--- Comment #5 from jakub at gcc dot gnu dot org 2009-05-06 12:14 ---
That just means you have no idea how a memory allocator works.
Returning all just freed pages immediately to the system is of course possible,
but terribly expensive performance wise, since often on the following mallo
--- Comment #4 from paolo dot carlini at oracle dot com 2009-05-06 12:07
---
Oh well, if valgrind is happy, definitely not a libstdc++ maintainers job.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40042
--- Comment #3 from simonracz at gmail dot com 2009-05-06 12:03 ---
(In reply to comment #2)
> By the way, valgrind, correctly, doesn't report anything, at least together
> with glibc 2.9:
You can see the difference in memory usage (after deleting) with top.
--
http://gcc.gnu.org/b
--- Comment #2 from paolo dot carlini at oracle dot com 2009-05-06 11:39
---
By the way, valgrind, correctly, doesn't report anything, at least together
with glibc 2.9:
==5250== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 3 from 1)
==5250== malloc/free: in use at exit: 0 bytes
--- Comment #1 from rguenth at gcc dot gnu dot org 2009-05-06 10:55 ---
This is a glibc "bug".
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
