nmedit on ppc64

Thread view:

Enable EMail Alerts

Start New Thread

Thread rating:

John Smith - 18 Jan 2006 17:09 GMT

I'm having a problem with nmedit when cross compiling for ppc64.

Whenever I use it with "-s" option I get

nmedit: don't have code to convert external relocation entry 0 in section
(__DATA,__const_coal) for global coalesced symbol: __ZTI15MBOGXDWOGPLCENX
in: Darwin_ppc64/to_be_fixed.o
makepp: error: target(s) failed

My best guess is that it tells me that it doesn't support ppc64. I couldn't
find anything on apple website or google though.

Can anyone tell more information? (The corresponding action worked perfectly
when using it for cross compiling for i386.)

Thanks in advance.

Reply to this Message

Eric Albert - 19 Jan 2006 03:20 GMT

> I'm having a problem with nmedit when cross compiling for ppc64.
>
[quoted text clipped - 10 lines]
> Can anyone tell more information? (The corresponding action worked perfectly
> when using it for cross compiling for i386.)

I think it should work. File a bug at <http://bugreport.apple.com/> and
attach your binary.

-Eric

Signature

Eric Albert ejalbert@cs.stanford.edu
http://outofcheese.org/

Reply to this Message

John Smith - 27 Jan 2006 09:34 GMT

> I think it should work. File a bug at <http://bugreport.apple.com/> and
> attach your binary.

Apple has the most slow bug report service in the entire world it seems.
I've posted this bug one week ago now and it still says open. They didn't
even bother to give any status back. Really useless. Do they ever listen to
their users?
Last time I had a bug report it took 2 months to find out the bug was a
duplicate and I had to email them several times because they didn't even
bother to tell me what the status was for the first report.

I wonder how many months I have to wait for getting an answer on this one.

Microsoft did the right thing and went open with the product feedback
report. You can see all status online and query to see if others has similar
issues.

-- John

Reply to this Message

Patrick Machielse - 27 Jan 2006 10:48 GMT

> > I think it should work. File a bug at <http://bugreport.apple.com/> and
> > attach your binary.
[quoted text clipped - 3 lines]
> even bother to give any status back. Really useless. Do they ever listen to
> their users?

Recently, they have been getting back to me within a few days (for
'duplicates'). However, my oldest 'open' bug is > 2 years old by now.

> Microsoft did the right thing and went open with the product feedback
> report. You can see all status online and query to see if others has similar
> issues.

Speaking for myself, I'd say we all want such a system. ;-)

After each 'duplicate' report, I feel Apple has stolen a bit of my life

:-(

patrick

Reply to this Message

Eric Albert - 28 Jan 2006 03:59 GMT

> > > I think it should work. File a bug at <http://bugreport.apple.com/> and
> > > attach your binary.
[quoted text clipped - 15 lines]
> After each 'duplicate' report, I feel Apple has stolen a bit of my life
> :-(

Duplicates are actually quite useful. It's one thing to know that one
developer has hit a problem; it's another to know that ten developers
care about it.

-Eric

Signature

Eric Albert ejalbert@cs.stanford.edu
http://outofcheese.org/

Reply to this Message

Chris Hanson - 30 Jan 2006 06:26 GMT

> After each 'duplicate' report, I feel Apple has stolen a bit of my life
> :-(

In addition to what Eric said, duplicates are also useful because a
duplicate might have a clearer test case, or it might contain that
crucial piece of information that leads to resolution. Duplicate is
absolutely not a "black hole" resolution.

-- Chris

Reply to this Message

Patrick Machielse - 30 Jan 2006 10:24 GMT

> > After each 'duplicate' report, I feel Apple has stolen a bit of my life
> > :-(
[quoted text clipped - 3 lines]
> crucial piece of information that leads to resolution. Duplicate is
> absolutely not a "black hole" resolution.

I realise that, but it still makes me _feel_ really stupid for
submitting it. And let's not forget the possibility that the bug may
already have been fixed while I'm carefully assembling my test case.
Thank you for submitting this bug report...

Apple has one of the most secretive and restrictive bug tracking systems
in place anywhere. It would be _really nice_ if you could check the
bug-database whenever you have a strong indication that you stumbled on
some issue. It could even suggest a workaround.

control-option-command-tripleclick on a class name in Xcode to open a
web page of known issues. There you have it: another feature request.

:-)

patrick

Reply to this Message

Michael Ash - 27 Jan 2006 21:32 GMT

>> I think it should work. File a bug at <http://bugreport.apple.com/> and
>> attach your binary.
[quoted text clipped - 6 lines]
> duplicate and I had to email them several times because they didn't even
> bother to tell me what the status was for the first report.

Report a bug with the idea that you're informing Apple, and letting them
decide what to do with it from there. Don't report it with the expectation
that it will be fixed immediately, or that you'll get any feedback
whatsoever about the status of the fix, as you'll just be setting yourself
up for disappointment. This is unfortunate but true. It *is* useful to
submit bug reports, and Apple does read them, but don't expect to find out
anything related to your particular bug until the fix ships.

Signature

Michael Ash
Rogue Amoeba Software

Reply to this Message

John Smith - 28 Jan 2006 21:52 GMT

> but don't expect to find out anything related to your particular bug until
> the fix ships.

... which is between now and never. If they don't bother to tell you updates
then why would I want to waste my time on reporting more bugs?

This problem I mentioned here prevents me from compiling our software for
ppc64. I cannot tell my customers to wait at least 6 months for a fix
because I don't know when or if Apple will fix it. Instead I tell them that
Apple is slow and it's their fault we cannot ship our software. Then I tell
them it's not our fault and that they cannot expect support if the platform
is broken. Naturally I explain my unhappyness and tell them why I hate them
very much to make sure we talk the same language.

Bottom line is people loose respect for the Apple if they don't take bugs
seriously. I mean how hard can it be to change their bug system to email me
with updates when something new happens? Even microsoft does that now.

It's quite redicules to have a bug system and not even be able to take
advantage of it.

If I would go tell my customers "yeah there might be bugs in our software
but you have to wait an unknown amount of time" I would surely not have them
use my product for very long. In fact some people have told me some
competitors wanted them to pay huge amount of money to get fixes done. Thats
why they contacted us instead.

Reply to this Message

Michael Ash - 28 Jan 2006 23:50 GMT

>> but don't expect to find out anything related to your particular bug until
>> the fix ships.
[quoted text clipped - 9 lines]
> is broken. Naturally I explain my unhappyness and tell them why I hate them
> very much to make sure we talk the same language.

This is exactly what I meant by expectations.

You're filing a bug with the expectation that it will help you out in the
near term. With Apple, this expectation is completely wrong. With Apple,
you should file bugs with the expectation that you're providing feedback
and helping to direct their overall development effort, nothing more.

If you have a problem that is actually preventing you from shipping, you
should contact DTS (Developer Technical Support), which is reachable by
e-mail at dts@ the obvious domain. Note that DTS charges for support
incidents. If you have a paid ADC membership (except the student one) you
will have some free incidents, otherwise you will have to pay. But this
will get you in touch with real, live Apple people who will work with you
on your problem and either provide you with a fix, a workaround, or a
refund.

Is this right? Probably not. But this is how it is, and if your goal is to
get things done, you have to work with what's provided.

Signature

Michael Ash
Rogue Amoeba Software

Reply to this Message

Eric Albert - 29 Jan 2006 08:38 GMT

> >> but don't expect to find out anything related to your particular bug until
> >> the fix ships.
[quoted text clipped - 29 lines]
> Is this right? Probably not. But this is how it is, and if your goal is to
> get things done, you have to work with what's provided.

Additionally, if nobody reports the bug to Apple, it won't be fixed.
It's as simple as that. Some bugs get fixed faster than others, of
course, but obviously Apple can only fix bugs that it knows about. If
you choose not to report bugs that's OK, but Apple's just going to end
up fixing other folks' bugs instead.

-Eric

Signature

Eric Albert ejalbert@cs.stanford.edu
http://outofcheese.org/

Reply to this Message

Mac Forum / Programming / Mac Programming / January 2006

nmedit on ppc64