All architectures define pte_index() as
(address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1)
and all architectures define pte_offset_kernel() as an entry in the array
of PTEs indexed by the pte_index().
For the most architectures the pte_offset_kernel() implementation relies
on the availability of pmd_page_vaddr() that converts a PMD entry value to
the virtual address of the page containing PTEs array.
Let's move x86 definitions of the PTE accessors to the generic place in
<linux/pgtable.h> and then simply drop the respective definitions from the
other architectures.
The architectures that didn't provide pmd_page_vaddr() are updated to have
that defined.
The generic implementation of pte_offset_kernel() can be overridden by an
architecture and alpha makes use of this because it has special ordering
requirements for its version of pte_offset_kernel().
[rppt@linux.ibm.com: v2]
Link: http://lkml.kernel.org/r/20200514170327.31389-11-rppt@kernel.org
[rppt@linux.ibm.com: update]
Link: http://lkml.kernel.org/r/20200514170327.31389-12-rppt@kernel.org
[rppt@linux.ibm.com: update]
Link: http://lkml.kernel.org/r/20200514170327.31389-13-rppt@kernel.org
[akpm@linux-foundation.org: fix x86 warning]
[sfr@canb.auug.org.au: fix powerpc build]
Link: http://lkml.kernel.org/r/20200607153443.GB738695@linux.ibm.com
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Cain <bcain@codeaurora.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Chris Zankel <chris@zankel.net>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Greentime Hu <green.hu@gmail.com>
Cc: Greg Ungerer <gerg@linux-m68k.org>
Cc: Guan Xuetao <gxt@pku.edu.cn>
Cc: Guo Ren <guoren@kernel.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Helge Deller <deller@gmx.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Ley Foon Tan <ley.foon.tan@intel.com>
Cc: Mark Salter <msalter@redhat.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Matt Turner <mattst88@gmail.com>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Michal Simek <monstr@monstr.eu>
Cc: Nick Hu <nickhu@andestech.com>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Richard Weinberger <richard@nod.at>
Cc: Rich Felker <dalias@libc.org>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Stafford Horne <shorne@gmail.com>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Vincent Chen <deanbo422@gmail.com>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Link: http://lkml.kernel.org/r/20200514170327.31389-10-rppt@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 31168f033e ("mips: drop __pXd_offset() macros that duplicate
pXd_index() ones") is correct that pud_index() & __pud_offset() are the
same when pud_index() is actually provided, however it does not take into
account the __PAGETABLE_PUD_FOLDED case. This has broken MIPS KVM
compilation because it relied on availability of pud_index().
Define pud_index() regardless of page table folded. It will evaluate to
actual index for 4-level pagetables and to 0 for folded PUD level.
Link: https://lore.kernel.org/lkml/20200331154749.5457-1-pbonzini@redhat.com
Reported-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Tested-by: Huacai Chen <chenhc@lemote.com>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Implement primitives necessary for the 4th level folding, add walks of p4d
level where appropriate, replace 5leve-fixup.h with pgtable-nop4d.h and
drop usage of __ARCH_USE_5LEVEL_HACK.
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Signed-off-by: Paul Burton <paulburton@kernel.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: James Hogan <jhogan@kernel.org>
Cc: linux-mips@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-mm@kvack.org
Cc: Mike Rapoport <rppt@kernel.org>
The __pXd_offset() macros are identical to the pXd_index() macros and there
is no point to keep both of them. All architectures define and use
pXd_index() so let's keep only those to make mips consistent with the rest
of the kernel.
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Signed-off-by: Paul Burton <paulburton@kernel.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: James Hogan <jhogan@kernel.org>
Cc: linux-mips@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-mm@kvack.org
Cc: Mike Rapoport <rppt@kernel.org>
With CONFIG_MIPS_VA_BITS_48=y the build fails miserably:
CC arch/mips/kernel/asm-offsets.s
In file included from arch/mips/include/asm/pgtable.h:644,
from include/linux/mm.h:99,
from arch/mips/kernel/asm-offsets.c:15:
include/asm-generic/pgtable.h:16:2: error: #error CONFIG_PGTABLE_LEVELS is not consistent with __PAGETABLE_{P4D,PUD,PMD}_FOLDED
#error CONFIG_PGTABLE_LEVELS is not consistent with __PAGETABLE_{P4D,PUD,PMD}_FOLDED
^~~~~
include/asm-generic/pgtable.h:390:28: error: unknown type name 'p4d_t'; did you mean 'pmd_t'?
static inline int p4d_same(p4d_t p4d_a, p4d_t p4d_b)
^~~~~
pmd_t
[ ... more such errors ... ]
scripts/Makefile.build:99: recipe for target 'arch/mips/kernel/asm-offsets.s' failed
make[2]: *** [arch/mips/kernel/asm-offsets.s] Error 1
This happens because when CONFIG_MIPS_VA_BITS_48 enables 4th level of the
page tables, but neither pgtable-nop4d.h nor 5level-fixup.h are included to
cope with the 5th level.
Replace #ifdef conditions around includes of the pgtable-nop{m,u}d.h with
explicit CONFIG_PGTABLE_LEVELS and add include of 5level-fixup.h for the
case when CONFIG_PGTABLE_LEVELS==4
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Signed-off-by: Paul Burton <paulburton@kernel.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: James Hogan <jhogan@kernel.org>
Cc: linux-mips@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-mm@kvack.org
Cc: Mike Rapoport <rppt@kernel.org>
This patch is borrowed from ARM64 to ensure pmd_present() returns false
after pmd_mknotpresent(). This is needed for THP.
References: 5bb1cc0ff9 ("arm64: Ensure pmd_present() returns false after pmd_mknotpresent()")
Reviewed-by: James Hogan <jhogan@kernel.org>
Signed-off-by: Huacai Chen <chenhc@lemote.com>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Patchwork: https://patchwork.linux-mips.org/patch/21135/
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: James Hogan <james.hogan@mips.com>
Cc: Steven J . Hill <Steven.Hill@cavium.com>
Cc: linux-mips@linux-mips.org
Cc: Fuxin Zhang <zhangfx@lemote.com>
Cc: Zhangjin Wu <wuzhangjin@gmail.com>
Cc: <stable@vger.kernel.org> # 3.8+
empty_bad_page() and empty_bad_pte_table() seem to be relics from old
days which is not used by any code for a long time. I have tried to
find when exactly but this is not really all that straightforward due to
many code movements - traces disappear around 2.4 times.
Anyway no code really references neither empty_bad_page nor
empty_bad_pte_table. We only allocate the storage which is not used by
anybody so remove them.
Link: http://lkml.kernel.org/r/20171004150045.30755-1-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Ralf Baechle <ralf@linus-mips.org>
Acked-by: Ingo Molnar <mingo@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: David Howells <dhowells@redhat.com>
Cc: Rich Felker <dalias@libc.org>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Richard Weinberger <richard@nod.at>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Some users must have 4K pages while needing a 48-bit VA space size.
The cleanest way do do this is to go to a 4-level page table for this
case. Each page table level using order-0 pages adds 9 bits to the
VA size (at 4K pages, so for four levels we get 9 * 4 + 12 == 48-bits.
For the 4K page size case only we add support functions for the PUD
level of the page table tree, also the TLB exception handlers get an
extra level of tree walk.
[david.daney@cavium.com: Forward port to v4.10.]
[david.daney@cavium.com: Forward port to v4.11.]
Signed-off-by: Alex Belits <alex.belits@cavium.com>
Signed-off-by: David Daney <david.daney@cavium.com>
Cc: James Hogan <james.hogan@imgtec.com>
Cc: Alex Belits <alex.belits@cavium.com>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/15312/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
If an architecture uses 4level-fixup.h we don't need to do anything as
it includes 5level-fixup.h.
If an architecture uses pgtable-nop*d.h, define __ARCH_USE_5LEVEL_HACK
before inclusion of the header. It makes asm-generic code to use
5level-fixup.h.
If an architecture has 4-level paging or folds levels on its own,
include 5level-fixup.h directly.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
With CONFIG_MIGRATION, the PFN of the migrating pages is stored in
__swp_offset(), so we must have enough bits to store the largest
possible PFN. OCTEON NUMA systems have 41 bits of physical address
space, so with 4K pages (12-bits), we need at least 29 bits to store
the PFN.
The current width of 24-bits is too narrow, so expand it all the way
out to 40-bits. This leaves the low order 16 bits as zero which does
not interfere with any of the PTE bits.
Signed-off-by: David Daney <david.daney@cavium.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/9315/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Having received another series of whitespace patches I decided to do this
once and for all rather than dealing with this kind of patches trickling
in forever.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
With CONFIG_TRANSPARENT_HUGEPAGE=y and CONFIG_HUGETLBFS=y we get the
following build failure:
CC mm/huge_memory.o
mm/huge_memory.c: In function 'set_huge_zero_page':
mm/huge_memory.c:780:2: error: implicit declaration of function 'pfn_pmd' [-Werror=implicit-function-declaration]
mm/huge_memory.c:780:8: error: incompatible types when assigning to type 'pmd_t' from type 'int'
Add a definition of pfn_pmd() for 64-bit kernels (the only place huge
pages are currently supported).
Signed-off-by: David Daney <david.daney@cavium.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/4813/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
The TASK_SIZE macro should reflect the size of a user process virtual
address space. Previously for 64-bit kernels, this was not the case.
The immediate cause of pain was in
hugetlbfs/inode.c:hugetlb_get_unmapped_area() where 32-bit processes
trying to mmap a huge page would be served a page with an address
outside of the 32-bit address range. But there are other uses of
TASK_SIZE in the kernel as well that would like an accurate value.
The new definition is nice because it now makes TASK_SIZE and
TASK_SIZE_OF() yield the same value for any given process.
For 32-bit kernels there should be no change, although I did factor
out some code in asm/processor.h that became identical for the 32-bit and
64-bit cases.
__UA_LIMIT is now set to ~((1 << SEGBITS) - 1) for 64-bit kernels.
This should eliminate the possibility of getting a
AddressErrorException in the kernel for addresses that pass the
access_ok() test.
With the patch applied, I can still run o32, n32 and n64 processes,
and have an o32 shell fork/exec both n32 and n64 processes.
Signed-off-by: David Daney <ddaney@caviumnetworks.com>
To: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/1701/
Since we no longer need to provide KM_type, the whole pte_*map_nested()
API is now redundant, remove it.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Chris Metcalf <cmetcalf@tilera.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: David Miller <davem@davemloft.net>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In the 64-bit kernel we use swapper_pg_dir for three different things.
1) xuseg mappings for kernel threads.
2) vmap mappings for all kernel-space accesses in xkseg.
3) vmap mappings for kernel modules in ksseg (kseg2).
Due to how the TLB refill handlers work, any mapping established in
xkseg or ksseg will also establish a xuseg mapping that should never
be used by the kernel.
In order to be able to use exceptions to trap NULL pointer
dereferences, we need to ensure that nothing is mapped at address
zero. Since vmap mappings in xkseg are reflected in xuseg, this means
we need to ensure that there are no vmap mappings established at the
start of xkseg. So we move back VMALLOC_START to avoid establishing
vmap mappings at the start of xkseg.
Signed-off-by: David Daney <ddaney@caviumnetworks.com>
To: linux-mips@linux-mips.org
Patchwork: http://patchwork.linux-mips.org/patch/1129/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
The SmartMIPS ASE specifies how Read Inhibit (RI) and eXecute Inhibit
(XI) bits in the page tables work. The upper two bits of EntryLo{0,1}
are RI and XI when the feature is enabled in the PageGrain register.
SmartMIPS only covers 32-bit systems. Cavium Octeon+ extends this to
64-bit systems by continuing to place the RI and XI bits in the top of
EntryLo even when EntryLo is 64-bits wide.
Because we need to carry the RI and XI bits in the PTE, the layout of
the PTE is changed. There is a two instruction overhead in the TLB
refill hot path to get the EntryLo bits into the proper position.
Also the TLB load exception has to probe the TLB to check if RI or XI
caused the exception.
Also of note is that the layout of the PTE bits is done at compile and
runtime rather than statically. In the 32-bit case this allows for
the same number of PFN bits as before the patch as the _PAGE_HUGE is
not supported in 32-bit kernels (we have _PAGE_NO_EXEC and
_PAGE_NO_READ instead of _PAGE_READ and _PAGE_HUGE).
The patch is tested on Cavium Octeon+, but should also work on 32-bit
systems with the Smart-MIPS ASE.
Signed-off-by: David Daney <ddaney@caviumnetworks.com>
To: linux-mips@linux-mips.org
Patchwork: http://patchwork.linux-mips.org/patch/952/
Patchwork: http://patchwork.linux-mips.org/patch/956/
Patchwork: http://patchwork.linux-mips.org/patch/962/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
For 64-bit kernels with 64KB pages and two level page tables, there are
42 bits worth of virtual address space This is larger than the 40 bits of
virtual address space obtained with the default 4KB Page size and three
levels, so there are no draw backs for using two level tables with this
configuration.
Signed-off-by: David Daney <ddaney@caviumnetworks.com>
Cc: linux-mips@linux-mips.org
Patchwork: http://patchwork.linux-mips.org/patch/761/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Linux kernel 2.6.32 and later allocate address space from the top of the
kernel virtual memory address space.
This patch implements virtual memory size detection for 64 bit MIPS CPUs
to avoid resulting crashes.
Signed-off-by: Guenter Roeck <guenter.roeck@ericsson.com>
Cc: linux-mips@linux-mips.org
Patchwork: http://patchwork.linux-mips.org/patch/935/
Reviewed-by: David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
By combining swapper_pg_dir and module_pg_dir, several if conditions
can be eliminated from the tlb exception handler. The reason they
can be combined is that, the effective virtual address of vmalloc
returned is at the bottom, and of module_alloc returned is at the
top. It also fixes the bug in vmalloc(), which happens when its
return address is not covered by the first pgd.
Signed-off-by: Wu Fei <at.wufei@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Beyond the requirements of the architecture standard Cavium also supports
8k and 32k pages.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Acked-by: David Daney <ddaney@caviumnetworks.com>
