On Thu, Oct 11, 2018 at 05:12:12PM +0000, Wei Yang wrote: > On Thu, Oct 11, 2018 at 08:28:01PM +0800, kernel test robot wrote: >> On Thu, Oct 11, 2018 at 09:59:24AM +0000, richard.weiyang(a)gmail.com wrote: >>> Oops, >>> >>> Thanks for testing. >>> >>> On Thu, Oct 11, 2018 at 04:33:28PM +0800, kernel test robot wrote: >>>> Greetings, >>>> >>>> 0day kernel testing robot got the below dmesg and the first bad commit is >>>> >>>> https://github.com/0day-ci/linux/commits/owner-linux-mm-kvack-org/mm-remo... >>>> >>>> commit cbb02697f838cc10002c3ac09f0468791d8bf328 >>>> Author: owner-linux-mm(a)kvack.org <owner-linux-mm(a)kvack.org&gt; >>>> AuthorDate: Wed Oct 10 20:53:27 2018 +0800 >>>> Commit: 0day robot <lkp(a)intel.com&gt; >>>> CommitDate: Thu Oct 11 06:15:36 2018 +0800 >>>> >>>> mm: remove a redundant check in do_munmap() >>>> >>>> A non-NULL vma returned from find_vma() implies: >>>> >>>> vma->vm_start <= start >>>> >>>> Since len != 0, the following condition always hods: >>>> >>>> vma->vm_start < start + len = end >>>> >>>> This means the if check would never be true. >>>> >>>> This patch removes this redundant check and fix two typo in comment. >>>> >>>> Signed-off-by: Wei Yang <richard.weiyang(a)gmail.com&gt; >>>> >>>> dd52cb8790 Merge tag 'platform-drivers-x86-v4.17-4' of git://git.infradead.org/linux-platform-drivers-x86 >>>> cbb02697f8 mm: remove a redundant check in do_munmap() >>>> +------------------------------------------------+------------+------------+ >>>> | | dd52cb8790 | cbb02697f8 | >>>> +------------------------------------------------+------------+------------+ >>>> | boot_successes | 52 | 7 | >>>> | boot_failures | 0 | 22 | >>>> | IP-Config:Auto-configuration_of_network_failed | 0 | 2 | >>>> | BUG:Bad_rss-counter_state_mm:(ptrval)idx:#val | 0 | 13 | >>>> | BUG:Bad_rss-counter_state_mm:#idx:#val | 0 | 5 | >>>> | kernel_BUG_at_mm/filemap.c | 0 | 5 | >>>> | invalid_opcode:#[##] | 0 | 5 | >>>> | RIP:unaccount_page_cache_page | 0 | 5 | >>>> | Kernel_panic-not_syncing:Fatal_exception | 0 | 5 | >>>> +------------------------------------------------+------------+------------+ >>>> >>>> [child2:625] ppoll (271) returned ENOSYS, marking as inactive. >>>> [child2:625] pkey_mprotect (329) returned ENOSYS, marking as inactive. >>>> [child2:625] swapon (167) returned ENOSYS, marking as inactive. >>>> [child2:625] splice (275) returned ENOSYS, marking as inactive. >>>> [child3:637] readlinkat (267) returned ENOSYS, marking as inactive. >>>> [ 64.240371] BUG: Bad rss-counter state mm: (ptrval) idx:1 val:1 >>>> [child2:652] faccessat (269) returned ENOSYS, marking as inactive. >>>> [child2:652] clock_adjtime (305) returned ENOSYS, marking as inactive. >>>> [child0:615] accept (43) returned ENOSYS, marking as inactive. >>>> [child0:615] io_submit (209) returned ENOSYS, marking as inactive. >>>> [child0:615] pwritev2 (328) returned ENOSYS, marking as inactive. >>>> >>>> # HH:MM RESULT GOOD BAD GOOD_BUT_DIRTY DIRTY_NOT_BAD >>>> git bisect start b3b6bb615d30ce1a710a3c3c415567af214cb34a 0238df646e6224016a45505d2c111a24669ebe21 -- >>>> git bisect bad 92cca9f16f670b4d5e78fa5fe118be92d2e2209b # 11:16 B 0 13 30 2 Merge 'tty/tty-next' into devel-spot-201810110745 >>>> git bisect bad c84cb0975077ea9b3b32190c55a8779241581209 # 11:16 B 0 11 29 3 Merge 'linux-review/Juergen-Gross/xen-drop-writing-error-messages-to-xenstore/20181011-032312' into devel-spot-201810110745 >>>> git bisect bad 1658f3952513eff47b1c63a76c74d974d1b9455a # 11:16 B 0 13 31 3 Merge 'radeon-alex/drm-next-4.20' into devel-spot-201810110745 >>>> git bisect bad b76535f30171e534a0dc48864c14c5fe6b726546 # 11:16 B 0 34 52 3 Merge 'userns/new-mount-api-testing' into devel-spot-201810110745 >>>> git bisect good 8a2c32fd1e09b34d6cf01ff8323c469e346b4dee # 11:16 G 11 0 0 6 Merge 'linux-review/Carl-Huang/ath10k-allocate-small-size-dma-memory-in-ath10k_pci_diag_write_mem/20181011-065025' into devel-spot-201810110745 >>>> git bisect good 4bb0052f3fa5f10e6820b48749d5b66200a806ac # 11:16 G 11 0 0 1 Merge 'ceph-client/testing' into devel-spot-201810110745 >>>> git bisect bad 54602f4b420feb7650f1589fa589bc5aeab49fed # 14:00 B 0 4 18 0 Merge 'linux-review/owner-linux-mm-kvack-org/mm-remove-a-redundant-check-in-do_munmap/20181011-061535' into devel-spot-201810110745 >>>> git bisect bad cbb02697f838cc10002c3ac09f0468791d8bf328 # 14:45 B 0 1 1 1 mm: remove a redundant check in do_munmap() >>>> # first bad commit: [cbb02697f838cc10002c3ac09f0468791d8bf328] mm: remove a redundant check in do_munmap() >>>> git bisect good dd52cb879063ca5470d5a093c32180ba74fa528e # 15:29 G 52 0 0 0 Merge tag 'platform-drivers-x86-v4.17-4' of git://git.infradead.org/linux-platform-drivers-x86 >>>> # extra tests with debug options >>>> git bisect bad cbb02697f838cc10002c3ac09f0468791d8bf328 # 15:48 B 0 3 17 0 mm: remove a redundant check in do_munmap() >>>> # extra tests on HEAD of linux-devel/devel-spot-201810110745 >>>> git bisect bad b3b6bb615d30ce1a710a3c3c415567af214cb34a # 15:48 B 0 24 44 2 0day head guard for 'devel-spot-201810110745' >>>> # extra tests on tree/branch linux-review/owner-linux-mm-kvack-org/mm-remove-a-redundant-check-in-do_munmap/20181011-061535 >>>> git bisect bad cbb02697f838cc10002c3ac09f0468791d8bf328 # 15:50 B 4 7 0 5 mm: remove a redundant check in do_munmap() >>>> # extra tests with first bad commit reverted >>>> git bisect good 217501d4a8346ca87cd5482bf151b5dc46944ab1 # 16:17 G 16 0 0 0 Revert "mm: remove a redundant check in do_munmap()" >>>> >>>> --- >>>> 0-DAY kernel test infrastructure Open Source Technology Center >>>> https://lists.01.org/pipermail/lkp Intel Corporation >>> Below is the testing script, right? >>> >> yes, please have a try. > I didn't manage to run the same test as in your script. > > The error message I get is: > > qemu: linux kernel too old to load a ram disk > > I build the qemu from source, build kernel in guest, wget the initrd as > described in your script. Hmm... confused about this behavior. Could > you share some light on this? > > Well I managed to boot the guest with a kernel containing this commit, > but the command line is more simple: > > qemu-system-x86_64 -m 8G -smp 8 --enable-kvm -drive file=fedora.img --nograpic > > Totally, reboot around 10 times and build kernel in the guest > successfully. > > Generally, the change in this commit is a general mmap function, which > means has little relationship about the guest configuration, as I > understand. And the configuration in your test environment looks more > related to the cpu/device type, which may not impact the mmap behavior. > > So the difference is in the kernel tree? > > Would you mind share your kernel tree? > > Thanks I also tried on a physical machine. Reboot and build kernel successfully. Curious about which situation leads to this error.