On 09/30/2013 12:45 PM, Stanislav Kholmanskikh wrote: > On 09/26/2013 06:06 PM, chru...@suse.cz wrote: >> Hi! >>>> If I remove this sleep and perform two sequential runs I get this error: >>>> root@debian7-x64:/opt/ltp/testcases/bin# ./cpuset_memory_spread_testset.sh >>>> 100+0 records in >>>> 100+0 records out >>>> 104857600 bytes (105 MB) copied, 0.0595533 s, 1.8 GB/s >>>> cpuset11 1 TFAIL : hog the memory on the unexpected >>>> node(FilePages_For_Nodes(KB): _0: 0 >>>> _1: 336 >>>> _2: 0 >>>> _3: 0, Expect Nodes: 0). >>>> >>>> But if I take a few seconds between two runs the error disappears. >>>> >>>> I suppose this is an environment issue but don't know how to manage it. >>> Hmm, strange indeed, so it looks like the memory is allocated under >>> wrong node. >>> >>> I would guess that there is someting wrong with the cpuset that moves >>> the test process into the new set. Maybe the system needs some time to >>> move the process memory to the node we specified. >> What about cating /proc/$test_pid/status before we allocate the memory >> and checking that Mems_allowed_list matches the expected node? >> > Hmmm, Mems_allowed* content of /proc/$test_pid/status before sending > SIGUSR1 to $test_pid is: > > Mems_allowed: 00000000,00000001 > Mems_allowed_list: 0 > cpuset11 1 TFAIL : hog the memory on the unexpected > node(FilePages_For_Nodes(KB): _0: 112 > _1: 0 > _2: 0 > _3: 0, Expect Nodes: 0). > > i.e. It is a failure on: > base_test "0" "0" "0" "0" > > So, may be, It's a testcase issue. I'm Looking more deeply. > >
It seems that just /bin/echo 3 > /proc/sys/vm/drop_caches is not sufficient to free memory. In accordance to Documentation/sysctl/vm.txt we need to invoke 'sync' before that. I removed all the 'sleep's from the testcases and add a single sync before sending SIGUSR1 to mem_hog. After that this testcase started to work reliable. ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60133471&iu=/4140/ostg.clktrk _______________________________________________ Ltp-list mailing list Ltp-list@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ltp-list
