The driver may sleep under a spinlock, and the function call paths are: gma_power_begin (acquire the spinlock) (drivers/gpu/drm/gma500/power.c) gma_resume_pci pci_set_power_state __pci_start_power_transition (drivers/pci/pci.c) msleep --> may sleep
gma_power_begin (acquire the spinlock) (drivers/gpu/drm/gma500/power.c) gma_resume_pci pci_enable_device pci_enable_device_flags (drivers/pci/pci.c) do_pci_enable_device pci_set_power_state __pci_start_power_transition msleep --> may sleep
To fix them, the spinlock is released before gma_resume_pci, and it is acquired again after gma_resume_pci.
This bug is found by my static analysis tool and my code review.
Signed-off-by: Jia-Ju Bai baijiaju1990@163.com --- drivers/gpu/drm/gma500/power.c | 2 ++ 1 file changed, 2 insertions(+)
diff --git a/drivers/gpu/drm/gma500/power.c b/drivers/gpu/drm/gma500/power.c index bea8578..c355d7e 100644 --- a/drivers/gpu/drm/gma500/power.c +++ b/drivers/gpu/drm/gma500/power.c @@ -264,7 +264,9 @@ bool gma_power_begin(struct drm_device *dev, bool force_on) goto out_false;
/* Ok power up needed */ + spin_unlock_irqrestore(&power_ctrl_lock, flags); ret = gma_resume_pci(dev->pdev); + spin_lock_irqsave(&power_ctrl_lock, flags); if (ret == 0) { psb_irq_preinstall(dev); psb_irq_postinstall(dev);