kernel-packages team mailing list archive
-
kernel-packages team
-
Mailing list archive
-
Message #137210
[Bug 1469829] Re: ppc64el should use 'deadline' as default io scheduler
This bug was fixed in the package linux-lts-utopic -
3.16.0-50.66~14.04.1
---------------
linux-lts-utopic (3.16.0-50.66~14.04.1) trusty; urgency=low
[ Luis Henriques ]
* Release Tracking Bug
- LP: #1494371
[ Chris J Arges ]
* [Config] DEFAULT_IOSCHED="deadline" for ppc64el
- LP: #1469829
[ Upstream Kernel Changes ]
* tcp: fix recv with flags MSG_WAITALL | MSG_PEEK
- LP: #1486146
* netfilter: nfnetlink_cthelper: Remove 'const' and '&' to avoid warnings
- LP: #1490901
* Bluetooth: ath3k: Add a new ID 0cf3:e006 to ath3k list
- LP: #1490901
* Btrfs: use kmem_cache_free when freeing entry in inode cache
- LP: #1490901
* Btrfs: fix race between caching kthread and returning inode to inode
cache
- LP: #1490901
* Btrfs: fix fsync data loss after append write
- LP: #1490901
* ext4: fix reservation release on invalidatepage for delalloc fs
- LP: #1490901
* ext4: be more strict when migrating to non-extent based file
- LP: #1490901
* ext4: correctly migrate a file with a hole at the beginning
- LP: #1490901
* ext4: replace open coded nofail allocation in ext4_free_blocks()
- LP: #1490901
* drm/radeon: Handle irqs only based on irq ring, not irq status regs.
- LP: #1490901
* drm/radeon: unpin cursor BOs on suspend and pin them again on resume
(v2)
- LP: #1490901
* hpfs: kstrdup() out of memory handling
- LP: #1490901
* hpfs: hpfs_error: Remove static buffer, use vsprintf extension %pV
instead
- LP: #1490901
* 9p: don't leave a half-initialized inode sitting around
- LP: #1490901
* MIPS: kernel: traps: Fix broken indentation
- LP: #1490901
* thermal: step_wise: fix: Prevent from binary overflow when trend is
dropping
- LP: #1490901
* spi: pl022: Specify 'num-cs' property as required in devicetree binding
- LP: #1490901
* iio: twl4030-madc: Pass the IRQF_ONESHOT flag
- LP: #1490901
* iio: inv-mpu: Specify the expected format/precision for write channels
- LP: #1490901
* iio: DAC: ad5624r_spi: fix bit shift of output data value
- LP: #1490901
* iio: adc: at91_adc: allow to use full range of startup time
- LP: #1490901
* ALSA: usb-audio: Add MIDI support for Steinberg MI2/MI4
- LP: #1490901
* iio: tmp006: Check channel info on write
- LP: #1490901
* dm btree remove: fix bug in redistribute3
- LP: #1490901
* kbuild: Allow arch Makefiles to override {cpp,ld,c}flags
- LP: #1490901
* ARC: Override toplevel default -O2 with -O3
- LP: #1490901
* crypto: omap-des - Fix unmapping of dma channels
- LP: #1490901
* USB: option: add 2020:4000 ID
- LP: #1490901
* USB: cp210x: add ID for Aruba Networks controllers
- LP: #1490901
* dm btree: silence lockdep lock inversion in dm_btree_del()
- LP: #1490901
* usb: musb: host: rely on port_mode to call musb_start()
- LP: #1490901
* usb: f_mass_storage: limit number of reported LUNs
- LP: #1490901
* drm: add a check for x/y in drm_mode_setcrtc
- LP: #1490901
* bio integrity: do not assume bio_integrity_pool exists if bioset exists
- LP: #1490901
* ARM: dts: mx23: fix iio-hwmon support
- LP: #1490901
* tracing: Have branch tracer use recursive field of task struct
- LP: #1490901
* drivers: net: cpsw: fix crash while accessing second slave ethernet
interface
- LP: #1490901
* USB: serial: Destroy serial_minors IDR on module exit
- LP: #1490901
* Btrfs: fix memory leak in the extent_same ioctl
- LP: #1490901
* Btrfs: fix list transaction->pending_ordered corruption
- LP: #1490901
* can: rcar_can: fix IRQ check
- LP: #1490901
* ARC: make sure instruction_pointer() returns unsigned value
- LP: #1490901
* Btrfs: fix file corruption after cloning inline extents
- LP: #1490901
* st: null pointer dereference panic caused by use after kref_put by
st_open
- LP: #1490901
* drm/radeon: add a dpm quirk for Sapphire Radeon R9 270X 2GB GDDR5
- LP: #1490901
* drm/radeon: Don't flush the GART TLB if rdev->gart.ptr == NULL
- LP: #1490901
* genirq: Prevent resend to interrupts marked IRQ_NESTED_THREAD
- LP: #1490901
* ARM: 8404/1: dma-mapping: fix off-by-one error in bitmap size check
- LP: #1490901
* ipv6: Make MLD packets to only be processed locally
- LP: #1490901
* bridge: mdb: start delete timer for temp static entries
- LP: #1490901
* net: graceful exit from netif_alloc_netdev_queues()
- LP: #1490901
* ip_tunnel: fix ipv4 pmtu check to honor inner ip header df
- LP: #1490901
* bridge: mdb: zero out the local br_ip variable before use
- LP: #1490901
* net: do not process device backlog during unregistration
- LP: #1490901
* net: dsa: Test array index before use
- LP: #1490901
* net: dsa: Fix off-by-one in switch address parsing
- LP: #1490901
* can: rcar_can: print signed IRQ #
- LP: #1490901
* perf symbols: Store if there is a filter in place
- LP: #1490901
* perf hists browser: Take the --comm, --dsos, etc filters into account
- LP: #1490901
* rds: rds_ib_device.refcount overflow
- LP: #1490901
* mm: avoid setting up anonymous pages into file mapping
- LP: #1490901
* evm: labeling pseudo filesystems exception
- LP: #1490901
* USB: usbfs: allow URBs to be reaped after disconnection
- LP: #1490901
* sg_start_req(): make sure that there's not too many elements in iovec
- LP: #1490901
* HID: cp2112: fix to force single data-report reply
- LP: #1490901
* ata: pmp: add quirk for Marvell 4140 SATA PMP
- LP: #1490901
* libata: add ATA_HORKAGE_BROKEN_FPDMA_AA quirk for HP 250GB SATA disk
VB0250EAVER
- LP: #1490901
* libata: add ATA_HORKAGE_NOTRIM
- LP: #1490901
* libata: force disable trim for SuperSSpeed S238
- LP: #1490901
* libata: increase the timeout when setting transfer mode
- LP: #1490901
* can: mcp251x: fix resume when device is down
- LP: #1490901
* libata: Do not blacklist M510DC
- LP: #1490901
* mac80211: clear subdir_stations when removing debugfs
- LP: #1490901
* iio: adc: vf610: fix the adc register read fail issue
- LP: #1490901
* net: mvneta: fix refilling for Rx DMA buffers
- LP: #1490901
* ALSA: hda - Add new GPU codec ID 0x10de007d to snd-hda
- LP: #1490901
* xdrm/i915: Use two 32bit reads for select 64bit REG_READ ioctls
- LP: #1490901
* usb: dwc3: gadget: return error if command sent to DEPCMD register
fails
- LP: #1490901
* usb: dwc3: Reset the transfer resource index on SET_INTERFACE
- LP: #1490901
* usb: xhci: Bugfix for NULL pointer deference in xhci_endpoint_init()
function
- LP: #1490901
* xhci: Calculate old endpoints correctly on device reset
- LP: #1490901
* xhci: report U3 when link is in resume state
- LP: #1490901
* xhci: prevent bus_suspend if SS port resuming in phase 1
- LP: #1490901
* xhci: do not report PLC when link is in internal resume state
- LP: #1490901
* usb: core: lpm: set lpm_capable for root hub device
- LP: #1490901
* USB: OHCI: Fix race between ED unlink and URB submission
- LP: #1490901
* usb-storage: ignore ZTE MF 823 card reader in mode 0x1225
- LP: #1490901
* blkcg: fix gendisk reference leak in blkg_conf_prep()
- LP: #1490901
* tile: use free_bootmem_late() for initrd
- LP: #1490901
* Input: usbtouchscreen - avoid unresponsive TSC-30 touch screen
- LP: #1490901
* block: Do a full clone when splitting discard bios
- LP: #1490901
* md/raid1: fix test for 'was read error from last working device'.
- LP: #1490901
* mmc: omap_hsmmc: Fix DTO and DCRC handling
- LP: #1490901
* mtd: nand: Fix NAND_USE_BOUNCE_BUFFER flag conflict
- LP: #1490901
* net/xen-netback: off by one in BUG_ON() condition
- LP: #1490901
* bridge: mdb: fix double add notification
- LP: #1490901
* isdn/gigaset: reset tty->receive_room when attaching ser_gigaset
- LP: #1490901
* usb: gadget: mv_udc_core: fix phy_regs I/O memory leak
- LP: #1490901
* bonding: fix destruction of bond with devices different from
arphrd_ether
- LP: #1490901
* bonding: correctly handle bonding type change on enslave failure
- LP: #1490901
* inet: frags: fix defragmented packet's IP header for af_packet
- LP: #1490901
* mmc: block: Add missing mmc_blk_put() in power_ro_lock_show()
- LP: #1490901
* mmc: sdhci-esdhc: Make 8BIT bus work
- LP: #1490901
* mmc: sdhci-pxav3: fix platform_data is not initialized
- LP: #1490901
* freeing unlinked file indefinitely delayed
- LP: #1490901
* s390/sclp: clear upper register halves in _sclp_print_early
- LP: #1490901
* s390/process: fix sfpc inline assembly
- LP: #1490901
* mmc: sdhci: Fix FSL ESDHC reset handling quirk
- LP: #1490901
* md: fix a build warning
- LP: #1490901
* Linux 3.16.7-ckt16
- LP: #1490901
-- Luis Henriques <luis.henriques@xxxxxxxxxxxxx> Thu, 10 Sep 2015
16:38:21 +0100
** Changed in: linux-lts-utopic (Ubuntu Trusty)
Status: Fix Committed => Fix Released
** Changed in: linux-lts-utopic (Ubuntu Trusty)
Status: Fix Committed => Fix Released
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1469829
Title:
ppc64el should use 'deadline' as default io scheduler
Status in linux package in Ubuntu:
Fix Released
Status in linux source package in Trusty:
Fix Released
Status in linux-lts-utopic source package in Trusty:
Fix Released
Status in linux source package in Utopic:
Won't Fix
Status in linux source package in Vivid:
Fix Released
Bug description:
[Impact]
Using cfq instead of deadline as the default io scheduler starves certain workloads and causes performance issues. In addition every other arch we build uses deadline as the default scheduler.
[Fix]
Change the configuration to the following for ppc64el:
CONFIG_DEFAULT_DEADLINE=y
CONFIG_DEFAULT_IOSCHED="deadline"
[Test Case]
Boot and cat /sys/block/*/queue/scheduler to see if deadline is being used.
--
-- Problem Description --
Firestone system given to DASD group failed HTX overnight test with miscompare error.
HTX mdt.hdbuster was running on secondary drive and failed about 12 hours into test
HTX miscompare analysis:
====================-==
Device under test: /dev/sdb
Stanza running: rule_3
miscompare offset: 0x40
Transfer size: Random Size
LBA number: 0x70fc
miscompare length: all the blocks in the transfer size
*- STANZA 3: Creates number of threads twice the queue depth. Each thread -*
*- doing 20000 num_oper with RC operation with xfer size between 1 block -*
*- to 256K. -*
This miscompare shows read operation is unable to get the expected
data from the disk. The re-read buffer also shows the same data as the
first read operation. Since the first read and next re-read shows same
data, there could be a write operation (of previous rule stanza to
initialize disk with pattern 007 ) failure on the disk. The same
miscompare behavior shows for all the blocks in the transfer size.
/dev/sdb Jun 2 02:29:43 2015 err=000003b6 sev=2 hxestorage <<=== device name (/dev/sdb)
rule_3_13 numopers= 20000 loop= 767 blk=0x70fc len=89088
min_blkno=0 max_blkno=0x74706daf, RANDOM access
Seed Values= 37303, 290, 23235
Data Pattern Seed Values = 37303, 291, 23235
BWRC LBA fencepost Detail:
th_num min_lba max_lba status
0 0 1c9be3ff R
1 1d1c1b6c 3a3836d7 F
2 3a3836d8 57545243 F
3 57545244 74706daf F
Miscompare at buffer offset 64 (0x40) <<=== miscompare offset (0x40)
(Flags: badsig=0; cksum=0x60000) Maximum LBA = 0x74706daf
wbuf (baseaddr 0x3ffe1c0e6600) b0ffffffffffffffffffffffffffffffffffffff
rbuf (baseaddr 0x3ffe1c0fc400) 850100fc700200fd700300fe700400ff70050000
Write buffer saved in /tmp/htxsdb.wbuf1
Read buffer saved in /tmp/htxsdb.rbuf1
Re-read fails compare at offset64; buffer saved in /tmp/htxsdb.rerd1
errno: 950(Unknown error 950)
Asghar reproduced that HTX hang he is seeing. Looking in the kernel
logs I see some messages from the kernel that there are user threads
blocked on getting reads serviced. So likely HTX is seeing the same
thing. I've asked Asghar to try using the deadline I/O scheduler
rather than CFQ to see if that makes any difference. If that does not
make any difference, the next thing to try is reducing the queue depth
of the device. Right now its 31, which I think is pretty high.
Step 1:
echo deadline > /sys/block/sda/queue/scheduler
echo deadline > /sys/block/sdb/queue/scheduler
If that reproduces the issue, go to step 2:
echo cfq > /sys/block/sda/queue/scheduler
echo cfq > /sys/block/sdb/queue/scheduler
echo 8 > /sys/block/sda/device/queue_depth
echo 8 > /sys/block/sdb/device/queue_depth
Breno - it looks like the default I/O scheduler + default queue depth
for the SATA disks in Firestone is not optimal, in that when running a
heavy I/O workload, we see read starvation occurring, which is making
the system nearly unusable.
Once we changed the I/O scheduler from cfq to deadline, all the issues
went away and the system is able to run the same workload yet still be
responsive. Suggest we either encourage Canonical to change the
default I/O scheduler to deadline or at the very least provide
documentation to encourage our customers to make this change
themselves.
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1469829/+subscriptions
