[PATCH] LLD: Remove YAML/Native round-trip passes

[Rui Ueyama]

On Thu, Feb 19, 2015 at 5:50 PM, Rafael Espíndola <
rafael.espindola at gmail.com> wrote:

> On 19 February 2015 at 19:41, Michael Spencer <bigcheesegs at gmail.com>
> wrote:
> > On Wed, Feb 11, 2015 at 4:37 PM, Rui Ueyama <ruiu at google.com> wrote:
> >> I found that Writing and reading Native files from YAML reader is
> actually
> >> tricky. Readers generally don't need a linking context, but writers do.
> So
> >> it's not easy to call a writer from a reader. Seems it needs a plumbing
> >> work.
> >>
> >> Then I tried to convert a YAML file to Native using yaml2obj
> -format=native.
> >> yaml2obj didn't actually support Native format because Native format
> lives
> >> only in LLD. Yet it's intended to be an alternative, hopefully better
> object
> >> file format, it's not supported by anyone except the linker. There's no
> >> consumer producing a file in that file format.
> >
> > We haven't taught clang to produce it yet because the format isn't
> > finalized. The performance gain from it is still rather large, and I
> > feel it's an important feature.
>
> Please note that we don't have evidence of this yet.
>
> I would be very opposed to adding a new file format to llvm proper until
>
> * lld is production quality
> * lld is as fast as gold on ELF
> * there is experimental evidence that a new format would make it even
> faster without compromising its ability to report errors.
>

I agree with all the points. If we eventually think that we need a new file
format for speed, the current native file format needs to be evaluated to
see if it's suitable for that purpose, along with other points such as
extendability of file format, complexity, etc.

For now let me go with this patch to unblock stuffs.

Thank you for reviewing.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150219/6787b6e9/attachment.html>