Kaffe/Alpha - bug fix ? finally?

Timothy Stack stack at cs.utah.edu
Mon Oct 9 14:45:51 PDT 2000

> Finally it seems like I might have fixed the
> assertion failure that has been bothering me
> on my Linux/Alpha box.

which assert?

> In gcRealloc():kaffe/kaffevm/mem/gc-incremental.c
> line 1026 it looks like the size of memcpy should
> be size instead of osize. If the osize (original/old
> size ?) is less than size+sizeof(gc_unit) isn't it
> logical to copy size number of bytes instead of
> osize? 

Look at the test on line 1020:

	if (osize >= size + sizeof(gc_unit)) {
		return (mem);

This will cause it to return if the old block size is
bigger than the requested size.  So, basically the collector
won't ever shrink the block so it should always copy the
old size.

> Anyways I made the fix in my local sources
> and ran 'make check'. 25 tests failed but atleast
> I seem to have gotten past the assertion failure.

With the previous change you will be copying bogus into
the end of the block where kaffe will normally expect
it to be initialized to zero.

> Bharadwaj
> PS : Do all the tests pass on Linux/Intel?

i believe so.

tim stack

More information about the kaffe mailing list