Project Home
Project Home		 Wiki
Wiki		 Discussion Forums
Discussions		 Project Information
Project Info
QNX RTOS v4 [EOL]/Discussion/General/second Fsys.atapi on same int /List of Posts
Forum Topic - second Fsys.atapi on same int : (5 Items)
   
Update
Klaus Pantleon(deleted)
07/06/2010 9:33 AM
post58667
second Fsys.atapi on same int   
Hello,

I'm actually trying to bring QNX4 on a Dell D6510 because these Laptop is an actual one with pccard port. Ans I need the
 pccard port to write on PCMCIA flash drives.

Therefore I start a second Fsys.atapi with the settings I get from pc-pin and the device will be atached. But as soon as
 I want to access the new device (e.g. fdisk /dev/pcmcia0) the Fsys.atapi from the real harddisk is consuming the most 
CPU usage. The command itself will work but only very very slow reaction.
Both Fsys.atapi are running purely on the same interrupt  and I have no chance to change it.
Is there a possibility to tell the Fsys.atapi that he should waste no time if there is an interrupt coming, but this int
 is not for him? There is another Fsys.atapi running for the accessed device.

The rest of the laptop is usable with QNX4 (net, graphic, usb, ...).
Oleg Bolshakov
07/09/2010 8:29 AM
post59053
Re: second Fsys.atapi on same int   
Hi Klaus,

Please provide us a bit more technical information:

sin ar
sin ver
sin ti
sin ir
sin pr
show_pci -vvv
pc-pin

Also try to start single Fsys.atapi driver for all controllers.

-- 
Respectfully,
Oleg

> Hello,
> 
> I'm actually trying to bring QNX4 on a Dell D6510 because these Laptop
> is an actual one with pccard port. Ans I need the pccard port to write
> on PCMCIA flash drives.
> 
> Therefore I start a second Fsys.atapi with the settings I get from
> pc-pin and the device will be atached. But as soon as I want to access
> the new device (e.g. fdisk /dev/pcmcia0) the Fsys.atapi from the real
> harddisk is consuming the most CPU usage. The command itself will work
> but only very very slow reaction. Both Fsys.atapi are running purely
> on the same interrupt  and I have no chance to change it. Is there a
> possibility to tell the Fsys.atapi that he should waste no time if
> there is an interrupt coming, but this int is not for him? There is
> another Fsys.atapi running for the accessed device.
> 
> The rest of the laptop is usable with QNX4 (net, graphic, usb, ...).
> 
> 
> 
> _______________________________________________
> 
> General
> http://community.qnx.com/sf/go/post58667
Klaus Pantleon(deleted)
07/12/2010 8:05 AM
post59188
Re: second Fsys.atapi on same int   
Hi Oleg,

I made a test to extract the problem as most clear as I can:
1) Boot over the network
2) start Fsys.atapi for cd
3) start Fsys.atapi for hd
4) checked speed for cd and hd
5) stopped Fsys.atapi
6) Start Fsys.atapi for cd and hd
7) checke speed for cd and hd

This test I made for the Dell and for a VMWare and attached the terminal aoutput in the following. Also I made the "sin 
ar" for both environment before starting the testprocedure. You will see that theses effect is not reproducable in 
VMWare. Please tell me at wht time in teh testprocedure you want the "sin ??" calls.

sin ar for Dell:

  PID USER NAME    ARGUMENTS
    1 System       Proc32 -l 179 -p2000 -P26 
    2 System       Slib32 
    4 System       Net -n500 -m 2 1 000C29 1EEDBD 
    5 System       Net.e1000 -I0 -l1 
    7 System       Not available.                  
  491 System       Pipe 
  509 System       Dev -n 100 
  527 System       Dev.ansi -Q -n 4 
  626 System       tinit -t /dev/con1 /dev/con2 /dev/con3 /dev/con4 /dev/config 
  636 System       /bin/nameloc 
  639 System       /bin/nameloc 
 1306 System       /bin/login -p 
 1321 System       /bin/login -p 
 1355 System       /bin/login -p 
 1381 System       /bin/login -p 
 5051 root         -sh 
 5781 root         sin ar 

sin ar for VMWare:

  PID USER NAME    ARGUMENTS
    1 System       Proc32 -l 182 -p2000 -P26 
    2 System       Slib32 
    4 System       Net -n500 -m 2 1 000C29 1EEDBD 
    5 System       Net.ether2100 -I0 -l1 
    7 System       Not available.                  
  489 System       Pipe 
  507 System       Dev -n 100 
  525 System       Dev.ansi -Q -n 4 
  541 System       Dev.ser -b38400 -t14 
  636 System       tinit -t /dev/con1 /dev/con2 /dev/con3 /dev/con4 /dev/config 
  646 System       /bin/nameloc 
  649 System       /bin/nameloc 
 1308 System       /bin/login -p 
 1333 System       /bin/login -p 
 1357 System       /bin/login -p 
 1390 System       /bin/login -p 
 2335 root         -sh 
 2826 root         sin ar 

Testprocedure Dell:

(root)//179/root>Fsys &
%1 +  7359 Running              Fsys
(root)//179/root>
(root)//179/root>Fsys.atapi fsys -Ncd -n0=cd atapi ioport=0x7090 &
%2 +  7582 Running              Fsys.atapi fsys -Ncd -n0=cd atapi ioport=0x7090
(root)//179/root>Path=0 - Generic IDE
 target=0 lun=0 CD-ROM(5) - MATSHITA DVD+-RW UJ892    Rev: 1.04

(root)//179/root>Fsys.atapi fsys -Nhd -n0=hd atapi ioport=0x70f0,irq=10 &
%3 +  7957 Running              Fsys.atapi fsys -Nhd -n0=hd atapi ioport=0x70f0
(root)//179/root>Path=0 - Generic IDE
 target=0 lun=0 Direct-Access(0) -          WDC WD2500BEKT-7 Rev: 01.0

(root)//179/root>cp -V /dev/cd0 /dev/null
cp: Copying /dev/cd0 to /dev/null
 20.70% (36160/174622 kbytes, 904 kb/s)

(root)//179/root>cp -V /dev/hd0 /dev/null
cp: Copying /dev/hd0 to /dev/null
 00.00% (6048/24419858 kbytes, 1745 kb/s)

(root)//179/root>rm /dev/cd0 /dev/hd0
%2 -  7582 Done                 Fsys.atapi fsys -Ncd -n0=cd atapi ioport=0x7090
%3    7957 Done                 Fsys.atapi fsys -Nhd -n0=hd atapi ioport=0x70f0
(root)//179/root>Fsys.atapi atapi ioport=0x70f0,irq=10 atapi ioport=0x7090 &
%2 + 12229 Running              Fsys.atapi atapi ioport=0x70f0,irq=10 atapi iop
(root)//179/root>Path=0 - Generic IDE
 target=0 lun=0 Direct-Access(0) -          WDC WD2500BEKT-7 Rev: 01.0
Path=1 - Generic IDE
 target=0 lun=0 CD-ROM(5) - MATSHITA DVD+-RW UJ892    Rev: 1.04

(root)//179/root>cp -V /dev/cd0 /dev/null
cp: Copying /dev/cd0 to /dev/null
 20.88% (36464/174622 kbytes, 1469 kb/s)

(root)//179/root>cp -V /dev/hd0 /dev/null
cp: Copying /dev/hd0 to /dev/null
 00.00% (6784/24419858 kbytes, 13789 kb/s)

(root)//179/root>


Tesprocedure VMWare:

(root)//182/root>
(root)//182/root>Fsys.atapi fsys -Ncd -n0=cd atapi ioport=0x170,irq=15 &
%2 + 11760 Running              Fsys.atapi fsys -Ncd -n0=cd...
View Full Message
Oleg Bolshakov
08/02/2010 6:37 AM
post61328
Re: second Fsys.atapi on same int   
Hi Klaus,

Sorry for delay with answer. I need output of

sin ar
sin ver
sin ti
sin ir
sin pr
show_pci -vvv
pc-pin

when your system works slow. Also I can't find the Dell D6510 specification, 
so post it (or link to it) here. May be using a USB stick instead of 
PCMCIA flash drive is a better idea.

-- 
Respectfully,
Oleg

> Hi Oleg,
> 
> I made a test to extract the problem as most clear as I can:
> 1) Boot over the network
> 2) start Fsys.atapi for cd
> 3) start Fsys.atapi for hd
> 4) checked speed for cd and hd
> 5) stopped Fsys.atapi
> 6) Start Fsys.atapi for cd and hd
> 7) checke speed for cd and hd
> 
> This test I made for the Dell and for a VMWare and attached the
> terminal aoutput in the following. Also I made the "sin ar" for both
> environment before starting the testprocedure. You will see that
> theses effect is not reproducable in VMWare. Please tell me at wht
> time in teh testprocedure you want the "sin ??" calls.
> 
> sin ar for Dell:
> 
>   PID USER NAME    ARGUMENTS
>     1 System       Proc32 -l 179 -p2000 -P26
>     2 System       Slib32
>     4 System       Net -n500 -m 2 1 000C29 1EEDBD
>     5 System       Net.e1000 -I0 -l1
>     7 System       Not available.
>   491 System       Pipe
>   509 System       Dev -n 100
>   527 System       Dev.ansi -Q -n 4
>   626 System       tinit -t /dev/con1 /dev/con2 /dev/con3 /dev/con4
> /dev/config 636 System       /bin/nameloc
>   639 System       /bin/nameloc
>  1306 System       /bin/login -p
>  1321 System       /bin/login -p
>  1355 System       /bin/login -p
>  1381 System       /bin/login -p
>  5051 root         -sh
>  5781 root         sin ar
> 
> sin ar for VMWare:
> 
>   PID USER NAME    ARGUMENTS
>     1 System       Proc32 -l 182 -p2000 -P26
>     2 System       Slib32
>     4 System       Net -n500 -m 2 1 000C29 1EEDBD
>     5 System       Net.ether2100 -I0 -l1
>     7 System       Not available.
>   489 System       Pipe
>   507 System       Dev -n 100
>   525 System       Dev.ansi -Q -n 4
>   541 System       Dev.ser -b38400 -t14
>   636 System       tinit -t /dev/con1 /dev/con2 /dev/con3 /dev/con4
> /dev/config 646 System       /bin/nameloc
>   649 System       /bin/nameloc
>  1308 System       /bin/login -p
>  1333 System       /bin/login -p
>  1357 System       /bin/login -p
>  1390 System       /bin/login -p
>  2335 root         -sh
>  2826 root         sin ar
> 
> Testprocedure Dell:
> 
> (root)//179/root>Fsys &
> %1 +  7359 Running              Fsys
> (root)//179/root>
> (root)//179/root>Fsys.atapi fsys -Ncd -n0=cd atapi ioport=0x7090 &
> %2 +  7582 Running              Fsys.atapi fsys -
Ncd -n0=cd atapi
> ioport=0x7090 (root)//179/root>Path=0 - Generic IDE
>  target=0 lun=0 CD-ROM(5) - MATSHITA DVD+-RW UJ892    Rev: 1.04
> 
> (root)//179/root>Fsys.atapi fsys -Nhd -n0=hd atapi ioport=0x70f0,irq=10
> & %3 +  7957 Running              Fsys.atapi fsys -Nhd -n0=hd atapi
> ioport=0x70f0 (root)//179/root>Path=0 - Generic IDE
>  target=0 lun=0 Direct-Access(0) -          WDC WD2500BEKT-7 Rev: 01.0
> 
> (root)//179/root>cp -V /dev/cd0 /dev/null
> cp: Copying /dev/cd0 to /dev/null
>  20.70% (36160/174622 kbytes, 904 kb/s)
> 
> (root)//179/root>cp -V /dev/hd0 /dev/null
> cp: Copying /dev/hd0 to /dev/null
>  00.00% (6048/24419858 kbytes, 1745 kb/s)
> 
> (root)//179/root>rm /dev/cd0 /dev/hd0
> %2 -  7582 Done                 Fsys.atapi fsys -Ncd -n0=cd atapi
> ioport=0x7090 %3    7957 Done                 Fsys.atapi fsys -Nhd
> -n0=hd atapi ioport=0x70f0 (root)//179/root>Fsys.atapi atapi
> ioport=0x70f0,irq=10 atapi ioport=0x7090 & %2 + 12229 Running        ...
View Full Message
Klaus Pantleon(deleted)
08/02/2010 8:42 AM
post61339
RE: second Fsys.atapi on same int   
Hi Oleg,

sorry but I dont have the hardware any more on my desktop.
As soon as I have the next Dell 6510 I will give you the "sin ??" info. 
The proposal with an USB2PCMCIA adapter I already tested and it is much much faster.
But I want to offer the service persons of my company a solution without any add on hardware.
These add on hardware I cannot control. The vandor and type of laptop is good to control.



Freundliche Grüße / Best regards

Klaus Pantleon
Softwarestandardisierung, Hardwarestandardisierung, Fernwartungslösungen
_________________________________________________

Voith Paper Automation GmbH & Co. KG
- prf
St. Pöltener Straße 43
89522 Heidenheim, Germany

Tel  +49 7321 37 2137
Fax  +49 7321 37 6109
Mobile  +49 172 7394158

Klaus.Pantleon@Voith.com
http://www.voithpaper.com
http://www.fieldinstruments.voithpaper.com
_________________________________________________

Handelsregister Ulm, HRA 661188
Sitz der Gesellschaft: Heidenheim
Geschäftsführung:
Dr. Antti Kaunonen (Vorsitzender), Ingolf Cedra, Alexander Kienle
Persönlich haftende Gesellschafterin: Voith Paper Automation Verwaltungs GmbH, Reg. Gericht Ulm, HRB 661292 

-----Original Message-----
From: Oleg Bolshakov [mailto:community-noreply@qnx.com] 
Sent: Monday, August 02, 2010 12:37 PM
To: general-qnx4
Subject: Re: second Fsys.atapi on same int

Hi Klaus,

Sorry for delay with answer. I need output of

sin ar
sin ver
sin ti
sin ir
sin pr
show_pci -vvv
pc-pin

when your system works slow. Also I can't find the Dell D6510 specification, so post it (or link to it) here. May be using a USB stick instead of PCMCIA flash drive is a better idea.

--
Respectfully,
Oleg

> Hi Oleg,
> 
> I made a test to extract the problem as most clear as I can:
> 1) Boot over the network
> 2) start Fsys.atapi for cd
> 3) start Fsys.atapi for hd
> 4) checked speed for cd and hd
> 5) stopped Fsys.atapi
> 6) Start Fsys.atapi for cd and hd
> 7) checke speed for cd and hd
> 
> This test I made for the Dell and for a VMWare and attached the 
> terminal aoutput in the following. Also I made the "sin ar" for both 
> environment before starting the testprocedure. You will see that 
> theses effect is not reproducable in VMWare. Please tell me at wht 
> time in teh testprocedure you want the "sin ??" calls.
> 
> sin ar for Dell:
> 
>   PID USER NAME    ARGUMENTS
>     1 System       Proc32 -l 179 -p2000 -P26
>     2 System       Slib32
>     4 System       Net -n500 -m 2 1 000C29 1EEDBD
>     5 System       Net.e1000 -I0 -l1
>     7 System       Not available.
>   491 System       Pipe
>   509 System       Dev -n 100
>   527 System       Dev.ansi -Q -n 4
>   626 System       tinit -t /dev/con1 /dev/con2 /dev/con3 /dev/con4
> /dev/config 636 System       /bin/nameloc
>   639 System       /bin/nameloc
>  1306 System       /bin/login -p
>  1321 System       /bin/login -p
>  1355 System       /bin/login -p
>  1381 System       /bin/login -p
>  5051 root         -sh
>  5781 root         sin ar
> 
> sin ar for VMWare:
> 
>   PID USER NAME    ARGUMENTS
>     1 System       Proc32 -l 182 -p2000 -P26
>     2 System       Slib32
>     4 System       Net -n500 -m 2 1 000C29 1EEDBD
>     5 System       Net.ether2100 -I0 -l1
>     7 System       Not available.
>   489 System       Pipe
>   507 System       Dev -n 100
>   525 System       Dev.ansi -Q -n 4
>   541 System       Dev.ser -b38400 -t14
>   636 System       tinit -t /dev/con1 /dev/con2 /dev/con3 /dev/con4
> /dev/config 646 System       /bin/nameloc
>   649 System       /bin/nameloc
>  1308 System       /bin/login -p
>  1333 System       /bin/login -p
>  1357 System       /bin/login -p
>  1390 System       /bin/login -p
>  2335 root         -sh
>  2826...
View Full Message

Return


  A subsidiary of BlackBerry             All content ©2004-2013, QNX Software Systems