mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Luis Henriques <luis.henriques@canonical.com>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org,
	kernel-team@lists.ubuntu.com
Cc: Robert Richter <rric@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Ingo Molnar <mingo@kernel.org>,
	Luis Henriques <luis.henriques@canonical.com>
Subject: [PATCH 3.5 10/88] perf/x86/amd/ibs: Fix waking up from S3 for AMD family 10h
Date: Fri,  7 Feb 2014 11:22:07 +0000	[thread overview]
Message-ID: <1391772205-22239-11-git-send-email-luis.henriques@canonical.com> (raw)
In-Reply-To: <1391772205-22239-1-git-send-email-luis.henriques@canonical.com>

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

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

From: Robert Richter <rric@kernel.org>

commit bee09ed91cacdbffdbcd3b05de8409c77ec9fcd6 upstream.

On AMD family 10h we see following error messages while waking up from
S3 for all non-boot CPUs leading to a failed IBS initialization:

 Enabling non-boot CPUs ...
 smpboot: Booting Node 0 Processor 1 APIC 0x1
 [Firmware Bug]: cpu 1, try to use APIC500 (LVT offset 0) for vector 0x400, but the register is already in use for vector 0xf9 on another cpu
 perf: IBS APIC setup failed on cpu #1
 process: Switch to broadcast mode on CPU1
 CPU1 is up
 ...
 ACPI: Waking up from system sleep state S3

Reason for this is that during suspend the LVT offset for the IBS
vector gets lost and needs to be reinialized while resuming.

The offset is read from the IBSCTL msr. On family 10h the offset needs
to be 1 as offset 0 is used for the MCE threshold interrupt, but
firmware assings it for IBS to 0 too. The kernel needs to reprogram
the vector. The msr is a readonly node msr, but a new value can be
written via pci config space access. The reinitialization is
implemented for family 10h in setup_ibs_ctl() which is forced during
IBS setup.

This patch fixes IBS setup after waking up from S3 by adding
resume/supend hooks for the boot cpu which does the offset
reinitialization.

Marking it as stable to let distros pick up this fix.

Signed-off-by: Robert Richter <rric@kernel.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/1389797849-5565-1-git-send-email-rric.net@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
[ luis: backported to 3.5: adjusted context ]
Signed-off-by: Luis Henriques <luis.henriques@canonical.com>
---
 arch/x86/kernel/cpu/perf_event_amd_ibs.c | 53 +++++++++++++++++++++++++++-----
 1 file changed, 45 insertions(+), 8 deletions(-)

diff --git a/arch/x86/kernel/cpu/perf_event_amd_ibs.c b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
index 4018f6e..6eea380 100644
--- a/arch/x86/kernel/cpu/perf_event_amd_ibs.c
+++ b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
@@ -10,6 +10,7 @@
 #include <linux/module.h>
 #include <linux/pci.h>
 #include <linux/ptrace.h>
+#include <linux/syscore_ops.h>
 
 #include <asm/apic.h>
 
@@ -777,6 +778,18 @@ out:
 	return ret;
 }
 
+static void ibs_eilvt_setup(void)
+{
+	/*
+	 * Force LVT offset assignment for family 10h: The offsets are
+	 * not assigned by the BIOS for this family, so the OS is
+	 * responsible for doing it. If the OS assignment fails, fall
+	 * back to BIOS settings and try to setup this.
+	 */
+	if (boot_cpu_data.x86 == 0x10)
+		force_ibs_eilvt_setup();
+}
+
 static inline int get_ibs_lvt_offset(void)
 {
 	u64 val;
@@ -812,6 +825,36 @@ static void clear_APIC_ibs(void *dummy)
 		setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_FIX, 1);
 }
 
+#ifdef CONFIG_PM
+
+static int perf_ibs_suspend(void)
+{
+	clear_APIC_ibs(NULL);
+	return 0;
+}
+
+static void perf_ibs_resume(void)
+{
+	ibs_eilvt_setup();
+	setup_APIC_ibs(NULL);
+}
+
+static struct syscore_ops perf_ibs_syscore_ops = {
+	.resume		= perf_ibs_resume,
+	.suspend	= perf_ibs_suspend,
+};
+
+static void perf_ibs_pm_init(void)
+{
+	register_syscore_ops(&perf_ibs_syscore_ops);
+}
+
+#else
+
+static inline void perf_ibs_pm_init(void) { }
+
+#endif
+
 static int __cpuinit
 perf_ibs_cpu_notifier(struct notifier_block *self, unsigned long action, void *hcpu)
 {
@@ -838,18 +881,12 @@ static __init int amd_ibs_init(void)
 	if (!caps)
 		return -ENODEV;	/* ibs not supported by the cpu */
 
-	/*
-	 * Force LVT offset assignment for family 10h: The offsets are
-	 * not assigned by the BIOS for this family, so the OS is
-	 * responsible for doing it. If the OS assignment fails, fall
-	 * back to BIOS settings and try to setup this.
-	 */
-	if (boot_cpu_data.x86 == 0x10)
-		force_ibs_eilvt_setup();
+	ibs_eilvt_setup();
 
 	if (!ibs_eilvt_valid())
 		goto out;
 
+	perf_ibs_pm_init();
 	get_online_cpus();
 	ibs_caps = caps;
 	/* make ibs_caps visible to other cpus: */
-- 
1.8.3.2


  parent reply	other threads:[~2014-02-07 11:24 UTC|newest]

Thread overview: 91+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-02-07 11:21 [3.5.y.z extended stable] Linux 3.5.7.30 stable review Luis Henriques
2014-02-07 11:21 ` [PATCH 3.5 01/88] ext4: fix deadlock when writing in ENOSPC conditions Luis Henriques
2014-02-07 11:21 ` [PATCH 3.5 02/88] writeback: Fix data corruption on NFS Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 03/88] md/raid5: Fix possible confusion when multiple write errors occur Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 04/88] md/raid10: fix two bugs in handling of known-bad-blocks Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 05/88] md/raid10: fix bug when raid10 recovery fails to recover a block Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 06/88] hwmon: (coretemp) Fix truncated name of alarm attributes Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 07/88] nilfs2: fix segctor bug that causes file system corruption Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 08/88] mm: fix crash when using XFS on loopback Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 09/88] vfs: In d_path don't call d_dname on a mount point Luis Henriques
2014-02-07 11:22 ` Luis Henriques [this message]
2014-02-07 11:22 ` [PATCH 3.5 11/88] staging: comedi: 8255_pci: fix for newer PCI-DIO48H Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 12/88] mm/memory-failure.c: recheck PageHuge() after hugetlb page migrate successfully Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 13/88] serial: amba-pl011: use port lock to guard control register access Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 14/88] rtlwifi: rtl8192cu: Fix W=1 build warning Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 15/88] rtlwifi: rtl8192cu: Add new firmware Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 16/88] rtlwifi: Set the link state Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 17/88] rtlwifi: rtl8192c: Add new definitions in the dm_common header Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 18/88] rtlwifi: rtl8192cu: Fix some code in RF handling Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 19/88] NFSv4: OPEN must handle the NFS4ERR_IO return code correctly Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 20/88] parport: parport_pc: remove double PCI ID for NetMos Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 21/88] staging: vt6656: [BUG] BBvUpdatePreEDThreshold Always set sensitivity on bScanning Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 22/88] [SCSI] bfa: Chinook quad port 16G FC HBA claim issue Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 23/88] usb: option: add new zte 3g modem pids to option driver Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 24/88] [media] dib8000: make 32 bits read atomic Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 25/88] serial: add support for 200 v3 series Titan card Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 26/88] usb: xhci: Check for XHCI_PLAT in xhci_cleanup_msix() Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 27/88] x86/efi: Fix off-by-one bug in EFI Boot Services reservation Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 28/88] perf kvm: Fix kvm report without guestmount Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 29/88] mtd: mxc_nand: remove duplicated ecc_stats counting Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 30/88] xen/pvhvm: If xen_platform_pci=0 is set don't blow up (v4) Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 31/88] USB: serial: add support for iBall 3.5G connect usb modem Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 32/88] USB: Nokia 502 is an unusual device Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 33/88] USB: cypress_m8: fix ring-indicator detection and reporting Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 34/88] ALSA: rme9652: fix a missing comma in channel_map_9636_ds[] Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 35/88] sunrpc: Fix infinite loop in RPC state machine Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 36/88] spidev: fix hang when transfer_one_message fails Luis Henriques
2014-02-07 18:17   ` Daniel Santos
2014-02-07 18:37     ` Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 37/88] dm thin: initialize dm_thin_new_mapping returned by get_next_mapping Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 38/88] SELinux: Fix memory leak upon loading policy Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 39/88] drm/radeon: warn users when hw_i2c is enabled (v2) Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 40/88] USB: ftdi_sio: added CS5 quirk for broken smartcard readers Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 41/88] serial: 8250: enable UART_BUG_NOMSR for Tegra Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 42/88] dm: wait until embedded kobject is released before destroying a device Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 43/88] dm space map common: make sure new space is used during extend Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 44/88] ASoC: adau1701: Fix ADAU1701_SEROCTL_WORD_LEN_16 constant Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 45/88] radeon/pm: Guard access to rdev->pm.power_state array Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 46/88] drm/radeon: skip colorbuffer checking if COLOR_INFO.FORMAT is set to INVALID Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 47/88] staging: r8712u: Set device type to wlan Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 48/88] ALSA: Enable CONFIG_ZONE_DMA for smaller PCI DMA masks Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 49/88] mmc: atmel-mci: fix timeout errors in SDIO mode when using DMA Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 50/88] rtlwifi: rtl8192cu: Add new device ID Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 51/88] mwifiex: add missing endian conversion for fw_tsf Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 52/88] b43: Fix lockdep splat Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 53/88] b43: Fix unload oops if firmware is not available Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 54/88] b43legacy: " Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 55/88] nfs4.1: properly handle ENOTSUP in SECINFO_NO_NAME Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 56/88] audit: correct a type mismatch in audit_syscall_exit() Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 57/88] md/raid5: fix long-standing problem with bitmap handling on write failure Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 58/88] drm/radeon: set the full cache bit for fences on r7xx+ Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 59/88] hp_accel: Add a new PnP ID HPQ6007 for new HP laptops Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 60/88] intel-iommu: fix off-by-one in pagetable freeing Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 61/88] fuse: fix pipe_buf_operations Luis Henriques
2014-02-07 11:22 ` [PATCH 3.5 62/88] drm/cirrus: correct register values for 16bpp Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 63/88] IB/qib: Fix QP check when looping back to/from QP1 Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 64/88] ore: Fix wrong math in allocation of per device BIO Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 65/88] b43: fix the wrong assignment of status.freq in b43_rx() Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 66/88] KVM: PPC: e500: Fix bad address type in deliver_tlb_misss() Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 67/88] Btrfs: handle EAGAIN case properly in btrfs_drop_snapshot() Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 68/88] ACPI / init: Flag use of ACPI and ACPI idioms for power supplies to regulator API Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 69/88] powerpc: Make sure "cache" directory is removed when offlining cpu Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 70/88] drm/radeon/DCE4+: clear bios scratch dpms bit (v2) Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 71/88] mm/page-writeback.c: fix dirty_balance_reserve subtraction from dirtyable memory Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 72/88] target/iscsi: Fix network portal creation race Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 73/88] mm, oom: base root bonus on current usage Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 74/88] x86, x32: Correct invalid use of user timespec in the kernel Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 75/88] alpha: fix broken network checksum Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 76/88] ARM: at91: smc: bug fix in sam9_smc_cs_read() Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 77/88] KVM: s390: fix diagnose code extraction Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 78/88] e752x_edac: Fix pci_dev usage count Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 79/88] lib/decompressors: fix "no limit" output buffer length Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 80/88] bnx2x: fix DMA unmapping of TSO split BDs Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 81/88] inet_diag: fix inet_diag_dump_icsk() timewait socket state logic Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 82/88] net: avoid reference counter overflows on fib_rules in multicast forwarding Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 83/88] net,via-rhine: Fix tx_timeout handling Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 84/88] mm: hugetlbfs: fix hugetlbfs optimization Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 85/88] usb: core: get config and string descriptors for unauthorized devices Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 86/88] tty/serial: at91: Handle shutdown more safely Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 87/88] slub: Fix calculation of cpu slabs Luis Henriques
2014-02-07 11:23 ` [PATCH 3.5 88/88] turbostat: Use GCC's CPUID functions to support PIC Luis Henriques

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=1391772205-22239-11-git-send-email-luis.henriques@canonical.com \
    --to=luis.henriques@canonical.com \
    --cc=kernel-team@lists.ubuntu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.org \
    --cc=rric@kernel.org \
    --cc=stable@vger.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®