[llvm-bugs] [Bug 34156] New: BPF target: implement support for emitting BPF_ALU ops (32 bit variants)

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=34156

            Bug ID: 34156
           Summary: BPF target: implement support for emitting BPF_ALU ops
                    (32 bit variants)
           Product: new-bugs
           Version: unspecified
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: new bugs
          Assignee: unassignedbugs at nondot.org
          Reporter: daniel at iogearbox.net
                CC: llvm-bugs at lists.llvm.org

Currently the BPF target only supports emitting BPF_ALU64 alu instructions, but
it doesn't support generating BPF_ALU (32 bit) alu instructions. It would be
really great if this could be implemented so that LLVM can potentially generate
more efficient code (e.g. doesn't need to do shifts to potentially clear upper
bits). The JITs could also emit smaller opcode images by using their 32bit ALU
mappings. There are also 32bit JITs that could translate the BPF insns into
more efficient code when LLVM can choose 32 over 64bit.

-- 
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/20170811/5d17bfb7/attachment-0001.html>