[llvm-bugs] [Bug 44113] New: _mm256_shuffle_ps(_, _, 0x44) compiles to vunpcklpd with -march= icelake-client

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=44113

            Bug ID: 44113
           Summary: _mm256_shuffle_ps(_,_,0x44) compiles to vunpcklpd with
                    -march= icelake-client
           Product: clang
           Version: 9.0
          Hardware: PC
                OS: All
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: C++
          Assignee: unassignedclangbugs at nondot.org
          Reporter: DrTroll at gmx.de
                CC: blitzrakete at gmail.com, dgregor at apple.com,
                    erik.pilkington at gmail.com, llvm-bugs at lists.llvm.org,
                    richard-llvm at metafoo.co.uk

In a recent Stackoverflow discussion (link:
https://stackoverflow.com/questions/58954801/avx-equivalent-for-mm-movelh-ps)
we found out that the instruction _mm256_shuffle_ps(_,_,0x44) is compiled to
vunpcklpd by Clang. This is a possible optimization for Skylake and other
processors which have identical throughput for shuffles and unpacks.
But as the Stackoverflow-user Peter Cordes mentioned in his answer, Ice Lake
processors have a higher throughput for shuffles than for unpacks:

unpack: 
https://www.uops.info/table.html?search=vunpcklp&cb_lat=on&cb_tp=on&cb_uops=on&cb_ports=on&cb_SKL=on&cb_ICL=on&cb_measurements=on&cb_iaca30=on&cb_doc=on&cb_avx=on&cb_avx2=on

shuffle:
https://www.uops.info/table.html?search=vshufp&cb_lat=on&cb_tp=on&cb_uops=on&cb_ports=on&cb_SKL=on&cb_ICL=on&cb_measurements=on&cb_iaca30=on&cb_doc=on&cb_avx=on&cb_avx2=on

Therefore, the performed replacement of the shuffle is contra-productive on an
Ice Lake processor.
Even with -march= icelake-client, Clang replaces the shuffle with vunpcklpd:

https://godbolt.org/z/MqtJXY


Same goes for _mm256_shuffle_ps(_,_,0xee) vunpckhpd.

-- 
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/20191122/170c6e3c/attachment.html>