Skip to content
GitLab
Commit d4748be6 authored by erwan's avatar erwan
Browse files

Fix: current on node calls that had more than one output were not handled properly. 

such as in the equation
 o1,o2 = current (n(x,y,z));
parent 50f8d179
Pipeline #71200 failed with stages
in 1 minute and 30 seconds
Hide whitespace changes
Inline Side-by-side
Makefile.dev
View file @ d4748be6
#######################################################################################
# debug with ocamldebug
debug:
cd bin; dune build main.bc
# generates build/default/bin/main.bc
#######################################################################################
......
bin/dune
View file @ d4748be6
(executable
(name main)
(modes byte exe)
(libraries lutils extlib lustre-v6)
)
......
lib/evalClock.ml
View file @ d4748be6
(* Time-stamp: <modified the 29/08/2019 (at 15:31) by Erwan Jahier> *)
(* Time-stamp: <modified the 19/06/2020 (at 14:43) by Erwan Jahier> *)
open LicEvalConst
......@@ -297,24 +297,24 @@ let rec (f : IdSolver.t -> subst -> Lic.val_exp -> Lxm.t list -> Lic.clock list
assert(ve.ve_clk <> []);
ve, inf_clks, s
and (f_aux : IdSolver.t -> subst -> Lic.val_exp ->
Lic.val_exp * Lic.id_clock list * subst) =
and (f_aux : IdSolver.t -> subst -> Lic.val_exp
-> Lic.val_exp * Lic.id_clock list * subst) =
fun id_solver s ve ->
let ve, cel, s, lxm =
match ve.ve_core with
| CallByPosLic ({it=posop; src=lxm}, args) -> (
let args, cel, s = eval_by_pos_clock id_solver posop lxm args s in
(* current of a constant: we can ignore the current *)
if
(match posop with CURRENT _ -> true | _ -> false) &&
(List.length args = 1) && Lic.val_exp_is_constant (List.hd args)
then
(* current of a constant: we can ignore the current *)
let ve,cel,s = f_aux id_solver s (List.hd args) in
ve, cel, s, lxm
else
let ve =
match posop,args with
| CURRENT None, { ve_clk = (On((cc,cv,ct),cv_clk))::_ ;_ }::_ ->
| CURRENT None, { ve_clk = (On((cc,cv,ct),cv_clk))::clks ;_ }::_ ->
(* We attach the clock constructor to CURRENT and the
clock var to the list of args. Indeed, the user does not need
to specify the clock when it uses current ; hence we add this
......@@ -325,10 +325,14 @@ and (f_aux : IdSolver.t -> subst -> Lic.val_exp ->
let cv_val_exp = { ve_core = cv_val_exp ; ve_typ = [ct] ;
ve_clk = [cv_clk];ve_src = lxm } in
let posop,args = CURRENT (Some cc), cv_val_exp::args in
List.iter
(* all clks should be the same *)
(fun clk -> assert(clk = On((cc,cv,ct),cv_clk))) clks;
let ve = { ve with
ve_core = CallByPosLic ({it=posop; src=lxm}, args) ;
ve_clk = [cv_clk]
} in
ve_clk = cv_clk::(List.map (fun _ -> cv_clk) clks)
}
in
ve
| _ -> { ve with ve_core = CallByPosLic ({it=posop; src=lxm}, args)}
in
......
lib/l2lSplit.ml
View file @ d4748be6
......@@ -66,69 +66,69 @@ let to_be_broken = function
let (break_it_do : val_exp -> val_exp list) =
fun ve ->
let nvel =
match ve.ve_core with
| CallByPosLic({it=Lic.PREDEF_CALL({ it = (("Lustre","if"),[]) ;_ });src=lxm}, [c;ve1;ve2]) ->
let vel1 = get_vel_from_tuple ve1
and vel2 = get_vel_from_tuple ve2
in
assert (List.length vel1 = List.length vel2);
List.map2
(fun ve1 ve2 ->
{ ve_core =
CallByPosLic({it=Lic.PREDEF_CALL(
{ it = (("Lustre","if"),[]);src=lxm });src=lxm},
[c;ve1;ve2]);
ve_typ = ve1.ve_typ;
ve_clk = ve1.ve_clk;
ve_src = lxm
}
)
vel1
vel2
| CallByPosLic({it=WHEN clk; src=lxm}, vel) -> (
let vel = List.flatten (List.map get_vel_from_tuple vel) in
List.map
(fun ve ->
{ ve with
ve_core=CallByPosLic({it=WHEN clk ; src=lxm }, [ve])})
vel
)
| CallByPosLic({it=Lic.TUPLE ; src=_lxm }, vel) -> (remove_tuple vel)
| CallByPosLic({it=op ; src=lxm }, [ve]) ->
let vel = get_vel_from_tuple ve in
List.map
(fun ve -> { ve with ve_core=CallByPosLic({it=op;src=lxm}, [ve])})
vel
| CallByPosLic({it=CURRENT c ; src=lxm }, [clk;ve]) ->
let vel = get_vel_from_tuple ve in
List.map
(fun ve -> { ve with ve_core=CallByPosLic({it=CURRENT c;src=lxm}, [clk;ve])})
vel
| CallByPosLic({it=op ; src=lxm }, [ve1;ve2]) ->
let vel1 = get_vel_from_tuple ve1
and vel2 = get_vel_from_tuple ve2
in
assert (List.length vel1 = List.length vel2);
List.map2
(fun ve1 ve2 ->
{ ve_core = CallByPosLic({it=op ; src=lxm}, [ve1;ve2]);
ve_typ = ve1.ve_typ;
ve_clk = ve1.ve_clk;
ve_src = lxm
}
)
vel1
vel2
| _ -> [ve]
(* assert false (* ougth to be dead code (guarded by to_be_broken...) *) *)
in
let tl = ve.ve_typ
and cl = ve.ve_clk in
assert (List.length ve.ve_typ = List.length nvel);
let nvel = List.map2 (fun nve t -> { nve with ve_typ = [t]; ve_clk=cl } ) nvel ve.ve_typ in
assert(ve.ve_typ = tl);
nvel
let nvel =
match ve.ve_core with
| CallByPosLic({it=Lic.PREDEF_CALL({ it = (("Lustre","if"),[]) ;_ });src=lxm}, [c;ve1;ve2]) ->
let vel1 = get_vel_from_tuple ve1
and vel2 = get_vel_from_tuple ve2
in
assert (List.length vel1 = List.length vel2);
List.map2
(fun ve1 ve2 ->
{ ve_core =
CallByPosLic({it=Lic.PREDEF_CALL(
{ it = (("Lustre","if"),[]);src=lxm });src=lxm},
[c;ve1;ve2]);
ve_typ = ve1.ve_typ;
ve_clk = ve1.ve_clk;
ve_src = lxm
}
)
vel1
vel2
| CallByPosLic({it=WHEN clk; src=lxm}, vel) -> (
let vel = List.flatten (List.map get_vel_from_tuple vel) in
List.map
(fun ve ->
{ ve with
ve_core=CallByPosLic({it=WHEN clk ; src=lxm }, [ve])})
vel
)
| CallByPosLic({it=Lic.TUPLE ; src=_lxm }, vel) -> (remove_tuple vel)
| CallByPosLic({it=op ; src=lxm }, [ve]) ->
let vel = get_vel_from_tuple ve in
List.map
(fun ve -> { ve with ve_core=CallByPosLic({it=op;src=lxm}, [ve])})
vel
| CallByPosLic({it=CURRENT c ; src=lxm }, [clk;ve]) ->
let vel = get_vel_from_tuple ve in
List.map
(fun ve -> { ve with ve_core=CallByPosLic({it=CURRENT c;src=lxm}, [clk;ve])})
vel
| CallByPosLic({it=op ; src=lxm }, [ve1;ve2]) ->
let vel1 = get_vel_from_tuple ve1
and vel2 = get_vel_from_tuple ve2
in
assert (List.length vel1 = List.length vel2);
List.map2
(fun ve1 ve2 ->
{ ve_core = CallByPosLic({it=op ; src=lxm}, [ve1;ve2]);
ve_typ = ve1.ve_typ;
ve_clk = ve1.ve_clk;
ve_src = lxm
}
)
vel1
vel2
| _ -> [ve]
(* assert false (* ougth to be dead code (guarded by to_be_broken...) *) *)
in
let tl = ve.ve_typ
and cl = ve.ve_clk in
assert (List.length ve.ve_typ = List.length nvel);
let nvel = List.map2 (fun nve t -> { nve with ve_typ = [t]; ve_clk=cl } ) nvel ve.ve_typ in
assert(ve.ve_typ = tl);
nvel
let rec (break_it : val_exp -> val_exp list) =
fun ve ->
......
lib/socPredef.ml
View file @ d4748be6
(* Time-stamp: <modified the 29/08/2019 (at 16:45) by Erwan Jahier> *)
(* Time-stamp: <modified the 19/06/2020 (at 14:56) by Erwan Jahier> *)
(** Synchronous Object Code for Predefined operators. *)
......@@ -556,7 +556,8 @@ let (soc_interface_of_pos_op:
let concrete_type = try List.nth types 1 with _ -> assert false in
let soc = of_soc_key lxm (("Lustre::current"), types@[concrete_type], Curr(cc)) in
instanciate_soc soc concrete_type
| Lic.CURRENT (_), _, _ -> assert false (* sno *)
| Lic.CURRENT _, _, _ ->
assert false (* sno *)
| Lic.ARROW, _, _ ->
let concrete_type = List.nth types 0 in
let soc = of_soc_key lxm (("Lustre::arrow"), types@[concrete_type],
......
lv6-ref-man/Makefile
View file @ d4748be6
......@@ -89,8 +89,6 @@ $(MAIN).pdf : $(PARSER) $(SRCS) $(LUS2TEX) $(SUMMARY) $(FIGS) $(OBJPDF)/version
(cd objpdf; pdflatex ../$(MAIN).tex )
mv objpdf/$(MAIN).pdf .
scp:$(MAIN).pdf
scp $(MAIN).pdf jahier@pressembois.imag.fr:/import/www/DIST-TOOLS/SYNCHRONE/lustre-v6/doc
#------------------------------
# Special : fig 2 latex 2 pdf
......
test/lus2lic.sum
View file @ d4748be6
==> lus2lic0.sum <==
Test run by jahier on Fri Jun 12 11:28:54
Test run by jahier on Fri Jun 19 15:04:42
Native configuration is x86_64-pc-linux-gnu
=== lus2lic0 tests ===
......@@ -66,7 +66,7 @@ XFAIL: Test bad programs (assert): test_lus2lic_no_node should_fail/assert/lecte
XFAIL: Test bad programs (assert): test_lus2lic_no_node should_fail/assert/s.lus
==> lus2lic1.sum <==
Test run by jahier on Fri Jun 12 11:28:55
Test run by jahier on Fri Jun 19 15:04:43
Native configuration is x86_64-pc-linux-gnu
=== lus2lic1 tests ===
......@@ -413,7 +413,7 @@ PASS: ./lus2lic {-2c multipar.lus -n multipar}
PASS: sh multipar.sh
==> lus2lic2.sum <==
Test run by jahier on Fri Jun 12 11:29:19
Test run by jahier on Fri Jun 19 15:05:07
Native configuration is x86_64-pc-linux-gnu
=== lus2lic2 tests ===
......@@ -753,7 +753,7 @@ PASS: sh zzz2.sh
PASS: /home/jahier/lus2lic/test/../utils/compare_exec_and_2c zzz2.lus {}
==> lus2lic3.sum <==
Test run by jahier on Fri Jun 12 11:29:48
Test run by jahier on Fri Jun 19 15:05:38
Native configuration is x86_64-pc-linux-gnu
=== lus2lic3 tests ===
......@@ -1267,7 +1267,7 @@ PASS: /home/jahier/lus2lic/test/../utils/test_lus2lic_no_node multipar.lus {}
==> lus2lic4.sum <==
Test run by jahier on Fri Jun 12 11:30:31
Test run by jahier on Fri Jun 19 15:06:23
Native configuration is x86_64-pc-linux-gnu
=== lus2lic4 tests ===
......@@ -1759,7 +1759,7 @@ PASS: /home/jahier/lus2lic/test/../utils/test_lus2lic_no_node zzz2.lus {}
# of expected failures 54
==> lus2lic1.sum <==
PASS: /home/jahier/lus2lic/test/../utils/compare_exec_and_2c multipar.lus 46317 {}
PASS: /home/jahier/lus2lic/test/../utils/compare_exec_and_2c multipar.lus 34137 {}
=== lus2lic1 Summary ===
......@@ -1789,12 +1789,12 @@ PASS: /home/jahier/lus2lic/test/../utils/compare_exec_and_2c multipar.lus 46317
# Total number of failures: 10
lus2lic0.log:testcase ./lus2lic.tests/test0.exp completed in 1 seconds
lus2lic1.log:testcase ./lus2lic.tests/test1.exp completed in 24 seconds
lus2lic2.log:testcase ./lus2lic.tests/test2.exp completed in 29 seconds
lus2lic3.log:testcase ./lus2lic.tests/test3.exp completed in 43 seconds
lus2lic4.log:testcase ./lus2lic.tests/test4.exp completed in 23 seconds
lus2lic2.log:testcase ./lus2lic.tests/test2.exp completed in 31 seconds
lus2lic3.log:testcase ./lus2lic.tests/test3.exp completed in 45 seconds
lus2lic4.log:testcase ./lus2lic.tests/test4.exp completed in 22 seconds
* Ref time:
71.69user 16.57system 1:59.66elapsed 73%CPU (0avgtext+0avgdata 42220maxresident)k
16inputs+147896outputs (0major+8462673minor)pagefaults 0swaps
74.43user 16.16system 2:02.49elapsed 73%CPU (0avgtext+0avgdata 42080maxresident)k
0inputs+151648outputs (0major+8408595minor)pagefaults 0swaps
* Quick time (-j 4):
91.89user 18.09system 0:57.95elapsed 189%CPU (0avgtext+0avgdata 42124maxresident)k
80inputs+139528outputs (0major+8011124minor)pagefaults 0swaps
86.00user 16.60system 1:10.17elapsed 146%CPU (0avgtext+0avgdata 42096maxresident)k
0inputs+146640outputs (0major+8181699minor)pagefaults 0swaps
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment