[llvm-bugs] [Bug 24512] uint32 ->float vector cast always returns 0.0f on x86 pre-avx with unsafe-fp-math

[via llvm-bugs]

https://llvm.org/bugs/show_bug.cgi?id=24512

Ahmed Bougacha <ahmed.bougacha at gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|---                         |FIXED

--- Comment #21 from Ahmed Bougacha <ahmed.bougacha at gmail.com> ---
r248965.

Sanjay: is there some sort of IACA equivalent for AMD microarchitectures, or do
you have an alternative data source (e.g., a well-informed engineer with an AMD
opt guide?) we could use to decide? I'd obviously be OK with only enabling this
for current Intel CPUs if this turns out to be slower elsewhere.

-- 
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/20151001/ce7b00fb/attachment.html>