This CL adds a "deadlocals" pass, which runs after inlining and before
escape analysis, to prune any unneeded local variables and
assignments. In particular, this helps avoid unnecessary Addrtaken
markings from unreachable closures.
Deadlocals is sensitive to "_ = ..." as a signal of explicit
use for testing. This signal occurs only if the entire
left-hand-side is "_" targets; if it is
`_, ok := someInlinedFunc(args)`
then the first return value is eligible for dead code elimination.
Use this (`_ = x`) to fix tests broken by deadlocals elimination.
Includes a test, based on one of the tests that required modification.
Matthew Dempsky wrote this, changing ownership to allow rebases, commits, tweaks.
Fixes#65158.
Old-Change-Id: I723fb69ccd7baadaae04d415702ce6c8901eaf4e
Change-Id: I1f25f4293b19527f305c18c3680b214237a7714c
Reviewed-on: https://go-review.googlesource.com/c/go/+/600498
Reviewed-by: Keith Randall <khr@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Keith Randall <khr@golang.org>
Auto-Submit: David Chase <drchase@google.com>
Commit-Queue: David Chase <drchase@google.com>
Most of the test cases in the test directory use the new go:build syntax
already. Convert the rest. In general, try to place the build constraint
line below the test directive comment in more places.
For #41184.
For #60268.
Change-Id: I11c41a0642a8a26dc2eda1406da908645bbc005b
Cq-Include-Trybots: luci.golang.try:gotip-linux-386-longtest,gotip-linux-amd64-longtest,gotip-windows-amd64-longtest
Reviewed-on: https://go-review.googlesource.com/c/go/+/536236
Reviewed-by: Ian Lance Taylor <iant@google.com>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Auto-Submit: Dmitri Shuralyov <dmitshur@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
This patch changes the inliner to use callsite scores when deciding to
inline as opposed to looking only at callee cost/hairyness.
For this to work, we have to relax the inline budget cutoff as part of
CanInline to allow for the possibility that a given function might
start off with a cost of N where N > 80, but then be called from a
callsites whose score is less than 80. Once a given function F in
package P has been approved by CanInline (based on the relaxed budget)
it will then be emitted as part of the export data, meaning that other
packages importing P will need to also need to compute callsite scores
appropriately.
For a function F that calls function G, if G is marked as potentially
inlinable then the hairyness computation for F will use G's cost for
the call to G as opposed to the default call cost; for this to work
with the new scheme (given relaxed cost change described above) we
use G's cost only if it falls below inlineExtraCallCost, otherwise
just use inlineExtraCallCost.
Included in this patch are a bunch of skips and workarounds to
selected 'errorcheck' tests in the <GOROOT>/test directory to deal
with the additional "can inline" messages emitted when the new inliner
is turned on.
Change-Id: I9be5f8cd0cd8676beb4296faf80d2f6be7246335
Reviewed-on: https://go-review.googlesource.com/c/go/+/519197
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
In go.dev/cl/517775, I moved the frontend's deadcode elimination pass
into unified IR. But I also made a small enhancement: a branch like
"if x || true" is now detected as always taken, so the else branch can
be eliminated.
However, the inliner also has an optimization for delaying the
introduction of the result temporary variables when there's a single
return statement (added in go.dev/cl/266199). Consequently, the
inliner turns "if x || true { return true }; return true" into:
if x || true {
~R0 := true
goto .i0
}
.i0:
// code that uses ~R0
In turn, this confuses phi insertion, because it doesn't recognize
that the "if" statement is always taken, and so ~R0 will always be
initialized.
With this CL, after inlining we instead produce:
_ = x || true
~R0 := true
goto .i0
.i0:
Fixes#62211.
Change-Id: Ic8a12c9eb85833ee4e5d114f60e6c47817fce538
Reviewed-on: https://go-review.googlesource.com/c/go/+/522096
Reviewed-by: Than McIntosh <thanm@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Auto-Submit: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
CL 415241 and CL 411935 break tests into unified/nounified variants, for
compatibility with old frontend while developing unified IR. Now the old
frontend was gone, so moving those tests back to the original files.
Change-Id: Iecdcd4e6ee33c723f6ac02189b0be26248e15f0f
Reviewed-on: https://go-review.googlesource.com/c/go/+/497275
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: Keith Randall <khr@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
CL 399694 added constant-fold switch early in compilation. So function:
func f() string {
switch intSize {
case 32:
return "32"
case 64:
return "64"
default:
panic("unreachable")
}
}
will be constant-fold to:
func f() string {
switch intSize {
case 64:
return "64"
}
}
When this function get inlined, there is a check whether we can delay
declaring the result parameter until the "return" statement. For the
original function, we can't delay the result, because there's more than
one return statement. However, the constant-fold one can, because
there's on one return statement in the body now. The result parameter
~R0 ends up declaring inside the switch statement scope.
Now, when walking the switch statement, it's re-written into if-else
statement. Without typecheck.EvalConst, the if condition "if 64 == 64"
is passed as-is to the ssa generation pass. Because "64 == 64" is not a
constant, the ssagen creates normal blocks for branching the results.
This confuses the liveness analysis, because ~R0 is only live inside the
if block. With typecheck.EvalConst, "64 == 64" is evaluated to "true",
so ssagen can branch the result without emitting conditional blocks.
Instead, the constant-fold can be re-written as:
switch {
case true:
// Body
}
So it does not depend on the delay results check during inlining. Adding
a test, which will fail when typecheck.EvalConst is removed, so we can
do the cleanup without breaking things.
Change-Id: I638730bb147140de84260653741431b807ff2f15
Reviewed-on: https://go-review.googlesource.com/c/go/+/484316
Reviewed-by: Keith Randall <khr@google.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: David Chase <drchase@google.com>
Reviewed-by: Keith Randall <khr@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
This patch changes the relative order of "CanInline" and "InlineCalls"
operations within the inliner for clumps of functions corresponding to
strongly connected components in the call graph. This helps increase
the amount of inlining within SCCs, particularly in Go's runtime
package, which has a couple of very large SCCs.
For a given SCC of the form { fn1, fn2, ... fnk }, the inliner would
(prior to this point) walk through the list of functions and for each
function first compute inlinability ("CanInline") and then perform
inlining ("InlineCalls"). This meant that if there was an inlinable
call from fn3 to fn4 (for example), this call would never be inlined,
since at the point fn3 was visited, we would not have computed
inlinability for fn4.
We now do inlinability analysis for all functions in an SCC first,
then do actual inlining for everything. This results in 47 additional
inlines in the Go runtime package (a fairly modest increase
percentage-wise of 0.6%).
Updates #58905.
Change-Id: I48dbb1ca16f0b12f256d9eeba8cf7f3e6dd853cd
Reviewed-on: https://go-review.googlesource.com/c/go/+/474955
Run-TryBot: Than McIntosh <thanm@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
There are a plenty of regression in 1.20 with this optimization. This CL
disable inline static init, so it's safer to backport to 1.20 branch.
The optimization will be enabled again during 1.21 cycle.
Updates #58293
Updates #58339
For #58293
Change-Id: If5916008597b46146b4dc7108c6b389d53f35e95
Reviewed-on: https://go-review.googlesource.com/c/go/+/467015
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: Keith Randall <khr@google.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
This was disabled in CL 452676 out of an abundance of caution,
but further analysis has shown that the failures were not being
caused by this optimization. Instead the sequence of commits was:
CL 450136 cmd/compile: handle simple inlined calls in staticinit
...
CL 449937 archive/tar, archive/zip: return ErrInsecurePath for unsafe paths
...
CL 451555 cmd/compile: fix static init for inlined calls
The failures in question became compile failures in the first CL
and started building again after the last CL.
But in the interim the code had been broken by the middle CL.
CL 451555 was just the first time that the tests could run and fail.
For #30820.
Change-Id: I65064032355b56fdb43d9731be2f9f32ef6ee600
Reviewed-on: https://go-review.googlesource.com/c/go/+/452817
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Cherry Mui <cherryyz@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Russ Cox <rsc@golang.org>
Auto-Submit: Russ Cox <rsc@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
This CL adds -d=inlstaticinit to control whether static initialization
of inlined function calls (added in CL 450136) is allowed.
We've needed to fix it once already (CL 451555) and Google-internal
testing is hitting additional failure cases, so putting this
optimization behind a feature flag seems appropriate regardless.
Also, while we diagnose and fix the remaining cases, this CL also
disables the optimization to avoid miscompilations.
Updates #56894.
Change-Id: If52a358ad1e9d6aad1c74fac5a81ff9cfa5a3793
Reviewed-on: https://go-review.googlesource.com/c/go/+/452676
Reviewed-by: Cherry Mui <cherryyz@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
Global variable initializers like
var myErr error = &myError{"msg"}
have been converted to statically initialized data
from the earliest days of Go: there is no init-time
execution or allocation for that line of code.
But if the expression is moved into an inlinable function,
the static initialization no longer happens.
That is, this code has always executed and allocated
at init time, even after we added inlining to the compiler,
which should in theory make this code equivalent to
the original:
func NewError(s string) error { return &myError{s} }
var myErr2 = NewError("msg")
This CL makes the static initialization rewriter understand
inlined functions consisting of a single return statement,
like in this example, so that myErr2 can be implemented as
statically initialized data too, just like myErr, with no init-time
execution or allocation.
A real example of code that benefits from this rewrite is
all globally declared errors created with errors.New, like
package io
var EOF = errors.New("EOF")
Package io no longer has to allocate and initialize EOF each
time a program starts.
Another example of code that benefits is any globally declared
godebug setting (using the API from CL 449504), like
package http
var http2server = godebug.New("http2server")
These are no longer allocated and initialized at program startup either.
The list of functions that are inlined into static initializers when
compiling std and cmd (along with how many times each occurs) is:
cmd/compile/internal/ssa.StringToAux (3)
cmd/compile/internal/walk.mkmapnames (4)
errors.New (360)
go/ast.NewIdent (1)
go/constant.MakeBool (4)
go/constant.MakeInt64 (3)
image.NewUniform (4)
image/color.ModelFunc (11)
internal/godebug.New (12)
vendor/golang.org/x/text/unicode/bidi.newBidiTrie (1)
vendor/golang.org/x/text/unicode/norm.newNfcTrie (1)
vendor/golang.org/x/text/unicode/norm.newNfkcTrie (1)
For the cmd/go binary, this CL cuts the number of init-time
allocations from about 1920 to about 1620 (a 15% reduction).
The total executable code footprint of init functions is reduced
by 24kB, from 137kB to 113kB (an 18% reduction).
The overall binary size is reduced by 45kB,
from 15.335MB to 15.290MB (a 0.3% reduction).
(The binary size savings is larger than the executable code savings
because every byte of executable code also requires corresponding
runtime tables for unwinding, source-line mapping, and so on.)
Also merge test/sinit_run.go, which had stopped testing anything
at all as of CL 161337 (Feb 2019) and initempty.go into a new test
noinit.go.
Fixes#30820.
Change-Id: I52f7275b1ac2a0a32e22c29f9095071c7b1fac20
Reviewed-on: https://go-review.googlesource.com/c/go/+/450136
Reviewed-by: Cherry Mui <cherryyz@google.com>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Joedian Reid <joedian@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: Than McIntosh <thanm@google.com>
Auto-Submit: Russ Cox <rsc@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Unified IR records the inline nodes position right at the position of
the inline call, while the old inliner always records at the position of
the original nodes.
We want to keep non-unified working up through go 1.20, thus this CL
extract the inline test case that is different in Unified IR and the old
inliner.
Updates #53058
Change-Id: I14b0ee99fe797d34f27cfec068982790c64ac263
Reviewed-on: https://go-review.googlesource.com/c/go/+/411935
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Cherry Mui <cherryyz@google.com>
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
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>
CL 394074 broke the noopt builder. Something about time.After's inlining
depends on the build flags to make.bash, not the build flags that run.go
passes.
Change-Id: Ib284c66ea2008a4d32829c055d57c54a34ec3fb4
Reviewed-on: https://go-review.googlesource.com/c/go/+/396037
Trust: Keith Randall <khr@golang.org>
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
This patch revises the fix for issue 46234, fixing a bug that was
accidentally introduced by CL 320913. When inlining a chunk of code
with a closure expression, we want to avoid updating the source
positions in the function being closed over, but we do want to update
the position for the ClosureExpr itself (since it is part of the
function we are inlining). CL 320913 unintentionally did away with the
closure expr source position update; here we restore it again.
Updates #46234.
Fixes#49171.
Change-Id: Iaa51bc498e374b9e5a46fa0acd7db520edbbbfca
Reviewed-on: https://go-review.googlesource.com/c/go/+/366494
Trust: Than McIntosh <thanm@google.com>
Trust: Dan Scales <danscales@google.com>
Reviewed-by: Dan Scales <danscales@google.com>
After CL 349012 and CL 350911, we can fully handle these
labeled statements, so we can allow them when inlining.
Updates #14768
Change-Id: I0ab3fd3f8d7436b49b1aedd946516b33c63f5747
Reviewed-on: https://go-review.googlesource.com/c/go/+/355497
Run-TryBot: David Chase <drchase@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Dan Scales <danscales@google.com>
Reviewed-by: David Chase <drchase@google.com>
Trust: Dan Scales <danscales@google.com>
This CL renames blank functions in the test/ directory so that they
don't rely on the compiler doing anything more than typechecking them.
In particular, I ran this search to find files that used blank
functions and methods:
$ git grep -l '^func.*\b_(' | xargs grep -n '^' | grep '\.go:1:' | grep -v '// errorcheck$'
I then skipped updating a few files:
* blank.go
* fixedbugs/issue11699.go
* fixedbugs/issue29870.go
These tests specifically check that blank functions/methods work.
* interface/fail.go
Not sure the motivation for the blank method here, but it's empty
anyway.
* typeparam/tparam1.go
Type-checking test, but uses "-G" (to use types2 instead of typecheck).
Updates #47446.
Change-Id: I9ec1714f499808768bd0dcd7ae6016fb2b078e5e
Reviewed-on: https://go-review.googlesource.com/c/go/+/338094
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>
As walk already create the wrapper if necessary.
With this change, test/inline.go need to be changed to use
errorcheckwithauto, for matching "inlining call to ..." in autogenerated
position for method value wrapper, since when we don't generate the
wrapper immediately during typecheck.
Change-Id: I9ffbec9ad3c2b7295546976e2fa517336c13c89b
Reviewed-on: https://go-review.googlesource.com/c/go/+/330838
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>
When inlining functions with closures, ensure that we don't mark the
body of the closure with a src.Pos marker that reflects the inline,
since this will result in the generation of an inltree table for the
closure itself (as opposed to the routine that the func-with-closure
was inlined into).
Fixes#46234.
Change-Id: I348296de6504fc4745d99adab436640f50be299a
Reviewed-on: https://go-review.googlesource.com/c/go/+/320913
Reviewed-by: Cherry Mui <cherryyz@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Cherry Mui <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Than McIntosh <thanm@google.com>
For the new export/import of node types, we were just missing setting
the types of the closure variables (which have the same types as the
captured variables) and the OCLOSURE node itself (which has the same
type as the Func node).
Re-enabled inlining of functions with closures.
Change-Id: I687149b061f3ffeec3244ff02dc6e946659077a9
Reviewed-on: https://go-review.googlesource.com/c/go/+/308974
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Include type information on exported function bodies, so that the
importer does not have to re-typecheck the body. This involves
including type information in the encoded output, as well as
avoiding some of the opcode rewriting and other changes that the
old exporter did assuming there would be a re-typechecking pass.
This CL could be considered a cleanup, but is more important than that
because it is an enabling change for generics. Without this CL, we'd
have to upgrade the current typechecker to understand generics. With
this CL, the current typechecker can mostly go away in favor of the
types2 typechecker.
For now, inlining of functions that contain closures is turned off.
We will hopefully resolve this before freeze.
Object files are only 0.07% bigger.
Change-Id: I85c9da09f66bfdc910dc3e26abb2613a1831634d
Reviewed-on: https://go-review.googlesource.com/c/go/+/301291
Trust: Keith Randall <khr@golang.org>
Trust: Dan Scales <danscales@google.com>
Reviewed-by: Dan Scales <danscales@google.com>
math.Float32bits was not being inlined across package boundaries.
Create a private func that can be inlined with -l.
Change-Id: Ic50bf4727dd8ade09d011eb204006b7ee88db34a
Reviewed-on: https://go-review.googlesource.com/c/go/+/295989
Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>
Reviewed-by: Bryan C. Mills <bcmills@google.com>
Reviewed-by: Keith Randall <khr@golang.org>
Trust: Josh Bleecher Snyder <josharian@gmail.com>
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Added a flag '-d=inlfuncswithclosures=1' to allow inlining functions with
closures, and change the default to off for now, until #44370 is fixed.
Updates #44370.
Change-Id: Ic17723aa5c091d91f5f5004d8b63ec7125257acf
Reviewed-on: https://go-review.googlesource.com/c/go/+/296049
Run-TryBot: Dan Scales <danscales@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Dan Scales <danscales@google.com>
OCONVOP doesn't have effect in the compiled code so, it can be safely
excluded from inline cost calculation.
Also make sequence ODEREF OCONVNOP* OADDR cost 1. This is rather common
conversion, such as *(*uint32)(unsafe.Pointer(&x)).
Fixes#42788
Change-Id: I5001f7e89d985c198b6405694cdd5b819cf3f47a
Reviewed-on: https://go-review.googlesource.com/c/go/+/281232
Reviewed-by: Keith Randall <khr@golang.org>
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Elias Naur <mail@eliasnaur.com>
I have exporting, importing, and inlining of functions with closures
working in all cases (issue #28727). all.bash runs successfully without
errors.
Approach:
- Write out the Func type, Dcls, ClosureVars, and Body when exporting
an OCLOSURE.
- When importing an OCLOSURE, read in the type, dcls, closure vars,
and body, and then do roughly equivalent code to (*noder).funcLit
- During inlining of a closure within inlined function, create new
nodes for all params and local variables (including closure
variables), so they can have a new Curfn and some other field
values. Must substitute not only on the Nbody of the closure, but
also the Type, Cvars, and Dcl fields.
Fixes#28727
Change-Id: I4da1e2567c3fa31a5121afbe82dc4e5ee32b3170
Reviewed-on: https://go-review.googlesource.com/c/go/+/283112
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Trust: Dan Scales <danscales@google.com>
This CL adds support for inlining type switches, including exporting
and importing them.
Type switches are represented mostly the same as expression switches.
However, if the type switch guard includes a short variable
declaration, then there are two differences: (1) there's an ONONAME
(in the OTYPESW's Left) to represent the overall pseudo declaration;
and (2) there's an ONAME (in each OCASE's Rlist) to represent the
per-case variables.
For simplicity, this CL simply writes out each variable separately
using iimport/iiexport's normal Vargen mechanism for disambiguating
identically named variables within a function. This could be improved
somewhat, but inlinable type switches are probably too uncommon to
merit the complexity.
While here, remove "case OCASE" from typecheck1. We only type check
"case" clauses as part of a "select" or "switch" statement, never as
standalone statements.
Fixes#37837
Change-Id: I8f42f6c9afdd821d6202af4a6bf1dbcbba0ef424
Reviewed-on: https://go-review.googlesource.com/c/go/+/266203
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Trust: Matthew Dempsky <mdempsky@google.com>
The inlining pass previously bailed upon encountering a go or defer statement, so it would not inline functions e.g. used to provide arguments to the deferred function. This change preserves the behavior of not inlining the
deferred function itself, but it allows the inlining walk to proceed into its arguments.
Fixes#42194
Change-Id: I4e82029d8dcbe69019cc83ae63a4b29af45ec777
Reviewed-on: https://go-review.googlesource.com/c/go/+/264997
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
OCALLPART is exported in its original form, which is as an OXDOT.
The body of the method value wrapper created in makepartialcall() was
not being typechecked, and that was causing a problem during escape
analysis, so I added code to typecheck the body.
The go executable got slightly bigger with this change (13598111 ->
13598905), because of extra exported methods with OCALLPART (I
believe), while the text size got slightly smaller (9686964 ->
9686643).
This is mainly part of the work to make sure all function bodies can
be exported (for purposes of generics), but might as well fix the
OCALLPART inlining bug as well.
Fixes#18493
Change-Id: If7aa055ff78ed7a6330c6a1e22f836ec567d04fd
Reviewed-on: https://go-review.googlesource.com/c/go/+/263620
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
This CL replaces the ad hoc and duplicated logic for detecting
inlinable calls with a single "inlCallee" function, which uses the
"staticValue" helper function introduced in an earlier commit.
Updates #41474.
Change-Id: I103d4091b10366fce1344ef2501222b7df68f21d
Reviewed-on: https://go-review.googlesource.com/c/go/+/256460
Reviewed-by: David Chase <drchase@google.com>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Trust: Matthew Dempsky <mdempsky@google.com>
We already allow inlining "if" and "goto" statements, so we might as
well allow "for" loops too. The majority of frontend support is
already there too.
The critical missing feature at the moment is that inlining doesn't
properly reassociate OLABEL nodes with their control statement (e.g.,
OFOR) after inlining. This eventually causes SSA construction to fail.
As a workaround, this CL only enables inlining for unlabeled "for"
loops. It's left to a (yet unplanned) future CL to add support for
labeled "for" loops.
The increased opportunity for inlining leads to a small growth in
binary size. For example:
$ size go.old go.new
text data bss dec hex filename
9740163 320064 230656 10290883 9d06c3 go.old
9793399 320064 230656 10344119 9dd6b7 go.new
Updates #14768.
Fixes#41474.
Change-Id: I827db0b2b9d9fa2934db05caf6baa463f0cd032a
Reviewed-on: https://go-review.googlesource.com/c/go/+/256459
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: David Chase <drchase@google.com>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
This allows the global initializers function to go through normal
mid-end optimizations (e.g., inlining, escape analysis) like any other
function.
Updates #33485.
Change-Id: I9bcfe98b8628d1aca09b4c238d8d3b74c69010a5
Reviewed-on: https://go-review.googlesource.com/c/go/+/254839
Reviewed-by: Keith Randall <khr@golang.org>
Trust: Matthew Dempsky <mdempsky@google.com>
We still disallow inlining for an immediately-recursive function, but allow
inlining if a function is in a recursion chain.
If all functions in the recursion chain are simple, then we could inline
forever down the recursion chain (eventually running out of stack on the
compiler), so we add a map to keep track of the functions we have
already inlined at a call site. We stop inlining when we reach a
function that we have already inlined in the recursive chain. Of course,
normally the inlining will have stopped earlier, because of the cost
function.
We could also limit the depth of inlining by a simple count (say, limit
max inlining of 10 at any given site). Would that limit other
opportunities too much?
Added a test in test/inline.go. runtime.BenchmarkStackCopyNoCache() is
also already a good test that triggers the check to stop inlining
when we reach the start of the recursive chain again.
For the bent benchmark suite, the performance improvement was mostly not
statistically significant, but the geomean averaged out to: -0.68%. The text size
increase was less than .1% for all bent benchmarks. The cmd/go text size increase
was 0.02% and the cmd/compile text size increase was .1%.
Fixes#29737
Change-Id: I892fa84bb07a947b3125ec8f25ed0e508bf2bdf5
Reviewed-on: https://go-review.googlesource.com/c/go/+/226818
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
This information is redundant with the position information already
provided. Also, no other -m diagnostics print out function name.
While here, report parameter leak diagnostics against the parameter
declaration position rather than the function, and use Warnl for
"moved to heap" messages.
Test cases updated programmatically by removing the first word from
every "no match for" error emitted by run.go:
go run run.go |& \
sed -E -n 's/^(.*):(.*): no match for `([^ ]* (.*))` in:$/\1!\2!\3!\4/p' | \
while IFS='!' read -r fn line before after; do
before=$(echo "$before" | sed 's/[.[\*^$()+?{|]/\\&/g')
after=$(echo "$after" | sed -E 's/(\&|\\)/\\&/g')
fn=$(find . -name "${fn}" | head -1)
sed -i -E -e "${line}s/\"${before}\"/\"${after}\"/" "${fn}"
done
Passes toolstash-check.
Change-Id: I6e02486b1409e4a8dbb2b9b816d22095835426b5
Reviewed-on: https://go-review.googlesource.com/c/go/+/195040
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
For most nodes (e.g., OPTRLIT, OMAKESLICE, OCONVIFACE), escape
analysis prints "escapes to heap" or "does not escape" to indicate
whether that node's allocation can be heap or stack allocated.
These messages are also emitted for OADDR, even though OADDR does not
actually allocate anything itself. Moreover, it's redundant because
escape analysis already prints "moved to heap" diagnostics when an
OADDR node like "&x" causes x to require heap allocation.
Because OADDR nodes don't allocate memory, my escape analysis rewrite
doesn't naturally emit the "escapes to heap" / "does not escape"
diagnostics for them. It's also non-trivial to replicate the exact
semantics esc.go uses for OADDR.
Since there are so many of these messages, I'm disabling them in this
CL by themselves. I modified esc.go to suppress the Warnl calls
without any other behavior changes, and then used a shell script to
automatically remove any ERROR messages mentioned by run.go in
"missing error" or "no match for" lines.
Fixes#16300.
Updates #23109.
Change-Id: I3993e2743c3ff83ccd0893f4e73b366ff8871a57
Reviewed-on: https://go-review.googlesource.com/c/go/+/170319
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Reviewed-by: David Chase <drchase@google.com>
This is a simple tweak to allow a bit more mid-stack inlining.
In cases like this:
func f() {
g()
}
We'd really like to inline f into its callers. It can't hurt.
We implement this optimization by making calls a bit cheaper, enough
to afford a single call in the function body, but not 2.
The remaining budget allows for some argument modification, or perhaps
a wrapping conditional:
func f(x int) {
g(x, 0)
}
func f(x int) {
if x > 0 {
g()
}
}
Update #19348
Change-Id: Ifb1ea0dd1db216c3fd5c453c31c3355561fe406f
Reviewed-on: https://go-review.googlesource.com/c/147361
Reviewed-by: Austin Clements <austin@google.com>
Reviewed-by: David Chase <drchase@google.com>
Calls to a closure held in a local, non-escaping,
variable can be inlined, provided the closure body
can be inlined and the variable is never written to.
The current implementation has the following limitations:
- closures with captured variables are not inlined because
doing so naively triggers invariant violation in the SSA
phase
- re-assignment check is currently approximated by checking
the Addrtaken property of the variable which should be safe
but may miss optimization opportunities if the address is
not used for a write before the invocation
Updates #15561
Change-Id: I508cad5d28f027bd7e933b1f793c14dcfef8b5a1
Reviewed-on: https://go-review.googlesource.com/65071
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Hugues Bruant <hugues.bruant@gmail.com>
Reviewed-by: Keith Randall <khr@golang.org>
We've supported inlining methods called as functions for a while now.
Change-Id: I53fba426e45f91d65a38f00456c2ae1527372b50
Reviewed-on: https://go-review.googlesource.com/69530
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
Allow inlining of functions with switch statements as long as they don't
contain a break or type switch.
Fixes#13071
Change-Id: I057be351ea4584def1a744ee87eafa5df47a7f6d
Reviewed-on: https://go-review.googlesource.com/20824
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Consider functions with an ODCLCONST for inlining and modify exprfmt to
ignore those nodes when exporting. Don't add symbols to the export list
if there is no definition. This occurs when OLITERAL symbols are looked
up via Pkglookup for non-exported symbols.
Fixes#7655
Change-Id: I1de827850f4c69e58107447314fe7433e378e069
Reviewed-on: https://go-review.googlesource.com/20773
Run-TryBot: Todd Neal <todd@tneal.org>
Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Robert Griesemer <gri@golang.org>
Some tests need to disable inlining of a function. It's currently done
in one of a few ways (adding a function call, an empty switch, or a
defer). Add support for a less fragile 'go:noinline' directive that
prevents inlining.
Fixes#12312
Change-Id: Ife444e13361b4a927709d81aa41e448f32eec8d4
Reviewed-on: https://go-review.googlesource.com/13911
Run-TryBot: Todd Neal <todd@tneal.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>
Inlining refuses to inline bodies containing an actual function call, so that
if that call or a child uses runtime.Caller it cannot observe
the inlining.
However, inlining was also refusing to inline bodies that contained
function calls that were themselves inlined away. For example:
func f() int {
return f1()
}
func f1() int {
return f2()
}
func f2() int {
return 2
}
The f2 call in f1 would be inlined, but the f1 call in f would not,
because f1's call to f2 blocked the inlining, despite itself eventually
being inlined away.
Account properly for this kind of transitive inlining and enable.
Also bump the inlining budget a bit, so that the runtime's
heapBits.next is inlined.
This reduces the time for '6g *.go' in html/template by around 12% (!).
(For what it's worth, closing Chrome reduces the time by about 17%.)
Change-Id: If1aa673bf3e583082dcfb5f223e67355c984bfc1
Reviewed-on: https://go-review.googlesource.com/5952
Reviewed-by: Austin Clements <austin@google.com>
