- Add support for stackleak feature. Also allow specifying
architecture-specific stackleak poison function to enable faster
implementation. On s390, the mvc-based implementation helps decrease
typical overhead from a factor of 3 to just 25%
- Convert all assembler files to use SYM* style macros, deprecating the
ENTRY() macro and other annotations. Select ARCH_USE_SYM_ANNOTATIONS
- Improve KASLR to also randomize module and special amode31 code
base load addresses
- Rework decompressor memory tracking to support memory holes and improve
error handling
- Add support for protected virtualization AP binding
- Add support for set_direct_map() calls
- Implement set_memory_rox() and noexec module_alloc()
- Remove obsolete overriding of mem*() functions for KASAN
- Rework kexec/kdump to avoid using nodat_stack to call purgatory
- Convert the rest of the s390 code to use flexible-array member instead
of a zero-length array
- Clean up uaccess inline asm
- Enable ARCH_HAS_MEMBARRIER_SYNC_CORE
- Convert to using CONFIG_FUNCTION_ALIGNMENT and enable
DEBUG_FORCE_FUNCTION_ALIGN_64B
- Resolve last_break in userspace fault reports
- Simplify one-level sysctl registration
- Clean up branch prediction handling
- Rework CPU counter facility to retrieve available counter sets just
once
- Other various small fixes and improvements all over the code
-----BEGIN PGP SIGNATURE-----
iQEzBAABCAAdFiEE3QHqV+H2a8xAv27vjYWKoQLXFBgFAmRM8pwACgkQjYWKoQLX
FBjV1AgAlvAhu1XkwOdwqdT4GqE8pcN4XXzydog1MYihrSO2PdgWAxpEW7o2QURN
W+3xa6RIqt7nX2YBiwTanMZ12TYaFY7noGl3eUpD/NhueprweVirVl7VZUEuRoW/
j0mbx77xsVzLfuDFxkpVwE6/j+tTO78kLyjUHwcN9rFVUaL7/orJneDJf+V8fZG0
sHLOv0aljF7Jr2IIkw82lCmW/vdk7k0dACWMXK2kj1H3dIK34B9X4AdKDDf/WKXk
/OSElBeZ93tSGEfNDRIda6iR52xocROaRnQAaDtargKFl9VO0/dN9ADxO+SLNHjN
pFE/9VD6xT/xo4IuZZh/Z3TcYfiLvA==
=Geqx
-----END PGP SIGNATURE-----
Merge tag 's390-6.4-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Pull s390 updates from Vasily Gorbik:
- Add support for stackleak feature. Also allow specifying
architecture-specific stackleak poison function to enable faster
implementation. On s390, the mvc-based implementation helps decrease
typical overhead from a factor of 3 to just 25%
- Convert all assembler files to use SYM* style macros, deprecating the
ENTRY() macro and other annotations. Select ARCH_USE_SYM_ANNOTATIONS
- Improve KASLR to also randomize module and special amode31 code base
load addresses
- Rework decompressor memory tracking to support memory holes and
improve error handling
- Add support for protected virtualization AP binding
- Add support for set_direct_map() calls
- Implement set_memory_rox() and noexec module_alloc()
- Remove obsolete overriding of mem*() functions for KASAN
- Rework kexec/kdump to avoid using nodat_stack to call purgatory
- Convert the rest of the s390 code to use flexible-array member
instead of a zero-length array
- Clean up uaccess inline asm
- Enable ARCH_HAS_MEMBARRIER_SYNC_CORE
- Convert to using CONFIG_FUNCTION_ALIGNMENT and enable
DEBUG_FORCE_FUNCTION_ALIGN_64B
- Resolve last_break in userspace fault reports
- Simplify one-level sysctl registration
- Clean up branch prediction handling
- Rework CPU counter facility to retrieve available counter sets just
once
- Other various small fixes and improvements all over the code
* tag 's390-6.4-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (118 commits)
s390/stackleak: provide fast __stackleak_poison() implementation
stackleak: allow to specify arch specific stackleak poison function
s390: select ARCH_USE_SYM_ANNOTATIONS
s390/mm: use VM_FLUSH_RESET_PERMS in module_alloc()
s390: wire up memfd_secret system call
s390/mm: enable ARCH_HAS_SET_DIRECT_MAP
s390/mm: use BIT macro to generate SET_MEMORY bit masks
s390/relocate_kernel: adjust indentation
s390/relocate_kernel: use SYM* macros instead of ENTRY(), etc.
s390/entry: use SYM* macros instead of ENTRY(), etc.
s390/purgatory: use SYM* macros instead of ENTRY(), etc.
s390/kprobes: use SYM* macros instead of ENTRY(), etc.
s390/reipl: use SYM* macros instead of ENTRY(), etc.
s390/head64: use SYM* macros instead of ENTRY(), etc.
s390/earlypgm: use SYM* macros instead of ENTRY(), etc.
s390/mcount: use SYM* macros instead of ENTRY(), etc.
s390/crc32le: use SYM* macros instead of ENTRY(), etc.
s390/crc32be: use SYM* macros instead of ENTRY(), etc.
s390/crypto,chacha: use SYM* macros instead of ENTRY(), etc.
s390/amode31: use SYM* macros instead of ENTRY(), etc.
...
- Remove diagnostics and adjust config for CSD lock diagnostics
- Add a generic IPI-sending tracepoint, as currently there's no easy
way to instrument IPI origins: it's arch dependent and for some
major architectures it's not even consistently available.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----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=Wp7f
-----END PGP SIGNATURE-----
Merge tag 'smp-core-2023-04-27' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull SMP cross-CPU function-call updates from Ingo Molnar:
- Remove diagnostics and adjust config for CSD lock diagnostics
- Add a generic IPI-sending tracepoint, as currently there's no easy
way to instrument IPI origins: it's arch dependent and for some major
architectures it's not even consistently available.
* tag 'smp-core-2023-04-27' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
trace,smp: Trace all smp_function_call*() invocations
trace: Add trace_ipi_send_cpu()
sched, smp: Trace smp callback causing an IPI
smp: reword smp call IPI comment
treewide: Trace IPIs sent via smp_send_reschedule()
irq_work: Trace self-IPIs sent via arch_irq_work_raise()
smp: Trace IPIs sent via arch_send_call_function_ipi_mask()
sched, smp: Trace IPIs sent via send_call_function_single_ipi()
trace: Add trace_ipi_send_cpumask()
kernel/smp: Make csdlock_debug= resettable
locking/csd_lock: Remove per-CPU data indirection from CSD lock debugging
locking/csd_lock: Remove added data from CSD lock debugging
locking/csd_lock: Add Kconfig option for csd_debug default
Allocate early async stack like other early stacks and get rid of
arch_early_irq_init(). This way the async stack is allocated earlier,
and handled like all other stacks.
Reviewed-by: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
s390 is the only architecture which switches from the initial stack to a
later on allocated different stack for the first process.
This is (at least) problematic for the stackleak feature, which instruments
functions to save the current stackpointer within the task structure of the
running process.
The stackleak code compares stack pointers of the current process - and
doesn't expect that the kernel stack of a task can change. Even though the
stackleak feature itself will not cause any harm, the assumption about
kernel stacks being consistent is there, and only s390 doesn't follow that.
Therefore switch back to use init_thread_union, just like all other
architectures.
Reviewed-by: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Make STACK_INIT_OFFSET also available for assembler code, and
use it everywhere instead of open-coding it at several places.
Reviewed-by: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
To be able to trace invocations of smp_send_reschedule(), rename the
arch-specific definitions of it to arch_smp_send_reschedule() and wrap it
into an smp_send_reschedule() that contains a tracepoint.
Changes to include the declaration of the tracepoint were driven by the
following coccinelle script:
@func_use@
@@
smp_send_reschedule(...);
@include@
@@
#include <trace/events/ipi.h>
@no_include depends on func_use && !include@
@@
#include <...>
+
+ #include <trace/events/ipi.h>
[csky bits]
[riscv bits]
Signed-off-by: Valentin Schneider <vschneid@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Guo Ren <guoren@kernel.org>
Acked-by: Palmer Dabbelt <palmer@rivosinc.com>
Link: https://lore.kernel.org/r/20230307143558.294354-6-vschneid@redhat.com
Direct access to the struct bus_type dev_root pointer is going away soon
so replace that with a call to bus_get_dev_root() instead, which is what
it is there for.
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Alexander Gordeev <agordeev@linux.ibm.com>
Cc: Christian Borntraeger <borntraeger@linux.ibm.com>
Cc: Sven Schnelle <svens@linux.ibm.com>
Cc: Ilya Leoshkevich <iii@linux.ibm.com>
Cc: linux-s390@vger.kernel.org
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Link: https://lore.kernel.org/r/20230313182918.1312597-19-gregkh@linuxfoundation.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There is no point in changing branch prediction state of a cpu shortly
before it enters stop state. Therefore remove __bpon().
Acked-by: Alexander Gordeev <agordeev@linux.ibm.com>
Reviewed-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Clear CPU state (e.g. all TLB entries, prefetched instructions, etc.)
of the target CPU, however without clearing register contents before
starting any work on it.
This puts the target CPU in a more defined state compared to the
current Stop + Restart sigp orders.
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
If a machine check interrupt hits while user process is
running __s390_handle_mcck() helper function is called
directly from the interrupt handler and terminates the
current process by calling make_task_dead() routine.
The make_task_dead() is not allowed to be called from
interrupt context which forces the machine check handler
switch to the kernel stack and enable local interrupts
first.
The __s390_handle_mcck() could also be called to service
pending work, but this time from the external interrupts
handler. It is the machine check handler that establishes
the work and schedules the external interrupt, therefore
the machine check interrupt itself should be disabled
while reading out the corresponding variable:
local_mcck_disable();
mcck = *this_cpu_ptr(&cpu_mcck);
memset(this_cpu_ptr(&cpu_mcck), 0, sizeof(mcck));
local_mcck_enable();
However, local_mcck_disable() does not have effect when
__s390_handle_mcck() is called directly form the machine
check handler, since the machine check interrupt is still
disabled. Therefore, it is not the opening bracket to the
following local_mcck_enable() function.
Simplify the user process termination flow by scheduling
the external interrupt and killing the affected process
from the interrupt context.
Assume a kernel-generated signal is always delivered and
ignore a value returned by do_send_sig_info() funciton.
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Reviewed-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Move Absolute Lowcore Area allocation to the decompressor.
As result, get_abs_lowcore() and put_abs_lowcore() access
brackets become really straight and do not require complex
execution context analysis and LAP and interrupts tackling.
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
The setup of the kernel virtual address space is spread
throughout the sources, boot stages and config options
like this:
1. The available physical memory regions are queried
and stored as mem_detect information for later use
in the decompressor.
2. Based on the physical memory availability the virtual
memory layout is established in the decompressor;
3. If CONFIG_KASAN is disabled the kernel paging setup
code populates kernel pgtables and turns DAT mode on.
It uses the information stored at step [1].
4. If CONFIG_KASAN is enabled the kernel early boot
kasan setup populates kernel pgtables and turns DAT
mode on. It uses the information stored at step [1].
The kasan setup creates early_pg_dir directory and
directly overwrites swapper_pg_dir entries to make
shadow memory pages available.
Move the kernel virtual memory setup to the decompressor
and start the kernel with DAT turned on right from the
very first istruction. That completely eliminates the
boot phase when the kernel runs in DAT-off mode, simplies
the overall design and consolidates pgtables setup.
The identity mapping is created in the decompressor, while
kasan shadow mappings are still created by the early boot
kernel code.
Share with decompressor the existing kasan memory allocator.
It decreases the size of a newly requested memory block from
pgalloc_pos and ensures that kernel image is not overwritten.
pgalloc_low and pgalloc_pos pointers are made preserved boot
variables for that.
Use the bootdata infrastructure to setup swapper_pg_dir
and invalid_pg_dir directories used by the kernel later.
The interim early_pg_dir directory established by the
kasan initialization code gets eliminated as result.
As the kernel runs in DAT-on mode only the PSW_KERNEL_BITS
define gets PSW_MASK_DAT bit by default. Additionally, the
setup_lowcore_dat_off() and setup_lowcore_dat_on() routines
get merged, since there is no DAT-off mode stage anymore.
The memory mappings are created with RW+X protection that
allows the early boot code setting up all necessary data
and services for the kernel being booted. Just before the
paging is enabled the memory protection is changed to
RO+X for text, RO+NX for read-only data and RW+NX for
kernel data and the identity mapping.
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Uninline copy_oldmem_kernel() function and make it consistent
with a very similar memcpy_real() implementation, by moving
to code to crash_dump.c, where it actually belongs.
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Function smp_save_dump_cpus() collects CPU state of a crashed
system for secondary CPUs and for the IPL CPU very differently.
The Signal Processor stop-and-store-status orders are used for
the former while Hardware System Area requests and memcpy_real()
routine are called for the latter. In addition a system reset is
triggered, which pins smp_save_dump_cpus() function call before
CPU and device initialization.
Move the collection of IPL CPU state to a later stage when DAT
becomes available. That is needed to allow a follow-up rework of
memcpy_real() routine.
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Temporary unsetting of the prefix page in memcpy_absolute() routine
poses a risk of executing code path with unexpectedly disabled prefix
page. This rework avoids the prefix page uninstalling and disabling
of normal and machine check interrupts when accessing the absolute
zero memory.
Although memcpy_absolute() routine can access the whole memory, it is
only used to update the absolute zero lowcore. This rework therefore
introduces a new mechanism for the absolute zero lowcore access and
scraps memcpy_absolute() routine for good.
Instead, an area is reserved in the virtual memory that is used for
the absolute lowcore access only. That area holds an array of 8KB
virtual mappings - one per CPU. Whenever a CPU is brought online, the
corresponding item is mapped to the real address of the previously
installed prefix page.
The absolute zero lowcore access works like this: a CPU calls the
new primitive get_abs_lowcore() to obtain its 8KB mapping as a
pointer to the struct lowcore. Virtual address references to that
pointer get translated to the real addresses of the prefix page,
which in turn gets swapped with the absolute zero memory addresses
due to prefixing. Once the pointer is not needed it must be released
with put_abs_lowcore() primitive:
struct lowcore *abs_lc;
unsigned long flags;
abs_lc = get_abs_lowcore(&flags);
abs_lc->... = ...;
put_abs_lowcore(abs_lc, flags);
To ensure the described mechanism works large segment- and region-
table entries must be avoided for the 8KB mappings. Failure to do
so results in usage of Region-Frame Absolute Address (RFAA) or
Segment-Frame Absolute Address (SFAA) large page fields. In that
case absolute addresses would be used to address the prefix page
instead of the real ones and the prefixing would get bypassed.
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Currently smp_reinit_ipl_cpu() is a pre-SMP early initcall.
That ensures no CPU is running in parallel, but still not
enough to assume the code is exclusive, since the scheduling
is already available.
Move the function call to arch_call_rest_init() callback
to ensure no thread could be preempted and allow lockless
allocation of the kernel page tables. That is needed to
allow a follow-up rework of the absolute lowcore access
mechanism.
Suggested-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Signal processor SIGP_SET_PREFIX command expects physical
address of the lowcore to be installed, but instead the
virtual address is provided.
Note: this does not fix a bug currently, since virtual and
physical addresses are identical.
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Macro mem_assign_absolute() is able to access the whole memory, but
is only used and makes sense when updating the absolute lowcore.
Instead, introduce get_abs_lowcore() and put_abs_lowcore() macros
that limit access to absolute lowcore addresses only.
Suggested-by: Heiko Carstens <hca@linux.ibm.com>
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Get rid of duplicate code and redundant data.
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Macro mem_assign_absolute() is used to initialize a target
CPU lowcore callback parameters. But despite the macro name
it writes to the absolute lowcore only if the target CPU is
offline. In case the CPU is online the macro does implicitly
write to the normal memory.
That behaviour is correct, but extremely subtle. Sacrifice
few program bits in favour of clarity and distinguish between
online vs offline CPUs and normal vs absolute lowcore pointer.
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Machine generations up to z9 (released in May 2006) have been officially
out of service for several years now (z9 end of service - January 31, 2019).
No distributions build kernels supporting those old machine generations
anymore, except Debian, which seems to pick the oldest supported
generation. The team supporting Debian on s390 has been notified about
the change.
Raising minimum supported machine generation to z10 helps to reduce
maintenance cost and effectively remove code, which is not getting
enough testing coverage due to lack of older hardware and distributions
support. Besides that this unblocks some optimization opportunities and
allows to use wider instruction set in asm files for future features
implementation. Due to this change spectre mitigation and usercopy
implementations could be drastically simplified and many newer instructions
could be converted from ".insn" encoding to instruction names.
Acked-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
With commit 5789284710 ("s390/smp: reallocate IPL CPU lowcore")
virtual addresses are wrongly passed to memblock_free_late() and
SPX instructions on IPL CPU reinitialization.
Note: this does not fix a bug currently, since virtual and
physical addresses are identical.
Fixes: 5789284710 ("s390/smp: reallocate IPL CPU lowcore")
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
There is a confusion with regard to the source address of
memcpy_real() and calling functions. While the declared
type for a source assumes a virtual address, in fact it
always called with physical address of the source.
This confusion led to bugs in copy_oldmem_kernel() and
copy_oldmem_user() functions, where __pa() macro applied
mistakenly to physical addresses. It does not lead to a
real issue, since virtual and physical addresses are
currently the same.
Fix both the bugs and memcpy_real() prototype by making
type of source address consistent to the function name
and the way it actually used.
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Remove my old invalid email address which can be found in a couple of
files. Instead of updating it, just remove my contact data completely
from source files.
We have git and other tools which allow to figure out who is responsible
for what with recent contact data.
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Signal processor STORE STATUS requires a physical address where register
contents are supposed to be written to, however the kernel must read the
data via the corresponding virtual address.
Also the allocated save_area, where register contents are copied to,
resides in virtual address space.
Fix this by using proper __pa() conversion, or correct memblock_alloc()
invocation.
Reviewed-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Updating of the pointer to machine check extended save area
on the IPL CPU needs the lowcore protection to be disabled.
Disable interrupts while the protection is off to avoid
unnoticed writes to the lowcore.
Suggested-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
memblock_phys_free() is used on a virtual address. Fix this by using
memblock_free().
Note: this doesn't fix a bug currently, since virtual and physical
addresses are identical.
Reviewed-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Add missing __pa/__va address conversion of machine check extended
save area designation, which is an absolute address.
Note: this currently doesn't fix a real bug, since virtual addresses
are indentical to physical ones.
Reported-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Tested-by: Vineeth Vijayan <vneethv@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Since memblock_free() operates on a physical range, make its name
reflect it and rename it to memblock_phys_free(), so it will be a
logical counterpart to memblock_phys_alloc().
The callers are updated with the below semantic patch:
@@
expression addr;
expression size;
@@
- memblock_free(addr, size);
+ memblock_phys_free(addr, size);
Link: https://lkml.kernel.org/r/20210930185031.18648-6-rppt@kernel.org
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Juergen Gross <jgross@suse.com>
Cc: Shahab Vahedi <Shahab.Vahedi@synopsys.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
memblock_free_early_nid() is unused and memblock_free_early() is an
alias for memblock_free().
Replace calls to memblock_free_early() with calls to memblock_free() and
remove memblock_free_early() and memblock_free_early_nid().
Link: https://lkml.kernel.org/r/20210930185031.18648-4-rppt@kernel.org
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Juergen Gross <jgross@suse.com>
Cc: Shahab Vahedi <Shahab.Vahedi@synopsys.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The cpu hotplug notifiers are called without updating the core/thread
masks when a new CPU is added. This causes problems with code setting
up data structures in a cpu hotplug notifier, and relying on that later
in normal code.
This caused a crash in the new core scheduling code (SCHED_CORE),
where rq->core was set up in a notifier depending on cpu masks.
To fix this, add a cpu_setup_mask which is used in update_cpu_masks()
instead of the cpu_online_mask to determine whether the cpu masks should
be set for a certain cpu. Also move update_cpu_masks() to update the
masks before calling notify_cpu_starting() so that the notifiers are
seeing the updated masks.
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Cc: <stable@vger.kernel.org>
[hca@linux.ibm.com: get rid of cpu_online_mask handling]
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
The secondary CPU start C routine uses nodat_stack as a
interim stack before finally switching to kernel_stack.
Such scheme is superfluous, since the assembler restart
interrupt handler (that secondary CPU starter is called
from) does not need to use any stack for switching into
DAT mode. Once DAT is on, any stack including virtually-
mapped one could be used.
Avoid the use of nodat_stack and smp_start_secondary()
helper. Instead, initiate kernel_stack directly from
the restart interrupt handler.
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
The restart interrupt is triggered whenever a secondary CPU is
brought online, a remote function call dispatched from another
CPU or a manual PSW restart is initiated and causes the system
to kdump. The handling routine is always called with DAT turned
off. It then initializes the stack frame and invokes a callback.
The existing callbacks handle DAT as follows:
* __do_restart() and __machine_kexec() turn in on upon entry;
* __ipl_run(), __reipl_run() and __dump_run() do not turn it
right away, but all of them call diag308() - which turns DAT
on, but only if kasan is enabled;
In addition to the described complexity all callbacks (and the
functions they call) should avoid kasan instrumentation while
DAT is off.
This update enables DAT in the assembler restart handler and
relieves any callbacks (which are mostly C functions) from
dealing with DAT altogether.
There are four types of CPU restart that initialize control
registers in different ways:
1. Start of secondary CPU on boot - control registers are
inherited from the IPL CPU;
2. Restart of online CPU - control registers of the CPU being
restarted are kept;
3. Hotplug of offline CPU - control registers are inherited
from the starting CPU;
4. Start of offline CPU triggered by manual PSW restart -
the control registers are read from the absolute lowcore
and contain the boot time IPL CPU values updated with all
follow-up calls of smp_ctl_set_bit() and smp_ctl_clear_bit()
routines;
In first three cases contents of the control registers is the
most recent. In the latter case control registers are good
enough to facilitate successful completion of kdump operation.
Suggested-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
The dma section name is confusing, since the code which resides within
that section has nothing to do with direct memory access. Instead the
limitation is that the code has to run in 31 bit addressing mode, and
therefore has to reside below 2GB. So the name was chosen since
ZONE_DMA is the same region.
To reduce confusion rename the section to amode31, which hopefully
describes better what this is about.
Note: this will also change vmcoreinfo strings
- SDMA=... gets renamed to SAMODE31=...
- EDMA=... gets renamed to EAMODE31=...
Acked-by: Vasily Gorbik <gor@linux.ibm.com>
Reviewed-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
The functions get_online_cpus() and put_online_cpus() have been
deprecated during the CPU hotplug rework. They map directly to
cpus_read_lock() and cpus_read_unlock().
Replace deprecated CPU-hotplug functions with the official version.
The behavior remains unchanged.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://lore.kernel.org/r/20210803141621.780504-5-bigeasy@linutronix.de
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
The new boot data struct shall replace global variables OLDMEM_BASE and
OLDMEM_SIZE. It is initialized in the decompressor and passed
to the decompressed kernel. In comparison to the old solution, this one
doesn't access data at fixed physical addresses which will become important
when the decompressor becomes relocatable.
Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
S390's init_idle_preempt_count(p, cpu) doesn't actually let us initialize the
preempt_count of the requested CPU's idle task: it unconditionally writes
to the current CPU's. This clearly conflicts with idle_threads_init(),
which intends to initialize *all* the idle tasks, including their
preempt_count (or their CPU's, if the arch uses a per-CPU preempt_count).
Unfortunately, it seems the way s390 does things doesn't let us initialize
every possible CPU's preempt_count early on, as the pages where this
resides are only allocated when a CPU is brought up and are freed when it
is brought down.
Let the arch-specific code set a CPU's preempt_count when its lowcore is
allocated, and turn init_idle_preempt_count() into an empty stub.
Fixes: f1a0a376ca ("sched/core: Initialize the idle task with preemption disabled")
Reported-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Link: https://lore.kernel.org/r/20210707163338.1623014-1-valentin.schneider@arm.com
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Lower case matches the call_on_stack() macro and is easier to read.
Reviewed-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
- Rework inline asm to get rid of error prone "register asm" constructs,
which are problematic especially when code instrumentation is enabled. In
particular introduce and use register pair union to allocate even/odd
register pairs. Unfortunately this breaks compatibility with older
clang compilers and minimum clang version for s390 has been raised to 13.
https://lore.kernel.org/linux-next/CAK7LNARuSmPCEy-ak0erPrPTgZdGVypBROFhtw+=3spoGoYsyw@mail.gmail.com/
- Fix gcc 11 warnings, which triggered various minor reworks all over
the code.
- Add zstd kernel image compression support.
- Rework boot CPU lowcore handling.
- De-duplicate and move kernel memory layout setup logic earlier.
- Few fixes in preparation for FORTIFY_SOURCE performing compile-time
and run-time field bounds checking for mem functions.
- Remove broken and unused power management support leftovers in s390
drivers.
- Disable stack-protector for decompressor and purgatory to fix buildroot
build.
- Fix vt220 sclp console name to match the char device name.
- Enable HAVE_IOREMAP_PROT and add zpci_set_irq()/zpci_clear_irq() in
zPCI code.
- Remove some implausible WARN_ON_ONCEs and remove arch specific counter
transaction call backs in favour of default transaction handling in
perf code.
- Extend/add new uevents for online/config/mode state changes of
AP card / queue device in zcrypt.
- Minor entry and ccwgroup code improvements.
- Other small various fixes and improvements all over the code.
-----BEGIN PGP SIGNATURE-----
iQEzBAABCAAdFiEE3QHqV+H2a8xAv27vjYWKoQLXFBgFAmDhuTEACgkQjYWKoQLX
FBjVlggAgDFBkDjlyfvrm4xzmHi7BJMmhrTJIONsSz+3tcA4/u5kE+Hrdrqxm0Uh
ZH4MXBxn4q4Fmoomhu5w5ZDe8o2ip0aN9fFNdsBoP8hurmQbL/IbdTnBETKMrKpV
XpogU2G7p+2nQ0+9+o6PS/vWlZhI88NVh8dWyRd2+5/XdMycgLv2Qm7NpQoACVw1
CbUvxP2PlpZ0wltLvNBKPg1xXMZa3GS0wbVUsS2jiWcr/3VzCqfTHenZJ/RadoE6
axG99QXCbLDMsJgVQcXtlI8K6Z461fAwbNtWZWC+Uq7o5pYuUFW1dovMg9WWF+7T
lFNqXyyNy5wwITRkvuzjlVTE8yzYYg==
=ADZ4
-----END PGP SIGNATURE-----
Merge tag 's390-5.14-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Pull s390 updates from Vasily Gorbik:
- Rework inline asm to get rid of error prone "register asm"
constructs, which are problematic especially when code
instrumentation is enabled.
In particular introduce and use register pair union to allocate
even/odd register pairs. Unfortunately this breaks compatibility with
older clang compilers and minimum clang version for s390 has been
raised to 13.
https://lore.kernel.org/linux-next/CAK7LNARuSmPCEy-ak0erPrPTgZdGVypBROFhtw+=3spoGoYsyw@mail.gmail.com/
- Fix gcc 11 warnings, which triggered various minor reworks all over
the code.
- Add zstd kernel image compression support.
- Rework boot CPU lowcore handling.
- De-duplicate and move kernel memory layout setup logic earlier.
- Few fixes in preparation for FORTIFY_SOURCE performing compile-time
and run-time field bounds checking for mem functions.
- Remove broken and unused power management support leftovers in s390
drivers.
- Disable stack-protector for decompressor and purgatory to fix
buildroot build.
- Fix vt220 sclp console name to match the char device name.
- Enable HAVE_IOREMAP_PROT and add zpci_set_irq()/zpci_clear_irq() in
zPCI code.
- Remove some implausible WARN_ON_ONCEs and remove arch specific
counter transaction call backs in favour of default transaction
handling in perf code.
- Extend/add new uevents for online/config/mode state changes of AP
card / queue device in zcrypt.
- Minor entry and ccwgroup code improvements.
- Other small various fixes and improvements all over the code.
* tag 's390-5.14-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (91 commits)
s390/dasd: use register pair instead of register asm
s390/qdio: get rid of register asm
s390/ioasm: use symbolic names for asm operands
s390/ioasm: get rid of register asm
s390/cmf: get rid of register asm
s390/lib,string: get rid of register asm
s390/lib,uaccess: get rid of register asm
s390/string: get rid of register asm
s390/cmpxchg: use register pair instead of register asm
s390/mm,pages-states: get rid of register asm
s390/lib,xor: get rid of register asm
s390/timex: get rid of register asm
s390/hypfs: use register pair instead of register asm
s390/zcrypt: Switch to flexible array member
s390/speculation: Use statically initialized const for instructions
virtio/s390: get rid of open-coded kvm hypercall
s390/pci: add zpci_set_irq()/zpci_clear_irq()
scripts/min-tool-version.sh: Raise minimum clang version to 13.0.0 for s390
s390/ipl: use register pair instead of register asm
s390/mem_detect: fix tprot() program check new psw handling
...
Per-CPU pointer to lowcore is stored in global lowcore_ptr[]
array and duplicated in struct pcpu::lowcore member. This
update removes the redundancy.
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Once the kernel is running the boot CPU lowcore becomes
freeable and does not differ from the secondary CPU ones
in any way. Make use of it and do not preserve the boot
CPU lowcore on unplugging. That allows returning unused
memory when the boot CPU is offline and makes the code
more clear.
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
The lowcore for IPL CPU is special. It is allocated early
in the boot process using memblock and never freed since.
The reason is pcpu_alloc_lowcore() and pcpu_free_lowcore()
routines use page allocator which is not available when
the IPL CPU is getting initialized.
Similar problem is already addressed for stacks - once the
virtual memory is available the early boot stacks get re-
allocated. Doing the same for lowcore will allow freeing
the IPL CPU lowcore and make no difference between the
boot and secondary CPUs.
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
With gcc-11, there are a lot of warnings because the facility functions
are accessing lowcore through a null pointer. Fix this by moving the
facility arrays away from lowcore.
Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
As pointed out by commit
de9b8f5dcb ("sched: Fix crash trying to dequeue/enqueue the idle thread")
init_idle() can and will be invoked more than once on the same idle
task. At boot time, it is invoked for the boot CPU thread by
sched_init(). Then smp_init() creates the threads for all the secondary
CPUs and invokes init_idle() on them.
As the hotplug machinery brings the secondaries to life, it will issue
calls to idle_thread_get(), which itself invokes init_idle() yet again.
In this case it's invoked twice more per secondary: at _cpu_up(), and at
bringup_cpu().
Given smp_init() already initializes the idle tasks for all *possible*
CPUs, no further initialization should be required. Now, removing
init_idle() from idle_thread_get() exposes some interesting expectations
with regards to the idle task's preempt_count: the secondary startup always
issues a preempt_disable(), requiring some reset of the preempt count to 0
between hot-unplug and hotplug, which is currently served by
idle_thread_get() -> idle_init().
Given the idle task is supposed to have preemption disabled once and never
see it re-enabled, it seems that what we actually want is to initialize its
preempt_count to PREEMPT_DISABLED and leave it there. Do that, and remove
init_idle() from idle_thread_get().
Secondary startups were patched via coccinelle:
@begone@
@@
-preempt_disable();
...
cpu_startup_entry(CPUHP_AP_ONLINE_IDLE);
Signed-off-by: Valentin Schneider <valentin.schneider@arm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210512094636.2958515-1-valentin.schneider@arm.com
When we intercept a DIAG_9C from the guest we verify that the
target real CPU associated with the virtual CPU designated by
the guest is running and if not we forward the DIAG_9C to the
target real CPU.
To avoid a diag9c storm we allow a maximal rate of diag9c forwarding.
The rate is calculated as a count per second defined as a new
parameter of the s390 kvm module: diag9c_forwarding_hz .
The default value of 0 is to not forward diag9c.
Signed-off-by: Pierre Morel <pmorel@linux.ibm.com>
Link: https://lore.kernel.org/r/1613997661-22525-2-git-send-email-pmorel@linux.ibm.com
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
The immediate need to have this is to have bpf_send_signal() send the
signal ASAP instead of during the next hrtimer interrupt. However, it
should also improve irq_work_queue() latencies in general, as well as
get s390 out of the lame architectures list [1].
[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/kernel/irq_work.c?h=v5.11#n45
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Make "cpumask_t cpumask" a static variable to avoid a potential large
stack frame. Also protect against potential concurrent callers by
introducing a local lock.
Note: smp_emergency_stop() gets only called with irqs and machine
checks disabled, therefore a cpu local deadlock is not possible. For
concurrent callers the first cpu which enters the critical section
wins and will stop all other cpus.
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
