Ros-dev

ros-dev@reactos.org

5788 discussions

Re: [ros-dev] [ros-diffs] [pschweitzer] 60871: [NTOSKRNL] - Implement MiUnlockProcessWorkingSetShared() and make use of it in MiUnlockProcessWorkingSetForFault(). - Complete implementation of MiLockProcessWorkingSetForFault()

by Alex Ionescu

What JIRA task is this associated with? -----Original Message----- From: ros-diffs-bounces(a)reactos.org [mailto:ros-diffs-bounces@reactos.org] On Behalf Of pschweitzer(a)svn.reactos.org Sent: Tuesday, November 5, 2013 2:04 PM To: ros-diffs(a)reactos.org Subject: [ros-diffs] [pschweitzer] 60871: [NTOSKRNL] - Implement MiUnlockProcessWorkingSetShared() and make use of it in MiUnlockProcessWorkingSetForFault(). - Complete implementation of MiLockProcessWorkingSetForFault() Author: pschweitzer Date: Tue Nov 5 22:03:42 2013 New Revision: 60871 URL: http://svn.reactos.org/svn/reactos?rev=60871&view=rev Log: [NTOSKRNL] - Implement MiUnlockProcessWorkingSetShared() and make use of it in MiUnlockProcessWorkingSetForFault(). - Complete implementation of MiLockProcessWorkingSetForFault() Modified: trunk/reactos/ntoskrnl/mm/ARM3/miarm.h Modified: trunk/reactos/ntoskrnl/mm/ARM3/miarm.h URL: http://svn.reactos.org/svn/reactos/trunk/reactos/ntoskrnl/mm/ARM3/miarm.h?r… ============================================================================== --- trunk/reactos/ntoskrnl/mm/ARM3/miarm.h [iso-8859-1] (original) +++ trunk/reactos/ntoskrnl/mm/ARM3/miarm.h [iso-8859-1] Tue Nov 5 22:03:42 2013 @@ -1197,6 +1197,30 @@ // FORCEINLINE VOID +MiUnlockProcessWorkingSetShared(IN PEPROCESS Process, + IN PETHREAD Thread) { + /* Make sure we are the owner of a safe acquisition (because shared) */ + ASSERT(MI_WS_OWNER(Process)); + ASSERT(!MI_IS_WS_UNSAFE(Process)); + + /* Ensure we are in a shared acquisition */ + ASSERT(Thread->OwnsProcessWorkingSetShared == TRUE); + ASSERT(Thread->OwnsProcessWorkingSetExclusive == FALSE); + + /* Don't claim the lock anylonger */ + Thread->OwnsProcessWorkingSetShared = FALSE; + + /* Release the lock and re-enable APCs */ + ExReleasePushLockShared(&Process->Vm.WorkingSetMutex); + KeLeaveGuardedRegion(); +} + +// +// Unlocks the working set for the given process // FORCEINLINE VOID MiUnlockProcessWorkingSetUnsafe(IN PEPROCESS Process, IN PETHREAD Thread) { @@ -1330,7 +1354,7 @@ else { /* Owner is shared (implies safe), release normally */ - ASSERT(FALSE); + MiUnlockProcessWorkingSetShared(Process, Thread); *Safe = TRUE; *Shared = TRUE; } @@ -1343,16 +1367,24 @@ IN BOOLEAN Safe, IN BOOLEAN Shared) { - ASSERT(Shared == FALSE); - /* Check if this was a safe lock or not */ if (Safe) { - /* Reacquire safely */ - MiLockProcessWorkingSet(Process, Thread); + if (Shared) + { + /* Reacquire safely & shared */ + MiLockProcessWorkingSetShared(Process, Thread); + } + else + { + /* Reacquire safely */ + MiLockProcessWorkingSet(Process, Thread); + } } else { + /* Unsafe lock cannot be shared */ + ASSERT(Shared == FALSE); /* Reacquire unsafely */ MiLockProcessWorkingSetUnsafe(Process, Thread); }

10 years, 10 months

Re: [ros-dev] [ros-diffs] [dgorbachev] 60832: [CMAKE] Reduce the disk space needed to build ReactOS (disabled for LTCG builds - linker bug #15143).

by Aleksey Bragin

Nice to see you committing, hto! Regards, Aleksey Bragin On 02.11.2013 17:56, dgorbachev(a)svn.reactos.org wrote: > Author: dgorbachev > Date: Sat Nov 2 13:56:47 2013 > New Revision: 60832 > > URL: http://svn.reactos.org/svn/reactos?rev=60832&view=rev > Log: > [CMAKE] > Reduce the disk space needed to build ReactOS (disabled for LTCG builds - linker bug #15143). > > Modified: > trunk/reactos/toolchain-gcc.cmake >

10 years, 10 months

Re: [ros-dev] [ros-diffs] [pschweitzer] 60802: [NTOSKRNL] Implement MiSetControlAreaSymbolsLoaded and MiLoadUserSymbols. These are responsible for marking the control area that we've loaded the symbols and for effectively l...

by Timo Kreuzer

Am 31.10.2013 00:17, schrieb pschweitzer(a)svn.reactos.org: > +VOID > +NTAPI > +MiSetControlAreaSymbolsLoaded(IN PCONTROL_AREA ControlArea) > +{ > + KIRQL OldIrql; > + > + ASSERT(KeGetCurrentIrql() <= APC_LEVEL); > + > + OldIrql = KeAcquireQueuedSpinLock(LockQueuePfnLock); > + ControlArea->u.Flags.DebugSymbolsLoaded |= 1; > + > + ASSERT(OldIrql <= APC_LEVEL); > + KeReleaseQueuedSpinLock(LockQueuePfnLock, OldIrql); > + ASSERT(KeGetCurrentIrql() <= APC_LEVEL); > +} No offence, but seriously? The PFN lock to set a flag?

10 years, 10 months

Status Meeting (October 2013)

by Aleksey Bragin

Hello, Let me invite you to the monthly status meeting taking place last Thursday of this month, 31st of October, 19:00 UTC. Put that into your calendars so you don't forget. That's the last day of this month too and that's tomorrow! IRC service will only be started shortly before the meeting. Your participation passwords and server address will be emailed to you shortly before the meeting starts, and they are going to be different once again as they are not stored in any database. Hopefully it's not much of inconvenience. If someone still is not getting passwords sent before a meeting - please email Pierre or Colin (depending who will handle this meeting) before the meeting started to get one. The agenda will be posted shortly before the meeting, suggestions are welcome (send them to me shortly before the meeting starts). Regards, Aleksey Bragin

10 years, 10 months

Servers unexpected downtime

by Pierre Schweitzer

Hi, North of Europa is facing a major storm. Currently, it is hitting Sweden where it is really violent. Unfortunately, some of our servers are there as well. Our datacenter has been affected (likely power issue) and two servers are currently offline and cannot be rebooted remotely. Our local sysadmin cannot make a move to the datacenter following the authorities recommandations to stay at home. So far, the following parts of our infrastructure are offline: -> Doxygen -> cppcheck -> ISO download -> ISO upload -> VMware testbots As a side effect, I will take the Linux builder & KVM testbot offline given it is impossible to upload any ISO at the moment. We will do our best to get the infrastructure back online as fast as possible. We will keep you informed. We are sorry for the caused inconvenience. With my best regards, -- Pierre Schweitzer<pierre at reactos.org> System Administrator ReactOS Foundation

10 years, 10 months

Building and installing ReactOS on newer computer

by Thomas Mueller

How can I build and install ReactOS on a new computer? Can I install to a USB stick or to a 1.2 GB IDE hard drive in a USB 2.0 hard-drive enclosure? Main hard drive is Western Digital Green 3 TB, GPT-partitioned. I would have to build from FreeBSD or less likely NetBSD or Linux. I downloaded the installation ISO for ReactOS 0.3.15, but it failed to boot, also failed to boot on an old computer. I already have subversion (svn), built from FreeBSD ports. Tom

10 years, 10 months

Re: [ros-dev] Building and installing ReactOS on newer computer

by Thomas Mueller

> [-- Type: multipart/alternative, Encoding: 7bit, Size: 3.0K --] > Content-Type: multipart/alternative; boundary=001a11c3c7dccc447f04e913cb34 > Content-Type: text/plain; charset=UTF-8 > You can find ready-to-use builds of recent SVN revisions at: > https://www.reactos.org/getbuilds > If you don't see builds for the latest, use the arrows to go back until you > find one that has files. > You can NOT install ReactOS on a USB hard drive, though. > If you do want to build from source regardless, you should look at RosBE, > although I'm not sure if it works well in FreeBSD, since I build from > Windows. Thanks for response. Linux and the BSDs are not so fussy about where they are installed on, USB stick is OK. But I remember DOS, Windows and OS/2 were fussy. Do you use MinGW in Windows? MinGW, subversion, ninja and GCC are all available in FreeBSD ports collection. >From what I read, I think ReactOS might not get along with USB 3.0 and GPT, hence the desire to use a USB stick. FreeDOS also is difficult to install, many failed attempts on my computer perhaps related to modern hardware. FAT32 file system is not really good for making use of today's big hard drives. But keeping up with NTFS is difficult because NTFS is a moving target (I believe). I really feel more at home now with Linux and FreeBSD than with DOS, Windows or OS/2. Tom

10 years, 11 months

Re: [ros-dev] [ros-diffs] [aandrejevic] 60757: [SOFTX86] Remove softx86, as it is no longer used by anything.

by Jérôme Gardou

Hello. Just nitpicking a bit: >> -target_link_libraries(ntvdm softx86 softx87 fast486) >> +target_link_libraries(ntvdm fast486) >> add_importlibs(ntvdm msvcrt user32 gdi32 kernel32 ntdll) >> -add_dependencies(ntvdm softx86 softx87) >> +add_dependencies(ntvdm fast486) The call to add_dependencies is useless here, as a call to target_link_libraries adds an implicit dependency. Regards Jérôme

10 years, 11 months

Re: [ros-dev] [ros-diffs] [jimtabor] 60683: [Win32ss] - Prevent a potential kernel exception.

by Alex Ionescu

Jim, this and your other patch can never happen. It's ListEntry that can be NULL, but Property will never be (in fact, it will be negative number). Also, ListEntry should never be NULL anyway, otherwise it means the list head is corrupted. Containing record is just math...not a dereference. -- Best regards, Alex Ionescu -----Original Message----- From: ros-diffs-bounces(a)reactos.org [mailto:ros-diffs-bounces@reactos.org] On Behalf Of jimtabor(a)svn.reactos.org Sent: Tuesday, October 15, 2013 3:17 PM To: ros-diffs(a)reactos.org Subject: [ros-diffs] [jimtabor] 60683: [Win32ss] - Prevent a potential kernel exception. Author: jimtabor Date: Tue Oct 15 22:16:31 2013 New Revision: 60683 URL: http://svn.reactos.org/svn/reactos?rev=60683&view=rev Log: [Win32ss] - Prevent a potential kernel exception. Modified: trunk/reactos/win32ss/user/ntuser/prop.c Modified: trunk/reactos/win32ss/user/ntuser/prop.c URL: http://svn.reactos.org/svn/reactos/trunk/reactos/win32ss/user/ntuser/prop.c… ============================================================================== --- trunk/reactos/win32ss/user/ntuser/prop.c [iso-8859-1] (original) +++ trunk/reactos/win32ss/user/ntuser/prop.c [iso-8859-1] Tue Oct 15 22:16:31 2013 @@ -21,6 +21,7 @@ while (ListEntry != &Window->PropListHead) { Property = CONTAINING_RECORD(ListEntry, PROPERTY, PropListEntry); + if (!Property) break; if (Property->Atom == Atom) { return(Property);

10 years, 11 months

Re: [ros-dev] [ros-diffs] [jimtabor] 60682: [Win32ss] - Prevent a potential kernel exception.

by Thomas Faber

On 2013-10-16 00:14, jimtabor(a)svn.reactos.org wrote: > URL: http://svn.reactos.org/svn/reactos/trunk/reactos/win32ss/user/user32/window… > ============================================================================== > --- trunk/reactos/win32ss/user/user32/windows/prop.c [iso-8859-1] (original) > +++ trunk/reactos/win32ss/user/user32/windows/prop.c [iso-8859-1] Tue Oct 15 22:14:48 2013 > @@ -53,6 +53,7 @@ > for (i = 0; i < pWnd->PropListItems; i++ ) > { > Property = CONTAINING_RECORD(ListEntry, PROPERTY, PropListEntry); > + if (!Property) break; > if (Property->Atom == Atom) > { > return(Property); > > That can't happen unless the list is corrupt. And even if the list contains a null pointer (which it's not supposed to), ListEntry would be NULL, not Property.

10 years, 11 months

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Ros-dev