[llvm-commits] [PATCH] Re-introduce llvm::sys::getHostTriple() function.

[Sebastian Pop]

Peter Collingbourne wrote:
> In r143502, we renamed getHostTriple() to getDefaultTargetTriple()
> as part of work to allow the user to supply a different default
> target triple at configure time.  This change also affected the JIT.
> However, it is inappropriate to use the default target triple in the
> JIT in most circumstances because this will not necessarily match
> the host architecture, leading to illegal instruction and other such
> errors at run time.
> 
> Reintroduce the getHostTriple() function for use in the JIT and its
> clients, and cause the JIT to use it.  At the same time, fix JIT
> for the 32-on-64-bit (and vice versa) build on non-Apple platforms
> by causing getHostTriple() to change the architecture variant if
> necessary.
> 
> In this email thread:
> http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20120102/134416.html
> an objection was raised against a similar patch on the grounds
> that the JIT could deliberately be used to compile binaries for a
> different architecture.  However, the normal usage scenario for the
> JIT is that the code is being compiled and run on the same machine,
> and if the client is doing anything out of the ordinary it should
> know the target architecture (which would not necessarily match the
> default target architecture) and set up the target triple itself.
> 

Your patch looks good to me.  Please wait for Joerg to comment on your patch
before committing.

Thanks,
Sebastian
-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation