Hello, This series converts the SuperH CPU to QOM. Subclasses, the feature flags bugfix and its consequences throughout code as well as the controversial QOM'ification of the SH7750 SoC are postponed.
I tested it using the sh-test-0.2 test image. Available at: git://github.com/afaerber/qemu-cpu.git qom-cpu-sh4.v4 https://github.com/afaerber/qemu-cpu/commits/qom-cpu-sh4.v4 Note: Depends on cris and microblaze series wrt Makefile.target; branch is for testing and cherry-picking, not for pulling. Cf. http://wiki.qemu.org/Features/QOM/CPU#Status Regards, Andreas Cc: Aurélien Jarno <aurel...@aurel32.net> Cc: Peter Maydell <peter.mayd...@linaro.org> v3 -> v4: * Move cpu_reset() to a new realizefn instead of the initfn (inspired by Peter). * Drop feature flags reset bugfix and its follow-up patches for now. v2 -> v3: * #include cpu.h rather than cpu-qom.h in cpu.c to get consistent include order. * Replace duplicated typedef in cpu.h by relocating #include of cpu-qom.h. v1 -> v2: * Rebased, now ordered after microblaze QOM'ification. * Fix LGPLv2.1 license notice ("Library" vs. "Lesser" bug in German GNU webpage). * Split QOM'ification into three steps to show code movements. * Drop subclasses and SoC QOM'ification for now. Andreas Färber (4): target-sh4: QOM'ify CPU target-sh4: QOM'ify CPU reset target-sh4: Start QOM'ifying CPU init target-sh4: Introduce realizefn for SuperHCPU Makefile.target | 1 + target-sh4/cpu-qom.h | 73 +++++++++++++++++++++++++++++++++++ target-sh4/cpu.c | 98 ++++++++++++++++++++++++++++++++++++++++++++++++ target-sh4/cpu.h | 2 + target-sh4/translate.c | 29 ++------------ 5 files changed, 179 insertions(+), 24 deletions(-) create mode 100644 target-sh4/cpu-qom.h create mode 100644 target-sh4/cpu.c -- 1.7.7
