Re: x86 TLB vs protection?

Posted on 2017-07-28, 18:49

by

by peterferrie

in PC Emulation

in PC Emulation

1. is not correct. Even if the address is in the TLB, you still need the protection check. If supervisor mode accessed a page that only supervisor can read, then user-mode accesses the same page, it should fail even though it's in the TLB. The TLB simply avoids the translation overhead.

---

Re: x86 TLB vs protection?

Posted on 2017-07-27, 20:11

by

by peterferrie

in PC Emulation

in PC Emulation

When a request is successful, the address and protection bits go into the TLB. The top 'x' bits (ways) are used to index into the array of 'y' entries (sets). Find the appropriate set, search within it for a match. The regular protection checks are still performed afterwards.

---

Re: Finding bugs in 8086-80386 emulation core?

Posted on 2017-07-24, 19:26

by

by peterferrie

in PC Emulation

in PC Emulation

There's no way to force ESP to be used in place of SP for stack access in 16-bit mode. It is used only for the parameters that are being read, not where it's being written.

---

Re: Finding bugs in 8086-80386 emulation core?

Posted on 2017-07-17, 19:16

by

by peterferrie

in PC Emulation

in PC Emulation

You have a bug in popa. Compare: 00:03:56:75.07504: 0000:7D3C (60)PUSHA 00:03:56:75.07504: Registers: 00:03:56:75.07520: EAX: 00000000, EBX: 00000700, ECX: 00000001, EDX: 00000000 00:04:10:57.06960: 0000:7D71 (61)POPA 00:04:11:20.06368: EAX: 00007c7a, EBX: 00000000, ECX: 00000000, EDX: 00000001 …

---

Re: Finding bugs in 8086-80386 emulation core?

Posted on 2017-07-17, 19:11

by

by peterferrie

in PC Emulation

in PC Emulation

It's used to find the first active partition, which might not be the first entry in the list.

---

Re: Finding bugs in 8086-80386 emulation core?

Posted on 2017-07-13, 17:00

by

by peterferrie

in PC Emulation

in PC Emulation

Is it writing to graphics memory at the time? If so, then that description sounds to me like the VESA technique, where page-faulting is used to know when to switch the planes.

---

Re: UniPCemu's 8086 vs newer CPU BCD instruction flags?

Posted on 2017-07-05, 18:20

by

by peterferrie

in PC Emulation

in PC Emulation

It seems that either that had a weird CPU that behaves in an undocumented way, or that their test results are actually wrong. I am certain that the values that you have are correct.

---

Re: UniPCemu's 8086 vs newer CPU BCD instruction flags?

Posted on 2017-06-30, 19:37

by

by peterferrie

in PC Emulation

in PC Emulation

Can you tell me some specific combinations of values that are failing the test? I have maps somewhere with every bit value for every possible combination of inputs, so if I can find them I can compare directly.

---

Re: UniPCemu's 8086 vs newer CPU BCD instruction flags?

Posted on 2017-06-29, 18:14

by

by peterferrie

in PC Emulation

in PC Emulation

This looks proper to me. AAD sets flags according to the 8-bit ADD.

---

Re: UniPCemu's 8086 vs newer CPU BCD instruction flags?

Posted on 2017-06-12, 06:51

by

by peterferrie

in PC Emulation

in PC Emulation

For if (((ALVAL&0xF)>9) || FLAG_AF) According to my notes, carry should not be altered here, AF should be the only flag that is set when true. OF is set on exit according to the result in all cases.

---

Re: UniPCemu 80286/80386+ protected mode problems?

Posted on 2017-05-29, 18:34

by

by peterferrie

in PC Emulation

in PC Emulation

- SHL/SHR/SAR always set with non-zero shift, cleared otherwise. Yes. - NEG sets when result low 4 bits not zero, cleared otherwise(essentially negated zero flag for low 4 bits). The flags result is equivalent to performing "0 SUB value". - INC sets when result low 4 bits is zero, cleared otherwise. …

---

Re: Pinball Construction Set

Posted on 2017-05-22, 18:57

by

by peterferrie

in Milliways

in Milliways

There are at least two .PB formats. The 1982 (pre-release version, Apple II only, the one that most pirates had, likely stolen, definitely incomplete) vs 1983 (final version on Apple II, RTM, possibly common to all other platforms).

---

Re: Pinball Construction Set

Posted on 2017-05-19, 16:13

by

by peterferrie

in Milliways

in Milliways

You can find some archives on Asimov, but they require the 1982 version of PCS to load them for editing.

---

Re: x86 loading CS after setting protected mode bit?

Posted on 2017-04-03, 06:13

by

by peterferrie

in PC Emulation

in PC Emulation

Even after switching to protected mode, old segment registers continue to work *exactly as before* until they are reloaded. That means that even after the jump to load CS, the other registers (DS, ES, FS, GS, SS) will still behave like real mode until they are reloaded. Note that reloading can …

---

Re: x64 sys files and exe files still contain a MS-DOS stub?

Posted on 2017-02-27, 23:30

by

by peterferrie

in DOS

in DOS

After looking at Windows 10's 64-bit WOW64 sys files and my own 64-bit version of UniPCemu(compiled using Visual Studio), I notice that it's still using a simply 32-bit EXE (MZ) header, as well as a complete MS-DOS stub? Why would 64-bit sys/exe files still contain a MS-DOS compatibility stub? Even …

---

Re: UniPCemu's 80286 emulation bug help?

Posted on 2017-02-16, 17:19

by

by peterferrie

in PC Emulation

in PC Emulation

I'm wondering about something. According to what I can find, opcode 8F with the ModR/M reg being 0 is POP modr/m. What about cases where the reg value is 1-7? Are they resulting in a #UD? Or are they counted as a normal POP modr/m instead(the opcode ignoring the reg field completely)? They trigger …

---

Re: UniPCemu's 80286 emulation bug help?

Posted on 2017-02-15, 20:21

by

by peterferrie

in PC Emulation

in PC Emulation

Cosmetic suggestion:00:05:37:54.09280: Writing to RAM: 00000A31=0D ( ) 00:05:37:54.09280: Writing to memory: 00000A32=0A ( ) 00:05:37:54.09280: Writing to RAM: 00000A32=0A ( ) Perhaps filter those so that they don't break the listing.

---

Re: UniPCemu's 80286 emulation bug help?

Posted on 2017-02-15, 20:17

by

by peterferrie

in PC Emulation

in PC Emulation

A problem is here: 00:05:37:46.09280: 020E:A555 (66C1E010)SHLD AX,10 00:05:37:46.09312: EAX: 0000020e, ... 00:05:37:46.09376: 020E:A559 (B850A6)MOVW AX, A650 00:05:37:46.09408: EAX: 00000000, EBX: 0000a700, ECX: 00000000, EDX: 00000000 -> should be SHLD EAX,10, so EAX=0x020E0000 Here's the other one …

---

Re: UniPCemu's 80286 emulation bug help?

Posted on 2017-02-14, 23:16

by

by peterferrie

in PC Emulation

in PC Emulation

Most of the 80386 instructions have been implemented(CRn/DRn register loads only checks for CPL in protected mode, the bits written to the registers are unchecked currently). Only a few (0F prefixed) instructions are left: - BT - SHLD(3 operands?) - BTS - SHRD(See SHLD) - IMUL - BTR - opcode BA( …

---

Re: UniPCemu's 80286 emulation bug help?

Posted on 2017-02-14, 23:12

by

by peterferrie

in PC Emulation

in PC Emulation

So CS DPL is checked against CS RPL and SS DPL is checked against SS RPL. But what about CPL? What is it based on and when is it reloaded? CPL is the name of the DPL in SS and CS (and they're supposed to be the same value). It is set when the selector is loaded if checks pass. Calls check SS first, …

---