Commit Graph

2045 Commits

Author SHA1 Message Date
Cary R 10ef0eb109 Fix class versus struct warning in clang 2017-06-19 21:26:16 -07:00
Martin Whitaker e2be64558b Ensure value passed to named event via vpi_put_value() is initialised.
The __vpiNamedEvent::vpi_put_value() function ignores the value, but
intermediate code looks at it.
2017-06-15 09:11:11 +01:00
Martin Whitaker 3de7c234f7 Implement vpi_put_value() for named events (GitHub issue #158). 2017-06-15 08:47:29 +01:00
Johann Klammer c79df7c44b fix failing VPI callbacks the vpip_cur_task was left dangling 2017-01-01 17:44:54 -08:00
Udi Finkelstein f48b0e3434 Allows compiling for Windows (32 or 64 bits) under Linux. 2016-12-11 03:40:35 +02:00
Udi Finkelstein 2127680e49 Fixed cross-compile using mingw 64 bit under Windows 10 Linux subsystem (Ubunto 14.04)
Should work with true Ubuntu 14.04 as-is.
2016-12-08 00:11:22 +02:00
Martin Whitaker 0fdf29e099 Fix for GitHub issue #117 - delays in VPI simulation time callbacks.
Change cbReadWriteSynch and cbReadOnlySynch to interpret the passed
time value as a relative delay, not an absolute time. This matches
the behaviour of other simulators.
2016-09-25 20:43:58 +01:00
Stephen Williams e56e52bc42 Merge branch 'master' of github.com:steveicarus/iverilog 2016-09-19 12:59:41 -07:00
Maciej Suminski f32b64f497 Fixed warnings about shifting a negative value 2016-09-19 12:54:15 -07:00
Cary R 5853e32f82 Fix some cppcheck warning issues 2016-09-13 23:45:25 -07:00
Stephen Williams d44c814bab Remove .alias records from vvp generated code.
Net arrays can be handled by nets directly, instead of creating
.alias records.
2016-08-31 14:05:09 -07:00
Cary R a2fbdeff78 Add some pass by reference to vvp 2016-08-14 22:26:16 -07:00
Cary R ad5a7f7ca3 Update cppcheck vvp suppression file 2016-08-14 21:18:55 -07:00
Cary R d0e0776910 Make the base class match the derived classes regarding passing by reference 2016-08-14 21:18:44 -07:00
Martin Whitaker 7bed181f68 Support timescales in design units that aren't inside a module.
SystemVerilog allows tasks, functions, and classes to be defined at the
root level or inside packages, so we can't rely on an enclosing module
being present to provide the timescale.
2016-07-22 22:48:20 +01:00
Yury Gribov 43cd693fe0 Put start events to proper queue. 2016-07-17 09:47:08 -07:00
Michele Castellana 9ad46044b4 Cleanup 2016-06-30 17:47:34 +02:00
Martin Whitaker 8234f1845d Fix for GitHub issue #106 - prevent VPI writes during read-only synch. 2016-06-09 21:27:20 +01:00
Martin Whitaker 61c82d2cb0 Handle mixed continuous and non-blocking assignments to same vector.
SystemVerilog allows a mixture of procedural and continuous assignments
to be applied to different parts of the same vector. The previous attempt
to make this work for non-blocking assignments was flawed (see preceding
fix for vvp_fun_part_pv::recv_vec4_pv). Instead, handle this case by
converting the non-blocking assignment into a delayed force statement,
which matches the way mixed continuous and blocking assignments are
handled.
2016-05-10 22:16:40 +01:00
Martin Whitaker 0c66116f51 Fix for GitHub issue #103 - vvp assertion failure on part select propagation.
The implementation of vvp_fun_part_pv::recv_vec4_pv was incorrect, and
propagated the incoming widths rather than the stored widths.
2016-05-10 21:59:21 +01:00
Martin Whitaker 6e5ed73b09 Extended fix for GitHub issue #99.
Most vvp functors need to support recv_vec4_pv. Any that are strength-aware
also need to support recv_vec8_pv. Note the simplifying assumption that is
documented in the base class recv_vec4_pv_ implementation.
2016-04-18 23:47:43 +01:00
Martin Whitaker b2f7d09f0d Fix GitHub issue #99 - recv_vec4_pv not implemented for arithmetic functors.
Also initialise the stored operand values to 'bz instead of 'bx to get
the correct results when bits are not driven.
2016-04-04 22:29:54 +01:00
Larry Doolittle 4521be4510 More spelling, comments only 2016-04-01 11:02:34 -07:00
Larry Doolittle 89edf62206 Capitalize Verilog in a few places 2016-04-01 11:02:24 -07:00
Martin Whitaker 0dcb46e4c4 Fix shadow warning. 2016-03-19 14:42:34 +00:00
Martin Whitaker 54feb89bf5 For SystemVerilog, run variable initialization before main simulation starts. 2016-03-19 13:46:09 +00:00
Stephen Williams 78a49eacef Merge branch 'master' of github.com:steveicarus/iverilog 2016-03-14 12:26:12 -07:00
Larry Doolittle 1784488096 Spelling and whitespace 2016-03-14 12:25:50 -07:00
Martin Whitaker dc189fec8c Added new vvp latch statement to README file. 2016-03-12 12:14:03 +00:00
Martin Whitaker 99afea6946 Added support for LPM latches to vvp. 2016-03-12 09:04:51 +00:00
Maciej Suminski 9cf1dcbae2 vvp: Corrected vvp_net_fun_t::recv_vec4_pv() (commit de968e18) 2016-03-07 09:54:28 +01:00
Maciej Suminski 69b7c000d0 vvp: Implemented vvp_net_fun_t::recv_vec4_pv(). 2016-03-07 09:54:28 +01:00
Stephen Williams 9a7f31c728 Functions that return strings pass the return value on the stack. 2016-03-01 15:38:28 -08:00
Cary R 7d21891147 Fix space issues. 2016-02-28 22:09:39 -08:00
Martin Whitaker 46105e0c5a Delayed output of vvp dff to end of current time slot.
Flip-flops are generally modelled in behavioural code using non-blocking
assignments. This change makes the synthesised code behave the same as
the behavioural code. It's a more realistic model of a real flip-flop
too, which will always have some clock-to-output delay.
2016-02-21 00:00:39 +00:00
Cary R 3e52b2be66 Fix valgrind compile 2016-02-03 22:25:57 -08:00
Cary R 841593d013 Remove some compile warnings 2016-02-03 21:40:18 -08:00
Stephen Williams 76ced1cf7c Fix %ret/* and %retload/* opcodes to search up for function thread. 2016-02-01 14:51:18 -08:00
Stephen Williams f494b478cc Merge branch 'master' into return-stack
# Conflicts:
#	sv_vpi_user.h
2016-02-01 14:47:44 -08:00
Stephen Williams ce692f90ad Implement and use the %retload/real opcode. 2016-02-01 09:31:06 -08:00
Stephen Williams c114edfa6c Handle void functions with new .scope format. 2016-02-01 09:29:49 -08:00
Stephen Williams 604a62379c Make .scope aware of return type, and extend %ret/vec4 operands
The .scope needs to be aware of return types so that the %call/vec4
function knows how to intialize the return value. We also need to
extend the %ret/vec4 to support writing parts of the return value.
2016-01-31 15:29:52 -08:00
Stephen Williams e435a879fc Add ability to read back return value / Add vec4 stacked user functions.
Also:
- handle functions as arguments to system tasks.
- Cleanup detect of signal as return value.
2016-01-24 18:36:26 -08:00
Stephen Williams 0c91a6b041 Add .ufunc/real functor to handle functions with return value on stack. 2016-01-17 16:23:28 -08:00
Stephen Williams bfc9cd8715 Make real functions in processes use parent stack for return value. 2016-01-10 17:09:33 -08:00
Maciej Suminski 517c9785e8 vvp: Code cleaning (vpi_tasks). 2016-01-05 14:23:25 +01:00
Maciej Suminski df6b24fd3a ivl & vvp: Enabled 'string' as the return type in VPI functions. 2016-01-05 14:23:25 +01:00
Cary R e44010c2a3 Remove redundant class keyword 2015-12-29 13:02:58 -08:00
Cary R 71ff9e70ea Fix valgrind release code to work with new scope definition 2015-12-29 12:46:42 -08:00
Cary R 8367285f3b Fix some struct versus class warnings for vpiScope 2015-12-29 12:29:10 -08:00
Stephen Williams 63ad15ee2e Rework user function calls to use specialized opcodes.
Create The %callf/* opcodes to invoke user defined functions in a
more specialized way. This allows for some sanity checking on the
way, and also is a step towards keeping return values on stacks.
2015-12-27 20:29:10 -08:00
Stephen Williams 1b33bf8615 Normalize the parsing of %fork and %disable opcodes. 2015-12-27 13:07:42 -08:00
Stephen Williams f275dac7a8 Kill a bunch of __vpiScope struct vs class warnings. 2015-12-27 11:37:10 -08:00
Stephen Williams 2fedb4942e C++-ify more __vpiScope members. 2015-12-23 17:38:13 -08:00
Stephen Williams fff69390ac C++-ify the __vpiScope classes. 2015-12-20 20:26:57 -08:00
Stephen Williams 3e3f245d8c Document the vvp .scope record.
Also clean up some dead code in the vvp parser.
2015-12-20 16:31:03 -08:00
Cary R e85d95a659 Cleanup any allocated event queue data 2015-12-14 21:08:17 -08:00
Cary R aacb4df1dd Fix memory leak found with valgrind 2015-12-13 14:55:05 -08:00
Christian Taedcke 6d5aabd4f0 Make a few constructors explicit.
This removes cppcheck warnings.
2015-10-22 12:33:33 +02:00
Cary R 97fca3fc4b Fix the modpath delay to ignore bits that are not changing 2015-10-21 22:01:24 -07:00
Cary R a65c007fdb Fix some cppcheck warnings 2015-10-02 09:43:54 -07:00
Stephen Williams 466b6464bc Some minor vvp runtime performance tweaks. 2015-10-01 12:43:56 -07:00
Stephen Williams 4338d43cea Opcodes documentation fixes. 2015-09-30 16:25:35 -07:00
Stephen Williams fe7156289b Remove some useless vvp opcodes. 2015-09-30 15:41:22 -07:00
Stephen Williams df7b1879dd Rework vvp file version checking. 2015-08-17 15:37:28 -07:00
Martin Whitaker 9f4b485c1c Updated copyright dates displayed for main programs and targets. 2015-08-17 22:05:08 +01:00
Larry Doolittle 4dd65926ae Spelling fixes 2015-08-17 11:35:10 -07:00
Martin Whitaker 94595fb5f2 Fixed version number in vvp man page. 2015-08-16 23:07:09 +01:00
Cary R 3082dceb6e Update vvp examples to have correct version number 2015-08-15 20:30:20 -07:00
Frederick C. Kurz d766e82aa1 Changes for vvp/vvp_net.h
"# include  <string>" was added so "Microsoft Visual Studio Express 2015 RC Web" could compile it without error. "static void operator delete[](void*); was preprocessed so "Microsoft Visual Studio Express 2015 RC Web" could link it without error for a function not yet implemented.
2015-07-22 18:02:52 -07:00
Frederick C. Kurz c35f2f8334 Changes to vvp/vthread.cc
"child->delay_delete = 1;" was added, for when building with "Microsoft Visual Studio Express 2015 RC Web" in DEBUG mode, so that pr2909555.v would pass with -strict, otherwise it would cause memory access error will trying to access the previously deleted "child" variable.
2015-07-22 18:02:45 -07:00
Frederick C. Kurz afd225a44d Changes to vvp/vpi_vthr_vector.cc
Changed for "Microsoft Visual Studio Express 2015 RC Web" so gold files would match.
2015-07-22 18:00:59 -07:00
Frederick C. Kurz b36a0a2c54 Changes to vvp/vpi_signal.cc
Changed the #ifdef structure so "Microsoft Visual Studio Express 2015 RC Web" could compile it without error.
2015-07-22 17:56:58 -07:00
Frederick C. Kurz a3728dcf20 Changes to vvp/vpi_mcd.cc
Changed for"Microsoft Visual Studio Express 2015 RC Web" so vpi/pr723.v would pass.
2015-07-22 17:56:20 -07:00
Frederick C. Kurz 5f906369b5 Changes to vvp/vpi_callback.cc
Changed for"Microsoft Visual Studio Express 2015 RC Web" so gold files would match.
2015-07-22 10:10:32 -07:00
Frederick C. Kurz 12f4d29df3 Changes to vvp/class_type.cc
Changed for "Microsoft Visual Studio Express 2015 RC Web" so sv_class14.v would pass without runtime memory error.
2015-07-22 10:06:19 -07:00
Martin Whitaker 97c6339241 Ensure VPI release on net connected to island returns correct value. 2015-07-01 09:00:43 +01:00
Martin Whitaker 8d2149f5aa Further fix for GitHub issue #73 - also handle part selects. 2015-06-30 23:59:00 +01:00
Martin Whitaker e8225bd39e Fix for GitHub issue #73 - allow for island ports in vpi_put_value.
In the special case that a net is attached to an island port, values
driven onto the net via the VPI must go to the functor, not the filter,
so that they propagate through the island.
2015-06-30 21:50:13 +01:00
Martin Whitaker 637fc40dd9 Fix various bugs in vpi_put_value.
When putting a value onto a wire, the value needs to be sent to the
filter, not the functor (the functor may be part of the expression
that drives the wire).

Force and release weren't implemented properly (or at all in the
case of real values). They need to behave the same as the force
and release operations in vthread.cc.
2015-06-26 08:49:41 +01:00
Martin Whitaker dad223316e Fix propagation of unforced bits when forcing a part select.
vvp_net_t::force_vec4 propagates all bits of the forced value passed
to it, regardless of the mask value. I can't see any way to fix this
directly, so instead make sure anything that calls force_vec4 sets
the unforced bits of the passed value to the correct value.
2015-06-25 00:13:57 +01:00
Larry Doolittle b23faff27c Just a few more spelling fixes
Includes some user-visible messages
2015-06-17 08:09:34 -07:00
Martin Whitaker b242663cae Support negedge flip-flops in synthesis and in vvp.
Also extend the support for FF asynchronous set values to vvp and
fix the dff functor in vvp to correctly model asynchronous set/clr
behaviour.
2015-06-13 16:47:57 +01:00
Larry Doolittle 2739f83702 Spelling fixes in C and C++ comments 2015-06-04 15:00:29 -07:00
Stephen Williams 05d591ccd6 Fix broken write to log files. 2015-06-03 14:32:04 -07:00
Stephen Williams e6c6f6c81e Use fwrite to write $display output, instead of fprintf
This change makes it safe to write non-ascii characters, which
is an issue when the %u format is used.
2015-06-03 10:00:35 -07:00
Larry Doolittle 33c651aa00 Spelling fixes in .txt files 2015-05-25 12:52:03 -07:00
Martin Whitaker c75498b1d4 Add -i (interactive) option to vvp.
This forces <stdout> to be unbuffered. This is useful when using the
mintty terminal emulator in Windows (as used by MSYS2 and CygWin),
which identifies as a pipe, not a tty.
2015-05-16 00:39:03 +01:00
Martin Whitaker b6304129fa Fix for GutHub issue #64 - vvp interrupt/continue in Windows.
Windows implements the original UNIX semantics for signal(), which
means you have to re-establish the signal handler each time a signal
is caught.
2015-05-15 22:30:03 +01:00
Martin Whitaker ac4f4cf1a0 Update man pages.
Removed obsolete -m32 and -ivl options from iverilog-vpi man page
and revised description of -mingw option. Also removed duplicate
descriptions of --cflags, --ldflags, and --ldlibs options. Updated
link to main iverilog web page in all man pages.
2015-05-10 13:19:51 +01:00
Martin Whitaker cfbc90812b Enable use of MinGW ANSI stdio routines.
Defining __USE_MINGW_ANSI_STDIO=1 provides C99 compatible printf
and scanf routines, which avoids the need for workarounds for the
various failings of the Microsoft C runtime library.
2015-05-10 11:45:42 +01:00
Martin Whitaker 6a3edc63d5 Fix printf format for size_t values when using 64-bit MinGW.
The Microsoft C runtime does not support the %zu and %zd formats.
Previously these were replaced with %u and %d, but for 64-bit we
need to use %llu and %lld.
2015-05-08 20:20:14 +01:00
Martin Whitaker 45dc13e496 Use uintptr_t/intptr_t when casting between pointer and integer.
The code was using (unsigned) long, but a long is 32-bits in the
Windows 64-bit ABI.
2015-05-05 23:00:09 +01:00
Larry Doolittle 5dc1396eea Spelling fixes 2015-04-13 11:35:12 -07:00
Martin Whitaker e006f9e132 Fix vvp crash when a part select of a wire is passed to $monitor.
When the PV expression is compiled, the parent net may not yet be
resolved, so we may not be able to get its VPI handle straight away.
2015-03-17 21:14:30 +00:00
Cary R feb710a186 Add support for any width 2-state darray objects 2015-02-27 11:43:25 -08:00
Cary R 62abb199d7 Add support for 4-state dynamic arrays 2015-02-26 17:56:49 -08:00
Cary R ae629f8c41 Add ability to put a string value to a SV string (from plusargs) 2015-02-25 19:20:50 -08:00
Cary R 47688d234b Assert if an unsupported dynamic array type is given. 2015-01-16 18:54:43 -08:00
Cary R 09439c55ab Add new object to the vpi_get_str(vpiType, handle) code. 2015-01-14 19:08:00 -08:00
Maciej Suminski 961d6a8f3b vpi: $left and $right functions for arrays and strings. 2015-01-12 11:08:31 +01:00
Cary R 4a41a53b42 For valgrind clean up the class definitions at the end. 2015-01-10 16:18:59 -08:00
Martin Whitaker a1630e1d30 Fix for br968.
__vpiVThrVec4Stack::vpi_get_value_int_ was always treating the thread variable
as unsigned, rather than observing the value of __vpiVThrVec4Stack::signed_flag_.
Not sure why this was done - none of the regression tests broke when I changed
this.
2015-01-10 17:24:50 +00:00
Cary R e61ed48914 Clean up vals_words when a dynamic array is accessed using the vpi 2015-01-09 17:38:39 -08:00
Cary R d29a6d134f Fix valgrind cleanup of dynamic and queue arrays 2015-01-09 15:05:48 -08:00
Cary R 6310b47538 Fix a memory leak when searching for an element by name 2015-01-08 16:14:20 -08:00
Stephen Williams 96d181efed Support vpiStrengthVal for vec4 stack objects. 2014-12-23 14:42:45 -08:00
Cary R 3ccc59eaa3 Fix compile with valgrind hooks after vec4-stack changes 2014-12-12 14:28:07 -08:00
Martin Whitaker 0282b8450c Fix potential memory leak when a thread is disabled.
A disable statement can terminate a thread whilst it still has
local variables on the stack (e.g. the loop counter for a repeat
statement). We need to clear the thread stacks when this happens.
2014-12-11 20:10:17 +00:00
Cary R 48b0fed29e Use uint64_t casting of constants since UL does not work on 32-bit machines
Using a UL constant in a unit64_t context does not work on a 32-bit
machine since UL is 32-bits. Instead create uint64_t constants using
static casts and the appropriate bit operators.
2014-12-10 14:41:18 -08:00
Cary R edf112b900 Update all the vvp examples to work correctly 2014-12-08 20:55:19 -08:00
Cary R c28188618b File example vvp code so make check passes. 2014-12-08 10:43:46 -08:00
Martin Whitaker 0d7daf5862 Fix vvp memory leak for user function calls in a continuous assignment. 2014-12-07 13:47:50 +00:00
Stephen Williams 105521b547 Merge branch 'master' into vec4-stack 2014-12-06 08:24:46 -08:00
Cary R fd3086f0a9 Fix compile warnings on RHEL5 2014-12-05 19:03:07 -08:00
Cary R d62a307c34 Fix 32-bit issue in vector4_to_value() 2014-12-05 18:00:08 -08:00
Stephen Williams 409f8c5823 Add the vec4 %subi instruction 2014-12-05 09:45:29 -08:00
Stephen Williams 03198356a5 vec4-stack stack manipulation improvements. 2014-12-04 17:15:27 -08:00
Stephen Williams b1d2393789 Optimize the %sub instruction by integrating it with vvp_vector4_t class 2014-12-04 17:01:16 -08:00
Stephen Williams 0f740289e9 Optimize %mul instructions by integrating with vvp_vector4_t class. 2014-12-04 16:00:57 -08:00
Stephen Williams 46ce236cfb Optimize the %add and %addi instructions
Tightly integrate with the vvp_vector4_t class to get much
better add performance.
2014-12-04 12:38:08 -08:00
Stephen Williams 86139c855d Optimize the vec4-stack %cmp/s and %cmpi/s instructions.
Magnitude compare is called a LOT, so it is worth putting some
special effort into it.
2014-12-04 10:42:48 -08:00
Stephen Williams eb070b061b Optimize the vec4_to_index, which implements %ix/vec4 instructions. 2014-12-03 17:53:45 -08:00
Stephen Williams 2b1393e7b9 Reduce some vector copies in %load/vec4 and %concat/vec4 instructions.
By clever stack manipulations, we can eliminate some vector copies,
which can improve performance.
2014-12-03 13:12:06 -08:00
Stephen Williams 85c7b07a9b Implement %cmp/ne and %cmpi/ne
These pull in the inversion of the output flags so that they more
efficiently implement != and !==, without %flag_inv instructions.
2014-12-03 11:06:11 -08:00
Stephen Williams a57e0a1e3f vvp implementation of copy from vec4 to vec2 can use subarray method. 2014-12-02 14:52:37 -08:00
Stephen Williams b83144c552 The __vpiArray::set_word method should take a reference.
Seem to have missed a case where a vvp_vector4_t can be passed by
const reference instead of by value.
2014-12-02 13:02:46 -08:00
Stephen Williams 58fb80aec4 Implement and put to use the %muli instruction. 2014-12-02 12:46:17 -08:00
Stephen Williams 38f277d81b Merge branch 'master' into vec4-stack
Conflicts:
	vvp/array.cc
	vvp/vthread.cc
2014-12-02 11:21:58 -08:00
Maciej Suminski 1834d470cb vvp: Refactored __vpiArrayVthrA & __vpiArrayVthrAPV. 2014-11-28 14:14:42 +01:00
Maciej Suminski 9bc463aac0 vvp: Moved array_word_change(), array_attach_word(), array_alias_word() to __vpiArray. 2014-11-28 14:14:42 +01:00
Maciej Suminski 7c77097f71 vvp: array_[set/get]_word changed to __vpiArray::[set/get]_word. 2014-11-28 14:14:42 +01:00
Maciej Suminski cc4c5f4998 vvp: Changed get_array_word_size(vvp_array_t) to __vpiArray::get_word_size(). 2014-11-28 14:14:42 +01:00
Maciej Suminski 230c435ae7 vvp: Added missing functions for __vpiDarray. 2014-11-28 14:14:42 +01:00
Maciej Suminski 4ddef32631 vvp: __vpiArray::get_word_value handles StrVal variants. 2014-11-28 14:14:42 +01:00
Maciej Suminski a236c274f3 vvp: Minor changes. 2014-11-28 14:14:42 +01:00
Maciej Suminski 21a8cb71ee vvp: Handles for dynamic array ranges. 2014-11-28 14:14:42 +01:00
Maciej Suminski b4ccaa1b0c vvp: Fixed iterators for dynamic arrays. 2014-11-28 14:14:42 +01:00
Maciej Suminski ca2ef5c956 vvp: Cleaning. 2014-11-28 14:14:39 +01:00
Maciej Suminski 6015aceda2 vvp: vpiArrayIterator::vpi_index core moved to vpiArrayBase::get_iter_index. 2014-11-28 14:14:37 +01:00
Maciej Suminski 9538501b54 vvp: Moved __vpiArrayBase::vpi_iterate to __vpiArayBase::vpi_array_base_iterate. 2014-11-28 14:12:49 +01:00
Maciej Suminski ea7fa24f9e vvp: Further extraction of common code pieces from ArrayBase. 2014-11-28 14:12:46 +01:00
Maciej Suminski 48d3701c02 vvp: get_word_index() & get_word_parent() moved to __vpiArrayWord. 2014-11-28 14:12:43 +01:00
Maciej Suminski c1164dcc33 vvp: Initial work on support for dynamic arrays in VPI.
Added array_common.[ch] to store shared code.
2014-11-28 14:12:37 +01:00
Maciej Suminski 1166cd932a vpi: Handle new formats in vpi_get_value() for unpacked arrays.
It is possible to use vpiIntVal, vpiVectorVal, vpiRealVal,
vpiStringVal and vpiObjTypeVal to retrieve data from unpacked
array cells.
2014-11-27 17:36:23 +01:00
Stephen Williams 679021a1b5 Optimize special cases of immediate value extraction. 2014-11-22 10:38:16 -08:00
Stephen Williams f3392561ed Optimize the vvp pad implementation. 2014-11-22 09:59:02 -08:00
Stephen Williams 7a9a022744 Remove some dead code. 2014-11-21 17:36:22 -08:00
Stephen Williams c222b3b6a4 Minor improvements to the %split/vec4 instruction. 2014-11-21 16:51:59 -08:00
Stephen Williams 663c79d4af Add the %cmp/e instructions, and put them to use.
When testing for == and ===, there is no need to also calculate <,
so it makes sense to have a special instruction for these cases.
2014-11-21 16:45:27 -08:00
Stephen Williams bea03db25d Minor vec4 stack manipulation optimizations. 2014-11-21 14:41:31 -08:00
Stephen Williams b96f04ccce Implement the %parti/X instructions
This allows part select with constant base to be handled optimally.
Also update some more instructions to more optimally work with
the vec4 stack.
2014-11-20 18:43:24 -08:00
Stephen Williams a9db765f98 Generate better code for condition expressions. 2014-11-20 14:47:44 -08:00
Stephen Williams 04bdfbccee Add %cmpi/s and %cmpi/u instructions for performance
These bypass the vec4 stack in some common cases, saving instructions
and vec4 manipulations.

Also, minor improvement to the %flag/set/vec4 statement.

Kill a few warnings.
2014-11-19 16:38:43 -08:00
Stephen Williams 2acc9fbdee Remove dead instructsion %set/qb and %set/qf / Kill some warnings. 2014-11-19 09:15:01 -08:00
Stephen Williams 725ed869ba Remove dead %load/vp0 instructions and related infrastructure
Also remove some &A<> and &PV<> symbols that use this now dead
infrastructure.
2014-11-19 09:02:21 -08:00
Stephen Williams 301edf69d3 Add and use %concati/vec4 and %addi instructions.
Also, clean up some warnings, and optimize some existing opcodes.
2014-11-18 12:27:55 -08:00
Stephen Williams 1612c6d638 Instruction %concat/vec4, %pad/u and %shiftl manipulate stack in place 2014-11-17 16:32:18 -08:00
Stephen Williams aadd67cd3b Some instructions can do stack manipultations in place.
By doing some stack manipulations in place, certain instructions
can eliminate, or optimize, vector copies.
2014-11-14 18:38:15 -08:00
Stephen Williams 0601b4e43b Optimizations for %pushi/vec4 and %load/vec4
- Have %pushi/vec4 handle some special cases optimally.
- Eliminate some duplicated method calls in %load/vec4.
- Optimize the vvp_vector4_t::copy_from_ method by inlining
some parts.
2014-11-14 14:41:04 -08:00
Stephen Williams 8aca66b109 Performance tweaks
Better performance for %cmp/s and the vvp_fun_part_sa node.
2014-11-14 11:48:36 -08:00
Martin Whitaker c2e53f0d63 Fix for br963 - vvp qpop instructions need to call thr_check_addr. 2014-11-02 15:36:52 +00:00
Stephen Williams 8ac06fb63f Support vpiObjTypeVal for vec4Stack values.
Also allow vpi_get_value outside of scope. (Return 'x)
2014-10-25 20:38:39 -07:00
Stephen Williams 702a8c3a05 Merge branch 'master' into vec4-stack 2014-10-24 15:17:31 -07:00
Stephen Williams 48d3972299 Remove a bunch of dead vvp opcodes. 2014-10-24 15:13:28 -07:00
Stephen Williams 09d3a5de59 Port %pushv/str to vec4-stack style. 2014-10-24 10:16:35 -07:00
Stephen Williams 97bde47c31 Remove dead %set/dar and %set/dar/obj instructions. 2014-10-24 10:03:22 -07:00
Stephen Williams 4588f7c615 vec4 stack versions of queue pop methods. 2014-10-24 09:03:20 -07:00
Cary R 006aef93d0 Fix a reference to an undefined element. 2014-10-23 16:56:13 -07:00
Stephen Williams 35f5d51028 Support testing nul for property that is an object. 2014-10-22 12:58:50 -07:00
Stephen Williams 87c019c65c Merge branch 'master' into vec4-stack 2014-10-21 12:19:40 -07:00
Stephen Williams 9964d67e5e Implicit crop of vectors assigned to vector properties.
The %store/prop/v takes a width so that it is guaranteed that
all assignments to a vector property will get the size right.
2014-10-21 11:46:59 -07:00
Stephen Williams 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
Cary R a63ca15735 Add support for putting a single delay from the VPI 2014-10-18 18:27:39 -07:00
Cary R af85d44d9f Add support for putting three and six delays from the VPI 2014-10-17 20:06:01 -07:00
Cary R c5e0507941 Fix the modpath edge and vpi_put_delay() code
The code to get the correct modpath delay for a given edge had the X and Z
entries swapped.

When putting a delay from the VPI the 2 delay to twelve delay mapping was
incorrect and the to/from X delays were also not being calculated correctly.
2014-10-17 19:42:55 -07:00
Cary R f36bebf0e1 Remove some compile warnings. 2014-10-13 16:31:53 -07:00
Maciej Suminski 1a4edcac48 vvp: Accepts continuous assignment for 2-state nets. 2014-10-10 16:16:46 +02:00
Stephen Williams 98799ff7fa Allow class properties to be arrayed.
This adds the runtime support for class properties that are classes
to be arrayed. Add a means to define the dimensions of a property
in the vvp format, and add functions for setting/extracting elements
of a property.
2014-09-15 17:37:30 -07:00
Cary R 1cfbd2db63 Fix some cppcheck warnings. 2014-09-10 18:54:01 -07:00
Stephen Williams facc982af4 Ability to compare arbitrary class-valued expression to nil. 2014-09-07 16:22:49 -07:00
Stephen Williams 0dc28c0fc3 Fix the %qpop/f instruction detection. 2014-08-30 10:18:57 -07:00
Stephen Williams 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
Stephen Williams 8b969202c0 Clean up some bugs introduced by recent patches. 2014-08-21 20:21:23 -07:00
Stephen Williams 335db49282 Add vvp implementations for pop_back and pop_front methods. 2014-08-21 16:44:46 -07:00
Stephen Williams 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
Stephen Williams 9b4681918b Array access to queue variables. 2014-08-21 16:44:45 -07:00
Stephen Williams 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
Stephen Williams d891285326 vvp code generation for push_back/push_front for queue objects. 2014-08-21 16:44:45 -07:00
Stephen Williams e8b8fcba57 Support declaring queue variables all the way to vvp.
Nothing actually useful happens here, but the declarations
are functional.
2014-08-21 16:44:45 -07:00
Cary R 0a6d75f1db Fix some cppcheck warnings (format string vs argument mismatches) 2014-08-01 18:55:44 -07:00
Cary R d6b6b76015 Update header files to use a more standard name to prevent rereading
This is from github report #16. There are likely a few more issues
that need to be addressed though this takes care of the major ones.
2014-07-23 13:42:56 -07:00
Stephen Williams fa38a8e08d Merge branch 'x-sizer6' 2014-07-17 08:17:05 -07:00
Stephen Williams b991a991a7 Implement vpi_handle(vpiLeft/RightRange, ...) for signals.
This returns a constant object that can be treated like an
expression, so following the letter of the LRM.
2014-07-14 17:15:16 -07:00
Stephen Williams 515137c8c7 Imlement vpi_handle(vpiScope, 0) for NULL handles
This is not part of the standard, but it convenient, and may be
used by the libveriuser to implement some bits of PLI1.
2014-07-14 17:15:16 -07:00
Stephen Williams d5fb0f4344 Handle some tricky conditions assignments to parts.
When for example assigning to foo[<x>] within a contitional, and
doing synthesis, we need to create a NetSubstitute device to manage
the l-value bit selects.
2014-07-14 16:46:58 -07:00
Larry Doolittle 850d8c3ee9 Eliminate a few more easy warnings
Testing with 32-bit clang 3.3, with additional compiler flags
 -Wsign-compare -Wundef
this patch eliminates the following warnings:

config.h💯6: warning: 'UINT64_T_AND_ULONG_SAME' is not defined, evaluates to 0 [-Wundef]
vcd_priv2.cc:233:12: warning: duplicate 'extern' declaration specifier [-Wduplicate-decl-specifier]
parse.cc:6496:9: warning: comparison of 0 <= unsigned expression is always true [-Wtautological-compare]
parse.cc:6499:13: warning: comparison of 0 <= unsigned expression is always true [-Wtautological-compare]
parse.cc:6502:9: warning: comparison of 0 <= unsigned expression is always true [-Wtautological-compare]
parse.cc:6510:53: warning: comparison of integers of different signs: 'const unsigned int' and 'int' [-Wsign-compare]

Changing the vlltype elements from unsigned to int reconciles their
type with the native bison YYLTYPE structure.
2014-07-10 12:30:08 -07:00
Larry Doolittle e221a8ced9 Rearrange compiler warning flags
Before this patch, WARNING_FLAGS applied to both C and C++,
and WARNING_FLAGS_CXX applied to C++ only.
This patch adds a WARNING_FLAGS_CC that applies to C only.
That change should be generally useful; in particular the C
code is almost ready for -Wstrict-prototypes, which does not
apply to C++.

-Wextra (or -W) used to only apply to C++ via WARNING_FLAGS_CXX.
This patch moves it to WARNING_FLAGS, to apply to both C and C++.
Unfortunately, that triggers a ton of warnings.

For now, cover most of the new warnings up by adding
-Wno-unused -Wno-sign-compare -Wno-type-limits
to WARNING_FLAGS_CC.  In the long run, I want to change the C coding
style, and take off these disable-warning flags.  But those changes
can dribble in as separate commits; this patch is big enough already.

Actually fix a couple missing-field-initializers in libveriuser/veriusertfs.c.
2014-07-09 09:04:17 -07:00
Larry Doolittle 6ebb57195b Fuss with C function prototypes
119 formal void parameters added to keep -Wstrict-prototypes happy.
Process found one real missing prototype in vpi/vcd_priv.h:
EXTERN void vcd_names_delete(struct vcd_names_list_s*tab);
8 such warnings left, all in Tony's code
2014-07-08 13:44:11 -07:00
Cary R 0611135758 Some more cppcheck cleanup/updates 2014-06-29 20:39:40 -07:00