On 12 February 2015 at 00:00, Chen Gang S <gang.c...@sunrus.com.cn> wrote: > After load elf64 tilegx binary for linux-user, the working flow reaches > the first correct instruction postion "__start". Next, we shall load all > instructions for qemu using.
> --- > configure | 7 + > default-configs/tile-linux-user.mak | 1 + > include/elf.h | 3 + > include/hw/elf_ops.h | 7 + > linux-user/elfload.c | 24 +++ > linux-user/main.c | 81 +++++++++ > linux-user/syscall_defs.h | 34 +++- > linux-user/tile/syscall.h | 90 ++++++++++ > linux-user/tile/syscall_nr.h | 327 > ++++++++++++++++++++++++++++++++++++ > linux-user/tile/target_cpu.h | 35 ++++ > linux-user/tile/target_signal.h | 28 +++ > linux-user/tile/target_structs.h | 48 ++++++ > linux-user/tile/termbits.h | 285 +++++++++++++++++++++++++++++++ > target-tile/Makefile.objs | 1 + > target-tile/cpu-qom.h | 72 ++++++++ > target-tile/cpu.c | 159 ++++++++++++++++++ > target-tile/cpu.h | 84 +++++++++ > target-tile/helper.h | 0 > target-tile/translate.c | 54 ++++++ > 19 files changed, 1337 insertions(+), 3 deletions(-) A diffstat like this for a single patch indicates that you *must* break this down into multiple patches for it to be sensibly reviewable. Each patch should provide a small but coherent part of the code. I suggest you look at the mailing list archives and git history for other new targets to see how this kind of "add a new architecture" can be broken up into multiple patches (Based purely on the number of lines added here I would expect this to end up at at least 5 and probably more like 10 patches or maybe more, but aim for a sensible split of the functionality rather than a particular number of patches. If in doubt, prefer more smaller patches to fewer larger ones.) thanks -- PMM
