2024-06-08 10:21:08

by Erick Archer

Subject: [PATCH] ethernet: sis900: use sizeof(*pointer) instead of sizeof(type)

It is preferred to use sizeof(*pointer) instead of sizeof(type)
due to the type of the variable can change and one needs not
change the former (unlike the latter).

At the same time remove some unnecessary initializations and
refactor a bit to make the code clearer.

This patch has no effect on runtime behavior.

Signed-off-by: Erick Archer <[email protected]>
---
drivers/net/ethernet/sis/sis900.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/sis/sis900.c b/drivers/net/ethernet/sis/sis900.c
index 85b850372efe..5b82c4763de9 100644
--- a/drivers/net/ethernet/sis/sis900.c
+++ b/drivers/net/ethernet/sis/sis900.c
@@ -614,11 +614,10 @@ static int sis900_mii_probe(struct net_device *net_dev)

/* search for total of 32 possible mii phy addresses */
for (phy_addr = 0; phy_addr < 32; phy_addr++) {
- struct mii_phy * mii_phy = NULL;
+ struct mii_phy *mii_phy;
u16 mii_status;
int i;

- mii_phy = NULL;
for(i = 0; i < 2; i++)
mii_status = mdio_read(net_dev, phy_addr, MII_STATUS);

@@ -630,7 +629,8 @@ static int sis900_mii_probe(struct net_device *net_dev)
continue;
}

- if ((mii_phy = kmalloc(sizeof(struct mii_phy), GFP_KERNEL)) == NULL) {
+ mii_phy = kmalloc(sizeof(*mii_phy), GFP_KERNEL);
+ if (!mii_phy) {
mii_phy = sis_priv->first_mii;
while (mii_phy) {
struct mii_phy *phy;
--
2.25.1

2024-06-13 00:55:26

by Jakub Kicinski

Subject: Re: [PATCH] ethernet: sis900: use sizeof(*pointer) instead of sizeof(type)

On Sat, 8 Jun 2024 12:20:33 +0200 Erick Archer wrote:
> It is preferred to use sizeof(*pointer) instead of sizeof(type)
> due to the type of the variable can change and one needs not
> change the former (unlike the latter).
>
> At the same time remove some unnecessary initializations and
> refactor a bit to make the code clearer.
>
> This patch has no effect on runtime behavior.

This is an ancient driver, unfortunately we prefer not to apply minor
code cleanups of this nature unless they are a part of other development
work. We'd be constantly tweaking the 200+ drivers we have, most of
which get zero use.
--
pw-bot: reject