linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Lorenzo Stoakes <lstoakes@gmail.com>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>,
	Christoph Hellwig <hch@lst.de>,
	Uladzislau Rezki <urezki@gmail.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Baoquan He <bhe@redhat.com>
Subject: Re: [V2 patch 3/6] mm/vmalloc: Prevent flushing dirty space over and over
Date: Sat, 27 May 2023 19:19:32 +0100	[thread overview]
Message-ID: <590a1dc3-ce57-4d13-a9c8-82ac5832ab99@lucifer.local> (raw)
In-Reply-To: <20230525124504.692056496@linutronix.de>

On Thu, May 25, 2023 at 02:57:05PM +0200, Thomas Gleixner wrote:
> vmap blocks which have active mappings cannot be purged. Allocations which
> have been freed are accounted for in vmap_block::dirty_min/max, so that
> they can be detected in _vm_unmap_aliases() as potentially stale TLBs.
>
> If there are several invocations of _vm_unmap_aliases() then each of them
> will flush the dirty range. That's pointless and just increases the
> probability of full TLB flushes.
>
> Avoid that by resetting the flush range after accounting for it. That's
> safe versus other invocations of _vm_unmap_aliases() because this is all
> serialized with vmap_purge_lock.
>
> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
> Reviewed-by: Baoquan He <bhe@redhat.com>
> ---
>  mm/vmalloc.c |    8 ++++++--
>  1 file changed, 6 insertions(+), 2 deletions(-)
>
> --- a/mm/vmalloc.c
> +++ b/mm/vmalloc.c
> @@ -2226,7 +2226,7 @@ static void vb_free(unsigned long addr,
>
>  	spin_lock(&vb->lock);
>
> -	/* Expand dirty range */
> +	/* Expand the not yet TLB flushed dirty range */
>  	vb->dirty_min = min(vb->dirty_min, offset);
>  	vb->dirty_max = max(vb->dirty_max, offset + (1UL << order));
>
> @@ -2264,7 +2264,7 @@ static void _vm_unmap_aliases(unsigned l
>  			 * space to be flushed.
>  			 */
>  			if (!purge_fragmented_block(vb, vbq, &purge_list) &&
> -			    vb->dirty && vb->dirty != VMAP_BBMAP_BITS) {
> +			    vb->dirty_max && vb->dirty != VMAP_BBMAP_BITS) {
>  				unsigned long va_start = vb->va->va_start;
>  				unsigned long s, e;
>
> @@ -2274,6 +2274,10 @@ static void _vm_unmap_aliases(unsigned l
>  				start = min(s, start);
>  				end   = max(e, end);
>
> +				/* Prevent that this is flushed again */

Sorry to be absurdly nitty, and I know there's other instances of this in
vmalloc.c but it'd be nice to correct the grammar here -> 'prevent this
from being flushed again'. However, far from urgent!

> +				vb->dirty_min = VMAP_BBMAP_BITS;
> +				vb->dirty_max = 0;
> +
>  				flush = 1;
>  			}
>  			spin_unlock(&vb->lock);
>
>

This is a really good catch, feel free to add:-

Reviewed-by: Lorenzo Stoakes <lstoakes@gmail.com>


  parent reply	other threads:[~2023-05-27 18:19 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-05-25 12:57 [V2 patch 0/6] mm/vmalloc: Assorted fixes and improvements Thomas Gleixner
2023-05-25 12:57 ` [V2 patch 1/6] mm/vmalloc: Prevent stale TLBs in fully utilized blocks Thomas Gleixner
2023-05-25 12:57 ` [V2 patch 2/6] mm/vmalloc: Avoid iterating over per CPU vmap blocks twice Thomas Gleixner
2023-05-26  7:55   ` Christoph Hellwig
2023-05-27 17:38   ` Lorenzo Stoakes
2023-05-25 12:57 ` [V2 patch 3/6] mm/vmalloc: Prevent flushing dirty space over and over Thomas Gleixner
2023-05-26  7:56   ` Christoph Hellwig
2023-05-27 18:19   ` Lorenzo Stoakes [this message]
2023-05-25 12:57 ` [V2 patch 4/6] mm/vmalloc: Check free space in vmap_block lockless Thomas Gleixner
2023-05-26  7:56   ` Christoph Hellwig
2023-05-27 18:27   ` Lorenzo Stoakes
2023-05-25 12:57 ` [V2 patch 5/6] mm/vmalloc: Add missing READ/WRITE_ONCE() annotations Thomas Gleixner
2023-05-26  7:56   ` Christoph Hellwig
2023-05-27 19:06   ` Lorenzo Stoakes
2023-05-25 12:57 ` [V2 patch 6/6] mm/vmalloc: Dont purge usable blocks unnecessarily Thomas Gleixner
2023-05-27 19:28   ` Lorenzo Stoakes
2023-05-27 10:21 ` [V2 patch 0/6] mm/vmalloc: Assorted fixes and improvements Baoquan He

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=590a1dc3-ce57-4d13-a9c8-82ac5832ab99@lucifer.local \
    --to=lstoakes@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=bhe@redhat.com \
    --cc=hch@lst.de \
    --cc=linux-mm@kvack.org \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --cc=urezki@gmail.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox