2019-03-21 13:20:57

by Anders Roxell

Subject: [PATCH] scsi: ufs-mediatek: fix uninitialized variable

When building the ufs-mediatek driver variable 'ret' may be used
uninitialized.

../drivers/scsi/ufs/ufs-mediatek.c: In function ‘ufs_mtk_setup_clocks’:
../drivers/scsi/ufs/ufs-mediatek.c:96:6: warning: ‘ret’ may be used uninitialized in this function [-Wmaybe-uninitialized]
int ret;
^~~

Rework to return directly instead of using a local variable 'ret'.

Fixes: ddd90623ce26 ("scsi: ufs-mediatek: Add UFS support for Mediatek SoC chips")
Signed-off-by: Anders Roxell <[email protected]>
---
drivers/scsi/ufs/ufs-mediatek.c | 9 ++++-----
1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/scsi/ufs/ufs-mediatek.c b/drivers/scsi/ufs/ufs-mediatek.c
index ce4767784937..330d68871b06 100644
--- a/drivers/scsi/ufs/ufs-mediatek.c
+++ b/drivers/scsi/ufs/ufs-mediatek.c
@@ -93,7 +93,6 @@ static int ufs_mtk_setup_clocks(struct ufs_hba *hba, bool on,
enum ufs_notify_change_status status)
{
struct ufs_mtk_host *host = ufshcd_get_variant(hba);
- int ret;

/*
* In case ufs_mtk_init() is not yet done, simply ignore.
@@ -106,18 +105,18 @@ static int ufs_mtk_setup_clocks(struct ufs_hba *hba, bool on,
switch (status) {
case PRE_CHANGE:
if (!on)
- ret = phy_power_off(host->mphy);
+ return phy_power_off(host->mphy);
break;
case POST_CHANGE:
if (on)
- ret = phy_power_on(host->mphy);
+ return phy_power_on(host->mphy);
break;
default:
- ret = -EINVAL;
+ return -EINVAL;
break;
}

- return ret;
+ return 0;
}

/**
--
2.20.1

2019-03-21 13:35:24

by Stanley Chu

Subject: Re: [PATCH] scsi: ufs-mediatek: fix uninitialized variable

Hi Anders,

On Thu, 2019-03-21 at 14:19 +0100, Anders Roxell wrote:
> When building the ufs-mediatek driver variable 'ret' may be used
> uninitialized.
>
> ../drivers/scsi/ufs/ufs-mediatek.c: In function ‘ufs_mtk_setup_clocks’:
> ../drivers/scsi/ufs/ufs-mediatek.c:96:6: warning: ‘ret’ may be used uninitialized in this function [-Wmaybe-uninitialized]
> int ret;
> ^~~
>
> Rework to return directly instead of using a local variable 'ret'.
>
> Fixes: ddd90623ce26 ("scsi: ufs-mediatek: Add UFS support for Mediatek SoC chips")
> Signed-off-by: Anders Roxell <[email protected]>
> ---
> drivers/scsi/ufs/ufs-mediatek.c | 9 ++++-----
> 1 file changed, 4 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/scsi/ufs/ufs-mediatek.c b/drivers/scsi/ufs/ufs-mediatek.c
> index ce4767784937..330d68871b06 100644
> --- a/drivers/scsi/ufs/ufs-mediatek.c
> +++ b/drivers/scsi/ufs/ufs-mediatek.c
> @@ -93,7 +93,6 @@ static int ufs_mtk_setup_clocks(struct ufs_hba *hba, bool on,
> enum ufs_notify_change_status status)
> {
> struct ufs_mtk_host *host = ufshcd_get_variant(hba);
> - int ret;
>
> /*
> * In case ufs_mtk_init() is not yet done, simply ignore.
> @@ -106,18 +105,18 @@ static int ufs_mtk_setup_clocks(struct ufs_hba *hba, bool on,
> switch (status) {
> case PRE_CHANGE:
> if (!on)
> - ret = phy_power_off(host->mphy);
> + return phy_power_off(host->mphy);
> break;
> case POST_CHANGE:
> if (on)
> - ret = phy_power_on(host->mphy);
> + return phy_power_on(host->mphy);
> break;
> default:
> - ret = -EINVAL;
> + return -EINVAL;
> break;
> }
>
> - return ret;
> + return 0;
> }
>
> /**

This was already fixed by Nathan Chancellor in below patch,
https://lkml.org/lkml/2019/3/20/784

Anyway thanks for sending this fix.

Stanley.