akkartik
/
mu
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
mu/cpp/023jump

116 lines
3.0 KiB
Plaintext
Raw Normal View History

1077
2015-04-17 18:22:59 +00:00
//: Jump primitives.
1073 - stop fixing the values of primitive recipes In the process I give up trace stability when I move files around, but I gain in exchange the ability to move files around.
2015-04-17 17:31:17 +00:00
:(before "End Primitive Recipe Declarations")
JUMP,
794 - first jump instruction
2015-02-20 01:28:25 +00:00
:(before "End Primitive Recipe Numbers")
Recipe_number["jump"] = JUMP;
:(before "End Primitive Recipe Implementations")
case JUMP: {
951 - done with break/loop
2015-03-18 07:01:43 +00:00
trace("run") << "ingredient 0 is " << instructions[pc].ingredients[0].value;
949 - paving the way for jumps to labels Addresses for reagents are now computed after all transforms.
2015-03-17 18:23:29 +00:00
pc += instructions[pc].ingredients[0].value;
951 - done with break/loop
2015-03-18 07:01:43 +00:00
trace("run") << "pc now " << pc;
794 - first jump instruction
2015-02-20 01:28:25 +00:00
break;
}
:(scenario "jump_can_skip_instructions")
recipe main [
jump 1:offset
1:integer <- copy 1:literal
]
832 - call-stack for C++ version These #defines and references now span many different layers. Let's see if the lack of encapsulation causes problems. Also interesting to run into a case where I need to modify a foundational layer and touch every single scenario/trace. Only alternative was to duplicate all the different layers that add instructions. Sign of problems with this model?
2015-02-25 09:24:11 +00:00
+run: instruction main/0
794 - first jump instruction
2015-02-20 01:28:25 +00:00
+run: ingredient 0 is 1
832 - call-stack for C++ version These #defines and references now span many different layers. Let's see if the lack of encapsulation causes problems. Also interesting to run into a case where I need to modify a foundational layer and touch every single scenario/trace. Only alternative was to duplicate all the different layers that add instructions. Sign of problems with this model?
2015-02-25 09:24:11 +00:00
-run: instruction main/1
968
2015-03-24 06:59:59 +00:00
-mem: storing 1 in location 1
794 - first jump instruction
2015-02-20 01:28:25 +00:00
:(scenario "jump_backward")
recipe main [
939 - c++: fix an old parsing hack
2015-03-17 05:47:28 +00:00
jump 1:offset # 0 -+
jump 1:offset # | +-+ 1
# \/ /\ |
jump -2:offset # 2 +-->+ |
] # \/ 3
832 - call-stack for C++ version These #defines and references now span many different layers. Let's see if the lack of encapsulation causes problems. Also interesting to run into a case where I need to modify a foundational layer and touch every single scenario/trace. Only alternative was to duplicate all the different layers that add instructions. Sign of problems with this model?
2015-02-25 09:24:11 +00:00
+run: instruction main/0
+run: instruction main/2
+run: instruction main/1
796 - jump instructions done
2015-02-20 02:25:25 +00:00
1073 - stop fixing the values of primitive recipes In the process I give up trace stability when I move files around, but I gain in exchange the ability to move files around.
2015-04-17 17:31:17 +00:00
:(before "End Primitive Recipe Declarations")
JUMP_IF,
796 - jump instructions done
2015-02-20 02:25:25 +00:00
:(before "End Primitive Recipe Numbers")
Recipe_number["jump-if"] = JUMP_IF;
:(before "End Primitive Recipe Implementations")
case JUMP_IF: {
vector<int> arg0 = read_memory(instructions[pc].ingredients[0]);
assert(arg0.size() == 1);
trace("run") << "ingredient 0 is " << arg0[0];
if (!arg0[0]) {
trace("run") << "jump-if fell through";
break;
}
trace("run") << "ingredient 1 is " << instructions[pc].ingredients[1].name;
949 - paving the way for jumps to labels Addresses for reagents are now computed after all transforms.
2015-03-17 18:23:29 +00:00
pc += instructions[pc].ingredients[1].value;
796 - jump instructions done
2015-02-20 02:25:25 +00:00
trace("run") << "jumping to instruction " << pc+1;
break;
}
:(scenario "jump_if")
recipe main [
jump-if 999:literal 1:offset
1:integer <- copy 1:literal
]
832 - call-stack for C++ version These #defines and references now span many different layers. Let's see if the lack of encapsulation causes problems. Also interesting to run into a case where I need to modify a foundational layer and touch every single scenario/trace. Only alternative was to duplicate all the different layers that add instructions. Sign of problems with this model?
2015-02-25 09:24:11 +00:00
+run: instruction main/0
796 - jump instructions done
2015-02-20 02:25:25 +00:00
+run: ingredient 1 is 1
+run: jumping to instruction 2
832 - call-stack for C++ version These #defines and references now span many different layers. Let's see if the lack of encapsulation causes problems. Also interesting to run into a case where I need to modify a foundational layer and touch every single scenario/trace. Only alternative was to duplicate all the different layers that add instructions. Sign of problems with this model?
2015-02-25 09:24:11 +00:00
-run: instruction main/1
968
2015-03-24 06:59:59 +00:00
-mem: storing 1 in location 1
796 - jump instructions done
2015-02-20 02:25:25 +00:00
:(scenario "jump_if_fallthrough")
recipe main [
jump-if 0:literal 1:offset
123:integer <- copy 1:literal
]
832 - call-stack for C++ version These #defines and references now span many different layers. Let's see if the lack of encapsulation causes problems. Also interesting to run into a case where I need to modify a foundational layer and touch every single scenario/trace. Only alternative was to duplicate all the different layers that add instructions. Sign of problems with this model?
2015-02-25 09:24:11 +00:00
+run: instruction main/0
796 - jump instructions done
2015-02-20 02:25:25 +00:00
+run: jump-if fell through
832 - call-stack for C++ version These #defines and references now span many different layers. Let's see if the lack of encapsulation causes problems. Also interesting to run into a case where I need to modify a foundational layer and touch every single scenario/trace. Only alternative was to duplicate all the different layers that add instructions. Sign of problems with this model?
2015-02-25 09:24:11 +00:00
+run: instruction main/1
968
2015-03-24 06:59:59 +00:00
+mem: storing 1 in location 123
796 - jump instructions done
2015-02-20 02:25:25 +00:00
1073 - stop fixing the values of primitive recipes In the process I give up trace stability when I move files around, but I gain in exchange the ability to move files around.
2015-04-17 17:31:17 +00:00
:(before "End Primitive Recipe Declarations")
JUMP_UNLESS,
796 - jump instructions done
2015-02-20 02:25:25 +00:00
:(before "End Primitive Recipe Numbers")
Recipe_number["jump-unless"] = JUMP_UNLESS;
:(before "End Primitive Recipe Implementations")
case JUMP_UNLESS: {
vector<int> arg0 = read_memory(instructions[pc].ingredients[0]);
assert(arg0.size() == 1);
trace("run") << "ingredient 0 is " << arg0[0];
if (arg0[0]) {
trace("run") << "jump-unless fell through";
break;
}
trace("run") << "ingredient 1 is " << instructions[pc].ingredients[1].name;
949 - paving the way for jumps to labels Addresses for reagents are now computed after all transforms.
2015-03-17 18:23:29 +00:00
pc += instructions[pc].ingredients[1].value;
796 - jump instructions done
2015-02-20 02:25:25 +00:00
trace("run") << "jumping to instruction " << pc+1;
break;
}
:(scenario "jump_unless")
recipe main [
jump-unless 0:literal 1:offset
1:integer <- copy 1:literal
]
832 - call-stack for C++ version These #defines and references now span many different layers. Let's see if the lack of encapsulation causes problems. Also interesting to run into a case where I need to modify a foundational layer and touch every single scenario/trace. Only alternative was to duplicate all the different layers that add instructions. Sign of problems with this model?
2015-02-25 09:24:11 +00:00
+run: instruction main/0
796 - jump instructions done
2015-02-20 02:25:25 +00:00
+run: ingredient 1 is 1
+run: jumping to instruction 2
832 - call-stack for C++ version These #defines and references now span many different layers. Let's see if the lack of encapsulation causes problems. Also interesting to run into a case where I need to modify a foundational layer and touch every single scenario/trace. Only alternative was to duplicate all the different layers that add instructions. Sign of problems with this model?
2015-02-25 09:24:11 +00:00
-run: instruction main/1
968
2015-03-24 06:59:59 +00:00
-mem: storing 1 in location 1
796 - jump instructions done
2015-02-20 02:25:25 +00:00
:(scenario "jump_unless_fallthrough")
recipe main [
jump-unless 999:literal 1:offset
123:integer <- copy 1:literal
]
832 - call-stack for C++ version These #defines and references now span many different layers. Let's see if the lack of encapsulation causes problems. Also interesting to run into a case where I need to modify a foundational layer and touch every single scenario/trace. Only alternative was to duplicate all the different layers that add instructions. Sign of problems with this model?
2015-02-25 09:24:11 +00:00
+run: instruction main/0
796 - jump instructions done
2015-02-20 02:25:25 +00:00
+run: ingredient 0 is 999
+run: jump-unless fell through
832 - call-stack for C++ version These #defines and references now span many different layers. Let's see if the lack of encapsulation causes problems. Also interesting to run into a case where I need to modify a foundational layer and touch every single scenario/trace. Only alternative was to duplicate all the different layers that add instructions. Sign of problems with this model?
2015-02-25 09:24:11 +00:00
+run: instruction main/1
968
2015-03-24 06:59:59 +00:00
+mem: storing 1 in location 123