ATTRIBUTES patch problems!

["Michael Kerrisk (man-pages)" <mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>]

[Was: Re: [PATCH v3] getutent.3: ATTRIBUTES: Note functions those 
aren't thread-safe]

On 04/01/2015 10:01 AM, Zeng Linggang wrote:
> The markings match glibc markings.

Hello Zeng Linggang and Ma Shimiao,

Arrgghhh! What is going on?

I started processing this queue of patches from Zeng Linggang, 
and now I see there are v2 and even now a v3 patch for some 
of the pages! With no explanation of the differences in 
one version of the patches from one version to the next...
And all of the different versions of the patches with a 
statement that the markings match the glibc markings...
That can't be true, because the different versions of the 
man-pages patches have different markings! (At least in the 
cases that I looked at.) Please explain.

And, all of the v1 patches are signed off by Ma Shimiao, implying
some degree of review by Ma Shimiao. Is this true? If so, why
do I get two or even three versions of the patches? Please 
explain.

I'm not happy about this. And I'd rather you could explain what's 
happening, and fix the problem. I'm sorry, but Zeng Linggang,
I have just reverted all the patches that I just applied from you,
and I've deleted the remaining v2 and v3 patches from you. Could you
get together with Ma Shimiao and send me new correct versions 
of these patches. And please, Ma Shimiao, if your name goes
into the Signed-off-by, I take this as some indication that you
checked the patches. If that's not the case, I should not see a 
Signed-off-by.

For the future, could you both, when sending patches, please 
in the commit message also quote the text of the glibc markings
so I can compare. (Just quoting the source text is fine.)

Thanks,

Michael



> Signed-off-by: Zeng Linggang <zenglg.jy-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
> Signed-off-by: Ma Shimiao <mashimiao.fnst-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
> ---
>  man3/getutent.3 | 55 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 55 insertions(+)
> 
> diff --git a/man3/getutent.3 b/man3/getutent.3
> index 4ff5581..1611d9b 100644
> --- a/man3/getutent.3
> +++ b/man3/getutent.3
> @@ -155,6 +155,61 @@ functions can also fail for the reasons described in
>  /var/run/utmp	database of currently logged-in users
>  .br
>  /var/log/wtmp	database of past user logins
> +.SH ATTRIBUTES
> +For an explanation of the terms used in this section, see
> +.BR attributes (7).
> +.TS
> +allbox;
> +lb lb lbw28
> +l l l.
> +Interface	Attribute	Value
> +T{
> +.BR getutent ()
> +T}	Thread safety	T{
> +MT-Unsafe init race:utent
> +.br
> +race:utentbuf sig:ALRM timer
> +T}
> +T{
> +.BR getutid (),
> +.br
> +.BR getutline ()
> +T}	Thread safety	T{
> +MT-Unsafe init race:utent
> +.br
> +sig:ALRM timer
> +T}
> +T{
> +.BR pututline ()
> +T}	Thread safety	T{
> +MT-Unsafe race:utent
> +.br
> +sig:ALRM timer
> +T}
> +T{
> +.BR setutent (),
> +.br
> +.BR endutent (),
> +.br
> +.BR utmpname ()
> +T}	Thread safety	MT-Unsafe race:utent
> +.TE
> +
> +In the above table,
> +.I utent
> +in
> +.I race:utent
> +signifies that if any of the functions
> +.BR setutent (3),
> +.BR getutent (3),
> +.BR getutid (3),
> +.BR getutline (3),
> +.BR pututline (3),
> +.BR utmpname (3),
> +or
> +.BR endutent (3)
> +are used in parallel in different threads of a program,
> +then data races could occur.
>  .SH CONFORMING TO
>  XPG2, SVr4.
>  .LP
> 


-- 
Michael Kerrisk
Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/
Linux/UNIX System Programming Training: http://man7.org/training/
--
To unsubscribe from this list: send the line "unsubscribe linux-man" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html