i would've searched for this, but the 'newsguy' news service doesnt seem to have a search function.  bleh.

can anyone post a link to a "d threads primer"?  i'm having trouble understanding what the thread methed 'run' wants from me.  i'm a c coder, and i understand function pointers, but..well, here is a code snippet:

void myfunc() {
printf("hi\n");
}
Thread mythread;
mythread.run(&myfunc);

thread.d(8): function std.thread.Thread.run () does not match argument types
(void(*)())
thread.d(8): Error: expected 0 arguments, not 1


but what gives?  the phobos docs say:
int run(void* p);
Entry point for a thread. If not overridden, it calls the function pointer fp
and argument arg passed in the constructor, or the delegate dg. The return value
is the thread exit code, which is normally 0.

anyway, am i being lame, or is it d?

"prefetch" <prefetch_member@pathlink.com> wrote in message news:ddqno7$1ebq$1@digitaldaemon.com...
>i would've searched for this, but the 'newsguy' news service doesnt seem to have
> a search function.  bleh.
>
> can anyone post a link to a "d threads primer"?  i'm having trouble
> understanding what the thread methed 'run' wants from me.  i'm a c coder,
> and i
> understand function pointers, but..well, here is a code snippet:
>
> void myfunc() {
> printf("hi\n");
> }
> Thread mythread;
> mythread.run(&myfunc);
>
> thread.d(8): function std.thread.Thread.run () does not match argument
> types
> (void(*)())
> thread.d(8): Error: expected 0 arguments, not 1

you probably want
 Thread mythread = new Thread(&myfunc);
 mythread.start();
Four points to notice:
1) you need to 'new' your Thread object
2) don't call run() but instead pass the worker delegate to the Thread
constructor
3) the new thread won't actually start until you call "start".
4) the phobos doc for the Thread constructors is wierd. I can't figure out
what it's trying to say :-P

> but what gives?  the phobos docs say:
> int run(void* p);
> Entry point for a thread. If not overridden, it calls the function pointer
> fp
> and argument arg passed in the constructor, or the delegate dg. The return
> value
> is the thread exit code, which is normally 0.
>
> anyway, am i being lame, or is it d?
>
> 

ben, thanks for the reply.  very helpful.
also, i managed to find this quick primer for future reference:

http://robertmuench.de/notes/d/

cheers.

In article <ddqotb$1fb6$1@digitaldaemon.com>, Ben Hinkle says...
>
>
>"prefetch" <prefetch_member@pathlink.com> wrote in message news:ddqno7$1ebq$1@digitaldaemon.com...
>>i would've searched for this, but the 'newsguy' news service doesnt seem to have
>> a search function.  bleh.
>>
>> can anyone post a link to a "d threads primer"?  i'm having trouble
>> understanding what the thread methed 'run' wants from me.  i'm a c coder,
>> and i
>> understand function pointers, but..well, here is a code snippet:
>>
>> void myfunc() {
>> printf("hi\n");
>> }
>> Thread mythread;
>> mythread.run(&myfunc);
>>
>> thread.d(8): function std.thread.Thread.run () does not match argument
>> types
>> (void(*)())
>> thread.d(8): Error: expected 0 arguments, not 1
>
>you probably want
> Thread mythread = new Thread(&myfunc);
> mythread.start();
>Four points to notice:
>1) you need to 'new' your Thread object
>2) don't call run() but instead pass the worker delegate to the Thread
>constructor
>3) the new thread won't actually start until you call "start".
>4) the phobos doc for the Thread constructors is wierd. I can't figure out
>what it's trying to say :-P
>
>> but what gives?  the phobos docs say:
>> int run(void* p);
>> Entry point for a thread. If not overridden, it calls the function pointer
>> fp
>> and argument arg passed in the constructor, or the delegate dg. The return
>> value
>> is the thread exit code, which is normally 0.
>>
>> anyway, am i being lame, or is it d?
>>
>> 
>
>

oops - i spoke too soon.  robermuench's code doesn't compile (i'm working the
bugs out now) and also, ben - your suggestion doesn't seem to work out (though
it makes more sense than what i was doing..)

could i trouble you to compile it and check?

In article <ddqpup$1g5n$1@digitaldaemon.com>, prefetch says...
>
>ben, thanks for the reply.  very helpful.
>also, i managed to find this quick primer for future reference:
>
>http://robertmuench.de/notes/d/
>
>cheers.
>
>In article <ddqotb$1fb6$1@digitaldaemon.com>, Ben Hinkle says...
>>
>>
>>"prefetch" <prefetch_member@pathlink.com> wrote in message news:ddqno7$1ebq$1@digitaldaemon.com...
>>>i would've searched for this, but the 'newsguy' news service doesnt seem to have
>>> a search function.  bleh.
>>>
>>> can anyone post a link to a "d threads primer"?  i'm having trouble
>>> understanding what the thread methed 'run' wants from me.  i'm a c coder,
>>> and i
>>> understand function pointers, but..well, here is a code snippet:
>>>
>>> void myfunc() {
>>> printf("hi\n");
>>> }
>>> Thread mythread;
>>> mythread.run(&myfunc);
>>>
>>> thread.d(8): function std.thread.Thread.run () does not match argument
>>> types
>>> (void(*)())
>>> thread.d(8): Error: expected 0 arguments, not 1
>>
>>you probably want
>> Thread mythread = new Thread(&myfunc);
>> mythread.start();
>>Four points to notice:
>>1) you need to 'new' your Thread object
>>2) don't call run() but instead pass the worker delegate to the Thread
>>constructor
>>3) the new thread won't actually start until you call "start".
>>4) the phobos doc for the Thread constructors is wierd. I can't figure out
>>what it's trying to say :-P
>>
>>> but what gives?  the phobos docs say:
>>> int run(void* p);
>>> Entry point for a thread. If not overridden, it calls the function pointer
>>> fp
>>> and argument arg passed in the constructor, or the delegate dg. The return
>>> value
>>> is the thread exit code, which is normally 0.
>>>
>>> anyway, am i being lame, or is it d?
>>>
>>> 
>>
>>
>
>

"prefetch" <prefetch_member@pathlink.com> wrote in message news:ddqrj4$1hi1$1@digitaldaemon.com...
> oops - i spoke too soon.  robermuench's code doesn't compile (i'm working
> the
> bugs out now) and also, ben - your suggestion doesn't seem to work out
> (though
> it makes more sense than what i was doing..)
>
> could i trouble you to compile it and check?

Did you try something like this?

import std.thread;
int main() {

 int myfunc() {
     printf("hi\n");
     return 0; // 0 means normal return
 }

 Thread mythread = new Thread(&myfunc);
 mythread.start();

 for (int k = 0; k<100000; k++)
     Thread.yield(); // give the other thread a chance

 return 0;
}

thanks ben.  that works. ;-)

it'll help me get started.  thanks again.

In article <ddqtcd$1j77$1@digitaldaemon.com>, Ben Hinkle says...
>
>
>"prefetch" <prefetch_member@pathlink.com> wrote in message news:ddqrj4$1hi1$1@digitaldaemon.com...
>> oops - i spoke too soon.  robermuench's code doesn't compile (i'm working
>> the
>> bugs out now) and also, ben - your suggestion doesn't seem to work out
>> (though
>> it makes more sense than what i was doing..)
>>
>> could i trouble you to compile it and check?
>
>Did you try something like this?
>
>import std.thread;
>int main() {
>
> int myfunc() {
>     printf("hi\n");
>     return 0; // 0 means normal return
> }
>
> Thread mythread = new Thread(&myfunc);
> mythread.start();
>
> for (int k = 0; k<100000; k++)
>     Thread.yield(); // give the other thread a chance
>
> return 0;
>}
>
>
>