Re: Loaded footgun open_datasync on Windows

On Tue, Jun 05, 2018 at 04:15:00PM +0200, Laurenz Albe wrote:

> --- a/src/bin/pg_test_fsync/pg_test_fsync.c
> +++ b/src/bin/pg_test_fsync/pg_test_fsync.c
> @@ -3,6 +3,8 @@
>   *           tests all supported fsync() methods
>   */
> 
> +/* we have to include c.h first so that pgwin32_open is used on Windows */
> +#include "c.h"
>  #include "postgres_fe.h"
> 
>  #include <sys/stat.h>

Ouch.  Including directly c.h as you do here is against project policy
code.  See recent commit a72f0365 for example.  pgwin32_open() is
visibly able to handle frontend code if I read this code correctly, so
could we just remove the "#ifndef FRONTEND" restriction? 

It could be
risky for existing callers of open() for tool maintainers, or on the
contrary people could welcome a wrapper of open() which is
concurrent-safe in their own tools.