Fix docstring <-> code mismatch of CompilerConfig
#422
Conversation
|
Ah, forgot to change the uses in the backends 🤦 |
|
Maybe it's better to sync the other way? I decided against renaming the kwarg to |
|
Either way is fine by me - I couldn't figure out which was intended, since both argument names stem from the same commit :') I'll change it to |
Seelengrab
force-pushed
the
entry_name_fix
branch
from
7df5275 to
f6b8cb2
Compare
Yikes. 👀 |
Codecov ReportPatch coverage has no change and project coverage change:
Additional details and impacted files@@ Coverage Diff @@
## master #422 +/- ##
==========================================
- Coverage 86.92% 79.14% -7.78%
==========================================
Files 24 24
Lines 2922 2892 -30
==========================================
- Hits 2540 2289 -251
- Misses 382 603 +221
... and 12 files with indirect coverage changes Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. ☔ View full report in Codecov by Sentry. |
| @@ -114,12 +114,12 @@ module LazyCodegen | |||
| @assert !cc.compiled | |||
| job = cc.job | |||
|
|
|||
| entry_name, jitted_mod = JuliaContext() do ctx | |||
| name, jitted_mod = JuliaContext() do ctx | |||
There was a problem hiding this comment.
FWIW, this introduced a bug. By assigning to name, the call to LLVM.name in the block is broken.
Fixed in #427
There was a problem hiding this comment.
Odd that this wasn't caught in CI 🤔 But yeah, shows that a quick sed -i is often more dangerous than one might think 😬
There was a problem hiding this comment.
It was caught, but resulted in a timeout which we ignored...
There was a problem hiding this comment.
Ah :| Mea culpa then! Still surprising that the others ran through, but it is what it is. Good thing you caught & fixed it 👍
The docstring said
entry_name, but the code usednameas the field - this PR makes them consistently useentry_name.