[PATCH 1/5] x86: Enable 5-level paging support
From: Kirill A. Shutemov
Date: Thu Jun 22 2017 - 08:26:58 EST
Most of things are in place and we can enable support of 5-level paging.
The patch makes XEN_PV dependent on !X86_5LEVEL. XEN_PV is not ready to
work with 5-level paging.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx>
Reviewed-by: Juergen Gross <jgross@xxxxxxxx>
Documentation/x86/x86_64/5level-paging.txt | 64 ++++++++++++++++++++++++++++++
arch/x86/Kconfig | 18 +++++++++
arch/x86/xen/Kconfig | 3 ++
3 files changed, 85 insertions(+)
create mode 100644 Documentation/x86/x86_64/5level-paging.txt
diff --git a/Documentation/x86/x86_64/5level-paging.txt b/Documentation/x86/x86_64/5level-paging.txt
new file mode 100644
@@ -0,0 +1,64 @@
+== Overview ==
+Original x86-64 was limited by 4-level paing to 256 TiB of virtual address
+space and 64 TiB of physical address space. We are already bumping into
+this limit: some vendors offers servers with 64 TiB of memory today.
+To overcome the limitation upcoming hardware will introduce support for
+5-level paging. It is a straight-forward extension of the current page
+table structure adding one more layer of translation.
+It bumps the limits to 128 PiB of virtual address space and 4 PiB of
+physical address space. This "ought to be enough for anybody" Â.
+QEMU 2.9 and later support 5-level paging.
+Virtual memory layout for 5-level paging is described in
+== Enabling 5-level paging ==
+CONFIG_X86_5LEVEL=y enables the feature.
+So far, a kernel compiled with the option enabled will be able to boot
+only on machines that supports the feature -- see for 'la57' flag in
+The plan is to implement boot-time switching between 4- and 5-level paging
+in the future.
+== User-space and large virtual address space ==
+On x86, 5-level paging enables 56-bit userspace virtual address space.
+Not all user space is ready to handle wide addresses. It's known that
+at least some JIT compilers use higher bits in pointers to encode their
+information. It collides with valid pointers with 5-level paging and
+leads to crashes.
+To mitigate this, we are not going to allocate virtual address space
+above 47-bit by default.
+But userspace can ask for allocation from full address space by
+specifying hint address (with or without MAP_FIXED) above 47-bits.
+If hint address set above 47-bit, but MAP_FIXED is not specified, we try
+to look for unmapped area by specified address. If it's already
+occupied, we look for unmapped area in *full* address space, rather than
+from 47-bit window.
+A high hint address would only affect the allocation in question, but not
+any future mmap()s.
+Specifying high hint address on older kernel or on machine without 5-level
+paging support is safe. The hint will be ignored and kernel will fall back
+to allocation from 47-bit address space.
+This approach helps to easily make application's memory allocator aware
+about large address space without manually tracking allocated virtual
+One important case we need to handle here is interaction with MPX.
+MPX (without MAWA extension) cannot handle addresses above 47-bit, so we
+need to make sure that MPX cannot be enabled we already have VMA above
+the boundary and forbid creating such VMAs once MPX is enabled.
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 72028a16327b..dc91c1763736 100644
@@ -320,6 +320,7 @@ config FIX_EARLYCON_MEM
+ default 5 if X86_5LEVEL
default 4 if X86_64
default 3 if X86_PAE
@@ -1392,6 +1393,23 @@ config X86_PAE
has the cost of more pagetable lookup overhead, and also
consumes more pagetable space per process.
+ bool "Enable 5-level page tables support"
+ depends on X86_64
+ 5-level paging enables access to larger address space:
+ upto 128 PiB of virtual address space and 4 PiB of
+ physical address space.
+ It will be supported by future Intel CPUs.
+ Note: kernel with the option enabled can only be booted
+ on machines that support the feature.
+ See Documentation/x86/x86_64/5level-paging.txt for more info.
+ Say N if unsure.
depends on X86_64 || X86_PAE
diff --git a/arch/x86/xen/Kconfig b/arch/x86/xen/Kconfig
index 027987638e98..72bd4c62b742 100644
@@ -17,6 +17,9 @@ config XEN_PV
bool "Xen PV guest support"
depends on XEN
+ # XEN_PV is not ready to work with 5-level paging.
+ # Changes to hypervisor are also required.
+ depends on !X86_5LEVEL