Re: [PATCH] ACPI: processor idle: Practically limit "Dummy wait" workaround to old Intel systems

Re: [PATCH] ACPI: processor idle: Practically limit "Dummy wait" workaround to old Intel systems

[Rafael J. Wysocki]

On 9/22/2022 8:47 PM, Dave Hansen wrote:
> Old, circa 2002 chipsets have a bug: they don't go idle when they are
> supposed to.  So, a workaround was added to slow the CPU down and
> ensure that the CPU waits a bit for the chipset to actually go idle.
> This workaround is ancient and has been in place in some form since
> the original kernel ACPI implementation.
>
> But, this workaround is very painful on modern systems.  The "inl()"
> can take thousands of cycles (see Link: for some more detailed
> numbers and some fun kernel archaeology).
>
> First and foremost, modern systems should not be using this code.
> Typical Intel systems have not used it in over a decade because it is
> horribly inferior to MWAIT-based idle.
>
> Despite this, people do seem to be tripping over this workaround on
> AMD system today.
>
> Limit the "dummy wait" workaround to Intel systems.  Keep Modern AMD
> systems from tripping over the workaround.  Remotely modern Intel
> systems use intel_idle instead of this code and will, in practice,
> remain unaffected by the dummy wait.
>
> Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
> Cc: Len Brown <lenb@kernel.org>
> Cc: Mario Limonciello <Mario.Limonciello@amd.com>
> Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
> Cc: Borislav Petkov <bp@alien8.de>
> Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> Reported-by: K Prateek Nayak <kprateek.nayak@amd.com>
> Link: https://lore.kernel.org/all/20220921063638.2489-1-kprateek.nayak@amd.com/

Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

or do you want me to pick this up?


> ---
>   drivers/acpi/processor_idle.c | 23 ++++++++++++++++++++---
>   1 file changed, 20 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
> index 16a1663d02d4..9f40917c49ef 100644
> --- a/drivers/acpi/processor_idle.c
> +++ b/drivers/acpi/processor_idle.c
> @@ -531,10 +531,27 @@ static void wait_for_freeze(void)
>   	/* No delay is needed if we are in guest */
>   	if (boot_cpu_has(X86_FEATURE_HYPERVISOR))
>   		return;
> +	/*
> +	 * Modern (>=Nehalem) Intel systems use ACPI via intel_idle,
> +	 * not this code.  Assume that any Intel systems using this
> +	 * are ancient and may need the dummy wait.  This also assumes
> +	 * that the motivating chipset issue was Intel-only.
> +	 */
> +	if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL)
> +		return;
>   #endif
> -	/* Dummy wait op - must do something useless after P_LVL2 read
> -	   because chipsets cannot guarantee that STPCLK# signal
> -	   gets asserted in time to freeze execution properly. */
> +	/*
> +	 * Dummy wait op - must do something useless after P_LVL2 read
> +	 * because chipsets cannot guarantee that STPCLK# signal gets
> +	 * asserted in time to freeze execution properly
> +	 *
> +	 * This workaround has been in place since the original ACPI
> +	 * implementation was merged, circa 2002.
> +	 *
> +	 * If a profile is pointing to this instruction, please first
> +	 * consider moving your system to a more modern idle
> +	 * mechanism.
> +	 */
>   	inl(acpi_gbl_FADT.xpm_timer_block.address);
>   }
>   

Re: [PATCH] ACPI: processor idle: Practically limit "Dummy wait" workaround to old Intel systems

[Limonciello, Mario]

On 9/22/2022 13:53, Rafael J. Wysocki wrote:
> On 9/22/2022 8:47 PM, Dave Hansen wrote:
>> Old, circa 2002 chipsets have a bug: they don't go idle when they are
>> supposed to.  So, a workaround was added to slow the CPU down and
>> ensure that the CPU waits a bit for the chipset to actually go idle.
>> This workaround is ancient and has been in place in some form since
>> the original kernel ACPI implementation.
>>
>> But, this workaround is very painful on modern systems.  The "inl()"
>> can take thousands of cycles (see Link: for some more detailed
>> numbers and some fun kernel archaeology).
>>
>> First and foremost, modern systems should not be using this code.
>> Typical Intel systems have not used it in over a decade because it is
>> horribly inferior to MWAIT-based idle.
>>
>> Despite this, people do seem to be tripping over this workaround on
>> AMD system today.
>>
>> Limit the "dummy wait" workaround to Intel systems.  Keep Modern AMD
>> systems from tripping over the workaround.  Remotely modern Intel
>> systems use intel_idle instead of this code and will, in practice,
>> remain unaffected by the dummy wait.
>>
>> Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
>> Cc: Len Brown <lenb@kernel.org>
>> Cc: Mario Limonciello <Mario.Limonciello@amd.com>
>> Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
>> Cc: Borislav Petkov <bp@alien8.de>
>> Suggested-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>> Reported-by: K Prateek Nayak <kprateek.nayak@amd.com>
>> Link: 
>> https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore.kernel.org%2Fall%2F20220921063638.2489-1-kprateek.nayak%40amd.com%2F&amp;data=05%7C01%7CMario.Limonciello%40amd.com%7C8460d9ef3add45bf571408da9ccbc58a%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637994696248641733%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=23k2wKPZaBrgOTtcHw8ByNzfsus1RSsdXMlCACjl%2Bmc%3D&amp;reserved=0 

If agreeable, I think this should be @stable too.

Either way:

Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>

>>
> 
> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> 
> or do you want me to pick this up?
> 
> 
>> ---
>>   drivers/acpi/processor_idle.c | 23 ++++++++++++++++++++---
>>   1 file changed, 20 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/acpi/processor_idle.c 
>> b/drivers/acpi/processor_idle.c
>> index 16a1663d02d4..9f40917c49ef 100644
>> --- a/drivers/acpi/processor_idle.c
>> +++ b/drivers/acpi/processor_idle.c
>> @@ -531,10 +531,27 @@ static void wait_for_freeze(void)
>>       /* No delay is needed if we are in guest */
>>       if (boot_cpu_has(X86_FEATURE_HYPERVISOR))
>>           return;
>> +    /*
>> +     * Modern (>=Nehalem) Intel systems use ACPI via intel_idle,
>> +     * not this code.  Assume that any Intel systems using this
>> +     * are ancient and may need the dummy wait.  This also assumes
>> +     * that the motivating chipset issue was Intel-only.
>> +     */
>> +    if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL)
>> +        return;
>>   #endif
>> -    /* Dummy wait op - must do something useless after P_LVL2 read
>> -       because chipsets cannot guarantee that STPCLK# signal
>> -       gets asserted in time to freeze execution properly. */
>> +    /*
>> +     * Dummy wait op - must do something useless after P_LVL2 read
>> +     * because chipsets cannot guarantee that STPCLK# signal gets
>> +     * asserted in time to freeze execution properly
>> +     *
>> +     * This workaround has been in place since the original ACPI
>> +     * implementation was merged, circa 2002.
>> +     *
>> +     * If a profile is pointing to this instruction, please first
>> +     * consider moving your system to a more modern idle
>> +     * mechanism.
>> +     */
>>       inl(acpi_gbl_FADT.xpm_timer_block.address);
>>   }
> 
> 

Re: [PATCH] ACPI: processor idle: Practically limit "Dummy wait" workaround to old Intel systems

[Dave Hansen]

On 9/22/22 11:53, Rafael J. Wysocki wrote:
> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> 
> or do you want me to pick this up?

I'll just stick it in x86/urgent.

It's modifying code in a x86 #ifdef.  I'll call it a small enclave of
sovereign x86 territory in ACPI land, just like an embassy. ;)

Re: [PATCH] ACPI: processor idle: Practically limit "Dummy wait" workaround to old Intel systems

[Kim Phillips]

On 9/22/22 2:01 PM, Dave Hansen wrote:
> On 9/22/22 11:53, Rafael J. Wysocki wrote:
>> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>>
>> or do you want me to pick this up?
> 
> I'll just stick it in x86/urgent.
> 
> It's modifying code in a x86 #ifdef.  I'll call it a small enclave of
> sovereign x86 territory in ACPI land, just like an embassy. ;)

Can it be cc:stable@vger.kernel.org, since it applies cleanly as far
back as this v5.4 commit?:

commit fa583f71a99c85e52781ed877c82c8757437b680

Author: Yin Fengwei <fengwei.yin@intel.com>

Date:   Thu Oct 24 15:04:20 2019 +0800



     ACPI: processor_idle: Skip dummy wait if kernel is in guest


Thanks,

Kim

Re: [PATCH] ACPI: processor idle: Practically limit "Dummy wait" workaround to old Intel systems

[Dave Hansen]

On 9/23/22 11:36, Kim Phillips wrote:
> Can it be cc:stable@vger.kernel.org, since it applies cleanly as far
> back as this v5.4 commit?:

I just sent the pull request to Linus for this fix.  I realized that I
didn't tag it for stable@.  If it gets applied, I'll send a request for
it to be picked up for stable@, via "Option 2":

> Option 2
> ********
> 
> After the patch has been merged to Linus' tree, send an email to
> stable@vger.kernel.org containing the subject of the patch, the commit ID,
> why you think it should be applied, and what kernel version you wish it to
> be applied to.

Sorry about that.