All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v4] mm: swap: properly update readahead statistics in unuse_pte_range()
@ 2020-04-18  8:47 Andrea Righi
  2020-04-19 23:55   ` Huang, Ying
  0 siblings, 1 reply; 3+ messages in thread
From: Andrea Righi @ 2020-04-18  8:47 UTC (permalink / raw)
  To: Andrew Morton
  Cc: Huang Ying, Minchan Kim, Anchal Agarwal, linux-mm, linux-kernel

In unuse_pte_range() we blindly swap-in pages without checking if the
swap entry is already present in the swap cache.

By doing this, the hit/miss ratio used by the swap readahead heuristic
is not properly updated and this leads to non-optimal performance during
swapoff.

Tracing the distribution of the readahead size returned by the swap
readahead heuristic during swapoff shows that a small readahead size is
used most of the time as if we had only misses (this happens both with
cluster and vma readahead), for example:

r::swapin_nr_pages(unsigned long offset):unsigned long:$retval
        COUNT      EVENT
        36948      $retval = 8
        44151      $retval = 4
        49290      $retval = 1
        527771     $retval = 2

Checking if the swap entry is present in the swap cache, instead, allows
to properly update the readahead statistics and the heuristic behaves in
a better way during swapoff, selecting a bigger readahead size:

r::swapin_nr_pages(unsigned long offset):unsigned long:$retval
        COUNT      EVENT
        1618       $retval = 1
        4960       $retval = 2
        41315      $retval = 4
        103521     $retval = 8

In terms of swapoff performance the result is the following:

Testing environment
===================

 - Host:
   CPU: 1.8GHz Intel Core i7-8565U (quad-core, 8MB cache)
   HDD: PC401 NVMe SK hynix 512GB
   MEM: 16GB

 - Guest (kvm):
   8GB of RAM
   virtio block driver
   16GB swap file on ext4 (/swapfile)

Test case
=========
 - allocate 85% of memory
 - `systemctl hibernate` to force all the pages to be swapped-out to the
   swap file
 - resume the system
 - measure the time that swapoff takes to complete:
   # /usr/bin/time swapoff /swapfile

Result (swapoff time)
======
                  5.6 vanilla   5.6 w/ this patch
                  -----------   -----------------
cluster-readahead      22.09s              12.19s
    vma-readahead      18.20s              15.33s

Conclusion
==========

The specific use case this patch is addressing is to improve swapoff
performance in cloud environments when a VM has been hibernated, resumed
and all the memory needs to be forced back to RAM by disabling swap.

This change allows to better exploits the advantages of the readahead
heuristic during swapoff and this improvement allows to to speed up the
resume process of such VMs.

Reviewed-by: "Huang, Ying" <ying.huang@intel.com>
Signed-off-by: Andrea Righi <andrea.righi@canonical.com>
---
Changes in v4:
 - added a "conclusion" section in the patch description to better
   explain the importance of this change (no functional changes)

 mm/swapfile.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/mm/swapfile.c b/mm/swapfile.c
index 5871a2aa86a5..f8bf926c9c8f 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -1937,10 +1937,14 @@ static int unuse_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
 
 		pte_unmap(pte);
 		swap_map = &si->swap_map[offset];
-		vmf.vma = vma;
-		vmf.address = addr;
-		vmf.pmd = pmd;
-		page = swapin_readahead(entry, GFP_HIGHUSER_MOVABLE, &vmf);
+		page = lookup_swap_cache(entry, vma, addr);
+		if (!page) {
+			vmf.vma = vma;
+			vmf.address = addr;
+			vmf.pmd = pmd;
+			page = swapin_readahead(entry, GFP_HIGHUSER_MOVABLE,
+						&vmf);
+		}
 		if (!page) {
 			if (*swap_map == 0 || *swap_map == SWAP_MAP_BAD)
 				goto try_next;
-- 
2.25.1



^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH v4] mm: swap: properly update readahead statistics in unuse_pte_range()
  2020-04-18  8:47 [PATCH v4] mm: swap: properly update readahead statistics in unuse_pte_range() Andrea Righi
@ 2020-04-19 23:55   ` Huang, Ying
  0 siblings, 0 replies; 3+ messages in thread
From: Huang, Ying @ 2020-04-19 23:55 UTC (permalink / raw)
  To: Andrea Righi
  Cc: Andrew Morton, Minchan Kim, Anchal Agarwal, linux-mm, linux-kernel

Andrea Righi <andrea.righi@canonical.com> writes:

> In unuse_pte_range() we blindly swap-in pages without checking if the
> swap entry is already present in the swap cache.
>
> By doing this, the hit/miss ratio used by the swap readahead heuristic
> is not properly updated and this leads to non-optimal performance during
> swapoff.
>
> Tracing the distribution of the readahead size returned by the swap
> readahead heuristic during swapoff shows that a small readahead size is
> used most of the time as if we had only misses (this happens both with
> cluster and vma readahead), for example:
>
> r::swapin_nr_pages(unsigned long offset):unsigned long:$retval
>         COUNT      EVENT
>         36948      $retval = 8
>         44151      $retval = 4
>         49290      $retval = 1
>         527771     $retval = 2
>
> Checking if the swap entry is present in the swap cache, instead, allows
> to properly update the readahead statistics and the heuristic behaves in
> a better way during swapoff, selecting a bigger readahead size:
>
> r::swapin_nr_pages(unsigned long offset):unsigned long:$retval
>         COUNT      EVENT
>         1618       $retval = 1
>         4960       $retval = 2
>         41315      $retval = 4
>         103521     $retval = 8
>
> In terms of swapoff performance the result is the following:
>
> Testing environment
> ===================
>
>  - Host:
>    CPU: 1.8GHz Intel Core i7-8565U (quad-core, 8MB cache)
>    HDD: PC401 NVMe SK hynix 512GB
>    MEM: 16GB
>
>  - Guest (kvm):
>    8GB of RAM
>    virtio block driver
>    16GB swap file on ext4 (/swapfile)
>
> Test case
> =========
>  - allocate 85% of memory
>  - `systemctl hibernate` to force all the pages to be swapped-out to the
>    swap file
>  - resume the system
>  - measure the time that swapoff takes to complete:
>    # /usr/bin/time swapoff /swapfile
>
> Result (swapoff time)
> ======
>                   5.6 vanilla   5.6 w/ this patch
>                   -----------   -----------------
> cluster-readahead      22.09s              12.19s
>     vma-readahead      18.20s              15.33s
>
> Conclusion
> ==========
>
> The specific use case this patch is addressing is to improve swapoff
> performance in cloud environments when a VM has been hibernated, resumed
> and all the memory needs to be forced back to RAM by disabling swap.
>
> This change allows to better exploits the advantages of the readahead
> heuristic during swapoff and this improvement allows to to speed up the
> resume process of such VMs.
>
> Reviewed-by: "Huang, Ying" <ying.huang@intel.com>
> Signed-off-by: Andrea Righi <andrea.righi@canonical.com>

Looks good to me, Thanks!

Best Regards,
Huang, Ying

> ---
> Changes in v4:
>  - added a "conclusion" section in the patch description to better
>    explain the importance of this change (no functional changes)
>
>  mm/swapfile.c | 12 ++++++++----
>  1 file changed, 8 insertions(+), 4 deletions(-)
>
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> index 5871a2aa86a5..f8bf926c9c8f 100644
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c
> @@ -1937,10 +1937,14 @@ static int unuse_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
>  
>  		pte_unmap(pte);
>  		swap_map = &si->swap_map[offset];
> -		vmf.vma = vma;
> -		vmf.address = addr;
> -		vmf.pmd = pmd;
> -		page = swapin_readahead(entry, GFP_HIGHUSER_MOVABLE, &vmf);
> +		page = lookup_swap_cache(entry, vma, addr);
> +		if (!page) {
> +			vmf.vma = vma;
> +			vmf.address = addr;
> +			vmf.pmd = pmd;
> +			page = swapin_readahead(entry, GFP_HIGHUSER_MOVABLE,
> +						&vmf);
> +		}
>  		if (!page) {
>  			if (*swap_map == 0 || *swap_map == SWAP_MAP_BAD)
>  				goto try_next;

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH v4] mm: swap: properly update readahead statistics in unuse_pte_range()
@ 2020-04-19 23:55   ` Huang, Ying
  0 siblings, 0 replies; 3+ messages in thread
From: Huang, Ying @ 2020-04-19 23:55 UTC (permalink / raw)
  To: Andrea Righi
  Cc: Andrew Morton, Minchan Kim, Anchal Agarwal, linux-mm, linux-kernel

Andrea Righi <andrea.righi@canonical.com> writes:

> In unuse_pte_range() we blindly swap-in pages without checking if the
> swap entry is already present in the swap cache.
>
> By doing this, the hit/miss ratio used by the swap readahead heuristic
> is not properly updated and this leads to non-optimal performance during
> swapoff.
>
> Tracing the distribution of the readahead size returned by the swap
> readahead heuristic during swapoff shows that a small readahead size is
> used most of the time as if we had only misses (this happens both with
> cluster and vma readahead), for example:
>
> r::swapin_nr_pages(unsigned long offset):unsigned long:$retval
>         COUNT      EVENT
>         36948      $retval = 8
>         44151      $retval = 4
>         49290      $retval = 1
>         527771     $retval = 2
>
> Checking if the swap entry is present in the swap cache, instead, allows
> to properly update the readahead statistics and the heuristic behaves in
> a better way during swapoff, selecting a bigger readahead size:
>
> r::swapin_nr_pages(unsigned long offset):unsigned long:$retval
>         COUNT      EVENT
>         1618       $retval = 1
>         4960       $retval = 2
>         41315      $retval = 4
>         103521     $retval = 8
>
> In terms of swapoff performance the result is the following:
>
> Testing environment
> ===================
>
>  - Host:
>    CPU: 1.8GHz Intel Core i7-8565U (quad-core, 8MB cache)
>    HDD: PC401 NVMe SK hynix 512GB
>    MEM: 16GB
>
>  - Guest (kvm):
>    8GB of RAM
>    virtio block driver
>    16GB swap file on ext4 (/swapfile)
>
> Test case
> =========
>  - allocate 85% of memory
>  - `systemctl hibernate` to force all the pages to be swapped-out to the
>    swap file
>  - resume the system
>  - measure the time that swapoff takes to complete:
>    # /usr/bin/time swapoff /swapfile
>
> Result (swapoff time)
> ======
>                   5.6 vanilla   5.6 w/ this patch
>                   -----------   -----------------
> cluster-readahead      22.09s              12.19s
>     vma-readahead      18.20s              15.33s
>
> Conclusion
> ==========
>
> The specific use case this patch is addressing is to improve swapoff
> performance in cloud environments when a VM has been hibernated, resumed
> and all the memory needs to be forced back to RAM by disabling swap.
>
> This change allows to better exploits the advantages of the readahead
> heuristic during swapoff and this improvement allows to to speed up the
> resume process of such VMs.
>
> Reviewed-by: "Huang, Ying" <ying.huang@intel.com>
> Signed-off-by: Andrea Righi <andrea.righi@canonical.com>

Looks good to me, Thanks!

Best Regards,
Huang, Ying

> ---
> Changes in v4:
>  - added a "conclusion" section in the patch description to better
>    explain the importance of this change (no functional changes)
>
>  mm/swapfile.c | 12 ++++++++----
>  1 file changed, 8 insertions(+), 4 deletions(-)
>
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> index 5871a2aa86a5..f8bf926c9c8f 100644
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c
> @@ -1937,10 +1937,14 @@ static int unuse_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
>  
>  		pte_unmap(pte);
>  		swap_map = &si->swap_map[offset];
> -		vmf.vma = vma;
> -		vmf.address = addr;
> -		vmf.pmd = pmd;
> -		page = swapin_readahead(entry, GFP_HIGHUSER_MOVABLE, &vmf);
> +		page = lookup_swap_cache(entry, vma, addr);
> +		if (!page) {
> +			vmf.vma = vma;
> +			vmf.address = addr;
> +			vmf.pmd = pmd;
> +			page = swapin_readahead(entry, GFP_HIGHUSER_MOVABLE,
> +						&vmf);
> +		}
>  		if (!page) {
>  			if (*swap_map == 0 || *swap_map == SWAP_MAP_BAD)
>  				goto try_next;


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2020-04-19 23:55 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-04-18  8:47 [PATCH v4] mm: swap: properly update readahead statistics in unuse_pte_range() Andrea Righi
2020-04-19 23:55 ` Huang, Ying
2020-04-19 23:55   ` Huang, Ying

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.