linux-arch.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrew Jones <ajones@ventanamicro.com>
To: Alexandre Ghiti <alexghiti@rivosinc.com>
Cc: Will Deacon <will@kernel.org>,
	"Aneesh Kumar K . V" <aneesh.kumar@linux.ibm.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Nick Piggin <npiggin@gmail.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Mayuresh Chitale <mchitale@ventanamicro.com>,
	Vincent Chen <vincent.chen@sifive.com>,
	Paul Walmsley <paul.walmsley@sifive.com>,
	Palmer Dabbelt <palmer@dabbelt.com>,
	Albert Ou <aou@eecs.berkeley.edu>,
	linux-arch@vger.kernel.org, linux-mm@kvack.org,
	linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 3/4] riscv: Make __flush_tlb_range() loop over pte instead of flushing the whole tlb
Date: Fri, 28 Jul 2023 15:32:35 +0200	[thread overview]
Message-ID: <20230728-f2cd8ddd252c2ece2e438790@orel> (raw)
In-Reply-To: <20230727185553.980262-4-alexghiti@rivosinc.com>

On Thu, Jul 27, 2023 at 08:55:52PM +0200, Alexandre Ghiti wrote:
> Currently, when the range to flush covers more than one page (a 4K page or
> a hugepage), __flush_tlb_range() flushes the whole tlb. Flushing the whole
> tlb comes with a greater cost than flushing a single entry so we should
> flush single entries up to a certain threshold so that:
> threshold * cost of flushing a single entry < cost of flushing the whole
> tlb.
> 
> This threshold is microarchitecture dependent and can/should be
> overwritten by vendors.
> 
> Co-developed-by: Mayuresh Chitale <mchitale@ventanamicro.com>
> Signed-off-by: Mayuresh Chitale <mchitale@ventanamicro.com>
> Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
> ---
>  arch/riscv/mm/tlbflush.c | 41 ++++++++++++++++++++++++++++++++++++++--
>  1 file changed, 39 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/riscv/mm/tlbflush.c b/arch/riscv/mm/tlbflush.c
> index 3e4acef1f6bc..8017d2130e27 100644
> --- a/arch/riscv/mm/tlbflush.c
> +++ b/arch/riscv/mm/tlbflush.c
> @@ -24,13 +24,48 @@ static inline void local_flush_tlb_page_asid(unsigned long addr,
>  			: "memory");
>  }
>  
> +/*
> + * Flush entire TLB if number of entries to be flushed is greater
> + * than the threshold below. Platforms may override the threshold
> + * value based on marchid, mvendorid, and mimpid.
> + */
> +static unsigned long tlb_flush_all_threshold __read_mostly = 64;
> +
> +static void local_flush_tlb_range_threshold_asid(unsigned long start,
> +						 unsigned long size,
> +						 unsigned long stride,
> +						 unsigned long asid)
> +{
> +	u16 nr_ptes_in_range = DIV_ROUND_UP(size, stride);
> +	int i;
> +
> +	if (nr_ptes_in_range > tlb_flush_all_threshold) {
> +		if (asid != -1)
> +			local_flush_tlb_all_asid(asid);
> +		else
> +			local_flush_tlb_all();
> +		return;
> +	}
> +
> +	for (i = 0; i < nr_ptes_in_range; ++i) {
> +		if (asid != -1)
> +			local_flush_tlb_page_asid(start, asid);
> +		else
> +			local_flush_tlb_page(start);
> +		start += stride;
> +	}
> +}
> +
>  static inline void local_flush_tlb_range(unsigned long start,
>  		unsigned long size, unsigned long stride)
>  {
>  	if (size <= stride)
>  		local_flush_tlb_page(start);
> -	else
> +	else if (size == (unsigned long)-1)

The more we scatter this -1 around, especially now that we also need to
cast it, the more I think we should introduce a #define for it.

>  		local_flush_tlb_all();
> +	else
> +		local_flush_tlb_range_threshold_asid(start, size, stride, -1);
> +
>  }
>  
>  static inline void local_flush_tlb_range_asid(unsigned long start,
> @@ -38,8 +73,10 @@ static inline void local_flush_tlb_range_asid(unsigned long start,
>  {
>  	if (size <= stride)
>  		local_flush_tlb_page_asid(start, asid);
> -	else
> +	else if (size == (unsigned long)-1)
>  		local_flush_tlb_all_asid(asid);
> +	else
> +		local_flush_tlb_range_threshold_asid(start, size, stride, asid);
>  }
>  
>  static void __ipi_flush_tlb_all(void *info)
> -- 
> 2.39.2
>

Otherwise,

Reviewed-by: Andrew Jones <ajones@ventanamicro.com>

Thanks,
drew

  reply	other threads:[~2023-07-28 13:34 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-07-27 18:55 [PATCH v2 0/4] riscv: tlb flush improvements Alexandre Ghiti
2023-07-27 18:55 ` [PATCH v2 1/4] riscv: Improve flush_tlb() Alexandre Ghiti
2023-07-28 13:35   ` Andrew Jones
2023-07-27 18:55 ` [PATCH v2 2/4] riscv: Improve flush_tlb_range() for hugetlb pages Alexandre Ghiti
2023-07-28 13:40   ` Andrew Jones
2023-07-27 18:55 ` [PATCH v2 3/4] riscv: Make __flush_tlb_range() loop over pte instead of flushing the whole tlb Alexandre Ghiti
2023-07-28 13:32   ` Andrew Jones [this message]
2023-07-28 13:51     ` Conor Dooley
2023-07-28 13:50   ` Conor Dooley
2023-07-27 18:55 ` [PATCH v2 4/4] riscv: Improve flush_tlb_kernel_range() Alexandre Ghiti
2023-07-28 13:48   ` Andrew Jones

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=20230728-f2cd8ddd252c2ece2e438790@orel \
    --to=ajones@ventanamicro.com \
    --cc=akpm@linux-foundation.org \
    --cc=alexghiti@rivosinc.com \
    --cc=aneesh.kumar@linux.ibm.com \
    --cc=aou@eecs.berkeley.edu \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-riscv@lists.infradead.org \
    --cc=mchitale@ventanamicro.com \
    --cc=npiggin@gmail.com \
    --cc=palmer@dabbelt.com \
    --cc=paul.walmsley@sifive.com \
    --cc=peterz@infradead.org \
    --cc=vincent.chen@sifive.com \
    --cc=will@kernel.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).