On Tue, Aug 25, 2020 at 01:58:33PM -0700, Richard Henderson wrote: > Well, this is larger than I expected. >
Wow, thanks for working on this Richard! I'll run some tests on it and go through the patches. Thanks, Edgar > I started off thinking conversion to decodetree would be quick, > after I reviewed the mttcg patches last week. Then I realized > that this could also use conversion to the generic translation loop. > Then I realized that there were a number of bugs, and some > inefficiencies, that could be fixed by using tcg exception unwind > properly. > > Hopefully most of these are small and easy to understand. > > I begin by adding enough stuff to test/tcg to make use of a > self-built musl cross-environment. I'll note that linuxtest > does not pass before or after this patch set -- I think that > linux-user/microblaze/signal.c needs work -- but that the > other tests do work. > > I also have an old copy of a microblaze system test image, > which I think used to be hosted on our wiki. After basic kernel > boot, it contains a "selftest" script which runs a bunch of > user-land isa tests. That still works fine too. > > HOWEVER: That's not nearly complete. There are cpu config options > that aren't default and I don't know how to change or test. > > In addition, the manual is really not clear on what's supposed to > happen under various edge conditions, especially with MSR[EE] unset. > E.g. unaligned access: Does the insn get nop-ed out? Do the low > bits of the address get ignored? Right now (before and after) the > access simply happens unaligned, which doesn't seem correct. > I assume the reason for having this configure option is to reduce > the size of the FPGA so that the unaligned access handling hw > simply isn't present. > > Lemme know what you think. > > > r~ > > > Richard Henderson (77): > tests/tcg: Add microblaze to arches filter > tests/tcg: Do not require FE_TOWARDZERO > tests/tcg: Do not require FE_* exception bits > target/microblaze: Tidy gdbstub > target/microblaze: Split out PC from env->sregs > target/microblaze: Split out MSR from env->sregs > target/microblaze: Split out EAR from env->sregs > target/microblaze: Split out ESR from env->sregs > target/microblaze: Split out FSR from env->sregs > target/microblaze: Split out BTR from env->sregs > target/microblaze: Split out EDR from env->sregs > target/microblaze: Split the cpu_SR array > target/microblaze: Fix width of PC and BTARGET > target/microblaze: Fix width of MSR > target/microblaze: Fix width of ESR > target/microblaze: Fix width of FSR > target/microblaze: Fix width of BTR > target/microblaze: Fix width of EDR > target/microblaze: Remove cpu_ear > target/microblaze: Tidy raising of exceptions > target/microblaze: Mark raise_exception as noreturn > target/microblaze: Remove helper_debug and env->debug > target/microblaze: Rename env_* tcg variables to cpu_* > target/microblaze: Tidy mb_tcg_init > target/microblaze: Split out MSR[C] to its own variable > target/microblaze: Use DISAS_NORETURN > target/microblaze: Check singlestep_enabled in gen_goto_tb > target/microblaze: Convert to DisasContextBase > target/microblaze: Convert to translator_loop > target/microblaze: Remove SIM_COMPAT > target/microblaze: Remove DISAS_GNU > target/microblaze: Remove empty D macros > target/microblaze: Remove LOG_DIS > target/microblaze: Ensure imm constant is always available > target/microblaze: Add decodetree infrastructure > target/microblaze: Convert dec_add to decodetree > target/microblaze: Convert dec_sub to decodetree > target/microblaze: Implement cmp and cmpu inline > target/microblaze: Convert dec_pattern to decodetree > target/microblaze: Convert dec_and, dec_or, dec_xor to decodetree > target/microblaze: Convert dec_mul to decodetree > target/microblaze: Convert dec_div to decodetree > target/microblaze: Unwind properly when raising divide-by-zero > target/microblaze: Convert dec_bit to decodetree > target/microblaze: Convert dec_barrel to decodetree > target/microblaze: Convert dec_imm to decodetree > target/microblaze: Convert dec_fpu to decodetree > target/microblaze: Fix cpu unwind for fpu exceptions > target/microblaze: Mark fpu helpers TCG_CALL_NO_WG > target/microblaze: Replace MSR_EE_FLAG with MSR_EE > target/microblaze: Cache mem_index in DisasContext > target/microblaze: Fix cpu unwind for stackprot > target/microblaze: Convert dec_load and dec_store to decodetree > target/microblaze: Assert no overlap in flags making up tb_flags > target/microblaze: Move bimm to BIMM_FLAG > target/microblaze: Store "current" iflags in insn_start > tcg: Add tcg_get_insn_start_param > target/microblaze: Use cc->do_unaligned_access > target/microblaze: Replace clear_imm with tb_flags_to_set > target/microblaze: Replace delayed_branch with tb_flags_to_set > target/microblaze: Tidy mb_cpu_dump_state > target/microblaze: Try to keep imm and delay slot together > target/microblaze: Convert brk and brki to decodetree > target/microblaze: Convert mbar to decodetree > target/microblaze: Reorganize branching > target/microblaze: Use tcg_gen_lookup_and_goto_ptr > target/microblaze: Convert dec_br to decodetree > target/microblaze: Convert dec_bcc to decodetree > target/microblaze: Convert dec_rts to decodetree > target/microblaze: Tidy do_rti, do_rtb, do_rte > target/microblaze: Convert msrclr, msrset to decodetree > target/microblaze: Convert dec_msr to decodetree > target/microblaze: Convert dec_stream to decodetree > target/microblaze: Remove last of old decoder > target/microblaze: Remove cpu_R[0] > target/microblaze: Add flags markup to some helpers > target/microblaze: Reduce linux-user address space to 32-bit > > include/tcg/tcg.h | 15 + > target/microblaze/cpu-param.h | 15 + > target/microblaze/cpu.h | 67 +- > target/microblaze/helper.h | 49 +- > tests/tcg/multiarch/float_helpers.h | 17 + > target/microblaze/insns.decode | 253 +++ > linux-user/elfload.c | 9 +- > linux-user/microblaze/cpu_loop.c | 26 +- > linux-user/microblaze/signal.c | 8 +- > target/microblaze/cpu.c | 9 +- > target/microblaze/gdbstub.c | 193 +- > target/microblaze/helper.c | 164 +- > target/microblaze/mmu.c | 4 +- > target/microblaze/op_helper.c | 175 +- > target/microblaze/translate.c | 3250 ++++++++++++++------------- > tests/tcg/multiarch/float_convs.c | 2 + > tests/tcg/multiarch/float_madds.c | 2 + > target/microblaze/meson.build | 3 + > tests/tcg/configure.sh | 2 +- > 19 files changed, 2309 insertions(+), 1954 deletions(-) > create mode 100644 target/microblaze/insns.decode > > -- > 2.25.1 >
