https://bugs.llvm.org/show_bug.cgi?id=41301
Bug ID: 41301
Summary: Optimizer fails to generate SIB displacements in
arithmetic expressions (x86_64)
Product: clang
Version: trunk
Hardware: PC
OS: Linux
Status: NEW
Severity: enhancement
Priority: P
Component: -New Bugs
Assignee: unassignedclangbugs at nondot.org
Reporter: bisqwit at iki.fi
CC: htmldeveloper at gmail.com, llvm-bugs at lists.llvm.org,
neeilans at live.com, richard-llvm at metafoo.co.uk
Consider this example code.
unsigned test(unsigned a,unsigned b)
{
return a*4 + b - 1;
}
GCC and ICC optimize this into a single instruction:
lea eax, [rsi + rdi*4 - 1]
However, Clang generates two instructions, no matter which optimization level
(tried -O3, -Os, -Ofast):
lea eax, [rsi + rdi*4]
add eax, -1
Not only is the latter longer in terms of instructions, it is longer in terms
of opcodes as well (4 bytes versus 6).
Same happens on both negative and positive displacements.
--
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/20190329/37e977ce/attachment.html>