Hi,
How hard would it be to add the url pointing to the diff for each commit message? I don't want to
have to subscribe to another mailing list or have to go hunting in websvn if I want to view the
diff of a svn commit.
Thanks
Steven
__________________________________
Do you Yahoo!?
Yahoo! Mail - now with 250MB free storage. Learn more.
http://info.mail.yahoo.com/mail_250
After starting up Shell=Explorer and crashing the seemingly
not-enough-(P)SEH-using Explorer (trying to view the RPC Control branch of
the NT Objects namespace is a sure way), I get the following. Disregarding
the crashing Explorer that I only included for completeness, could this be
related to the recent work of having a separate thread doing the shutdown?
(KERNEL32:except/except.c:132) Unhandled exception
(KERNEL32:except/except.c:133) Address:
(KERNEL32:except/except.c:136) 44a912 C:\ReactOS\explorer.exe
CS:EIP 1b:44a912
DS 23 ES 23 FS 3b GS 23
EAX: 00000000 EBX: 000001dc ECX: 007af1fa
EDX: 00000250 EBP: 007ae448 ESI: 007afb1c ESP: 007ae440
EDI: 007afa78 EFLAGS: 00000202
(KERNEL32:except/except.c:139) Frames:
(KERNEL32:except/except.c:147) 421b2c C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 40f5cc C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 4170b4 C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 415871 C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 416c86 C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 44ffb5 C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 77e71ef4 C:\ReactOS\system32\user32.dll
(KERNEL32:except/except.c:147) 77e72833 C:\ReactOS\system32\user32.dll
(KERNEL32:except/except.c:147) 77ea5934 C:\ReactOS\system32\user32.dll
(KERNEL32:except/except.c:147) 77ea7604 C:\ReactOS\system32\user32.dll
(KERNEL32:except/except.c:147) 77ea7839 C:\ReactOS\system32\user32.dll
(KERNEL32:except/except.c:147) 77e71ef4 C:\ReactOS\system32\user32.dll
(KERNEL32:except/except.c:147) 77e72059 C:\ReactOS\system32\user32.dll
(KERNEL32:except/except.c:147) 450459 C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 422e9d C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 450404 C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 77e71ef4 C:\ReactOS\system32\user32.dll
(KERNEL32:except/except.c:147) 77e72152 C:\ReactOS\system32\user32.dll
(KERNEL32:except/except.c:147) 4518b5 C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 40447b C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 404bcd C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 40460c C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 40120d C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 401258 C:\ReactOS\explorer.exe
(KERNEL32:except/except.c:147) 7c8279c1
C:\ReactOS\system32\kernel32.dll
(ntuser/class.c:114) Failed to lookup class atom!
(ntuser/class.c:114) Failed to lookup class atom!
(ntuser/class.c:114) Failed to lookup class atom!
(ntuser/class.c:114) Failed to lookup class atom!
(ntuser/class.c:114) Failed to lookup class atom!
(ntuser/class.c:114) Failed to lookup class atom!
(smss.c:90) SM: NtWaitForMultipleObjects failed!
(ke/error.c:51) Hard error c000021a
(api/wapi.c:175) CSR: NtListenPort() failed
(ke/catch.c:149) Unhandled UserMode exc(api/wapi.c:120) CSR:
NtReplyWaitReceivePort failed
eption, terminating thread
(api/wapi.c:120) CSR: NtReplyWaitReceivePort failed
(api/wapi.c:120) CSR: Nt(api/wapi.c:120) CSR: NtReplyWaitReceivePort failed
ReplyWaitReceivePort failed
ex/power.c:91
--
/Mike
Safedisc installs a kernel service called secdrv.sys. As wine is a
user-space
only implementation of the win32 API, we need to implement part of the
windows kernel API (ntoskrnl.exe) in user-space, and emulate what can't be
implemented.
Our biggest problem is the I/O Manager. The kernel services is started by
StartService, which currently only supports user-space services. We need to
understand how kernel services work in windows to fix that.
When a kernel service is started, it's DriverEntry routine is called with a
DRIVER_OBJECT. In which thread does that happen ? Is a new kernel thread
started for that ? Can DriverEntry block or must it return immediately ? If
it can block, can the driver dispatch routines be called before it
returns ?
When it returns, if it has been started in a new kernel thread, what
happens
to the thread ? In which thread do the dispatch routines run when they are
called from user-space ?
The MSDN documentation is not very clear about all that. I suppose it
hasn't
been written with windows cloners in mind :-)
Devs,
I installed the win32 vim package from vim.org back around
0.2.3-0.2.4, and was quite surprised when the console vim.exe mostly
worked. Recently, while testing PuTTY, I tried the command line tools
(psftp specifically) and they would not accept input. I could ^C out
of it and then all the input I had typed showed up on the command
line. So, I tried the console vim again and it doesn't accept input
also. I checked both vim.exe and the command line putty tools under
0.2.4 and 0.2.5-rc2, and they both work, so it is something after the
0.2.5 branch.
I'm reporting this to the ML because I'm not very familiar with
the console code. Any ideas or pointers would be appreciated.
Andy
--
The cheese stands alone.
> From: hyperion(a)svn.reactos.com
>
> honor __USE_W32API
>
>
> Updated files:
> trunk/reactos/include/tchar.h
I made the same change about half a year ago, but it turns out that it
breaks compilation on Linux.
Gé van Geldorp.
Hi,
Most of the developers I've talked to (including myself) want to get rid
of the /nt directory, since it is architecturally incorrect and should
instead be migrated into other parts of the kernel. Since the nt
functions that were in the directory were part of the Executive (note
the ExXXXObjectType), I've placed the functions in \ex, and renamed all
Ntp stuff by Exp, with the exception of some LPC stuff which I've
renamed from Ni to Lpcp. profiles were harder, since both the Executive
and Kernel parts were in the nt file (actually we're not implementing
the Executive part properly imo, we don't even have EPROFILE, but that's
another story). Anyways, attached is the patch so you can see what was
changed and where it was displaced. I don't think anyone wants to keep
the /nt directory, based on previous discussions, but I want to make
sure people will be aware of it.
Best regards,
Alex Ionescu
Hello,
I got this while compiling svn tree with DBG=1, gcc-3.4.1 :
mingw32-gcc -Wall -Werror -D__USE_W32API -I. -I../../../include
-I../../../w32api/include -pipe -march=i486 -D_M_IX86 -g -c tickcount.c
-o tickcount.o
tickcount.c: In function `parse_print':
tickcount.c:143: warning: implicit declaration of function `_ttoi64'
Thanks, David
I installed the SVN and could make it work, i
downloaded MinGW-3.1.0-1.exe ,gcc-core 3.3.3 , gcc-g++
3.3.3 and nasm-0.98-win32.zip Installed them and set
the path.
i tried to build ReactOs (at the reactos folder) with
make (I did previously a make clean)
And i get the following error:
-----------------------------------------------
...win32k: [DLLTOOL] libwin32k.a
win32csr: [DLLTOOL] libwin32csr.a
notifyhook: [DLLTOOL] libnotifyhook.a
regtests: [DLLTOOL] libregtests.a
=====================================================
Compiling deptool
=====================================================
Compiling bin2c
Tools are up to date.
Tools are up to date.
=====================================================
Assembling dosmbr
=====================================================
Assembling fat
=====================================================
Assembling fat32
=====================================================
Assembling isoboot
=====================================================
Assembling ext2
=====================================================
Assembling arch/i386/fathe
lp
=====================================================
Assembling arch/i386/arch
=====================================================
Assembling arch/i386/i386i
dt
=====================================================
Assembling arch/i386/i386t
rap
=====================================================
Assembling arch/i386/i386c
pu
=====================================================
Assembling arch/i386/i386p
np
=====================================================
Assembling arch/i386/boot
=====================================================
Assembling arch/i386/linux
=====================================================
Assembling arch/i386/mb
=====================================================
Compiling arch/i386/i386rt
l
cc1.exe: unrecognized option
`-fno-zero-initialized-in-bss'
make[2]: *** [arch/i386/i386rtl.o] Error 1
make[1]: *** [all] Error 2
make: *** [freeldr] Error 2
C:\Internet\ReactOS\Subersion\trunk\reactos>
------------------------------------------------
Nevertheless i try to make install and get the
following error:
---------------------------------------
expat: [OBJDUMP] expat.map
expat: [LD] expat.dll
expat.dll was successfully built.
expat: [INSTALL] expat.dll to system32/expat.dll
fmifs: [DEPENDS] .media.d
fmifs: [DEPENDS] .label.d
fmifs: [DEPENDS] .init.d
fmifs: [DEPENDS] .format.d
fmifs: [DEPENDS] .extend.d
fmifs: [DEPENDS] .diskcopy.d
fmifs: [DEPENDS] .compress.d
fmifs: [DEPENDS] .chkdsk.d
fmifs: [RC] fmifs.rc
fmifs: [CC] chkdsk.c
fmifs: [CC] compress.c
fmifs: [CC] diskcopy.c
fmifs: [CC] extend.c
fmifs: [CC] format.c
fmifs: [CC] init.c
fmifs: [CC] label.c
fmifs: [CC] media.c
fmifs: [AR] fmifs.a
make[1]: *** No rule to make target
`../../dk/w32/lib/libvfatlib.a', needed by `
fmifs.nostrip.dll'. Stop.
make: *** [fmifs_install] Error 2
C:\Internet\ReactOS\Subersion\trunk\reactos>
---------------------------------------------------
I dont know what i am making wrong, any idea?
Thanks in advance and my best wishes for 2005.
______________________________________________
Renovamos el Correo Yahoo!: ¡250 MB GRATIS!
Nuevos servicios, más seguridad
http://correo.yahoo.es
The attached patch fixes "warnings + -Werror = not compile" for
rosapps/tests/tokentest/tokentest.c
tokentest.c: In function `CreateInitialSystemToken':
tokentest.c:359: warning: braces around scalar initializer
tokentest.c:359: warning: (near initialization for `authId.LowPart')
tokentest.c:359: warning: excess elements in scalar initializer
tokentest.c:359: warning: (near initialization for `authId.LowPart')
I think SYSTEM_LUID and similar macros for initial values are intended
as initializers of LUID variables, not LARGE_INTEGER ones.
I've posted this (and not commited it myself) because I can be missing
something important. Is there other issue with LUID <-> LARGE_INTEGER ?
<#secure method=pgpmime mode=sign>
--
Nedko Arnaudov <GnuPG KeyID: DE1716B0>
