e71a76a1e2
Fix signed/unsigned comparison warning.
2018-10-06 20:40:54 +01:00
f1608e163f
Fix implicit fallthrough warnings when building with recent GCC.
2018-10-06 20:15:42 +01:00
87c01c55c0
Don't allow non-vectorable arguments to $signed/$unsigned.
...
This led to an assertion failure in many cases (see br1029).
1364-2012 indicates it is illegal.
2018-05-14 22:13:56 +01:00
c68a7fede3
Fix spacing
2017-11-20 07:30:52 -08:00
3fc9ad2db0
Add support for the wild compare operators ==? and !=?
2017-11-17 19:32:50 -08:00
d23b046203
Update the enumeration methods to set their width test attributes correctly
2017-11-17 13:09:14 -08:00
5ca058bfb5
Add support for darray initialisation from another darray.
...
Fixes GitHub issue #164 .
2017-10-08 17:51:33 +01:00
8a36849fda
Generate proper error messages for invalid dynamic array initialisers.
...
Invalid user code should not result in an "internal_error".
2017-10-08 12:59:30 +01:00
8f6f999ea7
Fix for GitHub issue #142 - assertion failure when parameter expression
...
contains undefined variable.
2017-01-29 10:47:54 +00:00
7cac4677bf
Fix for GitHub issue #130 part 2 : assertion failure on unsupported cast.
2016-11-25 22:11:13 +00:00
182c08b528
Fix segfault when error in multi-dimensional part select.
2016-10-27 10:59:31 -07:00
5853e32f82
Fix some cppcheck warning issues
2016-09-13 23:45:25 -07:00
c706c5dd90
The array properties can be available in a constant context
...
For most arrays the various properties are available as constant
values and can be evaluated in the compiler.
2016-09-11 14:25:00 -07:00
7f475d4210
Refactor to use new pad_to_width/cast_to_width functions.
2016-03-25 22:23:45 +00:00
dc1c3a4043
Fix for GitHub issue #94 - enhance support for SystemVerilog size casting.
...
Allow the size expression to be any constant expression. Also ensure that
the expression width and type are correctly calculated and applied.
2016-03-25 21:49:28 +00:00
0199ad129d
Fix expression type for compressed assignment statements.
...
A compressed assignment statement should give exactly the same
result as the equivalent uncompressed statement. This means
that the type (signed/unsigned) of the LHS affects the type of
the RHS expression (unlike in normal assignments). We need to
take care that bit/part selects and concatenations are correctly
identified as unsigned values, even in the cases where they
reduce to a single whole signal.
2016-02-23 16:53:01 +00:00
32fab21e95
ivl: Allow concatenating strings returned by function calls.
2016-01-07 11:11:04 +01:00
399384d81b
Fix some cppcheck warnings and bugs
2015-12-19 17:19:19 -08:00
6d5aabd4f0
Make a few constructors explicit.
...
This removes cppcheck warnings.
2015-10-22 12:33:33 +02:00
e682029648
A concatenation or the expression being replicated must have a non-zero width
2015-08-18 16:22:50 -07:00
3e2196651a
Remove unused parameter to indices_to_expressions
...
need_addr has been hanging around, unused, since commit 078a3fd4
2015-07-31 09:48:33 -07:00
44dfc41004
Detect and report excess function arguments.
...
Also enhance a couple of error messages.
2015-06-21 09:07:11 +01:00
b23faff27c
Just a few more spelling fixes
...
Includes some user-visible messages
2015-06-17 08:09:34 -07:00
2739f83702
Spelling fixes in C and C++ comments
2015-06-04 15:00:29 -07:00
ab688613cc
Fix for br978 - assertion involving addition of $ivlh_to_unsigned() result.
...
$ivlh_to_unsigned, unlike $signed and $signed, can cause a reduction
in width. The shared PECallFunction::cast_to_width_ method did not
support this.
2015-05-22 00:05:39 +01:00
301a7e587b
Fix for GitHub issue #62 - assertion failure on too many array/vector indices.
...
Provide an error message for invalid code that contains more indices than
there are dimensions.
2015-04-25 22:57:14 +01:00
22d15f9ca9
Fix handling of expression width with $ivlh_to_unsigned.
...
The first argument to $ivlh_to_unsigned should be treated as having
a self-determined width.
2015-04-25 09:20:31 +01:00
83d721232d
ivl: $ivlh_to_unsigned() also tests its argument width.
2015-03-27 20:53:14 +01:00
102d2d534f
Remove extra warning code.
2015-03-09 14:35:31 -07:00
bddfcac823
The === and !== operators cannot be used with a SV string
2015-03-02 15:34:22 -08:00
8b3667f76e
ivl: Casting vectors to dynamic arrays.
2015-02-04 16:02:38 +01:00
a52242745a
ivl: Casting dynamic arrays to vectors.
2015-02-04 16:02:38 +01:00
35401f0e2c
ivl: Functions may return dynamic arrays.
2015-01-12 11:08:31 +01:00
96472e5537
For a signed R-value we can use the full width when converting to long
...
When trying to get the value we can use the full width of a long if the
expression is signed.
2014-12-10 14:41:27 -08:00
38f277d81b
Merge branch 'master' into vec4-stack
...
Conflicts:
vvp/array.cc
vvp/vthread.cc
2014-12-02 11:21:58 -08:00
c3a318f14e
ivl: Casting vectors to strings.
2014-11-27 17:36:23 +01:00
b2deae7ba9
ivl: More meaningful errors messages. Minor code formatting.
2014-11-27 17:36:23 +01:00
675dd91403
ivl: Added sign checking & casting between integers of different size.
2014-11-27 17:36:23 +01:00
b12e00d875
ivl: String to vector casting.
2014-11-27 17:36:23 +01:00
103828577b
ivl: Fixed indentations.
2014-11-27 17:36:23 +01:00
89d1125979
ivl: Bit selection using variables in strings.
2014-11-27 17:36:23 +01:00
43c6a0bacd
ivl: Casting to int.
2014-11-27 17:36:23 +01:00
13f861a963
ivl: Added PECastType to handle type casting.
2014-11-27 17:36:23 +01:00
82ca4cf641
Add a warning if an expression pads itself to a crazy width.
2014-11-25 14:49:04 -08:00
a98f21aa65
Merge branch 'master' into vec4-stack
...
Conflicts:
elab_lval.cc
netmisc.cc
tgt-vvp/eval_object.c
tgt-vvp/vvp_process.c
vvp/codes.h
vvp/compile.cc
vvp/opcodes.txt
vvp/vpi_tasks.cc
vvp/vpi_vthr_vector.cc
vvp/vthread.cc
2014-10-21 09:12:02 -07:00
d85096c56a
Add support for implict this in class methods
2014-09-16 17:08:57 -07:00
480668fee6
Add support for classes defined in $root scope.
2014-09-15 17:37:30 -07:00
ea4b000be6
Get arrayed property expressions down to the ivl_target API.
2014-09-15 17:37:30 -07:00
697701a26e
Sorry messages for property arrays.
2014-09-07 17:48:19 -07:00
88e951418b
Handle elaboration of class properties referenced within sub-scopes.
2014-09-06 16:26:08 -07:00
d1a35d5152
Handle nil queue as a was to delete a queue.
...
Treat this like assigning null to a dynamic array. This deletes the
queue and thus has the effect of clearing it.
2014-09-02 12:19:58 -07:00
886b6cd349
Debug messages.
2014-08-30 10:18:57 -07:00
dec4cd5071
Handle arrays of class objects.
...
This goes all the way down to the vvp level, where we create support
for arrays of objects, generate the new code in the -tvvp code
generator, and elaborate the arrays in the first place.
2014-08-30 10:18:57 -07:00
b4d0c43a96
Remove some compile warnings.
2014-08-25 18:24:58 -07:00
e5f49baaac
Fix for GitHub issue #37 - bug in expression width pruning.
...
If an expression contains a division, remainder, or right shift operation,
set the expression min_width to UINT_MAX to flag that the expression width
cannot be pruned. Using UINT_MAX ensures that the min_width won't change
as we continue to elaborate the expression.
2014-08-25 20:27:22 +01:00
8c2d51142b
Elaborate pop_back/pop_front methods on dynamic arrays.
2014-08-21 16:44:46 -07:00
c9ff48bd4e
Add support for dynamic array/queue "last" index ($)
...
Internally, treat the "$" as a special expression type that takes
as an argument the signal that is being indexed. In the vvp target,
use the $last system function to implement this.
2014-08-21 16:44:45 -07:00
da63ef02d4
Implement the size method for queues.
...
This works by translating it to a $size() system function call.
The $size function is already implemented for dynamic queues and
it is easy enough to expand it for queues.
2014-08-21 16:44:45 -07:00
a730572e37
Elaborate size method of darray/queue objects.
2014-08-21 16:44:45 -07:00
6d052d4ff7
Handle queue types and expressions at pform level.
2014-08-21 16:44:45 -07:00
e82bafcb84
Fix for GitHub issue #30 : failed assertion in eval_tree.cc
...
The root cause was the constant propagation done during elaboration
of a multiply operation, which didn't check for mixed real/logic
expressions.
2014-06-20 20:52:15 +01:00
81947edaa5
A bit select is not the same as selecting part of a packed array
...
When adding the ability to select part of a multi-dimensional packed
array the sign and size information for a true bit select was broken.
2014-06-12 09:20:31 -07:00
40b36337e2
Fix some bugs with packed array dimensions.
...
The netparray_t::slice_dimensions bug was the most insidious and
caused all manner of confusion. Also fix some other packed array
and unpacked array (and mixed) indexing calculations.
2014-04-06 08:40:10 -07:00
df4889ba3d
Handle SEL_IDX_UP is some situations.
2014-04-06 08:40:09 -07:00
fc1f9ac6a5
Handle unpacked arrays as module input ports.
2014-04-06 08:40:09 -07:00
6caa41cc93
First pass at support for continuous assign of unpacked net arrays.
2014-04-06 08:40:09 -07:00
7ceb18fb37
Merge branch 'master' into vec4-stack
...
Conflicts:
netmisc.cc
tgt-vvp/eval_expr.c
vvp/vthread.cc
2014-03-01 09:02:14 -08:00
d2ff77d56c
More standard handling of unsized numbers as system task arguments.
2014-03-01 08:38:23 -08:00
d0d421ebe2
Remove compile warning
2014-02-28 17:29:55 -08:00
ed2e339dd6
Fix for GitHub issue #18 : undef propagation of const multiplies is incorrect.
...
When an expression is elaborated, the compiler converts multiplies with
one constamt zero operand into a constant zero value. This is only valid
if the other operand is not a 4-state variable.
2014-02-27 19:30:28 +00:00
5dcd2e8957
Added width cap for unsized expressions.
...
Unsized expressions can expand to extremely large widths. Usually this
is actually a mistake in the source code, but it can lead to the compiler
temporarily using extremely large amounts of memory, or in the worst
case, crashing. This adds a cap on the width of unsized expressions (by
default 65536 bits, but overridable by the user), and causes a warning
message to be output when the cap is reached.
2014-02-27 19:20:20 +00:00
a3450bf856
Fixes for GitHub issues 13 and 15.
...
The verinum arithmetic operators now observe the standard Verilog
rules for calculating the result width if all operands are sized.
If any operand is unsized, the result is lossless, as before.
They also now all observe the standard rules for handling partially
undefined operands (if any operand bit is 'x', the entire result is
'x').
I've also added the unary '-' operator, and renamed v_not() to be
the unary '~' operator. This has allowed some simplification in
other parts of the compiler.
2014-02-25 20:39:21 +00:00
ebed793dec
Merge branch 'master' into vec4-stack
...
Conflicts:
vvp/vthread.cc
Signed-off-by: Stephen Williams <steve@icarus.com>
2014-02-21 17:34:59 -08:00
66bdbb77ec
Fix for GitHub issue 9 part 2 : Efficiency of the verinum pow() function.
...
This changes the verinum pow() function to use the more efficient algorithm
used in the vvp runtime. It will still be slow if the left operand is unsized
and the right operand is large, as it will expand the result vector to avoid
overflow.
2014-02-15 22:06:31 +00:00
1805598eae
Fix some vec4 handling of putc, len, and other string methods.
2014-01-31 19:01:28 -08:00
0eca210722
Merge branch 'master' into vec4-stack
2014-01-30 17:07:54 -08:00
3e95966d70
More spelling fixes
...
Mostly comments
Changs "initilzers" in a string visible with debug_elaborate.
Includes a few British->American changes in the root directory only.
2014-01-30 16:43:17 -08:00
fdc9e8590d
Merge branch 'master' into vec4-stack
2014-01-30 15:59:51 -08:00
be17bfc0e9
Spelling fixes
...
Mostly comments.
One user-visible string ("Evalutated to ") changed in the debug_eval_tree case.
2014-01-30 15:34:20 -08:00
77a01f65d0
Remove fixable unused-parameter warnings
...
I see other warnings within vhdlpp/lexor.cc caused by a flex bug
(see http://sourceforge.net/p/flex/bugs/115/ ), which I won't try to work around.
2014-01-30 15:34:08 -08:00
1fc4093191
Fix possible overflow in calculation of canonical indices.
2014-01-15 14:26:53 -08:00
f906384423
Elaborate r-vlaue packed struct members pads as needed.
2014-01-11 19:19:15 -08:00
6ec31517a9
Implement $bits(type) to get the size of a type
...
In the process, I implemented a means to get at
previously elaborated types from the pform type
pointer.
2014-01-11 19:19:15 -08:00
8a4d769b3d
Type name expressions down to elaboration.
2014-01-11 19:19:15 -08:00
0b4056817a
Generalize struct member type
2013-12-07 12:20:28 -08:00
9544cb516d
Fix some compile warnings
2013-11-24 14:35:44 -08:00
f47af361cd
For default function arguments do not access an out of range real argument.
...
The code should only look at the real argument that exist any missing
arguments default to not being defined and hence needs to use the
default expression. Found with valgrind.
2013-11-20 16:52:17 -08:00
f16fd03dab
Elaborate PChainConstructor calls.
...
It is better to leave the handling of PChainConstructor calls to
the elaboration, instead of stripping them out early. This allows
for handling the arguments of the chain constructor in the correct
scope.
2013-11-10 18:03:19 -08:00
d192f545ac
Chain class instance constructors with arguments.
2013-11-10 18:03:19 -08:00
0157a156fb
Elaborate implicit chaining of constructors in class new.
2013-11-10 18:03:19 -08:00
be2595085f
Reorder calls to test_width() on binary operands for efficiency.
...
It is more common to find an unsized number on the right hand side
of a binary operator than on the left hand side, particularly for
comparison operations (e.g. x < 10 rather than 10 > x), so testing
the width of the right operand first is less likely to result in
the width needing to be retested. Counting the number of times an
operand width is retested when running the test suite confirms this;
before this change an operand width was retested 4869 times, after
the change an operand width was retested 99 times.
2013-10-28 22:55:10 +00:00
4625e7e2b6
Improvements to strict-expr-width mode.
...
Enable error reporting when an unsized number is used in a concatenation
operand. Allow greater pruning of expressions containing unsized numbers.
2013-10-28 22:07:09 +00:00
61fbfb562e
Merge branch 'x-mil11'
2013-10-21 08:55:13 -07:00
7e82a37fa3
Extend -gstrict-expr-width to cover unsized numbers.
...
To be strictly compliant with the standard and compatible with other
EDA tools, unsized numbers should be treated as having a fixed size
(the same size as an integer). The -gstrict-expr-width option is
extended to allow the user to enable this behaviour.
2013-10-20 23:16:11 +01:00
e60804cf41
Handle initialized darray of strings
...
This includes adding support for returning strings from functions,
adding initializing new darray with array_pattern strings, and
assigning an array_pattern of strings to a preallocated darray.
Also fix up support for initializing array with simple string
expression.
2013-10-19 15:34:15 -07:00
2464582587
Handle number literals as real-valued r-values.
2013-10-19 15:34:15 -07:00
82ebf6372c
Generate code to implement new element initializers.
2013-10-19 15:34:15 -07:00
3945b9df45
Elaborate darray new element initializer to ivl_target API
2013-10-19 15:34:15 -07:00
2030e06988
Parse/elaborate some array-patterns down to the ivl_target API.
2013-10-19 15:34:14 -07:00
cd85a42acc
Elaborate the null darray literal as a null expression.
...
This leads to the correct code being generated in the vvp
code generator.
2013-10-19 15:34:14 -07:00
Martin Whitaker
Cary R
Henry Wong
Maciej Suminski
Christian Taedcke
Larry Doolittle
Martin Whitaker
Stephen Williams