Recognize obsoleted options

[Bryce McKinlay]

Aaron Luchko wrote:

>The only real documentation of any of these I could find is an oreilly
>book.  Although the sun vm accepts all these options I don't believe
>they implement them since it's obsoleted, also looking at the size of
>diffs of straces of runs with and without the option there wasn't much
>of a difference with the -noverify.  
>The only mention of any of these options on the sun site I could find is
>this bug report which seems to suggest none of them are implemented
>http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4337532
>  
>

Actually, -noverify is still supported, even if they don't document it.

>If it's easy enough it could do as an optimization, I took a glimpse at
>links.cc and think it may be possible to do easily but I don't know the
>code that well.
>  
>

It is easy and I think we should implement it - occasionally it can come 
in useful when running bytecode produced by old, buggy compilers. I'm 
going to check in the following patch. Note that with this patch, 
-noverify also disables "type assertion" verification for BC-ABI 
compiled code.

Bryce


-------------- next part --------------
A non-text attachment was scrubbed...
Name: libjava-noverify.patch
Type: text/x-patch
Size: 4474 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/java-patches/attachments/20050512/52e68529/attachment.bin>