thank you. now, I understand.
Ian Lance Taylor於 2018年1月31日星期三 UTC+8下午1時29分09秒寫道：
>
> On Tue, Jan 30, 2018 at 9:13 PM, hǎi sū <uor...@gmail.com <javascript:>>
> wrote:
> > the code is below.
> > ```
> > package main
> >
> > import "math/rand"
> > import "fmt"
> >
> > /*
> > #include <stdio.h>
> > #include <stdlib.h>
> >
> > void t(const short *buf, size_t buf_size) {
> > printf("in t.");
> > printf("0: %d ", buf[0]);
> > printf("buf_size-1: %d\n", buf[buf_size-1]);
> > }
> >
> > */
> > import "C"
> >
> > func main() {
> > for {
> > a := make([]byte, 178209)
> > rand.Read(a)
> > fmt.Println(a[0], a[len(a)-1])
> > aPointer := C.CBytes(a)
> > C.t((*C.short)(aPointer), C.size_t(len(a)))
> > C.free(aPointer)
> > }
> > }
> > ```
>
> len(a) is the length in bytes, which is 178209. But the C code is
> using that length to index into a short*, meaning that it is looking
> at byte 356418, which is far past the end of the array. The C code
> should take unsigned char * or you should divide len(a) by 2 or
> something.
>
> Ian
>
--
You received this message because you are subscribed to the Google Groups
"golang-nuts" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to golang-nuts+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
