Ros-dev March 2009

ros-dev@reactos.org

32 participants
36 discussions

Recent changes
by Colin Finck 26 Mar '09

26 Mar '09

Hello everybody, As you might have noticed, I recently did some changes to our Website stuff and Regression Testing System. Finally, it's time to show some results now: == BuildBot upgrade == Christoph reinstalled the latest version of BuildBot on our Swedish server today, since the previous server system was way too old for such an upgrade. This also had the uncomfortable side-effect of the BuildBot URL changing from http://reactos.org:8010 to http://build.reactos.org:8010, which most of you already seem to have noticed ;-) Sorry for not giving a note about this before. I thought I could just finish my stuff and then write this E-Mail explaining all changes, but as always, it took longer than expected :-/ == Crash Recovery in rosautotest and sysreg2 == In the recent days, I rewrote rosautotest in C++ for getting a more maintainable code. This update also added a Crash Recovery feature. Together with my updated version of sysreg2, we can now resume the testing process with the next test in case a specific test made ReactOS crash. == Web Test Interface == As we recently set up a new VM, which will eventually replace our current outdated www.reactos.org server, I finally had a playground for my third component of our testing system: The "testman" Web Interface. It is available at http://79.99.5.181/testman/ Features include finding specific test results through a getbuilds-like interface and comparing them. The Comparison feature can compare up to 5 results side-by-side and show the differences. Clicking on "Show only changed results" will only show the lines that have differences. You can also use Drag & Drop to move the column headers with all associated results to get the table to look like you want it. Clicking a test result will show details, including a log of this specific test. Just play around with it a bit and you should easily get to know about all features. Regarding Lone_Rifle's interface at http://reactos.nxserve.net: From what he told me, this one was only meant as a decentral alternative as long as we don't have "testman" on an official server. I'm saying this, so nobody gets the impression that I just wanted to extinguish his work with my implementation. If some people prefer Lone_Rifle's version over mine, it would just need to be adjusted to the new BuildBot URL and those people could still use it. Comments and suggestions are welcome of course. Not everything works perfect yet (e.g. the Debug Buildslave needs quite some time now as all tests are run), but it should be ready for you to try now. Best regards, Colin

5 8

Buildbot page moved..
by Timo Kreuzer 25 Mar '09

25 Mar '09

Hi, Since today, I cannot access the old buildbot page (http://reactos.org:8010/) anymore. I found the new URL (http://build.reactos.org:8010) in Colin's commit message. Was this an intentional change? I am asking because noone announced this change and some people might now wonder where the buildbot page is... Or am I the only one accessing it via my favourites? Is there even a link from the website (I couldn't find one)? Ah... and the old page had an icon, the new one doesn't. Timo PS: The nice regtest pages at http://reactos.nxserve.net also don't work anymore. They were very useful and I'd like to see them up again. So a) do we have something like that for our website (or in work)? Or b) can this one be fixed? (I forgot who owns it.)

5 5

Re: [ros-dev] [ros-diffs] [dgorbachev] 40181: Add "REACTOS Mm Hack of Doom" again (removed in r39723). Bug #4296.
by Matthieu Suiche 24 Mar '09

24 Mar '09

In a previous message you said: >Alex Ionescu wrote: >> This new hack directly slows down every context switch instead of every >> page fault -- the latter is a less important perf hit. > >This slowdown is very small, especially when comparing with other >"Hack of Doom". >I just want to see, is it the only raison d'être of that old hack or not. Could you explain the reason you add Mm Hack of Doom again if the slowdown is very small? Regards, -- Matthieu Suiche On Mon, Mar 23, 2009 at 1:56 AM, <dgorbachev(a)svn.reactos.org> wrote: > Author: dgorbachev > Date: Mon Mar 23 03:56:01 2009 > New Revision: 40181 > > URL: http://svn.reactos.org/svn/reactos?rev=40181&view=rev > Log: > Add "REACTOS Mm Hack of Doom" again (removed in r39723). Bug #4296. > > Modified: > trunk/reactos/ntoskrnl/ke/i386/ctxswitch.S > trunk/reactos/ntoskrnl/ke/i386/trap.s > > Modified: trunk/reactos/ntoskrnl/ke/i386/ctxswitch.S > URL: http://svn.reactos.org/svn/reactos/trunk/reactos/ntoskrnl/ke/i386/ctxswitch… > ============================================================================== > --- trunk/reactos/ntoskrnl/ke/i386/ctxswitch.S [iso-8859-1] (original) > +++ trunk/reactos/ntoskrnl/ke/i386/ctxswitch.S [iso-8859-1] Mon Mar 23 03:56:01 2009 > @@ -389,9 +389,6 @@ > > /* Checking NPX, disable interrupts now */ > mov eax, [esi+KTHREAD_INITIAL_STACK] > - > - /* HACK */ > - mov ecx, [eax - 4] > cli > > /* Get the NPX State */ > > Modified: trunk/reactos/ntoskrnl/ke/i386/trap.s > URL: http://svn.reactos.org/svn/reactos/trunk/reactos/ntoskrnl/ke/i386/trap.s?re… > ============================================================================== > --- trunk/reactos/ntoskrnl/ke/i386/trap.s [iso-8859-1] (original) > +++ trunk/reactos/ntoskrnl/ke/i386/trap.s [iso-8859-1] Mon Mar 23 03:56:01 2009 > @@ -1924,10 +1924,14 @@ > NoFixUp: > mov edi, cr2 > > + /* REACTOS Mm Hack of Doom */ > + test dword ptr [ebp+KTRAP_FRAME_EFLAGS], EFLAGS_INTERRUPT_MASK > + je HandlePf > + > /* Enable interrupts and check if we got here with interrupts disabled */ > sti > - test dword ptr [ebp+KTRAP_FRAME_EFLAGS], EFLAGS_INTERRUPT_MASK > - jz IllegalState > + /* test dword ptr [ebp+KTRAP_FRAME_EFLAGS], EFLAGS_INTERRUPT_MASK > + jz IllegalState */ > > HandlePf: > /* Send trap frame and check if this is kernel-mode or usermode */ > >

2 1

Guidance
by മഹേഷ് മുകുന്ദന് | Mahesh M 24 Mar '09

24 Mar '09

Hi, I am doing a project on implementing VFS. Can anyone help me as a mentor? Since I'm more used to in Linux than in ROS code, it would be a great help if someone can help me through. I have gone through the ntos folder and read some documents and also the winvfs project that someone here suggested. I'm aware this is not a prime target in the ROS team, but once I complete this, I might be able to contribute to areas of higher priority. Hoping for a positive reply, Thank you. -- Mahesh M Happy hacking... , , / \ ((__-^^-,-^^-__)) `-_---' `---_-' `--|o` 'o|--' \ ` / ): :( :o_o: "-"

1 0

Re: [ros-dev] [ros-diffs] [mkupfer] 40173: - Disable "Back" in last page of syssetup, because it doesn't make any sense.
by Gabriel ilardi 22 Mar '09

22 Mar '09

This hides bug 2675. > Date: Sun, 22 Mar 2009 16:36:17 +0000 > To: ros-diffs(a)reactos.org > From: mkupfer(a)svn.reactos.org > Subject: [ros-diffs] [mkupfer] 40173: - Disable "Back" in last page of syssetup, because it doesn't make any sense. > > Author: mkupfer > Date: Sun Mar 22 19:36:16 2009 > New Revision: 40173 > > URL: http://svn.reactos.org/svn/reactos?rev=40173&view=rev > Log: > - Disable "Back" in last page of syssetup, because it doesn't make any sense. > > Modified: > trunk/reactos/dll/win32/syssetup/wizard.c > > Modified: trunk/reactos/dll/win32/syssetup/wizard.c > URL: http://svn.reactos.org/svn/reactos/trunk/reactos/dll/win32/syssetup/wizard.… > ============================================================================== > --- trunk/reactos/dll/win32/syssetup/wizard.c [iso-8859-1] (original) > +++ trunk/reactos/dll/win32/syssetup/wizard.c [iso-8859-1] Sun Mar 22 19:36:16 2009 > @@ -1990,22 +1990,13 @@ > { > case PSN_SETACTIVE: > /* Enable the correct buttons on for the active page */ > - PropSheet_SetWizButtons(GetParent(hwndDlg), PSWIZB_BACK | PSWIZB_FINISH); > + PropSheet_SetWizButtons(GetParent(hwndDlg), PSWIZB_FINISH); > > SendDlgItemMessage(hwndDlg, IDC_RESTART_PROGRESS, PBM_SETRANGE, 0, > MAKELPARAM(0, 300)); > SendDlgItemMessage(hwndDlg, IDC_RESTART_PROGRESS, PBM_SETPOS, 0, 0); > SetTimer(hwndDlg, 1, 50, NULL); > break; > - > - case PSN_WIZBACK: > - /* Handle a Back button click, if necessary */ > - KillTimer(hwndDlg, 1); > - > - /* Skip the progress page */ > - SetWindowLong(hwndDlg, DWL_MSGRESULT, IDD_DATETIMEPAGE); > - SetupData.UnattendSetup = FALSE; > - return TRUE; > > case PSN_WIZFINISH: > /* Handle a Finish button click, if necessary */ > _________________________________________________________________ Scopri le keyword più cercate della settimana! http://clk.atdmt.com/GBL/go/136430524/direct/01/

1 0

Re: [ros-dev] [ros-diffs] [cgutman] 40142: - Reorganize NDIS code
by Aleksey Bragin 21 Mar '09

21 Mar '09

What was wrong with a previous naming convention? x0gone.c were supposed to contain implementation of functions, which became legacy in that major version of NDIS. Now you substituted it with x0stubs.c, which means it's going to contain stubs only. Was that intentionally? On Mar 21, 2009, at 12:29 AM, cgutman(a)svn.reactos.org wrote: > Author: cgutman > Date: Sat Mar 21 00:29:53 2009 > New Revision: 40142 > > URL: http://svn.reactos.org/svn/reactos?rev=40142&view=rev > Log: > - Reorganize NDIS code > > Added: > trunk/reactos/drivers/network/ndis/ndis/30stubs.c > - copied, changed from r40140, trunk/reactos/drivers/network/ > ndis/ndis/40gone.c > trunk/reactos/drivers/network/ndis/ndis/40stubs.c > - copied, changed from r40141, trunk/reactos/drivers/network/ > ndis/ndis/50gone.c > trunk/reactos/drivers/network/ndis/ndis/50stubs.c > - copied, changed from r40113, trunk/reactos/drivers/network/ > ndis/ndis/stubs.c > trunk/reactos/drivers/network/ndis/ndis/misc.c (with props) > Removed: > trunk/reactos/drivers/network/ndis/ndis/40gone.c > trunk/reactos/drivers/network/ndis/ndis/50gone.c > trunk/reactos/drivers/network/ndis/ndis/stubs.c

2 1

advapi self forwarding and kernel32 self imports
by Timo Kreuzer 21 Mar '09

21 Mar '09

Hi, 1.) Is there a reason, why some exports of advapi32 are forwarded instead of just set to the proper function? Forwarding is more work for the loader. 2.) kernel32 imports from itself. Which will probably lead to the same result as bug 4292. This is because the mingw lib imports from kernel32 and requires kernel32 library to be added afterwards. But rbuild likes to add mingw lib at the very end of the library list which requires adding kernel32 as a library to mingw_common to resolve imports. This should be fixed and mingw added as the very first library. I wonder if those 2 are even allowed on Windows, as both don't make sense. Timo

1 0

Re: [ros-dev] [ros-diffs] [tkreuzer] 40115: rename DCs pDc_Attr to pdcattr, like in gdikdx. Make pdcattr alsways point to a DC_ATTR, either the user mode struct or the local part. Get rid of all the If (!pdcattr) pdcattr = &dc->Dc_Attr; That are not
by James Tabor 20 Mar '09

20 Mar '09

Hi All! Good to see it go, I had it in as a safeguard, and now we moved on to the next phase of the rewrite! Thanks Timo! James 8^D On Fri, Mar 20, 2009 at 4:47 AM, Aleksey Bragin <aleksey(a)reactos.org> wrote: > Oh, finally! ;) > > On Mar 20, 2009, at 7:51 AM, tkreuzer(a)svn.reactos.org wrote: > >> Author: tkreuzer >> Date: Fri Mar 20 07:51:26 2009 >> New Revision: 40115 >> >> URL: http://svn.reactos.org/svn/reactos?rev=40115&view=rev >> Log: >> rename DCs pDc_Attr to pdcattr, like in gdikdx. >> Make pdcattr alsways point to a DC_ATTR, either the user mode >> struct or the local part. Get rid of all the If (!pdcattr) pdcattr >> = &dc->Dc_Attr; That are not required anymore. >

1 0

20 Mar '09

Oh, finally! ;) On Mar 20, 2009, at 7:51 AM, tkreuzer(a)svn.reactos.org wrote: > Author: tkreuzer > Date: Fri Mar 20 07:51:26 2009 > New Revision: 40115 > > URL: http://svn.reactos.org/svn/reactos?rev=40115&view=rev > Log: > rename DCs pDc_Attr to pdcattr, like in gdikdx. > Make pdcattr alsways point to a DC_ATTR, either the user mode > struct or the local part. Get rid of all the If (!pdcattr) pdcattr > = &dc->Dc_Attr; That are not required anymore.

1 0

[ros-diffs] [fireball] 40106: - Don't try to be smarter than Windows and don't copy background/foreground color from the original DC.
by James Tabor 20 Mar '09

20 Mar '09

Introduced in revision 6042. Over think was a norm back then.

1 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Ros-dev March 2009