[llvm-dev] Should we split llvm Support and ADT?

Dimitry Andric via llvm-dev llvm-dev at lists.llvm.org
Sat May 27 03:16:40 PDT 2017


On 27 May 2017, at 05:21, Hal Finkel via llvm-dev <llvm-dev at lists.llvm.org> wrote:
> On 05/26/2017 10:06 PM, Zachary Turner wrote:
>> It would be better, because a debug tablegen is slower than an optimized tablegen, but it's still slow and it doesn't address the problem that tablegen runs *at all* when it doesn't really need to. I think if tablegen wasn't running all the time we could incremental builds down from 15 minutes (and that's on my really powerful machine) to under 5, which seemed like a big productivity boost averaged out over all users x time
>> 
>> We've got a lot of random stuff in Support, not all really related and not all relevant to every project. It seems like we should be able to do better.
> 
> That certainly makes sense. I wonder, however, if we could have a more principled decomposition (instead of just "What's used by TableGen" vs. "What's not used by TableGen").

In FreeBSD we use a 'minimal' llvm library for this purpose:

https://github.com/freebsd/freebsd/blob/master/lib/clang/libllvmminimal/Makefile

This contains just the .c and .cpp files which are dependencies of llvm-tbglen and clang-tblgen.  The header dependencies aren't in there, those are generated using -MD.

-Dimitry

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 194 bytes
Desc: Message signed with OpenPGP
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20170527/d8a440c9/attachment.sig>


More information about the llvm-dev mailing list