RPMs/gdb
Watch
1
0
Fork
You've already forked gdb
 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

VLA (Fortran dynamic arrays) strides (multi-dimensional subarrays) from Intel.

Browse source
This commit is contained in:
Jan Kratochvil 2016-01-09 11:58:30 +01:00
commit 046f33b589
10 changed files with 1944 additions and 19 deletions
Download patch file Download diff file Expand all files Collapse all files

402
gdb-fortran-stride-intel-5of6.patch Normal file
View file

@ -0,0 +1,402 @@
From: Christoph Weinmann <christoph.t.weinmann@intel.com>
[PATCH 5/6] fortran: calculate subarray with stride values.
https://sourceware.org/ml/gdb-patches/2015-12/msg00011.html
Message-Id: <1448976075-11456-6-git-send-email-christoph.t.weinmann@intel.com>
Calculate elements of a subarray using a provided stride value
The stride value can be a positive or negative integer, but may
not be zero. If no stride is provided, use the default value
1 to print all elements inside the range.
1| program prog
2| integer :: ary(10) = (/ (i, i=1, 10) /)
3| end program prog
(gdb) print ary(1:10:2)
$3 = (1, 3, 5, 7, 9)
2013-11-27 Christoph Weinmann <christoph.t.weinmann>
* eval.c (value_f90_subarray): Add range size calculation
for stride based ranges, and evaluation of user stride
parameters. Add check for matching user input to array
bounds.
* valops.c (value_slice): Add call parameter with default
stride value for calling value_slice_1.
* valops.c (value_slice_1): Add function parameter for
stride length in the return subarray. Calculate array
elements based on stride value.
* value.h: Add stride parameter to declaration of
value_slice_1.
Signed-off-by: Christoph Weinmann <christoph.t.weinmann@intel.com>
---
gdb/eval.c | 110 +++++++++++++++++++++++++++++++++++++++++++++------------
gdb/valops.c | 85 ++++++++++++++++++++++++++++++++------------
gdb/value.h | 2 +-
3 files changed, 150 insertions(+), 47 deletions(-)
diff --git a/gdb/eval.c b/gdb/eval.c
index 15b2ad4..b8cd080 100644
--- a/gdb/eval.c
+++ b/gdb/eval.c
@@ -437,8 +437,8 @@ value_f90_subarray (struct value *array, struct expression *exp,
{
struct subscript_range
{
- enum f90_range_type f90_range_type;
- LONGEST low, high, stride;
+ enum f90_range_type f90_range_type;
+ LONGEST low, high, stride;
}
range;
LONGEST number;
@@ -475,7 +475,7 @@ value_f90_subarray (struct value *array, struct expression *exp,
range = &index->range;
*pos += 3;
- range->f90_range_type = longest_to_int (exp->elts[pc].longconst);
+ range->f90_range_type = exp->elts[pc].longconst;
/* If a lower bound was provided by the user, the bit has been
set and we can assign the value from the elt stack. Same for
@@ -484,6 +484,7 @@ value_f90_subarray (struct value *array, struct expression *exp,
== SUBARRAY_LOW_BOUND)
range->low = value_as_long (evaluate_subexp (NULL_TYPE, exp,
pos, noside));
+
if ((range->f90_range_type & SUBARRAY_HIGH_BOUND)
== SUBARRAY_HIGH_BOUND)
range->high = value_as_long (evaluate_subexp (NULL_TYPE, exp,
@@ -496,6 +497,10 @@ value_f90_subarray (struct value *array, struct expression *exp,
/* Assign the default stride value '1'. */
else
range->stride = 1;
+
+ /* Check the provided stride value is illegal, aka '0'. */
+ if (range->stride == 0)
+ error (_("Stride must not be 0"));
}
/* User input is an index. E.g.: "p arry(5)". */
else
@@ -512,10 +517,8 @@ value_f90_subarray (struct value *array, struct expression *exp,
}
- /* Traverse the array from right to left and evaluate each corresponding
- user input. VALUE_SUBSCRIPT is called for every index, until a range
- expression is evaluated. After a range expression has been evaluated,
- every subsequent expression is also treated as a range. */
+ /* Traverse the array from right to left and set the high and low bounds
+ for later use. */
for (i = nargs - 1; i >= 0; i--)
{
struct subscript_store *index = &subscript_array[i];
@@ -548,6 +551,48 @@ value_f90_subarray (struct value *array, struct expression *exp,
|| range->high > TYPE_HIGH_BOUND (index_type))
error (_("provided bound(s) outside array bound(s)"));
+ /* For a negative stride the lower boundary must be larger than the
+ upper boundary.
+ For a positive stride the lower boundary must be smaller than the
+ upper boundary. */
+ if ((range->stride < 0 && range->low < range->high)
+ || (range->stride > 0 && range->low > range->high))
+ error (_("Wrong value provided for stride and boundaries"));
+
+ }
+ break;
+
+ case SUBSCRIPT_INDEX:
+ break;
+
+ }
+
+ array_type = TYPE_TARGET_TYPE (array_type);
+ }
+
+ /* Reset ARRAY_TYPE before slicing.*/
+ array_type = check_typedef (value_type (new_array));
+
+ /* Traverse the array from right to left and evaluate each corresponding
+ user input. VALUE_SUBSCRIPT is called for every index, until a range
+ expression is evaluated. After a range expression has been evaluated,
+ every subsequent expression is also treated as a range. */
+ for (i = nargs - 1; i >= 0; i--)
+ {
+ struct subscript_store *index = &subscript_array[i];
+ struct type *index_type = TYPE_INDEX_TYPE (array_type);
+
+ switch (index->kind)
+ {
+ case SUBSCRIPT_RANGE:
+ {
+
+ /* When we hit the first range specified by the user, we must
+ treat any subsequent user entry as a range. We simply
+ increment DIM_COUNT which tells us how many times we are
+ calling VALUE_SLICE_1. */
+ struct subscript_range *range = &index->range;
+
/* DIM_COUNT counts every user argument that is treated as a range.
This is necessary for expressions like 'print array(7, 8:9).
Here the first argument is a literal, but must be treated as a
@@ -555,10 +600,9 @@ value_f90_subarray (struct value *array, struct expression *exp,
dim_count++;
new_array
- = value_slice_1 (new_array,
- longest_to_int (range->low),
- longest_to_int (range->high - range->low + 1),
- dim_count);
+ = value_slice_1 (new_array, range->low,
+ range->high - range->low + 1,
+ range->stride, dim_count);
}
break;
@@ -572,27 +616,38 @@ value_f90_subarray (struct value *array, struct expression *exp,
to get the value offset right. */
if (dim_count == 0)
new_array
- = value_subscripted_rvalue (new_array, index->number,
+ = value_subscripted_rvalue (new_array, index->number,
f77_get_lowerbound (value_type
(new_array)));
else
{
- /* Check for valid index input. */
+ dim_count++;
+
+ /* We might end up here, because we have to treat the provided
+ index like a range. But now VALUE_SUBSCRIPTED_RVALUE
+ cannot do the range checks for us. So we have to make sure
+ ourselves that the user provided index is inside the
+ array bounds. Throw an error if not. */
if (index->number < TYPE_LOW_BOUND (index_type)
- || index->number > TYPE_HIGH_BOUND (index_type))
- error (_("error no such vector element"));
+ && index->number < TYPE_HIGH_BOUND (index_type))
+ error (_("provided bound(s) outside array bound(s)"));
+
+ if (index->number > TYPE_LOW_BOUND (index_type)
+ && index->number > TYPE_HIGH_BOUND (index_type))
+ error (_("provided bound(s) outside array bound(s)"));
- dim_count++;
new_array = value_slice_1 (new_array,
- longest_to_int (index->number),
- 1, /* length is '1' element */
+ index->number,
+ 1, /* COUNT is '1' element */
+ 1, /* STRIDE set to '1' */
dim_count);
}
}
break;
}
- }
+ array_type = TYPE_TARGET_TYPE (array_type);
+ }
/* With DIM_COUNT > 1 we currently have a one dimensional array, but expect
an array of arrays, depending on how many ranges have been provided by
@@ -617,7 +672,9 @@ value_f90_subarray (struct value *array, struct expression *exp,
the output array. So we traverse the SUBSCRIPT_ARRAY again, looking
for a range entry. When we find one, we use the range info to create
an additional range_type to set the correct bounds and dimensions for
- the output array. */
+ the output array. In addition, we may have a stride value that is not
+ '1', forcing us to adjust the number of elements in a range, according
+ to the stride value. */
for (i = 0; i < nargs; i++)
{
struct subscript_store *index = &subscript_array[i];
@@ -625,12 +682,19 @@ value_f90_subarray (struct value *array, struct expression *exp,
if (index->kind == SUBSCRIPT_RANGE)
{
struct type *range_type, *interim_array_type;
+ int new_length;
+
+ /* The length of a sub-dimension with all elements between the
+ bounds plus the start element itself. It may be modified by
+ a user provided stride value. */
+ new_length = index->range.high - index->range.low;
+ new_length /= index->range.stride;
range_type
= create_static_range_type (NULL,
- temp_type,
- 1,
- index->range.high - index->range.low + 1);
+ temp_type,
+ index->range.low,
+ index->range.low + new_length);
interim_array_type = create_array_type (NULL,
temp_type,
diff --git a/gdb/valops.c b/gdb/valops.c
index f8d23fb..6c9112f 100644
--- a/gdb/valops.c
+++ b/gdb/valops.c
@@ -3759,10 +3759,13 @@ value_of_this_silent (const struct language_defn *lang)
struct value *
value_slice (struct value *array, int lowbound, int length)
{
- /* Pass unaltered arguments to VALUE_SLICE_1, plus a CALL_COUNT of '1' as we
- are only considering the highest dimension, or we are working on a one
- dimensional array. So we call VALUE_SLICE_1 exactly once. */
- return value_slice_1 (array, lowbound, length, 1);
+ /* Pass unaltered arguments to VALUE_SLICE_1, plus a default stride
+ value of '1', which returns every element between LOWBOUND and
+ (LOWBOUND + LENGTH). We also provide a default CALL_COUNT of '1'
+ as we are only considering the highest dimension, or we are
+ working on a one dimensional array. So we call VALUE_SLICE_1
+ exactly once. */
+ return value_slice_1 (array, lowbound, length, 1, 1);
}
/* CALL_COUNT is used to determine if we are calling the function once, e.g.
@@ -3776,7 +3779,8 @@ value_slice (struct value *array, int lowbound, int length)
ranges in the calling function. */
struct value *
-value_slice_1 (struct value *array, int lowbound, int length, int call_count)
+value_slice_1 (struct value *array, int lowbound, int length,
+ int stride_length, int call_count)
{
struct type *slice_range_type, *slice_type, *range_type;
struct type *array_type = check_typedef (value_type (array));
@@ -3799,14 +3803,24 @@ value_slice_1 (struct value *array, int lowbound, int length, int call_count)
attributes of the underlying type. */
if (call_count > 1)
{
+ ary_low_bound = TYPE_LOW_BOUND (TYPE_INDEX_TYPE (elt_type));
+ ary_high_bound = TYPE_HIGH_BOUND (TYPE_INDEX_TYPE (elt_type));
elt_type = check_typedef (TYPE_TARGET_TYPE (elt_type));
row_count = TYPE_LENGTH (array_type)
/ TYPE_LENGTH (TYPE_TARGET_TYPE (array_type));
}
- elem_count = length;
+ /* With a stride of '1', the number of elements per result row is equal to
+ the LENGTH of the subarray. With non-default stride values, we skip
+ elements, but have to add the start element to the total number of
+ elements per row. */
+ if (stride_length == 1)
+ elem_count = length;
+ else
+ elem_count = ((length - 1) / stride_length) + 1;
+
elt_size = TYPE_LENGTH (elt_type);
- elt_offs = longest_to_int (lowbound - ary_low_bound);
+ elt_offs = lowbound - ary_low_bound;
elt_stride = TYPE_LENGTH (TYPE_INDEX_TYPE (array_type));
elt_offs *= elt_size;
@@ -3837,7 +3851,7 @@ value_slice_1 (struct value *array, int lowbound, int length, int call_count)
else
{
range_type = TYPE_INDEX_TYPE (TYPE_TARGET_TYPE (array_type));
- slice_range_size = (ary_low_bound + row_count - 1) * (elem_count);
+ slice_range_size = ary_low_bound + (row_count * elem_count) - 1;
ary_low_bound = TYPE_LOW_BOUND (range_type);
}
@@ -3849,8 +3863,9 @@ value_slice_1 (struct value *array, int lowbound, int length, int call_count)
{
struct type *element_type;
- /* When CALL_COUNT equals 1 we can use the legacy code for subarrays. */
- if (call_count == 1)
+ /* When both CALL_COUNT and STRIDE_LENGTH equal 1, we can use the legacy
+ code for subarrays. */
+ if (call_count == 1 && stride_length == 1)
{
element_type = TYPE_TARGET_TYPE (array_type);
@@ -3871,29 +3886,53 @@ value_slice_1 (struct value *array, int lowbound, int length, int call_count)
}
}
- /* When CALL_COUNT is larger than 1 we are working on a range of ranges.
- So we copy the relevant elements into the new array we return. */
+ /* With a CALL_COUNT or STRIDE_LENGTH are greater than 1 we are working
+ on a range of ranges. So we copy the relevant elements into the
+ new array we return. */
else
{
+ int j, offs_store = elt_offs;
LONGEST dst_offset = 0;
LONGEST src_row_length = TYPE_LENGTH (TYPE_TARGET_TYPE (array_type));
- element_type = TYPE_TARGET_TYPE (TYPE_TARGET_TYPE (array_type));
+ if (call_count == 1)
+ {
+ /* When CALL_COUNT is equal to 1 we are working on the current range
+ and use these elements directly. */
+ element_type = TYPE_TARGET_TYPE (array_type);
+ }
+ else
+ {
+ /* Working on an array of arrays, the type of the elements is the type
+ of the subarrays' type. */
+ element_type = TYPE_TARGET_TYPE (TYPE_TARGET_TYPE (array_type));
+ }
+
slice_type = create_array_type (NULL, element_type, slice_range_type);
- TYPE_CODE (slice_type) = TYPE_CODE (TYPE_TARGET_TYPE (array_type));
+ /* If we have a one dimensional array, we copy its TYPE_CODE. For a
+ multi dimensional array we copy the embedded type's TYPE_CODE. */
+ if (call_count == 1)
+ TYPE_CODE (slice_type) = TYPE_CODE (array_type);
+ else
+ TYPE_CODE (slice_type) = TYPE_CODE (TYPE_TARGET_TYPE (array_type));
v = allocate_value (slice_type);
- for (i = 0; i < longest_to_int (row_count); i++)
+
+ /* Iterate through the rows of the outer array and set the new offset
+ for each row. */
+ for (i = 0; i < row_count; i++)
{
- /* Fetches the contents of ARRAY and copies them into V. */
- value_contents_copy (v,
- dst_offset,
- array,
- elt_offs,
- elt_size * elem_count);
- elt_offs += src_row_length;
- dst_offset += elt_size * elem_count;
+ elt_offs = offs_store + i * src_row_length;
+
+ /* Iterate through the elements in each row to copy only those. */
+ for (j = 1; j <= elem_count; j++)
+ {
+ /* Fetches the contents of ARRAY and copies them into V. */
+ value_contents_copy (v, dst_offset, array, elt_offs, elt_size);
+ elt_offs += elt_size * stride_length;
+ dst_offset += elt_size;
+ }
}
}
diff --git a/gdb/value.h b/gdb/value.h
index 05939c4..d687468 100644
--- a/gdb/value.h
+++ b/gdb/value.h
@@ -1056,7 +1056,7 @@ extern struct value *varying_to_slice (struct value *);
extern struct value *value_slice (struct value *, int, int);
-extern struct value *value_slice_1 (struct value *, int, int, int);
+extern struct value *value_slice_1 (struct value *, int, int, int, int);
extern struct value *value_literal_complex (struct value *, struct value *,
struct type *);
--
1.7.0.7