A composite literal assignment
x = T{field: v}
may be compiled to
x = T{}
x.field = v
We already do not use this form is RHS uses LHS. If LHS is
address-taken, RHS may uses LHS implicitly, e.g.
v = &x.field
x = T{field: *v}
The lowering above would change the value of RHS (*v).
Fixes#52953.
Change-Id: I3f798e00598aaa550b8c17182c7472fef440d483
Reviewed-on: https://go-review.googlesource.com/c/go/+/407014
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Run-TryBot: Cherry Mui <cherryyz@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
Change-Id: I63eb42f3ce5ca452279120a5b33518f4ce16be45
GitHub-Last-Rev: a88f2f72be
GitHub-Pull-Request: golang/go#52951
Reviewed-on: https://go-review.googlesource.com/c/go/+/406843
Run-TryBot: Robert Griesemer <gri@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Ian Lance Taylor <iant@google.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Reviewed-by: Robert Griesemer <gri@google.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
For #51291
Change-Id: If47e4cbf899853ade5050852c3870b9500da4c63
Reviewed-on: https://go-review.googlesource.com/c/go/+/406916
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Cherry Mui <cherryyz@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Than McIntosh <thanm@google.com>
CL 406358 added test that use -buildmode=plugin. But plugin mode only
supports on some os/arch pairs, so this CL moving the test to
misc/cgo/testplugin directory instead.
Updates #52937
Change-Id: Iad049443c1f6539f6af1988bebd4dff56c6e1bf9
Reviewed-on: https://go-review.googlesource.com/c/go/+/406774
Reviewed-by: Michael Knyszek <mknyszek@google.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Cherry Mui <cherryyz@google.com>
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
CL 395854 made inline pass to not inlining function with shape params,
but pass no shape arguments. This is intended to be the reverse case of
CL 361260.
However, CL 361260 is using wider condition than necessary. Though it
only needs to check against function parameters, it checks whether the
function type has no shape. It does not cause any issue, because
!fn.Type().HasShape() implies !fn.Type().Params().HasShape().
But for the reverse case, it's not true. Function may have shape type,
but has no shape arguments. Thus, we must tighten the condition to
explicitly check against the function parameters only.
Fixes#52907
Change-Id: Ib87e87ff767c31d99d5b36aa4a6c1d8baf32746d
Reviewed-on: https://go-review.googlesource.com/c/go/+/406475
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Generic functions require instantiation, which package plugin doesn't
support, and likely never will. So instead, we can just skip writing
out any generic functions, which avoids an ICE in the plugin
generation code.
This issue doesn't affect GOEXPERIMENT=unified, because it avoids
leaking any non-instantiated types/functions to the rest of the
compiler backend.
Fixes#52937.
Change-Id: Ie35529c5c241e46b77fcb5b8cca48bb99ce7bfcb
Reviewed-on: https://go-review.googlesource.com/c/go/+/406358
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
This test is currently overly sensitive to compiler optimizations,
because inlining can affect the order in which cmd/link emits field
references. The order doesn't actually matter though, so this CL just
tweaks the test to sort the tracked fields before printing them.
Updates #51734.
Change-Id: I3b65ca265856b2e1102f40406d5ce34610c70d40
Reviewed-on: https://go-review.googlesource.com/c/go/+/406674
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: David Chase <drchase@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Since CL 391014, cmd/compile now requires the -p flag to be set the
build system. This CL changes it to initialize LocalPkg.Path to the
provided path, rather than relying on writing out `"".` into object
files and expecting cmd/link to substitute them.
However, this actually involved a rather long tail of fixes. Many have
already been submitted, but a few notable ones that have to land
simultaneously with changing LocalPkg:
1. When compiling package runtime, there are really two "runtime"
packages: types.LocalPkg (the source package itself) and
ir.Pkgs.Runtime (the compiler's internal representation, for synthetic
references). Previously, these ended up creating separate link
symbols (`"".xxx` and `runtime.xxx`, respectively), but now they both
end up as `runtime.xxx`, which causes lsym collisions (notably
inittask and funcsyms).
2. test/codegen tests need to be updated to expect symbols to be named
`command-line-arguments.xxx` rather than `"".foo`.
3. The issue20014 test case is sensitive to the sort order of field
tracking symbols. In particular, the local package now sorts to its
natural place in the list, rather than to the front.
Thanks to David Chase for helping track down all of the fixes needed
for this CL.
Updates #51734.
Change-Id: Iba3041cf7ad967d18c6e17922fa06ba11798b565
Reviewed-on: https://go-review.googlesource.com/c/go/+/393715
Reviewed-by: David Chase <drchase@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
CL 395854 made inline pass to not inlining function with shape params,
but pass no shape arguments. But it does not consider the case where
function has shape params, but passing zero arguments. In this case, the
un-safe interface conversion that may be applied to a shape argument can
not happen, so it's safe to inline the function.
Fixes#52907
Change-Id: Ifa7b23709bb47b97e27dc1bf32343d92683ef783
Reviewed-on: https://go-review.googlesource.com/c/go/+/406176
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: David Chase <drchase@google.com>
This is the same fix as CL 36126, but for the reverse case, function
with shape params but passed no shape arg. The same conversion problem
may occur in this case, see details explanation there.
Fixes#51909Fixes#51925
Change-Id: Ib0c1973c7511d85b4918a252c80060f1864180cf
Reviewed-on: https://go-review.googlesource.com/c/go/+/395854
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: David Chase <drchase@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
For #52841
Change-Id: If4723a70fba0dbedb5d1e70dab58f0b4612bf8b9
Reviewed-on: https://go-review.googlesource.com/c/go/+/405759
Auto-Submit: Ian Lance Taylor <iant@google.com>
Run-TryBot: Ian Lance Taylor <iant@google.com>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Reviewed-by: Cherry Mui <cherryyz@google.com>
CL 391014 requires the compiler to be invoked with the -p flag, to
specify the package path. Later, CL 394217 makes the compiler to
produce an unlinkable object file, so "go tool compile x.go" can
still be used on the command line. This CL does the same for the
assembler, requiring -p, otherwise generating an unlinkable object.
No special case for the main package, as the main package cannot
be only assembly code, and there is no way to tell if it is the
main package from an assembly file.
Now we guarantee that we always have an expanded package path in
the object file. A later CL will delete the name expansion code
in the linker.
Change-Id: I8c10661aaea2ff794614924ead958d80e7e2487d
Reviewed-on: https://go-review.googlesource.com/c/go/+/404298
Run-TryBot: Cherry Mui <cherryyz@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Than McIntosh <thanm@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
The compiler use to compile f()(g()) as:
t1, t2 := g()
f()(t1, t2)
That violates the Go spec, since when "..., all function calls, ... are
evaluated in lexical left-to-right order"
This PR fixes the bug by compiling f()(g()) as:
t0 := f()
t1, t2 := g()
t0(t1, t2)
to make "f()" to be evaluated before "g()".
Fixes#50672
Change-Id: I6a766f3dfc7347d10f8fa3a151f6a5ea79bcf818
Reviewed-on: https://go-review.googlesource.com/c/go/+/392834
TryBot-Result: Gopher Robot <gobot@golang.org>
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Keith Randall <khr@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
math/bits.Add64 and math/bits.Sub64 now lower and optimize
directly in SSA form.
The optimization of carry chains focuses around eliding
XER<->GPR transfers of the CA bit when used exclusively as an
input to a single carry operations, or when the CA value is
known.
This also adds support for handling XER spills in the assembler
which could happen if carry chains contain inter-dependencies
on each other (which seems very unlikely with practical usage),
or a clobber happens (SRAW/SRAD/SUBFC operations clobber CA).
With PPC64 Add64/Sub64 lowering into SSA and this patch, the net
performance difference in crypto/elliptic benchmarks on P9/ppc64le
are:
name old time/op new time/op delta
ScalarBaseMult/P256 46.3µs ± 0% 46.9µs ± 0% +1.34%
ScalarBaseMult/P224 356µs ± 0% 209µs ± 0% -41.14%
ScalarBaseMult/P384 1.20ms ± 0% 0.57ms ± 0% -52.14%
ScalarBaseMult/P521 3.38ms ± 0% 1.44ms ± 0% -57.27%
ScalarMult/P256 199µs ± 0% 199µs ± 0% -0.17%
ScalarMult/P224 357µs ± 0% 212µs ± 0% -40.56%
ScalarMult/P384 1.20ms ± 0% 0.58ms ± 0% -51.86%
ScalarMult/P521 3.37ms ± 0% 1.44ms ± 0% -57.32%
MarshalUnmarshal/P256/Uncompressed 2.59µs ± 0% 2.52µs ± 0% -2.63%
MarshalUnmarshal/P256/Compressed 2.58µs ± 0% 2.52µs ± 0% -2.06%
MarshalUnmarshal/P224/Uncompressed 1.54µs ± 0% 1.40µs ± 0% -9.42%
MarshalUnmarshal/P224/Compressed 1.54µs ± 0% 1.39µs ± 0% -9.87%
MarshalUnmarshal/P384/Uncompressed 2.40µs ± 0% 1.80µs ± 0% -24.93%
MarshalUnmarshal/P384/Compressed 2.35µs ± 0% 1.81µs ± 0% -23.03%
MarshalUnmarshal/P521/Uncompressed 3.79µs ± 0% 2.58µs ± 0% -31.81%
MarshalUnmarshal/P521/Compressed 3.80µs ± 0% 2.60µs ± 0% -31.67%
Note, P256 uses an asm implementation, thus, little variation is expected.
Change-Id: I88a24f6bf0f4f285c649e40243b1ab69cc452b71
Reviewed-on: https://go-review.googlesource.com/c/go/+/346870
Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Run-TryBot: Paul Murphy <murp@ibm.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@google.com>
This issue has been fixed in CL 403837.
Fixes#51840.
Change-Id: I282062bb06278696fe25e9ede333c64539dc964e
Reviewed-on: https://go-review.googlesource.com/c/go/+/404914
Run-TryBot: Wayne Zuo <wdvxdr@golangcn.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Auto-Submit: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: David Chase <drchase@google.com>
When a fully instantiated generic method is exported, be sure to also
export the types in its signature.
Fixes#52279.
Change-Id: Icc6bca05b01f914cf67faaf1bf184eaa5484f521
Reviewed-on: https://go-review.googlesource.com/c/go/+/405118
Run-TryBot: David Chase <drchase@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Following CL 405114, the extension rule is also wrong. It is safe
to drop the extension if the value is from a boolean-generating
instruction, but not a boolean-typed Value in general (e.g. a Phi
or a in-register parameter). Fix it.
Updates #52788.
Change-Id: Icf3028fe8e90806f9f57fbe2b38d47da27a97e2a
Reviewed-on: https://go-review.googlesource.com/c/go/+/405115
Reviewed-by: David Chase <drchase@google.com>
Run-TryBot: Cherry Mui <cherryyz@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
On ARM64, an If block is lowered to (NZ cond yes no). This is
incorrect because cond is a boolean value and therefore only the
last byte is meaningful (same as AMD64, see ARM64Ops.go). But here
we are comparing a full register width with 0. Correct it by
comparing only the last bit.
Fixes#52788.
Change-Id: I2cacf9f3d2f45e149c361a290f511b2d4ed845c4
Reviewed-on: https://go-review.googlesource.com/c/go/+/405114
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: David Chase <drchase@google.com>
Run-TryBot: Cherry Mui <cherryyz@google.com>
For this code:
z &= 63
_ = x<<z | x>>(64-z)
Now can prove 'x<<z' in bound. In ppc64 lowering pass, it will not
produce an extra '(ANDconst <typ.Int64> [63] z)' causing
codegen/rotate.go failed. Just remove the type check in rewrite rules
as the workaround.
Removes 32 bounds checks during make.bat.
Fixes#52563.
Change-Id: I14ed2c093ff5638dfea7de9bc7649c0f756dd71a
Reviewed-on: https://go-review.googlesource.com/c/go/+/404315
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: Keith Randall <khr@google.com>
Reviewed-by: David Chase <drchase@google.com>
Auto-Submit: Keith Randall <khr@golang.org>
Run-TryBot: Wayne Zuo <wdvxdr@golangcn.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
An invalid program may produce invalid types. If the program
calls unsafe.Sizeof on such a type, which is a compile-time
computation, the size-computation must be able to handle it.
Add the invalid type to the list of permissible basic types
and give it a size of 1 (word).
Fixes#52748.
Change-Id: I6c409628f9b77044758caf71cdcb199f9e77adea
Reviewed-on: https://go-review.googlesource.com/c/go/+/404894
Run-TryBot: Robert Griesemer <gri@google.com>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Robert Griesemer <gri@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: David Chase <drchase@google.com>
Noder pass will build a closure to implement generic function
instantiation which may produce `.dict` and `.rcvr` ident.
Since we allow `.dict` during exporting, we should allow `.rcvr` too.
Fixes#52241.
Change-Id: Ifc3912ba5155b5ac1887f20830da64f4fb3fceb6
Reviewed-on: https://go-review.googlesource.com/c/go/+/404314
Run-TryBot: Wayne Zuo <wdvxdr@golangcn.org>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: David Chase <drchase@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
For #52535
Change-Id: I6798a8379163497ebebcdadf836b8569735c282b
Reviewed-on: https://go-review.googlesource.com/c/go/+/404496
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Cherry Mui <cherryyz@google.com>
Reviewed-by: Than McIntosh <thanm@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Auto-Submit: Ian Lance Taylor <iant@google.com>
In walkCompare, any ir.OCONVNOP was removed from both operands. So when
constructing assignments for them to preserve any side-effects, using
temporary variables can cause type mismatched with original type.
Instead, using blank assignments will prevent that issue and still make
sure that the operands will be evaluated.
Fixes#52701
Change-Id: I229046acb154890bb36fe441d258563687fdce37
Reviewed-on: https://go-review.googlesource.com/c/go/+/403997
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Robert Griesemer <gri@google.com>
It is hit ~70k times building go.
This make the go binary, 0.04% smaller.
I didn't included benchmarks because this is just constant foldings
and is hard to mesure objectively.
For example, this enable rewriting things like:
if x == 20 {
return x + 30 + z
}
Into:
if x == 20 {
return 50 + z
}
It's not just fixing programer's code,
the ssa generator generate code like this sometimes.
Change-Id: I0861f342b27f7227b5f1c34d8267fa0057b1bbbc
GitHub-Last-Rev: 4c2f9b5216
GitHub-Pull-Request: golang/go#52669
Reviewed-on: https://go-review.googlesource.com/c/go/+/403735
Reviewed-by: Keith Randall <khr@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Keith Randall <khr@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: David Chase <drchase@google.com>
This shows up in a few crypto functions, and other
assorted places.
Change-Id: I5a7f4c25ddd4a6499dc295ef693b9fe43d2448ab
Reviewed-on: https://go-review.googlesource.com/c/go/+/404057
Run-TryBot: Paul Murphy <murp@ibm.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Lynn Boger <laboger@linux.vnet.ibm.com>
Reviewed-by: David Chase <drchase@google.com>
Reviewed-by: Russ Cox <rsc@golang.org>
When we convert a type to a shaped interface type, we are not able
to recognize the itab. So passing the itab by dictionary as the
workaround.
Fixes#52026.
Change-Id: I75c23c7dd215daf9761dc24116a8af2c28c6d948
Reviewed-on: https://go-review.googlesource.com/c/go/+/401034
Run-TryBot: Wayne Zuo <wdvxdr@golangcn.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Auto-Submit: Keith Randall <khr@golang.org>
Reviewed-by: Keith Randall <khr@google.com>
Reviewed-by: David Chase <drchase@google.com>
Reviewed-by: Keith Randall <khr@golang.org>
In #52529, we observed that checking types for duplicate fields and
methods during method collection can result in incorrect early expansion
of the base type. Fix this by delaying the check for duplicate fields.
Notably, we can't delay the check for duplicate methods as we must
preserve the invariant that added method names are unique.
After this change, it may be possible in the presence of errors to have
a type-checked type containing a method name that conflicts with a field
name. With the previous logic conflicting methods would have been
skipped. This is a change in behavior, but only for invalid code.
Preserving the existing behavior would likely require delaying method
collection, which could have more significant consequences.
As a result of this change, the compiler test fixedbugs/issue28268.go
started passing with types2, being previously marked as broken. The fix
was not actually related to the duplicate method error, but rather the
fact that we stopped reporting redundant errors on the calls to x.b()
and x.E(), because they are now (valid!) methods.
Fixes#52529
Change-Id: I850ce85c6ba76d79544f46bfd3deb8538d8c7d00
Reviewed-on: https://go-review.googlesource.com/c/go/+/403455
Reviewed-by: Robert Griesemer <gri@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
We need to use the same marker everywhere. My CL to rename the
marker (CL 241661) and the CL to add more uses of the marker
under the old name (CL 241678) weren't coordinated with each other.
Fixes#52612
Change-Id: I97023c0769e518491924ef457fe03bf64a2cefa6
Reviewed-on: https://go-review.googlesource.com/c/go/+/403094
Run-TryBot: Keith Randall <khr@golang.org>
Reviewed-by: Ian Lance Taylor <iant@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@google.com>
Accept ~x as ordinary unary expression in the parser but recognize
such expressions as invalid in the type checker.
This change opens the door to recognizing complex type constraint
literals such as `*E|~int` in `[P *E|~int]` and parse them correctly
instead of reporting a parse error because `P*E|~int` syntactically
looks like an incorrect array length expression (binary expression
where the RHS of | is an invalid unary expression ~int).
As a result, the parser is more forgiving with expressions but the
type checker will reject invalid uses as before.
We could pass extra information into the binary/unary expression
parse functions to prevent the use of ~ in invalid situations but
it doesn't seem worth the trouble. In fact it may be advantageous
to allow a more liberal expression syntax especially in the presence
of errors (better parser synchronization after an error).
Preparation for fixing #49482.
Change-Id: I119e8bd9445dfa6460fcd7e0658e3554a34b2769
Reviewed-on: https://go-review.googlesource.com/c/go/+/402255
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Reviewed-by: Robert Griesemer <gri@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
Change-Id: Ia0a4be56d4e1fbfc73e6ce24f01a658c89a74adb
GitHub-Last-Rev: dd95e50c4b
GitHub-Pull-Request: golang/go#52393
Reviewed-on: https://go-review.googlesource.com/c/go/+/400694
Reviewed-by: Ian Lance Taylor <iant@google.com>
Run-TryBot: Ian Lance Taylor <iant@google.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
This CL exports the existing ir.UintptrKeepAlive via the new directive
//go:uintptrkeepalive. This makes the compiler insert KeepAlives for
pointers converted to uintptr in calls, keeping them alive for the
duration of the call.
//go:uintptrkeepalive requires //go:nosplit, as stack growth can't
handle these arguments (it cannot know which are pointers). We currently
check this on the immediate function, but the actual restriction applies
to all transitive calls.
The existing //go:uintptrescapes is an extension of
//go:uintptrkeepalive which forces pointers to escape to the heap, thus
eliminating the stack growth issue.
This pragma is limited to the standard library.
For #51087
Change-Id: If9a19d484d3561b4219e5539b70c11a3cc09391e
Reviewed-on: https://go-review.googlesource.com/c/go/+/388095
Run-TryBot: Michael Pratt <mpratt@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
CL 388095 will change this file significantly. Move it preemptively to
ensure git tracks the move properly.
For #51087
Change-Id: I1408aecf8675723041b64e54cf44cdec38cc655c
Reviewed-on: https://go-review.googlesource.com/c/go/+/388094
Run-TryBot: Michael Pratt <mpratt@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: David Chase <drchase@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Fixes#52438.
Change-Id: I5cbf8c448dba037e9e0c5fe8f209401d6bf7d43f
Reviewed-on: https://go-review.googlesource.com/c/go/+/401134
Reviewed-by: Ian Lance Taylor <iant@google.com>
Reviewed-by: Robert Griesemer <gri@google.com>
The linker performs a global analysis of all nosplit call chains to
check they fit in the stack space ensured by splittable functions.
That analysis has two problems right now:
1. It's inefficient. It performs a top-down analysis, starting with
every nosplit function and the nosplit stack limit and walking *down*
the call graph to compute how much stack remains at every call. As a
result, it visits the same functions over and over, often with
different remaining stack depths. This approach is historical: this
check was originally written in C and this approach avoided the need
for any interesting data structures.
2. If some call chain is over the limit, it only reports a single call
chain. As a result, if the check does fail, you often wind up playing
whack-a-mole by guessing where the problem is in the one chain, trying
to reduce the stack size, and then seeing if the link works or reports
a different path.
This CL completely rewrites the nosplit stack check. It now uses a
bottom-up analysis, computing the maximum stack height required by
every function's call tree. This visits every function exactly once,
making it much more efficient. It uses slightly more heap space for
intermediate storage, but still very little in the scheme of the
overall link. For example, when linking cmd/go, the new algorithm
virtually eliminates the time spent in this pass, and reduces overall
link time:
│ before │ after │
│ sec/op │ sec/op vs base │
Dostkcheck 7.926m ± 4% 1.831m ± 6% -76.90% (p=0.000 n=20)
TotalTime 301.3m ± 1% 296.4m ± 3% -1.62% (p=0.040 n=20)
│ before │ after │
│ B/op │ B/op vs base │
Dostkcheck 40.00Ki ± 0% 212.15Ki ± 0% +430.37% (p=0.000 n=20)
Most of this time is spent analyzing the runtime, so for larger
binaries, the total time saved is roughly the same, and proportionally
less of the overall link.
If the new implementation finds an error, it redoes the analysis,
switching to preferring quality of error reporting over performance.
For error reporting, it computes stack depths top-down (like the old
algorithm), and reports *all* paths that are over the stack limit,
presented as a tree for compactness. For example, this is the output
from a simple test case from test/nosplit with two over-limit paths
from f1:
main.f1: nosplit stack overflow
main.f1
grows 768 bytes, calls main.f2
grows 56 bytes, calls main.f4
grows 48 bytes
80 bytes over limit
grows 768 bytes, calls main.f3
grows 104 bytes
80 bytes over limit
While we're here, we do a few nice cleanups:
- We add a debug output flag, which will be useful for understanding
what our nosplit chains look like and which ones are close to
running over.
- We move the implementation out of the fog of lib.go to its own file.
- The implementation is generally more Go-like and less C-like.
Change-Id: If1ab31197f5215475559b93695c44a01bd16e276
Reviewed-on: https://go-review.googlesource.com/c/go/+/398176
Run-TryBot: Austin Clements <austin@google.com>
Reviewed-by: Than McIntosh <thanm@google.com>
Reviewed-by: Cherry Mui <cherryyz@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
The nosplit test was originally written when the stack limit was a
mere 128 bytes. Now it's much larger, but rather than rewriting all of
the tests, we apply a hack to just add the extra space into the stack
frames of the existing tests.
Unfortunately, we add it in the wrong place. The extra space should be
added just once per chain of nosplit functions, but instead we add it
to every frame that appears first on a line in the test's little
script language. This means that for tests like
start 0 call f1
f1 16 nosplit call f2
f2 16 nosplit call f3
f3 16 nosplit call f4
f4 16 nosplit call f5
f5 16 nosplit call f6
f6 16 nosplit call f7
f7 16 nosplit call f8
f8 16 nosplit call end
end 1000
REJECT
we add 672 bytes to *every* frame, meaning that we wind up way over
the stack limit by the end of the stanza, rather than just a little as
originally intended.
Fix this by instead adding the extra space to the first nosplit
function in a stanza. This isn't perfect either, since we could have a
nosplit -> split -> nosplit chain, but it's the best we can do without
a graph analysis.
Change-Id: Ibf156c68fe3eb1b64a438115f4a17f1a6c7e2bd1
Reviewed-on: https://go-review.googlesource.com/c/go/+/398174
Run-TryBot: Austin Clements <austin@google.com>
Reviewed-by: Cherry Mui <cherryyz@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
So that the inliner knows all the other cases are dead and doesn't
accumulate any cost for them.
The canonical case for this is switching on runtime.GOOS, which occurs
several places in the stdlib.
Fixes#50253
Change-Id: I44823aaebb6c1b03c9b0c12d10086db81954350f
Reviewed-on: https://go-review.googlesource.com/c/go/+/399694
Run-TryBot: Keith Randall <khr@golang.org>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Reviewed-by: Russ Cox <rsc@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
runtime.getitab need filled fun[0] to identify whether
implemented the interface.
Fixes#51700Fixes#52228
Change-Id: I0173b98f4e1b45e3a0183a5b60229d289140d1e6
Reviewed-on: https://go-review.googlesource.com/c/go/+/399058
Reviewed-by: Keith Randall <khr@golang.org>
Run-TryBot: Keith Randall <khr@golang.org>
Auto-Submit: Keith Randall <khr@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Reviewed-by: David Chase <drchase@google.com>
gofmt is rewriting +build comments into //go:build anyway, so update
the test script to support both.
Change-Id: Ia6d950cfaa2fca9f184b8b2d3625a551bff88dde
Reviewed-on: https://go-review.googlesource.com/c/go/+/399794
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
Auto-Submit: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@google.com>
After CL 398014 fixed a compiler deadlock on syntax errors,
this CL adds a test case and more details for that.
How it was fixed:
CL 57751 introduced a channel "sem" to limit the number of
simultaneously open files.
Unfortunately, when the number of syntax processing goroutines
exceeds this limit, will easily trigger deadlock.
In the original implementation, "sem" only limited the number
of open files, not the number of concurrent goroutines, which
will cause extra goroutines to block on "sem". When the p.err
of the following iteration happens to be held by the blocking
goroutine, it will fall into a circular wait, which is a deadlock.
CL 398014 fixed the above deadlock, also see issue #52127.
First, move "sem <- struct{}{}" to the outside of the syntax
processing goroutine, so that the number of concurrent goroutines
does not exceed the number of open files, to ensure that all
goroutines in execution can eventually write to p.err.
Second, move the entire syntax processing logic into a separate
goroutine to avoid blocking on the producer side.
Change-Id: I1bb89bfee3d2703784f0c0d4ded82baab2ae867a
Reviewed-on: https://go-review.googlesource.com/c/go/+/399054
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Auto-Submit: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
