Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 06 Jan 2006 22:18:09 -0700
From:      Scott Long <scottl@samsco.org>
To:        John Baldwin <jhb@freebsd.org>
Cc:        cvs-src@freebsd.org, src-committers@freebsd.org, cvs-all@freebsd.org
Subject:   Re: cvs commit: src/sys/kern kern_descrip.c
Message-ID:  <43BF4F11.9000103@samsco.org>
In-Reply-To: <200601061253.51551.jhb@freebsd.org>
References:  <200601061630.k06GUUXW022868@repoman.freebsd.org> <43BE9C19.6070600@samsco.org> <200601061253.51551.jhb@freebsd.org>

next in thread | previous in thread | raw e-mail | index | archive | help
John Baldwin wrote:
> On Friday 06 January 2006 11:34 am, Scott Long wrote:
> 
>>John Baldwin wrote:
>>
>>>jhb         2006-01-06 16:30:30 UTC
>>>
>>>  FreeBSD src repository
>>>
>>>  Modified files:
>>>    sys/kern             kern_descrip.c
>>>  Log:
>>>  Return EBADF rather than EINVAL for FWRITE failure as per POSIX.
>>>
>>>  MFC after:      1 week
>>>
>>>  Revision  Changes    Path
>>>  1.290     +2 -3      src/sys/kern/kern_descrip.c
>>
>>Technically this will be an API change if it get's MFC'd, though a
>>minor one.  If it needs to go into RELENG_6 then it should be
>>clearly documented in the release notes.
> 
> 
> Not really because except for the hwpmc(4) driver and the mqueue code (which 
> isn't in 6.x yet IIRC), every other caller of fget_write() ignores the errno 
> value and hardcodes EBADF, so it won't actually change the errno for any 
> applications in 6.x.
> 

While it may not be used by much in the FreeBSD tree, the fact that
fget_write() is available via a kernel header means that it is part
of the kernel API.  It is trivial, so all I ask is that it be well
documented when it gets MFC'd.  Something along the lines of:

- fget_write() was changed to provide return values as per POSIX.

Scott



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?43BF4F11.9000103>