GNU bug report logs - #21098
uname man page

Previous Next

Full log

View this message in rfc822 format

From: Pádraig Brady <P <at> draigBrady.com>
To: Stephane Chazelas <stephane.chazelas <at> gmail.com>
Cc: Norbert de Jonge <mail <at> norbertdejonge.nl>, Assaf Gordon <assafgordon <at> gmail.com>, Paul Eggert <eggert <at> cs.ucla.edu>, 21098 <at> debbugs.gnu.org, djm <at> djmnet.org
Subject: bug#21098: uname man page
Date: Wed, 22 Jul 2015 11:27:08 +0100

On 22/07/15 10:33, Stephane Chazelas wrote:
> 2015-07-22 01:54:58 +0100, Pádraig Brady:
> [...]
>> On 21/07/15 14:34, Paul Eggert wrote:
>>> Thanks, that patch looks good, except for some nits.  POSIX spells the phrase 
>>> "non-portable" and we might as well be consistent.  The --help lines would look 
>>> better as:
>>>
>>>    -p, --processor          print the processor type (non-portable)\n\
>>>    -i, --hardware-platform  print the hardware platform (non-portable)\n\
>>>
>>> as the period would look funny after a non-capitalized sentence.
>>
>> Done and pushed.
>> I've closed the bugs now since we've discouraged use of these options.
>> Since they're platform specific, any logic changes should be
>> in uname(1) and/or the kernel.
> [...]
> 
> Note that for Solaris, it's -m that's discouraged
> http://docs.oracle.com/cd/E23824_01/html/821-1461/uname-1.html
> 
> -p is useful as it gives (is meant to give) the instruction-set. 
> 
> Although not POSIX, it's fairly portable. Among the modern (and
> less modern) Unix players, I could only find HP/UX not
> supporting it.

Thanks for the extra info.
Just to be clear, for portability I was referring
to the POSIX options, but mainly to the value itself.
I.E. it's meant as a warning to have users consider
use of these values.  Also it's better to align
with the standards and most popular platforms.

thanks,
Pádraig.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.