1.52.1 - Blog post explaining the fingerprint issue. #836
Conversation
|
Moving publish date to Monday morning. |
…diagnostic output.
Co-authored-by: Tyler Mandry <tmandry@gmail.com>
since this post isn't going out today, it should use more abstract language to refer to when 1.52.0 came out.
|
|
||
| We recommend that users of 1.52.0 disable incremental compilation, to avoid running into this problem. | ||
|
|
||
| We do *not* recommend that users of 1.52.0 downgrade to an earlier version of Rust in response to this problem. As noted above, there is at least one instance of a silent [miscompilation][issue-82920] caused by incremental compilation that was not caught until we added the fingerprint checking. |
There was a problem hiding this comment.
And we don't know how far back these bugs are present.
There was a problem hiding this comment.
(I think the text in the summary does a good job of explaining that aspect of things. I don't think we need to reiterate the point here.)
Just a first draft, need to re-read and adjust the structure a little still.
|
Just pushed an update in light of my expectation that we'll be releasing a point release on Monday - see rust-lang/rust#85097 - but I expect to continue working on the language tomorrow, thank you everyone for the comments so far. |
|
Hey @carols10cents I applied a bunch of feedback from your comments I think, though some of them are also a bit stale after some movement - I would really love a review if you have a chance to do so :) |
|
@Mark-Simulacrum looking now! |
There was a problem hiding this comment.
This is lots, lots better!!!
I think the only bits of my first round of feedback that I don't see incorporated are:
- At least a short statement that we will be investigating and retrospecting on the root causes of how this change got to stable before we expected it to and how to prevent similar situations in the future
- At least a short definition of what "miscompilation" means in this case to give a sense of what kind of "wrong" the compiler may have been silently generating, because it sounds pretty scary without further elaboration
Without that information, I would bet money that those two topics will be the ones that get the most comments/questions when this is posted.
Co-authored-by: Carol (Nichols || Goulding) <193874+carols10cents@users.noreply.github.com>
Co-authored-by: Carol (Nichols || Goulding) <193874+carols10cents@users.noreply.github.com>
|
Hopefully addressed all remaining points. Thanks! |
There was a problem hiding this comment.
This looks great now (with the few small edits @est31 pointed out that I agree with). Thank you so much for working on this ❤️
| our confidence in the fixes, may issue a 1.52.2 point release which backports | ||
| those fixes to the stable channel. Users wishing to help us test can use the | ||
| nightly channel, and report bugs to rust-lang/rust with any ICEs they are | ||
| seeing. We do not expect at this time to disable incremental by default on the |
There was a problem hiding this comment.
I think this raises the question "will incremental be disabled on beta?"
It might be good to preemptively address that question.
|
@wesleywiser I think I'm currently inclined to leave beta and nightly in the non-disabled incremental states, but it's definitely a good idea to call out their status. I think it's not unlikely 1.53 will need to also ship without incremental, though. |
Co-authored-by: est31 <est31@users.noreply.github.com>
Mark-Simulacrum
dismissed stale reviews from wesleywiser, steveklabnik, and carols10cents
via
c250c4e
Co-authored-by: est31 <est31@users.noreply.github.com>
Co-authored-by: est31 <est31@users.noreply.github.com>
|
Okay, merged in @est31's suggestions and a paragraph about the beta channel. @wesleywiser could you check that it seems good to you? |
Co-authored-by: Felix S Klock II <pnkfelix@pnkfx.org>
Co-authored-by: Felix S Klock II <pnkfelix@pnkfx.org>
pietroalbini
changed the title
No description provided.