On 1/26/21 9:44 AM, Stefan Weil wrote:
> I was not talking about the TODO assertions. When I wrote TCI, I only enabled
> and included code which was triggered by my testing - that's why I said the
> productive code lines have 100 % test coverage. TODO assertions are not
> productive code, but debug code which were made to detect new test cases. They
> were successful, too, because they were triggered by some tests in `make
> check-tcg`.
The TODO assertions are all bugs.
Any *real* dead code detection should have been done in
tcg/tci/tcg-target.c.inc. What's interpreted in tcg/tci.c should be exactly
what is produced on the other side, and you are producing more than you are
consuming.
> It should pass now unless you get timeouts for some tests. Please tell me if
> more TODO assertions are triggered by new tests.
case INDEX_op_ld8s_i32:
TODO();
break;
Can be triggered by
target/arm/translate-a64.c:1061:
tcg_gen_ld8s_i64(tcg_dest, cpu_env, vect_off);
target/arm/translate-a64.c:1090:
tcg_gen_ld8s_i32(tcg_dest, cpu_env, vect_off);
target/arm/translate.c:1210:
tcg_gen_ld8s_i32(dest, cpu_env, off);
target/s390x/translate_vx.c.inc:81:
tcg_gen_ld8s_i64(dst, cpu_env, offs);
target/s390x/translate_vx.c.inc:111:
tcg_gen_ld8s_i32(dst, cpu_env, offs);
case INDEX_op_ld16s_i32:
TODO();
break;
Can be triggered by
target/arm/translate-a64.c:1064:
tcg_gen_ld16s_i64(tcg_dest, cpu_env, vect_off);
target/arm/translate-a64.c:1093:
tcg_gen_ld16s_i32(tcg_dest, cpu_env, vect_off);
target/arm/translate.c:1216:
tcg_gen_ld16s_i32(dest, cpu_env, off);
target/s390x/translate_vx.c.inc:84:
tcg_gen_ld16s_i64(dst, cpu_env, offs);
target/s390x/translate_vx.c.inc:114:
tcg_gen_ld16s_i32(dst, cpu_env, offs);
All of which are target vector instructions.
I'm sure it would be trivial to whip up test cases for them, but I don't see
that as my job.
Please maintain this code properly or give it up.
r~
