From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 100392 invoked by alias); 26 Aug 2018 06:27:03 -0000 Mailing-List: contact systemtap-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Post: List-Help: , Sender: systemtap-owner@sourceware.org Received: (qmail 100093 invoked by uid 48); 26 Aug 2018 06:26:49 -0000 From: "agentzh at gmail dot com" To: systemtap@sourceware.org Subject: [Bug testsuite/23493] Test suite makes all CPU stuck forever on kernel 4.16.16 (Fedora 27) Date: Sun, 26 Aug 2018 06:27:00 -0000 X-Bugzilla-Reason: AssignedTo X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: systemtap X-Bugzilla-Component: testsuite X-Bugzilla-Version: unspecified X-Bugzilla-Keywords: X-Bugzilla-Severity: normal X-Bugzilla-Who: agentzh at gmail dot com X-Bugzilla-Status: UNCONFIRMED X-Bugzilla-Resolution: X-Bugzilla-Priority: P2 X-Bugzilla-Assigned-To: systemtap at sourceware dot org X-Bugzilla-Target-Milestone: --- X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://sourceware.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-SW-Source: 2018-q3/txt/msg00126.txt.bz2 https://sourceware.org/bugzilla/show_bug.cgi?id=3D23493 --- Comment #7 from agentzh --- It seems like that the stuck usually happens after running `uprobes_onthefly.exp` and/or `kprobes_onthefly.exp`. By checking the `/proc/buddyinfo` output during the nearby time window, the fragmentation becomes dramatically worse, for example: ``` $ cat /proc/buddyinfo Node 0, zone DMA 0 0 0 0 2 1 1 = 0=20=20 1 1 3 Node 0, zone DMA32 14 23 27 27 23 12 17 = 16=20=20 9 4 739 Node 0, zone Normal 379 4113 3506 3803 2461 733 221 = 0=20=20 0 0 0 $ cat /proc/buddyinfo Node 0, zone DMA 0 0 0 0 2 1 1 = 0=20=20 1 1 3 Node 0, zone DMA32 5 5 10 20 15 5 16 = 16=20=20 9 4 700 Node 0, zone Normal 344 1273 1414 1396 1047 473 143 = 0=20=20 0 0 0 ``` Even though there's a lot of free memory in the system at the same times: ``` $ free -m total used free shared buff/cache avail= able Mem: 11700 4897 3510 76 3293 = 5702 Swap: 2047 0 2047 ``` I'm thinking about enforcing VM compaction every few minutes during the test suite runs as a work around... --=20 You are receiving this mail because: You are the assignee for the bug.