qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: David Gibson <david@gibson.dropbear.id.au>
To: Peter Xu <peterx@redhat.com>
Cc: qemu-devel@nongnu.org, tianyu.lan@intel.com,
	Paolo Bonzini <pbonzini@redhat.com>,
	kevin.tian@intel.com, yi.l.liu@intel.com,
	Jason Wang <jasowang@redhat.com>,
	Alex Williamson <alex.williamson@redhat.com>
Subject: Re: [Qemu-devel] [RFC PATCH 6/8] memory: introduce AddressSpaceOps
Date: Mon, 1 May 2017 14:58:22 +1000	[thread overview]
Message-ID: <20170501045822.GM13773@umbus.fritz.box> (raw)
In-Reply-To: <1493285660-4470-7-git-send-email-peterx@redhat.com>

[-- Attachment #1: Type: text/plain, Size: 4151 bytes --]

On Thu, Apr 27, 2017 at 05:34:18PM +0800, Peter Xu wrote:
> This is something similar to MemoryRegionOps, it's just for address
> spaces to store arch-specific hooks.
> 
> The first hook I would like to introduce is iommu_get().
> 
> For systems that have IOMMUs, we will create a special address space per
> device which is different from system default address space for
> it (please refer to pci_device_iommu_address_space()). Normally when
> that happens, there will be one specific IOMMU (or say, translation
> unit) stands right behind that new address space.
> 
> This iommu_get() fetches that guy behind the address space. Here, the
> guy is defined as IOMMUObject, which is currently a (void *). In the
> future, maybe we can make it a better definition, but imho it's good
> enough for now, considering it's arch-dependent.
> 
> Signed-off-by: Peter Xu <peterx@redhat.com>

This doesn't make sense to me.  It would be entirely possible for a
single address space to have different regions mapped by different
IOMMUs.  Or some regions mapped by IOMMUs and others direct mapped to
a device or memory block.

> ---
>  include/exec/memory.h | 30 ++++++++++++++++++++++++++++++
>  memory.c              |  8 ++++++++
>  2 files changed, 38 insertions(+)
> 
> diff --git a/include/exec/memory.h b/include/exec/memory.h
> index e5707b3..0b0b58b 100644
> --- a/include/exec/memory.h
> +++ b/include/exec/memory.h
> @@ -183,6 +183,15 @@ struct MemoryRegionOps {
>      const MemoryRegionMmio old_mmio;
>  };
>  
> +/*
> + * This stands for an IOMMU unit. Normally it should be exactly the
> + * IOMMU device, however this can also be actually anything which is
> + * related to that translation unit. What it is should be totally
> + * arch-dependent. Maybe one day we can have something better than a
> + * (void *) here.
> + */
> +typedef void *IOMMUObject;
> +
>  typedef struct MemoryRegionIOMMUOps MemoryRegionIOMMUOps;
>  
>  struct MemoryRegionIOMMUOps {
> @@ -282,6 +291,19 @@ struct MemoryListener {
>  };
>  
>  /**
> + * AddressSpaceOps: callbacks structure for address space specific operations
> + *
> + * @iommu_get: returns an IOMMU object that backs the address space.
> + *             Normally this should be NULL for generic address
> + *             spaces, and it's only used when there is one
> + *             translation unit behind this address space.
> + */
> +struct AddressSpaceOps {
> +    IOMMUObject *(*iommu_get)(AddressSpace *as);
> +};
> +typedef struct AddressSpaceOps AddressSpaceOps;
> +
> +/**
>   * AddressSpace: describes a mapping of addresses to #MemoryRegion objects
>   */
>  struct AddressSpace {
> @@ -302,6 +324,7 @@ struct AddressSpace {
>      MemoryListener dispatch_listener;
>      QTAILQ_HEAD(memory_listeners_as, MemoryListener) listeners;
>      QTAILQ_ENTRY(AddressSpace) address_spaces_link;
> +    AddressSpaceOps as_ops;
>  };
>  
>  /**
> @@ -1800,6 +1823,13 @@ address_space_write_cached(MemoryRegionCache *cache, hwaddr addr,
>      address_space_write(cache->as, cache->xlat + addr, MEMTXATTRS_UNSPECIFIED, buf, len);
>  }
>  
> +/**
> + * address_space_iommu_get: Get the backend IOMMU for the address space
> + *
> + * @as: the address space to fetch IOMMU from
> + */
> +IOMMUObject *address_space_iommu_get(AddressSpace *as);
> +
>  #endif
>  
>  #endif
> diff --git a/memory.c b/memory.c
> index 6af523e..6aaad45 100644
> --- a/memory.c
> +++ b/memory.c
> @@ -2500,6 +2500,14 @@ void address_space_destroy(AddressSpace *as)
>      call_rcu(as, do_address_space_destroy, rcu);
>  }
>  
> +IOMMUObject *address_space_iommu_get(AddressSpace *as)
> +{
> +    if (!as->as_ops.iommu_get) {
> +        return NULL;
> +    }
> +    return as->as_ops.iommu_get(as);
> +}
> +
>  static const char *memory_region_type(MemoryRegion *mr)
>  {
>      if (memory_region_is_ram_device(mr)) {

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

  reply	other threads:[~2017-05-01  5:02 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-27  9:34 [Qemu-devel] [RFC PATCH 0/8] IOMMU: introduce common IOMMUObject Peter Xu
2017-04-27  9:34 ` [Qemu-devel] [RFC PATCH 1/8] memory: rename IOMMU_NOTIFIER_* Peter Xu
2017-05-01  4:50   ` David Gibson
2017-04-27  9:34 ` [Qemu-devel] [RFC PATCH 2/8] memory: rename IOMMUNotifier Peter Xu
2017-05-01  4:51   ` David Gibson
2017-04-27  9:34 ` [Qemu-devel] [RFC PATCH 3/8] memory: rename iommu_notifier_init() Peter Xu
2017-05-01  4:53   ` David Gibson
2017-05-08  5:50     ` Peter Xu
2017-04-27  9:34 ` [Qemu-devel] [RFC PATCH 4/8] memory: rename *_notify_iommu* Peter Xu
2017-05-01  4:55   ` David Gibson
2017-04-27  9:34 ` [Qemu-devel] [RFC PATCH 5/8] memory: rename *iommu_notifier* Peter Xu
2017-05-01  4:56   ` David Gibson
2017-04-27  9:34 ` [Qemu-devel] [RFC PATCH 6/8] memory: introduce AddressSpaceOps Peter Xu
2017-05-01  4:58   ` David Gibson [this message]
2017-05-08  5:48     ` Peter Xu
2017-05-08  6:07       ` David Gibson
2017-05-08  7:32         ` Peter Xu
2017-05-07  9:44           ` Liu, Yi L
2017-05-10  7:04           ` David Gibson
2017-05-11  5:04             ` Peter Xu
2017-05-15  5:32               ` David Gibson
2017-05-25  7:24                 ` Peter Xu
2017-05-26  5:30                   ` David Gibson
2017-06-08  8:24                     ` Liu, Yi L
2017-04-27  9:34 ` [Qemu-devel] [RFC PATCH 7/8] intel_iommu: provide AddressSpaceOps.iommu_get() Peter Xu
2017-04-27  9:34 ` [Qemu-devel] [RFC PATCH 8/8] iommu: introduce hw/core/iommu Peter Xu
2017-04-28 10:01   ` Liu, Yi L
2017-04-28 10:34     ` Peter Xu
2017-06-07  7:51   ` Liu, Yi L
2017-06-07  8:28     ` Peter Xu

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=20170501045822.GM13773@umbus.fritz.box \
    --to=david@gibson.dropbear.id.au \
    --cc=alex.williamson@redhat.com \
    --cc=jasowang@redhat.com \
    --cc=kevin.tian@intel.com \
    --cc=pbonzini@redhat.com \
    --cc=peterx@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=tianyu.lan@intel.com \
    --cc=yi.l.liu@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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).