GaloisInc/cryptol
1
1
Fork 0
mirror of https://github.com/GaloisInc/cryptol.git synced 2024-12-18 05:21:57 +03:00
Code Issues Projects Releases Wiki Activity
61 Commits 67 Branches 28 Tags 95 MiB
7e876be060
Commit Graph

61 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Brian Huffman
7e876be060 Remove additions to local fork of SBV that are no longer necessary 2014-07-23 10:10:54 -07:00
Brian Huffman
6bb2cd796d Unify Value datatypes for interpreter and symbolic simulator 
Some Value constructors and destructors are now polymorphic as well.
 2014-07-22 17:35:06 -07:00
Brian Huffman
ca53f226e5 Factor out separate type "BV" for words in evaluator's Value datatype 
Instead of constructor VWord Integer Integer, we use VWord BV,
where data BV = BV !Integer !Integer.
 2014-07-22 16:26:33 -07:00
Brian Huffman
c8abc76fd1 Remove unused length tag from interpreter tuple values 2014-07-22 15:43:29 -07:00
Brian Huffman
6a99eb99dd Modify symbolic Value type to match the interpreter's Value type 
This is a step towards unifying the two modules.
 2014-07-22 15:32:26 -07:00
Brian Huffman
907e4b55a9 Fix typo in comment 2014-07-22 15:16:00 -07:00
Brian Huffman
c5a14e388d Rewrite symbolic simulator in pure functional style 
We now take advantage of new features of SBV 3.1, including
the "sBranch" mechanism for evaluating reachability of conditional
branches.
 2014-07-22 13:44:17 -07:00
Brian Huffman
c2859d579b Update output for regression test 226 2014-07-21 16:18:47 -07:00
Brian Huffman
c8b3b8c134 Update output for regression test 086. 
(cf. revision 0ee396d434)
 2014-07-21 16:13:56 -07:00
Brian Huffman
ffebd5c330 Update output for regression test number 290 2014-07-21 16:02:38 -07:00
Brian Huffman
ec0395e9d0 Update regression test number 289, which now runs successfully. 
This test works as of revision db08bfafa9.

Versions of SBV prior to 3.1 did not distinguish the boolean type from
the size-1 bitvector type; this example uses literals like "0b0" and
"0b1", so it triggered the SBV bug, which manifested as a type error from
the external SMT solver.
 2014-07-21 16:01:43 -07:00
Brian Huffman
db08bfafa9 Merge changes from SBV release version 3.1 2014-07-21 15:23:34 -07:00
Brian Huffman
11d5c884b7 prove/sat commands avoid generating SBV bitvector variables of zero width 2014-07-21 14:59:33 -07:00
Brian Huffman
010d930638 Optimize symbolic "extract" to produce literal value if result has width 0 2014-07-21 11:28:40 -07:00
Brian Huffman
733c7932c5 Fix compiler warnings 2014-07-21 10:38:21 -07:00
Brian Huffman
ca14cff93a Pretty-print hex literals in parser AST using lowercase "abcdef". 
This makes the Cryptol.Parser.AST pretty printer consistent with the
one for values in Cryptol.Eval.Value.
 2014-07-15 16:43:20 -07:00
Brian Huffman
4f28f4bfed Update regression tests for :prove and :sat output 
As of ee8685e4f2, :sat and :prove now
parse and then pretty print the given predicate with appropriate
parenthesization.
 2014-07-15 16:26:37 -07:00
Brian Huffman
ee8685e4f2 REPL handles printing of :sat/prove results; parenthesize predicates as appropriate 2014-07-15 15:35:37 -07:00
Brian Huffman
f0ab651157 :sat and :prove print counterexamples in the user-configured number base 2014-07-15 14:32:14 -07:00
Brian Huffman
62cdfd453f Cryptol REPL prints a hint if the user omits "=" on a ":set" command. 
Old behavior:
Cryptol> :set base 16
base 16 = 10

New behavior:
Cryptol> :set base 16
Unknown user option: `base 16`
Did you mean: `:set base = 16`?
 2014-05-27 15:46:42 -07:00
Brian Huffman
0ee396d434 Pretty printer for Cryptol.Parser.AST.Expr parenthesizes infix operators. 
(The printer for Cryptol.TypeCheck.AST.Expr already does this.)

Previously:
Cryptol> :t (+)
+ : {a} (Arith a) => a -> a -> a

Now:
Cryptol> :t (+)
(+) : {a} (Arith a) => a -> a -> a
 2014-05-15 14:41:22 -07:00
Brian Huffman
682c130b11 Add "fin" constraint on index size for (@), (!) and (!!), like (@@) has. 
Prior to this changeset, any of the following three expressions

[1,2,3] @ (zero : [inf])
[1,2,3] ! (zero : [inf])
[1,2,3] !! [zero : [inf]]

would produce the following error:

  Location:  [Eval] fromVWord
  Message:   not a word
             [False, False, False, False, False, ...]
 2014-05-13 13:38:55 -07:00
Brian Huffman
b304eb2ef0 New module Cryptol.TypeCheck.TypeOf has functions to compute Type of an Expr 2014-05-09 16:27:07 -07:00
Brian Huffman
a6077dd59a Add more destructor functions for TypeCheck.AST.Type 2014-05-09 16:26:31 -07:00
Trevor Elliott
28043c395c Only print Loading when the module is going to be loaded 
Looks like we were re-loading everything afterall!  This closes #10
 2014-05-06 14:32:17 -07:00
Trevor Elliott
25f17d0202 Some bug fixes related to fenced code in markdown 
``` as cryptol wasn't supported, and fenced code blocks that weren't
understood were having their internals processed as markdown
 2014-05-06 14:29:51 -07:00
Trevor Elliott
f1f1991ce0 No plans to allow indented code fence blocks 
Support for the indented blocks isn't consistent, so it seems easier to just
avoid trying to support this feature.
 2014-05-06 10:59:29 -07:00
Dylan McNamee
87042d4604 fixing lone bad reference in doc, added syntax chapter, replaced Salsa spec 
PDF with pointer to it, fixed table in section 1.2.2
 2014-04-30 11:37:15 -07:00
Dylan McNamee
3188a212b5 Fixing a type constraint solver bug submitted by Jaak Randmets, 
verified by Iavor. Regression tests passed.
 2014-04-28 10:17:11 -07:00
Dylan McNamee
3e0ce32888 document that four failing tests is expected (for now) 2014-04-27 11:29:52 -07:00
Dylan McNamee
3e22966c1c spelling in README 2014-04-27 11:26:37 -07:00
Dylan McNamee
62acc96f97 Fixing makefile to be compatible with David Lazar's speck fix 2014-04-24 18:07:47 -07:00
Dylan McNamee
203b2c8596 Merge branch 'master' of https://github.com/davidlazar/cryptol into davidlazar-master 2014-04-24 18:01:46 -07:00
Adam C. Foltzer
2f229e0c9c reinsert book authors into acknowledgments 2014-04-24 14:47:55 -07:00
David Lazar
fe67268088 Simplify type constraints. 2014-04-24 17:33:43 -04:00
David Lazar
bda2596f30 Minor whitespace tweak. 2014-04-24 17:28:25 -04:00
David Lazar
e1c2ddec55 Fix formatting in speck.cry. 2014-04-24 17:27:34 -04:00
David Lazar
46c93d1a33 spec.cry should be speck.cry. 2014-04-24 17:19:20 -04:00
Dylan McNamee
bd578915ea incorporating "thanks" in main README 2014-04-24 10:59:06 -07:00
Adam C. Foltzer
8c97b35bb2 updated docs 2014-04-24 10:55:52 -07:00
Dylan McNamee
8fbb351063 final tweaks to the book 2014-04-24 10:53:00 -07:00
Adam C. Foltzer
934e0aae53 change system description parameter in makefile 2014-04-23 19:36:41 -07:00
Adam C. Foltzer
4a2da5686a temporarily check docs into the repo 
Getting the right TeX Live and pandoc environments on the build slaves
is turning out to be a Big Headache. For now, build docs locally and
keep the repo copies up to date.
 2014-04-23 19:16:57 -07:00
Adam C. Foltzer
417193b8a8 change glob for markdown docs 2014-04-23 18:54:45 -07:00
Adam C. Foltzer
72cd17c11b change name of book pdf 2014-04-23 18:52:26 -07:00
Adam C. Foltzer
af3e5ecb39 make package id configurable via env var 2014-04-23 18:52:26 -07:00
Dylan McNamee
288178dea2 markdown -> .md, Acks section, adding spec to contrib 2014-04-23 13:15:18 -07:00
Dylan McNamee
bba0c10cf7 keccak is version 1, and proving too tricky to convert for now. 
Moving simon and spec to contrib, to "seed the pool" of contributed
examples.
 2014-04-22 17:26:11 -07:00
Dylan McNamee
b7cad1c5b0 fixing issue290 - it was a bug in the brain of the reporter. 
imports need to be qualified to support this behavior.
 2014-04-22 17:03:24 -07:00
Dylan McNamee
72d690bd9f a little "how to run this" added to funstuff 2014-04-22 10:44:02 -07:00