This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [RFC] Mingw Windows 64-bit gdbserver
- From: Pedro Alves <pedro at codesourcery dot com>
- To: "Pierre Muller" <pierre dot muller at ics-cnrs dot unistra dot fr>
- Cc: gdb-patches at sourceware dot org
- Date: Fri, 16 Apr 2010 18:15:36 +0100
- Subject: Re: [RFC] Mingw Windows 64-bit gdbserver
- References: <000d01cadd79$efa9e2b0$cefda810$@muller@ics-cnrs.unistra.fr> <201004161727.26943.pedro@codesourcery.com> <001f01cadd83$3e5e6b70$bb1b4250$@muller@ics-cnrs.unistra.fr>
On Friday 16 April 2010 17:38:28, Pierre Muller wrote:
> GDb sources is still a mix:
>
> > Typo: I meant winsock2.h instead of winsock.h.
>
> $ grep "include.*<winso" *.h gdbserver/*.h *.c gdbserver/*.c
> gdb_select.h:#include <winsock2.h>
> serial.h:#include <winsock2.h>
> m32r-rom.c:#include <winsock.h>
> remote-m32r-sdi.c:#include <winsock.h>
> ser-base.c:#include <winsock2.h>
> ser-tcp.c:#include <winsock2.h>
> gdbserver/gdbreplay.c:#include <winsock.h>
> gdbserver/remote-utils.c:#include <winsock.h>
>
> should all be moved to winsock2.h?
Probably not a real problem as we only activate
Winsock 1:
gdb/
>grep WSAStart *.c -rn
m32r-rom.c:451: if (WSAStartup (MAKEWORD (1, 1), &wd))
ser-mingw.c:1255: if (WSAStartup (MAKEWORD (1, 0), &wsa_data) != 0)
gdb/gdbserver/
>grep WSAStart *.c -rn
gdbreplay.c:207: WSAStartup (MAKEWORD (1, 0), &wsad);
remote-utils.c:288: WSAStartup (MAKEWORD (1, 0), &wsad);
Presumably winsock2.h is a superset of winsock.h and so we're
likely fine. Wouldn't hurt to be consistent, moreso since we
include winsock2.h in some headers.
> and -lws2_32 as library?
You mean in gdbserver? Shouldn't hurt, yeah.
> By the way, do the m32r sources still link correctly currently?
No idea. There were recent changes to some of those
files though.
--
Pedro Alves