At 07:23 AM 11/24/2000, [log in to unmask] wrote:
>There has been a few comments, recently, from vendor compiler engineers,
>complaining that bug reports should be sent to the proper support
>provided by the vendor, and not to comp-fortran-90.
>
>I understand the frustration of the compiler engineers, when hearing of a
>bug for the first time on a mailing list.
>
>However, there is often another side of the story.
>Most of the time, a dysfunction is not clearly identified as a bug.
>Users do not always have a copy of the Standard, and even if they do,
>it is not so easy to understand.
Yet another side to this story occurs when one is using the Sun
compiler. Under our support agreement, the report must be made by
designated system administrators. Thus I've had to make phone calls and
write email to start the process. Then a Sun administrative person
communicates with me about the details. Then generally there is a long
delay until this person asks me for even more details. At no point do I
actually get any advice in the first few days of an encounter with Sun
support. At some point I solve the problem myself or get advice from
another source such comp-fortran-90.
In the cases when the Sun engineers have noticed my postings to
comp-fortran-90, the process led to some useful responses from Sun. (Also,
I've received communications from Sun managers stating that the process
will be improved.) In the other cases, I got regular requests to supply
more information or allow the "case" (I forget what Sun calls it) to be
closed. For each of these I allowed the case to be closed after developing
a work-around to the problem.
Having had this experience, I'm not surprised that some members would be
inclined to submit questions to comp-fortran-90 instead of the vendor.
>I had several cases in the last 2 or 3 years, where it took me some
>time, and help from the contributors of this list, to decide whether
>a problem was a compiler bug or not.
>I even saw one case where different compilers where acting differently,
>and the old-timer (the vendor that was present on my site for years) would
>dismiss my bug report, while the new-kid-on-the-block (the company that had
>just installed a new system on the site) was ready to "correct" its compiler
>from standard-compliant to old-timer compliant!
>
>This list permits developpers to ask the fortran community about
>unclear issues, get a strong, if not authoritative, opinion, and
>avoids us the embarassment of submitting a bug report when there is
>no bug at all.
>
>Bertrand MELTZ
>
>****************************************
>* *
>* Bertrand Meltz *
>* CEA / Bruyeres-le Chatel *
>* BP 12 *
>* 91680 Bruyeres-le-Chatel *
>* FRANCE *
>* *
>* Tel : (33) [0] 1 69 26 57 83 *
>* Fax : (33) [0] 1 69 26 70 93 *
>* *
>* e_mail : [log in to unmask] *
>* [log in to unmask] *
>* *
>****************************************
---
Harold P. Hjalmarson
(Computational Biology and Materials Technology Department, Dept. 9235)
Sandia National Labs, MS-1111, PO Box 5800, Albuquerque, NM 87185-1111
[log in to unmask] (Internet) 505-844-8888 (office) 505-845-7442 (fax)
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|