[llvm-bugs] [Bug 27972] New: missing optimization

[via llvm-bugs]

https://llvm.org/bugs/show_bug.cgi?id=27972

            Bug ID: 27972
           Summary: missing optimization
           Product: lld
           Version: unspecified
          Hardware: PC
                OS: All
            Status: NEW
          Severity: normal
          Priority: P
         Component: All Bugs
          Assignee: unassignedbugs at nondot.org
          Reporter: rafael.espindola at gmail.com
                CC: llvm-bugs at lists.llvm.org
    Classification: Unclassified

Reporting just for completeness, it is not clear if this is worth it.

Given

call foo at plt
call *foo at gotpcrel(%rip)

we know that there will be a non lazy reference to foo. Given that, we could
avoid the .got.plt entry and produce a non lazy plt that just uses the regular
got entry.

The reason I am not sure this is useful is that I don't expect it to show up a
lot. It would show up if mixing -fno-plt and -fplt code, but that is probably
not common.

The other case it would show up is something like

void f(void);
void *g(void) {return f; }

which is probably also not common.

I think this is the only missing optimization that is documented in the x86_64
abi.

-- 
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/20160601/8899eb97/attachment.html>