From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 12005 invoked by alias); 25 Feb 2016 01:06:38 -0000 Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org Received: (qmail 11994 invoked by uid 89); 25 Feb 2016 01:06:37 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-1.9 required=5.0 tests=BAYES_00,RP_MATCHES_RCVD,SPF_HELO_PASS autolearn=ham version=3.3.2 spammy=inconsistent, propogate, gary's, garys X-HELO: mx1.redhat.com Received: from mx1.redhat.com (HELO mx1.redhat.com) (209.132.183.28) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES256-GCM-SHA384 encrypted) ESMTPS; Thu, 25 Feb 2016 01:06:36 +0000 Received: from int-mx14.intmail.prod.int.phx2.redhat.com (int-mx14.intmail.prod.int.phx2.redhat.com [10.5.11.27]) by mx1.redhat.com (Postfix) with ESMTPS id 22CD8804F7; Thu, 25 Feb 2016 01:06:35 +0000 (UTC) Received: from [127.0.0.1] (ovpn01.gateway.prod.ext.phx2.redhat.com [10.5.9.1]) by int-mx14.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id u1P16XKo030177; Wed, 24 Feb 2016 20:06:33 -0500 Message-ID: <56CE5399.5060907@redhat.com> Date: Thu, 25 Feb 2016 01:06:00 -0000 From: Pedro Alves User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.7.0 MIME-Version: 1.0 To: Luis Machado , gdb-patches@sourceware.org CC: gbenson@redhat.com Subject: Re: [PATCH 1/2] Debugging without a binary (regression) References: <1456324014-17961-1-git-send-email-lgustavo@codesourcery.com> In-Reply-To: <1456324014-17961-1-git-send-email-lgustavo@codesourcery.com> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit X-SW-Source: 2016-02/txt/msg00763.txt.bz2 On 02/24/2016 02:26 PM, Luis Machado wrote: > When we attempt to debug a process using GDBserver in standard remote mode > without a symbol file on GDB's end, we may run into an issue where GDB cuts > the connection attempt short due to an error. The error is caused by not > being able to open a symbol file, like so: > > -- > > (gdb) set sysroot > (gdb) tar rem :2345 > Remote debugging using :2345 > /proc/23769/exe: Permission denied. > (gdb) i r > The program has no registers now. > (gdb) > > It should've been like this: > > (gdb) set sysroot > (gdb) tar rem :2345 > Remote debugging using :2345 > 0xf7ddb2d0 in ?? () > (gdb) i r > eax 0x0 0 > ecx 0x0 0 > edx 0x0 0 > ebx 0x0 0 > esp 0xffffdfa0 0xffffdfa0 > ebp 0x0 0x0 > esi 0x0 0 > edi 0x0 0 > eip 0xf7ddb2d0 0xf7ddb2d0 > eflags 0x200 [ IF ] > cs 0x33 51 > ss 0x2b 43 > ds 0x0 0 > es 0x0 0 > fs 0x0 0 > gs 0x0 0 > (gdb) > > This is caused by a couple of function calls within exec_file_locate_attach > that can potentially throw errors. > > The following patch guards both exec_file_attach and symbol_file_add_main to > prevent the errors from disrupting the connection process. > > There was also a case where native GDB tripped on this problem, but it was > mostly fixed by bf74e428bca61022bd5cdf6bf28789a184748b4d. > > Regression-tested on x86-64/Ubuntu. > > gdb/ChangeLog: > > 2016-02-24 Luis Machado > > * exec.c (exec_file_locate_attach): Guard a couple functions > that can throw errors. > --- > gdb/exec.c | 31 +++++++++++++++++++++++++++++-- > 1 file changed, 29 insertions(+), 2 deletions(-) > > diff --git a/gdb/exec.c b/gdb/exec.c > index 90811c0..3f77b3d 100644 > --- a/gdb/exec.c > +++ b/gdb/exec.c > @@ -176,8 +176,35 @@ exec_file_locate_attach (int pid, int from_tty) > > old_chain = make_cleanup (xfree, full_exec_path); > > - exec_file_attach (full_exec_path, from_tty); > - symbol_file_add_main (full_exec_path, from_tty); > + /* exec_file_attach and symbol_file_add_main may throw an error if the file > + cannot be opened either locally or remotely. This happens when, for > + example, GDB connects to a GDBserver that is running on a different > + filesystem and the sysroot is set to non-target-based (no "target:"). The second sentence no longer makes sense after Gary's change. I think it should read: This happens for example, when the file is first found in the local sysroot (above), and then disappears (a TOCTOU race), or when it doesn't exist in the target filesystem, or when the file does exist, but is not readable. > + > + Then GDB will neither load the binary from the target nor be able to > + load a binary from the local filesystem (it may not exist in the local > + filesystem in the same path as in the remote filesystem). No longer makes sense either. > + > + Even without a binary, the remote-based debugging session should > + continue normally instead of ending abruptly. Hence we catch thrown > + errors/exceptions in the following code. */ > + TRY > + { > + exec_file_attach (full_exec_path, from_tty); > + } > + CATCH (err, RETURN_MASK_ALL) This should be RETURN_MASK_ERROR instead. Silently swallowing ctrl-c is rarely the right thing to do, if ever. A ctrl-c during the initial remote connection is supposed to bring down the connection immediately: ~~~ /* Start the remote connection. If error() or QUIT, discard this target (we'd otherwise be in an inconsistent state) and then propogate the error on up the exception chain. This ensures that the caller doesn't stumble along blindly assuming that the function succeeded. ~~~ static void remote_start_remote (int from_tty, struct target_ops *target, int extended_p) { struct remote_state *rs = get_remote_state (); struct packet_config *noack_config; char *wait_status = NULL; immediate_quit++; /* Allow user to interrupt it. */ QUIT; ~~~ We should probably similarly completely tear down an incomplete attach interrupted by ctrl-c too, but that's a separate issue. > + { > + } > + END_CATCH > + > + TRY > + { > + symbol_file_add_main (full_exec_path, from_tty); > + } > + CATCH (err, RETURN_MASK_ALL) > + { > + } Throwing is bad, but, should we warn, including the exception message? > + END_CATCH > > do_cleanups (old_chain); > } > Thanks, Pedro Alves