vmstat

[Christopher Faylor]

On Thu, Jul 04, 2002 at 02:00:06PM -0400, Nicholas Wourms wrote:
>Chris January wrote:
>>>I suspect that this must have something to do with binutils.  I changed
>>>the alignments at David Billinghurst's suggestion to accommodate java.
>>>If this is causing problems, however, I'll change them back.
>>>
>>That's what I was referring to.
>>
>Is it possible for you to work around this?  I believe the alignment
>change is necessary for a working libgcj/gcj suite.  It would be nice
>to have both working properly...

Yes, I agree.  On thinking about this, it seems like the header files are
probably in error if they are sensitive to this.

For the record.  I changed this:

#define COFF_DEFAULT_SECTION_ALIGNMENT_POWER (2)

to this:

#define COFF_DEFAULT_SECTION_ALIGNMENT_POWER (4)

but I confess that I didn't give much thought to what it did.  I thought
it just changed the alignment of sections rather than the alignment
within sections.

Now that I think of it, the way I did things would change this only
for native compilers, not cross compilers.  That's not right.

Maybe this explains the gcc errors that I've been seeing.

cgf