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 network/19257] Per-thread memory leak in __res_vinit with IPv6 nameservers


https://sourceware.org/bugzilla/show_bug.cgi?id=19257

--- Comment #2 from mschiffer+misc@universe-factory.net ---
Created attachment 8808
  --> https://sourceware.org/bugzilla/attachment.cgi?id=8808&action=edit
Reproducer

While writing the reproducer, I've noticed that there is an additional
constraint for this bug to happen: The hostname must not actually need the
resolver - in this case, because the requested hostname "localhost" can be
found in /etc/hosts. As soon as a single request is actually sent to the
nameserver, the bug disappears, but as long as only requests are made that can
be resolved from /etc/hosts, the memory is leaked.

When the request is made in the main thread, valgrind reports "still reachable:
28 bytes in 1 blocks", when it is moved to another thread, it will become
"definitely lost".

-- 
You are receiving this mail because:
You are on the CC list for the bug.

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