Date: Sat, 15 Feb 2003 11:00:30 -0800 From: Arun Sharma <arun@sharma-home.net> To: Marcel Moolenaar <marcel@xcllnt.net> Cc: freebsd-ia64@FreeBSD.ORG Subject: Re: Maxmem should be bytes ? Message-ID: <20030215190030.GA31217@sharma-home.net> In-Reply-To: <20030215021642.GA1901@athlon.pn.xcllnt.net> References: <200302080111.h181Bkr26574@unix-os.sc.intel.com> <u1y2b3eb7.fsf@unix-os.sc.intel.com> <20030214002346.GA1267@athlon.pn.xcllnt.net> <uvfzn1yev.fsf@unix-os.sc.intel.com> <20030214010244.GA1344@athlon.pn.xcllnt.net> <ud6lu1g1q.fsf@unix-os.sc.intel.com> <20030215021642.GA1901@athlon.pn.xcllnt.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Feb 14, 2003 at 06:16:42PM -0800, Marcel Moolenaar wrote:
> > i386 does something equivalent.
>
> Your patch changes the granularity for Maxmem from pages to bytes.
> On all architectures it should be pages. On i386 you see atop(),
> which is the opposite of ia64_ptob().
Ah, you're right. I saw one incorrect usage of Maxmem and assumed that
Maxmem should be in bytes. So only necessary change then is:
--- pmap.c- Sat Feb 15 09:47:50 2003
+++ pmap.c Sat Feb 15 09:48:15 2003
@@ -363,7 +363,7 @@
* enough sparse, causing us to (try to) create a huge VHPT.
*/
vhpt_size = 15;
- while ((1<<vhpt_size) < ia64_btop(Maxmem) * 32)
+ while ((1<<vhpt_size) < (Maxmem * 32))
vhpt_size++;
vhpt_base = 0;
-Arun
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-ia64" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030215190030.GA31217>