rapidio-add-mport-char-device-driver-fix

fix printk warning on i386:

drivers/rapidio/devices/rio_mport_cdev.c: In function 'mport_mm_open':
include/linux/dynamic_debug.h:64:16: warning: format '%llx' expects argument of type 'long long unsigned int', but argument 4 has type 'dma_addr_t' [-Wformat=]

Cc: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Barry Wood <barry.wood@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio: add mport char device driver

Add mport character device driver to provide user space interface
to basic RapidIO subsystem operations.
See included Documentation/rapidio/mport_cdev.txt for more details.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Tested-by: Barry Wood <barry.wood@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Cc: Barry Wood <barry.wood@idt.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/tsi721_dma: fix hardware error handling

Add DMA channel re-initialization after an error to avoid termination
of all pending transfer requests.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Reported-by: Barry Wood <barry.wood@idt.com>
Tested-by: Barry Wood <barry.wood@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Cc: Barry Wood <barry.wood@idt.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/tsi721_dma: fix synchronization issues

Fix synchronization issues found during testing using multiple DMA
transfer requests to the same channel:
- lost MSI-X interrupt notifications
- non-synchronized attempts to start DMA channel HW resulting in error
  message from the driver
- cookie tracking/update race conditions resulting in incorrect DMA
  transfer status report

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Reported-by: Barry Wood <barry.wood@idt.com>
Tested-by: Barry Wood <barry.wood@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Cc: Barry Wood <barry.wood@idt.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/tsi721_dma: update error reporting from prep_sg callback

Switch to returning error-valued pointer instead of simple NULL pointer.
This allows to properly identify situation when request queue is full and
therefore gives to upper layer an option to retry operation later.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/tsi721: add filtered debug output

Replace "all-or-nothing" debug output with controlled debug output using
functional block masks.  This allows run time control of debug messages
through 'dbg_level' module parameter.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/tsi721: add outbound windows mapping support

Add device-specific callback functions to support outbound windows mapping
and release.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio: add outbound window support

Add RapidIO controller (mport) outbound window configuration operations.
  This patch is a part of the original patch submitted by Li Yang
  (https://lists.ozlabs.org/pipermail/linuxppc-dev/2009-April/071210.html).
  For some reason the original part was not applied to mainline code
  tree.  The inbound window mapping part has been applied later during
  tsi721 mport driver submission.  Now goes the second part with
  corresponding HW support.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Li Yang <leoli@freescale.com>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/tsi721: fix locking in OB_MSG processing

- Add spinlock protection into outbound message queuing routine.

- Change outbound message interrupt handler to avoid deadlock when
  calling registered callback routine.

- Allow infinite retries for outbound messages to avoid retry threshold
  error signaling in systems with nodes that have slow message receive
  queue processing.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio: add global inbound port write interfaces

Add new Port Write handler registration interfaces that attach PW handlers
to local mport device objects.  This is different from old interface that
attaches PW callback to individual RapidIO device.  The new interfaces are
intended for use for common event handling (e.g.  hot-plug notifications)
while the old interface is available for individual device drivers.

This patch is based on patch proposed by Andre van Herk but preserves
existing per-device interface and adds lock protection for list handling.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio: move rio_pw_enable into core code

Make rio_pw_enable() routine available to other RapidIO drivers.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio: move rio_local_set_device_id function to the common core

Make function rio_local_set_device_id() common for all components of
RapidIO subsystem.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio: add lock protection for doorbell list

Add lock protection around doorbell list handling to prevent list
corruption on SMP platforms.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/rionet: add mport removal handling

Add handling of a local mport device removal.

RIONET driver registers itself as class interface that supports only
removal notification, 'add_device' callback is not provided because RIONET
network device can be initialized only after enumeration is completed and
the existing method (using remote peer addition) satisfies this condition.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/rionet: add locking into add/remove device

Add spinlock protection when handling list of connected peers and ability
to handle new peer device addition after the RIONET device was open.
Before his update RIONET was sending JOIN requests only when it have been
opened, peer devices added later have been missing from this process.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

powerpc/fsl_rio: changes to mport registration

Change mport object initialization/registration sequence to match reworked
version of rio_register_mport() in the core code.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/tsi721: add HW specific mport removal

Add hardware-specific device removal support for Tsi721 PCIe-to-RapidIO
bridge.  To avoid excessive data type conversions, parameters passed to
some internal functions have been revised.  Dynamic memory allocations of
rio_mport and rio_ops have been replaced to reduce references between data
structures.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio: add core mport removal support

Add common mport removal support functions into the RapidIO subsystem core.

Changes to the existing mport registration process have been made to avoid
race conditions with active subsystem interfaces immediately after mport
device registration: part of initialization code from rio_register_mport()
have been moved into separate function rio_mport_initialize() to allow to
perform mport registration as the final step of setup process.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio: move net allocation into core code

Make net allocation/release routines available to all components of RapidIO
subsystem by moving code from rio-scan enumerator.
Make destination ID allocation method private to existing enumerator
because other enumeration methods can use their own algorithm.
Setup net device object as a parent of all RapidIO devices residing in it
and register net as a child of active mport device.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio: rework common RIO device add/delete routines

This patch moves per-net device list handling from rio-scan to common
RapidIO core and adds a matching device deletion routine. This makes
device object creation/removal available to other implementations of
enumeration/discovery process.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/rionet: add shutdown event handling

Add shutdown notification handler which terminates active connections with
remote RapidIO nodes.  This prevents remote nodes from sending packets to
the powered off node and eliminates hardware error events on remote nodes.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/tsi721: add shutdown notification callback

Add device driver specific shutdown notification callback.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio: add shutdown notification for RapidIO devices

Add bus-specific callback to stop RapidIO devices during a system shutdown.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/tsi721: add query_mport callback

Add device-specific implementation of query_mport callback function.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio: add query_mport operation

Add mport query operation to report master port RapidIO capabilities and
run time configuration to upper level drivers.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/tsi721_dma: fix pending transaction queue handling

Fix pending DMA request queue handling to avoid broken ordering during
concurrent request submissions.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/tsi721: add option to configure direct mapping of IB window

Add an option to configure mapping of Inbound Window without RIO-to-PCIe
address translation.

If a local memory buffer is not properly aligned to meet HW requirements
for RapidIO address mapping with address translation, caller can request
an inbound window with matching RapidIO address assigned to it.  This
implementation selects RapidIO base address and size for inbound window
that are capable to accommodate the local memory buffer.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/tsi721: add check for overlapped IB window mappings

Add check for attempts to request mapping of inbound RapidIO address space
that overlaps with existing active mapping windows.

Tsi721 device does not support overlapped inbound windows and SRIO address
decoding behavior is not defined in such cases.

This patch is applicable to kernel versions starting from v3.7.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/tsi721: fix hardcoded MRRS setting

Remove use of hardcoded setting for Maximum Read Request Size (MRRS) value
and use one set by PCIe bus driver.

Using hardcoded value can cause PCIe bus errors on platforms that have
tsi721 device on PCIe path that allows only smaller read request sizes.

This fix is applicable to kernel versions starting from v3.2.

Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Aurelien Jacquiot <a-jacquiot@ti.com>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/rionet: add capability to change MTU

These patches are the result of extensive collaboration within the
RapidIO.org Software Task Group between Texas Instruments, Freescale,
Prodrive Technologies, Nokia Networks, BAE and IDT.  Additional input was
received from other members of RapidIO.org.  The objective was to create a
character mode driver interface which exposes the capabilities of RapidIO
devices directly to applications, in a manner that allows the numerous and
varied RapidIO implementations to interoperate.

The Software Task Group has also developed fabric management, Remote
Memory Access, and sockets applications which make use of these interfaces
in user space.  Intensive testing with these applications prompted the
RapidIO subsystem updates provided within this set of patches.

This patch (of 29):

Replace default Ethernet-specific routine by the custom one to allow
setting of larger MTU supported by RapidIO messaging (max RIO packet size
is 4096 bytes).

Signed-off-by: Aurelien Jacquiot <a-jacquiot@ti.com>
Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

rapidio/rionet: fix deadlock on SMP

Fix deadlocking during concurrent receive and transmit operations on SMP
platforms caused by the use of incorrect lock: on transmit 'tx_lock'
spinlock should be used instead of 'lock' which is used for receive
operation.

This fix is applicable to kernel versions starting from v2.15.

Signed-off-by: Aurelien Jacquiot <a-jacquiot@ti.com>
Signed-off-by: Alexandre Bounine <alexandre.bounine@idt.com>
Cc: Matt Porter <mporter@kernel.crashing.org>
Cc: Andre van Herk <andre.van.herk@prodrive-technologies.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

kdump, vmcoreinfo: report actual value of phys_base

Currently, VMCOREINFO note information reports the virtual address of
phys_base that is assigned to symbol phys_base.  But this doesn't make
sense because to refer to phys_base, it's necessary to get the value of
phys_base itself we are now about to refer to.

Userland tools related to kdump such as makedumpfile and crash utility so
far have made some efforts to calculate phys_base on crash dump formats
generated by mechanisms running outside Linux kernel, such as virtual
machine hypervisor such as qemu dump, which ordinary users use via virsh
dump, or ones implemented on vendor specific firmware.

That is, find a kernel data whose virtual and physical addresses are
available via its note information and calculate phys_base from it.
However, such data structure is not the one prepared for phys_base
purpose.  There's no guarantee that other crash dump mechanisms include
such information that can be used to calculate phys_base similarly.

To get VMCOREINFO in vmcore, it's easy to use strings and grep commands
like this; VMCOREINFO consists of simple string:

$ strings vmcore-3.10.0-121.el7.x86_64 | grep -E ".*VMCOREINFO.*" -A 100
VMCOREINFO
OSRELEASE=3.10.0-121.el7.x86_64
PAGESIZE=4096
...

This is also useful to get value of phys_base in kdump 2nd kernel
contained in vmcore using the above-mentioned external crash dump
mechanism; kdump 2nd kernel is an inherently relocated kernel.

This commit doesn't remove VMCOREINFO_SYMBOL(phys_base) line because
makedumpfile refers to it and if removing it, old versions makedumpfile
doesn't work well.

Signed-off-by: HATAYAMA Daisuke <d.hatayama@jp.fujitsu.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Vivek Goyal <vgoyal@redhat.com>
Cc: Atsushi Kumagai <kumagai-atsushi@mxc.nes.nec.co.jp>
Cc: Dave Anderson <anderson@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

kexec-provide-arch_kexec_protectunprotect_crashkres-v4

v3->v4:
Add some comments according to Minfei's suggestion.

Signed-off-by: Xunlei Pang <xlpang@redhat.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Dave Young <dyoung@redhat.com>
Cc: Minfei Huang <mhuang@redhat.com>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

kexec: provide arch_kexec_protect(unprotect)_crashkres()

Implement the protection method for the crash kernel memory reservation
for the 64-bit x86 kdump.

Signed-off-by: Xunlei Pang <xlpang@redhat.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Dave Young <dyoung@redhat.com>
Cc: Minfei Huang <mhuang@redhat.com>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

kexec-introduce-a-protection-mechanism-for-the-crashkernel-reserved-memory-v4

Signed-off-by: Xunlei Pang <xlpang@redhat.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Dave Young <dyoung@redhat.com>
Cc: Minfei Huang <mhuang@redhat.com>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

kexec: introduce a protection mechanism for the crashkernel reserved memory

For the cases that some kernel (module) path stamps the crash reserved
memory(already mapped by the kernel) where has been loaded the second
kernel data, the kdump kernel will probably fail to boot when panic
happens (or even not happens) leaving the culprit at large, this is
unacceptable.

The patch introduces a mechanism for detecting such cases:

1) After each crash kexec loading, it simply marks the reserved memory
   regions readonly since we no longer access it after that.  When someone
   stamps the region, the first kernel will panic and trigger the kdump.
   The weak arch_kexec_protect_crashkres() is introduced to do the actual
   protection.

2) To allow multiple loading, once 1) was done we also need to remark
   the reserved memory to readwrite each time a system call related to
   kdump is made.  The weak arch_kexec_unprotect_crashkres() is introduced
   to do the actual protection.

The architecture can make its specific implementation by overriding
arch_kexec_protect_crashkres() and arch_kexec_unprotect_crashkres().

Signed-off-by: Xunlei Pang <xlpang@redhat.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Cc: Dave Young <dyoung@redhat.com>
Cc: Minfei Huang <mhuang@redhat.com>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

kernel/signal.c: add compile-time check for __ARCH_SI_PREAMBLE_SIZE

The value of __ARCH_SI_PREAMBLE_SIZE defines the size (including padding)
of the part of the struct siginfo that is before the union, and it is then
used to calculate the needed padding (SI_PAD_SIZE) to make the size of
struct siginfo equal to 128 (SI_MAX_SIZE) bytes.

Depending on the target architecture and word width it equals to either 3
or 4 times sizeof int.

Since the very beginning we had __ARCH_SI_PREAMBLE_SIZE wrong on the
parisc architecture for the 64bit kernel build.  It's even more
frustrating, because it can easily be checked at compile time if the value
was defined correctly.

This patch adds such a check for the correctness of
__ARCH_SI_PREAMBLE_SIZE in the hope that it will prevent existing and
future architectures from running into the same problem.

I refrained from replacing __ARCH_SI_PREAMBLE_SIZE by offsetof() in
copy_siginfo() in include/asm-generic/siginfo.h, because a) it doesn't
make any difference and b) it's used in the Documentation/kmemcheck.txt
example.

I ran this patch through the 0-DAY kernel test infrastructure and only the
parisc architecture triggered as expected.  That means that this patch
should be OK for all major architectures.

Signed-off-by: Helge Deller <deller@gmx.de>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

autofs4: fix string.h include in auto_dev-ioctl.h

Since including linux/string.h will now do the right thing remove the
conditional check.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

autofs4: use pr_xxx() macros directly for logging

Use the standard pr_xxx() log macros directly for log prints instead of
the AUTOFS_XXX() macros.

Signed-off-by: Ian Kent <ikent@redhat.com>
Cc: Joe Perches <joe@perches.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

autofs4: change log print macros to not insert newline

Common kernel coding practice is to include the newline of log prints
within the log text rather than hidden away in a macro.

To avoid introducing inconsistencies as changes are made change the log
macros to not include the newline.

Signed-off-by: Ian Kent <raven@themaw.net>
Cc: Joe Perches <joe@perches.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

autofs4: make autofs log prints consistent

Use the pr_*() print in AUTOFS_*() macros instead of printks and include
the module name in log message macros.  Also use the AUTOFS_*() macros
everywhere instead of raw printks.

Signed-off-by: Ian Kent <raven@themaw.net>
Cc: Joe Perches <joe@perches.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

autofs4: fix some white space errors

Fix some white space format errors.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

autofs4: fix invalid ioctl return in autofs4_root_ioctl_unlocked()

The return from an ioctl if an invalid ioctl is passed in should be EINVAL
not ENOSYS.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

autofs4: fix coding style line length in autofs4_wait()

The need for this is questionable but checkpatch.pl complains about the
line length and it's a straightfoward change.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

autofs4: fix coding style problem in autofs4_get_set_timeout()

Refactor autofs4_get_set_timeout() to eliminate coding style error.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

autofs4: coding style fixes

Try and make the coding style completely consistent throughtout the autofs
module and inline with kernel coding style recommendations.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

autofs: show pipe inode in mount options

This is required for CRIU (Checkpoint Restart In Userspace) to migrate a
mount point when write end in user space is closed.

Below is a brief description of the problem.

To migrate a non-catatonic autofs mount point, one has to restore the
control pipe between kernel and autofs master process.

One of the autofs masters is systemd, which closes pipe write end after
passing it to the kernel with mount call.

To be able to restore the systemd control pipe one has to know which read
pipe end in systemd corresponds to the write pipe end in the kernel.  The
pipe "fd" in mount options is not enough because it was closed and
probably replaced by some other descriptor.

Thus, some other attribute is required to be able to find the read pipe
end.  The best attribute to use to find the correct pipe end is inode
number becuase it's unique for the whole system and can't be reused while
the autofs mount exists.

This attribute can also be used to recognize a situation where an autofs
mount has no master (no process with specified "pgrp" or no file
descriptor with "pipe_ino", specified in autofs mount options).

Signed-off-by: Stanislav Kinsburskiy <skinsbursky@virtuozzo.com>
Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

init/main.c: use list_for_each_entry()

Use list_for_each_entry() instead of list_for_each() to simplify the code.

Signed-off-by: Geliang Tang <geliangtang@163.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

scripts/link-vmlinux.sh: force error on kallsyms failure

Since the output of the invocation of scripts/kallsyms is piped directly
into the assembler, error messages it emits are visible on stderr, but a
non-zero return code is ignored, and the build simply proceeds in that
case.  However, the resulting kernel is most likely broken, and will crash
at boot.

So instead, capture the output of kallsyms in a separate .S file, and pass
that to the assembler in a separate step.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Guenter Roeck <linux@roeck-us.net>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Michal Marek <mmarek@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

kallsyms: add support for relative offsets in kallsyms address table

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Guenter Roeck <linux@roeck-us.net>
Cc: Kees Cook <keescook@chromium.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Michal Marek <mmarek@suse.cz>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Sudip Mukherjee <sudipm.mukherjee@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

kallsyms: add support for relative offsets in kallsyms address table

Similar to how relative extables are implemented, it is possible to emit
the kallsyms table in such a way that it contains offsets relative to some
anchor point in the kernel image rather than absolute addresses.

On 64-bit architectures, it cuts the size of the kallsyms address table in
half, since offsets between kernel symbols can typically be expressed in
32 bits.  This saves several hundreds of kilobytes of permanent .rodata on
average.  In addition, the kallsyms address table is no longer subject to
dynamic relocation when CONFIG_RELOCATABLE is in effect, so the relocation
work done after decompression now doesn't have to do relocation updates
for all these values.  This saves up to 24 bytes (i.e., the size of a
ELF64 RELA relocation table entry) per value, which easily adds up to a
couple of megabytes of uncompressed __init data on ppc64 or arm64.  Even
if these relocation entries typically compress well, the combined size
reduction of 2.8 MB uncompressed for a ppc64_defconfig build (of which 2.4
MB is __init data) results in a ~500 KB space saving in the compressed
image.

Since it is useful for some architectures (like x86) to retain the ability
to emit absolute values as well, this patch also adds support for
capturing both absolute and relative values when KALLSYMS_ABSOLUTE_PERCPU
is in effect, by emitting absolute per-cpu addresses as positive 32-bit
values, and addresses relative to the lowest encountered relative symbol
as negative values, which are subtracted from the runtime address of this
base symbol to produce the actual address.

Support for the above is enabled by default for all architectures except
IA-64 and Tile-GX, whose symbols are too far apart to capture in this
manner.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Kees Cook <keescook@chromium.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Michal Marek <mmarek@suse.cz>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

kallsyms: don't overload absolute symbol type for percpu symbols

Commit c6bda7c988a5 ("kallsyms: fix percpu vars on x86-64 with
relocation") overloaded the 'A' (absolute) symbol type to signify that a
symbol is not subject to dynamic relocation.  However, the original A type
does not imply that at all, and depending on the version of the toolchain,
many A type symbols are emitted that are in fact relative to the kernel
text, i.e., if the kernel is relocated at runtime, these symbols should be
updated as well.

For instance, on sparc32, the following symbols are emitted as absolute
(kindly provided by Guenter Roeck):

  f035a420 A _etext
  f03d9000 A _sdata
  f03de8c4 A jiffies
  f03f8860 A _edata
  f03fc000 A __init_begin
  f041bdc8 A __init_text_end
  f0423000 A __bss_start
  f0423000 A __init_end
  f044457d A __bss_stop
  f044457d A _end

On x86_64, similar behavior can be observed:

  ffffffff81a00000 A __end_rodata_hpage_align
  ffffffff81b19000 A __vvar_page
  ffffffff81d3d000 A _end

Even if only a couple of them pass the symbol range check that results in
them to be taken into account for the final kallsyms symbol table, it is
obvious that 'A' does not mean the symbol does not need to be updated at
relocation time, and overloading its meaning to signify that is perhaps
not a good idea.

So instead, add a new percpu_absolute member to struct sym_entry, and when
--absolute-percpu is in effect, use it to record symbols whose addresses
should be emitted as final values rather than values that still require
relocation at runtime.  That way, we can drop the check against the 'A'
type.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Kees Cook <keescook@chromium.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Michal Marek <mmarek@suse.cz>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

x86-kallsyms-disable-absolute-percpu-symbols-on-smp-v5

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Kees Cook <keescook@chromium.org>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

x86: kallsyms: disable absolute percpu symbols on !SMP

scripts/kallsyms.c has a special --absolute-percpu command line
option which deals with the zero based per cpu offsets that are
used when building for SMP on x86_64. This means that the option
should only be passed in that case, so add a Kconfig symbol with
the correct predicate, and use that instead.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Kees Cook <keescook@chromium.org>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Michal Marek <mmarek@suse.cz>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

x86/compat: remove is_compat_task()

x86's is_compat_task always checked the current syscall type, not the task
type.  It has no non-arch users any more, so just remove it to avoid
confusion.

On x86, nothing should really be checking the task ABI.  There are
legitimate users for the syscall ABI and for the mm ABI.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

drivers/hid/uhid.c: check write() bitness using in_compat_syscall

uhid changes the format expected in write() depending on bitness.
It should check the syscall bitness directly.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: David Herrmann <dh.herrmann@googlemail.com>
Acked-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

input: redefine INPUT_COMPAT_TEST as in_compat_syscall()

The input compat code should work like all other compat code: for 32-bit
syscalls, use the 32-bit ABI and for 64-bit syscalls, use the 64-bit ABI.
We have a helper for that (in_compat_syscall()): just use it.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

drivers/gpu/drm/amd/amdkfd: use in_compat_syscall to check open() caller type

amdkfd wants to know syscall type, not task type.  Check directly.

Unfortunately, amdkfd is making nasty assumptions that a process' bitness
is a well-defined constant thing.  This isn't the case on x86.  I don't
know how much this matters, but this patch has no effect on generated code
on x86, so amdkfd is equally broken with and without this patch.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Oded Gabbay <oded.gabbay@gmail.com>
Cc: David Airlie <airlied@linux.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

drivers/firmware/efi/efivars.c: use in_compat_syscall() to check for compat callers

This should make no difference on any architecture, as x86's historical
is_compat_task behavior really did check whether the calling syscall was a
compat syscall.  x86's is_compat_task is going away, though.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Reviewed-by: Matt Fleming <matt@codeblueprint.co.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

firewire: Use in_compat_syscall to check ioctl compatness

Firewire was using is_compat_task to check whether it was in a
compat ioctl or a non-compat ioctl.  Use is_compat_syscall instead
so it works properly on all architectures.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

net/xfrm_user: Use in_compat_syscall to deny compat syscalls

The code wants to prevent compat code from receiving messages.  Use
in_compat_syscall for this.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Steffen Klassert <steffen.klassert@secunet.com>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

net/sctp: Use in_compat_syscall for sctp_getsockopt_connectx3

SCTP unfortunately has a different ABI for SCTP_SOCKOPT_CONNECTX3
for 32-bit and 64-bit callers.  Use in_compat_syscall to correctly
distinguish them on all architectures.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Vlad Yasevich <vyasevich@gmail.com>
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

ext4: In ext4_dir_llseek, check syscall bitness directly

ext4 treats directory offsets differently for 32-bit and 64-bit
callers.  Check the caller type using in_compat_syscall, not
is_compat_task.  This changes behavior on SPARC slightly.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

staging/lustre: switch from is_compat_task to in_compat_syscall

AFAICT, lustre is trying to determine syscall bitness.  Use the new
accessor.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Oleg Drokin <oleg.drokin@intel.com>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

auditsc: for seccomp events, log syscall compat state using in_compat_syscall

Except on SPARC, this is what the code always did.  SPARC compat
seccomp was buggy, although the impact of the bug was limited
because SPARC 32-bit and 64-bit syscall numbers are the same.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Paul Moore <paul@paul-moore.com>
Cc: Eric Paris <eparis@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

ptrace: in PEEK_SIGINFO, check syscall bitness, not task bitness

Users of the 32-bit ptrace() ABI expect the full 32-bit ABI.  siginfo
translation should check ptrace() ABI, not caller task ABI.

This is an ABI change on SPARC.  Let's hope that no one relied on the old
buggy ABI.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

seccomp: check in_compat_syscall, not is_compat_task, in strict mode

Seccomp wants to know the syscall bitness, not the caller task bitness,
when it selects the syscall whitelist.

As far as I know, this makes no difference on any architecture, so it's
not a security problem.  (It generates identical code everywhere except
sparc, and, on sparc, the syscall numbering is the same for both ABIs.)

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sparc/syscall: fix syscall_get_arch

Sparc's syscall_get_arch was buggy: it returned the task arch, not the
syscall arch.  This could confuse seccomp and audit.

I don't think this is as bad for seccomp as it looks: sparc's 32-bit and
64-bit syscalls are numbered the same.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sparc-compat-provide-an-accurate-in_compat_syscall-implementation-fix-fix

update comment, per Andy

Cc: Andy Lutomirski <luto@kernel.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sparc-compat-provide-an-accurate-in_compat_syscall-implementation-fix

add comment, per Sam

Cc: Andy Lutomirski <luto@kernel.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

sparc/compat: rrovide an accurate in_compat_syscall implementation

On sparc64 compat-enabled kernels, any task can make 32-bit and 64-bit
syscalls.  is_compat_task returns true in 32-bit tasks, which does not
necessarily imply that the current syscall is 32-bit.

Provide an in_compat_syscall implementation that checks whether the
current syscall is compat.

As far as I know, sparc is the only architecture on which is_compat_task
checks the compat status of the task and on which the compat status of a
syscall can differ from the compat status of the task.  On x86,
is_compat_task checks the syscall type, not the task type.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Acked-by: David S. Miller <davem@davemloft.net>
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

compat: add in_compat_syscall to ask whether we're in a compat syscall

A lot of code currently abuses is_compat_task to determine this.

Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Cc: Clemens Ladisch <clemens@ladisch.de>
Cc: David Airlie <airlied@linux.ie>
Cc: David Herrmann <dh.herrmann@googlemail.com>
Cc: David Miller <davem@davemloft.net>
Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Cc: Eric Paris <eparis@redhat.com>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Ingo Molnar <mingo@elte.hu>
Acked-by: Jiri Kosina <jkosina@suse.cz>
Cc: Matt Fleming <matt@codeblueprint.co.uk>
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: Oded Gabbay <oded.gabbay@gmail.com>
Cc: Oleg Drokin <oleg.drokin@intel.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Paul Moore <paul@paul-moore.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Steffen Klassert <steffen.klassert@secunet.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vlad Yasevich <vyasevich@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

lib: Add CRC64 ECMA module

Add implementation of CRC64 ECMA checksum.

We have an IP Acceleration driver for Freescale network processors which
is using this CRC64.  However, it still needs some work in order for it to
become upstreamable.

Signed-off-by: Marian Chereji <marian.chereji@freescale.com>
Reviewed-by: Varvara Andrei-B21317 <andrei.varvara@freescale.com>
Reviewed-by: Fleming Andrew-AFLEMING <AFLEMING@freescale.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

mm/util.c: add kstrimdup()

kstrimdup() creates a whitespace-trimmed duplicate of the passed in
null-terminated string.  This is useful for strings coming from sysfs that
often include trailing whitespace due to user input.

Thanks to Joe Perches for this implementation.

Signed-off-by: Sebastian Capella <sebastian.capella@linaro.org>
Cc: Joe Perches <joe@perches.com>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

param-convert-some-on-off-users-to-strtobool fix

This converts a missed __setup return (and silences the build warning it
was causing).

Signed-off-by: Kees Cook <keescook@chromium.org>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

param: convert some "on"/"off" users to strtobool

This changes several users of manual "on"/"off" parsing to use strtobool.

Some side-effects:
- these uses will now parse y/n/1/0 meaningfully too
- the early_param uses will now bubble up parse errors

Signed-off-by: Kees Cook <keescook@chromium.org>
Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Cc: Amitkumar Karwar <akarwar@marvell.com>
Cc: Andy Shevchenko <andy.shevchenko@gmail.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Joe Perches <joe@perches.com>
Cc: Kalle Valo <kvalo@codeaurora.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Nishant Sarmukadam <nishants@marvell.com>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: Steve French <sfrench@samba.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

lib: add "on"/"off" support to kstrtobool

Add support for "on" and "off" when converting to boolean.

Signed-off-by: Kees Cook <keescook@chromium.org>
Cc: Amitkumar Karwar <akarwar@marvell.com>
Cc: Andy Shevchenko <andy.shevchenko@gmail.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Joe Perches <joe@perches.com>
Cc: Kalle Valo <kvalo@codeaurora.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nishant Sarmukadam <nishants@marvell.com>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: Steve French <sfrench@samba.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

lib: update single-char callers of strtobool()

Some callers of strtobool() were passing a pointer to unterminated
strings.  In preparation of adding multi-character processing to
kstrtobool(), update the callers to not pass single-character pointers,
and switch to using the new kstrtobool_from_user() helper where possible.

Signed-off-by: Kees Cook <keescook@chromium.org>
Cc: Amitkumar Karwar <akarwar@marvell.com>
Cc: Nishant Sarmukadam <nishants@marvell.com>
Cc: Kalle Valo <kvalo@codeaurora.org>
Cc: Steve French <sfrench@samba.org>
Cc: Andy Shevchenko <andy.shevchenko@gmail.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Joe Perches <joe@perches.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

lib: move strtobool() to kstrtobool()

Create the kstrtobool_from_user() helper and move strtobool() logic into
the new kstrtobool() (matching all the other kstrto* functions).  Provides
an inline wrapper for existing strtobool() callers.

Signed-off-by: Kees Cook <keescook@chromium.org>
Cc: Joe Perches <joe@perches.com>
Cc: Andy Shevchenko <andy.shevchenko@gmail.com>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Amitkumar Karwar <akarwar@marvell.com>
Cc: Nishant Sarmukadam <nishants@marvell.com>
Cc: Kalle Valo <kvalo@codeaurora.org>
Cc: Steve French <sfrench@samba.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

include/linux/unaligned: force inlining of byteswap operations

Sometimes gcc mysteriously doesn't inline
very small functions we expect to be inlined. See
    https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66122

With this .config:
http://busybox.net/~vda/kernel_config_OPTIMIZE_INLINING_and_Os,
the following functions get deinlined many times.
Examples of disassembly:

<get_unaligned_be16> (24 copies, 108 calls):
       66 8b 07                mov    (%rdi),%ax
       55                      push   %rbp
       48 89 e5                mov    %rsp,%rbp
       86 e0                   xchg   %ah,%al
       5d                      pop    %rbp
       c3                      retq

<get_unaligned_be32> (25 copies, 181 calls):
       8b 07                   mov    (%rdi),%eax
       55                      push   %rbp
       48 89 e5                mov    %rsp,%rbp
       0f c8                   bswap  %eax
       5d                      pop    %rbp
       c3                      retq

<get_unaligned_be64> (23 copies, 94 calls):
       48 8b 07                mov    (%rdi),%rax
       55                      push   %rbp
       48 89 e5                mov    %rsp,%rbp
       48 0f c8                bswap  %rax
       5d                      pop    %rbp
       c3                      retq

<put_unaligned_be16> (2 copies, 11 calls):
       89 f8                   mov    %edi,%eax
       55                      push   %rbp
       c1 ef 08                shr    $0x8,%edi
       c1 e0 08                shl    $0x8,%eax
       09 c7                   or     %eax,%edi
       48 89 e5                mov    %rsp,%rbp
       66 89 3e                mov    %di,(%rsi)

<put_unaligned_be32> (8 copies, 43 calls):
       55                      push   %rbp
       0f cf                   bswap  %edi
       89 3e                   mov    %edi,(%rsi)
       48 89 e5                mov    %rsp,%rbp
       5d                      pop    %rbp
       c3                      retq

<put_unaligned_be64> (26 copies, 157 calls):
       55                      push   %rbp
       48 0f cf                bswap  %rdi
       48 89 3e                mov    %rdi,(%rsi)
       48 89 e5                mov    %rsp,%rbp
       5d                      pop    %rbp
       c3                      retq

This patch fixes this via s/inline/__always_inline/.

It only affects arches with efficient unaligned access insns, such as x86.
(arched which lack such ops do not include linux/unaligned/access_ok.h)

Code size decrease after the patch is ~8.5k:

    text     data      bss       dec     hex filename
92197848 20826112 36417536 149441496 8e84bd8 vmlinux
92189231 20826144 36417536 149432911 8e82a4f vmlinux6_unaligned_be_after

Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Cc: Thomas Graf <tgraf@suug.ch>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

include/uapi/linux/byteorder, swab: force inlining of some byteswap operations

Sometimes gcc mysteriously doesn't inline
very small functions we expect to be inlined. See
    https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66122

With this .config:
http://busybox.net/~vda/kernel_config_OPTIMIZE_INLINING_and_Os,
the following functions get deinlined many times.
Examples of disassembly:

<get_unaligned_be16> (12 copies, 51 calls):
       66 8b 07                mov    (%rdi),%ax
       55                      push   %rbp
       48 89 e5                mov    %rsp,%rbp
       86 e0                   xchg   %ah,%al
       5d                      pop    %rbp
       c3                      retq

<get_unaligned_be32> (12 copies, 135 calls):
       8b 07                   mov    (%rdi),%eax
       55                      push   %rbp
       48 89 e5                mov    %rsp,%rbp
       0f c8                   bswap  %eax
       5d                      pop    %rbp
       c3                      retq

<get_unaligned_be64> (2 copies, 20 calls):
       48 8b 07                mov    (%rdi),%rax
       55                      push   %rbp
       48 89 e5                mov    %rsp,%rbp
       48 0f c8                bswap  %rax
       5d                      pop    %rbp
       c3                      retq

<__swab16p> (16 copies, 146 calls):
       55                      push   %rbp
       89 f8                   mov    %edi,%eax
       86 e0                   xchg   %ah,%al
       48 89 e5                mov    %rsp,%rbp
       5d                      pop    %rbp
       c3                      retq

<__swab32p> (43 copies, ~560 calls):
       55                      push   %rbp
       89 f8                   mov    %edi,%eax
       0f c8                   bswap  %eax
       48 89 e5                mov    %rsp,%rbp
       5d                      pop    %rbp
       c3                      retq

<__swab64p> (21 copies, 119 calls):
       55                      push   %rbp
       48 89 f8                mov    %rdi,%rax
       48 0f c8                bswap  %rax
       48 89 e5                mov    %rsp,%rbp
       5d                      pop    %rbp
       c3                      retq

<__swab32s> (6 copies, 47 calls):
       8b 07                   mov    (%rdi),%eax
       55                      push   %rbp
       48 89 e5                mov    %rsp,%rbp
       0f c8                   bswap  %eax
       89 07                   mov    %eax,(%rdi)
       5d                      pop    %rbp
       c3                      retq

This patch fixes this via s/inline/__always_inline/.
Code size decrease after the patch is ~4.5k:

    text     data      bss       dec     hex filename
92202377 20826112 36417536 149446025 8e85d89 vmlinux
92197848 20826112 36417536 149441496 8e84bd8 vmlinux5_swap_after

Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Cc: Thomas Graf <tgraf@suug.ch>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

include/asm-generic/atomic-long.h: force inlining of some atomic_long operations

Sometimes gcc mysteriously doesn't inline
very small functions we expect to be inlined. See
    https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66122

With this .config:
http://busybox.net/~vda/kernel_config_OPTIMIZE_INLINING_and_Os,
atomic_long_inc(), atomic_long_dec() and atomic_long_add()
functions get deinlined about 40 times. Examples of disassembly:

<atomic_long_inc> (21 copies, 147 calls):
       55                      push   %rbp
       48 89 e5                mov    %rsp,%rbp
       f0 48 ff 07             lock incq (%rdi)
       5d                      pop    %rbp
       c3                      retq

<atomic_long_dec> (4 copies, 14 calls) is similar to inc.

<atomic_long_add> (11 copies, 41 calls):
       55                      push   %rbp
       48 89 e5                mov    %rsp,%rbp
       f0 48 01 3e             lock add %rdi,(%rsi)
       5d                      pop    %rbp
       c3                      retq

This patch fixes this via s/inline/__always_inline/.
Code size decrease after the patch is ~1.3k:

    text     data      bss       dec     hex filename
92203657 20826112 36417536 149447305 8e86289 vmlinux
92202377 20826112 36417536 149446025 8e85d89 vmlinux4_atomiclong_after

Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Thomas Graf <tgraf@suug.ch>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

err.h: allow IS_ERR_VALUE to handle properly more types

Current implementation of IS_ERR_VALUE works correctly only with following
types:

- unsigned long,

- short, int, long.

Other types are handled incorrectly either on 32-bit either on 64-bit
either on both architectures.

The patch fixes it by comparing argument with MAX_ERRNO casted to
argument's type for unsigned types and comparing with zero for signed
types.  As a result all integer types bigger than char are handled
properly.

I have analyzed usage of IS_ERR_VALUE using coccinelle and in about 35
cases it is used incorrectly, ie it can hide errors depending of 32/64 bit
architecture.  Instead of fixing usage I propose to enhance the macro to
cover more types.

And just for the record: the macro is used 101 times with signed
variables, I am not sure if it should be preferred over simple comparison
"ret < 0", but the new version can do it as well.

And below list of detected potential errors:
drivers/char/mem.c:698:45-46: WARNING: incorrect argument type in IS_ERR_VALUE(( unsigned long long ) offset)
drivers/media/platform/soc_camera/atmel-isi.c:1089:21-22: WARNING: incorrect argument type in IS_ERR_VALUE(irq)
drivers/net/ethernet/freescale/fs_enet/mac-scc.c:149:36-37: WARNING: incorrect argument type in IS_ERR_VALUE(fep -> ring_mem_addr)
drivers/net/ethernet/freescale/ucc_geth.c:2237:48-49: WARNING: incorrect argument type in IS_ERR_VALUE(ugeth -> tx_bd_ring_offset [ j ])
drivers/net/ethernet/freescale/ucc_geth.c:2314:48-49: WARNING: incorrect argument type in IS_ERR_VALUE(ugeth -> rx_bd_ring_offset [ j ])
drivers/net/ethernet/freescale/ucc_geth.c:2524:44-45: WARNING: incorrect argument type in IS_ERR_VALUE(ugeth -> tx_glbl_pram_offset)
drivers/net/ethernet/freescale/ucc_geth.c:2544:45-46: WARNING: incorrect argument type in IS_ERR_VALUE(ugeth -> thread_dat_tx_offset)
drivers/net/ethernet/freescale/ucc_geth.c:2571:46-47: WARNING: incorrect argument type in IS_ERR_VALUE(ugeth -> send_q_mem_reg_offset)
drivers/net/ethernet/freescale/ucc_geth.c:2612:42-43: WARNING: incorrect argument type in IS_ERR_VALUE(ugeth -> scheduler_offset)
drivers/net/ethernet/freescale/ucc_geth.c:2659:54-55: WARNING: incorrect argument type in IS_ERR_VALUE(ugeth -> tx_fw_statistics_pram_offset)
drivers/net/ethernet/freescale/ucc_geth.c:2696:44-45: WARNING: incorrect argument type in IS_ERR_VALUE(ugeth -> rx_glbl_pram_offset)
drivers/net/ethernet/freescale/ucc_geth.c:2715:45-46: WARNING: incorrect argument type in IS_ERR_VALUE(ugeth -> thread_dat_rx_offset)
drivers/net/ethernet/freescale/ucc_geth.c:2736:54-55: WARNING: incorrect argument type in IS_ERR_VALUE(ugeth -> rx_fw_statistics_pram_offset)
drivers/net/ethernet/freescale/ucc_geth.c:2756:53-54: WARNING: incorrect argument type in IS_ERR_VALUE(ugeth -> rx_irq_coalescing_tbl_offset)
drivers/net/ethernet/freescale/ucc_geth.c:2822:44-45: WARNING: incorrect argument type in IS_ERR_VALUE(ugeth -> rx_bd_qs_tbl_offset)
drivers/net/ethernet/freescale/ucc_geth.c:2908:47-48: WARNING: incorrect argument type in IS_ERR_VALUE(ugeth -> exf_glbl_param_offset)
drivers/net/ethernet/freescale/ucc_geth.c:292:36-37: WARNING: incorrect argument type in IS_ERR_VALUE(init_enet_offset)
drivers/net/ethernet/freescale/ucc_geth.c:3042:39-40: WARNING: incorrect argument type in IS_ERR_VALUE(init_enet_pram_offset)
drivers/soc/fsl/qe/ucc_fast.c:271:60-61: WARNING: incorrect argument type in IS_ERR_VALUE(uccf -> ucc_fast_tx_virtual_fifo_base_offset)
drivers/soc/fsl/qe/ucc_fast.c:284:60-61: WARNING: incorrect argument type in IS_ERR_VALUE(uccf -> ucc_fast_rx_virtual_fifo_base_offset)
drivers/soc/fsl/qe/ucc_slow.c:186:38-39: WARNING: incorrect argument type in IS_ERR_VALUE(uccs -> us_pram_offset)
drivers/soc/fsl/qe/ucc_slow.c:213:38-39: WARNING: incorrect argument type in IS_ERR_VALUE(uccs -> rx_base_offset)
drivers/soc/fsl/qe/ucc_slow.c:224:38-39: WARNING: incorrect argument type in IS_ERR_VALUE(uccs -> tx_base_offset)
drivers/tty/serial/clps711x.c:471:29-30: WARNING: incorrect argument type in IS_ERR_VALUE(s -> port . irq)
drivers/tty/serial/digicolor-usart.c:485:30-31: WARNING: incorrect argument type in IS_ERR_VALUE(dp -> port . irq)
drivers/usb/gadget/udc/fsl_qe_udc.c:2369:26-27: WARNING: incorrect argument type in IS_ERR_VALUE(tmp_addr)
drivers/video/fbdev/exynos/exynos_mipi_dsi.c:406:27-28: WARNING: incorrect argument type in IS_ERR_VALUE(dsim -> irq)
net/ipv4/netfilter/arp_tables.c:1427:39-40: WARNING: incorrect argument type in IS_ERR_VALUE(iter1 -> counters . pcnt)
net/ipv4/netfilter/arp_tables.c:530:34-35: WARNING: incorrect argument type in IS_ERR_VALUE(e -> counters . pcnt)
net/ipv4/netfilter/ip_tables.c:1614:34-35: WARNING: incorrect argument type in IS_ERR_VALUE(e -> counters . pcnt)
net/ipv4/netfilter/ip_tables.c:674:34-35: WARNING: incorrect argument type in IS_ERR_VALUE(e -> counters . pcnt)
net/ipv6/netfilter/ip6_tables.c:1624:34-35: WARNING: incorrect argument type in IS_ERR_VALUE(e -> counters . pcnt)
net/ipv6/netfilter/ip6_tables.c:687:34-35: WARNING: incorrect argument type in IS_ERR_VALUE(e -> counters . pcnt)
drivers/net/ethernet/freescale/fs_enet/mac-fcc.c:110:35-36: WARNING: unknown argument type in IS_ERR_VALUE(fpi -> dpram_offset)

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

usb: common: convert to use match_string() helper

The new helper returns index of the mathing string in an array.  We would
use it here.

Signed-off-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

ide: hpt366: convert to use match_string() helper

The new helper returns index of the mathing string in an array. We would use it
here.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

ata: hpt366: convert to use match_string() helper

The new helper returns index of the mathing string in an array.  We would
use it here.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

power: ab8500: convert to use match_string() helper

The new helper returns index of the mathing string in an array.  We would
use it here.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

power: charger_manager: convert to use match_string() helper

The new helper returns index of the mathing string in an array. We would use it
here.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Sebastian Reichel <sre@kernel.org>
Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

drm/edid: convert to use match_string() helper

The new helper returns index of the mathing string in an array.  We would
use it here.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: David Airlie <airlied@linux.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

pinctrl: convert to use match_string() helper

The new helper returns index of the mathing string in an array.  We would
use it here.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

device-property-convert-to-use-match_string-helper fix

Since match_string() returns -EINVAL let's convert it to -ENODATA as designed
by device property API.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

device property: convert to use match_string() helper

The new helper returns index of the mathing string in an array.  We would
use it here.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

lib-string-introduce-match_string-helper fix

Fix return code to be -EINVAL instead of -ENODATA.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

lib/string: introduce match_string() helper

Occasionally we have to search for an occurrence of a string in an array
of strings.  Make a simple helper for that purpose.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Cc: David Airlie <airlied@linux.ie>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Sebastian Reichel <sre@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

arm64: switch to relative exception tables

Instead of using absolute addresses for both the exception location and
the fixup, use offsets relative to the exception table entry values.  Not
only does this cut the size of the exception table in half, it is also a
prerequisite for KASLR, since absolute exception table entries are subject
to dynamic relocation, which is incompatible with the sorting of the
exception table that occurs at build time.

This patch also introduces the _ASM_EXTABLE preprocessor macro (which
exists on x86 as well) and its _asm_extable assembly counterpart, as
shorthands to emit exception table entries.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

ia64/extable: use generic search and sort routines

Replace the arch specific versions of search_extable() and sort_extable()
with calls to the generic ones, which now support relative exception
tables as well.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Acked-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

x86/extable: use generic search and sort routines

Replace the arch specific versions of search_extable() and sort_extable()
with calls to the generic ones, which now support relative exception
tables as well.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Acked-by: H. Peter Anvin <hpa@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

s390/extable: use generic search and sort routines

Replace the arch specific versions of search_extable() and sort_extable()
with calls to the generic ones, which now support relative exception
tables as well.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

alpha/extable: use generic search and sort routines

Replace the arch specific versions of search_extable() and sort_extable()
with calls to the generic ones, which now support relative exception
tables as well.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Acked-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

extable: add support for relative extables to search and sort routines

There are currently four architectures (x86, ia64, alpha and s390) whose
user-access exception tables are relative to the table entry address
rather than absolute.  Each of these architectures has its own
search_extable() and sort_extable() implementation, which are not only
mostly identical to each other, but also deviate very little from the
generic absolute implementations in lib/extable.c that they override.

So before making arm64 the fifth architecture that reimplements this,
let's refactor the existing code so that all of these architectures use
common code for searching and sorting the relative extables.  Archs may
set ARCH_HAS_RELATIVE_EXTABLE to indicate that the table consists of a
pair of relative ints, and may define swap_ex_entry_fixup() if the fixup
member needs special treatment in the swapping step of the sorting routine
(such as alpha).

This patch (of 6):

Add support to the generic search_extable() and sort_extable()
implementations for dealing with exception table entries whose fields
contain relative offsets rather than absolute addresses.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Acked-by: Helge Deller <deller@gmx.de>
Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: H. Peter Anvin <hpa@linux.intel.com>
Acked-by: Tony Luck <tony.luck@intel.com>
Acked-by: Will Deacon <will.deacon@arm.com>
Acked-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>