From: David Gibson <david@gibson.dropbear.id.au>
To: Alexey Kardashevskiy <aik@ozlabs.ru>
Cc: Alex Williamson <alex.williamson@redhat.com>,
qemu-ppc@nongnu.org, qemu-devel@nongnu.org,
Gavin Shan <gwshan@linux.vnet.ibm.com>,
Alexander Graf <agraf@suse.de>
Subject: Re: [Qemu-devel] [RFC PATCH v2 11/13] spapr_pci_vfio: Enable DDW
Date: Tue, 26 Aug 2014 17:19:19 +1000 [thread overview]
Message-ID: <20140826071919.GX9923@voom.redhat.com> (raw)
In-Reply-To: <1408097555-28126-12-git-send-email-aik@ozlabs.ru>
[-- Attachment #1: Type: text/plain, Size: 4826 bytes --]
On Fri, Aug 15, 2014 at 08:12:33PM +1000, Alexey Kardashevskiy wrote:
> This implements DDW for VFIO. Host kernel support is required for this.
>
> Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
> ---
> Changes:
> v2:
> * remove()/reset() callbacks use spapr_pci's ones
> ---
> hw/ppc/spapr_pci_vfio.c | 86 +++++++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 86 insertions(+)
>
> diff --git a/hw/ppc/spapr_pci_vfio.c b/hw/ppc/spapr_pci_vfio.c
> index 11b4272..79df716 100644
> --- a/hw/ppc/spapr_pci_vfio.c
> +++ b/hw/ppc/spapr_pci_vfio.c
> @@ -71,6 +71,88 @@ static void spapr_phb_vfio_finish_realize(sPAPRPHBState *sphb, Error **errp)
> spapr_tce_get_iommu(tcet));
>
> object_unref(OBJECT(tcet));
> +
> + if (sphb->ddw_enabled) {
> + sphb->ddw_enabled = !!(info.flags & VFIO_IOMMU_SPAPR_TCE_FLAG_DDW);
This overrides an explicit ddw= set by the user, which is a bit
counter-intuitive.
> + }
> +}
> +
> +static int spapr_pci_vfio_ddw_query(sPAPRPHBState *sphb,
> + uint32_t *windows_available,
> + uint32_t *page_size_mask)
> +{
> + sPAPRPHBVFIOState *svphb = SPAPR_PCI_VFIO_HOST_BRIDGE(sphb);
> + struct vfio_iommu_spapr_tce_query query = { .argsz = sizeof(query) };
> + int ret;
> +
> + ret = vfio_container_ioctl(&sphb->iommu_as, svphb->iommugroupid,
> + VFIO_IOMMU_SPAPR_TCE_QUERY, &query);
> + if (ret) {
> + return ret;
> + }
> +
> + *windows_available = query.windows_available;
> + *page_size_mask = query.page_size_mask;
> +
> + return ret;
> +}
> +
> +static int spapr_pci_vfio_ddw_create(sPAPRPHBState *sphb, uint32_t page_shift,
> + uint32_t window_shift, uint32_t liobn,
> + sPAPRTCETable **ptcet)
> +{
> + sPAPRPHBVFIOState *svphb = SPAPR_PCI_VFIO_HOST_BRIDGE(sphb);
> + struct vfio_iommu_spapr_tce_create create = {
> + .argsz = sizeof(create),
> + .page_shift = page_shift,
> + .window_shift = window_shift,
> + .start_addr = 0
> + };
> + int ret;
> +
> + ret = vfio_container_ioctl(&sphb->iommu_as, svphb->iommugroupid,
> + VFIO_IOMMU_SPAPR_TCE_CREATE, &create);
> + if (ret) {
> + return ret;
> + }
> +
> + *ptcet = spapr_tce_new_table(DEVICE(sphb), liobn,
> + create.start_addr, page_shift,
> + 1ULL << (window_shift - page_shift),
> + true);
> + memory_region_add_subregion(&sphb->iommu_root, (*ptcet)->bus_offset,
> + spapr_tce_get_iommu(*ptcet));
> +
> + return ret;
> +}
> +
> +static int spapr_pci_vfio_ddw_remove(sPAPRPHBState *sphb, sPAPRTCETable *tcet)
> +{
> + sPAPRPHBVFIOState *svphb = SPAPR_PCI_VFIO_HOST_BRIDGE(sphb);
> + struct vfio_iommu_spapr_tce_remove remove = {
> + .argsz = sizeof(remove),
> + .start_addr = tcet->bus_offset
> + };
> + int ret;
> +
> + spapr_pci_ddw_remove(sphb, tcet);
> + ret = vfio_container_ioctl(&sphb->iommu_as, svphb->iommugroupid,
> + VFIO_IOMMU_SPAPR_TCE_REMOVE, &remove);
> +
> + return ret;
> +}
> +
> +static int spapr_pci_vfio_ddw_reset(sPAPRPHBState *sphb)
> +{
> + sPAPRPHBVFIOState *svphb = SPAPR_PCI_VFIO_HOST_BRIDGE(sphb);
> + struct vfio_iommu_spapr_tce_reset reset = { .argsz = sizeof(reset) };
> + int ret;
> +
> + spapr_pci_ddw_reset(sphb);
> + ret = vfio_container_ioctl(&sphb->iommu_as, svphb->iommugroupid,
> + VFIO_IOMMU_SPAPR_TCE_RESET, &reset);
Unlike the non-VFIO version, this doesn't appear to reset ddw_num.
Also, there isn't call to explicitly call DDW reset on system reset.
Is that handled in kernel by the overall VFIO reset?
> + return ret;
> }
>
> static void spapr_phb_vfio_class_init(ObjectClass *klass, void *data)
> @@ -80,6 +162,10 @@ static void spapr_phb_vfio_class_init(ObjectClass *klass, void *data)
>
> dc->props = spapr_phb_vfio_properties;
> spc->finish_realize = spapr_phb_vfio_finish_realize;
> + spc->ddw_query = spapr_pci_vfio_ddw_query;
> + spc->ddw_create = spapr_pci_vfio_ddw_create;
> + spc->ddw_remove = spapr_pci_vfio_ddw_remove;
> + spc->ddw_reset = spapr_pci_vfio_ddw_reset;
> }
>
> static const TypeInfo spapr_phb_vfio_info = {
--
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: Type: application/pgp-signature, Size: 819 bytes --]
next prev parent reply other threads:[~2014-08-26 7:20 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-08-15 10:12 [Qemu-devel] [RFC PATCH v2 00/13] spapr: vfio: Enable Dynamic DMA windows (DDW) Alexey Kardashevskiy
2014-08-15 10:12 ` [Qemu-devel] [RFC PATCH v2 01/13] qom: Make object_child_foreach safe for objects removal Alexey Kardashevskiy
2014-08-19 0:39 ` David Gibson
2014-08-15 10:12 ` [Qemu-devel] [RFC PATCH v2 02/13] spapr_iommu: Disable in-kernel IOMMU tables for >4GB windows Alexey Kardashevskiy
2014-08-19 0:43 ` David Gibson
2014-08-20 8:09 ` Alexey Kardashevskiy
2014-08-27 9:27 ` Alexander Graf
2014-08-15 10:12 ` [Qemu-devel] [RFC PATCH v2 03/13] spapr_pci: Make find_phb()/find_dev() public Alexey Kardashevskiy
2014-08-15 10:12 ` [Qemu-devel] [RFC PATCH v2 04/13] spapr_iommu: Make spapr_tce_find_by_liobn() public Alexey Kardashevskiy
2014-08-15 10:12 ` [Qemu-devel] [RFC PATCH v2 05/13] spapr_pci: Introduce a liobn number generating macros Alexey Kardashevskiy
2014-08-19 0:44 ` David Gibson
2014-08-27 9:29 ` Alexander Graf
2014-08-15 10:12 ` [Qemu-devel] [RFC PATCH v2 06/13] spapr_iommu: Implement free_table() helper Alexey Kardashevskiy
2014-08-26 6:16 ` David Gibson
2014-08-26 7:04 ` Alexey Kardashevskiy
2014-08-15 10:12 ` [Qemu-devel] [RFC PATCH v2 07/13] spapr_rtas: Add Dynamic DMA windows (DDW) RTAS calls support Alexey Kardashevskiy
2014-08-26 7:06 ` David Gibson
2014-08-27 9:36 ` Alexander Graf
2014-08-27 13:56 ` Alexey Kardashevskiy
2014-08-15 10:12 ` [Qemu-devel] [RFC PATCH v2 08/13] spapr_pci: Enable DDW Alexey Kardashevskiy
2014-08-26 7:14 ` David Gibson
2014-08-26 8:11 ` Alexey Kardashevskiy
2014-08-15 10:12 ` [Qemu-devel] [RFC PATCH v2 09/13] spapr_pci_vfio: Call spapr_pci::reset on reset Alexey Kardashevskiy
2014-08-26 6:55 ` David Gibson
2014-08-15 10:12 ` [Qemu-devel] [RFC PATCH v2 10/13] linux headers update for DDW Alexey Kardashevskiy
2014-08-18 17:42 ` Alex Williamson
2014-08-20 7:49 ` Alexey Kardashevskiy
2014-08-20 19:44 ` Alex Williamson
2014-08-21 2:47 ` Alexey Kardashevskiy
2014-08-15 10:12 ` [Qemu-devel] [RFC PATCH v2 11/13] spapr_pci_vfio: Enable DDW Alexey Kardashevskiy
2014-08-26 7:19 ` David Gibson [this message]
2014-08-26 8:16 ` Alexey Kardashevskiy
2014-08-27 8:25 ` David Gibson
2014-08-15 10:12 ` [Qemu-devel] [RFC PATCH v2 12/13] vfio: Enable DDW ioctls to VFIO IOMMU driver Alexey Kardashevskiy
2014-08-26 7:20 ` David Gibson
2014-08-26 8:20 ` Alexey Kardashevskiy
2014-08-27 8:42 ` David Gibson
2014-08-15 10:12 ` [Qemu-devel] [RFC PATCH v2 13/13] spapr: Add pseries-2.2 machine with default "ddw" option Alexey Kardashevskiy
2014-08-27 9:44 ` Alexander Graf
2014-08-27 14:24 ` Alexey Kardashevskiy
2014-08-27 9:44 ` Alexander Graf
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=20140826071919.GX9923@voom.redhat.com \
--to=david@gibson.dropbear.id.au \
--cc=agraf@suse.de \
--cc=aik@ozlabs.ru \
--cc=alex.williamson@redhat.com \
--cc=gwshan@linux.vnet.ibm.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@nongnu.org \
/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).