DaemonForums  

Go Back   DaemonForums > OpenBSD > OpenBSD Installation and Upgrading

OpenBSD Installation and Upgrading Installing and upgrading OpenBSD.

Reply
 
Thread Tools Display Modes
  #1   (View Single Post)  
Old 28th October 2018
hanzer's Avatar
hanzer hanzer is offline
Real Name: Adam Jensen
just passing through
 
Join Date: Oct 2013
Location: EST USA
Posts: 314
Default Building the System from Source

It's been a while since I've built an OpenBSD system from source. This is my process:
Code:
doas user mod -G wsrc hanzer

cd /usr
doas mkdir -p   xenocara ports
doas chgrp wsrc xenocara ports
doas chmod 775  xenocara ports
doas umount /usr/ports/pobj
# add wxallowed to pobj mount options in /etc/fstab
doas mount /usr/ports/pobj
cd /usr/ports
doas chgrp wsrc pobj
doas chmod 775 pobj

# logout - login

cd /usr
cvs -qd anoncvs@anoncvs3.usa.openbsd.org:/cvs checkout -rOPENBSD_6_4 -P src ports xenocara

#### This part as root ####
cd /sys/arch/$(machine)/compile/GENERIC.MP
make obj 
make config 
make && make install
shutdown -r now

cd /usr/src 
make obj && make build
sysmerge 
cd /dev && ./MAKEDEV all

cd /usr/xenocara 
make bootstrap 
make obj 
make build

rm -rf /usr/obj/*
rm -rf /usr/xobj/*
shutdown -r now
Questions:
1) Is it safe to use something like # make -j4 to build the kernel and # make -j4 build to build the rest of the system?

2) $ dmesg | head -2 says:
Code:
OpenBSD 6.4 (GENERIC.MP) #364: Thu Oct 11 13:30:23 MDT 2018
    deraadt@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC.MP
A new kernel was built and installed last night, Oct 27, by root. Why does this dmesg say Oct 11 and deraadt?
Reply With Quote
  #2   (View Single Post)  
Old 28th October 2018
jggimi's Avatar
jggimi jggimi is offline
More noise than signal
 
Join Date: May 2008
Location: USA
Posts: 6,455
Default

1. There are no guarantees that forced multi-threading your build will function correctly. The only supported procedure can be found in the release(8) man page.

2. Either you are not looking at the most recent dmesg(8) in the buffer, which wraps, or you are not using the kernel you believe to have been built and installed. You can confirm the latter by inspecting the date and time stamp of the kernel file, then inspecting the contents with strings(1) and grep(1).
Reply With Quote
  #3   (View Single Post)  
Old 28th October 2018
hanzer's Avatar
hanzer hanzer is offline
Real Name: Adam Jensen
just passing through
 
Join Date: Oct 2013
Location: EST USA
Posts: 314
Default

Quote:
Originally Posted by jggimi View Post
2. Either you are not looking at the most recent dmesg(8) in the buffer, which wraps, or ...
Yep, that was it. The current dmesg is at the other end of the buffer.

Here it is: http://daemonforums.org/showthread.p...5428#post65428

Thanks!
Reply With Quote
  #4   (View Single Post)  
Old 28th October 2018
ibara's Avatar
ibara ibara is offline
Joined your Slack team
 
Join Date: Jan 2014
Posts: 646
Default

Yes you're fine running make -j<whatever>.
Reply With Quote
  #5   (View Single Post)  
Old 28th October 2018
hanzer's Avatar
hanzer hanzer is offline
Real Name: Adam Jensen
just passing through
 
Join Date: Oct 2013
Location: EST USA
Posts: 314
Default

Quote:
Originally Posted by jggimi View Post
1. There are no guarantees that forced multi-threading your build will function correctly.
Quote:
Originally Posted by ibara View Post
Yes you're fine running make -j<whatever>.
This is a sticky wicket. How do FOSS communities typically resolve such contradictions? Testing doesn't seem like a reliable way to determine if the build system was designed and is maintained to support a multi-process make. It doesn't seem to be explicit in the documentation (FAQ, man). Hmm...
Reply With Quote
  #6   (View Single Post)  
Old 28th October 2018
jggimi's Avatar
jggimi jggimi is offline
More noise than signal
 
Join Date: May 2008
Location: USA
Posts: 6,455
Default

It's pretty simple. If it works, great! If it doesn't, it is not explicitly supported, so rerun your build without it, per release(8).



Parallel build is not discussed in /usr/share/mk/bsd.README, nor is "-j" used in any of the /usr/share/mk/*.mk files.
Reply With Quote
  #7   (View Single Post)  
Old 28th October 2018
hanzer's Avatar
hanzer hanzer is offline
Real Name: Adam Jensen
just passing through
 
Join Date: Oct 2013
Location: EST USA
Posts: 314
Default

Quote:
Originally Posted by jggimi View Post
It's pretty simple. If it works, great! I
From the perspective of epistemological methodology and the sociotechnical infrastructure to support it, I don't think a shallow trial would be sufficient here. What if something is done out of order in a way that doesn't stop the build but does corrupt some part of the system? (This is just armchair chat while I wait for a ports compile to complete). Two common approaches to answer such a question are to reason from the design or to extensively test the implemented system. (This reminds me, the Make with Ada registration is open through Oct 16, 2018 – Feb 15, 2019). Is there a comprehensive test suite? What's going to happen to these large FOSS projects like OpenBSD when the people who understand those subtle, undocumented and unstated design requirements and specifications are gone?
Reply With Quote
  #8   (View Single Post)  
Old 28th October 2018
ibara's Avatar
ibara ibara is offline
Joined your Slack team
 
Join Date: Jan 2014
Posts: 646
Default

We're not going anywhere, don't worry!
Reply With Quote
  #9   (View Single Post)  
Old 28th October 2018
hanzer's Avatar
hanzer hanzer is offline
Real Name: Adam Jensen
just passing through
 
Join Date: Oct 2013
Location: EST USA
Posts: 314
Default

Quote:
Originally Posted by ibara View Post
We're not going anywhere, don't worry!
LOL, I guess some of the harder truths of reality haven't sunken in yet, aye kid?
Reply With Quote
Old 28th October 2018
ibara's Avatar
ibara ibara is offline
Joined your Slack team
 
Join Date: Jan 2014
Posts: 646
Default

No, I mean that from an institutional knowledge perspective.
Reply With Quote
Old 28th October 2018
hanzer's Avatar
hanzer hanzer is offline
Real Name: Adam Jensen
just passing through
 
Join Date: Oct 2013
Location: EST USA
Posts: 314
Default

Quote:
Originally Posted by ibara View Post
No, I mean that from an institutional knowledge perspective.
It's probably still fairly common for much knowledge to be maintained entirely in some kind of organic information system - in the minds of its members - and to be primarily transmitted through various customs, traditions, rights and rituals. Being so subjective and implicit rather than objective and explicit, that seems better classified as community culture than institutional knowledge. I suppose one simple way to begin to distinguish the two is to examine what information has been encoded into explicit, objective records.
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Building a 10.2 system from source hanzer FreeBSD Installation and Upgrading 1 1st September 2015 01:36 AM
Building OpenBSD userland from source bceverly OpenBSD General 6 6th April 2015 07:26 PM
*** Error code 1 building OpenBSD 5.1-stable from source comet--berkeley OpenBSD Installation and Upgrading 12 19th May 2012 02:18 AM
Blinkenlights for all: Cisco's building automation system vulnerable to manipulation J65nko News 0 27th May 2010 04:26 PM
Building a New System, Ideas ninjatux General Hardware 19 20th May 2009 09:54 AM


All times are GMT. The time now is 10:07 AM.


Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Content copyright © 2007-2010, the authors
Daemon image copyright ©1988, Marshall Kirk McKusick