Review request 8008310 - Some adjustments needed to minimal VM warnings and errors for unsupported command line options

[David Holmes]

Hi Joe,

On 18/02/2013 6:19 AM, JOSEPH PROVINO wrote:
> Webrev is here: http://cr.openjdk.java.net/~jprovino/8008310/webrev.00
>
> Only change is to arguments.cpp.

I'm sorry I didn't pick up on this with earlier changes (eg 
-Dcom.sun.management usage) but it seems to me that the argument parsing 
code is not supposed to call vm_exit_during_initialization, but should 
instead print an error message and return JNI_ERR. This will ultimately 
lead to CreateJavaVM returning JNI_ERR.

There is some existing use of vm_exit_during_initialization in 
arguments.cpp (CDS related and compilation policy related) but I would 
say those show bad examples of how to handle argument errors. There is 
an open RFE for the VM to never terminate during initialization but only 
return JNI_ERR. We're a long way from that, but I don't think we should 
be moving in the other direction.

So I would say that these changes from warnings to failures have to be 
redone. In addition the other uses of vm_exit_during_initialization 
related to the minimal VM, here:

2475       if (match_option(option, "-Dcom.sun.management", &tail)) {
2476 #if INCLUDE_MANAGEMENT
2477         FLAG_SET_CMDLINE(bool, ManagementServer, true);
2478 #else
2479         vm_exit_during_initialization(
2480           "-Dcom.sun.management is not supported in this VM.", NULL);
2481 #endif

here:

2826 #if !INCLUDE_MANAGEMENT
2827     } else if (match_option(option, "-XX:+ManagementServer", &tail)) {
2828       vm_exit_during_initialization(
2829         "ManagementServer is not supported in this VM.", NULL);
2830 #endif // INCLUDE_MANAGEMENT

and here

3132 #if INCLUDE_NMT
3133       MemTracker::init_tracking_options(tail);
3134 #else
3135       vm_exit_during_initialization(
3136         "Native Memory Tracking is not supported in this VM", NULL);
3137 #endif

should also be changed.

David
-----

> thanks.
>
> joe