| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8
Comments:
<0> and cout p
<1> oh, no this is not going to work
<1> you can't just pick a number and read/write it
<1> unless you are writing a device driver or running DOS
<1> otherwise there is no reason to do that
<0> Why not using a loop and dump secventially ?
<0> like before with pointers
<2> because you can't just read from any old memory
<2> this isn't 1986 ;)
<2> you should look for some OS utility to produce a dump of memory for the process
<2> is there something in particular that you'd like to do?
<2> you could always attach a debugger
<0> or some library....any links ?
<2> just google
<1> not to mention that you have no way of knowing where your pgoram really resides in RAM
<1> because of VM
<2> that's a lesser issue than knowing what parts of the address space are actually meaningful
<0> VM ... ?
<1> the addresses your program references are really only meaningful within your program's virtual address space
<1> VM = virtual memory
<2> yeah but cn28h, he would be addressing it from within his process
<2> it would really be helpful if you told us what you're really trying to do, nkr
<1> right, but he won't be able to reference memory not in this virtual space because the page table won't have entries corresponding to other memory. So basically he's locked inside his on process
<2> and why something like a debugger is useless
<0> i can do a overlow
<0> will that work
<0> After the overflow I will in the windows core address space ...I think
<2> cn28h what other memory would he be interested in?
<2> or is he trying to dump physical memory?
<1> I was under the impression he just wanted to dump the physical RAM
<1> like pick some spot, and check what's there
<1> and even write to it
<2> oh, well that's another loaf of bread entirely - you'd have to get code running in ring0 to help you there
<2> again, ask the SO
<2> OS
<0> I'll try...thx
<2> nkr WHAT are you trying to do???
<1> you might be able to get a physical memory dump if you cause a kernel panic :)
<2> do you really want a dump of your physical memory? and -why- ?
<2> ?
<2> nkr ?
<3> Solamente: You around?
<4> http://www.usatoday.com/news/washington/2006-04-02-federal-spending_x.htm
<5> hi
<5> i'm trying to add some MFC code with my non-mfc project and get "include 'stdafx.h' before including this file for PCH" and if i include afxwin.h and other MFC related includes i get: windows.h must not be included blablabla, can anyone help please ?
<6> hi,it is possible to run cl*** member function as a thread ?
<7> eh?
<8> good morning
<6> i wants to call CreateThread with function address of cl*** member function
<6> and msvc doing me problems
<7> you want a member function to runin a thread?
<9> dr3f - Not directly, in cl***ic C++. A traditional method is to create a static member function which takes an instance pointer as the thread parameter
<9> then invoke the member on that instance
<6> "None of the functions with this name in scope match the target type"
<6> asriel,you know any tutorial/example of that ?
<9> well
<9> something like
<7> its not particularly difficult if you make the thread function static
<9> cl*** Foo { public: static void ThreadEntryPoint(void *fooInstance) { (reinterpret_cast<Foo *>(fooInstance))->SomeMember(); } void SomeMember() {} };
<6> after that its possible to call CreateThread ?
<6> with that function ?
<9> on Foo::ThreadEntryPoint, yes
<6> and that function will have access to the cl*** varibles/other functions ?
<9> you can't call CreateThread on a member function, because all member functions have an implicit p***ed "this" pointer, which createthread can't provide
<6> yes
<6> :(
<6> but i wants to run stuff in threads
<9> No, it won't have access to the instance members - which is why you p*** in the instance you want to invoke the member function on
<9> and the static function invokes that
<9> most (all?) threading libraries allow you to p*** at least one void* into a thread entrypoint
<7> you can create a base cl*** and dervive from it to provide the threading functionality in the derived cl***es
<6> yes
<6> hmmm
<6> lets say i defined
<6> CmyCl*** xcl***;
<6> xcl***=xvoid;
<6> after that i shuld have abilty to change stuff in the xcl*** ?
<9> erm
<7> umm
<9> I'm not sure what you mean by that. Either way, I'm going out for a bit
<7> i made a little Thread clss on noidea...have a look at that (might be some errors but it will give you some ideas)
<6> thanks
<7> no synchronisation or anything
<6> nothing
<6> i have a simple thread
<7> thats an exercise for you
<10> What window message is sent when user resizes a window? I would've guessed WM_RESIZE but doesnt exist here
<7> does google "not exist" there
<11> bealtine: google-over-irc
<7> and windows messages have nothing whatsoever to do with c++
<7> hint drop the RE
<12> Off to office
<13> Hey, can someone explain to me how the use of functions like atoi() represent a security risk?
<14> homework, eh?
<15> jauncie: http://msdn.microsoft.com/library/default.asp?url=/library/en-us/intl/unicode_0ghf.asp
<7> eh?
<7> i dont see atoi as much of a risk...strcpy and strtok etc are a far far bigger risk
<7> any function that can overrun the supplied buffer is a risk
<13> How so?
<1> atoi()'s behavior is not defined if you give it bad input
<7> char s[1]; strcpy(s,"this is a big array");
<7> oops
<7> i always thought atoi returned 0 if the input was bad
<7> not worth it
<16> hi. what is the unmanaged windows API type for double**?
<16> if PDOUBLE is double* and PINT is int*
<16> and what's the difference between LPINT and PINT~
<17> all that is old ****
<16> if i'm writing a DLL, do i need this?
<17> relevant (maybe) back when you lad near pointers and far pointers
<7> no
<17> s/lad/had
<16> ok. so i don't need that lp mumbo jumbo
<7> lad pointers :)
<17> lol
<16> i thought u were talking british to me
<17> why do you need double** ?
<16> hmm. i'm building a DLL that takes a matrix
<16> in one of its functions as one of the parameters
<17> and what's your definition of a matrix ?
<16> hmm
<16> a 2d array
<16> an array of pointers to double
<17> k
<16> i was under the impression that i had to use these archaic type names
<17> you're writing this for some people to call from C, eh?
<16> lpwstr lpstr blah
<7> they are all typedefs
<16> hmm.. no. i am actually trying to write a modern dll, but i only got vc6
<17> vc2005 is free
<7> calc vc2005e
<17> vc6 ain't C++
<7> kniht must be mia
<16> i'll grab that as soon as i get more harddrive space.. i got like 400 MB left
<17> throwing away vc6 will clear some up
<7> and use the online msdn
<16> so.. what's the var type i should use for my typedef of double** in win API speak.. can someone tell me?
<17> yeah, that was 1.9Gb wasn't it?
<7> double**
<17> what's wrong with double**
<16> :|
<17> just because they were stupid in the past, doesn't mean you have to be now
<16> hmm.. i guess it was a no brainer.
<16> well i guess this solves a lot of problems then.. i hate flipping thru the web trying to find the old api type names..
<7> it was done for "portability" reasons
<7> win3.1 -> win32
<16> OHH that's.. stupid
Return to
#c++
or
Go to some related
logs:
#AllNiteCafe
pam_timestamp_check: pam_timestamp: `/' owner UID != 0
WHICH FLAG FLIES OVER THE CANARY ISLANDS
#linux
python for windows mobile
while running with -T switch at Daemonize.pm line 67.
#chatzone
synapsis linux
#windows
#php
|
|