Simplify the memory detection code a bit by removing the CHUNK_OLDMEM
and CHUNK_CRASHK memory types.
They are not needed. Everything that is needed is a mechanism to
insert holes into the detected memory.
Reviewed-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Add a notifier for kvm to get control before a page table entry is
invalidated. The notifier is only called for ptes of an address space
with pgstes that have been explicitly marked to require notification.
Kvm will use this to get control before prefix pages of virtual CPU
are unmapped.
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Commit 0c2c99b1 "memory hotplug: Allow memory blocks to span
multiple memory sections" introduced a weak memory_block_size_bytes()
function which can be used to set the size of a memory block as
seen in sysfs.
Provide an s390 specific override which makes sure that each
memory block has at least a size of 256MB or the increment size of
of a memory increment, whatever is larger.
This way we can make sure that the number of memory sysfs objects
doesn't explode for very large memory configurations.
Reported-by: NGerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The current memory detection loop will detect all present memory of
a machine. This is true even if the user specified the "mem=" parameter
on the kernel command line.
This can be a problem since the memory detection may cause a fully
populated host page table for the guest, even for those parts of the
memory that the guest will never use afterwards.

So fix this and only detect memory up to a user supplied "mem=" limit
if specified.
Reported-by: NMichael Johanssen <johanssn@de.ibm.com>
Reviewed-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

When in kdump mode the kernel may access only the first couple of
megabytes for execution, the rest contains the dump. However
the size of the bitmap used by the bootmem allocator was calculated
for the whole amount of memory of the machine.

For very large machines this can lead to the situation that the kdump
kernel will not come up because not enough memory is available.

So fix this and calculate the size of the bitmap only for the piece
of memory that the kdump kernel actually uses.

Call reserve_oldmem() before setup_memory_end() so that the memory_chunk
array already has been updated with respect to oldmem chunks.
Afterwards setup_memory_end() will ignore those chunks.
Reviewed-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

The variable real_memory_size has odd semantics and has been used in
a broken way by e.g. the old kvm code.
Therefore get rid of it before anybody else makes use of it.
Reviewed-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Add sanity check: verify if the passed in array resides in vmalloc space.
If so print a warning and return to caller.
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

When disabling and enabling interrupts we must tell lockdep.
So use local_irq_save()/restore() to disable and enable interrupts.
The DAT disabling/enabling get handled separately now.
Note: we may not call trace_hardirqs_on() with DAT disabled, since
the generic code may access vmalloc'ed data structures.
Reported-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Use the 'ipldev' and 'condev' cio_ignore keywords to setup the
command line for zfcpdump.
Reviewed-by: NMichael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: NSebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Add "fallthrough" comments so nobody wonders if a break statement is missing.
Reported-by: NJoe Perches <joe@perches.com>
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: NMartin Schwidefsky <schwidefsky@de.ibm.com>

Clean up the pseries scanlog driver's use of procfs:

 (1) Don't need to save the proc_dir_entry pointer as we have the filename to
     remove with.

 (2) Save the scan log buffer pointer in a static variable (there is only one
     of it) and don't save it in the PDE (which doesn't have a destructor).
Signed-off-by: NDavid Howells <dhowells@redhat.com>
cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
cc: Paul Mackerras <paulus@samba.org>
cc: linuxppc-dev@lists.ozlabs.org
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Clean up some of the problems with the rtas_flash driver:

 (1) It shouldn't fiddle with the internals of the procfs filesystem (altering
     pde->count).

 (2) If pid namespaces are in effect, then you can get multiple inodes
     connected to a single pde, thereby rendering the pde->count > 2 test
     useless.

 (3) The pde->count fudging doesn't work for forked, dup'd or cloned file
     descriptors, so add static mutexes and use them to wrap access to the
     driver through read, write and release methods.

 (4) The driver can only handle one device, so allocate most of the data
     previously attached to the pde->data as static variables instead (though
     allocate the validation data buffer with kmalloc).

 (5) We don't need to save the pde pointers as long as we have the filenames
     available for removal.

 (6) Don't try to multiplex what the update file read method does based on the
     filename.  Instead provide separate file ops and split the function.

Whilst we're at it, tabulate the procfile information and loop through it when
creating or destroying them rather than manually coding each one.

[Folded fixes from Vasant Hegde <hegdevasant@linux.vnet.ibm.com>]
Signed-off-by: NDavid Howells <dhowells@redhat.com>
cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
cc: Paul Mackerras <paulus@samba.org>
cc: Anton Blanchard <anton@samba.org>
cc: linuxppc-dev@lists.ozlabs.org
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Supply accessor functions to set attributes in proc_dir_entry structs.

The following are supplied: proc_set_size() and proc_set_user().
Signed-off-by: NDavid Howells <dhowells@redhat.com>
Acked-by: NMauro Carvalho Chehab <mchehab@redhat.com>
cc: linuxppc-dev@lists.ozlabs.org
cc: linux-media@vger.kernel.org
cc: netdev@vger.kernel.org
cc: linux-wireless@vger.kernel.org
cc: linux-pci@vger.kernel.org
cc: netfilter-devel@vger.kernel.org
cc: alsa-devel@alsa-project.org
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

With physical offsets, pa<->va translations aren't just based
on PAGE_OFFSET anymore.
Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

The Kconfig entries for HEXAGON_VM and HEXAGON_ANGEL_TRAPS were added,
together with the configuration and makefiles for the Hexagon
architecture, in v3.2. They have never been used. They can safely be
removed.
Signed-off-by: NPaul Bolle <pebolle@tiscali.nl>
[rkuo@codeaurora.org: adjust for line changes in Kconfig]
Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

CONFIG_GENERIC_FIND_NEXT_BIT was removed in v3.0, but reappeared in two
architectures. Remove it again.
Signed-off-by: NPaul Bolle <pebolle@tiscali.nl>
Acked-by: NJonas Bonn <jonas@southpole.se>
Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Based on feedback from Al Viro; previous-stack-pointer and
user reg for same should always be kept consistent.
Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Check return coming out of check_work_pending, and if copy_thread
passed us a function in r24, call it.  Based on feedback from Al
Viro.
Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Uncached on our architecture can still have side effects
such as canceled/replayed transactions; device type prevents
this.
Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Hardware single-step is only available on v4 and later
architectures.
Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Add multi-reg-write and unaligned-PC exceptions.
Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Use the exact number of pages needed to be mapped pre-VA-jump,
then map 896MB afterwards, which the arch mem init will fix up.
Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

This fixes the return value of sigreturn and moves the work pending check
into a c routine for readability and fixes the loop for multiple pending
signals.  Based on feedback from Al Viro.
Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Add support for a couple new v4+ registers, along with
newer save/restore pt_regs.
Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>

Signed-off-by: NRichard Kuo <rkuo@codeaurora.org>