Merge branch 'arm/queue' into 'master'

arm/arm64: Default to 'max' cpu type and add MTE test

See merge request kvm-unit-tests/kvm-unit-tests!76

arm/arm64: efi: Bump number of memregions

When using QEMU's 'max' cpu type additional EFI memory regions are
present. Just like riscv's commit f35fe0ee486a ("riscv: efi: Bump
number of memregions"),

arm/arm64: efi: Bump number of memregions

When using QEMU's 'max' cpu type additional EFI memory regions are
present. Just like riscv's commit f35fe0ee486a ("riscv: efi: Bump
number of memregions"), bump the number we allow to ensure we don't
hit the assert in memregions_add().

Signed-off-by: Andrew Jones <andrew.jones@linux.dev>

show more ...

Merge branch 'arm/queue' into 'master'

arm/arm64: LPA2 support and fpu/sve s/r test

See merge request kvm-unit-tests/kvm-unit-tests!61

arm: Detect FDT overlap with uninitialised data

If the FDT was placed in a region overlapping the bss/stack area, it
would have been overwritten at early boot. Assert this never happened
to detect t

arm: Detect FDT overlap with uninitialised data

If the FDT was placed in a region overlapping the bss/stack area, it
would have been overwritten at early boot. Assert this never happened
to detect the case.

Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
[Fixed EFI compile error.]
Signed-off-by: Andrew Jones <andrew.jones@linux.dev>

show more ...

Merge branch 'arm/queue' into 'master'

arm/arm64: EFI improvements and no more MAX_SMP probing

See merge request kvm-unit-tests/kvm-unit-tests!53

arm64: efi: Switch to our own stack

We don't want to map EFI_BOOT_SERVICES_DATA regions, so move the
stack from its EFI_BOOT_SERVICES_DATA region to EFI_LOADER_CODE,
which we always map. We'll still

arm64: efi: Switch to our own stack

We don't want to map EFI_BOOT_SERVICES_DATA regions, so move the
stack from its EFI_BOOT_SERVICES_DATA region to EFI_LOADER_CODE,
which we always map. We'll still map the stack as R/W instead of
R/X because we split EFI_LOADER_CODE regions on the _etext boundary
and map addresses before _etext as R/X and the rest as R/W.

Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Signed-off-by: Andrew Jones <andrew.jones@linux.dev>

show more ...

arm64: efi: Don't map reserved regions

We shouldn't need to map all the regions that the EFI memory map
contains. Just map EFI_LOADER_CODE and EFI_LOADER_DATA, since
those are for the loaded unit te

arm64: efi: Don't map reserved regions

We shouldn't need to map all the regions that the EFI memory map
contains. Just map EFI_LOADER_CODE and EFI_LOADER_DATA, since
those are for the loaded unit test, and any region types which
could be used by the unit test for its own memory allocations. We
still map EFI_BOOT_SERVICES_DATA since the primary stack is on a
region of that type. In a later patch we'll switch to a stack we
allocate ourselves to drop that one too.

Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Signed-off-by: Andrew Jones <andrew.jones@linux.dev>

show more ...

arm64: Add memregions_efi_init

Provide a memregions function which initialized memregions from an
EFI memory map. Add a new memregions flag (MR_F_PERSISTENT) for
EFI_PERSISTENT_MEMORY since that typ

arm64: Add memregions_efi_init

Provide a memregions function which initialized memregions from an
EFI memory map. Add a new memregions flag (MR_F_PERSISTENT) for
EFI_PERSISTENT_MEMORY since that type should not be reserved, but it
should also be distinct from conventional memory. The function also
points out the largest conventional memory region by returning a
pointer to it in the freemem parameter. Immediately apply this
function to arm64's efi_mem_init(). riscv will make use of it as well.

Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Signed-off-by: Andrew Jones <andrew.jones@linux.dev>

show more ...

arm64: Simplify efi_mem_init

Reduce the EFI mem_map loop to only setting flags and finding the
largest free memory region. Then, apply memregions_split() for
the code/data region split (which requir

arm64: Simplify efi_mem_init

Reduce the EFI mem_map loop to only setting flags and finding the
largest free memory region. Then, apply memregions_split() for
the code/data region split (which requires ensuring _etext is
page aligned). Finally, do the rest of the things that used to be
done in the EFI mem_map loop in a separate mem_region loop.

Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Signed-off-by: Andrew Jones <andrew.jones@linux.dev>

show more ...

arm/arm64: Factor out allocator init from mem_init

The allocator init is identical for mem_init() and efi_mem_init().
Share it.

Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Signed-off-by:

arm/arm64: Factor out allocator init from mem_init

The allocator init is identical for mem_init() and efi_mem_init().
Share it.

Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Signed-off-by: Andrew Jones <andrew.jones@linux.dev>

show more ...

arm/arm64: Factor out some initial setup

Factor out some initial setup code into separate functions in order
to share more code between setup() and setup_efi().

Reviewed-by: Nikos Nikoleris <nikos.

arm/arm64: Factor out some initial setup

Factor out some initial setup code into separate functions in order
to share more code between setup() and setup_efi().

Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Signed-off-by: Andrew Jones <andrew.jones@linux.dev>

show more ...

Merge branch 'riscv/initial-port-v3' into 'master'

riscv: Initial port

See merge request kvm-unit-tests/kvm-unit-tests!50

arm/arm64: Share memregions

arm/arm64 (and to a small extent powerpc) have memory regions which
get built from hardware descriptions (DT/ACPI/EFI) and then used to
build page tables. Move memregions

arm/arm64: Share memregions

arm/arm64 (and to a small extent powerpc) have memory regions which
get built from hardware descriptions (DT/ACPI/EFI) and then used to
build page tables. Move memregions to common code, tweaking the API
a bit at the same time, e.g. change 'mem_region' to 'memregions'.
The biggest change is there is now a default number of memory regions
which, if too small, should be overridden at setup time with a new
init function, memregions_init().

Signed-off-by: Andrew Jones <andrew.jones@linux.dev>
Acked-by: Thomas Huth <thuth@redhat.com>

show more ...

Merge branch 'arm/queue' into 'master'

arm/arm64: EFI support, arm64 backtrace support, PMU test improvements, and more

See merge request kvm-unit-tests/kvm-unit-tests!43

arm64: Use the provided fdt when booting through EFI

This patch uses the fdt pointer provided through efi_bootinfo_t for
tests that run as EFI apps. As in Linux, we give priority to the fdt.
First,

arm64: Use the provided fdt when booting through EFI

This patch uses the fdt pointer provided through efi_bootinfo_t for
tests that run as EFI apps. As in Linux, we give priority to the fdt.
First, we check if the pointer to the fdt is set and only if it isn't
we fallback to using the ACPI.

In addition, this patches changes the efi run script to generate and
use a fdt unless the user has set the enviroment variable EFI_USE_ACPI
to 'y'.

As a result:

$> ./arm/efi/run ./arm/selftest.efi -append "setup smp=2 mem=256" -smp 2 -m 256

will use an fdt, where as

$> EFI_USE_ACPI=y ./arm/efi/run ./arm/selftest.efi -append "setup smp=2 mem=256" -smp 2 -m 256

will run relying on ACPI.

Signed-off-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Signed-off-by: Andrew Jones <andrew.jones@linux.dev>

show more ...

arm64: Add a setup sequence for systems that boot through EFI

This change implements an alternative setup sequence for the system
when we are booting through EFI. The memory map is discovered throug

arm64: Add a setup sequence for systems that boot through EFI

This change implements an alternative setup sequence for the system
when we are booting through EFI. The memory map is discovered through
EFI boot services and devices through ACPI.

This change is based on a change initially proposed by
Andrew Jones <drjones@redhat.com>

Signed-off-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-by: Shaoqin Huang <shahuang@redhat.com>
[Changed __ALIGN to ALIGN as pointed out by Nadav Amit.]
Signed-off-by: Andrew Jones <andrew.jones@linux.dev>

show more ...

arm/arm64: Rename etext to _etext

Rename etext to the more popular _etext allowing different linker
scripts to more easily be used.

Signed-off-by: Andrew Jones <drjones@redhat.com>
Signed-off-by: N

arm/arm64: Rename etext to _etext

Rename etext to the more popular _etext allowing different linker
scripts to more easily be used.

Signed-off-by: Andrew Jones <drjones@redhat.com>
Signed-off-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-by: Ricardo Koller <ricarkol@google.com>
Signed-off-by: Andrew Jones <andrew.jones@linux.dev>

show more ...

arm64: Add support for cpu initialization through ACPI

In systems with ACPI support and when a DT is not provided, we can use
the MADTs to discover the number of CPUs and their corresponding MIDR.
T

arm64: Add support for cpu initialization through ACPI

In systems with ACPI support and when a DT is not provided, we can use
the MADTs to discover the number of CPUs and their corresponding MIDR.
This change implements this but retains the default behavior; we check
if a valid DT is provided, if not, we try to discover the cores in the
system using ACPI.

Signed-off-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Shaoqin Huang <shahuang@redhat.com>
Signed-off-by: Andrew Jones <andrew.jones@linux.dev>

show more ...

arm64: Add support for timer initialization through ACPI

For systems with ACPI support, we can discover timers through the ACPI
GTDT table. This change implements the code to discover timers through

arm64: Add support for timer initialization through ACPI

For systems with ACPI support, we can discover timers through the ACPI
GTDT table. This change implements the code to discover timers through
the GTDT and adds ACPI support in timer_save_state. This change
retains the default behavior; we check if a valid DT is provided, if
not, we try to discover timers using ACPI.

Signed-off-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Signed-off-by: Andrew Jones <andrew.jones@linux.dev>

show more ...

Merge branch 'arm/queue' into 'master'

arm/arm64: target-efi prep

This series mostly prepares kvm-unit-tests/arm for targeting EFI
platforms. The actually EFI support will come in another series,
b

Merge branch 'arm/queue' into 'master'

arm/arm64: target-efi prep

This series mostly prepares kvm-unit-tests/arm for targeting EFI
platforms. The actually EFI support will come in another series,
but these patches are good for removing assumptions from our memory
maps and about our PSCI conduit, even if we never merge EFI support.

See merge request kvm-unit-tests/kvm-unit-tests!8

show more ...

arm/arm64: psci: Don't assume method is hvc

The method can be smc in addition to hvc, and it will be when running
on bare metal. Additionally, we move the invocations to assembly so
we don't have to

arm/arm64: psci: Don't assume method is hvc

The method can be smc in addition to hvc, and it will be when running
on bare metal. Additionally, we move the invocations to assembly so
we don't have to rely on compiler assumptions. We also fix the
prototype of psci_invoke. function_id should be an unsigned int, not
an unsigned long.

Reviewed-by: Alexandru Elisei <alexandru.elisei@arm.com>
Tested-by: Alexandru Elisei <alexandru.elisei@arm.com>
Signed-off-by: Andrew Jones <drjones@redhat.com>

show more ...

arm/arm64: setup: Consolidate memory layout assumptions

Keep as much memory layout assumptions as possible in init::start
and a single setup function. This prepares us for calling setup()
from diffe

arm/arm64: setup: Consolidate memory layout assumptions

Keep as much memory layout assumptions as possible in init::start
and a single setup function. This prepares us for calling setup()
from different start functions which have been linked with different
linker scripts. To do this, stacktop is only referenced from
init::start, making freemem_start a parameter to setup(). We also
split mem_init() into three parts, one that populates the mem regions
per the DT, one that populates the mem regions per assumptions,
and one that does the mem init. The concept of a primary region
is dropped, but we add a sanity check for the absence of memory
holes, because we don't know how to deal with them yet.

Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-by: Alexandru Elisei <alexandru.elisei@arm.com>
Tested-by: Alexandru Elisei <alexandru.elisei@arm.com>
Signed-off-by: Andrew Jones <drjones@redhat.com>

show more ...

arm/arm64: mmu: Remove memory layout assumptions

Rather than making too many assumptions about the memory layout
in mmu code, just set up the page tables per the memory regions
(which means putting

arm/arm64: mmu: Remove memory layout assumptions

Rather than making too many assumptions about the memory layout
in mmu code, just set up the page tables per the memory regions
(which means putting all the memory layout assumptions in setup).
To ensure we get the right default flags set we need to split the
primary region into two regions for code and data.

We still only expect the primary regions to be present, but the
next patch will remove that assumption too.

(Unfortunately we still have an assumption in setup_mmu. We assume
the range 3G-4G is available for the virtual memory allocator. We'll
need to remove that assumption as well with another patch in order
to support arbitrary memory maps.)

Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-by: Alexandru Elisei <alexandru.elisei@arm.com>
Tested-by: Alexandru Elisei <alexandru.elisei@arm.com>
Signed-off-by: Andrew Jones <drjones@redhat.com>

show more ...

arm/arm64: Move setup_vm into setup

Consolidate our setup calls to reduce the amount we need to do from
init::start. Also remove a couple of pointless comments from setup().

Reviewed-by: Nikos Niko

arm/arm64: Move setup_vm into setup

Consolidate our setup calls to reduce the amount we need to do from
init::start. Also remove a couple of pointless comments from setup().

Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-by: Alexandru Elisei <alexandru.elisei@arm.com>
Tested-by: Alexandru Elisei <alexandru.elisei@arm.com>
Signed-off-by: Andrew Jones <drjones@redhat.com>

show more ...

Merge branch 'arm/queue' into 'master'

arm/arm64: Fixes, improvements, and prep for target-efi

See merge request kvm-unit-tests/kvm-unit-tests!6