This patch series does a lot of the missing misc work to get the disable-tcg flag working for the ppc architecture. most of the work is code motion, with the notable exception of overhauling the logic of dealing with fpscr.
The mmu logic overhaul has been removed from this patch series as it is pretty complex, and this way we can test the patches a bit better. One caveat is that while testing the commits for regressions, one test of the check-acceptance has failed on patch 2, but my automated script didn't give me any info and I couldn't reproduce it manually. Given how small that change is, I don't think that patch is the culprit, but rather there is a non deterministic edge case that may cause a problem in one of the acceptance tests. based on dgibson's ppc-for-6.1 tree Changelog for v2: * split the patch series * added a fix for 5d145639e, which no longer compiles with linux-user * removed patches ther were already accepted * applied rth's cleanup to ppc_store_sdr1 * changed destination of ppc_store_msr * undone change to helper-proto, now fewer files include it Bruno Larsen (billionai) (7): target/ppc: fix ppc_store_sdr1 for user-only compilation target/ppc: moved ppc_store_lpcr and ppc_store_msr to cpu.c target/ppc: reduce usage of fpscr_set_rounding_mode target/ppc: overhaul and moved logic of storing fpscr target/ppc: removed unnecessary inclusion of helper-proto.h target/ppc: moved ppc_cpu_do_interrupt to cpu.c target/ppc: wrapped some TCG only logic with ifdefs target/ppc/cpu.c | 90 +++++++++++++- target/ppc/cpu.h | 13 ++- target/ppc/cpu_init.c | 17 +-- target/ppc/excp_helper.c | 101 ++++++++-------- target/ppc/fpu_helper.c | 246 +++------------------------------------ target/ppc/gdbstub.c | 7 +- target/ppc/misc_helper.c | 16 --- target/ppc/mmu-hash32.c | 1 - target/ppc/mmu-hash64.c | 8 ++ target/ppc/mmu-radix64.c | 1 - 10 files changed, 181 insertions(+), 319 deletions(-) -- 2.17.1
