From: Ravikiran G Thirumalai <kiran@scalex86.org>
To: Andi Kleen <ak@suse.de>
Cc: linux-kernel@vger.kernel.org,
Christoph Lameter <clameter@engr.sgi.com>,
discuss@x86-64.org,
"Shai Fultheim (Shai@scalex86.org)" <shai@scalex86.org>,
Alok Kataria <alokk@calsoftinc.com>,
tony.luck@intel.com
Subject: [patch] Fix the node cpumask of a cpu going down
Date: Fri, 20 Jan 2006 17:56:33 -0800 [thread overview]
Message-ID: <20060121015633.GD3573@localhost.localdomain> (raw)
Currently, x86_64 and ia64 arches do not clear the corresponding bits
in the node's cpumask when a cpu goes down or cpu bring up is cancelled.
This is buggy since there are pieces of common code where the cpumask is
checked in the cpu down code path to decide on things (like in the slab
down path). PPC does the right thing, but x86_64 and ia64 don't (This
was the reason Sonny hit upon a slab bug during cpu offline on ppc and
could not reproduce on other arches). This patch fixes it for x86_64.
I won't attempt ia64 as I cannot test it.
Credit for spotting this should go to Alok.
Signed-off-by: Alok N Kataria <alokk@calsoftinc.com>
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Index: linux-2.6.16-rc1/arch/x86_64/kernel/smpboot.c
===================================================================
--- linux-2.6.16-rc1.orig/arch/x86_64/kernel/smpboot.c 2006-01-20 12:49:06.000000000 -0800
+++ linux-2.6.16-rc1/arch/x86_64/kernel/smpboot.c 2006-01-20 12:49:32.000000000 -0800
@@ -59,6 +59,7 @@
#include <asm/nmi.h>
#include <asm/irq.h>
#include <asm/hw_irq.h>
+#include <asm/numa.h>
/* Number of siblings per CPU package */
int smp_num_siblings = 1;
@@ -890,6 +891,7 @@ do_rest:
if (boot_error) {
cpu_clear(cpu, cpu_callout_map); /* was set here (do_boot_cpu()) */
clear_bit(cpu, &cpu_initialized); /* was set by cpu_init() */
+ clear_node_cpumask(cpu); /* was set by numa_add_cpu */
cpu_clear(cpu, cpu_present_map);
cpu_clear(cpu, cpu_possible_map);
x86_cpu_to_apicid[cpu] = BAD_APICID;
@@ -1187,6 +1189,7 @@ void remove_cpu_from_maps(void)
cpu_clear(cpu, cpu_callout_map);
cpu_clear(cpu, cpu_callin_map);
clear_bit(cpu, &cpu_initialized); /* was set by cpu_init() */
+ clear_node_cpumask(cpu);
}
int __cpu_disable(void)
Index: linux-2.6.16-rc1/include/asm-x86_64/numa.h
===================================================================
--- linux-2.6.16-rc1.orig/include/asm-x86_64/numa.h 2006-01-20 12:49:06.000000000 -0800
+++ linux-2.6.16-rc1/include/asm-x86_64/numa.h 2006-01-20 12:49:32.000000000 -0800
@@ -22,8 +22,15 @@ extern void numa_set_node(int cpu, int n
extern unsigned char apicid_to_node[256];
#ifdef CONFIG_NUMA
extern void __init init_cpu_to_node(void);
+
+static inline void clear_node_cpumask(int cpu)
+{
+ clear_bit(cpu, &node_to_cpumask[cpu_to_node(cpu)]);
+}
+
#else
#define init_cpu_to_node() do {} while (0)
+#define clear_node_cpumask(cpu) do {} while (0)
#endif
#define NUMA_NO_NODE 0xff
next reply other threads:[~2006-01-21 1:57 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-01-21 1:56 Ravikiran G Thirumalai [this message]
2006-01-22 21:21 ` [patch] Fix the node cpumask of a cpu going down Andi Kleen
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=20060121015633.GD3573@localhost.localdomain \
--to=kiran@scalex86.org \
--cc=ak@suse.de \
--cc=alokk@calsoftinc.com \
--cc=clameter@engr.sgi.com \
--cc=discuss@x86-64.org \
--cc=linux-kernel@vger.kernel.org \
--cc=shai@scalex86.org \
--cc=tony.luck@intel.com \
/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