ASoC: intel: cht_bsw_max98090_ti: Replace GFP_ATOMIC with GFP_KERNEL in snd_cht_mc_probe
authorJia-Ju Bai <baijiaju1990@gmail.com>
Mon, 9 Apr 2018 10:48:10 +0000 (18:48 +0800)
committerMark Brown <broonie@kernel.org>
Mon, 16 Apr 2018 10:53:40 +0000 (11:53 +0100)
snd_cht_mc_probe() is never called in atomic context.
This function is only set as ".probe" in "struct platform_driver".

Despite never getting called from atomic context,
snd_cht_mc_probe() calls devm_kzalloc() with GFP_ATOMIC,
which waits busily for allocation.
GFP_ATOMIC is not necessary and can be replaced with GFP_KERNEL,
to avoid busy waiting and improve the possibility of sucessful allocation.

This is found by a static analysis tool named DCNS written by myself.

Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
sound/soc/intel/boards/cht_bsw_max98090_ti.c

index d3e1c7e12004dded46fd24ac9345758cea556233..db6976f4ddaa28b9eb0e123b93d9952cd1869634 100644 (file)
@@ -391,7 +391,7 @@ static int snd_cht_mc_probe(struct platform_device *pdev)
        int ret_val = 0;
        struct cht_mc_private *drv;
 
-       drv = devm_kzalloc(&pdev->dev, sizeof(*drv), GFP_ATOMIC);
+       drv = devm_kzalloc(&pdev->dev, sizeof(*drv), GFP_KERNEL);
        if (!drv)
                return -ENOMEM;