On Jan 21, 11:50 pm, Dominique Pellé <[email protected]>
wrote:
>
> And graph after patch is here:
>
> http://dominique.pelle.free.fr/time-readfile.png

Thank you, an interesting tutorial.  Next time I want to graph
something I'll be searching for your post rather than firing up a
spreadsheet.

I can't help thinking that your linear times are not guaranteed with
the vagaries of heap fragmentation and memory allocator
implementation, and that calling into memory allocator code every 200
bytes of megabytes of data is to be avoided.  Such intuitions are
infamous, I suppose.

I said:

> > The old code's handling of allocation failures is haphazard, or at
> > least I can't see a clear outcome if an allocation fails.

> To test memory alloc failures, on Linux, you can limit the
> the amount of virtual memory of the shell & its subprocesses...
> with: ulimit -v <kbytes>.

Thank you.

Dominique said:
> ... in my opinion,
> when running out of memory, it's most of the time
> hopeless to try to recover, except in few critical places.

I agree, but that intuition says that if someone tries to do something
really over the top with memory and it fails, we could give up on the
big thing reliably.  I'm thinking of the user on a limited platform,
like DOS or even an iphone.

Regards, John

-- 
You received this message from the "vim_dev" maillist.
Do not top-post! Type your reply below the text you are replying to.
For more information, visit http://www.vim.org/maillist.php

Raspunde prin e-mail lui