Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.sys.acorn.misc > #4553 > unrolled thread

ROM modules

Started byDave Symes <dave@triffid.co.uk>
First post2012-03-23 08:04 +0000
Last post2012-03-27 18:21 +0100
Articles 5 — 3 participants

Back to article view | Back to comp.sys.acorn.misc

Contents

  ROM modules Dave Symes <dave@triffid.co.uk> - 2012-03-23 08:04 +0000
    Re: ROM modules Martin Wuerthner <spamtrap@mw-software.com> - 2012-03-23 10:50 +0100
      Re: ROM modules Dave Symes <dave@triffid.co.uk> - 2012-03-23 12:10 +0000
        Re: ROM modules Martin Wuerthner <spamtrap@mw-software.com> - 2012-03-23 16:23 +0100
          Re: ROM modules "Ste (news)" <steve@revi11.plus.com> - 2012-03-27 18:21 +0100

#4553 — ROM modules

FromDave Symes <dave@triffid.co.uk>
Date2012-03-23 08:04 +0000
SubjectROM modules
Message-ID<5274db99dcdave@triffid.co.uk>
SARPC RO 6.20
Some thoughts, advice please.

I've recently done a Sys Merge to update some modules to work with the
freeware versions of Fireworks.
The merge appears to have gone ok. 
But...

FPEmulator 4.22 is in ROM and should have been made dormant and replaced
with a soft load FPEmulator 4.28, but it has not.

Any attempt to Kill or Unplug FPE 4.22 results in the machine failing as
it auto processes through and ends all background system stuff, requiring
a back button reset.

FWIW: As far as possible, before attempting... All user background and
foreground running apps have been quit, leaving only Sys stuff running
background.

A lot of Trap handler and undefined instruction error messages arrive,
I've pasted in a few below.

*********************************
22 Mar 21:43:58 100   Message  : Internal error, trap while in trap
handler: Internal error: Undefined instruction at &038598D8, pc =
238598D8: registers at 0002968C
22 Mar 21:43:58 100 
22 Mar 21:44:07 100 ***Error***
22 Mar 21:44:07 100   Title    : Error
22 Mar 21:44:08 100   Task     : Display Manager
22 Mar 21:44:08 100   Message  : Internal error: Undefined instruction at
&03BEA450
22 Mar 21:44:08 100 
22 Mar 21:44:11 100 ***Error***
22 Mar 21:44:11 100   Title    : Error
22 Mar 21:44:11 100   Task     : Task Manager
22 Mar 21:44:11 100   Message  : Internal error: Undefined instruction at
&03BEA450
22 Mar 21:44:11 100 
22 Mar 21:44:12 100 ***Error***
22 Mar 21:44:12 100   Title    : Error from Filer
22 Mar 21:44:12 100   Task     : Filer
22 Mar 21:44:12 100   Message  : Internal error: Undefined instruction at
&03BEA450
22 Mar 21:44:12 100 
22 Mar 22:06:25 100 ***Error***
22 Mar 22:06:25 100   Title    : Error
22 Mar 22:06:25 100   Task     : Task Manager
22 Mar 22:06:25 100   Message  : Internal error: Undefined instruction at
&03BEA45C
22 Mar 22:06:25 100 
22 Mar 22:06:30 100 ***Error***
22 Mar 22:06:30 100   Title    : Error
22 Mar 22:06:30 100   Task     : Reporter
22 Mar 22:06:30 100   Message  : Internal error: Undefined instruction at
&03BEA450

********************************************

This hardware SARPC has an identical backup on a VRPC-AdjustSA (Win XP)
and the same process was undertaken on that, and the merge went correctly,
making  the FPE 4.22 Dormant and activating the softload 4.28

Any thoughts please?
Thanks
Dave

-- 

Dave Triffid

[toc] | [next] | [standalone]

#4556

FromMartin Wuerthner <spamtrap@mw-software.com>
Date2012-03-23 10:50 +0100
Message-ID<a145e57452.martin@bach.planiverse.com>
In reply to#4553
In message <5274db99dcdave@triffid.co.uk>
          Dave Symes <dave@triffid.co.uk> wrote:

> SARPC RO 6.20
> Some thoughts, advice please.

> I've recently done a Sys Merge to update some modules to work with the
> freeware versions of Fireworks.
> The merge appears to have gone ok.
> But...

> FPEmulator 4.22 is in ROM and should have been made dormant and replaced
> with a soft load FPEmulator 4.28, but it has not.

Should it? What makes you think so? Did the documentation advise you 
that that should happen? Does the program ask for 4.28? Does the 
program not work?

When a module is updated in !System, then precisely nothing happens, 
unless an application wants that later version and asks for it.

In this particular case, it would probably be utterly wrong to replace 
your 4.22 ROM version by a softloaded 4.28, but that subtlety should 
not be your concern.

> Any attempt to Kill or Unplug FPE 4.22 results in the machine failing as
> it auto processes through and ends all background system stuff, requiring
> a back button reset.

Why would you want to Kill or Unplug FPE 4.22?

Martin

> FWIW: As far as possible, before attempting... All user background and
> foreground running apps have been quit, leaving only Sys stuff running
> background.

> A lot of Trap handler and undefined instruction error messages arrive,
> I've pasted in a few below.

> *********************************
> 22 Mar 21:43:58 100   Message  : Internal error, trap while in trap
> handler: Internal error: Undefined instruction at &038598D8, pc =
> 238598D8: registers at 0002968C
> 22 Mar 21:43:58 100
> 22 Mar 21:44:07 100 ***Error***
> 22 Mar 21:44:07 100   Title    : Error
> 22 Mar 21:44:08 100   Task     : Display Manager
> 22 Mar 21:44:08 100   Message  : Internal error: Undefined instruction at
> &03BEA450
> 22 Mar 21:44:08 100
> 22 Mar 21:44:11 100 ***Error***
> 22 Mar 21:44:11 100   Title    : Error
> 22 Mar 21:44:11 100   Task     : Task Manager
> 22 Mar 21:44:11 100   Message  : Internal error: Undefined instruction at
> &03BEA450
> 22 Mar 21:44:11 100
> 22 Mar 21:44:12 100 ***Error***
> 22 Mar 21:44:12 100   Title    : Error from Filer
> 22 Mar 21:44:12 100   Task     : Filer
> 22 Mar 21:44:12 100   Message  : Internal error: Undefined instruction at
> &03BEA450
> 22 Mar 21:44:12 100
> 22 Mar 22:06:25 100 ***Error***
> 22 Mar 22:06:25 100   Title    : Error
> 22 Mar 22:06:25 100   Task     : Task Manager
> 22 Mar 22:06:25 100   Message  : Internal error: Undefined instruction at
> &03BEA45C
> 22 Mar 22:06:25 100
> 22 Mar 22:06:30 100 ***Error***
> 22 Mar 22:06:30 100   Title    : Error
> 22 Mar 22:06:30 100   Task     : Reporter
> 22 Mar 22:06:30 100   Message  : Internal error: Undefined instruction at
> &03BEA450

> ********************************************

> This hardware SARPC has an identical backup on a VRPC-AdjustSA (Win XP)
> and the same process was undertaken on that, and the merge went correctly,
> making  the FPE 4.22 Dormant and activating the softload 4.28

> Any thoughts please?
> Thanks
> Dave

[toc] | [prev] | [next] | [standalone]

#4571

FromDave Symes <dave@triffid.co.uk>
Date2012-03-23 12:10 +0000
Message-ID<5274f22180dave@triffid.co.uk>
In reply to#4556
In article <a145e57452.martin@bach.planiverse.com>,
   Martin Wuerthner <spamtrap@mw-software.com> wrote:
> In message <5274db99dcdave@triffid.co.uk>
>           Dave Symes <dave@triffid.co.uk> wrote:

> > SARPC RO 6.20
> > Some thoughts, advice please.

> > I've recently done a Sys Merge to update some modules to work with the
> > freeware versions of Fireworks.
> > The merge appears to have gone ok.
> > But...

> > FPEmulator 4.22 is in ROM and should have been made dormant and
> > replaced with a soft load FPEmulator 4.28, but it has not.

> Should it? What makes you think so? Did the documentation advise you 
> that that should happen? Does the program ask for 4.28? Does the 
> program not work?

Documentation, don't get to see any of that stuff these days.
Yes, in the Run file it has a RMEnsure FPEmulator 4.28

No the prog did not work, it was attempting to Run the proggy that caused
the same crash out that happens if an attempt is made to kill the FPE 4.22
module.

Changing the RMEnsure in the apps Run file so that it now only needs 4.22
and the app works.

> When a module is updated in !System, then precisely nothing happens, 
> unless an application wants that later version and asks for it.

In that case, as 4.28 is available in System Modules, and was put there by
the Config-System-Merge, why is this not happening?

xxxxxx (See end)

> In this particular case, it would probably be utterly wrong to replace 
> your 4.22 ROM version by a softloaded 4.28, but that subtlety should 
> not be your concern.

> > Any attempt to Kill or Unplug FPE 4.22 results in the machine failing
> > as it auto processes through and ends all background system stuff,
> > requiring a back button reset.

> Why would you want to Kill or Unplug FPE 4.22?

Because, when it's active the app, which requires 4.28 will not run, but
crashes the computer (Or something does).
> Martin

[Snippy]

> > This hardware SARPC has an identical backup on a VRPC-AdjustSA (Win
> > XP) and the same process was undertaken on that, and the merge went
> > correctly, making  the FPE 4.22 Dormant and activating the softload
> > 4.28

Why then on the VRPC version of this install was the FPE 4.22 made dormant
by the merge and the 4.28 softload shows as active?

Thanks for the input Martin, appreciated.

Dave

-- 

Dave Triffid

[toc] | [prev] | [next] | [standalone]

#4583

FromMartin Wuerthner <spamtrap@mw-software.com>
Date2012-03-23 16:23 +0100
Message-ID<a4d3037552.martin@bach.planiverse.com>
In reply to#4571
In message <5274f22180dave@triffid.co.uk>
          Dave Symes <dave@triffid.co.uk> wrote:

> In article <a145e57452.martin@bach.planiverse.com>,
>    Martin Wuerthner <spamtrap@mw-software.com> wrote:
>> In message <5274db99dcdave@triffid.co.uk>
>>           Dave Symes <dave@triffid.co.uk> wrote:

>>> FPEmulator 4.22 is in ROM and should have been made dormant and
>>> replaced with a soft load FPEmulator 4.28, but it has not.

>> Should it? What makes you think so? Did the documentation advise you
>> that that should happen? Does the program ask for 4.28? Does the
>> program not work?

> Documentation, don't get to see any of that stuff these days.
> Yes, in the Run file it has a RMEnsure FPEmulator 4.28

> No the prog did not work, it was attempting to Run the proggy that caused
> the same crash out that happens if an attempt is made to kill the FPE 4.22
> module.

Ah, I see. Not a good idea to have an RMEnsure FPEmulator in a !Run 
file. Softloading that module requires quite a bit more effort to make 
sure that it does not crash the system. I went through all that when 
putting the PostScript 3 release together. Having to softload 
FPEmulator is a huge inconvenience only justified by the fact that at 
the time of the PS3 release most installed versions of FPEmulator had 
a critical bug that broke PS3. That bug would not affect 
PipeDream/Fireworks though and I am not aware of anything being wrong 
with version 4.22 in the RO6 ROM, so I wonder why it was considered 
necessary to go to the trouble of ensuring a later FPEmulator.

Maybe that is a recommendation in the latest C/C++ tools?

Speaking of C/C++ tools, I do hope that the System update that is 
offered for download alongside Fireworks is the current official C/C++ 
tools release. Has anyone checked?

-- 
Martin
---------------------------------------------------------------------
Martin Wuerthner         MW Software      http://www.mw-software.com/
        RISC OS Software for Design, Printing and Publishing
---------------------------------------------------------------------

[toc] | [prev] | [next] | [standalone]

#4621

From"Ste (news)" <steve@revi11.plus.com>
Date2012-03-27 18:21 +0100
Message-ID<52771de945steve@revi11.plus.com>
In reply to#4583
In article <a4d3037552.martin@bach.planiverse.com>,
   Martin Wuerthner <spamtrap@mw-software.com> wrote:
> Ah, I see. Not a good idea to have an RMEnsure FPEmulator in a !Run 
> file.

That's not quite true - as Martin goes on to say.

It's fine to RMEnsure FPEmulator in a !Run file if all you're planning on
doing if a suitable version isn't running is to generate an error message
(or similar). If, however, you go on to try to softload a replacement,
you're into the world of pain Martin describes.

Ta,

Steve

-- 
Steve Revill @ Home
Note: All opinions expressed herein are my own.

[toc] | [prev] | [standalone]

Back to top | Article view | comp.sys.acorn.misc

csiph-web