The Landisk setup code maps the CF IDE area using ioremap_prot(), and
passes the resulting virtual addresses to the pata_platform driver,
disguising them as I/O port addresses.  Hence the pata_platform driver
translates them again using ioport_map().
As CONFIG_GENERIC_IOMAP=n, and CONFIG_HAS_IOPORT_MAP=y, the
SuperH-specific mapping code in arch/sh/kernel/ioport.c translates
I/O port addresses to virtual addresses by adding sh_io_port_base, which
defaults to -1, thus breaking the assumption of an identity mapping.

Fix this by setting sh_io_port_base to zero.

Fixes: 37b7a978 ("sh: machvec IO death.")
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: NRich Felker <dalias@libc.org>

Other architectures expect that syscall_set_return_value gets an already
negative value as error. That's also what kernel/seccomp.c provides.
Signed-off-by: NMichael Karcher <kernel@mkarcher.dialup.fu-berlin.de>
Tested-by: NJohn Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Signed-off-by: NRich Felker <dalias@libc.org>

Port sh to use the new SECCOMP_FILTER code.
Signed-off-by: NMichael Karcher <kernel@mkarcher.dialup.fu-berlin.de>
Tested-by: NJohn Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Signed-off-by: NRich Felker <dalias@libc.org>

This avoids out-of-range jumps that get auto-replaced by the assembler
and prepares for the changes needed to implement SECCOMP_FILTER cleanly.
Signed-off-by: NMichael Karcher <kernel@mkarcher.dialup.fu-berlin.de>
Tested-by: NJohn Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Signed-off-by: NRich Felker <dalias@libc.org>

Use the copy_thread_tls() calling convention which passes tls through a
register. This is required so we can remove the copy_thread{_tls}() split
and remove the HAVE_COPY_THREAD_TLS macro.

Cc: Rich Felker <dalias@libc.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: linux-sh@vger.kernel.org
Signed-off-by: NChristian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: NRich Felker <dalias@libc.org>

This switches to using common code for the DMA allocations, including
potential use of the CMA allocator if configured.

Switching to the generic code enables DMA allocations from atomic
context, which is required by the DMA API documentation, and also
adds various other minor features drivers start relying upon.  It
also makes sure we have on tested code base for all architectures
that require uncached pte bits for coherent DMA allocations.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NRich Felker <dalias@libc.org>

The code handling non-coherent DMA depends on being able to remap code
as non-cached.  But that can't be done without an MMU, so using this
option on NOMMU builds is broken.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NRich Felker <dalias@libc.org>

Have a single definition that architetures can select.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NRich Felker <dalias@libc.org>

Both functions are only used by compiled in core code.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NRich Felker <dalias@libc.org>

No need to expose the details of trapped I/O to drivers.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NRich Felker <dalias@libc.org>

Move the internal implementation details of ioremap out of line, no need
to expose any of this to drivers for a slow path API.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NRich Felker <dalias@libc.org>

ioremap_fixed is an internal implementation detail and should not be
exposed to drivers.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NRich Felker <dalias@libc.org>

There is no point in having __KERNEL__ ifdefs in headers not exported to
userspace.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NRich Felker <dalias@libc.org>

Ensure there is an order for the selects.  Also remove a duplicate
one.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NRich Felker <dalias@libc.org>

The sh build is full of warnings when building with gcc 9.2.1.  While
fixing those would be great, at least avoid failing the build.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NRich Felker <dalias@libc.org>

Rationale:
Reduces attack surface on kernel devs opening the links for MITM
as HTTPS traffic is much harder to manipulate.

Deterministic algorithm:
For each file:
  If not .svg:
    For each line:
      If doesn't contain `\bxmlns\b`:
        For each link, `\bhttp://[^# \t\r\n]*(?:\w|/)`:
	  If neither `\bgnu\.org/license`, nor `\bmozilla\.org/MPL\b`:
            If both the HTTP and HTTPS versions
            return 200 OK and serve the same content:
              Replace HTTP with HTTPS.
Signed-off-by: NAlexander A. Klimov <grandmaster@al2klimov.de>
Signed-off-by: NRich Felker <dalias@libc.org>

Drop all configs with the CONFIG_SOC_CAMERA prefix since those
have been removed.

SOC_CAMERA support for the sh architecture was removed a long time ago.
Drop it from the configs.
Signed-off-by: NHans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: NRich Felker <dalias@libc.org>

The SH implementation never called stacktrace_ops.stack().
Presumably this was copied from the x86 implementation.

Hence remove the method, and all implementations (most of them are
dummies).
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: NRich Felker <dalias@libc.org>

  - Convert from printk() to pr_*(),
  - Add missing continuations.
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Tested-by: NGuenter Roeck <linux@roeck-us.net>
Signed-off-by: NRich Felker <dalias@libc.org>

  - Convert from printk() to pr_*(),
  - Add missing continuations,
  - Join broken messages.

Note that printk(KERN_DEBUG ...) is retained, to preserve behavior
(pr_debug() is a dummy if DEBUG is not defined).
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Tested-by: NGuenter Roeck <linux@roeck-us.net>
Signed-off-by: NRich Felker <dalias@libc.org>

  - Convert from printk() to pr_*(),
  - Add missing continuation.
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Tested-by: NGuenter Roeck <linux@roeck-us.net>
Signed-off-by: NRich Felker <dalias@libc.org>

Rejoin the broken lines by using pr_cont().
Convert the remaining printk() calls to pr_*() while at it.
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Tested-by: NGuenter Roeck <linux@roeck-us.net>
Signed-off-by: NRich Felker <dalias@libc.org>

Rejoin the broken lines by dropping the log level parameters and using
pr_cont().
Use "%px" to print sensible addresses in call traces.
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Tested-by: NGuenter Roeck <linux@roeck-us.net>
Signed-off-by: NRich Felker <dalias@libc.org>

Rejoin the broken lines by using pr_cont().
Convert the remaining printk() calls to pr_*() while at it.
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Tested-by: NGuenter Roeck <linux@roeck-us.net>
Signed-off-by: NRich Felker <dalias@libc.org>

This reverts commit 8b92f348.

"data" became the log level in commit 539e786c ("sh: add loglvl
to show_trace()"), so we do need to keep the printk() before the
continuation in print_trace_address().
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: NRich Felker <dalias@libc.org>

This reverts commit 2deebe4d.

printk_address() is always used as a continuation of the previous
logging, hence it should not include a log level.
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: NRich Felker <dalias@libc.org>

Somewhere along the patch handling path, both the old "printk(KERN_ALERT
....)" and the new "pr_alert(...)" were retained, leading to the
duplicate printing of "PC:".

Drop the old one.

Fixes: eaabf98b ("sh: fault: modernize printing of kernel messages")
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: NRich Felker <dalias@libc.org>

As of commit 37744fee ("sh: remove sh5 support"), support for
the SH5-based Cayman platform can no longer be selected.
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: NRich Felker <dalias@libc.org>

Since the removal of core support for SH5, Cayman support can no longer
be selected.

Fixes: 37744fee ("sh: remove sh5 support")
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: NRich Felker <dalias@libc.org>

Fix typo: "triger" --> "trigger"
Signed-off-by: NFlavio Suligoi <f.suligoi@asem.it>
Signed-off-by: NRich Felker <dalias@libc.org>

This reverts commit d1f56f31.

__delay() is an internal implementation detail on several architectures.
Drivers should not call __delay() directly, as it has non-standardized
semantics, or may not even exist.
Hence there is no need to export __delay() to modules.

See also include/asm-generic/delay.h:

    /* Undefined functions to get compile-time errors */
    ...
    extern void __delay(unsigned long loops);
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: NRich Felker <dalias@libc.org>

Trying to build the kernel with CONFIG_INFINIBAND_USER_ACCESS enabled fails

     ERROR: "__get_user_unknown" [drivers/infiniband/core/ib_uverbs.ko] undefined!

with on SH since the kernel misses a 64-bit implementation of get_user().

Implement the missing 64-bit get_user() as __get_user_u64(), matching the
already existing __put_user_u64() which implements the 64-bit put_user().
Signed-off-by: NJohn Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Acked-by: NYoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: NRich Felker <dalias@libc.org>

Function dma_alloc_coherent use in buf already zeroes out memory,
so memset is not needed.
Signed-off-by: NChen Zhou <chenzhou10@huawei.com>
Signed-off-by: NRich Felker <dalias@libc.org>

The pgd kmem_cache allocation both specified __GFP_ZERO and had a
constructor which makes no sense.  Remove __GFP_ZERO and zero the user
parts of the pgd explicitly.
Signed-off-by: NMatthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: NRich Felker <dalias@libc.org>

Pull x86 fix from Ingo Molnar:
 "A single fix for a potential deadlock when printing a message about
  spurious interrupts"

* tag 'x86-urgent-2020-08-02' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/i8259: Use printk_deferred() to prevent deadlock

Merge tag 'kbuild-fixes-v5.8-4' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild

Pull Kbuild fixes from Masahiro Yamada:

 - clean the generated moc file for xconfig

 - fix xconfig bugs, and revert some bad commits

* tag 'kbuild-fixes-v5.8-4' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
  kbuild: remove redundant FORCE definition in scripts/Makefile.modpost
  kconfig: qconf: remove wrong ConfigList::firstChild()
  Revert "kconfig: qconf: don't show goback button on splitMode"
  Revert "kconfig: qconf: Change title for the item window"
  kconfig: qconf: remove "goBack" debug message
  kconfig: qconf: use delete[] instead of delete to free array
  kconfig: qconf: compile moc object separately
  kconfig: qconf: use if_changed for qconf.moc rule
  modpost: explain why we can't use strsep

Pull KVM fixes from Paolo Bonzini:
 "Bugfixes and strengthening the validity checks on inputs from new
  userspace APIs.

  Now I know why I shouldn't prepare pull requests on the weekend, it's
  hard to concentrate if your son is shouting about his latest Minecraft
  builds in your ear. Fortunately all the patches were ready and I just
  had to check the test results..."

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: SVM: Fix disable pause loop exit/pause filtering capability on SVM
  KVM: LAPIC: Prevent setting the tscdeadline timer if the lapic is hw disabled
  KVM: arm64: Don't inherit exec permission across page-table levels
  KVM: arm64: Prevent vcpu_has_ptrauth from generating OOL functions
  KVM: nVMX: check for invalid hdr.vmx.flags
  KVM: nVMX: check for required but missing VMCS12 in KVM_SET_NESTED_STATE
  selftests: kvm: do not set guest mode flag

The same code exists a few lines above.

Fixes: 436b2ac6 ("modpost: invoke modpost only when input files are updated")
Signed-off-by: NMasahiro Yamada <masahiroy@kernel.org>

This function returns the first child object, but the returned pointer
is not compatible with (ConfigItem *).

Commit cc1c08ed ("kconfig: qconf: don't show goback button on
splitMode") uncovered this issue because using the pointer from this
function would make qconf crash. (https://lkml.org/lkml/2020/7/18/411)

This function does not work. Remove.
Signed-off-by: NMasahiro Yamada <masahiroy@kernel.org>