| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20131213133430.GW18380@S2101-09.ap.freescale.net>
Date: Fri, 13 Dec 2013 21:34:32 +0800
From: Shawn Guo <shawn.guo@...aro.org>
To: Philippe De Muyter <phdm@...q.eu>
CC: <linux-kernel@...r.kernel.org>, <b29396@...escale.com>,
<cjb@...top.org>, Fabio Estevam <fabio.estevam@...escale.com>
Subject: Re: ARM,sdhci_esdhc_imx: INFO: HARDIRQ-safe -> HARDIRQ-unsafe lock
order detected
Hi Philippe,
On Thu, Dec 12, 2013 at 03:49:25PM +0100, Philippe De Muyter wrote:
> Hello,
>
> I have just booted 3.13-rc3 on my i.MX6DL board, done nothing but watch
> it start, and I have got this!
There is a patch [1] from Fabio for this issue.
Shawn
[1] http://thread.gmane.org/gmane.linux.kernel.mmc/23786
> ======================================================
> [ INFO: HARDIRQ-safe -> HARDIRQ-unsafe lock order detected ]
> 3.13.0-rc3+ #4 Not tainted
> ------------------------------------------------------
> ALSA device list:
> No soundcards found.
> kworker/u4:0/6 [HC0[0]:SC0[0]:HE0:SE1] is trying to acquire:
> (prepare_lock){+.+.+.}, at: [<804b55b4>] clk_prepare_lock+0x78/0xec
>
> and this task is already holding:
> (&(&host->lock)->rlock){-.-...}, at: [<804941c4>] sdhci_do_set_ios+0x20/0x728
> which would create a new lock dependency:
> (&(&host->lock)->rlock){-.-...} -> (prepare_lock){+.+.+.}
>
> but this new dependency connects a HARDIRQ-irq-safe lock:
> (&(&host->lock)->rlock){-.-...}
> ... which became HARDIRQ-irq-safe at:
> [<8005f640>] mark_lock+0x15c/0x6d8
> [<8006069c>] __lock_acquire+0xae0/0x1cc8
> [<80061d9c>] lock_acquire+0x68/0x7c
> [<80650b64>] _raw_spin_lock+0x34/0x44
> [<8049679c>] sdhci_irq+0x24/0xa30
> [<8006b64c>] handle_irq_event_percpu+0x54/0x1a4
> [<8006b7e0>] handle_irq_event+0x44/0x64
> [<8006e928>] handle_fasteoi_irq+0xc4/0x15c
> [<8006afb0>] generic_handle_irq+0x28/0x38
> [<8000f1f4>] handle_IRQ+0x54/0xb4
> [<80008644>] gic_handle_irq+0x30/0x64
> [<80012f64>] __irq_svc+0x44/0x5c
> [<806488c0>] printk+0x3c/0x44
> [<80493e20>] sdhci_add_host+0x83c/0xbc0
> [<80498bcc>] sdhci_esdhc_imx_probe+0x324/0x548
> [<8035f480>] platform_drv_probe+0x20/0x50
> [<8035d8ac>] driver_probe_device+0x118/0x22c
> [<8035da5c>] __driver_attach+0x9c/0xa0
> [<8035bfa0>] bus_for_each_dev+0x5c/0x90
> [<8035d380>] driver_attach+0x20/0x28
> [<8035cf80>] bus_add_driver+0xe0/0x1d8
> [<8035e114>] driver_register+0x80/0xfc
> [<8035edcc>] __platform_driver_register+0x50/0x64
> [<808979c0>] sdhci_esdhc_imx_driver_init+0x18/0x20
> [<800088b8>] do_one_initcall+0x40/0x170
> [<80867c78>] kernel_init_freeable+0x104/0x1d4
> [<80644934>] kernel_init+0x10/0x118
> [<8000e988>] ret_from_fork+0x14/0x2c
>
> to a HARDIRQ-irq-unsafe lock:
> (prepare_lock){+.+.+.}
> ... which became HARDIRQ-irq-unsafe at:
> ... [<8005f640>] mark_lock+0x15c/0x6d8
> [<80062594>] mark_held_locks+0xa0/0x138
> [<800626dc>] trace_hardirqs_on_caller+0xb0/0x1cc
> [<8006280c>] trace_hardirqs_on+0x14/0x18
> [<8064daa8>] mutex_trylock+0x180/0x20c
> [<804b5550>] clk_prepare_lock+0x14/0xec
> [<804b5d34>] clk_notifier_register+0x2c/0xf0
> [<80015f84>] twd_clk_init+0x50/0x68
> [<800088b8>] do_one_initcall+0x40/0x170
> [<80867c78>] kernel_init_freeable+0x104/0x1d4
> [<80644934>] kernel_init+0x10/0x118
> [<8000e988>] ret_from_fork+0x14/0x2c
>
> other info that might help us debug this:
>
> Possible interrupt unsafe locking scenario:
>
> CPU0 CPU1
> ---- ----
> lock(prepare_lock);
> local_irq_disable();
> lock(&(&host->lock)->rlock);
> lock(prepare_lock);
> <Interrupt>
> lock(&(&host->lock)->rlock);
>
> *** DEADLOCK ***
>
> 3 locks held by kworker/u4:0/6:
> #0: (kmmcd){.+.+.+}, at: [<8003ded4>] process_one_work+0x130/0x444
> #1: ((&(&host->detect)->work)){+.+.+.}, at: [<8003ded4>] process_one_work+0x130/0x444
> #2: (&(&host->lock)->rlock){-.-...}, at: [<804941c4>] sdhci_do_set_ios+0x20/0x728
>
> the dependencies between HARDIRQ-irq-safe lock and the holding lock:
> -> (&(&host->lock)->rlock){-.-...} ops: 410 {
> IN-HARDIRQ-W at:
> [<8005f640>] mark_lock+0x15c/0x6d8
> [<8006069c>] __lock_acquire+0xae0/0x1cc8
> [<80061d9c>] lock_acquire+0x68/0x7c
> [<80650b64>] _raw_spin_lock+0x34/0x44
> [<8049679c>] sdhci_irq+0x24/0xa30
> [<8006b64c>] handle_irq_event_percpu+0x54/0x1a4
> [<8006b7e0>] handle_irq_event+0x44/0x64
> [<8006e928>] handle_fasteoi_irq+0xc4/0x15c
> [<8006afb0>] generic_handle_irq+0x28/0x38
> [<8000f1f4>] handle_IRQ+0x54/0xb4
> [<80008644>] gic_handle_irq+0x30/0x64
> [<80012f64>] __irq_svc+0x44/0x5c
> [<806488c0>] printk+0x3c/0x44
> [<80493e20>] sdhci_add_host+0x83c/0xbc0
> [<80498bcc>] sdhci_esdhc_imx_probe+0x324/0x548
> [<8035f480>] platform_drv_probe+0x20/0x50
> [<8035d8ac>] driver_probe_device+0x118/0x22c
> [<8035da5c>] __driver_attach+0x9c/0xa0
> [<8035bfa0>] bus_for_each_dev+0x5c/0x90
> [<8035d380>] driver_attach+0x20/0x28
> [<8035cf80>] bus_add_driver+0xe0/0x1d8
> [<8035e114>] driver_register+0x80/0xfc
> [<8035edcc>] __platform_driver_register+0x50/0x64
> [<808979c0>] sdhci_esdhc_imx_driver_init+0x18/0x20
> [<800088b8>] do_one_initcall+0x40/0x170
> [<80867c78>] kernel_init_freeable+0x104/0x1d4
> [<80644934>] kernel_init+0x10/0x118
> [<8000e988>] ret_from_fork+0x14/0x2c
> IN-SOFTIRQ-W at:
> [<8005f640>] mark_lock+0x15c/0x6d8
> [<800601b4>] __lock_acquire+0x5f8/0x1cc8
> [<80061d9c>] lock_acquire+0x68/0x7c
> [<80650c44>] _raw_spin_lock_irqsave+0x4c/0x60
> [<80492f54>] sdhci_tasklet_finish+0x1c/0x120
> [<8002bb24>] tasklet_action+0x90/0x14c
> [<8002c07c>] __do_softirq+0x100/0x254
> [<8002c5a4>] irq_exit+0xb0/0x108
> [<8000f1f8>] handle_IRQ+0x58/0xb4
> [<80008644>] gic_handle_irq+0x30/0x64
> [<80012f64>] __irq_svc+0x44/0x5c
> [<806488c0>] printk+0x3c/0x44
> [<80493e20>] sdhci_add_host+0x83c/0xbc0
> [<80498bcc>] sdhci_esdhc_imx_probe+0x324/0x548
> [<8035f480>] platform_drv_probe+0x20/0x50
> [<8035d8ac>] driver_probe_device+0x118/0x22c
> [<8035da5c>] __driver_attach+0x9c/0xa0
> [<8035bfa0>] bus_for_each_dev+0x5c/0x90
> [<8035d380>] driver_attach+0x20/0x28
> [<8035cf80>] bus_add_driver+0xe0/0x1d8
> [<8035e114>] driver_register+0x80/0xfc
> [<8035edcc>] __platform_driver_register+0x50/0x64
> [<808979c0>] sdhci_esdhc_imx_driver_init+0x18/0x20
> [<800088b8>] do_one_initcall+0x40/0x170
> [<80867c78>] kernel_init_freeable+0x104/0x1d4
> [<80644934>] kernel_init+0x10/0x118
> [<8000e988>] ret_from_fork+0x14/0x2c
> INITIAL USE at:
> [<8005f640>] mark_lock+0x15c/0x6d8
> [<8005fee0>] __lock_acquire+0x324/0x1cc8
> [<80061d9c>] lock_acquire+0x68/0x7c
> [<80650c44>] _raw_spin_lock_irqsave+0x4c/0x60
> [<804941c4>] sdhci_do_set_ios+0x20/0x728
> [<804948f4>] sdhci_set_ios+0x28/0x34
> [<80481680>] mmc_power_up+0x6c/0xd0
> [<804823d8>] mmc_start_host+0x40/0x70
> [<804833f0>] mmc_add_host+0x60/0x88
> [<80493dd4>] sdhci_add_host+0x7f0/0xbc0
> [<80498bcc>] sdhci_esdhc_imx_probe+0x324/0x548
> [<8035f480>] platform_drv_probe+0x20/0x50
> [<8035d8ac>] driver_probe_device+0x118/0x22c
> [<8035da5c>] __driver_attach+0x9c/0xa0
> [<8035bfa0>] bus_for_each_dev+0x5c/0x90
> [<8035d380>] driver_attach+0x20/0x28
> [<8035cf80>] bus_add_driver+0xe0/0x1d8
> [<8035e114>] driver_register+0x80/0xfc
> [<8035edcc>] __platform_driver_register+0x50/0x64
> [<808979c0>] sdhci_esdhc_imx_driver_init+0x18/0x20
> [<800088b8>] do_one_initcall+0x40/0x170
> [<80867c78>] kernel_init_freeable+0x104/0x1d4
> [<80644934>] kernel_init+0x10/0x118
> [<8000e988>] ret_from_fork+0x14/0x2c
> }
> ... key at: [<80e517cc>] __key.24069+0x0/0x8
> ... acquired at:
> [<8005f13c>] check_usage+0x3e8/0x5f4
> [<8005f3a0>] check_irq_usage+0x58/0xb4
> [<80060ce0>] __lock_acquire+0x1124/0x1cc8
> [<80061d9c>] lock_acquire+0x68/0x7c
> [<8064f57c>] mutex_lock_nested+0x5c/0x3b4
> [<804b55b4>] clk_prepare_lock+0x78/0xec
> [<804b5fc8>] clk_get_rate+0x14/0x64
> [<80497c74>] esdhc_pltfm_set_clock+0x20/0x2a8
> [<80492b60>] sdhci_set_clock+0x4c/0x424
> [<804944b8>] sdhci_do_set_ios+0x314/0x728
> [<804948f4>] sdhci_set_ios+0x28/0x34
> [<80481560>] mmc_set_bus_mode+0x20/0x24
> [<804858d4>] mmc_attach_mmc+0x140/0x1a0
> [<80482348>] mmc_rescan+0x28c/0x2dc
> [<8003df48>] process_one_work+0x1a4/0x444
> [<8003ec7c>] worker_thread+0x120/0x3e4
> [<8004539c>] kthread+0xcc/0xe8
> [<8000e988>] ret_from_fork+0x14/0x2c
>
>
> the dependencies between the lock to be acquired and HARDIRQ-irq-unsafe lock:
> -> (prepare_lock){+.+.+.} ops: 337 {
> HARDIRQ-ON-W at:
> [<8005f640>] mark_lock+0x15c/0x6d8
> [<80062594>] mark_held_locks+0xa0/0x138
> [<800626dc>] trace_hardirqs_on_caller+0xb0/0x1cc
> [<8006280c>] trace_hardirqs_on+0x14/0x18
> [<8064daa8>] mutex_trylock+0x180/0x20c
> [<804b5550>] clk_prepare_lock+0x14/0xec
> [<804b5d34>] clk_notifier_register+0x2c/0xf0
> [<80015f84>] twd_clk_init+0x50/0x68
> [<800088b8>] do_one_initcall+0x40/0x170
> [<80867c78>] kernel_init_freeable+0x104/0x1d4
> [<80644934>] kernel_init+0x10/0x118
> [<8000e988>] ret_from_fork+0x14/0x2c
> SOFTIRQ-ON-W at:
> [<8005f640>] mark_lock+0x15c/0x6d8
> [<80062594>] mark_held_locks+0xa0/0x138
> [<80062720>] trace_hardirqs_on_caller+0xf4/0x1cc
> [<8006280c>] trace_hardirqs_on+0x14/0x18
> [<8064daa8>] mutex_trylock+0x180/0x20c
> [<804b5550>] clk_prepare_lock+0x14/0xec
> [<804b5d34>] clk_notifier_register+0x2c/0xf0
> [<80015f84>] twd_clk_init+0x50/0x68
> [<800088b8>] do_one_initcall+0x40/0x170
> [<80867c78>] kernel_init_freeable+0x104/0x1d4
> [<80644934>] kernel_init+0x10/0x118
> [<8000e988>] ret_from_fork+0x14/0x2c
> RECLAIM_FS-ON-W at:
> [<8005f640>] mark_lock+0x15c/0x6d8
> [<80062594>] mark_held_locks+0xa0/0x138
> [<80062dd8>] lockdep_trace_alloc+0x90/0x110
> [<800a8bf4>] __alloc_pages_nodemask+0x70/0x84c
> [<800d7558>] new_slab+0x6c/0x270
> [<80649880>] __slab_alloc.isra.71.constprop.73+0x3b8/0x504
> [<800d83d4>] kmem_cache_alloc+0x120/0x12c
> [<800f380c>] __d_alloc+0x24/0x180
> [<800f3b6c>] d_alloc+0x18/0x68
> [<800e652c>] lookup_dcache+0x8c/0xb0
> [<800e6570>] __lookup_hash+0x20/0x4c
> [<800eb344>] lookup_one_len+0xcc/0x120
> [<80256fc4>] __create_file+0x74/0x214
> [<80257238>] debugfs_create_file+0x30/0x38
> [<80257ffc>] debugfs_create_u32+0x44/0x54
> [<804b65d4>] clk_debug_create_subtree+0x98/0x11c
> [<804b6614>] clk_debug_create_subtree+0xd8/0x11c
> [<80898fb4>] clk_debug_init+0xac/0x13c
> [<800088b8>] do_one_initcall+0x40/0x170
> [<80867c78>] kernel_init_freeable+0x104/0x1d4
> [<80644934>] kernel_init+0x10/0x118
> [<8000e988>] ret_from_fork+0x14/0x2c
> INITIAL USE at:
> [<8005f640>] mark_lock+0x15c/0x6d8
> [<8005fee0>] __lock_acquire+0x324/0x1cc8
> [<80061d9c>] lock_acquire+0x68/0x7c
> [<8064da38>] mutex_trylock+0x110/0x20c
> [<804b5550>] clk_prepare_lock+0x14/0xec
> [<804b7548>] __clk_init+0x20/0x458
> [<804b7a50>] _clk_register+0xd0/0x168
> [<804b7b98>] clk_register+0x3c/0x88
> [<804b857c>] clk_register_fixed_rate+0x88/0xd4
> [<804b8630>] of_fixed_clk_setup+0x68/0x90
> [<80899088>] of_clk_init+0x44/0x6c
> [<8086b2c8>] time_init+0x2c/0x38
> [<808679f0>] start_kernel+0x1dc/0x360
> [<10008074>] 0x10008074
> }
> ... key at: [<808fd198>] prepare_lock+0x38/0x48
> ... acquired at:
> [<8005f170>] check_usage+0x41c/0x5f4
> [<8005f3a0>] check_irq_usage+0x58/0xb4
> [<80060ce0>] __lock_acquire+0x1124/0x1cc8
> [<80061d9c>] lock_acquire+0x68/0x7c
> [<8064f57c>] mutex_lock_nested+0x5c/0x3b4
> [<804b55b4>] clk_prepare_lock+0x78/0xec
> [<804b5fc8>] clk_get_rate+0x14/0x64
> [<80497c74>] esdhc_pltfm_set_clock+0x20/0x2a8
> [<80492b60>] sdhci_set_clock+0x4c/0x424
> [<804944b8>] sdhci_do_set_ios+0x314/0x728
> [<804948f4>] sdhci_set_ios+0x28/0x34
> [<80481560>] mmc_set_bus_mode+0x20/0x24
> [<804858d4>] mmc_attach_mmc+0x140/0x1a0
> [<80482348>] mmc_rescan+0x28c/0x2dc
> [<8003df48>] process_one_work+0x1a4/0x444
> [<8003ec7c>] worker_thread+0x120/0x3e4
> [<8004539c>] kthread+0xcc/0xe8
> [<8000e988>] ret_from_fork+0x14/0x2c
>
>
> stack backtrace:
> CPU: 0 PID: 6 Comm: kworker/u4:0 Not tainted 3.13.0-rc3+ #4
> Workqueue: kmmcd mmc_rescan
> Backtrace:
> [<80012218>] (dump_backtrace+0x0/0x10c) from [<800123b8>] (show_stack+0x18/0x1c)
> r6:00000000 r5:809c53f0 r4:00000000 r3:bf872d00
> [<800123a0>] (show_stack+0x0/0x1c) from [<8064a720>] (dump_stack+0x78/0x94)
> [<8064a6a8>] (dump_stack+0x0/0x94) from [<8005f190>] (check_usage+0x43c/0x5f4)
> r4:00000001 r3:bf872d00
> [<8005ed54>] (check_usage+0x0/0x5f4) from [<8005f3a0>] (check_irq_usage+0x58/0xb4)
> [<8005f348>] (check_irq_usage+0x0/0xb4) from [<80060ce0>] (__lock_acquire+0x1124/0x1cc8)
> r8:bf8730e8 r7:809c53f0 r6:80e46a54 r5:bf8730d0 r4:00000003
> [<8005fbbc>] (__lock_acquire+0x0/0x1cc8) from [<80061d9c>] (lock_acquire+0x68/0x7c)
> [<80061d34>] (lock_acquire+0x0/0x7c) from [<8064f57c>] (mutex_lock_nested+0x5c/0x3b4)
> r7:00000000 r6:80dfc7bc r5:804b55b4 r4:808fd160
> [<8064f520>] (mutex_lock_nested+0x0/0x3b4) from [<804b55b4>] (clk_prepare_lock+0x78/0xec)
> [<804b553c>] (clk_prepare_lock+0x0/0xec) from [<804b5fc8>] (clk_get_rate+0x14/0x64)
> r6:bf020c40 r5:000186a0 r4:bf828380 r3:80497c54
> [<804b5fb4>] (clk_get_rate+0x0/0x64) from [<80497c74>] (esdhc_pltfm_set_clock+0x20/0x2a8)
> r5:000186a0 r4:bf020c40
> [<80497c54>] (esdhc_pltfm_set_clock+0x0/0x2a8) from [<80492b60>] (sdhci_set_clock+0x4c/0x424)
> [<80492b14>] (sdhci_set_clock+0x0/0x424) from [<804944b8>] (sdhci_do_set_ios+0x314/0x728)
> [<804941a4>] (sdhci_do_set_ios+0x0/0x728) from [<804948f4>] (sdhci_set_ios+0x28/0x34)
> [<804948cc>] (sdhci_set_ios+0x0/0x34) from [<80481560>] (mmc_set_bus_mode+0x20/0x24)
> r5:806947d0 r4:bf020800
> [<80481540>] (mmc_set_bus_mode+0x0/0x24) from [<804858d4>] (mmc_attach_mmc+0x140/0x1a0)
> [<80485794>] (mmc_attach_mmc+0x0/0x1a0) from [<80482348>] (mmc_rescan+0x28c/0x2dc)
> r5:806947d0 r4:bf020af8
> [<804820bc>] (mmc_rescan+0x0/0x2dc) from [<8003df48>] (process_one_work+0x1a4/0x444)
> r7:bf9bc100 r6:bf80dc00 r5:bf020af8 r4:bf82d880
> [<8003dda4>] (process_one_work+0x0/0x444) from [<8003ec7c>] (worker_thread+0x120/0x3e4)
> [<8003eb5c>] (worker_thread+0x0/0x3e4) from [<8004539c>] (kthread+0xcc/0xe8)
> [<800452d0>] (kthread+0x0/0xe8) from [<8000e988>] (ret_from_fork+0x14/0x2c)
> r7:00000000 r6:00000000 r5:800452d0 r4:bf840580
> mmc2: BKOPS_EN bit is not set
> mmc2: new high speed MMC card at address 0001
>
> --
> Philippe De Muyter +32 2 6101532 Macq SA rue de l'Aeronef 2 B-1140 Bruxelles
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists