for clause
syntax/semi4.go:10:13: error: reference to undefined name ‘x’
syntax/semi4.go:12:17: error: reference to undefined name ‘z’

Change-Id: Ic88ff6e27d50bf70f5b2114383b84c42c0682f39
Reviewed-on: https://go-review.googlesource.com/c/go/+/273891
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>