[Coco] Accessing Vsync & Hsync interrups from Nitros9
Bill Pierce
ooogalapasooo at aol.com
Sat Apr 11 07:21:32 EDT 2020
Walter, the NitrOS9 repository is here:
https://sourceforge.net/p/nitros9/code/ci/default/tree/
-----Original Message-----
From: Walter Zambotti <zambotti at iinet.net.au>
To: coco at maltedmedia.com
Sent: Sat, Apr 11, 2020 6:28 am
Subject: Re: [Coco] Accessing Vsync & Hsync interrups from Nitros9
Where is the repo?
Walter
On 2020-04-11 16:04, L. Curtis Boyle wrote:
> For Walter Zambotti:
>
> There is some documentation in the source code on the repository. Also, this doc here came with the original package:
>
>
> VRN (VIRQ/RAM/NULL Driver) Documentation
>
>
> Copyright (C) 1989, 1990 Bruce Isted
>
> This program may be freely distributed as long as the copyright notice remains
> intact and the source, binary, and documentation files are distributed
> together.
>
> This program may not be sold for profit, individually or as part of a package
> without the prior written permission of the copyright holder.
>
>
> Note: Where the following text refers to "King's Quest III", the reader
> should remember that whatever is said applies also to "Leisure Suit Larry",
> another of Sierra's games for the CoCo 3.
>
>
> This ARchive should contain the following files:
> ftdd.asm - OS-9/6809 assembly source for Ftdd.dd descriptor.
> Ftdd.dd - Replacement for Flight Simulator II "FTDD" device descriptor.
> nil.asm - OS-9/6809 assembly source for Nil.dd descriptor.
> Nil.dd - Replacement for NilDrv "Nil" device descriptor.
> TstFS2 - Test program for Flight Simulator II type signals.
> tstfs2.asm - OS-9/6809 assembly source for TstFS2 program.
> TstKQ3 - Test program for King's Quest III type signals.
> tstkq3.asm - OS-9/6809 assembly source for TstKQ3 program.
> TstRAM - Test program for King's Quest III type RAM allocation.
> tstram.asm - OS-9/6809 assembly source for TstRAM program.
> vi.asm - OS-9/6809 assembly source for Vi.dd descriptor.
> Vi.dd - Replacement for King's Quest III "VI" device descriptor.
> vrn.asm - OS-9/6809 assembly source for VRN (VIRQ/RAM/NULL) driver.
> Vrn.dr - Replacement for AGIVIRQDr, FT, and NilDrv drivers.
> vrn.doc - This documentation file.
> read.me - A small information file that should be read first.
>
>
> The general features of each section of the VRN driver are explained in the
> "VRN (VIRQ/RAM/NULL Driver) Synopsis" section at the end of this documentation
> file. For more detail you'll have to consult the source file (vrn.asm).
>
> The three test programs supplied (TstFS2, TstKQ3, and TstRAM) are intended to
> both test the various functions of, and provide an example of how to interface
> to, the VRN driver.
>
> Using the test programs in combination with various other utilities we all
> should have (KD's utils, etc.) gives us a pretty complete set of tools to test
> VRN's features without requiring Flight Simulator II or King's Quest III. Of
> course, you can play the games if you want to! I've had four copies of FS2
> running simultaneously. However, it's impossible to run more than one copy of
> KQ3 at a time. Running more than two FS2s plus one KQ3 is also too much, the
> COCO can't keep up, and the stack (system, I think) overflows.
>
> The TstFS2 program is used to test both standard FS2 and FS2+ VIRQ signals.
> It allows one parameter character. If no parameters are given, then it
> defaults to standard FS2 VIRQ mode. If a parameter character is given, then
> the ASCII value of that character is used as the signal code. A special case
> is the numeral "0", which sets up a "one-shot" VIRQ. Pressing [.] (period)
> will return the total VIRQs and signal count, while pressing [SPACEBAR] will
> clear the FS2/FS2+ VIRQ while leaving the path open, and pressing [ENTER] will
> terminate the program.
>
> The TstKQ3 program is used to test KQ3 type signals. It ignores parameters.
> It simply counts the signals received, and reports whenever the one byte
> counter overflows (256 signals). Pressing [SPACEBAR] will clear the KQ3 VIRQ
> while leaving the path open, and pressing [ENTER] will terminate the program.
>
> The TstRAM program will allocate from 1 to 9 RAM blocks. It does nothing with
> them, not even mapping them into its address space. It allows one parameter
> character from "0" to "9". If no parameters are given, then it defaults to
> one RAM block. Pressing [SPACEBAR] will de-allocate the RAM block(s) while
> leaving the path open, and pressing [ENTER] will terminate the program.
>
> One thing that should be mentioned about the RAM allocation is that it simply
> reserves the RAM blocks. It's up to the user program to map the block(s) into
> the user's address space. Any number of RAM blocks may be reserved, up to the
> total available. It's possible for a program to reserve thirty or more
> consecutive RAM blocks for data, and map/unmap them as required in whatever
> chunks will fit in available user memory. Another option is to open several
> paths to a VRN descriptor, and allocate RAM block(s) for each path. This
> makes using large amounts of non-contiguous memory possible.
>
>
>
> VRN (VIRQ/RAM/NULL Driver) Synopsis
>
>
> _Init_
>
> Entry Conditions:
> DP = system direct page
> Y = address of the device descriptor
> U = address of the device memory area
> Note: All device memory except V.PAGE and V.PORT has been cleared
> (zeroed) before the _Init_ routine is entered.
>
> Exit Conditions:
> IRQ and VIRQ (repetetive one tick interval) polling table entries have
> been installed.
>
> Error Conditions:
> CC = carry set
> B = error code
>
>
> _Read_
>
> Entry Conditions:
> DP = system direct page
> Y = address of the path descriptor
> U = address of the device memory area
>
> Exit Conditions:
> No character is ever returned, _Read_ always exits with E$EOF error.
>
> Error Conditions:
> CC = carry set
> B = E$EOF
>
>
> _Write_
>
> Entry Conditions:
> A = character to write
> DP = system direct page
> Y = address of the path descriptor
> U = address of the device memory area
>
> Exit Conditions:
> The character to write is ignored, _Write_ always exits without error.
>
> Error Conditions:
> No error possible.
>
>
> _Get_Status_
>
> Entry Conditions:
> A = get status call code
> DP = system direct page
> Y = address of the path descriptor
> U = address of the device memory area
>
> Exit Conditions:
> Depends upon the call code. See "Supported Calls" below for further
> information. If the call code is not supported, _Get_Status_ exits
> with an E$UnkSvc error.
>
> Error Conditions:
> CC = carry set
> B = error code
>
> Supported Calls:
> Code $01, SS.Ready
> Always exits with E$NotRdy (no data ready) error.
> Code $80, FS2 defined
> Returns VIRQ total counter MSBs in caller's R$X, LSBs in R$Y, and
> clears VIRQ total. If entry for path is non-existant, exits with
> E$Unit error.
> Code $81, FS2 defined
> Returns signal total counter in caller's R$A and clears signal total.
> If entry for path is non-existant, exits with E$Unit error.
>
>
> _Set_Status_
>
> Entry Conditions:
> A = set status call code
> DP = system direct page
> Y = address of the path descriptor
> U = address of the device memory area
>
> Exit Conditions:
> Depends upon the call code. See "Supported Calls" below for further
> information. If the call code is not supported, _Set_Status_ exits
> with an E$UnkSvc error.
>
> Error Conditions:
> CC = carry set
> B = error code
>
> Supported Calls:
> Code $2A, SS.Close
> All KQ3 and FS2/FS2+ VIRQ entries for the closed path are cleared, and
> RAM block(s) allocated for the closed path are de-allocated.
> Code $81, FS2 defined
> If caller's R$Y LSB is not zero, an FS2 VIRQ entry for path is set.
> Caller's R$X contains initial timer count and reset count. The signal
> code to send is set to $80. If R$X is zero then the VIRQ is a
> one-shot, and will not be repeated. If entry for path already exists,
> it is replaced. If no table entries are available, exits with
> E$DevBsy error.
> If caller's R$Y LSB is zero, then the FS2/FS2+ VIRQ entry for path is
> cleared. If entry for path is non-existant and no table entries are
> available, exits with E$DevBsy error, otherwise exits without error.
> Code $C7, newly defined
> Sets FS2+ VIRQ entry for path. Caller's R$X contains initial timer
> count, R$Y contains reset count, and R$U LSB contains signal code to
> send. If R$Y is zero then the VIRQ is a one-shot, and will not be
> repeated. If entry for path already exists, it is replaced. If no
> table entries are available, exits with E$DevBsy error.
> Code $C8, KQ3 defined
> Sets KQ3 VIRQ entry for path. If entry for path already exists, exits
> without error. If no table entries are available, exits with E$DevBsy
> error.
> Code $C9, KQ3 defined
> Clears KQ3 VIRQ entry for path. If non-existant, exits without error.
> Code $CA, KQ3 defined
> Allocates caller's R$X LSB number of consecutive RAM block(s) for path
> and returns starting block number in caller's R$X. If RAM block(s)
> are already allocated for path, or if no table entries are available,
> exits with E$DevBsy error.
> Code $CB, KQ3 defined
> De-allocates RAM block(s) previously allocated for path. If none were
> allocated, exits without error.
>
>
> _Term_
>
> Entry Conditions:
> DP = system direct page
> U = address of the device memory area
>
> Exit Conditions:
> IRQ and VIRQ polling table entries have been removed.
>
> Error Conditions:
> CC = carry set
> B = error code
>
>
> _IRQ_Service_
>
> Entry Conditions:
> DP = system direct page
> U = address of the device memory area
>
> Exit Conditions:
> The VIRQ status flag has been cleared. All KQ3 entries in the VIRQ tables
> have been signalled ($80). All FS2/FS2+ entries in the VIRQ tables VIRQ
> total counters have been incremented, the VIRQ delay counters have been
> decremented, and if zero have been signalled (FS2 signal = $80, FS2+
> signal = user defined). Each signalled FS2/FS2+ entry's signal total
> counter has been incremented. If the signalled FS2/FS2+ entry's reset
> value is not zero then the counter is reset to the value, otherwise the
> FS2/FS2+ entry is cleared. _IRQ_Service_ always exits without error.
>
> Error Conditions:
> No error possible.
>
>
> Bruce Isted 90/11/29
>
>
>
--
Coco mailing list
Coco at maltedmedia.com
https://pairlist5.pair.net/mailman/listinfo/coco
More information about the Coco
mailing list