Ros-dev June 2007

ros-dev@reactos.org

25 participants
31 discussions

Re: [ros-dev] [ros-diffs] [dgorbachev] 27187: Improve FIXME comment. For future generations of developers. :)
by Alex Ionescu 15 Jun '07

15 Jun '07

Is it really that hard to realize the line of code should've been: LdrEntry->SectionPointer = Section; -----Original Message----- From: ros-diffs-bounces(a)reactos.org [mailto:ros-diffs-bounces@reactos.org] On Behalf Of dgorbachev(a)svn.reactos.org Sent: Friday, June 15, 2007 5:12 AM To: ros-diffs(a)reactos.org Subject: [ros-diffs] [dgorbachev] 27187: Improve FIXME comment. For future generations of developers. :) Author: dgorbachev Date: Fri Jun 15 16:11:48 2007 New Revision: 27187 URL: http://svn.reactos.org/svn/reactos?rev=27187&view=rev Log: Improve FIXME comment. For future generations of developers. :) Modified: trunk/reactos/ntoskrnl/mm/sysldr.c Modified: trunk/reactos/ntoskrnl/mm/sysldr.c URL: http://svn.reactos.org/svn/reactos/trunk/reactos/ntoskrnl/mm/sysldr.c?rev=2… ============================================================================== --- trunk/reactos/ntoskrnl/mm/sysldr.c (original) +++ trunk/reactos/ntoskrnl/mm/sysldr.c Fri Jun 15 16:11:48 2007 @@ -1767,7 +1767,7 @@ NtHeader->OptionalHeader.AddressOfEntryPoint); LdrEntry->SizeOfImage = DriverSize; LdrEntry->CheckSum = NtHeader->OptionalHeader.CheckSum; - LdrEntry->SectionPointer = NULL; // ? + LdrEntry->SectionPointer = NULL; // FIXME: Bug 2151. /* Now write the DLL name */ LdrEntry->BaseDllName.Buffer = (PVOID)(LdrEntry + 1);

2 1

Re: [ros-dev] [ros-diffs] [fireball] 27186: - Resurrect "old" bootvid. The plan is to * make it working in trunk * add framebuffer support for olpc and xbox kind of devices.
by Aleksey Bragin 15 Jun '07

15 Jun '07

Are you allright, Alex? :-) This is in a branch, OLPC ("branches/olpc/"). I had to disable critical warnings in freeldr so I don't have to do a BIOS-less keyboard support in olpc-freeldr in order to just hit ENTER one time during the bootup. And certainly, when trunk's bootvid has VGA dependency coded into the very heart of it, I had to get old bootvid so that I could do the framebuffer support (no VGA bios in OLPC hardware). Also in that same branch. As for hto's (Dmitry) patches - there are controversial opinions, but I'm trying to review and discuss commits he is doing. WBR, Aleksey Bragin. On Jun 15, 2007, at 10:51 PM, Alex Ionescu wrote: > Screwing up compatibility continues! Here's an idea, let's rename > all the kernel exports! > > -----Original Message----- > From: ros-diffs-bounces(a)reactos.org [mailto:ros-diffs- > bounces(a)reactos.org] On Behalf Of fireball(a)svn.reactos.org > Sent: Friday, June 15, 2007 3:59 AM > To: ros-diffs(a)reactos.org > Subject: [ros-diffs] [fireball] 27186: - Resurrect "old" bootvid. > The plan is to * make it working in trunk * add framebuffer support > for olpc and xbox kind of devices. > > Author: fireball > Date: Fri Jun 15 14:58:37 2007 > New Revision: 27186 > > URL: http://svn.reactos.org/svn/reactos?rev=27186&view=rev > Log: > - Resurrect "old" bootvid. The plan is to > * make it working in trunk > * add framebuffer support for olpc and xbox kind of devices. > > Added: > branches/olpc/drivers/base/bootvid/bar.bmp

1 0

Re: [ros-dev] [ros-diffs] [fireball] 27185: - Set esp in the beginning to freeldr's stack, fixes a crash during a call to the kernel's entrypoint. - Comment out some annoying / unnecessary dprints (including messagebox about not resolved imports for pciide).
by Alex Ionescu 15 Jun '07

15 Jun '07

So the new ReactOS is all about adding hacks to the kernel and hiding existing bugs instead of finding solutions. Cool beans. Maybe you should also comment out other DPRINTs. I mean, if a critical warning that a driver wasn't properly loaded because FreeLDR doesn't process import tables and that thus the device won't work right is annoying, maybe stuff like "File not found" should also be removed and return success instead. I can't wait 6 months from now when everyone wonders how come some driver doesn't load in ReactOS. Are you and hto actually the same person? -----Original Message----- From: ros-diffs-bounces(a)reactos.org [mailto:ros-diffs-bounces@reactos.org] On Behalf Of fireball(a)svn.reactos.org Sent: Friday, June 15, 2007 3:27 AM To: ros-diffs(a)reactos.org Subject: [ros-diffs] [fireball] 27185: - Set esp in the beginning to freeldr's stack, fixes a crash during a call to the kernel's entrypoint. - Comment out some annoying / unnecessary dprints (including messagebox about not resolved imports for pciide). Author: fireball Date: Fri Jun 15 14:27:27 2007 New Revision: 27185 URL: http://svn.reactos.org/svn/reactos?rev=27185&view=rev Log: - Set esp in the beginning to freeldr's stack, fixes a crash during a call to the kernel's entrypoint. - Comment out some annoying / unnecessary dprints (including messagebox about not resolved imports for pciide). Modified: branches/olpc/boot/freeldr/freeldr/arch/i386/arch.S branches/olpc/boot/freeldr/freeldr/arch/i386/loader.c branches/olpc/boot/freeldr/freeldr/arch/i386/macholpc.c Modified: branches/olpc/boot/freeldr/freeldr/arch/i386/arch.S URL: http://svn.reactos.org/svn/reactos/branches/olpc/boot/freeldr/freeldr/arch/… ============================================================================== --- branches/olpc/boot/freeldr/freeldr/arch/i386/arch.S (original) +++ branches/olpc/boot/freeldr/freeldr/arch/i386/arch.S Fri Jun 15 14:27:27 2007 @@ -395,7 +395,7 @@ //movw %dx,%fs //movw %dx,%gs //movw %dx,%ss - //movl $STACK32ADDR,%esp + movl $STACK32ADDR,%esp //movl $mb_info,%ebx /* See if the boot device was passed in */ Modified: branches/olpc/boot/freeldr/freeldr/arch/i386/loader.c URL: http://svn.reactos.org/svn/reactos/branches/olpc/boot/freeldr/freeldr/arch/… ============================================================================== --- branches/olpc/boot/freeldr/freeldr/arch/i386/loader.c (original) +++ branches/olpc/boot/freeldr/freeldr/arch/i386/loader.c Fri Jun 15 14:27:27 2007 @@ -487,7 +487,7 @@ } else { - DbgPrint("Don't yet support loading new modules from imports\n"); + //DbgPrint("Don't yet support loading new modules from imports\n"); Status = STATUS_NOT_IMPLEMENTED; } } Modified: branches/olpc/boot/freeldr/freeldr/arch/i386/macholpc.c URL: http://svn.reactos.org/svn/reactos/branches/olpc/boot/freeldr/freeldr/arch/… ============================================================================== --- branches/olpc/boot/freeldr/freeldr/arch/i386/macholpc.c (original) +++ branches/olpc/boot/freeldr/freeldr/arch/i386/macholpc.c Fri Jun 15 14:27:27 2007 @@ -253,7 +253,7 @@ BOOLEAN OlpcDiskGetDriveGeometry( ULONG DriveNumber, PGEOMETRY DriveGeometry ) { - ofwprintf("GetGeometry(%d)\n", DriveNumber); + //ofwprintf("GetGeometry(%d)\n", DriveNumber); DriveGeometry->BytesPerSector = 512; DriveGeometry->Heads = 16; DriveGeometry->Sectors = 63; @@ -262,7 +262,7 @@ ULONG OlpcDiskGetCacheableBlockCount( ULONG DriveNumber ) { - ofwprintf("GetCacheableBlockCount\n"); + //ofwprintf("GetCacheableBlockCount\n"); return 1; }

1 0

Re: [ros-dev] [ros-diffs] [dgorbachev] 27174: - Hackish solution of issue #2062. - Turn off title change, as it can cause a deadlock.
by Hervé Poussineau 15 Jun '07

15 Jun '07

Hi, > -----Message d'origine----- > De : ros-diffs-bounces(a)reactos.org > [mailto:ros-diffs-bounces@reactos.org] De la part de > dgorbachev(a)svn.reactos.org > Envoyé : jeudi 14 juin 2007 18:42 > À : ros-diffs(a)reactos.org > Objet : [ros-diffs] [dgorbachev] 27174: - Hackish solution of > issue #2062. - Turn off title change, as it can cause a deadlock. > > Author: dgorbachev > Date: Thu Jun 14 20:41:55 2007 > New Revision: 27174 > > URL: http://svn.reactos.org/svn/reactos?rev=27174&view=rev > Log: > - Hackish solution of issue #2062. > - Turn off title change, as it can cause a deadlock. > > Modified: > trunk/reactos/subsystems/win32/csrss/win32csr/conio.c > trunk/reactos/subsystems/win32/csrss/win32csr/dllmain.c > trunk/reactos/subsystems/win32/csrss/win32csr/guiconsole.c > Why hiding the real cause of bug 2062 by a hack, as you already did in r27107 for bug 2151? See desktop icons in explorer bar. They are not correct and I know when it appeared and I known how to hack a solution. But I won't do it because I don't want to hide the real problem. IMO, you should do the same thing and search the root problem or leave the effect visible, so it won't be forgotten. If you really need to add a hack in the code, please add a comment in the C code linking to bugzilla, and add a comment in BZ saying "hackish solution committed in rXXXXX" Hervé

2 1

(no subject)
by Cam 15 Jun '07

15 Jun '07

Way to go devs! Device manager finally works!

1 0

Re: [ros-dev] Smaller SVN REL ISO?
by carlo.bramix 11 Jun '07

11 Jun '07

Unfortunately, I'm not able to compile myself. So I download the SVN ISO images for testing. I was just wondering if it could be possible to reduce a bit my download times by including that flag into the REL builds. I also agree that it's better to leave the DEB builds without that flag. Thanks anyways, if it's not possible I can still download compressed ISO files at their current sizes. Sincerely, Carlo Bramini ---------- Initial Header ----------- >From : ros-dev-bounces(a)reactos.org To : "ReactOS Development List" ros-dev(a)reactos.org Cc : Date : Mon, 11 Jun 2007 14:28:37 +0400 Subject : Re: [ros-dev] Smaller SVN REL ISO? > It is possible to compile with this flag, but you need to build it > yourself then. Official ISOs are compiled with leanandmean set to no. > > > WBR, > Aleksey Bragin. > > On Jun 11, 2007, at 1:06 PM, carlo.bramix wrote: > > > Is the SVN REL Live CD compiled with "set ROS_LEAN_AND_MEAN = yes"? > > If not, is it possible to compile it with this flag? > > For other purposes beside the normal user testing, we have the SVN > > DEB Live CD image. > > > > Sincerely, > > > > LDChen > _______________________________________________ > Ros-dev mailing list > Ros-dev(a)reactos.org > http://www.reactos.org/mailman/listinfo/ros-dev > ------------------------------------------------------ Leggi GRATIS le tue mail con il telefonino i-mode di Wind http://i-mode.wind.it/

1 0

Re: [ros-dev] Smaller SVN REL ISO?
by Aleksey Bragin 11 Jun '07

11 Jun '07

It is possible to compile with this flag, but you need to build it yourself then. Official ISOs are compiled with leanandmean set to no. WBR, Aleksey Bragin. On Jun 11, 2007, at 1:06 PM, carlo.bramix wrote: > Is the SVN REL Live CD compiled with "set ROS_LEAN_AND_MEAN = yes"? > If not, is it possible to compile it with this flag? > For other purposes beside the normal user testing, we have the SVN > DEB Live CD image. > > Sincerely, > > LDChen

1 0

Smaller SVN REL ISO?
by carlo.bramix 11 Jun '07

11 Jun '07

Is the SVN REL Live CD compiled with "set ROS_LEAN_AND_MEAN = yes"? If not, is it possible to compile it with this flag? For other purposes beside the normal user testing, we have the SVN DEB Live CD image. Sincerely, LDChen ------------------------------------------------------ Leggi GRATIS le tue mail con il telefonino i-mode di Wind http://i-mode.wind.it/

1 0

Re: [ros-dev] [ros-diffs] [dgorbachev] 27107: Fix issue #2151.
by Thomas Weidenmueller 11 Jun '07

11 Jun '07

Do you even know what you're doing? I've seen many commits where you call something a FIX and then just comment code out and add another FIXME.... dgorbachev(a)svn.reactos.org wrote: > Author: dgorbachev > Date: Sun Jun 10 15:11:05 2007 > New Revision: 27107 > > URL: http://svn.reactos.org/svn/reactos?rev=27107&view=rev > Log: > Fix issue #2151. > > Modified: > trunk/reactos/ntoskrnl/mm/sysldr.c > > Modified: trunk/reactos/ntoskrnl/mm/sysldr.c > URL: http://svn.reactos.org/svn/reactos/trunk/reactos/ntoskrnl/mm/sysldr.c?rev=2… > ============================================================================== > --- trunk/reactos/ntoskrnl/mm/sysldr.c (original) > +++ trunk/reactos/ntoskrnl/mm/sysldr.c Sun Jun 10 15:11:05 2007 > @@ -1767,7 +1767,7 @@ > NtHeader->OptionalHeader.AddressOfEntryPoint); > LdrEntry->SizeOfImage = DriverSize; > LdrEntry->CheckSum = NtHeader->OptionalHeader.CheckSum; > - LdrEntry->SectionPointer = LdrEntry; > + LdrEntry->SectionPointer = NULL; // ? > > /* Now write the DLL name */ > LdrEntry->BaseDllName.Buffer = (PVOID)(LdrEntry + 1); > > >

2 1

Re: [ros-dev] [ros-diffs] [dgorbachev] 26483: Undo r26482 change, waste more stack space.
by Alex Ionescu 07 Jun '07

07 Jun '07

Perhaps you could try convincing the GCC folks to actually properly support declspec_align like MSVC does, instead of claiming to support it, but actually ignoring it. Then you could also convince them to support SEH, PDBs, STDCALL-default calling convention, etc ;) -----Original Message----- From: ros-diffs-bounces(a)reactos.org [mailto:ros-diffs-bounces@reactos.org] On Behalf Of dgorbachev(a)svn.reactos.org Sent: April-24-07 1:06 AM To: ros-diffs(a)reactos.org Subject: [ros-diffs] [dgorbachev] 26483: Undo r26482 change, waste more stack space. Author: dgorbachev Date: Tue Apr 24 09:05:49 2007 New Revision: 26483 URL: http://svn.reactos.org/svn/reactos?rev=26483&view=rev Log: Undo r26482 change, waste more stack space. Modified: trunk/reactos/ntoskrnl/include/internal/ex.h Modified: trunk/reactos/ntoskrnl/include/internal/ex.h URL: http://svn.reactos.org/svn/reactos/trunk/reactos/ntoskrnl/include/internal/… ============================================================================== --- trunk/reactos/ntoskrnl/include/internal/ex.h (original) +++ trunk/reactos/ntoskrnl/include/internal/ex.h Tue Apr 24 09:05:49 2007 @@ -106,9 +106,15 @@ // #ifdef __GNUC__ -#define DEFINE_WAIT_BLOCK(x) \ - PEX_PUSH_LOCK_WAIT_BLOCK x = __builtin_alloca(sizeof(EX_PUSH_LOCK_WAIT_BLOCK)); - +#define DEFINE_WAIT_BLOCK(x) \ + struct _AlignHack \ + { \ + UCHAR Hack[15]; \ + EX_PUSH_LOCK_WAIT_BLOCK UnalignedBlock; \ + } WaitBlockBuffer; \ + PEX_PUSH_LOCK_WAIT_BLOCK x = (PEX_PUSH_LOCK_WAIT_BLOCK) \ + ((ULONG_PTR)&WaitBlockBuffer.UnalignedBlock &~ 0xF); + #else // @@ -116,8 +122,8 @@ // local variable (the actual pointer) away, so we don't take any perf hit // by doing this. // -#define DEFINE_WAIT_BLOCK(x) \ - EX_PUSH_LOCK_WAIT_BLOCK WaitBlockBuffer; \ +#define DEFINE_WAIT_BLOCK(x) \ + EX_PUSH_LOCK_WAIT_BLOCK WaitBlockBuffer; \ PEX_PUSH_LOCK_WAIT_BLOCK x = &WaitBlockBuffer; #endif

11 19

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Ros-dev June 2007