On 20.02.2023 05:41, Herbert Xu wrote:
> On Sat, Feb 18, 2023 at 09:58:07PM +0100, Heiner Kallweit wrote:
>> Use the priv data member of struct hwrng to make the iomem base address
>> available in meson_rng_read(). This allows for removing struct
>> meson_rng_data completely in the next step.
>> __force is used to silence sparse warnings.
>>
>> Signed-off-by: Heiner Kallweit <hkallweit1@xxxxxxxxx>
>> ---
>> drivers/char/hw_random/meson-rng.c | 14 +++++++-------
>> 1 file changed, 7 insertions(+), 7 deletions(-)
>>
>> diff --git a/drivers/char/hw_random/meson-rng.c b/drivers/char/hw_random/meson-rng.c
>> index a4eb8e35f..bf7a6e594 100644
>> --- a/drivers/char/hw_random/meson-rng.c
>> +++ b/drivers/char/hw_random/meson-rng.c
>> @@ -17,16 +17,14 @@
>> #define RNG_DATA 0x00
>>
>> struct meson_rng_data {
>> - void __iomem *base;
>> struct hwrng rng;
>> };
>
> This is too ugly for words. If you're trying to save memory we
> should instead get rid of rng.priv and convert the drivers that
> user it over to this model of embedding struct hwrng inside the
> driver struct.
>
OK, then let's omit patches 4 and 5.
Patches 1-3 have a Reviewed-by, can you apply them as-is or do
you need a resubmit of the series with patch 1-3 only?
> Thanks,
