Hi!
> >With fbcon on machine with loaded PCI bus (usb modem), N can be ~500,
> >even more with really full screen. You know, fbcon is running with
> >interrupts disabled.
>
> I don't know why people keep on mentioning fbcon.
>
> The answer is really simple: if you want low latencies, don't use fbcon.
> (Or, if you want to, fix it). End of story. This is not a "Linux
> problem" - it's a "don't do that then" problem.
Problem is, that fixing fbcon is hard. Scrolling is operation that can
take 500msec. If printk() wants guarantees about being able to print
any time, whole fbcon has to be cli-ed even for normal
operation. Ouch.
Or we could give something up. Like giving up printk altogether [not
good] or declaring that printk is emergency-only and may mess up
screen.
> Adding ugly code to fix othe rugly code is not the answer. Fixing the
> ugly code is, or just deciding that something is unusable. But fbcon is
> not an argument for adding new low-latency features to the kernel.
This was not argument in favor of new low-latency features (or I did
not want it to be).
Pavel
-- I'm pavel@ucw.cz. "In my country we have almost anarchy and I don't care." Panos Katsaloulis describing me w.r.t. patents at discuss@linmodems.org- 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/
This archive was generated by hypermail 2b29 : Fri Jul 07 2000 - 21:00:21 EST