On Sat, Aug 06, 2011 at 05:25:04PM +0000, Mindaugas Rasiukevicius wrote:
> Module Name: src
> Committed By: rmind
> Date: Sat Aug 6 17:25:04 UTC 2011
>
> Modified Files:
> src/sys/uvm: uvm_amap.c uvm_anon.c uvm_anon.h uvm_fault.c uvm_loan.c
> uvm_map.c
>
> Log Message:
> - Rework uvm_anfree() into uvm_anon_freelst(), which always drops the lock.
> - Free anons in uvm_anon_freelst() without lock held.
> - Mechanic sync to unused loaning code.
These changes caused a regression;
ump/rumpkern/t_sp (393/449): 10 test cases
basic: Passed.
fork_fakeauth: Passed.
fork_pipecomm: Passed.
fork_simple: Passed.
reconnect: Passed.
signal: Passed.
sigsafe: Passed.
stress_killer: panic: kernel diagnostic assertion "anon->an_page ==
NULL" failed: file
"/bracket/i386/work/2011.08.06.19.23.38/src/sys/uvm/uvm_anon.c", line
202
cpu0: Begin traceback...
?(c0e86374,dc6,c5548940,1,c0ea46e4,0,c4ecbf80,1,c0ea3914,0) at 0
cpu0: End traceback...
dumping to dev 0,1 offset 1984
dump 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1
succeeded
The same assertion is catched with other tests too (see [1]).
- Jukka.
[1] Examples:
http://releng.netbsd.org/b5reports/i386/build/2011.08.06.19.23.38/test.log
http://releng.netbsd.org/b5reports/i386/build/2011.08.06.19.47.54/test.log
http://releng.netbsd.org/b5reports/i386/build/2011.08.06.19.53.24/test.log
http://releng.netbsd.org/b5reports/i386/build/2011.08.07.02.18.56/test.log
