Re: Top briefly updated too fast

On May 12, 2026 11:42:21 PM GMT+03:00, bob prohaska <fbsd@www.zefox.net> wrote:
>On Tue, May 12, 2026 at 11:35:17PM +0300, Sulev-Madis Silber wrote:
>> 
>> 
>> On May 12, 2026 10:32:16 PM GMT+03:00, bob prohaska <fbsd@www.zefox.net> wrote:
>> >While glancing at a top display from a Pi2 running buildworld
>> >on -current the display update rate went briefly from its 
>> >normal two second interval to several, perhaps five, updates
>> >per second. In less than a minute it then reverted to its
>> >customary 2 second interval. Top had been started with 8
>> >as the number of processes to report. 
>> >
>> >The top session had been running for hours (on the serial console,
>> >if that could matter). No sign that anything special had happened.
>> >
>> >I've never seen such behavior and wonder how it's possible....
>> >
>> >Thanks for reading!
>> >
>> >bob prohaska 
>> >
>> 
>> 
>> are you sure it didn't somehow lag? eg resource exhaustion or ssh over jittery links. then you would get huge bursts of data all at once which would look like top is speeding up due all the data in queue now also wants to be displayed. once that's done, it returns to normal update speed
>> 
>
>I'm not sure of anything 8-) and didn't notice what happened
>immediately before. I've seen such delays over ssh, though.
>The link in use was an FTDI  usb-serial adapter accessed via
>cu. 
>
>That's a much simpler explanation than top running too fast,
>which I find hard to imagine.
>
>Thanks for writing!
>
>bob prohaska
>
>


well who knows what lagged there

there's also chance of some hw or sw bug or combination of them too

oh and i only now noticed you run buildworld on rpi2 at same time of top

so more like resources and less of current bug. latters do exist too of course