logs archiveBotHelp.net / Freenode / #43oh / 2015 / September / 9 / 3
it was a lot easier to do web stuff on unix at first
perl and GGI.pm
all the pbm stuff
back when there wasn't any real cgi
* dynamic web pages
i remember using pbm to dynamically generate zoomed in images for fax
with a web page
back when people had 16MB of memory
the sun was a little faster :)
fu*k I can't believe how much space gcc 5.2.0 takes to compile
my /home FS had 1.8GB available and it ran out of space
gonna have to build off a beefier FS now
1.8GB hmmm
i think i have that for my swap space
that aint much
I mean I have plenty of FS's with 40+GB space..... just never thought gcc would swallow up all that
common problem in the past was /tmp was too small
this is something like my 4th attempt b/c of running out of space lol
finally figured it's time to move the build dir to another FS
* looks at fs available space /.. 192.5GB, 155.2, 16.4, 98.6, 106.8, 289.9 ...
diskspace .. we got diskspace
of course home is the only one with barely any
i have no idea what is taking up space in the millions of files i have
also crazy to think of, it's pretty easy for me to d/l big files "from" my home server to my work servers since my comcast uplink is reasonably fast
10Mbps or so
that is pretty decent
i used to hoard downloads in the past because they would disappear
ok rx-gcc build dir is up to 2.5GB already
I see now why it ran out of space :)
in the meantime, I bet the 12-core, hyperthreaded box at work will be done the whole build before the final gcc build is done on my home server
working from home today?
slow day at work though ;)
lol lack of makeinfo is boxing the work machine
got that installed now
hmm nope, home server finished first :)
although the work server was hampered by newlib bi**hing about no makeinfo and actually killing the build for that
(that's stupid)
there is a long list or prereqs to build that crap
of prereqs
aaand done
I found a good guide on doing it with OSX that I'm following under linux
just had to toss makeinfo in there to patch that up
but this procedure has you manually build gmp, mpfr and mpc
ok well it does appear that GCC 5.2.0 has the bug too
though I haven't actually tried running this code on my RX yet
and it appears to screw up the call for any function that also had a virtual variant in the parent class
well that was painful but easy in the end
when i figure out what i was doing wrong
if you checkout nuttx
and the apps git things
and then build it ..
it just works
* looking at a nsh prompt running on the tm4c1294
« prev 1 2 3 next »