Date: Thu, 09 Jan 2025 00:00:06 +0000 From: bugzilla-noreply@freebsd.org To: wireless@FreeBSD.org Subject: [Bug 283903] rtw88: possible skb leak Message-ID: <bug-283903-21060-nY9rPC1IkD@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-283903-21060@https.bugs.freebsd.org/bugzilla/> References: <bug-283903-21060@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D283903 --- Comment #3 from Guillaume Outters <guillaume-freebsd@outters.eu> --- Here it seems to happen only after memory has been exhausted at least once: after a reboot there *always* (see (1)) have to happen a bunch of "kernel: = pid xxx (xxxxx), jid 0, uid 1001, was killed: failed to reclaim memory" (2) bef= ore any "skb alloc failed" appears. Could it be that the leaks only happen after some allocations have been confined to the swap, and then even when mem gets freed the driver is unabl= e to regain its ability to allocate elsewhere than to swap? (please forgive my FreeBSD and VM newbiness if my hypothesis is filled with inaccuracies and nowhere near reality)=20 (1) { bunzip2 < messages.0.bz2 ; cat messages ; } | egrep alloc\|memory | awk '/real memory/{s=3D"0 boot"}/failed to reclaim memory/{s=3D"1 killed"}/skb = alloc failed/{s=3D"2 skb"}old!=3Ds{if(old)print n" x\t"old;n=3D0;old=3Ds}{++n}' (2) Repetitively building a full LLVM for all its architectures + clang + compiler-rt, with make -j4, on a 8 GB laptop. --=20 You are receiving this mail because: You are on the CC list for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-283903-21060-nY9rPC1IkD>