Re: cvs commit: src/lib/libthread_xu Makefile pthread.map src/lib/libthread_xu/arch Makefile.inc src/lib/libthread_xu/arch/alpha Makefile.inc src/lib/libthread_xu/arch/alpha/alpha pthread_md.c src/lib/libthread_xu/arch/alpha/include pthread_md.h src/lib/libthread_xu/arch/amd64 ...

[User Davidxu <xuyifeng@xxxxxxxxx>]

Matthew Dillon wrote:
> :The management thread should be avoided, that introduces complex into
> :userland, I don't want to go old evil linuxthread way.
> :I only need one flag to indicate that the thread now executes in
> :kernel(pthread_exit() calls _exit()), so its userland resource is no 
> :longer needed, e.g. its userland stack and user thread control block, so 
> :other threads can reuse them when creating new thread or recycle it at 
> :sometime later, but without this flag, there is always a race condition, 
> :it is not safe to reuse it because that thread may be still using the 
> :stack, although it called _exit(), but that does not say it is not using 
> :the stack! calling _exit() pushes return address on stack, and it 
> :destroys another thread's memory if the stack was reused by other 
> :threads very quickly.
> :In detail, I want following feature:
> :a syscall allows userland to set an userland address, for example:
> :__sys_set_thread_exit_addr(int *addr), it remembers the addr
> :in kernel thread structure, if the thread calls _exit(),
> :in kernel, kernel code writes a value to addr, that's the only
> :thing I need to do userland garbage collection.
> :In this way I can avoid complex of management thread.
> :
> :Cheers,
> :David Xu
>
>     Hmm.  Well, I really dislike adding single-purpose system calls.  If I
>     do this we need to try to make the system call a bit more general-purpose.
>     For example, we could use it to detect abnormal thread termination as
>     well as deal with the stack resource race in the normal thread-exit case.
>
>     So something like this:
>
>     __sys_set_exit_data(int *addr, int data);
>
>     Defined to be:
>
>     "When a process exits the system will store the specified data into the
>     specified address.  Only one address may be registered per process.
>     The feature can be disabled by passing addr = NULL.
>
>     In an rfork/thread baesd system this may be used to detect abnormal 
>     process exits and to synchronize stack resource reuse with normal thread
>     termination."
>
>     How's that?

This is fine to me, did you consider how to exit whole process not
just a single thread ? right now, I am using _exit(), but it only
exits current thread, other threads can not be shutdown.

> 					-Matt
> 					Matthew Dillon 
> 					<dillon@xxxxxxxxxxxxx>