I don't know the Mac well enough... What could cause the below?

From
https://forum.lazarus.freepascal.org/index.php/topic,44950.msg317604.html#msg317604

The logfile tells that the units have been compiled right before lldb was launched. So the date 15.4 should be correct.

LLDB complains
error: project1 debug map object file
'/Users/andy/Lazarus/lcl/units/i386-darwin/carbon/carbonint.o' has changed (actual time is 2019-04-15 12:17:05.000000000, debug map time is 2019-04-11 08:51:18.000000000) since this executable was linked, file will be ignored

Our code in the IDE also rejects the files because of the wrong timestamp.

Apparently the files may be on an NAS. Not sure if relevant... I.e. maybe if there is some cache...

I suspect it might be the linker (from apples developper tools?) that writes the timestamp into the debug map? So not fpc? Would the linker maybe have it's own cache, that may cause something like this?

I also (strongly) suspect, that the files have been recompiled once or more between the 11th Apr, and the current date. If so, then the linker (if it is the linker) must have some cache somewhere. This is because the same issue is in yesterdays log. The files also where rejected. With the 11th April in the debug map. But yesterday they would not have had todays timestamp...

Or maybe it is creation vs modification time?

I don't know the Mac well enough... What could cause the below?

From
https://forum.lazarus.freepascal.org/index.php/topic,44950.msg317604.html#msg317604

The logfile tells that the units have been compiled right before lldb was launched. So the date 15.4 should be correct.

LLDB complains
error: project1 debug map object file
'/Users/andy/Lazarus/lcl/units/i386-darwin/carbon/carbonint.o' has changed (actual time is 2019-04-15 12:17:05.000000000, debug map time is 2019-04-11 08:51:18.000000000) since this executable was linked, file will be ignored

Our code in the IDE also rejects the files because of the wrong timestamp.

Apparently the files may be on an NAS. Not sure if relevant... I.e. maybe if there is some cache...

I suspect it might be the linker (from apples developper tools?) that writes the timestamp into the debug map? So not fpc? Would the linker maybe have it's own cache, that may cause something like this?

I also (strongly) suspect, that the files have been recompiled once or more between the 11th Apr, and the current date. If so, then the linker (if it is the linker) must have some cache somewhere. This is because the same issue is in yesterdays log. The files also where rejected. With the 11th April in the debug map. But yesterday they would not have had todays timestamp...

Or maybe it is creation vs modification time?

Martin wrote

I suspect it might be the linker (from apples developper tools?) that writes the timestamp into the debug map? So not fpc?

FPC indeed has nothing to do with the debug map.

Would the linker maybe have it's own cache, that may cause something like this?

I have no idea.

Jonas

Op 2019-04-15 om 13:36 schreef Martin:

Apparently the files may be on an NAS. Not sure if relevant... I.e. maybe if there is some cache...

That might be the reason.  If you have remote files, you have three date systems, the local PC one, the NAS one, and the one of the filesharing protocol (like SMB).  Roundings and time differences between the systems can cause strange effects.