[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [AUCTeX] Text corruption

From: David Kastrup
Subject: Re: [AUCTeX] Text corruption
Date: Mon, 17 Dec 2012 13:11:56 +0100
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3.50 (gnu/linux)

Tassilo Horn <address@hidden> writes:

> "Axel E. Retif" <address@hidden> writes:
> Hi Axel,
>>>>    GNU Emacs 24.2.1 (x86_64-pc-linux-gnu, GTK+ Version 2.24.10)
>>>>     of 2012-11-08 on lakoocha, modified by Debian
>>> by Robert Browning, from Damien Cassou's ppa. I will install AUCTeX
>>> on it and use this exclusively, and see what happens.
>> Well... this is rather unfortunate!
>> With this new set up Emacs, right in front my eyes, capitalized a word
>> a few lines below where I was typing some text (undoing the typing one
>> step at a time suddenly un-capitalized the word),
> That you can undo the change speaks against memory corruption, IMO.  And
> that it happes only with Emacs, and then only in AUCTeX buffers, also
> seems to imply some other problem.
>> and then it changed the name of a figure a few lines above where I was
>> typing... all this in less than half an hour!
>>  I will start using TeXmaker --- if this is an issue of memory
>> corruption, it should happen sooner or later, right?
> If it was some memory corruption, I'd expect that you encounter strange
> problems and crashes with more important applications, e.g., crashes of
> X, kernel oops, etc.

Not really.  Most of the run-of-the-mill applications don't react all
that much to the occasionally flipping bit.

I had once debugged a problem for hours in a program of mine until
finding the problem.  When doing arithmetic right shifts of negative
numbers, the sign randomly became positive.  Exchanged the processor,
problem gone.  Apart from causing this one program to malfunction, there
was no previous sighting of any problems, likely for years.

Similar with the systems that were not able to finish several hours of
kernel compilation.  No other signs of problems specific enough to
warrant attention.

David Kastrup

reply via email to

[Prev in Thread] Current Thread [Next in Thread]