[OE-core] [PATCH 0/2] perf: enable man pages for 'help' functionality

[Randy MacLeod]

As Ross asked in a separate reply, we should discuss
the pros and cons of making the host-based generation of docs
be optional. I think that's the way to go since, in the common case
of not enabling docs, the builds would be faster and the output smaller.

On 2016-07-31 09:33 PM, dengke.du at windriver.com wrote:
> Hi Randy
>
> 1.  I have checked the asciidoc-native: 3.1M
>
> 2.  real    4m41.858s
>      user    27m3.280s
>      sys    6m32.372s



Those numbers, on their own, are not useful.

How long does it take to build another native package on your system?
What's the overall time  and time difference to build an image without
and then with asciidoc and perf (and maybe ccache, git, trace-cmd)
docs being generated? Alternatively just make the generation of docs
be optional and then the time isn't a critical concern.

You should follow-up, ***later*** by enabling doc generation for:

commit 40627f5c334544178b056078da5e1d645ebd2a38
Author: Robert Yang <liezhi.yang at windriver.com>
Date:   Mon Jul 25 01:16:31 2016 -0700

     ccache: 3.2.4 -> 3.2.5

     Add Revert-Create-man-page-in-the-make-install-from-git-.patch to
     disable asciidoc since we don't have it.


and perhaps:


commit 9aba4bf2143c228d58aac06764f87ace5dd21d02
Author: Paul Gortmaker <paul.gortmaker at windriver.com>
Date:   Tue Feb 10 14:17:37 2015 -0500

     git: expand recipe to take advantage of pre-gen'd manpages

     These could be created from scratch from git itself, but it
     requires asciidoc, xsltproc, python bits and too much other
     baggage.  Since the git folks issue a tarball with the manpages
     for each release, it is simpler to just go get that.

and:

commit 73ac48377491561151658617d8cc45936242eb0c
Author: Darren Hart <dvhart at linux.intel.com>
Date:   Wed Nov 30 17:58:52 2011 -0800

     trace-cmd: Update to 1.2 (includes kernelshark)

>
> 3.  Yes, the build system using python3.

so what about python2? Does asciidoc require a specific major version
of python?

../Randy

>
> //dengke
>
> On 2016年07月28日 02:23, Randy MacLeod wrote:
>> On 2016-07-27 01:02 AM, Dengke Du wrote:
>>> The following changes since commit
>>> 36feb38045b7a2af86ece147fec54b0db3bf491f:
>>>
>>>   linux-yocto: Update genericx86* SRCREVs for linux-yocto 4.4
>>> (2016-07-21 07:48:53 +0100)
>>>
>>> are available in the git repository at:
>>>
>>>   git://git.openembedded.org/openembedded-core-contrib
>>> dengke/enable-help-man-pages-for-perf
>>> http://cgit.openembedded.org/cgit.cgi/openembedded-core-contrib/log/?h=dengke/enable-help-man-pages-for-perf
>>>
>>>
>>> Dengke Du (2):
>>>   Asciidoc: add it
>>>   perf: enable man pages for 'help' functionality
>>>
>>>  meta/recipes-extended/asciidoc/asciidoc_8.6.9.bb | 20
>>> ++++++++++++++++++++
>>>  meta/recipes-kernel/perf/perf.bb                 | 11 ++++++++++-
>>>  2 files changed, 30 insertions(+), 1 deletion(-)
>>>  create mode 100644 meta/recipes-extended/asciidoc/asciidoc_8.6.9.bb
>>>
>>
>> Looks good .
>>
>> I was under the impression that asciidoc was larger but it's just
>>  ~ 900KB plus it's all python so it's not going to add to the system
>> build time directly.
>>
>> How much longer does it take to build perf?
>>
>> Did you test for python3 only?
>>
>


-- 
# Randy MacLeod. SMTS, Linux, Wind River
Direct: 613.963.1350 | 350 Terry Fox Drive, Suite 200, Ottawa, ON, 
Canada, K2K 2W5