reactos-kernel

Re: [ros-kernel] midl compiler

[Steven E. <ste...@ya...>]

The mingw-DTK also contains flex/bison and it is on
sourceforge on the mingw project. I use to compile the
wine resource compiler.

> I hacked the idl compiler to build properly on
> Windows. I also fixed the
> preprocessor invocation for Windows. Flex and bison
> are needed to build
> midl. A MinGW port is available at José Fonseca's
> MinGW site
> (http://mefriss1.swan.ac.uk/~jfonseca/gnu-win32/index.html).

__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup
http://fifaworldcup.yahoo.com

Re: [ros-kernel] CVS Commit: reactos

[Eric K. <ek...@rz...>]

"Hartmut Birr" <har...@te...> wrote:

> drivers\storage\scsiport\scsiport.edf:
>   Fixed a typo.
Ooops!

> drivers\fs\vfat\blockdev.c:
> drivers\fs\vfat\create.c:
> drivers\fs\vfat\fat.c:
> drivers\fs\vfat\fcb.c:
> drivers\fs\vfat\fsctl.c:
> drivers\fs\vfat\makefile:
> drivers\fs\vfat\rw.c:
> drivers\fs\vfat\vfat.h:
>   Changed the sector size from BLOCKSIZE to the size from the boot sector.

This is great! Now I can read 640MB MO-Disks (2048 byte sectors) again.

Eric

Re: [ros-kernel] ReactOS not booting correctly

[Eric K. <ek...@rz...>]

"Casper Hornstrup" <ch...@us...> wrote:


> With latest FreeLoader I cannot successfully boot latest ReactOS.
>
> "\SystemRoot" is not created, but all boot drivers are loaded and
initialized.
>
[...]
> (ke/main.c:714) CommandLine: multi(0)disk(0)rdisk(0)partition(1)\reactos
/DEBUGPORT=COM1 /BAUDRATE=115200
> (io/arcname.c:304) NtOpenSymbolicLinkObject() failed to open '\SystemRoot'
(Status c0000001)

I rebuilt the current ReactOS and FreeLoader but I cannot reproduce this
bug.
Do you have the latest DJGPP?
If not, get it! I couldn't get the latest FreeLoader running with an older
DJGPP.

Eric

Re: [ros-kernel] midl compiler

[Steven E. <ste...@ya...>]

Have you taken a look at the ReWind projets IDL
compiler? 


--- Eric Kohl <ek...@rz...> wrote:
> Hi!
> 
> I hacked the idl compiler to build properly on
> Windows. I also fixed the
> preprocessor invocation for Windows. Flex and bison
> are needed to build
> midl. A MinGW port is available at José Fonseca's
> MinGW site
>
(http://mefriss1.swan.ac.uk/~jfonseca/gnu-win32/index.html).


__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup
http://fifaworldcup.yahoo.com

Re: [ros-kernel] midl compiler

[Eric K. <ek...@rz...>]

<we...@bo...> wrote:

> Work as in actually generate correct code or just compile?

It is more or less a 'it compiles' and doesn't crash. Bison didn't like the
original idl.y file because tokens like UUID_LITERAL were defined but
UUID_TOKEN was used in the grammar rules.


> One thing to note is that whatever idl compiler we use will need to
> generate code that works with the rpcrt4 interfaces.

Sure! That's why I described the alternatives as 'unusable'. They either
don't generate rpcrt4-compatible code or don't generate code at all.

Eric

[ros-kernel] CVS Commit: reactos

[Hartmut B. <har...@te...>]

include\ddk\iotypes.h:
  Added PCM_PARTIAL_RESOURCE_LIST.

subsys\win32k\misc\object.c:
  Fixed some calculatations in ObmpGetObjectHandle.

drivers\storage\scsiport\scsiport.edf:
  Fixed a typo.

drivers\fs\vfat\blockdev.c:
drivers\fs\vfat\create.c:
drivers\fs\vfat\fat.c:
drivers\fs\vfat\fcb.c:
drivers\fs\vfat\fsctl.c:
drivers\fs\vfat\makefile:
drivers\fs\vfat\rw.c:
drivers\fs\vfat\vfat.h:
  Changed the sector size from BLOCKSIZE to the size from the boot sector.

- Hartmut

[ros-kernel] ReactOS not booting correctly

[Casper H. <ch...@us...>]

With latest FreeLoader I cannot successfully boot latest ReactOS.

"\SystemRoot" is not created, but all boot drivers are loaded and initialized.


(ke/main.c:621) Module: 'ntoskrnl.exe' at c0000000, length 0x0009e000
(ke/main.c:621) Module: 'HAL.DLL' at c009e000, length 0x0000fb09
(ke/main.c:621) Module: 'SYSTEM.HIV' at c00ae000, length 0x00001c3c
(ke/main.c:621) Module: 'Scsiport.sys' at c00b0000, length 0x000098f5
(ke/main.c:621) Module: 'Atapi.sys' at c00ba000, length 0x00007abc
(ke/main.c:621) Module: 'Class2.sys' at c00c2000, length 0x0000a877
(ke/main.c:621) Module: 'Disk.sys' at c00cd000, length 0x00007f2e
(ke/main.c:621) Module: 'Vfatfs.sys' at c00d5000, length 0x000164ba
(ke/main.c:669) Process registry chunk at c00ae000
(ke/main.c:697) Initializing driver 'Scsiport.sys' at c00b0000, length 0x000098f5
(ldr/loader.c:1429) DriverBase: dc9c0000
(ke/main.c:697) Initializing driver 'Atapi.sys' at c00ba000, length 0x00007abc
(ldr/loader.c:1429) DriverBase: dc9c8000
(ke/main.c:697) Initializing driver 'Class2.sys' at c00c2000, length 0x0000a877
(ldr/loader.c:1429) DriverBase: dc9cf000
(ke/main.c:697) Initializing driver 'Disk.sys' at c00cd000, length 0x00007f2e
(ldr/loader.c:1429) DriverBase: dc9d8000
(ke/main.c:697) Initializing driver 'Vfatfs.sys' at c00d5000, length 0x000164ba
(ldr/loader.c:1429) DriverBase: dc9df000
(ke/main.c:714) CommandLine: multi(0)disk(0)rdisk(0)partition(1)\reactos /DEBUGPORT=COM1 /BAUDRATE=115200
(io/arcname.c:304) NtOpenSymbolicLinkObject() failed to open '\SystemRoot' (Status c0000001)
(ldr/loader.c:1332) LdrpGetModuleObject: Failed to find module \SystemRoot\system32\drivers\cdrom.sys
(ldr/loader.c:1429) DriverBase: dc9fe000
(ldr/loader.c:1332) LdrpGetModuleObject: Failed to find module \SystemRoot\system32\drivers\fs_rec.sys
(ldr/loader.c:1429) DriverBase: dca04000
(io/create.c:90) Parent is a Directory which is not a device type
(io/create.c:90) Parent is a Directory which is not a device type
(ldr/loader.c:1332) LdrpGetModuleObject: Failed to find module \SystemRoot\system32\drivers\beep.sys
(ldr/loader.c:833) Could not open module file: \SystemRoot\system32\drivers\beep.sys
(io/driver.c:204) LdrLoadModule() failed (Status c0000034)
(ldr/loader.c:1332) LdrpGetModuleObject: Failed to find module \SystemRoot\system32\drivers\null.sys
(ldr/loader.c:1429) DriverBase: dca0a000
(ldr/loader.c:1332) LdrpGetModuleObject: Failed to find module \SystemRoot\system32\drivers\keyboard.sys
(ldr/loader.c:1429) DriverBase: dca11000
(ldr/loader.c:1332) LdrpGetModuleObject: Failed to find module \SystemRoot\system32\drivers\blue.sys
(ldr/loader.c:1429) DriverBase: dca1a000
(ldr/loader.c:1332) LdrpGetModuleObject: Failed to find module \SystemRoot\system32\drivers\vgamp.sys
(ldr/loader.c:1429) DriverBase: dca21000
(ldr/loader.c:1332) LdrpGetModuleObject: Failed to find module vidport.sys
(ldr/loader.c:1332) LdrpGetModuleObject: Failed to find module \SystemRoot\system32\drivers\vidport.sys
(ldr/loader.c:1429) DriverBase: dca2b000
(ldr/loader.c:1332) LdrpGetModuleObject: Failed to find module \SystemRoot\system32\drivers\msfs.sys
(ldr/loader.c:1429) DriverBase: dca33000
Mailslot FSD 0.0.1
(ldr/loader.c:1332) LdrpGetModuleObject: Failed to find module \SystemRoot\system32\drivers\npfs.sys
(ldr/loader.c:1429) DriverBase: dca3a000
Named Pipe FSD 0.0.2
(ldr/loader.c:1332) LdrpGetModuleObject: Failed to find module \SystemRoot\system32\drivers\ndis.sys
(ldr/loader.c:1429) DriverBase: dca43000
Page Fault Exception: 14(4)
CS:EIP 1b:77f765b7 <ntdll.dll: 165b7>
CR2 280cc CR3 3e92000 Process: c01f0902 Pid: 2 <> Thrd: c0205b7a Tid: 13
DS 23 ES 23 FS 3b GS 23
EAX: 000280cc   EBX: 7ffdf000   ECX: 00607930
EDX: 000000a6   EBP: 00606da8   ESI: 00000000
EDI: 00000000   EFLAGS: 00000216 SS:ESP 23:606da8
Frames:   <ntdll.dll: 166c9><ntdll.dll: 12ec7><ntdll.dll: 13869><smss.exe: 1054><smss.exe: 1689><ntdll.dll: f1f7><smss.exe: 187a><smss.exe: 2193><smss.exe: 1132><ntdll.dll: 2c0a>

Re: [ros-kernel] midl compiler

[<we...@bo...>]

Quoting Eric Kohl <ek...@rz...>:
> I hacked the idl compiler to build properly on Windows. I also fixed
> the preprocessor invocation for Windows. 
>
Work as in actually generate correct code or just compile?

> As part of the recent cleanup activities I want to move midl into the
> tools directory and rename it to ridlc. Renaming is needed to a) avoid any
> conflicts with an installed M$ Visual C++ and b) to keep the Micro-boys
> from complaining about stealing 'their' compiler and gpl'ing it.
> 
This is a good idea.

> 
> For the past few days I was searching the net for alternative idl
> compilers:
[...]
> 
One thing to note is that whatever idl compiler we use will need to
generate code that works with the rpcrt4 interfaces.

[ros-kernel] midl compiler

[Eric K. <ek...@rz...>]

Hi!

I hacked the idl compiler to build properly on Windows. I also fixed the
preprocessor invocation for Windows. Flex and bison are needed to build
midl. A MinGW port is available at Jos=E9 Fonseca's MinGW site
(http://mefriss1.swan.ac.uk/~jfonseca/gnu-win32/index.html).

As part of the recent cleanup activities I want to move midl into the too=
ls
directory and rename it to ridlc. Renaming is needed to a) avoid any
conflicts with an installed M$ Visual C++ and b) to keep the Micro-boys f=
rom
complaining about stealing 'their' compiler and gpl'ing it.


For the past few days I was searching the net for alternative idl compile=
rs:

I found sidlc, which is part of the samba-tng project
(http://www.samba-tng.org). This compiler already builds a parse tree but
doesn't generate any c code yet. Unfortunately, there is a bug in a parse=
r
action that crashes it. IMO, the scanner and parser seem to be quite poor
compared to our idl compiler. Result: not usable and sources seem to be
unmaintained for about a year but the parse tree looks good.

Of course there is the FreeDCE project. They have a working idl compiler
which is based on the OpenGroup idl compiler. This is a DCE-RPC idl
compiler, not a MS-RPC (aka DCE-RPC plus M$ extensions), and will most
likely not produce MS-RPC compatible c code. IMO, the source code cannot =
be
GPL'd because it is still owned by the OpenGroup. Result: not usable but =
it
might be worth to have a look at.

Any complaints, ideas, comments?


Eric

[ros-kernel] CVS Commit: reactos

[Eric K. <ek...@rz...>]

drivers/fs/mup/.cvsignore
drivers/fs/mup/create.c
drivers/fs/mup/makefile
drivers/fs/mup/mup.c
drivers/fs/mup/mup.h
drivers/fs/mup/mup.rc
drivers/fs/ntfs/.cvsignore
drivers/fs/ntfs/blockdev.c
drivers/fs/ntfs/create.c
drivers/fs/ntfs/dirctl.c
drivers/fs/ntfs/fcb.c
drivers/fs/ntfs/fsctl.c
drivers/fs/ntfs/makefile
drivers/fs/ntfs/ntfs.c
drivers/fs/ntfs/ntfs.h
drivers/fs/ntfs/ntfs.rc
drivers/fs/ntfs/volinfo.c
install.bat
makefile

Added 'Multi UNC Provider' driver stub.
Added NTFS driver stub.

Eric

[ros-kernel] CVS Commit: reactos

[Eric K. <ek...@rz...>]

services/eventlog/eventlog.c
services/eventlog/eventlog.h
services/eventlog/eventlog.rc
services/eventlog/logport.c
services/eventlog/makefile
services/rpcss/endpoint.c
services/rpcss/makefile
services/rpcss/rpcss.c
services/rpcss/rpcss.h
services/rpcss/rpcss.rc
makefile

    Added eventlog service stub.
    Added RPC server stub.

Both services are not usable yet.
If you want to improve one of them, feel free to do so. :-)

Eric

[ros-kernel] 0.0.20

[Jason F. <jas...@ya...>]

I guess I don't have a problem releasing 0.0.20 - but is there
anything anyone wanted to add first?

- Jason







__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup
http://fifaworldcup.yahoo.com

Re: [ros-kernel] ReactOS 0.0.20

[David K. <kr...@mb...>]

Commit them, please.

David


> PS.:
> I want to move the midl compiler in 'lib/rpcrt4/midl' to 'tools/midl' and
> resurrect rpcrt4. I also have stub-implementations of ntfs.sys, mup.sys,
> eventlog service and rpcss service on my harddisk. Should I commit them and
> flood the cvs tree or keep them to myself? Does anyone want to adopt them?
> 
> Eric
> 
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by: Jabber Inc.
> Don't miss the IM event of the season | Special offer for OSDN members! 
> JabConf 2002, Aug. 20-22, Keystone, CO http://www.jabberconf.com/osdn
> _______________________________________________
> reactos-kernel mailing list
> rea...@li...
> https://lists.sourceforge.net/lists/listinfo/reactos-kernel
> 

Re: [ros-kernel] ReactOS 0.0.20

[Joseph G. <gal...@va...>]

> PS.:
> I want to move the midl compiler in 'lib/rpcrt4/midl' to 'tools/midl' and
> resurrect rpcrt4. I also have stub-implementations of ntfs.sys, mup.sys,
> eventlog service and rpcss service on my harddisk. Should I commit them
and
> flood the cvs tree or keep them to myself? Does anyone want to adopt them?

I would say commit them; that way people
will know they exist and maybe someone
with time will decide to work on one of
them.

At the very least, if they are committed and
being built as part of the make, they shouldn't
fall out of date quite so easily.

Also, stubs provide good guidance for anyone
looking for something to implement.

- Joseph

[ros-kernel] CVS Commit: reactos

[Eric K. <ek...@rz...>]

lib/kernel32/file/npipe.c
    Implemented CallNamedPipeW() and TransactNamedPipe().

Eric

[ros-kernel] ReactOS 0.0.20

[Eric K. <ek...@rz...>]

Hi!

I think we should release ReactOS 0.0.20 soon. A lot fixes and improvements
happened since 0.0.19 was released. From a realistic point-of-view, I do not
think we will be able to make 0.0.20 a bootable and installable OS but
0.0.21 could finally reach this goal and become ReactOS 0.1. Therefore, the
following topics will have to be addressed:
    - floppy driver
    - registry
    - setup application
    - setup loader

Floppy driver:
AFAIK, the floppy driver doesn't work. I don't have any more information
about this.

Registry:
We need a writable registry to get the reactos setup rolling. If an
NT-compatible sollution is not available within a reasonable time, we should
think about using the flat text format as an intermediate sollution. Later a
setup/update application can convert the format. I'm in favor of the
intermediate sollution because FreeLoader already supports it.

Setup application:
I started to write a prototype. It looks like the NT setup but it doesn't
work yet. ;-)

Setup loader:
FreeLoader needs to be changed. I tried to add setup features to the
existing FreeLoader but it makes the code kind of ugly since it has to deal
with two entirely different jobs. I suggest to write a second front end
(setupldr.c) that does not include the ability to boot multiple OSs but
focuses on booting ReactOS for setup. IMHO, the current architecture of
FreeLoader supports multiple front ends perfectly.

Other important topics:
- Vfatfs needs to support large sector disks (1024/2048 byte sectors).
- Better support for removable disks.

PS.:
I want to move the midl compiler in 'lib/rpcrt4/midl' to 'tools/midl' and
resurrect rpcrt4. I also have stub-implementations of ntfs.sys, mup.sys,
eventlog service and rpcss service on my harddisk. Should I commit them and
flood the cvs tree or keep them to myself? Does anyone want to adopt them?

Eric

Re: [ros-kernel] ping

[Eric K. <ek...@rz...>]

> ping

Pong! ;-)

[ros-kernel] ping

[Steven E. <ste...@ya...>]

ping

__________________________________________________
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup
http://fifaworldcup.yahoo.com

Re: [ros-kernel] Another Free C Runtime Library

[Aidan K. <ke...@pa...>]

 Ar an 19ú lá de mí 6, scríobh KJK::Hyperion :

 > At 07.15 19/06/2002, you wrote:
 > >Do you know where to find the documentaion(specs) for the ANSI C library?
 > 
 > The ANSI C draft is available for free on the net as a PDF document. The 
 > final version costs (a lot, too), but the draft should be enough in most 
 > cases. I don't remember the original filename, but I can send it to you if 
 > you cannot find it

Original file name is n869.{txt,pdf}; the txt version is more useful
for quick reference on screen, in my experience. 

-- 
I'm not a pheasant plucker / I'm a pheasant plucker's son.
I'm just a'plucking pheasants / 'Til the pheasant plucker comes.

Re: [ros-kernel] Build...

[Robert D. <od...@pn...>]

Latest CVS (not a fresh snapshot) builds from clean and runs
in latest bochs from built from a recent CVS snapshot and
booting using freedos.bat GDITest runs and exits nicely.

freeldr.bat fails with the dump shown below. Although I can
build freeldr using DJGPP I havn't delved into the procedure
for creating a new freeldr.img The current one .zip's to around
360kb. I know it's discouraged but would anybody care to post
a latest freeldr floppy image to the list?

----------------------------------------------------------------------------------------------------
Found 1 system processor. [32 MB 
Memory]
No boot drivers 
available.

Bug detected (code 0 param 0 0 0 
0)
   The bug code is undefined. Please use an existing code 
instead.

Pid: 1 <SYSTEM> Thrd: c01106f4 Tid: 
1
Frames: <ntoskrnl.exe: 855e><ntoskrnl.exe: aa14><ntoskrnl.exe: 
aaad><ntoskrnl.exe: b052><ntoskrnl.exe: 
117c>
Bug detected code: 
0x1D
Divide Error Exception: 
0(0)
Processor: 0 CS:EIP 8:c0004cf8 <ntoskrnl.exe: 
4cf8>
cr2 0 cr3 271000 Proc: c00ffaa6 Pid: 1 <SYSTEM> Thrd: c01106f4 Tid: 
1
DS 10 ES 10 FS 30 GS 
10
EAX: 00000001   EBX: c0085750   ECX: 
c00a7000
EDX: 00000001   EBP: c007fd80   ESI: 
00200000
EDI: 00000000   EFLAGS: 00000202 kESP c007fd0c kernel stack base 
c007d000
ESP 
c007fd0c
Frames: <ntoskrnl.exe: 8539><ntoskrnl.exe: 855e><ntoskrnl.exe: 
aa14><ntoskrnl.exe: aaad><ntoskrnl.exe: b052><ntoskrnl.exe: 
117c>

Re: [ros-kernel] Build...

[Robert D. <od...@pn...>]

Sorry guys,

Eric was spot on! Forgot to commit the two files I changed
from the include directory, winsock2.h and ndis.h

Now committed. I am pulling down any latest changes and
will perform a clean build shortly to check for any other problems.

I just found 'Flat Mode' in WinCVS which makes it much
easier to find the files one has modified. Shouldn't have this
problem again I hope.

Rgs, Robert.

At 08:57 PM 21/06/2002 +0200, you wrote:

>"Joseph Galbraith" <gal...@va...> wrote:
>
> > Indeed, though I failed to mention it, I did
> > try a make clean, and didn't have any luck with
> > it.
> >
> > I haven't tried a fresh checkout from CVS though --
> > I may try that next, if others aren't seeing the
> > build failure.
>
>I've seen the unresolved references too. IMO, the ndis.h patch wasn't
>checked in.
>
>Eric
>
>
>
>
>-------------------------------------------------------
>Sponsored by:
>ThinkGeek at http://www.ThinkGeek.com/
>_______________________________________________
>reactos-kernel mailing list
>rea...@li...
>https://lists.sourceforge.net/lists/listinfo/reactos-kernel

Re: [ros-kernel] Build...

[Robert D. <od...@pn...>]

Hi Joseph,

I'm the one who committed the changes. It all builds ok on my
tree which I used to sync my changes into CVS.

For a quick fix simply remove the reference to 'packet' from the
NET_DRIVERS list in the master makefile.

NET_DRIVERS = afd ndis packet tcpip tdi wshtcpip

The packet driver probably will not work as is anyhow and so it
has not been added to the registry for loading. Therefore it is
not required to be built.

Rgs, Robert.


At 09:31 AM 21/06/2002 -0600, you wrote:
>Is anyone else having trouble building most
>recent cvs?
>
>I'm dying in drivers/net/packet with undefined symbols
>for various ndis symbosl.  Most seem to be out
>of control.c -- NdisResetEvent, etc.
>
>It looks like for some reason it is failing to link
>with ndis.  I haven't tracked it further than this.
>
>- Joseph
>
>
>
>
>-------------------------------------------------------
>Sponsored by:
>ThinkGeek at http://www.ThinkGeek.com/
>_______________________________________________
>reactos-kernel mailing list
>rea...@li...
>https://lists.sourceforge.net/lists/listinfo/reactos-kernel

[ros-kernel] Build failed - packet driver on clean source tree

[Steven E. <Ste...@ya...>]

        ./packet.coff packet.o read.o openclos.o write.o
win_bpf_filter.o  ../..
/../dk/nkm/lib/ntoskrnl.a ../../../dk/nkm/lib/hal.a
../../../dk/nkm/lib/ndis.a
packet.o(.text+0x220):packet.c: undefined reference to
`NdisSystemProcessorCount
'
packet.o(.text+0x177a):packet.c: undefined reference to `NdisResetEvent'
packet.o(.text+0x1807):packet.c: undefined reference to `NdisWaitEvent'
packet.o(.text+0x1921):packet.c: undefined reference to `NdisSetEvent'
read.o(.text+0x13):read.c: undefined reference to `NdisAcquireSpinLock'
read.o(.text+0x7f):read.c: undefined reference to `NdisReleaseSpinLock'
read.o(.text+0x36b):read.c: undefined reference to `NdisAcquireSpinLock'
read.o(.text+0x3ad):read.c: undefined reference to `NdisReleaseSpinLock'
read.o(.text+0x42d):read.c: undefined reference to `NdisAcquireSpinLock'
read.o(.text+0x471):read.c: undefined reference to `NdisReleaseSpinLock'
read.o(.text+0x56a):read.c: undefined reference to `NdisAcquireSpinLock'
read.o(.text+0x59d):read.c: undefined reference to `NdisReleaseSpinLock'
read.o(.text+0x64b):read.c: undefined reference to `NdisAcquireSpinLock'
read.o(.text+0x67e):read.c: undefined reference to `NdisReleaseSpinLock'
read.o(.text+0x87e):read.c: undefined reference to `NdisAcquireSpinLock'
read.o(.text+0x937):read.c: undefined reference to `NdisReleaseSpinLock'
read.o(.text+0x96d):read.c: undefined reference to `NdisAcquireSpinLock'
read.o(.text+0x9a5):read.c: undefined reference to `NdisReleaseSpinLock'
read.o(.text+0xc84):read.c: undefined reference to `NdisAcquireSpinLock'
read.o(.text+0xcb0):read.c: undefined reference to `NdisReleaseSpinLock'
read.o(.text+0xd11):read.c: undefined reference to
`NdisReinitializePacket'
openclos.o(.text+0x31b):openclos.c: undefined reference to
`NdisInitializeEvent'

openclos.o(.text+0x32f):openclos.c: undefined reference to
`NdisInitializeEvent'

openclos.o(.text+0x52b):openclos.c: undefined reference to
`NdisAllocateSpinLock
'
openclos.o(.text+0x53f):openclos.c: undefined reference to
`NdisAllocateSpinLock
'
openclos.o(.text+0x6d5):openclos.c: undefined reference to
`NdisWaitEvent'
openclos.o(.text+0x902):openclos.c: undefined reference to
`NdisSetEvent'
openclos.o(.text+0x935):openclos.c: undefined reference to
`NdisResetEvent'
write.o(.text+0x7b):write.c: undefined reference to `NdisResetEvent'
write.o(.text+0x1dc):write.c: undefined reference to `NdisWaitEvent'
write.o(.text+0x1f0):write.c: undefined reference to `NdisResetEvent'
write.o(.text+0x272):write.c: undefined reference to `NdisSetEvent'
write.o(.text+0x29f):write.c: undefined reference to
`NdisReinitializePacket'
make[1]: *** [packet.nostrip.sys] Error 1
make[1]: Leaving directory `D:/src/reactos/drivers/net/packet'
make: *** [packet] Error 2

D:\src\reactos>

Thanks
Steven

"Every revolution was once a thought in one man's mind"
- Ralph Waldo Emerson 

Re: [ros-kernel] Build...

[Eric K. <ek...@rz...>]

"Joseph Galbraith" <gal...@va...> wrote:

> Indeed, though I failed to mention it, I did
> try a make clean, and didn't have any luck with
> it.
>
> I haven't tried a fresh checkout from CVS though --
> I may try that next, if others aren't seeing the
> build failure.

I've seen the unresolved references too. IMO, the ndis.h patch wasn't
checked in.

Eric

Re: [ros-kernel] Build...

[Joseph G. <gal...@va...>]

Indeed, though I failed to mention it, I did
try a make clean, and didn't have any luck with
it.

I haven't tried a fresh checkout from CVS though --
I may try that next, if others aren't seeing the
build failure.

Thanks,

Joseph

----- Original Message ----- 
From: "Casper Hornstrup" <ch...@us...>
To: <rea...@li...>
Sent: Friday, June 21, 2002 9:44 AM
Subject: Re: [ros-kernel] Build...


> fre, 2002-06-21 kl. 17:31 skrev Joseph Galbraith:
> > Is anyone else having trouble building most
> > recent cvs?
> 
> A good strategy when experiencing build failures is to first do a "make
> clean". Then a full "make". In this case it may be sufficient to do a
> "make ndis_implib" since the ndis.def files have changed. Of course, it
> may be that this build failure cannot be solved by this approach. I
> haven't tried building the latest CVS version so I couldn't say.
> 
> Casper
> 
> 
> 
> 
> -------------------------------------------------------
> Sponsored by:
> ThinkGeek at http://www.ThinkGeek.com/
> _______________________________________________
> reactos-kernel mailing list
> rea...@li...
> https://lists.sourceforge.net/lists/listinfo/reactos-kernel
> 
>