qemu/hw
Joao Martins 5b1e96e654 vfio/iommufd: Introduce auto domain creation
There's generally two modes of operation for IOMMUFD:

1) The simple user API which intends to perform relatively simple things
with IOMMUs e.g. DPDK. The process generally creates an IOAS and attaches
to VFIO and mainly performs IOAS_MAP and UNMAP.

2) The native IOMMUFD API where you have fine grained control of the
IOMMU domain and model it accordingly. This is where most new feature
are being steered to.

For dirty tracking 2) is required, as it needs to ensure that
the stage-2/parent IOMMU domain will only attach devices
that support dirty tracking (so far it is all homogeneous in x86, likely
not the case for smmuv3). Such invariant on dirty tracking provides a
useful guarantee to VMMs that will refuse incompatible device
attachments for IOMMU domains.

Dirty tracking insurance is enforced via HWPT_ALLOC, which is
responsible for creating an IOMMU domain. This is contrast to the
'simple API' where the IOMMU domain is created by IOMMUFD automatically
when it attaches to VFIO (usually referred as autodomains) but it has
the needed handling for mdevs.

To support dirty tracking with the advanced IOMMUFD API, it needs
similar logic, where IOMMU domains are created and devices attached to
compatible domains. Essentially mimicking kernel
iommufd_device_auto_get_domain(). With mdevs given there's no IOMMU domain
it falls back to IOAS attach.

The auto domain logic allows different IOMMU domains to be created when
DMA dirty tracking is not desired (and VF can provide it), and others where
it is. Here it is not used in this way given how VFIODevice migration
state is initialized after the device attachment. But such mixed mode of
IOMMU dirty tracking + device dirty tracking is an improvement that can
be added on. Keep the 'all of nothing' of type1 approach that we have
been using so far between container vs device dirty tracking.

Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Reviewed-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
[ clg: Added ERRP_GUARD() in iommufd_cdev_autodomains_get() ]
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
2024-07-23 17:14:52 +02:00
..
9pfs hw/xen: Make XenDevOps structures const 2024-06-04 11:53:43 +02:00
acpi hw/acpi/ich9: Remove dead code related to 'acpi_memory_hotplug' 2024-06-19 12:40:49 +02:00
adc aspeed/adc: Add AST2700 support 2024-07-21 07:46:38 +02:00
alpha alpha: switch boards to "default y" 2024-05-03 15:47:47 +02:00
arm aspeed: fix coding style 2024-07-21 07:46:38 +02:00
audio hw/audio/virtio-snd: Always use little endian audio format 2024-06-19 12:40:49 +02:00
avr avr: switch boards to "default y" 2024-05-03 15:47:47 +02:00
block hw/block: m25p80: support quad mode for w25q01jvq 2024-07-09 08:05:44 +02:00
char hw/char/pl011: Avoid division-by-zero in pl011_get_baudrate() 2024-07-11 11:41:33 +01:00
core hw: Fix crash that happens when introspecting scsi-block on older machine types 2024-07-22 13:56:11 +02:00
cpu hw: Add a Kconfig switch for the TYPE_CPU_CLUSTER device 2024-04-25 12:48:12 +02:00
cris cris: switch boards to "default y" 2024-05-03 15:47:47 +02:00
cxl hw/cxl/cxl-mailbox-utils: Add superset extent release mailbox support 2024-07-01 17:16:04 -04:00
display virtio-gpu-gl: declare dependency on ui-opengl 2024-07-22 12:47:28 +04:00
dma hw/dma: Add a trace log for a description loading failure 2024-06-18 14:52:00 +02:00
fsi hw/fsi: Aspeed APB2OPB & On-chip peripheral bus 2024-02-01 08:33:18 +01:00
gpio hw/gpio/aspeed: Add reg_table_count to AspeedGPIOClass 2024-07-02 07:52:43 +02:00
hppa hw/hppa/machine: Replace g_memdup() by g_memdup2() 2024-05-08 19:42:45 +02:00
hyperv kvm: move target-dependent interrupt routing out of kvm-all.c 2024-05-03 15:47:48 +02:00
i2c hw/i2c/aspeed: rename the I2C class pool attribute to share_pool 2024-07-21 07:46:38 +02:00
i386 trivial patches for 2024-07-17 2024-07-18 10:07:23 +10:00
ide hw/ide/macio: switch from using qemu_allocate_irq() to qdev input GPIOs 2024-07-02 06:58:48 +02:00
input hw: arm: Remove use of tabs in some source files 2024-05-28 14:20:48 +01:00
intc virtio: features,fixes 2024-07-03 20:54:17 -07:00
ipack hw/ipack: Constify VMState 2023-12-29 11:17:30 +11:00
ipmi hw/ipmi: Constify VMState 2023-12-29 11:17:30 +11:00
isa hw/isa/vt82c686: Turn "intr" irq into a named gpio 2024-07-16 20:04:08 +02:00
loongarch hw/loongarch/virt: Remove unused assignment 2024-07-12 09:41:18 +08:00
m68k hw: skip registration of outdated versioned machine types 2024-07-02 06:58:37 +02:00
mem hw/cxl/events: Improve QMP interfaces and documentation for add/release dynamic capacity. 2024-07-03 18:14:07 -04:00
microblaze microblaze: switch boards to "default y" 2024-05-03 15:47:47 +02:00
mips hw/isa/vt82c686: Turn "intr" irq into a named gpio 2024-07-16 20:04:08 +02:00
misc hw/misc: In STM32L4x5 EXTI, handle direct interrupts 2024-07-11 11:41:34 +01:00
net hw/net:ftgmac100: update TX and RX packet buffers address to 64 bits 2024-07-09 08:05:44 +02:00
nubus hw/nubus: add nubus-virtio-mmio device 2024-02-27 09:36:39 +01:00
nvme hw/nvme: remove useless type cast 2024-07-22 14:43:17 +02:00
nvram hw/nvram: Add BCM2835 OTP device 2024-07-01 12:48:55 +01:00
openrisc kconfig: express dependency of individual boards on libfdt 2024-05-10 15:45:15 +02:00
pci hw/pci: Replace -1 with UINT32_MAX for romsize 2024-07-03 18:14:07 -04:00
pci-bridge Misc HW patch queue 2024-04-25 09:43:29 -07:00
pci-host hw/ppc: Avoid using Monitor in pnv_phb4_pic_print_info() 2024-06-19 12:40:49 +02:00
pcmcia hw/pcmcia/pxa2xx: Inline pxa2xx_pcmcia_init() 2023-10-27 12:48:57 +01:00
ppc hw/isa/vt82c686: Turn "intr" irq into a named gpio 2024-07-16 20:04:08 +02:00
remote hw/remote/vfio-user: Fix config space access byte order 2024-05-08 19:43:15 +02:00
riscv hw/riscv/virt.c: re-insert and deprecate 'riscv,delegate' 2024-07-18 12:08:45 +10:00
rtc hw/i386: move rtc-reset-reinjection command out of hw/rtc 2024-05-10 15:45:15 +02:00
rx kconfig: express dependency of individual boards on libfdt 2024-05-10 15:45:15 +02:00
s390x virtio: features,fixes 2024-07-03 20:54:17 -07:00
scsi * target/i386/tcg: fixes for seg_helper.c 2024-07-17 15:40:28 +10:00
sd hw/sd/sdcard: Support boot area in emmc image 2024-07-16 20:30:15 +02:00
sensor hw, target: Add ResetType argument to hold and exit phase methods 2024-04-25 10:21:06 +01:00
sh4 hw/sh4: Remove newline character in trace events 2024-06-10 13:05:27 -04:00
smbios hw/smbios: Remove 'smbios_uuid_encoded', simplify smbios_encode_uuid() 2024-06-19 12:40:49 +02:00
sparc sparc: switch boards to "default y" 2024-05-03 15:47:48 +02:00
sparc64 qemu-sparc queue 2024-05-06 10:19:56 -07:00
ssi aspeed/smc: Fix possible integer overflow 2024-07-21 07:46:38 +02:00
timer hpet: fix HPET_TN_SETVAL for high 32-bits of the comparator 2024-07-16 18:18:24 +02:00
tpm hw/tpm: Remove HOST_PAGE_ALIGN from tpm_ppi_init 2024-02-29 11:35:36 -10:00
tricore tricore: switch boards to "default y" 2024-05-03 15:47:48 +02:00
ufs hw/ufs: Fix mcq register range check logic 2024-07-14 17:11:21 +09:00
usb hw/usb/u2f-passthru: Get rid of qemu_open_old() 2024-07-17 14:04:15 +03:00
vfio vfio/iommufd: Introduce auto domain creation 2024-07-23 17:14:52 +02:00
virtio virtio-mem: improve error message when unplug of device fails due to plugged memory 2024-07-10 18:06:24 +02:00
watchdog aspeed/wdt: Add AST2700 support 2024-06-16 21:08:54 +02:00
xen xen: mapcache: Fix unmapping of first entries in buckets 2024-07-12 00:17:36 +02:00
xenpv hw/xen: Register framebuffer backend via xen_backend_init() 2024-06-04 11:53:43 +02:00
xtensa hw/xtensa: require libfdt 2024-05-10 15:45:15 +02:00
Kconfig hw: Fix problem with the A*MPCORE switches in the Kconfig files 2024-04-25 12:48:12 +02:00
meson.build hw/rdma: Remove deprecated pvrdma device and rdmacm-mux helper 2024-04-24 16:03:38 +02:00