GNU bug report logs - #75209
30.0.93; Emacs reader failed to read data in "/home/nlj/.cache/org-persist/gc-lock.eld"

Previous Next

Full log

View this message in rfc822 format

From: "N. Jackson" <njackson <at> posteo.net>
To: Ihor Radchenko <yantar92 <at> posteo.net>
Cc: Eli Zaretskii <eliz <at> gnu.org>, 75209 <at> debbugs.gnu.org
Subject: bug#75209: 30.0.93; Emacs reader failed to read data in "/home/nlj/.cache/org-persist/gc-lock.eld"
Date: Thu, 24 Apr 2025 22:15:11 +0000

At 17:24 +0000 on Sunday 2025-04-20, Ihor Radchenko wrote:
>
> "N. Jackson" <njackson <at> posteo.net> writes:
>
>>> -  (let ((write-region-inhibit-fsync t)
>>> +  (let ((write-region-inhibit-fsync nil)
>>>          ;; We set UTF-8 here and in `org-persist--read-elisp-file'
>>>          ;; to avoid the overhead from `find-auto-coding'.
>>>          (coding-system-for-write 'emacs-internal)
>>
>> I will test this diff and report back.
>
> It has been a while since the last update in this thread.
> Does it mean that the diff fixed the problem for good?

I haven't seen the bug since installing the change quoted above that
let binds `write-region-inhibit-fsync' to nil.

Unfortunately, however, there is a confounding factor because around
the same time as I made this change, I switched from mainly using a
very slow desktop system to using a considerably faster laptop.


A slight tangent follows about the patch to do the atomic write
juggle for Windows systems.

At 12:55 +0000 on Saturday 2025-03-15, Ihor Radchenko wrote:
>
> diff --git a/lisp/org-persist.el b/lisp/org-persist.el
> index a639699d93..202c5e645b 100644
> --- a/lisp/org-persist.el
> +++ b/lisp/org-persist.el
> @@ -506,7 +506,12 @@ (defun org-persist--write-elisp-file
>              (let ((pp-use-max-width nil)) ; Emacs bug#58687
>                (pp data (current-buffer)))
>            (prin1 data (current-buffer))))
> -      (rename-file tmp-file file 'overwrite)
> +      (let ((tmp-file-2 (make-temp-file "org-persist-")))
> +        ;; Just renaming may still not be atomic on Windows, so we do
> +        ;; a bit more complex juggle.
> +        (rename-file file tmp-file-2)
> +        (rename-file tmp-file file)
> +        (delete-file tmp-file-2))
>        (org-persist--display-time
>         (- (float-time) start-time)
>         "Writing to %S" file))))

At 17:43 +0000 on Saturday 2025-03-15, Ihor Radchenko wrote:
>
> `make-temp-name' instead of `make-temp-file'.

After the fix using `make-temp-name' instead of `make-temp-file',
there are still two problems with the code:

Problem 1. `rename-file' produces an error if the file doesn't exist
-- which it doesn't the first time the code runs.

Problem 2. Unlike `make-temp-file', `make-temp-name' requires an
absolute file name otherwise the file ends up in the current
directory, whatever that happens to be at the time of the write.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.