Re: CFD: bogomips CPU performance metric
On Sun, 10 Oct 1999, Chris Costello wrote: On Sun, Oct 10, 1999, Laurence Berland wrote: I like the idea as an optional LINT parameter that is NOT in the generic kernel. Might make some linux people feel comfortable with the switch, or might prove useful under some odd circumstances, but I agree it'd be silly to include it by default (kindof on the level of a splash screen) I disagree. BogoMIPS is a completely meaningless measurement and does not belong in our source tree as it will only produce repository bloat. Right. If you're worried that there's something wrong with your CPU speed, you can always bootverbose and check out the clocks and bzero tests. -- Brian Fundakowski Feldman \ FreeBSD: The Power to Serve! / [EMAIL PROTECTED]`--' To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
I disagree. BogoMIPS is a completely meaningless measurement and does not belong in our source tree as it will only produce repository bloat. I would agree.. BogoMIPS actually stands for "Bogus, Misleading Indication of Processor Speed"... In an old Linux Journal article I have (will dig it up upon request), it describes how BogoMIPS is calculated -- it's VERY processor-dependant, and only really (sort of) useful for comparing motherboards with the same processor on them. Switch from an Intel 486/66 to an AMD 486/100, and the numbers do NOT match up to what you'd think And if you change to a Pentium, forget it. The biggest thing it's good for is as an ego-booster Of course, my dual-PII/450 box has a bigger BogoMIPS rating than any machine at work.. :) --mike To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
I like the idea as an optional LINT parameter that is NOT in the generic kernel. Might make some linux people feel comfortable with the switch, or might prove useful under some odd circumstances, but I agree it'd be silly to include it by default (kindof on the level of a splash screen) Robert Sexton wrote: On Thu, Sep 02, 1999 at 10:40:30AM -0700, Nick Sayer wrote: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. We don't require doing any such thing, and so adding it would be purely cosmetic. However, I allege that cosmetic things aren't in and of themselves evil, so long as they don't break anything in the process. I'd have to agree with the "Lets be more professional" crowd. How about as a LINT option? "If you need something so banal, you can turn it on yourself" -- Robert Sexton, [EMAIL PROTECTED] "Imagine if every Thursday your shoes exploded if you tied them the usual way. This happens to us all the time with computers, and nobody thinks of complaining." -- Jeff Raskin, interviewed in Doctor Dobb's Journal To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message -- Laurence Berland, Stuyvesant HS Debate Windows 98: n. useless extension to a minor patch release for 32-bit extensions and a graphical shell for a 16-bit patch to an 8-bit operating system originally coded for a 4-bit microprocessor, written by a 2-bit company that can't stand for 1 bit of competition. http://stuy.debate.net icq #7434346aol imer E1101 The above email Copyright (C) 1999 Laurence Berland All rights reserved To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
On Sun, Oct 10, 1999, Laurence Berland wrote: I like the idea as an optional LINT parameter that is NOT in the generic kernel. Might make some linux people feel comfortable with the switch, or might prove useful under some odd circumstances, but I agree it'd be silly to include it by default (kindof on the level of a splash screen) I disagree. BogoMIPS is a completely meaningless measurement and does not belong in our source tree as it will only produce repository bloat. -- |Chris Costello [EMAIL PROTECTED] |A bug in the code is worth two in the documentation. ` To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
Wilko Bulte [EMAIL PROTECTED] writes: "The Wrath of Satoshi" (free interpretation of "The Wrath of Khan") 8-) The question is, does "The Wrath of Satoshi" also have Kirstie Alley in the role of Lt. Saavik? And if it doesn't, what else does it have that makes it worth watching? Too bad she's a scientologist. DES (http://www.moviebbs.com/gallery/samples/s-025-ka.jpg) -- Dag-Erling Smorgrav - [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
Wilko Bulte wi...@yedi.iaf.nl writes: The Wrath of Satoshi (free interpretation of The Wrath of Khan) 8-) The question is, does The Wrath of Satoshi also have Kirstie Alley in the role of Lt. Saavik? And if it doesn't, what else does it have that makes it worth watching? Too bad she's a scientologist. DES (http://www.moviebbs.com/gallery/samples/s-025-ka.jpg) -- Dag-Erling Smorgrav - d...@flood.ping.uio.no To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
As Satoshi - Ports Wraith - Asami wrote ... * From: Wilko Bulte [EMAIL PROTECTED] * As Chris Costello wrote ... * I don't think the local Ports Wraith would be amazingly happy * * No. He will turn into a Ports Wrath Hey, if you're going to make fun of me, at least CC: me or something so I know! ;) "The Wrath of Satoshi" (free interpretation of "The Wrath of Khan") 8-) -- | / o / / _ Arnhem, The Netherlands- Powered by FreeBSD - |/|/ / / /( (_) BulteWWW : http://www.tcja.nl http://www.freebsd.org To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
Oliver Fromme wrote: Nick Sayer wrote in list.freebsd-hackers: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. It's not a metric of CPU performance. It's just a meaningless number, and its relation to the actual performance of the machine is very questionable. My evil streak makes me want to do something like this under boot -v, but with a twist... Figure out a vague partity with the Linux numbers and then make sure it reports about 3 times higher along with something like this: "This cpu rates X (meaningless) bogomips.." .. where the reported number is significantly higher than linux would report on the same hardware... That would serve two purposes... one to undermine the ``benchmark'' and the other to tease the various clueless folks with. Cheers, -Peter To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
* Satoshi - Ports Wraith - Asami ([EMAIL PROTECTED]) [990903 12:14]: * From: Jonathan Lemon [EMAIL PROTECTED] * hw.clockrate: 132 * Doing this for Pentium and better systems should be trivial. Doing * it for 486 and lower would just add a timing loop. Doing it for SMP * would be harder. * * hw.cpu0.clockrate: 233 * hw.cpu1.clockrate: 233 * * Possibly? The implementer gets to pick a better name than these. hw.clockrate.cpu0: 233 hw.clockrate.cpu1: 233 (cpu0 only for uni-procs, of course). This sounds like a good solution. I don't know why we can't detect speed on SMP boxes and I doubt the code will make things clear to me [seeing my third rank coding abilities]. But we could probe the first CPU, get it's clockspeed and since it's SMP be certain that all other CPU's would have the same speed. This might not be the cleanest solution should AMP type systems ever surface. [If they ever] Else, anyone care to detail exactly why we can't detect it? cheers, -- Jeroen Ruigrok van der Werven asmodai(at)wxs.nl The BSD Programmer's Documentation Project http://home.wxs.nl/~asmodai Network/Security SpecialistBSD: Technical excellence at its best Knowledge comes, but wisdom lingers. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
As Satoshi - Ports Wraith - Asami wrote ... * From: Wilko Bulte wi...@yedi.iaf.nl * As Chris Costello wrote ... * I don't think the local Ports Wraith would be amazingly happy * * No. He will turn into a Ports Wrath Hey, if you're going to make fun of me, at least CC: me or something so I know! ;) The Wrath of Satoshi (free interpretation of The Wrath of Khan) 8-) -- | / o / / _ Arnhem, The Netherlands- Powered by FreeBSD - |/|/ / / /( (_) BulteWWW : http://www.tcja.nl http://www.freebsd.org To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
Oliver Fromme wrote: Nick Sayer wrote in list.freebsd-hackers: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. It's not a metric of CPU performance. It's just a meaningless number, and its relation to the actual performance of the machine is very questionable. My evil streak makes me want to do something like this under boot -v, but with a twist... Figure out a vague partity with the Linux numbers and then make sure it reports about 3 times higher along with something like this: This cpu rates X (meaningless) bogomips.. .. where the reported number is significantly higher than linux would report on the same hardware... That would serve two purposes... one to undermine the ``benchmark'' and the other to tease the various clueless folks with. Cheers, -Peter To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
* Satoshi - Ports Wraith - Asami (as...@freebsd.org) [990903 12:14]: * From: Jonathan Lemon jle...@americantv.com * hw.clockrate: 132 * Doing this for Pentium and better systems should be trivial. Doing * it for 486 and lower would just add a timing loop. Doing it for SMP * would be harder. * * hw.cpu0.clockrate: 233 * hw.cpu1.clockrate: 233 * * Possibly? The implementer gets to pick a better name than these. hw.clockrate.cpu0: 233 hw.clockrate.cpu1: 233 (cpu0 only for uni-procs, of course). This sounds like a good solution. I don't know why we can't detect speed on SMP boxes and I doubt the code will make things clear to me [seeing my third rank coding abilities]. But we could probe the first CPU, get it's clockspeed and since it's SMP be certain that all other CPU's would have the same speed. This might not be the cleanest solution should AMP type systems ever surface. [If they ever] Else, anyone care to detail exactly why we can't detect it? cheers, -- Jeroen Ruigrok van der Werven asmodai(at)wxs.nl The BSD Programmer's Documentation Project http://home.wxs.nl/~asmodai Network/Security SpecialistBSD: Technical excellence at its best Knowledge comes, but wisdom lingers. To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
* From: Jonathan Lemon [EMAIL PROTECTED] * What I want is a simple new readable sysctl, something like: * * hw.clockrate: 132 * * I think that this would be useful both for development (how fast * is that stupid machine down in the bunker?), and system admininstration * (who needs a cpu upgrade this year?). I like that. * Doing this for Pentium and better systems should be trivial. Doing * it for 486 and lower would just add a timing loop. Doing it for SMP * would be harder. * * hw.cpu0.clockrate: 233 * hw.cpu1.clockrate: 233 * * Possibly? The implementer gets to pick a better name than these. How about hw.clockrate.cpu0: 233 hw.clockrate.cpu1: 233 (cpu0 only for uni-procs, of course). Satoshi To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
As Chris Costello wrote ... On Thu, Sep 02, 1999, Karl Pielorz wrote: Chris Costello wrote: No, since it would just be useless bloat in the source tree. If we must have it, how about a port? - I'm definitely for the "this isn't a good idea" crowd, When I was using Linux, I thought it was 'cute'... I've grown up a bit since then... I don't think the local Ports Wraith would be amazingly happy No. He will turn into a Ports Wrath :) -- | / o / / _ Arnhem, The Netherlands- Powered by FreeBSD - |/|/ / / /( (_) BulteWWW : http://www.tcja.nl http://www.freebsd.org To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
* From: Wilko Bulte [EMAIL PROTECTED] * As Chris Costello wrote ... * I don't think the local Ports Wraith would be amazingly happy * * No. He will turn into a Ports Wrath Hey, if you're going to make fun of me, at least CC: me or something so I know! ;) -PW (W?) To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
Hey, if you're going to make fun of me, at least CC: me or something so I know! ;) -PW (W?) oh oh! He's upgraded his acronym ... He's gonna take over the world ! -- ISIS/STA, T.P.270, Joint Research Centre, 21020 Ispra, Italy To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
* From: Jonathan Lemon jle...@americantv.com * What I want is a simple new readable sysctl, something like: * * hw.clockrate: 132 * * I think that this would be useful both for development (how fast * is that stupid machine down in the bunker?), and system admininstration * (who needs a cpu upgrade this year?). I like that. * Doing this for Pentium and better systems should be trivial. Doing * it for 486 and lower would just add a timing loop. Doing it for SMP * would be harder. * * hw.cpu0.clockrate: 233 * hw.cpu1.clockrate: 233 * * Possibly? The implementer gets to pick a better name than these. How about hw.clockrate.cpu0: 233 hw.clockrate.cpu1: 233 (cpu0 only for uni-procs, of course). Satoshi To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
As Chris Costello wrote ... On Thu, Sep 02, 1999, Karl Pielorz wrote: Chris Costello wrote: No, since it would just be useless bloat in the source tree. If we must have it, how about a port? - I'm definitely for the this isn't a good idea crowd, When I was using Linux, I thought it was 'cute'... I've grown up a bit since then... I don't think the local Ports Wraith would be amazingly happy No. He will turn into a Ports Wrath :) -- | / o / / _ Arnhem, The Netherlands- Powered by FreeBSD - |/|/ / / /( (_) BulteWWW : http://www.tcja.nl http://www.freebsd.org To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
* From: Wilko Bulte wi...@yedi.iaf.nl * As Chris Costello wrote ... * I don't think the local Ports Wraith would be amazingly happy * * No. He will turn into a Ports Wrath Hey, if you're going to make fun of me, at least CC: me or something so I know! ;) -PW (W?) To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
Hey, if you're going to make fun of me, at least CC: me or something so I know! ;) -PW (W?) oh oh! He's upgraded his acronym ... He's gonna take over the world ! -- ISIS/STA, T.P.270, Joint Research Centre, 21020 Ispra, Italy To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
CFD: bogomips CPU performance metric
Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. We don't require doing any such thing, and so adding it would be purely cosmetic. However, I allege that cosmetic things aren't in and of themselves evil, so long as they don't break anything in the process. I would like to generate a number that will hopefully be reasonably compatible with the one Linux spits out. The best method I have come up with is to have a similar (the same?) count down loop in assembler. Have it count down from 1,000,000 and see how much nanotime() has gone by. NANSPERSEC/nansused = bogomips. A 1 bogomips machine will take an extra second to do this (anything likely to be even able to run FreeBSD should exceed 1 BM - yes, ha ha), and a kBM CPU can do it in 1 ms. Perhaps in the future a prescaler might be required, but this whole thing is just really chrome anyway. Would anyone scream and projectile-vomit if I added this to identcpu.c? S/MIME Cryptographic Signature
Re: CFD: bogomips CPU performance metric
On Thu, Sep 02, 1999 at 10:40:30AM -0700, Nick Sayer wrote: Description of BogoMIPS deleted Would anyone scream and projectile-vomit if I added this to identcpu.c? I might. :-) Why exactly, except to keep up with the Linux kidz, do we need this? I recognize that this is solely a cosmetic change, but one of the things I hold over the heads of the Linux folks I deal with is the fact that FreeBSD is a professional quality operating system which doesn't need useless blinking lights like BogoMIPS. Chalk me up as one of the people who considers "Linux works like that" as a negative. Regards, --Keith Stevenson-- -- Keith Stevenson System Programmer - Data Center Services - University of Louisville [EMAIL PROTECTED] PGP key fingerprint = 4B 29 A8 95 A8 82 EA A2 29 CE 68 DE FC EE B6 A0 To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
There was such a thing in 386BSD and FreeBSD1.0 I certainly thing it was a worth-while thing. I'd try make the loop as similar to the Linux one so that they are comparable. On Thu, 2 Sep 1999, Nick Sayer wrote: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. We don't require doing any such thing, and so adding it would be purely cosmetic. However, I allege that cosmetic things aren't in and of themselves evil, so long as they don't break anything in the process. I would like to generate a number that will hopefully be reasonably compatible with the one Linux spits out. The best method I have come up with is to have a similar (the same?) count down loop in assembler. Have it count down from 1,000,000 and see how much nanotime() has gone by. NANSPERSEC/nansused = bogomips. A 1 bogomips machine will take an extra second to do this (anything likely to be even able to run FreeBSD should exceed 1 BM - yes, ha ha), and a kBM CPU can do it in 1 ms. Perhaps in the future a prescaler might be required, but this whole thing is just really chrome anyway. Would anyone scream and projectile-vomit if I added this to identcpu.c? To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
Description of BogoMIPS deleted Would anyone scream and projectile-vomit if I added this to identcpu.c? I might. :-) Why exactly, except to keep up with the Linux kidz, do we need this? I recognize that this is solely a cosmetic change, but one of the things I hold over the heads of the Linux folks I deal with is the fact that FreeBSD is a professional quality operating system which doesn't need useless blinking lights like BogoMIPS. Chalk me up as one of the people who considers "Linux works like that" as a negative. I'm with you on this, Keith. I'd rather that we kept the professional FreeBSD look and feel. If we look too much like Linux, then people will just use Linux. -DG David Greenman Co-founder/Principal Architect, The FreeBSD Project - http://www.freebsd.org Creator of high-performance Internet servers - http://www.terasolutions.com Pave the road of life with opportunities. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
Nick Sayer wrote in list.freebsd-hackers: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. It's not a metric of CPU performance. It's just a meaningless number, and its relation to the actual performance of the machine is very questionable. We don't require doing any such thing, and so adding it would be purely cosmetic. However, I allege that cosmetic things aren't in and of themselves evil, so long as they don't break anything in the process. It would break user's confidence in the seriousness of the system. Would anyone scream and projectile-vomit if I added this to identcpu.c? I would, FWIW. Regards Oliver -- Oliver Fromme, Leibnizstr. 18/61, 38678 Clausthal, Germany (Info: finger userinfo:[EMAIL PROTECTED]) "In jedem Stück Kohle wartet ein Diamant auf seine Geburt" (Terry Pratchett) To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
As Nick Sayer wrote ... so long as they don't break anything in the process. I would like to generate a number that will hopefully be reasonably compatible with the one Linux spits out. The best method I have come up with is to have a similar (the same?) count down loop in assembler. Have it count down from 1,000,000 and see how much nanotime() has gone by. NANSPERSEC/nansused = bogomips. A 1 bogomips machine will take an extra second to do this (anything likely to be even able to run FreeBSD should exceed 1 BM - yes, ha ha), and a kBM CPU can do it in 1 ms. Perhaps in the future a prescaler might be required, but this whole thing is just really chrome anyway. Would anyone scream and projectile-vomit if I added this to identcpu.c? I really don't see any merit in adding Linux-isms like this. Sounds a bit like the 'hack of the day' to me. I feel FreeBSD is well respected for it's stability etc. Not for it's 'me too' Linux-isms without any practical use. Mind you, I don't say Linux does not have stuff that is useful for inclusion in FreeBSD. I just stay BogoMips is what it calls itself: bogus, and should be kept from the FreeBSD kernel. Just my Dfl 0.02 W/ -- | / o / / _ Arnhem, The Netherlands- Powered by FreeBSD - |/|/ / / /( (_) BulteWWW : http://www.tcja.nl http://www.freebsd.org To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
There was such a thing in 386BSD and FreeBSD1.0 I remember no such thing doing a 'bogomips' to compare against Linux. Certainly not in 386BSD. Nate I certainly thing it was a worth-while thing. I'd try make the loop as similar to the Linux one so that they are comparable. On Thu, 2 Sep 1999, Nick Sayer wrote: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. We don't require doing any such thing, and so adding it would be purely cosmetic. However, I allege that cosmetic things aren't in and of themselves evil, so long as they don't break anything in the process. I would like to generate a number that will hopefully be reasonably compatible with the one Linux spits out. The best method I have come up with is to have a similar (the same?) count down loop in assembler. Have it count down from 1,000,000 and see how much nanotime() has gone by. NANSPERSEC/nansused = bogomips. A 1 bogomips machine will take an extra second to do this (anything likely to be even able to run FreeBSD should exceed 1 BM - yes, ha ha), and a kBM CPU can do it in 1 ms. Perhaps in the future a prescaler might be required, but this whole thing is just really chrome anyway. Would anyone scream and projectile-vomit if I added this to identcpu.c? To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
On Thu, Sep 02, 1999, Robert Sexton wrote: I'd have to agree with the "Lets be more professional" crowd. How about as a LINT option? "If you need something so banal, you can turn it on yourself" No, since it would just be useless bloat in the source tree. -- |Chris Costello [EMAIL PROTECTED] |Supercomputer: Turns a CPU-bound problem into |an I/O-bound problem. - Ken Batcher `--- To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
On Thu, Sep 02, 1999, Nick Sayer wrote: Would anyone scream and projectile-vomit if I added this to identcpu.c? Yes, I would. The way I interpret it, along with "useless blinking light", is as follows: BogoMIPS is but the combination of "Bogus" and an acronym for "Meaningless Indicator of Processor Speed." :) -- |Chris Costello [EMAIL PROTECTED] |You might have mail. `-- To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
It was there... when I added the code to calibrate the delay loops originally and added the DELAY macro, it printed out the callibration factor.. (DELAY was originally a spin loop) It wasn't called 'BOGOMIPS...' On Thu, 2 Sep 1999, Nate Williams wrote: There was such a thing in 386BSD and FreeBSD1.0 I remember no such thing doing a 'bogomips' to compare against Linux. Certainly not in 386BSD. Nate I certainly thing it was a worth-while thing. I'd try make the loop as similar to the Linux one so that they are comparable. On Thu, 2 Sep 1999, Nick Sayer wrote: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. We don't require doing any such thing, and so adding it would be purely cosmetic. However, I allege that cosmetic things aren't in and of themselves evil, so long as they don't break anything in the process. I would like to generate a number that will hopefully be reasonably compatible with the one Linux spits out. The best method I have come up with is to have a similar (the same?) count down loop in assembler. Have it count down from 1,000,000 and see how much nanotime() has gone by. NANSPERSEC/nansused = bogomips. A 1 bogomips machine will take an extra second to do this (anything likely to be even able to run FreeBSD should exceed 1 BM - yes, ha ha), and a kBM CPU can do it in 1 ms. Perhaps in the future a prescaler might be required, but this whole thing is just really chrome anyway. Would anyone scream and projectile-vomit if I added this to identcpu.c? To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
On Thu, 2 Sep 1999, Oliver Fromme wrote: Nick Sayer wrote in list.freebsd-hackers: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. It's not a metric of CPU performance. It's just a meaningless number, and its relation to the actual performance of the machine is very questionable. It is not as meaningless as it seems. The computed value is used for some driver timing. It can also detect some CPU misconfigurations. Quite valuable thing. Milan Kopacka To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
We have this for 586+ class machines: CPU: Pentium/P54C (132.73-MHz 586-class CPU) Origin = "GenuineIntel" Id = 0x52c Stepping=12 Features=0x1bfFPU,VME,DE,PSE,TSC,MSR,MCE,CX8 Seems more precise and informative. For 386/486 based hardware someone could adapt one of the numerous CPU speed detection routines out there. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
On Thu, 2 Sep 1999, Julian Elischer wrote: There was such a thing in 386BSD and FreeBSD1.0 I certainly thing it was a worth-while thing. I'd try make the loop as similar to the Linux one so that they are comparable. My vote is to make the number printed in parity with the number printed by a Netware server. -- | Matthew N. Dodd | '78 Datsun 280Z | '75 Volvo 164E | FreeBSD/NetBSD | | [EMAIL PROTECTED] | 2 x '84 Volvo 245DL| ix86,sparc,pmax | | http://www.jurai.net/~winter | This Space For Rent | ISO8802.5 4ever | To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
According to Nick Sayer: Would anyone scream and projectile-vomit if I added this to identcpu.c? Yes. We are also FreeBSD users/developers because we don't follow the Linux way. Bogomips are [as it says] bogus and many people acknoledge this but far too often you see in some Linux list/newsgroup some dick sizing^W^Wbogomips comparisons. Here lies madness. -- Ollivier ROBERT -=- FreeBSD: The Power to Serve! -=- [EMAIL PROTECTED] FreeBSD keltia.freenix.fr 4.0-CURRENT #73: Sat Jul 31 15:36:05 CEST 1999 To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
"Matthew N. Dodd" wrote: On Thu, 2 Sep 1999, Julian Elischer wrote: There was such a thing in 386BSD and FreeBSD1.0 I certainly thing it was a worth-while thing. I'd try make the loop as similar to the Linux one so that they are comparable. My vote is to make the number printed in parity with the number printed by a Netware server. I think we should just hack SPECint and SPECfp into the system startup and be done with it. That'll make us look professional! For those who don't want to add 10 or 15 minutes to their system we'll add an rc.conf knob to turn it off. This has got to be the dumbest discussion I've ever seen on -hackers. -- "Where am I, and what am I doing in this handbasket?" Wes Peters Softweyr LLC http://softweyr.com/ [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
On Thu, 2 Sep 1999, Ollivier Robert wrote: According to Nick Sayer: Would anyone scream and projectile-vomit if I added this to identcpu.c? Yes. We are also FreeBSD users/developers because we don't follow the Linux way. Bogomips are [as it says] bogus and many people acknoledge this but far too often you see in some Linux list/newsgroup some dick sizing^W^Wbogomips comparisons. oh, please... what do you call a "world-stone"? :) Seriously, I do agree with you that it's a bad idea. "Just say no to bogomips" -Alfred Here lies madness. -- Ollivier ROBERT -=- FreeBSD: The Power to Serve! -=- [EMAIL PROTECTED] FreeBSD keltia.freenix.fr 4.0-CURRENT #73: Sat Jul 31 15:36:05 CEST 1999 To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
CPU: Pentium/P54C (132.73-MHz 586-class CPU) Origin = "GenuineIntel" Id = 0x52c Stepping=12 Features=0x1bfFPU,VME,DE,PSE,TSC,MSR,MCE,CX8 Seems more precise and informative. For 386/486 based hardware someone could adapt one of the numerous CPU speed detection routines out there. Indeed. In fact, if someone is truly motivated to go and find something to do, rather than adding a BogoMIPS counter why not instead figure out some way to add CPU speed detection to SMP machines? As anyone with an SMP box knows, that speed information is disabled for various reasons. - Jordan To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
In article local.mail.freebsd-hackers/19990902233418$[EMAIL PROTECTED] you write: CPU: Pentium/P54C (132.73-MHz 586-class CPU) Origin = "GenuineIntel" Id = 0x52c Stepping=12 Features=0x1bfFPU,VME,DE,PSE,TSC,MSR,MCE,CX8 Seems more precise and informative. For 386/486 based hardware someone could adapt one of the numerous CPU speed detection routines out there. Indeed. In fact, if someone is truly motivated to go and find something to do, rather than adding a BogoMIPS counter why not instead figure out some way to add CPU speed detection to SMP machines? As anyone with an SMP box knows, that speed information is disabled for various reasons. What I want is a simple new readable sysctl, something like: hw.clockrate: 132 I think that this would be useful both for development (how fast is that stupid machine down in the bunker?), and system admininstration (who needs a cpu upgrade this year?). Doing this for Pentium and better systems should be trivial. Doing it for 486 and lower would just add a timing loop. Doing it for SMP would be harder. hw.cpu0.clockrate: 233 hw.cpu1.clockrate: 233 Possibly? The implementer gets to pick a better name than these. -- Jonathan To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
Ollivier Robert wrote: According to Nick Sayer: Would anyone scream and projectile-vomit if I added this to identcpu.c? Yes. We are also FreeBSD users/developers because we don't follow the Linux way. Bogomips are [as it says] bogus and many people acknoledge this but far too often you see in some Linux list/newsgroup some dick sizing^W^Wbogomips comparisons. Of course, that's what RC5DES is for. -- Kris Kirby [EMAIL PROTECTED] --- TGIFreeBSD... 'Nuff said. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: CFD: bogomips CPU performance metric
Chris Costello wrote: On Thu, Sep 02, 1999, Robert Sexton wrote: I'd have to agree with the "Lets be more professional" crowd. How about as a LINT option? "If you need something so banal, you can turn it on yourself" No, since it would just be useless bloat in the source tree. I'm going to have to side with Chris on this one. I spent half of last night trying to cut and trim as much out of the kernel as I could so it would boot on a 386SX-20 with 3 MB of RAM. Needless bloat is just that. (And FWIW, I made it to 997K, with just wdc0, fdc0, sio(4), ppp(4), and MSDOS support. I just wish I had PCMCIA slots so I could BOOTP FreeBSD instead.) -- Kris Kirby [EMAIL PROTECTED] --- TGIFreeBSD... 'Nuff said. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
CFD: bogomips CPU performance metric
Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. We don't require doing any such thing, and so adding it would be purely cosmetic. However, I allege that cosmetic things aren't in and of themselves evil, so long as they don't break anything in the process. I would like to generate a number that will hopefully be reasonably compatible with the one Linux spits out. The best method I have come up with is to have a similar (the same?) count down loop in assembler. Have it count down from 1,000,000 and see how much nanotime() has gone by. NANSPERSEC/nansused = bogomips. A 1 bogomips machine will take an extra second to do this (anything likely to be even able to run FreeBSD should exceed 1 BM - yes, ha ha), and a kBM CPU can do it in 1 ms. Perhaps in the future a prescaler might be required, but this whole thing is just really chrome anyway. Would anyone scream and projectile-vomit if I added this to identcpu.c? smime.p7s Description: S/MIME Cryptographic Signature
Re: CFD: bogomips CPU performance metric
On Thu, Sep 02, 1999 at 10:40:30AM -0700, Nick Sayer wrote: Description of BogoMIPS deleted Would anyone scream and projectile-vomit if I added this to identcpu.c? I might. :-) Why exactly, except to keep up with the Linux kidz, do we need this? I recognize that this is solely a cosmetic change, but one of the things I hold over the heads of the Linux folks I deal with is the fact that FreeBSD is a professional quality operating system which doesn't need useless blinking lights like BogoMIPS. Chalk me up as one of the people who considers Linux works like that as a negative. Regards, --Keith Stevenson-- -- Keith Stevenson System Programmer - Data Center Services - University of Louisville k.steven...@louisville.edu PGP key fingerprint = 4B 29 A8 95 A8 82 EA A2 29 CE 68 DE FC EE B6 A0 To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
There was such a thing in 386BSD and FreeBSD1.0 I certainly thing it was a worth-while thing. I'd try make the loop as similar to the Linux one so that they are comparable. On Thu, 2 Sep 1999, Nick Sayer wrote: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. We don't require doing any such thing, and so adding it would be purely cosmetic. However, I allege that cosmetic things aren't in and of themselves evil, so long as they don't break anything in the process. I would like to generate a number that will hopefully be reasonably compatible with the one Linux spits out. The best method I have come up with is to have a similar (the same?) count down loop in assembler. Have it count down from 1,000,000 and see how much nanotime() has gone by. NANSPERSEC/nansused = bogomips. A 1 bogomips machine will take an extra second to do this (anything likely to be even able to run FreeBSD should exceed 1 BM - yes, ha ha), and a kBM CPU can do it in 1 ms. Perhaps in the future a prescaler might be required, but this whole thing is just really chrome anyway. Would anyone scream and projectile-vomit if I added this to identcpu.c? To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
Description of BogoMIPS deleted Would anyone scream and projectile-vomit if I added this to identcpu.c? I might. :-) Why exactly, except to keep up with the Linux kidz, do we need this? I recognize that this is solely a cosmetic change, but one of the things I hold over the heads of the Linux folks I deal with is the fact that FreeBSD is a professional quality operating system which doesn't need useless blinking lights like BogoMIPS. Chalk me up as one of the people who considers Linux works like that as a negative. I'm with you on this, Keith. I'd rather that we kept the professional FreeBSD look and feel. If we look too much like Linux, then people will just use Linux. -DG David Greenman Co-founder/Principal Architect, The FreeBSD Project - http://www.freebsd.org Creator of high-performance Internet servers - http://www.terasolutions.com Pave the road of life with opportunities. To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
On Thu, Sep 02, 1999 at 10:40:30AM -0700, Nick Sayer wrote: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. We don't require doing any such thing, and so adding it would be purely cosmetic. However, I allege that cosmetic things aren't in and of themselves evil, so long as they don't break anything in the process. I'd have to agree with the Lets be more professional crowd. How about as a LINT option? If you need something so banal, you can turn it on yourself -- Robert Sexton, rob...@kudra.com Imagine if every Thursday your shoes exploded if you tied them the usual way. This happens to us all the time with computers, and nobody thinks of complaining. -- Jeff Raskin, interviewed in Doctor Dobb's Journal To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
Nick Sayer wrote in list.freebsd-hackers: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. It's not a metric of CPU performance. It's just a meaningless number, and its relation to the actual performance of the machine is very questionable. We don't require doing any such thing, and so adding it would be purely cosmetic. However, I allege that cosmetic things aren't in and of themselves evil, so long as they don't break anything in the process. It would break user's confidence in the seriousness of the system. Would anyone scream and projectile-vomit if I added this to identcpu.c? I would, FWIW. Regards Oliver -- Oliver Fromme, Leibnizstr. 18/61, 38678 Clausthal, Germany (Info: finger userinfo:o...@dorifer.heim3.tu-clausthal.de) In jedem Stück Kohle wartet ein Diamant auf seine Geburt (Terry Pratchett) To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
As Nick Sayer wrote ... so long as they don't break anything in the process. I would like to generate a number that will hopefully be reasonably compatible with the one Linux spits out. The best method I have come up with is to have a similar (the same?) count down loop in assembler. Have it count down from 1,000,000 and see how much nanotime() has gone by. NANSPERSEC/nansused = bogomips. A 1 bogomips machine will take an extra second to do this (anything likely to be even able to run FreeBSD should exceed 1 BM - yes, ha ha), and a kBM CPU can do it in 1 ms. Perhaps in the future a prescaler might be required, but this whole thing is just really chrome anyway. Would anyone scream and projectile-vomit if I added this to identcpu.c? I really don't see any merit in adding Linux-isms like this. Sounds a bit like the 'hack of the day' to me. I feel FreeBSD is well respected for it's stability etc. Not for it's 'me too' Linux-isms without any practical use. Mind you, I don't say Linux does not have stuff that is useful for inclusion in FreeBSD. I just stay BogoMips is what it calls itself: bogus, and should be kept from the FreeBSD kernel. Just my Dfl 0.02 W/ -- | / o / / _ Arnhem, The Netherlands- Powered by FreeBSD - |/|/ / / /( (_) BulteWWW : http://www.tcja.nl http://www.freebsd.org To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
There was such a thing in 386BSD and FreeBSD1.0 I remember no such thing doing a 'bogomips' to compare against Linux. Certainly not in 386BSD. Nate I certainly thing it was a worth-while thing. I'd try make the loop as similar to the Linux one so that they are comparable. On Thu, 2 Sep 1999, Nick Sayer wrote: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. We don't require doing any such thing, and so adding it would be purely cosmetic. However, I allege that cosmetic things aren't in and of themselves evil, so long as they don't break anything in the process. I would like to generate a number that will hopefully be reasonably compatible with the one Linux spits out. The best method I have come up with is to have a similar (the same?) count down loop in assembler. Have it count down from 1,000,000 and see how much nanotime() has gone by. NANSPERSEC/nansused = bogomips. A 1 bogomips machine will take an extra second to do this (anything likely to be even able to run FreeBSD should exceed 1 BM - yes, ha ha), and a kBM CPU can do it in 1 ms. Perhaps in the future a prescaler might be required, but this whole thing is just really chrome anyway. Would anyone scream and projectile-vomit if I added this to identcpu.c? To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
On Thu, Sep 02, 1999, Robert Sexton wrote: I'd have to agree with the Lets be more professional crowd. How about as a LINT option? If you need something so banal, you can turn it on yourself No, since it would just be useless bloat in the source tree. -- |Chris Costello ch...@calldei.com |Supercomputer: Turns a CPU-bound problem into |an I/O-bound problem. - Ken Batcher `--- To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
On Thu, Sep 02, 1999, Nick Sayer wrote: Would anyone scream and projectile-vomit if I added this to identcpu.c? Yes, I would. The way I interpret it, along with useless blinking light, is as follows: BogoMIPS is but the combination of Bogus and an acronym for Meaningless Indicator of Processor Speed. :) -- |Chris Costello ch...@calldei.com |You might have mail. `-- To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
Chris Costello wrote: No, since it would just be useless bloat in the source tree. If we must have it, how about a port? - I'm definitely for the this isn't a good idea crowd, When I was using Linux, I thought it was 'cute'... I've grown up a bit since then... -Kp To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
It was there... when I added the code to calibrate the delay loops originally and added the DELAY macro, it printed out the callibration factor.. (DELAY was originally a spin loop) It wasn't called 'BOGOMIPS...' On Thu, 2 Sep 1999, Nate Williams wrote: There was such a thing in 386BSD and FreeBSD1.0 I remember no such thing doing a 'bogomips' to compare against Linux. Certainly not in 386BSD. Nate I certainly thing it was a worth-while thing. I'd try make the loop as similar to the Linux one so that they are comparable. On Thu, 2 Sep 1999, Nick Sayer wrote: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. We don't require doing any such thing, and so adding it would be purely cosmetic. However, I allege that cosmetic things aren't in and of themselves evil, so long as they don't break anything in the process. I would like to generate a number that will hopefully be reasonably compatible with the one Linux spits out. The best method I have come up with is to have a similar (the same?) count down loop in assembler. Have it count down from 1,000,000 and see how much nanotime() has gone by. NANSPERSEC/nansused = bogomips. A 1 bogomips machine will take an extra second to do this (anything likely to be even able to run FreeBSD should exceed 1 BM - yes, ha ha), and a kBM CPU can do it in 1 ms. Perhaps in the future a prescaler might be required, but this whole thing is just really chrome anyway. Would anyone scream and projectile-vomit if I added this to identcpu.c? To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
On Thu, Sep 02, 1999, Karl Pielorz wrote: Chris Costello wrote: No, since it would just be useless bloat in the source tree. If we must have it, how about a port? - I'm definitely for the this isn't a good idea crowd, When I was using Linux, I thought it was 'cute'... I've grown up a bit since then... I don't think the local Ports Wraith would be amazingly happy with a port which modifies identcpu.c, which is part of the kernel. -Kp -- |Chris Costello ch...@calldei.com |Base 8 is just like base 10, if you are missing two fingers. - Tom Lehrer `-- To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
If we must have it, how about a port? - I'm definitely for the this isn't a good idea crowd, When I was using Linux, I thought it was 'cute'... I've grown up a bit since then... Create /usr/ports/useless_linux_utils Add this and code for making the keyboard lights blink in time to whatever audio CD is playing ;-) I agree with what seems to be the consenus: not only are BogoMIPS useless in themselves, but it does detract from the 'professional' feel of FreeBSD. -marc To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
On Thu, 2 Sep 1999, Oliver Fromme wrote: Nick Sayer wrote in list.freebsd-hackers: Linux generates a meric of CPU performance as a byproduct of calibrating a delay loop. It's not a metric of CPU performance. It's just a meaningless number, and its relation to the actual performance of the machine is very questionable. It is not as meaningless as it seems. The computed value is used for some driver timing. It can also detect some CPU misconfigurations. Quite valuable thing. Milan Kopacka To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
We have this for 586+ class machines: CPU: Pentium/P54C (132.73-MHz 586-class CPU) Origin = GenuineIntel Id = 0x52c Stepping=12 Features=0x1bfFPU,VME,DE,PSE,TSC,MSR,MCE,CX8 Seems more precise and informative. For 386/486 based hardware someone could adapt one of the numerous CPU speed detection routines out there. To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
On Thu, 2 Sep 1999, Julian Elischer wrote: There was such a thing in 386BSD and FreeBSD1.0 I certainly thing it was a worth-while thing. I'd try make the loop as similar to the Linux one so that they are comparable. My vote is to make the number printed in parity with the number printed by a Netware server. -- | Matthew N. Dodd | '78 Datsun 280Z | '75 Volvo 164E | FreeBSD/NetBSD | | win...@jurai.net | 2 x '84 Volvo 245DL| ix86,sparc,pmax | | http://www.jurai.net/~winter | This Space For Rent | ISO8802.5 4ever | To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
According to Nick Sayer: Would anyone scream and projectile-vomit if I added this to identcpu.c? Yes. We are also FreeBSD users/developers because we don't follow the Linux way. Bogomips are [as it says] bogus and many people acknoledge this but far too often you see in some Linux list/newsgroup some dick sizing^W^Wbogomips comparisons. Here lies madness. -- Ollivier ROBERT -=- FreeBSD: The Power to Serve! -=- robe...@keltia.freenix.fr FreeBSD keltia.freenix.fr 4.0-CURRENT #73: Sat Jul 31 15:36:05 CEST 1999 To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
Matthew N. Dodd wrote: On Thu, 2 Sep 1999, Julian Elischer wrote: There was such a thing in 386BSD and FreeBSD1.0 I certainly thing it was a worth-while thing. I'd try make the loop as similar to the Linux one so that they are comparable. My vote is to make the number printed in parity with the number printed by a Netware server. I think we should just hack SPECint and SPECfp into the system startup and be done with it. That'll make us look professional! For those who don't want to add 10 or 15 minutes to their system we'll add an rc.conf knob to turn it off. This has got to be the dumbest discussion I've ever seen on -hackers. -- Where am I, and what am I doing in this handbasket? Wes Peters Softweyr LLC http://softweyr.com/ w...@softweyr.com To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
On Thu, 2 Sep 1999, Ollivier Robert wrote: According to Nick Sayer: Would anyone scream and projectile-vomit if I added this to identcpu.c? Yes. We are also FreeBSD users/developers because we don't follow the Linux way. Bogomips are [as it says] bogus and many people acknoledge this but far too often you see in some Linux list/newsgroup some dick sizing^W^Wbogomips comparisons. oh, please... what do you call a world-stone? :) Seriously, I do agree with you that it's a bad idea. Just say no to bogomips -Alfred Here lies madness. -- Ollivier ROBERT -=- FreeBSD: The Power to Serve! -=- robe...@keltia.freenix.fr FreeBSD keltia.freenix.fr 4.0-CURRENT #73: Sat Jul 31 15:36:05 CEST 1999 To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
CPU: Pentium/P54C (132.73-MHz 586-class CPU) Origin = GenuineIntel Id = 0x52c Stepping=12 Features=0x1bfFPU,VME,DE,PSE,TSC,MSR,MCE,CX8 Seems more precise and informative. For 386/486 based hardware someone could adapt one of the numerous CPU speed detection routines out there. Indeed. In fact, if someone is truly motivated to go and find something to do, rather than adding a BogoMIPS counter why not instead figure out some way to add CPU speed detection to SMP machines? As anyone with an SMP box knows, that speed information is disabled for various reasons. - Jordan To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
In article local.mail.freebsd-hackers/19990902233418$5...@fish.pcs you write: CPU: Pentium/P54C (132.73-MHz 586-class CPU) Origin = GenuineIntel Id = 0x52c Stepping=12 Features=0x1bfFPU,VME,DE,PSE,TSC,MSR,MCE,CX8 Seems more precise and informative. For 386/486 based hardware someone could adapt one of the numerous CPU speed detection routines out there. Indeed. In fact, if someone is truly motivated to go and find something to do, rather than adding a BogoMIPS counter why not instead figure out some way to add CPU speed detection to SMP machines? As anyone with an SMP box knows, that speed information is disabled for various reasons. What I want is a simple new readable sysctl, something like: hw.clockrate: 132 I think that this would be useful both for development (how fast is that stupid machine down in the bunker?), and system admininstration (who needs a cpu upgrade this year?). Doing this for Pentium and better systems should be trivial. Doing it for 486 and lower would just add a timing loop. Doing it for SMP would be harder. hw.cpu0.clockrate: 233 hw.cpu1.clockrate: 233 Possibly? The implementer gets to pick a better name than these. -- Jonathan To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
Ollivier Robert wrote: According to Nick Sayer: Would anyone scream and projectile-vomit if I added this to identcpu.c? Yes. We are also FreeBSD users/developers because we don't follow the Linux way. Bogomips are [as it says] bogus and many people acknoledge this but far too often you see in some Linux list/newsgroup some dick sizing^W^Wbogomips comparisons. Of course, that's what RC5DES is for. -- Kris Kirby k...@airnet.net --- TGIFreeBSD... 'Nuff said. To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message
Re: CFD: bogomips CPU performance metric
Chris Costello wrote: On Thu, Sep 02, 1999, Robert Sexton wrote: I'd have to agree with the Lets be more professional crowd. How about as a LINT option? If you need something so banal, you can turn it on yourself No, since it would just be useless bloat in the source tree. I'm going to have to side with Chris on this one. I spent half of last night trying to cut and trim as much out of the kernel as I could so it would boot on a 386SX-20 with 3 MB of RAM. Needless bloat is just that. (And FWIW, I made it to 997K, with just wdc0, fdc0, sio(4), ppp(4), and MSDOS support. I just wish I had PCMCIA slots so I could BOOTP FreeBSD instead.) -- Kris Kirby k...@airnet.net --- TGIFreeBSD... 'Nuff said. To Unsubscribe: send mail to majord...@freebsd.org with unsubscribe freebsd-hackers in the body of the message