Re: [PATCH] virtio: a new vcpu watchdog driver

From: Michael S. Tsirkin
Date: Thu Aug 10 2023 - 16:56:01 EST

Next message: Ian Rogers: "Re: [PATCH] perf tools: Handle old data in PERF_RECORD_ATTR"
Previous message: Raghavendra Rao Ananta: "Re: [PATCH v8 02/14] KVM: Declare kvm_arch_flush_remote_tlbs() globally"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Mon, Jul 31, 2023 at 09:25:12AM +0800, zhanghao1 wrote:
> A new virtio pci driver is added for listening to vcpus
> inside guest. Each vcpu creates a corresponding thread to
> periodically send data to qemu's back-end watchdog device.
> If a vCPU is in the stall state, data cannot be sent to
> back-end virtio device. As a result, the back-end device
> can detect that the guest is in the stall state.
>
> The driver is mainly used with the back-end watchdog device of qemu.
>
> The qemu backend watchdog device is implemented as follow:
> https://lore.kernel.org/qemu-devel/20230705081813.411526-1-zhanghao1@xxxxxxxxxx/
>
> Signed-off-by: zhanghao1 <zhanghao1@xxxxxxxxxx>
> ---
> drivers/virtio/Kconfig | 9 +
> drivers/virtio/Makefile | 1 +
> drivers/virtio/virtio_vcpu_stall_detector.c | 299 ++++++++++++++++++++
> 3 files changed, 309 insertions(+)
> create mode 100644 drivers/virtio/virtio_vcpu_stall_detector.c
>
> diff --git a/drivers/virtio/Kconfig b/drivers/virtio/Kconfig
> index 0a53a61231c2..869323e345a1 100644
> --- a/drivers/virtio/Kconfig
> +++ b/drivers/virtio/Kconfig
> @@ -173,4 +173,13 @@ config VIRTIO_DMA_SHARED_BUFFER
> This option adds a flavor of dma buffers that are backed by
> virtio resources.
>
> +config VIRTIO_VCPU_WATCHDOG
> + tristate "Virtio vcpu watchdog driver"
> + depends on VIRTIO_PCI
> + help
> + When this driver is bound inside a KVM guest, it will
> + periodically "pet" an PCI virtio watchdog device from each vCPU
> + and allow the host to detect vCPU stalls.
> +
> + If you do not intend to run this kernel as a guest, say N.
> endif # VIRTIO_MENU
> diff --git a/drivers/virtio/Makefile b/drivers/virtio/Makefile
> index 8e98d24917cc..c7341f078a34 100644
> --- a/drivers/virtio/Makefile
> +++ b/drivers/virtio/Makefile
> @@ -12,3 +12,4 @@ obj-$(CONFIG_VIRTIO_INPUT) += virtio_input.o
> obj-$(CONFIG_VIRTIO_VDPA) += virtio_vdpa.o
> obj-$(CONFIG_VIRTIO_MEM) += virtio_mem.o
> obj-$(CONFIG_VIRTIO_DMA_SHARED_BUFFER) += virtio_dma_buf.o
> +obj-$(CONFIG_VIRTIO_VCPU_WATCHDOG) += virtio_vcpu_stall_detector.o
> diff --git a/drivers/virtio/virtio_vcpu_stall_detector.c b/drivers/virtio/virtio_vcpu_stall_detector.c
> new file mode 100644
> index 000000000000..58344ca528be
> --- /dev/null
> +++ b/drivers/virtio/virtio_vcpu_stall_detector.c
> @@ -0,0 +1,299 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +//
> +// VCPU stall detector.
> +// Copyright (C) Kylin Software, 2023
> +
> +#include <linux/cpu.h>
> +#include <linux/init.h>
> +#include <linux/io.h>
> +#include <linux/kernel.h>
> +
> +#include <linux/device.h>
> +#include <linux/interrupt.h>
> +#include <linux/module.h>
> +#include <linux/nmi.h>
> +#include <uapi/linux/virtio_ids.h>
> +#include <linux/virtio_config.h>
> +#include <linux/param.h>
> +#include <linux/percpu.h>
> +#include <linux/slab.h>
> +
> +#define VCPU_STALL_REG_STATUS (0x00)
> +#define VCPU_STALL_REG_LOAD_CNT (0x04)
> +#define VCPU_STALL_REG_CURRENT_CNT (0x08)
> +#define VCPU_STALL_REG_CLOCK_FREQ_HZ (0x0C)
> +#define VCPU_STALL_REG_LEN (0x10)
> +#define VCPU_STALL_REG_TIMEOUT_SEC (0x14)
> +
> +#define VCPU_STALL_DEFAULT_CLOCK_HZ (10)
> +#define VCPU_STALL_MAX_CLOCK_HZ (100)
> +#define VCPU_STALL_DEFAULT_TIMEOUT_SEC (8)
> +#define VCPU_STALL_MAX_TIMEOUT_SEC (600)
> +
> +struct vcpu_stall_detect_config {
> + u32 clock_freq_hz;
> + u32 stall_timeout_sec;
> +
> + enum cpuhp_state hp_online;
> +};
> +
> +struct vcpu_stall_priv {
> + struct hrtimer vcpu_hrtimer;
> + struct virtio_device *vdev;
> + u32 cpu_id;
> +};
> +
> +struct vcpu_stall {
> + struct vcpu_stall_priv *priv;
> + struct virtqueue *vq;
> + spinlock_t lock;
> + struct pet_event {
> + u32 cpu_id;
> + bool is_initialized;
> + u32 ticks;
> + } pet_event;

should all be LE. and formatting here is very compiler dependent.
also put this in a header under uapi/

> +};
> +
> +static const struct virtio_device_id vcpu_stall_id_table[] = {
> + { VIRTIO_ID_WATCHDOG, VIRTIO_DEV_ANY_ID },
> + { 0, },
> +};
> +
> +/* The vcpu stall configuration structure which applies to all the CPUs */
> +static struct vcpu_stall_detect_config vcpu_stall_config;
> +static struct vcpu_stall *vcpu_stall;
> +
> +static struct vcpu_stall_priv __percpu *vcpu_stall_detectors;
> +
> +static enum hrtimer_restart
> +vcpu_stall_detect_timer_fn(struct hrtimer *hrtimer)
> +{
> + u32 ticks, ping_timeout_ms;
> + struct scatterlist sg;
> + int unused, err = 0;
> +
> + struct vcpu_stall_priv *vcpu_stall_detector =
> + this_cpu_ptr(vcpu_stall->priv);
> +
> + /* Reload the stall detector counter register every
> + * `ping_timeout_ms` to prevent the virtual device
> + * from decrementing it to 0. The virtual device decrements this
> + * register at 'clock_freq_hz' frequency.
> + */
> + ticks = vcpu_stall_config.clock_freq_hz *
> + vcpu_stall_config.stall_timeout_sec;
> +
> + spin_lock(&vcpu_stall->lock);
> + while (virtqueue_get_buf(vcpu_stall->vq, &unused))
> + ;
> + vcpu_stall->pet_event.ticks = cpu_to_virtio32(vcpu_stall_detector->vdev, ticks);
> + vcpu_stall->pet_event.is_initialized = true;
> + vcpu_stall->pet_event.cpu_id = vcpu_stall_detector->cpu_id;
> +
> + sg_init_one(&sg, &vcpu_stall->pet_event, sizeof(vcpu_stall->pet_event));
> + err = virtqueue_add_outbuf(vcpu_stall->vq, &sg, 1, vcpu_stall, GFP_ATOMIC);
> + if (!err)
> + virtqueue_kick(vcpu_stall->vq);
> + else
> + pr_err("cpu:%d failed to add outbuf, err:%d\n", vcpu_stall_detector->cpu_id, err);

this can happen if device is slow to use the buffers. and then what?
will missing the pet even mean host will reset the guest?
pr_err also might make guest crash, depending on config.

> +
> + spin_unlock(&vcpu_stall->lock);
> +
> + ping_timeout_ms = vcpu_stall_config.stall_timeout_sec *
> + MSEC_PER_SEC / 2;
> + hrtimer_forward_now(hrtimer,
> + ms_to_ktime(ping_timeout_ms));
> + return HRTIMER_RESTART;
> +}
> +
> +static int start_stall_detector_cpu(unsigned int cpu)
> +{
> + u32 ticks, ping_timeout_ms;
> + struct scatterlist sg;
> + struct hrtimer *vcpu_hrtimer;
> + int err = 0;
> +
> + struct vcpu_stall_priv *vcpu_stall_detector =
> + this_cpu_ptr(vcpu_stall->priv);
> +
> + vcpu_stall_detector->cpu_id = cpu;
> +
> + vcpu_hrtimer = &vcpu_stall_detector->vcpu_hrtimer;
> +
> + /* Compute the number of ticks required for the stall detector
> + * counter register based on the internal clock frequency and the
> + * timeout value given from the device tree.
> + */
> + ticks = vcpu_stall_config.clock_freq_hz *
> + vcpu_stall_config.stall_timeout_sec;
> + vcpu_stall->pet_event.ticks = cpu_to_virtio32(vcpu_stall_detector->vdev, ticks);
> +
> + /* Pet the stall detector at half of its expiration timeout
> + * to prevent spurious resets.
> + */
> + ping_timeout_ms = vcpu_stall_config.stall_timeout_sec *
> + MSEC_PER_SEC / 2;
> +
> + hrtimer_init(vcpu_hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
> + vcpu_hrtimer->function = vcpu_stall_detect_timer_fn;
> +
> + vcpu_stall->pet_event.is_initialized = true;
> +
> + spin_lock(&vcpu_stall->lock);
> + vcpu_stall->pet_event.cpu_id = cpu;
> + sg_init_one(&sg, &vcpu_stall->pet_event, sizeof(vcpu_stall->pet_event));
> + err = virtqueue_add_outbuf(vcpu_stall->vq, &sg, 1, vcpu_stall, GFP_ATOMIC);
> + if (!err)
> + virtqueue_kick(vcpu_stall->vq);
> +
> + spin_unlock(&vcpu_stall->lock);
> +
> + hrtimer_start(vcpu_hrtimer, ms_to_ktime(ping_timeout_ms),
> + HRTIMER_MODE_REL_PINNED);
> + return err;
> +}
> +
> +static int stop_stall_detector_cpu(unsigned int cpu)
> +{
> + int err = 0;
> + struct scatterlist sg;
> +
> + struct vcpu_stall_priv *vcpu_stall_detector =
> + per_cpu_ptr(vcpu_stall_detectors, cpu);
> +
> + /* Disable the stall detector for the current CPU */
> + hrtimer_cancel(&vcpu_stall_detector->vcpu_hrtimer);
> + vcpu_stall->pet_event.is_initialized = false;
> + vcpu_stall->pet_event.cpu_id = cpu;
> +
> + spin_lock(&vcpu_stall->lock);
> + sg_init_one(&sg, &vcpu_stall->pet_event, sizeof(vcpu_stall->pet_event));
> + err = virtqueue_add_outbuf(vcpu_stall->vq, &sg, 1, vcpu_stall, GFP_ATOMIC);
> + if (!err)
> + virtqueue_kick(vcpu_stall->vq);
> +
> + spin_unlock(&vcpu_stall->lock);
> +
> + return err;
> +}
> +
> +static int vcpu_stall_detect_probe(struct virtio_device *vdev)
> +{
> + int ret, cpu;
> + u32 clock_freq_hz = VCPU_STALL_DEFAULT_CLOCK_HZ;
> + u32 stall_timeout_sec = VCPU_STALL_DEFAULT_TIMEOUT_SEC;
> +
> + vcpu_stall = kzalloc(sizeof(struct vcpu_stall), GFP_KERNEL);
> + if (!vcpu_stall) {
> + ret = -ENOMEM;
> + goto err;
> + }
> + vdev->priv = vcpu_stall;
> +
> + vcpu_stall->priv = devm_alloc_percpu(&vdev->dev,
> + typeof(struct vcpu_stall_priv));
> + if (!vcpu_stall->priv) {
> + ret = -ENOMEM;
> + goto failed_priv;
> + }
> +
> + for_each_possible_cpu(cpu) {
> + struct vcpu_stall_priv *priv;
> +
> + priv = per_cpu_ptr(vcpu_stall->priv, cpu);
> + priv->vdev = vdev;
> + }
> +
> + ret = virtio_cread_feature(vdev, VCPU_STALL_REG_CLOCK_FREQ_HZ,
> + struct vcpu_stall_detect_config, clock_freq_hz,
> + &clock_freq_hz);
> + if (ret || !clock_freq_hz) {
> + if (!(clock_freq_hz > 0 &&
> + clock_freq_hz < VCPU_STALL_MAX_CLOCK_HZ)) {
> + dev_warn(&vdev->dev, "clk out of range\n");
> + clock_freq_hz = VCPU_STALL_DEFAULT_CLOCK_HZ;
> + }
> + }
> + ret = virtio_cread_feature(vdev, VCPU_STALL_REG_TIMEOUT_SEC,
> + struct vcpu_stall_detect_config, stall_timeout_sec,
> + &stall_timeout_sec);
> + if (ret || !stall_timeout_sec) {
> + if (!(stall_timeout_sec > 0 &&
> + stall_timeout_sec < VCPU_STALL_MAX_TIMEOUT_SEC)) {
> + dev_warn(&vdev->dev, "stall timeout out of range\n");
> + stall_timeout_sec = VCPU_STALL_DEFAULT_TIMEOUT_SEC;

how does host know what is the range?

> + }
> + }
> +
> + vcpu_stall_config = (struct vcpu_stall_detect_config) {
> + .clock_freq_hz = clock_freq_hz,
> + .stall_timeout_sec = stall_timeout_sec
> + };
> +
> + /* find virtqueue for guest to send pet event to host */
> + vcpu_stall->vq = virtio_find_single_vq(vdev, NULL, "pet-event");
> + if (IS_ERR(vcpu_stall->vq)) {
> + dev_err(&vdev->dev, "failed to find vq\n");
> + goto failed_priv;
> + }
> +
> + spin_lock_init(&vcpu_stall->lock);
> +
> + ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN,
> + "virt/vcpu_stall_detector:online",
> + start_stall_detector_cpu,
> + stop_stall_detector_cpu);
> + if (ret < 0) {
> + dev_err(&vdev->dev, "failed to install cpu hotplug\n");
> + goto failed_priv;
> + }
> +
> + vcpu_stall_config.hp_online = ret;
> + return 0;
> +
> +
> +failed_priv:
> + kfree(vcpu_stall);
> +err:
> + return ret;
> +}
> +
> +static void vcpu_stall_detect_remove(struct virtio_device *vdev)
> +{
> + int cpu;
> +
> + cpuhp_remove_state(vcpu_stall_config.hp_online);
> +
> + for_each_possible_cpu(cpu)
> + stop_stall_detector_cpu(cpu);

you are sending a bunch of messages ok but never even
wait for them to be processed. normally there's
need to reset device, destroy vqs ... not necessary here? why?

> +}
> +
> +static unsigned int features_legacy[] = {
> + VCPU_STALL_REG_STATUS, VCPU_STALL_REG_LOAD_CNT, VCPU_STALL_REG_CURRENT_CNT,
> + VCPU_STALL_REG_CLOCK_FREQ_HZ, VCPU_STALL_REG_LEN, VCPU_STALL_REG_TIMEOUT_SEC
> +};
> +

this should be a modern only device, no way this needs legacy support.

> +
> +static unsigned int features[] = {
> + VCPU_STALL_REG_STATUS, VCPU_STALL_REG_LOAD_CNT, VCPU_STALL_REG_CURRENT_CNT,
> + VCPU_STALL_REG_CLOCK_FREQ_HZ, VCPU_STALL_REG_LEN, VCPU_STALL_REG_TIMEOUT_SEC
> +};
> +
> +static struct virtio_driver vcpu_stall_detect_driver = {
> + .feature_table = features,
> + .feature_table_size = ARRAY_SIZE(features),
> + .feature_table_legacy = features_legacy,
> + .feature_table_size_legacy = ARRAY_SIZE(features_legacy),
> + .driver.name = KBUILD_MODNAME,
> + .driver.owner = THIS_MODULE,
> + .id_table = vcpu_stall_id_table,
> + .probe = vcpu_stall_detect_probe,
> + .remove = vcpu_stall_detect_remove,
> +};
> +
> +module_virtio_driver(vcpu_stall_detect_driver);
> +
> +MODULE_LICENSE("GPL");
> +MODULE_DEVICE_TABLE(virtio, vcpu_stall_id_table);
> +MODULE_AUTHOR("zhanghao1 <zhanghao1@xxxxxxxxxx>");
> +MODULE_DESCRIPTION("VCPU stall detector");
> --
> 2.25.1
>
>
> No virus found
> Checked by Hillstone Network AntiVirus

Next message: Ian Rogers: "Re: [PATCH] perf tools: Handle old data in PERF_RECORD_ATTR"
Previous message: Raghavendra Rao Ananta: "Re: [PATCH v8 02/14] KVM: Declare kvm_arch_flush_remote_tlbs() globally"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]