On Fri, 5 Apr 2024 15:08:56 +0900 Honggyu Kim wrote:
> This patch adds "damon_migrate_{hot,cold}" under node specific vmstat
> counters at the following location.
>
> /sys/devices/system/node/node*/vmstat
>
> The counted values are accumulcated to the global vm
This patch adds "damon_migrate_{hot,cold}" under node specific vmstat
counters at the following location.
/sys/devices/system/node/node*/vmstat
The counted values are accumulcated to the global vmstat so it also
introduces the same counter at /proc/vmstat as well.
Signed-off-by: H
Before the new slab memory controller with per object byte charging,
charging and vmstat data update happen only when new slab pages are
allocated or freed. Now they are done with every kmem_cache_alloc()
and kmem_cache_free(). This causes additional overhead for workloads
that generate a lot of
On 4/19/21 12:38 PM, Johannes Weiner wrote:
On Sun, Apr 18, 2021 at 08:00:29PM -0400, Waiman Long wrote:
Before the new slab memory controller with per object byte charging,
charging and vmstat data update happen only when new slab pages are
allocated or freed. Now they are done with every
On Sun, Apr 18, 2021 at 08:00:29PM -0400, Waiman Long wrote:
> Before the new slab memory controller with per object byte charging,
> charging and vmstat data update happen only when new slab pages are
> allocated or freed. Now they are done with every kmem_cache_alloc()
> and kme
__count_numa_event is small enough to be treated similarly to
__count_vm_event so inline it.
Signed-off-by: Mel Gorman
Acked-by: Vlastimil Babka
---
include/linux/vmstat.h | 10 +-
mm/vmstat.c| 9 -
2 files changed, 9 insertions(+), 10 deletions(-)
diff --git a/inc
maintain perfect accuracy similar to what is required for a vmstat like
NR_FREE_PAGES. There even is a sysctl vm.numa_stat to allow userspace to
turn off the collection of NUMA statistics like NUMA_HIT.
This patch converts NUMA_HIT and friends to be NUMA events with similar
accuracy to VM events
Before the new slab memory controller with per object byte charging,
charging and vmstat data update happen only when new slab pages are
allocated or freed. Now they are done with every kmem_cache_alloc()
and kmem_cache_free(). This causes additional overhead for workloads
that generate a lot of
On Thu, Apr 15, 2021 at 01:08:29PM -0400, Waiman Long wrote:
> On 4/15/21 12:50 PM, Johannes Weiner wrote:
> > On Tue, Apr 13, 2021 at 09:20:25PM -0400, Waiman Long wrote:
> > > Before the new slab memory controller with per object byte charging,
> > > charging and vms
On 4/15/21 12:50 PM, Johannes Weiner wrote:
On Tue, Apr 13, 2021 at 09:20:25PM -0400, Waiman Long wrote:
Before the new slab memory controller with per object byte charging,
charging and vmstat data update happen only when new slab pages are
allocated or freed. Now they are done with every
On Tue, Apr 13, 2021 at 09:20:25PM -0400, Waiman Long wrote:
> Before the new slab memory controller with per object byte charging,
> charging and vmstat data update happen only when new slab pages are
> allocated or freed. Now they are done with every kmem_cache_alloc()
> and kme
s there's no such thing for
> >> the numa
> >> counters.
> >>
> >
> > That is a problem I missed. Even if zonestats was preserved on
> > hot-remove, fold_vm_zone_numa_events would not be reading the CPU so
> > hotplug events jump all over
On Wed, Apr 14, 2021 at 06:26:25PM +0200, Vlastimil Babka wrote:
> On 4/14/21 6:20 PM, Vlastimil Babka wrote:
> > On 4/14/21 3:39 PM, Mel Gorman wrote:
> >> __count_numa_event is small enough to be treated similarly to
> >> __count_vm_event so inline it.
> >>
> >> Signed-off-by: Mel Gorman
> >
>
On Tue, Apr 13, 2021 at 09:20:25PM -0400, Waiman Long wrote:
> Before the new slab memory controller with per object byte charging,
> charging and vmstat data update happen only when new slab pages are
> allocated or freed. Now they are done with every kmem_cache_alloc()
> and kme
On 4/14/21 6:20 PM, Vlastimil Babka wrote:
> On 4/14/21 3:39 PM, Mel Gorman wrote:
>> __count_numa_event is small enough to be treated similarly to
>> __count_vm_event so inline it.
>>
>> Signed-off-by: Mel Gorman
>
> Acked-by: Vlastimil Babka
>
>> ---
>> include/linux/vmstat.h | 9 +
On 4/14/21 3:39 PM, Mel Gorman wrote:
> __count_numa_event is small enough to be treated similarly to
> __count_vm_event so inline it.
>
> Signed-off-by: Mel Gorman
Acked-by: Vlastimil Babka
> ---
> include/linux/vmstat.h | 9 +
> mm/vmstat.c| 9 -
> 2 files change
ers.
>>
>
> That is a problem I missed. Even if zonestats was preserved on
> hot-remove, fold_vm_zone_numa_events would not be reading the CPU so
> hotplug events jump all over the place.
>
> So some periodic folding is necessary. I would still prefer not to do it
> by t
The counters are used by userspace to get a general overview
> > of a workloads NUMA behaviour but the page allocator incurs a high cost to
> > maintain perfect accuracy similar to what is required for a vmstat like
> > NR_FREE_PAGES. There even is a sysctl vm.numa_stat to allow u
__count_numa_event is small enough to be treated similarly to
__count_vm_event so inline it.
Signed-off-by: Mel Gorman
---
include/linux/vmstat.h | 9 +
mm/vmstat.c| 9 -
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/include/linux/vmstat.h b/include/l
maintain perfect accuracy similar to what is required for a vmstat like
NR_FREE_PAGES. There even is a sysctl vm.numa_stat to allow userspace to
turn off the collection of NUMA statistics like NUMA_HIT.
This patch converts NUMA_HIT and friends to be NUMA events with similar
accuracy to VM events
MA behaviour but the page allocator incurs a high cost to
> maintain perfect accuracy similar to what is required for a vmstat like
> NR_FREE_PAGES. There even is a sysctl vm.numa_stat to allow userspace to
> turn off the collection of NUMA statistics like NUMA_HIT.
>
> This patch conve
Before the new slab memory controller with per object byte charging,
charging and vmstat data update happen only when new slab pages are
allocated or freed. Now they are done with every kmem_cache_alloc()
and kmem_cache_free(). This causes additional overhead for workloads
that generate a lot of
On Mon, Apr 12, 2021 at 3:55 PM Waiman Long wrote:
>
> Before the new slab memory controller with per object byte charging,
> charging and vmstat data update happen only when new slab pages are
> allocated or freed. Now they are done with every kmem_cache_alloc()
> and kmem_cac
Before the new slab memory controller with per object byte charging,
charging and vmstat data update happen only when new slab pages are
allocated or freed. Now they are done with every kmem_cache_alloc()
and kmem_cache_free(). This causes additional overhead for workloads
that generate a lot of
On 4/12/21 2:22 PM, Roman Gushchin wrote:
On Fri, Apr 09, 2021 at 07:18:40PM -0400, Waiman Long wrote:
Before the new slab memory controller with per object byte charging,
charging and vmstat data update happen only when new slab pages are
allocated or freed. Now they are done with every
On Fri, Apr 09, 2021 at 07:18:40PM -0400, Waiman Long wrote:
> Before the new slab memory controller with per object byte charging,
> charging and vmstat data update happen only when new slab pages are
> allocated or freed. Now they are done with every kmem_cache_alloc()
> and kme
Before the new slab memory controller with per object byte charging,
charging and vmstat data update happen only when new slab pages are
allocated or freed. Now they are done with every kmem_cache_alloc()
and kmem_cache_free(). This causes additional overhead for workloads
that generate a lot of
Allow page counters to be more readily modified by callers which have
a folio. Name these wrappers with 'stat' instead of 'state' as requested
by Linus here:
https://lore.kernel.org/linux-mm/CAHk-=wj847sudr-kt+46ft3+xffgiwpgthvm7djwgdi4cvr...@mail.gmail.com/
Signed-off-by: Matthew Wilcox (Oracle)
__count_numa_event is small enough to be treated similarly to
__count_vm_event so inline it.
Signed-off-by: Mel Gorman
---
include/linux/vmstat.h | 9 +
mm/vmstat.c| 9 -
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/include/linux/vmstat.h b/include/l
maintain perfect accuracy similar to what is required for a vmstat like
NR_FREE_PAGES. There even is a sysctl vm.numa_stat to allow userspace to
turn off the collection of NUMA statistics like NUMA_HIT.
This patch converts NUMA_HIT and friends to be NUMA events with similar
accuracy to VM events
On Wed, Mar 31, 2021 at 07:47:04PM +0100, Matthew Wilcox (Oracle) wrote:
> Allow page counters to be more readily modified by callers which have
> a folio. Name these wrappers with 'stat' instead of 'state' as requested
> by Linus here:
Looks good,
Reviewed-by: Christoph Hellwig
Allow page counters to be more readily modified by callers which have
a folio. Name these wrappers with 'stat' instead of 'state' as requested
by Linus here:
https://lore.kernel.org/linux-mm/CAHk-=wj847sudr-kt+46ft3+xffgiwpgthvm7djwgdi4cvr...@mail.gmail.com/
Signed-off-by: Matthew Wilcox (Oracle)
__count_numa_event is small enough to be treated similarly to
__count_vm_event so inline it.
Signed-off-by: Mel Gorman
---
include/linux/vmstat.h | 9 +
mm/vmstat.c| 9 -
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/include/linux/vmstat.h b/include/l
high cost to maintain perfect accuracy similar to what is required for
a vmstat like NR_FREE_PAGES. There even is a sysctl vm.numa_stat to allow
userspace to turn off the collection of NUMA statistics like NUMA_HIT.
This patch converts NUMA_HIT and friends to be NUMA events with similar
accuracy
Allow page counters to be more readily modified by callers which have
a folio. Name these wrappers with 'stat' instead of 'state' as requested
by Linus here:
https://lore.kernel.org/linux-mm/CAHk-=wj847sudr-kt+46ft3+xffgiwpgthvm7djwgdi4cvr...@mail.gmail.com/
Signed-off-by: Matthew Wilcox (Oracle)
> +static inline
> +void __inc_zone_folio_stat(struct folio *folio, enum zone_stat_item item)
This prototype style is weird and doesn't follow either of the preffered
styles..
> static inline void __mod_zone_freepage_state(struct zone *zone, int nr_pages,
>
On Sun, 14 Mar 2021 04:11:55 + Matthew Wilcox wrote:
> On Sat, Mar 13, 2021 at 12:37:07PM -0800, Andrew Morton wrote:
> > On Fri, 5 Mar 2021 04:18:39 + "Matthew Wilcox (Oracle)"
> > wrote:
> >
> > > Allow page counters to be more readily modified by callers which have
> > > a folio.
On Sat, Mar 13, 2021 at 12:37:07PM -0800, Andrew Morton wrote:
> On Fri, 5 Mar 2021 04:18:39 + "Matthew Wilcox (Oracle)"
> wrote:
>
> > Allow page counters to be more readily modified by callers which have
> > a folio. Name these wrappers with 'stat' instead of 'state' as requested
> > by
On Fri, 5 Mar 2021 04:18:39 + "Matthew Wilcox (Oracle)"
wrote:
> Allow page counters to be more readily modified by callers which have
> a folio. Name these wrappers with 'stat' instead of 'state' as requested
> by Linus here:
> https://lore.kernel.org/linux-mm/CAHk-=wj847sudr-kt+46ft3+xff
Allow page counters to be more readily modified by callers which have
a folio. Name these wrappers with 'stat' instead of 'state' as requested
by Linus here:
https://lore.kernel.org/linux-mm/CAHk-=wj847sudr-kt+46ft3+xffgiwpgthvm7djwgdi4cvr...@mail.gmail.com/
Signed-off-by: Matthew Wilcox (Oracle)
On 3/2/21 10:33, Minchan Kim wrote:
Since CMA is used more widely, it's worth to have CMA
allocation statistics into vmstat. With it, we could
know how agressively system uses cma allocation and
how often it fails.
Signed-off-by: Minchan Kim
---
* from v1 -
https://lore.kernel.org/lin
Since CMA is used more widely, it's worth to have CMA
allocation statistics into vmstat. With it, we could
know how agressively system uses cma allocation and
how often it fails.
Signed-off-by: Minchan Kim
---
* from v1 -
https://lore.kernel.org/linux-mm/20210217170025.512704-1
On Mon, Mar 01, 2021 at 04:17:39PM -0500, Zi Yan wrote:
> On 28 Jan 2021, at 2:03, Matthew Wilcox (Oracle) wrote:
> > Allow page counters to be more readily modified by callers which have
> > a folio. Name these wrappers with 'stat' instead of 'state' as requested
>
> Shouldn’t we change the stats
On 28 Jan 2021, at 2:03, Matthew Wilcox (Oracle) wrote:
> Allow page counters to be more readily modified by callers which have
> a folio. Name these wrappers with 'stat' instead of 'state' as requested
> by Linus here:
> https://lore.kernel.org/linux-mm/CAHk-=wj847sudr-kt+46ft3+xffgiwpgthvm7djwg
, so we would have
> > > > > more time to find a solution, acceptable for all?
> > > >
> > > > No probs. I already had a big red asterisk on it ;)
> > >
> > > I've a suspicion that Andrew might be tiring of his big red asterisk,
> >
> No probs. I already had a big red asterisk on it ;)
> >
> > I've a suspicion that Andrew might be tiring of his big red asterisk,
> > and wanting to unload
> > mm-vmstat-fix-proc-sys-vm-stat_refresh-generating-false-warnings.patch
> > mm-vmstat-fix-pr
haven't reached a consensus here.
> > > Can, you, please, not merge this patch into 5.9, so we would have
> > > more time to find a solution, acceptable for all?
> >
> > No probs. I already had a big red asterisk on it ;)
>
> I've a suspicion that Andr
we would have
> > more time to find a solution, acceptable for all?
>
> No probs. I already had a big red asterisk on it ;)
I've a suspicion that Andrew might be tiring of his big red asterisk,
and wanting to unload
mm-vmstat-fix-proc-sys-vm-stat_refresh-generating-false-
On Tue, Feb 09, 2021 at 11:33:04AM -0500, Johannes Weiner
wrote:
> ---
> tools/testing/selftests/cgroup/test_kmem.c | 22 ++
> 1 file changed, 14 insertions(+), 8 deletions(-)
Reviewed-by: Michal Koutný
signature.asc
Description: Digital signature
On Wed, Feb 17, 2021 at 12:57:25PM -0800, John Hubbard wrote:
> On 2/17/21 9:00 AM, Minchan Kim wrote:
> > Since CMA is used more widely, it's worth to have CMA
> > allocation statistics into vmstat. With it, we could
> > know how agressively system uses cma allocation
On 2/17/21 9:00 AM, Minchan Kim wrote:
Since CMA is used more widely, it's worth to have CMA
allocation statistics into vmstat. With it, we could
know how agressively system uses cma allocation and
how often it fails.
Signed-off-by: Minchan Kim
---
include/linux/vm_event_item.h | 3 +++
Since CMA is used more widely, it's worth to have CMA
allocation statistics into vmstat. With it, we could
know how agressively system uses cma allocation and
how often it fails.
Signed-off-by: Minchan Kim
---
include/linux/vm_event_item.h | 3 +++
mm/cma.c
On Tue, Feb 09, 2021 at 11:33:04AM -0500, Johannes Weiner wrote:
> With memcg having switched to rstat, memory.stat output is precise.
> Update the cgroup selftest to reflect the expectations and error
> tolerances of the new implementation.
>
> Also add newly tracked types of memory to the memory
tests/cgroup/test_kmem.c
index 0941aa16157e..22b31ebb3513 100644
--- a/tools/testing/selftests/cgroup/test_kmem.c
+++ b/tools/testing/selftests/cgroup/test_kmem.c
@@ -19,12 +19,12 @@
/*
- * Memory cgroup charging and vmstat data aggregation is performed using
- * percpu batches 32 pages big (lo
On Fri, Feb 5, 2021 at 10:28 AM Johannes Weiner wrote:
>
> With memcg having switched to rstat, memory.stat output is precise.
> Update the cgroup selftest to reflect the expectations and error
> tolerances of the new implementation.
>
> Also add newly tracked types of memory to the memory.stat si
index 0941aa16157e..22b31ebb3513 100644
--- a/tools/testing/selftests/cgroup/test_kmem.c
+++ b/tools/testing/selftests/cgroup/test_kmem.c
@@ -19,12 +19,12 @@
/*
- * Memory cgroup charging and vmstat data aggregation is performed using
- * percpu batches 32 pages big (look at MEMCG_CHARGE_BATCH). S
Byte-accounted items are used for slab object accounting at the cgroup
level, because the objects in a slab page can belong to different
cgroups. At the global level these items always change in multiples of
whole slab pages. The vmstat code exploits this and stores these items
as pages internally
On NOHZ, the periodic vmstat flushers on each CPU can go to sleep and
won't wake up until stat changes are detected in the per-cpu deltas of
the zone vmstat counters.
In commit 75ef71840539 ("mm, vmstat: add infrastructure for per-node
vmstats") per-node counters were introduced, a
Allow page counters to be more readily modified by callers which have
a folio. Name these wrappers with 'stat' instead of 'state' as requested
by Linus here:
https://lore.kernel.org/linux-mm/CAHk-=wj847sudr-kt+46ft3+xffgiwpgthvm7djwgdi4cvr...@mail.gmail.com/
Signed-off-by: Matthew Wilcox (Oracle)
On Tue 19-01-21 11:58:42, Johannes Weiner wrote:
> On Tue, Jan 19, 2021 at 01:27:44PM +0800, Muchun Song wrote:
> > The current memcg will be freed soon, so updating it's vmstat and
> > vmevent values is pointless. Just skip updating it.
> >
> > Signed-off-by: Much
On Tue, Jan 19, 2021 at 01:27:44PM +0800, Muchun Song wrote:
> The current memcg will be freed soon, so updating it's vmstat and
> vmevent values is pointless. Just skip updating it.
>
> Signed-off-by: Muchun Song
Oof, that's pretty subtle! Somebody trying to refact
The current memcg will be freed soon, so updating it's vmstat and
vmevent values is pointless. Just skip updating it.
Signed-off-by: Muchun Song
---
mm/memcontrol.c | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index c0a9076
Allow page counters to be more readily modified by callers which have
a folio. Name these wrappers with 'stat' instead of 'state' as requested
by Linus here:
https://lore.kernel.org/linux-mm/CAHk-=wj847sudr-kt+46ft3+xffgiwpgthvm7djwgdi4cvr...@mail.gmail.com/
Signed-off-by: Matthew Wilcox (Oracle)
This patch series is aimed to convert all THP vmstat counters to pages.
The unit of some vmstat counters are pages, some are bytes, some are
HPAGE_PMD_NR, and some are KiB. When we want to expose these vmstat
counters to the userspace, we have to know the unit of the vmstat counters
is which one
This patch series is aimed to convert all THP vmstat counters to pages.
The unit of some vmstat counters are pages, some are bytes, some are
HPAGE_PMD_NR, and some are KiB. When we want to expose these vmstat
counters to the userspace, we have to know the unit of the vmstat counters
is which one
Hi,
This patch series is aimed to convert all THP vmstat counters to pages.
The unit of some vmstat counters are pages, some are bytes, some are
HPAGE_PMD_NR, and some are KiB. When we want to expose these vmstat
counters to the userspace, we have to know the unit of the vmstat counters
is which
Hi,
This patch series is aimed to convert all THP vmstat counters to pages.
The unit of some vmstat counters are pages, some are bytes, some are
HPAGE_PMD_NR, and some are KiB. When we want to expose these vmstat
counters to the userspace, we have to know the unit of the vmstat counters
is which
On Tue, Dec 8, 2020 at 3:46 AM Roman Gushchin wrote:
>
> On Sun, Dec 06, 2020 at 06:14:48PM +0800, Muchun Song wrote:
> > the global and per-node counters are stored in pages, however memcg
> > and lruvec counters are stored in bytes. This scheme looks weird.
> > So
On Tue, Dec 8, 2020 at 4:33 AM Hugh Dickins wrote:
>
> On Mon, 7 Dec 2020, Roman Gushchin wrote:
> > On Mon, Dec 07, 2020 at 04:02:54PM +0100, Michal Hocko wrote:
> > >
> > > As I've said the THP accounting change makes more sense to me because it
> > > allows future changes which are already unde
On Mon, Dec 7, 2020 at 11:02 PM Michal Hocko wrote:
>
> On Mon 07-12-20 22:52:30, Muchun Song wrote:
> > On Mon, Dec 7, 2020 at 9:00 PM Michal Hocko wrote:
> > >
> > > On Sun 06-12-20 18:14:39, Muchun Song wrote:
> > > > Hi,
> > > >
> &
ng wrote:
> >>>> Hi,
> >>>>
> >>>> This patch series is aimed to convert all THP vmstat counters to pages
> >>>> and some KiB vmstat counters to bytes.
> >>>>
> >>>> The unit of some vmstat counters are pages, s
On Mon, 7 Dec 2020, Roman Gushchin wrote:
> On Mon, Dec 07, 2020 at 04:02:54PM +0100, Michal Hocko wrote:
> >
> > As I've said the THP accounting change makes more sense to me because it
> > allows future changes which are already undergoing so there is more
> > merit in those.
>
> +1
> And this
On Mon, Dec 07, 2020 at 11:46:22AM -0800, Roman Gushchin wrote:
> On Sun, Dec 06, 2020 at 06:14:48PM +0800, Muchun Song wrote:
> > the global and per-node counters are stored in pages, however memcg
> > and lruvec counters are stored in bytes. This scheme looks weird.
> >
t; > > > This patch series is aimed to convert all THP vmstat counters to pages
> > > > and some KiB vmstat counters to bytes.
> > > >
> > > > The unit of some vmstat counters are pages, some are bytes, some are
> > > > HPAGE_PMD_NR, and some
On Sun, Dec 06, 2020 at 06:14:48PM +0800, Muchun Song wrote:
> the global and per-node counters are stored in pages, however memcg
> and lruvec counters are stored in bytes. This scheme looks weird.
> So convert all vmstat slab counters to bytes.
There is a reason for this weird schem
On 12/7/20 7:02 AM, Michal Hocko wrote:
> On Mon 07-12-20 22:52:30, Muchun Song wrote:
>> On Mon, Dec 7, 2020 at 9:00 PM Michal Hocko wrote:
>>>
>>> On Sun 06-12-20 18:14:39, Muchun Song wrote:
>>>> Hi,
>>>>
>>>> This patch series is
On Mon 07-12-20 22:52:30, Muchun Song wrote:
> On Mon, Dec 7, 2020 at 9:00 PM Michal Hocko wrote:
> >
> > On Sun 06-12-20 18:14:39, Muchun Song wrote:
> > > Hi,
> > >
> > > This patch series is aimed to convert all THP vmstat counters to pages
&
On Mon, Dec 7, 2020 at 9:00 PM Michal Hocko wrote:
>
> On Sun 06-12-20 18:14:39, Muchun Song wrote:
> > Hi,
> >
> > This patch series is aimed to convert all THP vmstat counters to pages
> > and some KiB vmstat counters to bytes.
> >
> > The unit of some
On Sun 06-12-20 18:14:39, Muchun Song wrote:
> Hi,
>
> This patch series is aimed to convert all THP vmstat counters to pages
> and some KiB vmstat counters to bytes.
>
> The unit of some vmstat counters are pages, some are bytes, some are
> HPAGE_PMD_NR, and some are KiB. W
Some vmstat counters are being accounted in bytes not pages, so the
stat_threshold should also scale to bytes.
The vmstat counters are already long type for memcg (can reference
to struct lruvec_stat). For the global per-node vmstat counters
also can scale to long. But the maximum vmstat
the global and per-node counters are stored in pages, however memcg
and lruvec counters are stored in bytes. This scheme looks weird.
So convert all vmstat slab counters to bytes.
Signed-off-by: Muchun Song
---
include/linux/vmstat.h | 17 ++---
mm/vmstat.c| 21
Hi,
This patch series is aimed to convert all THP vmstat counters to pages
and some KiB vmstat counters to bytes.
The unit of some vmstat counters are pages, some are bytes, some are
HPAGE_PMD_NR, and some are KiB. When we want to expose these vmstat
counters to the userspace, we have to know
I am very sorry that I have hit 'git send-email *' in a directory
containing both v1 and v2 patchs. Please ignore this. I will
resend this version. Very sorry for the noise. Thanks.
On Sun, Dec 6, 2020 at 4:25 PM Muchun Song wrote:
>
> Hi,
>
> This patch series is aimed to
the global and per-node counters are stored in pages, however memcg
and lruvec counters are stored in bytes. This scheme looks weird.
So convert all vmstat slab counters to bytes.
Signed-off-by: Muchun Song
---
include/linux/vmstat.h | 17 ++---
mm/vmstat.c| 21
Some vmstat counters are being accounted in bytes not pages, so the
stat_threshold should also scale to bytes.
The vmstat counters are already long type for memcg (can reference
to struct lruvec_stat). For the global per-node vmstat counters
also can scale to long. But the maximum vmstat
Hi,
This patch series is aimed to convert all THP vmstat counters to pages
and some KiB vmstat counters to bytes.
The unit of some vmstat counters are pages, some are bytes, some are
HPAGE_PMD_NR, and some are KiB. When we want to expose these vmstat
counters to the userspace, we have to know
The kernel stack is being accounted in KiB not page, so the
stat_threshold should also adjust to byte.
Signed-off-by: Muchun Song
---
mm/vmstat.c | 4
1 file changed, 4 insertions(+)
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 8d77ee426e22..f7857a7052e4 100644
--- a/mm/vmstat.c
+++ b/mm/v
The kernel stack is being accounted in KiB not page, so the
stat_threshold should also adjust to byte.
Signed-off-by: Muchun Song
---
mm/vmstat.c | 4
1 file changed, 4 insertions(+)
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 8d77ee426e22..f7857a7052e4 100644
--- a/mm/vmstat.c
+++ b/mm/v
Hi,
This patch series is aimed to convert all THP vmstat counters to pages
and the kernel stack vmstat counter to bytes.
The unit of some vmstat counters are pages, the unit of some vmstat counters
are bytes, the unit of some vmstat counters are HPAGE_PMD_NR, and the unit
of some vmstat counters
On Mon, Nov 23 2020 at 17:56, Alex Belits wrote:
> This function checks to see if a vmstat worker is not running,
> and the vmstat diffs don't require an update. The function is
> called from the task-isolation code to see if we need to
> actually do some work to quiet vmstat.
A
the project name you are using but the affected code
belongs to the memory management subsystem and if you run
git log mm/vmstat.c
you might get a hint what the proper prefix is, i.e. 'mm/vmstat: '
> In commit f01f17d3705b ("mm, vmstat: make quiet_vmstat lighter")
>
From: Chris Metcalf
This function checks to see if a vmstat worker is not running,
and the vmstat diffs don't require an update. The function is
called from the task-isolation code to see if we need to
actually do some work to quiet vmstat.
Signed-off-by: Chris Metcalf
Signed-off-by:
From: Chris Metcalf
In commit f01f17d3705b ("mm, vmstat: make quiet_vmstat lighter")
the quiet_vmstat() function became asynchronous, in the sense that
the vmstat work was still scheduled to run on the core when the
function returned. For task isolation, we need a synchronous
vers
在 2020/11/19 上午4:46, Jonathan Corbet 写道:
> On Mon, 16 Nov 2020 17:51:22 +0800
> Alex Shi wrote:
>
>> Commit 5647bc293ab1 ("mm: compaction: Move migration fail/success
>> stats to migrate.c"), removed 3 items in /proc/vmstat. but the docs
>> still has
On Mon, 16 Nov 2020 17:51:22 +0800
Alex Shi wrote:
> Commit 5647bc293ab1 ("mm: compaction: Move migration fail/success
> stats to migrate.c"), removed 3 items in /proc/vmstat. but the docs
> still has their explanation. let's remove them.
>
> "compact_bl
On 16 Nov 2020, at 4:51, Alex Shi wrote:
> Commit 5647bc293ab1 ("mm: compaction: Move migration fail/success
> stats to migrate.c"), removed 3 items in /proc/vmstat. but the docs
> still has their explanation. let's remove them.
>
> "compact_bl
Commit 5647bc293ab1 ("mm: compaction: Move migration fail/success
stats to migrate.c"), removed 3 items in /proc/vmstat. but the docs
still has their explanation. let's remove them.
"compact_blocks_moved",
"compact_pages_moved",
"compact_pagemigrate_failed&
:49, Anshuman Khandual wrote:
> >>>> Add following new vmstat events which will track HugeTLB page migration.
> >>>>
> >>>> 1. HUGETLB_MIGRATION_SUCCESS
> >>>> 2. HUGETLB_MIGRATION_FAILURE
> >>>>
> >>>> It fo
On 10/05/2020 11:35 AM, Michal Hocko wrote:
> On Mon 05-10-20 07:59:12, Anshuman Khandual wrote:
>>
>>
>> On 10/02/2020 05:34 PM, Michal Hocko wrote:
>>> On Wed 30-09-20 11:30:49, Anshuman Khandual wrote:
>>>> Add following new vmstat events which will
On Mon 05-10-20 07:59:12, Anshuman Khandual wrote:
>
>
> On 10/02/2020 05:34 PM, Michal Hocko wrote:
> > On Wed 30-09-20 11:30:49, Anshuman Khandual wrote:
> >> Add following new vmstat events which will track HugeTLB page migration.
> >>
> &g
1 - 100 of 957 matches
Mail list logo
