Re: [RFA] Re: DOS/Windows-specific code: cli-cmds.c

Re: [RFA] Re: DOS/Windows-specific code: cli-cmds.c

[David Taylor]

    Date: Sun, 10 Jun 2001 16:28:57 +0300 (IDT)
    From: Eli Zaretskii <eliz@is.elta.co.il>

    > Date: Tue, 8 May 2001 14:41:53 +0300 (IDT)
    > From: Eli Zaretskii <eliz@is.elta.co.il>
    > 
    >     * cli-cmds.c:
    > 
    >     /* FIXME: this should be auto-configured!  */
    >     #ifdef __MSDOS__
    >     # define CANT_FORK
    >     #endif
    > 
    > This seems to cry for either an Autoconf test (based on whether the
    > compiler defines __MSDOS__ or not) or perhaps the whole snippet should
    > be moved to config/i386/xm-go32.h.
    > 
    > 
    >     * cli-cmds.c:shell_escape()
    > 
    >     #ifdef __DJGPP__
    >       /* Make sure to return to the directory GDB thinks it is, in case the
    >        shell command we just ran changed it.  */
    >       chdir (current_directory);
    >     #endif
    > 
    > This code is there because the current working directory is a global
    > notion (as opposed to being private to each process on Posix
    > systems).  Windows ports, at least the non-Cygwin ones, probably want
    > this as well.  Suggestions how to test this, anyone?  Should we define
    > a GLOBAL_CURDIR macro (zero by default)?

    Here's the patch to fix these two issues.  Okay to commit?


    2001-06-10  Eli Zaretskii  <eliz@is.elta.co.il>

	    * cli/cli-cmds.c (shell_escape) [GLOBAL_CURDIR]: Condition the
	    call to chdir on this symbol rather than on __DJGPP__.
	    (CANT_FORK) [__MSDOS__]: Move from here...
	    * defs.h (CANT_FORK) [__MSDOS__]: ...to here.
	    (GLOBAL_CURDIR) [__MSDOS__]: Define.

Sorry for the delay.  The defs.h part is approved.

Re: [RFA] Re: DOS/Windows-specific code: cli-cmds.c

[Eli Zaretskii]

> Date: Thu, 21 Jun 2001 12:47:14 -0400
> From: David Taylor <taylor@candd.org>
> 
>     2001-06-10  Eli Zaretskii  <eliz@is.elta.co.il>
> 
> 	    * cli/cli-cmds.c (shell_escape) [GLOBAL_CURDIR]: Condition the
> 	    call to chdir on this symbol rather than on __DJGPP__.
> 	    (CANT_FORK) [__MSDOS__]: Move from here...
> 	    * defs.h (CANT_FORK) [__MSDOS__]: ...to here.
> 	    (GLOBAL_CURDIR) [__MSDOS__]: Define.
> 
> Sorry for the delay.  The defs.h part is approved.

Thanks.

Fernando, is the patch to cli-cmds.c okay with you?

Re: [RFA] Re: DOS/Windows-specific code: cli-cmds.c

[Eli Zaretskii]

> Date: Thu, 21 Jun 2001 12:47:14 -0400
> From: David Taylor <taylor@candd.org>
> 
>     Date: Sun, 10 Jun 2001 16:28:57 +0300 (IDT)
>     From: Eli Zaretskii <eliz@is.elta.co.il>
> 
>     > Date: Tue, 8 May 2001 14:41:53 +0300 (IDT)
>     > From: Eli Zaretskii <eliz@is.elta.co.il>
>     > 
>     >     * cli-cmds.c:
>     > 
>     >     /* FIXME: this should be auto-configured!  */
>     >     #ifdef __MSDOS__
>     >     # define CANT_FORK
>     >     #endif
>     > 
>     > This seems to cry for either an Autoconf test (based on whether the
>     > compiler defines __MSDOS__ or not) or perhaps the whole snippet should
>     > be moved to config/i386/xm-go32.h.
>     > 
>     > 
>     >     * cli-cmds.c:shell_escape()
>     > 
>     >     #ifdef __DJGPP__
>     >       /* Make sure to return to the directory GDB thinks it is, in case the
>     >        shell command we just ran changed it.  */
>     >       chdir (current_directory);
>     >     #endif
>     > 
>     > This code is there because the current working directory is a global
>     > notion (as opposed to being private to each process on Posix
>     > systems).  Windows ports, at least the non-Cygwin ones, probably want
>     > this as well.  Suggestions how to test this, anyone?  Should we define
>     > a GLOBAL_CURDIR macro (zero by default)?
> 
>     Here's the patch to fix these two issues.  Okay to commit?
> 
> 
>     2001-06-10  Eli Zaretskii  <eliz@is.elta.co.il>
> 
> 	    * cli/cli-cmds.c (shell_escape) [GLOBAL_CURDIR]: Condition the
> 	    call to chdir on this symbol rather than on __DJGPP__.
> 	    (CANT_FORK) [__MSDOS__]: Move from here...
> 	    * defs.h (CANT_FORK) [__MSDOS__]: ...to here.
> 	    (GLOBAL_CURDIR) [__MSDOS__]: Define.
> 
> Sorry for the delay.  The defs.h part is approved.

Fernando, could you please review this?  Thanks.

Re: [RFA] Re: DOS/Windows-specific code: cli-cmds.c

[Fernando Nasser]

I cant't find the original message.  Can someone send me a copy?

Thanks.

Fernando


Eli Zaretskii wrote:
> 
> > Date: Thu, 21 Jun 2001 12:47:14 -0400
> > From: David Taylor <taylor@candd.org>
> >
> >     Date: Sun, 10 Jun 2001 16:28:57 +0300 (IDT)
> >     From: Eli Zaretskii <eliz@is.elta.co.il>
> >
> >     > Date: Tue, 8 May 2001 14:41:53 +0300 (IDT)
> >     > From: Eli Zaretskii <eliz@is.elta.co.il>
> >     >
> >     >     * cli-cmds.c:
> >     >
> >     >     /* FIXME: this should be auto-configured!  */
> >     >     #ifdef __MSDOS__
> >     >     # define CANT_FORK
> >     >     #endif
> >     >
> >     > This seems to cry for either an Autoconf test (based on whether the
> >     > compiler defines __MSDOS__ or not) or perhaps the whole snippet should
> >     > be moved to config/i386/xm-go32.h.
> >     >
> >     >
> >     >     * cli-cmds.c:shell_escape()
> >     >
> >     >     #ifdef __DJGPP__
> >     >       /* Make sure to return to the directory GDB thinks it is, in case the
> >     >        shell command we just ran changed it.  */
> >     >       chdir (current_directory);
> >     >     #endif
> >     >
> >     > This code is there because the current working directory is a global
> >     > notion (as opposed to being private to each process on Posix
> >     > systems).  Windows ports, at least the non-Cygwin ones, probably want
> >     > this as well.  Suggestions how to test this, anyone?  Should we define
> >     > a GLOBAL_CURDIR macro (zero by default)?
> >
> >     Here's the patch to fix these two issues.  Okay to commit?
> >
> >
> >     2001-06-10  Eli Zaretskii  <eliz@is.elta.co.il>
> >
> >           * cli/cli-cmds.c (shell_escape) [GLOBAL_CURDIR]: Condition the
> >           call to chdir on this symbol rather than on __DJGPP__.
> >           (CANT_FORK) [__MSDOS__]: Move from here...
> >           * defs.h (CANT_FORK) [__MSDOS__]: ...to here.
> >           (GLOBAL_CURDIR) [__MSDOS__]: Define.
> >
> > Sorry for the delay.  The defs.h part is approved.
> 
> Fernando, could you please review this?  Thanks.

-- 
Fernando Nasser
Red Hat Canada Ltd.                     E-Mail:  fnasser@redhat.com
2323 Yonge Street, Suite #300
Toronto, Ontario   M4P 2C9

Re: [RFA] Re: DOS/Windows-specific code: cli-cmds.c

[Eli Zaretskii]

On Thu, 12 Jul 2001, Fernando Nasser wrote:

> I cant't find the original message.  Can someone send me a copy?

I don't have a copy, but you can find the original message here:

  http://sources.redhat.com/ml/gdb-patches/2001-06/msg00196.html

Re: [RFA] Re: DOS/Windows-specific code: cli-cmds.c

[Christopher Faylor]

On Thu, Jul 12, 2001 at 10:51:59AM +0300, Eli Zaretskii wrote:
>> Date: Thu, 21 Jun 2001 12:47:14 -0400
>> From: David Taylor <taylor@candd.org>
>> 
>>     Date: Sun, 10 Jun 2001 16:28:57 +0300 (IDT)
>>     From: Eli Zaretskii <eliz@is.elta.co.il>
>> 
>>     > Date: Tue, 8 May 2001 14:41:53 +0300 (IDT)
>>     > From: Eli Zaretskii <eliz@is.elta.co.il>
>>     > 
>>     >     * cli-cmds.c:
>>     > 
>>     >     /* FIXME: this should be auto-configured!  */
>>     >     #ifdef __MSDOS__
>>     >     # define CANT_FORK
>>     >     #endif
>>     > 
>>     > This seems to cry for either an Autoconf test (based on whether the
>>     > compiler defines __MSDOS__ or not) or perhaps the whole snippet should
>>     > be moved to config/i386/xm-go32.h.
>>     > 
>>     > 
>>     >     * cli-cmds.c:shell_escape()
>>     > 
>>     >     #ifdef __DJGPP__
>>     >       /* Make sure to return to the directory GDB thinks it is, in case the
>>     >        shell command we just ran changed it.  */
>>     >       chdir (current_directory);
>>     >     #endif
>>     > 
>>     > This code is there because the current working directory is a global
>>     > notion (as opposed to being private to each process on Posix
>>     > systems).  Windows ports, at least the non-Cygwin ones, probably want
>>     > this as well.  Suggestions how to test this, anyone?  Should we define
>>     > a GLOBAL_CURDIR macro (zero by default)?
>> 
>>     Here's the patch to fix these two issues.  Okay to commit?
>> 
>> 
>>     2001-06-10  Eli Zaretskii  <eliz@is.elta.co.il>
>> 
>> 	    * cli/cli-cmds.c (shell_escape) [GLOBAL_CURDIR]: Condition the
>> 	    call to chdir on this symbol rather than on __DJGPP__.
>> 	    (CANT_FORK) [__MSDOS__]: Move from here...
>> 	    * defs.h (CANT_FORK) [__MSDOS__]: ...to here.
>> 	    (GLOBAL_CURDIR) [__MSDOS__]: Define.
>> 
>> Sorry for the delay.  The defs.h part is approved.
>
>Fernando, could you please review this?  Thanks.

FWIW, Cygwin doesn't need any special chdir considerations.  This
should be a MSDOS only issue.

cgf

Re: [RFA] Re: DOS/Windows-specific code: cli-cmds.c

[Eli Zaretskii]

> Date: Thu, 12 Jul 2001 13:21:45 -0400
> From: Christopher Faylor <cgf@redhat.com>
> >> 
> >>     2001-06-10  Eli Zaretskii  <eliz@is.elta.co.il>
> >> 
> >> 	    * cli/cli-cmds.c (shell_escape) [GLOBAL_CURDIR]: Condition the
> >> 	    call to chdir on this symbol rather than on __DJGPP__.
> >> 	    (CANT_FORK) [__MSDOS__]: Move from here...
> >> 	    * defs.h (CANT_FORK) [__MSDOS__]: ...to here.
> >> 	    (GLOBAL_CURDIR) [__MSDOS__]: Define.
> >> 
> >> Sorry for the delay.  The defs.h part is approved.
> >
> >Fernando, could you please review this?  Thanks.
> 
> FWIW, Cygwin doesn't need any special chdir considerations.  This
> should be a MSDOS only issue.

Well, that's why GLOBAL_CURDIR is not defined for Cygwin... ;-)

Re: [RFA] Re: DOS/Windows-specific code: cli-cmds.c

[Christopher Faylor]

On Thu, Jul 12, 2001 at 09:22:33PM +0300, Eli Zaretskii wrote:
>> Date: Thu, 12 Jul 2001 13:21:45 -0400
>> From: Christopher Faylor <cgf@redhat.com>
>> >> 
>> >>     2001-06-10  Eli Zaretskii  <eliz@is.elta.co.il>
>> >> 
>> >> 	    * cli/cli-cmds.c (shell_escape) [GLOBAL_CURDIR]: Condition the
>> >> 	    call to chdir on this symbol rather than on __DJGPP__.
>> >> 	    (CANT_FORK) [__MSDOS__]: Move from here...
>> >> 	    * defs.h (CANT_FORK) [__MSDOS__]: ...to here.
>> >> 	    (GLOBAL_CURDIR) [__MSDOS__]: Define.
>> >> 
>> >> Sorry for the delay.  The defs.h part is approved.
>> >
>> >Fernando, could you please review this?  Thanks.
>> 
>> FWIW, Cygwin doesn't need any special chdir considerations.  This
>> should be a MSDOS only issue.
>
>Well, that's why GLOBAL_CURDIR is not defined for Cygwin... ;-)

Ok.  I'm just responding since I am being Cc'ed in this email.

I've redirected future responses back to gdb-patches only.

cgf

Re: [RFA] Re: DOS/Windows-specific code: cli-cmds.c

[Fernando Nasser]

Eli Zaretskii wrote:
> 
> > Date: Thu, 21 Jun 2001 12:47:14 -0400
> > From: David Taylor <taylor@candd.org>
> >
> >     Date: Sun, 10 Jun 2001 16:28:57 +0300 (IDT)
> >     From: Eli Zaretskii <eliz@is.elta.co.il>
> >
> >     > Date: Tue, 8 May 2001 14:41:53 +0300 (IDT)
> >     > From: Eli Zaretskii <eliz@is.elta.co.il>
> >     >
> >     >     * cli-cmds.c:
> >     >
> >     >     /* FIXME: this should be auto-configured!  */
> >     >     #ifdef __MSDOS__
> >     >     # define CANT_FORK
> >     >     #endif
> >     >
> >     > This seems to cry for either an Autoconf test (based on whether the
> >     > compiler defines __MSDOS__ or not) or perhaps the whole snippet should
> >     > be moved to config/i386/xm-go32.h.
> >     >
> >     >
> >     >     * cli-cmds.c:shell_escape()
> >     >
> >     >     #ifdef __DJGPP__
> >     >       /* Make sure to return to the directory GDB thinks it is, in case the
> >     >        shell command we just ran changed it.  */
> >     >       chdir (current_directory);
> >     >     #endif
> >     >
> >     > This code is there because the current working directory is a global
> >     > notion (as opposed to being private to each process on Posix
> >     > systems).  Windows ports, at least the non-Cygwin ones, probably want
> >     > this as well.  Suggestions how to test this, anyone?  Should we define
> >     > a GLOBAL_CURDIR macro (zero by default)?
> >
> >     Here's the patch to fix these two issues.  Okay to commit?
> >
> >
> >     2001-06-10  Eli Zaretskii  <eliz@is.elta.co.il>
> >
> >           * cli/cli-cmds.c (shell_escape) [GLOBAL_CURDIR]: Condition the
> >           call to chdir on this symbol rather than on __DJGPP__.
> >           (CANT_FORK) [__MSDOS__]: Move from here...
> >           * defs.h (CANT_FORK) [__MSDOS__]: ...to here.
> >           (GLOBAL_CURDIR) [__MSDOS__]: Define.
> >
> > Sorry for the delay.  The defs.h part is approved.
> 
> Fernando, could you please review this?  Thanks.

OK to commit.

Sorry for the delay.

Regards,
Fernando

-- 
Fernando Nasser
Red Hat - Toronto                       E-Mail:  fnasser@redhat.com
2323 Yonge Street, Suite #300
Toronto, Ontario   M4P 2C9

Re: [RFA] Re: DOS/Windows-specific code: cli-cmds.c

[Eli Zaretskii]

On Thu, 12 Jul 2001, Fernando Nasser wrote:

> > >     2001-06-10  Eli Zaretskii  <eliz@is.elta.co.il>
> > >
> > >           * cli/cli-cmds.c (shell_escape) [GLOBAL_CURDIR]: Condition the
> > >           call to chdir on this symbol rather than on __DJGPP__.
> > >           (CANT_FORK) [__MSDOS__]: Move from here...
> > >           * defs.h (CANT_FORK) [__MSDOS__]: ...to here.
> > >           (GLOBAL_CURDIR) [__MSDOS__]: Define.
> > >
> > > Sorry for the delay.  The defs.h part is approved.
> > 
> > Fernando, could you please review this?  Thanks.
> 
> OK to commit.

Committed.  Thanks.

[RFA] Re: DOS/Windows-specific code: cli-cmds.c

[Eli Zaretskii]

> Date: Tue, 8 May 2001 14:41:53 +0300 (IDT)
> From: Eli Zaretskii <eliz@is.elta.co.il>
> 
>     * cli-cmds.c:
> 
>     /* FIXME: this should be auto-configured!  */
>     #ifdef __MSDOS__
>     # define CANT_FORK
>     #endif
> 
> This seems to cry for either an Autoconf test (based on whether the
> compiler defines __MSDOS__ or not) or perhaps the whole snippet should
> be moved to config/i386/xm-go32.h.
> 
> 
>     * cli-cmds.c:shell_escape()
> 
>     #ifdef __DJGPP__
>       /* Make sure to return to the directory GDB thinks it is, in case the
> 	 shell command we just ran changed it.  */
>       chdir (current_directory);
>     #endif
> 
> This code is there because the current working directory is a global
> notion (as opposed to being private to each process on Posix
> systems).  Windows ports, at least the non-Cygwin ones, probably want
> this as well.  Suggestions how to test this, anyone?  Should we define
> a GLOBAL_CURDIR macro (zero by default)?

Here's the patch to fix these two issues.  Okay to commit?


2001-06-10  Eli Zaretskii  <eliz@is.elta.co.il>

	* cli/cli-cmds.c (shell_escape) [GLOBAL_CURDIR]: Condition the
	call to chdir on this symbol rather than on __DJGPP__.
	(CANT_FORK) [__MSDOS__]: Move from here...
	* defs.h (CANT_FORK) [__MSDOS__]: ...to here.
	(GLOBAL_CURDIR) [__MSDOS__]: Define.

--- gdb/defs.h~0	Mon Jun  4 09:38:04 2001
+++ gdb/defs.h	Sun Jun 10 16:08:22 2001
@@ -1372,6 +1372,11 @@
 #endif
 #endif
 
+#ifdef __MSDOS__
+# define CANT_FORK
+# define GLOBAL_CURDIR
+#endif
+
 /* Provide default definitions of PIDGET, TIDGET, and MERGEPID.
    The name ``TIDGET'' is a historical accident.  Many uses of TIDGET
    in the code actually refer to a lightweight process id, i.e,
--- gdb/cli/cli-cmds.c~0	Wed Jun  6 12:25:24 2001
+++ gdb/cli/cli-cmds.c	Sun Jun 10 16:10:18 2001
@@ -39,11 +39,6 @@
 #define GDBINIT_FILENAME        ".gdbinit"
 #endif
 
-/* FIXME: this should be auto-configured!  */
-#ifdef __MSDOS__
-# define CANT_FORK
-#endif
-
 /* From gdb/top.c */
 
 extern void dont_repeat (void);
@@ -454,7 +449,7 @@
       fprintf_unfiltered (gdb_stderr, "%s exited with status %d\n", arg, rc);
       gdb_flush (gdb_stderr);
     }
-#ifdef __DJGPP__
+#ifdef GLOBAL_CURDIR
   /* Make sure to return to the directory GDB thinks it is, in case the
      shell command we just ran changed it.  */
   chdir (current_directory);