Your message dated Sun, 8 Jan 2017 22:17:28 +0100
with message-id <20170108211726.v5fjjwvvgjfg6...@breakpoint.cc>
and subject line Re: blocks listed by valgrind
has caused the Debian Bug report #351684,
regarding libssl0.9.8: Please free remaining memory blocks
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact ow...@bugs.debian.org
immediately.)
--
351684: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=351684
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems
--- Begin Message ---
Package: libcurl3
Version: 7.15.1-1
Severity: normal
Tags: upstream
curl gives me valgrind warnings, even using something like
% valgrind --show-reachable=yes --leak-check=full curl google.com
==16911== at 0x4010FA0: (within /lib/ld-2.3.5.so)
==16911== by 0x40049C1: (within /lib/ld-2.3.5.so)
==16911== by 0x400651A: (within /lib/ld-2.3.5.so)
==16911== by 0x43F8A5F: (within /lib/tls/libc-2.3.5.so)
==16911== by 0x400B026: (within /lib/ld-2.3.5.so)
==16911== by 0x43F941A: _dl_open (in /lib/tls/libc-2.3.5.so)
==16911== by 0x43FA92C: (within /lib/tls/libc-2.3.5.so)
==16911== by 0x400B026: (within /lib/ld-2.3.5.so)
==16911== by 0x43FAA8D: __libc_dlopen_mode (in /lib/tls/libc-2.3.5.so)
==16911== by 0x43D55CD: __nss_lookup_function (in /lib/tls/libc-2.3.5.so)
==16911== by 0x43AB3F6: (within /lib/tls/libc-2.3.5.so)
==16911== by 0x43ABE15: getaddrinfo (in /lib/tls/libc-2.3.5.so)
==16911== Address 0x4456770 is 24 bytes inside a block of size 27 alloc'd
==16911== at 0x401B41A: malloc (vg_replace_malloc.c:149)
==16911== by 0x4006722: (within /lib/ld-2.3.5.so)
==16911== by 0x43F8A5F: (within /lib/tls/libc-2.3.5.so)
==16911== by 0x400B026: (within /lib/ld-2.3.5.so)
==16911== by 0x43F941A: _dl_open (in /lib/tls/libc-2.3.5.so)
==16911== by 0x43FA92C: (within /lib/tls/libc-2.3.5.so)
==16911== by 0x400B026: (within /lib/ld-2.3.5.so)
==16911== by 0x43FAA8D: __libc_dlopen_mode (in /lib/tls/libc-2.3.5.so)
==16911== by 0x43D55CD: __nss_lookup_function (in /lib/tls/libc-2.3.5.so)
==16911== by 0x43AB3F6: (within /lib/tls/libc-2.3.5.so)
==16911== by 0x43ABE15: getaddrinfo (in /lib/tls/libc-2.3.5.so)
==16911== by 0x41276E0: (within /usr/lib/libcurl.so.3.0.0)
==16911== Conditional jump or move depends on uninitialised value(s)
==16911== at 0x4008B13: (within /lib/ld-2.3.5.so)
==16911== by 0x43F8D69: (within /lib/tls/libc-2.3.5.so)
==16911== by 0x400B026: (within /lib/ld-2.3.5.so)
==16911== by 0x43F941A: _dl_open (in /lib/tls/libc-2.3.5.so)
==16911== by 0x43FA92C: (within /lib/tls/libc-2.3.5.so)
==16911== by 0x400B026: (within /lib/ld-2.3.5.so)
==16911== by 0x43FAA8D: __libc_dlopen_mode (in /lib/tls/libc-2.3.5.so)
==16911== by 0x43D55CD: __nss_lookup_function (in /lib/tls/libc-2.3.5.so)
==16911== by 0x43AB3F6: (within /lib/tls/libc-2.3.5.so)
==16911== by 0x43ABE15: getaddrinfo (in /lib/tls/libc-2.3.5.so)
==16911== by 0x41276E0: (within /usr/lib/libcurl.so.3.0.0)
==16911== by 0x4105A03: (within /usr/lib/libcurl.so.3.0.0)
==16911== Conditional jump or move depends on uninitialised value(s)
==16911== at 0x40087D3: (within /lib/ld-2.3.5.so)
==16911== by 0x43F8D69: (within /lib/tls/libc-2.3.5.so)
==16911== by 0x400B026: (within /lib/ld-2.3.5.so)
==16911== by 0x43F941A: _dl_open (in /lib/tls/libc-2.3.5.so)
==16911== by 0x43FA92C: (within /lib/tls/libc-2.3.5.so)
==16911== by 0x400B026: (within /lib/ld-2.3.5.so)
==16911== by 0x43FAA8D: __libc_dlopen_mode (in /lib/tls/libc-2.3.5.so)
==16911== by 0x43D55CD: __nss_lookup_function (in /lib/tls/libc-2.3.5.so)
==16911== by 0x43AB3F6: (within /lib/tls/libc-2.3.5.so)
==16911== by 0x43ABE15: getaddrinfo (in /lib/tls/libc-2.3.5.so)
==16911== by 0x41276E0: (within /usr/lib/libcurl.so.3.0.0)
==16911== by 0x4105A03: (within /usr/lib/libcurl.so.3.0.0)
<HTML><HEAD><meta http-equiv="content-type"
content="text/html;charset=us-ascii">
...
curl_global_init is also not freeing some memory used indirectly via
libcrypto:
==16917== 756 bytes in 11 blocks are still reachable in loss record 1 of 1
==16917== at 0x401B41A: malloc (vg_replace_malloc.c:149)
==16917== by 0x41DA0BD: (within /usr/lib/i686/cmov/libcrypto.so.0.9.8)
==16917== by 0x41DA738: CRYPTO_malloc (in
/usr/lib/i686/cmov/libcrypto.so.0.9.8)
==16917== by 0x4231628: ENGINE_new (in /usr/lib/i686/cmov/libcrypto.so.0.9.8)
==16917== by 0x4235CDB: ENGINE_load_dynamic (in
/usr/lib/i686/cmov/libcrypto.so.0.9.8)
==16917== by 0x42341F6: ENGINE_load_builtin_engines (in
/usr/lib/i686/cmov/libcrypto.so.0.9.8)
==16917== by 0x4118D25: (within /usr/lib/libcurl.so.3.0.0)
==16917== by 0x41285EB: (within /usr/lib/libcurl.so.3.0.0)
==16917== by 0x4121CA0: curl_global_init (in /usr/lib/libcurl.so.3.0.0)
==16917== by 0x804D124: (within /usr/bin/curl)
==16917== by 0x4307ECF: __libc_start_main (in /lib/tls/libc-2.3.5.so)
==16917==
--- End Message ---
--- Begin Message ---
On 2006-02-16 23:20:12 [-0500], Justin Pryzby wrote:
> ==11395== 756 bytes in 11 blocks are still reachable in loss record 1 of 1
> ==11395== at 0x401B422: malloc (vg_replace_malloc.c:149)
> ==11395== by 0x434C0BD: ??? (mem.c:79)
> ==11395== by 0x434C738: CRYPTO_malloc (mem.c:304)
> ==11395== by 0x43A3628: ENGINE_new (eng_lib.c:68)
> ==11395== by 0x43A7CDB: ENGINE_load_dynamic (eng_dyn.c:270)
> ==11395== by 0x43A61F6: ENGINE_load_builtin_engines (eng_all.c:71)
> ==11395== by 0x4081D25: ??? (ssluse.c:548)
> ==11395== by 0x40915EB: ??? (sslgen.c:167)
> ==11395== by 0x408ACA0: curl_global_init (easy.c:204)
> ==11395== by 0x804C2D6: usno_query (usnoquery.c:359)
> ==11395== by 0x804941D: main (usnoquery_main.c:81)
>
> ENGINE_cleanup is not being called, as per the curl #ifdef, but
> forcing it to be called doesn't actually fully solve the problem (nor
> am I sure that it should). So there does seem to be some mem that
> isn't freed by ENGINE_cleanup.
curl is no longer linked against openssl for the main binary. Also with
the 1.1.0 rewrite it might been fixed.
If this still exists, please open a new bug.
Closing.
Sebastian
--- End Message ---
