Re: G1 STW phases and FGC column in jstat

Hi Yasumasa, Looks good. Regards, Kirk > On Mar 3, 2016, at 2:23 PM, Yasumasa Suenaga wrote: > > Hi all, > > I've uploaded PoC of PerfCounter for STW phase in concurrent GC: > > hotspot: http://cr.openjdk.java.net/~ysuenaga/perfcounter-for-cgc/hotspot/ > jdk:

Re: G1 STW phases and FGC column in jstat

Hi all, I've uploaded PoC of PerfCounter for STW phase in concurrent GC: hotspot: http://cr.openjdk.java.net/~ysuenaga/perfcounter-for-cgc/hotspot/ jdk: http://cr.openjdk.java.net/~ysuenaga/perfcounter-for-cgc/jdk/ These patches makes new jstat column CGC and CGCT as below:

Re: G1 STW phases and FGC column in jstat

It is not difficult to create webrev for this proposal. However, this proposal affects to jstat output, and I cannot access JPRT. So I want to find sponsor for it at first. Could somebody be a sponsor? Yasumasa On 2016/03/02 22:29, k...@kodewerk.com wrote: Hi Yasumasa, It all sounds

Re: G1 STW phases and FGC column in jstat

Hi Yasumasa, It all sounds reasonable and of course the impact on downstream tooling vs the value of the change must be evaluated. In this case I think adding a CGC counter makes sense especially in prep for G1 becoming default in 9. Regards, Kirk > On Mar 2, 2016, at 6:58 AM, Yasumasa

Re: G1 STW phases and FGC column in jstat

Hi Kirk, Agree to you. However, I use occasional FGC counter at CMS as below: 1. Check major collection occurrence Some production systems have large memory as Java heap, and they are not set GC log. If their CPU usage becomes high, I want to check GC execution. (Of

Re: G1 STW phases and FGC column in jstat

Hi Yasumasa, Good question. I’ve never considered CMS to be a Full GC. This implies that there should be separate performance counters for CMS pause phases as it is possible to have FGC. Of course a FGC during CMS maybe user triggered, triggered outside a CMS cycle, interrupts a CMS cycle, or

Re: G1 STW phases and FGC column in jstat

Hi, Would adopting MiGC (minor) and MaGC (major) in jstat output make more sense to cover all collectors' algorithms if we need a change here? Thanks. Tao Mao On Tue, Mar 1, 2016 at 3:03 PM, Yasumasa Suenaga wrote: > Hi Kirk, > > It is also incorrect to count initial mark

Re: G1 STW phases and FGC column in jstat

Hi Kirk, It is also incorrect to count initial mark and remark in CMS as a FGC. Though, how can we check execution of major collection without GC log? Should we add new PerfCounter for CGC (and add CGC column to jstat output)? Yasumasa On 2016/03/02 6:35, k...@kodewerk.com wrote: Hi, I

Re: G1 STW phases and FGC column in jstat

Hi, I think it is incorrect to count remark and cleanup as FGC. They are not full collections. It is also incorrect to count initial mark and remark in CMS as a FGC. It is unfortunate that this is counted this way. Regards, Kirk > On Mar 1, 2016, at 8:56 AM, Yasumasa Suenaga

G1 STW phases and FGC column in jstat

Hi all, I wonder that STW phases (Remark and Cleanup) at G1 are not counted in jstat FGC column. For example, Initial Mark and Remark at CMS are counted as FGC. For consistency, I think that G1 STW phases should be counted as FGC. What do you think about it? If it is accepted, I will file it