[PATCH] x86, cpuinfo fix cpu_data(0) x86_model_id field truncation

[PATCH] x86, cpuinfo fix cpu_data(0) x86_model_id field truncation

[Prarit Bhargava]

When comparing 'model name' fields in /proc/cpuinfo it was noticed that
a simple test comparing the model name fields was failing.  After some
simple investigation it was noticed that, in fact, the model name fields
are different for each processor.  Processor 0's model name field had
white space removed, while the other processors did not.

Another way of seeing this behaviour is to convert spaces into underscores
in the output of /proc/cpuinfo,

[thetango@prarit ~]# grep "^model name" /proc/cpuinfo | uniq -c | sed 's/\ /_/g'
______1_model_name      :_AMD_Opteron(TM)_Processor_6272
_____63_model_name      :_AMD_Opteron(TM)_Processor_6272_________________

which shows two different model name fields even though they should be the
same.

This occurs because the kernel calls strim() on cpu 0's x86_model_id field
to output a pretty message to the console in print_cpu_info(), and as a
result truncates the whitespace at the end of the x86_model_id field.

The x86_model_id field should be the same for the same processors.  This
patch adds a local copy of the x86_model_id field and modifies the value
there so that the output in dmesg still looks like

smpboot: CPU0: AMD Opteron(TM) Processor 6272 (fam: 15, model: 01, stepping: 02)

and the x86_model_id field is correct across all processors:

[thetango@prarit ~]# grep "^model name" /proc/cpuinfo | uniq -c | sed 's/\ /_/g'
_____64_model_name      :_AMD_Opteron(TM)_Processor_6272_________________

Signed-off-by: Prarit Bhargava <prarit@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
---
 arch/x86/kernel/cpu/common.c |    8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index a62cf04..61ac5c3 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -1110,6 +1110,7 @@ __setup("noclflush", setup_noclflush);
 void print_cpu_info(struct cpuinfo_x86 *c)
 {
 	const char *vendor = NULL;
+	char x86_model_id[64];
 
 	if (c->x86_vendor < X86_VENDOR_NUM) {
 		vendor = this_cpu->c_vendor;
@@ -1121,9 +1122,10 @@ void print_cpu_info(struct cpuinfo_x86 *c)
 	if (vendor && !strstr(c->x86_model_id, vendor))
 		printk(KERN_CONT "%s ", vendor);
 
-	if (c->x86_model_id[0])
-		printk(KERN_CONT "%s", strim(c->x86_model_id));
-	else
+	if (c->x86_model_id[0]) {
+		strcpy(x86_model_id, c->x86_model_id);
+		printk(KERN_CONT "%s", strim(x86_model_id));
+	} else
 		printk(KERN_CONT "%d86", c->x86);
 
 	printk(KERN_CONT " (fam: %02x, model: %02x", c->x86, c->x86_model);
-- 
1.7.9.3

Re: [PATCH] x86, cpuinfo fix cpu_data(0) x86_model_id field truncation

[Borislav Petkov]

On Mon, May 18, 2015 at 02:21:00PM -0400, Prarit Bhargava wrote:
> When comparing 'model name' fields in /proc/cpuinfo it was noticed that
> a simple test comparing the model name fields was failing.  After some
> simple investigation it was noticed that, in fact, the model name fields
> are different for each processor.  Processor 0's model name field had
> white space removed, while the other processors did not.
> 
> Another way of seeing this behaviour is to convert spaces into underscores
> in the output of /proc/cpuinfo,
> 
> [thetango@prarit ~]# grep "^model name" /proc/cpuinfo | uniq -c | sed 's/\ /_/g'
> ______1_model_name      :_AMD_Opteron(TM)_Processor_6272
> _____63_model_name      :_AMD_Opteron(TM)_Processor_6272_________________
> 
> which shows two different model name fields even though they should be the
> same.
> 
> This occurs because the kernel calls strim() on cpu 0's x86_model_id field

I'd actually prefer this much simpler patch:

---
diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c
index e7d8c7608471..d215e9b26567 100644
--- a/arch/x86/kernel/cpu/proc.c
+++ b/arch/x86/kernel/cpu/proc.c
@@ -67,7 +67,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 		   c->x86_vendor_id[0] ? c->x86_vendor_id : "unknown",
 		   c->x86,
 		   c->x86_model,
-		   c->x86_model_id[0] ? c->x86_model_id : "unknown");
+		   c->x86_model_id[0] ? strim(c->x86_model_id) : "unknown");
 
 	if (c->x86_mask || c->cpuid_level >= 0)
 		seq_printf(m, "stepping\t: %d\n", c->x86_mask);
---

-- 
Regards/Gruss,
    Boris.

ECO tip #101: Trim your mails when you reply.
--

Re: [PATCH] x86, cpuinfo fix cpu_data(0) x86_model_id field truncation

[Brian Gerst]

On Mon, May 18, 2015 at 3:45 PM, Borislav Petkov <bp@alien8.de> wrote:
> On Mon, May 18, 2015 at 02:21:00PM -0400, Prarit Bhargava wrote:
>> When comparing 'model name' fields in /proc/cpuinfo it was noticed that
>> a simple test comparing the model name fields was failing.  After some
>> simple investigation it was noticed that, in fact, the model name fields
>> are different for each processor.  Processor 0's model name field had
>> white space removed, while the other processors did not.
>>
>> Another way of seeing this behaviour is to convert spaces into underscores
>> in the output of /proc/cpuinfo,
>>
>> [thetango@prarit ~]# grep "^model name" /proc/cpuinfo | uniq -c | sed 's/\ /_/g'
>> ______1_model_name      :_AMD_Opteron(TM)_Processor_6272
>> _____63_model_name      :_AMD_Opteron(TM)_Processor_6272_________________
>>
>> which shows two different model name fields even though they should be the
>> same.
>>
>> This occurs because the kernel calls strim() on cpu 0's x86_model_id field
>
> I'd actually prefer this much simpler patch:
>
> ---
> diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c
> index e7d8c7608471..d215e9b26567 100644
> --- a/arch/x86/kernel/cpu/proc.c
> +++ b/arch/x86/kernel/cpu/proc.c
> @@ -67,7 +67,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
>                    c->x86_vendor_id[0] ? c->x86_vendor_id : "unknown",
>                    c->x86,
>                    c->x86_model,
> -                  c->x86_model_id[0] ? c->x86_model_id : "unknown");
> +                  c->x86_model_id[0] ? strim(c->x86_model_id) : "unknown");
>
>         if (c->x86_mask || c->cpuid_level >= 0)
>                 seq_printf(m, "stepping\t: %d\n", c->x86_mask);

The problem here is that strim() modifies the string in place,
replacing the first trailing space with a null.  I think the best
solution is to do the trimming in get_model_name().  It already trims
leading spaces for Intel.

--
Brian Gerst

Re: [PATCH] x86, cpuinfo fix cpu_data(0) x86_model_id field truncation

[Borislav Petkov]

On Mon, May 18, 2015 at 04:27:02PM -0400, Brian Gerst wrote:
> The problem here is that strim() modifies the string in place,
> replacing the first trailing space with a null.  I think the best
> solution is to do the trimming in get_model_name().  It already trims
> leading spaces for Intel.

Sounds good - start from the 48th position forward to the first non \s
char. Yeah.

-- 
Regards/Gruss,
    Boris.

ECO tip #101: Trim your mails when you reply.
--

Re: [PATCH] x86, cpuinfo fix cpu_data(0) x86_model_id field truncation

[H. Peter Anvin]

On 05/18/2015 01:36 PM, Borislav Petkov wrote:
> On Mon, May 18, 2015 at 04:27:02PM -0400, Brian Gerst wrote:
>> The problem here is that strim() modifies the string in place,
>> replacing the first trailing space with a null.  I think the best
>> solution is to do the trimming in get_model_name().  It already trims
>> leading spaces for Intel.
> 
> Sounds good - start from the 48th position forward to the first non \s
> char. Yeah.
> 

Yes, we should trim both leading and trailing spaces.

	-hpa

Re: [PATCH] x86, cpuinfo fix cpu_data(0) x86_model_id field truncation

[Prarit Bhargava]

On 05/18/2015 04:53 PM, H. Peter Anvin wrote:
> On 05/18/2015 01:36 PM, Borislav Petkov wrote:
>> On Mon, May 18, 2015 at 04:27:02PM -0400, Brian Gerst wrote:
>>> The problem here is that strim() modifies the string in place,
>>> replacing the first trailing space with a null.  I think the best
>>> solution is to do the trimming in get_model_name().  It already trims
>>> leading spaces for Intel.
>>
>> Sounds good - start from the 48th position forward to the first non \s
>> char. Yeah.
>>
> 
> Yes, we should trim both leading and trailing spaces.

Oh sorry, I missed this as it just landed in my mailbox.  So a backward search
for the first non \s char.

I can do that and I'll nack my new patch.

P.

> 
> 	-hpa
>