Add various defines like e.g. _REGION1_SHIFT to reflect the hardware
mmu. We have quite a bit code that does not make use of the Linux
memory management primitives but directly modifies page, segment and
region values.
Most of this is open-coded like e.g. "1UL << 53". In order to clean
this up introduce a couple of new defines. The existing Linux memory
management defines are changed, so the mapping to the hardware
implementation is reflected.
Reviewed-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Add the logic to upgrade the page table for a 64-bit process to
five levels. This increases the TASK_SIZE from 8PB to 16EB-4K.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
The ptff() and clear_table() functions use the gcc extension "variable
length arrays in structures" (VLAIS) to define in the inline assembler
constraints the area of the clobbered memory. This extension will most
likely never be supported by LLVM/Clang.
Since currently BPF programs are compiled with LLVM, this leads to the
following compile errors:
$ cd samples/bpf
$ make
In file included from /root/linux-master/samples/bpf/tracex1_kern.c:8:
In file included from ./include/linux/netdevice.h:44:
...
In file included from ./arch/s390/include/asm/mmu_context.h:10:
./arch/s390/include/asm/pgalloc.h:30:24: error: fields must have a
constant size: 'variable length array in structure' extension will never
be supported
typedef struct { char _[n]; } addrtype;
In file included from /root/linux-master/samples/bpf/tracex1_kern.c:7:
In file included from ./include/linux/skbuff.h:18:
...
In file included from ./include/linux/jiffies.h:8:
In file included from ./include/linux/timex.h:65:
./arch/s390/include/asm/timex.h:105:24: error: fields must have a
constant size: 'variable length array in structure' extension will never
be supported
typedef struct { char _[len]; } addrtype;
To fix this do the following:
- Convert ptff() into a macro that then uses a fixed size array
when expanded.
- Convert the clear_table() function and use an inline assembly
with fixed size array in a loop.
The runtime performance of the new version is even better than
the old version (tested with EC12/z13 and gcc 4.8.5/6.2.1 with
"-march=z196 -O2").
Reported-by: Zvonko Kosic <zvonko.kosic@de.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
For a nested KVM guest the outer KVM host needs to create shadow
page tables for the nested guest. This patch adds the basic support
to the guest address space (gmap) code.
For each guest address space the inner KVM host creates, the first
outer KVM host needs to create shadow page tables. The address space
is identified by the ASCE loaded into the control register 1 at the
time the inner SIE instruction for the second nested KVM guest is
executed. The outer KVM host creates the shadow tables starting with
the table identified by the ASCE on a on-demand basis. The outer KVM
host will get repeated faults for all the shadow tables needed to
run the second KVM guest.
While a shadow page table for the second KVM guest is active the access
to the origin region, segment and page tables needs to be restricted
for the first KVM guest. For region and segment and page tables the first
KVM guest may read the memory, but write attempt has to lead to an
unshadow. This is done using the page invalid and read-only bits in the
page table of the first KVM guest. If the first guest re-accesses one of
the origin pages of a shadow, it gets a fault and the affected parts of
the shadow page table hierarchy needs to be removed again.
PGSTE tables don't have to be shadowed, as all interpretation assist can't
deal with the invalid bits in the shadow pte being set differently than
the original ones provided by the first KVM guest.
Many bug fixes and improvements by David Hildenbrand.
Reviewed-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
There is a race with multi-threaded applications between context switch and
pagetable upgrade. In switch_mm() a new user_asce is built from mm->pgd and
mm->context.asce_bits, w/o holding any locks. A concurrent mmap with a
pagetable upgrade on another thread in crst_table_upgrade() could already
have set new asce_bits, but not yet the new mm->pgd. This would result in a
corrupt user_asce in switch_mm(), and eventually in a kernel panic from a
translation exception.
Fix this by storing the complete asce instead of just the asce_bits, which
can then be read atomically from switch_mm(), so that it either sees the
old value or the new value, but no mixture. Both cases are OK. Having the
old value would result in a page fault on access to the higher level memory,
but the fault handler would see the new mm->pgd, if it was a valid access
after the mmap on the other thread has completed. So as worst-case scenario
we would have a page fault loop for the racing thread until the next time
slice.
Also remove dead code and simplify the upgrade/downgrade path, there are no
upgrades from 2 levels, and only downgrades from 3 levels for compat tasks.
There are also no concurrent upgrades, because the mmap_sem is held with
down_write() in do_mmap, so the flush and table checks during upgrade can
be removed.
Reported-by: Michael Munday <munday@ca.ibm.com>
Reviewed-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Pull s390 updates from Martin Schwidefsky:
- Add the CPU id for the new z13s machine
- Add a s390 specific XOR template for RAID-5 checksumming based on the
XC instruction. Remove all other alternatives, XC is always faster
- The merge of our four different stack tracers into a single one
- Tidy up the code related to page tables, several large inline
functions are now out-of-line. Bloat-o-meter reports ~11K text size
reduction
- A binary interface for the priviledged CLP instruction to retrieve
the hardware view of the installed PCI functions
- Improvements for the dasd format code
- Bug fixes and cleanups
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (31 commits)
s390/pci: enforce fmb page boundary rule
s390: fix floating pointer register corruption (again)
s390/cpumf: add missing lpp magic initialization
s390: Fix misspellings in comments
s390/mm: split arch/s390/mm/pgtable.c
s390/mm: uninline pmdp_xxx functions from pgtable.h
s390/mm: uninline ptep_xxx functions from pgtable.h
s390/pci: add ioctl interface for CLP
s390: Use pr_warn instead of pr_warning
s390/dasd: remove casts to dasd_*_private
s390/dasd: Refactor dasd format functions
s390/dasd: Simplify code in format logic
s390/dasd: Improve dasd format code
s390/percpu: remove this_cpu_cmpxchg_double_4
s390/cpumf: Improve guest detection heuristics
s390/fault: merge report_user_fault implementations
s390/dis: use correct escape sequence for '%' character
s390/kvm: simplify set_guest_storage_key
s390/oprofile: add z13/z13s model numbers
s390: add z13s model number to z13 elf platform
...
The fork of a process with four page table levels is broken since
git commit 6252d702c5 "[S390] dynamic page tables."
All new mm contexts are created with three page table levels and
an asce limit of 4TB. If the parent has four levels dup_mmap will
add vmas to the new context which are outside of the asce limit.
The subsequent call to copy_page_range will walk the three level
page table structure of the new process with non-zero pgd and pud
indexes. This leads to memory clobbers as the pgd_index *and* the
pud_index is added to the mm->pgd pointer without a pgd_deref
in between.
The init_new_context() function is selecting the number of page
table levels for a new context. The function is used by mm_init()
which in turn is called by dup_mm() and mm_alloc(). These two are
used by fork() and exec(). The init_new_context() function can
distinguish the two cases by looking at mm->context.asce_limit,
for fork() the mm struct has been copied and the number of page
table levels may not change. For exec() the mm_alloc() function
set the new mm structure to zero, in this case a three-level page
table is created as the temporary stack space is located at
STACK_TOP_MAX = 4TB.
This fixes CVE-2016-2143.
Reported-by: Marcin Kościelnicki <koriakin@0x04.net>
Reviewed-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: stable@vger.kernel.org
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
The pgtable.c file is quite big, before it grows any larger split it
into pgtable.c, pgalloc.c and gmap.c. In addition move the gmap related
header definitions into the new gmap.h header and all of the pgste
helpers from pgtable.h to pgtable.c.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Replacing a 2K page table with a 4K page table while a VMA is active
for the affected memory region is fundamentally broken. Rip out the
page table reallocation code and replace it with a simple system
control 'vm.allocate_pgste'. If the system control is set the page
tables for all processes are allocated as full 4K pages, even for
processes that do not need it.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Remove the 31 bit support in order to reduce maintenance cost and
effectively remove dead code. Since a couple of years there is no
distribution left that comes with a 31 bit kernel.
The 31 bit kernel also has been broken since more than a year before
anybody noticed. In addition I added a removal warning to the kernel
shown at ipl for 5 minutes: a960062e58 ("s390: add 31 bit warning
message") which let everybody know about the plan to remove 31 bit
code. We didn't get any response.
Given that the last 31 bit only machine was introduced in 1999 let's
remove the code.
Anybody with 31 bit user space code can still use the compat mode.
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
- spring cleaning: removed support for IA64, and for hardware-assisted
virtualization on the PPC970
- ARM, PPC, s390 all had only small fixes
For x86:
- small performance improvements (though only on weird guests)
- usual round of hardware-compliancy fixes from Nadav
- APICv fixes
- XSAVES support for hosts and guests. XSAVES hosts were broken because
the (non-KVM) XSAVES patches inadvertently changed the KVM userspace
ABI whenever XSAVES was enabled; hence, this part is going to stable.
Guest support is just a matter of exposing the feature and CPUID leaves
support.
Right now KVM is broken for PPC BookE in your tree (doesn't compile).
I'll reply to the pull request with a patch, please apply it either
before the pull request or in the merge commit, in order to preserve
bisectability somewhat.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)
iQEcBAABAgAGBQJUkpg+AAoJEL/70l94x66DUmoH/jzXYkptSW9NGgm79KqxGJlD
lzLnLBkitVvx++Mz5YBhdJEhKKLUlCtifFT1zPJQ/pthQhIRSaaAwZyNGgUs5w5x
yMGKHiPQFyZRbmQtZhCInW0BftJoYHHciO3nUfHCZnp34My9MP2D55W7/z+fYFfQ
DuqBSE9ThyZJtZ4zh8NRA9fCOeuqwVYRyoBs820Wbsh4cpIBoIK63Dg7k+CLE+ZV
MZa/mRL6bAfsn9W5bnOUAgHJ3SPznnWbO3/g0aV+roL/5pffblprJx9lKNR08xUM
6hDFLop2gDehDJesDkY/o8Ckp1hEouvfsVpSShry4vcgtn0hgh2O5/6Orbmj6vE=
=Zwq1
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM update from Paolo Bonzini:
"3.19 changes for KVM:
- spring cleaning: removed support for IA64, and for hardware-
assisted virtualization on the PPC970
- ARM, PPC, s390 all had only small fixes
For x86:
- small performance improvements (though only on weird guests)
- usual round of hardware-compliancy fixes from Nadav
- APICv fixes
- XSAVES support for hosts and guests. XSAVES hosts were broken
because the (non-KVM) XSAVES patches inadvertently changed the KVM
userspace ABI whenever XSAVES was enabled; hence, this part is
going to stable. Guest support is just a matter of exposing the
feature and CPUID leaves support"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (179 commits)
KVM: move APIC types to arch/x86/
KVM: PPC: Book3S: Enable in-kernel XICS emulation by default
KVM: PPC: Book3S HV: Improve H_CONFER implementation
KVM: PPC: Book3S HV: Fix endianness of instruction obtained from HEIR register
KVM: PPC: Book3S HV: Remove code for PPC970 processors
KVM: PPC: Book3S HV: Tracepoints for KVM HV guest interactions
KVM: PPC: Book3S HV: Simplify locking around stolen time calculations
arch: powerpc: kvm: book3s_paired_singles.c: Remove unused function
arch: powerpc: kvm: book3s_pr.c: Remove unused function
arch: powerpc: kvm: book3s.c: Remove some unused functions
arch: powerpc: kvm: book3s_32_mmu.c: Remove unused function
KVM: PPC: Book3S HV: Check wait conditions before sleeping in kvmppc_vcore_blocked
KVM: PPC: Book3S HV: ptes are big endian
KVM: PPC: Book3S HV: Fix inaccuracies in ICP emulation for H_IPI
KVM: PPC: Book3S HV: Fix KSM memory corruption
KVM: PPC: Book3S HV: Fix an issue where guest is paused on receiving HMI
KVM: PPC: Book3S HV: Fix computation of tlbie operand
KVM: PPC: Book3S HV: Add missing HPTE unlock
KVM: PPC: BookE: Improve irq inject tracepoint
arm/arm64: KVM: Require in-kernel vgic for the arch timers
...
Define get_guest_storage_key which can be used to get the value of a guest
storage key. This compliments the functionality provided by the helper function
set_guest_storage_key. Both functions are needed for live migration of s390
guests that use storage keys.
Signed-off-by: Jason J. Herne <jjherne@linux.vnet.ibm.com>
Reviewed-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Replace the s390 specific page table walker for the pgste updates
with a call to the common code walk_page_range function.
There are now two pte modification functions, one for the reset
of the CMMA state and another one for the initialization of the
storage keys.
Signed-off-by: Dominik Dingel <dingel@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Store the target address for the gmap segments in a radix tree
instead of using invalid segment table entries. gmap_translate
becomes a simple radix_tree_lookup, gmap_fault is split into the
address translation with gmap_translate and the part that does
the linking of the gmap shadow page table with the process page
table.
A second radix tree is used to keep the pointers to the segment
table entries for segments that are mapped in the guest address
space. On unmap of a segment the pointer is retrieved from the
radix tree and is used to carry out the segment invalidation in
the gmap shadow page table. As the radix tree can only store one
pointer, each host segment may only be mapped to exactly one
guest location.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
page_table_reset_pgste() already does a complete page table walk to
reset the pgste. Enhance it to initialize the storage keys to
PAGE_DEFAULT_KEY if requested by the caller. This will be used
for lazy storage key handling. Also provide an empty stub for
!CONFIG_PGSTE
Lets adopt the current code (diag 308) to not clear the keys.
Signed-off-by: Dominik Dingel <dingel@linux.vnet.ibm.com>
Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Add the pgtable_pmd_page_ctor/pgtable_pmd_page_dtor calls to the pmd
allocation and free functions and enable ARCH_ENABLE_SPLIT_PMD_PTLOCK
for 64 bit.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
The guest page state needs to be reset to stable for all pages
on initial program load via diagnose 0x308.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
From time to time we need to set the guest storage key. Lets
provide a helper function that handles the changes with all the
right locking and checking.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Remove the file name from the comment at top of many files. In most
cases the file name was wrong anyway, so it's rather pointless.
Also unify the IBM copyright statement. We did have a lot of sightly
different statements and wanted to change them one after another
whenever a file gets touched. However that never happened. Instead
people start to take the old/"wrong" statements to use as a template
for new files.
So unify all of them in one go.
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Replace __s390x__ with CONFIG_64BIT in all places that are not exported
to userspace or guarded with #ifdef __KERNEL__.
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Git commit 36409f6353 "use generic RCU
page-table freeing code" introduced a tlb flushing bug. Partially revert
the above git commit and go back to s390 specific page table flush code.
For s390 the TLB can contain three types of entries, "normal" TLB
page-table entries, TLB combined region-and-segment-table (CRST) entries
and real-space entries. Linux does not use real-space entries which
leaves normal TLB entries and CRST entries. The CRST entries are
intermediate steps in the page-table translation called translation paths.
For example a 4K page access in a three-level page table setup will
create two CRST TLB entries and one page-table TLB entry. The advantage
of that approach is that a page access next to the previous one can reuse
the CRST entries and needs just a single read from memory to create the
page-table TLB entry. The disadvantage is that the TLB flushing rules are
more complicated, before any page-table may be freed the TLB needs to be
flushed.
In short: the generic RCU page-table freeing code is incorrect for the
CRST entries, in particular the check for mm_users < 2 is troublesome.
This is applicable to 3.0+ kernels.
Cc: <stable@vger.kernel.org>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Add code that allows KVM to control the virtual memory layout that
is seen by a guest. The guest address space uses a second page table
that shares the last level pte-tables with the process page table.
If a page is unmapped from the process page table it is automatically
unmapped from the guest page table as well.
The guest address space mapping starts out empty, KVM can map any
individual 1MB segments from the process virtual memory to any 1MB
aligned location in the guest virtual memory. If a target segment in
the process virtual memory does not exist or is unmapped while a
guest mapping exists the desired target address is stored as an
invalid segment table entry in the guest page table.
The population of the guest page table is fault driven.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Replace the s390 specific rcu page-table freeing code with the
generic variant. This requires to duplicate the definition for the
struct mmu_table_batch as s390 does not use the generic tlb flush
code.
While we are at it remove the restriction that page table fragments
can not be reused after a single fragment has been freed with rcu
and split out allocation and freeing of page tables with pgstes.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Rework the architecture page table functions to access the bits in the
page table extension array (pgste). There are a number of changes:
1) Fix missing pgste update if the attach_count for the mm is <= 1.
2) For every operation that affects the invalid bit in the pte or the
rcp byte in the pgste the pcl lock needs to be acquired. The function
pgste_get_lock gets the pcl lock and returns the current pgste value
for a pte pointer. The function pgste_set_unlock stores the pgste
and releases the lock. Between these two calls the bits in the pgste
can be shuffled.
3) Define two software bits in the pte _PAGE_SWR and _PAGE_SWC to avoid
calling SetPageDirty and SetPageReferenced from pgtable.h. If the
host reference backup bit or the host change backup bit has been
set the dirty/referenced state is transfered to the pte. The common
code will pick up the state from the pte.
4) Add ptep_modify_prot_start and ptep_modify_prot_commit for mprotect.
5) Remove pgd_populate_kernel, pud_populate_kernel, pmd_populate_kernel
pgd_clear_kernel, pud_clear_kernel, pmd_clear_kernel and ptep_invalidate.
6) Rename kvm_s390_test_and_clear_page_dirty to
ptep_test_and_clear_user_dirty and add ptep_test_and_clear_user_young.
7) Define mm_exclusive() and mm_has_pgste() helper to improve readability.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
The noexec support on s390 does not rely on a bit in the page table
entry but utilizes the secondary space mode to distinguish between
memory accesses for instructions vs. data. The noexec code relies
on the assumption that the cpu will always use the secondary space
page table for data accesses while it is running in the secondary
space mode. Up to the z9-109 class machines this has been the case.
Unfortunately this is not true anymore with z10 and later machines.
The load-relative-long instructions lrl, lgrl and lgfrl access the
memory operand using the same addressing-space mode that has been
used to fetch the instruction.
This breaks the noexec mode for all user space binaries compiled
with march=z10 or later. The only option is to remove the current
noexec support.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Introduce user_mode to replace the two variables switch_amode and
s390_noexec. There are three valid combinations of the old values:
1) switch_amode == 0 && s390_noexec == 0
2) switch_amode == 1 && s390_noexec == 0
3) switch_amode == 1 && s390_noexec == 1
They get replaced by
1) user_mode == HOME_SPACE_MODE
2) user_mode == PRIMARY_SPACE_MODE
3) user_mode == SECONDARY_SPACE_MODE
The new kernel parameter user_mode=[primary,secondary,home] lets
you choose the address space mode the user space processes should
use. In addition the CONFIG_S390_SWITCH_AMODE config option
is removed.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Suzuki Poulose reported the following recursive locking bug on s390:
Here is the stack trace : (see Appendix I for more info)
[<0000000000406ed6>] _spin_lock+0x52/0x94
[<0000000000103bde>] crst_table_free+0x14e/0x1a4
[<00000000001ba684>] __pmd_alloc+0x114/0x1ec
[<00000000001be8d0>] handle_mm_fault+0x2cc/0xb80
[<0000000000407d62>] do_dat_exception+0x2b6/0x3a0
[<0000000000114f8c>] sysc_return+0x0/0x8
[<00000200001642b2>] 0x200001642b2
The page_table_lock is already acquired in __pmd_alloc (mm/memory.c) and
it tries to populate the pud/pgd with a new pmd allocated. If another
thread populates it before we get a chance, we free the pmd using
pmd_free().
On s390x, pmd_free(even pud_free ) is #defined to crst_table_free(),
which acquires the page_table_lock to protect the crst_table index updates.
Hence this ends up in a recursive locking of the page_table_lock.
The solution suggested by Dave Hansen is to use a new spin lock in the mmu
context to protect the access to the crst_list and the pgtable_list.
Reported-by: Suzuki Poulose <suzuki@in.ibm.com>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Tell the compile that the clear_table inline assembly writes to the
memory referenced by *s.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
