► (3/3) ReactOS Design Talk ♪♪♪

The place to bring up any design issues, or post your own creations

Moderator: Moderator Team

User avatar
Blue
Posts: 53
Joined: Mon Apr 21, 2014 8:14 pm

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by Blue »

Build 63607 -dbgwin
Wine Gecko (32-bit) v2.24

Info
- This problem only happens in the ReactOS < Bootup pick option:

1. Machine
2. Settings
3. Network tab ( \/ Image \/ )

[ external image ]

4. Promiscuous mode kills OS

[ external image ]


A Missing library in OS.
[ external image ]
Last edited by Blue on Sat Jun 21, 2014 12:44 pm, edited 14 times in total.

justincase
Posts: 441
Joined: Sat Nov 15, 2008 4:13 pm

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by justincase »

You'll need to supply more information if you want that BSoD to mean much of anything.
See viewtopic.php?f=4&t=13361.
I reserve the right to ignore any portion of any post if I deem it not constructive or likely to cause the discussion to degenerate.

User avatar
Blue
Posts: 53
Joined: Mon Apr 21, 2014 8:14 pm

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by Blue »

justincase,

Added the rest of the info.

justincase
Posts: 441
Joined: Sat Nov 15, 2008 4:13 pm

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by justincase »

First,
It's considered rude to cross-post (make the same post in multiple threads)

Second,
This is really something that should get it's own thread and/or a JIRA report

Third,
That still doesn't meet the standards put forth in viewtopic.php?f=4&t=13361
vicmarcal wrote: - We need to know if it you are testing the LIVE CD or the BOOT CD version.
- We need to know WHEN the BSOD happens(i.e: "It passed the first installation stage without issues, but when beginning the graphical second stage, it BSODed"). Please be descriptive. There are no longs descriptions, but incomplete ones. We love longs descriptions. You are our eyes.
Please get a Debug log.
Since you seem to be using VirtualBox, check the VirtualBox wiki page
(If you don't want to install Putty you can get a Portable version from Portableapps.com)
Preferably get a backtrace too (once you see "kdbg:>" type "bt" (without quotes) and press enter)

And seriously, if you want help remember:
vicmarcal wrote:You are our eyes.
It's really hard to fix something you can't see, but it can be made much easier if someone who can see it does a good job of describing it.
I reserve the right to ignore any portion of any post if I deem it not constructive or likely to cause the discussion to degenerate.

User avatar
Blue
Posts: 53
Joined: Mon Apr 21, 2014 8:14 pm

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by Blue »

justincase,

The problem stopped Ros from working & i repeated the step 6 times.. I even fully closed VM DOWN to see if VM was the problem... I can not repeat the problem I was having... after I setup the ports & removed all settings that where changed to get it to repeat... The problem looks like it has fixed itself.. No clue why VM was going down for changing this setting..... ... Now Ros shows no issue/problem after taken your suggestion above. :evil:

First,
The thread you linked to me was also apart of the testing team information... So posting on a persons thread that is apart of the test team... Should not be considered rude/disrespectful... If it was on a normal user thread.. Than linking over would be something else.. your going to take this information in anyway you feel. I'm Not being disrespectful to you or others.

I did not really have much time today as I was working, but when I was testing the OS out this is what was happening... Until I ran debug mode in the OS pick options & changed the port settings... It magically fixed the issue from happening again.. Than I removed all the changes and tried to see if the problem would repeat itself again and nothing so I could make a JIRA Report.. Now It changes fine... No clue why it was going down & now can bridge all the network adapters fine, using the promiscuous Mode options. This information up top is helpful for debug testing for reports. I'll make a debug report next time I have a problem like this.... I still found this odd that it just fixed itself...

Second,
This is really something that should get it's own thread and/or a JIRA report - Agree I should of done that.

Third,
I did not really have the time to read the thread so I just posted it before I left & only read the top intro...

"As you may know there are thousands of millions of zillions of combinations of the hardware out there.
Sometimes the problem is a specific piece of hardware, but other times the problem is how those pieces of hardware talk each other!" > than looked at the name and said (Test Team) Meaning problems you have post here! As the thread was not locked. Anyways, JustineCase Thanks for your time.

solid_black
Posts: 33
Joined: Fri Apr 04, 2014 8:33 am

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by solid_black »

justincase wrote:Since you seem to be using VirtualBox, check the VirtualBox wiki page
(If you don't want to install Putty you can get a Portable version from Portableapps.com)
By the way, there is no need to use PuTTY to get a debug log from VBox VM. Just create a COM port and set Port Mode to 'Raw File', then enter a path where you want your debug.log to be (on your host machine).
I don't know why wiki recommends to use pipes and PuTTY. Maybe there's a reason for that, but redirecting the COM port to a file is obviously easier and works just fine.

User avatar
Blue
Posts: 53
Joined: Mon Apr 21, 2014 8:14 pm

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by Blue »

solid_black
If I can do that it would make it easier than having to use a load of extra tools. I would like to know why wiki recommends it also. Just guessing here : It might be from not all VMs are setup the same way. So its just easier without explain each step to users as they can throw a tool at them. No clue really it could also be due to a better debug report from the app....

User avatar
Blue
Posts: 53
Joined: Mon Apr 21, 2014 8:14 pm

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by Blue »

==================================================================
IN OTHER NEWS on Jun 20, 2014
/* No clue if OpenGL works under Ros! never tested it yet. */
==================================================================

GLIntercept 1.3 adds OpenGL 4.4 logging + bug fixes
The OpenGL logging/debugging tool GLIntercept has been updated to include the latest OpenGL XML registry headers and fixes several minor bugs.
https://code.google.com/p/glintercept/

1. Mesa3D <QUESTION>
3rd Party File.txt - Reports that Ros is using v8.0.4 from July 10, 2012 & why has it not been updated if that is true? so they have a newer one just released Jun 9, 2014.

Title: Mesa3D
Used by Ros Version: 8.0.4
Website: http://www.mesa3d.org

> Current Version: Mesa 3D Graphics Library - (v10.2.1)
ftp://ftp.freedesktop.org/pub/mesa/

2. DirctX10
DX10 has not been added inside the Ros build .svn other than inside Wine... Please add DX10 inside the Ros Build and yes. I've did look inside subsystems and DX8 and DX9 was the only DX hosted and 10 was left out Unless I'm still not spotting the DX10. The idea is not for me to port over DX10 myself but have it inside the build so everyone can use it.


1 and 2 is what I would like to see updated inside the .svn

hbelusca
Developer
Posts: 1188
Joined: Sat Dec 26, 2009 10:36 pm
Location: Zagreb, Croatia

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by hbelusca »

solid_black wrote:By the way, there is no need to use PuTTY to get a debug log from VBox VM. Just create a COM port and set Port Mode to 'Raw File', then enter a path where you want your debug.log to be (on your host machine).
I don't know why wiki recommends to use pipes and PuTTY. Maybe there's a reason for that, but redirecting the COM port to a file is obviously easier and works just fine.
If you want to be able to do a backtrace (and more generally, interact with the kernel debugger) and see what it shows to you, I'm not sure redirecting the COM port to a file log is the best thing you need to do... (hence the PuTTY / pipes options).

Frontier
Posts: 70
Joined: Fri Sep 20, 2013 10:29 am

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by Frontier »

In version 9, mesa moved to gallium based ICDs, which were unsuitable for ROS. DirectX 10 is Vista tech.

User avatar
Blue
Posts: 53
Joined: Mon Apr 21, 2014 8:14 pm

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by Blue »

hbelusca
Thanks that explains PuTTY questions.

Frontier
Version 9, mesa moved to gallium based ICDs, which were unsuitable for ROS. (I see)
I ask the developer and s/he said look inside wine sub in svn... Than I was thinking there is no DirectX 10 listed... in svn...
DirectX information: http://www.falconfly.de/directx.htm

middings
Posts: 1036
Joined: Tue May 07, 2013 9:18 pm
Location: California, USA

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by middings »

Frontier wrote:In version 9, mesa moved to gallium based ICDs, which were unsuitable for ROS.
Anyone who wants more details can read z98's Newsletter 95 article "OpenGL".

milon
Posts: 969
Joined: Sat Sep 05, 2009 9:26 pm

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by milon »

hbelusca wrote:
solid_black wrote:By the way, there is no need to use PuTTY to get a debug log from VBox VM. Just create a COM port and set Port Mode to 'Raw File', then enter a path where you want your debug.log to be (on your host machine).
I don't know why wiki recommends to use pipes and PuTTY. Maybe there's a reason for that, but redirecting the COM port to a file is obviously easier and works just fine.
If you want to be able to do a backtrace (and more generally, interact with the kernel debugger) and see what it shows to you, I'm not sure redirecting the COM port to a file log is the best thing you need to do... (hence the PuTTY / pipes options).
Would you make a note in the wiki that raw file output is possible but not recommended? That would be more user-friendly, IMO. I'd do it myself, but I don't understand it well enough to clarify why it's not recommended.

User avatar
Blue
Posts: 53
Joined: Mon Apr 21, 2014 8:14 pm

Re: ► (3/3) ReactOS Design Talk ♪♪♪

Post by Blue »

Its not a big deal, at all but I would like the SVN documents to be changed as a standard on Ros. When you've first open a document as the example below details. It should not be this way. even though it does not hurt anything it should be improved for better standards... At times even a start of a document is smashed together like so...

; ACPI.INF
; Installation file for ACPI driver
[Version]

--------------------------------------------
----------- Example: -----------
--------------------------------------------

; ACPI.INF

; Installation file for ACPI driver

[Version]
Signature = "$Windows NT$"
;Signature = "$ReactOS$"
LayoutFile = layout.inf
Class = System
ClassGUID = {4D36E97D-E325-11CE-BFC1-08002BE10318}
Provider = %ReactOS%
DriverVer = 10/11/2005,1.01.0.0


--------------------------------------------
All Documents that start like above should be changed to look like below.
--------------------------------------------

;
; ACPI.INF
;
; Installation file for ACPI driver
;


[Version]
Signature = "$Windows NT$"
;Signature = "$ReactOS$"
LayoutFile = layout.inf
Class = System
ClassGUID = {4D36E97D-E325-11CE-BFC1-08002BE10318}
Provider = %ReactOS%
DriverVer = 10/11/2005,1.01.0.0

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest