Re: Bogomips - Centrino
On Wed, Feb 09, 2005 at 03:10:25PM +0100, Christian Kerbetz wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hallo, beim anzeigen der Prozessorinformationen auf meinem Notebook wird jedesmal eine andere bogomips-zahl ausgegeben! kann mir evtl. jemand sagen warum das so ist? http://ftp.gwdg.de/LDP/HOWTO/BogoMips/x78.html Quoted from the Internet, origin unknown but brought to the attention by Eric S Raymond [EMAIL PROTECTED], and Geoff Mackenzie [EMAIL PROTECTED], there is an humourously illustrative definition of BogoMips as ''the number of million times per second a processor can do absolutely nothing.'' Ich gehe mal davon aus das sich durch automatisches speed stepping oder so sich die staendig wechselnden1 Werte ergeben. (Ist nur geraten - kann auch was voellig anderes sein!) Sven -- If God passed a mic to me to speak I'd say stay in bed, world Sleep in peace [The Cardigans - No sleep] -- Haeufig gestellte Fragen und Antworten (FAQ): http://www.de.debian.org/debian-user-german-FAQ/ Zum AUSTRAGEN schicken Sie eine Mail an [EMAIL PROTECTED] mit dem Subject unsubscribe. Probleme? Mail an [EMAIL PROTECTED] (engl)
Re: Bogomips - Centrino
ja hallo erstmal,... Am Mittwoch, 9. Februar 2005 17:41 schrieb Sven Hoexter: On Wed, Feb 09, 2005 at 03:10:25PM +0100, Christian Kerbetz wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hallo, beim anzeigen der Prozessorinformationen auf meinem Notebook wird jedesmal eine andere bogomips-zahl ausgegeben! kann mir evtl. jemand sagen warum das so ist? http://ftp.gwdg.de/LDP/HOWTO/BogoMips/x78.html Quoted from the Internet, origin unknown but brought to the attention by Eric S Raymond [EMAIL PROTECTED], and Geoff Mackenzie [EMAIL PROTECTED], there is an humourously illustrative definition of BogoMips as ''the number of million times per second a processor can do absolutely nothing.'' Wo ist denn dann der Unterschied zur Taktrate? 1 GHZ heißt doch eigentlich auch 1 Milliarde ASM-Nopes pro Sek. Keep smiling yanosz -- Achtung: Die E-Mail-Adresse [EMAIL PROTECTED] wird in Kürze deaktiviert werden. Bitte nutzen Sie die Adresse [EMAIL PROTECTED]
Re: Bogomips - Centrino
Jan Lühr schrieb: ja hallo erstmal,... dito, Wo ist denn dann der Unterschied zur Taktrate? 1 GHZ heißt doch eigentlich auch 1 Milliarde ASM-Nopes pro Sek. Auf [1] findest du folgendes: BogoMips was named BogoMips for a reason; it's totally bogus as a benchmark. It's nothing more than a timer loop -- the kernel runs a null loop which does nothing but count down to 0. It calibrates this loop against the real time clock to see how fast it executes. From that point onwards the calibrated loop is used as a way for the kernel to do timing delay loops; for example a device driver might want to delay for 10 microseconds while hardware does something. The reported BogoMips figure is just how many times the delay loop can be executed in one second. So it really tells you nothing about how fast the processor is for useful work. Keep smiling yanosz MfG Jan [1] http://www.obsolyte.com/bogomips.html -- OpenPGP Key-Fingerprint: 0E9B 4052 C661 5018 93C3 4E46 651A 7A28 4028 FF7A pgpTNzBtZ51Ng.pgp Description: PGP signature
Re: bogomips ?
Egor Tur said: Hi. A little question: programme `linux_logo' show 1264.84 Bogomips Total on my system, but `bogomips' - 634.00 BogoMips. Why do these values be different? Thanx. maybe they use different methods to get the result. maybe there was some program running in the background that skewed results, in any case you can get your bogomips from /proc/cpuinfo note this really is a bogus(hence the bogo) measure of performance I think it's only done to configure some sort of timing mechanism in the kernel during boot(not sure though). nate -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: bogomips ?
On Thu, Feb 20, 2003 at 08:05:24AM +0200, Egor Tur wrote: Hi. A little question: programme `linux_logo' show 1264.84 Bogomips Total on my system, but `bogomips' - 634.00 BogoMips. Why do these values be different? Possibly you have a dual-CPU machine since 634*2 ~= 1264. Cat /proc/cpuinfo for the actual value(s) and add up each bogomips line. Regards, Anand -- `` We are shaped by our thoughts, we become what we think. When the mind is pure, joy follows like a shadow that never leaves. '' -- Buddha, The Dhammapada -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: BogoMIPS ?
[EMAIL PROTECTED] wrote: could anyone please tell me what BogoMIPS at bootup means? See the BogoMips mini-HOWTO, which you should be able to find in /usr/doc/HOWTO or /usr/share/doc/HOWTO depending on your version of Debian. `MIPS is short for Millions of Instructions Per Second. It is a measure for the computation speed of a program. Like most such measures, it is more often abused than used prop erly (it is very difficult to justly compare MIPS for dif ferent kinds of computers). BogoMips are Linus's invention. The kernel (or was it a device driver?) needs a timing loop (the time is too short and/or needs to be too exact for a non-busy-loop method of waiting), which must be calibrated to the processor speed of the machine. Hence, the kernel measures at boot time how fast a certain kind of busy loop runs on a computer. Bogo comes from bogus, i.e, something which is a fake. Hence, the BogoMips value gives some indication of the processor speed, but it is way too unscientific to be called anything but BogoMips. The reasons (there are two) it is printed during bootup is that a) it is slightly useful for debugging and for checking that the computers caches and turbo button work, and b) Linus loves to chuckle when he sees confused people on the news.' -- Colin Watson [EMAIL PROTECTED]
Re: BogoMIPS ?
http://metalab.unc.edu/pub/Linux/docs/HOWTO/mini/BogoMips Regards, Onno At 12:59 PM 2/19/00 +0100, [EMAIL PROTECTED] wrote: Hi there, could anyone please tell me what BogoMIPS at bootup means? Thanks in advance, Uwe -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: bogomips
Alberto Maurizi wrote: Is this normal? I mean, that so low bogomips value. model name : Pentium 60/66 cpu MHz : 59.999660 bogomips: 23.91 I've got an Intel P133. Approximating pro rata using your data gives 23.91 / 60 * 133 = 53. Actual reported value is 53.25 So I'd say it looks exactly right. Regards, Paul
Re: bogomips
Alberto Maurizi wrote: Is this normal? I mean, that so low bogomips value. model name : Pentium 60/66 cpu MHz : 59.999660 bogomips: 23.91 I've got an Intel P133. Approximating pro rata using your data gives 23.91 / 60 * 133 = 53. Actual reported value is 53.25 So I'd say it looks exactly right. Regards, Paul So mine doesn't make much sence then does it??? this is my /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 5 model : 1 model name : AMD-K5(tm) Processor stepping: 1 cpu MHz : 100.230957 fdiv_bug: no hlt_bug : no sep_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr mce cx8 pge bogomips: 199.88 -- Tim Nicholas [EMAIL PROTECTED] Let the frantic Goddess and delerious drunk cry together in shadow for the puppy's sad stare, the forest and the death of the moon.
Re: bogomips
On Thu, Dec 16, 1999 at 10:59:18AM +1300, Tim Nicholas wrote: Alberto Maurizi wrote: Is this normal? I mean, that so low bogomips value. model name : Pentium 60/66 cpu MHz : 59.999660 bogomips: 23.91 I've got an Intel P133. Approximating pro rata using your data gives 23.91 / 60 * 133 = 53. Actual reported value is 53.25 So I'd say it looks exactly right. So mine doesn't make much sence then does it??? this is my /proc/cpuinfo vendor_id : AuthenticAMD model name: AMD-K5(tm) Processor cpu MHz : 100.230957 bogomips : 199.88 It makes perfect sense, for an AMD-K5. Bogomips are as their name implies, bogus. Bogomips only make sense with respect to one particular CPU family. There is a bogomips explanation somewhere out there, in a FAQ somewhere else, too, no doubt. -- William Burrow -- New Brunswick, Canada How the Internet explodes myths... - the GOOD TIMES email virus hoax, brought to life courtesy Microsoft - MAKE MONEY FAST brought to reality on Wall Street by dot-coms and Linux
Re: bogomips
-BEGIN PGP SIGNED MESSAGE- On Thu, 16 Dec 1999, Tim Nicholas wrote: Alberto Maurizi wrote: Is this normal? I mean, that so low bogomips value. model name : Pentium 60/66 cpu MHz : 59.999660 bogomips: 23.91 I've got an Intel P133. Approximating pro rata using your data gives 23.91 / 60 * 133 = 53. Actual reported value is 53.25 So mine doesn't make much sence then does it??? vendor_id : AuthenticAMD model name: AMD-K5(tm) Processor cpu MHz : 100.230957 bogomips : 199.88 There is a BogoMips mini-HOWTO in /usr/doc/HOWTO ... Martin - -- Where do you want to go today? - As far from Redmond as possible! For public PGP-key: finger [EMAIL PROTECTED] -BEGIN PGP SIGNATURE- Version: 2.6.3ia Charset: noconv iQCVAwUBOFgfX7CGSMW7I2etAQHnDwQAgCklbt6Tbut8V/XFtI/eBWbkIMoO7eBS 27eafJUDqFYnQNiGOVNFNFXgSBsGph2q5OqDjH6CpjS1NMaM/OhIuJmEKUSFjFg1 hkcbUkga+/kv5CtfrXgq16XTFx9WSBdRSI4qCSHqKiqY1FcFP+kg29N6f8PcAJPe IbG6G+nvIN0= =Db2x -END PGP SIGNATURE-
Re: Bogomips disparity
I remember this being discussed a month or so ago. You might take a look at the User's archives at, http://www.debian.org/Lists-Archives/ Good luck, Kent Pann McCuaig wrote: I'll be looking for some docs to explain this, but in the meantime perhaps someone knows off the top . . . Just screwed together a couple of new boxes. AMD K6/2-350 CPUs. The kernel on the slink rescue disk (2.1.8) as well as the kernel installed as slink:kernel-image-2.0.36 both report ~350 bogomips. The kernel on Tom's Root Boot Disk reports ~700 bogomips. We're talking 2.0.36 in all cases. Any ideas? -- your man pann -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: bogomips
Subject: Re: bogomips Date: Thu, Nov 12, 1998 at 12:56:21AM +0100 In reply to:Martin Bialasinski Quoting Martin Bialasinski([EMAIL PROTECTED]): w == wtopa [EMAIL PROTECTED] writes: w Yes that would seem about right. The problem I have is that my w 166MMX gives me 149.9 BogoMips on my Slackware partition and only w 130.6 on my Debian partition. Same hardware (same box). Same kernelversion? The bogomips implementation changes sometimes. Yes. 2.0.34/35 on Slackware 149.9. On debian 2.0.34 120 Bogomips and on 2.0.35 130.6. The .config's on both dists are the same as well. Interesting, isn't it. NOT! I am confused by the change in Debian kernels. On Slackware I have gone from 2.0.30 - 35 with no change at all, allways 149.9. I did note that on Slackware 3.5 (just installed) on a scsi drive on sba1, kernel 2.1.127 gets 149.5. This is the first run on that kernel and I have a bunch of new stuff compiled into the kernel, so that _may_ account for the change. Would appreciate it if anyone could give me some reason for the difference in the Debian kernel. Wayne Ciao, Martin -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- Real Programmers think better when playing Adventure or Rogue. ___ Wayne T. Topa [EMAIL PROTECTED]
Re: bogomips
- w Yes that would seem about right. The problem I have is that my - w 166MMX gives me 149.9 BogoMips on my Slackware partition and only - w 130.6 on my Debian partition. Same hardware (same box). - - Same kernelversion? The bogomips implementation changes sometimes. - - Yes. 2.0.34/35 on Slackware 149.9. On debian 2.0.34 120 Bogomips and - on 2.0.35 130.6. The .config's on both dists are the same as well. - Interesting, isn't it. NOT! I am confused by the change in Debian - kernels. On Slackware I have gone from 2.0.30 - 35 with no change at - all, allways 149.9. I did note that on Slackware 3.5 (just installed) - on a scsi drive on sba1, kernel 2.1.127 gets 149.5. This is the first - run on that kernel and I have a bunch of new stuff compiled into the - kernel, so that _may_ account for the change. get kernel source and recompile it; do kernels say other number by booting or by runnning bogomips command ? does someone apply any patches to kernels used by debian ? -- Matus fantomas Uhlar, sysadmin at NETLAB+ Kosice, Slovakia BIC coord for *.sk; admin of netlab.irc.sk; co-admin of irc.felk.cvut.cz
Re: bogomips
Subject: Re: bogomips Date: Fri, Nov 13, 1998 at 07:27:12AM -0600 I remember back in '98 when [EMAIL PROTECTED] wrote: Yes. 2.0.34/35 on Slackware 149.9. On debian 2.0.34 120 Bogomips and on 2.0.35 130.6. The .config's on both dists are the same as well. Interesting, isn't it. NOT! ... Would appreciate it if anyone could give me some reason for the difference in the Debian kernel. As someone mentioned, it could be that there are some patches in the Debian kernel, not in the slackware. Realize that Bogomips are really sensative to alignment in memory, if the loop is in the wrong place it could change the numbers. Also you could be using different versions of the C compiler on the different machines. try typing gcc --version on both. gcc 2.7.2.x is the official compiler for the linux kernel, at least in Linus' mind. There have been on-and-off report of breakage with gcc 2.8.x and (especially) egcc. Go look on linux-kernel for anything by Linux with egcc in the subject line for more info.. Matt Matt may have found the link. I just went to the 3 dists and got the following: Debian slink %gcc --version gcc 2.7.2.3 Slackware 3.3 %gcc --version gcc-2.7.2.2 Slackware 3.5 %gcc --version egcs-2.9.29 980515 egcs 1.0.3 NOTE: Debian uses glibc while Slackware is still using libc. So I got the source for gcc-2.7.2.3 and compiled it for the Slack3.3 dist. (Yea, I know, I have too much spare time!) I installed the new version (%gcc --version gcc-2.7.2.3 ), and recompilied linux-2.0.35. and rebooted. Slackware _again_ reports BogoMips at 149.91, now with kernel compiled with gcc-2.7.2.3. So that doesn't look like it is/was the cause. Now I will take the kernel source from ftp.kernel.org for 2.0.35 and compile it on Debian with make dep,clean and zImage, make a boot floppy and see if that changes anything. No, it didn't. Debian still says that BogoMips = 130.66. So that leads me to believe the kernel-source.deb files are ok, so is gcc-2.7.2.3. Leaves glibc. That I will not change! OK, someone said that the BogoMips are meaningless. It just might be that he is correct. I have just noticed something I hadn't found before. I have saytime running in cron and it reports the time on the hour. On Slackware (149.91 Bogomips) the time is the hour and 1 sec, on Debian (130.66 BogoMips) the time is the hour exactly! Well I had fun tracking this down but, to me anyway, BogoMips is a nice number but I won't lose any sleep over differences between distributions. Hope this answers my question? Or does it??? Wayne -- /* Matt Sayler -- [EMAIL PROTECTED] -- atwork?astronomy:cs http://www.cs.utexas.edu/users/mpsayler -- (512)471-7450 Have you ever imagined a world with no hypothetical situations? */ -- If you put garbage in a computer nothing comes out but garbage. But this garbage, having passed through a very expensive machine, is somehow enobled and none dare criticize it. ___ Wayne T. Topa [EMAIL PROTECTED]
Re: bogomips
w == wtopa [EMAIL PROTECTED] writes: w Yes that would seem about right. The problem I have is that my w 166MMX gives me 149.9 BogoMips on my Slackware partition and only w 130.6 on my Debian partition. Same hardware (same box). Same kernelversion? The bogomips implementation changes sometimes. Ciao, Martin
Re: bogomips
On Tue, 10 Nov 1998, Oz Dror wrote: Hi I have a pentium-II 400Mhz and a pentium MMX 200Mhz both have 400.59 bogomips Why? SWAG (Silly Wild-Assed Guess): The memory is the same speed? Since Intel guesses wrong in the branch prediction, it must refetch the instruction residing at the address of the result of the branch from cache. The bottleneck is not CPU speed, it is the cache RAM. In other words, the CPU is sees a branch instruction coming and assumes that the resulting target of the branch will be ahead ... it does not expect that someone is going to be running a tight loop. It never even looks for that since it is so rare in real life. As a result, the next instruction is not in its prefetch store and it has to go out to external cache to get it ... again and again and again. The memory is not the same. the P-II has 256M dimm (8nsec), the P-I MMX has 128MB EDO (60nsec) -Oz -- NAME Oz Dror, Los Angeles, California EMAIL [EMAIL PROTECTED] Linux since 8/15/94 PHONE Fax (310) 474-3126 -BEGIN PGP PUBLIC KEY BLOCK- Version: 2.6.2 mQBtAzA/tLQAAAEDAKUy/TEjQ/jiZ+9/WJb/+NHxqkvOxGZ3W/F2JCNm5v5ZTZz+ BVZC9GM/I+plQ8xz+7B+KhDSVax8gxNTAkJ+I7P/zAP2ZDMwVf4lq5ZFxMJC+7c7 ET+hNtmQUt8vCVR8hQAFEbQZT3ogRHJvciA8ZHJvckBuZXRjb20uY29tPg== =EU23 -END PGP PUBLIC KEY BLOCK-
Re: bogomips
On Tue, 10 Nov 1998, Oz Dror wrote: Hi I have a pentium-II 400Mhz and a pentium MMX 200Mhz both have 400.59 bogomips Why? Because bogomips are bogus and meaningless? Andrew Tarr If you're not part of the solution, you're part of the precipitate |___ http://multinet.co.nz/personalhomepages/locusmeus/antechamber.html |~~~
Re: bogomips
On Tue, Nov 10, 1998 at 10:24:46PM -0800, Oz Dror wrote: Hi I have a pentium-II 400Mhz and a pentium MMX 200Mhz both have 400.59 bogomips Why? MMX doubles your bogomips. MMX CPUs are evidently great at running empty loops. miket
Re: bogomips
On Tue, Nov 10, 1998 at 10:24:46PM -0800, Oz Dror wrote: Hi I have a pentium-II 400Mhz and a pentium MMX 200Mhz both have 400.59 bogomips Why? MMX doubles your bogomips. MMX CPUs are evidently great at running empty loops. Every Pentium II has MMX. Anyway, for 166 and 300 MHz PII's I have seen a bogomips value that approximately equals the MHz number. So that seems OK. HTH, Eric -- E.L. Meijer ([EMAIL PROTECTED]) | tel. office +31 40 2472189 Eindhoven Univ. of Technology | tel. lab. +31 40 2475032 Lab. for Catalysis and Inorg. Chem. (TAK) | tel. fax+31 40 2455054
Re: bogomips
i'm running a 2.0.33 kernel from the installation disks and it gives me 400.59 Bogomips for a pentium 200MMX running on a compaq laptop. I recompiled the kernel and it gave me about 208 bogomips after the new kernel was recompiled, and the bogomips mystery continued. btw, i did read the bogomips mini howto. just my lil contribution jd? On Wed, 11 Nov 1998, E.L. Meijer (Eric) wrote: On Tue, Nov 10, 1998 at 10:24:46PM -0800, Oz Dror wrote: Hi I have a pentium-II 400Mhz and a pentium MMX 200Mhz both have 400.59 bogomips Why? MMX doubles your bogomips. MMX CPUs are evidently great at running empty loops. Every Pentium II has MMX. Anyway, for 166 and 300 MHz PII's I have seen a bogomips value that approximately equals the MHz number. So that seems OK. HTH, Eric -- E.L. Meijer ([EMAIL PROTECTED]) | tel. office +31 40 2472189 Eindhoven Univ. of Technology | tel. lab. +31 40 2475032 Lab. for Catalysis and Inorg. Chem. (TAK) | tel. fax+31 40 2455054 -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: bogomips
Subject: Re: bogomips Date: Wed, Nov 11, 1998 at 01:19:22PM +0100 In reply to:[EMAIL PROTECTED] Quoting [EMAIL PROTECTED]([EMAIL PROTECTED]): On Tue, Nov 10, 1998 at 10:24:46PM -0800, Oz Dror wrote: Hi I have a pentium-II 400Mhz and a pentium MMX 200Mhz both have 400.59 bogomips Why? MMX doubles your bogomips. MMX CPUs are evidently great at running empty loops. Every Pentium II has MMX. Anyway, for 166 and 300 MHz PII's I have seen a bogomips value that approximately equals the MHz number. So that seems OK. Yes that would seem about right. The problem I have is that my 166MMX gives me 149.9 BogoMips on my Slackware partition and only 130.6 on my Debian partition. Same hardware (same box). The numbers may be meaningless but I wonder why my Debian kernel runs slower! Slackware kernel made with the ole standby, make dep, clean, zImage. Debian with make-kpkg. I have tried different kernels on Slackware and the 149.9 stays the same. For different Debian kernels the numbers vary from 120 -130.6. Courious about why this should be. HTH, Eric -- E.L. Meijer ([EMAIL PROTECTED]) | tel. office +31 40 2472189 Eindhoven Univ. of Technology | tel. lab. +31 40 2475032 Lab. for Catalysis and Inorg. Chem. (TAK) | tel. fax+31 40 2455054 -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- Software, n.: Formal evening attire for female computer analysts. ___ Wayne T. Topa [EMAIL PROTECTED]