A bug in ifort 11.1.067? Variable 'class' is messed up with CLASS definition

[bruceibm]

We have fortran codes with variables like 'class0', 'class1' and 'classes'. It was always OK with ifort until the latest release of ifort 11.1.067.

When I compile the code with the new ifort, it complains about the wrong definition of CLASS. The new ifort takes all the names begining with 'class' as an object.

Could this be fixed in the next release? Otherwise a lot of code need to be re-written.

Here are some typical error messages from the latest release of ifort:
--------
error #5082: Syntax error, found INTEGER_CONSTANT '0' when expecting one of: ( % : . = =>
class0(nhybrid) = atmcls(it0)
xtalfit.f(398): error #5082: Syntax error, found IDENTIFIER 'ES' when expecting one of: ( % : . = =>
classes(i,ixtal) = class(i)
-----------------^
xtalfit.f(398): error #6535: This variable or component must be of a derived or structure type [CLASS]
classes(i,ixtal) = class(i)
------------^

------------------

Thx!

[Steven_L_Intel1 · Accepted solution]

That's a Linux version number. The current Windows version is 048. This bug is fixed in Update 3, which should be out in the next week.

[bruceibm]

Quoting - Steve Lionel (Intel)

That's a Linux version number. The current Windows version is 048. This bug is fixed in Update 3, which should be out in the next week.

I am using Snow Leopard. So next week this problem will be gone? Thanks! I will wait for this.

[Steven_L_Intel1]

Oh, Mac. Ok - we have a separate forum for Mac and Linux. Yes, I hope that this bug should go away with an update soon.

[bruceibm]

Quoting - Steve Lionel (Intel)

Oh, Mac. Ok - we have a separate forum for Mac and Linux. Yes, I hope that this bug should go away with an update soon.

Oh, Sorry, I did not notice this is a forum for Windows. I should post there. Thanks!

[Steven_L_Intel1]

Next time...