Hello,
I have updated my kernel to 2.6.7-bk2 this morning and it felt faster
than 2.6.7! It feels faster in firefox and it is noticeably faster in
gnome-terminal. Goodness!
But, there is downside as well: the clock runs faster. 125-150% faster ;(
It is weird since the time server does not correct it.
Can I get the speed with a correct clock? Please?
florin
--
Don't question authority: they don't know either!
El S?bado 19 Junio 2004 23:07, Florin Iucha escribi?:
> Hello,
>
> I have updated my kernel to 2.6.7-bk2 this morning and it felt faster
> than 2.6.7! It feels faster in firefox and it is noticeably faster in
> gnome-terminal. Goodness!
>
> But, there is downside as well: the clock runs faster. 125-150% faster ;(
> It is weird since the time server does not correct it.
>
> Can I get the speed with a correct clock? Please?
>
> florin
Are you sure that Firefox is faster with 2.6.7? I haven't noticed it yet...
Cheers
Manuel.
--
Manuel Arostegui Ramirez #Linux Registered User 200896
On Saturday 19 June 2004 23:07, Florin Iucha wrote:
> Hello,
>
> I have updated my kernel to 2.6.7-bk2 this morning and it felt faster
> than 2.6.7! It feels faster in firefox and it is noticeably faster in
> gnome-terminal. Goodness!
>
> But, there is downside as well: the clock runs faster. 125-150% faster ;(
> It is weird since the time server does not correct it.
>
> Can I get the speed with a correct clock? Please?
>
> florin
Yes, same here! I had to decrease the speed of the keyboard in KDE config,
because I couldn't type a character without that the character was printed
hundred times.
The jumping icon on startup of several KDE apps, is now jumping much faster
too and apps are started much faster!
I'm working on a Intel(R) Pentium(R) 4 CPU.
greets dominik
Dominik Karall <[email protected]> wrote:
>
> On Saturday 19 June 2004 23:07, Florin Iucha wrote:
> > Hello,
> >
> > I have updated my kernel to 2.6.7-bk2 this morning and it felt faster
> > than 2.6.7! It feels faster in firefox and it is noticeably faster in
> > gnome-terminal. Goodness!
> >
> > But, there is downside as well: the clock runs faster. 125-150% faster ;(
> > It is weird since the time server does not correct it.
> >
> > Can I get the speed with a correct clock? Please?
> >
> > florin
>
> Yes, same here!
dmesg and .config, please.
On Sunday 20 June 2004 01:05, Andrew Morton wrote:
> Dominik Karall <[email protected]> wrote:
> > On Saturday 19 June 2004 23:07, Florin Iucha wrote:
> > > Hello,
> > >
> > > I have updated my kernel to 2.6.7-bk2 this morning and it felt faster
> > > than 2.6.7! It feels faster in firefox and it is noticeably faster in
> > > gnome-terminal. Goodness!
> > >
> > > But, there is downside as well: the clock runs faster. 125-150% faster
> > > ;( It is weird since the time server does not correct it.
> > >
> > > Can I get the speed with a correct clock? Please?
> > >
> > > florin
> >
> > Yes, same here!
>
> dmesg and .config, please.
attached
cpuinfo:
processor : 0
vendor_id : GenuineIntel
cpu family : 15
model : 2
model name : Intel(R) Pentium(R) 4 CPU 2.66GHz
stepping : 7
cpu MHz : 2672.862
cache size : 512 KB
physical id : 0
siblings : 1
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 pbe cid
bogomips : 5275.64
greets dominik
Dominik Karall <[email protected]> wrote:
>
> > dmesg and .config, please.
> attached
The dmesg output is incomplete. You'll need to use `dmesg -s 1000000' to
get all of it.
I wish that would get fixed actually. Seems a bit silly, and current
kernels permit querying of the ringbuffer size.
On Sunday 20 June 2004 01:23, Andrew Morton wrote:
> Dominik Karall <[email protected]> wrote:
> > > dmesg and .config, please.
> >
> > attached
>
> The dmesg output is incomplete. You'll need to use `dmesg -s 1000000' to
> get all of it.
>
> I wish that would get fixed actually. Seems a bit silly, and current
> kernels permit querying of the ringbuffer size.
'dmesg -s 1000000' output attached.
greets dominik
On Sun, 20 Jun 2004, Dominik Karall wrote:
> On Sunday 20 June 2004 01:23, Andrew Morton wrote:
> > Dominik Karall <[email protected]> wrote:
> > > > dmesg and .config, please.
> > >
> > > attached
> >
> > The dmesg output is incomplete. You'll need to use `dmesg -s 1000000' to
> > get all of it.
> >
> > I wish that would get fixed actually. Seems a bit silly, and current
> > kernels permit querying of the ringbuffer size.
> 'dmesg -s 1000000' output attached.
Try booting with 'acpi=off'
IRQ to pin mappings:
IRQ0 -> 0:0-> 0:2
On Sunday 20 June 2004 02:30, Zwane Mwaikambo wrote:
> On Sun, 20 Jun 2004, Dominik Karall wrote:
> > On Sunday 20 June 2004 01:23, Andrew Morton wrote:
> > > Dominik Karall <[email protected]> wrote:
> > > > > dmesg and .config, please.
> > > >
> > > > attached
> > >
> > > The dmesg output is incomplete. You'll need to use `dmesg -s 1000000'
> > > to get all of it.
> > >
> > > I wish that would get fixed actually. Seems a bit silly, and current
> > > kernels permit querying of the ringbuffer size.
> >
> > 'dmesg -s 1000000' output attached.
>
> Try booting with 'acpi=off'
>
> IRQ to pin mappings:
> IRQ0 -> 0:0-> 0:2
After booting with 'acpi=off' it works as normal.
dmesg
...
IRQ to pin mappings:
IRQ0 -> 0:2
....
Let me know if you need more information.
greets dominik
Hi,
I have the same problem.
Everything seems to be double speed. The keys, the little moving icon in the
top right off the konqueror window and the cursor is blinking like mad.
glxgears fps are halved, like the fps in unreal tournament 2004 demo.
There everything was incredidbly fast, but fps halved.
testgart showed only half of the ususal throughput.
Oh, and between a boot, recompiling some modules and reboot I lost over 500
secs.
Attached are dmesg output and config.
Gl?ck Auf,
Volker
--
Conclusions
In a straight-up fight, the Empire squashes the Federation like a bug. Even
with its numerical advantage removed, the Empire would still squash the
Federation like a bug. Accept it. -Michael Wong
On Sat, Jun 19, 2004 at 04:05:32PM -0700, Andrew Morton wrote:
> dmesg and .config, please.
Attached.
florin
--
Don't question authority: they don't know either!
"Hemmann, Volker Armin" <[email protected]> said on Sun, 20 Jun 2004 04:31:44 +0200:
> Hi,
>
> I have the same problem.
>
> Everything seems to be double speed. The keys, the little moving icon in th=
> e=20
> top right off the konqueror window and the cursor is blinking like mad.=20
> glxgears fps are halved, like the fps in unreal tournament 2004 demo.
>
> There everything was incredidbly fast, but fps halved.
You realise that FPS has the "per second" part as measured by the
computer, right? And what will happen if the computer thinks a second
takes less time?
Of course FPS will go down, but the real Frames per Human-Second
probably hasn't changed at all.
--
TimC -- http://astronomy.swin.edu.au/staff/tconnors/
If I'd known computer science was going to be like this, I'd never have
given up being a rock 'n' roll star. -- G. Hirst
Hi List,
I traced the current double-speed issue for 2.6.7-bk2 on x86-64 back to an
ACPI-change in mpparse.c. The small attached patch solved the issue at least
on my MSI K8T Neo (Athlon 64 3200+) system.
--- linux-2.6.7-bk2.orig/arch/x86_64/kernel/mpparse.c 2004-06-19
17:17:15.824315000 +0200
+++ linux-2.6.7-bk2/arch/x86_64/kernel/mpparse.c 2004-06-20 16:02:34.974446912
+0200
@@ -861,7 +861,7 @@
for (idx = 0; idx < mp_irq_entries; idx++)
if (mp_irqs[idx].mpc_srcbus == MP_ISA_BUS &&
- (mp_irqs[idx].mpc_dstapic == ioapic) &&
+ (mp_irqs[idx].mpc_dstapic == intsrc.mpc_dstapic) &&
(mp_irqs[idx].mpc_srcbusirq == i ||
mp_irqs[idx].mpc_dstirq == i))
break;
--
--
Hans-Frieder Vogt e-mail: hfvogt (at) arcor (dot) de
On Sat, Jun 19, 2004 at 04:23:22PM -0700, Andrew Morton wrote:
> The dmesg output is incomplete. You'll need to use `dmesg -s 1000000' to
> get all of it.
>
> I wish that would get fixed actually. Seems a bit silly, and current
> kernels permit querying of the ringbuffer size.
Hi Andrew - not so impatient.
Of course I fixed dmesg before I fixed the kernel.
From: Matthew Gregan <[email protected]>
: Hi Andries,
:
: Attached is a small patch to dmesg that adds support for the new
: functionality in kernel 2.6.6 and upwards to read the size of the kernel
: message ring buffer.
Thanks!
However, dmesg already has this code.
Andries