Currently, the KVM page-table allocator uses a mix of put_page() and
free_page() calls depending on the context even though page-allocation
is always achieved using variants of __get_free_page().
Make the code consistent by using put_page() throughout, and reduce the
memory management API surface used by the page-table code. This will
ease factoring out page-allocation from pgtable.c, which is a
pre-requisite to creating page-tables at EL2.
Acked-by: Will Deacon <will@kernel.org>
Signed-off-by: Quentin Perret <qperret@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20210319100146.1149909-6-qperret@google.com
When walking the page tables at a given level, and if the start
address for the range isn't aligned for that level, we propagate
the misalignment on each iteration at that level.
This results in the walker ignoring a number of entries (depending
on the original misalignment) on each subsequent iteration.
Properly aligning the address before the next iteration addresses
this issue.
Cc: stable@vger.kernel.org
Reported-by: Howard Zhang <Howard.Zhang@arm.com>
Acked-by: Will Deacon <will@kernel.org>
Signed-off-by: Jia He <justin.he@arm.com>
Fixes: b1e57de62c ("KVM: arm64: Add stand-alone page-table walker infrastructure")
[maz: rewrite commit message]
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20210303024225.2591-1-justin.he@arm.com
Message-Id: <20210305185254.3730990-9-maz@kernel.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
(1) During running time of a a VM with numbers of vCPUs, if some vCPUs
access the same GPA almost at the same time and the stage-2 mapping of
the GPA has not been built yet, as a result they will all cause
translation faults. The first vCPU builds the mapping, and the followed
ones end up updating the valid leaf PTE. Note that these vCPUs might
want different access permissions (RO, RW, RX, RWX, etc.).
(2) It's inevitable that we sometimes will update an existing valid leaf
PTE in the map path, and we perform break-before-make in this case.
Then more unnecessary translation faults could be caused if the
*break stage* of BBM is just catched by other vCPUS.
With (1) and (2), something unsatisfactory could happen: vCPU A causes
a translation fault and builds the mapping with RW permissions, vCPU B
then update the valid leaf PTE with break-before-make and permissions
are updated back to RO. Besides, *break stage* of BBM may trigger more
translation faults. Finally, some useless small loops could occur.
We can make some optimization to solve above problems: When we need to
update a valid leaf PTE in the map path, let's filter out the case where
this update only change access permissions, and don't update the valid
leaf PTE here in this case. Instead, let the vCPU enter back the guest
and it will exit next time to go through the relax_perms path without
break-before-make if it still wants more permissions.
Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20210114121350.123684-3-wangyanan55@huawei.com
Procedures of hyp stage-1 map and guest stage-2 map are quite different,
but they are tied closely by function kvm_set_valid_leaf_pte().
So adjust the relative code for ease of code maintenance in the future.
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20210114121350.123684-2-wangyanan55@huawei.com
When dirty logging is enabled, we collapse block entries into tables
as necessary. If dirty logging gets canceled, we can end-up merging
tables back into block entries.
When this happens, we must not only free the non-huge page-table
pages but also invalidate all the TLB entries that can potentially
cover the block. Otherwise, we end-up with multiple possible translations
for the same physical page, which can legitimately result in a TLB
conflict.
To address this, replease the bogus invalidation by IPA with a full
VM invalidation. Although this is pretty heavy handed, it happens
very infrequently and saves a bunch of invalidations by IPA.
Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
[maz: fixup commit message]
Signed-off-by: Marc Zyngier <maz@kernel.org>
Acked-by: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201201201034.116760-3-wangyanan55@huawei.com
When installing a new leaf PTE onto an invalid ptep, we need to
get_page(ptep) to account for the new mapping.
However, simply updating a valid PTE shouldn't result in any
additional refcounting, as there is new mapping. This otherwise
results in a page being forever wasted.
Address this by fixing-up the refcount in stage2_map_walker_try_leaf()
if the PTE was already valid, balancing out the later get_page()
in stage2_map_walk_leaf().
Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
[maz: update commit message, add comment in the code]
Signed-off-by: Marc Zyngier <maz@kernel.org>
Acked-by: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20201201201034.116760-2-wangyanan55@huawei.com
stage2_pte_cacheable() tries to figure out whether the mapping installed
in its 'pte' parameter is cacheable or not. Unfortunately, it fails
miserably because it extracts the memory attributes from the entry using
FIELD_GET(), which returns the attributes shifted down to bit 0, but then
compares this with the unshifted value generated by the PAGE_S2_MEMATTR()
macro.
A direct consequence of this bug is that cache maintenance is silently
skipped, which in turn causes 32-bit guests to crash early on when their
set/way maintenance is trapped but not emulated correctly.
Fix the broken masks by avoiding the use of FIELD_GET() altogether.
Fixes: 6d9d2115c4 ("KVM: arm64: Add support for stage-2 map()/unmap() in generic page-table")
Reported-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Cc: Quentin Perret <qperret@google.com>
Link: https://lore.kernel.org/r/20201029144716.30476-1-will@kernel.org
For consistency with the rest of the stage-2 page-table page allocations
(performing using a kvm_mmu_memory_cache), ensure that __GFP_ACCOUNT is
included in the GFP flags for the PGD pages.
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Gavin Shan <gshan@redhat.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Quentin Perret <qperret@google.com>
Link: https://lore.kernel.org/r/20201026144423.24683-1-will@kernel.org
Alex pointed out that we don't pass a level hint to the TLBI instruction
when handling a stage-2 permission fault, even though the walker does
at some point have the level information in its hands.
Rework stage2_update_leaf_attrs() so that it can optionally return the
level of the updated pte to its caller, which can in turn be used to
provide the correct TLBI level hint.
Reported-by: Alexandru Elisei <alexandru.elisei@arm.com>
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Alexandru Elisei <alexandru.elisei@arm.com>
Reviewed-by: Gavin Shan <gshan@redhat.com>
Cc: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/595cc73e-636e-8b3a-f93a-b4e9fb218db8@arm.com
Link: https://lore.kernel.org/r/20200930131801.16889-1-will@kernel.org
Add support for relaxing the permissions of a stage-2 mapping (i.e.
adding additional permissions) to the generic page-table code.
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Gavin Shan <gshan@redhat.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Quentin Perret <qperret@google.com>
Link: https://lore.kernel.org/r/20200911132529.19844-17-will@kernel.org
Add support for cache flushing a range of the stage-2 address space to
the generic page-table code.
Signed-off-by: Quentin Perret <qperret@google.com>
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Gavin Shan <gshan@redhat.com>
Cc: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20200911132529.19844-15-will@kernel.org
Introduce alloc() and free() functions to the generic page-table code
for guest stage-2 page-tables and plumb these into the existing KVM
page-table allocator. Subsequent patches will convert other operations
within the KVM allocator over to the generic code.
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Gavin Shan <gshan@redhat.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Quentin Perret <qperret@google.com>
Link: https://lore.kernel.org/r/20200911132529.19844-6-will@kernel.org
The generic page-table walker is pretty useless as it stands, because it
doesn't understand enough to allocate anything. Teach it about stage-1
page-tables, and hook up an API for allocating these for the hypervisor
at EL2.
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Gavin Shan <gshan@redhat.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Quentin Perret <qperret@google.com>
Link: https://lore.kernel.org/r/20200911132529.19844-4-will@kernel.org
The KVM page-table code is intricately tied into the kernel page-table
code and re-uses the pte/pmd/pud/p4d/pgd macros directly in an attempt
to reduce code duplication. Unfortunately, the reality is that there is
an awful lot of code required to make this work, and at the end of the
day you're limited to creating page-tables with the same configuration
as the host kernel. Furthermore, lifting the page-table code to run
directly at EL2 on a non-VHE system (as we plan to to do in future
patches) is practically impossible due to the number of dependencies it
has on the core kernel.
Introduce a framework for walking Armv8 page-tables configured
independently from the host kernel.
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Reviewed-by: Gavin Shan <gshan@redhat.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Quentin Perret <qperret@google.com>
Link: https://lore.kernel.org/r/20200911132529.19844-3-will@kernel.org
