Commit Graph

49 Commits

Author SHA1 Message Date
Damien George 5a2599d962 py: Replace POP_BLOCK and POP_EXCEPT opcodes with POP_EXCEPT_JUMP.
POP_BLOCK and POP_EXCEPT are now the same, and are always followed by a
JUMP.  So this optimisation reduces code size, and RAM usage of bytecode by
two bytes for each try-except handler.
2019-03-05 16:09:58 +11:00
Damien George e1fb03f3e2 py: Fix VM crash with unwinding jump out of a finally block.
This patch fixes a bug in the VM when breaking within a try-finally.  The
bug has to do with executing a break within the finally block of a
try-finally statement.  For example:

    def f():
        for x in (1,):
            print('a', x)
            try:
                raise Exception
            finally:
                print(1)
                break
            print('b', x)
    f()

Currently in uPy the above code will print:

    a 1
    1
    1
    segmentation fault (core dumped)  micropython

Not only is there a seg fault, but the "1" in the finally block is printed
twice.  This is because when the VM executes a finally block it doesn't
really know if that block was executed due to a fall-through of the try (no
exception raised), or because an exception is active.  In particular, for
nested finallys the VM has no idea which of the nested ones have active
exceptions and which are just fall-throughs.  So when a break (or continue)
is executed it tries to unwind all of the finallys, when in fact only some
may be active.

It's questionable whether break (or return or continue) should be allowed
within a finally block, because they implicitly swallow any active
exception, but nevertheless it's allowed by CPython (although almost never
used in the standard library).  And uPy should at least not crash in such a
case.

The solution here relies on the fact that exception and finally handlers
always appear in the bytecode after the try body.

Note: there was a similar bug with a return in a finally block, but that
was previously fixed in b735208403
2019-03-05 16:05:05 +11:00
Damien George c2074e7b66 tests/cmdline/cmd_showbc.py: Fix test to explicitly declare nonlocal.
The way it was written previously the variable x was not an implicit
nonlocal, it was just a normal local (but the compiler has a bug which
incorrectly makes it a nonlocal).
2018-10-27 23:57:14 +11:00
Damien George 165aab12a3 py/repl: Generalise REPL autocomplete to use qstr probing.
This patch changes the way REPL autocomplete finds matches.  It now probes
the target object for all qstrs via mp_load_method_maybe to look for a
match with the given input string.  Similar to how the builtin dir()
function works, this new algorithm now find all methods and instances of
user-defined classes including attributes of their parent classes.  This
helps a lot at the REPL prompt for user-discovery and to autocomplete names
even for classes that are derived.

The downside is that this new algorithm is slower than the previous one,
and in particular will be slower the more qstrs there are in the system.
But because REPL autocomplete is primarily used in an interactive way it is
not that important to make it fast, as long as it is "fast enough" compared
to human reaction.

On a slow microcontroller (CPU running at 16MHz) the autocomplete time for
a list of 35 names in the outer namespace (pressing tab at a bare prompt)
takes about 160ms with this algorithm, compared to about 40ms for the
previous implementation (this time includes the actual printing of the
names as well).  This time of 160ms is very reasonable especially given the
new functionality of listing all the names.

This patch also decreases code size by:

   bare-arm:    +0
minimal x86:  -128
   unix x64:  -128
unix nanbox:  -224
      stm32:   -88
     cc3200:   -80
    esp8266:   -92
      esp32:   -84
2018-02-19 16:12:44 +11:00
Damien George 35a759dc1d tests: Add some more tests to improve coverage of py/parse.c. 2017-12-19 16:13:00 +11:00
Damien George 0864a6957f py: Clean up unary and binary enum list to keep groups together.
2 non-bytecode binary ops (NOT_IN and IN_NOT) are moved out of the
bytecode group, so this change will change the bytecode format.
2017-10-05 10:49:44 +11:00
Paul Sokolovsky 9d836fedbd py: Clarify which mp_unary_op_t's may appear in the bytecode.
Not all can, so we don't need to reserve bytecodes for them, and can
use free slots for something else later.
2017-09-25 16:35:19 -07:00
Paul Sokolovsky b8ee7ab5b9 py/runtime0.h: Put inplace arith ops in front of normal operations.
This is to allow to place reverse ops immediately after normal ops, so
they can be tested as one range (which is optimization for reverse ops
introduction in the next patch).
2017-09-08 00:10:10 +03:00
Paul Sokolovsky 50b9329eba py/runtime0.h: Move MP_BINARY_OP_DIVMOD to the end of mp_binary_op_t.
It starts a dichotomy of mp_binary_op_t values which can't appear in the
bytecode. Another reason to move it is to VALUES of OP_* and OP_INPLACE_*
nicely adjacent. This also will be needed for OP_REVERSE_*, to be soon
introduced.
2017-09-07 11:26:42 +03:00
Paul Sokolovsky d4d1c45a55 py/runtime0.h: Move relational ops to the beginning of mp_binary_op_t.
This is to allow to encode arithmetic operations more efficiently, in
preparation to introduction of __rOP__ method support.
2017-09-07 10:55:43 +03:00
Damien George 30badd1ce1 tests: Add tests for calling super and loading a method directly. 2017-04-22 23:39:38 +10:00
Damien George 5335942b59 py/compile: Refactor handling of special super() call.
This patch refactors the handling of the special super() call within the
compiler.  It removes the need for a global (to the compiler) state variable
which keeps track of whether the subject of an expression is super.  The
handling of super() is now done entirely within one function, which makes
the compiler a bit cleaner and allows to easily add more optimisations to
super calls.

Changes to the code size are:

   bare-arm: +12
    minimal:  +0
   unix x64: +48
unix nanbox: -16
     stmhal:  +4
     cc3200:  +0
    esp8266: -56
2017-04-22 21:46:32 +10:00
Damien George 5255255fb9 py: Create str/bytes objects in the parser, not the compiler.
Previous to this patch any non-interned str/bytes objects would create a
special parse node that held a copy of the str/bytes data.  Then in the
compiler this data would be turned into a str/bytes object.  This actually
lead to 2 copies of the data, one in the parse node and one in the object.
The parse node's copy of the data would be freed at the end of the compile
stage but nevertheless it meant that the peak memory usage of the
parse/compile stage was higher than it needed to be (by an amount equal to
the number of bytes in all the non-interned str/bytes objects).

This patch changes the behaviour so that str/bytes objects are created
directly in the parser and the object stored in a const-object parse node
(which already exists for bignum, float and complex const objects).  This
reduces peak RAM usage of the parse/compile stage, simplifies the parser
and compiler, and reduces code size by about 170 bytes on Thumb2 archs,
and by about 300 bytes on Xtensa archs.
2017-02-24 13:43:43 +11:00
Damien George c889f01b8d tests/cmdline/cmd_parsetree: Update to work with changes to grammar. 2017-02-17 12:29:51 +11:00
Damien George 0ec957d7c5 tests/cmdline: Update cmd_parsetree test for changes to grammar order. 2017-02-16 19:45:07 +11:00
Damien George 86b3db9cd0 tests/cmdline/cmd_showbc: Update to work with recent changes. 2017-02-16 18:38:07 +11:00
Damien George 861b001783 tests/cmdline: Update tests to pass with latest changes to bytecode. 2017-02-16 18:38:07 +11:00
Damien George f4df3aaa72 py: Allow bytecode/native to put iter_buf on stack for simple for loops.
So that the "for x in it: ..." statement can now work without using the
heap (so long as the iterator argument fits in an iter_buf structure).
2017-02-16 18:38:06 +11:00
Rami Ali b7024f0f64 tests/cmdline: Improve repl.c autocomplete test coverage. 2016-12-29 17:14:04 +11:00
Damien George cdec7ba30e tests: Add a coverage test for printing the parse-tree. 2016-12-22 11:26:06 +11:00
Damien George 453c2e8f55 tests/cmdline: Improve coverage test for printing bytecode. 2016-10-17 11:23:37 +11:00
Damien George 5e22afce41 tests: Improve test coverage of py/compile.c. 2016-10-11 12:30:32 +11:00
stijn 7f19b1c3eb tests: Fix expected output of verbose cmdline test
The output might contain more than one line ending in 5b so properly skip
everything until the next known point.
This fixes test failures in appveyor debug builds.
2016-10-05 12:58:50 +02:00
Damien George f65e4f0b8f tests/cmdline/cmd_showbc: Fix test now that 1 value is stored on stack.
This corresponds to the change in the way exception values are stored on
the Python value stack.
2016-09-27 13:22:06 +10:00
Damien George bb954d80a4 tests: Get cmdline verbose tests running again.
The showbc function now no longer uses the system printf so works
correctly.
2016-09-20 11:33:19 +10:00
Damien George 3b5affa0d1 tests/cmdline: Add test for -O option to check optimisation value. 2016-08-26 22:29:38 +10:00
Alex March 69d9e7d27d py/repl: Check for an identifier char after the keyword.
- As described in the #1850.
- Add cmdline tests.
2016-02-17 08:56:15 +00:00
Damien George 3aa7dd23c9 unix: Add exit and paste-mode hints to shell startup banner.
Thanks to @nyov for the initial patch.
2015-10-12 00:19:00 +01:00
Damien George 0334058fa4 Rename "Micro Python" to "MicroPython" in REPL, help, readme's and misc. 2015-10-12 00:06:25 +01:00
Alex March 15018291b2 tests: Add escaped quotes tests for REPL.
Test possible combinations of single and double quotes with escaped
quotes and parenthesis with and without function calls in REPL.

Covers: #1419
2015-09-19 14:06:23 +01:00
Damien George 3ca84026db unix: Enable REPL auto-indent. 2015-09-12 22:09:18 +01:00
Paul Sokolovsky a3fe307400 tests: Consolidate all feature check snippets under feature_check/. 2015-08-30 11:36:42 +03:00
Tom Soulanille 20f40c3229 tests: Remove over-specification of startup banner 2015-08-08 12:37:36 +01:00
Tom Soulanille 89852d38ef run-tests: Test REPL emacs keys, but only if present.
Uses cmdline/repl_emacs_check.py to check for presence of emacs keys in
repl before doing full feature test.
2015-08-06 00:03:44 +01:00
Tom Soulanille 3dd0b69e46 run-tests: Use PTY when running REPL tests. 2015-07-26 15:23:11 +01:00
Damien George 59fba2d6ea py: Remove mp_load_const_bytes and instead load precreated bytes object.
Previous to this patch each time a bytes object was referenced a new
instance (with the same data) was created.  With this patch a single
bytes object is created in the compiler and is loaded directly at execute
time as a true constant (similar to loading bignum and float objects).
This saves on allocating RAM and means that bytes objects can now be
used when the memory manager is locked (eg in interrupts).

The MP_BC_LOAD_CONST_BYTES bytecode was removed as part of this.

Generated bytecode is slightly larger due to storing a pointer to the
bytes object instead of the qstr identifier.

Code size is reduced by about 60 bytes on Thumb2 architectures.
2015-06-25 14:42:13 +00:00
Damien George c5029bcbf3 py: Add MP_BINARY_OP_DIVMOD to simplify and consolidate divmod builtin. 2015-06-13 23:36:30 +01:00
Damien George 9ae3fc6523 unix: Add option to use uPy readline, and enable by default.
This gets uPy readline working with unix port, with tab completion and
history.  GNU readline is still supported, configure using
MICROPY_USE_READLINE variable.
2015-05-27 15:59:43 +01:00
Damien George c2a4e4effc py: Convert hash API to use MP_UNARY_OP_HASH instead of ad-hoc function.
Hashing is now done using mp_unary_op function with MP_UNARY_OP_HASH as
the operator argument.  Hashing for int, str and bytes still go via
fast-path in mp_unary_op since they are the most common objects which
need to be hashed.

This lead to quite a bit of code cleanup, and should be more efficient
if anything.  It saves 176 bytes code space on Thumb2, and 360 bytes on
x86.

The only loss is that the error message "unhashable type" is now the
more generic "unsupported type for __hash__".
2015-05-12 22:46:02 +01:00
Damien George 9a42eb541e py: Fix naming of function arguments when function is a closure.
Addresses issue #1226.
2015-05-06 13:55:33 +01:00
Damien George 367d4d1098 tests: Fix cmd_showbc now that LOAD_CONST_ELLIPSIS bytecode is gone. 2015-05-05 23:58:52 +01:00
Damien George 8c1d23a0e2 py: Modify bytecode "with" behaviour so it doesn't use any heap.
Before this patch a "with" block needed to create a bound method object
on the heap for the __exit__ call.  Now it doesn't because we use
load_method instead of load_attr, and save the method+self on the stack.
2015-04-24 01:52:28 +01:00
Damien George c9aa1883ed py: Simplify bytecode prelude when encoding closed over variables. 2015-04-07 00:08:17 +01:00
Damien George 9dd3640464 tests: Add missing tests for builtins, and many other things. 2015-04-04 22:05:30 +01:00
Damien George d478fc75b3 tests: Adjust expected output, since Travis can't do git describe. 2015-03-20 21:31:50 +00:00
Damien George 1004535237 tests: Make cmdline tests more stable by using regex for matching. 2015-03-20 17:25:25 +00:00
Damien George 0683c1ceef tests: Don't try to verify amount of memory used in cmd_showbc test. 2015-03-14 17:38:41 +00:00
Damien George 703c009681 tests: Add cmdline test to test showbc code. 2015-03-14 14:06:20 +00:00
Damien George 143c34109c tests: Add ability to test uPy cmdline executable.
This allows to test options passed to cmdline executable, as well as the
behaviour of the REPL.
2015-03-13 10:58:34 +00:00