X-Recipient: archive-cygwin AT delorie DOT com
X-SWARE-Spam-Status: No, hits=2.2 required=5.0	tests=AWL,BAYES_00,BUG6152_INVALID_DATE_TZ_ABSURD,INVALID_DATE_TZ_ABSURD,L_SPAM_TOOL_13,TO_NO_BRKTS_PCNT
X-Spam-Check-By: sourceware.org
Date: Mon, 1 Aug 2011 16:12:54 +0159
From: Denis Excoffier <Cygwin AT Denis-Excoffier DOT org>
To: cygwin AT cygwin DOT com
Subject: Re: bash-4.2 and symlink to folder that turns to be not executable
Message-ID: <20110801141318.GA1696@po8371>
References: <20110801115922 DOT GA1124 AT po8371> <20110801122425 DOT GX26203 AT calimero DOT vinschen DOT de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20110801122425.GX26203@calimero.vinschen.de>
User-Agent: Mutt/1.5.21 (2010-09-15)
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
Precedence: bulk
List-Id: <cygwin.cygwin.com>
List-Unsubscribe: <mailto:cygwin-unsubscribe-archive-cygwin=delorie DOT com AT cygwin DOT com>
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com


On Mon, Aug 01, 2011 at 02:24:25PM +0200, Corinna Vinschen wrote:
>> 
>> However, at a first glance I doubt this is a Cygwin bug.  Consider:
>> 
>> - bash 4.1.10 returns "execut"
>> - /bin/test from coreustils returns "execut"
>> - My highly professional and streamlined testcase returns the correct
>>   result:
>> 
>>     #include <stdio.h>
>>     #include <unistd.h>
>>     #include <errno.h>
>>     #include <string.h>
>> 
>>     void
>>     my_access (const char *file, int flag, const char *fname, int effective)
>>     {
>>       int ret = effective ? access (file, flag) : eaccess (file, flag);
>>       printf ("%saccess (%s, %s) = %d",
>> 	      effective ? "e" : " ", file, fname, ret);
>>       if (ret)
>> 	printf (" <%s>", strerror (errno));
>>       fputc ('\n', stdout);
>>     }
>> 
>>     int
>>     main (int argc, char**argv)
>>     {
>>       int eff;
>> 
>>       if (argc > 1)
>> 	for (eff = 0; eff < 2; ++eff)
>> 	  {
>> 	    my_access (argv[1], F_OK, "F_OK", eff);
>> 	    my_access (argv[1], R_OK, "R_OK", eff);
>> 	    my_access (argv[1], W_OK, "W_OK", eff);
>> 	    my_access (argv[1], X_OK, "X_OK", eff);
>> 	  }
>>       return 0;
>>     }
>> 
>>   $ gcc -g -o access access.c
>>   $ ./access /tmp/xxx/folder
>>    access (/tmp/xxx/folder, F_OK) = 0
>>    access (/tmp/xxx/folder, R_OK) = 0
>>    access (/tmp/xxx/folder, W_OK) = 0
>>    access (/tmp/xxx/folder, X_OK) = 0
>>   eaccess (/tmp/xxx/folder, F_OK) = 0
>>   eaccess (/tmp/xxx/folder, R_OK) = 0
>>   eaccess (/tmp/xxx/folder, W_OK) = 0
>>   eaccess (/tmp/xxx/folder, X_OK) = 0
>> 
>> 

Right, and me too. But if i replace in your testcase:
eaccess(file, flag) with
faccessat(0 /* not used if file is absolute */, file, flag, AT_EACCESS)
(and also "e" : " " with " " : "f" for clarity)
(and i have to include <fcntl.h>)

i obtain:

% ./corinna++ /tmp/xxx/folder
faccess (/tmp/xxx/folder, F_OK) = 0
faccess (/tmp/xxx/folder, R_OK) = 0
faccess (/tmp/xxx/folder, W_OK) = 0
faccess (/tmp/xxx/folder, X_OK) = -1 <Permission denied>
 access (/tmp/xxx/folder, F_OK) = 0
 access (/tmp/xxx/folder, R_OK) = 0
 access (/tmp/xxx/folder, W_OK) = 0
 access (/tmp/xxx/folder, X_OK) = 0
%

(faccess is what i find in bash-4.2/lib/sh/eaccess.c)

Is this under control?


>> Corinna
Denis Excoffier.

--
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