WinUAE 2.5.0 Beta series (Was 2.4.2) - Page 9

Marcuz · 01 September 2012, 13:33

Ok:

first Ti

Code:

 1 1: 00200858
  [H] 00000000
 2 1: 00200864
  [H] 00203704
      00203752 (C=00F95424 D=00203704) 'disk.resource'
 3 1: 00200870
  [H] 00000000
 4 2: 0020087c
  [S] 00208032 (C=00f8b0d2 D=00207FD0) 'card.resource'
  [S] 0020348a (C=00f8ea72 D=00203460) 'ciaa.resource'
  [S] 002002c8 (C=00f002a8 D=00F002A8) 'UAE filesystem'
 5 3: 00200888
  [S] <none>
 6 3: 00200894
  [S] 00203c8e (C=00faecfa D=00203C44) 'graphics.library'
  [S] 00207d78 (C=00fc745c D=00207C44) 'gameport.device'
  [S] 00207eba (C=00f84e94 D=00207DDC) 'timer.device'
  [S] 002002e8 (C=00f003b0 D=00F0FFFC) 'UAE heart beat'
 7 3: 002008a0
  [H] 00203c44
      00203cba (C=00FAEF2E D=00203C44) 'graphics.library'
 8 4: 002008ac
  [H] 00000000
 9 4: 002008b8
  [H] 00000000
10 4: 002008c4
  [H] 00000000
11 4: 002008d0
  [H] 00000000
12 5: 002008dc
  [H] 00000000
13 5: 002008e8
  [H] 00203704
      00203768 (C=00F9543A D=00203704) 'disk.resource'
14 6: 002008f4
  [S] 0020352a (C=00f8eb18 D=00203500) 'ciab.resource'
  [S] 0020801c (C=00f8afb6 D=00207FD0) 'card.resource'
15 6: 00200900
  [H] 00000000
16 7: 0020090c
  [S] <none>

second Ti

Code:

 1 1: ee1fb2eb
  [H] 00000000
 2 1: ee1fb2f7
  [H] 00000000
 3 1: ee1fb303
  [H] 00000000
 4 2: ee1fb30f
  [S] <none>
 5 3: ee1fb31b
  [S] <none>
 6 3: ee1fb327
  [S] <none>
 7 3: ee1fb333
  [H] 00000000
 8 4: ee1fb33f
  [H] 00000000
 9 4: ee1fb34b
  [H] 00000000
10 4: ee1fb357
  [H] 00000000
11 4: ee1fb363
  [H] 00000000
12 5: ee1fb36f
  [H] 00000000
13 5: ee1fb37b
  [H] 00000000
14 6: ee1fb387
  [S] <none>
15 6: ee1fb393
  [H] 00000000
16 7: ee1fb39f
  [S] <none>

r

Code:

  D0 00000000   D1 00C00020   D2 40000000   D3 00008000 
  D4 00000000   D5 00000000   D6 80000000   D7 C0000000 
  A0 00DFF000   A1 00000000   A2 00FE7350   A3 0020099E 
  A4 00F813F4   A5 00000000   A6 EE1FB297   A7 00202256 
USP  00209628 ISP  00202256 SFC  00000000 DFC  00000000 
CACR 00000009 VBR  00000000 CAAR 00000000 MSP  00000000 
T=00 S=1 M=0 X=0 N=1 Z=0 V=0 C=0 IMASK=3 STP=0
00000000 0000 0000                OR.B #$00,D0
Next PC: 00000004

m ra6

Code:

EE1FB297 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB2A7 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB2B7 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB2C7 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB2D7 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB2E7 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB2F7 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB307 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB317 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB327 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB337 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB347 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB357 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB367 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB377 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB387 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB397 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB3A7 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB3B7 0000 0000 0000 0000 0000 0000 0000 0000  ................
EE1FB3C7 0000 0000 0000 0000 0000 0000 0000 0000  ................

m ra1

Code:

00000000 0000 0000 EE1F B297 03FF 0A3E 00F8 0A3E  ...........>...>
00000010 0313 0313 0E10 0384 0E10 0A06 00F8 0A08  ................
00000020 00F8 0B28 00F8 0A0C 00F8 0A0E 00F8 0A10  ...(............
00000030 00F8 0A12 00F8 0A14 00F8 0A16 00F8 0A18  ................
00000040 00F8 0A1A 00F8 0A1C 00F8 0A1E 00F8 0A20  ............... 
00000050 00F8 0A22 00F8 0A24 00F8 0A26 00F8 0A28  ..."...$...&...(
00000060 00F8 0A2A 00F8 1090 00F8 10E8 00F8 1128  ...*...........(
00000070 00F8 1180 00F8 1220 00F8 126A 00F8 340A  ....... ...j..4.
00000080 00F8 0A98 00F8 0A9A 00F8 0A9C 00F8 0A9E  ................
00000090 00F8 0AA0 00F8 0AA2 00F8 0AA4 00F8 0AA6  ................
000000A0 00F8 0AA8 00F8 0AAA 00F8 0AAC 00F8 0AAE  ................
000000B0 00F8 0AB0 00F8 0AB2 00F8 0AB4 00F8 0AB6  ................
000000C0 00F8 0AB8 00F8 0ABA 00F8 0ABC 00F8 0ABE  ................
000000D0 00F8 0AC0 00F8 0AC2 00F8 0AC4 00F8 0AC6  ................
000000E0 00F8 0AC8 00F8 0ACA 00F8 0ACC 00F8 0ACE  ................
000000F0 00F8 0AD0 00F8 0AD2 00F8 0AD4 00F8 0AD6  ................
00000100 0000 0000 0000 0000 0000 0000 0000 0000  ................
00000110 0000 0000 0000 0000 0000 0000 0000 0000  ................
00000120 0000 0000 0000 0000 0000 0000 0000 0000  ................
00000130 0000 0000 0000 0000 0000 0000 0000 0000  ................

Toni Wilen · 01 September 2012, 13:40

Thanks. Now it makes more sense, something corrupts first 8 or so bytes of RAM, for example execbase is totally corrupted which explains the crash.

Still no idea what causes it and what does it have to do with screen change.

Next test: enter debugger again, "f 0" (no need for f 8 anymore), new command "w 0 0 8 w" (break to debugger if something writes to addresses from 0 to 7)

Trigger guru again.

Attach "r" and "H 50" outputs again. Should finally show the real cause for corruption.

Attach your configuration file too.

Marcuz · 01 September 2012, 14:24

Done, here the results.

r

Code:

  D0 00000000   D1 00DC0020   D2 40000000   D3 00008000 
  D4 00000000   D5 00000000   D6 80000000   D7 C0000000 
  A0 00DFF000   A1 00000000   A2 00FE7350   A3 0020099E 
  A4 00F813F4   A5 00000000   A6 EE1FB297   A7 00202256 
USP  00209628 ISP  00202256 SFC  00000000 DFC  00000000 
CACR 00000009 VBR  00000000 CAAR 00000000 MSP  00000000 
T=00 S=1 M=0 X=0 N=1 Z=0 V=0 C=0 IMASK=3 STP=0
00000000 0000 0000                OR.B #$00,D0
Next PC: 00000004

H 50

Code:

00F8139C 08ee 0007 0124           BSET.B #$0007,(A6, $0124) == $00200928
00F813A2 4e72 2000                STOP.L #$2000
00F813A6 60e6                     BT .B #$ffffffe6 == $00F8138E (T)
00F8138E 46fc 2700                MV2SR.W #$2700
00F81392 2650                     MOVEA.L (A0),A3
00F81394 2013                     MOVE.L (A3),D0
00F81396 6610                     BNE.B #$00000010 == $00F813A8 (T)
00F81398 52ae 0118                ADD.L #$00000001,(A6, $0118) == $0020091c
00F8139C 08ee 0007 0124           BSET.B #$0007,(A6, $0124) == $00200928
00F813A2 4e72 2000                STOP.L #$2000
00F813A6 60e6                     BT .B #$ffffffe6 == $00F8138E (T)
00F8138E 46fc 2700                MV2SR.W #$2700
00F81392 2650                     MOVEA.L (A0),A3
00F81394 2013                     MOVE.L (A3),D0
00F81396 6610                     BNE.B #$00000010 == $00F813A8 (T)
00F81398 52ae 0118                ADD.L #$00000001,(A6, $0118) == $0020091c
00F8139C 08ee 0007 0124           BSET.B #$0007,(A6, $0124) == $00200928
00F813A2 4e72 2000                STOP.L #$2000
00F813A6 60e6                     BT .B #$ffffffe6 == $00F8138E (T)
00F8138E 46fc 2700                MV2SR.W #$2700
00F81392 2650                     MOVEA.L (A0),A3
00F81394 2013                     MOVE.L (A3),D0
00F81396 6610                     BNE.B #$00000010 == $00F813A8 (T)
00F81398 52ae 0118                ADD.L #$00000001,(A6, $0118) == $0020091c
00F8139C 08ee 0007 0124           BSET.B #$0007,(A6, $0124) == $00200928
00F813A2 4e72 2000                STOP.L #$2000
00F813A6 60e6                     BT .B #$ffffffe6 == $00F8138E (T)
00F8138E 46fc 2700                MV2SR.W #$2700
00F81392 2650                     MOVEA.L (A0),A3
00F81394 2013                     MOVE.L (A3),D0
00F81396 6610                     BNE.B #$00000010 == $00F813A8 (T)
00F81398 52ae 0118                ADD.L #$00000001,(A6, $0118) == $0020091c
00F8139C 08ee 0007 0124           BSET.B #$0007,(A6, $0124) == $00200928
00F813A2 4e72 2000                STOP.L #$2000
00F81128 48e7 c0c6                MOVEM.L D0-D1/A0-A1/A5-A6,-(A7)
00F8112C 41f9 00df f000           LEA.L $00dff000,A0
00F81132 3228 001c                MOVE.W (A0, $001c) == $00dff01c,D1
00F81136 0801 000e                BTST.L #$000e,D1
00F8113A 67e2                     BEQ.B #$ffffffe2 == $00F8111E (F)
00F8113C c268 001e                AND.W (A0, $001e) == $00dff01e,D1
00F81140 2c78 0004                MOVEA.L $00000004,A6
00F81144 0801 0006                BTST.L #$0006,D1
00F81148 670e                     BEQ.B #$0000000e == $00F81158 (F)
00F81158 0801 0005                BTST.L #$0005,D1
00F8115C 670e                     BEQ.B #$0000000e == $00F8116C (F)
00F8115E 4cee 2200 0090           MOVEM.L (A6, $0090) == $ee1fb327,A1/A5
00F81164 2f0e                     MOVE.L A6,-(A7)
00F81166 486e ffdc                PEA.L (A6, -$0024) == $ee1fb273
00F8116A 4ed5                     JMP.L (A5)
00000000 0000 0000                OR.B #$00,D0

In the archive, two different configurations that both go guru, one with a harddrive directory the other with a hardfile as a boot disk: the one used in this dump above is the "- 2012 EoB.uae" with the harddrive directory.

Toni Wilen · 01 September 2012, 14:45

Hmm.. It appears something outside of WinUAE corrupts the data because memwatch breakpoint didn't trigger (You did get "Memwatch breakpoints enabled
0: 00000000 - 00000007 (8) W" message after "w 0 0 8 w" command?)

Does http://www.winuae.net/files/b/winuae.zip change anything? (Reserves virtual memory space from top of address space = different memory allocation)

Have you tried updating display drivers?

Marcuz · 01 September 2012, 15:00

Quote:

Originally Posted by Toni Wilen

You did get "Memwatch breakpoints enabled
0: 00000000 - 00000007 (8) W" message after "w 0 0 8 w" command?

Yes I noticed that line; and no, the winuae in the zip attached doesn't affect the guru happening; and the guru message is the same 80000003 00002121DA.
Do you need the dumps of this version too?

I haven't tried to update the graphic drivers because these ones (Catalyst 10.3) are stable, and the changelogs of newer ones seems to affect only more recent cards: since updating the Catalyst requires aggressive uninstall of the present one, i prefer to leave that as last possibility, so, do I really need to?

Toni Wilen · 01 September 2012, 15:11

Quote:

Originally Posted by Marcuz

Yes I noticed that line; and no, the winuae in the zip attached doesn't affect the guru happening; and the guru message is the same 80000003 00002121DA.
Do you need the dumps of this version too?

Bleh

Ok, lets try to adjust configuration more. Same in non-cycle exact? Immediate blitter?

Quote:

I haven't tried to update the graphic drivers because these ones (Catalyst 10.3) are stable, and the changelogs of newer ones seems to affect only more recent cards: since updating the Catalyst requires aggressive uninstall of the present one, i prefer to leave that as last possibility, so, do I really need to?

I never uninstalled anything when I had ATI card (last had one in beginning of 2012). And changes do change other cards, even if they are not mentioned, drivers are partially unified.

Would at least confirm if it is driver problem or not. Short hangs may be but guru problem probably isn't.

Marcuz · 01 September 2012, 15:14

I did also this test: as you know the emulation does not goes guru if booting from ADF or if just using an empty quickstart mode; so i made a clean A1200 quickstart and just added a system hardfile (the one used in the other tests).

This way there's no lag at boot time, differently to its own configuration (and I can't figure where this difference comes from by looking at the configuration file) but the emulation go guru nevertheless when changing the display.

Marcuz · 01 September 2012, 15:15

Quote:

Originally Posted by Toni Wilen

Bleh

Ok, lets try to adjust configuration more. Same in non-cycle exact? Immediate blitter?

I never uninstalled anything when I had ATI card (last had one in beginning of 2012). And changes do change other cards, even if they are not mentioned, drivers are partially unified.

Would at least confirm if it is driver problem or not. Short hangs may be but guru problem probably isn't.

I will test in 5 minutes [edit: sorry i misread your last sentence, got it now]

Marcuz · 01 September 2012, 15:25

Tested non-cycle exact, with and without immediate blitter, no change.
Ok, I will update the Catalyst.

Toni Wilen · 01 September 2012, 15:26

Did you try misc panel directdraw/direct3d setting?

Marcuz · 01 September 2012, 15:30

Quote:

Originally Posted by Toni Wilen

Did you try misc panel directdraw/direct3d setting?

yes, tried yesterday, and now again: no change.

Toni Wilen · 01 September 2012, 15:35

Ok.. Did you confirm what was first beta that has this problem? b13?

EDIT: What if you set all filter settings to default?

Marcuz · 01 September 2012, 16:23

Ok, first of all I've updated the card drivers to no difference;

then I've backtracked the betas and the first one to cause the guru is the b6: I'm real sorry because while I was sure the b14 was the first one, evidently the beta I was using before the 1GB hack was older: sorry Toni;

third, resetting the filters to default doesn't change anything.

Lastly, I've gotta run now, I'll test later tomorrow.

dlfrsilver · 01 September 2012, 18:49

Can you post the config file you use ? I have an ATI card myself with one of the latest driver (12.3).

Marcuz · 01 September 2012, 20:22

They are here in the quoted post:

Quote:

Originally Posted by Marcuz

In the archive, two different configurations that both go guru, one with a harddrive directory the other with a hardfile as a boot disk: the one used in this dump above is the "- 2012 EoB.uae" with the harddrive directory.

But as said the guru happens even with a quickstart mode + any HD as boot unit

Toni Wilen · 01 September 2012, 21:17

Quote:

Originally Posted by Marcuz

They are here in the quoted post:

But as said the guru happens even with a quickstart mode + any HD as boot unit

Even if "HD" is empty directory?
HDFs too? HDF + IDE0 controller? (not "UAE")

Marcuz · 01 September 2012, 22:48

Quote:

Originally Posted by Toni Wilen

Even if "HD" is empty directory?
HDFs too? HDF + IDE0 controller? (not "UAE")

Boot from the following and changing display mode:

- HDF ---> guru;
- empty directory ---> guru;
- workbench disk with any HD as side disk ---> guru;

workbench disk with no HD at all ---> works;
- HDF + IDE0 ---> works;

Toni Wilen · 01 September 2012, 23:04

I am now quite sure it is mouse driver related. Check that you don't have magic mouse or virtual mouse driver enabled.

(Quickstart won't affect these options = it won't remove requirement to attach config files

)

Marcuz · 01 September 2012, 23:40

ehm sorry but they are disabled even in the configurations: everything in the game ports panel is defaulted, mouse speed 100.
[edit]: tried to opt in magic mouse and to load mouse driver too, nothing changes

dlfrsilver · 02 September 2012, 00:53

i have found something really strange. I use the latest beta toni did, and it appears that in both config files, the mounted drives are bootable. Only the system partition must be like that, otherwise it can't work.

Strangely i made a CF 8gb under winuae, and i got myself strange result if more than one drive is bootable.

On a real amiga, having all partition bootable is a nonsense.

I modified a bit them, can you try them ?

01 September 2012, 15:35	#172
Toni Wilen WinUAE developer Join Date: Aug 2001 Location: Hämeenlinna/Finland Age: 49 Posts: 26,540	Ok.. Did you confirm what was first beta that has this problem? b13? EDIT: What if you set all filter settings to default? Last edited by Toni Wilen; 01 September 2012 at 15:57.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
WinUAE 2.5.1 beta series	Toni Wilen	support.WinUAE	69	22 December 2012 10:22
WinUAE 2.3.3 beta series	Toni Wilen	support.WinUAE	124	17 September 2011 15:48
WinUAE 2.3.2 beta series	Toni Wilen	support.WinUAE	79	31 May 2011 19:39
WinUAE 2.3.0 beta series (was 2.2.1)	Toni Wilen	support.WinUAE	229	22 September 2010 19:20

01 September 2012, 13:40	#162
Toni Wilen WinUAE developer Join Date: Aug 2001 Location: Hämeenlinna/Finland Age: 49 Posts: 26,540	Thanks. Now it makes more sense, something corrupts first 8 or so bytes of RAM, for example execbase is totally corrupted which explains the crash. Still no idea what causes it and what does it have to do with screen change. Next test: enter debugger again, "f 0" (no need for f 8 anymore), new command "w 0 0 8 w" (break to debugger if something writes to addresses from 0 to 7) Trigger guru again. Attach "r" and "H 50" outputs again. Should finally show the real cause for corruption. Attach your configuration file too.

01 September 2012, 14:45	#164
Toni Wilen WinUAE developer Join Date: Aug 2001 Location: Hämeenlinna/Finland Age: 49 Posts: 26,540	Hmm.. It appears something outside of WinUAE corrupts the data because memwatch breakpoint didn't trigger (You did get "Memwatch breakpoints enabled 0: 00000000 - 00000007 (8) W" message after "w 0 0 8 w" command?) Does http://www.winuae.net/files/b/winuae.zip change anything? (Reserves virtual memory space from top of address space = different memory allocation) Have you tried updating display drivers?

01 September 2012, 15:14	#167
Marcuz Registered User Join Date: Jun 2002 Location: . Age: 48 Posts: 5,562	I did also this test: as you know the emulation does not goes guru if booting from ADF or if just using an empty quickstart mode; so i made a clean A1200 quickstart and just added a system hardfile (the one used in the other tests). This way there's no lag at boot time, differently to its own configuration (and I can't figure where this difference comes from by looking at the configuration file) but the emulation go guru nevertheless when changing the display.

01 September 2012, 15:25	#169
Marcuz Registered User Join Date: Jun 2002 Location: . Age: 48 Posts: 5,562	Tested non-cycle exact, with and without immediate blitter, no change. Ok, I will update the Catalyst.

01 September 2012, 15:26	#170
Toni Wilen WinUAE developer Join Date: Aug 2001 Location: Hämeenlinna/Finland Age: 49 Posts: 26,540	Did you try misc panel directdraw/direct3d setting?

01 September 2012, 16:23	#173
Marcuz Registered User Join Date: Jun 2002 Location: . Age: 48 Posts: 5,562	Ok, first of all I've updated the card drivers to no difference; then I've backtracked the betas and the first one to cause the guru is the b6: I'm real sorry because while I was sure the b14 was the first one, evidently the beta I was using before the 1GB hack was older: sorry Toni; third, resetting the filters to default doesn't change anything. Lastly, I've gotta run now, I'll test later tomorrow.

01 September 2012, 18:49	#174
dlfrsilver CaptainM68K-SPS France Join Date: Dec 2004 Location: Melun nearby Paris/France Age: 46 Posts: 10,497	Can you post the config file you use ? I have an ATI card myself with one of the latest driver (12.3).

01 September 2012, 23:04	#178
Toni Wilen WinUAE developer Join Date: Aug 2001 Location: Hämeenlinna/Finland Age: 49 Posts: 26,540	I am now quite sure it is mouse driver related. Check that you don't have magic mouse or virtual mouse driver enabled. (Quickstart won't affect these options = it won't remove requirement to attach config files )

01 September 2012, 23:40	#179
Marcuz Registered User Join Date: Jun 2002 Location: . Age: 48 Posts: 5,562	ehm sorry but they are disabled even in the configurations: everything in the game ports panel is defaulted, mouse speed 100. [edit]: tried to opt in magic mouse and to load mouse driver too, nothing changes

Currently Active Users Viewing This Thread: 3 (0 members and 3 guests)