You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've gave a try to "trunk" build from ~25 DEC on WL500GPv1 (brcm47xx - legacy). It generally worked (except for swap). But it seems to have memleak. Haven't happened in early versions.
To reproduce:
Install trunk on WL500GP V1. As non-standard difference I have ath9k card installed in minipci, therefore kmod-ath9k package is installed.
Login by ssh and look on "free" output (ofc line including buff/cache).
I had about 16 MB of free RAM total with freshly booted firmware at this point (no luci, etc, just stock trunk package set + ath9k).
Let it run for a while without reboots. I ran it about 5 days.
Repeat step 2 and look free RAM again.
At this point I had just 10 MB free RAM without obvious users of rather large 6MB chunk. Let's remind, it inlcudes buffers/cache.
Result:
In 5 days about 6MB RAM gone to unknown use I failed to identify. It does not seems to be attributed to processes, ps gives nothing useful. I would surely spot 6MB-sized process but there was none.
The text was updated successfully, but these errors were encountered:
jsmith8042:
I've gave a try to "trunk" build from ~25 DEC on WL500GPv1 (brcm47xx - legacy). It generally worked (except for swap). But it seems to have memleak. Haven't happened in early versions.
To reproduce:
Result:
In 5 days about 6MB RAM gone to unknown use I failed to identify. It does not seems to be attributed to processes, ps gives nothing useful. I would surely spot 6MB-sized process but there was none.
The text was updated successfully, but these errors were encountered: