Date: Fri, 26 Jan 2001 19:36:15 +0100 From: Xavier Galleri <xgalleri@enition.com> To: Alfred Perlstein <bright@wintelcom.net> Cc: Peter Wemm <peter@netplex.com.au>, freebsd-hackers@FreeBSD.ORG Subject: Re: Kernel memory allocation bug ... Message-ID: <3A71C39F.8060109@enition.com> References: <200101191524.f0JFOtk14877@mobile.wemm.org> <3A6861DF.8060108@enition.com> <3A706788.5040509@enition.com> <20010125101659.N26076@fw.wintelcom.net> <3A707211.2080801@enition.com> <20010125114602.Q26076@fw.wintelcom.net> <3A7151D5.2040405@enition.com>
index | next in thread | previous in thread | raw e-mail
[-- Attachment #1 --]
Xavier Galleri wrote:
[...]
>> You've been complaining about this problem for at least a week.
>> Producing some code so that we can test couldn't take more than a
>> couple of hours and would have probably had your issue resolved
>> by now.
>>
>> This is why I'm irritated that you still haven't provided any code
>> to reproduce it.
>
[...]
> Well, actually, it sounds that I did go to far to be able to retract
> myself, so I will let you know of any progress I will make asap ;-)
Ok, I'm back again ;-) and this time, I got something for you ...
Basically, what I did, is to write a little SYSCALL kld module to allow
controlling calls to MALLOC/FREE from user space. This way, I can issue
some filesystem-intensive command (tar
something-bigger-than-memory-size) in order to create the free list
shortage. Then, I use my test program to issue several MALLOC in kernel
space. What I actually noticed is that, when the free list felt below
120 or alike, I can issue several thousands successful MALLOC as long as
the asked size is no more than one or a few pages. But, when I ask for
numerous pages (32Kb), then the process is falling asleep (this is
analysed with my manual stack analysis process ;-), whilst the M_NOWAIT
flag has been set (of course !).
Everything is almost entirely self-documented in the attached
kmem.tar.gz file. I hope this will help and I am eager to hear from you
soon.
Have a nice WE,
Regards
Xavier
[-- Attachment #2 --]
Gq:kmem.tar kSHWMXae#lq
m%KFFgKI߯{fdH.>
gRz!w֞^VaͰF"~+KV2튉φaZ5=5L3'X:>K'?#unw
Jos?g,ovr͊>e5ET@_/E4|z:+ȯ9l.DKF2(p,nGp
2(S
@%C*mt]NV$
e_"J8KWDmګ)J"f)#Q#)}PzO8L̏+352N" LciF?|,NL(!|c%DUa$NhsC&3ߣDb{8{$[<z%v6Ҡ1wM}OnX0bB<KbM"RE6:GKCwnq[w4dʸz_qEs8(̛+AIzd5p))27wVgFV&{/h/=⦴!t989lQKqĂf(O olI& 7y>pVc3gɸzQ|.!]xrHZMyWPj1?gUVYHv!O\F}8<yGgeuaMѶcG9Q{8WJ?òWgS/Ɯ-_gM]姓CP~:k~w^t|H]'ǷN{xn*+Ƿמ-}Ut$?x
"czI=lstv7[UQփˤ1pE~ϟEnj/ڧo$k6EluGq:U]ygۤ?g/ }ֻP3fQ$bY<
Cd(edSzkvr\~*7omYLgCJH"n1ϒ[0@:o:3/Ŋ|ax0N%Z^gU5-]^o_ފRKO>_zn˼bM<GO]^s%bnHNr wʼnӠ>>:m۳ENJGe]pZG?),ѕRu|Џ/:?
ySkBZ5`Q-iB_\[Rj'o.?=6l/_y+yq Y0OA4}vIT9ʣ6m&!4,D1g5jp[!9=;iܫ:~@$&*wS6eUø?H $ų`*yLwR)l2Ճqqypк(XW)7!.'f}K
~0W
=7_ci(*mTrK05NԅuҠ0q#ߨ y3_G:"Il4ychxZiKIdZIK!wskSB}eW?vI%r-`2_OI5Iq=9;l>dS.N
I
R_\x_o
R.m$%+ATZ1=OZOäCotUY0l3fi
Y
o2DGBRpR3ݤb`t|3 ŦP?Hayx&@ o
`o[US&qhiXgxqS+(
7?6B?|(
Ge ǧ^AXɡ$YbP"FA@"_ChUb:P6YQ`o˺>}%]ӱe| &P}tW;cHbQw.CIvqwGR
X:`nKr6( q})G,{V.5g++E1'!gܠ[
]e_}ոJn]ݯ1B_6A SD2E5QYUyjVcF&sUxF3t(1j9S\+mw6n{ĉ%\ém̭XN5Cq<q1eʤMDO$#3)4FSv#0DR}SSLEcf#מ]+d}mW];>|HȁKLQ^@`<0l"hYWr^3ឆ8OGɤc&fQ $*8j?C@#!\.&%qQ8$\89p$[8}' s32wq'o;cLQlH8lvl6!ŐN<sxPie _sLxT^'{ML]\/p9%o
<V;txozTmԀ`nJ. 8!i@jMQzFAF
>u"qIн0Rh4m:(=~F?Dص^w0wӨ|>d9I/ģ(c!~DS}T'pMY2bX)
ԴvXZve^n=WeL`QH\ ڴs[R]:*ߚbnt<4m3/@_^
[72($59n3縷7
[7*%DR㖮=KuF+$iW(uv\/#֫uuQx|
1A֩*
EdaIYҭzq2ѷ8<Q$&\g2͝-zXBU"+->8E?ĩ(d ɸm9Ygl Clr$AKhb`^}ˀQdCp7R
trmSl6iqGa;_l0ƘLبZyPiVZ
wSqϐ=
|[+0|orvŶ]Π>̽9h[fDN6@mlEqQ7tnG%s#{\B5HX;ݴi,J"GR,-pXv?R_jvY^S:mro#I.HJ"S-{9Az,ʳ{Bh?bPGs:vu2T4iL(M(q͏铄
>³W0ʗhkU3̺&ra(!|DvôkvՠQ)
Q4ήnR5?gC0<V8(&$HAB))ykW#̥Y&jCnHeS;pUō(!0L[T%pzE8Ty ۺޞPQ=B^wH0; tR*MĘ2@$¢QQHa?(H('wxn]ȫcQ@eYgY]kuյVZ]ku?G@ P
help
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3A71C39F.8060109>
