[llvm-commits] [llvm] r69350 - in /llvm/trunk: lib/Target/X86/X86ISelLowering.cpp lib/Target/X86/X86Instr64bit.td lib/Target/X86/X86InstrInfo.td test/CodeGen/X86/2009-04-17-tls-fast.ll

[Rafael Espindola]

>
> Would you mind explaining why this is?

The modes have a very strict definition:

http://people.redhat.com/drepper/tls.pdf

I think the rationale is to make  it easier for the linker to do
optimizations. In particular, the testcase include in the patch would
produces errors in both GNU ld and gold.

We could (should?) define our own models once doing LTO is easier and
we don't really care which optimizations the linker wants to do.

> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>

Cheers,
-- 
Rafael Avila de Espindola

Google | Gordon House | Barrow Street | Dublin 4 | Ireland
Registered in Dublin, Ireland | Registration Number: 368047