Skipping ttyS2??

Skipping ttyS2??

[Jeroen Akershoek]

Hello,

I've got a system with one serial port on the motherboard and four other
ports on two PCI slots (2 each)
The kernel detects these as ttyS0 (onboard), ttyS14/S15 (1st PCI card)
and ttyS47/S48 (2nd PCI card).
We've got some pretty old proprietary software that insists on using
ttyS0-3, so I thought I'd just symlink
the higher ports.

I've made a rules file with the following:

KERNEL="ttyS14", SYMLINK="ttyS1"
KERNEL="ttyS15", SYMLINK="ttyS2"
KERNEL="ttyS47", SYMLINK="ttyS3"
KERNEL="ttyS48", SYMLINK="ttyS4"

This works fine with one exception: ttyS2 isn't symlinked

crw-rw----  1 root dialout 4, 64 Jun 14 10:07 /dev/ttyS0
lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS1 -> ttyS14
crw-rw----  1 root dialout 4, 66 Jun 14 12:15 /dev/ttyS2
lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS3 -> ttyS47
lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS4 -> ttyS48

udevtest says that the rule matches and that it's creating the link.
Changing the order, changing the symlink, changing SYMLINK to NAME or
removing
the other rules have no effect either. It just seems that this device is
skipped or maybe
later overridden by the original 4,66 device (udevtest doesn't say
anything about it
though).

Anybody got ideas on this?

Regards,
  Jeroen


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r 
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: Skipping ttyS2??

[Kay Sievers]

On Tue, Jun 14, 2005 at 12:27:03PM +0200, Jeroen Akershoek wrote:
> Hello,
> 
> I've got a system with one serial port on the motherboard and four other
> ports on two PCI slots (2 each)
> The kernel detects these as ttyS0 (onboard), ttyS14/S15 (1st PCI card)
> and ttyS47/S48 (2nd PCI card).
> We've got some pretty old proprietary software that insists on using
> ttyS0-3, so I thought I'd just symlink
> the higher ports.
> 
> I've made a rules file with the following:
> 
> KERNEL="ttyS14", SYMLINK="ttyS1"
> KERNEL="ttyS15", SYMLINK="ttyS2"
> KERNEL="ttyS47", SYMLINK="ttyS3"
> KERNEL="ttyS48", SYMLINK="ttyS4"
> 
> This works fine with one exception: ttyS2 isn't symlinked
> 
> crw-rw----  1 root dialout 4, 64 Jun 14 10:07 /dev/ttyS0
> lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS1 -> ttyS14
> crw-rw----  1 root dialout 4, 66 Jun 14 12:15 /dev/ttyS2
> lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS3 -> ttyS47
> lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS4 -> ttyS48
> 
> udevtest says that the rule matches and that it's creating the link.
> Changing the order, changing the symlink, changing SYMLINK to NAME or
> removing
> the other rules have no effect either. It just seems that this device is
> skipped or maybe
> later overridden by the original 4,66 device (udevtest doesn't say
> anything about it though).

Does renaming the "original" nodes to something else prevent that
conflict?

Kay


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r 
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: Skipping ttyS2??

[Jeroen Akershoek]

Kay Sievers wrote:

>On Tue, Jun 14, 2005 at 12:27:03PM +0200, Jeroen Akershoek wrote:
>  
>
>>Hello,
>>
>>I've got a system with one serial port on the motherboard and four other
>>ports on two PCI slots (2 each)
>>The kernel detects these as ttyS0 (onboard), ttyS14/S15 (1st PCI card)
>>and ttyS47/S48 (2nd PCI card).
>>We've got some pretty old proprietary software that insists on using
>>ttyS0-3, so I thought I'd just symlink
>>the higher ports.
>>
>>I've made a rules file with the following:
>>
>>KERNEL="ttyS14", SYMLINK="ttyS1"
>>KERNEL="ttyS15", SYMLINK="ttyS2"
>>KERNEL="ttyS47", SYMLINK="ttyS3"
>>KERNEL="ttyS48", SYMLINK="ttyS4"
>>
>>This works fine with one exception: ttyS2 isn't symlinked
>>
>>crw-rw----  1 root dialout 4, 64 Jun 14 10:07 /dev/ttyS0
>>lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS1 -> ttyS14
>>crw-rw----  1 root dialout 4, 66 Jun 14 12:15 /dev/ttyS2
>>lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS3 -> ttyS47
>>lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS4 -> ttyS48
>>
>>udevtest says that the rule matches and that it's creating the link.
>>Changing the order, changing the symlink, changing SYMLINK to NAME or
>>removing
>>the other rules have no effect either. It just seems that this device is
>>skipped or maybe
>>later overridden by the original 4,66 device (udevtest doesn't say
>>anything about it though).
>>    
>>
>
>Does renaming the "original" nodes to something else prevent that
>conflict?
>
>  
>
Nope. I've even tried removing them entirely. It gets recreated somehow
by restarting
udev, but I can't figure out with what rule. I've also tried commented
out all the default
rules referencing ttyS*, but they still get recreated.

Regards
  Jeroen


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r 
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: Skipping ttyS2??

[Jeroen Akershoek]

Kay Sievers wrote:

>On Tue, Jun 14, 2005 at 12:27:03PM +0200, Jeroen Akershoek wrote:
>  
>
>>Hello,
>>
>>I've got a system with one serial port on the motherboard and four other
>>ports on two PCI slots (2 each)
>>The kernel detects these as ttyS0 (onboard), ttyS14/S15 (1st PCI card)
>>and ttyS47/S48 (2nd PCI card).
>>We've got some pretty old proprietary software that insists on using
>>ttyS0-3, so I thought I'd just symlink
>>the higher ports.
>>
>>I've made a rules file with the following:
>>
>>KERNEL="ttyS14", SYMLINK="ttyS1"
>>KERNEL="ttyS15", SYMLINK="ttyS2"
>>KERNEL="ttyS47", SYMLINK="ttyS3"
>>KERNEL="ttyS48", SYMLINK="ttyS4"
>>
>>This works fine with one exception: ttyS2 isn't symlinked
>>
>>crw-rw----  1 root dialout 4, 64 Jun 14 10:07 /dev/ttyS0
>>lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS1 -> ttyS14
>>crw-rw----  1 root dialout 4, 66 Jun 14 12:15 /dev/ttyS2
>>lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS3 -> ttyS47
>>lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS4 -> ttyS48
>>
>>udevtest says that the rule matches and that it's creating the link.
>>Changing the order, changing the symlink, changing SYMLINK to NAME or
>>removing
>>the other rules have no effect either. It just seems that this device is
>>skipped or maybe
>>later overridden by the original 4,66 device (udevtest doesn't say
>>anything about it though).
>>    
>>
>
>Does renaming the "original" nodes to something else prevent that
>conflict?
>  
>
Uhm... ok, forget my last reply :)
doing something like:
KERNEL="ttyS2", NAME="ttybackup2"
actually works yes

Why it does that I have no idea though. And why only with ttyS2?

Jeroen


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r 
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel

Re: Skipping ttyS2??

[Kay Sievers]

On Tue, Jun 14, 2005 at 02:37:39PM +0200, Jeroen Akershoek wrote:
> Kay Sievers wrote:
> 
> >On Tue, Jun 14, 2005 at 12:27:03PM +0200, Jeroen Akershoek wrote:
> >  
> >
> >>Hello,
> >>
> >>I've got a system with one serial port on the motherboard and four other
> >>ports on two PCI slots (2 each)
> >>The kernel detects these as ttyS0 (onboard), ttyS14/S15 (1st PCI card)
> >>and ttyS47/S48 (2nd PCI card).
> >>We've got some pretty old proprietary software that insists on using
> >>ttyS0-3, so I thought I'd just symlink
> >>the higher ports.
> >>
> >>I've made a rules file with the following:
> >>
> >>KERNEL="ttyS14", SYMLINK="ttyS1"
> >>KERNEL="ttyS15", SYMLINK="ttyS2"
> >>KERNEL="ttyS47", SYMLINK="ttyS3"
> >>KERNEL="ttyS48", SYMLINK="ttyS4"
> >>
> >>This works fine with one exception: ttyS2 isn't symlinked
> >>
> >>crw-rw----  1 root dialout 4, 64 Jun 14 10:07 /dev/ttyS0
> >>lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS1 -> ttyS14
> >>crw-rw----  1 root dialout 4, 66 Jun 14 12:15 /dev/ttyS2
> >>lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS3 -> ttyS47
> >>lrwxrwxrwx  1 root root        6 Jun 14 12:15 /dev/ttyS4 -> ttyS48
> >>
> >>udevtest says that the rule matches and that it's creating the link.
> >>Changing the order, changing the symlink, changing SYMLINK to NAME or
> >>removing
> >>the other rules have no effect either. It just seems that this device is
> >>skipped or maybe
> >>later overridden by the original 4,66 device (udevtest doesn't say
> >>anything about it though).
> >>    
> >>
> >
> >Does renaming the "original" nodes to something else prevent that
> >conflict?
> >
> >  
> >
> Nope. I've even tried removing them entirely. It gets recreated somehow
> by restarting
> udev, but I can't figure out with what rule. I've also tried commented
> out all the default
> rules referencing ttyS*, but they still get recreated.

_Rename_ the devices with an explicit rule! Commenting out does not help.

Kay


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r 
_______________________________________________
Linux-hotplug-devel mailing list  http://linux-hotplug.sourceforge.net
Linux-hotplug-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-hotplug-devel