Subject: Linux-Development Digest #154
From: Digestifier <Linux-Development-Request@senator-bedfellow.MIT.EDU>
To: Linux-Development@senator-bedfellow.MIT.EDU
Reply-To: Linux-Development@senator-bedfellow.MIT.EDU
Date:     Sun, 10 Oct 93 21:27:05 EDT

Linux-Development Digest #154, Volume #1         Sun, 10 Oct 93 21:27:05 EDT

Contents:
  Re: [PATCH] BogoBoost speedup for Linux (Superuser)
  Re: NO QUESTIONS HERE PLEASE / RTF FAQ (Re: -m486 doing nothing?) (rich@mulvey.com)
  Re: setitimer bug ? (Brandon S. Allbery)
  RFD: Removal of group "comp.os.linux.development" (Tim Peoples)
  Re: RFD: Removal of group "comp.os.linux.development" (Michael Griffith)
  WANTED: linux ioctl and tty doccumentation (Le Mauvais Sophiste)
  Re: RFD: Removal of group "comp.os.linux.development" (Lars Wirzenius)

----------------------------------------------------------------------------

From: root@fusion.cuc.ab.ca (Superuser)
Subject: Re: [PATCH] BogoBoost speedup for Linux
Date: Sun, 10 Oct 1993 19:58:17 GMT

pfau@cnj.digex.com (Thomas Pfau) writes:

>In article <CEFu8D.Aun@iat.holonet.net>,
>Thomas McWilliams <tmcwill@iat.holonet.net> wrote:
>>[ a long description of a speed enhancing patch which I won't reproduce
>> here ]

>This had absolutely no effect on my system.  BogoMips stayed at 9.98,
>dhryston gives the same result.  Is it possible that some newer systems
>are shipping with higher DRAM refresh timer values?


Hmm, I've been using a similar patch on my box for years..  I had a patch
for Coherent that did this, and I put it in Linux as well.  I haven't bothered
to figure out the internals of SCO Unix- so I wrote a little device driver
that just sets these values and de-installs itself.. (I use the same approach
with Linux to change the 16550 UART trigger).   I've also got a little 14 byte
.COM file for DOS too- it's the first program run on bootup- I get about a
5-6% performance increase on my machine by "fixing" refresh rates.  There
was a little program for DOS that basically kept increasing the refresh rate,
then running a little benchmark and displaying the results..  you ran it
until your machine croaked with a parity error, and then you could pick a
value to use based on those results..  As I remember it, the performance
vs. refresh timer looked something like

        Performance
        increase (%)
            ^
            |
          6-|                   *    *    * ~  *
            |             *                 ~
            |         *                     ~
            |      *                        ~
            |    *                          ~
            |  *                            ~
            | *                             ~
            |*                              ~
          0-+-------------------------------~---*------> refresh timer
                                                ^
                                              Crash!

The performance increase was asymptotic- once you reached a certain point,
increasing the refresh timer a *lot* would get you a tiny bit more performance,
you couldn't do better than about 6% increase, and most systems would lose
integrity (parity error) long before then..  The values I use on my systems
are: low byte=0x00, high byte=0x02.

My suggestion in regards to this patch are try it!  Then, if your system craps
out with parity errors, get rid of it.  It can't do any harm and if it works,
can only increase your system performance..

Regards,
c4
-- 
Christopher Lau- "Mr. Unix"    |     /       Fusion: Playing With Fire!
StarBright Research            |    / /      H + H -> He + 24 MeV
            --                 |   /_/_/_    "Bring back Trudeau!"
root,lauc@fusion.cuc.ab.ca     |____________ "This space for rent"

------------------------------

From: rich@mulvey.com
Subject: Re: NO QUESTIONS HERE PLEASE / RTF FAQ (Re: -m486 doing nothing?)
Date: Sun, 10 Oct 1993 20:49:01 GMT

Stefan Lukka (slukka@nyx.cs.du.edu) wrote:
: In article <CEo1tE.K3t@freenet.carleton.ca> ag010@Freenet.carleton.ca (Mark Swanson) writes:
: >
: >What Ian said was correct.  This IS the development channel as well... like Ian said.
: >Also, I found your rebuttle childish.

: More childish than calling someone who simply raised a
: development-related issue "Idiot who knows no netiquette"? Read
: Jackason's post carefully. Such an unprovoked attack is simply
: unacceptable, and should not be condoned.

   Wait a second here... I would love to see where you got that quote
from.  Ian never said ANYTHING about an "Idiot who knows no netiquette."
Quite frankly, you must be one incredibly paranoid person if you can read
THAT into his response to what has *unanimously* been acknowledged as
an incorrectly posted question.  Get a clue, friend.  You are wrong.

- Rich

-- 
Rich Mulvey                 Amateur Radio: N2VDS              Rochester, NY
rich@mulvey.com         "Ignorance should be painful."

------------------------------

Crossposted-To: comp.unix.programmer
From: bsa@kf8nh.wariat.org (Brandon S. Allbery)
Subject: Re: setitimer bug ?
Date: Sun, 10 Oct 1993 21:08:05 GMT

In article <299eui$a0n@mail.fwi.uva.nl> casper@fwi.uva.nl (Casper H.S. Dik) writes:
>Of course, other interface designs didn't do to well either.
>(The sigset/sighold/sigrelse interface as found in SysV has
>a fatal flaw that makes sighold()/sigrelse() pairs unusable).

Guess what?  System V took that from 4.1BSD.  Of course, one would think that
they'd have realized there was a reason it changed again in 4.2... :-)

++Brandon
-- 
Brandon S. Allbery         kf8nh@kf8nh.ampr.org          bsa@kf8nh.wariat.org
"MSDOS didn't get as bad as it is overnight -- it took over ten years
of careful development."  ---dmeggins@aix1.uottawa.ca

------------------------------

Crossposted-To: news.groups
From: tep@rzrbyte.fay.ar.us (Tim Peoples)
Subject: RFD: Removal of group "comp.os.linux.development"
Date: Sun, 10 Oct 1993 20:46:42 GMT




                        Request For Discussion

                    Removal of the USENET newsgroup
                      "comp.os.linux.development"


   This is an official Request for Discussion for the removal of
the group "comp.os.linux.development".  When and if this discussion
deems it necessary, a vote will be called on this matter.


Whereas,

     the USENET newsgroup "comp.os.linux.development" (c.o.l.d) was
     brought into being for the open (and unmoderated) discussion of
     software and system development issues as applied to the Linux
     operating system; and

Whereas,

     it is the convention on USENET newsgroups that a major part of
     any open discussion is the broad distribution of "questions" as
     they pertain to the agreed upon subject matter of any given
     newsgroup; and

Whereas,

     a small yet vocal group of active readers of c.o.l.d have ever
     so rightiously taken it upon themselves to be the self appointed
     guardians of content for messages posted to c.o.l.d; and

Whereas,

     this group of readers has made public announcement that the
     asking of questions will not be tolerated on the group c.o.l.d; and

Whereas,

     there is no need for a USENET newsgroup in which questions are
     not tolerated;

Therefore,

     I propose that the newsgroup "comp.os.linux.development" be
     removed from the USENET system.

Followups to this discussion have been directed to news.groups.


-- 

Timothy E. Peoples
tep@rzrbyte.fay.ar.us


------------------------------

From: grif@ucrengr.ucr.edu (Michael Griffith)
Crossposted-To: news.groups
Subject: Re: RFD: Removal of group "comp.os.linux.development"
Date: 10 Oct 1993 22:45:06 GMT

In article <1993Oct10.204642.6749@rzrbyte.fay.ar.us>,
Tim Peoples <tep@rzrbyte.fay.ar.us> wrote:
>
>                        Request For Discussion
>
>                    Removal of the USENET newsgroup
>                      "comp.os.linux.development"
>
[fluff deleted]
>Whereas,
>
>     this group of readers has made public announcement that the
>     asking of questions will not be tolerated on the group c.o.l.d; and
>
>Whereas,
>
>     there is no need for a USENET newsgroup in which questions are
>     not tolerated;
>
>Therefore,
>
>     I propose that the newsgroup "comp.os.linux.development" be
>     removed from the USENET system.
>

I think you (and your opponents) may have missed the entire point.
comp.os.linux.development should be for developers.  Developers should
be able to openly ask questions of each other.  I doubt that anyone has a
problem with that.  

The real issue is that crossposting newbie questions invariably make
their way to c.o.l.d where they don't belong since we have c.o.l.help.
Why should developer exchange have to be constantly interrupted with
messages about how to remove LILO or write an Xconfig?

Developer questions should be allowed.  Dumb questions should not.

--
                                                Michael A. Griffith
                                                grif@cs.ucr.edu




------------------------------

From: myddryn@netcom.com (Le Mauvais Sophiste)
Subject: WANTED: linux ioctl and tty doccumentation
Date: Sun, 10 Oct 1993 23:14:56 GMT


Does anyone have a manpage for ioctl? Or at least a list of commands to it?
Also, is tyy handling in linux posix? If not, what kind of doccumentation
gives a good overview?

thanks

--
        "Maturity is a slow death."

        "If Jesus died for our sins, would it not be disrespectful to live
         a life without sin, thus rendering his sacrifice in vain?"
                        - Jules Feiffer

Adam Sussman    - myddryn@netcom.COM
                - asussman@ucsd.edu

Chairman, UCSD chapter of the ACM


------------------------------

From: wirzeniu@klaava.Helsinki.FI (Lars Wirzenius)
Crossposted-To: news.groups
Subject: Re: RFD: Removal of group "comp.os.linux.development"
Date: 11 Oct 1993 01:21:13 +0200

tep@rzrbyte.fay.ar.us (Tim Peoples) writes:
>                         Request For Discussion
> 
>                     Removal of the USENET newsgroup
>                       "comp.os.linux.development"
> 
>    This is an official Request for Discussion for the removal of
> the group "comp.os.linux.development".  When and if this discussion
> deems it necessary, a vote will be called on this matter.

If it's official, why do I get the feeling it should have been posted
via news.announce.newgroups, like all other RFDs?

If it's not official, it looks too much like a hoax to be pleasant.

> Whereas,
> 
>      the USENET newsgroup "comp.os.linux.development" (c.o.l.d) was
>      brought into being for the open (and unmoderated) discussion of
>      software and system development issues as applied to the Linux
>      operating system; 

comp.os.linux.development is for discussion about development OF
LINUX, not random programs for Linux.  Porting issues are a borderline
case, but not particularly frequent as far as I have noticed.

> Whereas,
> 
>      a small yet vocal group of active readers of c.o.l.d have ever
>      so rightiously taken it upon themselves to be the self appointed
>      guardians of content for messages posted to c.o.l.d; and

Quite.  It should not be allowed to point out that some posting would
be better received in another group, or that the answer is in the FAQ.

> Whereas,
> 
>      this group of readers has made public announcement that the
>      asking of questions will not be tolerated on the group c.o.l.d; and

The assertation that questions are not tolerated is -- of course --
false, but the spirit of the post that started this round of
Linux-newsgroup-content-flaming was good.  c.o.l.d should not let be
drowned in postings about bugs in gcc (that were actually bugs in the
posters own code), or questions about how to use signal handlers, or
what the problem in someone's uucp setup is, or whether there's Motif
for Linux.

The Linux newsgroups have almost always (ever since comp.os.linux was
created?) had trouble with people who have little understanding about
how to select a newsgroup in which to post, whether to post at all,
and how to behave when posting.  One of the purposes of the recent
split -- which I posted the first RFD to (even though I voted against
the split) -- was to make the new Linux newsgroups work better than
the old single group.  Better meaning here that the amount of traffic
(per group) would be lower, it would be meaningfully divided, and
there would be less noise.  The wish was that there would be rather
less of the write-only attitude that permeated comp.os.linux of old,
i.e., that nobody was reading the group.

It isn't working as well as it could.

Some people think that if it is pointed out frequently enough that
certain kinds of posts aren't appropriate for a group, fewer people
will make mistakes.  I am in this camp.

Other people seem to think that doing this is rude (regardless of how
it is done, and I'm not saying that Ian's postings were completely
polite), arrogant, and should be stopped.

Instant flame war.

(Personally, I'm getting tired of the Linux newsgroups.  There's too
much noise, too little signal.  For several months now I've been
skipping almost everything based on subject lines.  This recent flame
war has convinced me it's time to take a vacation from these groups:
if I don't read them for a while, I'm likely to be able to read them
again without getting my blood pressure up.  Those who feel the need
can reach me by mail.)

--
Lars.Wirzenius@helsinki.fi  (finger wirzeniu@klaava.helsinki.fi)
It doesn't matter who you are, it's what you do that takes you far. --Madonna

------------------------------


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: Linux-Development-Request@NEWS-DIGESTS.MIT.EDU

You can send mail to the entire list (and comp.os.linux.development) via:

    Internet: Linux-Development@NEWS-DIGESTS.MIT.EDU

Linux may be obtained via one of these FTP sites:
    nic.funet.fi				pub/OS/Linux
    tsx-11.mit.edu				pub/linux
    sunsite.unc.edu				pub/Linux

End of Linux-Development Digest
******************************
