home *** CD-ROM | disk | FTP | other *** search
- Path: news.uni-stuttgart.de!fleischr
- From: fleischr@izfm.uni-stuttgart.de (Matthias Fleischer)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: ln_Type of a new IORequest
- Date: 10 Apr 1996 07:35:10 GMT
- Organization: None. I'm not organized at all.
- Message-ID: <4kfobe$2oec@info4.rus.uni-stuttgart.de>
- References: <4kdi28$n8a@taiwan.informatik.uni-rostock.de>
- NNTP-Posting-Host: sunserv.izfm.uni-stuttgart.de
-
- In article <4kdi28$n8a@taiwan.informatik.uni-rostock.de>,
- Gunther Nikl <gnikl@informatik.uni-rostock.de> wrote:
- >
- >I mentioned in "exec/nodes.h" that a new IORequest should have
- >a ln_Type of NT_UNKNOWN. Being surprised I checked the type of
- >the result from a CreateIORequest() call. The IOrequest had the
- >type "NT_REPLYMSG" (V39). Sometimes I create an IORequest by hand
- >setting ln_Type to "NT_MESSAGE". Now I wonder what is really
- >correct?
- >
- AFAIK the node types for IORequests have the following meanings:
- NT_MESSAGE - IORequest is currently active (in use by the device).
- NT_REPLYMSG - IORequest is finished (the device replied to it).
-
- OpenDevice() always sets the type to NT_REPLYMSG - so the
- value after CreateIORequest() probably doesn't matter.
-
- Matthias
- --
- Ever noticed that %=0/0=NaN?
-