[PATCH] x86/perf: Fixed kernel panic during boot on Nano processor.

[PATCH] x86/perf: Fixed kernel panic during boot on Nano processor.

[Cody Yao-oc]

From: CodyYao-oc <CodyYao-oc@zhaoxin.com>

Nano processor may not fully support rdpmc instruction, it works well
for reading general pmc counter, but will lead to GP(general protection)
when accessing fixed pmc counter. Futhermore, family/model information
is same between Nano processor and ZX-C processor, it leads to zhaoxin
pmu driver is wrongly loaded for Nano processor, which resulting boot
kernal fail.

To solve this problem, stepping information will be checked to distinguish
between Nano processor and ZX-C processor.

[https://bugzilla.kernel.org/show_bug.cgi?id=212389]

Reported-by: Arjan <8vvbbqzo567a@nospam.xutrox.com>
Signed-off-by: CodyYao-oc <CodyYao-oc@zhaoxin.com>
---
 arch/x86/events/zhaoxin/core.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/x86/events/zhaoxin/core.c b/arch/x86/events/zhaoxin/core.c
index 949d845c922b..cef1de251613 100644
--- a/arch/x86/events/zhaoxin/core.c
+++ b/arch/x86/events/zhaoxin/core.c
@@ -541,7 +541,8 @@ __init int zhaoxin_pmu_init(void)
 
 	switch (boot_cpu_data.x86) {
 	case 0x06:
-		if (boot_cpu_data.x86_model == 0x0f || boot_cpu_data.x86_model == 0x19) {
+		if ((boot_cpu_data.x86_model == 0x0f && boot_cpu_data.x86_stepping >= 0x0e) ||
+			boot_cpu_data.x86_model == 0x19) {
 
 			x86_pmu.max_period = x86_pmu.cntval_mask >> 1;
 
-- 
2.17.1

Re: [PATCH] x86/perf: Fixed kernel panic during boot on Nano processor.

[Codyyao-oc]

Dear Mingo and Peter,

Thank you for taking your precious time to read this letter, I am very 
graterful.


Last month, I fixed the bug that boot failed on Nano processor which 
introduced by

"Fixes: 3a4ac121c2ca ("x86/perf: Add hardware performance events support 
for Zhaoxin CPU.")"

with Arjan's help and submitted this patch.  But I haven't got back. 
Greatly appreciate if you could kindly

check it and reply at your convenience.

Many Thanks!

Cody

On 2021/6/7 上午9:31, Cody Yao-oc wrote:
> From: CodyYao-oc <CodyYao-oc@zhaoxin.com>
>
> Nano processor may not fully support rdpmc instruction, it works well
> for reading general pmc counter, but will lead to GP(general protection)
> when accessing fixed pmc counter. Futhermore, family/model information
> is same between Nano processor and ZX-C processor, it leads to zhaoxin
> pmu driver is wrongly loaded for Nano processor, which resulting boot
> kernal fail.
>
> To solve this problem, stepping information will be checked to distinguish
> between Nano processor and ZX-C processor.
>
> [https://bugzilla.kernel.org/show_bug.cgi?id=212389]
>
> Reported-by: Arjan <8vvbbqzo567a@nospam.xutrox.com>
> Signed-off-by: CodyYao-oc <CodyYao-oc@zhaoxin.com>
> ---
>   arch/x86/events/zhaoxin/core.c | 3 ++-
>   1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/arch/x86/events/zhaoxin/core.c b/arch/x86/events/zhaoxin/core.c
> index 949d845c922b..cef1de251613 100644
> --- a/arch/x86/events/zhaoxin/core.c
> +++ b/arch/x86/events/zhaoxin/core.c
> @@ -541,7 +541,8 @@ __init int zhaoxin_pmu_init(void)
>   
>   	switch (boot_cpu_data.x86) {
>   	case 0x06:
> -		if (boot_cpu_data.x86_model == 0x0f || boot_cpu_data.x86_model == 0x19) {
> +		if ((boot_cpu_data.x86_model == 0x0f && boot_cpu_data.x86_stepping >= 0x0e) ||
> +			boot_cpu_data.x86_model == 0x19) {
>   
>   			x86_pmu.max_period = x86_pmu.cntval_mask >> 1;
>   

Re: [PATCH] x86/perf: Fixed kernel panic during boot on Nano processor.

[Thorsten Leemhuis]

Hi perf maintainers and Codyyao-oc! What happened to below patch, which
was posted many moons ago? It wasn't merged afaics. Did it fall through
the cracks or is there something wrong with it?

I'm asking because a user who reported this regression asked what's up:
https://bugzilla.kernel.org/show_bug.cgi?id=212389

On 30.06.21 06:38, Codyyao-oc wrote:
> 
> Thank you for taking your precious time to read this letter, I am very
> graterful.
> 
> Last month, I fixed the bug that boot failed on Nano processor which
> introduced by
> 
> "Fixes: 3a4ac121c2ca ("x86/perf: Add hardware performance events support
> for Zhaoxin CPU.")"

Just BTW: You want to add that tag to your patch description.

> with Arjan's help and submitted this patch.  But I haven't got back.
> Greatly appreciate if you could kindly
> 
> check it and reply at your convenience.
> 
> Many Thanks!
> 
> Cody
> 
> On 2021/6/7 上午9:31, Cody Yao-oc wrote:
>> From: CodyYao-oc <CodyYao-oc@zhaoxin.com>
>>
>> Nano processor may not fully support rdpmc instruction, it works well
>> for reading general pmc counter, but will lead to GP(general protection)
>> when accessing fixed pmc counter. Futhermore, family/model information
>> is same between Nano processor and ZX-C processor, it leads to zhaoxin
>> pmu driver is wrongly loaded for Nano processor, which resulting boot
>> kernal fail.
>>
>> To solve this problem, stepping information will be checked to
>> distinguish
>> between Nano processor and ZX-C processor.

And this...

>> [https://bugzilla.kernel.org/show_bug.cgi?id=212389]

...should look like this:

Link: https://bugzilla.kernel.org/show_bug.cgi?id=212389

Ohh, and you might want to add this to ensure backporting:

Cc: <stable@vger.kernel.org> # 5.10.x

Guess adding those and submitting it again might be wise and help to
finally get this regression resolved.

Ciao, Thorsten

>> Reported-by: Arjan <8vvbbqzo567a@nospam.xutrox.com>
>> Signed-off-by: CodyYao-oc <CodyYao-oc@zhaoxin.com>
>> ---
>>   arch/x86/events/zhaoxin/core.c | 3 ++-
>>   1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/x86/events/zhaoxin/core.c
>> b/arch/x86/events/zhaoxin/core.c
>> index 949d845c922b..cef1de251613 100644
>> --- a/arch/x86/events/zhaoxin/core.c
>> +++ b/arch/x86/events/zhaoxin/core.c
>> @@ -541,7 +541,8 @@ __init int zhaoxin_pmu_init(void)
>>         switch (boot_cpu_data.x86) {
>>       case 0x06:
>> -        if (boot_cpu_data.x86_model == 0x0f ||
>> boot_cpu_data.x86_model == 0x19) {
>> +        if ((boot_cpu_data.x86_model == 0x0f &&
>> boot_cpu_data.x86_stepping >= 0x0e) ||
>> +            boot_cpu_data.x86_model == 0x19) {
>>                 x86_pmu.max_period = x86_pmu.cntval_mask >> 1;
>>   

Re: [PATCH] x86/perf: Fixed kernel panic during boot on Nano processor.

[Peter Zijlstra]

On Mon, Jun 07, 2021 at 09:31:09AM +0800, Cody Yao-oc wrote:
> From: CodyYao-oc <CodyYao-oc@zhaoxin.com>
> 
> Nano processor may not fully support rdpmc instruction, it works well
> for reading general pmc counter, but will lead to GP(general protection)
> when accessing fixed pmc counter. Futhermore, family/model information
> is same between Nano processor and ZX-C processor, it leads to zhaoxin
> pmu driver is wrongly loaded for Nano processor, which resulting boot
> kernal fail.
> 
> To solve this problem, stepping information will be checked to distinguish
> between Nano processor and ZX-C processor.
> 
> [https://bugzilla.kernel.org/show_bug.cgi?id=212389]
> 
> Reported-by: Arjan <8vvbbqzo567a@nospam.xutrox.com>
> Signed-off-by: CodyYao-oc <CodyYao-oc@zhaoxin.com>

I suppose I'll queue it up for perf/urgent post -rc1

Re: [PATCH] x86/perf: Fixed kernel panic during boot on Nano processor.

[Peter Zijlstra]

On Mon, Jun 07, 2021 at 09:31:09AM +0800, Cody Yao-oc wrote:
> From: CodyYao-oc <CodyYao-oc@zhaoxin.com>
> 
> Nano processor may not fully support rdpmc instruction, it works well
> for reading general pmc counter, but will lead to GP(general protection)
> when accessing fixed pmc counter. Futhermore, family/model information
> is same between Nano processor and ZX-C processor, it leads to zhaoxin
> pmu driver is wrongly loaded for Nano processor, which resulting boot
> kernal fail.
> 
> To solve this problem, stepping information will be checked to distinguish
> between Nano processor and ZX-C processor.
> 
> [https://bugzilla.kernel.org/show_bug.cgi?id=212389]
> 
> Reported-by: Arjan <8vvbbqzo567a@nospam.xutrox.com>
> Signed-off-by: CodyYao-oc <CodyYao-oc@zhaoxin.com>

*sigh*.. so this email address doesn't exist, as such I can't apply this
patch. Consider it dropped.

Re: [PATCH] x86/perf: Fixed kernel panic during boot on Nano processor.

[Arjan]

On 13-10-2022 17:07, Peter Zijlstra wrote:
> On Mon, Jun 07, 2021 at 09:31:09AM +0800, Cody Yao-oc wrote:
>> From: CodyYao-oc <CodyYao-oc@zhaoxin.com>
>>
>> Nano processor may not fully support rdpmc instruction, it works well
>> for reading general pmc counter, but will lead to GP(general protection)
>> when accessing fixed pmc counter. Futhermore, family/model information
>> is same between Nano processor and ZX-C processor, it leads to zhaoxin
>> pmu driver is wrongly loaded for Nano processor, which resulting boot
>> kernal fail.
>>
>> To solve this problem, stepping information will be checked to distinguish
>> between Nano processor and ZX-C processor.
>>
>> [https://bugzilla.kernel.org/show_bug.cgi?id=212389]
>>
>> Reported-by: Arjan <8vvbbqzo567a@nospam.xutrox.com>
>> Signed-off-by: CodyYao-oc <CodyYao-oc@zhaoxin.com>
> 
> *sigh*.. so this email address doesn't exist, as such I can't apply this
> patch. Consider it dropped.

If it's about my email address: The address exists and works.
If the nospam part bothers you, that part can be left out. You may leave the reported-by line out if you want to.

Re: [PATCH] x86/perf: Fixed kernel panic during boot on Nano processor.

[Peter Zijlstra]

On Sun, Oct 16, 2022 at 11:53:14AM +0200, Arjan wrote:
> On 13-10-2022 17:07, Peter Zijlstra wrote:
> > On Mon, Jun 07, 2021 at 09:31:09AM +0800, Cody Yao-oc wrote:
> > > From: CodyYao-oc <CodyYao-oc@zhaoxin.com>
> > > 
> > > Nano processor may not fully support rdpmc instruction, it works well
> > > for reading general pmc counter, but will lead to GP(general protection)
> > > when accessing fixed pmc counter. Futhermore, family/model information
> > > is same between Nano processor and ZX-C processor, it leads to zhaoxin
> > > pmu driver is wrongly loaded for Nano processor, which resulting boot
> > > kernal fail.
> > > 
> > > To solve this problem, stepping information will be checked to distinguish
> > > between Nano processor and ZX-C processor.
> > > 
> > > [https://bugzilla.kernel.org/show_bug.cgi?id=212389]
> > > 
> > > Reported-by: Arjan <8vvbbqzo567a@nospam.xutrox.com>
> > > Signed-off-by: CodyYao-oc <CodyYao-oc@zhaoxin.com>
> > 
> > *sigh*.. so this email address doesn't exist, as such I can't apply this
> > patch. Consider it dropped.
> 
> If it's about my email address: The address exists and works.
> If the nospam part bothers you, that part can be left out. You may leave the reported-by line out if you want to.

The SoB address (CodyYao-oc@zhaoxin.com) bounced for me -- since that's
the patch author that is somewhat important.

Re: [PATCH] x86/perf: Fixed kernel panic during boot on Nano processor.

[8vvbbqzo567a]

On 16-10-2022 11:59, Peter Zijlstra wrote:
> On Sun, Oct 16, 2022 at 11:53:14AM +0200, Arjan wrote:
>> On 13-10-2022 17:07, Peter Zijlstra wrote:
>>> On Mon, Jun 07, 2021 at 09:31:09AM +0800, Cody Yao-oc wrote:
>>>> From: CodyYao-oc <CodyYao-oc@zhaoxin.com>
>>>>
>>>> Nano processor may not fully support rdpmc instruction, it works well
>>>> for reading general pmc counter, but will lead to GP(general protection)
>>>> when accessing fixed pmc counter. Futhermore, family/model information
>>>> is same between Nano processor and ZX-C processor, it leads to zhaoxin
>>>> pmu driver is wrongly loaded for Nano processor, which resulting boot
>>>> kernal fail.
>>>>
>>>> To solve this problem, stepping information will be checked to distinguish
>>>> between Nano processor and ZX-C processor.
>>>>
>>>> [https://bugzilla.kernel.org/show_bug.cgi?id=212389]
>>>>
>>>> Reported-by: Arjan <8vvbbqzo567a@nospam.xutrox.com>
>>>> Signed-off-by: CodyYao-oc <CodyYao-oc@zhaoxin.com>
>>>
>>> *sigh*.. so this email address doesn't exist, as such I can't apply this
>>> patch. Consider it dropped.
>>
>> If it's about my email address: The address exists and works.
>> If the nospam part bothers you, that part can be left out. You may leave the reported-by line out if you want to.
> 
> The SoB address (CodyYao-oc@zhaoxin.com) bounced for me -- since that's
> the patch author that is somewhat important.

It now bounced for me too.
It was still valid when Cody submitted the patch in 2021, because we
exchanged messages while debugging and testing the patch.

[PATCH] x86/perf: Fixed kernel panic during boot on Nano processor.

[silviazhao-oc]

Nano processor may not fully support rdpmc instruction, it works well
for reading general pmc counter, but will lead GP(general protection)
when accessing fixed pmc counter. Furthermore, family/mode information
is same between Nano processor and ZX-C processor, it leads to zhaoxin
pmu driver is wrongly loaded for Nano processor, which resulting boot
kernal fail.

To solve this problem, stepping information will be checked to distinguish
between Nano processor and ZX-C processor.

[https://bugzilla.kernel.org/show_bug.cgi?id=212389]

Reported-by: Arjan <8vvbbqzo567a@nospam.xutrox.com>
Signed-off-by: silviazhao-oc <silviazhao-oc@zhaoxin.com>
---
 arch/x86/events/zhaoxin/core.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/x86/events/zhaoxin/core.c b/arch/x86/events/zhaoxin/core.c
index 949d845c922b..cef1de251613 100644
--- a/arch/x86/events/zhaoxin/core.c
+++ b/arch/x86/events/zhaoxin/core.c
@@ -541,7 +541,8 @@ __init int zhaoxin_pmu_init(void)
 
 	switch (boot_cpu_data.x86) {
 	case 0x06:
-		if (boot_cpu_data.x86_model == 0x0f || boot_cpu_data.x86_model == 0x19) {
+		if ((boot_cpu_data.x86_model == 0x0f && boot_cpu_data.x86_stepping >= 0x0e) ||
+			boot_cpu_data.x86_model == 0x19) {
 
 			x86_pmu.max_period = x86_pmu.cntval_mask >> 1;
 
-- 
2.17.1

Re: [PATCH] x86/perf: Fixed kernel panic during boot on Nano processor.

[Borislav Petkov]

On Thu, Nov 03, 2022 at 11:23:04AM +0800, silviazhao-oc wrote:
> Nano processor may not fully support rdpmc instruction,

What does that even mean? Not fully support?

> it works well for reading general pmc counter, but will lead
> GP(general protection) when accessing fixed pmc counter.

RDPMC will #GP when the perf counter specified cannot be read.

AFAICT, that is RCX: 0000000040000001 which looks like perf counter
index 1 with INTEL_PMC_FIXED_RDPMC_BASE ORed in.

> Furthermore, family/mode information is same between Nano processor
> and ZX-C processor, it leads to zhaoxin pmu driver is wrongly loaded
> for Nano processor, which resulting boot kernal fail.

So *that* is the real problem - it tries to access perf counters
thinking it is running on architectural perf counters implementation but
nano doesn't have that.

> To solve this problem, stepping information will be checked to distinguish
> between Nano processor and ZX-C processor.

Why doesn't that ZXC thing doesn't have a CPUID flag to check instead of
looking at models and steppings and thus confusing it with a nano CPU?

> [https://bugzilla.kernel.org/show_bug.cgi?id=212389]
> 
> Reported-by: Arjan <8vvbbqzo567a@nospam.xutrox.com>

Does Arjan have a last name?

> Signed-off-by: silviazhao-oc <silviazhao-oc@zhaoxin.com>

I'm assuming your name is properly spelled "Silvia Zhao" and not in a
single word with a "-oc" string appended at the end, yes?

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette