On Sun, 18 Sep 2022, Philippe Mathieu-Daudé wrote: > Hi Zoltan, > > On 18/9/22 22:24, BALATON Zoltan wrote: >> Move the PPC4xx DDR and DDR2 SDRAM contrller models into a new file >> called ppc4xx_sdram to separate from other device models and put them >> in one place allowing sharing some code between them. >> >> Signed-off-by: BALATON Zoltan >> --- >> hw/ppc/meson.build | 3 +- >> hw/ppc/ppc440_uc.c | 332 ----------------- >> hw/ppc/ppc4xx_devs.c | 414 --------------------- >> hw/ppc/ppc4xx_sdram.c | 771 ++++++++++++++++++++++++++++++++++++++++ >> include/hw/ppc/ppc4xx.h | 24 +- >> 5 files changed, 785 insertions(+), 759 deletions(-) > > This seems a proper cleanup, but even using `git-diff > --color-moved=dimmed-zebra` I'm having hard time reviewing this single > patch. > Looking at the changes in the ppc4xx_sdram_types[] array, it > seems we can be move one model at a time, right? I could try to break this patch up some more if absolutely necessary but I've already spent a lot of time rebasing this series as every little change in earlier patch leads to conflicts later then patches get squashed during rebase and I have to redo it again to separate them. So if you can't review it looking at it some more I could try a v6 but don't want if can be avoided. I'll wait for Cédric's comments too in any case to only do as many respins as needed. Regards, BALATON Zoltan > [...] >> +static const TypeInfo ppc4xx_sdram_types[] = { >> + { >> + .name = TYPE_PPC4xx_SDRAM_DDR, >> + .parent = TYPE_PPC4xx_DCR_DEVICE, >> + .instance_size = sizeof(Ppc4xxSdramDdrState), >> + .class_init = ppc4xx_sdram_ddr_class_init, >> + }, { >> + .name = TYPE_PPC4xx_SDRAM_DDR2, >> + .parent = TYPE_PPC4xx_DCR_DEVICE, >> + .instance_size = sizeof(Ppc4xxSdramDdr2State), >> + .class_init = ppc4xx_sdram_ddr2_class_init, >> + } >> +}; >> + >> +DEFINE_TYPES(ppc4xx_sdram_types) > [...] > >