From: Dave Hansen <[email protected]>
I went to go add a new RECLAIM_* mode for the zone_reclaim_mode
sysctl. Like a good kernel developer, I also went to go update the
documentation. I noticed that the bits in the documentation didn't
match the bits in the #defines.
The VM never explicitly checks the RECLAIM_ZONE bit. The bit is,
however implicitly checked when checking 'node_reclaim_mode==0'.
The RECLAIM_ZONE #define was removed in a cleanup. That, by itself
is fine.
But, when the bit was removed (bit 0) the _other_ bit locations also
got changed. That's not OK because the bit values are documented to
mean one specific thing and users surely rely on them meaning that one
thing and not changing from kernel to kernel. The end result is that
if someone had a script that did:
sysctl vm.zone_reclaim_mode=1
That script went from doing nothing to writing out pages during
node reclaim after the commit in question. That's not great.
Put the bits back the way they were and add a comment so something
like this is a bit harder to do again. Update the documentation to
make it clear that the first bit is ignored.
Signed-off-by: Dave Hansen <[email protected]>
Fixes: 648b5cf368e0 ("mm/vmscan: remove unused RECLAIM_OFF/RECLAIM_ZONE")
Cc: Ben Widawsky <[email protected]>
Cc: Alex Shi <[email protected]>
Cc: Daniel Wagner <[email protected]>
Cc: "Tobin C. Harding" <[email protected]>
Cc: Christoph Lameter <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Huang Ying <[email protected]>
Cc: Dan Williams <[email protected]>
Cc: Qian Cai <[email protected]>
Cc: Daniel Wagner <[email protected]>
Cc: [email protected]
---
b/Documentation/admin-guide/sysctl/vm.rst | 10 +++++-----
b/mm/vmscan.c | 9 +++++++--
2 files changed, 12 insertions(+), 7 deletions(-)
diff -puN Documentation/admin-guide/sysctl/vm.rst~mm-vmscan-restore-old-zone_reclaim_mode-abi Documentation/admin-guide/sysctl/vm.rst
--- a/Documentation/admin-guide/sysctl/vm.rst~mm-vmscan-restore-old-zone_reclaim_mode-abi 2020-07-01 08:22:11.354955336 -0700
+++ b/Documentation/admin-guide/sysctl/vm.rst 2020-07-01 08:22:11.360955336 -0700
@@ -948,11 +948,11 @@ that benefit from having their data cach
left disabled as the caching effect is likely to be more important than
data locality.
-zone_reclaim may be enabled if it's known that the workload is partitioned
-such that each partition fits within a NUMA node and that accessing remote
-memory would cause a measurable performance reduction. The page allocator
-will then reclaim easily reusable pages (those page cache pages that are
-currently not used) before allocating off node pages.
+Consider enabling one or more zone_reclaim mode bits if it's known that the
+workload is partitioned such that each partition fits within a NUMA node
+and that accessing remote memory would cause a measurable performance
+reduction. The page allocator will take additional actions before
+allocating off node pages.
Allowing zone reclaim to write out pages stops processes that are
writing large amounts of data from dirtying pages on other nodes. Zone
diff -puN mm/vmscan.c~mm-vmscan-restore-old-zone_reclaim_mode-abi mm/vmscan.c
--- a/mm/vmscan.c~mm-vmscan-restore-old-zone_reclaim_mode-abi 2020-07-01 08:22:11.356955336 -0700
+++ b/mm/vmscan.c 2020-07-01 08:22:11.362955336 -0700
@@ -4090,8 +4090,13 @@ module_init(kswapd_init)
*/
int node_reclaim_mode __read_mostly;
-#define RECLAIM_WRITE (1<<0) /* Writeout pages during reclaim */
-#define RECLAIM_UNMAP (1<<1) /* Unmap pages during reclaim */
+/*
+ * These bit locations are exposed in the vm.zone_reclaim_mode sysctl
+ * ABI. New bits are OK, but existing bits can never change.
+ */
+#define RECLAIM_ZONE (1<<0) /* Run shrink_inactive_list on the zone */
+#define RECLAIM_WRITE (1<<1) /* Writeout pages during reclaim */
+#define RECLAIM_UNMAP (1<<2) /* Unmap pages during reclaim */
/*
* Priority for NODE_RECLAIM. This determines the fraction of pages
_
On Wed, 1 Jul 2020, Dave Hansen wrote:
>
> From: Dave Hansen <[email protected]>
>
> I went to go add a new RECLAIM_* mode for the zone_reclaim_mode
> sysctl. Like a good kernel developer, I also went to go update the
> documentation. I noticed that the bits in the documentation didn't
> match the bits in the #defines.
>
> The VM never explicitly checks the RECLAIM_ZONE bit. The bit is,
> however implicitly checked when checking 'node_reclaim_mode==0'.
> The RECLAIM_ZONE #define was removed in a cleanup. That, by itself
> is fine.
>
> But, when the bit was removed (bit 0) the _other_ bit locations also
> got changed. That's not OK because the bit values are documented to
> mean one specific thing and users surely rely on them meaning that one
> thing and not changing from kernel to kernel. The end result is that
> if someone had a script that did:
>
> sysctl vm.zone_reclaim_mode=1
>
> That script went from doing nothing to writing out pages during
> node reclaim after the commit in question. That's not great.
>
> Put the bits back the way they were and add a comment so something
> like this is a bit harder to do again. Update the documentation to
> make it clear that the first bit is ignored.
>
> Signed-off-by: Dave Hansen <[email protected]>
> Fixes: 648b5cf368e0 ("mm/vmscan: remove unused RECLAIM_OFF/RECLAIM_ZONE")
> Cc: Ben Widawsky <[email protected]>
> Cc: Alex Shi <[email protected]>
> Cc: Daniel Wagner <[email protected]>
> Cc: "Tobin C. Harding" <[email protected]>
> Cc: Christoph Lameter <[email protected]>
> Cc: Andrew Morton <[email protected]>
> Cc: Huang Ying <[email protected]>
> Cc: Dan Williams <[email protected]>
> Cc: Qian Cai <[email protected]>
> Cc: Daniel Wagner <[email protected]>
> Cc: [email protected]
Acked-by: David Rientjes <[email protected]>
Dave Hansen <[email protected]> writes:
> From: Dave Hansen <[email protected]>
>
> I went to go add a new RECLAIM_* mode for the zone_reclaim_mode
> sysctl. Like a good kernel developer, I also went to go update the
> documentation. I noticed that the bits in the documentation didn't
> match the bits in the #defines.
>
> The VM never explicitly checks the RECLAIM_ZONE bit. The bit is,
> however implicitly checked when checking 'node_reclaim_mode==0'.
> The RECLAIM_ZONE #define was removed in a cleanup. That, by itself
> is fine.
>
> But, when the bit was removed (bit 0) the _other_ bit locations also
> got changed. That's not OK because the bit values are documented to
> mean one specific thing and users surely rely on them meaning that one
> thing and not changing from kernel to kernel. The end result is that
> if someone had a script that did:
>
> sysctl vm.zone_reclaim_mode=1
>
> That script went from doing nothing
Per my understanding, this script would have enabled node reclaim for
clean unmapped pages before commit 648b5cf368e0 ("mm/vmscan: remove
unused RECLAIM_OFF/RECLAIM_ZONE"). So we should revise the description
here?
> to writing out pages during
> node reclaim after the commit in question. That's not great.
>
> Put the bits back the way they were and add a comment so something
> like this is a bit harder to do again. Update the documentation to
> make it clear that the first bit is ignored.
>
Best Regards,
Huang, Ying
On 7/2/20 4:28 AM, Huang, Ying wrote:
>> But, when the bit was removed (bit 0) the _other_ bit locations also
>> got changed. That's not OK because the bit values are documented to
>> mean one specific thing and users surely rely on them meaning that one
>> thing and not changing from kernel to kernel. The end result is that
>> if someone had a script that did:
>>
>> sysctl vm.zone_reclaim_mode=1
>>
>> That script went from doing nothing
> Per my understanding, this script would have enabled node reclaim for
> clean unmapped pages before commit 648b5cf368e0 ("mm/vmscan: remove
> unused RECLAIM_OFF/RECLAIM_ZONE"). So we should revise the description
> here?
Yes, you're right. I updated the patch with the updated understanding
about the implicit use of the bit but didn't update the changelog. I'll
do that for v3.