<br><font size=2><tt>Dave Hansen <haveblue@us.ibm.com> wrote on 28/09/2005
17:44:18:<br>
<br>
> On Wed, 2005-09-28 at 15:10 -0300, Gustavo Sverzut Barbieri wrote:<br>
> > Here are some patches to clean up lmb.c, numa.c and init.c, as
requested by <br>
> > Dave Hansen.<br>
> <br>
> Let me elaborate a bit. I created _another_ loop to run through
all of<br>
> the LMBs in the NUMA init code a few weeks back for memory_present()<br>
> calls. I thought this this is a decent way to atone for that
messy<br>
> code.<br>
> <br>
> This first patch is a pretty simple cleanup.<br>
> <br>
> The LMB macro patch certainly makes some of the code more readable.<br>
> But, it might not be worth it to use the macro in lmb.c. It
just seems<br>
> to make it more complicated. Any thoughts?<br>
</tt></font>
<br><font size=2><tt>I could use that macro just where it loops from 0
to "cnt", with that we can leave the _reversed() version out
and leave coalesce and the other one unchanged. Maybe leave the _reversed()
(it's used in 2 places).</tt></font>
<br>
<br><font size=2><tt>PS: If this mail doesn't respect any internet standards,
it's because it was sent by Lotus Notes. Sorry, my imap account is still
in progress!</tt></font>
<br>
<br><font size=2><tt>Cheers from Brazil,</tt></font>
<br>
<br><font size=2><tt>---</tt></font>
<br><font size=2><tt>Gustavo Sverzut Barbieri</tt></font>