* Re: WARNING: at arch/x86/kernel/kvmclock.c:127 @ 2012-06-10 15:47 Sedat Dilek 2012-06-10 16:08 ` Sedat Dilek 2012-06-11 21:47 ` Eric B Munson 0 siblings, 2 replies; 9+ messages in thread From: Sedat Dilek @ 2012-06-10 15:47 UTC (permalink / raw) To: Eric B Munson Cc: kvm, LKML, Avi Kivity, Marcelo Tosatti, Glauber Costa, Thomas Gleixner, Ingo Molnar, x86 Hi, I see the same warning especially when resuming from a suspend (see timestamps >=30-35sec) between Linux v3.5-rc1..v3.5-rc2. $ sudo grep kvmclock.c /var/log/kern.log Jun 4 21:11:32 fambox kernel: [ 784.037237] WARNING: at arch/x86/kernel/kvmclock.c:127 kvm_check_and_clear_guest_paused+0x52/0x60() Jun 5 20:35:57 fambox kernel: [ 1928.458060] WARNING: at arch/x86/kernel/kvmclock.c:127 kvm_check_and_clear_guest_paused+0x52/0x60() Jun 8 09:35:52 fambox kernel: [ 3290.134637] WARNING: at arch/x86/kernel/kvmclock.c:127 kvm_check_and_clear_guest_paused+0x52/0x60() Jun 8 09:35:52 fambox kernel: [ 3290.238592] WARNING: at arch/x86/kernel/kvmclock.c:127 kvm_check_and_clear_guest_paused+0x52/0x60() Jun 8 12:11:20 fambox kernel: [ 5777.023571] WARNING: at arch/x86/kernel/kvmclock.c:127 kvm_check_and_clear_guest_paused+0x52/0x60() Jun 9 13:32:50 fambox kernel: [ 2778.842695] WARNING: at arch/x86/kernel/kvmclock.c:127 kvm_check_and_clear_guest_paused+0x52/0x60() >From [1]: "...The warning itself is not required for the check_and_clear function and can be removed as far as I am concerned." >From [2] commit 3b5d56b9317fa7b5407dff1aa7b115bf6cdbd494 ("kvmclock: Add functions to check if the host has stopped the vm") ... +bool kvm_check_and_clear_guest_paused(void) +{ + bool ret = false; + struct pvclock_vcpu_time_info *src; + + /* + * per_cpu() is safe here because this function is only called from + * timer functions where preemption is already disabled. + */ + WARN_ON(!in_atomic()); + src = &__get_cpu_var(hv_clock); + if ((src->flags & PVCLOCK_GUEST_STOPPED) != 0) { + __this_cpu_and(hv_clock.flags, ~PVCLOCK_GUEST_STOPPED); + ret = true; + } + + return ret; +} +EXPORT_SYMBOL_GPL(kvm_check_and_clear_guest_paused); + ... ( The export macro was dropped in a followup commit. ) So you mean "WARN_ON(!in_atomic());" can be deleted? Regards, - Sedat - [1] http://www.spinics.net/lists/kvm/msg73732.html [2] http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=blobdiff;f=arch/x86/kernel/kvmclock.c;h=4ba090ca689db5322fb242b98b53158de5447dc9;hp=f8492da65bfcb03e2775638ad5ce9502099d62c8;hb=3b5d56b9317fa7b5407dff1aa7b115bf6cdbd494;hpb=eae3ee7d8a7c59cf63441dedf28674889f5fc477 ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: WARNING: at arch/x86/kernel/kvmclock.c:127 2012-06-10 15:47 WARNING: at arch/x86/kernel/kvmclock.c:127 Sedat Dilek @ 2012-06-10 16:08 ` Sedat Dilek 2012-06-11 21:47 ` Eric B Munson 1 sibling, 0 replies; 9+ messages in thread From: Sedat Dilek @ 2012-06-10 16:08 UTC (permalink / raw) To: Eric B Munson Cc: kvm, LKML, Avi Kivity, Marcelo Tosatti, Thomas Gleixner, Ingo Molnar, x86, Rafael J. Wysocki [-- Attachment #1: Type: text/plain, Size: 2983 bytes --] On Sun, Jun 10, 2012 at 5:47 PM, Sedat Dilek <sedat.dilek@googlemail.com> wrote: > Hi, > > I see the same warning especially when resuming from a suspend (see > timestamps >=30-35sec) between Linux v3.5-rc1..v3.5-rc2. > > $ sudo grep kvmclock.c /var/log/kern.log > Jun 4 21:11:32 fambox kernel: [ 784.037237] WARNING: at > arch/x86/kernel/kvmclock.c:127 > kvm_check_and_clear_guest_paused+0x52/0x60() > Jun 5 20:35:57 fambox kernel: [ 1928.458060] WARNING: at > arch/x86/kernel/kvmclock.c:127 > kvm_check_and_clear_guest_paused+0x52/0x60() > Jun 8 09:35:52 fambox kernel: [ 3290.134637] WARNING: at > arch/x86/kernel/kvmclock.c:127 > kvm_check_and_clear_guest_paused+0x52/0x60() > Jun 8 09:35:52 fambox kernel: [ 3290.238592] WARNING: at > arch/x86/kernel/kvmclock.c:127 > kvm_check_and_clear_guest_paused+0x52/0x60() > Jun 8 12:11:20 fambox kernel: [ 5777.023571] WARNING: at > arch/x86/kernel/kvmclock.c:127 > kvm_check_and_clear_guest_paused+0x52/0x60() > Jun 9 13:32:50 fambox kernel: [ 2778.842695] WARNING: at > arch/x86/kernel/kvmclock.c:127 > kvm_check_and_clear_guest_paused+0x52/0x60() > > From [1]: > > "...The warning itself is not required for the check_and_clear > function and can be removed as far as I am concerned." > > From [2] commit 3b5d56b9317fa7b5407dff1aa7b115bf6cdbd494 ("kvmclock: > Add functions to check if the host has stopped the vm") > ... > +bool kvm_check_and_clear_guest_paused(void) > +{ > + bool ret = false; > + struct pvclock_vcpu_time_info *src; > + > + /* > + * per_cpu() is safe here because this function is only called from > + * timer functions where preemption is already disabled. > + */ > + WARN_ON(!in_atomic()); > + src = &__get_cpu_var(hv_clock); > + if ((src->flags & PVCLOCK_GUEST_STOPPED) != 0) { > + __this_cpu_and(hv_clock.flags, ~PVCLOCK_GUEST_STOPPED); > + ret = true; > + } > + > + return ret; > +} > +EXPORT_SYMBOL_GPL(kvm_check_and_clear_guest_paused); > + > ... > ( The export macro was dropped in a followup commit. ) > > So you mean "WARN_ON(!in_atomic());" can be deleted? > > Regards, > - Sedat - > > [1] http://www.spinics.net/lists/kvm/msg73732.html > [2] http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=blobdiff;f=arch/x86/kernel/kvmclock.c;h=4ba090ca689db5322fb242b98b53158de5447dc9;hp=f8492da65bfcb03e2775638ad5ce9502099d62c8;hb=3b5d56b9317fa7b5407dff1aa7b115bf6cdbd494;hpb=eae3ee7d8a7c59cf63441dedf28674889f5fc477 [ Removed Glauber Costa - Email-address @ RH does not exist anymore ] [ Added CC to Rafael J. Wysocki (S/R) ] I added for the sake of correctness the last call-trace I have seen with Linux v3.5-rc2. And... I had loaded kvm and kvm_intel kernel-modules, but no KVM instance actively running or active while doing suspend/resume (S/R). - Sedat - [-- Attachment #2: CALL-TRACE_kvmclock_kern-log.txt --] [-- Type: text/plain, Size: 9731 bytes --] Jun 9 13:32:50 fambox kernel: [ 2777.787365] Freezing user space processes ... (elapsed 0.01 seconds) done. Jun 9 13:32:50 fambox kernel: [ 2777.803351] Freezing remaining freezable tasks ... (elapsed 0.01 seconds) done. Jun 9 13:32:50 fambox kernel: [ 2777.803354] PM: Entering mem sleep Jun 9 13:32:50 fambox kernel: [ 2777.803420] Suspending console(s) (use no_console_suspend to debug) Jun 9 13:32:50 fambox kernel: [ 2777.803806] sd 1:0:0:0: [sdb] Synchronizing SCSI cache Jun 9 13:32:50 fambox kernel: [ 2777.803850] sd 0:0:0:0: [sda] Synchronizing SCSI cache Jun 9 13:32:50 fambox kernel: [ 2777.804168] sd 0:0:0:0: [sda] Stopping disk Jun 9 13:32:50 fambox kernel: [ 2777.806550] sd 1:0:0:0: [sdb] Stopping disk Jun 9 13:32:50 fambox kernel: [ 2778.422876] PM: suspend of devices complete after 619.779 msecs Jun 9 13:32:50 fambox kernel: [ 2778.422878] PM: suspend devices took 0.620 seconds Jun 9 13:32:50 fambox kernel: [ 2778.423006] PM: late suspend of devices complete after 0.126 msecs Jun 9 13:32:50 fambox kernel: [ 2778.438920] r8169 0000:02:00.0: wake-up capability enabled by ACPI Jun 9 13:32:50 fambox kernel: [ 2778.502850] PM: noirq suspend of devices complete after 79.895 msecs Jun 9 13:32:50 fambox kernel: [ 2778.503082] ACPI: Preparing to enter system sleep state S3 Jun 9 13:32:50 fambox kernel: [ 2778.526894] PM: Saving platform NVS memory Jun 9 13:32:50 fambox kernel: [ 2778.531564] Disabling non-boot CPUs ... Jun 9 13:32:50 fambox kernel: [ 2778.634664] CPU 1 is now offline Jun 9 13:32:50 fambox kernel: [ 2778.738600] CPU 2 is now offline Jun 9 13:32:50 fambox kernel: [ 2778.739230] Broke affinity for irq 9 Jun 9 13:32:50 fambox kernel: [ 2778.739247] Broke affinity for irq 16 Jun 9 13:32:50 fambox kernel: [ 2778.842535] CPU 3 is now offline Jun 9 13:32:50 fambox kernel: [ 2778.842688] ------------[ cut here ]------------ Jun 9 13:32:50 fambox kernel: [ 2778.842695] WARNING: at arch/x86/kernel/kvmclock.c:127 kvm_check_and_clear_guest_paused+0x52/0x60() Jun 9 13:32:50 fambox kernel: [ 2778.842696] Hardware name: <HIDDEN> Jun 9 13:32:50 fambox kernel: [ 2778.842723] Modules linked in: sch_fq_codel sch_htb snd_hda_codec_hdmi snd_hda_codec_realtek rfcomm parport_pc bnep ppdev joydev snd_hda_intel arc4 snd_hda_codec coretemp snd_hwdep kvm_intel snd_pcm snd_page_alloc kvm i915 snd_seq_midi iwlwifi snd_seq_midi_event snd_rawmidi snd_seq ghash_clmulni_intel snd_seq_device drm_kms_helper aesni_intel snd_timer uvcvideo drm aes_x86_64 mac80211 cryptd videobuf2_vmalloc snd videobuf2_memops psmouse i2c_algo_bit btusb videobuf2_core samsung_laptop soundcore microcode cfg80211 videodev lp bluetooth serio_raw mei hid_generic mac_hid lpc_ich video parport usbhid hid r8169 Jun 9 13:32:50 fambox kernel: [ 2778.842725] Pid: 5475, comm: pm-suspend Not tainted 3.5.0-rc2-1-iniza-generic #1 Jun 9 13:32:50 fambox kernel: [ 2778.842727] Call Trace: Jun 9 13:32:50 fambox kernel: [ 2778.842731] [<ffffffff810527bf>] warn_slowpath_common+0x7f/0xc0 Jun 9 13:32:50 fambox kernel: [ 2778.842734] [<ffffffff8105281a>] warn_slowpath_null+0x1a/0x20 Jun 9 13:32:50 fambox kernel: [ 2778.842736] [<ffffffff8103fc82>] kvm_check_and_clear_guest_paused+0x52/0x60 Jun 9 13:32:50 fambox kernel: [ 2778.842739] [<ffffffff810dda6d>] watchdog_timer_fn+0x8d/0x180 Jun 9 13:32:50 fambox kernel: [ 2778.842742] [<ffffffff8107a506>] __run_hrtimer+0x76/0x1f0 Jun 9 13:32:50 fambox kernel: [ 2778.842744] [<ffffffff810dd9e0>] ? __touch_watchdog+0x30/0x30 Jun 9 13:32:50 fambox kernel: [ 2778.842746] [<ffffffff8101b349>] ? read_tsc+0x9/0x20 Jun 9 13:32:50 fambox kernel: [ 2778.842748] [<ffffffff8107adc3>] hrtimer_interrupt+0xe3/0x200 Jun 9 13:32:50 fambox kernel: [ 2778.842750] [<ffffffff8107af0b>] __hrtimer_peek_ahead_timers.part.26+0x2b/0x30 Jun 9 13:32:50 fambox kernel: [ 2778.842754] [<ffffffff816527c2>] hrtimer_cpu_notify+0x1c3/0x1eb Jun 9 13:32:50 fambox kernel: [ 2778.842757] [<ffffffff81668f5d>] notifier_call_chain+0x4d/0x70 Jun 9 13:32:50 fambox kernel: [ 2778.842758] [<ffffffff8107c54e>] __raw_notifier_call_chain+0xe/0x10 Jun 9 13:32:50 fambox kernel: [ 2778.842761] [<ffffffff81055620>] __cpu_notify+0x20/0x40 Jun 9 13:32:50 fambox kernel: [ 2778.842762] [<ffffffff81055775>] cpu_notify_nofail+0x15/0x20 Jun 9 13:32:50 fambox kernel: [ 2778.842765] [<ffffffff816365ad>] _cpu_down+0x10d/0x270 Jun 9 13:32:50 fambox kernel: [ 2778.842767] [<ffffffff81055927>] disable_nonboot_cpus+0xa7/0x150 Jun 9 13:32:50 fambox kernel: [ 2778.842770] [<ffffffff8109952c>] suspend_devices_and_enter+0x22c/0x310 Jun 9 13:32:50 fambox kernel: [ 2778.842771] [<ffffffff8109979f>] pm_suspend+0x18f/0x1f0 Jun 9 13:32:50 fambox kernel: [ 2778.842773] [<ffffffff810987ae>] state_store+0x8e/0xf0 Jun 9 13:32:50 fambox kernel: [ 2778.842775] [<ffffffff81319aff>] kobj_attr_store+0xf/0x30 Jun 9 13:32:50 fambox kernel: [ 2778.842778] [<ffffffff811eddcf>] sysfs_write_file+0xef/0x170 Jun 9 13:32:50 fambox kernel: [ 2778.842781] [<ffffffff8117ed93>] vfs_write+0xb3/0x180 Jun 9 13:32:50 fambox kernel: [ 2778.842782] [<ffffffff8117f0ba>] sys_write+0x4a/0x90 Jun 9 13:32:50 fambox kernel: [ 2778.842785] [<ffffffff8166d229>] system_call_fastpath+0x16/0x1b Jun 9 13:32:50 fambox kernel: [ 2778.842786] ---[ end trace c32bb8d548ae6e7f ]--- Jun 9 13:32:50 fambox kernel: [ 2778.843016] Extended CMOS year: 2000 Jun 9 13:32:50 fambox kernel: [ 2778.844289] ACPI: Low-level resume complete Jun 9 13:32:50 fambox kernel: [ 2778.844335] PM: Restoring platform NVS memory Jun 9 13:32:50 fambox kernel: [ 2778.845559] Extended CMOS year: 2000 Jun 9 13:32:50 fambox kernel: [ 2778.845604] Enabling non-boot CPUs ... Jun 9 13:32:50 fambox kernel: [ 2778.845694] Booting Node 0 Processor 1 APIC 0x1 Jun 9 13:32:50 fambox kernel: [ 2778.856634] Disabled fast string operations Jun 9 13:32:50 fambox kernel: [ 2778.859002] NMI watchdog: enabled, takes one hw-pmu counter. Jun 9 13:32:50 fambox kernel: [ 2778.859289] CPU1 is up Jun 9 13:32:50 fambox kernel: [ 2778.859424] Booting Node 0 Processor 2 APIC 0x2 Jun 9 13:32:50 fambox kernel: [ 2778.870461] Disabled fast string operations Jun 9 13:32:50 fambox kernel: [ 2778.872826] NMI watchdog: enabled, takes one hw-pmu counter. Jun 9 13:32:50 fambox kernel: [ 2778.873148] CPU2 is up Jun 9 13:32:50 fambox kernel: [ 2778.873243] Booting Node 0 Processor 3 APIC 0x3 Jun 9 13:32:50 fambox kernel: [ 2778.884278] Disabled fast string operations Jun 9 13:32:50 fambox kernel: [ 2778.886641] NMI watchdog: enabled, takes one hw-pmu counter. Jun 9 13:32:50 fambox kernel: [ 2778.886944] CPU3 is up Jun 9 13:32:50 fambox kernel: [ 2778.891780] ACPI: Waking up from system sleep state S3 Jun 9 13:32:50 fambox kernel: [ 2778.943274] PM: noirq resume of devices complete after 2.200 msecs Jun 9 13:32:50 fambox kernel: [ 2778.943379] PM: early resume of devices complete after 0.065 msecs Jun 9 13:32:50 fambox kernel: [ 2778.943451] i915 0000:00:02.0: setting latency timer to 64 Jun 9 13:32:50 fambox kernel: [ 2778.943461] ehci_hcd 0000:00:1a.0: setting latency timer to 64 Jun 9 13:32:50 fambox kernel: [ 2778.943497] ehci_hcd 0000:00:1d.0: setting latency timer to 64 Jun 9 13:32:50 fambox kernel: [ 2778.943509] ahci 0000:00:1f.2: setting latency timer to 64 Jun 9 13:32:50 fambox kernel: [ 2778.943585] r8169 0000:02:00.0: wake-up capability disabled by ACPI Jun 9 13:32:50 fambox kernel: [ 2778.943638] iwlwifi 0000:01:00.0: RF_KILL bit toggled to enable radio. Jun 9 13:32:50 fambox kernel: [ 2778.943651] snd_hda_intel 0000:00:1b.0: irq 50 for MSI/MSI-X Jun 9 13:32:50 fambox kernel: [ 2778.943682] usb usb3: root hub lost power or was reset Jun 9 13:32:50 fambox kernel: [ 2778.943684] usb usb4: root hub lost power or was reset Jun 9 13:32:50 fambox kernel: [ 2778.948575] xhci_hcd 0000:03:00.0: irq 41 for MSI/MSI-X Jun 9 13:32:50 fambox kernel: [ 2778.948581] xhci_hcd 0000:03:00.0: irq 42 for MSI/MSI-X Jun 9 13:32:50 fambox kernel: [ 2778.948586] xhci_hcd 0000:03:00.0: irq 43 for MSI/MSI-X Jun 9 13:32:50 fambox kernel: [ 2778.948592] xhci_hcd 0000:03:00.0: irq 44 for MSI/MSI-X Jun 9 13:32:50 fambox kernel: [ 2778.948597] xhci_hcd 0000:03:00.0: irq 45 for MSI/MSI-X Jun 9 13:32:50 fambox kernel: [ 2779.001863] [drm] Enabling RC6 states: RC6 on, RC6p off, RC6pp off Jun 9 13:32:50 fambox kernel: [ 2779.276755] ata4: SATA link down (SStatus 0 SControl 300) Jun 9 13:32:50 fambox kernel: [ 2779.284757] ata5: SATA link down (SStatus 0 SControl 300) Jun 9 13:32:50 fambox kernel: [ 2779.292760] ata2: SATA link up 3.0 Gbps (SStatus 123 SControl 300) Jun 9 13:32:50 fambox kernel: [ 2779.297042] ata2.00: configured for UDMA/133 Jun 9 13:32:50 fambox kernel: [ 2779.312786] sd 1:0:0:0: [sdb] Starting disk Jun 9 13:32:50 fambox kernel: [ 2779.380807] usb 2-1.4: reset low-speed USB device number 3 using ehci_hcd Jun 9 13:32:50 fambox kernel: [ 2779.732579] usb 1-1.4: reset high-speed USB device number 3 using ehci_hcd Jun 9 13:32:50 fambox kernel: [ 2781.483191] ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300) Jun 9 13:32:50 fambox kernel: [ 2781.485651] ata1.00: configured for UDMA/133 Jun 9 13:32:50 fambox kernel: [ 2781.499271] sd 0:0:0:0: [sda] Starting disk Jun 9 13:32:50 fambox kernel: [ 2781.501362] Extended CMOS year: 2000 Jun 9 13:32:50 fambox kernel: [ 2781.528530] PM: resume of devices complete after 2586.977 msecs Jun 9 13:32:50 fambox kernel: [ 2781.528604] PM: resume devices took 2.588 seconds Jun 9 13:32:50 fambox kernel: [ 2781.528646] PM: Finishing wakeup. Jun 9 13:32:50 fambox kernel: [ 2781.531291] Restarting tasks ... done. Jun 9 13:32:50 fambox kernel: [ 2781.537436] video LNXVIDEO:00: Restoring backlight state ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: WARNING: at arch/x86/kernel/kvmclock.c:127 2012-06-10 15:47 WARNING: at arch/x86/kernel/kvmclock.c:127 Sedat Dilek 2012-06-10 16:08 ` Sedat Dilek @ 2012-06-11 21:47 ` Eric B Munson 2012-06-11 22:07 ` Marcelo Tosatti 1 sibling, 1 reply; 9+ messages in thread From: Eric B Munson @ 2012-06-11 21:47 UTC (permalink / raw) To: sedat.dilek Cc: kvm, LKML, Avi Kivity, Marcelo Tosatti, Glauber Costa, Thomas Gleixner, Ingo Molnar, x86 On Sun, 10 Jun 2012 17:47:24 +0200, Sedat Dilek wrote: > Hi, > > I see the same warning especially when resuming from a suspend (see > timestamps >=30-35sec) between Linux v3.5-rc1..v3.5-rc2. > > $ sudo grep kvmclock.c /var/log/kern.log > Jun 4 21:11:32 fambox kernel: [ 784.037237] WARNING: at > arch/x86/kernel/kvmclock.c:127 > kvm_check_and_clear_guest_paused+0x52/0x60() > Jun 5 20:35:57 fambox kernel: [ 1928.458060] WARNING: at > arch/x86/kernel/kvmclock.c:127 > kvm_check_and_clear_guest_paused+0x52/0x60() > Jun 8 09:35:52 fambox kernel: [ 3290.134637] WARNING: at > arch/x86/kernel/kvmclock.c:127 > kvm_check_and_clear_guest_paused+0x52/0x60() > Jun 8 09:35:52 fambox kernel: [ 3290.238592] WARNING: at > arch/x86/kernel/kvmclock.c:127 > kvm_check_and_clear_guest_paused+0x52/0x60() > Jun 8 12:11:20 fambox kernel: [ 5777.023571] WARNING: at > arch/x86/kernel/kvmclock.c:127 > kvm_check_and_clear_guest_paused+0x52/0x60() > Jun 9 13:32:50 fambox kernel: [ 2778.842695] WARNING: at > arch/x86/kernel/kvmclock.c:127 > kvm_check_and_clear_guest_paused+0x52/0x60() > > From [1]: > > "...The warning itself is not required for the check_and_clear > function and can be removed as far as I am concerned." > > From [2] commit 3b5d56b9317fa7b5407dff1aa7b115bf6cdbd494 ("kvmclock: > Add functions to check if the host has stopped the vm") > ... > +bool kvm_check_and_clear_guest_paused(void) > +{ > + bool ret = false; > + struct pvclock_vcpu_time_info *src; > + > + /* > + * per_cpu() is safe here because this function is only > called from > + * timer functions where preemption is already disabled. > + */ > + WARN_ON(!in_atomic()); > + src = &__get_cpu_var(hv_clock); > + if ((src->flags & PVCLOCK_GUEST_STOPPED) != 0) { > + __this_cpu_and(hv_clock.flags, > ~PVCLOCK_GUEST_STOPPED); > + ret = true; > + } > + > + return ret; > +} > +EXPORT_SYMBOL_GPL(kvm_check_and_clear_guest_paused); > + > ... > ( The export macro was dropped in a followup commit. ) > > So you mean "WARN_ON(!in_atomic());" can be deleted? When I wrote the patch originally, I was under the (incorrect) assumption that the watch dog was only ever reset in an atomic context. Given that this is not the case, the warning can be removed. Though before that happens, I have a question: if this is called outside of an atomic context, is the use of __get_cpu_Var() and __this_cpu_and() invalid? Thanks, Eric ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: WARNING: at arch/x86/kernel/kvmclock.c:127 2012-06-11 21:47 ` Eric B Munson @ 2012-06-11 22:07 ` Marcelo Tosatti 2012-06-11 22:10 ` Eric B Munson 0 siblings, 1 reply; 9+ messages in thread From: Marcelo Tosatti @ 2012-06-11 22:07 UTC (permalink / raw) To: Eric B Munson Cc: sedat.dilek, kvm, LKML, Avi Kivity, Glauber Costa, Thomas Gleixner, Ingo Molnar, x86 On Mon, Jun 11, 2012 at 05:47:00PM -0400, Eric B Munson wrote: > On Sun, 10 Jun 2012 17:47:24 +0200, Sedat Dilek wrote: > >Hi, > > > >I see the same warning especially when resuming from a suspend (see > >timestamps >=30-35sec) between Linux v3.5-rc1..v3.5-rc2. > > > >$ sudo grep kvmclock.c /var/log/kern.log > >Jun 4 21:11:32 fambox kernel: [ 784.037237] WARNING: at > >arch/x86/kernel/kvmclock.c:127 > >kvm_check_and_clear_guest_paused+0x52/0x60() > >Jun 5 20:35:57 fambox kernel: [ 1928.458060] WARNING: at > >arch/x86/kernel/kvmclock.c:127 > >kvm_check_and_clear_guest_paused+0x52/0x60() > >Jun 8 09:35:52 fambox kernel: [ 3290.134637] WARNING: at > >arch/x86/kernel/kvmclock.c:127 > >kvm_check_and_clear_guest_paused+0x52/0x60() > >Jun 8 09:35:52 fambox kernel: [ 3290.238592] WARNING: at > >arch/x86/kernel/kvmclock.c:127 > >kvm_check_and_clear_guest_paused+0x52/0x60() > >Jun 8 12:11:20 fambox kernel: [ 5777.023571] WARNING: at > >arch/x86/kernel/kvmclock.c:127 > >kvm_check_and_clear_guest_paused+0x52/0x60() > >Jun 9 13:32:50 fambox kernel: [ 2778.842695] WARNING: at > >arch/x86/kernel/kvmclock.c:127 > >kvm_check_and_clear_guest_paused+0x52/0x60() > > > >From [1]: > > > >"...The warning itself is not required for the check_and_clear > >function and can be removed as far as I am concerned." > > > >From [2] commit 3b5d56b9317fa7b5407dff1aa7b115bf6cdbd494 ("kvmclock: > >Add functions to check if the host has stopped the vm") > >... > >+bool kvm_check_and_clear_guest_paused(void) > >+{ > >+ bool ret = false; > >+ struct pvclock_vcpu_time_info *src; > >+ > >+ /* > >+ * per_cpu() is safe here because this function is only > >called from > >+ * timer functions where preemption is already disabled. > >+ */ > >+ WARN_ON(!in_atomic()); > >+ src = &__get_cpu_var(hv_clock); > >+ if ((src->flags & PVCLOCK_GUEST_STOPPED) != 0) { > >+ __this_cpu_and(hv_clock.flags, > >~PVCLOCK_GUEST_STOPPED); > >+ ret = true; > >+ } > >+ > >+ return ret; > >+} > >+EXPORT_SYMBOL_GPL(kvm_check_and_clear_guest_paused); > >+ > >... > >( The export macro was dropped in a followup commit. ) > > > >So you mean "WARN_ON(!in_atomic());" can be deleted? Yes. > When I wrote the patch originally, I was under the (incorrect) > assumption that the watch dog was only ever reset in an atomic > context. Given that this is not the case, the warning can be > removed. Though before that happens, I have a question: if this is > called outside of an atomic context, is the use of __get_cpu_Var() > and __this_cpu_and() invalid? It remains valid because its called with interrupts disabled (see migrate_hrtimers). ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: WARNING: at arch/x86/kernel/kvmclock.c:127 2012-06-11 22:07 ` Marcelo Tosatti @ 2012-06-11 22:10 ` Eric B Munson 2012-06-12 2:13 ` Marcelo Tosatti 0 siblings, 1 reply; 9+ messages in thread From: Eric B Munson @ 2012-06-11 22:10 UTC (permalink / raw) To: Marcelo Tosatti Cc: sedat.dilek, kvm, LKML, Avi Kivity, Thomas Gleixner, Ingo Molnar, x86 On Mon, 11 Jun 2012 19:07:19 -0300, Marcelo Tosatti wrote: > On Mon, Jun 11, 2012 at 05:47:00PM -0400, Eric B Munson wrote: >> On Sun, 10 Jun 2012 17:47:24 +0200, Sedat Dilek wrote: >> >Hi, >> > >> >I see the same warning especially when resuming from a suspend (see >> >timestamps >=30-35sec) between Linux v3.5-rc1..v3.5-rc2. >> > >> >$ sudo grep kvmclock.c /var/log/kern.log >> >Jun 4 21:11:32 fambox kernel: [ 784.037237] WARNING: at >> >arch/x86/kernel/kvmclock.c:127 >> >kvm_check_and_clear_guest_paused+0x52/0x60() >> >Jun 5 20:35:57 fambox kernel: [ 1928.458060] WARNING: at >> >arch/x86/kernel/kvmclock.c:127 >> >kvm_check_and_clear_guest_paused+0x52/0x60() >> >Jun 8 09:35:52 fambox kernel: [ 3290.134637] WARNING: at >> >arch/x86/kernel/kvmclock.c:127 >> >kvm_check_and_clear_guest_paused+0x52/0x60() >> >Jun 8 09:35:52 fambox kernel: [ 3290.238592] WARNING: at >> >arch/x86/kernel/kvmclock.c:127 >> >kvm_check_and_clear_guest_paused+0x52/0x60() >> >Jun 8 12:11:20 fambox kernel: [ 5777.023571] WARNING: at >> >arch/x86/kernel/kvmclock.c:127 >> >kvm_check_and_clear_guest_paused+0x52/0x60() >> >Jun 9 13:32:50 fambox kernel: [ 2778.842695] WARNING: at >> >arch/x86/kernel/kvmclock.c:127 >> >kvm_check_and_clear_guest_paused+0x52/0x60() >> > >> >From [1]: >> > >> >"...The warning itself is not required for the check_and_clear >> >function and can be removed as far as I am concerned." >> > >> >From [2] commit 3b5d56b9317fa7b5407dff1aa7b115bf6cdbd494 >> ("kvmclock: >> >Add functions to check if the host has stopped the vm") >> >... >> >+bool kvm_check_and_clear_guest_paused(void) >> >+{ >> >+ bool ret = false; >> >+ struct pvclock_vcpu_time_info *src; >> >+ >> >+ /* >> >+ * per_cpu() is safe here because this function is only >> >called from >> >+ * timer functions where preemption is already disabled. >> >+ */ >> >+ WARN_ON(!in_atomic()); >> >+ src = &__get_cpu_var(hv_clock); >> >+ if ((src->flags & PVCLOCK_GUEST_STOPPED) != 0) { >> >+ __this_cpu_and(hv_clock.flags, >> >~PVCLOCK_GUEST_STOPPED); >> >+ ret = true; >> >+ } >> >+ >> >+ return ret; >> >+} >> >+EXPORT_SYMBOL_GPL(kvm_check_and_clear_guest_paused); >> >+ >> >... >> >( The export macro was dropped in a followup commit. ) >> > >> >So you mean "WARN_ON(!in_atomic());" can be deleted? > > Yes. > >> When I wrote the patch originally, I was under the (incorrect) >> assumption that the watch dog was only ever reset in an atomic >> context. Given that this is not the case, the warning can be >> removed. Though before that happens, I have a question: if this is >> called outside of an atomic context, is the use of __get_cpu_Var() >> and __this_cpu_and() invalid? > > It remains valid because its called with interrupts > disabled (see migrate_hrtimers). Thank you, that was my assumption but I wanted to confirm. I do not have access to my working kernel tree and won't for 2 days. I can't get to the patch until late Wednesday so if someone else wants to work that up I'd appreciate it. Thanks, Eric ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: WARNING: at arch/x86/kernel/kvmclock.c:127 2012-06-11 22:10 ` Eric B Munson @ 2012-06-12 2:13 ` Marcelo Tosatti 2012-06-12 8:48 ` Sedat Dilek 2012-06-15 9:35 ` Sedat Dilek 0 siblings, 2 replies; 9+ messages in thread From: Marcelo Tosatti @ 2012-06-12 2:13 UTC (permalink / raw) To: Eric B Munson Cc: sedat.dilek, kvm, LKML, Avi Kivity, Thomas Gleixner, Ingo Molnar, x86 On Mon, Jun 11, 2012 at 06:10:34PM -0400, Eric B Munson wrote: > On Mon, 11 Jun 2012 19:07:19 -0300, Marcelo Tosatti wrote: > >On Mon, Jun 11, 2012 at 05:47:00PM -0400, Eric B Munson wrote: > >>On Sun, 10 Jun 2012 17:47:24 +0200, Sedat Dilek wrote: > >>>Hi, > >>> > >>>I see the same warning especially when resuming from a suspend (see > >>>timestamps >=30-35sec) between Linux v3.5-rc1..v3.5-rc2. > >>> > >>>$ sudo grep kvmclock.c /var/log/kern.log > >>>Jun 4 21:11:32 fambox kernel: [ 784.037237] WARNING: at > >>>arch/x86/kernel/kvmclock.c:127 > >>>kvm_check_and_clear_guest_paused+0x52/0x60() > >>>Jun 5 20:35:57 fambox kernel: [ 1928.458060] WARNING: at > >>>arch/x86/kernel/kvmclock.c:127 > >>>kvm_check_and_clear_guest_paused+0x52/0x60() > >>>Jun 8 09:35:52 fambox kernel: [ 3290.134637] WARNING: at > >>>arch/x86/kernel/kvmclock.c:127 > >>>kvm_check_and_clear_guest_paused+0x52/0x60() > >>>Jun 8 09:35:52 fambox kernel: [ 3290.238592] WARNING: at > >>>arch/x86/kernel/kvmclock.c:127 > >>>kvm_check_and_clear_guest_paused+0x52/0x60() > >>>Jun 8 12:11:20 fambox kernel: [ 5777.023571] WARNING: at > >>>arch/x86/kernel/kvmclock.c:127 > >>>kvm_check_and_clear_guest_paused+0x52/0x60() > >>>Jun 9 13:32:50 fambox kernel: [ 2778.842695] WARNING: at > >>>arch/x86/kernel/kvmclock.c:127 > >>>kvm_check_and_clear_guest_paused+0x52/0x60() > >>> > >>>From [1]: > >>> > >>>"...The warning itself is not required for the check_and_clear > >>>function and can be removed as far as I am concerned." > >>> > >>>From [2] commit 3b5d56b9317fa7b5407dff1aa7b115bf6cdbd494 > >>("kvmclock: > >>>Add functions to check if the host has stopped the vm") > >>>... > >>>+bool kvm_check_and_clear_guest_paused(void) > >>>+{ > >>>+ bool ret = false; > >>>+ struct pvclock_vcpu_time_info *src; > >>>+ > >>>+ /* > >>>+ * per_cpu() is safe here because this function is only > >>>called from > >>>+ * timer functions where preemption is already disabled. > >>>+ */ > >>>+ WARN_ON(!in_atomic()); > >>>+ src = &__get_cpu_var(hv_clock); > >>>+ if ((src->flags & PVCLOCK_GUEST_STOPPED) != 0) { > >>>+ __this_cpu_and(hv_clock.flags, > >>>~PVCLOCK_GUEST_STOPPED); > >>>+ ret = true; > >>>+ } > >>>+ > >>>+ return ret; > >>>+} > >>>+EXPORT_SYMBOL_GPL(kvm_check_and_clear_guest_paused); > >>>+ > >>>... > >>>( The export macro was dropped in a followup commit. ) > >>> > >>>So you mean "WARN_ON(!in_atomic());" can be deleted? > > > >Yes. > > > >>When I wrote the patch originally, I was under the (incorrect) > >>assumption that the watch dog was only ever reset in an atomic > >>context. Given that this is not the case, the warning can be > >>removed. Though before that happens, I have a question: if this is > >>called outside of an atomic context, is the use of __get_cpu_Var() > >>and __this_cpu_and() invalid? > > > >It remains valid because its called with interrupts > >disabled (see migrate_hrtimers). > > Thank you, that was my assumption but I wanted to confirm. I do not > have access to my working kernel tree and won't for 2 days. I can't > get to the patch until late Wednesday so if someone else wants to > work that up I'd appreciate it. > > Thanks, > Eric Removed the warning in kvm.git master. ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: WARNING: at arch/x86/kernel/kvmclock.c:127 2012-06-12 2:13 ` Marcelo Tosatti @ 2012-06-12 8:48 ` Sedat Dilek 2012-06-15 9:35 ` Sedat Dilek 1 sibling, 0 replies; 9+ messages in thread From: Sedat Dilek @ 2012-06-12 8:48 UTC (permalink / raw) To: Marcelo Tosatti Cc: Eric B Munson, kvm, LKML, Avi Kivity, Thomas Gleixner, Ingo Molnar, x86 On Tue, Jun 12, 2012 at 4:13 AM, Marcelo Tosatti <mtosatti@redhat.com> wrote: > On Mon, Jun 11, 2012 at 06:10:34PM -0400, Eric B Munson wrote: >> On Mon, 11 Jun 2012 19:07:19 -0300, Marcelo Tosatti wrote: >> >On Mon, Jun 11, 2012 at 05:47:00PM -0400, Eric B Munson wrote: >> >>On Sun, 10 Jun 2012 17:47:24 +0200, Sedat Dilek wrote: >> >>>Hi, >> >>> >> >>>I see the same warning especially when resuming from a suspend (see >> >>>timestamps >=30-35sec) between Linux v3.5-rc1..v3.5-rc2. >> >>> >> >>>$ sudo grep kvmclock.c /var/log/kern.log >> >>>Jun 4 21:11:32 fambox kernel: [ 784.037237] WARNING: at >> >>>arch/x86/kernel/kvmclock.c:127 >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() >> >>>Jun 5 20:35:57 fambox kernel: [ 1928.458060] WARNING: at >> >>>arch/x86/kernel/kvmclock.c:127 >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() >> >>>Jun 8 09:35:52 fambox kernel: [ 3290.134637] WARNING: at >> >>>arch/x86/kernel/kvmclock.c:127 >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() >> >>>Jun 8 09:35:52 fambox kernel: [ 3290.238592] WARNING: at >> >>>arch/x86/kernel/kvmclock.c:127 >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() >> >>>Jun 8 12:11:20 fambox kernel: [ 5777.023571] WARNING: at >> >>>arch/x86/kernel/kvmclock.c:127 >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() >> >>>Jun 9 13:32:50 fambox kernel: [ 2778.842695] WARNING: at >> >>>arch/x86/kernel/kvmclock.c:127 >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() >> >>> >> >>>From [1]: >> >>> >> >>>"...The warning itself is not required for the check_and_clear >> >>>function and can be removed as far as I am concerned." >> >>> >> >>>From [2] commit 3b5d56b9317fa7b5407dff1aa7b115bf6cdbd494 >> >>("kvmclock: >> >>>Add functions to check if the host has stopped the vm") >> >>>... >> >>>+bool kvm_check_and_clear_guest_paused(void) >> >>>+{ >> >>>+ bool ret = false; >> >>>+ struct pvclock_vcpu_time_info *src; >> >>>+ >> >>>+ /* >> >>>+ * per_cpu() is safe here because this function is only >> >>>called from >> >>>+ * timer functions where preemption is already disabled. >> >>>+ */ >> >>>+ WARN_ON(!in_atomic()); >> >>>+ src = &__get_cpu_var(hv_clock); >> >>>+ if ((src->flags & PVCLOCK_GUEST_STOPPED) != 0) { >> >>>+ __this_cpu_and(hv_clock.flags, >> >>>~PVCLOCK_GUEST_STOPPED); >> >>>+ ret = true; >> >>>+ } >> >>>+ >> >>>+ return ret; >> >>>+} >> >>>+EXPORT_SYMBOL_GPL(kvm_check_and_clear_guest_paused); >> >>>+ >> >>>... >> >>>( The export macro was dropped in a followup commit. ) >> >>> >> >>>So you mean "WARN_ON(!in_atomic());" can be deleted? >> > >> >Yes. >> > >> >>When I wrote the patch originally, I was under the (incorrect) >> >>assumption that the watch dog was only ever reset in an atomic >> >>context. Given that this is not the case, the warning can be >> >>removed. Though before that happens, I have a question: if this is >> >>called outside of an atomic context, is the use of __get_cpu_Var() >> >>and __this_cpu_and() invalid? >> > >> >It remains valid because its called with interrupts >> >disabled (see migrate_hrtimers). >> >> Thank you, that was my assumption but I wanted to confirm. I do not >> have access to my working kernel tree and won't for 2 days. I can't >> get to the patch until late Wednesday so if someone else wants to >> work that up I'd appreciate it. >> >> Thanks, >> Eric > > Removed the warning in kvm.git master. > I am reading your email approx. 6hrs after you have sent it and I do NOT see the patch in [1]. Can you please send me the patch as an attachment, Thanks! - Sedat - [1] http://git.kernel.org/?p=virt/kvm/kvm.git;a=shortlog;h=refs/heads/master ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: WARNING: at arch/x86/kernel/kvmclock.c:127 2012-06-12 2:13 ` Marcelo Tosatti 2012-06-12 8:48 ` Sedat Dilek @ 2012-06-15 9:35 ` Sedat Dilek 2012-06-15 11:08 ` Marcelo Tosatti 1 sibling, 1 reply; 9+ messages in thread From: Sedat Dilek @ 2012-06-15 9:35 UTC (permalink / raw) To: Marcelo Tosatti Cc: Eric B Munson, kvm, LKML, Avi Kivity, Thomas Gleixner, Ingo Molnar, x86 On Tue, Jun 12, 2012 at 4:13 AM, Marcelo Tosatti <mtosatti@redhat.com> wrote: > On Mon, Jun 11, 2012 at 06:10:34PM -0400, Eric B Munson wrote: >> On Mon, 11 Jun 2012 19:07:19 -0300, Marcelo Tosatti wrote: >> >On Mon, Jun 11, 2012 at 05:47:00PM -0400, Eric B Munson wrote: >> >>On Sun, 10 Jun 2012 17:47:24 +0200, Sedat Dilek wrote: >> >>>Hi, >> >>> >> >>>I see the same warning especially when resuming from a suspend (see >> >>>timestamps >=30-35sec) between Linux v3.5-rc1..v3.5-rc2. >> >>> >> >>>$ sudo grep kvmclock.c /var/log/kern.log >> >>>Jun 4 21:11:32 fambox kernel: [ 784.037237] WARNING: at >> >>>arch/x86/kernel/kvmclock.c:127 >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() >> >>>Jun 5 20:35:57 fambox kernel: [ 1928.458060] WARNING: at >> >>>arch/x86/kernel/kvmclock.c:127 >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() >> >>>Jun 8 09:35:52 fambox kernel: [ 3290.134637] WARNING: at >> >>>arch/x86/kernel/kvmclock.c:127 >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() >> >>>Jun 8 09:35:52 fambox kernel: [ 3290.238592] WARNING: at >> >>>arch/x86/kernel/kvmclock.c:127 >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() >> >>>Jun 8 12:11:20 fambox kernel: [ 5777.023571] WARNING: at >> >>>arch/x86/kernel/kvmclock.c:127 >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() >> >>>Jun 9 13:32:50 fambox kernel: [ 2778.842695] WARNING: at >> >>>arch/x86/kernel/kvmclock.c:127 >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() >> >>> >> >>>From [1]: >> >>> >> >>>"...The warning itself is not required for the check_and_clear >> >>>function and can be removed as far as I am concerned." >> >>> >> >>>From [2] commit 3b5d56b9317fa7b5407dff1aa7b115bf6cdbd494 >> >>("kvmclock: >> >>>Add functions to check if the host has stopped the vm") >> >>>... >> >>>+bool kvm_check_and_clear_guest_paused(void) >> >>>+{ >> >>>+ bool ret = false; >> >>>+ struct pvclock_vcpu_time_info *src; >> >>>+ >> >>>+ /* >> >>>+ * per_cpu() is safe here because this function is only >> >>>called from >> >>>+ * timer functions where preemption is already disabled. >> >>>+ */ >> >>>+ WARN_ON(!in_atomic()); >> >>>+ src = &__get_cpu_var(hv_clock); >> >>>+ if ((src->flags & PVCLOCK_GUEST_STOPPED) != 0) { >> >>>+ __this_cpu_and(hv_clock.flags, >> >>>~PVCLOCK_GUEST_STOPPED); >> >>>+ ret = true; >> >>>+ } >> >>>+ >> >>>+ return ret; >> >>>+} >> >>>+EXPORT_SYMBOL_GPL(kvm_check_and_clear_guest_paused); >> >>>+ >> >>>... >> >>>( The export macro was dropped in a followup commit. ) >> >>> >> >>>So you mean "WARN_ON(!in_atomic());" can be deleted? >> > >> >Yes. >> > >> >>When I wrote the patch originally, I was under the (incorrect) >> >>assumption that the watch dog was only ever reset in an atomic >> >>context. Given that this is not the case, the warning can be >> >>removed. Though before that happens, I have a question: if this is >> >>called outside of an atomic context, is the use of __get_cpu_Var() >> >>and __this_cpu_and() invalid? >> > >> >It remains valid because its called with interrupts >> >disabled (see migrate_hrtimers). >> >> Thank you, that was my assumption but I wanted to confirm. I do not >> have access to my working kernel tree and won't for 2 days. I can't >> get to the patch until late Wednesday so if someone else wants to >> work that up I'd appreciate it. >> >> Thanks, >> Eric > > Removed the warning in kvm.git master. > Unfortunately, I was busy this week with familiar affairs. I have seen too late that Marcelo sent me the patch via PM. Marcelo, could you please give credits to people next time, Thanks. I am missing sth. like this in the commit now in Linus upstream GIT... Reported-by: Frank Arnold <frank.arnold@...> (hidden - me reads KVM ML offline) Reported-by: Sedat Dilek <sedat.dilek@gmail.com> (lazy guy - could have reported earlier) Acked-by or Reviewed-by: Eric B Munson <emunson@mgebm.net> - Sedat - [1] http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=commitdiff;h=e32025a56403df4386cd61a741c0a36afe79ae8a ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: WARNING: at arch/x86/kernel/kvmclock.c:127 2012-06-15 9:35 ` Sedat Dilek @ 2012-06-15 11:08 ` Marcelo Tosatti 0 siblings, 0 replies; 9+ messages in thread From: Marcelo Tosatti @ 2012-06-15 11:08 UTC (permalink / raw) To: sedat.dilek Cc: Eric B Munson, kvm, LKML, Avi Kivity, Thomas Gleixner, Ingo Molnar, x86 On Fri, Jun 15, 2012 at 11:35:58AM +0200, Sedat Dilek wrote: > On Tue, Jun 12, 2012 at 4:13 AM, Marcelo Tosatti <mtosatti@redhat.com> wrote: > > On Mon, Jun 11, 2012 at 06:10:34PM -0400, Eric B Munson wrote: > >> On Mon, 11 Jun 2012 19:07:19 -0300, Marcelo Tosatti wrote: > >> >On Mon, Jun 11, 2012 at 05:47:00PM -0400, Eric B Munson wrote: > >> >>On Sun, 10 Jun 2012 17:47:24 +0200, Sedat Dilek wrote: > >> >>>Hi, > >> >>> > >> >>>I see the same warning especially when resuming from a suspend (see > >> >>>timestamps >=30-35sec) between Linux v3.5-rc1..v3.5-rc2. > >> >>> > >> >>>$ sudo grep kvmclock.c /var/log/kern.log > >> >>>Jun 4 21:11:32 fambox kernel: [ 784.037237] WARNING: at > >> >>>arch/x86/kernel/kvmclock.c:127 > >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() > >> >>>Jun 5 20:35:57 fambox kernel: [ 1928.458060] WARNING: at > >> >>>arch/x86/kernel/kvmclock.c:127 > >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() > >> >>>Jun 8 09:35:52 fambox kernel: [ 3290.134637] WARNING: at > >> >>>arch/x86/kernel/kvmclock.c:127 > >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() > >> >>>Jun 8 09:35:52 fambox kernel: [ 3290.238592] WARNING: at > >> >>>arch/x86/kernel/kvmclock.c:127 > >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() > >> >>>Jun 8 12:11:20 fambox kernel: [ 5777.023571] WARNING: at > >> >>>arch/x86/kernel/kvmclock.c:127 > >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() > >> >>>Jun 9 13:32:50 fambox kernel: [ 2778.842695] WARNING: at > >> >>>arch/x86/kernel/kvmclock.c:127 > >> >>>kvm_check_and_clear_guest_paused+0x52/0x60() > >> >>> > >> >>>From [1]: > >> >>> > >> >>>"...The warning itself is not required for the check_and_clear > >> >>>function and can be removed as far as I am concerned." > >> >>> > >> >>>From [2] commit 3b5d56b9317fa7b5407dff1aa7b115bf6cdbd494 > >> >>("kvmclock: > >> >>>Add functions to check if the host has stopped the vm") > >> >>>... > >> >>>+bool kvm_check_and_clear_guest_paused(void) > >> >>>+{ > >> >>>+ bool ret = false; > >> >>>+ struct pvclock_vcpu_time_info *src; > >> >>>+ > >> >>>+ /* > >> >>>+ * per_cpu() is safe here because this function is only > >> >>>called from > >> >>>+ * timer functions where preemption is already disabled. > >> >>>+ */ > >> >>>+ WARN_ON(!in_atomic()); > >> >>>+ src = &__get_cpu_var(hv_clock); > >> >>>+ if ((src->flags & PVCLOCK_GUEST_STOPPED) != 0) { > >> >>>+ __this_cpu_and(hv_clock.flags, > >> >>>~PVCLOCK_GUEST_STOPPED); > >> >>>+ ret = true; > >> >>>+ } > >> >>>+ > >> >>>+ return ret; > >> >>>+} > >> >>>+EXPORT_SYMBOL_GPL(kvm_check_and_clear_guest_paused); > >> >>>+ > >> >>>... > >> >>>( The export macro was dropped in a followup commit. ) > >> >>> > >> >>>So you mean "WARN_ON(!in_atomic());" can be deleted? > >> > > >> >Yes. > >> > > >> >>When I wrote the patch originally, I was under the (incorrect) > >> >>assumption that the watch dog was only ever reset in an atomic > >> >>context. Given that this is not the case, the warning can be > >> >>removed. Though before that happens, I have a question: if this is > >> >>called outside of an atomic context, is the use of __get_cpu_Var() > >> >>and __this_cpu_and() invalid? > >> > > >> >It remains valid because its called with interrupts > >> >disabled (see migrate_hrtimers). > >> > >> Thank you, that was my assumption but I wanted to confirm. I do not > >> have access to my working kernel tree and won't for 2 days. I can't > >> get to the patch until late Wednesday so if someone else wants to > >> work that up I'd appreciate it. > >> > >> Thanks, > >> Eric > > > > Removed the warning in kvm.git master. > > > > Unfortunately, I was busy this week with familiar affairs. > I have seen too late that Marcelo sent me the patch via PM. > > Marcelo, could you please give credits to people next time, Thanks. > > I am missing sth. like this in the commit now in Linus upstream GIT... > > Reported-by: Frank Arnold <frank.arnold@...> (hidden - me reads KVM ML offline) > Reported-by: Sedat Dilek <sedat.dilek@gmail.com> (lazy guy - could > have reported earlier) > Acked-by or Reviewed-by: Eric B Munson <emunson@mgebm.net> > > - Sedat - > > [1] http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=commitdiff;h=e32025a56403df4386cd61a741c0a36afe79ae8a Sedat, You are right, sorry for not giving appropriate credit. Unfortunately we cannot change the commit message now. Thanks anyway for reporting. ^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2012-06-15 13:45 UTC | newest] Thread overview: 9+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2012-06-10 15:47 WARNING: at arch/x86/kernel/kvmclock.c:127 Sedat Dilek 2012-06-10 16:08 ` Sedat Dilek 2012-06-11 21:47 ` Eric B Munson 2012-06-11 22:07 ` Marcelo Tosatti 2012-06-11 22:10 ` Eric B Munson 2012-06-12 2:13 ` Marcelo Tosatti 2012-06-12 8:48 ` Sedat Dilek 2012-06-15 9:35 ` Sedat Dilek 2012-06-15 11:08 ` Marcelo Tosatti
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox