Christopher Faylor [EMAIL PROTECTED] wrote:
On Sat, Sep 21, 2002 at 08:34:06PM +0100, Conrad Scott wrote:
The attached patch is the (small) subset of the cygwin_daemon merge
that
I need clearance on.
Looks ok. Go ahead and check it in.
Thanks: it's in now. Sorry for the mangled Makefile
The attached patch is the (small) subset of the cygwin_daemon merge that
I need clearance on. The files affected are:
src/winsup/cygwin/Makefile.in
src/winsup/cygwin/dcrt0.cc
src/winsup/cygwin/fhandler_tty.cc
src/winsup/cygwin/tty.cc
There are just two changes reflected here:
* The code only
Francois de Campagnolle [EMAIL PROTECTED] wrote:
The cygwin binaries try for some reason to talk to the 1543 port on
localhost since I last run setup.
What's gives ?
Since in your last message on the list you asked about cygserver, my
guess is that you have a windows 9x/ME box and you are
Nicholas Wourms [EMAIL PROTECTED] wrote:
I can't believe there are people who still use cpio! Is this a
emacs
build thing or do you honestly prefer to use cpio?
Nah -- it's one of them sociological woojits :-)
// Conrad
Thomas Chadwick [EMAIL PROTECTED] wrote:
After doing this, select is still non-blocking, returning -1,
and the value
returned by WSAGetLastError() is 10038, which seems to
correspond to the
#define WSAENOTSOCK in winsock.h.
Any idea what this error code means and what steps I might take
to
Chuck,
The wrapper scripts for autoconf and autoheader use temporary files
that they (are supposed to) remove on exit (using trap 0). But I tend
to find *lots* of these directories in /tmp (e.g. /tmp/acX and
/tmp/ahX). Just running autoconf or autoheader with no arguments
creates a
Charles Wilson [EMAIL PROTECTED] wrote:
Yeah, I've noticed that too. Perhaps I outsmarted myself:
the wrapper script never actually exits
-- it *execs* the real autoconf/make/whatever.
So perhaps the trap doesn't get triggered when you exec.
Yep: that's what it is. I constructed various
Corinna Vinschen [EMAIL PROTECTED] wrote:
More problematic is the approach to use cygserver for this. I've
talked
to Chris about passing descriptors and we agree in that we want to
try
under all circumstances to find a solution which doesn't need
cygserver.
Corinna,
I thought that the main
egor duda [EMAIL PROTECTED] wrote:
Btw, libstdc++ in gcc 3.* is configured so that classes in std::
namespace are not visible unless one specify std:: via 'using' or
explicitly. I feel this can be the problem that will make most
noise.
Cygwin setup is just one example of program affected. I
Robert Collins [EMAIL PROTECTED] wrote:
Ok, I've got some feedback for you...
I need to have a good think about some of what's being presented.
The following things are unconditionally good:
The pure virtual transport changes
The recoverable approach, and instance detection changes.
Robert Collins [EMAIL PROTECTED] wrote:
There seems to be a lot of code duplication - definitions copied to
make
private versions, that sort of thing. Can you elaborate on why? I
strongly prefer to only have one instance of such things to prevent
skew
occuring.
Rob, as I was having a bath I
/pipe).
Just to be sure, I gave this the same i_mode as previously but with
S_IFIFO.
This patch also makes the output of `ls -l /dev/tcp' a bit strange,
but I think this is a relatively unimportant detail, what with a
pending /dev vfs.
Anyhow, enjoy?
// Conrad
2002-06-22 Conrad Scott [EMAIL
Earnie Boyd [EMAIL PROTECTED] wrote:
MSDN says that this is Win2000 SP2 and XP only. So you need to
guard it
with the appropriate WINVER constant.
Earnie, I'm a bit confused about these WINVER guards. The use of a
new Win2000 SP2 and XP only flag like this is not a compile-time but a
Earnie Boyd [EMAIL PROTECTED] wrote:
MSDN says that this is Win2000 SP2 and XP only. So you need to
guard it
with the appropriate WINVER constant.
Sorry about my earlier querulous email: I was blissfully ignorant of
the whole WINVER system.
My problem now is to choose a relevant version
Before anyone else gets there, I'll reply to my own message (what a
change).
The patch I just sent for path.cc breaks (at least) find(1) on
/proc/registry -- it doesn't descend into it at all. Sorry.
So I've got something wrong in my understanding of the cygwin/win32
stuff (which doesn't
Pierre A. Humblet [EMAIL PROTECTED] wrote:
I also tried to build the current cvs but get an error
In file included from ../../../../src/winsup/cygwin/assert.cc:13:
/src/winsup/w32api/include/wingdi.h:2521: `HENMETAFILE' was not declared
in
this scope
I just stumbled over that one: it's a
Conrad Scott [EMAIL PROTECTED]
* woutsup.h: New file for cygserver compilations outside the
cygwin dll. Also collects the common debugging code from the
various cygserver files.
* cygserver.cc: Change to use woutsup.h. Use new XXX_printf
functions
Conrad Scott [EMAIL PROTECTED] wrote:
I've made some changes to the cygserver code to harmonize the debugging
output.
I also realise, now that I come to run things *without* cygserver running,
that I've made things *far* too noisy. So, here's a replacement for the
previous patch that calms
, compatible with the previous versions).
I'll go back to looking to see if I can get the shm sub-system working with
the socket transport next (or, at least, trying to understand why it doesn't
work at the moment).
Cheers.
// Conrad
2002-06-10 Conrad Scott [EMAIL PROTECTED]
* include/sys
).
Cheers.
// Conrad
2002-06-10 Conrad Scott [EMAIL PROTECTED]
* include/sys/ipc.h: Reorganize file. Add IPC_INFO flag.
* include/sys/shm.h: Reorganize file. Add SHM_STAT flag.
(struct shmid_ds): Add spare fields.
(struct shminfo): New struct.
(class cygshmid_ds): New class to replace
and paste from the usage message in strace.cc.
ChangeLog appended here and attached (to keep its format out of Outlook's
clutches).
// Conrad
2002-06-07 Conrad Scott [EMAIL PROTECTED]
* strace.cc (toggle): New global variable.
(error): Use exit instead of ExitProcess so that stdio buffers get
[Second attempt once I'd calmed down enough to get it right.]
Attached is a patch that adds sub-second resolution to the access,
modification, and creation times returned by stat(2) etc. I thought this
would make a nice companion to Corinna's work on making other things in
stat(2) be 64-bit.
Christopher Faylor [EMAIL PROTECTED] wrote:
I'm unclear whether this is the best naming / type scheme but it is one
recognised by both the make and fileutils packages available from the
cygwin setup (i.e. make this patch and re-compile those packages and they
detect the new fields).
As
Christopher Faylor [EMAIL PROTECTED] wrote:
The ChangeLog is closer, but still not complete. You should provide a
*complete* ChangeLog, including time/date name, email address. The
entries should also be capitalized (Ditto. not ditto.).
If you can also indent the code, as per a normal
I've been playing around with the strace program some more and noticed a
minor glitch: it's only meant to trace forked children if the -f flag is
given on the command line. Unfortunately it currently always traces
children, and the this flag has no effect.
Having read the MS documentation, I
Christopher Faylor [EMAIL PROTECTED] wrote:
I think you're running run.exe from Chuck Wilson's site.
Umm . . . (quick check). No: it's the copy from the current version of
XFree-startup-scripts (4.2.0-2). I was using it in this case without an X
server running (it's always happily worked either
Chris,
I've just picked up the latest changes from CVS and I'm having a problem
with run.exe from a .BAT file (i.e., from my current cygwin.bat mechanism).
After a bit of tracing, I've found that it comes down to NtQueryObject in
handle_to_fn() in dtable.cc. NtQueryObject succeeds but the
Christopher Faylor [EMAIL PROTECTED] wrote:
On Tue, May 28, 2002 at 11:00:28PM +0100, Conrad Scott wrote:
I've just picked up the latest changes from CVS and I'm having a problem
with run.exe from a .BAT file (i.e., from my current cygwin.bat
mechanism).
What's run.exe?
cgf
Sorry: it's
Just to clarify: This email is not about bugs in setup.exe.
I've just spent all evening re-installing cygwin from scratch on my machine
and I thought I'd share some of my excess fun with everyone else:
(BTW I'm using setup.exe version 2.218.2.9, if and where that's relevant.)
(*) The gdbm and
[I'm crossposting to the cygwin and the rxvt-bugs lists.]
I've just changed over to using rxvt with cygwin (rather than the default
Windows console) and I'm loving it. (I'm not using X11, just the 'raw'
window manager.)
Anyhow, I've discovered a repeatable crash in rxvt 2.7.2 under cygwin
I'm befuddled and I can't find anything on any of the cygwin mailing lists
to help me, so I hope someone here can spot my mistake.
My system has been working fine for a long time but just recently I noticed
that when I do an `ls' of `/usr' or `echo /usr/*', `/usr/bin' doesn't show
up. It's there
31 matches
Mail list logo
