search for: khugepag

...le. > > > Please test with it. > > > this is what I'm getting with the [zsmalloc: keep first object offset in struct page] > applied: "count:0 mapcount:-127". which may be not related to zsmalloc at this point. > > kernel: BUG: Bad page state in process khugepaged pfn:101db8 > kernel: page:ffffea0004076e00 count:0 mapcount:-127 mapping: (null) index:0x1 Hm, it seems double free. It doen't happen if you disable zram? IOW, it seems to be related zsmalloc migration? How easy can you reprodcue it? Could you bisect it? > kernel: flags: 0...

...le. > > > Please test with it. > > > this is what I'm getting with the [zsmalloc: keep first object offset in struct page] > applied: "count:0 mapcount:-127". which may be not related to zsmalloc at this point. > > kernel: BUG: Bad page state in process khugepaged pfn:101db8 > kernel: page:ffffea0004076e00 count:0 mapcount:-127 mapping: (null) index:0x1 Hm, it seems double free. It doen't happen if you disable zram? IOW, it seems to be related zsmalloc migration? How easy can you reprodcue it? Could you bisect it? > kernel: flags: 0...

...d_hda_codec_realtek snd_hda_codec_generic snd_hda_intel snd_hda_codec snd_hda_core acpi_cpufreq snd_pcm snd_timer snd soundcore lpc_ich mfd_core processor sch_fq_codel sd_mod hid_generic usbhid hid ahci libahci libata ehci_pci ehci_hcd scsi_mod usbcore usb_common [ 315.146785] CPU: 3 PID: 38 Comm: khugepaged Not tainted 4.7.0-rc3-next-20160614-dbg-00004-ga1c2cbc-dirty #488 [ 315.146841] task: ffff8800bfaf2900 ti: ffff880112468000 task.ti: ffff880112468000 [ 315.146859] RIP: 0010:[<ffffffffa02c413d>] [<ffffffffa02c413d>] zs_page_migrate+0x355/0xaa0 [zsmalloc] [ 315.146892] RSP: 0000:ff...

...ltek snd_hda_codec_generic snd_hda_intel snd_hda_codec snd_hda_core acpi_cpufreq snd_pcm snd_timer snd soundcore lpc_ich mfd_core processor sch_fq_codel sd_mod hid_generic usbhid hid ahci libahci libata ehci_pci ehci_hcd scsi_mod usbcore usb_common > > > [ 315.146785] CPU: 3 PID: 38 Comm: khugepaged Not tainted 4.7.0-rc3-next-20160614-dbg-00004-ga1c2cbc-dirty #488 > > > [ 315.146841] task: ffff8800bfaf2900 ti: ffff880112468000 task.ti: ffff880112468000 > > > [ 315.146859] RIP: 0010:[<ffffffffa02c413d>] [<ffffffffa02c413d>] zs_page_migrate+0x355/0xaa0 [zsmallo...

...ltek snd_hda_codec_generic snd_hda_intel snd_hda_codec snd_hda_core acpi_cpufreq snd_pcm snd_timer snd soundcore lpc_ich mfd_core processor sch_fq_codel sd_mod hid_generic usbhid hid ahci libahci libata ehci_pci ehci_hcd scsi_mod usbcore usb_common > > > [ 315.146785] CPU: 3 PID: 38 Comm: khugepaged Not tainted 4.7.0-rc3-next-20160614-dbg-00004-ga1c2cbc-dirty #488 > > > [ 315.146841] task: ffff8800bfaf2900 ti: ffff880112468000 task.ti: ffff880112468000 > > > [ 315.146859] RIP: 0010:[<ffffffffa02c413d>] [<ffffffffa02c413d>] zs_page_migrate+0x355/0xaa0 [zsmallo...

...e: > [..] > > > this is what I'm getting with the [zsmalloc: keep first object offset in struct page] > > > applied: "count:0 mapcount:-127". which may be not related to zsmalloc at this point. > > > > > > kernel: BUG: Bad page state in process khugepaged pfn:101db8 > > > kernel: page:ffffea0004076e00 count:0 mapcount:-127 mapping: (null) index:0x1 > > > > Hm, it seems double free. > > > > It doen't happen if you disable zram? IOW, it seems to be related > > zsmalloc migration? > > need...

...e: > [..] > > > this is what I'm getting with the [zsmalloc: keep first object offset in struct page] > > > applied: "count:0 mapcount:-127". which may be not related to zsmalloc at this point. > > > > > > kernel: BUG: Bad page state in process khugepaged pfn:101db8 > > > kernel: page:ffffea0004076e00 count:0 mapcount:-127 mapping: (null) index:0x1 > > > > Hm, it seems double free. > > > > It doen't happen if you disable zram? IOW, it seems to be related > > zsmalloc migration? > > need...

...own I see these messages in /var/log/messages (this is hand typed, as I can't copy/paste or send email from this system, although for a change I do have remote access via 2 VPN hops): kernel: BUG: soft lockup - CPU#5 stuck for 67s! [khungtaskd] kernel: BUG: soft lockup - CPU#7 stuck for 67s! [khugepaged] These messages started appearing today around 7 AM, which is when users started reporting the slowdown. They are still occurring periodically and the system is still slow. Are these messages caused by the slowdown or are they the reason for it? Are they useful in tracking down the problem?

...13:47), Minchan Kim wrote: [..] > > this is what I'm getting with the [zsmalloc: keep first object offset in struct page] > > applied: "count:0 mapcount:-127". which may be not related to zsmalloc at this point. > > > > kernel: BUG: Bad page state in process khugepaged pfn:101db8 > > kernel: page:ffffea0004076e00 count:0 mapcount:-127 mapping: (null) index:0x1 > > Hm, it seems double free. > > It doen't happen if you disable zram? IOW, it seems to be related > zsmalloc migration? need to test more, can't confidently ans...

...y get_first_obj_offset. sounds reasonable. > Please test with it. this is what I'm getting with the [zsmalloc: keep first object offset in struct page] applied: "count:0 mapcount:-127". which may be not related to zsmalloc at this point. kernel: BUG: Bad page state in process khugepaged pfn:101db8 kernel: page:ffffea0004076e00 count:0 mapcount:-127 mapping: (null) index:0x1 kernel: flags: 0x8000000000000000() kernel: page dumped because: nonzero mapcount kernel: Modules linked in: lzo zram zsmalloc mousedev coretemp hwmon crc32c_intel snd_hda_codec_realtek i2c_i801 snd...

..._codec_realtek snd_hda_codec_generic snd_hda_intel snd_hda_codec snd_hda_core acpi_cpufreq snd_pcm snd_timer snd soundcore lpc_ich mfd_core processor sch_fq_codel sd_mod hid_generic usbhid hid ahci libahci libata ehci_pci ehci_hcd scsi_mod usbcore usb_common > [ 315.146785] CPU: 3 PID: 38 Comm: khugepaged Not tainted 4.7.0-rc3-next-20160614-dbg-00004-ga1c2cbc-dirty #488 > [ 315.146841] task: ffff8800bfaf2900 ti: ffff880112468000 task.ti: ffff880112468000 > [ 315.146859] RIP: 0010:[<ffffffffa02c413d>] [<ffffffffa02c413d>] zs_page_migrate+0x355/0xaa0 [zsmalloc] Thanks for the r...

..._codec_realtek snd_hda_codec_generic snd_hda_intel snd_hda_codec snd_hda_core acpi_cpufreq snd_pcm snd_timer snd soundcore lpc_ich mfd_core processor sch_fq_codel sd_mod hid_generic usbhid hid ahci libahci libata ehci_pci ehci_hcd scsi_mod usbcore usb_common > [ 315.146785] CPU: 3 PID: 38 Comm: khugepaged Not tainted 4.7.0-rc3-next-20160614-dbg-00004-ga1c2cbc-dirty #488 > [ 315.146841] task: ffff8800bfaf2900 ti: ffff880112468000 task.ti: ffff880112468000 > [ 315.146859] RIP: 0010:[<ffffffffa02c413d>] [<ffffffffa02c413d>] zs_page_migrate+0x355/0xaa0 [zsmalloc] Thanks for the r...

...is is what I'm getting with the [zsmalloc: keep first object offset in struct page] > > > > > applied: "count:0 mapcount:-127". which may be not related to zsmalloc at this point. > > > > > > > > > > kernel: BUG: Bad page state in process khugepaged pfn:101db8 > > > > > kernel: page:ffffea0004076e00 count:0 mapcount:-127 mapping: (null) index:0x1 > > > > > > > > Hm, it seems double free. > > > > > > > > It doen't happen if you disable zram? IOW, it seems to be rel...

...is is what I'm getting with the [zsmalloc: keep first object offset in struct page] > > > > > applied: "count:0 mapcount:-127". which may be not related to zsmalloc at this point. > > > > > > > > > > kernel: BUG: Bad page state in process khugepaged pfn:101db8 > > > > > kernel: page:ffffea0004076e00 count:0 mapcount:-127 mapping: (null) index:0x1 > > > > > > > > Hm, it seems double free. > > > > > > > > It doen't happen if you disable zram? IOW, it seems to be rel...

...ng information more fine grained, it became clear that nearly trivial loops like DO I=1,GRIDC%RL%NP DWORK1(I,ISP)= REAL( DWORK(I,ISP) ,KIND=q) ENDDO were a large part of the problem. The only thing I could find after poking around that was at all odd is that when this happened, khugepaged would be taking 100% of cpu for tens of seconds on the affected nodes. It appears, although it?s not completely certain yet (since the problem appears rarely enough to that it?s hard to definitively rule out), that disabling THP by putting ?never? in /sys/kernel/mm/transparent_hugepage/enabled a...

...MA, but I guess this breaks the VMA. Is this recommended? The mirror vma should not be an anon vma. > > Then, mapping anon pages from one VMA to another without fixing the > refcount and the mapcount breaks the daemons that think they're working > on a pure anon VMA (kcompactd, khugepaged). Note here the target vma ie the mirroring one is a mmap of device file and thus is skip by all of the above (kcompactd, khugepaged, ...) it is fully ignore by core mm. Thus you do not need to fix the refcount in any way. If any of the core mm try to reclaim memory from the original vma then y...

Recently, I got many reports about perfermance degradation in embedded system(Android mobile phone, webOS TV and so on) and easy fork fail. The problem was fragmentation caused by zram and GPU driver mainly. With memory pressure, their pages were spread out all of pageblock and it cannot be migrated with current compaction algorithm which supports only LRU pages. In the end, compaction cannot

Recently, I got many reports about perfermance degradation in embedded system(Android mobile phone, webOS TV and so on) and easy fork fail. The problem was fragmentation caused by zram and GPU driver mainly. With memory pressure, their pages were spread out all of pageblock and it cannot be migrated with current compaction algorithm which supports only LRU pages. In the end, compaction cannot

...e VMA. Is this recommended? > > The mirror vma should not be an anon vma. > >> >> Then, mapping anon pages from one VMA to another without fixing the >> refcount and the mapcount breaks the daemons that think they're working >> on a pure anon VMA (kcompactd, khugepaged). > > Note here the target vma ie the mirroring one is a mmap of device file > and thus is skip by all of the above (kcompactd, khugepaged, ...) it is > fully ignore by core mm. > > Thus you do not need to fix the refcount in any way. If any of the core > mm try to reclaim...

...e VMA. Is this recommended? > > The mirror vma should not be an anon vma. > >> >> Then, mapping anon pages from one VMA to another without fixing the >> refcount and the mapcount breaks the daemons that think they're working >> on a pure anon VMA (kcompactd, khugepaged). > > Note here the target vma ie the mirroring one is a mmap of device file > and thus is skip by all of the above (kcompactd, khugepaged, ...) it is > fully ignore by core mm. > > Thus you do not need to fix the refcount in any way. If any of the core > mm try to reclaim...