Hi Krzysztof Kozlowski, > -----Original Message----- > From: Krzysztof Kozlowski <krzk@xxxxxxxxxx> > Sent: 24 January 2025 13:41 > Subject: Re: [PATCH 3/5] watchdog: Make RZV2HWDT driver depend on ARCH_R9A09G47 > > On 24/01/2025 14:24, Biju Das wrote: > >>>> > >>>>> > >>>>> > >>>>> config K3_RTI_WATCHDOG > >>>>> tristate "Texas Instruments K3 RTI watchdog" > >>>>> depends on ARCH_K3 || COMPILE_TEST > >>>> > >>>> Dependency on ARCH. > >>>> > >>>> Do you understand the difference between ARCH and SoC > >>>> (ARCH_R9A09G47 is the SoC - individual or family)? > >>> > >>> ARCH_R9A09G47 --> Is a SoC (RZ/G3E) > >>> ARCH_R9A09G57 --> Is a SoC (RZ/V2H) > >>> > >>> 90%of IP between these SoCs are same. So can't this belongs to same family of SoCs(eg: > >> ARCH_RZ_G3E_V2H family)? > >> > >> > >> We do not discuss what these SoCs belong to. How does it matter if > >> you create ARCH_RZ_ONE_TWO_THREE_SOCS? Your dependency is ARCH, so unified kernel image will be > easier to create. > >> This is not helping in unified image and Greg was talking about this *multiple times*. > > > Renesas needs to finally start integrating the drivers and treating your SoCs in unified way leading > to unified kernel image. You have four watchdog drivers, which look like could be combined into two > drivers (so each pair of drivers could be combined). For entire Qualcomm family supporting more SoCs > than Renesas there is only one driver. You mean combining SoC IP differences into a hw info struct and call function pointers to achieve same functionality?? What will be the image size saving in that case? Cheers, Biju
