Skip to contentSkip to conversation

i2c: designware: use u64 for clock freq to avoid u32 multiplication overflow

PATCH Superseded Hanna Hawa 2 participants 2 messages Dec 14, 2022

Subsystems (3)

v1 of 5 Superseded Latest is v5 →

lore
Share & tools
lore

Hanna Hawa

author
#0 lore raw

PATCH 1/1 i2c: designware: use u64 for clock freq to avoid u32 multiplication overflow

From: Lareine Khawaly <[email protected]>

In functions i2c_dw_scl_lcnt() and i2c_dw_scl_hcnt() may have overflow by depending on the values of the given parameters including the ic_clk. For example in our usecase where ic_clk >= 1000000, multiplication of ic_clk * 4700 will result in 32 bit overflow.

This commit change the ic_clk to be u64 parameter to avoid the overflow.

---
 drivers/i2c/busses/i2c-designware-common.c | 4 ++--
 drivers/i2c/busses/i2c-designware-core.h   | 4 ++--
 drivers/i2c/busses/i2c-designware-master.c | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)
drivers/i2c/busses/i2c-designware-common.c +2 -2
index c023b691441e..61a6b7bb8935 100644
@@ -332,7 +332,7 @@ void i2c_dw_adjust_bus_speed(struct dw_i2c_dev *dev)
 }
 EXPORT_SYMBOL_GPL(i2c_dw_adjust_bus_speed);
 
-u32 i2c_dw_scl_hcnt(u32 ic_clk, u32 tSYMBOL, u32 tf, int cond, int offset)
+u32 i2c_dw_scl_hcnt(u64 ic_clk, u32 tSYMBOL, u32 tf, int cond, int offset)
 {
 	/*
 	 * DesignWare I2C core doesn't seem to have solid strategy to meet
@@ -370,7 +370,7 @@ u32 i2c_dw_scl_hcnt(u32 ic_clk, u32 tSYMBOL, u32 tf, int cond, int offset)
 		return DIV_ROUND_CLOSEST(ic_clk * (tSYMBOL + tf), MICRO) - 3 + offset;
 }
 
-u32 i2c_dw_scl_lcnt(u32 ic_clk, u32 tLOW, u32 tf, int offset)
+u32 i2c_dw_scl_lcnt(u64 ic_clk, u32 tLOW, u32 tf, int offset)
 {
 	/*
 	 * Conditional expression:
drivers/i2c/busses/i2c-designware-core.h +2 -2
index 4d3a3b464ecd..aaba6f9977b6 100644
@@ -319,8 +319,8 @@ struct i2c_dw_semaphore_callbacks {
 };
 
 int i2c_dw_init_regmap(struct dw_i2c_dev *dev);
-u32 i2c_dw_scl_hcnt(u32 ic_clk, u32 tSYMBOL, u32 tf, int cond, int offset);
-u32 i2c_dw_scl_lcnt(u32 ic_clk, u32 tLOW, u32 tf, int offset);
+u32 i2c_dw_scl_hcnt(u64 ic_clk, u32 tSYMBOL, u32 tf, int cond, int offset);
+u32 i2c_dw_scl_lcnt(u64 ic_clk, u32 tLOW, u32 tf, int offset);
 int i2c_dw_set_sda_hold(struct dw_i2c_dev *dev);
 unsigned long i2c_dw_clk_rate(struct dw_i2c_dev *dev);
 int i2c_dw_prepare_clk(struct dw_i2c_dev *dev, bool prepare);
drivers/i2c/busses/i2c-designware-master.c +1 -1
index 478318b1d35f..d6960507ea46 100644
@@ -44,7 +44,7 @@ static int i2c_dw_set_timings_master(struct dw_i2c_dev *dev)
 	u32 sda_falling_time, scl_falling_time;
 	struct i2c_timings *t = &dev->timings;
 	const char *fp_str = "";
-	u32 ic_clk;
+	u64 ic_clk;
 	int ret;
 
 	ret = i2c_dw_acquire_lock(dev);
Signature · 1 line
2.38.1
#1 lore raw
Quoted 5 lines · Hanna Hawa

From: Lareine Khawaly <[email protected]>

In functions i2c_dw_scl_lcnt() and i2c_dw_scl_hcnt() may have overflow by depending on the values of the given parameters including the ic_clk. For example in our usecase where ic_clk >= 1000000, multiplication of

It's hard to count 0:s, can you use plain words instead (like million, billion)?

Quoted 3 lines

ic_clk * 4700 will result in 32 bit overflow.

This commit change the ic_clk to be u64 parameter to avoid the overflow.

Read Submitting Patches about imperative mood in the commit messages.

...

Smells like you are missing the Fixes tag.

Signature · 2 lines
With Best Regards,
Andy Shevchenko