[Bug fortran/23905] New: misbehavior of function "gfc_copy_array_spec" in gcc/fortran/array.c

[skyhover at hotmail dot com]

The function is supposed to copy expressions from "src" to "dest", however, at  
line 486 and 487:  
  dest->lower[i] = gfc_copy_expr (dest->lower[i]);  
  dest->upper[i] = gfc_copy_expr (dest->upper[i]);  
"dest" are misused as the arguments in the parentheses. Both of them should be  
"src", I think. Also, such mistyping are still in the CVS. 
  
Thank you.

-- 
   Summary: misbehavior of function "gfc_copy_array_spec" in
gcc/fortran/array.c
   Product: gcc
   Version: 4.0.1
Status: UNCONFIRMED
  Severity: normal
  Priority: P2
 Component: fortran
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: skyhover at hotmail dot com
CC: gcc-bugs at gcc dot gnu dot org
  GCC host triplet: *-*-*


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23905

[Bug fortran/23906] New: inappropriate branch condition in function "transform_sections" in "dependency.c"

[skyhover at hotmail dot com]

Line 422 in the file is a if-statement:  
  if (l_stride != NULL) 
mpz_cdiv_q (X2, X2, r_stride->value.integer); 
 
I am not clear about the functionality of "mpz_cdiv_q", but I think the 
condition should be "r_stride != NULL" by looking at its surroundings and 
complying to certain consistency rules.  
 
Thank you.

-- 
   Summary: inappropriate branch condition in function
"transform_sections" in "dependency.c"
   Product: gcc
   Version: 4.1.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P2
 Component: fortran
AssignedTo: unassigned at gcc dot gnu dot org
    ReportedBy: skyhover at hotmail dot com
CC: gcc-bugs at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23906

[Bug fortran/23907] New: missing switch-case in function "gfc_simplify_radix" in gcc/fortran/simplify.c?

[skyhover at hotmail dot com]

We noticed the function has no switch case for "BT_COMPLEX" by similarity check 
 
against other functions such as "gfc_simplify_range" in the same file. I think 
it's reasonable for a complex to have a radix. So one "case BT_COMPLEX:" may be 
needed just after "case BT_REAL:" at line 2513. Could you please inspect my 
observations? 
 
I also have similar doubts on functions "gfc_simplify_dim", 
"gfc_simplify_huge", "gfc_simplify_sign", and etc. 
 
Thanks a lot.

-- 
   Summary: missing switch-case in function "gfc_simplify_radix" in
gcc/fortran/simplify.c?
   Product: gcc
   Version: 4.0.1
Status: UNCONFIRMED
  Severity: normal
  Priority: P2
 Component: fortran
    AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: skyhover at hotmail dot com
CC: gcc-bugs at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23907

[Bug fortran/23907] missing switch-case in function "gfc_simplify_radix" in gcc/fortran/simplify.c?

[skyhover at hotmail dot com]

--- Additional Comments From skyhover at hotmail dot com  2005-09-16 05:45 
---
Thanks for your information and clarification. I agree now.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23907

[Bug fortran/23905] misbehavior of function "gfc_copy_array_spec" in gcc/fortran/array.c

[skyhover at hotmail dot com]

--- Additional Comments From skyhover at hotmail dot com  2005-09-16 05:53 
---
There are some similar code in function "gfc_copy_array_ref" at line 53-55. 
There, "src" are kind of recursively copied into "dest".

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23905

[Bug fortran/23905] misbehavior of function "gfc_copy_array_spec" in gcc/fortran/array.c

[skyhover at hotmail dot com]

--- Additional Comments From skyhover at hotmail dot com  2005-09-24 22:32 
---
Yes, you made me clear now. "dest->lower" points to the same thing as "src-
>lower" after "*dest=*src". So the loop is semantically right, creating copies 
of expressions pointed to by "src->lower". 

On the other hand, IMHO, it is a "syntax" misuse which obfuscates the meaning 
of the code?

Thanks for your clarification.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23905