With the previous patch to switch to 64-bit mode after returning from
RTAS and before doing any memory accesses, the RMA limit need not be
clamped to 1GB to avoid RTAS bugs.
Keep the 1GB limit for older firmware (although this is more of a kernel
concern than RTAS), and remove it starting with POWER9.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
With the previous patch to switch to 64-bit mode after returning from
RTAS and before doing any memory accesses, the RMA limit need not be
clamped to 1GB to avoid RTAS bugs.
Keep the 1GB limit for older firmware (although this is more of a kernel
concern than RTAS), and remove it starting with POWER9.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The radix guest is not subject to the paravirtualized HPT VRMA limit,
so remove that from ppc64_rma_size calculation for that platform.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This removes the RMA limit on powernv platform, which constrains
early allocations such as PACAs and stacks. There are still other
restrictions that must be followed, such as bolted SLB limits, but
real mode addressing has no constraints.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
There are several cases outside the normal address space management
where a CPU's entire local TLB is to be flushed:
1. Booting the kernel, in case something has left stale entries in
the TLB (e.g., kexec).
2. Machine check, to clean corrupted TLB entries.
One other place where the TLB is flushed, is waking from deep idle
states. The flush is a side-effect of calling ->cpu_restore with the
intention of re-setting various SPRs. The flush itself is unnecessary
because in the first case, the TLB should not acquire new corrupted
TLB entries as part of sleep/wake (though they may be lost).
This type of TLB flush is coded inflexibly, several times for each CPU
type, and they have a number of problems with ISA v3.0B:
- The current radix mode of the MMU is not taken into account, it is
always done as a hash flushn For IS=2 (LPID-matching flush from host)
and IS=3 with HV=0 (guest kernel flush), tlbie(l) is undefined if
the R field does not match the current radix mode.
- ISA v3.0B hash must flush the partition and process table caches as
well.
- ISA v3.0B radix must flush partition and process scoped translations,
partition and process table caches, and also the page walk cache.
So consolidate the flushing code and implement it in C and inline asm
under the mm/ directory with the rest of the flush code. Add ISA v3.0B
cases for radix and hash, and use the radix flush in radix environment.
Provide a way for IS=2 (LPID flush) to specify the radix mode of the
partition. Have KVM pass in the radix mode of the guest.
Take out the flushes from early cputable/dt_cpu_ftrs detection hooks,
and move it later in the boot process after, the MMU registers are set
up and before relocation is first turned on.
The TLB flush is no longer called when restoring from deep idle states.
This was not be done as a separate step because booting secondaries
uses the same cpu_restore as idle restore, which needs the TLB flush.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Trap numbers can have extra bits at the bottom that need to
be filtered out. There are a few cases where we don't do that.
It's possible that we got lucky but better safe than sorry.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
When CONFIG_SWAP is set, the TLB miss handlers have to also take
into account _PAGE_ACCESSED flag. At the moment it is done by
anding _PAGE_ACCESSED into _PAGE_PRESENT using 3 instructions.
This patch uses APG for handling _PAGE_ACCESSED, allowing to
just copy _PAGE_ACCESSED bit into APG field, hence reducing the
action to a single instruction.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
As Linux kernel separates KERNEL and USER address spaces, there is
therefore no need to flag USER access at page level.
Today, the 8xx TLB handlers already handle user access in the L1 entry
through Access Protection Groups, it is then natural to move the user
access handling at PMD level once _PAGE_NA allows to handle PAGE_NONE
protection without _PAGE_USER
In the mean time, as we free up one bit in the PTE, we can use it to
include SPS (page size flag) in the PTE and avoid handling it at every
TLB miss hence removing special handling based on compiled page size.
For _PAGE_EXEC, we rework it to use PP PTE bits, avoiding the copy
of _PAGE_EXEC bit into the L1 entry. Unfortunatly we are not
able to put it at the correct location as it conflicts with
NA/RO/RW bits for data entries.
Upper bits of APG in L1 entry overlap with PMD base address. In
order to avoid having to filter that out, we set up all groups so that
upper bits can have any value.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Today, PAGE_NONE is defined as a page not having _PAGE_USER.
In some circunstances, when the CPU supports it, it might be
better to be able to flag a page with NO ACCESS.
In a following patch, the 8xx will switch user access being flagged
in the PMD, therefore it will not be possible anymore to use
_PAGE_USER as a way to flag a page with no access.
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
commit ac29c64089 ("powerpc/mm: Replace _PAGE_USER with
_PAGE_PRIVILEGED") introduced _PAGE_PRIVILEGED for BOOK3S/64
This patch generalises _PAGE_PRIVILEGED for all CPUs, allowing
to have either _PAGE_PRIVILEGED or _PAGE_USER or both.
PPC_8xx has a _PAGE_SHARED flag which is set for and only for
all non user pages. Lets rename it _PAGE_PRIVILEGED to remove
confusion as it has nothing to do with Linux shared pages.
On BookE, there's a _PAGE_BAP_SR which has to be set for kernel
pages: defining _PAGE_PRIVILEGED as _PAGE_BAP_SR will make
this generic
Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The Power Hypervisor has introduced a new device tree format for
the property describing the dynamic reconfiguration LMBs for a system,
ibm,dynamic-memory-v2. This new format condenses the size of the
property, especially on large memory systems, by reporting sets
of LMBs that have the same properties (flags and associativity array
index).
This patch updates the powerpc/mm/drmem.c code to provide routines
that can parse the new device tree format during the walk_drmem_lmb*
routines used during boot, the creation of the LMB array, and updating
the device tree to create a new property in the proper format for
ibm,dynamic-memory-v2.
Signed-off-by: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Now that the powerpc code parses dynamic reconfiguration memory
LMB information from the LMB array and not the device tree
directly we can move the of_drconf_cell struct to drmem.h where
it fits better.
In addition, the struct is renamed to of_drconf_cell_v1 in
anticipation of upcoming support for version 2 of the dynamic
reconfiguration property and the members are typed as __be*
values to reflect how they exist in the device tree.
Signed-off-by: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Update the pseries memory hotplug code to use the newly added
dynamic reconfiguration LMB array. Doing this is required for the
upcoming support of version 2 of the dynamic reconfiguration
device tree property.
In addition, making this change cleans up the code that parses the
LMB information as we no longer need to worry about device tree
format. This allows us to discard one of the first steps on memory
hotplug where we make a working copy of the device tree property and
convert the entire property to cpu format. Instead we just use the
LMB array directly while holding the memory hotplug lock.
This patch also moves the updating of the device tree property to
powerpc/mm/drmem.c. This allows to the hotplug code to work without
needing to know the device tree format and provides a single
routine for updating the device tree property. This new routine
will handle determination of the proper device tree format and
generate a properly formatted device tree property.
Signed-off-by: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Update code in powerpc/numa.c to use the walk_drmem_lmbs()
routine instead of parsing the device tree directly. This is
in anticipation of introducing a new ibm,dynamic-memory-v2
property with a different format. This will allow the numa code
to use a single initialization routine per-LMB irregardless of
the device tree format.
Additionally, to support additional routines in numa.c that need
to look up LMB information, an late_init routine is added to drmem.c
to allocate the array of LMB information. This LMB array will provide
per-LMB information to separate the LMB data from the device tree
format.
Signed-off-by: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
We currently have code to parse the dynamic reconfiguration LMB
information from the ibm,dynamic-meory device tree property in
multiple locations; numa.c, prom.c, and pseries/hotplug-memory.c.
In anticipation of adding support for a version 2 of the
ibm,dynamic-memory property this patch aims to separate the device
tree information from the device tree format.
Doing this requires a two step process to avoid a possibly very large
bootmem allocation early in boot. During initial boot, new routines
are provided to walk the device tree property and make a call-back
for each LMB.
The second step (introduced in later patches) will allocate an
array of LMB information that can be used directly without needing
to know the DT format.
This approach provides the benefit of consolidating the device tree
property parsing to a single location and (eventually) providing
a common data structure for retrieving LMB information.
This patch introduces a routine to walk the ibm,dynamic-memory
property in the flattened device tree and updates the prom.c code
to use this to initialize memory.
Signed-off-by: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Look up the associativity arrays in of_drconf_to_nid_single when
deriving the nid for a LMB instead of having it passed in as a
parameter.
Signed-off-by: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Look up the device node for the usable memory property instead
of having it passed in as a parameter. This changes precedes an update
in which the calling routines for of_get_usable_memory() will not have
the device node pointer to pass in.
Signed-off-by: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Look up the device node for the associativity array property instead
of having it passed in as a parameter. This changes precedes an update
in which the calling routines for of_get_assoc_arrays() will not have
the device node pointer to pass in.
Signed-off-by: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
No functional change in this patch. This update gup_hugepte to use the
helper. This will help later when we add memory keys.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The H_PAGE_F_SECOND,H_PAGE_F_GIX are not in the 64K main-PTE.
capture these changes in the dump pte report.
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
replace redundant code in __hash_page_4K() and flush_hash_page()
with helper functions pte_get_hash_gslot() and pte_set_hidx()
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Rearrange 64K PTE bits to free up bits 3, 4, 5 and 6
in the 64K backed HPTE pages. This along with the earlier
patch will entirely free up the four bits from 64K PTE.
The bit numbers are big-endian as defined in the ISA3.0
This patch does the following change to 64K PTE backed
by 64K HPTE.
H_PAGE_F_SECOND (S) which occupied bit 4 moves to the
second part of the pte to bit 60.
H_PAGE_F_GIX (G,I,X) which occupied bit 5, 6 and 7 also
moves to the second part of the pte to bit 61,
62, 63, 64 respectively
since bit 7 is now freed up, we move H_PAGE_BUSY (B) from
bit 9 to bit 7.
The second part of the PTE will hold
(H_PAGE_F_SECOND|H_PAGE_F_GIX) at bit 60,61,62,63.
NOTE: None of the bits in the secondary PTE were not used
by 64k-HPTE backed PTE.
Before the patch, the 64K HPTE backed 64k PTE format was
as follows
0 1 2 3 4 5 6 7 8 9 10...........................63
: : : : : : : : : : : :
v v v v v v v v v v v v
,-,-,-,-,--,--,--,--,-,-,-,-,-,------------------,-,-,-,
|x|x|x| |S |G |I |X |x|B| |x|x|................|x|x|x|x| <- primary pte
'_'_'_'_'__'__'__'__'_'_'_'_'_'________________'_'_'_'_'
| | | | | | | | | | | | |..................| | | | | <- secondary pte
'_'_'_'_'__'__'__'__'_'_'_'_'__________________'_'_'_'_'
After the patch, the 64k HPTE backed 64k PTE format is
as follows
0 1 2 3 4 5 6 7 8 9 10...........................63
: : : : : : : : : : : :
v v v v v v v v v v v v
,-,-,-,-,--,--,--,--,-,-,-,-,-,------------------,-,-,-,
|x|x|x| | | | |B |x| | |x|x|................|.|.|.|.| <- primary pte
'_'_'_'_'__'__'__'__'_'_'_'_'_'________________'_'_'_'_'
| | | | | | | | | | | | |..................|S|G|I|X| <- secondary pte
'_'_'_'_'__'__'__'__'_'_'_'_'__________________'_'_'_'_'
The above PTE changes is applicable to hugetlbpages aswell.
The patch does the following code changes:
a) moves the H_PAGE_F_SECOND and H_PAGE_F_GIX to 4k PTE
header since it is no more needed b the 64k PTEs.
b) abstracts out __real_pte() and __rpte_to_hidx() so the
caller need not know the bit location of the slot.
c) moves the slot bits to the secondary pte.
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Rearrange 64K PTE bits to free up bits 3, 4, 5 and 6,
in the 4K backed HPTE pages.These bits continue to be used
for 64K backed HPTE pages in this patch, but will be freed
up in the next patch. The bit numbers are big-endian as
defined in the ISA3.0
The patch does the following change to the 4k HTPE backed
64K PTE's format.
H_PAGE_BUSY moves from bit 3 to bit 9 (B bit in the figure
below)
V0 which occupied bit 4 is not used anymore.
V1 which occupied bit 5 is not used anymore.
V2 which occupied bit 6 is not used anymore.
V3 which occupied bit 7 is not used anymore.
Before the patch, the 4k backed 64k PTE format was as follows
0 1 2 3 4 5 6 7 8 9 10...........................63
: : : : : : : : : : : :
v v v v v v v v v v v v
,-,-,-,-,--,--,--,--,-,-,-,-,-,------------------,-,-,-,
|x|x|x|B|V0|V1|V2|V3|x| | |x|x|................|x|x|x|x| <- primary pte
'_'_'_'_'__'__'__'__'_'_'_'_'_'________________'_'_'_'_'
|S|G|I|X|S |G |I |X |S|G|I|X|..................|S|G|I|X| <- secondary pte
'_'_'_'_'__'__'__'__'_'_'_'_'__________________'_'_'_'_'
After the patch, the 4k backed 64k PTE format is as follows
0 1 2 3 4 5 6 7 8 9 10...........................63
: : : : : : : : : : : :
v v v v v v v v v v v v
,-,-,-,-,--,--,--,--,-,-,-,-,-,------------------,-,-,-,
|x|x|x| | | | | |x|B| |x|x|................|.|.|.|.| <- primary pte
'_'_'_'_'__'__'__'__'_'_'_'_'_'________________'_'_'_'_'
|S|G|I|X|S |G |I |X |S|G|I|X|..................|S|G|I|X| <- secondary pte
'_'_'_'_'__'__'__'__'_'_'_'_'__________________'_'_'_'_'
the four bits S,G,I,X (one quadruplet per 4k HPTE) that
cache the hash-bucket slot value, is initialized to
1,1,1,1 indicating -- an invalid slot. If a HPTE gets
cached in a 1111 slot(i.e 7th slot of secondary hash
bucket), it is released immediately. In other words,
even though 1111 is a valid slot value in the hash
bucket, we consider it invalid and release the slot and
the HPTE. This gives us the opportunity to determine
the validity of S,G,I,X bits based on its contents and
not on any of the bits V0,V1,V2 or V3 in the primary PTE
When we release a HPTE cached in the 1111 slot
we also release a legitimate slot in the primary
hash bucket and unmap its corresponding HPTE. This
is to ensure that we do get a HPTE cached in a slot
of the primary hash bucket, the next time we retry.
Though treating 1111 slot as invalid, reduces the
number of available slots in the hash bucket and may
have an effect on the performance, the probabilty of
hitting a 1111 slot is extermely low.
Compared to the current scheme, the above scheme
reduces the number of false hash table updates
significantly and has the added advantage of releasing
four valuable PTE bits for other purpose.
NOTE:even though bits 3, 4, 5, 6, 7 are not used when
the 64K PTE is backed by 4k HPTE, they continue to be
used if the PTE gets backed by 64k HPTE. The next
patch will decouple that aswell, and truely release the
bits.
This idea was jointly developed by Paul Mackerras,
Aneesh, Michael Ellermen and myself.
4K PTE format remains unchanged currently.
The patch does the following code changes
a) PTE flags are split between 64k and 4k header files.
b) __hash_page_4K() is reimplemented to reflect the
above logic.
Acked-by: Balbir Singh <bsingharora@gmail.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Introduce pte_get_hash_gslot()() which returns the global slot number of
the HPTE in the global hash table.
This function will come in handy as we work towards re-arranging the PTE
bits in the later patches.
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Ram Pai <linuxram@us.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
At some point, pr_warning will be removed so all logging messages use
a consistent <prefix>_warn style.
Update arch/powerpc/
Miscellanea:
o Coalesce formats
o Realign arguments
o Use %s, __func__ instead of embedded function names
o Remove unnecessary line continuations
Signed-off-by: Joe Perches <joe@perches.com>
Acked-by: Geoff Levand <geoff@infradead.org>
[mpe: Rebase due to some %pOF changes.]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Two fixes for nasty kexec/kdump crashes in certain configurations.
A couple of minor fixes for the new TIDR code.
A fix for an oops in a CXL error handling path.
Thanks to:
Andrew Donnellan, Christophe Lombard, David Gibson, Mahesh Salgaonkar, Vaibhav Jain.
-----BEGIN PGP SIGNATURE-----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=/LD/
-----END PGP SIGNATURE-----
Merge tag 'powerpc-4.15-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc fixes from Michael Ellerman:
"Two fixes for nasty kexec/kdump crashes in certain configurations.
A couple of minor fixes for the new TIDR code.
A fix for an oops in a CXL error handling path.
Thanks to: Andrew Donnellan, Christophe Lombard, David Gibson, Mahesh
Salgaonkar, Vaibhav Jain"
* tag 'powerpc-4.15-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc: Do not assign thread.tidr if already assigned
powerpc: Avoid signed to unsigned conversion in set_thread_tidr()
powerpc/kexec: Fix kexec/kdump in P9 guest kernels
powerpc/powernv: Fix kexec crashes caused by tlbie tracing
cxl: Check if vphb exists before iterating over AFU devices
A small batch of fixes, about 50% tagged for stable and the rest for recently
merged code.
There's one more fix for the >128T handling on hash. Once a process had
requested a single mmap above 128T we would then always search above 128T. The
correct behaviour is to consider the hint address in isolation for each mmap
request.
Then a couple of fixes for the IMC PMU, a missing EXPORT_SYMBOL in VAS, a fix
for STRICT_KERNEL_RWX on 32-bit, and a fix to correctly identify P9 DD2.1 but in
code that is currently not used by default.
Thanks to:
Aneesh Kumar K.V, Christophe Leroy, Madhavan Srinivasan, Sukadev Bhattiprolu.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=gv70
-----END PGP SIGNATURE-----
Merge tag 'powerpc-4.15-2' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc fixes from Michael Ellerman:
"A small batch of fixes, about 50% tagged for stable and the rest for
recently merged code.
There's one more fix for the >128T handling on hash. Once a process
had requested a single mmap above 128T we would then always search
above 128T. The correct behaviour is to consider the hint address in
isolation for each mmap request.
Then a couple of fixes for the IMC PMU, a missing EXPORT_SYMBOL in
VAS, a fix for STRICT_KERNEL_RWX on 32-bit, and a fix to correctly
identify P9 DD2.1 but in code that is currently not used by default.
Thanks to: Aneesh Kumar K.V, Christophe Leroy, Madhavan Srinivasan,
Sukadev Bhattiprolu"
* tag 'powerpc-4.15-2' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc/64s: Fix Power9 DD2.1 logic in DT CPU features
powerpc/perf: Fix IMC_MAX_PMU macro
powerpc/perf: Fix pmu_count to count only nest imc pmus
powerpc: Fix boot on BOOK3S_32 with CONFIG_STRICT_KERNEL_RWX
powerpc/perf/imc: Use cpu_to_node() not topology_physical_package_id()
powerpc/vas: Export chip_to_vas_id()
powerpc/64s/slice: Use addr limit when computing slice mask
Rebooting into a new kernel with kexec fails in trace_tlbie() which is
called from native_hpte_clear(). This happens if the running kernel
has CONFIG_LOCKDEP enabled. With lockdep enabled, the tracepoints
always execute few RCU checks regardless of whether tracing is on or
off. We are already in the last phase of kexec sequence in real mode
with HILE_BE set. At this point the RCU check ends up in
RCU_LOCKDEP_WARN and causes kexec to fail.
Fix this by not calling trace_tlbie() from native_hpte_clear().
mpe: It's not safe to call trace points at this point in the kexec
path, even if we could avoid the RCU checks/warnings. The only
solution is to not call them.
Fixes: 0428491cba ("powerpc/mm: Trace tlbie(l) instructions")
Cc: stable@vger.kernel.org # v4.13+
Signed-off-by: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Reported-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Suggested-by: Michael Ellerman <mpe@ellerman.id.au>
Acked-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
While computing slice mask for the free area we need make sure we only
search in the addr limit applicable for this mmap. We update the
slb_addr_limit after we request for a mmap above 128TB. But the
following mmap request with hint addr below 128TB should still limit
its search to below 128TB. ie. we should not use slb_addr_limit to
compute slice mask in this case. Instead, we should derive high addr
limit based on the mmap hint addr value.
Fixes: f4ea6dcb08 ("powerpc/mm: Enable mappings above 128TB")
Cc: stable@vger.kernel.org # v4.12+
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Non-highlights:
- Five fixes for the >128T address space handling, both to fix bugs in our
implementation and to bring the semantics exactly into line with x86.
Highlights:
- Support for a new OPAL call on bare metal machines which gives us a true NMI
(ie. is not masked by MSR[EE]=0) for debugging etc.
- Support for Power9 DD2 in the CXL driver.
- Improvements to machine check handling so that uncorrectable errors can be
reported into the generic memory_failure() machinery.
- Some fixes and improvements for VPHN, which is used under PowerVM to notify
the Linux partition of topology changes.
- Plumbing to enable TM (transactional memory) without suspend on some Power9
processors (PPC_FEATURE2_HTM_NO_SUSPEND).
- Support for emulating vector loads form cache-inhibited memory, on some
Power9 revisions.
- Disable the fast-endian switch "syscall" by default (behind a CONFIG), we
believe it has never had any users.
- A major rework of the API drivers use when initiating and waiting for long
running operations performed by OPAL firmware, and changes to the
powernv_flash driver to use the new API.
- Several fixes for the handling of FP/VMX/VSX while processes are using
transactional memory.
- Optimisations of TLB range flushes when using the radix MMU on Power9.
- Improvements to the VAS facility used to access coprocessors on Power9, and
related improvements to the way the NX crypto driver handles requests.
- Implementation of PMEM_API and UACCESS_FLUSHCACHE for 64-bit.
Thanks to:
Alexey Kardashevskiy, Alistair Popple, Allen Pais, Andrew Donnellan, Aneesh
Kumar K.V, Arnd Bergmann, Balbir Singh, Benjamin Herrenschmidt, Breno Leitao,
Christophe Leroy, Christophe Lombard, Cyril Bur, Frederic Barrat, Gautham R.
Shenoy, Geert Uytterhoeven, Guilherme G. Piccoli, Gustavo Romero, Haren
Myneni, Joel Stanley, Kamalesh Babulal, Kautuk Consul, Markus Elfring, Masami
Hiramatsu, Michael Bringmann, Michael Neuling, Michal Suchanek, Naveen N. Rao,
Nicholas Piggin, Oliver O'Halloran, Paul Mackerras, Pedro Miraglia Franco de
Carvalho, Philippe Bergheaud, Sandipan Das, Seth Forshee, Shriya, Stephen
Rothwell, Stewart Smith, Sukadev Bhattiprolu, Tyrel Datwyler, Vaibhav Jain,
Vaidyanathan Srinivasan, William A. Kennington III.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=Rq81
-----END PGP SIGNATURE-----
Merge tag 'powerpc-4.15-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc updates from Michael Ellerman:
"A bit of a small release, I suspect in part due to me travelling for
KS. But my backlog of patches to review is smaller than usual, so I
think in part folks just didn't send as much this cycle.
Non-highlights:
- Five fixes for the >128T address space handling, both to fix bugs
in our implementation and to bring the semantics exactly into line
with x86.
Highlights:
- Support for a new OPAL call on bare metal machines which gives us a
true NMI (ie. is not masked by MSR[EE]=0) for debugging etc.
- Support for Power9 DD2 in the CXL driver.
- Improvements to machine check handling so that uncorrectable errors
can be reported into the generic memory_failure() machinery.
- Some fixes and improvements for VPHN, which is used under PowerVM
to notify the Linux partition of topology changes.
- Plumbing to enable TM (transactional memory) without suspend on
some Power9 processors (PPC_FEATURE2_HTM_NO_SUSPEND).
- Support for emulating vector loads form cache-inhibited memory, on
some Power9 revisions.
- Disable the fast-endian switch "syscall" by default (behind a
CONFIG), we believe it has never had any users.
- A major rework of the API drivers use when initiating and waiting
for long running operations performed by OPAL firmware, and changes
to the powernv_flash driver to use the new API.
- Several fixes for the handling of FP/VMX/VSX while processes are
using transactional memory.
- Optimisations of TLB range flushes when using the radix MMU on
Power9.
- Improvements to the VAS facility used to access coprocessors on
Power9, and related improvements to the way the NX crypto driver
handles requests.
- Implementation of PMEM_API and UACCESS_FLUSHCACHE for 64-bit.
Thanks to: Alexey Kardashevskiy, Alistair Popple, Allen Pais, Andrew
Donnellan, Aneesh Kumar K.V, Arnd Bergmann, Balbir Singh, Benjamin
Herrenschmidt, Breno Leitao, Christophe Leroy, Christophe Lombard,
Cyril Bur, Frederic Barrat, Gautham R. Shenoy, Geert Uytterhoeven,
Guilherme G. Piccoli, Gustavo Romero, Haren Myneni, Joel Stanley,
Kamalesh Babulal, Kautuk Consul, Markus Elfring, Masami Hiramatsu,
Michael Bringmann, Michael Neuling, Michal Suchanek, Naveen N. Rao,
Nicholas Piggin, Oliver O'Halloran, Paul Mackerras, Pedro Miraglia
Franco de Carvalho, Philippe Bergheaud, Sandipan Das, Seth Forshee,
Shriya, Stephen Rothwell, Stewart Smith, Sukadev Bhattiprolu, Tyrel
Datwyler, Vaibhav Jain, Vaidyanathan Srinivasan, and William A.
Kennington III"
* tag 'powerpc-4.15-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux: (151 commits)
powerpc/64s: Fix Power9 DD2.0 workarounds by adding DD2.1 feature
powerpc/64s: Fix masking of SRR1 bits on instruction fault
powerpc/64s: mm_context.addr_limit is only used on hash
powerpc/64s/radix: Fix 128TB-512TB virtual address boundary case allocation
powerpc/64s/hash: Allow MAP_FIXED allocations to cross 128TB boundary
powerpc/64s/hash: Fix fork() with 512TB process address space
powerpc/64s/hash: Fix 128TB-512TB virtual address boundary case allocation
powerpc/64s/hash: Fix 512T hint detection to use >= 128T
powerpc: Fix DABR match on hash based systems
powerpc/signal: Properly handle return value from uprobe_deny_signal()
powerpc/fadump: use kstrtoint to handle sysfs store
powerpc/lib: Implement UACCESS_FLUSHCACHE API
powerpc/lib: Implement PMEM API
powerpc/powernv/npu: Don't explicitly flush nmmu tlb
powerpc/powernv/npu: Use flush_all_mm() instead of flush_tlb_mm()
powerpc/powernv/idle: Round up latency and residency values
powerpc/kprobes: refactor kprobe_lookup_name for safer string operations
powerpc/kprobes: Blacklist emulate_update_regs() from kprobes
powerpc/kprobes: Do not disable interrupts for optprobes and kprobes_on_ftrace
powerpc/kprobes: Disable preemption before invoking probe handler for optprobes
...
Most callers users of free_hot_cold_page claim the pages being released
are cache hot. The exception is the page reclaim paths where it is
likely that enough pages will be freed in the near future that the
per-cpu lists are going to be recycled and the cache hotness information
is lost. As no one really cares about the hotness of pages being
released to the allocator, just ditch the parameter.
The APIs are renamed to indicate that it's no longer about hot/cold
pages. It should also be less confusing as there are subtle differences
between them. __free_pages drops a reference and frees a page when the
refcount reaches zero. free_hot_cold_page handled pages whose refcount
was already zero which is non-obvious from the name. free_unref_page
should be more obvious.
No performance impact is expected as the overhead is marginal. The
parameter is removed simply because it is a bit stupid to have a useless
parameter copied everywhere.
[mgorman@techsingularity.net: add pages to head, not tail]
Link: http://lkml.kernel.org/r/20171019154321.qtpzaeftoyyw4iey@techsingularity.net
Link: http://lkml.kernel.org/r/20171018075952.10627-8-mgorman@techsingularity.net
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
On a machine with 5-level paging support a process can allocate
significant amount of memory and stay unnoticed by oom-killer and memory
cgroup. The trick is to allocate a lot of PUD page tables. We don't
account PUD page tables, only PMD and PTE.
We already addressed the same issue for PMD page tables, see commit
dc6c9a35b6 ("mm: account pmd page tables to the process").
Introduction of 5-level paging brings the same issue for PUD page
tables.
The patch expands accounting to PUD level.
[kirill.shutemov@linux.intel.com: s/pmd_t/pud_t/]
Link: http://lkml.kernel.org/r/20171004074305.x35eh5u7ybbt5kar@black.fi.intel.com
[heiko.carstens@de.ibm.com: s390/mm: fix pud table accounting]
Link: http://lkml.kernel.org/r/20171103090551.18231-1-heiko.carstens@de.ibm.com
Link: http://lkml.kernel.org/r/20171002080427.3320-1-kirill.shutemov@linux.intel.com
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: Rik van Riel <riel@redhat.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull timer updates from Thomas Gleixner:
"Yet another big pile of changes:
- More year 2038 work from Arnd slowly reaching the point where we
need to think about the syscalls themself.
- A new timer function which allows to conditionally (re)arm a timer
only when it's either not running or the new expiry time is sooner
than the armed expiry time. This allows to use a single timer for
multiple timeout requirements w/o caring about the first expiry
time at the call site.
- A new NMI safe accessor to clock real time for the printk timestamp
work. Can be used by tracing, perf as well if required.
- A large number of timer setup conversions from Kees which got
collected here because either maintainers requested so or they
simply got ignored. As Kees pointed out already there are a few
trivial merge conflicts and some redundant commits which was
unavoidable due to the size of this conversion effort.
- Avoid a redundant iteration in the timer wheel softirq processing.
- Provide a mechanism to treat RTC implementations depending on their
hardware properties, i.e. don't inflict the write at the 0.5
seconds boundary which originates from the PC CMOS RTC to all RTCs.
No functional change as drivers need to be updated separately.
- The usual small updates to core code clocksource drivers. Nothing
really exciting"
* 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (111 commits)
timers: Add a function to start/reduce a timer
pstore: Use ktime_get_real_fast_ns() instead of __getnstimeofday()
timer: Prepare to change all DEFINE_TIMER() callbacks
netfilter: ipvs: Convert timers to use timer_setup()
scsi: qla2xxx: Convert timers to use timer_setup()
block/aoe: discover_timer: Convert timers to use timer_setup()
ide: Convert timers to use timer_setup()
drbd: Convert timers to use timer_setup()
mailbox: Convert timers to use timer_setup()
crypto: Convert timers to use timer_setup()
drivers/pcmcia: omap1: Fix error in automated timer conversion
ARM: footbridge: Fix typo in timer conversion
drivers/sgi-xp: Convert timers to use timer_setup()
drivers/pcmcia: Convert timers to use timer_setup()
drivers/memstick: Convert timers to use timer_setup()
drivers/macintosh: Convert timers to use timer_setup()
hwrng/xgene-rng: Convert timers to use timer_setup()
auxdisplay: Convert timers to use timer_setup()
sparc/led: Convert timers to use timer_setup()
mips: ip22/32: Convert timers to use timer_setup()
...
Radix keeps no meaningful state in addr_limit, so remove it from radix
code and rename to slb_addr_limit to make it clear it applies to hash
only.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Radix VA space allocations test addresses against mm->task_size which
is 512TB, even in cases where the intention is to limit allocation to
below 128TB.
This results in mmap with a hint address below 128TB but address +
length above 128TB succeeding when it should fail (as hash does after
the previous patch).
Set the high address limit to be considered up front, and base
subsequent allocation checks on that consistently.
Fixes: f4ea6dcb08 ("powerpc/mm: Enable mappings above 128TB")
Cc: stable@vger.kernel.org # v4.12+
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
While mapping hints with a length that cross 128TB are disallowed,
MAP_FIXED allocations that cross 128TB are allowed. These are failing
on hash (on radix they succeed). Add an additional case for fixed
mappings to expand the addr_limit when crossing 128TB.
Fixes: f4ea6dcb08 ("powerpc/mm: Enable mappings above 128TB")
Cc: stable@vger.kernel.org # v4.12+
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Hash unconditionally resets the addr_limit to default (128TB) when the
mm context is initialised. If a process has > 128TB mappings when it
forks, the child will not get the 512TB addr_limit, so accesses to
valid > 128TB mappings will fail in the child.
Fix this by only resetting the addr_limit to default if it was 0. Non
zero indicates it was duplicated from the parent (0 means exec()).
Fixes: f4ea6dcb08 ("powerpc/mm: Enable mappings above 128TB")
Cc: stable@vger.kernel.org # v4.12+
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
When allocating VA space with a hint that crosses 128TB, the SLB
addr_limit variable is not expanded if addr is not > 128TB, but the
slice allocation looks at task_size, which is 512TB. This results in
slice_check_fit() incorrectly succeeding because the slice_count
truncates off bit 128 of the requested mask, so the comparison to the
available mask succeeds.
Fix this by using mm->context.addr_limit instead of mm->task_size for
testing allocation limits. This causes such allocations to fail.
Fixes: f4ea6dcb08 ("powerpc/mm: Enable mappings above 128TB")
Cc: stable@vger.kernel.org # v4.12+
Reported-by: Florian Weimer <fweimer@redhat.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Currently userspace is able to request mmap() search between 128T-512T
by specifying a hint address that is greater than 128T. But that means
a hint of 128T exactly will return an address below 128T, which is
confusing and wrong.
So fix the logic to check the hint is greater than *or equal* to 128T.
Fixes: f4ea6dcb08 ("powerpc/mm: Enable mappings above 128TB")
Cc: stable@vger.kernel.org # v4.12+
Suggested-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Suggested-by: Nicholas Piggin <npiggin@gmail.com>
[mpe: Split out of Nick's bigger patch]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
When using the radix MMU on Power9 DD1, to work around a hardware
problem, radix__pte_update() is required to do a two stage update of
the PTE. First we write a zero value into the PTE, then we flush the
TLB, and then we write the new PTE value.
In the normal case that works OK, but it does not work if we're
updating the PTE that maps the code we're executing, because the
mapping is removed by the TLB flush and we can no longer execute from
it. Unfortunately the STRICT_RWX code needs to do exactly that.
The exact symptoms when we hit this case vary, sometimes we print an
oops and then get stuck after that, but I've also seen a machine just
get stuck continually page faulting with no oops printed. The variance
is presumably due to the exact layout of the text and the page size
used for the mappings. In all cases we are unable to boot to a shell.
There are possible solutions such as creating a second mapping of the
TLB flush code, executing from that, and then jumping back to the
original. However we don't want to add that level of complexity for a
DD1 work around.
So just detect that we're running on Power9 DD1 and refrain from
changing the permissions, effectively disabling STRICT_RWX on Power9
DD1.
Fixes: 7614ff3272 ("powerpc/mm/radix: Implement STRICT_RWX/mark_rodata_ro() for Radix")
Cc: stable@vger.kernel.org # v4.13+
Reported-by: Andrew Jeffery <andrew@aj.id.au>
[Changelog as suggested by Michael Ellerman <mpe@ellerman.id.au>]
Signed-off-by: Balbir Singh <bsingharora@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Unmaps that free page tables always flush the entire PID, which is
sub-optimal. Provide TLB range flushing with an additional PWC flush
that can be use for va range invalidations with PWC flush.
Time to munmap N pages of memory including last level page table
teardown (after mmap, touch), local invalidate:
N 1 2 4 8 16 32 64
vanilla 3.2us 3.3us 3.4us 3.6us 4.1us 5.2us 7.2us
patched 1.4us 1.5us 1.7us 1.9us 2.6us 3.7us 6.2us
Global invalidate:
N 1 2 4 8 16 32 64
vanilla 2.2us 2.3us 2.4us 2.6us 3.2us 4.1us 6.2us
patched 2.1us 2.5us 3.4us 5.2us 8.7us 15.7us 6.2us
Local invalidates get much better across the board. Global ones have
the same issue where multiple tlbies for va flush do get slower than
the single tlbie to invalidate the PID. None of this test captures
the TLB benefits of avoiding killing everything.
Global gets worse, but it is brought in to line with global invalidate
for munmap()s that do not free page tables.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The single page flush ceiling is the cut-off point at which we switch
from invalidating individual pages, to invalidating the entire process
address space in response to a range flush.
Introduce a local variant of this heuristic because local and global
tlbie have significantly different properties:
- Local tlbiel requires 128 instructions to invalidate a PID, global
tlbie only 1 instruction.
- Global tlbie instructions are expensive broadcast operations.
The local ceiling has been made much higher, 2x the number of
instructions required to invalidate the entire PID (i.e., 256 pages).
Time to mprotect N pages of memory (after mmap, touch), local invalidate:
N 32 34 64 128 256 512
vanilla 7.4us 9.0us 14.6us 26.4us 50.2us 98.3us
patched 7.4us 7.8us 13.8us 26.4us 51.9us 98.3us
The behaviour of both is identical at N=32 and N=512. Between there,
the vanilla kernel does a PID invalidate and the patched kernel does
a va range invalidate.
At N=128, these require the same number of tlbiel instructions, so
the patched version can be sen to be cheaper when < 128, and more
expensive when > 128. However this does not well capture the cost
of invalidated TLB.
The additional cost at 256 pages does not seem prohibitive. It may
be the case that increasing the limit further would continue to be
beneficial to avoid invalidating all of the process's TLB entries.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Currently for radix, flush_tlb_range flushes the entire PID, because
the Linux mm code does not tell us about page size here for THP vs
regular pages. This is quite sub-optimal for small mremap / mprotect
/ change_protection.
So implement va range flushes with two flush passes, one for each
page size (regular and THP). The second flush has an order of matnitude
fewer tlbie instructions than the first, so it is a relatively small
additional cost.
There is still room for improvement here with some changes to generic
APIs, particularly if there are mostly THP pages to be invalidated,
the small page flushes could be reduced.
Time to mprotect 1 page of memory (after mmap, touch):
vanilla 2.9us 1.8us
patched 1.2us 1.6us
Time to mprotect 30 pages of memory (after mmap, touch):
vanilla 8.2us 7.2us
patched 6.9us 17.9us
Time to mprotect 34 pages of memory (after mmap, touch):
vanilla 9.1us 8.0us
patched 9.0us 8.0us
34 pages is the point at which the invalidation switches from va
to entire PID, which tlbie can do in a single instruction. This is
why in the case of 30 pages, the new code runs slower for this test.
This is a deliberate tradeoff already present in the unmap and THP
promotion code, the idea is that the benefit from avoiding flushing
entire TLB for this PID on all threads in the system.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Move the barriers and range iteration down into the _tlbie* level,
which improves readability.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Short range flushes issue a sequences of tlbie(l) instructions for
individual effective addresses. These do not all require individual
barrier sequences, only one covering all tlbie(l) instructions.
Commit f7327e0ba3 ("powerpc/mm/radix: Remove unnecessary ptesync")
made a similar optimization for tlbiel for PID flushing.
For tlbie, the ISA says:
The tlbsync instruction provides an ordering function for the
effects of all tlbie instructions executed by the thread executing
the tlbsync instruction, with respect to the memory barrier
created by a subsequent ptesync instruction executed by the same
thread.
Time to munmap 30 pages of memory (after mmap, touch):
local global
vanilla 10.9us 22.3us
patched 3.4us 14.4us
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
We have some dependencies & conflicts between patches in fixes and
things to go in next, both in the radix TLB flush code and the IMC PMU
driver. So merge fixes into next.
In commit e6f81a9201 ("powerpc/mm/hash: Support 68 bit VA") the
masking is folded into ASM_VSID_SCRAMBLE but the comment about masking
is removed only from the firt use of ASM_VSID_SCRAMBLE.
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Currently if the hardware supports the radix MMU we will use
it, *unless* "disable_radix" is passed on the kernel command line.
However some users would like the reverse semantics. ie. The kernel
uses the hash MMU by default, unless radix is explicitly requested on
the command line.
So add a CONFIG option to choose whether we use radix by default or
not, and expand the disable_radix command line option to allow
"disable_radix=no" which *enables* radix.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
CONFIG_PPC_STD_MMU_64 indicates support for the "standard" powerpc MMU
on 64-bit CPUs. The "standard" MMU refers to the hash page table MMU
found in "server" processors, from IBM mainly.
Currently CONFIG_PPC_STD_MMU_64 is == CONFIG_PPC_BOOK3S_64. While it's
annoying to have two symbols that always have the same value, it's not
quite annoying enough to bother removing one.
However with the arrival of Power9, we now have the situation where
CONFIG_PPC_STD_MMU_64 is enabled, but the kernel is running using the
Radix MMU - *not* the "standard" MMU. So it is now actively confusing
to use it, because it implies that code is disabled or inactive when
the Radix MMU is in use, however that is not necessarily true.
So s/CONFIG_PPC_STD_MMU_64/CONFIG_PPC_BOOK3S_64/, and do some minor
formatting updates of some of the affected lines.
This will be a pain for backports, but c'est la vie.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>