Re: [PATCH v12 1/2] win32-pthread: prepare pthread.c for change by formatting

[Junio C Hamano <gitster@pobox.com>]

"Seija Kijin via GitGitGadget" <gitgitgadget@gmail.com> writes:

>  int pthread_create(pthread_t *thread, const void *unused,
> -		   void *(*start_routine)(void*), void *arg)
> +		   void *(*start_routine)(void *), void *arg)

Good.

>  {
>  	thread->arg = arg;
>  	thread->start_routine = start_routine;
> -	thread->handle = (HANDLE)
> -		_beginthreadex(NULL, 0, win32_start_routine, thread, 0, NULL);
> +	thread->handle = (HANDLE)_beginthreadex(NULL, 0, win32_start_routine,
> +						thread, 0, NULL);

I would say this is good either way---preimage may slightly be
easier to read, but it is not worth a patch churn.

> -	DWORD result = WaitForSingleObject(thread->handle, INFINITE);
> -	switch (result) {
> -		case WAIT_OBJECT_0:
> -			if (value_ptr)
> -				*value_ptr = thread->arg;
> -			return 0;
> -		case WAIT_ABANDONED:
> -			return EINVAL;
> -		default:
> -			return err_win_to_posix(GetLastError());
> +	switch (WaitForSingleObject(thread->handle, INFINITE)) {
> +	case WAIT_OBJECT_0:
> +		if (value_ptr)
> +			*value_ptr = thread->arg;
> +		return 0;
> +	case WAIT_ABANDONED:
> +		return EINVAL;
> +	default:
> +		return err_win_to_posix(GetLastError());
>  	}
>  }

Loss of a single-use variable is, strictly speaking, a "while at it"
change that has nothing to do with "reformatting", but otherwise the
result of the hunk looks good.

Will queue.

Thanks.



P.S.  An address that is designed to bounce or sent to black hole
appears on the CC: of the original e-mail message.  Can you stop
doing that (or is that what GGG does without user's control)?

  Rose <83477269+AtariDreams@users.noreply.github.com>