[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAFULd4ZP58hPacGp+n_bBZV0e7nomzLMdX7OOu3DEeGe2OxomA@mail.gmail.com>
Date: Wed, 26 Feb 2025 20:36:00 +0100
From: Uros Bizjak <ubizjak@...il.com>
To: Brian Gerst <brgerst@...il.com>
Cc: linux-kernel@...r.kernel.org, x86@...nel.org,
Ingo Molnar <mingo@...nel.org>, "H . Peter Anvin" <hpa@...or.com>, Thomas Gleixner <tglx@...utronix.de>,
Borislav Petkov <bp@...en8.de>, Ard Biesheuvel <ardb@...nel.org>,
Linus Torvalds <torvalds@...uxfoundation.org>, Andy Lutomirski <luto@...nel.org>,
Peter Zijlstra <peterz@...radead.org>, Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [PATCH v2 01/11] percpu: Introduce percpu hot section
On Wed, Feb 26, 2025 at 7:05 PM Brian Gerst <brgerst@...il.com> wrote:
>
> Add a subsection to the percpu data for frequently accessed variables
> that should remain cached on each processor. These varables should not
> be accessed from other processors to avoid cacheline bouncing.
>
> This will replace the pcpu_hot struct on x86, and open up similar
> functionality to other architectures and the kernel core.
>
> Signed-off-by: Brian Gerst <brgerst@...il.com>
> ---
> include/asm-generic/vmlinux.lds.h | 10 ++++++++++
> include/linux/percpu-defs.h | 12 ++++++++++++
> 2 files changed, 22 insertions(+)
>
> diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
> index 92fc06f7da74..92dd6065fd0a 100644
> --- a/include/asm-generic/vmlinux.lds.h
> +++ b/include/asm-generic/vmlinux.lds.h
> @@ -385,6 +385,11 @@ defined(CONFIG_AUTOFDO_CLANG) || defined(CONFIG_PROPELLER_CLANG)
> . = ALIGN(PAGE_SIZE); \
> __nosave_end = .;
>
> +#define CACHE_HOT_DATA(align) \
> + . = ALIGN(align); \
> + *(SORT_BY_ALIGNMENT(.data..hot.*)) \
> + . = ALIGN(align);
> +
> #define PAGE_ALIGNED_DATA(page_align) \
> . = ALIGN(page_align); \
> *(.data..page_aligned) \
> @@ -1065,6 +1070,10 @@ defined(CONFIG_AUTOFDO_CLANG) || defined(CONFIG_PROPELLER_CLANG)
> . = ALIGN(PAGE_SIZE); \
> *(.data..percpu..page_aligned) \
> . = ALIGN(cacheline); \
> + __per_cpu_hot_start = .; \
> + *(SORT_BY_ALIGNMENT(.data..percpu..hot.*)) \
> + . = ALIGN(cacheline); \
> + __per_cpu_hot_end = .; \
> *(.data..percpu..read_mostly) \
> . = ALIGN(cacheline); \
> *(.data..percpu) \
> @@ -1112,6 +1121,7 @@ defined(CONFIG_AUTOFDO_CLANG) || defined(CONFIG_PROPELLER_CLANG)
> INIT_TASK_DATA(inittask) \
> NOSAVE_DATA \
> PAGE_ALIGNED_DATA(pagealigned) \
> + CACLE_HOT_DATA(cacheline) \
There is a typo in the above macro name.
Uros.
> CACHELINE_ALIGNED_DATA(cacheline) \
> READ_MOSTLY_DATA(cacheline) \
> DATA_DATA \
> diff --git a/include/linux/percpu-defs.h b/include/linux/percpu-defs.h
> index 40d34e032d5b..eb3393f96e5a 100644
> --- a/include/linux/percpu-defs.h
> +++ b/include/linux/percpu-defs.h
> @@ -112,6 +112,18 @@
> #define DEFINE_PER_CPU(type, name) \
> DEFINE_PER_CPU_SECTION(type, name, "")
>
> +/*
> + * Declaration/definition used for per-CPU variables that are frequently
> + * accessed and should be in a single cacheline.
> + *
> + * For use only by architecture and core code.
> + */
> +#define DECLARE_PER_CPU_CACHE_HOT(type, name) \
> + DECLARE_PER_CPU_SECTION(type, name, "..hot.." #name)
> +
> +#define DEFINE_PER_CPU_CACHE_HOT(type, name) \
> + DEFINE_PER_CPU_SECTION(type, name, "..hot.." #name)
> +
> /*
> * Declaration/definition used for per-CPU variables that must be cacheline
> * aligned under SMP conditions so that, whilst a particular instance of the
> --
> 2.48.1
>
Powered by blists - more mailing lists