[PATCH] D26042: Consolidate BumpPtrAllocators.

[Mehdi Amini via llvm-commits]

> On Oct 28, 2016, at 12:58 PM, Rafael Espíndola <rafael.espindola at gmail.com> wrote:
> 
>> Be careful with your use of "we". You never got consensus with the way
>> things went in lld.
>> I'm just expressing my own disagreement here, as a reminder of this, but I
>> know that I'm not the only one in the community that believe that there was
>> a "steam roll" there.
> 
> I am most certainly part of the "we”.

Sure, I had a pretty good idea of who is “we”. I just want to point that “we” isn’t “the LLVM community”. 
When used generically as above, it can too easily slip toward “generality”. (I tried to carefully use a single “I” to express my opnion, and added “a few other folks” where I have data without insinuating any consensus or majority).

> Pretending we are writing
> something generic in here would be a complete waste of effort. We are
> writing a linker. Just a linker.
> So having global state is not technical debt. It is not something we
> want to change in the future, it is in line with the lld design as a
> stand alone linker.

Right, I still disagree that this is acceptable for an LLVM project, and I don’t think I'm the only one, and you guys haven’t really convinced us of the opposite. Just something to keep it in mind.


I’m not sure if you’re trying to restart the topic here, it’s pretty clear that we’re in disagreement, and re-asserting your opinion like “facts” doesn’t make it more “right”.

This “design” choice you made (against many other in the community) prevents other from contributing. *I* certainly don’t feel comfortable working on lld-ELF at least (I am possibly not the only one, I don’t pretend having any data).

Best,

— 
Mehdi