Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-07-05 Thread Joerg Bornemann
On 6/25/19 11:51 AM, Alexandru Croitor wrote:

> If the stash file was in higher in the directory tree than you expected, it 
> might have been due to this bug report 
> https://bugreports.qt.io/browse/QTBUG-76140
> 
> which should be fixed by 
> https://codereview.qt-project.org/c/qt/qtbase/+/263519

Which had to be reverted, because it caused QTBUG-76907. :-(


BR,

Joerg
___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-07-04 Thread Thiago Macieira
On Wednesday, 3 July 2019 18:09:36 -03 Matthew Woehlke wrote:
> Can't you make that depend on *the compiler*?
> 
>   .qmake.stash: /usr/bin/g++
> 
> (I guess this wouldn't behave well with ccache, but oh well...)

And icecc.

And the actual compiler is /usr/lib64/gcc/x86_64-suse-linux/9/cc1plus. That 
file can change and the g++ front-end remain the same.

Not to mention that outside of macOS, we call the compiler without a full 
path, so you can change $PATH and find a different one. Quite handy to test a 
new GCC version.

So, no, depending on the compiler path won't help.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel System Software Products



___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-07-03 Thread Matthew Woehlke
On 26/06/2019 21.17, Thiago Macieira wrote:
> The whole reason the file exists is to avoid 
> having to call the compiler to find out its details every time that qmake is 
> started. In order to do what you are asking for, we need to call out to the 
> compiler, which defeats the purpose.

Can't you make that depend on *the compiler*?

  .qmake.stash: /usr/bin/g++

(I guess this wouldn't behave well with ccache, but oh well...)

-- 
Matthew
___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-07-03 Thread Thiago Macieira
On Wednesday, 3 July 2019 06:28:04 -03 Christian Kandeler wrote:
> On Tue, 2 Jul 2019 21:08:49 -0300
> 
> Thiago Macieira  wrote:
> > On Tuesday, 2 July 2019 16:41:30 -03 Patrick Stinson wrote:
> > > `rm -rf` on your build does not solve this issue because the
> > > .qmake.stash
> > > file often exists in a path above the build/source dir.
> > 
> > That means you ran qmake there, which makes it a build dir.
> 
> But see https://bugreports.qt.io/browse/QTBUG-76140.

Yes, qmake has a very old bug (much older than your report) in calculating 
relative paths. That's why you can't have the build inside the sources, it 
gets the number of .. wrong.

And no one is going to fix this. Just put up with it until Qt switches to 
something else.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel System Software Products



___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-07-03 Thread Christian Kandeler
On Tue, 2 Jul 2019 21:08:49 -0300
Thiago Macieira  wrote:

> On Tuesday, 2 July 2019 16:41:30 -03 Patrick Stinson wrote:
> > `rm -rf` on your build does not solve this issue because the .qmake.stash
> > file often exists in a path above the build/source dir.
> 
> That means you ran qmake there, which makes it a build dir.

But see https://bugreports.qt.io/browse/QTBUG-76140.


Christian
___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-07-02 Thread Patrick Stinson
Aha! Very interesting. That does seem to make it a user-issue, albeit a very 
convoluted issue.

> On Jul 2, 2019, at 4:08 PM, Thiago Macieira  wrote:
> 
> On Tuesday, 2 July 2019 16:41:30 -03 Patrick Stinson wrote:
>> `rm -rf` on your build does not solve this issue because the .qmake.stash
>> file often exists in a path above the build/source dir.
> 
> That means you ran qmake there, which makes it a build dir.
> 
> You should remove your build dirs.
> 
> -- 
> Thiago Macieira - thiago.macieira (AT) intel.com
>  Software Architect - Intel System Software Products
> 
> 
> 
> ___
> Interest mailing list
> Interest@qt-project.org
> https://lists.qt-project.org/listinfo/interest

___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-07-02 Thread Thiago Macieira
On Tuesday, 2 July 2019 16:41:30 -03 Patrick Stinson wrote:
> `rm -rf` on your build does not solve this issue because the .qmake.stash
> file often exists in a path above the build/source dir.

That means you ran qmake there, which makes it a build dir.

You should remove your build dirs.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel System Software Products



___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-07-02 Thread Patrick Stinson
`rm -rf` on your build does not solve this issue because the .qmake.stash
file often exists in a path above the build/source dir.

On Wed, Jun 26, 2019 at 5:23 PM Thiago Macieira 
wrote:

> On Wednesday, 26 June 2019 16:22:48 PDT Hamish Moffatt wrote:
> > My point was that the file is entirely managed by qmake, so (IMHO) qmake
> > should deal with it if the data is now stale.
>
> It's very difficult to do that. The whole reason the file exists is to
> avoid
> having to call the compiler to find out its details every time that qmake
> is
> started. In order to do what you are asking for, we need to call out to
> the
> compiler, which defeats the purpose.
>
> Also, changing the compiler does require erasing your build and starting
> over
> anyway. For example, if you're using precompiled headers, a version update
> of
> the compiler (even patch level) can invalidate those files and neither
> qmake
> nor Make will notice that.
>
> My point is that this is a bigger problem than just the .qmake.stash file.
> I
> really recommend just rm -rf your build.
>
> PS: make your life easier and don't configure Qt from inside the sources.
>
> --
> Thiago Macieira - thiago.macieira (AT) intel.com
>   Software Architect - Intel System Software Products
>
>
>
> ___
> Interest mailing list
> Interest@qt-project.org
> https://lists.qt-project.org/listinfo/interest
>
___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-26 Thread Thiago Macieira
On Wednesday, 26 June 2019 16:22:48 PDT Hamish Moffatt wrote:
> My point was that the file is entirely managed by qmake, so (IMHO) qmake
> should deal with it if the data is now stale.

It's very difficult to do that. The whole reason the file exists is to avoid 
having to call the compiler to find out its details every time that qmake is 
started. In order to do what you are asking for, we need to call out to the 
compiler, which defeats the purpose.

Also, changing the compiler does require erasing your build and starting over 
anyway. For example, if you're using precompiled headers, a version update of 
the compiler (even patch level) can invalidate those files and neither qmake 
nor Make will notice that.

My point is that this is a bigger problem than just the .qmake.stash file. I 
really recommend just rm -rf your build.

PS: make your life easier and don't configure Qt from inside the sources.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel System Software Products



___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-26 Thread Hamish Moffatt

On 25/6/19 7:28 pm, Christian Kandeler wrote:

On Mon, 24 Jun 2019 23:25:38 +
Hamish Moffatt  wrote:


On 25/6/19 8:46 am, Thiago Macieira wrote:

On Monday, 24 June 2019 08:24:25 PDT Thiago Macieira wrote:

Anyway, let's stop speculating. Until we can reliably reproduce this problem
to understand what's going on, we're shooting in the dark.

I managed to reproduce this once. It was a left-over .qmake.stash file up in
the directory tree that was telling qmake to use an SDK version that was no
longer installed. I don't know how that file got created...


I also had problems with a stale SDK version in .qmake.stash when I
first upgraded to Qt 5.12. https://bugreports.qt.io/browse/QTBUG-71620

Despite this being entirely qmake's problem it was closed as invalid.

"Out of scope", actually.



My point was that the file is entirely managed by qmake, so (IMHO) qmake 
should deal with it if the data is now stale.



Hamish

___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-26 Thread Patrick Stinson
Yep, it was a .qmake.stash file two dirs up. I deleted it and everything builds 
fine again.

> On Jun 25, 2019, at 8:03 AM, Thiago Macieira  
> wrote:
> 
> On Tuesday, 25 June 2019 02:51:58 PDT Alexandru Croitor wrote:
>> If the stash file was in higher in the directory tree than you expected, it
>> might have been due to this bug report
>> https://bugreports.qt.io/browse/QTBUG-76140
>> 
>> which should be fixed by
>> https://codereview.qt-project.org/c/qt/qtbase/+/263519
> 
> I have a slightly unusual set up that I configure in qtbase, but then can 
> cd .. and make in the top-level. This isn't supported with the official Qt 
> but 
> used to be sometime before 5.5. My set up remained.
> 
> The intruding .qmake.stash was the one in the top-level. I did rsync from the 
> old machine to the new one, so that might explain how the bad data came to be.
> 
> -- 
> Thiago Macieira - thiago.macieira (AT) intel.com
>  Software Architect - Intel System Software Products
> 
> 
> 
> ___
> Interest mailing list
> Interest@qt-project.org
> https://lists.qt-project.org/listinfo/interest

___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-25 Thread Thiago Macieira
On Tuesday, 25 June 2019 02:51:58 PDT Alexandru Croitor wrote:
> If the stash file was in higher in the directory tree than you expected, it
> might have been due to this bug report
> https://bugreports.qt.io/browse/QTBUG-76140
> 
> which should be fixed by
> https://codereview.qt-project.org/c/qt/qtbase/+/263519

I have a slightly unusual set up that I configure in qtbase, but then can 
cd .. and make in the top-level. This isn't supported with the official Qt but 
used to be sometime before 5.5. My set up remained.

The intruding .qmake.stash was the one in the top-level. I did rsync from the 
old machine to the new one, so that might explain how the bad data came to be.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel System Software Products



___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-25 Thread Alexandru Croitor


> On 25. Jun 2019, at 00:46, Thiago Macieira  wrote:
> 
> On Monday, 24 June 2019 08:24:25 PDT Thiago Macieira wrote:
>> Anyway, let's stop speculating. Until we can reliably reproduce this problem
>> to understand what's going on, we're shooting in the dark.
> 
> I managed to reproduce this once. It was a left-over .qmake.stash file up in 
> the directory tree that was telling qmake to use an SDK version that was no 
> longer installed. I don't know how that file got created...
> 
> -- 
> Thiago Macieira - thiago.macieira (AT) intel.com
>  Software Architect - Intel System Software Products
> 

If the stash file was in higher in the directory tree than you expected, it 
might have been due to this bug report 
https://bugreports.qt.io/browse/QTBUG-76140

which should be fixed by https://codereview.qt-project.org/c/qt/qtbase/+/263519
___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-25 Thread Christian Kandeler
On Mon, 24 Jun 2019 23:25:38 +
Hamish Moffatt  wrote:

> On 25/6/19 8:46 am, Thiago Macieira wrote:
> > On Monday, 24 June 2019 08:24:25 PDT Thiago Macieira wrote:
> >> Anyway, let's stop speculating. Until we can reliably reproduce this 
> >> problem
> >> to understand what's going on, we're shooting in the dark.
> > I managed to reproduce this once. It was a left-over .qmake.stash file up in
> > the directory tree that was telling qmake to use an SDK version that was no
> > longer installed. I don't know how that file got created...
> >
> I also had problems with a stale SDK version in .qmake.stash when I 
> first upgraded to Qt 5.12. https://bugreports.qt.io/browse/QTBUG-71620
> 
> Despite this being entirely qmake's problem it was closed as invalid.

"Out of scope", actually. 


Christian
___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-24 Thread Thiago Macieira
On Monday, 24 June 2019 16:25:38 PDT Hamish Moffatt wrote:
> I also had problems with a stale SDK version in .qmake.stash when I
> first upgraded to Qt 5.12. https://bugreports.qt.io/browse/QTBUG-71620
> 
> Despite this being entirely qmake's problem it was closed as invalid.

When you upgrade the compiler or the SDK, you have to delete your build and 
start over.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel System Software Products



___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-24 Thread Hamish Moffatt

On 25/6/19 8:46 am, Thiago Macieira wrote:

On Monday, 24 June 2019 08:24:25 PDT Thiago Macieira wrote:

Anyway, let's stop speculating. Until we can reliably reproduce this problem
to understand what's going on, we're shooting in the dark.

I managed to reproduce this once. It was a left-over .qmake.stash file up in
the directory tree that was telling qmake to use an SDK version that was no
longer installed. I don't know how that file got created...

I also had problems with a stale SDK version in .qmake.stash when I 
first upgraded to Qt 5.12. https://bugreports.qt.io/browse/QTBUG-71620


Despite this being entirely qmake's problem it was closed as invalid.


Hamish

___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-24 Thread Thiago Macieira
On Monday, 24 June 2019 08:24:25 PDT Thiago Macieira wrote:
> Anyway, let's stop speculating. Until we can reliably reproduce this problem
> to understand what's going on, we're shooting in the dark.

I managed to reproduce this once. It was a left-over .qmake.stash file up in 
the directory tree that was telling qmake to use an SDK version that was no 
longer installed. I don't know how that file got created...

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel System Software Products



___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-24 Thread Thiago Macieira
On Sunday, 23 June 2019 22:23:56 PDT Patrick Stinson wrote:
> The qt configure script is predictably unable to find it on the first
> configure run but able to find it on the second run. The ask handy changed.
> Or were you considering that as well?

The configure script runs the compiler. If the compiler fails once then 
passes, with the exact same command-line, it's the compiler's fault. On the 
other hand, if the configure script is running something different on the 
second run, then it's the configure script's fault.

Anyway, let's stop speculating. Until we can reliably reproduce this problem 
to understand what's going on, we're shooting in the dark.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel System Software Products



___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-23 Thread Patrick Stinson
The qt configure script is predictably unable to find it on the first configure 
run but able to find it on the second run. The ask handy changed. Or were you 
considering that as well?

> On Jun 23, 2019, at 9:07 PM, Thiago Macieira  
> wrote:
> 
>> On Sunday, 23 June 2019 13:23:04 PDT Patrick Stinson wrote:
>> Hmm, what indicates to you that it is a compiler problem? It seems to me
>> that the fact that it (predictably) works unpredictably depending on the
>> state of the Qt conf/build system points more strongly to the problem being
>> in Qt the Qt conf/build system.
> 
> The fact that the compiler failed to find a header that comes from the SDK 
> that comes with the compiler. And that this problem did not happen in older 
> XCode versions (my previous Mac was still High Sierra, which didn't get the 
> latest XCode update).
> 
> -- 
> Thiago Macieira - thiago.macieira (AT) intel.com
>  Software Architect - Intel System Software Products
> 
> 
> 
> ___
> Interest mailing list
> Interest@qt-project.org
> https://lists.qt-project.org/listinfo/interest
___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-23 Thread Thiago Macieira
On Sunday, 23 June 2019 13:23:04 PDT Patrick Stinson wrote:
> Hmm, what indicates to you that it is a compiler problem? It seems to me
> that the fact that it (predictably) works unpredictably depending on the
> state of the Qt conf/build system points more strongly to the problem being
> in Qt the Qt conf/build system.

The fact that the compiler failed to find a header that comes from the SDK 
that comes with the compiler. And that this problem did not happen in older 
XCode versions (my previous Mac was still High Sierra, which didn't get the 
latest XCode update).

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel System Software Products



___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-23 Thread Patrick Stinson
Hmm, what indicates to you that it is a compiler problem? It seems to me that 
the fact that it (predictably) works unpredictably depending on the state of 
the Qt conf/build system points more strongly to the problem being in Qt the Qt 
conf/build system.

> On Jun 23, 2019, at 7:59 AM, Thiago Macieira  
> wrote:
> 
>> On Saturday, 22 June 2019 09:50:56 PDT Patrick Stinson wrote:
>> Wow. Very strange. Maybe I should file a bug report, though I bet it will be
>> very low priority.
> 
> To whom? To Qt or to Apple? As far as I can tell, this is a compiler problem.
> 
> -- 
> Thiago Macieira - thiago.macieira (AT) intel.com
>  Software Architect - Intel System Software Products
> 
> 
> 
> ___
> Interest mailing list
> Interest@qt-project.org
> https://lists.qt-project.org/listinfo/interest
___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-22 Thread Patrick Stinson
Wow. Very strange. Maybe I should file a bug report, though I bet it will be 
very low priority.

> On Jun 22, 2019, at 8:07 AM, Thiago Macieira  
> wrote:
> 
>> On Friday, 21 June 2019 21:50:39 PDT Patrick Stinson wrote:
>> So how do you account for that? Seems very odd to me. This was from a
>> freshly extracted qt source archive.
> 
> I don't know why it happens. But I had a similar situation yesterday setting 
> up my new Mac, with Xcode 10.2. This did not happen on my old one.
> 
> It complained that it couldn't find the GL headers. From the SDK provided by 
> Xcode...
> 
> -- 
> Thiago Macieira - thiago.macieira (AT) intel.com
>  Software Architect - Intel System Software Products
> 
> 
> 
> ___
> Interest mailing list
> Interest@qt-project.org
> https://lists.qt-project.org/listinfo/interest
___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-21 Thread Patrick Stinson
I am building qt from a couple of different scripts so it would be great to be 
able to have them work on first run.

> On Jun 21, 2019, at 8:50 PM, Patrick Stinson  wrote:
> 
> My config.log did not contain the string “std-atomic64.” It did report that 
> it could not find libatomic. Deleting config.cache and running config.status 
> worked though!
> 
> So how do you account for that? Seems very odd to me. This was from a freshly 
> extracted qt source archive.
> 
>> On Jun 21, 2019, at 12:13 PM, Thiago Macieira  
>> wrote:
>> 
>> On Friday, 21 June 2019 10:35:36 PDT Patrick Stinson wrote:
>>> #define QT_FEATURE_std_atomic64 -1
>> 
>> Can you check the config.log in the top-level dir to see what how the std-
>> atomic64 test failed? If you don't see that in the config.log, remove 
>> config.cache and rerun configure (./config.status)
>> 
>> -- 
>> Thiago Macieira - thiago.macieira (AT) intel.com
>> Software Architect - Intel System Software Products
>> 
>> 
>> 
>> ___
>> Interest mailing list
>> Interest@qt-project.org
>> https://lists.qt-project.org/listinfo/interest
> 

___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-21 Thread Thiago Macieira
On Friday, 21 June 2019 10:35:36 PDT Patrick Stinson wrote:
> #define QT_FEATURE_std_atomic64 -1

Can you check the config.log in the top-level dir to see what how the std-
atomic64 test failed? If you don't see that in the config.log, remove 
config.cache and rerun configure (./config.status)

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel System Software Products



___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-21 Thread Jean-Michaël Celerier
Every time I had this error recently it's because I cleaned my build
directory via rm -rf * which did not remove the hidden .qmake.cache files
in there which then thoroughly destroys any attemptto rebuild qt in this
directory.

Maybe it can help ? (it was the exact same symptoms in any case)
Best,
Jean-Michaël

On Fri, Jun 21, 2019 at 7:37 PM Patrick Stinson 
wrote:

> #define QT_FEATURE_std_atomic64 -1
>
> On Jun 21, 2019, at 9:32 AM, Thiago Macieira 
> wrote:
>
>  src/corelib/qtcore-config.h
>
>
> ___
> Interest mailing list
> Interest@qt-project.org
> https://lists.qt-project.org/listinfo/interest
>
___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-21 Thread Patrick Stinson
#define QT_FEATURE_std_atomic64 -1

> On Jun 21, 2019, at 9:32 AM, Thiago Macieira  
> wrote:
> 
>  src/corelib/qtcore-config.h

___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


Re: [Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-21 Thread Thiago Macieira
On Friday, 21 June 2019 10:00:45 PDT Patrick Stinson wrote:
> thread/qatomic.cpp:1624:4: error: "Q_ATOMIC_INT64_IS_SUPPORTED must be
> defined on a 64-bit platform"

The macro is defined by this block in qatomic_cxx11.h:

#if QT_CONFIG(std_atomic64)
template<> struct QAtomicOpsSupport<8> { enum { IsSupported = 1 }; };
#  define Q_ATOMIC_INT64_IS_SUPPORTED

So the question is whether QT_FEATURE_std_atomic64 is defined to 1 or not. Can 
you open the src/corelib/qtcore-config.h in the build dir and see if it is 
there?

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel System Software Products



___
Interest mailing list
Interest@qt-project.org
https://lists.qt-project.org/listinfo/interest


[Interest] Q_ATOMIC_INT64_IS_SUPPORTED: Qt Compile errors on macOS

2019-06-21 Thread Patrick Stinson
(NOTE: I have started a new thread about this from an old one incorrectly 
associating this problem with “make -jn”)

I am having a hard time pinning down the source of this error. I am getting it 
sporadically on Qt-5.12.0 - Qt-5.13.0. using make -jn flags or not using make 
-jn flags doesn’t seem to have an effect. Any ideas?



/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/clang++
 -c -include.pch/Qt5Core/c++_x86_64 -pipe -stdlib=libc++ -O3 -fPIC -std=c++1y 
-fapplication-extension  -arch x86_64 -isysroot 
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.14.sdk
 -mmacosx-version-min=10.12 -fvisibility=hidden -fvisibility-inlines-hidden 
-ffunction-sections -fdata-sections -Wall -W -Wobjc-interface-ivars 
-Wobjc-method-access -Wobjc-multiple-method-names -DQT_NO_USING_NAMESPACE 
-DQT_NO_FOREACH -DQT_NO_NARROWING_CONVERSIONS_IN_CONNECT -DQT_BUILD_CORE_LIB 
-DQT_BUILDING_QT -DQT_NO_CAST_TO_ASCII -DQT_ASCII_CAST_WARNINGS -DQT_MOC_COMPAT 
-DQT_USE_QSTRINGBUILDER -DQT_DEPRECATED_WARNINGS 
-DQT_DISABLE_DEPRECATED_BEFORE=0x05 -D_LARGEFILE64_SOURCE 
-D_LARGEFILE_SOURCE -DQT_NO_DEBUG -DPCRE2_CODE_UNIT_WIDTH=16 -I. 
-I../3rdparty/zlib/src -Iglobal -I../3rdparty/harfbuzz/src -I../3rdparty/md5 
-I../3rdparty/md4 -I../3rdparty/sha3 -I../3rdparty 
-I../3rdparty/double-conversion/include 
-I../3rdparty/double-conversion/include/double-conversion -I../3rdparty/forkfd 
-I../3rdparty/tinycbor/src -I../../include -I../../include/QtCore 
-I../../include/QtCore/5.13.0 -I../../include/QtCore/5.13.0/QtCore -I.moc 
-I.tracegen -I../3rdparty/pcre2/src 
-I/Users/patrick/dev/vendor/sysroot-masos-64/include -I../../mkspecs/macx-clang 
-o .obj/qatomic.o thread/qatomic.cpp
thread/qatomic.cpp:1624:4: error: "Q_ATOMIC_INT64_IS_SUPPORTED must be defined 
on a 64-bit platform"
#  error "Q_ATOMIC_INT64_IS_SUPPORTED must be defined on a 64-bit platform"
   ^
In file included from thread/qatomic.cpp:1:
In file included from 
/Users/patrick/dev/vendor/sysroot-masos-64/build/qt-everywhere-src-5.13.0/qtbase/src/corelib/global/qt_pch.h:56:
In file included from 
../../include/QtCore/../../src/corelib/global/qglobal.h:1224:
In file included from 
/Users/patrick/dev/vendor/sysroot-masos-64/build/qt-everywhere-src-5.13.0/qtbase/src/corelib/../../include/QtCore/qatomic.h:1:
In file included from thread/qatomic.h:46:
In file included from 
/Users/patrick/dev/vendor/sysroot-masos-64/build/qt-everywhere-src-5.13.0/qtbase/src/corelib/../../include/QtCore/qbasicatomic.h:1:
/Users/patrick/dev/vendor/sysroot-masos-64/build/qt-everywhere-src-5.13.0/qtbase/src/corelib/../../include/QtCore/../../src/corelib/thread/qbasicatomic.h:97:5:
 error: static_assert failed due to requirement 
'bool(QAtomicOpsSupport::IsSupported)' "template parameter is an 
integral of a size not supported on this platform"
Q_STATIC_ASSERT_X(QAtomicOpsSupport::IsSupported, "template 
parameter is an integral of a size not supported on this platform");

^~
../../include/QtCore/../../src/corelib/global/qglobal.h:121:49: note: expanded 
from macro 'Q_STATIC_ASSERT_X'
#  define Q_STATIC_ASSERT_X(Condition, Message) static_assert(bool(Condition), 
Message)
^ ~~~
thread/qatomic.h:55:31: note: in instantiation of template class 
'QBasicAtomicInteger' requested here
class QAtomicInteger : public QBasicAtomicInteger
  ^
thread/qatomic.cpp:1631:17: note: in instantiation of template class 
'QAtomicInteger' requested here
Q_STATIC_ASSERT(sizeof(QAtomicInteger));
^
In file included from thread/qatomic.cpp:1:
In file included from 
/Users/patrick/dev/vendor/sysroot-masos-64/build/qt-everywhere-src-5.13.0/qtbase/src/corelib/global/qt_pch.h:56:
In file included from 
../../include/QtCore/../../src/corelib/global/qglobal.h:1224:
In file included from 
/Users/patrick/dev/vendor/sysroot-masos-64/build/qt-everywhere-src-5.13.0/qtbase/src/corelib/../../include/QtCore/qatomic.h:1:
In file included from thread/qatomic.h:46:
In file included from 
/Users/patrick/dev/vendor/sysroot-masos-64/build/qt-everywhere-src-5.13.0/qtbase/src/corelib/../../include/QtCore/qbasicatomic.h:1:
/Users/patrick/dev/vendor/sysroot-masos-64/build/qt-everywhere-src-5.13.0/qtbase/src/corelib/../../include/QtCore/../../src/corelib/thread/qbasicatomic.h:97:5:
 error: static_assert failed due to requirement 
'bool(QAtomicOpsSupport::IsSupported)' "template 
parameter is an integral of a size not supported on this platform"
Q_STATIC_ASSERT_X(QAtomicOpsSupport::IsSupported, "template 
parameter is an integral of a size not supported on this platform");