This is the mail archive of the glibc-bugs@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug libc/4349] _int_malloc extremely slow with ordblks free chunks


------- Additional Comments From tomash dot brechko at gmail dot com  2007-04-19 10:13 -------
(In reply to comment #0)
> If this is a known limitation, is there a good work around?

Based on your description it seems you hit the problem addressed by the named patch.

You may download ptmalloc from malloc.de and link with it.  Basically, malloc in
glibc 2.4 (and in 2.5 :-/) is based on ptmalloc2.  You could use
ptmalloc2+patch, but better to just use ptmalloc3.  While the patch implements
naive skip lists to speed up the search of the chunk of the right size,
ptmalloc3 comes with bitwise digital trees (aka tries), which is a much better
solution for the same problem.

> Searching on the web, this problem might be related to the issue Tomash Brechko
> had a patch for in Dec 2004
> (http://sourceware.org/ml/libc-alpha/2004-12/msg00041.html) His patch is
> apparently not included in the glibc 2.4 tree.  Would this patch been eventually
> considered?

Probably no.  As ptmalloc3 is the better solution, we have but to wait until it
will be adopted to glibc.  I guess binary incompatibility is the main obstacle
for that.  Until then, link with ptmalloc3 yourself.

  Tomash

-- 


http://sourceware.org/bugzilla/show_bug.cgi?id=4349

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]