You are not logged in.
Hello there, I now got some progress to show on the VFS and FAT12 code. I can now load files!
After spending almost 2 days in bochs debugger - fixing bugs not only in the new code but also in my floppy driver - it sure feels great to finally being able to read a file!
Haha, nice to see you! That was unexpected to say the least. When i got back to coding on BOS I actually checked your webpage and stuff but all i got was a message about your blog being removed..
Btw, the triple fault is corrected, and I've tried the code on larger files - more then one cluster. Everything seems to work fine for read access
Might settle with read for now and do write at a later version, just have to clean everything up into user-callable functions. Also would have liked to have a executable format before releasing it so the FAT12 code can be useful (loading programs), but we'll see... at least something is happening now
Nice to see your triple fault is fixed and you're able to read files from a FAT12 FS :-). Having read only support is certainly nice and opens up the option to have user program support. Having write support will enable you to have a self hosting environment.
About the blog being removed, yeah.. I got fed up with my "blog" and I suck at updating everything so I decided to remove it. But, OS development got my attention since a few months. I didn't progress as much as I would like (job, study, personal life), but I'll have more time later in a few weeks.
Any plans btw to port your OS to a 64 bit platform, or are you just going to stick with 32 bit? :-)
I think that I will try to get the 32-bit version to a point where it's usable first. Maybe have support for loading 64-bit executables even with the OS being 32-bit. Anyway, eventually I might port it, but nothing planned right now.
Cool, i been away so have not been able to get any news (web free holiday) .
I see you have made good progress 8).
I will now debug my write code in bochs to see, if i can find why it does not work, in that environment.
But its great to see you are making progress.
One step forward, 2 steps back... or that's what it feels like now anyway. I've integrated the floppy driver into the VFS design so that in it's init function it registers itself as a device driver to the VFS, being added in the driver list. It also registers any found floppy drive into the VFS logical drive-list, so the VFS later can identify it as FAT12 and connect the fat12 driver to it.
but before all this my test where made with both fat12 code and floppy driver as freestanding from the VFS, and everything worked. the fat12 driver is still freestanding, but i tried moving all the low level reads to go via the VFS->floppy and everything breaked.
I guess i'll have another day or 2 in bochs debugger to look forward to.
Wow, I hate debugging!
Figured it out now finally, so back on track. Had written maybe 10k of new code and simple things like mixing up a pointer from edi->esi at one point and using sector read "1" instead of "0" in LBA mode (used to it from CHS) really set me back 10 hours or so.
Well, now to integrate the fat12 code into my VFS
haha, yeah it's painfully slow when you have to step through the code in bochs, checking register values at every other instruction but i'm not sure what i would have done without bochs debugger, would have taken me weeks to manually debug it with output to the screen and reboots for changes all the time
PS i have tried my code on larger files now and it works fine with for example the fasm change log that's 9kb