linux-kernel - Re: [PATCH 2/2] iommu/vt-d: Share DMAR fault IRQ to prevent vector exhaustion

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: <20240408103844.1b567bce@jacob-builder>
Date: Mon, 8 Apr 2024 10:38:44 -0700
From: Jacob Pan <jacob.jun.pan@...ux.intel.com>
To: "Tian, Kevin" <kevin.tian@...el.com>
Cc: "sivanich@....com" <sivanich@....com>, Thomas Gleixner
 <tglx@...utronix.de>, LKML <linux-kernel@...r.kernel.org>,
 "iommu@...ts.linux.dev" <iommu@...ts.linux.dev>, Lu Baolu
 <baolu.lu@...ux.intel.com>, Joerg Roedel <joro@...tes.org>, "Liu, Yi L"
 <yi.l.liu@...el.com>, "steve.wahl@....com" <steve.wahl@....com>, "Anderson,
 Russ" <russ.anderson@....com>, "Peter Zijlstra" <peterz@...radead.org>,
 Will Deacon <will@...nel.org>, Robin Murphy <robin.murphy@....com>,
 jacob.jun.pan@...ux.intel.com
Subject: Re: [PATCH 2/2] iommu/vt-d: Share DMAR fault IRQ to prevent vector
 exhaustion

Hi Jacob,

On Mon, 8 Apr 2024 09:05:56 -0700, Jacob Pan
<jacob.jun.pan@...ux.intel.com> wrote:

> Hi Kevin,
> 
> On Mon, 8 Apr 2024 08:48:54 +0000, "Tian, Kevin" <kevin.tian@...el.com>
> wrote:
> 
> > > From: Jacob Pan <jacob.jun.pan@...ux.intel.com>
> > > Sent: Thursday, April 4, 2024 7:46 AM
> > > 
> > > DMAR fault interrupt is used for per-IOMMU unrecoverable fault
> > > reporting, it occurs only if there is a kernel programming error or
> > > serious hardware failure. In other words, they should never occur
> > > under normal circumstances.    
> > 
> > this is not accurate. When a device is assigned to a malicious guest
> > then it's not unusual to observe faults.
> >   
> Right, a malicious guest kernel could cause unrecoverable faults, e.g.
> wrong privilege.
> 
> > in this context you probably meant that it's not a performance path
> > hence sharing the vector is acceptable.
> >   
> Yes.
> > >
> > > @@ -1182,7 +1182,6 @@ static void free_iommu(struct intel_iommu
> > > *iommu)
> > >  			iommu->pr_irq = 0;
> > >  		}
> > >  		free_irq(iommu->fault_irq, iommu);
> > > -		dmar_free_hwirq(iommu->fault_irq);    
> > 
> > You still want to free the vector for the iommu which first gets the
> > vector allocated.
> >   
> I think we always want to keep this vector since the system always needs
> one vector to share. We will never offline all the IOMMUs, right?
> 
> > > @@ -1956,9 +1955,8 @@ void dmar_msi_mask(struct irq_data *data)
> > >  	raw_spin_unlock_irqrestore(&iommu->register_lock, flag);
> > >  }
> > > 
> > > -void dmar_msi_write(int irq, struct msi_msg *msg)
> > > +static void dmar_msi_write_msg(struct intel_iommu *iommu, int irq,
> > > struct msi_msg *msg)
> > >  {    
> > 
> > what about iommu_msi_write_msg() to match the first parameter?
> > 
> > otherwise it leads to a slightly circled calltrace:
> > 	dmar_msi_write_msg()
> > 		dmar_msi_write()
> > 			dmar_msi_write_msg()
> >   
> Good point, will do.
> 
> > > +
> > > +	/*
> > > +	 * Only the owner IOMMU of the shared IRQ has its fault event
> > > +	 * interrupt unmasked after request_irq(), the rest are
> > > explicitly
> > > +	 * unmasked.
> > > +	 */
> > > +	if (!(iommu->flags & VTD_FLAG_FAULT_IRQ_OWNER))
> > > +		dmar_fault_irq_unmask(iommu);
> > > +    
> > 
> > em there is a problem in dmar_msi_mask() and dmar_msi_mask()
> > which only touches the owner IOMMU. With this shared vector
> > approach we should mask/unmask all IOMMU's together.   
> I thought about this as well, in addition to fault_irq,
> dmar_msi_mask/unmask() are used for other DMAR irqs, page request and
> perfmon. So we need a special case for fault_irq there, it is not pretty.
> 
> I added a special case here in this patch, thinking we never mask the
> fault_irq since we need to cover the lifetime of the system. I have looked
> at:
> 1.IOMMU suspend/resume, no mask/unmask
Actually, we do call mask/unmask in suspend/unmask noirq phase.
And DMAR-MSI chip has IRQCHIP_SKIP_SET_WAKE flag.

So you are right, I am missing this case where non-owner IOMMU's fault_irqs
are not masked/unmasked.

> 2.IRQ migration, added IRQF_NOBALANCING
> 
> maybe I missed some cases?
> 
> 
> Thanks,
> 
> Jacob


Thanks,

Jacob