summaryrefslogtreecommitdiffstats
path: root/kernel/locking/mcs_spinlock.h
diff options
context:
space:
mode:
authorLv Zheng <lv.zheng@intel.com>2017-01-19 08:21:34 +0100
committerRafael J. Wysocki <rafael.j.wysocki@intel.com>2017-01-20 03:44:58 +0100
commit7a37052adb5e843bcfff6c98aee9b60bb087b910 (patch)
tree7c49947ea8cfad64268f049684e6f067cd72e0d3 /kernel/locking/mcs_spinlock.h
parentACPI / osl: Remove deprecated acpi_get_table_with_size()/early_acpi_os_unmap_... (diff)
downloadlinux-7a37052adb5e843bcfff6c98aee9b60bb087b910.tar.xz
linux-7a37052adb5e843bcfff6c98aee9b60bb087b910.zip
ACPICA: Tables: Fix hidden logic related to acpi_tb_install_standard_table()
There is a hidden logic for acpi_tb_install_standard_table() as it can be invoked from the boot stage and during runtime. 1. When it is invoked from the OS boot stage, the ACPICA mutex may not have been initialized yet and so acpi_ut_acquire_mutex()/acpi_ut_release_mutex() are not invoked in these code paths: acpi_initialize_tables acpi_tb_parse_root_table acpi_tb_install_standard_table (4 invocations) acpi_install_table acpi_tb_install_standard_table 2. When it is invoked during the runtime, ACPICA mutex is used as appropriate: acpi_ex_load_op acpi_tb_install_and_load_table acpi_tb_install_standard_table acpi_load_table acpi_tb_install_and_load_table acpi_tb_install_standard_table The mutex is now used in acpi_tb_install_and_load_table(), while it actually should be in acpi_tb_install_standard_table(). This introduces another problem in acpi_tb_install_standard_table() where acpi_gbl_table_handler is invoked from and the lock contexts are thus not consistent for the table handlers. This triggers a regression when acpi_get_table()/acpi_put_table() start to hold table mutex during runtime. The regression is noticed by LKP as new errors reported by ACPICA mutex debugging facility. [ 2.043693] ACPI Error: Mutex [ACPI_MTX_Tables] already acquired by this thread [497483776] (20160930/utmutex-254) [ 2.054084] ACPI Error: Mutex [0x2] is not acquired, cannot release (20160930/utmutex-326) And it triggers a deadlock: [ 247.066214] INFO: task swapper/0:1 blocked for more than 120 seconds. ... [ 247.091271] Call Trace: ... [ 247.121523] down_timeout+0x47/0x50 [ 247.125065] acpi_os_wait_semaphore+0x47/0x62 [ 247.129475] acpi_ut_acquire_mutex+0x43/0x81 [ 247.133798] acpi_get_table+0x2d/0x84 [ 247.137513] acpi_table_attr_init+0xcd/0x100 [ 247.146590] acpi_sysfs_table_handler+0x5d/0xb8 [ 247.151174] acpi_bus_table_handler+0x23/0x2a [ 247.155583] acpi_tb_install_standard_table+0xe0/0x213 [ 247.164489] acpi_tb_install_and_load_table+0x3a/0x82 [ 247.169592] acpi_ex_load_op+0x194/0x201 ... [ 247.200108] acpi_ns_evaluate+0x1bb/0x247 [ 247.204170] acpi_evaluate_object+0x178/0x274 [ 247.213249] acpi_processor_set_pdc+0x154/0x17b ... The table mutex is held in acpi_tb_install_and_load_table() and is re-visited by acpi_get_table(). Noticing that the early mutex requirement actually belongs to the OSL layer and has already been handled in acpi_os_wait_semaphore()/acpi_os_signal_semaphore(), the regression canbe fixed by removing this hidden logic from the ACPICA core to the OS-specific code. Fixes: 174cc7187e6f ("ACPICA: Tables: Back port acpi_get_table_with_size() and early_acpi_os_unmap_memory() from Linux kernel") Reported-and-tested-by: Tomi Sarvela <tomi.p.sarvela@intel.com> Reported-by: Ye Xiaolong <xiaolong.ye@intel.com> Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Diffstat (limited to 'kernel/locking/mcs_spinlock.h')
0 files changed, 0 insertions, 0 deletions