[Info-vax] Running Alpha VMS under the ES40 emulator

Simon Clubley clubley at remove_me.eisner.decus.org-Earth.UFP
Sat Oct 12 20:47:01 EDT 2013


On 2013-10-12, Johnny Billquist <bqt at softjar.se> wrote:
> On 2013-10-12 17:27, Simon Clubley wrote:
>> On 2013-10-12, Johnny Billquist <bqt at softjar.se> wrote:
>>>
>>> While I will not exclude memory leaks from likely problems, that is a
>>> problem that should be equally bad daytime, and much worse during active
>>> use...
>>>
>>
>> It is and it is.
>>
>> When I leave Firefox downloading something during the day while I do
>> something else and then come back to it a couple or so hours later I also
>> encounter some thrashing.
>
> Do I even need to point out that "doing something else" is pretty 
> equivalent to the system doing something else...
> "Some trashing" essentially just means that pages have been paged out 
> and the physical memory have been put to other use by the OS. I would be 
> surprised if that did not happen if you do something else on the machine.
>

"Doing something else" does not mean working on the same box in this case.
It means physically leaving the box alone with only the normal routine
background processes running and nothing else running interactively apart
from Firefox.

It's not like I am trying to build a cross compiler toolchain at the same
time as running Firefox. :-)

>> When using Firefox heavily while doing some research, I also encounter
>> that same problem after a period of time.
>
> This does not really smell like memory leaks or big bugs, just the 
> normal expected behavior. If this is all you get, then it would be 
> reasonable for things to be stable over night when the system is not 
> touched.
>
> Ah well. I do not feel like arguing over this. I *know* that the nightly 
> jobs will push processes out of memory for file cache use.
> You can easily simulate it by just (as root) do a "find / -name '*.core' 
> -print" twice, and once they have completed, check how responsive your 
> web browser is.

Oh, I agree with you on that; it's just that the same thing happens when
I don't do anything like that.

One would expect memory to be reclaimed after a period of time as tabs
and windows are closed. Instead, apart from some minor changes, the
overall trend appears to be generally upwards and a continued slowing
of Firefox even after multiple tabs and windows have been closed.

Simon.

-- 
Simon Clubley, clubley at remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world



More information about the Info-vax mailing list