search

PetteriAimonen / AlterBIOS

Extensions and fixes for DSO Quad BIOS (SYS.HEX)
11 stars 7 forks source link

AlterBIOS v0.2 causes severe issues in GCC v1.24 Community Edition App #1

Open benryanau opened 11 years ago

benryanau commented 11 years ago

(Further to "stalkervan"'s posts on the seeedstudio DSO Quad forum thread..)

I've developed two severe issues after loading AlterBIOS v0.2 (AlterBIOS 0.2 (includes SYS 1.52) Running HW v2.70, and main app as GCC v1.24 APP Comm. Edition.

The measurements panel's readings all show 0.00 until I perform a screenshot - then many of the values show in kilovolts (kV). Calibration fails as well. Attached are two screenshots. NO_READING_IMAG000 KV_READINGS_IMAG001

I booted to DFU, loaded SYS_152, rebooted, and all was well again. Loaded ALT_02.HEX and the problem reoccured. Tried deleting the calibration/cfg file via USB - no change.

You're done some awesome work here and I really want to use your platform and apps (specially the frq response app!) but can't with this crippling bug. If I can be of assistance with more info, testing etc drop me a line Thanks

PetteriAimonen commented 11 years ago

Could you try reformatting the drive from PC (either Windows or Linux should be ok)? Use the normal FAT format utility, do not use 'dd' with the binary file.

Also check when you boot if it says "AlterBIOS 0.2: ok" on the top row.

I am very interested in receiving any feedback. For me, the SYS 1.52 filesystem corruption occurs very often and AlterBIOS fixes it for me. I'm currently using the "APP P100", but the calibration has never worked for me for some reason.

benryanau commented 11 years ago

Reformatted.. No joy, while the kilovolts has gone for now (althought I haven't tried a calibration) the meter readings are all zero, even when the scope is showing a DC positive voltage. Something is going on but I can't explain what it is. I did get filesystem corruption but I think it's unrelated to the problem with readouts that I'm talking about here. Is there a debug dump or something that would help you work out what's going on? Ah yes calibration, while it used to work now it doesn't, usually crashed the DSO.

benryanau commented 11 years ago

To reiterate, AlterBios 0.2 renders all metering functions inoperable - either the readings are zero'es, or nonsense numbers. The kilovolt readings are back. Appears only on certain imput ranges which is why I didn't notice at first. If it "works ok for you", perhaps it'd help if you sent me a package (rom/fpga/etc) that works for you and see if the problem appears here with the same codeset.

PetteriAimonen commented 11 years ago

Sorry I haven't had time to debug this yet. It may be because the original firmware does some funny tricks (storing the sector address in an integer and plugging it back), and maybe the community edition does the same.

I did some changes to the initialization routines. It is now possible (with QuadPawn 0.06) to use it without patching the official BIOS. QuadPawn now includes AlterBIOS, but without the BIOS patch. I.e. if you only install QuadPawn, you can use it but it won't patch the original BIOS.

benryanau commented 11 years ago

No worries mate, it's Christmas time :) I don't grasp the detail of what you're saying with the actual cause of the problem but I don't really need to I guess.

That's good news that QuadPAWN can run without AlterBIOS patching - I'll try the standalone version soon.

Just to be sure, we're taling about the issue being the weird vaules in readings here, correct? Not the FAT corruption you mentioned... that issue is seperate, and not as important as the "kilovolts" problem.

PetteriAimonen commented 11 years ago

Yeah, FAT corruption is what AlterBIOS fixes. Weird values is probably it somehow confusing the application. It wasn't entirely easy to hook the filesystem stuff, because the original is quite hackish.

benryanau commented 11 years ago

Ah, i thought AlterBIOS was needed to support the PAWN platform via some extended functions. I've browsed the code, no idea how the weird values are caused by AlterBIOS, I didn't think it touched anything dealing with measurements.

PetteriAimonen commented 11 years ago

Well, that also. But the reason for the bios hooks was the corruption caused by original BIOS. But now I made the hooks optional so that you can use Pawn without hooking the bios.

AlterBIOS doesn't touch the measurements, so probably the only way is that it somehow messes up the calibration file when the app tries to read it.

benryanau commented 11 years ago

Right - I actually suspected the calibration to be somehow central to the problem. Is it stored in a file on the FS or in NVRAM? I've tried backup/restore of FS files in case calibration was inside the config file but it didn't help.

So assuming it's related to calibration, where do we go from here to debug and resole the issue?

PetteriAimonen commented 11 years ago

Calibration is inside the config file. One thing you can try is whether the problem occurs after you remove the config file, i.e. run without calibration. But the application even checksums the file so I'm not sure what is going wrong. I think some people on the forum have got kilovolt readings even with original SYS.