From: "Michael S. Tsirkin" <mst@redhat.com>
To: qemu-devel@nongnu.org
Cc: Peter Maydell <peter.maydell@linaro.org>,
Cornelia Huck <cohuck@redhat.com>,
Eric Auger <eric.auger@redhat.com>,
Jean-Philippe Brucker <jean-philippe@linaro.org>
Subject: [PULL 23/45] virtio-iommu: Support bypass domain
Date: Fri, 4 Mar 2022 08:40:12 -0500 [thread overview]
Message-ID: <20220304133556.233983-24-mst@redhat.com> (raw)
In-Reply-To: <20220304133556.233983-1-mst@redhat.com>
From: Jean-Philippe Brucker <jean-philippe@linaro.org>
The driver can create a bypass domain by passing the
VIRTIO_IOMMU_ATTACH_F_BYPASS flag on the ATTACH request. Bypass domains
perform slightly better than domains with identity mappings since they
skip translation.
Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Message-Id: <20220214124356.872985-4-jean-philippe@linaro.org>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
hw/virtio/virtio-iommu.c | 39 ++++++++++++++++++++++++++++++++++-----
1 file changed, 34 insertions(+), 5 deletions(-)
diff --git a/hw/virtio/virtio-iommu.c b/hw/virtio/virtio-iommu.c
index 4ca36db4ac..239fe97b12 100644
--- a/hw/virtio/virtio-iommu.c
+++ b/hw/virtio/virtio-iommu.c
@@ -43,6 +43,7 @@
typedef struct VirtIOIOMMUDomain {
uint32_t id;
+ bool bypass;
GTree *mappings;
QLIST_HEAD(, VirtIOIOMMUEndpoint) endpoint_list;
} VirtIOIOMMUDomain;
@@ -258,12 +259,16 @@ static void virtio_iommu_put_endpoint(gpointer data)
}
static VirtIOIOMMUDomain *virtio_iommu_get_domain(VirtIOIOMMU *s,
- uint32_t domain_id)
+ uint32_t domain_id,
+ bool bypass)
{
VirtIOIOMMUDomain *domain;
domain = g_tree_lookup(s->domains, GUINT_TO_POINTER(domain_id));
if (domain) {
+ if (domain->bypass != bypass) {
+ return NULL;
+ }
return domain;
}
domain = g_malloc0(sizeof(*domain));
@@ -271,6 +276,7 @@ static VirtIOIOMMUDomain *virtio_iommu_get_domain(VirtIOIOMMU *s,
domain->mappings = g_tree_new_full((GCompareDataFunc)interval_cmp,
NULL, (GDestroyNotify)g_free,
(GDestroyNotify)g_free);
+ domain->bypass = bypass;
g_tree_insert(s->domains, GUINT_TO_POINTER(domain_id), domain);
QLIST_INIT(&domain->endpoint_list);
trace_virtio_iommu_get_domain(domain_id);
@@ -334,11 +340,16 @@ static int virtio_iommu_attach(VirtIOIOMMU *s,
{
uint32_t domain_id = le32_to_cpu(req->domain);
uint32_t ep_id = le32_to_cpu(req->endpoint);
+ uint32_t flags = le32_to_cpu(req->flags);
VirtIOIOMMUDomain *domain;
VirtIOIOMMUEndpoint *ep;
trace_virtio_iommu_attach(domain_id, ep_id);
+ if (flags & ~VIRTIO_IOMMU_ATTACH_F_BYPASS) {
+ return VIRTIO_IOMMU_S_INVAL;
+ }
+
ep = virtio_iommu_get_endpoint(s, ep_id);
if (!ep) {
return VIRTIO_IOMMU_S_NOENT;
@@ -356,7 +367,12 @@ static int virtio_iommu_attach(VirtIOIOMMU *s,
}
}
- domain = virtio_iommu_get_domain(s, domain_id);
+ domain = virtio_iommu_get_domain(s, domain_id,
+ flags & VIRTIO_IOMMU_ATTACH_F_BYPASS);
+ if (!domain) {
+ /* Incompatible bypass flag */
+ return VIRTIO_IOMMU_S_INVAL;
+ }
QLIST_INSERT_HEAD(&domain->endpoint_list, ep, next);
ep->domain = domain;
@@ -419,6 +435,10 @@ static int virtio_iommu_map(VirtIOIOMMU *s,
return VIRTIO_IOMMU_S_NOENT;
}
+ if (domain->bypass) {
+ return VIRTIO_IOMMU_S_INVAL;
+ }
+
interval = g_malloc0(sizeof(*interval));
interval->low = virt_start;
@@ -464,6 +484,11 @@ static int virtio_iommu_unmap(VirtIOIOMMU *s,
if (!domain) {
return VIRTIO_IOMMU_S_NOENT;
}
+
+ if (domain->bypass) {
+ return VIRTIO_IOMMU_S_INVAL;
+ }
+
interval.low = virt_start;
interval.high = virt_end;
@@ -780,6 +805,9 @@ static IOMMUTLBEntry virtio_iommu_translate(IOMMUMemoryRegion *mr, hwaddr addr,
entry.perm = flag;
}
goto unlock;
+ } else if (ep->domain->bypass) {
+ entry.perm = flag;
+ goto unlock;
}
found = g_tree_lookup_extended(ep->domain->mappings, (gpointer)(&interval),
@@ -1139,8 +1167,8 @@ static const VMStateDescription vmstate_endpoint = {
static const VMStateDescription vmstate_domain = {
.name = "domain",
- .version_id = 1,
- .minimum_version_id = 1,
+ .version_id = 2,
+ .minimum_version_id = 2,
.pre_load = domain_preload,
.fields = (VMStateField[]) {
VMSTATE_UINT32(id, VirtIOIOMMUDomain),
@@ -1149,6 +1177,7 @@ static const VMStateDescription vmstate_domain = {
VirtIOIOMMUInterval, VirtIOIOMMUMapping),
VMSTATE_QLIST_V(endpoint_list, VirtIOIOMMUDomain, 1,
vmstate_endpoint, VirtIOIOMMUEndpoint, next),
+ VMSTATE_BOOL_V(bypass, VirtIOIOMMUDomain, 2),
VMSTATE_END_OF_LIST()
}
};
@@ -1186,7 +1215,7 @@ static const VMStateDescription vmstate_virtio_iommu_device = {
.version_id = 2,
.post_load = iommu_post_load,
.fields = (VMStateField[]) {
- VMSTATE_GTREE_DIRECT_KEY_V(domains, VirtIOIOMMU, 1,
+ VMSTATE_GTREE_DIRECT_KEY_V(domains, VirtIOIOMMU, 2,
&vmstate_domain, VirtIOIOMMUDomain),
VMSTATE_UINT8_V(config.bypass, VirtIOIOMMU, 2),
VMSTATE_END_OF_LIST()
--
MST
next prev parent reply other threads:[~2022-03-04 14:18 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-03-04 13:36 [PULL 00/45] virtio,pc,pci: features, cleanups, fixes Michael S. Tsirkin
2022-03-04 13:37 ` [PULL 01/45] qom: assert integer does not overflow Michael S. Tsirkin
2022-03-04 13:37 ` [PULL 02/45] ACPI ERST: specification for ERST support Michael S. Tsirkin
2022-03-04 13:37 ` [PULL 03/45] MAINTAINERS: no need to add my name explicitly as a reviewer for VIOT tables Michael S. Tsirkin
2022-03-04 13:37 ` [PULL 04/45] docs/acpi/erst: add device id for ACPI ERST device in pci-ids.txt Michael S. Tsirkin
2022-03-04 13:37 ` [PULL 05/45] hw/acpi/erst: clean up unused IS_UEFI_CPER_RECORD macro Michael S. Tsirkin
2022-03-04 13:37 ` [PULL 06/45] hw/smbios: code cleanup - use macro definitions for table header handles Michael S. Tsirkin
2022-03-04 13:38 ` [PULL 07/45] hw/smbios: fix overlapping table handle numbers with large memory vms Michael S. Tsirkin
2022-03-04 13:38 ` [PULL 08/45] hw/smbios: add assertion to ensure handles of tables 19 and 32 do not collide Michael S. Tsirkin
2022-03-04 13:38 ` [PULL 09/45] vhost-user: remove VirtQ notifier restore Michael S. Tsirkin
2022-03-04 13:38 ` [PULL 10/45] vhost-user: fix VirtQ notifier cleanup Michael S. Tsirkin
2022-03-04 13:38 ` [PULL 11/45] virtio: fix the condition for iommu_platform not supported Michael S. Tsirkin
2022-03-04 13:38 ` [PULL 12/45] hw/vhost-user-i2c: Add support for VIRTIO_I2C_F_ZERO_LENGTH_REQUEST Michael S. Tsirkin
2022-03-04 13:39 ` [PULL 13/45] tests/qtest/libqos/pci: Introduce pio_limit Michael S. Tsirkin
2022-03-04 13:39 ` [PULL 14/45] tests/qtest/libqos: Skip hotplug tests if pci root bus is not hotpluggable Michael S. Tsirkin
2022-03-04 13:39 ` [PULL 15/45] tests/qtest/vhost-user-blk-test: Temporary hack to get tests passing on aarch64 Michael S. Tsirkin
2022-03-04 13:39 ` [PULL 16/45] tests/qtest/libqos: Add generic pci host bridge in arm-virt machine Michael S. Tsirkin
2022-03-07 10:02 ` Michael S. Tsirkin
2022-03-07 11:07 ` Eric Auger
2022-03-04 13:39 ` [PULL 17/45] hw/virtio: vdpa: Fix leak of host-notifier memory-region Michael S. Tsirkin
2022-03-04 13:39 ` [PULL 18/45] vhost-vdpa: make notifiers _init()/_uninit() symmetric Michael S. Tsirkin
2022-03-04 13:39 ` [PULL 19/45] intel_iommu: support snoop control Michael S. Tsirkin
2022-03-04 13:39 ` [PULL 20/45] hw/i386: Improve bounds checking in OVMF table parsing Michael S. Tsirkin
2022-03-04 13:40 ` [PULL 21/45] hw/i386: Replace magic number with field length calculation Michael S. Tsirkin
2022-03-04 13:40 ` [PULL 22/45] virtio-iommu: Default to bypass during boot Michael S. Tsirkin
2022-03-04 13:40 ` Michael S. Tsirkin [this message]
2022-03-04 13:40 ` [PULL 24/45] tests/qtest/virtio-iommu-test: Check bypass config Michael S. Tsirkin
2022-03-04 13:40 ` [PULL 25/45] hw/i386/pc_piix: Mark the machine types from version 1.4 to 1.7 as deprecated Michael S. Tsirkin
2022-03-04 13:40 ` [PULL 26/45] hw/pci-bridge/pxb: Fix missing swizzle Michael S. Tsirkin
2022-03-04 13:40 ` [PULL 27/45] virtio-net: Unlimit tx queue size if peer is vdpa Michael S. Tsirkin
2022-03-04 13:40 ` [PULL 28/45] pcie: Add support for Single Root I/O Virtualization (SR/IOV) Michael S. Tsirkin
2022-03-04 13:40 ` [PULL 29/45] pcie: Add some SR/IOV API documentation in docs/pcie_sriov.txt Michael S. Tsirkin
2022-03-04 13:40 ` [PULL 30/45] pcie: Add a helper to the SR/IOV API Michael S. Tsirkin
2022-03-04 13:41 ` [PULL 31/45] pcie: Add 1.2 version token for the Power Management Capability Michael S. Tsirkin
2022-03-04 13:41 ` [PULL 32/45] pci-bridge/xio3130_upstream: Fix error handling Michael S. Tsirkin
2022-03-04 13:41 ` [PULL 33/45] pci-bridge/xio3130_downstream: " Michael S. Tsirkin
2022-03-04 13:41 ` [PULL 34/45] headers: Add pvpanic.h Michael S. Tsirkin
2022-03-04 13:41 ` [PULL 35/45] hw/misc/pvpanic: Use standard headers instead Michael S. Tsirkin
2022-03-04 13:41 ` [PULL 36/45] pci: show id info when pci BDF conflict Michael S. Tsirkin
2022-03-04 13:42 ` [PULL 37/45] pci: expose TYPE_XIO3130_DOWNSTREAM name Michael S. Tsirkin
2022-03-04 13:42 ` [PULL 38/45] acpi: pcihp: pcie: set power on cap on parent slot Michael S. Tsirkin
2022-03-04 13:42 ` [PULL 39/45] pc: add option to disable PS/2 mouse/keyboard Michael S. Tsirkin
2022-03-04 13:42 ` [PULL 40/45] vhost-vsock: detach the virqueue element in case of error Michael S. Tsirkin
2022-03-04 13:42 ` [PULL 41/45] x86: cleanup unused compat_apic_id_mode Michael S. Tsirkin
2022-03-04 13:42 ` [PULL 42/45] event_notifier: add event_notifier_get_wfd() Michael S. Tsirkin
2022-03-04 13:42 ` [PULL 43/45] vhost: use wfd on functions setting vring call fd Michael S. Tsirkin
2022-03-04 13:42 ` [PULL 44/45] configure, meson: allow enabling vhost-user on all POSIX systems Michael S. Tsirkin
2022-03-04 13:42 ` [PULL 45/45] docs: vhost-user: add subsection for non-Linux platforms Michael S. Tsirkin
2022-03-04 22:13 ` [PULL 00/45] virtio,pc,pci: features, cleanups, fixes Peter Maydell
2022-03-06 10:37 ` Michael S. Tsirkin
2022-03-07 10:48 ` Eric Auger
2022-03-07 11:55 ` Peter Maydell
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=20220304133556.233983-24-mst@redhat.com \
--to=mst@redhat.com \
--cc=cohuck@redhat.com \
--cc=eric.auger@redhat.com \
--cc=jean-philippe@linaro.org \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@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).