Re: [PATCH v2] mfd: mt6370: Add the out-of-bound check to prevent the null pointer

[Andy Shevchenko]

On Fri, Oct 21, 2022 at 12:58 PM ChiYuan Huang <u0084500@xxxxxxxxx> wrote:
> Andy Shevchenko <andy.shevchenko@xxxxxxxxx> 於 2022年10月21日 週五 下午5:14寫道：
> > On Fri, Oct 21, 2022 at 12:02 PM ChiYuan Huang <u0084500@xxxxxxxxx> wrote:
> > > Andy Shevchenko <andy.shevchenko@xxxxxxxxx> 於 2022年10月21日 週五 下午4:34寫道：
> > > > On Fri, Oct 21, 2022 at 5:17 AM cy_huang <u0084500@xxxxxxxxx> wrote:

...

> > > > https://www.kernel.org/doc/html/latest/process/submitting-patches.html#backtraces-in-commit-messages
> >
> > Have you had a chance to read this section of the document?
> >
> OK, get it.
> Is the below text enough to express this problem?

It is you who decides, because I don't know what exact problem this
will represent.

> Ex. Testing as below (mt6370 register range from 0 to 0x1ff)
>    rg_bound_show()
>         regmap_raw_read(regmap, 0x200, &val, sizeof(val));

> pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)

For example, why is this line here?

> pc : i2c_smbus_xfer+0x58/0x120
> lr : i2c_smbus_read_i2c_block_data+0x74/0xc0
> Call trace:
>  i2c_smbus_xfer+0x58/0x120
>  i2c_smbus_read_i2c_block_data+0x74/0xc0
>  mt6370_regmap_read+0x40/0x60
>  _regmap_raw_read+0xe4/0x278
>  regmap_raw_read+0xec/0x240
>  rg_bound_show+0xb0/0x120

The rule of thumb is that it's okay to shrink to ~4-5 lines (in most cases).

-- 
With Best Regards,
Andy Shevchenko