Helge Skrivervik

> > I was under the impression that reducing the fs size to 32M would be sufficient to have fsck work OK? > > We should be able to check...

More thoughts on multiple opens of `/dev/eth`: >>we could probably just allow multiple opens on /dev/eth, period. >>Let me check that out, maybe start by allowing two /dev/eth opens >Ok,...

OK; thanks @ghaerr - I'll follow this model, test it with the ne2k driver and add it to the others when it seems solid. —M > 14. jun. 2022 kl....

Wow @ghaerr - that was fast (again). Thank you for diving into this - in a way that is significantly more ambitious than I had envisioned, and will deliver a...

Thank you @ghaerr, I like the speed of this! > I'm having reservations about the suggested initialization scheme though. > > I think I miscommunicated the issue with IRQs -...

OK, sounds good. I'm quite looking forward to it. I was pondering if expanding the eth struct would be good for placing the NIC params, but given the speed at...

NO, I was referring to the struct in drivers/char/eth.c where you keep the file ops, one per driver. Then again I don't have the big picture at this time. -M...

A lot of heavy ELKS usage brings up an old 'friend' - `mv`, or rather, the `rename` system call, so I'm adding it to my wishlist. An old friend because...

Thank you @ghaerr, and I agree with you, this has to start with (and possibly be limited to) minix fs. Actually - I'm not sure whether the DOS MOVE command...

Thank you @ghaerr - that was a cold shower indeed. Very thorough - much appreciated. Maybe we could lower the level of ambition, ignore the havoc a -9 signal may...