[lkp-robot] [mm/swap] 73c6573f40: vm-scalability.throughput 67.1% improvement
by kernel test robot
Greeting,
FYI, we noticed a 67.1% improvement of vm-scalability.throughput due to commit:
commit: 73c6573f407836b36bd8b2ebcd41895ffad4cca8 ("mm/swap: Split swap cache into 64MB trunks")
git://bee.sh.intel.com/git/yhuang/linux.git swap_optimize_v5.1
in testcase: vm-scalability
on test machine: 72 threads Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz with 128G memory
with following parameters:
thp_enabled: never
thp_defrag: never
nr_task: 32
disk: 1pmem
test: swap-w-seq
unit_size: 96G
size: 96G
cpufreq_governor: performance
test-description: The motivation behind this suite is to exercise functions and regions of the mm/ of the Linux kernel which are of interest to us.
test-url: https://git.kernel.org/cgit/linux/kernel/git/wfg/vm-scalability.git/
In addition to that, the commit also has significant impact on the following tests:
+------------------+-----------------------------------------------------------------------+
| testcase: change | vm-scalability: vm-scalability.throughput 70.1% improvement |
| test machine | 72 threads Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz with 128G memory |
| test parameters | cpufreq_governor=performance |
| | disk=1pmem |
| | nr_task=16 |
| | size=96G |
| | test=swap-w-seq |
| | thp_defrag=never |
| | thp_enabled=never |
| | unit_size=96G |
+------------------+-----------------------------------------------------------------------+
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
testcase/path_params/tbox_group/run: vm-scalability/never-never-32-1pmem-swap-w-seq-96G-96G-performance/lkp-hsw-ep4
4d4f250ef3a5b1cc 73c6573f407836b36bd8b2ebcd
---------------- --------------------------
3110660 67% 5199065 vm-scalability.throughput
61.98 15% 71.43 ± 3% vm-scalability.time.user_time
96063 -9% 87145 vm-scalability.time.involuntary_context_switches
89.37 -28% 64.40 vm-scalability.time.elapsed_time
89.37 -28% 64.40 vm-scalability.time.elapsed_time.max
2662 -28% 1905 vm-scalability.time.system_time
25022160 34% 33611704 interrupts.CAL:Function_call_interrupts
56.18 9% 61.05 turbostat.RAMWatt
47.24 7% 50.71 turbostat.%Busy
1320 7% 1417 turbostat.Avg_MHz
24568 ± 74% -2e+04 718 ±183% latency_stats.max.pipe_read.__vfs_read.vfs_read.SyS_read.entry_SYSCALL_64_fastpath
39577 ± 81% -4e+04 1022 ±179% latency_stats.sum.pipe_read.__vfs_read.vfs_read.SyS_read.entry_SYSCALL_64_fastpath
686 ± 48% 152% 1732 ± 9% vmstat.swap.si
1427204 47% 2104822 ± 3% vmstat.swap.so
351747 67% 587050 vmstat.system.in
4329 19% 5156 vmstat.system.cs
4.728e+08 54% 7.273e+08 perf-stat.node-stores
6.219e+08 ± 3% 45% 9.022e+08 perf-stat.node-store-misses
19.40 ± 5% 42% 27.60 perf-stat.iTLB-load-miss-rate%
7576 ± 19% 42% 10763 ± 6% perf-stat.major-faults
1.483e+11 13% 1.672e+11 ± 4% perf-stat.dTLB-stores
3.474e+09 10% 3.829e+09 perf-stat.cache-misses
1.085e+10 7% 1.165e+10 perf-stat.cache-references
0.35 5% 0.36 perf-stat.ipc
32.00 32.86 perf-stat.cache-miss-rate%
74.65 -4% 71.49 perf-stat.node-load-miss-rate%
2.878e+08 -8% 2.66e+08 ± 5% perf-stat.dTLB-store-misses
1.745e+09 -12% 1.537e+09 perf-stat.node-load-misses
389085 -14% 335970 perf-stat.context-switches
8.723e+08 -15% 7.427e+08 perf-stat.branch-misses
7.205e+11 -17% 5.981e+11 ± 4% perf-stat.dTLB-loads
0.19 ± 3% -18% 0.16 ± 3% perf-stat.dTLB-store-miss-rate%
3.027e+12 -19% 2.466e+12 ± 3% perf-stat.instructions
7.719e+11 -19% 6.267e+11 perf-stat.branch-instructions
8.748e+12 -22% 6.802e+12 ± 4% perf-stat.cpu-cycles
210261 ± 6% -24% 158831 ± 6% perf-stat.instructions-per-iTLB-miss
60024629 -32% 40803777 perf-stat.iTLB-loads
turbostat.RAMWatt
76 ++---------------------------------------------------------------------+
74 O+ O |
| O O |
72 ++ |
70 ++ |
68 ++ O |
66 ++ |
| |
64 ++ |
62 ++ O O |
60 ++ O O |
58 ++ |
| ..*...*..*... ..*...*...*... |
56 *+..*...*...*...*...*...*...*. *. *...*...*...*
54 ++---------------------------------------------------------------------+
vm-scalability.time.user_time
95 ++---------------------------------------------------------------------+
| O |
90 ++ |
85 ++ |
| |
80 ++ |
| |
75 ++ O |
| O |
70 O+ O O O O O |
65 ++ |
| ..*...*...*...*...*...*...*...*...*..*... ..*...*...*.. .*... |
60 *+ *. . .. *...*
| * |
55 ++---------------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
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,
Xiaolong
4 years
[lkp-robot] [block] 210734751f: BUG:sleeping_function_called_from_invalid_context_at_mm/slab.h
by kernel test robot
FYI, we noticed the following commit:
commit: 210734751fd01e57d335e3b81804d43c0743d663 ("block: Don't register a registered bdi device")
url: https://github.com/0day-ci/linux/commits/Yijing-Wang/block-Don-t-register...
base: https://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git for-next
in testcase: boot
on test machine: qemu-system-x86_64 -enable-kvm -cpu host -smp 2 -m 4G
caused below changes:
+-----------------------------------------------------------------------------+------------+------------+
| | cdb98c2698 | 210734751f |
+-----------------------------------------------------------------------------+------------+------------+
| boot_successes | 154 | 13 |
| boot_failures | 70 | 150 |
| BUG:kernel_hang_in_test_stage | 12 | |
| BUG:kernel_reboot-without-warning_in_test_stage | 36 | |
| invoked_oom-killer:gfp_mask=0x | 22 | 6 |
| Mem-Info | 22 | 6 |
| Kernel_panic-not_syncing:Out_of_memory_and_no_killable_processes | 22 | 6 |
| BUG:sleeping_function_called_from_invalid_context_at_mm/slab.h | 0 | 144 |
| BUG:scheduling_while_atomic | 0 | 3 |
| BUG:sleeping_function_called_from_invalid_context_at_kernel/locking/mutex.c | 0 | 10 |
+-----------------------------------------------------------------------------+------------+------------+
[ 61.299048] piix4_smbus 0000:00:01.3: SMBus Host Controller at 0x700, revision 0
[ 61.334072] FDC 0 is a S82078B
[ 61.343793] BUG: sleeping function called from invalid context at mm/slab.h:408
[ 61.343796] in_atomic(): 1, irqs_disabled(): 0, pid: 161, name: systemd-udevd
[ 61.343823] CPU: 1 PID: 161 Comm: systemd-udevd Not tainted 4.9.0-07151-g21073475 #1
[ 61.343825] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.9.3-20161025_171302-gandalf 04/01/2014
[ 61.343827] Call Trace:
[ 61.343860] dump_stack+0x63/0x8a
[ 61.343887] ___might_sleep+0xd3/0x120
[ 61.343891] __might_sleep+0x4a/0x80
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email
Thanks,
Xiaolong
4 years
[lkp-robot] [locking/mutex] e274795ea7: WARNING:at_arch/x86/include/asm/fpu/internal.h:#__switch_to
by kernel test robot
FYI, we noticed the following commit:
commit: e274795ea7b7caa0fd74ef651594382a69e2a951 ("locking/mutex: Fix mutex handoff")
https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git locking/core
in testcase: boot
on test machine: qemu-system-i386 -enable-kvm -cpu Haswell,+smep,+smap -m 360M
caused below changes:
+------------------------------------------------------------------------+------------+------------+
| | 52b94129f2 | e274795ea7 |
+------------------------------------------------------------------------+------------+------------+
| boot_successes | 840 | 87 |
| boot_failures | 4 | 132 |
| WARNING:at_arch/x86/include/asm/fpu/internal.h:#fpu__restore | 2 | 1 |
| invoked_oom-killer:gfp_mask=0x | 2 | 2 |
| Mem-Info | 2 | 2 |
| WARNING:at_arch/x86/include/asm/fpu/internal.h:#__switch_to | 0 | 131 |
| WARNING:at_arch/x86/include/asm/fpu/internal.h:#copy_fpregs_to_fpstate | 0 | 131 |
| WARNING:at_arch/x86/include/asm/fpu/internal.h:#copy_kernel_to_fpregs | 0 | 131 |
| kernel_BUG_at_include/linux/mm.h | 0 | 1 |
| invalid_opcode:#[##]PREEMPT_DEBUG_PAGEALLOC | 0 | 1 |
| Kernel_panic-not_syncing:Fatal_exception | 0 | 1 |
+------------------------------------------------------------------------+------------+------------+
[ 7.744273] Write protecting the kernel text: 21308k
[ 7.745224] Write protecting the kernel read-only data: 7016k
[ 7.747379] ------------[ cut here ]------------
[ 7.748061] WARNING: CPU: 0 PID: 1 at arch/x86/include/asm/fpu/internal.h:348 __switch_to+0x102d/0x1670
[ 7.749520] Modules linked in:
[ 7.750018] CPU: 0 PID: 1 Comm: init Not tainted 4.10.0-rc3-00137-ge274795 #2
[ 7.750923] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.9.3-20161025_171302-gandalf 04/01/2014
[ 7.752259] Call Trace:
[ 7.752692] ---[ end trace 67461e5452e5849e ]---
[ 7.774744] random: init: uninitialized urandom read (12 bytes read)
[ 7.784423] ------------[ cut here ]------------
[ 7.785523] WARNING: CPU: 0 PID: 7 at arch/x86/include/asm/fpu/internal.h:363 __switch_to+0x4bf/0x1670
[ 7.787335] Modules linked in:
[ 7.787892] CPU: 0 PID: 7 Comm: rcu_preempt Tainted: G W 4.10.0-rc3-00137-ge274795 #2
[ 7.789237] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.9.3-20161025_171302-gandalf 04/01/2014
[ 7.790671] Call Trace:
[ 7.791114] ---[ end trace 67461e5452e5849f ]---
[ 7.793802] ------------[ cut here ]------------
[ 7.794519] WARNING: CPU: 0 PID: 91 at arch/x86/include/asm/fpu/internal.h:348 copy_fpregs_to_fpstate+0x1dc/0x450
[ 7.797175] Modules linked in:
[ 7.797758] CPU: 0 PID: 91 Comm: rc.local Tainted: G W 4.10.0-rc3-00137-ge274795 #2
[ 7.799328] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.9.3-20161025_171302-gandalf 04/01/2014
[ 7.801004] Call Trace:
[ 7.801827] dump_stack+0x32/0x4f
[ 7.802789] __warn+0x222/0x260
[ 7.803734] ? copy_fpregs_to_fpstate+0x1dc/0x450
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email
Thanks,
Xiaolong
4 years
[lkp-robot] [blk] 738042d5db: general_protection_fault:#[##]SMP_KASAN
by kernel test robot
FYI, we noticed the following commit:
commit: 738042d5db927fe70d947a54bb8966d5c68e1195 ("blk-mq: abstract out helpers for allocating/freeing tag maps")
https://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git blk-mq-sched.6
in testcase: boot
on test machine: qemu-system-x86_64 -enable-kvm -cpu kvm64,+ssse3 -smp 2 -m 8G
caused below changes:
+------------------------------------------------------------------+------------+------------+
| | 8f1f3eabd4 | 738042d5db |
+------------------------------------------------------------------+------------+------------+
| boot_successes | 4 | 1 |
| boot_failures | 16 | 21 |
| BUG:workqueue_lockup-pool | 16 | 14 |
| invoked_oom-killer:gfp_mask=0x | 0 | 3 |
| Mem-Info | 0 | 3 |
| Kernel_panic-not_syncing:Out_of_memory_and_no_killable_processes | 0 | 2 |
| general_protection_fault:#[##]SMP_KASAN | 0 | 4 |
| Kernel_panic-not_syncing:Fatal_exception | 0 | 4 |
| Out_of_memory:Kill_process | 0 | 1 |
+------------------------------------------------------------------+------------+------------+
[ 373.421849] FDC 0 is a S82078B
[ 373.515156] kasan: CONFIG_KASAN_INLINE enabled
[ 373.521315] kasan: GPF could be caused by NULL-ptr deref or user memory access
[ 373.532987] general protection fault: 0000 [#1] SMP KASAN
[ 373.536302] Modules linked in:
[ 373.536302] CPU: 1 PID: 1 Comm: swapper/0 Not tainted 4.10.0-rc3-00059-g738042d #1
[ 373.552229] task: ffff8801f51f4000 task.stack: ffff8801f51f8000
[ 373.552229] RIP: 0010:blk_mq_free_rqs+0x3c/0x510
[ 373.552229] RSP: 0000:ffff8801f51ffa78 EFLAGS: 00010212
[ 373.552229] RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffff8801f51f47b8
[ 373.552229] RDX: 0000000000000010 RSI: 0000000000000000 RDI: ffff8801f4161310
[ 373.552229] RBP: ffff8801f51ffad8 R08: 0000000000000000 R09: 0000000000000000
[ 373.552229] R10: 0000000000000000 R11: 0000000000000000 R12: ffff8801f4161310
[ 373.552229] R13: ffff8801ce073248 R14: ffff8801f4161348 R15: 0000000000000001
[ 373.552229] FS: 0000000000000000(0000) GS:ffff8801f6700000(0000) knlGS:0000000000000000
[ 373.552229] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 373.552229] CR2: 0000000000000000 CR3: 0000000005422000 CR4: 00000000000006a0
[ 373.552229] Call Trace:
[ 373.552229] ? trace_hardirqs_on+0x1d/0x30
[ 373.552229] blk_mq_free_map_and_requests+0x79/0x1e0
[ 373.552229] blk_mq_free_tag_set+0x8c/0x1b0
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email
Thanks,
Xiaolong
4 years
[lkp-robot] [mac80211] 96aa2e7cf1: BUG:sleeping_function_called_from_invalid_context_at_kernel/locking/mutex.c
by kernel test robot
FYI, we noticed the following commit:
commit: 96aa2e7cf126773b16c6c19b7474a8a38d3c707e ("mac80211: calculate min channel width correctly")
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master
in testcase: hwsim
with following parameters:
group: hwsim-06
on test machine: 8 threads Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz with 8G memory
caused below changes:
+-----------------------------------------------------------------------------+------------+------------+
| | 06f7c88c10 | 96aa2e7cf1 |
+-----------------------------------------------------------------------------+------------+------------+
| boot_successes | 3 | 0 |
| boot_failures | 3 | 4 |
| BUG:sleeping_function_called_from_invalid_context_at_kernel/printk/printk.c | 2 | 1 |
| BUG:kernel_reboot-without-warning_in_test_stage | 1 | |
| BUG:sleeping_function_called_from_invalid_context_at_kernel/locking/mutex.c | 0 | 4 |
+-----------------------------------------------------------------------------+------------+------------+
[ 89.509631] ieee80211 phy1: BASIC_RATES: 0x1
[ 89.514783] wlan0 waiting for 4-way handshake completion with wlan1 02:00:00:00:01:00
[ 89.514787]
[ 89.550888] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:232
[ 89.552459] in_atomic(): 1, irqs_disabled(): 0, pid: 353, name: kworker/u16:6
[ 89.554097] CPU: 1 PID: 353 Comm: kworker/u16:6 Not tainted 4.10.0-rc2-00138-g96aa2e7 #1
[ 89.555804] Hardware name: /DH67GD, BIOS BLH6710H.86A.0132.2011.1007.1505 10/07/2011
[ 89.557631] Workqueue: phy1 ieee80211_iface_work [mac80211]
[ 89.559276] Call Trace:
[ 89.560756] dump_stack+0x63/0x8a
[ 89.562277] ___might_sleep+0xd3/0x120
[ 89.563777] __might_sleep+0x4a/0x80
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
Thanks,
Xiaolong
4 years
[lkp-robot] [kernel, mm] 2295815a88: WARNING:at_include/linux/refcount.h:#bdi_init
by kernel test robot
FYI, we noticed the following commit:
commit: 2295815a8895c673fcabc6b85ca13749da97f99a ("kernel, mm: convert from atomic_t to refcount_t")
https://git.kernel.org/pub/scm/linux/kernel/git/kees/linux.git kspp/hardened-atomic
in testcase: trinity
with following parameters:
runtime: 300s
test-description: Trinity is a linux system call fuzz tester.
test-url: http://codemonkey.org.uk/projects/trinity/
on test machine: qemu-system-i386 -enable-kvm -m 320M
caused below changes:
+-------------------------------------------------+------------+------------+
| | 1854813c7e | 2295815a88 |
+-------------------------------------------------+------------+------------+
| boot_successes | 15 | 0 |
| boot_failures | 5 | 31 |
| BUG:kernel_hang_in_boot_stage | 4 | 3 |
| BUG:kernel_reboot-without-warning_in_test_stage | 1 | |
| WARNING:at_include/linux/refcount.h:#bdi_init | 0 | 31 |
| WARNING:at_include/linux/refcount.h:#bdi_exit | 0 | 27 |
| BUG:workqueue_lockup-pool | 0 | 1 |
+-------------------------------------------------+------------+------------+
[ 3.931914] PCI: PCI BIOS revision 2.10 entry at 0xfd3e3, last bus=0
[ 3.950088] PCI: Using configuration type 1 for base access
[ 4.051083] ------------[ cut here ]------------
[ 4.073510] WARNING: CPU: 0 PID: 1 at include/linux/refcount.h:109 bdi_init+0x387/0x3b0
[ 4.119068] refcount_t: increment on 0; use-after-free.
[ 4.156622] CPU: 0 PID: 1 Comm: swapper Not tainted 4.10.0-rc2-00008-g2295815 #45
[ 4.194578] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.9.3-20161025_171302-gandalf 04/01/2014
[ 4.230291] Call Trace:
[ 4.245875] dump_stack+0x16/0x22
[ 4.260843] __warn+0xd6/0xf0
[ 4.275332] ? bdi_init+0x387/0x3b0
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email
Thanks,
Xiaolong
4 years
[lkp-robot] [locking/ww_mutex] 3c7f9edadc: WARNING:at_kernel/locking/mutex.c:#__ww_mutex_wakeup_for_backoff
by kernel test robot
FYI, we noticed the following commit:
commit: 3c7f9edadceda905e9b92655bc0522c2d56bb6e9 ("locking/ww_mutex: Wake at most one waiter for back off when acquiring the lock")
https://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git locking/testing
in testcase: trinity
with following parameters:
runtime: 300s
test-description: Trinity is a linux system call fuzz tester.
test-url: http://codemonkey.org.uk/projects/trinity/
on test machine: qemu-system-x86_64 -enable-kvm -m 512M
caused below changes:
+------------------------------------------------------------------+------------+------------+
| | 0200165943 | 3c7f9edadc |
+------------------------------------------------------------------+------------+------------+
| boot_successes | 4 | 0 |
| boot_failures | 4 | 8 |
| BUG:kernel_hang_in_test_stage | 4 | 4 |
| WARNING:at_kernel/locking/mutex.c:#__ww_mutex_wakeup_for_backoff | 0 | 8 |
+------------------------------------------------------------------+------------+------------+
[ 0.000000] ---[ end trace f68728a0d3053b52 ]---
[ 0.000000] ok |
[ 0.000000] ------------[ cut here ]------------
[ 0.000000] WARNING: CPU: 0 PID: 0 at kernel/locking/mutex.c:305 __ww_mutex_wakeup_for_backoff+0xa3/0xc0
[ 0.000000] Modules linked in:
[ 0.000000] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G W 4.10.0-rc3-00065-g3c7f9ed #1
[ 0.000000] Call Trace:
[ 0.000000] dump_stack+0xc4/0x131
[ 0.000000] __warn+0xfa/0x120
[ 0.000000] warn_slowpath_null+0x18/0x20
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email
Thanks,
Xiaolong
4 years
[lkp-developer] [mm/swap] 735bab895e: vm-scalability.throughput 69.3% improvement
by kernel test robot
Greeting,
FYI, we noticed a 69.3% improvement of vm-scalability.throughput due to commit:
commit: 735bab895e64c930581ffb0a05b661e01da82bc5 ("mm/swap: Split swap cache into 64MB trunks")
git://bee.sh.intel.com/git/yhuang/linux.git swap_optimize_v4.4
in testcase: vm-scalability
on test machine: 72 threads Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz with 128G memory
with following parameters:
thp_enabled: never
thp_defrag: never
nr_task: 16
disk: 1pmem
test: swap-w-seq
unit_size: 96G
size: 96G
cpufreq_governor: performance
test-description: The motivation behind this suite is to exercise functions and regions of the mm/ of the Linux kernel which are of interest to us.
test-url: https://git.kernel.org/cgit/linux/kernel/git/wfg/vm-scalability.git/
In addition to that, the commit also has significant impact on the following tests:
+------------------+-----------------------------------------------------------------------+
| testcase: change | vm-scalability: vm-scalability.throughput 72.1% improvement |
| test machine | 72 threads Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz with 128G memory |
| test parameters | cpufreq_governor=performance |
| | disk=1pmem |
| | nr_task=32 |
| | size=96G |
| | test=swap-w-seq |
| | thp_defrag=never |
| | thp_enabled=never |
| | unit_size=96G |
+------------------+-----------------------------------------------------------------------+
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
testcase/path_params/tbox_group/run: vm-scalability/never-never-16-1pmem-swap-w-seq-96G-96G-performance/lkp-hsw-ep4
dbb860bbd825b1aa 735bab895e64c930581ffb0a05
---------------- --------------------------
%stddev change %stddev
\ | \
2829217 69% 4790718 vm-scalability.throughput
89726 -22% 70110 vm-scalability.time.involuntary_context_switches
100.35 -32% 68.37 vm-scalability.time.elapsed_time
100.35 -32% 68.37 vm-scalability.time.elapsed_time.max
1515 -33% 1013 vm-scalability.time.system_time
27758463 22% 33891449 interrupts.CAL:Function_call_interrupts
54.14 10% 59.59 turbostat.RAMWatt
751 10% 830 turbostat.Avg_MHz
26.64 11% 29.47 turbostat.%Busy
179 4% 186 turbostat.PkgWatt
536 ± 17% 85% 994 ± 15% vmstat.swap.si
1361688 48% 2016716 vmstat.swap.so
346054 61% 556228 vmstat.system.in
3763 23% 4619 vmstat.system.cs
6883 ± 70% -6e+03 550 ±110% latency_stats.max.pipe_read.__vfs_read.vfs_read.SyS_read.entry_SYSCALL_64_fastpath
11358 ± 24% 1e+05 125592 ± 38% latency_stats.sum.devkmsg_read.__vfs_read.vfs_read.SyS_read.entry_SYSCALL_64_fastpath
32663 ± 25% 1e+05 136542 ± 34% latency_stats.sum.ep_poll.SyS_epoll_wait.do_syscall_64.return_from_SYSCALL_64
11110 ± 66% -1e+04 622 ±112% latency_stats.sum.pipe_read.__vfs_read.vfs_read.SyS_read.entry_SYSCALL_64_fastpath
5.161e+08 60% 8.271e+08 perf-stat.node-stores
6.492e+08 28% 8.282e+08 perf-stat.node-store-misses
37.54 5% 39.47 perf-stat.iTLB-load-miss-rate%
6491 ± 6% 24% 8067 ± 6% perf-stat.major-faults
1.491e+11 ± 3% 11% 1.655e+11 ± 4% perf-stat.dTLB-stores
3.981e+09 ± 4% 5% 4.172e+09 perf-stat.cache-misses
0.39 11% 0.43 perf-stat.ipc
0.17 -6% 0.16 perf-stat.branch-miss-rate%
31.15 5% 32.65 perf-stat.cache-miss-rate%
55.71 -10% 50.03 perf-stat.node-store-miss-rate%
75.99 -5% 71.87 perf-stat.node-load-miss-rate%
2.767e+08 -13% 2.404e+08 ± 6% perf-stat.dTLB-store-misses
378648 -16% 318561 perf-stat.context-switches
1.787e+09 ± 3% -18% 1.457e+09 ± 4% perf-stat.node-load-misses
1.007e+09 -20% 8.057e+08 perf-stat.branch-misses
5.108e+08 ± 7% -27% 3.733e+08 ± 12% perf-stat.dTLB-load-misses
0.19 ± 4% -22% 0.14 perf-stat.dTLB-store-miss-rate%
2.36e+12 ± 3% -14% 2.02e+12 perf-stat.instructions
6.043e+11 ± 3% -15% 5.158e+11 perf-stat.branch-instructions
60039 ± 3% 8% 64890 perf-stat.instructions-per-iTLB-miss
39308015 -21% 31137783 perf-stat.iTLB-load-misses
3191 ± 9% -23% 2466 ± 7% perf-stat.cpu-migrations
0.11 ± 8% -23% 0.09 ± 8% perf-stat.dTLB-load-miss-rate%
6.055e+12 ± 3% -23% 4.653e+12 ± 3% perf-stat.cpu-cycles
65389663 -27% 47781728 perf-stat.iTLB-loads
vm-scalability.throughput
9e+06 ++--O--O------O-----------------------------------------------------+
O O |
8e+06 ++ |
| |
7e+06 ++ |
| |
6e+06 ++ |
| |
5e+06 ++ O O O |
| O |
4e+06 ++ |
| |
3e+06 ++ ..*..*...*..*...*
*...*..*...*..*...*..*...*...*..*...*..*...*...*..*. |
2e+06 ++------------------------------------------------------------------+
vm-scalability.time.involuntary_context_switches
95000 ++------------------------------------------------------------------+
| .*.. .*.. ..*..*... .*...|
90000 *+..*. . .*...*.. .. *. *...*..*...*..*...*. *
85000 ++ *. *...* |
| |
80000 ++ |
| |
75000 ++ O |
O O |
70000 ++ O O |
65000 ++ O O |
| O |
60000 ++ |
| O |
55000 ++------------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
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,
Xiaolong
4 years
[lkp-developer] [mm] f6d5cfde88: aim9.shell_rtns_1.ops_per_sec -20.1% regression
by kernel test robot
Greeting,
FYI, we noticed a -20.1% regression of aim9.shell_rtns_1.ops_per_sec due to commit:
commit: f6d5cfde888b9e0356719fabe8754fdfe6fe236b ("mm: async free vma")
git://bee.sh.intel.com/git/aaron/linux.git parallel_free
in testcase: aim9
on test machine: 72 threads Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz with 128G memory
with following parameters:
testtime: 300s
test: shell_rtns_1
cpufreq_governor: performance
test-description: Suite IX is the "AIM Independent Resource Benchmark:" the famous synthetic benchmark.
test-url: https://sourceforge.net/projects/aimbench/files/aim-suite9/
In addition to that, the commit also has significant impact on the following tests:
+------------------+---------------------------------------------------------------------+
| testcase: change | unixbench: unixbench.score -18.4% regression |
| test machine | 24 threads Nehalem-EP with 24G memory |
| test parameters | cpufreq_governor=performance |
| | nr_task=1 |
| | runtime=300s |
| | test=shell1 |
+------------------+---------------------------------------------------------------------+
| testcase: change | ltp: will-it-scale.per_process_ops -90.0% regression |
| test machine | qemu-system-x86_64 -enable-kvm -cpu host -smp 4 -m 5G |
| test parameters | test=admin_tools |
+------------------+---------------------------------------------------------------------+
| testcase: change | unixbench: aim9.page_test.ops_per_sec -3.6% regression |
| test machine | 4 threads Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz with 4G memory |
| test parameters | cpufreq_governor=performance |
| | nr_task=100% |
| | runtime=300s |
| | test=shell1 |
+------------------+---------------------------------------------------------------------+
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
testcase/path_params/tbox_group/run: aim9/300s-shell_rtns_1-performance/lkp-hsw-ep4
v4.9 f6d5cfde888b9e0356719fabe8
---------------- --------------------------
475 -20% 379 aim9.shell_rtns_1.ops_per_sec
288815 505% 1747076 interrupts.CAL:Function_call_interrupts
17432 517% 107550 vmstat.system.cs
73370 8% 79016 vmstat.system.in
43.59 4% 45.37 turbostat.RAMWatt
7.09 14% 8.06 turbostat.%Busy
209 14% 238 turbostat.Avg_MHz
131.03 9% 143.03 turbostat.PkgWatt
5256757 518% 32469045 perf-stat.context-switches
659508 -9% 599405 perf-stat.cpu-migrations
1.648e+12 ± 3% 12% 1.846e+12 perf-stat.instructions
3.846e+11 9% 4.18e+11 perf-stat.dTLB-loads
1.011e+08 4% 1.055e+08 perf-stat.node-store-misses
3.15e+08 31% 4.138e+08 perf-stat.iTLB-loads
2.682e+10 16% 3.106e+10 perf-stat.cache-references
4.952e+12 ± 3% 13% 5.605e+12 perf-stat.cpu-cycles
2.611e+08 -8% 2.405e+08 perf-stat.dTLB-store-misses
5.708e+08 5% 5.989e+08 perf-stat.node-load-misses
23903847 5% 25039190 perf-stat.node-loads
0.14 -11% 0.12 perf-stat.dTLB-store-miss-rate%
65047421 6% 69250977 perf-stat.node-stores
0.40 -9% 0.36 perf-stat.dTLB-load-miss-rate%
8.001e+08 4% 8.324e+08 perf-stat.cache-misses
7.832e+09 -7% 7.294e+09 perf-stat.branch-misses
3.631e+11 ± 3% 14% 4.131e+11 perf-stat.branch-instructions
43158719 -19% 34759720 perf-stat.page-faults
43158703 -19% 34759720 perf-stat.minor-faults
2.98 -10% 2.68 perf-stat.cache-miss-rate%
0.33 0.33 perf-stat.ipc
3950 ± 3% 10% 4328 perf-stat.instructions-per-iTLB-miss
56.98 -11% 50.75 perf-stat.iTLB-load-miss-rate%
2.16 -18% 1.77 perf-stat.branch-miss-rate%
6258 ± 70% -5e+03 977 ± 48% latency_stats.avg.wait_on_page_bit.__filemap_fdatawait_range.filemap_fdatawait_keep_errors.sync_inodes_sb.sync_inodes_one_sb.iterate_supers.sys_sync.entry_SYSCALL_64_fastpath
9261 ± 51% -7e+03 1900 ± 13% latency_stats.max.wait_on_page_bit.__filemap_fdatawait_range.filemap_fdatawait_keep_errors.sync_inodes_sb.sync_inodes_one_sb.iterate_supers.sys_sync.entry_SYSCALL_64_fastpath
0 6e+06 5715232 latency_stats.sum.flush_work.unmap_vmas.exit_mmap.mmput.flush_old_exec.load_elf_binary.search_binary_handler.do_execveat_common.SyS_execve.do_syscall_64.return_from_SYSCALL_64
0 3e+06 3269078 latency_stats.sum.flush_work.unmap_vmas.unmap_region.do_munmap.SyS_munmap.entry_SYSCALL_64_fastpath
0 3e+06 2592048 latency_stats.sum.flush_work.unmap_vmas.unmap_region.do_munmap.mmap_region.do_mmap.vm_mmap_pgoff.SyS_mmap_pgoff.SyS_mmap.entry_SYSCALL_64_fastpath
0 2e+06 2008032 latency_stats.sum.flush_work.unmap_vmas.exit_mmap.mmput.flush_old_exec.load_elf_binary.search_binary_handler.load_script.search_binary_handler.do_execveat_common.SyS_execve.do_syscall_64
0 7e+05 701288 latency_stats.sum.flush_work.unmap_vmas.unmap_region.do_munmap.vm_munmap.elf_map.load_elf_binary.search_binary_handler.do_execveat_common.SyS_execve.do_syscall_64.return_from_SYSCALL_64
0 4e+05 402039 latency_stats.sum.flush_work.unmap_vmas.unmap_region.do_munmap.vm_munmap.elf_map.load_elf_binary.search_binary_handler.load_script.search_binary_handler.do_execveat_common.SyS_execve
9450 ± 51% -7e+03 1992 ± 20% latency_stats.sum.wait_on_page_bit.__filemap_fdatawait_range.filemap_fdatawait_keep_errors.sync_inodes_sb.sync_inodes_one_sb.iterate_supers.sys_sync.entry_SYSCALL_64_fastpath
perf-stat.iTLB-loads
5e+08 ++----------------------------------------------------------------+
4.5e+08 ++ O O O O O |
O O O O O O O O O O O O O O O O O O O O O O
4e+08 ++ |
3.5e+08 ++ |
*..*.*..*.*..* *.*..*.*.. .*..*..*.*..*.*.. .*..*.*..*.*..* |
3e+08 ++ : : * * |
2.5e+08 ++ : : |
2e+08 ++ : : |
| : : |
1.5e+08 ++ : : |
1e+08 ++ : : |
| :: |
5e+07 ++ : |
0 ++-------------*--------------------------------------------------+
perf-stat.page-faults
4.5e+07 *+-*-*--*-*--*------*--*-*------------*-----------------*---------+
| : * *.*..*..* *.*..*.*..*.*. *..* |
4e+07 ++ : : |
3.5e+07 O+ O O O O O O O O O O O O O O O O O O O O O O O O O O
| : : |
3e+07 ++ : : |
2.5e+07 ++ : : |
| : : |
2e+07 ++ : : |
1.5e+07 ++ : : |
| : : |
1e+07 ++ :: |
5e+06 ++ : |
| : |
0 ++-------------*--------------------------------------------------+
perf-stat.context-switches
3.5e+07 ++----------------------------------------------------------------+
O O O O O O O O O O O O O O O O O O O O O O O O O O O
3e+07 ++ |
| |
2.5e+07 ++ |
| |
2e+07 ++ |
| |
1.5e+07 ++ |
| |
1e+07 ++ |
| |
5e+06 *+.*.*..*.*..* *.*..*.*..*.*..*..*.*..*.*..*.*..*.*..*.*..* |
| + .. |
0 ++-------------*--------------------------------------------------+
perf-stat.minor-faults
4.5e+07 *+-*-*--*-*--*------*--*-*------------*-----------------*---------+
| : * *.*..*..* *.*..*.*..*.*. *..* |
4e+07 ++ : : |
3.5e+07 O+ O O O O O O O O O O O O O O O O O O O O O O O O O O
| : : |
3e+07 ++ : : |
2.5e+07 ++ : : |
| : : |
2e+07 ++ : : |
1.5e+07 ++ : : |
| : : |
1e+07 ++ :: |
5e+06 ++ : |
| : |
0 ++-------------*--------------------------------------------------+
perf-stat.iTLB-load-miss-rate_
70 ++---------------------------------------------------------------------+
| |
60 *+.*.*..*..*..* *..*..*.*..*.. .*.. |
| : : *..*.*..*..*.*. *..*.*..*..*..* |
50 O+ O O O O O O O O O O O O O O O O O O O O O O O O O O
| : : |
40 ++ : : |
| : : |
30 ++ : : |
| : : |
20 ++ : : |
| :: |
10 ++ :: |
| : |
0 ++--------------*------------------------------------------------------+
aim9.shell_rtns_1.ops_per_sec
500 ++--------------------------------------------------------------------+
450 *+.*.*..*..*.* *..*.*..*..*.*..*..*.*..*..*.*..*..*..*.*..*..* |
| : : |
400 ++ O O O O O O O O O O O O O O O O O O O O O O O
350 O+ O O O : : |
| : : |
300 ++ : : |
250 ++ : : |
200 ++ : : |
| : : |
150 ++ : : |
100 ++ : : |
| : |
50 ++ : |
0 ++--------------*-----------------------------------------------------+
aim9.time.voluntary_context_switches
7e+06 ++------------------------------------------------------------------+
| |
6e+06 O+ O O O O O O O O O O O O O O O O O O O O O O O O O O
| |
5e+06 ++ |
| |
4e+06 ++ |
| |
3e+06 ++ |
| |
2e+06 ++ |
| |
1e+06 *+.*.*..*.*..*.. *..*..*.*..* *.*..*.. *..* |
| + + .. + |
0 ++--------------*--------------*--*--*-*----------*-*--*--*---------+
interrupts.CAL:Function_call_interrupts
1.8e+06 ++----------------------------------------------O--O-O--O-O--O-O--O
O O O O O O O O O O O O O O O O O O O |
1.6e+06 ++ |
1.4e+06 ++ |
| |
1.2e+06 ++ |
1e+06 ++ |
| |
800000 ++ |
600000 ++ |
| |
400000 ++ |
200000 *+.*.*..*.*..* *.*..*.*..*.*..*..*.*..*.*..*.*..*.*..*.*..* |
| + .. |
0 ++-------------*--------------------------------------------------+
vmstat.system.in
80000 O+-O-O--O-O--O--O-O--O--O-O--O-O--O--O-O--O-O--O--O-O--O--O-O--O-O--O
*..*.*..*.*..* *..*..*.*..*.*..*..*.*..*.*..*..*.*..*..*.*..* |
70000 ++ : : |
60000 ++ : : |
| : : |
50000 ++ : : |
| : : |
40000 ++ : : |
| : : |
30000 ++ : : |
20000 ++ : : |
| :: |
10000 ++ : |
| : |
0 ++--------------*---------------------------------------------------+
vmstat.system.cs
120000 ++-----------------------------------------------------------------+
| O O O O O O O O O O O O O O O O O
100000 O+ O O O O O O O O O |
| |
| |
80000 ++ |
| |
60000 ++ |
| |
40000 ++ |
| |
| |
20000 *+.*.*..*.*..* *..*.*..*.*..*..*.*..*.*..*.*..*..*.*..*.*..* |
| + .. |
0 ++-------------*---------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
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,
Xiaolong
4 years
[lkp-developer] [perf/core] 49c04ee1a7: WARNING:at_kernel/events/core.c:#perf_iterate_ctx_matching
by kernel test robot
FYI, we noticed the following commit:
commit: 49c04ee1a704ad7fe785474d9d17a4341dcb50a3 ("perf/core: use rb-tree index to optimize filtered perf_iterate_ctx")
url: https://github.com/0day-ci/linux/commits/David-Carrillo-Cisneros/optimize...
in testcase: trinity
with following parameters:
runtime: 300s
test-description: Trinity is a linux system call fuzz tester.
test-url: http://codemonkey.org.uk/projects/trinity/
on test machine: qemu-system-i386 -enable-kvm -m 320M
caused below changes:
+------------------------------------------------------------+------------+------------+
| | 746b9b80c7 | 49c04ee1a7 |
+------------------------------------------------------------+------------+------------+
| boot_successes | 11 | 0 |
| boot_failures | 5 | 14 |
| BUG:kernel_reboot-without-warning_in_test_stage | 3 | |
| BUG:unable_to_handle_kernel | 2 | 14 |
| Oops | 2 | 14 |
| Kernel_panic-not_syncing:Fatal_exception | 2 | 14 |
| WARNING:at_kernel/events/core.c:#perf_iterate_ctx_matching | 0 | 14 |
+------------------------------------------------------------+------------+------------+
[ 293.129762] do_int80_syscall_32+0x4a/0x5c
[ 293.129788] entry_INT80_32+0x33/0x33
[ 293.129792] EIP: 0xa7704dae
[ 293.129794] EFLAGS: 00000246 CPU: 0
[ 293.129819] EAX: ffffffda EBX: 080615c0 ECX: 00000002 EDX: 080515b8
[ 293.129823] ESI: 08061694 EDI: 00000599 EBP: af8abad8 ESP: af8abab8
[ 293.129826] DS: 007b ES: 007b FS: 0000 GS: 0033 SS: 007b
[ 293.276233] ------------[ cut here ]------------
[ 293.276293] WARNING: CPU: 0 PID: 1257 at kernel/events/core.c:6428 perf_iterate_ctx_matching+0x34/0xfe
[ 293.276298] Modules linked in:
[ 293.276326] CPU: 0 PID: 1257 Comm: trinity-main Not tainted 4.9.0-02688-g49c04ee #12
[ 293.276330] Call Trace:
[ 293.276363] dump_stack+0x73/0xa5
[ 293.276390] ? perf_iterate_ctx_matching+0x34/0xfe
[ 293.276397] __warn+0xb2/0xc9
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email
Thanks,
Xiaolong
4 years