linux-kernel - Re: [PATCH v5 15/23] iommu: handle page response timeout

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180529092058.1942b223@jacob-builder>
Date:   Tue, 29 May 2018 09:20:58 -0700
From:   Jacob Pan <jacob.jun.pan@...ux.intel.com>
To:     Lu Baolu <baolu.lu@...ux.intel.com>
Cc:     iommu@...ts.linux-foundation.org,
        LKML <linux-kernel@...r.kernel.org>,
        Joerg Roedel <joro@...tes.org>,
        David Woodhouse <dwmw2@...radead.org>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Alex Williamson <alex.williamson@...hat.com>,
        Jean-Philippe Brucker <jean-philippe.brucker@....com>,
        Rafael Wysocki <rafael.j.wysocki@...el.com>,
        "Liu, Yi L" <yi.l.liu@...el.com>,
        "Tian, Kevin" <kevin.tian@...el.com>,
        Raj Ashok <ashok.raj@...el.com>,
        Jean Delvare <khali@...ux-fr.org>,
        Christoph Hellwig <hch@...radead.org>,
        jacob.jun.pan@...ux.intel.com
Subject: Re: [PATCH v5 15/23] iommu: handle page response timeout

On Mon, 14 May 2018 15:43:54 +0800
Lu Baolu <baolu.lu@...ux.intel.com> wrote:

> Hi,
> 
> On 05/12/2018 04:54 AM, Jacob Pan wrote:
> > When IO page faults are reported outside IOMMU subsystem, the page
> > request handler may fail for various reasons. E.g. a guest received
> > page requests but did not have a chance to run for a long time. The
> > irresponsive behavior could hold off limited resources on the
> > pending device.
> > There can be hardware or credit based software solutions as
> > suggested in the PCI ATS Ch-4. To provide a basic safty net this
> > patch introduces a per device deferrable timer which monitors the
> > longest pending page fault that requires a response. Proper action
> > such as sending failure response code could be taken when timer
> > expires but not included in this patch. We need to consider the
> > life cycle of page groupd ID to prevent confusion with reused group
> > ID by a device. For now, a warning message provides clue of such
> > failure.
> >
> > Signed-off-by: Jacob Pan <jacob.jun.pan@...ux.intel.com>
> > Signed-off-by: Ashok Raj <ashok.raj@...el.com>
> > ---
> >  drivers/iommu/iommu.c | 53
> > +++++++++++++++++++++++++++++++++++++++++++++++++++
> > include/linux/iommu.h |  4 ++++ 2 files changed, 57 insertions(+)
> >
> > diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
> > index 02fed3e..1f2f49e 100644
> > --- a/drivers/iommu/iommu.c
> > +++ b/drivers/iommu/iommu.c
> > @@ -827,6 +827,37 @@ int iommu_group_unregister_notifier(struct
> > iommu_group *group, }
> >  EXPORT_SYMBOL_GPL(iommu_group_unregister_notifier);
> >  
> > +static void iommu_dev_fault_timer_fn(struct timer_list *t)
> > +{
> > +	struct iommu_fault_param *fparam = from_timer(fparam, t,
> > timer);
> > +	struct iommu_fault_event *evt;
> > +
> > +	u64 now;
> > +
> > +	now = get_jiffies_64();
> > +
> > +	/* The goal is to ensure driver or guest page fault
> > handler(via vfio)
> > +	 * send page response on time. Otherwise, limited queue
> > resources
> > +	 * may be occupied by some irresponsive guests or drivers.
> > +	 * When per device pending fault list is not empty, we
> > periodically checks
> > +	 * if any anticipated page response time has expired.
> > +	 *
> > +	 * TODO:
> > +	 * We could do the following if response time expires:
> > +	 * 1. send page response code FAILURE to all pending PRQ
> > +	 * 2. inform device driver or vfio
> > +	 * 3. drain in-flight page requests and responses for this
> > device
> > +	 * 4. clear pending fault list such that driver can
> > unregister fault
> > +	 *    handler(otherwise blocked when pending faults are
> > present).
> > +	 */
> > +	list_for_each_entry(evt, &fparam->faults, list) {
> > +		if (time_after64(now, evt->expire))
> > +			pr_err("Page response time expired!, pasid
> > %d gid %d exp %llu now %llu\n",
> > +				evt->pasid,
> > evt->page_req_group_id, evt->expire, now);
> > +	}
> > +	mod_timer(t, now + prq_timeout);
> > +}
> > +  
> 
> This timer scheme is very rough.
> 
yes, the timer is a rough safety net for misbehaved PRQ handlers such
as a guest.
> The timer expires every 10 seconds (by default).
> 
> 0                   10                 20
> 30                 40
> +---------------+---------------+---------------+---------------+ ^
> ^   ^  ^                        ^ |   |     |
> |                         | F0 F1  F2 F3
> (F1,F2,F3 will not be handled until here!)
> 
> F0, F1, F2, F3 are four page faults happens during [0, 10s) time
> window. F1, F2, F3 timeout won't be handled until the timer expires
> again at 20s. That means a fault might be pending there until about
> (2 * prq_timeout) seconds later.
> 
correct. it could be 2x for the worst case. I should explain in
comments.
> Out of curiosity, Why not adding a timer in iommu_fault_event,
> starting it in iommu_report_device_fault() and removing it in
> iommu_page_response()?
> 
I thought about that also but since we are just trying to have a broad
and rough safety net (in addition to potential HW mechanism or credit
based solution), my thought was that having a per device timer is more
economical than per event.
Thanks for the in-depth check!

> Best regards,
> Lu Baolu
> 
> 
>  [...]  
> 

[Jacob Pan]