Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed, Jan 27, 2016 at 03:13:12PM -0800, David Rientjes wrote: > On Thu, 26 Nov 2015, Joonsoo Kim wrote: > > > I understand design decision, but, it is better to get value as much > > as accurate if there is no performance problem. My patch would not > > cause much performance degradation

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Thu, 26 Nov 2015, Joonsoo Kim wrote: > I understand design decision, but, it is better to get value as much > as accurate if there is no performance problem. My patch would not > cause much performance degradation because it is just adding one > this_cpu_read(). > > Consider about following

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed, Jan 27, 2016 at 03:13:12PM -0800, David Rientjes wrote: > On Thu, 26 Nov 2015, Joonsoo Kim wrote: > > > I understand design decision, but, it is better to get value as much > > as accurate if there is no performance problem. My patch would not > > cause much performance degradation

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Thu, 26 Nov 2015, Joonsoo Kim wrote: > I understand design decision, but, it is better to get value as much > as accurate if there is no performance problem. My patch would not > cause much performance degradation because it is just adding one > this_cpu_read(). > > Consider about following

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Thu, Nov 26, 2015 at 10:52:52AM +0900, Joonsoo Kim wrote: > On Wed, Nov 25, 2015 at 10:04:44AM -0600, Christoph Lameter wrote: > > > Although vmstat values aren't designed for accuracy, these are already > > > used by some sensitive places so it is better to be more accurate. > > > > The

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Thu, Nov 26, 2015 at 10:52:52AM +0900, Joonsoo Kim wrote: > On Wed, Nov 25, 2015 at 10:04:44AM -0600, Christoph Lameter wrote: > > > Although vmstat values aren't designed for accuracy, these are already > > > used by some sensitive places so it is better to be more accurate. > > > > The

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Thu 26-11-15 10:56:12, Joonsoo Kim wrote: > On Wed, Nov 25, 2015 at 01:00:22PM +0100, Michal Hocko wrote: > > On Tue 24-11-15 15:22:03, Joonsoo Kim wrote: > > > When I tested compaction in low memory condition, I found that > > > my benchmark is stuck in congestion_wait() at

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Thu 26-11-15 10:56:12, Joonsoo Kim wrote: > On Wed, Nov 25, 2015 at 01:00:22PM +0100, Michal Hocko wrote: > > On Tue 24-11-15 15:22:03, Joonsoo Kim wrote: > > > When I tested compaction in low memory condition, I found that > > > my benchmark is stuck in congestion_wait() at

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Thu, Nov 26, 2015 at 7:26 AM, Joonsoo Kim wrote: > On Wed, Nov 25, 2015 at 01:00:22PM +0100, Michal Hocko wrote: >> On Tue 24-11-15 15:22:03, Joonsoo Kim wrote: >> > When I tested compaction in low memory condition, I found that >> > my benchmark is stuck in congestion_wait() at

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed, Nov 25, 2015 at 01:00:22PM +0100, Michal Hocko wrote: > On Tue 24-11-15 15:22:03, Joonsoo Kim wrote: > > When I tested compaction in low memory condition, I found that > > my benchmark is stuck in congestion_wait() at shrink_inactive_list(). > > This stuck last for 1 sec and after then it

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed, Nov 25, 2015 at 10:04:44AM -0600, Christoph Lameter wrote: > > Although vmstat values aren't designed for accuracy, these are already > > used by some sensitive places so it is better to be more accurate. > > The design is to sacrifice accuracy and the time the updates occur for >

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed, 25 Nov 2015, Michal Hocko wrote: > > Simply remove the counter from the vmstat handling and do it differently > > then. > > We definitely do not want yet another set of counters. vmstat counters > are not only to be exported into the userspace. We have in kernel users > as well. I do

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed 25-11-15 10:04:44, Christoph Lameter wrote: > On Wed, 25 Nov 2015, Joonsoo Kim wrote: > > > I think that maintaining duplicate counter to guarantee accuracy isn't > > reasonable solution. It would cause more overhead to the system. > > Simply remove the counter from the vmstat handling

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed, 25 Nov 2015, Joonsoo Kim wrote: > I think that maintaining duplicate counter to guarantee accuracy isn't > reasonable solution. It would cause more overhead to the system. Simply remove the counter from the vmstat handling and do it differently then. > Although vmstat values aren't

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed 25-11-15 14:43:38, Vlastimil Babka wrote: > On 11/25/2015 01:00 PM, Michal Hocko wrote: > > On Tue 24-11-15 15:22:03, Joonsoo Kim wrote: > >> When I tested compaction in low memory condition, I found that > >> my benchmark is stuck in congestion_wait() at shrink_inactive_list(). > >> This

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On 11/25/2015 01:00 PM, Michal Hocko wrote: > On Tue 24-11-15 15:22:03, Joonsoo Kim wrote: >> When I tested compaction in low memory condition, I found that >> my benchmark is stuck in congestion_wait() at shrink_inactive_list(). >> This stuck last for 1 sec and after then it can escape. More

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Tue 24-11-15 15:22:03, Joonsoo Kim wrote: > When I tested compaction in low memory condition, I found that > my benchmark is stuck in congestion_wait() at shrink_inactive_list(). > This stuck last for 1 sec and after then it can escape. More investigation > shows that it is due to stale vmstat

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed, 25 Nov 2015, Joonsoo Kim wrote: > I think that maintaining duplicate counter to guarantee accuracy isn't > reasonable solution. It would cause more overhead to the system. Simply remove the counter from the vmstat handling and do it differently then. > Although vmstat values aren't

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed, Nov 25, 2015 at 10:04:44AM -0600, Christoph Lameter wrote: > > Although vmstat values aren't designed for accuracy, these are already > > used by some sensitive places so it is better to be more accurate. > > The design is to sacrifice accuracy and the time the updates occur for >

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed, Nov 25, 2015 at 01:00:22PM +0100, Michal Hocko wrote: > On Tue 24-11-15 15:22:03, Joonsoo Kim wrote: > > When I tested compaction in low memory condition, I found that > > my benchmark is stuck in congestion_wait() at shrink_inactive_list(). > > This stuck last for 1 sec and after then it

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed, 25 Nov 2015, Michal Hocko wrote: > > Simply remove the counter from the vmstat handling and do it differently > > then. > > We definitely do not want yet another set of counters. vmstat counters > are not only to be exported into the userspace. We have in kernel users > as well. I do

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed 25-11-15 10:04:44, Christoph Lameter wrote: > On Wed, 25 Nov 2015, Joonsoo Kim wrote: > > > I think that maintaining duplicate counter to guarantee accuracy isn't > > reasonable solution. It would cause more overhead to the system. > > Simply remove the counter from the vmstat handling

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Thu, Nov 26, 2015 at 7:26 AM, Joonsoo Kim wrote: > On Wed, Nov 25, 2015 at 01:00:22PM +0100, Michal Hocko wrote: >> On Tue 24-11-15 15:22:03, Joonsoo Kim wrote: >> > When I tested compaction in low memory condition, I found that >> > my benchmark is stuck in

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Tue 24-11-15 15:22:03, Joonsoo Kim wrote: > When I tested compaction in low memory condition, I found that > my benchmark is stuck in congestion_wait() at shrink_inactive_list(). > This stuck last for 1 sec and after then it can escape. More investigation > shows that it is due to stale vmstat

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Wed 25-11-15 14:43:38, Vlastimil Babka wrote: > On 11/25/2015 01:00 PM, Michal Hocko wrote: > > On Tue 24-11-15 15:22:03, Joonsoo Kim wrote: > >> When I tested compaction in low memory condition, I found that > >> my benchmark is stuck in congestion_wait() at shrink_inactive_list(). > >> This

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On 11/25/2015 01:00 PM, Michal Hocko wrote: > On Tue 24-11-15 15:22:03, Joonsoo Kim wrote: >> When I tested compaction in low memory condition, I found that >> my benchmark is stuck in congestion_wait() at shrink_inactive_list(). >> This stuck last for 1 sec and after then it can escape. More

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Tue, Nov 24, 2015 at 09:36:09AM -0600, Christoph Lameter wrote: > On Tue, 24 Nov 2015, Joonsoo Kim wrote: > > > When I tested compaction in low memory condition, I found that > > my benchmark is stuck in congestion_wait() at shrink_inactive_list(). > > This stuck last for 1 sec and after then

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Tue, 24 Nov 2015, Joonsoo Kim wrote: > When I tested compaction in low memory condition, I found that > my benchmark is stuck in congestion_wait() at shrink_inactive_list(). > This stuck last for 1 sec and after then it can escape. More investigation > shows that it is due to stale vmstat

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Tue, Nov 24, 2015 at 09:36:09AM -0600, Christoph Lameter wrote: > On Tue, 24 Nov 2015, Joonsoo Kim wrote: > > > When I tested compaction in low memory condition, I found that > > my benchmark is stuck in congestion_wait() at shrink_inactive_list(). > > This stuck last for 1 sec and after then

Re: [PATCH] mm/vmstat: retrieve more accurate vmstat value

On Tue, 24 Nov 2015, Joonsoo Kim wrote: > When I tested compaction in low memory condition, I found that > my benchmark is stuck in congestion_wait() at shrink_inactive_list(). > This stuck last for 1 sec and after then it can escape. More investigation > shows that it is due to stale vmstat

[PATCH] mm/vmstat: retrieve more accurate vmstat value

When I tested compaction in low memory condition, I found that my benchmark is stuck in congestion_wait() at shrink_inactive_list(). This stuck last for 1 sec and after then it can escape. More investigation shows that it is due to stale vmstat value. vmstat is updated every 1 sec so it is stuck

[PATCH] mm/vmstat: retrieve more accurate vmstat value

When I tested compaction in low memory condition, I found that my benchmark is stuck in congestion_wait() at shrink_inactive_list(). This stuck last for 1 sec and after then it can escape. More investigation shows that it is due to stale vmstat value. vmstat is updated every 1 sec so it is stuck