Message-Id: <20250627044805.945491-1-ankur.a.arora@oracle.com>
Date: Thu, 26 Jun 2025 21:48:00 -0700
From: Ankur Arora <ankur.a.arora@...cle.com>
To: linux-kernel@...r.kernel.org, linux-arch@...r.kernel.org,
        linux-arm-kernel@...ts.infradead.org, bpf@...r.kernel.org
Cc: arnd@...db.de, catalin.marinas@....com, will@...nel.org,
        peterz@...radead.org, akpm@...ux-foundation.org, mark.rutland@....com,
        harisokn@...zon.com, cl@...two.org, ast@...nel.org, memxor@...il.com,
        zhenglifeng1@...wei.com, xueshuai@...ux.alibaba.com,
        joao.m.martins@...cle.com, boris.ostrovsky@...cle.com,
        konrad.wilk@...cle.com
Subject: [PATCH v3 0/5] barrier: Add smp_cond_load_*_timewait()

Hi,

This series adds waited variants of the smp_cond_load() primitives:
smp_cond_load_relaxed_timewait(), and smp_cond_load_acquire_timewait().

Why?: as the name suggests, the new interfaces are meant for contexts
where you want to wait on a condition variable for a finite duration.
This is easy enough to do with a loop around cpu_relax(). However,
some architectures (ex. arm64) also allow waiting on a cacheline. So,
these interfaces handle a mixture of spin/wait with a smp_cond_load()
thrown in.

There are two known users for these interfaces:

 - poll_idle() [1]
 - resilient queued spinlocks [2]

The interfaces are:
   smp_cond_load_relaxed_spinwait(ptr, cond_expr,
                                  time_expr, time_limit, slack)
   smp_cond_load_acquire_spinwait(ptr, cond_expr,
                                  time_expr, time_limit, slack)

The added parameters pertain to the timeout checks and a measure of how
much slack the caller can tolerate in the timeout. The slack is useful
when in the wait state and thus dependent on an asynchronous event.

Changelog:
  v2 [3]:
    - simplified the interface (suggested by Catalin Marinas)
       - get rid of wait_policy, and a multitude of constants
       - adds a slack parameter
      This helped remove a fair amount of duplicated code duplication and in hindsight
      unnecessary constants.

  v1 [4]:
     - add wait_policy (coarse and fine)
     - derive spin-count etc at runtime instead of using arbitrary
       constants.

Haris Okanovic had tested an earlier version of this series with 
poll_idle()/haltpoll patches. [5]

Any comments appreciated!

Ankur

[1] https://lore.kernel.org/lkml/20241107190818.522639-3-ankur.a.arora@oracle.com/
[2] Uses the smp_cond_load_acquire_timewait() from v1
    https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/arm64/include/asm/rqspinlock.h
[3] https://lore.kernel.org/lkml/20250502085223.1316925-1-ankur.a.arora@oracle.com/
[4] https://lore.kernel.org/lkml/20250203214911.898276-1-ankur.a.arora@oracle.com/
[5] https://lore.kernel.org/lkml/f2f5d09e79539754ced085ed89865787fa668695.camel@amazon.com

Cc: Arnd Bergmann <arnd@...db.de>
Cc: Will Deacon <will@...nel.org>
Cc: Catalin Marinas <catalin.marinas@....com>
Cc: Peter Zijlstra <peterz@...radead.org>
Cc: Kumar Kartikeya Dwivedi <memxor@...il.com>
Cc: Alexei Starovoitov <ast@...nel.org>
Cc: linux-arch@...r.kernel.org

Ankur Arora (5):
  asm-generic: barrier: Add smp_cond_load_relaxed_timewait()
  asm-generic: barrier: Handle spin-wait in
    smp_cond_load_relaxed_timewait()
  asm-generic: barrier: Add smp_cond_load_acquire_timewait()
  arm64: barrier: Support waiting in smp_cond_load_relaxed_timewait()
  arm64: barrier: Handle waiting in smp_cond_load_relaxed_timewait()

 arch/arm64/include/asm/barrier.h    |  54 +++++++++++
 arch/arm64/include/asm/rqspinlock.h |   2 +-
 include/asm-generic/barrier.h       | 137 ++++++++++++++++++++++++++++
 3 files changed, 192 insertions(+), 1 deletion(-)

-- 
2.43.5

Powered by blists - more mailing lists