bug#68083: 30.0.50; Intermittent build failure with native compilation

[Aaron Jensen <aaronjensen@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1561 bytes --]

 Works for me, though is it still an outstanding question as to whether or
not we should add macro expand back?

Also yeah if we add the t we can do what you suggested where it’s either a
delete or rename force.


Aaron

On Sat, Dec 30 2023 at 1:46 PM, Eli Zaretskii <eliz@gnu.org> wrote:

> From: Aaron Jensen <aaronjensen@gmail.com>
> Date: Sat, 30 Dec 2023 13:06:55 -0500
> Cc: acorallo@gnu.org, 68083@debbugs.gnu.org, jschmidt4gnu@vodafonemail.de
>
> On Sat, Dec 30, 2023 at 12:43 PM Eli Zaretskii <eliz@gnu.org> wrote:
> >
> > What happens if, instead of wrapping rename-file in condition-case,
> > you change that to say
> >
> > (t (if newfile
> > (rename-file newfile oldfile)
> > (delete-file oldfile))
>
> I don't really understand this change. The previous version of the code
> wraps the rename in a (when newfile) and deletes the oldfile right before
> that. The deletion would always be necessary if the oldfile exists unless
> OK-IF-ALREADY-EXISTS is specified in rename-file.
>
> Sorry, I meant to add the OK-IF-ALREADY-EXISTS argument non-nil, of
> course.
>
> The point is that you can rename a file if the OLDFILE exists with no
> problem, and that is supposed to be an atomic operation, so no race
> conditions.
>
> Indeed, but we aren't specifying OK-IF-ALREADY-EXISTS, so if another
> process does the same compile at the same time (and therefore the same
> rename), they will conflict. We could specify OK-IF-ALREADY-EXISTS instead
> and that works as well:
>
> Then I prefer this version.
>

[-- Attachment #2: Type: text/html, Size: 2522 bytes --]