Skip to contentSkip to conversation

mmotm 2010-11-23-16-12 uploaded

PATCH New [email protected] 10 participants 27 messages Nov 24, 2010 – Dec 3, 2010

Subsystems (12)
lore
Share & tools
lore

[email protected]

author maintainer
#0 lore raw

The mm-of-the-moment snapshot 2010-11-23-16-12 has been uploaded to

http://userweb.kernel.org/~akpm/mmotm/

and will soon be available at

git://zen-kernel.org/kernel/mmotm.git

It contains the following patches against 2.6.37-rc3:

leds-fix-bug-with-reading-nas-ss4200-dmi-code.patch
include-linux-fsh-fix-userspace-build.patch
nommu-yield-cpu-while-disposing-vm.patch
uml-disable-winch-irq-before-freeing-handler-data.patch
arch-x86-kernel-entry_64s-fix-build-with-gas-2161.patch
memcg-fix-false-positive-vm_bug-on-non-smp.patch
memcg-fix-false-positive-vm_bug-on-non-smp-fix.patch
linux-next.patch
next-remove-localversion.patch
i-need-old-gcc.patch
aesni-nfg.patch
arch-alpha-kernel-systblss-remove-debug-check.patch
sgi-xpc-xpc-fails-to-discover-partitions-with-all-nasids-above-128.patch
fuse-fix-attributes-after-openo_trunc.patch
drivers-leds-leds-lp5521c-change-some-macros-to-functions.patch drivers-leds-leds-lp5523c-change-some-macros-to-functions.patch drivers-leds-leds-lp5521c-adjust-delays-and-add-comments-to-them.patch drivers-leds-leds-lp5523c-adjust-delays-and-add-comments-to-them.patch drivers-leds-leds-lp5521c-perform-sw-reset-before-detection.patch drivers-leds-leds-lp5523c-perform-sw-reset-before-detection.patch memcg-avoid-deadlock-between-move-charge-and-try_charge.patch cgroups-make-swap-accounting-default-behavior-configurable.patch cgroups-make-swap-accounting-default-behavior-configurable-update.patch mm-page_allocc-fix-build_all_zonelist-where-percpu_alloc-is-wrongly-called-under-stop_machine_run.patch mm-page_allocc-fix-build_all_zonelist-where-percpu_alloc-is-wrongly-called-under-stop_machine_run-cleanup.patch mm-remove-call-to-find_vma-in-pagewalk-for-non-hugetlbfs.patch
pagemap-set-pagemap-walk-limit-to-pmd-boundary.patch
drivers-misc-isl29020c-remove-incorrect-kfree-in-isl29020_remove.patch
backlight-grab-ops_lock-before-testing-bd-ops.patch
reiserfs-fix-inode-mutex-reiserfs-lock-misordering.patch
scripts-fix-gfp-translate-for-recent-changes-to-gfph.patch
scripts-fix-gfp-translate-for-recent-changes-to-gfph-fix.patch
mm-vmap-area-cache.patch
arch-arm-plat-omap-iovmmc-fix-end-address-of-vm-area-comparation-in-alloc_iovm_area.patch backlight-fix-88pm860x_bl-macro-collision.patch
cciss-fix-botched-tag-masking-for-scsi-tape-commands.patch
arch-x86-kernel-entry_32s-i386-too.patch
arch-x86-include-asm-fixmaph-mark-__set_fixmap_offset-as-__always_inline.patch
ibm_rtl-fix-printk-format-warning.patch
acerhdf-add-support-for-aspire-1410-bios-v13314.patch
arch-x86-kernel-apic-io_apicc-fix-warning.patch
x86-olpc-add-xo-1-suspend-resume-support.patch
fs-btrfs-inodec-eliminate-memory-leak.patch
btrfs-dont-dereference-extent_mapping-if-null.patch
cifs-dont-overwrite-dentry-name-in-d_revalidate.patch cpufreq-fix-ondemand-governor-powersave_bias-execution-time-misuse.patch drivers-dma-use-the-ccflag-y-instead-of-extra_cflags.patch
drivers-dma-ioat-use-the-ccflag-y-instead-of-extra_cflags.patch
jfs-dont-overwrite-dentry-name-in-d_revalidate.patch
powerpc-enable-arch_dma_addr_t_64bit-with-arch_phys_addr_t_64bit.patch
debugfs-remove-module_exit.patch
drivers-gpu-drm-radeon-atomc-fix-warning.patch
irq-use-per_cpu-kstat_irqs.patch
irq-use-per_cpu-kstat_irqs-checkpatch-fixes.patch
drivers-leds-leds-lp5521c-fix-potential-buffer-overflow.patch
leds-route-kbd-leds-through-the-generic-leds-layer.patch mips-enable-arch_dma_addr_t_64bit-with-highmem-64bit_phys_addr-64bit.patch isdn-capi-unregister-capictr-notifier-after-init-failure.patch
isdn-capi-make-kcapi-use-a-separate-workqueue.patch drivers-video-backlight-l4f00242t03c-make-1-bit-signed-field-unsigned.patch drivers-video-backlight-l4f00242t03c-full-implement-fb-power-states-for-this-lcd.patch btusb-patch-add_apple_macbookpro62.patch
atmel_serial-fix-rts-high-after-initialization-in-rs485-mode.patch atmel_serial-fix-rts-high-after-initialization-in-rs485-mode-fix.patch
drivers-message-fusion-mptsasc-fix-warning.patch
hpsa-remove-incorrect-redefinition-of-pci_device_id_hp_cissf.patch drivers-block-makefile-replace-the-use-of-module-objs-with-module-y.patch drivers-block-aoe-makefile-replace-the-use-of-module-objs-with-module-y.patch
vfs-remove-a-warning-on-open_fmode.patch
vfs-add-__fmode_exec.patch
n_hdlc-fix-read-and-write-locking.patch
n_hdlc-fix-read-and-write-locking-update.patch
mm.patch
mm-page-allocator-adjust-the-per-cpu-counter-threshold-when-memory-is-low.patch mm-vmstat-use-a-single-setter-function-and-callback-for-adjusting-percpu-thresholds.patch mm-vmstat-use-a-single-setter-function-and-callback-for-adjusting-percpu-thresholds-fix.patch mm-vmstat-use-a-single-setter-function-and-callback-for-adjusting-percpu-thresholds-update.patch mm-vmstat-use-a-single-setter-function-and-callback-for-adjusting-percpu-thresholds-fix-set_pgdat_percpu_threshold-dont-use-for_each_online_cpu.patch mm-mempolicyc-add-rcu-read-lock-to-protect-pid-structure.patch
writeback-integrated-background-writeback-work.patch
writeback-trace-wakeup-event-for-background-writeback.patch writeback-stop-background-kupdate-works-from-livelocking-other-works.patch writeback-stop-background-kupdate-works-from-livelocking-other-works-update.patch writeback-avoid-livelocking-wb_sync_all-writeback.patch
writeback-avoid-livelocking-wb_sync_all-writeback-update.patch
writeback-check-skipped-pages-on-wb_sync_all.patch
writeback-check-skipped-pages-on-wb_sync_all-update.patch
writeback-check-skipped-pages-on-wb_sync_all-update-fix.patch
writeback-io-less-balance_dirty_pages.patch
writeback-consolidate-variable-names-in-balance_dirty_pages.patch
writeback-per-task-rate-limit-on-balance_dirty_pages.patch
writeback-per-task-rate-limit-on-balance_dirty_pages-fix.patch writeback-prevent-duplicate-balance_dirty_pages_ratelimited-calls.patch writeback-account-per-bdi-accumulated-written-pages.patch
writeback-bdi-write-bandwidth-estimation.patch
writeback-bdi-write-bandwidth-estimation-fix.patch
writeback-show-bdi-write-bandwidth-in-debugfs.patch
writeback-quit-throttling-when-bdi-dirty-pages-dropped-low.patch
writeback-reduce-per-bdi-dirty-threshold-ramp-up-time.patch writeback-make-reasonable-gap-between-the-dirty-background-thresholds.patch writeback-scale-down-max-throttle-bandwidth-on-concurrent-dirtiers.patch writeback-add-trace-event-for-balance_dirty_pages.patch
writeback-make-nr_to_write-a-per-file-limit.patch
writeback-make-nr_to_write-a-per-file-limit-fix.patch
sync_inode_metadata-fix-comment.patch
mm-page-writebackc-fix-__set_page_dirty_no_writeback-return-value.patch vmscan-factor-out-kswapd-sleeping-logic-from-kswapd.patch
mm-find_get_pages_contig-fixlet.patch
fs-mpagec-consolidate-code.patch
fs-mpagec-consolidate-code-checkpatch-fixes.patch
mm-convert-sprintf_symbol-to-%ps.patch
mm-smaps-export-mlock-information.patch
mm-compaction-add-trace-events-for-memory-compaction-activity.patch
mm-vmscan-convert-lumpy_mode-into-a-bitmask.patch
mm-vmscan-reclaim-order-0-and-use-compaction-instead-of-lumpy-reclaim.patch mm-vmscan-reclaim-order-0-and-use-compaction-instead-of-lumpy-reclaim-fix.patch mm-migration-allow-migration-to-operate-asynchronously-and-avoid-synchronous-compaction-in-the-faster-path.patch mm-migration-allow-migration-to-operate-asynchronously-and-avoid-synchronous-compaction-in-the-faster-path-fix.patch mm-migration-cleanup-migrate_pages-api-by-matching-types-for-offlining-and-sync.patch mm-compaction-perform-a-faster-migration-scan-when-migrating-asynchronously.patch mm-vmscan-rename-lumpy_mode-to-reclaim_mode.patch
mm-deactivate-invalidated-pages.patch
mm-deactivate-invalidated-pages-fix.patch
mm-remove-unused-get_vm_area_node.patch
mm-remove-gfp-mask-from-pcpu_get_vm_areas.patch
mm-unify-module_alloc-code-for-vmalloc.patch
oom-allow-a-non-cap_sys_resource-proces-to-oom_score_adj-down.patch
mm-clear-pageerror-bit-in-msync-fsync.patch
frv-duplicate-output_buffer-of-e03.patch
frv-duplicate-output_buffer-of-e03-checkpatch-fixes.patch
hpet-factor-timer-allocate-from-open.patch
kernel-power-changed-makefile-to-use-proper-ccflag-flag.patch
um-mark-config_highmem-as-broken.patch
arch-um-drivers-linec-safely-iterate-over-list-of-winch-handlers.patch kmsg_dump-constrain-mtdoops-and-ramoops-to-perform-their-actions-only-for-kmsg_dump_panic.patch kmsg_dump-add-kmsg_dump-calls-to-the-reboot-halt-poweroff-and-emergency_restart-paths.patch set_rtc_mmss-show-warning-message-only-once.patch
include-linux-kernelh-abs-fix-handling-of-32-bit-unsigneds-on-64-bit.patch include-linux-kernelh-abs-fix-handling-of-32-bit-unsigneds-on-64-bit-fix.patch add-the-common-dma_addr_t-typedef-to-include-linux-typesh.patch
dca-remove-unneeded-null-check.patch
scripts-get_maintainerpl-make-rolestats-the-default.patch
scripts-get_maintainerpl-use-git-fallback-more-often.patch
maintainers-intel-gfx-is-a-subscribers-only-mailing-list.patch percpucounter-optimize-__percpu_counter_add-a-bit-through-the-use-of-this_cpu-operations.patch drivers-mmc-host-omapc-use-resource_size.patch
drivers-mmc-host-omap_hsmmcc-use-resource_size.patch scripts-checkpatchpl-add-check-for-multiple-terminating-semicolons-and-casts-of-vmalloc.patch checkpatchpl-fix-cast-detection.patch
fs-select-fix-information-leak-to-userspace.patch
fs-select-fix-information-leak-to-userspace-fix.patch
epoll-convert-max_user_watches-to-long.patch
binfmt_elf-cleanups.patch
drivers-rtc-rtc-omapc-fix-a-memory-leak.patch
rtc-add-real-time-clock-driver-for-nvidia-tegra.patch drivers-gpio-cs5535-gpioc-add-some-additional-cs5535-specific-gpio-functionality.patch drivers-staging-olpc_dcon-convert-to-new-cs5535-gpio-api.patch
cyber2000fb-avoid-palette-corruption-at-higher-clocks.patch
jbd-remove-dependency-on-__gfp_nofail.patch
memcg-add-page_cgroup-flags-for-dirty-page-tracking.patch
memcg-document-cgroup-dirty-memory-interfaces.patch
memcg-document-cgroup-dirty-memory-interfaces-fix.patch
memcg-create-extensible-page-stat-update-routines.patch
memcg-add-lock-to-synchronize-page-accounting-and-migration.patch
memcg-use-zalloc-rather-than-mallocmemset.patch
fs-proc-basec-kernel-latencytopc-convert-sprintf_symbol-to-%ps.patch fs-proc-basec-kernel-latencytopc-convert-sprintf_symbol-to-%ps-checkpatch-fixes.patch proc-use-unsigned-long-inside-proc-statm.patch
exec_domain-establish-a-linux32-domain-on-config_compat-systems.patch rapidio-use-common-destid-storage-for-endpoints-and-switches.patch
rapidio-integrate-rio_switch-into-rio_dev.patch
fs-execc-provide-the-correct-process-pid-to-the-pipe-helper.patch
nfc-driver-for-nxp-semiconductors-pn544-nfc-chip.patch
nfc-driver-for-nxp-semiconductors-pn544-nfc-chip-update.patch
remove-dma64_addr_t.patch
pps-trivial-fixes.patch
pps-declare-variables-where-they-are-used-in-switch.patch
pps-fix-race-in-pps_fetch-handler.patch
pps-unify-timestamp-gathering.patch
pps-access-pps-device-by-direct-pointer.patch
pps-convert-printk-pr_-to-dev_.patch
pps-move-idr-stuff-to-ppsc.patch
pps-add-async-pps-event-handler.patch
pps-add-async-pps-event-handler-fix.patch
pps-dont-disable-interrupts-when-using-spin-locks.patch
pps-use-bug_on-for-kernel-api-safety-checks.patch
pps-simplify-conditions-a-bit.patch
ntp-add-hardpps-implementation.patch
pps-capture-monotonic_raw-timestamps-as-well.patch
pps-add-kernel-consumer-support.patch
pps-add-parallel-port-pps-client.patch
pps-add-parallel-port-pps-signal-generator.patch
memstick-a-few-changes-to-core.patch
memstick-add-support-for-legacy-memorysticks.patch
memstick-add-driver-for-ricoh-r5c592-card-reader.patch
memstick-add-driver-for-ricoh-r5c592-card-reader-fix.patch
memstick-core-fix-device_register-error-handling.patch
w1-ds2423-counter-driver-and-documentation.patch
w1-ds2423-counter-driver-and-documentation-fix.patch
romfs-have-romfs_fsh-pull-in-necessary-headers.patch
decompressors-add-missing-init-ie-__init.patch
decompressors-get-rid-of-set_error_fn-macro.patch
decompressors-include-linux-slabh-in-linux-decompress-mmh.patch decompressors-remove-unused-function-from-lib-decompress_unlzmac.patch
make-sure-nobodys-leaking-resources.patch
journal_add_journal_head-debug.patch
releasing-resources-with-children.patch
make-frame_pointer-default=y.patch
mutex-subsystem-synchro-test-module.patch
mutex-subsystem-synchro-test-module-add-missing-header-file.patch
slab-leaks3-default-y.patch
put_bh-debug.patch
add-debugging-aid-for-memory-initialisation-problems.patch
workaround-for-a-pci-restoring-bug.patch
prio_tree-debugging-patch.patch
single_open-seq_release-leak-diagnostics.patch
add-a-refcount-check-in-dput.patch
getblk-handle-2tb-devices.patch
memblock-add-input-size-checking-to-memblock_find_region.patch memblock-add-input-size-checking-to-memblock_find_region-fix.patch

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

#1 lore raw

mmotm 2010-11-23 - lockdep whinge in e1000e driver

Quoted 3 lines · On Tue, 23 Nov 2010 16:13:06 PST, [email protected] said:

The mm-of-the-moment snapshot 2010-11-23-16-12 has been uploaded to

http://userweb.kernel.org/~akpm/mmotm/

Whinges during boot while bringing up the ethernet interface:

[    1.081504] ===================================================
[    1.081507] [ INFO: suspicious rcu_dereference_check() usage. ]
[    1.081509] ---------------------------------------------------
[    1.081512] include/linux/inetdevice.h:208 invoked rcu_dereference_check() without protection!
[    1.081514] 
[    1.081515] other info that might help us debug this:
[    1.081516] 
[    1.081518] 
[    1.081518] rcu_scheduler_active = 1, debug_locks = 1
[    1.081521] 3 locks held by swapper/1:
[    1.081523]  #0:  (&__lockdep_no_validate__){+.+.+.}, at: [<ffffffff812d0b57>] device_lock+0xf/0x11
[    1.081534]  #1:  (&__lockdep_no_validate__){+.+.+.}, at: [<ffffffff812d0b57>] device_lock+0xf/0x11
[    1.081541]  #2:  (rtnl_mutex){+.+.+.}, at: [<ffffffff8142dee8>] rtnl_lock+0x12/0x14
[    1.081549] 
[    1.081550] stack backtrace:
[    1.081553] Pid: 1, comm: swapper Not tainted 2.6.37-rc3-mmotm1123 #3
[    1.081555] Call Trace:
[    1.081562]  [<ffffffff81069580>] lockdep_rcu_dereference+0x9d/0xa5
[    1.081567]  [<ffffffff8147b235>] __in_dev_get_rcu.clone.12+0x3f/0x47
[    1.081571]  [<ffffffff8147b24d>] inet_get_link_af_size+0x10/0x1f
[    1.081575]  [<ffffffff8142ce16>] if_nlmsg_size+0xd5/0x111
[    1.081579]  [<ffffffff8142ecf6>] rtmsg_ifinfo+0x1f/0xeb
[    1.081584]  [<ffffffff8105d78e>] ? raw_notifier_call_chain+0xf/0x11
[    1.081589]  [<ffffffff81421ee7>] register_netdevice+0x3ea/0x410
[    1.081593]  [<ffffffff81421f47>] register_netdev+0x3a/0x4c
[    1.081599]  [<ffffffff81551cc2>] e1000_probe+0x986/0xb6f
[    1.081604]  [<ffffffff81237b2e>] local_pci_probe+0x3f/0x70
[    1.081608]  [<ffffffff81237eae>] pci_device_probe+0x65/0x96
[    1.081614]  [<ffffffff8115a82a>] ? sysfs_create_link+0xe/0x10
[    1.081617]  [<ffffffff812d0fe0>] driver_probe_device+0xe8/0x182
[    1.081621]  [<ffffffff812d10c4>] __driver_attach+0x4a/0x6b
[    1.081625]  [<ffffffff812d107a>] ? __driver_attach+0x0/0x6b
[    1.081629]  [<ffffffff812d01cf>] bus_for_each_dev+0x57/0x83
[    1.081633]  [<ffffffff812d0ca5>] driver_attach+0x19/0x1b
[    1.081637]  [<ffffffff812d08e7>] bus_add_driver+0xae/0x205
[    1.081641]  [<ffffffff812d1324>] driver_register+0xb5/0x122
[    1.081646]  [<ffffffff81b455cb>] ? e1000_init_module+0x0/0x3e
[    1.081650]  [<ffffffff812380e4>] __pci_register_driver+0x61/0xcd
[    1.081654]  [<ffffffff81b455cb>] ? e1000_init_module+0x0/0x3e
[    1.081658]  [<ffffffff81b45607>] e1000_init_module+0x3c/0x3e
[    1.081663]  [<ffffffff810002ff>] do_one_initcall+0x7a/0x12f
[    1.081668]  [<ffffffff81b1fd08>] kernel_init+0x15d/0x1e7
[    1.081672]  [<ffffffff810035d4>] kernel_thread_helper+0x4/0x10
[    1.081678]  [<ffffffff8102f845>] ? finish_task_switch+0x3f/0xe3
[    1.081682]  [<ffffffff8155b5c0>] ? restore_args+0x0/0x30
[    1.081686]  [<ffffffff81b1fbab>] ? kernel_init+0x0/0x1e7
[    1.081690]  [<ffffffff810035d0>] ? kernel_thread_helper+0x0/0x10
[    1.081731] e1000e 0000:00:19.0: eth0: (PCI Express:2.5GB/s:Width x1) 00:24:e8:c6:ad:17
#2 lore raw

mmotm 2010-11-23 - WARNING: at drivers/tty/tty_io.c:1331

In reply to [email protected]
Quoted 3 lines · On Tue, 23 Nov 2010 16:13:06 PST, [email protected] said:

The mm-of-the-moment snapshot 2010-11-23-16-12 has been uploaded to

http://userweb.kernel.org/~akpm/mmotm/

Seen during boot:

[   22.859616] SELinux: initialized (dev tmpfs, type tmpfs), uses transition SIDs
[   23.015434] ------------[ cut here ]------------
[   23.015443] WARNING: at drivers/tty/tty_io.c:1331 tty_open+0x2a2/0x49a()
[   23.015446] Hardware name: Latitude E6500                  
[   23.015448] Modules linked in:
[   23.015453] Pid: 1207, comm: plymouthd Not tainted 2.6.37-rc3-mmotm1123 #3
[   23.015455] Call Trace:
[   23.015461]  [<ffffffff8103b189>] warn_slowpath_common+0x80/0x98
[   23.015465]  [<ffffffff8103b1b6>] warn_slowpath_null+0x15/0x17
[   23.015469]  [<ffffffff8128a3ab>] tty_open+0x2a2/0x49a
[   23.015475]  [<ffffffff810fd53f>] chrdev_open+0x11d/0x146
[   23.015479]  [<ffffffff810fd422>] ? chrdev_open+0x0/0x146
[   23.015483]  [<ffffffff810f7b4c>] __dentry_open+0x31a/0x483
[   23.015488]  [<ffffffff810f88fe>] nameidata_to_filp+0x50/0x57
[   23.015492]  [<ffffffff81105e53>] do_last+0x448/0x5b2
[   23.015497]  [<ffffffff81229229>] ? __raw_spin_lock_init+0x31/0x50
[   23.015501]  [<ffffffff81106205>] do_filp_open+0x248/0x64a
[   23.015507]  [<ffffffff810a5c4d>] ? trace_preempt_on+0x15/0x28
[   23.015511]  [<ffffffff81110cda>] ? alloc_fd+0x17c/0x18e
[   23.015516]  [<ffffffff8155adfc>] ? _raw_spin_unlock+0x30/0x69
[   23.015521]  [<ffffffff8155e3b8>] ? sub_preempt_count+0x35/0x49
[   23.015525]  [<ffffffff81110cda>] ? alloc_fd+0x17c/0x18e
[   23.015529]  [<ffffffff810f8965>] do_sys_open+0x60/0xfb
[   23.015533]  [<ffffffff8155a64b>] ? trace_hardirqs_on_thunk+0x3a/0x3f
[   23.015537]  [<ffffffff810f8a1b>] sys_open+0x1b/0x1d
[   23.015542]  [<ffffffff8100277b>] system_call_fastpath+0x16/0x1b
[   23.015545] ---[ end trace 12db3a7ab6675b51 ]---
#3 lore raw

Re: mmotm 2010-11-23 - WARNING: at drivers/tty/tty_io.c:1331

Quoted 10 lines · On Tue, Nov 23, 2010 at 11:55:39PM -0500, [email protected] wrote:
Quoted 3 lines · On Tue, 23 Nov 2010 16:13:06 PST, [email protected] said:

The mm-of-the-moment snapshot 2010-11-23-16-12 has been uploaded to

http://userweb.kernel.org/~akpm/mmotm/

Seen during boot:

[   23.015448] Modules linked in:
[   23.015453] Pid: 1207, comm: plymouthd Not tainted 2.6.37-rc3-mmotm1123 #3
[   23.015455] Call Trace:

I've been trying to figure this one out for a while, without much luck. (Users are seeing it in 2.6.36 as well.)

I *think* (I added a rawhide debugging patch to print the tty->name) that plymouth is always opening tty7 to cause this. My guess is the BKL removal has exposed some kind of race, but it's not obvious to me (and there's many other bugs to sort through too. :(

CC-ing Jiri since he seems to be the poor guy who's been poking this recently (there's a good few threads about this (though the others look like an ldisc attach race...)) I wouldn't think that's the case here since N_TTY is the default...

--Kyle

#4 lore raw

Re: mmotm 2010-11-23 - WARNING: at drivers/tty/tty_io.c:1331

Quoted 20 lines · Kyle McMartin
Quoted 10 lines · On Tue, Nov 23, 2010 at 11:55:39PM -0500, [email protected] wrote:
Quoted 3 lines · On Tue, 23 Nov 2010 16:13:06 PST, [email protected] said:

The mm-of-the-moment snapshot 2010-11-23-16-12 has been uploaded to

http://userweb.kernel.org/~akpm/mmotm/

Seen during boot:

[   23.015448] Modules linked in:
[   23.015453] Pid: 1207, comm: plymouthd Not tainted 2.6.37-rc3-mmotm1123 #3
[   23.015455] Call Trace:

I've been trying to figure this one out for a while, without much luck. (Users are seeing it in 2.6.36 as well.)

I *think* (I added a rawhide debugging patch to print the tty->name) that plymouth is always opening tty7 to cause this. My guess is the BKL removal has exposed some kind of race, but it's not obvious to me (and there's many other bugs to sort through too. :(

Quoted 4 lines

CC-ing Jiri since he seems to be the poor guy who's been poking this recently (there's a good few threads about this (though the others look like an ldisc attach race...)) I wouldn't think that's the case here since N_TTY is the default...

Ok, tty_reopen is called without TTY_LDISC set. For further considerations, note tty_lock is held in tty_open. TTY_LDISC is cleared in:

1) __tty_hangup from tty_ldisc_hangup to tty_ldisc_enable. During this section tty_lock is held.

2) tty_release via tty_ldisc_release till the end of tty existence. If tty->count <= 1, tty_lock is taken, TTY_CLOSING bit set and then tty_ldisc_release called. tty_reopen checks TTY_CLOSING before checking TTY_LDISC.

3) tty_set_ldisc from tty_ldisc_halt to tty_ldisc_enable. We take tty_lock, set TTY_LDISC_CHANGING, put tty_lock, do some other work, take tty_lock, call tty_ldisc_enable, put tty_lock.

So the only option I see is 3) and we should do:

drivers/tty/tty_io.c +2 -1
@@ -1310,7 +1310,8 @@ static int tty_reopen(struct tty_struct *tty)
 {
        struct tty_driver *driver = tty->driver;
 
-       if (test_bit(TTY_CLOSING, &tty->flags))
+       if (test_bit(TTY_CLOSING, &tty->flags) ||
+                       test_bit(TTY_LDISC_CHANGING, &tty->flags))
                return -EIO;
 
        if (driver->type == TTY_DRIVER_TYPE_PTY &&

Alan, Greg?

thanks,

Signature · 2 lines
js
suse labs
#5 lore raw

Re: mmotm 2010-11-23 - WARNING: at drivers/tty/tty_io.c:1331

Quoted 32 lines · Jiri Slaby
Quoted 20 lines · Kyle McMartin
Quoted 10 lines · On Tue, Nov 23, 2010 at 11:55:39PM -0500, [email protected] wrote:
Quoted 3 lines · On Tue, 23 Nov 2010 16:13:06 PST, [email protected] said:

The mm-of-the-moment snapshot 2010-11-23-16-12 has been uploaded to

http://userweb.kernel.org/~akpm/mmotm/

Seen during boot:

[   23.015448] Modules linked in:
[   23.015453] Pid: 1207, comm: plymouthd Not tainted 2.6.37-rc3-mmotm1123 #3
[   23.015455] Call Trace:

I've been trying to figure this one out for a while, without much luck. (Users are seeing it in 2.6.36 as well.)

I *think* (I added a rawhide debugging patch to print the tty->name) that plymouth is always opening tty7 to cause this. My guess is the BKL removal has exposed some kind of race, but it's not obvious to me (and there's many other bugs to sort through too. :(

Quoted 4 lines

CC-ing Jiri since he seems to be the poor guy who's been poking this recently (there's a good few threads about this (though the others look like an ldisc attach race...)) I wouldn't think that's the case here since N_TTY is the default...

Ok, tty_reopen is called without TTY_LDISC set. For further considerations, note tty_lock is held in tty_open. TTY_LDISC is cleared in:

1) __tty_hangup from tty_ldisc_hangup to tty_ldisc_enable. During this section tty_lock is held.

Quoted 8 lines

2) tty_release via tty_ldisc_release till the end of tty existence. If tty->count <= 1, tty_lock is taken, TTY_CLOSING bit set and then tty_ldisc_release called. tty_reopen checks TTY_CLOSING before checking TTY_LDISC.

3) tty_set_ldisc from tty_ldisc_halt to tty_ldisc_enable. We take tty_lock, set TTY_LDISC_CHANGING, put tty_lock, do some other work, take tty_lock, call tty_ldisc_enable, put tty_lock.

Oh, "do some other work" includes tty_ldisc_halt where TTY_LDISC is cleared and tty_lock is _not_ held.

Quoted 17 lines

So the only option I see is 3) and we should do:

drivers/tty/tty_io.c +2 -1
@@ -1310,7 +1310,8 @@ static int tty_reopen(struct tty_struct *tty)
 {
        struct tty_driver *driver = tty->driver;
 
-       if (test_bit(TTY_CLOSING, &tty->flags))
+       if (test_bit(TTY_CLOSING, &tty->flags) ||
+                       test_bit(TTY_LDISC_CHANGING, &tty->flags))
                return -EIO;
 
        if (driver->type == TTY_DRIVER_TYPE_PTY &&

Alan, Greg?

thanks,

Signature · 2 lines
js
suse labs
#6 lore raw

PATCH 1/1 TTY: don't allow reopen when ldisc is changing

There are many WARNINGs like the following reported nowadays:

WARNING: at drivers/tty/tty_io.c:1331 tty_open+0x2a2/0x49a()
Hardware name: Latitude E6500
Modules linked in:
Pid: 1207, comm: plymouthd Not tainted 2.6.37-rc3-mmotm1123 #3
Call Trace:
 [<ffffffff8103b189>] warn_slowpath_common+0x80/0x98
 [<ffffffff8103b1b6>] warn_slowpath_null+0x15/0x17
 [<ffffffff8128a3ab>] tty_open+0x2a2/0x49a
 [<ffffffff810fd53f>] chrdev_open+0x11d/0x146

...

This means tty_reopen is called without TTY_LDISC set. For further considerations, note tty_lock is held in tty_open. TTY_LDISC is cleared in:
1) __tty_hangup from tty_ldisc_hangup to tty_ldisc_enable. During this section tty_lock is held.

2) tty_release via tty_ldisc_release till the end of tty existence. If tty->count <= 1, tty_lock is taken, TTY_CLOSING bit set and then tty_ldisc_release called. tty_reopen checks TTY_CLOSING before checking TTY_LDISC.

3) tty_set_ldisc from tty_ldisc_halt to tty_ldisc_enable. We:
* take tty_lock, set TTY_LDISC_CHANGING, put tty_lock
* call tty_ldisc_halt (clear TTY_LDISC), tty_lock is _not_ held
* do some other work
* take tty_lock, call tty_ldisc_enable (set TTY_LDISC), put tty_lock

So the only option I see is 3). The solution is to check TTY_LDISC_CHANGING along with TTY_CLOSING in tty_reopen.

Nicely reproducible with two processes:

while (1) {
	fd = open("/dev/ttyS1", O_RDWR);
	if (fd < 0) {
		warn("open");
		continue;
	}
	close(fd);
}
--------
while (1) {
        fd = open("/dev/ttyS1", O_RDWR);
        ld1 = 0; ld2 = 2;
        while (1) {
                ioctl(fd, TIOCSETD, &ld1);
                ioctl(fd, TIOCSETD, &ld2);
        }
        close(fd);
}
Cc Kyle McMartin <[email protected]>
Cc Alan Cox <[email protected]>
---
 drivers/tty/tty_io.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)
drivers/tty/tty_io.c +2 -1
index c05c5af..878f6d6 100644
@@ -1310,7 +1310,8 @@ static int tty_reopen(struct tty_struct *tty)
 {
 	struct tty_driver *driver = tty->driver;
 
-	if (test_bit(TTY_CLOSING, &tty->flags))
+	if (test_bit(TTY_CLOSING, &tty->flags) ||
+			test_bit(TTY_LDISC_CHANGING, &tty->flags))
 		return -EIO;
 
 	if (driver->type == TTY_DRIVER_TYPE_PTY &&
Signature · 1 line
1.7.3.1
#7 lore raw

Re: [PATCH 1/1] TTY: don't allow reopen when ldisc is changing

drivers/tty/tty_io.c static int tty_reopen(struct tty_struct *tty) · line 1315

Quoted 5 lines · Jiri Slaby
drivers/tty/tty_io.c +2 -1
-	if (test_bit(TTY_CLOSING, &tty->flags))
+	if (test_bit(TTY_CLOSING, &tty->flags) ||
+			test_bit(TTY_LDISC_CHANGING, &tty->flags))
 		return -EIO;

Doh, nice catch. I just built a couple test images and sent them out to the reporters for confirmation.

Thanks!
Kyle

#8 lore raw

Re: [PATCH 1/1] TTY: don't allow reopen when ldisc is changing

In reply to Jiri Slaby

drivers/tty/tty_io.c static int tty_reopen(struct tty_struct *tty) · line 1317

Quoted 6 lines · Jiri Slaby
drivers/tty/tty_io.c +2 -1
-	if (test_bit(TTY_CLOSING, &tty->flags))
+	if (test_bit(TTY_CLOSING, &tty->flags) ||
+			test_bit(TTY_LDISC_CHANGING, &tty->flags))
 		return -EIO;
 
 	if (driver->type == TTY_DRIVER_TYPE_PTY &&

Unfortunately, users report this doesn't seem to fix things for them (built against 2.6.36 (plus another patch you wrote iirc.))

https://bugzilla.redhat.com/show_bug.cgi?id=630464#c27

I tried reverting the TTY patches between 2.6.36 and 2.6.35 and getting them to test that, and it seems ok:

https://bugzilla.redhat.com/show_bug.cgi?id=630464#c30

So I guess there must be a race here somewhere... I'll keep looking. :/

I would imagine it's something that's probably existed since the dawn of time but the BKL has just papered over entirely.

Thanks for trying!
--Kyle

#9 lore raw

Re: [PATCH 1/1] TTY: don't allow reopen when ldisc is changing

Quoted 10 lines · Kyle McMartin

drivers/tty/tty_io.c static int tty_reopen(struct tty_struct *tty) · line 1317

Quoted 6 lines · Jiri Slaby
drivers/tty/tty_io.c +2 -1
-	if (test_bit(TTY_CLOSING, &tty->flags))
+	if (test_bit(TTY_CLOSING, &tty->flags) ||
+			test_bit(TTY_LDISC_CHANGING, &tty->flags))
 		return -EIO;
 
 	if (driver->type == TTY_DRIVER_TYPE_PTY &&

Unfortunately, users report this doesn't seem to fix things for them (built against 2.6.36 (plus another patch you wrote iirc.))

Which patches exactly do you have? You need three of mine in 2.6.36.

Quoted 1 line

https://bugzilla.redhat.com/show_bug.cgi?id=630464#c27

regards,

Signature · 2 lines
js
suse labs
#10 lore raw

Re: [PATCH 1/1] TTY: don't allow reopen when ldisc is changing

Quoted 7 lines · Jiri Slaby
Quoted 6 lines · Kyle McMartin
Quoted 5 lines · Jiri Slaby
drivers/tty/tty_io.c +2 -1
-	if (test_bit(TTY_CLOSING, &tty->flags))
+	if (test_bit(TTY_CLOSING, &tty->flags) ||
+			test_bit(TTY_LDISC_CHANGING, &tty->flags))
 		return -EIO;
Quoted 7 lines
Quoted 4 lines
Quoted 1 line

if (driver->type == TTY_DRIVER_TYPE_PTY &&

Unfortunately, users report this doesn't seem to fix things for them (built against 2.6.36 (plus another patch you wrote iirc.))

Which patches exactly do you have? You need three of mine in 2.6.36.

Just tty-restore-tty_ldisc_wait_idle.patch on top of 2.6.36.1, I'll grab the other two now.

--Kyle

#11 lore raw

Re: [PATCH 1/1] TTY: don't allow reopen when ldisc is changing

In reply to Jiri Slaby
Quoted 5 lines · Jiri Slaby
Quoted 5 lines
Quoted 5 lines
drivers/tty/tty_io.c +2 -1
-	if (test_bit(TTY_CLOSING, &tty->flags))
+	if (test_bit(TTY_CLOSING, &tty->flags) ||
+			test_bit(TTY_LDISC_CHANGING, &tty->flags))
 		return -EIO;
Quoted 9 lines
Quoted 4 lines
Quoted 1 line

if (driver->type == TTY_DRIVER_TYPE_PTY &&

Unfortunately, users report this doesn't seem to fix things for them (built against 2.6.36 (plus another patch you wrote iirc.))

Which patches exactly do you have? You need three of mine in 2.6.36.

Quoted 1 line

https://bugzilla.redhat.com/show_bug.cgi?id=630464#c27

Hrm, I'm still seeing it on top of Linus' latest with that patch. :/

Even more bizarrely, I tried to come up with ways this could be failing, and decided to test a few things...

I set_bit(TTY_DEBUG, &tty->flags) just before returning from tty_init_dev (which, afaict, should be called for vc/tty$n and ptys?) and then checked it with a similar WARN_ON in tty_reopen, and found that I was hitting it fairly regularly.

As far as I can tell, for this to occur, we'd need something to open /dev/tty1 first, which hits the tty_init_dev, and something else to very closely follow that, hit the linking of driver->ttys[idx] and so skip into tty_reopen, and smack into my WARN_ON.

Of course, given the locking, I have no idea how it could possibly be happening.

I'm poking around to see, I think maybe something might be dropping locks in the callchain that gives us a window where this might be possible... I don't see any other way we could end up with tty1 having TTY_LDISC unset.

(I'm poking in some more debugging, and moving the 'linking in' of the
device until after tty_ldisc_setup in tty_init_dev, but I'm not particularly hopeful.)

--Kyle

#12 lore raw

Re: [PATCH 1/1] TTY: don't allow reopen when ldisc is changing

Quoted 3 lines · Kyle McMartin

I'm poking around to see, I think maybe something might be dropping locks in the callchain that gives us a window where this might be possible...

Of course, that's the case:

        clear_bit(TTY_LDISC, &tty->flags);
        tty_unlock();
        cancel_delayed_work_sync(&tty->buf.work);
        mutex_unlock(&tty->ldisc_mutex);

        tty_lock();
        mutex_lock(&tty->ldisc_mutex);

in tty_ldisc_hangup. Hence my point 1) from previous posts doesn't hold too:
1) __tty_hangup from tty_ldisc_hangup to tty_ldisc_enable. During this section tty_lock is held.

I will check, how to fix this.

thanks,

Signature · 2 lines
js
suse labs
#13 lore raw

Re: [PATCH 1/1] TTY: don't allow reopen when ldisc is changing

Quoted 14 lines · Jiri Slaby
Quoted 3 lines · Kyle McMartin

I'm poking around to see, I think maybe something might be dropping locks in the callchain that gives us a window where this might be possible...

Of course, that's the case:

        clear_bit(TTY_LDISC, &tty->flags);
        tty_unlock();
        cancel_delayed_work_sync(&tty->buf.work);
        mutex_unlock(&tty->ldisc_mutex);

        tty_lock();
        mutex_lock(&tty->ldisc_mutex);
Quoted 5 lines

in tty_ldisc_hangup. Hence my point 1) from previous posts doesn't hold too:
1) __tty_hangup from tty_ldisc_hangup to tty_ldisc_enable. During this section tty_lock is held.

I will check, how to fix this.

Reproducible with 2 running processes from the attachment.

regards,

Signature · 2 lines
js
suse labs

#include <err.h>
#include <fcntl.h>
#include <signal.h>
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>

#include <sys/ioctl.h>
#include <sys/stat.h>
#include <sys/types.h>
#include <sys/wait.h>

static void do_work(const char *tty)

{
	char buf[256];
	unsigned int cnt = 0;
	unsigned int errc = 0;
	int fd, con;

	if (signal(SIGHUP, SIG_IGN) == SIG_ERR)
		err(1, "signal(SIGHUP)");

	setsid();

	con = open("/tmp/aaa", O_WRONLY|O_NOCTTY|O_CREAT);
	if (con < 0)
		err(2, "open cons");

	while (1) {
		if (!(cnt++ % 10000)) {
			int len = sprintf(buf, "err=%x\n", errc);
			write(con, buf, len);
			errc = 0;
		}
		fd = open(tty, O_RDWR|O_NOCTTY);
		if (fd < 0) {
			errc |= 1;
			continue;
		}
		if (ioctl(fd, TIOCSCTTY)) {
			errc |= 2;
			continue;
		}

		if (vhangup()) {
			errc |= 4;
			continue;
		}
		close(fd);
	}
	close(con);
	exit(errc);
}

int main(int argc, char **argv)

{
	pid_t pid;

	switch (pid = fork()) {
	case 0:
		do_work(argv[1]);
		break;
	case -1:
		err(1, "fork");
		break;
	default:
	{
		int stat;
		waitpid(pid, &stat, 0);
		if (stat) {
			fprintf(stderr, "exited with: %d sig=%d signr=%u\n",
					WEXITSTATUS(stat), WIFSIGNALED(stat),
					WTERMSIG(stat));
		}
		break;
	}
	}

	return 0;
}
#14 lore raw

Re: [PATCH 1/1] TTY: don't allow reopen when ldisc is changing

Quoted 15 lines · Jiri Slaby
Quoted 14 lines · Jiri Slaby
Quoted 3 lines · Kyle McMartin

I'm poking around to see, I think maybe something might be dropping locks in the callchain that gives us a window where this might be possible...

Of course, that's the case:

        clear_bit(TTY_LDISC, &tty->flags);
        tty_unlock();
        cancel_delayed_work_sync(&tty->buf.work);
        mutex_unlock(&tty->ldisc_mutex);

        tty_lock();
        mutex_lock(&tty->ldisc_mutex);
Quoted 7 lines
Quoted 5 lines

in tty_ldisc_hangup. Hence my point 1) from previous posts doesn't hold too:
1) __tty_hangup from tty_ldisc_hangup to tty_ldisc_enable. During this section tty_lock is held.

I will check, how to fix this.

Reproducible with 2 running processes from the attachment.

Is it fixed with the attached proof-of-concept patch?

So you need:
THIS ONE
TTY: don't allow reopen when ldisc is changing
TTY: ldisc, fix open flag handling
Char: TTY, restore tty_ldisc_wait_idle

The last one is in 2.6.37-rc2 already.

thanks,

Signature · 2 lines
js
suse labs
Quoted 1 line

From 9e88e8b9915b5e067507a087437d80e6a133d612 Mon Sep 17 00:00:00 2001

From: Jiri Slaby <[email protected]>
Date: Sat, 27 Nov 2010 16:06:46 +0100
Subject: [PATCH 1/1] TTY: open/hangup race fixup

---
 drivers/tty/tty_io.c |   10 +++++++++-
 include/linux/tty.h  |    1 +
 2 files changed, 10 insertions(+), 1 deletions(-)
drivers/tty/tty_io.c +9 -1
index 878f6d6..35480dd 100644
@@ -559,6 +559,9 @@ void __tty_hangup(struct tty_struct *tty)
 
 	tty_lock();
 
+	/* some functions below drop BTM, so we need this bit */
+	set_bit(TTY_HUPPING, &tty->flags);
+
 	/* inuse_filps is protected by the single tty lock,
 	   this really needs to change if we want to flush the
 	   workqueue with the lock held */
@@ -578,6 +581,10 @@ void __tty_hangup(struct tty_struct *tty)
 	}
 	spin_unlock(&tty_files_lock);
 
+	/*
+	 * it drops BTM and thus races with reopen
+	 * we protect the race by TTY_HUPPING
+	 */
 	tty_ldisc_hangup(tty);
 
 	read_lock(&tasklist_lock);
@@ -615,7 +622,6 @@ void __tty_hangup(struct tty_struct *tty)
 	tty->session = NULL;
 	tty->pgrp = NULL;
 	tty->ctrl_status = 0;
-	set_bit(TTY_HUPPED, &tty->flags);
 	spin_unlock_irqrestore(&tty->ctrl_lock, flags);
 
 	/* Account for the p->signal references we killed */
@@ -641,6 +647,7 @@ void __tty_hangup(struct tty_struct *tty)
 	 * can't yet guarantee all that.
 	 */
 	set_bit(TTY_HUPPED, &tty->flags);
+	clear_bit(TTY_HUPPING, &tty->flags);
 	tty_ldisc_enable(tty);
 
 	tty_unlock();
@@ -1311,6 +1318,7 @@ static int tty_reopen(struct tty_struct *tty)
 	struct tty_driver *driver = tty->driver;
 
 	if (test_bit(TTY_CLOSING, &tty->flags) ||
+			test_bit(TTY_HUPPING, &tty->flags) ||
 			test_bit(TTY_LDISC_CHANGING, &tty->flags))
 		return -EIO;
 
include/linux/tty.h +1 -0
index 032d79f..54e4eaa 100644
@@ -366,6 +366,7 @@ struct tty_file_private {
 #define TTY_HUPPED 		18	/* Post driver->hangup() */
 #define TTY_FLUSHING		19	/* Flushing to ldisc in progress */
 #define TTY_FLUSHPENDING	20	/* Queued buffer flush pending */
+#define TTY_HUPPING 		21	/* ->hangup() in progress */
 
 #define TTY_WRITE_FLUSH(tty) tty_write_flush((tty))
 
Signature · 1 line
1.7.3.1
#15 lore raw

Re: [PATCH 1/1] TTY: don't allow reopen when ldisc is changing

Quoted 9 lines · Jiri Slaby

Is it fixed with the attached proof-of-concept patch?

So you need:
THIS ONE
TTY: don't allow reopen when ldisc is changing
TTY: ldisc, fix open flag handling
Char: TTY, restore tty_ldisc_wait_idle

The last one is in 2.6.37-rc2 already.

Shoved them all into a build and sent it out for testing, the tester who was hitting it very frequently reports he hasn't seen it, so I think you've managed to close the race window, awesome!

I rebooted my laptop continuously and didn't hit the WARN_ON as well to confirm.

--Kyle

#16 lore raw

mmotm 2010-11-23 + autogroups -> inconsistent lock state

In reply to [email protected]
Quoted 3 lines · On Tue, 23 Nov 2010 16:13:06 PST, [email protected] said:

The mm-of-the-moment snapshot 2010-11-23-16-12 has been uploaded to

http://userweb.kernel.org/~akpm/mmotm/

(I appear to be on a roll tonight - 3 splats before I even had a chance to login. :)

mmotm + Ingo's cleanup of Mike's autogroups patch.

[  114.569222] =================================
[  114.578171] [ INFO: inconsistent lock state ]
[  114.578171] 2.6.37-rc3-mmotm1123 #3
[  114.578171] ---------------------------------
[  114.578171] inconsistent {HARDIRQ-ON-W} -> {IN-HARDIRQ-W} usage.
[  114.578171] kworker/0:0/0 [HC1[1]:SC0[0]:HE0:SE1] takes:
[  114.578171]  (&(&sighand->siglock)->rlock){?.+...}, at: [<ffffffff8104bfb1>] __lock_task_sighand+0x88/0xd6
[  114.578171] {HARDIRQ-ON-W} state was registered at:
[  114.578171]   [<ffffffff8106a9a9>] __lock_acquire+0x358/0xd4e
[  114.578171]   [<ffffffff8106b8b1>] lock_acquire+0x100/0x126
[  114.578171]   [<ffffffff8155a849>] _raw_spin_lock+0x36/0x45
[  114.578171]   [<ffffffff81030bc6>] sched_autogroup_fork+0x30/0x61
[  114.578171]   [<ffffffff8103995a>] copy_process+0x994/0x1325
[  114.578171]   [<ffffffff8103a4ca>] do_fork+0x1ae/0x3e3
[  114.578171]   [<ffffffff81009603>] kernel_thread+0x6b/0x6d
[  114.578171]   [<ffffffff8105832e>] kthreadd+0xdd/0x11f
[  114.578171]   [<ffffffff810035d4>] kernel_thread_helper+0x4/0x10
[  114.578171] irq event stamp: 1137212
[  114.578171] hardirqs last  enabled at (1137209): [<ffffffff8155ae6f>] _raw_spin_unlock_irqrestore+0x3a/0x80
[  114.578171] hardirqs last disabled at (1137210): [<ffffffff8155b467>] save_args+0x67/0x70
[  114.578171] softirqs last  enabled at (1137212): [<ffffffff810414f3>] _local_bh_enable+0xe/0x10
[  114.578171] softirqs last disabled at (1137211): [<ffffffff81041edd>] irq_enter+0x3d/0x6f
[  114.578171] 
[  114.578171] other info that might help us debug this:
[  114.578171] 3 locks held by kworker/0:0/0:
[  114.578171]  #0:  (&(&new_timer->it_lock)->rlock){-.....}, at: [<ffffffff81056e7b>] posix_timer_fn+0x24/0xc7
[  114.578171]  #1:  (rcu_read_lock){.+.+..}, at: [<ffffffff81056d77>] rcu_read_lock+0x0/0x35
[  114.578171]  #2:  (rcu_read_lock){.+.+..}, at: [<ffffffff8104a6e6>] rcu_read_lock+0x0/0x35
[  114.578171] 
[  114.578171] stack backtrace:
[  114.578171] Pid: 0, comm: kworker/0:0 Tainted: G        W   2.6.37-rc3-mmotm1123 #3
[  114.578171] Call Trace:
[  114.578171]  <IRQ>  [<ffffffff8106a467>] valid_state+0x17c/0x18e
[  114.578171]  [<ffffffff81069d2c>] ? check_usage_forwards+0x0/0x87
[  114.578171]  [<ffffffff8106a558>] mark_lock+0xdf/0x1d8
[  114.578171]  [<ffffffff81069d2c>] ? check_usage_forwards+0x0/0x87
[  114.578171]  [<ffffffff8106a928>] __lock_acquire+0x2d7/0xd4e
[  114.578171]  [<ffffffff8106a4a6>] ? mark_lock+0x2d/0x1d8
[  114.578171]  [<ffffffff8104bfb1>] ? __lock_task_sighand+0x88/0xd6
[  114.578171]  [<ffffffff8106b8b1>] lock_acquire+0x100/0x126
[  114.578171]  [<ffffffff8104bfb1>] ? __lock_task_sighand+0x88/0xd6
[  114.578171]  [<ffffffff8155a942>] _raw_spin_lock_irqsave+0x44/0x57
[  114.578171]  [<ffffffff8104bfb1>] ? __lock_task_sighand+0x88/0xd6
[  114.578171]  [<ffffffff8104bfb1>] __lock_task_sighand+0x88/0xd6
[  114.578171]  [<ffffffff8104c6b3>] send_sigqueue+0x51/0x162
[  114.578171]  [<ffffffff81056e42>] posix_timer_event+0x3f/0x54
[  114.578171]  [<ffffffff81056ea1>] posix_timer_fn+0x4a/0xc7
[  114.578171]  [<ffffffff812294fd>] ? do_raw_spin_unlock+0xd0/0xfa
[  114.578171]  [<ffffffff8105bb7e>] __run_hrtimer+0x13e/0x27a
[  114.578171]  [<ffffffff81056e57>] ? posix_timer_fn+0x0/0xc7
[  114.578171]  [<ffffffff8105c5f3>] hrtimer_interrupt+0xea/0x1d6
[  114.578171]  [<ffffffff8101ad4f>] smp_apic_timer_interrupt+0x74/0x87
[  114.578171]  [<ffffffff81003193>] apic_timer_interrupt+0x13/0x20
[  114.578171]  <EOI>  [<ffffffff81000cf5>] ? cpu_idle+0x42/0x14e
[  114.578171]  [<ffffffff81000dd5>] ? cpu_idle+0x122/0x14e
[  114.578171]  [<ffffffff81b57170>] start_secondary+0x1a9/0x1ad

~

#17 lore raw

Re: mmotm 2010-11-23 + autogroups -> inconsistent lock state

Quoted 8 lines · On Wed, 2010-11-24 at 00:01 -0500, [email protected] wrote:
Quoted 3 lines · On Tue, 23 Nov 2010 16:13:06 PST, [email protected] said:

The mm-of-the-moment snapshot 2010-11-23-16-12 has been uploaded to

http://userweb.kernel.org/~akpm/mmotm/

(I appear to be on a roll tonight - 3 splats before I even had a chance to login. :)

mmotm + Ingo's cleanup of Mike's autogroups patch.

...

Sorry for slow response, been trying to use some of my last few vacation days on vacation stuff ;-)

The below should run gripe free. Suppose I should learn to turn on lockdep and whatnot when tinkering/testing.

Unfortunately, tip's update_shares() changes are still being difficult.

static void update_shares(int cpu)

{
        struct cfs_rq *cfs_rq;
        struct rq *rq = cpu_rq(cpu);

        rcu_read_lock();
        for_each_leaf_cfs_rq(rq, cfs_rq)
                update_shares_cpu(cfs_rq->tg, cpu);
        rcu_read_unlock();
}

Despite task groups being freed via rcu, update_shares_cup() hits freed memory and explodes, and nothing I've tried has been able to stop it. The only thing I haven't tried (aside from the right thing;) is to take rcu out of the picture entirely.

From: Mike Galbraith <[email protected]>
Date: Sat, 20 Nov 2010 12:35:00 -0700
Subject: [PATCH] sched: Improve desktop interactivity: Implement automated per session task groups

A recurring complaint from CFS users is that parallel kbuild has a negative impact on desktop interactivity. This patch implements an idea from Linus, to automatically create task groups. Currently, only per session autogroups are implemented, but the patch leaves the way open for enhancement.

Implementation: each task's signal struct contains an inherited pointer to a refcounted autogroup struct containing a task group pointer, the default for all tasks pointing to the init_task_group. When a task calls setsid(), a new task group is created, the process is moved into the new task group, and a reference to the preveious task group is dropped. Child processes inherit this task group thereafter, and increase it's refcount. When the last thread of a process exits, the process's reference is dropped, such that when the last process referencing an autogroup exits, the autogroup is destroyed.

At runqueue selection time, IFF a task has no cgroup assignment, its current autogroup is used.

Autogroup bandwidth is controllable via setting it's nice level through the proc filesystem. cat /proc/<pid>/autogroup displays the task's group and the group's nice level. echo <nice level> > /proc/<pid>/autogroup Sets the task group's shares to the weight of nice <level> task. Setting nice level is rate limited for !admin users due to the abuse risk of task group locking.

The feature is enabled from boot by default if CONFIG_SCHED_AUTOGROUP=y is selected, but can be disabled via the boot option noautogroup, and can also be turned on/off on the fly via..
echo [01] > /proc/sys/kernel/sched_autogroup_enabled.
..which will automatically move tasks to/from the root task group.

Cc Oleg Nesterov <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Markus Trippelsdorf <[email protected]>
Cc Mathieu Desnoyers <[email protected]>

LKML-Reference: <[email protected]>

---
 Documentation/kernel-parameters.txt |    2 
 fs/proc/base.c                      |   79 +++++++++++
 include/linux/sched.h               |   23 +++
 init/Kconfig                        |   12 +
 kernel/fork.c                       |    5 
 kernel/sched.c                      |   13 +
 kernel/sched_autogroup.c            |  240 ++++++++++++++++++++++++++++++++++++
 kernel/sched_autogroup.h            |   23 +++
 kernel/sched_debug.c                |   29 ++--
 kernel/sys.c                        |    4 
 kernel/sysctl.c                     |   11 +
 11 files changed, 423 insertions(+), 18 deletions(-)
include/linux/sched.h +23 -0
Index: linux-2.6.37.git/include/linux/sched.h
===================================================================
@@ -509,6 +509,8 @@ struct thread_group_cputimer {
 	spinlock_t lock;
 };
 
+struct autogroup;
+
 /*
  * NOTE! "signal_struct" does not have it's own
  * locking, because a shared signal_struct always
@@ -576,6 +578,9 @@ struct signal_struct {
 
 	struct tty_struct *tty; /* NULL if no tty */
 
+#ifdef CONFIG_SCHED_AUTOGROUP
+	struct autogroup *autogroup;
+#endif
 	/*
 	 * Cumulative resource counters for dead threads in the group,
 	 * and for reaped dead child processes forked by this group.
@@ -1931,6 +1936,24 @@ int sched_rt_handler(struct ctl_table *t
 
 extern unsigned int sysctl_sched_compat_yield;
 
+#ifdef CONFIG_SCHED_AUTOGROUP
+extern unsigned int sysctl_sched_autogroup_enabled;
+
+extern void sched_autogroup_create_attach(struct task_struct *p);
+extern void sched_autogroup_detach(struct task_struct *p);
+extern void sched_autogroup_fork(struct signal_struct *sig);
+extern void sched_autogroup_exit(struct signal_struct *sig);
+#ifdef CONFIG_PROC_FS
+extern void proc_sched_autogroup_show_task(struct task_struct *p, struct seq_file *m);
+extern int proc_sched_autogroup_set_nice(struct task_struct *p, int *nice);
+#endif
+#else
+static inline void sched_autogroup_create_attach(struct task_struct *p) { }
+static inline void sched_autogroup_detach(struct task_struct *p) { }
+static inline void sched_autogroup_fork(struct signal_struct *sig) { }
+static inline void sched_autogroup_exit(struct signal_struct *sig) { }
+#endif
+
 #ifdef CONFIG_RT_MUTEXES
 extern int rt_mutex_getprio(struct task_struct *p);
 extern void rt_mutex_setprio(struct task_struct *p, int prio);
kernel/sched.c +11 -2
Index: linux-2.6.37.git/kernel/sched.c
===================================================================
@@ -78,6 +78,7 @@
 
 #include "sched_cpupri.h"
 #include "workqueue_sched.h"
+#include "sched_autogroup.h"
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/sched.h>
@@ -268,6 +269,10 @@ struct task_group {
 	struct task_group *parent;
 	struct list_head siblings;
 	struct list_head children;
+
+#ifdef CONFIG_SCHED_AUTOGROUP
+	struct autogroup *autogroup;
+#endif
 };
 
 #define root_task_group init_task_group
@@ -605,11 +610,14 @@ static inline int cpu_of(struct rq *rq)
  */
 static inline struct task_group *task_group(struct task_struct *p)
 {
+	struct task_group *tg;
 	struct cgroup_subsys_state *css;
 
 	css = task_subsys_state_check(p, cpu_cgroup_subsys_id,
 			lockdep_is_held(&task_rq(p)->lock));
-	return container_of(css, struct task_group, css);
+	tg = container_of(css, struct task_group, css);
+
+	return autogroup_task_group(p, tg);
 }
 
 /* Change a task's cfs_rq and parent entity if it moves across CPUs/groups */
@@ -2006,6 +2014,7 @@ static void sched_irq_time_avg_update(st
 #include "sched_idletask.c"
 #include "sched_fair.c"
 #include "sched_rt.c"
+#include "sched_autogroup.c"
 #include "sched_stoptask.c"
 #ifdef CONFIG_SCHED_DEBUG
 # include "sched_debug.c"
@@ -7979,7 +7988,7 @@ void __init sched_init(void)
 #ifdef CONFIG_CGROUP_SCHED
 	list_add(&init_task_group.list, &task_groups);
 	INIT_LIST_HEAD(&init_task_group.children);
-
+	autogroup_init(&init_task);
 #endif /* CONFIG_CGROUP_SCHED */
 
 #if defined CONFIG_FAIR_GROUP_SCHED && defined CONFIG_SMP
kernel/fork.c +4 -1
Index: linux-2.6.37.git/kernel/fork.c
===================================================================
@@ -174,8 +174,10 @@ static inline void free_signal_struct(st
 
 static inline void put_signal_struct(struct signal_struct *sig)
 {
-	if (atomic_dec_and_test(&sig->sigcnt))
+	if (atomic_dec_and_test(&sig->sigcnt)) {
+		sched_autogroup_exit(sig);
 		free_signal_struct(sig);
+	}
 }
 
 void __put_task_struct(struct task_struct *tsk)
@@ -904,6 +906,7 @@ static int copy_signal(unsigned long clo
 	posix_cpu_timers_init_group(sig);
 
 	tty_audit_fork(sig);
+	sched_autogroup_fork(sig);
 
 	sig->oom_adj = current->signal->oom_adj;
 	sig->oom_score_adj = current->signal->oom_score_adj;
kernel/sys.c +3 -1
Index: linux-2.6.37.git/kernel/sys.c
===================================================================
@@ -1080,8 +1080,10 @@ SYSCALL_DEFINE0(setsid)
 	err = session;
 out:
 	write_unlock_irq(&tasklist_lock);
-	if (err > 0)
+	if (err > 0) {
 		proc_sid_connector(group_leader);
+		sched_autogroup_create_attach(group_leader);
+	}
 	return err;
 }
 
kernel/sched_debug.c +15 -14
Index: linux-2.6.37.git/kernel/sched_debug.c
===================================================================
@@ -87,6 +87,20 @@ static void print_cfs_group_stats(struct
 }
 #endif
 
+#if defined(CONFIG_CGROUP_SCHED) && \
+	(defined(CONFIG_FAIR_GROUP_SCHED) || defined(CONFIG_RT_GROUP_SCHED))
+static void task_group_path(struct task_group *tg, char *buf, int buflen)
+{
+	/* may be NULL if the underlying cgroup isn't fully-created yet */
+	if (!tg->css.cgroup) {
+		if (!autogroup_path(tg, buf, buflen))
+			buf[0] = '\0';
+		return;
+	}
+	cgroup_path(tg->css.cgroup, buf, buflen);
+}
+#endif
+
 static void
 print_task(struct seq_file *m, struct rq *rq, struct task_struct *p)
 {
@@ -115,7 +129,7 @@ print_task(struct seq_file *m, struct rq
 		char path[64];
 
 		rcu_read_lock();
-		cgroup_path(task_group(p)->css.cgroup, path, sizeof(path));
+		task_group_path(task_group(p), path, sizeof(path));
 		rcu_read_unlock();
 		SEQ_printf(m, " %s", path);
 	}
@@ -147,19 +161,6 @@ static void print_rq(struct seq_file *m,
 	read_unlock_irqrestore(&tasklist_lock, flags);
 }
 
-#if defined(CONFIG_CGROUP_SCHED) && \
-	(defined(CONFIG_FAIR_GROUP_SCHED) || defined(CONFIG_RT_GROUP_SCHED))
-static void task_group_path(struct task_group *tg, char *buf, int buflen)
-{
-	/* may be NULL if the underlying cgroup isn't fully-created yet */
-	if (!tg->css.cgroup) {
-		buf[0] = '\0';
-		return;
-	}
-	cgroup_path(tg->css.cgroup, buf, buflen);
-}
-#endif
-
 void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq)
 {
 	s64 MIN_vruntime = -1, min_vruntime, max_vruntime = -1,
fs/proc/base.c +79 -0

91 lines not on this page · raw message

kernel/sched_autogroup.h new file +23 -0
Index: linux-2.6.37.git/kernel/sched_autogroup.h
===================================================================
@@ -0,0 +1,23 @@
+#ifdef CONFIG_SCHED_AUTOGROUP
+
+static inline struct task_group *
+autogroup_task_group(struct task_struct *p, struct task_group *tg);
+
+#else /* !CONFIG_SCHED_AUTOGROUP */
+
+static inline void autogroup_init(struct task_struct *init_task) {  }
+
+static inline struct task_group *
+autogroup_task_group(struct task_struct *p, struct task_group *tg)
+{
+	return tg;
+}
+
+#ifdef CONFIG_SCHED_DEBUG
+static inline int autogroup_path(struct task_group *tg, char *buf, int buflen)
+{
+	return 0;
+}
+#endif
+
+#endif /* CONFIG_SCHED_AUTOGROUP */
kernel/sched_autogroup.c new file +240 -0

240 lines not on this page · raw message

kernel/sysctl.c +11 -0
Index: linux-2.6.37.git/kernel/sysctl.c
===================================================================
@@ -382,6 +382,17 @@ static struct ctl_table kern_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec,
 	},
+#ifdef CONFIG_SCHED_AUTOGROUP
+	{
+		.procname	= "sched_autogroup_enabled",
+		.data		= &sysctl_sched_autogroup_enabled,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+		.extra1		= &zero,
+		.extra2		= &one,
+	},
+#endif
 #ifdef CONFIG_PROVE_LOCKING
 	{
 		.procname	= "prove_locking",
init/Kconfig +12 -0
Index: linux-2.6.37.git/init/Kconfig
===================================================================
@@ -728,6 +728,18 @@ config NET_NS
 
 endif # NAMESPACES
 
+config SCHED_AUTOGROUP
+	bool "Automatic process group scheduling"
+	select CGROUPS
+	select CGROUP_SCHED
+	select FAIR_GROUP_SCHED
+	help
+	  This option optimizes the scheduler for common desktop workloads by
+	  automatically creating and populating task groups.  This separation
+	  of workloads isolates aggressive CPU burners (like build jobs) from
+	  desktop applications.  Task group autogeneration is currently based
+	  upon task session.
+
 config MM_OWNER
 	bool
 
Documentation/kernel-parameters.txt +2 -0
Index: linux-2.6.37.git/Documentation/kernel-parameters.txt
===================================================================
@@ -1622,6 +1622,8 @@ and is between 256 and 4096 characters.
 	noapic		[SMP,APIC] Tells the kernel to not make use of any
 			IOAPICs that may be present in the system.
 
+	noautogroup	Disable scheduler automatic task group creation.
+
 	nobats		[PPC] Do not use BATs for mapping kernel lowmem
 			on "Classic" PPC cores.
 
#18 lore raw

Re: mmotm 2010-11-23 + autogroups -> inconsistent lock state

Quoted 1 line · Mike Galbraith

The below should run gripe free.

In <= 2.6.37-rc3 kernel I mean. The tip version is still explosive.

-Mike

#19 lore raw

Re: mmotm 2010-11-23 + autogroups -> inconsistent lock state

In reply to Mike Galbraith
Quoted 8 lines · Mike Galbraith
Quoted 7 lines · On Wed, 2010-11-24 at 00:01 -0500, [email protected] wrote:
Quoted 3 lines · On Tue, 23 Nov 2010 16:13:06 PST, [email protected] said:

The mm-of-the-moment snapshot 2010-11-23-16-12 has been uploaded to

http://userweb.kernel.org/~akpm/mmotm/

(I appear to be on a roll tonight - 3 splats before I even had a chance to login. :)

Quoted 8 lines
Quoted 1 line

mmotm + Ingo's cleanup of Mike's autogroups patch.

...

Sorry for slow response, been trying to use some of my last few vacation days on vacation stuff ;-)

The below should run gripe free. Suppose I should learn to turn on lockdep and whatnot when tinkering/testing.

Yes, this version runs quietly, thanks.

#20 lore raw

Re: mmotm 2010-11-23 + autogroups -> inconsistent lock state

In reply to Mike Galbraith
Quoted 30 lines · Mike Galbraith
Quoted 8 lines · On Wed, 2010-11-24 at 00:01 -0500, [email protected] wrote:
Quoted 3 lines · On Tue, 23 Nov 2010 16:13:06 PST, [email protected] said:

The mm-of-the-moment snapshot 2010-11-23-16-12 has been uploaded to

http://userweb.kernel.org/~akpm/mmotm/

(I appear to be on a roll tonight - 3 splats before I even had a chance to login. :)

mmotm + Ingo's cleanup of Mike's autogroups patch.

...

Sorry for slow response, been trying to use some of my last few vacation days on vacation stuff ;-)

The below should run gripe free. Suppose I should learn to turn on lockdep and whatnot when tinkering/testing.

Unfortunately, tip's update_shares() changes are still being difficult.

static void update_shares(int cpu)

{
        struct cfs_rq *cfs_rq;
        struct rq *rq = cpu_rq(cpu);

        rcu_read_lock();
        for_each_leaf_cfs_rq(rq, cfs_rq)
                update_shares_cpu(cfs_rq->tg, cpu);
        rcu_read_unlock();
}
Quoted 4 lines

Despite task groups being freed via rcu, update_shares_cup() hits freed memory and explodes, and nothing I've tried has been able to stop it. The only thing I haven't tried (aside from the right thing;) is to take rcu out of the picture entirely.

Is your new autogroup structure retaining a pointer to memory that is freed by RCU?

If so, you will need to NULL out that pointer before the memory in question is passed to call_rcu(). (Or before the call to synchronize_rcu(), as the case may be.)

Thanx, Paul

Quoted 691 lines

From: Mike Galbraith <[email protected]>
Date: Sat, 20 Nov 2010 12:35:00 -0700
Subject: [PATCH] sched: Improve desktop interactivity: Implement automated per session task groups

A recurring complaint from CFS users is that parallel kbuild has a negative impact on desktop interactivity. This patch implements an idea from Linus, to automatically create task groups. Currently, only per session autogroups are implemented, but the patch leaves the way open for enhancement.

Implementation: each task's signal struct contains an inherited pointer to a refcounted autogroup struct containing a task group pointer, the default for all tasks pointing to the init_task_group. When a task calls setsid(), a new task group is created, the process is moved into the new task group, and a reference to the preveious task group is dropped. Child processes inherit this task group thereafter, and increase it's refcount. When the last thread of a process exits, the process's reference is dropped, such that when the last process referencing an autogroup exits, the autogroup is destroyed.

At runqueue selection time, IFF a task has no cgroup assignment, its current autogroup is used.

Autogroup bandwidth is controllable via setting it's nice level through the proc filesystem. cat /proc/<pid>/autogroup displays the task's group and the group's nice level. echo <nice level> > /proc/<pid>/autogroup Sets the task group's shares to the weight of nice <level> task. Setting nice level is rate limited for !admin users due to the abuse risk of task group locking.

The feature is enabled from boot by default if CONFIG_SCHED_AUTOGROUP=y is selected, but can be disabled via the boot option noautogroup, and can also be turned on/off on the fly via..
echo [01] > /proc/sys/kernel/sched_autogroup_enabled.
..which will automatically move tasks to/from the root task group.

Cc Oleg Nesterov <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Linus Torvalds <[email protected]>
Cc Markus Trippelsdorf <[email protected]>
Cc Mathieu Desnoyers <[email protected]>

LKML-Reference: <[email protected]>

---
 Documentation/kernel-parameters.txt |    2 
 fs/proc/base.c                      |   79 +++++++++++
 include/linux/sched.h               |   23 +++
 init/Kconfig                        |   12 +
 kernel/fork.c                       |    5 
 kernel/sched.c                      |   13 +
 kernel/sched_autogroup.c            |  240 ++++++++++++++++++++++++++++++++++++
 kernel/sched_autogroup.h            |   23 +++
 kernel/sched_debug.c                |   29 ++--
 kernel/sys.c                        |    4 
 kernel/sysctl.c                     |   11 +
 11 files changed, 423 insertions(+), 18 deletions(-)

Index: linux-2.6.37.git/include/linux/sched.h
===================================================================

include/linux/sched.h +23 -0
@@ -509,6 +509,8 @@ struct thread_group_cputimer {
 	spinlock_t lock;
 };
 
+struct autogroup;
+
 /*
  * NOTE! "signal_struct" does not have it's own
  * locking, because a shared signal_struct always
@@ -576,6 +578,9 @@ struct signal_struct {
 
 	struct tty_struct *tty; /* NULL if no tty */
 
+#ifdef CONFIG_SCHED_AUTOGROUP
+	struct autogroup *autogroup;
+#endif
 	/*
 	 * Cumulative resource counters for dead threads in the group,
 	 * and for reaped dead child processes forked by this group.
@@ -1931,6 +1936,24 @@ int sched_rt_handler(struct ctl_table *t
 
 extern unsigned int sysctl_sched_compat_yield;
 
+#ifdef CONFIG_SCHED_AUTOGROUP
+extern unsigned int sysctl_sched_autogroup_enabled;
+
+extern void sched_autogroup_create_attach(struct task_struct *p);
+extern void sched_autogroup_detach(struct task_struct *p);
+extern void sched_autogroup_fork(struct signal_struct *sig);
+extern void sched_autogroup_exit(struct signal_struct *sig);
+#ifdef CONFIG_PROC_FS
+extern void proc_sched_autogroup_show_task(struct task_struct *p, struct seq_file *m);
+extern int proc_sched_autogroup_set_nice(struct task_struct *p, int *nice);
+#endif
+#else
+static inline void sched_autogroup_create_attach(struct task_struct *p) { }
+static inline void sched_autogroup_detach(struct task_struct *p) { }
+static inline void sched_autogroup_fork(struct signal_struct *sig) { }
+static inline void sched_autogroup_exit(struct signal_struct *sig) { }
+#endif
+
 #ifdef CONFIG_RT_MUTEXES
 extern int rt_mutex_getprio(struct task_struct *p);
 extern void rt_mutex_setprio(struct task_struct *p, int prio);

Index: linux-2.6.37.git/kernel/sched.c
===================================================================

kernel/sched.c +11 -2
@@ -78,6 +78,7 @@
 
 #include "sched_cpupri.h"
 #include "workqueue_sched.h"
+#include "sched_autogroup.h"
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/sched.h>
@@ -268,6 +269,10 @@ struct task_group {
 	struct task_group *parent;
 	struct list_head siblings;
 	struct list_head children;
+
+#ifdef CONFIG_SCHED_AUTOGROUP
+	struct autogroup *autogroup;
+#endif
 };
 
 #define root_task_group init_task_group
@@ -605,11 +610,14 @@ static inline int cpu_of(struct rq *rq)
  */
 static inline struct task_group *task_group(struct task_struct *p)
 {
+	struct task_group *tg;
 	struct cgroup_subsys_state *css;
 
 	css = task_subsys_state_check(p, cpu_cgroup_subsys_id,
 			lockdep_is_held(&task_rq(p)->lock));
-	return container_of(css, struct task_group, css);
+	tg = container_of(css, struct task_group, css);
+
+	return autogroup_task_group(p, tg);
 }
 
 /* Change a task's cfs_rq and parent entity if it moves across CPUs/groups */
@@ -2006,6 +2014,7 @@ static void sched_irq_time_avg_update(st
 #include "sched_idletask.c"
 #include "sched_fair.c"
 #include "sched_rt.c"
+#include "sched_autogroup.c"
 #include "sched_stoptask.c"
 #ifdef CONFIG_SCHED_DEBUG
 # include "sched_debug.c"
@@ -7979,7 +7988,7 @@ void __init sched_init(void)
 #ifdef CONFIG_CGROUP_SCHED
 	list_add(&init_task_group.list, &task_groups);
 	INIT_LIST_HEAD(&init_task_group.children);
-
+	autogroup_init(&init_task);
 #endif /* CONFIG_CGROUP_SCHED */
 
 #if defined CONFIG_FAIR_GROUP_SCHED && defined CONFIG_SMP

Index: linux-2.6.37.git/kernel/fork.c
===================================================================

kernel/fork.c +4 -1
@@ -174,8 +174,10 @@ static inline void free_signal_struct(st
 
 static inline void put_signal_struct(struct signal_struct *sig)
 {
-	if (atomic_dec_and_test(&sig->sigcnt))
+	if (atomic_dec_and_test(&sig->sigcnt)) {
+		sched_autogroup_exit(sig);
 		free_signal_struct(sig);
+	}
 }
 
 void __put_task_struct(struct task_struct *tsk)
@@ -904,6 +906,7 @@ static int copy_signal(unsigned long clo
 	posix_cpu_timers_init_group(sig);
 
 	tty_audit_fork(sig);
+	sched_autogroup_fork(sig);
 
 	sig->oom_adj = current->signal->oom_adj;
 	sig->oom_score_adj = current->signal->oom_score_adj;

Index: linux-2.6.37.git/kernel/sys.c
===================================================================

kernel/sys.c +3 -1
@@ -1080,8 +1080,10 @@ SYSCALL_DEFINE0(setsid)
 	err = session;
 out:
 	write_unlock_irq(&tasklist_lock);
-	if (err > 0)
+	if (err > 0) {
 		proc_sid_connector(group_leader);
+		sched_autogroup_create_attach(group_leader);
+	}
 	return err;
 }
 

Index: linux-2.6.37.git/kernel/sched_debug.c
===================================================================

kernel/sched_debug.c +15 -14
@@ -87,6 +87,20 @@ static void print_cfs_group_stats(struct
 }
 #endif
 
+#if defined(CONFIG_CGROUP_SCHED) && \
+	(defined(CONFIG_FAIR_GROUP_SCHED) || defined(CONFIG_RT_GROUP_SCHED))
+static void task_group_path(struct task_group *tg, char *buf, int buflen)
+{
+	/* may be NULL if the underlying cgroup isn't fully-created yet */
+	if (!tg->css.cgroup) {
+		if (!autogroup_path(tg, buf, buflen))
+			buf[0] = '\0';
+		return;
+	}
+	cgroup_path(tg->css.cgroup, buf, buflen);
+}
+#endif
+
 static void
 print_task(struct seq_file *m, struct rq *rq, struct task_struct *p)
 {
@@ -115,7 +129,7 @@ print_task(struct seq_file *m, struct rq
 		char path[64];
 
 		rcu_read_lock();
-		cgroup_path(task_group(p)->css.cgroup, path, sizeof(path));
+		task_group_path(task_group(p), path, sizeof(path));
 		rcu_read_unlock();
 		SEQ_printf(m, " %s", path);
 	}
@@ -147,19 +161,6 @@ static void print_rq(struct seq_file *m,
 	read_unlock_irqrestore(&tasklist_lock, flags);
 }
 
-#if defined(CONFIG_CGROUP_SCHED) && \
-	(defined(CONFIG_FAIR_GROUP_SCHED) || defined(CONFIG_RT_GROUP_SCHED))
-static void task_group_path(struct task_group *tg, char *buf, int buflen)
-{
-	/* may be NULL if the underlying cgroup isn't fully-created yet */
-	if (!tg->css.cgroup) {
-		buf[0] = '\0';
-		return;
-	}
-	cgroup_path(tg->css.cgroup, buf, buflen);
-}
-#endif
-
 void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq)
 {
 	s64 MIN_vruntime = -1, min_vruntime, max_vruntime = -1,

Index: linux-2.6.37.git/fs/proc/base.c
===================================================================

fs/proc/base.c +79 -0
@@ -1407,6 +1407,82 @@ static const struct file_operations proc
 
 #endif
 
+#ifdef CONFIG_SCHED_AUTOGROUP
+/*
+ * Print out autogroup related information:
+ */
+static int sched_autogroup_show(struct seq_file *m, void *v)
+{
+	struct inode *inode = m->private;
+	struct task_struct *p;
+
+	p = get_proc_task(inode);
+	if (!p)
+		return -ESRCH;
+	proc_sched_autogroup_show_task(p, m);
+
+	put_task_struct(p);
+
+	return 0;
+}
+
+static ssize_t
+sched_autogroup_write(struct file *file, const char __user *buf,
+	    size_t count, loff_t *offset)
+{
+	struct inode *inode = file->f_path.dentry->d_inode;
+	struct task_struct *p;
+	char buffer[PROC_NUMBUF];
+	long nice;
+	int err;
+
+	memset(buffer, 0, sizeof(buffer));
+	if (count > sizeof(buffer) - 1)
+		count = sizeof(buffer) - 1;
+	if (copy_from_user(buffer, buf, count))
+		return -EFAULT;
+
+	err = strict_strtol(strstrip(buffer), 0, &nice);
+	if (err)
+		return -EINVAL;
+
+	p = get_proc_task(inode);
+	if (!p)
+		return -ESRCH;
+
+	err = nice;
+	err = proc_sched_autogroup_set_nice(p, &err);
+	if (err)
+		count = err;
+
+	put_task_struct(p);
+
+	return count;
+}
+
+static int sched_autogroup_open(struct inode *inode, struct file *filp)
+{
+	int ret;
+
+	ret = single_open(filp, sched_autogroup_show, NULL);
+	if (!ret) {
+		struct seq_file *m = filp->private_data;
+
+		m->private = inode;
+	}
+	return ret;
+}
+
+static const struct file_operations proc_pid_sched_autogroup_operations = {
+	.open		= sched_autogroup_open,
+	.read		= seq_read,
+	.write		= sched_autogroup_write,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+#endif /* CONFIG_SCHED_AUTOGROUP */
+
 static ssize_t comm_write(struct file *file, const char __user *buf,
 				size_t count, loff_t *offset)
 {
@@ -2733,6 +2809,9 @@ static const struct pid_entry tgid_base_
 #ifdef CONFIG_SCHED_DEBUG
 	REG("sched",      S_IRUGO|S_IWUSR, proc_pid_sched_operations),
 #endif
+#ifdef CONFIG_SCHED_AUTOGROUP
+	REG("autogroup",  S_IRUGO|S_IWUSR, proc_pid_sched_autogroup_operations),
+#endif
 	REG("comm",      S_IRUGO|S_IWUSR, proc_pid_set_comm_operations),
 #ifdef CONFIG_HAVE_ARCH_TRACEHOOK
 	INF("syscall",    S_IRUSR, proc_pid_syscall),

Index: linux-2.6.37.git/kernel/sched_autogroup.h
===================================================================

kernel/sched_autogroup.h new file +23 -0
@@ -0,0 +1,23 @@
+#ifdef CONFIG_SCHED_AUTOGROUP
+
+static inline struct task_group *
+autogroup_task_group(struct task_struct *p, struct task_group *tg);
+
+#else /* !CONFIG_SCHED_AUTOGROUP */
+
+static inline void autogroup_init(struct task_struct *init_task) {  }
+
+static inline struct task_group *
+autogroup_task_group(struct task_struct *p, struct task_group *tg)
+{
+	return tg;
+}
+
+#ifdef CONFIG_SCHED_DEBUG
+static inline int autogroup_path(struct task_group *tg, char *buf, int buflen)
+{
+	return 0;
+}
+#endif
+
+#endif /* CONFIG_SCHED_AUTOGROUP */

Index: linux-2.6.37.git/kernel/sched_autogroup.c
===================================================================

kernel/sched_autogroup.c new file +240 -0
@@ -0,0 +1,240 @@
+#ifdef CONFIG_SCHED_AUTOGROUP
+
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/kallsyms.h>
+#include <linux/utsname.h>
+
+unsigned int __read_mostly sysctl_sched_autogroup_enabled = 1;
+
+struct autogroup {
+	struct kref		kref;
+	struct task_group	*tg;
+	struct rw_semaphore	lock;
+	unsigned long		id;
+	int			nice;
+};
+
+static struct autogroup autogroup_default;
+static atomic_t autogroup_seq_nr;
+
+static void autogroup_init(struct task_struct *init_task)
+{
+	autogroup_default.tg = &init_task_group;
+	init_task_group.autogroup = &autogroup_default;
+	kref_init(&autogroup_default.kref);
+	init_rwsem(&autogroup_default.lock);
+	init_task->signal->autogroup = &autogroup_default;
+}
+
+static inline void autogroup_free(struct task_group *tg)
+{
+	kfree(tg->autogroup);
+}
+
+static inline void autogroup_destroy(struct kref *kref)
+{
+	struct autogroup *ag = container_of(kref, struct autogroup, kref);
+
+	sched_destroy_group(ag->tg);
+}
+
+static inline void autogroup_kref_put(struct autogroup *ag)
+{
+	kref_put(&ag->kref, autogroup_destroy);
+}
+
+static inline struct autogroup *autogroup_kref_get(struct autogroup *ag)
+{
+	kref_get(&ag->kref);
+	return ag;
+}
+
+static inline struct autogroup *autogroup_create(void)
+{
+	struct autogroup *ag = kzalloc(sizeof(*ag), GFP_KERNEL);
+	struct task_group *tg;
+
+	if (!ag)
+		goto out_fail;
+
+	tg = sched_create_group(&init_task_group);
+
+	if (IS_ERR(tg))
+		goto out_free;
+
+	kref_init(&ag->kref);
+	init_rwsem(&ag->lock);
+	ag->id = atomic_inc_return(&autogroup_seq_nr);
+	ag->tg = tg;
+	tg->autogroup = ag;
+
+	return ag;
+
+out_free:
+	kfree(ag);
+out_fail:
+	if (printk_ratelimit())
+		printk(KERN_WARNING "autogroup_create: %s failure.\n",
+			ag ? "sched_create_group()" : "kmalloc()");
+
+	return autogroup_kref_get(&autogroup_default);
+}
+
+static inline bool
+task_wants_autogroup(struct task_struct *p, struct task_group *tg)
+{
+	if (tg != &root_task_group)
+		return false;
+
+	if (p->sched_class != &fair_sched_class)
+		return false;
+
+	/*
+	 * We can only assume the task group can't go away on us if
+	 * autogroup_move_group() can see us on ->thread_group list.
+	 */
+	if (p->flags & PF_EXITING)
+		return false;
+
+	return true;
+}
+
+static inline struct task_group *
+autogroup_task_group(struct task_struct *p, struct task_group *tg)
+{
+	int enabled = ACCESS_ONCE(sysctl_sched_autogroup_enabled);
+
+	if (enabled && task_wants_autogroup(p, tg))
+		return p->signal->autogroup->tg;
+
+	return tg;
+}
+
+static void
+autogroup_move_group(struct task_struct *p, struct autogroup *ag)
+{
+	struct autogroup *prev;
+	struct task_struct *t;
+	unsigned long flags;
+
+	if (!lock_task_sighand(p, &flags)) {
+		WARN_ON(1);
+		return;
+	}
+
+	prev = p->signal->autogroup;
+	if (prev == ag) {
+		unlock_task_sighand(p, &flags);
+		return;
+	}
+
+	p->signal->autogroup = autogroup_kref_get(ag);
+	t = p;
+
+	do {
+		sched_move_task(t);
+	} while_each_thread(p, t);
+
+	unlock_task_sighand(p, &flags);
+	autogroup_kref_put(prev);
+}
+
+/* Allocates GFP_KERNEL, cannot be called under any spinlock */
+void sched_autogroup_create_attach(struct task_struct *p)
+{
+	struct autogroup *ag = autogroup_create();
+
+	autogroup_move_group(p, ag);
+	/* drop extra refrence added by autogroup_create() */
+	autogroup_kref_put(ag);
+}
+EXPORT_SYMBOL(sched_autogroup_create_attach);
+
+/* Cannot be called under siglock.  Currently has no users */
+void sched_autogroup_detach(struct task_struct *p)
+{
+	autogroup_move_group(p, &autogroup_default);
+}
+EXPORT_SYMBOL(sched_autogroup_detach);
+
+void sched_autogroup_fork(struct signal_struct *sig)
+{
+	struct task_struct *p = current;
+
+	spin_lock_irq(&p->sighand->siglock);
+	sig->autogroup = autogroup_kref_get(p->signal->autogroup);
+	spin_unlock_irq(&p->sighand->siglock);
+}
+
+void sched_autogroup_exit(struct signal_struct *sig)
+{
+	autogroup_kref_put(sig->autogroup);
+}
+
+static int __init setup_autogroup(char *str)
+{
+	sysctl_sched_autogroup_enabled = 0;
+
+	return 1;
+}
+
+__setup("noautogroup", setup_autogroup);
+
+#ifdef CONFIG_PROC_FS
+
+/* Called with siglock held. */
+int proc_sched_autogroup_set_nice(struct task_struct *p, int *nice)
+{
+	static unsigned long next = INITIAL_JIFFIES;
+	struct autogroup *ag;
+	int err;
+
+	if (*nice < -20 || *nice > 19)
+		return -EINVAL;
+
+	err = security_task_setnice(current, *nice);
+	if (err)
+		return err;
+
+	if (*nice < 0 && !can_nice(current, *nice))
+		return -EPERM;
+
+	/* this is a heavy operation taking global locks.. */
+	if (!capable(CAP_SYS_ADMIN) && time_before(jiffies, next))
+		return -EAGAIN;
+
+	next = HZ / 10 + jiffies;
+	ag = autogroup_kref_get(p->signal->autogroup);
+
+	down_write(&ag->lock);
+	err = sched_group_set_shares(ag->tg, prio_to_weight[*nice + 20]);
+	if (!err)
+		ag->nice = *nice;
+	up_write(&ag->lock);
+
+	autogroup_kref_put(ag);
+
+	return err;
+}
+
+void proc_sched_autogroup_show_task(struct task_struct *p, struct seq_file *m)
+{
+	struct autogroup *ag = autogroup_kref_get(p->signal->autogroup);
+
+	down_read(&ag->lock);
+	seq_printf(m, "/autogroup-%ld nice %d\n", ag->id, ag->nice);
+	up_read(&ag->lock);
+
+	autogroup_kref_put(ag);
+}
+#endif /* CONFIG_PROC_FS */
+
+#ifdef CONFIG_SCHED_DEBUG
+static inline int autogroup_path(struct task_group *tg, char *buf, int buflen)
+{
+	return snprintf(buf, buflen, "%s-%ld", "/autogroup", tg->autogroup->id);
+}
+#endif /* CONFIG_SCHED_DEBUG */
+
+#endif /* CONFIG_SCHED_AUTOGROUP */

Index: linux-2.6.37.git/kernel/sysctl.c
===================================================================

kernel/sysctl.c +11 -0
@@ -382,6 +382,17 @@ static struct ctl_table kern_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec,
 	},
+#ifdef CONFIG_SCHED_AUTOGROUP
+	{
+		.procname	= "sched_autogroup_enabled",
+		.data		= &sysctl_sched_autogroup_enabled,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+		.extra1		= &zero,
+		.extra2		= &one,
+	},
+#endif
 #ifdef CONFIG_PROVE_LOCKING
 	{
 		.procname	= "prove_locking",

Index: linux-2.6.37.git/init/Kconfig
===================================================================

init/Kconfig +12 -0
@@ -728,6 +728,18 @@ config NET_NS
 
 endif # NAMESPACES
 
+config SCHED_AUTOGROUP
+	bool "Automatic process group scheduling"
+	select CGROUPS
+	select CGROUP_SCHED
+	select FAIR_GROUP_SCHED
+	help
+	  This option optimizes the scheduler for common desktop workloads by
+	  automatically creating and populating task groups.  This separation
+	  of workloads isolates aggressive CPU burners (like build jobs) from
+	  desktop applications.  Task group autogeneration is currently based
+	  upon task session.
+
 config MM_OWNER
 	bool
 

Index: linux-2.6.37.git/Documentation/kernel-parameters.txt
===================================================================

Documentation/kernel-parameters.txt +2 -1
@@ -1622,6 +1622,8 @@ and is between 256 and 4096 characters.
 	noapic		[SMP,APIC] Tells the kernel to not make use of any
 			IOAPICs that may be present in the system.
 
+	noautogroup	Disable scheduler automatic task group creation.
+
 	nobats		[PPC] Do not use BATs for mapping kernel lowmem
 			on "Classic" PPC cores.
 
 
 
 
 
--

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

#21 lore raw

Re: mmotm 2010-11-23 + autogroups -> inconsistent lock state

Quoted 9 lines · Paul E. McKenney
Quoted 4 lines · Mike Galbraith

Despite task groups being freed via rcu, update_shares_cup() hits freed memory and explodes, and nothing I've tried has been able to stop it. The only thing I haven't tried (aside from the right thing;) is to take rcu out of the picture entirely.

Is your new autogroup structure retaining a pointer to memory that is freed by RCU?

That turned out to be a typo that left freed cfs_rq registered. No dark elves (memory ordering), just a defenseless little typo.

-Mike

#22 lore raw
In reply to [email protected]

Ave

2010/11/24 <[email protected]>:

Quoted 1 line

The mm-of-the-moment snapshot 2010-11-23-16-12 has been uploaded to

So far, so good - eight builds and one error (AFAICS known issue)

'make CONFIG_DEBUG_SECTION_MISMATCH=y'

  GEN     .version
  CHK     include/generated/compile.h
  UPD     include/generated/compile.h
  CC      init/version.o
  LD      init/built-in.o
  LD      .tmp_vmlinux1

drivers/built-in.o: In function `timblogiw_close':
/home/test/linux-2.6-mm/drivers/media/video/timblogiw.c:704: undefined reference to `dma_release_channel'
drivers/built-in.o: In function `buffer_release':
/home/test/linux-2.6-mm/drivers/media/video/timblogiw.c:595: undefined reference to `dma_sync_wait'
drivers/built-in.o: In function `timblogiw_open':
/home/test/linux-2.6-mm/drivers/media/video/timblogiw.c:671: undefined reference to `__dma_request_channel'
make[1]: *** [.tmp_vmlinux1] Error 1
make: *** [sub-make] Error 2

Signature · 16 lines
Slawa!
N.P.S.

Chwała tobie, Szatanie, cześć na wysokościach
Nieba, gdzie królowałeś, chwała w głębokościach
Piekła, gdzie zwyciężony, trwasz w dumnym milczeniu!
Uczyń, niechaj ma dusza spocznie z Tobą w cieniu
Drzewa Wiedzy, gdy swoje konary rozwinie,
Jak sklepienie kościoła, który nie przeminie!

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
#23 lore raw

Re: mmotm 2010-11-23-16-12 uploaded (olpc)

In reply to [email protected]
Quoted 7 lines · On Tue, 23 Nov 2010 16:13:06 -0800 [email protected] wrote:

The mm-of-the-moment snapshot 2010-11-23-16-12 has been uploaded to

http://userweb.kernel.org/~akpm/mmotm/

and will soon be available at

git://zen-kernel.org/kernel/mmotm.git

make[4]: *** No rule to make target arch/x86/platform/olpc/olpc-xo1-wakeup.c', needed by arch/x86/platform/olpc/olpc-xo1-wakeup.o'.

It's olpc-xo1-wakeup.S, so I guess it needs a special makefile rule ??

---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

#24 lore raw

Re: mmotm 2010-11-23-16-12 uploaded (olpc)

Quoted 11 lines · On Wed, 24 Nov 2010 10:51:26 -0800 Randy Dunlap <[email protected]> wrote:
Quoted 7 lines · On Tue, 23 Nov 2010 16:13:06 -0800 [email protected] wrote:

The mm-of-the-moment snapshot 2010-11-23-16-12 has been uploaded to

http://userweb.kernel.org/~akpm/mmotm/

and will soon be available at

git://zen-kernel.org/kernel/mmotm.git=

Quoted 7 lines

make[4]: *** No rule to make target arch/x86/platform/olpc/olpc-xo1-wakeup.c', needed by arch/x86/platform/olpc/olpc-xo1-wakeup.o'.

It's olpc-xo1-wakeup.S, so I guess it needs a special makefile rule ??

I had trouble with this as well (and after flailing at it a bit, ended up just dropping the olpc pm stuff from my tree for now). The build failure is definitely config-specific. I suspected that it needs something like the following, but failed to figure it out:

foo-y := olpc-xo1-wakeup.o
obj-$(CONFIG_OLPC_XO1) += olpc-xo1.o foo.o

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

#25 lore raw

Re: mmotm 2010-11-23-16-12 uploaded (olpc)

In reply to Randy Dunlap
Quoted 4 lines · Randy Dunlap

make[4]: *** No rule to make target arch/x86/platform/olpc/olpc-xo1-wakeup.c', needed by arch/x86/platform/olpc/olpc-xo1-wakeup.o'.

It's olpc-xo1-wakeup.S, so I guess it needs a special makefile rule ??

Works if you build it in, but fails as above as a module.

And it looks like making it work as a module is not as easy as we thought. I'll discuss this with Andres and get a new patch submitted soon.

Daniel

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

#26 lore raw

PATCH 1/1 media: fix timblogiw kconfig & build error

In reply to [email protected]

From: Randy Dunlap <[email protected]>

timblogiw uses dma() interfaces and it selects TIMB_DMA for that support. However, drivers/dma/ is not built unless CONFIG_DMA_ENGINE is enabled, so select/enable that symbol also.

drivers/built-in.o: In function `timblogiw_close':
timblogiw.c:(.text+0x4419fe): undefined reference to dma_release_channel' drivers/built-in.o: In function buffer_release':
timblogiw.c:(.text+0x441a8d): undefined reference to dma_sync_wait' drivers/built-in.o: In function timblogiw_open':
timblogiw.c:(.text+0x44212b): undefined reference to `__dma_request_channel'

---
 drivers/media/video/Kconfig |    1 +
 1 file changed, 1 insertion(+)
drivers/media/video/Kconfig +1 -0
@@ -669,6 +669,7 @@ config VIDEO_HEXIUM_GEMINI
 config VIDEO_TIMBERDALE
 	tristate "Support for timberdale Video In/LogiWIN"
 	depends on VIDEO_V4L2 && I2C
+	select DMA_ENGINE
 	select TIMB_DMA
 	select VIDEO_ADV7180
 	select VIDEOBUF_DMA_CONTIG

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>