Re: [Patch2/2] fix wrong proc cpuinfo on x64

2007-11-06 Thread Andreas Herrmann 
On Tue, Nov 06, 2007 at 04:28:20PM +0800, Zou Nan hai wrote:
> in 2.6.24-rc1 kernel, 
> The /proc/cpuinfo display is wrong.
> 
> Another issue is that it will display bogus cpus with wrong information
> if the kernel is compiled with a big CONFIG_NR_CPU.
> 
> That is because before a cpu in cpu_present_map is up, c->cpu_index of
> that cpu is 0.
> thus the cpu_online(c->cpu_index) check in show_cpuinfo is invalid.

This issue should already be fixed with that patch

http://marc.info/?l=linux-kernel=119394201230954

> This patch will let cpuinfo_op use cpu_online_map instead of
> cpu_present_map to iterate cpus.

BTW, the problem affects i386 as well.


Regards,

Andreas

-- 
Operating | AMD Saxony Limited Liability Company & Co. KG,
  System  | Wilschdorfer Landstr. 101, 01109 Dresden, Germany
 Research | Register Court Dresden: HRA 4896, General Partner authorized
  Center  | to represent: AMD Saxony LLC (Wilmington, Delaware, US)
  (OSRC)  | General Manager of AMD Saxony LLC: Dr. Hans-R. Deppe, Thomas McCoy



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[Patch2/2] fix wrong proc cpuinfo on x64

2007-11-06 Thread Zou Nan hai 
in 2.6.24-rc1 kernel, 
The /proc/cpuinfo display is wrong.

Another issue is that it will display bogus cpus with wrong information
if the kernel is compiled with a big CONFIG_NR_CPU.

That is because before a cpu in cpu_present_map is up, c->cpu_index of
that cpu is 0.
thus the cpu_online(c->cpu_index) check in show_cpuinfo is invalid.

This patch will let cpuinfo_op use cpu_online_map instead of
cpu_present_map to iterate cpus.


Signed-off-by: Zou Nan hai <[EMAIL PROTECTED]>

--- linux-2.6.24-rc1/arch/x86/kernel/setup_64.c 2007-10-29 22:03:05.0 
-0400
+++ b/arch/x86/kernel/setup_64.c2007-11-05 23:09:06.0 -0500
@@ -1078,8 +1078,6 @@ static int show_cpuinfo(struct seq_file 
 
 
 #ifdef CONFIG_SMP
-   if (!cpu_online(c->cpu_index))
-   return 0;
cpu = c->cpu_index;
 #endif
 
@@ -1171,15 +1169,15 @@ static int show_cpuinfo(struct seq_file 
 static void *c_start(struct seq_file *m, loff_t *pos)
 {
if (*pos == 0)  /* just in case, cpu 0 is not the first */
-   *pos = first_cpu(cpu_possible_map);
-   if ((*pos) < NR_CPUS && cpu_possible(*pos))
+   *pos = first_cpu(cpu_online_map);
+   if ((*pos) < NR_CPUS && cpu_online(*pos))
return _data(*pos);
return NULL;
 }
 
 static void *c_next(struct seq_file *m, void *v, loff_t *pos)
 {
-   *pos = next_cpu(*pos, cpu_possible_map);
+   *pos = next_cpu(*pos, cpu_online_map);
return c_start(m, pos);
 }
 

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[Patch2/2] fix wrong proc cpuinfo on x64

2007-11-06 Thread Zou Nan hai 
in 2.6.24-rc1 kernel, 
The /proc/cpuinfo display is wrong.

Another issue is that it will display bogus cpus with wrong information
if the kernel is compiled with a big CONFIG_NR_CPU.

That is because before a cpu in cpu_present_map is up, c-cpu_index of
that cpu is 0.
thus the cpu_online(c-cpu_index) check in show_cpuinfo is invalid.

This patch will let cpuinfo_op use cpu_online_map instead of
cpu_present_map to iterate cpus.


Signed-off-by: Zou Nan hai [EMAIL PROTECTED]

--- linux-2.6.24-rc1/arch/x86/kernel/setup_64.c 2007-10-29 22:03:05.0 
-0400
+++ b/arch/x86/kernel/setup_64.c2007-11-05 23:09:06.0 -0500
@@ -1078,8 +1078,6 @@ static int show_cpuinfo(struct seq_file 
 
 
 #ifdef CONFIG_SMP
-   if (!cpu_online(c-cpu_index))
-   return 0;
cpu = c-cpu_index;
 #endif
 
@@ -1171,15 +1169,15 @@ static int show_cpuinfo(struct seq_file 
 static void *c_start(struct seq_file *m, loff_t *pos)
 {
if (*pos == 0)  /* just in case, cpu 0 is not the first */
-   *pos = first_cpu(cpu_possible_map);
-   if ((*pos)  NR_CPUS  cpu_possible(*pos))
+   *pos = first_cpu(cpu_online_map);
+   if ((*pos)  NR_CPUS  cpu_online(*pos))
return cpu_data(*pos);
return NULL;
 }
 
 static void *c_next(struct seq_file *m, void *v, loff_t *pos)
 {
-   *pos = next_cpu(*pos, cpu_possible_map);
+   *pos = next_cpu(*pos, cpu_online_map);
return c_start(m, pos);
 }
 

-
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [Patch2/2] fix wrong proc cpuinfo on x64

2007-11-06 Thread Andreas Herrmann 
On Tue, Nov 06, 2007 at 04:28:20PM +0800, Zou Nan hai wrote:
 in 2.6.24-rc1 kernel, 
 The /proc/cpuinfo display is wrong.
 
 Another issue is that it will display bogus cpus with wrong information
 if the kernel is compiled with a big CONFIG_NR_CPU.
 
 That is because before a cpu in cpu_present_map is up, c-cpu_index of
 that cpu is 0.
 thus the cpu_online(c-cpu_index) check in show_cpuinfo is invalid.

This issue should already be fixed with that patch

http://marc.info/?l=linux-kernelm=119394201230954

 This patch will let cpuinfo_op use cpu_online_map instead of
 cpu_present_map to iterate cpus.

BTW, the problem affects i386 as well.


Regards,

Andreas

-- 
Operating | AMD Saxony Limited Liability Company  Co. KG,
  System  | Wilschdorfer Landstr. 101, 01109 Dresden, Germany
 Research | Register Court Dresden: HRA 4896, General Partner authorized
  Center  | to represent: AMD Saxony LLC (Wilmington, Delaware, US)
  (OSRC)  | General Manager of AMD Saxony LLC: Dr. Hans-R. Deppe, Thomas McCoy



-
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/