summaryrefslogtreecommitdiff
path: root/src/test
AgeCommit message (Collapse)Author
2014-09-30Executing an actual elf fileKathy Gray
2014-09-29Connect up to elf model; which doesn't seem to be representing the elf file ↵Kathy Gray
properly yet but run_power.native is connected to it
2014-09-29add run_power to last commitKathy Gray
2014-09-04Refine overloading resolution with respect to vectors and rangesKathy Gray
Fix a few other bugs as well
2014-08-28fixes to bugs exposed by arm modelKathy Gray
2014-08-27Changes to get another (slightly larger) executable running;Kathy Gray
adding executable as a test as well
2014-08-26small changes for armKathy Gray
2014-08-20Add ability to track register dependencies in interactive stepper; thus ↵Kathy Gray
testing register tracking/tainting
2014-08-19make test_power_interactive working again; now using interp_interface ↵Kathy Gray
instead of internal actions
2014-08-19Add file that actually drives command line interpreterKathy Gray
2014-08-18Handling many register reads, writes, and memory reads.Kathy Gray
There are problems that warrant discussion about handling special registers that hold records or data structures previously.
2014-08-18Switch run_power to use interp_interface. Compilers and begins running, and ↵Kathy Gray
then hits an error. So still in progress Note: moved barrier kind changes to interp_interface, along with other memory kinds
2014-08-07Fix bug introduced by last commitKathy Gray
2014-08-07Track taints across machine calls.Kathy Gray
This introduced a bug in vector.sail, commented out and needs to be fixed.
2014-08-01Support separated memory read/write functions.Kathy Gray
Also allows register writing functions to be on the left hand side of an assignment in the same way. The last parameter to a writing function is the value to be written, and should appear on the right hand side of an assignment expression.
2014-07-30working dec vectorsKathy Gray
2014-07-18Writing to concatenated aliasesKathy Gray
2014-07-16Reading from an alias to two concatenated registers; not writing yet.Kathy Gray
2014-07-15Add failing test for type of aliasesGabriel Kerneis
File "test/regbits.sail", line 23, character 3 to line 23, character 4 Type error: Can only assign to identifiers with type register or reg, found identifier CA with type bit
2014-07-14Initial support for aliases and exit through the type system and the ↵Kathy Gray
interpreter. An alias can be read within the interpreter, but not written to. Exits aren't yet taken in the interpreter.
2014-07-04Correct error of forgetting type information that was allowing more programs ↵Kathy Gray
than it should have
2014-07-04Update power.sailGabriel Kerneis
2014-07-02Update Power exampleGabriel Kerneis
2014-06-27Update type conformance for overloading resolution on order arguments.Kathy Gray
Also, correctly type check nondet blocks
2014-06-26Adding better support for unspecified values in indexed vectorsKathy Gray
Also begining to add support for nondeterministic blocks and cleaning up some of the Many warnings on pattern matches
2014-06-25Add support for actions that read just a slice or single bit of a registerKathy Gray
2014-06-24Get vector length for to_inc_vec and to_dec_vec from the type system after ↵Kathy Gray
constraint solving (instead of hardcoding 64 as the default).
2014-06-23Get indexed vectors, particularly with default values, workingKathy Gray
2014-06-18Correct effect matching bug; and print out effect listsKathy Gray
2014-06-18Make hex constants work; improve utility of casts for selecting overloaded ↵Kathy Gray
functions
2014-06-11power.sail: store mode64bit in a bool registerGabriel Kerneis
This is now possible because we handle coercions better.
2014-06-11Update power.sail: remove some hacksGabriel Kerneis
2014-06-09Improve interaction after chat with PeterGabriel Kerneis
- remember mode (run, step or next) between instructions - display continuation by default in step mode - start in step mode by default - incompatible change: the shorthand for stack is now bt (=backtrace), since s becomes the shorthand for step - incompatible change: pressing enter now repeats the current mode, instead of "step"
2014-06-07Don't use cycle for fetch-decode-executeGabriel Kerneis
2014-06-07exts returns bit[64] instead of natGabriel Kerneis
Update power.sail to new, pretty-printer-based version
2014-06-07Copy main.bin in Sail repository for Power testGabriel Kerneis
2014-06-07Pattern-matching exhaustivenessGabriel Kerneis
2014-06-06Add wrapper script and --interactive for demoGabriel Kerneis
2014-06-05Shorten debug output for powerGabriel Kerneis
2014-06-05Enable single-stepping for test/power.sailGabriel Kerneis
2014-06-04Fixup type coercions and overloadingKathy Gray
Reduce the number of implicit coercions we're doing, expanding overloading and fixing up types of functions. Warning: test_power does not run as not all overloaded funcitons are implemented Warning: vector concatenation does not pretty print to sail source yet
2014-06-02Fix dependent-type for MEM in power.sailGabriel Kerneis
Now, constraint resolution works for test/power.sail
2014-05-30Add pretty-printer test (idempotence)Gabriel Kerneis
Fails in most cases because of "fresh" variables in type annotations that cannot be parsed again.
2014-05-30Remove bit-rotted, abonned test fileGabriel Kerneis
2014-04-15Use type information in vector concatenation pattern matching in interpreter.Kathy Gray
Does not properly bind variables (i.e. in the pattern 0b01:(bit[3]) a: 0b0001 subsequent uses of a will not be bound in the interpreter, though they are in the type checker), so until bug is fixed, treat all such identifiers as _
2014-04-04Cosmetic improvements (statistics, --quiet)Gabriel Kerneis
2014-04-04Improve Power executionGabriel Kerneis
- Move FDE loop to the OCaml side of the Power model (avoid leaking memory due to lack of TCO in interpreter) - Display cycle count - Check the value of CIA at the end of each cycle and stop if it is equal to the initial value of LR, returning the value in GPR3.
2014-04-03Power example now EXECUTES (almost) CORRECTLY! \o/Gabriel Kerneis
Warning: we set the link register to the address of the first instruction, which means the program is effectively an infinite loop returning 42 forever. (Except it slows down and leaks memory, probably because the interpreter doesn't optimise the tail call in fde_loop.) 42, 42, 42, 42, 42, 42, 42, 42, ...
2014-04-03Remove workarounds and update Power modelGabriel Kerneis
2014-04-03Exhibit two bugs about bitwise operators and vector castsGabriel Kerneis