Re: linux-kernel-digest V1 #106

Drew Eckhardt (
Fri, 07 Jul 1995 02:50:18 -0600

In message <199507060613.JAA28675@keos.cs.Helsinki.FI>, Linus.Torvalds@cs.helsi writes:
>If you want to protect some critical section _from_ a fast interrupt,
>you should do:
> unsigned long flags;
> save_flags(flags);
> cli();
> .. critical code ..
> sti();

You mean restore_flags(flags) instead of sti().

>> The x86 linux allows raw device access. Why the #$^&! doesn't 680x0 Lin
>> BTW, I only get 150K/sec through my A3000 SCSI to a quantum and a seacrate.
>> seems disgustingly slow. This is why I'd like to have raw device access so I
> can see
>> whether the driver or the cache is slow...Under the native OS, (still pre-em
>> with a buffer cache) I can see megabytes per second. What gives?

Probably a broken SCSI driver (broken == doesn't support enough scatter/gather
segments, where "enough" is at least 16 if you use clustering, 64 otherwise;
or is doing something silly like handshaking itself).