[PATCH] Cygwin: exec: check execute bit prior to evaluating script

[Ken Brown]

On 8/6/2019 4:53 AM, Corinna Vinschen wrote:
> From: Corinna Vinschen <corinna-cygwin@cygwin.com>
> 
> When the exec family of functions is called for a script-like
> file, the av::setup function handles the exec[vl]p case as
> well.  The execve case for files not starting with a she-bang
> is handled first by returning ENOEXEC.  Only after that, the
> file's executability is checked.
> 
> This leads to the problem that ENOEXEC is returned for non-executable
> files as well.  A calling shell interprets this as a file it should try
> to run as script.  This is not desired for non-executable files.
> 
> Fix this problem by checking the file for executability first.  Only
> after that, follow the other potential code paths.
> ---
>   winsup/cygwin/spawn.cc | 12 ++++++------
>   1 file changed, 6 insertions(+), 6 deletions(-)
> 
> diff --git a/winsup/cygwin/spawn.cc b/winsup/cygwin/spawn.cc
> index 7f7af4449da1..d95772802f8f 100644
> --- a/winsup/cygwin/spawn.cc
> +++ b/winsup/cygwin/spawn.cc
> @@ -1172,6 +1172,12 @@ av::setup (const char *prog_arg, path_conv& real_path, const char *ext,
>   	  }
>   	UnmapViewOfFile (buf);
>     just_shell:
> +	/* Check if script is executable.  Otherwise we start non-executable
> +	   scripts successfully, which is incorrect behaviour. */
> +	if (real_path.has_acls ()
> +	    && check_file_access (real_path, X_OK, true) < 0)
> +	  return -1;	/* errno is already set. */
> +
>   	if (!pgm)
>   	  {
>   	    if (!p_type_exec)
> @@ -1188,12 +1194,6 @@ av::setup (const char *prog_arg, path_conv& real_path, const char *ext,
>   	    arg1 = NULL;
>   	  }
>   
> -	/* Check if script is executable.  Otherwise we start non-executable
> -	   scripts successfully, which is incorrect behaviour. */
> -	if (real_path.has_acls ()
> -	    && check_file_access (real_path, X_OK, true) < 0)
> -	  return -1;	/* errno is already set. */
> -
>   	/* Replace argv[0] with the full path to the script if this is the
>   	   first time through the loop. */
>   	replace0_maybe (prog_arg);

LGTM, and I've confirmed that it fixes the problem reported in 
http://www.cygwin.org/ml/cygwin/2019-08/msg00054.html.

Ken