From: Kamal Mostafa <kamal@canonical.com>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org,
kernel-team@lists.ubuntu.com
Cc: 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>, Jiri Slaby <jslaby@suse.cz>,
Kamal Mostafa <kamal@canonical.com>
Subject: [PATCH 3.13.y-ckt 73/95] sched/idle/x86: Restore mwait_idle() to fix boot hangs, to improve power savings and to improve performance
Date: Wed, 27 May 2015 12:00:21 -0700 [thread overview]
Message-ID: <1432753243-16151-74-git-send-email-kamal@canonical.com> (raw)
In-Reply-To: <1432753243-16151-1-git-send-email-kamal@canonical.com>
3.13.11-ckt21 -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: 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: Jiri Slaby <jslaby@suse.cz>
Signed-off-by: Kamal Mostafa <kamal@canonical.com>
---
arch/x86/kernel/process.c | 49 +++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 49 insertions(+)
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 3fb8d95..c5db2a4 100644
--- 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 *)¤t_thread_info()->flags);
+ mb();
+ }
+
+ __monitor((void *)¤t_thread_info()->flags, 0, 0);
+ if (!need_resched())
+ __sti_mwait(0, 0);
+ else
+ local_irq_enable();
+ } else
+ local_irq_enable();
+ __current_clr_polling();
+}
+
void select_idle_routine(const struct cpuinfo_x86 *c)
{
#ifdef CONFIG_SMP
@@ -411,6 +457,9 @@ void select_idle_routine(const struct cpuinfo_x86 *c)
/* 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;
}
--
1.9.1
next prev parent reply other threads:[~2015-05-27 19:02 UTC|newest]
Thread overview: 96+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-05-27 18:59 [3.13.y-ckt stable] Linux 3.13.11-ckt21 stable review Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 01/95] [3.13-stable only] Revert "gianfar: Carefully free skbs in functions called by netpoll." Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 02/95] Bluetooth: ath3k: Add support Atheros AR5B195 combo Mini PCIe card Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 03/95] power_supply: twl4030_madc: Check return value of power_supply_register Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 04/95] power_supply: lp8788-charger: Fix leaked power supply on probe fail Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 05/95] ARM: dts: dove: Fix uart[23] reg property Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 06/95] xtensa: xtfpga: fix hardware lockup caused by LCD driver Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 07/95] Drivers: hv: vmbus: Fix a bug in the error path in vmbus_open() Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 08/95] xtensa: provide __NR_sync_file_range2 instead of __NR_sync_file_range Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 09/95] KVM: s390: Zero out current VMDB of STSI before including level3 data Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 10/95] usb: musb: core: fix TX/RX endpoint order Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 11/95] drm/radeon: fix doublescan modes (v2) Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 12/95] usb: phy: Find the right match in devm_usb_phy_match Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 13/95] x86: kvm: Revert "remove sched notifier for cross-cpu migrations" Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 14/95] tools lib traceevent kbuffer: Remove extra update to data pointer in PADDING Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 15/95] ring-buffer: Replace this_cpu_*() with __this_cpu_*() Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 16/95] ASoC: wm8741: Fix rates constraints values Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 17/95] cdc-wdm: fix endianness bug in debug statements Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 18/95] staging: panel: fix lcd type Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 19/95] UBI: account for bitflips in both the VID header and data Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 20/95] UBI: fix out of bounds write Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 21/95] UBI: initialize LEB number variable Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 22/95] UBI: fix check for "too many bytes" Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 23/95] ARM: S3C64XX: Use fixed IRQ bases to avoid conflicts on Cragganmore Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 24/95] ASoC: davinci-evm: drop un-necessary remove function Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 25/95] iscsi-target: Convert iscsi_thread_set usage to kthread.h Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 26/95] Drivers: hv: vmbus: Don't wait after requesting offers Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 27/95] Btrfs: fix log tree corruption when fs mounted with -o discard Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 28/95] btrfs: don't accept bare namespace as a valid xattr Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 29/95] ARM: 8320/1: fix integer overflow in ELF_ET_DYN_BASE Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 30/95] rtlwifi: rtl8192cu: Add new USB ID Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 31/95] MIPS: Hibernate: flush TLB entries earlier Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 32/95] ASoC: cs4271: Increase delay time after reset Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 33/95] [media] stk1160: Make sure current buffer is released Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 34/95] mnt: Improve the umount_tree flags Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 35/95] ext4: make fsync to sync parent dir in no-journal for real this time Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 36/95] Input: elantech - fix absolute mode setting on some ASUS laptops Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 37/95] usb: define a generic USB_RESUME_TIMEOUT macro Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 38/95] usb: host: xhci: use new USB_RESUME_TIMEOUT Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 39/95] usb: host: ehci: " Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 40/95] usb: host: uhci: " Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 41/95] usb: musb: " Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 42/95] usb: host: isp116x: " Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 43/95] usb: host: fotg210: " Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 44/95] usb: host: fusbh200: " Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 45/95] usb: host: oxu210hp: " Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 46/95] usb: host: r8a66597: " Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 47/95] usb: host: sl811: " Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 48/95] usb: dwc2: hcd: " Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 49/95] usb: isp1760: " Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 50/95] usb: core: hub: " Kamal Mostafa
2015-05-27 18:59 ` [PATCH 3.13.y-ckt 51/95] iser-target: Fix possible deadlock in RDMA_CM connection error Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 52/95] x86: vdso: fix pvclock races with task migration Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 53/95] gpio: mvebu: Fix mask/unmask managment per irq chip type Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 54/95] scsi: storvsc: Fix a bug in copy_from_bounce_buffer() Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 55/95] ALSA: emu10k1: don't deadlock in proc-functions Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 56/95] xtensa: ISS: fix locking in TAP network adapter Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 57/95] s390/hibernate: fix save and restore of kernel text section Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 58/95] Btrfs: fix inode eviction infinite loop after extent_same ioctl Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 59/95] Btrfs: fix inode eviction infinite loop after cloning into it Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 60/95] powerpc/perf: Cap 64bit userspace backtraces to PERF_MAX_STACK_DEPTH Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 61/95] ACPICA: Utilities: split IO address types from data type models Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 62/95] drm/i915: Dont enable CS_PARSER_ERROR interrupts at all Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 63/95] target: Fix COMPARE_AND_WRITE with SG_TO_MEM_NOALLOC handling Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 64/95] mm/hugetlb: use pmd_page() in follow_huge_pmd() Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 65/95] fs/binfmt_elf.c: fix bug in loading of PIE binaries Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 66/95] IB/core: disallow registering 0-sized memory region Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 67/95] IB/core: don't disallow registering region starting at 0x0 Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 68/95] ptrace: fix race between ptrace_resume() and wait_task_stopped() Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 69/95] mvsas: fix panic on expander attached SATA devices Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 70/95] drm/i915: cope with large i2c transfers Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 71/95] RCU pathwalk breakage when running into a symlink overmounting something Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 72/95] compal-laptop: Check return value of power_supply_register Kamal Mostafa
2015-05-27 19:00 ` Kamal Mostafa [this message]
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 74/95] nfs: don't call blocking operations while !TASK_RUNNING Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 75/95] nfs: fix high load average due to callback thread sleeping Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 76/95] e1000: add dummy allocator to fix race condition between mtu change and netpoll Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 77/95] wl18xx: show rx_frames_per_rates as an array as it really is Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 78/95] lib: memzero_explicit: use barrier instead of OPTIMIZER_HIDE_VAR Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 79/95] driver core: bus: Goto appropriate labels on failure in bus_add_device Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 80/95] C6x: time: Ensure consistency in __init Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 81/95] crypto: omap-aes - Fix support for unequal lengths Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 82/95] jhash: Update jhash_[321]words functions to use correct initval Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 83/95] KVM: use slowpath for cross page cached accesses Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 84/95] powerpc: Fix missing L2 cache size in /sys/devices/system/cpu Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 85/95] NFS: fix BUG() crash in notify_change() with patch to chown_common() Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 86/95] i2c: core: Export bus recovery functions Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 87/95] IB/mlx4: Fix WQE LSO segment calculation Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 88/95] mlx5_core: Fix PowerPC support Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 89/95] mlx5: wrong page mask if CONFIG_ARCH_DMA_ADDR_T_64BIT enabled for 32Bit architectures Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 90/95] skbuff: Do not scrub skb mark within the same name space Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 91/95] firmware/ihex2fw.c: restore missing default in switch statement Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 92/95] memstick: mspro_block: add missing curly braces Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 93/95] tools/power turbostat: Use $(CURDIR) instead of $(PWD) and add support for O= option in Makefile Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 94/95] ext4: fix data corruption caused by unwritten and delayed extents Kamal Mostafa
2015-05-27 19:00 ` [PATCH 3.13.y-ckt 95/95] powerpc: Add vr save/restore functions Kamal Mostafa
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=1432753243-16151-74-git-send-email-kamal@canonical.com \
--to=kamal@canonical.com \
--cc=bp@alien8.de \
--cc=efault@gmx.de \
--cc=hpa@zytor.com \
--cc=ibmalone@gmail.com \
--cc=jslaby@suse.cz \
--cc=jwboyer@redhat.com \
--cc=kernel-team@lists.ubuntu.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
all inboxes | Powered by JetHome®