F From:	SMTP%"MACRO32%WKUVX1.BITNET@uu7.psi.com" 20-MAY-1993 09:41:40.97 To:	EVERHART CC:	/ Subj:	Re: Buffered or Direct I/O in DECnet-VAX?   O X-Listname: "VMS Internals, MACRO, & BLISS Discussions" <MACRO32@WKUVX1.BITNET>  Warnings-To: <> - Errors-To: MacroMan%WKUVX1.BITNET@uu7.psi.com * Sender: MacroMan%WKUVX1.BITNET@uu7.psi.com* From: jeh%cmkrnl.com@ULKYVM.LOUISVILLE.EDU+ Reply-To: MACRO32%WKUVX1.BITNET@uu7.psi.com * X-Newsgroups: comp.os.vms,vmsnet.internals2 Subject: Re: Buffered or Direct I/O in DECnet-VAX?. Message-Id: <1993May19.190214.2020@cmkrnl.com> Date: 19 May 93 19:02:14 PDT0 Organization: Kernel Mode Systems, San Diego, CA	 Lines: 78  To: MACRO32@WKUVX1.BITNET  X-Gateway-Source-Info: USENET   ) In article <C7An2y.rp@news.cso.uiuc.edu>, 4  mme32505@uxa.cso.uiuc.edu (Matthew England) writes:: > I'm looking into message-size limitations in DECnet-VAX. > D > I want to be able to know whether or not the DECnet-VAX operations@ > QIO(READVBLK, WRITEVBLK) are using direct or buffered I/O when4 > exchanging data messages between DECnet-VAX tasks.  N For writes, the behavior of the implementation which I most recently looked atO (VMS V5.5) is that it's normally buffered; but if both tasks are running on the O same node, and if the buffer is large enough to make it worthwhile, it's a form N of direct I/O (the process buffer is double-mapped into system address space).  M But you aren't supposed to know any of this, and you certainly shouldn't take J advantage of it in your code; it's undocumented behavior, therefore it mayO change.  ie if you are queueing multiple writes to the link (not that that will K buy you anything), you should use a separate process-space buffer for each, ' even though buffered I/O is being used.    > It would be evenI > nicer if I could _control_ whether or not it does so.  Is there any way * > to support either of these capabilities?   No, it's up to NETDRIVER.   F > I'd like to do this in order to know whether or not there might be aH > problem with the BYTLM, BIOLM, or DIOLM quotas (since these seem to beE > the only limits in terms of buffer transfer size for $QIO) before I : > haul off and start exchaning messages between two tasks.  I $QIOs to DECnet links aren't limited by BYTLM even though buffered I/O is K the norm.  The BIOLM and DIOLM is complicated.  Normally they count against F BIOLM, but if NETDRIVER decides to make it a direct I/O, the charge isM returned to your BIOLM and your DIOLM is checked and decremented!  So for the < truly general case you need one available BIOLM *and* DIOLM.  H But there is NO performance advantage in doing multiple queued writes orM reads to DECnet links, so why worry about it?  Just leave one read queued all 1 the time, and write to the link when you need to.    > It would also E > be nice to limit the possibilities of what might have happened upon G > receiving a SS$_EXQUOTA status returned from a QIO call (lets say I'm  > using QIOW).  H Why don't you just leave process resource wait enabled?  (And it doesn't+ matter whether you're using $QIO or $QIOW.)   F > (I am familiar with the IO_MULTIPLE function for DECnet-VAX QIO; I'dH > like to avoid having to use that in my program unless I have to, i.e.,8 > if I know that I failed because of a BYTLM violation.)  N If there is a limit these days as to the size of buffers that may be sent overN DECnet logical links, I haven't found it.  It's definitely much more than 64K,N and it's definitely more than the PIPELINE QUOTA.  I've sent 100K buffers with, no problems (o SS$_EXQUOTA messages either).  L The main reason for the existence of IO$M_MULTIPLE, as far as I can tell, isM to let a writing program send multiple messages that are received as a single H message, without requiring the writing program to "assemble" the messageL itself.  Or, to deal with messages in "pieces" when the single message would' be too large for the program's buffers.   + > The main purpose that I can see for using G > The only thing that I could think of was maybe there is a correlation G > between the virtual, logical, and physcial I/O classification set and B > the buffered and direct I/O classication set, but I haven't read0 > anything in any documentation to suggest that.  O Nor will you ever.  V, L, and P I/O has nothing to do with direct vs. buffered. L (I suppose someone could write a driver so that this relationship does hold,F but it'd be counter to the way things are supposed to be done on VMS.)  =         --- Jamie Hanrahan, Kernel Mode Systems, San Diego CA O drivers, internals, networks, applications, and training for VMS and Windows NT M uucp 'g' protocol guru and release coordinator, VMSnet (DECUS uucp) W.G., and J Chair, Programming and Internals Working Group, U.S. DECUS VMS Systems SIGB Internet:  jeh@cmkrnl.com  Uucp: uunet!cmkrnl!jeh  CIS: 74140,2055