Ros-dev

ros-dev@reactos.org

5789 discussions

Re: [ros-dev] [ros-diffs] [fireball] 36426: [FORMATTING] - Reformat to the kernel coding style.

by Alex Ionescu

Please show me where the developers of this code agreed to having their code GPL V3 licensed. I would like to see a full trail of every developer that wrote this code, as well as written permission from them for you to slap on this license. Thank you. On 23-Sep-08, at 7:45 AM, fireball(a)svn.reactos.org wrote: > + * LICENSE: GPL v2 or later - See COPYING in the top level > directory Best regards, Alex Ionescu

16 years, 7 months

Re: [ros-dev] Reformat to the kernel coding style.

by chris evans

Pasting gpl notices in every file of a project, takes up space. better to put in license.txt in project root to encompass all source files. but most that don't like a license will refuse to use it and just copy paste the code anyways, and not say where they got it. (happens all the time). kinda makes me wonder why we got licenses for software in the first place, seems to be silly constructs of obstructionism. GPLv3 don't allow you to control the software that runs on the custom hardware, inability to link to the code with out upsetting the original author. ce From: James Tabor <jimtabor.rosdev(a)gmail.com> To: ReactOS Development List <ros-dev(a)reactos.org> Sent: Saturday, October 4, 2008 12:38:23 AM Subject: Re: Reformat to the kernel coding style. 1st: No to v3! 2nd: No to v3! 3rd: No to v3! On Fri, Oct 3, 2008 at 11:44 PM, Alex Ionescu <ionucu(a)videotron.ca> wrote: > You basically assumed that a header from another file (probably copy- > pasted) applies to the file you modified. > > While I'm certainly being pedantic, and I admit it -- this shouldn't > be a change to brush over. > > Someone needs to figure out if the devs agree with v3 or not -- I

16 years, 7 months

[ros-diffs] [sginsberg] 36624: - Add missing APIENTRY to ntgdi.h, syscalls are always stdcall.

by James Tabor

Hi! Something looks wrong with this patch..... >From the latest SDK I have Version 6. __kernel_entry W32KAPI BOOL NtGdiUpdateTransform( __in HDC hdc ); __kernel_entry W32KAPI BOOL NtGdiSetPUMPDOBJ( __in_opt HUMPD humpd, __in BOOL bStoreID, __inout_opt HUMPD *phumpd, __out_opt BOOL *pbWOW64 ); __kernel_entry W32KAPI BOOL NtGdiBRUSHOBJ_DeleteRbrush( __in_opt BRUSHOBJ *pbo, __in_opt BRUSHOBJ *pboB ); __kernel_entry W32KAPI BOOL NtGdiUMPDEngFreeUserMem( __in KERNEL_PVOID *ppv ); Thanks, James

16 years, 7 months

Re: [ros-dev] [ros-diffs] [hyperion] 36610: modified ke/i386/cpu.c Added CMPXCHG8B enabling/detection code for TransMeta, Centaur and Rise (source: <URL: http://www.geoffchappell.com/notes/windows/kernel/cpu/cx8.htm>). Code dead until each vendor will be officially supported Alex and others: please review

by Aleksey Bragin

Why wrap into #ifdefs? I thought it's supposed to be runtime, not compile time. WBR, Aleksey. On Oct 1, 2008, at 3:54 PM, hyperion(a)svn.reactos.org wrote: > Author: hyperion > Date: Wed Oct 1 06:54:29 2008 > New Revision: 36610 > > URL: http://svn.reactos.org/svn/reactos?rev=36610&view=rev > Log: > modified ke/i386/cpu.c > Added CMPXCHG8B enabling/detection code for TransMeta, Centaur > and Rise (source: <URL: http://www.geoffchappell.com/notes/windows/ > kernel/cpu/cx8.htm>). Code dead until each vendor will be > officially supported > Alex and others: please review

16 years, 7 months

CMPXHG8B Support

by Alex Ionescu

I was asked on IRC to add support for non-Intel and AMD CPUs to the kernel. Obviously I can't do that -- but I mentioned that some special "magic tricks" need to be done to enable support for the instruction on other CPUs, which is why I made the kernel bugcheck. Thankfully, someone documented the steps required, so I won't get in trouble for pointing out this link for you: http://www.geoffchappell.com/notes/windows/kernel/cpu/cx8.htm. I can confirm that's what the kernel does. Do that too, and you can then support Transmeta/Rise/VIA CPUs. Best regards, Alex Ionescu

16 years, 7 months

Re: [ros-dev] [ros-diffs] [jimtabor] 36596: - Patch by Vincent Povirk, Handle undocumented WM_POPUPSYSTENNEBU message.

by Alex Ionescu

James, if it's undocumented, please don't add it to the SDK headers. On 29-Sep-08, at 8:27 PM, jimtabor(a)svn.reactos.org wrote: > WM_POPUPSYSTEMMENU Best regards, Alex Ionescu

16 years, 7 months

Re: [ros-dev] [ros-diffs] [hyperion] 36586: modified dll/win32/kernel32/kernel32.def modified dll/win32/kernel32/kernel32.rbuild added dll/win32/kernel32/misc/icustubs.cpp Export NormalizeString and IsStringNormalized from kernel32.dll. This (finally) makes normaliz.dll (50%) functional. Link kernel32.dll to ICU. Finally. Binary size increased by about 300 KB on a debug build (this is only the beginning). Umpteenth copy of C++ and Win32 stubs required to get ICU to link properly added to kernel32.dll. This

by Aleksey Bragin

It's going (if not already) to overflow the baseaddress range given to it, thus making ntdll.dll relocated, which leads to obvious problems. WBR, Aleksey. On Sep 29, 2008, at 2:39 AM, hyperion(a)svn.reactos.org wrote: > Author: hyperion > Date: Sun Sep 28 17:39:45 2008 > New Revision: 36586 > > URL: http://svn.reactos.org/svn/reactos?rev=36586&view=rev > Log: > modified dll/win32/kernel32/kernel32.def > modified dll/win32/kernel32/kernel32.rbuild > added dll/win32/kernel32/misc/icustubs.cpp > Export NormalizeString and IsStringNormalized from kernel32.dll. > This (finally) makes normaliz.dll (50%) functional. > Link kernel32.dll to ICU. Finally. Binary size increased by > about 300 KB on a debug build (this is only the beginning). > Umpteenth copy of C++ and Win32 stubs required to get ICU to > link properly added to kernel32.dll. > This commit dedicated to Timo Kreuzer. <3.

16 years, 7 months

Re: [ros-dev] [ros-diffs] [cgutman] 36580: - Raise IRQL to DISPATCH_LEVEL before using KeAcquireSpinLockAtDpcLevel/KeReleaseSpinLockFromDpcLevel

by Aleksey Bragin

Is it really supposed to be so? Why don't you just use KeAcquireSpinlock, which would save you the current IRQL level? For me it looks like a hack. WBR, Aleksey Bragin. On Sep 28, 2008, at 11:58 PM, cgutman(a)svn.reactos.org wrote: > Author: cgutman > Date: Sun Sep 28 14:58:09 2008 > New Revision: 36580 > > URL: http://svn.reactos.org/svn/reactos?rev=36580&view=rev > Log: > - Raise IRQL to DISPATCH_LEVEL before using > KeAcquireSpinLockAtDpcLevel/KeReleaseSpinLockFromDpcLevel > > Modified: > branches/aicom-network-fixes/drivers/network/ndis/ndis/protocol.c > > Modified: branches/aicom-network-fixes/drivers/network/ndis/ndis/ > protocol.c > URL: http://svn.reactos.org/svn/reactos/branches/aicom-network- > fixes/drivers/network/ndis/ndis/protocol.c? > rev=36580&r1=36579&r2=36580&view=diff > ====================================================================== > ======== > --- branches/aicom-network-fixes/drivers/network/ndis/ndis/ > protocol.c [iso-8859-1] (original) > +++ branches/aicom-network-fixes/drivers/network/ndis/ndis/ > protocol.c [iso-8859-1] Sun Sep 28 14:58:09 2008 > @@ -67,6 +67,7 @@ > { > UINT BufferedLength; > UINT PacketLength; > + KIRQL OldIrql; > > NDIS_DbgPrint(MAX_TRACE, ("Called.\n")); > > @@ -75,6 +76,8 @@ > #endif > > NdisQueryPacket(Packet, NULL, NULL, NULL, &PacketLength); > + > + KeRaiseIrql(DISPATCH_LEVEL, &OldIrql); > > NDIS_DbgPrint(MAX_TRACE, ("acquiring miniport block lock\n")); > KeAcquireSpinLockAtDpcLevel(&Adapter->NdisMiniportBlock.Lock); > @@ -102,6 +105,8 @@ > Adapter->NdisMiniportBlock.IndicatedPacket > [KeGetCurrentProcessorNumber()] = NULL; > } > KeReleaseSpinLockFromDpcLevel(&Adapter->NdisMiniportBlock.Lock); > + > + KeLowerIrql(OldIrql); > > return STATUS_SUCCESS; > } >

16 years, 7 months

VFS implementation

by മഹേഷ് മുകുന്ദന് | Mahesh M

Hi, I am doing a project on "Implementation of VFS in ReactOS". What I am doing is extract the VFS layer from Linux Kernel and create a driver to ROS kernel. I am aware of the Installable FS but I feel that VFS is a better option. I would like to request for suggestions and help on the same. -- Mahesh M Happy hacking... , , / \ ((__-^^-,-^^-__)) `-_---' `---_-' `--|o` 'o|--' \ ` / ): :( :o_o: "-"

16 years, 7 months

Re: [ros-dev] [ros-diffs] [pschweitzer] 36539: Get rid off recursive FsRtlIsNameInExpression once again. Fix a typo in iterative function that fixes it properly. We now get ReactOS up to 2nd stage with that fix. 2nd stage is broken...

by Aleksey Bragin

That's probably because FsRtlIsNameInExpression is not really used that much in the 1st stage, but in the 2nd stage it's used to load fonts (*.ttf from the media fonts, resulting in quite a few calls to this FsRtl function). WBR, Aleksey. On Sep 27, 2008, at 12:05 AM, pschweitzer(a)svn.reactos.org wrote: > Author: pschweitzer > Date: Fri Sep 26 14:50:39 2008 > New Revision: 36539 > > URL: http://svn.reactos.org/svn/reactos?rev=36539&view=rev > Log: > Get rid off recursive FsRtlIsNameInExpression once again. > Fix a typo in iterative function that fixes it properly. We now get > ReactOS up to 2nd stage with that fix. 2nd stage is broken...

16 years, 7 months

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