![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to README.PROGRAMMERS CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [UNIX Amiga Emulator] / uae / docs|
Return to README.PROGRAMMERS CVS log | Up to [UNIX Amiga Emulator] / uae / docs |
1.1 root 1: You can help to make this program better. If you fix bugs or implement new
2: features, I'd be grateful if you send me patches. For a list of interesting
3: projects, and for a brief summary on how UAE works, see below.
4:
5: A few guidelines for anyone who wants to help:
6: - Please contact me first before you implement major new features. Someone
7: else might be doing the same thing already. This has already happened :-(
8: Even if no one else is working on this feature, there might be alternative
9: and better/easier/more elegant ways to do it.
10: - If you have more than one Kickstart, try your code with each one.
11: - Patches are welcome in any form, but diff -u or diff -c output is preferred.
12: If I get whole source files, the first thing I do is to run diff on it. You
13: can save me some work here (and make my mailbox smaller).
14:
15: Some possible projects, in order of estimated difficulty:
1.1.1.2 root 16: - Add gamma correction
1.1.1.3 root 17: - If the serial port still isn't working (I've got no idea, I don't use it),
18: fix it.
1.1 root 19: - Someone with a 68020 data sheet might check whether all opcodes are
20: decoded correctly and whether all instructions really do what they are
21: supposed to do (I'm pretty sure it's OK by now, but you never know...).
1.1.1.3 root 22: - Add more 2.0 packets to filesys.c
1.1.1.2 root 23: - Multi-thread support is there now, it just needs someone to test it on a SMP
24: machine and to fix it so it improves speed instead of slowing the thing
25: down.
1.1 root 26: - Improve the Kickstart replacement to boot more demos.
27: - Snapshots as in CPE. Will need to collect all the variables containing
28: important information. Fairly easy, but boring. (Use core dumps instead :-)
1.1.1.2 root 29: _If_ someone attempts this, please be more clever than the various CPC
30: emulators and dump state only at one fixed point in the frame, preferrably
1.1.1.3 root 31: the vsync point. Also talk with Petter about this.
1.1 root 32: - Find out why uae.device has to be mounted manually with Kick 1.3.
33: The problem seems to be that we don't have a handler for it. I _think_ what
34: we need is the seglist of the standard filesystem handler. Problem is,
35: DOS hasn't been started when the devices are initialized and so we can't get
36: to the DosBase->RootNode->FileHandlerSeg pointer, and then there is the
37: confusing matter of BCPL GlobVecs and other weird stuff...
1.1.1.2 root 38: - Some incompatibilities might be fixed with user-modifiable fudge variables
39: the same way it's done in various C64 emulators.
40: - With the new display code, it would probably be easier than before to
41: implement ECS resolutions - however, a lot of places rely on the OCS timing
42: parameters and display sizes.
1.1 root 43: - Figure out a diskfile format that supports every possible non-standard
1.1.1.6 root 44: format [I'll work on this soon (I hope)].
45: - Port bsdsocket library code to other operating systems than Win32.
1.1 root 46: - Implement 68551 MMU. I have docs now. Not among the most necessary things.
1.1.1.3 root 47: Should be done like exception 3 handling: add code to genamode in gencpu.c.
1.1 root 48: - Implement AGA support. Some bits and pieces exist.
49: - Reimplement Amiga OS. (Well-behaved) Amiga programs could then be made
50: to use the X Window System as a "public screen". Of course, not all the
1.1.1.3 root 51: OS would have to be re-done, only Intuition/GFX/Layers (which is enough).
52: [Started, look at gfxlib.c - not usable yet.]
1.1.1.2 root 53: - Find some extremely clever ways to optimize the smart update methods. Some
54: ideas:
55: a) Always use memcmpy() to check for bitplane differences. If no differences
56: are found, see if BPLxDELAY got modified, if so, scroll.
1.1.1.3 root 57: Problems:
1.1.1.2 root 58: * You'd still have to draw a few pixels around the DIW borders. Not very
59: hard.
60: * Scrolling with memcpy in video memory can be terribly slow (no, I
61: shouldn't have bought the cheaper video card with DRAMs)
62: * At least every 15 pixels a full update has to be done since the
63: bitplane pointers get updated after that. And that's with the slowest
64: scrolling - if the playfield scrolls faster, the benefit converges
65: against zero.
66: You could also do vertical scrolling tests, but similar problems arise -
67: where should one check? One line above/below? What about faster
68: scrolling? You could use the bitplane pointers as hints, but with
69: double/triple buffering this gets problematic, too.
70: On the whole, I don't think it would be worth the effort, even if it
71: works very well for a few games.
72: b) Well, there is no b). If I thought of something I forgot it while
73: writing a).
1.1 root 74: - Port it to Java and Emacs Lisp
75: - A formal proof of correctness would be nice.
76:
77:
1.1.1.3 root 78: Source file layout
79:
80: src/ contains (mostly) machine-independent C code.
81: include/ contains header files included by C code.
82: md-*/ CPU and compiler dependent files, linked to machdep by configure
83: od-*/ operating system dependent files, linked to osdep by configure
84: td-*/ thread library dependent files, linked to threaddep by configure
85: sd-*/ Sound code. sd-* is only for sound systems which are not OS specific
86: or for which no "od-*" directory exists. Linked to sounddep
1.1.1.7 ! root 87: scg-*/ SCSI code.
! 88: This is the same code as in the cdrecord libscg, so this will become
! 89: obsolete as soon as the libscg is released seperately. In the meantime,
! 90: UAE will be distributed together with the code itself, because it
! 91: also requires patching: not every libscg implementation is thread-safe
! 92: in the current cdrecord version.
1.1.1.3 root 93: targets/ Contains header files which contain some information about which
94: options a specific port of UAE understands.
95:
96:
97: Coding style
98:
99: As long as your code is hidden in a file buried in md-*/ or od-*/ where I
100: never have a look at it, you can probably get away with not following these
101: guidelines.
102:
103: * Do not include CR characters.
104: * Do not use GNU C extensions if you can't hide them in a macro or in a
105: system-specific file so that an alternative implementation is available
106: when GNU C is not used.
107: This applies to _all_ OS/CPU/compiler specific details. Basically, nothing
108: of that sort should appear in src/*.c (we're a bit away from that goal at
109: the moment, but it's getting better).
110: * Make sure your code does not make assumption about type sizes other than
111: the minimum widths allowed by C. If you need specific type sizes, use the
112: uae_u32 type and its friends.
113: * Set up your editor so that tab characters round up to the next position
114: where ((cursorx-1) % 8) == 0, i.e. 8 space tabs. Do not use 4 space tabs,
115: that makes the code awful to read on other machines and worse to edit.
1.1.1.5 root 116: (I'm talking about the tab character here, not indentation!)
1.1.1.3 root 117: * Lines can be up to 132 characters wide. Use SVGATextMode for the Linux
118: console, or use a windowing system in a high resolution.
119: * C++ comments are a no-no in C code.
120: * Indentation - look at some code in custom.c and try to follow it. Don't
121: use GNU 2-space-in-weird-places indentation, I find it awful. But _do_
122: follow the GNU rules for adding whitespace in expressions, and those for
123: breaking up multiple-line if statements.
124: Fixed indentation rules almost never make sense - break the rules if that
125: makes your code more readable.
126: Hint: Get jed from space.mit.edu, /pub/davis. It can indent your code
127: automatically. Put the following into your .jedrc, and it will come out
128: right:
129: C_INDENT = 4;
130: C_BRACE = 0;
131: C_BRA_NEWLINE = 0;
132: C_Colon_Offset = 1;
133: C_CONTINUED_OFFSET = 4;
134:
135:
1.1 root 136: How it works
137:
138: Let's start with the memory emulation. All addressable memory is split into
139: banks of 64K each. Each bank can define custom routines accessing bytes,
140: words, and longwords. All banks that really represent physical memory just
141: define these routines to write/read the specified amount of data to a chunk
1.1.1.3 root 142: of memory. This memory area is organized as an array of uae_u8, which means
1.1 root 143: that those parts of the emulator that want to access memory in a linear
1.1.1.3 root 144: fashion can get a (uae_u8 *) pointer and use it to circumvent the overhead of
1.1.1.2 root 145: the put_*() and get_*() calls. That is done, for example, in the
1.1 root 146: pfield_doline() function which handles screen refreshes.
147: Memory banks that represent hardware registers (such as the custom chip bank
148: at 0xDF0000) can trap reads/writes and take any necessary actions.
149:
150: To provide a good emulation of graphical effects, only one thing is vital:
151: Copper and playfield emulation have to be kept absolutely synchronous. If the
152: copper writes to (say) a color register in a specific cycle, the playfield
1.1.1.2 root 153: hardware needs to use the new information in the next word of data it
1.1 root 154: processes.
155: UAE 0.1 used to call routines like do_pfield() and do_copper() each time the
156: CPU emulator had finished an instruction. That was one of the reasons why it
157: was so slow. Recent versions try to draw complete scanlines in one piece. This
158: is possible if the copper does not write to any registers affecting the
159: display during that scanline. Therefore, drawing the line is deferred until
1.1.1.2 root 160: the last cycle of the line. However, sometimes a register which affects how
161: the screen will look is modified before the end of the line (think of copper
162: plasmas). That's what "struct decision thisline_decision" is for. It is
163: initialized at the start of each line. During the line, whenever a vital
164: register is changed, one of the decide_*() functions is called and may modify
165: thisline_decision. There are several independent decisions:
166: - which DIW should be used
167: - where does data fetch start/stop (or is the line in the border altogether)
168: - where should sprites be drawn (note: the same sprite can appear more than
169: once on one scanline, see Turrican I world 3 levels 1 and 3 for the best
170: example)
171: - what are the playfield pointers at the start of DDF. Related, what data do
172: they point to.
173: - what are the playfield modulos at the end of DDF
174: - coppermagic with the colors is remembered for later use
175: - so is copper magic with the bitplane delay values. I used to think there
176: was no useful application for modifying BPLCON1 while data is being
177: displayed, but Sanity demos can make Amiga emulator programmers look real
178: old.
179:
180: All of this is remembered while the raster line is processed by the hardware.
181: After the line (at hsync), all the decisions are made if they weren't made
182: before. At that point the line can be drawn by playfield_draw_line.
183: Additionally, all the decisions from the previous displayed frame are saved
184: and compared with the new ones, since often lines are not modified between
185: frames. This saves a lot of redrawing work.
1.1 root 186:
187: The CPU emulator no longer has to call all sorts of functions after each
188: instruction. Instead, it keeps a list of events that are scheduled (timer
189: interrupts, hsync and vsync events) and their "arrival time". Only the time
190: for the next event is checked after each CPU instruction. If it's higher than
191: the current cycle counter, the CPU can continue to execute.
192:
1.1.1.2 root 193: Things that can't be supported with the current "decision" model:
194: - Changes in lores/hires mode during one line. Dunno whether that was ever
195: used in reality.
196: - Changes to the bitplane DMA bit during one line. Hardly useful and not
1.1.1.3 root 197: likely to be used. [but there are at least two programs which do ugly
1.1.1.2 root 198: things like that, and there are some hacks in UAE that make those programs
199: work (Magic 12 Ray of Hope 2 is one of these demos)]
200: - Changes in bitplane data during one line. If programs do this kind of
201: thing, it's most likely accidental and the program is broken. Can happen
202: with programs that use the blitter incorrectly, like all the Andromeda
203: demos.
204: - others? (fill in if you can think of anything)
205:
206: All in all, it's unlikely that this causes compatibility problems. If it does,
207: fudge values could be introduced (although that sort of thing gets messy
208: quickly).
1.1 root 209:
210:
1.1.1.3 root 211: * Native code vs. 68k code
1.1.1.2 root 212:
1.1.1.3 root 213: It is possible to call native code from 68k code; autoconf.c has some routines
214: which make setting up a call trap very easy. However, it is not as easy to
215: call 68k code from native C code, at least not while Amiga Exec multitasking
216: is running. You ask why?
217:
218: Amiga process1 calls native function foo
219: Native function foo calls some 68k function and goes into 68k mode
220: Amiga context switch happens, process1 is put to sleep and process2 gets run.
221: Amiga process2 calls native function foo
222: Native function foo calls some 68k function and goes into 68k mode
223: Amiga context switch happens, process2 is put to sleep and process1 gets run.
224: Process 1 completes the 68k function called by foo and returns from 68k mode.
225:
226: There. Now we are in function foo again. When it called the 68k code, process2
227: was active. Now process1 is active, and the function we called in process2
228: hasn't completed yet. What a mess.
229:
230: To get around this, you need to do some stack magic. Code to do this exists,
231: but it must be adapted for each port, since setting up a different stack is
232: completely non-portable.
233:
234:
235: * How multithreading in filesys.c works
236:
237: AmigaOS is nice enough to start one processes for each mounted filesystem. All
238: of these run in the 68k emulation code, i.e. in the main UAE thread. This is
239: the reason why multithreading is desirable: if the main UAE thread blocks
240: waiting for I/O, the CPU emulation can't continue to run. Since the Amiga OS
241: is capable of multi-tasking, it is possible that other code could run until
242: the I/O operation is complete. The most important bit of code that can run is
243: the code that moves the mouse pointer - it's unpleasant if the pointer does
244: not follow mouse movement during disk/CD accesses.
245:
246: When a packet is received by the filesys.asm code, filesys_handler is called.
247: This function always runs in the main UAE thread.
248: - In the single-threaded case, this function performs the action that was
249: requested, then returns 0 to indicate "action completed, reply packet".
250: Nothing else is performed.
251: - In the multi-threaded case, filesys_handler figures out which unit the
252: packet was for and sends the packet to the UAE thread responsible for
253: handling this unit. filesys_handler returns 0 to indicate: queue the
254: packet. Also, one (at that point unused) field in the packet is set to
255: 0 to indicate that the action was not completed.
256:
257: The latter case is the interesting one. The thread that got the packet does
258: the following:
259: - perform the action as usual
260: - set the "command complete" field in the packet to -1.
261: - send a message to the AmigaOS (!) filesystem process. However, it can't do
262: that without some effort. We can't call 68k code from the emulator easily.
263: So we have to use an Amiga interrupt. The filesystem init code sets up an
264: Exec IntServer for the EXTER interrupt, and hsync_handler() checks
265: periodically whether the filesystem needs an interrupt and raises one if
266: necessary.
267: Only one dummy message is used per filesystem unit, which is allocated at
268: startup. This means that there must be some locking to prevent the unit
269: thread from sending the same message twice to the same port. To determine
270: whether the message is free, three counts are kept. "cmds_sent" is
271: incremented by the UAE thread whenever it has completed a command.
272: "cmds_acked" is set to the same value of cmds_sent at the point that the
273: interrupt handler got invoked and decided it must send a message. Finally,
274: cmds_complete is set to this value at the time the AmigaOS process receives
275: the dummy message. Whenever cmds_acked == cmds_complete, the dummy message
276: is free to be sent again.
277:
278: The EXTER interrupt basically walks through the units, looks at the cmds_*
279: fields and sends the dummy message to the Amiga filesystem process when
280: possible and necessary.
281:
282: When the Amiga filesystem process receives such a dummy message, it does the
283: following:
284: - increment cmds_complete as described above.
285: - walk through the queue of unprocessed commands and see which ones now have
286: a status of -1, indicating that they are finished. These are removed from
287: the queue and replied to.
288:
289:
290: * Calltraps at fixed locations
291:
292: F0FF00: return from 68k mode.
293: F0FF10: must have gotten lost somewhere ;)
294: F0FF20: used by filesys.c to store away some information from the startup
295: packet.
296: F0FF30: filesys_handler().
297: F0FF40: startup_handler(), handles only the startup packet for each
298: filesystem.
299: F0FF50: used by the EXTER interrupt which we set up for the filesystem.
300: F0FF60: used by the uaectrl/uae-control programs (see uaelib.c)
301: F0FF70: used by the task that gets set up for the mouse emulation.
1.1.1.7 ! root 302: F0FF80: getchipmemsize()
! 303: F0FF90: uaeexe
! 304: F0FFA0: timehack
1.1.1.5 root 305:
1.1.1.3 root 306: * How the compiler works
1.1.1.2 root 307:
308: .. yet to be written. To be decided, in fact.
1.1.1.3 root 309:
310:
311: Portability
312:
313: This section was out of date. I'll rewrite it.
314: Some day.
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.