Re: [1.7] Undocumented change in accessing by dos drive letters?

2009-11-03 Thread Andy Koppe
2009/11/4 Linda Walsh: >> C: and C:\ aren't the same thing in DOS/cmd.exe. C: means the current >> directory of the C drive, whereas C:\ means the root directory of the >> C drive. Within each cmd.exe session, each drive has its own current >> directory. > > --- >        Right.  That's a cmd.exe-is

Re: incomplete installs on AspireOne netbook

2009-11-03 Thread Larry Hall (Cygwin)
On 11/03/2009 10:54 PM, David T-G wrote: Hi, all -- I am having a terrible time trying to install cygwin on our new Acer AspireOne netbook (Win XP SP3 @ 1G). I have tried downloading to a local directory (my usual preference), installing from the 'net, and -- most interestingly, to me -- instal

incomplete installs on AspireOne netbook

2009-11-03 Thread David T-G
Hi, all -- I am having a terrible time trying to install cygwin on our new Acer AspireOne netbook (Win XP SP3 @ 1G). I have tried downloading to a local directory (my usual preference), installing from the 'net, and -- most interestingly, to me -- installing from a copied download tree from anoth

Re: [1.7] Updated: cygwin-1.7.0-63

2009-11-03 Thread Jon TURNEY
On 03/11/2009 19:13, Jim Reisert AD1C wrote: Unfortunately, once I installed this version, I couldn't get the Cygwin/X server to start (using 1.7.1). I had to roll back to 1.7.0-62 This is quite probably the same problem as [1], a consequence of the change in default locale to C.UTF-8 (and so

Re: [1.7] Undocumented change in accessing by dos drive letters?

2009-11-03 Thread Linda Walsh
Andy Koppe wrote: C: and C:\ aren't the same thing in DOS/cmd.exe. C: means the current directory of the C drive, whereas C:\ means the root directory of the C drive. Within each cmd.exe session, each drive has its own current directory. --- Right. That's a cmd.exe-ism -- As Christofph

[1.7] RCS - co -l fails if *,v not owned by user -

2009-11-03 Thread Yeng Chen
RCS co -l fails if the ,v file is not owned by the user. I get a permission denied error. In my case, the RCS directory is on samba. It seems to fix it if in conf.h I change #define bad_b_rename 1 The flag makes it delete the target(*,v) before renaming the work file (,*,v -> *,v) Thanks

binutils: dlltool DATA exports incompatible with auto-import

2009-11-03 Thread Yaakov (Cygwin/X)
GNOME Evolution contains several interdependent libraries, so on Cygwin/MinGW the build creates artificial import libraries with dlltool from .def files. While trying to build 2.28 I came across a possible incompatibility between the implib generated by dlltool and auto-import linkage: $ gcc

Is it worth it to run rebaseall?

2009-11-03 Thread Jim Reisert AD1C
During the brief period I ran Windows Vista, it was recommended to run rebaseall. Now that I'm running Windows 7 and Cygwin 1.7, seemingly without problems, is there any reason/need/advantage to running it? Thanks - Jim -- Jim Reisert AD1C, , http://www.ad1c.us -- Problem reports: http:/

Re: [1.7] RCS - co -l fails if *,v not owned by user -

2009-11-03 Thread Larry Hall (Cygwin)
On 11/03/2009 08:33 PM, Yeng Chen wrote: RCS co -l fails if the ,v file is not owned by the user. I get a permission denied error. In my case, the RCS directory is on samba. It seems to fix it if in conf.h I change #define bad_b_rename 1 The flag makes it delete the target(*,v) before renaming

Re: Shall dlopen("foo") succeeed if only "foo.dll" exists?

2009-11-03 Thread Linda Walsh
Corinna Vinschen wrote: For 1.7 our choice is to keep dlopen() checking for the .dll suffix to be more Windows-like, or to be more Linux-like by dropping the check for the .dll suffix so that dlopen() fails if the filename isn't specified fully. --- Does anybody know such a tool? If so, is the

source package for atlas 3.9.16 on cygwin 1-7

2009-11-03 Thread Marco Atzeri
Hi All, I prepared a (very preliminary) source package for Automatically Tuned Linear Algebra Software (ATLAS). It is a fast alternative to the reference BLAS/LAPACK build. ATLAS 3.9.16 packages, source and Core232SSE3 lib, for cygwin-1.7 are at http://matzeri.altervista.org/cygwin-1.7/atlas/

Re: Shall dlopen("foo") succeeed if only "foo.dll" exists?

2009-11-03 Thread Charles Wilson
Corinna Vinschen wrote: > On Nov 2 20:05, Eric Blake wrote: >> According to Corinna Vinschen on 11/2/2009 9:48 AM: >>> While we tend to change the implementation to be more Linux-like, >>> there could be some tools out there which erroneously depend on the >>> Windows-like behaviour of Cygwin's dl

Re: [1.7] Updated: cygwin-1.7.0-63

2009-11-03 Thread Jim Reisert AD1C
I'm up and running again, thanks! On Tue, Nov 3, 2009 at 12:54 PM, Christopher Faylor wrote: > On Tue, Nov 03, 2009 at 07:50:41PM +, Jon TURNEY wrote: >>On 03/11/2009 19:13, Jim Reisert AD1C wrote: >>>Unfortunately, once I installed this version, I couldn't get the >>>Cygwin/X server to star

Re: [1.7] Updated: cygwin-1.7.0-63

2009-11-03 Thread Christopher Faylor
On Tue, Nov 03, 2009 at 07:50:41PM +, Jon TURNEY wrote: >On 03/11/2009 19:13, Jim Reisert AD1C wrote: >>Unfortunately, once I installed this version, I couldn't get the >>Cygwin/X server to start (using 1.7.1). I had to roll back to 1.7.0-62 > >This is quite probably the same problem as [1], a

Re: [1.7] Updated: cygwin-1.7.0-63

2009-11-03 Thread Jim Reisert AD1C
Unfortunately, once I installed this version, I couldn't get the Cygwin/X server to start (using 1.7.1). I had to roll back to 1.7.0-62 I can provide debug info if someone tells me what is needed. - Jim On Tue, Nov 3, 2009 at 7:54 AM, Corinna Vinschen wrote: > Hi folks, > > > I just uploaded a

Re: Is it worth it to run rebaseall?

2009-11-03 Thread Larry Hall (Cygwin)
On 11/03/2009 06:41 PM, Jim Reisert AD1C wrote: During the brief period I ran Windows Vista, it was recommended to run rebaseall. Now that I'm running Windows 7 and Cygwin 1.7, seemingly without problems, is there any reason/need/advantage to running it? Not generally, no. You should only run

Re: Parallel downloading of Cygwin packages

2009-11-03 Thread paul . hermeneutic
On Mon, Nov 2, 2009 at 22:39, Chris Cormie wrote: > Christopher Faylor wrote: >> >> On Mon, Nov 02, 2009 at 02:25:03PM +1100, Chris Cormie wrote: There are barriers to implementing rpm in Cygwin, the most frequently mentioned being the fact that a cygwin process can't easily replace

Re: Accessing GLOBALROOT paths - a potential compromise???

2009-11-03 Thread Corinna Vinschen
On Nov 3 16:54, Corinna Vinschen wrote: > On Nov 3 16:44, Corinna Vinschen wrote: > > $ ./DefDosDevice X: '\??\GLOBALROOT\Device\HarddiskVolumeShadowCopy1 > > $ cd /mnt/x > > Oops... replace /mnt/x with /cygdrive/x. I have changed my cygdrive > prefix to /mnt locally. Sorry for any confusi

Re: Accessing GLOBALROOT paths - a potential compromise???

2009-11-03 Thread Corinna Vinschen
On Nov 3 16:44, Corinna Vinschen wrote: > $ ./DefDosDevice X: '\??\GLOBALROOT\Device\HarddiskVolumeShadowCopy1 > $ cd /mnt/x Oops... replace /mnt/x with /cygdrive/x. I have changed my cygdrive prefix to /mnt locally. Sorry for any confusion. Corinna -- Corinna Vinschen

Re: Accessing GLOBALROOT paths - a potential compromise???

2009-11-03 Thread Corinna Vinschen
On Nov 3 01:49, Jeffrey J. Kosowsky wrote: > I know that there have been multiple threads about the pros/cons of > being able to access XP/Vista style \\?\GLOBALROOT paths. > > However, not being able to access them limits one's abilities to use > things like shadow copies since they create shado

[ANNOUNCEMENT] [1.7] Updated: cygwin-1.7.0-63

2009-11-03 Thread Corinna Vinschen
Hi folks, I just uploaded a new Cygwin 1.7 test release, 1.7.0-63. Changes in relation to 1.7.0-62: - Add a bigger patch which allows by default to run multiple Cygwin installations in parallel without interferring and without interaction between processes

Bonjour Monsieur

2009-11-03 Thread Hadja Zanab Diallo
Bonjour Monsieur. Je tiens à appliquer par ce moyen pour votre coopération pour assurer la possibilité d'investir et faire des affaires avec vous dans votre pays, j'ai reçu votre contact dans la chambre de commerce et d'industrie du Ghana. Je prie et choisi votre nom entre d'autres noms en

Re: Build a toolchain for ARM with cygwin 1.5.25-15 ?

2009-11-03 Thread Dave Korn
Marcel Rutten wrote: > building the toolchain is done in 4 steps: > > 1 build binutils, > 2 build gcc stage 1, a barebones gcc, for building glibc > 3 build glibc > 4 build gcc stage 2, with c and c++ support. > > Things go wrong in step 3, the linker starts complaining about > undefined referen

Errors when I first run .bat file

2009-11-03 Thread radiohacker
I just installed cygwin I get the following errors when I run the batch file bash: /usr/bin/tr: No such file or directory bash: /usr/bin/sed: No such file or directory Your group is currently "mkpasswd". This indicates that the /etc/passwd (and possibly /etc/group) files should be rebuilt. Se

Re: Parenthood seems to be lost when the parent is exeeced

2009-11-03 Thread Christopher Faylor
On Tue, Nov 03, 2009 at 11:58:53AM +0100, Laurent Najman wrote: >The parenthood seems to be lost when the parent is execed. Here is a >sample code Yes, it's a longstanding limitation of Cygwin that an execed process does not remember the child processes associated with the previous pid. Sorry.

Build a toolchain for ARM with cygwin 1.5.25-15 ?

2009-11-03 Thread Marcel Rutten
Dear all, I  am trying to set up a toolchain for an ARM platform (arm-linux-gnueabi, my TV) using cygwin, without much success so far. The tools I'm using are: binutils 20080624-2 gcc4 4.3.2-2 The toolchain code consists of: gcc 4.2.0, glibc 2.5.90, binutils 2.17.50,

Parenthood seems to be lost when the parent is exeeced

2009-11-03 Thread Laurent Najman
Hi, The parenthood seems to be lost when the parent is execed. Here is a sample code #include #include #include #include #include int main(int argc, char **argv) { int status; if (argc > 1) { printf ("Execed Parent: Cygwin: %d Windows: %lu \n", getpid (), GetCurrentProcessId

Re: unable to switch to gcc 4

2009-11-03 Thread Thomas Wolff
Andy Koppe schrieb: 2009/11/1 Hans Horn: I think I figured it out myself; seemingly I need to have /etc/alternatives in my path before /usr/bin. No, that shouldn't be necessary. Thes issue is with this: lrwxrwxrwx 1 Hans None9 Jul 14 16:30 /usr/bin/gcc.exe -> gcc-3.exe That

Re: Shall dlopen("foo") succeeed if only "foo.dll" exists?

2009-11-03 Thread Corinna Vinschen
On Nov 2 20:05, Eric Blake wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > According to Corinna Vinschen on 11/2/2009 9:48 AM: > > Weird question, right? > > > > Here's the problem. > > > > Assume you have a file "foo.so" on Linux. If you call > > > > dlopen ("./foo.so", RTLD_L

Re: Shall dlopen("foo") succeeed if only "foo.dll" exists?

2009-11-03 Thread Yaakov (Cygwin/X)
On 03/11/2009 02:21, Thomas Wolff wrote: In terms of compatibility and least surprising behavior, what about accepting dlopen ("bla.so") and looking for "bla.dll"? No, Ruby and Apache2 (among others) use .so on all platforms, including native Win32. Yaakov -- Problem reports: http://

Re: Shall dlopen("foo") succeeed if only "foo.dll" exists?

2009-11-03 Thread Thomas Wolff
Corinna Vinschen schrieb: On Nov 2 14:17, Larry Hall (Cygwin) wrote: On 11/02/2009 11:48 AM, Corinna Vinschen wrote: For 1.7 our choice is to keep dlopen() checking for the .dll suffix to be more Windows-like, or to be more Linux-like by dropping the check for the .dll suffix so that d