mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Greg KH <gregkh@suse.de>
To: linux-kernel@vger.kernel.org, stable@kernel.org
Cc: stable-review@kernel.org, torvalds@linux-foundation.org,
	akpm@linux-foundation.org, alan@lxorguk.ukuu.org.uk,
	Len Brown <len.brown@intel.com>
Subject: [18/55] intel_idle: disable NHM/WSM HW C-state auto-demotion
Date: Fri, 25 Mar 2011 16:47:11 -0700	[thread overview]
Message-ID: <20110325234840.810402470@clark.kroah.org> (raw)
In-Reply-To: <20110325234853.GA21145@kroah.com>

2.6.37-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Len Brown <len.brown@intel.com>

commit 14796fca2bd22acc73dd0887248d003b0f441d08 upstream.

Hardware C-state auto-demotion is a mechanism where the HW overrides
the OS C-state request, instead demoting to a shallower state,
which is less expensive, but saves less power.

Modern Linux should generally get exactly the states it requests.
In particular, when a CPU is taken off-line, it must not be demoted, else
it can prevent the entire package from reaching deep C-states.

https://bugzilla.kernel.org/show_bug.cgi?id=25252

Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>


---
 arch/x86/include/asm/msr-index.h |    4 ++++
 drivers/idle/intel_idle.c        |   20 ++++++++++++++++++++
 2 files changed, 24 insertions(+)

--- a/arch/x86/include/asm/msr-index.h
+++ b/arch/x86/include/asm/msr-index.h
@@ -36,6 +36,10 @@
 #define MSR_IA32_PERFCTR1		0x000000c2
 #define MSR_FSB_FREQ			0x000000cd
 
+#define MSR_NHM_SNB_PKG_CST_CFG_CTL	0x000000e2
+#define NHM_C3_AUTO_DEMOTE		(1UL << 25)
+#define NHM_C1_AUTO_DEMOTE		(1UL << 26)
+
 #define MSR_MTRRcap			0x000000fe
 #define MSR_IA32_BBL_CR_CTL		0x00000119
 
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -62,6 +62,7 @@
 #include <linux/notifier.h>
 #include <linux/cpu.h>
 #include <asm/mwait.h>
+#include <asm/msr.h>
 
 #define INTEL_IDLE_VERSION "0.4"
 #define PREFIX "intel_idle: "
@@ -85,6 +86,12 @@ static int intel_idle(struct cpuidle_dev
 static struct cpuidle_state *cpuidle_state_table;
 
 /*
+ * Hardware C-state auto-demotion may not always be optimal.
+ * Indicate which enable bits to clear here.
+ */
+static unsigned long long auto_demotion_disable_flags;
+
+/*
  * States are indexed by the cstate number,
  * which is also the index into the MWAIT hint array.
  * Thus C0 is a dummy.
@@ -276,6 +283,15 @@ static struct notifier_block setup_broad
 	.notifier_call = setup_broadcast_cpuhp_notify,
 };
 
+static void auto_demotion_disable(void *dummy)
+{
+	unsigned long long msr_bits;
+
+	rdmsrl(MSR_NHM_SNB_PKG_CST_CFG_CTL, msr_bits);
+	msr_bits &= ~auto_demotion_disable_flags;
+	wrmsrl(MSR_NHM_SNB_PKG_CST_CFG_CTL, msr_bits);
+}
+
 /*
  * intel_idle_probe()
  */
@@ -319,6 +335,8 @@ static int intel_idle_probe(void)
 	case 0x25:	/* Westmere */
 	case 0x2C:	/* Westmere */
 		cpuidle_state_table = nehalem_cstates;
+		auto_demotion_disable_flags =
+			(NHM_C1_AUTO_DEMOTE | NHM_C3_AUTO_DEMOTE);
 		break;
 
 	case 0x1C:	/* 28 - Atom Processor */
@@ -431,6 +449,8 @@ static int intel_idle_cpuidle_devices_in
 			return -EIO;
 		}
 	}
+	if (auto_demotion_disable_flags)
+		smp_call_function(auto_demotion_disable, NULL, 1);
 
 	return 0;
 }



  parent reply	other threads:[~2011-03-26  0:25 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-03-25 23:48 [00/55] 2.6.37.6-stable review Greg KH
2011-03-25 23:46 ` [01/55] ALSA: hda - VIA: Fix stereo mixer recording no sound issue Greg KH
2011-03-25 23:46 ` [02/55] ALSA: hda - VIA: Fix independent headphone " Greg KH
2011-03-25 23:46 ` [03/55] ALSA: hda - VIA: Add missing support for VT1718S in A-A path Greg KH
2011-03-25 23:46 ` [04/55] ALSA: hda - VIA: Fix invalid A-A path volume adjust issue Greg KH
2011-03-25 23:46 ` [05/55] ALSA: hda - VIA: Fix codec type for VT1708BCE at the right timing Greg KH
2011-03-25 23:46 ` [06/55] ALSA: hda - VIA: Fix VT1708 cant build up Headphone control issue Greg KH
2011-03-25 23:47 ` [07/55] ethtool: Compat handling for struct ethtool_rxnfc Greg KH
2011-03-25 23:47 ` [08/55] Revert "slab: Fix missing DEBUG_SLAB last user" Greg KH
2011-03-25 23:47 ` [09/55] aio: wake all waiters when destroying ctx Greg KH
2011-03-25 23:47 ` [10/55] cgroups: if you list_empty() a head then dont list_del() it Greg KH
2011-03-25 23:47 ` [11/55] shmem: let shared anonymous be nonlinear again Greg KH
2011-03-25 23:47 ` [12/55] mm: swap: unlock swapfile inode mutex before closing file on bad swapfiles Greg KH
2011-03-25 23:47 ` [13/55] oom: prevent unnecessary oom kills or kernel panics Greg KH
2011-03-25 23:47 ` [14/55] oom: skip zombies when iterating tasklist Greg KH
2011-03-25 23:47 ` [15/55] oom: avoid deferring oom killer if exiting task is being traced Greg KH
2011-03-25 23:47 ` [16/55] PCI hotplug: acpiphp: set current_state to D0 in register_slot Greg KH
2011-03-25 23:47 ` [17/55] xen: set max_pfn_mapped to the last pfn mapped Greg KH
2011-03-25 23:47 ` Greg KH [this message]
2011-03-25 23:47 ` [19/55] intel_idle: disable Atom/Lincroft HW C-state auto-demotion Greg KH
2011-03-25 23:47 ` [20/55] Prevent rt_sigqueueinfo and rt_tgsigqueueinfo from spoofing the signal code Greg KH
2011-03-25 23:47 ` [21/55] ALSA: HDA: Fix internal mic on Dell E5420/E5520 Greg KH
2011-03-25 23:47 ` [22/55] ext3: skip orphan cleanup on rocompat fs Greg KH
2011-03-25 23:47 ` [23/55] sysctl: restrict write access to dmesg_restrict Greg KH
2011-03-25 23:47 ` [24/55] procfs: fix /proc/<pid>/maps heap check Greg KH
2011-03-25 23:47 ` [25/55] proc: protect mm start_code/end_code in /proc/pid/stat Greg KH
2011-03-25 23:47 ` [26/55] fbcon: Bugfix soft cursor detection in Tile Blitting Greg KH
2011-03-25 23:47 ` [27/55] nfsd41: modify the members value of nfsd4_op_flags Greg KH
2011-03-25 23:47 ` [28/55] nfsd4: minor nfs4state.c reshuffling Greg KH
2011-03-25 23:47 ` [29/55] nfsd4: fix struct file leak Greg KH
2011-03-25 23:47 ` [30/55] nfsd: wrong index used in inner loop Greg KH
2011-03-25 23:47 ` [31/55] [media] uvcvideo: Fix uvc_fixup_video_ctrl() format search Greg KH
2011-03-25 23:47 ` [32/55] [media] uvcvideo: Fix descriptor parsing for video output devices Greg KH
2011-03-25 23:47 ` [33/55] sh: Fix ptrace fpu state initialisation Greg KH
2011-03-25 23:47 ` [34/55] sh: Fix ptrace hw_breakpoint handling Greg KH
2011-03-25 23:47 ` [35/55] USB: Do not pass negative length to snoop_urb() Greg KH
2011-03-25 23:47 ` [36/55] ehci-hcd: Bug fix: dont set a QHs Halt bit Greg KH
2011-03-25 23:47 ` [37/55] USB: uss720 fixup refcount position Greg KH
2011-03-25 23:47 ` [38/55] USB: Fix bad dma problem on WDM device disconnect Greg KH
2011-03-25 23:47 ` [39/55] USB: cdc-acm: fix memory corruption / panic Greg KH
2011-03-25 23:47 ` [40/55] USB: cdc-acm: fix potential null-pointer dereference Greg KH
2011-03-25 23:47 ` [41/55] USB: cdc-acm: fix potential null-pointer dereference on disconnect Greg KH
2011-03-25 23:47 ` [42/55] fs: assign sb->s_bdi to default_backing_dev_info if the bdi is going away Greg KH
2011-03-25 23:47 ` [43/55] Input: xen-kbdfront - advertise either absolute or relative coordinates Greg KH
2011-03-25 23:47 ` [44/55] x86: Cleanup highmap after brk is concluded Greg KH
2011-03-25 23:47 ` [45/55] drm: check for modesetting on modeset ioctls Greg KH
2011-03-25 23:47 ` [46/55] drm/i915: Prevent racy removal of request from client list Greg KH
2011-03-25 23:47 ` [47/55] drm: Fix use-after-free in drm_gem_vm_close() Greg KH
2011-03-25 23:47 ` [48/55] drm/radeon/kms: prefer legacy pll algo for tv-out Greg KH
2011-03-25 23:47 ` [49/55] drm/radeon/kms: fix hardcoded EDID handling Greg KH
2011-03-25 23:47 ` [50/55] perf: Fix tear-down of inherited group events Greg KH
2011-03-25 23:47 ` [51/55] NFS: Fix a hang/infinite loop in nfs_wb_page() Greg KH
2011-03-25 23:47 ` [52/55] SUNRPC: Never reuse the socket port after an xs_close() Greg KH
2011-03-25 23:47 ` [53/55] fs: call security_d_instantiate in d_obtain_alias V2 Greg KH
2011-03-25 23:47 ` [54/55] dcdbas: force SMI to happen when expected Greg KH
2011-03-25 23:47 ` [55/55] ext4: skip orphan cleanup if fs has unknown ROCOMPAT features Greg KH

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20110325234840.810402470@clark.kroah.org \
    --to=gregkh@suse.de \
    --cc=akpm@linux-foundation.org \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=len.brown@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable-review@kernel.org \
    --cc=stable@kernel.org \
    --cc=torvalds@linux-foundation.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®