Messages in this thread | | | Subject | Re: [ANNOUNCE] block device interfaces changes | Date | Sun, 9 Jan 2000 20:46:20 +0000 (GMT) | From | Alan Cox <> |
| |
> > This wouldn't be flexible enough. Many ioctls pass multiple objects around. > "Many"? Which would those be? Haven't seen any yet.
Things like SIOCDEVPRIVATE are regularly used to pass data with pointers to other data.
> > For the standard form the existing ioctls encode their size. > Except that there are tons of legacy ioctls that don't ...
Yep. But we are stuck with those. Hindsight is a glorious thing.
> > size for the copy it has no code size improvement > > Code size?! I see no reason to even think about that in this context.
Always think about code size and speed.
> > devicecontrol in the form you give has no value outside a message passing > > microkernel. > > Somehow, my memory tells me that people (including Linus) liked something > very similar a short time back ...
Maybe, but I dont 8)
Alan
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |