mame icon indicating copy to clipboard operation
mame copied to clipboard

a7000p - riscpc.cpp does not show any boot text

Open Tafoid opened this issue 3 years ago • 9 comments

Per request of @MooglyGuy and related to changes detailed in https://github.com/mamedev/mame/commit/e6d4824a8d42c9b9f5bfa2c9137c6422b370bd3c, this machine does not do much anymore.

a7000p-diff

Tafoid avatar Nov 26 '21 02:11 Tafoid

Posting a comment so that the issue can be assigned to me.

MooglyGuy avatar Nov 26 '21 09:11 MooglyGuy

Just so that the information doesn't get lost, I had a look at what's being executed by the machine on boot, and I suspect that it showing anything at all in 0.226 was out of pure chance, not out of any inherent validity of the emulation before my change versus after it.

This is a disassembly of what the machine is executing from power-on, and it does this in both 0.226 and current top-of-tree. Looking at the values in memory, it's pretty clearly executing what's meant to be data, not code. At that point, all bets are off.

https://gist.github.com/MooglyGuy/8448b4666966633c65c548710d9b974b

MooglyGuy avatar Nov 27 '21 11:11 MooglyGuy

The initial POST being odd has nothing to do with the current breakage. This must have more to do with the fact that it's trying to r/w the unmapped SIMM 1 slots at $0x18000000-$1cffffff and not throwing a data abort exception. May be worth to map them and see what happens.

angelosa avatar Dec 17 '21 03:12 angelosa

I don't care what you think, it's literally executing garbage as of the second instruction with -bios 0. Your opinion is invalid.

MooglyGuy avatar Dec 17 '21 08:12 MooglyGuy

I really don’t see how jumping straight into a data table has anything to do with unmapped SIMM slots,

cuavas avatar Dec 17 '21 08:12 cuavas

Incidentally, it is a valid analysis of why -bios 1 fails to boot, but that's not what this issue is about. Never has been. It's about the default BIOS, -bios 0, failing to so much as show anything on-screen.

MooglyGuy avatar Dec 17 '21 08:12 MooglyGuy

The source for RISC OS 3.71 is available, any clues at https://github.com/barryc-ro/RiscOS_371/blob/master/Sources/OS_Core/Kernel/s/Morris

Pernod70 avatar Dec 17 '21 13:12 Pernod70

I really don’t see how jumping straight into a data table has anything to do with unmapped SIMM slots,

It’s what it does at the end of the rainbow. We base our suppositions about what’s it’s doing at the end, not at the start. cfr. ARM reference

angelosa avatar Dec 18 '21 02:12 angelosa

Looks like it's now testing the PS/2 interface, failing because the hookup is wrong from my misunderstanding of how it works (it's direct serial from what I can tell). https://github.com/barryc-ro/RiscOS_371/blob/f6c81db7e90f727f3258f874692f751b692a14cb/Sources/OS_Core/Kernel/s/KbdResPC#L154

angelosa avatar Nov 28 '23 22:11 angelosa