[llvm-bugs] [Bug 30543] 4.0 Regression in llvm.powi.f64 lowering on ARM

[via llvm-bugs]

https://llvm.org/bugs/show_bug.cgi?id=30543

Yichao Yu <yyc1992 at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|INVALID                     |---

--- Comment #4 from Yichao Yu <yyc1992 at gmail.com> ---
In another word if this is a frontend bug, it's also a bug in clang since the
following C program compiled with clang on ARM also produce 2.0

```
#include <stdio.h>

__attribute__((noinline)) double f(double v, int i)
{
    return __builtin_powi(v, i);
}

int main()
{
    printf("%f\n", f(2.0, 62));
    return 0;
}
```

I'll be happy to follow the clang frontend fix.

(Request for reopening while keeping it as llvm arm backend for now since I
currently can't get any IR to produce the correct result.)

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20160928/20d330c7/attachment-0001.html>