After removal of CONFIG_HAVE_MEMBLOCK_NODE_MAP we have two equivalent
functions that call memory_present() for each region in memblock.memory:
sparse_memory_present_with_active_regions() and membocks_present().
Moreover, all architectures have a call to either of these functions
preceding the call to sparse_init() and in the most cases they are called
one after the other.
Mark the regions from memblock.memory as present during sparce_init() by
making sparse_init() call memblocks_present(), make memblocks_present()
and memory_present() functions static and remove redundant
sparse_memory_present_with_active_regions() function.
Also remove no longer required HAVE_MEMORY_PRESENT configuration option.
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20200712083130.22919-1-rppt@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "mm: cleanup usage of <asm/pgalloc.h>"
Most architectures have very similar versions of pXd_alloc_one() and
pXd_free_one() for intermediate levels of page table. These patches add
generic versions of these functions in <asm-generic/pgalloc.h> and enable
use of the generic functions where appropriate.
In addition, functions declared and defined in <asm/pgalloc.h> headers are
used mostly by core mm and early mm initialization in arch and there is no
actual reason to have the <asm/pgalloc.h> included all over the place.
The first patch in this series removes unneeded includes of
<asm/pgalloc.h>
In the end it didn't work out as neatly as I hoped and moving
pXd_alloc_track() definitions to <asm-generic/pgalloc.h> would require
unnecessary changes to arches that have custom page table allocations, so
I've decided to move lib/ioremap.c to mm/ and make pgalloc-track.h local
to mm/.
This patch (of 8):
In most cases <asm/pgalloc.h> header is required only for allocations of
page table memory. Most of the .c files that include that header do not
use symbols declared in <asm/pgalloc.h> and do not require that header.
As for the other header files that used to include <asm/pgalloc.h>, it is
possible to move that include into the .c file that actually uses symbols
from <asm/pgalloc.h> and drop the include from the header file.
The process was somewhat automated using
sed -i -E '/[<"]asm\/pgalloc\.h/d' \
$(grep -L -w -f /tmp/xx \
$(git grep -E -l '[<"]asm/pgalloc\.h'))
where /tmp/xx contains all the symbols defined in
arch/*/include/asm/pgalloc.h.
[rppt@linux.ibm.com: fix powerpc warning]
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Pekka Enberg <penberg@kernel.org>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> [m68k]
Cc: Abdul Haleem <abdhalee@linux.vnet.ibm.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Joerg Roedel <joro@8bytes.org>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Satheesh Rajendran <sathnaga@linux.vnet.ibm.com>
Cc: Stafford Horne <shorne@gmail.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Joerg Roedel <jroedel@suse.de>
Cc: Matthew Wilcox <willy@infradead.org>
Link: http://lkml.kernel.org/r/20200627143453.31835-1-rppt@kernel.org
Link: http://lkml.kernel.org/r/20200627143453.31835-2-rppt@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull ARM updates from Russell King:
- add arch/arm/Kbuild from Masahiro Yamada.
- simplify act_mm macro, since it contains an open-coded
get_thread_info.
- VFP updates for Clang from Stefan Agner.
- Fix unwinder for Clang from Nathan Huckleberry.
- Remove unused it8152 PCI host controller, used by the removed cm-x2xx
platforms from Mike Rapoport.
- Further explanation of __range_ok().
- Remove kimage_voffset that isn't used anymore from Marc Zyngier.
- Drop ancient Thumb-2 workaround for old binutils from Ard Biesheuvel.
- Documentation cleanup for mach-* from Pete Zaitcev.
* tag 'for-linus' of git://git.armlinux.org.uk/~rmk/linux-arm:
ARM: 8996/1: Documentation/Clean up the description of mach-<class>
ARM: 8995/1: drop Thumb-2 workaround for ancient binutils
ARM: 8994/1: mm: drop kimage_voffset which was only used by KVM
ARM: uaccess: add further explanation of __range_ok()
ARM: 8993/1: remove it8152 PCI controller driver
ARM: 8992/1: Fix unwind_frame for clang-built kernels
ARM: 8991/1: use VFP assembler mnemonics if available
ARM: 8990/1: use VFP assembler mnemonics in register load/store macros
ARM: 8989/1: use .fpu assembler directives instead of assembler arguments
ARM: 8982/1: mm: Simplify act_mm macro
ARM: 8981/1: add arch/arm/Kbuild
Now that KVM support has been removed from the 32-bit ARM port,
drop the export kimage_voffset symbol, which no longer has any
users.
Acked-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Commit
84e6ffb2c4 ("arm: add support for folded p4d page tables")
updated create_mapping_late() to take folded P4Ds into account when
creating mappings, but inverted the p4d_alloc() failure test, resulting
in no mapping to be created at all.
When the EFI rtc driver subsequently tries to invoke the EFI GetTime()
service, the memory regions covering the EFI data structures are missing
from the page tables, resulting in a crash like
Unable to handle kernel paging request at virtual address 5ae0cf28
pgd = (ptrval)
[5ae0cf28] *pgd=80000040205003, *pmd=00000000
Internal error: Oops: 207 [#1] SMP THUMB2
Modules linked in:
CPU: 0 PID: 7 Comm: kworker/u32:0 Not tainted 5.7.0+ #92
Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
Workqueue: efi_rts_wq efi_call_rts
PC is at efi_call_rts+0x94/0x294
LR is at efi_call_rts+0x83/0x294
pc : [<c0b4f098>] lr : [<c0b4f087>] psr: 30000033
sp : e6219ef0 ip : 00000000 fp : ffffe000
r10: 00000000 r9 : 00000000 r8 : 30000013
r7 : e6201dd0 r6 : e6201ddc r5 : 00000000 r4 : c181f264
r3 : 5ae0cf10 r2 : 00000001 r1 : e6201dd0 r0 : e6201ddc
Flags: nzCV IRQs on FIQs on Mode SVC_32 ISA Thumb Segment none
Control: 70c5383d Table: 661cc840 DAC: 00000001
Process kworker/u32:0 (pid: 7, stack limit = 0x(ptrval))
...
[<c0b4f098>] (efi_call_rts) from [<c0448219>] (process_one_work+0x16d/0x3d8)
[<c0448219>] (process_one_work) from [<c0448581>] (worker_thread+0xfd/0x408)
[<c0448581>] (worker_thread) from [<c044ca7b>] (kthread+0x103/0x104)
...
Fixes: 84e6ffb2c4 ("arm: add support for folded p4d page tables")
Reviewed-by: Mike Rapoport <rppt@linux.ibm.com>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Better describe what this helper does, and match the naming of
copy_from_kernel_nofault.
Also switch the argument order around, so that it acts and looks
like get_user().
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The powerpc 32-bit implementation of pgtable has nice shortcuts for
accessing kernel PMD and PTE for a given virtual address. Make these
helpers available for all architectures.
[rppt@linux.ibm.com: microblaze: fix page table traversal in setup_rt_frame()]
Link: http://lkml.kernel.org/r/20200518191511.GD1118872@kernel.org
[akpm@linux-foundation.org: s/pmd_ptr_k/pmd_off_k/ in various powerpc places]
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
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/20200514170327.31389-9-rppt@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull ARM SoC updates from Arnd Bergmann:
"One new platform gets added, the Realtek RTD1195, which is an older
Cortex-a7 based relative of the RTD12xx chips that are already
supported in arch/arm64. The platform may also be extended to support
running 32-bit kernels on those 64-bit chips for memory-constrained
machines.
In the Renesas shmobile platform, we gain support for "RZ/G1H" or
R8A7742, an eight-core chip based on Cortex-A15 and Cortex-A7 cores,
originally released in 2016 as one of the last high-end 32-bit
designs.
There is ongoing cleanup for the integrator, tegra, imx, and omap2
platforms, with integrator getting very close to the goal of having
zero code in arch/arm/, and omap2 moving more of the chip specifics
from old board code into device tree files.
The Versatile Express platform is made more modular, with built-in
drivers now becoming loadable modules. This is part of a greater
effort for the Android OS to have a common kernel binary for all
platforms and any platform specific code in loadable modules.
The PXA platform drops support for Compulab's pxa2xx boards that had
rather unusual flash and PCI drivers but no known users remaining. All
device drivers specific to those boards can now get removed as well.
Across platforms, there is ongoing cleanup, with Geert and Rob
revisiting some a lot of Kconfig options"
* tag 'arm-soc-5.8' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (94 commits)
ARM: omap2: fix omap5_realtime_timer_init definition
ARM: zynq: Don't select CONFIG_ICST
ARM: OMAP2+: Fix regression for using local timer on non-SMP SoCs
clk: versatile: Fix kconfig dependency on COMMON_CLK_VERSATILE
ARM: davinci: fix build failure without I2C
power: reset: vexpress: fix build issue
power: vexpress: cleanup: use builtin_platform_driver
power: vexpress: add suppress_bind_attrs to true
Revert "ARM: vexpress: Don't select VEXPRESS_CONFIG"
MAINTAINERS: pxa: remove Compulab arm/pxa support
ARM: pxa: remove Compulab pxa2xx boards
bus: arm-integrator-lm: Fix return value check in integrator_ap_lm_probe()
soc: imx: move cpu code to drivers/soc/imx
ARM: imx: move cpu definitions into a header
ARM: imx: use device_initcall for imx_soc_device_init
ARM: imx: pcm037: make pcm970_sja1000_platform_data static
bus: ti-sysc: Timers no longer need legacy quirk handling
ARM: OMAP2+: Drop old timer code for dmtimer and 32k counter
ARM: dts: Configure system timers for omap2
ARM: dts: Configure system timers for ti81xx
...
Every single architecture (including !CONFIG_HIGHMEM) calls...
pagefault_enable();
preempt_enable();
... before returning from __kunmap_atomic(). Lift this code into the
kunmap_atomic() macro.
While we are at it rename __kunmap_atomic() to kunmap_atomic_high() to
be consistent.
[ira.weiny@intel.com: don't enable pagefault/preempt twice]
Link: http://lkml.kernel.org/r/20200518184843.3029640-1-ira.weiny@intel.com
[akpm@linux-foundation.org: coding style fixes]
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Christian König <christian.koenig@amd.com>
Cc: Chris Zankel <chris@zankel.net>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Helge Deller <deller@gmx.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Guenter Roeck <linux@roeck-us.net>
Link: http://lkml.kernel.org/r/20200507150004.1423069-8-ira.weiny@intel.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Every arch has the same code to ensure atomic operations and a check for
!HIGHMEM page.
Remove the duplicate code by defining a core kmap_atomic() which only
calls the arch specific kmap_atomic_high() when the page is high memory.
[akpm@linux-foundation.org: coding style fixes]
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Christian König <christian.koenig@amd.com>
Cc: Chris Zankel <chris@zankel.net>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Helge Deller <deller@gmx.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20200507150004.1423069-7-ira.weiny@intel.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
All architectures do exactly the same thing for kunmap(); remove all the
duplicate definitions and lift the call to the core.
This also has the benefit of changing kmap_unmap() on a number of
architectures to be an inline call rather than an actual function.
[akpm@linux-foundation.org: fix CONFIG_HIGHMEM=n build on various architectures]
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Christian König <christian.koenig@amd.com>
Cc: Chris Zankel <chris@zankel.net>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Helge Deller <deller@gmx.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20200507150004.1423069-5-ira.weiny@intel.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Recent work with KASan exposed the folling hard-coded bitmask
in arch/arm/mm/proc-macros.S:
bic rd, sp, #8128
bic rd, rd, #63
This forms the bitmask 0x1FFF that is coinciding with
(PAGE_SIZE << THREAD_SIZE_ORDER) - 1, this code was assuming
that THREAD_SIZE is always 8K (8192).
As KASan was increasing THREAD_SIZE_ORDER to 2, I ran into
this bug.
Fix it by this little oneline suggested by Ard:
bic rd, sp, #(THREAD_SIZE - 1) & ~63
Where THREAD_SIZE is defined using THREAD_SIZE_ORDER.
We have to also include <linux/const.h> since the THREAD_SIZE
expands to use the _AC() macro.
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Suggested-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
We would be trying to print the kernel virtual address of the base
register address which is not very useful and is not displayed by
default because of pointer restriction. Print the Device Tree node name
instead which is what was originally intended.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Pull dma-mapping updates from Christoph Hellwig:
- fix an integer overflow in the coherent pool (Kevin Grandemange)
- provide support for in-place uncached remapping and use that for
openrisc
- fix the arm coherent allocator to take the bus limit into account
* tag 'dma-mapping-5.7' of git://git.infradead.org/users/hch/dma-mapping:
ARM/dma-mapping: merge __dma_supported into arm_dma_supported
ARM/dma-mapping: take the bus limit into account in __dma_alloc
ARM/dma-mapping: remove get_coherent_dma_mask
openrisc: use the generic in-place uncached DMA allocator
dma-direct: provide a arch_dma_clear_uncached hook
dma-direct: make uncached_kernel_address more general
dma-direct: consolidate the error handling in dma_direct_alloc_pages
dma-direct: remove the cached_kernel_address hook
dma-coherent: fix integer overflow in the reserved-memory dma allocation
Pull kvm updates from Paolo Bonzini:
"ARM:
- GICv4.1 support
- 32bit host removal
PPC:
- secure (encrypted) using under the Protected Execution Framework
ultravisor
s390:
- allow disabling GISA (hardware interrupt injection) and protected
VMs/ultravisor support.
x86:
- New dirty bitmap flag that sets all bits in the bitmap when dirty
page logging is enabled; this is faster because it doesn't require
bulk modification of the page tables.
- Initial work on making nested SVM event injection more similar to
VMX, and less buggy.
- Various cleanups to MMU code (though the big ones and related
optimizations were delayed to 5.8). Instead of using cr3 in
function names which occasionally means eptp, KVM too has
standardized on "pgd".
- A large refactoring of CPUID features, which now use an array that
parallels the core x86_features.
- Some removal of pointer chasing from kvm_x86_ops, which will also
be switched to static calls as soon as they are available.
- New Tigerlake CPUID features.
- More bugfixes, optimizations and cleanups.
Generic:
- selftests: cleanups, new MMU notifier stress test, steal-time test
- CSV output for kvm_stat"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (277 commits)
x86/kvm: fix a missing-prototypes "vmread_error"
KVM: x86: Fix BUILD_BUG() in __cpuid_entry_get_reg() w/ CONFIG_UBSAN=y
KVM: VMX: Add a trampoline to fix VMREAD error handling
KVM: SVM: Annotate svm_x86_ops as __initdata
KVM: VMX: Annotate vmx_x86_ops as __initdata
KVM: x86: Drop __exit from kvm_x86_ops' hardware_unsetup()
KVM: x86: Copy kvm_x86_ops by value to eliminate layer of indirection
KVM: x86: Set kvm_x86_ops only after ->hardware_setup() completes
KVM: VMX: Configure runtime hooks using vmx_x86_ops
KVM: VMX: Move hardware_setup() definition below vmx_x86_ops
KVM: x86: Move init-only kvm_x86_ops to separate struct
KVM: Pass kvm_init()'s opaque param to additional arch funcs
s390/gmap: return proper error code on ksm unsharing
KVM: selftests: Fix cosmetic copy-paste error in vm_mem_region_move()
KVM: Fix out of range accesses to memslots
KVM: X86: Micro-optimize IPI fastpath delay
KVM: X86: Delay read msr data iff writes ICR MSR
KVM: PPC: Book3S HV: Add a capability for enabling secure guests
KVM: arm64: GICv4.1: Expose HW-based SGIs in debugfs
KVM: arm64: GICv4.1: Allow non-trapping WFI when using HW SGIs
...
The idea comes from a discussion between Linus and Andrea [1].
Before this patch we only allow a page fault to retry once. We achieved
this by clearing the FAULT_FLAG_ALLOW_RETRY flag when doing
handle_mm_fault() the second time. This was majorly used to avoid
unexpected starvation of the system by looping over forever to handle the
page fault on a single page. However that should hardly happen, and after
all for each code path to return a VM_FAULT_RETRY we'll first wait for a
condition (during which time we should possibly yield the cpu) to happen
before VM_FAULT_RETRY is really returned.
This patch removes the restriction by keeping the FAULT_FLAG_ALLOW_RETRY
flag when we receive VM_FAULT_RETRY. It means that the page fault handler
now can retry the page fault for multiple times if necessary without the
need to generate another page fault event. Meanwhile we still keep the
FAULT_FLAG_TRIED flag so page fault handler can still identify whether a
page fault is the first attempt or not.
Then we'll have these combinations of fault flags (only considering
ALLOW_RETRY flag and TRIED flag):
- ALLOW_RETRY and !TRIED: this means the page fault allows to
retry, and this is the first try
- ALLOW_RETRY and TRIED: this means the page fault allows to
retry, and this is not the first try
- !ALLOW_RETRY and !TRIED: this means the page fault does not allow
to retry at all
- !ALLOW_RETRY and TRIED: this is forbidden and should never be used
In existing code we have multiple places that has taken special care of
the first condition above by checking against (fault_flags &
FAULT_FLAG_ALLOW_RETRY). This patch introduces a simple helper to detect
the first retry of a page fault by checking against both (fault_flags &
FAULT_FLAG_ALLOW_RETRY) and !(fault_flag & FAULT_FLAG_TRIED) because now
even the 2nd try will have the ALLOW_RETRY set, then use that helper in
all existing special paths. One example is in __lock_page_or_retry(), now
we'll drop the mmap_sem only in the first attempt of page fault and we'll
keep it in follow up retries, so old locking behavior will be retained.
This will be a nice enhancement for current code [2] at the same time a
supporting material for the future userfaultfd-writeprotect work, since in
that work there will always be an explicit userfault writeprotect retry
for protected pages, and if that cannot resolve the page fault (e.g., when
userfaultfd-writeprotect is used in conjunction with swapped pages) then
we'll possibly need a 3rd retry of the page fault. It might also benefit
other potential users who will have similar requirement like userfault
write-protection.
GUP code is not touched yet and will be covered in follow up patch.
Please read the thread below for more information.
[1] https://lore.kernel.org/lkml/20171102193644.GB22686@redhat.com/
[2] https://lore.kernel.org/lkml/20181230154648.GB9832@redhat.com/
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Suggested-by: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Tested-by: Brian Geffon <bgeffon@google.com>
Cc: Bobby Powers <bobbypowers@gmail.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Denis Plotnikov <dplotnikov@virtuozzo.com>
Cc: "Dr . David Alan Gilbert" <dgilbert@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: "Kirill A . Shutemov" <kirill@shutemov.name>
Cc: Martin Cracauer <cracauer@cons.org>
Cc: Marty McFadden <mcfadden8@llnl.gov>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Maya Gokhale <gokhale2@llnl.gov>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Mike Rapoport <rppt@linux.vnet.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Link: http://lkml.kernel.org/r/20200220160246.9790-1-peterx@redhat.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge __dma_supported into its only caller, and move the resulting
function so that it doesn't need a forward declaration. Also mark
it static as there are no callers outside of dma-mapping.c.
Signed-off-by: Christoph Hellwig <hch@lst.de>
The DMA coherent allocator needs to take bus limits into account for
picking the zone that the memory is allocated from.
Reported-by: Roger Quadros <rogerq@ti.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Tested-by: Roger Quadros <rogerq@ti.com>
The core DMA code already checks for valid DMA masks earlier on, and
doesn't allow NULL struct device pointers. Remove the now not required
checks.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Pull ARM updates from Russell King:
- decompressor updates
- prevention of out-of-bounds access while stacktracing
- fix a section mismatch warning with free_memmap()
- make kexec depend on MMU to avoid some build errors
- remove swapops stubs
* tag 'for-linus' of git://git.armlinux.org.uk/~rmk/linux-arm:
ARM: 8954/1: NOMMU: remove stubs for swapops
ARM: 8952/1: Disable kmemleak on XIP kernels
ARM: 8951/1: Fix Kexec compilation issue.
ARM: 8949/1: mm: mark free_memmap as __init
ARM: 8948/1: Prevent OOB access in stacktrace
ARM: 8945/1: decompressor: use CONFIG option instead of cc-option
ARM: 8942/1: Revert "8857/1: efi: enable CP15 DMB instructions before cleaning the cache"
ARM: 8941/1: decompressor: enable CP15 barrier instructions in v7 cache setup code
max_pfn, as set in arch/arm/mm/init.c:
static void __init find_limits(unsigned long *min,
unsigned long *max_low,
unsigned long *max_high)
{
*max_low = PFN_DOWN(memblock_get_current_limit());
*min = PFN_UP(memblock_start_of_DRAM());
*max_high = PFN_DOWN(memblock_end_of_DRAM());
}
with memblock_end_of_DRAM() pointing to the next byte after DRAM. As
such, max_pfn points to the PFN after the end of DRAM.
Thus when using max_pfn to check DMA masks, we should subtract one when
checking DMA ranges against it.
Commit 8bf1268f48 ("ARM: dma-api: fix off-by-one error in
__dma_supported()") fixed the same issue, but missed this spot.
This issue was found while working on the sun4i-csi v4l2 driver on the
Allwinner R40 SoC. On Allwinner SoCs, DRAM is offset at 0x40000000, and
we are starting to use of_dma_configure() with the "dma-ranges" property
in the device tree to have the DMA API handle the offset.
In this particular instance, dma-ranges was set to the same range as the
actual available (2 GiB) DRAM. The following error appeared when the
driver attempted to allocate a buffer:
sun4i-csi 1c09000.csi: Coherent DMA mask 0x7fffffff (pfn 0x40000-0xc0000)
covers a smaller range of system memory than the DMA zone pfn 0x0-0xc0001
sun4i-csi 1c09000.csi: dma_alloc_coherent of size 307200 failed
Fixing the off-by-one error makes things work.
Link: http://lkml.kernel.org/r/20191224030239.5656-1-wens@kernel.org
Fixes: 11a5aa3256 ("ARM: dma-mapping: check DMA mask against available memory")
Fixes: 9f28cde0bc ("ARM: another fix for the DMA mapping checks")
Fixes: ab746573c4 ("ARM: dma-mapping: allow larger DMA mask than supported")
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Robin Murphy <robin.murphy@arm.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
As of commit ac7c3e4ff4 ("compiler: enable CONFIG_OPTIMIZE_INLINING
forcibly"), free_memmap() might not always be inlined, and thus is
triggering a section warning:
WARNING: vmlinux.o(.text.unlikely+0x904): Section mismatch in reference from the function free_memmap() to the function .meminit.text:memblock_free()
Mark it as __init, since the faller (free_unused_memmap) already is.
Fixes: ac7c3e4ff4 ("compiler: enable CONFIG_OPTIMIZE_INLINING forcibly")
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Pull ARM SoC platform updates from Olof Johansson:
"Most of these are for MMP (seeing a bunch of cleanups and refactorings
for the first time in a while), and for OMAP (a bunch of cleanups and
added support for voltage controller on OMAP4430)"
* tag 'armsoc-soc' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (51 commits)
ARM: OMAP2+: Add missing put_device() call in omapdss_init_of()
OMAP2: fixup doc comments in omap_device
ARM: OMAP1: drop duplicated dependency on ARCH_OMAP1
ARM: ASPEED: update default ARCH_NR_GPIO for ARCH_ASPEED
ARM: imx: use generic function to exit coherency
ARM: tegra: Use WFE for power-gating on Tegra30
ARM: tegra: Fix FLOW_CTLR_HALT register clobbering by tegra_resume()
ARM: exynos: Enable exynos-asv driver for ARCH_EXYNOS
ARM: s3c: Rename s5p_usb_phy functions
ARM: s3c: Rename s3c64xx_spi_setname() function
ARM: imx: Add serial number support for i.MX6/7 SoCs
ARM: imx: Drop imx_anatop_usb_chrg_detect_disable()
arm64: Introduce config for S32
ARM: hisi: drop useless depend on ARCH_MULTI_V7
arm64: realtek: Select reset controller
ARM: shmobile: rcar-gen2: Drop legacy DT clock support
ARM: OMAP2+: Remove duplicated include from pmic-cpcap.c
ARM: OMAP1: ams-delta FIQ: Fix a typo ("Initiaize")
MAINTAINERS: Add logicpd-som-lv and logicpd-torpedo to OMAP TREE
ARM: OMAP2+: pdata-quirks: drop TI_ST/KIM support
...
Pull ARM updates from Russell King:
- improve ARM implementation of pfn_valid()
- various sparse fixes
- spelling fixes
- add further ARMv8 debug architecture versions
- clang fix for decompressor
- update to generic vDSO
- remove Brahma-B53 from spectre hardening
- initialise broadcast hrtimer device
- use correct nm executable in decompressor
- remove old mcount et.al.
* tag 'for-linus' of git://git.armlinux.org.uk/~rmk/linux-arm: (26 commits)
ARM: 8940/1: ftrace: remove mcount(),ftrace_caller_old() and ftrace_call_old()
ARM: 8939/1: kbuild: use correct nm executable
ARM: 8938/1: kernel: initialize broadcast hrtimer based clock event device
ARM: 8937/1: spectre-v2: remove Brahma-B53 from hardening
ARM: 8933/1: replace Sun/Solaris style flag on section directive
ARM: 8932/1: Add clock_gettime64 entry point
ARM: 8931/1: Add clock_getres entry point
ARM: 8930/1: Add support for generic vDSO
ARM: 8929/1: use APSR_nzcv instead of r15 as mrc operand
ARM: 8927/1: ARM/hw_breakpoint: add more ARMv8 debug architecture versions support
ARM: 8918/2: only build return_address() if needed
ARM: 8928/1: ARM_ERRATA_775420: Spelling s/date/data/
ARM: 8925/1: tcm: include <asm/tcm.h> for missing declarations
ARM: 8924/1: tcm: make dtcm_end and itcm_end static
ARM: 8923/1: mm: include <asm/vga.h> for vga_base
ARM: 8922/1: parse_dt_topology() rate is pointer to __be32
ARM: 8920/1: share get_signal_page from signal.c to process.c
ARM: 8919/1: make unexported functions static
ARM: 8917/1: mm: include <asm/set_memory.h>
ARM: 8916/1: mm: make set_section_perms() static
...
Pull dma-mapping updates from Christoph Hellwig:
- improve dma-debug scalability (Eric Dumazet)
- tiny dma-debug cleanup (Dan Carpenter)
- check for vmap memory in dma_map_single (Kees Cook)
- check for dma_addr_t overflows in dma-direct when using DMA offsets
(Nicolas Saenz Julienne)
- switch the x86 sta2x11 SOC to use more generic DMA code (Nicolas
Saenz Julienne)
- fix arm-nommu dma-ranges handling (Vladimir Murzin)
- use __initdata in CMA (Shyam Saini)
- replace the bus dma mask with a limit (Nicolas Saenz Julienne)
- merge the remapping helpers into the main dma-direct flow (me)
- switch xtensa to the generic dma remap handling (me)
- various cleanups around dma_capable (me)
- remove unused dev arguments to various dma-noncoherent helpers (me)
* 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux:
* tag 'dma-mapping-5.5' of git://git.infradead.org/users/hch/dma-mapping: (22 commits)
dma-mapping: treat dev->bus_dma_mask as a DMA limit
dma-direct: exclude dma_direct_map_resource from the min_low_pfn check
dma-direct: don't check swiotlb=force in dma_direct_map_resource
dma-debug: clean up put_hash_bucket()
powerpc: remove support for NULL dev in __phys_to_dma / __dma_to_phys
dma-direct: avoid a forward declaration for phys_to_dma
dma-direct: unify the dma_capable definitions
dma-mapping: drop the dev argument to arch_sync_dma_for_*
x86/PCI: sta2x11: use default DMA address translation
dma-direct: check for overflows on 32 bit DMA addresses
dma-debug: increase HASH_SIZE
dma-debug: reorder struct dma_debug_entry fields
xtensa: use the generic uncached segment support
dma-mapping: merge the generic remapping helpers into dma-direct
dma-direct: provide mmap and get_sgtable method overrides
dma-direct: remove the dma_handle argument to __dma_direct_alloc_pages
dma-direct: remove __dma_direct_free_pages
usb: core: Remove redundant vmap checks
kernel: dma-contiguous: mark CMA parameters __initdata/__initconst
dma-debug: add a schedule point in debug_dma_dump_mappings()
...
Pull generic ioremap support from Christoph Hellwig:
"This adds the remaining bits for an entirely generic ioremap and
iounmap to lib/ioremap.c. To facilitate that, it cleans up the giant
mess of weird ioremap variants we had with no users outside the arch
code.
For now just the three newest ports use the code, but there is more
than a handful others that can be converted without too much work.
Summary:
- clean up various obsolete ioremap and iounmap variants
- add a new generic ioremap implementation and switch csky, nds32 and
riscv over to it"
* tag 'ioremap-5.5' of git://git.infradead.org/users/hch/ioremap: (21 commits)
nds32: use generic ioremap
csky: use generic ioremap
csky: remove ioremap_cache
riscv: use the generic ioremap code
lib: provide a simple generic ioremap implementation
sh: remove __iounmap
nios2: remove __iounmap
hexagon: remove __iounmap
m68k: rename __iounmap and mark it static
arch: rely on asm-generic/io.h for default ioremap_* definitions
asm-generic: don't provide ioremap for CONFIG_MMU
asm-generic: ioremap_uc should behave the same with and without MMU
xtensa: clean up ioremap
x86: Clean up ioremap()
parisc: remove __ioremap
nios2: remove __ioremap
alpha: remove the unused __ioremap wrapper
hexagon: clean up ioremap
ia64: rename ioremap_nocache to ioremap_uc
unicore32: remove ioremap_cached
...
