public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Mike Galbraith <bitbucket@online.de>,
	Len Brown <len.brown@intel.com>, Borislav Petkov <bp@alien8.de>,
	"H. Peter Anvin" <hpa@zytor.com>, Ian Malone <ibmalone@gmail.com>,
	Josh Boyer <jwboyer@redhat.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Mike Galbraith <efault@gmx.de>,
	Peter Zijlstra <peterz@infradead.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@kernel.org>
Subject: [PATCH 3.10 06/65] sched/idle/x86: Restore mwait_idle() to fix boot hangs, to improve power savings and to improve performance
Date: Sat,  2 May 2015 21:03:36 +0200	[thread overview]
Message-ID: <20150502190115.038330647@linuxfoundation.org> (raw)
In-Reply-To: <20150502190114.555225285@linuxfoundation.org>

3.10-stable review patch.  If anyone has any objections, please let me know.

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

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

commit b253149b843f89cd300cbdbea27ce1f847506f99 upstream.

In Linux-3.9 we removed the mwait_idle() loop:

  69fb3676df33 ("x86 idle: remove mwait_idle() and "idle=mwait" cmdline param")

The reasoning was that modern machines should be sufficiently
happy during the boot process using the default_idle() HALT
loop, until cpuidle loads and either acpi_idle or intel_idle
invoke the newer MWAIT-with-hints idle loop.

But two machines reported problems:

 1. Certain Core2-era machines support MWAIT-C1 and HALT only.
    MWAIT-C1 is preferred for optimal power and performance.
    But if they support just C1, cpuidle never loads and
    so they use the boot-time default idle loop forever.

 2. Some laptops will boot-hang if HALT is used,
    but will boot successfully if MWAIT is used.
    This appears to be a hidden assumption in BIOS SMI,
    that is presumably valid on the proprietary OS
    where the BIOS was validated.

       https://bugzilla.kernel.org/show_bug.cgi?id=60770

So here we effectively revert the patch above, restoring
the mwait_idle() loop.  However, we don't bother restoring
the idle=mwait cmdline parameter, since it appears to add
no value.

Maintainer notes:

  For 3.9, simply revert 69fb3676df
  for 3.10, patch -F3 applies, fuzz needed due to __cpuinit use in
  context For 3.11, 3.12, 3.13, this patch applies cleanly

Tested-by: Mike Galbraith <bitbucket@online.de>
Signed-off-by: Len Brown <len.brown@intel.com>
Acked-by: Mike Galbraith <bitbucket@online.de>
Cc: <stable@vger.kernel.org> # 3.9+
Cc: Borislav Petkov <bp@alien8.de>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Ian Malone <ibmalone@gmail.com>
Cc: Josh Boyer <jwboyer@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/345254a551eb5a6a866e048d7ab570fd2193aca4.1389763084.git.len.brown@intel.com
[ Ported to recent kernels. ]
[ Mike: 3.10 backport ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 arch/x86/kernel/process.c |   49 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 49 insertions(+)

--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -398,6 +398,52 @@ static void amd_e400_idle(void)
 		default_idle();
 }
 
+/*
+ * Intel Core2 and older machines prefer MWAIT over HALT for C1.
+ * We can't rely on cpuidle installing MWAIT, because it will not load
+ * on systems that support only C1 -- so the boot default must be MWAIT.
+ *
+ * Some AMD machines are the opposite, they depend on using HALT.
+ *
+ * So for default C1, which is used during boot until cpuidle loads,
+ * use MWAIT-C1 on Intel HW that has it, else use HALT.
+ */
+static int prefer_mwait_c1_over_halt(const struct cpuinfo_x86 *c)
+{
+	if (c->x86_vendor != X86_VENDOR_INTEL)
+		return 0;
+
+	if (!cpu_has(c, X86_FEATURE_MWAIT))
+		return 0;
+
+	return 1;
+}
+
+/*
+ * MONITOR/MWAIT with no hints, used for default default C1 state.
+ * This invokes MWAIT with interrutps enabled and no flags,
+ * which is backwards compatible with the original MWAIT implementation.
+ */
+
+static void mwait_idle(void)
+{
+	if (!current_set_polling_and_test()) {
+		if (static_cpu_has(X86_FEATURE_CLFLUSH_MONITOR)) {
+			mb();
+			clflush((void *)&current_thread_info()->flags);
+			mb();
+		}
+
+		__monitor((void *)&current_thread_info()->flags, 0, 0);
+		if (!need_resched())
+			__sti_mwait(0, 0);
+		else
+			local_irq_enable();
+	} else
+		local_irq_enable();
+	__current_clr_polling();
+}
+
 void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c)
 {
 #ifdef CONFIG_SMP
@@ -411,6 +457,9 @@ void __cpuinit select_idle_routine(const
 		/* E400: APIC timer interrupt does not wake up CPU from C1e */
 		pr_info("using AMD E400 aware idle routine\n");
 		x86_idle = amd_e400_idle;
+	} else if (prefer_mwait_c1_over_halt(c)) {
+		pr_info("using mwait in idle threads\n");
+		x86_idle = mwait_idle;
 	} else
 		x86_idle = default_idle;
 }



  parent reply	other threads:[~2015-05-02 19:50 UTC|newest]

Thread overview: 83+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-05-02 19:03 [PATCH 3.10 00/65] 3.10.77-stable review Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 01/65] ip_forward: Drop frames with attached skb->sk Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 02/65] tcp: fix possible deadlock in tcp_send_fin() Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 03/65] tcp: avoid looping " Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 04/65] Btrfs: fix log tree corruption when fs mounted with -o discard Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 05/65] Btrfs: fix inode eviction infinite loop after cloning into it Greg Kroah-Hartman
2015-05-02 19:03 ` Greg Kroah-Hartman [this message]
2015-05-02 19:03 ` [PATCH 3.10 07/65] usb: gadget: composite: enable BESL support Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 08/65] KVM: s390: Zero out current VMDB of STSI before including level3 data Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 09/65] s390/hibernate: fix save and restore of kernel text section Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 11/65] MIPS: Hibernate: flush TLB entries earlier Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 12/65] cdc-wdm: fix endianness bug in debug statements Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 13/65] spi: spidev: fix possible arithmetic overflow for multi-transfer message Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 14/65] ring-buffer: Replace this_cpu_*() with __this_cpu_*() Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 15/65] power_supply: lp8788-charger: Fix leaked power supply on probe fail Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 16/65] ARM: 8320/1: fix integer overflow in ELF_ET_DYN_BASE Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 17/65] ARM: S3C64XX: Use fixed IRQ bases to avoid conflicts on Cragganmore Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 18/65] usb: phy: Find the right match in devm_usb_phy_match Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 19/65] usb: define a generic USB_RESUME_TIMEOUT macro Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 20/65] usb: host: r8a66597: use new USB_RESUME_TIMEOUT Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 21/65] usb: host: isp116x: " Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 22/65] usb: host: xhci: " Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 23/65] usb: host: sl811: " Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 24/65] usb: core: hub: " Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 25/65] ALSA: emu10k1: dont deadlock in proc-functions Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 26/65] Input: elantech - fix absolute mode setting on some ASUS laptops Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 27/65] fs/binfmt_elf.c: fix bug in loading of PIE binaries Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 28/65] ptrace: fix race between ptrace_resume() and wait_task_stopped() Greg Kroah-Hartman
2015-05-02 19:03 ` [PATCH 3.10 29/65] rtlwifi: rtl8192cu: Add new USB ID Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 30/65] rtlwifi: rtl8192cu: Add new device ID Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 31/65] parport: disable PC-style parallel port support on cris Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 32/65] drivers: parport: Kconfig: exclude h8300 for PARPORT_PC Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 33/65] console: Disable VGA text console support on cris Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 34/65] video: vgacon: Dont build on arm64 Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 35/65] arm64: kernel: compiling issue, need delete read_current_timer() Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 36/65] ext4: make fsync to sync parent dir in no-journal for real this time Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 37/65] powerpc/perf: Cap 64bit userspace backtraces to PERF_MAX_STACK_DEPTH Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 38/65] tools/power turbostat: Use $(CURDIR) instead of $(PWD) and add support for O= option in Makefile Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 39/65] UBI: account for bitflips in both the VID header and data Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 40/65] UBI: fix out of bounds write Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 41/65] UBI: initialize LEB number variable Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 42/65] UBI: fix check for "too many bytes" Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 43/65] scsi: storvsc: Fix a bug in copy_from_bounce_buffer() Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 44/65] drivers: parport: Kconfig: exclude arm64 for PARPORT_PC Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 45/65] ACPICA: Utilities: split IO address types from data type models Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 46/65] xtensa: xtfpga: fix hardware lockup caused by LCD driver Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 47/65] xtensa: provide __NR_sync_file_range2 instead of __NR_sync_file_range Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 48/65] gpio: mvebu: Fix mask/unmask managment per irq chip type Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 49/65] Drivers: hv: vmbus: Fix a bug in the error path in vmbus_open() Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 50/65] mvsas: fix panic on expander attached SATA devices Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 51/65] [media] stk1160: Make sure current buffer is released Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 52/65] IB/core: disallow registering 0-sized memory region Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 53/65] IB/core: dont disallow registering region starting at 0x0 Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 54/65] IB/mlx4: Fix WQE LSO segment calculation Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 55/65] tracing: Handle ftrace_dump() atomic context in graph_trace_open() Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 56/65] i2c: core: Export bus recovery functions Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 57/65] drm/radeon: fix doublescan modes (v2) Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 58/65] drm/i915: cope with large i2c transfers Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 59/65] RCU pathwalk breakage when running into a symlink overmounting something Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 60/65] ksoftirqd: Enable IRQs and call cond_resched() before poking RCU Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 61/65] e1000: add dummy allocator to fix race condition between mtu change and netpoll Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 62/65] lib: memzero_explicit: use barrier instead of OPTIMIZER_HIDE_VAR Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 63/65] wl18xx: show rx_frames_per_rates as an array as it really is Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 64/65] C6x: time: Ensure consistency in __init Greg Kroah-Hartman
2015-05-02 19:04 ` [PATCH 3.10 65/65] memstick: mspro_block: add missing curly braces Greg Kroah-Hartman
2015-05-03 19:49 ` [PATCH 3.10 00/65] 3.10.77-stable review Guenter Roeck
2015-05-04  4:40   ` Guenter Roeck
2015-05-04  8:09     ` Heiko Carstens
2015-05-04  9:54       ` Jiri Slaby
2015-05-04 10:40         ` Heiko Carstens
2015-05-04 21:38           ` Greg Kroah-Hartman
2015-05-04 18:51       ` Guenter Roeck
2015-05-05  4:42     ` Guenter Roeck
2015-05-05 21:59       ` Greg Kroah-Hartman
2015-05-04 21:40   ` Greg Kroah-Hartman
2015-05-04 16:16 ` Shuah Khan
2015-05-04 21:42   ` Greg Kroah-Hartman
2015-05-04 22:13 ` Shuah Khan
2015-05-04 22:24   ` Guenter Roeck
2015-05-05 21:58   ` Greg Kroah-Hartman
2015-05-05 22:05 ` Greg Kroah-Hartman
2015-05-06  3:31   ` Guenter Roeck
2015-05-06 16:01   ` Shuah Khan

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=20150502190115.038330647@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=bitbucket@online.de \
    --cc=bp@alien8.de \
    --cc=efault@gmx.de \
    --cc=hpa@zytor.com \
    --cc=ibmalone@gmail.com \
    --cc=jwboyer@redhat.com \
    --cc=len.brown@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.org \
    --cc=stable@vger.kernel.org \
    --cc=tglx@linutronix.de \
    --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