Installation on small hard disks FAT size mismatch between partition table and VBR

[andrewbird]

When using the install program on SHIPMOS/1_MOS5userSystemSN0530041410.zip with a small harddisk (IBM type1 10MB) the partition table created has entry

/tmp/hdtype1.img *       17 20807   20791 10.2M  1 FAT12

But the v4 BPB within its boot block has FAT16 type

$ ./exbootblk /tmp/hdtype1.img
Wrote bootblock into file /tmp/boot-306-4-17.blk

$ od /tmp/boot-306-4-17.blk
000000 eb 5a 90 54 53 4c 20 35 2e 30 31 00 02 08 01 00  >.Z.TSL 5.01.....<
000010 02 00 02 37 51 f8 08 00 11 00 04 00 11 00 00 00  >...7Q...........<
000020 00 00 00 00 80 00 29 1d 0a 46 11 4e 4f 20 4e 41  >......)..F.NO NA<
000030 4d 45 20 20 20 20 46 41 54 31 36 20 20 20 80 1c  >ME    FAT16   ..<

[the-grue]

Hi @andrewbird! That is a text field and it doesn't actually get used to identify the true FAT type. To do that, there is some math involved. I won't go into detail here, but if you're interested in the incantation, check this out:

FAT on disk format

Starts on page 14, FAT Type Determination.

Now, this IS an issue, as it is obviously confusing. Anyone want to take this on?

[ghost]

That .pdf link does not work in my browser. Do you have the file? If legal, we should keep some of these files on the project. Internet links tend to disappear.

[the-grue]

Good idea. Here it is.

fatgen103.pdf

[andrewbird]

@the-grue thanks for looking at this, I'm no longer subscribed to github pc-mos, so feel free to do as you like with this ticket. I do note from a quick glance, that the document you reference is only expressing reality as far as MS usage of the field, but does say this.

This string is informational only and is not used by Microsoft file system drivers to determine FAT type
because it is frequently not set correctly or is not present. See the FAT Type Determination section of
this document. This string should be set based on the FAT type though, because some non-Microsoft
FAT file system drivers do look at it.

[ghost]

This is suitable material for issues. Blogging about GEM and other things is not. Please use the Wiki for that.

[andrewbird]

Look closer!!! I said nothing about gem nor have i. You are the reason I unsubscribed.

[ghost]

No YOU are the reason you unsubscribed. Don't blame me for your personal problems. You don't know what I was referring to, because you unsubscribed.

@the-grue plays rough. If that doesn't suit you, happy trails.

[roelandjansen]

If andrewbird states that the reason to unsubscribe is you, you will have to take that story.
I feel that your comment @src153 is NOK.

So again: again - calm down. I won't tolerate this.

It's sad to hear Andrew is gone here.

[andrewbird]

@roelandjansen thanks and I don't want to cause trouble here. I only responded because @the-grue mentioned me directly, it seemed ignorant otherwise.

[ghost]

If you're running a project for pansies, I won't waste my time. You can delete me as a collaborator.

[ghost]

This is not a negotiation. I'm done. @sub205, good luck with your work.

[the-grue]

W O W
Someone pass the popcorn. Or the preparation-H because the butthurt is getting a little out of control in here.

Seriously people? @the-grue plays rough? Project for pansies? Driving away contributors?

I thought this was supposed to be fun and educational?

Anyway, I won't rant on. I will keep contributing because I am enjoying what I am doing. @roelandjansen has an interesting project here and I feel my work and contributions are valued.

GEM was my comment, btw, and I am spinning it up as a bolt-on to MOS if I can get it to play nice.

[andrewbird]

@the-grue, further to my last note I don't actually remember if after doing the install, I checked if the resultant partition after asking for FAT12 was either:
Partition marked FAT12, BPB FAT16, actual file structure FAT12
or
Partition marked FAT12, BPB FAT16, actual file structure FAT16

So the problem may not be the Partition table / VBR mismatch, it may be a Partition table / (VBR & filesystem) mismatch.

And just to be clear I never had a problem with anything you said to me, so no worries on my part 😄

[the-grue]

@andrewbird I didn't think so. Your presence has been missed. I am sure we would all like to have you back on the project.

[stsp]

If andrewbird states that the reason to unsubscribe is you, you will have to take that story.

Me too, but now, after reading this thread, seems like its
a good time to re-subscribe. :) Was the date bug fixed in
a mean time?

[the-grue]

Yeah, I fixed that awhile ago. (See #2 )

I have some RCs out for testing and set up a google group. Information is in the Wiki.

Welcome back!

[stsp]

Thanks!
I actually looked into the commits page initially,
but found nothing. Now it appears this fix can be
reached only by clicking the "Older" button a few
times, which means the development was quite
high. Which wasn't the case any of the time I was
subscribed! :)
Cool.