Re: [PATCH] ath10k: fix the status check and wrong return

3 messages, 3 authors, 2020-08-18 · open the first message on its own page

Re: [PATCH] ath10k: fix the status check and wrong return

From: Kalle Valo <hidden>
Date: 2020-08-17 14:27:49

Tang Bin [off-list ref] writes:
In the function ath10k_ahb_clock_init(), devm_clk_get() doesn't
return NULL. Thus use IS_ERR() and PTR_ERR() to validate
the returned value instead of IS_ERR_OR_NULL().
Why? What's the benefit of this patch? Or what harm does
IS_ERR_OR_NULL() create?
quoted hunk
Signed-off-by: Zhang Shengju <redacted>
Signed-off-by: Tang Bin <redacted>
---
 drivers/net/wireless/ath/ath10k/ahb.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/drivers/net/wireless/ath/ath10k/ahb.c b/drivers/net/wireless/ath/ath10k/ahb.c
index ed87bc00f..ea669af6a 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.c
+++ b/drivers/net/wireless/ath/ath10k/ahb.c
@@ -87,24 +87,24 @@ static int ath10k_ahb_clock_init(struct ath10k *ar)
 	dev = &ar_ahb->pdev->dev;
 
 	ar_ahb->cmd_clk = devm_clk_get(dev, "wifi_wcss_cmd");
-	if (IS_ERR_OR_NULL(ar_ahb->cmd_clk)) {
+	if (IS_ERR(ar_ahb->cmd_clk)) {
 		ath10k_err(ar, "failed to get cmd clk: %ld\n",
 			   PTR_ERR(ar_ahb->cmd_clk));
-		return ar_ahb->cmd_clk ? PTR_ERR(ar_ahb->cmd_clk) : -ENODEV;
+		return PTR_ERR(ar_ahb->cmd_clk);
 	}
devm_clk_get() can return NULL if CONFIG_HAVE_CLK is disabled:

static inline struct clk *devm_clk_get(struct device *dev, const char *id)
{
	return NULL;
}

-- 
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches

Re: [PATCH] ath10k: fix the status check and wrong return

From: Tang Bin <hidden>
Date: 2020-08-18 01:43:00

Hi Kalle:

在 2020/8/17 22:26, Kalle Valo 写道:
quoted
In the function ath10k_ahb_clock_init(), devm_clk_get() doesn't
return NULL. Thus use IS_ERR() and PTR_ERR() to validate
the returned value instead of IS_ERR_OR_NULL().
Why? What's the benefit of this patch? Or what harm does
IS_ERR_OR_NULL() create?
Thanks for you reply, the benefit of this patch is simplify the code, 
because in

this function, I don't think the situation of 'devm_clk_get() return 
NULL' exists.

So please think about it, thanks.


Tang Bin
quoted
Signed-off-by: Zhang Shengju <redacted>
Signed-off-by: Tang Bin <redacted>
---
  drivers/net/wireless/ath/ath10k/ahb.c | 12 ++++++------
  1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/drivers/net/wireless/ath/ath10k/ahb.c b/drivers/net/wireless/ath/ath10k/ahb.c
index ed87bc00f..ea669af6a 100644
--- a/drivers/net/wireless/ath/ath10k/ahb.c
+++ b/drivers/net/wireless/ath/ath10k/ahb.c
@@ -87,24 +87,24 @@ static int ath10k_ahb_clock_init(struct ath10k *ar)
  	dev = &ar_ahb->pdev->dev;
  
  	ar_ahb->cmd_clk = devm_clk_get(dev, "wifi_wcss_cmd");
-	if (IS_ERR_OR_NULL(ar_ahb->cmd_clk)) {
+	if (IS_ERR(ar_ahb->cmd_clk)) {
  		ath10k_err(ar, "failed to get cmd clk: %ld\n",
  			   PTR_ERR(ar_ahb->cmd_clk));
-		return ar_ahb->cmd_clk ? PTR_ERR(ar_ahb->cmd_clk) : -ENODEV;
+		return PTR_ERR(ar_ahb->cmd_clk);
  	}
devm_clk_get() can return NULL if CONFIG_HAVE_CLK is disabled:

static inline struct clk *devm_clk_get(struct device *dev, const char *id)
{
	return NULL;
}

Re: [PATCH] ath10k: fix the status check and wrong return

From: Steve deRosier <hidden>
Date: 2020-08-18 04:41:22

On Mon, Aug 17, 2020 at 6:43 PM Tang Bin [off-list ref] wrote:
Hi Kalle:

在 2020/8/17 22:26, Kalle Valo 写道:
quoted
quoted
In the function ath10k_ahb_clock_init(), devm_clk_get() doesn't
return NULL. Thus use IS_ERR() and PTR_ERR() to validate
the returned value instead of IS_ERR_OR_NULL().
Why? What's the benefit of this patch? Or what harm does
IS_ERR_OR_NULL() create?
Thanks for you reply, the benefit of this patch is simplify the code,
because in

this function, I don't think the situation of 'devm_clk_get() return
NULL' exists.
I admit I'm not looking at HEAD, but at least in the two versions I've
got checked out, devm_clk_get() can theoretically return NULL. This
feels like a gratuitous change anyway, but in any case it's wrong and
could cause wrong behavior.

- Steve
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help