how to fix a powerpc bug... on intel

"Gregory Weston" <uce DeleteThis @splook.com> a écrit dans le message de news:
uce-4373B4.10221426052008 DeleteThis @newsclstr02.news.prodigy.com...
> In article <483ac183$0$890$ba4acef3@news.orange.fr>,
> "Armel" <armelasselin DeleteThis @hotmail.com> wrote:
>
>> Hello,
>>
>> I have here only Intel Mac OS X... and build for PowerPC as well.
>> The PowerPC build works as expected on the Intel (thanks to Rosetta
>> emulation I suppose)... but it does not work on a PowerPC
>
> Are you sure about that? You're not building a Universal binary? Are you
> sure you're building a deployment release, instead of a debug build that
> will only work on *your* specific machine?
I build PPC and Intel separately (the exe is quite big).
about the deployment release with respect to debug release... I have been
quite careful to not use anything specific from my own machine, and the
Intel build which differ only by the '-arch ppc' works well on other Intel
machines.
Are there any explicit command to run?

>
>> Is the only solution to test on a true PowerPC, or is there a possibility
>> with an emulator (which?)
>> if the only way is the true-PowerPC way, is there any place to find
>> PowerPC
>> based Macs [i'm in France]?
>
> Find a developer with a PPC machine who can test for you or even let you
> use his/her machine.
not easy

Regards
Armel

In article <483acc4d$0$871$ba4acef3@news.orange.fr>,
"Armel" <armelasselin.RemoveThis@hotmail.com> wrote:

> "Gregory Weston" <uce.RemoveThis@splook.com> a écrit dans le message de news:
> uce-4373B4.10221426052008.RemoveThis@newsclstr02.news.prodigy.com...
> > In article <483ac183$0$890$ba4acef3@news.orange.fr>,
> > "Armel" <armelasselin.RemoveThis@hotmail.com> wrote:
> >
> >> Hello,
> >>
> >> I have here only Intel Mac OS X... and build for PowerPC as well.
> >> The PowerPC build works as expected on the Intel (thanks to Rosetta
> >> emulation I suppose)... but it does not work on a PowerPC
> >
> > Are you sure about that? You're not building a Universal binary? Are you
> > sure you're building a deployment release, instead of a debug build that
> > will only work on *your* specific machine?
> I build PPC and Intel separately (the exe is quite big).
> about the deployment release with respect to debug release... I have been
> quite careful to not use anything specific from my own machine,

Yes, but depending on the tools you're using and how they're configured,
it's possible to generate a build that does make assumptions the
ultimately restrict it to your machine even if your code is the most
beautiful and robust thing ever written. It's even the default behavior
for debug builds in certain versions of XCode.

> and the
> Intel build which differ only by the '-arch ppc' works well on other Intel
> machines.
> Are there any explicit command to run?
>
> >
> >> Is the only solution to test on a true PowerPC, or is there a possibility
> >> with an emulator (which?)
> >> if the only way is the true-PowerPC way, is there any place to find
> >> PowerPC
> >> based Macs [i'm in France]?
> >
> > Find a developer with a PPC machine who can test for you or even let you
> > use his/her machine.
> not easy

Ultimately, you should never claim to support a machine you haven't
tested on (or at least something moderately close). If you have no
access to a real PPC, I wouldn't bother deploying a PPC build. But I
still consider it very strange that something would run under Rosetta on
an Intel box and fail on a real PowerMac.

What's the nature of "does not work?"

G

--
"Harry?" Ron's voice was a mere whisper. "Do you smell something ... burning?"
- Harry Potter and the Odor of the Phoenix

"Gregory Weston" <uce.RemoveThis@splook.com> a écrit dans le message de news:
uce-475988.13194926052008.RemoveThis@newsclstr03.news.prodigy.net...
> In article <483acc4d$0$871$ba4acef3@news.orange.fr>,
> "Armel" <armelasselin.RemoveThis@hotmail.com> wrote:
>
>> "Gregory Weston" <uce.RemoveThis@splook.com> a écrit dans le message de news:
>> uce-4373B4.10221426052008.RemoveThis@newsclstr02.news.prodigy.com...
>> > In article <483ac183$0$890$ba4acef3@news.orange.fr>,
>> > "Armel" <armelasselin.RemoveThis@hotmail.com> wrote:
>> >
>> >> Hello,
>> >>
>> >> I have here only Intel Mac OS X... and build for PowerPC as well.
>> >> The PowerPC build works as expected on the Intel (thanks to Rosetta
>> >> emulation I suppose)... but it does not work on a PowerPC
>> >
>> > Are you sure about that? You're not building a Universal binary? Are
>> > you
>> > sure you're building a deployment release, instead of a debug build
>> > that
>> > will only work on *your* specific machine?
>> I build PPC and Intel separately (the exe is quite big).
>> about the deployment release with respect to debug release... I have been
>> quite careful to not use anything specific from my own machine,
>
> Yes, but depending on the tools you're using and how they're configured,
> it's possible to generate a build that does make assumptions the
> ultimately restrict it to your machine even if your code is the most
> beautiful and robust thing ever written. It's even the default behavior
> for debug builds in certain versions of XCode.
>
>> and the
>> Intel build which differ only by the '-arch ppc' works well on other
>> Intel
>> machines.
>> Are there any explicit command to run?
>>
>> >
>> >> Is the only solution to test on a true PowerPC, or is there a
>> >> possibility
>> >> with an emulator (which?)
>> >> if the only way is the true-PowerPC way, is there any place to find
>> >> PowerPC
>> >> based Macs [i'm in France]?
>> >
>> > Find a developer with a PPC machine who can test for you or even let
>> > you
>> > use his/her machine.
>> not easy
>
> Ultimately, you should never claim to support a machine you haven't
> tested on (or at least something moderately close). If you have no
> access to a real PPC, I wouldn't bother deploying a PPC build. But I
> still consider it very strange that something would run under Rosetta on
> an Intel box and fail on a real PowerMac.
memory alignment problems won't be catched by an emulator optimized for
speed (PowerPC can access packets of more than 8 bits only on boundaries of
the given packets size, whereas Intel processors can access memory without
alignment, even if it's a bit slower than when acces is aligned).

> What's the nature of "does not work?"
that's my exact problem... as we have no PowerPC we decide to not support it
at all. But as some customers asked again and again to have a PowerPC build
we did one and asked them to test the program... unfortunately the only
information that I coudl get from them is 'it does not run'.

if I could put a hand on a true machine and run a few commands, i'm pretty
sure that I can could determine in a few minutes or hours what is wrong...
maybe it's just a dylib missing or an entry point, maybe it's a memory
alignment problem, just running otool or gdb could tell, but without a
machine, i'm stuck.

Armel

In article <483b1332$0$928$ba4acef3@news.orange.fr>,
"Armel" <armelasselin.DeleteThis@hotmail.com> wrote:

> > Ultimately, you should never claim to support a machine you haven't
> > tested on (or at least something moderately close). If you have no
> > access to a real PPC, I wouldn't bother deploying a PPC build. But I
> > still consider it very strange that something would run under Rosetta on
> > an Intel box and fail on a real PowerMac.
> memory alignment problems won't be catched by an emulator optimized for
> speed (PowerPC can access packets of more than 8 bits only on boundaries of
> the given packets size, whereas Intel processors can access memory without
> alignment, even if it's a bit slower than when acces is aligned).
>
> > What's the nature of "does not work?"
> that's my exact problem... as we have no PowerPC we decide to not support it
> at all. But as some customers asked again and again to have a PowerPC build
> we did one and asked them to test the program... unfortunately the only
> information that I coudl get from them is 'it does not run'.

"Does not run" actually narrows down the scope of the problem a lot. It
is, for example, a much smaller problem space than a post-launch crash.

Around here, a G4 mini isn't that difficult to buy used, and can be
fairly cheap. What's the used market like in your area?

> if I could put a hand on a true machine and run a few commands, i'm pretty
> sure that I can could determine in a few minutes or hours what is wrong...
> maybe it's just a dylib missing or an entry point, maybe it's a memory
> alignment problem, just running otool or gdb could tell, but without a
> machine, i'm stuck.

Ask them to examine the system and/or console log files as they attempt
to run it.

--
"Harry?" Ron's voice was a mere whisper. "Do you smell something ... burning?"
- Harry Potter and the Odor of the Phoenix

"Gregory Weston" <uce.DeleteThis@splook.com> a écrit dans le message de news:
uce-968385.21030926052008.DeleteThis@newsclstr03.news.prodigy.net...
> In article <483b1332$0$928$ba4acef3@news.orange.fr>,
> "Armel" <armelasselin.DeleteThis@hotmail.com> wrote:
>
>> > Ultimately, you should never claim to support a machine you haven't
>> > tested on (or at least something moderately close). If you have no
>> > access to a real PPC, I wouldn't bother deploying a PPC build. But I
>> > still consider it very strange that something would run under Rosetta
>> > on
>> > an Intel box and fail on a real PowerMac.
>> memory alignment problems won't be catched by an emulator optimized for
>> speed (PowerPC can access packets of more than 8 bits only on boundaries
>> of
>> the given packets size, whereas Intel processors can access memory
>> without
>> alignment, even if it's a bit slower than when acces is aligned).
>>
>> > What's the nature of "does not work?"
>> that's my exact problem... as we have no PowerPC we decide to not support
>> it
>> at all. But as some customers asked again and again to have a PowerPC
>> build
>> we did one and asked them to test the program... unfortunately the only
>> information that I coudl get from them is 'it does not run'.
>
> "Does not run" actually narrows down the scope of the problem a lot. It
> is, for example, a much smaller problem space than a post-launch crash.
>
> Around here, a G4 mini isn't that difficult to buy used, and can be
> fairly cheap. What's the used market like in your area?
>
>> if I could put a hand on a true machine and run a few commands, i'm
>> pretty
>> sure that I can could determine in a few minutes or hours what is
>> wrong...
>> maybe it's just a dylib missing or an entry point, maybe it's a memory
>> alignment problem, just running otool or gdb could tell, but without a
>> machine, i'm stuck.
>
> Ask them to examine the system and/or console log files as they attempt
> to run it.
that's what I'll try to do.

I dumped otool -L for my executable:

/System/Library/Frameworks/QuickTime.framework/Versions/A/QuickTime
(compatibility version 1.0.0, current version 50.0.0)
/System/Library/Frameworks/IOKit.framework/Versions/A/IOKit (compatibility
version 1.0.0, current version 275.0.0)
/System/Library/Frameworks/Carbon.framework/Versions/A/Carbon
(compatibility version 2.0.0, current version 136.0.0)
/System/Library/Frameworks/Cocoa.framework/Versions/A/Cocoa (compatibility
version 1.0.0, current version 12.0.0)
/usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current version
111.0.0)
/System/Library/Frameworks/WebKit.framework/Versions/A/WebKit
(compatibility version 1.0.0, current version 525.13.0)
/usr/lib/libiconv.2.dylib (compatibility version 7.0.0, current version
7.0.0)
/usr/lib/libcharset.1.dylib (compatibility version 2.0.0, current version
2.0.0)
@loader_path/libcurl.3.dylib (compatibility version 4.0.0, current version
4.0.0)
/usr/lib/libstdc++.6.dylib (compatibility version 7.0.0, current version
7.4.0)
/usr/lib/libgcc_s.1.dylib (compatibility version 1.0.0, current version
1.0.0)
/System/Library/Frameworks/CoreServices.framework/Versions/A/CoreServices
(compatibility version 1.0.0, current version 32.0.0)
/System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation(compatibility version 150.0.0, current version 476.10.0) /System/Library/Frameworks/ApplicationServices.framework/Versions/A/ApplicationServices (compatibility version 1.0.0, current version 34.0.0)can I determine from that if my app will load on a 10.4 PPC Mac OS X ?RegardsArmel