How to check whether a variable is unreachable?

[Cholerae Hu]

https://play.golang.org/p/WJTH-lUukeb

Do I need to uncomment line 23? IMHO, variable data and ds will not be used after line 16, so they can be collected in runtime.GC. But running it with gccheckmark=1 and clobberfree=1 doesn't show any problems.

Are there any documents about the details or behavior of liveness analysis? 

[Ian Lance Taylor]

I don't see any reason why that program should need runtime.KeepAlive.

The local variable sh will point to data (aka ds), so the object will
stay alive. This is not different than p := &data. After that point,
if data is no longer mentioned, then the variable data is no longer
live, but the value that p points to will remain alive.

Ian

[Cholerae Hu]

Thanks, I see. How about this one? I use a temp variable to store sh.Data, sh/ds/data are all no longer mentioned at the point when unsafe.Slice is called. Do I need to use runtime.KeepAlive? 

package main

import (

"fmt"

"reflect"

"runtime"

"unsafe"

)

var b []byte

func f(data interface{}) {

switch ds := data.(type) {

case []int32:

sh := (*reflect.SliceHeader)(unsafe.Pointer(&ds))

l := len(ds)

d := sh.Data

runtime.GC()

b = unsafe.Slice((*byte)(unsafe.Pointer(d)), l*4)

}

// runtime.KeepAlive(data)

}

func main() {

f([]int32{1, 2})

runtime.GC()

fmt.Println(b)

}

[jake...@gmail.com]

I'm pretty sure new example is not valid Go code to begin with. You violate the rules on converting from uintptr to unsafe.Pointer. If you read the rules at https://pkg.go.dev/unsafe#Pointer, a unitptr can not generally be converted to an unsafe.Pointer. Since this code is not valid, the question is pretty much moot.

[peterGo]

$ go vet ch.go
# command-line-arguments
./ch.go:19:28: possible misuse of unsafe.Pointer

$

https://play.golang.org/p/n9A_sR1yuXC

Peter

On Thursday, September 23, 2021 at 2:23:47 AM UTC-4 chole...@gmail.com wrote:

[peterGo]

Package runtime

https://pkg.go.dev/runtime

Documentation

func KeepAlive

Note: KeepAlive should only be used to prevent finalizers from running prematurely.

Peter

On Thursday, September 23, 2021 at 2:23:47 AM UTC-4 chole...@gmail.com wrote:

[Cholerae Hu]

I get it. Thanks for all your responses!