Pull x86 mm cleanup from Thomas Gleixner:
"Use PAGE_ALIGNED() instead of open coding it in the x86/mm code"
* tag 'x86-mm-2022-06-05' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mm: Use PAGE_ALIGNED(x) instead of IS_ALIGNED(x, PAGE_SIZE)
Pull libnvdimm and DAX updates from Dan Williams:
"New support for clearing memory errors when a file is in DAX mode,
alongside with some other fixes and cleanups.
Previously it was only possible to clear these errors using a truncate
or hole-punch operation to trigger the filesystem to reallocate the
block, now, any page aligned write can opportunistically clear errors
as well.
This change spans x86/mm, nvdimm, and fs/dax, and has received the
appropriate sign-offs. Thanks to Jane for her work on this.
Summary:
- Add support for clearing memory error via pwrite(2) on DAX
- Fix 'security overwrite' support in the presence of media errors
- Miscellaneous cleanups and fixes for nfit_test (nvdimm unit tests)"
* tag 'libnvdimm-for-5.19' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
pmem: implement pmem_recovery_write()
pmem: refactor pmem_clear_poison()
dax: add .recovery_write dax_operation
dax: introduce DAX_RECOVERY_WRITE dax access mode
mce: fix set_mce_nospec to always unmap the whole page
x86/mce: relocate set{clear}_mce_nospec() functions
acpi/nfit: rely on mce->misc to determine poison granularity
testing: nvdimm: asm/mce.h is not needed in nfit.c
testing: nvdimm: iomap: make __nfit_test_ioremap a macro
nvdimm: Allow overwrite in the presence of disabled dimms
tools/testing/nvdimm: remove unneeded flush_workqueue
Pull MM updates from Andrew Morton:
"Almost all of MM here. A few things are still getting finished off,
reviewed, etc.
- Yang Shi has improved the behaviour of khugepaged collapsing of
readonly file-backed transparent hugepages.
- Johannes Weiner has arranged for zswap memory use to be tracked and
managed on a per-cgroup basis.
- Munchun Song adds a /proc knob ("hugetlb_optimize_vmemmap") for
runtime enablement of the recent huge page vmemmap optimization
feature.
- Baolin Wang contributes a series to fix some issues around hugetlb
pagetable invalidation.
- Zhenwei Pi has fixed some interactions between hwpoisoned pages and
virtualization.
- Tong Tiangen has enabled the use of the presently x86-only
page_table_check debugging feature on arm64 and riscv.
- David Vernet has done some fixup work on the memcg selftests.
- Peter Xu has taught userfaultfd to handle write protection faults
against shmem- and hugetlbfs-backed files.
- More DAMON development from SeongJae Park - adding online tuning of
the feature and support for monitoring of fixed virtual address
ranges. Also easier discovery of which monitoring operations are
available.
- Nadav Amit has done some optimization of TLB flushing during
mprotect().
- Neil Brown continues to labor away at improving our swap-over-NFS
support.
- David Hildenbrand has some fixes to anon page COWing versus
get_user_pages().
- Peng Liu fixed some errors in the core hugetlb code.
- Joao Martins has reduced the amount of memory consumed by
device-dax's compound devmaps.
- Some cleanups of the arch-specific pagemap code from Anshuman
Khandual.
- Muchun Song has found and fixed some errors in the TLB flushing of
transparent hugepages.
- Roman Gushchin has done more work on the memcg selftests.
... and, of course, many smaller fixes and cleanups. Notably, the
customary million cleanup serieses from Miaohe Lin"
* tag 'mm-stable-2022-05-25' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (381 commits)
mm: kfence: use PAGE_ALIGNED helper
selftests: vm: add the "settings" file with timeout variable
selftests: vm: add "test_hmm.sh" to TEST_FILES
selftests: vm: check numa_available() before operating "merge_across_nodes" in ksm_tests
selftests: vm: add migration to the .gitignore
selftests/vm/pkeys: fix typo in comment
ksm: fix typo in comment
selftests: vm: add process_mrelease tests
Revert "mm/vmscan: never demote for memcg reclaim"
mm/kfence: print disabling or re-enabling message
include/trace/events/percpu.h: cleanup for "percpu: improve percpu_alloc_percpu event trace"
include/trace/events/mmflags.h: cleanup for "tracing: incorrect gfp_t conversion"
mm: fix a potential infinite loop in start_isolate_page_range()
MAINTAINERS: add Muchun as co-maintainer for HugeTLB
zram: fix Kconfig dependency warning
mm/shmem: fix shmem folio swapoff hang
cgroup: fix an error handling path in alloc_pagecache_max_30M()
mm: damon: use HPAGE_PMD_SIZE
tracing: incorrect isolate_mote_t cast in mm_vmscan_lru_isolate
nodemask.h: fix compilation error with GCC12
...
Pull dma-mapping updates from Christoph Hellwig:
- don't over-decrypt memory (Robin Murphy)
- takes min align mask into account for the swiotlb max mapping size
(Tianyu Lan)
- use GFP_ATOMIC in dma-debug (Mikulas Patocka)
- fix DMA_ATTR_NO_KERNEL_MAPPING on xen/arm (me)
- don't fail on highmem CMA pages in dma_direct_alloc_pages (me)
- cleanup swiotlb initialization and share more code with swiotlb-xen
(me, Stefano Stabellini)
* tag 'dma-mapping-5.19-2022-05-25' of git://git.infradead.org/users/hch/dma-mapping: (23 commits)
dma-direct: don't over-decrypt memory
swiotlb: max mapping size takes min align mask into account
swiotlb: use the right nslabs-derived sizes in swiotlb_init_late
swiotlb: use the right nslabs value in swiotlb_init_remap
swiotlb: don't panic when the swiotlb buffer can't be allocated
dma-debug: change allocation mode from GFP_NOWAIT to GFP_ATIOMIC
dma-direct: don't fail on highmem CMA pages in dma_direct_alloc_pages
swiotlb-xen: fix DMA_ATTR_NO_KERNEL_MAPPING on arm
x86: remove cruft from <asm/dma-mapping.h>
swiotlb: remove swiotlb_init_with_tbl and swiotlb_init_late_with_tbl
swiotlb: merge swiotlb-xen initialization into swiotlb
swiotlb: provide swiotlb_init variants that remap the buffer
swiotlb: pass a gfp_mask argument to swiotlb_init_late
swiotlb: add a SWIOTLB_ANY flag to lift the low memory restriction
swiotlb: make the swiotlb_init interface more useful
x86: centralize setting SWIOTLB_FORCE when guest memory encryption is enabled
x86: remove the IOMMU table infrastructure
MIPS/octeon: use swiotlb_init instead of open coding it
arm/xen: don't check for xen_initial_domain() in xen_create_contiguous_region
swiotlb: rename swiotlb_late_init_with_default_size
...
Pull misc x86 updates from Borislav Petkov:
"A variety of fixes which don't fit any other tip bucket:
- Remove unnecessary function export
- Correct asm constraint
- Fix __setup handlers retval"
* tag 'x86_misc_for_v5.19_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mm: Cleanup the control_va_addr_alignment() __setup handler
x86: Fix return value of __setup handlers
x86/delay: Fix the wrong asm constraint in delay_loop()
x86/amd_nb: Unexport amd_cache_northbridges()
Pull x86 mm fixlet from Borislav Petkov:
- A sparse address space annotation fix
* tag 'x86_mm_for_v5.19_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/fault: Cast an argument to the proper address space in prefetch()
Pull x86 cleanups from Borislav Petkov:
- Serious sanitization and cleanup of the whole APERF/MPERF and
frequency invariance code along with removing the need for
unnecessary IPIs
- Finally remove a.out support
- The usual trivial cleanups and fixes all over x86
* tag 'x86_cleanups_for_v5.19_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (21 commits)
x86: Remove empty files
x86/speculation: Add missing srbds=off to the mitigations= help text
x86/prctl: Remove pointless task argument
x86/aperfperf: Make it correct on 32bit and UP kernels
x86/aperfmperf: Integrate the fallback code from show_cpuinfo()
x86/aperfmperf: Replace arch_freq_get_on_cpu()
x86/aperfmperf: Replace aperfmperf_get_khz()
x86/aperfmperf: Store aperf/mperf data for cpu frequency reads
x86/aperfmperf: Make parts of the frequency invariance code unconditional
x86/aperfmperf: Restructure arch_scale_freq_tick()
x86/aperfmperf: Put frequency invariance aperf/mperf data into a struct
x86/aperfmperf: Untangle Intel and AMD frequency invariance init
x86/aperfmperf: Separate AP/BP frequency invariance init
x86/smp: Move APERF/MPERF code where it belongs
x86/aperfmperf: Dont wake idle CPUs in arch_freq_get_on_cpu()
x86/process: Fix kernel-doc warning due to a changed function name
x86: Remove a.out support
x86/mm: Replace nodes_weight() with nodes_empty() where appropriate
x86: Replace cpumask_weight() with cpumask_empty() where appropriate
x86/pkeys: Remove __arch_set_user_pkey_access() declaration
...
Pull x86 CPU feature updates from Borislav Petkov:
- Remove a bunch of chicken bit options to turn off CPU features which
are not really needed anymore
- Misc fixes and cleanups
* tag 'x86_cpu_for_v5.19_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/speculation: Add missing prototype for unpriv_ebpf_notify()
x86/pm: Fix false positive kmemleak report in msr_build_context()
x86/speculation/srbds: Do not try to turn mitigation off when not supported
x86/cpu: Remove "noclflush"
x86/cpu: Remove "noexec"
x86/cpu: Remove "nosmep"
x86/cpu: Remove CONFIG_X86_SMAP and "nosmap"
x86/cpu: Remove "nosep"
x86/cpu: Allow feature bit names from /proc/cpuinfo in clearcpuid=
Pull Intel TDX support from Borislav Petkov:
"Intel Trust Domain Extensions (TDX) support.
This is the Intel version of a confidential computing solution called
Trust Domain Extensions (TDX). This series adds support to run the
kernel as part of a TDX guest. It provides similar guest protections
to AMD's SEV-SNP like guest memory and register state encryption,
memory integrity protection and a lot more.
Design-wise, it differs from AMD's solution considerably: it uses a
software module which runs in a special CPU mode called (Secure
Arbitration Mode) SEAM. As the name suggests, this module serves as
sort of an arbiter which the confidential guest calls for services it
needs during its lifetime.
Just like AMD's SNP set, this series reworks and streamlines certain
parts of x86 arch code so that this feature can be properly
accomodated"
* tag 'x86_tdx_for_v5.19_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (34 commits)
x86/tdx: Fix RETs in TDX asm
x86/tdx: Annotate a noreturn function
x86/mm: Fix spacing within memory encryption features message
x86/kaslr: Fix build warning in KASLR code in boot stub
Documentation/x86: Document TDX kernel architecture
ACPICA: Avoid cache flush inside virtual machines
x86/tdx/ioapic: Add shared bit for IOAPIC base address
x86/mm: Make DMA memory shared for TD guest
x86/mm/cpa: Add support for TDX shared memory
x86/tdx: Make pages shared in ioremap()
x86/topology: Disable CPU online/offline control for TDX guests
x86/boot: Avoid #VE during boot for TDX platforms
x86/boot: Set CR0.NE early and keep it set during the boot
x86/acpi/x86/boot: Add multiprocessor wake-up support
x86/boot: Add a trampoline for booting APs via firmware handoff
x86/tdx: Wire up KVM hypercalls
x86/tdx: Port I/O: Add early boot support
x86/tdx: Port I/O: Add runtime hypercalls
x86/boot: Port I/O: Add decompression-time support for TDX
x86/boot: Port I/O: Allow to hook up alternative helpers
...
Pull AMD SEV-SNP support from Borislav Petkov:
"The third AMD confidential computing feature called Secure Nested
Paging.
Add to confidential guests the necessary memory integrity protection
against malicious hypervisor-based attacks like data replay, memory
remapping and others, thus achieving a stronger isolation from the
hypervisor.
At the core of the functionality is a new structure called a reverse
map table (RMP) with which the guest has a say in which pages get
assigned to it and gets notified when a page which it owns, gets
accessed/modified under the covers so that the guest can take an
appropriate action.
In addition, add support for the whole machinery needed to launch a
SNP guest, details of which is properly explained in each patch.
And last but not least, the series refactors and improves parts of the
previous SEV support so that the new code is accomodated properly and
not just bolted on"
* tag 'x86_sev_for_v5.19_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (60 commits)
x86/entry: Fixup objtool/ibt validation
x86/sev: Mark the code returning to user space as syscall gap
x86/sev: Annotate stack change in the #VC handler
x86/sev: Remove duplicated assignment to variable info
x86/sev: Fix address space sparse warning
x86/sev: Get the AP jump table address from secrets page
x86/sev: Add missing __init annotations to SEV init routines
virt: sevguest: Rename the sevguest dir and files to sev-guest
virt: sevguest: Change driver name to reflect generic SEV support
x86/boot: Put globals that are accessed early into the .data section
x86/boot: Add an efi.h header for the decompressor
virt: sevguest: Fix bool function returning negative value
virt: sevguest: Fix return value check in alloc_shared_pages()
x86/sev-es: Replace open-coded hlt-loop with sev_es_terminate()
virt: sevguest: Add documentation for SEV-SNP CPUID Enforcement
virt: sevguest: Add support to get extended report
virt: sevguest: Add support to derive key
virt: Add SEV-SNP guest driver
x86/sev: Register SEV-SNP guest request platform device
x86/sev: Provide support for SNP guest request NAEs
...
The set_memory_uc() approach doesn't work well in all cases.
As Dan pointed out when "The VMM unmapped the bad page from
guest physical space and passed the machine check to the guest."
"The guest gets virtual #MC on an access to that page. When
the guest tries to do set_memory_uc() and instructs cpa_flush()
to do clean caches that results in taking another fault / exception
perhaps because the VMM unmapped the page from the guest."
Since the driver has special knowledge to handle NP or UC,
mark the poisoned page with NP and let driver handle it when
it comes down to repair.
Please refer to discussions here for more details.
https://lore.kernel.org/all/CAPcyv4hrXPb1tASBZUg-GgdVs0OOFKXMXLiHmktg_kFi7YBMyQ@mail.gmail.com/
Now since poisoned page is marked as not-present, in order to
avoid writing to a not-present page and trigger kernel Oops,
also fix pmem_do_write().
Fixes: 284ce4011b ("x86/memory_failure: Introduce {set, clear}_mce_nospec()")
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Jane Chu <jane.chu@oracle.com>
Acked-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/165272615484.103830.2563950688772226611.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Calls to change_protection_range() on THP can trigger, at least on x86,
two TLB flushes for one page: one immediately, when pmdp_invalidate() is
called by change_huge_pmd(), and then another one later (that can be
batched) when change_protection_range() finishes.
The first TLB flush is only necessary to prevent the dirty bit (and with a
lesser importance the access bit) from changing while the PTE is modified.
However, this is not necessary as the x86 CPUs set the dirty-bit
atomically with an additional check that the PTE is (still) present. One
caveat is Intel's Knights Landing that has a bug and does not do so.
Leverage this behavior to eliminate the unnecessary TLB flush in
change_huge_pmd(). Introduce a new arch specific pmdp_invalidate_ad()
that only invalidates the access and dirty bit from further changes.
Link: https://lkml.kernel.org/r/20220401180821.1986781-4-namit@vmware.com
Signed-off-by: Nadav Amit <namit@vmware.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Andrew Cooper <andrew.cooper3@citrix.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will@kernel.org>
Cc: Yu Zhao <yuzhao@google.com>
Cc: Nick Piggin <npiggin@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
__setup() handlers should return 1 to obsolete_checksetup() in
init/main.c to indicate that the boot option has been handled. A return
of 0 causes the boot option/value to be listed as an Unknown kernel
parameter and added to init's (limited) argument (no '=') or environment
(with '=') strings. So return 1 from these x86 __setup handlers.
Examples:
Unknown kernel command line parameters "apicpmtimer
BOOT_IMAGE=/boot/bzImage-517rc8 vdso=1 ring3mwait=disable", will be
passed to user space.
Run /sbin/init as init process
with arguments:
/sbin/init
apicpmtimer
with environment:
HOME=/
TERM=linux
BOOT_IMAGE=/boot/bzImage-517rc8
vdso=1
ring3mwait=disable
Fixes: 2aae950b21 ("x86_64: Add vDSO for x86-64 with gettimeofday/clock_gettime/getcpu")
Fixes: 77b52b4c5c ("x86: add "debugpat" boot option")
Fixes: e16fd002af ("x86/cpufeature: Enable RING3MWAIT for Knights Landing")
Fixes: b8ce335906 ("x86_64: convert to clock events")
Reported-by: Igor Zhbanov <i.zhbanov@omprussia.ru>
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lore.kernel.org/r/64644a2f-4a20-bab3-1e15-3b2cdd0defe3@omprussia.ru
Link: https://lore.kernel.org/r/20220314012725.26661-1-rdunlap@infradead.org
Drop lookup_address_in_mm() now that KVM is providing it's own variant
of lookup_address_in_pgd() that is safe for use with user addresses, e.g.
guards against page tables being torn down. A variant that provides a
non-init mm is inherently dangerous and flawed, as the only reason to use
an mm other than init_mm is to walk a userspace mapping, and
lookup_address_in_pgd() does not play nice with userspace mappings, e.g.
doesn't disable IRQs to block TLB shootdowns and doesn't use READ_ONCE()
to ensure an upper level entry isn't converted to a huge page between
checking the PAGE_SIZE bit and grabbing the address of the next level
down.
This reverts commit 13c72c060f.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <YmwIi3bXr/1yhYV/@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Move enabling SWIOTLB_FORCE for guest memory encryption into common code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Tested-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Various mm code calls nodes_weight() to check if any bit of a given
nodemask is set.
This can be done more efficiently with nodes_empty() because nodes_empty()
stops traversing the nodemask as soon as it finds first set bit, while
nodes_weight() counts all bits unconditionally.
Signed-off-by: Yury Norov <yury.norov@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20220210224933.379149-26-yury.norov@gmail.com
In some cases, x86 code calls cpumask_weight() to check if any bit of a
given cpumask is set.
This can be done more efficiently with cpumask_empty() because
cpumask_empty() stops traversing the cpumask as soon as it finds first set
bit, while cpumask_weight() counts all bits unconditionally.
Signed-off-by: Yury Norov <yury.norov@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Steve Wahl <steve.wahl@hpe.com>
Link: https://lore.kernel.org/r/20220210224933.379149-17-yury.norov@gmail.com
Intel TDX doesn't allow VMM to directly access guest private memory.
Any memory that is required for communication with the VMM must be
shared explicitly. The same rule applies for any DMA to and from the
TDX guest. All DMA pages have to be marked as shared pages. A generic way
to achieve this without any changes to device drivers is to use the
SWIOTLB framework.
The previous patch ("Add support for TDX shared memory") gave TDX guests
the _ability_ to make some pages shared, but did not make any pages
shared. This actually marks SWIOTLB buffers *as* shared.
Start returning true for cc_platform_has(CC_ATTR_GUEST_MEM_ENCRYPT) in
TDX guests. This has several implications:
- Allows the existing mem_encrypt_init() to be used for TDX which
sets SWIOTLB buffers shared (aka. "decrypted").
- Ensures that all DMA is routed via the SWIOTLB mechanism (see
pci_swiotlb_detect())
Stop selecting DYNAMIC_PHYSICAL_MASK directly. It will get set
indirectly by selecting X86_MEM_ENCRYPT.
mem_encrypt_init() is currently under an AMD-specific #ifdef. Move it to
a generic area of the header.
Co-developed-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
Signed-off-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Dave Hansen <dave.hansen@linux.intel.com>
Link: https://lkml.kernel.org/r/20220405232939.73860-28-kirill.shutemov@linux.intel.com
early_set_memory_{encrypted,decrypted}() are used for changing the page
state from decrypted (shared) to encrypted (private) and vice versa.
When SEV-SNP is active, the page state transition needs to go through
additional steps.
If the page is transitioned from shared to private, then perform the
following after the encryption attribute is set in the page table:
1. Issue the page state change VMGEXIT to add the page as a private
in the RMP table.
2. Validate the page after its successfully added in the RMP table.
To maintain the security guarantees, if the page is transitioned from
private to shared, then perform the following before clearing the
encryption attribute from the page table.
1. Invalidate the page.
2. Issue the page state change VMGEXIT to make the page shared in the
RMP table.
early_set_memory_{encrypted,decrypted}() can be called before the GHCB
is setup so use the SNP page state MSR protocol VMGEXIT defined in the
GHCB specification to request the page state change in the RMP table.
While at it, add a helper snp_prep_memory() which will be used in
probe_roms(), in a later patch.
[ bp: Massage commit message. ]
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Venu Busireddy <venu.busireddy@oracle.com>
Link: https://lore.kernel.org/r/20220307213356.2797205-19-brijesh.singh@amd.com
Commit in Fixes uses accessors based on the access mode, i.e., it
distinguishes its access if instr carries a user address or a kernel
address.
Since that commit, sparse complains about passing an argument without
__user annotation to get_user(), which expects a pointer of the __user
address space:
arch/x86/mm/fault.c:152:29: warning: incorrect type in argument 1 (different address spaces)
arch/x86/mm/fault.c:152:29: expected void const volatile [noderef] __user *ptr
arch/x86/mm/fault.c:152:29: got unsigned char *[assigned] instr
Cast instr to __user when accessing user memory.
No functional change. No change in the generated object code.
[ bp: Simplify commit message. ]
Fixes: 35f1c89b0c ("x86/fault: Fix AMD erratum #91 errata fixup for user code")
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lore.kernel.org/r/20220201144055.5670-1-lukas.bulwahn@gmail.com
0day reported a regression on a microbenchmark which is intended to
stress the TLB flushing path:
https://lore.kernel.org/all/20220317090415.GE735@xsang-OptiPlex-9020/
It pointed at a commit from Nadav which intended to remove retpoline
overhead in the TLB flushing path by taking the 'cond'-ition in
on_each_cpu_cond_mask(), pre-calculating it, and incorporating it into
'cpumask'. That allowed the code to use a bunch of earlier direct
calls instead of later indirect calls that need a retpoline.
But, in practice, threads can go idle (and into lazy TLB mode where
they don't need to flush their TLB) between the early and late calls.
It works in this direction and not in the other because TLB-flushing
threads tend to hold mmap_lock for write. Contention on that lock
causes threads to _go_ idle right in this early/late window.
There was not any performance data in the original commit specific
to the retpoline overhead. I did a few tests on a system with
retpolines:
https://lore.kernel.org/all/dd8be93c-ded6-b962-50d4-96b1c3afb2b7@intel.com/
which showed a possible small win. But, that small win pales in
comparison with the bigger loss induced on non-retpoline systems.
Revert the patch that removed the retpolines. This was not a
clean revert, but it was self-contained enough not to be too painful.
Fixes: 6035152d8e ("x86/mm/tlb: Open-code on_each_cpu_cond_mask() for tlb_is_not_lazy()")
Reported-by: kernel test robot <oliver.sang@intel.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Nadav Amit <namit@vmware.com>
Cc: <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/164874672286.389.7021457716635788197.tip-bot2@tip-bot2
Pull ptrace cleanups from Eric Biederman:
"This set of changes removes tracehook.h, moves modification of all of
the ptrace fields inside of siglock to remove races, adds a missing
permission check to ptrace.c
The removal of tracehook.h is quite significant as it has been a major
source of confusion in recent years. Much of that confusion was around
task_work and TIF_NOTIFY_SIGNAL (which I have now decoupled making the
semantics clearer).
For people who don't know tracehook.h is a vestiage of an attempt to
implement uprobes like functionality that was never fully merged, and
was later superseeded by uprobes when uprobes was merged. For many
years now we have been removing what tracehook functionaly a little
bit at a time. To the point where anything left in tracehook.h was
some weird strange thing that was difficult to understand"
* tag 'ptrace-cleanups-for-v5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace:
ptrace: Remove duplicated include in ptrace.c
ptrace: Check PTRACE_O_SUSPEND_SECCOMP permission on PTRACE_SEIZE
ptrace: Return the signal to continue with from ptrace_stop
ptrace: Move setting/clearing ptrace_message into ptrace_stop
tracehook: Remove tracehook.h
resume_user_mode: Move to resume_user_mode.h
resume_user_mode: Remove #ifdef TIF_NOTIFY_RESUME in set_notify_resume
signal: Move set_notify_signal and clear_notify_signal into sched/signal.h
task_work: Decouple TIF_NOTIFY_SIGNAL and task_work
task_work: Call tracehook_notify_signal from get_signal on all architectures
task_work: Introduce task_work_pending
task_work: Remove unnecessary include from posix_timers.h
ptrace: Remove tracehook_signal_handler
ptrace: Remove arch_syscall_{enter,exit}_tracehook
ptrace: Create ptrace_report_syscall_{entry,exit} in ptrace.h
ptrace/arm: Rename tracehook_report_syscall report_syscall
ptrace: Move ptrace_report_syscall into ptrace.h
On x86, prior to ("mm: handle uninitialized numa nodes gracecully"), NUMA
nodes could be allocated at three different places.
- numa_register_memblks
- init_cpu_to_node
- init_gi_nodes
All these calls happen at setup_arch, and have the following order:
setup_arch
...
x86_numa_init
numa_init
numa_register_memblks
...
init_cpu_to_node
init_memory_less_node
alloc_node_data
free_area_init_memoryless_node
init_gi_nodes
init_memory_less_node
alloc_node_data
free_area_init_memoryless_node
numa_register_memblks() is only interested in those nodes which have
memory, so it skips over any memoryless node it founds. Later on, when
we have read ACPI's SRAT table, we call init_cpu_to_node() and
init_gi_nodes(), which initialize any memoryless node we might have that
have either CPU or Initiator affinity, meaning we allocate pg_data_t
struct for them and we mark them as ONLINE.
So far so good, but the thing is that after ("mm: handle uninitialized
numa nodes gracefully"), we allocate all possible NUMA nodes in
free_area_init(), meaning we have a picture like the following:
setup_arch
x86_numa_init
numa_init
numa_register_memblks <-- allocate non-memoryless node
x86_init.paging.pagetable_init
...
free_area_init
free_area_init_memoryless <-- allocate memoryless node
init_cpu_to_node
alloc_node_data <-- allocate memoryless node with CPU
free_area_init_memoryless_node
init_gi_nodes
alloc_node_data <-- allocate memoryless node with Initiator
free_area_init_memoryless_node
free_area_init() already allocates all possible NUMA nodes, but
init_cpu_to_node() and init_gi_nodes() are clueless about that, so they
go ahead and allocate a new pg_data_t struct without checking anything,
meaning we end up allocating twice.
It should be mad clear that this only happens in the case where
memoryless NUMA node happens to have a CPU/Initiator affinity.
So get rid of init_memory_less_node() and just set the node online.
Note that setting the node online is needed, otherwise we choke down the
chain when bringup_nonboot_cpus() ends up calling
__try_online_node()->register_one_node()->... and we blow up in
bus_add_device(). As can be seen here:
BUG: kernel NULL pointer dereference, address: 0000000000000060
#PF: supervisor read access in kernel mode
#PF: error_code(0x0000) - not-present page
PGD 0 P4D 0
Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC PTI
CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.17.0-rc4-1-default+ #45
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.0.0-prebuilt.qemu-project.org 04/4
RIP: 0010:bus_add_device+0x5a/0x140
Code: 8b 74 24 20 48 89 df e8 84 96 ff ff 85 c0 89 c5 75 38 48 8b 53 50 48 85 d2 0f 84 bb 00 004
RSP: 0000:ffffc9000022bd10 EFLAGS: 00010246
RAX: 0000000000000000 RBX: ffff888100987400 RCX: ffff8881003e4e19
RDX: ffff8881009a5e00 RSI: ffff888100987400 RDI: ffff888100987400
RBP: 0000000000000000 R08: ffff8881003e4e18 R09: ffff8881003e4c98
R10: 0000000000000000 R11: ffff888100402bc0 R12: ffffffff822ceba0
R13: 0000000000000000 R14: ffff888100987400 R15: 0000000000000000
FS: 0000000000000000(0000) GS:ffff88853fc00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000060 CR3: 000000000200a001 CR4: 00000000001706b0
Call Trace:
device_add+0x4c0/0x910
__register_one_node+0x97/0x2d0
__try_online_node+0x85/0xc0
try_online_node+0x25/0x40
cpu_up+0x4f/0x100
bringup_nonboot_cpus+0x4f/0x60
smp_init+0x26/0x79
kernel_init_freeable+0x130/0x2f1
kernel_init+0x17/0x150
ret_from_fork+0x22/0x30
The reason is simple, by the time bringup_nonboot_cpus() gets called, we
did not register the node_subsys bus yet, so we crash when
bus_add_device() tries to dereference bus()->p.
The following shows the order of the calls:
kernel_init_freeable
smp_init
bringup_nonboot_cpus
...
bus_add_device() <- we did not register node_subsys yet
do_basic_setup
do_initcalls
postcore_initcall(register_node_type);
register_node_type
subsys_system_register
subsys_register
bus_register <- register node_subsys bus
Why setting the node online saves us then? Well, simply because
__try_online_node() backs off when the node is online, meaning we do not
end up calling register_one_node() in the first place.
This is subtle, broken and deserves a deep analysis and thought about
how to put this into shape, but for now let us have this easy fix for
the leaking memory issue.
[osalvador@suse.de: add comments]
Link: https://lkml.kernel.org/r/20220221142649.3457-1-osalvador@suse.de
Link: https://lkml.kernel.org/r/20220218224302.5282-2-osalvador@suse.de
Fixes: da4490c958ad ("mm: handle uninitialized numa nodes gracefully")
Signed-off-by: Oscar Salvador <osalvador@suse.de>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Rafael Aquini <raquini@redhat.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Wei Yang <richard.weiyang@gmail.com>
Cc: Dennis Zhou <dennis@kernel.org>
Cc: Alexey Makhalov <amakhalov@vmware.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull x86 perf event updates from Ingo Molnar:
- Fix address filtering for Intel/PT,ARM/CoreSight
- Enable Intel/PEBS format 5
- Allow more fixed-function counters for x86
- Intel/PT: Enable not recording Taken-Not-Taken packets
- Add a few branch-types
* tag 'perf-core-2022-03-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86/intel/uncore: Fix the build on !CONFIG_PHYS_ADDR_T_64BIT
perf: Add irq and exception return branch types
perf/x86/intel/uncore: Make uncore_discovery clean for 64 bit addresses
perf/x86/intel/pt: Add a capability and config bit for disabling TNTs
perf/x86/intel/pt: Add a capability and config bit for event tracing
perf/x86/intel: Increase max number of the fixed counters
KVM: x86: use the KVM side max supported fixed counter
perf/x86/intel: Enable PEBS format 5
perf/core: Allow kernel address filter when not filtering the kernel
perf/x86/intel/pt: Fix address filter config for 32-bit kernel
perf/core: Fix address filter parser for multiple filters
x86: Share definition of __is_canonical_address()
perf/x86/intel/pt: Relax address filter validation
Pull x86 cleanups from Borislav Petkov:
- Remove a misleading message and an unused function
* tag 'x86_cleanups_for_v5.18_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/nmi: Remove the 'strange power saving mode' hint from unknown NMI handler
x86/pat: Remove the unused set_pages_array_wt() function
Pull x86 confidential computing updates from Borislav Petkov:
- Add shared confidential computing code which will be used by both
vendors instead of proliferating home-grown solutions for
technologies (SEV/SNP and TDX) which are pretty similar
* tag 'x86_cc_for_v5.18_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mm/cpa: Generalize __set_memory_enc_pgtable()
x86/coco: Add API to handle encryption mask
x86/coco: Explicitly declare type of confidential computing platform
x86/cc: Move arch/x86/{kernel/cc_platform.c => coco/core.c}
Break a header file circular dependency by removing the unnecessary
include of task_work.h from posix_timers.h.
sched.h -> posix-timers.h
posix-timers.h -> task_work.h
task_work.h -> sched.h
Add missing includes of task_work.h to:
arch/x86/mm/tlb.c
kernel/time/posix-cpu-timers.c
Reviewed-by: Kees Cook <keescook@chromium.org>
Link: https://lkml.kernel.org/r/20220309162454.123006-6-ebiederm@xmission.com
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
As documented, the setup_indirect structure is nested inside
the setup_data structures in the setup_data list. The code currently
accesses the fields inside the setup_indirect structure but only
the sizeof(struct setup_data) is being memremapped. No crash
occurred but this is just due to how the area is remapped under the
covers.
Properly memremap both the setup_data and setup_indirect structures
in these cases before accessing them.
Fixes: b3c72fc9a7 ("x86/boot: Introduce setup_indirect")
Signed-off-by: Ross Philipson <ross.philipson@oracle.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Daniel Kiper <daniel.kiper@oracle.com>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/1645668456-22036-2-git-send-email-ross.philipson@oracle.com
The kernel provides infrastructure to set or clear the encryption mask
from the pages for AMD SEV, but TDX requires few tweaks.
- TDX and SEV have different requirements to the cache and TLB
flushing.
- TDX has own routine to notify VMM about page encryption status change.
Modify __set_memory_enc_pgtable() and make it flexible enough to cover
both AMD SEV and Intel TDX. The AMD-specific behavior is isolated in the
callbacks under x86_platform.guest. TDX will provide own version of said
callbacks.
[ bp: Beat into submission. ]
Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Link: https://lore.kernel.org/r/20220223043528.2093214-1-brijesh.singh@amd.com
AMD SME/SEV uses a bit in the page table entries to indicate that the
page is encrypted and not accessible to the VMM.
TDX uses a similar approach, but the polarity of the mask is opposite to
AMD: if the bit is set the page is accessible to VMM.
Provide vendor-neutral API to deal with the mask: cc_mkenc() and
cc_mkdec() modify given address to make it encrypted/decrypted. It can
be applied to phys_addr_t, pgprotval_t or page table entry value.
pgprot_encrypted() and pgprot_decrypted() reimplemented using new
helpers.
The implementation will be extended to cover TDX.
pgprot_decrypted() is used by drivers (i915, virtio_gpu, vfio).
cc_mkdec() called by pgprot_decrypted(). Export cc_mkdec().
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lore.kernel.org/r/20220222185740.26228-5-kirill.shutemov@linux.intel.com
The kernel derives the confidential computing platform
type it is running as from sme_me_mask on AMD or by using
hv_is_isolation_supported() on HyperV isolation VMs. This detection
process will be more complicated as more platforms get added.
Declare a confidential computing vendor variable explicitly and set it
via cc_set_vendor() on the respective platform.
[ bp: Massage commit message, fixup HyperV check. ]
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lore.kernel.org/r/20220222185740.26228-4-kirill.shutemov@linux.intel.com
