CF and SD clash

[AshtonPete]

We have tried lots of trial setups on the Amiga 1200 over the weeks. One such test run was an 64 gb SD card srt up in winuae and 3.9, it works like a dream and I now use that as my master or primary card. But we like to continue to experiment, so I bought a 64 CF, and the mate ops it into WinUAE and sets it all up hoping for more 7million RAWREAD in sysinfo.

So we next copied at the perfect working 64g SD contents over the 64g CF.
But shock horror It doesn't work and keeps crashing with all too common reg guru screen.
Does anyone have a good idea what we are doing wrong? We just want to see if we can get more Rawread of at least 10 or more from the CF.
I am told its a possible Startup Sequence issue? TBH its swings and roundabouts and don't what to do next.
Both cards have been thoroughly tested with H2TestW and spot on.

I'll give a better example of prob, lets say the mate took home my CF, he would set it up on his WINUAE, spend a few hours prepping it and then trys it in his 1200, it works perfect up at his house on his Amiga1200, but walk down here and plug it in mine and we get the dreaded crash followed by red guru meditation bar. Any help would be much appreciated as the men in white have put me in a straight jacket and sedated me.

[thomas]

Which program gurus (check with set echo on in startup-sequence)?
What's the contents of the guru?

What's the difference between your 1200 and his? Do you have a better CPU? Does he have more RAM? Does he have other harddrives in the same machine? Do you have other harddrives?

Have you checked MaxTransfer?
Do you use third-party file systems? Have you put the file system into the RDB?

[AshtonPete]

He is standard 1200 with 3.1 roms which I just gave him, he has no accelerator.

At the moment I have the Blizzard 030 with 256 ram both chips running at 60ns

I have the FastATA cr (latest)

I only have one hard drive at 64 gig. he too only uses one HD... A lot what he preps is done in WINUAE

I will pass your post onto him when he gets out of bed LoL

[AshtonPete]

Ok Thomas, these are the crashing numbers after it states a software failure.
8000 0004 followed by 000103F0
Block size is 1024 and max Transfer is 1FE00
Mask = 0X76FE
File Sys = SFS

Hope this is helpful for you as its just over my head...

Thanks
Pete

[Amiga1992]

Does SFS run on an EC020 setup?

[AshtonPete]

Mate says Yes

[thomas]

Originally Posted by AshtonPete

Ok Thomas, these are the crashing numbers after it states a software failure.
8000 0004 followed by 000103F0

When does it crash? Can you enter the early startup menu? Can you boot without startup-Sequence?

If you can, enter set echo on and execute s:startup-sequence. Then watch the output and remember the last line before it gurus.

Mask = 0X76FE

Is this real? Or shall that mean there are 6 F's between 0x7 and E?

If it's real, then please change it to 0x7ffffffe.

Originally Posted by Akira

Does SFS run on an EC020 setup?

As I understand it, the person with the EC020 prepared and tested the drive, so yes, it must work. The one which crashes is the 030 + FastATA.

Originally Posted by AshtonPete

I have the FastATA cr (latest)

Are the FastATA drivers installed?

Or anything else for making more than 4GB available on the IDE bus? IDEfix for example?

Is the AmigaOS 3.9 ROM update deactivated?

How big is the boot partition?

How big is (are) the other partition(s)?

[AshtonPete]

My mate will log in later this evening and give all details better than I can. It seems I don't cut the mustard with all this stuff and Its aggravating the life out of me, I just type what I am told to... So please hang fire and the mate will be deffo be on later... Thanks Pete

PS 0x76FE was a communication error and 0x7ffffffe is correct...

[steveh]

Ok Here go's

First off I began with Petes OS3.9 install that AmigaKit had done for him on a 30GB PC SSD , I made a backup of it using Winuae, then I checked what filesystem was on the RDB and other settings Like
Block size 1024
Max transfer 0x0001fe00
Mask 0x7FFFFFFE, and noted them down.

Then Swapped over to the 64GB SDHC card, then prepped and partitioned it with the desired size partitions, installed SFS on to the RDB set all the above settings to match.

Then of course simply copied the OS3.9 install over to the newly prepped SDHC card, I then quit WinUae removed the SDHC card put it into the SD to Ide adapter in his 1200 and powered on, and it booted as I expected it to.

So I then returned the SDHC card back to the PC fired up Winuae and followed the procedure to install Classic Workbench for OS3.9 which went flawlessly and rebooted within winuae without any problems.

So out come the SDHC card put it back in the 1200 powered on and it booted without any problems as I expected it to do.

While in the 1200 we then updated the FastATA drivers and it rebooted again without any problems.

Now Please bear in mind this IS A 64GB SDHC card and it works 100% perfect.

Then I did all the above, Prepping EXACTLY the same but this time with a 64GB Kingston X266 Compact Flash Card, then simply copied the contents of the boot partition from the now 100% working SDHC card to the boot partition of the Compact Flash Card, which then Booted 100% perfect in winuae, but it crashes as soon as the card is accessed in the real A1200, Probably in the 1st micro second during boot.


BTW a fresh install Of standard 3.1 workbench works and boots fine also a Fresh install of OS3.9 works fine.

So whats wrong ? its a bloody head scratcher I know that.

[thomas]

Please answer my questions.

Immediate crash after reading the partition table usually means that the boot partition is not accessible by the file system or the file system is missing. Most likely it is bigger than 4GB.

If it crashes after loading fastata.driver then probably the PIO settings are not supported by the CF card. Use FastATAPrefs to reduce the PIO modes.

IDEfix, patched scsi.device and the like should not be installed when using a FastATA.

[steveh]

Boot Partition is 2GB on Both Card Types , Yeah I know its overkill.

And yet if these " IDEfix, patched scsi.device " are present in the startup sequence along with fast ata, which boots fine on the SD Card, then it can only be the PIO mode that is causing the problem with the CF, TBH I never thought about PIO I just assumed PIO 5 what the SD card is set to and gives the fastest raw read times of almost 7ml bytes or is it bits per second under sysinfo, would also work at pio 5, This is all new to me this CF SD stuff as the last time I used my 1200 was 1997 with an old 6GB hitachi Laptop HD.

Will check all this tomorrow.

Thanks for your reply.

P.s I will post the startup sequence tomorrow if that helps.

[thomas]

I am sorry but it is very difficult to help you. It seems like you are not willing to narrow the problem systematically. All your assumptions are wrong as long as they have not been proven.

You have not checked whether you can reach the early startup menu. You have not checked whether you can boot without startup-sequence. You have not checked whether it crashes before or after fastata.driver. You have not checked if unnecessary and possibly incompatible device drivers have been deactivated/uninstalled.

"which boots fine on the SD Card" means nothing. SD and CF cards are completely different devices. That a combination of software works with an SD card does not mean that it will work with a CF card, too. You should deactivate all device drivers which could interfere with fastata.driver.

" it can only be the PIO mode" is wrong as long as you haven't proven that the crash is caused solely by fastata.driver and not by a combination of other software.

You wrote that a clean install of 3.1 or 3.9 boots fine. Have you installed fastata.driver to these installations? And did it still boot after that? If yes, then it's not fastata.driver or the PIO mode but rather someting in ClassicWB.

[steveh]

I can understand your frustration Thomas but I couldn't answer your questions last night because I had already left peters home, the Amiga is his not mine I am reluctant to bring it home with me in case anything drastically go's wrong with it and I am then left with a hefty repair bill, but I will be back there in the next hour and check all that you have asked me to check.

So if you want me to continue that's fine, if not, that's also fine.

Regards Steve

[AshtonPete]

Can anyone plz tell me how I can get rid of this?

At bootup there is a System request box, in it it says: "Please Insert Volume ENV in any drive.) Retry or Cancel buttons.

If I press cancel, it reboots and I get normal screen :dunno:

[daxb]

Read post #7 again and do the "set echo on" thing. Then you will see why it happens.

[AshtonPete]

I'll have a crack at it and see what happens. Ta!

[thomas]

Originally Posted by AshtonPete

Can anyone plz tell me how I can get rid of this?

At bootup there is a System request box, in it it says: "Please Insert Volume ENV in any drive.) Retry or Cancel buttons.

If I press cancel, it reboots and I get normal screen :dunno:

Please post your Startup-Sequence from the top up to SetPatch. There is something in it which requires environment variables but ENV has not yet been set up.

[AshtonPete]

I just phoned Steveh, he is coming down to mine doing that for me Shouldn't be too long.

[AshtonPete]

Got into early startup boot with no startup

In cli window I Type in
Set echo on
execute startup-sequence

"
C:ATA3.driver QUIET
C:stack 8192
;ClassicWB Startup-Sequence
;--------------------------
;OS39

;Start OS3.9 Checks
IF EXISTS C:SquirrelSCSI
C:SquirrelSCSI
EndIF
Set SCSIUpdate 1
IF EXISTS C:IDEFix
C:IDEFix
Set SCSIUpdate 0
EndIF
IF EXISTS SYS:Expansion/XSurfIDE
Set SCSIUpdate 0
EndIF
IF
"

Thats all that is echoed then if I cancel it continues to boot after it soft resets itself then boots fine.

Here is the full startup sequence below
C:ATA3.driver QUIET
C:stack 8192
;ClassicWB Startup-Sequence
;--------------------------
;OS39

;Start OS3.9 Checks
IF EXISTS C:SquirrelSCSI
C:SquirrelSCSI
EndIF
Set SCSIUpdate 1
IF EXISTS C:IDEFix
C:IDEFix
Set SCSIUpdate 0
EndIF
IF EXISTS SYS:Expansion/XSurfIDE
Set SCSIUpdate 0
EndIF
IF EXISTS SYS:Prefs/Env-Archive/NOSCSIUPDATE
Set SCSIUpdate 0
EndIf
IF $SCSIUpdate EQ 1
C:SetPatch QUIET
ELSE
C:SetPatch SKIPROMUPDATES "scsi.device" QUIET
EndIF
Unset SCSIUpdate
;End OS3.9 Checks

C:GetMouseInput LOCAL

IF $MouseInput EQ 1
Unset MouseInput
C:Assign >NIL: ENV: RAM:
C:Assign >NIL: T: RAM:
C:Execute S:Maxmem-Sequence
EndCLI >NIL:
EndIF

IF $MouseInput EQ 2
Unset MouseInput
C:Execute S:Bootmenu-Sequence
EndCLI >NIL:
EndIF

Unset MouseInput

IF EXISTS C:StackAttack
C:StackAttack
EndIF

C:Version >NIL:
FailAt 21

IF EXISTS SYS:WBstartupD
C:Rename >NIL: SYS:WBStartup SYS:WBStartupM
C:Rename >NIL: SYS:WBStartupD SYS:WBStartup
EndIF

Resident >NIL: C:IconX PURE
Resident >NIL: C:WBRun PURE
Resident >NIL: C:Assign PURE
Resident >NIL: C:Execute PURE
Resident >NIL: C:SetEnv PURE REPLACE

IF EXISTS L:env-handler
C:MakeDir RAM:T RAM:Clipboards
L:env-handler
ELSE
C:MakeDir RAM:T RAM:Clipboards RAM:ENV RAM:ENV/Sys
C:Copy >NIL: ENVARC: RAM:ENV ALL NOREQ
Assign >NIL: ENV: RAM:ENV
EndIF

IF EXISTS C:PatchRAM
C:PatchRAM
EndIF

IF EXISTS C:TagLiFE
C:TagLiFE
EndIF

Assign >NIL: T: RAM:T
Assign >NIL: CLIPS: RAM:Clipboards
Assign >NIL: REXX: S:
Assign >NIL: PRINTERS: DEVS:Printers
Assign >NIL: KEYMAPS: DEVS:Keymaps
Assign >NIL: LOCALE: SYS:Locale
Assign >NIL: LIBS: SYS:Classes ADD
Assign >NIL: HELP: LOCALE:Help DEFER

IF EXISTS C:BlazeWCP
C:BlazeWCP
EndIF

IF EXISTS SYS:System/FBlit/FBlit
SYS:System/FBlit/FBlit
SYS:System/FBlit/FText
EndIF

BindDrivers
C:Mount >NIL: DEVSOSDrivers/~(#?.info)

C:LoadMonDrvs

SetEnv Language "english"
SetEnv Workbench $Workbench
SetEnv Kickstart $Kickstart
UnSet Workbench
UnSet Kickstart

IF EXISTS C:CyberBugFix
C:CyberBugFix
EndIF

IF EXISTS C:IconBeFast
C:IconBeFast 0
EndIF

C:AddDataTypes REFRESH QUIET
IF EXISTS SYS:System/VPrefs/VisualPrefs
Run > NIL: SYS:System/VPrefs/VisualPrefs
EndIF
C:IPrefs
SYS:System/FPalette/FPPrefs
C:ConClip

Path >NIL: RAM: C: SYS:Utilities SYS:Rexxc SYS:System S: SYS:Prefs SYS:WBStartup SYS:Tools SYS:Tools/Commodities

SYS:System/REXXMast >NIL:

IF EXISTS S:User-Startup
Execute S:User-Startup
EndIF

Resident Execute REMOVE
Resident Assign REMOVE

IF EXISTS C:StackAttack
C:StackAttack AUTO
EndIF

C:LoadWB

EndCLI >NIL:

Regards Steve.

[daxb]

I see more or less at least one problem. IDEFix and ata3.driver are launched and that is a very bad idea.

It seems (I guess) that following line is responsible for your ENV Reqeuster:
IF EXISTS SYS:Prefs/Env-Archive/NOSCSIUPDATE

Who the fuck had done all the IF EXISTS stuff? If I`m not completely wrong this startup-sequence has been posted already somewhere on this forum and I did some comments about.