qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Zhenzhong Duan <zhenzhong.duan@intel.com>
To: qemu-devel@nongnu.org
Cc: alex.williamson@redhat.com, clg@redhat.com,
	eric.auger@redhat.com, mst@redhat.com, jasowang@redhat.com,
	peterx@redhat.com, ddutile@redhat.com, jgg@nvidia.com,
	nicolinc@nvidia.com, joao.m.martins@oracle.com,
	clement.mathieu--drif@eviden.com, kevin.tian@intel.com,
	yi.l.liu@intel.com, chao.p.peng@intel.com,
	Zhenzhong Duan <zhenzhong.duan@intel.com>
Subject: [PATCH v5 14/21] intel_iommu: Stick to system MR for IOMMUFD backed host device when x-fls=on
Date: Fri, 22 Aug 2025 02:40:52 -0400	[thread overview]
Message-ID: <20250822064101.123526-15-zhenzhong.duan@intel.com> (raw)
In-Reply-To: <20250822064101.123526-1-zhenzhong.duan@intel.com>

When guest in scalable mode and x-flts=on, we stick to system MR for IOMMUFD
backed host device. Then its default hwpt contains GPA->HPA mappings which is
used directly if PGTT=PT and used as nested parent if PGTT=FLT. Otherwise
fallback to original processing.

Suggested-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
---
 hw/i386/intel_iommu.c | 34 ++++++++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c
index af384ce7f0..15582977b8 100644
--- a/hw/i386/intel_iommu.c
+++ b/hw/i386/intel_iommu.c
@@ -1773,6 +1773,28 @@ static bool vtd_dev_pt_enabled(IntelIOMMUState *s, VTDContextEntry *ce,
 
 }
 
+static VTDHostIOMMUDevice *vtd_find_hiod_iommufd(IntelIOMMUState *s,
+                                                 VTDAddressSpace *as)
+{
+    struct vtd_as_key key = {
+        .bus = as->bus,
+        .devfn = as->devfn,
+    };
+    VTDHostIOMMUDevice *vtd_hiod = g_hash_table_lookup(s->vtd_host_iommu_dev,
+                                                       &key);
+
+    if (vtd_hiod && vtd_hiod->hiod &&
+        object_dynamic_cast(OBJECT(vtd_hiod->hiod),
+                            TYPE_HOST_IOMMU_DEVICE_IOMMUFD)) {
+        return vtd_hiod;
+    }
+    return NULL;
+}
+
+/*
+ * vtd_switch_address_space() calls vtd_as_pt_enabled() to determine which
+ * MR to switch to. Switch to system MR if return true, iommu MR otherwise.
+ */
 static bool vtd_as_pt_enabled(VTDAddressSpace *as)
 {
     IntelIOMMUState *s;
@@ -1781,6 +1803,18 @@ static bool vtd_as_pt_enabled(VTDAddressSpace *as)
     assert(as);
 
     s = as->iommu_state;
+
+    /*
+     * When guest in scalable mode and x-flts=on, we stick to system MR
+     * for IOMMUFD backed host device. Then its default hwpt contains
+     * GPA->HPA mappings which is used directly if PGTT=PT and used as
+     * nested parent if PGTT=FLT. Otherwise fallback to original
+     * processing.
+     */
+    if (s->root_scalable && s->flts && vtd_find_hiod_iommufd(s, as)) {
+        return true;
+    }
+
     if (vtd_dev_to_context_entry(s, pci_bus_num(as->bus), as->devfn,
                                  &ce)) {
         /*
-- 
2.47.1



  parent reply	other threads:[~2025-08-22  6:43 UTC|newest]

Thread overview: 113+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-08-22  6:40 [PATCH v5 00/21] intel_iommu: Enable stage-1 translation for passthrough device Zhenzhong Duan
2025-08-22  6:40 ` [PATCH v5 01/21] intel_iommu: Rename vtd_ce_get_rid2pasid_entry to vtd_ce_get_pasid_entry Zhenzhong Duan
2025-08-22 22:19   ` Nicolin Chen via
2025-08-25  6:01     ` Duan, Zhenzhong
2025-08-22  6:40 ` [PATCH v5 02/21] hw/pci: Introduce pci_device_get_viommu_cap() Zhenzhong Duan
2025-08-22 22:22   ` Nicolin Chen
2025-08-27 11:13   ` Yi Liu
2025-08-27 11:22     ` Eric Auger
2025-08-27 12:30       ` Yi Liu
2025-08-27 12:32         ` Eric Auger
2025-08-27 15:30           ` Nicolin Chen
2025-08-28  8:26             ` Yi Liu
2025-08-28  9:06               ` Duan, Zhenzhong
2025-08-29  1:54                 ` Duan, Zhenzhong
2025-08-29  3:26                   ` Nicolin Chen
2025-09-01  2:35                     ` Duan, Zhenzhong
2025-09-01  2:59                       ` Nicolin Chen
2025-09-01  3:31                         ` Duan, Zhenzhong
2025-08-22  6:40 ` [PATCH v5 03/21] intel_iommu: Implement get_viommu_cap() callback Zhenzhong Duan
2025-08-22 22:23   ` Nicolin Chen
2025-08-22  6:40 ` [PATCH v5 04/21] vfio: Introduce helper vfio_pci_from_vfio_device() Zhenzhong Duan
2025-08-22 22:40   ` Nicolin Chen via
2025-08-25  6:06     ` Duan, Zhenzhong
2025-08-27 11:13   ` Yi Liu
2025-08-27 11:34   ` Eric Auger
2025-09-01 16:36   ` Cédric Le Goater
2025-09-02  2:12     ` Duan, Zhenzhong
2025-08-22  6:40 ` [PATCH v5 05/21] vfio/iommufd: Force creating nested parent domain Zhenzhong Duan
2025-08-22 23:12   ` Nicolin Chen
2025-08-25  8:28     ` Duan, Zhenzhong
2025-08-27 11:51       ` Eric Auger
2025-08-27 11:48   ` Eric Auger
2025-08-28  9:53     ` Duan, Zhenzhong
2025-08-28 13:00       ` Eric Auger
2025-08-29  1:40         ` Duan, Zhenzhong
2025-08-29  3:47           ` Nicolin Chen
2025-08-22  6:40 ` [PATCH v5 06/21] hw/pci: Export pci_device_get_iommu_bus_devfn() and return bool Zhenzhong Duan
2025-08-22 23:13   ` Nicolin Chen
2025-08-27 11:14   ` Yi Liu
2025-08-22  6:40 ` [PATCH v5 07/21] intel_iommu: Introduce a new structure VTDHostIOMMUDevice Zhenzhong Duan
2025-08-22 23:17   ` Nicolin Chen
2025-08-26 17:21   ` Nicolin Chen
2025-08-27  6:45     ` Duan, Zhenzhong
2025-08-27  8:51       ` Nicolin Chen
2025-08-27 16:36     ` Eric Auger
2025-08-27 16:57       ` Nicolin Chen
2025-08-27 11:14   ` Yi Liu
2025-08-28  9:17     ` Duan, Zhenzhong
2025-08-29  2:57       ` Yi Liu
2025-08-22  6:40 ` [PATCH v5 08/21] intel_iommu: Check for compatibility with IOMMUFD backed device when x-flts=on Zhenzhong Duan
2025-08-27 11:42   ` Yi Liu
2025-08-28  9:37     ` Duan, Zhenzhong
2025-08-27 11:55   ` Eric Auger
2025-08-22  6:40 ` [PATCH v5 09/21] intel_iommu: Fail passthrough device under PCI bridge if x-flts=on Zhenzhong Duan
2025-08-28 10:33   ` Yi Liu
2025-09-01  5:14     ` Duan, Zhenzhong
2025-08-22  6:40 ` [PATCH v5 10/21] intel_iommu: Introduce two helpers vtd_as_from/to_iommu_pasid_locked Zhenzhong Duan
2025-08-28 11:36   ` Yi Liu
2025-09-01  5:33     ` Duan, Zhenzhong
2025-09-03  6:30       ` Yi Liu
2025-09-03  7:13         ` Duan, Zhenzhong
2025-08-22  6:40 ` [PATCH v5 11/21] intel_iommu: Handle PASID entry removal and update Zhenzhong Duan
2025-08-27 14:25   ` Eric Auger
2025-09-01  3:17     ` Duan, Zhenzhong
2025-08-28 12:05   ` Yi Liu
2025-09-01  3:31     ` Duan, Zhenzhong
2025-09-03  7:58       ` Yi Liu
2025-09-04  2:37         ` Duan, Zhenzhong
2025-08-22  6:40 ` [PATCH v5 12/21] intel_iommu: Handle PASID entry addition Zhenzhong Duan
2025-08-27 16:22   ` Eric Auger
2025-09-01  9:03     ` Duan, Zhenzhong
2025-09-03  8:52       ` Yi Liu
2025-09-04  2:45         ` Duan, Zhenzhong
2025-08-29  5:46   ` Yi Liu
2025-08-22  6:40 ` [PATCH v5 13/21] intel_iommu: Introduce a new pasid cache invalidation type FORCE_RESET Zhenzhong Duan
2025-08-27 16:28   ` Eric Auger
2025-08-29  5:56     ` Yi Liu
2025-09-01  9:04     ` Duan, Zhenzhong
2025-08-22  6:40 ` Zhenzhong Duan [this message]
2025-08-27 17:14   ` [PATCH v5 14/21] intel_iommu: Stick to system MR for IOMMUFD backed host device when x-fls=on Eric Auger
2025-08-29  6:06   ` Yi Liu
2025-08-22  6:40 ` [PATCH v5 15/21] intel_iommu: Bind/unbind guest page table to host Zhenzhong Duan
2025-08-28  8:37   ` Eric Auger
2025-08-29  7:05   ` Yi Liu
2025-08-22  6:40 ` [PATCH v5 16/21] intel_iommu: Replay pasid bindings after context cache invalidation Zhenzhong Duan
2025-08-28  9:43   ` Eric Auger
2025-08-29  7:35     ` Yi Liu
2025-09-01  8:11       ` Duan, Zhenzhong
2025-09-03 10:18         ` Yi Liu
2025-09-04  6:42           ` Duan, Zhenzhong
2025-08-22  6:40 ` [PATCH v5 17/21] intel_iommu: Propagate PASID-based iotlb invalidation to host Zhenzhong Duan
2025-08-28 10:00   ` Eric Auger
2025-08-28 12:11     ` Yi Liu
2025-09-01  8:32     ` Duan, Zhenzhong
2025-08-22  6:40 ` [PATCH v5 18/21] intel_iommu: Replay all pasid bindings when either SRTP or TE bit is changed Zhenzhong Duan
2025-08-28 10:02   ` Eric Auger
2025-08-22  6:40 ` [PATCH v5 19/21] vfio: Add a new element bypass_ro in VFIOContainerBase Zhenzhong Duan
2025-08-28 12:47   ` Eric Auger
2025-08-22  6:40 ` [PATCH v5 20/21] Workaround for ERRATA_772415_SPR17 Zhenzhong Duan
2025-08-22 23:55   ` Nicolin Chen
2025-08-25  9:21     ` Duan, Zhenzhong
2025-08-25 16:58       ` Nicolin Chen
2025-08-27  7:11         ` Duan, Zhenzhong
2025-08-27  8:42           ` Nicolin Chen
2025-08-27 11:56     ` Yi Liu
2025-08-27 15:09       ` Nicolin Chen
2025-08-29  8:16         ` Yi Liu
2025-08-29  8:54           ` Nicolin Chen
2025-08-22  6:40 ` [PATCH v5 21/21] intel_iommu: Enable host device when x-flts=on in scalable mode Zhenzhong Duan
2025-08-28 12:51   ` Eric Auger
2025-08-29  7:42   ` Yi Liu
2025-08-27 11:13 ` [PATCH v5 00/21] intel_iommu: Enable stage-1 translation for passthrough device Yi Liu
2025-08-28  5:53   ` Duan, Zhenzhong

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=20250822064101.123526-15-zhenzhong.duan@intel.com \
    --to=zhenzhong.duan@intel.com \
    --cc=alex.williamson@redhat.com \
    --cc=chao.p.peng@intel.com \
    --cc=clement.mathieu--drif@eviden.com \
    --cc=clg@redhat.com \
    --cc=ddutile@redhat.com \
    --cc=eric.auger@redhat.com \
    --cc=jasowang@redhat.com \
    --cc=jgg@nvidia.com \
    --cc=joao.m.martins@oracle.com \
    --cc=kevin.tian@intel.com \
    --cc=mst@redhat.com \
    --cc=nicolinc@nvidia.com \
    --cc=peterx@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --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).