From: Ozkan Sezer <sezeroz@gmail.com>
To: gdb-patches@sourceware.org
Subject: gdbserver with -Werror, win64 socket type
Date: Fri, 27 Aug 2010 13:56:00 -0000 [thread overview]
Message-ID: <AANLkTi=BqExXKgwgVS5cXgrjK-g-0P=GYEOhK9P6ciM_@mail.gmail.com> (raw)
Hi:
gdb assumes that the windows socket handles and file descriptors can be
used interchangeably as in unix, but it can not: the SOCKET type in
windows is UINT_PTR (because it is a handle) and INVALID_SOCKET used in
gdbserver/remote-utils.c is (SOCKET)(~0), but it is assigned to signed
int typed data.
Now, for w32 this wraps to -1. For w64, however, it causes the following
warnings:
../../../gdb-cvs/gdb/gdbserver/remote-utils.c:110: error: overflow in
implicit constant conversion
../../../gdb-cvs/gdb/gdbserver/remote-utils.c:111: error: overflow in
implicit constant conversion
../../../gdb-cvs/gdb/gdbserver/remote-utils.c: In function 'remote_close':
../../../gdb-cvs/gdb/gdbserver/remote-utils.c:357: error: overflow in
implicit constant conversion
Normally, we at mingw-w64 define the SOCKET type as INT_PTR, ie. signed
intptr to just _workaround_ such issues. But I got bit by the above
warnings + -Werror when I was testing compilation using the correct
UINT_PTR SOCKET type because of definitions in remote-utils.c line #79.
I know that, with the gdb source as it is now, the file descriptor and
socket interchangability issue is not an easy fix. However, we can just
add an INT_PTR cast to INVALID_DESCRIPTOR definition along with a fixme
note, like:
Index: remote-utils.c
===================================================================
RCS file: /cvs/src/src/gdb/gdbserver/remote-utils.c,v
retrieving revision 1.80
diff -u -p -r1.80 remote-utils.c
--- remote-utils.c 26 Aug 2010 16:24:41 -0000 1.80
+++ remote-utils.c 27 Aug 2010 13:50:11 -0000
@@ -77,7 +77,9 @@ typedef int socklen_t;
#ifndef IN_PROCESS_AGENT
#if USE_WIN32API
-# define INVALID_DESCRIPTOR INVALID_SOCKET
+/* FIXME: we are using windows socket handles and
+ file descriptors interchangeably, it is wrong. */
+# define INVALID_DESCRIPTOR (INT_PTR)INVALID_SOCKET
#else
# define INVALID_DESCRIPTOR -1
#endif
Ideas?
--
Ozkan
next reply other threads:[~2010-08-27 13:56 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-08-27 13:56 Ozkan Sezer [this message]
2010-08-27 14:45 ` Pedro Alves
2010-08-27 15:09 ` Ozkan Sezer
2010-08-27 15:34 ` Pedro Alves
2010-08-27 15:46 ` Ozkan Sezer
2010-08-27 16:06 ` Pedro Alves
2010-08-27 16:12 ` Ozkan Sezer
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='AANLkTi=BqExXKgwgVS5cXgrjK-g-0P=GYEOhK9P6ciM_@mail.gmail.com' \
--to=sezeroz@gmail.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