linux-kernel - Re: [PATCH v3 3/3] intel_idle: Provide enter

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <b6ab357c-1562-4035-ad3a-2159d2c8c1fa@intel.com>
Date: Fri, 8 Nov 2024 08:21:24 -0800
From: Dave Hansen <dave.hansen@...el.com>
To: Patryk Wlazlyn <patryk.wlazlyn@...ux.intel.com>, x86@...nel.org
Cc: linux-kernel@...r.kernel.org, linux-pm@...r.kernel.org,
 rafael.j.wysocki@...el.com, len.brown@...el.com,
 artem.bityutskiy@...ux.intel.com, dave.hansen@...ux.intel.com
Subject: Re: [PATCH v3 3/3] intel_idle: Provide enter_dead() handler for SRF

On 11/8/24 04:29, Patryk Wlazlyn wrote:
> Intel's Sierra Forest report two C6 substates in cpuid leaf 5:
>     C6S  (hint 0x22)
>     C6SP (hint 0x23)
> 
> Hints 0x20 and 0x21 are skipped entirely, causing the generic
> implementation in mwait_play_dead() to compute the wrong hint, when
> looking for the deepest cstate. As a result, package with an offlined
> CPU can never reach PC6.

This series has said multiple times how the old algorithm is wrong.  But
it never actually _fixed_ the bad algorithm, only worked around it.

Does mwait_play_dead() itself need to get fixed?

> Define the enter_dead() handler for SRF.

This effectively gets the mwait hints from ______ instead of using the
calculation in mwait_play_dead().

> diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
> index 9aab7abc2ae9..bd67959e5e8b 100644
> --- a/drivers/idle/intel_idle.c
> +++ b/drivers/idle/intel_idle.c
> @@ -56,6 +56,7 @@
>  #include <asm/mwait.h>
>  #include <asm/spec-ctrl.h>
>  #include <asm/fpu/api.h>
> +#include <asm/smp.h>
>  
>  #define INTEL_IDLE_VERSION "0.5.1"
>  
> @@ -221,6 +222,17 @@ static __cpuidle int intel_idle_s2idle(struct cpuidle_device *dev,
>  	return 0;
>  }
>  
> +static __cpuidle int intel_idle_enter_dead(struct cpuidle_device *dev,
> +					   int index)
> +{
> +	struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev);
> +	struct cpuidle_state *state = &drv->states[index];
> +	unsigned long eax = flg2MWAIT(state->flags);
> +
> +	/* Retruns only in case of an error. */

		^ returns?

> +	return mwait_play_dead_with_hint(eax);
> +}
> +
>  /*
>   * States are indexed by the cstate number,
>   * which is also the index into the MWAIT hint array.
> @@ -1303,6 +1315,7 @@ static struct cpuidle_state srf_cstates[] __initdata = {
>  		.exit_latency = 1,
>  		.target_residency = 1,
>  		.enter = &intel_idle,
> +		.enter_dead = &intel_idle_enter_dead,
>  		.enter_s2idle = intel_idle_s2idle, },
>  	{
>  		.name = "C1E",
> @@ -1311,6 +1324,7 @@ static struct cpuidle_state srf_cstates[] __initdata = {
>  		.exit_latency = 2,
>  		.target_residency = 10,
>  		.enter = &intel_idle,
> +		.enter_dead = &intel_idle_enter_dead,
>  		.enter_s2idle = intel_idle_s2idle, },
>  	{
>  		.name = "C6S",
> @@ -1319,6 +1333,7 @@ static struct cpuidle_state srf_cstates[] __initdata = {
>  		.exit_latency = 270,
>  		.target_residency = 700,
>  		.enter = &intel_idle,
> +		.enter_dead = &intel_idle_enter_dead,
>  		.enter_s2idle = intel_idle_s2idle, },
>  	{
>  		.name = "C6SP",
> @@ -1327,6 +1342,7 @@ static struct cpuidle_state srf_cstates[] __initdata = {
>  		.exit_latency = 310,
>  		.target_residency = 900,
>  		.enter = &intel_idle,
> +		.enter_dead = &intel_idle_enter_dead,
>  		.enter_s2idle = intel_idle_s2idle, },
>  	{
>  		.enter = NULL }