QEMU-Devel Archive on lore.kernel.org
 help / color / Atom feed
From: "Liu, Yi L" <yi.l.liu@intel.com>
To: Peter Xu <peterx@redhat.com>
Cc: "jean-philippe@linaro.org" <jean-philippe@linaro.org>,
	"Tian, Kevin" <kevin.tian@intel.com>,
	Jacob Pan <jacob.jun.pan@linux.intel.com>,
	Yi Sun <yi.y.sun@linux.intel.com>,
	Eduardo Habkost <ehabkost@redhat.com>,
	"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
	"mst@redhat.com" <mst@redhat.com>,
	"Tian, Jun J" <jun.j.tian@intel.com>,
	"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
	"eric.auger@redhat.com" <eric.auger@redhat.com>,
	"alex.williamson@redhat.com" <alex.williamson@redhat.com>,
	"pbonzini@redhat.com" <pbonzini@redhat.com>,
	"Wu, Hao" <hao.wu@intel.com>, "Sun, Yi Y" <yi.y.sun@intel.com>,
	Richard Henderson <rth@twiddle.net>,
	"david@gibson.dropbear.id.au" <david@gibson.dropbear.id.au>
Subject: RE: [PATCH v1 07/22] intel_iommu: add set/unset_iommu_context callback
Date: Tue, 24 Mar 2020 11:15:24 +0000
Message-ID: <A2975661238FB949B60364EF0F2C25743A20041A@SHSMSX104.ccr.corp.intel.com> (raw)
In-Reply-To: <20200323212911.GQ127076@xz-x1>

> From: Peter Xu <peterx@redhat.com>
> Sent: Tuesday, March 24, 2020 5:29 AM
> To: Liu, Yi L <yi.l.liu@intel.com>
> Subject: Re: [PATCH v1 07/22] intel_iommu: add set/unset_iommu_context
> callback
> 
> On Sun, Mar 22, 2020 at 05:36:04AM -0700, Liu Yi L wrote:
> > This patch adds set/unset_iommu_context() impelementation in Intel
> > vIOMMU. For Intel platform, pass-through modules (e.g. VFIO) could
> > set HostIOMMUContext to Intel vIOMMU emulator.
> >
> > Cc: Kevin Tian <kevin.tian@intel.com>
> > Cc: Jacob Pan <jacob.jun.pan@linux.intel.com>
> > Cc: Peter Xu <peterx@redhat.com>
> > Cc: Yi Sun <yi.y.sun@linux.intel.com>
> > Cc: Paolo Bonzini <pbonzini@redhat.com>
> > Cc: Richard Henderson <rth@twiddle.net>
> > Cc: Eduardo Habkost <ehabkost@redhat.com>
> > Signed-off-by: Liu Yi L <yi.l.liu@intel.com>
> > ---
> >  hw/i386/intel_iommu.c         | 70
> +++++++++++++++++++++++++++++++++++++++----
> >  include/hw/i386/intel_iommu.h | 17 +++++++++--
> >  2 files changed, 80 insertions(+), 7 deletions(-)
> >
> > diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c
> > index 4b22910..8d9204f 100644
> > --- a/hw/i386/intel_iommu.c
> > +++ b/hw/i386/intel_iommu.c
> > @@ -3354,23 +3354,35 @@ static const MemoryRegionOps vtd_mem_ir_ops =
> {
> >      },
> >  };
> >
> > -VTDAddressSpace *vtd_find_add_as(IntelIOMMUState *s, PCIBus *bus, int
> devfn)
> > +/**
> > + * Fetch a VTDBus instance for given PCIBus. If no existing instance,
> > + * allocate one.
> > + */
> > +static VTDBus *vtd_find_add_bus(IntelIOMMUState *s, PCIBus *bus)
> >  {
> >      uintptr_t key = (uintptr_t)bus;
> >      VTDBus *vtd_bus = g_hash_table_lookup(s->vtd_as_by_busptr, &key);
> > -    VTDAddressSpace *vtd_dev_as;
> > -    char name[128];
> >
> >      if (!vtd_bus) {
> >          uintptr_t *new_key = g_malloc(sizeof(*new_key));
> >          *new_key = (uintptr_t)bus;
> >          /* No corresponding free() */
> > -        vtd_bus = g_malloc0(sizeof(VTDBus) + sizeof(VTDAddressSpace *) * \
> > -                            PCI_DEVFN_MAX);
> > +        vtd_bus = g_malloc0(sizeof(VTDBus) + PCI_DEVFN_MAX * \
> > +                            (sizeof(VTDAddressSpace *) + \
> > +                             sizeof(VTDHostIOMMUContext *)));
> 
> IIRC I commented on this before...  Shouldn't sizeof(VTDBus) be
> enough?

Right. My bad. Will do it in next version.

> >          vtd_bus->bus = bus;
> >          g_hash_table_insert(s->vtd_as_by_busptr, new_key, vtd_bus);
> >      }
> > +    return vtd_bus;
> > +}
> > +
> > +VTDAddressSpace *vtd_find_add_as(IntelIOMMUState *s, PCIBus *bus, int
> devfn)
> > +{
> > +    VTDBus *vtd_bus;
> > +    VTDAddressSpace *vtd_dev_as;
> > +    char name[128];
> >
> > +    vtd_bus = vtd_find_add_bus(s, bus);
> >      vtd_dev_as = vtd_bus->dev_as[devfn];
> >
> >      if (!vtd_dev_as) {
> > @@ -3436,6 +3448,52 @@ VTDAddressSpace
> *vtd_find_add_as(IntelIOMMUState *s, PCIBus *bus, int devfn)
> >      return vtd_dev_as;
> >  }
> >
> > +static int vtd_dev_set_iommu_context(PCIBus *bus, void *opaque,
> > +                                     int devfn,
> > +                                     HostIOMMUContext *host_icx)
> > +{
> > +    IntelIOMMUState *s = opaque;
> > +    VTDBus *vtd_bus;
> > +    VTDHostIOMMUContext *vtd_dev_icx;
> > +
> > +    assert(0 <= devfn && devfn < PCI_DEVFN_MAX);
> > +
> > +    vtd_bus = vtd_find_add_bus(s, bus);
> > +
> > +    vtd_iommu_lock(s);
> > +    vtd_dev_icx = vtd_bus->dev_icx[devfn];
> > +
> > +    if (!vtd_dev_icx) {
> 
> We can assert this directly I think, in case we accidentally set the
> context twice without notice.

good idea. will add it.

> 
> > +        vtd_bus->dev_icx[devfn] = vtd_dev_icx =
> > +                    g_malloc0(sizeof(VTDHostIOMMUContext));
> > +        vtd_dev_icx->vtd_bus = vtd_bus;
> > +        vtd_dev_icx->devfn = (uint8_t)devfn;
> > +        vtd_dev_icx->iommu_state = s;
> > +        vtd_dev_icx->host_icx = host_icx;
> > +    }
> > +    vtd_iommu_unlock(s);
> > +
> > +    return 0;
> > +}
> > +
> > +static void vtd_dev_unset_iommu_context(PCIBus *bus, void *opaque, int
> devfn)
> > +{
> > +    IntelIOMMUState *s = opaque;
> > +    VTDBus *vtd_bus;
> > +    VTDHostIOMMUContext *vtd_dev_icx;
> > +
> > +    assert(0 <= devfn && devfn < PCI_DEVFN_MAX);
> > +
> > +    vtd_bus = vtd_find_add_bus(s, bus);
> > +
> > +    vtd_iommu_lock(s);
> > +
> > +    vtd_dev_icx = vtd_bus->dev_icx[devfn];
> > +    g_free(vtd_dev_icx);
> 
> Better set it as NULL, and can also drop vtd_dev_icx which seems
> meaningless..

right. it's missed.

>        g_free(vtd_bus->dev_icx[devfn]);
>        vtd_bus->dev_icx[devfn] = NULL;
> 
> > +
> > +    vtd_iommu_unlock(s);
> > +}
> > +
> >  static uint64_t get_naturally_aligned_size(uint64_t start,
> >                                             uint64_t size, int gaw)
> >  {
> > @@ -3731,6 +3789,8 @@ static AddressSpace *vtd_host_dma_iommu(PCIBus
> *bus, void *opaque, int devfn)
> >
> >  static PCIIOMMUOps vtd_iommu_ops = {
> >      .get_address_space = vtd_host_dma_iommu,
> > +    .set_iommu_context = vtd_dev_set_iommu_context,
> > +    .unset_iommu_context = vtd_dev_unset_iommu_context,
> >  };
> >
> >  static bool vtd_decide_config(IntelIOMMUState *s, Error **errp)
> > diff --git a/include/hw/i386/intel_iommu.h b/include/hw/i386/intel_iommu.h
> > index 3870052..9b4fc0a 100644
> > --- a/include/hw/i386/intel_iommu.h
> > +++ b/include/hw/i386/intel_iommu.h
> > @@ -64,6 +64,7 @@ typedef union VTD_IR_TableEntry VTD_IR_TableEntry;
> >  typedef union VTD_IR_MSIAddress VTD_IR_MSIAddress;
> >  typedef struct VTDPASIDDirEntry VTDPASIDDirEntry;
> >  typedef struct VTDPASIDEntry VTDPASIDEntry;
> > +typedef struct VTDHostIOMMUContext VTDHostIOMMUContext;
> >
> >  /* Context-Entry */
> >  struct VTDContextEntry {
> > @@ -112,10 +113,20 @@ struct VTDAddressSpace {
> >      IOVATree *iova_tree;          /* Traces mapped IOVA ranges */
> >  };
> >
> > +struct VTDHostIOMMUContext {
> > +    VTDBus *vtd_bus;
> > +    uint8_t devfn;
> > +    HostIOMMUContext *host_icx;
> > +    IntelIOMMUState *iommu_state;
> > +};
> > +
> >  struct VTDBus {
> > -    PCIBus* bus;		/* A reference to the bus to provide translation
> for */
> > +    /* A reference to the bus to provide translation for */
> > +    PCIBus *bus;
> >      /* A table of VTDAddressSpace objects indexed by devfn */
> > -    VTDAddressSpace *dev_as[];
> > +    VTDAddressSpace *dev_as[PCI_DEVFN_MAX];
> > +    /* A table of VTDHostIOMMUContext objects indexed by devfn */
> > +    VTDHostIOMMUContext *dev_icx[PCI_DEVFN_MAX];
> >  };
> >
> >  struct VTDIOTLBEntry {
> > @@ -271,6 +282,8 @@ struct IntelIOMMUState {
> >      /*
> >       * Protects IOMMU states in general.  Currently it protects the
> >       * per-IOMMU IOTLB cache, and context entry cache in VTDAddressSpace.
> > +     * Protect the update/usage of HostIOMMUContext pointer cached in
> > +     * VTDBus->dev_icx array as array elements may be updated by hotplug
> 
> I think the context update does not need to be updated, because they
> should always be with the BQL, right?

Hmmmm, maybe I used bad description. My purpose is to protect the stored
HostIOMMUContext pointer in vIOMMU. With pci_device_set/unset_iommu_context,
vIOMMU have a copy of HostIOMMUContext. If VFIO container is released
(e.g. hotpulg out device), HostIOMMUContext will alos be released. This
will trigger the pci_device_unset_iommu_context() to clean the copy. To
avoid using a staled HostIOMMUContext in vIOMMU, vIOMMU should have a
lock to block the pci_device_unset_iommu_context() calling until other
threads finished their HostIOMMUContext usage. Do you want a description
update here or other preference?

Regards,
Yi Liu

  reply index

Thread overview: 80+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-22 12:35 [PATCH v1 00/22] intel_iommu: expose Shared Virtual Addressing to VMs Liu Yi L
2020-03-22 12:35 ` [PATCH v1 01/22] scripts/update-linux-headers: Import iommu.h Liu Yi L
2020-03-22 12:35 ` [PATCH v1 02/22] header file update VFIO/IOMMU vSVA APIs Liu Yi L
2020-03-29 16:32   ` Auger Eric
2020-03-30  7:06     ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 03/22] vfio: check VFIO_TYPE1_NESTING_IOMMU support Liu Yi L
2020-03-22 12:36 ` [PATCH v1 04/22] hw/iommu: introduce HostIOMMUContext Liu Yi L
2020-03-23 20:58   ` Peter Xu
2020-03-24 10:00     ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 05/22] hw/pci: modify pci_setup_iommu() to set PCIIOMMUOps Liu Yi L
2020-03-22 12:36 ` [PATCH v1 06/22] hw/pci: introduce pci_device_set/unset_iommu_context() Liu Yi L
2020-03-23 21:15   ` Peter Xu
2020-03-24 10:02     ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 07/22] intel_iommu: add set/unset_iommu_context callback Liu Yi L
2020-03-23 21:29   ` Peter Xu
2020-03-24 11:15     ` Liu, Yi L [this message]
2020-03-24 15:24       ` Peter Xu
2020-03-25  9:37         ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 08/22] vfio: init HostIOMMUContext per-container Liu Yi L
     [not found]   ` <20200323213943.GR127076@xz-x1>
2020-03-24 13:03     ` Liu, Yi L
2020-03-24 14:45       ` Peter Xu
2020-03-25  9:30         ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 09/22] vfio/common: check PASID alloc/free availability Liu Yi L
2020-03-23 22:06   ` Peter Xu
2020-03-24 11:18     ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 10/22] intel_iommu: add virtual command capability support Liu Yi L
2020-03-22 12:36 ` [PATCH v1 11/22] intel_iommu: process PASID cache invalidation Liu Yi L
2020-03-22 12:36 ` [PATCH v1 12/22] intel_iommu: add PASID cache management infrastructure Liu Yi L
2020-03-24 17:32   ` Peter Xu
2020-03-25 12:20     ` Liu, Yi L
2020-03-25 14:52       ` Peter Xu
2020-03-26  6:15         ` Liu, Yi L
2020-03-26 13:57           ` Liu, Yi L
2020-03-26 15:53             ` Peter Xu
2020-03-27  1:33               ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 13/22] vfio: add bind stage-1 page table support Liu Yi L
2020-03-24 17:41   ` Peter Xu
2020-03-25  9:49     ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 14/22] intel_iommu: bind/unbind guest page table to host Liu Yi L
2020-03-24 17:46   ` Peter Xu
2020-03-25 12:42     ` Liu, Yi L
2020-03-25 14:56       ` Peter Xu
2020-03-26  3:04         ` Liu, Yi L
2020-03-25 12:47     ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 15/22] intel_iommu: replay guest pasid bindings " Liu Yi L
2020-03-24 18:00   ` Peter Xu
2020-03-25 13:14     ` Liu, Yi L
2020-03-25 15:06       ` Peter Xu
2020-03-26  3:17         ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 16/22] intel_iommu: replay pasid binds after context cache invalidation Liu Yi L
2020-03-24 18:07   ` Peter Xu
2020-03-25 13:18     ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 17/22] intel_iommu: do not pass down pasid bind for PASID #0 Liu Yi L
2020-03-24 18:13   ` Peter Xu
2020-03-25 10:42     ` Liu, Yi L
2020-03-25 15:12       ` Peter Xu
2020-03-26  2:42         ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 18/22] vfio: add support for flush iommu stage-1 cache Liu Yi L
2020-03-24 18:19   ` Peter Xu
2020-03-25 10:40     ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 19/22] intel_iommu: process PASID-based iotlb invalidation Liu Yi L
2020-03-24 18:26   ` Peter Xu
2020-03-25 13:36     ` Liu, Yi L
2020-03-25 15:15       ` Peter Xu
2020-03-29 11:17         ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 20/22] intel_iommu: propagate PASID-based iotlb invalidation to host Liu Yi L
2020-03-24 18:34   ` Peter Xu
2020-03-25 13:21     ` Liu, Yi L
2020-03-26  5:41       ` Liu, Yi L
2020-03-26 13:02         ` Peter Xu
2020-03-26 13:22           ` Peter Xu
2020-03-26 13:33             ` Liu, Yi L
2020-03-26 13:23           ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 21/22] intel_iommu: process PASID-based Device-TLB invalidation Liu Yi L
2020-03-24 18:36   ` Peter Xu
2020-03-25  9:19     ` Liu, Yi L
2020-03-22 12:36 ` [PATCH v1 22/22] intel_iommu: modify x-scalable-mode to be string option Liu Yi L
2020-03-24 18:39   ` Peter Xu
2020-03-25 13:22     ` Liu, Yi L
2020-03-22 13:25 ` [PATCH v1 00/22] intel_iommu: expose Shared Virtual Addressing to VMs no-reply

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=A2975661238FB949B60364EF0F2C25743A20041A@SHSMSX104.ccr.corp.intel.com \
    --to=yi.l.liu@intel.com \
    --cc=alex.williamson@redhat.com \
    --cc=david@gibson.dropbear.id.au \
    --cc=ehabkost@redhat.com \
    --cc=eric.auger@redhat.com \
    --cc=hao.wu@intel.com \
    --cc=jacob.jun.pan@linux.intel.com \
    --cc=jean-philippe@linaro.org \
    --cc=jun.j.tian@intel.com \
    --cc=kevin.tian@intel.com \
    --cc=kvm@vger.kernel.org \
    --cc=mst@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=peterx@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=rth@twiddle.net \
    --cc=yi.y.sun@intel.com \
    --cc=yi.y.sun@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

QEMU-Devel Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/qemu-devel/0 qemu-devel/git/0.git
	git clone --mirror https://lore.kernel.org/qemu-devel/1 qemu-devel/git/1.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 qemu-devel qemu-devel/ https://lore.kernel.org/qemu-devel \
		qemu-devel@nongnu.org
	public-inbox-index qemu-devel

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.nongnu.qemu-devel


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git