Commit Graph

10116 Commits

Author SHA1 Message Date
Stephen Williams 6902bf44e3 Improved error handling in parser around typedefs. 2014-02-16 17:04:52 -08:00
Stephen Williams 8f849a9944 Sorry message for unpacked array module ports. 2014-02-16 15:56:32 -08:00
Stephen Williams b1ef0997ed Merge branch 'x-sizer3' 2014-02-15 15:19:55 -08:00
Martin Whitaker ecce1d25bc Fix for GitHub issue 12 : Ternary lval-rval width mismatch.
A simple typing error that caused the expression to be incorrectly
evaluated as a real constant.
2014-02-15 23:05:41 +00:00
Stephen Williams f43080fbcb Detect proper support for missing else_ without latches (synthesis) 2014-02-15 14:54:58 -08:00
Stephen Williams 463407572f Sizer pay attention to ivl_synthesis_off attribute. 2014-02-15 14:44:55 -08:00
Stephen Williams 6b6574dd8a sizer support for simple LPM MUX devices. 2014-02-15 14:44:15 -08:00
Stephen Williams 4e81939eda Handle asynchronous if-without-else synthesis. 2014-02-15 14:16:22 -08:00
Martin Whitaker 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
Martin Whitaker 5853f7d867 Fix for GitHub issue 9 part 1 : Efficiency of vvp_vector2_t::pow() function.
The vvp_vector2_t::pow() function is recursive, and performs a multiplication
operation on each step. The multiplication operator was expanding the result
vector to accomodate the maximum possible result value for the given operand
vectors, thus causing the execution time of the power operation to be
exponentially proportional to the exponent value. Both in this case and
in general, it is unnecessary for the multiplication result vector to be
expanded, as the compiler has already determined the required vector width
during elaboration, and sizes the operand vectors to match.
2014-02-15 21:40:55 +00:00
Martin Whitaker 68f8de28af Fix for GitHub issue 7 : Undef propagation in power operator.
The vvp_vector2_t constructor that takes a vvp_vector4_t value was
documented as creating a NaN value if the supplied vector contained
any X or Z bits, but instead used the standard Verilog 4-state to
2-state conversion semantics (X or Z translate to 0). I've added an
optional second parameter to the constructor to allow the user to
choose which semantics they want, as both are needed.
2014-02-15 13:14:45 +00:00
Martin Whitaker ac3aee0172 Fix for GitHub issue 8 : Signedness of constant binary bitwise operations.
When doing constant evaluation of binary bitwise operations, the result
value must be signed if the expression type is signed.
2014-02-15 11:39:05 +00:00
Martin Whitaker 8135f369a5 Fix for GitHub issue 11 : Syntax error on operator attributes.
Attributes are allowed to be attached to unary, binary, and ternary
operators in expressions. For now just accept and discard them.
2014-02-14 23:29:22 +00:00
Stephen Williams be5bfeb172 Elaborate widths of parameters correctly for unsigned values
If there is no size in the parameter type declaration or on the
value itself, then give the parameter value a minimum width of
integer bits.
2014-02-11 18:24:12 -08:00
Stephen Williams 801e795112 Better job of matching adder operand sizes for vec4 code generator. 2014-02-10 18:06:56 -08:00
Stephen Williams e66fc7e5cc Port UWIRE assignments to vec4 branch. 2014-02-10 17:19:52 -08:00
Stephen Williams 1d63875e5d Merge branch 'master' into vec4-stack
Conflicts:
	elab_lval.cc
2014-02-10 16:26:22 -08:00
Stephen Williams be1130ddbf More robust case statement synthesis
Handle the cases that the case statement carries complet sub-
statements. This is just a generalization of what is already
there.
2014-02-09 17:30:30 -08:00
Stephen Williams 9f2b7d6553 Rearrange counters to me more realistic
Instead of counting gate types, convert to gate count estimates
and some other interesting statistics. Also handling the descent
into child scopes.
2014-02-09 08:59:51 -08:00
Stephen Williams f5041e6c09 Collect some actual sizer statistics. 2014-02-08 18:53:42 -08:00
Stephen Williams 959ac3229e Start a sizer backend. 2014-02-08 10:16:11 -08:00
Stephen Williams 554fb7ebdd Various internal vec4 size mismatches fixed. 2014-02-07 17:50:13 -08:00
Stephen Williams c9e8392dc7 Fix dangling vec4 stack when force assignment is suppressed.
Also improve the %debug/thr instruction.
2014-02-07 16:10:28 -08:00
Stephen Williams 60d37e1f53 More vec4 support for various things. 2014-02-07 11:24:41 -08:00
Colin e0ac4893f0 Fixed homepage link
The link was pointing to a page that said:

This page has been moved to http://iverilog.icarus.com, you will be forwarded there automatically in 3 seconds.
Please update your links.

So now it's updated
2014-02-07 12:22:53 +00:00
Stephen Williams 401fccdf6e vec4 handling of DARRAY of vec4 vectors. 2014-02-06 15:05:26 -08:00
Stephen Williams f5564a195f vec4 versions of compressed assignment statements. 2014-02-06 09:55:25 -08:00
Stephen Williams c7e61f3aa4 Merge branch 'master' into vec4-stack 2014-02-05 15:18:25 -08:00
Stephen Williams a012406ca4 Fix broken search where member names accidentally match variables. 2014-02-03 20:04:24 -08:00
Stephen Williams 52a9fdde8a Handle packed structs as module outputs. 2014-02-03 19:22:59 -08:00
Stephen Williams 4f1c43b690 Account for force l-values that are uwires. 2014-02-02 17:05:42 -08:00
Stephen Williams 7f59c51ca2 Handle proceedural writes to UWIRE objects.
The elaborator allows this only if it determines that the bits
that are procedural written are not also continuously written.
2014-02-02 16:43:48 -08:00
Stephen Williams 0be577cc44 Allow some behavioral assignments to unresolved wires.
If the l-value is an unresolved wire, then elaboration can allow
the assignment as long as it is to bits that are not otherwise
driven. Handle this in some simple cases.
2014-02-02 11:08:43 -08:00
Stephen Williams 49cf5556a2 Fix primitive table lexical analysis.
Sequences of digits in the table of a primitive may be matched
as decimal numbers instead of digits, and this breaks some
primitives.
2014-02-02 10:57:53 -08:00
Stephen Williams 1805598eae Fix some vec4 handling of putc, len, and other string methods. 2014-01-31 19:01:28 -08:00
Stephen Williams 0eca210722 Merge branch 'master' into vec4-stack 2014-01-30 17:07:54 -08:00
Larry Doolittle 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
Stephen Williams fdc9e8590d Merge branch 'master' into vec4-stack 2014-01-30 15:59:51 -08:00
Larry Doolittle f76d56beb8 Remove lint detected by clang
With this change, iverilog builds and runs with no extra warnings when using clang 3.0
2014-01-30 15:39:55 -08:00
Larry Doolittle 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
Larry Doolittle 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
Stephen Williams e497f63e29 Implement string compares for vec4 support. 2014-01-28 17:11:21 -08:00
Stephen Williams 75233a9bca vec4 implementation of auto-increment/decrement. 2014-01-28 16:24:26 -08:00
Stephen Williams 11197baf86 The vec4 %cvt/vr requires a <wid> 2014-01-28 15:51:29 -08:00
Stephen Williams 23ba0bc019 various vec4 fixes. 2014-01-25 19:25:21 -08:00
Stephen Williams 9cfb15a302 Implement assign/vec4/off/e 2014-01-23 18:39:29 -08:00
Stephen Williams 607b63a758 Merge pull request #5 from cliffordwolf/master
Added missing support for binary ^~ in eval_tree.cc
2014-01-23 17:12:05 -08:00
Stephen Williams 8bebe59e20 Fix vec4 abs() generating duplicat labels. 2014-01-22 17:15:26 -08:00
Stephen Williams 20d415463a vec4 non-blocking assign to array swapped delay and part offset! 2014-01-21 20:58:48 -08:00
Stephen Williams 9c1dcad994 vec4 vvp support for non-blocking assign with delay and part offset. 2014-01-21 19:37:05 -08:00