Hi Andreas,
Well, it was indeed caused by the timer refactoring. I didn't notice
that PJ_TIMER_HEAP_USE_COPY had been renamed to PJ_TIMER_USE_COPY in a
later revision, so setting PJ_TIMER_HEAP_USE_COPY to 0 had no effect at all.
Now with PJ_TIMER_USE_COPY set to 0 I get the old memory usage back.
(~70MByte compared to ~88MByte on the same system with the same
configuration).
Best regards,
Pirmin
Am 15.04.20 um 07:10 schrieb Andreas Wehrmann:
On 14/04/2020 16:09, Pirmin Walthert wrote:
Hi there,
A few days ago I've tested Asterisk 16.9 with Pjsip 2.10 (the bundled
version is still 2.9 and so is the version we usually use).
One thing I've noticed was that the combination with Pjsip 2.10 was
using a lot more memory according to top (compiled with the same
config_site.h and the same configure options, the usage was ~90MByte
compared to ~68MByte after boot). This was even the case with
PJ_TIMER_DEBUG 0 and PJ_TIMER_HEAP_USE_COPY 0
So with these two options disabled I'd expect about the same memory
footprint. Does anyone have an idea about which changeset could be
responsible for this?
Best regards,
Pirmin
Hey there,
is the memory usage higher, right from the beginning?
I'm asking because I'm currently dealing with memory usage problems as
well, with my own app though:
https://github.com/pjsip/pjproject/issues/2389
Regards,
Andreas
_______________________________________________
Visit our blog: http://blog.pjsip.org
pjsip mailing list
pjsip@xxxxxxxxxxxxxxx
http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org
_______________________________________________
Visit our blog: http://blog.pjsip.org
pjsip mailing list
pjsip@xxxxxxxxxxxxxxx
http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org
