FYI, we noticed stderr.Run:"Process_Creation":Fork_failed_at_iteration in unixbench test on https://github.com/alexshi/power-scheduling.git x86-tlb commit ff9b625d26ba7750745d235dc6f1792fb36fea48 ("x86/tlb: just do tlb flush on one of siblings of SMT") 2016-03-17 06:04:39 ./Run spawn -c 1 # # # # # # # ##### ###### # # #### # # # # ## # # # # # # # ## # # # # # # # # # # # ## ##### ##### # # # # ###### # # # # # # ## # # # # # # # # # # # # ## # # # # # # # ## # # # # #### # # # # # ##### ###### # # #### # # Version 5.1.3 Based on the Byte Magazine Unix Benchmark Multi-CPU version Version 5 revisions by Ian Smith, Sunnyvale, CA, USA January 13, 2011 johantheghost at yahoo period com 1 x Process Creation 1 ========================================================================================= compiler/cpufreq_governor/kconfig/nr_task/rootfs/tbox_group/test/testcase: gcc-4.9/performance/x86_64-rhel/1/debian-x86_64-2015-02-07.cgz/lkp-ivb-d02/spawn/unixbench commit: v4.5 ff9b625d26ba7750745d235dc6f1792fb36fea48 v4.5 ff9b625d26ba7750745d235dc6 ---------------- -------------------------- fail:runs %reproduction fail:runs | | | :4 100% 4:4 last_state.failed :4 100% 4:4 last_state.is_incomplete_run :4 100% 4:4 last_state.unixbench.exit_code.1 :4 100% 4:4 stderr.********************************************** :4 50% 2:4 stderr.Reason:Success;aborting :4 100% 4:4 stderr.Run:"Process_Creation":Fork_failed_at_iteration :4 50% 2:4 stderr.Segmentation_fault;aborting ========================================================================================= compiler/cpufreq_governor/kconfig/nr_task/rootfs/tbox_group/test/testcase: gcc-4.9/performance/x86_64-rhel/1/debian-x86_64-2015-02-07.cgz/lkp-ivb-d04/spawn/unixbench commit: v4.5 ff9b625d26ba7750745d235dc6f1792fb36fea48 v4.5 ff9b625d26ba7750745d235dc6 ---------------- -------------------------- fail:runs %reproduction fail:runs | | | :8 100% 8:8 last_state.failed :8 100% 8:8 last_state.is_incomplete_run :8 100% 8:8 last_state.unixbench.exit_code.1 :8 100% 8:8 stderr.********************************************** :8 12% 1:8 stderr.Aborted;aborting :8 62% 5:8 stderr.Reason:Success;aborting :8 75% 6:8 stderr.Run:"Process_Creation":Fork_failed_at_iteration :8 12% 1:8 stderr.Run:"Process_Creation":Segmentation_fault;aborting :8 12% 1:8 stderr.Run:"Process_Creation":spawn:../nptl/sysdeps/unix/sysv/linux/x86_64/../fork.c:#:__libc_fork:Assertion`({__typeof(({struct_pthread*__self;asm("mov%%fs:%c1,%#":"=r"(__self):"i"(__builtin_offsetof(struct_pthread,header.self)));__self;})->tid)__value;if(sizeof(__value)==#)asm_volatile("movb%%fs:%P2,%b0":"=q"(__value):"#"(#),"i"(__builtin_offsetof(struct_pthread,tid)));else_if(sizeof(__value)==#)asm_volatile("movl%%fs:%P1,%#":"=r"(__value):"i"(__builtin_offsetof(struct_pthread,tid)));else{if(sizeof(__value)!=#)abort();asm_volatile("movq%%fs:%P1,%q0":"=r"(__value):"i"(__builtin_offsetof(struct_pthread,tid)));}__value;})==ppid'failed :8 12% 1:8 stderr.Segmentation_fault;aborting lkp-ivb-d02: Ivy Bridge Memory: 8G lkp-ivb-d04: Ivy Bridge Memory: 4G 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 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 Ye

Hi Xiaolong, I tried this benchmark on my native haswell box, but it works. and get unstable/better performance value. This patch may be has some problem when it run in guest OS. Did you run the case on a guest OS or a native OS?

Any other benchmark did you test on any other machine? If so, what's the result you get?

Regards Alex On 03/17/2016 02:00 PM, kernel test robot wrote: > FYI, we noticed stderr.Run:"Process_Creation":Fork_failed_at_iteration in unixbench test on > > https://github.com/alexshi/power-scheduling.git x86-tlb > commit ff9b625d26ba7750745d235dc6f1792fb36fea48 ("x86/tlb: just do tlb flush on one of siblings of SMT") > > 2016-03-17 06:04:39 ./Run spawn -c 1 > > # # # # # # # ##### ###### # # #### # # > # # ## # # # # # # # ## # # # # # > # # # # # # ## ##### ##### # # # # ###### > # # # # # # ## # # # # # # # # # > # # # ## # # # # # # # ## # # # # > #### # # # # # ##### ###### # # #### # # > > Version 5.1.3 Based on the Byte Magazine Unix Benchmark > > Multi-CPU version Version 5 revisions by Ian Smith, > Sunnyvale, CA, USA > January 13, 2011 johantheghost at yahoo period com > > > 1 x Process Creation 1 > > > ========================================================================================= > compiler/cpufreq_governor/kconfig/nr_task/rootfs/tbox_group/test/testcase: > gcc-4.9/performance/x86_64-rhel/1/debian-x86_64-2015-02-07.cgz/lkp-ivb-d02/spawn/unixbench > > commit: > v4.5 > ff9b625d26ba7750745d235dc6f1792fb36fea48 > > v4.5 ff9b625d26ba7750745d235dc6 > ---------------- -------------------------- > fail:runs %reproduction fail:runs > | | | > :4 100% 4:4 last_state.failed > :4 100% 4:4 last_state.is_incomplete_run > :4 100% 4:4 last_state.unixbench.exit_code.1 > :4 100% 4:4 stderr.********************************************** > :4 50% 2:4 stderr.Reason:Success;aborting > :4 100% 4:4 stderr.Run:"Process_Creation":Fork_failed_at_iteration > :4 50% 2:4 stderr.Segmentation_fault;aborting > > ========================================================================================= > compiler/cpufreq_governor/kconfig/nr_task/rootfs/tbox_group/test/testcase: > gcc-4.9/performance/x86_64-rhel/1/debian-x86_64-2015-02-07.cgz/lkp-ivb-d04/spawn/unixbench > > commit: > v4.5 > ff9b625d26ba7750745d235dc6f1792fb36fea48 > > v4.5 ff9b625d26ba7750745d235dc6 > ---------------- -------------------------- > fail:runs %reproduction fail:runs > | | | > :8 100% 8:8 last_state.failed > :8 100% 8:8 last_state.is_incomplete_run > :8 100% 8:8 last_state.unixbench.exit_code.1 > :8 100% 8:8 stderr.********************************************** > :8 12% 1:8 stderr.Aborted;aborting > :8 62% 5:8 stderr.Reason:Success;aborting > :8 75% 6:8 stderr.Run:"Process_Creation":Fork_failed_at_iteration > :8 12% 1:8 stderr.Run:"Process_Creation":Segmentation_fault;aborting > :8 12% 1:8 stderr.Run:"Process_Creation":spawn:../nptl/sysdeps/unix/sysv/linux/x86_64/../fork.c:#:__libc_fork:Assertion`({__typeof(({struct_pthread*__self;asm("mov%%fs:%c1,%#":"=r"(__self):"i"(__builtin_offsetof(struct_pthread,header.self)));__self;})->tid)__value;if(sizeof(__value)==#)asm_volatile("movb%%fs:%P2,%b0":"=q"(__value):"#"(#),"i"(__builtin_offsetof(struct_pthread,tid)));else_if(sizeof(__value)==#)asm_volatile("movl%%fs:%P1,%#":"=r"(__value):"i"(__builtin_offsetof(struct_pthread,tid)));else{if(sizeof(__value)!=#)abort();asm_volatile("movq%%fs:%P1,%q0":"=r"(__value):"i"(__builtin_offsetof(struct_pthread,tid)));}__value;})==ppid'failed > :8 12% 1:8 stderr.Segmentation_fault;aborting > > lkp-ivb-d02: Ivy Bridge > Memory: 8G > > lkp-ivb-d04: Ivy Bridge > Memory: 4G > > > 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 > > > 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 Ye >

Hi Xiaolong, Thanks for the info. Do you have any other benchmark data for this branch?

Regards Alex On 03/21/2016 12:43 PM, Ye Xiaolong wrote: > The unixbench is launched on a native OS. >> > >> >Any other benchmark did you test on any other machine? If so, what's the >> >result you get? > So far, we have just found issues with unixbench on two Ivy Bridge > machine.

Hi Xiaolong, Thanks for the info. Do you have any other benchmark data for this branch? Regards Alex On 03/21/2016 12:43 PM, Ye Xiaolong wrote: > The unixbench is launched on a native OS. >> > >> >Any other benchmark did you test on any other machine? If so, what's the >> >result you get? > So far, we have just found issues with unixbench on two Ivy Bridge > machine. _______________________________________________ LKP mailing list LKP(a)lists.01.org https://lists.01.org/mailman/listinfo/lkp

Long time no see, Huang Ying. :) How are you?

On 03/21/2016 02:33 PM, Huang, Ying wrote: > Usually we will merge all developer's git together to do performance > test. Because there are too many developer's git. This is the only > possible way to do test. Yes, it would reduce much of testing. but may miss some finding some performance changes covered each other...