| Age | Commit message (Collapse) | Author |
|---|
|
David Hildenbrand says:
===================
This series fixes one issue with uffd + shared zeropages on s390x and
fixes that "ordinary" KVM guests can make use of shared zeropages again.
userfaultfd could currently end up mapping shared zeropages into processes
that forbid shared zeropages. This only apples to s390x, relevant for
handling PV guests and guests that use storage kets correctly. Fix it
by placing a zeroed folio instead of the shared zeropage during
UFFDIO_ZEROPAGE instead.
I stumbled over this issue while looking into a customer scenario that
is using:
(1) Memory ballooning for dynamic resizing. Start a VM with, say, 100 GiB
and inflate the balloon during boot to 60 GiB. The VM has ~40 GiB
available and additional memory can be "fake hotplugged" to the VM
later on demand by deflating the balloon. Actual memory overcommit is
not desired, so physical memory would only be moved between VMs.
(2) Live migration of VMs between sites to evacuate servers in case of
emergency.
Without the shared zeropage, during (2), the VM would suddenly consume
100 GiB on the migration source and destination. On the migration source,
where we don't excpect memory overcommit, we could easilt end up crashing
the VM during migration.
Independent of that, memory handed back to the hypervisor using "free page
reporting" would end up consuming actual memory after the migration on the
destination, not getting freed up until reused+freed again.
While there might be ways to optimize parts of this in QEMU, we really
should just support the shared zeropage again for ordinary VMs.
We only expect legcy guests to make use of storage keys, so let's handle
zeropages again when enabling storage keys or when enabling PV. To not
break userfaultfd like we did in the past, don't zap the shared zeropages,
but instead trigger unsharing faults, just like we do for unsharing
KSM pages in break_ksm().
Unsharing faults will simply replace the shared zeropage by a zeroed
anonymous folio. We can already trigger the same fault path using GUP,
when trying to long-term pin a shared zeropage, but also when unmerging
a KSM-placed zeropages, so this is nothing new.
Patch #1 tested on 86-64 by forcing mm_forbids_zeropage() to be 1, and
running the uffd selftests.
Patch #2 tested on s390x: the live migration scenario now works as
expected, and kvm-unit-tests that trigger usage of skeys work well, whereby
I can see detection and unsharing of shared zeropages.
Further (as broken in v2), I tested that the shared zeropage is no
longer populated after skeys are used -- that mm_forbids_zeropage() works
as expected:
./s390x-run s390x/skey.elf \
-no-shutdown \
-chardev socket,id=monitor,path=/var/tmp/mon,server,nowait \
-mon chardev=monitor,mode=readline
Then, in another shell:
# cat /proc/`pgrep qemu`/smaps_rollup | grep Rss
Rss: 31484 kB
# echo "dump-guest-memory tmp" | sudo nc -U /var/tmp/mon
...
# cat /proc/`pgrep qemu`/smaps_rollup | grep Rss
Rss: 160452 kB
-> Reading guest memory does not populate the shared zeropage
Doing the same with selftest.elf (no skeys)
# cat /proc/`pgrep qemu`/smaps_rollup | grep Rss
Rss: 30900 kB
# echo "dump-guest-memory tmp" | sudo nc -U /var/tmp/mon
...
# cat /proc/`pgrep qemu`/smaps_rollup | grep Rsstmp/mon
Rss: 30924 kB
-> Reading guest memory does populate the shared zeropage
===================
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
These are generated files. Prefix them with $(obj)/ instead of $(src)/.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Acked-by: Helge Deller <deller@gmx.de>
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>
|
|
KVM allows issuing the KVM_CHECK_EXTENSION ioctl either on the /dev/kvm
fd or the VM fd. In the first case, kvm_vm_ioctl_check_extension() is
called with kvm==NULL. Ensure we don't dereference the pointer in that
case.
Fixes: 40ebdb8e59df ("KVM: s390: Make huge pages unavailable in ucontrol VMs")
Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Message-ID: <20240419160723.320910-2-jean-philippe@linaro.org>
Signed-off-by: Christian Borntraeger <borntraeger@linux.ibm.com>
|
|
With commit ed6776c96c60 ("s390/crypto: remove retry
loop with sleep from PAES pkey invocation") the retry
loop to retry derivation of a protected key from a
secure key has been removed. This was based on the
assumption that theses retries are not needed any
more as proper retries are done in the zcrypt layer.
However, tests have revealed that there exist some
cases with master key change in the HSM and immediately
(< 1 second) attempt to derive a protected key from a
secure key with exact this HSM may eventually fail.
The low level functions in zcrypt_ccamisc.c and
zcrypt_ep11misc.c detect and report this temporary
failure and report it to the caller as -EBUSY. The
re-established retry loop in the paes implementation
catches exactly this -EBUSY and eventually may run
some retries.
Fixes: ed6776c96c60 ("s390/crypto: remove retry loop with sleep from PAES pkey invocation")
Signed-off-by: Harald Freudenberger <freude@linux.ibm.com>
Reviewed-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Holger Dengler <dengler@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
The address of the crypto control block in the (shadow) SIE block is
absolute/physical.
Convert from virtual to physical when shadowing the guest's control
block during VSIE.
Signed-off-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Link: https://lore.kernel.org/r/20240429171512.879215-1-nsg@linux.ibm.com
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
Currently kernel image relocation tables and other ELF
data are set to base zero. Since kernel virtual and
physical address spaces are uncoupled the kernel is
mapped at the top of the virtual address space, hence
making the information contained in vmlinux ELF tables
inconsistent.
That does not pose any issue with regard to the kernel
booting and operation, but makes it difficult to use a
generated vmlinux with some debugging tools (e.g. gdb).
Relocate vmlinux image base address from zero to a base
address in the virtual address space. It is the address
that kernel is mapped to in cases KASLR is disabled.
The vmlinux ELF header before and after this change looks
like this:
Elf file type is EXEC (Executable file)
Entry point 0x100000
There are 3 program headers, starting at offset 64
Program Headers:
Type Offset VirtAddr PhysAddr
FileSiz MemSiz Flags Align
LOAD 0x0000000000001000 0x0000000000100000 0x0000000000100000
0x0000000001323378 0x0000000001323378 R E 0x1000
LOAD 0x0000000001325000 0x0000000001424000 0x0000000001424000
0x00000000003a4200 0x000000000048fdb8 RWE 0x1000
NOTE 0x00000000012a33b0 0x00000000013a23b0 0x00000000013a23b0
0x0000000000000054 0x0000000000000054 0x4
Elf file type is EXEC (Executable file)
Entry point 0x3ffe0000000
There are 3 program headers, starting at offset 64
Program Headers:
Type Offset VirtAddr PhysAddr
FileSiz MemSiz Flags Align
LOAD 0x0000000000001000 0x000003ffe0000000 0x000003ffe0000000
0x0000000001323378 0x0000000001323378 R E 0x1000
LOAD 0x0000000001325000 0x000003ffe1324000 0x000003ffe1324000
0x00000000003a4200 0x000000000048fdb8 RWE 0x1000
NOTE 0x00000000012a33b0 0x000003ffe12a23b0 0x000003ffe12a23b0
0x0000000000000054 0x0000000000000054 0x4
Suggested-by: Vasily Gorbik <gor@linux.ibm.com>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
When the kernel is built with CONFIG_PIE_BUILD option enabled it
uses dynamic symbols, for which the linker does not allow more
than 64K number of entries. This can break features like kpatch.
Hence, whenever possible the kernel is built with CONFIG_PIE_BUILD
option disabled. For that support of unaligned symbols generated by
linker scripts in the compiler is necessary.
However, older compilers might lack such support. In that case the
build process resorts to CONFIG_PIE_BUILD option-enabled build.
Compile object files with -fPIC option and then link the kernel
binary with -no-pie linker option.
As result, the dynamic symbols are not generated and not only kpatch
feature succeeds, but also the whole CONFIG_PIE_BUILD option-enabled
code could be dropped.
[ agordeev: Reworded the commit message ]
Suggested-by: Ulrich Weigand <ulrich.weigand@de.ibm.com>
Signed-off-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Reviewed-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
Sections .hash and .gnu.hash are only created when CONFIG_PIE_BUILD
option is enabled. Drop these for the case CONFIG_PIE_BUILD is disabled.
[ agordeev: Reworded the commit message ]
Fixes: 778666df60f0 ("s390: compile relocatable kernel without -fPIE")
Suggested-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Reviewed-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
Using __builtin_return_address(n) might return undefined values
when used with values of n outside of the stack. This was noticed
when __builtin_return_address() was called in ftrace on top level
functions like the interrupt handlers.
As this behaviour cannot be fixed, use the s390 stack unwinder and
remove the ftrace compilation flags for unwind_bc.c and stacktrace.c
to prevent the unwinding function polluting function traces.
Another advantage is that this also works with clang.
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
The S/390 architecture doesn't support SMBIOS, so CONFIG_DMI will
never be defined there. So we can simply omit these preprocessing
directives and speed up the build a bit.
Signed-off-by: Jean Delvare <jdelvare@suse.de>
Cc: Niklas Schnelle <schnelle@linux.ibm.com>
Cc: Gerald Schaefer <gerald.schaefer@linux.ibm.com>
Acked-by: Niklas Schnelle <schnelle@linux.ibm.com>
Link: https://lore.kernel.org/r/20240423162724.3966265a@endymion.delvare
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
gcc's -Warray-bounds warned about an out-of-bounds access to
the entry array contained in struct os_info. This doesn't trigger
a bug right now because there's a large reserved space after the
array. Nevertheless fix this, and also add a BUILD_BUG_ON to make
sure struct os_info is always exactly on page in size.
Fixes: f4cac27dc0d6 ("s390/crash: Use old os_info to create PT_LOAD headers")
Reviewed-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
The commit be42660d0c13 ("s390/crash: use old os_info to create PT_LOAD headers")
introduced use of the old os_info into standalone dump kernel.
Before this change os_info_old_init() expected to be called only from
a regular kdump kernel although the function itself is able to work
in standalone dump kernels as well (because copy_oldmem_kernel() is able
to handle both use cases). Therefore, fix the expectation of os_info_old_init()
and enable it to be called from a standalone dump kernel.
Fixes: f4cac27dc0d6 ("s390/crash: Use old os_info to create PT_LOAD headers")
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
The return-address (RA) register r14 is specified as volatile in the
s390x ELF ABI [1]. Nevertheless proper CFI directives must be provided
for an unwinder to restore the return address, if the RA register
value is changed from its value at function entry, as it is the case.
[1]: s390x ELF ABI, https://github.com/IBM/s390x-abi/releases
Fixes: 4bff8cb54502 ("s390: convert to GENERIC_VDSO")
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
Nowadays, we call it "GUP-fast", the external interface includes functions
like "get_user_pages_fast()", and we renamed all internal functions to
reflect that as well.
Let's make the config option reflect that.
Link: https://lkml.kernel.org/r/20240402125516.223131-3-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Mike Rapoport (IBM) <rppt@kernel.org>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: John Hubbard <jhubbard@nvidia.com>
Cc: Peter Xu <peterx@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The vm_flags of vma already checked under per-VMA lock, if it is a bad
access, directly handle error, no need to retry with mmap_lock again.
Since the page faut is handled under per-VMA lock, count it as a vma lock
event with VMA_LOCK_SUCCESS.
Link: https://lkml.kernel.org/r/20240403083805.1818160-7-wangkefeng.wang@huawei.com
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: Alexander Gordeev <agordeev@linux.ibm.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Gerald Schaefer <gerald.schaefer@linux.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Future changes will need to add a new member to struct
vm_unmapped_area_info. This would cause trouble for any call site that
doesn't initialize the struct. Currently every caller sets each member
manually, so if new ones are added they will be uninitialized and the core
code parsing the struct will see garbage in the new member.
It could be possible to initialize the new member manually to 0 at each
call site. This and a couple other options were discussed. Having some
struct vm_unmapped_area_info instances not zero initialized will put those
sites at risk of feeding garbage into vm_unmapped_area(), if the
convention is to zero initialize the struct and any new field addition
missed a call site that initializes each field manually. So it is useful
to do things similar across the kernel.
The consensus (see links) was that in general the best way to accomplish
taking into account both code cleanliness and minimizing the chance of
introducing bugs, was to do C99 static initialization. As in: struct
vm_unmapped_area_info info = {};
With this method of initialization, the whole struct will be zero
initialized, and any statements setting fields to zero will be unneeded.
The change should not leave cleanup at the call sides.
While iterating though the possible solutions a few archs kindly acked
other variations that still zero initialized the struct. These sites have
been modified in previous changes using the pattern acked by the
respective arch.
So to be reduce the chance of bugs via uninitialized fields, perform a
tree wide change using the consensus for the best general way to do this
change. Use C99 static initializing to zero the struct and remove and
statements that simply set members to zero.
Link: https://lkml.kernel.org/r/20240326021656.202649-11-rick.p.edgecombe@intel.com
Link: https://lore.kernel.org/lkml/202402280912.33AEE7A9CF@keescook/#t
Link: https://lore.kernel.org/lkml/j7bfvig3gew3qruouxrh7z7ehjjafrgkbcmg6tcghhfh3rhmzi@wzlcoecgy5rs/
Link: https://lore.kernel.org/lkml/ec3e377a-c0a0-4dd3-9cb9-96517e54d17e@csgroup.eu/
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Aneesh Kumar K.V <aneesh.kumar@kernel.org>
Cc: Borislav Petkov (AMD) <bp@alien8.de>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Deepak Gupta <debug@rivosinc.com>
Cc: Guo Ren <guoren@kernel.org>
Cc: Helge Deller <deller@gmx.de>
Cc: H. Peter Anvin (Intel) <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Liam R. Howlett <Liam.Howlett@oracle.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Naveen N. Rao <naveen.n.rao@linux.ibm.com>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The mm_struct contains a function pointer *get_unmapped_area(), which is
set to either arch_get_unmapped_area() or arch_get_unmapped_area_topdown()
during the initialization of the mm.
Since the function pointer only ever points to two functions that are
named the same across all arch's, a function pointer is not really
required. In addition future changes will want to add versions of the
functions that take additional arguments. So to save a pointers worth of
bytes in mm_struct, and prevent adding additional function pointers to
mm_struct in future changes, remove it and keep the information about
which get_unmapped_area() to use in a flag.
Add the new flag to MMF_INIT_MASK so it doesn't get clobbered on fork by
mmf_init_flags(). Most MM flags get clobbered on fork. In the
pre-existing behavior mm->get_unmapped_area() would get copied to the new
mm in dup_mm(), so not clobbering the flag preserves the existing behavior
around inheriting the topdown-ness.
Introduce a helper, mm_get_unmapped_area(), to easily convert code that
refers to the old function pointer to instead select and call either
arch_get_unmapped_area() or arch_get_unmapped_area_topdown() based on the
flag. Then drop the mm->get_unmapped_area() function pointer. Leave the
get_unmapped_area() pointer in struct file_operations alone. The main
purpose of this change is to reorganize in preparation for future changes,
but it also converts the calls of mm->get_unmapped_area() from indirect
branches into a direct ones.
The stress-ng bigheap benchmark calls realloc a lot, which calls through
get_unmapped_area() in the kernel. On x86, the change yielded a ~1%
improvement there on a retpoline config.
In testing a few x86 configs, removing the pointer unfortunately didn't
result in any actual size reductions in the compiled layout of mm_struct.
But depending on compiler or arch alignment requirements, the change could
shrink the size of mm_struct.
Link: https://lkml.kernel.org/r/20240326021656.202649-3-rick.p.edgecombe@intel.com
Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Acked-by: Dave Hansen <dave.hansen@linux.intel.com>
Acked-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Reviewed-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Aneesh Kumar K.V <aneesh.kumar@kernel.org>
Cc: Borislav Petkov (AMD) <bp@alien8.de>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Deepak Gupta <debug@rivosinc.com>
Cc: Guo Ren <guoren@kernel.org>
Cc: Helge Deller <deller@gmx.de>
Cc: H. Peter Anvin (Intel) <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Mark Brown <broonie@kernel.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Naveen N. Rao <naveen.n.rao@linux.ibm.com>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
The comment in the code explains the reasons. We took a different
approach comparing to pmd_pfn() by providing a fallback function.
Another option is to provide some lower level config options (compare to
HUGETLB_PAGE or THP) to identify which layer an arch can support for such
huge mappings. However that can be an overkill.
[peterx@redhat.com: fix loongson defconfig]
Link: https://lkml.kernel.org/r/20240403013249.1418299-4-peterx@redhat.com
Link: https://lkml.kernel.org/r/20240327152332.950956-6-peterx@redhat.com
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Tested-by: Ryan Roberts <ryan.roberts@arm.com>
Cc: Mike Rapoport (IBM) <rppt@kernel.org>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Andrew Jones <andrew.jones@linux.dev>
Cc: Aneesh Kumar K.V (IBM) <aneesh.kumar@kernel.org>
Cc: Axel Rasmussen <axelrasmussen@google.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: David Hildenbrand <david@redhat.com>
Cc: James Houghton <jthoughton@google.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Kirill A. Shutemov <kirill@shutemov.name>
Cc: Lorenzo Stoakes <lstoakes@gmail.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Rik van Riel <riel@surriel.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
All implementations that aren't no-ops just set a bit in the flags, and we
want to use the folio flags rather than the page flags for that. Rename
it to arch_clear_hugetlb_flags() while we're touching it so nobody thinks
it's used for THP.
[willy@infradead.org: fix arm64 build]
Link: https://lkml.kernel.org/r/ZgQvNKGdlDkwhQEX@casper.infradead.org
Link: https://lkml.kernel.org/r/20240326171045.410737-8-willy@infradead.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Patch series "Memory allocation profiling", v6.
Overview:
Low overhead [1] per-callsite memory allocation profiling. Not just for
debug kernels, overhead low enough to be deployed in production.
Example output:
root@moria-kvm:~# sort -rn /proc/allocinfo
127664128 31168 mm/page_ext.c:270 func:alloc_page_ext
56373248 4737 mm/slub.c:2259 func:alloc_slab_page
14880768 3633 mm/readahead.c:247 func:page_cache_ra_unbounded
14417920 3520 mm/mm_init.c:2530 func:alloc_large_system_hash
13377536 234 block/blk-mq.c:3421 func:blk_mq_alloc_rqs
11718656 2861 mm/filemap.c:1919 func:__filemap_get_folio
9192960 2800 kernel/fork.c:307 func:alloc_thread_stack_node
4206592 4 net/netfilter/nf_conntrack_core.c:2567 func:nf_ct_alloc_hashtable
4136960 1010 drivers/staging/ctagmod/ctagmod.c:20 [ctagmod] func:ctagmod_start
3940352 962 mm/memory.c:4214 func:alloc_anon_folio
2894464 22613 fs/kernfs/dir.c:615 func:__kernfs_new_node
...
Usage:
kconfig options:
- CONFIG_MEM_ALLOC_PROFILING
- CONFIG_MEM_ALLOC_PROFILING_ENABLED_BY_DEFAULT
- CONFIG_MEM_ALLOC_PROFILING_DEBUG
adds warnings for allocations that weren't accounted because of a
missing annotation
sysctl:
/proc/sys/vm/mem_profiling
Runtime info:
/proc/allocinfo
Notes:
[1]: Overhead
To measure the overhead we are comparing the following configurations:
(1) Baseline with CONFIG_MEMCG_KMEM=n
(2) Disabled by default (CONFIG_MEM_ALLOC_PROFILING=y &&
CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=n)
(3) Enabled by default (CONFIG_MEM_ALLOC_PROFILING=y &&
CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=y)
(4) Enabled at runtime (CONFIG_MEM_ALLOC_PROFILING=y &&
CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=n && /proc/sys/vm/mem_profiling=1)
(5) Baseline with CONFIG_MEMCG_KMEM=y && allocating with __GFP_ACCOUNT
(6) Disabled by default (CONFIG_MEM_ALLOC_PROFILING=y &&
CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=n) && CONFIG_MEMCG_KMEM=y
(7) Enabled by default (CONFIG_MEM_ALLOC_PROFILING=y &&
CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=y) && CONFIG_MEMCG_KMEM=y
Performance overhead:
To evaluate performance we implemented an in-kernel test executing
multiple get_free_page/free_page and kmalloc/kfree calls with allocation
sizes growing from 8 to 240 bytes with CPU frequency set to max and CPU
affinity set to a specific CPU to minimize the noise. Below are results
from running the test on Ubuntu 22.04.2 LTS with 6.8.0-rc1 kernel on
56 core Intel Xeon:
kmalloc pgalloc
(1 baseline) 6.764s 16.902s
(2 default disabled) 6.793s (+0.43%) 17.007s (+0.62%)
(3 default enabled) 7.197s (+6.40%) 23.666s (+40.02%)
(4 runtime enabled) 7.405s (+9.48%) 23.901s (+41.41%)
(5 memcg) 13.388s (+97.94%) 48.460s (+186.71%)
(6 def disabled+memcg) 13.332s (+97.10%) 48.105s (+184.61%)
(7 def enabled+memcg) 13.446s (+98.78%) 54.963s (+225.18%)
Memory overhead:
Kernel size:
text data bss dec diff
(1) 26515311 18890222 17018880 62424413
(2) 26524728 19423818 16740352 62688898 264485
(3) 26524724 19423818 16740352 62688894 264481
(4) 26524728 19423818 16740352 62688898 264485
(5) 26541782 18964374 16957440 62463596 39183
Memory consumption on a 56 core Intel CPU with 125GB of memory:
Code tags: 192 kB
PageExts: 262144 kB (256MB)
SlabExts: 9876 kB (9.6MB)
PcpuExts: 512 kB (0.5MB)
Total overhead is 0.2% of total memory.
Benchmarks:
Hackbench tests run 100 times:
hackbench -s 512 -l 200 -g 15 -f 25 -P
baseline disabled profiling enabled profiling
avg 0.3543 0.3559 (+0.0016) 0.3566 (+0.0023)
stdev 0.0137 0.0188 0.0077
hackbench -l 10000
baseline disabled profiling enabled profiling
avg 6.4218 6.4306 (+0.0088) 6.5077 (+0.0859)
stdev 0.0933 0.0286 0.0489
stress-ng tests:
stress-ng --class memory --seq 4 -t 60
stress-ng --class cpu --seq 4 -t 60
Results posted at: https://evilpiepirate.org/~kent/memalloc_prof_v4_stress-ng/
[2] https://lore.kernel.org/all/20240306182440.2003814-1-surenb@google.com/
This patch (of 37):
The next patch drops vmalloc.h from a system header in order to fix a
circular dependency; this adds it to all the files that were pulling it in
implicitly.
[kent.overstreet@linux.dev: fix arch/alpha/lib/memcpy.c]
Link: https://lkml.kernel.org/r/20240327002152.3339937-1-kent.overstreet@linux.dev
[surenb@google.com: fix arch/x86/mm/numa_32.c]
Link: https://lkml.kernel.org/r/20240402180933.1663992-1-surenb@google.com
[kent.overstreet@linux.dev: a few places were depending on sizes.h]
Link: https://lkml.kernel.org/r/20240404034744.1664840-1-kent.overstreet@linux.dev
[arnd@arndb.de: fix mm/kasan/hw_tags.c]
Link: https://lkml.kernel.org/r/20240404124435.3121534-1-arnd@kernel.org
[surenb@google.com: fix arc build]
Link: https://lkml.kernel.org/r/20240405225115.431056-1-surenb@google.com
Link: https://lkml.kernel.org/r/20240321163705.3067592-1-surenb@google.com
Link: https://lkml.kernel.org/r/20240321163705.3067592-2-surenb@google.com
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Pasha Tatashin <pasha.tatashin@soleen.com>
Tested-by: Kees Cook <keescook@chromium.org>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Alex Gaynor <alex.gaynor@gmail.com>
Cc: Alice Ryhl <aliceryhl@google.com>
Cc: Andreas Hindborg <a.hindborg@samsung.com>
Cc: Benno Lossin <benno.lossin@proton.me>
Cc: "Björn Roy Baron" <bjorn3_gh@protonmail.com>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Dennis Zhou <dennis@kernel.org>
Cc: Gary Guo <gary@garyguo.net>
Cc: Miguel Ojeda <ojeda@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Wedson Almeida Filho <wedsonaf@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
This API is not used anymore, drop it for the whole tree.
Link: https://lkml.kernel.org/r/20240318200404.448346-13-peterx@redhat.com
Signed-off-by: Peter Xu <peterx@redhat.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Andreas Larsson <andreas@gaisler.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@kernel.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Bjorn Andersson <andersson@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Fabio Estevam <festevam@denx.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jason Gunthorpe <jgg@nvidia.com>
Cc: Konrad Dybcio <konrad.dybcio@linaro.org>
Cc: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Cc: Lucas Stach <l.stach@pengutronix.de>
Cc: Mark Salter <msalter@redhat.com>
Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Mike Rapoport (IBM) <rppt@kernel.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Naoya Horiguchi <nao.horiguchi@gmail.com>
Cc: "Naveen N. Rao" <naveen.n.rao@linux.ibm.com>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Shawn Guo <shawnguo@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Cross-merge networking fixes after downstream PR.
Conflicts:
drivers/net/ethernet/ti/icssg/icssg_prueth.c
net/mac80211/chan.c
89884459a0b9 ("wifi: mac80211: fix idle calculation with multi-link")
87f5500285fb ("wifi: mac80211: simplify ieee80211_assign_link_chanctx()")
https://lore.kernel.org/all/20240422105623.7b1fbda2@canb.auug.org.au/
net/unix/garbage.c
1971d13ffa84 ("af_unix: Suppress false-positive lockdep splat for spin_lock() in __unix_gc().")
4090fa373f0e ("af_unix: Replace garbage collection algorithm.")
drivers/net/ethernet/ti/icssg/icssg_prueth.c
drivers/net/ethernet/ti/icssg/icssg_common.c
4dcd0e83ea1d ("net: ti: icssg-prueth: Fix signedness bug in prueth_init_rx_chns()")
e2dc7bfd677f ("net: ti: icssg-prueth: Move common functions into a separate file")
No adjacent changes.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
Complete switching the __iowriteXX_copy() routines over to use #define and
arch provided inline/macro functions instead of weak symbols.
S390 has an implementation that simply calls another memcpy
function. Inline this so the callers don't have to do two jumps.
Link: https://lore.kernel.org/r/3-v3-1893cd8b9369+1925-mlx5_arm_wc_jgg@nvidia.com
Acked-by: Niklas Schnelle <schnelle@linux.ibm.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
It is trivial to implement an inline to do this, so provide it in the s390
headers. Like the 64 bit version it should just invoke zpci_memcpy_toio()
with the correct size.
Link: https://lore.kernel.org/r/2-v3-1893cd8b9369+1925-mlx5_arm_wc_jgg@nvidia.com
Acked-by: Niklas Schnelle <schnelle@linux.ibm.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
The function __storage_key_init_range() expects the end address to be
the first byte outside the range to be initialized. I.e. end - start
should be the size of the area to be initialized.
The current code works because __storage_key_init_range() will still loop
over every page in the range, but it is slower than using sske_frame().
Fixes: 3afdfca69870 ("s390/mm: Clear skeys for newly mapped huge guest pmds")
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Link: https://lore.kernel.org/r/20240416114220.28489-3-imbrenda@linux.ibm.com
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
The function __storage_key_init_range() expects the end address to be
the first byte outside the range to be initialized. I.e. end - start
should be the size of the area to be initialized.
The current code works because __storage_key_init_range() will still loop
over every page in the range, but it is slower than using sske_frame().
Fixes: 964c2c05c9f3 ("s390/mm: Clear huge page storage keys on enable_skey")
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Link: https://lore.kernel.org/r/20240416114220.28489-2-imbrenda@linux.ibm.com
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
commit fa41ba0d08de ("s390/mm: avoid empty zero pages for KVM guests to
avoid postcopy hangs") introduced an undesired side effect when combined
with memory ballooning and VM migration: memory part of the inflated
memory balloon will consume memory.
Assuming we have a 100GiB VM and inflated the balloon to 40GiB. Our VM
will consume ~60GiB of memory. If we now trigger a VM migration,
hypervisors like QEMU will read all VM memory. As s390x does not support
the shared zeropage, we'll end up allocating for all previously-inflated
memory part of the memory balloon: 50 GiB. So we might easily
(unexpectedly) crash the VM on the migration source.
Even worse, hypervisors like QEMU optimize for zeropage migration to not
consume memory on the migration destination: when migrating a
"page full of zeroes", on the migration destination they check whether the
target memory is already zero (by reading the destination memory) and avoid
writing to the memory to not allocate memory: however, s390x will also
allocate memory here, implying that also on the migration destination, we
will end up allocating all previously-inflated memory part of the memory
balloon.
This is especially bad if actual memory overcommit was not desired, when
memory ballooning is used for dynamic VM memory resizing, setting aside
some memory during boot that can be added later on demand. Alternatives
like virtio-mem that would avoid this issue are not yet available on
s390x.
There could be ways to optimize some cases in user space: before reading
memory in an anonymous private mapping on the migration source, check via
/proc/self/pagemap if anything is already populated. Similarly check on
the migration destination before reading. While that would avoid
populating tables full of shared zeropages on all architectures, it's
harder to get right and performant, and requires user space changes.
Further, with posctopy live migration we must place a page, so there,
"avoid touching memory to avoid allocating memory" is not really
possible. (Note that a previously we would have falsely inserted
shared zeropages into processes using UFFDIO_ZEROPAGE where
mm_forbids_zeropage() would have actually forbidden it)
PV is currently incompatible with memory ballooning, and in the common
case, KVM guests don't make use of storage keys. Instead of zapping
zeropages when enabling storage keys / PV, that turned out to be
problematic in the past, let's do exactly the same we do with KSM pages:
trigger unsharing faults to replace the shared zeropages by proper
anonymous folios.
What about added latency when enabling storage kes? Having a lot of
zeropages in applicable environments (PV, legacy guests, unittests) is
unexpected. Further, KSM could today already unshare the zeropages
and unmerging KSM pages when enabling storage kets would unshare the
KSM-placed zeropages in the same way, resulting in the same latency.
[ agordeev: Fixed sparse and checkpatch complaints and error handling ]
Reviewed-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Tested-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Fixes: fa41ba0d08de ("s390/mm: avoid empty zero pages for KVM guests to avoid postcopy hangs")
Signed-off-by: David Hildenbrand <david@redhat.com>
Link: https://lore.kernel.org/r/20240411161441.910170-3-david@redhat.com
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
The recently added check to figure out if a fault happened on gmap ASCE
dereferences the gmap pointer in lowcore without checking that it is not
NULL. For all non-KVM processes the pointer is NULL, so that some value
from lowcore will be read. With the current layouts of struct gmap and
struct lowcore the read value (aka ASCE) is zero, so that this doesn't lead
to any observable bug; at least currently.
Fix this by adding the missing NULL pointer check.
Fixes: 64c3431808bd ("s390/entry: compare gmap asce to determine guest/host fault")
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
Currently, kernel modules contain their own set of expoline thunks. In
the case of EXPOLINE_EXTERN, this involves postlinking of precompiled
expoline.o. expoline.o is also necessary for out-of-source tree module
builds.
Now that the kernel modules area is less than 4 GB away from
kernel expoline thunks, make modules use kernel expolines. Also make
EXPOLINE_EXTERN the default if the compiler supports it. This simplifies
build and aligns with the approach adopted by other architectures.
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
Fix offset calculation when branch target is more then 2Gb away.
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
The .vmlinux.relocs section is moved in front of the compressed
kernel. The interim section rescue step is avoided as result.
Suggested-by: Sumanth Korikkar <sumanthk@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
Rework deployment of kernel image for both compressed and
uncompressed variants as defined by CONFIG_KERNEL_UNCOMPRESSED
kernel configuration variable.
In case CONFIG_KERNEL_UNCOMPRESSED is disabled avoid uncompressing
the kernel to a temporary buffer and copying it to the target
address. Instead, uncompress it directly to the target destination.
In case CONFIG_KERNEL_UNCOMPRESSED is enabled avoid moving the
kernel to default 0x100000 location when KASLR is disabled or
failed. Instead, use the uncompressed kernel image directly.
In case KASLR is disabled or failed .amode31 section location in
memory is not randomized and precedes the kernel image. In case
CONFIG_KERNEL_UNCOMPRESSED is disabled that location overlaps the
area used by the decompression algorithm. That is fine, since that
area is not used after the decompression finished and the size of
.amode31 section is not expected to exceed BOOT_HEAP_SIZE ever.
There is no decompression in case CONFIG_KERNEL_UNCOMPRESSED is
enabled. Therefore, rename decompress_kernel() to deploy_kernel(),
which better describes both uncompressed and compressed cases.
Introduce AMODE31_SIZE macro to avoid immediate value of 0x3000
(the size of .amode31 section) in the decompressor linker script.
Modify the vmlinux linker script to force the size of .amode31
section to AMODE31_SIZE (the value of (_eamode31 - _samode31)
could otherwise differ as result of compiler options used).
Introduce __START_KERNEL macro that defines the kernel ELF image
entry point and set it to the currrent value of 0x100000.
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
Since kernel virtual and physical address spaces are
uncoupled the kernel is mapped at the top of the virtual
address space in case KASLR is disabled.
That does not pose any issue with regard to the kernel
booting and operation, but makes it difficult to use a
generated vmlinux with some debugging tools (e.g. gdb),
because the exact location of the kernel image in virtual
memory is unknown. Make that location known and introduce
CONFIG_KERNEL_IMAGE_BASE configuration option.
A custom CONFIG_KERNEL_IMAGE_BASE value that would break
the virtual memory layout leads to a build error.
The kernel image size is defined by KERNEL_IMAGE_SIZE
macro and set to 512 MB, by analogy with x86.
Suggested-by: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
The uncoupling physical vs virtual address spaces brings
the following benefits to s390:
- virtual memory layout flexibility;
- closes the address gap between kernel and modules, it
caused s390-only problems in the past (e.g. 'perf' bugs);
- allows getting rid of trampolines used for module calls
into kernel;
- allows simplifying BPF trampoline;
- minor performance improvement in branch prediction;
- kernel randomization entropy is magnitude bigger, as it is
derived from the amount of available virtual, not physical
memory;
The whole change could be described in two pictures below:
before and after the change.
Some aspects of the virtual memory layout setup are not
clarified (number of page levels, alignment, DMA memory),
since these are not a part of this change or secondary
with regard to how the uncoupling itself is implemented.
The focus of the pictures is to explain why __va() and __pa()
macros are implemented the way they are.
Memory layout in V==R mode:
| Physical | Virtual |
+- 0 --------------+- 0 --------------+ identity mapping start
| | S390_lowcore | Low-address memory
| +- 8 KB -----------+
| | |
| | identity | phys == virt
| | mapping | virt == phys
| | |
+- AMODE31_START --+- AMODE31_START --+ .amode31 rand. phys/virt start
|.amode31 text/data|.amode31 text/data|
+- AMODE31_END ----+- AMODE31_END ----+ .amode31 rand. phys/virt start
| | |
| | |
+- __kaslr_offset, __kaslr_offset_phys| kernel rand. phys/virt start
| | |
| kernel text/data | kernel text/data | phys == kvirt
| | |
+------------------+------------------+ kernel phys/virt end
| | |
| | |
| | |
| | |
+- ident_map_size -+- ident_map_size -+ identity mapping end
| |
| ... unused gap |
| |
+---- vmemmap -----+ 'struct page' array start
| |
| virtually mapped |
| memory map |
| |
+- __abs_lowcore --+
| |
| Absolute Lowcore |
| |
+- __memcpy_real_area
| |
| Real Memory Copy|
| |
+- VMALLOC_START --+ vmalloc area start
| |
| vmalloc area |
| |
+- MODULES_VADDR --+ modules area start
| |
| modules area |
| |
+------------------+ UltraVisor Secure Storage limit
| |
| ... unused gap |
| |
+KASAN_SHADOW_START+ KASAN shadow memory start
| |
| KASAN shadow |
| |
+------------------+ ASCE limit
Memory layout in V!=R mode:
| Physical | Virtual |
+- 0 --------------+- 0 --------------+
| | S390_lowcore | Low-address memory
| +- 8 KB -----------+
| | |
| | |
| | ... unused gap |
| | |
+- AMODE31_START --+- AMODE31_START --+ .amode31 rand. phys/virt start
|.amode31 text/data|.amode31 text/data|
+- AMODE31_END ----+- AMODE31_END ----+ .amode31 rand. phys/virt end (<2GB)
| | |
| | |
+- __kaslr_offset_phys | kernel rand. phys start
| | |
| kernel text/data | |
| | |
+------------------+ | kernel phys end
| | |
| | |
| | |
| | |
+- ident_map_size -+ |
| |
| ... unused gap |
| |
+- __identity_base + identity mapping start (>= 2GB)
| |
| identity | phys == virt - __identity_base
| mapping | virt == phys + __identity_base
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
+---- vmemmap -----+ 'struct page' array start
| |
| virtually mapped |
| memory map |
| |
+- __abs_lowcore --+
| |
| Absolute Lowcore |
| |
+- __memcpy_real_area
| |
| Real Memory Copy|
| |
+- VMALLOC_START --+ vmalloc area start
| |
| vmalloc area |
| |
+- MODULES_VADDR --+ modules area start
| |
| modules area |
| |
+- __kaslr_offset -+ kernel rand. virt start
| |
| kernel text/data | phys == (kvirt - __kaslr_offset) +
| | __kaslr_offset_phys
+- kernel .bss end + kernel rand. virt end
| |
| ... unused gap |
| |
+------------------+ UltraVisor Secure Storage limit
| |
| ... unused gap |
| |
+KASAN_SHADOW_START+ KASAN shadow memory start
| |
| KASAN shadow |
| |
+------------------+ ASCE limit
Unused gaps in the virtual memory layout could be present
or not - depending on how partucular system is configured.
No page tables are created for the unused gaps.
The relative order of vmalloc, modules and kernel image in
virtual memory is defined by following considerations:
- start of the modules area and end of the kernel should reside
within 4GB to accommodate relative 32-bit jumps. The best way
to achieve that is to place kernel next to modules;
- vmalloc and module areas should locate next to each other
to prevent failures and extra reworks in user level tools
(makedumpfile, crash, etc.) which treat vmalloc and module
addresses similarily;
- kernel needs to be the last area in the virtual memory
layout to easily distinguish between kernel and non-kernel
virtual addresses. That is needed to (again) simplify
handling of addresses in user level tools and make __pa()
macro faster (see below);
Concluding the above, the relative order of the considered
virtual areas in memory is: vmalloc - modules - kernel.
Therefore, the only change to the current memory layout is
moving kernel to the end of virtual address space.
With that approach the implementation of __pa() macro is
straightforward - all linear virtual addresses less than
kernel base are considered identity mapping:
phys == virt - __identity_base
All addresses greater than kernel base are kernel ones:
phys == (kvirt - __kaslr_offset) + __kaslr_offset_phys
By contrast, __va() macro deals only with identity mapping
addresses:
virt == phys + __identity_base
.amode31 section is mapped separately and is not covered by
__pa() macro. In fact, it could have been handled easily by
checking whether a virtual address is within the section or
not, but there is no need for that. Thus, let __pa() code
do as little machine cycles as possible.
The KASAN shadow memory is located at the very end of the
virtual memory layout, at addresses higher than the kernel.
However, that is not a linear mapping and no code other than
KASAN instrumentation or API is expected to access it.
When KASLR mode is enabled the kernel base address randomized
within a memory window that spans whole unused virtual address
space. The size of that window depends from the amount of
physical memory available to the system, the limit imposed by
UltraVisor (if present) and the vmalloc area size as provided
by vmalloc= kernel command line parameter.
In case the virtual memory is exhausted the minimum size of
the randomization window is forcefully set to 2GB, which
amounts to in 15 bits of entropy if KASAN is enabled or 17
bits of entropy in default configuration.
The default kernel offset 0x100000 is used as a magic value
both in the decompressor code and vmlinux linker script, but
it will be removed with a follow-up change.
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
This is a preparatory rework to allow uncoupling virtual
and physical addresses spaces.
The vmcore ELF program headers describe virtual memory
regions of a crashed kernel. User level tools use that
information for the kernel text and data analysis (e.g
vmcore-dmesg extracts the kernel log).
Currently the kernel image is covered by program headers
describing the identity mapping regions. But in the future
the kernel image will be mapped into separate region outside
of the identity mapping. Create the additional ELF program
header that covers kernel image only, so that vmcore tools
could locate kernel text and data.
Further, the identity mapping in crashed and capture kernels
will have different base address. Due to that __va() macro
can not be used in the capture kernel. Instead, read crashed
kernel identity mapping base address from os_info and use
it for PT_LOAD type program headers creation.
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
This is a preparatory rework to allow uncoupling virtual
and physical addresses spaces.
The virtual memory layout is needed for address translation
by crash tool when /proc/kcore device is used as the memory
image.
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
This is a preparatory rework to allow uncoupling virtual
and physical addresses spaces.
The virtual memory layout will be read out by makedumpfile,
crash and other user tools for virtual address translation.
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
Introduce entries that do not reference any data in memory,
but rather provide values. Set the size of such entries to
zero and do not compute checksum for them, since there is no
data which integrity needs to be checked. The integrity of
the value entries itself is still covered by the os_info
checksum.
Reserve the lowest unused entry index OS_INFO_RESERVED for
future use - presumably for the number of entries present.
That could later be used by user level tools. The existing
tools would not notice any difference.
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
This is a preparatory rework to allow uncoupling virtual
and physical addresses spaces.
Introduce .amode31 section address range AMODE31_START
and AMODE31_END macros for later use.
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
This is a preparatory rework to allow uncoupling virtual
and physical addresses spaces.
Currently the identity mapping base address is implicit
and is always set to zero. Make it explicit by putting
into __identity_base persistent boot variable and use it
in proper context - which is the value of PAGE_OFFSET.
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
This is a preparatory rework to allow uncoupling virtual
and physical addresses spaces.
Currently __kaslr_offset is the kernel offset in both
physical memory on boot and in virtual memory after DAT
mode is enabled.
Uncouple these offsets and rename the physical address
space variant to __kaslr_offset_phys while keep the name
__kaslr_offset for the offset in virtual address space.
Do not use __kaslr_offset_phys after DAT mode is enabled
just yet, but still make it a persistent boot variable
for later use.
Use __kaslr_offset and __kaslr_offset_phys offsets in
proper contexts and alter handle_relocs() function to
distinguish between the two.
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
This is a preparatory rework to allow uncoupling virtual
and physical addresses spaces.
Put virtual memory layout information into a structure
to improve code generation when accessing the structure
members, which are currently only ident_map_size and
__kaslr_offset.
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
Move everyting KASLR related to <asm/page.h>,
similarly to many other architectures.
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Suggested-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
This is a preparatory rework to allow uncoupling virtual
and physical addresses spaces.
Currently the order of virtual memory areas is (the lowcore
and .amode31 section are skipped, as it is irrelevant):
identity mapping (the kernel is contained within)
vmemmap
vmalloc
modules
Absolute Lowcore
Real Memory Copy
In the future the kernel will be mapped separately and placed
to the end of the virtual address space, so the layout would
turn like this:
identity mapping
vmemmap
vmalloc
modules
Absolute Lowcore
Real Memory Copy
kernel
However, the distance between kernel and modules needs to be as
little as possible, ideally - none. Thus, the Absolute Lowcore
and Real Memory Copy areas would stay in the way and therefore
need to be moved as well:
identity mapping
vmemmap
Absolute Lowcore
Real Memory Copy
vmalloc
modules
kernel
To facilitate such layout swap the vmalloc and Absolute Lowcore
together with Real Memory Copy areas. As result, the current
layout turns into:
identity mapping (the kernel is contained within)
vmemmap
Absolute Lowcore
Real Memory Copy
vmalloc
modules
This will allow to locate the kernel directly next to the
modules once it gets mapped separately.
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
In case vmemmap array could overlap with vmalloc area on
virtual memory layout setup, the size of vmalloc area
is decreased. That could result in less memory than user
requested with vmalloc= kernel command line parameter.
Instead, reduce the size of identity mapping (and the
size of vmemmap array as result) to avoid such overlap.
Further, currently the virtual memmory allocation "rolls"
from top to bottom and it is only VMALLOC_START that could
get increased due to the overlap. Change that to decrease-
only, which makes the whole allocation algorithm more easy
to comprehend.
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
The maximum mappable physical address (as returned by
arch_get_mappable_range() callback) is limited by the
value of (1UL << MAX_PHYSMEM_BITS).
The maximum physical address available to a DCSS segment
is 512GB.
In case the available online or offline memory size is less
than the DCSS limit arch_get_mappable_range() would include
never used [512GB..(1UL << MAX_PHYSMEM_BITS)] range.
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
vmemmap is forcefully set to start at MAX_PHYSMEM_BITS at most.
That could be needed in the past to limit ident_map_size to
MAX_PHYSMEM_BITS. However since commit 75eba6ec0de1 ("s390:
unify identity mapping limits handling") ident_map_size is
limited in setup_ident_map_size() function, which is called
earlier.
Another reason to limit vmemmap start to MAX_PHYSMEM_BITS is
because it was returned by arch_get_mappable_range() as the
maximum mappable physical address. Since commit f641679dfe55
("s390/mm: rework arch_get_mappable_range() callback") that
is not required anymore.
As result, there is no neccessity to limit vmemmap starting
address with MAX_PHYSMEM_BITS.
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
In order for SIE to interpretively execute STFLE, it requires the real
or absolute address of a facility-list control block.
Before writing the location into the shadow SIE control block, convert
it from a virtual address.
We currently do not run into this bug because the lower 31 bits are the
same for virtual and physical addresses.
Signed-off-by: Nina Schoetterl-Glausch <nsg@linux.ibm.com>
Link: https://lore.kernel.org/r/20240319164420.4053380-3-nsg@linux.ibm.com
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Message-Id: <20240319164420.4053380-3-nsg@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
|
|
update_timer_sys() and update_timer_mcck() are inlines used for
CPU time accounting from the interrupt and machine-check handlers.
These routines are specific to s390 architecture, but included
via <linux/vtime.h> header implicitly. Avoid the extra loop and
include <asm/vtime.h> header directly.
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Link: https://lore.kernel.org/r/3fb696637c0eb7e9d6ffd6cbf9e647d7c5986b3d.1712760275.git.agordeev@linux.ibm.com
|
|
__ARCH_HAS_VTIME_TASK_SWITCH macro is not used anymore.
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Acked-by: Nicholas Piggin <npiggin@gmail.com>
Link: https://lore.kernel.org/r/b1055852eab0ffea33ad16c92d6a825c83037c3e.1712760275.git.agordeev@linux.ibm.com
|
Alexander Gordeev
Masahiro Yamada
Jean-Philippe Brucker
Harald Freudenberger
Nina Schoetterl-Glausch
Sumanth Korikkar
Sven Schnelle
Jean Delvare
Alexander Egorenkov
Jens Remus
David Hildenbrand
Kefeng Wang
Rick Edgecombe
Peter Xu
Matthew Wilcox (Oracle)
Kent Overstreet
Jakub Kicinski
Jason Gunthorpe
Claudio Imbrenda
Vasily Gorbik