Перейти к содержанию

Билды CE из репозитария


Рекомендуемые сообщения


make sure iexplore closes

fix onethread breakpoint cleanup

fix for 64-bit speedhack

fix the structure

add some AMD info

forgot some breaks here

some cleanup

add some stuff that makes testing easier and remove psod

forgot some breaks on vmcalls that are hardly used

test commit

some gui changes

add a find next to the auto assembler window

Add lua calculation/execution to memoryrecord setValue . use [ ] to use

Add a custom range to the assembly scan

fix create process breakpoint on entry

make the stackview resize on window change instead of the hexview if stackview is visible

add option to name single addresses

Ссылка на комментарий
Поделиться на другие сайты


only shiftselect entries when visible

add a refresh to the foundlist update

fix the no-serial build

fix kernelmodule unloading when global debug was used

implement a few more realmode instructions (vmware still doesn't fully boot. Could be a vmware issue as one of their asserts fails)

and don't output debug data if it's not required

add a lua filter to the pointerscan rescan for whatever reason you might want one

fix opening a pointerfile with the wrong process open

change the order of rescan so the base requirement fits better with the offset rule

increase the percentage that the user's ad show. Fix the auto rotation. And make the supportCheatEngine command rotate to a new ad immediately

add structurename.elementname to the symbol handler

+ скомпилировал CheatEngine.chm, dbk32.sys и vmdisk.img

Ссылка на комментарий
Поделиться на другие сайты


Cheat Engine 6.3 Beta 1

let's not skip 6.3

Prepare for first beta test

fix the luaclient template


fix out of sequence access of a unknown initial value saved type and fix the pagerange upgrade for exception breakpoints

add autosigning to the ces buildscript (usable only by me)

add a new ceregreset that can save the custom types

commit the changelist

trying to import the custom lua source ce uses into the svn

remove a file that shouldn't have been commited

fix the "boot without vm" option (didn't take care of idle AP cpu's)

Ссылка на комментарий
Поделиться на другие сайты


improve performance by telling the cpu which fields didn't get changed

fix "changed addresses" for exceptions and improve performance

fix readmsr (Every time I forget that the documentation is missing the fact that it clears the bits of 64-bit registers)

protect the efer.svme bit from being set to 0 by the guest

remove some debug stuff. Also worth mentioning that dbvm now boots up windows 8 on an AMD system

implement protection of the VM_HSAVE_PA msr

add struct/endstruct support to the auto assembler

fix TCEForm.LoadFromFile

emulate the bios calls for amd that determine which memory is available so it won't overwrite dbvm's memory (dbvm boots in dbvm now)

wasting hours of time just to find out that the 4KB page at 00000000 has some bits changed on vmexit/entry because I didn't set one msr

implement the vmxloop for amd

Now just need to handle the events (not as many as with intel)

fix instruction preference for some instructions

AMD's boots up in bochs... (This is pathetically more easier than intel's implementation)


add a custom background color to structure elements

fix the listview

Ссылка на комментарий
Поделиться на другие сайты


only move selected entries if the parent is not selected

fix selecting the function from the functionlist

committing some of the dll's to the svn

changelist cleanup

helpfile update

remove some debug info

make sure no exceptions are sent to ce while the initial threadlist is sent to ce

fix compiler option where it generates a .dat.exe instead of .dat

configure build of ceregreset

change dbvm offloading to kernelmode

change dbvm offloading to kernelmode

automatically add the d3dhook dll's to the exe trainer generator

fix new method declarations when using the formdesigner

implment the d3d toggle hotkey

fix new method declarations when using the formdesigner

fix the checkbox position on resize

change the VEH debugger to emulate an breakpoint event on attach. This fixes ce from prematurely setting breakpoints before the vehhandler is setup and all threads are signaled

activate multiple cpu support

Ссылка на комментарий
Поделиться на другие сайты


Cheat Engine 6.3 Beta 2

clear the GD bit in dr7 before signaling the interrupt

fix a stupid bug

fix another bug (somehow ss dpl was 0)

add dbvm support for amd in the driver

implement int1 redirect for amd

some fixes to the trainergenerator and implement the changes needed to get dbvm offloading to work on AMD systems

fix a bug that might be responsible for the crappy offload crashes

fix selecting visible entries only

add support for runtime offloading into dbvm for amd

DBVM теперь доступна и на процессорах AMD


Правда загрузиться с ней мне не удалось.

Если кто хочет попытаться, запишите файл "vmdisk.img" на дискету (оО) или флешку при помощи "Win32 Disk Image", и затем стартуйте систему именно с неё, дальше появится меню и там хз =)

Ссылка на комментарий
Поделиться на другие сайты

Я ради интереса скачал Cheat Engine 6.3b2 rev1874 (x86), и попробовал включить DBVM ради интереса, но послевключение комп завись, после перезагрузки комп глючить, то зависает, то экран " :( " с таким значком выходить и проценты считает (у меня вин8).

Как отключить этот DBVM?

Сейчас, после включение при загрузке "Обязательню проверку подписи драйверов" (не помню как оно назывался) вроде работаеть. Но это не правиьлно наверно и не знаю дальше как будет работать.

Извените за флуд, просто этот комп на работе, если не исправлю началника меня убъет).

Ссылка на комментарий
Поделиться на другие сайты

Вообще, если включить проверку подписи драйверов, то DBVM не работает. Так что такой метод верный.

Ещё нужно зайти в CE, в диалог "О программе", и щёлкнуть мышкой на надпись "На вашей системе запущена DBVM версии...", если таковая будет (По крайней мере, она так включается, возможно и выключается так же)

Не зря я писал 1-ом сообщении:


Используйте на свой страх и риск.

DB постоянно допиливаем DBVM под amd, так что врят ли она ещё готова на 100%. Лично я бы не рискнул её запускать на рабочем компьютере с важной информацией. Не делайте больше таких глупостей.

Ссылка на комментарий
Поделиться на другие сайты

Благодарю за ответ SER[G]ANT. Утром пришел на работу и вроди пока работает.

ANT' timestamp='1370020757' post='14518']

Ещё нужно зайти в CE, в диалог "О программе", и щёлкнуть мышкой на надпись "На вашей системе запущена DBVM версии...", если таковая будет (По крайней мере, она так включается, возможно и выключается так же)

Я тоже думал, что там же отключение но такого надписи не было до и после включение "проверку подписи драйверов". Может еще не дороботали... не знаю.

Ссылка на комментарий
Поделиться на другие сайты


remove a debug label and some more alignment stuff

some more alignment fixes

add commonmodulelist.txt to the svn

remove some debug messages

change the way icons are handled in the menu processlist

fix bootup on intel (test msr doesn't exist in intel)

update changes list

fix "flashing" when selecting a different entry in the found code dialog

still giving he finger to AMD users when it comes to ultimap

remove debug message and fix memory leak

implement int14 intercept for amd

implement int3 intercept for amd and test for int3

some debugging to find out why it would randomly crash. (found it: didn't initialize a variable)

properly fix the 64-bit handling of GD

Ссылка на комментарий
Поделиться на другие сайты

rev 1907

CheatEngine 6.3 RC1

change position

fix the structure spider so it can handle structures bigger than 4096.

Also fix support for 64-bit targets and fix where end of memory addresses where marked as readable

deal with interference of users

fix the Changed Addresses form from showing when it shouldn't

compile fix for 32-bit and update version number

add the manual stacktrace form and source

fix save/load traces

add a manual stackview and override to the threadlist to show 32-bit states

add option to show what addresses an opcode accesses when using a "find out what accesses this address"

(max 8)

this debug var isn't needed anymore

fix the vehdebug

fix last address bug in the structure spider

Disable the hex checkbox for groupscan and add the ability to let the user to pick which items to add to the addresslist when using a groupscan

make the hex checkbox invisible when grouped scan is used

don't forget unregisterFormAddNotification

Implement registerFormAddNotification for lua (makes adding features to forms easier)

deal with the fact that the lcl groupbox click bug has been moved from mouseclick back to mousedown, so update the event handler again

fix continue from execute breakpoint in xp

quick untested fix for msr c0000080 (might need more testing)

Ссылка на комментарий
Поделиться на другие сайты


fix escape not exiting the MultilineInputQuery

add ctrl+enter shortcut to edit comments

add ctrl+alt+c shortcut to copy bytes only

doc update

update changelist

might as well update the 32-bit d3d hook dll's with debug stuff removed

adding the 64-bit d3dhook dll's to the svn

fix compiling in debug mode for dxhookbase 64-bit

fix the snapshot counter not incrementing in 64-bit

remove some debug code

some more 64-bit compile fixes (and make ce inject the appropriate dll)

fix compilation for 64-bit

remove a debug flag

make the stacktrace a bit easier to read/work with

fix selecting non-visible entries

createD3DHook was correct, it just wasn't lowercase

detect direct3d hook

apply hotkey description updates

add easier trace opening in the memory view window

fix opening traces with stack snapshots

link the manual stacktrace with the stackview window for automatic lockviews

only update the stackview when it's visible when going through a trace

add res* support to the assembler

add aa highlighting to some new keywords

keep last scan when assembly scanning

apply the background color when creating a new entry as well

fix setting the selected group different color text

remove breakpoint after 8 hits when launched using the foundcode dialog

Ссылка на комментарий
Поделиться на другие сайты

Ну, финал - это я поспешил, т.к. уже нет приписки в виде "beta", "rc", "alpha". Скорее это еще и не финал, т.к. и после 1929 ревизии есть обновления, а на офф. сайте пока тихо. Там багов еще хватает)

100% Final будет после релиза на офф. сайте.

Ссылка на комментарий
Поделиться на другие сайты


Cheat Engine 6.3 Final

fix xmplayer call

fix parsing 64-bit numbers when using (int)

fix opcodes A0 to A3 for 64-bit

fix symbols for aa self

fix LUA_functioncall

document stream better

fix readBytesFromTable

add the Memory property of the memorystream class

fix the old d3d lua commands to return class objects instead of raw userdata


Наверное из-за того, что я собирал драйвер под Windows 7. Но уже не важно, на сайте вышла офф. версия.

Ссылка на комментарий
Поделиться на другие сайты

  • 3 недели спустя...

rev1966 (Это все ещё 6.3 Final. DB часто обновляет установщик на офф. сайте)

fix dq highlighting

fix threading bug

change GetThreadSelectorEntry to a dynamic linked entry (even though according to docs vista 64 DOES have it)

fix defines not accepting previous defines

Ultimap: Enter critical section when actually needed

zero init some variables and add hotkey timing functions to lua and the genericHotkey object

fix writeLocal for some functions

fix parsing groupscan strings


fix unchecking hex when changing scantype

forgot that copying 0 bytes from a stream copies ALL the bytes of that stream

fix the stacktrace parameters

add a "todo" for the pointerscan files

fix export/import memory in 64-bit

fix generichotkeys from interfering with the "default" hotkeys in settings

fix getItems

some more fixes to the array handler

add a default array handler for the menuItem

FIX the default array handler

Ссылка на комментарий
Поделиться на другие сайты

  • 3 недели спустя...

rev1974->rev2001 (большая часть относится к серверной части CE, которую я не знаю как компилировать)

remove redirected symbol support for now (needs better implementation)

Started with network debugger interface

implement jgoemat's AA templates ( http://forum.cheaten...ic.php?t=566415 )

made a few minor adjustments, not that it matters much, but it'd be a shame not to use the disassembler object when it already provides a lot of functionality used here

might as well mark as a + version

forgot one last stupid mistake

fix script generation for xmplayer stop when trainer loses focus

Implemented GetThreadContext (untested and probably has bugs)

implement suspend/resume thread (freezes whole process, but useful for delayed operations like suspend, get context, set context, resume )

redesign the server part so internal debug messages can get handled more efficient (without having to send a random sigstop)

test and fix watchpoint breakpoints for arm (find what writes/accesses ....)

note to self: Opposed to intel, these breakpoints seem to break before execution

some arm tests

quote symbols with space and comma

implement code for arm breakpoints. (not yet tested)

implement removeBreakpoint

add breakpoint setting for x86 arch

fix the Component property

deal with forwarded exports

add local var display and reg+offset notation

parameters are now shown in the disassembler view

add cvconst.pas

start adding support for extra pdb data like function parameters and local vars

some more experimenting


implement debugserverfd selecting. todo: See if I can make waitfordebugevent NOT return to the client on local rpm

add a critical section to setfilepointer and read/writefile

add some tests and dealing with an extra annoyance

apparently you can't call printf inside a signal handler.

fix breakpoint detection

better event handling

Ссылка на комментарий
Поделиться на другие сайты

  • 4 недели спустя...


fix Description property

implement network getThreadContext (untested)

small change in the way the context is sent back

fix entry selection when switching tabs right after a scan

fix CheckForAndDispatchCommand

implement dealing with compressed results of rpm

implement compression of readprocessmemory

implement network symbols in the client

fix and change the header generation


start with symbol stuff

I don't know why, buit I keep on making this same mistake

implement client side of module32first/module32next, and fix closehandle to be not so slow...

improve speed dramatically (stupid naggle)

fix realloc bug

check for module list instead

it's 35C

really stupid bug

implement module32first/module32next

and there's SWI

fix multi and implement SWPParser

implement LDM/STM

implement SingleDataParser and fix post/pre indexed disassembling for the same instruction class

implement mul parser

implement MSR and MRS

implement branch parser

and a compiling fix

fix record properties

fix getpropertylist

also commit the changes to the disassemblerarm.pas which the assembler relies on

start with arm assembler

fix setting unsupported method to nil

fix the legacy opendialog_execute

implement an ARM disassembler

implement more network debug options. (trap works properly now)

some more fixes here and there

fix calculation with multiple negatives

fix dline.GetHexBytes

some fixes here and there

update about form and some network bugfixes and decrease network usage

add reference count to the handle and some bugfixes

implement the basic stuff for the network debugger interface (not tested)

add some initial debug events for thread creation

add undocumented shl/shr reg

close the thread handle when creating a remote thread. (Some people never close ce and constantly create threads)

some syntax check fixes

add message in case of wrong parameter

implment "reassemble" aa command

forgot a space

Ссылка на комментарий
Поделиться на другие сайты

  • 1 месяц спустя...


best keep a record of methods that aren't implemented in 6.3

add makeVisible and commuit some stuff I might have forgotten to commit

deal with math errors

fix some bugs with queued events and the difference between the official debug state shown to the user and the internal state

some debug fixes

some fixes for arm

forgot to save

fix B producing a 12 byte branch instead of an 8 byte (uninitialized var)

add support for rewriting conditional branches

some arm and auto assembler stuff. (I should really do something about that nopping between variable size instructions )

add missing file

fix some bugs and add an alternative to B(L)

start with adding support for arm instructions that break when relocated

some arm assembler and disassembler fixes

fix r32,mm,imm and r32,xmm,imm (PEXTRW)

implement psubsb and psubsw

might as well disassemble it

add pmulld just because I can


implement RCPSS and RCPPS

implement RSQRTPS and RSQRTSS

fix fyl2xp1

fix crash when breaking in a 32-bit app (not arm)

fix the index lookup not handling non alphabetical characters in the index (f2xm1)

fix fnclex

fix lpd/lps parameter order

fix movhps<>movlhps


fix movlps<>movhlps

fix rsqrtss

fix shld

fix fsubr, fdiv and fdivr

fix faddp st(0), st(0)

fix disassembling fldl2e and improve the description of some instructions

fix dsr disassembly

forgot to put some code back

Implement clock_gettime


add isNetwork to the processhandler

Fix injectdll for network

Implement the speedhack for network

start implementing the liux speedhack

fix FindFreeBlockForRegion when in network mode

implement network version of injectdll



implement a network version of CreateRemoteThread

implement CMD_CREATETHREAD and fix possible error with module injection (readlink doesn't return a 0-terminated string)


Implement network variants of VirtualAllocEx and VirtualFreeEx



implement the command to load the extension and start with alloc

redefined the ext_ command defines

and some cleanup

improve the flow

implement enabling the extension and calling alloc

Add detection if the extension module has been loaded

Implement communication between ceserver and ceserver_extension

constructor test

start with the extension module

Share project "ceserver-extension" into "https://cheat-engine.googlecode.com/svn"

filter by typing directly into the processlist

implement moduleinject for arm as well

switch from pread to an lseek64/read combination as I couldn't get pread to read addresses that could be interpreted as negative (e.g: b0000000)

merge the experimental code with the svn just to make sure it's safe since it's working on 64-bit (need to add setting the proper parameters for arm)

Found a faster injection method without the need to wait for the thread to wake

documentation and experiment results

Strip version numbers from linux modulenames

start with research and design on making alloc/createthread and other useful options available

deal with new isExe flag

fix symbols on program files

implement reactivating a breakpoint in arm after it got disabled for next step

fix arm compilation

fix the thread specific check

set and remove breakpoint now deal with multiple/user specified breakpoint registers. (untested)

Some arm disassembler fixes and adding code to deal with the debugger capabilities of the server

Transfer hardware breakpoint capabilities to ce client

Add support for x86 breakpoint address results

some fixes and start with support for multiple bp's

add armdisassembly to the foundcode dialog

some debug and symbol fixes

implement arm disassembling

implement breakpoint handling for x86 network (tested)

fix breakpoint removal and return values

Ссылка на комментарий
Поделиться на другие сайты

  • 4 недели спустя...


clean up the scandata uploader when the scan finishes

add extra safeguard

disable the timer and clean up the list when new is clicked

you can just clear this var

small fixes

cleanup handles

fix listener

fix error

commit something that might cause a weird compile error

fix the character range to accept

start with implementation of scan operation broadcasts

map the apic instead of blindly guessing

fix inserting return each time ctrl+f is used

implement out of diskspace checking and handling that status

some more fixes. Distributed rescan works now

some bugfixes and implement client->server status updates

implement downloadpage and add some missing files to the svn

first implementation for rescan (untested, 100% bugged out)

replace map with pagemap as it's a lot more efficient than the normal Map object (greatly increases the rescan speed)

add a pagemap class (just for testing purposes)

implement the server side for rescans and a few bugfixes

improve pointer rescan speed a bit, and combine all rpm calls into one location

some fixes, adding more data to be visible to the user, started with designing the rescan

forgot the structsize

add some extra data to the disassembler

fix hostname lookups

implement scandata uploader/downloader

fix compile for release

implement client part of the protocol (rescan still needs to be added)

implement the server part of the protocol

Start implementing support for distributed pointerscanning

Add option to export the pointermap and made it possible to do a pointerscan without attaching to a process

improve pointerscan speed

commit some last fixes before royally screwing up my own copy

somewhere along the line I (or fpc) might have fixed the threading bug that blocked rescan from being multithreaded

Ссылка на комментарий
Поделиться на другие сайты

  • 4 недели спустя...


fix releaseObjectHandle and also enumerate all other .net modules

implement code to add the first .net module to the symbol lookup and implement it's usage

implement the ReleaseObject method command to the dotnetpipe client class

And adding the DotNetDataCollector binaries for those without visual studio

also report if .net 4.5 is supported for structure layout lookup and add a simple hack that tells windows the app didn't crash

start merging the .net data collector with the symbol handler and make some adjustments

Commit the DotNetDataCollector

fix nextscan

fix next scan having the possibility of causing a buffer overflow in some situations

Add an option to make the addresslist display values using the dropdown list format

commit changes in mainunit.frm

fix bug caused by lazarus autorename feature

better detection if a module is 64-bit or not

fix compiling for 32-bit users

fix stupid bug

add a donottrace.txt to the tracer

don't edit when rightclick

fix maxpd, maxss and maxps

fix assembling fmul in 64-bit (it has no rex_w )

make "show previous value column" persistent

add data size recognition to the disassembler and make the tracer use those sizes when guessing the type

adding some extra dll's

Add a group option that shows a +/- sign in front of entries

Save useless data to the cheat table that people are interested in for some unknown reason

Implement dropdown list capability to memory records

Make some components resize if they can't contain the text

some beautifications

fix script editing

implement inputQuery for lua

Add the ability to change values without using an inputquery

only affect the base types

fix changing the hex display while frozen

fix merge (link)

gui fixes and kick rescan workers if their results have already been merged

make "resyncModulelist" also resync stacks

fix getting the wrong filepath

error out when the user gives an invalid multiplication value

continue with implementing merge and distributed rescan (still untested)

Ссылка на комментарий
Поделиться на другие сайты

  • 5 недель спустя...


minor change to errorOnLookupFailure

fix compile error

add field and method enum

Commit the mono data collector. Still WIP (similar to the DotNetDataCollector, but this one will be controlled from the lua engine)

fix compilation for 32-bit

change the behavior of the pipe function and add a destructor

add pipe communication ability to lua. (useful for talking to injected dll's / helper processes )

Optimized the cuda scan a bit more. Speed for a level 5 tutorial scan is still equal to a single threaded non-optimized scan. This project can most likely be abandoned for now

fix the addresslist documentation

fix a bug. It's now only slightly slower to a single threaded pointerscan on a normal cpu

committing some cuda pointerscan stuff

Committing some code i've been experimenting with

bug the user only when he uses a different structname than the default no structname

fix a small bug that caused some bugreports

give it a break

compile fix for 32-bit

Make the disassembler show some more data in the comment field

Adding some extra safety checks to the structure dissect

Fix the structure settings from closing after giving invalid input

(last commit meant physical of course)

Add an option to make virtual memory scanning safer. (on by default)

add code to do big endian file access (not used, but it's there)

fix a bug in the symbol handler where it filtered out certain symbols

Add a driver function to get the physical memory layout and make stopping ultimap a bit more stable

update the documentation and make two functions more forgiving

add a safety check to the pe-info dissector

fix GetThreadList

fix ultimap hotkeys

fix non defined symbols from getting function parameters

Change the way exceptions that reach the gui are handled

Fix the 64-bit registers not aligning properly

make the code injection template make use of the caller memory view window

Fix find, and implement entry highlighting based on the address

add a find to the stackview

Apply only the changed fields when changing multiple records

fix the deselection when rightclicking an multiselected entry

fix problem where deleting an non-visible component wouldn't work

make setAddress call ReinterpretAddress

Added the much requested "Pointer must start with" feature. (it will not speed up the scan one bit, probably makes it slower, but it may help with diskspace)

change the way the dotnetpipe is being used so it's more thread safe and export a lua function to specifically reload a module's symbols

export fillFromDotNetAddress to lua

assembler: fix the extra data

structure dissect: Make the autogenerated child structures use .net lookup info as well when possible

make structure dissector get the layout from .net

fix the .net modulename/base setup and make the dll and symbol window show the .net symbols

Ссылка на комментарий
Поделиться на другие сайты

  • Создать...

Важная информация

Находясь на нашем сайте, Вы автоматически соглашаетесь соблюдать наши Условия использования.