| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
| |
Signed-off-by: Josh Morman <jmorman@perspectalabs.com>
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
First batch of changes:
```
find […] -print0 | xargs -0 sed -i -r '/get_initial_sptr/{:nxt N;/;/!b nxt;s/get_initial_sptr\(\s*new ([^(]+)(.*)\)\)/make_block_sptr<\1>\2)/}'
```
Then:
* Back out `sptr_magic.h` edits
* Change some `friend` clauses
* clang-format the whole thing
* Update checksums in the pybind11 bindings files
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Both of these defects are for the second parameter of the `constellation`
class' `map_to_points` function that accepts a pointer to an array of
`gr_complex` values. In both of these defects, a class is calling this function
and passing the address of a single `gr_complex` value in place of an array
pointer. The only reason this isn't exploding with SEGFAULTs is because both of
these functions happen to use the default constructor of `constellation`, which
sets the loop limit in `map_to_points` to `1`. It's generally a dangerous
design, but changing the function prototype seems heavy heanded, and adding
additional conditionals will end up affecting the fast-path. For now, I am just
documenting this oddity in the code.
|
| | |
|
| |
|
|
| |
filter: updated fir_filter_* by adding a method that allows the update of a single tap in the filter in a way that propagates to all of the aligned tap vectors. Mostly to support the adaptive equalizers.
|
| |
|
