Re: [PATCH v3 3/5] tty: serial: samsung: Remove USI initialization

[Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>]

On 13/12/2021 12:31, Sam Protsenko wrote:
> On Mon, 13 Dec 2021 at 13:00, Krzysztof Kozlowski
> <krzysztof.kozlowski@canonical.com> wrote:
>>
>> On 04/12/2021 20:57, Sam Protsenko wrote:
>>> USI control is now extracted to the dedicated USI driver. Remove USI
>>> related code from serial driver to avoid conflicts and code duplication.
>>>
>>> Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
>>> ---
>>> Changes in v3:
>>>   - Spell check fixes in commit message
>>>
>>> Changes in v2:
>>>   - (none)
>>>
>>>  drivers/tty/serial/samsung_tty.c | 36 ++++----------------------------
>>>  include/linux/serial_s3c.h       |  9 --------
>>>  2 files changed, 4 insertions(+), 41 deletions(-)
>>>
>>
>> Hi Sam,
>>
>> Does this patch depend on USI driver? In cover letter you did not
>> mention any dependency, so this can go via Greg's tree, right?
>>
> 
> Hi Krzysztof,
> 
> TL;DR: Serial patches from this series don't really depend on USI
> driver and can go via Greg's tree.
> 
> I'd say ideally those should be applied right after USI driver
> patches. Because otherwise we'd have double initialization of USI
> block (from USI driver and from serial driver), which wasn't tested
> and may lead to not functional USI (though I think it should be fine).
> That's why I decided to keep those in one single patch series. But if
> we don't care about some short period of not tested behavior -- then
> those can go via Greg's tree.

I think it is fine considering that only one board will be affected and
it will be already non-bisectable due to DTS/driver changes going
separate branches. More important is to get USI and serial driver
changes in the same kernel release, so in v5.17.


Best regards,
Krzysztof