Incorrect processing of "type var[]" constructions

ID0000629: This issue was created automatically from Mantis Issue 629. Further discussion may take place here.

---

Id	Project	Category	View	Due Date	Updated
ID0000629	Frama-C	Plug-in > jessie	public	2010-11-16	2010-11-17

Reporter	Astra	Assigned To	cmarche	Resolution	open
Priority	normal	Severity	minor	Reproducibility	always
Platform	-	OS	-	OS Version	-
Product Version	Frama-C Boron-20100401	Target Version	-	Fixed in Version	-

Description :

During the processing of axiomatic Permut (taken from the ACSL specification file, page 51) Frama gives the next message:

tester@ubuntu-fm:~/workspace/prac1$ frama-c -jessie -jessie-atp gui qsorti.c [kernel] preprocessing with "gcc -C -E -I. -dD qsorti.c" [jessie] Starting Jessie translation qsorti.c:27:[jessie] failure: Unexpected failure. Please submit bug report (Ref. "norm.ml:1533:10"). [kernel] The full backtrace is: Raised at file "src/kernel/log.ml", line 506, characters 30-31 Called from file "src/kernel/log.ml", line 500, characters 2-9 Re-raised at file "src/kernel/log.ml", line 503, characters 8-9 Called from file "src/lib/type.ml", line 746, characters 40-45 Called from file "queue.ml", line 134, characters 6-20 Called from file "src/kernel/boot.ml", line 50, characters 4-20 Called from file "src/kernel/cmdline.ml", line 170, characters 4-8

    Plug-in jessie aborted because of an internal error.
    Please report as 'crash' at http://bts.frama-c.com

Specified line number corresponds to the next axiom:

@ axiom permut_exchange{L1,L2} : @ \forall double t1[], double t2[], integer i, integer j, integer n; @ \at(t1[i],L1) == \at(t2[j],L2) && @ \at(t1[j],L1) == \at(t2[i],L2) && @ (\forall integer k; 0 <= k < n && k != i && k != j ==> @ \at(t1[k],L1) == \at(t2[k],L2)) @ ==> permut{L1,L2}(t1,t2,n);

The same message is displayed while attempting to process the predicate Swap:

@ predicate Swap{L1,L2}(int a[], integer i, integer j) = @ \at(a[i],L1) == \at(a[j],L2) && @ \at(a[j],L1) == \at(a[i],L2) && @ \forall integer k; k != i && k != j @ ==> \at(a[k],L1) == \at(a[k],L2);

The reason is the incorrect processing of constructions like "type var[]". The equal construction "type* var" causes no error.