Dear Davide,
This is the first test with valgrind (15 minutes running) !
Do you think my setup of it is correct and how many time I have to
keep running before to read interesting result ?
==30326== discard syms in /lib/libnss_files-2.3.2.so due to munmap()
==30326== discard syms in /lib/libnss_dns-2.3.2.so due to munmap()
==30326== discard syms in /lib/libresolv-2.3.2.so due to munmap()
==30326==
==30326== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 10 from 1)
--30326--
--30326-- supp: 10 __pthread_mutex_unlock/_IO_funlockfile
==30326== malloc/free: in use at exit: 6808 bytes in 233 blocks.
==30326== malloc/free: 959148 allocs, 958915 frees, 77707206 bytes allocated.
==30326==
==30326== searching for pointers to 233 not-freed blocks.
==30326== checked 4873372 bytes.
==30326==
==30326== 200 bytes in 1 blocks are definitely lost in loss record 1 of 4
==30326== at 0x40218D6C: my_malloc (vg_libpthread.c:263)
==30326== by 0x4021A6FF: get_or_allocate_specifics_ptr (vg_libpthread.c:1405)
==30326== by 0x4021A818: __pthread_key_create (vg_libpthread.c:1442)
==30326== by 0x807A374: (within /var/MailRoot/bin/XMail)
==30326== by 0x8052470: (within /var/MailRoot/bin/XMail)
==30326== by 0x4021AB04: __pthread_once (vg_libpthread.c:1551)
==30326== by 0x805248D: (within /var/MailRoot/bin/XMail)
==30326== by 0x8052557: (within /var/MailRoot/bin/XMail)
==30326==
==30326==
==30326== 392 bytes in 14 blocks are still reachable in loss record 2 of 4
==30326== at 0x400218C6: malloc (vg_replace_malloc.c:153)
==30326== by 0x42F9AF4B: ???
==30326== by 0x42F99C06: ???
==30326== by 0x42F9A35A: ???
==30326== by 0x42F99F21: ???
==30326== by 0x42F8E05E: ???
==30326== by 0x42F8E292: ???
==30326== by 0x40414A77: gethostbyname_r@@GLIBC_2.1.2 (in /lib/libc-2.3.2.so)
==30326==
==30326==
==30326== 2928 bytes in 37 blocks are definitely lost in loss record 3 of 4
==30326== at 0x400218C6: malloc (vg_replace_malloc.c:153)
==30326== by 0x807A414: (within /var/MailRoot/bin/XMail)
==30326== by 0x807358C: (within /var/MailRoot/bin/XMail)
==30326== by 0x8055562: (within /var/MailRoot/bin/XMail)
==30326== by 0x805621B: (within /var/MailRoot/bin/XMail)
==30326== by 0x80809BE: (within /var/MailRoot/bin/XMail)
==30326== by 0x80809EE: (within /var/MailRoot/bin/XMail)
==30326== by 0x40332916: __libc_start_main (in /lib/libc-2.3.2.so)
==30326==
==30326==
==30326== 3288 bytes in 181 blocks are still reachable in loss record 4 of 4
==30326== at 0x400218C6: malloc (vg_replace_malloc.c:153)
==30326== by 0x807A414: (within /var/MailRoot/bin/XMail)
==30326== by 0x80524B6: (within /var/MailRoot/bin/XMail)
==30326== by 0x8052557: (within /var/MailRoot/bin/XMail)
==30326== by 0x80628E2: (within /var/MailRoot/bin/XMail)
==30326== by 0x8062B14: (within /var/MailRoot/bin/XMail)
==30326== by 0x8062C58: (within /var/MailRoot/bin/XMail)
==30326== by 0x8062DA6: (within /var/MailRoot/bin/XMail)
==30326==
==30326== LEAK SUMMARY:
==30326== definitely lost: 3128 bytes in 38 blocks.
==30326== possibly lost: 0 bytes in 0 blocks.
==30326== still reachable: 3680 bytes in 195 blocks.
==30326== suppressed: 0 bytes in 0 blocks.
==30326==
--30326-- TT/TC: 0 tc sectors discarded.
--30326-- 11015 chainings, 0 unchainings.
--30326-- translate: new 13157 (204203 -> 1695560; ratio 83:10)
--30326-- discard 713 (11322 -> 79824; ratio 70:10).
--30326-- dispatch: 319950000 jumps (bb entries), of which 47874658 (14%) were
unchained.
--30326-- 674911/8468687 major/minor sched events. 93572 tt_fast misses.
--30326-- reg-alloc: 25 t-req-spill, 322215+76 orig+spill uis, 32137 total-reg-r.
--30326-- sanity: 558464 cheap, 22339 expensive checks.
--30326-- ccalls: 65280 C calls, 60% saves+restores avoided (232540 bytes)
--30326-- 65298 args, avg 0.63 setup instrs each (47598 bytes)
--30326-- 0% clear the stack (195840 bytes)
--30326-- 0 retvals, 100% of reg-reg movs avoided (0 bytes)
samedi 27 d�cembre 2003 at 21:05:54, you said :
Davide> On Sat, 27 Dec 2003, Pascal de R. wrote:
>>
>> vendredi 26 d�cembre 2003 at 22:45:34, you said :
>>
>> Davide> Are you able to run valgrind on it?
>>
>> I didn't know this product!
>>
>> Davide> http://valgrind.kde.org/
>>
>> Davide> Basically you have to build XMail with -O0 -g and remove the strip
>> Davide> commands from Makefile.lnx. Then :
>>
>> Davide> # make -f Makefile.lnx
>> Davide> # export MAIL_ROOT=...
>> Davide> # export MAIL_CMD_LINE=...
>> Davide> # valgrind -v ./XMail -Md
>>
>>
>> Thank you for your suggestion but do you think I can do that on the
>> production machine with any risk ?
Davide> There's no risk but a little perf penalty. You should use a debug version
Davide> of XMail (-O0 -g) and you should remove the "strip" lines from
Davide> Makefile.lnx. This is what you have to run:
Davide> # export MAIL_ROOT=...
Davide> # export MAIL_CMD_LINE=...
Davide> # valgrind -v --skin=addrcheck --num-callers=8
Davide> --leak-check=yes --show-reachable=yes ./XMail -Md
Davide> The "addrcheck" skin should give you same results for what we are
Davide> searching for, w/out huge performance degradation.
Davide> - Davide
Davide> -
Davide> To unsubscribe from this list: send the line "unsubscribe xmail" in
Davide> the body of a message to [EMAIL PROTECTED]
Davide> For general help: send the line "help" in the body of a message to
Davide> [EMAIL PROTECTED]
Best regards,
Pascal, [EMAIL PROTECTED]
-
To unsubscribe from this list: send the line "unsubscribe xmail" in
the body of a message to [EMAIL PROTECTED]
For general help: send the line "help" in the body of a message to
[EMAIL PROTECTED]
