From mboxrd@z Thu Jan  1 00:00:00 1970
From: walter harms <wharms@bfs.de>
Subject: Re: Documenting sigaltstack SS_AUTODISRM
Date: Mon, 30 Oct 2017 11:21:45 +0100
Message-ID: <59F6FD39.40502@bfs.de>
References: <CAKgNAkjw_OBxcL627V9xMrKUei69hwjVCr6sbHbG2riBe7VVYw@mail.gmail.com> <08467ae1-7187-3b2a-9a78-8af0c10bf816@list.ru> <3907bc2a-0645-8d93-6ee5-3f99874e7022@gmail.com> <32d95303-5839-9279-a1d3-a06f34e3484e@list.ru> <CAKgNAkgw6P5RAsA2RSFJX57b=DHM=eNZ+ZOoagcO3ydSHzBcQA@mail.gmail.com> <50de8f3b-8a1e-df50-b5dd-d1b74cb77fad@list.ru> <026308b5-4e92-4439-1eb2-82b67584d548@gmail.com> <3a4f9f3e-fc33-cf98-2322-27087664813f@list.ru> <a5565692-7f23-d236-3b01-142088310741@gmail.com>
Reply-To: wharms@bfs.de
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Return-path: <linux-kernel-owner@vger.kernel.org>
In-Reply-To: <a5565692-7f23-d236-3b01-142088310741@gmail.com>
Sender: linux-kernel-owner@vger.kernel.org
To: "Michael Kerrisk (man-pages)" <mtk.manpages@gmail.com>
Cc: Stas Sergeev <stsp@list.ru>, linux-man <linux-man@vger.kernel.org>, Andy Lutomirski <luto@amacapital.net>, Oleg Nesterov <oleg@redhat.com>, lkml <linux-kernel@vger.kernel.org>
List-Id: linux-man@vger.kernel.org



Am 30.10.2017 11:04, schrieb Michael Kerrisk (man-pages):
> [So, things fell on the floor, a while back.]
> 
> On 05/25/2017 11:17 AM, Stas Sergeev wrote:
>> 24.05.2017 14:09, Michael Kerrisk (man-pages) пишет:
>>> One could do this I suppose, but I read POSIX differently from
>>> you and, more importantly, SS_ONSTACK breaks portability on
>>> numerous other systems and is a no-op on Linux. So, the Linux man
>>> page really should warn against its use in the strongest terms.
>> So how about instead of the strongest terms towards
>> the code's author, just explain that SS_ONSTACK is a
>> bit-value on some/many OSes, and as such, 0 is a
>> valid value to enable sas on them, plus all the other
>> values would give EINVAL?
>> No strongest terms will help w/o an explanation,
>> because people will keep looking for something that
>> suits as a missing SS_ENABLE.
> 
> Fair enough. I've removed the statement in the manual page
> about "confusion". By now the page says:
> 
>     BUGS
>        In the lead up to the release of the Linux 2.4  kernel,  a  change
>        was   made   to   allow  sigaltstack()  to  accept  SS_ONSTACK  in
>        ss.ss_flags, which results in behavior that is the  same  as  when
>        ss_flags is 0 (i.e., the inclusion of SS_ONSTACK in ss.ss_flags is
>        a no-op).  On other implementations,  and  according  to  POSIX.1,

i am confused, i understand that:
           ss.ss_sp = malloc(SIGSTKSZ);

           ss.ss_size = SIGSTKSZ;
           ss.ss_flags = 0;
           if (sigaltstack(&ss, NULL) == -1)

is equivalent to:
           ss.ss_sp = malloc(SIGSTKSZ);

           ss.ss_size = SIGSTKSZ;
           ss.ss_flags = SS_ONSTACK ;
           if (sigaltstack(&ss, NULL) == -1)

but also to
           ss.ss_sp = malloc(SIGSTKSZ);

           ss.ss_size = SIGSTKSZ;
           ss.ss_flags = SS_ONSTACK | SOMETHING_FLAG ;
           if (sigaltstack(&ss, NULL) == -1)

so the use of SS_ONSTACK would result in ss.ss_flags = 0 no matter what.
OR
SS_ONSTACK is a no-op in Linux

re,
 wh

>        SS_ONSTACK appears only as a reported flag in old_ss.ss_flags.  On
>        Linux, there is no need ever to specify this flag in  ss.ss_flags,
>        and indeed doing so should be avoided on portability grounds: var‐
>        ious other systems give an error if  SS_ONSTACK  is  specified  in
>        ss.ss_flags.
> 
> Cheers,
> 
> Michael
>