-
I'm getting this crash from LLVM:
```
LLVM ERROR: underestimated function size
PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace.
Stack dum…
-
The GASNet team reports that when building the Chapel compiler with Intel classic compilers (which are admittedly stale), they see:
```bash
/scratch/upcnightly/EX-dirac-ibv-intel/bld/dbg/chapel/th…
-
```
Right now our eval loop is ~60% slower than CPython's because we collect
feedback. In theory, this isn't a problem, because if we have a good
hotness metric, then the eval loop performance doesn'…
-
```
Right now our eval loop is ~60% slower than CPython's because we collect
feedback. In theory, this isn't a problem, because if we have a good
hotness metric, then the eval loop performance doesn'…
-
Hello!
It would be great if we could have another variant of `boost::unordered_flat_map` where the keys and values are stored in separate arrays - "split" storage.
The main motivation for this is …
-
I wrote an llvm pass myself to generate some code. Sometimes these codes can run normally through the lli command, and sometimes they can't run normally, but each time the error is different, I am ver…
-
**Is your feature request related to a problem? Please describe.**
We would like the ability to run over the list of registered subsystems to check for annotations and any implementations.
**Descr…
-
Using LLVM 14.0.6 we are experiencing a consistent AddressSanitizer: use-after-poison error w/ stack trace:
```console
==32563==ERROR: AddressSanitizer: use-after-poison on address 0x621001049b60 at…
-
| | |
| --- | --- |
| Bugzilla Link | [38716](https://llvm.org/bz38716) |
| Version | unspecified |
| OS | Linux |
| Attachments | [dgetri.c (preprocessed)](https://user-images.githubusercontent.com…
-
```
Right now our eval loop is ~60% slower than CPython's because we collect
feedback. In theory, this isn't a problem, because if we have a good
hotness metric, then the eval loop performance doesn'…