Stream: t-compiler/const-eval

Topic: miri-rustc-tests


RalfJ (Oct 24 2018 at 12:22, on Zulip):

@Oli any idea why the miri-rustc-tests binary passes -Zmir-opt-level=3?

oli (Oct 24 2018 at 14:48, on Zulip):

nope, have you tried git blaming the line?

RalfJ (Oct 24 2018 at 15:01, on Zulip):

not yet

RalfJ (Oct 24 2018 at 15:02, on Zulip):

@Oli here it is: https://github.com/solson/miri/commit/8b81f5430f5f95f19e2c3d452971ef3380537a49#diff-5f266410c072c63b36718d107e25c370R152

oli (Oct 24 2018 at 15:03, on Zulip):

oh. right. a few PRs before that we were manually registering the optimization passes

oli (Oct 24 2018 at 15:03, on Zulip):

which caused us to diverge from the rustc passes

oli (Oct 24 2018 at 15:04, on Zulip):

so this change was to not change behavior

RalfJ (Oct 24 2018 at 15:10, on Zulip):

seems odd though, to make level 3 the default

RalfJ (Oct 24 2018 at 15:10, on Zulip):

I think level 0 should be default, and it actually is in my latest miri branch

RalfJ (Oct 24 2018 at 15:10, on Zulip):

and in making that consistent I realized this oddity in the rustc test runner

RalfJ (Oct 24 2018 at 15:15, on Zulip):

I'd prefer to remove it, make things more consistent?

oli (Oct 24 2018 at 15:19, on Zulip):

seems fine

RalfJ (Oct 24 2018 at 15:26, on Zulip):

done

Last update: Nov 15 2019 at 20:45UTC