2023-05-29 08:26:53

by Carlos Song

[permalink] [raw]
Subject: [PATCH] i2c: imx-lpi2c: fix type char overflow issue when calculating the clock cycle

From: Clark Wang <[email protected]>

Claim clkhi and clklo as integer type to avoid possible calculation
errors caused by data overflow.

Fixes: a55fa9d0e42e ("i2c: imx-lpi2c: add low power i2c bus driver")
Signed-off-by: Clark Wang <[email protected]>
Signed-off-by: Carlos Song <[email protected]>
---
drivers/i2c/busses/i2c-imx-lpi2c.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/i2c/busses/i2c-imx-lpi2c.c b/drivers/i2c/busses/i2c-imx-lpi2c.c
index 40a4420d4c12..7dd33564cbe1 100644
--- a/drivers/i2c/busses/i2c-imx-lpi2c.c
+++ b/drivers/i2c/busses/i2c-imx-lpi2c.c
@@ -211,8 +211,8 @@ static void lpi2c_imx_stop(struct lpi2c_imx_struct *lpi2c_imx)
/* CLKLO = I2C_CLK_RATIO * CLKHI, SETHOLD = CLKHI, DATAVD = CLKHI/2 */
static int lpi2c_imx_config(struct lpi2c_imx_struct *lpi2c_imx)
{
- u8 prescale, filt, sethold, clkhi, clklo, datavd;
- unsigned int clk_rate, clk_cycle;
+ u8 prescale, filt, sethold, datavd;
+ unsigned int clk_rate, clk_cycle, clkhi, clklo;
enum lpi2c_imx_pincfg pincfg;
unsigned int temp;

--
2.34.1



2023-05-30 18:54:28

by Andi Shyti

[permalink] [raw]
Subject: Re: [PATCH] i2c: imx-lpi2c: fix type char overflow issue when calculating the clock cycle

On Mon, May 29, 2023 at 04:02:51PM +0800, [email protected] wrote:
> From: Clark Wang <[email protected]>
>
> Claim clkhi and clklo as integer type to avoid possible calculation
> errors caused by data overflow.
>
> Fixes: a55fa9d0e42e ("i2c: imx-lpi2c: add low power i2c bus driver")
> Signed-off-by: Clark Wang <[email protected]>
> Signed-off-by: Carlos Song <[email protected]>
> ---
> drivers/i2c/busses/i2c-imx-lpi2c.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/i2c/busses/i2c-imx-lpi2c.c b/drivers/i2c/busses/i2c-imx-lpi2c.c
> index 40a4420d4c12..7dd33564cbe1 100644
> --- a/drivers/i2c/busses/i2c-imx-lpi2c.c
> +++ b/drivers/i2c/busses/i2c-imx-lpi2c.c
> @@ -211,8 +211,8 @@ static void lpi2c_imx_stop(struct lpi2c_imx_struct *lpi2c_imx)
> /* CLKLO = I2C_CLK_RATIO * CLKHI, SETHOLD = CLKHI, DATAVD = CLKHI/2 */
> static int lpi2c_imx_config(struct lpi2c_imx_struct *lpi2c_imx)
> {
> - u8 prescale, filt, sethold, clkhi, clklo, datavd;
> - unsigned int clk_rate, clk_cycle;
> + u8 prescale, filt, sethold, datavd;
> + unsigned int clk_rate, clk_cycle, clkhi, clklo;

looks OK.

Reviewed-by: Andi Shyti <[email protected]>

there is, though, another part I would take care of, this bit:

clklo = clk_cycle - clkhi;
if (clklo < 64)
break;

It might be unlikely, but if "clk_cycle > clkhi" then all the
calculation. Do you mind adding a check:

if (unlikely(clk_cycle > clkhi))
break;

Andi

> enum lpi2c_imx_pincfg pincfg;
> unsigned int temp;
>
> --
> 2.34.1
>

2023-06-01 13:55:53

by Uwe Kleine-König

[permalink] [raw]
Subject: Re: [PATCH] i2c: imx-lpi2c: fix type char overflow issue when calculating the clock cycle

On Tue, May 30, 2023 at 08:50:36PM +0200, Andi Shyti wrote:
> On Mon, May 29, 2023 at 04:02:51PM +0800, [email protected] wrote:
> > From: Clark Wang <[email protected]>
> >
> > Claim clkhi and clklo as integer type to avoid possible calculation
> > errors caused by data overflow.
> >
> > Fixes: a55fa9d0e42e ("i2c: imx-lpi2c: add low power i2c bus driver")
> > Signed-off-by: Clark Wang <[email protected]>
> > Signed-off-by: Carlos Song <[email protected]>
> > ---
> > drivers/i2c/busses/i2c-imx-lpi2c.c | 4 ++--
> > 1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/i2c/busses/i2c-imx-lpi2c.c b/drivers/i2c/busses/i2c-imx-lpi2c.c
> > index 40a4420d4c12..7dd33564cbe1 100644
> > --- a/drivers/i2c/busses/i2c-imx-lpi2c.c
> > +++ b/drivers/i2c/busses/i2c-imx-lpi2c.c
> > @@ -211,8 +211,8 @@ static void lpi2c_imx_stop(struct lpi2c_imx_struct *lpi2c_imx)
> > /* CLKLO = I2C_CLK_RATIO * CLKHI, SETHOLD = CLKHI, DATAVD = CLKHI/2 */
> > static int lpi2c_imx_config(struct lpi2c_imx_struct *lpi2c_imx)
> > {
> > - u8 prescale, filt, sethold, clkhi, clklo, datavd;
> > - unsigned int clk_rate, clk_cycle;
> > + u8 prescale, filt, sethold, datavd;
> > + unsigned int clk_rate, clk_cycle, clkhi, clklo;
>
> looks OK.
>
> Reviewed-by: Andi Shyti <[email protected]>
>
> there is, though, another part I would take care of, this bit:
>
> clklo = clk_cycle - clkhi;
> if (clklo < 64)
> break;
>
> It might be unlikely, but if "clk_cycle > clkhi" then all the
> calculation. Do you mind adding a check:
>
> if (unlikely(clk_cycle > clkhi))
> break;

With clk_cycle being unsigned and clkhi = DIV_ROUND_UP(clk_cycle, 3)
this will be true unless clk_cycle 0 or 1, so I'd say this is quite
likely to trigger?!

I guess you meant something different than you wrote. (And your sentence
no verb.)

Best regards
Uwe

--
Pengutronix e.K. | Uwe Kleine-K?nig |
Industrial Linux Solutions | https://www.pengutronix.de/ |


Attachments:
(No filename) (2.03 kB)
signature.asc (499.00 B)
Download all attachments

2023-06-23 08:17:35

by Wolfram Sang

[permalink] [raw]
Subject: Re: [PATCH] i2c: imx-lpi2c: fix type char overflow issue when calculating the clock cycle

On Mon, May 29, 2023 at 04:02:51PM +0800, [email protected] wrote:
> From: Clark Wang <[email protected]>
>
> Claim clkhi and clklo as integer type to avoid possible calculation
> errors caused by data overflow.
>
> Fixes: a55fa9d0e42e ("i2c: imx-lpi2c: add low power i2c bus driver")
> Signed-off-by: Clark Wang <[email protected]>
> Signed-off-by: Carlos Song <[email protected]>

Applied to for-current, thanks!


Attachments:
(No filename) (442.00 B)
signature.asc (849.00 B)
Download all attachments