public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: Rik van Riel <riel@surriel.com>
Cc: x86@kernel.org, linux-kernel@vger.kernel.org,
	kernel-team@meta.com, dave.hansen@linux.intel.com,
	luto@kernel.org, tglx@linutronix.de, mingo@redhat.com,
	bp@alien8.de, hpa@zytor.com, akpm@linux-foundation.org
Subject: Re: [PATCH 04/10] x86,mm: use INVLPGB for kernel TLB flushes
Date: Sun, 22 Dec 2024 12:16:06 +0100	[thread overview]
Message-ID: <20241222111606.GU11133@noisy.programming.kicks-ass.net> (raw)
In-Reply-To: <20241222040717.3096835-5-riel@surriel.com>

On Sat, Dec 21, 2024 at 11:06:36PM -0500, Rik van Riel wrote:
> Use broadcast TLB invalidation for kernel addresses when available.
> 
> This stops us from having to send IPIs for kernel TLB flushes.
> 
> Signed-off-by: Rik van Riel <riel@surriel.com>
> ---
>  arch/x86/mm/tlb.c | 32 ++++++++++++++++++++++++++++++++
>  1 file changed, 32 insertions(+)
> 
> diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
> index 6cf881a942bb..09980fb17907 100644
> --- a/arch/x86/mm/tlb.c
> +++ b/arch/x86/mm/tlb.c
> @@ -1077,6 +1077,33 @@ void flush_tlb_all(void)
>  	on_each_cpu(do_flush_tlb_all, NULL, 1);
>  }
>  
> +static void broadcast_kernel_range_flush(unsigned long start, unsigned long end)
> +{
> +	unsigned long addr;
> +	unsigned long maxnr = boot_cpu_data.invlpgb_count_max;
> +	unsigned long threshold = tlb_single_page_flush_ceiling * maxnr;
> +
> +	/*
> +	 * TLBSYNC only waits for flushes originating on the same CPU.
> +	 * Disabling migration allows us to wait on all flushes.
> +	 */
> +	migrate_disable();

So how expensive is all this? That is, I think I would feel better is
this were preempt_disable().

> +
> +	if (end == TLB_FLUSH_ALL ||
> +	    (end - start) > threshold << PAGE_SHIFT) {
> +		invlpgb_flush_all();
> +	} else {
> +		unsigned long nr;
> +		for (addr = start; addr < end; addr += nr << PAGE_SHIFT) {
> +			nr = min((end - addr) >> PAGE_SHIFT, maxnr);
> +			invlpgb_flush_addr(addr, nr);
> +		}
> +	}
> +
> +	tlbsync();
> +	migrate_enable();
> +}
> +
>  static void do_kernel_range_flush(void *info)
>  {
>  	struct flush_tlb_info *f = info;
> @@ -1089,6 +1116,11 @@ static void do_kernel_range_flush(void *info)
>  
>  void flush_tlb_kernel_range(unsigned long start, unsigned long end)
>  {
> +	if (static_cpu_has(X86_FEATURE_INVLPGB)) {
> +		broadcast_kernel_range_flush(start, end);
> +		return;
> +	}
> +
>  	/* Balance as user space task's flush, a bit conservative */
>  	if (end == TLB_FLUSH_ALL ||
>  	    (end - start) > tlb_single_page_flush_ceiling << PAGE_SHIFT) {
> -- 
> 2.47.1
> 

  reply	other threads:[~2024-12-22 11:16 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-12-22  4:06 [RFC PATCH 00/10] AMD broadcast TLB invalidation Rik van Riel
2024-12-22  4:06 ` [PATCH 01/10] Add X86_FEATURE_INVLPGB definition Rik van Riel
2024-12-22  4:06 ` [PATCH 02/10] x86,tlb: get INVLPGB count max from CPUID Rik van Riel
2024-12-22 11:46   ` Borislav Petkov
2024-12-22 14:39     ` Rik van Riel
2024-12-22  4:06 ` [PATCH 03/10] x86,mm: add INVLPGB support code Rik van Riel
2024-12-22 11:05   ` Peter Zijlstra
2024-12-22 21:24     ` Rik van Riel
2024-12-22  4:06 ` [PATCH 04/10] x86,mm: use INVLPGB for kernel TLB flushes Rik van Riel
2024-12-22 11:16   ` Peter Zijlstra [this message]
2024-12-22 15:12     ` Rik van Riel
2024-12-24 18:13       ` Peter Zijlstra
2024-12-22 11:47   ` Borislav Petkov
2024-12-22  4:06 ` [PATCH 05/10] x86,tlb: use INVLPGB in flush_tlb_all Rik van Riel
2024-12-22 11:17   ` Peter Zijlstra
2024-12-22  4:06 ` [PATCH 06/10] x86,mm: use broadcast TLB flushing for page reclaim TLB flushing Rik van Riel
2024-12-22 11:18   ` Peter Zijlstra
2024-12-22  4:06 ` [PATCH 07/10] x86,mm: enable broadcast TLB invalidation for multi-threaded processes Rik van Riel
2024-12-22 11:36   ` Peter Zijlstra
2024-12-22 22:45     ` Rik van Riel
2024-12-22  4:06 ` [PATCH 08/10] x86,tlb: do targeted broadcast flushing from tlbbatch code Rik van Riel
2024-12-22  4:06 ` [PATCH 09/10] x86/mm: enable AMD translation cache extensions Rik van Riel
2024-12-22 11:38   ` Peter Zijlstra
2024-12-22 15:37     ` Rik van Riel
2024-12-24 18:25       ` Peter Zijlstra
2024-12-22  4:06 ` [PATCH 10/10] x86/mm: only invalidate final translations with INVLPGB Rik van Riel
2024-12-22 11:11 ` [RFC PATCH 00/10] AMD broadcast TLB invalidation Peter Zijlstra
2024-12-22 15:06   ` Rik van Riel

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=20241222111606.GU11133@noisy.programming.kicks-ass.net \
    --to=peterz@infradead.org \
    --cc=akpm@linux-foundation.org \
    --cc=bp@alien8.de \
    --cc=dave.hansen@linux.intel.com \
    --cc=hpa@zytor.com \
    --cc=kernel-team@meta.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@kernel.org \
    --cc=mingo@redhat.com \
    --cc=riel@surriel.com \
    --cc=tglx@linutronix.de \
    --cc=x86@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