Re: reproducible memory problem and crashes on cygwin

[emacs user]

> From: Eli Zaretskii <address@hidden>
>
> > From: "emacs user" <address@hidden>
>
> Thank you for your report.
>
> > this problem leads to consistent and reproducible crashes of emacs.
>
> You didn't show any crashes.  Please do (with meaningful backtraces).

so emacs uses more and more memory, and eventually slows down my laptop so 
much that I either have to kill it (OK, not exactly a crash, I suppose), or 
that it somehow dies.  wouldnt this be sufficient information to the cygwin 
maintainer to work with?  even without the crash, this seems like a fairly 
significant bug...?

> > using latest cvs gnu emacs (22.0.50) on latest cygwin.
>
> Which version of Cygwin is that?  We had in the past reports about
> various problems in specific versions; please try to search the
> archives of this list and of emacs-devel, and see if your crashes
> resemble some reported in the past.

the cygwin home page suggests that  Latest Cygwin DLL release version is 
1.5.20-1
is this what you were looking for?  how do I find the cygwin version number? 
 I am trying to be well intentioned but am certainly a total ignorant as a 
non fortran programmer

I did follow the links you suggested in a previous email, but couldnt 
understand if they are relevant or not to this specific problem, or the 
previous one I reported.

> > start with
> > emacs -q --no-site-file&
> > edit a jpeg file
> > kill buffer
> > edit another jpeg file
> > kill buffer again.
> >
> > every such jpeg file edited and exited leaves the memory used by emacs 
> (as
> > seen in the list of processes of the windows task manager) increased.
>
> Please report the numbers: the memory usage figures reported by the
> Task Manager before, during, and after the image buffer is killed.

an example:

before editing the image: 33Mb
during: 39Mb
after killing the buffer: 39Mb
image file is 186263 bytes large (jpeg)

> > Memory does not decrease once buffer with image is killed.  when a text 
> file
> > is edited and exited, the memory does return to its original size.
>
> I don't think Emacs returns the memory to the OS immediately, but GC
> should cause that.

I'll take your word for that, I hardly know what GC is...

_________________________________________________________________
Don’t just search. Find. Check out the new MSN Search! 
http://search.msn.click-url.com/go/onm00200636ave/direct/01/