From: Michael Ellerman <mpe@ellerman.id.au>
To: alexander.levin@verizon.com,
"linux-kernel\@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"stable\@vger.kernel.org" <stable@vger.kernel.org>
Cc: Nicholas Piggin <npiggin@gmail.com>, alexander.levin@verizon.com
Subject: Re: [PATCH AUTOSEL for 3.18 36/40] powerpc/xmon: Avoid tripping SMP hardlockup watchdog
Date: Thu, 14 Dec 2017 00:10:39 +1100 [thread overview]
Message-ID: <87lgi6ai40.fsf@concordia.ellerman.id.au> (raw)
In-Reply-To: <20171213015905.7059-36-alexander.levin@verizon.com>
alexander.levin@verizon.com writes:
> From: Nicholas Piggin <npiggin@gmail.com>
>
> [ Upstream commit 064996d62a33ffe10264b5af5dca92d54f60f806 ]
>
> The SMP hardlockup watchdog cross-checks other CPUs for lockups, which
> causes xmon headaches because it's assuming interrupts hard disabled
> means no watchdog troubles. Try to improve that by calling
> touch_nmi_watchdog() in obvious places where secondaries are spinning.
>
> Also annotate these spin loops with spin_begin/end calls.
These macros didn't exist until 4.13, and haven't been backported AFAIK.
cheers
> diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c
> index e23f559faa47..6cec1081cb67 100644
> --- a/arch/powerpc/xmon/xmon.c
> +++ b/arch/powerpc/xmon/xmon.c
> @@ -436,14 +436,19 @@ static int xmon_core(struct pt_regs *regs, int fromipi)
>
> waiting:
> secondary = 1;
> + spin_begin();
> while (secondary && !xmon_gate) {
> if (in_xmon == 0) {
> - if (fromipi)
> + if (fromipi) {
> + spin_end();
> goto leave;
> + }
> secondary = test_and_set_bit(0, &in_xmon);
> }
> - barrier();
> + spin_cpu_relax();
> + touch_nmi_watchdog();
> }
> + spin_end();
>
> if (!secondary && !xmon_gate) {
> /* we are the first cpu to come in */
> @@ -470,21 +475,25 @@ static int xmon_core(struct pt_regs *regs, int fromipi)
> mb();
> xmon_gate = 1;
> barrier();
> + touch_nmi_watchdog();
> }
>
> cmdloop:
> while (in_xmon) {
> if (secondary) {
> + spin_begin();
> if (cpu == xmon_owner) {
> if (!test_and_set_bit(0, &xmon_taken)) {
> secondary = 0;
> + spin_end();
> continue;
> }
> /* missed it */
> while (cpu == xmon_owner)
> - barrier();
> + spin_cpu_relax();
> }
> - barrier();
> + spin_cpu_relax();
> + touch_nmi_watchdog();
> } else {
> cmd = cmds(regs);
> if (cmd != 0) {
> --
> 2.11.0
next prev parent reply other threads:[~2017-12-13 13:10 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-12-13 1:59 [PATCH AUTOSEL for 3.18 01/40] ALSA: hda - add support for docking station for HP 820 G2 alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 03/40] r8152: fix the list rx_done may be used without initialization alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 02/40] cpuidle: Validate cpu_dev in cpuidle_add_sysfs() alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 05/40] net: qmi_wwan: Add USB IDs for MDM6600 modem on Motorola Droid 4 alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 04/40] crypto: deadlock between crypto_alg_sem/rtnl_mutex/genl_mutex alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 07/40] usb: gadget: udc: remove pointer dereference after free alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 06/40] usb: gadget: f_uvc: Sanity check wMaxPacketSize for SuperSpeed alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 11/40] KVM: x86: correct async page present tracepoint alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 08/40] netfilter: nfnl_cthelper: fix runtime expectation policy updates alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 09/40] netfilter: nfnl_cthelper: Fix memory leak alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 10/40] scsi: lpfc: Fix PT2PT PRLI reject alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 12/40] ARM: dts: ti: fix PCI bus dtc warnings alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 13/40] hwmon: (asus_atk0110) fix uninitialized data access alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 14/40] i2c: mux: pca954x: Add missing pca9546 definition to chip_desc alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 15/40] HID: xinmo: fix for out of range for THT 2P arcade controller alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 19/40] irda: vlsi_ir: fix check for DMA mapping errors alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 16/40] s390/qeth: no ETH header for outbound AF_IUCV alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 17/40] net: Do not allow negative values for busy_read and busy_poll sysctl interfaces alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 18/40] i40e: Do not enable NAPI on q_vectors that have no rings alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 21/40] netfilter: nf_nat_snmp: Fix panic when snmp_trap_helper fails to register alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 23/40] isdn: kcapi: avoid uninitialized data alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 20/40] netfilter: nfnl_cthelper: fix a race when walk the nf_ct_helper_hash table alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 22/40] ARM: dts: am335x-evmsk: adjust mmc2 param to allow suspend alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 25/40] ARM: dma-mapping: disallow dma_get_sgtable() for non-kernel managed memory alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 26/40] cpuidle: powernv: Pass correct drv->cpumask for registration alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 27/40] backlight: pwm_bl: Fix overflow condition alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 24/40] xhci: plat: Register shutdown for xhci_plat alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 29/40] net: phy: at803x: Change error to EINVAL for invalid MAC alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 30/40] PCI: Avoid bus reset if bridge itself is broken alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 31/40] scsi: cxgb4i: fix Tx skb leak alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 28/40] crypto: crypto4xx - increase context and scatter ring buffer elements alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 35/40] percpu: don't forget to free the temporary struct pcpu_alloc_info alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 32/40] PCI: Create SR-IOV virtfn/physfn links before attaching driver alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 34/40] PCI/AER: Report non-fatal errors only to the affected endpoint alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 33/40] igb: check memory allocation failure alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 36/40] powerpc/xmon: Avoid tripping SMP hardlockup watchdog alexander.levin
2017-12-13 13:10 ` Michael Ellerman [this message]
2017-12-13 13:44 ` alexander.levin
2018-01-30 4:35 ` Michael Ellerman
2018-02-03 5:58 ` Nicholas Piggin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 39/40] fm10k: ensure we process SM mbx when processing VF mbx alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 38/40] scsi: lpfc: PLOGI failures during NPIV testing alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 37/40] scsi: lpfc: Fix secure firmware updates alexander.levin
2017-12-13 1:59 ` [PATCH AUTOSEL for 3.18 40/40] tcp: fix under-evaluated ssthresh in TCP Vegas alexander.levin
2017-12-21 9:54 ` 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=87lgi6ai40.fsf@concordia.ellerman.id.au \
--to=mpe@ellerman.id.au \
--cc=alexander.levin@verizon.com \
--cc=linux-kernel@vger.kernel.org \
--cc=npiggin@gmail.com \
--cc=stable@vger.kernel.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®