Alexander Viro writes:
> On Sun, 30 Jul 2000, Albert D. Cahalan wrote:
>> So, incoherent one, tell me how in hell I am supposed to write software
>> that tolerates people who set HZ to 666 one day and 42 the next.
>>
>> Note that previous procps releases have solidly demonstrated that
>> config files _will_ be wrong... It is hard enough getting people
>> to keep the System.map file updated, never mind a HZ file.
>> The tty name mapping file was eliminated partly for this reason.
>>
>> As an incentive for you, I'll show you the current horror:
>
> [snip the disgusting mess]
>
> So you can't speak decent C. Tell me something I wouldn't know, will you?
I called it a "horror". It is easy to fling shit, but much harder to
write code that tolerates a kernel with non-standard HZ. There are
real live users running with HZ set to 1000 and higher. Now people are
starting to pick multiples of MIDI frequencies.
> The only incentive it gives me is to say rm -rf /usr/src/base/procps
> and write a sane replacement, that would not be choke-full of useless
> features gathered from every friggin' abortion of Missed'em'V, would be
BSD has plenty to vomit about:
ps t? Cool, a shell wildcard means "no tty"
ps u12345 Some BSD won't accept a space, so users expect this
ps h One header per page, conflicting with the old Linux ps
ps axg If SunOS 4 is BSD to you, 'g' is used, else not
ps u42 The "user" display format, for process 42
ps t Processes on the current tty
ps t42 Processes on tty 42, NOT anything related to process 42
ps tu42 This horror is supposed to be tty 42 also, not PID 42
Going the BSD way, you are supposed to read kernel memory as root.
Have fun. I eagerly await your code for non-standard HZ support.
-
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 : Mon Jul 31 2000 - 21:00:32 EST