system clock speed too high?

DISCUSSION Andreas Haumer 8 participants 22 messages Jun 2, 2003 – Jun 4, 2003

lore

Share & tools

lore

Andreas Haumer

author

2003-06-02 19:27 UTC #0 lore raw

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi!

(I've already reported this previously as post scriptum to a different bugreport, so it might have slipped through unnoticed...)

I have a quite strange phenomenon here: I see a ~2.5 times speed up of system time on a Asus AP1700-S5 server with Linux-2.4.21-rc6-ac1.
Simple proof: a "sleep 300" command terminates after exactly 120 seconds of wall clock time.

Or just look at this:
root@setup:~ {544} $ ntpdate ntp.xss.co.at; ntpdate ntp.xss.co.at; sleep 300; ntpdate ntp.xss.co.at
2 Jun 21:24:45 ntpdate[3768]: step time server 194.152.162.17 offset -38.679740 sec 2 Jun 21:24:45 ntpdate[3769]: adjust time server 194.152.162.17 offset -0.069998 sec 2 Jun 21:26:45 ntpdate[3786]: step time server 194.152.162.17 offset -180.053781 sec

Impressive, isn't it?

This "acceleration" of time shows with other time measures, like "date", system uptime, process time, etc., too.

(HZ is still defined as 100, I haven't changed anything in the kernel sources)

System info:

Asus AP1700-S5, Asus PR-DLS533 Motherboard, Intel Xeon 2.4GHz CPU, 512MB RAM

root@setup:~ {545} $ uname -a
Linux setup 2.4.21-rc6-ac1 #1 SMP Sat May 31 20:04:18 CEST 2003 i686 unknown

root@setup:~ {546} $ lspci -v
00:00.0 Host bridge: ServerWorks CNB20-HE Host Bridge (rev 31)
Flags: fast devsel

00:00.1 Host bridge: ServerWorks CNB20-HE Host Bridge
Flags: fast devsel

00:00.2 Host bridge: ServerWorks CNB20-HE Host Bridge
Flags: fast devsel

00:02.0 Ethernet controller: Intel Corp. 82540EM Gigabit Ethernet Controller (rev 02)
Subsystem: Intel Corp. 82540EM Gigabit Ethernet Controller
Flags: bus master, 66Mhz, medium devsel, latency 32, IRQ 18 Memory at fd800000 (32-bit, non-prefetchable) [size=128K]
I/O ports at d800 [size=64]
Capabilities: [dc] Power Management version 2
Capabilities: [e4] PCI-X non-bridge device.
Capabilities: [f0] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable-

00:03.0 VGA compatible controller: ATI Technologies Inc Rage XL (rev 27) (prog-if 00 [VGA])
Subsystem: ATI Technologies Inc: Unknown device 8008
Flags: bus master, stepping, medium devsel, latency 32, IRQ 46 Memory at fc000000 (32-bit, non-prefetchable) [size=16M]
I/O ports at d400 [size=256]
Memory at fb800000 (32-bit, non-prefetchable) [size=4K]
Expansion ROM at febe0000 [disabled] [size=128K]
Capabilities: [5c] Power Management version 2

00:0f.0 ISA bridge: ServerWorks CSB5 South Bridge (rev 93)
Subsystem: ServerWorks CSB5 South Bridge
Flags: bus master, medium devsel, latency 32

00:0f.1 IDE interface: ServerWorks CSB5 IDE Controller (rev 93) (prog-if 8a [Master SecP PriP])
Subsystem: ServerWorks CSB5 IDE Controller
Flags: bus master, medium devsel, latency 64
I/O ports at <ignored>
I/O ports at <ignored>
I/O ports at <ignored>
I/O ports at <ignored>
I/O ports at a800 [size=16]

00:0f.2 USB Controller: ServerWorks OSB4/CSB5 OHCI USB Controller (rev 05) (prog-if 10 [OHCI])
Subsystem: ServerWorks OSB4/CSB5 OHCI USB Controller
Flags: bus master, medium devsel, latency 32, IRQ 11
Memory at fb000000 (32-bit, non-prefetchable) [size=4K]

00:0f.3 Host bridge: ServerWorks GCLE Host Bridge
Subsystem: ServerWorks: Unknown device 0230
Flags: bus master, medium devsel, latency 0

00:10.0 Host bridge: ServerWorks: Unknown device 0101 (rev 03)
Flags: 66Mhz, medium devsel
Capabilities: [60]
00:10.2 Host bridge: ServerWorks: Unknown device 0101 (rev 03)
Flags: 66Mhz, medium devsel
Capabilities: [60]
00:11.0 Host bridge: ServerWorks: Unknown device 0101 (rev 03)
Flags: 66Mhz, medium devsel
Capabilities: [60]
00:11.2 Host bridge: ServerWorks: Unknown device 0101 (rev 03)
Flags: 66Mhz, medium devsel
Capabilities: [60]
02:04.0 SCSI storage controller: LSI Logic / Symbios Logic 53c1030 (rev 07)

        Subsystem: LSI Logic / Symbios Logic: Unknown device 1000
        Flags: bus master, 66Mhz, medium devsel, latency 72, IRQ 22
        I/O ports at a000 [size=256]
        Memory at fa000000 (64-bit, non-prefetchable) [size=64K]
        Memory at f9800000 (64-bit, non-prefetchable) [size=64K]
        Expansion ROM at fe900000 [disabled] [size=1M]
        Capabilities: [50] Power Management version 2
        Capabilities: [58] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable-
        Capabilities: [68]
02:04.1 SCSI storage controller: LSI Logic / Symbios Logic 53c1030 (rev 07)
        Subsystem: LSI Logic / Symbios Logic: Unknown device 1000
        Flags: bus master, 66Mhz, medium devsel, latency 72, IRQ 23
        I/O ports at 9800 [size=256]
        Memory at f9000000 (64-bit, non-prefetchable) [size=64K]
        Memory at f8800000 (64-bit, non-prefetchable) [size=64K]
        Expansion ROM at fe800000 [disabled] [size=1M]
        Capabilities: [50] Power Management version 2
        Capabilities: [58] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable-
        Capabilities: [68]

03:02.0 Ethernet controller: Intel Corp. 82544GC Gigabit Ethernet Controller (LOM) (rev 02)
Subsystem: Intel Corp.: Unknown device 110d
Flags: bus master, 66Mhz, medium devsel, latency 32, IRQ 19
Memory at f8000000 (64-bit, non-prefetchable) [size=128K]
Memory at f7800000 (64-bit, non-prefetchable) [size=128K]
I/O ports at 9400 [size=32]
Expansion ROM at fe7e0000 [disabled] [size=128K]
Capabilities: [dc] Power Management version 2
Capabilities: [e4] PCI-X non-bridge device.
Capabilities: [f0] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable-

root@setup:~ {547} $ cat /proc/interrupts
CPU0 CPU1

  0:    1737732    1760169    IO-APIC-edge  timer
  1:        260        260    IO-APIC-edge  keyboard
  2:          0          0          XT-PIC  cascade
  4:         10         10    IO-APIC-edge  serial
 15:          5         17    IO-APIC-edge  ide1
 18:   22903530   23147515   IO-APIC-level  eth0
 22:    1695902    1726918   IO-APIC-level  ioc0
 23:         25         17   IO-APIC-level  ioc1
NMI:          0          0
LOC:    1398848    1398847
ERR:          0
MIS:          0

Any ideas?

- - andreas

- --

Andreas Haumer                     | mailto:[email protected]
*x Software + Systeme              | http://www.xss.co.at/
Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
A-1100 Vienna, Austria             | Fax: +43-1-6060114-71

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQE+26g5xJmyeGcXPhERAj4tAJ98iFyniQ2N3P0jKkfQqla2GBVKXwCbBr36 +2kB0HBIKcVXA+lkxAXeVMU=
=Xk1O
-----END PGP SIGNATURE-----

Alan Cox

2003-06-02 20:20 UTC #1 lore raw

Quoted 8 lines · Andreas Haumer

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi!
(I've already reported this previously as post scriptum to a different bugreport, so it might have slipped through unnoticed...)

Does this occur if you boot with "notsc" or if you run a non ACPI kernel ?

Andreas Haumer

author

2003-06-03 09:41 UTC #2 lore raw

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

Quoted 7 lines · Alan Cox

Quoted 5 lines · Andreas Haumer
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi!

Quoted 8 lines

Quoted 3 lines
(I've already reported this previously as post scriptum to a different bugreport, so it might have slipped through unnoticed...)
Does this occur if you boot with "notsc" or if you run a non ACPI kernel ?

I did some more tests on this machine, with the following results:

1.) Kernel 2.4.21-rc6-ac1 (ACPI compiled as module)
a) no special kernel commandline option
-> System doesn't boot, fusion MPT driver doesn't find
the hardware listed on the PCI bus (problem already reported in a previous mail)

b) boot with "acpi=off"
-> System does boot, fusion MTP driver finds the controller -> time runs 2.5 times too fast

c) boot with "acpi=off notsc"
-> same as 1b)

2.) Kernel 2.4.21-rc4 (ACPI compiled as module)

    a) no special kernel commandline option
       -> System does boot, fusion MTP driver finds the controller
       -> time runs 2.5 times too fast

    b) boot with "acpi=off"
       -> Same as 2a)

    c) boot with "acpi=off notsc"
       -> Same as 2a)

3.) Kernel 2.4.21-rc2-ac2 (ACPI compiled as module)
a) no special kernel commandline option
-> System doesn't boot, fusion MPT driver doesn't find
the hardware listed on the PCI bus

b) boot with "acpi=off"
-> System does boot, fusion MTP driver finds the controller -> time runs 2.5 times too fast

c) boot with "acpi=off notsc"
-> same as 3b)

4.) Kernel 2.4.21-rc6-ac1 (without ACPI)

    a) no special kernel commandline option
       -> System does boot, fusion MTP driver finds the controller
       -> time runs 2.5 times too fast

    b) boot with "notsc"
       -> Same as 4a)

5.) Kernel 2.4.21-rc4 (ACPI compiled as module)
floppy-minisystem, booting into initrd only
modules loaded: unix.o. ext2.o, floppy.o
-> time runs 2.5 times too fast

Time "acceleration" always seems to be by a factor of exactly 2.5 Proof:
root@setup:~ {502} $ ntpdate ntp.xss.co.at; sleep 300; ntpdate ntp.xss.co.at
3 Jun 11:31:52 ntpdate[1185]: step time server 194.152.162.17 offset -895.898772 sec 3 Jun 11:33:52 ntpdate[1187]: step time server 194.152.162.17 offset -180.088187 sec

(I first sync the system clock with our NTP timeserver, then sleep 300 seconds, and then again sync against our NTP server)

Sleeping 300 "system seconds" takes 120 "wall clock seconds" and brings the system clock ahead of 180 seconds, which gives a speed up factor of 2.5. Does this number ring any bell?

Additional info: I'm booting with LILO and have set LILO timeout to 10 seconds. This works fine: "LILO time" is the same as wall clock time. So this speed up must be triggered somewhere inside the kernel, it seems...

Any idea someone?

- - andreas

- --

Andreas Haumer                     | mailto:[email protected]
*x Software + Systeme              | http://www.xss.co.at/
Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
A-1100 Vienna, Austria             | Fax: +43-1-6060114-71

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQE+3HBPxJmyeGcXPhERAplBAKCYqTbno0EnP9WqKwtzXyUjBCUosgCfT7qw vfPbba5yqYD0qUI9BDJBD40=
=i+JF
-----END PGP SIGNATURE-----

Andreas Haumer

author

2003-06-03 11:46 UTC #3 lore raw

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi!

I have some more information

6.) Kernel 2.4.19-ac4 (ACPI compiled as modules)

    a) no special kernel commandline option
       -> System does boot, fusion MTP driver finds the controller
       -> time runs 2.5 times too fast

    b) boot with "acpi=off"
       -> Same as 6a)

    c) boot with "acpi=off notsc"
       -> Same as 6a)

I tried also the following BIOS settings, but without improving the situation:

*) Hyperthreading "disabled"
*) MPS 1.4 Support "disabled"
*) BIOS Update "disabled"
*) Plug & Play "yes" (this makes it even worse: linux crashes
eventually as it gets confused by IRQ setting)

- - andreas

- --

Andreas Haumer                     | mailto:[email protected]
*x Software + Systeme              | http://www.xss.co.at/
Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
A-1100 Vienna, Austria             | Fax: +43-1-6060114-71

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQE+3I2/xJmyeGcXPhERAjXOAKCc7lD6Y+oYUyvNtqVT/dj+HBpP+wCfcWI4 /Pfay3OP/7/XPxpIbBAE9X4=
=cIUN
-----END PGP SIGNATURE-----

Zephaniah E. Hull

2003-06-03 12:03 UTC #4 lore raw

Quoted 3 lines · Andreas Haumer

Hi!
I have some more information

Just a wild hunch, do you still have the 2.5x speed if you boot with init=/bin/sh and don't start anything?

(If you're wondering, I suspect something like tickadj being run in the init scripts with really really wrong numbers.)

Zephaniah E. Hull.

Signature · 6 lines

	1024D/E65A7801 Zephaniah E. Hull <[email protected]>
	   92ED 94E4 B1E6 3624 226D  5727 4453 008B E65A 7801
	    CCs of replies from mailing lists are requested.

"I am ecstatic that some moron re-invented a 1995 windows fuckup."
        -- Alan Cox

Andreas Haumer

author

2003-06-03 12:25 UTC #5 lore raw

In reply to Andreas Haumer

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi!

Andreas Haumer wrote:
[...]

Quoted 6 lines

*) Hyperthreading "disabled"
*) MPS 1.4 Support "disabled"
*) BIOS Update "disabled"
*) Plug & Play "yes" (this makes it even worse: linux crashes
eventually as it gets confused by IRQ setting)

I've found it:

It's BIOS setting "USB legacy support", which I had on "Auto" (default setting).

As soon as I disable "USB legacy support", system time is now in sync with wall clock time!

root@setup:~ {508} $ ntpdate ntp.xss.co.at; sleep 60; ntpdate ntp.xss.co.at
3 Jun 14:35:32 ntpdate[1141]: adjust time server 194.152.162.17 offset 0.005909 sec 3 Jun 14:36:32 ntpdate[1146]: adjust time server 194.152.162.17 offset -0.002158 sec

Uaaahhhhhhhh!!!!!

Linux also now reports an interesting error on bootup, which isn't there when booting with "USB legacy support=Auto"

[...]
..MP-BIOS bug: 8254 timer not connected to IO-APIC [...]

(This is 2.4.21-rc6-ac1 booting with "acpi=off")

What the hell does USB legacy support have to do with all this??? This is weird!

- - andreas

- --

Andreas Haumer                     | mailto:[email protected]
*x Software + Systeme              | http://www.xss.co.at/
Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
A-1100 Vienna, Austria             | Fax: +43-1-6060114-71

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQE+3JbkxJmyeGcXPhERAmzLAJ48RMa7o6Zn0Ko6YxFn4y5UVeGl+wCgwQwN DSjZtxe4QDEL8I812fz2h1U=
=8zBV
-----END PGP SIGNATURE-----

Alan Cox

2003-06-03 13:45 UTC #6 lore raw

USB legacy generates SMI events. It could be that something the BIOS SMI magic is doing is disrupting the system clock or causing extra interrupts

Andreas Haumer

author

2003-06-03 14:10 UTC #7 lore raw

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi!

Quoted 4 lines · Alan Cox

USB legacy generates SMI events. It could be that something the BIOS SMI magic is doing is disrupting the system clock or causing extra interrupts

Thanks for your support.

The "USB legacy support" settings slipped through when I was preparing the system. I was too much concentrated on the ACPI/Fusion MPT problems...

But something must be wrong in this area, as the kernel prints this "..MP-BIOS bug: 8254 timer not connected to IO-APIC" message when USB legacy support is disabled.

Anyway, I'll have this system available for tests in the next few days (it is to be installed as production server next week), and it _will_ suffer a lot from the stress tests I've planned... ;-)

So if you want me to test anything (ACPI patches, perhaps?) you're very welcome!

- - andreas

- --

Andreas Haumer                     | mailto:[email protected]
*x Software + Systeme              | http://www.xss.co.at/
Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
A-1100 Vienna, Austria             | Fax: +43-1-6060114-71

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQE+3K9fxJmyeGcXPhERAsqfAJkBJS/W5DxjPLYh2f7795j9Xx2pDACfcFXG rAKbPhQrxpGnAW3xFv1G3u4=
=J0bc
-----END PGP SIGNATURE-----

Martin Schlemmer

2003-06-03 14:48 UTC #8 lore raw

Quoted 8 lines · Andreas Haumer

The "USB legacy support" settings slipped through when I was preparing the system. I was too much concentrated on the ACPI/Fusion MPT problems...
But something must be wrong in this area, as the kernel prints this "..MP-BIOS bug: 8254 timer not connected to IO-APIC" message when USB legacy support is disabled.

Quoted 8 lines

Anyway, I'll have this system available for tests in the next few days (it is to be installed as production server next week), and it _will_ suffer a lot from the stress tests I've planned... ;-)
So if you want me to test anything (ACPI patches, perhaps?) you're very welcome!

What about contacting the manufacturer to fix the bios ?

Regards,

Signature · 1 line

Martin Schlemmer

Andreas Haumer

author

2003-06-03 15:18 UTC #9 lore raw

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Quoted 11 lines · Martin Schlemmer

Quoted 8 lines · Andreas Haumer
The "USB legacy support" settings slipped through when I was preparing the system. I was too much concentrated on the ACPI/Fusion MPT problems...
But something must be wrong in this area, as the kernel prints this "..MP-BIOS bug: 8254 timer not connected to IO-APIC" message when USB legacy support is disabled.

[...]

Quoted 3 lines

What about contacting the manufacturer to fix the bios ?

It's on my TODO list, but I want to do some more tests first.

The Asus AP1700-S5 does officially list Linux as supported OS, (and it's a really nice piece of server hardware, too), so they should be definitely interested in fixing this. It might be not so clear who's to blame for this problem, though.

Is anyone from ASUSTeK Computers Inc. around here on this list?

- - andreas

- --

Andreas Haumer                     | mailto:[email protected]
*x Software + Systeme              | http://www.xss.co.at/
Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
A-1100 Vienna, Austria             | Fax: +43-1-6060114-71

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQE+3L9gxJmyeGcXPhERAh8OAJ9uCh3S60u9RTGYyfg2TN3PDKc8RwCfWu7S 1fdoACA6B+7FjKxzXsM+xQs=
=OoCm
-----END PGP SIGNATURE-----

Martin Schlemmer

2003-06-04 05:42 UTC #10 lore raw

Quoted 17 lines · Andreas Haumer

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 11 lines · Martin Schlemmer
Quoted 8 lines · Andreas Haumer
The "USB legacy support" settings slipped through when I was preparing the system. I was too much concentrated on the ACPI/Fusion MPT problems...
But something must be wrong in this area, as the kernel prints this "..MP-BIOS bug: 8254 timer not connected to IO-APIC" message when USB legacy support is disabled.
[...]
Quoted 1 line

Quoted 9 lines

Quoted 2 lines
What about contacting the manufacturer to fix the bios ?
It's on my TODO list, but I want to do some more tests first.
The Asus AP1700-S5 does officially list Linux as supported OS, (and it's a really nice piece of server hardware, too), so they should be definitely interested in fixing this. It might be not so clear who's to blame for this problem, though.

I am assuming you have not tried another OS just to verify if it is a linux only problem ?

Quoted 2 lines

Is anyone from ASUSTeK Computers Inc. around here on this list?

I might know a email or two, but I do not know if I am allowed .... If you can get more info (also maybe if its a linux only problem), Drop me a mail ... maybe I can just forward it if you list all the info.

Also ... have you tried to look if there is a BIOS update out yet?

Regards,

Signature · 1 line

Martin Schlemmer

Andreas Haumer

author

2003-06-04 08:36 UTC #11 lore raw

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi!

Martin Schlemmer wrote:
[...]

Quoted 4 lines

I am assuming you have not tried another OS just to verify if it is a linux only problem ?

You're right :-)
We're a Linux-only company and we don't even have another OS available capable of running on this system...

Quoted 3 lines

Also ... have you tried to look if there is a BIOS update out yet?

It has BIOS Revision 1007, which is current as of today (I just looked at the Asus website)

- - andreas

- --

Andreas Haumer                     | mailto:[email protected]
*x Software + Systeme              | http://www.xss.co.at/
Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
A-1100 Vienna, Austria             | Fax: +43-1-6060114-71

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQE+3bKgxJmyeGcXPhERAqDIAJ9+FIxHmW6smLFS2aUpJrquSVWTEQCfcsIm cwOmzyH/jfdtJfWydzbvsUU=
=yjD7
-----END PGP SIGNATURE-----

[email protected]

2003-06-03 15:03 UTC #12 lore raw

In reply to Andreas Haumer

I reported this same problem for kernel 2.5.69/70 on pentiumIII with chipset i810, but it seems the reports has been unnoticed.

I replicated your tests with 2.5.70 without acpi, same results here.

Quoted 123 lines · Andreas Haumer

Date: Tue, 03 Jun 2003 11:54:26 +0200
From: Andreas Haumer <[email protected]>
To: Alan Cox <[email protected]>
Cc: Linux Kernel Mailing List <[email protected]>
Subject: Re: system clock speed too high?
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi,
Quoted 7 lines · Alan Cox
Quoted 5 lines · Andreas Haumer
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi!
Quoted 8 lines
Quoted 3 lines
(I've already reported this previously as post scriptum to a different bugreport, so it might have slipped through unnoticed...)
Does this occur if you boot with "notsc" or if you run a non ACPI kernel ?
I did some more tests on this machine, with the following results:
1.) Kernel 2.4.21-rc6-ac1 (ACPI compiled as module)
a) no special kernel commandline option
-> System doesn't boot, fusion MPT driver doesn't find
the hardware listed on the PCI bus (problem already reported in a previous mail)
b) boot with "acpi=off"
-> System does boot, fusion MTP driver finds the controller -> time runs 2.5 times too fast
c) boot with "acpi=off notsc"
-> same as 1b)
2.) Kernel 2.4.21-rc4 (ACPI compiled as module)
    a) no special kernel commandline option
       -> System does boot, fusion MTP driver finds the controller
       -> time runs 2.5 times too fast

    b) boot with "acpi=off"
       -> Same as 2a)

    c) boot with "acpi=off notsc"
       -> Same as 2a)
3.) Kernel 2.4.21-rc2-ac2 (ACPI compiled as module)
a) no special kernel commandline option
-> System doesn't boot, fusion MPT driver doesn't find
the hardware listed on the PCI bus
b) boot with "acpi=off"
-> System does boot, fusion MTP driver finds the controller -> time runs 2.5 times too fast
c) boot with "acpi=off notsc"
-> same as 3b)
4.) Kernel 2.4.21-rc6-ac1 (without ACPI)
    a) no special kernel commandline option
       -> System does boot, fusion MTP driver finds the controller
       -> time runs 2.5 times too fast

    b) boot with "notsc"
       -> Same as 4a)
5.) Kernel 2.4.21-rc4 (ACPI compiled as module)
floppy-minisystem, booting into initrd only
modules loaded: unix.o. ext2.o, floppy.o
-> time runs 2.5 times too fast
Time "acceleration" always seems to be by a factor of exactly 2.5 Proof:
root@setup:~ {502} $ ntpdate ntp.xss.co.at; sleep 300; ntpdate ntp.xss.co.at
3 Jun 11:31:52 ntpdate[1185]: step time server 194.152.162.17 offset -895.898772 sec 3 Jun 11:33:52 ntpdate[1187]: step time server 194.152.162.17 offset -180.088187 sec
(I first sync the system clock with our NTP timeserver, then sleep 300 seconds, and then again sync against our NTP server)
Sleeping 300 "system seconds" takes 120 "wall clock seconds" and brings the system clock ahead of 180 seconds, which gives a speed up factor of 2.5. Does this number ring any bell?
Additional info: I'm booting with LILO and have set LILO timeout to 10 seconds. This works fine: "LILO time" is the same as wall clock time. So this speed up must be triggered somewhere inside the kernel, it seems...
Any idea someone?
- - andreas
- --
Andreas Haumer                     | mailto:[email protected]
*x Software + Systeme              | http://www.xss.co.at/
Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
A-1100 Vienna, Austria             | Fax: +43-1-6060114-71
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQE+3HBPxJmyeGcXPhERAplBAKCYqTbno0EnP9WqKwtzXyUjBCUosgCfT7qw vfPbba5yqYD0qUI9BDJBD40=
=i+JF
-----END PGP SIGNATURE-----
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Andreas Haumer

author

2003-06-03 15:24 UTC #13 lore raw

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi!

Quoted 5 lines · [email protected] wrote:

I reported this same problem for kernel 2.5.69/70 on pentiumIII with chipset i810, but it seems the reports has been unnoticed.
I replicated your tests with 2.5.70 without acpi, same results here.

Hm, do you have "USB legacy support" enabled on your system, too? Does the problem go away if you disable it?

What motherboard and BIOS does your computer have (manufacturer, version numbers)?

(This is a totally different chipset, so the problems might be completely unrelated)

- - andreas

- --

Andreas Haumer                     | mailto:[email protected]
*x Software + Systeme              | http://www.xss.co.at/
Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
A-1100 Vienna, Austria             | Fax: +43-1-6060114-71

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQE+3MDWxJmyeGcXPhERAmk+AKDER5RkBffYSYFpnNFq7nKhaqPwyACgr1rJ UxspwdElGOOGQs58Ol1N5v4=
=Gb0Z
-----END PGP SIGNATURE-----

[email protected]

2003-06-03 15:30 UTC #14 lore raw

No, I do not have USB enabled at all.
anyway I see exactly the same results, also if the problem could be somewhere else.

Luigi

Quoted 43 lines · Andreas Haumer

Date: Tue, 03 Jun 2003 17:38:09 +0200
From: Andreas Haumer <[email protected]>
To: [email protected]
Cc: Alan Cox <[email protected]>,
Linux Kernel Mailing List <[email protected]>
Subject: Re: system clock speed too high?
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi!
Quoted 5 lines · [email protected] wrote:
I reported this same problem for kernel 2.5.69/70 on pentiumIII with chipset i810, but it seems the reports has been unnoticed.
I replicated your tests with 2.5.70 without acpi, same results here.
Hm, do you have "USB legacy support" enabled on your system, too? Does the problem go away if you disable it?
What motherboard and BIOS does your computer have (manufacturer, version numbers)?
(This is a totally different chipset, so the problems might be completely unrelated)
- - andreas
- --
Andreas Haumer                     | mailto:[email protected]
*x Software + Systeme              | http://www.xss.co.at/
Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
A-1100 Vienna, Austria             | Fax: +43-1-6060114-71
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQE+3MDWxJmyeGcXPhERAmk+AKDER5RkBffYSYFpnNFq7nKhaqPwyACgr1rJ UxspwdElGOOGQs58Ol1N5v4=
=Gb0Z
-----END PGP SIGNATURE-----

Randy.Dunlap

2003-06-03 15:35 UTC #15 lore raw

Hi,

Did you check to see if USB is enabled in your BIOS setup? Nothing to do with Linux USB kernel support....

~Randy

On Tue, 3 Jun 2003 17:43:42 +0200 (CEST) [email protected] wrote:

| 
| No, I do not have USB enabled at all.
| anyway I see exactly the same results, also if the problem could be somewhere
| else.
| 
| Luigi
| 
| On Tue, 3 Jun 2003, Andreas Haumer wrote:
| 
| > Date: Tue, 03 Jun 2003 17:38:09 +0200
| > From: Andreas Haumer <[email protected]>
| > To: [email protected]
| > Cc: Alan Cox <[email protected]>,
| >      Linux Kernel Mailing List <[email protected]>
| > Subject: Re: system clock speed too high?
| >
| > -----BEGIN PGP SIGNED MESSAGE-----
| > Hash: SHA1
| >
| > Hi!
| >
| > [email protected] wrote:
| > > I reported this same problem for kernel 2.5.69/70 on pentiumIII with chipset
| > > i810, but it seems the reports has been unnoticed.
| > >
| > > I replicated your tests with 2.5.70 without acpi, same results here.
| > >
| > Hm, do you have "USB legacy support" enabled on your
| > system, too? Does the problem go away if you disable it?
| >
| > What motherboard and BIOS does your computer have (manufacturer,
| > version numbers)?
| >
| > (This is a totally different chipset, so the problems
| > might be completely unrelated)
| >
| > - - andreas
| >
| > - --
| > Andreas Haumer                     | mailto:[email protected]
| > *x Software + Systeme              | http://www.xss.co.at/
| > Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
| > A-1100 Vienna, Austria             | Fax: +43-1-6060114-71

[email protected]

2003-06-03 15:58 UTC #16 lore raw

USB is disabled in BIOS and ACPI is disabled too, but the problem persists.
It is even difficoult to type the keyb as fast as it is on alll irqs.

Quoted 65 lines · Randy.Dunlap

Date: Tue, 3 Jun 2003 08:47:37 -0700
From: Randy.Dunlap <[email protected]>
To: [email protected]
Cc: [email protected], [email protected], [email protected]
Subject: Re: system clock speed too high?
Hi,
Did you check to see if USB is enabled in your BIOS setup? Nothing to do with Linux USB kernel support....
~Randy
On Tue, 3 Jun 2003 17:43:42 +0200 (CEST) [email protected] wrote:
|
| No, I do not have USB enabled at all.
| anyway I see exactly the same results, also if the problem could be somewhere
| else.
|
| Luigi
|
| On Tue, 3 Jun 2003, Andreas Haumer wrote:
|
| > Date: Tue, 03 Jun 2003 17:38:09 +0200
| > From: Andreas Haumer <[email protected]>
| > To: [email protected]
| > Cc: Alan Cox <[email protected]>,
| >      Linux Kernel Mailing List <[email protected]>
| > Subject: Re: system clock speed too high?
| >
| > -----BEGIN PGP SIGNED MESSAGE-----
| > Hash: SHA1
| >
| > Hi!
| >
| > [email protected] wrote:
| > > I reported this same problem for kernel 2.5.69/70 on pentiumIII with chipset
| > > i810, but it seems the reports has been unnoticed.
| > >
| > > I replicated your tests with 2.5.70 without acpi, same results here.
| > >
| > Hm, do you have "USB legacy support" enabled on your
| > system, too? Does the problem go away if you disable it?
| >
| > What motherboard and BIOS does your computer have (manufacturer,
| > version numbers)?
| >
| > (This is a totally different chipset, so the problems
| > might be completely unrelated)
| >
| > - - andreas
| >
| > - --
| > Andreas Haumer                     | mailto:[email protected]
| > *x Software + Systeme              | http://www.xss.co.at/
| > Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
| > A-1100 Vienna, Austria             | Fax: +43-1-6060114-71
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

george anzinger

2003-06-02 20:23 UTC #17 lore raw

In reply to Andreas Haumer

Quoted 9 lines · Andreas Haumer

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi!
(I've already reported this previously as post scriptum to a different bugreport, so it might have slipped through unnoticed...)

Quoted 5 lines

I have a quite strange phenomenon here: I see a ~2.5 times speed up of system time on a Asus AP1700-S5 server with Linux-2.4.21-rc6-ac1.
Simple proof: a "sleep 300" command terminates after exactly 120 seconds of wall clock time.

Just as a wild shot in the dark, what speed does the kernel think the cpu is running at and does this match what the BIOS thinks?

It sounds like the CLOCK_TICK_RATE is wrong. This would show up as the kernel thinking the cpu was fast also.

You pin this to a particular kernel version. Do other kernel versions do a better job?

-g

Quoted 154 lines

Or just look at this:
root@setup:~ {544} $ ntpdate ntp.xss.co.at; ntpdate ntp.xss.co.at; sleep 300; ntpdate ntp.xss.co.at
2 Jun 21:24:45 ntpdate[3768]: step time server 194.152.162.17 offset -38.679740 sec 2 Jun 21:24:45 ntpdate[3769]: adjust time server 194.152.162.17 offset -0.069998 sec 2 Jun 21:26:45 ntpdate[3786]: step time server 194.152.162.17 offset -180.053781 sec
Impressive, isn't it?
This "acceleration" of time shows with other time measures, like "date", system uptime, process time, etc., too.
(HZ is still defined as 100, I haven't changed anything in the kernel sources)
System info:
Asus AP1700-S5, Asus PR-DLS533 Motherboard, Intel Xeon 2.4GHz CPU, 512MB RAM
root@setup:~ {545} $ uname -a
Linux setup 2.4.21-rc6-ac1 #1 SMP Sat May 31 20:04:18 CEST 2003 i686 unknown
root@setup:~ {546} $ lspci -v
00:00.0 Host bridge: ServerWorks CNB20-HE Host Bridge (rev 31)
Flags: fast devsel
00:00.1 Host bridge: ServerWorks CNB20-HE Host Bridge
Flags: fast devsel
00:00.2 Host bridge: ServerWorks CNB20-HE Host Bridge
Flags: fast devsel
00:02.0 Ethernet controller: Intel Corp. 82540EM Gigabit Ethernet Controller (rev 02)
Subsystem: Intel Corp. 82540EM Gigabit Ethernet Controller
Flags: bus master, 66Mhz, medium devsel, latency 32, IRQ 18 Memory at fd800000 (32-bit, non-prefetchable) [size=128K]
I/O ports at d800 [size=64]
Capabilities: [dc] Power Management version 2
Capabilities: [e4] PCI-X non-bridge device.
Capabilities: [f0] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable-
00:03.0 VGA compatible controller: ATI Technologies Inc Rage XL (rev 27) (prog-if 00 [VGA])
Subsystem: ATI Technologies Inc: Unknown device 8008
Flags: bus master, stepping, medium devsel, latency 32, IRQ 46 Memory at fc000000 (32-bit, non-prefetchable) [size=16M]
I/O ports at d400 [size=256]
Memory at fb800000 (32-bit, non-prefetchable) [size=4K]
Expansion ROM at febe0000 [disabled] [size=128K]
Capabilities: [5c] Power Management version 2
00:0f.0 ISA bridge: ServerWorks CSB5 South Bridge (rev 93)
Subsystem: ServerWorks CSB5 South Bridge
Flags: bus master, medium devsel, latency 32
00:0f.1 IDE interface: ServerWorks CSB5 IDE Controller (rev 93) (prog-if 8a [Master SecP PriP])
Subsystem: ServerWorks CSB5 IDE Controller
Flags: bus master, medium devsel, latency 64
I/O ports at <ignored>
I/O ports at <ignored>
I/O ports at <ignored>
I/O ports at <ignored>
I/O ports at a800 [size=16]
00:0f.2 USB Controller: ServerWorks OSB4/CSB5 OHCI USB Controller (rev 05) (prog-if 10 [OHCI])
Subsystem: ServerWorks OSB4/CSB5 OHCI USB Controller
Flags: bus master, medium devsel, latency 32, IRQ 11
Memory at fb000000 (32-bit, non-prefetchable) [size=4K]
00:0f.3 Host bridge: ServerWorks GCLE Host Bridge
Subsystem: ServerWorks: Unknown device 0230
Flags: bus master, medium devsel, latency 0
00:10.0 Host bridge: ServerWorks: Unknown device 0101 (rev 03)
Flags: 66Mhz, medium devsel
Capabilities: [60]
00:10.2 Host bridge: ServerWorks: Unknown device 0101 (rev 03)
Flags: 66Mhz, medium devsel
Capabilities: [60]
00:11.0 Host bridge: ServerWorks: Unknown device 0101 (rev 03)
Flags: 66Mhz, medium devsel
Capabilities: [60]
00:11.2 Host bridge: ServerWorks: Unknown device 0101 (rev 03)
Flags: 66Mhz, medium devsel
Capabilities: [60]
02:04.0 SCSI storage controller: LSI Logic / Symbios Logic 53c1030 (rev 07)
        Subsystem: LSI Logic / Symbios Logic: Unknown device 1000
        Flags: bus master, 66Mhz, medium devsel, latency 72, IRQ 22
        I/O ports at a000 [size=256]
        Memory at fa000000 (64-bit, non-prefetchable) [size=64K]
        Memory at f9800000 (64-bit, non-prefetchable) [size=64K]
        Expansion ROM at fe900000 [disabled] [size=1M]
        Capabilities: [50] Power Management version 2
        Capabilities: [58] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable-
        Capabilities: [68]
02:04.1 SCSI storage controller: LSI Logic / Symbios Logic 53c1030 (rev 07)
        Subsystem: LSI Logic / Symbios Logic: Unknown device 1000
        Flags: bus master, 66Mhz, medium devsel, latency 72, IRQ 23
        I/O ports at 9800 [size=256]
        Memory at f9000000 (64-bit, non-prefetchable) [size=64K]
        Memory at f8800000 (64-bit, non-prefetchable) [size=64K]
        Expansion ROM at fe800000 [disabled] [size=1M]
        Capabilities: [50] Power Management version 2
        Capabilities: [58] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable-
        Capabilities: [68]
03:02.0 Ethernet controller: Intel Corp. 82544GC Gigabit Ethernet Controller (LOM) (rev 02)
Subsystem: Intel Corp.: Unknown device 110d
Flags: bus master, 66Mhz, medium devsel, latency 32, IRQ 19
Memory at f8000000 (64-bit, non-prefetchable) [size=128K]
Memory at f7800000 (64-bit, non-prefetchable) [size=128K]
I/O ports at 9400 [size=32]
Expansion ROM at fe7e0000 [disabled] [size=128K]
Capabilities: [dc] Power Management version 2
Capabilities: [e4] PCI-X non-bridge device.
Capabilities: [f0] Message Signalled Interrupts: 64bit+ Queue=0/0 Enable-
root@setup:~ {547} $ cat /proc/interrupts
CPU0 CPU1
  0:    1737732    1760169    IO-APIC-edge  timer
  1:        260        260    IO-APIC-edge  keyboard
  2:          0          0          XT-PIC  cascade
  4:         10         10    IO-APIC-edge  serial
 15:          5         17    IO-APIC-edge  ide1
 18:   22903530   23147515   IO-APIC-level  eth0
 22:    1695902    1726918   IO-APIC-level  ioc0
 23:         25         17   IO-APIC-level  ioc1
NMI:          0          0
LOC:    1398848    1398847
ERR:          0
MIS:          0
Any ideas?
- - andreas
- --
Andreas Haumer                     | mailto:[email protected]
*x Software + Systeme              | http://www.xss.co.at/
Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
A-1100 Vienna, Austria             | Fax: +43-1-6060114-71
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQE+26g5xJmyeGcXPhERAj4tAJ98iFyniQ2N3P0jKkfQqla2GBVKXwCbBr36 +2kB0HBIKcVXA+lkxAXeVMU=
=Xk1O
-----END PGP SIGNATURE-----
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Signature · 3 lines

George Anzinger   [email protected]
High-res-timers:  http://sourceforge.net/projects/high-res-timers/
Preemption patch: http://www.kernel.org/pub/linux/kernel/people/rml

Andreas Haumer

author

2003-06-03 09:51 UTC #18 lore raw

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi!

Quoted 2 lines · george anzinger

Andreas Haumer wrote:

[...]

Quoted 7 lines

Quoted 5 lines
I have a quite strange phenomenon here: I see a ~2.5 times speed up of system time on a Asus AP1700-S5 server with Linux-2.4.21-rc6-ac1.
Simple proof: a "sleep 300" command terminates after exactly 120 seconds of wall clock time.

Quoted 6 lines

Just as a wild shot in the dark, what speed does the kernel think the cpu is running at and does this match what the BIOS thinks?
It sounds like the CLOCK_TICK_RATE is wrong. This would show up as the kernel thinking the cpu was fast also.

Hm, I don't think this is the case.
BIOS reports (correctly) a single, hyperthreaded Xeon CPU with 2.4GHz

Kernel reports the same:
root@setup:~ {503} $ uname -a
Linux setup 2.4.21-rc6-ac1 #2 SMP Tue Jun 3 09:45:13 CEST 2003 i686 unknown

root@setup:~ {504} $ cat /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 15
model : 2
model name : Intel(R) Xeon(TM) CPU 2.40GHz
stepping : 7
cpu MHz : 2392.065
cache size : 512 KB
physical id : 0
siblings : 2
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 2
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm
bogomips : 4771.02

processor : 1
vendor_id : GenuineIntel
cpu family : 15
model : 2
model name : Intel(R) Xeon(TM) CPU 2.40GHz
stepping : 7
cpu MHz : 2392.065
cache size : 512 KB
physical id : 0
siblings : 2
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 2
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm
bogomips : 4771.02

root@setup:~ {505} $ ntpdate ntp.xss.co.at; sleep 500; ntpdate ntp.xss.co.at
3 Jun 11:58:16 ntpdate[1118]: step time server 194.152.162.17 offset -177.268071 sec 3 Jun 12:01:36 ntpdate[1120]: step time server 194.152.162.17 offset -300.106769 sec

(Sleeping 500 "system seconds" takes 200 "wall clock seconds")

Quoted 3 lines

You pin this to a particular kernel version. Do other kernel versions do a better job?

So far I tried with:

2.4.21-rc2-ac2 (ACPI compiled as module) 2.4.21-rc4 (ACPI compiled as module) 2.4.21-rc6-ac1 (ACPI compiled as module) 2.4.21-rc6-ac1 (no ACPI comiled at all)

Time acceleration is the same for all kernels.

I'll try it with some older kernels, too.

- - andreas

- --

Andreas Haumer                     | mailto:[email protected]
*x Software + Systeme              | http://www.xss.co.at/
Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
A-1100 Vienna, Austria             | Fax: +43-1-6060114-71

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQE+3HKxxJmyeGcXPhERAkZ/AJ9tjgI6K2HtM/tL15nKDFIKzdcvQgCfYOGY i96/o6Nyxlem5yOp1A2Q//s=
=plFh
-----END PGP SIGNATURE-----

george anzinger

2003-06-03 18:59 UTC #19 lore raw

Re: system clock speed too high? hyperthread problem?

Quoted 22 lines · Andreas Haumer

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi!
Quoted 2 lines · george anzinger
Andreas Haumer wrote:
[...]
Quoted 10 lines
Quoted 5 lines
I have a quite strange phenomenon here: I see a ~2.5 times speed up of system time on a Asus AP1700-S5 server with Linux-2.4.21-rc6-ac1.
Simple proof: a "sleep 300" command terminates after exactly 120 seconds of wall clock time.
Just as a wild shot in the dark, what speed does the kernel think the cpu is running at and does this match what the BIOS thinks?

Quoted 7 lines

Quoted 3 lines
It sounds like the CLOCK_TICK_RATE is wrong. This would show up as the kernel thinking the cpu was fast also.
Hm, I don't think this is the case.
BIOS reports (correctly) a single, hyperthreaded Xeon CPU with 2.4GHz

Ok, that means that the PIT is doing the right thing. Since the clock is only advanced on PIT interrupts and it is doing the right thing, it follows that you are servicing the interrupt more than once. This leads one to suspect the hyperthread stuff. We need someone who understands APIC/ hyperthread issues.

-g

Quoted 92 lines

Kernel reports the same:
root@setup:~ {503} $ uname -a
Linux setup 2.4.21-rc6-ac1 #2 SMP Tue Jun 3 09:45:13 CEST 2003 i686 unknown
root@setup:~ {504} $ cat /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 15
model : 2
model name : Intel(R) Xeon(TM) CPU 2.40GHz
stepping : 7
cpu MHz : 2392.065
cache size : 512 KB
physical id : 0
siblings : 2
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 2
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm
bogomips : 4771.02
processor : 1
vendor_id : GenuineIntel
cpu family : 15
model : 2
model name : Intel(R) Xeon(TM) CPU 2.40GHz
stepping : 7
cpu MHz : 2392.065
cache size : 512 KB
physical id : 0
siblings : 2
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 2
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm
bogomips : 4771.02
root@setup:~ {505} $ ntpdate ntp.xss.co.at; sleep 500; ntpdate ntp.xss.co.at
3 Jun 11:58:16 ntpdate[1118]: step time server 194.152.162.17 offset -177.268071 sec 3 Jun 12:01:36 ntpdate[1120]: step time server 194.152.162.17 offset -300.106769 sec
(Sleeping 500 "system seconds" takes 200 "wall clock seconds")
Quoted 3 lines
You pin this to a particular kernel version. Do other kernel versions do a better job?
So far I tried with:
2.4.21-rc2-ac2 (ACPI compiled as module)
2.4.21-rc4 (ACPI compiled as module)
2.4.21-rc6-ac1 (ACPI compiled as module)
2.4.21-rc6-ac1 (no ACPI comiled at all)
Time acceleration is the same for all kernels.
I'll try it with some older kernels, too.
- - andreas
- --
Andreas Haumer                     | mailto:[email protected]
*x Software + Systeme              | http://www.xss.co.at/
Karmarschgasse 51/2/20             | Tel: +43-1-6060114-0
A-1100 Vienna, Austria             | Fax: +43-1-6060114-71
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQE+3HKxxJmyeGcXPhERAkZ/AJ9tjgI6K2HtM/tL15nKDFIKzdcvQgCfYOGY i96/o6Nyxlem5yOp1A2Q//s=
=plFh
-----END PGP SIGNATURE-----
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Signature · 3 lines

George Anzinger   [email protected]
High-res-timers:  http://sourceforge.net/projects/high-res-timers/
Preemption patch: http://www.kernel.org/pub/linux/kernel/people/rml

Herbert Xu

2003-06-04 09:34 UTC #20 lore raw

In reply to Andreas Haumer

Quoted 4 lines · Andreas Haumer

I have a quite strange phenomenon here: I see a ~2.5 times speed up of system time on a Asus AP1700-S5 server with Linux-2.4.21-rc6-ac1.

I see exactly the opposite. The time is ~2.5 times slower with 2.4.21-rc6.

This is actually in a VMware setup on a P4 machine. The same setup runs 2.4.20, 2.5.69 and 2.5.70 correctly.

I'm going to play with it right after I get IDE modules working properly.

Signature · 4 lines

Debian GNU/Linux 3.0 is out! ( http://www.debian.org/ )
Email:  Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Herbert Xu

2003-06-04 12:50 UTC #21 lore raw

Quoted 6 lines · Herbert Xu

I see exactly the opposite. The time is ~2.5 times slower with 2.4.21-rc6.
This is actually in a VMware setup on a P4 machine. The same setup runs 2.4.20, 2.5.69 and 2.5.70 correctly.

That was my stupidity by not connecting the RTC. 2.5 just copes with it much better so I didn't notice.

Signature · 4 lines

Debian GNU/Linux 3.0 is out! ( http://www.debian.org/ )
Email:  Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt