The upstream pprof implements the readline feature using
the github.com/chzyer/readline package in its pprof.go main.
It would be ideal to use the same readline support package as
the upstream for better user experience and code maintenance.
However, bringing in third-party packages requires more work
than I envisioned (e.g. clean up the vendored code to meet the
expected standard - iow don't break builders).
As a result, this change implements the similar feature
for the pprof command included in the go distribution
(cmd/pprof/pprof.go) using golang.org/x/crypto/ssh/terminal
for now.
Auto-completion is not yet supported (same in the upstream).
The feature is enabled only in linux, windows, darwin, and
only when terminal support is available.
This change brings in new vendored packages,
golang.org/x/crypto/ssh/terminal and
golang.org/x/sys/{unix,windows}.
For #14041
Change-Id: If4a790796acf2ab20f7e81268b9d9354c5a5cd2b
Reviewed-on: https://go-review.googlesource.com/112436
Run-TryBot: Hyang-Ah Hana Kim <hyangah@gmail.com>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
This partially reverts commit 3bdbb5df76.
The latest CL of js/wasm's file system support does not use
file descriptor mapping any more.
Change-Id: Iaec9c84b392366282cddc69acc75c8a3eb556824
Reviewed-on: https://go-review.googlesource.com/114195
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
The current documentation of the WriteTo function is very poor and it
is difficult to deduce how to use it correctly. A good example will
make things much easier.
Fixes#25456
Change-Id: Ibf0c0e153afae8f3e0d7d765d0dc9bcbfd69bfb1
Reviewed-on: https://go-review.googlesource.com/113775
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
When a test binary runs for too long, the go command sends it a
SIGQUIT to force a backtrace dump. On Android, the exec wrapper
will instead receive the signal and dump its backtrace.
Forward SIGQUIT signals from the wrapper to the wrapped process
to gain useful backtraces.
Inspired by issuse 25519; this CL would have revealed the hanging
test directly in the builder log.
Change-Id: Ic362d06940d261374343a1dc09366ef54edaa631
Reviewed-on: https://go-review.googlesource.com/114137
Run-TryBot: Elias Naur <elias.naur@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
While next-ing over a call in gdb, if execution of that call
causes a goroutine's stack to grow (i.e., be moved), gdb loses
track and runs ahead to the next breakpoint, or to the end of
the program, whichever comes first.
Prevent this by preemptively growing the stack so that
ssa/debug_test.go will reliably measure what is intended,
the goodness of line number placement and variable printing.
Fixes#25497.
Change-Id: I8daf931650292a8c8faad2285d7fd405f2157bd2
Reviewed-on: https://go-review.googlesource.com/114080
Run-TryBot: David Chase <drchase@google.com>
Reviewed-by: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
golang.org/cl/37879 unintentionally changed the way NXDOMAIN errors were
handled. Before that change, resolution would fail on the first NXDOMAIN
error and return to the user. After that change, the next server would
be consulted and resolution would fail only after all servers had been
consulted. This change restores the old behavior.
Go 10.10.2:
BenchmarkGoLookupIP-12 10000 174883 ns/op 11450 B/op 163 allocs/op
BenchmarkGoLookupIPNoSuchHost-12 3000 670140 ns/op 52189 B/op 544 allocs/op
BenchmarkGoLookupIPWithBrokenNameServer-12 1 5002568137 ns/op 163792 B/op 375 allocs/op
before this change:
BenchmarkGoLookupIP-12 10000 165501 ns/op 8585 B/op 94 allocs/op
BenchmarkGoLookupIPNoSuchHost-12 1000 1204117 ns/op 83661 B/op 674 allocs/op
BenchmarkGoLookupIPWithBrokenNameServer-12 1 5002629186 ns/op 159128 B/op 275 allocs/op
after this change:
BenchmarkGoLookupIP-12 10000 158102 ns/op 8585 B/op 94 allocs/op
BenchmarkGoLookupIPNoSuchHost-12 2000 645364 ns/op 42990 B/op 356 allocs/op
BenchmarkGoLookupIPWithBrokenNameServer-12 1 5002163437 ns/op 159144 B/op 275 allocs/op
Fixes#25336
Change-Id: I315cd70330d1f66e54ce5a189a61c99f095bc138
Reviewed-on: https://go-review.googlesource.com/113815
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
This function checks Request.PostForm, which now includes values parsed
from a PATCH request.
Change-Id: I5d0af58d9c0e9111d4e822c45f0fb1f511bbf0d5
Reviewed-on: https://go-review.googlesource.com/114009
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
This change adds Yi8 forms for every ytab that had them before AVX-512 patch.
The rationale is backwards-compatibility.
EVEX forms remain strict and unchanged as they're not bound to any
backwards-compatibility issues.
Fixes#25510
Change-Id: Icd692266010ed64c9fe47cc837afc2edf2ad2d1d
Reviewed-on: https://go-review.googlesource.com/114136
Run-TryBot: Iskander Sharipov <iskander.sharipov@intel.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ilya Tocar <ilya.tocar@intel.com>
The existing code just used timestamps. The new code uses more fields
when timestamps are equal.
Revised to shorten code per reviewer comments.
Change-Id: Ibd0824d0acd7644484d536b1a754a0da156fac3d
Reviewed-on: https://go-review.googlesource.com/113721
Reviewed-by: Hyang-Ah Hana Kim <hyangah@gmail.com>
Fixes builders that do not have cgo installed.
Change-Id: I719b7959226b0e67c3ffc11e071784787cabc5ab
Reviewed-on: https://go-review.googlesource.com/114235
Run-TryBot: Alex Brainman <alex.brainman@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
CL 110555 introduced some changes which were not properly gofmt'ed.
Because the CL was sent via Github the gofmt checks usually performed by
git-codereview didn't catch this (see #24946).
Change-Id: I65c1271620690dbeec88b4ce482d158f7d6df45d
Reviewed-on: https://go-review.googlesource.com/114255
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Alex Brainman <alex.brainman@gmail.com>
Windows does not allow to delete opened file.
Fixes#24704
Change-Id: Idfca2d00a2c46bdd9bd2a721478bfd003c474ece
Reviewed-on: https://go-review.googlesource.com/113935
Run-TryBot: Alex Brainman <alex.brainman@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
There is no NL at the end of README, and that make it strange
when doing "cat misc/android/README".
Change-Id: Ib47953d7b16e8927a4d6be7d5be8de8f2ddbcc39
Reviewed-on: https://go-review.googlesource.com/114010
Reviewed-by: Elias Naur <elias.naur@gmail.com>
The splice syscall is buggy prior to linux 2.6.29. Instead of returning
0 when reading a closed socket, it returns EAGAIN. While it is possible
to detect this (HAProxy falls back to recv), it is simpiler to avoid
using splice all together. the "fcntl(fd, F_GETPIPE_SZ)" syscall is used
detect buggy versions of splice as the syscall returns EINVAL on
versions prior to 2.6.35.
Fixes#25486
Change-Id: I860c029f13de2b09e95a7ba39b76ac7fca91a195
Reviewed-on: https://go-review.googlesource.com/113999
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
In rare circumstances that we don't yet fully understand, the g
register can be spilled to the stack and then reloaded. If this
happens, liveness analysis sees a pointer load into a
non-general-purpose register and panics.
We should fix the root cause of this, but fix the build for now by
ignoring pointer loads into the g register.
For #25504.
Change-Id: I0dfee1af9750c8e9157c7637280cdf07118ef2ca
Reviewed-on: https://go-review.googlesource.com/114081
Run-TryBot: Austin Clements <austin@google.com>
Reviewed-by: Keith Randall <khr@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
There's semantically-but-not-literally equivalent code in
two cases for joining blocks' value lists in ssa/fuse.go.
It can be made literally equivalent, then commoned up.
Updates #25426.
Change-Id: Id1819366c9d22e5126f9203dcd4c622423994110
Reviewed-on: https://go-review.googlesource.com/113719
Run-TryBot: David Chase <drchase@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com>
This moves the bvec hash table logic out of Liveness.compact and into
a bvecSet type. Furthermore, the bvecSet type has the ability to grow
dynamically, which the current implementation doesn't. In addition to
making the code cleaner, this will make it possible to incrementally
compact liveness bitmaps.
Passes toolstash -cmp
Updates #24543.
Change-Id: I46c53e504494206061a1f790ae4a02d768a65681
Reviewed-on: https://go-review.googlesource.com/110176
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Currently Liveness.epilogue makes three passes over the Blocks, but
there's no need to do this. Combine them into a single pass. This
eliminates the need for blockEffects.lastbitmapindex, but, more
importantly, will let us incrementally compact the liveness bitmaps
and significantly reduce allocatons in Liveness.epilogue.
Passes toolstash -cmp.
Updates #24543.
Change-Id: I27802bcd00d23aa122a7ec16cdfd739ae12dd7aa
Reviewed-on: https://go-review.googlesource.com/110175
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: David Chase <drchase@google.com>
Until vgo sorts out and cleans up the vendoring process.
Ran govendor to update packages the cmd/pprof depends on
which resulted in deletion of some of unnecessary files.
Change-Id: Idfba53e94414e90a5e280222750a6df77e979a16
Reviewed-on: https://go-review.googlesource.com/114079
Run-TryBot: Hyang-Ah Hana Kim <hyangah@gmail.com>
Reviewed-by: Daniel Theophanes <kardianos@gmail.com>
On OSX 10.12 and earlier, paired with XCode 9.0,
specifying DWARF version 3 causes dsymutil to misbehave.
Version 2 appears to be good enough to allow processing
of the prologue_end opcode on (at least one version of)
Linux and OSX 10.13.
Fixes#25451.
Change-Id: Ic760e34248393a5386be96351c8e492da1d3413b
Reviewed-on: https://go-review.googlesource.com/114015
Reviewed-by: Alessandro Arzilli <alessandro.arzilli@gmail.com>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Traceback matches the defer stack with the function call stack using
the SP recorded in defer frames when the defer frame is created.
However, on LR machines this is ambiguous: if function A pushes a
defer and then calls function B, where B is a leaf function with a
zero-sized frame, then both A and B have the same SP and will *both*
match the defer on the defer stack. Since traceback unwinds through B
first, it will incorrectly match up the defer with B's frame instead
of A's frame.
Where this goes particularly wrong is if function B causes a signal
that turns into a panic (e.g., a nil pointer dereference). In order to
handle the fact that we may not have a liveness map at the location
that caused the signal and injected a sigpanic call, traceback has
logic to unwind the panicking frame's continuation PC to the PC where
the most recent defer was pushed (this is safe because the frame is
dead other than any defers it pushed). However, if traceback
mis-matches the defer stack, it winds up reporting the B's
continuation PC is in A. If the runtime then uses this continuation PC
to look up PCDATA in B, it will panic because the PC is out of range
for B. This failure mode can be seen in
sync/atomic/atomic_test.go:TestNilDeref. An example failure is:
https://build.golang.org/log/8e07a762487839252af902355f6b1379dbd463c5
This CL fixes all of this by recognizing that a function that pushes a
defer must also have a non-zero-sized frame and using this fact to
refine the defer matching logic.
Fixes the build for arm64, mips, mipsle, ppc64, ppc64le, and s390x.
Fixes#25499.
Change-Id: Iff7c01d08ad42f3de22b3a73658cc2f674900101
Reviewed-on: https://go-review.googlesource.com/114078
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Needs the go compiler to be build with GOEXPERIMENT=debugcpu to be active.
The GODEBUGCPU environment variable can be used to disable usage of
specific processor features in the Go standard library.
This is useful for testing and benchmarking different code paths that
are guarded by internal/cpu variable checks.
Use of processor features can not be enabled through GODEBUGCPU.
To disable usage of AVX and SSE41 cpu features on GOARCH amd64 use:
GODEBUGCPU=avx=0,sse41=0
The special "all" option can be used to disable all options:
GODEBUGCPU=all=0
Updates #12805
Updates #15403
Change-Id: I699c2e6f74d98472b6fb4b1e5ffbf29b15697aab
Reviewed-on: https://go-review.googlesource.com/91737
Run-TryBot: Martin Möhrmann <moehrmann@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
This makes the checking of build tags in file names consistent to that of the build tags in `// +build` line.
Fixed#25461
Change-Id: Iba14d1050f8aba44e7539ab3b8711af1980ccfe4
GitHub-Last-Rev: 11b14e239d
GitHub-Pull-Request: golang/go#25480
Reviewed-on: https://go-review.googlesource.com/113818
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
The added fields are used in buildExtensions so
should be documented too.
Fixes#21363
Change-Id: Ifcc11da5b690327946c2488bcf4c79c60175a339
Reviewed-on: https://go-review.googlesource.com/113916
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
It's easier to skim a list of items visually when the
items are each on a separate line. Separate lines also
help reduce diff size when items are added/removed.
The list is indented so that it's displayed preformatted
in HTML output as godoc doesn't support formatting lists
natively yet (see #7873).
Change-Id: Ibf9e92437e4b464ba58ea3ccef579e8df4745d75
Reviewed-on: https://go-review.googlesource.com/113915
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Some functions in log/syslog depend on syslogd running. Instead of
treating errors caused by the daemon not running as test failures,
ignore them and skip the test.
Fixes the longtest builder.
Change-Id: I628fe4aab5f1a505edfc0748861bb976ed5917ea
Reviewed-on: https://go-review.googlesource.com/113838
Run-TryBot: Alberto Donizetti <alb.donizetti@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Immediately following the conditional block removed here is a loop
which checks exactly what the conditional already checked, so the
entire conditional is redundant.
Change-Id: I892fd9f2364d87e2c1cacb0407531daec6643183
Reviewed-on: https://go-review.googlesource.com/114000
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
When rulegen complains about a missing type, report the line number
in the rules file.
Change-Id: Ic7c19e1d5f29547911909df5788945848a6080ff
Reviewed-on: https://go-review.googlesource.com/114004
Reviewed-by: David Chase <drchase@google.com>
This adds a mechanism for debuggers to safely inject calls to Go
functions on amd64. Debuggers must participate in a protocol with the
runtime, and need to know how to lay out a call frame, but the runtime
support takes care of the details of handling live pointers in
registers, stack growth, and detecting the trickier conditions when it
is unsafe to inject a user function call.
Fixes#21678.
Updates derekparker/delve#119.
Change-Id: I56d8ca67700f1f77e19d89e7fc92ab337b228834
Reviewed-on: https://go-review.googlesource.com/109699
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
This extends the liveness analysis to track registers containing live
pointers. We do this by tracking bitmaps for live pointer registers
in parallel with bitmaps for stack variables.
This does not yet do anything with these liveness maps, though they do
appear in the debug output for -live=2.
We'll optimize this in later CLs:
name old time/op new time/op delta
Template 193ms ± 5% 195ms ± 2% ~ (p=0.050 n=9+9)
Unicode 97.7ms ± 2% 98.4ms ± 2% ~ (p=0.315 n=9+10)
GoTypes 674ms ± 2% 685ms ± 1% +1.72% (p=0.001 n=9+9)
Compiler 3.21s ± 1% 3.28s ± 1% +2.28% (p=0.000 n=10+9)
SSA 7.70s ± 1% 7.79s ± 1% +1.07% (p=0.015 n=10+10)
Flate 130ms ± 3% 133ms ± 2% +2.19% (p=0.003 n=10+10)
GoParser 159ms ± 3% 161ms ± 2% +1.51% (p=0.019 n=10+10)
Reflect 444ms ± 1% 450ms ± 1% +1.43% (p=0.000 n=9+10)
Tar 181ms ± 2% 183ms ± 2% +1.45% (p=0.010 n=10+9)
XML 230ms ± 1% 234ms ± 1% +1.56% (p=0.000 n=8+9)
[Geo mean] 405ms 411ms +1.48%
No effect on binary size because we're not yet emitting the register
maps.
For #24543.
Change-Id: Ieb022f0aea89c0ea9a6f035195bce2f0e67dbae4
Reviewed-on: https://go-review.googlesource.com/109352
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
For register maps, we need a dense numbering of registers that may
contain pointers of interest to the garbage collector. Add this to
Register and compute it from the GP register set.
For #24543.
Change-Id: If6f0521effca5eca4d17895468b1fc52d67e0f32
Reviewed-on: https://go-review.googlesource.com/109351
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Compiling without optimizations (-N) can result in write barrier
blocks that have been optimized away but not actually pruned from the
block set. Fix unsafe-point analysis to recognize and ignore these.
For #24543.
Change-Id: I2ca86fb1a0346214ec71d7d6c17b6a121857b01d
Reviewed-on: https://go-review.googlesource.com/114076
Run-TryBot: Austin Clements <austin@google.com>
Reviewed-by: David Chase <drchase@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
- Uncomment tests for AVX512 encoder
- Permit instruction suffixes for x86
- Permit limited reg list [reg-reg] syntax for x86 for multi-source ops
- EVEX encoding support in obj/x86 (Z-cases, asmevex, etc.)
- optabs and ytabs generated by x86avxgen (https://golang.org/cl/107216)
Note: suffix formatting implemented with updated CConv function.
Now arch asm backend should register formatting function by
calling RegisterOpSuffix.
Updates #22779
Change-Id: I076a167ee49582700e058c56ad74e6696710c8c8
Reviewed-on: https://go-review.googlesource.com/113315
Run-TryBot: Iskander Sharipov <iskander.sharipov@intel.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
1. Some incorrect test cases are disabled.
2. Some wrong test cases are corrected.
3. Some new test cases are added.
Change-Id: Ib5d0473d55159f233ddab79f96967eaec7b08597
Reviewed-on: https://go-review.googlesource.com/113736
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
This modifies issafepoint in liveness analysis to report almost every
operation as a safe point. There are four things we don't mark as
safe-points:
1. Runtime code (other than at calls).
2. go:nosplit functions (other than at calls).
3. Instructions between the load of the write barrier-enabled flag and
the write.
4. Instructions leading up to a uintptr -> unsafe.Pointer conversion.
We'll optimize this in later CLs:
name old time/op new time/op delta
Template 185ms ± 2% 190ms ± 2% +2.95% (p=0.000 n=10+10)
Unicode 96.3ms ± 3% 96.4ms ± 1% ~ (p=0.905 n=10+9)
GoTypes 658ms ± 0% 669ms ± 1% +1.72% (p=0.000 n=10+9)
Compiler 3.14s ± 1% 3.18s ± 1% +1.56% (p=0.000 n=9+10)
SSA 7.41s ± 2% 7.59s ± 1% +2.48% (p=0.000 n=9+10)
Flate 126ms ± 1% 128ms ± 1% +2.08% (p=0.000 n=10+10)
GoParser 153ms ± 1% 157ms ± 2% +2.38% (p=0.000 n=10+10)
Reflect 437ms ± 1% 442ms ± 1% +0.98% (p=0.001 n=10+10)
Tar 178ms ± 1% 179ms ± 1% +0.67% (p=0.035 n=10+9)
XML 223ms ± 1% 229ms ± 1% +2.58% (p=0.000 n=10+10)
[Geo mean] 394ms 401ms +1.75%
No effect on binary size because we're not yet emitting these extra
safe points.
For #24543.
Change-Id: I16a1eebb9183cad7cef9d53c0fd21a973cad6859
Reviewed-on: https://go-review.googlesource.com/109348
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: David Chase <drchase@google.com>
Currently liveness only produces a stack map index at each safe point,
so the information is summarized in a map[*ssa.Value]int. We're about
to have both a stack map index and a register map index, so replace
the int with a LivenessIndex type we can extend, and replace the map
with a LivenessMap that we can also change more easily in the future.
This also gives us an easy hook for defining the value that means "not
a safe point".
Passes toolstash -cmp.
For #24543.
Change-Id: Ic4c069839635efed4fd0f603899b80f8be3b56ec
Reviewed-on: https://go-review.googlesource.com/109347
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
The obj package needs to emit the PCDATA to select the entry stack map
before calling morestack. Currently this is copied for every
architecture. Since we're about to change how this works, consolidate
all of these copies into a single helper function.
For #24543.
Change-Id: Ia92d94de78f8e23fd06dba747c43e03e5989f67b
Reviewed-on: https://go-review.googlesource.com/109346
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Currently, range loops over slices and arrays are compiled roughly
like:
for i, x := range s { b }
⇓
for i, _n, _p := 0, len(s), &s[0]; i < _n; i, _p = i+1, _p + unsafe.Sizeof(s[0]) { b }
⇓
i, _n, _p := 0, len(s), &s[0]
goto cond
body:
{ b }
i, _p = i+1, _p + unsafe.Sizeof(s[0])
cond:
if i < _n { goto body } else { goto end }
end:
The problem with this lowering is that _p may temporarily point past
the end of the allocation the moment before the loop terminates. Right
now this isn't a problem because there's never a safe-point during
this brief moment.
We're about to introduce safe-points everywhere, so this bad pointer
is going to be a problem. We could mark the increment as an unsafe
block, but this inhibits reordering opportunities and could result in
infrequent safe-points if the body is short.
Instead, this CL fixes this by changing how we compile range loops to
never produce this past-the-end pointer. It changes the lowering to
roughly:
i, _n, _p := 0, len(s), &s[0]
if i < _n { goto body } else { goto end }
top:
_p += unsafe.Sizeof(s[0])
body:
{ b }
i++
if i < _n { goto top } else { goto end }
end:
Notably, the increment is split into two parts: we increment the index
before checking the condition, but increment the pointer only *after*
the condition check has succeeded.
The implementation builds on the OFORUNTIL construct that was
introduced during the loop preemption experiments, since OFORUNTIL
places the increment and condition after the loop body. To support the
extra "late increment" step, we further define OFORUNTIL's "List"
field to contain the late increment statements. This makes all of this
a relatively small change.
This depends on the improvements to the prove pass in CL 102603. With
the current lowering, bounds-check elimination knows that i < _n in
the body because the body block is dominated by the cond block. In the
new lowering, deriving this fact requires detecting that i < _n on
*both* paths into body and hence is true in body. CL 102603 made prove
able to detect this.
The code size effect of this is minimal. The cmd/go binary on
linux/amd64 increases by 0.17%. Performance-wise, this actually
appears to be a net win, though it's mostly noise:
name old time/op new time/op delta
BinaryTree17-12 2.80s ± 0% 2.61s ± 1% -6.88% (p=0.000 n=20+18)
Fannkuch11-12 2.41s ± 0% 2.42s ± 0% +0.05% (p=0.005 n=20+20)
FmtFprintfEmpty-12 41.6ns ± 5% 41.4ns ± 6% ~ (p=0.765 n=20+19)
FmtFprintfString-12 69.4ns ± 3% 69.3ns ± 1% ~ (p=0.084 n=19+17)
FmtFprintfInt-12 76.1ns ± 1% 77.3ns ± 1% +1.57% (p=0.000 n=19+19)
FmtFprintfIntInt-12 122ns ± 2% 123ns ± 3% +0.95% (p=0.015 n=20+20)
FmtFprintfPrefixedInt-12 153ns ± 2% 151ns ± 3% -1.27% (p=0.013 n=20+20)
FmtFprintfFloat-12 215ns ± 0% 216ns ± 0% +0.47% (p=0.000 n=20+16)
FmtManyArgs-12 486ns ± 1% 498ns ± 0% +2.40% (p=0.000 n=20+17)
GobDecode-12 6.43ms ± 0% 6.50ms ± 0% +1.08% (p=0.000 n=18+19)
GobEncode-12 5.43ms ± 1% 5.47ms ± 0% +0.76% (p=0.000 n=20+20)
Gzip-12 218ms ± 1% 218ms ± 1% ~ (p=0.883 n=20+20)
Gunzip-12 38.8ms ± 0% 38.9ms ± 0% ~ (p=0.644 n=19+19)
HTTPClientServer-12 76.2µs ± 1% 76.4µs ± 2% ~ (p=0.218 n=20+20)
JSONEncode-12 12.2ms ± 0% 12.3ms ± 1% +0.45% (p=0.000 n=19+19)
JSONDecode-12 54.2ms ± 1% 53.3ms ± 0% -1.67% (p=0.000 n=20+20)
Mandelbrot200-12 3.71ms ± 0% 3.71ms ± 0% ~ (p=0.143 n=19+20)
GoParse-12 3.22ms ± 0% 3.19ms ± 1% -0.72% (p=0.000 n=20+20)
RegexpMatchEasy0_32-12 76.7ns ± 1% 75.8ns ± 1% -1.19% (p=0.000 n=20+17)
RegexpMatchEasy0_1K-12 245ns ± 1% 243ns ± 0% -0.72% (p=0.000 n=18+17)
RegexpMatchEasy1_32-12 71.9ns ± 0% 71.7ns ± 1% -0.39% (p=0.006 n=12+18)
RegexpMatchEasy1_1K-12 358ns ± 1% 354ns ± 1% -1.13% (p=0.000 n=20+19)
RegexpMatchMedium_32-12 105ns ± 2% 105ns ± 1% -0.63% (p=0.007 n=19+20)
RegexpMatchMedium_1K-12 31.9µs ± 1% 31.9µs ± 1% ~ (p=1.000 n=17+17)
RegexpMatchHard_32-12 1.51µs ± 1% 1.52µs ± 2% +0.46% (p=0.042 n=18+18)
RegexpMatchHard_1K-12 45.3µs ± 1% 45.5µs ± 2% +0.44% (p=0.029 n=18+19)
Revcomp-12 388ms ± 1% 385ms ± 0% -0.57% (p=0.000 n=19+18)
Template-12 63.0ms ± 1% 63.3ms ± 0% +0.50% (p=0.000 n=19+20)
TimeParse-12 309ns ± 1% 307ns ± 0% -0.62% (p=0.000 n=20+20)
TimeFormat-12 328ns ± 0% 333ns ± 0% +1.35% (p=0.000 n=19+19)
[Geo mean] 47.0µs 46.9µs -0.20%
(https://perf.golang.org/search?q=upload:20180326.1)
For #10958.
For #24543.
Change-Id: Icbd52e711fdbe7938a1fea3e6baca1104b53ac3a
Reviewed-on: https://go-review.googlesource.com/102604
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Reviewed-by: David Chase <drchase@google.com>
Currently, we compile range loops into for loops with the obvious
initialization and update of the index variable. In this form, the
prove pass can see that the body is dominated by an i < len condition,
and findIndVar can detect that i is an induction variable and that
0 <= i < len.
GOEXPERIMENT=preemptibleloops compiles range loops to OFORUNTIL and
we're preparing to unconditionally switch to a variation of this for
#24543. OFORUNTIL moves the increment and condition *after* the body,
which makes the bounds on the index variable much less obvious. With
OFORUNTIL, proving anything about the index variable requires
understanding the phi that joins the index values at the top of the
loop body block.
This interferes with both prove's ability to see that i < len (this is
true on both paths that enter the body, but from two different
conditional checks) and with findIndVar's ability to detect the
induction pattern.
Fix this by teaching prove to detect that the index in the pattern
constructed by OFORUNTIL is an induction variable and add both bounds
to the facts table. Currently this is done separately from findIndVar
because it depends on prove's factsTable, while findIndVar runs before
visiting blocks and building the factsTable.
Without any GOEXPERIMENT, this has no effect on std or cmd. However,
with GOEXPERIMENT=preemptibleloops, this change becomes necessary to
prove 90 conditions in std and cmd.
Change-Id: Ic025d669f81b53426309da5a6e8010e5ccaf4f49
Reviewed-on: https://go-review.googlesource.com/102603
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
