From: Steve Sistare <steven.sistare@oracle.com>
To: qemu-devel@nongnu.org
Cc: Yi Liu <yi.l.liu@intel.com>, Eric Auger <eric.auger@redhat.com>,
Zhenzhong Duan <zhenzhong.duan@intel.com>,
Alex Williamson <alex.williamson@redhat.com>,
Cedric Le Goater <clg@redhat.com>,
"Michael S. Tsirkin" <mst@redhat.com>,
Peter Xu <peterx@redhat.com>, Fabiano Rosas <farosas@suse.de>,
Philippe Mathieu-Daude <philmd@linaro.org>,
David Hildenbrand <david@redhat.com>,
Steve Sistare <steven.sistare@oracle.com>
Subject: [RFC V1 02/12] iommufd: no DMA to BARs
Date: Sat, 20 Jul 2024 12:15:27 -0700 [thread overview]
Message-ID: <1721502937-87102-3-git-send-email-steven.sistare@oracle.com> (raw)
In-Reply-To: <1721502937-87102-1-git-send-email-steven.sistare@oracle.com>
Do not map VFIO PCI BARs for DMA. This stops a raft of warnings of the
following form at QEMU start time when using -object iommufd:
qemu-kvm: warning: IOMMU_IOAS_MAP failed: Bad address, PCI BAR?
qemu-kvm: vfio_container_dma_map(0x555558282db0, 0x8800010000, 0x4000, 0x7ffff7ff0000) = -14 (Bad address)
Signed-off-by: Steve Sistare <steven.sistare@oracle.com>
---
hw/vfio/common.c | 3 ++-
hw/vfio/helpers.c | 1 +
include/exec/memory.h | 1 +
3 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/hw/vfio/common.c b/hw/vfio/common.c
index da2e0ec..403d45a 100644
--- a/hw/vfio/common.c
+++ b/hw/vfio/common.c
@@ -248,7 +248,8 @@ static bool vfio_listener_skipped_section(MemoryRegionSection *section)
* are never accessed by the CPU and beyond the address width of
* some IOMMU hardware. TODO: VFIO should tell us the IOMMU width.
*/
- section->offset_within_address_space & (1ULL << 63);
+ section->offset_within_address_space & (1ULL << 63) ||
+ section->mr->no_dma;
}
/* Called with rcu_read_lock held. */
diff --git a/hw/vfio/helpers.c b/hw/vfio/helpers.c
index b14edd4..e4cfdd2 100644
--- a/hw/vfio/helpers.c
+++ b/hw/vfio/helpers.c
@@ -435,6 +435,7 @@ int vfio_region_mmap(VFIORegion *region)
memory_region_owner(region->mem),
name, region->mmaps[i].size,
region->mmaps[i].mmap);
+ region->mmaps[i].mem.no_dma = true;
g_free(name);
memory_region_add_subregion(region->mem, region->mmaps[i].offset,
®ion->mmaps[i].mem);
diff --git a/include/exec/memory.h b/include/exec/memory.h
index ea03ef2..850cc8c 100644
--- a/include/exec/memory.h
+++ b/include/exec/memory.h
@@ -794,6 +794,7 @@ struct MemoryRegion {
bool unmergeable;
uint8_t dirty_log_mask;
bool is_iommu;
+ bool no_dma;
RAMBlock *ram_block;
Object *owner;
/* owner as TYPE_DEVICE. Used for re-entrancy checks in MR access hotpath */
--
1.8.3.1
next prev parent reply other threads:[~2024-07-20 19:16 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-20 19:15 [RFC V1 00/12] Live update: iommufd Steve Sistare
2024-07-20 19:15 ` [RFC V1 01/12] vfio: move cpr_exec_notifier Steve Sistare
2024-07-20 19:15 ` Steve Sistare [this message]
2024-08-12 22:05 ` [RFC V1 02/12] iommufd: no DMA to BARs Alex Williamson
2024-08-13 1:39 ` Yi Liu
2024-08-13 14:53 ` Steven Sistare
2024-07-20 19:15 ` [RFC V1 03/12] iommufd: pass name to connect Steve Sistare
2024-07-20 19:15 ` [RFC V1 04/12] migration: cpr_find_fd_any Steve Sistare
2024-07-20 19:15 ` [RFC V1 05/12] iommufd: preserve device fd Steve Sistare
2024-07-20 19:15 ` [RFC V1 06/12] iommufd: export iommufd_cdev_get_info_iova_range Steve Sistare
2024-07-20 19:15 ` [RFC V1 07/12] iommufd: change_process kernel interface Steve Sistare
2024-07-20 19:15 ` [RFC V1 08/12] vfio/iommufd: register container for cpr Steve Sistare
2024-07-20 19:15 ` [RFC V1 09/12] vfio/iommufd: rebuild device Steve Sistare
2024-07-20 19:15 ` [RFC V1 10/12] migration/ram: old host address Steve Sistare
2024-08-16 17:57 ` Fabiano Rosas
2024-08-16 18:13 ` Steven Sistare
2024-07-20 19:15 ` [RFC V1 11/12] iommufd: update DMA virtual addresses Steve Sistare
2024-07-20 19:15 ` [RFC V1 12/12] vfio: mdev blocker Steve Sistare
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=1721502937-87102-3-git-send-email-steven.sistare@oracle.com \
--to=steven.sistare@oracle.com \
--cc=alex.williamson@redhat.com \
--cc=clg@redhat.com \
--cc=david@redhat.com \
--cc=eric.auger@redhat.com \
--cc=farosas@suse.de \
--cc=mst@redhat.com \
--cc=peterx@redhat.com \
--cc=philmd@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=yi.l.liu@intel.com \
--cc=zhenzhong.duan@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).