> No problem. I will alter that, since I've got a bunch of other minor
> changes to make that get this all working with O3CPU. What's the best
> thing for me to do - add everything into this patch and repost or create
> separate patches for the new stuff that can be applied after this one?
Sorry I didn't get back sooner.  In general, I think it is better to
fix patches that have code that you want fixed, but I do prefer
keeping separate changes in separate patches.   Reposting works.  An
alternative is to add a patch to your queue on top of the one that you
want to fix, fix the problems in the new patch, e-mail that patch out,
then you can qfold it back in later before it is committed.  Whichever
is easier for you.

> Ah, right. The disassembly code is one area that could do with some
> improvement. To tell you the truth, I put this together in dribs and drabs
> so the choice of << or ccprintf was just whatever was quick to get it
> working, then I didn't fix it up properly. Again, I can fix this so it's
> consistent, but at the same time I'd like to also change the disassembly
> output so that it is fairly similar to what gdb or objdump prints out.
> This means fixing some of the mnemonics under certain constraints and is
> kinda off my critical path at the moment. But it would be useful to do.
No problem.  I just really wanted to know if something was broken with cprintf.

It seems that things are pretty close to done, so I don't think we
need to spend forever perfecting this set of patches.  You could fix
future issues in the tree.

Thanks,

   Nate
_______________________________________________
m5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/m5-dev

Reply via email to