/[gxemul]/trunk/TODO
This is repository of my old source code which isn't updated any more. Go to git.rot13.org for current projects!
ViewVC logotype

Contents of /trunk/TODO

Parent Directory Parent Directory | Revision Log Revision Log


Revision 34 - (show annotations)
Mon Oct 8 16:21:17 2007 UTC (16 years, 5 months ago) by dpavlin
File size: 22112 byte(s)
++ trunk/HISTORY	(local)
$Id: HISTORY,v 1.1480 2007/02/19 01:34:42 debug Exp $
20061029	Changing usleep(1) calls in the debugger to usleep(10000)
20061107	Adding a new disk image option (-d o...) which sets the ISO9660
		filesystem base offset; also making some other hacks to allow
		NetBSD/dreamcast and homebrew demos/games to boot directly
		from a filesystem image.
		Moving Dreamcast-specific stuff in the documentation to its
		own page (dreamcast.html).
		Adding a border to the Dreamcast PVR framebuffer.
20061108	Adding a -T command line option (again?), for halting the
		emulator on unimplemented memory accesses.
20061109	Continuing on various SH4 and Dreamcast related things.
		The emulator should now halt on more unimplemented device
		accesses, instead of just printing a warning, forcing me to
		actually implement missing stuff :)
20061111	Continuing on SH4 and Dreamcast stuff.
		Adding a bogus Landisk (SH4) machine mode.
20061112	Implementing some parts of the Dreamcast GDROM device. With
		some ugly hacks, NetBSD can (barely) mount an ISO image.
20061113	NetBSD/dreamcast now starts booting from the Live CD image,
		but crashes randomly quite early on in the boot process.
20061122	Beginning on a skeleton interrupt.h and interrupt.c for the
		new interrupt subsystem.
20061124	Continuing on the new interrupt system; taking the first steps
		to attempt to connect CPUs (SuperH and MIPS) and devices
		(dev_cons and SH4 timer interrupts) to it. Many things will
		probably break from now on.
20061125	Converting dev_ns16550, dev_8253 to the new interrupt system.
		Attempting to begin to convert the ISA bus.
20061130	Incorporating a patch from Brian Foley for the configure
		script, which checks for X11 libs in /usr/X11R6/lib64 (which
		is used on some Linux systems).
20061227	Adding a note in the man page about booting from Dreamcast
		CDROM images (i.e. that no external kernel is needed).
20061229	Continuing on the interrupt system rewrite: beginning to
		convert more devices, adding abort() calls for legacy interrupt
		system calls so that everything now _has_ to be rewritten!
		Almost all machine modes are now completely broken.
20061230	More progress on removing old interrupt code, mostly related
		to the ISA bus + devices, the LCA bus (on AlphaBook1), and
		the Footbridge bus (for CATS). And some minor PCI stuff.
		Connecting the ARM cpu to the new interrupt system.
		The CATS, NetWinder, and QEMU_MIPS machine modes now work with
		the new interrupt system :)
20061231	Connecting PowerPC CPUs to the new interrupt system.
		Making PReP machines (IBM 6050) work again.
		Beginning to convert the GT PCI controller (for e.g. Malta
		and Cobalt emulation). Some things work, but not everything.
		Updating Copyright notices for 2007.
20070101	Converting dev_kn02 from legacy style to devinit; the 3max
		machine mode now works with the new interrupt system :-]
20070105	Beginning to convert the SGI O2 machine to the new interrupt
		system; finally converting O2 (IP32) devices to devinit, etc.
20070106	Continuing on the interrupt system redesign/rewrite; KN01
		(PMAX), KN230, and Dreamcast ASIC interrupts should work again,
		moving out stuff from machine.h and devices.h into the
		corresponding devices, beginning the rewrite of i80321
		interrupts, etc.
20070107	Beginning on the rewrite of Eagle interrupt stuff (PReP, etc).
20070117	Beginning the rewrite of Algor (V3) interrupts (finally
		changing dev_v3 into devinit style).
20070118	Removing the "bus" registry concept from machine.h, because
		it was practically meaningless.
		Continuing on the rewrite of Algor V3 ISA interrupts.
20070121	More work on Algor interrupts; they are now working again,
		well enough to run NetBSD/algor. :-)
20070122	Converting VR41xx (HPCmips) interrupts. NetBSD/hpcmips
		can be installed using the new interrupt system :-)
20070123	Making the testmips mode work with the new interrupt system.
20070127	Beginning to convert DEC5800 devices to devinit, and to the
		new interrupt system.
		Converting Playstation 2 devices to devinit, and converting
		the interrupt system. Also fixing a severe bug: the interrupt
		mask register on Playstation 2 is bitwise _toggled_ on writes.
20070128	Removing the dummy NetGear machine mode and the 8250 device
		(which was only used by the NetGear machine).
		Beginning to convert the MacPPC GC (Grand Central) interrupt
		controller to the new interrupt system.
		Converting Jazz interrupts (PICA61 etc.) to the new interrupt
		system. NetBSD/arc can be installed again :-)
		Fixing the JAZZ timer (hardcoding it at 100 Hz, works with
		NetBSD and it is better than a completely dummy timer as it
		was before).
		Converting dev_mp to the new interrupt system, although I
		haven't had time to actually test it yet.
		Completely removing src/machines/interrupts.c, cpu_interrupt
		and cpu_interrupt_ack in src/cpu.c, and
		src/include/machine_interrupts.h! Adding fatal error messages
		+ abort() in the few places that are left to fix.
		Converting dev_z8530 to the new interrupt system.
		FINALLY removing the md_int struct completely from the
		machine struct.
		SH4 fixes (adding a PADDR invalidation in the ITLB replacement
		code in memory_sh.c); the NetBSD/dreamcast LiveCD now runs
		all the way to the login prompt, and can be interacted with :-)
		Converting the CPC700 controller (PCI and interrupt controller
		for PM/PPC) to the new interrupt system.
20070129	Fixing MACE ISA interrupts (SGI IP32 emulation). Both NetBSD/
		sgimips' and OpenBSD/sgi's ramdisk kernels can now be
		interacted with again.
20070130	Moving out the MIPS multi_lw and _sw instruction combinations
		so that they are auto-generated at compile time instead.
20070131	Adding detection of amd64/x86_64 hosts in the configure script,
		for doing initial experiments (again :-) with native code
		generation.
		Adding a -k command line option to set the size of the dyntrans
		cache, and a -B command line option to disable native code
		generation, even if GXemul was compiled with support for
		native code generation for the specific host CPU architecture.
20070201	Experimenting with a skeleton for native code generation.
		Changing the default behaviour, so that native code generation
		is now disabled by default, and has to be enabled by using
		-b on the command line.
20070202	Continuing the native code generation experiments.
		Making PCI interrupts work for Footbridge again.
20070203	More native code generation experiments.
		Removing most of the native code generation experimental code,
		it does not make sense to include any quick hacks like this.
		Minor cleanup/removal of some more legacy MIPS interrupt code.
20070204	Making i80321 interrupts work again (for NetBSD/evbarm etc.),
		and fixing the timer at 100 Hz.
20070206	Experimenting with removing the wdc interrupt slowness hack.
20070207	Lowering the number of dyntrans TLB entries for MIPS from
		192 to 128, resulting in a minor speed improvement.
		Minor optimization to the code invalidation routine in
		cpu_dyntrans.c.
20070208	Increasing (experimentally) the nr of dyntrans instructions per
		loop from 60 to 120.
20070210	Commenting out (experimentally) the dyntrans_device_danger
		detection in memory_rw.c.
		Changing the testmips and baremips machines to use a revision 2
		MIPS64 CPU by default, instead of revision 1.
		Removing the dummy i960, IA64, x86, AVR32, and HP PA-RISC
		files, the PC bios emulation, and the Olivetti M700 (ARC) and
		db64360 emulation modes.
20070211	Adding an "mp" demo to the demos directory, which tests the
		SMP functionality of the testmips machine.
		Fixing PReP interrupts some more. NetBSD/prep now boots again.
20070216	Adding a "nop workaround" for booting Mach/PMAX to the
		documentation; thanks to Artur Bujdoso for the values.
		Converting more of the MacPPC interrupt stuff to the new
		system.
		Beginning to convert BeBox interrupts to the new system.
		PPC603e should NOT have the PPC_NO_DEC flag! Removing it.
		Correcting BeBox clock speed (it was set to 100 in the NetBSD
		bootinfo block, but should be 33000000/4), allowing NetBSD
		to start without using the (incorrect) PPC_NO_DEC hack.
20070217	Implementing (slow) AltiVec vector loads and stores, allowing
		NetBSD/macppc to finally boot using the GENERIC kernel :-)
		Updating the documentation with install instructions for
		NetBSD/macppc.
20070218-19	Regression testing for the release.

==============  RELEASE 0.4.4  ==============


1 $Id: TODO,v 1.453 2007/02/19 01:34:42 debug Exp $
2
3 -------------------------------------------------------------------------------
4
5 Fix after the 0.4.4 release:
6 Fix the PowerPC DECR interrupt speed!
7
8 -------------------------------------------------------------------------------
9
10 Possible (relatively large) work packages to concentrate on in the future:
11
12 x) SMP:
13 Get SMP working again. It is pretty much broken since I started
14 the conversion from the old bintrans system to the new dyntrans system.
15 Add better Test machine demos for SMP in the demos directory.
16
17 x) Network:
18 Redesign of the networking subsystem, at least the NAT translation
19 part. The current way of allowing raw ethernet frames to be
20 transfered to/from the emulator via UDP should probably be extended
21 to allow the frames to be transmitted other ways as well.
22 Also adding support for connecting ttys (either to xterms, or to
23 pipes/sockets etc, or even to PPP->NAT or SLIP->NAT :-).
24
25 x) PCI:
26 Pretty much everything related to runtime configuration, device
27 slots, interrupts, whatever. The current code is very hardcoded
28 and ugly.
29
30 x) Debugging:
31 Think more about SMP debugging, etc. Right now, the
32 debugger is a mess. Also, a better connection to GDB would be
33 very nice to have.
34
35 x) Userland emulation:
36 Primary goals would be NetBSD and Linux syscall emulation.
37
38 And of course, there are _LOTS_ of minor TODOs spread out throughout
39 the source code, which must be fixed sooner or later.
40
41 -------------------------------------------------------------------------------
42
43 Some other things, in random order, that I'd like to fix: (Some items in
44 this list are probably out-to-date by now.)
45
46 Dyntrans:
47 x) Instruction combination collisions? How to avoid easily...
48 x) Think about how to do both SHmedia and SHcompact in a reasonable
49 way! (Or AMD64 long/protected/real, for that matter.)
50 x) 68K emulation; think about how to do variable instruction
51 lengths across page boundaries.
52 x) Dyntrans with valgrind-inspired memory checker. (In memory_rw,
53 it would be reasonably simple to add; in each individual fast
54 load/store routine = a lot more work, and it would become
55 kludgy very fast.)
56 x) Dyntrans with SMP... lots of work to be done here.
57 x) Dyntrans with cache emulation... lots of work here as well.
58 o) dev_mp doesn't work well with dyntrans yet
59 o) In general, IPIs, CAS, LL/SC etc must be made to work with dyntrans
60 x) Redesign/rethink the delay slot mechanism used for e.g. MIPS,
61 so that it caches a translation (that is, an instruction
62 word and the instr_call it was translated to the last
63 time), so that it doesn't need to do slow
64 to_be_translated for each end of page?
65 x) Program Counter statistics:
66 Per machine? What about SMP? All data to the same file?
67 A debugger command should be possible to use to enable/
68 disable statistics gathering.
69 Configuration file option!
70 x) Breakpoints:
71 o) Physical vs virtual addresses!
72 o) 32-bit vs 64-bit sign extension for MIPS, and others?
73 x) INVALIDATION should cause translations in _all_ cpus to be
74 invalidated, e.g. on a write to a write-protected page
75 (containing code)
76 x) 16-bit encodings? (MIPS16, ARM Thumb, 32-bit SH on SH64)
77 x) Lots of other stuff: see src/cpus/README_DYNTRANS
78 x) true recompilation backend? think carefully about this.
79 o) abstract syntax for emitting opcopdes
80 o) convert into native code only after an entire
81 block has been translated? probably best.
82 o) x86/amd64 code generator can be very similar... perhaps
83 o) branches to already translated code blocks can
84 link the blocks together
85 o) load/store are the most important.
86
87 Simple Valgrind-like checks?
88 o) Mark every address with bits which tell whether or not the address
89 has been written to.
90 o) What should happen when programs are loaded? Text/data, bss (zero
91 filled). But stack space and heap is uninitialized.
92 o) Uninitialized local variables:
93 A load from a place on the stack which has not previously
94 been stored to => warning. Increasing the stack pointer using
95 any available means should reset the memory to uninitialized.
96 o) If calls to malloc() and free() can be intercepted:
97 o) Access to a memory area after free() => warning.
98 o) Memory returned by malloc() is marked as not-initialized.
99 o) Non-passive, but good to have: Change the argument
100 given to malloc, to return a slightly larger memory
101 area, i.e. margin_before + size + margin_after,
102 and return the pointer + margin_before.
103 Any access to the margin_before or _after space results
104 in warnings. (free() must be modified to free the
105 actually allocated address.)
106
107 MIPS:
108 o) Nicer MIPS status bits in register dumps.
109 o) Alignment exceptions.
110 o) Floating point exception correctness.
111 o) Fix this? Triggered by NetBSD/sgimips? Hm:
112 to_be_translated(): TODO: unimplemented instruction:
113 000000000065102c: 00200800 (d) rot_00 at,zr,0
114 o) Some more work on opcodes.
115 x) MIPS64 revision 2.
116 o) Find out which actual CPUs implement the rev2 ISA!
117 o) DROTR32 and similar MIPS64 rev 2 instructions,
118 which have a rotation bit which differs from
119 previous ISAs.
120 o) EI and DI instructions for MIPS64/32 rev 2.
121 NOTE: These are _NOT_ the same as for R5900!
122 x) _MAYBE_ TX79 and R5900 actually differ in their
123 opcodes? Check this carefully!
124 o) Dyntrans: Count register updates are probably not 100% correct yet.
125 o) Refactor code for performance and readability/maintainability.
126 o) (Re)implement 128-bit loads/stores for R5900.
127 o) R4000 and others:
128 x) watchhi/watchlo exceptions, and other exception
129 handling details
130 o) R10000 and others: (R12000, R14000 ?)
131 x) The code before the line
132 /* reg[COP0_PAGEMASK] = cpu->cd.mips.coproc[0]->tlbs[0].mask & PAGEMASK_MASK; */
133 in cpu_mips.c is not correct for R10000 according to
134 Lemote's Godson patches for GXemul. TODO: Go through all
135 register definitions according to http://techpubs.sgi.com/library/tpl/cgi-bin/getdoc.cgi/hdwr/bks/SGI_Developer/books/R10K_UM/sgi_html/t5.Ver.2.0.book_263.html#HEADING334
136 and make sure everything works with R10000.
137 Then test with OpenBSD/sgi?
138 x) memory space, exceptions, ...
139 x) use cop0 framemask for tlb lookups
140 (http://techpubs.sgi.com/library/tpl/cgi-bin/getdoc.cgi/hdwr/bks/SGI_Developer/books/R10K_UM/sgi_html/t5.Ver.2.0.book_284.html)
141
142 SuperH:
143 x) SH4 interrupt controller:
144 x) Implement correct priorities of interrupts
145 x) SH4 DMA (0xffa00000)
146 x) SH4 UBC (0xff200000)
147 x) Store queues can copy 32 bytes at a time, there's no need to
148 copy individual 32-bit words. (Performance improvement.)
149 x) SH4 BSC (Bus State Controller)
150 x) SH4 RTC: Read the host's clock.
151 x) SH4 SCIF: Serial _interrupts_
152 x) Instruction tracing should include symbols for branch targets,
153 and so on, to make the output more human readable.
154 x) NetBSD/evbsh3, dreamcast, mmeye, hpcsh! Linux?
155 x) Replace pc-relative loads with immediate load, if within the
156 same page. (Similar to the same optimization for ARM.)
157 x) Floating point speed!
158 x) Floating point exception correctness.
159 x) Think carefully about how to implement SH5/SH64 (for evbsh5).
160
161 Dreamcast:
162 x) G2 DMA
163 x) LAN adapter (dev_mb8696x.c). NetBSD root-on-nfs.
164 x) PVR: Lots of stuff. See dev_pvr.c.
165 x) GDROM
166 x) Modem
167 x) PCI bridge/bus?
168 x) Maple bus:
169 x) Correct controller input
170 x) Mouse input
171 x) Software emulation of BIOS calls:
172 x) GD-ROM emulation: Use the GDROM device.
173 x) Use the VGA font as a fake ROM font. (Better than
174 nothing.)
175 x) Linux/dreamcast? (The gentoo kernel currently crashes.)
176 x) Make as many as possible of the KOS examples run!
177 x) More homebrew demos/games.
178 x) SPU: Sound emulation (ARM cpu).
179 x) VME processor emulation? "(Sanyo LC8670 "Potato")" according to
180 Wikipedia, LC86K87 according to Comstedt's page. See
181 http://www.maushammer.com/vmu.html for a good description of
182 the differences between LC86104C and the one used in the VME.
183
184 Transputer:
185 x) Implement support for Helios binaries.
186 x) Stack and register contents at startup?
187 x) Figure out how to boot an entire Helios distribution.
188 x) Implement all instructions. :)
189
190 RCA1802/RCA1805, CHIP8:
191 x) CHIP8 -> RCA180x conversion
192 x) Think about how to do dual-mode, variable-instr-length
193 ISAs, and switch between modes.
194 x) 1805 "extended" opcode -> trigger CHIP8 emulation?
195 That is, all calls 0NNN could point to 0x68 opcodes,
196 which, if running on a 1802 in CHIP8-emulation-mode,
197 would be manually interpreted.
198 x) Better solution:
199 CHIP8 calls to 00xx => handle at high level,
200 calls to 0xxx in general = call 180X machine code
201 (0000 = reboot?)
202 x) 1802 info: http://www.nyx.net/~lturner/public_html/Cosmac.html
203 and: http://www.elf-emulation.com/1802.html
204 x) 1805 extended opcodes: Implement at least disassembly support!
205 x) Keyboard input.
206 x) Sound (beep only).
207 x) Slow-down to correct speed? Wikipedia: "it was usually operated
208 at 3.58 MHz/2 to suit the requirements of the 1861 chip which
209 gave a speed of a little over 100,000 instructions per second"
210 (Note that _CHIP8_ emulation would then be even slower.)
211 x) SCHIP48 (Super) emulation:
212 Some more opcodes, 128x64 framebuffer, larger
213 sprites and fonts.
214
215 Alpha:
216 x) OSF1 PALcode, Virtual memory support.
217 x) PALcode replacement! PAL1E etc opcodes...?
218 x) Interrupt/exception/trap handling.
219 x) Floating point exception correctness.
220 x) More work on bootup memory and register contents.
221 x) More Alpha machine types, so it could work with
222 OpenBSD, FreeBSD, and Linux too?
223
224 SPARC:
225 o) Implement Adress space identifiers; load/stores etc.
226 o) Save/restore register windows etc! Both v9 and pre-v9!
227 o) Finish the subcc and addcc flag computation code.
228 o) Add more registers (floating point, control regs etc)
229 o) Exception/trap handling.
230 o) Disassemly of some more instructions?
231 o) Are sll etc 32-bit sign-extending or zero-extending?
232 o) Finish the GDB register stuff.
233 x) Floating point exception correctness.
234 o) SPARC v8, v7 etc?
235
236 Debugger:
237 o) How does SMP debugging work? Does it simply use "threads"?
238 What if the guest OS (running on an emulated SMP machine)
239 has a usertask running, with userland threads?
240 o) Try to make the debugger more modular and, if possible, reentrant!
241 o) Remove the emul command? (But show network info if showing
242 machines?)
243 o) Evaluate expressions within []? That would allow stuff like
244 cpu[x] where x is an expression.
245 o) Settings:
246 x) Special handlers for Write!
247 +) MIPS coproc regs
248 +) Alpha/MIPS/SPARC zero registers
249 +) x86 64/32/16-bit registers
250 x) Value formatter for resulting output.
251 o) see src/debugger.c for more
252
253 POWER/PowerPC:
254 x) Fix DECR timer speed, so it matches the host.
255 x) NetBSD/prep 3.x triggers a possible bug in the emulator:
256 <wdc_exec_command(0xd005e514,0xd60cdd30,0,8,..)>
257 <ata_get_xfer(0,0xd60cdd30,0,8,..)>
258 <0x26c550(&ata_xfer_pool,2,0,8,..)>
259 <0x35c71c(0x3f27000,0,52,8,..)>
260 <ata_exec_xfer(0xd005e4c8,0x3f27000,0,13,..)>
261 <atastart(0xd005e4c8,0x3f27000,0,13,..)>
262 <__wdccommand_start(0xd005e4c8,0x3f27000,0,13,..)>
263 <bsw1(&prep_isa_io_space_tag,0x800001f6,0,176,..)>
264 [ wdc: write to SDH: 0xb0 (sectorsize 2, lba=1, drive 1, head 0) ]
265 <wdcwait(0xd005e4c8,72,64,0xbb8,..)>
266 <0x198120(0xd005e4c8,72,64,0xbb8,..)>
267 <bsr1(&prep_isa_io_space_tag,0,0,0xbb8,..)>
268 <delay(100,0,0,0xbb8,..)>
269 Note: <bsr1(&prep_isa_io_space_tag,0,0,0xbb8,..)>
270 x) PPC optimizations; instr combs
271 x) 64-bit stuff: either Linux on G5, or perhaps some hobbyist
272 version of AIX? (if there exists such a thing)
273 x) macppc: adb controller; keyboard (for framebuffer mode)
274 x) make OpenBSD/macppc work (PCI controller stuff)
275 x) Floating point exception correctness.
276 x) Alignment exceptions.
277
278 PReP:
279 Clock time! ("Bad battery blah blah")
280
281 Algor:
282 o) Other models than the P5064?
283 o) PCI interrupts... needed for stuff like the tlp NIC?
284
285 BeBox:
286 o) Interrupts. There seems to be a problem with WDC interrupts
287 "after a short while", although a few interrupts get through?
288 o) Perhaps find a copy of BeOS and try it?
289
290 HPCmips:
291 x) Mouse/pad support! :)
292 x) A NIC? (As a PCMCIA device?)
293
294 AVR:
295 o) Everything.
296
297 ARM:
298 o) See netwinder_reset() in NetBSD; the current "an internal error
299 occured" message after reboot/halt is too ugly.
300 o) ARM "wait"-like instruction?
301 o) try to get netbsd/evbarm 3.x or 4.x running (iq80321)
302 o) make the xscale counter registers (ccnt) work
303 o) make the ata controller usable for FreeBSD!
304 o) Zaurus emulation, for e.g. OpenBSD/zaurus
305 o) Debian/cats crashes because of unimplemented coproc stuff.
306 fix this?
307
308 Test machines:
309 o) dev_fb block fill and copy
310 o) dev_fb draw characters (from the built-in font)?
311 o) dev_fb input device? mouse pointer coordinates and buttons
312 (allow changes in these to cause interrupts as well?)
313 o) Redefine the halt() function so that it stops "sometimes
314 soon", i.e. usage in demo code should be:
315 for (;;) {
316 halt();
317 }
318
319 Better CD Image file support:
320 x) Support CD formats that contain more than 1 track, e.g.
321 CDI files (?). These can then contain a mixture of e.g. sound
322 and data tracks, and booting from an ISO filesystem path
323 would boot from [by default] the first data track.
324 (This would make sense for e.g. Dreamcast CD images, or
325 possibly other live-CD formats.)
326
327 Networking:
328 x) Fix performance problems caused by only allowing a
329 single TCP packet to be unacked.
330 x) Don't hardcode offsets into packets!
331 x) Test with lower than 100 max tcp/udp connections,
332 to make sure that reuse works!
333 x) Make OpenBSD work better as a guest OS!
334 x) DHCP? Debian doesn't actually send DHCP packets, even
335 though it claims to? So it is hard to test.
336 x) Multiple networks per emulation, and let different
337 NICs in machines connect to different networks.
338 x) Support VDE (vde.sf.net)? Easiest/cleanest (before a
339 redesign of the network framework has been done) is
340 probably to connect it using the current (udp) solution.
341 x) Allow SLIP connections, possibly PPP, in addition to
342 ethernet?
343
344 Cache simulation:
345 o) Command line flags for:
346 o) CPU endianness?
347 o) Cache sizes? (multiple levels)
348 o) Separate from the CPU concept, so that multi-core CPUs sharing
349 e.g. a L2 cache can be simulated (?)
350 o) Instruction cache emulation is easiest (if separate from the
351 data cache); similar hack as the S;I; hack in cpu_dyntrans.c.
352 NOTE: if the architecture has a delay slot, then an instruction
353 slot can actually be executed as 2 instructions.
354 o) Data cache emulation = harder; each arch's load/store routines
355 must include support? running one instruction at a time and
356 having a cpu-dependant lookup function for each instruction
357 is another option (easier to implement, but very very slow).
358
359 Documentation:
360 x) Note about sandboxing/security:
361 Not all emulated instructions fail in the way they would
362 do on real hardware (e.g. a userspace program writing to
363 a system register might work in GXemul, but it would
364 fail on real hardware). Sandbox = contain from the
365 host OS. But the emulated programs will run "less
366 securely".
367 x) Try NetBSD/arc 4.x! (It seems to work with disk images!)
368 x) NetBSD/pmax 4 install instructions: xterm instead of vt100!
369 x) BETTER DEVICE EXAMPLES!
370 o) Move away from technical.html to somewhere new.
371 o) DEVICE_TICK
372 o) Implement example devices using interrupts, dyntrans
373 memory access, etc.?
374 x) Document the dyntrans core?
375 x) Rewrite the section about experimental devices, after the
376 framebuffer acceleration has been implemented, and demos
377 written. (Symbolic names instead of numbers; example
378 use cases, etc. Mention demo files that use the various
379 features?)
380 x) "a very simple linear framebuffer device (for graphics output)"
381 under "which machines does gxemul emulate" ==> better
382 description?
383 x) Better description on how to set up a cross compiler?
384 Example for MIPS64.
385 o) Automagic documentation generation?
386 x) machines, cpus, devices.
387 x) REMEMBER that several machines/devices can be in
388 the same source file!
389 o) Try to rewrite the install instructions for those machines
390 that use 3MAX into using CATS or hpcmips? (To remove the need
391 to use a raw ffs partition, using up all of the disk image.)
392
393 More generic out_of_memory error reporting, and check everywhere!
394 Causes: OpenBSD has low default limits for normal users.
395 Host is 32-bit? (32-bit hosts are limited to 4 GB or less
396 of userspace memory.)
397 You are actually low on RAM. (As trivial as this might sound,
398 Unix systems usually allow processes to allocate virtual
399 memory beyond the amount of RAM in the machine.)
400
401 The Device subsystem:
402 x) allow devices to be moved and/or changed in size (down to a
403 minimum size, etc, or up to a max size); if there is a collision,
404 return false. It is up to the caller to handle this situation!
405 x) NOTE: Translations must be invalidated, both for
406 registering new devices, and for moving existing ones.
407 cpu->invalidate translation caches, for all CPUs that
408 are connected to a specific memory.
409 x) keep track of interrupts and busses? actually, allowing any device
410 to be a bus might be a nice idea.
411 x) turn interrupt controllers into devices? :-)
412 x) refactor various clocks/nvram/cmos into one device?
413
414 PCI:
415 x) last write was ffffffff ==> fix this, it should be used
416 together with a mask to get the correct bits. also, not ALL
417 bits are size bits! (lowest 4 vs lowest 2?)
418 x) add support for address fixups
419 x) generalize the interrupt routing stuff (lines etc)
420
421 Clocks and timers:
422 x) DON'T HARDCODE 100 HZ IN cpu_mips_coproc.c!
423 x) Test the 8253? Right now it doesn't seem to be used?
424 x) NetWinder timeofday is incorrect!
425 x) Cobalt TOD is incorrect!
426 x) Go through all other machines, one by one, and fix them.
427
428 Config file parser:
429 o) Rewrite it from scratch!
430 o) Usage of any expression available through the debugger
431 o) Allow interrupt controllers to be added! and interrupts
432 to be used in more ways than before
433 o) Support for running debugger commands (like the -c
434 command line option)
435
436 Floating point layer:
437 o) make it common enough to be used by _all_ emulation modes
438 o) implement correct error/exception handling and rounding modes
439 o) implement more helper functions (i.e. add, sub, mul...)
440 o) non-IEEE modes (i.e. x86)?
441
442 Userland emulation:
443 x) Lots of stuff; freebsd and netbsd (and linux?) syscalls.
444 x) Dynamic linking? Hm.
445
446 Sound:
447 x) generic sound framework
448 x) add one or more sound cards as devices; add a testmachine
449 sound card first?
450 x) Dreamcast sound? Generic PCI sound cards?
451
452 ASC SCSI controller:
453 x) NetBSD/arc 2.0 uses the ASC controller in a way which GXemul
454 cannot yet handle. (NetBSD 1.6.2 works ok.) (Possibly a problem
455 in NetBSD itself, http://mail-index.netbsd.org/source-changes/
456 2005/11/06/0024.html suggests that.)
457 NetBSD 4.x seems to work? :)
458
459 Caches / memory hierarchies: (this is mostly MIPS-specific)
460 o) src/memory*.c: Implement correct cache emulation for
461 all CPU types. (currently only R2000/R3000 is implemented)
462 (per CPU, multiple levels should be possible, associativity etc!)
463 o) R2000/R3000 isn't _100%_ correct, just almost correct :)
464 o) Move the -S (fill mem with random) functionality into the
465 memory.c subsystem, not machine.c or wherever it is now
466 o) ECC stuff, simulation of memory errors? (Machine dependent)
467 o) More than 4GB of emulated RAM, when run on a 32-bit host?
468 (using manual swap-out of blocks to disk, ugly)
469 o) A global command line option should be used to turn
470 cache emulation on or off. When off, caches should be
471 faked like they are right now. When on, caches and
472 memory latencies should be emulated as correctly as
473 possible.
474
475 File/disk/symbol handling:
476 o) Make sure that disks can be added/removed during runtime!
477 (Perhaps this needs a reasonably large re-write.)
478 o) Remove some of the complexity in file format guessing, for
479 Ultrix kernels that are actually disk images?
480 o) Better handling of tape files
481 o) Read function argument count and types from binaries? (ELF?)
482 o) Better demangling of C++ names. Note: GNU's C++ differs from e.g.
483 Microsoft's C++, so multiple schemes must be possible. See
484 URL at top of src/symbol_demangle.c for more info.
485
486 Userland ABI emulation:
487 o) see src/useremul.c
488
489 Better framebuffer and X-windows functionality:
490 o) Generalize the update_x1y1x2y2 stuff to an extend-region()
491 function...
492 o) -Yx sometimes causes crashes.
493 o) Simple device access to framebuffer_blockcopyfill() etc,
494 and text output (using the built-in fonts), for dev_fb.
495 o) CLEAN UP the ugly event code
496 o) Mouse clicks can be "missed" in the current system; this is
497 not good. They should be put on a stack of some kind.
498 o) More 2D and 3D framebuffer acceleration.
499 o) Non-resizable windows? Or choose scaledown depending
500 on size (and center the image, with a black border).
501 o) Different scaledown on different windows?
502 o) Non-integral scale-up? (E.g. 640x480 -> 1024x768)
503 o) Switch scaledown during runtime? (Ala CTRL-ALT-plus/minus)
504 o) Bug reported by Elijah Rutschman on MacOS with weird
505 keys (F5 = cursor down?).
506 o) Keyboard and mouse events:
507 x) Do this for more machines than just DECstation
508 x) more X11 cursor keycodes
509 x) Keys like CTRL, ALT, SHIFT do not get through
510 by themselves (these are necessary for example
511 to change the font of an xterm in X in the
512 emulator)
513 o) Generalize the framebuffer stuff by moving _ALL_ X11
514 specific code to src/x11.c!
515

  ViewVC Help
Powered by ViewVC 1.1.26