David,
On 12/16/2012 02:44 PM, David Kastrup wrote:
Tassilo Horn <[email protected]> writes:
[...]
That's a really strange problem. It doesn't sound like an Emacs or
AUCTeX problem but more like a colleague playing a trick on you.
More like the computer playing a trick on him. i (0x69) and m (0x6d)
differ by a single bit, bit 2. Time to run some memory checker.
I ran memtester for about an hour, without any problem.
The most reliable memory check I know is compiling the Linux kernel. I
had computers running a standard memory checker for a whole night, but
not surviving an hour of kernel compilation (gcc bombs out with
segmentation fault astonishingly reliably with flaky memory). Of
course, keeping today's computers busy for hours is no longer feasible
with a single kernel compile, but you might put it in a loop.
I'll try to do that and keep you posted.
I'm a copy-editor/typesetter, that is, the books I work on are not my
own, so you and Tassilo can understand why this is so important to me
---I simply can't introduce typos on other people's work.
Thank you very much.
Axel
_______________________________________________
auctex mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/auctex
