Am 25.06.2012 13:46, schrieb Evgeny Voevodin: > On 25.06.2012 13:24, Andreas Färber wrote: >> Am 25.06.2012 09:55, schrieb Oleg Ogurtsov: >>> Signed-off-by: Oleg Ogurtsov<o.ogurt...@samsung.com> >>> --- >>> hw/arm/Makefile.objs | 1 + >>> hw/exynos4210.c | 8 + >>> hw/exynos4210_rtc.c | 607 >>> ++++++++++++++++++++++++++++++++++++++++++++++++++ >>> 3 files changed, 616 insertions(+), 0 deletions(-) >>> create mode 100644 hw/exynos4210_rtc.c >> This RTC like many other Exynos devices has no dependency on the CPU. I >> have a patch in preparation that moves such devices from >> hw/arm/Makefile.objs to hw/Makefile.objs. >> I don't object to this patch, not even minor style nits spotted, >> compliment, but if you have to respin for some reason, it would be nice >> if you could consider that improvement. > > These devices are SOC specific and this SOC is based on ARM only. > Do we really need to move them?
For one, they do not need to be rebuilt when cpu.h changes and they should get the usual device poisoning for proper modeling. For another, someone on IRC started work on an armeb-softmmu, for which we would probably not want to compile in the Exynos devices. Or if we do, we certainly don't want to compile everything twice (cf. xilinx). If devices are ARM-specific and need access to the CPU (e.g., machines, PICs) then according to Paolo they should be placed in hw/arm/ with the new scheme. I'm trying to stay away from moving other people's files around, but the Makefile changes are pretty non-intrusive and can go through arm-devs.next. Cheers, Andreas -- SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg
