There are 3 virtqueues (ctrl, event and cmd) for virtio scsi device,
but seabios will only set the physical address for the 3rd one (cmd).
Then in vhost_virtqueue_start(), virtio_queue_get_desc_addr()
will be 0 for ctrl and event vq.
In this case, ctrl and event vq are not initialized.
vhost_verify_ring_mappings may use uninitialized vhost_virtqueue
such that vhost_verify_ring_part_mapping returns ENOMEM.
When encountered this problem, we got the following logs:
qemu-system-x86_64: Unable to map available ring for ring 0
qemu-system-x86_64: Verify ring failure on region 0
Signed-off-by: Forrest Liu <forrestl@synology.com>
Signed-off-by: yuchenlin <yuchenlin@synology.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
It seems that the intel link is unavailable, change it to point to the
qemu site.
Signed-off-by: Li Qiang <liq3ea@163.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Make them more QOMConventional.
Cc:qemu-trivial@nongnu.org
Signed-off-by: Li Qiang <liq3ea@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Cc: qemu-trivial@nongnu.org
Signed-off-by: Li Qiang <liq3ea@gmail.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Mao Zhongyi <maozhongyi@cmss.chinamobile.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Noted while refactoring:
CC mips-softmmu/hw/mips/gt64xxx_pci.o
In file included from include/hw/pci-host/gt64xxx.h:2,
from hw/mips/gt64xxx_pci.c:30:
include/hw/pci/pci_bus.h:23:5: error: unknown type name ‘PCIIOMMUFunc’
PCIIOMMUFunc iommu_fn;
^~~~~~~~~~~~
include/hw/pci/pci_bus.h:27:5: error: unknown type name ‘pci_set_irq_fn’
pci_set_irq_fn set_irq;
^~~~~~~~~~~~~~
include/hw/pci/pci_bus.h:28:5: error: unknown type name ‘pci_map_irq_fn’
pci_map_irq_fn map_irq;
^~~~~~~~~~~~~~
include/hw/pci/pci_bus.h:29:5: error: unknown type name ‘pci_route_irq_fn’
pci_route_irq_fn route_intx_to_irq;
^~~~~~~~~~~~~~~~
include/hw/pci/pci_bus.h:31:24: error: ‘PCI_SLOT_MAX’ undeclared here (not in a function)
PCIDevice *devices[PCI_SLOT_MAX * PCI_FUNC_MAX];
^~~~~~~~~~~~
include/hw/pci/pci_bus.h:31:39: error: ‘PCI_FUNC_MAX’ undeclared here (not in a function)
PCIDevice *devices[PCI_SLOT_MAX * PCI_FUNC_MAX];
^~~~~~~~~~~~
make[1]: *** [rules.mak:69: hw/mips/gt64xxx_pci.o] Error 1
make: *** [Makefile:482: subdir-mips-softmmu] Error 2
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Introduced in 48ebf2f90f and faf1e708d5, these functions
were never used. Remove them.
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
In commit 9fa99d2519 ("hw/pci-host: Fix x86 Host Bridges 64bit PCI
hole", 2017-11-16), we meant to expose such a 64-bit PCI MMIO aperture in
the ACPI DSDT that would be at least as large as the new "pci-hole64-size"
property (2GB on i440fx, 32GB on q35). The goal was to offer "enough"
64-bit MMIO aperture to the guest OS for hotplug purposes.
Previous patch fixed the issue that the aperture is extended relative to
a possibly incorrect base. This may result in an aperture size that is
smaller than the intent of commit 9fa99d2519.
This patch adds a test to make sure it won't happen again.
In the test case being added:
- use 128 MB initial RAM size,
- ask for one DIMM hotplug slot,
- ask for 2 GB maximum RAM size,
- use a pci-testdev with a 64-bit BAR of 2 GB size.
Consequences:
(1) In pc_memory_init() [hw/i386/pc.c], the DIMM hotplug area size is
initially set to 2048-128 = 1920 MB. (Maximum RAM size minus initial
RAM size.)
(2) The DIMM area base is set to 4096 MB (because the initial RAM is only
128 MB -- there is no initial "high RAM").
(3) Due to commit 085f8e88ba ("pc: count in 1Gb hugepage alignment when
sizing hotplug-memory container", 2014-11-24), we add 1 GB for the one
DIMM hotplug slot that was specified. This sets the DIMM area size to
1920+1024 = 2944 MB.
(4) The reserved-memory-end address (exclusive) is set to 4096 + 2944 =
7040 MB (DIMM area base plus DIMM area size).
(5) The reserved-memory-end address is rounded up to GB alignment,
yielding 7 GB (7168 MB).
(6) Given the 2 GB BAR size of pci-testdev, SeaBIOS allocates said 64-bit
BAR in 64-bit address space.
(7) Because reserved-memory-end is at 7 GB, it is unaligned for the 2 GB
BAR. Therefore SeaBIOS allocates the BAR at 8 GB. QEMU then
(correctly) assigns the root bridge aperture base this BAR address, to
be exposed in \_SB.PCI0._CRS.
(8) The intent of commit 9fa99d2519 dictates that QEMU extend the
aperture size to 32 GB, implying a 40 GB end address. However, QEMU
performs the extension relative to reserved-memory-end (7 GB), not
relative to the bridge aperture base that was correctly deduced from
SeaBIOS's BAR programming (8 GB). Therefore we see 39 GB as the
aperture end address in \_SB.PCI0._CRS:
> QWordMemory (ResourceProducer, PosDecode, MinFixed, MaxFixed, Cacheable, ReadWrite,
> 0x0000000000000000, // Granularity
> 0x0000000200000000, // Range Minimum
> 0x00000009BFFFFFFF, // Range Maximum
> 0x0000000000000000, // Translation Offset
> 0x00000007C0000000, // Length
> ,, , AddressRangeMemory, TypeStatic)
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
In commit 9fa99d2519 ("hw/pci-host: Fix x86 Host Bridges 64bit PCI
hole", 2017-11-16), we meant to expose such a 64-bit PCI MMIO aperture in
the ACPI DSDT that would be at least as large as the new "pci-hole64-size"
property (2GB on i440fx, 32GB on q35). The goal was to offer "enough"
64-bit MMIO aperture to the guest OS for hotplug purposes.
In that commit, we added or modified five functions:
- pc_pci_hole64_start(): shared between i440fx and q35. Provides a default
64-bit base, which starts beyond the cold-plugged 64-bit RAM, and skips
the DIMM hotplug area too (if any).
- i440fx_pcihost_get_pci_hole64_start(), q35_host_get_pci_hole64_start():
board-specific 64-bit base property getters called abstractly by the
ACPI generator. Both of these fall back to pc_pci_hole64_start() if the
firmware didn't program any 64-bit hole (i.e. if the firmware didn't
assign a 64-bit GPA to any MMIO BAR on any device). Otherwise, they
honor the firmware's BAR assignments (i.e., they treat the lowest 64-bit
GPA programmed by the firmware as the base address for the aperture).
- i440fx_pcihost_get_pci_hole64_end(), q35_host_get_pci_hole64_end():
these intended to extend the aperture to our size recommendation,
calculated relative to the base of the aperture.
Despite the original intent, i440fx_pcihost_get_pci_hole64_end() and
q35_host_get_pci_hole64_end() currently only extend the aperture relative
to the default base (pc_pci_hole64_start()), ignoring any programming done
by the firmware. This means that our size recommendation may not be met.
Fix it by honoring the firmware's address assignments.
The strange extension sizes were spotted by Alex, in the log of a guest
kernel running on top of OVMF (which prefers to assign 64-bit GPAs to
64-bit BARs).
This change only affects DSDT generation, therefore no new compat property
is being introduced.
Using an i440fx OVMF guest with 5GB RAM, an example _CRS change is:
> @@ -881,9 +881,9 @@
> QWordMemory (ResourceProducer, PosDecode, MinFixed, MaxFixed, Cacheable, ReadWrite,
> 0x0000000000000000, // Granularity
> 0x0000000800000000, // Range Minimum
> - 0x000000080001C0FF, // Range Maximum
> + 0x000000087FFFFFFF, // Range Maximum
> 0x0000000000000000, // Translation Offset
> - 0x000000000001C100, // Length
> + 0x0000000080000000, // Length
> ,, , AddressRangeMemory, TypeStatic)
> })
> Device (GPE0)
(On i440fx, the low RAM split is at 3GB, in this case. Therefore, with 5GB
guest RAM and no DIMM hotplug range, pc_pci_hole64_start() returns 4 +
(5-3) = 6 GB. Adding the 2GB extension to that yields 8GB, which is below
the firmware-programmed base of 32GB, before the patch. Therefore, before
the patch, the extension is ineffective. After the patch, we add the 2GB
extension to the firmware-programmed base, namely 32GB.)
Using a q35 OVMF guest with 5GB RAM, an example _CRS change is:
> @@ -3162,9 +3162,9 @@
> QWordMemory (ResourceProducer, PosDecode, MinFixed, MaxFixed, Cacheable, ReadWrite,
> 0x0000000000000000, // Granularity
> 0x0000000800000000, // Range Minimum
> - 0x00000009BFFFFFFF, // Range Maximum
> + 0x0000000FFFFFFFFF, // Range Maximum
> 0x0000000000000000, // Translation Offset
> - 0x00000001C0000000, // Length
> + 0x0000000800000000, // Length
> ,, , AddressRangeMemory, TypeStatic)
> })
> Device (GPE0)
(On Q35, the low RAM split is at 2GB. Therefore, with 5GB guest RAM and no
DIMM hotplug range, pc_pci_hole64_start() returns 4 + (5-2) = 7 GB. Adding
the 32GB extension to that yields 39GB (0x0000_0009_BFFF_FFFF + 1), before
the patch. After the patch, we add the 32GB extension to the
firmware-programmed base, namely 32GB.)
The ACPI test data for the bios-tables-test case that we added earlier in
this series are corrected too, as follows:
> @@ -3339,9 +3339,9 @@
> QWordMemory (ResourceProducer, PosDecode, MinFixed, MaxFixed, Cacheable, ReadWrite,
> 0x0000000000000000, // Granularity
> 0x0000000200000000, // Range Minimum
> - 0x00000009BFFFFFFF, // Range Maximum
> + 0x00000009FFFFFFFF, // Range Maximum
> 0x0000000000000000, // Translation Offset
> - 0x00000007C0000000, // Length
> + 0x0000000800000000, // Length
> ,, , AddressRangeMemory, TypeStatic)
> })
> Device (GPE0)
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Fixes: 9fa99d2519
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Expose the calculated "hole64 start" GPAs as plain uint64_t values,
extracting the internals of the current property getters.
This patch doesn't change behavior.
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Add memory bar to pci-testdev. Size is configurable using the membar
property. Setting the size to zero (default) turns it off. Can be used
to check whether guests handle large pci bars correctly.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Tested-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The "tests/acpi-test-data" files are currently not covered by any section
in MAINTAINERS, and "scripts/checkpatch.pl" complains when new data files
are added.
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Alex Williamson <alex.williamson@redhat.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Now that amd-iommu support interrupt remapping, enable the GASup in IVRS
table and GASup in extended feature register to indicate that IOMMU
support guest virtual APIC mode. GASup provides option to guest OS to
make use of 128-bit IRTE.
Note that the GAMSup is set to zero to indicate that amd-iommu does not
support guest virtual APIC mode (aka AVIC) which would be used for the
nested VMs.
See Table 21 from IOMMU spec for interrupt virtualization controls
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Cc: Tom Lendacky <Thomas.Lendacky@amd.com>
Cc: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Emulate the interrupt remapping support when guest virtual APIC is
enabled.
For more information refer: IOMMU spec rev 3.0 (section 2.2.5.2)
When VAPIC is enabled, it uses interrupt remapping as defined in
Table 22 and Figure 17 from IOMMU spec.
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Cc: Tom Lendacky <Thomas.Lendacky@amd.com>
Cc: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
When interrupt remapping is enabled, add a special IVHD device
(type IOAPIC).
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Acked-by: Peter Xu <peterx@redhat.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Cc: Tom Lendacky <Thomas.Lendacky@amd.com>
Cc: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Emulate the interrupt remapping support when guest virtual APIC is
not enabled.
For more info Refer: AMD IOMMU spec Rev 3.0 - section 2.2.5.1
When VAPIC is not enabled, it uses interrupt remapping as defined in
Table 20 and Figure 15 from IOMMU spec.
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Cc: Tom Lendacky <Thomas.Lendacky@amd.com>
Cc: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Register the interrupt remapping callback and read/write ops for the
amd-iommu-ir memory region.
amd-iommu-ir is set to higher priority to ensure that this region won't
be masked out by other memory regions.
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Cc: Tom Lendacky <Thomas.Lendacky@amd.com>
Cc: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
To be consistent with intel-iommu:
- rename the address space to use '_' instead of '-'
- update the memory region relationships
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Cc: Tom Lendacky <Thomas.Lendacky@amd.com>
Cc: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Currently, the amdvi_validate_dte() assumes that a valid DTE will
always have V=1. This is not true. The V=1 means that bit[127:1] are
valid. A valid DTE can have IV=1 and V=0 (i.e address translation
disabled and interrupt remapping enabled)
Remove the V=1 check from amdvi_validate_dte(), make the caller
responsible to check for V or IV bits.
This also fixes a bug in existing code that when error is
detected during the translation we'll fail the translation
instead of assuming a passthrough mode.
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Cc: Tom Lendacky <Thomas.Lendacky@amd.com>
Cc: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The vtd_generate_msi_message() in intel-iommu is used to construct a MSI
Message from IRQ. A similar function will be needed when we add interrupt
remapping support in amd-iommu. Moving the function in common file to
avoid the code duplication. Rename it to x86_iommu_irq_to_msi_message().
There is no logic changes in the code flow.
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Suggested-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Cc: Tom Lendacky <Thomas.Lendacky@amd.com>
Cc: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Interrupt remapping needs kernel-irqchip={off|split} on both Intel and AMD
platforms. Move the check in common place.
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
Cc: Tom Lendacky <Thomas.Lendacky@amd.com>
Cc: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Some old guests (before commit 7a11370e5: "virtio_blk: enable VQs early")
kick virtqueue before setting VIRTIO_CONFIG_S_DRIVER_OK. This violates
the virtio spec. But virtio 1.0 transitional devices support this behaviour.
So we should start vhost when guest kicks in this case.
Signed-off-by: Yongji Xie <xieyongji@baidu.com>
Signed-off-by: Chai Wen <chaiwen@baidu.com>
Signed-off-by: Ni Xun <nixun@baidu.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
We should handle VTD_FR_CONTEXT_ENTRY_P properly when synchronizing
shadow page tables. Having invalid context entry there is perfectly
valid when we move a device out of an existing domain. When that
happens, instead of posting an error we invalidate the whole region.
Without this patch, QEMU will crash if we do these steps:
(1) start QEMU with VT-d IOMMU and two 10G NICs (ixgbe)
(2) bind the NICs with vfio-pci in the guest
(3) start testpmd with the NICs applied
(4) stop testpmd
(5) rebind the NIC back to ixgbe kernel driver
The patch should fix it.
Reported-by: Pei Zhang <pezhang@redhat.com>
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1627272
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
There are two callers for vtd_sync_shadow_page_table_range(): one
provided a valid context entry and one not. Move that fetching
operation into the caller vtd_sync_shadow_page_table() where we need to
fetch the context entry.
Meanwhile, remove the error_report_once() directly since we're already
tracing all the error cases in the previous call. Instead, return error
number back to caller. This will not change anything functional since
callers are dropping it after all.
We do this move majorly because we want to do something more later in
vtd_sync_shadow_page_table().
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
QEMU is not handling the global DMAR switch well, especially when from
"on" to "off".
Let's first take the example of system reset.
Assuming that a guest has IOMMU enabled. When it reboots, we will drop
all the existing DMAR mappings to handle the system reset, however we'll
still keep the existing memory layouts which has the IOMMU memory region
enabled. So after the reboot and before the kernel reloads again, there
will be no mapping at all for the host device. That's problematic since
any software (for example, SeaBIOS) that runs earlier than the kernel
after the reboot will assume the IOMMU is disabled, so any DMA from the
software will fail.
For example, a guest that boots on an assigned NVMe device might fail to
find the boot device after a system reboot/reset and we'll be able to
observe SeaBIOS errors if we capture the debugging log:
WARNING - Timeout at nvme_wait:144!
Meanwhile, we should see DMAR errors on the host of that NVMe device.
It's the DMA fault that caused a NVMe driver timeout.
The correct fix should be that we do proper switching of device DMA
address spaces when system resets, which will setup correct memory
regions and notify the backend of the devices. This might not affect
much on non-assigned devices since QEMU VT-d emulation will assume a
default passthrough mapping if DMAR is not enabled in the GCMD
register (please refer to vtd_iommu_translate). However that's required
for an assigned devices, since that'll rebuild the correct GPA to HPA
mapping that is needed for any DMA operation during guest bootstrap.
Besides the system reset, we have some other places that might change
the global DMAR status and we'd better do the same thing there. For
example, when we change the state of GCMD register, or the DMAR root
pointer. Do the same refresh for all these places. For these two
places we'll also need to explicitly invalidate the context entry cache
and iotlb cache.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1625173
CC: QEMU Stable <qemu-stable@nongnu.org>
Reported-by: Cong Li <coli@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
--
v2:
- do the same for GCMD write, or root pointer update [Alex]
- test is carried out by me this time, by observing the
vtd_switch_address_space tracepoint after system reboot
v3:
- rewrite commit message as suggested by Alex
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Provide the function and use it in vtd_init(). Used to reset both
context entry cache and iotlb cache for the whole IOMMU unit.
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Here should be submit_requests, there is no submit_merged_requests
function.
Signed-off-by: Yaowei Bai <baiyaowei@cmss.chinamobile.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
A few places in configure were doing ad-hoc calls to
the symlink function to set up symlinks from the build tree
back to the source tree. We have a loop that does this
already for all files and directories listed in the LINKS
environment variable; use that instead.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The FILES variable is used to accumulate a list of things to symlink
from the source tree into the build tree. These don't have to be
individual files; symlinking an entire directory of data files is
also fine. Rename it to something less confusing before we add a few
directories to it.
Improve the comment to clarify what DIRS and LINKS do and why
it's not a good idea to add things to LINKS with wildcarding.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Currently tests/hex-loader-check-data contains data files used
by the hexloader-test, and configure individually symlinks those
data files into the build directory using a wildcard.
Using a wildcard like this is a bad idea, because if a new
data file is added, nothing causes configure to be rerun,
and so no symlink is added for the new file. This can cause
tests to spuriously fail when they can't find their data.
Instead, it's better to symlink an entire directory of
data files. We already have such a directory: tests/data.
Move the data files from tests/hex-loader-check-data/ to
tests/data/hex-loader/, and remove the unnecessary symlinking.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Currently tests/acpi-test-data contains data files used by the
bios-tables-test, and configure individually symlinks those
data files into the build directory using a wildcard.
Using a wildcard like this is a bad idea, because if a new
data file is added, nothing causes configure to be rerun,
and so no symlink is added for the new file. This can cause
tests to spuriously fail when they can't find their data.
Instead, it's better to symlink an entire directory of
data files. We already have such a directory: tests/data.
Move the data files from tests/acpi-test-data/ to
tests/data/acpi/, and remove the unnecessary symlinking.
We can remove entirely the note in rebuild-expected-aml.sh
about copying any new data files, because now they will
be in the source directory, not the build directory, and
no copying is required.
(We can't just change the existing tests/acpi-test-data/
to being a symlinked directory, because if we did that and
a developer switched git branches from one after that change
to one before it then configure would end up trashing all
the test files by making them symlinks to themselves.
Changing their path avoids this annoyance.)
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
- auto-read-only option to fix commit job when used with -blockdev
- Fix help text related qemu-iotests failure (by improving the help text
and updating the reference output)
- quorum: Add missing checks when adding/removing child nodes
- Don't take address of fields in packed structs
- vvfat: Fix crash when reporting error about too many files in directory
-----BEGIN PGP SIGNATURE-----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=/9Om
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging
Block layer patches:
- auto-read-only option to fix commit job when used with -blockdev
- Fix help text related qemu-iotests failure (by improving the help text
and updating the reference output)
- quorum: Add missing checks when adding/removing child nodes
- Don't take address of fields in packed structs
- vvfat: Fix crash when reporting error about too many files in directory
# gpg: Signature made Mon 05 Nov 2018 15:35:25 GMT
# gpg: using RSA key 7F09B272C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>"
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74 56FE 7F09 B272 C88F 2FD6
* remotes/kevin/tags/for-upstream: (36 commits)
include: Add a comment to explain the origin of sizes' lookup table
vdi: Use a literal number of bytes for DEFAULT_CLUSTER_SIZE
fw_cfg: Drop newline in @file description
object: Make option help nicer to read
qdev-monitor: Make device options help nicer
chardev: Indent list of chardevs
option: Make option help nicer to read
qemu-iotests: Test auto-read-only with -drive and -blockdev
block: Make auto-read-only=on default for -drive
iscsi: Support auto-read-only option
gluster: Support auto-read-only option
curl: Support auto-read-only option
file-posix: Support auto-read-only option
nbd: Support auto-read-only option
block: Require auto-read-only for existing fallbacks
rbd: Close image in qemu_rbd_open() error path
block: Add auto-read-only option
block: Update flags in bdrv_set_read_only()
iotest: Test x-blockdev-change on a Quorum
quorum: Forbid adding children in blkverify mode
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
The lookup table for power-of-two sizes was added in commit 540b849261
for the purpose of having convenient shortcuts for these sizes in cases
when the literal number has to be present at compile time, and
expressions as '(1 * KiB)' can not be used. One such case is the
stringification of sizes. Beyond that, it is convenient to use these
shortcuts for all power-of-two sizes, even if they don't have to be
literal numbers.
Despite its convenience, this table introduced 55 lines of "dumb" code,
the purpose and origin of which are obscure without reading the message
of the commit which introduced it. This patch fixes that by adding a
comment to the code itself with a brief explanation for the reasoning
behind this table. This comment includes the short AWK script that
generated the table, so that anyone who's interested could make sure
that the values in it are correct (otherwise these values look as if
they were typed manually).
Signed-off-by: Leonid Bloch <lbloch@janustech.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
If an expression is used to define DEFAULT_CLUSTER_SIZE, when compiled,
it will be embedded as a literal expression in the binary (as the
default value) because it is stringified to mark the size of the default
value. Now this is fixed by using a defined number to define this value.
Signed-off-by: Leonid Bloch <lbloch@janustech.com>
Reviewed-by: Stefan Weil <sw@weilnetz.de>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
There is no good reason why there should be a newline in this
description, so remove it.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Just like in qemu_opts_print_help(), print the object name as a caption
instead of on every single line, indent all options, add angle brackets
around types, and align the descriptions after 24 characters.
Also, indent every object name in the list of available objects.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Just like in qemu_opts_print_help(), print the device name as a caption
instead of on every single line, indent all options, add angle brackets
around types, and align the descriptions after 24 characters. Also,
separate the descriptions with " - " instead of putting them in
parentheses, because that is what we do everywhere else. This does look
a bit funny here because basically all bits have the description
"on/off", but funny does not mean it is less readable.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Following the example of qemu_opts_print_help(), indent all entries in
the list of character devices.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
This adds some whitespace into the option help (including indentation)
and puts angle brackets around the type names. Furthermore, the list
name is no longer printed as part of every line, but only once in
advance, and only if the caller did not print a caption already.
This patch also restores the description alignment we had before commit
9cbef9d68e, just at 24 instead of 16 characters like we used to.
This increase is because now we have the type and two spaces of
indentation before the description, and with a usual type name length of
three chracters, this sums up to eight additional characters -- which
means that we now need 24 characters to get the same amount of padding
for most options. Also, 24 is a third of 80, which makes it kind of a
round number in terminal terms.
Finally, this patch amends the reference output of iotest 082 to match
the changes (and thus makes it pass again).
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
While we want machine interfaces like -blockdev and QMP blockdev-add to
add as little auto-detection as possible so that management tools are
explicit about their needs, -drive is a convenience option for human
users. Enabling auto-read-only=on by default there enables users to use
read-only images for read-only guest devices without having to specify
read-only=on explicitly. If they try to attach the image to a read-write
device, they will still get an error message.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
If read-only=off, but auto-read-only=on is given, open the volume
read-write if we have the permissions, but instead of erroring out for
read-only volumes, just degrade to read-only.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
If read-only=off, but auto-read-only=on is given, open the file
read-write if we have the permissions, but instead of erroring out for
read-only files, just degrade to read-only.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Niels de Vos <ndevos@redhat.com>
If read-only=off, but auto-read-only=on is given, just degrade to
read-only.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
If read-only=off, but auto-read-only=on is given, open the file
read-write if we have the permissions, but instead of erroring out for
read-only files, just degrade to read-only.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
If read-only=off, but auto-read-only=on is given, open a read-write NBD
connection if the server provides a read-write export, but instead of
erroring out for read-only exports, just degrade to read-only.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Some block drivers have traditionally changed their node to read-only
mode without asking the user. This behaviour has been marked deprecated
since 2.11, expecting users to provide an explicit read-only=on option.
Now that we have auto-read-only=on, enable these drivers to make use of
the option.
This is the only use of bdrv_set_read_only(), so we can make it a bit
more specific and turn it into a bdrv_apply_auto_read_only() that is
more convenient for drivers to use.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Commit e2b8247a32 introduced an error path in qemu_rbd_open() after
calling rbd_open(), but neglected to close the image again in this error
path. The error path should contain everything that the regular close
function qemu_rbd_close() contains.
This adds the missing rbd_close() call.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
