3.16.83-rc1 review patch. If anyone has any objections, please let me know. ------------------ From: Takashi Iwai <tiwai@xxxxxxx> commit 0aec96f5897ac16ad9945f531b4bef9a2edd2ebd upstream. Jia-Ju Bai reported a possible sleep-in-atomic scenario in the ice1724 driver with Infrasonic Quartet support code: namely, ice->set_rate callback gets called inside ice->reg_lock spinlock, while the callback in quartet.c holds ice->gpio_mutex. This patch fixes the invalid call: it simply moves the calls of ice->set_rate and ice->set_mclk callbacks outside the spinlock. Reported-by: Jia-Ju Bai <baijiaju1990@xxxxxxxxx> Link: https://lore.kernel.org/r/5d43135e-73b9-a46a-2155-9e91d0dcdf83@xxxxxxxxx Link: https://lore.kernel.org/r/20191218192606.12866-1-tiwai@xxxxxxx Signed-off-by: Takashi Iwai <tiwai@xxxxxxx> Signed-off-by: Ben Hutchings <ben@xxxxxxxxxxxxxxx> --- sound/pci/ice1712/ice1724.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) --- a/sound/pci/ice1712/ice1724.c +++ b/sound/pci/ice1712/ice1724.c @@ -663,6 +663,7 @@ static int snd_vt1724_set_pro_rate(struc unsigned long flags; unsigned char mclk_change; unsigned int i, old_rate; + bool call_set_rate = false; if (rate > ice->hw_rates->list[ice->hw_rates->count - 1]) return -EINVAL; @@ -686,7 +687,7 @@ static int snd_vt1724_set_pro_rate(struc * setting clock rate for internal clock mode */ old_rate = ice->get_rate(ice); if (force || (old_rate != rate)) - ice->set_rate(ice, rate); + call_set_rate = true; else if (rate == ice->cur_rate) { spin_unlock_irqrestore(&ice->reg_lock, flags); return 0; @@ -694,12 +695,14 @@ static int snd_vt1724_set_pro_rate(struc } ice->cur_rate = rate; + spin_unlock_irqrestore(&ice->reg_lock, flags); + + if (call_set_rate) + ice->set_rate(ice, rate); /* setting master clock */ mclk_change = ice->set_mclk(ice, rate); - spin_unlock_irqrestore(&ice->reg_lock, flags); - if (mclk_change && ice->gpio.i2s_mclk_changed) ice->gpio.i2s_mclk_changed(ice); if (ice->gpio.set_pro_rate)
