Re: [PATCH 3/3] vhost: apply cpumask and cgroup to vhost workers
From: "Michael S. Tsirkin" <mst@redhat.com>
Date: 2010-06-01 10:22:03
Also in:
kvm, lkml
On Tue, Jun 01, 2010 at 11:35:15AM +0200, Tejun Heo wrote:
Apply the cpumask and cgroup of the initializing task to the created vhost worker. Based on Sridhar Samudrala's patch. Li Zefan spotted a bug in error path (twice), fixed (twice). Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Michael S. Tsirkin <mst@redhat.com> Cc: Sridhar Samudrala <redacted> Cc: Li Zefan <redacted>
Something that I wanted to figure out - what happens if the CPU mask limits us to a certain CPU that subsequently goes offline? Will e.g. flush block forever or until that CPU comes back? Also, does singlethreaded workqueue behave in the same way?
quoted hunk ↗ jump to hunk
--- drivers/vhost/vhost.c | 34 ++++++++++++++++++++++++++++++---- 1 file changed, 30 insertions(+), 4 deletions(-) Index: work/drivers/vhost/vhost.c ===================================================================--- work.orig/drivers/vhost/vhost.c +++ work/drivers/vhost/vhost.c@@ -23,6 +23,7 @@ #include <linux/highmem.h> #include <linux/slab.h> #include <linux/kthread.h> +#include <linux/cgroup.h> #include <linux/net.h> #include <linux/if_packet.h>@@ -187,11 +188,29 @@ long vhost_dev_init(struct vhost_dev *de struct vhost_virtqueue *vqs, int nvqs) { struct task_struct *worker; - int i; + cpumask_var_t mask; + int i, ret = -ENOMEM; + + if (!alloc_cpumask_var(&mask, GFP_KERNEL)) + goto out_free_mask; worker = kthread_create(vhost_worker, dev, "vhost-%d", current->pid); - if (IS_ERR(worker)) - return PTR_ERR(worker); + if (IS_ERR(worker)) { + ret = PTR_ERR(worker); + goto out_free_mask; + } + + ret = sched_getaffinity(current->pid, mask); + if (ret) + goto out_stop_worker; + + ret = sched_setaffinity(worker->pid, mask); + if (ret) + goto out_stop_worker; + + ret = cgroup_attach_task_current_cg(worker); + if (ret) + goto out_stop_worker; dev->vqs = vqs; dev->nvqs = nvqs;@@ -214,7 +233,14 @@ long vhost_dev_init(struct vhost_dev *de } wake_up_process(worker); /* avoid contributing to loadavg */ - return 0; + ret = 0; + goto out_free_mask; + +out_stop_worker: + kthread_stop(worker); +out_free_mask: + free_cpumask_var(mask); + return ret; } /* Caller should have device mutex */