2014-10-11 18:00:11

by Larry Finger

Subject: [PATCH V3.18] rtlwifi: rtl8192ee: Prevent log spamming for switch statements

The driver logs a message when the default branch of switch statements are
taken. Such information is useful when debugging, but these log items should
not be seen for standard usage.

Signed-off-by: Larry Finger <[email protected]>
---
drivers/net/wireless/rtlwifi/rtl8192ee/hw.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/rtlwifi/rtl8192ee/hw.c b/drivers/net/wireless/rtlwifi/rtl8192ee/hw.c
index dfdc9b2..1a87edc 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192ee/hw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192ee/hw.c
@@ -362,7 +362,7 @@ void rtl92ee_get_hw_reg(struct ieee80211_hw *hw, u8 variable, u8 *val)
}
break;
default:
- RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG,
+ RT_TRACE(rtlpriv, COMP_ERR, DBG_DMESG,
"switch case not process %x\n", variable);
break;
}
@@ -591,7 +591,7 @@ void rtl92ee_set_hw_reg(struct ieee80211_hw *hw, u8 variable, u8 *val)
acm_ctrl &= (~ACMHW_BEQEN);
break;
default:
- RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG,
+ RT_TRACE(rtlpriv, COMP_ERR, DBG_DMESG,
"switch case not process\n");
break;
}
@@ -710,7 +710,7 @@ void rtl92ee_set_hw_reg(struct ieee80211_hw *hw, u8 variable, u8 *val)
}
break;
default:
- RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG,
+ RT_TRACE(rtlpriv, COMP_ERR, DBG_DMESG,
"switch case not process %x\n", variable);
break;
}
@@ -2424,7 +2424,7 @@ void rtl92ee_set_key(struct ieee80211_hw *hw, u32 key_index,
enc_algo = CAM_AES;
break;
default:
- RT_TRACE(rtlpriv, COMP_ERR, DBG_EMERG,
+ RT_TRACE(rtlpriv, COMP_ERR, DBG_DMESG,
"switch case not process\n");
enc_algo = CAM_TKIP;
break;
--
1.8.4.5

2014-10-11 18:14:49

by Joe Perches

Subject: Re: [PATCH V3.18] rtlwifi: rtl8192ee: Prevent log spamming for switch statements

On Sat, 2014-10-11 at 12:59 -0500, Larry Finger wrote:
> The driver logs a message when the default branch of switch statements are
> taken. Such information is useful when debugging, but these log items should
> not be seen for standard usage.

Hey Larry.

At some point, it'd be good to make RT_TRACE use
the generic dynamic_debug facility.

Something like the below, but I believe there are
some issues with include ordering and #define DEBUG
when dynamic_debug is not #defined.

---

drivers/net/wireless/rtlwifi/debug.h | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/rtlwifi/debug.h b/drivers/net/wireless/rtlwifi/debug.h
index fc794b3..fbc8185 100644
--- a/drivers/net/wireless/rtlwifi/debug.h
+++ b/drivers/net/wireless/rtlwifi/debug.h
@@ -178,17 +178,14 @@ do { \
do { \
if (unlikely(((comp) & rtlpriv->dbg.global_debugcomponents) && \
((level) <= rtlpriv->dbg.global_debuglevel))) { \
- printk(KERN_DEBUG KBUILD_MODNAME ":%s():<%lx-%x> " fmt, \
- __func__, in_interrupt(), in_atomic(), \
- ##__VA_ARGS__); \
+ pr_debug(fmt, ##__VA_ARGS__); \
} \
} while (0)

#define RTPRINT(rtlpriv, dbgtype, dbgflag, fmt, ...) \
do { \
if (unlikely(rtlpriv->dbg.dbgp_type[dbgtype] & dbgflag)) { \
- printk(KERN_DEBUG KBUILD_MODNAME ": " fmt, \
- ##__VA_ARGS__); \
+ pr_debug(fmt, ##__VA_ARGS__); \
} \
} while (0)