Reply-To: "Rossen Dimitrov" From: "Rossen Dimitrov" To: ,"William Gropp" References: <4.2.2.20010604152321.036ed0b8@XXXXXXXXX> Subject: Re: Handling errors in handle transfer functions Date: Mon, 4 Jun 2001 16:20:57 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.3018.1300 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.3018.1300 Bill, The standard actually states that an invalid Fortran/C handle is translated into an invalid C/Fortran handle (p. 51, lines 15 and 20). It does not say that translating an invalid handle is erroneous, which would imply that an error handler should be called. So, I don't think that handle translation between languages should result in calling error handlers. I am not sure how an invalid handle is exactly defined in the standard. So, maybe the proposed clarification should explicitly state what an invalid handle is for the purposes of handle translation and following this definition how the implementation should generate such a handle from the translation functions. Since the translated handle is "returned" form the translation functions (as opposed to passed by reference), it must be set to something specific, which could not be mistaken for a valid handle. One candidate value for this purpose could be MPI_UNDEFINED. Rossen MPI Software Technology, Inc. ----- Original Message ----- From: "William Gropp" To: Sent: Monday, June 04, 2001 3:23 PM Subject: Handling errors in handle transfer functions > What should a high-quality implementation do if an invalid handle is given > to one of the transfer functions (e.g., MPI_Comm_c2f)? I believe that the > error handler on MPI_COMM_WORLD should be invoked (MPI_FILE_NULL for > MPI_FILE handle transfers) because that is the specified error handler if > no valid communicator/window/file is available. A user asked this > recently; I'd like to propose this as a clarification. > > Bill >