This is the mail archive of the java@gcc.gnu.org mailing list for the Java project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

RE: overhead of reflection info


There is a claim in _Jv_MarkObj comments that some data structures
referenced from class objects are allocated pointerfree (even though they are
not?) but then traced correctly by _Jv_MarkObj.  If this is indeed still correct,
this would have to be looked at carefully.

Hans

> -----Original Message-----
> From: java-owner@gcc.gnu.org 
> [mailto:java-owner@gcc.gnu.org]On Behalf Of
> Tom Tromey
> Sent: Tuesday, February 24, 2004 8:32 AM
> To: Adam Megacz
> Cc: java@gcc.gnu.org
> Subject: Re: overhead of reflection info
> 
> 
> >>>>> "Adam" == Adam Megacz <adam@megacz.com> writes:
> 
> Adam> Ah, interesting.  Since it is only called in the rare 
> case, could I
> Adam> just have it assume that every word of the object is a 
> pointer in
> Adam> order to play it safe?
> 
> Yes, with some changes in boehm.cc you should be able to set things up
> to do a completely conservative scan.  Or you could modify the
> compiler to crash if it encounters a class for which it can't generate
> a bitmap descriptor.
> 
> Tom
> 


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]