PCem has bug in ATA HDD geometry

[Jeditobe]

handled wrong (zero) HDD geometry info returned by PCem VM.


http://alter.org.ua/soft/win/uni_ata/

Version 0.46e3

[ppgrainbow]

How did you know if the ATA HDD CHS geometry is not handled correctly?

[SarahWalker]

Would you care to say what this bug is? Your post is probably the most vague bug report I've ever seen.

[Battler]

I suspect the driver the user linked to is simply too new for PCem - it seems to be a driver for 21st century hard disks, not for the 1990's hard disk PCem emulates that is supported out of the box by all of the OS'es listed on the driver's page.

[Jeditobe]

This driver is default in ReactOS. And ReactOS works in PCem

[Jeditobe]

Would you care to say what this bug is? Your post is probably the most vague bug report I've ever seen.

Looks like PCem sets Current Sectors/Heads to zero in HDD identify block.
This caused div-by-zero exception.

[Jeditobe]

https://jira.reactos.org/browse/CORE-12909

[Battler]

It's still a bug in ReactOS, since the ATA standard has a flag in the identify block that tells whether or not the current sector/heads values can be trusted to be valid. ReactOS seems to completely ignore that and just assume those values can be trusted to be valid. I'd file a bug report with ReactOS, and point them at the ATA standard.

[Jeditobe]

Battler
To report a bug you need to register first at the main site (reactos.org), then wait an hour, then use the same login:pass for jira.reactos.org.
Thanx in advance.

[ecksemmess]

You're not seriously asking Battler to go file the bug report, are you? What possible reason could you have for not just doing it yourself?

[Jeditobe]

I am 100% serious.
Probably Battler knows this thing better than me to submit more accurate report than I could.

[gen_angry]

^ lol.

Or you can do it yourself and point them to this thread if you want, it's not a PCem issue.

[Jeditobe]

I can not because, I am not fully familiar with information Battler was talking about.