FYI, we noticed the below changes on commit 982792c782ef337381e982fd2047391886f89693 ("x86, mm: probe memory block size for generic x86 64bit") I think this is use memory size reduction, so should be a good thing. testbox/testcase/testparams: brickland3/boot/1 c46a7c817e662a82 982792c782ef337381e982fd20 ---------------- -------------------------- %stddev %change %stddev \ | \ 128348 ± 0% -12.3% 112541 ± 1% boot-meminfo.SUnreclaim 903719 ± 0% -13.9% 778082 ± 0% boot-slabinfo.num_objs 169283 ± 0% -9.7% 152819 ± 1% boot-meminfo.Slab 45691 ± 0% -8.2% 41954 ± 1% boot-slabinfo.num_pages brickland3: Brickland Ivy Bridge-EX Memory: 512G boot-slabinfo.num_objs 940000 ++-----------------------------------------------------------------+ 920000 ++ *. * | *.**.**.*. : * + *.* .*.**. *.*. .* .* | 900000 ++ * *.* * * ** * * | 880000 ++ | | | 860000 ++ | 840000 ++ | 820000 ++ | | | 800000 O+ O O | 780000 ++OO O O OO OO OO O O O OO O O OO O | OO O O OO O O OO O O O O OO | 760000 ++ O O O O | 740000 ++-----------------------------------------------------------------+ [*] bisect-good sample [O] bisect-bad sample To reproduce: apt-get install ruby ruby-oj git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git cd lkp-tests bin/setup-local job.yaml # the job file attached in this email bin/run-local job.yaml Disclaimer: Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. Thanks, Huang, Ying