linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Anton Blanchard <anton@samba.org>
To: akpm@osdl.org
Cc: paulus@samba.org, linux-kernel@vger.kernel.org, nathanl@austin.ibm.com
Subject: [PATCH] ppc64: Make NUMA code handle unexpected layouts
Date: Tue, 11 Jan 2005 02:19:50 +1100	[thread overview]
Message-ID: <20050110151950.GY14239@krispykreme.ozlabs.ibm.com> (raw)
In-Reply-To: <20050110144339.GW14239@krispykreme.ozlabs.ibm.com>


From: Nathan Lynch <nathanl@austin.ibm.com>

Ran into this on a 4GB partition - all but about ~300MB was thrown away.

Does this look ok?  It works for me, but I've not tested on firmware
without the bug.

Fall back to non-numa setup upon discovering unexpected memory layout
as presented by firmware, instead of throwing away regions.

Signed-off-by: Nathan Lynch <nathanl@austin.ibm.com>
Acked-by: Anton Blanchard <anton@samba.org>

diff -puN arch/ppc64/mm/numa.c~numa_broke_fix arch/ppc64/mm/numa.c
--- foobar2/arch/ppc64/mm/numa.c~numa_broke_fix	2005-01-10 23:36:39.922337007 +1100
+++ foobar2-anton/arch/ppc64/mm/numa.c	2005-01-10 23:46:46.069205815 +1100
@@ -345,8 +345,6 @@ new_range:
 			numa_domain = 0;
 		}
 
-		node_set_online(numa_domain);
-
 		if (max_domain < numa_domain)
 			max_domain = numa_domain;
 
@@ -361,14 +359,19 @@ new_range:
 				init_node_data[numa_domain].node_start_pfn +
 				init_node_data[numa_domain].node_spanned_pages;
 			if (shouldstart != (start / PAGE_SIZE)) {
-				printk(KERN_ERR "WARNING: Hole in node, "
-						"disabling region start %lx "
-						"length %lx\n", start, size);
-				continue;
+				/* Revert to non-numa for now */
+				printk(KERN_ERR
+				       "WARNING: Unexpected node layout: "
+				       "region start %lx length %lx\n",
+				       start, size);
+				printk(KERN_ERR "NUMA is disabled\n");
+				goto err;
 			}
 			init_node_data[numa_domain].node_spanned_pages +=
 				size / PAGE_SIZE;
 		} else {
+			node_set_online(numa_domain);
+
 			init_node_data[numa_domain].node_start_pfn =
 				start / PAGE_SIZE;
 			init_node_data[numa_domain].node_spanned_pages =
@@ -388,6 +391,14 @@ new_range:
 		node_set_online(i);
 
 	return 0;
+err:
+	/* Something has gone wrong; revert any setup we've done */
+	for_each_node(i) {
+		node_set_offline(i);
+		init_node_data[i].node_start_pfn = 0;
+		init_node_data[i].node_spanned_pages = 0;
+	}
+	return -1;
 }
 
 static void __init setup_nonnuma(void)
_


      reply	other threads:[~2005-01-10 15:23 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-01-10 13:38 [PATCH] ppc64: PCI cleanup Anton Blanchard
2005-01-10 13:41 ` [PATCH] ppc64: Remove flush_instruction_cache Anton Blanchard
2005-01-10 14:35   ` [PATCH] ppc64: interrupt code cleanup Anton Blanchard
2005-01-10 14:43     ` [PATCH] ppc64: Fix rtas_set_indicator(9005) Anton Blanchard
2005-01-10 15:19       ` Anton Blanchard [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20050110151950.GY14239@krispykreme.ozlabs.ibm.com \
    --to=anton@samba.org \
    --cc=akpm@osdl.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nathanl@austin.ibm.com \
    --cc=paulus@samba.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).