Update the association of a memory section with a numa node that
occurs during hotplug add of a memory section. This adds a check in
the hot_add_scn_to_nid() routine for the
ibm,dynamic-reconfiguration-memory node in the device tree. If
present the new hot_add_drconf_scn_to_nid() routine is invoked, which
can properly parse the ibm,dynamic-reconfiguration-memory node of the
device tree and make the proper numa node associations.
This also introduces the valid_hot_add_scn() routine as a helper
function for code that is common to the hot_add_scn_to_nid() and
hot_add_drconf_scn_to_nid() routines.
Signed-off-by: Nathan Fontenot <nfont@austin.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This splits off several pieces of code that parse the
ibm,dynamic-reconfiguration-memory node of the device tree into separate
helper routines. This is in preparation for the next commit that will
use these helper routines. There are no functional changes in this patch.
Signed-off-by: Nathan Fontenot <nfont@austin.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Currently the kernel fails to build with the above config options with:
CC arch/powerpc/mm/mem.o
arch/powerpc/mm/mem.c: In function 'arch_add_memory':
arch/powerpc/mm/mem.c:130: error: implicit declaration of function 'create_section_mapping'
This explicitly includes asm/sparsemem.h in arch/powerpc/mm/mem.c and
moves the guards in include/asm-powerpc/sparsemem.h to protect the
SPARSEMEM specific portions only.
Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Dave Kleikamp <shaggy@linux.vnet.ibm.com>
Cc: Jon Tollefson <kniht@linux.vnet.ibm.com>
Cc: Adam Litke <agl@us.ibm.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
At present, if we have a kernel with a 64kB page size, and some
process maps something that has to be mapped with 4kB pages (such as a
cache-inhibited mapping on POWER5+, or the eHCA infiniband queue-pair
pages), we change the process to use 4kB pages everywhere. This hurts
the performance of HPC programs that access eHCA from userspace.
With this patch, the kernel will only demote the slice(s) containing
the eHCA or cache-inhibited mappings, leaving the remaining slices
able to use 64kB hardware pages.
This also changes the slice_get_unmapped_area code so that it is
willing to place a 64k-page mapping into (or across) a 4k-page slice
if there is no better alternative, i.e. if the program specified
MAP_FIXED or if there is not sufficient space available in slices that
are either empty or already have 64k-page mappings in them.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
While working on the 36-bit physical support, I noticed that there
was exactly one line of code that actually referenced the bitfields.
So I got rid of them and redefined ppc_bat as a struct of 2 u32's:
batu and batl. I also got rid of the previous union that held the
bitfield structs and a word representation of the batu/l values.
This seems like a nicer solution than adding in a bunch of
new bitfields to support extended bat addressing that would never
get used, and just leaving the struct as-is would have been
incomplete in the face of large physical addressing.
Signed-off-by: Becky Bruce <becky.bruce@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Currently, the physical address is an unsigned long, but it should
be phys_addr_t in set_bat, [v/p]_mapped_by_bat. Also, create a
macro that can convert a large physical address into the correct
format for programming the BAT registers.
Signed-off-by: Becky Bruce <becky.bruce@freescale.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This frees a PTE bit when using 64K pages on ppc64. This is done
by getting rid of the separate _PAGE_HASHPTE bit. Instead, we just test
if any of the 16 sub-page bits is set. For non-combo pages (ie. real
64K pages), we set SUB0 and the location encoding in that field.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
When we demote a slice from 64k to 4k, and we are about to insert an
HPTE for a 4k subpage and we notice that there is an existing 64k
HPTE, we first invalidate that HPTE before inserting the new 4k
subpage HPTE. Since the bits that encode which hash bucket the old
HPTE was in overlap with the bits that encode which of the 16 subpages
have HPTEs, we need to clear out the subpage HPTE-present bits before
starting to insert HPTEs for the 4k subpages. If we don't do that, we
can erroneously think that a subpage already has an HPTE when it
doesn't.
That in itself wouldn't be such a problem except that when we go to
update the HPTE that we think is present on machines with a
hypervisor, the hypervisor can tell us that the HPTE we think is there
is actually there even though it isn't, which can lead to a process
getting stuck in a loop, continually faulting. The reason for the
confusion is that the AVPN (abbreviated virtual page number) we are
looking for in the HPTE for a 4k subpage can actually match the AVPN
in a stale HPTE for another 64k page. For example, the HPTE for
the 4k subpage at 0x84000f000 will be in the same hash bucket and have
the same AVPN as the HPTE for the 64k page at 0x8400f0000.
This fixes the code to clear out the subpage HPTE-present bits.
Signed-off-by: Paul Mackerras <paulus@samba.org>
The ehea driver was recently changed[1] to use walk_memory_resource() to
detect the system's memory layout. However, walk_memory_resource() is
available only when memory hotplug is enabled. So CONFIG_EHEA was
made to depend on MEMORY_HOTPLUG [2], but it is inappropriate for a
network driver to have such a dependency.
Make the declaration of walk_memory_resource() and its powerpc
implementation (ehea is powerpc-specific) unconditionally available.
[1] 48cfb14f8b
"ehea: Add DLPAR memory remove support"
[2] fb7b6ca2b6
"ehea: Add dependency to Kconfig"
Signed-off-by: Nathan Lynch <ntl@pobox.com>
Acked-by: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
__set_fixmap() in pgtable_32.c currently fails to compile if
STRICT_MM_TYPECHECKS is defined. This fixes it.
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This removes a CVS keyword that wasn't updated for a long time from a
comment.
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This changes vmemmap to use a different region (region 0xf) of the
address space, and to configure the page size of that region
dynamically at boot.
The problem with the current approach of always using 16M pages is that
it's not well suited to machines that have small amounts of memory such
as small partitions on pseries, or PS3's.
In fact, on the PS3, failure to allocate the 16M page backing vmmemmap
tends to prevent hotplugging the HV's "additional" memory, thus limiting
the available memory even more, from my experience down to something
like 80M total, which makes it really not very useable.
The logic used by my match to choose the vmemmap page size is:
- If 16M pages are available and there's 1G or more RAM at boot,
use that size.
- Else if 64K pages are available, use that
- Else use 4K pages
I've tested on a POWER6 (16M pages) and on an iSeries POWER3 (4K pages)
and it seems to work fine.
Note that I intend to change the way we organize the kernel regions &
SLBs so the actual region will change from 0xf back to something else at
one point, as I simplify the SLB miss handler, but that will be for a
later patch.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Somewhere along the way (e28f7faf05,
"Four level pagetables for ppc64") we ended up with duplicate
definitions for pte_freelist_cur and pte_freelist_force_free.
Somehow this compiles, but it would be better to just have one
definition for each.
The two definitions we end up with can be static too!
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
... instead of having extern declarations in a .c file.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Make two vmemmap helpers static in init_64.c
Make stab variables static in stab.c
Make psize defs static in hash_utils_64.c
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
... instead of having an extern declaration in a .c file.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This fixes a regression reported by Kamalesh Bulabel where a POWER4
machine would crash because of an SLB miss at a point where the SLB
miss exception was unrecoverable. This regression is tracked at:
http://bugzilla.kernel.org/show_bug.cgi?id=10082
SLB misses at such points shouldn't happen because the kernel stack is
the only memory accessed other than things in the first segment of the
linear mapping (which is mapped at all times by entry 0 of the SLB).
The context switch code ensures that SLB entry 2 covers the kernel
stack, if it is not already covered by entry 0. None of entries 0
to 2 are ever replaced by the SLB miss handler.
Where this went wrong is that the context switch code assumes it
doesn't have to write to SLB entry 2 if the new kernel stack is in the
same segment as the old kernel stack, since entry 2 should already be
correct. However, when we start up a secondary cpu, it calls
slb_initialize, which doesn't set up entry 2. This is correct for
the boot cpu, where we will be using a stack in the kernel BSS at this
point (i.e. init_thread_union), but not necessarily for secondary
cpus, whose initial stack can be allocated anywhere. This doesn't
cause any immediate problem since the SLB miss handler will just
create an SLB entry somewhere else to cover the initial stack.
In fact it's possible for the cpu to go quite a long time without SLB
entry 2 being valid. Eventually, though, the entry created by the SLB
miss handler will get overwritten by some other entry, and if the next
access to the stack is at an unrecoverable point, we get the crash.
This fixes the problem by making slb_initialize create a suitable
entry for the kernel stack, if we are on a secondary cpu and the stack
isn't covered by SLB entry 0. This requires initializing the
get_paca()->kstack field earlier, so I do that in smp_create_idle
where the current field is initialized. This also abstracts a bit of
the computation that mk_esid_data in slb.c does so that it can be used
in slb_initialize.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Arrange for a syntax check to always be done on the powerpc/mm/slb.c
DBG() macro by defining it to pr_debug() for non-debug builds.
Also, fix these related compile warnings:
slb.c:273: warning: format '%04x' expects type 'unsigned int', but argument 2 has type 'long unsigned int
slb.c:274: warning: format '%04x' expects type 'unsigned int', but argument 2 has type 'long unsigned int'
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Provide walk_memory_resource() for 64-bit powerpc. PowerPC maintains
logical memory region mapping in the lmb.memory structure. Walk
through these structures and do the callbacks for the contiguous
chunks.
Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
All architectures use an effectively identical definition of online_page(), so
just make it common code. x86-64, ia64, powerpc and sh are actually
identical; x86-32 is slightly different.
x86-32's differences arise because it puts its hotplug pages in the highmem
zone. We can handle this in the generic code by inspecting the page to see if
its in highmem, and update the totalhigh_pages count appropriately. This
leaves init_32.c:free_new_highpage with a single caller, so I folded it into
add_one_highpage_init.
I also removed an incorrect comment referring to the NUMA case; any NUMA
details have already been dealt with by the time online_page() is called.
[akpm@linux-foundation.org: fix indenting]
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Acked-by: Dave Hansen <dave@linux.vnet.ibm.com>
Reviewed-by: KAMEZAWA Hiroyuki <kamez.hiroyu@jp.fujitsu.com>
Tested-by: KAMEZAWA Hiroyuki <kamez.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: Christoph Lameter <clameter@sgi.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Use (31-THREAD_SHIFT) to get to thread_info from stack pointer. This makes
the code a bit easier to read and more robust if we ever change THREAD_SHIFT.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The fixmap code from x86 allows us to have compile time virtual addresses
that we change the physical addresses of at run time.
This is useful for applications like kmap_atomic, PCI config that is done
via direct memory map, kexec/kdump.
We got ride of CONFIG_HIGHMEM_START as we can now determine a more optimal
location for PKMAP_BASE based on where the fixmap addresses start and
working back from there.
Additionally, the kmap code in asm-powerpc/highmem.h always had debug
enabled. Moved to using CONFIG_DEBUG_HIGHMEM to determine if we should
have the extra debug checking.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Added support to allow an 85xx kernel to be run from a non-zero physical
address (useful for cooperative asymmetric multiprocessing situations and
kdump). The support can be configured at compile time by setting
CONFIG_PAGE_OFFSET, CONFIG_KERNEL_START, and CONFIG_PHYSICAL_START as
desired.
Alternatively, the kernel build can set CONFIG_RELOCATABLE. Setting this
config option causes the kernel to determine at runtime the physical
addresses of CONFIG_PAGE_OFFSET and CONFIG_KERNEL_START. If
CONFIG_RELOCATABLE is set, then CONFIG_PHYSICAL_START has no meaning.
However, CONFIG_PHYSICAL_START will always be used to set the LOAD program
header physical address field in the resulting ELF image.
Currently we are limited to running at a physical address that is a
multiple of 256M. This is due to how we map TLBs to cover
lowmem. This should be fixed to allow 64M or maybe even 16M alignment
in the future. It is considered an error to try and run a kernel at a
non-aligned physical address.
All the magic for this support is accomplished by proper initialization
of the kernel memory subsystem and use of ARCH_PFN_OFFSET.
The use of ARCH_PFN_OFFSET only affects normal memory and not IO mappings.
ioremap uses map_page and isn't affected by ARCH_PFN_OFFSET.
/dev/mem continues to allow access to any physical address in the system
regardless of how CONFIG_PHYSICAL_START is set.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
numa.c requires routines declared in linux/of.h, so should include it.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Remove the __max_memory variable, as it is not referenced anywhere
in the tree besides some code in arch/ppc.
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
When we moved to arch/powerpc we actively tried to avoid using the
ppc_md.setup_io_mappings(). Currently no board ports use it so let's
remove it to avoid any new boards using it.
Also, remove early_serial_map() since we don't even have a call out for
it in arch/powerpc.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
We always use __initial_memory_limit as an address so rename it
to be clear.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
* Determine the RPN we are running the kernel at runtime rather
than using compile time constant for initial TLB
* Cleanup adjust_total_lowmem() to respect memstart_addr and
be a bit more clear on variables that are sizes vs addresses.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
total_lowmem represents the amount of low memory, not the physical
address that low memory ends at. If the start of memory is at 0 it
happens that total_lowmem can be used as both the size and the address
that lowmem ends at (or more specifically one byte beyond the end).
To make the code a bit more clear and deal with the case when the start of
memory isn't at physical 0, we introduce lowmem_end_addr that represents
one byte beyond the last physical address in the lowmem region.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
A number of users of PPC_MEMSTART (40x, ppc_mmu_32) can just always
use 0 as we don't support booting these kernels at non-zero physical
addresses since their exception vectors must be at 0 (or 0xfffx_xxxx).
For the sub-arches that support relocatable interrupt vectors
(book-e), it's reasonable to have memory start at a non-zero physical
address. For those cases use the variable memstart_addr instead of
the #define PPC_MEMSTART since the only uses of PPC_MEMSTART are for
initialization and in the future we can set memstart_addr at runtime
to have a relocatable kernel.
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This eliminates a warning in builds that don't define
CONFIG_MEMORY_HOTPLUG.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
hash_page_sync() takes and releases the low level mmu hash
lock in order to sync with other processors disposing of page
tables. Because that lock can be needed to service hash misses
triggered by interrupt handlers, taking it must be done with
interrupts off. However, hash_page_sync() appears to be called
with interrupts enabled, thus causing occasional deadlocks.
We fix it by making sure hash_page_sync() masks interrupts while
holding the lock.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
show_mem() has no need to print the amount of free swap space manually
because show_free_areas() does this already and is called by the
former.
The two outputs only differ in text formatting:
printk("Free swap = %lukB\n", ...);
printk("Free swap: %6ldkB\n", ...);
Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
__FUNCTION__ is gcc-specific, use __func__
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
WARNING: vmlinux.o(.text+0xb41b0): Section mismatch in reference from the
function .add_memory() to the function .devinit.text:.arch_add_memory()
The function .add_memory() references
the function __devinit .arch_add_memory().
This is often because .add_memory lacks a __devinit
annotation or the annotation of .arch_add_memory is wrong.
arch_add_memory() is also not __devinit on other architectures
Signed-off-by: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
If the platform doesn't support hpte_removebolted(), gracefully
return failure rather than success.
Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
On pSeries, the hypervisor doesn't let us map in the eHEA ethernet
adapter using 64k pages, and thus the ehea driver will fail if 64k
pages are configured. This works around the problem by always
using 4k pages for ioremap on pSeries (but not on other platforms).
A better fix would be to check whether the partition could ever
have an eHEA adapter, and only force 4k pages if it could, but this
will do for 2.6.25.
This is based on an earlier patch by Tony Breeds.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Since the PMU is an NMI now, it can come at any time we are only soft
disabled. We must hard disable around the two places we allow the kernel
stack SLB and r1 to go out of sync. Otherwise the PMU exception can
force a kernel stack SLB into another slot, which can lead to it
getting evicted, which can lead to a nasty unrecoverable SLB miss
in the exception entry code.
Signed-off-by: Anton Blanchard <anton@samba.org>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
My recent hack to allocate the hash table under 1GB on cell was poorly
tested, *cough*. It turns out on blades with large amounts of memory we
fail to allocate the hash table at all. This is because RTAS has been
instantiated just below 768MB, and 0-x MB are used by the kernel,
leaving no areas that are both large enough and also naturally-aligned.
For the cell IOMMU hack the page tables must be under 2GB, so use that
as the limit instead. This has been tested on real hardware and boots
happily.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
For memory remove, we need to clean up htab mappings for the
section of the memory we are removing.
This implements support for removing htab bolted mappings for pSeries
logical partitions. Other sub-archs may need to implement similar
functionality for hotplug memory remove to work on them.
Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
* 'for-2.6.25' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc:
[POWERPC] Add arch-specific walk_memory_remove() for 64-bit powerpc
[POWERPC] Enable hotplug memory remove for 64-bit powerpc
[POWERPC] Add remove_memory() for 64-bit powerpc
[POWERPC] Make cell IOMMU fixed mapping printk more useful
[POWERPC] Fix potential cell IOMMU bug when switching back to default DMA ops
[POWERPC] Don't enable cell IOMMU fixed mapping if there are no dma-ranges
[POWERPC] Fix cell IOMMU null pointer explosion on old firmwares
[POWERPC] spufs: Fix timing dependent false return from spufs_run_spu
[POWERPC] spufs: No need to have a runnable SPU for libassist update
[POWERPC] spufs: Update SPU_Status[CISHP] in backing runcntl write
[POWERPC] spufs: Fix state_mutex leaks
[POWERPC] Disable G5 NAP mode during SMU commands on U3
