DomainKey-Signature: a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id
	:list-unsubscribe:list-subscribe:list-archive:list-post
	:list-help:sender:from:to:subject:date:message-id:references
	:in-reply-to:content-type:content-id:content-transfer-encoding
	:mime-version; q=dns; s=default; b=L/hHAzn01Y9uHpTdJi9xjqUCE18AO
	2Y+WB2MKj3G9QbuT6fJH8LVK0ejeTbUbRA3wgNDbNyv36e/Zd11TjP5419LTQHkR
	3BK5+QhO1HVfBdxaTs8/VCqGVHSlrqap44rMqukFhBiGe8ZRjrK7c2mOmfgbFqtt
	BX4D9RDNCAXy9w=
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
From: Ken Brown <kbrown AT cornell DOT edu>
To: "cygwin AT cygwin DOT com" <cygwin AT cygwin DOT com>
Subject: Re: MC no longer working with cygwin 3.1.x
Date: Wed, 18 Dec 2019 21:14:42 +0000
Message-ID: <c3809f52-1842-83d6-6692-53f6caec3491@cornell.edu>
References: <DB8PR01MB5996973955794728D12078E6BC530 AT DB8PR01MB5996 DOT eurprd01 DOT prod DOT exchangelabs DOT com> <97c8a719-73b9-e0ec-9651-97c2c684a086 AT gmail DOT com> <20191218193205 DOT GP10310 AT calimero DOT vinschen DOT de>
In-Reply-To: <20191218193205.GP10310@calimero.vinschen.de>
user-agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:60.0) Gecko/20100101 Thunderbird/60.9.1
received-spf: None (protection.outlook.com: cornell.edu does not designate permitted sender hosts)
Content-Type: text/plain; charset="Windows-1252"
Content-ID: <676776F443BEE94BB19651D761A877F8 AT namprd04 DOT prod DOT outlook DOT com>
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-userprincipalname: EYy5MYoS4YWkSMC0ojHLCgqT/VcZ66luxVVkx7k0iTO/mf8+R37sTS4YaYRSu7rkcCrBPTqDXPbksOkV1y3zkQ==
X-IsSubscribed: yes
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by delorie.com id xBILFK6F005277

On 12/18/2019 2:32 PM, Corinna Vinschen wrote:
> On Dec 18 17:23, Marco Atzeri wrote:
>> Am 18.12.2019 um 15:42 schrieb Laurent R:
>>> Dear all,
>>>
>>> After the update to cygwin-3.1.1, mc (aka Midnight Commander) no longer starts. It goes into an infinite loop, if I go back to cygwin-3.0.7, mc works perfectly.
>>>
>>> I tested with cygwin-3.1.0, same problem.
>>>
>>> The MC version is 4.8.23 and the operating system is Windows 10 Pro x64 1903.
>>>
>>> Regards.
>>>
>>
>> confirmed
>>
>> there must be a race somewhere as running /usr/bin/mc.exe from strace
>> works fine.
> 
> But that's not the only problem running mc AFAICS.  I'm seeing an
> error message in the background:
> 
> Cannot open named pipe /tmp/mc-corinna/mc.pipe.384
>    /cygdrive/d/cyg_pub/devel/mc/mc-4.8.23-1.x86_64/src/mc-4.8.23/src/subshell/common.c: open: Permission denied
> 
> It looks like there's another problem, this time with the new FIFO
> code.
> 
> Ken, care to take a look?

I've just looked at the mc source code, and it seems that (under tcsh only), mc 
creates a fifo and tries to open it twice as O_RDWR.  This is not supported on 
Cygwin, so the second attempt fails.

Here's the relevant code from src/subshell/common.c:

         if (mc_global.shell->type == SHELL_TCSH)
         {
             g_snprintf (tcsh_fifo, sizeof (tcsh_fifo), "%s/mc.pipe.%d",
                         mc_tmpdir (), (int) getpid ());
             if (mkfifo (tcsh_fifo, 0600) == -1)
             [...]
             /* Opening the FIFO as O_RDONLY or O_WRONLY causes deadlock */

             if ((subshell_pipe[READ] = open (tcsh_fifo, O_RDWR)) == -1
                 || (subshell_pipe[WRITE] = open (tcsh_fifo, O_RDWR)) == -1)
             {
                 fprintf (stderr, _("Cannot open named pipe %s\n"), tcsh_fifo);
                 perror (__FILE__ ": open");
                 mc_global.tty.use_subshell = FALSE;
                 [...]

Under bash, zsh, and some other shells, mc uses ordinary pipes rather than 
FIFOs.  I have no idea why tcsh is treated differently.

Until someone[*] gets around to making it possible to open a FIFO twice for 
reading, I don't think there's anything we can do about it.  I also don't know 
how important this is for mc.  I don't use tcsh or mc, so I can't judge.

Ken

[*] I will probably look into this at some point if there's enough demand for 
it, but I'd like the new FIFO code to get lots of testing first.

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple