The cost of changing a cacheline from shared to exclusive state can be
significant, especially when this is triggered by an exclusive store,
since it may result in having to retry the transaction.
This patch makes use of prefetch.w to prefetch cachelines for write
prior to lr/sc loops when using the xchg_small atomic routine.
This patch is inspired by commit 0ea366f5e1 ("arm64: atomics:
prefetch the destination word for write prior to stxr").
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Signed-off-by: Guo Ren <guoren@kernel.org>
Link: https://lore.kernel.org/r/20231231082955.16516-4-guoren@kernel.org
Tested-by: Andrea Parri <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20250421142441.395849-5-alexghiti@rivosinc.com
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Signed-off-by: Palmer Dabbelt <palmer@dabbelt.com>
RISC-V code uses the queued spinlock implementation, which calls
the macros smp_cond_load_acquire for one byte. So, complement the
implementation of byte and halfword versions.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Signed-off-by: Guo Ren <guoren@kernel.org>
Cc: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Link: https://lore.kernel.org/r/20241217013910.1039923-1-guoren@kernel.org
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Sign extend also an unsigned compare value to match what lr.w is doing.
Otherwise try_cmpxchg may spuriously return true when used on a u32 value
that has the sign bit set, as it happens often in inode_set_ctime_current.
Do this in three conversion steps. The first conversion to long is needed
to avoid a -Wpointer-to-int-cast warning when arch_cmpxchg is used with a
pointer type. Then convert to int and back to long to always sign extend
the 32-bit value to 64-bit.
Fixes: 6c58f25e69 ("riscv/atomic: Fix sign extension for RV64I")
Signed-off-by: Andreas Schwab <schwab@suse.de>
Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Tested-by: Xi Ruoyao <xry111@xry111.site>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/mvmed0k4prh.fsf@suse.de
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
This adds runtime support for Zabha in xchg8/16() operations.
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Andrea Parri <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20241103145153.105097-9-alexghiti@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Now that Zacas is supported in the kernel, let's use the double word
atomic version of amocas to improve the SLUB allocator.
Note that we have to select fixed registers, otherwise gcc fails to pick
even registers and then produces a reserved encoding which fails to
assemble.
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Andrea Parri <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20241103145153.105097-8-alexghiti@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The current fully-ordered cmpxchgXX() implementation results in:
amocas.X.rl a5,a4,(s1)
fence rw,rw
This provides enough sync but we can actually use the following better
mapping instead:
amocas.X.aqrl a5,a4,(s1)
Suggested-by: Andrea Parri <andrea@rivosinc.com>
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Andrea Parri <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20241103145153.105097-7-alexghiti@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
This adds runtime support for Zabha in cmpxchg8/16() operations.
Note that in the absence of Zacas support in the toolchain, CAS
instructions from Zabha won't be used.
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Andrea Parri <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20241103145153.105097-6-alexghiti@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
This adds runtime support for Zacas in cmpxchg operations.
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Andrea Parri <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20241103145153.105097-4-alexghiti@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
riscv does not have lr instructions on byte and halfword but the
qspinlock implementation actually uses such atomics provided by the
Zabha extension, so those sizes are legitimate.
Then instead of failing to build, just fallback to the !Zawrs path.
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Andrea Parri <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20241103145153.105097-3-alexghiti@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
* Support for various new ISA extensions:
* The Zve32[xf] and Zve64[xfd] sub-extensios of the vector
extension.
* Zimop and Zcmop for may-be-operations.
* The Zca, Zcf, Zcd and Zcb sub-extensions of the C extension.
* Zawrs,
* riscv,cpu-intc is now dtschema.
* A handful of performance improvements and cleanups to text patching.
* Support for memory hot{,un}plug
* The highest user-allocatable virtual address is now visible in
hwprobe.
-----BEGIN PGP SIGNATURE-----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=+ICi
-----END PGP SIGNATURE-----
Merge tag 'riscv-for-linus-6.11-mw1' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux
Pull RISC-V updates from Palmer Dabbelt:
- Support for various new ISA extensions:
* The Zve32[xf] and Zve64[xfd] sub-extensios of the vector
extension
* Zimop and Zcmop for may-be-operations
* The Zca, Zcf, Zcd and Zcb sub-extensions of the C extension
* Zawrs
- riscv,cpu-intc is now dtschema
- A handful of performance improvements and cleanups to text patching
- Support for memory hot{,un}plug
- The highest user-allocatable virtual address is now visible in
hwprobe
* tag 'riscv-for-linus-6.11-mw1' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux: (58 commits)
riscv: lib: relax assembly constraints in hweight
riscv: set trap vector earlier
KVM: riscv: selftests: Add Zawrs extension to get-reg-list test
KVM: riscv: Support guest wrs.nto
riscv: hwprobe: export Zawrs ISA extension
riscv: Add Zawrs support for spinlocks
dt-bindings: riscv: Add Zawrs ISA extension description
riscv: Provide a definition for 'pause'
riscv: hwprobe: export highest virtual userspace address
riscv: Improve sbi_ecall() code generation by reordering arguments
riscv: Add tracepoints for SBI calls and returns
riscv: Optimize crc32 with Zbc extension
riscv: Enable DAX VMEMMAP optimization
riscv: mm: Add support for ZONE_DEVICE
virtio-mem: Enable virtio-mem for RISC-V
riscv: Enable memory hotplugging for RISC-V
riscv: mm: Take memory hotplug read-lock during kernel page table dump
riscv: mm: Add memory hotplugging support
riscv: mm: Add pfn_to_kaddr() implementation
riscv: mm: Refactor create_linear_mapping_range() for memory hot add
...
Andrew Jones <ajones@ventanamicro.com> says:
Zawrs provides two instructions (wrs.nto and wrs.sto), where both are
meant to allow the hart to enter a low-power state while waiting on a
store to a memory location. The instructions also both wait an
implementation-defined "short" duration (unless the implementation
terminates the stall for another reason). The difference is that while
wrs.sto will terminate when the duration elapses, wrs.nto, depending on
configuration, will either just keep waiting or an ILL exception will be
raised. Linux will use wrs.nto, so if platforms have an implementation
which falls in the "just keep waiting" category (which is not expected),
then it should _not_ advertise Zawrs in the hardware description.
Like wfi (and with the same {m,h}status bits to configure it), when
wrs.nto is configured to raise exceptions it's expected that the higher
privilege level will see the instruction was a wait instruction, do
something, and then resume execution following the instruction. For
example, KVM does configure exceptions for wfi (hstatus.VTW=1) and
therefore also for wrs.nto. KVM does this for wfi since it's better to
allow other tasks to be scheduled while a VCPU waits for an interrupt.
For waits such as those where wrs.nto/sto would be used, which are
typically locks, it is also a good idea for KVM to be involved, as it
can attempt to schedule the lock holding VCPU.
This series starts with Christoph's addition of the riscv
smp_cond_load_relaxed function which applies wrs.sto when available.
That patch has been reworked to use wrs.nto and to use the same approach
as Arm for the wait loop, since we can't have arbitrary C code between
the load-reserved and the wrs. Then, hwprobe support is added (since the
instructions are also usable from usermode), and finally KVM is
taught about wrs.nto, allowing guests to see and use the Zawrs
extension.
We still don't have test results from hardware, and it's not possible to
prove that using Zawrs is a win when testing on QEMU, not even when
oversubscribing VCPUs to guests. However, it is possible to use KVM
selftests to force a scenario where we can prove Zawrs does its job and
does it well. [4] is a test which does this and, on my machine, without
Zawrs it takes 16 seconds to complete and with Zawrs it takes 0.25
seconds.
This series is also available here [1]. In order to use QEMU for testing
a build with [2] is needed. In order to enable guests to use Zawrs with
KVM using kvmtool, the branch at [3] may be used.
[1] https://github.com/jones-drew/linux/commits/riscv/zawrs-v3/
[2] https://lore.kernel.org/all/20240312152901.512001-2-ajones@ventanamicro.com/
[3] https://github.com/jones-drew/kvmtool/commits/riscv/zawrs/
[4] cb2beccebc
Link: https://lore.kernel.org/r/20240426100820.14762-8-ajones@ventanamicro.com
* b4-shazam-merge:
KVM: riscv: selftests: Add Zawrs extension to get-reg-list test
KVM: riscv: Support guest wrs.nto
riscv: hwprobe: export Zawrs ISA extension
riscv: Add Zawrs support for spinlocks
dt-bindings: riscv: Add Zawrs ISA extension description
riscv: Provide a definition for 'pause'
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
RISC-V code uses the generic ticket lock implementation, which calls
the macros smp_cond_load_relaxed() and smp_cond_load_acquire().
Introduce a RISC-V specific implementation of smp_cond_load_relaxed()
which applies WRS.NTO of the Zawrs extension in order to reduce power
consumption while waiting and allows hypervisors to enable guests to
trap while waiting. smp_cond_load_acquire() doesn't need a RISC-V
specific implementation as the generic implementation is based on
smp_cond_load_relaxed() and smp_acquire__after_ctrl_dep() sufficiently
provides the acquire semantics.
This implementation is heavily based on Arm's approach which is the
approach Andrea Parri also suggested.
The Zawrs specification can be found here:
https://github.com/riscv/riscv-zawrs/blob/main/zawrs.adoc
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
Co-developed-by: Andrew Jones <ajones@ventanamicro.com>
Signed-off-by: Andrew Jones <ajones@ventanamicro.com>
Link: https://lore.kernel.org/r/20240426100820.14762-11-ajones@ventanamicro.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The fully ordered versions of xchg[8|16]() using LR/SC lack the
necessary memory barriers to guarantee the order.
Fix this by matching what is already implemented in the fully ordered
versions of cmpxchg() using LR/SC.
Suggested-by: Andrea Parri <parri.andrea@gmail.com>
Reported-by: Andrea Parri <parri.andrea@gmail.com>
Closes: https://lore.kernel.org/linux-riscv/ZlYbupL5XgzgA0MX@andrea/T/#u
Fixes: a8ed2b7a2c ("riscv/cmpxchg: Implement xchg for variables of size 1 and 2")
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Andrea Parri <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20240530145546.394248-1-alexghiti@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Jisheng Zhang <jszhang@kernel.org> says:
This series selects ARCH_USE_CMPXCHG_LOCKREF to enable the
cmpxchg-based lockless lockref implementation for riscv. Then,
implement arch_cmpxchg64_{relaxed|acquire|release}.
After patch1:
Using Linus' test case[1] on TH1520 platform, I see a 11.2% improvement.
On JH7110 platform, I see 12.0% improvement.
After patch2:
on both TH1520 and JH7110 platforms, I didn't see obvious
performance improvement with Linus' test case [1]. IMHO, this may
be related with the fence and lr.d/sc.d hw implementations. In theory,
lr/sc without fence could give performance improvement over lr/sc plus
fence, so add the code here to leave performance improvement room on
newer HW platforms.
* b4-shazam-merge:
riscv: cmpxchg: implement arch_cmpxchg64_{relaxed|acquire|release}
riscv: select ARCH_USE_CMPXCHG_LOCKREF
Link: http://marc.info/?l=linux-fsdevel&m=137782380714721&w=4 [1]
Link: https://lore.kernel.org/r/20240325111038.1700-1-jszhang@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Leonardo Bras <leobras@redhat.com> says:
While studying riscv's cmpxchg.h file, I got really interested in
understanding how RISCV asm implemented the different versions of
{cmp,}xchg.
When I understood the pattern, it made sense for me to remove the
duplications and create macros to make it easier to understand what exactly
changes between the versions: Instruction sufixes & barriers.
Also, did the same kind of work on atomic.c.
After that, I noted both cmpxchg and xchg only accept variables of
size 4 and 8, compared to x86 and arm64 which do 1,2,4,8.
Now that deduplication is done, it is quite direct to implement them
for variable sizes 1 and 2, so I did it. Then Guo Ren already presented
me some possible users :)
I did compare the generated asm on a test.c that contained usage for every
changed function, and could not detect any change on patches 1 + 2 + 3
compared with upstream.
Pathes 4 & 5 were compiled-tested, merged with guoren/qspinlock_v11 and
booted just fine with qemu -machine virt -append "qspinlock".
(tree: https://gitlab.com/LeoBras/linux/-/commits/guo_qspinlock_v11)
Latest tests happened based on this tree:
https://github.com/guoren83/linux/tree/qspinlock_v12
* b4-shazam-lts:
riscv/cmpxchg: Implement xchg for variables of size 1 and 2
riscv/cmpxchg: Implement cmpxchg for variables of size 1 and 2
riscv/atomic.h : Deduplicate arch_atomic.*
riscv/cmpxchg: Deduplicate cmpxchg() asm and macros
riscv/cmpxchg: Deduplicate xchg() asm functions
Link: https://lore.kernel.org/r/20240103163203.72768-2-leobras@redhat.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
After selecting ARCH_USE_CMPXCHG_LOCKREF, one straight futher
optimization is implementing the arch_cmpxchg64_relaxed() because the
lockref code does not need the cmpxchg to have barrier semantics. At
the same time, implement arch_cmpxchg64_acquire and
arch_cmpxchg64_release as well.
However, on both TH1520 and JH7110 platforms, I didn't see obvious
performance improvement with Linus' test case [1]. IMHO, this may
be related with the fence and lr.d/sc.d hw implementations. In theory,
lr/sc without fence could give performance improvement over lr/sc plus
fence, so add the code here to leave performance improvement room on
newer HW platforms.
Link: http://marc.info/?l=linux-fsdevel&m=137782380714721&w=4 [1]
Signed-off-by: Jisheng Zhang <jszhang@kernel.org>
Reviewed-by: Andrea Parri <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20240325111038.1700-3-jszhang@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
xchg for variables of size 1-byte and 2-bytes is not yet available for
riscv, even though its present in other architectures such as arm64 and
x86. This could lead to not being able to implement some locking mechanisms
or requiring some rework to make it work properly.
Implement 1-byte and 2-bytes xchg in order to achieve parity with other
architectures.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Tested-by: Guo Ren <guoren@kernel.org>
Link: https://lore.kernel.org/r/20240103163203.72768-7-leobras@redhat.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
cmpxchg for variables of size 1-byte and 2-bytes is not yet available for
riscv, even though its present in other architectures such as arm64 and
x86. This could lead to not being able to implement some locking mechanisms
or requiring some rework to make it work properly.
Implement 1-byte and 2-bytes cmpxchg in order to achieve parity with other
architectures.
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Tested-by: Guo Ren <guoren@kernel.org>
Link: https://lore.kernel.org/r/20240103163203.72768-6-leobras@redhat.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
In this header every cmpxchg define (_relaxed, _acquire, _release,
vanilla) contain it's own asm file, both for 4-byte variables an 8-byte
variables, on a total of 8 versions of mostly the same asm.
This is usually bad, as it means any change may be done in up to 8
different places.
Unify those versions by creating a new define with enough parameters to
generate any version of the previous 8.
Then unify the result under a more general define, and simplify
arch_cmpxchg* generation
(This did not cause any change in generated asm)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Guo Ren <guoren@kernel.org>
Reviewed-by: Andrea Parri <parri.andrea@gmail.com>
Tested-by: Guo Ren <guoren@kernel.org>
Link: https://lore.kernel.org/r/20240103163203.72768-4-leobras@redhat.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
In this header every xchg define (_relaxed, _acquire, _release, vanilla)
contain it's own asm file, both for 4-byte variables an 8-byte variables,
on a total of 8 versions of mostly the same asm.
This is usually bad, as it means any change may be done in up to 8
different places.
Unify those versions by creating a new define with enough parameters to
generate any version of the previous 8.
Then unify the result under a more general define, and simplify
arch_xchg* generation.
(This did not cause any change in generated asm)
Signed-off-by: Leonardo Bras <leobras@redhat.com>
Reviewed-by: Guo Ren <guoren@kernel.org>
Reviewed-by: Andrea Parri <parri.andrea@gmail.com>
Tested-by: Guo Ren <guoren@kernel.org>
Link: https://lore.kernel.org/r/20240103163203.72768-3-leobras@redhat.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Disparate fence implementations are consolidated into fence.h.
Also introduce RISCV_FENCE_ASM to make fence macro more reusable.
Signed-off-by: Eric Chan <ericchancf@google.com>
Reviewed-by: Andrea Parri <parri.andrea@gmail.com>
Reviewed-by: Samuel Holland <samuel.holland@sifive.com>
Tested-by: Samuel Holland <samuel.holland@sifive.com>
Link: https://lore.kernel.org/r/20240217131316.3668927-1-ericchancf@google.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Introduce RISCV_FULL_BARRIER and use in arch_atomic* function.
like RISCV_ACQUIRE_BARRIER and RISCV_RELEASE_BARRIER, the fence
instruction can be eliminated When SMP is not enabled.
Signed-off-by: Eric Chan <ericchancf@google.com>
Reviewed-by: Andrea Parri <parri.andrea@gmail.com>
Reviewed-by: Samuel Holland <samuel.holland@sifive.com>
Tested-by: Samuel Holland <samuel.holland@sifive.com>
Link: https://lore.kernel.org/r/20240217131302.3668481-1-ericchancf@google.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Decrease the probability of this internal facility to be used by
driver code.
Signed-off-by: Andrzej Hajda <andrzej.hajda@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> [m68k]
Acked-by: Palmer Dabbelt <palmer@rivosinc.com> [riscv]
Link: https://lore.kernel.org/r/20230118154450.73842-1-andrzej.hajda@intel.com
Cc: Linus Torvalds <torvalds@linux-foundation.org>
The cmpxchg32 & cmpxchg32_local are not used in Linux anymore. So
clean up asm/cmpxchg.h.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Signed-off-by: Guo Ren <guoren@kernel.org>
Link: https://lore.kernel.org/r/20220505035526.2974382-2-guoren@kernel.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
We'd like all architectures to convert to ARCH_ATOMIC, as once all
architectures are converted it will be possible to make significant
cleanups to the atomics headers, and this will make it much easier to
generically enable atomic functionality (e.g. debug logic in the
instrumented wrappers).
As a step towards that, this patch migrates riscv to ARCH_ATOMIC. The
arch code provides arch_{atomic,atomic64,xchg,cmpxchg}*(), and common
code wraps these with optional instrumentation to provide the regular
functions.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: Palmer Dabbelt <palmerdabbelt@google.com>
Acked-by: Palmer Dabbelt <palmerdabbelt@google.com>
Cc: Albert Ou <aou@eecs.berkeley.edu>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Paul Walmsley <paul.walmsley@sifive.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210525140232.53872-29-mark.rutland@arm.com
The argument passed to cmpxchg is not guaranteed to be sign
extended, but lr.w sign extends on RV64I. This makes cmpxchg
fail on clang built kernels when __old is negative.
To fix this, we just cast __old to long which sign extends on
RV64I. With this fix, clang built RISC-V kernels now boot.
Link: https://github.com/ClangBuiltLinux/linux/issues/867
Signed-off-by: Nathan Huckleberry <nhuck@google.com>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Based on 1 normalized pattern(s):
this program is free software you can redistribute it and or modify
it under the terms of the gnu general public license as published by
the free software foundation version 2 this program is distributed
in the hope that it will be useful but without any warranty without
even the implied warranty of merchantability or fitness for a
particular purpose see the gnu general public license for more
details
extracted by the scancode license scanner the SPDX license identifier
GPL-2.0-only
has been chosen to replace the boilerplate/reference in 97 file(s).
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Allison Randal <allison@lohutok.net>
Reviewed-by: Alexios Zavras <alexios.zavras@intel.com>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190529141901.025053186@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Atomics present the same issue with locking: release and acquire
variants need to be strengthened to meet the constraints defined
by the Linux-kernel memory consistency model [1].
Atomics present a further issue: implementations of atomics such
as atomic_cmpxchg() and atomic_add_unless() rely on LR/SC pairs,
which do not give full-ordering with .aqrl; for example, current
implementations allow the "lr-sc-aqrl-pair-vs-full-barrier" test
below to end up with the state indicated in the "exists" clause.
In order to "synchronize" LKMM and RISC-V's implementation, this
commit strengthens the implementations of the atomics operations
by replacing .rl and .aq with the use of ("lightweigth") fences,
and by replacing .aqrl LR/SC pairs in sequences such as:
0: lr.w.aqrl %0, %addr
bne %0, %old, 1f
...
sc.w.aqrl %1, %new, %addr
bnez %1, 0b
1:
with sequences of the form:
0: lr.w %0, %addr
bne %0, %old, 1f
...
sc.w.rl %1, %new, %addr /* SC-release */
bnez %1, 0b
fence rw, rw /* "full" fence */
1:
following Daniel's suggestion.
These modifications were validated with simulation of the RISC-V
memory consistency model.
C lr-sc-aqrl-pair-vs-full-barrier
{}
P0(int *x, int *y, atomic_t *u)
{
int r0;
int r1;
WRITE_ONCE(*x, 1);
r0 = atomic_cmpxchg(u, 0, 1);
r1 = READ_ONCE(*y);
}
P1(int *x, int *y, atomic_t *v)
{
int r0;
int r1;
WRITE_ONCE(*y, 1);
r0 = atomic_cmpxchg(v, 0, 1);
r1 = READ_ONCE(*x);
}
exists (u=1 /\ v=1 /\ 0:r1=0 /\ 1:r1=0)
[1] https://marc.info/?l=linux-kernel&m=151930201102853&w=2https://groups.google.com/a/groups.riscv.org/forum/#!topic/isa-dev/hKywNHBkAXMhttps://marc.info/?l=linux-kernel&m=151633436614259&w=2
Suggested-by: Daniel Lustig <dlustig@nvidia.com>
Signed-off-by: Andrea Parri <parri.andrea@gmail.com>
Cc: Palmer Dabbelt <palmer@sifive.com>
Cc: Albert Ou <albert@sifive.com>
Cc: Daniel Lustig <dlustig@nvidia.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Jade Alglave <j.alglave@ucl.ac.uk>
Cc: Luc Maranget <luc.maranget@inria.fr>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Akira Yokosawa <akiyks@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: linux-riscv@lists.infradead.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Palmer Dabbelt <palmer@sifive.com>
This contains all the code that directly interfaces with the RISC-V
memory model. While this code corforms to the current RISC-V ISA
specifications (user 2.2 and priv 1.10), the memory model is somewhat
underspecified in those documents. There is a working group that hopes
to produce a formal memory model by the end of the year, but my
understanding is that the basic definitions we're relying on here won't
change significantly.
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Palmer Dabbelt <palmer@dabbelt.com>
