Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1500309 > unrolled thread

[PATCH v3 2/4] Add enabling of the R3 MWAIT during boot for KNL

Started byGrzegorz Andrejczuk <grzegorz.andrejczuk@intel.com>
First post2016-10-13 16:50 +0200
Last post2016-10-13 17:40 +0200
Articles 2 — 2 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  [PATCH v3 2/4] Add enabling of the R3 MWAIT during boot for KNL Grzegorz Andrejczuk <grzegorz.andrejczuk@intel.com> - 2016-10-13 16:50 +0200
    Re: [PATCH v3 2/4] Add enabling of the R3 MWAIT during boot for  KNL Thomas Gleixner <tglx@linutronix.de> - 2016-10-13 17:40 +0200

#1500309 — [PATCH v3 2/4] Add enabling of the R3 MWAIT during boot for KNL

FromGrzegorz Andrejczuk <grzegorz.andrejczuk@intel.com>
Date2016-10-13 16:50 +0200
Subject[PATCH v3 2/4] Add enabling of the R3 MWAIT during boot for KNL
Message-ID<srPux-4ML-13@gated-at.bofh.it>
If processor is Intel Xeon Phi we enable user-level mwait feature.
Enabling this feature suppreses invalid-opcode error, when MONITOR/MWAIT
is called from ring 3.

Signed-off-by: Grzegorz Andrejczuk <grzegorz.andrejczuk@intel.com>
---
 arch/x86/kernel/cpu/intel.c | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index fcd484d..464df22 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -26,6 +26,8 @@
 #include <asm/apic.h>
 #endif
 
+static int phi_r3mwait_disabled;
+
 /*
  * Just in case our CPU detection goes bad, or you have a weird system,
  * allow a way to override the automatic disabling of MPX.
@@ -61,6 +63,34 @@ void check_mpx_erratum(struct cpuinfo_x86 *c)
 	}
 }
 
+static int __init phir3mwait_disable(char *__unused)
+{
+	phi_r3mwait_disabled = 1;
+
+	return 1;
+}
+__setup("phir3mwait=disable", phir3mwait_disable);
+
+static void __init probe_xeon_phi_r3mwait(struct cpuinfo_x86 *c)
+{
+	if (phi_r3mwait_disabled)
+		return;
+
+	/*
+	* Setting ring 3 MONITOR/MWAIT for all threads
+	* when CPU is Xeon Phi Family x200
+	* This can be disabled with phir3mwait=disable cmdline switch.
+	* We preserve the reserved values and set only 2nd bit.
+	*/
+	if (c->x86 == 6 && c->x86_model == INTEL_FAM6_XEON_PHI_KNL) {
+		u64 msr;
+
+		rdmsrl(MSR_PHI_MISC_THD_FEATURE, msr);
+		msr |= MSR_PHI_MISC_THD_FEATURE_R3MWAIT;
+		wrmsrl(MSR_PHI_MISC_THD_FEATURE, msr);
+	}
+}
+
 static void early_init_intel(struct cpuinfo_x86 *c)
 {
 	u64 misc_enable;
@@ -211,6 +241,7 @@ static void early_init_intel(struct cpuinfo_x86 *c)
 	}
 
 	check_mpx_erratum(c);
+	probe_xeon_phi_r3mwait(c);
 }
 
 #ifdef CONFIG_X86_32
-- 
2.5.1

[toc] | [next] | [standalone]


#1500360 — Re: [PATCH v3 2/4] Add enabling of the R3 MWAIT during boot for KNL

FromThomas Gleixner <tglx@linutronix.de>
Date2016-10-13 17:40 +0200
SubjectRe: [PATCH v3 2/4] Add enabling of the R3 MWAIT during boot for KNL
Message-ID<srQgV-5jv-9@gated-at.bofh.it>
In reply to#1500309
On Thu, 13 Oct 2016, Grzegorz Andrejczuk wrote:
>  
> +static int phi_r3mwait_disabled;

__read_mostly please.

> +
>  /*
>   * Just in case our CPU detection goes bad, or you have a weird system,
>   * allow a way to override the automatic disabling of MPX.
> @@ -61,6 +63,34 @@ void check_mpx_erratum(struct cpuinfo_x86 *c)
>  	}
>  }
>  
> +static int __init phir3mwait_disable(char *__unused)
> +{
> +	phi_r3mwait_disabled = 1;
> +
> +	return 1;
> +}
> +__setup("phir3mwait=disable", phir3mwait_disable);

This needs proper documentation in Documentation/kernel-parameters.txt

> +static void __init probe_xeon_phi_r3mwait(struct cpuinfo_x86 *c)
> +{
> +	if (phi_r3mwait_disabled)
> +		return;
> +
> +	/*
> +	* Setting ring 3 MONITOR/MWAIT for all threads
> +	* when CPU is Xeon Phi Family x200


> +	* This can be disabled with phir3mwait=disable cmdline switch.

And why is this information useful in this comment? 

> +	* We preserve the reserved values and set only 2nd bit.

This last sentence does not make any sense at all.

Thanks,

	tglx

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web