Re: mbuf statistics, tracking of drops

2017-11-16 Thread David Gwynne
On Thu, Nov 16, 2017 at 08:13:39PM +0100, Gregor Best wrote: > On Thu, Nov 16, 2017 at 11:13:04AM +1000, David Gwynne wrote: > > > > > On 16 Nov 2017, at 7:23 am, Gregor Best wrote: > > > > > > Hi, > > > > > > On Mon, Nov 13, 2017 at 01:47:01PM +1000, David Gwynne wrote: >

Re: mbuf statistics, tracking of drops

2017-11-16 Thread Gregor Best
On Thu, Nov 16, 2017 at 11:13:04AM +1000, David Gwynne wrote: > > > On 16 Nov 2017, at 7:23 am, Gregor Best wrote: > > > > Hi, > > > > On Mon, Nov 13, 2017 at 01:47:01PM +1000, David Gwynne wrote: > >> [...] > >> pools maintain count of how many times they failed to provide

Re: mbuf statistics, tracking of drops

2017-11-15 Thread David Gwynne
> On 16 Nov 2017, at 7:23 am, Gregor Best wrote: > > Hi, > > On Mon, Nov 13, 2017 at 01:47:01PM +1000, David Gwynne wrote: >> [...] >> pools maintain count of how many times they failed to provide an >> allocation. you can watch this with vmstat -m or systat pool. >>

Re: mbuf statistics, tracking of drops

2017-11-15 Thread Gregor Best
Hi, On Mon, Nov 13, 2017 at 01:47:01PM +1000, David Gwynne wrote: > [...] > pools maintain count of how many times they failed to provide an > allocation. you can watch this with vmstat -m or systat pool. > however, we could use that to populate mb_drops too. > [...] That's certainly smarter

Re: mbuf statistics, tracking of drops

2017-11-12 Thread Martin Pieuchot
On 13/11/17(Mon) 13:47, David Gwynne wrote: > On Sun, Nov 12, 2017 at 11:23:31PM +0100, Gregor Best wrote: > > Hi Martin, > > > > On Sun, Nov 12, 2017 at 03:40:59PM +0100, Martin Pieuchot wrote: > > > [...] > > > It does, some comments below. > > > [...] > > > > Wonderful. > > > > > [...] > > >

Re: mbuf statistics, tracking of drops

2017-11-12 Thread David Gwynne
On Sun, Nov 12, 2017 at 11:23:31PM +0100, Gregor Best wrote: > Hi Martin, > > On Sun, Nov 12, 2017 at 03:40:59PM +0100, Martin Pieuchot wrote: > > [...] > > It does, some comments below. > > [...] > > Wonderful. > > > [...] > > This would be an approximation because it might happen that after >

Re: mbuf statistics, tracking of drops

2017-11-12 Thread Gregor Best
Hi Martin, On Sun, Nov 12, 2017 at 03:40:59PM +0100, Martin Pieuchot wrote: > [...] > It does, some comments below. > [...] Wonderful. > [...] > This would be an approximation because it might happen that after > returning NULL the second pool_get(9) won't sleep. However I think > it's the way

Re: mbuf statistics, tracking of drops

2017-11-12 Thread Martin Pieuchot
On 11/11/17(Sat) 13:05, Gregor Best wrote: > Hi people, > > while reading around in /sys/kern/uipc_mbuf.c to try to track down a > problem with my iwm(4) that seems to correlate with mbuf allocation > failures, I noticed that the MBSTAT_DROPS counter and its friends > MBSTAT_{WAIT,DRAIN} don't

mbuf statistics, tracking of drops

2017-11-11 Thread Gregor Best
Hi people, while reading around in /sys/kern/uipc_mbuf.c to try to track down a problem with my iwm(4) that seems to correlate with mbuf allocation failures, I noticed that the MBSTAT_DROPS counter and its friends MBSTAT_{WAIT,DRAIN} don't seem to get increased anywhere in /sys. Does the patch