From: Emmanuel Grumbach Date: Sun, 23 May 2010 07:14:08 +0000 (-0700) Subject: iwlwifi: move sysfs_create_group to post request firmware X-Git-Url: http://drtracing.org/?a=commitdiff_plain;h=7d47618a2ade0cb6d8a0b2597029c383c1662fa0;hp=7d47618a2ade0cb6d8a0b2597029c383c1662fa0;p=deliverable%2Flinux.git iwlwifi: move sysfs_create_group to post request firmware Move the sysfs_create_group to iwl_ucode_callback after we have safely got the firmware. The motivation to do this comes from a warning from lockdep which detected that we request priv->mutex while holding s_active during a sysfs request (show_statistics in the example copy pasted). The reverse order exists upon request_firmware: request_firmware which is a sysfs operation that requires s_active is run under priv->mutex. This ensures that we don't get sysfs request before we finish to request the firmware, avoiding this deadlock. ======================================================= [ INFO: possible circular locking dependency detected ] ------------------------------------------------------- cat/2595 is trying to acquire lock: (&priv->mutex){+.+.+.}, at: [] show_statistics+0x48/0x100 [iwlagn] but task is already holding lock: (s_active){++++.+}, at: [] sysfs_get_active_two+0x1d/0x50 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (s_active){++++.+}: [] __lock_acquire+0xc44/0x1230 [] lock_acquire+0x8d/0x110 [] sysfs_addrm_finish+0xe9/0x180 [] sysfs_hash_and_remove+0x4a/0x80 [] sysfs_remove_group+0x44/0xd0 [] dpm_sysfs_remove+0x15/0x20 [] device_del+0x38/0x170 [] device_unregister+0x1e/0x60 [] _request_firmware+0x29d/0x550 [] request_firmware+0x17/0x20 [] iwl_mac_start+0xb1/0x1230 [iwlagn] [] ieee80211_open+0x436/0x6f0 [mac80211] [] dev_open+0x92/0xf0 [] dev_change_flags+0x7b/0x190 [] do_setlink+0x178/0x3b0 [] rtnl_setlink+0xf9/0x130 [] rtnetlink_rcv_msg+0x1bb/0x1f0 [] netlink_rcv_skb+0x86/0xa0 [] rtnetlink_rcv+0x1c/0x30 [] netlink_unicast+0x263/0x290 [] netlink_sendmsg+0x1c8/0x2a0 [] sock_sendmsg+0xcd/0x100 [] sys_sendmsg+0x15d/0x290 [] sys_socketcall+0xeb/0x2a0 [] sysenter_do_call+0x12/0x38 -> #0 (&priv->mutex){+.+.+.}: [] __lock_acquire+0x1054/0x1230 [] lock_acquire+0x8d/0x110 [] __mutex_lock_common+0x58/0x470 [] mutex_lock_nested+0x3a/0x50 [] show_statistics+0x48/0x100 [iwlagn] [] dev_attr_show+0x29/0x50 [] sysfs_read_file+0xdd/0x190 [] vfs_read+0x9f/0x190 [] sys_read+0x42/0x70 [] sysenter_do_call+0x12/0x38 other info that might help us debug this: 3 locks held by cat/2595: #0: (&buffer->mutex){+.+.+.}, at: [] sysfs_read_file+0x35/0x190 #1: (s_active){++++.+}, at: [] sysfs_get_active_two+0x2d/0x50 #2: (s_active){++++.+}, at: [] sysfs_get_active_two+0x1d/0x50 stack backtrace: Pid: 2595, comm: cat Not tainted 2.6.33-tp-rc4 #2 Call Trace: [] ? printk+0x1d/0x22 [] print_circular_bug+0xc2/0xd0 [] __lock_acquire+0x1054/0x1230 [] ? sched_clock_cpu+0x121/0x180 [] lock_acquire+0x8d/0x110 [] ? show_statistics+0x48/0x100 [iwlagn] [] __mutex_lock_common+0x58/0x470 [] ? show_statistics+0x48/0x100 [iwlagn] [] mutex_lock_nested+0x3a/0x50 [] ? show_statistics+0x48/0x100 [iwlagn] [] show_statistics+0x48/0x100 [iwlagn] [] ? sysfs_get_active+0x69/0xb0 [] ? show_statistics+0x0/0x100 [iwlagn] [] dev_attr_show+0x29/0x50 [] sysfs_read_file+0xdd/0x190 [] ? security_file_permission+0x14/0x20 [] ? rw_verify_area+0x62/0xd0 [] vfs_read+0x9f/0x190 [] ? up_read+0x1b/0x30 [] ? sysfs_read_file+0x0/0x190 [] ? audit_syscall_entry+0x1f4/0x220 [] sys_read+0x42/0x70 [] sysenter_do_call+0x12/0x38 Signed-off-by: Emmanuel Grumbach Signed-off-by: Reinette Chatre ---