GNU bug report logs - #62009
29.0.60; Emacs crashes on setf symbol-name

Previous Next

Full log

View this message in rfc822 format

From: Eli Zaretskii <eliz <at> gnu.org>
To: Gregory Heytings <gregory <at> heytings.org>
Cc: philipk <at> posteo.net, michael_heerdegen <at> web.de, mail <at> daniel-mendler.de, monnier <at> iro.umontreal.ca, 62009 <at> debbugs.gnu.org, arstoffel <at> gmail.com
Subject: bug#62009: 29.0.60; Emacs crashes on setf symbol-name
Date: Fri, 10 Mar 2023 13:59:28 +0200

> Date: Fri, 10 Mar 2023 10:59:03 +0000
> From: Gregory Heytings <gregory <at> heytings.org>
> cc: Eli Zaretskii <eliz <at> gnu.org>, Philip Kaludercic <philipk <at> posteo.net>, 
>     michael_heerdegen <at> web.de, monnier <at> iro.umontreal.ca, 62009 <at> debbugs.gnu.org, 
>     Augusto Stoffel <arstoffel <at> gmail.com>
> 
> > Creating a string is not a good idea since it will lead to an 
> > unacceptably large performance overhead.
> 
> Is "symbol-name" a function that is used in performance-critical code? 

Yes, it is.  Just grep for it.  We even call it from C quite a few
times.  And processing is just one aspect of that; memory and GC is
another, not less important.

> And did you actually measure that performance overhead before concluding 
> that it it "unacceptably large"?

Anything greater than zero is unacceptably large from where I stand,
when the other side of the balance is the use case against which this
protects.

> > Raising an exception upon modification would be the best approach.
> 
> That would also come with a performance overhead, as there is currently no 
> way to distinguist strings that are used for symbol names from other 
> strings.  Not to mention the added complexity in the code.

Which is why we should do neither.

Previous Next

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