Ros-dev December 2005

ros-dev@reactos.org

81 participants
142 discussions

Usage of RtlAllocateHeap
by M Bealby 11 Dec '05

11 Dec '05

Hi, First of all I would like to introduce myself. My name is Martin and I am currently attempting to go through and audit the ReactOS source code for security vulnerabilities. I've been keeping an eye on the project for about a year now and I think as development is now rapidly increasing, security must take more and more of a priority. I started with the kernel32 library on Friday (9th) and have found a number of possible security vulnerabilies. These have been filed with bugzilla and i'm very impressed with the rate that these were fixed. So well done on that count! However, after submitting bugs #1102 and #1103 (now fixed by BrandonTurner) it became apparent that incorrect usage of the RtlAllocateHeap function (and its alias HeapAlloc) is quite common. It may return NULL if memory cannot be allocated and in some parts of the code this is not checked. This would lead to a segmentation fault on reading / writing. Through some judicious use of grep I went through the source code and extracted calls to this function with their context. After going through each one and investigating it I have now found another twenty occurances of the same bug (This is just in the kernel32 library). I am writing this to let all developers know that they must check their return values. Errors can exist, resources may be starved and the call may fail. Make sure your code can cope with this. Finally, as I am fairly new to bugzilla, should I submit each of these twenty bugs individually or should I submit them all as one bug? Cheers, Martin

4 4

Re: [ros-svn] [gvg] 20062: Files to create Mozilla Control installer with correct DLL
by David Hinz 11 Dec '05

11 Dec '05

Can't we include Mozilla Control in ReactOS? So that we build it with our ROS Build Environment or whatever is used to compile ReactOS, put it on bootcd and install it with ReactOS? Greets, David Hinz gvg(a)svn.reactos.com schrieb: > Files to create Mozilla Control installer with correct DLL > > > > Added files: > trunk/tools/ > trunk/tools/MozillaControl/ > trunk/tools/MozillaControl/README.txt > trunk/tools/MozillaControl/control-ros.nsi > trunk/tools/MozillaControl/files.nsh > trunk/tools/MozillaControl/local.nsh > trunk/tools/MozillaControl/redist/ > > _______________________________________________ > Ros-svn mailing list > Ros-svn(a)reactos.org > http://www.reactos.org/mailman/listinfo/ros-svn >

6 10

RE: [ros-svn] [gvg] 20062: Files to create Mozilla Control installerwith correct DLL
by Ge van Geldorp 11 Dec '05

11 Dec '05

> From: gvg(a)svn.reactos.com > > Files to create Mozilla Control installer with correct DLL I've created a "fixed" version of the Mozilla Control installer which includes the correct MSVCP60.DLL and put it in our SourceForge download area. shdocvw.dll will now download/install that fixed version. GvG

1 0

RE: [ros-diffs] [arty] 20055: A little library for making it easier toget to PASSIVE_LEVEL. Just contains
by Casper Hornstrup 11 Dec '05

11 Dec '05

We don't have the REVISIONS section anymore and a copyright notice is missing. Casper _____ From: ros-diffs-bounces(a)reactos.org [mailto:ros-diffs-bounces@reactos.org] On Behalf Of arty(a)svn.reactos.com Sent: 11. december 2005 09:26 To: ros-diffs(a)reactos.com Subject: [ros-diffs] [arty] 20055: A little library for making it easier toget to PASSIVE_LEVEL. Just contains _____ Added: trunk/reactos/drivers/lib/chew/workqueue.c --- trunk/reactos/drivers/lib/chew/workqueue.c 2005-12-11 08:11:21 UTC (rev 20054) +++ trunk/reactos/drivers/lib/chew/workqueue.c 2005-12-11 08:26:16 UTC (rev 20055) @@ -0,0 +1,101 @@ +/* + * COPYRIGHT: See COPYING in the top level directory + * PROJECT: ReactOS TCP/IP protocol driver + * FILE: tcpip/main.c + * PURPOSE: Common Highlevel Executive Worker + * PROGRAMMERS: Art Yerkes + * REVISIONS: + * CSH 10/12-2005 Created + */

1 0

Boch 2.2.1 keyboard buffer overflow with React OS 0.2.8
by Jerry 11 Dec '05

11 Dec '05

I tried to run both the live ISO and the intstall ISO (reactos 0.2.8) in bochs 2.2.1 and I got a keyboard buffer overflow. I don't know what is wrong. Here is apart of the Bochs core dump: 00117791507e[KBD ] keyboard disabled & send of byte f2 to kbd 00117791507i[KBD ] identify keyboard command received 00118580248e[KBD ] keyboard disabled & send of byte f2 to kbd 00118580248i[KBD ] identify keyboard command received 00119368173e[KBD ] keyboard disabled & send of byte f2 to kbd 00119368173i[KBD ] identify keyboard command received 00120156914e[KBD ] keyboard disabled & send of byte f2 to kbd 00120156914i[KBD ] identify keyboard command received 00120946888e[KBD ] keyboard disabled & send of byte f2 to kbd 00120946888i[KBD ] identify keyboard command received 00121734813e[KBD ] keyboard disabled & send of byte f2 to kbd 00121734813i[KBD ] identify keyboard command received 00121734813i[KBD ] internal keyboard buffer full, ignoring scancode.(ab) 00121734813i[KBD ] internal keyboard buffer full, ignoring scancode.(41) 00122523554e[KBD ] keyboard disabled & send of byte f2 to kbd 00122523554i[KBD ] identify keyboard command received 00122523554i[KBD ] internal keyboard buffer full, ignoring scancode.(fa) 00122523554i[KBD ] internal keyboard buffer full, ignoring scancode.(ab) 00122523554i[KBD ] internal keyboard buffer full, ignoring scancode.(41) 00123313528e[KBD ] keyboard disabled & send of byte f2 to kbd 00123313528i[KBD ] identify keyboard command received 00123313528i[KBD ] internal keyboard buffer full, ignoring scancode.(fa) 00123313528i[KBD ] internal keyboard buffer full, ignoring scancode.(ab) 00123313528i[KBD ] internal keyboard buffer full, ignoring scancode.(41) 00124101453e[KBD ] keyboard disabled & send of byte f2 to kbd 00124101453i[KBD ] identify keyboard command received 00124101453i[KBD ] internal keyboard buffer full, ignoring scancode.(fa) 00124101453i[KBD ] internal keyboard buffer full, ignoring scancode.(ab) 00124101453i[KBD ] internal keyboard buffer full, ignoring scancode.(41) 00124891427e[KBD ] keyboard disabled & send of byte f2 to kbd 00124891427i[KBD ] identify keyboard command received 00124891427i[KBD ] internal keyboard buffer full, ignoring scancode.(fa) 00124891427i[KBD ] internal keyboard buffer full, ignoring scancode.(ab) 00124891427i[KBD ] internal keyboard buffer full, ignoring scancode.(41) 00125680168e[KBD ] keyboard disabled & send of byte f2 to kbd 00125680168i[KBD ] identify keyboard command received 00125680168i[KBD ] internal keyboard buffer full, ignoring scancode.(fa) 00125680168i[KBD ] internal keyboard buffer full, ignoring scancode.(ab) 00125680168i[KBD ] internal keyboard buffer full, ignoring scancode.(41) 00126470143e[KBD ] keyboard disabled & send of byte ed to kbd 00126470143p[KBD ] >>PANIC<< internal keyboard buffer full (imm)

2 2

RE: [gvg] 20024: Work around lack of MSG_PEEK
by Ge van Geldorp 10 Dec '05

10 Dec '05

With r20024 I can get ibrowser to actually download the Mozilla webbrowser control. It usually takes a few tries, often the download hangs halfway through. I have the feeling that it helps if you move the mouse constantly during download, but that could just be my imagination. To install the control, you will need a copy of MSVCP60.DLL, which the author forgot to include in the install image (Wine has the same problem). So first copy MSVCP60.DLL from your Windows installation to \ReactOS\System32, then start ibrowser, download the control and install it. After that ibrowser (and the "Web" button in ROSExplorer) becomes functional. GvG

4 5

Industry support
by David Hinz 10 Dec '05

10 Dec '05

When I first throw in the idea, that we could try to get some big companys supporting us, I was amazed by the huge response. A lot of people seem to be very interested in this, but after some days, I noticed that there are nearly no devs who replied to my mail, or the numerous replies on it. So what I'm asking for is how the devs and especially our PC Steven Edwards think about asking some companys like IBM, Intel, AMD, nVIDIA, ATI, Dell,... to support our project. There are a lot of ways, they could support us, either by money, or by developers, or by knowledge/documentations. Every of the mentioned companys would be able to support us, but I think the only person who has the right to ask any of them is our PC, or maybe somebody he authorized to do so. I'm waiting for your replies. Greets, David Hinz BTW, the mail I mentioned above is http://www.reactos.org/archives/public/ros-dev/2005-December/006645.html

5 7

Re: [ros-svn] [gvg] 20026: reactos.org is the preferred name these days
by TwoTailedFox 10 Dec '05

10 Dec '05

Does this fix the reference to 'reactos.com' with the 'ver' command, in the Command Interpreter? On 12/9/05, gvg(a)svn.reactos.com <gvg(a)svn.reactos.com> wrote: > reactos.org is the preferred name these days > > > > Updated files: > trunk/reactos/subsys/system/explorer/shell/mainframe.cpp > > _______________________________________________ > Ros-svn mailing list > Ros-svn(a)reactos.org > http://www.reactos.org/mailman/listinfo/ros-svn > -- "I had a handle on life, but then it broke"

3 2

Compilation error
by David Hinz 09 Dec '05

09 Dec '05

When I try to compile ReactOS on Windows, I get the following error: ./subsys/win32k/objects/path.c: In function `PATH_PathToRegion': ./subsys/win32k/objects/path.c:1024: warning: 'hrgn' might be used uninitialized in this function mingw32-make[1]: *** [obj-i386\subsys\win32k\objects.o] Error 1 Important: I'm compiling with optimisations for amd64 and dbg and kdbg=0. Revision is 20025 If I'm not the only one with these problems, I hope you get that fixed soon, I don't think this is because if my platform optimizations. Greets, David Hinz

1 0

process token problem
by Hartmut Birr 09 Dec '05

09 Dec '05

Hi, after compiling ros on ros, I see over 45000 memory blocks with the tags TOKd, TOKp, TOKu and TOKp. It seems, something is wrong with the referencing/dereferencing of tokens. I've add some debug code in SeDeassignPrimaryToken (only called by PspDeleteProcess), which prints the ref count of the token object. It is always greater than 1. The process token is never deleted. - Hartmut

1 0

← Newer
1
...
6
7
8
9
10
11
12
...
15
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Ros-dev December 2005