Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: "Pierre Muller" <pierre.muller@ics-cnrs.unistra.fr>
To: "'Joel Brobecker'" <brobecker@adacore.com>
Cc: <gdb-patches@sourceware.org>
Subject: RE: [RFA] mingw port: Allow use of cvs GDB on Windows 95
Date: Mon, 21 Feb 2011 11:58:00 -0000	[thread overview]
Message-ID: <009301cbd1bd$88ed0550$9ac70ff0$@muller@ics-cnrs.unistra.fr> (raw)
In-Reply-To: <20110221102151.GG2600@adacore.com>

> > 2011-02-05  Pierre Muller  <muller@ics.u-strasbg.fr>
> >
> > 	Allow use of mingw native on Windows 95 OS.
> > 	* src/gdb/ser-mingw.c (CancelIo): New macro for dynamically
> loaded
> > 	DLL entry.
> > 	(ser_windows_close): Only call CancelIo if function exists.
> > 	(_initialize_ser_windows): Use LoadLirary/GetProcAddress
> > 	to check for existence of CancelIo function in kernel32 DLL.
> 
> I was hoping that someone with more Windows knowledge would be able
> to review this patch, but oh well...

  Thanks for doing this!
 
> This patch looks OK to me (ie, no risk that I can see for the other
> platforms).  Just a couple of remarks:
> 
> > +/* CancelIo is not available for Windows 95 OS,
> > +   so we need to use LoadLibrary/GetProcAddress to avoid
> > +   a startup failure.  */
> 
> I find that the length of your lines are consistently really short
> and inconsistent.  I think it makes it harder to read.  The above
> can be reformatted to:
> 
> /* CancelIo is not available for Windows 95 OS, so we need to use
>    LoadLibrary/GetProcAddress to avoid a startup failure.  */
> 
> (but I think you could also look at your editor to see if there is
> anything that's contributing to this, especially the inconsistency
> - are you using fixed fonts?)

  No, it's just that I do like to have sentences in their own lines...
I should try to adhere closer to GNU standards...

 
> > +#define CancelIo dyn_CancelIo
> > +
> > +static BOOL WINAPI (*CancelIo) (HANDLE);
> 
> I would *personally* keep these two together, to make it clearer
> that the comment above applies to both.  Just a thought, so don't
> feel obligated to follow the suggestion if you don't agree.

I removed the empty line in between.
 
> >    /* Stop any pending selects.  */
> > -  CancelIo ((HANDLE) _get_osfhandle (scb->fd));
> > +  if (CancelIo)
> > +    CancelIo ((HANDLE) _get_osfhandle (scb->fd));
> 
> Can you add a comment explaining why it's OK to not call CancelIo
> if it is not defined? I think you were trying to get more info
> about this on this list - whatever information that you might have
> gleaned...

  I added some explanation, see below.

> The patch is pre-approved once the comments are addressed.

Thanks for the approval,
for the record, here is what I just committed.


Pierre Muller
GDB pascal language maintainer

Index: ChangeLog
===================================================================
RCS file: /cvs/src/src/gdb/ChangeLog,v
retrieving revision 1.12615
diff -u -p -r1.12615 ChangeLog
--- ChangeLog	21 Feb 2011 08:38:10 -0000	1.12615
+++ ChangeLog	21 Feb 2011 11:46:15 -0000
@@ -1,3 +1,12 @@
+2011-02-21  Pierre Muller  <muller@ics.u-strasbg.fr>
+
+	Allow use of mingw native on Windows 95 OS.
+	* src/gdb/ser-mingw.c (CancelIo): New macro for dynamically loaded
+	DLL entry.
+	(ser_windows_close): Only call CancelIo if function exists.
+	(_initialize_ser_windows): Use LoadLirary/GetProcAddress
+	to check for existence of CancelIo function in kernel32 DLL.
+
 2011-02-21  Hui Zhu  <teawater@gmail.com>
 
 	* Makefile.in (HFILES_NO_SRCDIR): Add printcmd.h.
Index: ser-mingw.c
===================================================================
RCS file: /cvs/src/src/gdb/ser-mingw.c,v
retrieving revision 1.26
diff -u -p -r1.26 ser-mingw.c
--- ser-mingw.c	11 Jan 2011 21:53:23 -0000	1.26
+++ ser-mingw.c	21 Feb 2011 11:46:15 -0000
@@ -45,6 +45,11 @@ struct ser_windows_state
   HANDLE except_event;
 };
 
+/* CancelIo is not available for Windows 95 OS, so we need to use
+   LoadLibrary/GetProcAddress to avoid a startup failure.  */
+#define CancelIo dyn_CancelIo
+static BOOL WINAPI (*CancelIo) (HANDLE);
+
 /* Open up a real live device for serial I/O.  */
 
 static int
@@ -216,8 +221,12 @@ ser_windows_close (struct serial *scb)
 {
   struct ser_windows_state *state;
 
-  /* Stop any pending selects.  */
-  CancelIo ((HANDLE) _get_osfhandle (scb->fd));
+  /* Stop any pending selects. On Windows 95 OS, CancelIo function does not
+     exist. In that case, it can be replaced by a call to CloseHandle, but
+     this is not necessary here as we do close the Windows handle by
calling
+     close (scb->fd) below.  */
+  if (CancelIo)
+    CancelIo ((HANDLE) _get_osfhandle (scb->fd));
   state = scb->state;
   CloseHandle (state->ov.hEvent);
   CloseHandle (state->except_event);
@@ -1208,8 +1217,19 @@ _initialize_ser_windows (void)
   WSADATA wsa_data;
   struct serial_ops *ops;
 
-  /* First register the serial port driver.  */
+  HMODULE hm = NULL;
+
+  /* First find out if kernel32 exports CancelIo function.  */
+  hm = LoadLibrary ("kernel32.dll");
+  if (hm)
+    {
+      CancelIo = (void *) GetProcAddress (hm, "CancelIo");
+      FreeLibrary (hm);
+    }
+  else
+    CancelIo = NULL;
 
+  /* Now register the serial port driver.  */
   ops = XMALLOC (struct serial_ops);
   memset (ops, 0, sizeof (struct serial_ops));
   ops->name = "hardwire";


  reply	other threads:[~2011-02-21 11:50 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-02-04 13:06 [RFC] " Compte ICS
2011-02-04 13:57 ` Pierre Muller
2011-02-05 20:38   ` [RFA] " Pierre Muller
2011-02-21 11:06     ` Joel Brobecker
2011-02-21 11:58       ` Pierre Muller [this message]
2011-02-21 12:52         ` Joel Brobecker
2011-02-21 13:35           ` Pierre Muller
2011-02-21 14:59             ` Joel Brobecker

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='009301cbd1bd$88ed0550$9ac70ff0$@muller@ics-cnrs.unistra.fr' \
    --to=pierre.muller@ics-cnrs.unistra.fr \
    --cc=brobecker@adacore.com \
    --cc=gdb-patches@sourceware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox