From: Jisheng Zhang <jszhang@kernel.org>
To: Samuel Holland <samuel.holland@sifive.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>,
linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, Alexandre Ghiti <alexghiti@rivosinc.com>
Subject: Re: [PATCH v4 12/12] riscv: mm: Always use an ASID to flush mm contexts
Date: Wed, 3 Jan 2024 23:02:24 +0800 [thread overview]
Message-ID: <ZZV3AAY3WIAq+sl9@xhacker> (raw)
In-Reply-To: <20240102220134.3229156-13-samuel.holland@sifive.com>
On Tue, Jan 02, 2024 at 02:00:49PM -0800, Samuel Holland wrote:
> Even if multiple ASIDs are not supported, using the single-ASID variant
> of the sfence.vma instruction preserves TLB entries for global (kernel)
> pages. So it is always more efficient to use the single-ASID code path.
>
> Signed-off-by: Samuel Holland <samuel.holland@sifive.com>
> ---
>
> Changes in v4:
> - There is now only one copy of __flush_tlb_range()
>
> Changes in v2:
> - Update both copies of __flush_tlb_range()
>
> arch/riscv/include/asm/mmu_context.h | 2 --
> arch/riscv/mm/context.c | 3 +--
> arch/riscv/mm/tlbflush.c | 3 +--
> 3 files changed, 2 insertions(+), 6 deletions(-)
>
> diff --git a/arch/riscv/include/asm/mmu_context.h b/arch/riscv/include/asm/mmu_context.h
> index 7030837adc1a..b0659413a080 100644
> --- a/arch/riscv/include/asm/mmu_context.h
> +++ b/arch/riscv/include/asm/mmu_context.h
> @@ -33,8 +33,6 @@ static inline int init_new_context(struct task_struct *tsk,
> return 0;
> }
>
> -DECLARE_STATIC_KEY_FALSE(use_asid_allocator);
> -
> #include <asm-generic/mmu_context.h>
>
> #endif /* _ASM_RISCV_MMU_CONTEXT_H */
> diff --git a/arch/riscv/mm/context.c b/arch/riscv/mm/context.c
> index 3ca9b653df7d..20057085ab8a 100644
> --- a/arch/riscv/mm/context.c
> +++ b/arch/riscv/mm/context.c
> @@ -18,8 +18,7 @@
>
> #ifdef CONFIG_MMU
>
> -DEFINE_STATIC_KEY_FALSE(use_asid_allocator);
> -
> +static DEFINE_STATIC_KEY_FALSE(use_asid_allocator);
One of my optimization "riscv: tlb: avoid tlb flushing if fullmm == 1"
will make use of use_asid_allocator, so could we remove this modification?
> static unsigned long num_asids;
>
> static atomic_long_t current_version;
> diff --git a/arch/riscv/mm/tlbflush.c b/arch/riscv/mm/tlbflush.c
> index 5ec621545c69..39d80f56d292 100644
> --- a/arch/riscv/mm/tlbflush.c
> +++ b/arch/riscv/mm/tlbflush.c
> @@ -84,8 +84,7 @@ static void __flush_tlb_range(struct mm_struct *mm, unsigned long start,
> if (cpumask_empty(cmask))
> return;
>
> - if (static_branch_unlikely(&use_asid_allocator))
> - asid = cntx2asid(atomic_long_read(&mm->context.id));
> + asid = cntx2asid(atomic_long_read(&mm->context.id));
> } else {
> cmask = cpu_online_mask;
> }
> --
> 2.42.0
>
>
> _______________________________________________
> linux-riscv mailing list
> linux-riscv@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-riscv
_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv
WARNING: multiple messages have this Message-ID (diff)
From: Jisheng Zhang <jszhang@kernel.org>
To: Samuel Holland <samuel.holland@sifive.com>
Cc: Palmer Dabbelt <palmer@dabbelt.com>,
linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, Alexandre Ghiti <alexghiti@rivosinc.com>
Subject: Re: [PATCH v4 12/12] riscv: mm: Always use an ASID to flush mm contexts
Date: Wed, 3 Jan 2024 23:02:24 +0800 [thread overview]
Message-ID: <ZZV3AAY3WIAq+sl9@xhacker> (raw)
In-Reply-To: <20240102220134.3229156-13-samuel.holland@sifive.com>
On Tue, Jan 02, 2024 at 02:00:49PM -0800, Samuel Holland wrote:
> Even if multiple ASIDs are not supported, using the single-ASID variant
> of the sfence.vma instruction preserves TLB entries for global (kernel)
> pages. So it is always more efficient to use the single-ASID code path.
>
> Signed-off-by: Samuel Holland <samuel.holland@sifive.com>
> ---
>
> Changes in v4:
> - There is now only one copy of __flush_tlb_range()
>
> Changes in v2:
> - Update both copies of __flush_tlb_range()
>
> arch/riscv/include/asm/mmu_context.h | 2 --
> arch/riscv/mm/context.c | 3 +--
> arch/riscv/mm/tlbflush.c | 3 +--
> 3 files changed, 2 insertions(+), 6 deletions(-)
>
> diff --git a/arch/riscv/include/asm/mmu_context.h b/arch/riscv/include/asm/mmu_context.h
> index 7030837adc1a..b0659413a080 100644
> --- a/arch/riscv/include/asm/mmu_context.h
> +++ b/arch/riscv/include/asm/mmu_context.h
> @@ -33,8 +33,6 @@ static inline int init_new_context(struct task_struct *tsk,
> return 0;
> }
>
> -DECLARE_STATIC_KEY_FALSE(use_asid_allocator);
> -
> #include <asm-generic/mmu_context.h>
>
> #endif /* _ASM_RISCV_MMU_CONTEXT_H */
> diff --git a/arch/riscv/mm/context.c b/arch/riscv/mm/context.c
> index 3ca9b653df7d..20057085ab8a 100644
> --- a/arch/riscv/mm/context.c
> +++ b/arch/riscv/mm/context.c
> @@ -18,8 +18,7 @@
>
> #ifdef CONFIG_MMU
>
> -DEFINE_STATIC_KEY_FALSE(use_asid_allocator);
> -
> +static DEFINE_STATIC_KEY_FALSE(use_asid_allocator);
One of my optimization "riscv: tlb: avoid tlb flushing if fullmm == 1"
will make use of use_asid_allocator, so could we remove this modification?
> static unsigned long num_asids;
>
> static atomic_long_t current_version;
> diff --git a/arch/riscv/mm/tlbflush.c b/arch/riscv/mm/tlbflush.c
> index 5ec621545c69..39d80f56d292 100644
> --- a/arch/riscv/mm/tlbflush.c
> +++ b/arch/riscv/mm/tlbflush.c
> @@ -84,8 +84,7 @@ static void __flush_tlb_range(struct mm_struct *mm, unsigned long start,
> if (cpumask_empty(cmask))
> return;
>
> - if (static_branch_unlikely(&use_asid_allocator))
> - asid = cntx2asid(atomic_long_read(&mm->context.id));
> + asid = cntx2asid(atomic_long_read(&mm->context.id));
> } else {
> cmask = cpu_online_mask;
> }
> --
> 2.42.0
>
>
> _______________________________________________
> linux-riscv mailing list
> linux-riscv@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-riscv
next prev parent reply other threads:[~2024-01-03 16:16 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-02 22:00 [PATCH v4 00/12] riscv: ASID-related and UP-related TLB flush enhancements Samuel Holland
2024-01-02 22:00 ` Samuel Holland
2024-01-02 22:00 ` [PATCH v4 01/12] riscv: Flush the instruction cache during SMP bringup Samuel Holland
2024-01-02 22:00 ` Samuel Holland
2024-01-04 11:58 ` Alexandre Ghiti
2024-01-04 11:58 ` Alexandre Ghiti
2024-01-02 22:00 ` [PATCH v4 02/12] riscv: Use IPIs for remote cache/TLB flushes by default Samuel Holland
2024-01-02 22:00 ` Samuel Holland
2024-01-04 12:09 ` Alexandre Ghiti
2024-01-04 12:09 ` Alexandre Ghiti
2024-01-02 22:00 ` [PATCH v4 03/12] riscv: mm: Broadcast kernel TLB flushes only when needed Samuel Holland
2024-01-02 22:00 ` Samuel Holland
2024-01-04 12:15 ` Alexandre Ghiti
2024-01-04 12:15 ` Alexandre Ghiti
2024-01-02 22:00 ` [PATCH v4 04/12] riscv: Only send remote fences when some other CPU is online Samuel Holland
2024-01-02 22:00 ` Samuel Holland
2024-01-03 14:57 ` Jisheng Zhang
2024-01-03 14:57 ` Jisheng Zhang
2024-01-03 15:04 ` Jisheng Zhang
2024-01-03 15:04 ` Jisheng Zhang
2024-01-04 12:33 ` Alexandre Ghiti
2024-01-04 12:33 ` Alexandre Ghiti
2024-01-04 15:33 ` Samuel Holland
2024-01-04 15:33 ` Samuel Holland
2024-01-02 22:00 ` [PATCH v4 05/12] riscv: mm: Combine the SMP and UP TLB flush code Samuel Holland
2024-01-02 22:00 ` Samuel Holland
2024-01-04 12:36 ` Alexandre Ghiti
2024-01-04 12:36 ` Alexandre Ghiti
2024-01-02 22:00 ` [PATCH v4 06/12] riscv: Apply SiFive CIP-1200 workaround to single-ASID sfence.vma Samuel Holland
2024-01-02 22:00 ` Samuel Holland
2024-01-02 22:00 ` [PATCH v4 07/12] riscv: Avoid TLB flush loops when affected by SiFive CIP-1200 Samuel Holland
2024-01-02 22:00 ` Samuel Holland
2024-01-02 22:00 ` [PATCH v4 08/12] riscv: mm: Introduce cntx2asid/cntx2version helper macros Samuel Holland
2024-01-02 22:00 ` Samuel Holland
2024-01-04 12:39 ` Alexandre Ghiti
2024-01-04 12:39 ` Alexandre Ghiti
2024-01-04 15:42 ` Samuel Holland
2024-01-04 15:42 ` Samuel Holland
2024-01-02 22:00 ` [PATCH v4 09/12] riscv: mm: Use a fixed layout for the MM context ID Samuel Holland
2024-01-02 22:00 ` Samuel Holland
2024-01-04 12:42 ` Alexandre Ghiti
2024-01-04 12:42 ` Alexandre Ghiti
2024-01-02 22:00 ` [PATCH v4 10/12] riscv: mm: Make asid_bits a local variable Samuel Holland
2024-01-02 22:00 ` Samuel Holland
2024-01-03 15:00 ` Jisheng Zhang
2024-01-03 15:00 ` Jisheng Zhang
2024-01-04 15:49 ` Samuel Holland
2024-01-04 15:49 ` Samuel Holland
2024-01-04 12:47 ` Alexandre Ghiti
2024-01-04 12:47 ` Alexandre Ghiti
2024-01-02 22:00 ` [PATCH v4 11/12] riscv: mm: Preserve global TLB entries when switching contexts Samuel Holland
2024-01-02 22:00 ` Samuel Holland
2024-01-04 12:55 ` Alexandre Ghiti
2024-01-04 12:55 ` Alexandre Ghiti
2024-01-02 22:00 ` [PATCH v4 12/12] riscv: mm: Always use an ASID to flush mm contexts Samuel Holland
2024-01-02 22:00 ` Samuel Holland
2024-01-03 15:02 ` Jisheng Zhang [this message]
2024-01-03 15:02 ` Jisheng Zhang
2024-01-04 15:50 ` Samuel Holland
2024-01-04 15:50 ` Samuel Holland
2024-01-04 13:01 ` Alexandre Ghiti
2024-01-04 13:01 ` Alexandre Ghiti
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ZZV3AAY3WIAq+sl9@xhacker \
--to=jszhang@kernel.org \
--cc=alexghiti@rivosinc.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-riscv@lists.infradead.org \
--cc=palmer@dabbelt.com \
--cc=samuel.holland@sifive.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.