Test-case snippet on modified kernel on ESM (3.13.0-174.225 + patch)
The problem doesn't happen anymore.
[ 169.972017] cpu 4/pid 1765/task TWO :: change_prot_numa() :: address =
0x7f534f800000, end = 0x7f534fa00000
[ 171.976313] cpu 5/pid 1766/task ONE :: do_huge_pmd_numa_page() :: addr/mask
= 0x7f534f800000, addr = 0x7f534f800000, pmd = 0x80003fff4a0009e6, *pmdp =
0x80003fff4a0009e6, pmdp = 0xffff8800375cb3e0
[ 171.976319] cpu 4/pid 1765/task TWO :: do_huge_pmd_numa_page() :: addr/mask
= 0x7f534f800000, addr = 0x7f534f800000, pmd = 0x80003fff4a0009e6, *pmdp =
0x80003fff4a0009e6, pmdp = 0xffff8800375cb3e0
[ 171.976320] cpu 4/pid 1765/task TWO :: Stage 1. T2 awaking up T1. Task 2 at
entry point.
[ 171.976321] cpu 4/pid 1765/task TWO :: Stage 2. T2 waiting on T1. Task 2
waiting for Task 1 at exit/migration point.
[ 171.980026] cpu 5/pid 1766/task ONE :: Stage 1. T1 waiting on T2. Task 1 at
entry point, checking for Task 2 at entry point too.
[ 171.980026] cpu 5/pid 1766/task ONE :: Stage 1. T1 resumed by T2
[ 172.003775] cpu 5/pid 1766/task ONE :: migrate_misplaced_transhuge_page() ::
addr/mask = 0x7f534f800000, addr = 0x7f534f800000, pmd = 0x80003fff4a0009e6,
*pmdp = 0x80003fff4a0009e6, pmdp = 0xffff8800375cb3e0, node = 1
[ 172.007754] cpu 5/pid 1766/task ONE :: Stage 2. T1 awaking up T2. Task 1 at
migration point.
[ 172.007754] cpu 5/pid 1766/task ONE :: Stage 3. T1 waiting on T2. Task 1
waiting for Task 2 to check migration status.
[ 171.980086] cpu 4/pid 1765/task TWO :: Stage 2. T2 resumed by T1
[ 172.013721] cpu 4/pid 1765/task TWO :: Stage 3. T2 awaking up T1. Task 2 has
checked migration status.
[ 172.013722] cpu 4/pid 1765/task TWO :: Stage 4. T2 waiting on T1. Task 2
waiting for Task 1 to clear the PMD.
[ 172.007754] cpu 5/pid 1766/task ONE :: Stage 3. T1 resumed by T2
[ 172.030364] cpu 5/pid 1766/task ONE :: Stage 4. T1 awaking up T2. Task 1 has
cleared the PMD.
[ 172.033374] cpu 5/pid 1766/task ONE :: Stage 4. T1 warning up T2: it may
BUG!
[ 172.033698] cpu 5/pid 1766/task ONE :: Stage 5. T1 sleep for 5s.
[ 173.004133] cpu 4/pid 1765/task TWO :: Stage 4. T2 resumed by T1
[ 173.008102] cpu 4/pid 1765/task TWO :: Stage 4. T2 sleep for 1s. any BUG
afterward?
[ 177.040138] cpu 5/pid 1766/task ONE :: Stage 5. T1 slept for 5s.
[ 181.044020] cpu 4/pid 1765/task TWO :: change_prot_numa() :: address =
0x7f534f800000, end = 0x7f534fa00000
[ 183.977694] cpu 5/pid 1766/task ONE :: do_huge_pmd_numa_page() :: addr/mask
= 0x7f534f800000, addr = 0x7f534f800000, pmd = 0x80003ffdc9c009e6, *pmdp =
0x80003ffdc9c009e6, pmdp = 0xffff8800375cb3e0
[ 183.977703] cpu 4/pid 1765/task TWO :: do_huge_pmd_numa_page() :: addr/mask
= 0x7f534f800000, addr = 0x7f534f800000, pmd = 0x80003ffdc9c009e6, *pmdp =
0x80003ffdc9c009e6, pmdp = 0xffff8800375cb3e0
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1813018
Title:
Kernel Oops - unable to handle kernel paging request; RIP is at
wait_migrate_huge_page+0x51/0x70
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1813018/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
