Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

10.0.4 printting

9 views
Skip to first unread message

Barry Landy

unread,
May 8, 2012, 5:00:07 PM5/8/12
to
Both the preview and the releeas of 10.0.4

printing to a CUPS USB connected printer works fine; printing to a LPT1
connected printer (HP5Llaserjet in case it matters) causes FF10 to end
with a beep.

--
Barry Landy Email: Remove nospam in from address
192, Gilbert Road, Cambridge CB4 3PB









Dave Yeo

unread,
May 8, 2012, 10:12:49 PM5/8/12
to
Barry Landy wrote:
> Both the preview and the releeas of 10.0.4
>
> printing to a CUPS USB connected printer works fine; printing to a LPT1
> connected printer (HP5Llaserjet in case it matters) causes FF10 to end
> with a beep.
>

Can you look in your Firefox program directory and post the trp file to
cr...@e-vertise.com? preferably as a text attachment. As far as I know
it shouldn't matter how the printer is connected but I'm not the expert.
Perhaps Rich will become involved again and the trp file may help
diagnose the problem or something will stand out in it.
Dave

Dave Yeo

unread,
May 8, 2012, 10:15:03 PM5/8/12
to
Barry Landy wrote:
> Both the preview and the releeas of 10.0.4
>
> printing to a CUPS USB connected printer works fine; printing to a LPT1
> connected printer (HP5Llaserjet in case it matters) causes FF10 to end
> with a beep.

Also has anyone else had experience printing to a printer connected to
anything besides USB?
Dave

Steve Wendt

unread,
May 8, 2012, 10:57:47 PM5/8/12
to
On 5/8/2012 7:15 PM, Dave Yeo wrote:

>> printing to a CUPS USB connected printer works fine; printing to a LPT1
>> connected printer (HP5Llaserjet in case it matters) causes FF10 to end
>> with a beep.
>
> Also has anyone else had experience printing to a printer connected to
> anything besides USB?

I haven't tried lately, but my HP Laserjet is connected to LPT1.

Steve Wendt

unread,
May 8, 2012, 11:23:05 PM5/8/12
to
On 05/08/12 07:57 pm, Steve Wendt wrote:

> I haven't tried lately, but my HP Laserjet is connected to LPT1.

Works fine, using the PostScript driver. I have both of these values
set to true: print.os2.postscript.use_builtin and ...use_ibmnull.

From what I recall of earlier testing, other values worked, but
produced grainy results.

I do seem to recall at least one other user having problems with
printing to an HP Laserjet, but I don't know if there was ever a resolution.

FWIW, mine is a 4250n, using the 4/4M Plus PS 300 version 30.827
PostScript driver.

Lewis Rosenthal

unread,
May 9, 2012, 12:12:48 AM5/9/12
to
On 05/08/12 10:15 pm, Dave Yeo thus wrote :
I print to network-attached printers, either directly to the IP or via
captured parallel port to a NetWare queue. (Note that a NetWare-captured
port is not the same as a redirected port in the spooler, though the end
result is, obviously, the same.)

I just tested here under SM & FF (to an HP Color LaserJet 3800, using
the native driver), and no crashes printing to a captured port. I don't
have a parallel (or serial!) printer around at the moment to test via a
cable.

Barry, is your spooler enabled or disabled? I suppose that it might be
possible to overload the print stream with the spooler disabled going
over a parallel port, depending upon the amount of data being sent. I'm
not sure of the buffering available in the application, in that case.

--
Lewis
-------------------------------------------------------------
Lewis G Rosenthal, CNA, CLP, CLE, CWTS
Rosenthal & Rosenthal, LLC www.2rosenthals.com
Need a managed Wi-Fi hotspot? www.hautspot.com
visit my IT blog www.2rosenthals.net/wordpress
-------------------------------------------------------------

Barry Landy

unread,
May 9, 2012, 2:21:24 AM5/9/12
to
On Tue, 8 May 2012, Dave Yeo wrote:
:>

I have attached the .trp file which indicates an C0000005 trap.

Barry Landy

unread,
May 9, 2012, 2:23:12 AM5/9/12
to
On Tue, 8 May 2012, Steve Wendt wrote:
:>

I should have said that I have not had preblems printing with earlier
releases.

I use the HP LASERJET 5L driver as AFAIK spooling is on.

Barry Landy

unread,
May 9, 2012, 5:14:30 AM5/9/12
to
On Tue, 8 May 2012, Barry Landy wrote:

:>Both the preview and the releeas of 10.0.4
:>
:>printing to a CUPS USB connected printer works fine; printing to a LPT1
:>connected printer (HP5Llaserjet in case it matters) causes FF10 to end with a
:>beep.
:>
:>

extra points:
1) it happens on two different computer with very different specs
2) Firefox 10.0 does NOT show this effect (printing to the HPL5 via LPT1
works fine).

Dave Saville

unread,
May 9, 2012, 5:37:52 AM5/9/12
to
On Wed, 9 May 2012 02:15:03 UTC, Dave Yeo <dave....@gmail.com>
wrote:
Printing to my HP Laserjet 4L which is network attached (slpr) works
fine.

--
Regards
Dave Saville

David McKenna

unread,
May 9, 2012, 7:23:19 AM5/9/12
to dev-po...@lists.mozilla.org
Works fine (both your latest SeaMonkey and FF) printing over the network to
a CUPS server (which itself sends the job over the network).

Dave McKenna



Barbara

unread,
May 9, 2012, 9:13:06 AM5/9/12
to
On Wed, 9 May 2012 02:15:03 UTC, Dave Yeo <dave....@gmail.com>
wrote:

Parallel, LPT1 to Brother B/W Laser and HP2550L Color Laserjet work
fine.

--
Barbara

Dave Yeo

unread,
May 9, 2012, 10:56:50 AM5/9/12
to
Barry Landy wrote:
> On Tue, 8 May 2012, Dave Yeo wrote:
>
> :>Barry Landy wrote:
> :>> Both the preview and the releeas of 10.0.4
> :>>
> :>> printing to a CUPS USB connected printer works fine; printing to a LPT1
> :>> connected printer (HP5Llaserjet in case it matters) causes FF10 to end
> :>> with a beep.
> :>>
> :>
> :>Can you look in your Firefox program directory and post the trp file to
> :>cr...@e-vertise.com? preferably as a text attachment. As far as I know it
> :>shouldn't matter how the printer is connected but I'm not the expert. Perhaps
> :>Rich will become involved again and the trp file may help diagnose the problem
> :>or something will stand out in it.
> :>Dave
> :>
>
> I have attached the .trp file which indicates an C0000005 trap.
>

Try disabling the Innotek Font engine.
Dave

Dave Yeo

unread,
May 9, 2012, 10:57:34 AM5/9/12
to
Barry Landy wrote:
> On Tue, 8 May 2012, Barry Landy wrote:
>
> :>Both the preview and the releeas of 10.0.4
> :>
> :>printing to a CUPS USB connected printer works fine; printing to a LPT1
> :>connected printer (HP5Llaserjet in case it matters) causes FF10 to end with a
> :>beep.
> :>
> :>
>
> extra points:
> 1) it happens on two different computer with very different specs
> 2) Firefox 10.0 does NOT show this effect (printing to the HPL5 via LPT1
> works fine).
>

What about 10.02?
Dave

Barry Landy

unread,
May 9, 2012, 2:08:58 PM5/9/12
to
On Wed, 9 May 2012, Dave Yeo wrote:

:>Barry Landy wrote:
:>

OK. I found MOZINST which includes fechg.exe which enables switching
on.off the font engine and that does cure the exception.

Barry Landy

unread,
May 9, 2012, 3:32:28 PM5/9/12
to
On Wed, 9 May 2012, Dave Yeo wrote:

:>

Same as 10.0 ( no exception)

Barry Landy

unread,
May 9, 2012, 3:55:18 PM5/9/12
to
On Wed, 9 May 2012, Barry Landy wrote:

:>On Wed, 9 May 2012, Dave Yeo wrote:
:>
:>:>Barry Landy wrote:
:>:>> On Tue, 8 May 2012, Barry Landy wrote:
:>:>>
:>:>> :>Both the preview and the releeas of 10.0.4
:>:>> :>
:>:>> :>printing to a CUPS USB connected printer works fine; printing to a LPT1
:>:>> :>connected printer (HP5Llaserjet in case it matters) causes FF10 to end
:>:>> with a
:>:>> :>beep.
:>:>> :>
:>:>> :>
:>:>>
:>:>> extra points:
:>:>> 1) it happens on two different computer with very different specs
:>:>> 2) Firefox 10.0 does NOT show this effect (printing to the HPL5 via LPT1
:>:>> works fine).
:>:>>
:>:>
:>:>What about 10.02?
:>:>Dave
:>:>
:>
:>Same as 10.0 ( no exception)
:>
:>

Spoke too soon about fechg: even after fechg -d firefox the exception
still happens (still C0...05 in PM MERGE)

Allan

unread,
May 9, 2012, 4:52:33 PM5/9/12
to
On Wed, 9 May 2012 19:55:18 UTC, Barry Landy <bl10n...@cam.ac.uk> wrote:

> On Wed, 9 May 2012, Barry Landy wrote:
>
> :>On Wed, 9 May 2012, Dave Yeo wrote:
> :>
> :>:>Barry Landy wrote:
> :>:>> On Tue, 8 May 2012, Barry Landy wrote:
> :>:>>
> :>:>> :>Both the preview and the releeas of 10.0.4
> :>:>> :>
> :>:>> :>printing to a CUPS USB connected printer works fine; printing to a LPT1
> :>:>> :>connected printer (HP5Llaserjet in case it matters) causes FF10 to end
> :>:>> with a
> :>:>> :>beep.
> :>:>> :>
> :>:>> :>
> :>:>>
> :>:>> extra points:
> :>:>> 1) it happens on two different computer with very different specs
> :>:>> 2) Firefox 10.0 does NOT show this effect (printing to the HPL5 via LPT1
> :>:>> works fine).
> :>:>>
> :>:>
> :>:>What about 10.02?
> :>:>Dave
> :>:>
> :>
> :>Same as 10.0 ( no exception)
> :>
> :>
>
> Spoke too soon about fechg: even after fechg -d firefox the exception
> still happens (still C0...05 in PM MERGE)

You need to reboot after such a change.


--
Allan.

It is better to close your mouth, and look like a fool,
than to open it, and remove all doubt.

Barry Landy

unread,
May 9, 2012, 6:27:45 PM5/9/12
to
On Wed, 9 May 2012, Allan wrote:
Ah..... "documentation" does not say so. Implies it can be switched off
and on.

yes, after reboot the exception does not happen although printing is not
very happy (perhaps something wrong with the document?)

Barry Landy

unread,
May 10, 2012, 5:10:25 AM5/10/12
to
On Wed, 9 May 2012, Dave Yeo wrote:

:>Barry Landy wrote:
:>

Contrary to my previous email I also get an exception printing to HPL5
from FF10.0 (still C05 but this time from a dfferent location; start of
.trp below)


______________________________________________________________________

Exception Report - created 2012/02/22 12:29:24
______________________________________________________________________

Firefox v10.0 - build 20120130135311

OS2/eCS Version: 2.45
# of Processors: 4
Physical Memory: 511 mb
Virt Addr Limit: 1536 mb
Exceptq Version: 7.10 (Mar 1 2011)

______________________________________________________________________

Exception C0000005 - Access Violation
______________________________________________________________________

Process: F:\FIREFOX\FIREFOX.EXE
PID: 91 (145)
TID: 01 (1)
Priority: 200

Filename: F:\FIREFOX\XUL.DLL
Address: 005B:1D9FC56B (0001:012FC56B)
Cause: Attempted to read from 00000000
(not a valid address)

______________________________________________________________________

Failing Instruction
______________________________________________________________________

1D9FC55B MOV ECX, [ESP+0x24] (8b4c24 24)
1D9FC55F MOV [ESP+0x18], ESI (897424 18)
1D9FC563 MOV EBX, [ESP+0x2c] (8b5c24 2c)
1D9FC567 MOV ESI, [ESP+0x28] (8b7424 28)
1D9FC56B >MOV EDX, [EAX] (8b10)
1D9FC56D MOV [ESP+0x4], ECX (894c24 04)
1D9FC571 MOV [ESP], EAX (890424)
1D9FC574 CALL [EDX+0x20] (ff52 20)

Barry Landy

unread,
May 10, 2012, 6:20:24 AM5/10/12
to
On Thu, 10 May 2012, Barry Landy wrote:

:>On Wed, 9 May 2012, Dave Yeo wrote:
:>
:>:>Barry Landy wrote:
:>:>> On Tue, 8 May 2012, Dave Yeo wrote:
:>:>>
:>:>> :>Barry Landy wrote:
:>:>> :>> Both the preview and the releeas of 10.0.4
:>:>> :>>
:>:>> :>> printing to a CUPS USB connected printer works fine; printing to a LPT1
:>:>> :>> connected printer (HP5Llaserjet in case it matters) causes FF10 to end
:>:>> :>> with a beep.
:>:>> :>>
:>:>> :>
:>:>> :>Can you look in your Firefox program directory and post the trp file to
:>:>> :>cr...@e-vertise.com? preferably as a text attachment. As far as I know it
:>:>> :>shouldn't matter how the printer is connected but I'm not the expert.
:>:>> Perhaps
:>:>> :>Rich will become involved again and the trp file may help diagnose the
:>:>> problem
:>:>> :>or something will stand out in it.
:>:>> :>Dave
:>:>> :>
:>:>>
:>:>> I have attached the .trp file which indicates an C0000005 trap.
:>:>>
:>:>
:>:>Try disabling the Innotek Font engine.
:>:>Dave
:>:>
:>

So my final analysis is that all versions 10 give an exception when
printing to my HP5L unless Innotek font engine is off for firefox, when
other funny things happen (documents go to the printer but the printer
does not like them).

I am now back on Aurora (FF9.0) which seems fine (with FE ONN)

[Thanks BTW for all the hard work; I really appreciated getting my
cursor back!]

Steven Levine

unread,
May 10, 2012, 7:24:12 PM5/10/12
to
On Thu, 10 May 2012 09:10:25 UTC, Barry Landy <bl10n...@cam.ac.uk>
wrote:

Hi Barry,

> Contrary to my previous email I also get an exception printing to HPL5
> from FF10.0 (still C05 but this time from a dfferent location; start of
> .trp below)

I you wish me to look at this, you need to attach the entire .trp file
or e-mail to me directly. It does not appear that your attempts to
attach the .trp file to prior messages have succeeded.

Post pieces of the .trp file is not useful. You don't know which
parts are important.

Steven


--
---------------------------------------------------------------------
Steven Levine <ste...@earthlink.bogus.net>
eCS/Warp/DIY etc. www.scoug.com www.ecomstation.com
---------------------------------------------------------------------

Dave Yeo

unread,
May 10, 2012, 7:32:30 PM5/10/12
to
Steven Levine wrote:
> On Thu, 10 May 2012 09:10:25 UTC, Barry Landy<bl10n...@cam.ac.uk>
> wrote:
>
> Hi Barry,
>
>> > Contrary to my previous email I also get an exception printing to HPL5
>> > from FF10.0 (still C05 but this time from a dfferent location; start of
>> > .trp below)
> I you wish me to look at this, you need to attach the entire .trp file
> or e-mail to me directly. It does not appear that your attempts to
> attach the .trp file to prior messages have succeeded.

I've forwarded it to you. Perhaps if you ever talk to Rich, you can get
access to cr...@e-vertise.com
Dave

Steven Levine

unread,
May 10, 2012, 9:00:39 PM5/10/12
to
On Thu, 10 May 2012 23:32:30 UTC, Dave Yeo <dave....@gmail.com>
wrote:

Hi Dave,

> I've forwarded it to you.

A quick look says Barry still has the Innotek font library enabled. I
see ft2lib.dll in the call path.

Barry, what is the bldlevel of your pmmerge.dll?

Dave Yeo

unread,
May 10, 2012, 9:41:06 PM5/10/12
to
Steven Levine wrote:
> On Thu, 10 May 2012 23:32:30 UTC, Dave Yeo<dave....@gmail.com>
> wrote:
>
> Hi Dave,
>
>> > I've forwarded it to you.
> A quick look says Barry still has the Innotek font library enabled. I
> see ft2lib.dll in the call path.

I had him disable it and
Barry Landy wrote:
> So my final analysis is that all versions 10 give an exception when
> printing to my HP5L unless Innotek font engine is off for firefox, when
> other funny things happen (documents go to the printer but the printer
> does not like them).
>
> I am now back on Aurora (FF9.0) which seems fine (with FE ONN)

Dave

Steven Levine

unread,
May 10, 2012, 10:46:54 PM5/10/12
to
On Fri, 11 May 2012 01:41:06 UTC, Dave Yeo <dave....@gmail.com>
wrote:

> I had him disable it and
> Barry Landy wrote:
> > So my final analysis is that all versions 10 give an exception when
> > printing to my HP5L unless Innotek font engine is off for firefox, when
> > other funny things happen (documents go to the printer but the printer
> > does not like them).

I was under the impression that the Innotek font engine is both not
compatible and not useful with recent firefoxes. If so, there's not a
lot of reason to look at a setup that still has this enabled for
firefox. Am I confused?

Dave Yeo

unread,
May 10, 2012, 11:51:07 PM5/10/12
to
Steven Levine wrote:
> On Fri, 11 May 2012 01:41:06 UTC, Dave Yeo<dave....@gmail.com>
> wrote:
>
>> > I had him disable it and
>> > Barry Landy wrote:
>> > > So my final analysis is that all versions 10 give an exception when
>> > > printing to my HP5L unless Innotek font engine is off for firefox, when
>> > > other funny things happen (documents go to the printer but the printer
>> > > does not like them).
> I was under the impression that the Innotek font engine is both not
> compatible and not useful with recent firefoxes. If so, there's not a
> lot of reason to look at a setup that still has this enabled for
> firefox. Am I confused?

No, you are not confused. After Mozilla switched to Cairo for drawing,
Cairo took care of anti-aliasing and it was unneeded. Peter also stated
that it was incompatible and the Firefox etc key should be removed.
The problem is that eCS ships with it enabled for Firefox, which while
it doesn't seem to hurt most of the time, I'm always advising to disable
it and Lars even wrote a small program to fix the registry entries.
In Barry's case, I guess that there is an interaction between his
printer driver and the Cairo printing surface that caused a crash with
the font engine and still causes problems without it. Most people seem
to be able to print successfully and not much changed between 9 and 10.
Steve vaguely recalls others having problems with the Laserjet.

The next question is who to contact to get the Font engine not enabled
for Mozilla. I take it that it was enabled as eCS also ships an older
version of Firefox which supports the Java plugin. I'd guess if the new
plugin works then the old Firefox won't ship in 2.2.
Dave

Steve Wendt

unread,
May 11, 2012, 1:00:58 AM5/11/12
to
On 05/10/12 08:51 pm, Dave Yeo wrote:

> Steve vaguely recalls others having problems with the Laserjet.

Switching to the Postscript driver will help:
https://bugzilla.mozilla.org/show_bug.cgi?id=682952#c8

> I take it that it was enabled as eCS also ships an older
> version of Firefox which supports the Java plugin.

I don't know what version was included, but 3.5.x still supported the
OJI plugins, and also used Cairo.

Dave Yeo

unread,
May 11, 2012, 1:53:41 AM5/11/12
to
I can't find 3.5.x on the eCS 2.1 CD though I do remember having the
offer to install it. The question is did it use Cairo for displaying text?
I'm posting from SM 1.19 which does use Cairo and get the thumbnail
thing when mouse overing a tab but the text is not anti-aliased and
looks terrible. This is built on Gecko 1.8 which I think FF 3.5.x was as
well but I can't remember for sure.
Dave

Barry Landy

unread,
May 11, 2012, 2:37:18 AM5/11/12
to
On Thu, 10 May 2012, Steven Levine wrote:

:>On Thu, 10 May 2012 23:32:30 UTC, Dave Yeo <dave....@gmail.com>
:>wrote:
:>
:>Hi Dave,
:>
:>> I've forwarded it to you.
:>
:>A quick look says Barry still has the Innotek font library enabled. I
:>see ft2lib.dll in the call path.
:>
:>Barry, what is the bldlevel of your pmmerge.dll?

In case it is still relevant PMMERGE is 14.106 (10 04 2007)

Barry Landy

unread,
May 11, 2012, 4:47:14 AM5/11/12
to
On Thu, 10 May 2012, Steve Wendt wrote:

:>On 05/10/12 08:51 pm, Dave Yeo wrote:
:>
:>> Steve vaguely recalls others having problems with the Laserjet.
:>
:>Switching to the Postscript driver will help:
:>https://bugzilla.mozilla.org/show_bug.cgi?id=682952#c8
:>

But the HP 5L is a PCL5 printer. Surely the postscript driver would not
work?

:>> I take it that it was enabled as eCS also ships an older
:>> version of Firefox which supports the Java plugin.
:>
:>I don't know what version was included, but 3.5.x still supported the OJI
:>plugins, and also used Cairo.
:>

Barry Landy

unread,
May 11, 2012, 5:51:01 AM5/11/12
to
On Thu, 10 May 2012, Dave Yeo wrote:
:>

I have found that even after using fechg -d firefox (and rebooting) I
can geet a trap in PMMRGE from both Aurora and FF10.4 when trying to
print to the HPL5. The page I am trying to print is
http://www.os2bbs.com/os2news/
is OS2 Warp news and rumours
(not all pages cause a trap)

I have posted the .trp file to cr...@e-vertise.com

I notice that FT2LIB is in the "DLLs accessible" list if that is
relevant.
Message has been deleted

Dave Yeo

unread,
May 11, 2012, 10:09:28 AM5/11/12
to
Allan wrote:
> On Fri, 11 May 2012 05:53:41 UTC, Dave Yeo<dave....@gmail.com> wrote:
>
>> On 05/10/12 10:00 pm, Steve Wendt wrote:
>>> On 05/10/12 08:51 pm, Dave Yeo wrote:
>>>
>>>> Steve vaguely recalls others having problems with the Laserjet.
>>>
>>> Switching to the Postscript driver will help:
>>> https://bugzilla.mozilla.org/show_bug.cgi?id=682952#c8
>>>
>>>> I take it that it was enabled as eCS also ships an older
>>>> version of Firefox which supports the Java plugin.
>>>
>>> I don't know what version was included, but 3.5.x still supported the
>>> OJI plugins, and also used Cairo.
>>
>> I can't find 3.5.x on the eCS 2.1 CD though I do remember having the
>> offer to install it.
>
> There is a link from http://www.mozilla.org/ports/os2/
> :-)

I'm bandwidth constrained :)

>
> >The question is did it use Cairo for displaying text?
>> I'm posting from SM 1.19 which does use Cairo
>
> AFAIK Cairo was not included in SM before 2.0
> My SM 1.19 shows really ugly fonts if I disable IFE.

I looked at the source of 1.19, it uses Cairo, but not for everything so
it still has the transparent png bug and as you said, the fonts look ugly.
Dave

Dave Yeo

unread,
May 11, 2012, 11:15:38 AM5/11/12
to
Barry Landy wrote:
> I have found that even after using fechg -d firefox (and rebooting) I
> can geet a trap in PMMRGE from both Aurora and FF10.4 when trying to
> print to the HPL5. The page I am trying to print is
> http://www.os2bbs.com/os2news/
> is OS2 Warp news and rumours
> (not all pages cause a trap)
>
> I have posted the .trp file toc...@e-vertise.com

I forwarded it to Steven.

>
> I notice that FT2LIB is in the "DLLs accessible" list if that is
> relevant.

That just means it is on the LIBPATH
Dave

Steve Wendt

unread,
May 11, 2012, 1:36:59 PM5/11/12
to
On 5/11/2012 5:10 AM, Allan wrote:

>>> I don't know what version was included, but 3.5.x still supported the
>>> OJI plugins, and also used Cairo.
>>
>> The question is did it use Cairo for displaying text?
>
> AFAIK Cairo was not included in SM before 2.0

Right, SeaMonkey 2.0.x and Firefox 3.5.x were the versions that started
using Thebes.

Steve Wendt

unread,
May 11, 2012, 1:39:22 PM5/11/12
to
On 5/11/2012 1:47 AM, Barry Landy wrote:

> :>Switching to the Postscript driver will help:
> :>https://bugzilla.mozilla.org/show_bug.cgi?id=682952#c8
>
> But the HP 5L is a PCL5 printer. Surely the postscript driver would not
> work?

I didn't realize HP made laser printers without PostScript, but they
sure did:
http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?lang=en&cc=us&taskId=130&prodSeriesId=240170&prodTypeId=18972&objectID=bpl11943

Why would you buy one? :-)

Steven Levine

unread,
May 11, 2012, 2:16:39 PM5/11/12
to
On Fri, 11 May 2012 15:15:38 UTC, Dave Yeo <dave....@gmail.com>
wrote:

Hi Dave,

> > I have posted the .trp file toc...@e-vertise.com
>
> I forwarded it to Steven.

I assume is this 0052_01.TRP. This one removes the ft2lib usage, but
still traps. I'll take a closer look later today.

FWIW, the trap is in a pmmerge at SavePath32+4E. Unfortunately for
use, IBM chose to build many DLLs without stack frames so it will take
some manual effort to determine the actual call path.

Barry and others, please grab copies of the .xqs files you don't have
from the zip files at

http://www.warpcave.com/symbols

For exceptq .trp reports you only need to install the .xqs files. As
with all .xqs files, install the .xqs file in the directory that
contains the corresponding DLL.

> > I notice that FT2LIB is in the "DLLs accessible" list if that is
> > relevant.
>
> That just means it is on the LIBPATH

No, it means that the DLL has been loaded into the application's
memory space and is available for use. This is quite different.

The ft2lib.dll is loaded whenever the Innotek font library is
installed. This is controlled by the SYS_DLLS->LoadPerProcess INI
key.

Steven






> Dave

Steven Levine

unread,
May 11, 2012, 7:37:19 PM5/11/12
to
On Fri, 11 May 2012 18:16:39 UTC, "Steven Levine"
<ste...@nomail.earthlink.net> wrote:

Hi,

> I assume is this 0052_01.TRP. This one removes the ft2lib usage, but
> still traps. I'll take a closer look later today.

OK. The failing call path is

__moz_cairo_os2_printing_surface_create + 2B4A
->GpiSavePS
->IdpSavePS
->SaveDC32
->SavePath32

The call is not really from __moz_cairo_os2_printing_surface_create.
Rather it's from one of the static functions that follow it in the
sources. This is cairo_os2_printing_surface_create() in the sources.

The call path orginates from __cairo_surface_show_text_glyphs() and is
probably calling _cairo_os2_printing_surface_show_glyphs() which is
doing the failing GpiSavePS.

I guess the question is what is wrong with the PS.

Steven

Dave Yeo

unread,
May 11, 2012, 9:02:23 PM5/11/12
to
SM 1.19 used a minimal Thebes
Directory of I:\mozilla\gfx\thebes\src

5-17-10 9:53p <DIR> 0 ---- .
4-03-10 9:53a <DIR> 0 ---- ..
4-03-10 9:53a <DIR> 0 ---- CVS
7-01-05 6:05p 10,786 0 a--- gfxContext.cpp
6-29-05 9:58p 3,088 0 a--- gfxImageSurface.cpp
7-01-05 6:05p 1,868 0 a--- gfxPattern.cpp
6-28-05 2:18a 2,019 0 a--- gfxWindowsSurface.cpp
7-01-05 6:36p 3,579 0 a--- gfxXlibSurface.cpp
6-29-05 9:58p 596 0 a--- Makefile.in

Dave

Steve Wendt

unread,
May 11, 2012, 9:13:06 PM5/11/12
to
On 5/11/2012 4:37 PM, Steven Levine wrote:

> The call path orginates from __cairo_surface_show_text_glyphs() and is
> probably calling _cairo_os2_printing_surface_show_glyphs() which is
> doing the failing GpiSavePS.
>
> I guess the question is what is wrong with the PS.

This sounds similar to what Rich found:
https://bugzilla.mozilla.org/show_bug.cgi?id=682952#c8

If you get any farther, you may want to add the info to Bugzilla.

Dave Yeo

unread,
May 11, 2012, 9:33:19 PM5/11/12
to dev-po...@lists.mozilla.org
On 05/11/12 04:37 pm, Steven Levine wrote:
> On Fri, 11 May 2012 18:16:39 UTC, "Steven Levine"
> <ste...@nomail.earthlink.net> wrote:
>
> Hi,
>
>> I assume is this 0052_01.TRP. This one removes the ft2lib usage, but
>> still traps. I'll take a closer look later today.
>
> OK. The failing call path is
>
> __moz_cairo_os2_printing_surface_create + 2B4A
> ->GpiSavePS
> ->IdpSavePS
> ->SaveDC32
> ->SavePath32
>
> The call is not really from __moz_cairo_os2_printing_surface_create.
> Rather it's from one of the static functions that follow it in the
> sources. This is cairo_os2_printing_surface_create() in the sources.
>
> The call path orginates from __cairo_surface_show_text_glyphs() and is
> probably calling _cairo_os2_printing_surface_show_glyphs() which is
> doing the failing GpiSavePS.
>
> I guess the question is what is wrong with the PS.
>

I wonder if it is related to the failure of printing to PDF or PS on the second run.
Ignoring white space changes, here is a diff between ver 6 and the current one.
Ver 6 was the last one Rich worked on with the updates mostly from Walter. Updating the patch when it failed was a bitch as the rej file was 40kb or so.

--- G:\comm-central.6\mozilla\gfx\cairo\cairo\src\cairo-os2-printing-surface.c Mon Aug 1 17:41:40 2011
+++ cairo-os2-printing-surface.c Thu May 3 21:46:42 2012
@@ -716,13 +716,10 @@

old_content = surface->content;
if (recording_surface->base.content == CAIRO_CONTENT_COLOR) {
- cairo_pattern_t *source;
- cairo_solid_pattern_t black;

- surface->content = CAIRO_CONTENT_COLOR;
- _cairo_pattern_init_solid (&black, CAIRO_COLOR_BLACK);
- source = (cairo_pattern_t*) &black;
- status = _cairo_os2_printing_surface_paint_solid_pattern (surface, source);
+ surface->content = CAIRO_CONTENT_COLOR;
+ status = _cairo_os2_printing_surface_paint_solid_pattern (surface,
+ &_cairo_pattern_black.base);
if (status)
return status;
}
@@ -1055,9 +1052,9 @@
cairo_operator_t op,
const cairo_pattern_t *source,
cairo_path_fixed_t *path,
- cairo_stroke_style_t *style,
- cairo_matrix_t *stroke_ctm,
- cairo_matrix_t *stroke_ctm_inverse,
+ const cairo_stroke_style_t *style,
+ const cairo_matrix_t *stroke_ctm,
+ const cairo_matrix_t *stroke_ctm_inverse,
double tolerance,
cairo_antialias_t antialias,
cairo_clip_t *clip)

Dave

Barry Landy

unread,
May 12, 2012, 5:03:48 PM5/12/12
to
On Fri, 11 May 2012, Steve Wendt wrote:
:>

When I bought it it was the latest and best (must be more than 15
years back)

Barry Landy

unread,
May 12, 2012, 5:11:55 PM5/12/12
to
On Fri, 11 May 2012, Steve Wendt wrote:

:>

I did try the "printer specific format" option mentioned there with very
much the same effects are mentioned in the reply (printer blinking for
ever).

Lewis Rosenthal

unread,
May 12, 2012, 7:19:50 PM5/12/12
to
On 05/12/12 05:03 pm, Barry Landy thus wrote :
> On Fri, 11 May 2012, Steve Wendt wrote:
>
> :>On 5/11/2012 1:47 AM, Barry Landy wrote:
> :>
> :>> :>Switching to the Postscript driver will help:
> :>> :>https://bugzilla.mozilla.org/show_bug.cgi?id=682952#c8
> :>>
> :>> But the HP 5L is a PCL5 printer. Surely the postscript driver would not
> :>> work?
> :>
> :>I didn't realize HP made laser printers without PostScript, but they sure did:
> :>http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?lang=en&cc=us&taskId=130&prodSeriesId=240170&prodTypeId=18972&objectID=bpl11943
> :>
> :>Why would you buy one? :-)
> :>
>
> When I bought it it was the latest and best (must be more than 15
> years back)
>

...adm at the time, only bigger LaserJets came with PostScript emulation
as a default (5Si, 4000 series on up, etc.). Smaller LaserJets only used
PCL, to keep the price down, though you could purchase a PostScript SIMM
for most of them.

There were versions of the 4L & 5L (4ML & 5ML, respectively) which were
"Mac-compatible," and thus, were native PostScript (emulation) printers
(not PCL, because on Mac OS up through version 9, PostScript was the
native language).

Barry, what version is your LaserJet driver? The one I have installed
here is 30.827.

--
Lewis
-------------------------------------------------------------
Lewis G Rosenthal, CNA, CLP, CLE, CWTS
Rosenthal & Rosenthal, LLC www.2rosenthals.com
Need a managed Wi-Fi hotspot? www.hautspot.com
visit my IT blog www.2rosenthals.net/wordpress
-------------------------------------------------------------

Barry Landy

unread,
May 12, 2012, 8:20:18 PM5/12/12
to
On Fri, 11 May 2012, Steven Levine wrote:

:>On Fri, 11 May 2012 15:15:38 UTC, Dave Yeo <dave....@gmail.com>
:>wrote:
:>
:>Hi Dave,
:>
:>> > I have posted the .trp file toc...@e-vertise.com
:>>
:>> I forwarded it to Steven.
:>
:>I assume is this 0052_01.TRP. This one removes the ft2lib usage, but
:>still traps. I'll take a closer look later today.
:>
:>FWIW, the trap is in a pmmerge at SavePath32+4E. Unfortunately for
:>use, IBM chose to build many DLLs without stack frames so it will take
:>some manual effort to determine the actual call path.
:>
:>Barry and others, please grab copies of the .xqs files you don't have
:>from the zip files at
:>
:> http://www.warpcave.com/symbols
:>
:>For exceptq .trp reports you only need to install the .xqs files. As
:>with all .xqs files, install the .xqs file in the directory that
:>contains the corresponding DLL.

DONE

:>
:>> > I notice that FT2LIB is in the "DLLs accessible" list if that is
:>> > relevant.
:>>
:>> That just means it is on the LIBPATH
:>
:>No, it means that the DLL has been loaded into the application's
:>memory space and is available for use. This is quite different.
:>
:>The ft2lib.dll is loaded whenever the Innotek font library is
:>installed. This is controlled by the SYS_DLLS->LoadPerProcess INI
:>key.
:>
:>Steven

Barry Landy

unread,
May 13, 2012, 2:17:01 AM5/13/12
to
On Sat, 12 May 2012, Lewis Rosenthal wrote:

:>On 05/12/12 05:03 pm, Barry Landy thus wrote :
:>> On Fri, 11 May 2012, Steve Wendt wrote:
:>>
:>> :>On 5/11/2012 1:47 AM, Barry Landy wrote:
:>> :>
:>> :>> :>Switching to the Postscript driver will help:
:>> :>> :>https://bugzilla.mozilla.org/show_bug.cgi?id=682952#c8
:>> :>>
:>> :>> But the HP 5L is a PCL5 printer. Surely the postscript driver would not
:>> :>> work?
:>> :>
:>> :>I didn't realize HP made laser printers without PostScript, but they sure
:>> did:
:>> :>http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?lang=en&cc=us&taskId=130&prodSeriesId=240170&prodTypeId=18972&objectID=bpl11943
:>> :>
:>> :>Why would you buy one? :-)
:>> :>
:>>
:>> When I bought it it was the latest and best (must be more than 15
:>> years back)
:>>
:>
:>...adm at the time, only bigger LaserJets came with PostScript emulation as a
:>default (5Si, 4000 series on up, etc.). Smaller LaserJets only used PCL, to
:>keep the price down, though you could purchase a PostScript SIMM for most of
:>them.
:>
:>There were versions of the 4L & 5L (4ML & 5ML, respectively) which were
:>"Mac-compatible," and thus, were native PostScript (emulation) printers (not
:>PCL, because on Mac OS up through version 9, PostScript was the native
:>language).
:>
:>Barry, what version is your LaserJet driver? The one I have installed here is
:>30.827.
:>
:>

Looks like 30.827 of 31jan2005 (2370214 bytes)

Barry Landy

unread,
May 14, 2012, 3:08:23 AM5/14/12
to
On Fri, 11 May 2012, Steven Levine wrote:

:>On Fri, 11 May 2012 18:16:39 UTC, "Steven Levine"
:><ste...@nomail.earthlink.net> wrote:
:>
:>Hi,
:>
:>> I assume is this 0052_01.TRP. This one removes the ft2lib usage, but
:>> still traps. I'll take a closer look later today.
:>
:>OK. The failing call path is
:>
:> __moz_cairo_os2_printing_surface_create + 2B4A
:> ->GpiSavePS
:> ->IdpSavePS
:> ->SaveDC32
:> ->SavePath32
:>
:>The call is not really from __moz_cairo_os2_printing_surface_create.
:>Rather it's from one of the static functions that follow it in the
:>sources. This is cairo_os2_printing_surface_create() in the sources.
:>
:>The call path orginates from __cairo_surface_show_text_glyphs() and is
:>probably calling _cairo_os2_printing_surface_show_glyphs() which is
:>doing the failing GpiSavePS.
:>
:>I guess the question is what is wrong with the PS.
:>
:>Steven
:>
:>
:>

FWIW I tried the same print in version 8.0.1 and got the same trap.

If it helps I can let you have the .trp output (with full sym info) for
both the version 10.0.4 trap and the version 8 trap.

Dave Yeo

unread,
May 14, 2012, 6:50:53 PM5/14/12
to
Barry Landy wrote:
> FWIW I tried the same print in version 8.0.1 and got the same trap.

Can you test with 4 and maybe 6?
Dave

Barry Landy

unread,
May 15, 2012, 2:35:57 AM5/15/12
to
On Mon, 14 May 2012, Dave Yeo wrote:
:>

I wanted to do that and couldnt find the binaries. Perhaps you can give
me a URL.

Barry Landy

unread,
May 15, 2012, 4:08:27 AM5/15/12
to
On Tue, 15 May 2012, Barry Landy wrote:

:>On Mon, 14 May 2012, Dave Yeo wrote:
:>
:>:>Barry Landy wrote:
:>:>> FWIW I tried the same print in version 8.0.1 and got the same trap.
:>:>
:>:>Can you test with 4 and maybe 6?
:>:>Dave
:>:>
:>
:>I wanted to do that and couldnt find the binaries. Perhaps you can give
:>me a URL.

I found 4.0.1 in the ECS2.1 dist but I cant find 6.

I do NOT get a trap with 4.0.1

Barry Landy

unread,
May 15, 2012, 8:27:22 AM5/15/12
to
On Tue, 15 May 2012, Barry Landy wrote:

:>On Tue, 15 May 2012, Barry Landy wrote:
:>
:>:>On Mon, 14 May 2012, Dave Yeo wrote:
:>:>
:>:>:>Barry Landy wrote:
:>:>:>> FWIW I tried the same print in version 8.0.1 and got the same trap.
:>:>:>
:>:>:>Can you test with 4 and maybe 6?
:>:>:>Dave
:>:>:>
:>:>
:>:>I wanted to do that and couldnt find the binaries. Perhaps you can give
:>:>me a URL.
:>
:>I found 4.0.1 in the ECS2.1 dist but I cant find 6.
:>
:>I do NOT get a trap with 4.0.1
:>
:>

I found 6 on a different computer (office) and will try at home later.

Dave Yeo

unread,
May 15, 2012, 10:37:12 AM5/15/12
to
Barry Landy wrote:
> On Mon, 14 May 2012, Dave Yeo wrote:
>
> :>Barry Landy wrote:
> :>> FWIW I tried the same print in version 8.0.1 and got the same trap.
> :>
> :>Can you test with 4 and maybe 6?
> :>Dave
> :>
>
> I wanted to do that and couldnt find the binaries. Perhaps you can give
> me a URL.

They were uploaded to the Mozilla ftp server.
Dave

Barry Landy

unread,
May 15, 2012, 12:23:44 PM5/15/12
to
On Tue, 15 May 2012, Dave Yeo wrote:

:>Barry Landy wrote:
Couldnt find them there but found copies on my office computer.

FF6 traps in PMMERGE and I have emailed the crash to cr...@e-vertise.com

FF8 also crashed but FF4 (from ECS21 distro) did not.

Lewis Rosenthal

unread,
May 15, 2012, 12:59:48 PM5/15/12
to
On 05/15/12 12:23 pm, Barry Landy thus wrote :
> On Tue, 15 May 2012, Dave Yeo wrote:
>
> :>Barry Landy wrote:
> :>> On Mon, 14 May 2012, Dave Yeo wrote:
> :>>
> :>> :>Barry Landy wrote:
> :>> :>> FWIW I tried the same print in version 8.0.1 and got the same trap.
> :>> :>
> :>> :>Can you test with 4 and maybe 6?
> :>> :>Dave
> :>> :>
> :>>
> :>> I wanted to do that and couldnt find the binaries. Perhaps you can give
> :>> me a URL.
> :>
> :>They were uploaded to the Mozilla ftp server.
>
> Couldnt find them there but found copies on my office computer.
>
> FF6 traps in PMMERGE and I have emailed the crash to cr...@e-vertise.com
>
> FF8 also crashed but FF4 (from ECS21 distro) did not.
>
I set up a virtual HP 5L printer with the 30.827 LaserJet driver,
directed to print to file. Printing from both SM 2.7.2.4esr and FF
10.0.4esr yielded no crashes and clearly legible and complete PCL files.

PMMERGE.DLL is 14.95

Barry, can you try setting up a printer object to print to file to see
if you get the same crash, or whether you *only* get it when printing
through the LaserJet driver to LPT1?

Lewis Rosenthal

unread,
May 15, 2012, 1:08:06 PM5/15/12
to
On 05/15/12 12:59 pm, Lewis Rosenthal thus wrote :
> On 05/15/12 12:23 pm, Barry Landy thus wrote :
>> On Tue, 15 May 2012, Dave Yeo wrote:
>>
>> :>Barry Landy wrote:
>> :>> On Mon, 14 May 2012, Dave Yeo wrote:
>> :>>
>> :>> :>Barry Landy wrote:
>> :>> :>> FWIW I tried the same print in version 8.0.1 and got the same
>> trap.
>> :>> :>
>> :>> :>Can you test with 4 and maybe 6?
>> :>> :>Dave
>> :>> :>
>> :>>
>> :>> I wanted to do that and couldnt find the binaries. Perhaps you can
>> give
>> :>> me a URL.
>> :>
>> :>They were uploaded to the Mozilla ftp server.
>>
>> Couldnt find them there but found copies on my office computer.
>>
>> FF6 traps in PMMERGE and I have emailed the crash to cr...@e-vertise.com
>>
>> FF8 also crashed but FF4 (from ECS21 distro) did not.
>>
> I set up a virtual HP 5L printer with the 30.827 LaserJet driver,
> directed to print to file. Printing from both SM 2.7.2.4esr and FF
> 10.0.4esr yielded no crashes and clearly legible and complete PCL files.
>

I lied. I mistakenly viewed the same output file. In fact, what hapened
was that SM output a file but FF did not (zero-length; I just tried
creating a file of a completely different name).

> PMMERGE.DLL is 14.95
>
> Barry, can you try setting up a printer object to print to file to see
> if you get the same crash, or whether you *only* get it when printing
> through the LaserJet driver to LPT1?
>

Barry, if you're game to try, do you want to test with SM to see what
you get? From there, perhaps it may be easier to nail down what's
different between the two apps.

Barry Landy

unread,
May 15, 2012, 2:59:31 PM5/15/12
to
On Tue, 15 May 2012, Lewis Rosenthal wrote:

:>On 05/15/12 12:59 pm, Lewis Rosenthal thus wrote :
I dont have SM installed. Will think about it.

Dave Yeo

unread,
May 16, 2012, 1:59:11 AM5/16/12
to
Barry Landy wrote:
> FF6 traps in PMMERGE and I have emailed the crash toc...@e-vertise.com
>
> FF8 also crashed but FF4 (from ECS21 distro) did not.

FF6 was the last one that Rich worked on and as he couldn't figure out
the crash (according to bug#682952) I doubt that I can. (Rich wrote the
code) If I find the time I'll try to diff them but if the changes
weren't in OS/2 specific code I doubt that I can figure it out.
Dave

Dave Yeo

unread,
May 16, 2012, 2:04:54 AM5/16/12
to
Lewis Rosenthal wrote:
> On 05/15/12 12:59 pm, Lewis Rosenthal thus wrote :
>> On 05/15/12 12:23 pm, Barry Landy thus wrote :
>>> On Tue, 15 May 2012, Dave Yeo wrote:
>>>
>>> :>Barry Landy wrote:
>>> :>> On Mon, 14 May 2012, Dave Yeo wrote:
>>> :>>
>>> :>> :>Barry Landy wrote:
>>> :>> :>> FWIW I tried the same print in version 8.0.1 and got the same
>>> trap.
>>> :>> :>
>>> :>> :>Can you test with 4 and maybe 6?
>>> :>> :>Dave
>>> :>> :>
>>> :>>
>>> :>> I wanted to do that and couldnt find the binaries. Perhaps you can
>>> give
>>> :>> me a URL.
>>> :>
>>> :>They were uploaded to the Mozilla ftp server.
>>>
>>> Couldnt find them there but found copies on my office computer.
>>>
>>> FF6 traps in PMMERGE and I have emailed the crash to cr...@e-vertise.com
>>>
>>> FF8 also crashed but FF4 (from ECS21 distro) did not.
>>>
>> I set up a virtual HP 5L printer with the 30.827 LaserJet driver,
>> directed to print to file. Printing from both SM 2.7.2.4esr and FF
>> 10.0.4esr yielded no crashes and clearly legible and complete PCL files.
>>
>
> I lied. I mistakenly viewed the same output file. In fact, what hapened
> was that SM output a file but FF did not (zero-length; I just tried
> creating a file of a completely different name).

Can you compare your prefs.js. Also about:support will show any changed
prefs. All the printing code is shared between FF and SM so I'd expect
the same results.
It wouldn't hurt for Barry and perhaps you to try printing from a virgin
profile as well.
[...]

Dave

Barry Landy

unread,
May 16, 2012, 3:18:35 AM5/16/12
to
On Tue, 15 May 2012, Dave Yeo wrote:

:>

good thinking but it made no difference. Still a C05 trap in PMMERGE

Lewis Rosenthal

unread,
May 16, 2012, 10:59:15 AM5/16/12
to
On 05/16/12 03:18 am, Barry Landy thus wrote :
> On Tue, 15 May 2012, Dave Yeo wrote:
>
> :>Lewis Rosenthal wrote:
> :>> On 05/15/12 12:59 pm, Lewis Rosenthal thus wrote :
> :>> > On 05/15/12 12:23 pm, Barry Landy thus wrote :
> :>> > > On Tue, 15 May 2012, Dave Yeo wrote:
> :>> > >
> :>> > > :>Barry Landy wrote:
> :>> > > :>> On Mon, 14 May 2012, Dave Yeo wrote:
> :>> > > :>>
> :>> > > :>> :>Barry Landy wrote:
<snip>
> :>> > >
> :>> > > FF6 traps in PMMERGE and I have emailed the crash to cr...@e-vertise.com
> :>> > >
> :>> > > FF8 also crashed but FF4 (from ECS21 distro) did not.
> :>> > >
> :>> > I set up a virtual HP 5L printer with the 30.827 LaserJet driver,
> :>> > directed to print to file. Printing from both SM 2.7.2.4esr and FF
> :>> > 10.0.4esr yielded no crashes and clearly legible and complete PCL files.
> :>> >
> :>>
> :>> I lied. I mistakenly viewed the same output file. In fact, what happened
> :>> was that SM output a file but FF did not (zero-length; I just tried
> :>> creating a file of a completely different name).
> :>
> :>Can you compare your prefs.js. Also about:support will show any changed prefs.
> :>All the printing code is shared between FF and SM so I'd expect the same
> :>results.
> :>It wouldn't hurt for Barry and perhaps you to try printing from a virgin
> :>profile as well.
> :>[...]
> :>
> :>Dave
> :>
>
> good thinking but it made no difference. Still a C05 trap in PMMERGE
>
Tried with two virgin, default profiles.

about:support for FF shows:



Application Basics

Name
Firefox

Version
10.0.4

User Agent
Mozilla/5.0 (OS/2; Warp 4.5; rv:10.0.4) Gecko/20120422
Firefox/10.0.4

Profile Directory

Open Containing Folder

Enabled Plugins

about:plugins

Build Configuration

about:buildconfig

Crash Reports

about:crashes

Memory Use

about:memory

Extensions

Name

Version

Enabled

ID

Modified Preferences

Name

Value

browser.places.smartBookmarksVersion
2

browser.startup.homepage_override.buildID
20120422120931

browser.startup.homepage_override.mstone
rv:10.0.4

extensions.lastAppVersion
10.0.4

network.cookie.prefsMigrated
true

places.history.expiration.transient_current_max_pages
51969

print.os2.postscript.use_builtin
false

print.printer_HP Color LaserJet 3800 PS - LPR.printer_description
HP Color LaserJet 3700 PS (PSCRIPT)

print.printer_HP Color LaserJet 3800 PS.printer_description
HP Color LaserJet 3700 PS (PSCRIPT)

print.printer_HP LaserJet 5L.printer_description
HP LaserJet 5/5M (LASERJET)

print.printer_HP LaserJet 9000 MFP PCL 6.printer_description
HP LaserJet 9000L MFP (LASERJET)

print.printer_HP LaserJet 9000 MFP PS.printer_description
HP LaserJet 9000L MFP PS (PSCRIPT)

print.printer_HP LaserJet 9000 PS - LPR.printer_description
HP LaserJet 9000 PS (PSCRIPT)

print.printer_HP Mopier 320 PCL 6.printer_description
HP Mopier 320 (LASERJET)

print.printer_HP Mopier 320 PS - VA LPR.printer_description
HP Mopier 320 (PSCRIPT)

print.printer_HP Mopier 320 PS.printer_description
HP Mopier 320 (PSCRIPT)

print.printer_HP PhotoSmart D7260.printer_description
HP PhotoSmart 7150 (PSCRIPT)

print.printer_PDF Writer.printer_description
Apple Color LaserWriter 12/600 (PSCRIPT)

privacy.sanitize.migrateFx3Prefs
true

Graphics

and for SM:

Application Basics

Name
SeaMonkey

Version
2.7.2

User Agent
Mozilla/5.0 (OS/2; Warp 4.5; rv:10.0.4esrpre) Gecko/20120407
Firefox/10.0.4esrpre SeaMonkey/2.7.2

Profile Directory

Open Containing Folder

Enabled Plugins

about:plugins

Build Configuration

about:buildconfig

Crash Reports

about:crashes

Memory Use

about:memory

Extensions

Name

Version

Enabled

ID

ChatZilla
0.9.88.2
true
{59c81df5-4b7a-477b-912d-4e0fdf64e5f2}

DOM Inspector
2.0.12pre
true
insp...@mozilla.org

JavaScript Debugger
0.9.89
true
{f13b157f-b174-47e7-a34d-4815ddfdfeb8}

Modified Preferences

Name

Value

browser.places.smartBookmarksVersion
2

browser.startup.homepage_override.mstone
rv:10.0.4esrpre

extensions.lastAppVersion
2.7.2

network.cookie.prefsMigrated
true

places.history.expiration.transient_current_max_pages
51969

print.os2.postscript.use_builtin
false

print.printer_HP Color LaserJet 3800 PS - LPR.printer_description
HP Color LaserJet 3700 PS (PSCRIPT)

print.printer_HP Color LaserJet 3800 PS.printer_description
HP Color LaserJet 3700 PS (PSCRIPT)

print.printer_HP LaserJet 5L.printer_description
HP LaserJet 5/5M (LASERJET)

print.printer_HP LaserJet 9000 MFP PCL 6.printer_description
HP LaserJet 9000L MFP (LASERJET)

print.printer_HP LaserJet 9000 MFP PS.printer_description
HP LaserJet 9000L MFP PS (PSCRIPT)

print.printer_HP LaserJet 9000 PS - LPR.printer_description
HP LaserJet 9000 PS (PSCRIPT)

print.printer_HP Mopier 320 PCL 6.printer_description
HP Mopier 320 (LASERJET)

print.printer_HP Mopier 320 PS - VA LPR.printer_description
HP Mopier 320 (PSCRIPT)

print.printer_HP Mopier 320 PS.printer_description
HP Mopier 320 (PSCRIPT)

print.printer_HP PhotoSmart D7260.printer_description
HP PhotoSmart 7150 (PSCRIPT)

print.printer_PDF Writer.printer_description
Apple Color LaserWriter 12/600 (PSCRIPT)

Graphics

No matter what I do, I can't get any data in the output file from FF,
but it consistently generates from SM. I do not get a crash in FF,
though. I also haven't tried TB.

I tried toggling print.os2.postscript.use_builtin to true in FF &
restarting; this made no difference (as expected).

Barry Landy

unread,
May 16, 2012, 11:08:37 AM5/16/12
to
On Wed, 16 May 2012, Lewis Rosenthal wrote:

:>On 05/16/12 03:18 am, Barry Landy thus wrote :
:>
:>

Mine has basically the same stuff with (obviously) different printer
things.

trying the print experiment stuffs up my printer until reboot. I dont
know why. (neither cancelling not switching the printer off and on
improve things).

Dave Yeo

unread,
May 16, 2012, 11:21:03 AM5/16/12
to
Barry Landy wrote:
> trying the print experiment stuffs up my printer until reboot. I dont
> know why. (neither cancelling not switching the printer off and on
> improve things).

Anything stuck in your printer spool?
Dave

Barry Landy

unread,
May 16, 2012, 3:06:39 PM5/16/12
to
On Wed, 16 May 2012, Dave Yeo wrote:
:>

No. It was all deleted.
Anyway if there was rebooting would not have helped

Lewis Rosenthal

unread,
May 16, 2012, 3:59:44 PM5/16/12
to
On 05/16/12 03:06 pm, Barry Landy thus wrote :
> On Wed, 16 May 2012, Dave Yeo wrote:
>
> :>Barry Landy wrote:
> :>> trying the print experiment stuffs up my printer until reboot. I dont
> :>> know why. (neither cancelling not switching the printer off and on
> :>> improve things).
> :>
> :>Anything stuck in your printer spool?
> :>Dave
> :>
>
> No. It was all deleted.
> Anyway if there was rebooting would not have helped
>

I just changed my printer object to point to LPT3, which is captured to
a NetWare queue and redirected to a Color LaserJet 3800 (which handles
PCL as well as PostScript). Sending the print job to the printer this
time instead of to a file caused me to have to answer a prompt at the
printer to override the paper size, even though looking at the printer
properties, Letter is selected, but the second job went through right
away. I'll try FF next.

Lewis Rosenthal

unread,
May 16, 2012, 4:15:40 PM5/16/12
to
On 05/16/12 03:59 pm, Lewis Rosenthal thus wrote :
> On 05/16/12 03:06 pm, Barry Landy thus wrote :
>> On Wed, 16 May 2012, Dave Yeo wrote:
>>
>> :>Barry Landy wrote:
>> :>> trying the print experiment stuffs up my printer until reboot. I dont
>> :>> know why. (neither cancelling not switching the printer off and on
>> :>> improve things).
>> :>
>> :>Anything stuck in your printer spool?
>> :>Dave
>> :>
>>
>> No. It was all deleted.
>> Anyway if there was rebooting would not have helped
>>
>
> I just changed my printer object to point to LPT3, which is captured to
> a NetWare queue and redirected to a Color LaserJet 3800 (which handles
> PCL as well as PostScript). Sending the print job to the printer this
> time instead of to a file caused me to have to answer a prompt at the
> printer to override the paper size, even though looking at the printer
> properties, Letter is selected, but the second job went through right
> away. I'll try FF next.
>

Some consistency, at least, and more puzzlement:

FF produced no output whatsoever, which is consistent with the
zero-length PCL file it was producing.

TB, however, crashed with what I believe is the same symptom you've been
seeing, Barry. TRP file is here:

ftp://ftp.2rosenthals.com/pub/Mozilla/Thunderbird/exceptq/0087_01.TRP

Thunderbird v10.0.3 - build 20120410192118

However, printing to an HP LaserJet Mopier 320, using the *same*
LaserJet driver (30.827), directed to a parallel port and similarly
captured, prints (from TB) without error and no crash.

How about them apples?

Barry Landy

unread,
May 16, 2012, 4:22:05 PM5/16/12
to
On Wed, 16 May 2012, Lewis Rosenthal wrote:

:>On 05/16/12 03:59 pm, Lewis Rosenthal thus wrote :
rotten apples!

That is certainly the same as one of my crashes (some are illegal reads
and some writes).

I notice you havenot installed the sym tables as requested by Steven.

-----

Barry and others, please grab copies of the .xqs files you don't have
from the zip files at

http://www.warpcave.com/symbols

For exceptq .trp reports you only need to install the .xqs files. As
with all .xqs files, install the .xqs file in the directory that
contains the corresponding DLL.

------------

Lewis Rosenthal

unread,
May 16, 2012, 6:05:38 PM5/16/12
to
On 05/16/12 04:22 pm, Barry Landy thus wrote :
Indeed!

I also needed to upgrade TB to the latest build (I don't normally use it
for things other than testing, so I don't keep as current with it as I
should).

So, new version, symbols in place, fresh TRP file:

ftp://ftp.2rosenthals.com/ftp/pub/Mozilla/Thunderbird/exceptq/00B8_01.TRP

Now to figure out why SM works & FF does nothing. ;-)

Lewis Rosenthal

unread,
May 16, 2012, 6:35:54 PM5/16/12
to
On 05/16/12 06:05 pm, Lewis Rosenthal thus wrote :
> On 05/16/12 04:22 pm, Barry Landy thus wrote :
>> On Wed, 16 May 2012, Lewis Rosenthal wrote:
>>
>> :>On 05/16/12 03:59 pm, Lewis Rosenthal thus wrote :

<snip>

>> :>> I just changed my printer object to point to LPT3, which is
>> captured to
>> :>> a NetWare queue and redirected to a Color LaserJet 3800 (which
>> handles
>> :>> PCL as well as PostScript). Sending the print job to the printer this
>> :>> time instead of to a file caused me to have to answer a prompt at the
>> :>> printer to override the paper size, even though looking at the
>> printer
>> :>> properties, Letter is selected, but the second job went through right
>> :>> away. I'll try FF next.
>> :>>
<snip>

I removed all plugins from MOZ_PLUG & started FF again (safe mode did
not want to start in safe mode, even after accepting the dialog that it
was supposed to be in safe mode - but that's for another discussion).

From FF, the best I get to the Mopier 320 PCL object is a blink on the
data light at the printer (no output). PostScript seems to do fine.

So, the trend I'm seeing, at least is:

- consistent output from SM, using either PCL or PS
- consistent output from FF, using PS
- consistent output from TB, using PS
- consistent crash from TB, using PCL to LJ5
- consistent output from TB, using PCL to Mopier 320 (same driver)
- no output from FF, using PCL to LJ5
- garbage output from FF, using PCL to Mopier 320

The garbage output, directed to a file, when printing about: yields the
file located here:

ftp://ftp.2rosenthals.com/ftp/pub/Mozilla/Firefox/printing/PCL/about-page-ff10-0-4esr-PCL-mopier-320.pcl

(not all garbage characters would copy to the clipboard)

whereas about: printed under SM to the same printer object yields:

ftp://ftp.2rosenthals.com/ftp/pub/Mozilla/Firefox/printing/PCL/about-page-sm2-7-2-4esr-mopier-320.pcl

Hmmm...

Steven Levine

unread,
May 16, 2012, 8:41:58 PM5/16/12
to
On Wed, 16 May 2012 22:05:38 UTC, Lewis Rosenthal
<lgros...@2-de-sp-am-2rosenthals.com> wrote:

Hi Lewis,

>
> So, new version, symbols in place, fresh TRP file:
>
> ftp://ftp.2rosenthals.com/ftp/pub/Mozilla/Thunderbird/exceptq/00B8_01.TRP

Although it is not exactly the same trap as Barry's it does look
similar enough to warrant some looking.

I recommend you grab the symbols you don't have from
home.earthlink.net/~steve53/symbols.

Is this the system that's running the 14.095 pmmerge? If so, we
should either track down a copy of the .sym files or convince you to
update to 14.106.

Can you disable to automatic print to file pdf conversion get the trap
to occur with a print to file?

I suspect that eventually, I am going to need to see a dump file to
understand exactly how this is failing, but we need to collect some
more data first.

Steven


--
---------------------------------------------------------------------
Steven Levine <ste...@earthlink.bogus.net>
eCS/Warp/DIY etc. www.scoug.com www.ecomstation.com
---------------------------------------------------------------------

Steven Levine

unread,
May 16, 2012, 11:34:14 PM5/16/12
to
On Thu, 17 May 2012 00:41:58 UTC, "Steven Levine"
<ste...@nomail.earthlink.net> wrote:

Hi,

> > So, new version, symbols in place, fresh TRP file:
> >
> > ftp://ftp.2rosenthals.com/ftp/pub/Mozilla/Thunderbird/exceptq/00B8_01.TRP

I can replicate this with Firefox

http://home.earthlink.net/~steve53/mozilla/0037_01.TRP

If I believe the symbol decode, it appears that the Laserjet driver
has a problem drawing splines or possibly ovals. The exception occurs
only when printing directly. If I print to a file and drop the
resulting file on the printer, it seems to print OK.

Lewis Rosenthal

unread,
May 17, 2012, 12:59:05 AM5/17/12
to
Hey, Steve...

On 05/16/12 08:41 pm, Steven Levine thus wrote :
> On Wed, 16 May 2012 22:05:38 UTC, Lewis Rosenthal
> <lgros...@2-de-sp-am-2rosenthals.com> wrote:
>
> Hi Lewis,
>
>>
>> So, new version, symbols in place, fresh TRP file:
>>
>> ftp://ftp.2rosenthals.com/ftp/pub/Mozilla/Thunderbird/exceptq/00B8_01.TRP
>
> Although it is not exactly the same trap as Barry's it does look
> similar enough to warrant some looking.
>
> I recommend you grab the symbols you don't have from
> home.earthlink.net/~steve53/symbols.
>
> Is this the system that's running the 14.095 pmmerge? If so, we
> should either track down a copy of the .sym files or convince you to
> update to 14.106.
>
LOL... Well I would imagine that we would get more useful information if
the symbols matched the dll... ;-) I really do need to upgrade this
machine soon, too. Anyway, I just bumped PMMERGE to 14.106.

> Can you disable to automatic print to file pdf conversion get the trap
> to occur with a print to file?
>
Now that you have been able to replicate the crash, do you still need
this from me? I'm happy to keep going, though .

> I suspect that eventually, I am going to need to see a dump file to
> understand exactly how this is failing, but we need to collect some
> more data first.
>
Just let me know how I may help. I find it interesting that I can get
zero output in one instance and garbage elsewhere, while utilizing the
same driver yet again for another printer model, it behaves as expected.

Cheers/2

Dave Yeo

unread,
May 17, 2012, 2:08:39 AM5/17/12
to
Lewis Rosenthal wrote:
> Now to figure out why SM works & FF does nothing.

This is the really weird part as the code is shared between the two.
Dave

Barry Landy

unread,
May 17, 2012, 2:13:01 AM5/17/12
to
On Wed, 16 May 2012, Steven Levine wrote:

:>On Thu, 17 May 2012 00:41:58 UTC, "Steven Levine"
:><ste...@nomail.earthlink.net> wrote:
:>
:>Hi,
:>
:>> > So, new version, symbols in place, fresh TRP file:
:>> >
:>> > ftp://ftp.2rosenthals.com/ftp/pub/Mozilla/Thunderbird/exceptq/00B8_01.TRP
:>
:>I can replicate this with Firefox
:>
:> http://home.earthlink.net/~steve53/mozilla/0037_01.TRP
:>
:>If I believe the symbol decode, it appears that the Laserjet driver
:>has a problem drawing splines or possibly ovals. The exception occurs
:>only when printing directly. If I print to a file and drop the
:>resulting file on the printer, it seems to print OK.
:>

great that you can replicate it.

I have no joy if I print to file (works fine) and then drop the file on
the printer - the printer light (usually) just flashes.

It should be noted that printing from FF4 works just fine... maybe the
change to cairo is drawing things differently and triggering some bug?

Steven Levine

unread,
May 17, 2012, 3:07:41 AM5/17/12
to
On Thu, 17 May 2012 04:59:05 UTC, Lewis Rosenthal
<lgros...@2-de-sp-am-2rosenthals.com> wrote:

Hi Lewis,


> LOL... Well I would imagine that we would get more useful information if
> the symbols matched the dll... ;-)

Having symbols helps. :-) If the code does not change all that much,
I can get by with the symbols being a little off.

> Now that you have been able to replicate the crash, do you still need
> this from me?

Not at the moment. I'll generate a process dump and see what I can
see. I'll probaby also give the PM debugger a try.

> Just let me know how I may help. I find it interesting that I can get
> zero output in one instance and garbage elsewhere, while utilizing the
> same driver yet again for another printer model, it behaves as expected.

I suspect it has to do with sizing things to fit.

What I find interesting is that printing to a file does not trap.
Perhaps there is a difference in the presentation spaces used for
direct and to file printing.

Steve Wendt

unread,
May 17, 2012, 1:28:29 PM5/17/12
to
On 5/16/2012 11:13 PM, Barry Landy wrote:

> It should be noted that printing from FF4 works just fine... maybe the
> change to cairo is drawing things differently and triggering some bug?

Cairo has been used for a *long* time. You may recall that there was no
direct printing support at all for Firefox 3.0.x through 3.6.x.

Barry Landy

unread,
May 17, 2012, 3:03:31 PM5/17/12
to
On Thu, 17 May 2012, Steve Wendt wrote:
:>

yes indeed. Quite a surprise when I first used V3
"where did the print go?"
- took a google search before I realised I had to look for something on
te desktop :-)
0 new messages