all: multiple packages fail go test when not in short mode #24464

alexd765 · 2018-03-20T20:43:29Z

What version of Go are you using (`go version`)?

go version devel +be371edd67 Tue Mar 20 19:38:06 2018 +0000 linux/amd64

What operating system and processor architecture are you using (`go env`)?

linux/amd64

What did you do?

Test all packages via all.bash (in short mode) and then via go test ./... (not in short mode)

~/gotip/src$ ./all.bash && go test ./...

Inspired by #24436

What did you expect to see?

All tests pass (in short mode)
All tests pass (not in short mode)

What did you see instead?

All tests pass (in short mode)
Multiple packages fail go test (not in short mode)

FAIL cmd/compile/internal/ssa (discussed here)
FAIL cmd/doc (fixed in cl 101836)
FAIL cmd/go (#24436)
FAIL cmd/gofmt (#24472)
FAIL go/types (fixed in cl 101815)
FAIL net

Edit: i have added links to the fixes and other discussions so far

The text was updated successfully, but these errors were encountered:

andybons · 2018-03-20T21:53:09Z

@rsc

gopherbot · 2018-03-21T10:57:56Z

Change https://golang.org/cl/101815 mentions this issue: cmd/trace: remove unused variable in tests

Unused variables in closures are currently not diagnosed by the compiler (this is Issue #3059), while go/types catches them. One unused variable in the cmd/trace tests is causing the go/types test that typechecks the whole standard library to fail: FAIL: TestStdlib (8.05s) stdlib_test.go:223: cmd/trace/annotations_test.go:241:6: gcTime declared but not used FAIL Remove it. Updates #24464 Change-Id: I0f1b9db6ae1f0130616ee649bdbfdc91e38d2184 Reviewed-on: https://go-review.googlesource.com/101815 Reviewed-by: Daniel Martí <mvdan@mvdan.cc>

ALTree · 2018-03-21T11:11:49Z

go/types failure fixed in 041c5d8.

gopherbot · 2018-03-21T12:25:27Z

Change https://golang.org/cl/101835 mentions this issue: test: fmt fixedbugs/issue22662.go

mvdan · 2018-03-21T12:27:22Z

The failure in cmd/doc is because it may fail depending on your GOPATH. For example:

$ go test
2018/03/21 12:26:36 no buildable Go source files in /home/mvdan/go/land/src/github.com/gopherjs/gopherjs/compiler/natives/src/crypto/rand
exit status 1
FAIL    cmd/doc 0.119s
$ GOPATH= go test
PASS
ok      cmd/doc 1.857s

The tests should be made more robust. Will give it a go, but there are many ways to fix this.

gopherbot · 2018-03-21T12:41:19Z

Change https://golang.org/cl/101836 mentions this issue: cmd/doc: use empty GOPATH when running the tests

ALTree · 2018-03-21T13:05:41Z

The cmd/gofmt failure is caused by gofmt not being idempotent on certain programs. I opened #24472 about this, with a small reproducer.

Otherwise, a populated GOPATH might result in failures such as: $ go test [...] no buildable Go source files in [...]/gopherjs/compiler/natives/src/crypto/rand exit status 1 Move the initialization of the dirs walker out of the init func, so that we can control its behavior in the tests. Updates #24464. Change-Id: I4b26a7d3d6809bdd8e9b6b0556d566e7855f80fe Reviewed-on: https://go-review.googlesource.com/101836 Run-TryBot: Daniel Martí <mvdan@mvdan.cc> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>

randall77 · 2018-03-21T16:51:29Z

I can't get cmd/compile/internal/ssa to fail.

$ git checkout be371edd677abe6de310c9ffc225b9e8b052d2b8
$ ./make.bash
Building Go cmd/dist using /Users/khr/go1.4.
Building Go toolchain1 using /Users/khr/go1.4.
Building Go bootstrap cmd/go (go_bootstrap) using Go toolchain1.
Building Go toolchain2 using go_bootstrap and Go toolchain1.
Building Go toolchain3 using go_bootstrap and Go toolchain2.
Building packages and commands for darwin/amd64.
---
Installed Go for darwin/amd64 in /Users/khr/sandbox/readonly
Installed commands in /Users/khr/sandbox/readonly/bin
$ ../bin/go test cmd/compile/internal/ssa/...
ok  	cmd/compile/internal/ssa	71.866s

(same thing on linux)

ALTree · 2018-03-21T16:55:34Z

I can (linux/amd64), here's the log:

--- FAIL: TestNexting (7.94s)
    --- FAIL: TestNexting/gdb-dbg-hist (2.37s)
    	debug_test.go:234: step/next histories differ, diff=
    		--- testdata/hist.gdb-dbg.nexts	2018-03-19 11:37:08.131407088 +0100
    		+++ /tmp/debug_test231459976/test-hist.gdb-dbg.nexts	2018-03-21 17:54:58.587602381 +0100
    		@@ -9,7 +9,7 @@
    		 61:		sink = dx + dy            //gdb-opt=(dx,dy)
    		 63:		hist := make([]int, 7)                                //gdb-opt=(dx/O,dy/O) // TODO sink is missing if this code is in 'test' instead of 'main'
    		 64:		var reader io.Reader = strings.NewReader(cannedInput) //gdb-dbg=(hist/A) // TODO cannedInput/A is missing if this code is in 'test' instead of 'main'
    		-hist =  []int = {0, 0, 0, 0, 0, 0, 0}
    		+hist = {array = <A>, len = 7, cap = 7}
    		 65:		if len(os.Args) > 1 {
    		 73:		scanner := bufio.NewScanner(reader)
    		 74:		for scanner.Scan() { //gdb-opt=(scanner/A)
    --- FAIL: TestNexting/gdb-opt-hist (0.95s)
    	debug_test.go:234: step/next histories differ, diff=
    		--- testdata/hist.gdb-opt.nexts	2018-03-19 11:37:08.139407145 +0100
    		+++ /tmp/debug_test577668668/test-hist.gdb-opt.nexts	2018-03-21 17:55:03.655623768 +0100
    		@@ -25,7 +25,7 @@
    		 76:			i, err := strconv.ParseInt(s, 10, 64)
    		 77:			if err != nil { //gdb-dbg=(i) //gdb-opt=(err,hist,i)
    		 err = {tab = 0x0, data = 0x0}
    		-hist =  []int = {0, 0, 0, 0, 0, 0, 0}
    		+hist = {array = 0xc00003be98, len = 7, cap = 7}
    		 i = 1
    		 81:			hist = ensure(int(i), hist)
    		 82:			hist[int(i)]++
    		@@ -35,7 +35,7 @@
    		 76:			i, err := strconv.ParseInt(s, 10, 64)
    		 77:			if err != nil { //gdb-dbg=(i) //gdb-opt=(err,hist,i)
    		 err = {tab = 0x0, data = 0x0}
    		-hist =  []int = {0, 1, 0, 0, 0, 0, 0}
    		+hist = {array = 0xc00003be98, len = 7, cap = 7}
    		 i = 1
    		 81:			hist = ensure(int(i), hist)
    		 82:			hist[int(i)]++
    		@@ -45,7 +45,7 @@
    		 76:			i, err := strconv.ParseInt(s, 10, 64)
    		 77:			if err != nil { //gdb-dbg=(i) //gdb-opt=(err,hist,i)
    		 err = {tab = 0x0, data = 0x0}
    		-hist =  []int = {0, 2, 0, 0, 0, 0, 0}
    		+hist = {array = 0xc00003be98, len = 7, cap = 7}
    		 i = 1
    		 81:			hist = ensure(int(i), hist)
    		 82:			hist[int(i)]++
    		@@ -55,7 +55,7 @@
    		 76:			i, err := strconv.ParseInt(s, 10, 64)
    		 77:			if err != nil { //gdb-dbg=(i) //gdb-opt=(err,hist,i)
    		 err = {tab = 0x0, data = 0x0}
    		-hist =  []int = {0, 3, 0, 0, 0, 0, 0}
    		+hist = {array = 0xc00003be98, len = 7, cap = 7}
    		 i = 2
    		 81:			hist = ensure(int(i), hist)
    		 82:			hist[int(i)]++
    		@@ -65,7 +65,7 @@
    		 76:			i, err := strconv.ParseInt(s, 10, 64)
    		 77:			if err != nil { //gdb-dbg=(i) //gdb-opt=(err,hist,i)
    		 err = {tab = 0x0, data = 0x0}
    		-hist =  []int = {0, 3, 1, 0, 0, 0, 0}
    		+hist = {array = 0xc00003be98, len = 7, cap = 7}
    		 i = 2
    		 81:			hist = ensure(int(i), hist)
    		 82:			hist[int(i)]++
    		@@ -75,7 +75,7 @@
    		 76:			i, err := strconv.ParseInt(s, 10, 64)
    		 77:			if err != nil { //gdb-dbg=(i) //gdb-opt=(err,hist,i)
    		 err = {tab = 0x0, data = 0x0}
    		-hist =  []int = {0, 3, 2, 0, 0, 0, 0}
    		+hist = {array = 0xc00003be98, len = 7, cap = 7}
    		 i = 2
    		 81:			hist = ensure(int(i), hist)
    		 82:			hist[int(i)]++
    		@@ -85,7 +85,7 @@
    		 76:			i, err := strconv.ParseInt(s, 10, 64)
    		 77:			if err != nil { //gdb-dbg=(i) //gdb-opt=(err,hist,i)
    		 err = {tab = 0x0, data = 0x0}
    		-hist =  []int = {0, 3, 3, 0, 0, 0, 0}
    		+hist = {array = 0xc00003be98, len = 7, cap = 7}
    		 i = 4
    		 81:			hist = ensure(int(i), hist)
    		 82:			hist[int(i)]++
    		@@ -95,7 +95,7 @@
    		 76:			i, err := strconv.ParseInt(s, 10, 64)
    		 77:			if err != nil { //gdb-dbg=(i) //gdb-opt=(err,hist,i)
    		 err = {tab = 0x0, data = 0x0}
    		-hist =  []int = {0, 3, 3, 0, 1, 0, 0}
    		+hist = {array = 0xc00003be98, len = 7, cap = 7}
    		 i = 4
    		 81:			hist = ensure(int(i), hist)
    		 82:			hist[int(i)]++
    		@@ -105,7 +105,7 @@
    		 76:			i, err := strconv.ParseInt(s, 10, 64)
    		 77:			if err != nil { //gdb-dbg=(i) //gdb-opt=(err,hist,i)
    		 err = {tab = 0x0, data = 0x0}
    		-hist =  []int = {0, 3, 3, 0, 2, 0, 0}
    		+hist = {array = 0xc00003be98, len = 7, cap = 7}
    		 i = 5
    		 81:			hist = ensure(int(i), hist)
    		 82:			hist[int(i)]++
FAIL
exit status 1
FAIL	cmd/compile/internal/ssa	93.549s

randall77 · 2018-03-21T17:44:39Z

Hmm, might be a gdb version issue. I'm running 7.9.
@dr2chase

mvdan · 2018-03-21T17:45:06Z

I get the same failure, gdb version 8.1.

ALTree · 2018-03-21T17:47:32Z

My system comes with gdb 7.12

dr2chase · 2018-03-21T18:35:05Z

Is that gdb version or presence/absence of Python scripts?
I'll try to figure this out, I could at least make it more picky about gdb version if that is the case.

dr2chase · 2018-03-21T19:59:12Z

I have gdb versions 7.9 and 8.0.1, seems to work for me.
I will try to figure out how to install a version that triggers the wrong printing (which I have clearly seen in the past, because of the "/A" following hist).

I am not super-fluent in Python, but this from runtime-gdb.py seems semi-relevant:

class SliceTypePrinter:
	"Pretty print slices."

	pattern = re.compile(r'^struct \[\]')

	def __init__(self, val):
		self.val = val

	def display_hint(self):
		return 'array'

	def to_string(self):
		return str(self.val.type)[6:]  # skip 'struct '

	def children(self):
		sval = SliceValue(self.val)
		if sval.len > sval.cap:
			return
		for idx, item in enumerate(sval):
			yield ('[{0}]'.format(idx), item)

alexd765 · 2018-03-21T22:02:40Z

i have added links to the fixes and other discussions so far to the main post

alexd765 · 2018-03-21T22:06:21Z

cmd/compile/internal/ssa fails for me with gdb 8.0.1, so the gdb version doesn't seem to be the issue.

josharian · 2018-03-22T22:30:38Z

The underlying problem here is #12508

dr2chase · 2018-03-23T15:52:11Z

@alexd765 the SliceTypePrinter code is really old, so that's probably the same.
My Python versions are 2.7.14 and 2.7.13. Yours?

mvdan · 2018-03-23T16:28:26Z

@dr2chase I can reproduce it and I only have versions 2.7.14 and 3.6.4 installed. I'm running Arch Linux, so I would assume that any somewhat recent Linux system should have this test failure.

I tried playing with that runtime-gdb.py file, but not even emptying the file changes the behaviour of the test.

alexd765 · 2018-03-23T18:30:34Z

I am on Ubuntu 17.10 with python 2.7.14

dr2chase · 2018-03-23T18:38:16Z

Except that I am failing to reproduce the bug on Linux....
I am learning a little Python.

alexd765 · 2018-03-23T19:29:55Z

@dr2chase I also started to look into this a little bit:

First observation: if i force the test to use delve with ~/gotip/src/cmd/compile/internal/ssa$ go test -d -v it passes.

Then i saw that there is an option to regenerate the testdata with ~/gotip/src/cmd/compile/internal/ssa$ go test -u. It changed some things in 2 testdata files and after that the tests are passing.

I can send the changes as a cl if that helps

dr2chase · 2018-03-23T19:44:11Z

Mucking with the testdata files w/o understanding why we have this difference isn't going to help -- you'll break it for me, for example.

Also, I have several CLs stacked up that will mess with the testdata in a big way, alleged positive. (If you actually look at what's recorded there, there's plenty of room for improvement.)

dr2chase · 2018-03-23T21:19:23Z

Perhaps it is not autoloading runtime-gdb.py?
Are there any warnings at the beginning about auto-load and safe-path?

I definitely need to modify the test to be better about this.

gopherbot · 2018-03-26T17:42:05Z

Change https://golang.org/cl/102598 mentions this issue: cmd/compile: invoke gdb more carefully in ssa/debug_test.go

Gdb can be sensitive to contents of .gdbinit, and to run this test properly needs to have runtime/runtime-gdb.py on the auto load safe path. Therefore, turn off .gdbinit loading and explicitly add $GOROOT/runtime to the safe load path. This should make ssa/debug_test.go run more consistently. Updates #24464. Change-Id: I63ed17c032cb3773048713ce51fca3a3f86e79b6 Reviewed-on: https://go-review.googlesource.com/102598 Run-TryBot: David Chase <drchase@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>

mvdan · 2018-03-27T15:12:01Z

@dr2chase all the TestNexting subtests are passing on my machine, so we can fairly safely assume that it's fixed for @alexd765 too :)

Out of the initial 5 issues, 3 have been fixed and 2 have been split into separate issues. So it seems like there's nothing else to do here. Thanks again @alexd765 for reporting them.

dr2chase · 2018-03-27T15:37:39Z

Thanks for verifying. I wasn't 100% sure, but I was hopeful.

alexd765 · 2018-03-27T17:58:13Z

yeah i can confirm that cmd/compile/internal/ssa is fixed for me as well

andybons added Testing An issue that has been verified to require only test changes, not just a test failure. NeedsInvestigation Someone must examine and confirm this is a valid issue and not a duplicate of an existing one. labels Mar 20, 2018

andybons added this to the Unplanned milestone Mar 20, 2018

josharian mentioned this issue Mar 22, 2018

build: run all tests on some builder #12508

Closed

mvdan closed this as completed Mar 27, 2018

golang locked and limited conversation to collaborators Mar 27, 2019

gopherbot added the FrozenDueToAge label Mar 27, 2019

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

all: multiple packages fail go test when not in short mode #24464

all: multiple packages fail go test when not in short mode #24464

alexd765 commented Mar 20, 2018 •

edited

andybons commented Mar 20, 2018

gopherbot commented Mar 21, 2018

ALTree commented Mar 21, 2018

gopherbot commented Mar 21, 2018

mvdan commented Mar 21, 2018 •

edited

gopherbot commented Mar 21, 2018

ALTree commented Mar 21, 2018

randall77 commented Mar 21, 2018

ALTree commented Mar 21, 2018

randall77 commented Mar 21, 2018

mvdan commented Mar 21, 2018

ALTree commented Mar 21, 2018

dr2chase commented Mar 21, 2018

dr2chase commented Mar 21, 2018

alexd765 commented Mar 21, 2018

alexd765 commented Mar 21, 2018

josharian commented Mar 22, 2018

dr2chase commented Mar 23, 2018

mvdan commented Mar 23, 2018

alexd765 commented Mar 23, 2018

dr2chase commented Mar 23, 2018

alexd765 commented Mar 23, 2018

dr2chase commented Mar 23, 2018

dr2chase commented Mar 23, 2018

gopherbot commented Mar 26, 2018

mvdan commented Mar 27, 2018

dr2chase commented Mar 27, 2018

alexd765 commented Mar 27, 2018

all: multiple packages fail go test when not in short mode #24464

all: multiple packages fail go test when not in short mode #24464

Comments

alexd765 commented Mar 20, 2018 • edited

What version of Go are you using (go version)?

What operating system and processor architecture are you using (go env)?

What did you do?

What did you expect to see?

What did you see instead?

andybons commented Mar 20, 2018

gopherbot commented Mar 21, 2018

ALTree commented Mar 21, 2018

gopherbot commented Mar 21, 2018

mvdan commented Mar 21, 2018 • edited

gopherbot commented Mar 21, 2018

ALTree commented Mar 21, 2018

randall77 commented Mar 21, 2018

ALTree commented Mar 21, 2018

randall77 commented Mar 21, 2018

mvdan commented Mar 21, 2018

ALTree commented Mar 21, 2018

dr2chase commented Mar 21, 2018

dr2chase commented Mar 21, 2018

alexd765 commented Mar 21, 2018

alexd765 commented Mar 21, 2018

josharian commented Mar 22, 2018

dr2chase commented Mar 23, 2018

mvdan commented Mar 23, 2018

alexd765 commented Mar 23, 2018

dr2chase commented Mar 23, 2018

alexd765 commented Mar 23, 2018

dr2chase commented Mar 23, 2018

dr2chase commented Mar 23, 2018

gopherbot commented Mar 26, 2018

mvdan commented Mar 27, 2018

dr2chase commented Mar 27, 2018

alexd765 commented Mar 27, 2018

alexd765 commented Mar 20, 2018 •

edited

What version of Go are you using (`go version`)?

What operating system and processor architecture are you using (`go env`)?

mvdan commented Mar 21, 2018 •

edited