[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <498136FC.2020206@kernel.org>
Date: Wed, 28 Jan 2009 20:56:28 -0800
From: Yinghai Lu <yinghai@...nel.org>
To: Ingo Molnar <mingo@...e.hu>, Thomas Gleixner <tglx@...utronix.de>,
"H. Peter Anvin" <hpa@...or.com>,
Andrew Morton <akpm@...ux-foundation.org>
CC: "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"Dave Jones" <davej@...hat.com>,
"Langsdorf, Mark" <mark.langsdorf@....com>,
Rusty Russell <rusty@...tcorp.com.au>,
Mike Travis <travis@....com>
Subject: [PATCH] x86/powernow: fix cpus_allowed brokage when acpi=off
Impact: fix current->cpus_allowed overwriting.
on one AMD Fam10h SMP system:
when checking numactl output, when acpi=off
found output is not right.
# numactl --show
policy: default
preferred node: current
physcpubind: 5 7 10 11 12 13
cpubind: 1 2 3
nodebind: 1 2 3
membind: 0 1 2 3
it turns out in powernowk8_cpu_init there is path ( ACPI is off or _PSS is not there)
it will try to overwrite current->cpus_allowed with uninitialized oldmask.
caused by
| commit 2fdf66b491ac706657946442789ec644cc317e1a
| Author: Rusty Russell <rusty@...tcorp.com.au>
| Date: Wed Dec 31 18:08:47 2008 -0800
|
| cpumask: convert shared_cpu_map in acpi_processor* structs to cpumask_var_t
need to get oldmask early.
also remove some wrong warning when acpi is disabled.
and don't call exit_acpi if _PSS is not found.
with patch get numactl correct.
# numactl --show
policy: default
preferred node: current
physcpubind: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
cpubind: 0 1 2 3
nodebind: 0 1 2 3
membind: 0 1 2 3
Signed-off-by: Yinghai Lu <yinghai@...nel.org>
---
arch/x86/kernel/cpu/cpufreq/powernow-k8.c | 17 +++++++++++++----
1 file changed, 13 insertions(+), 4 deletions(-)
Index: linux-2.6/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
===================================================================
--- linux-2.6.orig/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
+++ linux-2.6/arch/x86/kernel/cpu/cpufreq/powernow-k8.c
@@ -1124,6 +1124,7 @@ static int powernowk8_verify(struct cpuf
static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
{
struct powernow_k8_data *data;
+ int k8_cpu_acpi_inited = 0;
cpumask_t oldmask;
int rc;
@@ -1141,14 +1142,19 @@ static int __cpuinit powernowk8_cpu_init
data->cpu = pol->cpu;
data->currpstate = HW_PSTATE_INVALID;
+ oldmask = current->cpus_allowed;
rc = powernow_k8_cpu_init_acpi(data);
- if (rc) {
+ if (!rc) {
+ k8_cpu_acpi_inited = 1;
+ } else {
/*
* Use the PSB BIOS structure. This is only availabe on
* an UP version, and is deprecated by AMD.
*/
if (num_online_cpus() != 1) {
+ if (acpi_disabled)
+ goto err_out;
#ifndef CONFIG_ACPI_PROCESSOR
printk(KERN_ERR PFX "ACPI Processor support is required "
"for SMP systems but is absent. Please load the "
@@ -1164,6 +1170,8 @@ static int __cpuinit powernowk8_cpu_init
goto err_out;
}
if (pol->cpu != 0) {
+ if (acpi_disabled)
+ goto err_out;
printk(KERN_ERR FW_BUG PFX "No ACPI _PSS objects for "
"CPU other than CPU0. Complain to your BIOS "
"vendor.\n");
@@ -1176,7 +1184,6 @@ static int __cpuinit powernowk8_cpu_init
}
/* only run on specific CPU from here on */
- oldmask = current->cpus_allowed;
set_cpus_allowed_ptr(current, &cpumask_of_cpu(pol->cpu));
if (smp_processor_id() != pol->cpu) {
@@ -1218,7 +1225,8 @@ static int __cpuinit powernowk8_cpu_init
/* min/max the cpu is capable of */
if (cpufreq_frequency_table_cpuinfo(pol, data->powernow_table)) {
printk(KERN_ERR FW_BUG PFX "invalid powernow_table\n");
- powernow_k8_cpu_exit_acpi(data);
+ if (k8_cpu_acpi_inited)
+ powernow_k8_cpu_exit_acpi(data);
kfree(data->powernow_table);
kfree(data);
return -EINVAL;
@@ -1238,7 +1246,8 @@ static int __cpuinit powernowk8_cpu_init
err_out:
set_cpus_allowed_ptr(current, &oldmask);
- powernow_k8_cpu_exit_acpi(data);
+ if (k8_cpu_acpi_inited)
+ powernow_k8_cpu_exit_acpi(data);
kfree(data);
return -ENODEV;
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists