In particular, i64/double are actually supposed to be aligned
to their size on RISC-V (at least according to the ELF psABI),
though it is unclear to me whether this actually caused any
issues.
The representation of TList(T) is changed from `{T*, u32}` to
`{T*, u32}*`. The old representation forbids changing the length of a
list when the list is passed as a parameter into functions, as the
length is passed by value. The representation now matches with nac3.
We don't need to know whether there's a outer finally block
that's already implicit in the current break and continue
target.
Signed-off-by: Michael Birtwell <michael.birtwell@oxionics.com>
When we have a trys inside a loop then we want to make sure any
finallys are executed by break and continue inside this try. But
this shouldn't pull finallys defined outside the loop in to the
loop. This change resets the `outer_final` attribute when
visiting for and while loops so that this doesn't happen.
Signed-off-by: Michael Birtwell <michael.birtwell@oxionics.com>
Otherwise, the exception message might be allocated on a stack, and will
become a dangling pointer when the exception is raised.
This will break some code that constructs exceptions with a function by
passing the message as a parameter because we cannot know if the parameter
is a constant. A way to mitigate this would be to defer this check to
LLVM IR codegen stage, and do inlining first for those exception
allocation functions, but I am not sure if we will guarantee inlining
for certain functions, and whether this is really needed.
Note that because we changed exception representation from using string
names as exception identifier into using integer IDs, we need to
initialize the embedding map in order to allocate the integer IDs. Also,
we can no longer print the exception names and messages from the kernel,
we will need the host to map exception IDs to names, and may need the
host to map string IDs to actual strings (messages can be static strings
in the firmware, or strings stored in the host only).
We now check for exception IDs for lit tests, which are fixed because we
preallocated all builtin exceptions.
Instead of removing basic blocks with no predecessor, we will now mark
and remove all blocks that are unreachable from the entry block. This
can handle loops that are dead code. This is needed as we will now
generate more complicated code for exception handling which the old dead
code eliminator failed to handle.
Exceptions are now allocated in the runtime when we raise the exception,
and destroyed when we exit the catch block. Nested exception and try
block is now supported, and should behave the same as in CPython.
Exceptions raised in except blocks will now unwind through finally
blocks, matching the behavior in CPython. Reraise will now preserve
backtrace.
Phi block LLVM IR generation is modified to handle landingpads, which
one ARTIQ IR will map to multiple LLVM IR.
Exception name is replaced by exception ID, which requires no
allocation. Other strings in the exception can now be 'host-only'
strings, which is represented by a CSlice with len = usize::MAX and
ptr = key, to avoid the need for allocation when raising exceptions
through RPC.
LLVM 6 seemed not to mind the mismatch, but more recent
versions produce miscompilations without this.
Needs llvmlite support (GitHub: numba/llvmlite#702).
In the origin implementation, the `nowrite` flag literally means not writing memory at all.
Due to the usage of flags on certain functions, it results in the same issues found in artiq-zynq after optimization passes. (M-Labs/artiq-zynq#119)
A fix wrote by @dnadlinger can resolve this issue. (c1e46cc7c8)
The previous code could have never worked as-is, as the result slot
went unused, and it tried to append the load instruction to the
block just terminated with the invoke.
GitHub: Fixes#1506, #1531.
Since we don't implement any integer-like operations for TBool
(addition, bitwise not, etc.), TBool is currently neither
strictly equivalent to builtin bool nor numpy.bool_, but through
very obvious compiler errors (operation not supported) rather than
silently different runtime behaviour.
Just mapping both to TBool thus is a huge improvement over the
current behaviour (where numpy.False_ is a true-like object). In
the future, we could still implement more operations for TBool,
presumably following numpy.bool_ rather than the builtin type,
just like builtin integers get translated to the numpy-like
TInt{32,64}.
GitHub: Fixes#1275.
Previously, any type would be accepted for the test expression,
leading to internal errors in the code generator if the passed
value wasn't in fact a bool.
array([...]), the constructor for NumPy arrays, currently has the
status of some weird kind of macro in ARTIQ Python, as it needs
to determine the number of dimensions in the resulting array
type, which is a fixed type parameter on which inference cannot
be performed.
This leads to an ambiguity for empty lists, which could contain
elements of arbitrary type, including other lists (which would
add to the number of dimensions).
Previously, I had chosen to make array([]) to be of completely
indeterminate type for this reason. However, this is different
to how the call behaves in host NumPy, where this is a well-formed
call creating an empty 1D array (or 2D for array([[], []]), etc.).
This commit adds special matching for (recursive lists of) empty
ListT AST nodes to treat them as scalar dimensions, with the
element type still unknown.
This also happens to fix type inference for embedding empty 1D
NumPy arrays from host object attributes, although multi-dimensional
arrays will still require work (see GitHub #1633).
GitHub: Fixes#1626.
Strided slicing of one-dimensional arrays (i.e. with non-trivial
steps) might have previously been working, but would have had
different semantics, as all slices were copies rather than a view
into the original data.
Fixing this in the future will require adding support for an index
stride field/tuple to our array representation (and all the
associated indexing logic).
GitHub: Fixes#1627.
This was a long-standing issue affecting both lists and
the new NumPy array implementation, just caused by the
generic inference passes not being run on the slice
subexpressions (and thus e.g. ints not being monomorphized).
GitHub: Fixes#1632.
This allows assert() to be used on Zynq, where abort() is not
currently implemented for kernels. Furthermore, this is arguably
the more natural implementation of assertions on all kernel targets
(i.e. where embedding into host Python is used), as it matches host
Python behavior, and the exception information actually makes it to
the user rather than leading to a ConnectionClosed error.
Since this does not implement printing of the subexpressions, I
left the old print+abort implementation as default for the time
being.
The lit/integration/instance.py diff isn't just a spurious change;
the exception-based assert implementation exposes a limitation in
the existing closure lifetime tracking algorithm (which is not
supposed to be what is tested there).
GitHub: Fixes#1539.