Re: [PROBLEM] hard-lock with kmemtrace, relayfs, and splice

["Peter Teoh" <htmldeveloper@gmail.com>]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=GB2312, Size: 3806 bytes --]

ok.   just ignore it.   i think based on "git log" of"git://repo.or.cz/linux-2.6/kmemtrace.git":
commit 0441e5ff6ab71cf7a3e9ec3116f14d0fd7d20d51Author: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>Date:   Thu Jul 10 20:20:05 2008 +0300
    kmemtrace: SLOB hooks.
the repo seemed way out of date.   But a clone of Pekka's slab-2.6'stopic/kmemtrace branch have no problem.   The steps are as follows:
1.   git clone git://git.kernel.org/pub/scm/linux/kernel/git/penberg/slab-2.6.git2.   git checkout -m  origin/topic/kmemtrace3.   git checkout -b kmemtrace
git log gives:
commit 51b19be3535c8fbcce6b6f838d89b9a6a4cc5b92Author: Tom Zanussi <zanussi@comcast.net>Date:   Fri Oct 10 23:58:51 2008 -0500
    relayfs: fix infinite loop with splice()
But the userspace tool I still get it from: git://repo.or.cz/kmemtrace-user.git
Correct?
Now I have some problems:
a.   I would like to extract out all the commit as diff - may I knowhow to do that? ("git log" only gives the descriptive part).b.   boot-time memory profiling....how can it be done (or extractedout)?  (kmemtrace-user does not have that?)c.   please provide some pointers to documentation:   how do Iinterpret the following:
Allocation #83468 (CPU0) already exists!        by __kmalloc_track_caller+25        last touched by __kmalloc_track_callerAllocation #83740 (CPU0) already exists!        by __kmalloc_track_caller+25        last touched by __kmalloc_track_caller
For my purpose, I would like to trace how and where is memoryallocated - just one one single userspace program which access themmap("/dev/mem") to read the memory.   How can it be done?   "Tracingmemory" in my context will mean list all the kernel-function +memory-allocated-within-the-function + its-PFN-or-PTE-information (topinpoint the exact physical page).   My understanding is that thesecan be done, via checking the value of "current", as each userspaceprogram started will have a unique "current" context value, andtherefore within any kernel function, just check this value, to focustracing on just one particular userspace program.   Correct?
d.   The following:
/sys/kernel/debug/kmemtrace>ls -al×Ü¼Æ 0drwxr-xr-x 2 root root          0 2008-10-25 .drwxr-xr-x 9 root root          0 2008-10-25 ..-r-------- 1 root root          0 2008-10-25 abi_version-r-------- 1 root root 2946936192 2008-10-25 cpu0-r-------- 1 root root 1701263952 2008-10-25 cpu1-rw------- 1 root root          0 2008-10-25 enabled-r-------- 1 root root          0 2008-10-25 total_overruns
So the information is stored in memory, right?   Is it possible toreset it?   I don't want these information to clog up the memory.
e.   The balance between cpu0 and cpu1 does not seemed equal, anyimplications on the scheduler/ IO Scheduler?   or memory allocationscheduler (if there exists one :-))....sorry about that....myknowledge is limited.
On Fri, Oct 24, 2008 at 10:15 PM, Pekka Enberg <penberg@cs.helsinki.fi> wrote:> On Fri, 2008-10-24 at 12:44 +0800, Peter Teoh wrote:>> after doing a>>>> git clone git://repo.or.cz/linux-2.6/kmemtrace.git>>>> I reboot the OS and encountered several application crashes.....the>> logs are as per attached....please comment.>>>> for bug0 and bug1, the system go into a state of complete>> non-responsive, even connecting via SSH into the system becomes>> impossible, and bug0 and bug1 was generated just before it goes into>> this state.>>>> for bug2, the mouse seemed to respond, i managed to output the dmesg>> (which is bug2 itself).>>>> any comments?>> Hmm, you're hitting out-of-memory. I can't immediately see how kmemtrace> is at fault here. Eduard, thoughts?>>


-- Regards,Peter Teohÿôèº{.nÇ+‰·Ÿ®‰­†+%ŠËÿ±éݶ\x17¥Šwÿº{.nÇ+‰·¥Š{±þG«éÿŠ{ayº\x1dʇڙë,j\a­¢f£¢·hšïêÿ‘êçz_è®\x03(­éšŽŠÝ¢j"ú\x1a¶^[m§ÿÿ¾\a«þG«éÿ¢¸?™¨è­Ú&£ø§~á¶iO•æ¬z·švØ^\x14\x04\x1a¶^[m§ÿÿÃ\fÿ¶ìÿ¢¸?–I¥