Nick Piggin wrote:
> +static void free_unmap_vmap_area(struct vmap_area *va)
> +{
> + va->flags |= VM_LAZY_FREE;
> + atomic_add((va->va_end - va->va_start) >> PAGE_SHIFT, &vmap_lazy_nr);
> + if (unlikely(atomic_read(&vmap_lazy_nr) > LAZY_MAX))
> + purge_vmap_area_lazy();
> +}
Could you check here if we are in an atomic context and then simply not purge
the vmap area? That may we may get to a vfree that can be run in an atomic
context.
Or run purge_vma_area_lazy from keventd?
--
To unsubscribe from this list: send the line "unsubscribe linux-arch" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
