rpmsg: mtk_rpmsg: Fix circular locking dependency

During execution of the worker that's used to register rpmsg devices
we are safely locking the channels mutex but, when creating a new
endpoint for such devices, we are registering a IPI on the SCP, which
then makes the SCP to trigger an interrupt, lock its own mutex and in
turn register more subdevices.
This creates a circular locking dependency situation, as the mtk_rpmsg
channels_lock will then depend on the SCP IPI lock.

[   15.447736] ======================================================
[   15.460158] WARNING: possible circular locking dependency detected
[   15.460161] 5.17.0-next-20220324+ #399 Not tainted
[   15.460165] ------------------------------------------------------
[   15.460166] kworker/0:3/155 is trying to acquire lock:
[   15.460170] ffff5b4d0eaf1308 (&scp->ipi_desc[i].lock){+.+.}-{4:4}, at: scp_ipi_lock+0x34/0x50 [mtk_scp_ipi]
[   15.504958]
[]                but task is already holding lock:
[   15.504960] ffff5b4d0e8f1918 (&mtk_subdev->channels_lock){+.+.}-{4:4}, at: mtk_register_device_work_function+0x50/0x1cc [mtk_rpmsg]
[   15.504978]
[]                which lock already depends on the new lock.

[   15.504980]
[]                the existing dependency chain (in reverse order) is:
[   15.504982]
[]               -> #1 (&mtk_subdev->channels_lock){+.+.}-{4:4}:
[   15.504990]        lock_acquire+0x68/0x84
[   15.504999]        __mutex_lock+0xa4/0x3e0
[   15.505007]        mutex_lock_nested+0x40/0x70
[   15.505012]        mtk_rpmsg_ns_cb+0xe4/0x134 [mtk_rpmsg]
[   15.641684]        mtk_rpmsg_ipi_handler+0x38/0x64 [mtk_rpmsg]
[   15.641693]        scp_ipi_handler+0xbc/0x180 [mtk_scp]
[   15.663905]        mt8192_scp_irq_handler+0x44/0xa4 [mtk_scp]
[   15.663915]        scp_irq_handler+0x6c/0xa0 [mtk_scp]
[   15.685779]        irq_thread_fn+0x34/0xa0
[   15.685785]        irq_thread+0x18c/0x240
[   15.685789]        kthread+0x104/0x110
[   15.709579]        ret_from_fork+0x10/0x20
[   15.709586]
[]               -> #0 (&scp->ipi_desc[i].lock){+.+.}-{4:4}:
[   15.731271]        __lock_acquire+0x11e4/0x1910
[   15.740367]        lock_acquire.part.0+0xd8/0x220
[   15.749813]        lock_acquire+0x68/0x84
[   15.757861]        __mutex_lock+0xa4/0x3e0
[   15.766084]        mutex_lock_nested+0x40/0x70
[   15.775006]        scp_ipi_lock+0x34/0x50 [mtk_scp_ipi]
[   15.785503]        scp_ipi_register+0x40/0xa4 [mtk_scp_ipi]
[   15.796697]        scp_register_ipi+0x1c/0x30 [mtk_scp]
[   15.807194]        mtk_rpmsg_create_ept+0xa0/0x108 [mtk_rpmsg]
[   15.818912]        rpmsg_create_ept+0x44/0x60
[   15.827660]        cros_ec_rpmsg_probe+0x15c/0x1f0
[   15.837282]        rpmsg_dev_probe+0x128/0x1d0
[   15.846203]        really_probe.part.0+0xa4/0x2a0
[   15.855649]        __driver_probe_device+0xa0/0x150
[   15.865443]        driver_probe_device+0x48/0x150
[   15.877157]        __device_attach_driver+0xc0/0x12c
[   15.889359]        bus_for_each_drv+0x80/0xe0
[   15.900330]        __device_attach+0xe4/0x190
[   15.911303]        device_initial_probe+0x1c/0x2c
[   15.922969]        bus_probe_device+0xa8/0xb0
[   15.933927]        device_add+0x3a8/0x8a0
[   15.944193]        device_register+0x28/0x40
[   15.954970]        rpmsg_register_device+0x5c/0xa0
[   15.966782]        mtk_register_device_work_function+0x148/0x1cc [mtk_rpmsg]
[   15.983146]        process_one_work+0x294/0x664
[   15.994458]        worker_thread+0x7c/0x45c
[   16.005069]        kthread+0x104/0x110
[   16.014789]        ret_from_fork+0x10/0x20
[   16.025201]
[]               other info that might help us debug this:

[   16.047769]  Possible unsafe locking scenario:

[   16.063942]        CPU0                    CPU1
[   16.075166]        ----                    ----
[   16.086376]   lock(&mtk_subdev->channels_lock);
[   16.097592]                                lock(&scp->ipi_desc[i].lock);
[   16.113188]                                lock(&mtk_subdev->channels_lock);
[   16.129482]   lock(&scp->ipi_desc[i].lock);
[   16.140020]
[]                *** DEADLOCK ***

[   16.158282] 4 locks held by kworker/0:3/155:
[   16.168978]  #0: ffff5b4d00008748 ((wq_completion)events){+.+.}-{0:0}, at: process_one_work+0x1fc/0x664
[   16.190017]  #1: ffff80000953bdc8 ((work_completion)(&mtk_subdev->register_work)){+.+.}-{0:0}, at: process_one_work+0x1fc/0x664
[   16.215269]  #2: ffff5b4d0e8f1918 (&mtk_subdev->channels_lock){+.+.}-{4:4}, at: mtk_register_device_work_function+0x50/0x1cc [mtk_rpmsg]
[   16.242131]  #3: ffff5b4d05964190 (&dev->mutex){....}-{4:4}, at: __device_attach+0x44/0x190

To solve this, simply unlock the channels_lock mutex before calling
mtk_rpmsg_register_device() and relock it right after, as safety is
still ensured by the locking mechanism that happens right after
through SCP.

Fixes: 7017996951 ("rpmsg: add rpmsg support for mt8183 SCP.")
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Link: https://lore.kernel.org/r/20220525091201.14210-1-angelogioacchino.delregno@collabora.com
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
This commit is contained in:
AngeloGioacchino Del Regno 2022-05-25 11:12:01 +02:00 committed by Mathieu Poirier
parent abe13e9a56
commit 353d921468
1 changed files with 2 additions and 0 deletions

View File

@ -234,7 +234,9 @@ static void mtk_register_device_work_function(struct work_struct *register_work)
if (info->registered)
continue;
mutex_unlock(&subdev->channels_lock);
ret = mtk_rpmsg_register_device(subdev, &info->info);
mutex_lock(&subdev->channels_lock);
if (ret) {
dev_err(&pdev->dev, "Can't create rpmsg_device\n");
continue;