Undefined dependence type raised by Pluto

38 views
Skip to first unread message

Ruizhe Zhao

unread,
Oct 20, 2020, 5:44:56 AM10/20/20
to Pluto development
Hi there,

I'm trying to use Pluto to process an OpenScop input (see the end of this email) that corresponds to this code snippet - 

for ( i=0; i<N; i++) {
  for ( j=0; j<N; j++) {
    a[ i ][ j ] = a[i ][ j−1] + a[i−1][j ];
  }
}

But while loading this OpenScop file into Pluto, the dependence on this statement turns out to be PLUTO_DEP_UNDEFINED. I'm just wondering is Pluto able to process this kind of input or not, or I should attach some other configuration to make it work.

Thanks in advance for any help!

Best regards
Ruizhe

------------------

# [File generated by the OpenScop Library 0.9.2]

<OpenScop>

# =============================================== Global
# Language
C

# Context
CONTEXT
0 2 0 0 0 0

# Parameters are not provided
0


# Number of statements
1

# =============================================== Statement 1
# Number of relations describing the statement:
5

# ----------------------------------------------  1.1 Domain
DOMAIN
4 4 2 0 0 0
# e/i| i0   i1 |  1  
   1    1    0   -1    ## i0-1 >= 0
   1   -1    0   63    ## -i0+63 >= 0
   1    0    1   -1    ## i1-1 >= 0
   1    0   -1   63    ## -i1+63 >= 0

# ----------------------------------------------  1.2 Scattering
SCATTERING
5 9 5 2 0 0
# e/i| c1   c2   c3   c4   c5 | i0   i1 |  1  
   0   -1    0    0    0    0    0    0    0    ## c1 == 0
   0    0   -1    0    0    0    1    0    0    ## c2 == i0
   0    0    0   -1    0    0    0    0    0    ## c3 == 0
   0    0    0    0   -1    0    0    1    0    ## c4 == i1
   0    0    0    0    0   -1    0    0    0    ## c5 == 0

# ----------------------------------------------  1.3 Access
WRITE
3 7 3 2 0 0
# e/i| Arr  [1]  [2]| i0   i1 |  1  
   0   -1    0    0    0    0    1    ## Arr == A1
   0    0   -1    0    1    0    0    ## [1] == i0
   0    0    0   -1    0    1    0    ## [2] == i1

READ
3 7 3 2 0 0
# e/i| Arr  [1]  [2]| i0   i1 |  1  
   0   -1    0    0    0    0    1    ## Arr == A1
   0    0   -1    0    1    0    0    ## [1] == i0
   0    0    0   -1    0    1   -1    ## [2] == i1-1

READ
3 7 3 2 0 0
# e/i| Arr  [1]  [2]| i0   i1 |  1  
   0   -1    0    0    0    0    1    ## Arr == A1
   0    0   -1    0    1    0   -1    ## [1] == i0-1
   0    0    0   -1    0    1    0    ## [2] == i1

# ----------------------------------------------  1.4 Statement Extensions
# Number of Statement Extensions
1
<body>
# Number of original iterators
2
# List of original iterators
i0 i1
# Statement body expression
S0(A1, 2, A1, 2, A1, 2, i0, i1)
</body>

# =============================================== Extensions
<arrays>
# Number of arrays
1
# Mapping array-identifiers/array-names
1 A1
</arrays>

<scatnames>
c0 i0 c2 i1 c4
</scatnames>

</OpenScop>


Ruizhe Zhao

unread,
Oct 20, 2020, 4:05:30 PM10/20/20
to Pluto development
FYI: After switching to ISL dep (by setting context->options->isldep to 1) Pluto could work (I'm using Pluto as a library and all the options are initialised to 0, so isldep was not set to its default value). So it might be candl's problem.
Reply all
Reply to author
Forward
0 new messages