[llvm-bugs] [Bug 42079] New: [X86] avx512vl vcvtps2pd isel patterns can narrow a volatile load

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=42079

            Bug ID: 42079
           Summary: [X86] avx512vl vcvtps2pd isel patterns can narrow a
                    volatile load
           Product: libraries
           Version: trunk
          Hardware: PC
                OS: Windows NT
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: Backend: X86
          Assignee: unassignedbugs at nondot.org
          Reporter: craig.topper at gmail.com
                CC: craig.topper at gmail.com, llvm-bugs at lists.llvm.org,
                    llvm-dev at redking.me.uk, spatel+llvm at rotateright.com

This IR will contains a volatile 128-bit load

define <2 x double> @test(<4 x float>* %x) {
  %a = load volatile <4 x float>, <4 x float>* %x
  %b = shufflevector <4 x float> %a, <4 x float> %a, <2 x i32> <i32 0, i32 1>
  %c = fpext <2 x float> %b to <2 x double>
  ret <2 x double> %c
}


When compiled with avx512f this assembly is produced which only loads 128 bits.

vcvtps2pd       (%rdi), %xmm0


This is probably not the only example of this bug in our isel patterns.

-- 
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/20190530/5306df22/attachment.html>