Peter Vreman wrote:
The example that is provided doesn't fail. Because when calling foo the value
of -4(%ebp) is
loaded in %eax and that value is used in the function foo. The setlength() will
create a new
string and store that at -4(%ebp) of the calling function. But doesn't change
the original
parameter value in %eax (and also stored also at (%esp) in foo).
Hmm... It doesn't fail for me either now, although I've seen it failing
with my own eyes before submitting it the first time :/
Probably I've oversimplified the example.
The failing function in Lazarus code is CreateRelativePath in
components/codetools/fileprocs.pas, being called (second occurrence)
from TBaseCompilerOptions.MakeOptionsString (ide/compileroptions.pp).
That code is much more complex than the example, and what's happening
there is not in the testsuite, because my patch wasn't causing testsuite
failures.
I'll be digging further in.
Sergei
_______________________________________________
fpc-devel maillist - fpc-devel@lists.freepascal.org
http://lists.freepascal.org/mailman/listinfo/fpc-devel
