[Build] Refactor OPE code to adhere to stricter build warnings #443
Conversation
VivekPanyam
force-pushed
the
backends_cc_warnings
branch
from
d07374b to
ad04349
Compare
VivekPanyam
force-pushed
the
multiprocess_cc_warnings
branch
from
81a1700 to
63939bd
Compare
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #443 +/- ##
=======================================
Coverage 88.23% 88.23%
=======================================
Files 101 102 +1
Lines 6356 6356
=======================================
Hits 5608 5608
Misses 748 748 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
|
|
||
| // Null terminated char * array | ||
| char *env_arr[env.size() + 1]; | ||
| std::vector<char *> env_arr(env.size() + 1); |
There was a problem hiding this comment.
in case of vector, all elements are initialized with 0 (nullptr) already because it calls (char*) init for every element, so next line is redundant (it was necessary for regular array though)
|
|
||
| auto base_ptr = byte_block->get_raw_data_ptr(); | ||
| auto max_len = dims[dims.size() - 1]; | ||
| auto max_len = static_cast<size_t>(dims[dims.size() - 1]); |
There was a problem hiding this comment.
I think this cast is not needed and as result you don't need to cast it then at line 229 again
| // with `allocator` without making a copy, you could use this function | ||
| std::shared_ptr<NeuropodTensor> wrap_existing_tensor(NeuropodTensorAllocator & allocator, | ||
| std::shared_ptr<NeuropodTensor> tensor) | ||
| static std::shared_ptr<NeuropodTensor> wrap_existing_tensor(NeuropodTensorAllocator & allocator, |
There was a problem hiding this comment.
Interesting why static is necessary here, did compiler require it to be static?
Some thoughts:
In C++, anonymous namespace was recommended and "static" was even deprecated for vars and functions. It was undeprecated in C++11
Some discussion clarifies details https://stackoverflow.com/questions/154469/unnamed-anonymous-namespaces-vs-static-functions
Anyway, please share details on it
vkuzmin-uber
left a comment
vkuzmin-uber
left a comment
There was a problem hiding this comment.
Looks good. Please check I left comment and question.
VivekPanyam
force-pushed
the
backends_cc_warnings
branch
from
ad04349 to
39a9c4c
Compare
Summary:
Refactor code under
source/neuropod/multiprocessto adhere to the stricter set of warnings introduced in #437Test Plan:
CI